Particulate-Filled Polymer Composites,

Particulate-Filled Polymer Composites Second Edition

Editor

R.N. Rothon

Rapra Technology Limited

Particulate-Filled Polymer Composites 2nd Edition

Editor: Roger N. Rothon

rapra TECHNOLOGY

Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.rapra.net

First Published in 2003 by

Rapra Technology Limited Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

©2003, Rapra Technology Limited

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 authors and publishers apologize if any have been overlooked. Cover micrograph of aramid fibre-reinforced polyamide 6,6, reproduced with permission from P.R. Hornsby, Brunel University, UK.

ISBN: 1-85957-382-7

Typeset by Rapra Technology Limited Cover printed by The Printing House, Crewe, Cheshire, UK Printed and bound by Rapra Technology Limited, Shrewsbury, UK

Contents

Preface ................................................................................................................... 1 Contributors .......................................................................................................... 3 1

General Principles Guiding Selection and Use of Particulate Materials ........... 5 1.1

Introduction ........................................................................................... 5

1.2

Basic Characteristics of Particulate Fillers .............................................. 5 1.2.1

Cost ........................................................................................... 6

1.2.2

Chemistry, Composition and Impurities..................................... 7

1.2.3

Density or Specific Gravity ........................................................ 9

1.2.4

Hardness .................................................................................. 10

1.2.5

Abrasiveness ............................................................................ 11

1.2.6

Optical Properties .................................................................... 11

1.2.7

Thermal Properties .................................................................. 14

1.2.8

Particle Shape and Size............................................................. 16

1.2.9

Shape ....................................................................................... 17

1.2.10 Particle Size .............................................................................. 20 1.3 Surface Modification ............................................................................... 26

1.4

1.3.1

Stearic Acid and Stearates ........................................................ 26

1.3.2

Coupling Agents ...................................................................... 27

1.3.3

Polymer Modifications............................................................. 27

1.3.4

Direct Bonding ......................................................................... 28

Particle Packing and the Maximum Packing Fraction .......................... 29 1.4.1

Introduction ............................................................................. 29

1.4.2

Determination of Maximum Packing Fraction (Pf) by Oil Absorption Procedures............................................................. 29

1.4.3

Particle Packing Theory ........................................................... 30 i

Particulate-Filled Polymer Composites

1.4.4

Applications of Packing Principles to Particulate Filled Composites .............................................................................. 33

1.5

Interparticle Spacing ............................................................................ 34

1.6

Particle Effects on the Structure of Polymers ........................................ 36 1.6.1

Introduction ............................................................................. 36

1.6.2

Molecular Weight Reduction During Processing ...................... 36

1.6.3

Molecular Weight and Crosslinking Changes due to Cure Modifications .................................................................. 37

1.6.4

Preferential Adsorption of Polar Species .................................. 37

1.6.5

Formation of an Interphase of Immobilised Polymer ............... 38

1.6.6

Effects on Polymer Conformation due to the Presence of Particle Surfaces and Interparticle Spacing ............................... 42

1.6.7

Effects on Crystallinity............................................................. 42

References ..................................................................................................... 45 2

Principal Types of Particulate Fillers ............................................................. 53 2.1

Introduction ......................................................................................... 53

2.2

Particulate Fillers from Natural Origins (Mineral Fillers) .................... 53 2.2.1

Introduction ............................................................................. 53

2.2.2

Minerals and Rocks ................................................................. 54

2.2.3

Rocks ....................................................................................... 55

2.2.4

Calcium Carbonate Minerals ................................................... 57

2.2.5

Dolomite .................................................................................. 61

2.2.6

China Clay or Kaolin ............................................................... 61

2.2.7

Calcined Clay .......................................................................... 66

2.2.8

Mica ........................................................................................ 69

2.2.9

Talc .......................................................................................... 70

2.2.10 Montmorillonite (AlMg)8(Si4O10)3-(OH)10.12H2O ................... 72 2.2.11 Barite (BaSO4).......................................................................... 73 2.2.12 Calcium Sulfate Products ......................................................... 74 2.2.13 Wollastonite (CaSiO3) .............................................................. 74 2.2.14 Crystalline Silicas ..................................................................... 76

ii

Contents

2.3

Synthetic Particulate Fillers .................................................................. 78 2.3.1

Carbon Black ........................................................................... 78

2.3.2

Synthetic Silicas ....................................................................... 81

