Rubber Analysis:

Characterisation, Failure Diagnosis

and Reverse Engineering

Martin Forrest
 Rubber Analysis:
Characterisation, Failure Diagnosis
and Reverse Engineering

Martin Forrest

A Smithers Group Company

Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom

Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118
http://www.polymer-books.com
 First Published in 2018 by

Smithers Rapra
Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

©Smithers Information Ltd., 2018

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.

Every effort has been made to contact copyright holders of any material reproduced
within the text and the author and publishers apologise if
any have been overlooked.

ISBN: 978-1-91108-840-0 (softback)

978-1-91108-841-7 (ebook)

Typeset by Argil Services

P
reface

This book provides a comprehensive review of rubber analysis and its many
applications and uses in the rubber industry and in other industrial sectors. These
applications include fundamental characterisation work that can be carried out on
rubber polymers, the various levels of investigative work that can be undertaken
to obtain quality control and compositional information on rubber products, and
the contributions that rubber analysis makes to occupational hygiene studies in the
rubber industry, regulatory studies in the medical and pharmaceutical industries, and
regulatory compliance work in the food industry. An overview of the wide range of
analytical techniques (e.g., spectroscopic, chromatographic, thermal and elemental)
that can be used in the pursuit of the different objectives of these industrially orientated
applications of rubber analysis is included. The analytical techniques section also
provides extensive referencing for readers who wish to find out more information on
the fundamental principles and attributes of the individual techniques.

A practical, application-based approach has been used throughout the book, with
the emphasis on how the analytical techniques can be used, both individually
and in combination, to provide analysts and researchers in industry, consultancy
organisations and test houses with important and commercially useful information.
For example, in addition to the use of these techniques for routine quality control and
quality assurance purposes, their application to the identification and quantification
of specific additives is described, as is their important contribution to the industrially
important areas of reverse engineering, failure diagnosis and the identification of
contaminants.

The majority of rubber products are cured, crosslinked materials, and a specific section
is devoted to the study of the curing process and characterisation of cure state due to
their vital role in assuring that a rubber product will achieve its optimum properties
and provide satisfactory long-term performance in service.

In order to fully reflect the important role that rubber analysis and the analytical
techniques associated with it have today, a number of important specific applications
have been detailed. These include the studies that have been undertaken to characterise
rubber process fumes and to identify relationships between fumes and the composition

iii
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

of the rubber compounds that generate them. Also included is risk assessment work
such as the profiling and determination of extractables and leachables from rubber
products used in medical devices and pharmaceutical packaging, as well as the
identification and quantification of potential and actual migrants from food-contact
rubber products.

Finally, the inclusion of a section on the analysis and characterisation of rubber latex,
both the natural and synthetic versions, and the products that are manufactured from
them, ensures that this important class of rubber products is not overlooked.

In writing this book, the up-to-date, extensive information contained in the Smithers
Rapra Polymer Library has been used. This has been complemented by many other
sources of information, including the published findings of relevant research projects,
trade literature references, and the considerable experience that the author has
acquired during 40 years in the rubber industry.

iv
C
ontents

1 Introduction................................................................................................ 1

2 Analytical Techniques................................................................................. 7
2.1 Introduction..................................................................................... 7
2.2 Wet Chemistry Techniques............................................................... 7
2.3 Elemental Analysers and Elemental Analysis.................................... 9
2.4 Spectroscopic Techniques [15–17].................................................. 10
2.4.1 Infrared Spectroscopy [18]................................................ 10
2.4.2 Infrared Microspectrometry.............................................. 13
2.4.3 Raman Spectroscopy [26, 27]........................................... 14
2.4.4 Ultraviolet Light Spectroscopy [32].................................. 15
2.4.5 Nuclear Magnetic Resonance Spectroscopy [33–36]......... 15
2.4.6 Mass Spectroscopy Techniques [36].................................. 15
2.4.7 Atomic Absorption Spectrometry [38].............................. 17
2.4.8 Inductively Coupled Plasma–Atomic Emission
Spectroscopy [39]............................................................. 17
2.4.9 Inductively Coupled Plasma–Mass Spectrometry.............. 18
2.4.10 X-Ray Fluorescence Spectroscopy and X-Ray
Diffraction [40]................................................................ 18
2.4.11 Energy-Dispersive X-Ray Spectroscopy [41]..................... 19
2.4.12 Chemiluminescence Spectroscopy .................................... 20
2.5 Chromatographic Techniques [42] ................................................ 20
2.5.1 Gas Chromatography [43]................................................ 20
2.5.2 Gas Chromatography–Mass Spectrometry [44]................ 21

v
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

2.5.3 Gas Chromatography–Gas chromatography–Time-of-

flight Mass Spectrometry.................................................. 22
2.5.4 High-performance Liquid Chromatography [47].............. 25
2.5.5 High-performance Liquid Chromatography–Mass
Spectrometry [48–51]....................................................... 25
2.5.6 Ion Chromatography........................................................ 27
2.5.7 Supercritical Fluid Chromatography [52]......................... 27
2.5.8 Gel-permeation Chromatography/Size Exclusion
Chromatography [53, 54]................................................. 28
2.5.9 Thin-layer Chromatography and Paper
Chromatography [56] . .................................................... 30
2.6 Thermal Techniques [57, 58]......................................................... 30
2.6.1 Differential Scanning Calorimetry [59]............................. 30
2.6.1.1 Modulated Differential Scanning Calorimetry
[64]...................................................................... 32
2.6.1.2 HyperDSC™........................................................ 34
2.6.2 Dynamic Mechanical Analysis [71]................................... 35
2.6.3 Microthermal Analysis..................................................... 35
2.6.3.1 Determination of the Homogeneity of Polymer
Blends................................................................... 36
2.6.3.2 Studying Surfaces................................................. 36
2.6.3.3 Characterisation of Localised Thermal History.... 36
2.6.4 Thermal Mechanical Analysis........................................... 37
2.6.5 Thermogravimetric Analysis [9]........................................ 37
2.6.5.1 High-resolution Thermal Mechanical Analysis..... 38
2.6.5.2 Modulated Thermal Mechanical Analysis............. 38
2.6.5.3 Thermal Mechanical Analysis with Evolved Gas
Analysis................................................................ 39
2.6.6 Dielectric Analysis............................................................ 40
2.6.7 Thermal Conductivity Analysis......................................... 41
2.7 Microscopy Techniques [83, 84].................................................... 41

vi
 Contents

2.8 Rheological Instruments ............................................................... 42

2.9 Miscellaneous Analytical Techniques ............................................ 43

3 Principal Components and Bulk Composition........................................... 51

3.1 Introduction................................................................................... 51
3.2 Identification and Quantification of the Polymer in Rubber
Compounds................................................................................... 52
3.3 Characterisation of Rubbery Polymers . ........................................ 65
3.3.1 Determination of Molecular Weight [19] . ....................... 65
3.3.1.1 Gel Permeation Chromatography/Size Exclusion
Chromatography.................................................. 66
3.3.1.2 Viscosity .............................................................. 69
3.3.1.3 Osmometry ......................................................... 70
3.3.1.4 Light Scattering ................................................... 70
3.3.1.5 Miscellaneous Techniques . .................................. 71
3.3.2 Monomer Types and Microstructure ............................... 71
3.3.3 Specific Heat Values.......................................................... 73
3.3.4 Crystallinity . ................................................................... 74
3.3.5 Analysis of Rubber Blends................................................ 76
3.3.6 Degradation Studies and Changes in Chemical Structure.. 79
3.4 Determination of Plasticisers or Process Oils ................................ 80
3.4.1 Quantification ................................................................. 80
3.4.2 Identification ................................................................... 82
3.4.3 Determination of Polyaromatic Hydrocarbons in
Hydrocarbon Oils and Rubber Products.......................... 84
3.5 Determination of Carbon Black .................................................... 86
3.5.1 Quantification of Carbon Black in Rubber Vulcanisates .. 87
3.5.2 Dispersion of Carbon Black Within Rubber Products . .... 90
3.5.2.1 Cabot Dispersion Test ......................................... 90
3.5.2.2 Cut Surface and Torn Surface Methods ............... 91
3.5.2.3 Transmission Electron Microscopy ...................... 91

vii
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

3.5.2.4 Online Measurement during Processing................ 91

3.5.3 Isolation of Carbon Black from a Vulcanisate ................. 91
3.5.3.1 Digestion of the Matrix Followed by Filtration ... 91
3.5.3.2 Isolation by Non-oxidative Pyrolysis ................... 92
3.5.4 Surface Area Tests on Recovered Black . .......................... 92
3.6 Characterisation of Carbon Black.................................................. 93
3.6.1 Particulate and Aggregate Size ......................................... 93
3.6.2 Total Surface Area ........................................................... 94
3.6.2.1 Brunauer–Emmett–Teller Method (Nitrogen
Adsorption) ......................................................... 94
3.6.2.2 Iodine Adsorption ............................................... 94
3.6.2.3 Cetyltrimethylammonium Bromide Adsorption ... 94
3.6.3 External Surface Area ...................................................... 95
3.6.4 Determination of Polyaromatic Hydrocarbons in
Carbon Black................................................................... 95
3.6.5 Miscellaneous Analytical Techniques . ............................. 95
3.7 Determination of Inorganic Fillers ................................................ 95
3.8 Determination of the Bulk Composition of a Rubber..................... 98

4 Additives................................................................................................. 113
4.1 Introduction................................................................................. 113
4.2 Determination of Antidegradants in a Rubber Compound........... 116
4.2.1 Ultraviolet Stabilisers...................................................... 117
4.2.2 Antioxidants................................................................... 118
4.2.3 Antiozonants.................................................................. 123
4.2.4 Degradation Studies....................................................... 124
4.3 Determination of Cure System Additives and their Breakdown
Products....................................................................................... 128
4.3.1 Analysis of Uncured Samples.......................................... 129
4.3.2 Analysis of Cured Samples.............................................. 129
4.3.3 Cure State and Cure-state Studies................................... 138

viii
 Contents

4.4 Determination of Miscellaneous Additives................................... 138

4.4.1 Blowing Agents.............................................................. 138
4.4.2 Flame Retardants........................................................... 139
4.4.3 Process Aids................................................................... 141
4.4.3.1 Plasticisers and Oils............................................ 141
4.4.3.2 Resins................................................................. 141
4.4.3.3 Pine Tar and Factice........................................... 143
4.4.4 Peroxide Co-Agents........................................................ 143
4.4.5 Stearic acid..................................................................... 143
4.5 General Analysis Work for the Identification and Quantification
of Additives................................................................................. 144

5 Reverse Engineering and Product Deformulation.................................... 153

5.1 Introduction................................................................................. 153
5.2 Use of GC×GC–ToF–MS for Reverse Engineering........................ 156
5.3 Case Study – Deformulation of a Rubber Sample........................ 166
5.3.1 Introduction................................................................... 166
5.3.2 Initial Suite of Tests........................................................ 166
5.3.2.1 Quantitative Solvent Extraction.......................... 166
5.3.2.2 Ash Content....................................................... 167
5.3.2.3 Thermogravimetric Analysis............................... 167
5.3.2.4 Fourier–Transform Infrared Spectroscopy.......... 167
5.3.2.5 X-Ray Fluorescence Spectroscopy....................... 167
5.3.2.6 Gas Chromatography–Mass Spectrometry.......... 168
5.3.3 Results from the Initial Suite of Tests.............................. 168
5.3.4 Additional Tests.............................................................. 174
5.3.4.1 Quantitative Elemental Determinations.............. 174
5.3.4.2 Solid-state Nuclear Magnetic Resonance............ 175
5.3.5 Results from the Additional Tests................................... 175
5.3.6 Summary of the Information Obtained on the Unknown
Sample............................................................................ 177

ix
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

5.4 Published Reverse Engineering Studies......................................... 179

6 Curing and Cure State Studies................................................................. 187

6.1 Introduction................................................................................. 187
6.2 Thermal Analytical Techniques for Curing and Cure State
Studies......................................................................................... 189
6.2.1 Differential Scanning Calorimetry.................................. 189
6.2.2 Dynamic Mechanical Analysis........................................ 194
6.2.3 Other Thermal Techniques............................................. 196
6.3 Use of Chromatographic and Spectroscopic Techniques for
Curing and Cure State Studies..................................................... 198
6.4 Rheometric Studies...................................................................... 199
6.5 Finite Element Analysis and Mathematical Modelling................. 202
6.6 Other Techniques and Methodologies for Studying Curing and
Cure State.................................................................................... 204
6.7 Comparative Studies.................................................................... 205

7 Characterisation of Rubber Process Fume............................................... 211

7.1 Introduction................................................................................. 211
7.2 Composition of Rubber Process Fume......................................... 213
7.3 Principal Factors Affecting the Composition of Rubber Fume in
the Workplace.............................................................................. 214
7.3.1 Influence of Rubber Compound Formulation on
Composition of Rubber Fume........................................ 214
7.3.2 Influence of Different Processes and Processing
Temperatures on the Composition of Rubber Fumes...... 215
7.4 Capture and Analyses of Rubber Process Fumes.......................... 216
7.4.1 Different Sampling Techniques and Analytical
Methods......................................................................... 216
7.4.2 Influence of Different Sampling and Analytical
Techniques on Rubber Fumes Data................................ 218
7.5 Analytical Studies Carried Out on Rubber Fumes........................ 220
7.5.1 Rubber Fume Data Resulting from Factory Studies........ 220

x
 Contents

7.5.2 Rubber Fume Data Resulting from Laboratory Studies.. 223

7.6 Conclusions................................................................................. 228

8 Compliance with Food-contact Regulations............................................ 235

8.1 Introduction................................................................................. 235
8.2 Rubbers for Food-contact Applications....................................... 237
8.3 Overview of the European Union and United States Food and
Drug Administration Regulations................................................ 239
8.3.1 European Union Legislation........................................... 239
8.3.2 German Regulations....................................................... 239
8.3.3 Requirements in France.................................................. 242
8.3.4 Requirements in the United Kingdom............................. 242
8.3.5 United States Food and Drug Administration
Regulations.................................................................... 243
8.3.6 Council of Europe Resolutions....................................... 245
8.3.6.1 CoE Resolution on Rubber Products.................. 245
8.3.6.2 CoE Resolution on Silicone Materials................. 247
8.4 Analysis Work to Characterise Food-contact Rubbers................. 248
8.4.1 Introduction................................................................... 248
8.4.2 Fingerprinting Potential Migrants from Rubber
Compounds.................................................................... 249
8.4.2.1 Use of Gas Chromatography–Mass
Spectrometry to Fingerprint Food-contact
Rubber Samples.................................................. 249
8.4.2.2 Use of Liquid Chromatography–Mass
Spectrometry to Fingerprint Food-contact
Rubber Samples.................................................. 251
8.4.3 Determination of Specific Substances in Food-Contact
Rubbers.......................................................................... 255
8.4.3.1 Monomers.......................................................... 255
8.4.3.2 Plasticisers and Process Oils................................ 256
8.4.3.3 Cure System Species, Accelerators and their
Reaction Products.............................................. 256

xi
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

8.4.3.4 Antidegradants and their Reaction Products....... 256

8.4.3.5 Oligomers........................................................... 257
8.5 Analysis Work to Detect and Quantify Specific Migrants in
Food Simulants and Food............................................................ 257
8.5.1 Introduction................................................................... 257
8.5.2 Chemical Type of the Potential Migrant......................... 258
8.5.2.1 Alkylphenol and Bisphenol A............................. 258
8.5.2.2 Peroxide Breakdown Products............................ 259
8.5.2.3 Dimethylsiloxanes and other Components
from Silicone Rubbers........................................ 259
8.5.2.4 Accelerators and Antidegradants........................ 260
8.5.2.5 Oligomers........................................................... 261
8.5.3 Food-contact Products.................................................... 261
8.5.3.1 Teats and Soothers.............................................. 261
8.5.3.2 Rubber Adhesives............................................... 263
8.5.3.3 Meat Netting...................................................... 263
8.5.3.4 Rubber Gloves for Handling Food..................... 264
8.5.3.5 Thermoplastic Rubbers....................................... 264
8.5.3.6 Natural and Synthetic Rubber Products.............. 264
8.5.3.7 Silicone Rubber Products.................................... 266

9 Extractable and Leachable Studies on Rubber Products.......................... 275

9.1 Introduction................................................................................. 275
9.1.1 Extractables.................................................................... 276
9.1.2 Leachables...................................................................... 276
9.2 Origins of Substances that can contribute to Extractable and
Leachable Profiles of Rubbers...................................................... 278
9.2.1 Polymer.......................................................................... 278
9.2.1.1 Polymerisation Residues, Monomers, Oligomers
and Impurities.................................................... 278
9.2.1.2 Degradation Products of the Polymer................. 280

xii
 Contents

9.2.2 Additives in the Unaltered State...................................... 281

9.2.2.1 Plasticisers and Fillers......................................... 281
9.2.2.2 Antidegradants, Process Aids and Pigments........ 281
9.2.3 Breakdown and Reaction Products of Cure System
Components................................................................... 282
9.2.3.1 Peroxides............................................................ 282
9.2.3.2 Sulfur Donor-type Cure Systems......................... 283
9.2.3.3 Accelerators used in Sulfur-type Cure Systems.... 284
9.2.4 Breakdown and Reaction Products of Antidegradants
used in Rubbers.............................................................. 286
9.2.4.1 Antioxidants....................................................... 286
9.2.4.2 Antiozonants used in Rubber Products............... 287
9.2.5 Carbon Black Fillers and the Formation of
Nitrosamines.................................................................. 287
9.3 Overview of Extractable and Leachable Testing of Rubber
Products....................................................................................... 288
9.4 Case Study – Potential and Actual Extractables from an Nitrile
Rubber Compound...................................................................... 293
9.4.1 Composition of the Nitrile Rubber Compound.............. 293
9.4.2 Potential Extractables..................................................... 294
9.4.3 Gas Chromatography–Mass Spectrometry Analysis of
the Nitrile Rubber Sample.............................................. 296
9.4.3.1 Headspace Gas Chromatography–Mass
Spectrometry...................................................... 296
9.4.3.2 Solution Injection Gas Chromatography–Mass
Spectrometry...................................................... 296
9.4.4 Analysis of the Nitrile Rubber Sample by Liquid
Chromatography–Mass Spectrometry............................. 298
9.4.5 Summery of the Extractables Detected by Gas
Chromatography–Mass Spectrometry and Liquid
Chromatography–Mass Spectrometry Analysis............... 299
9.5 Determination of Extractables and Leachables............................ 301

xiii
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

10 Analysis of Surface Blooms and Contaminants........................................ 313

10.1 Introduction................................................................................. 313
10.2 Surface Blooms............................................................................ 313
10.2.1 Introduction................................................................... 313
10.2.2 Different Types of Bloom or Bloom-like Phenomena...... 317
10.2.2.1 True Blooms....................................................... 317
10.2.2.2 Modified Blooms................................................ 317
10.2.2.3 Pseudo-blooms................................................... 318
10.2.2.4 Surface Contamination....................................... 318
10.2.2.5 Staining or Discolouration.................................. 318
10.2.2.6 Hazing................................................................ 319
10.2.3 Analysis and Identification of Blooms............................. 319
10.2.3.1 Simple Investigative Tests................................... 319
10.2.3.2 In Situ Tests........................................................ 320
10.2.3.3 Modelling of Blooming....................................... 326
10.3 Contaminants.............................................................................. 327
10.3.1 Introduction................................................................... 327
10.3.2 Identification of Different Types of Contaminant........... 327
10.3.2.1 Particulates......................................................... 327
10.3.2.2 Fibres................................................................. 328
10.3.2.3 Liquids............................................................... 329

11 Analysis of Rubber Latices...................................................................... 333

11.1 Introduction................................................................................. 333
11.2 Quality Control Tests................................................................... 335
11.2.1 Sampling........................................................................ 335
11.2.2 Coagulum Content......................................................... 335
11.2.3 Total Solids Content....................................................... 335
11.2.4 Dry Rubber Content....................................................... 335
11.2.5 Dry Polymer Content..................................................... 336

xiv
 Contents

11.2.6 Density........................................................................... 336

11.2.7 Determination of pH...................................................... 336
11.2.8 Alkalinity....................................................................... 336
11.2.9 Potassium Hydroxide Number....................................... 337
11.2.10 Volatile Fatty Acid Number............................................ 337
11.2.11 Carbon Dioxide Number................................................ 337
11.2.12 Non-volatile Acids.......................................................... 338
11.2.13 Higher Fatty Acids......................................................... 338
11.2.14 Electrical Conductivity................................................... 338
11.2.15 Latex Allergy Tests, Protein Tests and Sensitivity Tests... 338
11.3 Chemical Stability Tests............................................................... 339
11.3.1 Natural Rubber Latices.................................................. 340
11.3.2 Synthetic Rubber Latices................................................ 340
11.4 Polymer Composition of Synthetic Latices................................... 340
11.4.1 Bound Styrene Content................................................... 340
11.4.2 Total Bound Styrene Content.......................................... 340
11.4.3 Bound Acrylonitrile Content........................................... 341
11.4.4 Volatile Monomers......................................................... 341
11.5 Particulate Property Tests............................................................. 342
11.5.1 Viscosity......................................................................... 342
11.5.2 Surface Tension (Surface Free Energy)............................ 342
11.5.3 Soap Content.................................................................. 343
11.5.4 Soap Deficiency.............................................................. 343
11.5.5 Particle Size.................................................................... 343
11.6 Miscellaneous Chemical Tests...................................................... 344
11.6.1 Nitrogen and Trace Metals............................................. 344
11.6.2 Zinc Diethyldithiocarbamate.......................................... 344
11.6.3 Boric acid....................................................................... 345
11.6.4 Determination of Additives and Compounding
Ingredients...................................................................... 345

xv
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

11.7 Characterisation Studies Undertaken on Rubber Latices.............. 347

12 Conclusions............................................................................................. 355

Appendix 1 Standard Nomenclature System for Rubbers................................... 357

Appendix 2 International Rubber Analysis Standards (ISO)............................... 361

Appendix 3 Specific Gravities of Rubbers and Compounding Ingredients.......... 369

Abbreviations..................................................................................................... 373

Index ............................................................................................................... 383

xvi
1
Introduction

This book follows on from a review of rubber analysis published by the author as
a Rapra Review Report in 2001 [1]. In the intervening 17 years, there have been
number of important developments in the field. Some of these were associated with
the development and commercialisation of new analytical techniques and others led
to the appearance of improved versions of established techniques, both of which
have increased the speed, efficiency and capability available to the rubber analyst.
Other developments were associated with new testing methodologies created to help
protect the environment and to safeguard human health. Alongside these advances
and developments, there has also been a significant amount of new research and
investigative work carried out and published. Taking everything into account, it was,
therefore, considered to be timely and important to reflect and represent all of the
these changes in a new, and very much enlarged edition of the book.

The analysis of rubber products to demonstrate compliance with regulations

is one area that has grown significantly since 2001. In addition to the many
regulations that existed prior to 2001, new regulations have been published, such
as those concerning the presence of polyaromatic hydrocarbons in extender oils
[2] and a new Chinese standard for the use of rubber products for food-contact
applications [3]. As well as these specific, rubber-related regulations, major new
European Union regulations have been published which have had a significant
and wider impact on industry, such as the Registration, Evaluation, Authorisation
and restriction of Chemicals [4] and the Restriction of Hazardous Substances [5].
Regulatory bodies, such as the US Food and Drug Administration, have put more
emphasis since 2001 on identifying and quantifying the substances that could
migrate from the rubber products used in medical devices, and in drug storage,
dispensing and manufacturing systems.

It is for these reasons that dedicated sections that deal specifically with the analysis
work that has been undertaken in these subject areas have been included in this book,
as well as other sections that address additional specific areas of interest, such as the
characterisation of rubber process fume, studying the curing behaviour of rubbers,
and the identification of blooms and contaminants in rubber products.

1
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

To ensure that the major ‘themes’ of rubber analysis work are fully represented, there
are also extensive sections in the book that deal with the identification, quantification
and characterisation of the principal constituents in a rubber compound (i.e., polymer,
plasticiser/oil and filler), and of the wide range of additives that can be used in
rubber compounds (e.g., cure system and antidegradant system species). To provide
a thorough background to all the investigative and themed sections, there is also a
major section which describes the extensive range of analytical techniques available
to rubber analysts when they are in pursuit of their various and diverse goals.

The overall objective of this book, therefore, is to present an introduction and overview
of rubber analysis and the analytical methods and techniques used to achieve a wide
range of objectives, including:

• Quality-control work

• Failure investigations

• Deformulation studies

• Characterisation work (e.g., curing behaviour)

• Work on regulatory compliance (e.g., for food-contact applications)

• Work to address health and safety or environmental concerns (e.g., characterisation

of rubber fume)

To set the scene for a book on rubber analysis, it is useful to provide a brief introduction
to rubber technology and, hence, provide the background to the complexity of the
challenges that the rubber analyst often encounters. This is the intended function of
the next six paragraphs.

To obtain useful products that perform under various demanding conditions, a

rubber compound can be ‘tailor-made’ by selecting from a broad range of polymers
and additives. With rubbers, the possible compositional permutations are made more
numerous by an extensive array of plasticisers, fillers, process aids, antidegradants
and cure systems. The technology of rubbers is, therefore, a mature one, and allows
‘fine tuning’ of a compound to fit a number of seemingly conflicting design criteria
and product requirements. Hence, rubber compounds and products present analysts
with one of their most difficult, but satisfying, challenges.

Natural and synthetic rubber compounds are used in a highly diverse range of rubber
products for many applications and are manufactured throughout the world for
various sectors of industry and end users. The examples of the products that can be
manufactured from rubber compounds vary from bridge bearings to rubber seals,

2
 Introduction

gaskets, adhesives and elastic cord (Table 1.1). The majority of these products are
manufactured using traditional thermoset rubbers (i.e., vulcanised rubbers), but
there are also a number of thermoplastic rubbers available in which the crosslinks
are physical as a result of crystalline domains, or those that have a glass transition
temperature above ambient. Of the sectors where rubber is used, the automotive
industry is of particular importance in that tyre-related products, account for ~60%
of the synthetic rubber and ~75% of the natural rubber used worldwide [6].

Table 1.1 Examples of commercial rubber products

Generic class of product Examples of commercial products
Tyres Passenger car, truck tyres, racing, cycle, off-road tyres, inner tubes,
curing bladders
Conveyor/transmission Steel cord conveyor belting, repair material for conveyor belting,
belting scrapers, mining conveyors, V-belts, flat belts, synchronous belts
Industrial hose Water hose, high-pressure hose, welding hose, hydraulic hose,
spiral hose, offshore hose, oil hose, chemical hoses
Automotive products Hoses and gaskets, anti-vibration mounts, timing belts, window
and door seals, transmission and engine components, wiper blades
General mouldings/ Seals and gaskets, floor and roller coverings, sheeting, protection
sheeting linings, microcellular products, rubberised fabric, wire and cable
jackets, pump impellors, rail mounts, bridge bearings
Medical/pharma Surgeons’ gloves, medical tubing, gaskets, catheters, dialysis
products products, implants, contraceptives, baby soothers, tubing and
valves, masks and respirators
Clothing Footwear, protective suits, household gloves, industrial gloves,
footwear/boot heels and soling, cellular rubber soles, diving suits,
coated fabrics, sports footwear and clothing
Food contact products Transportation belts and hoses, gloves, pipe and machinery seals
and gaskets, plate heat exchanger gaskets, seals/gaskets for cans
and bottles, tank linings
Miscellaneous products Adhesives, rubberised asphalt, high-vacuum and radiation
components, carpet backing, latex thread, sealants and caulking,
toys

Table 1.1 provides an indication of the diverse range of rubber products that are
manufactured throughout the world today. Both the composition of a rubber
compound and the manufacturing processes that are used to produce it will depend
upon the individual product that is being manufactured.

3
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

As mentioned above, rubber products have complex compositions and the company
or individual who designs a rubber formulation for a specific product has a large
number of ingredients to choose from [7]. Although there are general guidelines for
designing rubber compounds, the final formulation depends upon the knowledge and
expertise of each rubber compounder, i.e., there are no ‘standard’ rubber compounds.

One of the reasons that the composition of a rubber compound can vary enormously
is because there are many types of base polymer (>20), each having a unique chemical
structure which has a direct influence on its inherent chemical and physical properties,
processing behaviour and the applications and products in which it can be used
[8]. It is also the case that for each generic type of rubbery polymer (e.g., ethylene
propylene diene monomer rubber), there are a significant number of different grades
available, each of which may differ in a number of aspects (e.g., molecular weight
or comonomer type). For the ethylene propylene class of rubbers, for example, over
100 different grades are available.

In addition to the base polymer, rubbers are formulated with a range of additives
(typically between 5 and 20) to achieve a compound that has the desired processing
and service properties [9]. There are also many cases where two or more base polymers
are mixed and blended together in order to achieve the desired properties (e.g.,
some of the compounds used in the manufacture of tyres). Also, within each class of
ingredient (e.g., antioxidant) there is a wide range of different chemical compounds
available on the market, each having different structures and properties, from a
number of manufacturers. Within another additive group, the carbon blacks, there
are in excess of 30 possible products. The composition of rubber compounds can
also alter during manufacturing because some of the ingredients are volatile at typical
processing temperatures and others (e.g., cure accelerators and antioxidants) generate
breakdown and reaction products [10–12], all of which makes accurate quantification
of the original concentration extremely difficult to derive. It is also common industry
practice to use more than one ingredient from a particular class in a formulation
to achieve the desired properties (e.g., two or more accelerators). Certain rubber
compounds may also contain reactive ingredients (e.g., pre-vulcanisation inhibitors)
at a relatively low (<0.5%) level, which can make detection difficult.

The rubber analyst, therefore, has to bear in mind all these complexities to ensure an
effective analytical strategy is devised and, also consider the possible interferences and
difficulties that may arise and influence the data. The usual result is that when, for
example, reverse engineering work is carried out on a rubber compound, a number of
wet chemistry, elemental, spectroscopic, chromatographic and thermal techniques have
to be used in an integrated and structured approach in order to ensure a successful
result. It is still the case, though, that even with the greatest care, and in the most
advantageous situations, >90% of a complex rubber formulation can be completely

4
 Introduction

elucidated by analytical means alone. Some compound development work will still
be required in order to fill knowledge gaps and find a satisfactory match for the
processing, curing, physical properties and in-service performance of the target sample.

To assist the practicing analyst, the emphasis throughout the book is of an empirical
nature, with the author drawing on his 40 years of technical experience within the
polymer industry, to provide practical illustrations of what can be achieved using
certain analytical techniques and methodologies. For example, to assist analysts
with certain specific areas of investigation (e.g., pyrolysis gas chromatography–mass
spectrometry analysis), examples of the experimental conditions that have been used
to carry out particular analyses have been provided and the results obtained by the
use of these conditions displayed to illustrate the capability and diagnostic power
of a technique.

In order for the analyst to have a full understanding of the capability and efficacy of the
analytical methods and techniques that are available today, it is extremely useful for them
to have an understanding of rubber technology itself, beyond the very brief introduction
that is provided in the paragraphs above. Also, with respect to characterisation and
deformulation studies, the analysts’ job when analysing ‘unknowns’ is always made
easier if at least some information on the sample is available at the outset (e.g., the
generic rubber type and the intended application). Armed with this information, and
guided by their background knowledge of rubber technology, analysts can work more
efficiently and often obtain more comprehensive results.

To assist the reader in obtaining a greater understanding of rubber analysis and

rubber technology, relevant books, academic articles, guidance documents, testing
standards and other information references are cited throughout each of the sections
in this book. There are also Appendices at the end of the book which summarise the
standard nomenclature system for rubbers (Appendix 1), International Standards
(International Organization for Standardization) for rubber analysis (Appendix 2),
and the specific gravities of rubbers and compounding ingredients (Appendix 3).

References

1. M.J Forrest in Rubber Analysis – Polymers, Compounds and Products, Rapra

Review Report No.139, Smithers Rapra, Shawbury, UK, 2001.

2. EU Directive 2005/69/EC: PAH’s in Tread Rubber.

3. GB 4806.11-2016: Chinese National Food Safety Standard – Food Contact

Rubber Materials and Articles.

5
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

4. EU Regulation (EC) 1907/2006: REACH.

5. EU Directive 2002/95/EC: Restriction of Hazardous Substances.

6. Rubber Statistical Bulletin, 2008, 60, 4–5, 2.

7. Rubber Compounding Ingredients Sourcebook, Rapra Technology Ltd,

Shawbury, UK, 2000.

8. J.R. White and S.K. De in Rubber Technologists Handbook, Smithers Rapra,

Shawbury, UK, 2001.

9. H. Long in Basic Compounding and Processing of Rubber, ACS Rubber

Division, Akron, OH, USA, 1985.

10. N. Grassie and G. Scott in Polymer Degradation and Stabilisation,

Cambridge University Press, Cambridge, UK, 1985.

11. A.G. Davies in Organic Peroxides, Butterworths, Oxford, UK, 1961.

12. A.Y. Coran in Vulcanisation, Science and Technology of Rubber, Ed.,

F.R. Eirich, Academic Press, New York, NY, USA, 1978.

6
2
Analytical Techniques

2.1 Introduction

The objective of this section is to provide an introduction and overview to the extensive
range of analytical techniques that are available to the rubber analyst. Although there
are a very large number available, an attempt has been made to include as many
types and classes, and variants within specific classes, as possible. If the reader would
like more information on a particular analytical technique and its capabilities and
relevant applications, there are a number of specialist books available, and examples
have been provided throughout this section. For a broader coverage, there are books
that cover a wide range of techniques and methodologies within general topics, such
as the characterisation of polymers [1–3], or address particular themes, including:

• Practical techniques [4, 5].

• Determination of additives [6, 7].

• Application of thermal techniques [8, 9].

• Application of specific analytical techniques, for example, infrared (IR) and Raman
spectroscopy [10].

• Study of particular fundamental properties, for example, polymer morphology

[11].

2.2 Wet Chemistry Techniques

Before the advent of modern instrumental techniques, rubber analysis relied on

traditional wet chemistry (e.g., colorimetric) techniques to determine information
such as the generic polymer type of a compound and the nature of the cure system
present; classical organic chemistry methods were used constantly. The amount of wet
chemistry carried out in the modern rubber analysis laboratory is minimal, but there
are some tests which can still be grouped under this heading, although even some of
these employ an instrumental technique to provide the final answer.

7
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The determination of total sulfur and combined sulfur in rubber can be carried out
using essentially the same method; the only difference is that the combined sulfur is
the sulfur that remains after a solvent extraction has been carried out on the sample.
There are a number of methods that can be used and workers have been developing
approaches to the determination of sulfur in organic compounds since the 19th century.
Two standardised methods that are still in use today in some laboratories are the tube
furnace combustion method and oxygen flask combustion method.

The tube furnace method, which is described in International Organization for

Standardizations’ (ISO) standard, ISO 6528-3, can be used on rubber samples that
are fully compounded and contain chlorine and/or nitrogen. The sample is burnt in
a stream of oxygen at very high temperatures (e.g., 1,350 °C) to ensure that all the
sulfur in the compound (even that in inorganic fillers such as barium sulfate) has
reacted and is included in the final value. Adding catalysts into the method can reduce
this temperature to ≈1,000 °C. If the sulfur contribution from inorganic additives is
not desirable then catalysts can be left out and the test carried out at 1,000 °C. The
volatile products are trapped in a solution of hydrogen peroxide (H2O2) and then a
titration carried out to a permanent pink colour using 0.01 M barium perchlorate
in a mixture of 80:20 (v/v) propan-2-ol and water.

The oxygen flask method, described in ISO 6528-1 and the British Standard (BS),
BS 7164-23.1, also uses an initial burn in oxygen, although the sample is trapped
between two platinum gauge ‘flags’, and the trapping of the volatile products in an
absorbent solution of H2O2. Once the volatiles have been trapped, however, there
is a choice of analytical method. A titration can be undertaken in the same way as
for the tube furnace method, or ion chromatography (IC) can be used to quantify
the sulfate ion. In the case of the titration approach, care has to be taken because a
number of interferences can take place. For example, zinc sulfate [produced during
the combustion of zinc oxide (ZnO) or due to the oxidation of zinc sulfide] will
dissolve in the H2O2 solution, and zinc ions, in general, will interfere with the end
point; these have to be removed by sending the H2O2 solution down through an
ion-exchange column. Even small amounts (e.g., 0.5%) of chlorine in a rubber will
give poor results due to the volatility of zinc chloride and hence its take-up by the
absorption solution. If the sample contains calcium carbonate (CaCO3), a change
in the absorption solution is required to a mixture of 0.25 ml of concentrated
hydrochloric acid, 2 ml of distilled water and 1 ml of 6% H2O2. This will dissolve
the calcium sulfate formed and decompose any CaCO3. An ion-exchange column is
then used prior to the titration. These problems, which are also common to the tube
furnace titration step, are completely removed by using IC to quantify the sulfur via
the sulfate ion (Section 2.5.7).

The oxygen flask combustion technique can also be used for the determination of

8
 Analytical Techniques

total halogens (titration method) or quantification of specific halogens (IC method) in

a rubber. The absorbing solution will typically comprise 1.5 ml of 0.05 M potassium
hydroxide, 0.2 ml of 30% H2O2 and 10 ml of distilled water.

It is sometimes necessary to determine the amount of elemental sulfur remaining in

a rubber. One method, the copper spiral method, is the subject of a standard [12];
other methods include the sulfite method, polarography, and reverse-phase high-
performance liquid chromatography (HPLC) using acetonitrile/water gradient elution
and ultraviolet (UV) detection at 270 nm.

2.3 Elemental Analysers and Elemental Analysis

The quantification of certain elements is required so often in so many branches of

science that for a number of years now commercial instruments have been available
for their specific determination.

Several configurations are available depending on user requirements. Examples which

are useful for the analysis of rubbers and other polymeric samples include:

• Carbon:hydrogen:nitrogen

• Carbon:nitrogen:sulfur

• Carbon:hydrogen:nitrogen:sulfur

• Carbon:hydrogen:nitrogen:oxygen:sulfur

These dedicated analysers are able to provide quick, accurate and easy quantification
of these elements on samples which do not require much preparation work and, in
some cases, at sample masses as low as 1 mg.

An international standard has been published [13] which deals with the determination
of the total nitrogen content of a sample using an automated analyser. No specific
manufacturer’s instrument is specified, and the standard focuses on the experimental
procedure to be used. A standard has also been published [14] which deals with the
determination of the total sulfur content of a sample using an automated analyser.
Again, no specific instrument is specified and the standard focuses on the experimental
procedure.

These instruments, therefore, offer an efficient and cost-effective way to obtain

important quantitative elemental data that can be used for a variety of purposes,
for example:

9
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Quantifying the amount of acrylonitrile in a nitrile rubber (NBR).

• Calculating the ratio of NBR and polyvinyl chloride in blends of those polymers.

• Determinations of total sulfur and bound sulfur.

• Differentiating between a sulfur-cured rubber and peroxide-cured rubber.

The determination of the concentration of other elements can also be very useful in
rubber analysis work. In particular, it is often necessary to determine the amount
of a halogen, such as chlorine or bromine, in the calculation of polymer content or
polymer-blend proportions. Also, the targeting of specific elements can enable the
quantity of certain additives (e.g., phosphate flame retardant or titanium dioxide
pigment) to be calculated. The other techniques and instruments that can be used to
identify and quantify elements within rubber samples are covered in Section 2.2 and
in all the sections from 2.4.7–2.4.11.

2.4 Spectroscopic Techniques [15–17]

2.4.1 Infrared Spectroscopy [18]

IR spectroscopy or, Fourier–Transform infrared (FTIR) spectroscopy as it is often

called due to the ubiquitous nature of the Fourier–Transform version, is a relatively
cost-effective, quick analytical technique that is extensively used in the analysis of
rubbers to identify the base polymer as well as the main (i.e., bulk) additives (e.g.,
plasticisers, oils and fillers).

Although the IR spectra of polymers and other organic substances can be recorded
using lasers emitting IR light in the near-, far- and mid-ranges, the majority are
recorded in the mid-range (i.e., 400–4,000 cm−1) and commercial instruments and
IR libraries reflect this fact. To record transmission IR data it is necessary to place
a sample on an IR transparent ‘window’ and, typically, sodium chloride plates are
used in standard IR accessories, whereas diamond ‘windows’ are commonly used
in ‘single bounce’-type attenuated total reflectance (ATR) accessories due to their
greater durability. A number of other sampling accessories are available, and one that
can be useful for direct rubber analysis work is a ‘multi-reflectance’ ATR accessory
equipped with a germanium crystal due to its relatively small degree of penetration
into the sample. This property can help reduce the interference that results due to
the presence of carbon black (CB), which is also referred to below.

10
 Analytical Techniques

To assist in the identification of polymers, IR spectral libraries are available, such as

those published by Forrest and co-workers [19] and the Smithsonian Institute [20]. In
addition to these IR spectral libraries, there is a major multi-volume library of spectra
published by Hummel and Scholl which, in addition to polymers, also includes a wide
range of additives used in rubbers, plastics and other polymer-based products [21].

The majority of rubber compounds contain additives such as CB and plasticisers

that can interfere with the IR data obtained if an attempt is made to analyse the
compound directly. It is also common for a rubber product to have been subjected to
a vulcanisation procedure, which renders it insoluble. For both reasons it is, therefore,
necessary to undertake some simple preparation work prior to analysis. For example, it
is possible to remove plasticisers from the sample matrix using solvent extraction (e.g.,
using acetone or methanol) and then to pyrolyse the extracted portion and record an
IR spectrum from the collected pyrolysis condensates. This operation eliminates any
interference from the CB, inorganic additives (e.g., ZnO) or fillers. For this approach
to be successful, a database of pyrolysate IR spectra must be available. It is a useful
feature of the database published by Forrest and co-workers [19] that a large number
of pyrolysis spectra of pre-extracted rubbers are included. An example of a pyrolysis
IR spectrum, in this case of a natural rubber (NR) sample, is shown in Figure 2.1.

96
88
80
Transmittance (%)

72
64
56
48
40
32
24
16
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm-1)

Figure 2.1 Pyrolysis IR spectrum of a NR sample. Reproduced with permission

from M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,
Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.7.
©2001, Smithers Rapra [22]

11
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Analysis of the solvent extract that was obtained in preparing the rubber sample for
the pyrolysis work by transmission FTIR spectroscopy could, with the assistance of the
IR databases mentioned above, identify the plasticiser or oil present in the compound.
IR spectroscopy can also be used in this way to identify specialist organic additives,
such as fire retardants; the inorganic types of fire retardants can be determined by IR
using a similar approach to that described below for inorganic fillers.

With regard to the identification of inorganic additives, although FTIR can be used
directly, usually employing an ATR accessory, to identify the inorganic fillers that are
present in a polymer [23], the CB in rubbers can cause problems by absorbing the
IR radiation. Hence, a better approach is to isolate the inorganic filler by ashing. An
FTIR analysis of the ash of a rubber sample obtained by heating to 550 °C for ≈60
min can identify the inorganic filler that is present. If it is suspected that a mixture
of inorganic fillers has been used, or that other inorganic additives are also present,
it is advisable to complement the FTIR data with elemental data on the ash, for
example, by using X-ray fluorescence spectroscopy (XRF) or inductively coupled
plasma (ICP) analysis.

If the raw, uncompounded polymer is to be investigated, it is possible to record an

IR spectrum from either a film cast from solution, by direct analysis using ATR, or
the pyrolysate approach. The transmission IR spectrum obtained by casting a film
of NR from a chloroform solution is shown in Figure 2.2.

The examples mentioned above refer to the common form of IR spectroscopy

which relies on absorption of the IR radiation. However, it is also possible to use
IR emission spectroscopy in the analysis of rubbers. This is essentially a qualitative
technique in which an IR beam causes vibrations in the polymer molecules and the
radiation emitted is analysed by a highly sensitive spectrometer. The advantage of the
technique is that it can be used on CB-filled vulcanisates, and it has been employed in
degradation studies. It is possible to qualitatively determine chemical changes, such
as carbonyl formation (e.g., in acid groups), weight loss and volatilisation (via loss
of emission intensity due to the ‘unzipping’ of the polymer), as well as the formation
of conjugated double bonds and specific polymer reactions. The technique has been
used to study surface oxidation after natural and accelerated heat ageing and the
results obtained provided a rate of degradation by studying the vibration of chemical
groups associated with the degradation. Although the data obtained by emission
IR spectroscopy is not quantitative due to interferences such as sample reflectivity
and re-absorbance in thicker samples, Pell and co-workers [24] have suggested an
approximate relationship between absorbance and emittance.

12
 Analytical Techniques

94

86
78
Transmittance (%)

70

62

54

46

38
16
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm-1)

Figure 2.2 Transmission IR spectrum of a NR sample. Reproduced with

permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds and
Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001,
p.7. ©2001, Smithers Rapra [22]

Another use for FTIR, this time in the near-infrared (NIR) region, has been reported
by researchers drawn from Chiang Mai University and Maejo University [25]. Those
researchers have assessed the capability of Fourier–Transform (FT)–NIR to determine
the mechanical properties of rubbers as an alternative to using conventional physical
tests (e.g., hardness, tensile tests). A range of commercially important rubbers [e.g.,
NR, styrene-butadiene rubber (SBR), NBR and ethylene propylene diene monomer
(EPDM) rubber], and their blends, were analysed by an FT–NIR instrument that was
fitted with an integration sphere working in diffused-reflectance mode. The spectra
that resulted were found to correlate with hardness and tensile properties, and the
workers concluded that FT–NIR could be used to determine these properties of
rubbers and rubber blends.

2.4.2 Infrared Microspectrometry

For over 30 years it has been possible to obtain systems in which an FTIR instrument
has been interfaced with a microscope. Initially optical microscopes, such as those
marketed by Olympus, were used to examine and target the sample to be analysed
and then an IR objective lens and a sensitive IR detector brought into line to record

13
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

an IR spectrum. Modern instruments are now integrated packages that use digital
imaging and have additional capabilities (e.g., the mapping of surfaces).

IR microspectrometry is an extremely useful tool for the following types of

investigative work:

• Identification of fibres.

• Characterisation of surface blooms and discolourations.

• Identification of contaminants and inclusions in samples.

• Analysis of any sample that is too small to enable a standard IR accessory/bench

combination to be used [e.g., extracts from microextractions, small drops of
fluid, particles of dust, debris and other small solid particles, fractions separated
by chromatographic analysis, and small amounts of inorganic residue from
thermogravimetric analysis (TGA)].

2.4.3 Raman Spectroscopy [26, 27]

Raman spectroscopy, available in both dispersive and FT forms, has a number of uses
and is often used in conjunction with the other vibrational spectroscopy technique, IR
spectroscopy, due to their complementary nature. The Raman effect occurs when a
sample is irradiated by monochromatic light from a laser, causing a small fraction of
radiation to exhibit shifted frequencies that correspond to the vibrational transitions
of the sample.

Some absorption bands that are relatively weak in an IR spectrum (e.g., those due
to sulfur–sulfur, carbon–carbon, and halogen–carbon bonds) are more intense in
Raman spectroscopy and this is one of its advantages. The reverse is the case for
substances such as silicate fillers, and this means that they are much less likely to
obscure important diagnostic bands as they are prone to in an IR spectrum. This
capability of Raman spectroscopy can aid the determination of additives in a rubber,
but its sensitivity to particular absorption bands can also make it a useful tool for
structural analysis [28–30].

A library of dispersive Raman spectra of polymers and plasticisers has been published
by the Smithsonian Institute [20] and a handbook published by Kuptsov and Zhizhin
[31] includes examples of FT Raman spectra of polymers.

14
 Analytical Techniques

2.4.4 Ultraviolet Light Spectroscopy [32]

These instruments can usually scan over both the UV and ultraviolet visible (UV-
Vis) portions of the electromagnetic spectrum and so are referred to as ‘UV-Vis
spectrometers’. Although not as universally used as IR spectroscopy, or as versatile,
UV-Vis spectroscopy still has a part to play in the analysis of polymers because it is
well suited for particular tasks. For example, it is often used in the qualitative and
quantitative analysis of additives that possess strong chromophores (e.g., antioxidants
with aromatic rings) and in analytical methods that employ a derivatisation step
to produce such species with this characteristic (e.g., determination of unreacted
peroxides in silicone rubber).

One of the principal limitations of the technique is that the analyte needs to be in
the form of a solution and the UV absorbance of the solvent must not interfere with
that of the analyte.

2.4.5 Nuclear Magnetic Resonance Spectroscopy [33–36]

Nuclear magnetic resonance (NMR) spectroscopy is a powerful tool for the

determination of polymer structure. It is at its most sensitive when used on analytes in
solution and so is very useful for the analysis of the raw polymers. The vast majority
of rubber compounds are, however, vulcanised. Hence, either an NMR analysis
has to be carried out on the fraction of the rubber matrix which can be extracted
using a strong solvent (e.g., chloroform) and, because rubbers are lightly crosslinked
structures, a sufficient amount of sample can usually be obtained to make this possible,
or a solid-state NMR (e.g., magic angle) instrument has to be used. The disadvantage
of the latter is its lack of sensitivity, which can make the analysis of blends in which
one polymer is present at a low level difficult, as well as the detection of cure site
monomers in rubbers such as halobutyl rubbers and EPDM.

Of the conventional NMR instruments, both proton and carbon-13 variants are
commonly used in rubber analysis.

2.4.6 Mass Spectroscopy Techniques [36]

Mass spectrometry (MS) is an extremely useful and powerful tool for identifying
chemical compounds and for elucidating their chemical structure. It has a number
of features and advantages, including:

• Only small amounts of sample are required.

15
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• The molar mass of a substance can be obtained directly by measuring the mass
of the molecular ion.

• Molecular structures can be derived by means of the molar masses and ion-
fragmentation patterns.

• Atomic compositions can be derived using accurate mass instruments.

• Mixtures of substances can be characterised using ‘soft’ ionisation methods in

conjunction with chromatography techniques.

With regard to the last example in the list above, the use of mass spectrometers [e.g.,
quadrupole or time-of-flight (ToF)] to enhance the information that can be provided
by techniques such gas chromatography (GC), liquid chromatography (LC) and ICP
is mentioned in Sections 2.5.2, 2.5.3, 2.5.6 and 2.4.9.

Due to the advances made in the field in recent years (e.g., development of powerful
FT instruments) and the increase in the use of ionisation methods for the desorption
of ions direct from the surface of a sample [e.g., matrix-assisted laser desorption/
ionisation (MALDI)], the technique has gained in importance, with its contribution
to structural analysis complementing the data provided by IR and NMR. Various
MS techniques have been applied to the analysis of rubbers and polymer additives,
either as extracts or on the sample surface by laser techniques, and these have been
reviewed by Lattimer and Harris [37].

As well as those examples where a mass spectrometer is used as an integral part of

an analytical system (e.g., ICP–MS) There are now a large number of MS-based
analytical systems available to the analyst:

• Electrospray ionisation MS
• Chemical ionisation MS
• Electron-impact MS
• Direct probe MS
• Field desorption–MS
• Fast atom bombardment–MS
• Tandem MS–MS
• MALDI–MS
• Laser desorption/laser photoionisation ToF MS
• Secondary-ion mass spectrometry (SIMS)

16
 Analytical Techniques

2.4.7 Atomic Absorption Spectrometry [38]

Atomic absorption spectrometry (AAS) has been around since 1955 and for most
of this time it has been the standard analytical tool employed by analysts for the
determination of trace amounts of metals. The elements in an aqueous solution are
converted to atomic vapour in a flame (e.g., oxygen-acetylene flame) and radiation at
the appropriate wavelength is passed through the vapour to excite the ground-state
atoms to the first excited electronic level. The amount of radiation that is absorbed
is directly related to the atom concentration.

AAS cannot ‘survey’ or ‘screen’ for a range of elements in a single operation, unlike
ICP-based techniques (Section 2.4.8), but it can determine a particular element with
little interference from other elements. Another limitation of AAS is that it is not as
sensitive as ICP, and it is these two main limitations which have restricted its use in
modern laboratories.

Variants of the AAS technique that are available commercially include:

• Graphite furnace AAS

• Vapour generation AAS

• Zeeman AAS

These variants provide advantages in important areas such as improved sensitivity

and enhancement of the elemental range.

2.4.8 Inductively Coupled Plasma–Atomic Emission Spectroscopy [39]

This technique can also be referred to as ICP–optical emission spectroscopy and has
a number of advantages over AAS, including:

• Greater sensitivity.

• It can survey a sample for a large number (e.g., >50) of elements in a single
operation.

As in the case of the AAS technique, initially, the rubber sample is digested in an
strong acid, or mixture of acids, to create a solution, which is then filtered to remove
particulates and insoluble matter. This operation can be labour-intensive and involve
a considerable amount of heating if a rubber has a good inherent resistance to acids,
but fortunately there are automated digestion systems available on the market.

17
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Different ICP–atomic emission spectroscopy (AES) instruments available include

those that operate in a sequential manner or which can scan a number of elements
simultaneously. As well as providing advantages relating to efficiency and low limits
of detection, these instruments can be used to obtain large amounts of elemental data
on unknown samples, which is particularly useful for reverse engineering and failure
diagnosis. In this mode, it is usual to operate the instrument semi-quantitatively and
then to repeat the analysis using a full calibration if specific elements need to be
quantified.

Applications of semi-quantitative work include the determination of a flame-retardant

system in a rubber where additives such as antimony trioxide, zinc borate, hydrated
alumina, or triphenyl phosphate may have been used, or the search for metals that may
have catalysed the degradation of a rubber (e.g., copper or manganese). Regarding
quantitative work, as well as the above examples, ICP–AES systems are used regularly
for the accurate quantification of substances to check for compliance with regulations
(e.g., food-contact regulations).

2.4.9 Inductively Coupled Plasma–Mass Spectrometry

ICP–MS uses the established ICP technique to break the sample into a stream of
positively charged ions that are subsequently analysed on the basis of their mass.
The ICP–MS systems use a compact quadruple mass analyser of the type employed
by bench-top GC–MS and LC–MS systems.

The technique has a number of attractive attributes, including:

• Large dynamic range (e.g., solutions containing 2% solids can be analysed).

• High sensitivity.

• Unambiguous spectra.

• Ability to directly measure different isotopes of the same element.

2.4.10 X-Ray Fluorescence Spectroscopy and X-Ray Diffraction [40]

Characteristic fluorescence X-rays are emitted from a sample when it has been
excited by bombardment with high-energy X-rays or gamma rays, and these can be
used to obtain elemental data of a sample. The standard technique usually requires a
relatively large amount of sample (≈0.5 g). In principle, the lightest element that can

18
 Analytical Techniques

be detected is beryllium (atomic number = 4) but, due to instrumental limitations and

the low X-ray yields for the light elements, it has not yet been possible to achieve this.
However, over the years the working range of XRF instruments has increased and
it is now possible for routine elemental scans to be carried out from at least sodium
(atomic number = 11) upwards.

The XRF instrument is usually operated in a semi-quantitative way and care must be
taken in the interpretation of the results because they can be influenced by the nature
of the matrix. In order to use the technique quantitatively, a range of standards of the
element of interest in the same matrix as is present in the sample has to be prepared.

XRF spectroscopy, in conjunction with IR spectroscopy, is used routinely to identify

inorganic fillers and inorganic pigments in rubber samples and their ashes. The semi-
quantitative elemental data from XRF spectroscopy indicates the types of elements
present in the sample, and the IR spectrum helps identify the type of compound(s).
X-ray diffraction is also mentioned in this section because it can be used in conjunction
with XRF and IR spectroscopy on these types of samples if it is important to identify
the particular crystalline form (and hence type) of certain inorganic fillers (e.g., silica
and silicates).

2.4.11 Energy-Dispersive X-Ray Spectroscopy [41]

Energy-dispersive X-ray spectroscopy (EDX) has a number of other names (e.g.,

energy dispersive spectroscopy and X-ray energy dispersive spectroscopy) and can
also be referred to as energy-dispersive X-ray analysis or energy-dispersive X-ray
microanalysis.

As in the case of XRF, characteristic fluorescence X-rays are emitted from a sample
when it is impacted using a high-energy beam of charged particles (e.g., electrons,
protons or X-rays). However, in EDX, the fluorescence radiation is analysed by
sorting the energies of the photons as opposed to separating the wavelengths of the
radiation as with XRF.

In the instruments that are commonly used to analyse rubber samples and other
polymeric samples, a scanning electron microscope is often employed as the source
of electrons. Using this source provides the added benefit of obtaining scanning
electron microscopy (SEM) images of the sample or feature within the sample (e.g.,
a contaminant) as well as elemental data.

The EDX technique is extremely useful for the characterisation of very small samples
(e.g, contaminants and inclusions removed from the surface and bulk of rubber

19
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

samples), the analysis of surface defects, such as stains and blemishes, and areas of
discolouration (e.g., blooms). In addition to obtaining data on these features, it is
common to obtain data on a freshly cut or undiscoloured portion of the sample in
order to obtain background or baseline values. In this way, it is possible to determine
if the defect feature is chemically unrelated to the rubber sample, or if it may have
originated from it in some way.

2.4.12 Chemiluminescence Spectroscopy

Chemiluminescence analysis is suitable for studying the early stages of the thermal
oxidation of rubbers. A weak emission of light formed by chemical reactions appears
during the oxidative degradation of hydrocarbons. The weak light may originate from
the deactivation of an excited carbonyl group formed during the oxidation, or from
the formation of peroxide radicals. The use of a sensitive photon detector enables
the chemiluminescence to be detected accurately and with good reproducibility. The
intensity of the light emitted by the polymer is proportional to the concentration of
the originating source. Thus, the degree of oxidation of the polymer can be assessed
and the effectiveness of antioxidants evaluated. From the decrease in luminescence
intensity it is possible to calculate the rate constant of the decay of the radicals in the
polymer undergoing degradation and the effective activation energy for the initiation
of oxidation.

The imaging chemiluminescence technique has proved to be a useful method to

depth-profile the oxidation of rubbers.

The advantages of chemiluminescence are the relative simplicity of the equipment

required, its high sensitivity and the short duration of the experiments. Tests can be
carried out either dynamically (at heating rates from 1 to 15 °C/min) or isothermally,
with an upper temperature limit of 250 °C.

2.5 Chromatographic Techniques [42]

2.5.1 Gas Chromatography [43]

GC can be used for the identification and quantification of a large range of low-
molecular weight (MW) substances and additives. Two important criteria that species
need to have in order to be amenable to its application are thermal stability and
relatively volatility. Standards of high purity also need to be available to enable the
identification of retention time (RT) and plotting of calibration curves. The mobile

20
 Analytical Techniques

phase is often helium and the modern form of the technique employs capillary columns
to achieve higher resolution than that possible with traditional packed columns. The
standard detector is the flame ionisation type and other detectors are available which
improve the specificity and detection limit of the technique (e.g., nitrogen phosphorus
detector).

GC can be used as a relatively quick and simple technique to identify and quantify
a number of analytes, including:

• Residual monomers and oligomers

• Solvents

• Monomeric antioxidants

• Antiozonants

• Monomeric plasticisers

• Organic flame retardants

GC can also be used to obtain polymer-type information by use of a pyrolysis

approach, although it is less powerful than GC–MS because of the absence of a
mass spectrometer. A pyrogram ‘fingerprint’ must be used and a comparison made
against a library of known pyrolysis standards recorded under the same conditions
(e.g., pyrolysis temperature and column temperature programme) as the ‘unknown’.

2.5.2 Gas Chromatography–Mass Spectrometry [44]

Although it is bound by the same sample requirements as GC (e.g., thermally stability),

this powerful, versatile technique can be used to obtain a significant amount of
compositional information on rubber samples. It is particularly useful in identifying
minor components of the sample, such as the breakdown products of the cure system
and antidegradants. GC–MS also plays a part in failure diagnosis by being able to
provide information on chemicals which may cause odours, and organic contaminants
that may have degraded the rubber matrix.

There are a variety of ways that the sample or fractions of the sample (e.g., an extract
or volatiles) can be analysed by GC–MS:

• Static headspace

• Dynamic headspace

21
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Solution injection

• Pyrolysis

With regard to the last example listed above, this approach can be used to investigate
the polymer fraction within a sample [45]. The GC–MS instrument is fitted with a
pyrolysis unit set at ≈600 °C and the pyrolysis products are ‘swept’ into the instrument
using an inert gas (e.g., helium). This method has the advantage over the pyrolysis
FTIR approach in that it has higher resolution and sensitivity and so has greater
capability in the identification of complex polymer blends.

One of the limitations of the standard mass spectrometer detectors (e.g., quadrupole
or ion trap) used on common GC–MS instruments is that they have a limited accuracy
when it comes to determining the actual mass of the ions in the mass spectrum of a
compound (e.g., ±0.5 mass units). Hence, compounds are not identified by the mass
of their molecular ions, but by obtaining matches on commercially available 70-eV
electron-impact ionisation mass spectral libraries. If no match is found, it is usually
only possible to comment on the possible generic class of compound from the overall
appearance of the mass spectrum and from knowledge of how chemical structures
within molecules fragment under 70 eV.

More sophisticated and expensive GC–MS systems that have high-resolution mass
spectrometers are commercially available. These mass spectrometers can provide the
value of the molecular ion to an accuracy that enables the empirical formula of the
organic compound to be obtained. More accurate masses of the fragment ions in a
mass spectrum are also obtained. In addition to GC–MS, high-resolution detectors
are also available for other techniques that use MS, including LC–MS (Section 2.5.5).
The range and capability of GC–MS instruments can also be enhanced by adding to
electron-impact ionisation the option of chemical ionisation.

2.5.3 Gas Chromatography–Gas chromatography–Time-of-flight Mass

Spectrometry

This technique has been commercially available for ≈20 years and can be referred
to as a three-dimensional (3D) chromatography technique. This comes about
because of the combination of two GC columns in the instrument (a main, primary
one and a shorter, secondary one), the mass spectrometer, and powerful software
that can be used to reconstruct chromatograms and assist in the deconvolution
of species peaks.

Despite its high resolution, GC–MS (Section 2.5.2) can encounter problems when
trying to resolve all of the components present in complex samples, such as rubber

22
 Analytical Techniques

extracts. Although improvements in resolution can be achieved (e.g., by using longer

columns) the results obtained are limited because only one type of column (i.e., polar
or non-polar) can be used at a time and so any improvements are mainly achieved by
influencing the RT due to volatility or polarity. A GC×GC–ToF–MS instrument has the
potential to overcome this limitation by interfacing two columns using a modulator
at the end of the first, non-polar column (30 m × 0.25 mm), which concentrates
the retained compounds over a short (1–5 s) period to focus them and send them
down the second, polar (2 m × 0.10 mm) column. The latter is situated in its own
compartment within the main oven.

This focusing operation is repeated throughout the length of the analytical run and,
typically, 1,500 separate modulations can be undertaken. The mass spectrometer
is located at the end of the second column and this has to scan the complete mass
spectrum range selected (e.g., 35 to 600 Da) at a very fast rate (e.g., 50 scans/s) as the
analysis of the second column is completed within the modulation period. A ToF mass
spectrometer is used in the instrument because it can scan faster than a quadrupole
mass spectrometer. The software present in the instrument also aids data analysis and
interpretation by possessing a powerful deconvolution programme.

The two GC columns allow separation by volatility on the first analytical column
and by polarity on the second column so that peaks with a similar (or identical) RT
on the first column can be separated by the second column. The practical benefit
of this scenario to the rubber analyst is that complex mixtures can be separated
more thoroughly. Hence, higher-quality mass spectra are obtained on the resolved
components, assisting in the identification of the components via mass spectral library
searches or by manual interpretation. Mass spectra are produced by these instruments
using electron-impact ionisation at 70 eV and so matches can be found using the same
commercial libraries that are used with conventional GC–MS systems.

Because of its ability to separate complex mixtures of substances, GC×GC–ToF–MS

is a particularly useful tool for the analysis of rubber extracts in reverse engineering
and food simulant samples for migration studies (Chapters 5 and 8).

To provide a simple illustration of the data obtained using GC×GC–ToF–MS, a

chromatogram of a test mixture, in this case a Grob Mix, is shown in Figure 2.3.
The composition of the Grob Mix, together with the relevant RTs of the first and
second column, is shown in Table 2.1.

23
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

3
2285
2n 2 1785
d
Co
lu RT
m 1285 lumn
n 1 Co
RT 1st
785

0 285

Figure 2.3 GC×GC–ToF–MS chromatogram of a 10-ppm solution of the Grob

Mix. Reproduced with permission from M.J. Forrest, S. Holding and D. Howells,
Polymer Testing, 2006, 25, 63. ©2006, Elsevier [46]

Table 2.1 First and second column retention times of substances in a Grob
mix
Retention (s) Retention time (s)
Decane 713.0 1.97
Octanal 796.9 2.35
Undecane 828.8 1.94
Nonanol 836.9 2.34
Dimethylphenol 840.9 2.83
Ethylhexanoic acid 848.9 2.34
Dimethylbenzamine 904.9 3.02
Methyldecanoate 1052.9 2.20
Methyldodecanoate 1220.9 2.18
Cyclohexylcyclohexamine 1144.9 2.34
Reproduced with permission from M. Forrest, S. Holding and D. Howells, Polymer Testing,
2006, 25, 63. ©2006, Elsevier [46]

The Grob Mix is obviously a relatively simple mixture but the chromatogram in
Figure 2.3 clearly illustrates the separation that is obtained by virtue of volatility on
the first, main column and by polarity on the second column.

24
 Analytical Techniques

2.5.4 High-performance Liquid Chromatography [47]

HPLC and LC–MS (Section 2.5.5) instruments are usually employed to identify and
quantify additives in rubber and plastic compounds that are either too involatile
or insufficiently thermally stable to be analysed by either GC or GC–MS. HPLC
is, therefore, the first-choice instrument for antioxidants that have high MW (e.g.,
Irganox® 1010) or are oligomeric (e.g., poly-2,2,4-trimethyl-1,2-dihydroquinoline).
Such antioxidants are often used in rubber products to minimise loses during curing
or in long-term end uses at high temperatures.

HPLC is also very useful for the detection and quantification of other relatively
high-MW additives, for example, those that fall into the plasticiser (e.g., diisononyl
phthalate) or process aid category (e.g., polyethylene glycol). The main limitation of
HPLC, however, is that the standard detector is UV and so the analyte needs to have
a chromophore to be detected. Hence, standard HPLC cannot be used to determine
so high-MW additives such as paraffin waxes.

As with the majority of analytical techniques, a number of modes of operation are

available relating to variables such as the column mobile-phase combination (e.g.,
reverse phase or normal phase), the make-up of the mobile phase during an analysis
(e.g., isocratic or gradient elution) and the type of detector (e.g., UV, fluorescence, and
mass spectrometer). The use of mass spectrometers as detectors for the technique are
now so common and important that this variant, LC–MS, with optional UV detection
in situ, is often the instrument of choice and is covered in Section 2.5.5.

2.5.5 High-performance Liquid Chromatography–Mass Spectrometry

[48–51]

As mentioned in Section 2.5.4, because of their advantages, these instruments have

mainly replaced HPLC instruments and are now found in the majority of commercial
laboratories. The LC section of LC–MS can be operated in a very similar way, with
similar choices of operating mode, to the less sophisticated HPLC technique, although
care has to be taken that complex sample media (e.g., salt solutions) do not pose a
threat of contamination to the mass spectrometer.

Substances are detected and identified as they elute from the LC column via both the
mass spectrometer (e.g., a quadrupole type) and the in-line UV detector. It is standard
practice to use both detection systems because the UV detector is more sensitive to
some chemical compounds and, in other cases, they are not ionised by the mass
spectrometer and so are not detected.

25
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

There are a number of choices when it comes to the operating mode of the MS section,
some options being:

• Atmospheric chemical ionisation – two variants:

• Negative-ion ionisation (good for compounds with labile hydrogens such as
alcohols).
•  ositive-ion ionisation (good for compounds that will accept a hydrogen ion
P
such as amines).
• Electrospray (good for ionic compounds).
• Atmospheric pressure photoionisation (good for aromatic hydrocarbons).

These MS options, which tend to be ‘soft techniques’ result in little fragmentation of

the molecular ion, and obviously extend the scope of the technique. However, one
drawback of the lack of standardisation is that is that there are no commercially
available LC–MS libraries (unlike GC–MS) and so laboratories have to generate their
own in-house libraries to identify unknown samples.

A variant on the LC–MS technique, which can be seen in a number of commercial

laboratories, is HPLC–MS/MS. This configuration enables a charged ion from the
initial fragmentation of a substance to be selected and then subjected to further
fragmentation to produce another mass spectrum. In this way, the technique can
improve the specificity available to the analyst as the same initial fragment ion from
different substances will not produce the same secondary fragmentation pattern.
The technique, therefore, improves the ability to differentiate between the different
structural isomers of a compound.

A more recent development in LC–MS that is now accessible to commercial

laboratories is high-resolution accurate mass (HRAM) LC–MS instruments that also
have the capability of ion mobility. The high-resolution mass spectrometer enables
rapid assignment of the molecular formula and structure of compounds having MW
up to 50,000 Da. The addition of ion mobility, which provides information on the
3D structure of a molecule, assists data interpretation by contributing an important
additional dimension in the confirmation of an assigned molecular structure. The
enhanced confidence in assignments that results from this detector combination
reduces the need for confirmatory experiments. The performance of these instruments
can also be improved by the upgrading of the LC component to one that has an
ultra-performance liquid chromatography (UPLC) status, with the acronym then
becoming UPLC–HRAM/MS.

26
 Analytical Techniques

2.5.6 Ion Chromatography

IC, which can also be referred to as ion-exchange chromatography, separates ions

and polar molecules based on their affinity to an ion-exchange resin present in a
column. Laboratory-scale instruments usually use chromatographic columns which
are designed to separate either cations or anions and the mobile phases that are
used are buffered aqueous solutions. The columns are typically resins (e.g., cellulose
beads) with covalent-bonded charged functional groups. The target analytes are
retained on the column for a time but are displaced and eluted by increasing the
concentration of a similarly charged species in the mobile phase (e.g., in cation IC,
a positively-charged analyte can be replaced by positively charged sodium ions).
Detection of the eluted ions is usually carried out using either a conductivity detector
or a UV-Vis detector.

Although not used extensively for the analysis of rubber, one of the applications for IC
is to provide accurate quantifications of the amount of sulfur or halogen species (e.g.,
chlorine) in the reaction mixture than that resulting from an oxygen flask combustion
experiment (Section 2.2). It is more accurate and sensitive for these determinations
than using a titration-based approach and so it is the preferred route, particularly if
the levels of analyte are expected to be low.

2.5.7 Supercritical Fluid Chromatography [52]

Supercritical fluid chromatography (SFC) uses the unusual properties of supercritical

fluids. It offers the dual benefits of the solubility behaviour of liquids and the diffusion
and viscosity behaviour of gases. Using mobile phases such as carbon dioxide, it
provides unprecedented versatility in obtaining high-resolution separation of mixtures
and substances that other chromatographic techniques (e.g., GC and LC) struggle
with (e.g., high-MW oligomers and additives).

SFC can be used in combination with the majority of the detectors used in GC and
LC in a preparative way or to isolate samples for subsequent offline analysis using
techniques such as IR microscopy. Another advantage of the technique is that the
loss of the mobile phase due to decompression once it exits the instrument makes
deposition onto a substrate or into a sample vial relatively easy and removes this
source of potential interference. It is also possible to interface SFC with a number of
spectroscopy techniques (e.g., IR and MS).

27
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

2.5.8 Gel-permeation Chromatography/Size Exclusion Chromatography

[53, 54]

Used in its conventional form, with mixed pore-size column packings, gel-permeation
chromatography (GPC), or size-exclusion chromatography as it can also be referred
to, is used extensively to characterise the MW and molecular distribution of polymers.
By using column packings that have small pore sizes (e.g., 50 or 100 Å) it is also
able to separate oligomers, although it does not have the same resolving capability
in this regard as SFC. Also, by employing large-bore columns, it can be used in a
preparative way to accumulate samples for subsequent analysis by GC–MS, LC–MS
or IR microscopy.

For GPC to be successful, the sample must be readily soluble in an organic solvent.
Hence, by definition, it cannot be carried out on the polymer phase in a rubber
vulcanisate and even unvulcanised fully compounded rubber samples can present
problems due to filler–rubber interactions. Therefore, it tends to be restricted to the
analysis of raw rubbery polymers or thermoplastic rubbers and in this, using solvents
such as toluene and tetrahydrofuran, it is usually very successful. However, given
that a number of rubbers are free radical-polymerised (e.g., NBR) or have very high
MW (e.g., NR), there can be a problem with the insoluble fractions (i.e., ‘gel’) that
result from physical entanglements and chemical interactions.

It is possible to use GPC to obtain information on high-MW additives present in

vulcanised rubbers by the analysis of solvent extracts. This approach is particularly
useful if it is suspected that oligomeric or polymeric additives may have been used
in the compound. Figure 2.4 shows how high-resolution GPC (using 100 A, 10-µm
columns) has been used to separate the polymeric and other additives extracted from
a NBR sample. Use of an IR detector or LC-Transform® system, as mentioned below,
can then assist in identifying these substances.

During the last 30 years, the range of detectors available for use with GPC has
increased and a ‘state-of-the-art’ system now often possesses the following three
detectors:

• Refractive index (RI)

• Light scattering

• Viscosity

28
 Analytical Techniques

0
10
Low-MW
additives

20
30
40
%
50
Polymeric polyester plasticiser
60
70
80
90
100

Figure 2.4 Separation of polymeric and monomeric additives extracted from a

NBR using high-resolution GPC. Reproduced with permission from M.J. Forrest
in Rubber Analysis – Polymers, Compounds and Products, Rapra Review Report
No.139, Smithers Rapra, Shawbury, UK, 2001, p.10. ©2001, Smithers Rapra [55]

Another detection system that is becoming available for GPC is IR, and some
commercially available systems have these in conjunction with a RI detector. One
drawback of IR as an in-line detection system for GPC is that the IR absorption of
the mobile phase will reduce the amount of the spectrum that is available for use.

29
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Before the advent of in-line IR detectors for GPC, the development of the Lab
Connections LC-Transform® interface offered the capability of obtaining IR spectra
on sections or fractions of a GPC chromatogram. The LC-Transform® consisted
of two sampling accessories (modules) and operating software. The first module
collected the species present in the GPC chromatogram as they eluted from the
system. They were deposited onto a revolving germanium disc with the mobile
phase being volatilised by a combination of heat, vacuum and ultrasonic agitation.
The other module enabled the prepared disc to be placed into a standard IR bench
for analysis. In the case of some rubber systems, by using the correct choice of
separating column, polymer blends could be separated and identified, as well as
plasticisers and other high-MW additives.

2.5.9 Thin-layer Chromatography and Paper Chromatography [56]

Thin-layer chromatography (TLC) and paper chromatography are relatively simple

techniques that have been in existence for many years now and for which little new
work has been published since the 1970s; both are rarely used in modern analytical
laboratories.

Historically, before the advent of modern HPLC-based instruments, TLC and paper
chromatography were used for both the separation and identification of additives
such as amine and phenolic antioxidants, and UV stabilisers, and they can still fulfil
that function today if required. A number of approaches are possible using different
developing media, solvents and spraying agents. The standard method of detection
is the use of a UV light source.

By the use of plates that have thick layers of the stationary phase, TLC can be used in a
preparative way for the isolation of sample fractions for subsequent chromatographic
(e.g., GC–MS) or spectroscopic (e.g., IR) analysis.

2.6 Thermal Techniques [57, 58]

2.6.1 Differential Scanning Calorimetry [59]

Differential scanning calorimetry (DSC) can be used for a number of purposes in the
analysis of rubbers. For example, it is possible to use it to provide quality-control
fingerprint thermograms from which deviations in formulation can be ascertained.
These will contain both endotherms from the volatilisation of ingredients such as
plasticisers, together with exotherms from the curing of the rubber and the degradation

30
 Analytical Techniques

of the sample. Analytical work can be performed in both a non-oxidising atmosphere

(e.g., nitrogen), which is the most commonly used environment and, if appropriate,
such as for investigations involving oxidation stability, an oxidising atmosphere
(e.g., air).

A typical two-stage DSC programme used in a nitrogen atmosphere to investigate

the composition and properties of a rubber sample is shown below:

1. 2 min at −100 °C.

2. Heat from −100 to +250 °C at 20 °C/min (creates an ‘as-received’ thermogram).

3. Hold for 2 min at +250 °C.

4. Cool from +250 to −100 °C at 20 °C/min.

5. Hold for 2 min at −100 °C .

6. Heat from −100 to +250 °C at 20 °C/min (creates ‘reheat’ thermogram).

With rubber samples, the information that is obtained during the initial heating stage
(i.e., stage 2 as shown above) is usually the most useful. During this stage, a number
of events may be present in the thermogram depending upon the history of the sample
and the objective of the analysis, including:

• Volatilisation of absorbed substances (e.g., water or solvents)*.

• Change in specific heat due to the glass transition temperature (Tg) of the
polymer(s).

• Melting of crystalline ingredients*.

• Loss of plasticiser and other ingredients*.

• Curing exotherm (if the rubber is compounded but unvulcanised).

• Non-oxidising degradation exotherms.

*endothermic changes

The reheat thermogram (stage 6 as shown above) can sometimes provide evidence
of additional compositional changes and reactions, such as further non-oxidising
degradation.

An indication of the range of events which can be observed by analysing an uncured

compounded rubber by DSC using a nitrogen atmosphere, a temperature ramp of
20 °C/min and a temperature range −100 to around +700 °C, is provided below:

31
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• −100 to 0 °C – Tg.

• 0 to 150 °C – melting of additives and loss of plasticiser/oil.

• 150 to 250 °C – curing.

• >250 °C – degradation and antioxidant function.

Using appropriate standard materials it is possible to employ DSC to quantify the level
of certain additives (e.g., peroxide curatives in uncured rubbers) in a sample. DSC
can also be used in thermal stability studies of a rubber compound, and to investigate
the effectiveness of antidegradants and, to an extent, fire retardants.

DSC can also be used to profile the degree of cure through thick rubber articles, to
study cure kinetics, and the compatibility of polymers in a blend. Two examples of the
latter studies are those of a blend of EPDM and NR, published by Abou-Helal and El-
Sabbagh [60], and that of a blend of polypropylene (PP) and EPDM, a thermoplastic
rubber, published by Qixia and co-workers [61].

Crosslink density can influence the Tg of rubber samples. Cook and co-workers
[62], using unmodified rubbers (i.e., compounds without plasticisers or fillers),
demonstrated that a linear relationship exists between crosslink density and Tg for
three types of rubber: NR, polybutadiene (BR) rubber and SBR.

Thermoplastic rubbers, such as those produced by combining EPDM with PP, have
a semi-crystalline component (i.e., the PP) and so can be identified and characterised
(e.g., determining the proportions of PP and EPDM) by DSC. To assist with this,
it is important to have reference spectra that have been recorded under standard
conditions (e.g., those in stages 1–6 shown above) and an example of such a library
has been published by Price [63].

2.6.1.1 Modulated Differential Scanning Calorimetry [64]

The advent of modulated temperature DSC (also described as alternating or oscillating

DSC) over 15 years ago offered advantages to the polymer analyst in that its ability
to separate the heat flow into its reversing (heat capacity-related) and non-reversing
(kinetic) events improved the quality of the data that could be obtained by DSC and,
hence, increased its usefulness [65, 66].

The major reversing events that occur in plastics and rubbers include the Tg in both
cases, and the melting of the crystalline region in the case of semi-crystalline plastics.
There are a number of non-reversing events (including re-crystallisation processes)
occurring during melting, and the enthalpic relaxation that occurs in the region of the

32
 Analytical Techniques

Tg for amorphous polymers that have undergone physical ageing. In addition, with
crystalline plastics, the situation is complicated by the fact that the heating process
can alter the crystalline structure of the material, with imperfect crystals melting
and becoming absorbed into larger crystals prior to these melting. Conventional
DSC cannot separate these two events because they often happen simultaneously,
but modulated DSC can resolve the reversing heat flow of crystal melting and the
non-reversing heat flow of crystallisation. Modulated DSC is, therefore, capable of
providing more accurate data on the latent heat of fusion for the crystal melting
event. This is important because the study of the crystallinity of polymers is one of
the principal uses of DSC.

Other advantages that have been claimed for the use of modulated DSC in the analysis
of plastics and rubbers include:

• Improvements in the accuracy of Tg measurements in plastics due to the greater

separation of the Tg (reversing) and enthalpic relaxation event (non-reversing)
which occurs just after the Tg [67].
• The inherent greater sensitivity of the technique makes it easier to detect the Tg
in highly filled/reinforced polymers, or highly crystalline plastics, where these
features reduce the change in the heat capacity that occurs as the materials passes
through this transition. HyperDSC™ (Section 2.6.1.2) also offers this advantage
over conventional DSC.
• It can be useful in the analysis of certain polymer blends. For example, in the
case of a polycarbonate (PC)/polybutylene terephthalate (PBT) blend, the Tg of
PC may overlap with the re-crystallisation of parts of the PBT phase.
• It provides, along with power-compensation DSC, a means of directly measuring
the heat capacity of a material. It has been demonstrated that the method can
also be used to produce thermal conductivity data that agree well with values
established by more conventional techniques. This and other applications are
covered by a TA Instruments guide to modulated DSC [66].
• Providing that the calibration has taken into account the heat loss through the
purge gas surrounding the sample, the thermal conductivity of polymers can also
be measured in the range from 0.1 to 1.5 W/°C.
• Its ability to generate an instantaneous heating rate during quasi-isothermal
experiments allows cure measurements on rubbers to be made which are not
possible by conventional DSC. This can assist in obtaining a greater understanding
of the cure mechanism.
Hourston and Song [68] have proposed a modulated DSC method for the measurement
of the weight fraction of the interface in a wide range of rubber–rubber blends. A

33
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

quantitative analysis using the differential of the heat capacity versus the temperature
signal from the modulated temperature DSC allowed the weight fraction of the
interface to be calculated. The data obtained showed how essentially miscible the
different rubber types were, as well as revealing the influence of properties such as
stereo regularity (with BR blends) and the proportions of comonomers (with SBR
blends) on the extent of mixing.

2.6.1.2 HyperDSC™

This variant on DSC involves recording the heat flow cure of a sample as it is
heated very rapidly (e.g., at 500 °C/min), as opposed to a standard 20 °C/min
heating ramp. This technique can usually be used only with power-compensation
systems as an analysis chamber having a low thermal mass is essential; and the
ability to record data very quickly is another pre-requisite. Several publications
provide information on the use and application of HyperDSC™ [69, 70].

The advantages that this technique offers in the analysis of polymers come from
its twin attributes of very low short analysis times and increased accuracy when in
comes to measuring crystallinity. In the former case, the instrument has been used
effectively on the production lines of semi-crystalline materials to record data, such
as Tg and melting temperature (Tm), almost as soon as they are produced; a full
characterisation is possible within 60 s. This means that any quality-control problems
can be identified very quickly and rectified before too much production has been lost.
Although the samples are heated very rapidly, it has been demonstrated that this
results in no significant increase in Tg. The data obtained from these HyperDSC™
analyses can, therefore, be compared with library standards that have been run using
more traditional conditions.

With respect to the latter case, many semi-crystalline polymers exist in several
potentially unstable states. A fast heating rate enables data to be recorded on them
before they change due, for example, to microcrystalline regions melting and then
re-crystallising during the analysis. If this type of effect occurs under a slow heating
rate, it means that the extent of crystallinity of the polymer can change during an
experiment, and so the data obtained are not totally representative of the original
material.

The two other advantages of HyperDSC™ are:

1. The increased scan rate provides much higher sensitivity due to the resulting
higher heat flow, and this enables the analyst to identify transitions which would
not be apparent using standard conditions.

34
 Analytical Techniques

2. The ability to have fast heating and cooling rates means that it is possible to
closely mimic the conditions that some samples experience in their manufacture
and production.

2.6.2 Dynamic Mechanical Analysis [71]

Dynamic mechanical analysis (DMA) is a very useful technique for rubbers. It is

able to generate modulus versus temperature data under a range of conditions (e.g.,
deformation modes and frequencies) and, as a result of this, provide important
information on a sample, for example, dynamic mechanical properties, Tg, thermal
stability, and cure state.

The effects of temperature on various types of moduli can be obtained over a wide
temperature range (typically −150 to +200 °C). A typical set of conditions that can
be used to analyse rubber samples by DMA is shown below:

• Deformation mode: double cantilever

• Sample dimensions: 2 × 10 × 25 mm

• Heating rate: 3 °C/min

• Frequency of deformation: 1 Hz

• Start temperature: −150 °C

• Finish temperature: 250 °C

• Amplitude of deformation: 50 µm

The sensitivity of DMA to Tg (≈1,000-times greater than DSC, which measures changes
in heat capacity rather than modulus) is very useful in the analysis of unknown
samples. For example, it can be difficult to detect the presence of 5–10% of BR in SBR
by IR, which would only provide the value of the total butadiene content. However,
the large difference in the Tg values of the two polymers makes the presence of BR
in such a matrix much easier to detect and, with appropriate standards, quantify.

2.6.3 Microthermal Analysis

Microthermal analysis is a characterisation technique that can be regarded as

combining the thermal analysis principles of DSC with the high spatial resolution of
scanning probe microscopy (SPM). It is possible to create a microthermal analyser
by modifying an atomic force microscope by replacing its tip with an ultraminiature-

35
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

resistive heater that also serves as a temperature sensor. Using this device, it is possible
to obtain localised thermal analysis data on the surface of polymer samples, and this
feature enabled the technique to be launched commercially in 1998 [72].

An example of the commercial instruments that have been produced is the one
marketed by TA Instruments [73, 74]. This device is useful for the analysis of polymers
in certain applications because it can obtain DSC and thermal mechanical analysis
(TMA) data on very small (i.e., micron-sized) areas. Examples of the applications
that the instrument can be used for are presented below.

2.6.3.1 Determination of the Homogeneity of Polymer Blends

The surface of a rubber product can be mapped and occurrence of the events which
are characteristic of individual polymers can be obtained. The relative intensity of
these will indicate whether some areas of the sample are richer in one polymer than
others. For example, in a thermoplastic rubber gasket produced from a mixture of
EPDM rubber and PP, the Tg of the EPDM rubber (TMA trace) and the Tm (DSC
trace) of the PP could be focused upon.

2.6.3.2 Studying Surfaces

Examples of possible applications include rubber products that are modified by

processes such as corona discharge or are sterilised by radiation. These processes
change the physical properties (e.g., modulus) of the surface layer of the polymer due
to crosslinking and/or oxidation reactions. The microthermal analyser can be used
to determine how uniform across the surface these changes are.

2.6.3.3 Characterisation of Localised Thermal History

Although this capability will be more useful for studying plastics, where localised
stresses can become ‘frozen’ into the product if they are cooled too quickly during
manufacture, it could have some application for thermoplastic rubbers, and for
determining if certain areas of a rubber product are undercured due to cold spots in
a mould. In the first case, a microthermal analyser would be able to map the heat
capacity of the thermoplastic rubber and see if frozen-in irregularities in the structure
can be detected (e.g., by a reduction in crystallinity in certain regions). In the second
case, DSC could detect a residual curing agent via an exotherm, and TMA could
detect differences in the modulus.

36
 Analytical Techniques

A thorough review of the techniques and applications of microthermal analysis has

been published by Pollock and Hammiche [75].

2.6.4 Thermal Mechanical Analysis

TMA is a method for measuring linear or volumetric changes in polymers as a

function of temperature, time or force. It is often used in conjunction with DSC to
investigate the structure and properties of polymers. This combination is often found
in commercial microthermal analysers (Section 2.6.3). While DSC is concerned with
the energetics of physical and chemical changes, TMA measures the dimensional
effects associated with these changes, and so it can be used to obtain the coefficient
of thermal expansion for rubber samples.

TMA can also measure properties such as Tg and Tm, and determine the modulus of
rubbers, and so is similar to DMA in this regard.

2.6.5 Thermogravimetric Analysis [9]

TGA is extremely useful for the analysis of compounded rubber samples because it can
provide an extensive amount of information relating to their bulk composition and
other important characteristics (e.g., the absorption of a fluid in service). It is possible
to obtain accurate quantification of the principal ingredients such as plasticisers,
polymers, CB and inorganic species. These data can be obtained on small (e.g., 10 mg)
samples in a relatively short time (≈45 min). By the use of the temperature maxima at
which weight loss events occur, it is also possible to use the technique to obtain some
qualitative assignments for the bulk ingredients, particularly the polymer and the CB,
and to identify the presence of substances that have reproducible and characteristic
thermal-breakdown patterns (e.g., CaCO3).

A typical TGA programme for the analysis of a rubber would be:

1. Heat from 40 to 550 °C in a nitrogen atmosphere at 20 °C/min

2. Hold at 550 °C for 10 min

3. Reduce the temperature to 300 at 20 °C/min

4. Change the atmosphere to air

5. Heat from 300 to 900 °C in an air atmosphere at 20 °C/min

37
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

For rubbers that are very thermally stable (e.g., silicone rubbers), it is necessary to
modify the programme stated above to ensure that all of the organic material is
pyrolysed from the sample before the temperature is reduced to 300 °C. This is done
by increasing the final temperature of step 1 to ≈700 °C and holding the sample in
nitrogen at this temperature for a longer period of time (e.g., 20 min).

It is usually common to plot the weight loss derivative in conjunction with the weight
loss. The derivative can be used to detect the presence of a polymer blend of two or
more components and to indicate if the sample contains more than one type of CB.

Other uses for TGA include the production of compositional fingerprints for quality-
control purposes, and the investigation of thermal stability, including the effects that
additives such as flame retardants and antioxidants have upon it.

Modern TGA instruments have the ability to carry out high-resolution TGA
experiments (Section 2.6.5.1) and modulated TGA experiments (Section 2.6.5.2). It
is also possible to interface modern TGA instruments directly to gas analysers (e.g.,
FTIR instruments or mass spectrometers) to obtain both quantitative and qualitative
data in one analytical step. These combinations can be referred to as ‘hyphenated
TGA’ or ‘evolved gas analysis TGA’ and add a significant degree of capability to the
technique (Section 2.6.5.3).

2.6.5.1 High-resolution Thermal Mechanical Analysis

High-resolution TGA was developed to enhance the resolution of multiple components

in complex systems. It, therefore, offers particular potential advantages in the analysis
of complex rubber formulations. The principle of its operation is that rather than
having a constant heating rate (e.g., 20 °C/min) throughout the analytical run, the TGA
furnace is actively controlled by the software and the rate of heating is determined
by the rate that the sample loses weight. A review of the use of high-resolution
TGA for the analysis of polymer samples has been published by Forrest [76]. This
review demonstrates that, in practice, high resolution offers distinct advantages in
some applications (e.g., quantification of the plasticiser and polymer components in
rubbers), but can cause problems with others (e.g., quantification of two or more
rubbers in a blend). For example, Forrest found that NR gave one weight loss in a
conventional TGA experiment, but four in a high-resolution operation [76].

2.6.5.2 Modulated Thermal Mechanical Analysis

Modulated TGA was also intended to improve the accuracy of the data resulting

38
 Analytical Techniques

from TGA experiments. It was developed by TA Instruments [77, 78] and operates
using a similar principle to that of modulated DSC (Section 2.6.1.1), with a sinusoidal
modulation overlaid over the linear heating rate so that the average sample temperature
changes continuously with time in a non-linear way. Gamlin and co-workers [79]
used a combination of high-resolution TGA and modulated TGA to study the effects
of the ethylene content and maleated EPDM content on the thermal stability and
degradation kinetics of EPDM rubbers. A comparison was made between the data
generated from isothermal and non-isothermal TGA experiments, and other values
reported in the literature.

2.6.5.3 Thermal Mechanical Analysis with Evolved Gas Analysis

Combining TGA with an instrument capable of analysing the evolved gaseous products
that are emitted from it during an analysis extends and enhances the usefulness and
diagnostic power of the technique. Two popular choices of such an instrument are an
FTIR or a mass spectrometer, and these are directly coupled to the TGA by the use
of a heated interface. The nature of the interface varies according to the instrument
being employed. If it is a mass spectrometer, it could be a fused silica capillary tube
heated to ≈200 °C to prevent condensation of the gaseous products and only a
small quantity (e.g., 1%) of the volatiles will be diverted into the instrument. If the
instrument is an FTIR, a less sensitive technique, all of the gaseous products in the
purge gas (e.g., nitrogen) are transferred through a heated interface line (e.g., glass-
coated steel) into a heated gas cell.

In addition to TGA–IR and TGA–MS, other hyphenated TGA techniques that are
available to the analyst include TGA–GC, TGA–GC–IR and TGA–GC–MS.

TGA with evolved gas analysis can have a number of applications in the analysis of
rubbers, including:

• Thermal degradation processes

• Studies of vulcanisation

• Identification of polymer type

• Identification of additives (e.g., cure system ingredients)

• Analysis of substances (e.g., oils) absorbed in service

39
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

2.6.6 Dielectric Analysis

Dielectric analysis (DEA) has proved to be very useful over the years. It measures
the response of a polymer to an applied alternating voltage signal. It, therefore,
measures two fundamental electrical characteristics of a polymer – capacitance and
conductance – as a function of time, temperature and frequency. A polymer sample
is placed in contact with a sensor (or electronic array for samples in the form of
pellets or powders) and the oscillating voltage signal applied at frequencies between
0.001 and 100,000 Hz. Capacitance and conductance can be related to changes in
the molecular state of a polymer:

• Capacitance: the ability of a polymer to store an electrical change – dominates

at temperatures below Tg or Tm.

• Conductance: the ability of a polymer to transfer an electrical charge – dominates

at temperatures above Tg or Tm.