2.3.3

Hydroxides and Basic Carbonates ........................................... 84

2.3.4

Precipitated Calcium Carbonate (PCC) .................................... 96

Acknowledgements ....................................................................................... 96 References ..................................................................................................... 97 3

Analytical Techniques for Characterising Filler Surfaces ............................. 101 3.1

Introduction ....................................................................................... 101

3.2

Acid-Base Theory ............................................................................... 104 3.2.1

Introduction ........................................................................... 104

3.2.2

Gutmann Approach ............................................................... 105

3.2.3

Drago Approach .................................................................... 106

3.2.4

Use in Characterising Fillers .................................................. 106

3.3

Analytical Techniques ........................................................................ 108

3.4

Reactive Techniques ........................................................................... 109

3.5

3.6

3.7

3.4.1

Flow Microcalorimetry .......................................................... 109

3.4.2

Inverse Gas Chromatography ................................................ 119

Spectroscopic Techniques ................................................................... 124 3.5.1

Introduction ........................................................................... 124

3.5.2

X-Ray Photoelectron Spectroscopy ........................................ 124

3.5.3

Secondary Ion Mass Spectrometry ......................................... 130

3.5.4

Diffuse Reflectance Fourier Transform Infrared Spectroscopy . 134

Methods for Examining Structural Order in Filler Coatings .............. 145 3.6.1

Wide Angle X-Ray Diffraction ............................................... 145

3.6.2

Differential Scanning Calorimetry ......................................... 146

Summary ............................................................................................ 147

References ................................................................................................... 148

iii

Particulate-Filled Polymer Composites

4

Surface Modification and Surface Modifiers ............................................... 153 4.1

Introduction ....................................................................................... 153

4.2

Reasons for Using Surface Modifiers ................................................. 153

4.3

General Principles of Surface Modification ........................................ 154

4.4

Methods of Using Surface Modifiers .................................................. 155

4.5

Choice of Coating Level ..................................................................... 156

4.6

Techniques for Determining the Amount of Coating Present, and Assessing the Amount Needed for Mono-layer Coverage ........... 158

4.7

4.6.1

Determination of Amount of Additive and it’s Distribution ... 158

4.6.2

The Monolayer and it’s Determination .................................. 160

4.6.3

Effects of Processing on the Coating Structure ....................... 163

Surface Modifier Types ...................................................................... 163 4.7.1

Monomeric Organic Acids and their Salts ............................. 163

4.7.2

Stearic Acid (CH3(CH2)16COOH ........................................... 166

4.7.3

Other Saturated Fatty Acids and Related Substances ............. 170

4.7.4

Effects of Stearic Acid Coating in Composites ....................... 171

4.7.5

Fatty Acid Salts ...................................................................... 173

4.7.6

Unsaturated and other Functional Organic Acids in Composites ........................................................................ 173

4.7.7

Polymeric Acids and Anhydrides ........................................... 175

4.7.8

Organo-silicon Compounds ................................................... 177

4.7.9

Examples of Silane Coupling Agent Effects in Filled Polymers ...................................................................... 190

4.7.10 Organo-Titanates and Zirconates .......................................... 191 4.7.11 Aluminates and Zircoaluminates ........................................... 198 4.7.12 Phosphates and Borates ......................................................... 199 4.7.13 Organic Amines and Amino-acids ......................................... 200 4.8

Conclusions ....................................................................................... 200

References ................................................................................................... 201

iv

Contents

5

Compound Preparation, Mixture Characterisation and Process Enhancement of Particulate-Filled Polymer Compounds ............................. 207 5.1

Introduction ....................................................................................... 207

5.2

Functional Characteristics of Compounding Machinery .................... 209

5.3

5.4

5.5

5.2.1

Transport of Feedstock .......................................................... 210

5.2.2

Melting and Shear Heating .................................................... 213

5.2.3

Mixing ................................................................................... 214

5.2.4

Melt Devolatilisation ............................................................. 217

5.2.5

Melt Pumping and Pressurisation .......................................... 218

Constructional Design of Compounding Plant ................................... 219 5.3.1

Low and Medium Intensity Premixing Procedures ................. 220

5.3.2

High-Intensity Compounding Machinery .............................. 221

Characterisation of Filled Compounds .............................................. 228 5.4.1

Introduction ........................................................................... 228

5.4.2

Residence Time Distribution .................................................. 229

5.4.3

Specific Energy Input ............................................................. 231

5.4.4

Screen Pack Analysis .............................................................. 231

5.4.5

Rheological Analysis .............................................................. 232

5.4.6

Ultrasonic Measurement ........................................................ 233

5.4.7

Microstructural Analysis........................................................ 235

5.4.8

Miscellaneous Methods of Analysis ....................................... 239

Process Enhancement of Particulate Polymer Composites .................. 240 5.5.1

Addition of Rigid Particulate Fillers ....................................... 241

5.5.2

Effects on Polymer Molecular Weight .................................... 243

5.5.3

Short Fibre-Reinforced Thermoplastics Composites .............. 244

5.6

Woodflour and Natural Fibre-Filled Thermoplastics ......................... 246

5.7

Supercritical Fluid Assisted Processing of Filled Compounds ............. 249

5.8

Processing of Thermoset Recyclate Waste Materials .......................... 250

5.9

Preparation of Silicate Layer Polymer Nanocomposites ..................... 251

v

Particulate-Filled Polymer Composites

5.10 Conclusions ....................................................................................... 254 References ................................................................................................... 254 6

Effects of Particulate Fillers on Flame Retardant Properties of Composites ... 263 6.1

Introduction ....................................................................................... 263

6.2

General Effects of Fillers on Polymer Flammability ........................... 263

6.3

Fire Retardant Testing........................................................................ 264

6.4

6.3.1

Oxygen Index Test (ASTM D2863-87) .................................. 266

6.3.2

Underwriters Laboratory Vertical Burn Test (UL94 -1980) .... 266

6.3.3

Horizontal Burn Test (ASTM D635-88) ................................ 267

6.3.4

Ignitability Test (ISO5657 1986) ........................................... 267

6.3.5

Cone Calorimeter (ASTM E1354, ISO 5660) ........................ 267

6.3.6

Smoke and Corrosive Gas Tests ............................................. 268

Fire Retardant Fillers that Rely on Endothermic Decomposition ....... 269 6.4.1

Historical Background ........................................................... 269

6.4.2

Potential Endothermic Flame Retardant Fillers ...................... 270

6.4.3

Performance of Endothermic Flame Retardant Fillers ............ 273

6.4.4

Smoke and Corrosive and Toxic Gases .................................. 290

6.5

Nano-Clays ........................................................................................ 296

6.6

Ammonium Polyphosphate (APP) ...................................................... 297

6.7

Fillers for Use in Conjunction with Halogens .................................... 297

References ................................................................................................... 298 7

Particulate Fillers in Elastomers .................................................................. 303 7.1

Introduction ....................................................................................... 303

7.2

Uses of Elastomers ............................................................................. 303

7.3

Elasticity of Rubber ........................................................................... 303

7.4

Formulation of Elastomers ................................................................. 306 7.4.1

vi

General .................................................................................. 306

Contents

7.5

7.6

7.7

7.4.2

Selection of Polymer .............................................................. 306

7.4.3

The Curing System ................................................................. 308

7.4.4

Antioxidants and Antiozonants ............................................. 310

7.4.5

Coupling Agents .................................................................... 311

7.4.6

Process Oils and Plasticisers ................................................... 311

7.4.7

Fillers ..................................................................................... 313

7.4.8

Specialty Additives ................................................................. 314

The Performance of the Polymer ........................................................ 317 7.5.1

Specification of the Polymer .................................................. 317

7.5.2

Processing Considerations ..................................................... 321

7.5.3

Strength Characteristics of Polymers ...................................... 321

7.5.4

Compounding Considerations ............................................... 322

The Performance of Fillers ................................................................. 326 7.6.1

Reinforcement of Rubber by Fillers ....................................... 326

7.6.2

Processing Considerations ..................................................... 330

7.6.3

Compounding Considerations ............................................... 333

Filler Types ........................................................................................ 338 7.7.1

Specification of Fillers for Elastomers .................................... 338

7.7.2

Carbon Black ......................................................................... 340

7.7.3

Synthetic Silicas and Silicates ................................................. 343

7.7.4

Clay Minerals ........................................................................ 344

7.7.5

Calcium Carbonates .............................................................. 346

7.7.6

Alumina Trihydrate ............................................................... 348

7.7.7

Talcs ...................................................................................... 349

7.7.8

Natural Silicas ....................................................................... 349

7.7.9

Barytes and Blanc Fixe ........................................................... 350

7.7.10 Miscellaneous ........................................................................ 350 Acknowledgements ..................................................................................... 350 References ................................................................................................... 351

vii

Particulate-Filled Polymer Composites

8

Filled Thermoplastics .................................................................................. 357 8.1