The actual polymeric properties monitored by DEA are:

• E' (permittivity) – a measure of the degree of alignment of the molecular dipoles

to the applied electrical field (analogous to the elastic modulus in DMA).

• E'' (loss factor) – represents the energy required to align the dipoles (analogous
to the viscous modulus in DMA).

Tan δ in DEA is the ratio of these two properties and the technique can be used for
a similar range of applications as DSC and DMA, but it has a number of practical
advantages over those two techniques:

• Data are not affected by evolution of volatile species during such events as curing
reactions.

• Data can be obtained on samples in a wide range of physical forms (e.g., solid,
liquid, pellets or powder).

• It is more sensitive than DSC and so it can be used to obtain superior data on
samples where there is only a small polymer fraction (e.g., highly extended or
filled polymer products).

• Data can be obtained by probe samplers situated away from the instrument, and
so it can be used for ‘in-line’ monitoring of events such as curing reactions.

There are a number of application notes on the use of DEA to characterise polymers
available from instrument suppliers [80].

40
 Analytical Techniques

2.6.7 Thermal Conductivity Analysis

Accurate thermal conductivity data are important for assessing the suitability of a
particular polymer for a given application. Such data are also being used extensively
in the application of computer modelling packages to predict the flow of plastics and
rubber compounds during processing operations.

The ability to use modulated DSC to determine such data has been mentioned in
Section 2.6.1.1. More traditional methods involve the use of the standard guarded
hot plate device that is described in the American Society for Testing of Materials
(ASTM) standard, ASTM F-433. In addition to this, Lobo and Cohen [81] described
an in-line method that can be applied to polymer melts. Also, Oehmke and Wiegmann
[82] described an apparatus that can carry out thermal conductivity measurements
as a function of temperature as well as pressure.

2.7 Microscopy Techniques [83, 84]

High-magnification (e.g., 150×) optical microscopy can be used to examine the

dispersion of fillers, such as CB, within the rubber matrix. The success of this operation
is usually dependent on the generation of high-quality microtome sections of the sample.
The degree of dispersion can be subjectively compared with photographs of standard
examples (e.g., a Cabot Dispersion Classification Chart for CB).

Modern digital microscopes and their associated software have greatly enhanced the
capability of the technique, including the measurement of features of interest and
the superimposing of multiple images that allow 3D pictures to be created of items
such as contaminants and defect structures. Digital imaging also allows the easy and
rapid transfer of images from the microscopy to online computers for use in reports
and presentations.

SEM has been mentioned in Section 2.3.11 in relation to the elemental analysis of
surfaces using EDX. It can also be used in its own right, where its high-magnification
powers can be employed to determine the particle size of fillers and in the detailed
examination of fracture surfaces. It can also be used to investigate the morphology
of rubber compounds, particularly filled rubber compounds [85] and rubber blends
[86]. Transmission electron microscopy can also be used to investigate the phase
morphology of polymer blends and the dispersion of fillers.

The use of IR microspectrometry to generate IR data on a variety of small samples

(e.g., polymers, inclusions, fibres and contaminants) has been covered in Section 2.4.2.

41
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Atomic force microscopy [also called scanning force microscopy (SFM)] involves
use of a very high-resolution scanning probe microscope capable of resolution of the
order of fractions of a nanometre, which is >1,000-times better than the detection
limit of an optical microscope. A review of the use of SFM to visualise and study the
morphology of polymers has been published by Vancso and Schonherr [87].

There are is wide range of instrumental techniques and methods that can be regarded
as being based on SPM. One of these, microthermal analysis using a modified atomic
force microscope, has been mentioned in Section 2.6.3. The use of SPM to characterise
a variety of polymeric materials has been reviewed by Tsukruk [88].

2.8 Rheological Instruments

This class of instrument can be used in a number of ways to characterise rubber

samples. One of the most important, which is essential for any quality-control
function within a rubber manufacturing environment, is the determination of cure
characteristics (e.g., scorch time, cure rate and cure time). The use of these instruments
for this type of application is covered within Chapter 5 – curing and cure state studies.

These types of instrument can also be used to obtain other important information
on rubber samples, such as those related to their viscosity, viscoelastic properties as
well as the interaction between fillers and polymers.

Three of the principal types of instrument that are commercially available include
the moving die rheometer, the rubber process analyser, and Mooney viscometer. A
publication produced by one of the manufacturers of these types of instruments [89]
provides a convenient listing of their capabilities and applications, including:

• Mooney viscosity, Mooney stress relaxation and Mooney scorch determinations.

• Determination of isothermal curing characteristics.

• Strain sweep measurements for filler loading, filler type, filler dispersion and
rubber–filler interaction information.

• Monitoring of the breakdown of structure within CB during processing (e.g.,

milling).

• Studying stress–strain response at high strain to characterise filler content and

structure, as well as polymer architecture.

• Monitoring simultaneous curing and blowing agent action during the production
of rubber foams.

42
 Analytical Techniques

2.9 Miscellaneous Analytical Techniques

There are a number of other analytical techniques that can be used to characterise
rubber samples. These include specific techniques applied to the analysis of certain
additives (e.g., CB), rubber systems (e.g., latex) and to the rubber itself (e.g., viscosity
measurement for determining MW). The application and information obtained by
such techniques are covered in subsequent sections of this book.

Surface analytical techniques such as SIMS [90, 91], laser ionisation mass analysis
[92, 93] and X-ray photoelectron spectroscopy [94–96] can be used for failure
diagnosis associated with the poor bonding of rubber products to metal substrates
and investigation of surface contaminates.

References

1. T.R. Crompton in Characterisation of Polymers, Volume 1 (2008) and

Volume 2 (2009), Smithers Rapra, Shawbury, UK.

2. N.P. Cheremisinoff in Polymer Characterisation, University Press of

Mississippi, Jackson, MS, USA, 1996.

3. B.J. Hunt and M.I. James in Polymer Characterisation, Springer Science &
Business Media, Berlin, Germany, 2012.

4. T.R. Crompton in Practical Polymer Analysis, Springer Science & Business

Media, Berlin, Germany, 2012.

5. T.R Crompton in Polymer Reference Book, iSmithers, Smithers Rapra

Technology Ltd, Shawbury, UK, 2006.

6. T.R. Crompton in Determination of Additives in Polymers and Rubbers,

Smithers Rapra Technology Ltd, Shawbury, UK, 2007.

7. J.C.J. Bart in Additives in Polymers, John Wiley & Sons, New York, NY,
USA, 2005.

8. P.P. De, N.R. Choudhury and N.K. Dutta in Thermal Analysis of Rubbers
and Rubbery Materials, Smithers Rapra Technology Ltd, Shawbury, UK,
2010.

43
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

9. M.J. Forrest in Application to Thermoplastics and Rubbers, Principles and

Applications of Thermal Analysis, Ed., P. Gabbott, Blackwell Publishing,
Oxford, UK, 2008, Chapter 6.

10. J.L. Koenig in Infrared and Raman Spectroscopy of Polymers, Rapra Review
Report No.134, Smithers Rapra Technology Ltd, Shawbury, UK, 2001.

11. Q. Guo in Polymer Morphology: Principles, Characterisation and Processing,

John Wiley & Sons, New York, NY, USA, 2016.

12. ISO 7269:1995 – Rubber – Determination of free sulfur.

13. ISO 15672:2000 – Rubber and rubber additives – Determination of total

nitrogen content using an automatic analyser.

14. ISO 15671:2000 – Rubber and rubber additives – Determination of total

sulfur content using an automatic analyser.

15. V.M. Litvinov and P.P. De in Spectroscopy of Rubbers and Rubbery Materials,
Rapra Technology Ltd, Shawbury, UK, 2002.

16. J.L. Koenig in Spectroscopy of Polymers, 2nd Edition, Elsevier, Amsterdam,

The Netherlands, 1999.

17. W. Klopffer in Introduction to Polymer Spectroscopy, Springer Science &

Business Media, Berlin, Germany, 2012.

18. N. Everall, P.R. Griffiths and J.M. Chalmers in Vibrational Spectroscopy of

Polymers: Principals and Practice, John Wiley & Sons, New York, NY, USA,
2007.

19. M.J. Forrest, Y. Davies and J. Davies in The Rapra Collection of Infrared
Spectra of Rubbers, Plastics and Thermoplastic Elastomers, 3rd Edition,
Rapra Technology Ltd, Shawbury, UK, 2007.

20. M. Jefcoat, H. Tubb and O. Madden in Proceedings of the 243rd National

ACS Meeting, 25–29th March, San Diego, CA, USA and the Smithsonian
Museum Conservation Institute 2012, 53, 1, 735.

21. D.O. Hummel and F.K Scholl in Infrared Analysis of Polymers, Resins
and Additives: An Atlas, Volumes 1–3, 2nd Edition, Carl Hanser, Munich,
Germany, 1979–1983.

44
 Analytical Techniques

22. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.7.

23. S. Elliott, British Plastics and Rubber, 2013, September, p.43.

24. R.J. Pell, J.B. Callis and B.R. Kowalski, Applied Spectroscopy, 1991, 45, 5,
808

25. R. Pornprasit, P. Pornprasit, P. Boonma and J. Natwichai, Journal of

Spectroscopy, 2016, Paper No.4024783, p.7

26. P.J. Hendra and J.L. Agbenyega in The Raman Spectra of Polymers, Wiley,
Chichester, UK, 1993.

27. Fourier Transform Raman Spectroscopy, Eds., D.B. Chase and J.F. Rabolt,
Academic Press, Orlando, FL, USA, 1994.

28. C.G. Smith, P.B. Smith, A.J. Pasztor, M.L. McKelvy, D.M. Meunier and
S.W. Froelicher, Analytical Chemistry, 1995, 67, 12, 97R.

29. P.J. Hendra and K.D.O. Jackson in Proceedings of the 144th IRC Conference,
26–29th October, Orlando, FL, USA, 1993, Paper No.105.

30. P.J. Hendra, C.H. Jones, P.J. Wallen, G. Ellis, B.J. Kip, M. van Duin,
K.D.O. Jackson and M.J.R. Loadman, Kautchuk und Gummi Kunststoffe,
1992, 45, 11, 910.

31. A.H. Kuptsov and G.N. Zhizhin in Handbook of Fourier Transform Raman
and Infrared Spectra of Polymers, Elsevier, Amsterdam, The Netherlands,
1998.

32. B.J. Clark, T. Frost and M.A. Russell in UV Spectroscopy, Springer Science &
Business Media, Berlin, Germany, 1993.

33. J. Keeler in Understanding NMR Spectroscopy, John Wiley & Sons,

New York, NY, USA, 2011.

34. T. Kitayama and K. Hatada in NMR Spectroscopy of Polymers, Springer

Science & Business Media, Berlin, Germany, 2013.

35. R.N. Ibbett in NMR Spectroscopy of Polymers, Springer Science & Business
Media, Berlin, Germany, 2012.

45
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

36. E. de Hoffmann in Mass Spectrometry: Principals and Applications,

John Wiley & Sons, New York, NY, USA, 2007.

37. R.P Lattimer and R.E. Harris, Mass Spectrometry Reviews, 1985, 4, 3, 369.

38. B. Welz and M. Sperling in Atomic Absorption Spectrometry, 3rd Edition,

1999, Wiley-VCH GmbH, Weinheim, Germany, 1999.

39. J. Nolte in ICP Emission Spectrometry, Wiley-VCH, Weinheim, Germany, 2003.

40. R. Jenkins in Chemical Analysis: A series of Monographs on Analytical

Chemistry and its Applications, Volume 152, 2nd Edition, 1999, Wiley,
Hoboken, NJ, USA, 1999.

41. J.C. Russ, M. Ashby Frs, R. Kiessling and J. Charles in Fundamentals of

Energy Dispersive X-ray Analysis, Elsevier, Amsterdam, The Netherlands,
1984.

42. T. Provder in Chromatography of Polymers: Hyphenated and

Multidimensional Techniques, Volume 731, American Chemical Society,
Washington, DC, USA, 1999.

43. V.G. Berezkin, V.R. Alishoyev and I.B. Nemirovskaya in Gas

Chromatography of Polymers, Elsevier, Amsterdam, The Netherlands, 1977.

44. H-J. Hubschmann in Handbook of GC-MS, John Wiley & Sons, New York,
NJ, USA, 2015.

45. S. Tsuge, H. Ohtani and C. Watanabe in Pyrolysis – GC/MS Data Book of

Synthetic Polymers, 1st Edition, Elsevier, Amsterdam, The Netherlands, 2011.

46. M.J. Forrest, S. Holding and D. Howells, Polymer Testing, 2006, 25, 63.

47. H. Pasch in HPLC of Polymers, Springer Science & Business Media, Berlin,
Germany, 1999.

48. H. Pasch and B. Trathnigg in Multidimensional HPLC of Polymers, Springer

Science & Business Media, Berlin, Germany, 2013.

49. W. Radke and J. Falkenhagen in Liquid Interaction Chromatography of

Polymers, Elsevier, Amsterdam, The Netherlands, 2013, Chapter 5.

50. M.C. McMaster in LC/MS: A Practical Users Guide, John Wiley & Sons,
New York, NY, USA, 2005.

46
 Analytical Techniques

51. W.M.A. Niessen in Liquid Chromatography-Mass Spectroscopy, 3rd Edition,

CRC Press, Boca Raton, FL, USA, 2006.

52. K. Ute in Encyclopedia of Analytical Chemistry, John Wiley & Sons Ltd,
New York, NY, USA, 2006, DOI:10.1002/9780470027318.a2033.

53. S. Mori and H.G. Barth in Size Exclusion Chromatography, Springer Science
& Business Media, Berlin, Germany, 2013.

54. C-S. Wu in Handbook of Size Exclusion Chromatography and Related

Techniques: Revised and Expanded, CRC Press, Boca Raton, FL, USA, 2003.

55. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.10.

56. J.G. Kreiner and W.C. Warner, Journal of Chromatography A, 1969, 44, 315.

57. J.D. Menczei and R.B. Prime in Thermal Analysis of Polymers: Fundamentals
and Applications, John Wiley & Sons, New York, NY, USA, 2014.

58. W.M. Groenewoud in Characterisation of Polymers by Thermal Analysis,

Elsevier, Amsterdam, The Netherlands, 2001.

59. V.A. Bershtein and V.M. Egorov in Differential Scanning Colaorimetry

of Polymers: Physics, Chemistry, Analysis, Technology, Ellis Horwood,
Hemstead, UK, 1994.

60. M.O. Abou-Helal and S.H. El-Sabbagh, Journal of Elastomers and Plastics,
2005, 37, 4, 319.

61. Z. Qixia, F. Hong, B. Zhiyang and L. Bogeng, China Synthetic Rubber

Industry, 2004, 27, 3, 1000.

62. S. Cook, S. Groves and A.J. Tinker, Journal Rubber Research, 2003, 5, 3,
121.

63. C. Price in The Rapra Collection of DSC Thermograms of Semi-crystalline

Thermoplastic Materials, Rapra Technology Ltd, Shawbury, UK, 1997.

64. M. Reading and D.J. Hourston in Modulated Temperature Differential

Scanning Colorimetry: Theoretical and Practical Applications in Polymer
Characterisation, Springer Science & Business Media, Berlin, Germany, 2006.

47
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

65. Modulated DSC Compendium; Basic Theory and Experimental

Considerations, TA Instruments Application Note TA210, New Castle, DE,
USA, 1996.

66. Modulated DSCTM Theory & Applications Basic Overview, TA Instruments

Application Note TA211, New Castle, DE, USA, 1995.

67. J.M. Hutchinson, Journal of Thermal Analysis and Calorimetry, 2003, 72,
619.

68. D.J. Hourston and M. Song, Journal of Applied Polymer Science, 2000, 76,
12, 1791,

69. T.F.J. Pijpers, V.B.E. Mathot and B. Goderis, Macromolecular, 2002, 35, 3601.

70. P. Gabbott, P. Clarke, T. Mann, P. Royall and S. Shergill, American

Laboratory, 2003, 35, 16, 17.

71. K.P. Menard in Dynamic Mechanical Analysis: A Practical Introduction, 2nd

Edition, CRC Press, Boca Raton, FL, USA, 2008.

72. H.M. Pollock and A. Hammiche, Journal of Physics D: Applied Physics,

2001, 34, R24.

73. Polymer Blend Study by Micro-Thermal Analysis, TA Instruments

Application Note TS58, New Castle, DE, USA.
http://www.tainstruments.com/pdf/literature/Hotline_99v1.pdf

74. Micro-Thermal Analysis - An Holistic Approach to Materials

Characterization, TA Instruments Application Note TA257, New Castle, DE,
USA.

75. H.M. Pollock and A. Hammiche, Journal Physics D: Applied Physics, 2001,
34, R23.

76. M.J. Forrest in Proceedings of Polymer Testing 1997, 7–11th April Rapra
Technology Ltd, Shawbury, UK, 1997, Paper No.5.

77. Modulated Thermogravimetric Analysis: A New Approach for Obtaining

Kinetic Parameters, TA Instruments Application Note TA237, New Castle,
DE, USA.

78. Obtaining Kinetic Parameters by MTGA, TA Instruments Application Note

TA251, New Castle, DE, USA.

48
 Analytical Techniques

79. C. Gamlin, M.G. Markovic, N.K. Datta, N.R. Choudhury and J.G. Matisons,
Journal of Thermal Analysis and Calorimetry, 2000, 59, 319.

80. Thermal Analysis Technical Literature (Theory and Applications),

TA Instruments, New Castle, DE, USA, 1994.

81. H. Lobo and C. Cohen, Polymer Engineering and Science, 1990, 30, 2, 65.

82. F. Oehmke and T. Wiegmann in Proceedings of ANTEC 1994, 1–5th May,

San Francisco, CA, USA, 1994, p.2240.

83. G.H. Michler in Electron Microscopy of Polymers, Springer Science &

Business Media, Berlin, Germany, 2008.

84. L.C. Sawyer, Polymer Microscopy, Springer Science & Business Media,
Berlin, Germany, 2012.

85. Y. Changjie, Q. Zhang, G. Junwei, Z. Junping, S. Youqiang and W. Yuhang,

Journal of Polymer Research, 2011, 18, 6, 2487.

86. D. Ma, J. Tan, Y. Li, C. Bian, G. Wang and S. Feng, Journal of Applied
Polymer Science, 2014, 131, 15, 7.

87. G.J. Vancso and H. Schonherr in Scanning Force Microscopy of Polymers,

Springer Science & Business Media, Berlin, Germany, 2010.

88. V.V. Tsukruk in Scanning Probe Microscopy of Polymers, Eds., B.D. Ratner
and V.V. Tsukruk, ACS – Division of Polymer Chemistry, Washington, DC,
USA, 1998.

89. Rubber Testing, TA Instruments, New Castle, DE, USA, 2014.

90. P. Albers, B. Freund, G. Prescher, K. Siebold and S. Wolff in Proceedings

of the 2nd International Conference on Carbon Black, 27–30th September,
Mulhouse, France, 1993, p.119.

91. W. van Ooij and M. Nahmias in Proceedings of the 134th ACS Fall Meeting,
18–21st October, Cincinnatti, OH, USA, 1988, Paper No.26.

92. K. Harrison and L.B. Hazell, Materials World, 1994, 1, 10, 532.

93. R.S. Lehrle and K.J. Niderost in Proceedings of the Rubbercon 1992
Conference, 15–19th June, Brighton, UK, 1992, p.281.

49
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

94. M. Dimopoulos, N.R. Choudhury, M. Ginic-Markovic, J. Matisons and

D.R.G. Williams, Journal of Adhesion Science and Technology, 1998, 12, 12,
1377.

95. E. Dujko and S.C. Cerelast in Proceedings of the IRC Conference,

12–14th May, Paris, France, 1998, p.87.

96. J. Jacks, Adhesives Age, 1997, 40, 11, 36.

50
3
Principal Components and Bulk
Composition

3.1 Introduction

In this section, the principal components, or ingredients, of a rubber compound are

defined as being:

• Polymer

• Plasticiser/oil

• Carbon black (CB)

• Inorganic filler

Work of the type described in this section yields important, fundamental information
on a rubber sample and so it can be carried out for a number of reasons, including:

• Characterisation of individual component(s) for identification purposes.

• Obtain fundamental information as part of a research and development

programme.

• Quality control.

• Defence of a patent.

• Reverse engineering.

• Failure diagnosis.

The information obtained from a bulk compositional analysis obviously plays an

important part in a full characterisation (i.e., reverse engineering) study, but there
are practical reasons why it can be regarded as having a stand-alone purpose and
so be covered in a specific section of this book. This is because for a number of the
exercises listed above, it is often convenient to look upon reverse engineering work
as comprising of two distinct stages, as defined below:

51
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Stage 1: Determination of bulk composition and the identity of the principal

components.

• Stage 2: Determination of additives, such as antidegradants, curatives and

accelerators.

This chapter covers stage 1, Chapter 4 deals with stage 2, and Chapter 5 presents
guidance and examples of how the various approaches and techniques can be brought
together to achieve the best results in reverse engineering and product deformulation
studies.

One of the practical reasons why splitting the work into two stages is useful is
because failure diagnosis work can often reach a successful conclusion by carrying
out the relatively cost effective and quick tests that comprise stage 1, and to have
set off a comprehensive analysis programme would have wasted time and money.
Unfortunately, it is often the case that rubber products fail in service because the
wrong type of rubber has been used or the level of one of the principal components
is not correct. If the initial work shows that nothing obvious is wrong with the bulk
composition, then a more in-depth analysis can proceed to target those aspects of
the formulation that are more likely to provide a solution to the problem given the
nature of the failure.

It will be evident that splitting the work into two stages, rather than setting off a
comprehensive analysis programme right from the start, can also be advantageous
in other instances, particularly if a large number of samples is concerned and some
‘screening’ has to take place to make the task more manageable.

The work that can be carried out to identify, quantify and characterise the principal
components in a rubber is described in Sections 3.2–3.7 and then the approaches
that can be used to obtain an overall measure of the bulk composition of a rubber
sample by a single operation [e.g., thermogravimetric analysis (TGA) analysis], are
brought together in Section 3.8.

3.2 Identification and Quantification of the Polymer in Rubber

Compounds

One of the most basic and useful tests to carry out on a rubber sample is to determine
the generic type of polymer (or blend of polymers) that is present. This has resulted
in some useful textbooks being published that address this particular aim [1].

Because a typical rubber compound contains only ≈50% polymer, a direct analysis by
the most convenient technique, attenuated total reflectance (ATR) Fourier–Transform

52
 Principal Components and Bulk Composition

infrared (FTIR) spectroscopy, rarely provides a definitive answer. It is also the case
that often the majority of the work carried out by laboratories is on samples that
have been vulcanised and so are insoluble in solvents, meaning that transmission
FTIR work cannot be carried out.

In order to obtain an FTIR spectrum, a certain amount of preparation work is required

to remove the interference of the organic additives. After milling or commuting the
sample to increase the surface area, a solvent extraction is performed to remove
organic additives such as plasticisers/oils (Section 3.4.1) and the species related to
the cure system and antidegradant systems. The extracted portion of the sample can
then be analysed by an ATR route (e.g., using a single bounce ATR accessory having
a diamond window), although the data obtained will still be compromised by the
absorption of the infrared (IR) light by any CB present, and any inorganic fillers will
contribute IR absorption bands to the sample spectrum.

The usual practice is, therefore, to carry out an additional preparation step to remove
the influence of the CB and the inorganic fillers and additives, such as zinc oxide
(ZnO). This is achieved by pyrolysing the solvent extracted portion (e.g., in a modified
ignition tube) and to record a FTIR spectrum of the collected pyrolysis condensates.
One example of a pyrolysate FTIR spectra of a rubber, a natural rubber (NR) is
provided in Chapter 2 (Figure 2.1), and another, that of a castable polyurethane (PU)
rubber, is shown in Figure 3.1. Providing that an extensive database of pyrolysate
spectra is available, for example, the one published by Forrest and co-workers [2], this
method can be very successful in both differentiating between generic polymer types
[e.g., NR and styrene-butadiene rubber (SBR)]. Also, in some cases, it can show the
differences between the grades of a polymer, such as the copolymer, terpolymer, and
tetrapolymer versions of fluorocarbon rubber. It is also possible to use the technique
to identify certain rubber blends, particularly blends of diene rubbers [e.g., SBR
and polybutadiene (BR)], and the results can be expressed semi-quantitatively if the
appropriate standards are available.

In cases where the rubber is not vulcanised or compounded, the situation is obviously
much simpler and either the sample can be analysed directly using an ATR technique
or by transmission IR if a suitably thin film can be prepared. It may be necessary
to dissolve an amount of the sample in a suitable solvent (e.g., chloroform) and
obtain a transmission IR spectrum from a cast film. Identifications can then be made
by reference to FTIR libraries containing transmission and ATR spectra [2]. The
transmission approach can also be used to obtain information on the microstructure
of a rubber (Section 3.3.2). An example of a transmission FTIR spectrum of a rubber,
NR, is shown in Figure 2.2 in Chapter 2, and that of a millable-type polyether-type
PU rubber, is shown in Figure 3.2.

53
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

RP28.0501 Urethane castable ether

94
86
78
70
Transmittance (%)

62
54
46
38
30
22
14
6
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm )
-1

Figure 3.1 Pyrolysate IR spectrum of a polyether-type castable elastomer.

Reproduced with permission from M.J. Forrest in Chemical Characterisation of
Polyurethanes, Rapra Review Report No.108, Smithers Rapra, Shawbury, UK,
1999, p.19. ©2001, Smithers Rapra [3]

RT28.0801 Urethane millable

100
92
84
76
Transmittance (%)

68
60
52
44
36
28
20
12
4
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm )
-1

Figure 3.2 Millable polyether-type PU rubber. Reproduced with permission from

M.J. Forrest in Chemical Characterisation of Polyurethanes, Rapra Review Report
No.108, Smithers Rapra, Shawbury, UK, 1999, p.15. ©2001, Smithers Rapra [4]

54
 Principal Components and Bulk Composition

Another method which enables the polymer type within a rubber sample to be
identified is pyrolysis gas chromatography (GC). This method can be applied to
either the extracted portion of the sample or the sample as-received. In the latter, the
relatively low-molecular weight(s) (MW) oils elute early on and do not tend to have a
significant influence on the pyrogramme as a whole. However, to avoid complications
it is advisable to use samples that have been solvent-extracted. As with pyrolysis IR
spectroscopy, a database of standards needs to be put together and then the pyrogram
fingerprints of unknowns can be matched.

The temperature that the pyrolysis unit is set at, and the design of the unit (e.g., Curie
point or microfurnace), will affect the pyrograms produced from a particular sample.
As with pyrolysis FTIR, it is possible to identify and quantify blends and blend ratios.
As an illustration of the data produced by this approach, reference fingerprint GC
pyrograms for SBR and NR obtained at 750 °C are shown in Figures 3.3 and 3.4,
respectively, and typical experimental conditions for pyrolysis GC experiments are
shown below.

µV
1000000

800000

600000

400000

200000

0 5 10 15 20 25 min

Figure 3.3 GC pyrogram obtained at 750 °C for SBR. Reproduced with permission
from M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,
Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.16.
©2001, Smithers Rapra [5]

55
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

µV

1400000

1200000

1000000

800000

600000

400000

200000

0
0 5 10 15 20 25 min

Figure 3.4 GC pyrogram obtained at 750 °C for NR. Reproduced with permission
from M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,
Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.16.
©2001, Smithers Rapra [5]

General requirements for pyrolysis GC are:

• Sample weights: 0.1–5.0 mg.

• Pyrolysis temperatures: 400–800 °C (microfurnace and Curie point pyrolysers).

• 800–1,200 °C (platinum filament pyrolyser).

Microfurnace-type pyrolyser – specific conditions using a commercial example: a

SGE™ Pyrojector:

• Weight of sample: 0.4 mg.

• Pyrolysis temperature: 600 °C.

• Pyrolysis gas: helium at 15 psi.

56
 Principal Components and Bulk Composition

• GC column: 5% diphenyl siloxane 30 m × 0.25 mm, 1.0-µm film.

• GC temperature programme: 2 mins at 50 °C followed by 10 °C/min to 280 °C

and then 10 min at 280 °C.

• GC carrier gas: helium at 10 psi.

• GC injector temperature: 320 °C.

• Detector: flame ionisation detector at 350 °C.

Adding a mass spectrometer to the system so that the technique becomes pyrolysis
GC–mass spectrometry (MS) further increases the usefulness of the method and the
information that can be obtained. For example, Table 3.1 lists the principal diagnostic
pyrolysis species that can be detected in the pyrograms of some common rubbers.

Table 3.1 Principal diagnostic pyrolysis species for common rubbers

Rubber type Diagnostic pyrolysis species
PI rubber and NR Isoprene and dipentene
SBR Butadiene, 4-vinyl-1-cyclohexene and styrene
BR Butadiene and 4-vinyl-1-cyclohexene
Isobutylene-isoprene rubber Isobutene and isoprene
NBR Butadiene, acrylonitrile and 4-vinyl-1-cyclohexene
NBR: Nitrile rubber
PI: Polyisoprene
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.17.
©2001, Smithers Rapra [6]

It can be seen from Table 3.1 that rubbers will often regenerate monomers and yield
dimers and other oligomers upon pyrolysis under non-oxidising conditions (e.g., under
helium). However, care needs to be taken in the interpretation of the results, as the
presence of various ‘monomer type’ substances in the following list of the pyrolysis
products of a polychloroprene (CR) rubber demonstrate:

• Chloroethene

• Butadiene

57
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Chloroprene

• Toluene

• Styrene

• Chlorobenzene

• 1-Chloro-3-methylbenzene

• 4-Chloro-1,2-dimethylbenzene

In the case of the aromatic compounds in this list, these non-chlorinated species
originate from the loss of hydrogen chloride (HCl) followed by cyclisation. Common
pyrolysis products can cause problems in trying to detect amounts of, for example,
SBR, in CR. Hence, it is important to target ‘secondary’ diagnostic pyrolysis products
as well as ‘primary’.

In addition to identifying the polymer type of a rubber, pyrolysis GC can also be

used to quantify the polymers in a blend. Fuh and Wang [7] used the peak ratio of
1-chloro-4-(1-chloroethenyl)-cyclohexene from CR, and benzonitrile from NBR, to
accurately quantify the two polymers in standard samples having blend compositions
varying from 80:20 to 20:80.

A review covering the use of pyrolysis IR spectroscopy, pyrolysis GC, solid-state

magic-angle spinning 13C-nuclear magnetic resonance (NMR) and 300-MHz proton
NMR to identify polymers in rubber samples was presented at the International Tyre
Conference in Ohio in 1996 [8]. In addition to reviewing the subject, the analytical
results obtained using the four techniques on six prepared rubber samples by twelve
laboratories were detailed.

In addition to being used to identify the polymer type, solid-state 13C-NMR can also
be used in conjunction with physical testing data to characterise crosslinked rubber
samples [9]. The data obtained at a constant cure temperature and time, but with a
varying amount of sulfur curative, was found to correlate well. However, keeping
the sulfur level constant but varying the cure temperature and time was found to
produce large changes in the results obtained by physical testing (e.g., tensile strength,
elongation and modulus), but led to only small differences in the 13C-NMR data.

It is also possible to obtain an indication of the type of rubber in a sample using

dynamic mechanical analysis (DMA) and utilising its high sensitivity with regard to
recording the glass transition temperature (Tg). As the rubber sample is heated from
well below sub-ambient (e.g., −150 °C) to above ambient (e.g., +200 °C) temperatures,
a major change in its modulus occurs as it passes through its Tg due to segmental

58
 Principal Components and Bulk Composition

rotation taking place within its polymer molecules. This change is most easily detected
by recording the change in the tan δ plot (i.e., loss modulus/elastic modulus) and the
peak in this plot is a measure of the Tg of the rubber. This is demonstrated in Figure
3.5, which shows the DMA thermogram of a CR-based rubber sample.

To carry out this type of work, it is necessary to have a reference list of the Tg of
a range of rubbers. An example of such a list, obtained using differential scanning
calorimetry (DSC) data, is provided in Table 3.2. However, care needs to be taken
because there are no definitive values for Tg because the value obtained for a given
rubber or plastic is dependent on both the type of test that is performed (e.g., DMA,
DSC, torsion pendulum) and the precise operating conditions of the instrument (e.g.,
the heating rate, rate of deformation). It is, therefore, important, as with so many
other analytical procedures, to standardise the DMA conditions and run standard
samples in order to make the fullest use of this capability.

AL0047 rubber sample F7B new

Scales not UKAS calibrated
2.5E+09 1.00
Temperature (ºC) = -300
0.90
2.0E+09 0.80
Modulus (Pa)

0.70
1.5E+09 0.60 Modulus
tan δ

0.50 tan δ
1.0E+09 0.40
0.30
5.0E+08 0.20
0.10
0.0E+00 0.00
-100 -75 -50 -25 0 25 50 75 100 125 150 175 200 225 250
Temperature (ºC)

Figure 3.5 Typical DMA trace for a rubber sample (a CR rubber). The modulus
shows a significant reduction at Tg, while tan δ shows a large peak. Reproduced
with permission from M.J. Forrest in Principles and Applications of Thermal
Analysis, Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.214. ©2008,
Blackwell Publishing [10]

59
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The value for a Tg recorded by DMA will also be dependent to varying degrees on
sample-related factors, such as the presence and level of a plasticiser, because the
value reflects the Tg of the complete sample and not just the polymer that is present.
However, even so, it can often provide a useful indication of polymer type, and is
particularly valuable in detecting if a blend of polymers is present. An example of
this type of application is shown in Figure 3.6, in which the DMA thermogram of
a blend of NBR and SBR is shown. The ratio of the two rubbers in this blend was
NBR (85%):SBR (15%) and so, as mentioned above, by the use of such standard
materials, it is possible to use DMA to obtain the approximate blend ratio of an
unknown sample. Depending upon the particular blend of polymers present, in can
be possible to detect the presence of a polymer in a blend down to ≈5% of the total
polymer fraction. Due to this capability, carrying out DMA can often corroborate
and help to reinforce the data generated by FTIR.

Mix C-NBR 85%: SBR 15%

Scales not UKAS calibrated
1.4E+09 1.80
Temperature (ºC) = -9.4
1.60
1.2E+09
1.40
1.0E+09
1.20
Modulus (Pa)

8.0E+08 1.00
Modulus
tan δ

tan δ
6.0E+08 0.80

0.60
4.0E+08
Temperature (ºC) = -37 0.40
2.0E+08
0.20

0.0E+00 0.00
-150 -100 -50 0 50 100 150 200 250
Temperature (ºC)

Figure 3.6 DMA trace showing the presence of 15% SBR within a NBR matrix
– see the small peak in the tan δ just above −50 °C. Reproduced with permission
from M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,
P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.218. ©2008,
Blackwell Publishing [11]

60
 Principal Components and Bulk Composition

One specific area where DMA is also very useful is in differentiating between polymeric
materials and samples that have the same monomers in their polymer molecules, for
example, between vulcanisable SBR and styrene-butadiene-styrene (SBS) thermoplastic
rubbers. FTIR can struggle in such cases, both in terms of differentiating between these
two distinct types of rubbers and if the characteristic absorption bands for styrene
and butadiene units are present for other reasons (e.g., a blend of SBR and BR) or
the presence of a styrene-butadiene thermoplastic resin in either SBR or BR. The
differences in the Tg between these different types of styrene-butadiene based polymers
(e.g., −40 °C for the SBR; −80 and +100 °C for the styrene and butadiene units in
the SBS thermoplastic rubber, respectively) means that DMA can provide answers.

As indicated above with the example of styrene-butadiene resin, it is also possible to

use DMA to detect the presence of a rubber within another matrix, such as a plastic.
A common example of this situation is when BR has been mixed with polystyrene
(PS) to produce a high-impact grade. The large difference between the Tg of the two
polymers (−80 and +100 °C, respectively) and its high sensitivity makes the technique
ideal for this type of investigation. Another example of DMA being used to detect the
presence of a rubber within a rubber–plastic blend is shown in the DMA thermogram
in Figure 3.7. In this case, butyl rubber and a semi-crystalline thermoplastic, low-
density polyethylene (LDPE), have been mixed in a ratio of 50:50. The data plotted
in the tan δ curve show very clearly the Tg of the butyl rubber phase (about −56 °C)
and the softening and melting of the LDPE phase (≈60 to 120 °C). It was possible to
continue to record data on the sample after the LDPE phase had melted due to the
presence of the butyl rubber; a sample of pure LDPE would have lost dimensional
stability once past its melting point and dropped from the instruments grips.

DMA can also be used as a means of characterising blends of rubbers in terms of

their compatibility, degree of mixing, and of assessing which rubber is present as the
dispersed phase and which as the continuous phase [13, 14]. Another application
involves its use to investigate the reinforcement of rubbers by different fillers and the
influence of coupling agents in this process [15].

61
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Gasket sample – blend of butyl rubber and LDPE

3.0E+09 1.40
Temperature (ºC) = -55.8
1.20
2.5E+09

1.00
2.0E+09
Modulus (Pa)

0.80
Modulus

tan δ
1.5E+09 tan δ
0.60

1.0E+09
0.40

5.0E+08 0.20

0.0E+00 0.00
-150 -100 -50 0 50 100 150 200 250
Temperature (ºC)

Figure 3.7 DMA thermogram of a 50:50 blend of butyl rubber and LDPE. The
butyl rubber supports the material through the melting range of the LDPE.
Reproduced with permission from M.J. Forrest in Principles and Applications of
Thermal Analysis, Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.222.
©2008, Blackwell Publishing [12]

Another common thermal technique that can also be used to detect the Tg is DSC
(Table 3.2). This is done by looking for a characteristic inflexion in the heat capacity
curve (i.e., specific heat curve) within the thermogram. A similar temperature range is
employed as that used for DMA and again standard materials should be analysed to
obtain in-house reference data. Because the change in the heat capacity of a sample as
it moves through its Tg is far less pronounced than the change in its modulus, the DSC
is a much less sensitive technique than DMA for this type of work. This makes the
identification of polymers in blends much harder, particularly if one of the polymers is
only present at the relatively low level (e.g., ≤20%). A guide to the relative sensitivity
of different thermal analytical techniques to detect the Tg is shown in Table 3.3.

62
 Principal Components and Bulk Composition

Table 3.2 Approximate Tg values for a range of commercial rubbers in the

uncompounded, uncured state obtained using DSC
Rubber Type Approximate values for Tg (°C)
Polyisobutylene −70
EPM and EPDM −65
Fluorocarbon −20
Hydrin −50
Polydimethyl siloxane −120
BR −100
CR −45
Butyl rubber −70
Halogenated butyl rubber −70
Synthetic polyisoprene −70
NR −70
NBR −65 to 0*
SBR −50
Polysulfide −60
PU (ether and ester types) −40**
* Depends upon level of acrylonitrile monomer
**Varies with chemical structure
EPM: Ethylene propylene monomer
EPDM: Ethylene propylene diene monomer
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.54.
©2001, Smithers Rapra [16]

Table 3.3 Relative sensitivity of a range of thermal analytical techniques for

the detection of Tg
Technique Measured parameter Relative sensitivity*
DSC Heat capacity 1
Thermal mechanical analysis Coefficient of expansion 5

DMA Modulus 10
Dielectric analysis Permittivity 10
*where a value of 10 represents the most sensitive
Reproduced with permission from M.J. Forrest in Principles and Applications of Thermal Analysis,
Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.236. ©2008, Blackwell Publishing [17]

63
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

With respect to the quantification of the polymer component within a rubber, the
easiest technique to use is TGA. In a number of cases it is also possible to detect a
blend of polymers using TGA and to obtain a semi-quantification of the blend ratio
by reference to the splitting of the weight loss derivative. Unfortunately, some rubber
types (e.g., CR) undergo two-stage weight loss when heated and the first stage (loss
of HCl in the case of CR) will overlap with the weight loss due to the volatilisation
of any low-MW organic substances in the rubber (e.g., plasticiser). It is, therefore,
difficult to use TGA alone to determine the polymer content of these samples and
other analytical techniques (e.g., elemental analysis) often have to be used as well.

Table 3.4 Carbonaceous residues obtained from a range of commercial

rubbers by TGA
Polymer type Carbonaceous residue (%)*
Chlorosulfonated PE rubber 2–4
Fluorocarbon rubber 3–10
Ethyl acrylate rubber 6–8
NBR (20–50% ACN) 2–12
Epichlorohydrin rubber 5–15
CR rubber 20–25
PVC (used in blends with NBR) 15–20
*Carbonaceous residue obtained from 100% polymer and a range is given because the exact
amount obtained depends upon the actual grade of polymer
PE: Polyethylene
PVC: Polyvinyl chloride
Reproduced with permission from M.J. Forrest in Principles and Applications of Thermal
Analysis, Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.204. ©2008,
Blackwell Publishing [18]

Another complicating factor is that a number of rubbers (e.g., halogenated rubbers,

NBR, acrylic rubbers) do not completely pyrolyse in the nitrogen atmosphere region
and leave a carbonaceous residue (Table 3.4) which is lost due to oxidisation in
the air atmosphere region. This carbonaceous residue, which may also overlap to a
degree with the CB weight loss in this region, has to be included in order to obtain an
accurate value for the polymer content. It is, however, possible, in some limited, specific
cases involving co-polymers, to use the fact that some polymers give a carbonaceous
residue during TGA run to obtain some approximate quantitative information on

64
 Principal Components and Bulk Composition

the monomer ratio and, hence, the generic ‘grade’ of rubber is present. One classic
example of this is NBR, a copolymer of acrylonitrile (ACN) and butadiene. NBR is
available in a range of grades in which the ACN level can vary from ≈20 to ≈50%
according to the level of oil resistance, low-temperature flexibility and other properties
that are desired. Any carbonaceous residue obtained in the TGA trace is only derived
from the ACN units in the NBR and so the amount, adjusted for the total amount of
polymer in the rubber compound, is indicative of the level of this monomer (Table 3.4).

The use of TGA to determine the polymer content of a rubber sample and provide
an indication of the type of polymer that is present within it is described in more
detail in Section 3.8. This section also describes how TGA can be used to determine
the bulk composition of rubber samples.

3.3 Characterisation of Rubbery Polymers

Section 3.2 covered the determination of polymer type, one of the most basic
requirements of rubber analysis, and this section will cover the analytical techniques
that can be applied both to raw, uncompounded polymers and fully compounded
materials in order to obtain a fuller characterisation (e.g., MW, ratio of monomers)
of the polymer.

3.3.1 Determination of Molecular Weight [19]

In common with other polymers, rubbery polymers have a distribution of MW. This
distribution will be widest for polymers produced by free-radical polymerisation (e.g.,
NBR) and narrowest for ionic polymers such as EPDM rubbers.

It is possible to work with both raw (i.e., uncompounded) rubbers, as well as

compounded, unvulcanised ones, although the production of an insoluble, infinite
network by vulcanisation obviously excludes the analysis of cured rubbers.

There are a number of ways in which the molecular weight distribution (MWD) of
a polymer can be expressed:

Σ NM
Mn = (3.1)
ΣN

Σ NM2
Mw = (3.2)
Σ NM

65
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Σ NM3
Mz = (3.3)
Σ NM2

where ‘N’ is the total number of molecules of molecular mass ‘M’ in all cases; Mn is
the number average molecular weight, Mw is the weight average molecular weight,
and Mz is the Z average molecular weight.

The ‘Z’ average is very sensitive to high-MW molecules within the polymer. The weight
average is always greater than the number average. A measure of the breadth of the
molecular distribution [i.e., polydispersity (PDI)] of the polymer can be obtained by
dividing the weight average by the number average:

PDI = Mw/Mn (3.4)

As a polymer tends towards a uniform MWD, its PDI approaches unity and it is said
to be ‘monodisperse’.

There are a wide variety of methods that can be used to obtain the MW. The principal
ones are described in Sections 3.3.1.1–3.3.1.4.

3.3.1.1 Gel Permeation Chromatography/Size Exclusion Chromatography

This is the pre-eminent technique and has been refined and developed extensively
over the past 45 years. It is possible to use gel permeation chromatography (GPC)
for both raw polymers and compounded ones, and Mn, Mw and Mz values can
be obtained. GPC is essentially a liquid chromatography (LC) technique with the
sample being dissolved in a suitable solvent [e.g., toluene for SBR; tetrahydrofuran
(THF) for NBR] and injected into the system where the molecules are separated
according to their molecular size by a column that contains a swollen, particulate,
porous gel packing.

The standard detector for GPC systems is the refractive index (RI) detector, but double
(i.e., RI and viscosity) or triple (i.e., RI, viscosity and light scattering) detection systems
have been developed recently. These multiple detection systems provide more accurate
measurements of the MW by obtaining actual Mark–Houwink parameters, via the

66
 Principal Components and Bulk Composition

viscosity detector (Section 3.3.1.2), for the polymer being analysed, as opposed to
PS-equivalent MW from the sole use of monodisperse PS calibrants. In addition to
the advantage that the viscosity detector brings to GPC, an inherent property of the
light-scattering detector is that it provides ‘true’ MW data. Equipping a GPC system
with triple detection also has other advantages, such as the capability to provide
structural information on polymers.

It is also possible to use spectroscopic detectors such as IR and ultraviolet (UV) with
GPC systems. One practical use of the latter is to determine if chemical modifiers (e.g.,
silane coupling agents) are bound to polymer molecules. This is done by measuring
the MWD of the modified rubber at a UV wavelength where the rubber itself does
not contribute to the absorbance and comparing this with the distribution of the
unmodified rubber.

There can be two problems encountered with the GPC of rubbers. The first is that
rubbers can have relatively high MW (>1,000,000) and care has to be taken that the
flow rates used with modern small particle size packings (i.e., <10-µm diameter) do
not cause the rubber molecules to break up under the imposed shear forces. Flow
rates <0.5 ml/min may have to be used to minimise back pressure. The other problem
is that free radical-polymerised rubbers, and NR, can have a degree of insoluble-
crosslinked gel associated with them.

GPC is often carried out on a comparative basis and Figure 3.8 illustrates this, where
the GPC chromatograms of two NBR samples produced using THF as the solvent
for the samples and the mobile phase have been overlaid. In this case, the work was
carried out because it was thought that differences in processing performance were
related to MW differences. However, the MW data computed from Figure 3.8 are
shown in Table 3.1 and it can be seen that the difference between the samples is
relatively insignificant.

Table 3.5 MW data for two NBR samples (Figure 3.8)

Mw Mn PDI
Sample A 281,000 89,800 3.1
Sample B 331,000 92,000 3.6
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.13.
©2001, Smithers Rapra [20]

67
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

MWD

1.00

0.75
Frequency

0.50

0.25

0.00
3.00 4.00 5.00 6.00 7.00 8.00
Log M

Figure 3.8 Overlaid GPC chromatogram of two NBR. Reproduced with

permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds and
Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001,
p.13. ©2001, Smithers Rapra [20]

Some of the typical applications that can be conducted on rubbers by GPC/size

exclusion chromatography (SEC) are described by Agilent Technologies in a recent
Application Compendium [21]. This publication describes applications in which
various GPC/SEC systems are used to obtain the MW and, in some cases, structural
information, on various rubber samples. The applications that are covered include:

• High-sensitivity analysis of NR with evaporative light scattering (ELS) detection

– This example demonstrates how ELS detection can improve the data that are
obtained on NR samples because it eliminates the interference of the ‘system
peaks’ that occur with standard RI detection.

• The analysis of NR and synthetic BR with triple detection – The example provided
show how both MW data and structural information can be obtained on NR and
BR samples using systems that are equipped with triple detection capability.

• General analysis of synthetic elastomers – Use of GPC/SEC systems with columns

capable of resolving MW up to 10,000,000 Da to provide MW data on PI, BR,
SBR and butyl rubbers is demonstrated.

68
 Principal Components and Bulk Composition

• Polydimethylsiloxane analysis using GPC/viscometry – Illustrates how two silicone

rubbers, although quite different in MW, are structurally similar and so their
viscoelastic behaviour as a function of MW would be expected to be comparable.

• Analysis of SBS block copolymers – Demonstrates how ‘conventional’ GPC

(i.e., THF solvent at 40 °C using RI detection) can be used to detect the presence
of homopolymers, which can significantly affect the end properties in SBS
thermoplastic rubbers.

Prior to the advent of advanced detection systems, the use of modular accessories like
the LC–Transform with GPC extended its capabilities by allowing IR spectroscopy
data to be generated on the molecules eluted during an analysis. The LC–Transform
system worked by impinging the solute that elutes from a GPC column onto a
germanium disc by removing the solvent using a combination vacuum and high
temperature. The germanium disc is then placed into a module fitted into the sample
compartment of an IR bench and IR data collected on the entire chromatogram
using continuous collection software. This procedure, in conjunction with elution
data obtained from a GPC chromatogram recorded on the same sample with an RI
detector, enables the IR spectrum of any particular retention time, or range of retention
times, to be viewed. This technique is very useful for compounded rubbers because
it allows polymer blends to be identified, as well as relatively high MW additives
such as plasticisers, and its full capabilities have yet to be replaced by a commercial
in-line detection system.

3.3.1.2 Viscosity

Viscosity is a standard procedure for MW determinations and involves the use of

specially designed viscometers to accurately measure the viscosity of a polymer
solution. From this the intrinsic viscosity is determined and, hence, the MW. The
time taken for the polymer solution to pass between two marks on the viscometer
is compared with that of pure solvent and the ratio is the viscosity of the solution.
Successive dilutions give a range of concentrations and times from which the intrinsic
viscosity can be calculated. The value for this is then entered into the Mark–Houwink–
Sakwada equation:

Intrinsic viscosity = KMX (3.5)

where ‘K’ and ‘X’ are constants for the particular solvent being used and are referred
to as the ‘Mark–Houwink parameters’.

69
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Depending on the source of the Mark–Houwink parameters, the MW can be expressed

as either the number average or weight average.

3.3.1.3 Osmometry

There are two principal osmometry techniques [22, 23]: vapour pressure osmometry
and membrane osmometry.

Vapour pressure osmometry involves the indirect measurement of the lowering of

the vapour pressure of a solvent due to the presence of a solute. It is based on the
measurement of the temperature difference between droplets of pure solvent and of
the polymer solution maintained in an isothermal atmosphere saturated with the
solvent vapour. Calibration involves the analysis of standards of known MW and
should be over the entire MW of interest to ensure the best results. The technique is
useful for polymers that have MW in the range 500–50,000 Da.

Membrane osmometry involves the use of a membrane between the solvent and the
polymer solution that is permeable to the solvent molecules, but not to the solvated
polymer molecules. This is the major restriction on the technique and means that it
cannot be used for low-MW polymers (i.e., those below ≈50,000 Da).

3.3.1.4 Light Scattering

Recent developments in instrumentation have led to a resurgence of this technique:

the Mw is obtained from the light-scattering behaviour of polymer solutions. These
developments have centred on low-angle laser light scattering [24], multi-angle laser
light scattering (MALLS) [25, 26], and ELS. As well as being stand-alone instruments,
light scattering detectors have been developed for GPC systems (Section 3.3.1.1).
MALLS detectors have the additional benefit of being able to provide direct
information on the branching ratio of a polymer [27].

To reduce variability, care has to be taken to ensure that the polymer solution is clear
and low cell volumes (e.g., 0.1 µl) and dilute solutions help with this. Solvent choice
is also critical, with the best results being obtained with solvents that have a very
different RI to the polymer and have low-light scattering properties.

ELS involves the removal of the solvent from the solvated polymer as it elutes down
a drift tube. The isolated polymer particles then scatter light from the light source.

Light scattering can produce precise data and has been used successfully on polymers
with MW ranging from 10,000 to 10,000,000.

70
 Principal Components and Bulk Composition

3.3.1.5 Miscellaneous Techniques

Other methods of determining or investigating MW include:

• End group analysis – provides Mn data.

• Ultra-centrifugation – provides Mw data.

• Comparative ‘indirect’ techniques (e.g., DSC or DMA) that work by monitoring

properties that can be influenced by MW (e.g., Tg and crystallinity).

3.3.2 Monomer Types and Microstructure

The use of pyrolysis FTIR spectroscopy and pyrolysis GC to determine the generic
polymer type of a rubber, and hence information on the monomers used to produce
it, is covered in Section 3.2. This section also mentions the use of FTIR in the
transmission mode for identification of a polymer where the rubber is uncompounded
or unvulcanised. As with all FTIR work, access to a comprehensive FTIR library
is important and the example published by Forrest and co-workers [2], already
mentioned in Section 3.2 in relation to its collection of pyrolysate spectra, also contains
an extensive collection of transmission spectra. With respect to quantitative work,
where appropriate standards are available it is also possible to use FTIR spectroscopy
to quantify the amounts of monomers in copolymers and terpolymers, providing a
reasonably large amount of each monomer is present.

Regarding the minor monomers (e.g., those at <10% w/w) FTIR in the transmission
mode with uncompounded and unvulcanised rubber samples can, in some cases, enable
qualitative information to be obtained. For example, in the case of the terpolymer
EPDM, transmission FTIR can be used on the pure polymer to identify the diene
monomer. However, if this rubber is compounded, it may not be possible to remove
enough of the additives to stop them masking the diene monomer in the rubber and,
if the sample is vulcanised, additional detection difficulties are encountered as a large
proportion of the diene monomer will have reacted with the cure system components.

Other analytical techniques which are more sensitive than FTIR can be used for the
determination of minor monomers, even in rubber samples that have been vulcanised.
For example, Yamada and co-workers showed that high-resolution pyrolysis GC
can be used to both identify and quantify ethylidene norbornene diene monomer in
vulcanised EPDM rubbers [28].

To assist workers in their efforts to characterise the structural features present in cis-
and trans-1,4-polybutadiene, Nallasamy and co-workers [29] analysed both these

71
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

polymers using Raman spectroscopy and FTIR. The analyses were performed over
the following frequency ranges:

• Raman spectroscopy: 100–4,000 cm−1

• FTIR spectroscopy: 200–4,000 cm−1

Tables are provided in this reference which list probable assignments for the absorption
bands that the research team observed in the data, made using the help of intensities
and normal co-ordinate analyses. The research team found that the calculated normal
frequencies were in good agreement with those determined experimentally.

Another technique that is very effective at identifying the exact monomers present
in a rubber is NMR. Although the best results are obtained on a solution of the
pure polymer, it is possible to use the solid-state technique on vulcanised samples
and also to analyse solvent extracts of these by 13C-NMR; the chemical composition
of the extracted high-MW oligomers usually provide an accurate enough reflection
of the original polymer. The NMR technique will also provide information on the
proportions of the various monomers in copolymers and terpolymers. In contrast
to IR spectroscopy, this information can be obtained without having to analyse or
having available polymer standards of known monomer composition. It is possible
for experienced specialists to interpret the NMR data using fundamental principles
and the areas under the diagnostic chemical shifts.

NMR instruments can also be used to obtain fundamental structural information

on polymers such as end group chemistry, branch points and structural isomerism
(e.g., cis- and trans-1,4-isomer ratios in diene rubbers). It is also possible to use
FTIR spectroscopy for the latter example, although some of the isomeric structures
have relatively weak absorptions which makes detection difficult if they are at a low
concentration.

Two sets of workers who have carried out a lot of work on the use of solid-state
13
C-NMR to investigate the structure of vulcanised rubbers are Koenig and his
team at Case Western Reserve University [30–32] and Gronski and co-workers at
the University of Freiberg [33–35]. The low density of the crosslinks present in a
vulcanisate challenge the sensitivity of the technique, but it has still been possible
to distinguish modifications to the main chain (isomerism and cyclic sulfides), the
crosslinks themselves, and the reactions of cure and reversion [36].

Sung-Seen Choi and co-workers [37] at Sejong University used both FTIR and
H-NMR to determine the microstructural composition of unbound polymer (i.e.,
solvent extractable) that was present in filled SBR/BR blends. They concluded that
H-NMR had fewer experimental errors than transmission FTIR. The research team

72
 Principal Components and Bulk Composition

used H-NMR to determine the 1,2- and 1,4-butadiene contents of the unbound rubber
to obtain average SBR/BR ratios. The bound rubber microstructural composition in
the filled SBR/BR blends was then obtained using a combination of the bound rubber
content and the average unbound microstructural compositions. The results showed
that the bound rubber had a higher proportion of SBR to BR than that present in
the rubber formulation. Three of the authors of this work also carried out a study
on sulfur-cured solution SBR, the aim of which was to obtain information on their
microstructure [38]. The cured rubbers were analysed by both solution H-NMR and
transmission FTIR and the results obtained were compared with that of raw solution
SBR. The comparison revealed that the microstructures were different because of
the pendent group formed by the cure accelerator and cis–trans isomerism. Also, the
styrene and trans-1,4 unit contents of the cured SBR were larger than those of the raw
SBR, whereas the 1,2 unit contents of the cured SBR tended to be smaller than those
of the raw SBR. The research team also compared the H-NMR and FTIR results,
and found that the NMR data had narrower error ranges and better correlation than
that obtained by FTIR.

Shield and Ghebremeskel [39] investigated the possibility of using TGA to determine
the styrene content of SBR. An alternative method to the traditional methods
for determining the styrene content of SBR in tyre treads was thought desirable
because these could be hampered by the presence of additives and other polymers
in the samples. The approach that they used involved monitoring the shifts in the
decomposition temperature of SBR as a function of the styrene content. It was found
that the magnitude of the shift could be used to determine the styrene content of
an SBR. The effect of emulsifiers, CB filler, curing agents and other polymers on
this shift was investigated and ways of minimising their influence on the data were
suggested.

For further information in this area, the reader is directed to the work of Brame, who
provides a review on the use of spectroscopic techniques (NMR, FTIR and MS) for
the determination of polymer structure [40].

3.3.3 Specific Heat Values

DSC can be used to determine the specific heat of polymers and the effect that bulk
ingredients (e.g., CB) and the vulcanisation process have on these values. Such a study
has been published by Changwoon Nah and co-workers [41] and encompassed a
number of rubbers, including NR, SBR, BR and butyl rubber.

73
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

3.3.4 Crystallinity

Conventional rubbers are, to a very large extent, amorphous materials. This feature
is essential if they are to exhibit the properties that define them as a class of material
(e.g., very high elasticity and high resilience). Hence, analyses to acquire information
on their crystallinity are relatively rare, but can be important in some cases. One
of these cases is in the analysis of EPDM, both the conventional rubber and when
it is blended with polypropylene (PP) to produce a thermoplastic elastomer (TPE).
A commercial example of an EPDM/PP blend is marketed under the Santoprene™
tradename, and a DSC thermogram of a Santoprene™ TPE (grade 103–40) in
the ‘as-received’ state (i.e., without any annealing) is shown in Figure 3.9. In this
thermogram, the endotherm due to the semi-crystalline PP phase is very obvious
and has a melting point maximum at 154.27 °C.

2.00
Endothermic →

Wt. 6.29 mg Max. 154.27

Scan rate: 20.00 deg/min

Peak from: 109.9

to: 184.97
Onset: 139.42
mcal/sec

Cal/gram: 9.95

1.00

0.00
90.00 100.00 110.00 120.00 130.00 140.00 150.00 160.00 170.00 180.00
Temperature (ºC)

Figure 3.9 DSC thermogram of Santoprene™ 103-40 in the ‘as-received’ state.

Reproduced with permission from M.J. Forrest, Y. Davies and J. Davies in The
Rapra Collection of Infrared Spectra of Rubbers, Plastics and Thermoplastic
Elastomers, 3rd Edition, Rapra Technology Ltd, Shawbury, UK, 2007, p.74.
©2007, Rapra Technology Ltd [2]

Within the structure of EPDM rubber there can be crystalline regions due to the
ethylene component. The ratio of ethylene to propylene in EPDM can vary between
the different grades and, as the amount of ethylene increases, the degree of crystallinity
within the rubber increases and this increases its Tg. An application note published by
PerkinElmer [42] describes how DSC can be used to characterise EPDM elastomers.
In particular, it demonstrates how DSC can be used to show distinct and significant
differences between two EPDM TPE samples in terms of crystalline content and their

74
 Principal Components and Bulk Composition

Tg values. The results displayed show how the EPDM with the greatest concentration
of ethylene has a higher Tg, melting temperature and heat of melting; this is valuable
information because the thermo-physical properties of the EPDM will affect the end
use characteristics of the TPE.

Other rubbers that exhibit crystalline regions include members of the PU group.
Figures 3.10–3.12 show the DSC thermograms obtained on a PU rubber in three
distinct ways in order to characterise its morphology and obtain information on its
‘heat history’. The following information is present in these three DSC thermograms:

• Figure 3.10 – ‘As-received’ DSC thermogram of the PU rubber sample providing

information how its crystalline phase has been influenced by its processing (i.e.,
its heat history).

• Figure 3.11 – ‘Reheat’ DSC thermogram obtained after the PU rubber had been
slowly cooled from the melt state to obtain information on its true melting
characteristics.

• Figure 3.12 – Reheat DSC thermogram obtained on the PU rubber after it had
been quenched from the melt state to obtain information on the formation of its
crystalline phase (e.g., crystallisation temperature) and the subsequent melting
behaviour of this phase.

1.00
Weight: 5.62 mg
Scan rate: 20.00 deg/min Melting peaks
Peak from: 110.1
Max. 183.72
Endothermic →

to: 212.2
Onset: 166.82
Cal/gram: 4.91

0.50
mcal/sec

0.00
30.00 60.00 90.00 120.00 150.00 180.00 210.00 240.00

Temperature (ºC)

Figure 3.10 As-received DSC thermogram of a PU elastomer. Reproduced with

permission from M.J. Forrest in Chemical Characterisation of Polyurethanes,
Rapra Review Report No.108, Smithers Rapra, Shawbury, UK, 1999, p.23.
©2001, Smithers Rapra [43]

75
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

1.00

Weight: 5.62 mg
Scan rate: 20.00 deg/min Max. 184.96

Endothermic →
Peak from: 125.45
to: 214.34
Onset: 169.54
Cal/gram: 3.98

mcal/sec 0.50

0.00
30.00 60.00 90.00 120.00 150.00 180.00 210.00 240.00

Temperature (ºC)

Figure 3.11 Reheat DSC thermogram of the PU elastomer (Figure 3.10) after slow
cooling from the melt. Reproduced with permission from M.J. Forrest in
Chemical Characterisation of Polyurethanes, Rapra Review Report No.108,
Smithers Rapra, Shawbury, UK, 1999, p.23. ©2001, Smithers Rapra [43]

2.00

Max. 203.63
Endothermic →

1.00 Weight: 5.62 mg

Scan rate: 20.00 deg/min
Peak from: 132.95
to: 217.91
Onset: 194.97
mcal/sec

Crustallisation
peak Cal/gram: 5.63

0.00
30.00 60.00 90.00 120.00 150.00 180.00 210.00 240.00

Temperature (ºC)

Figure 3.12 Reheat DSC thermogram of the PU elastomer (Figure 3.10) after
quenching from the melt. Reproduced with permission from M.J. Forrest in
Chemical Characterisation of Polyurethanes, Rapra Review Report No.108,
Smithers Rapra, Shawbury, UK, 1999, p.23. ©2001, Smithers Rapra [43]

3.3.5 Analysis of Rubber Blends

When two or more rubbers are blended together, in order to obtain the best properties
from the resulting compound it is important to ensure that they are well mixed and
analytical techniques can assist with monitoring this. For example, as the degree of

76
 Principal Components and Bulk Composition

mixing of two rubbers increases, this will influence a number of physical properties,
one of which will be the Tg of the blend. Due to its high sensitivity to this property,
this change can usually be detected by DMA and an assessment as to the degree of
mixing can be obtained (Section 3.2).

Another important consideration when blending rubbers is to ensure that they are
compatible with one another. Theraftil and co-workers [44] assessed the compatibility
of blends of EPDM and chlorobutyl rubber (CIIR) in sulfur-cured compounds. The
research team used a range of physico-chemical analytical techniques in their study,
including tensile tests, DSC and DMA, and they also analysed the fracture surfaces
of test pieces using scanning electron microscopy (SEM). The effects of adding
chlorosulfonated PE as a compatibilisation agent on the phase separation of EPDM
and CIIR was also evaluated by mechanical testing (e.g., tensile testing and tear testing)
and by an examination of the fracture surface morphology by SEM.

DeCastro and co-workers [45] investigated the effect on dynamic mechanical

properties that results from varying the sequence that additives are incorporated
into a 50:50 blend of NR/BR. In their work, four addition sequences were used,
the curing conditions were kept constant, and the blend samples were analysed by
both DMA and SEM. Their work revealed that a selective partition of the additives
within each rubber phase, and that this resulted in a correlation between mechanical
performance and blend morphology.

It is common practice to blend rubbers with other polymers (e.g., plastics and resins)
in order to create a material with enhanced properties (e.g., toughness). One typical
example of such a blend is when rubber is blended with an epoxy resin. Nigram and
co-workers [46] used a combination of thermal techniques (DSC, TGA and DMA)
to study a blend of an epoxy cresol novolac resin that had been toughened using
carboxyl-terminated nitrile rubber (CTNR). The research team used these techniques
to obtain data on several properties of the blend. For example, they found that the
exothermic heat of reaction due to crosslinking of the resin using diaminodiphenyl
methane as the curing agent showed a decreasing trend as the rubber content of the
material increased, and an increase in rubber content was also found to provide
an increase in thermal stability and a decrease in the storage modulus. The loss
modulus and loss tangent values of the material were found to initially increase as
the temperature was increased, but then fell away, and addition of 10% of CTNR
was found to give the maximum values for both the loss modulus and tan δ.

Jana and co-workers [47] at the Indian Institute of Technology used TGA to evaluate
the thermal stability of blends of silicone rubber and LDPE that contained four
ethylene copolymer compatibilisers. The TGA data showed that the blends underwent
degradation in two stages and that it followed first-order reaction kinetics. The
activation energy at 10% degradation, determined using the Freeman–Carroll method,

77
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

was at a maximum for the 25:75 LDPE/silicone rubber blend, and this blend also
yielded the maximum half-life at 200 °C (812.5 days) calculated using the Flynn–Wall
method. Of the four compatibilisation agents evaluated, ethylene methyl acrylate
copolymer (EMA) was found to function the most effectively. In a related project by
members of the same research team [48] TGA and derivative TGA were used to study
compatibilised 50:50 blends of silicone rubber and LDPE that contained different
levels (0 to 15%) of EMA copolymer compatibiliser. The TGA and derivative TGA
data of the different blend samples showed that thermal degradation took place in
two stages, whereas pure LDPE showed a three-stage degradation, and a mixture of
only EMA and silicone rubber degraded in a single step. The research team calculated
the activation energies of degradation at the 10% point using the Freeman–Carroll
method and found that the maximum value (171 kJ/mol) was obtained with the blend
that contained 6% of EMA. This level of EMA was also found to give the maximum
half-life period at 200 °C using the Flynn–Wall method of calculation.

Amash and co-workers [49] investigated the solubility behaviour of various elastomeric
processing promoter additives in EPDM, NBR and PP using thermal analysis (e.g.,
DMA). The research team then prepared binary blends of rubber/PP-containing selected
promoter additives and examined their phase morphology, compatibilisation and
certain final properties that related to these characteristics. The results showed that the
promoter additives acted as effective homogenising agents and contributed to a significant
improvement in the dispersion and compatibility of the two blend components.

Blends are used extensively in the tyre industry (e.g., for tread compounds) and
researchers are continually looking at how improvements in performance can be
achieved. A research team from South Korea used DMA to characterise rubber
compounds that contained 5 phr of various phenolic resins, some of which were
modified resins that had been prepared specifically for this research work. Overall,
the investigation showed that because alkyl phenolic resins have a higher softening
point, phenolic resins that had been modified with either terpene or rosin showed
potential as new additives for improving the grip properties of tyres, although not all
the resins could enhance their rolling resistance, which is another important property.

Varghese and co-workers [50] used TGA to study the effect that a number of compounding
variations had on the thermal behaviour of NBR/ethylene-vinyl acetate (EVA) blends.
The following aspects of the formulation were altered during the course of their work:

• Ratio of NBR to EVA

• Filler type (e.g., silica, clay and CB)

• Amount of filler

• Cure system type (e.g., sulfur and peroxide)

78
 Principal Components and Bulk Composition

With respect to the initial decomposition temperature, it was found to increase as the
amount of NBR increased in the blend and the peroxide cure system provided the
highest value, which was thought to be due to the high-bond dissociation energy of
the carbon–carbon crosslinks. The peroxide-crosslinking system was also found to
exhibit the best retention of properties upon heat ageing and the addition of fillers
improved thermal stability. The data obtained by the research team on the blend
samples included the activation energies of degradation and mass loss at different
temperatures.

3.3.6 Degradation Studies and Changes in Chemical Structure

Aoyagi and Sano [51] analysed a fluorocarbon rubber shaft seal after 300,000 km
of service by micro-hardness and ATR–FTIR. The workers found that the hardness
of the lip seal increased due to both a structural change within the polymer and an
ageing reaction within the material. The ATR–FTIR results showed that the peak
width of the CF2 absorption band had increased due to the ageing and the structural
changes that had resulted. Simulation work to create the in-service environment
showed that it was the additives within the engine oil that were mainly responsible
for the degradation observed.

Under certain conditions, rubber products can fail in service due to a phenomenon
known as ‘blow-out’ whereby gaseous substances produced within the product can
escape in an explosive manner, causing it to rupture. A research team at Sejong
University [52] used a microwave oven, along with GC–MS analysis of the volatile
substances generated, to investigate the causes of blow-out from CB-filled NR and SBR
samples. The species that were identified by GC–MS in the blow-out gas from the NR
sample included 1-methyl-4-(1-methylethenyl)cyclohexene, and 4-vinylcyclohexene
and styrene were detected from the SBR sample. These three substances were
considered to be the thermal degradation products of the NR and SBR, respectively.
Upon examination of the burst region of the sample, it was not possible to find
evidence of any organic additives that had been used in the compounding stage,
but a large number of hydrocarbon substances were detected for both samples. The
research team concluded that the principal cause of blow-out was the decomposition
of the polymer, which caused cavities to form in the samples, thereby reducing the
resistance to the increased internal pressure.

Pappa and co-workers [53] analysed the volatile products that result from the
thermal decomposition of NBR/PVC blends using TGA–MS. Such blends are used
to manufacture products, such as forest fire hoses, and experimental compounds
containing various inorganic fillers (e.g., magnesium hydroxide, organic-modified
kaolin and a nanoclay material) were prepared to determine how they affected the
nature of the decomposition products. The TGA–MS data revealed no significant

79
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

differences between these experimental compounds and an unfilled NBR/PVC control

compound, apart from the absence of HCl in the gaseous products from the blend
that contained the magnesium hydroxide filler.

3.4 Determination of Plasticisers or Process Oils

3.4.1 Quantification

Mention has been made in Section 3.2 of the isolation of plasticisers from a rubber
matrix by solvent extraction. In that case, the objective was to clean up the polymer
sample for further work (e.g., FTIR analysis for polymer identification). However, it
can be an end in its own right and it is the subject of the International Organization
for Standardization (ISO) standard, ISO 1407, which shows how it is possible to
obtain accurate quantification of the plasticiser present in a rubber compound.

Table 3.6 Solubility parameters of common solvents

Solvent Solubility parameter (cal1/2 cm−3/2)
n-Hexane 7.0
Diethyl ether 7.4
Cyclohexane 8.2
Carbon tetrachloride 8.6
Xylene 8.8
Toluene 8.9
THF 9.1
Chloroform 9.3
Methyl ethyl ketone 9.3
Trichloroethane 9.6
Dichloromethane 9.7
Cyclohexanone 9.7
N,N-diethylacetamide 9.9
Acetone 9.9
o-Dichlorobenzene 10.0
Carbon disulfide 10.0
N,N-diethylformamide 10.6
n-Butanol 11.4
Cyclohexanol 11.4
ACN 11.9
N,N-Dimethylformamide 12.1
Ethanol 12.7
Dimethylsulfoxide 12.9
Methanol 14.5
Ethylene glycol 14.6
Water 23.4
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.56.
©2001, Smithers Rapra [54]

80
 Principal Components and Bulk Composition

The choice of solvent is an important factor in obtaining accurate quantification of

the plasticiser or oil and a good-quality extract for identification work (Section 3.4.2).
Ideally a solvent should be used that is a good solvent for the plasticiser/oil to
facilitate efficient removal, but a poor solvent for the vulcanised rubber polymer to
avoid a high proportion of low-MW polymer being extracted. In reality, due to the
similarities in the solubility parameter values of the plasticiser/oil and the polymer
(essential if they are to be compatible) a compromise often has to be reached and
some contribution from the polymer in the extract cannot be avoided completely,
particularly as vulcanised rubbers are only lightly crosslinked systems.

The solubility parameters of a range of rubbers and common solvents are shown
in Tables 3.6 and 3.7, respectively. In general, the closer the solubility parameter
values of a rubber and solvent are, the more compatible they will be and the better
the solvent will be for the rubber. These tables can also be used, therefore, to select a
solvent to dissolve an unvulcanised rubber to create a solution (e.g., to enable a film
to be cast for a transmission IR analysis).

Table 3.7 Solubility parameters of typical rubbers

Rubber type Solubility parameter (cal1/2 cm−3/2)
NR 7.9–8.5
Polyisoprene 7.9–8.5
SBR (4 to 40% styrene content) 8.1–8.6
Butadiene rubber 8.0–8.6
Butyl rubber 7.5–8.0
EPDM 7.5–8.6
Chloroprene rubber 8.1–9.4
NBR (18 to 30% ACN) 8.7–9.3
NBR (40% ACN) 10.4–10.5
Silicone rubbers 7.0–11.0
Polysulfide rubbers 9.0–10.0
PU rubbers 9.8–10.3
PVC (plastic used in blends with NBR) 8.5–11.0
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.55.
©2001, Smithers Rapra [55]

81
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Because they are viscous liquids with relatively low MW and are volatile within
the temperature range 150–300 °C, the TGA technique can usually provide a
good quantification of plasticiser or process oil content, particularly with synthetic
plasticisers, such as phthalates and adipates. These substances are lost from a rubber
sample during a TGA experiment by volatilisation and this is true whether or not an
oxidising atmosphere is used in the experiment.

3.4.2 Identification

Having quantified the plasticiser/oil present by solvent extraction, it is then possible

to obtain an identification by use of transmission FTIR spectroscopy. To assist with
this process, there are extensive IR spectral databases available to the polymer analyst
[42]. For example, the data shown in the spectra in Figures 3.13 and 3.14 are taken
from standard spectra present in the in-house collection at Smithers Rapra and are
consistent for the plasticisers dioctyl phthalate and dibutyl sebacate, respectively. Also,
where the plasticisers have distinct IR spectra (e.g., phthalates and sulfonamide) it is
possible to detect the presence of a blend of plasticisers. If quantification of the blend
composition is required, though, this is usually easier to achieve by high-performance
LC or LC–MS, than it is by FTIR.

The use of TGA to quantify the amount of a plasticiser or oil in a rubber sample has
been mentioned in Section 3.4.1. It is also possible to use the technique to obtain
some qualitative information. For example, by the use of standardised conditions
and control samples, the TGA derivative peak temperature at which the maximum
rate of volatilisation occurs can be used to obtain some qualitative information on
the type of plasticiser present in a sample. For example, under the same analytical
conditions, dibutyl phthalate (DBP) will have a derivative peak temperature of
220 °C, whereas di-(2-ethylhexyl)phthalate will have a peak at 265 °C. The chemical
structure of the plasticiser also has an influence, with di-(2-ethylhexyl)adipate and
di-(2-ethylhexyl) sebacate having peak temperatures of 250 and 280 °C, respectively.

Some of the mineral process oils and plasticisers that have higher MW (e.g., polymeric
plasticisers) can give problems because the weight loss event due to their volatilisation
from the sample can merge with the pyrolysis weight loss event of the polymer.
However, the relative positions of the derivative peak temperature, and the temperature
at which volatilisation begins, can continue to be used to obtain comparative data.

82
 Principal Components and Bulk Composition

Transmission IR of dioctyl phthalate 'as-received'

iS10-01402
120

115

110

105
Transmittance (%)

100

95

90

85

80

75

70

65

60

55

50

45

40

4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600

Wavenumbers (cm -1)

Smithers Rapra – Collection time: Monday September 18 15:29:21 2017 (GMT+01:00)

Figure 3.13 Transmission FTIR spectrum of dioctyl phthalate

Transmission IR of dibutyl sebacate 'as-received'

iS10-01403
125
120
115
110
105
100
95
Transmittance (%)

90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600

Wavenumbers (cm -1)

Smithers Rapra – Collection time: Monday September 18 15:37:21 2017 (GMT+01:00)

Figure 3.14 Transmission FTIR spectrum of dibutyl sebacate

83
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Modifying the conditions under which a TGA experiment is performed can improve
the quality of the data obtained. Knappe [57] described how operating a TGA under
vacuum can improve the separation between plasticisers and the polymer component
in a rubber sample, and also how coupling a TGA to an FTIR instrument can assist
in the identification of additives such as plasticisers. Interfacing a TGA to an FTIR
is covered in more detail in Section 3.8.

As mentioned in Section 3.2, rubber compounds containing polymers that lose small
molecules by heat degradation (e.g., the HCl lost by CR) give more complicated
TGA traces in which the early weight loss events overlap. These samples require a
solvent extraction step to determine their plasticiser level. It is also harder to obtain
qualitative information on the type of plasticiser; often an IR analysis of the solvent
extract is required.

3.4.3 Determination of Polyaromatic Hydrocarbons in Hydrocarbon Oils

and Rubber Products

For many years it was a common and widespread practice to use aromatic mineral
oils, also called ‘extender oils’, in large tonnage products like passenger tyres. In fact,
the incorporation of relatively large amounts of these types of mineral oil in tyres
was beneficial to certain prime properties of the final product (e.g., road holding
and good grip).

However, around 20 years ago, health and safety concerns began to grow over the
polyaromatic hydrocarbons (PAH) present in these types of mineral oils [e.g., the
distillate aromatic (DAE) types]. These concerns included both their potential threat
to the environment, as a result of tyre rubber being deposited onto road surfaces
and substances leaching from it, and to the workers who came into contact with
them in industry. This pressure on the industry to substitute these types of aromatic
oils with safer mineral oils led to the development of low-PAH content extender oils
(e.g., mild-extraction solvate types) for use in tyres and other rubber products. The
European Commission (EC) also published a regulation addressing the use of these
oils: Directive (EC) 2005/69. This Directive came into force on 1st January 2010 and
prohibited the use of DAE-type oils.

Another approach to the regulatory problems associated with highly-aromatic oils was
to look for renewable alternatives. One such alternative is cardanol (m-pentadecenyl
phenol), a cheap and abundant byproduct of the cashew nut industry, which has
been shown to be capable of acting as a plasticiser and multifunctional additive in
polymers. Researchers in India [58] carried out comparative analyses using a range of
techniques (e.g., thermal, physical and microscopic) on NR that had been chemically

84
 Principal Components and Bulk Composition

grafted with cardonol (CGNR) and an oil-plasticised NR compound. The compound

based on CGNR was found to be superior to the NR compound in the following ways:

• Physico-mechanical properties

• Crosslink density

• Bound rubber content

• Lower Payne effect

• Better and more uniform dispersion of CB

Another possible source of PAH in rubbers is CB, a material that can also be used
in other products, such as plastics. As a consequence, the EC published another
regulation in 2013 [59] which restricted PAH in several consumer articles that
were manufactured using both rubber and plastics materials. The presence of PAH
in extender oils and tyres has also been addressed by the Registration, Evaluation,
Authorisation and Restriction of CHemicals (REACH) regulation [60]. REACH, in
Annex XVII (i.e., EC 1907/2006 XVII amendment), placed a specific restriction on
the amount of benzo-alpha-pyrene (a common PAH compound) that could be present
(1 mg/kg) and a limit (10 mg/kg) on eight specified PAH compounds, one of which
is benzo-alpha-pyrene. It also designated ISO 21461 as the method that should be
used to determine the combined level of these eight PAH compounds.

Because of the specific concern over extender oils and tyres, and the general concern
over PAH, work was carried out to develop analytical methods that were capable
of accurately determining the amount of specific PAH compounds in products and
in determining the overall level of PAH compounds. One example of a specific
method is the method set by the American Society for Testing of Materials (ASTM)
for quantifying benzo-alpha-pyrene in CB [61]. Important general methods for the
determination of PAH include:

• IP 346 – total amount of PAH in extender oils.

• ISO 21461 – NMR-based method for determining the PAH level of tyres (eight
PAH compounds are targeted).

• German ZEK method (16 PAH compounds).

• US Environmental Protection Agency (18 PAH compounds).

With regard to ISO 21461, the NMR instrument determines the level of PAH in an
extender oil or the extract of a rubber sample by measuring the amount of Bay protons
that are present. According to REACH (EC) 552/2009, extender oils and tyres are

85
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

compliant with the restriction on PAH (re: the limit introduced on 1st January 2010)
if they do not exceed 0.35 Bay proton as measured and calculated by ISO 21461.

Hamm and co-workers [62] investigated the extraction and migration behaviour of
PAH from cured, CB-filled rubber compounds. The research team used standard PI-
based formulations to develop and validate a GC–MS method for the identification
and quantification of PAH compounds that had migrated into various aqueous media.
The objective was to develop a method that could be used to assess the exposure to
PAH compounds from rubber products.

3.5 Determination of Carbon Black

CB is by far the most commonly used filler for rubber compounds and has been in
constant use in formulations since just before the First World War. Although CB can be
added to rubber compounds to make them less expensive by reducing the proportion
of the more expensive polymer component (e.g., in highly ‘extended’ EPM and EPDM
compounds), it is more often added to improve the physical properties (e.g., tensile
strength and tear strength) of a rubber compound, something which few other filler
materials can do. In common with other fillers, it also increases the hardness and
viscosity of a compound.

Historically, there have been three main types of CB. Each type is designated by
reference to the manufacturing technique used to produce it from oil, namely:

• Furnace black
• Thermal black
• Channel black

The first of these is the most common type of CB (95% of the market). It is available
in a large range of grades (more than 20), and is used in a wide range of rubber
compounds. The second type is available in only a few grades and tends to be used
in a limited range of compounds (e.g., fluorocarbon rubbers). The third type is rarely
seen today, although it has some specific uses due to its relatively high purity (e.g.,
food-grade rubbers for approval by the US Food and Drug Administration).

In addition to the three types shown above, two specialist CB exist. These are
lampblacks (which have been developed by the Chinese for ink and lacquer
formulations) and acetylene black (which is conductive and used to formulate
conducting rubber compounds). As mentioned in Section 3.5.1, graphite can also be
added to increase conductivity and carbon fibres can be present in some compounds
to alter physical properties. Another carbonaceous material that may become more

86
 Principal Components and Bulk Composition

common in rubber compounds in the future are carbon nanotubes, which are showing
some promise in improving the electrical conductivity of plastics [63].

3.5.1 Quantification of Carbon Black in Rubber Vulcanisates

It is possible to quantify the amount of black in a vulcanisate by using a pyrolysis

approach in a tube furnace in a way that is analogous to that described in Section 2.1
for the determination of halogens. The initial, non-oxidative pyrolysis stage involves
heating an accurately weighed sample in a stream of nitrogen and then weighing
the resulting residue. The final stage involves heating this residue in an oxidising
atmosphere (either air or oxygen) to oxidise the CB to carbon dioxide (CO2). The final
inorganic residue is then weighed and the amount of black in the sample calculated
using Equation 3.6:

A-B × 100
Amount of CB (%) = (3.6)
original weight

where ‘A’ is the residue after non-oxidative pyrolysis stage and ‘B’ is the final residue.

Carrying out the above experiment in a tube furnace has the advantage that relatively
large (e.g., 1–2 g) samples can be used, and so a relatively accurate result is obtained.

It is more usual, however, to use the automated and sensitive technique of TGA and
this instrument can determine the amount of CB in a sample, along with the amount of
the other principal ingredients (e.g., plasticiser and polymer) in one seamless operation
with automatic gas switching and data collection (Sections 3.8). Compared with the
tube furnace, small samples are analysed (1–10 mg), but this has the advantage that
the technique can be used on small samples, such as contaminants or, if the sample
is large enough, in a virtually non-destructive way.

A TGA trace of a NBR formulation showing the quantification of carbonaceous

residue and CB is shown in Figure 3.15.

Its inherent advantages mean that TGA is an excellent technique for the quantification
of CB in rubber compounds, and the results obtained have a reasonable degree of
accuracy, as shown by the data generated using a range of CB in different rubber
compounds by Pautrat and co-workers [64]. In general, the research team found that,

87
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

in all cases, the amount of CB recorded by the TGA was slightly above (≈5%) the
quantity that had been added to the rubber compounds at the mixing stage.

120

6.164% Plasticiser
(0.7557 mg) 55.89% Polymer
100
(6.853 mg)

80
Weight (%)

60
3.526% Carbonaceous residue
(0.4324 mg)
40

30.69% CB
(3.763 mg)
20

0
0 20 40 60 80 100
Time (min)

Figure 3.15 TGA trace of NBR. Reproduced with permission from M.J. Forrest
in Rubber Analysis – Polymers, Compounds and Products, Rapra Review Report
No.139, Smithers Rapra, Shawbury, UK, 2001, p.16. ©2001, Smithers Rapra [5]

Although TGA is useful for quantifying the amount of CB, it has a limited capability
when it comes to distinguishing between the different types of CB. This is because there
is no clear and definitive relationship between particle size and/or particle structure
and the oxidation temperature of a CB. There is, however, a loose relationship in
that the higher surface areas of the small particle size blacks (e.g., super abrasion
furnace) results in lower oxidation temperatures than the larger particle size blacks.
The temperature at which a CB oxidises also varies between its virgin form and

88
 Principal Components and Bulk Composition

when it has been incorporated into a rubber compound. For example, the oxidation
temperature of a virgin semi-reinforcing furnace black has been reported as being
≈670 °C, as opposed to ≈580 °C when it is present in a cured butyl rubber compound
[65]. Such a result shows that it is important to have access to a number of appropriate
rubber compound reference standards.

Table 3.8 Indicative-TGA oxidation temperatures of a selection of CB when

present in rubber compounds
Type of CB Indicative-oxidation temperature (°C)
Medium thermal 600
Semi-reinforcing furnace 575
High-abrasion furnace 550
Medium process channel 520
Reproduced with permission from M.J. Forrest in Principles and Applications of Thermal
Analysis, Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.196. ©2008,
Blackwell Publishing [66]

For reference, some indications of the oxidation temperatures for a selected range of
CB, when they are present in a rubber compound, is provided in Table 3.8.

A number of workers, for example Pautrat and co-workers [64] and Maurer [67],
used the time taken for 15% of the weight of the CB to oxidise (T15) to investigate if
the type of CB in a rubber compound could be ascertained by TGA. Again, a loose
relationship has been found but no absolute correlation exists, and some of the
problems that are encountered in assigning CB type by TGA can be summarised as
shown below:

• Blacks vary in structure as well as particle size and this can blur the distinction
between different grades resulting in overlap within a TGA experiment.

• Analytical conditions (e.g., temperature ramp) can affect the data.

• The cure system (i.e., sulfur or peroxide) used in the compound can influence the
data.

• The polymer can affect the data if it generates a carbonaceous residue that overlaps
with the black weight loss.

89
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The last point is important as a number of commercially important rubbers (e.g.,

halogenated, nitrile and PU) produce carbonaceous residues during the nitrogen
atmosphere region and care must be taken to ensure that these do not affect the
quantification of CB in the air atmosphere region. Use of a high-resolution TGA
instrument may assist in this because it should improve the separation of the two
weight loss events. However, one must be beware of the difficulties that can be
encountered with the operation of certain high-resolution TGA instruments because an
incorrect choice of programme can cause greater problems, as described in Section 3.8.

As mentioned previously, in addition to CB, rubbers can contain other types of

carbonaceous fillers, for example:

• Acetylene black

• Graphite

• Carbon fibres

The first two products are used to create a rubber that is conductive and the third,
which is found relatively rarely in rubber compounds, is a filler to improve strength
and increase hardness. All three products are purer forms of carbon than CB so they
have higher oxidation temperatures. CB, which contains various trace substances
(e.g., sulfur compounds) that originate from the oil which it is derived from, usually
oxidises in rubber compounds over the range 500–600 °C, whereas the three purer
forms of carbon all oxidise ≈700 °C in rubber compounds, and this additional stability
makes it possible to differentiate between the two classes in a TGA experiment. In
both cases, the temperatures cited are for a TGA programme in which the temperature
of the rubber sample is increased at a rate of 20 °C/min.

3.5.2 Dispersion of Carbon Black Within Rubber Products

To achieve optimum properties, it is important that the CB within a rubber is uniformly

dispersed. There are a number of tests that have been developed to assess the degree
of dispersion.

3.5.2.1 Cabot Dispersion Test

This test [68] involves the cold-stage microtoming of thin (≈2-µm thick) sections from
the rubber, which are then immersed in xylene and examined between glass slides
using a light transmission microscope fitted with a Cabot graticule. The degree of
dispersion is graded using a Cabot Dispersion Classification Chart.

90
 Principal Components and Bulk Composition

3.5.2.2 Cut Surface and Torn Surface Methods

The surfaces of the rubber are viewed using incident illumination, thereby avoiding
the need to prepare microtome sections, and a degree of automation is provided by
equipment such as the Optigrade Dispergrader. The torn surface method has been
described by Sweitzer and co-workers [69] and Stumpe and Railsback [70] and is,
in effect, a method for the determination of the agglomeration of CB within the
sample. Grading is possible and a scale of 1 (very poor dispersion) to 10 (excellent
dispersion) is used.

3.5.2.3 Transmission Electron Microscopy

Very thin sections are required (<100 nm) and the small area examined makes it
possible to evaluate the dispersion of black within phases of polymer blends. Care
has to be taken though that the small area analysed does not lead to unrepresentative
results. A detailed description of the microscopy of rubber samples has been provided
by Kruse [71].

3.5.2.4 Online Measurement during Processing

Le and co-workers [72] used an online measurement method to investigate the

effect of curing additives on the dispersion kinetics of CB within SBR compounds.
The online method used involved the measurement of electrical conductance during
mixing and the data obtained showed that curing agents such as stearic acid and
diphenyl guanidine (DPG) greatly accelerated the dispersion of CB within the rubber
compound. The research team postulated that stearic acid and DPG might alter filler–
filler interactions and that this subsequently accelerated the dispersion process. Their
work also revealed that ZnO was not capable of improving dispersion because of its
limited amount of interaction with the CB, that sulfur and N-cyclohexyl benzothiazole-
2-sulfenamide decelerated the dispersion process and that, as the amount of styrene in
the SBR increased, the influence that the curing additives had on dispersion decreased.

3.5.3 Isolation of Carbon Black from a Vulcanisate

Historically, there are two main routes for the recovery of CB from rubber vulcanisates.

3.5.3.1 Digestion of the Matrix Followed by Filtration

In this method, the polymer in the rubber vulcanisate is destroyed and solubilised
by a chemical such as a strong (e.g., nitric) acid in the presence of an organic solvent

91
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

(e.g., nitrobenzene) and heat. The CB is then recovered by filtering the mixture. The
drawback in the method is the difficulty of achieving a sufficient degree of degradation,
especially with the more chemically resistant rubbers.

3.5.3.2 Isolation by Non-oxidative Pyrolysis

In a non-oxidising atmosphere, heating of the rubber compound to ≈500 °C results

in the pyrolysis and loss of the polymer fraction as volatile products. This leaves a
residue comprising CB and inorganic species. CB is then isolated by an acid wash,
a water wash and then low-temperature drying (≈105 °C), with light grinding to
break up aggregates.

3.5.4 Surface Area Tests on Recovered Black

All three test methods to obtain total surface area (Section 3.6.2) have been applied
to CB that has been recovered from vulcanisates. The results obtained suggest that
with the Brunauer–Emmett–Teller method (BET) the type of rubber that the black has
been recovered from can influence the data (Table 3.9). The results obtained using
the cetyltrimethylammonium bromide (CTAB) method were more independent of
polymer type (Table 3.10).

Table 3.9 Total surface area (m2/g) of CB determined by the BET method
CB type Original black Recovered from Recovered from Recovered from
NR SBR SBR/BR
N110 144.6 132.5 127.5 123.9
N220 124.2 124.3 112.9 110.0
N330 82.9 86.4 78.2 74.5
Adapted from data published in European Rubber Journal

Table 3.10 Total surface area (m2/g) of CB determined by the CTAB method
CB type Original black Recovered from Recovered from Recovered from
NR SBR SBR/BR
N110 103.9 98.2 99.1 100.1
N220 95.9 96.1 92.0 91.0
N330 65.0 71.5 62.0 68.9
Adapted from data published in European Rubber Journal

92
 Principal Components and Bulk Composition

The DBP method to obtain the external surface area has been carried out on CB
samples recovered from vulcanised rubber using the preferred pyrolysis route and,
as with the total surface area tests, reasonable results have been obtained enabling
the type of black used to be determined.

3.6 Characterisation of Carbon Black

CB is an extremely pure form of carbon which consists of very small, mostly spherical
particles which fuse together in clusters referred to as ‘aggregates’. The latter group
together as ‘agglomerates’ which break up during the mixing process. The aggregates
tend to remain intact in the rubber matrix and the type of aggregate defines the
structure of a CB; the higher the structure the greater the number of particles in an
aggregate. The greater the number of particles in an aggregate, the larger will be the
volume within it that is not accessible to rubber molecules.

The properties that define a particular CB are as follows:

• Sphere size

• Structure (aggregate size and shape)

• Total surface area

• Surface area available to rubber molecules.

3.6.1 Particulate and Aggregate Size

The most popular technique for the examination of the particles of CB is SEM.
However, other techniques have been used, such as X-ray diffraction (XRD),
transmission electron microscopy, atomic force microscopy and scanning tunneling
microscopy, and these feature in a review of the subject [73]. These last two techniques
can be used to examine the surface structure of CB [74]. In addition to the examination
of individual particles, high-resolution electron microscopy is a useful tool for the
studying particle aggregates [75].

A team drawn from industry and academia in Japan [76] used contrast variation
small-angle neutron scattering (SANS) to study the surface of CB dispersed within SBR
samples. The team prepared the SBR samples by swelling them in solvents (mixtures
of toluene and deuterated toluene) that had different scattering length densities.
Analysis of the swollen vulcanisates by SANS revealed that there was an adsorption
layer of SBR, of ≈10 nm in thickness, on the surface of the CB, and the results also

93
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

provided information on the CB aggregate structure. One of this group of researchers,

Takenaka, also published a review of the use of SANS to analyse the structure of
sulfur-crosslinked rubber and used the technique to characterise the structure of PI
rubber and SBR, with a non-uniform structure being observed in both cases [77].

3.6.2 Total Surface Area

The total surface area of a CB sample (i.e., the area that can be accessed by rubber
molecules and that within pores and between particles) provides an indication of
particle size. There are three main methods for the determination of total surface area.

3.6.2.1 Brunauer–Emmett–Teller Method (Nitrogen Adsorption)

Named after Brunauer and co-workers, who developed it in 1938, this method is
described in ISO 4652 and is carried out using a Ni-Count-1 instrument. It also
features in ASTM D3037, in which there are four accepted procedures, one of which
uses the Ni-Count-1.

3.6.2.2 Iodine Adsorption

This method uses simple laboratory equipment and produces results that correlate
well with the nitrogen adsorption method. The procedure to be followed is described
in ISO 1304 (ASTM D1510). Reinforcing grades of CB have iodine adsorption
numbers typically in the range 70–160 mg/g whereas semi-reinforcing grades are in
the range 30–45 mg/g.

3.6.2.3 Cetyltrimethylammonium Bromide Adsorption

This method, in which a large molecule is used as an adsorption species, has been
described by Lamond and Gillingham [78]. It is thought to be more accurate than
the other two methods because it is less influenced by the chemical nature of blacks
and is less sensitive to particle porosity. A similar procedure to that suggested by
Lamond and Gillingham is presented in ASTM D3765. Reinforcing grades of CB
have CTAB values in the range 80–140 m2/g, whereas semi-reinforcing grades are in
the range 30–45 m2/g.

94
 Principal Components and Bulk Composition

3.6.3 External Surface Area

This test enables a measurement of the amount of structure in a CB to be established.

The DBP test is described in ASTM D2414 and involves adding phthalate from a
burette into a weighed portion of CB which is stirred constantly. As DBP is added,
the CB powder changes to a semi-liquid mass with an increase in torque. The torque
peaks at the limit of DBP absorption. The result is expressed as ml of DBP per 100 g
of CB. Low-structure CB have values in the range 60–80 ml/100 g, with high-structure
types in excess of 120 ml/100 g.

3.6.4 Determination of Polyaromatic Hydrocarbons in Carbon Black

Due to their method of manufacture (i.e., combustion of mineral oil) it is possible

for CB to contain some PAH compounds. A statement published by the International
Carbon Black Association (ICBA) stated that, based on laboratory assessments
involving vigorous Soxhlet extraction using solvents such as toluene, most grades of
CB typically have PAH levels not exceeding 0.1%.

The ICBA contributed to the development of an ASTM method [61] for the
determination of a particular PAH compound, benzo-alpha-pyrene, in CB. This
method, and other analytical methods to determine PAH compounds in extender oils
and final rubber products, such as tyres, have been covered in Section 3.4.3.

3.6.5 Miscellaneous Analytical Techniques

A number of analytical techniques have been used to characterise CB, and a

representative selection of these, along with the references in which information can
be sourced, are shown in Table 3.11.

3.7 Determination of Inorganic Fillers

In addition to being extremely useful for the determination of polymers, plasticiser/

oil and CB, TGA is also a very rapid and convenient technique for the quantification
of inorganic fillers, such as barium sulfate (BaSO4) (i.e., barytes), silica, silicates (e.g.,
clays) and calcium carbonate (CaCO3), in rubber compounds and products.

95
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 3.11 Examples of analytical techniques used to characterise CB

Technique Application References
IGC Surface energies and Wang, Wolff and Donnet [79]; Wang
thermodynamic parameters and Wolff [80]
Neutron scattering Particle structure Hjelm and co-workers [81]
Raman spectroscopy Particle microstructure Gruber, Zerda and Gerspacher [82];
and X-ray scattering Gruber, Zerda and Gerspacher [83];
and Gerspacher and Lasinger [84]
GC–MS, secondary Surface chemistry Ayala and co-workers [85]
ion mass spectrometry,
X-ray photoelectron
spectroscopy and IGC
IGC: Inverse gas chromatography
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.22.
©2001, Smithers Rapra [86]

One limitation of the TGA technique is that the total inorganic content of the rubber
is obtained, with no indication of the relative proportion if a blend of inorganic
fillers or a mixture of inorganic filler and other inorganic additives is present. Most
rubber compounds contain ZnO as part of the cure system (sulfur systems) or as an
acid acceptor (peroxide cure systems), so it is common practice to complement the
TGA data with a semi-quantitative elemental technique, such as X-ray fluorescence
spectroscopy (XRF), in order to obtain qualitative as well as quantitative data.
Although it is possible to use the ash obtained by the TGA experiment for this
analysis, it is more common to use the ash that has been obtained by a qualitative or
quantitative ashing process, such as the one described below.

CaCO3 can be detected and quantified more effectively by TGA than most of the
other common inorganic fillers, which are stable up to 1,000 °C, because of the
quantitative way that it decomposes upon heating. It will undergo a characteristic
quantitative degradation at around 680–700 °C (at a heating rate of 20 °C/min)
into CO2 and calcium oxide, which remains as a stable residue up to 1,000 °C. It
is, therefore, possible to both identify the presence of CaCO3 in a sample by this
characteristic weight loss, and to quantify it because the CO2 weight loss represents
44% of the CaCO3 by mass.

The value of the residue that remains will also show if any other inorganic compounds
are present in the sample. Information can then be obtained on these using the
analytical techniques (e.g., XRF) described elsewhere in this section.

96
 Principal Components and Bulk Composition

The other technique commonly used for the quantification and isolation of inorganic
substances from a rubber sample is furnace ashing. One common method used for
this process is described in ISO 247 and two procedures are provided. The first, dry
ashing, may be unsuitable for rubbers containing halogens due to the loss of volatile
halides, such as zinc chloride, formed by the reaction of ZnO with the HCl liberated
from the polymer. The second, acid ashing, is not recommended for raw rubbery
polymers. There are two temperature options in the standard: 550 and 950 °C. Care
has to be taken with the higher temperature if knowledge of the types of inorganic
compounds present is not available. For example, if CaCO3 is present, as mentioned
above, it will breakdown at ≈700 °C. With this in mind, 550 °C is the more common
ashing temperature used in commercial laboratories.

Although it is rarely used for this purpose because of the practical and time-efficient
benefits of TGA and standardised ashing techniques, it is possible to use the tube
furnace approach if relatively large samples (e.g., 2–5 g) are available. In order to
separate the inorganic species from the rubber sample, the tube furnace is employed
using a similar set of operating conditions to the TGA. Values for total organics, total
carbonaceous material and total inorganics are obtained by the experiment and the
isolated ash can be analysed to obtain qualitative and quantitative information on
the inorganic compounds present as described below.

Once an ash has been obtained, by either TGA, tube furnace, or a standard ashing
process, qualitative information on the inorganic compounds present can be obtained
by a variety of analytical techniques. One of the most common and convenient is
FTIR, and it is standard practice to use a liquid paraffin mull technique. If only a
small amount of ash is available and the quantity of ash is too small to use the paraffin
mull technique effectively, other FTIR techniques can be used such as ‘single bounce’
ATR or FTIR microscopy. This eventuality is very rare in the case of inorganic filler
additives which by nature are bulk additives. Hence, these alternatives are usually
required only for additives such as inorganic pigments (Chapter 4).

The use of FTIR alone to identify inorganic fillers in a sample ash can be problematic,
particularly if a mixture is present, so it is often necessary to compliment this work
with an elemental technique. For qualitative or semi-quantitative work, XFR is
very useful in providing the elements present which, in conjunction with the FTIR
spectrum, will usually enable specific compounds to be identified. If it is suspected that
a compound, such as a silicate is present, use of XRD can often provide an assignment
of the particular type by virtue of its characteristic crystalline fragmentation pattern.
If quantification data are obtained on a particular element, or if it is thought that
a semi-quantitative multi-element scan would be useful, then inductively coupled
plasma can be employed.

97
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The use of FTIR to assist in the identification of inorganic fillers in a rubber sample,
via analysis of its ash, is mentioned above and some of the characteristic, diagnostic
IR absorption bands that can be employed in this task are shown in Table 3.12.

Table 3.12 Examples of characteristic IR absorption bands for some typical

inorganic fillers
Inorganic filler Diagnostic IR absorption bands
Silica Broad band at 950–1,330 cm−1 peaking at 1,050–1,100 cm−1
Silicates Broad band at 850–1,300 cm−1 peaking at 950–1,100 cm−1
CaCO3 1,420, 870 and 710 cm−1
BaSO4 1,080 and 610 cm−1

3.8 Determination of the Bulk Composition of a Rubber

Sections 3.2–3.7 have shown how analytical work can be carried out on a rubber
sample to obtain information on the principal components which make up its bulk
composition. In addition to obtaining this information, it is also important in a number
of cases (e.g., checking a formulation against a specification) to obtain information
on the overall composition of a rubber sample (i.e., the proportions of principal
components). This kind of information can be referred to as the ‘bulk composition’
because it addresses only the major components and does not provide information on
the many types of additives that can be added in relatively small quantities; obtaining
this type of information is the subject of Chapter 4.

It is possible to obtain this information by combining the qualitative and quantitative

data obtained using particular techniques and approaches in Sections 3.2–3.7.
However, it is often the case that a simplified approach is preferred, and one of the
most useful and effective techniques for obtaining bulk compositional data on a rubber
sample is TGA. This is because the technique has a number of inherent advantages:

• It only requires a small sample (e.g., 10 mg).

• It is relatively quick – a complete analysis can be completed within 90 min.

• It is relatively accurate (e.g., ±0.5% w/w).

• It is relatively sensitive (e.g., it can indicate the presence of two polymers or two
types of CB).

98
 Principal Components and Bulk Composition

A typical TGA analysis will involve two stages:

• Stage 1 – The sample is heated from 40 °C under a nitrogen atmosphere at a

heating rate of 20 °C/min to 550 °C and held at that temperature for 10 min to
ensure that constant weight is achieved.

• Stage 2 – The temperature is reduced from 550 to 300 °C, the atmosphere changed
to air, and then the sample heated to 850 °C.

Because of its effectiveness and popularity for a range of important applications,

including quality control, failure diagnosis, and reverse engineering, the use of TGA
to determine bulk composition has been incorporated into international standards.
An example is ISO 9924-1, which describes the use of TGA for the determination
of the composition of butadiene, ethylene propylene, butyl, isoprene and SBR. The
technique also lends itself to being standardised because, apart from differences in
optional features (e.g., high-resolution programmes), the majority of the commercial
instruments use the same fundamental design.

Because of the small sample size that is used, and the practical options that are
available (e.g., heating rate), it is best to standardise the analytical conditions to
ensure that meaningful comparisons between samples can be made, even if there is
a time delay in their analysis.

A representative TGA trace of a rubber sample, in this case a NBR, recorded using
the analysis programme shown above, is shown in Figure 3.16.

The data in Figure 3.16, with both the weight loss and the derivative weight loss
curves displayed, show the main weight loss events present in such traces. From this
TGA trace it is possible to obtain the following bulk compositional information:

• Plasticiser content – 5.7%

• Polymer content – 67.3%

• Carbonaceous residue* – 3.2%

• CB – 10.1%

• Inorganic material – 13.7%

*Polymers that contain atoms such as nitrogen, oxygen, sulfur or a halogen

produce a carbonaceous residue during reactions such as cyclisation during stage
1 (see the programme above). This residue oxidises to CO2 during stage 2 (see
above).

99
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Plasticiser = 5.7% 2.761

108.1
2
100
0
90
-2

Derivative weight (%/min)

80
-4
70
-6
Weight (%)

60 Polymer = 67.3%
-8
50 Carbonaceous residue = 3.2%
-10
40 CB = 10.1%
-12
30
-14
20
-16
10 Inorganic constituents = 13.7%
5 -17.78
0 10 20 30 40 50 60 70 80 90.25
Time (min)

Figure 3.16 Typical TGA trace for a rubber (NBR) compound. The weight loss
is shown as the full curve and the derivative curve is dashed. Reproduced with
permission from M.J. Forrest in Principles and Applications of Thermal Analysis,
Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.193. ©2008, Blackwell
Publishing [87]

The general application of TGA to determine the bulk composition of rubber samples
is also covered in the section on reverse engineering (Chapter 5) and the data generated
on tyre samples using the AutoStepwise method is included.

An important point to make regarding the carbonaceous residues referred to above is

that, because their oxidation temperature is usually less than that of CB, this helps to
resolve these events within the TGA trace and so aids their respective quantifications.
Also, the ability to detect the presence of a carbonaceous residue in a TGA trace
and its size can, in a limited sense, assist in the diagnostic value of the technique to
identify the generic polymer type of the rubber present in a sample. This information
can then be used to corroborate data obtained by the more standard approaches,
such as the use of FTIR (Section 3.2). In addition to the example of NBR (which
is discussed in more detail in Section 3.2), there are a number of other polymers
which give reproducible amounts of carbonaceous residue that can be used to assist
with generic polymer identifications. The most popular commercial polymers used
in rubbers and rubber blends and which yield carbonaceous residues are shown in
Table 3.4 (Section 3.2).

100
 Principal Components and Bulk Composition

104.4 2.761
100 2
90 0

80 -2

Derivative weight (%/min)

Plasticiser = 5.7%
70 -4
Weight (%)

60 -6
Polymer = 67.3%
50 Carbonaceous residue = 3.2% -8

40 -10
CB = 10.1% -12
30
-14
20
-16
10 Inorganic constituents = 13.7%
5 -17.78
24.03 100 200 300 400 500 600 700 800 882.5
Temperature (ºC)

Figure 3.17 The TGA data presented in Figure 3.16 shown in the form of weight
loss against temperature. Reproduced with permission from M.J. Forrest in
Principles and Applications of Thermal Analysis, Ed., P. Gabbot,
Blackwell Publishing, Oxford, UK, 2008, p.194. ©2008, Blackwell Publishing [88]

Figure 3.16 shows the TGA trace plotted in the form of weight loss against time.
This is the standard format because it makes the information presented easier to see
by eliminating the influence of the two independent heating stages. However, it is
often useful to obtain temperature data on weight loss events, for example, the onset
temperature of the polymer weight loss, to see if a reduction in thermal stability can
be detected (e.g., due to conditions in service). The temperature at the maximum rate
of weight loss (i.e., the peak in the derivative curve) is also useful for using the TGA
trace to provide some qualitative information on a rubber compound. Examples of
this particular application include indications to the polymer type and CB type present
by reference to data obtained on rubber compounds of known composition analysed
under standardised conditions (Sections 3.2 and 3.5). Figure 3.17 shows how the data
in Figure 3.16 would look if displayed as weight loss versus temperature.

Polymers in rubber compounds, due to their high MW, will be lost during a TGA
experiment by a pyrolysis-type reaction and this is true whether a non-oxidising
(e.g., nitrogen) or oxidising (e.g., air) atmosphere is used. By using a standard TGA
programme, for example, the one shown above, sufficient data can been generated on
reference rubber compounds in the nitrogen atmosphere region to enable some limited,

101
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

qualitative information on the types of polymers present in ‘unknown samples’ to be

obtained. This task is performed using the temperature at the peak of the derivative
curve. Indicative examples of the data obtained are provided below:

• NR – peak maximum 370 °C

• Butyl rubber – peak maximum 390 °C

• SBR, BR or EPDM – peak maximum 460 °C

• Fluorocarbon rubber – peak maximum 480 °C

Process oil = 5.6% 0.87

109.6
100 0

90
-2

Derivative weight (%/min)

80
70 -4
SBR = 14.0%
Weight (%)

60
50 -6
NR = 44.6%
40
30 -8
CB = 31.7%
20
-10
10
0 Inorganic constituents = 4.1% -12
-5.579 -12.49
0 10 20 30 40 50 60 70 80 90.25
Time (min)

Figure 3.18 TGA trace for a tyre tread compound containing both NR and SBR.
The weight loss derivative curve has been crucial in this application to determine
the crossover point of the weight losses. Reproduced with permission from
M.J. Forrest in Principles and Applications of Thermal Analysis, Ed., P. Gabbot,
Blackwell Publishing, Oxford, UK, 2008, p.202. ©2008, Blackwell Publishing [89]

102
 Principal Components and Bulk Composition

The list shown above illustrates that the presence of polymers that have relatively low
amounts of thermal stability (e.g., NR) can be detected quite readily in a compound.
However, for a number of polymers, this approach has little or no value and does
not begin to replace the more effective methods of polymer identification reviewed
in Section 3.2. However, in those cases, if a rubber sample contains more than
one polymer (i.e., a blend of polymers) and one of those is NR, these temperature
differences can result in TGA being be a useful tool for, firstly, identifying that a blend
exists and, secondly, in obtaining an approximate measure of the different proportions
of each polymer. This type of work is illustrated in Figure 3.18, where the TGA trace
for a tyre tread compound that contains a blend of NR and SBR is shown. In this
trace, the weight loss events for the two polymers are partially resolved due to the
differences in their respective pyrolysis peak maximum temperatures (i.e., 360 and
460 °C) and so it is possible, by using the inflexion on the derivative weight loss
curve, to obtain an approximate quantification of their amounts in the compound:
44.6% for the NR and 14.0% for the SBR.

It might be thought that the use of a high-resolution TGA programme might assist
with this kind of analysis. However, some experiences have shown that because
these can result in additional weight loss events for the individual polymers, causing
additional overlapping within the TGA trace, their use does not always result in more
accurate quantifications. An example of this effect has been presented by Forrest [90],
who showed that whereas in a conventional TGA experiment only one weight loss
event is obtained for NR, analysing the same sample by using a high-resolution TGA
programme resulted in four weight loss events being observed. Another example in
this published work that indicated that use of a high-resolution TGA programme
can cause problems, was based on its ability to detect a blend of CB in a compound.
Operating the instrument using this mode showed that additional weight loss events
could be created, which gave the impression that more than one type of black was
present in a rubber sample, when in fact only one had been used in the compound.

The examples shown in Figures 3.16 and 3.18 show that if a sample has a composition
that is typical for rubber compounds, the resulting TGA trace is relatively simple
to interpret. In some cases, however, a rubber compound contains a larger number
of components, some with multiple weight loss events, and the situation can be
much more complex. Such a case can be encountered with a flame-retardant rubber
compound and, in these cases, knowledge of the behaviour of the components within
the compound and of the technique itself is crucial to understanding the data. To
illustrate this point, the TGA of a flame-retardant rubber compound based on a blend
of NBR and PVC, and which contains both organic and inorganic flame-retardant
additives in addition to other ingredients, would be expected to exhibit the weight
loss events shown in Table 3.13.

103
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 3.13 Weight loss position in the TGA trace for the components present
in a flame-retardant nitrile/PVC compound
Ingredient/function Position in the TGA trace*
Pentabromodiphenylether 1st weight loss
Phosphate ester 1st weight loss
Phthalate ester 1st weight loss
NBR 2nd and 3rd weight losses
PVC 1st, 2nd and 3rd weight losses
CB 4th weight loss
Hydrated alumina 1st weight loss and inorganic residue
Antimony trioxide Inorganic residue
ZnO Inorganic residue
Cure system, antidegradant system, pre- Obscured by weight losses of principal components
vulcanisation inhibitor and so on due to low level of addition
*TGA experiment run according to the stage 1 and 2 conditions shown above
Reproduced with permission from M.J. Forrest in Principles and Applications of Thermal
Analysis, Ed., P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.206. ©2008,
Blackwell Publishing [91]

TGA is not capable of detecting and quantifying a number of common additives and
ingredients due to their relatively low level in a rubber compound and the relatively
poor detection limit of conventional TGA. Another reason why it cannot detect and
quantify certain ingredients is that they are reactive towards the polymer and each
other (e.g., cure system and antidegradant system). A summary of the most important
additives that cannot usually be determined directly by conventional TGA is:

• Curatives and accelerators

• Antioxidants and antiozonants

• Process aids

• Friction reducing agents

• Pre-vulcanisation inhibitors and cure retarders

The analytical techniques and methods that can be used to detect and quantify these
types of additives are reviewed in Chapter 4. However, the increasing use of TGA
instruments that are interfaced to either an FTIR spectrometer or mass spectrometer
has improved the chances of rubber analysts obtaining at least some of this

104
 Principal Components and Bulk Composition

information. These combinations can also be used to obtain qualitative information

on the plasticiser/oil and the polymer fractions in a rubber sample (Section 2.6.5.3).

It is also possible to use the tube furnace method to obtain bulk compositional
information on a rubber sample, but this has a number of disadvantages over TGA
because it:

• Is often more time-consuming

• Requires a larger sample

• Is not as informative*

*Because there is no record of the weight loss, only values for the total amount of
organic material (i.e., plasticiser and polymer), the total amount of carbonaceous
material (i.e., all CB and any carbonaceous material), and the total amount of
inorganic material are obtained.

The advantage that the tube furnace method has over TGA is that, because a larger
sample size is used, it is usually more accurate, particularly if quantifying small
amounts (e.g., 1–5%) of CB, or other carbonaceous additives, in rubbers that do not
produce carbonaceous residues.

References

1. G.A.L. Verleye, N.P.G. Roeges and M.O. De Moor in Easy Identification of

Plastics and Rubbers, Smithers Rapra Publishing, Shawbury, UK, 2001.

2. M.J. Forrest, Y. Davies and J. Davies in The Rapra Collection of Infrared

Spectra of Rubbers, Plastics and Thermoplastic Elastomers, 3rd Edition,
Rapra Technology Ltd, Shawbury, UK, 2007.

3. M.J. Forrest in Chemical Characterisation of Polyurethanes, Rapra Review

Report No.108, Smithers Rapra, Shawbury, UK, 1999, p.19.

4. M.J. Forrest in Chemical Characterisation of Polyurethanes, Rapra Review

Report No.108, Smithers Rapra, Shawbury, UK, 1999, p.15.

5. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.16.

6. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.17.

105
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

7. M.R.S. Fuh and G-Y. Wang, Analytica Chimica Acta, 1998, 371, 1, 89.

8. G.J. Frisone, D.L. Schwarz and R.A. Ludwigsen in Proceedings of the

International Tire Exhibition & Conference (ITEC) 1996 Select, Akron, OH,
USA, 1996, p.21.

9. D.D. Parker, J.L. Koenig and M. Mori, Rubber Chemistry and Technology,
2003, 76, 1, 212.

10. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.214.

11. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.218.

12. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.222.

13. K.N. Pandey, D.K. Setua and G.N. Mathur, Polymer Engineering Science,
2005, 45, 9, 1265.

14. M. Carlberg, D. Colombini and F.H.J. Maurer, Journal of Applied Polymer

Science, 2004, 94, 5, 2240.

15. B.L. Lopez, L.D. Perez and M. Mesa, E-Polymer, 2005, 18, 1.

16. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.54.

17. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.236

18. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.204.

19. S.R. Holding and E. Meehan in Molecular Weight Characterisation of

Synthetic Polymers, Rapra Review Report No.83, Rapra Technology Ltd,
Shawbury, UK, 1995.

20. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.13.

21. G. Cleaver in Application Compendium: Analysis of elastomers by GPC/SEC,

Agilent Technologies Inc., Santa Clara, CA, USA, 1st August 2015.

106
 Principal Components and Bulk Composition

22. E. Aik-Hwee, S. Ejiri, S. Kawahara and Y. Tanaka in Proceedings of the

International Seminar on Elastomers: Applied Polymer Symposium,
14–16th July 1993, Akron, OH, USA, 1993, p.5.

23. M.R. Ambler, Journal of Applied Polymer Science, 1980, 25, 5, 901.

24. L. Mrkvickova, Macromolecules, 1997, 30, 17, 5175.

25. W.S. Fulton, W.M.H. Thrope and R.J. White in Proceedings of the
International Rubber Conference (IRC) 1996, 17–21st June, Manchester, UK,
1996.

26. W.S. Fulton, W.M.H. Thrope and R.J. White, European Rubber Journal,
1996, 178, 9, 30.

27. P.J. Wyatt, Analytica Chimica Acta, 1992, 272, 1.

28. T. Yamada, T. Okumoto, H. Ohtani and S. Tsuge, Rubber Chemistry and

Technology, 1991, 64, 5, 708.

29. P. Nallasamy, P.M. Anbarasan and S. Mohan, Turkish Journal of Chemistry,

2002, 26, 105.

30. D.D. Parker and J.L. Koenig, Journal of Applied Polymer Science, 1998, 70,
7, 1371.

31. M. Mori and J.L. Koenig, Rubber Chemistry and Technology, 1997, 70, 4,
671.

32. D.J. Patterson, J.L. Koenig and J.R. Shelton in Proceedings of the ACS 123rd
Spring Meeting, 9–13th May, Toronto, Canada, American Chemical Society,
Washington, DC, USA, 1983, Paper No.67, p.30.

33. U. Hoffmann, W. Gronski, G. Simon and A. Wutzler, Angewandte

Makromolekulare Chemie, 1992, 202–203, 283.

34. W. Gronski, U. Hoffmann, G. Simon, A. Wutzler and E. Straube, Rubber

Chemistry and Technology, 1992, 65, 1, 63.

35. W. Gronski, U. Hoffmann, B. Freund and S. Wolff in Proceedings of the

International Rubber Conference (IRC) 1991, 24–27th June 1991, Essen,
Germany, 1991, p.311.

107
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

36. M.R. Krejsa and J.L. Koenig, Rubber Chemistry and Technology, 1991, 64,
1, 40.

37. S-S. Choi, H-M. Kwon, Y. Kim, E. Ko and E. Kim, Polymer Testing, 2017,
59, May, 414.

38. S-S. Choi, Y. Kim and H-M. Kwon, Polymer International, 2107, 66, 6, 803.

39. S.R. Shield and G.N. Ghebremeskel, Rubber World, 2000, 223, 2, 24.

40. E.G. Brame in Applications of Polymer Spectroscopy, Elsevier, Amsterdam,

The Netherlands, 2012.

41. C. Nah, J.H. Park, C.T. Cho, Y-W. Chang and S. Kaang, Journal of Applied
Polymer Science, 1999, 72, 12, 1513.

42. W.J. Sichina, Thermal Analysis, Application Note, Perkin Elmer, PETech-45,
2000.

43. M.J. Forrest in Chemical Characterisation of Polyurethanes, Rapra Review

Report, No.108, Smithers Rapra, Shawbury, UK, 1999, p.23.

44. S.J. Theraftil, A.A. Kuzhuppully and R. Joseph, Iranian Polymer Journal,
2008, 17, 6, 419.

45. D.F. DeCastro, A.F. Martins, J.C.M. Suarez and R.C. Reis Nunes, Kautschuk,
Gummi, Kunststoffe, 2003, 56, 1–2, 49.

46. V. Nigam, D.K. Setua and G.N. Mathur, Journal of Thermal Analysis and
Calorimetry, 2001, 64, 2, 521.

47. R.N. Jana and G.B. Nando, Journal of Applied Polymer Science, 2003, 90, 3,
635.

48. R.N. Jana, P.G. Mukunda and G.B. Nando, Polymer Degradation and
Stabilisation, 2003, 80, 1, 75.

49. A. Amash, R.H. Schuster and T. Frueh, Kautchuk und Gummi Kunststoffe,
2001, 54, 6, 315.

50. H. Varghese, S.S. Bhagawan and S. Thomas, Journal of Thermal Analysis and
Colorimetry, 2001, 63, 3, 749.

108
 Principal Components and Bulk Composition

51. Yuichi Aoyagi and H. Sano in the Proceedings of the ACS Rubber Division
184th Technical Fall Meeting and Education Symposium, 8–10th October,
Cleveland, OH, USA, American Chemical Society: Rubber Division,
Washington, DC, USA, 2013, Paper No.89, p.14.

52. S-S. Choi, J-C. Kim and H-M. Lee, Journal of Applied Polymer Science,
2008, 110, 5, 3068.

53. A. Pappa, K. Mikedi, A. Agapiou, S. Karma and G.C. Pallis, Journal of

Analytical and Applied Pyrolysis, 2011, 92, 1, 106.

54. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.56.

55. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.55.

56. Hummel and F. Scholl in Atlas of Polymer and Plastics Analysis, Volume 3,
2nd Revised Edition, Verlag Chemie GmbH, Weinheim, Germany, 1981.

57. S. Knappe, Rubber World, 2002, January, 33.

58. S. Mohapatra and G.B. Nando, Rubber Chemistry and Technology, 2015,
88, 2, 289.

59. Commission Regulation (EU) 1272/2013: Amending Annex XVII to

Regulation (EC) No 1907/2006 of the European Parliament and of the
Council on the Registration, Evaluation, Authorisation and Restriction of
Chemicals (REACH) as regards polycyclic aromatic hydrocarbons Text with
EEA relevance.

60. EC REACH Regulation (EC) 552/2009: Amending Regulation (EC)

No 1907/2006 of the European Parliament and of the Council on the
Registration, Evaluation, Authorisation and Restriction of Chemicals
(REACH) as regards Annex XVII (Text with EEA relevance).

61. ASTM D7771-11 – Standard test method for determination of benzo-alpha-

pyrene (BaP) content in carbon black.

62. S. Hamm, T. Frey, R. Weinand, G. Moninot and N. Petiniot, Rubber

Chemistry and Technology, 2009, 82, 2, 214.

63. Polycond – An EU Integrated Project funded under FP6-NMP.

http://www.polycond.info

109
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

64. R. Pautrat, B. Metivier and J. Marteau, Rubber Chemistry and Technology,

1976, 49, 4, 1060.

65. M.J.R. Loadman in Analysis of Rubber and Rubber-like Polymers,

4th Edition, Kulwer Academic Publishers, Dordrecht, The Netherlands, 1998,
Chapter 11.

66. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.196.

67. J.J. Maurer, Rubber Age, 1970, 102, 47.

68. A.I. Medalia and D.F. Walker, Technical Report RG-124, Revision 2,
Cabot Corporation: Carbon Black Division, Boston, MA, USA, 1970.

69. C.W. Sweitzer, W.M. Hess and J.E. Callan, Rubber World, 1958, 138, 6, 869.

70. N.A. Stumpe and H.E. Railsback, Rubber World, 1964, 151, 3, 41.

71. J. Kruse, Rubber Chemistry and Technology, 1973, 46, 3, 653.

72. H.H. Le, S. Ilisch, E. Hamann, M. Keller and H-J. Radusch, Rubber
Chemistry and Technology, 2011, 84, 3, 415.

73. W.M. Hess and C.R. Herd in Carbon Black Science and Technology,
2nd Edition, Marcel Dekker, New York, NY, USA, 1993, p.89.

74. H. Raab, J. Froehlich and D. Goeritz in Proceedings of the International

Rubber Conference (IRC) 1997, 6–9th October, Kuala Lumpur, Malaysia,
1997, p.171.

75. W.M. Hess, L.L. Ban and G.C. McDonald, Rubber Chemistry and
Technology, 1969, 42, 4, 1209.

76. M. Takenaka, S. Nishitsuji, S. Fujii, N. Amino and Y. Ishika, International

Polymer Science and Technology, 2011, 38, 5, T/1.

77. M. Takenaka, International Polymer Science and Technology, 2015, 42, 3,

pT/43.

78. T.G. Lamond and C.R. Gillingham, Rubber Journal, 1970, 152, 65.

79. M.J. Wang, S. Wolff and J-B. Donnet, Rubber Chemistry and Technology,
1991, 64, 714.

110
 Principal Components and Bulk Composition

80. M.J. Wang and S. Wolff, Rubber Chemistry and Technology, 1992, 65, 5,
890.

81. R.P. Hjelm, W.A. Wampler, P.A. Seeger and M. Gerspacher, Journal Material
Research, 1994, 9, 3210.

82. T.C. Gruber, T.C. Zerda and M. Gerspacher, Carbon, 1993, 31, 1209.

83. T.C. Gruber, T.C. Zerda and M. Gerspacher, Carbon, 1994, 32, 1377.

84. M. Gerspacher and C.M Lansinger in Proceedings of the 133rd ACS Spring
Meeting, 19–22nd April, Dallas, TX, USA, American Chemical Society,
Washington, DC, USA, 1988, Paper No.7.

85. J.A. Ayala, W.M. Hess, A.O. Dotsan and G.A. Joyce, Rubber Chemistry and
Technology, 1990, 63, 747.

86. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.22.

87. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.193.

88. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.194.

89. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.202.

90. M.J. Forrest in Proceedings of the Polymer Testing ’97 Conference,

7–11th April, Rapra Technology Ltd, Shawbury, UK, 1997, Paper No.5.

91. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.206.

111
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

112
4
Additives

4.1 Introduction

The determination of the principal components of a rubber sample has been covered
in Chapter 3. Two of these components, fillers and plasticisers/oils, can also be
considered, quite rightly, additives. However, due to levels that they are added to a
rubber compound and, as described in Chapter 3 because it is often not necessary
in practice to delve further into the details of a formulation to solve problems (e.g.,
failures in service), it was thought better to cover these particular additives in their own
section, along with the investigations that take place into the nature of the polymer.

This chapter includes two very important groups of additives, antidegradants and
curatives, both of which contain many members. Members of the antidegradant group
can also be referred to using other names (e.g., stabilisers, protection agents) and it
can also include additives that have designations that relate to their functions (e.g.,
antioxidants and antiozonants). Curing agents are also referred to as ‘vulcanisation
agents’ and ‘crosslinking agents’.

With the exception of those that have inherently excellent resistance to degrading
agencies such as high temperatures and ozone (e.g., silicones and fluorocarbons),
rubbers need additives (i.e., antidegradants) that prevent them from suffering
degradation during processing and in service. In addition to heat and ozone,
degradation is also possible from other sources (e.g., radiation and chemicals).

As mentioned above, different rubbers have differing degrees of inherent stability

to these degrading agencies. Tables 4.1–4.3 provide some comparative data on the
resistance of rubbers to degradation due to gamma radiation (Table 4.1), ozone
(Table 4.2) and provide an approximate guide to their maximum service temperature
when they have been compounded with antidegradants (Table 4.3).

In addition to the need for antidegradants to protect them, rubbers also need curatives
to generate the three-dimensional (3D) network which provides them with the inherent
properties (e.g., high elasticity) that are such important and defining characteristics.

113
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 4.1 Gamma radiation resistance of a range of rubbers

Rubber type Insignificant damage [radiation dose (Gy)]
Butyl rubber Up to 10,000
Acrylic rubber Up to 100,000
Silicone rubber Up to 100,000
Chlorosulfonated PE Up to 100,000
NBR Up to 100,000
Fluorocarbon rubber Up to 100,000
CR Up to 100,000
SBR Up to 500,000
EPM and EPDM Up to 500,000
PU rubber Up to 500,000
CR: Polychloroprene
EPDM: Ethylene propylene diene monomer
EPM: Ethylene propylene monomer
NBR: Nitrile rubber
PE: Polyethylene
PU: Polyurethane
SBR: Styrene-butadiene rubber
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.55.
©2001, Smithers Rapra [1]

Table 4.2 Relative resistance of a range of rubbers in the unstabilised state to

ozone
Rubbers with very poor Diene rubbers (e.g., NR, SBR, NBR, butadiene rubber)
resistance
Rubbers having some Butyl rubbers, hydrin rubber, hydrogenated nitrile,
resistance polythioethers, CR
Rubbers having good Acrylics, chlorosulfonated PE, EPR, EPDM, fluorocarbon
resistance rubbers, silicone rubbers
NR: Natural rubber
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.55.
©2001, Smithers Rapra [1]

114
 Additives

Table 4.3 Approximate maximum service temperatures of a range of rubbers

Rubber type Approximate maximum service temperature (°C)
PU rubber 80
NR and polyisoprene rubber 80
Butadiene rubber and SBR 100
NBR and CR rubber 120
Butyl rubber 130
EPR and EPDM rubbers 140
Chlorosulfonated PE 150
Hydrogenated NBR 160
Acrylic rubber 160
Silicone rubber 225
Fluorocarbon rubber 250
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.55.
©2001, Smithers Rapra [1]

There have been many technological developments associated with these two classes
of additive over the last 80 years. These developments, together with the development
of new rubbers, often requiring novel curing and antidegradant systems, means that
having a good technological background in rubber technology is of great assistance to
the rubber analyst. This is a recurring theme throughout this book and its importance
cannot be overestated.

In addition to antidegradants and curatives, there are many classes of additives that
may, or may not, be used in a particular rubber compound depending upon the
properties that it needs to perform satisfactorily during processing and in service. Some
of the more commonly encountered classes of additive that are used in this way are:

• Fire retardants

• Blowing agents

• Process aids

• Resins

The analytical techniques and approaches that can be used to determine antidegradants,
curatives and these other types of additives will be reviewed in this section. For this
type of work, because of the relatively low levels and complex mixtures involved,

115
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

it is combinations involving chromatographic techniques and certain spectroscopic

techniques [e.g., gas chromatography (GC)–mass spectrometry (MS) and liquid
chromatography (LC)–MS] that are usually the most effective due to their sensitivity,
specificity and diagnostic power. However, it is also possible to use other many
approaches and techniques. For example, the relatively volatile nature of a many
rubber additives enabled Lattimer [2] to use a direct probe method MS to analyse
the additives present in a vulcanised rubber sample.

The use of the thermal technique thermogravimetric analysis (TGA) to identify and
quantify principal additives (i.e., plasticisers and fillers) has been addressed in Chapter
3, and it has a part to play in investigative work of this type (e.g., stability studies),
but unless it is coupled with an infrared (IR) spectrometer or MS it does not have
the sensitivity or diagnostic power to be capable of providing specific information
on many of the additives mentioned above. Similar comments apply to other popular
thermal analysis techniques, such as differential scanning calorimetry (DSC) and
dynamic mechanical analysis (DMA).

As with all the sections in this book, this is a large subject in its own right and a
number of specialist texts exists, for example, the book written by Crompton [3].
There are also published articles which provide useful background information and
look to the challenges and needs of the future [4].

In addition to the information in this section, there are also many other examples in
this book in which additives have been determined in rubber products. This work
has been carried out for a wide range of purposes, including reverse engineering
(Chapter 5), curing studies (Chapter 6), rubber fume studies (Chapter 7) food-contact
work (Chapter 8) and extractables and leachables (E&L) work (Chapter 9), but often
the analytical methodologies and approaches that have been used are flexible enough
to be adapted and applied to other applications and objectives.

4.2 Determination of Antidegradants in a Rubber Compound

In common with other polymer systems, to maintain the optimal physical properties
imparted to a rubber compound by a judicious choice of polymer type, filler, plasticiser
and cure system, stabilisers need to be added to stop, or at least retard, degradation
by a number of agencies [e.g., heat and ultraviolet (UV) light].

The use of an initial solvent extraction step is common in the analysis of a rubber
matrix for antidegradants, and other types of additives because these species may not
be volatile enough to be removed by heating the sample. The use of organic solvents to
remove additives (e.g., plasticisers) from rubber samples was introduced in Section 3.2

116
 Additives

and, particularly when quantification as well as identification is required, similar

considerations apply in these studies in that the solvent should be compatible with the
additive of interest. Table 3.6 provides, as a guide, a list of the solubility parameters
of a range of solvents, a number of which can be used for this type of work.

In the case of some relatively volatile antidegradants [e.g., butylated hydroxytoluene

(BHT) and isopropyl p-phenylenediamine (IPPD)], heating a sample liberates enough
of the species to enable an identification to be made using GC–MS. However,
quantification of an antidegradant, even volatile examples, will usually require
extraction by a solvent to as near to 100% efficiency as possible. Analysis of the
extract can then be carried out by high-performance liquid chromatography (HPLC)
or spectroscopy (UV or IR) or, providing the species has sufficient thermal stability
and volatility, GC. A chromatographic approach to prevent ‘masking’ by other
co-extractants will usually be required if the extract is relatively complex. For this
application, relatively non-specific solvents (i.e., those having mid-range solubility
parameters), such as chloroform and acetone, are good extracting solvents. If a high
proportion of oligomeric material is present in the extract (sometimes a problem with
unvulcanised rubber samples), it can be necessary to add a precipitating solvent (e.g.,
methanol or heptane) to prevent this from interfering with the chromatographic or
spectroscopic step.

If the type of organic antidegradant in a compound is completely unknown, analysis

of the solvent extract by GC–MS is an excellent method for identification. Once the
identification has been achieved, quantification can be carried out using a preferred
method and a validated procedure (e.g., a published standard or in-house method)
if one is available. To ensure confidence in the data, a validated method will usually
have to be developed if one is not already available.

If the antidegradant is not volatile enough to be identified using a GC-based technique,

direct probe MS can be used and the antidegradant identified by reference to its
molecular ion and a database of molecular weights (MW). This technique can be
carried out on samples in the as-received state as well as dried sample extracts.

4.2.1 Ultraviolet Stabilisers

The widespread use of carbon black (CB) in rubber compounds gives the additional
advantage of stabilisation against UV light, with the filler preferentially absorbing
the energy and so protecting the polymer. The determination and quantification of
CB (usually regarded correctly as a filler) in a rubber is described in Section 3.5.

The white pigment titanium dioxide (TiO2) also provides protection against UV light

117
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

and can be incorporated into non-black articles. The amount of TiO2 in a rubber can
be determined after acid digestion of the matrix by the use of an elemental technique
such as atomic absorption spectroscopy or inductively coupled plasma.

Benzotriazoles and hindered amines can be used as UV stabilisers. The UV-absorbing

qualities of these species can be employed for their detection and quantification by
UV spectroscopic analysis of solvent extracts of rubber samples.

4.2.2 Antioxidants

The International Organization for Standardization (ISO) standard, ISO 11089, deals
with the determination of N-phenyl-β-naphthylamine and poly-2,2,4-trimethyl-1,2-
dihydroquinoline (TMQ) as well as two generic types of antiozonant (Section 4.2.3).
This method has been tested on CR, NBR and SBR and involves an initial extraction
step using a 2:1 mixture of isopropanol and dichloromethane (DCM). The extract
obtained is then analysed by reverse-phase HPLC using a mixed solvent programme
with UV detection.

In addition to the above, HPLC has been widely used for many years by analysts for
the determination of antioxidants in rubbers, and reviews have been provided by
Sidwell [5] and Sullivan and co-workers [6].

A useful HPLC method, which is a modification of a method developed by

an antidegradant manufacturer, can be used on rubbers for the detection and
quantification of a wide range (i.e., 20 species) of antidegradants. The sample is
cryogenically ground up into a fine powder and then extracted with diethyl ether
for 30 min with ultrasonic agitation. The extract is then analysed by HPLC using
the following conditions:

• Injection volume: 25 µl

• HPLC column: aqua 5 C18 ODS reverse-phase column 150 × 3.0 mm (or similar)

• Mobile phase: A = 75/25 methanol/water and B = 50/50 ethyl acetate/acetonitrile:

Gradient %A %B
Initial 90 10
After 25 min 0 100

• Flow rate: 2.0 ml/min

118
 Additives

• Detection: UV detector set at 270 nm

• Temperature: 40 °C

The following antioxidants can be determined using this technique:

• Irganox® 1010

• Irganox® 1098

• Irganox® 1076

• Irganox® 1330

• BHT

The wide applicability of this HPLC method is due to the non-specific nature of the
mobile-phase mixture.

The capability of HPLC has been greatly extended by the commercialisation of

affordable LC–MS instruments. The mass selective detector enables unknown samples
to be analysed and the antioxidants present to be identified. The LC–MS technique,
therefore, compliments GC–MS and, with it, enables an extensive MW/volatility
range to be covered.

Thin-layer chromatography (TLC) is comparatively easy and cheap to use and it is

both quick and accurate for the identification of antioxidants in rubber extracts.
Irrespective of the complexity of the extract composition and the amount used, this
method can be used to give a sharp separation for identification purposes. Distinctive
retardation factor values and colours can be seen for each antioxidant depending
on the stationary phase, the developing solvent used and the detection agent [7, 8].
Normal silica gel has proved effective as a stationary phase, as have others such as
alumina starch, gypsum–silica gel and starch–silica gel, and where identification
is difficult because of the interferences of co-extracted additives, two-dimensional
development methods using two types of solvent system have been employed. [9].

The TLC technique for antidegradants is also described in the American Society for
Testing of Materials (ASTM) standard, ASTM D3156 and ISO 4645. The rubber
extract is taken up in DCM and then a spot of this solution developed in the developing
tank using the nominated solvent. In addition to being used for identification purposes,
TLC can be used in a semi-quantitative way; the logarithm of the spot area being
proportional to the quantity of the analyte in the spot, and this value being compared
with spots containing standard amounts of the analyte. The complexity of rubber
extracts usually limits the accuracy of the quantifications obtained.

119
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

GC has been used for the analysis of phenolic and amine-type antioxidants.
Antioxidants which have high boiling points cannot be directly analysed by GC but
they can be analysed in the form of derivatives, such as acetates, trifluoroacetates,
trimethyl silylethers, and methyl ethers. Trimethyl silane-based antioxidants, for
example, give good separations on standard siloxane-based stationary-phase columns
[10].

The newer, more sophisticated variants of GC–MS-type instruments also offer great
potential in this area. An example is GC×GC–time-of-flight (ToF)–MS and this type
of instrument has been used to analyse rubber samples by Forrest and co-workers
[11]. Data were obtained by this research team from an acetone extract of a cured
butyl rubber sample using the analytical conditions shown below:

• Sample for analysis:

• Extract obtained from 0.3 g of finely divided sample which was ultrasonicated
with acetone for 30 min.

• Instrument details and analysis conditions:

• I nstrument: Agilent 6890 gas chromatograph with LECO Pegasus III GC×GC–
ToF–MS.
• Injection: programmed temperature vaporiser injection, 10 °C above primary
oven temperature; 1 µl of the acetone extract.
•  rimary column: J&W scientific HP-5MS (30 m × 0.250 mm with film
P
thickness of 0.25 µm).
• Secondary column: SGE BPX-50 (1.8 m × 0.100 mm with film thickness of
0.10 µm).
• Carrier gas: helium at 1.0 ml/min with constant flow.
• Primary oven program: 40 °C for 10 min then 10 °C/min to 320 °C and hold
for 15 min.
• S econdary oven program: 50 °C for 10 min then 10 °C/min to 330 °C and
hold for 15 min.
• Modulator offset: 30 °C.
• Modulator frequency: 4 s.
• Hot time of the modulator: 0.30 s.
• Mass spectrometer: scanning 30 to 650 Da at 76 spectra/s.

120
 Additives

The data generated by a GC×GC–ToF–MS instrument can be presented in a number

of ways (Section 5.2.1). For example, Figure 4.1 shows the 3D chromatogram total
ion current (TIC) that resulted from the analysis of the butyl rubber acetone extract.

3
2n 2365
d 2
Co T
lu nR
m 1365 lum
n 1 Co
RT 1st

0 365

Figure 4.1 3D GC×GC–ToF–MS TIC chromatogram for the butyl rubber acetone
extract (RT: retention time). Reproduced with permission from M.J. Forrest,
S. Holding and D. Howells, Polymer Testing, 2006, 25, 63. ©2006, Elsevier
[11] and M.J. Forrest, S. Holding and D. Howells in Proceedings of the High
Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland, Rapra
Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd
[12]

The information shown in Figure 4.1 can be contrasted with the chromatogram
obtained on the same butyl rubber extract by conventional GC–MS (Figure 4.2) using
the analytical conditions shown below:

• Sample for analysis:

• E
 xtract obtained from 0.3 g of finely divided and ultrasonicated with acetone
for 30 min.

• Instrument details and analysis conditions:

• Instrument: Agilent 6890/5973 GC–MS.

121
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Injection: 1 µl splitless at 310 °C.

• Carrier: helium at 1.0 ml/min.

• Column: Restek RTX-5 amine 30 m × 0.32 mm and film of 1.0 µm.

• Column oven: 40 °C for 5 min; 20 °C/min increase up to 300 °C held for

15 min (30-min run time).

• MS: scanning 35 up to 650 Da every 0.25 s.

Abundance
TIC: 0804031.D
3000000
2800000
2600000
2400000
2200000
2000000
1800000
1600000
1400000
1200000
1000000
800000
600000
400000
200000
0
Time--> 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00

Figure 4.2 Conventional GC–MS TIC chromatogram for the butyl rubber acetone
extract. Reproduced with permission from M.J. Forrest, S. Holding and
D. Howells, Polymer Testing, 2006, 25, 63. ©2006, Elsevier [11] and M.J. Forrest,
S. Holding and D. Howells in Proceedings of the High Performance and Specialty
Elastomers, 20–21st April, Geneva, Switzerland, Rapra Technology Ltd, Shawbury,
UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [12]

122
 Additives

In the case of this sample, both the GC×GC–ToF–MS and conventional GC–MS
analyses were successful and by analysing and interpreting the data it was possible to
identify the antidegradant that was present in the butyl rubber: 6-ethoxy-1,2-dihydro-
2,2,4-trimethylquinoline. If the rubber sample, and hence its solvent extract, had been
more complex, then the GC×GC–ToF–MS system would have provided the analyst
with a better chance of success due to its superior resolving power.

4.2.3 Antiozonants

To date, the only organic compounds shown to be very effective in the protection of
diene rubbers against chain scission by ozone (via the cleavage of the carbon–carbon
double bond) are p-phenylenediamine compounds.

In addition to the analysis of antioxidants (Section 4.2.2), the standard ISO 11089
can be used to determine N-alkyl-N'-phenyl-p-phenylenediamines [e.g., IPPD and
N-(1,3-dimethyl butyl)-N'-phenyl-p-phenylenediamine (6PPD)] and N-aryl-N'-aryl-
p-phenylenediamines (e.g., diphenyl-p-phenylenediamine) by HPLC. The HPLC
chromatogram of a methanol extract from a CR compound, obtained using a UV
detector, is shown in Figure 4.3. From the data in Figure 4.3 it is possible to identify
and quantify a number of peaks due to diaryl-p-phenylenediamine antiozonants
(highlighted in the chromatogram) in the sample.

A microcrystalline PE wax is often added to a rubber formulation along with a

p-phenylene-type antiozonant. These types of wax have a limited solubility in the
rubber matrix and so bloom to the surface. In doing so they help the antiozonant
migrate to the surface where it is needed to function but, more importantly, they
provide a physical barrier to ozone. This approach is most successful in static
applications, but does not work for dynamic applications because as the rubber
product is stressed the protective film will be ruptured. The fact that these waxes
bloom to the surface can be used for their detection by surface analysis techniques
such as IR microspectrometry. Unfortunately, these waxes are added only at relatively
low doses to the compounds (e.g., 2–5 phr) and so their detection by bulk analytical
techniques [e.g., DSC, pyrolysis (Py) IR] is unlikely to succeed, and their relatively
high MW and relatively poor solubility makes isolation by solvent extraction difficult.

TLC can be used for the analysis of p-phenylenediamines. The best results for the
isolation and identification from extracts are obtained using isopropanol/chlorobenzene/
water/25% aqueous ammonia (52:33:10:5) as the developing solvent, and diazotised
4-amino-benzene sulfuric acid or 3,5-dibromo-p-benzoquinonechlorimine as the
colour-forming reagent.

123
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Methanol extract from sample

Diaryl-p-phenylenediamines
Absorbance

present

0 4 8
Time (min)
Figure 4.3 HPLC UV chromatogram of the methanol extract from a CR rubber
compound. Reproduced with permission from M.J. Forrest in Rubber Analysis
– Polymers, Compounds and Products, Rapra Review Report No.139, Smithers
Rapra, Shawbury, UK, 2001, p.26. ©2001, Smithers Rapra [13]

4.2.4 Degradation Studies

Sections 4.2.1–4.2.3 have covered the analytical techniques and methods used to
identify and quantify the different types of antidegradant used in rubber products.
A significant amount of investigative work has been carried out on the function of
these additives in rubbers as well as the breakdown products and reaction products

124
 Additives

that can result. To illustrate the findings of this work, a selection of it will be included
here. However, because the products of stabilisation and antidegradant activity are
often substances with relatively low MW, they are of interest in other important areas
of research and characterisation work. Hence, related information will also be found
in the sections of this book that deal with those, for example:

• Composition of rubber fume (Chapter 7)

• Migration from rubber products into food (Chapter 8)

• E&L studies on rubber products (Chapter 9)

The analytical techniques that can be employed to study the degradation of rubber
and some examples of the literature that has published recently in this area are
introduced below.

Although they cannot be used to identity the type of antioxidant in a rubber sample,
it is possible to use the thermal analysis techniques TGA and DSC to obtain a
measure of the thermal oxidative stability of the rubber compound that results
from their use and, in the case of DSC, the energetics of their protective action.
In the latter case, the studies that have been published include work on the use
of p-phenylenediamines to protect polyisoprene rubber [14, 15] and that of Parra
and Matos [16], who investigated the synergistic effects of a range of different
antioxidants in NR.

In addition to being able to contribute towards research studies, this capability enables
these techniques to be used in industry as quality-control tools and to assist in failure
diagnosis where it is suspected that a sample does not possess sufficient protection
because of the absence or an insufficient quantity of the antioxidant. One of the
ways to use DSC for this type of work is to employ a relatively high experimental
temperature (e.g., 100 °C), along with an oxidising atmosphere (e.g., air), and monitor
the change in heat capacity as a means of indicating the onset of oxidation. When
DSC is used for this purpose for plastics, the result obtained is referred to as the
‘oxidation induction time’, and there are standards that can be used to carry out the
work (e.g., ISO 11357-6). With rubber compounds, because they tend to have more
complex compositions and contain a higher proportion of volatile species, this type
of investigative work can be more difficult because a change in heat capacity can also
take place because of the volatilisation of substances or, if the sample is uncured,
due to the volatiles released during curing. The use of sealed sample pans can help
overcome these problems although, in the case of curing, sometimes the pressure
that builds up within the pan is too great and it splits open. Another way in which
DSC can be used to the study oxidation of rubbers is to use the fact that this type
of event results in a change in heat capacity, with the glass transition temperature

125
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

(Tg) occurring over a broader temperature range and at a higher temperature. This
phenomenon is usually more obvious with rubbers that undergo some crosslinking
during oxidation (e.g., diene-type rubbers).

The observation made above with regard to the disruptive effects of compounding
ingredients volatilising during DSC is also applicable when TGA is used to obtain
information on the stability of a rubber. In the case of the TGA experiment, it is a
reduction in the temperature at which the rubber compound starts to pyrolyse that
can be used. Because this event takes place at a relatively high temperature (e.g.,
>250 °C), a high experimental temperature in either an isothermal or temperature
ramp mode has to be used, and so the influence of volatile ingredients often cannot
be avoided. If all other factors are the same, the earlier and faster that the rubber
starts to pyrolyse, the less thermally stable it is. It is important to point out, though,
that this is a measure of non-oxidative thermal stability because an non-oxidising
atmosphere (e.g., nitrogen) must be used to prevent the situation becoming even more
complicated due to the oxidation of additives, particularly CB.

It is complications, such as those described above, with the use of TGA that means that
DSC tends to be the more accepted technique for thermal stability work. However,
researchers have shown how TGA can be used effectively in thermal stability studies,
and this type of investigation can obviously have important contributions in the
designing of a rubber compound to ensure that it has the appropriate stability for a
particular end-use application. An example of this type of study has been performed
by Denardin and co-workers [17], who used TGA with a ramped heating rate to
show how the dynamic, non-isothermal Kissinger and Osawa methods can be used
to determine the activation energies of the degradation processes that occur with CR
rubber. Also, an extensive study of EPDM rubbers by Gamli and co-workers [18]
involved using high-temperature (410 to 440 °C) isothermal TGA experiments and
mathematical models based on proposed degradation mechanisms to establish the
kinetic parameters of degradation.

Another thermal analysis technique that has a part to play in degradation and
oxidation investigations and, hence provide information on the effectiveness of
antioxidants, is DMA. In this work, use can be made of the ability of DMA to record
comparative data on samples that have been through an ageing process, either an
accelerated one in an oven or during use in service. When a rubber or plastic ages,
free radicals are usually generated and, if not stopped by an antioxidant, these can
lead to one of two effects:

• Scission of polymer chains – resulting in a decrease in the modulus

• Creation of crosslinks – resulting in an increase in the modulus

126
 Additives

Often, these effects can both occur but usually one of them (the choice of which
being dependant upon the type of rubber) will predominate. This means that the
progress of ageing can be followed by DMA by monitoring the modulus of the rubber
samples [19].

A number of researchers have used a combination of analytical techniques to study

thermal stability. For example, Skachkova and co-workers [20] used isothermal
TGA in conjunction with atomic force microscopy to study the thermo-oxidation of
oil-extended EPDM/polypropylene thermoplastic rubbers. They found that the oil
helped to stabilise the rubber by degrading and combining with the oxidised surface
of the sample to form a partial barrier to oxygen. Pruneda and co-workers [21] used
a number of techniques, including TGA, DSC and scanning electron microscopy
(SEM), to obtain the activation energies associated with the degradation processes
of NBR/polyvinyl chloride blends that contain the antidegradant TMQ.

Ogawa and co-workers [22] used PyGC–MS to identify the volatile substances
produced by NBR sheets that had been aged in an ozone-rich environment. The
data they obtained revealed three substances related to the antioxidant in the NBR
rubber. By focusing on the quinoline-type antioxidant itself, they also showed that
when the concentration of this additive was reduced to ≤50% of its starting value
by a ozone-ageing regimen, the mechanical strength of the 0.1-mm surface layer of
the rubber became critically low.

Hakkarainen and co-workers [23] evaluated the behaviour of NBR during long-term
thermal ageing by following the migration of low-MW additives and the thermal
degradation products of additives. Volatile substances were detected and identified
using headspace GC–MS analysis and the oxidation of the butoxyethyl phosphate
plasticiser was followed by Fourier–Transform infrared (FTIR) spectroscopy.

PyGC–MS was mentioned in Section 3.2 as a means of identifying the polymers

present in rubber samples. It can also be used to identify antioxidants and other
additives and to obtain information on the changes that can occur to these ingredients
when a rubber ages in service or under accelerated conditions. An example of this
application is provided by Yang and co-workers [24]. They used a flash evaporation
PyGC–MS technique to analyse the volatile additives and macromolecular structure
of a number of polymeric materials, including a NBR in aged and unaged forms.
The research team found that during the thermal oxidation of NBR seals, additional
crosslinks were formed, but there were no additional volatile degradation products.
The principal change that they detected was a decrease in the level of volatile additives,
particularly paraffins, the antioxidant, and hindered phenol compounds, which was
accompanied by the hardening of the rubber and a reduced level of protection from
further oxidation. Depth profiling showed that different additives had different

127
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

migration speeds within the rubber and, from the levels of additives that remained,
it was possible to predict the degradation status of the rubber. Overall, the research
team considered that PyGC–MS could provide extensive information on how
polymers degrade and so could make a valuable contribution towards research into
degradation mechanisms.

Tawfic and Hussein [25] prepared an adduct amine-epoxy (Am-Ep) by reacting

together N-methyl-p-phenylenediamine and epoxidised oil. They then added this
adduct in differing amounts to EPDM and NBR rubbers and assessed its influence
on thermal stability, blend compatibility and mechanical properties. The analytical
techniques used in the work included SEM and DSC and the results obtained revealed
that the optimum addition level of the Am-Ep adduct was 7.5 phr.

4.3 Determination of Cure System Additives and their Breakdown

Products

This section looks at the analysis of rubbers to obtain information on their cure
systems. There are two principal types of cure system used in rubber products: sulfur or
peroxide. Where either of the systems could have been used (e.g., EPDM rubbers) it is
usually possible to differentiate between them by carrying out a sulfur determination.
Although additives such as CB will contribute to the overall sulfur level in a rubber
compound, a low result (e.g., <0.3% w/w) is usually indicative of a peroxide-cure
system. Techniques to quantify the amount of sulfur in a rubber are discussed in
Sections 2.2 and 2.3. In common with most other aspects of rubber analysis, there
can be complications, and it is possible for technologists to add elemental sulfur
into a rubber as a peroxide cure co-agent. Therefore, a relatively high sulfur level
cannot always be taken to indicate a sulfur-type cure system and further work (e.g.,
GC–MS) needs to be carried out. The identification of peroxide co-agents such as
triallyl cyanurate (TAC) are covered in Section 4.4.4.

The fact that some sulfur-cure system accelerators [e.g., tetramethylthiuram disulfide
(TMTD)] produce nitrosamines during the curing process has given cause for concern
for a number of years. Because of the potential health and safety concerns, a number
of studies have been carried out on these species. Analysis of rubber compounds for
nitrosamines usually takes the form of extraction followed by GC analysis using a
specific detector (e.g., electrolytic conductivity or thermal electron analyser). A review
of the subject of nitrosamines in rubber is provided by Willoughby and Scott [26].

In addition to the information presented in this chapter, there is also information

that is applicable and relevant to the determination of curing system additives in
Section 4.5 and, as mentioned in Section 4.1, in many other chapters of this book

128
 Additives

that deal with particular themes such as reverse engineering (Chapter 5) and curing
studies (Chapter 6).

4.3.1 Analysis of Uncured Samples

In the case of failure and deformulation work it is usually only cured rubber which
is available for analysis because the sample will usually have originated from the a
final product. However, there are examples, and quality-control work is one, where
the uncured sample will be available. This will provide the opportunity to look for
un-reacted accelerators and curing agents such as peroxides. Particular care has to
be taken because these chemicals are, by their nature, thermally labile and reactive.
It is not possible, therefore, to use hot solvent extraction (e.g., reflux or Soxhlet
extraction) or high-temperature analysis techniques, such as GC–MS, because these
will bring about the same types of reactions that occur during curing. Rather, cold
extraction is used along with an ambient temperature separation and identification
technique such as TLC or HPLC. It should also be borne in mind that, even at
ambient temperature, the choice of extraction solvent can be important. The use
of acetone can cause thiuram disulfide accelerators (e.g., TMTD) to decompose to
dithiocarbamates. For this reason, the use of solvents such as methanol, 2-propanol
or DCM is preferred.

Gross and Strauss [27] described the use of HPLC for the analysis of accelerators.
There are many reviews concerning the use of TLC, including those by McSweeny
[28] and Hummel and Scholl [29].

In recent years, advances in technology have reduced the price and, hence, increased
the accessibility of LC–MS instruments and these are often used along with GC–MS
instruments to identify cure systems ingredients and their breakdown products and
reaction products. Unlike GC–MS, which can use the published 70-eV electron-impact
mass spectra libraries, there are no commercial libraries available for LC–MS and so
in-house libraries need to be generated under specific conditions of use. Once these are
available, however, these instruments offer the analyst an increased range of options
over traditional HPLC systems (Section 2.5.5).

4.3.2 Analysis of Cured Samples

The analysis of cured samples is the more usual state of affairs. It is necessary to
have a certain amount of knowledge on the types of species that are formed during
vulcanisation reactions if success in identifying a cure system is to be forthcoming.

129
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The most useful generally available technique for this type of work is GC–MS. There
are two possible modes of sample introduction: solution injection and headspace
(dynamic or static mode). In the former, a useful method of sample preparation
involves cryogenic grinding of 0.3 g of the cured sample, followed by extraction
using 2 ml of a non-selective solvent (e.g., diethyl ether or acetone) in an ultrasonic
bath for 30 min.

The resulting solvent extract is then analysed, for example, under the following
conditions:

• Instrument: Hewlett Packard 6890/ 5973 GC–MSD

• Carrier: helium at 2.0 ml/min constant flow

• Injection mode: 1 µl splitless (0.75 min; 50 ml/min)

• Injection temperature: 320 °C

• Column type: RTX5-MS 30 m × 0.25 mm, 0.25 µm film thickness

• Column temperature: 40 °C for 5 min, 20 °C/min up to 300 °C held for 12 min

• MS settings: 20 to 620 Da scanned every 0.33 s

The solvent extraction method enables the large majority of the diagnostic species
of interest to be detected. Other non-cure system species such as antidegradants and
process aids will also be identified. The main drawback with this approach is that
volatile diagnostic compounds such as low-MW amines (e.g., dimethylamine) can be
masked by the large solvent peak at the start of the chromatogram.

The headspace techniques provide a means of identifying the largest range of cure
system breakdown products, particularly in conjunction with cryogenic cooling of
the GC oven to improve resolution of the early eluters. In this approach, a relatively
small amount of sample (e.g., 3 mg) is heated in the oven of a dynamic desorption
unit and the volatiles produced are collected in a cryogenically cooled Tenax™-filled
trap. The experimental conditions are, typically, as follows:

• Dynamic headspace:

• Instrument: Perkin Elmer ATD 400

• Desorption temperature: 150 °C for 10 min

• Desorption flow: helium at 20 ml/min

• Trap collection temperature: −30 °C

130
 Additives

• Trap injection temperature: 250 °C

• Trap outlet split flow: 9 ml/min

• Trap inlet split flow: off

• Split: 10:1

• GC–MS – The conditions used are the same as for the solvent extraction method
with the exception of the following:

•  C column temperature: −30 to 50 °C at 5 °C/min, following by 20 °C/min

G
to 300 °C and then held at 300 °C for 12 min

• Carrier: helium at 10.5 psi

3
2n 2365
d 2
Co
lu RT
m 1365 lumn
n 1 Co
RT 1st

0 365

Figure 4.4 3D GC×GC–ToF–MS TIC chromatogram for the acrylic rubber acetone
extract. Reproduced with permission from M.J. Forrest, S. Holding and
D. Howells, Polymer Testing, 2006, 25, 63. ©2006, Elsevier [11] and M.J. Forrest,
S. Holding and D. Howells in Proceedings of the High Performance and Specialty
Elastomers, 20–21st April, Geneva, Switzerland, Rapra Technology Ltd, Shawbury,
UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [12]

131
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

As discussed in the case of antioxidants (Section 4.2.2), the newer GC–MS-type

instruments offer great potential in the area of additive analysis. Forrest and co-
workers [11] used GC×GC–ToF–MS to determine the additives, including the
curatives, in an acrylic rubber sample. Data were obtained from an acetone extract
of the cured acrylic rubber sample using the same sample preparation conditions and
instrument conditions as those shown in Section 4.2.2 for the analysis of the butyl
rubber extract.

Figure 4.4 shows the 3D GC×GC–ToF–MS TIC chromatogram for the cured acrylic
rubber extract. Figure 4.5 shows the TIC chromatogram recorded on the same extract
with a conventional GC–MS instrument operated under the same conditions shown
in Section 4.2.2.

Abundance
TIC: 0804011.D
1e+07
9500000
9000000
8500000
8000000
7500000
7000000
6500000
6000000
5500000
5000000
4500000
4000000
3500000
3000000
2500000
2000000
1500000
1000000
500000
0
Time--> 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00

Figure 4.5 Conventional GC–MS TIC chromatogram for the acrylic rubber
acetone extract. Reproduced with permission from M.J. Forrest, S. Holding and
D. Howells, Polymer Testing, 2006, 25, 63. ©2006, Elsevier [11] and M.J. Forrest,
S. Holding and D. Howells in Proceedings of the High Performance and Specialty
Elastomers, 20–21st April, Geneva, Switzerland, Rapra Technology Ltd, Shawbury,
UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [12]

132
 Additives

By analysing and interpreting the data produced by these two techniques it is possible
to obtain information on the cure system that was used to cure the acrylic rubber.
This was found to include the amine N,N'-di-o-tolyl guanidine.

For either analytical approach, identification of the diagnostic cure system breakdown
species is by reference to mass spectral libraries, such as those produced by the
National Institute of Standards and Technology, the Royal Society of Chemistry,
Wiley/NBS [30] and the USA Environmental Protection Agency/National Institutes
of Health. The mass spectra of the majority of species of interest are in these libraries
and so, providing that in-house knowledge to reconstruct cure systems is available,
successful attempts can be made to interpret the results.

Table 4.4 Diagnostic species associated with different cure systems

Sulfur cure system Diagnostic species
2-(4-Morpholinothio)benzothiazole BT
Morpholine
Sulfrasan R Morpholine
MBT BT
2,2'-Dithiobis(benzothiazole) BT
TMTD Dimethylamine
Tetramethylthiourea
Dimethylformamide
Zinc dimethyldithiocarbamate Dimethylformamide
Dimethylamine
Peroxide curative Diagnostic species
Perkadox 14/40 Tertiary butanol
1,3-Di-(1,1-dimethylmethanol) benzene
methanol
Tert-butyl cumyl peroxide 1,1-Dimethyl benzene methanol
Acetophenone
Tertiary butanol
Dicumyl peroxide 1,1-Dimethyl benzene methanol
Acetophenone
BT: Benzothiazole
MBT: 2-Mercaptobenzothiazole
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.28.
©2001, Smithers Rapra [32]

133
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

One problem associated with this task is that most accelerators and curing agents,
such as peroxides, break down to give more than one product and a number of these
are common to more than one cure system species. This is demonstrated in Table 4.4
and these types of diagnostic data are included in a book by Willoughby [31].

Another example of GC–MS being used to identify cure system breakdown products
and, hence, provide possible identities to the compounds that have been used in the
original compound is provided by Sidwell [33] in the results obtained from the analysis
of water that had contacted a NR compound (designated NR2). The formulation of
this NR2 compound is shown in Table 4.5 and samples for analysis were obtained
by curing for 6.25 min at 155 °C.

Table 4.5 Formulation of NR compound – NR2

Ingredient phr
NR (SMR-CV60) 100
ZnO (cure activator) 5
Stearic acid (cure activator) 1
4,4'-Thio-bis-(3-methyl-6-tert-butyl phenol) (antioxidant) 1
6PPD (antiozonant) 1
Hydrocarbon wax (Okerin 1944) 2
High abrasion furnace CB N330 (reinforcing filler) 35
Naphthenic oil (petroleum-based oil lubricant) 5
Sulfur (curative) 1
CBS 2.5
CBS: N-cyclohexyl-2-benzothiazole sulfenamide (cure accelerator)
phr: parts per hundred of rubber
Reproduced with permission from J. Sidwell in Proceedings of the Food Contact Polymers
Conference, 21–22nd April, Brussels, Belgium, Smithers Rapra, Shawbury, UK, 2009,
Paper No.7. ©2009, Smithers Rapra [33]

The GC–MS TIC chromatogram that was obtained on a DCM-partitioned, distilled

water sample that had contacted a cured sample of NR2 for 24 h at 40 °C is shown
in Figure 4.6. The peaks of the three substances highlighted in this chromatogram
(i.e., cyclohexanamine, BT and mercaptobenzothiazole) are the breakdown/reaction
products of the CBS accelerator in the compound and can be regarded as diagnostic
species, useful for the identification of this compound during reverse engineering

134
 Additives

work. However, care has to be taken during such evaluations because two of these
species (BT and mercaptobenzothiazole) can also be formed by other accelerators
(Sections 5.3.3 and 9.2.3.3). It is, therefore, important to build up an in-house database
of such breakdown products and to consult other sources (e.g., the database generated
during the FSA research project A03038) (Section 8.1).

TIC: RUN007.D
3e+07 Cyclohexanamine

NR2 SAMPLE
2.5e+07
DCM extraction
BT
Abundance

2e+07

Mercaptobenzothiazole
1.5e+07

1e+07

5000000

0
4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00
Time

Figure 4.6 Water extractables obtained at 24 h at 40 °C from compound NR2

(partitioned into DCM and examined by GC–MS) [33]. Reproduced with
permission from M.J. Forrest in Food Contact Rubbers 2 – Products, Migration
and Regulation, Rapra Review Report No.182, Smithers Rapra, Shawbury, UK,
2006, p.28. ©2006, Smithers Rapra [34]

In addition to the three major breakdown substances shown in Figure 4.6, a number
of other products that were also related to the CBS accelerator were detected in the
distilled water by GC–MS and these are all shown in Figure 4.7.

135
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

O
H2N O C
N
Cyclohexanamine Cyclohexanone Cyclohexane, isocyanato

H
N N N
S O
S S S
Benzothiazole Benzothiazole, 2-methylthio)- 2(3H)-Benzothiazolone

N N
HS H2N
S S
Mercaptobenzothiazole 2-Benzothiazolamine

Figure 4.7 Breakdown products of CBS detected by GC–MS in distilled water that
had contacted NR2 for 24 h at 40 °C. Reproduced with permission from M.J.
Forrest in Food Contact Rubbers 2 – Products, Migration and Regulation, Rapra
Review Report No.182, Smithers Rapra, Shawbury, UK, 2006, p.29. ©2006,
Smithers Rapra [35]

In the same study, Sidwell also reported on the use of headspace GC–MS for the
detection of low-MW monomers, breakdown products of accelerators and reaction
byproducts of the cure system in rubber samples. A sulfur-cured food-grade NBR
sample was prepared and the volatiles that were released at 150 °C were examined
by headspace GC–MS. Among the substances detected were:

• Acrylonitrile (ACN) (residual monomer)

• Dimethylamine [breakdown product of the accelerator tetramethylthiuram

monosulfide (TMTM)]

• Carbon disulfide (reaction product of the sulfur cure system)

The presence of these substances eluting early on in headspace GC–MS chromatogram

is shown in Figure 4.8.

136
 Additives

Abundance
TIC: RUN064.D
9000000 NBR3 TCT @ 150 °C
8000000

7000000
6000000
Dimethylamine, ACN
5000000 monomer, carbon disulfide,
4000000
3000000

2000000

1000000

0
2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00
Time-->

Figure 4.8 Headspace GC–MS TIC chromatogram of a cured NBR sample (TCT:
thermal desorption cold trap injector). Reproduced with permission from
J. Sidwell in Proceedings of the Food Contact Polymers Conference, 21–22nd April,
Brussels, Belgium, Smithers Rapra, Shawbury, UK, 2009, Paper No.7.
©2009, Smithers Rapra [33]

LC–MS systems also assist in the analysis of vulcanised rubber to detect cure system
species because the types of species being investigated will be amenable to reverse-
phase HPLC. Examples of how LC–MS, in conjunction with GC–MS, has been
applied to food-contact rubbers (e.g., EPDM) as part of an overall analysis of the
low-MW species present are provided in Section 8.4.2.2. Another example is provided
in Section 9.4, where LC–MS, again used in conjunction with GC–MS, is employed
to identify the substances, including cure system species, that can be extracted from
an NBR compound.

It is also possible to use PyGC to obtain information on the type of cure (e.g., sulfur
or sulfurless) that has been used to cure rubbers. For example, a study carried out in
China [36] showed how PyGC can be used to distinguish between the different cure
systems employed to vulcanise NR. The method that was developed was based on
the approximate linear relationship between sulfur content and the relative areas of
characteristics Py products in the chromatogram. A standard deviation of 1.02 and
a coefficient of variation of 2.96% was claimed for the method.

137
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

4.3.3 Cure State and Cure-state Studies

The sections above have described the various analytical approaches that can be
used to identify and quantify the curatives and cure system components present in
rubber samples. Due to the large amount of information that has been published,
and its fundamental importance to rubber technology, rather than produce a sub-
section within this section, analogous to that on ageing studies that appears at the
end of the antidegradant and stabiliser section (i.e., Section 4.2.3), the subject of
cure state has been given its own section in this book: Chapter 6.

4.4 Determination of Miscellaneous Additives

4.4.1 Blowing Agents

There are two main types of blowing agent used with polymer systems physical and
chemical, and these are defined below:

1. Chemical type – chemical compounds (principally organic) that undergo

chemical decomposition at the processing temperature to form a gaseous species
(e.g., nitrogen) which will create the cellular structure. Examples include azo
compounds, nitroso compounds, sulfonyl hydrazide compounds and sodium
bicarbonate.

2. Physical type – low boiling point organic compounds (e.g., DCM or pentane)
which volatilise at the processing temperature to create the cellular structure.

In the case of rubbers, by far the most widely used types are chemical blowing agents,
the majority of which decompose during curing to produce nitrogen. The reactivity
of these compounds can make analysis difficult even if the rubber sample is uncured
(e.g., they will decompose in the injection port of a GC–MS instrument), and the
approach that is often employed to identify these additives is an indirect one via
the analysis of the decomposition products. A GC–MS instrument is the best choice
for this, used in the headspace mode, and an in-house library of typical breakdown
fragments obtained by the analysis of standard rubber compounds is required.

Quantification of blowing agents in an uncured product is very difficult because their

polarity and reactivity (e.g., hydrogen-bonding reactions and thermal instability) can
cause problems in their isolation from the compound and subsequent analysis. In
terms of the reverse engineering of commercial samples, this is unlikely to be required
because the product will invariably be in the blown state. For laboratories that carry

138
 Additives

out quality control, other properties such as density and expansion ratios can be used
to determine the level of a blowing agent in a compound.

If it is suspected that a physical-type blowing agent has been used to produce a foam,
then it is also possible to use headspace GC–MS to obtain information on it. Even
in the blown state, the high sensitivity of the technique enables an assignment of
the decomposition gas to be made from the small amount of the residual blowing
agent that will be present. It is uncommon for these types of blowing agents to be
used in rubber compounds because the curing/processing temperatures of standard
rubbers are usually too high to give controlled blowing. They tend to be used in
amorphous plastics (e.g., polystyrene) which have a relatively low Tg (≈100 °C).
Given this application, there is some scope, therefore, for their use in thermoplastic
rubbers.

Because the breakdown of a blowing agent (either physical or chemical type) is an

energetic event, it is possible to use DSC to study their performance within a rubber
compound. In the case of chemical types, an exotherm with result in the specific heat
trace and, with the physical types, an endotherm because they remove heat from the
sample to breakdown. By the use of reference samples, it is possible to use DSC to
establish the:

• Type of blowing agent

• Amount of a particular blowing agent

It is important when using DSC for this kind of work that, as with curing studies
(Chapter 6), sealed sample pans are used to prevent the gas that is generated escaping
and influencing the specific heat data. In addition, a major complication that occurs
with rubber samples that is rarely encountered with plastics is that, in order to achieve
a satisfactory product, the blowing reaction occurs within the same temperature
range as the curing reaction. The data obtained are, therefore, a composite of these
two processes and empirical steps have to be taken to try and separate out the two
events. One possible route is to analyse control samples that are equivalent in all
respects except that they do not contain any blowing agent. The obvious potential
flaw in this approach is that the blowing agent may influence the chemistry of the
cure as it breaks down (e.g., by changing the pH or undergoing specific reactions)
and so care has to be taken to obtain meaningful data.

4.4.2 Flame Retardants

There are two main categories of flame retardant: organic and inorganic. The type
of organic flame retardant present in a rubber compound can be determined in an

139
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

analogous way to the plasticiser-type additives (Section 3.4) because this type of flame
retardant is usually similar in MW and polarity to a number of these substances and,
hence, they have similar properties (e.g., extractability and amenability to analysis
by IR).

Once the type of organic flame retardant has been determined by FTIR spectroscopy,
an accurate quantification can be obtained by carrying out an elemental analysis for a
characteristic element (e.g., halogen or phosphorus) on the un-extracted sample. The
extract value itself will rarely provide an accurate quantification because other species,
particularly any plasticiser/oil, will contribute to it as well. In some instances (e.g.,
organophosphates), it may be possible to use an HPLC of the extract to quantify the
flame retardant. GC-based techniques can also be used in some cases, but bromide
types present problems for the GC-based techniques because of their relatively high
MW (>900).

Care must be taken over the quantification of inorganic flame retardants because
some of them (e.g., antimony trioxide) can react with any organic flame retardant
present, or breakdown to produce volatile products (e.g., hydrated alumina), under
quantitative ashing conditions, and also during analysis by thermal techniques such
as TGA. A good initial approach involves carrying out a semi-quantitative elemental
analysis on the rubber sample by X-ray fluorescence spectroscopy to see which type(s)
of flame retardant are present. Accurate quantifications can then be obtained by
precise elemental determinations.

Some inorganic flame retardants in a rubber sample can be quantified by making use
of their specific characteristic properties once their presence has been confirmed. For
example, if it is known that hydrated alumina is present in a sample, it is possible to
quantify it reasonably well using TGA. This can be done by determining the early
weight loss event that results from the loss of the water of hydration from the hydrated
alumina and using the fact that this additive loses close to 35% of its original weight
when it is heated to 550 °C.

It is possible to use other techniques to study flame retardants. Yang and co-workers
[37] used a diverse combination (e.g., X-ray diffraction, SEM, TGA and oxygen index
measurements) to study the morphological, mechanical, thermal and flammability
properties of montmorillonite (MMT)-filled silicone rubber samples containing
the synergistic flame retardants magnesium hydroxide and red phosphorous. The
results showed that it required only 1% of the MMT nanomaterial to increase the
decomposition temperature of the silicone rubber above the value of the control sample
and that this high thermal stability was matched by good flame-retardant properties.

140
 Additives

Ismawi and co-workers [38] evaluated the effects of various flame-retardant additives
on the flammability and other properties of NR. The flame retardants that were used
in the work, alone or in combination with one another, included the following:

• Decabromodiphenyl oxide

• Antimony trioxide

• Aluminium trihydroxide

• Organoclay

• Zinc hydroxystannate

• Chlorinated paraffin

The properties of the resulting NR samples were investigated by vertical burning

tests, SEM, TGA, smoke density measurements, determination of the limiting oxygen
index value, and cure characterisation work.

4.4.3 Process Aids

4.4.3.1 Plasticisers and Oils

The addition of hydrocarbon oil or a synthetic plasticiser (e.g., dioctyl phthalate)

into a compound will improve its processability, in addition to changing a wide range
of physical properties. The techniques used to identify and quantify these types of
process aid are described in Section 3.4.

4.4.3.2 Resins

High-styrene resins are thermoplastic polymers of styrene and butadiene, and they
are used to modify the hardness of rubber compounds. They are used in preference
to high levels of CB because they do not increase the processing viscosity to the same
extent. The high level of styrene (≤85%) makes them easy to detect by looking for
their characteristic Tg by DMTA or thermal mechanical analysis. The DMTA trace of
a SBR containing a high-styrene resin is shown in Figure 4.9. The Tg for the rubber
(at about −20 °C) and the Tg for the resin (at ≈50 °C) are clearly visible in the data.
PyGC–MS is also an option and can be employed in way that is analogous to the
identification of polymers (Section 4.2) by looking for characteristic monomer species

141
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

and other Py products in the pyrogram. This approach can obviously be compromised
if the high-styrene resin is in a SBR rubber compound.

Log E' (PA) Tan δ

9.75
.9
9.25 DMA

8.75 .7

8.25
.5
7.75
1 Hz
7.25 .3
3 ºC/min
Double
6.75 cantilever
.1
6.25

-100 -50 0 50 100 150 200

Temp (ºC)

Figure 4.9 DMTA trace of a SBR containing a high-styrene resin. Reproduced with
permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds and
Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001,
p.30. ©2001, Smithers Rapra [39]

Coumarone resins are manufactured by the polymerisation of styrene, coumarone

and indene, and are used as tackifiers and plasticisers. Varying the polymerisation
conditions leads to a range of resins having melting points in the range 65–110 °C.
If sufficient resin is present in a sample, its melting point can be detected by DSC.
PyGC–MS can also be used to detect the presence of these resins in a compound.

Thermosetting alkyl phenol formaldehyde resins can be used instead of the high-
styrene type to give excellent flow characteristics in moulding and extrusion. Non-
reactive phenolic resins can be used as tackifiers instead of coumarone resins. With

142
 Additives

regard to analysis, phenolic fragments can be detected in both solvent extracts and
pyrograms by GC–MS.

4.4.3.3 Pine Tar and Factice

Pine tar can be used as a processing aid in rubber compounds and an analysis of the
solvent extract of a rubber sample by GC–MS will reveal diagnostic species (e.g.,
pinene) if this additive is present.

Factice, of which there are both ‘white’ and ‘brown’ versions depending upon their
colour and manufacturing history, is used in rubber compounds and is particularly
good at controlling the amount of die swell that occurs during extrusion. It is produced
by crosslinking vegetable oil with sulfur and so a relatively-high sulfur content will give
a good indication as to whether it is present or not in a rubber compound. Whether
any sulfur has been contributed by a sulfur-based cure system will obviously have
to be considered.

The presence of both pine tar oil and factice in a sample can be detected by the use of
nuclear magnetic resonance due to its relative sensitivity and diagnostic capabilities,
whereas FTIR may have problems in providing a definitive result.

4.4.4 Peroxide Co-Agents

Peroxide co-agents such as TAC and triallyl isocyanurate can be detected in solvent
extracts of either the unvulcanised or vulcanised rubber by GC–MS because a
sufficient amount of these additives remains in the ‘unbound’ state (i.e., not part
of the crosslinked matrix) in the case of vulcanised samples. It is also possible to
detect these species by headspace GC–MS because they are reasonably volatile and
thermally stable.

4.4.5 Stearic acid

Stearic acid is an important ingredient in rubbers, being used in the majority of

sulfur-cured compounds due to its role as a co-agent, along with zinc oxide, in the
cure system. The quantitative analysis of stearic acid and its associated fatty acids
(the industrial chemical usually also contains palmitic acid and myristic acid) can,
therefore, be important from a quality-control viewpoint and Watanabe and co-
workers [40] investigated two analytical methodologies for this purpose. In place of
the conventional liquid-phase extraction procedure, the research team determined

143
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

the fatty acids in an SBR sample directly by thermal desorption (TD) GC–MS and
thermally-assisted hydrolysis and methylation (THM) GC–MS. The results showed
that the precision of the analytical data was only fair (i.e., relative standard deviation
(RSD) of 7.8%) for the TD GC–MS method due to the interaction between the polar
fatty acids and the basic sites in the chromatographic system. The problem with
interactions was overcome with the THM GC–MS approach because the fatty acids
were derivatised to the methyl esters using tetramethylammonium hydroxide and the
average result of 0.62% (RSD of 3.2%) of total fatty acids in the SBR agreed well
with the actual, added weight of fatty acids of 0.64%.

4.5 General Analysis Work for the Identification and

Quantification of Additives

Sections 4.2–4.4 have addressed the determination of different additives according

to their specific classes (e.g., UV stabiliser, antioxidant, antiozonant). This section
includes the techniques and methods used by industrial scientists and researchers to
provide more general information, including information on the:

• Additives present in a rubber for the purposes of reverse engineering.

• Function, behaviour and performance of additives to assist with degradation

studies and curing studies.

The examples given below are taken from studies published within the last 15 years
and have been selected because they are considered to provide a good overview to
the industrially relevant work that is possible using the technology that is presently
available.

A research team from Brazil [41] used a Py-based technique to characterise a rubber
compound. In this case, FTIR was employed to investigate the gaseous Py products
of two EPDM rubbers with a view to evaluating its potential to characterise the
additives that were present within the samples. Both of the EPDM rubbers contained
additives that would usually be expected to be employed in such a material and Py–
FTIR characterisation data were obtained for the following ingredients:

• Paraffin oil

• Stearic acid

• TMQ

• TMTM

144
 Additives

• Tetraethylthiuram disulfide (TETD)

• MBT

The work undertaken by this research team also included a comparative study involving
the characterisation of uncompounded EPDM, unvulcanised EPDM and vulcanised
EPDM. The Py–FTIR results obtained showed that the vulcanisation process did
not interfere with the data and that it was possible to identify the functional groups
of the additives in both cases and, thus, eliminated the need for a solvent extraction
process to isolate them. As well as being shown as a fast and cost-effective analytical
method for the determination of additives, another benefit was that it was sensitive
enough to be capable of detecting sulfur-containing additives down to 1.26% (1.4
phr) in both unvulcanised and vulcanised EPDM. However, one limitation was its
inability to differentiate between TMTM and TETD because of overlapping within
the FTIR spectral information.

Il’yasov and co-workers [42] investigated the high-temperature volatility of 39 ingredients

that can be used in different tyre rubber formulations (e.g., tread, sidewall). The research
team used a gravimetric method that was capable of the quantitative analysis of the
ingredients when they were heated under different regimens. The ingredients that were
studied in this way included fillers, curing agents, resins, plasticisers, antioxidants,
activators and modifiers, and the research team thought that the results could be used
to develop a procedure for determining the weight loss of rubber mixes during their
preparation and processing so that the formulation could be optimised.

FTIR, in its various forms, is an extremely versatile technique and its use to identify
polymers is covered in some detail in Section 3.2. Tikhomirov and Kimstach [43]
showed how it can also be used for the quantitative and qualitative analysis of
polymers and their additives for quality control and other purposes. Their overview
of the subject provides examples of the application of a number of FTIR-based
techniques, including:

• Attenuated total reflection spectroscopy

• Diffuse-reflection FTIR

• GC–FTIR

• Near IR

Kuzdzal [44, 45] described the advantages of using LC–MS-ion-trap-ToF–MS for the
identification and structural analysis of polymer additives, such as plasticisers and
viscosity modifiers. The application of the technique to different scenarios for the
purposes of quality control, quality assurance of raw materials, migration testing, or

145
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

chemical modification of additives due to degradation or oxidation was presented.

Two of the scenarios included were the important areas of the potential interaction
and migration of species between packaging and drug products as well as food-contact
materials and food products.

As Chapter 2 demonstrates, many analytical methods are available for the study and
characterisation of rubbers and their compounds and products, including a number
that are interfaced with MS, for example, GC–MS, LC–MS, Py–MS, and TGA–MS.
These methods work well but some, particularly if initial preparation steps such as
solvent extraction are taken into account, can be relatively time-consuming and so
not always suited to a high-throughput industrial laboratory. The search for fast and
accurate analytical methods is, therefore, always of interest, and Trimpin and co-
workers [46] have discussed the use of three such methods. All three of these rapid
methods are based on MS and are:

• Multi-sample matrix-assisted laser desorption/ionisation (MALDI) MS – for the

analysis of low-MW polymers.
• Atmospheric pressure (AP)-solids analysis probe MS – for the analysis of additives.
• Atmospheric pressure Py MS – for the determination of polymer type.
The last two methods are stated as being able to provide information on a sample
regardless of its composition or the MW of the polymer.

Hakkarainen [47] published a review of the use of the solid-phase microextraction

technique in different polymer-related applications. Within the review, the analysis of
various polymer additives was discussed, for example, plasticisers, flame retardants
and the extraction of polymer additives from biological fluids. The analysis of
polymer degradation products was also covered, with reference to thermo-oxidation,
photooxidation and thermal oxidation products and the degradation products of
degradable polymers. In addition, the analysis of residual monomers and residual
solvents was included, as was the analysis of migrants from food-contact materials,
pharmaceutical packaging and medicinal materials.

Buchberger and Stiftinger [48] reviewed the methods available for the analysis of the
stabilisers present in polymeric materials. In their review they made the following points:

• The degradation products of stabilisers should be identified to understand the

reactions that are occurring.

• Adequate analytical methods are required to ensure the optimisation of the

performance of a polymer in service and ensuring that its manufacturing costs
are well controlled.

146
 Additives

• LC–MS with atmospheric pressure ionisation (API) has become the state-of-the-art
for the identification of the additives and components within polymeric materials.

• The API techniques that are available include: electrospray ionisation, AP

photoionisation and API.

• API has the advantages of low detection limits and wide applicability in terms of
different additive structures.

• As an alternative to combinations of chromatography and MS (e.g., LC–MS),

direct mass spectrometric techniques for solid samples are emerging because they
offer advantages in terms of speed (e.g., for screening work) and avoidance of
time-consuming preparation work.

• In addition to chromatography, capillary electrophoresis has demonstrated potential

for the separation of polymer stabilisers and the characterisation of polymers.

A review provided by Geissler [49] described the use of PyGC–MS to quantitatively

analyse the additives in polymers.

Aminlashgari and Hakkarainen [50] reviewed the emerging mass spectrometric

techniques available for the analysis of polymers and polymer additives. The emerging
techniques that are covered in this review include:

• New developments in laser desorption ionisation techniques (e.g., solvent-free

MALDI and surface-assisted laser desorption ionisation).

• Developments in secondary-ion mass spectrometry (SIMS) such as gentle-SIMS

and cluster SIMS.

• Desorption electrospray ionisation MS and direct analysis in real time MS for

the analysis of solid samples.

• Ion-mobility spectrometry mass spectrometric analysis for evaluation of complex

structures.

Although the potential of these analytical techniques still has to be fully explored,
the authors stated that they will strengthen the position of MS as an irreplaceable
tool for polymer characterisation.

Sanches and co-workers [51] used TGA–FTIR to analyse both unvulcanised and
vulcanised EPDM rubbers with the objective of identifying some of the additives that
are frequently used in these compounds. The data generated on standard compounds
showed that the technique was capable of detecting the following additives at a level
of 0.7 phr (0.63%):

147
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• TMTM

• MBT

However, the technique was found to have some limitations because it was not capable
of detecting a number of other additives, including:

• TMQ

• Paraffin oil

• Stearic acid

Despite these limitations, the authors believed that the technique had value because
it was particularly suited for the detection of sulfur-containing additives, and the fact
that no initial solvent extraction procedure was required reduced the time and effort
needed to obtain compositional information on rubber compounds.

References

1. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.55

2. R.P. Lattimer in Proceedings of the ACS 132nd Fall Meeting, 6–9th October,
Cleveland, Ohio, USA, American Chemical Society Rubber Division,
Washington, DC, USA, 1987, Paper No.86.

3. T.R. Crompton in Determination of Additives in Polymers and Rubbers,

Rapra Technology, Shawbury, UK, 2007.