8.2

8.3

8.4

8.5

viii

Introduction ....................................................................................... 357 8.1.1

Thermoplastics and Typical Applications .............................. 357

8.1.2

Thermoplastic Composites..................................................... 359

Bulk and Process Related Properties .................................................. 361 8.2.1

Specific Gravity or Relative Density ....................................... 361

8.2.2

Acoustic Properties ................................................................ 362

8.2.3

Melt Viscosity (MFI) .............................................................. 363

8.2.4

Compounding and Extrusion ................................................. 364

8.2.5

Thermal Conductivity and Specific Heat Capacity ................ 367

8.2.6

Thermal Expansion ................................................................ 368

8.2.7

Electrical Properties ............................................................... 369

8.2.8

Barrier Properties ................................................................... 370

Mechanical Properties ........................................................................ 371 8.3.1

Introduction ........................................................................... 371

8.3.2

Modulus – Tensile and Flexural ............................................. 372

8.3.3

Heat Deflection Temperature (HDT) ..................................... 374

8.3.4

Yield Strength ........................................................................ 375

8.3.5

Impact Strength (Toughness) .................................................. 376

Effects of Filler on the Polymer Phase ................................................ 380 8.4.1

Introduction ........................................................................... 380

8.4.2

Nucleation ............................................................................. 380

8.4.3

Transcrystallinity ................................................................... 381

8.4.4

Interphase .............................................................................. 382

Surface Science Aspects ...................................................................... 383 8.5.1

Introduction ........................................................................... 383

8.5.2

Surface Energy and Surface Tension....................................... 383

8.5.3

Wetting and Spreading ........................................................... 384

8.5.4

Adhesion ................................................................................ 384

8.5.5

Dispersion and Agglomeration .............................................. 387

8.5.6

Surface Treatments – Dispersants and Coupling Agents ........ 388

Contents

8.6

8.7

8.8

8.9

Aesthetics ........................................................................................... 390 8.6.1

Introduction ........................................................................... 390

8.6.2

Colour/Pigmentation .............................................................. 390

8.6.3

Surface Finish and Gloss ........................................................ 390

8.6.4

Scratch and Abrasion Resistance ........................................... 391

Stabilisation and Recycleability ......................................................... 392 8.7.1

Introduction ........................................................................... 392

8.7.2

The Effect of Filler Chemistry and Impurities on Stability ..... 393

8.7.3

The Effect of Antioxidant Adsorption on Stability ................ 394

8.7.4

Recycleability ......................................................................... 395

Uses of Filled Thermoplastics............................................................. 396 8.8.1

Uses of Fillers ......................................................................... 396

8.8.2

Fillers in PVC ......................................................................... 397

8.8.2

Uses of Fillers in Unplasticised PVC ....................................... 401

8.8.3

Uses of Fillers in Polypropylene ............................................. 403

8.8.4

Uses of Fillers in Polyethylene ................................................ 406

8.8.5

The Use of Fillers in Polyamides ............................................ 408

8.8.6

Polybutylene Terephthalate .................................................... 409

8.8.7

Polyethylene Terephthalate .................................................... 410

Conclusions ....................................................................................... 411

Acknowledgements ..................................................................................... 413 References ................................................................................................... 413 9

Filled Thermosets ........................................................................................ 425 9.1

Introduction ....................................................................................... 425

9.2

Brief Chemistry of Thermoset Polymers ............................................. 427

9.3

9.2.1

Free-Radical Chain-Growth Curing Resins ............................ 427

9.2.2

Step Addition Curing Resins .................................................. 435

9.2.3

Condensation Resins .............................................................. 440

Mechanical Properties ........................................................................ 444

ix

Particulate-Filled Polymer Composites

9.4

9.3.1

Modulus ................................................................................ 444

9.3.2

Fracture Toughness and Fracture Energy ............................... 450

9.3.3

Failure Stress .......................................................................... 463

9.3.4

Fatigue ................................................................................... 469

Applications ....................................................................................... 477 9.4.1

Cost Reduction ...................................................................... 477

9.4.2

Modified Mechanical Properties ............................................ 477

9.4.3

Exotherm Control .................................................................. 478

9.4.4

Shrinkage Control .................................................................. 480

9.4.5

Processing Aids ...................................................................... 481

9.4.6

Flame Retardants ................................................................... 482

9.4.7

Metal Fillers ........................................................................... 482

9.4.8

Structural Adhesives .............................................................. 483