4. J.C.J Bart, Polymer Degradation and Stability, 2003, 82, 2, 197.

5. J.A. Sidwell in High Performance Liquid Chromatography – Analytical

Applications in the Rubber and Plastics Industries, Members Report
Number 49, Rapra Technology Ltd, Shawbury, UK, 1980.

6. A.B. Sullivan, G.H. Kuhls and R.H. Campbell, Rubber Age, 1976, 108, 3, 41.

7. W.C. Warner, Journal of Chromatography, 1969, 44, 315.

8. J.G. Kriener, Rubber Chemistry and Technology, 1971, 44, 381.

9. K. Nagasawa and K. Ohta, Rubber Chemistry and Technology, 1969, 42,

625.

148
 Additives

10. H.B.S. Conacher and B.D. Page, Journal of Chromatographic Science, 1979,
17, 188.

11. M.J. Forrest, S.R. Holding and D. Howells, Polymer Testing, 2006, 25, 63.

12. M.J. Forrest, S. Holding and D. Howells in Proceedings of the High

Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2.

13. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.26.

14. E. Klein, Z. Cibulkova and V.A. Lukes, Polymer Degradation and

Stabilisation, 2005, 88, 3, 548.

15. Z. Cibulkova, P. Simon, P. Lehocky and J. Balko, Polymer Degradation and

Stabilisation, 2005, 87, 3, 479.

16. D.F. Parra and J.R. Matos, Journal of Thermochemistry and Calorimetry,
2002, 67, 2, 287.

17. E.L.G. Denardin, D. Samos, P.R. Janissek and G.P. Souza, Rubber Chemistry
and Technology, 2001, 74, 4, 622.

18. C.D, Gamlin, N.K. Dutta and N.R. Choudhury, Polymer Degradation and
Stabilisation, 2003, 80, 3, 525.

19. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbott, Blackwell Publishing Ltd, Oxford, UK, 2008, p.225.

20. V.K. Skachkova, N.A. Erina, L.M. Chepal and E.V. Prut, Polymer Science
Series A, 2003, 45, 12, 1220.

21. F. Pruneda, J.J. Sunol, F. Andreu-Mateu and X. Colom, Journal of Thermal

Analytical Colorimetry, 2005, 80, 1, 187.

22. T. Ogawa, T. Yamagata, T. Hara, S. Osawa, Y. Yoshida and M. Kanazawa,

Kobunshi Ronbunshu, 2003, 60, 2, 64.

23. M. Hakkarainen, A-C Albertsson and S. Karlsson, International Journal of

Polymer Analysis and Characterisation, 2003, 8, 4, 279.

24. R. Yang, Z. Jiaohong and Y. Liu, Polymer Degradation and Stability, 2013,
98, 12, 2466.

149
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

25. M. L. Tawfic and A.I. Hussein, Kautschuk, Gummi, Kunststoffe, 2015, 68, 9,
30.

26. B.G. Willoughby and K.W. Scott in Nitrosamines in Rubber, Rapra

Technology Ltd, Shawbury, UK, 1997.

27. D. Gross and K. Strauss, Kautschuk, Gummi, Kunststoffe, 1979, 32, 1, 18.

28. G.P. McSweeny, Journal of the Institute of the Rubber Institute, 1970, 4, 243.

29. D.O. Hummel and F.K. Scholl in Atlas of Polymer and Plastics Analysis,
Volume 3 – Additives and Processing Aids: Spectra and Methods of
Identification, 2nd Revised Edition, Verlag Chemie GmbH, Weinheim,
Germany, 1981.

30. F. McLafferty and D.B. Stauffer in Wiley/NBS Registry of Mass Spectral

Data, John Wiley and Sons, New York, NY, USA, 1989.

31. B. Willoughby in Rubber Fume: Ingredients/Emissions Relationships,

Rapra Technology Ltd, Shawbury, UK, 1994.

32. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.28.

33. J.A Sidwell in Proceedings of the Food Contact Polymers Conference,

21–22nd April, Brussels, Belgium, Smithers Rapra, Shawbury, UK, 2009,
Paper No.7.

34. M.J. Forrest in Food Contact Rubbers 2 – Products, Migration and

Regulation, Rapra Review Report No.182, Smithers Rapra, Shawbury, UK,
2006, p.28.

35. M.J. Forrest in Food Contact Rubbers 2 – Products, Migration and

Regulation, Rapra Review Report No.182, Smithers Rapra, Shawbury, UK,
2006, p.29.

36. Z. Naidong and Z. Qingshan, China Rubber Industry, 1991, 38, 1, 33.

37. L. Yang, Y. Hu, H. Lu and L. Song, Journal of Applied Polymer Science,

2006, 99, 6, 3275.

38. D.H.A. Ismawi, J.F. Harper and A. Ansarifar, Journal of Rubber Research,
2008, 11, 4, 223.

150
 Additives

39. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.30.

40. A. Watanabe, C. Watanabe, R. Freeman, M. Nakajima, N. Teramae and

H. Ohtani, Rubber Chemistry and Technology, 2014, 87, 3, 516.

41. N.B. Sanches, S.N. Cassu, M.F. Diniz and R. Dutra de Cassia Lazzarini,
Polimeros: Ciencia e Tecnologia, 2014, 24, 3, 269.

42. R.S. Il’yasov, E.G. Mokhnatkina, E.E. Potapov and E.V. Sakharova,
International Polymer Science and Technology, 2009, 36, 11, T/29.

43. S.V. Tikhomirov and T.B. Kimstach, International Polymer Science and
Technology, 2008, 35, 8, T/59.

44. S. Kuzdzal, Rubber World, 2010, 241, 4, 19.

45. S. Kuzdzal, Adhesives and Sealants Industry, 2010, 17, 3, 21.

46. S. Trimpin, K. Wijerathne and C.N. McEwen, Analytica Chimica Acta, 2009,
654, 1, 20.

47. M. Hakkarainen in Advances in Polymer Science, Eds., A. Alberson and

M. Hakkarainen, Springer-Verlag, Berlin, Germany, 2008, 11, 23.

48. W. Buchberger and M. Stiftinger, Advances in Polymer Science, 2012, 248,

39.

49. M. Geissler in Proceedings of the 7th European Conference on Additives

and Colors, 16–17th March, Bonn, Germany, Society of Plastics Engineers,
Antwerp, Belgium, 2011, Paper No.3.

50. N. Aminlashgari and M. Hakkarainen, Advances in Polymer Science, 2012,

248, 1.

51. N.B. Sanches, S.N. Cassu and R.L. Dutra, Polimeros, 2015, 25, 3, 247.

151
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

152
5
Reverse Engineering and Product
Deformulation

5.1 Introduction

One of the most frequent tasks asked of a rubber analyst is to deformulate a rubber
compound in order to obtain as much information of the types of ingredients that
have been used in its manufacture and the level of these ingredients. This work
can be regarded as being analogous to being asked to piece together a jigsaw and
is a satisfying and challenging task to undertake, particularly if the rubber analyst
has sufficient knowledge of rubber technology to compliment his/her skills as an
investigative analyst.

In order to provide the greatest opportunity for success in this area it is important that
in addition to the skills and knowledge mentioned above that the rubber analyst has
access, either solely in-house, or from a combination of in-house capability and that
of a trusted service provided, of a full range of analytical techniques. These should
include as a minimum:

• Fourier–Transform infrared (FTIR) spectroscopy for providing information on

the generic type of base polymer(s), bulk additives (plasticiser/oil and inorganic
fillers), and providing an initial insight into the complexity of the sample.

• Nuclear magnetic resonance (NMR) to provide information on the polymer

microstructure (e.g., ratio of comonomers, cure site monomers) and, in the case
of polymer blends, approximate blend ratio.

• Thermal techniques that include thermogravimetric analysis (TGA) for bulk

composition analysis and dynamic mechanical analysis for determination of
glass transition temperature and generation of modulus versus temperature
plots.

• Chromatographic techniques that are capable of identifying and quantifying

substances possessing a range of molecular weights (MW) and degrees of thermal
stability, for example, gas chromatography (GC)–mass spectrometry (MS) and
liquid chromatography (LC)–MS.

153
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• An elemental technique that can be used in a semi-quantitative, scanning mode,

such as X-ray fluorescence spectroscopy (XRF), to assist in identifying the type(s)
or inorganic additives and fillers present.

The combined capability of the techniques stated above will provide a significant
amount of information on a sample, but they may have to be complimented by some
of the other specific or specialist techniques that are described in Chapter 2 depending
upon the particular rubber compound under investigation and how the deformulation
work progresses. Typical examples of these techniques include X-ray diffraction
to determine the specific type of a particular inorganic filler (e.g., a silicate), or a
quantitative elemental technique [e.g., inductively coupled plasma (ICP)] to quantify
a metal or a hetero-element as a means to quantify a polymer in a blend or the level
of a specific additive. It is also often possible to obtain additional compositional
information by complementing the chemical analysis work with a physical test. For
example, combining the bulk compositional data obtained from TGA with a hardness
determination can help in deciding the approximate type of carbon black (CB) used
by referring to the compounding guidelines of rubber technologists.

There can be a number of reasons why a reverse engineering programme is embarked

upon, including:

• To ensure the correct formulation was used initially because a product has failed
in service.

• To assist in a research and development programme by obtaining information on

products that are already in the market place.

• To investigate the regulatory compliance of part of the formulation (e.g., cure

system ingredients).*

*Particular care needs to be taken in this case due to the limitations discussed
below.

It is important to understand the limitations associated with these types of investigative

study in order to ensure that unrealistic goals are not laid down at the outset. In the
first two cases listed above, experience has shown that it is not possible from analytical
work alone to determine with 100% accuracy the complete composition of the vast
majority of rubber compounds. Although the degree of success depends upon a number
of factors (see below), as a rough estimate, it is usually possible to obtain between 70
and 85% of the compositional information. These comments apply even if a substantial
budget, extended time-scale and wide range of modern analytical instrumentation
are available to a knowledgeable and experienced analyst. They also apply to both
unvulcanised and vulcanised samples.

154
 Reverse Engineering and Product Deformulation

There are a number of reasons why there are practical restrictions on the degree of
success, and some of these are listed below:

• Presence of thermally-labile substances (e.g., cure system compounds) that will

breakdown during vulcanisation and often during analysis (e.g., during a solvent
extraction step or when analysed using instruments such as GC–MS).

• Presence of substances that have no single diagnostic species, a good example

being oligomeric antioxidants.

• Presence of additives that are volatile, or have a volatile fraction, and so are lost
to a degree during vulcanisation or analysis.

• Variations in the manufacturing specifications of some ingredients (e.g., the

amount of bromine and chlorine in halobutyl rubbers due to variable amounts
of cure site monomer being used in different grades).

• Complications arising from the presence of two or more ingredients of the same
‘type’ being used, for example, process oils, CB, inorganic fillers, and a blend of
two grades of polymer of the same type [e.g., styrene-butadiene rubber (SBR)].

• The fact that only the chemical identify of substances can be determined not the
specific manufacturers grade that has been used.

• Once vulcanised, it is not possible to determine the MW characteristics of the

original polymer(s) and, even in the unvulcanised state, complete isolation of the
polymer(s) from the other ingredients in a rubber compound may be difficult due
to strong polymer–filler interactions.

It is these types of challenges that can cause problems if reverse engineering work is
used to try to assess compliance with regulations and regulatory guidelines even in a
relatively minor way (e.g., cure system ingredients) and mean that it is not possible
to use reverse engineering to show complete compliance with, say, a ‘positive list’ of
ingredients in a food-contact regulation (e.g., BfR Recommendation XXI) because
substances that were not on the list could have been missed.

Specific applications of the mainstream spectroscopic, chromatographic, thermal

and elemental techniques to partial or extensive reverse engineering-type work
is covered in other chapters of this book, particularly Chapters 3 and 4. Taken
together, these show that by using combinations of these techniques a significant
amount of qualitative and quantitative information can be obtained on the bulk
components (i.e., plasticiser/oil, polymer and filler) and specific classes of additives
(e.g., antidegradants and curatives).

155
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

As mentioned above, many examples of the specific aspects of reverse engineering

work are covered in other chapters in this book and so, rather than repeat some of
them in this chapter, we will instead review the capability of one of the more advanced
versions of the mainstream techniques of which the author has had personal experience
(Section 5.2). It is for this reason, as well as the fact that it is widely applicable to
rubber compounds and products and offers advantages in this area, that this technique
has been chosen and not because it is necessarily any better than the many others
that are available and which are overviewed in Chapter 2.

This chapter also uses an in-depth case study to provide a practical illustration of an
approach to reverse engineering that can be taken with a complex, real-life sample
(Section 5.3). In addition, some published examples of reverse engineering work
that has been carried out on rubbers and related polymer systems are provided in
Section 5.4. For additional assistance on how to undertake these types of exercises,
reference books that have been published on this subject include those of Gupta and
co-workers [1] and Scheirs [2].

With regard to guidance as to what additives and ingredients are being used in the
rubber industry, there are a number of books that provide extensive coverage of the
range of additives and chemicals that can be compounded into rubber products [3, 4].

5.2 Use of GC×GC–ToF–MS for Reverse Engineering

Despite its high-resolution, GC–MS can encounter problems in trying to resolve all
of the components present in rubber extracts. This is particularly true if the sample
contains a relatively large amount of a multi-component species, such as a mineral
oil. Inevitably, there is a considerable amount of co-retention in the middle part of
the GC chromatogram and the ability to identify components within this region can
be severely compromised, thereby limiting the success of reverse engineering work.

The GC×GC–time-of-flight (ToF)–MS technique offers the rubber analyst a greater

potential capability for reverse engineering work and the background to how it
delivers this is covered in Section 2.5.3. One of the principal features of the technique
is that two types of GC column are used and this enables separations to be performed
according to both volatility and polarity. Powerful deconvolution software also aids
the processing and interpretation of the data.

The advantages of using GC×GC–ToF–MS in reverse engineering work are highlighted

in this chapter by using the results obtained on five food-grade, high-performance
rubber samples [5]. Firstly, the reverse engineering data that can be generated by
the GC×GC–ToF–MS are illustrated using a hydrogenated-nitrile rubber (NBR)

156
 Reverse Engineering and Product Deformulation

compound as an example. The compound was a peroxide-cured, hydrogenated NBR,

designated 349N, which had the formulation shown in Table 5.1.

Table 5.1 Formulation of peroxide-cured hydrogenated NBR – 349N

Ingredient phr
Zetpol® 2000L (hydrogenated NBR) 100
High-abrasion furnace N330 (CB filler) 35
ZnO 3
Perkadox 14/40 [1,3-bis(tert-butyl-peroxy-isopropyl)benzene]
®
6
Antioxidant 2246 [2,2'-methylene-bis(4-methyl-6-tert-butylphenol] 1
ZnO: Zinc oxide
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

A test sheet was prepared from the hydrogenated NBR 349N compound by curing
a portion for 22 min at 165 °C, conditions which gave a degree of cure of ≈90%,
as determined by the moving die rheometer (MDR), of the theoretical maximum.

A solvent extract was obtained for the analysis by finely cutting up 0.3 g of the cured
test sheet, placing it into a vial along with 2 ml of acetone, and then subjecting it to
ultrasonic agitation for 30 min. At the end of this period, the extract was removed
and placed into a GC sample vial. A 20-ppm eicosane ‘internal standard’ was included
in the acetone used for the extraction. Acetone was the preferred choice of solvent
because it is a good, non-selective solvent for the removal of low-MW compounds
from a wide range of rubber types. It is also a good solvent for GC–MS-type work
due to its relatively high volatility and reasonable stability (i.e., it does not contain
stabilisers that may interfere with analyte peaks).

The acetone extract of the hydrogenated NBR 349N compound was analysed by
GC×GC–ToF–MS using the following experimental conditions:

• Instrument: Agilent 6890 gas chromatograph with LECO Pegasus III GC×GC–
ToF–MS.

• Injection: PTV injection, 10 °C above primary oven temperature; 1 µl.

• Primary column: J&W scientific HP-5MS (30 m × 0.250 mm with film thickness
of 0.25 µm).

157
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Secondary column: SGE BPX-50 (1.8 m × 0.100 mm with film thickness of 0.10 µm).

• Carrier gas: helium at 1.0 ml/min with constant flow.

• Primary oven programme: 40 °C for 10 min, then 10°C/min to 320 °C and hold
for 15 min.

• Secondary oven programme: 50 °C for 10 min, then 10 °C/min to 330 °C and

hold for 15 min.

• Modulator offset: 30 °C.

• Modulator frequency: 4 s.

• Hot time: 0.30 s.

• Mass spectrometer: scanning 30 to 650 Da at 76 spectra/s.

The data generated by the GC×GC–ToF–MS analysis of the acetone extract are shown
in Figures 5.1–5.3.

3
2n 2365
d 2
Co
lu T
m mnR
n 1365 olu
RT 1 C
1st

0 365

Figure 5.1 Three-dimensional (3D) GC×GC–ToF–MS total ion current (TIC)

chromatogram of hydrogenated NBR 349N acetone extract (RT: retention time).
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in
Proceedings of the High Performance and Specialty Elastomers, 20–21st April,
Geneva, Switzerland, Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2.
©2005, Rapra Technology Ltd [6]

158
 Reverse Engineering and Product Deformulation

Increasing
2nd Column RT

polarity

Column bleed

0
365 1365 1st Column RT 2365

Figure 5.2 Plan view of GC×GC–ToF–MS TIC chromatogram of hydrogenated

NBR 349N acetone extract – species due to ‘column bleed’ and a polarity marker
are shown. Reproduced with permission from M.J. Forrest, S. Holding and
D. Howells in Proceedings of the High Performance and Specialty Elastomers,
20–21st April, Geneva, Switzerland, Rapra Technology Ltd, Shawbury, UK, 2005,
Paper No.2. ©2005, Rapra Technology Ltd [6]

3
2nd Column RT

0
365 1365 1st Column RT 2365

Figure 5.3 Plot view of GC×GC–ToF–MS TIC chromatogram of hydrogenated

NBR 349N acetone extract with peak markers. Reproduced with permission from
M.J. Forrest, S. Holding and D. Howells in Proceedings of the High Performance
and Specialty Elastomers, 20–21st April, Geneva, Switzerland, Rapra Technology
Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

159
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

For comparison, the acetone extract from hydrogenated NBR 349N was also analysed
by conventional GC–MS, using the conditions below, and the results obtained are
shown in Figure 5.4:

• Instrument: Agilent 6890/5973 GC–MS.

• Injection: 1 µl splitless at 310 °C.

• Carrier: helium at 1.0 ml/min.

• Column: Restek RTX-5 amine (30 m × 0.32 mm with film of 1.0 µm).

• Column oven: 40 °C for 5 min, then 20 °C/min increase up to 300 °C held for
15 min (30-min run time).

• MS: scanning 35 to 650 Da every 0.25 s.

Abundance
TIC: 0904045.D
1e+07
9500000
9000000
8500000
8000000
7500000
7000000
6500000
6000000
5500000
5000000
4500000
4000000
3500000
3000000
2500000
2000000
1500000
1000000
500000
0
Time--> 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00

Figure 5.4 Conventional GC–MS TIC chromatogram of a hydrogenated NBR

349N acetone extract. Reproduced with permission from M.J. Forrest, S. Holding
and D. Howells in Proceedings of the High Performance and Specialty Elastomers,
20–21st April, Geneva, Switzerland, Rapra Technology Ltd, Shawbury, UK, 2005,
Paper No.2. ©2005, Rapra Technology Ltd [6]

160
 Reverse Engineering and Product Deformulation

The conventional GC–MS chromatogram (Figure 5.4) confirmed that this technique
would struggle to separate a large number of substances when they elute closely
together.

The solvent extracts from a further four food-grade, high-performance rubbers

were also analysed using both GC×GC–ToF–MS and conventional GC–MS. The
formulations of these four compounds are shown in Tables 5.2–5.5.

Table 5.2 Formulation of fluorocarbon rubber – 49V

Ingredient phr
Viton TM
GBL 200 (peroxide-crosslinkable fluorocarbon rubber tetrapolymer) 100
ZnO 3
Medium thermal N990 (CB filler) 30
Triallyl cyanurate 3
Luperco 101XL [2,5-dimethyl-2,5-(di-tert-butylperoxy)hexane] 4
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

Table 5.3 Formulation of acrylic rubber – Vamac® 8

Ingredient phr
Vamac G [ethylene-methyl acrylate (acrylic) rubber] 100
Stearic acid 1.5
FEF N550 (CB filler) 50
Diak No.1 (hexamethylenediamine carbamate)
™
1.5
DOTG 4
Antioxidant 2246 1
FEF: Fast extrusion furnace
DOTG: N,N'-di-ortho-tolyl guanidine
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

161
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 5.4 Formulation of hydrin rubber – 20D

Ingredient phr
Hydrin 200 [epichlorohydrin-ethylene oxide copolymer (hydrin) rubber] 100
FEF N550 (CB filler) 50
Maglite® DE (magnesium oxide) 7
ZnO 3
Diak No.1
™
1.5
Antioxidant 2-mercaptobenzimidazole 2
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

Table 5.5 Formulation of butyl rubber 50B

Ingredient phr
Butyl 268 [a isobutylene-isoprene (butyl) rubber] 100
ZnO 3
Stearic acid 1
Vistanex LM-MS (low-molecular-weight butyl rubber process aid)
™
10
Intermediate super-abrasion furnace N220 (CB filler) 50
Sulfur 1
Rhenogran MPTD 70 (dimethyldiphenylthiuram disulfide)
®
1.43
DOTG 0.3
Nocrac AW (6-ethoxy-1,2-dihydro-2,2,4-trimethylquinoline) 1
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

The cure temperatures and times used to produce test sheets from these four rubber
compounds are given in Table 5.6.

162
 Reverse Engineering and Product Deformulation

Table 5.6 Curing conditions for the four rubber compounds (Tables 5.2–5.5)
Rubber compound Cure conditions*

Hydrin rubber – 20D 31 min at 170 °C

Fluorocarbon rubber – 49V 11 min at 170 °C

Butyl rubber – 50B 21 min at 155 °C

Acrylic rubber – Vamac® 8 25 min at 170 °C

*These curing conditions gave a degree of cure of around 90%, as determined by MDR, of the
theoretical maximum
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

The five rubber compounds shown above represent a range of high-performance

elastomers that could be used for food-contact type applications. They display a
range of complexity, with the hydrogenated NBR 349N having the greatest number
of constituents within it. In all five cases, the greater diagnostic power of the
GC×GC–ToF–MS technique resulted in a larger number of species being detected in
the acetone extract. Hence, more compositional information was provided compared
with conventional GC–MS, which is invaluable in reverse engineering work where
the degree of success is dependent on a good set of diagnostic compounds.

The difference in performance between GC×GC–ToF–MS and conventional GC–MS

is demonstrated in Table 5.7, where a comparison of the approximate number of
diagnostic substances found in the acetone extract of the five rubber compounds is
shown. Diagnostic substances can be regarded as those that are the most useful in
revealing information on the original formulation of a rubber. They can be defined as:

• Compounds that are the known breakdown products of certain classes of additive
(e.g., curatives and antidegradants).

• Original ingredients in their unchanged form.

This definition of diagnostic substances does not include non-specific compounds

(e.g., general aliphatic and aromatic hydrocarbons) and other common compounds,
such as siloxanes, which can originate from a number of sources, including processing
materials (e.g., moulding-release agents).

163
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 5.7 Comparison of the number of diagnostic substances found by

GC×GC–ToF–MS and conventional GC–MS
Rubber compound GC×GC–ToF–MS Conventional GC–MS

NBR – 349N >35 >30

Fluorocarbon – 49V >20 >10

Acrylic – Vamac® 8 >30 >20

Hydrin – 20D >10 >10

Butyl – 50B >15 >5

Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

The data in Table 5.7 show that, in most cases, the GC×GC–ToF–MS system has
identified significantly more diagnostic substances than conventional GC–MS
and, because the better resolution results in higher-quality mass spectra, a greater
proportion of these will be positively identified by either automatic searching using
a mass spectral database or by manual interpretation.

To sum up the results presented above, the GC×GC–ToF–MS technique offers two
principal advantages over conventional GC–MS that can make it potentially more
effective for reverse engineering-type work. These advantages are:

1. A greater number of diagnostic compounds from sources such as the cure and
antidegradant system can be identified.

2. The greater sensitivity of the technique, coupled with its greater resolving power
and the deconvolution software offers a greater chance that small amounts of
original additives (e.g., accelerators) can be detected.

With regard to comparing the ability of the two techniques to positively identify all
the low-MW compounds (i.e., including the diagnostic compounds mentioned above)
in the five rubber extracts, Table 5.8 gives the total number of compounds detected
by each of the techniques and, of these, how many have been positively identified
from their mass spectra.

164
 Reverse Engineering and Product Deformulation

Table 5.8 Comparison of the total number of substances detected and the
number positively identified for GC×GC–ToF–MS and conventional GC–MS
Compound GC×GC–ToF–MS Conventional GC–MS
NBR – 349N 102 (49) 66 (34)
Fluorocarbon – 49V 184 (66) 36 (14)
Acrylic – Vamac 8
®
217 (138) 55 (26)
Hydrin – 20D 46 (23) 37 (17)
Butyl – 50B 11 (70) 27 (12)
( ): Number of substances positively identified using the mass spectral database
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells in Proceedings of
the High Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2. ©2005, Rapra Technology Ltd [6]

Care has to be taken in compiling the type of data shown in Table 5.8 because a
significant degree of subjective interpretation can be required to carry out the matching
process. However, it can be seen that the GC×GC–ToF–MS technique has produced a
more extensive list of identified compounds and so, in addition to its value in reverse
engineering work, offers greater advantages in the following type of studies:

• Those that involve profiling rubber compounds to assess the low-MW compounds
that have the potential to migrate into food and pharmaceutical products
(Chapters 8 and 9).

• Analysis of food and drug products which have contacted rubber components to
determine if any species have migrated into them (Chapters 8 and 9).

The data in Table 5.8 also show that the proportion of compounds for which no
assignment, or only a tentative assignment, can be found is approximately the same
for the two techniques.

Finally, any improvement in the ability to detect and identify low-MW components in
rubber extracts is welcomed by rubber analysts, and sophisticated chromatographic
techniques, such as GC×GC–ToF–MS, have the potential capability to demonstrate
positive technical advantages over conventional GC–MS systems.

165
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

5.3 Case Study – Deformulation of a Rubber Sample

5.3.1 Introduction

In order to illustrate how the various analytical techniques and methodologies that
have been presented and discussed in Chapters 2–4 can be combined to deformulate
a rubber compound, a reasonably complex example, designated ‘unknown’, will be
worked through in this chapter. It should be remembered that there is considerable
latitude available to the analyst in the approach that he/she can take in this type of
work and that one single, ‘correct’ way does not exist.

The example chosen is a highly flame-retardant rubber compound used to produce a

foam-insulating product for use in the cable industry. The overall objective that was
set at the outset of this analysis work was to obtain information on the following:

• Polymer content and type(s)

• Flame-retardant content and type(s)

• Filler content and type(s)

• CB content and identification of cure system and antidegradants

• Plasticiser content and type(s)

The methodology and approach used to reverse engineer this ‘unknown’ sample
are described in Sections 5.3.2 and 5.3.4 and the results obtained are presented in
Sections 5.3.3, 5.3.5 and 5.3.6.

This case study shows that the work programme was broken down into two stages,
with an initial suite of tests being performed on the unknown sample to obtain as
much compositional information as possible in a time and cost-effective way. Once
the results from this initial programme had been evaluated, additional tests were
performed to fill in gaps in the information or clarify points that had been raised.

5.3.2 Initial Suite of Tests

5.3.2.1 Quantitative Solvent Extraction

One part of the sample was milled on a two-roll laboratory mill and then extracted

166
 Reverse Engineering and Product Deformulation

with methanol for 16 h. At the end of the extraction period the extract was dried at
105 °C for 30 min and then weighed. This procedure was carried out in duplicate
and both the extract and extracted portion retained for further work.

5.3.2.2 Ash Content

The ash content of the sample was determined using the method described in
International Organization for Standardizations’ ISO 247:2006. This procedure was
carried out in duplicate.

5.3.2.3 Thermogravimetric Analysis

One part of the sample (≈10 mg) was analysed by TGA using a specific programme.
Initially, the sample was heated from ambient temperature to 550 °C in a nitrogen
atmosphere. When a constant weight had been obtained, the temperature was reduced
to 300 °C, the atmosphere changed to air and then the temperature raised to 950 °C.
Weight loss and derivative weight loss curves were plotted throughout. The heating
rate at each stage was 20 °C/min and the flow rate of nitrogen and air 50 ml/min.

5.3.2.4 Fourier–Transform Infrared Spectroscopy

The following analytical procedures were carried out using FTIR:

• A transmission infrared (IR) spectrum was recorded of the sample extract prepared
in Section 5.3.2.1.

• A pyrolysis IR spectrum was recorded of the methanol-extracted portion of the

sample prepared in Section 5.3.2.1.

• A transmission IR spectrum was recorded of a liquid paraffin mull of the ash

prepared in Section 5.3.2.2.

5.3.2.5 X-Ray Fluorescence Spectroscopy

A semi-quantitative elemental scan was carried out by XRF on the following:

• Sample in the ‘as-received’ state

• Methanol-extracted portion of the sample

167
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Sample ash

5.3.2.6 Gas Chromatography–Mass Spectrometry

One part of the sample was qualitatively extracted with chloroform for 16 h and
then the concentrated extract analysed by GC–MS under the following conditions:

• Injection: 1 µl at a split of 30:1.

• Injection temperature: 320 °C.

• GC column: SGE BPX5 (25 m × 0.32 mm with a film of 0.2 µm).

• GC column temperature: 40 °C for 2 mins, followed by 20 °C/min to 320 °C.

• Mass spectrometer: scanning 25 to 400 Da every 2 s.

A GC–MS experiment was also conducted whereby the sample was heated to 180 °C
for 10 min in a tube furnace and the volatile species trapped onto an adsorbent packed
with Tenax™. The latter was then transferred to an automated thermal desorption
unit and desorbed for 10 min at 300 °C. The desorbed species were trapped in a cold
trap set at −30 °C and, once the desorption was complete, injected into the GC–MS
by rapid heating of the trap to 300 °C. The compounds injected into the GC–MS
were then analysed under the same conditions as those shown above.

5.3.3 Results from the Initial Suite of Tests

The methanol-extract value of the sample was 34.73% and the FTIR spectrum of
the extract was a composite, with evidence for the presence of following two flame
retardants:

• Pentabromodiphenylether

• A phosphate ester

The pyrolysate FTIR spectrum for the methanol-extracted portion of the sample
(Figure 5.5) gave data that were consistent for a blend of the following two polymers:

• NBR

• Polyvinyl chloride (PVC)

168
 Reverse Engineering and Product Deformulation

98
90
82
74
Transmittance (%)

66
58
50
42
34
26
18
10
2
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm )
-1

Figure 5.5 FTIR spectrum of the methanol extract of an unknown sample.

Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers,
Compounds and Products, Rapra Review Report No.139, Smithers Rapra,
Shawbury, UK, 2001, p.34. ©2001, Smithers Rapra [7]

The ash content for the sample was found to be 18.28% and the FTIR spectrum of
the ash was a composite with evidence for the presence of the following:

• Aluminium oxide (from the flame-retardant hydrated alumina)

• Metallic oxides

The semi-quantitative XRF data obtained on the three portions of the unknown
sample are shown in Table 5.9.

The elemental data in Table 5.9 confirmed the qualitative FTIR assignments and in
addition, showed that two metallic oxides were present in the sample:

• Antimony trioxide (a flame retardant)

• ZnO (part of the cure system for the NBR)

The TGA trace of the sample was unusually complex for a rubber product due to the
multiple weight loss events of the PVC (dehydrochlorination, main chain breakdown
and additional carbonaceous residue – the NBR also yields some) and the loss of
water of the hydrated alumina. These events meant that it was not possible to obtain
accurate quantifications of plasticiser, polymer, CB and inorganics and so it was only
possible to obtain the data shown in Table 5.10.

169
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 5.9 Semi-quantitative XRF data on the three portions of the unknown
sample
Element Sample in the ‘as Methanol extracted portion of the Sample ash
received’ state (%) sample (%) (%)
Silicon 0.25 0.10 0.10
Titanium <0.01 <0.01 <0.01
Aluminium 23.5 30.0 77.0
Iron <0.01 <0.01 <0.01
Calcium 0.04 0.06 0.15
Magnesium 0.45 1.10 0.35
Potassium 0.05 <0.01 <0.01
Sodium <0.03 <0.03 0.95
Phosphorus 0.25 <0.02 0.65
Zinc 3.5 2.7 6.1
Chlorine 23.5 23.5 0.2
Bromine 70.5 2.4 5.9
Antimony 6.5 12.0 5.0
Sulfur 2.2 2.5 0.45
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.35.
©2001, Smithers Rapra [8]

Table 5.10 Data obtained from the TGA trace of the unknown sample
Identification Proportion of the unknown sample
(w/w)
Plasticiser, organic flame retardants, pyrolysis fraction 72%
of PVC and NBR, and released water of hydration
Total carbonaceous material (from PVC, NBR and 7%
carbon black pigment)
Inorganic residue (inorganic additives) 21%

170
 Reverse Engineering and Product Deformulation

21
100.0

R1C
543

574

347
446 481 513
43
140 224 427
86 307 330 371
204
100 200 300 400 500 600 SCAN
3:20 6:40 10:00 13:20 16:40 20:00 TIME

Figure 5.6 GC–MS TIC chromatogram of the chloroform extract from the
unknown sample (RIC: reconstructed ion current). Reproduced with permission
from M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,
Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.36.
©2001, Smithers Rapra [9]

The GC–MS TIC chromatogram of the chloroform extract from the unknown sample
is shown in Figure 5.6 and the qualitative assignments for the species peaks present
in this chromatogram were as follows:

• Trichloromethane

• Formamide

• N,N-dimethylformamide

• Cyclohexanethiol

• Phenol

• N,N-dimethylurea

• 3-(Bromomethyl)-heptane

171
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Benzothiazole (BT)

• Urea

• Phthalimide

• 4-(1,1,3,3-Tetramethylbutyl)-phenol

• Aliphatic hydrocarbons

• Aliphatic carboxylic acids

• Di-(2-ethylhexyl)phthalate

• 2-Ethylhexyl diphenyl phosphate

The GC–MS TIC chromatogram of the volatile species from the unknown sample that
were trapped onto Tenax™ is shown in Figure 5.7 and the qualitative assignments
for the species peaks present in this chromatogram were as follows:

• Carbon disulfide

• Benzene

• Toluene

• Aliphatic hydrocarbons

• N,N-dimethylformamide

• Cyclohexanone

• Cyclohexanethiol

• Phenol

• Acetophenone

• 3-(Bromomethyl)-heptane

• BT

• 2-Methylpropyl butanoate

• 4-(1,1,3,3-Tetramethylbutyl)-phenol

• N-phenylbenzeneamine

• Phthalimide

172
 Reverse Engineering and Product Deformulation

100.0 252

419
466
R10
227
373
39

137
527

315 552
72 272
190

100 200 300 400 500 600 SCAN

3:25 6:50 10:15 13:40 17:05 20:30 TIME

Figure 5.7 GC–MS TIC chromatogram of the volatile species from the unknown
sample that was trapped onto Tenax™. Reproduced with permission from
M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,
Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.36.
©2001, Smithers Rapra [9]

The two sets of GC–MS data confirmed the presence of a phosphate in the unknown
sample, and identified it as 2-ethylhexyl diphenyl phosphate, and identified the
presence of a phthalate, namely di-ethylhexylphthalate. In addition, the identification
of urea in the chloroform extract could mean that a flame retardant such as guanyl
urea was present in the unknown sample.

The GC–MS data also provided information on the cure and antidegradant systems,
and these are shown in Table 5.11.

173
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 5.11 Cure system and antidegradant system information obtained on

the unknown sample by GC–MS
Identified species Possible origin
Cyclohexanethiol CBS accelerator
BT MBT/MBTS (also the CBS) accelerators
N,N-dimethylformamide TMTD accelerator
Phthalimide Santoguard PVI
N-phenylbenzene amine Possibly an acetone/diphenyl amine antioxidant
CBS: N-cyclohexyl-2-benzothiazole sulfenamide
MBT: 2-Mercaptobenzothiazole
MBTS: 2,2'-Dithiobis(benzothiazole)
PVI: Prevulcanisation inhibitor
TMTD: Tetramethylthiuram disulfide

5.3.4 Additional Tests

Having determined from the results obtained from the initial suite of tests that the
sample was based on a blend of polymers and contained a number of flame retardants
(both organic and inorganic), a number of tests could be carried out to obtain
quantification data on the constituents identified in Section 5.3.3.

These additional tests were also used to look for substances that may, because of
logical assumptions based on the initial findings, be present in the unknown sample,
but which had not been apparent in the results (e.g., other flame retardants).

5.3.4.1 Quantitative Elemental Determinations

A number of tests were conducted on the unknown sample in order to obtain

quantitative elemental information and these are listed below:

a) The following elements were quantified by ICP in the unknown sample in the as-
received state and the ash from the unknown sample:

• Antimony

• Zinc

174
 Reverse Engineering and Product Deformulation

• Aluminium

• Phosphorus

• Boron

b) The combined chlorine and bromine content of the unknown sample as-received
state and in the methanol-extracted portion were determined using the oxygen flask
combustion technique with titration as the final stage. The method used was that
described in British Standard BS 7164-22.2:1992.

c) The bromine content of the as-received sample was determined using the oxygen
flask combustion technique, but with ion chromatography as the final step to separate
the bromine- and chlorine-containing species.

d) The sulfur content of the methanol extracted portion of the sample was also
determined using the oxygen flask combustion technique.

e) The nitrogen content of the sample in the as-received state was determined using
a PerkinElmer 2400 CHN analyser.

5.3.4.2 Solid-state Nuclear Magnetic Resonance

The methanol-extracted portion of the unknown sample was analysed by cross-

polarised solid-state NMR in order to provide additional data on the polymers present
within it, and to ensure that the FTIR work had not missed anything.

5.3.5 Results from the Additional Tests

The chlorine content of the extracted portion of the unknown sample was 13.73%.
This chlorine will have originated from the PVC in the sample and could be used
to calculate the amount present in the original sample by adjusting for the level of
methanol extract and then using the fact that PVC contains 56.8% chlorine. The
calculation that was performed and the result obtained is shown below:

(100-extract value) × 13.73

=15.8% PVC in the unknown sample (5.1)
56.8

175
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Assuming the acrylonitrile (ACN) content of the NBR in the PVC/NBR blend was
36% (i.e., a medium ACN content NBR), the amount of NBR in the unknown sample
could be calculated using the same approach as that for PVC content. The nitrogen
content of the extracted portion of the sample was 3.39% and the percentage of
nitrogen in 36% NBR was 9.5%:

(100-extract value) × 3.39

= 23.3% NBR in the unknown sample (5.2)
9.5

The ICP quantifications for the other elements are shown in Table 5.12.

Table 5.12 ICP quantification results for other elements in the unknown sample
Element Sample in the ‘as received’ state Sample ash
Antimony 1.38% 2.12%
Zinc 0.41% 2.35%
Boron <0.05% <0.05%
Aluminium 7.03% 37.8%
Phosphorous 0.22% 1.36%
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.38.
©2001, Smithers Rapra [10]

The data in Table 5.12 show that the flame-retardant zinc borate was not present
in the unknown sample. However, it confirmed that the flame retardants hydrated
alumina and antimony trioxide were present, at 20 and 2%, respectively, and that
ZnO was present at 2%.

The bromine content of the as-received sample was 8.63%. This will have originated
from the brominated flame retardant, pentabromodiphenylether, and showed that
the sample contained ≈13% of this compound by weight.

With regard to the NMR work that was carried out, the solid-state NMR spectrum
for the methanol extracted portion of the unknown sample (Figure 5.8) showed
peaks at shift values of 45.6 and 57 ppm, which were due to PVC, and shifts at 33

176
 Reverse Engineering and Product Deformulation

and 130 ppm, which were due to NBR. No other polymers were apparent in the
unknown sample from this data and so it corroborated the assignments made by
pyrolysis FTIR (Section 5.3.2.4).

45.515
182.940

130.302

57.396

33.195

240 220 200 180 160 140 120 100 80 60 40 20 0

PPM

Figure 5.8 Solid-state NMR spectrum for the methanol extracted portion of the
unknown sample. Reproduced with permission from M.J. Forrest in Rubber
Analysis – Polymers, Compounds and Products, Rapra Review Report No.139,
Smithers Rapra, Shawbury, UK, 2001, p.39. ©2001, Smithers Rapra [11]

5.3.6 Summary of the Information Obtained on the Unknown Sample

By combining all of the information that had been obtained on the unknown
sample using the initial suite of tests (Sections 5.3.2 and 5.3.3) and the additional

177
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

tests (Sections 5.3.4 and 5.3.5), it was possible to present an approximation of

its composition as it stood at this stage and this is shown in Table 5.13. For the
reasons that have been discussed during this chapter and in Chapters 3 and 4, the
quantifications in Table 5.7 are only approximate and it is not possible to provide
quantifications for a number of the compound ingredients (e.g., accelerators).

Table 5.13 Summary of the compositional data obtained on the ‘unknown’

rubber sample
Polymers present NBR and PVC in a blend having the approximate ratio: NBR
55:PVC 45
Total polymer content: 39%
Organic flame retardants* Pentabromodiphenyl ether (13%)
2-Ethylhexyl diphenyl phosphate

Plasticiser Di-(2-ethylhexyl)phthalate
Carbon black content 2%
Total inorganic content 24%
Inorganic compounds ZnO (2%) Hydrated alumina (20%) Antimony trioxide (2%)
Antidegradants Diphenylamine/acetone condensation product
Cure system Sulfur-based system
Accelerators CBS TMTD MBT/MBTS
Other cure system species PVI
*In addition to the flame retardants given above, guanyl urea could be in the sample (see GC–
MS data in Section 5.3.3). If present, this type of additive can be present in rubber compounds
at around the 3% level
Reproduced with permission from M.J. Forrest in Rubber Analysis – Polymers, Compounds
and Products, Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.38.
©2001, Smithers Rapra [10]

As with all investigations of this type, the characterisation work on the unknown
sample could continue, and one of the next steps could be to carry out further work
by solvent extraction LC–MS to obtain quantifications for 2-ethylhexyl diphenyl
phosphate and di-(2-ethylhexyl)phthalate. The nature and direction of any additional
work would depend upon the original objectives of the reverse engineering programme.

It is worth reiterating, though, that the complexity of rubber compounds precludes

a complete characterisation by chemical analytical techniques alone. It is always
necessary for the analytical data to be supplemented by input from a rubber

178
 Reverse Engineering and Product Deformulation

technologist in order for a satisfactory copy of the compound to be obtained. In

order to achieve this, it is usually necessary to use the analytical data as a basis for a
number of test mixes to ensure that all the relevant properties (e.g., cure behaviour,
processibility, durability) and, in this case, flame retardancy, are met.

5.4 Published Reverse Engineering Studies

Because of the deficiencies that still exist regarding the amount of information that
can be recovered from a rubber compound by reverse engineering work (Section 5.1),
research is still being undertaken to improve the accuracy that can be obtained
from these investigations. For example, Datta and co-workers [12] described how
they carried out reverse engineering work on four rubber compounds of known
composition in order to investigate the effectiveness of a new analytical methodology.
This methodology, which used a range of analytical techniques and involved
fractional mass transfer from the acetone extract to the TGA data, produced bulk
compositional data that were very consistent with the formulations and were thought
to offer advantages in areas such as the analysis of waste tyre rubbers of unknown
composition.

A recent example of how a range of analytical techniques can be combined in a reverse

engineering study was provided by Harada [13] and, in this study, the author also
covered the analysis of vulcanised rubber to obtain information on crosslink structure
and the dispersion of fillers.

Another study describing an extensive analytical programme carried out to

deformulate rubber compounds was published by Coz and co-workers [14]. In this
work, reverse engineering work was carried out on four rubber compounds that were
used to manufacture the following automotive products:

• Radial passenger tyre tread

• Radiator hose

• Oil pan seal

• Engine gasket

A computer programme was used to reconstruct the formulations from all the
analytical data and these were found to be very similar to the actual recipes.

Datta and co-workers [15] carried out a quantitative analysis on three batches of end-
of-life ground tyre rubber using a non-pyrolytic FTIR method and derivative TGA.

179
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Two batches were based on truck tyres and the other batch was based on passenger
tyres. The primary objective of the work was to establish an analytical method to
characterise the composition of the rubber blend in waste tyres to assist with efforts
to recycle them. A secondary objective was to evaluate the validity of the method’s
independence of the particle size of the ground rubber. The group used attenuated total
reflectance (ATR)–FTIR to analyse the rubber samples. The data generated revealed
the presence of a blend of NR, SBR and polybutadiene (BR) in all three batches by
use of the following characteristic IR absorption bands:

• NR – 1,375 cm−1

• SBR – 699 cm−1

• BR – 738 cm−1

The same group of workers had derived a general IR blend parameter in a previous
study [16] and this was used, in conjunction with the derivative TGA information, to
quantify the three rubbers in the samples. The blend parameter had been calculated
by using the exact height for the ATR–FTIR absorption bands of NR (1,375 cm−1)
and SBR (699 cm−1) obtained by modifying the spectra using a novel numerical
algorithmic method of baseline creation on, and subsequent subtraction from, the
original spectrum.

Wesdemiotis [17] described a multi-dimensional MS method for the analysis of

polymers. The technique interfaces a suitable ionisation technique and mass analysis
with fragmentation by tandem mass spectrometry (MS2) and an orthogonal online
separation method. Examples of techniques that can be used for the separation stage
include LC for liquid-phase work or ion-mobility spectrometry (IMS), in which
separation takes place in the solution state or post-ionisation in the gas phase. The
mass analysis step in the process provides elemental composition information, whereas
the MS2 step exploits differences in the bond stabilities of a polymer to provide
connectivity and sequence information. The LC work can achieve separations using
differences in polarity, end-group functionality, or hydrodynamic volume, whereas
IMS adds selectivity according to macromolecular shape and architecture. Together
these techniques can be combined to determine the information listed below on a
wide range of polymeric materials, including homopolymers, copolymers, polymer
blends and crosslinked structures:

• Constituents

• Structures

• End groups

180
 Reverse Engineering and Product Deformulation

• Sequences

• Architecture

A review of the analytical procedures and techniques that can be used to separate
and analyse the constituents in rubber and plastic formulations has been published
by Bart [18].

Bacher and co-workers [19] presented a multi-technique methodology to reverse

engineer rubber compounds to identify the polymers and ingredients that are present
in order to assist with duplication or enhancement work. The combined approach
that was employed used the following techniques:

• TGA to determine the quantities of polymer, plasticiser and CB.

• FTIR to verify the identity of the polymer, plasticiser and some of the additives.

• ICP–optical emission spectrometry to determine the elementary filler composition.

• Range of ‘wet chemistry’ methods to determine certain additives and fillers.

Nimkar and co-workers [20] used an integrated multi-spectral range Fourier–

Transform (FT) spectrometer to characterise a range of polymeric materials. The
combination that was developed included IR, Raman spectroscopy and TGA
to provide a versatile and cost-effective tool for laboratories engaged in reverse
engineering work. Some attributes of the workstation included:

• Extended range ATR–FTIR that enabled IR spectra to be acquired below 100 cm−1.

• FT–Raman accessory configured with a 1,064-nm laser that enabled spectra to

be recorded from materials that fluoresce with visible excitation.

• TGA–IR that provided an easy way to identify the volatile components.

The research team hopes that their results will demonstrate the complementary
nature of the various analytical techniques towards providing a complete solution
to a challenging problem.

Bhatt and co-workers [21] carried out work to establish a correlation between the
FTIR data and TGA data obtained on the composition of rubber blends used in tyre
formulations. Determining this correlation then enabled the blend ratio of rubbers
such as NR, SBR and BR in passenger tyres and off-the-road tyres to be calculated,
which could be used to assist with important industrial activities, such as quality
assurance work.

181
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Nawale and co-workers [22] explored the feasibility of using a mathematical

enhancement, such as the second-order derivative of ATR–FTIR spectra, for carrying
out compositional analysis work on NR, polychloroprene and their blends in CB-
filled vulcanisates. A germanium crystal was used to record the ATR spectra and the
second-order derivative approach was found to be successful for the qualitative and
quantitative analysis of the two rubbers and their blends. The research team also
considered that this method and approach was applicable to other rubber blends,
including those of NR/SBR and NR/BR.

Work carried out at PerkinElmer showed how the AutoStepwise TGA method could
be used for the compositional analysis of tyre samples [23]. This technique is a
subset of the controlled rate thermal analysis technique developed independently by
Rouquerol [24] and Paulik and Paulik [25]. They used the following conditions and
TGA programme for the analyses:

• Sample mass: 13 mg (approximate weight).

• Instrument: Pyris 1 TGA.

• Method: AutoStepwise.

• Heating rate: 20 °C/min from 30 to 900 °C.

• AutoStepwise entrance threshold: 4%/min.

• AutoStepwise exit threshold: 0.5%/min.

• Pan open: aluminium pan in a ceramic holder.

• Purge gas: nitrogen to 700 °C and air to 900 °C (both 30 ml/min).

The workers regarded the results as showing excellent resolution between all the
weight loss events, including the difficult oil and polymer steps, and this enabled the
following compositional data to be recorded on the tyre sample:

• Oil: 21.8 %

• Polymer: 43.9%

• CB: 32.2%

• Inert filler: 2.11%

The AutoStepwise approach was regarded by the PerkingElmer workers as having

an advantage over the variable heating rate TGA method (i.e., the high-resolution
TGA method). This advantage was due to it handling diffusion-controlled weight loss

182
 Reverse Engineering and Product Deformulation

steps much better because the high-resolution method tends to move through the oil
weight loss step much quicker and so may not provide the full evolution of oil before
the polymer degradation step is encountered. Due to its speed and applicability for a
range of different oil types, the AutoStepwise method was also thought to be a better
approach, as would high-resolution TGA, for quantifying the oil-in-rubber samples
than the following methodologies covered by Maurer [26]:

• Isothermal analysis at a temperature significantly below the temperature at which

polymer degradation occurs.

• Use of a reduced pressure (or vacuum) to aid volatilisation of the oil.

• Quantitative extraction of the rubber with a solvent to remove the oil prior to
TGA analysis.

References

1. S.D. Gupta, R. Mukhopadhyay, K.C. Baranwal and A.K. Bhowmick in

Reverse Engineering of Rubber Products, CRC Press, Taylor and Francis
Group, Boca Raton, FL, USA, 2013.

2. J. Scheirs in Compositional and Failure Analysis of Polymers: A Practical

Approach, John Wiley & Sons, New York, NY, USA, 2000.

3. The Complete Book on Rubber Chemicals, NPCS Board of Consultants &

Engineers, ASIA Pacific Business Press, Inc., Kamla Nagar, Delhi, India, 2009.

4. The Blue Book – Materials, Compounding Ingredients and Services for the
Rubber Industry, Rubber World Magazine, Akron, OH, USA, 2016.

5. M.J. Forrest, S. Holding and D. Howells, Polymer Testing, 2006, 25, 63.

6. M.J. Forrest, S. Holding and D. Howells in Proceedings of the High

Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2.

7. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.34.

8. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.35.

183
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

9. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.36.

10. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.38.

11. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK, 2001, p.39.

12. S. Datta, R. Stocek, Ivo Kuritka and P. Saha, Polymer Engineering and
Science, 2015, 55, 6, 1450.

13. M. Harada, International Polymer Science and Technology, 2016, 43, 2,

T/45.

14. D. Coz. K. Baranwal and T.M Knowles in Proceedings of the ACS Rubber
Division Fall Meeting, 8–11th October, Louisville, KY, USA,
American Chemical Society Rubber Division, Washington, DC, USA, 1996,
Paper No.68.

15. S. Datta, J. Antos and R. Stocek, Polymer Testing, 2017, 59, 308.

16. S. Datta, J. Antos and R. Stocek, Polymer Testing, 2017, 57, 192.

17. C. Wesdemiotis, Angewandte Chemie, International Edition, 2017, 56, 6,

1452.

18. J.C.J. Bart in Proceedings of the 60th ANTEC Technical Conference,

5–9th May, San Francisco, CL, USA, Society of Plastics Engineers, Brookfield,
CT, USA, 2002, Paper 012.

19. B. Bacher, M. Walker and A. Riga, Rubber World, 2013, 247, 4, 22.

20. A. Nimkar, T. Strother and S. Lowry in Proceedings of the ANTEC Technical

Conference, Mumbai, India, 6–7th December, Society of Plastics Engineers,
Brookfield, CT, USA, 2012, p.467.

21. J. Bhatt, B.K. Roy, A.K. Chandra, S.K. Mustafi and P.K. Mohame, Rubber
India, 2003, 55, 9, 7.

22. M. Nawale, J. Jarugala, S. Sahoo, K. Rajkumar, B. Abbavaram and R. Sadiku,

R. Pritti and D.J. Maurya, Polymer Testing, 2017, 22, 447.

184
 Reverse Engineering and Product Deformulation

23. Compositional Analysis of Tire Elastomers using AutoStepwise TGA,

PerkinElmer, Waltham, MA, USA, 2009.

24. J. Rocquerol, Bulletin of the Society de Chemie France, 1964, p.31.

25. F. Paulik and J. Paulik, Analytica Chemica Acta, 1971, 56, 2, 328.

26. J.J. Maurer in Thermal Characterisation of Polymeric Materials, Ed., E. Turi,

Academic Press, New York, NY, USA, 1981, p.653.

185
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

186
6
Curing and Cure State Studies

6.1 Introduction

This chapter provides an overview of the various chemical and thermal techniques
that have been used to characterise the curing behaviour and cure state of rubber
compounds and rubber products. It is an essential element of quality control (QC)
work to routinely monitor and assess these characteristics, but the need also arises for
a number of other reasons. For example, during research and development (R&D)
activities, and it often plays a crucial part in failure diagnosis investigations.

There are standard pieces of equipment within a rubber laboratory that are used to
measure and characterise curing behaviour (e.g., rate of cure at a given temperature)
and cure state (i.e., degree of cure) and these include the oscillating disc rheometer
(ODR), the moving die rheometer (MDR), and the rubber process analyser (RPA).
These are excellent QC and R&D instruments, but they have a practical constraint
concerning the minimum amount of material that is required for an examination, and
they also tend to be used on rubber samples that start off in the unvulcanised state.
The information that can be obtained by these instruments can be complimented by
the use of thermal analysis techniques [e.g., differential scanning calorimetry (DSC)
and dynamic mechanical analysis (DMA)] and, in the case of liquid samples, specialist
equipment [e.g., the RAPRA Vibrating Needle Curemeter (VNC)]. Thermal analysis
techniques offer a number of advantages, including being able to record data on small
samples (e.g., 5 mg in the case of DSC) and being able to analyse cured, partially
cured, and uncured samples.

It is also possible to use physical testing techniques to establish the state of cure
(i.e., crosslink density) of a rubber sample. Two classic approaches are equilibrium
swelling in a solvent and extensional experiments involving a tensometer. The former
method has the advantage in that it can be carried out on a sample having an irregular
geometry; the latter requires standard test sheets and test pieces (e.g., dumbbells). The
use of these techniques for this type of work has been discussed at length in other
publications [1] and, although they can often contribute to analytical investigations,
they will not be covered in detail in this chapter.

187
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

When a rubber cures, a wide range of volatile organic compounds are released as
rubber vulcanisation fume. The composition of this fume has been studied (Chapter 7)
to assess its affect on the health of employees in the rubber industry and general
environment. This research work has established relationships between the fume
and the compounding ingredients in a rubber compound. This has given the analysis
of rubber volatiles a role in the determination of additives (Chapter 4) and reverse
engineering work (Chapter 5).

Complete cure

Curing

Before cure

Figure 6.1 Typical ODR curve for a rubber compound. Reproduced with
permission from A. Ciesielaki in An Introduction to Rubber Technology,
Rapra Technology Ltd, Shawbury, UK, 1999, p.85. ©1999, Rapra Technology Ltd
[2]

Two illustrations of an ODR curve that has been recorded on a rubber compound
are shown in Figures 6.1 and 6.2. Both versions of this curve have been obtained by
plotting the change in torque as a function of time and demonstrate how the torque
increases as the rubber cures and a three-dimensional crosslink network develops
within the material. The optimum cure time for a particular rubber compound is
usually regarded as the time to achieve 90% maximum torque at a given temperature
and is designated ‘t90’. The version of the ODR trace shown in Figure 6.2 provides
examples of the other properties that can be obtained from such a trace, such as the
scorch time, minimum viscosity value (i.e., minimum torque value) and maximum
torque value. Figures 6.1 and 6.2 show the torque value attaining a plateau once the
rubber has become fully cured but, in some cases [e.g., with natural rubber (NR)
compounds], the rubber undergoes ‘reversion’, and the value of the torque begins to
decrease as the matrix undergoes thermal degradation.

188
 Curing and Cure State Studies

R&D organisations such as Smithers Rapra have been very active in studying the
cure behaviour of rubber compounds and other polymeric systems. A comprehensive
review of the use of chemical and physical analytical techniques to assess cure has
been published by Willoughby [3]. Additional information on the techniques and
methodologies that can be used to study curing can be found in a number of other
publications, such as those edited by Gabbott [4] and De and co-workers [5].
Torque, dN.m

Scorch time, ts2

Maximum torque, MH

ML Minimum viscosity

Time, minutes

Figure 6.2 The trace in Figure 6.1 with important features highlighted. Reproduced
with permission from A. Ciesielski in An Introduction to Rubber Technology,
Rapra Technology Ltd, Shawbury, UK, 1999, p.85. ©1999, Rapra Technology Ltd
[2]

6.2 Thermal Analytical Techniques for Curing and Cure State

Studies

6.2.1 Differential Scanning Calorimetry

One analytical technique that is particularly useful for this type of work is DSC. A

189
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

review of the use of DSC to study sulfur and peroxide vulcanisation curing processes
in rubbers has been provided by Brazier [6].

DSC is a very effective tool for this application because the curing reaction of a
rubber, in common with many other types of chemical reactions, is very exothermic.
This exotherm, which results from a number of reactions that can take place during
vulcanisation, is easily detected by a significant change in the specific heat plot of a
sample.

A number of different methods for using DSC to characterise the cure of rubbers
have been developed [7]:

• Borchardt and Daniels method [8].

• American Society for Testing and Materials’ (ASTM) E698 method [9, 10].

• Isothermal method [11, 12].

The size and position (i.e., temperature range) of the curing exotherm obtained in a
DSC analysis is dependent upon the:

• Type of polymer.

• Proportion of polymer in the sample.

• Type of cure system (e.g., peroxide or sulfur).

• Composition of the cure system (e.g., type of peroxide or type of accelerator(s)

if the system is sulfur-based).

• Overall amount of curative(s).

• Level of individual curatives in the blend if a blend of curatives is present.

In addition to the factors stated above, which are sample-related, it has also
been demonstrated using inter-laboratory trials that experimental conditions and
instrumental effects can influence the cure data that are obtained [13, 14].

An example of a curing exotherm (from ≈90 to ≈240 °C) is shown in Figure 6.3. This
is a DSC thermogram of a fluorocarbon rubber undergoing vulcanisation due to the
presence of a peroxide-based cure system. It is apparent from this thermogram that
values for the cure onset temperature and total area of the exotherm can be obtained
which, by use of the sample weight, can be converted into J/g. The size of the sample
is this case was 12.9 mg and the value for the curing exotherm was calculated to be
110 J/g.

190
 Curing and Cure State Studies

In Figure 6.3, the exotherm is given a negative value but this is only due to the style
of presentation that has been used – with exotherms being plotted in this case in
a negative direction and endotherms in a positive one. The endotherm that occurs
after the curing exotherm, with a peak at ≈270 °C, is due to the curing volatiles lost
from the sample. This analysis was carried out using a sealed sample pan to try and
contain these volatiles, but a large quantity were generated and the resulting pressure
ruptured it. This situation could be addressed by reducing the size of the sample,
but it can also occur if there is a relatively high proportion of additives in the rubber
compound (e.g., plasticisers or oils) that are particularly volatile.

Sample ID: 41V BT1

Operator ID: cdp
Comment: CCA7 @ 0 °C, 40 µL pans
Project no. J0316
Sample weight: 12.910 mg
Data collected: 22/08/2003 01:39:20

20

18

16

14 Onset = 171.26 °C
Heat flow endo up (mW)

12

10
Area = -1505.381 mJ
8 ∆H = -116.6058 J/g

2
Peak = 211.70 °C
0
60 80 100 120 140 160 180 200 220 240 260 280 300
Temperature (°C)
(1) Hold for 1.0 min at 40.00 °C (4) Cool from 500.00 °C to 40.00 °C at 20.00 °C/min
(2) Heat from 40.00 °C to 500.00 °C at 20.00 °C/min (2) Hold for 3.0 min at 400.00 °C
(3) Hold for 1.0 min at 500.00 °C (3) Heat from 40.00 °C to 40.00 °C at 20.00 °C/min

Figure 6.3 DSC trace showing the curing exotherm of a peroxide-cured

fluorocarbon rubber. Reproduced with permission from M.J. Forrest in Principles
and Applications of Thermal Analysis, Ed., P. Gabbot, Blackwell Publishing,
Oxford, UK, 2008, p.242. ©2008, Blackwell Publishing [15]

191
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Two advantages of using DSC for curing investigations are that only small sample
sizes are required (e.g., 5 mg) and no specific sample geometry is necessary. This type
of investigation can be carried out for a number of purposes, for example:

• Quantification of cure system species

• Qualitative data on cure system type

• Characterisation of a cure (e.g., initiation temperature and kinetics)

As listed above, a number of factors, including a number of other components in a

rubber compound and experimental conditions, will influence the DSC data and so
this type of work, and other DSC applications, are usually carried out in a comparative
way. For example, to use DSC to determine the amount of cure system in a rubber,
because the exotherm obtained is very specific to the rubber matrix and the cure
system used, standards need to be available if quantitative data are to be obtained on
a test sample from the J/g exotherm value. Once these standards have been obtained
and the reference data generated, this type of application obviously has uses from a
QC point-of-view as well as other applications, such as failure diagnosis.

If an inconsistent cure state is thought to exist through the cross-section of a relatively

large rubber product, by taking and analysing small samples by DSC it is possible to
obtain a depth profile of this property. The ability of DSC to detect residual unreacted
curatives can be used in a more general sense to investigate if a product has been cured
appropriately as part of a QC process. In these examples, the exotherm generated
can be relatively small and can occur in a similar region in the thermogram to where
additives, such as plasticisers and oils, are volatilising and creating an endotherm.
For these reasons, such exotherms can be very difficult to detect and it is, therefore,
important to modify the experimental conditions to optimise the system in favour
of their detection. For example, larger samples should be used in sealed sample pans
to stop substances volatilising, although care has to be taken because rupturing of
the sample pan can occur.

The examples given above illustrate how DSC can be used for practical purposes
that benefit the day-to-day activities of industry and test houses. It can also be used
in more fundamental R&D work to assist researchers who are interested in cure
kinetics, as well as other important properties, such as optimising the degree of cure
and the cure state that exists within different phases of polymer blends.

Information on cure kinetics can be of great value to R&D chemists trying to

development new curing agents or curing systems. The achievement of an optimised
state of cure within a rubber product is vital to it realising its full potential with respect
to its properties and performance in service. Using DSC for this type of quantitative

192
 Curing and Cure State Studies

work, and others, can be challenging and a number of researchers and scientists have
discussed its conduct and the possible problems that can be encountered [13, 14, 16].

One example of the curing kinetics studies that have been carried out is the one
reported by Simon and Kucma [17]. They vulcanised rubber compounds using both
isothermal and non-isothermal (i.e., temperature ramp) conditions and, among other
features, described the temperature dependence of the cure induction period using
an Arrhenius-type equation. Another example is the work carried out by Ou and co-
workers [18], who used DSC and ODR to study the crosslinking reaction of silicone
rubber. From the experimental data that were obtained, different kinetic analytical
methods were used to determine the activation energy of silicone rubbers, such as
the Kissinger, Ozawa, Friedman, Flynn–Wall–Ozawa, Kissinger–Akahira–Sunose and
integral methods. The results obtained by their study showed how the composition
of the silicone rubber, the analytical technique and kinetic analytical method affected
the activation energy and its evolution during the crosslinking reaction.

Even though sulfur vulcanisation has been used by the rubber industry for over well
over 100 years, it has often been refined and advanced in a mainly empirical way.
Over the years, various studies have been conducted to advance the fundamental
understanding of how elemental sulfur, the various classes of accelerators (e.g.,
thiurams) and different co-agents [e.g., zinc oxide (ZnO)] that can be used to vulcanise
rubber react with one another and the rubbers that they are compounded into.
One major contribution to this body of work is the extensive series of experiments
conducted by researchers at Port Elizabeth University, who used DSC extensively and
its contribution features in a number of publications [19, 20].

DSC can be used to study the cure state within different phases of polymer blends.
For example, Wong-on and Wootthikanokkan [21] used DSC to determine the degree
of cure that existed within the rubber phase of a plastic–rubber blend. To carry out
the work, the two researchers prepared PVC/acrylic rubber blends using a twin-screw
extruder and then cured samples for testing using compression moulding. The curing
characteristics of blends consisting of two rubbers have also been investigated by DSC.
For example, Baba and co-workers [22] used DSC to evaluate the crosslink density
of rubber blends. Their work involved ethylene propylene diene monomer (EPDM)
rubber and polybutadiene (BR) rubbers that had been crosslinked to varying degrees
by either photooxidation or dicumyl peroxide.

To understand the influence that additives have on curing chemistry, Karbhari and
Kabalnova [23] used DSC to investigate the curing kinetics of rubber-modified, carbon
fibre (CF)-reinforced vinyl ester resins. Their work reported on how the quantity of
CF, and the amount of sizing present on its surface, affected the cure rate, degree of
cure, time to achieve the maximum cure state, and cure rate constant.

193
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

It is well known that the crosslinking of polymer chains increases the glass transition
temperature (Tg) of the polymer [24]. For example, Zeng and Ko [25] reported
that crosslinking cis-1,4-BR to a higher degree increased the Tg of the rubber from
−103 to −96 °C. Cook and co-workers [26] studied the Tg of various rubbers with
different crosslinking densities using both DSC and nuclear magnetic resonance
(NMR), and found that the Tg versus crosslinking data followed a linear regression.
In addition, they found that the NR and BR vulcanisates displayed a similar Tg versus
crosslinking density plot, but that the increase in Tg of the styrene-butadiene rubber
(SBR) vulcanisate with crosslink density was much greater than that of NR or BR.

In addition to being used to study the vulcanisation of solid rubbers, DSC can also
be employed to characterise the cure of rubber latex. Peres and Lopes [27] used DSC
to investigate the efficiency of the vulcanisation of NR latex using two types of cure
system – a conventional sulfur cure system and an efficient [i.e., efficient vulcanisation
(EV)] cure system. The DSC data were complemented with tensile strength (TS) tests
and optical analysis. The results showed that the conventional sulfur cure system
provided a higher curing rate, better processing safety (i.e., longer scorch time) and
superior physical properties.

The cure of rubber-modified epoxy resin systems has also been investigated by
Restrepo-Zapata and co-workers [28]. The curing reaction of an aliphatic epoxy resin/
EPDM rubber system was modelled from DSC data using a methodology proposed
by Hernandez-Ortiz and Osswald. The kinetics of the reaction were represented by a
Kamal–Sourour model, with and without diffusion reaction control, and was extracted
using a non-linear regression method coupled with heat and mass balance equations.

6.2.2 Dynamic Mechanical Analysis

Another technique which can be used to investigate the degree of cure of a rubber
product is DMA. It suffers from two drawbacks when compared with the DSC
technique: a certain sample size and geometry (normally 2 cm × 1 cm × 1 mm) is
usually required and it is often less sensitive, relying on the effect that crosslink density
has on the Tg of the rubber compound (as shown by the tan δ plot) or its modulus.
Although the sample dimension described above is often regarded as the ‘standard’
one, modern instruments have become more accommodating as new sampling devices
have come onto the market. For example, it is now possible to obtain comparative
DMA data on smaller samples and even those with a non-specific geometry (e.g., in
the form of a powder).

Use of the tan δ plot in the DMA thermogram to obtain an indication of cure state
is possible because the reduction in storage (i.e., elastic) modulus as the rubber

194
 Curing and Cure State Studies

passes through its Tg is reduced successively as the crosslink density is increased.

The degree of reduction that is observed is not uniform and is dependent upon the
type of rubber (i.e., polymer) that the compound is based upon. The approximate
percentage reduction in the height of the tan δ peak moving from an uncured state
to the optimum cured state for three types of rubber is shown below [29]:

• Nitrile rubber (NBR) – 15%

• Polyisoprene – 76%

• BR – 64%

Having established the nature of the relationship shown above by the use of standard
compounds, it can then be used to provide an assessment of the degree of cure of a
test sample. Although this task could be carried out using other tests (e.g., hardness
determination, solvent swelling), DMA offers the advantages of a relatively quick
diagnosis on, potentially, a relatively small portion of sample.

With respect to using the change in the modulus to reflect the cure state of a rubber
compound, the effectiveness of this approach can be influenced by the levels of the
principal additives (e.g., fillers and plasticisers) that are present. This is because high
levels of these additives will have a significant influence on the modulus of the rubber
compound and reduce the polymer fraction of the sample; the faction altered by
crosslinking. What is found in practice is that for a highly filled rubber compound,
the difference in modulus between it being fully cured and uncured is less than for
the unfilled version. This illustrates that these types of studies and a lot of others in
this chapter are best carried out in a comparative way on samples that have been
fully characterised.

It is possible in some cases when using the typical temperature ramp (e.g., 3 °C/min)
to observe a rubber sample that is a poor state of cure initially suffering a decrease
in modulus once it has passed its Tg due to softening and then, as the temperature
continues to rise, to gain modulus as the crosslinking reaction takes place. This
effect is more likely to be observed when the poorly cured rubber sample contains
curatives that have relatively low temperatures of initiation (e.g., certain peroxides)
and is lightly modified by fillers.

As shown in Figures 6.1 and 6.2, the ODR and MDR can be used to plot graphs
of torque versus time at a constant temperature to show how a cure is progressing
and, based on the results obtained, the optimum cure time is usually regarded as
the time required for the torque to reach 90% of its maximum achievable value,
and is designated ‘t90’. Khimi and Pickering [30] assessed the use of DMA in
the determination of t90. The DMA experiments were carried out using shear

195
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

mode isothermal tests to measure the changes in material properties caused by

vulcanisation. The results obtained revealed that the shear storage modulus (G'),
shear loss modulus (G'') and tan δ all reflected the vulcanisation process, but the
tan δ plot gave the best representative level of vulcanisation. In fact, the tan δ plot
could be used to obtain a value for t90, and this value was in the good agreement
with the value of t90 obtained using a MDR.

Formela and co-workers [31] used DMA to contribute to a programme of work

designed to characterise ground-tyre rubber samples that had been reclaimed using
a twin-screw extruder and then re-vulcanised using different types of standard
rubber industry vulcanisation accelerators [e.g., 2-mercaptobenzothiazole and
tetramethylthiuram disulfide (TMTD)] formulated into conventional sulfur and EV
cure systems. The results obtained showed that the static and dynamic mechanical
properties of the re-vulcanised rubber depended strongly on which type of cure
system had been used. The highest crosslink densities were obtained with the samples
that had been cured using a TMTD-based EV system, whereas the best processing
and mechanical properties resulted from the use of a N-tert-butyl-2-benzothiazole
sulfenamide (TBBS)/N-cyclohexyl-2-benzothiazole sulfenamide (CBS) accelerated
conventional sulfur cure system.

The dynamic mechanical properties of electron beam-irradiated NBR vulcanisates

containing varying levels of sulfur have been studied by DMA [32]. DMA was
performed from −80 to +80 °C, at frequencies that varied from 0.32 to 32 Hz, and
at strains of between 0.001 and 10%. The results showed that the irradiation of the
samples caused significant changes in the tan δ peak temperature and storage modulus.
The vulcanisates containing high amounts of sulfur formed intense crosslink networks
and crosslink rearrangements, which were supported by an increase in storage modulus
and the shift in tan δ towards higher temperatures compared with the control. The
research team also found that there was an increase in the tan δ peak height due to
chain scission and subsequent plasticisation.

6.2.3 Other Thermal Techniques

It is possible to use thermal mechanical analysis (TMA), either in penetrometry or

expansion mode, to evaluate the state of cure of finished rubber products, and its use
for this purpose has been discussed by a number of authors [33–36]. In the expansion
approach, the sample is heated from ambient to a high temperature (e.g., 200 °C)
and the small dimensional change that occurs is monitored. This dimensional change
increases as the cure state of the sample increases. For example, a degree of cure of
50% (based on a rheometer reading) might give a change of ≈20 µm, which increases
to ≈40 µm for a fully cured sample. Providing that reference data for a particular

196
 Curing and Cure State Studies

system are available, the technique can be used to investigate the state of cure of a
test sample.

The use of TMA via the penetrometry mode for calculating crosslink density was
advanced by Prime [36], who demonstrated a correlation between Young’s modulus
and the solvent swelling ratio of a sample. This correlation then enabled the crosslink
density ‘e’ of a sample to be obtained by, firstly, calculation of its elastic modulus
‘Em’ from the TMA penetrometer measurements using the equation developed by
Gent [37], and then by use of the following equation [38]:

e = Em/3RT (6.1)

where ‘T’ is 239 K.

R (universal gas constant) = 8.314 × 107 ergs/mol/K

Provder and co-workers [39] used diffusing mechanical analysis, thermal analysis, and
DMA to study the cure of a polydimethylsiloxane (PDMS) top coat/epoxy coating
for primer marine applications. The data obtained by this combination of analytical
techniques provided the research team with information on a wide range of properties,
including curing, drying and relaxation behaviour, film-forming parameters, coating
thickness, thermal transitions, stress–strain properties, elastic modulus, substrate
effects and dynamic mechanical properties.

Zhao and co-workers [40] investigated the use of a number of novel compounds for
the high-temperature curing of brominated butyl rubber. In addition to recording
thermogravimetric analysis data on the compounds, their curing behaviour and physical
properties were evaluated and the results showed that N,N'-phenylenedimaleimide,
N,N'-(4,4'-methylenediphenyl)dimaleimide and diammonium phosphate were suitable
compounds for curing brominated butyl rubber at high temperatures.

Maistros and co-workers [41] used dynamic dielectric analysis (DDA) to monitor the
isothermal cure in rubber-modified epoxy systems. Their results showed that in the
frequency range 0.1 to 20 kHz, DDA was capable of showing the onset of gelation
in the modified resins via a sudden increase in the relaxation time. The separation of
phases within the material was detected by an increase in relative permittivity due
to interfacial polarisation. Overall, the results obtained by DDA were found by the
workers to be in good agreement with those obtained by turbidity measurements
and solubility determination.

197
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

6.3 Use of Chromatographic and Spectroscopic Techniques for

Curing and Cure State Studies

Andersson and co-workers [42] used pyrolysis (Py) gas chromatography (GC) to study
the sulfur bridges in filled NR vulcanisates. The gas chromatograph was fitted with a
sulfur-selective flame photometric detector and the main Py products detected were
carbon disulfide and thiophenes. Two NR formulations were analysed, one cured
with sulfur and CBS, and the other with the sulfur donor compound TMTD. The
yields of the Py products were found to be different for the two rubber compounds
and to vary with the cure time in both cases.

It is possible to use spectroscopic techniques to study cure state and crosslinked

structures within rubbers as demonstrated by the work reported by Ito and Sawanobori
[43]. The two researchers used NMR to determine the crosslink density of both
solid and foamed sulfur-cured EPDM compounds. With the solid EPDM samples,
the results showed a good correlation between the spin-relaxation time (T2) at high
temperatures (e.g., 120 °C) and the crosslink density obtained by equilibrium swelling
and extensional experiments. They also found that the T2 of the EPDM foams was not
affected by their cellular structure and, hence, it was possible to use NMR to obtain
the crosslink density of these products, something that is not possible by equilibrium
swelling or extensional experiments.

Brown and co-workers [44] demonstrated how Fourier–Transform NMR can be

used to estimate the crosslink densities of unfilled and carbon black (CB)-filled
rubber blends. The different rubbers that were included in the study included NR,
BR, NBR and EPDM. Using the Fourier–Transform approach was shown to improve
the resolution of NMR data and enabled work to be performed on filled compounds
that contained filler additives at realistic, commercial levels.

Pazur and Walker [45] discussed the issues surrounding the use of NMR spin echo
measurements for assessing the state of cure of crosslinked rubber. They suggested
that a technique known as the ‘state of cure number’ could be used for the analysis
of the data and that this overcame the problems (e.g., variable test results) found
with spin echo measurements. The state of cure number correlated to the crosslink
density in a rubber sample and the published work showed that it could be applied
to the measurement of the state of cure of specialty rubbers for automotive sealing
applications and optimisation of the injection moulding curing conditions of a
developmental EPDM part.

A research team at Shandong University [46] used in situ Fourier–Transform infrared

(FTIR) spectroscopy to study the curing kinetics of a two-part, room temperature-
cured, silicone rubber. The two-part rubber, which had excellent low-temperature

198
 Curing and Cure State Studies

properties, was prepared by mixing a random silicone copolymer (PDMS-co-

diethylsiloxane), synthesised using anionic ring-opening polymerisation, with a
hydrosilicone. The cure kinetics that resulted from the in situ FTIR analysis were used
to forecast the tack-free time of the silicone rubber at different curing temperatures,
and these were found to agree with the measured tack-free time of test samples.

6.4 Rheometric Studies

The data shown in Figure 6.4 illustrate how rheometer curves (e.g., ODR curves) can
change as the formulation of a rubber compound is varied, and how the information
obtained (e.g., scorch time, cure time and rate of cure) can be compared directly.
The curves shown in Figure 6.4 are for a series of NR compounds that contain a
sulfur cure system accelerated using five types of sulfenamide accelerator, each at a
concentration of 0.6 phr. The other formulation details are:

• NR: 100

• ZnO: 5.0

• Stearic acid: 2.0

• Sulfur: 2.5

Mention has been made in Section 6.1 of the use of RPAs as a QC tool in the rubber
process laboratory. ASTM D6204 Part C describes a test method for studying
cure kinetics and conducting variable temperature analysis (VTA) using the RPA
manufactured by Alpha Technologies. Two articles by Dick and Norton [48, 49]
provide reviews of how reaction cure kinetics data and VTA data, generated using
the Alpha Technologies RPA, can be used to investigate rubber curing problems
and improve QC in manufacturing operations. To illustrate this situation, several
laboratory designs of experimental studies were carried out on selected rubber
compounds to compare the effectiveness of the cure kinetic and VTA approaches.
These model compounds were based on sulfur-cured SBR and NR compounds
and peroxide/promoter-cured EPDM, chlorinated polyethylene, polychloroprene
and fluorinated compounds. In addition, in an article written in 2015, Dick and
Xue [50] provided an overview of the improvement of the capabilities of the
RPA made by Alpha Technologies to undertake a range of analyses, including
the analysis of scorch and curing profiles, and the prediction of the performance
of cured rubber products by measuring their dynamic properties. The report also
describes how improvements in the range and precision of strain and frequency
sweeps resulted in improvements in the predictions that can be made using the
RPA. The examples provided included the better prediction of the state-of-the-mix

199
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

through ASTM D6204 Part B (high strain) and better predictions of the degree
of silanisation through improved, wider-range strain sweeps for the Payne effect.

Compound recipe phr

CBS TBBS
MBS NR 100.0
0.75 Zinc oxide 5.0
Torque, Nm

Stearic acid 2.0

Sulphenamide 0.6
0.50
Sulfur 2.5
TBSI

0.25 DCBS

0.00 15.00 30.00 45.00

Time, min

Figure 6.4 Overlaid rheometer curves showing how varying the type
of sulfenamide accelerator within a NR compound can alter its curing
characteristics (DCBS: N,N-dicyclohexyl-2-benzothiazyl sulfenamide; MBS:
2-(4-Morpholinothio)benzothiazole and TBSI: N-tert-butyl-2-benzothiazole
sulfenimide). Reproduced with permission from R.N. Datta and F.A.A. Ingham
in Rubber Technologist’s Handbook, Eds., J.R. White and S.K. De, Rapra
Technology Ltd, Shawbury, UK, 2001, p.175. ©2001, Rapra Technology Ltd [47]

Rosa and Vergnaud [51] developed a new method for using an MDR. By increasing the
temperature of the instrument with the square root of time, the kinetic parameters of
the curing reaction were determined. These included the activation energy of the cure,
the pre-exponential factor, and the order of the overall reaction. Other information
that was obtained using this method included the determination of the temperature
profiles developed through the thickness of the rubber sample and profiles of the
state of cure.

Torque curves were used by Marzocca and Mansilla [52] to characterise the
vulcanisation of SBR compounds prepared using different cure systems based on

200
 Curing and Cure State Studies

sulfur and TBBS. The time to achieve maximum torque was evaluated and the density
and type of elastically active crosslinks were examined by swelling measurements.

Madhuri Nanda and Deba Kumar Tripathy [53] carried out a comparative study
involving four cure systems for chlorosulfonated polyethylene rubber (CSM). The
systems investigated were based on sulfur, peroxide (dicumyl peroxide), metal oxide
(lead oxide) and epoxy resin, and their curing characteristics were determined using a
cure rheometer. The highest values of maximum rheometric torque and scorch safety
were found with the peroxide cure system, whereas the sulfur system gave superior
mechanical and rheological properties. Analyses of the cured CSM rubbers by FTIR
using the attenuated total reflectance technique enabled the researchers to propose
possible crosslink structures for the different types of cure system.

A research team in China carried out a rheological study on the cure kinetics of a
two-component addition-cured silicone rubber [54] and monitored the effect of the
reaction temperature. Their research showed that the Kamal–Sourour (autocatalyst)
model fitted the cure kinetics of the two-part silicone rubber better than the Kissinger
model. A second-order reaction was derived for the time period up the point in the
reaction where the crosslinked network started to restrict the diffusion of reaction
components; after this point the order of reaction was >2 because of this restriction.
The reaction rate constants at different cure temperatures and the activation energy
were also calculated.

Willoughby [55] examined the rheological outputs that can be produced by a cure
rheometer using the Cole–Cole plot, a procedure that is usually applied to a reaction
system. The Cole–Cole plot was applied to rheological data obtained during the
curing of a NR sample at ≥160 °C and above, conditions which ensured that both
curing and reversion behaviour could be covered. This approach was considered
by Willoughby to be capable of identifying the competition between intermolecular
and intramolecular sulfurisation, and between crosslink formation and main chain
scission, and so to have the potential to expand the capabilities of cure rheology.

Ghoreishy and Asghari [56] used a rheometer, a RPA, and a multi-recorder to study
the cure kinetics of an SBR compound under both non-isothermal and isothermal
conditions. Changes to the degree of curing were followed using kinetic parameters,
direct determination of the temperature of the compound in the centre of the mould,
and time–temperature profiles. The results obtained showed that curing behaviour
under both non-isothermal and isothermal conditions could be explained using an
autocatalytic model.

An oscillating disc rheometer was used by da Costa and co-workers [57] to study the
vulcanisation of novel silica-filled NR vulcanisates at different curing temperatures.

201
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Overall, the objective of the study was to evaluate the suitability of caster oil fatty
acid to function as an activator in the NR compounds, and the levels of stearic acid,
castor oil and polyethylene glycol were varied while the amounts of silica filler and
other ingredients were kept constant.

6.5 Finite Element Analysis and Mathematical Modelling

Another approach used to study state of cure is finite element analysis (FEA). During
the curing of actual rubber articles, as opposed to standard 2 mm-thick test sheets,
there is a temperature gradient (i.e., non-isothermal and non-uniform cure conditions)
within the product. Therefore, the physical properties of the product may vary from
those obtained from the test sheet. FEA was used by Gregory and Muhr [58] and
they reviewed the theoretical framework for calculating isothermal cures that were
representative of particular temperature histories. The theory was implemented for
specific cases by solving the transient thermal problems using FEA together with user-
defined variables representing the ‘state of cure’ and the ‘representative temperature’
of an equivalent isothermal cure. The predictions that were obtained were compared
with experimental results and, although the overall agreement was good, there were
limitations and these were discussed in their article.

A team of researchers from the Qingdao University of Science and Technology used
FEA to study the curing process in rubber [59]. They used the technique to solve
the Maxwell equation and heat transfer equation, employing ANSYS finite element
software to calculate and analyse the process of conventional heating in which heat
transfers from outside to the inside of a rubber sample and microwave heating (in
which the reverse occurs). The results obtained showed that the time of microwave
heating was shorter and efficiency higher. Another team from the South China
University of Technology and Shanghai Tire & Rubber Group used thermoelectric
couple cure temperature, DSC, and FEA to study the effect of the heat of reaction
on the cure temperature of rubber blocks and tyres [60]. This approach enabled cure
temperatures to be predicted and these were then compared with those determined
experimentally.

A team from Jiangsu University of Science and Technology [61] presented a systematic
model for the FEA of the rubber curing process. By monitoring how the thermal
conductivity and specific heat capacity varied with temperature they were able to
assess the accuracy of five models (Nth-order, Piloyan, Kamal–Ryan, Kamal–Sourour
and Refei) in providing the optimum kinetics from the curing curves. By using a
rubber piston washer as an example, and ABAQUS software, the workers were able to
calculate the curing model and analysed the induction time, temperature and state of
cure during the curing process. The findings of the study enabled them to demonstrate

202
 Curing and Cure State Studies

that FEA is suitable for providing cost-effective and practical predictions for the
rubber curing process. ABAQUS software was also employed by a team at Qingdao
Technological University, who used FEA to simulate and analyse the curing of radial
tyres [62]. The output of their research included the temperature distributions for
different curing stages and curing temperatures, and the degree of cure at various
time points during the curing process.

Researchers drawn from Shandong University, Changchun Institute of Applied

Chemistry and the Chinese Academy of Sciences developed a mathematical model
for predicting the vulcanisation kinetics of silicone rubber [63]. The thermal
effects of the curing process were computed using the increment method, and the
vulcanisation process was described by FEA. The work by this team illustrated
the importance of the exothermic curing reaction for ensuring that energy was
used efficiently, and the optimum cure temperature for achieving an effective
and uniform cure. The effect of increasing the forced convective heat transfer
coefficient above a certain level was also investigated during the work programme.
Further studies on the curing of silicone rubbers using FEA was reported by Jia
and co-workers [64] and the team presented a methodology for the integrated
processing–structure–property analysis of the vulcanisation of silicone rubber
within a mould. This enabled the temporal evolution, the spatial distribution
characteristics of the hot processing parameters, the crosslinking structure
parameters, and the mechanical property parameters to be obtained by FEA.
The benefits of this approach included optimisation of the curing conditions and
ability to design the curing process to meet specific requirements.

Gough [65] used FEA and a model to simulate the curing behaviour of a laminated
rubber bearing. He made a comparison between the temperatures in a real bearing with
those of the finite element simulation, and the results obtained enabled guidelines to be
composed that can be used to assist with the curing of thick rubber articles.

Karaagac and Deniz [66] described how Adaptive Neuro–Fuzzy Interference Analysis
Systems (ANFIS) can be used to predict the cure of complex and thick rubber articles.
These workers carried out an experimental study that involved 11 rubber compounds
cured using 10 curing temperatures. The obtained data showed that ANFIS methods,
when compared with conventional techniques, gave the lowest average errors for
optimum cure times.

Vebber and Pereira [67] developed a predictive model for choosing cure agents for
rubber blends. The objective was to enable curing agents to be chosen that would
not show a preference for one or other of the rubbers in a blend (as can often happen
due to differences in polarity and double bond concentration) and which can result
in an over-curing of one phase and under-curing of the other. The successful method

203
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

that was developed included a new tool based on a genetic optimisation algorithm
for assessing the partitioning of cure agents in different rubbers and their blends.
The quantitative data that resulted enabled a series of analyses of the solubility of
the cure agents in the different phases, and was validated by correlation with the
physico-chemical and mechanical properties of the prepared blends.

6.6 Other Techniques and Methodologies for Studying Curing and

Cure State

Dinzburg [68] described how a modified version of the Flory equation in combination
with the Cluff–Gladding–Pariser method could be used to provide information about
the cure state and structure of both filled and unfilled hydrogenated-NBR compounds.
The results of the investigation showed that rubber–filler interactions play a significant
part in the network structure. Dinzburg derived an empirical correlation to make it
easier to find the molecular weight between crosslinks without the determination of
the interaction parameter. The study also evaluated the influence of unsaturation in
the polymer on the properties of filled and unfilled compounds.

Nichetti [69] described the use of mechanical property data and rheometric data
to develop a curing kinetic model based on a constitutive equation to predict the
mechanical properties of silica-filled SBR compounds during a curing cycle. The results
obtained using a model for isothermal curing were compared with experimental data,
and the application of the model to non-isothermal cures was considered.

A team in China used the solvent swelling approach to investigate the crosslink density
of different types of rubbers. They published work on unfilled and CB-filled EPDM
compounds cured using both the sulfur-donor compound TMTD and conventional
sulfur cures [70] and on unfilled and CB-filled SBR rubbers that were cured using
different cure systems [71]. Their investigations looked into a number of parameters,
including the relationship between the number of moles of sulfur donor and the
numbers of crosslinks formed, and how the swelling behaviour was influenced by
the type of cure used.

Arrillage and co-workers [72] also used solvent swelling as a means of studying
vulcanisation and the degree of cure of a rubber sample. They also attempted to
combine the use of solvent swelling with DSC to measure the residual exothermicity,
but found that the level of residual curatives was insufficient to provide a measurable
exotherm. Swelling in cyclohexane was used to investigate the cure of a NR
compound vulcanised using a sulfur-based cure and an EPDM compound that had
been crosslinked using peroxides.

204
 Curing and Cure State Studies

El Labban and co-workers [73] developed an experimental device capable of

measuring and controlling the heat transfer within a mould cavity and moulded
part with a view to optimising the curing cycles of thick-sectioned NR mouldings.
An optimisation method based on the calculation of the vulcanisation field and its
sensitivity was established. This enabled a performance index to be derived which
qualified the process by considering a compromise between quality (assessed using
the state of cure) and productivity (which was influenced by the cure time).

6.7 Comparative Studies

Comparative studies have been published of the different techniques that can be used
to investigate the cure state of rubbers. An example is that produced by Warley and
Del Vecchio [74], in which they compared a number of physical tests (e.g., TS and
modulus values) with DSC, solvent swelling, and the Mooney–Rivlin equilibrium
modulus. Their results showed that the DSC method had advantages (e.g., small
samples and relatively rapid), but was limited to materials with sufficiently high heats
of vulcanisation, and the Mooney–Rivlin method proved useful when employing the
small strain modulus equation. The two-solvent technique of the swelling method
was good for determining the polymer–solvent interaction parameter, but gave poor
resolution at high states of cure.

Jaunich and co-workers [75] have investigated the cure state of ethylene-vinyl acetate
copolymers using rheology, DSC and FTIR to see if they were suitable alternatives
to the time-consuming method of gel content determination. The results that the
group obtained showed that all of these methods were capable of being used for
this type of analysis and some of the alternatives gave very similar results to the gel
content approach. Another aspect that they evaluated was the ability of methods to
differentiate between samples that had been aged for a year prior to curing and those
that had not been aged. They found that the alternative methods were able to detect
differences in cure behaviour between the samples, whereas the determination of the
gel content could not.

References

1. R Brown in Physical Testing of Rubber, Chapman and Hall, London, UK,

1996.

2. A. Ciesielski in An Introduction to Rubber Technology, Rapra Technology

Ltd, Shawbury, UK, 1999, p.85.

205
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

3. B.G. Willoughby in Cure Assessment by Physical and Chemical Techniques,

Rapra Review Report No.68, Rapra Technology, Shawbury, UK, 1993.

4. M.J. Forrest in Application to Thermoplastics and Rubbers, Ed., P Gabbott,

Blackwell Publishing, Oxford, UK, 2008, Chapter 6.

5. Thermal Analysis of Rubbers and Rubbery Materials, Eds., P. De,

N.R. Choudhury and N.K. Dutta, Smithers Rapra, Shawbury, UK, 2010.

6. D.W. Brazier, Rubber Chemistry and Technology, 1980, 53, 457

7. A.K. Naskar and P.P. De in Thermal Analysis of Rubbers and Rubbery

Materials, Eds., P.P. De, N. Roy Choudhury and N.K. Dutta, iSmithers,
Shawbury, UK, 2010, Chapter 3.

8. H.J. Borchardt and F. Daniels, Journal of the American Chemical Society,

1957, 79, 1, 41.

9. N.S. Schneider, J.F. Sprouse, G.L. Hagnauer and J.K. Gillham, Polymer
Engineering and Science, 1979, 19, 4, 304.

10. T. Ozawa, Journal of Thermal Analysis, 1970, 2, 3, 301.

11. R.B. Prime in Thermal Characterisation of Polymeric Materials, Ed.,

E.A. Turi, Academic Press Inc, New York, NY, USA, 1981, p.431.

12. A.A. Duswalt, Thermochimica Acta, 1974, 8, 1–2, 57.

13. M.J. Richardson, Pure and Applied Chemistry, 1992, 64, 11, 1789.

14. S. Affolter and M. Schmid, International Journal of Polymer Analysis and

Charaterisation, 2000, 6, 35.

15. M.J. Forrest in Principles and Applications of Thermal Analysis, Ed.,

P. Gabbot, Blackwell Publishing, Oxford, UK, 2008, p.242.

16. M.P. Sepe in Elastomer Technology Handbook, CRC Press, Boca Raton, FL,
USA, 1993, p.105.

17. P. Simon and A. Kucma, Journal of Thermal Analytical Calorimetry, 1999,

56, 3, 1107.

18. H. Ou, M. Sahli, T. Barriere and J-C Gelin in Proceedings of the MATEC
Web of Conferences, 2016, 80, Paper No.16007.

206
 Curing and Cure State Studies

19. M.H.S. Gradwell and D. Grooff, Journal of Applied Polymer Science, 2001,
80, 12, 2292.

20. K.G. Hendrikse, W.J. McGill and J. Reedijik, Journal of Applied Polymer
Science, 2000, 78, 13, 2290.

21. J. Wong-on and J. Wootthikanokkan, Journal of Applied Polymer Science,

2003, 88, 11, 2657.

22. M. Baba, S.C. George, J.L. Gardette and J. Lacoste, Rubber Chemistry and
Technology, 2002, 75, 1, 143.

23. V.M. Karbhari and L. Kabalnova, Journal of Reinforced Plastic Composites,

2001, 20, 2, 90.

24. T.G. Fox and S. Loshaek, Journal of Polymer Science, 1955, 15, 371.

25. X-R. Zeng and T-M. Ko, Journal of Applied Polymer Science, 1998, 67, 13,
2131.

26. S. Cook, S. Groves and A.J. Tinker, Journal of Rubber Research, 2003, 6, 3,
121.

27. A.C.C. Peres and L.M.A. Lopes, Polimeros: Ciencia e Tecnologia, 2006, 16,
1, 61.

28. N.C. Restrepo-Zapata, A.T. Osswald and J. Pablo in Proceedings of the

ANTEC 69th SPE Annual Technical Conference, 1–5th May, Boston, MA,
USA, Society of Plastics Engineers, Brookfield, CT, USA, 2011, p.2264.

29. M.J. Forrest in Rubber Analysis – Polymers, Compounds and Products,

Rapra Review Report No.139, Smithers Rapra, Shawbury, UK.

30. S.R. Khimi and K.L Pickering, Journal of Applied Polymer Science, 2014,
131, 6, 7.

31. K. Formela, D. Wasowicz, M. Formela, A. Hejna and J. Haponiuk, Iranian

Polymer Journal, 2015, 24, 4, 289.

32. V. Vijayabasker and A.K. Bhowmick, Journal of Materials Science, 2005, 40,
11, 2823.

33. E.M. Barrall, M.A. Flandera and J.A. Logan, Thermochima Acta, 1973, 5, 4,
415.

207
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

34. P.M. James, E.M. Barrall, B. Dawson and J.A. Logan, Journal of
Macromolecular Science, Part A; Pure and Applied Chemistry, 1974, A8, 1,
135.

35. E.M. Barral and M.A. Flandera, Journal of Elastomers and Plastics, 1974, 6,
1, 16.

36. R.B. Prime, Thermochimica Acta, 1978, 26, 1–3, 165.

37. A.N. Gent, Transactions of the Institute of the Rubber Industry, 1958, 34, 1,
46.

38. R.S, Rajeev and P.P. De in Thermal Analysis of Rubbers and Rubbery
Materials, Eds., P.P. De, N. Roy Choudhury and N.K. Dutta, iSmithers,
Shawbury, UK, 2010, Chapter 5.

39. T. Provder, S. Malliprakash, S.H. Amin and A. Majid, Macromolecular

Symposia, 2006, 242, 279.

40. T-Q. Zhao, Q-C.Wang, Z-F. Zhou, M-X. Chen, R-J. Chen and S-M. Cai,
Luntai Gongye, 2015, 35, 7, 440.

41. G.M. Maistros, H. Block, C. Bucknall and I.K. Partridge in Proceedings of

the Polymer Blends Symposium, 24–26th July, Cambridge, UK, 1990,
p A7/1–4, 6125.

42. E.M. Andersson, I. Ericsson and L. Trojer, Journal of Applied Polymer

Science, 1982, 27, 7, 2527.

43. M. Ito and J. Swanobori, Nippon Gomu Kyokaishi, 2003, 76, 3, 81.

44. P.S Brown, M.J.R. Loadman and A.J. Tinker in Proceedings of the ACS Fall
Meeting, 8–11th October, Detroit, MI, USA, American Chemical Society
Rubber Division, Washington, DC, USA, 1991, Paper No.36.

45. R.J. Pazur and F.J. Walker in Proceedings of the 176th ACS Fall Technical
Meeting, 13–15th October, Pittsburgh, PA, USA, American Chemical Society
Rubber Division, Washington, DC, USA, 2009, Paper No.45.

46. B. Li, Z. Zhang, D. Ma, Q. Zhai, S. Feng and J. Zhang, Journal of Applied
Polymer Science, 2015, 132, 41, 7.

47. R.N. Datta and F.A.A. Ingham in Rubber Technologist’s Handbook, Eds.,
J.R. White and S.K. De, Rapra Technology Ltd, Shawbury, UK, 2001, p.175.

208
 Curing and Cure State Studies

48. J.S. Dick and E. Norton, Kautschuk, Gummi, Kunststoffe, 2014, 67, 4, 12.

49. J.S. Dick and E. Norton in the Proceedings of the 182nd ACS Fall Technical
Meeting, 9–11th October, Cincinnati, OH, USA, American Chemical Society
Rubber Division, Washington, DC, USA, 2012, Paper No.90.

50. J.S. Dick and T. Xue, Rubber Fibres Plastics International, 2015, 10, 3 186.

51. I.D. Rosca and J.M. Vergnaud, Polymer, 2003, 44, 14, 4067.

52. A.J. Marzocca and M.A. Mansilla, Journal of Applied Polymer Science, 2007,
103, 2, 1105.

53. M. Nanda and D.K. Tripathy, Polymers and Polymer Composites, 2010, 18,
8, 417.

54. D-B. Guan, Z-Y. Cai, X-C. Liu, B. Lou, Y-L. Dou, D-H. Xu and W-G. Yao,
Chinese Journal of Polymer Science, 2016, 34, 10, 1290.

55. B.G. Willoughby, Journal of Applied Polymer Science, 2016, 133, 41, 9.

56. M.H.R. Ghoreishy and G.H. Asghari, Iranian Journal of Polymer Science,
2007, 20, 1, 11.

57. H.M. da Costa, T.A.S Abrantes, R.C.R. Nunes and L.L.Y. Viscont, Polymer
Testing, 2003, 22, 7, 769.

58. I.H. Gregory and A.H. Muhr in Proceedings of the International Rubber
Conference (IRC) 1997, 6–9th October, Kuala Lumpur, Malaysia, 1997,
p.858.

59. T. Li, G-Z. Yang, H-L. Chen and Q-L. Li, China Rubber Industry, 2013, 60,
1, 42.

60. L-T. Kou, Q. Gu, J-P. Ding and Z-Y. Yao, China Rubber Industry, 2007, 54,
10, 619.

61. J. Zhang, W-X. Tang, G-L. Wang and X. Zhang, China Rubber Industry,
2012, 59, 7, 401.

62. H-J. Lu, T. Chen, Y-S. Wang and H. Ma, Luntai Gongye, 2009, 29, 3, 172.

63. X. Wang, Y. Jia, L. Feng and L. An, Macromolecular Theory and Simulations,
2009, 18, 4–5, 268.

209
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

64. Y. Jia, X. Wang, L. Feng and L. An, European Polymer Journal, 2009, 45, 6,
1759.

65. J. Gough, Malaysian Rubber Technology Developments, 2007, 7, 1, 4.

66. B. Karaagac and V. Deniz in the Proceedings of the International Rubber

Conference (IRC) 2010, 17–19th November, Mumbai, India, 2010,
Paper No.20.

67. C. Canete Vebber and C. Nunes Pereira, Journal of Applied Polymer Science,
2015, 132, 18, 10.

68. B.N. Dinzburg, Kautchuk, Gummi, Kunststoffe, 1999, 52, 6, 413.

69. D. Nichetti, European Polymer Journal, 2004, 40, 10, 2401.

70. T. Songchao, L. Chong, L. Xuedai and W. Zhangyong, China Synthetic

Rubber Industry, 2002, 25, 5, 289.

71. T. Songchao, L. Chong, L. Xuedai and W. Zhangyong, China Synthetic

Rubber Industry, 2003, 26, 1, 21.

72. A. Arrillaga, Z. Kareaga, E. Retolaza and A.M. Zaldua, Gummi Fasern

Kunststoffe, 2011, 64, 11, 669.

73. A. El Labban, P. Mousseau, R. Deterre and J.L Bailleul, Rubber Chemistry

and Technology, 2010, 83, 4, 331.

74. R.L. Warley and R.J. Vecchio in Proceedings of the ACS Spring Meeting,
26–29th May, Akron, OH, USA, American Chemical Society, Washington,
DC, USA, 1987, Paper No.29

75. M. Jaunich, M. Bohning, U. Braun, G. Teteris and W. Stark, Polymer Testing,

2016, 52, 133.

210
7
Characterisation of Rubber Process Fume

7.1 Introduction

The composition of rubber processing fumes is complex because it is generated by the

activities of an industry, which in itself is very complex in a number of ways (e.g., in
the ingredients that it uses, the products that it produces, and the processing methods
that it employs to manufacture them). It is because of this complexity, and the pressing
need to obtain accurate information on its composition due to increasing health and
safety legislation, that its study has attracted so much attention from rubber chemists/
analysts over the last 50 years. This chapter assists the rubber analyst who is tasked
with studying rubber fumes by providing a background to their origin, nature and
composition as well as reviewing the various approaches and analytical techniques
that have been used to characterise them over the last 40 to 50 years.

Due to their chemical complexity, there are a large number of reactions and interactions
that occur when rubber compounds are subjected to the high temperatures used during
their processing and these can generate low-molecular weight (MW) compounds
that contribute to rubber fumes [1]. Another principal variable that can influence
the composition of rubber fumes is the way that rubber is processed, and the main
processing stages for the majority of rubber components involve the following
operations [2]:

• Mixing of the ingredients [3]

• Shaping operations (e.g., extrusion) [2]

• Vulcanisation to crosslink the rubber [4, 5]

• Post-vulcanisation processes (e.g., post-curing in an oven) [5]

Of the factors stated above, the processes where fumes are most likely to be generated
are during the mixing, vulcanisation and post-curing stages.

The only international definition of rubber fumes believed to be available at this time
is that produced by the International Organization for Standardization (ISO) during its
work on standards for the study of rubber fumes. This ISO definition is: ‘A variety of

211
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

substances emitted from rubber compounds into a workplace atmosphere as a result

of industrial processing, the composition of which depends on the formulation of
the compounds concerned, the process technology in use and the associated process
parameters’. In 2013, they published a standard, ISO/TS 17796:2013 [6], for the
qualitative screening of rubber fumes in the workplace and storage environments
using a thermal desorption (TD) gas chromatography (GC)–mass spectrometry (MS)
method. ISO have also published a standard for the quantitative analysis of the
substances in rubber process fumes [7]. This standard, which should be published
within the next couple of years, will include the hazardous substances that have
been identified in rubber fumes and, for each one, provide examples of validated
test methods.

Another definition of rubber fumes is provided by the UK Health and Safety Executive
(HSE) in its EH40 publication and in its method for the determination of rubber dust
and fume: Method for Determination of Hazardous Substance (MDHS) 47/2:1999
[8]. This HSE method has been used to determine the levels of rubber fume in the
UK rubber industry for many years. The HSE definition of rubber fumes is: ‘Rubber
fumes is evolved in the mixing, milling and blending of natural rubber and rubber
or synthetic elastomers, or of natural rubber and synthetic polymers combined with
chemicals, and in the processes which convert the resultant blends into finished
products or parts thereof, and including any inspection procedures where fumes
continues to be evolved’.

From the comprehensive research work that has been carried out over the years, it
is apparent that any analytical programme to characterise the chemical nature and
composition of rubber fumes should embrace the following classes of species:

1. Gases.

2. Vapours.

3. Aerosols.

4. Special case compounds – N-nitrosamines, aromatic amines and polyaromatic

hydrocarbons (PAH).

Considering all of these physical forms and classes of chemical compound ensures
that a comprehensive analysis of rubber fumes is undertaken, with no species being
overlooked or disregarded. A comprehensive review of the composition and nature
of vulcanisation fumes in the rubber industry was published recently by Forrest [9].
Another comprehensive review of the composition of rubber fumes, based on the
information provided by 95 publications, was prepared by ISO/TC 45 (rubber and
rubber products) and published as an ISO Technical Report in February 2017 [10].

212
 Characterisation of Rubber Process Fume

The information presented in Section 7.2 provides a summary of the nature and
composition of rubber fumes with regard to the variety of chemical compounds that
can be present and their origins, the different physical forms in which they can exist
within the factory environment (i.e., gases, aerosols) and their origins. The principal
factors that influence the composition of rubber fumes and the sophisticated analytical
approaches that have been developed to capture and analyse them are introduced in
Sections 7.3 and 7.4, respectively. Finally, in Section 7.5 an overview is presented of
the characterisation data that has been published on rubber fumes, both as a result
of laboratory studies and from carrying out monitoring work at commercial rubber
processing sites.

7.2 Composition of Rubber Process Fume

This section provides a brief overview of the chemical substances that can be found
in rubber process fumes and where they have originated from within the rubber
compound.

Experimental work that has been carried out, both in laboratory and factory
environments, has shown that rubber fumes comprise a wide range of chemical species
with respect to MW and compound type (i.e., polarity and structure). A number of
studies have reported over 30 species in the fumes from a particular rubber compound
(Section 7.5).

The substances that are present in rubber fumes can be from a wide range of different
chemical classes and a number of scientists [11, 12] have identified a large range of
chemical species, with the following types being represented:

• Aliphatic hydrocarbons – straight chain and cyclic

• Aromatic hydrocarbons
• Halogenated compounds
• Isothiocyanates
• Ketones
• Nitrosamines
• Thiazoles
• Aldehydes and alcohols

• Esters and ethers

213
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Amines

• Sulfur compounds

The relative contribution of these substances to any given sample of rubber fumes is
dependent on the ingredients present in each rubber compound that is contributing to
the rubber fumes at a particular moment in time. Also, these compounds are present
in rubber fumes in a number of physical forms (i.e., gases, vapour and aerosols) and
the relative levels of these different physical forms vary in different areas (e.g., mixing,
moulding, storage) within a rubber factory.

The different physical forms of the compounds in rubber fumes leads to a way of
referring to these species based on the ability of the human eye to detect them in the
atmosphere: ‘visible’ or ‘invisible’ fractions. The definition of these two classes is
provided in Section 7.4.1.

Research work has shown that there are three distinct sources of rubber fume
compounds [13]:

• Volatile ingredients (e.g., antioxidants and plasticisers).

• Volatile impurities in ingredients (e.g., residual monomers and manufacturing

impurities).

• Volatile reaction and breakdown products of chemical reactions that occur during
processing (e.g., vulcanisation and antioxidant activity [14, 15]).

7.3 Principal Factors Affecting the Composition of Rubber Fume in

the Workplace

As mentioned in Section 7.1, there are a number of factors that can alter the
composition of rubber fumes and which need to be considered to understand their
variable nature. This section provides an overview of the influence of two of the most
important factors: rubber composition and processing temperature.

7.3.1 Influence of Rubber Compound Formulation on Composition of

Rubber Fume

A number of authors have demonstrated that the chemical species in rubber fumes directly
relate to the polymer and compounding ingredients present in a rubber compound.

214
 Characterisation of Rubber Process Fume

With respect to the influence of the polymer in the rubber compound, Donskaya and
co-workers [16] listed some of the volatile compounds that can be present in the gas
phase of rubber fumes when different synthetic rubbers are in the compound:

• Styrene-butadiene rubber (SBR): styrene, butadiene, ethyl benzene.

• Polybutadiene: butadiene, 4-vinylcyclohexene, cyclododecatriene.

• Nitrile rubber (NBR): butadiene, acrylonitrile, butadiene trimers, 4-cyclohexenitrile.

An example of the published work that has demonstrated the influence of the
compounding ingredients is that of Blanden and Isherwood [17]. They found that
the substances collected were members of a number of different classes of compound
(e.g., acid, basic, neutral) and some examples from four rubber compounds that were
in the basic (i.e., alkaline) class are shown below:

• SBR: aniline, phthalimide, dicyclohexylamine, 2-(4-morpholinyl)benzothiazole,

N-(1,3-dimethyl butyl)-N'-phenyl-p-phenylenediamine (6PPD).

• Ethylene propylene diene monomer (EPDM): morpholine, N-cyclohexyl

formamide, N-butyl-1-butanamide, 2-cyclohexyl benzothiazole.

• Butyl rubber: benzidine, N-(2,2-dimethylpropyl)-N-methyl benzenamine, N-ethyl-

2-benzothiazolamine, benzothiazole (BT), phenyl benzimidazole.

• NBR: aniline, dimethyl-2-butanamine, tetrabutyl urea.

Further information on the work carried out by Blanden and Isherwood is provided in
Section 7.5.1; this section and Section 7.5.2 provide other examples that illustrate how
the composition of a rubber compound has a profound influence on the substances
that are present in its process fumes.

7.3.2 Influence of Different Processes and Processing Temperatures on

the Composition of Rubber Fumes

The type of rubber process that is being carried out in a rubber factory will have an
influence on the rubber fumes that are generated. Taking into account both visible
and invisible fractions of fume, Dost and co-workers [18] showed that in the general
rubber goods (GRG) sector, the overall exposure to rubber fumes varies according
to the process that is being carried out, as shown below:

moulding > extrusion > milling

215
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The differences found for these three generic processes relate to factors such as the
process temperature and the amount of the surface area of the rubber exposed to
the atmosphere.

Regarding the influence of processing temperature, work by Willoughby [19] using

the Rapra gas transfer mould (GTM)/GC–MS combination with an SBR compound
illustrated how it can affect the types and levels of hydrocarbon species that are
detectable in its fumes. For example, in the case of styrene, increasing the vulcanisation
temperature from 145–165 °C and 180–240 °C increased the amount present in the
fumes from 2–180 to 9–500 µg/m3. For 4-vinylcyclohexane, none was found at the
lower temperature, but 30–210 µm/m3 was observed at the higher temperature. The
data also showed that the situation was complicated, with the levels of some compounds
reducing and others increasing as the vulcanisation temperature was increased.

Changing the moulding temperature can also affect the amount of a particular
reaction product. For example, Maisey [20] reported the effect that changing the
moulding temperature has on the amount of carbon disulfide (CS2) released from a
sulfur-cured EPDM rubber containing the accelerator tetramethylthiuram disulfide
(TMTD). Increasing the temperature from 150 to 200 °C and then 250 °C increased
the amount of CS2 released from 1.03 µmol to 1.91 µmol and, finally, to 2.34 µmol.
This work, which was carried out at Rapra, also showed that by modifying the cure
system, CS2 was not liberated.

The total amount of rubber fumes can be influenced by processing temperature as

well, as demonstrated by Willoughby [21]. In his study, as expected, the total mass of
fumes released from a rubber sample was found to increase with temperature, rising
from 0.18% w/w at 160 °C to 0.75% w/w at 190 °C.

7.4 Capture and Analyses of Rubber Process Fumes

7.4.1 Different Sampling Techniques and Analytical Methods

A number of methods and techniques have been used and they have been designed
bearing in mind the compositional complexity of rubber fumes, the objective being
to achieve the most accurate characterisation possible.

In addition to the comments made regarding their composition in Sections 7.1–7.3, in

general, rubber fumes can also be considered as being comprised of visible and invisible
fractions [22]. Gases and vapours in the fumes make up what can be regarded as the
invisible component of rubber fumes, with aerosols forming the visible component.

216
 Characterisation of Rubber Process Fume

The visible fraction is believed to be produced at two stages in the production process:
the mixing stage (due to the large mechanical stresses and local overheating within the
rubber compound), and the vulcanisation stage (due to the condensation of vapours
of high-boiling-point substances). In both of these cases, the aerosol particles contain
numerous substances that are captured by the oil-like material that is forming the
aerosol. The 1974/75 BRMA/Rapra environmental survey reported that the solvent-
soluble portion of airborne particulates provides an effective measure of visible fumes
from hot rubber. MDHS Method 47/2:1999 [8] is the standard method used in the
UK and regards the cyclohexane-soluble fraction of the airborne contamination
collected on a glass fibre filter as being the rubber fumes portion, with the insoluble
fraction being regarded as the rubber dust fraction. The analytical methods used in
a number of other countries are based on this MDHS method. However, difficulties
can be encountered if cyclohexane-soluble organic dust and rubber chemicals, such
as stearic acid, are present, resulting in overestimation of rubber fumes. Also, an
analysis of the rubber dust and fumes data on the European Union Concerted Action
programme database [23] has shown that using Teflon™ filters, rather than the glass
fibre filters recommended by MDHS 47/2:1999, can influence the results.

To obtain analytical data on the invisible (i.e., gases and vapours) fraction of rubber
fumes, two principal approaches have been used by analysts over the years. The first
involves trapping the fume compounds (e.g., using adsorbents such Tenax™) in a
rubber factory and then using an analytical technique (e.g., GC–MS) in a laboratory
to separate and identify the compounds trapped. This type of approach, which is
covered in more detail in Section 7.5.1, is incorporated into an ISO standard published
in 2013, ISO/TS 17796:2013 [6], which describes a method for the trapping and
identification of volatile components of rubber fumes using active sampling on a
poly(2,6-diphenylphenylene oxide)-type sorbent (i.e., Tenax™) and subsequent
analysis by TD GC–MS. This method can be used to screen emissions from the
processing of rubber compounds in the workplace and in storage environments, but
one potential limitation is that the data will be influenced by the specificity of the
sorbent (Section 7.4.2). Data generated on a number of rubber compounds using this
ISO method have been published [24].

The second type of approach is laboratory-based and specialised equipment (e.g.,

the Rapra GTM) has been manufactured that closely mimics the curing processes
used in industry (e.g., compression moulding). This equipment is interfaced with the
analytical equipment (e.g., GC–MS) so that the fumes are transported directly into
it, thereby minimising the losses that may take place if an intermediate ‘trapping’
stage is employed. More information is provided on this approach, and the results
obtained as a result of it, in Section 7.5.2.

217
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

7.4.2 Influence of Different Sampling and Analytical Techniques on

Rubber Fumes Data

The complex chemical composition of rubber fumes means that ensuring that all of
their components (i.e., gases, volatile compounds, aerosols) are trapped in an efficient
manner to enable a full analysis to be performed is a challenging task. Accordingly, a
number of scientists have carried out research to obtain information on how various
‘sampling trains’ of absorbent tubes, other sampling media, and sampling systems,
can influence the data obtained. Work has also been carried out to investigate which
analytical techniques and methodologies offer the most effective means of identifying
and quantifying the compounds present in rubber fumes.

Giese and co-workers at the German Institute for Rubber Technology [25] described
the development of a sampling technique, the selection and characterisation of
adsorbents, and the development and testing of a sampling apparatus for the sampling
and analysis of vulcanisation vapours and emissions. In this way, they developed
and reported on a multi-faceted sampling and analysis methodology capable of
collecting, and then identifying and quantifying, the various types of components
present in rubber fumes. The overall procedure has been validated for ≈50 single
chemical substances.

A generic illustration of the types of sampling analysis combinations used by Giese

and co-workers and others in the occupational health and fire gas analytical fields
is provided in Table 7.1.

Table 7.1 Examples of sampling-analysis combinations for trapping various

types of airborne substances
Substance Collection media Analysis method
Gaseous compounds (e.g., Gas sampling bag Direct injection GC–FID or GC–MS
carbon disulfide)
Volatile aliphatic and aromatic Activated charcoal Thermal desorption GC–FID or GC–
hydrocarbon compounds adsorbent sampling MS
tube
Amines and amides Silica gel adsorbentSolvent desorption GC–nitrogen-
sampling tube phosphorus detector
Aldehydes and ketones Modified silica gel Solvent desorption high-performance
sampling tube liquid chromatography–ultraviolet
detection
General organic compounds, Tenax/XAD-2/Porapak Thermal desorption GC–MS
including those with lower adsorbent sampling
volatility tubes
Aerosol compounds Glass fibre filter Solvent extraction – FTIR analysis
FID: Flame ionisation detector
FTIR: Fourier–Transform infrared

218
 Characterisation of Rubber Process Fume

The existence of so many types of substances can result in one species interfering
with the collection and identification/quantification of one or more of the others. In
order to check on the validity and accuracy of the 'Giese' multi-faceted system, both
laboratory trials and workplace monitoring at tyre and GRG factories have taken
place. A good level of reliability for the system, and accuracy of the results obtained
using it, has been demonstrated by these trials and a wide diversity of compounds
from the following substance groups were detected in rubber factories [25]:

• Highly volatile hydrocarbons

• Amines

• Aldehydes and ketones

• Chlorinated hydrocarbons

• Low-volatility vapours

• Aerosols

A number of other studies have been published by the scientists at the German Institute
for Rubber Technology. Some of these address more specific components that can be
found in rubber fumes. For example, Kuhn-Stoffers and co-workers [26] described
investigations into the use of activated carbon and silica gels (surface-modified with
dinitrophenyl hydrazine) in terms of their suitability for sampling aldehydes and
ketones in vulcanisation fumes. The study demonstrated that the derivitisation reaction
was preceded by adsorption on the substrate material and that subsequent quantitative
transformation to the hydrazone required different reaction times depending on the
reactivity of the carbonyl components.

A study has been published which looks at the influence of plasticiser aerosols on
the sampling and analysis of rubber fumes in the rubber industry [27]. A sampling
system for fumes and aerosols from rubber products needs to be developed with
special attention to the influence of plasticiser aerosols on the sampling process itself.
In this study, the efficiency of sampling with a combination of selective adsorbents
in the presence of plasticiser aerosols was evaluated using test gas mixtures and
test aerosols in experiments that simulated the atmospheres that may be found in
rubber factories. The analytical determination of individual substances and aerosols
was carried out by GC and FTIR after desorption of the adsorbents with a solvent.
Interferences caused by plasticiser aerosols were found by the scientists to occur only
at high concentrations.

The collaborative study presented by Mai (LRCC in France) and Giese (DIK in
Germany) [28] at the International Rubber Conference in 2013 included the recovery

219
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

rate data for a number of the collection systems available for sampling rubber fumes.
Some examples of those presented are shown below:

• Highly volatile aliphatic and aromatic compounds collected on activated charcoal:

90.2–100% recovery.

• Highly volatile sulfur compounds (e.g., CS2) collected using a sampling bag:
80.0–97.5% recovery.

• Aerosols collected using a glass fibre filter: 94.0–98.9% recovery.

7.5 Analytical Studies Carried Out on Rubber Fumes

This section provides an overview of the analytical results obtained on rubber fumes
by the two principal two routes employed by research scientists, namely, studies
conducted on the fumes present in rubber factories (Section 7.5.1) and those carried
out on fumes generated in laboratories (Section 7.5.2).

7.5.1 Rubber Fume Data Resulting from Factory Studies

The Nordic Curing Fumes Project [29] investigated the effect that curing fumes have
on the environment. This project was set up by the Swedish Board for Environmental
Protection, supported by Norsam (the trade association of the Nordic rubber
industries), and looked into the:

• Composition of curing fumes.

• Environmental consequences of curing fumes.

• Possibilities to control the release of curing fumes from factories.

Overall, the results showed that amines and sulfur compounds were the main
compounds in the vapour phase of rubber fumes from sulfur curing, and that amines,
together with aliphatic and aromatic hydrocarbons, were the main components of
the vapour phase of rubber fumes from peroxide curing. The authors also referred to
the generation of N-nitrosamines from the nitrosation of secondary amines formed
as a result of the action of certain rubber accelerators and the way that this could be
avoided. Carbon black (CB) and aromatic process oils as sources of PAH in rubber
fumes were also mentioned, and the authors stated that 14 PAH compounds had
been found in the air in a Russian tyre works [30] and that these had the potential
to react with nitrogen in the air to form nitro-derivatives [31] which are carcinogens
and mutagens.

220
 Characterisation of Rubber Process Fume

A group of industry specialists, on behalf of the EU-EXASRUB consortium, used

the rubber dust and rubber fumes data from European Union Concerted Action to
establish exposure trends in the EU rubber industry between the 1970s and 2003
[32]. One of the overall conclusions was that the average level of exposure to rubber
dust and its cyclohexane soluble fraction (i.e., rubber fume) had declined, on average,
by 4.3 and 3.3%, respectively, per year from the mid-1970s to 2003. This study also
indicated that differences in sampling methodology, for example, the use of either
glass fibre or Teflon™ filters, could influence the results obtained. The results of this
study were broadly in line with those reported by other authors, such as Kromhout
and co-workers [33].

Frolikova and co-workers [34, 35] reported on the levels of various substances in
the air in different departments of a tyre factory. The quantity of an important PAH
compound, benz(a)pyrene, was determined in the working environment in four areas
of the factory – the CB storehouse, mixing department, mixing area, and during
transportation and mixing. The concentrations found varied between 0 and 0.45 mg/
m3, with the highest levels detected in the CB storehouse. Also, levels of N-nitrosamines
in vulcanisation fumes and in the vulcanisates after curing were determined, with the
levels shown to vary between 0.1 and 9.1 µg/m3 in fumes, and between 10 and 482
µg/kg in the vulcanisates. The total levels of N-nitrosamines present in the working
atmosphere of different areas within the factory were established; levels between
2.1 µg/m3 (buffing area), 11.65 µg/m3 (vulcanisation area) and 94.27 µg/m3 (finished
product storehouse) were found.

Nudel’man reported on both the aerosol and gaseous components of the rubber fumes
generated by a number of rubber compounds, including those based on NBR and
NBR/polychloroprene (CR) blends [36]. Aerosols were found to be formed at two
stages of production: mixing (due to large mechanical stresses and local overheating)
and vulcanisation (due the condensation of vapours of high-boiling-point substances).
In both cases the aerosol particles contained numerous substances captured by the
drops of oil-like material forming the aerosol. Nudel’man listed the components that
had been trapped and identified in an aerosol formed during the vulcanisation of a
NBR and, although nine compounds were detected, 90% of the fumes comprised
dibutyl phthalate.

Blanden and Isherwood [17] published data generated on rubber fumes collected at
five unnamed sites in the UK engaged in different rubber manufacturing operations
(e.g., car tyre curing, injection moulding and compression moulding). Particulate
samples were collected on glass fibre filters and an electrostatic precipitator. Tenax™
absorption tubes fitted with glass fibre pre-filters were used to collect compounds of
medium volatility. No attempt was made to collect gases or highly volatile compounds.
Prior to analyses, each extract was separated into five fractions (basic, acidic, polar

221
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

neutral, aromatic and aliphatic) and then the substances in each fraction identified
qualitatively by GC–MS. In all cases, the fumes samples were complex mixtures of
chemical substances with a number of compound classes represented:

• Aliphatic compounds
• Aromatic compounds
• Polar, neutral compounds
• Acidic compounds
• Basic compounds

Some of the compounds detected in the ‘basic compound’ class for three types of
rubber have been presented in Section 7.3.1.

Cocheo and co-workers [37] reported the results of an industry study that involved
the collection of samples in the vulcanisation areas of a shoe-sole factory and a tyre
retreading operation. Samples were also taken in the extrusion areas of the retreading
company and an insulated cable manufacturer. In all cases, the fumes samples were
trapped on activated charcoal, desorbed in the laboratory with trichlorofluoromethane
and analysed by GC–MS. A very large number of species were identified in the four
areas, with some overlapping in the chromatograms, thereby resulting in a number
of ‘unidentified’ peaks. An extensive table of assignments was provided for all of the
individual compounds found in the four areas, together with an indication of their
probable source within the rubber compound (i.e., the ingredient). The total number
of compounds identified and quantified was 99, with some examples of the number
in each class given below:

• Alkanes (16 compounds)

• Aliphatic hydrocarbons (32 compounds)

• Chlorinated compounds (4 compounds)

• Phenols (6 compounds)

• Esters (6 compounds)

The authors appreciated that charcoal is not a universal sorbent and some compounds
would not have been trapped in detectable quantities. However, the study did produce
results that were in general agreement with other studies that had been carried out.

In a Russian study [38], the gases produced during the manufacture of silicone rubber
tubes for medical applications were found to include a range of siloxanes, ethyl

222
 Characterisation of Rubber Process Fume

formate, methanol, ethylene and o-dichlorobenzene, but at concentrations below the

maximum permitted in the factory. In addition, ventilating discharges from the site
were found to contain mainly formaldehyde.

In a study presented at the International Rubber Conference in 2013, Khalfoune [24]

of the LRCCP Rubber and Plastics Research and Testing Laboratory described how
ISO/TS 17796:2013 for the trapping and analysis of rubber fumes (Section 7.1) had
been used to determine the substances present in the fumes generated by a number of
rubber compounds of known composition. The presentation also included information
on laboratory studies that had been carried out on standard compounds, and this
work enabled a link to be made between the volatile species in rubber fumes and
specific compounding ingredients in the rubber. The results presented by Khalfoune
for a peroxide-cured EPDM included the following ingredient/substance relationships:

• Ethylene norbornene from the Vistalon EPDM rubber.

• Six organic compounds from the Peroximon F 40 peroxide curative.

• Ethanol from the Silane ST69 coupling agent.

• Ethyl palmitate from the stearic acid.

The information presented by Khalfoune was consistent with the work carried out
by other scientists, such as Willoughby [13, 39] (Section 7.5.2).

7.5.2 Rubber Fume Data Resulting from Laboratory Studies

The composition of fumes evolved from rubber compounds when they are heated
to processing temperatures using laboratory-based systems has been investigated
by a number of scientists. Using these types of experiments, a considerable amount
of fundamental scientific work was carried out on the nature and composition of
rubber vulcanisation fumes by Rapra Technology. These studies, which began in the
1970s and carried on through into the late-1990s, were led by Bryan Willoughby
[40–42]. The results obtained, and the understanding of the chemical relationships
that resulted, enabled the composition of rubber fumes from a given rubber compound
to be predicted with a reasonable degree of accuracy.

Rapra carried out a series of over 40 vulcanisation experiments in their GTM using
industrially relevant formulations produced from a ‘pool’ of 75 ingredients. The
species that were given off in the fumes were studied by interfacing the GTM with
GC–MS. This approach ensured that representative fumes samples were generated
and then transferred into the GC–MS system quickly with negligible loss. More than

223
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

150 chemical species were detected and important ingredient/process conditions/

emission relationships established, for example:

• A list of the formulations which yielded a single, specific compound and the
common ingredients in such formulations.

• A list of the formulations which contained a specific ingredient, and the common
compounds emitted from these formulations.

• The overall effect of the formulation, curing temperature and temperature of the
rubber at the time of analyses.

The complexity of rubber compounds meant that the relationship between the
ingredients and the emissions obtained was not necessarily simple. For example,
an individual component of rubber fumes may have more than one source in a
formulation. This work at Rapra did reveal certain key trends however. For example,
three distinct sources of volatile emissions were recognised (Section 7.2). This
information was provided by an in-depth study published by Willoughby [13] that
contained sections on the origin of specific individual compounds, such as CS2, and a
table that listed the potential origins of over 40 volatiles species that could be detected
in rubber fumes. The results of the Rapra Vulcanisation Project and description of
the predictive rubber fumes software that resulted from it are described in a book
authored by Willoughby [39].

In addition to these summary publications, a large number of other studies and

publications have been published by Willoughby and other scientists at Rapra
during the course of their work on rubber fumes. For example, in an article entitled
‘Prediction of on-site performance for vulcanisation fume’, Smith and Willoughby
[43] described rubber fumes as comprising two distinct phases. These two phases,
which have been referred to in Section 7.2 and 7.3.2, were:

• Visible portion – composed of aerosols formed by the condensation of hot vapours,

which can be trapped onto filters providing that the sampling flow rate is high
enough.

• Invisible portion – composed of a complex mixture of species that remain in the

gaseous phase. The wide range of species present mean that either a range of
adsorbents, or a range of liquid media for trapping by absorption, have to be used
to ensure that the data subsequently obtained by analysis are fully representative.

Levin [11] and Asplund [12] both reported on a study carried out by the company
Trelleborg Industri AB on the composition of curing fumes released from a range
of rubber compounds under both industrial and laboratory conditions. All five
compounds were CB-filled with four containing typical sulfur-curing systems and

224
 Characterisation of Rubber Process Fume

one compound with a peroxide-curing system. The laboratory-based phase of the

work was carried out at Rapra Technology using their GTM/GC–MS combination.
A combination of sampling media had to be used to capture fume samples in the
factory environment due to the multi-component, complex nature of rubber fumes.
The following sampling train was used in the study:

• Adsorbent tubes for sulfur compounds

• Tenax™ tubes for organic compounds

• Charcoal tubes for less volatile organic compounds

• Quartz wool filters for aerosols and particulates

• Impinger tubes for amines

• Continuous registering instrument for total hydrocarbon content (THC)

The analysis of the fumes that were captured by this sampling train from the rubber
compounds resulted in the identification of 221 substances which, as introduced in
Section 7.1, could be divided into a large range of chemical groups.

Becklin and co-workers [44, 45] described how vulcanisation fumes from a number
of rubber compounds were collected directly from a laboratory-scale moulding
press and introduced into a GC–MS system for identification and quantification of
the organic compounds present. This system was similar to the one set up by Rapra
Technology and enabled rapid screening of various rubber formulations [e.g., natural
rubber (NR), EPDM, SBR, CR and NBR] for the identification and quantification of
organic compounds present in their vulcanisation fumes. The effect of curing time and
pressure were also investigated as well as recipe changes within a given compound
series. The analytical results of over 35 organic compounds were presented and the
pathways for the generation of some of these chemicals described.

Rozynov and co-workers [46] used a mini-curing system/GC–MS combination for

characterising rubber vulcanisation fumes. The system enabled the rapid screening of
various rubber formulations to identify and quantify the organic compounds present
in their vulcanisation fumes. Some of the compounds that were detected in rubber
fumes using this system were bis(2-ethylhexyl)phthalate, dimethylformamide, CS2,
acetophenone and isopropylbenzene.

Zietlow and Schuster [47] applied analytical methods based on dynamic headspace
sampling in combination with GC–MS to the analysis of the components present in
fumes generated during laboratory-scale vulcanisation experiments with a ‘heatable
reactor cell’ and a compression mould that simulated workplace conditions.

225
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The results obtained from a small number of rubber compounds made it possible to
identify, or classify, ≈300 individual compounds and, in common with other studies
of this type, ingredient/emission relationships were apparent, enabling the assignment
of a given compound to a specific raw material.

Lucas and co-workers [48] carried out laboratory investigations into the crosslinking
efficiency and emissions behaviour of a number of peroxides used in EPDM-based
compounds and compared the results obtained with those achieved with a N-tert-
butyl-2-benzothiazole sulfonamide/tetrabenzylthiuram disulfide accelerated sulfur-
cured EPDM. The choice of peroxide was shown to be important because using one
with a high crosslinking efficiency meant that less had to be employed to achieve a
successful cure with a rubber compound, and so the of level moulding fumes was
lower. In addition to fogging tests for Deutsches Institut für Normung e.V., DIN 75201,
the authors used thermodeposition GC–MS [as specified by the German Association
of the Automotive Industry (VDA) 278 method] to identify the emissions produced
by the rubber compounds, particularly the breakdown and reaction products of the
peroxides. A number of principal cleavage products for five commercial peroxides were
identified and their share of the total emissions estimated. The number of compounds
and their respective levels in the emissions, along with their chemical nature and MW,
varied considerably according to the type of peroxide used. However, the results
obtained showed, in common with the work carried out by Rapra Technology, that
it is possible to predict the range of breakdown products that will be produced by a
peroxide of known chemical structure.

Nudel’man, in conjunction with Antonovski, [49, 50], also looked into the volatile
decomposition products of three commercially available peroxides used to vulcanise
rubber compounds. The investigation was performed by compounding the peroxides
into an EPDM rubber formulation and then curing sheets of the rubber using a typical
commercial moulding temperature (i.e., 150 °C). This moulding process was carried
out between aluminium sheets that trapped all of the volatile compounds that were
formed during the vulcanisation process within the rubber. Once the rubber sheets
had cooled, samples were removed and placed into a TD unit that was connected
to a GC–MS system. In addition to low-MW substances that were unrelated to the
peroxide (e.g., oligomers from the rubber, antioxidants), around ≥10 compounds were
found to have been formed from each of the peroxides investigated. The formation
of cyclic compounds, such as tetrahydrofuran compounds, was thought to be due to
the cyclisation of intermediate breakdown products.

A couple of studies have looked into the presence of specific species in rubber fumes.
Chikishev and co-workers reported the results of a Russian study [51] that used
experiments using MS to determine qualitatively the accelerator and antioxidant
content of gases evolved from rubber compounds heated at 75 °C. Four compounds

226
 Characterisation of Rubber Process Fume

based on SKN-26 NBR with three types of crosslinking accelerator and two
antioxidants were used as models. Over 15 types of compound were detected by
their mass spectra and assigned, where possible, to the various antioxidants and
accelerators used in the compounds.

In a study by Wommelsdorff and co-workers [52], trace amounts of carcinogenic

PAH were found to be present in the fumes resulting from the vulcanisation of rubber
products. Results of laboratory-based curing studies on 17 rubber compounds based
on two rubber types (EPDM and SBR) and containing different filler and oil additives
were presented. They showed that a reduction of the PAH content of vulcanisation
fumes was achievable through the use of a mineral oil with low-PAH levels. Varying
the compounding ingredients (e.g., use of silica as opposed to CB) also influenced
PAH emissions.

One of the contributory factors to rubber fumes is the low-MW compounds released
from the base polymer at high temperature. Sakdapipanich and co-workers [53] and
Hoven and co-workers [54] looked into the volatile components released from NR.
These scientists from Mahidol University and Chulalongkorn University, respectively,
used headspace GC–MS to identify the volatile compounds emitted by seven grades of
raw NR when they were heated at 60 °C for 2 h. A large range of chemical compounds
was detected and these were generated by a number of reactions, for example, esters by
microbial esterification and sulfur-containing compounds by amino acid degradation.
The types of volatile compounds detected were found to vary according to the type of
NR. For example, the sulfur-containing compounds found in the skim crumb rubber,
which are thought to be formed due to amino acid degradation, were not found in
the other six types of NR. These results suggested that a wide range of compounds
was present in the emissions and that the types and levels were dependent upon the
grade of NR used in the rubber formulation.

Aarts and Davies [55] reviewed the published information on the presence of harmful
components in fumes produced during the vulcanisation of rubber compounds and the
conclusions that have been reached regarding the mechanisms associated with their
formation. They also reported on the results that they produced from the analysis of
fumes produced during the curing of four rubber compounds – two NR-based and
two oil-extended-SBR based – in a cure simulation apparatus that enabled continuous
sampling to take place. The total amount of condensate within the system and total
volatiles produced by the four compounds were reported, with a high proportion
of the condensate being due to the oil extender in the case of the SBR compounds.
Major contributory components of both the condensate and volatiles were found to
be the compounding ingredient 6PPD, and the breakdown products BT and ‘parent’
amines from the accelerators.

227
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

After reviewing various systems for analysing vulcanisation fumes, Schuster and co-
workers [56] selected two reproducible and quantitative methods for the analysis of
the vulcanisation fumes of rubber mixtures using laboratory curing equipment. Among
the results presented were analytical results obtained on unvulcanised rubbers, with
over 40 volatile compounds being detected in the vapours emitted by the NR, SMR
CV 50, and over 45 from the SBR rubber, Buna 1712, at 180 °C. The objective was
to show the contribution that base polymers make to vulcanisation fumes. The species
detected in the fumes of other rubbers (e.g., EPDM and CR) were also discussed. Some
of the data presented by Schuster were also published by Linde in a dissertation [57].

A collaborative presentation by Mai (LRCC in France) and Giese (DIK in Germany)

[28] at the International Rubber Conference in 2013 in Paris described the results
obtained by analysing rubber fumes in the laboratory using a dynamic headspace
GC–MS method. To carry out the work, an EPDM compound was prepared and
heated to 200 °C in a TD instrument to generate fumes. Examples of the reaction
products generated by the accelerators in the compound and detected by the GC–MS
are listed below:

• Benzylisothiocyanate and N-benzylidenebenzylamine from zinc

diethyldithiocarbamate. BT and methylthiobenzothiazole from
2-mercaptobenzothiazole.

• Tetramethylthiourea from the TMTD.

In order to obtain a link between the substances detected in rubber fumes and
ingredients present in a compound, Mai and Giese used a methodology that involved
the successive addition of ingredients to the base polymer and analysed the compounds
produced using the TD GC–MS system. This study also included data that had been
obtained on rubber fumes produced by EPDM, NR and NBR compounds using the
ISO standard for rubber fumes analysis, ISO/TS 17796:2013 [6] (Section 7.5.1).
In addition, it included information on the sampling options for the collection and
identification of rubber fumes in factories and the recovery rates for certain types
of organic substances from different types of collection media e.g., (sorbents, filters
and sampling bags) (Table 7.1).

7.6 Conclusions

Rubber fumes are an extremely complex mix of chemical substances (e.g., over 30
from a single rubber compound) that have a wide range of possible sources and
origins. In addition, the physical form (i.e., gas, vapour or aerosol) that these species
exist in the factory environment can be influenced by the type of process carried out

228
 Characterisation of Rubber Process Fume

and the process temperature. The type of process and process temperature can also
influence the reactions that take place in a rubber compound and, hence, the types of
species that can be emitted and the proportion of the rubber compound which falls
within the volatile fraction and so has the potential to be released as process fumes.

There are three distinct sources of species in rubber fumes:

• Volatile ingredients (e.g., antioxidants and plasticisers).

• Volatile impurities in ingredients (e.g., residual monomers and manufacturing

impurities).

• Products from chemical reactions that occur during processing (e.g., vulcanisation
and antioxidant activity).

The relative contribution of the above to any given sample of rubber fumes is
dependent on the ingredients present in each rubber formulation.

The information published demonstrates that rubber fumes are not a single chemical
entity, but a highly complex mixture of chemical species that varies depending upon
the factors described above. Hence, the analytical methods that evaluate only the
aerosol portion of rubber fumes (i.e., the visible fraction), like MDHS 47/2:1999 [8],
and provide no information on the gas and vapour portion (i.e., invisible fraction) are
unsuitable for evaluation of the full chemical composition of the fumes. It is for this
reason that combinations of sampling media (e.g., different types of adsorbent tubes,
impingers, filters) have had to be used in order to efficiently capture rubber fumes in
factory atmospheres and enable representative compositional data to be obtained.

The studies that have been carried out have shown that when only the composition
of a rubber compound is known, it may be difficult to compile an accurate predictive
list of the species that are likely to be present in its process fumes because there are
too many variables that influence their composition. It is necessary to consider the
process conditions because these could cause the generation of specific species.

It is possible to have reasonable knowledge of the chemical composition of rubber

fumes by adopting a case-by-case strategy based on three main steps.

1. Knowledge of the chemical composition of the rubber compound(s), the process

technology in use and the associated process parameters – see ISO definition for
rubber fumes in ISO/TS 17796:2013 [6] in Section 7.1.

2. Evaluation of the visible and invisible fractions of the fumes using qualitative
methods.

229
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

3. Quantification evaluation (e.g., by GC–MS) of the components whose presence

has been detected using qualitative methods.

Such an approach has been applied to improve knowledge of the chemical composition
of rubber fumes and to discontinue use of dangerous substances in the production
processes. The recent introduction of ISO/TS 17796:2013 [6] should assist with this
work.

References

1. M.J. Forrest and B.G. Willoughby in Proceedings of RubberChem 2004,

9–10th November Birmingham, UK, Rapra Technology Ltd, Shawbury, UK,
2004, Paper No.1.

2. R.B. Simpson in Rubber Basics, Rapra Technology Limited, Shawbury, UK,

2002.

3. J.F. Funt in Mixing of Rubber, Smithers Rapra Technology Limited,

Shawbury, UK, 2009.

4. V.A. Shershnev, Rubber Chemistry and Technology, 1982, 55, 537.

5. S. Majumdar in Rubber Vulcanisation Processes: An Overview, Chemical

Weekly, 26th August 2008, No.2, p.211.

6. ISO/TS 17796:2013 – Rubber – Trapping and identification of volatile

components of rubber fumes with active sampling on a poly(2,6-
diphenylphenylene oxide) type sorbent, using TD and gas chromatography
method with mass spectrometric detection.

7. ISO/TS 21522:2017 – Rubber process fumes components – Quantitative test

methods.

8. MDHS 47/2:1999 – Determination of rubber process dust and rubber fumes

(measured as cyclohexane-soluble material) in air.

9. M.J. Forrest, Progress in Rubber, Plastics and Recycling Technology, 2015,

31, 4, 219.

10. ISO/DTR 21275:2017 – Rubber – Comprehensive review of the composition

and nature of process fumes in the rubber industry.

230
 Characterisation of Rubber Process Fume

11. N.M. Levin in Proceedings of the 144th International Rubber Conference ‘93,
26–29th October, Orlando, FL, USA, American Chemical Society, Rubber
Division, Akron, OH, USA, 1993, Paper No.11.

12. J. Asplund, Kautschuk, Gummi, Kunststoffe, 1995, 48, 4, 276.

13. B.G. Willoughby in Proceedings of Hazards in the Rubber Industry,

28–29th September, Rapra Technology Ltd, Shawbury, UK, 1999, Paper No.7.

14. F. Pilati, S. Masoni and C. Berti, Polymer Communications, 1985, 26, 280.

15. J. Pospisil in Aromatic Amine Antidegradants, Developments in Polymer

Stabilisation – 7, Elsevier Applied Science, London, UK, 1984.

16. M.M. Donskaya, S.M. Kavun, A.V. Krokhin, V.G. Frolokova and
Y.A. Khazanova, International Polymer Science and Technology, 1994, 21, 3,
T/38.

17. C.R Blanden and S.A. Isherwood in Proceedings of Health and Safety in the
Plastics & Rubber Industries, 15–16th September, York, UK, 1987, p.3/1–
3/10.

18. A.A. Dost, D. Redman and G. Cox in Proceedings of International Rubber

Exhibition and Conference, 7–10th June, Manchester, UK, 1999, Paper No.5.

19. B.G Willoughby in Rubber Technology – 4, Eds., A. Whelan and K.S. Lee,
Elsevier Applied Science, Amsterdam, The Netherlands, 1987, p.253.

20. L.J. Maisey in Proceedings of SRC-81 Conference, Helsinki, Finland,

21–22nd May, 1981, p.215.

21. B.G. Willoughby in Proceedings of Health and Safety in the Plastics and
Rubber Industries, Warwick, UK, 29th September– 1st October, 1980,
Paper No.12.

22. J. A. Worwood in Proceedings of the 3rd International Conference on Health

and Safety in the Plastics & Rubber Industry, 15–16th September, York, UK,
1987, p.4/1–4/9.

23. F. De Vocht, K. Straif, N. Szeszenia-Dabrowska, L. Hagmar, T. Sorahan,

I. Burstyn, R. Vermeulen and H. Kromhout, Annals of Occupational Hygiene,
2005, 49, 8, 691.

231
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

24. H. Khalfoune in Proceedings of International Rubber Conference 2013, Paris,

France, 20–22nd March, 2013, Paper No.54.

25. U. Giese in Proceedings of Hazards in the European Rubber Industry,

28–29th September, Manchester, UK, 1999, Paper No.9.

26. P. Kuhn-Stoffers, U. Giese, R.H. Schuster and G. Wunsch, Kautschuk,

Gummi, Kunststoffe, 1997, 50, 5, 380.

27. T. Will and U. Giese, Kautschuk, Gummi, Kunststoffe, 1996, 49, 3, 200.

28. Le Huy Mai and U. Giese in Proceedings of International Rubber Conference

2013, Paris, France, 20–22th March, 2013, Paper No.74.

29. Internal Report from the Nordic Rubber Industry 1991, Summary Kautsch,
Gummi, Kunststoffe, 1993, 46, 858.

30. T. Rogszewska and co-workers, Polymer Journal Occupational Medicine,

1989, 2, 4, 366.

31. R.H. Schuster, F. Nabholz and M. Gmünder, Kautschuk, Gummi, Kunststoffe,

1990, 43, 2, 95.

32. F. De Vocht, R. Vermeulen, I. Burstyn, W. Sobala, A. Dost, D. Taeger,

U. Bergendorf, K. Straif, P. Swuste and H. Kromhout, Occupational and
Environmental Medicine, 2008, 65, 6, 384.

33. H. Kromhout, P. Swuste and J.S.M. Boleij, Annals of Occupational Hygiene,

1994, 38, 1, 3.

34. V.O Frolikova, M.M. Donskaya, L.I. Yalovaya, A.M. Pichugin and
I.I. Vishnyakov, International Polymer Science and Technology, 2009, 36, 8,
29.

35. V.O. Frolikova, M.M. Donskaya, L.I. Yalovaya, A.M. Pichugin and
I.I. Vishnyakov, Kauchuk I Rezina, 2008, 5, 20.

36. Z.N. Nudel’man, International Polymer Science and Technology, 2001, 28,
2, T/38.

37. V. Cocheo, M.L. Bellomo and G.G. Bombi, American Industrial Hygiene
Association Journal, 1983, 44, 7, 521.

38. M.I. Novokovskaya, V.B. Saltanova, K. Yu and Shaposhnikov,

Kauchuk I Rezina, 1976, 6, 48.

232
 Characterisation of Rubber Process Fume

39. B.G. Willoughby in Rubber Fume: Ingredient/Emission Relationships,

Rapra Technology Ltd, Shawbury, UK, 1994, p.105.

40. K.G. Ashness, G. Lawson, R.E, Wetton and B.G. Willoughby, Plastics Rubber
Processing Applications, 1984, 4, 69.

41. B.G. Willoughby and K.W. Scott, Rubber Chemistry and Technology, 1998,
71, 310.

42. B. Willoughby in Proceedings of the 8th Scandinavian Rubber Conference,

10–12th June, Copenhagen, Denmark, 1985, p.593.

43. R.W.B Smith and B.G Willoughby in Proceedings of the 3rd Health and Safety
in the Plastics and Rubber Industries, 15–16th September, York, UK, 1987,
p.4/1–4/9.

44. D. Becklin, T. Herman, S. Ponto and B. Rozynov, Rubber and Plastics News,
1995, 25, 2, 15/8.

45. D. Becklin, T. Herman, S. Ponto and B. Rozynov in Proceedings of the

147th ACS Spring Meeting, 2–5th May, American Chemical Society, Rubber
Division, Philadelphia, PA, USA, 1995, Paper No.31, p.14.

46. B.V Rozynov, R.J. Liukkonen, D.O. Becklin, A.L. Noreen and S.D. Ponto,
ACS Polymer Preprints, American Chemical Society, Division of Polymer
Chemistry, 2000, 41, 1, 692.

47. J. Zietlow and R.H. Schuster in Proceedings of the 144th International

Rubber Conference ‘93, 26–29th October, American Chemical Society, Rubber
Division, Orlando, Fl, USA, 1993, Paper No.10, p.8.

48. G. Lucas and U Giese, Kautschuk, Gummi, Kunststoffe, 2008, 61, 4, 180.

49. Z.N. Nudel’man and V.L. Antonovskii, International Polymer Science and
Technology, 1994, 21, 1, p.T/52.

50. Z.N. Nudel’man and V.L. Antonovskii, Kauchuk I Rezina, 1993, 6, 14.

51. Y. G. Chikishev, N.A. Klyuev, G.A. Vakhtberg and V.G. Zhil’Nikov, Kauchuk
I Rezina, 1975, 12, 42.

52. R. Wommelsdorff, U. Giese, C. Thomas and A. Hill, Kautschuk, Gummi,

Kunststoffe, 1994, 47, 8, 549.

233
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

53. J. Sakdapipanich and K. Insom, Kautschuk, Gummi, Kunststoffe, 2006, 59,

7–8, 382.

54. V.P. Hoven, K. Rattanakaran and Y. Tanaka, Rubber Chemistry and

Technology, 2003, 76, 5, 1128.

55. A. J. Aarts and K.M Davies in Proceedings of Rubbercon 92 – A Vision for

Europe, 15–19th June, Plastics and Rubber Institute, Brighton, UK, 1992,
p.455.

56. R.H. Schuster, H. Linde H and G. Wuensch, Kautschuk, Gummi, Kunststoffe,

1991, 44, 3, 222.

57. H.Linde in Untersuchung von bei der Vulkanisation enstehenden Gasen und
Dampfen, Universitat Hannover, Germany, 1990. [Dissertation]

234
8
Compliance with Food-contact Regulations

8.1 Introduction

For rubber products that are intended for use in certain sectors of industry, there are
regulations and guidelines that they must comply with before they can be placed on
the market. Three of the most highly regulated of these sectors are:

• Food-contact materials and articles

• Pharmaceutical products

• Medical devices

This chapter describes the tests that rubber products undergo to ensure that they pass
food-contact regulations. Chapter 9 addresses the analytical work carried out during
the conduct of extractables and leachables (E&L) studies aimed at addressing the
requirements of the pharmaceutical and medical sectors.

Regulations that govern the use of rubber products for food-contact applications in
Europe and the United States (US) have been in existence for many years; for example,
the German BfR Recommendations were first published in the 1950s. Over the last
30 years, the number of countries and regions that have published their own food-
contact regulations have grown to the extent that few developed and emerging nations
do not have them. Information on the food-contact regulations that apply to rubber
products in the European Union (EU), US and other parts of the world is available
in publications such as the ones edited by Rijk and Veraart [1] and Baughan [2].

One feature common to all regulations, and most guidance documents, is an inventory
list, sometimes called a ‘positive list’, of substances that are allowed to be used in
the manufacture of the food-contact material. A recent comparison of a number of
regulatory or non-regulatory inventory lists for food-contact materials was published
by Geueke and co-workers [3]. The authors estimated that >6,000 substances appear
on these lists and stated that some have been linked to chronic diseases and others
lack sufficient toxicological evaluation. The aim of their study was the identification
of known food-contact substances that were also considered to be ‘chemicals of
concern’ and the following three inventory lists were chosen for the investigation:

235
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• The 2013 Pew Charitable Trusts database of direct and indirect food additives
used legally in the US.

• The 2014 EU list for plastic food-contact materials and articles; the ‘Union List’
provided in Annex 1 of Regulation (EU) 10/2011 and updated by subsequent
amendments.

• The 2011 non-plastics food-contact substances database published by the

European Food Safety Authority.

The conclusion of the study was that 175 chemicals of concern were identified in the
three lists and information about their applications, regulatory status and potential
hazards was provided in the article.

In Section 9.2, a description is provided of the potential origins for the substances,
both added intentionally (i.e., ingredients) and non-intentionally (e.g., reaction
products) that have the potential to migrate from rubber products and, hence, should
be addressed by E&L study programmes. The substances that have the potential to
migrate into food from rubber products have the same origins and so the information
in this description is also very relevant to this chapter. One important difference
between an E&L study and food migration work, however, is that more direction is
provided to the analyst, via the regulations, of the specific substances that need to
be identified and quantified. For readers interested in obtaining more information on
how these regulations compare with one another, an article discussing the parallels
and differences between the food, medical and pharmaceutical regulations has been
published by Creese [4].

A review of the regulations that apply to rubbers intended for food-contact

applications has been published by Forrest [5]. In addition to providing information
on the regulations, this publication also provides information on the testing that has
to be carried out to meet them. It also presents the information generated by the work
carried out during the course of four research projects for the UK Food Standards
Agency (FSA) [6–9].

A copy of the final report for all of these projects can be obtained via the FSA website:
http://www.foodstandards.gov.uk.

A review of the testing of rubber products (e.g., seals, gaskets, hoses) for food-contact
applications such as transportation, pipe work, pumps, storage vessels, handling,
preparation and manufacturing, has been provided by Isa [10]. This review includes
the testing of rubber products for the migration of additives (e.g., plasticisers, process
oils) according to national and international standards and regulations.

236
 Compliance with Food-contact Regulations

8.2 Rubbers for Food-contact Applications

Many types of rubber can be used to manufacture food-contact materials and articles.
These include:

• Natural rubber (NR).

• Nitrile rubber (NBR).

• Ethylene propylene rubbers [i.e., ethylene propylene monomer and ethylene

propylene diene monomer (EPDM)].

• Fluorocarbon rubber (FR).

• Silicone rubber.

• Thermoplastic elastomers (TPE).

• Others – including butyl rubber, polychloroprene (CR), acrylic and hydrin.

There is also a wide range of products that are manufactured from these rubber
compounds. These products make contact with food in a variety of end-use conditions
(i.e., contact times, areas and temperatures).

Although rubber products can be used to a limited extent for the packaging of food
(e.g., seals within bottle closures), and cooking of food, they mainly come into contact
with food during its manufacture, processing, storage and transportation. Information
on the rubber products that make contact with food (e.g., types of rubber compounds
and conditions of contact) was collected by Rapra Technology during the course of a
FSA project [6]. Some of the information collected during an industry survey is shown
in Tables 8.1 and 8.2. Specifically, Table 8.1 shows examples of rubber products
and components that make contact with food and Table 8.2 provides representative
examples of typical applications for food-contact rubbers.

With regards to the analysis of the low-molecular-weight (MW) substances in rubbers

that have the potential to migrate into food, the development of new instrumentation
has provided the rubber analyst with much greater capability. A prime example is
the fact that the past 15 years have seen the proliferation of liquid chromatography
(LC)–mass spectrometry (MS) instruments to the extent that they have now
essentially replaced conventional high-performance liquid chromatography (HPLC)
instruments in the majority of laboratories. The LC–MS technique compliments gas
chromatography (GC)–MS and enables the analyst, for the first time, to routinely
generate data on both the thermally-labile and thermally-stable compounds present
in rubber products, food and food simulants up to and beyond the absorption limit
of the gastrointestinal tract (1,000 Da).

237
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 8.1 Examples of rubber products and components that contact food
Location/Usage Examples of products and components
Food transportation Conveyor belts, hoses, rubber skirting and rubber paddle lips
Pipe work components Seals, gaskets, flexible connectors and butterfly valves
Pumps Progressive cavity pump stators, diaphragm pumps
Plate heat exchangers Gaskets
Machinery/storage vessels General seals and gaskets
Cans/bottles Bottle seals and can seals
Food handling/preparation Gloves and feather pluckers
Food manufacturing Silicone sweet moulds, rubber squeeze rollers
Food wrapping Meat and poultry nets
Reproduced with permission from M.J. Forrest in Food Contact Materials – Rubbers, Silicones,
Coatings and Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.15. ©2009,
Smithers Rapra Technology Ltd [11]

Table 8.2 Representative contact conditions for typical applications of food-

contact rubbers
Component Contact area Contact area Typical Maximum Contact
– individual – component contact contact temperature (oC)
components assemblies time time General Extreme
(cm3) (cm3)
Food transportation Up to 50,000
1
– <1 h – <85 –
Pipework <1,000 <10,000 <1 h 2 weeks2 <140 2503
components
Pumps <10,000 – <1 h 2 weeks2 <85 –
Plate heat exchanger <1,000 Up to 56,000 <3 min – <140 2503
gaskets
General seals and <1,000 Up to 30,000 <1 h 12 weeks2 <85 –
gaskets
Packaging seals and <1,000 – Up to 5 – Ambient 1404
closures years
Miscellaneous <1,000 – <1 h 4 weeks5 <85 2506
1. Potential contact areas could be up to 1,500,000 cm for some of the longer conveyor belts
3

2. May have extended contact times during shutdowns and beer-keg seals may have contact
times of up to 12 weeks
3. Refining of vegetable oil
4. During pasteurisation/sterilisation
5. Meat and poultry nets
6. Meat and poultry nets if not removed before cooking
Reproduced with permission from M.J. Forrest in Food Contact Materials – Rubbers, Silicones,
Coatings and Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.17. ©2009,
Smithers Rapra Technology Ltd [12]

238
 Compliance with Food-contact Regulations

In addition, development work continues to advance analytical instrumentation so

that there are improvements in important areas such as MW range, detection limits,
software-assisted peak deconvolution, analysis speed, accuracy of library searching,
and species selectivity. The advent of multi-hyphenated techniques such as GC×GC–
time-of-flight (ToF)–MS and LC–MS/MS are examples of this. These instruments, with
their greater resolving power and selectivity, are also improving the direct analysis
of food products. The large range of low-MW compounds in these food products
have caused problems in the past and caused workers to rely quite heavily on food
simulants.

These developments in instrumentation, together with the research work being

carried out on rubbers (e.g., the FSA projects mentioned above and in Section 8.1),
has enabled more accurate conformity checks to be performed on compounds, as
well as continuing to add to the understanding of the migration behaviour of rubber-
related substances.

Food-contact materials are a very heavily regulated sector with, as mentioned

in Section 8.1, all of the developed countries in the world having food-contact
regulations. To set the scene, Section 8.3 provides a brief overview of the regulations
for food-contact rubber products in Europe and the US.

8.3 Overview of the European Union and United States Food and
Drug Administration Regulations

8.3.1 European Union Legislation

At the moment, there is no harmonised EU legislation for rubber food-contact

materials and articles. There is only a specific regulation published in 1993 concerning
nitrosamines in baby’s dummies [13]. However, all rubber materials and articles need
to comply with the EU Framework Regulation (EC) 1935/2004 and the EU Good
Manufacturing Practice Regulation (EC) 2023/2006 so that in normal use they will
not transfer their constituents to food in quantities that could endanger health or cause
unacceptable changes in the composition of food or deterioration in its organoleptic
properties (i.e., taste, texture, aroma, or appearance).

8.3.2 German Regulations

Within Germany, the food-contact legislation for natural and synthetic rubbers is

239
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

described in BfR Recommendation XXI ‘Commodity Articles Based on Natural and

Synthetic Rubber’. There are separate requirements for silicone rubbers and these
are contained within BfR Recommendation XV.

In the case of Recommendation XXI, four usage categories and a special category
are defined as follows:

• Category 1 (test conditions: 10 days at 40 °C): Rubber articles which come into
contact with food for periods of >24 h to several months (e.g., storage containers,
container linings, seals for cans and bottles).

• Category 2 (test conditions: 24 h at 40 °C): Rubber articles which come into

contact with food for <24 h (e.g., food conveying belts, tubes and hoses, sealing
rings for cooking pots, lock seals for milk can lids).

• Category 3 (test conditions: 10 min at 40 °C): Rubber articles which come into
contact with food for <10 min (e.g., milk liners and milking machine tubes, roller
coatings and conveyor belts (fatty foods only in both cases), gloves and aprons
for food handling).

• Category 4 (no migration testing required): Rubber articles which are used only
under conditions where no migration into food is to be expected (i.e., if the
articles come into contact with the food for a very short time or only over a very
small area). Examples of rubber products in this category include: conveyor belts,
suction and pressure hosing for moving and loading/unloading dried food, and
pump parts.

• Special Category (test conditions: 24 h at 40 °C): Rubber articles directly associated

with the consumption of food and which are being, or can expected to be, taken
into the mouth (e.g., toys according to BfR Recommendation XLVII, teats,
soothers, gum shields, balloons).

The following food simulants are used in connection with the German regulations:
distilled water, 10% ethanol and 3% acetic acid. There is no simulant stated in the
BfR Recommendation XXI for fatty food, although the document mentions the
possibility of one being available in the future. The permissible overall migration
limits (OML) vary according to the category and simulant (Table 8.3).

To carry out the overall migration test to the requirements of BfR Recommendation
XXI, test pieces of 50 × 50 mm to give a total area of 50 cm2 (both surfaces) are
immersed in 100 ml of the appropriate simulant for the intended end use, the test
performed and then the simulant dried down quantitatively.

240
 Compliance with Food-contact Regulations

Table 8.3 OML for BfR Recommendation XXI

Food simulant Category and OML (mg/dm2)
1 2 3 Special
Distilled water 50 20 10 10 or 50#
10% Ethanol 50 20 10 –
3% Acetic acid 150 (50) 100 (20) 50 (10) –
(value) = permissible organics within total
#dependent on product type
NB: there is no migration limit for category 4 (see above)
Reproduced with permission from M.J. Forrest in Food Contact Materials – Rubbers,
Silicones, Coatings and Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.28.
©2009, Smithers Rapra Technology Ltd [14]

The BfR regulations also include specific compositional limits and specific migration
limits (SML) for a number of substances, including:

1. N-Nitrosamines and nitrosatable substances

2. Amines (all categories)

3. Milking liners and milking tubes

4. Formaldehyde

5. Acrylonitrile (ACN)

6. Zinc dibenzyldithiocarbamate

In addition to migration testing and compositional requirements, BfR Recommendation

XX1 also contains inventory lists for the various end-use categories, and some of the
substances within these lists have restrictions concerning their maximum level within
a rubber compound. Pysklo and co-workers [15] compared the BfR inventory list of
ingredients approved for food-contact rubbers with the equivalent Polish list. The
latter was originally based on the German list and this exercise showed that it was
in need of updating.

The BfR Recommendation XV for Silicones covers silicone oils, silicone resins and
silicone rubbers. The chapter on silicone rubbers has an inventory list stipulating
acceptable starting materials and the additives that may be used in processing and
manufacture; some of the latter having restrictions on their maximum levels.

241
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Separate restrictions are stated in Recommendation XV, where silicone rubber is to

be used for teats, dummies, nipple caps, teething rings or dental guards. Dummies
and bottle teats must also comply with the requirements laid down in the German
Commodities Regulation (Bedarfsgegenständeverordnung).

The amount of volatile organic material in the finished silicone rubber product is
restricted to a maximum of 0.5%, as is the total extractable material. Test methods are
referenced in Recommendation XV for these determinations as well as a test that needs
to be carried out for residual, un-decomposed peroxides, which should be negative.

8.3.3 Requirements in France

French requirements for food-contact elastomers (excluding silicones) are given in

the Arrete of 9th November 1994, which was published in the Journal Officiel de la
République Française, 2nd December 1994 (pages 17029–17036). Four usage categories
(A to D) and a special category (designated T) are described in this document together
with an inventory list detailing permitted ingredients in each category. There is an
OML set at 10 mg/dm2 (60 mg/kg), the same as for plastics, and specific restrictions
also apply, such as an SML for primary and secondary aromatic amines of <1 mg/kg.

8.3.4 Requirements in the United Kingdom

The UK legislation on food-contact materials is published as a number of Statutory

Instruments. The use of rubbers in contact with food is covered by the legislation
included in Statutory Instrument 1987 No.1523 ‘Materials and Articles in Contact
With Foodstuffs’. This states that any food-contact material should not be injurious
to the health of the consumer and that any contamination should not have an adverse
effect on the organoleptic properties of the food. The absence of any inventory list for
compounding ingredients means that UK rubber compounders usually refer to either
the US Food and Drugs Administration (FDA) or the BfR regulations depending on
the market to be addressed.

There are separate rules in the UK for the use of rubber in contact with potable
water. These are given in the UK water-fitting bylaws scheme and include tests for
the following:

• Taste

• Appearance

• Growth of aquatic micro-organisms

242
 Compliance with Food-contact Regulations

• Migration of substances that may be of concern to public health

• Migration of toxic metals

The test methods for the above are given in the British Standard, BS 6920, Parts 1 to 4.

8.3.5 United States Food and Drug Administration Regulations

In the US, the FDA produces a guidance for industry document entitled ‘Preparation of
Food Contact Notifications and Food Additive Petitions for Food Contact Substances:
Chemistry Recommendations’. This is in addition to the Code of Federal Regulations
(CFR) Title 21, Parts 170 to 199 Food and Drugs, which contains the FDA food-
contact regulations. Copies of both the guidance documents and the regulations can
be assessed within the appropriate sections of the FDA website: http://www.fda.gov.

The CFR is published annually on 1st April and the main requirements for rubber
products for use with food are covered in Part 170, specifically:

• Rubber articles intended for repeat use: Part 177.2600

• Closures with sealing gaskets for food containers: Part 177.1210.

Part 177.2600 is the principal section that needs to be addressed and which covers
the majority of rubber compounds and products. In addition to the sections that
need to be considered when compounding a food-contact rubber, there are other
sections that are sometimes applicable, such as Parts 180.22 and 181.32, that place
a restriction (i.e., 0.003 mg/inch2) on the amount of ACN monomer that can migrate
from NBR rubbers and blends that contain NBR rubber (e.g., polyvinyl chloride
(PVC)/NBR blends).

In addition to listing acceptable compounding ingredients in these two ‘rubber

sections’ of Part 177, the FDA regulations also allow the use in rubbers of ‘prior
sanctioned ingredients’ (Part 181), additives that are generally recognised as safe
(Parts 184 to 186) and, providing that the specified restrictions allow, substances
listed in other Parts. Hence, it is the case with the FDA regulations that not all the
ingredients that can be used in food-contact rubber products are to be found in the
same place, and compounders and assessors may need to address a number of Parts
in CFR Title 21. For example, relevant substances listed in Part 182 include zinc
oxide (ZnO) (182.8991), zinc stearate (182.8994) and calcium silicate (182.2227),
substances listed in Part 184 include calcium carbonate, calcium stearate and
calcium oxide, and those in Part 186 include kaolin clay (186.1256) and iron oxide
(186.1374). With respect to exclusions, the FDA regulations specifically prohibit

243
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

the use of the following, relatively common, compounding ingredients in food-

contact rubbers:

• Part 189.220 – polymerised 1,2-dihydro-2,2,4-trimethylquinoline

• Part 189.250 – mercaptoimidazoline and 2-mercaptoimidazoline

A review and comparison of the FDA and BfR regulations (Section 8.3.2), with
particular emphasis on the inventory lists of approved ingredients, has been carried
out by Pysklo [16]. A schedule of the commercial rubber compounding ingredients
which meet the requirements of the two regulations is provided in the article.

The migration testing that needs to be carried out on rubber products is described
in Part 177.2600. Test pieces are cut from the rubber test product to provide a
known surface area (cut edges are included in the calculation) and immersed in an
appropriate amount (e.g., 100 ml) of food simulant, either hexane for fatty food-
contact applications or distilled water for aqueous food-contact applications. The
samples are refluxed for 7 h in pre-cleaned glassware and then removed and placed
into fresh simulant and refluxed for a further 2 h. The test pieces are then removed
and both the 7- and 2-h test portions evaporated separately to dryness in conditioned
crucibles and the residues weighed. Blank determinations on equivalent volumes of
the food simulant used are also performed. In order to be acceptable for food use,
the rubber has to pass the requirements shown in Table 8.4.

Table 8.4 FDA Part 177.2600 migration limits for repeat-use articles
Food type First 7 h Succeeding 2 h
Fatty foods – hexane extractables under reflux 175 mg inch –2
4 mg inch–2
Aqueous foods – distilled water extractables under reflux 20 mg inch–2 1 mg inch–2
Reproduced with permission from M.J. Forrest in Food Contact Materials – Rubbers,
Silicones, Coatings and Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.22.
©2009, Smithers Rapra Technology Ltd [17]

To summarise, providing that the ingredients in a rubber product are listed within
the applicable parts of CFR Title 21 as being approved, the water (for aqueous food
use) and hexane (for fatty food use) extractables are within the prescribed limits, and
any applicable restrictions for its composition (e.g., migration of residual monomer)
or end-use conditions [e.g., food type(s), contact time and temperature] are met, then
the product is considered suitable for food use.

244
 Compliance with Food-contact Regulations

8.3.6 Council of Europe Resolutions

The Council of Europe (CoE) is an international organisation with over 45 member

countries that has a committee of Experts on Materials and Articles Coming into
Contact with Food that meets under the auspices of the Partial Agreement in the Social
and Public Health Field. Over the period 1989 to 2005, this committee published
nine resolutions concerning a wide range of food-contact materials and articles, from
‘Colourants’ [Resolution AP(89)1] to ‘Cork Stoppers and Other Cork Materials’
[AP(2004)2]. The resolutions that are applicable to this section, the ‘CoE Resolution
on Rubber Products’ [AP(2004)4] and the ‘CoE Resolution on Silicone Products’
[AP(2004)5], were published in 2004.

Since 2008, there has been a change within the CoE, and its activities relating to food-
contact materials have been transferred to the European Directorate for the Quality
of Medicines & HealthCare (EDQM). The EDQM is a Directorate of the CoE and
is an organisation that protects public health in the areas of food, medicines, and
organ transplants. It has its own laboratory and publishes well-respect standards,
such as the European Pharmacopoeia, which is legally binding within the member
states of the CoE.

Once adopted, the food-contact material resolutions and supporting documents

drawn up by the CoE are not legally binding, but member states, of which the UK is
one, are expected to take note of them.

8.3.6.1 CoE Resolution on Rubber Products

The ‘CoE Rubber Resolution’ [AP(2004)4] on food-contact elastomers has an

inventory list of additives within it and a small section that deals with breakdown
products – nitrosamines and amines. The inventory list is described as Technical
Document No.1 – List of substances to be used in the manufacture of rubber products
intended to come into contact with foodstuffs. An FSA project [8] was commissioned
to study the breakdown and reaction products from the curatives and antidegradants
present in this inventory list.

In addition to Technical Document No.1, there are four other documents associated
with this CoE Resolution:

• Technical Document No.2: Guidelines concerning the manufacture of rubber

products intended to come into contact with foodstuffs.

• Technical Document No.3: Good manufacturing practices of rubber products

intended to come into contact with foodstuffs.

245
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Technical Document No.4: Test conditions and methods of analysis for rubber
products intended to come into contact with foodstuffs.

• Technical Document No.5: Practical guide for users of Resolution AP(2004)4 on

rubber products intended to come into contact with foodstuffs.

The resolution places rubber products into one of three categories:

• Category I comprises the following rubber products for which migration testing
is required:

• Feeding teats.

• Rubber products to come into contact with baby food, for which the R-total
is ≥0.001 (a definition of R-total is given below).

• Category II comprises rubber products for which R-total is ≥0.001 and for which
migration testing is required.

• Category III comprises rubber products for which R-total is <0.001 and for
which migration testing is not required, except for rubber products containing
nitrosamines, nitrosatable substances or aromatic amines and Category III
substances having an SML in Technical Document No.1.

These three categories take into account the wide variety of applications for which
rubber products are used and the fact that migration may vary with the application.
The level of migration for rubber products may be estimated by taking into account
four factors: R1, R2, R3 and R4. Categories are based on the intended use or on the
result of the multiplication of the four factors (R1 × R2 × R3 × R4 = R total).

The factors R1, R2, R3 and R4 can be defined as follows:

‘R1 refers to the relative contact area (AR) between rubber products and food or
beverage, expressed in cm2 of rubber surface per kg of food or beverage. For a relative
area smaller or equal to 100 cm2/kg foodstuffs, R1 has a value calculated according
to the formula: R1 = AR/100. For a relative surface larger than 100 cm2/kg, R1 always
has the value of 1.00.

R2 refers to the temperature during the contact period of the rubber product with the
food or beverage. At a temperature lower than or equal to 130 °C, R2 has a value
calculated according to the formula: R2 = 0.05e0.023T. Where ‘e’ is the base of the
natural or Napierian logarithms and T is the contact temperature, expressed in °C.
For temperatures higher than 130 °C, R2 always has the value 1.00.

R3 refers to the time, t, expressed in hours, during which a rubber product is in contact

246
 Compliance with Food-contact Regulations

with the food or beverage. For a contact time shorter than or equal to 10 hours, R3
has a value calculated according to the formula: R3 = t/10. For a contact time of more
than 10 hours, R3 has the value 1.00.

R4 refers to the number of times, N, that one and the same rubber product, or part of
that rubber product comes into recurrent contact with a quantity of food or beverage.
If the number of contact times is greater than 1000, then R4 is calculated according
to the formula: 10log R4 = 6 – 2 10log N. If the number of contact times is smaller
than or equal to 1000, then R4 always has the value 1.00.’

The resolution also states that, among other things:

1. Rubber products in Categories I and II should not transfer their constituents to

foodstuffs or food simulants in total quantities >60 mg/kg food or food simulant
(i.e., an OML).

2. Rubber products of Categories I and II should comply with the restrictions laid
down in Technical Document No 1. In addition these rubber products should
comply with the requirements set out in Table 1 of the resolution, except rubber
teats, which should comply with EU Directive 93/11/EEC.

3. Rubber products intended for repeated use should be subjected to tests according
to EU Directive 2002/72/EC Annex 1 [Note: this EU Directive has now been
replaced by EU Regulation (EU) 10/2001].

4. Rubber products belonging to Category III do not require migration testing, unless
otherwise specified.

5. Verification of compliance with the quantitative restrictions should be carried

out according to the requirements laid down in Technical Document No.2. This
document provides more detail regarding the following:

a) Further definitions and data surrounding the ‘R’ values.

b) E
 xamples of rubber products that fall into Categories II and III and the
calculations applied to them.

c) Migration tests.

8.3.6.2 CoE Resolution on Silicone Materials

There is a separate CoE Resolution, designated AP(2004)5, which addresses silicone

materials for food contact. The resolution defines the silicone product group
comprising silicone rubbers, silicone liquids, silicone pastes and silicone resins. Blends

247
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

of silicone rubber with organic polymers are covered by the resolution where the
silicone monomer units are the pre-dominate species by weight.

Silicones that are used as food additives (e.g., as defoamers) in the manufacture of
substances such as wine are not covered by this resolution, but polysiloxanes used
as emulsifiers are. The resolution gives an OML of 10 mg/dm2 of the surface area
of the product or material, or 60 mg/kg of food, depending on the intended end-use
of the silicone rubber product (see below). There are restrictions on the types of
monomers that can be used to produce the silicone polymers and there is an inventory
list – Technical Document No.1 – of substances used in the manufacture of silicone
products (including rubbers) for food-contact applications.

An OML of 60 mg/kg of food or food simulant is stipulated for rubber products

that are regarded as being in Categories I and II, as defined in the ‘CoE Resolution
on Rubber Products’ AP(2004)4 (Section 8.3.6.1). The choice of food stimulant
and the conditions that are used for the overall migration experiment (i.e., time and
temperature) should be appropriate bearing in mind the conditions that the rubber
product will see in service. Guidance for the designing of these tests is given in
Technical Document No.4 of the Resolution.

8.4 Analysis Work to Characterise Food-contact Rubbers

8.4.1 Introduction

Although there are strict rules and guidelines about how a food-contact rubber can
be formulated and manufactured, the complexity of rubber technology, particularly
the chemistry associated with vulcanisation, has meant that considerable effort has
been made over the last 15 years to fully characterise the substances that have the
potential to migrate into food. As mentioned in Section 8.1, the UK FSA funded a
series of research projects from 1994 [6–9] to achieve this goal and this activity, and
the efforts of other funding bodies and research institutions, also encouraged other
work in this area. This section provides an overview of the analytical techniques that
have been developed to fingerprint the chemical substances that are present in food-
contact rubbers (Section 8.4.2) and the work that has been carried out to identify
and quantify substances that have been targeted specifically (Section 8.4.3).

248
 Compliance with Food-contact Regulations

8.4.2 Fingerprinting Potential Migrants from Rubber Compounds

8.4.2.1 Use of Gas Chromatography–Mass Spectrometry to Fingerprint

Food-contact Rubber Samples

It is often useful to produce a qualitative or semi-quantitative fingerprint of the low-

MW species in a rubber compound that have the potential to migrate into food.
GC–MS is often used for this purpose due to its high resolving power (important with
rubbers due to their complexity) and the identification power of the mass spectrometer.
In order to obtain data on as large a range of species as possible, it is often advisable
to use both headspace GC–MS (the solid rubber samples being heated to ≈150 °C)
and direct injection GC–MS, where an extract solution of the rubber, produced using
a non-selective solvent (e.g., dichloromethane or acetone), is injected straight into the
instrument. In the latter case, semi-quantitative data can be obtained on the species
that are detected by use of a single calibrant compound such as eicosane.

Work by Forrest and co-workers [18] using typical food-contact rubber compounds
has shown that, on average, between 20 and 30 compounds can be detected by
conventional GC–MS using these types of approaches. The commercialisation of
multi-dimensional GC–MS instruments has provided the analyst with greater resolving
power, coupled with improved detection limits and enhanced deconvolution software.
When a two-dimensional GC–MS system (GC×GC–ToF–MS) was applied to the same
food-contact rubber compounds by the research team, its use enabled the detection
of over 100 compounds.

To illustrate the GC–MS results obtained by Forrest and co-workers [18] using the
two types of GC–MS system, the data presented on a food-grade FR has been selected.
The formulation of the peroxide-cured food-grade FR (designated 49V) that was
compounded and vulcanised for 11 min at 170 °C for this exercise is shown in Table 8.5.

Table 8.5 Formulation of food-grade FR 49V

Ingredient phr
Viton™ GBL-200 (peroxide cross-linkable fluorocarbon tetrapolymer) 100
ZnO 3
Medium thermal N990 (CB filler) 30
TAC 3
Luperco 101-XL [2,5-dimethyl-2,5(di-tert-butyl-peroxy)hexane] 4
CB: Carbon black
TAC: Triallyl cyanurate
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells, Polymer Testing,
2006, 25, 63. ©2006, Elsevier Ltd [19]

249
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

An acetone extract, obtained by ultrasonicating 0.3 g in 10 ml of solvent for 30 min,

from this FR compound was analysed using the GC×GC–ToF–MS experimental
conditions listed below, and the three-dimensional chromatogram that resulted from
the analysis is shown in Figure 8.1.

• Instrument: Agilent 6890 gas chromatograph with LECO Pegasus III GC×GC–
ToF–MS.

• Injection: PTV injection set at 10 °C above the primary oven. Injection volume
of 1 µl.

• Primary column: J and W scientific HP-5MS (30 m × 0.250 mm with film thickness
of 0.25 µm).

• Secondary column: SGE BPX-50 (1.8 m × 0.100 mm with film thickness of

0.10 µm).

• Carrier Gas: helium, 1.0 ml/min at constant flow.

• Primary oven programme: 40° C for 10 min and the 10 °C/min to 320° C. Hold
at 320 °C for 15 min.

• Secondary oven programme: 50 °C for 10 min and then 10 °C/min to 330 °C.
Hold at 330 °C for 15 min.

• Modulator offset: 30 °C.

• Modulator frequency: 4 s.

• Hot time of modulator: 0.30 s.

• MS setting: 30–650 Da scanned at 76 spectra/s.

As demonstrated by Figure 8.1, the GC×GC–ToF–MS instrument provides a high

degree of resolution and, hence, separation of the species, resulting in a large number
being detected and identified in the acetone extract. Over 20 diagnostic compounds
related directly to specific formulation ingredients were identified and over 150
compounds in general. These numbers compared with only 10 and 36, respectively,
for the conventional GC–MS analysis of the same acetone extract, illustrating the
much greater profiling power of the GC×GC–ToF–MS system.

250
 Compliance with Food-contact Regulations

3
2n 2365
d
Co 2
lu
m RT
n mn
RT 1365 lu
1 Co
1st

0 365

Figure 8.1 GC×GC–ToF–MS total ion current (TIC) chromatogram of the acetone
extract of peroxide-cured FR 49V (RT: retention time). M.J. Forrest, S. Holding
and D. Howells in Proceedings of the High Performance and Specialty Elastomers,
20–21st April, Geneva, Switzerland, Rapra Technology Ltd, Shawbury, UK, 2005,
Paper No.2. ©2005, Rapra Technology Ltd [20]

8.4.2.2 Use of Liquid Chromatography–Mass Spectrometry to Fingerprint

Food-contact Rubber Samples

In-house databases have been developed as a result of laboratories using LC–MS

systems on rubber samples. Hence, the inclusion of this technique into the fingerprinting
process has complemented the data generated by GC–MS by contributing information
on thermally-labile and/or relatively-large (e.g., oligomeric) potential migrants.

Acetone, acetonitrile and diethyl ether solutions of rubber extracts can be analysed
directly by LC–MS. This was demonstrated by Sidwell [21], who described how
LC–MS was used to provide additional information on the species present in an ether
extract of a food-grade EPDM rubber. The formulation of this EPDM compound,
which was designated ‘EPDM 2’, is provided in Table 8.6.

251
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 8.6 Formulation of food-grade EPDM

Ingredient phr
Keltan® 720 100
ZnO 5
Stearic acid 1
Self-reinforcing CB N762 50
Strukpar 2280 paraffinic oil 8
Struktol WB16
®
1
Sulfur 2
MBT 1.5
Tetramethylthiuram disulfide 0.8
TDEC 0.8
Dipentamethylene thiuram hexasulfide 0.8
MBT: 2-Mercaptobenzothiazole
TDEC: Tellurium diethyldithiocarbamate
Reproduced with permission from M.J. Forrest, S. Holding and D. Howells, Polymer Testing,
2006, 25, 63. ©2006, Elsevier Ltd [19]

Some of the results presented by Sidwell [21], from both LC–MS and GC–MS analyses,
are reproduced here. For example, Figure 8.2 shows the TIC GC–MS trace for the
examination of the diethyl ether extractable species from a sample of the vulcanised
sulfur-cured EPDM 2 compound that had not been post-cured, a procedure that is
sometimes used to ‘clean-up’ food-contact rubber products. The species detected in
this GC–MS chromatogram are shown in Table 8.7 and, importantly, all of these
species are of a relatively low MW.

In an attempt to obtain more information on this diethyl ether extract, it was also
analysed by LC–MS using a C18 reverse-phase gradient elution separation with
atmospheric pressure chemical ionisation (APCI). Using the APCI head in the positive
mode gave the LC–MS TIC chromatogram shown in Figure 8.3. The last three peaks
showed ions of masses 538, 566 and 594, respectively, and are believed to relate to
the presence of tellurium dithiocarbamates in the extract (i.e., ions + 2H+ from the
ionised protonic solvent), which originated from the TDEC accelerator. Information
on the proposed chemical structure of these three species is presented below:

252
 Compliance with Food-contact Regulations

• Mass 538 – tellurium4+ with two dimethyldithiocarbamate groups and two

diethyldithiocarbamate groups (tellurium4+ being detected separately).

• Mass 566 – tellurium4+ with three dimethyldithiocarbamate groups and one

diethyldithiocarbamate groups.

• Mass 594 – tellurium4+ with four diethyldithiocarbamate groups.

Clearly, the presence of such high-MW species in the diethyl ether extract was not
detected by the GC–MS analyses (Table 8.7).

Abundance
TIC: RUN101.D
1000000

900000 Sample EPDM 2 Not Post Cured

Ether Extract
800000

700000

600000

500000

400000

300000

200000

100000

0
4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00
Time-->

Figure 8.2 GC–MS TIC chromatogram of the diethyl ether extract of EPDM 2.
Reproduced with permission from M.J. Forrest in Food Contact Materials –
Rubbers, Silicones, Coatings and Inks, Smithers Rapra Technology Ltd, Shawbury,
UK, 2009, p.37. ©2009, Smithers Rapra Technology Ltd [22]

253
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 8.7 Species extracted from EPDM 2 by diethyl ether and identified by
GC–MS (Figure 8.2)
Peak time (mins) Assignment Origin of the species

5.11 Isothiocyanato-ethane From cure system

7.52 Dicyclopentadiene Monomer

8.50 N-formylpiperidine From cure system

9.19 Tetramethylthiourea From cure system

9.27 Benzothiazole From cure system

11.92 2-(Methylthio)-benzothiazole From cure system

12.16 N-ethyl-2-benzothiazolamine From cure system

14.48 N-(2-hydroxyethyl)-dodecanamide From process aid

14.88 Pyrene From CB

15.09 N,N-dimethylpalmitamide From process aid

15.42 N-(2-hydroxyethyl)-dodecanamide From process aid

16.57 Di(2-ethylhexyl)phthalate Contaminant

17.18 4-Ethyl-2-propyl-thiazole From cure system

Reproduced with permission from M.J. Forrest in Food Contact Materials – Rubbers,
Silicones, Coatings and Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.38.
©2009, Smithers Rapra Technology Ltd [23]

Other studies have demonstrated the power of LC–MS-based techniques to fingerprint

low-MW species in food-contact rubbers. For example, Fichtner and Giese [24]
showed how both LC–MS and LC–MS/MS can be used to profile the low-MW species
present in the rubbers and plastic components used in food and pharmaceutical
industries. The advantages that these techniques offer over HPLC were emphasised
in their article, as is the work that will be required in the future to ensure that these
techniques are exploited to their full potential.

254
 Compliance with Food-contact Regulations

MSD2 TIC, MS File (J0329\LCMS0003.D) APCI, Pos, Scan, Frag: 190

0.543
1600000 APCI positive total ion chromatogram of an ether extracted rubber sample

20.119
1400000

1200000
0.821

1000000 7.940
1.855

21.441
18.709
800000
5.632
1.390

600000
10.328
10.076
9.828

400000

30.405
7.125

12.453

27.714

31.472
32.319
4.551

17.172
200000

5 10 15 20 25 30 min

Figure 8.3 LC–MS APCI (positive mode) TIC chromatogram of the diethyl ether
extract of EPDM 2. Reproduced with permission from M.J. Forrest in
Food Contact Materials – Rubbers, Silicones, Coatings and Inks, Smithers Rapra
Technology Ltd, Shawbury, UK, 2009, p.38. ©2009, Smithers Rapra Technology
Ltd [23]

8.4.3 Determination of Specific Substances in Food-Contact Rubbers

The main classes of potential specific migrants as well as the analytical techniques
used to detect and quantify them are described in Sections 8.4.3.1–8.4.3.5.

8.4.3.1 Monomers

Monomers are either in the gaseous state or relatively volatile liquids. Hence, GC and
GC–MS-based techniques are used to determine them in both rubber compounds and
food simulants/food products. To simplify the analysis, a static headspace sampler is
often used to isolate the monomer from the sample matrix because an extract-based
procedure often presents chromatographic problems, with the extraction solvent and
co-extractants often obscuring the analyte.