Acknowledgements ..................................................................................... 483 References ................................................................................................... 484 10 Composites Using Nano-Fillers ................................................................... 489 10.1 Introduction ....................................................................................... 489 10.2 Scope ................................................................................................. 489 10.3 General Comments ............................................................................ 490 10.4 Nano-Filler Forms ............................................................................. 490 10.4.1 Regular Shapes ...................................................................... 491 10.4.2 Rods, Fibres, etc. ................................................................... 492 10.4.3 Platy Nano-Fillers (Nano-Clays and Related Materials) ........ 493 10.5 Summary and Future Perspectives ...................................................... 510 Acknowledgements ..................................................................................... 511 References ................................................................................................... 511 Abbreviations and Acronyms ...................................................................... 515 Author Index ............................................................................................... 521 Index ........................................................................................................... 527 x

Preface

I was delighted to accept the chance offered by Rapra Technology to produce this updated second edition of Particulate Filled Polymer Composites, a book first published in 1995. The first edition had been very well received but has been out of print for some time. Despite the relatively short time since the publication of the first edition, much has changed and hence considerable new material has been introduced, including a completely new chapter covering the latest developments in nano-filler technology. The use of particulate fillers in polymers has a long history, and they continue to play a very important role today. Despite the apparent commodity status, the area is still very dynamic and considerable changes have occurred in applications over the last few years. The most significant of these have been the dramatic growth in the use of precipitated silica in energy efficient tyre applications and the emergence of ‘nano-filler’ technologies. The multidisciplinary nature of this topic is well illustrated by the scope of the references in the book. These are drawn from many fields, including: mineralogy, crystallography, precipitation and crystal growth, powder technology, surface and colloid science, organic and organo-metallic chemistry, optics, materials science, and polymer science and technology. It also brings together people from many backgrounds, such as filler producers, machinery manufacturers, polymer compounders and suppliers of processing aids and surface modifiers. Together with my co-authors, I have set out to describe the fundamentals involved in producing, characterising and using particulate fillers in as clear and concise a way as possible. The authors have been encouraged to pay particular attention to those areas of their subjects which their experience shows cause the most confusion, or which are poorly covered elsewhere. Most importantly, where a topic is controversial, or poorly understood, they have been encouraged to interpret the existing literature, not merely quote opposing views, as is often the case. Where possible, use has been made of previously unpublished work to illustrate key points and extensive literature references are provided so that any subject can be followed up in depth, if needed. Chapter 4, on surface modification, is a good example of the approach taken. Most works concentrate on the organo-silanes. While they are still treated in detail here, significant space is also given to fatty acids and organo-titanates. Fatty acids are of considerable commercial importance and exhibit complexities that can cause problems

1

Particulate-Filled Polymer Composites for the unwary. Considerable controversy exists over the mode of action of the organotitanates and the available scientific literature is critically reviewed here. As with any editor, I am very much indebted to my co-authors for their hard work. I am also indebted to the staff at Rapra Technology, especially Frances Powers, for their help and support in turning the various contributions into a very well presented composite. Finally, I would like to express my particular thanks to the late Professor Derek Birchall, FRS, who first aroused my interest in this area and to the late Hugh Olmstead founder of Intertech Corporation and to my many friends and colleagues, especially those at Manchester Metropolitan University, Avecia Limited, Colin Stewart Minchem, CSIRO Division of Minerals, Electrolux and the Intertech Corporation, who have helped me to continue this interest in recent years. Roger N. Rothon July 2003

2

Contributors

David Ashton ICI Acrylics PO Box 34 Darwen Lancashire BB3 1QB David Briggs Siacon Consultants Ltd. 21 Woodfarm Road Malvern Worcestershire WR14 4RL Chris DeArmitt BASF Aktiengesellschaft KSP/S-E100 67056 Ludwigshafen Germany Michael Hancock 6 Poltair Avenue St Austell Cornwall PL25 4LY Graham Jackson LGC Ltd. The Heath Runcorn Cheshire WA7 4QD Peter Hornsby The Wolfson Centre Brunel University Uxbridge Middlesex UB8 3BH

Christopher Liauw Manchester Metropolitan University Centre For Materials Science Research John Dalton Building Chester Street Manchester M1 5GD Michael Orton 24, School Lane Hartford Cheshire CW8 1PE Roger Rothon Rothon Consultants and Manchester Metropolitan University 3 Orchard Croft Guilden Sutton Chester CH3 7SL David Skelhorn Performance Materials Division Imerys 100 Mansell Court East Suite 300 Roswell Georgia 30076 USA Howard Taylor Manchester Metropolitan University Department of Engineering and Technology John Dalton Building Chester Street Manchester M1 5GD