255
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

8.4.3.2 Plasticisers and Process Oils

These additives are essentially viscous, high-boiling-point liquids and so the most
appropriate technique to use is LC–MS. A range of synthetic plasticisers such as
phthalates, adipates, mellitates and sebacates can be detected using APCI ionisation.
Process oils are hydrocarbon mineral oils and require either an atmospheric
pressure photoionisation head (which can ionise non-polar species) or, if the oil has
sufficient aromatic character, the use of in-line ultraviolet or fluorescence detectors.
A fluorescence detector is particularly sensitive in the detection of polyaromatic
hydrocarbon compounds in such oils.

8.4.3.3 Cure System Species, Accelerators and their Reaction Products

This class of additive can present problems as it is often thermally labile, reactive and,
in some cases, has a degree of ionic character (e.g., dithiocarbamate salts). In some
cases, the reaction products [e.g., methyl aniline from N,N'-di-ortho-tolyl guanidine
and cyclohexylamine from N-cyclohexyl-2-benzothiazole sulfenamide (CBS)] are stable
and so GC and GC–MS can be used. Peroxides are popular curatives for food-use
rubbers and the stable, breakdown products of these can be easily detected by GC–MS.

In a number of cases, LC–MS is a more appropriate technique than GC–MS. It is

also easier to use LC–MS with most of the approved food simulants because, being
compatible with the mobile phase, they can be injected directly into the instrument,

Nitrosamines are derived from the secondary amines that are the breakdown
products of a number of commonly used accelerators (Section 7.5.1 and 8.5.3). These
potentially carcinogenic species can be determined at low-ppm levels by the use of a
combined GC–thermal energy analyser (TEA) instrument. Samples can be prepared
from rubber compounds by either solvent extraction or food migration studies and
then, after a concentration step, injected into the GC. The separated nitrosamines
enter a catalytic pyrolyser where nitrosyl radicals are generated. These react with
ozone introduced into the system to form a new radical that is chemiluminescent as
it returns to the ground state. The emitted light generated by this loss of energy is
detected and quantified. A number of factors have been found to affect the levels of
nitrosamines found in a particular rubber sample [25].

8.4.3.4 Antidegradants and their Reaction Products

This class of additive is less thermally labile and reactive than the preceding one and
GC-based methods can be used for a number of its members. However, due to the

256
 Compliance with Food-contact Regulations

relatively large number of high-temperature processing steps used with rubbers (i.e.,
mixing, extruding/calendaring, moulding and, in some cases, post-curing), a number
of low-volatility, oligomeric antidegradants are commercially available and the higher
oligomers of these far exceed the MW limits of GC. LC–MS methods, therefore, have
to be used and the technique has proved to be of great value in determining a range
of amine and phenolic types.

8.4.3.5 Oligomers

Low-MW oligomers have the potential to migrate from rubbers and those that
have a MW of ≤1,000 Da are of particular interest because, like other chemical
substances in this weight range, they have the potential to be absorbed by the human
gastrointestinal tract. A number of chromatographic techniques have the capability
to detect and quantify oligomers, including GC–MS, LC–MS and supercritical fluid
chromatography (SFC). Their effectiveness depends upon the chemical nature of the
oligomers (e.g., structure and thermal stability) and the weight range being targeted.

8.5 Analysis Work to Detect and Quantify Specific Migrants in

Food Simulants and Food

8.5.1 Introduction

National regulations and CoE Resolutions stipulate concentration limits for certain
species within food-contact rubber compounds (Section 8.3) and they also have SMLs
for certain migrant compounds.

Analytical work is, therefore, required on a quality-assurance basis to ensure that

a food-approved rubber compound remains fit for purpose (e.g. by checking the
monomer level) and to ensure that compounds having SMLs do not exceed them
in food simulants or food samples prepared using appropriate contact conditions.

These tests are, therefore, used to target specific chemical compounds for which there
is a toxicological concern. The test methodology varies according to the regulations
being applied, but some species (e.g., nitrosamines and nitrosatable substances) appear
regularly due to the degree of concern associated with them. Other specific migrants
that are often targeted include:

• Aromatic amines

257
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Other amines (e.g., cycloaliphatic amines)

• Peroxides and their breakdown products

• Formaldehyde

• Monomers (e.g., ACN)

• Accelerators (e.g., zinc dibutyldithiocarbamate, CBS, MBT)

This list is not complete because it is recognised that rubber products contain two
important classes of ingredients (antidegradants and curatives) that are reactive
and so produce a large range of reaction products and breakdown products. Work
carried out at Rapra for the FSA [8] showed that there are >1,000 of these products
originating from the 200 compounds in the 2004 CoE Rubber Resolution AP(2004)4
inventory list.

For convenience, in this section, the work that has been carried out in this area has
been divided into two sections:

• The chemical type of the potential migrant – Section 8.5.2.

• The specific food-contact product – Section 8.5.3.

8.5.2 Chemical Type of the Potential Migrant

8.5.2.1 Alkylphenol and Bisphenol A

Concerns over their potential to function as endocrine disruptors led to a Japanese

study on the levels of alkylphenols in 60 rubber products [26]. Such compounds
are used as starting materials in the manufacture of a number of rubber additives,
particularly oligomeric phenolic antioxidants. The work concentrated on four
compounds: p-tert-butylphenol (PTBP), p-tert-octylphenol (PTOP), p-nonylphenol
(NP) and bisphenol A. The results showed the presence of PTOP in three samples
in the range 2.2 to 37 µg/g, NP in 15 samples in the range 2.6 to 513 µg/g, and
no PTBP or NP in any samples. Some specific migration experiments for NP were
also carried out using water, 20% ethanol and n-heptane. The levels were found
to vary from 0.004 to 1.519 µg/ml, with the highest results being obtained with
the n-heptane.

Hakkarainen covered the use of solid-phase microextraction for the analysis of

migratable substances from food-contact materials in a chapter within a book on

258
 Compliance with Food-contact Regulations

chromatography [27]. In particular, he described how the technique can be used to

target a number of substances, including:

• Butylated hydroxyltoluene (BHT) in bottled drinking water.

• Bisphenol-type contaminants.

• Acetaldehydes and terephthalic acid in polyethylene terephthalate bottles.

• Butyl tin compounds in food and beverage packaging.

8.5.2.2 Peroxide Breakdown Products

Peroxides can be used to cure silicones and a number of other rubbers. Work carried
out by Novitiskaya and Ivanova [28] on a peroxide-cured NBR detected between 0.82
and 6.41 mg/l of the breakdown product diisopropyl benzene in an aqueous food
simulant (distilled water). Work has also been carried out at Smithers Rapra on the
migration of substances from silicone rubbers, and among the substances detected
were peroxide breakdown products (Section 8.5.3).

8.5.2.3 Dimethylsiloxanes and other Components from Silicone Rubbers

A test report produced by the Fraunhofer Institute of Food Technology and Packaging
[29] commented on the migration of siloxanes from three silicone rubbers – a high-
temperature curing material, a room temperature curing material, and a cured liquid
silicone rubber. Five food simulants (isooctane, ethanol, ethanol/water, ethyl acetate
and olive oil) were used and one of the experimental parameters investigated was the
degree to which the thickness of the sample effected overall migration. In the case
of the hydrophobic solvents, it was found to be more important than the polarity of
the simulant. As expected, the results obtained with ethanol/water mixtures showed
that the amount of migrating oligomeric material reduced markedly with increasing
water content, a virtually zero result being obtained above the level of 30%. The
migrants were characterised by SFC using both a flame ionisation detector and MS
detection. A homologous series of methyl-terminated linear siloxane oligomers of
≤20 SiMe2O units was identified.

A review of the use of a number of analytical techniques [e.g., infrared, GC–MS,

nuclear magnetic resonance (NMR), atomic absorption spectroscopy] to identify
and quantify polydimethylsiloxanes (PDMS) in a wide range of matrixes (e.g., food,
pharmaceuticals and cosmetics) was published by Mojsiewicz-Pienkowska and
Lukasiak [30]. Their study also considered the toxicological issues surrounding PDMS.

259
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Smithers Rapra have carried out studies of the migration of silicone oligomers from
food-contact silicone elastomers as part of the FSA project A03046 [9, 31]. They
found that a range of cyclic and linear siloxane oligomers could be detected. Further
information of the findings of this research is provided in Section 8.5.3.

8.5.2.4 Accelerators and Antidegradants

An investigation has been carried out and reported by Kazarinova and Ledovskikh
[32] on the migration of the accelerator diphenyl guanidine and its reaction products
from rubber compounds into food simulants. The concentration of migrants was
found to be influenced by the fillers present in the rubber compounds, with non-black
vulcanisates giving the highest values.

The results of a study presented in 1999 by Kretzchmar [33] suggested that when
guanidine accelerators are used in sulfur vulcanisation with phenylenediamine-based
antioxidants, carcinogenic aromatic amines and toxic isothiocyanates may be formed.
Similarly, when N-(1,3-dimethyl butyl)-N'-phenyl-p-phenylenediamine is used as an
antiozonant, it may decompose and react in aqueous food types to form aromatic
amines. This study also included the effects that compound formulation and curing
conditions can have on these reactions.

Barnes and co-workers [34] developed a LC–MS method to identify vulcanisation

agents and their breakdown products in food and drink samples. A large sample
of 236 retail foodstuffs were analysed for the presence of MBT and its breakdown
product benzothiazole (BT). The accelerators 2,2'-dithiobis(benzothiazole) (MBTS)
and CBS, which are commonly used in food-contact rubbers, were also searched for,
and MBT and BT are also known to be breakdown products of these two compounds.
The detection limit for these species was found to dependent on the food product type
and ranged from 0.005 to 0.043 mg/kg. MBT, BT, MBTS or CBS was not detected
by the analytical work in any of the samples above these values.

Kruger and co-workers [35] studied the migration of primary aromatic amines (e.g.,
aniline) and secondary amines into food simulants that came into contact with styrene-
butadiene rubber (SBR) rubber compounds containing a series of phenylenediamine
antiozonants having thioether and sulfoxide groups. The results that the team
obtained were discussed in comparison with those obtained using N-phenyl-N'-(1,3-
dimethylbutyl)-p-phenylenediamine as an antioxonant in the SBR compound.

260
 Compliance with Food-contact Regulations

8.5.2.5 Oligomers

The wide availability of LC–MS instruments means that they are now rivalling SFC
for the analysis of oligomers. The MW range of LC–MS instruments can be extended
to ≥4,000 Da, and this capability makes them ideal for the characterisation of
oligomers. For example, it has been shown [9] that silicone oligomers can be detected
by this technique in food simulants and some of the results obtained are shown in
Section 8.5.3.

8.5.3 Food-contact Products

8.5.3.1 Teats and Soothers

Historically, a variety of rubbers has been used in the manufacture of teats for baby
feeding bottles, including NR, silicone and styrene-butadiene block copolymers. The
rubbers that were vulcanised using sulfur-based cure systems presented particular
concerns in this product area due to the potential for nitrosamines to be present. The
EU published legislation on the level of nitrosamines permitted in babies’ teats and
soothers in 1993 in the form of a European directive (Section 8.3.1).

The publication of this directive came at a time when there were concerns over the
presence of nitrosamines in the processing environment within the rubber industry
(Section 7.5.1) and in rubber products, as well as from studies conducted on teats and
soothers. An example of the latter was carried out by Mizuishi and co-workers [36].
They focused on the migration of nitrosamines from teats and soothers made from
silicone, NR, synthetic polyisoprene and SBR, and showed that several nitrosamine
compounds were detected and that, predictably, the products made from NR yielded
the highest levels.

In addition to the targeting of nitrosamines, analytical work of a more general nature

has been carried out on teats and soothers. For example, a survey of the extractables
present in rubber teats was published in 1991 [37]. The samples were extracted with
diethyl ether or acetone and the extracts analysed by GC and GC–MS. Data was
obtained on 49 rubber teats commercially available in the Netherlands and a number
of compounds not permitted in the Dutch regulations were identified: dibenzylamine,
acetophenone, zinc dibenzyldithiocarbamate, 4,4'-thio-bis(2-tert-butyl-5-methyl)
phenol and bis(2-hydroxy-3-tert-butyl 5-ethylphenyl)methane.

A survey from the Netherlands in 2003 [38] looked at the migration of N-nitrosamines,
N-nitrosatable substances and MBT from 19 samples of teats and soothers. In addition

261
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

to these species, screening work was carried out for any other potential migrants.
In finding that the majority of teats and soothers were manufactured from silicone
rubber, this study highlighted how the industry was moving away from conventional,
sulfur-cured synthetic rubbers and towards cleaner rubbers that could not contain
substances like nitrosamines. However, NR products were present in the sample and
only one was found to be above the permissible limit, and that was for nitrosatable
substances at 0.23 mg/kg. MBT was found in only one of the NR products, and this
was below the migration limit of 0.3 mg/teat.

Dopico-Garcia and co-workers [39] evaluated the stability of some phenolic

antioxidants and the oxidation product of one phosphate antioxidant in four food
simulants (distilled water, 3% acetic acid, 10% ethanol and olive oil) at different
temperatures and time intervals. The results showed that, in general, the antioxidants
were more stable in olive oil than in the aqueous food simulants. Of the aqueous
simulants, distilled water and 10% ethanol allowed the highest stability at low
temperatures. Some differences were observed according to MW with the 3%
acetic acid simulant. The low-MW antioxidants had good stability in this simulant,
even at high temperatures, but the high-MW examples decomposed rapidly at low
temperature.

Di Feng and co-workers [40] analysed non-target compounds in silicone rubber teats
with a view to providing the means to carry out a safety evaluation and early-warning
mechanism for these and similar polymeric materials. They used two analytical
approaches to extract non-target compounds in 30 silicone rubber teats: purge and
trap and solid-phase microextraction. This strategy enabled 140 compounds in 12
categories to be separated and detected by GC–MS. Three qualitative methods were
used to identify the compounds: spectral matching using a library, retention index,
and confirmation by use of standards. Consideration was given to how many of these
140 compounds represented a potential danger to infant health, and 53 were selected
based on particular criteria, such as their concentration and potential toxicity. These
53 compounds included aromatic compounds, siloxanes, BHT, trimethyl silanol,
N,N-dibutylformamide and BT.

Sung and co-workers [41] used a LC–MS/MS method as an alternative to the GC–
TEA method recommended by the European Committee on Standardisation for the
simultaneous determination of eight N-nitrosamines released into artificial saliva
from rubber teats and soothers. The team found that the method could be validated,
using N-nitrosodipropylamine-d14 as the internal standard, with relatively-good
analytical results, including a sufficiently low detection limit of 0.1 to 2 µg/kg of
sample. The results also showed that the LC–MS/MS method was sufficiently rugged
and successful to be used for the routine analysis to demonstrate compliance with
European Directive 93/11/EEC (Section 8.3.1).

262
 Compliance with Food-contact Regulations

8.5.3.2 Rubber Adhesives

Rubber adhesives, both natural and synthetic, are used widely in the manufacture
of multi-laminates commonly employed as food packaging. For this reason, it is
important to identify the compounds that have the potential to migrate from the
laminate into food. Nerin and co-workers [42] identified 29 compounds in two rubber
adhesives, some of them with high toxicity levels according to the theoretical model
of Cramer (e.g., benzene 4-cyanocyclohexene and benzene isothiocyanate). The team
carried out work to determine the partition of these compounds between adhesive
and different substrates. The diffusion in both media was found to be very variable
depending upon the properties of the compounds and substrates used. The results
of the study also showed that only three compounds (benzene 4-cyanocyclohexene,
BHT and 2-cyclopentyl-1,3,5-trimethylbenzene) migrated into food and these were
below the SML set by the EU, and the values recommended by Cramer.

8.5.3.3 Meat Netting

NR has been the traditional material for elastomeric meat netting to hold joints
of meat together during packaging and cooking for many years. This has led to a
number of studies on the levels of N-nitrosamines, nitrosatable and other compounds.
Work has been carried out in the US by Marsden and Pesselman [43] using a typical
product from NR latex in contact with a 50% ethanol simulant for 150 min at
152 °C. The results showed that the most abundant zinc carbamate compound was
zinc dibenzyldithiocarbamate (860 µg/g netting) and the most abundant secondary
amine was dimethylamine (8.8 ng/g netting).

A collaborative survey of ten meat-netting samples obtained from four manufacturers

was carried out in the Netherlands [44] by workers from the Inspectorate for Health
Protection and Veterinary Public Health as well as the National Institute of Public
Health and the Environment. All ten samples consisted of both NR and vegetable
fibres and, in addition to nitrosamines and N-nitrosatable substances, the samples were
screened for other potential migrants. Nitrosamines were detected in concentrations
up to 2 mg/kg of netting and the two N-nitrosatable compounds dimethylamine
and dibenzylamine were found in up to 0.4 mg/kg of netting. These values were not
considered to be of concern to public health because of the ratio of meat netting to
food product. The other potential migrants identified included alkanes, alkenes, acids,
antioxidants, plasticisers and sterols, several of which were not authorised for food
contact in the Netherlands, but were allowed in other countries.

263
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

8.5.3.4 Rubber Gloves for Handling Food

Datta and Gonlag [45, 46] looked at a number of issues involving latex gloves,
including nitrosamines, zinc, allergy problems and the food-contact legislation and
regulations that exist in Europe. Additional information concerning the allergy
problems that can be associated with latex gloves is provided in Section 11.2.15.

A study was published in 2001 [47] that reported GC–MS results from solvents
extracts (e.g., n-heptane and n-hexane extracts) obtained on a range of gloves,
including those produced from NR and NBR. A range of accelerator and plasticiser-
type species were identified, but it was apparent in the case of the nitrile samples that
a relatively large number of extracted compounds could not be identified by GC–MS,
with no match being found in commercially available libraries.

Another study was then carried out on NBR gloves by members of the same team with
a view to improving the overall quality of the data [48]. Six compounds, which were
common to a number of the nitrile gloves used in the original work, were isolated
from an n-hexane extract by silica gel chromatography, and then these compounds
were identified by NMR and high-resolution MS.

8.5.3.5 Thermoplastic Rubbers

Although the market share for TPE in food-contact situations is relatively low, they
find niche applications where their ease of processing and physical properties enables
the replacement of vulcanised rubber. Such applications can include conveyor belting,
diaphragms and kitchen utensils. Because of their intermediate status, it is often the
case that legislation for both rubbers and plastics has to be consulted to enable a full
regulatory assessment to be made. A study by Sidwell [49] addressed the approach to
the regulation of TPE in Europe and the US. He also reviewed the use of mathematical
modelling to assess the migration of substances from TPE into food.

A review of the use of thermoplastic rubbers for the food packaging, medical and
pharmaceutical industry was presented by Siepmann at the 2012 ‘Thermoplastic
Elastomers’ conference [50]. In this review it was highlighted that these materials
were demonstrating an annual growth of ≈7.5% in these regulated industries, and
they were being chosen as substitutes for vulcanised rubbers, PVC or silicones.

8.5.3.6 Natural and Synthetic Rubber Products

As part of the work carried out on the FSA research project that targeted breakdown

264
 Compliance with Food-contact Regulations

products and reaction products in food-contact rubbers [8], GC–MS and LC–MS were
used to identify substances that had migrated from the rubbers into food simulants
and food products. For the preparation of the migration samples, food simulants,
food products and test conditions were selected that were representative of the types
of end-use conditions that the different rubber compounds could experience in service.
Table 8.8 shows some of the results that were obtained using GC–MS to analyse the
selected food products.

Table 8.8 Identification of migrants into different food products from a

variety of rubber compounds by GC–MS
Rubber compound Food Examples of migrants detected by GC–MS**
and condition* product
Hydrin 20D aged Olive oil 4,4'-Thiobis[2-(1,1-dimethylethyl-5-methyl-phenol)]
EPDM 239C unaged Beer 4-Oxopentanoic acid, n-butyl ester
Olive oil TAC
EPDM 240C unaged Olive 9,10-Dihydro-9,9-dimethyl-acridine
N-phenyl-1-naphthylamine
NR 504A aged Olive oil N-isopropyl-N'-phenyl-p-phenylenediamine
4,4'-Thiobis[2-(1,1-dimethylethyl-5-methyl-phenol)]
EPDM 242C unaged White wine N-butyl-1-butanamine
Olive oil N-phenyl-1-naphthylamine
SBR 195S aged Apple juice N-isopropyl-N'-phenyl-p-phenylenediamine
Olive oil N-isopropyl-N'-phenyl-p-phenylenediamine
Butyl 51B unaged Olive oil Aniline
N-phenyl-octadecanamide
CR 201E aged Olive oil N-phenyl-1-naphthylamine
*Some cured rubber samples were aged by placing in an oven prior to the migration test to
reflect ageing in service
**The foodstuffs were extracted (e.g., with a pentane/diethyl ether mixture) to provide
samples for analyses
Adapted from M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.85 [51]

An example of the data obtained on the food simulants using LC–MS, in both positive
and negative APCI mode, is shown in Table 8.9. This table lists the migrants that

265
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

were detected in 3% acetic acid that had contacted a sample of a cured SBR rubber
compound, designated 195S, for 4 h at 100 °C.

Table 8.9 Identification of migrants from the SBR 195S into 3% acetic acid
by LC–MS
APCI mode Retention time Mass (m/e) Migrant
of detection (min)
Positive 18.6 185.0 N-phenyl-1,4-phenylenediamine
Positive 20.2 397.2 MBT zinc salt
Positive 23.8 227.1 N-isopropyl-N'-phenyl-p-phenylenediamine
Negative 16.8 165.9 2-Mercaptobenzothiazole
Adapted from M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and Inks,
Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.311 [52]

It can be seen from the assignments in these two tables that the migrants detected by
GC–MS and LC–MS were either breakdown/reaction products of antidegradants or
accelerators, or were examples of these types of compounding ingredients in their
unchanged form. These two sets of data are only a selection of the results that were
obtained during this 3-year study. A full set of results is presented in the final report
for the project [8], which is available on the FSA website (Section 8.1). In addition,
Sidwell reviewed some of the findings of this research in a paper presented to Rapra’s
‘Food Contact Polymers’ conference in 2009 [53]. Some of the examples described
in the paper included the use of GC–MS and LC–MS to detect curative breakdown
products and volatile substances from NR and NBR compounds, and extractable
siloxane oligomers from food-grade silicone rubbers.

8.5.3.7 Silicone Rubber Products

Silicone rubbers are in wide use as food-contact materials for various applications.
This fact is reflected in the amount of food-related research and investigative work
that has been published. A review on the use of silicone products for food-contact
applications was published by Forrest in 2006 [54]. This publication includes an
extensive literature search. Some of the relevant literature that has been published
since 2006 is presented below.

Smithers Rapra carried out studies on the migration from food-contact silicone
elastomers as part of the FSA project A03046 [9, 31]. A combination of GC×GC–
ToF–MS and LC–MS was used to examine, detect, and identify the migrants in a

266
 Compliance with Food-contact Regulations

number of food simulants, including olive oil and 95% ethanol. The results revealed
that the migrating species could include:

• Silicone oligomers:

• Cyclic siloxanes

• Linear siloxanes (methyl and hydroxyl terminated)

• Cure system breakdown products:

• Peroxide breakdown products (e.g., ketones and aldehydes)

• Metal catalysts (e.g., platinum compounds)

• Oxidation products:

• Cyclic oligomers – from main chain scission

• Aldehydes – oxidation of alkyl groups (e.g., formaldehyde from methyl groups)

Examples of the data obtained are shown in Figures 8.4 (the GC×GC–ToF–MS TIC
chromatogram of the acetonitrile extract of the olive oil food simulant sample) and
Figure 8.5 [the LC–MS APCI (positive mode) TIC chromatogram of the 95% ethanol
food simulant sample]. A wide range of species peaks, including those of siloxane
oligomers, can be observed in Figure 8.4 due to the high resolving power of the
GC×GC–ToF–MS instrument. In Figure 8.5, the sequence of species peaks are due to
a series of oligomers from n = 7 (i.e., heptamer) to around n = 20 with a difference
of 74 mass units between each oligomer, equal to the dimethylsiloxane repeat unit.

Helling and Simat [55] of the Saxon Institute for Public and Veterinary Health
presented an overview of the migration behaviour of silicone rubbers used as baking
moulds. The migration results were obtained using proton NMR analysis of a range
of foodstuffs that had contacted the silicone rubbers. The study also contained a
critical assessment of the suitability of food simulants for evaluation of the migration
from silicone rubbers into cakes and muffins. Other aspects of the subject addressed
by the study included:

• Evaluation of migration from silicone rubbers into cakes, meatloaf and crème brûlée.

• Migration from silicone rubbers into pizza during long-term usage.

• Migration of fat from food into silicone rubbers.

• Influence of temperature treatment of silicone moulds of migration of siloxanes

into food.

267
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

1590
3

1090
2n
d 2 RT
Co n
lu m
m olu
n
RT 590 tC
1 1s

0 365

Figure 8.4 GC×GC–ToF–MS TIC chromatogram – olive oil food simulant sample
(acetonitrile extract). Reproduced with permission from J. Sidwell in
Proceedings of the Food Contact Polymers 2007, 21–22nd February, Brussels,
Belgium, Smithers Rapra, Shawbury, UK, 2007, Paper No.5. ©2007,
Smithers Rapra [31]

Intens.
x107

0 20 40 60 80 100 Time [min]

Figure 8.5 LC–MS APCI (positive mode) TIC chromatogram – 95% ethanol food
simulant sample. Reproduced with permission from J. Sidwell in Proceedings of
the Food Contact Polymers 2007, 21–22nd February, Brussels, Belgium, Smithers
Rapra, Shawbury, UK, 2007, Paper No.5. ©2007, Smithers Rapra [31]

268
 Compliance with Food-contact Regulations

In another study, Helling and Simat [56] reported the results obtained on ≈100
silicone rubber food-contact products (e.g., silicone teats and moulds) with respect to
migration behaviour, loss of volatile substances, amount of platinum catalyst, take-up
of fat from the foods, and the hygienic considerations surrounding the absorption
of fat. In the case of silicone teats, a high level of chemical and physical stability was
observed, with only minor levels (0.26%) of fat absorption. The sterilisation of silicone
rubber teats using a microwave up to 100-times was not found to have a significant
effect on its migration behaviour in contact with milk or its physical properties. In
the case of silicone moulds, a high level of oligomers was found to migrate and the
levels exceeded the current limits set by a number of EU national regulations and the
CoE Resolution (i.e., 60 mg/kg). The level of oligomer migration from the moulds
was found to decrease with repeated use, but the amount of absorbed fat from the
food (e.g., flaky pastry and soufflés) continued to increase up to 1.2%.

The results of other studies on the migration performance of silicone rubber baking
moulds have been published by Helling and other workers at the Saxon Institute for
Public and Veterinary Health. For example, in an article published in 2010 in Food
Additives & Contaminants [57], the researchers reported on the migration results
obtained using two food simulants, 95 and 50% ethanol, and also foodstuffs that
had different fat contents (e.g., creamed cake and meatloaf). Data were generated
on the substances that had migrated using the proton NMR method that the team
had previously validated and employed for this purpose (see below). Other areas
addressed by the article included the effect of tempering on siloxane migration, the
determination of volatile organic compounds (restricted in regulations such as the BfR
Recommendation XV), and the take-up of fat potentially leading to organoleptic and
hygiene issues. Another study [58] on silicone baking moulds published by the team
in the same journal described how the substances that migrated were characterised
using proton NMR, LC–MS and GC–MS, and were identified as cyclic organosiloxane
oligomers and linear partly hydroxyl-terminated organosiloxanes. These migration
studies were carried out using cakes that came into contact with a range of food
simulants (olive oil, isooctane, ethanol and Tenax™) and a validated proton NMR
method enabled quantification of the migrating siloxanes.

Helling and co-workers [59] reported on the work they carried out on the migration
properties of silicone materials during typical long-term commercial and household
use applications. This work was a continuation of earlier studies, examples of which
are provided in the paragraphs and references above, in that pre-characterised silicone
products were assessed during these long-term trials. One case in particular involved
silicone moulds that were used in a commercial pizza bakery on a daily basis up to
1,700 times. The following properties were monitored by the team over this period:

• Migration behaviour

269
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Uptake of fat

• Amount of volatiles and extractables

• Physical properties (e.g., elongation and tensile strength)

The migration of substances from the silicone mould into the food was determined
using proton NMR, and was found to decrease during the experimental exercise
from values between 11 to 18 mg/kg to below the limit of detection (<1 mg/kg). No
evidence was found for the formation of new migratable siloxanes by degradation
during the study period, and the loss of the siloxanes was compensated for by the
uptake of fat and other lipophilic food constituents. Also, the release of volatile organic
compounds decreased from 0.44% for the new moulds to 0.14% for the oldest ones
(i.e., those with ≈1,700 uses – equivalent to 400 h at 180 °C). The physical properties
of the mould remained reasonably constant during the study. Similar results for the
volatile compounds and physical properties were obtained for a long-term household
study that utilised baby teats sterilised in a microwave oven. However, there was a
difference with regard to the extractable siloxane content of new and old samples due
to the weak extracting power of milk, with no significant differences being detected.

References

1. Global Legislation for Food Packaging Materials, Eds., R. Rijk and R. Veraat,
Wiley-VCH Verlag GmbH & Co., Weinheim, Germany, 2010.

2. Global Legislation for Food Contact Materials, Volume 278, Ed.,

J.S. Baughan, Woodhead Publishing Series in Food Science, Elsevier Science
and Technology, Amsterdam, The Netherlands, 2015.

3. B. Geueke, C.C. Wagner and J. Muncke, Food Additives & Contaminants:

Part A, 2014, 31, 8, 1438.

4. M. Creese, Pharmaceutical Manufacturing and Packing Sourcer, 2012,

February, 30.

5. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and

Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009.

6. Project FS2219 – Migration Studies – Food Contact Materials, FSA, UK,

1994–1997.

7. Project FS2248 – Further research on chemical migration from food contact

rubbers and other elastomers FSA, UK, 1999–2001.

270
 Compliance with Food-contact Regulations

8. Project A03038 – An investigation of the breakdown products of curatives

and antidegradants used to produce food contact elastomers, FSA, UK,
2002–2004.

9. Project A03046 – Chemical migration from silicones used in connection with

food contact materials and articles, FSA, UK, 2004–2005.

10. R.C. Isa, Malaysian Rubber Technology Developments, 2011, Special

Edition, 11, 11.

11. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.15.

12. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.17.

13. European Union Directive 93/11/EEC: Release of N-nitrosamines from

rubber teats.

14. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.28.

15. L. Pysklo, T. Kleps and K. Cwiek-Ludwicka, Elastomery, 2002, 6, 4–5, 39.

16. L. Pysklo, Elastomery, 2003, 7, 1, 26.

17. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.22.

18. M.J. Forrest, S. Holding and D. Howells in Proceedings of the High

Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005.

19. M.J. Forrest, S. Holding and D. Howells, Polymer Testing, 2006, 25, 63.

20. M.J. Forrest, S. Holding and D. Howells in Proceedings of the High

Performance and Specialty Elastomers, 20–21st April, Geneva, Switzerland,
Rapra Technology Ltd, Shawbury, UK, 2005, Paper No.2.

21. J.A. Sidwell in Proceedings of RubberChem, 11–12th June, Munich, Germany,

Rapra Technology Ltd, Shawbury, UK, 2002.

22. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.37.

271
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

23. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.38.

24. S. Fichtner and U. Giese, Kautschuk, Gummi, Kunststoffe, 2004, 57, 3, 116.

25. B.G. Willoughby and K.W. Scott in Nitrosamines in Rubber, Rapra

Technology Ltd, Shawbury, UK, 1997.

26. A. Ozaki and T. Baba, Food Additives & Contaminants, 2003, 20, 1, 92.

27. M. Hakkarainen in Chromatography for Sustainable Polymeric Materials,

Renewable, Degradable and Recyclable, Eds., A. Albertsson and
M. Hakkarainen, Springer-Verlag, Berlin, Germany, 2008.

28. L.P. Novitiskaya and T.P. Ivanova, Gigiena i Sanitariya, 1989, 5, 88.

29. O. Piringer and T. Bucherl in Extraction and Migration Measurements of

Silicone Articles and Materials Coming into Contact with Foodstuffs,
FhG Test Report, 1994.

30. K. Mojsiewicz-Pienkowska and J. Lukasiak, Polimery, 2003, 48, 6, 401.

31. J.A Sidwell in Proceedings of Food Contact Polymers, 21–22nd February,

Brussels, Belgium, Smithers Rapra, Shawbury, UK, 2007, Paper No.5.

32. N.F. Kazarinova and N.G. Ledovskikh, Kauchuk I Rezina, 1978, 1, 26.

33. H-J. Kretzchmar in Proceedings of Hazards in the European Rubber Industry,

28–29th September, Manchester, UK, Rapra Technology Ltd, Shawbury, UK,
1999, Paper No.6.

34. K.A. Barnes, L. Castle, A.P. Damant, W.A. Read and D.R. Speck, Food
Additives & Contaminants, 2003, 20, 2, 196.

35. R.H. Kruger, C. Boissiere, K. Klein-Hartwig and H-J. Kretzschmar, Food

Additives & Contaminants, 2005, 22, 10, 968.

36. K. Mizuishi, M. Takeuchi, H. Yamanobe and Y. Watanabe, Annual Report of

Tokyo Metropolitan Research Laboratory of Public Health, 1986, 37, 145.

37. J.B.H. van Lierop in Food Policy Trends in Europe: Nutrition, Technology,
Analysis and Safety, Eds., H. Deelstra, M. Fondu, W. Ooghe and
R. Van Havere, Woodhead Publishing Ltd, Cambridge, UK, 1991.

272
 Compliance with Food-contact Regulations

38. K. Bouma, F.M. Nab and R.C. Schothorst, Food Additives & Contaminants,
2003, 20, 9, 853.

39. M.S. Dopico-García, J.M. López-Vilariño and M.V. González-Rodriguez,

Journal of Applied Polymer Science, 2006, 100, 1, 656.

40. D. Feng, H. Yang, D. Qi and Z. Li, Polymer Testing, 2016, 56, 91.

41. J.H. Sung, I.S. Kwak, S.K. Park, H.I. Kim, H.S. Lim, H.J. Park and S.H. Kim,
Food Additives & Contaminants: Part A, 2010, 27, 12, 1745.

42. C. Nerin, J. Gaspar, P. Vera, E. Canellas, M. Azner and P. Mercea,

International Journal of Adhesion and Adhesives, 2013, 40, 1, 56 .

43. J. Marsden and R. Pesselman, Food Technology, 1993, 47, 3, 131.

44. K. Bouma and R.C. Schothorst, Food Additives & Contaminants, 2003, 20,
3, 300.

45. R. Datta and A.T. Gonlag, Gummi Fasern Kunststoffe, 2003, 56, 12, 768.

46. R. Datta and A.T. Gonlag, Gummi Fasern Kunststoffe, 2004, 57, 6, 310.

47. C. Wakui, Y. Kawamura and T. Maitani, Journal of the Food Hygienics

Society of Japan, 2001, 42, 322.

48. M. Mutsuga, C. Wakui, Y. Kawamura and T. Maitani, Food Additives &

Contaminants, 2002, 19, 11.

49. J.A. Sidwell in Proceedings of Thermoplastic Elastomers, 8–9th November,

Brussels, Belgium, Smithers Rapra Technology Ltd, Shawbury, UK, 2011.

50. D. Siepmann in Proceedings of Thermoplastic Elastomers, 13–14th November

2012, Berlin, Germany, Smithers Rapra Limited, Shawbury, UK, 2012, Paper
No.19.

51. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.85.

52. M.J. Forrest in Food Contact Materials – Rubbers, Silicones, Coatings and
Inks, Smithers Rapra Technology Ltd, Shawbury, UK, 2009, p.311.

53. J.A. Sidwell in Proceedings of Food Contact Polymers, 21–22nd April,

Brussels, Belgium, Smithers Rapra Technology Ltd, Shawbury, UK, 2009,
Paper No.7.

273
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

54. M.J. Forrest in Silicone Products for Food Contact, Rapra Review Report
No.188, Smithers Rapra Technology Ltd, Shawbury, UK, 2006.

55. R. Helling and T.J. Simat in Proceedings of Silicone Elastomers,

7–8th October, Hamburg, Germany, Smithers Rapra, Shawbury, UK, 2009,
Paper No.9.

56. R. Helling and T.J. Simat in Proceedings of Silicone Elastomers,

30th–31st March, Cologne, Germany, Smithers Rapra, Shawbury, UK, 2011,
Paper No.11.

57. R. Helling, K. Kutschbach and T.J Simat, Food Additives & Contaminants,
2010, 27, 3, 396.

58. R. Helling, A. Mieth, S. Altmann and T.J. Simat, Food Additives &
Contaminants, 2009, 26, 3, 395.

59. R. Helling, P. Seilfried, D. Fritzche and T. Simat, Food Additives &

Contaminants: Part A, 2012, 29, 9, 1489.

274
9
Extractable and Leachable Studies on
Rubber Products

9.1 Introduction

The tests that have to be carried out on rubber products intended for food-contact
applications, many of which target migrating substances, are covered in Chapter 8.
This chapter describes the extractables and leachables (E&L) studies carried out on
rubber products to obtain information to support regulatory submissions for the
medical and pharmaceutical industries.

Although these two types of exercises often target and characterise the same set of
primarily, low-molecular weight (MW) substances, there are significant and important
differences in the methodologies and protocols. Creese [1] took this as his theme
and compared and contrasted the difference between the two sets of requirements.
There are also significant regulatory differences in that a specific set of regulations
cover food-contact materials and articles and others, for example, the European
Union Medicinal Products Directive 2001/83/EC, apply to products that contain
pharmaceutical products.

For a number of years, manufacturers of rubber products that were intended for
use in pharmaceutical applications (e.g., seals in single-use inhalers) and for certain
classes of medical devices (e.g., those that were not invasive) tended to use food-
approved rubber products as one of the starting points in their selection process
for an appropriate product. In recent years, a number of guidance documents
and standards have been published, such as those by the US Food and Drug
Administration (FDA) [2], the Product Quality Research Institute (PQRI) [3],
the United States Pharmacopeia (USP) [4–6] and the International Council for
Harmonisation of Technical Requirements for Pharmaceuticals for Human Use
(ICH) [7]. The objective of these guidance documents is to enable manufacturers
and end-users to carry out specific E&L test programmes to establish if the rubber
products are acceptable for their intended end-use [8]. This is a dynamic area and,
in 2016, representatives of the PQRI reported to the E&L USA Conference on their
toxicology teams’ approach to thresholds for leachables in parenteral drug products
[9]. Included in the presentation was mention of a database of 606 chemicals
that had been shown to potentially leach from container closure systems. These

275
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

chemicals had been compiled, analysed and sorted into classes using a modified
Cramer approach and a final classification scheme had been developed following
an initial risk assessment of the most toxic leachables. A database referred to as
the ‘E&L materials and safety database’ has been compiled by the Extractables and
Leachables Safety Information Exchange [10].

One of the fundamental differences between the extractables and leachables parts
of an E&L work programme is that solvents (e.g., isopropanol) that simulate drug
products are usually employed for the extractable study, whereas the actual drug
products are often used for the leachables study. It is also usual practice to use ‘worst
case’ conditions in terms of temperature and extraction process (e.g., refluxing) in
extractables work to ensure that everything that has the potential to be extracted and,
hence, ultimately leach into a pharmaceutical product in service, is removed from the
rubber. Leachable work is usually carried out using the complete drug packaging and
under conditions that, although still representing the worst case, are much closer to
the actual storage temperature and time.

The leachables that can be determined from a rubber, or other component of the
pharmaceutical packaging (e.g., plastic, label, adhesive and so on), can be regarded
as an imperfect sub-set of the extractables profile. It is an imperfect sub-set because
there is the potential for substances that are present in the extractables profile to be
chemically changed into new compounds during the leachables study, for example,
by a reaction with other components of the drug packaging or with the drug itself
(e.g., the ‘active component’ or other constituents within its formulation).

With regard to the definitions that apply to ‘extractable’ and ‘leachable’, general ones
that have proved useful in describing the two groups of substances are detailed below.

9.1.1 Extractables

Extractables are chemical compounds that migrate from any product (e.g., a drug)
contacting containing/closure material(s) when exposed to a solvent under exaggerated
conditions such as the use of a challenging solvent and/or high temperature and time.
These compounds can contaminant the drug product. Extractables are an intrinsic
property of the contact material.

9.1.2 Leachables

Leachables are chemical compounds, typically a sub-set of extractables, which migrate

into a drug formulation from any product (e.g., a drug) contact material(s) as a result

276
 Extractable and Leachable Studies on Rubber Products

of direct contact under normal process conditions or accelerated storage conditions:

• A property of the contact material, drug product and conditions of use.

• Leachables are generally a sub-set of extractables, but additional substances can be

generated by reactions that occur between leachable substances from the contact
material(s) and species within the drug media.

The two groups can be represented pictorially as shown in Figure 9.1.

Extractables

Leachables

Figure 9.1 Schematic representation of E&L

As with the majority of analytical investigations, prior knowledge of the specimen

to be tested is invaluable in assisting the analyst in choosing the correct and most
appropriate approach. With regulatory work that uses comprehensive documents
that provide in-depth detail on the methodology that must be used to demonstrate
compliance (e.g., food-contact regulations) this prior knowledge is not so important.
However, with E&L work, the guidance documents that are available assist the
analyst in putting together a programme of work, but this is often complemented
by input concerning the composition of the rubber so that an effective ‘study plan’
can be devised. Section 9.2 provides an overview of the origins of the substances and
species that have the potential to contribute to the E&L profiles of rubber products.

An introduction to the subject of E&L has been provided by Feilden [11] in the form
of an presentation in 2012 to the E&L USA Conference and also in a book published
in 2011 [12]. In addition, further useful background information is provided by Bailey
[13] that outlines a strategy for evaluating E&L to minimise risk when changing
materials in primary packaging components.

277
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

9.2 Origins of Substances that can contribute to Extractable and

Leachable Profiles of Rubbers

The complex compositional nature of vulcanised rubber products means that there
are numerous potential sources for the substances that can contribute to E&L profiles
and these have been the subject of a review by Forrest [14] and Hulme and Forrest
[15]. Also, Norwood and co-workers [16] provided a general overview of the E&L
issues that can be associated with materials, particularly rubber products, used in
contact with therapeutic products.

The species that can leach from rubber products can be reduced by undertaking
measures associated with their compounding and their manufacture, and these aspects
have been addressed by Toynbee [17] and Janssen [18].

An overview of the potential origin of the substances that can contribute to an E&L
profile is provided in Sections 9.2.1–9.2.5 and this overview underlines how complex
the situation is because, in addition to species associated with the polymerisation
stage, the polymer itself and the additives incorporated into the final compound, there
are chemical changes within the rubber that can take place for a number a reasons.
Such chemical changes can include:

1. Reaction products from intermediates and byproducts in the steps leading up to

the formation of the polymer.

2. Reaction products arising from the curing chemistry.

3. Products formed during antidegradant and stabiliser activity.

4. Thermal decomposition products of the polymer and additives.

Because it is often these same, relatively low-MW substances, that have the potential to
migrate into food (Chapter 8) and that contribute to vulcanisation fume (Chapter 7),
this review also has relevance to these subject areas and provides some important
background chemistry.

9.2.1 Polymer

9.2.1.1 Polymerisation Residues, Monomers, Oligomers and Impurities

Polymerisation residues are largely unavoidable during polymer production, but

can be controlled. For pharmaceutical, medical-grade polymers, stringent quality

278
 Extractable and Leachable Studies on Rubber Products

assurance controls are put in place and are often highlighted by the manufacturers
and are a big selling point – although at a cost premium.

Polymerisation residues can originate from many sources, but may include:

• Monomers and oligomers

• Catalyst remnants

• Polymerisation solvents and surfactants

• Non-polymerisable impurities

• Impurities from the manufacturing plant and polymerisation vessel

• MW modifiers (e.g., chain transfer agents)

Among the polymerisation modifiers used in emulsion polymerisation processes [e.g.,

for styrene-butadiene rubber (SBR) or nitrile rubber (NBR)] are long-chain mercaptans
(thiols) or disulfides. One such example is a mixture of C12 mercaptans with high
tertiary (i.e., R1R2R3CSH) content. This substance has been found to break down in
the rubber during curing to the respective alkenes [19]:

mixed dodecyl mercaptans → mixed dodecenes

Another example of a disulfide used for molecular control in polymerisation reactions

is the thiuram disulfide compound used in the production of chloroprene–sulfur
copolymers [20]. Copolymerising sulfur in this way improves the crystallisation
resistance of the resulting polychloroprene (CR), and these types of CR (e.g.,
Neoprene® G types) contain residual thiuram disulfide that can react further in the
subsequent curing reaction.

In natural rubber (NR), a number of species may remain from the biosynthesis or
subsequent treatment steps (Chapter 11). Examples include a number of fatty acids
(e.g., formic, acetic, propionic, lactic succinic and malic) [21] and the presence of
these create a number of possibilities for the formation of amides.

All polymers have a MW distribution, the width of which is dependent upon the
type of polymerisation technique used in their manufacture (e.g., free radical, ionic
or condensation). This means that at the end of a polymerisation there will be both
residual monomer species remaining in the polymer and oligomers composed of a
variable number of monomer units. The relatively small MW of these monomers and
oligomers means that they have the potential to migrate into food and pharmaceutical
products in their unchanged form, or they can undergo reactions in the rubber and
the reaction products can have the potential to migrate.

279
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

In some rubbers, for example, SBR, it is possible to detect residual monomers (e.g.,
styrene) at 100 ppm. In other cases (e.g., NBR), where the monomer is regarded
as being more toxic, the permissible level of residual monomer is much lower, for
example, acrylonitrile (ACN) has a maximum limit of 50 ppb for materials intended
for food-contact use. Residual monomers, such as ACN, may also play a part in
chemical reactions if additives such as accelerators are present. An example is the
involvement of ACN in thiuram reactions during NBR vulcanisation [22] and the
new products that can be formed in this way can include:

• 2-Cyanoethyl diethyldithiocarbamate from tetraethylthiuram disulfide (TETD).

• 2-Cyanoethyl dimethyldithiocarbamate from tetramethylthiuram disulfide

(TMTD) or tetramethylthiuram monosulfide.

With regard to oligomers, the situation with pharmaceutical/medical products is

different from food-contact studies due to the greater number of ways that species
can enter into the blood stream and contact bodily fluids (e.g., absorption through a
nasal membrane or direct, long-term implantation) and so a different range of MW
may be of interest depending upon their chemical nature.

9.2.1.2 Degradation Products of the Polymer

Some polymers will begin to ‘unzip’ to their monomers if heated (e.g., those containing
styrene such as SBR and styrene-butadiene-styrene). However, if such monomers
are found, not all of them may have arisen from degradation because, as shown in
Section 9.2.1.1, the residual monomer from the polymerisation may also be present.

In some cases, the presence of dissimilar units in the backbone can compromise
stability and lead to the formation of new substances. For example, in the case of
rubber copolymers:

• Acetic acid from ethylene-vinyl acetate copolymers.

• Hydrogen fluoride from vinylidene fluoride copolymers (e.g., FKM-type

fluorocarbon rubbers).

In considering polymer breakdown due to heating, thermo-oxidative attack during

processing or in service cannot be ruled out. In fact, thermo-oxidation may also occur
to the products of thermal degradation, or an initial oxidative step may trigger the
thermal degradation, so the two processes may not be easily separated. The products
of thermo-oxidation may include aldehydes and ketones.

280
 Extractable and Leachable Studies on Rubber Products

9.2.2 Additives in the Unaltered State

9.2.2.1 Plasticisers and Fillers

Plasticisers are used to soften a rubber product (e.g., by reducing the proportion of filler
and polymer) and/or improve its low-temperature flexibility by reducing their glass
transition temperatures. Such additives also assist in the processing and manufacture
of rubber products by improving their rheological properties.

There are many types of plasticisers that can be used in rubber products, including
hydrocarbon oils, esters and phosphates, and these substances are all relatively
thermally stable and non-reactive, so will be present in E&L profiles in their
unchanged form. Because they can be present at relatively high levels (e.g.,
15% w/w) they may also contribute to the chemical species in these profiles due
to the presence of the starting compounds that they are manufactured from.
A typical example is 2-ethylhexanol, a starting material for the manufacture of
di(2-ethylhexyl)phthalate.

Fillers, when added to rubbers, have the opposite effect to plasticisers in that they
harden the compound, but they are also used for other important reasons (e.g., to
increase tensile strength and tear strength). They are usually inorganic compounds
[e.g., calcium carbonate (CaCO3), silica and aluminium silicates] or different forms
of carbon black (CB). Hence, they are essentially insoluble in the solvents and drug
preparations used E&L studies, but may still have an impact (e.g., ppm levels of silica)
on the profiles, particularly the ‘extractables’ profile. Impurities may also be present
in fillers (e.g., hydrocarbons from CB, Section 3.6.4) and whether these contribute
to an E&L profile depends upon their chemical nature and how well they are bound
into the filler matrix. Overall, fillers can usually be expected to have much less of an
impact on E&L than plasticisers.

9.2.2.2 Antidegradants, Process Aids and Pigments

Although antidegradants (e.g., hindered phenolic compounds and amine compounds)

are designed to be reactive and stabilise the rubber during processing by undergoing
reactions, such as those involved in the chain breaking of free-radical reactions and
the dissociation of hydroperoxide groups, they are added at a level that will mean
that a high proportion of the initial stabiliser is present in a rubber in its unchanged
form so that it can continue to protect the rubber in service. For this reason, this class
of additive, and the ones stated below, can contribute to E&L profiles.

281
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Flow promoters and lubricants improve the processibility of rubbers by reducing

the friction between the rubber compound and the processing equipment. Examples
of these types of additives are stearate compounds (e.g., calcium stearate and zinc
stearate).

Although rubber compounds are usually black due to the very common use of CB
filler, they can be other colours (e.g., white or green) and these compounds will
contain pigments. Both organic pigments (e.g., azo dyes) and inorganic pigments
(e.g., titanium dioxide) can be used in rubber compounds.

9.2.3 Breakdown and Reaction Products of Cure System Components

Curatives are designed to break down in rubbers during vulcanisation to initiate

chemical reactions that will result in covalent bonds being formed as crosslinks
between polymer molecules. This breakdown ensures that only low levels of the
original curative compound(s) are present in the rubber. Indeed, in order to obtain the
best service-life properties, and to ‘clean-up’ the rubber product, a post-cure operation
is often employed after an initial curing cycle to ensure that the majority (i.e., >95%)
of the curative has reacted within the rubber. Some examples of the principal curative
systems used in rubbers, and the breakdown products/reaction products that may be
formed as a result of their use, are given below.

9.2.3.1 Peroxides

The decomposition of peroxides proceeds at measurable rates usually at temperatures

above 100 °C. The generic products of this decomposition, depending upon the
structure of the original peroxide compound, include: alcohols, alkenes and ketones.
The actual compounds formed, and their relative amounts, will depend on the peroxide
used and the environment within which it has reacted (e.g., cure temperature, type
of rubber).

For simplicity, the various dialkyl peroxides can be described in terms of two generic
formulae:

X−O−O−X X−O−O−Y−O−O−X (9.1)

where X is a monovalent organic group and Y is a divalent organic group.

282
 Extractable and Leachable Studies on Rubber Products

In the case of most commercial peroxides, X is either a tert-butyl group or the cumyl
group (e.g., di-tert-butyl peroxide, dicumyl peroxide). In the former case, the principal
decomposition products include tert-butanol, iso-butene and acetone. In the latter
case the principal decomposition products include cumyl alcohol, α-methyl styrene
and acetophenone.

Divalent groupings of the type Y decompose in a similar fashion, but not necessarily
to give the relatively simple molecules for which toxicological information is
documented. Probably the simplest commercial example of this group is 1,1-bis(tert-
butylperoxy)cyclohexane, in which the divalent grouping breaks down principally
to cyclohexanone.

Diacyl peroxides decompose more readily than dialkyl peroxides and this breakdown
is also amenable to catalysis (e.g., by tertiary amines such as N,N-dimethylaniline).
The mode of breakdown is also different from that of dialkyl peroxides in that the
acyloxy radical that is initially formed decomposes rapidly to an aryl radical by the
loss of carbon dioxide. The principal byproducts are, therefore, those of this second
radical and for aromatic types of these peroxides they include single-ring compounds
(by hydrogen abstraction) and double-ring compounds (by radical coupling).

The thermal breakdown characteristics and products of different types of diacyl

peroxide have been described by Davies [23]. Examples of the abstraction and coupling
products of two common peroxides are as follows:

• Dibenzoyl peroxide – benzene (abstraction); biphenyl (coupling).

• Bis(2,4-dichlorobenzoyl) – m-dichlorobenzene (abstraction);

2,2',4,4'-tetrachlorobiphenyl (coupling).

9.2.3.2 Sulfur Donor-type Cure Systems

Asymmetric cleavage in molecules of the type X-S-S-X (to give X radicals) is

believed to play a part in the sulfur donor curing reactions of agents such as
4,4'-dithiodimorpholine (DTDM) and TMTD [24]. Hydrogen abstraction or coupling
may then follow to give a range of products and, using DTDM and TMTD to illustrate
this, examples of these products can include:

DTDM → morpholine

TMTD → tetramethylthiourea

283
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

9.2.3.3 Accelerators used in Sulfur-type Cure Systems

Many studies have addressed the mechanisms and byproducts of accelerator action in
sulfur cures. Both sequential and competitive processes are involved, and the nature
and yield of byproducts may be highly dependent on the proportions of the ingredients
used in the rubber compound and on its thermal history. Amines and organo-sulfur
compounds may be expected to be formed and the scheme shown in Figure 9.2
illustrates just some of the possibilities when TMTD is used as an accelerator [25]. The
reaction scheme in Figure 9.2 also includes reactions with 2-mercaptobenzothiazole
(MBT), an accelerator in its own right, but also a breakdown product of thiazole
and sulfenamide accelerators, which are often used in cure systems in conjunction
with TMTD.

MBT
ZDMDC

N
CH
vulc S
TMTD TMTD
insol sol

∆ MBT
Me2NCS2H + MBTS

Me2NCSNMe2 + CS2 + S

∆ Me2NH + CS2
MeNCS + Me3N

Figure 9.2 Some potential reactions associated with the accelerator TMTD
[MBTS: 2,2'-dithiobis(benzothiazole) and ZDMC: zinc dimethyldithiocarbamate].
Reproduced with permission from M.J. Forrest and B. Willoughby in Proceedings
of the RubberChem Conference, 9–10th November, Birmingham, UK,
Rapra Technology Ltd, Shawbury, UK, 2004, Paper No.1. ©2004,
Rapra Technology Ltd [26]

284
 Extractable and Leachable Studies on Rubber Products

The scheme shown above includes possible interactions with MBT. It should be
recognised that MBT is an active agent in sulfenamide [e.g., N-cyclohexyl-2-
benzothiazole sulfenamide (CBS), 2-(4-Morpholinothio)benzothiazole (MBS)]
acceleration [27] and in these cases both MBT and MBTS are formed. The compound
MBT is important because it is a ‘special case’ compound for E&L studies as defined
by organisations like the PQRI (Section 9.3).

Amines are also formed in these types of reactions [e.g., cyclohexylamine from CBS and
tertiary butylamine from N-tert-butyl-2-benzothiazole sulfenamide (TBBS)]. Indeed,
amine formation extends beyond the thiuram and sulfenamide examples because
many accelerator compounds are manufactured from amines and some reversion to
these starting compounds appears possible. Examples of some of the amines formed
in accelerator action are listed in Table 9.1 [26]. These are distinguished in terms of
primary (RNH2) or secondary (R2NH) amine character.

Table 9.1 Primary and secondary amines from accelerators and curing agents
Rubber compounding ingredient Amine formed Boiling point of
amine (°C)
TMTD Dimethylamine (2°) 7.4
TBBS tert-Butylamine (1°) 44
TETD Diethylamine (2°) 56
N,N-diisopropyl-2-benzothiazole sulfenamide Di-isopropylamine (2°) 84
DTDM Morpholine (2°) 128
N-oxydiethylene-2-benzothiazyl sulfenamide Morpholine (2°) 128
CBS Cyclohexylamine (1°) 135
TBTD Dibutylamine (2°) 159
DPG Aniline (1°) 184
(1°) = Primary amine
(2°) = Secondary amine
DPG: Diphenyl guanidine
TBTD: Tetrabutylthiuram disulfide
Reproduced with permission from M.J. Forrest and B. Willoughby in Proceedings of the
RubberChem Conference, 9–10th November, Birmingham, UK, Rapra Technology Ltd,
Shawbury, UK, 2004, Paper No.1. ©2004, Rapra Technology Ltd [26]

285
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Amines are also highly reactive and so they would be expected to react further. Possible
reactions of this sort include:

• Amide formation with fatty acids:

RNH2 + HO2CR' → RNHCOR' (with primary or secondary amines) (9.2)

• Nitrosamine formation with nitrosating agents (NOX):

R2NH + NOX → R2N − N = O + HOX (with secondary amines only) (9.3)

Note: nitrosamines are another PQRI ‘special case’ compound, Section 9.3.

Fatty acids for reaction (i) can include the components of commercial stearic acid
(mainly octadecanoic, hexadecanoic and tetradecanoic) and any introduced into the
rubber compound as impurities. Sources of NOX for reaction (ii) may include the
nitrite salt bath medium used in continuous vulcanisation processes or impurities
introduced by other ingredients of the rubber compound (e.g., CB).

Other examples of potential amine reactions include symmetrical and unsymmetrical

thiourea formation from exchange reactions in mixed accelerator systems.

9.2.4 Breakdown and Reaction Products of Antidegradants used in

Rubbers

9.2.4.1 Antioxidants

Antioxidants can be subdivided by their mode of action into ‘primary’ or ‘secondary’

types [28]. Primary antioxidants function by ‘chain breaking’ mechanisms whereas
secondary antioxidants function by a form of ‘sacrificial oxidation’. The former are
phenols and aromatic amines, whereas the latter are species such as phosphites and
thioesters. In practice, both may produce oxidised products, and examples of some
of the possibilities are listed in Table 9.2 [14].

286
 Extractable and Leachable Studies on Rubber Products

Table 9.2 Some possible oxidation products of antioxidants

Antioxidants Possible reaction product
Butylated hydroxytoluene 2,6-Di-tert-butyl-1,4-benzoquinone
Trisnonylphenyl phosphite Trisnonylphenyl phosphate
Phenyl-beta-naphthylamine 1,2-Napthoquinone-2-anil
6PPD N-(1,3-dimethylbutyl)-N'-phenyl-1,4-p-
benzoquinone diimine
6PPD: N-(1,3-dimethylbutyl)-N'-phenyl-phenylenediamine
Reproduced with permission from M.J Forrest in Proceedings of Extractables & Leachables
for Pharmaceutical Products, 14–15th September, London, UK, Smithers Rapra Technology
Ltd, Shawbury, UK, 2010, Paper No.1. ©2010, Smithers Rapra [14]

Another reaction to consider in this context is thermal decomposition of the

antioxidant. For example, chemical compounds which are the products of Freidel–
Crafts reactions may undergo a catalysed reversion which selectively de-alkylates the
aromatic ring. Potential catalysts include acids, metal ions or amides [29]. Antioxidants
made by Freidel–Crafts chemistry include hindered phenols and alkylated (or arylated)
diphenylamines.

9.2.4.2 Antiozonants used in Rubber Products

Some rubbers are particularly prone to attack by ozone (e.g., diene-type rubbers such
as NBR) and so antiozonants have to be incorporated into the compounds to ensure
good ageing resistance in service. Antiozonants act by a sacrificial reaction with ozone
[30]. A byproduct of their chemistry in rubber is side-chain oxidation and this can
lead to the release of ketones as illustrated below for two antiozonants [isopropyl
p-phenylenediamine (IPPD) and 6PPD] in common use in the rubber industry:

IPPD → acetone

6PPD → methyl ethyl ketone

9.2.5 Carbon Black Fillers and the Formation of Nitrosamines

The involvement of CB in nitrosamine formation has been controversial and the co-
agents are believed to be the oxides of nitrogen adsorbed on the surface of the CB.
Their action in the formation of nitroamines is notably temperature-dependent and

287
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

has been investigated by Willoughby and Scott [31]. This effect arises because not
all the oxides are active in nitrosation. Dinitrogen trioxide (N2O3) and dinitrogen
tetroxide (N2O4) are NOX, but nitrogen dioxide (NO2) is not. The respective reactions
with secondary amines can be depicted as:

2R2NH + N2O3 → 2R2NNO + H2O (9.4)

R2NH + N2O4 → R2NNO + HNO3 (9.5)

The origin of the secondary amine from the use of certain accelerators is discussed
in Section 9.2.3.3.

Given that N2O4 and NO2 are interconvertible (N2O4 dissociates to NO2 on heating
and reforms on cooling), it is perhaps not surprising that nitrosamine formation in
CB-filled, accelerated sulfur vulcanisates is dependent on thermal history [31].

9.3 Overview of Extractable and Leachable Testing of Rubber

Products

Mention has been made in Section 9.1 of the guidance documents and protocols
concerning E&L that have been published in recent years. One of these, in addition
to a number of other publications of relevance in this field, has been produced by
the PQRI and they have produced a summary of their recommendations that address
the following areas:

• Safety thresholds

• Analytical evaluation threshold (AET)

• Integration of safety evaluation

• Componentry

• Controlled extraction studies

288
 Extractable and Leachable Studies on Rubber Products

• Leachables studies

• Routine extractables testing

The last of these areas is where three important special case compounds (i.e.,
polyaromatic hydrocarbons, MBT and nitrosamines) are listed, along with the
recommendation that they should be evaluated ‘by specific analytical techniques and
technology defined thresholds for leachable studies and routine extraction testing’.
These three compounds are particularly relevant to rubber compounds (Section 9.2)
and are often cited in the literature (Section 9.5).

Information on the potential origins of E&L from rubber products is provided

in Section 9.2. It can be seen that these compounds tend to be low- or relatively
low-MW species. Section 9.4 provides a case study detailing the approach used to
carry out an extractables study on a rubber product (e.g., a seal) of the type that
could be used in pharmaceutical packaging, a drug dispensing device or a medical
device, and illustrates the possible substances that could be found. Once such an
extractables profile has been obtained, the next step involves its examination by a
toxicologist, who would pay particular attention to the substances detected at the
highest concentrations. This toxicological assessment would highlight substances that
should be specifically targeted as ‘leachables’ during the next stage in the process. In
this leachables study, the analytical programme would be devised to include analytical
techniques and protocols optimised and validated to provide accurate quantifications
of these substances, often in the drug media. These data would then be used, along
with the end-use information associated with the medicinal product, to calculate the
maximum human daily exposure from normal clinical use to establish if safe daily
intake levels of the leachables were being adhered to [33, 33]. A keynote paper was
presented by Robison [34] of the FDA at the 2016 E&L USA Conference in which the
design of E&L studies and the toxicological risk assessment of E&L in primary and
secondary packaging were addressed. Case studies were covered in the presentation
that related to the following three products and a guide to the best practices provided:

• Pre-filled auto-injector

• Flexible bags for use with intravenous drug products

• Pre-filled syringe

A number of articles have addressed the design of analytical strategies to assess the
safety of E&L of pharmaceutical packaging, such as those produced by Van Dongen
[35], Schulz [36] and Feilden [37, 38].

The complexity of the rubber products that were traditionally used in sealing
applications in the medical/pharmaceutical industry [e.g., ethylene propylene diene

289
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

monomer (EPDM) and nitrile] meant that potential migrants could be present in
complex mixtures, which represented a challenge to the analyst (Section 9.4). Because
of the complexity of conventional rubber compounds, there has been a trend over
recent years for manufacturers to produce ‘clean’ rubbers for pharmaceutical and
medical applications. These are designed to address the concerns of the industry
by containing novel, ‘low extractables’ cure systems and fewer additives that can
contribute to E&L profiles, as well as avoiding classes of materials (e.g., those that
are animal-derived) and additives that could produce species of high concern, (e.g.,
MBT or nitrosamines). An example of these specialist clean rubbers is the bromobutyl
rubbers, marketed by ExxonMobil Chemical Company, which are cured using
proprietary cure systems.

With regard to changes in materials in the search of lower levels of E&L, other
possibilities are available to manufacturers of packaging and sealing materials.
One of these is to use thermoplastic elastomers (TPE) and high-performance TPE.
Examples of these materials are commercially available and can be used for disposable
devices, such as pre-filled syringes and vial stoppers [39–41]. Another option is to
use elastomeric cyclic olefin copolymers (COC) as alternatives to TPE. These COC
rubbers are regarded as having exceptional toughness and impact properties, as well
as the transparency, dielectric properties, low E&L and chemical resistance that have
come to be associated with this class of material. They can be used for a broad range
of applications in the medical, pharmaceutical packaging, food packaging, optics
and electronics sectors [42]. Despite the introduction of this new generation of clean
rubbers, there is still a need at times, in order to satisfy particular criteria, for more
traditional rubbers.

Irrespective of the type of rubber product being studied, the fact that a mixture
of compounds will result from the extractables phase of the work means that gas
chromatography (GC)–mass spectrometry (MS), both headspace GC–MS and
solution injection (also called ‘direct injection’) GC–MS, is an ideal analytical
technique for these mixtures with its low detection limit, high resolving power, and
good identification and quantification capabilities. The ability of headspace GC–
MS, particularly dynamic headspace GC–MS, to make an important contribution
in the E&L screening of pharma and medical grade polymers has been reviewed by
Jeanguyot [43].

Some problems, however, can be encountered with GC–MS and these occur if
an analyte is thermally labile (degradation problems), highly polar (adsorption
problems) or of a reasonably high MW (poor elution from the column). Because
of these considerations, liquid chromatography (LC)–MS is employed alongside
GC–MS, although care has to be taken because the mobile phase/separating column
combination is crucial to ensuring good chromatographic performance with this

290
 Extractable and Leachable Studies on Rubber Products

technique. As yet, there are no comprehensive commercial libraries available to assist

with identification and so there is not the same ease of identification that exists with
GC–MS, for which comprehensive 70-eV fragmentation voltage libraries have been
available for many years. The lack of a standardised approach means that this is
likely to remain the case for a long time to come. Identification is, therefore, based
on the analysis of standards and an understanding of the chemical processes (e.g.,
formation of adduct ions) that occur within the mass spectrometer.

Recent developments in instrumentation have provided the analyst with even more
powerful tools for this type of work. In the case of GC–MS, developments in the
GC oven area have enabled the GC×GC–time-of-flight (ToF)–MS instrument to be
developed. This technique is essentially a two-dimensional chromatographic operation
where both a non-polar and a polar column can be used in series for all the peaks
in the chromatogram via a modulation device. Hence, mixed peaks (i.e., those that
contain two or more species due to co-elution occurring with the first column),
or an area of the chromatogram where a series of peaks are unresolved, can be
separated into their individual components by use of the second column. ToF–MS is
complimentary to this technology due to its very fast data collection rate and this,
coupled with enhancements in the software package, provides an additional peak
deconvolution capability. This can sometimes be referred to as the ‘third dimension’
in these instruments.

In the case of LC–MS instruments, there have been a number of developments that
have made sophisticated versions of the technique accessible to test house laboratories.
For example, the LC–trap, which is essentially an LC–MS×MS instrument in which
fragment ions (or molecular ions) resulting from the first analysis, can be fragmented
a second time to provide a further series of fragments. This provides the benefits
of enabling structural information to be obtained on the original ions (good for
identifying unknowns) and definitive assignments in complex LC chromatograms
where peak resolution can be a problem and mixed spectra result. These benefits
are important in LC–MS because molecular ions (or molecular adduct) ions are
often formed and these are not always specific enough [e.g., where two compounds
(or adducts) have the same MW]. The second fragmentation pattern, together with
the initial ion(s) and the peak retention time, will assist the analyst in obtaining an
unequivocal identification.

Another important development is the greater availability of LC–MS instruments

equipped with a high-resolution accurate mass (HRAM) mass spectrometer, which
again enables a greater number of substances in extractables studies to be identified
and increases confidence in these assignments (Section 2.5.5). The performance of
these instruments can also be improved by the upgrading of the LC component to
one that has ultra-performance liquid chromatography (UPLC) status, the acronym

291
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

for the technique then becoming UPLC–HRAM/MS. UPLC–HRAM/MS instruments

are highly regarded for the rapid and sensitive screening of non-volatile organic
compounds, such as antioxidants, lubricants and larger polymer degradation products.
As mentioned in Section 2.5.5, there are options in the mode of operation of the MS
component of LC–MS systems. Using these instruments in both the atmospheric
pressure chemical ionisation (APCI) and electron ionisation (EI) mode offers the
chance of generating the most complete profile because the former is more sensitive
to compounds with low-to-medium polarity, whereas the latter is more sensitive to
highly polar compounds.

Ion mobility-MS is another novel approach to screening for E&L components from
packaging materials. Cabovska and Riches [44] and Cabovska [45] provided an
introduction to this technique along with collisional cross-section (CCS). Ion mobility-
MS measures the drift time of an ion and the application of a calibration provides
the CCS, a key physiochemical property of a species. According to the authors,
the advantages of CCS include a reduction in the number of false-positive data, an
additional dimension of separation to obtain cleaner spectra in complex samples, and
an improvement in reliability for the assignment of isomers, which are not separated
chromatographically.

As is apparent from the above information, this is a very active area for developments
and a recent presentation by D’Silva of ThermoFisher Scientific [46] provided an
overview of the analytical capabilities that are available for E&L studies, together with
an introduction to a new technology, called Q Exactive GC–MS from AstraZeneca,
for identifying and quantifying unknown impurities in support of pharmaceutical
applications.

To complete the picture when carrying out screening work for extractables on a rubber
product, a number of other analytical tests are usually employed. For example, a
semi-quantitative 36-element scan is carried out on solvent extracts (e.g., hexane or
isopropanol) of the product by inductively coupled plasma (ICP). The non-volatile
residue (NVR) of these solvent extracts is also obtained by quantitatively drying
down in an oven, and the bulk chemical characteristics of this NVR are obtained
by recording an infrared (IR) spectrum using a Fourier–Transform infrared (FTIR)
spectrometer.

To summarise, the analytical techniques that are usually brought together in a

complementary way to characterise the extractables that are present in rubber products
are listed, together with their contributions to the data obtained, in Table 9.3.

292
 Extractable and Leachable Studies on Rubber Products

Table 9.3 Analytical techniques used in an extractables study and their

contribution
Analytical technique Substances targeted

Headspace–GC–MS Volatile organic compounds

GC–MS* Semi-volatile organic compounds

LC–MS and LC– ultraviolet Non-volatile organic compounds

ICP–OES or ICP–MS Heavy metals and other elements

Ion chromatography Anions and cations

IR General chemical profile of dried, and weighed

extracted material (i.e., NVR)

*Solution injection GC–MS

OES: Optical emission spectrometry

Additional examples of the general and specific analytical approaches used to

undertake E&L studies to ensure that a comprehensive assessment results are provided
in Section 9.5. This section also contains a number of published examples of the E&L
work that has been carried out on rubber products, and provides specific information
on the substances that the authors have identified and quantified.

9.4 Case Study – Potential and Actual Extractables from an Nitrile

Rubber Compound

9.4.1 Composition of the Nitrile Rubber Compound

Section 9.2 discussed the origins of E&L species in rubber products. To provide a
reasonably in-depth illustration of some of these substances, and how analytical
techniques such as GC–MS and LC–MS can be employed to identify them in a
screening study, an example of a relatively complex rubber formulation is provided
in this section and the formulation of this rubber compound is shown in Table 9.4.

293
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 9.4 Formulation for the NBR compound

Ingredients phr
Breon N36C60 [acrylonitrile-butadiene copolymer (nitrile) rubber] 100
ZnO 5
Stearic acid 2
HAF N330 (CB filler) 15
Translink 77 (calcined and surface-modified clay filler with vinyl functional surface 15
modification)
Sulfur (MC grade) 1.5
DPG (curative) 0.15
MBTS (curative) 1.5
Rhenogran MPTD70 (curative) 0.29
Rhenofit PAN (antidegradant) 1
Wingstay 29 [mixture of styrenated diphenylamines (antidegradant)]
®
1.43
HAF: High abrasion furnace
MPTD: Dimethyl diphenyl thiuram disulfide
PAN: N-phenyl-1-naphthylamine
Reproduced with permission from M.J Forrest in Proceedings of Extractables & Leachables
for Pharmaceutical Products, 14–15th September, London, UK, Smithers Rapra Technology
Ltd, Shawbury, UK, 2010, Paper No.1. ©2010, Smithers Rapra [14]

The NBR compound in Table 9.4 is a traditional one that meets European food-contact
legislation with respect to its composition and could be used to produce gaskets and
seals for pharmaceutical packaging and medical devices although, as mentioned in
Section 9.3, these industries have started to move away from compounds like these
in recent years due to their desire to use cleaner rubbers. However, due to its relative
complexity, it provides a useful illustration of the extractables data that can be
obtained using a combination of GC–MS and LC–MS.

9.4.2 Potential Extractables

Given the ingredients in the NBR formulation (Table 9.4, Section 9.4.1), and
considering the theoretical background to the origins of extractable species in rubber
compounds presented in Section 9.2, the potentially extractables species, including
reaction products and breakdown products, that might be detected by a combination
of GC–MS and LC–MS are shown in Table 9.5.

294
 Extractable and Leachable Studies on Rubber Products

Table 9.5 Possible extractable species from the cured NBR compound that
could be detected by a combination of GC–MS and LC–MS
Ingredient Description Potential extractable species
Breon N36C60 NBR polymer ACN monomer, oligomers and dodecenes
Stearic acid Activator Mixture of carboxylic acids (C14, C16 and C18)
Amides by reaction with amines listed within this table
DPG Accelerator Unchanged ingredient
N,N'-diphenylthiourea
Trisphenyl amino-1,3,5-triazine
Diphenylcarbodiimide
Aniline
Ammonia
N-phenyl-N-methyl-N'-phenylthiourea
N,N'-diphenylurea
Diphenylamine
Phenyl isothiocyanate
Phenyl isocyanate
MBTS Accelerator Unchanged ingredient
MBT
Mercaptobenzothiazole zinc salt
Dibenzothiazyl monosulfide
Benzothiazole (BT)
Hydrogen sulfide
Aniline
Carbon disulfide
Phenyl mercaptan
MPTD Accelerator Unchanged ingredient
Methylphenyldithiocarbamic acid
methylphenylammonium salt
Methylphenyldithiocarbamic acid zinc salt
N-methylaniline
Carbon disulfide
Carbonyl sulfide
Sym-dimethyldiphenylthiourea
N-methyldiphenylamine
N,N-dimethylaniline
Methyl isothiocyanate
Phenyl isothiocyanate
N-nitroso-N-methylamine
2-Cyanoethyl methylphenyl-dithiocarbamate
PAN Antidegradant Unchanged ingredient
1,2-Naphthoquinone-1-anil

295
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Styrenated Antidegradant Unchanged ingredient

diphenylamines Monostyrenated diphenylamine
Distyrenated diphenylamine-N-oxide
Diphenylamine
Aniline
Styrene
Note: the full chemical names for the acronyms of the ingredients are shown in Table 9.4
(Section 9.4.1)
Reproduced with permission from M.J Forrest in Proceedings of Extractables & Leachables
for Pharmaceutical Products, 14–15th September, London, UK, Smithers Rapra Technology
Ltd, Shawbury, UK, 2010, Paper No.1. ©2010, Smithers Rapra [14]

9.4.3 Gas Chromatography–Mass Spectrometry Analysis of the Nitrile

Rubber Sample

As mentioned in Section 9.3, it is common practice to use both headspace GC–MS

and solution injection GC–MS to ensure as wide a range of extractable substances is
detected as possible. Once a cured test sheet of the NBR compound in Table 9.4 was
produced, samples for GC–MS analysis were prepared as shown below.

9.4.3.1 Headspace Gas Chromatography–Mass Spectrometry

A small amount (e.g., 0.2 g) of the sample was cut finely, placed into a headspace vial
and then sealed. The vial was heated at 150 °C for ≈30 min and then the headspace
injected into the GC–MS.

9.4.3.2 Solution Injection Gas Chromatography–Mass Spectrometry

A small amount of the sample (e.g., 0.3 g) was cut finely, placed into a vial and 2 ml
of acetone added. The vial was then placed into an ultrasonic bath and agitated for
30 min.

The headspace GC–MS chromatogram and solution injection (acetone extract)

GC–MS chromatogram for the vulcanised NBR sample are shown in Figures 9.3
and 9.4, respectively.

296
 Extractable and Leachable Studies on Rubber Products

20-Dec-2003 ATD Rubber 351N A83097 (10mins@150°C)

j0329-029
RB
100 Scan EI+
TIC
3.76e6

15.92
%

18.20
9.52

14.54
15.62

17.49
10.61

13.72
11.65
1.65

0 Time
2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00

Figure 9.3 Headspace GC–MS chromatogram for the NBR sample (ATD:
automated thermal desorption; TIC: total ion current). Reproduced with
permission from M.J Forrest in Proceedings of Extractables & Leachables for
Pharmaceutical Products, 14–15th September, London, UK, Smithers Rapra
Technology Ltd, Shawbury, UK, 2010, Paper No.1. ©2010, Smithers Rapra [14]

Abundance
TIC: 0804014.D

1.6e+07

1.4e+07

1.2e+07

1e+07

8000000

6000000

4000000

2000000

0
4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00
Time-->

Figure 9.4 Solution injection (acetone extract) GC–MS chromatogram for the
NBR sample. Reproduced with permission from M.J Forrest in Proceedings of
Extractables & Leachables for Pharmaceutical Products, 14–15th September,
London, UK, Smithers Rapra Technology Ltd, Shawbury, UK, 2010, Paper No.1.
©2010, Smithers Rapra [14]

297
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

9.4.4 Analysis of the Nitrile Rubber Sample by Liquid Chromatography–

Mass Spectrometry

An extract of the NBR sample was obtained using a solvent (methanol) that was
compatible with the mobile phase (methanol/water gradient) of the LC–MS and
injected into the instrument. As mentioned in Section 9.3, the use of LC–MS
compliments GC–MS in that species which are thermally labile (e.g., MBTS) and/
or of a high MW (e.g., species in oligomeric antidegradants like Wingstay® 29) can
be identified. Because of the lack of commercial LC–MS libraries, identification was
based on the analysis of standards and an understanding of the chemical processes
(e.g., formation of adduct ions with the mobile phase) that occur within the mass
spectrometer source.

The LC–MS chromatogram for the methanol extract of the NBR sample is shown
in Figure 9.5.

MSD2 TIC, MS File (D:\J0329\48030058.D) APCI, Neg, Scan, Frag: 150, "neg, APCI 100-1250"

9000000

8000000

7000000

6000000

5000000

4000000

3000000

2000000

1000000

2 4 6 8 10 12 14 16 18 min

Figure 9.5 LC–MS chromatogram for the methanol extract of the NBR sample.
Reproduced with permission from M.J. Forrest in Proceedings of Extractables
& Leachables for Pharmaceutical Products, 14–15th September, London, UK,
Smithers Rapra Technology Ltd, Shawbury, UK, 2010, Paper No.1. ©2010,
Smithers Rapra [14]

298
 Extractable and Leachable Studies on Rubber Products

9.4.5 Summery of the Extractables Detected by Gas Chromatography–

Mass Spectrometry and Liquid Chromatography–Mass Spectrometry
Analysis

The data obtained on the ‘extractables’ from the NBR sample by both GC–MS and
LC–MS revealed the presence of three main types of diagnostic (i.e., of traceable
origin and hence useful) species:

• Original compounding ingredients in the unchanged form.

• Impurities and substances from compounding ingredients.

• Breakdown products and reaction products.

To illustrate these results, some selected examples of these three categories are shown
below:

1. Examples of the original ingredients included:

• MBTS.

• PAN.

• Styrenated diphenylamines (adducts of 1:1, 2:1 and 3:1 stoichiometry).

2. Examples of impurities in the original ingredients included:

• ACN (residual monomer in the NBR polymer).

• Vinyl cylohexene and cyanocyclohexene (polymerisation intermediates).

• Alkylbenzenes (possible impurities in the CB).

•  -Octanethiol/dioctyl disulfide (polymerisation residues – substances used to

1
control MW).

3. Examples of breakdown products/reaction products included:

• Aniline.

• N-phenylpalmitamide (aniline/C15CO2H reaction product).

• N-phenylstearamide (aniline/C17CO2H reaction product).

• BT.

• Carbon disulfide.

299
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Diphenylamine.

• Dodecenes.

• 2-Mercaptobenzothiazole.

• N-methylaniline.

Not all of the theoretically predicted breakdown products (Section 9.4.2) were found
by GC–MS and LC–MS. It may be that some others were present in amounts below
the detection limits of the GC–MS and LC–MS analyses, or that the conditions for
their formation were not favourable for this particular NBR compound with its
particular thermal history.

In some cases, species were seen that are recognised derivatives of some of the
predicted breakdown products. One example is the 1:1 condensation product of
an aromatic primary amine (ArNH2) with the solvent that was used to prepare the
extract, acetone, i.e.:

ArNH2 + O = CMe2 → ArN = CMe2 + H2O (9.6)

The actual species that was detected in this case was isopropylidene aniline, which
is also called acetone ‘anil’.

As is normal with this type of analytical work, a number of non-diagnostic, commonly

detected species, were present in the extractables data (e.g., aliphatic hydrocarbons)
that were of uncertain origins, hence the term ‘non-diagnostic’.

One point worth mentioning is that this is essentially a trace component analysis, and
the presence of components from external sources (e.g., vulcanisation fume species
from other rubber samples being cured and/or stored nearby) in the assignment list
should not be surprising. Thus, while the low-MW species within the rubber may
have an impact on the environment around the rubber, that environment may also
have an impact on the low-MW species within the rubber.

As mentioned in Section 9.3, it would be common practice during an extractables

study to use other analytical techniques in order to provide a fuller picture of the
substances that could potentially leach into a medicinal product during storage or in
service. Such additional work would include:

300
 Extractable and Leachable Studies on Rubber Products

• Semi-quantitative multi-element (e.g., 36 element) analysis of the extraction media

by ICP.

• Determination of the NVR of the extraction media by quantitative drying down

in pre-weighed glassware (e.g., a petri-dish).

• Analysis of the NVR residue obtained above by FTIR. This could be carried out
using transmission or attenuated total reflectance techniques or, if insufficient
material was available for these approaches, an FTIR microscope would be used.

9.5 Determination of Extractables and Leachables

This section contains examples of E&L investigations and studies on rubber products
and samples that have been published. It also contains articles and other publications
that illustrate how the analytical techniques presented in Table 9.3, and discussed in
Section 9.3, have been shown to be of great value in the determination of low-MW
substances from rubber samples and so confirms their relevance to E&L work.

Ruberto [47] described how carrying out a comprehensive materials assessment

prior to conducting an E&L study can be a useful tool to predict migrants, design
a suitable controlled extraction study, facilitate data analysis, verify vendor supply,
and evaluate supply chain integrity. Also included in this presentation made to the
2012 E&L Conference were typical extractables profiles that can be obtained from
plastics and elastomers, and sources of unexpected migrants.

The PQRI publications and guidance documents for E&L work were introduced in
Sections 9.1 and 9.3. Researchers within the pharmaceutical industry and associated
networks have commented on how these relate to specific commercial situations. For
example, Ball and co-workers [48] discussed the derivation and application of the
PQRI safety thresholds for leachables in orally-inhaled nasal drug products. In their
study, experimental data was included, the conversion of the thresholds to analytical
thresholds in an E&L characterisation process described, and a summary of best
practice recommendations provided.

Both GC×GC–ToF–MS and LC–MSxMS have been shown to be very useful in a

number of publications, including those produced by Forrest and co-workers [49,
50] for the characterisation of low-MW constituents in rubber products. Their
strengths in this area, particularly when they are combined, mean that they can make
a significant contribution to E&L studies. Weil [51] of the LC–MS manufacturer
Agilent Technologies discussed the impact of mobile phase, chromatography,
ionisation mode (i.e., EI, APCI and atmospheric pressure photoionisation), and
the MS interface on data obtained during E&L investigations. Also included in

301
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

the study was specific information concerning the following aspects of the LC–MS
analytical system:

• Effect of mobile phase buffers on ionisation and in-source fragmentation.

• Influence of ion sampling on detection limits.

• Suitability of automated data mining with statistical analysis as a simplified E&L

differential analysis and multiple class comparison method.

Christiaens [52] considered the use of water for injection as an extraction solvent
as well as the relationship between E&E studies and physico-chemical compendial
tests. Also included within the presentation to the E&L for Pharmaceutical Products
Conference in 2009 were data generated by GC–MS studies and, in relation to rubber
samples, the results of a comparison between the sample preparation conditions in
compendial tests.

Christiaens [53] also determined the amount of halobutyl and non-halogenated

oligomers in pharmaceutical rubbers. The origin and formation of these species
was examined and as well as the differences between their postulated chemical
structures and actual structures. With respect to the determination of these
substances, the analytical techniques and methods that could be used for their
detection in E&L studies were described, and consideration was given to the toxicity
of these substances and the need for analytical standards to assist with detection
and validation work.

Kirchmeyer [54] carried out an in-depth analysis of the inorganic extractables from
rubbers intended for pharmaceutical applications in response to the publication of
the ICH Q3D ‘Guideline for Metal Impurities’ and the announcement that USP 232
‘Elemental Impurities – Limits’ and USP 233 ‘Elemental Impurities – Procedures’
will replace USP 231 ‘Heavy Metals’. For the analytical work, atomic absorption
spectrometry (zinc only), ICP–OES and ICP–MS were used and testing was carried
out on water extracts from a range of starting materials (e.g., CaCO3, talc, calcined
clay, zinc oxide and silica) and rubber compounds that contained these ingredients.
The results obtained using standard elemental test methods (e.g., International
Organizations for Standardizations’ ISO 8871-1:2003) demonstrated the ability
of ICP–OES and ICP–MS to determine the inorganic E&L from raw materials and
rubber compounds, and so could be used for the testing of rubbers intended for
pharmaceutical applications. The results also demonstrated how the washing process
used for the rubber samples could be responsible for increasing the concentration of
some water-soluble substances and that brominated butyl rubber samples had very
low levels of the elements listed in USP 232.

302
 Extractable and Leachable Studies on Rubber Products

A number of authors have reported on E&L work carried out on rubber products
used in the packaging of different types of pharmaceutical products. For example,
Nielsen [55] presented at paper at the 2015 E&L Conference which addressed the
identification of leachables in a topical product during its development using, as a
case study, an ointment in a prototype packaging composed of polymers and rubbers.
In the case of ophthalmic drug products, Houston [56] in a paper to the 2016 E&L
USA Conference addressed the challenges, important study parameters, key variables
and potential pitfalls in ophthalmic drug product E&L testing and looked at how
assessments of product packaging changes could be conducted in an effective and
time-efficient manner. Akapo [57] described a systematic approach to identify the
E&L from three polymeric components used in the dispensing device for aqueous-
based parenteral medications:

• Chlorobutyl rubber (vial stopper)

• COC plastic (syringe barrel)
• Bromobutyl rubber (stopper/tip cap)

The study also provided a discussion on the estimation of AET for the three
materials, stated the results that were calculated in these cases, namely 0.15 µg/vial
for the vial stopper and 0.03 µg/syringe for the syringe components (i.e., barrel, tip
cap and plunger stopper), and provided details of the analytical methods that had
been developed to monitor the levels of the leachables. Also, in regard to parenteral
products, Janssen [58] reported on the impact that the extraction conditions and time
had on the extractable results obtained on a chlorobutyl rubber compound. His study
assessed the influence of a number of the variables, including:

• Surface area of the rubber sample.

• Extraction technique (e.g., reflux, Soxhlet, ultrasonication).
• Extraction time (0.5 to 8.0 h).
• Extraction solvent (isopropanol, water at pH 11 and water at pH 3).

A range of analytical techniques, similar to that shown in Table 9.3 (Section 9.3),
was used to identify and quantify the extractable species. The results showed that
extending the extraction time (e.g., from 1 to 8 h) had an influence on the quantitative
data, but with evidence that the 8-h period was still not exhaustive. With respect to
qualitative data, extending the extraction time was found to have less impact, with
a similar set of extractables being detected.

The materials involved in the manufacture of e-cigarettes have been the subject of
E&L testing. Kesselring [59] provided an overview of the regulatory requirements for

303
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

medical devices together with a brief overview of their critical parts and e-cigarettes.
In his study, test results for e-cigarettes extracted with solvent and gas were provided,
as were data for medical devices made from silicone, SBR, polyurethane, polyethylene
and polyvinyl chloride. Finally, safety concerns over e-cigarettes were considered.

Section 9.1 provided a definition of leachables and this presented the possibility
that a sub-set of substances could be generated due to a reaction between leachables
from the contact material and substances within the drug product. One of the
scenarios that could allow this type of event to occur is when thiuram disulfide-type
accelerators are used in the sulfur cure system for rubber stoppers. By undergoing
thiol–disulfide exchange reactions, such compounds can undergo rapid reactions
between themselves and thiols to yield a mixture of disulfide products. Corredor
and co-workers [60] investigated the possibility that thiuram disulfides from rubber
stoppers undergo exchange reactions with drug products containing pendant thiol
groups. The results of their investigation enabled them to report the formation and
identification of mixed thiuram disulfides of TMTD, TBTD and captopril (a thiol-
containing drug). To detect these substances they developed a reversed-phase LC–MS
method that utilised an ODS-AQ column (from YMC Co Ltd.) at 35 °C in a mobile
phase consisting of a) (acetonitrile 20: water 80: trifluoroacetic acid: 0.1) and b)
(acetonitrile 100: trifluoroacetic acid: 0.1). MS using molecular ions at 420.9 and
337.1 m/z, respectively, identified captopril–TMTD and captopril–TBTD disulfides.
Possible structures for these compounds were provided and pH was found to have
an effect on their formation, with the captopril–TMTD disulfide being enhanced at
pH 6.0 and reaching a maximum concentration after 4.1 h. At lower pH values (e.g.,
4.0 and 2.2), the mixed captopril adduct product was detected in solution after 20 h.

Rushing [61] also reported on the E&L profile that can be obtained from a rubber
stopper, in this case in combination with a glass vial, and related to the early phase
of the development of drug products. The study also covered the regulatory questions
and actions that need to be addressed and undertaken, and the considerations that
need to be included in the design of an E&L study.

The trend away from the use of traditional rubbers that contain accelerated sulfur cure
systems (e.g., NR, NBR, EPDM) for the manufacture of pharmaceutical packaging
components (e.g., stoppers) was introduced in Section 9.3. Wong of ExxonMobil
Chemical Company [62] emphasised this trend to improve cleanliness in the industry
due to the increasing awareness of the potential impact that impurities have on
drugs (e.g., deactivation) and, hence, the health of patients. As a consequence,
the level of E&L components has become an important parameter with regard to
pharmaceutical stoppers being qualified. Wong pointed out that the cleanliness of the
stoppers themselves was determined by the manufacturing process, the cleanliness of
each ingredient within the rubber compound (i.e., its purity), and various residual

304
 Extractable and Leachable Studies on Rubber Products

ingredients (e.g., curatives), and the reaction products and breakdown products that
can be generated during processing, curing and ageing. Bromobutyl can be effectively
cured via low levels of sulfur-free and/or zinc-free curing systems, so it is becoming
a very popular rubber for pharmaceutical stoppers and seals. Specific examples of
clean curing systems for bromobutyl rubbers were provided in the study and the
extractables profiles of cured samples produced from these compounds described.

In addition to being considered for individual container/drug combinations,

information on leachables are also of considerable importance when it comes to the
items and components within a manufacturing site. Wilhelm [63] reported on the scope
of the leachables project at Roche’s Penzberg site, the leachables assessment strategy,
the opportunities for assessing a large-scale production site, and laboratory testing.
In addition, the study, which was presented at 2015 E&L Europe Conference, also
included the leachable results obtained for pumps, flow meters, valve membranes,
tubing systems and chromatography systems.

Cecchini [64] provided an introduction to the regulatory expectations with regard to

extractables from the polymeric materials used in the storage of drug products or drug
delivery systems. A number of other subjects were also covered in the study, including:

• An overview of biotech product manufacturing.

• The ‘decision tree’ used in the initiation of an E&L evaluation.

• An E&L study for a developmental product.

• Case studies involving an extraction study of the contact materials for a

downstream processing product; an E&L study on a drug substance/active
pharmaceutical ingredient storage container; and an extraction study on F&F
product-contact materials.

There have been recent developments regarding E&L in the biopharmaceutical

industry and these have been reviewed by Menzel [65] in a presentation that dealt
with the BioPhorum Operations Group’s (BPOG) best practice guide [66] BPOG
‘Standardised Extractables Testing Protocol, Pharmaceutical Engineering (2014)'
for leachables that might originate from single-use manufacturing components. This
best practice guide included risk assessment using a standard model approach, the
design of a leachable study, the test methods that should be employed and applicable
analytical considerations.

Single-use systems (SUS) are used increasingly in bio-manufacturing processes due

to their multitude of advantages. However, one of the main concerns for their use
are their E&L profile and this was addressed by Ding [67] of Pall Life Sciences. As
a SUS provider, Pall Life Sciences have systemically studied the extractables from

305
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

single-use components, including filters, tubing, connectors and bio-containers. A

study presented by Ding summarised the test design and strategy employed by the
company. This work enabled a library of extractable substances to be generated and
this facilitated product-specific E&L evaluation during process validation work.
Martin [68] reported on the regulatory requirements for disposable SUS for the
manufacture of bio-pharmaceuticals and pharmaceuticals in the US and Europe and
outlined the recommended approach of Bio-Process Systems Alliance for suppliers
and users to address E&L from single-use process contact equipment.

References

1. M. Creese, Pharmaceutical Manufacturing and Packing Sourcer, 2012,

February, 30.

2. Guidance for Industry: Elemental Impurities in Drug Products: Guidance for

Industry, Food and Drug Administration, Rockville, MD, USA, 2016.
https://www.fda.gov/downloads/Drugs/Guidances/UCM509432.pdf

3. Product Quality Research Institute in Thresholds and Best Practices for

Extractables and Leachables in Proceedings of the 3rd PQRI/FDA Conference
on Advancing Product Quality, 22nd March, Washington, DC, USA, 2017.

4. USP 38–NF33 – General chapters on Extractables and leachables.

5. USP 1663 – Assessment of extractables associated with pharmaceutical

packaging/delivery systems.

6. USP 1664 – Assessment of drug product leachables associated with

pharmaceutical/packaging delivery systems.

7. ICH Guideline for Elemental Impurities Q3D, ICH Harmonised Guideline,

2014.

8. M.N. Eakins in Proceedings of Extractables & Leachables Europe

Conference, 10–12th November, London, UK, Smithers Rapra Technology
Ltd, Shawbury, UK, 2015, Paper No.3.

9. J. Bones in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra, Akron, OH, USA, 2016,
Paper No.14.

306
 Extractable and Leachable Studies on Rubber Products

10. D.J. Ball in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 13–14th May, Barcelona, Spain, Smithers Rapra
Technology Ltd, Shawbury, UK, 2009, Paper No.7.

11. A. Feilden in Proceedings of Extractables & Leachables USA Conference,

15–17th May, Baltimore, MD, USA, Smithers Rapra, Akron, OH, USA, 2012,
Paper No.1.

12. A. Feilden in Update on Undertaking Extractable and Leachable Testing,

Smithers Rapra Technology Ltd, Shawbury, UK, 2011.

13. S. Bailey in Proceedings of Extractables & Leachables Conference,

5–6th March, Dublin, Ireland, Smithers Rapra Technology Ltd, Shawbury,
UK, 2008, Paper No.14.

14. M.J. Forrest in Proceedings of Extractables & Leachables for Pharmaceutical

Products, 14–15th September, London, UK, Smithers Rapra Technology Ltd,
Shawbury, UK, 2010, Paper No.1.

15. T. Hulme and M.J. Forrest in Proceedings of Extractables & Leachables for
Pharmaceutical Products Conference, 27–28th September, Dublin, Ireland,
Smithers Rapra Technology Ltd, Shawbury, UK, 2011, Paper No.1.

16. D.L Norwood, M. Ruberto, F.L. DeGrazio, J. Castner and W. Boehringer,

Rubber and Plastics News, 2009, 38, 24, 15.

17. J. Toynbee in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 14–15th September 2010, London, UK, Smithers Rapra
Technology Ltd, Shawbury, UK, 2010, Paper No.19.

18. R. Janssen in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 27–28th September 2011, Dublin, Ireland,
Smithers Rapra Technology Ltd, Shawbury, UK, 2011, Paper No.10.

19. K.G. Ashness, G. Lawson, R.E. Wetton and B.G. Willoughby, Plastics and
Rubber Processing and Applications, 1984, 4, 69.

20. K. J. Saunder in Organic Polymer Chemistry, Chapman and Hall, London,

UK, 1973.

21. R.C. Crafts, J.E. Davey, G.P. McSweeney and I.S. Stevens, Journal of National
Rubber Research, 1990, 5, 275.

307
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

22. B.G. Willoughby in Health and Safety, in Developments in Rubber Technology

4, Elsevier Applied Science, London, UK, 1987.

23. A.G. Davies in Organic Peroxides, Butterworth’s London, London, UK,

1961.

24. V.A. Shershnev, Rubber Chemistry and Technology, 1982, 55, 537.

25. B.G. Willoughby in Proceedings of the Hazards in the European Rubber

Industry Conference, Shawbury, 28–29th September, Rapra Technology Ltd,
Shawbury, UK, 1999.

26. M.J. Forrest and B. Willoughby in Proceedings of the RubberChem

Conference, 9–10th November, Birmingham, UK, Rapra Technology Ltd,
Shawbury, UK, 2004, Paper No.1.

27. A.Y. Coran in Vulcanisation, in Science and Technology of Rubber, Ed.,

F.R. Eirich, Academic Press, New York, NY, USA,1978.

28. N. Grassie and G. Scott in Polymer Degradation and Stabilisation,

Cambridge University Press, Cambridge, UK, 1985.

29. F. Pilati, S. Masoni and C. Berti, Polymer Communications, 1985, 26, 280.

30. J. Pospisil in Developments in Polymer Stabilisation – 7, Elsevier Applied

Science, London, UK, 1984.

31. B.G. Willoughby and K.W. Scott, Rubber Chemistry & Technology, 1998,
71, 310.

32. R. Hawkins in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 14–15th September, London, UK, Smithers Rapra
Technology Ltd, Shawbury, UK, 2010, Paper No.2.

33. R. Hawkins in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 13–14th May, Barcelona, Spain, Smithers Rapra
Technology Ltd, Shawbury, UK, 2009, Paper No.1.

34. T.W. Robison in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra, Akron, OH, USA, 2016,
Paper No.1.

35. W. Van Dongen, Flexible and Plastic Packaging Innovation News, 2008, 2, 5,
174.

308
 Extractable and Leachable Studies on Rubber Products

36. H. Schulz in Proceedings of Extractables and Leachables Conference,

5–6th March, Dublin, Ireland, Smithers Rapra Technology, Shawbury, UK,
2008, Paper No.9.

37. A. Feilden in Proceedings of Extractables and Leachables Conference,

5–6th March, Dublin, Ireland, Smithers Rapra Technology Ltd, Shawbury,
UK, 2008, Paper No.8.

38. A. Feilden in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 13–14th May, Barcelona, Spain, Smithers Rapra
Technology Ltd, Shawbury, UK, 2009, Paper No.5.

39. R. Varma, Rubber World, 2008, 238, 3, 17.

40. Anon, High-Barrier TPE for Medical Uses in Plastics Technology, 2007, 53,
8, 23.

41. British Plastics and Rubber, 2009, February, 22.

42. P.D. Tatarka in Proceedings of the 188th ACS Fall Technical Meeting,
13–15th October, Cleveland, OH, USA, ACS Rubber Division, Akron, OH,
USA, 2015, Paper No.47.

43. V. Jeanguyot in Proceedings of Extractables & Leachables for Pharmaceutical

Products Conference, 27–28th September, Smithers Rapra Technology Ltd,
Shawbury, UK, 2011, Paper No.14.

44. B. Cabovska and E. Riches in Proceedings of Extractables & Leachables

Europe Conference, 10–12th November, London, UK, Smithers Rapra
Technology Ltd, Shawbury, UK, 2015, Paper No.6.

45. B. Cabovska in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra Technology Ltd, Akron,
OH, USA, 2016, Paper No.17.

46. K. D’Silva in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra Technology, Akron, OH,
USA, 2016, Paper No.16.

47. M.A. Ruberto in Proceedings of Extractables & Leachables USA Conference,

15–17th May, Baltimore, MD, USA, Smithers Rapra, Akron, OH, USA, 2012,
Paper No.8.

309
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

48. D.J. Ball, D.L. Norwood and L. Nagao in Proceedings of Extractables

& Leachables Conference, 5–6th March, Dublin, Ireland, Smithers Rapra
Technology Ltd, Shawbury, UK, 2008, Paper No.3.

49. M. Forrest, S. Holding and D. Howells, Polymer Testing, 2006, 25, 63.

50. M.J. Forrest, S.R. Holding, D. Howells and M. Eardley in Proceedings of

Silicone Elastomers, 19–20th September, Frankfurt, Germany, Smithers Rapra
Technology, Shawbury, UK, 2006.

51. D.A. Weil in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra Technology, Akron, OH,
USA, 2016, Paper No.15.

52. P. Christiaens in Proceedings of Extractables and Leachables for

Pharmaceutical Products Conference, 13–14th May, Barcelona, Spain,
Smithers Rapra Technology, Shawbury, UK, 2009, Paper No.10.

53. P. Christiaens in Proceedings of Extractables & Leachables Conference,

12–13th December, Vienna, Austria, Smithers Rapra Technology Ltd,
Shawbury, UK, 2012, Paper No.15.

54. H. Kirchmeyer in Proceedings of Extractables & Leachables Conference,

12–13th December, Vienna, Austria, Smithers Rapra Technology Ltd,
Shawbury, UK, 2012, Paper No.14.

55. L.D. Nielsen in Proceedings of Extractables & Leachables USA Conference,

12–14th May, Bethesda, MD, USA, Smithers Rapra Technology Ltd, Akron,
OH, USA, 2015, Paper No.8.

56. C.T. Houston in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra, Akron, OH, USA, 2016,
Paper No.11.

57. S.O. Akapo in Proceedings of Extractables & Leachables USA Conference,

7–9th May, Providence, Rhode Island, USA, Smithers Rapra Technology,
Akron, USA, 2013, Paper No.10.

58. R. Janssen in Proceedings in Extractables & Leachables Europe,

10–11th December, Smithers Rapra Technology Ltd, Shawbury, UK, 2013.

59. A. Kesselring in Proceedings of Extractables & Leachables USA Conference,

9–11th May, Bethesda, MD, USA, Smithers Rapra, Akron, OH, USA, 2016,
Paper No.4.

310
 Extractable and Leachable Studies on Rubber Products

60. C. Corredor, F.P. Tomasella and J. Young, Journal of Chromatography A,

2009, 1216, 1, 43.

61. W. Rushing in Proceedings of Extractables & Leachables USA Conference,

12–14th May, Besthesda, MD, USA, Smithers Rapra Technology Ltd, Akron,
OH, USA, 2015, Paper No.11.

62. W.K. Wong in Proceedings of Extractables and Leachables for

Pharmaceutical Products Conference, 14–15th September, London, UK,
Smithers Rapra Technology Ltd, Shawbury, UK, 2010, Paper No.22.

63. J. Wilhelm in Proceedings of Extractables & Leachables Europe Conference,

10–12th November, London, UK, Smithers Rapra Technology Limited,
Shawbury, UK, 2015, Paper No.9.

64. I. Cecchini in Proceedings of Extractables & Leachables Europe Conference,

10–12th November, London, UK, Smithers Rapra Technology, Shawbury, UK,
2015, Paper No.7.

65. C. Menzel in Proceedings of Extractables & Leachables Europe Conference,

10–12th November, London, UK, Smithers Rapra Technology Ltd, Shawbury,
UK, 2015, Paper No.10.

66. Standardised Extractables Testing Protocol, Pharmaceutical Engineering,

BioPhorum Operation Group, Sheffield, UK, 2014.

67. W. Ding in Proceedings of Extractables and Leachables for Pharmaceutical

Products Conference, 27–28th September, Dublin, Ireland, Smithers Rapra
Technology, Shawbury, UK, 2011, Paper No.12.

68. J. Martin in Proceedings of Extractables and Leachables for Pharmaceutical

Products Conference, 13–14th May, Barcelona, Spain, Smithers Rapra
Technology Ltd, Shawbury, UK, 2009, Paper No.3.

311
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

312
10
Analysis of Surface Blooms and
Contaminants

10.1 Introduction

Rubber products and compounds can fail in service, or prior to service, for a number
of reasons and some of these have been covered by other sections in this book, such
as incorrect composition (Chapters 3–5).

This section deals specifically with two other important reasons for failure:
1. Surface blooms
2. Contamination

The first of these has caused problems for rubber technologists since the beginning
of the industry and they can range from a displeasing aesthetic appearance to serious
product failure. An example of the latter is when a surface bloom interferes with the
rubber bonding to another substrate, such as metal. The second can by its nature take
many forms and the form that the contamination is in (e.g., liquid, solid particle, fibre)
will have a significant influence on the analytical approach and instrumentation that
will be employed to identify it and, hopefully, by doing so, trace its source.

Section 10.1 provides an overview of blooming on the surface of rubbers, covering

aspects of its science, such as the reasons for its formation, to the analytical techniques
and methodologies that have been found to be the most effective in characterising
it. Some ways in which blooms can be avoided are also presented in this section.
Section 10.2 provides examples of the most commonly encountered contaminants
and the ways in which they can be identified, both in situ and after removal from the
rubber compounding ingredient, compound or product.

10.2 Surface Blooms

10.2.1 Introduction

The appearance of a bloom on the surface of a rubber product or compound is a

313
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

relatively common phenomenon that can lead to a number of problems. It is often,

therefore, the subject of an analytical investigation in order to obtain information
on its identity and, hence, the possible reason(s) for its appearance. To facilitate
this investigation, as with most areas of rubber analysis, it is useful to have an
understanding of what causes blooming, the substances that may be present in a
bloom, and the reasons for their formation. Therefore, in addition to describing the
analytical techniques and methods that have been used to characterise blooms, this
section also provides a brief overview of the background and chemistry associated
with their formation.

It is useful to start with a definition and one, relatively narrow, definition of a surface
bloom is ‘Formation on the surface of an uncured or cured rubber of a thin coating
of a compounding ingredient which has limited solubility in the rubber matrix’.

As will be seen during the course of this section, this is a narrow definition for a
number of reasons, one of which is that a bloom can be a mixture of compounding
ingredients, or a reaction product of a compounding ingredient.

A bloom can appear during storage or service, and it can be an aesthetic problem
which effectively categorises the product as a ‘failure’ as much as other deficiencies,
such as a poor cure state or air blisters. Blooming can also cause bonding problems
in composite systems (e.g., rubber to metal) by interfering with the wetting of the
adhesive, and it can cause other problems, such as contaminating a nearby surface
or altering the profile of an extractable and leachables investigation (Chapter 9).

In general, for a substance to bloom it must have a relatively low degree of solubility
in a rubber matrix, and be present in excess of that solubility. This excess will be
present as discrete particles dispersed throughout the matrix and the molecules within
these particles will migrate (i.e., diffuse) to the surface of the rubber. This migration
is dependent upon the:

• Overall polarity of the rubber matrix

• Solubility of the substance in the rubber

• Concentration of the substance in the rubber

• Temperature

• Time

At a given temperature, the solubility of a substance is influenced by its molecular

weight (MW) and chemical structure. Also, in some cases, as the temperature drops
the solubility of a substance will be reduced and the occurrence of a bloom made more

314
 Analysis of Surface Blooms and Contaminants

likely. Typical examples of some of the rubber ingredients that have been known to
bloom are hydrocarbon waxes and antidegradants, and other examples are provided
in Section 10.2.2.1.

Blooming is not always undesirable. For example, the blooming of paraffinic waxes
onto the surface of a rubber product provides a physical barrier to ozone attack
and assists in the migration of antidegradants to the surface, where they can protect
the rubber from ozone and oxygen. It is also possible for reaction products and/or
breakdown products of rubber ingredients such as peroxide curatives, accelerators
or antidegradants to bloom to the surface of vulcanised products.

Some examples of a white surface bloom on rubber components are shown in

Figures 10.1–10.3. Figure 10.3 illustrates how mechanical pressure, in this case due
to a finger, can cause or exacerbate a bloom due to a ‘seeding’ effect. Mechanical
pressure may also result in localised reactions that can cause a bloom.

Figure 10.1 Example of a white bloom on the surface of a cured rubber product

315
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Figure 10.2 Example of a white bloom on the surface of uncured rubber extrudate

Figure 10.3 Illustration of how mechanical pressure (finger impression) can

encourage a bloom, in this case a white bloom, to appear on the surface of a cured
rubber product

316
 Analysis of Surface Blooms and Contaminants

In addition to ‘true’ blooms which can be said to conform to the definition and
description above, there are also a number of other examples of bloom-like phenomena
and these are included in Section 10.2.2.

10.2.2 Different Types of Bloom or Bloom-like Phenomena

The different types of bloom or bloom-like phenomena can be classified and described
as shown below.

10.2.2.1 True Blooms

True blooms can occur if a substance within the rubber compound has limited
solubility in the rubber matrix and is present in amounts that exceed this solubility.
This substance can then migrate to the surface and crystallise out onto it when the
product cools down after the vulcanisation process. Examples of substances that can
bloom in this way include:

• Paraffinic waxes.

• Antioxidants (e.g., p-phenylenediamines).

• Accelerators [e.g., zinc diethyldithiocarbamate (ZDEC), 2-mercaptobenzothiazole].

• Elemental sulfur.

• Zinc salts of carboxylic acids (e.g., zinc stearate).

• Reaction products of sulfur cure systems (e.g., accelerators).

• Reaction products of other cure systems (e.g., peroxide cure systems).

• Retarders, lubricants and colourants.

Examples of this type of bloom may dissolve back into the rubber matrix if the
product is heated.

10.2.2.2 Modified Blooms

Modified blooms can result from the presence of a chemical substance in the rubber
matrix which reacts deliberately, or accidentally, with the environment. Examples of
substances which can create modified blooms include:

317
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• p-Phenylenediamine antidegradants if used in excess.

• Zinc stearate when in excess – creates an initial ‘oily’ bloom which can then
solidify if moisture is present in the atmosphere.

10.2.2.3 Pseudo-blooms

Pseudo-blooms can occur if degradation of the rubber surface has occurred to expose
a filler (e.g., an inorganic filler) that has been incorporated into the rubber. This results
in the appearance of a pseudo-bloom which can look like ‘frosting’.

Inorganic fillers, such as calcium carbonate, are not soluble in the rubber matrix and
so cannot migrate to the surface and form true blooms, but their particles can appear
on the surface if the original rubber surface has been removed by degradation.

10.2.2.4 Surface Contamination

An undesirable deposit on the surface of a rubber component can appear due to a

number of possible events:

• Inorganic or organic substances in washing or rinsing solutions.

• Silicone-release agents giving a greasy bloom which attract airborne materials.

• Airborne dusts or mists in a factory atmosphere.

• Handling the product with contaminated gloves.

10.2.2.5 Staining or Discolouration

These effects can result if substances on the surface of rubber products react with
substances in other products or materials (e.g., textiles, plastics) that they come into
contact with. Examples include:

• Elemental sulfur or dithiocarbamates in the presence of copper or iron.

• Zinc oxide (ZnO) in the presence of lead.
• When copper or iron are present in a contact medium.
• Reactions of breakdown products of phenolic-type antioxidants.
• Reactions of breakdown products of p-phenylenediamines.

318
 Analysis of Surface Blooms and Contaminants

10.2.2.6 Hazing

Hazing is rare and is due to the cloudy appearance within a transparent rubber
product, or at the surface of it, causing opacity. It can appear as a result of insoluble
solid particles or, rarely, droplets of a liquid. Examples include the incorrect grade
or level of ZnO, calcium oxide or other poorly dispersed ingredients.

10.2.3 Analysis and Identification of Blooms

10.2.3.1 Simple Investigative Tests

There are a number of simple tests that can be applied to a rubber to obtain some
information on the substance(s) that may be causing a bloom. Two examples of
such tests are those that involve a change in colour taking place on the surface of
the rubber, and the affect that heating a rubber product exhibiting a bloom has on
the appearance of that bloom. Examples of these two classes of test are listed below.

Surface colour change:

• Reactions occurring due to exposure to light:

• Formation of a grey/brown colour indicates an amine antioxidant.

• Formation of a pink colour indicates a phenolic-type antioxidant.

• Patchy and inconsistent colouration over the surface:

• I ndicates that rubber ingredients on the surface have reacted with chemicals
that they have contacted in the environment.

• Reactions occurring on contact with a metal:

•  opper and iron will react with dithiocarbamates on the surface resulting in
C
coloured salt products.

Heating the sample:

• Does the bloom disappear upon heating?

• Yes – this result indicates that it is a true bloom and the substances have re-
dissolved in the rubber.

•  o – this result indicates that it is a ‘modified bloom’ or a pseudo-bloom due

N
to surface oxidation or the reaction of an ingredient on the surface.

319
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

10.2.3.2 In Situ Tests

One of the first approaches taken if a bloom needs to be investigated is to examine it

using optical microscopy. This initial assessment will enable some information on the
colour, structure, complexity, depth and overall nature of the bloom to be obtained.
The close-up photographic images of a white bloom shown in Figures 10.1–10.3
illustrates the results that can be achieved using this approach.

It is also possible to analyse rubber blooms in situ on the surface of a rubber compound
using various surface specific analytical techniques, but both Fourier-Transform
infrared (FTIR) spectroscopy and energy-dispersive X-ray spectroscopy (EDX) are
particularly useful, especially if used in combination.

With FTIR spectroscopy, an attenuated total reflectance (ATR) spectroscopy sampling

accessory can be used and, although an amount of the infrared (IR) radiation will
penetrate through the bloom into the surface of the rubber, it is possible to obtain a
spectrum. Unfortunately, the fact that the majority of rubber samples are carbon black-
filled (a substance that absorbs IR light and so reduces the strength of the reflected
response) is another limitation of this technique. Use of an ATR crystal that reduces
the penetration depth of the IR radiation (e.g., germanium crystal) can improve the
surface specificity of the technique and improve the quality of the data obtained.

An example of where ATR–FTIR has been used to identify a bloom is illustrated by

Figure 10.4. This work was conducted on a rubber glove exhibiting a bloom on its
surface. There are two ATR–FTIR spectra in Figure 10.4, the top one shows an area
of the glove which showed no bloom (‘control’ spectrum) and the one below is the
spectrum recorded of an area with the bloom (‘sample’ spectrum). The difference in
the IR data between the control and sample spectra are consistent for the accelerator
ZDEC and it is possible that this ingredient has bloomed to the surface because it
has exceeded its solubility within the rubber matrix.

Using the EDX technique will enable both electron micrographs of the bloom to be
obtained, which will provide important pictorial images of the bloom that are useful
for assessing its structure (e.g., evidence for crystals), and semi-quantitative elemental
data. When carrying out this work, a baseline determination from a clean, freshly
cut rubber surface is obtained in order to gauge the degree of enhancement of each
element within the area of the bloom.

320
 Analysis of Surface Blooms and Contaminants

Transmittance (%)

4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm-1)
Transmittance (%)

4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm-1)

Figure 10.4 ATR–FTIR spectra of glove surfaces. a) control surface (above) and b)
surface with ZDEC bloom (below)

Table 10.1 shows the EDX data obtained on the yellow areas of bloom present on
the surface of a moulded rubber component. Duplicate sets of data are displayed for
both a freshly cut surface and the surface exhibiting the yellow bloom. These show a
significant enhancement in the response for sulfur in the surface with the bloom. This
relatively complex set of data also shows the importance of obtaining background
data to aid interpretation. The likely source(s) of the elements in Table 10.1 are:

• Carbon – base rubber and any organic ingredients.

• Oxygen – any ingredients that is an oxide or contains oxygen.

321
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

• Aluminium – possibly from a silicate-type filler.

• Silicon – possibly from a silicate-type filler.

• Titanium – probably from the pigment titanium dioxide.

• Zinc – probably from curing co-agent ZnO.

• Sulfur – elemental sulfur and any other compounds (e.g., accelerators) that contain
sulfur.

Table 10.1 EDX data obtained on the yellow areas of bloom present on the
surface of a moulded rubber component
Sample Semi-quantitative elemental data (%)
Carbon Oxygen Aluminium Silicon Sulfur Titanium Zinc
Freshly cut surface (1) 65.0 21.2 5.4 5.1 1.7 1.1 1.1
Freshly cut surface (2) 64.0 22.0 5.5 5.2 1.7 1.1 1.0
Yellow bloom (1) 46.0 2.7 0.5 1.0 49.2 0.1 0.5
Yellow bloom (2) 53.1 3.4 0.9 1.3 41.0 0.2 0.5

In the case of this particular bloom, the images provided by EDX were extremely
useful in identifying its cause. Two of the images obtained on the yellow bloom in situ
are shown in Figures 10.5 and 10.6. The close-up image in particular (Figure 10.6)
shows crystalline structures which, together with the discovery of sulfur (Table 10.1),
and its yellow colour, confirmed that the bloom on the moulded rubber component
was due to elemental sulfur.

Because it does not provide information on the actual compounds that the elements
in a bloom are present in, it is unlikely that EDX data alone will positively identify
the culprit species. As mentioned above, ATR–FTIR can encounter problems
due to absorption of IR radiation. Fortunately, there is a viable alternative to the
ATR–FTIR route: FTIR microspectrometry. Investigative work using this technique
can be conducted using two main approaches, with IR spectral information either
being obtained in situ using the reflectance mode (statically or in the ‘mapping’
configuration), or on a small amount of the bloom that has been removed from the
surface of the rubber. The latter, which can be regarded as an ‘indirect’ approach, is
dependent upon there being sufficient bloom material to function effectively but, due
to the sensitivity of the technique, this is not usually a problem.

322
 Analysis of Surface Blooms and Contaminants

600 µm Rubber disc 323/0753 Yellow Deposit

Figure 10.5 Scanning electron micrograph of a yellow bloom on the moulded

rubber component

With the indirect approach, a small amount of the bloom is carefully taken from
the surface of the rubber using a sharp instrument (e.g., scalpel tip), placed on the
window of the sampling accessory in the FTIR microscope, and a transmission FTIR
spectrum recorded. This FTIR spectrum, in combination with the EDX data that
have been obtained on the bloom, is usually very effective at characterising it, even
if more than one compound is present.

If problems of identification are still encountered, it is possible to remove the bloom

using a dry-swab technique, which avoids contamination of the bloom constituents
by solvent-soluble extractables in the rubber product, and the bloom material taken
off the swab using a general-purpose solvent (e.g., chloroform). Having prepared
a solution of the bloom, it is then possible to analyse it using a sensitive analytical
technique, such as gas chromatography (GC)–mass spectrometry (MS). Such an

323
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

approach has the benefit of utilising a chromatographic separation step and so this
aids the characterisation of complex blooms.

60 µm Rubber Disc 323/0753 Yellow Deposit

Figure 10.6 Close-up scanning electron micrograph of a yellow bloom on the

moulded rubber component

The three-dimensional (3D) GC×GC–time-of-flight (ToF)–MS chromatogram of

the species present in an ‘oily’ bloom removed from the surface of a diene rubber
sample is shown in Figure 10.7. The assignments obtained for the species present in
this chromatogram, together with its overall configuration, suggested that the bloom
was a hydrocarbon oil. This conclusion was supported by the data in the FTIR
spectrum obtained on the bloom (Figure 10.8) and its oily appearance on the surface
of the rubber. The data in Figure 10.8 also suggested that the bloom was due to a
naphthenic-type hydrocarbon oil. An oil of this type was one of the compounding
ingredients used in the formulation of this rubber product.

324
 Analysis of Surface Blooms and Contaminants

1400
3
RT
n
2nd 2 900 um
Co Col
lum st
nR 1
T 1

0 400

Figure 10.7 3D GC×GC–ToF–MS chromatogram of the species present in an ‘oily’

bloom on the surface of a diene rubber sample (RT: retention time)

98
96
94
92
90
88
86
Reflectance (%)

84
82
80
78
76
74
72
70
68
66
64
62
60
4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm )-1

Figure 10.8 FTIR spectrum of an oily bloom present on the surface of a diene
rubber sample

325
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Once a solution of the bloom constituents has been obtained, it is also possible to
analyse it using other chromatographic techniques, for example, high-performance
liquid chromatography (LC) or supercritical fluid chromatography. Also, on some
occasions, it might be considered informative to use ion chromatography to investigate
the possibility of salts (e.g., nitrates or sulfates) being present in a bloom.

Other analytical techniques capable of providing the structural and elemental data
of blooms that are in situ on the surface of rubber compounds and products include:

• Secondary-ion mass spectrometry (SIMS)

• Laser-ionisation mass analysis (LIMA)

If larger quantities of bloom are available, it is possible to use techniques that can
function with a couple of milligrams, such as differential scanning calorimetry
(DSC). Use of a thermal technique like DSC can be beneficial if it is suspected that
a bloom is composed of substances that have characteristic melting points, such as
microcystalline waxes.

As discussed above, elemental sulfur can be responsible for blooms and its poor
dispersion within a rubber compound can be a contributory factor. Jurkowski and co-
workers [1] studied the dispersion of sulfur in rubber compounds using radioisotopes.
They introduced the radioactive sulfur-35 isotope into compounds and measured the
radioactivity by volumetric and surface analytical methods. This approach enabled the
physical and chemical processes in the rubber to be investigated as well as the study
of sulfur blooming. The research team also obtained experimental data on the effect
that the mixing conditions and storage temperatures had on the generation of blooms.

10.2.3.3 Modelling of Blooming

In addition to chemical analytical work, researchers have also used modelling to

study the phenomena associated with blooming. An example of this work is that of
Hoei [2], who developed a general kinetic model for the blooming of crystallisable
additives in rubber and applied it to the analysis of the blooming of docosane in
natural rubber that had been crosslinked using dicumyl peroxide. Use of this model
enabled the cause of the blooming of docosame to be related to the melting of crystal
particles as a result of a localised depression in their melting point. A reduction in
elastic stress around the crystal particles was also believed to occur, and the data
generated by the model enabled an initial diffusion coefficient of the docosane during
blooming to be calculated.

326
 Analysis of Surface Blooms and Contaminants

10.3 Contaminants

10.3.1 Introduction

Contaminants can take many forms and can enter the product stream either in the
ingredients used in the production, or during one of the stages of manufacture (e.g.,
mixing, extruding, moulding, storage or transportation).

Rubber products can fail, both aesthetically and physically, because of particulate
contaminants and sometimes the physical failures can be catastrophic if the rubber
part is essentially in maintaining a seal, or part of a mechanical support apparatus.
This section will also cover contaminants that can become part of a rubber product in
service, either due to physical attachment or, in the case of liquids, by being absorbed
into the surface or throughout the bulk.

If the contaminants are in the form of solid, discreet entities (e.g., fibres), their detection
in the product and removal are relatively easy. However, if they are in the form of
absorbed liquids, their detection is sometimes dependent upon indirect methods, such
as seeing if a drop in hardness is apparent, and their removal can be more involved
(e.g., by volatilisation or extraction). The most commonly encountered contaminants
and the techniques that have been found to be the most effective at identifying them
are covered in Sections 10.3.2.1–10.3.2.3.

10.3.2 Identification of Different Types of Contaminant

10.3.2.1 Particulates

Whether they are on the surface of a rubber product, or below it and have to accessed
by sectioning, it is often possible to remove particulate contamination (e.g., using
tweezers or the end of a scalpel blade) and transfer it to an analytical instrument.

A technique that is ideal for the analysis of particulate contamination is FTIR

microscopy and its use, providing the particle is not black, will deliver a significant
amount of structural and qualitative information on its composition. This is often
enough to identify its source, but if the data are complex, indicating that a mixture of
substances may be present (e.g., two or more inorganic compounding ingredients), it
is useful to also apply EDX in order to provide complementary elemental information.
It is then often possible using the combined data to identify the contamination and
its source.

327
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Other techniques can be used to compliment FTIR and EDX data to characterise the
particulate matter more thoroughly:

• Optical microscopy with image capture

• Scanning electron microscopy (SEM)

• GC–MS

GC–MS can be very useful if a complex mixture of substances is present, but there
are obvious practical constraints and it can be used only if the particle is composed
of a soluble, organic substance (e.g., cure system or antidegradant system ingredient),
within a certain MW range (e.g., 35 to 400 Da), and if a sufficient mass is present
to enable detection.

If the particulate contamination is too small, or too well embedded in the product to
allow removal, then it is usually possible to analyse it in situ. This can be carried out
using a similar range of techniques to those shown above, although FTIR microscope
analyses would have to be carried out in reflectance mode and the particle would
have to have an area >10 um2 in order for good-quality spectra to be obtained. It
may also be possible to obtain structural information on the contaminant using SIMS
and LIMA, although these are very sensitive, surface-specific techniques and care has
to be taken on sample preparation and handling in order to avoid contamination.

If large particles are present, possibly in larger rubber products, it may be

possible, if the particle can be removed and it has sufficient mass (e.g., >3 mg), to
use thermogravimetric analysis (TGA) to obtain its bulk composition. Also, if a
hyphenated TGA technique is available (e.g., TGA–IR or TGA–MS) then this will
enable both quantitative and qualitative data on the particle to be obtained.

10.3.2.2 Fibres

FTIR microscopy is also an excellent tool for the analysis of fibrous contamination,
particularly if it can be removed from the product, although the relatively fragile
nature of some fibrous materials may make this difficult. Also, due to the large aspect
ratio of fibres, the minimum area constraint referred to in Section 10.3.2.1 for FITR
microscopy may be more problematic in the case of very fine fibres. Analysing the
fibre in situ may be the best option in a number of cases and in addition to FTIR
microscopy other microscopic techniques (e.g., optical and SEM) can make important
contributions towards finding the origin of the contamination.

SEM and optical microscopy are particularly useful because they can differentiate

328
 Analysis of Surface Blooms and Contaminants

synthetic fibres (e.g., polyester fibres) from natural fibres (e.g., cotton), and possibly
the type of natural fibre, by their surface appearance: synthetic fibres are smooth
and natural fibres have characteristic topographies. They can also show the exact
positioning of the fibre in relation to the product surface (e.g., on the surface or
partially embedded in the surface) and this may provide some clue as to which stage
in its life the product encountered the fibre.

10.3.2.3 Liquids

A rubber product can absorb a solvent or other liquid (e.g., water) in service and
this can cause a reduction in physical properties (e.g., hardness and tensile strength)
and, possibly, failure. It is necessary in these cases to identify the fluid that has been
absorbed so that appropriate measures can be taken (e.g., a change in the rubber’s
composition to make it more resistant). Identifying the fluid can also go a long way
to solving a failure problem if some abuse of the rubber component is suspected in
that it had been exposed to an environment it should not have come into contact
with, or was not expected to come into contact with.

One important practical consideration of this kind of work is that, due to their
inherent volatility, and the fact that their absorption is primarily a physical effect with
only low intermolecular forces (e.g., van der Waals forces) at work, the majority of
low-MW liquids (e.g., solvents) will be slowly lost from a rubber sample over time.
Losses once the sample is in the laboratory can be minimised by refrigeration until
the time of analyses.

There are a range of possible analytical options for identifying the liquid that the
rubber product has absorbed and some of the principal ones, with their advantages
and disadvantages, are presented below.

A relatively simple approach that can be used to identify an involatile liquid (e.g., an
oil) in a rubber involves the extraction of the rubber with a solvent (e.g., acetone),
removal of the solvent by drying (e.g., in an oven), and analysis of the dried extract
by FTIR, either using a standard bench or a microscope. This approach has the major
disadvantage that the solvent-extraction step will also remove many other substances
from the rubber sample (e.g., oligomers, plasticiser, process aids, and substances
relating to the cure system and antidegradant system) and these can mask the IR
data from the absorbed liquid. Manipulation of the IR data using, for example,
background subtraction of a control sample, may assist the identification process,
but an approach that is more targeted may still be required, particularly if the level
of the absorbed fluid is relatively low.

329
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

If problems are encountered with the solvent extraction–FTIR approach, another

route that can be taken is to use a solvent to extract the rubber that is compatible
with a chromatographic system and then, rather than drying down the extract and
potentially losing the absorbed fluid, analyse it using that system. Chromatographic
system-extraction solvent combinations that could be used are:

• GC–MS – acetone, methanol or dichloromethane

• LC–MS – methanol or acetonitrile

As discussed in Chapter 2, both of these chromatographic techniques have their

strengths and weaknesses, and this will be a relatively costly and time-consuming
route, but combining both techniques should provide a high chance of success.

The use of TGA to determine low-MW additives, such as plasticisers, has been covered
in Section 3.4.1 and it can also be used to identify and quantify the presence of a fluid.
One advantage of the TGA technique for this type of investigation is that it requires
only a relatively small sample size (e.g., 5 to 10 mg) and so it is possible to use it, in
a limited sense, to ‘depth profile’ the sample by quantifying the fluid at the surface
of a sample and at different depths through its bulk. Due to its more sophisticated
nature, TGA can also be more effective at identifying the presence of a low-MW fluid
than the solvent extraction–FTIR approach.

The temperature at which an absorbed fluid will be lost from a rubber sample during
a TGA experiment will depend upon its MW and its chemical properties, but the
majority of solvents and water will show up in a TGA trace before, or at the start,
of a weight loss event due to a plasticiser or oil. It has been shown that the use of a
high-resolution programme or other approaches (e.g., use of a slower heating rate)
can assist in improving the resolution between these events [3].

The use of sealed pans, which are punctured just prior to TGA analysis, is
recommended to help prevent the loss of the absorbed liquid from the rubber
sample, and use of a TGA–IR or TGA–MS combination can enable identification
of the absorbed liquid and so may remove the need to use a more time-consuming
and sophisticated technique, such as headspace GC–MS (see below) or extraction
GC–MS and/or LC–MS (see above). If a hyphenated TGA option is not available to
the analyst, or in more complex circumstances whereby a mixture of fluids has been
absorbed, it is necessary to use a chromatographic approach.

As discussed in Chapters 4 and 5, the headspace GC–MS technique is a very powerful

tool for identifying and, if standards are available, quantifying low-MW substances
(e.g., plasticisers and cure system species) in rubber samples. It can, therefore, be
very useful for identifying liquid contaminants if they are sufficiently volatile; the

330
 Analysis of Surface Blooms and Contaminants

temperature limit (usually around 150 to 200 °C) of the transfer line from the
headspace sampler to the GC–MS instrument is the limiting factor. The high sensitivity
and excellent resolving power of the GC–MS technique mean that it is usually possible
to identify the liquid in the presence of the other volatile substances (e.g., additives
and their breakdown products) that will be in the headspace of the rubber sample.

References

1. B. Jurkowski, E. Koczorowska, W. Goraczko and J. Manuszak, Journal of

Applied Polymer Science, 1996, 59, 4, 639.

2. Y. Hoei, Rubber Chemistry and Technology, 2010, 83, 1, 46.

3. M.J. Forrest in Proceedings of the Polymer Testing Conference, 7–11th April,

Shawbury, UK, 1997, Paper No.5.

331
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

332
11
Analysis of Rubber Latices

11.1 Introduction

The majority of this book is concerned with the analysis of solid rubber samples or
products (e.g., gum rubbers, compounded rubbers, foamed rubbers) but it was thought
necessary, in order to fully reflect the diversity of rubbery materials, to include a section
on the analysis of rubber latices. Latices can be readily split into two principal types:

• Natural rubber (NR) latices from Hevea brasiliensis.

• Synthetic rubber latices produced by the emulsion polymerisation of monomers,

such as styrene, butadiene, chloroprene and acrylonitrile (ACN).

NR latex straight from Hevea brasiliensis usually has ≈30% solids content and one
of the first processes that it undergoes is a concentration step (e.g., by centrifuging)
to increase its solids content to ≈65%. Ammonia, at a ‘high’ or ‘low’ level is added
to the latex to stabilise it and keep it free from micro-organisms. In the case of the
low ammonia latices, at least one other preservative is added. The addition of these
stabilisers results in NR latices being subdivided into the following classifications [1].

• High ammonia latex.

• Low ammonia with 0.2% sodium pentachlorophenate latex.

• Low ammonia with 0.25% boric acid and 0.05% sodium pentachlorophenate
(LABA) latex.

• Low ammonia with up to 0.1% zinc diethyldithiocarbamate (ZDEC) latex.

With respect to synthetic latices, the polymerisation of organic monomers via the
emulsion polymerisation method to create these latices is a huge technical subject in
its own right and has been the subject of numerous books [2, 3]. This section will
provide an overview of the analytical work that can be carried out on both natural
and synthetic latices. However, the detailed technology and science associated with
these products is outside the scope of this book, and references have been provided
throughout to help the interested reader research the subject.

333
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The rubber latex branch of the rubber products family is a very important one,
including as it does many high-tonnage and instantly recognisable products, as well as
those that have a lower profile, and so are less well known. Latex gloves and condoms
fall into the former category and adhesives and elastic cord into the latter. There have
been a number of textbooks describing in detail the manufacture of products from
rubber latices, including those published by Blackley [4, 5]. There are many physical
tests that have been developed to test the final latex products (e.g., burst pressure of
condoms and permeation performance of safety gloves) which are outside the remit
of this book, but are described in the literature for those who are interested [4, 5].

This section is concerned with the analysis of rubber latices and within this subject
there are two discreet areas:

• Analysis of the latex itself – including natural latex and synthetic latices.

• Analysis of products manufactured from both natural and synthetic latices.

It will concentrate on the chemical analytical tests used to characterise latex and to
test its quality and composition. Some physical tests that are used for these purposes
(e.g., total solids tests and viscosity tests) will also be included.

With regard to the chemical analysis of the final products produced from latices,
because the water component of the latex is invariably removed (by coagulation and
then by drying and vulcanisation) during the manufacturing process, what remains
is a solid rubber product that can be subjected to the same analytical tests and
testing protocols as other solid rubber products when it comes to the determination
of, for example, the principal components and additives. The same techniques and
approaches can also be applied for the purposes of reverse engineering, failure
diagnosis and for demonstrating compliance with regulations.

One area of analysis and quality control which has been particularly active for NR
latex in recent years is the determination of residual protein in products such as
gloves (particularly medical gloves) and condoms due to the well publicised allergy
and sensitisation problems. For example, the US Center for Disease Control and
Prevention has estimated that up to 6% of the population has a sensitivity or allergy
to the proteins in NR latex, and it is thought that there has been an increase in the
sensitivity to the accelerators used in NR latex and synthetic rubber latex [6]. To
address these problems and concerns, a number of tests that can be carried out on latex
and latex-derived products have been developed. One of the tests that can be used on
latex products is a latex allergy test and this particular test, as well as examples of the
others that are available (e.g., protein test), have been included in Section 11.2.15.

334
 Analysis of Rubber Latices

11.2 Quality Control Tests

11.2.1 Sampling

The sampling of rubber latex is covered by the International Organization for

Standardizations’ ISO 123:2001, which describes the procedures to be adopted to
obtain representative samples of latex contained in drums or in bulk tanks. In this
standard, it is emphasised that the material of the container in which the latex is kept
must be impermeable and chemically resistant to the latex. Also, space should be left
in the container to allow for thermal expansion of the latex and the container should
be kept closed as much as possible to prevent evaporation and interaction with the
air. During handling operations, care should be taken to prevent the introduction of
air into the latex because foam on the surface will rapidly form a skin.

11.2.2 Coagulum Content

ISO 123:2001 specifies that if the coagulum of the latex exceeds 0.05%, it shall be
filtered before any other tests are carried out on it. The coagulum content of rubber
latex is defined as the material retained on stainless-steel wire cloth having an average
aperture width of 180 ± 15 µm under the conditions of the test, which are provided
in ISO 706:2004. The test method in ISO 706:2004 involves diluting 200 g of latex
with soap solution and passing it through the filter, thoroughly washing the filter and
any retained matter, and then drying to constant weight at 100 °C. The coagulum
content is expressed as a percentage by mass of the latex.

11.2.3 Total Solids Content

The total solids content of rubber latex is described in ISO 124:2014 and is determined
by drying ≈2.0 g of latex to constant mass at a temperature of 100 or 70 °C for NR
latex. In the case of synthetic rubber latex, an additional option of 125 °C is available
if a reduced pressure of 20 kPa is used.

11.2.4 Dry Rubber Content

The dry rubber content of NR latex is determined by coagulating ≈10 g of latex,

after dilution with water, with dilute acetic acid. The coagulated rubber is digested,
washed thoroughly with water, and pressed into a film of up to 2-mm thickness before
drying to constant weight at 70 °C.

335
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The dry rubber content of NR latex is a more accurate determination than the total
solids content because a much larger mass of latex is tested and the dry rubber content
is less hygroscopic that the total solids content. The difference between these two
measurements can be regarded as the ‘non-rubber solids’ content of NR latex, and
will include substances such as proteins.

11.2.5 Dry Polymer Content

The dry polymer content test method is applicable to high solids ‘cold’ polymerised
synthetic rubber latices, particularly certain styrene-butadiene rubber (SBR) types.
The method, which is available in the British Standard, BS 3397:1976, measures the
total styrene-butadiene content, including any other polymeric material present in
the latex. To carry out the determination, ≈6 g of latex is coagulated by the addition
of acetone, refluxed briefly and then washed and dried to constant weight at 100 °C.

11.2.6 Density

The method to determine the density of NR latex is provided in ISO 705:2015, and
requires the use of a 50-cm3 density bottle. It emphasises the importance of ensuring
that no air is trapped and that the temperature is accurately recorded.

11.2.7 Determination of pH

The pH of rubber latices is measured using the test method provided by ISO 976:2013.
This method uses a glass electrode and saturated calomel cell after standardising
the pH meter with borax and potassium hydrogen phthalate solutions. A suitable
combination electrode may be used in place of the single electrodes. The method
should give a result that is accurate to 0.1 pH.

11.2.8 Alkalinity

The alkalinity of NR latex is defined as the percentage by mass of ammonia [or

potassium hydroxide (KOH) if it has been used as a preservative] that it contains.
The alkalinity of a latex is determined by titrating to pH 6.0 in the presence of a
stabiliser. The method, described in ISO 125:2011, usually employs electrometric
titration, but it is also possible to use methyl red as an indicator.

336
 Analysis of Rubber Latices

11.2.9 Potassium Hydroxide Number

This was the first test to be developed as an indicator of latex quality. The test is
described in ISO 127:1984 and KOH number is defined as the number of grams of
KOH equivalent to the acidic moieties combined with ammonia in latex containing
100 g of total solids. It is determined by potentiometric titration of NR latex with
KOH solution, after adjustment of the alkalinity and dilution of the latex. The
endpoint of the titration is the point of inflection of the titration curve of pH against
the volume of KOH added.

The KOH number collectively determines volatile fatty acid (VFA), carbonate/
bicarbonate [carbon dioxide (CO2) number – Section 11.2.11], non-volatile acids
(NVA), higher fatty acid (HFA) and other acidic species. Given that HFA are stabilising,
whereas the other acidic species are all destabilising, reduces the value of the KOH
number as a measure of latex quality because it adds together both stabilising and
destabilising components.

11.2.10 Volatile Fatty Acid Number

This test is described in ISO 506:1992. The VFA number is determined after
coagulating the latex with ammonium sulfate and acidifying the resulting serum by
steam-distilling the serum in a Markham still and measuring the volatile acids in
the distillate by titration with barium hydroxide solution. Prior to the titration, it is
essential to purge the distillate of CO2 with nitrogen. The most predominant VFA in
natural latex is acetic acid.

11.2.11 Carbon Dioxide Number

The CO2 number refers to the carbonate and bicarbonate content of NR latex. It can
be determined by a gravimetric method [7] and other methods (e.g., the macrobaryta
technique after acidifying stabilised latex).

Sundaram and Calvert [8] discussed the changes that can occur in the CO2 number of
field latex during a number of stages in its production, for example, before and after
ammoniation, and the effects of centrifuging and storage of the concentrate prior
to shipping, as well as changes that may occur during shipping. The CO2 number is
usually ≈0.12 greater than the VFA number, and so is an alternative indicator of the
level of preservation of NR latex.

337
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

11.2.12 Non-volatile Acids

NVA contribute more to the KOH number than any of the other acidic groups. NVA
are determined by passing latex serum, obtained by coagulation with acetic acid,
through a strong cation-exchange resin, followed by evaporation and acidimetric
titration [9]. This method is time-consuming (≈4 h) and requires a measure of expertise.

11.2.13 Higher Fatty Acids

These compounds are determined, after extraction of latex total solids with acetone,
by acidification with sulfuric acid and, after work-up, titration with KOH solution [9].

The VFA number, CO2 number, HFA number and NVA number are all expressed in
the same units as the KOH number.

11.2.14 Electrical Conductivity

Electrical conductivity testing of latices has been of interest for a number of reasons:

• As a measure of the quality of NR latex.

• For the conductometric titration of synthetic latices for total soap content.
• As an aid for the soap titration of synthetic latices for the determination of average
particle size.
Its potential use in the first example arises because it has been established that as NR
latex deteriorates its electrical conductivity tends to increase due to an increase in the
ionic strength of the aqueous phase of the latex. However, the comparative insensitivity
of electrical conductivity to the deterioration in quality means that determination of
the VFA number (Section 11.2.10) is often regarded as more useful.

Measuring electrical conductivity has proved to be of practical use for both the
determination of total soap, in which a diluted, acidified solution of latex is back-
titrated with alkali, and in providing assistance to the soap titration method for
average particle size determinations [4].

11.2.15 Latex Allergy Tests, Protein Tests and Sensitivity Tests

Although the Food and Drug Administration in the United States have stated that,
at present, they are not aware of any tests that can show a product is completely free

338
 Analysis of Rubber Latices

of NR latex proteins that can cause allergic reactions, tests are available to detect
allergens [10]. These tests, which can be performed using bespoke testing kits, are an
immunological test for measuring NR latex allergens from a variety of rubber products
(e.g., gloves). Separate testing kits have been made available by manufacturers for
measuring each of the major allergens: Hev b 1, Hex b 3 and Hex b 6.02. These tests
function by using specific monoclonal antibodies that have been developed against
the clinically relevant latex allergens present in NR latex products.

In addition to allergen tests, there are also standard test methods that have been
developed by standardisation organisations [e.g., American Society for Testing and
Materials (ASTM)] for the determination of proteins in NR latex products. Examples
of these tests include:

• ASTM D5712 – Modified Lowry/total protein – Standard for the quantification of

total aqueous extractable protein associated with natural rubber, natural rubber
latex, and elastomeric products.

• ASTM D6499 – Inhibition enzyme-linked immunosorbent assay (ELISA) is an

immunological method which determines the amount of antigenic protein in
natural rubber and its products by using rabbit antisera specific for natural rubber
latex proteins.

• ASTM D7427 – Allergen ELISA is a test method that uses four ELISA kits to
quantify four of the most common Hevea allergens (i.e., b 1, 3, 5 and 6.02) in
natural rubber latex and its products.

Other tests that are available include:

• ASTM Guayule ELISA ASTM WG 25943 – Measures the levels of extractable

Guayule proteins from commercial products. At the time of writing, this method
is being validated by ASTM.

• ASTM D6214 – Powder content – Determines the average powder or filter-retained

mass found on a sample of medical gloves.

• ASTM D7558 – Chemical sensitivity – Determines the amount of total extractable

accelerators in natural rubber latex gloves and nitrile latex gloves.

11.3 Chemical Stability Tests

Chemical stability tests have been slow to achieve international acceptance because
they have been found to be difficult to reproduce, have relatively poor repeatability,
or of questionable relevance to industrial latex processes.

339
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

11.3.1 Natural Rubber Latices

Zinc oxide (ZnO) is added to NR latices in a number of industrial processes. Hence,

the stability of the latex-to-zinc ammine ions (from the reaction of ZnO with the
ammonia preservative) has been the focus of attention. Two particular tests which
have gained prominence are the zinc stability time (ZST) test [11] and the zinc oxide
viscosity test [12].

Other chemical stability tests for NR latex tend to be methods that are modifications
of the ZST test or methods that use preformed zinc ammonium acetate.

11.3.2 Synthetic Rubber Latices

The chemical stability tests that have been proposed for synthetic latices are more
varied and diverse. One widely regarded method involves the addition of a surfactant
and ZnO dispersion to the latex, and measurement of the time taken to initiate
coagulation. Other methods determine the amount of coagulum formed by the
addition of an electrolyte (e.g., calcium chloride) or methanol.

11.4 Polymer Composition of Synthetic Latices

11.4.1 Bound Styrene Content

The ISO 3136:1983 method for bound styrene is applicable only to straight SBR
latices and not to modified versions such as carboxylated SBR latex. In comparison,
the ISO 4655:1985 method applies to all SBR-type latices.

In the case of ISO 3136:1983, the solid polymer is first prepared by coagulating the
latex with sodium chloride and sulfuric acid in the presence of methanol. The polymer
is then dried, extracted with ethanol–toluene azeotrope, and then pressed into a thin
sheet. The bound styrene value is obtained by refractive index measurements.

11.4.2 Total Bound Styrene Content

ISO 4655:1985 describes two methods for the determination of total bound styrene.
This standard can be applied to ‘reinforced’ SBR latices, whereas ISO 3136:1983
(Section 11.4.1) cannot. Both methods in ISO 4655:1985 start with the coagulation
of the latex with isopropanol, followed by extensive washing and thorough drying of

340
 Analysis of Rubber Latices

the coagulum. In the carbon/hydrogen method, the coagulum is burnt in a specified

apparatus and the total bound styrene determined from the amount of CO2 and
water that are quantitatively absorbed. In the nitration method, the coagulum is
nitrated and oxidised to convert its total bound styrene content to p-nitrodenzoic
acid, which is then separated by multiple extractions and determined quantitatively by
measuring its ultraviolet (UV) absorption. The two methods are capable of producing
comparable results.

11.4.3 Bound Acrylonitrile Content

The ISO 3900:1995 method for bound ACN can be applied to nitrile rubber (NBR)
latices, carboxylated versions of NBR latices, and nitrile–isoprene latices.

The preparative stage of this method involves extraction of an air-dried film of the
nitrile latex with water to remove water-soluble nitrogen-containing material and
drying to constant mass. The analytical procedure is then the same as that described
in ISO 1656:1996 for the determination of the nitrogen content of NR.

11.4.4 Volatile Monomers

Instrumental methods for the determination of residual monomers have largely

replaced the older international standards that were essentially based on wet chemistry
techniques, for example:

• ASTM D4026-06 for styrene in SBR latices.

• ISO 3899:2005 for ACN in NBR latices.

• ISO 3499:1976 for polyvinyl acetate latices.

One principal reason for this replacement is that gas chromatography (GC) methods
are more sensitive and so give a lower limit of detection. The preferred technique is
static headspace GC, in which the latex is heated in a sealed vial and the headspace
analysed for the presence of the monomer. Calibration is usually achieved using the
standard addition approach, where known amounts of the monomer are added to
aliquots of the same latex to ensure that the headspace partitioning remains constant.
For monomers such as ACN, a specific detector (a nitrogen phosphorus detector) can
be used to achieve very low levels (i.e., ppb) of detection. This compares with the old
ISO 3899:2005 method, which has a detection limit of only 100 ppm.

341
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

11.5 Particulate Property Tests

Test methods for particle properties cover determinations of viscosity, surface tension,
soap content, soap deficiency and particle size.

11.5.1 Viscosity

There are three international standards for the determination of viscosity:

ISO 1652:2011, ISO 2555:1989 and ISO 3219:1993.

ISO 1652:2011 is applicable for the determination of the viscosity of rubber

latex. In the standard, the Brookfield L instrument is specified for low viscosities
(≤200 mPa.s), the Brookfield R instrument for high viscosities (>2,000 mPa.s) and
either instrument can be used for intermediate viscosities. The speed of rotation of
the viscometer is restricted to 60 rpm for the Brookfield L instrument and 20 rpm
for the Brookfield R instrument.

ISO 2555:1989 is for the determination of the viscosity of dispersions of resins and
differs from ISO 1652:2011 in two respects: only the R version of the Brookfield
instrument is specified and there is no restriction on the speed of rotation. However,
speeds of 10 and 20 rpm are recommended.

ISO 3219:1993 is a general standard for the determination of the viscosity of

dispersions of polymers. This standard uses a rotational viscometer working at a
defined shear rate and seven shear rates from 1 to 250 s−1 are recommended.

The viscosity of latex increases with its total solids content so, as well as depending
on the rate of shear used, both total solids content and shear rate should be appended
to all viscosity results.

11.5.2 Surface Tension (Surface Free Energy)

ISO 1409:2006 describes the determination of the surface tension of rubber latex.
A du Nouy tensiometer with a platinum ring of 40 or 60 mm in circumference is
specified. The latex is required to be at a total solids content of ≤40%. The units
of surface tension are mN/m. Surface free energy is synonymous and has the same
numerical value as surface tension if expressed in units of mJ/m2.

342
 Analysis of Rubber Latices

11.5.3 Soap Content

The soap content of synthetic latices that have been produced using potassium oleate
can be determined by potentiometric titration with sulfuric acid, after stabilisation
with a non-ionic stabiliser and neutralised isopropanol, and adjustment of the pH
to at least 11.0 with KOH. Two inflection points occur at around pH 9.4 and 5.0.
The volume of acid in between these two inflection points is calculated for a latex
sample and a blank, and the potassium oleate content is regarded as the difference
between the two.

Other methods exist for different fatty acid and rosin acid soap systems.

11.5.4 Soap Deficiency

The soap deficiency of some latices can be determined by surface tension titration
or conductometric titration with the same soap that is already present in the latex.

With surface tension titration, the surface tension is lowered linearly as soap is added.
Once the surface of the polymer particles is saturated with adsorbed soap, a break
in the line (i.e., an inflexion) occurs and further addition of soap has only a minor
effect on surface tension. The amount of soap added to reach this inflexion point
represents the soap deficiency of the latex.

With conductometric titration, the initial electrical conductance of the latex increases
rapidly in a linear way with soap added. As with the former method, once saturation
of the particle surface has occurred, the conductance increases more gradually. Again,
the soap deficiency is the amount of soap added to reach the break in linearity.

11.5.5 Particle Size

There are a number of experimental methods that can be used to determine particle
size: electron microscopy, soap adsorption, light scattering, centrifugation, fractional
creaming and counting methods.

To be amenable to electron microscopy, rubbery latex particles have to be hardened,

for example, by bromination, to prevent distortion or coalescence.

There are a number of soap adsorption methods [13] and these require the latex
to have a soap deficiency. The initial soap content of the latex and the molecular
adsorption area of the soap need to be known.

343
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Light-scattering methods measure the intensity of radiation that is scattered by the

latex at various angles to the direction of incidence.

Centrifuge methods depend on the difference in density between the particles and
the serum in which they are dispersed.

The Coulter Counter is a useful instrument for measuring the size distribution of
latex particles in the diameter range 40 to 3,000 µm.

‘Fractional creaming’ was developed by Schmidt and Biddison [14]. The method uses
the quantitative inverse relationship between the concentration of a creaming agent
(sodium alginate) and the size of the creamed particles.

11.6 Miscellaneous Chemical Tests

11.6.1 Nitrogen and Trace Metals

The nitrogen content of NR latex is determined using the method described in

ISO 1656:1996. This test applies only to the analysis of NR latices, and the nitrogen
content that is obtained may be used to obtain an estimate of the protein content of
a latex by applying a factor of 6.25.

There are international standards for the determination of copper (ISO 1654:1971),
manganese (ISO 1655:1975), and iron (ISO 1657:1986). These tests can be applied
to both NR and synthetic rubber latices.

More contemporary methods are also available. For example, a method for the
determination of the nitrogen content of natural and synthetic rubber latex using
a photometric method is described in ISO 8053:1995 and a sodium periodate
photometric method for the determination of manganese is specified in ISO 7780:1998.

It is possible to determine other metals and elements in both types of rubber latex
by using atomic absorption spectrometry (AAS) or one of the inductively coupled
plasma-based methods. For example, the determination of metals in rubber latex
samples by AAS is described in the six-part standard ISO 6101.

11.6.2 Zinc Diethyldithiocarbamate

It is possible to use ZDEC as an auxiliary preservative in NR latex and to estimate

344
 Analysis of Rubber Latices

the amount of ZDEC in a latex sample by quantitatively decomposing it to carbon

disulfide, converting this compound into a water-soluble dithiocarbamate (e.g.,
diethylammonium diethyldithiocarbamate), and then forming the corresponding
copper derivative which is estimated absorptiometrically.

11.6.3 Boric acid

ISO 1802:1985 describes the determination of the boric acid content of the LABA
type of NR latex. The test involves complexing the boric acid with mannitol and
titrating the liberated hydrogen ions with alkali. Importantly, LABA types of NR
latex have a stabiliser system that employs boric acid and sodium pentachlorphenate
as secondary stabilisers, with ammonia as the primary stabiliser.

11.6.4 Determination of Additives and Compounding Ingredients

Because of its colloidal properties, rubber latex is compounded using a different

approach to that used for solid rubber. For example, water-soluble materials can be
added to the latex as solutions, whereas insoluble liquids are added as emulsions,
and insoluble solid materials after they have gone through a grinding stage to reduce
their particle size to ≤5 µm. If these constituents are added, their concentrations must
be adjusted to control viscosity, reduce ionic shock and their pH has to be matched
to that of the rubber latex. In general, the types of additives that are used to stabilise
and cure rubber latex are the same as those used in solid rubbers, although there
will be specific classes of additive that are only used in latex (e.g., preservatives,
surfactants and stabilisers).

The types of antioxidants and curatives that are commonly used in rubber latex
formulations are summarised in Table 11.1.

Once the latex has been destabilised, dried and cured and a solid product produced,
the identification of the antioxidants and curatives can be undertaken using the same,
or similar, analytical techniques and approaches to those described in Chapters 2 and
4. For example, Faridah [15] reviewed the high-performance liquid chromatography
(HPLC) and thin-layer chromatography procedures that can be used to detect and
identify the amount of residual accelerators and antioxidants in rubber products and
rubber latex dipped products. The article concentrated on the different extraction
procedures that can be used to isolate the analytes from the rubber products, and
provided the types of accelerators and antioxidants, and their residues, that can be
detected by HPLC.

345
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Table 11.1 Examples of antioxidants and curatives used in rubber latices [15]
Antioxidants used in rubber latex Curatives used in rubber latex
Hindered phenols Sulfur
Zinc 2-mercaptotoluimidazole ZnO
Styrenated phenols Guanadines
p-Phenylenediamines Thioureas
Paraffinic waxes Thiazoles
Dithiocarbamates (residual quantities from cure) Dithiocarbamates
Network-bound antioxidants Thiurams
Xanthates
Reproduced with permission from D. Hill in Proceedings of Latex 2006, 24–25th January
2006, Frankfurt, Germany, Rapra Technology Ltd, Shawbury, UK, 2006, Paper No.8.
©2006, Rapra Technology Ltd [15]

Another very important class of ingredient that is used in the compounding of rubber
latex is stabilisers. Some examples of the stabilisers that can be used in rubber latex
have been provided by Hill [16], and are listed below:

• Alkali (e.g., KOH and ammonium hydroxide)

• Fatty acid soaps (e.g., potassium laurate)

• Fatty alcohol ethoxylates

• Alkyl aryl sulfonates

• Alkyl sulfates

• Rosin acid soaps

• Proteins (e.g., casein)

Specific information on the stabilisers that have been used in a rubber latex can be
obtained using some of the analytical techniques described in Chapter 2. Guidance on
the methods and approaches that can be employed are included in general polymer
analysis books such as those published by Crompton [17] and publications specifically
addressing latices, for example, those published by Blackley [4], Daniels and Anderson
[18], Warson and Finch [19] and Poehlein and co-workers [20].

Some bulk test methods to determine stabilisers and their effect on the bulk properties
of latices are described in Sections 11.2 and 11.5. Finally, the identification and

346
 Analysis of Rubber Latices

characterisation of the polymer within the rubber latex is possible using the relevant
parts of Chapter 2, Chapter 3 and Section 11.4.

11.7 Characterisation Studies Undertaken on Rubber Latices

The quality control tests and analytical methods described in Sections 11.2–11.6 have
provided an overview of the work that can be carried out to characterise and evaluate
specific properties of rubber latices and the additives and ingredients that are used in
their formulations. This section will provide some examples of the work that has been
published over the last 15 years to characterise rubber latices and their constituents.
Often such studies are an essential part of research programmes aimed at developing
new products with attractive properties. On other occasions they have been carried
out in order to extend and enhance the knowledge that exists in certain areas (e.g.,
chemical properties or chemical structure). It, therefore, performs a similar function
to Chapter 5, which provides examples of published reverse engineering programmes
that have been carried out on solid rubber products, in that it demonstrates the multi-
technique approach that analysts use in order to obtain as much data as possible on
test samples.

Yan and co-workers [21] synthesised zinc ion self-crosslinkable polyacrylate (PA)
latices that can be cured at room temperature. They used a seeded semi-continuous
emulsion polymerisation method to produce the latices and ZnO as the crosslinking
agent. In order to control the amount of crosslinking that took place within the latex,
the quantity of ZnO that was employed was related to the quantity of methacrylic
acid (MAA) monomer. Once the latices were produced, they were characterised using
a range of analytical techniques that included Fourier–Transform infrared (FTIR),
differential scanning calorimetry (DSC), transmission electron microscopy (TEM)
and thermogravimetric analysis (TGA). The results obtained showed that, within
a certain concentration range of MAA, the average particle size decreased with
increasing amounts of MAA, and the stability of the latex increased. The optimum
values of the variables were found to be:

• Proportion of MAA within the monomer feed – 12%.

• Amount of ZnO – 25% mole fraction of MAA.

• Temperature at which ZnO is introduced – 60 °C.

The TEM results showed that the latex particles were coarse spherical particles that
were surface-enriched with carboxyl groups, and the zinc ions were dissociated as
a zinc ammine complex in the aqueous phase. The FTIR data confirmed that the
chelate crosslinking occurred between zinc ions and carboxylic acid during the film-

347
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

forming process, with the resulting films exhibiting excellent hardness and the ability
to be sanded. The DSC thermograms revealed that the glass transition temperature
(Tg) of the latices increased as a function of the formation of a coordinate structure,
and TGA analyses confirmed that the use of ZnO enhanced the thermal stability of
crosslinkable PA latices.

A range of self-crosslinking latices has also been produced by researchers from

the Ministry of Eduction in China and Nankai University [22]. The research team
used seed emulsion polymerisation with various amounts of vinyl triethoxysilane in
conjunction with methyl methacylate, n-butyl acrylate and styrene. The properties
of the latices were characterised by employing a combination of techniques (e.g.,
FTIR, DSC and TGA) and films from the latices were prepared and their mechanical
properties and water repellency assessed. Together, the results provided the workers
with information on the structure, degree of crosslinking and thermal properties of
the latices.

Microemulsion photopolymerisation has been used to prepare silicone-modified

styrene-butyl acrylate copolymer latices by a group of Chinese researchers [23]. The
group used a combination of FTIR, UV, TEM, TGA, water swelling tests and acid/
alkali resistance tests to study the effect that modifying the monomer-to-surfactant
mass ratio and the silicone content had on the properties of the latices and coatings
that were produced from them using the casting technique. The data obtained showed
that the latex particles were almost spherical and had a uniform particle size. Also, as
the monomer-to-surfactant mass ratio or silicone content increased, the acid, alkali
and water resistance also increased.

A team from Wuhan University of Technology [24] synthesised the silicone monomer
methacryloxypropyltris(trimethylsiloxy)silane and used it as a monomer in a co-
polymerisation with methyl methacrylate. The co-polymerisation reaction was used
to form emulsifier-free latices of silicone acrylate copolymer using sodium 3-allyloxy-
2-hydroxypropanesulfonate, potassium persulfate and ultrasonic emulsification. The
emulsifier-free latices that resulted were analysed using a wide range of techniques
(e.g., spectroscopic and microscopic) and were found to be better dispersed and
have a lower surface free energy than latices produced using conventional emulsion
polymerisation.

Yuan and co-workers [25] prepared, by emulsion polymerisation, a silicone-

modified styrene-butyl acrylate copolymer. This was achieved by grafting silicone
oil to a styrene-butyl acrylate copolymer through a reaction with a functional
monomer (gamma-propylmethacrylate trimethoxysilane). The surface properties,
hydrophobicity, water repellency and Tg of the latices were determined by FTIR,
contact angle measurements, water absorption measurements and DSC. As well as

348
 Analysis of Rubber Latices

yielding primary characterisation information, the results also enabled the team to
evaluate the effects that the grafted silicone oil had on the properties of the latices.

McMahan and co-workers [26] prepared guayule and NR latices and characterised
them to determine their solids content, pH, particle size, molecular characteristics,
nitrogen content and sulfur content. The team also cast films from the two types of
latices and determined their curing, rheological and dynamic mechanical properties
using the Advanced Polymer Analyzer 2000. Within the programme of work, a
comparison was also made of the curing properties and mechanical properties of
films that had been subjected to heat treatment.

Stephen and co-workers [27] studied the thermal degradation and ageing behaviour
of microcomposites of NR, carboxylated styrene-butadiene rubber (XSBR) latices
and their blends. During their work they used TGA, SEM, X-ray photoelectron
spectroscopy and mechanical property tests, before and after ageing, to examine the
effect of clay and silica microfillers on the thermal stability of the latices in the blends.
The results obtained showed that the fillers improved the thermal stability and ageing
resistance of the XSBR latices and their blends. During the study, the surfaces of
samples were exposed to ion beam irradiation and the surface analysis results showed
some redistribution of elements, but no changes in the binding energies.

The chemical structures and the thermal and thermooxidative stabilities of the gel
and sol fractions from chlorinated natural rubber latex (CNR) have been analysed
by a Chinese group of researchers [28] using the following techniques:

• Chemical analysis

• High-resolution pyrolysis GC–mass spectrometry

• Differential thermal analysis

• TGA

From the data that resulted from the test programme, the researchers concluded that
it was the presence of the carbonyl group and tertiary C–Cl group that resulted in
the CNR from latex having worse stability than CNR produced from solution. Also,
the thermal and themooxidative stabilities of the sol fraction were better than those
of the gel fraction.

Nawamawat and co-workers [29] studied the structure of branch-points in NR after

washing fresh and aged NR latices by centrifugation with sodium dodecyl sulfate.
The team found that the nitrogen content of the NR isolated from both the fresh
and aged latex decreased to 0.2% after centrifuging thrice, a value similar to that
of enzymatically deproteinated NR. The gel content of the NR from the two types

349
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

of latex was found to decrease in a stepwise manner, to ≈6% for the aged latex and
≈3% for the fresh latex. In addition, the decrease in the molecular weight and the
Huggins constant brought about by washing was considered to indicate a decrease in
the branch points in the NR latex and to suggest that the proteins participated in gel
formation. Analysis of the proteins that had been removed by the washing process
by gel electrophoresis showed that the majority had not been degraded.

Rubber products that are used as protective clothing (e.g., gloves) are often
manufactured from latex and one of the most important properties is resistance
to permeation by liquids (e.g., solvents). A research team from New Zealand [30]
evaluated an attenuated total reflectance (ATR)–FTIR method for measuring the
permeation of a number of chemicals (e.g., solvents, a commercial pesticide mix
and a volatile solid) through gloves that were made of different rubbers (e.g.,
polychloroprene, NBR and NR). Prior to conducting the measurements, the team
ensured that good contact existed between the sample and the ATR crystal by using a
low-pressure gas. The results showed that it was possible to measure the permeation
of the chemicals through the samples by collecting a series of FTIR spectra, but that
the diffusion coefficients could only be estimated. Their study showed that the ATR
method had both advantages and disadvantages over the traditional two-compartment
cell method that is used for permeation work.

A Malaysian team [31] carried out work on NR latex products to investigate the
effect that residual chemicals have on their biocompatibility and physical properties.
The approach involved the preparation of several in-house compounds of differing
chemical composition and testing these using liquid chromatography and tensile
tests to identify an optimum formulation, which contained a low level of residual
chemicals and had acceptable physical properties. The biocompatibility of the
specimens produced from this optimum formulation was also evaluated using an
in vitro cytotoxicity method. The effect of pre-washing the products with acetone
on the residual chemical content, physical properties and biocompatibility was also
investigated, and comparative work on commercial, powder-free gloves was included
in the test programme.

Potter [32] reported a pilot study to investigate the use of various chemical, physical
and spectroscopic methods to assess the chemical changes occurring in NR latex
condoms aged under different conditions. The tests employed included ATR–FTIR
spectroscopy, determination of residual accelerators, and the assessment of crosslink
type and crosslink density using solvent swelling and chemical probes. During the
ageing process, the condoms were packed in air-impermeable aluminium foil laminate
packaging. The comparative results obtained on un-aged and aged condoms are
summarised below:

350
 Analysis of Rubber Latices

• ATR–FTIR did not detect any significant differences in chemical structure.

• Residual accelerator data did not provide information on the chemical changes
that might be occurring during ageing, but was useful background information.

• Statistically significant changes in crosslink density and crosslink type were

observed depending upon the ageing conditions.

The conclusion reached from the study was that chemical probe studies (e.g., using
propane-2-thiol) offered significant promise as a method for investigating the chemical
changes that occur as NR latex condoms age. The study also provided further evidence
to that already in the public domain that high-temperature ageing (i.e., at 70 °C)
resulted in changes that were not typical of those that were seen at intermediate and
ambient temperatures.

Bluemich and co-workers [33] described how a bar-magnet nuclear magnetic

resonance–mobile universal surface explorer was designed with a new coil which
made it particularly useful for investigating samples with a thickness <1 mm and for
examining the surfaces of polymer samples. The team applied the technique to the
study of a number of different samples and reported on the findings. The examples
included in their study are shown below:

• Swelling and drying of a latex membrane exposed to cyclohexane vapour

• Surface damage on rubber samples

• Drying of a thin, sprayed adhesive layer

References

1. K.O. Calvert in Polymer Latices and their Application, Ed., K.O. Calvert,
Applied Science Publishers Ltd., Barking, UK, 1982, Chapter 2.

2. D.C. Blackley in Emulsion Polymerisation: Theory and Practice,

Applied Science Publishers Ltd., Barking, UK, 1975.

3. Chemistry and Technology of Emulsion Polymerisation, 2nd Edition, Ed.,

A.M. van Herk, John Wiley & Sons Limited, Chichester, UK, 2013.

4. D.C. Blackley in Polymer Latices Science and Technology, Volumes 1–3,

2nd Edition, 1997, Chapman and Hall, London, UK, 1997.

5. Polymer Latices and their Application, Ed., K.O. Calvert, Applied Science
Publishers Ltd., Barking, UK, 1982.

351
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

6. Latex Testing, Bureau Veritas Group, Paris, France.

http://www.bureauveritas.com/services+sheet/latex-testing

7. K.O. Calvert and R.K. Smith, Rubber India, 1974, February, 31.

8. P. Sundaram and K.O. Calvert in Proceedings of International Rubber

Conference RRI Malaysia, Kuala Lumpur, Malaysia, 1975, 5, 313.

9. K.O. Calvert, Plastics and Rubber: Materials and Applications, 1977, 2, 2,

59.

10. Don’t be Misled by ‘Latex Free’ Claims, US Food and Drug Administration,
Silver Spring, MD, USD, 2015.
https://www.fda.gov/ForConsumers/ConsumerUpdates/ucm342641.htm

11. J.L.M. Newnham, K.O. Calvert and D.J. Simcox in Proceedings of Natural
Rubber Research Conference, Kuala Lumpar, Malaysia, 1961, p.668.

12. H.G. Dawson, Rubber World, 1956, 135, 239.

13. S.H. Maron, M.E. Elder, I.N. Ulevitch and C. Moore, Journal of Colloid
Science, 1954, 9, pages 89, 104, 263, 347, 353 and 382.

14. E. Schmidt and P.H. Biddison, Rubber Age, 1960, 88, 484.

15. H.A.H. Faridah, Malaysian Rubber Technology Developments, 2005, 5, 1, 8.

16. D. Hill in Proceedings of Latex 2006, 24–25th January 2006, Frankfurt,

Germany, Rapra Technology Ltd, Shawbury, UK, 2006, Paper No.8.

17. T.R. Crompton in Determination of Additives in Polymers and Rubbers,

Smithers Rapra Technology Limited, Shawbury, UK, 2007.

18. E.S. Daniels and C.D. Anderson in Emulsion Polymerisation and Latex
Applications, Rapra Review Report No.160, Rapra Technology Limited,
Shawbury, UK, 2003.

19. H. Warson and C.A. Finch in Applications of Synthetic Resin Latices,

Volumes 1–3, John Wiley & Sons, Chichester, UK, 2001.

20. Proceedings of the NATO Advanced Study Institute on Polymer Colloids,

28th June–8th July 1982, University of Bristol, Bristol, UK, Eds.,
G. W. Poehlein, R.H. Ottewill and J.W. Goodwin, Springer Science + Business
Media, Dordrecht, The Netherlands, 1983.

352
 Analysis of Rubber Latices

21. W. Yan, X. Zhang, Y. Zhu and H. Chen, Iranian Polymer Journal, 2012, 21,
9, 631.

22. T.Y. Guo, C. Xi, G.J. Hao and M.D. Son, Advances in Polymer Technology,
2005, 24, 4, 288.

23. T. Wan, C. Wu, X.L. Ma, L. Wang and J. Yao, Polymer Bulletin, 2009, 62, 6,
801.

24. C. Zhang, L. Hu and Y. Hu, ACS Polymer Preprints, 2009, 50, 2, 599.

25. J. Yuan, G. Gu, S Zhou and L. Wu, High Performance Polymers, 2004, 16, 1,
69.

26. C.M. McMahan, K. Cornish, H. Pawlowski and J. Williams in Proceedings of

the Spring 167th ACS Rubber Division Meeting, 16–18th May, San Antonio,
TX, USA, American Chemical Society Rubber Division, Akron, OH, USA,
2005, Paper No.62.

27. R. Stephen, A.M. Siddique, F. Singh, L. Kailas and S. Jose, Journal of Applied
Polymer Science, 2007, 105, 2, 341.

28. Y. Dan, L. Sidong, Z. Jieping and J. Demin, China Synthetic Rubber Industry,
2003, 26, 1, 47.

29. K. Nawamawat, J.T. Sakdapipanich, D. Mekkriengkrai and Y. Tanaka,

Kautschuk, Gummi, Kunststoffe, 2008, 61, 10, 518.

30. K. O’Callaghan, P.M. Fredericks and D. Bromwich, Applied Spectroscopy,

2001, 55, 5, 555.

31. M.S.N. Qamarina, K.L. Mok, A.Y. Tajul and R.N. Fadilah, Journal of
Rubber Research, 2010, 13, 4, 240.

32. B. Potter in Proceedings of the 6th Latex and Synthetic Polymer Dispersions,
23–24th March, Amsterdam, The Netherlands, Smithers Rapra Technology
Ltd, Shawbury, 2010, Paper No.7.

33. B. Bluemich, V. Anferov, S. Anferova, M. Klein and F. Fechete,

Macomolecular Materials and Engineering, 2003, 288, 4, 312.

353
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

354
12
Conclusions

Since publication of the Rapra Review Report version of this book, there has been a
significant number of developments in the field of analytical chemistry. These have
resulted in considerable improvement in the results that can now be achieved by the
rubber analyst; this extensive up-to-date review has reflected these developments and
provided many practical examples of their application to rubber samples.

This review has also provided an in-depth description of every important area of rubber
analysis. It has dealt with the various compositional objectives, both specific (i.e., single
substance) and collective (i.e., bulk composition), which need to be undertaken to
ensure consistently high quality in manufacturing and to understand and characterise
rubber samples and products. It has also covered the important themes (e.g., H&S
and regulatory studies) which have had a major influence on the nature of the work
that is being carried out today in a large number of R&D and test house laboratories
and which, in many cases, has led to the setting up of specialist teams. With the
increasing proliferation and importance of regulations, and as pressures continue to
grow and every effort is made to reduce risk to society, it is to be expected that such
activities will increase in the coming years, with analysts having a crucial function
within surveillance and enforcement work.

Also covered in this book are the analytical applications, such as reverse engineering,
cure state studies, failure diagnosis work, and the analysis of contaminants, which are
so important for industry because they seek to develop and improve the products that
they put into the marketplace and to understand why, on some occasions, whether
due to abuse, poor manufacturing or incorrect choice, rubber products can fail to
perform satisfactorily in service.

It is hoped that the strong emphasis on the practical application of the various
analytical techniques, procedures and methods to solve problems and answer questions
will enhance this book’s use to practicing analysts and, in some ways, it can be looked
upon as taking the form of a ‘how to’ guide to rubber analysis. The many examples
of the information that has been obtained by carrying out certain tests in particular
ways should also have the benefit of illustrating the capability of these approaches if
applied to real samples to achieve stated goals and objectives.

355
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

The additional reading references provided in each section should greatly assist the
analyst who requires more information, both specific and general, on any particular
topic and who would like, or requires, more theoretical background on how particular
analytical instruments and experimental methods function and achieve results.

Although an attempt has been made to include as much information as possible, the
subject of rubber analysis is so large that it is inevitable that certain subject areas,
analytical techniques and methods will have been excluded, or only given a cursory
mention; again, the references provided will assist here by filling in any gaps.

For the future, it is to be expected that R&D activities by the instrument manufacturers,
aimed towards more ever more sophisticated, faster, sensitive and higher resolution
equipment, will continue and that these advances will continue to be of benefit to
rubber analysts and further enhance the already significant contribution they make
to industry and academia.

Due to the time and money constraints and pressures that are endemic in the modern
world, one area which is expected to continue to attract attention is the development
of techniques (e.g., those based on mass spectrometry) that can be used for the direct
analysis of rubber samples, hence by-passing the often time-consuming and expensive
sample preparation steps (e.g., solvent extraction or ashing) that have traditionally
been employed.

With regard to rubber technology itself, in some industrial sectors there have
been developments aimed at simplifying the composition of rubber products, an
example being the development of relatively ‘clean’ halobutyl rubber compounds
for pharmaceutical applications to reduce the level and range of extractable and
leachable compounds. In general, there have also been a number of changes to
eliminate particular classes of additive (e.g., high-polyaromatic hydrocarbon content
extender oils), or reduce potentially harmful breakdown products (e.g., nitrosamines
from certain accelerators used in sulfur vulcanisation systems).

However, despite these changes, rubber technology is a very complex science and it is
clear that, even with the development of new and more advanced analytical techniques,
the analysis of rubber compounds and products will continue to present the analyst
with complex and difficult challenges. It is in the meeting and overcoming of these
challenges that makes the role of the rubber analyst one of the most satisfying within
the analytical chemistry profession.

356
A
ppendix 1 Standard Nomenclature System
for Rubbers

The nomenclature system used in the rubber industry is based on International

Organization for Standardisation, ISO 1629. The last letter of the identification code
defines the basic group to which the polymer belongs, whereas the first letters provide
more specific information relating to the polymer’s structure.

Appendix 1.1 – M Group: Rubbers having a saturated carbon–carbon

main chain

CFM Fluorochlorocarbon rubbers (e.g., copolymers of vinylidene fluoride

and chlorotrifluoroethylene).

CM Chlorinated polyethylene – two grades are available; one containing

36% chlorine by weight and the other 42% chlorine by weight.

CSM Chlorosulfonated polyethylene. The polymers contain varying

amounts of chlorine (20–45%) and sulfur (0.5–2.5%). The optimum
values are Cl (30%): S (1.5%).

EPDM Terpolymer of ethylene, propylene and a small amount of a third

diene monomer allowing the use of a sulfur-based cure system.

EPM Copolymer of ethylene and propylene having monomer ratios of

70–30 and 30–70 wt/wt.

FPM Fluorocarbon rubbers having fluoro/fluoroalkyl groups on the

saturated carbon backbone (e.g., copolymers of hexafluoropropylene
and vinylidene fluoride and copolymers of vinylidene fluoride and
1-hydropentafluoropropylene).

IM Polyisobutylene.

357
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Appendix 1.2 – O Group: Rubbers having carbon and oxygen in

the main chain

CO Polyepichlorohydrin.

ECO Copolymer of epichlorohydrin and ethylene oxide.

GPO Copolymer of propylene oxide and allyl glycidyl ether.

Appendix 1.3 – Q Group: Silicone rubbers

MQ Polydimethylsiloxane – can be an oil, wax or rubber depending on

molecular weight.

MFQ Fluorinated MQ – degrees of fluorination can vary.

MPQ MQ with the presence of some phenylmethyl siloxane units in the
backbone.

MPVQ MPQ with the presence of some vinylmethylsiloxane units in the

backbone.

Appendix 1.4 – R Group: Rubbers having an unsaturated carbon

backbone

ABR Copolymers of butadiene and methyl methacrylate. Also, terpolymer

of butadiene, methyl methacrylate and acrylonitrile and tetrapolymer
of butadiene, methyl methacrylate, acrylonitrile and styrene.

BIIR Brominated isobutylene-isoprene rubber 2–3% w/w.

BR Polybutadiene – can vary greatly in trans-1,4, cis-1,4 and vinyl content.

General purpose rubbers are usually 90% cis-1,4 or approximately
45% cis-1,4, 45% trans-1,4 and 10% vinyl.

CIIR Chlorinated isobutylene-isoprene rubber 2–3% w/w.

CR Polychlorobutadiene – there are two principal types: G types, which

are amber copolymers with sulfur and have a molecular weight of
≈100,000 and W types, which are white homopolymers and have a
molecular weight of ≈200,000.

358
 Appendix 1 Standard Nomenclature System for Rubbers

IIR Copolymer of isobutylene and a small amount of isoprene (i.e., butyl

rubber) to provide cure sites.

IR Synthetic cis-1,4-polyisoprene. Cis-1,4 level 90–99% with the

remainder trans-1,4 and vinyl.

NBR Copolymer of acrylonitrile and butadiene – acrylonitrile content of

the random copolymer can vary between 15 to 50% to change the
degree of oil resistance. Two important modified versions of nitrile
rubber are hydrogenated NBR (in which the double bonds in the
butadiene segments are removed by hydrogenation) and carboxylated
NBR which, in addition to acrylonitrile and butadiene units, contains
some carboxyl groups.

NR Natural rubber – NR consists of 96% polymer which is virtually

100% cis-1,4 with trans-1,4 and vinyl content at <0.1%. The other
4% is made of natural products such as proteins. There is a wide
variety of NR grades [e.g., Rib Smoked Sheet (RSS) and Standard
Malaysian Rubber (SMR)]. These denominations are accompanied
by numbers denoting purity and quality (e.g., RSS 1 and SMR 10).
There are also modified natural rubbers available, such as epoxidised
natural rubber (ENR).

SBR Copolymer of styrene and butadiene – the styrene level in the random
copolymer can vary between 10 and 80%, but the general purpose
grade contains ≈23%. Many grades are available and these are
identified by a specific coding system.

Appendix 1.5 – T Group: Rubbers having carbon, oxygen and

sulfur in the main chain

EOT OT monomers polymerised with ethylene dichloride (e.g., polyethylene

disulfide and polybutyl ether disulfide).

OT Copolymer of bis-chloroalkyether (or formal) with sulfur (e.g., bis-

2-chloroethylformal, sulfur and 1,2,3-trichloropropane as a cure site
monomer).

359
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Appendix 1.6 – U Group: Rubbers containing carbon, oxygen and

nitrogen in the main chain

AU Polyester-type polyurethanes.

EU Polyether-type polyurethanes.

A wide range of polyurethanes can be produced by making use of a number of different

isocyanates, polyols and chain extenders.

Appendix 1.7 – Others Rubbers – General designations and

descriptions

ACM Acrylic rubbers.

Polynorbornene.

Polyfluorophophazene rubbers.

Ethylene acrylic rubber.

EVA Ethylene vinyl acetate (elastomeric in the range 40–60% vinyl acetate).

Appendix 1.8 – Other polymer-related acronyms which are

commonly encountered in the rubber industry include

HNBR Hydrogenated nitrile rubber.

TPR Thermoplastic rubber (e.g., MG Rubber, NR that has had methyl

methacrylate grafted onto it to produce a material containing 30 or
49% methyl methacrylate).

XNBR Carboxylated nitrile rubber.