3

Particulate-Filled Polymer Composites

4

1

General Principles Guiding Selection and Use of Particulate Materials Roger N. Rothon and Michael Hancock

1.1 Introduction Particulate-filled polymer composites have a long history and consequently newcomers to the field usually expect to find an area of well-understood science with few intellectual challenges remaining. However, they are usually amazed to find that this is far from the truth in many areas, with few reliable generalisations (but several unreliable ones) available and much basic information yet to be established. This is largely due to the way in which the technology has developed, with different filler and polymer combinations tending to be developed largely piecemeal to meet the specific demands of various industries. The initial, and to some extent continuing, emphasis on cost reduction has also meant that many fillers have been poorly characterised. The purpose of this chapter is to give readers sufficient background so that they may approach the subject with an open mind and see the whole area as one with stimulating intellectual challenges, not as an area of mystery and witchcraft. This chapter discusses how fillers can be characterised and defined with the least ambiguity, and hence how their behaviour in composites can be understood and hopefully predicted. We also highlight the principal sources of misunderstandings and the limitations imposed by commonly used methods of measurement, and explain why and where deviations from general rules are likely to occur.

1.2 Basic Characteristics of Particulate Fillers In developing a particulate-filled composite, the formulator needs to be able to answer the following questions: 1. What property benefits are being sought? 2. What deleterious changes may also occur and can they be tolerated? 3. How easy is the filler to handle and how might it affect processing?

5

Particulate-Filled Polymer Composites 4. Are special additives needed? 5. What is the true cost of using the filler, is it justifiable, and are there more cost effective alternatives? Important information for answering the above questions includes, cost, purity, particle size and shape, density, hardness, optical and thermal properties and chemistry. The primary source of this information should be the filler supplier, although frequently data may be sparse. A brief description of each factor, its measurement and significance follows.

1.2.1 Cost In theory this is a fairly simple topic, but one that causes considerable problems for the unwary. It is widely assumed that polymers are expensive and fillers are cheap, and many articles on filled polymers start with the statement that fillers are used primarily to reduce costs. While this can often be the case, these savings are frequently not as great as anticipated and, in quite a few instances, compound costs, even with the lowest cost filler, can be higher than the unmodified polymer. There are two principal reasons for this. Firstly, the process of compounding filler into polymer costs money in the form of capital investment, manpower and energy [1]. In cases where compounding is essential, because other additives such as stabilisers or curatives have to be added to the polymer, then the cost of incorporating a filler is markedly reduced. In these cases, exemplified by elastomers, polyvinyl chloride (PVC) and thermosets, the use of fillers is the rule rather than the exception, unlike the case with, say, polyethylene. Secondly, prices of fillers and polymers are usually quoted in weight terms, while the majority of their composites are used on a volume basis. The specific gravity of most fillers is two to three times that of common polymers and, while raw material savings may accrue on a weight basis, more mass of the final compound will have to be used to achieve the same volume than would be the case with the unfilled polymer. When looking at potential cost savings, comparison of polymer and filler costs must therefore be made on a volume basis. On this basis one should generally regard the effective raw material costs of mineral fillers as two to three times their price (by weight). As a consequence many fillers will not give significant cost savings in the commodity plastics, such as polypropylene (PP) and this explains why the emphasis today is on achieving specific effects. However, while fillers may not ‘cheapen’ a filled composite they may well produce a cost-property performance that allows the composite to compete against and often replace more expensive systems.

6

General Principles Guiding Selection and Use of Particulate Materials

1.2.2 Chemistry, Composition and Impurities While the chemical nature of the filler is frequently of little direct importance to its use in composites, it plays two important roles, in that it determines the structure of the mineral, and also the nature of the interaction between polymer and filler. However, away from these a priori considerations, only in a few cases does the chemical reactivity of the filler play a significant role in the properties of a composite.

1.2.2.1 Bulk Chemistry Intrinsically fillers can be divided into two types, reactive and inert. Reactive fillers will react with their environment. A good example of this is gibbsite (aluminium hydroxide), which will react with both acidic and basic substances. Aluminium hydroxide also loses its water of crystallisation at around 200 °C and this enables it to provide fire retardancy in polymer formulations. The silicate minerals: (kaolin, mica, talc, quartz, etc.), are, in classical chemical terms, virtually inert, only being attacked by very strong acids and alkalis. The carbonate minerals and the hydroxide minerals are very reactive to acids. The interactions between the constituent elements in a filler determine its molecular structure and hence crystallinity, which then dictates all the intrinsic properties of the filler.

1.2.2.2 Surface Chemistry Filler surface chemistries are of more significance than the bulk ones, as they determine both the rate of wetting and the strength of interaction with polymers. They are invariably different from bulk chemistry but, unfortunately, they are poorly characterised for many fillers. Because of the interest in this very important topic, techniques for surface characterisation are covered in detail in Chapter 3.

1.2.2.3 Surface Interactions Polymers have a very much higher (20-30 times) thermal expansion coefficient than mineral fillers. Thus, in many well-dispersed, hot processed composites, a radial compressive stress develops as the polymer cools leading to an intimate interaction between matrix and filler. The value of this stress can be calculated, and depends on both polymer and filler [2]. Because of this, the interaction will at the very least be mechanical, but other types will exist depending on the surface chemistry of the filler and also the chemistry of the polymer.

7

Particulate-Filled Polymer Composites Wettability of filler by polymer is an indication of compatibility between the two. Wettability by polymer is not readily measured directly, but the effect that the material has on surface tension of a liquid is a measure of its surface energy. Usually water is chosen as the medium. Schlumf [3] has reported surface energies for a variety of fillers (Table 1.1). There have been a wide number of claims that surface energies determine the forces between polymer chains and different phases, determining mechanical properties of the composite. Adhesion at a polymer-filler interface has been shown to exert a considerable influence on mechanical responses, and a correlation between acid-base characteristics of filler and polymer (as determined by inverse gas chromatography) and properties has been established [4]. Lewis acid-base interactions between filler and polymer are claimed to be the most important component in the adhesion of a filler to polymers and thus in determining properties of composites [5]. For some filler-polymer systems, the strength of interfacial acid-base bonding may be appreciably enhanced by surface modification of the filler, or by modification of the polymer, giving large increases in properties.

Table 1.1 Surface energies of fillers and plastics Material

Surface energy (m Jm-2)

Diamond

10,000

Mica

240-500

Glass

1,200

Titanium dioxide Kaolin

650 500-600

Calcium carbonate

65-70

Stearate coated calcium carbonate

25-30

Talc

65-70

Polymers

15-60

Polypropylene

31

1.2.2.4 Chemical Analysis and Impurities The purity of a filler is of importance both commercially and technically, i.e., the users need to know what they are buying, whether it contains components that may be detrimental to the properties of their product and whether it will cause environmental problems. Impurities include trace elements that may be on the filler surface or in the structure of the filler and ancillary minerals, which will have been formed at the same

8

General Principles Guiding Selection and Use of Particulate Materials time as the main mineral (see Chapter 2). Sometimes additives from the filler manufacture may be present, (e.g., surfactants). The form of the impurity can be very important in determining it’s importance. Sometimes a potentially detrimental impurity can be tolerated in relatively high levels, if present in an inert form, but small traces of the same impurity, if present on the surface in an active form, can be very deleterious. Examples are certain transition metals, (e.g., iron, manganese, copper), which can seriously affect colour and thermal stability if present in an active form, even at levels of a few parts per million. Thus some micas and talcs can contain high levels of iron in an inert form, which is not detrimental to polymer ageing, while others can have lower levels of more active material. Some impurities may also pose health hazards, examples being crystalline silica and asbestos. Fillers are not always analysed for such low-level impurities and it is a mistake to believe that, if a possible deleterious impurity is not quoted in a data sheet, it is not present in the product. Conversely, it must be stressed that, if an element or substance is present in a non-active form, it may have little effect on polymer properties [6]. It is also necessary to point out that the way chemical composition is presented, can, and does, cause misunderstandings. Because of the methods of analysis used, many mineral compositions are quoted as a list of oxides rather than the actual constituent chemical. Thus a calcium carbonate may be specified as so many percent CaO or a clay may be reported in terms of SiO2 and Al2O3. This does not imply that any of the oxides are present in the free state. It should also be recognised that some standard methods of analysis do not detect very light elements.