Appendix 1.9 – Other Other naturally occurring polyisoprenes

Balata 100% trans-1,4 polyisoprene

Chicle Mixture of 25% cis-1,4 and 75% trans-1,4 polyisoprene

Guayule 100 cis-1,4 polyisoprene

GuttaPercha 100% trans-1,4 polyisoprene

360
A
ppendix 2 International Rubber Analysis
Standards (ISO)

Appendix 2.1 General ISO Standards

Standard Title
No.
565 Test sieves – Metal wire cloth, perforated metal plate and
electroformed sheet – Nominal sizes of openings
1382 Rubber - Vocabulary
1407 Rubber – Determination of solvent extract
1408 Rubber – Determination of carbon black content – Pyrolytic and
chemical degradation methods
1629 Rubber and latices – Nomenclature
2194 Industrial screens – Woven wire cloth, perforated plate and
electroformed sheet – Designation and nominal sizes of openings
2781 Rubber, vulcanised – Determination of density
3104 Petroleum products – Transparent and opaque liquids –
Determination of kinematic viscosity and calculation of dynamic
viscosity
3105 Glass capillary kinematic viscometers – Specifications and operating
instructions
3865 Rubber, vulcanised or thermoplastic – Methods of test for staining in
contact with organic material
4645 Rubber and rubber products – Guide to the identification of
antidegradants – Thin-layer chromatographic methods
4648 Rubber, vulcanised or thermoplastic – Determination of dimensions of
test pieces and products for test purposes
4650 Rubber – Identification – Infrared spectroscopic methods
4661-1 Rubber, vulcanised or thermoplastic – Preparation of samples and test
pieces – Part 1: Physical tests
4661-2 Rubber, vulcanised - Preparation of samples and test pieces – Part 2:
Chemical tests
5945 Rubber – Determination of polyisoprene content

361
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

6528-1 Rubber – Determination of total sulfur content – Part 1: Oxygen

combustion flask method
6528-2 Rubber – Determination of total sulfur content – Part 2: Sodium
peroxide fusion method
6528-3 Rubber – Determination of total sulfur content – Part 3: Furnace
combustion method
7269 Rubber – Determination of free sulfur
7725 Rubber and rubber products – Determination of bromine and chlorine
content – Oxygen flask combustion technique
8054 Rubber, compounded or vulcanised – Determination of sulfide sulfur
content - Iodometric method
9924-1 Rubber and rubber products – Determination of the composition of
vulcanisates and uncured compounds by thermogravimetry – Part 1:
Butadiene, ethylene-propylene copolymer and terpolymer, isobutene-
isoprene, isoprene and styrene-butadiene rubbers
9924-2 Rubber and rubber products – Determination of the composition of
vulcanisates and uncured compounds by thermogravimetry – Part 2:
Acrylonitrile-butadiene and halobutyl rubbers
9924-3 Rubber and rubber products – Determination of the composition of
vulcanisates and uncured compounds by thermogravimetry – Part 3:
Hydrocarbon rubbers, halogenated rubbers and polysiloxane rubbers
after extraction
10398 Rubber – Identification of accelerators in cured and uncured
compounds
10638 Rubber – identification of antidegradants by gas chromatography/
mass spectroscopy
11089 Rubber, raw synthetic – Determination of anti-degradants by high-
performance liquid chromatography
15671 Rubber and rubber additives – Determination of total sulfur content
using an automatic analyser
15672 Rubber and rubber additives – Determination of total nitrogen
content using an automatic analyser

Appendix 2.2 Latex ISO Standards

Standard Title
No.
123 Rubber latex – Sampling
124 Latex, rubber – Determination of total solids content

362
 Appendix 2 International Rubber Analysis Standards (ISO)

125 Natural rubber latex concentrate – Determination of alkalinity

126 Latex, rubber, natural concentrate – Determination of dry rubber
content
127 Rubber, natural latex concentrate – Determination of KOH number
498 Natural rubber latex concentrate – Preparation of dry films
506 Rubber latex, natural, concentrate – Determination of volatile fatty
acid number
705 Rubber latex – Determination of density between 5 and 40 °C
706 Rubber latex – Determination of coagulum content (sieve residue)
976 Rubber and plastics – Polymer dispersions and rubber latices –
Determination of pH
1147 Plastics/rubber – Polymer dispersions and synthetic rubber lattices –
freeze–thaw cycle stability test
1652 Rubber latex – Determination of apparent viscosity by the Brookfield
test method
1657 Rubber, raw and rubber latex – Determination of iron content –
1,10-Phenanthroline photometric method
1802 Natural rubber latex concentrate – Determination of boric acid
content
2004 Natural rubber latex concentrate – Centrifuged or creamed,
ammonia-preserved types – Specification
2005 Rubber latex, natural, concentrate – Determination of sludge content
2008 Rubber latex, styrene-butadiene – Determination of volatile
unsaturates
2028 Synthetic rubber latex – Preparation of dry polymer
2438 Rubber latex, synthetic – Codification
3136 Rubber latex – Styrene-butadiene – Determination of bound styrene
content
6101 Rubber – Determination of metal content by atomic absorption
spectrometry (5 parts)
7780 Rubbers and rubber latices – Determination of manganese content –
Sodium periodate photometric methods
8053 Rubber and latex – Determination of copper content – Photometric
method
11852 Rubber – Determination of magnesium content of field natural rubber
latex by titration
12000 Plastics/rubber – Polymer dispersions and rubber lattices (natural and
synthetic) – Definitions and review of test methods

363
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

12243 Medical gloves made from natural rubber latex – Determination of

water-extractable protein using the modified Lowry method
13741-1 Plastics/rubber – Polymer dispersions and rubber latices (natural and
synthetic) – Determination of residual monomers and other organic
components by capillary-column gas chromatography – Part 1 Direct
liquid injection method
13741-2 Plastics/rubber – Polymer dispersions and rubber latices (natural and
synthetic) – Determination of residual monomers and other organic
components by capillary-column gas chromatography – Part 2
Headspace method
13773 Rubber – Polychloroprene latex – Determination of alkalinity

Appendix 2.3 Carbon Black ISO Standards

Standard Title
No.
1124 Rubber compounding ingredients – Carbon black shipment sampling
procedures
1125 Rubber compounding ingredients – Carbon black – Determination of
ash
1126 Rubber compounding ingredients – Carbon black – Determination of
loss on heating
1138 Rubber compounding ingredients – Carbon black – Determination of
sulphur content
1304 Rubber compounding ingredients – Carbon black – Determination of
iodine adsorption number – Titrimetric method
1437 Rubber compounding ingredients – Carbon black – Determination of
sieve residue
1867 Carbon black for use in the rubber industry – Specification for sieve
residue
1868 Rubber compounding ingredients – Carbon black – Specification
limits for loss on heating
3858-1 Carbon black for use in the rubber industry – Determination of light
transmittance of toluene extract – Part 1: Rapid method
3858-2 Carbon black for use in the rubber industry – Determination of
light transmittance of toluene extract – Part 2: Method for product
evaluation
4652-1 Rubber compounding ingredients – Carbon black – Determination of
specific surface area by nitrogen adsorption methods – Part 1: Single-
point procedures

364
 Appendix 2 International Rubber Analysis Standards (ISO)

4656-1 Rubber compounding ingredients – Carbon black – Determination

of dibutyl phthalate absorption number – Part 1: Method using
absorptometer
5435 Rubber compounding ingredients – Carbon black – Determination of
tinting strength
6209 Rubber compounding ingredients – Carbon black – Determination of
solvent extractable material
6810 Rubber compounding ingredients – Carbon black – Determination of
surface area – CTAB adsorption methods
6894 Rubber compounding ingredients – Carbon black – Preparation of
samples for determination of dibutyl phthalate absorption number
(compressed sample)
12245 Carbon blacks used in rubber products – Classification system

Appendix 2.4 Standard ISO Test Formulations

Standard Title
No.
1658 Natural rubber (NR) – Evaluation procedure
2302 Isobutene-isoprene rubber (IIR) – Evaluation procedures
2322 Styrene-butadiene rubber (SBR) – Emulsion and solution-polymerised
types – Evaluation procedures
2475 Chloroprene rubber (CR) – General-purpose types – Evaluation
procedure
2476 Rubber, butadiene (BR) – Solution polymerised types – Evaluation
procedures
3257 Rubber compounding ingredients – Carbon black – Method of
evaluation in styrene butadiene rubbers
4097 Rubber, ethylene-propylene-diene (EPDM) – Evaluation procedure
4659 Rubber, styrene-butadiene (carbon black or carbon black and oil
masterbatches) – Evaluation procedure

Appendix 2.5 Statistical Standards

Standard Title
No.
2602 Statistical interpretation of test results – Estimation of the mean –
Confidence interval

365
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

2854 Statistical interpretation of data – Techniques of estimation and tests

relating to means and variances
2859 Sampling procedures for inspection by attributes (5 parts)
3207 Statistical interpretation of data – Determination of a statistical
tolerance interval
3301 Statistical interpretation of data – Comparison of two means in the
case of paired observations
3494 Statistical interpretation of data – Power of tests relating to means
and variances
3534 Statistics – Vocabulary and symbols (3 parts)
5725 Accuracy (trueness and precision) of measurement methods and
results (6 parts)

Appendix 2.6 Raw and Compounded Rubber Standards

Standard Title
No.
247 Rubber – Determination of ash
248 Rubbers, raw – Determination of volatile matter content
249 Rubber, raw natural – Determination of dirt content
1407 Rubber – Determination of solvent extract
1434 Natural rubber in bales – Amount of bale coating – Determination
1656 Rubber, raw natural, and rubber latex, natural – Determination of
nitrogen content
1657 Rubber, raw and rubber latex – Determination of iron content –
1,10-Phenanthroline photometric method
1795 Rubber, raw natural and raw synthetic – Sampling and further
preparative procedures
2000 Rubber, raw natural – Specification
2453 Rubber, raw styrene-butadiene, emulsion polymerised –
Determination of bound styrene content – Refractive index method
2454 Rubber products – Determination of zinc content – EDTA titrimetric
method
3899 Rubber – Nitrile latex – Determination of residual acrylonitrile
content
3900 Rubber – Nitrile latex – Determination of bound acrylonitrile content

366
 Appendix 2 International Rubber Analysis Standards (ISO)

4655 Rubber – Reinforced styrene-butadiene latex – Determination of total

bound styrene content
4660 Rubber, raw natural – Colour index test
5945 Rubber – Determination of polyisoprene content
6101 Rubber – Determination of metal content by atomic absorption
spectrometry (5 parts)
6235 Rubber, raw – Determination of block polystyrene content –
Ozonolysis method
7270 Rubber – Identification of polymers (single polymers and blends) –
Pyrolytic gas chromatographic method
7780 Rubbers and rubber latices – Determination of manganese content –
Sodium periodate photometric methods
7781 Rubber, raw styrene-butadiene – Determination of soap and organic-
acid content
8053 Rubber and latex – Determination of copper content – Photometric
method
11089 Rubber, raw synthetic – Determination of anti-degradants by high-
performance liquid chromatography
11344 Rubber, raw synthetic – Determination of the molecular-mass
distribution of solution polymers by gel permeation chromatography
12492 Rubber, raw – Determination of water content by Karl Fischer
method
19050 Rubber, raw, vulcanised – Determination of metal content by ICP–
OES
22768 Rubber, raw – Determination of the glass transition temperature by
differential scanning calorimetry

367
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

368
A
ppendix 3 Specific Gravities of Rubbers and
Compounding Ingredients

Appendix 3.1 Specific gravities of polymers

Polymer Specific gravity (g/cm3)

Acrylic rubbers 1.1

Bromobutyl rubber 0.91

Butyl rubber 0.91

Chlorobutyl rubber 0.91

Chlorosulfonated polyethylene 1.11–1.28*

Epichlorohydrin copolymer 1.27

Epichlorohydrin homopolymer 1.38

Ethylene-propylene diene monomer rubber 0.86

Ethylene-propylene rubber 0.85

Ethylene-vinyl acetate copolymer 0.94

Fluorocarbon rubbers 1.41–1.86*

High styrene-butadiene rubber 1.05

Natural rubber 0.92–0.93

Nitrile rubbers 0.95–1.00*

Nitrile/polyvinyl chloride blends 1.0–1.11*

Polybutadiene rubbers 0.91–0.93

369
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Polychloroprene rubbers 1.23–1.25*

Polyisoprene rubber 0.92–0.93

Polynorbornene 0.96

Polysulfide rubbers 1.27–1.60*

Styrene-butadiene rubbers 0.94

*According to grade

Appendix 3.2 Specific Gravities of Compounding Ingredients

Ingredient Specific gravity (g/cm3)

Aluminium silicate 2.10

Antimony trioxide 5.40

Barium sulfate 4.30

Bitumen 1.04

Calcined china clay 2.50

Calcium carbonate 2.65

Calcium oxide (lime) 2.19

Calcium silicate 2.10

Carbon black 1.80

Carnauba wax 0.99

China clay (kaolin) 2.60

Cotton fibre 1.05

Diethylene glycol 1.12

Factice 1.05

370
 Appendix 3 Specific Gravities of Rubbers and Compounding Ingredients

Graphite 2.04

Indene-coumarone resin 1.09

Iron oxide 4.5–5.1

Lanolin 1.08

Litharge 9.30

Magnesium carbonate 2.21

Magnesium oxide 3.60

Mica 2.80

Mineral oil 0.91

Oleic acid 0.90

Paraffin oil 0.80

Paraffin wax 0.90

Pine tar 1.08

Rosin 1.07

Silica 1.95

Stearic acid 0.85

Sulfur 2.05

Talc (magnesium silicate) 2.80

Titanium dioxide (anatase) 3.90

Titanium dioxide (rutile) 4.20

Zinc oxide 5.55

Zinc stearate 1.10

371
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

372
A
bbreviations

3D Three-dimensional

6PPD N-(1,3-dimethyl butyl)-N'-phenyl-p-phenylenediamine

AAS Atomic absorption spectrometry

ACN Acrylonitrile

AES Atomic emission spectroscopy

AET Analytical evaluation threshold

Am-Ep Amine-epoxy

AP Atmospheric pressure

APCI Atmospheric pressure chemical ionisation

API Atmospheric pressure ionisation

ASTM American Society for Testing and Materials

ATD Automated thermal desorption

ATR Attenuated total reflectance

BET Brunauer–Emmett–Teller

BHT Butylated hydroxyltoluene

BPOG BioPhorum Operations Group

BR Polybutadiene

BS British Standard

373
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

BT Benzothiazole

CB Carbon black(s)

CF Carbon fibre

CBS N-cyclohexyl-2-benzothiazole sulfenamide

CCS Collisional cross-section

CFR Code of Federal Regulations

CGNR Chemically grafted with cardonol

CIIR Chlorobutyl rubber

CNR Chlorinated natural rubber latex

COC Cyclic olefin copolymers

CoE Council of Europe

CR Polychloroprene

CSM Chlorosulfonated polyethylene rubber

CTAB Cetyltrimethylammonium bromide

CTNR Carboxyl-terminated nitrile rubber

DAE Distillate aromatic

DBP Dibutyl phthalate

DCBS N,N-dicyclohexyl-2-benzothiazyl sulfenamide

DCM Dichloromethane

DDA Dynamic dielectric analysis

DEA Dielectric analysis

DIN Deutsches Institut für Normung e.V.

DMA Dynamic mechanical analysis

374
 Abbreviations

DOTG N,N'-di-ortho-tolyl guanidine

DPG Diphenyl guanidine

DSC Differential scanning calorimetry

DTDM 4,4'-Dithiodimorpholine

E&L Extractables and leachables

EC European Commission

EDQM European Directorate for the Quality of Medicines & Healthcare

EDX Energy-dispersive X-ray spectroscopy

EI Electron ionisation

ELISA Enzyme-linked immunosorbent assay

ELS Evaporative light scattering

EMA Ethylene methyl acrylate copolymer

EPDM Ethylene propylene diene monomer

EPM Ethylene propylene monomer

EU European Union

EV Efficient vulcanisation

EVA Ethylene-vinyl acetate

FDA Food and Drug Administration

FEA Finite element analysis

FEF Fast extrusion furnace

FID Flame ionisation detector

FR Fluorocarbon rubber

FSA Food Standards Agency

375
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

FT Fourier–Transform

FTIR Fourier–Transform infrared

GC Gas chromatography

GPC Gel permeation chromatography

GRG General rubber goods

GTM Gas transfer mould

HAF High-abrasion furnace

HFA Higher fatty acid

HPLC High-performance liquid chromatography

HRAM High-resolution accurate mass

HSE Health and Safety Executive

IC Ion chromatography

ICBA International Carbon Black Association

ICH International Council for Harmonisation of Technical Requirements

for Pharmaceuticals for Human Use

ICP Inductively coupled plasma

IGC Inverse gas chromatography

IMS Ion-mobility spectrometry

IPPD Isopropyl p-phenylenediamine

IR Infrared

ISO International Organization for Standardization

LABA Low ammonia with 0.25% boric acid and 0.05% sodium
pentachlorophenate

376
 Abbreviations

LC Liquid chromatography

LDPE Low-density polyethylene

LIMA Laser-ionisation mass analysis

MAA Methacrylic acid

MALDI Matrix-assisted laser desorption/ionisation

MALLS Multi-angle laser light scattering

MBS 2-(4-Morpholinothio)benzothiazole

MBT 2-Mercaptobenzothiazole

MBTS 2,2'-Dithiobis(benzothiazole)

MDHS Method for Determination of Hazardous Substance

MDR Moving die rheometer

MMT Montmorillonite

Mn Number average molecular weight

MPTD Dimethyl diphenyl thiuram disulfide

MS Mass spectrometry

MS2 Tandem mass spectrometry

MW Molecular weight(s)

Mw Weight average molecular weight

MWD Molecular weight distribution

Mz Z average molecular weight

NBR Nitrile rubber(s)

NIR Near-infrared

NMR Nuclear magnetic resonance

377
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

NOX Nitrosating agents

NP p-nonylphenol

NR Natural rubber

NVA Non-volatile acids

NVR Non-volatile residue

ODR Oscillating disc rheometer

OES Optical emission spectrometry

OML Overall migration limits

PA Polyacrylate

PAH Polyaromatic hydrocarbons

PAN N-phenyl-1-naphthylamine

PBT Polybutylene terephthalate

PC Polycarbonate

PDI Polydispersity index

PDMS Polydimethylsiloxane

PE Polyethylene

PI Polyisoprene

PP Polypropylene

PQRI Product Quality Research Institute

PS Polystyrene

PTBP p-tert-butylphenol

PTOP p-tert-octylphenol

PU Polyurethane

378
 Abbreviations

PVC Polyvinyl chloride

PVI Prevulcanisation inhibitor

Py Pyrolysis

QC Quality control

R&D Research and development

REACH Registration, Evaluation, Authorisation and Restriction of Chemicals

RI Refractive index

RIC Reconstructed ion current

RPA Rubber process analyser

RSD Relative standard deviation

RT Retention time

SANS Small-angle neutron scattering

SBR Styrene-butadiene rubber(s)

SBS Styrene-butadiene-styrene

SEC Size exclusion chromatography

SEM Scanning electron microscopy

SFC Supercritical fluid chromatography

SFM Scanning force microscopy

SIMS Secondary-ion mass spectrometry

SML Specific migration limits

SPM Scanning probe microscopy

SUS Single-use systems

TAC Triallyl cyanurate

379
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

TBBS N-tert-butyl-2-benzothiazole sulfenamide

TBSI N-tert-butyl-2-benzothiazole sulfenimide

TBTD Tetrabutylthiuram disulfide

TC Technical Committee

TCT Thermal desorption cold trap injector

TD Thermal desorption

TDEC Tellurium diethyldithiocarbamate

TEA Thermal energy analyser

TEM Transmission electron microscopy

TETD Tetraethylthiuram disulfide

Tg Glass transition temperature

TGA Thermogravimetric analysis

THF Tetrahydrofuran

THM Thermally-assisted hydrolysis and methylation

TIC Total ion current

TLC Thin-layer chromatography

Tm Melting temperature

TMA Thermal mechanical analysis

TMQ Poly-2,2,4-trimethyl-1,2-dihydroquinoline

TMTD Tetramethylthiuram disulfide

TMTM Tetramethylthiuram monosulfide

ToF Time-of-flight

TPE Thermoplastic elastomer(s)

380
 Abbreviations

TS Tensile strength

UPLC Ultra-performance liquid chromatography

US United States

USP United States Pharmacopeia

UV Ultraviolet

UV-Vis Ultraviolet visible

VFA Volatile fatty acid

VTA Variable temperature analysis

XRD X-ray diffraction

XRF X-ray fluorescence spectroscopy

XSBR Carboxylated styrene-butadiene rubber

ZDEC Zinc diethyldithiocarbamate

ZDMC Zinc dimethyldithiocarbamate

ZnO Zinc oxide

ZST Zinc stability time

381
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

382
 I
ndex

α-Methylstyrene, 283
1,1-Bis(tert-butylperoxy)cyclohexane, 283
1,2-and1,4-butadiene, 73
1,2-Dihydro-2,2,4-trimethylquinoline, 244
1,2-Naphthoquinone-1-anil, 295
1,2-Napthoquinone-2-anil, 287
1,2,3-Trichloropropane, 359
1,10-Phenanthroline photometric method, 363, 366
1974/75BRMA/Rapraenvironmentalsurvey, 217
1-Chloro-3-methylbenzene, 58
1-Chloro-4-(1-chloroethenyl)cyclohexene, 58
abcd 1-Hydropentafluoropropylene, 357
1-Methyl-4-(1-methylethenyl)cyclohexene, 79
abcd 1-Octanethiol/dioctyldisulfide, 299
2-(4-Morpholinothio)benzothiazole (MBS), 133, 200, 285
abcd 2-(4-Morpholinyl)benzothiazole, 215
2-(Methylthio)-benzothiazole, 254
abcd 2,2',4,4'-Tetrachlorobiphenyl, 283
2-Cyanoethyldiethyldithiocarbamate, 280
a b c d 2-Cyanoethyldimethyldithiocarbamate, 280
2-Cyanoethylmethylphenyl-dithiocarbamate, 295
abcd 2-Cyclohexyl benzothiazole, 215
2-Cyclopentyl-1,3,5-trimethylbenzene, 263
a 2,2'-Dithiobis(benzothiazole) (MBTS), 133, 174, 260, 284, 295, 299
2,2'-Methylene-bis(4-methyl-6-tert-butylphenol, 157
αβχδ 2,5-Dimethyl-2,5-(di-tert-butylperoxy)hexane, 161, 249
2,6-Di-tert-butyl-1,4-benzoquinone, 287
❁ 2-Ethylhexanol, 281
2-Ethylhexyldiphenylphosphate, 172-173, 178
2-Mercaptobenzothiazole (MBT), 133, 145, 148, 162, 174, 196, 252, 258, 260-
262, 266, 284-285, 289-290, 295, 300, 317
2-Mercaptoimidazoline, 244
2-Methylpropylbutanoate, 172

383
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

3-(Bromomethyl)-heptane, 171-172
3,5-Dibromo-p-benzoquinonechlorimine, 123
3DChromatogram total ioncurrent, 121
4-(1,1,3,3-Tetramethylbutyl)-phenol, 172
4-Amino-benzene sulfuricacid, 123
4-Chloro-1,2-dimethylbenzene, 58
4-Cyclohexenitrile, 215
4,4'-Dithiodimorpholine (DTDM), 283, 285
4,4'-Thio-bis(2-tert-butyl-5-methyl) phenol, 261
4,4'-Thio-bis(3-methyl-6-tert-butylphenol), 134
4-Ethyl-2-propyl-thiazole, 254
4-Vinylcyclohexene, 79, 215-216
6-Ethoxy-1,2-dihydro-2,2,4-trimethylquinoline, 123, 162
A
ABAQUS software, 202-203
Absorption bands, 14, 53, 61, 72, 98, 180
Absorptometer, 365
Accelerated sulfur cure systems, 304
Accelerator diphenyl guanidine, 260
Accelerators, 4, 52, 104, 128-129, 134-136, 164, 174, 178, 193, 220, 227-228,
256, 260, 266, 280, 285, 288, 315, 317, 322, 334, 339, 345, 350, 356, 362
Acetaldehydes, 259
Acetates, 120
Acetic acid, 240-241, 262, 266, 280, 335, 337-338
Acetone, 11, 80, 117, 120-122, 129-130, 132, 157-160, 163, 174, 179, 249-251,
261, 283, 287, 296-297, 300, 329, 330, 336, 338, 350
Acetonitrile, 9, 118, 251, 267-268, 304, 330
Acetonitrile/water gradient elution, 9
Acetophenone, 172, 225, 261, 283
Acetylene black, 86, 90
Acidic compounds, 222
Acidifying stabilised latex, 337
Acidimetric titration, 338
Acrylic–Vamac®, 8, 164, 165
Acrylic rubber, 114, 115, 131-133, 161, 163, 193, 360, 369
acetone extract, 131
Acrylonitrile (ACN), 64-65, 80-81, 136-137, 176, 215, 241, 243, 258, 280, 295,
299, 333, 341, 358-359
Activated charcoal, 220, 222
adsorbent sampling tube, 218
Activators, 145

384
 Index

Adaptive Neuro–Fuzzy Interference Analysis Systems (ANFIS), 203

Additives, 2, 4, 7, 10-12, 14-15, 18, 20, 25, 27-30, 32, 38-39, 43-44, 52-53, 69,
71, 73, 77-79, 84, 91, 96, 98, 104, 113, 115-116, 119, 124, 126-128, 132, 138,
140, 143-148, 154-156, 164, 170, 181, 188, 191-192, 193, 195, 198, 227, 236,
241, 243, 245, 248, 256, 278, 280-282, 290, 326, 330-331, 334, 345, 347
Adhesives, 3, 334
Adipates, 82, 256
Advanced Polymer Analyzer 2000, 349
Aerosol, 213-214, 216-219, 221, 224-225, 228-229
compounds, 218
Agilent 6890 gas chromatograph, 157, 250
with LECO Pegasus III GC×GC–time-of-flight–mass spectrometry, 120
Agilent 6890/5973 gas chromatography–mass spectrometry, 121, 160
Alcohols, 26, 213, 282
Aldehydes, 213, 218-219, 267
Aliphatic
carboxylic acids, 172
compounds, 222
hydrocarbons, 172, 212-222, 300
Alkalinity, 336-337, 364
Alkanes, 222
Alkenes, 263, 279, 282
Alkyl
aryl sulfonates, 346
phenol formaldehyde resins, 142
phenolic resins, 78
sulfates, 346
Alkylbenzenes, 299
Alkylphenols, 258
Allergen tests, 339
Allyl glycidyl ether, 358
Alumina, 119, 140, 178
Aluminium, 141, 169-170, 175-176, 322
foillaminate packaging, 350
oxide, 169
sheets, 226
silicates, 281, 370
trihydroxide, 141
American Society for Testing of Materials (ASTM), 41, 85
D1510, 94
D2414, 95
D3037, 94

385
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

D3156, 119
D3765, 94
D4026-06, 341
D5712, 339
D6204 Part B, 200
D6204 Part C, 199
D6214, 339
D6499, 339
D7427, 339
D7558, 339
D7771-11, 109
E698, 190
F-433, 41
Amine (ArNH2), 300
antioxidant, 319
compounds, 281
N,N'-di-o-tolyl guanidine, 133
Amines, 214, 218-219, 241, 284-286
-epoxy (Am-Ep), 128
Amino acid degradation, 227
Ammonia, 295, 333
Ammonium
hydroxide, 346
sulfate, 337
Amorphous
materials, 74
plastics, 139
polymers, 33
Analyte, 15, 21, 25, 27, 119, 157, 255, 290, 345
Analytical evaluation threshold (AET), 288
Aniline, 215, 260, 265, 285, 295, 296, 299
ANSYS finite element software, 202
Antidegradant, 2, 21, 32, 52, 113, 115-119, 130, 155, 163, 166, 245, 258, 266,
271, 281, 298, 315, 362
manufacturer, 118
poly-2,2,4-trimethyl-1,2-dihydroquinoline, 127
system, 104
Antigenic protein, 339
Antimony, 170, 174, 176
trioxide, 18, 104, 140-141, 169, 176, 178, 370
Antioxidant(s), 4, 15, 20, 21, 25, 30, 32, 38, 104, 113, 118-120, 123, 125-127,
132, 134, 144-145, 174, 214, 226-227, 229, 263, 286-287, 292, 317, 345, 346

386
 Index

Antiozonant, 104, 113, 118, 123, 134, 144, 260, 287

Aqueous foods, 244
Aromatic
amines, 212, 242, 246, 260, 286
compounds, 220, 222
hydrocarbons, 213
processoils, 220
Arrete, 242
Arrhenius-type equation, 193
Artificial saliva, 262
Ashing, 12, 96,-97, 140, 356
As-received thermogram, 31
Guayule ELISA ASTM WG 25943, 339
method, 95
Atmospheric chemical ionisation, 26
Atmospheric pressure chemical ionisation (APCI), 252, 256, 292
Atmospheric pressure (AP), 26, 146-147, 256, 301
-solidsanalysis, 146
ionisation (API), 147
photoionisation, 26, 256
pyrolysis mass spectrometry, 146
Atomic absorption spectrometry (AAS), 17, 259, 302, 344, 363, 367
Atomic force microscopy, 42, 93
Atomic absorption spectroscopy, 118
Atomic force microscope, 35, 42
Atomic number, 19
Attenuated total reflectance (ATR), 10, 12, 53, 79, 320-322
absorption bands, 180
–Fourier–Transform infrared method, 52, 181, 350-351
spectra, 180
spectroscopy, 145, 320, 350
techniques, 301
Autocatalytic model, 201
Automated analyser, 9
Automated thermal desorption (ATD), 297
Automatic
analyser, 362
gas switching, 87
industry, 3
sealing applications, 198
AutoStepwise approach, 100, 182-182
thermogravimetric analysis method, 182

387
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Auxiliary preservative, 344

Azo
dyes, 282
compounds, 138
B
Benzylisothiocyanate, 228
Balata, 360
Bale coating, 366
Barium
perchlorate, 8
sulfate (BaSO4), 8, 95, 98, 370
Barytes, 95
Basic compounds, 222
Bay proton, 85-86
Bedarfsgegenständeverordnung, 242
Benz(a)pyrene, 221
Benzene, 172, 283
4-cyanocyclohexene, 263
isothiocyanate, 263
Benzidine, 215
Benzo-alpha-pyrene, 85, 95
Benzonitrile, 58
Benzothiazole (BT), 118, 133, 136, 172, 174, 215, 254, 260, 295, 299
Beryllium, 19
BfR inventory, 241
BfR Recommendation
XLVII, 240
XV, 241-242, 269
XXI, 155, 240-241
regulations, 241-242, 244
Biopharmaceutical industry, 305
BioPhorum Operations Group’s (BPOG), 305
Biphenyl, 283
Bis(2,4-dichlorobenzoyl)–m-dichlorobenzene, 283
Bis(2-ethylhexyl)phthalate, 225
Bis(2-hydroxy-3-tert-butyl-5-ethylphenyl)methane, 261
Bisphenol A, 258
Bisphenol-type contaminants, 259
Bitumen, 370
Blooming, 313-315, 319, 326
Blowing agents, 115, 139

388
 Index

Borchardt and Daniels method, 190

Boric acid, 345, 363
Boron, 175, 176
Bound styrene value, 340
Breon N36C60 [acrylonitrile-butadiene copolymer (nitrile) rubber, 294-295
British Standard (BS),
BS 3397:1976, 336
BS 7164-22.2:1992, 175
BS 7164-23, 8
Brominated butyl rubber, 197, 302
Brominated isobutylene-isoprene rubber, 358
Bromination, 343
Bromine, 170
Bromobutyl, 305
rubber, 290, 303, 369
Brookfield
instrument, 342
L instrument, 342
R instrument, 342
test method, 363
Brunauer–Emmett–Teller method (BET), 92
Bulk composition analysis, 153
Buna 1712, 228
Butadiene, 13, 35, 53, 57, 61, 65, 99, 114-115, 141, 215, 333
Butadiene, 57, 115, 358-359
rubber, 81
trimers, 215
Butoxyethyl phosphate plasticiser, 127
Butyl, 99
50B, 164, 165
51B unaged, 265
compound, 89
rubber, 61-63, 73, 81, 89, 102, 114-116, 120-123, 132, 162-163, 197, 215,
237, 369
tin compounds, 259
Butylated hydroxyltoluene (BHT), 117, 119, 259, 262, 287
C
C-nuclear magnetic resonance (C-NMR), 58, 72
Cabot Dispersion Classification Chart, 41, 90
Cabot graticule, 90
Calcined

389
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

china clay, 370

clay, 302
Calcium, 170
carbonate (CaCO3), 8, 37, 95-98, 281, 302, 318, 370
chloride, 340
oxide (lime), 96, 243, 319, 370
silicate, 243, 370
stearate, 243, 282
sulfate, 8
Capillary-column gas chromatography, 364
Captopril, 304
Captopril–tetrabutylthiuram disulfide disulfides, 304
Carbon
black (CB), 10-12, 37, 38, 41-43, 49, 51, 53, 64, 73, 78-79, 85-96, 98-105,
110, 117, 126, 128, 141, 154-155, 157, 161-162, 166, 169, 181-182, 220,
224, 227, 249, 254, 281-282, 286-288, 299, 361, 364-365, 370
dioxide (CO2), 337, 338
disulfide (CS2), 80, 136, 172, 216, 220, 224-225, 284, 295, 299, 345
fibres (CF), 90
   -reinforced vinyl ester resins, 193
   nanotubes, 87
   tetrachloride, 80
-filled ethylene propylene diene monomer compounds, 204
-filled rubber blends, 198
-filled rubber compounds, 86
-filled styrene-butadiene rubber rubbers, 204
-filled vulcanisates, 12, 182
filler, 161, 162
Carbon/hydrogen method, 341
Carbonaceous, 64-65, 86-87, 90, 97, 105, 170
additives, 105
material, 86, 97, 105, 170
residue, 64, 89-90, 99-100, 105, 169
Carbonyl
formation, 12
sulfide, 295
Carboxyl groups, 347
Carboxylated
nitrile rubber, 359-360
styrene-butadiene rubber (XSBR), 340, 349
Carboxylic acid, 295, 317, 347
(C14, C16 and C18), 295

390
 Index

Carboxyl-terminated nitrile rubber (CTNR), 77

Carcinogens, 220]
Carcinogenic aromatic amines, 260
Carcinogenic polyaromatic hydrocarbons, 227
Cardanol (m-pentadecenyl phenol), 84
Carnuba wax, 370
Casein, 346
Casting technique, 348
Castor oil, 202
Catalyst remnants, 279
Catalytic pyrolyser, 256
Centrifugation, 343, 349
Centrifuge methods, 344
Cetyltrimethylammonium bromide (CTAB) method, 92
adsorption methods, 365
CFR Title 21, 243-244
Charcoal tubes, 225
Chemical ionisation mass spectrometry, 16
Chemically grafted with Cardonol (CGNR), 85
Chemiluminescence analysis, 20
Chicle, 360
China Clay (kaolin), 370
Chlorinated
compounds, 222
hydrocarbons, 219
isobutylene-isoprene rubber, 358
natural rubber latex (CNR), 349
   paraffin, 141
polyethylene, 199, 357
Chlorine, 170, 357
Chlorobenzene, 58
Chlorobutyl rubber (CIIR), 76-77, 303, 358, 369
Chloroethene, 57
Chloroform, 12, 15, 53, 80, 117, 168, 171, 173, 323
solution, 12
Chloroprene, 58, 81, 279, 333
Chloroprene rubber (CR), 57-59, 63-64, 84, 114-115, 118, 123-124, 126, 225,
228, 237, 279, 358, 365
201E aged, 265
Chloroprene–sulfur copolymers, 2795
Chlorosulfonated polyethylene, 64, 77, 114-115, 357, 369
Chlorosulfonated polyethylene rubber (CSM), 64, 201, 357

391
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Chlorotrifluoroethylene, 357
Chromatographic techniques, 153
Chromophore, 15, 25
Cis-1,4-vinyl, 358
Cis-1,4-polyisoprene, 359-360
Classical organic chemistry methods, 7
Clays, 95
Cluff–Gladding–Pariser method, 204
Co-polymerisation, 348
Coagulum, 335, 340-341, 363
Coalescence, 343
Code of Federal Regulations (CFR), 243
Codification, 363
Cole–Coleplot, 201
Colour index test, 367
Collisional cross-section (CCS), 292
Column
mobile-phasecombination, 25
packings, 28
Combined sulfur, 8
Commercial rubber products, 3
Commission Regulation (EU) 1272/2013, 109
Commodity Articles Based on Natural and Synthetic Rubber, 240
Common rubbers, 57
Compact quadruple mass analyser, 18
Compatibilisation agent, 77
Compression mould(ing), 193, 217, 221, 225
Concentrated hydrochloric acid, 8
Condoms, 334
Conductivity detector, 27
Conductometric titration, 338, 343
Contact angle measurements, 348
Contaminants, 270, 272-274, 313, 327
Continuous collection software, 69
Controlled rate thermal analysis technique, 182
Conventional
differential scanning calorimetry, 33
emulsion polymerisation, 348
physical tests, 13
rubbers, 74
thermogravimetric analysis, 104
Copper spiral method, 9

392
 Index

Cotton, 329
fibre, 370
Coulter Counter, 344
Coumarone resins, 142
Council of Europe (CoE), 245
Resolution, 245, 247, 257, 269
Colourants [AP(89)1], 245
Cork Stoppers and Other Cork Materials [AP(2004)2], 245
Rubber Products [AP(2004)4], 245, 248, 258
Rubber Resolution [AP(2004)4], 245, 258
Silicone Products [AP(2004)5], 245
Coupled plasma-based methods, 344
Coupling agents, 61
Creaming agent, 344
Crosslinkable polyacrylate latices, 348
Crosslinking agent, 113, 347
Crystalline
plastics, 33
structures, 322
Crystallinity, 33-34, 36, 71, 74
Cumyl alcohol, 283
Curatives, 32, 52, 113, 115, 132, 138, 155, 163, 190, 192, 195, 204, 245, 256,
258, 271, 282, 305, 345-346
Cure
accelerators, 4
mechanism, 33
retarders, 104
rheometer, 201
Cure system, 2, 7, 21, 39, 53, 71, 78, 89, 96, 104, 116, 128-130, 133-134, 136-
138, 143, 154-155, 166, 169, 174, 178, 190, 192, 194, 196, 201, 216, 254,
267, 304, 328-330
Cured liquid silicone rubber, 259
Curedrubber, 129, 195, 199, 265, 314-316
Curie point, 55
pyrolysers, 56
Curing agents, 113, 129, 134, 145, 192, 203, 285
Cyanocyclohexene, 263, 299
Cyclic
olefin copolymers (COC), 290
   plastic, 303
oligomers, 267
organosiloxane oligomers, 269

393
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

siloxanes, 267
Cyclododecatriene, 215
Cyclohexanamine, 134
Cyclohexane, 80, 136, 204, 217, 221, 230, 351
Cyclohexane-soluble organic dust, 217
Cyclohexanethiol, 171-172, 174
Cyclohexanol, 80
Cyclohexanone, 80, 136, 172, 283
Cyclohexylamine, 256, 285
Cyclohexylcyclohexamine, 24
D
Decabromodiphenyl oxide, 141
Decane, 24
Defoamers, 248
Degree
of accuracy, 87, 223
of crystallinity, 74
of dispersion, 41, 90
of oxidation, 20
of silanisation, 200
Dehydrochlorination, 169
Derivative curve, 100, 101, 102
Desorption electrospray ionisation mass spectrometry, 147
Deuterated toluene, 93
Di-(2-ethylhexyl)sebacate, 82
Di-(2-ethylhexyl)adipate, 82
Di(2-ethylhexyl)phthalate, 82, 172, 178, 254, 281
Diacyl peroxides, 283
DiakTM, 161-162
Dialkyl peroxides, 282, 283
Diaminodiphenyl methane, 77
Diammonium phosphate, 197
Diamond ‘windows’, 10, 53
Diaryl-p-phenylenediamine antiozonants, 123
Dibenzothiazyl monosulfide, 295
Dibenzoyl peroxide, 283
Dibenzylamine, 261, 263
Dibutyl phthalate (DBP), 82, 93, 95, 221
absorption number, 365
Dibutyl sebacate, 82, 83
Dichloromethane (DCM), 80, 118, 249, 330

394
 Index

Dicumyl peroxide, 133, 193, 283, 326

Dicyclohexylamine, 215
Dicyclopentadiene, 254
Dielectric analysis (DEA), 40, 63
Diene rubbers, 53, 72, 324
Diene-type rubbers, 126, 287
Diethyl ether, 80, 118, 251, 261
Diethylammonium diethyldithiocarbamate, 345
Diethyldithiocarbamate, 253
Diethylene glycol, 370
Di-ethylhexylphthalate, 173
Differential scanning calorimetry (DSC), 30-37, 39, 40, 47-48, 59, 62-63, 71, 73-
77, 116, 123, 125-128, 139, 142, 189-194, 202, 204-205, 326, 348-347, 367
programme, 31
thermogram, 74
Differential thermal analysis, 349
Diffused-reflectance mode, 13
Diffuse-reflection Fourier–Transform infrared, 145
Diffusing mechanic alanalysis, 197
Digital imaging, 14, 41
Diisononyl phthalate, 25
Diisopropyl benzene, 259
Dimers, 57
Dimethyl-2-butanamine, 215
Dimethylamine, 130, 136, 263
Dimethylbenzamine, 24
Dimethyldiphenylthiuram disulfide, 162
Dimethyldithiocarbamate, 253
Dimethylformamide, 225
Dimethylphenol, 24
Dimethyl diphenyl thiuram disulfide (MPTD), 295
Dimethylsiloxane, 267
Dimethylsulfoxide, 80
Dinitrogen tetroxide (N2O4), 288
Dinitrogentrioxide (N2O3), 288
Dinitrophenyl hydrazine, 219
Dioctylphthalate, 82-83, 141
Dipentamethylene thiuram hexasulfide, 252
Diphenyl guanidine (DPG), 91, 285, 295
Diphenyl siloxane, 57
Diphenylamine, 178, 295, 296, 300
Diphenylamine/acetone, 178

395
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Diphenylcarbodiimide, 295
Diphenyl-p-phenylenediamine, 123
Direct
analysis, 12, 52, 239, 356
injection, 218
liquid injection method, 364
mass spectrometric techniques, 147
probe mass spectrometry, 16, 117
Disposable SUS, 306
Distillate aromatic (DAE), 84
Distortion, 343
Distyrenated diphenylamine-N-oxide, 296
Disulfides, 279
Di-tert-butyl peroxide, 283
Dithiocarbamates, 129, 318-319, 346
salts, 256
Divalent organic group, 282
Docosane, 326
Dodecenes, 300
Double-ring compounds, 283
Dry polymer, 363
Dry rubber, 363
Du Nouy tensiometer, 342
Dumbbells, 187
Dynamic dielectric analysis (DDA), 197
Dynamic headspace, 21, 130
gas chromatography–mass spectrometry method, 228
Dynamic mechanical analysis (DMA), 35, 37, 40, 58-59, 60, 61, 62-63, 71, 77-78,
116, 126, 127, 142, 153, 187, 194-197
Dynamic mechanical thermal analysis (DMTA), 141-142
Dynamic viscosity, 361
E
Efficient vulcanisation (EV), 194
EH40 publication, 212
Eicosane, 157, 249
Elastic modulus, 40, 59, 197
Elastic cord, 3, 334
Electrical conductivity testing, 338
Electromagnetic spectrum, 15
Electrometric titration, 336
Electron ionisation (EI) mode, 292

396
 Index

Electron-impact mass spectrometry, 16

Electrospray, 16, 26
ionisation, 147
ionisation mass spectrometry, 16
Elemental
impurities, 302
technique, 64, 97, 140
sulfur, 317, 318
Elongation, 58, 270
Emission infrared spectroscopy, 12
Emulsifier-free latices, 348
Emulsifiers, 248
Emulsion polymerisation, 279, 333
Emulsions, 345
End group analysis, 71
Endotherms, 30, 191
Energy-dispersive X-ray spectroscopy (EDX), 19, 41, 320-323, 327, 328
Engine gasket, 179
Epichlorohydrin, 358
copolymer, 369
homopolymer, 369
rubber, 64
Epichlorohydrin-ethylene oxide copolymer (hydrin) rubber, 162
Epoxidised oil, 128
Epoxidised natural rubber (ENR), 359
Epoxy
cresol novolac resin, 77
resin, 77, 194, 201
resin/ethylene propylene diene monomer rubber system, 194
Esters, 144, 213, 222, 227, 281
Ethanol, 80, 223, 241, 258, 269
Ethanol/water, 259
Ethanol–toluene azeotrope, 340
Ethers, 213
Ethyl
acetate, 259
acrylate rubber, 64
benzene, 215
palmitate, 223
Ethylene, 4, 13, 39, 74-75, 77, 99, 193, 223, 357
acrylic rubber, 360
dichloride, 359

397
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

glycol, 80
methyl acrylate copolymer (EMA), 78
norbornene, 223
oxide, 358
propylene diene, 4, 13, 193, 237, 289
propylene diene monomer (EPDM) rubber, 13, 15, 32, 36, 39, 63, 65, 71,
74-75, 77-78, 81, 86, 102, 114-115, 126, 128, 137, 144, 193, 198-199,
204, 215, 225-228, 237, 251-255, 290, 304, 357, 365, 369
   239C unaged, 265
   240C unaged, 265
   242C unaged, 265
   foams, 198
   polypropylene blend, 74
   thermoplastic rubbers, 127
propylene monomer (EPM), 63, 237, 357
propylene rubbers, 237, 369
vinyl acetate (EVA), 360
Ethylene-methyl acrylate (acrylic) rubber, 161
Ethylene-vinylacetate copolymers, 205, 280, 368
Ethyl hexanoic acid, 24
EU Directive
10/2001, 247
10/2011, 236
52/2009, 85, 109
93/11/EEC, 247, 262
1907/2006, 109
1935/2004, 239
2001/83/EC, 275
2002/72/EC, 247
2002/95/EC, 6
2005/69, 84
2023/2006, 239
EU-EXASRUB consortium, 221
European Pharmacopoeia, 245
European Directorate for the Quality of Medicines & Health Care (EDQM), 245
European Union regulations, 1
Evaporation, 338
Evaporative light scattering (ELS) detection, 68
Extender oils, 1, 84-85
Extensive washing, 340
Extractable Guayule proteins, 339

398
 Index

Extractables, 116, 135, 235, 244, 261, 270, 275-276, 277, 281, 289-294, 299-
303, 305-306, 323
Extractables and leachables (E&L), 116, 275
materials and safety database, 276
screening, 290
test, 275
Extraction gas chromatography–mass spectrometry, 330
Extruding/calendaring, 257
Extrusion, 211, 215
F
Factice, 143, 370
Failure diagnosis, 51
Fast atom bombardment–mass spectrometry, 16
Fatty acid soaps, 346
Fatty acids, 144, 279, 286
Fatty acid number, 363
Fatty alcohol ethoxylates, 346
Fatty foods, 244
Fast extrusion furnace (FEF), 161
N550, 161-162
Field
desorption–mass spectrometry, 16
latex, 337
natural rubber latex, 363
Fillers, 2, 10, 11, 14, 19, 32, 41-42, 53, 61, 79, 86, 90, 95, 98, 113, 116, 145,
154, 179, 181, 195, 260, 281, 318, 349
Finished rubber products, 196
Finite element analysis (FEA), 202-203
Fire retardants, 12, 32, 115
FKM-type, 280
Flame ionisation detector, 57, 259
Flame retardants, 38, 140-141, 146, 168, 174, 176, 178
additives, 103, 141
hydrated alumina, 169
rubber compound, 103
Flory equation, 204
Fluorescence detector, 256
Fluorinated
compounds, 199
Fluorination, 358
Fluoro/fluoroalkyl, 357

399
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Fluorocarbon, 63
Fluorocarbon - 49V, 164, 165
Fluorocarbon rubber, 53, 64, 79, 86, 102, 114-115, 161, 163, 190, 237, 280, 357,
369
Fluorochlorocarbon rubbers, 357
Flynn–Wall method, 78
Flynn–Wall–Ozawa, 193
Fogging tests, 226
Food
Additives & Contaminants, 269
contact materials, 238, 253, 255, 266
-contact rubbers, 238
simulants, 237, 239-240, 247, 255-257, 259-262, 265, 267, 269
Food-contact
applications, 1, 2, 235, 236, 244, 248, 266, 275
elastomers, 242, 245
materials, 146, 235-237, 239, 242, 245, 258, 266, 275
regulations, 18, 235, 239, 277
rubber, 137, 237, 241, 248, 254, 260, 265
   compounds, 257
silicone elastomers, 260, 266
substances, 235, 236
Food-grade
rubbers, 86
silicone rubbers, 266
Food Standards Agency (FSA), 135, 236-237, 239, 245, 248, 258, 260, 264, 266
Project A03038, 271
Project A03046, 271
Project FS2219, 270
Project FS2248, 270
Forest fire hoses, 79
Formaldehyde, 258
Formamide, 171
Fourier–Transform infrared (FTIR), 10, 12-13, 22, 38, 39, 53, 55, 60, 61, 71-73,
79-80, 82-83, 97-98, 100, 127, 140, 143-145, 153, 167-169, 175, 177, 181,
198-199, 201, 205, 218-219, 292, 301, 320-322, 324-325, 328-330, 347-348
instrument, 16, 84
microscope, 301, 323
microscopy, 97, 327-328
microspectrometry, 322
–near infrared, 13
nuclear magnetic resonance, 198

400
 Index

spectrometer, 104
spectroscopy, 71
Fractional creaming, 343-344
Free radical-polymerised rubbers, 67
Free sulfur, 362
Freeman–Carroll method, 77-78
Free-radical polymerisation, 65
Freeze–thaw cycle stability test, 363
Freidel–Crafts reactions, 287
Fresh and aged latex, 349
Friction reducing agents, 104
Friedman, 193
Full curve, 100
Furnace ashing, 97
Furnace black, 86, 89
Furnace combustion method, 362
G
Gamma radiation, 113
Gamma-propylmethacrylate trimethoxysilane, 348
Gas chromatography (GC), 16, 55, 198, 237, 341
analysis, 128
–Fourier–Transform infrared, 145
–mass spectrometry, 18, 21-23, 25, 26, 28, 30, 39, 57, 79, 86, 96, 116-117,
119-123, 127-138, 143, 146, 153, 155-157, 160-161, 163-165, 168, 171,
173-174, 212, 217-218, 222-223, 225-228, 230, 249-257, 259, 261-262,
264-266, 269, 290-300, 302, 323, 328, 330, 362
   analysis, 296
instrument, 22, 138, 249, 331
total ion current chromatogram, 171-172
pyrogram, 55, 56
–thermal energy analyser method, 256, 262
GC×GC–time-of-flight–MS, 23-24, 121, 123, 131-132, 156-158, 159, 161,
163-165, 239, 249-251, 267-268, 291, 301, 324-325
Gas sampling bag, 218
Gel permeation chromatography (GPC), 28, 66-67, 70, 367
General rubber goods (GRG), 215
Genetic optimisation algorithm, 204
German Association of the Automotive Industry
VDA 278, 226
German ZEK method, 85

401
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

German Commodities Regulation, 242

Germanium
crystal, 10, 182, 320
disc, 30, 69
Giese multi-faceted system, 219
Glass capillary kinematic viscometers, 361
Glass transition temperature (Tg), 3, 31, 58, 125, 194, 281, 348, 367
Glass fibre filter, 217-218, 220-221
Glass fibre pre-filters, 221
Graphite, 17, 86, 90, 371
furnace atomic absorption spectrometry, 17
Gravimetric method, 145
Grob Mix, 23-24
Gas transfer mould (GTM)/gas chromatography–mass spectrometry, 225
Guanadines, 346
accelerators, 260
Guayule, 349, 360
GuttaPercha, 360
Gypsum–silica gel, 119
H
H-nuclear magnetic resonance, 72-73
High abrasion furnace (HAF), 294
N330, 134, 157, 294
Halobutyl rubbers, 15, 155, 362
Halogen, 10, 27, 99, 140
Halogenated
butyl rubber, 63
compounds, 213
rubbers, 64, 362
Halogens, 9, 87, 97
Hazing, 319
Headspace gas chromatography–mass spectometry, 136, 139, 143, 228, 249, 290,
293, 296, 330
Headspace techniques, 130, 364
Health and Safety Executive (HSE), 212
Heat transfer equation, 202
Heptane, 117
Hevea allergens, 339
Hevea brasiliensis, 333
Hexafluoropropylene, 357
Hexamethylenediamine carbamate, 161

402
 Index

High ammonia latex, 333

Higher fatty acid (HFA), 337
number, 338
Highly sensitive spectrometer, 12
High styrene-butadiene rubber, 369
High-performance liquid chromatography (HPLC), 9, 25, 26, 46, 117-119,
123-124, 129, 137, 140, 237, 254, 345, 362, 367
instruments, 30, 237
High-polyaromatic hydrocarbon content extender oils, 356
High-resolution accurate mass (HRAM) mass spectrometer, 291
High-resolution
accurate mass liquid chromatography–mass spectrometry instruments, 26
gel permeation chromatography, 28-29
programmes, 99
pyrolysis gas chromatography, 71
–mass spectrometry, 349
thermogravimetric analysis, 38, 182
High-sensitivity analysis, 68
High-styrene resins, 141
Hindered
amines, 118
phenolic compounds, 281
phenols, 346
Hot solvent extraction, 129
Huggins constant, 350
Hydrated alumina, 18, 104, 140, 169, 176
Hydrin, 63
20D, 162, 164, 165, 200, 265
rubber, 163
Hydrocarbon
mineral oils, 256
oil, 141, 281, 324
rubbers, 362
wax, 134, 315
waxes, 315
Hydrogen
abstraction, 283
fluoride, 280
peroxide (H2O2), 8
sulfide, 295
Hydrogenated
-nitrile rubber, 115, 156-158, 160, 204, 359-360

403
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

349N, 157-158, 160

Hydrophobicity, 348
Hydrosilicone, 199
Hydroxyl-terminated organosiloxanes, 269
HyperDSCTM analyses, 33-34
I
Imaging chemiluminescence technique, 20
Impingers, 229
tubes, 225
Insitu Fourier–Transform infrared (FTIR) spectroscopy, 198
In vitro cytotoxicity method, 350
Increment method, 203
Indene, 142
-coumarone resin, 371
Inductively coupled plasma (ICP) analysis, 12, 16, 46, 154, 174, 176, 292, 301
–atomic emission spectroscopy, 18
-based techniques, 17
–mass spectrometry, 16, 18, 293, 302
–optical emission spectrometry, 17, 181, 293, 302, 367
Industrial
latex, 339
screens, 361
Inert filler, 182
Inflection points, 343
Infrared (IR), 293, 359
data, 10-12, 41, 69, 320, 329
detector, 13
emission spectroscopy, 12
libraries, 10
microscopy, 27, 28
microspectrometry, 14, 41, 123
objective lens, 13
radiation, 12, 320, 322
spectral libraries, 11
spectroscopy, 10, 12, 14-15, 19, 55, 58, 69, 72, 361
spectrum, 14
Inhibition enzyme-linked immunosorbent assay (ELISA), 339
Inorganic
additives, 8, 12
compounds, 97, 281
filler, 8, 12, 19, 51, 79, 96-98, 153, 155, 318

404
 Index

filler additives, 97
flame retardants, 140
pigments, 97, 282
Insoluble solid materials, 345
International Carbon Black Association (ICBA), 95
International Council for Harmonisation of Technical Requirements for
Pharmaceuticals for Human Use (ICH), 275
Guideline for Elemental Impurities Q3D, 306
Guideline for MetalImpurities Q3D, 302
International Organization for Standardization (ISO), 211
ISO 123, 362
ISO 124, 362
ISO 125, 363
ISO 126, 363
ISO 127, 363
ISO 247, 366
ISO 248, 366
ISO 249, 366
ISO 498, 363
ISO 506, 363
ISO 565, 361
ISO 705, 363
ISO 706, 363
ISO 976, 363
ISO 1124, 364
ISO 1125, 364
ISO 1126, 364
ISO 1138, 364
ISO 1147, 363
ISO 1304, 94, 364
ISO 1382, 361
ISO 1407, 80, 361, 366
ISO 1408, 361
ISO 1434, 366
ISO 1437, 364
ISO 1629, 357, 361
ISO 1652, 363
ISO 1656, 366
ISO 1657, 344, 363, 366
ISO 1658, 365
ISO 1795, 366
ISO 1802, 363

405
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

ISO 1867, 364

ISO 1868, 364
ISO 2000, 366
ISO 2004, 363
ISO 2005, 363
ISO 2008, 363
ISO 2194, 361
ISO 2302, 365
ISO 2322, 365
ISO 2438, 363
ISO 2453, 366
ISO 2454, 366
ISO 2475, 365
ISO 2476, 365
ISO 2602, 365
ISO 2781, 361
ISO 2854, 366
ISO 2859, 366
ISO 3104, 361
ISO 3105, 361
ISO 3136, 363
ISO 3207, 366
ISO 3257, 365
ISO 3301, 366
ISO 3494, 366
ISO 3534, 366
ISO 3865, 361
ISO 3899, 366
ISO 3900, 366
ISO 4097, 365
ISO 4645, 119, 361
ISO 4648, 361
ISO 4650, 361
ISO 4655, 367
ISO 4659, 365
ISO 4660, 367
ISO 5435, 365
ISO 5725, 366
ISO 5945, 361, 367
ISO 6101, 344, 363, 367
ISO 6209, 365
ISO 6235, 367

406
 Index

ISO 6810, 365

ISO 6894, 365
ISO 7269, 362
ISO 7270, 367
ISO 7725, 362
ISO 7780, 363, 367
ISO 7781, 367
ISO 8053, 363, 367
ISO 8054, 362
ISO 10398, 362
ISO 10638, 362
ISO 11089, 118, 123, 362, 367
ISO 11344, 367
ISO 11852, 363
ISO 12000, 363
ISO 12243, 364
ISO 12245, 365
ISO 12492, 367
ISO 13773, 364
ISO 15671, 362
ISO 15672, 362
ISO 19050, 367
ISO 22768, 367
ISO 13741-1, 364
ISO 13741-2, 364
ISO 3858-1, 364
ISO 3858-2, 364
ISO 4652-1, 364
ISO 4656-1, 365
ISO 4661-1, 361
ISO 4661-2, 361
ISO 6528-1, 8, 362
ISO 6528-2, 362
ISO 6528-3, 8, 362
ISO 9924-1, 99, 362
ISO 9924-2, 362
ISO 9924-3, 362
ISO 123:2001, 335
ISO 127:1984, 337
ISO 506:1992, 337
ISO 705:2015, 336
ISO 706:2004, 335

407
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

ISO 976:2013, 336

ISO 1409:2006, 342
ISO 15671:2000, 44
ISO 15672:2000, 44
ISO 1652:2011, 342
ISO 1654:1971, 344
ISO 1655:1975, 344
ISO 1656:1996, 344
ISO 1802:1985, 345
ISO 247:2006, 167
ISO 2555:1989, 342
ISO 3136:1983, 340
ISO 3219:1993, 342
ISO 3499:1976, 341
ISO 3899:2005, 341
ISO 3900:1995, 341
ISO 4655:1985, 340
ISO 7269:1995, 44
ISO 7780:1998, 344
ISO 8053:1995, 344
ISO 8871-1:2003, 302
ISO/DTR21275:2017, 230
ISO/TC45, 212
ISO/TS17796:2013, 212, 217, 223, 228-230
ISO/TS21522:2017, 230
Intrinsic viscosity, 69
Inverse gas chromatography (IGC), 96
Iodine adsorption number, 364
Iodometric method, 362
Ion
chromatography (IC), 8, 175, 293, 326
mass spectrometry, 16, 96, 147
-exchange chromatography, 27
mobility–mass spectrometry, 292
Ionic shock, 345
Ion-mobility spectrometry (IMS), 180
mass spectrometric analysis, 147
Irganox®
1010, 25, 119
1076, 119
1098, 119
1330, 119

408
 Index

Iron, 170
Iron oxide, 371
(186.1374), 243
Isobutene-isoprene rubber (IIR), 57, 359, 365
Isobutylene, 359
Isooctane, 259
Isoprene, 57, 99, 359
Isopropanol, 276, 340
Isopropyl p-phenylenediamine (IPPD), 117, 123, 287
Isopropylbenzene, 225
Isopropylidene aniline, 300
Isothermal
method, 190
thermogravimetric anaylsis, 127
Isothiocyanates, 213, 260
Isothiocyanato-ethane, 254
K
Kamal–Ryan, 202
Kamal–Sourour model, 194, 201-202
Kaolin clay (186.1256), 243
Karl Fischer method, 367
Keltan®, 720, 252
Ketones, 213, 218-219, 267, 280, 282, 287
Kinetic analytical methods, 193
Kinematic viscosity, 361
Kissinger model, 193, 201
Kissinger–Akahira–Sunose, 193
L
Laboratory-scale instruments, 27
Laminated rubber bearing, 203
Lampblacks, 86
Lanolin, 371
Laser desorption ionisation techniques, 147
Laser desorption/laser photo ionisation–time-of-flight–mass spectrometry, 16
Laser-ionisation mass analysis (LIMA), 43, 326, 328
Laser techniques, 16
Latex, 3, 43, 264, 334-352
allergens, 339
allergy test, 334
gloves, 264, 334

409
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

products, 334, 339

serum, 338
Leachable work, 276
Leachables, 276, 277, 287, 289, 294, 296-298, 301, 306-311
project, 305
LECO Pegasus III GC×GC–time-of-flight–mass spectrometry, 157, 250
Light scattering, 28, 70, 343
detectors, 70
Light transmission microscope, 90
Light-scattering methods, 344
Linear siloxanes, 267
Litharge, 371
Liquid
chromatography (LC), 16, 66, 290, 326, 350
–mass spectrometry, 18, 22, 25-26, 28, 82, 116, 119, 129, 137, 146-147, 153,
178, 237, 251-252, 254, 255-257, 260, 261-262, 265-269, 291-295, 298,
299-302, 304, 330
instruments, 261, 291
   libraries, 26, 298
–mass spectrometry/mass spectrometry method, 262
–mass spectrometry×mass spectrometry instrument, 291, 301
–mass spectrometry–ion-trap–time-of-flight–mass spectrometry, 145
paraffin mull technique, 97
-Transform®, 28, 30
Liquid-phase extraction, 143
Long-term thermal ageing, 127
Loss modulus, 59, 77
Low ammonia, 333
with 0.25% boric acid and 0.05% sodium pentachlorophenate (LABA), 345,
345
Low-density polyethylene (LDPE), 61-62, 77-78
Low-angle laser light scattering, 70
Low-molecular weight
amines, 130
compounds, 211
liquids, 329
oligomers, 257
polymers, 70, 146
Low-polyaromatic hydrocarbons content extender oils, 84
LRCCP Rubber and Plastics Research and Testing Laboratory, 223
Luperco 101-XL, 161, 249

410
 Index

M
Methylthiobenzothiazole, 228
Macrobaryta technique, 337
Magicangle, 15
Maglite® DE, 162
Magnesium, 170
carbonate, 371
hydroxide, 79-80 140
filler, 80
Magnesium oxide, 162, 371
Mark–Houwinkparameters, 66, 69-70
Mark–Houwink–Sakwada equation, 69
Mass spectrometry (MS), 15, 358
detection, 259
techniques, 16
-based analytical systems, 16
Mass spectrometer detectors, 22
Mathematical enhancement, 182
Matrix-assisted laser desorption/ionisation (MALDI), 16, 146-147
mass spectrometry, 16, 146
Maxwell equation, 202
Mechanical testing, 77
Medical
devices, 275
gloves, 334, 339, 364
Medium thermal N990, 161, 249
Mellitates, 256
Melting temperature (Tm), 34
Membrane osmometry, 70
Mercaptans (thiols), 279
Mercaptobenzothiazole 2,2'-dithiobis(benzothiazole), 134, 135, 174
Mercaptobenzothiazole zinc salt, 295
Mercaptoimidazoline, 244
Metal catalysts, 267
Metal oxide (lead oxide), 201
Methacrylic acid (MAA), 347
Methacryloxypropyltris(trimethylsiloxy)silane, 348
Methanol, 11, 80, 117-118, 123-124, 129, 133, 167-169, 170, 175-176, 177, 223,
298, 330, 340
Method for Determination of Hazardous Substance (MDHS), 212
47/2:1999, 217, 229, 230
Methyl

411
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

aniline, 256
ethers, 120
ethyl ketone, 80, 287
isothiocyanate, 295
methacrylate, 348
red, 336
Methyldecanoate, 24
Methyldodecanoate, 24
Methylphenyldithiocarbamic acid, 295
Methyl methacrylate, 358, 360
Methyl-terminated linear siloxane oligomers, 259
Mica, 371
Microbial esterification, 227
Microcystalline waxes, 326
Microemulsion photopolymerisation, 348
Microfurnace, 55, 56
-type pyrolyser, 56
Microthermal
analyser, 35, 36
analysis, 35, 37, 42
Microtoming, 90
Migratable substances, 258
Migration testing, 145, 240-241, 244, 246-247
Millable polyether-type polyurethane rubber, 54
Milling, 215
Mineral oil, 156, 227, 371
Mixed
dodecenes, 279
dodecyl mercaptans, 279
Mixing, 257
Modern rubber analysis laboratory, 7
Modified
blooms, 317
ignition tube, 53
Lowry method, 364
Modifiers, 67, 145
Modulated
differential scanning calorimetry, 33, 41
thermogravimetric analysis, 38, 39
Moleculart weight (MW) modifiers, 279
Molecular weight distribution (MWD), 65
Monoclonal antibodies, 339

412
 Index

Monodisperse, 66, 67
Monostyrenated diphenylamine, 296
Monovalent organic group, 282
Montmorillonite (MMT)-filled silicone rubber, 140
Mooney
scorch determinations, 42
stress relaxation, 42
viscometer, 42
viscosity, 42
–Rivlin equilibrium modulus, 205
–Rivlin method, 205
Morpholine, 215, 283
Moulding, 215, 257
temperature, 216, 226
-release agents, 163
Moving die rheometer (MDR), 42, 157, 163, 187, 195-196, 200
Multi-angle laser lights cattering (MALLS), 70
Mutagens, 220
Myristic acid, 143
N
N-benzylidenebenzylamine, 228
N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylene diamine (6PPD), 123, 134, 215,
227, 260, 287
N-(1,3-dimethylbutyl)-N'-phenyl-1,4-p-benzoquinonediimine, 287
N-(2,2-dimethylpropyl)-N-methyl benzenamine, 215
N-(2-hydroxyethyl)-dodecanamide, 254
N,N'-(4,4'-methylenediphenyl)dimaleimide, 197
N,N-dibutylformamide, 262
N,N-dicyclohexyl-2-benzothiazyl sulfenamide (DCBS), 200
N,N-diethylacetamide, 80
N,N-diethylformamide, 80
N,N-diisopropyl-2-benzothiazole sulfenamide, 285
N,N-dimethylaniline, 283, 295
N,N-dimethylformamide, 80, 171-172, 174
N,N-dimethylpalmitamide, 254
N,N-dimethylurea, 171
N,N'-di-ortho-tolyl guanidine (DOTG), 161-162, 256
N,N’-diphenylthiourea, 295
N,N'-diphenylurea, 295
N,N'-phenylenedimaleimide, 197
N-alkyl-N'-phenyl-p-phenylenediamines, 123

413
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Nanoclay material, 79
Naphthenic oil, 134
Naphthenic-type hydrocarbon oil, 324
Napierian logarithms, 246
N-aryl-N'-aryl-p-phenylenediamines, 123
Natural latices 333
Natural fibres, 329
Natural rubber (NR), 3, 11-13, 28, 32, 38, 53, 55-57, 63, 67-68, 73, 77, 79, 81,
84-85, 92, 102, 103, 114-115, 125, 134, 137, 141, 180-182, 188, 194, 198-202,
204, 212, 225, 227-228, 237, 239, 261-264, 266, 279, 304, 326, 333, 335-339,
341, 344-345, 349-350, 359, 363-366, 369
504A aged, 265
condoms, 350-351
compound, 85, 134
latex, 263, 333-340, 344-345, 349-350, 363-364
   gloves, 339
   proteins, 339
mouldings, 205
/polybutadiene, 182
/styrene-butadiene rubber, 182
N-butanol, 80
N-butyl acrylate, 348
N-butyl-1-butanamide, 215
N-cyclohexyl benzothiazole, 91
N-cyclohexyl formamide, 215
N-cyclohexyl-2-benzothiazole sulfenamide (CBS), 134-136, 174, 178, 196, 198,
200, 256, 258, 260, 285
accelerator, 134-135, 174
Near-infrared (NIR) region, 13, 145
Negative-ionionisation, 26
N-ethyl-2-benzothiazolamine, 215, 254
Network-bound antioxidants, 346
Neutron scattering, 96
N-formylpiperidine, 254
n-Heptane, 258
n-Hexane, 80
Ni-Count-1 instrument, 94
Nitration method, 341
Nitrile
latex, 341, 366
gloves, 339
gloves, 264

414
 Index

Nitrile/polyvinyl chloride blends, 369

Nitrile rubber (NBR), 10, 13, 28-29, 57-58, 60, 63-68, 78-81, 87-88, 99-100,
103-104, 114-115, 118, 127-128, 136-137, 157, 159, 163, 168-170, 176-178,
195-196, 198, 215, 221, 225, 228, 237, 259, 264, 266, 279, 280, 287, 294-300,
304, 341, 350, 359, 369
gloves, 264
latices, 341
rubber, 127, 243
vulcanisation, 280
–349N, 164, 165
/ethylene-vinyl acetate blends, 78
/polychloroprene blends, 221
/polyvinyl chloride blends, 127
Nitrile–isoprene latices, 341
Nitrobenzene, 92
Nitrogen adsorption methods, 364
Nitrogen dioxide (NO2), 288
Nitrogen phosphorus detector, 21, 341
Nitrosamines, 128, 150, 212-213, 239, 245-246, 256-257, 261-264, 271-272,
286-287, 289-290, 356
Nitrosatable substances, 241, 257, 262
Nitrosating agents (NOX), 286, 288
Nitrosation, 220, 288
Nitroso compounds, 138
Nitrosyl radicals, 256
N-methylaniline, 300
N-methyldiphenylamine, 295
N-methyl-p-phenylenediamine, 128
N-nitrosamines, 220-221, 241, 261-263
N-nitrosatable
compounds, 263
substances, 261, 263
N-nitroso-N-methylamine, 295
Nocrac AW, 162
Nonanol, 24
Non-linear regression method, 194
Non-pyrolytic Fourier–Transform infrared method, 179
Non-volatile acids (NVA), 337-338
Non-volatile residue (NVR), 292, 301
Nordic Curing Fumes Project, 220
N-oxydiethylene-2-benzothiazyl sulfenamide, 285
N-phenylbenzeneamine, 172, 174

415
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

N-phenyl-1-naphthylamine (PAN), 294-295, 299

N-phenyl-N'-(1,3-dimethylbutyl)-p-phenylenediamine, 260
N-phenyl-N-methyl-N'-phenylthiourea, 295
N-phenylpalmitamide, 299
N-phenylstearamide, 299
N-phenyl-β-naphthylamine, 118
NR2 compound, 134-136
N-tert-butyl-2-benzothiazole sulfonamide/tetrabenzylthiuram disulfide, 226
N-tert-butyl-2-benzothiazole sulfenamide (TBBS), 196, 200-201, 285
/N-cyclohexyl-2-benzothiazole sulfenamide (CBS), 196
N-tert-butyl-2-benzothiazole sulfenimide (TBSI), 200
Nuclear magnetic resonance (NMR) spectroscopy, 15-16, 45, 58, 72, 85, 143,
153, 175-177, 194, 198, 259, 264, 351
instruments, 15, 72
Numerical algorithmic method, 180
O
Octanal, 24
o-Dichlorobenzene, 80, 229
Oil pan seal, 179
Okerin 1944, 134
Oleic acid, 371
Oligomeric
antidegradants, 257
antioxidants, 155
phenolic antioxidants, 258
Oligomers, 21, 27-28, 57, 72, 226, 257, 261, 269, 279-280, 295, 302, 329
Olive oil, 259
Olympus, 13
Optical emission spectrometry (OES), 293, 302
Optical microscopy, 328
with image capture, 328
Optical microscopes, 13
Optigrade Dispergrader, 91
Organic
compounds, 8, 123,
flame retardants, 8, 21, 123, 139-140, 178, 218, 223, 225, 269-270, 292, 293
material, 361
pigments, 282
polymers, 248
solvents, 116
Organic-modified kaolin, 79

416
 Index

Organoclay, 141
Organophosphates, 140
Organo-sulfur compounds, 284
Orthogonal online separation method, 180
Oscillating disc rheometer (ODR), 187-188, 199, 201
Oscillating differential scanning calorimetry, 32
Osmometrytechniques, 70
Overallmigration limits (OML), 240-242, 248
Oxidation temperature, 88, 89, 100
Oxygen
flask combustion method, 8, 175, 362
flask method, 8
index value, 141
Ozawa, 193
Ozonolysis method, 367
P
Palmitic acid, 143
Paper chromatography, 30
Paraffin
mull technique, 97
oil, 144, 148, 371
wax, 25, 317, 346, 371
Particle size:electron microscopy, 343
Payne effect, 85, 200
Penetrometry, 196, 197
Pentabromodiphenylether, 104, 168, 176
Perkadox®, 14/40, 133, 157
Perkin Elmer ATD 400, 130
Perkin Elmer 2400 CHN analyser, 175
Peroxide (dicumyl peroxide), 201
Peroxide compound, 282
Peroxide curatives, 315
Peroxide co-agents, 143
Peroxide-cured
ethylene propylene diene monomer, 223
fluorocarbon rubber, 191
rubber, 10
Peroxides, 259
Peroximon F 40 peroxide curative, 223
Petroleum products, 361
Pew Charitable Trusts database, 236

417
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Pharmaceutical
applications, 275
packaging, 146, 276, 289, 290, 294, 304, 306
products, 165, 275, 279, 303
stoppers, 304
Phenol, 171, 172
Phenolic
antioxidants, 262
resins, 78, 142
Phenolic-type antioxidant, 318-319
Phenols, 222
phenyl
benzimidazole, 215
isocyanate, 295
isothiocyanate, 295
mercaptan, 295
Phenyl-beta-naphthylamine, 287
Phenylenediamine-based antioxidants, 260
Phenylmethyl siloxane, 358
Phosphate
antioxidant, 262
ester, 104, 168
flame retardant, 10
Phosphates, 281
Phosphites, 286
Phosphorous, 21, 140, 170, 175-176, 341
Photoionisation, 16, 147, 301
Photometric method, 344, 363, 367
Photo oxidation, 146, 193
Phthalate ester, 104
Phthalates, 82, 256
Phthalimide, 172, 174, 215
Physical
ageing, 33
testing, 58
Physico-chemical analytical techniques, 77
Piloyan, 202
Pine tar, 143, 371
oil, 143
Pinene, 143
Plasticisation, 196

418
 Index

Plasticiser, 2, 10-12, 14, 21, 25, 29, 30-32, 37-38, 60, 64, 69, 80-82, 84, 87, 95,
105, 113, 116, 140-142, 145-146, 155, 169, 181, 191-192, 214, 219, 229, 236,
256, 263-264, 281, 329-330
Plasticiser/oil, 153
Plastic–rubber blend, 193
Platinum
gauge ‘flags’, 8
ring, 342
p-nonylphenol (NP), 258
Polar, neutral compounds, 222
Polarography, 9
Poly(2,6-diphenylphenylene oxide)-type sorbent, 217
Poly-2,2,4-trimethyl-1,2-dihydroquinoline (TMQ), 25, 118, 144, 148
Polyacrylate (PA) latices, 347
Polyaromatic
hydrocarbon compounds, 256
hydrocarbons (PAH), 1, 5, 84-86, 95, 212, 220
compound, 85-86, 220-221
Polybutadiene (BR), 32, 34-35, 53, 57, 61, 63, 68, 72-73, 77, 92, 102, 180-181,
193-195, 198, 358
rubber, 32, 369
Polybutadiene, 215
Polybutylene terephthalate (PBT), 33
Polybutyl ether disulfide, 359
Polycarbonate (PC), 33
/polybutylene terephthalate blend, 33
Polychlorobutadiene, 358
Polychloroprene (CR), 57, 182, 199, 237, 279, 350
rubber, 57, 370
latex, 364
Polydimethylsiloxane (PDMS), 63, 197, 259, 358
Polydimethylsiloxane analysis, 69
Polydispersity (PDI), 66
Polyepichlorohydrin, 358
Polyester fibres, 329
Polyester-type polyurethanes, 360
Polyethylene, 25, 61, 64, 114, 201-202, 259, 304
disulfide, 359
glycol, 25, 202
wax, 123
Polyethylene terephthalate bottles, 259
Polyfluorophophazene rubbers, 360

419
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Polyisobutylene, 63
Polyisoprene (PI), 57, 81, 195, 361, 367
rubber, 57, 94, 115, 125, 370
Polymer
additives, 16, 145-147
blends, 22, 30, 33, 41, 69, 91, 153, 180, 192-193
industry, 5
degradation, 146, 183, 292
morphology, 7
rubbers, 4
stabilisers, 147
structure, 15, 73
Polymer-based products, 11
Polymeric plasticisers, 82
Polymerisation, 142, 199, 278-280, 299, 333, 351-352
modifiers, 279
residues, 278
vessel, 279
Polynorbornene, 360, 370
Polypropylene (PP), 32, 74
Polysiloxanes, 248
rubbers, 362
Polystyrene, 61, 139
Polysulfide rubbers, 81, 370
Polyurethane, 53, 114, 304
elastomer, 75-76
rubber, 53, 75, 81, 114, 115
Polyvinyl
acetate latices, 341
chloride (PVC), 10, 64, 79-81, 103-104, 168-169, 170, 175-176, 178, 193,
243, 264, 304
/nitrile rubber blend, 79, 176, 243
Positive-ion ionisation, 26
Positive list, 155, 235
Potassium, 170
hydroxide (KOH), 9, 336-338, 343, 346
   number, 337, 338, 363
laurate, 346
oleate, 343
persulfate, 348
Potentiometric titration, 337, 343
Powder-free gloves, 350

420
 Index

p-Phenylenediamines, 123, 125, 317, 318, 346

antidegradants, 318
compounds, 123
Practicing analyst, 5
Precipitating solvent, 117
Pre-extracted rubbers, 11
Preparation of Food Contact Notifications and Food Additive Petitions for Food
Contact Substances: Chemistry Recommendations, 243
Preservatives, 345
Pre-vulcanisation inhibitors, 104
Principal decomposition products, 283
Process
aids, 2, 104, 115, 130, 329
oils, 82, 155, 236
Product Quality Research Institute (PQRI), 275, 288
Propane-2-thiol, 351
Propylene, 4, 13, 63, 74, 99, 114, 193, 357
Propylene oxide, 358
Protection agents, 113
Protein test, 334
Proteins, 346
Proton nuclear magnetic resonance analysis, 267, 269-270
Pseudo-bloom, 318-319
p-Tert-butylphenol (PTBP), 258
p-Tert-octylphenol (PTOP), 258
PTV injection, 157, 250
Pyrene, 254
Pyrolytic gas chromatographic method, 367
Pyrolysate approach, 12
Pyrolysis, 5, 11-12, 21-22, 46, 53, 55-58, 71, 82, 87, 92-93, 101, 103, 109, 123,
142, 170, 177, 198
–gas chromatography, 58
   –mass spectrometry analysis, 5, 127-128, 137, 141-142, 147
–mass spectrometry, 146
infrared, 123
   spectrum, 167
temperature, 21
Q
Q Exactive gas chromatography–mass spectrometry, 292
Quadrupole mass spectrometer, 23
Quality control (QC), 187

421
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Quartz wool filters, 225

R
Rabbitanti sera, 339
Radial passenger tyre tread, 179
Radiator hose, 179
Radical coupling, 283
Raman spectroscopy, 14, 72, 96, 181
Rapid method, 364
Rapra
gas transfer mould (GTM)/gas chromarography–mass spectrometry, 216, 225
Review Report, 1, 355
Vibrating Needle Curemeter (VNC), 187
Vulcanisation Project, 224
Rate of degradation, 12
Raw
polymers, 15
rubbery polymers, 97
styrene-butadiene, 366-367
synthetic, 362, 366-367
Red phosphorous, 140
Refei, 202
Reflectance attenuated total reflectance–Fourier–Transform infrared, 10, 180
Refractive index (RI), 28, 366
detector, 66
Registration, Evaluation, Authorisation and Restriction of CHemicals (REACH)
regulation, 85
Reheat thermogram, 31
Reinforced styrene-butadiene latex, 367
Relative standard deviation (RSD), 144
Research and development (R&D), 187
Residual
monomers, 146, 214, 229, 280, 364
solvents, 146
Resins, 44, 115, 141
Restek RTX-5 amine, 160
Retention time (RT), 20, 158
Reverse engineering, 4, 18, 23, 51-52, 99-100, 116, 129, 134, 138, 144, 154-155,
156, 163, 164, 165, 178, 179, 181, 188, 334, 347, 355
Reverse engineering work, 4, 51, 155, 156, 165, 179, 181, 188
Reverse-phase high-performance liquid chromatography, 9
Rhenofit PAN (antidegradant), 294

422
 Index

Rhenogran® MPTD 70 (curative), 162, 294

Rheology, 205
Rib Smoked Sheet (RSS), 359
Rosin, 371
soaps, 343, 346
Rotational viscometer, 342
Routine extractables testing, 289
Rubber
additives, 258
adhesives, 263
analysis, 1, 2, 5, 7, 10, 15, 65, 128, 314, 355, 356
analyst, 4, 7, 115, 355, 356
blends, 13, 41, 53, 100, 181, 182, 193, 203
blooms, 320
compound, 2-4, 11, 15, 32, 41, 51-52, 78, 80, 84, 86-92, 95-96, 101, 103-
104, 113, 115-117, 124-126, 128, 138-139, 141, 143-144, 153-156, 162-
163, 165-166, 178-179, 181, 187-189, 191-192, 194, 195, 198-199, 203,
211-212, 214-215, 217, 221-223, 225-229, 237, 241, 243, 249, 255-257,
260, 265, 282, 284, 286, 289-290, 293-294, 302-304, 317, 320, 326, 356
compounder, 4
compounding, 364-365
curing process, 202
formulation, 4, 73, 123, 227, 229, 293
fume, 125
fumes, 211-219, 221, 223-225, 228-230
industry, 156, 193, 196, 212, 219, 221, 230, 261, 287, 364
latex, 194, 334-335, 339, 342, 344-347, 362-364, 366
   dipped products, 345
latices, 347, 363-364, 367
manufacturing environment, 42
matrix, 15, 21, 41, 80, 93, 116, 123, 192, 314, 317-318, 320
piston washer, 202
process analyser (RPA), 42, 187
products, 1-4, 25, 36, 43, 52, 79, 84, 86, 95, 116, 124, 125, 128, 156, 187,
199, 212, 219, 227, 235-240, 243, 244-248, 252, 258, 261, 275, 277-278,
281, 289, 292-293, 301, 303, 318, 334, 339, 345, 355-356
seals, 2
teats, 262
technology, 2, 5, 115, 138, 153, 248, 356
technology, analysts, 5
-related regulations, 1
–rubber blends, 33

423
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Rubbery
latex particles, 343
polymers, 65
S
SantopreneTM, 74
Scalpel blade, 327
Scanning electron microscopy (SEM), 19, 41-42, 77, 93, 127-128, 140-141, 328,
349
Scanning tunneling microscopy, 93
Sealed pans, 330
Sealing applications, 289
Sebacates, 256
Secondary-ion mass spectrometry (SIMS), 16, 43, 147, 326, 328
cluster, 147
gentle, 147
Seed emulsion polymerisation, 348
Seeded semi-continuous emulsion polymerisation method, 347
Seeding effect, 315
Self-reinforcing CB N762, 252
Semi-crystalline plastics, 32
SGETM Pyrojector, 56
BPX5, 168
BPX-50, 158
Shear loss modulus (G''), 196
Shear storage modulus (G'), 196
Sieve residue, 363-364
Silane coupling agents, 67
Silane ST69, 223
Silica, 98, 218, 281, 302, 371
gel, 119, 218
gel chromatography, 264
microfillers, 349
-filled natural rubber vulcanisates, 201
Silicates, 98
Silicon, 170, 322
Silicone, 304
acrylate copolymer, 348
copolymer (polydimethylsiloxane-co-diethylsiloxane), 199
liquids, 247
materials, 247, 269
oils, 241

424
 Index

oligomers, 260-261, 267

pastes, 247
polymers, 248
resins, 241, 247
rubber, 15, 38, 69, 77-78, 81, 86, 114-115, 140, 193, 198-199, 201, 203,
222, 237, 240-242, 247-248, 259, 262, 266-267, 269
food-contact products, 269
-modified styrene-butyl acrylate copolymer, 348
-release agents, 318
Siloxane
migration, 269
oligomers, 260, 266, 267
Siloxanes, 163, 222, 259, 262, 267, 269-270
Single bounce attenuated total reflectance, 10, 53, 97
Single-ring compounds, 283
Single-use
components, 306
inhalers, 275
systems (SUS), 305
Size-exclusion chromatography, 28
Skim crumb rubber, 227
SKN-26 nitrile rubber, 227
Sludge content, 363
Small-angle neutron scattering (SANS), 93-94
SMR CV 50, 228
Soap
adsorption, 343
content, 338, 342-343
deficiency, 343
titration, 338
Sodium, 170
3-allyloxy-, 348
alginate, 344
bicarbonate, 138
chloride, 340
   plates, 10
dodecyl sulfate, 349
pentachlorophenate, 345
   latex, 333
periodate photometric method, 344, 363, 367
peroxide fusion method, 362
Softening point, 78

425
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Solid rubber, 249, 333-345

products 334, 347
Solid-phase microextraction technique, 146, 258, 262
Solids tests, 334
Solid-state magic-angle spinning, 58
Solution injection, 130, 290
gas chromatography–mass spectrometry, 296
Solvent desorption gas chromatography, 218
Solvent extraction, 8, 11, 80, 82, 84, 116, 123, 130, 131, 145, 146, 148, 155,
256, 356
–Fourier–Transform infrared analysis, 218, 330
Solvent extractable material, 365
Soothers, 262
Soxhlet extraction, 95, 129
Specific migration, 241
limit (SML), 242
Spraying agents, 30
Stabilisers, 113, 116, 146, 157, 333, 345-346
Standard Malaysian Rubber (SMR), 359
Standard thermogravimetric analysis programme, 101
Standardised ashing techniques, 97
Standardised Extractables Testing Protocol, Pharmaceutical Engineering, 305
Starch, 119
Starch–silica gel, 119
Starting materials, 258
Static
headspace, 21
   gas chromatography, 341
Stearate compounds, 282
Stearic acid, 91, 134, 143-144, 148, 161, 162, 199, 200, 202, 217, 223, 252, 286,
294-295, 371
Sterilisation, 269
Stoichiometry, 299
Storage modulus, 194, 196
Strukpar 2280 paraffinic oil, 252
Struktol® WB16, 252
Styrenated
diphenylamines, 294, 296, 299
phenols, 346
Styrene, 13, 53, 57-58, 61, 73, 79, 81, 91, 114, 141-142, 215, 216, 280, 296, 333,
340-341, 348, 358

426
 Index

-butadiene rubber (SBR) rubber, 13, 32, 35, 53, 55, 57-58, 60-61, 63, 66, 68,
73, 79, 81, 91-94, 99, 102-103, 114-115, 118, 141-142, 144, 155, 180,
181, 194, 199-201, 204, 215-216, 225, 227, 260-261, 266, 279-280, 304,
336, 359, 365, 370
195S aged, 265
blends, 34
block copolymers, 69
latices, 340, 341
rubber, 142, 228, 266
/polybutadiene blends, 72, 73
-type latices, 340
-butadiene based polymers, 61
-butadiene block copolymers, 261
-butadiene-styrene, 280
   thermoplastic rubbers, 61
-butyl acrylate copolymer, 348
Sulfenamide, 285
accelerator, 199-200, 284
Sulfite method, 9
Sulfonamide, 82
Sulfonyl hydrazide compounds, 138
Sulfrasan R, 133
Sulfur, 133-134, 162, 170, 178, 199-200, 214, 252, 284, 322, 346, 357, 371
(MC grade), 294
blooming, 326
compounds, 220, 225
cure system, 136, 194, 196, 199, 317
curing systems, 224
donor-type cure systems, 283
vulcanisation, 193, 260, 356
   systems, 356
-based cure, 204
   system, 143, 261, 357
-containing additives, 145, 148
-containing compounds, 227
-cured compounds, 77, 143
-cured ethylene propylene diene monomer rubber, 216, 226
-cured rubber, 10
-cured solution, 73
-cured synthetic rubbers, 262
-donor compound tetramethylthiuram disulfide, 204
Sulfuric acid, 338, 340, 343

427
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Sulfenamide, 200
Super abrasion furnace, 88, 162
Supercritical fluid chromatography (SFC), 27, 257, 259, 326
Surface
analysis techniques, 123
analytical techniques, 43
blooms, 313
free energy, 342
tension, 342, 343
-assisted laser desorption ionisation, 147
Surfactants, 279, 345
Sym-dimethyldiphenylthiourea, 295
Synthetic
elastomers, 68, 212
fibres, 329
latices, 334, 338, 340, 343
plasticiser, 82, 141, 256
polyisoprene, 63, 261
rubber, 2, 3, 215, 239, 335
rubber latices, 333-334, 336, 344, 363
Synthetic polyisoprene, 63
T
Talc (magnesium silicate), 302, 371
Tan δ, 59-62, 77, 194-196
plot, 59, 194, 196
Tandem mass spectrometry (MS2), 16, 180
Tear testing, 77
TeflonTM filters, 217, 221
Tellurium, 253
diethyldithiocarbamate (TDEC), 252
accelerator, 252
dithiocarbamates, 252
Temperature sensor, 36
Tenax/XAD-2/Porapak adsorbent sampling tubes, 218
TenaxTM, 130, 168, 172-173, 217, 225, 269
absorption tubes, 221
-filled trap, 130
Tensile
strength, 58, 86, 194, 270, 281, 329
testing, 77
tests, 13, 77, 350

428
 Index

Tensometer, 187
Terephthalic acid, 259
Terpene, 78
Terpolymers, 71, 72
Tert-butanol, 283
Tert-butyl cumyl peroxide, 133
Tertiary amines, 283
Test gas mixtures, 219
Testing kits, 339
Tetrabutyl urea, 215
Tetrabutylthiuram disulfide (TBTD), 285
Tetraethylthiuram disulfide (TETD), 145, 280, 285
Tetrahydrofuran (THF), 28, 66, 67, 69, 80, 226
Tetramethylthiourea, 133, 228, 254, 283
Tetramethylthiuram disulfide (TMTD), 128, 129, 133, 174, 196, 198, 216, 228,
252, 280, 283-285, 304
-based EV system, 196
Tetramethylthiuram monosulfide (TMTM), 136, 144, 145, 148, 280
Theoretical model of Cramer, 263
Therapeutic products, 278
Thermal
analysis technique: TGA, 77-78, 116, 126, 187, 197
analysis techniques, 62, 125
black, 86
degradation, 78-79, 188, 280, 349
desorption, 218
   gas chromatography–mass spectrometry, 218
electron analyser, 128
mechanical analysis (TMA), 36, 37, 63, 196-197
oxidation, 20, 127, 146
stability, 20, 38, 77, 79, 101, 103, 117, 126-128, 140, 257, 348
technique, 7, 62, 77, 116, 140, 187, 326
Thermally-assisted hydrolysis and methylation (THM) gas chromatography–mass
spectrometry, 144
Thermodeposition gas chromatography–mass spectrometry, 226
Thermogravimetric analysis (TGA), 14, 37-39, 52, 64-65, 73, 77-79, 82, 84, 87-
90, 95-105, 116, 126-127, 140, 141, 153-154, 167, 169-170, 179-183, 185,
197, 328, 330, 347-349, 362
instruments, 38, 90, 104
–Fourier–Transform infrared, 147
–gas chromatography, 39
–infrared, 39, 181, 328, 330

429
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

–mass spectrometry, 39, 79, 146, 328, 330

Thermo-oxidation, 127, 146, 280
Thermoplastic
elastomer (TPE), 74-75, 237, 264, 290, 309
polymers, 141
rubber, 3, 28, 32, 36, 61, 69, 139, 264, 360
Thiazole, 213, 284, 346
Thin-layer chromatography (TLC), 30, 119, 123, 345, 361
Thioesters, 286
Thioureas, 346
Thiuram, 193, 285, 346
disulfide, 129, 279, 294, 304
accelerators, 129
-type accelerators, 304
Thorough drying, 340
Three-dimensional chromatography technique, 22
Three-dimensional network, 113
Time-of-flight (ToF), 16, 120
mass spectrometer, 23
Tinting strength, 365
Titanium, 170, 322
Titanium dioxide (TiO2), 10, 117, 282, 322
pigment, 10
Titanium dioxide (anatase), 371
Titanium dioxide (rutile), 371
Titration, 8, 9, 27, 175, 337, 338, 343, 363
Titrimetric method, 364, 366
Toluene, 28, 58, 66, 80, 93, 96, 172
Torn surface method, 91
Torque, 95, 188, 195, 201
curves, 200
Torsion pendulum, 59
Total bound styrene, 340-341
Total sulfur, 8, 9
Total ion current (TIC), 158, 297
Toluene extract, 364
Toxicity, 262
Traditional
rubbers, 290, 304
thermoset rubbers, 3
wet chemistry, 7
Trans-1,4-polyisoprene, 360

430
 Index

Trans-1,4-vinyl, 358
Translink, 294
Transmission electron microscopy (TEM), 41, 93, 348
Transmission Fourier–Transform infrared spectroscopy, 12, 82
Transmission infrared (IR) spectrum, 167
Triallyl cyanurate (TAC), 128, 143, 161, 249, 265
Triallyl isocyanurate, 143
Trichloroethane, 80
Trichlorofluoromethane, 222
Trichloromethane, 171
Trifluoroacetates, 120
Trifluoroacetic acid, 304
Trimethyl
silane-based antioxidants, 120
silanol, 262
silylethers, 120
Triphenyl phosphate, 18
Trisnonylphenyl
phosphate, 287
phosphite, 287
Trisphenyl amino-1,3,5-triazine, 295
Tube furnace, 8, 87, 97, 105, 168
combustion method, 8
method, 8
Tweezers, 327
Twin-screw extruder, 193, 196
Tyre industry, 78
U
Ultra-centrifugation, 71
Ultra-performance liquid chromatography (UPLC), 26, 291
–high-resolution accurate mass/mass spectrometry and ultra-performance
liquid chromatography–high-resolution accurate mass/mass spectrometry
instruments, 292
Ultrasonic
bath, 130, 296
emulsification, 348
Ultraviolet absorption, 341
Ultraviolet detection, 9
Ultraviolet detector, 123
Ultraviolet stabiliser, 30, 118, 144

431
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

Ultraviolet visible (UV-Vis), 15

detector, 27
spectrometers, 15
spectroscopy, 15
Uncompounded
ethylene propylene diene monomer, 145
polymer, 12, 65
Undecane, 24
United States Pahrmacopeia
USP 1663, 306
USP 1664, 306
USP 38–NF33, 306
Universal gas constant, 197
Unmodified rubber, 32, 67
Un-reacted accelerators, 129
Unvulcanised ethylene propylene diene monomer, 145, 147
Unvulcanised rubber, 71, 81, 117, 228
Urea, 172-173
US Environmental Protection Agency, 85
US Food and Drug Administration (FDA), 1, 86, 275, 352
Part 177.2600, 244
Part 189.220, 244
Part 189.250, 244
V
Vamac G, 161
Vander Waals forces, 329
Vapour generation atomic absorption spectrometry, 17
Vapour pressure osmometry, 70
Variable temperature analysis (VTA), 199
Vinyl
cylohexene, 299
triethoxysilane, 348
Vinylidene fluoride, 357
copolymers, 280
Vinylmethylsiloxane, 358
Viscometer, 69, 342
Viscosity, 27-28, 42-43, 66-67, 69, 86, 141, 145, 188-189, 342, 345
tests, 334
Vistalon ethylene propylene diene monomer rubber, 223
VistanexTM LM-MS, 162
VitonTM GBL200, 161, 249

432
 Index

Volatile fatty acid (VFA), 337

Volatilisation, 12, 30-31, 64, 82, 125, 183, 327
Volatility, 8, 20, 23, 24, 117, 119, 145, 156-157, 218-219, 221, 257, 329
Vulcanisation, 4, 11, 39, 65, 73, 104, 129, 145, 155, 188, 190, 194, 196, 200-
201, 203-205, 211-212, 214, 216-219, 221-223, 225, 226-229, 248, 260, 278,
282, 286, 300, 317, 334
accelerators, 196
agents’, 113
Vulcanised
ethylene propylene diene monomer, 145, 147
rubber products, 278
rubber compounds, 193
rubbers, 3, 28, 72, 81, 116, 137, 143, 179, 264
sulfur-cured ethylene propylene diene monomer 2 compound, 252
W
Water absorption measurements, 348
Water-soluble
dithiocarbamate, 345
materials, 345
Weight loss, 12, 37-38, 64, 82, 84, 89-90, 96, 99, 100-105, 140, 145, 167, 169,
182-183, 330
Wet chemistry, 4, 7, 181, 341
Wingstay®, 29, 294
X
Xanthates, 346
X-ray diffraction (XRD), 19, 93, 97, 140, 154
X-ray energy dispersive spectroscopy, 19
X-ray fluorescence spectroscopy (XRF), 12, 19, 96, 154, 167, 169-170
instrument, 19
X-ray photo electron spectroscopy, 349
Xylene, 80
Y
Yellow bloom, 321, 322, 323, 324
Young’s modulus, 197
Z
Zeeman atomic absorption spectrometry, 17
Zetpol® 2000L, 157
Zinc, 133, 141, 157, 170, 174, 176, 200, 241, 317-318, 322, 344

433
Rubber Analysis: Characterisation, Failure Diagnosis and Reverse Engineering

2-mercaptotoluimidazole, 346
ammine complex, 347
borate, 18, 176
chloride, 8, 97
dibenzyldithiocarbamate, 241, 261, 263
diethyldithiocarbamate (ZDEC), 228, 258, 317, 320-321, 333, 344-345
dimethyldithiocarbamate (ZDMC), 133, 284
hydroxystannate, 141
oxide (ZnO), 8, 11, 53, 91, 96-97, 104, 134, 143, 157, 161-162, 169, 176,
178, 193, 199, 243, 249, 252, 294, 302, 318-319, 322, 340, 346-348, 371
oxide viscosity test, 340
stability time (ZST) test, 340
stearate, 243, 282, 317-318, 371
sulfide, 8

434
 Published by Smithers Rapra, 2018

Rubber analysis plays a vital part in ensuring that the many products manufactured from the material are fit
for purpose, as well as contributing to many other important investigative activities within the rubber industry
and in other industrial sectors.

This comprehensive, application-based review, with extensive up-to-date referencing, covers all the important
applications and subject areas associated with the analysis of rubber compounds and rubber products.

The characterisation of rubber polymers is described, as is the identification and quantification of specific
additives within rubber samples. The approaches and methodologies that can be used to carry out partial or
complete reverse engineering work on fully compounded products are also included.

The many ways that rubber analysis can contribute to failure diagnosis studies are featured throughout the
book and its use in the identification of contaminants in final products and the characterisation of surface
blooms are highlighted.

The roles that rubber analysis has in quality control work, curing studies, and the characterisation of latex
and latex products are described. Also included are its application to the characterisation of rubber fumes,
identification of extractables and leachables from rubber products used in medical devices and pharmaceutical
packaging, as well as the identification of potential and actual migrants from food contact rubbers.

The author has drawn on 40 years of experience within the rubber industry to ensure that this book is
essential reading for anyone requiring up-to-date information on how a wide range of analytical techniques
can be applied to obtain industrially relevant information on many different rubber products at all stages in
their manufacture and life cycle.

Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

Telephone: +44 (0)1939 250383
Fax: +44 (0)1939 251118
Web: www.polymer-books.com