1.2.3 Density or Specific Gravity This is one of the simplest of properties to define, understand and use. Density is mass per unit volume with SI (Système Internationale) units of kilograms per cubic meter. Specific gravity and relative density, the weight of the substance in relation to the weight of an equivalent volume of water, are often used synonymously. They are dimensionless but have the same numerical values as density. Density is determined by the size of the atoms forming the mineral, the closeness with which they are packed together and by impurities present in the crystal lattice: the heavier the atoms and the more tightly they are packed together, the higher the density. Thus, smithsonite, ZnCO3, has a higher specific gravity (4.3) than calcite, CaCO3 (2.7) because zinc has a much higher atomic mass than calcium, but magnesite, MgCO3, has a higher specific gravity (3.0-3.2) than calcite despite the fact that magnesium is lighter than calcium. This is because its ions pack much more closely together. In some fillers air is deliberately introduced into a matrix to reduce its density.

9

Particulate-Filled Polymer Composites True specific gravities of particulate additives can range from less than 1.0 for hollow glass beads to over 6.0 for some metallic fillers, but most lie in the range 1.6-2.8, while common polymers range from 0.9 to 1.4. Powder densities are of commercial importance, indicating ease of handling, and polymer compounding, and even affecting plant design. Values for bulk and packing densities are usually supplied by filler producers and are a measure of how the powder packs under various conditions. They are determined to some extent by the true density, particle shape and size of the filler, but the processing involved in its production also plays an important role. Surface treatments, and the methods used to apply them, can also be important. A powder with a low bulk density relative to its true density will contain a great deal of air. It will flow easily but will be difficult to incorporate into polymers, especially in equipment such as internal mixers and screw compounders.

1.2.4 Hardness Hardness is the resistance of a mineral to scratching. It is related to the structure of the mineral, the strength of the chemical bonds and the density of packing of its constituent atoms. F. Mohs, a mineralogist drew up a table of hardness in 1812 called the Mohs Hardness Scale (Table 1.2), which is still used to rank minerals by their resistance or susceptibility to scratching by other minerals. This scale is widely used for mineral fillers, but care must be taken in using the data because the scale is not linear, with the differences becoming greater at the high end. The hardness of most common particulate fillers range from 1 to 4. Talc is the softest, with a value of one.

Table 1.2 Mohs scale of mineral hardness 1

Talc (softest)

5

Apatite

8

Topaz

2

Gypsum

6

Orthoclase

9

Corundum

3

Calcite

7

Quartz

10

Diamond

4

Fluorite

Hardness must not be confused with toughness, which relates to the ability of the mineral to resist fracture. Hard minerals can often fracture very easily, for example, by cleavage along crystal planes.

10

General Principles Guiding Selection and Use of Particulate Materials Where mixed fillers are used, if one is much harder than the other, then abrasion of the softer one may occur. This is of particular importance when hard particulates are used with easily damaged glass fibres. Minerals such as talc are soft because they readily delaminate due to weak structural features (see Chapter 2). Such soft particles can fracture or cleave when polymer composites are deformed, thus limiting their reinforcing properties [7].

1.2.5 Abrasiveness Abrasion due to the filler has quite an important role in the processing of filled polymers. It has several origins. Mohs hardness is a major factor with the harder minerals, such as quartz and wollastonite (CaSiO3), approaching, or even exceeding the hardness of the metals used in processing. Large particles and angular particles with sharp edges are particularly detrimental, probably due to a scratching mechanism, followed by attrition of the exposed edges. Sometimes the coarse component of a filler may be predominantly a harder impurity, which leads unfortunately to a combination of the two mechanisms. Chemical attack may also be caused by some reactive fillers, leading to etching and progressive erosion.

1.2.6 Optical Properties The colour, opacity and gloss of a composite are very important, both aesthetically and functionally, and will be strongly affected by the incorporation of fillers. When light strikes any surface, it is subjected to several effects: absorption, scattering, reflection, polarisation and interference. The extent of each effect is dependent on the nature of both exit and entrance media, but for all practical purposes only three media, air, filler and polymer, will be considered here in a superficial treatment. The reader is referred to works such as by Bohren and Huffman for a detailed explanation [8]. A light ray incident upon an air-solid interface at an angle, i, will be bent or refracted somewhat normal to the interface, at an angle of refraction, r. A relationship of the incident angle, i, and the refracted angle, r, sin i/sin r, is constant for a given isotropic solid. The constant always denoted as n is called the refractive index. The refractive index of a filler is one of the most important parameters affecting the optical properties of a filled polymer system and is always quoted relative to air. It is, in fact, a manifestation of the ratio of the speed of light in air to its speed in the solid.

11

Particulate-Filled Polymer Composites The refractive index of a mineral is the sum of the specific refractive indices of all individual components normalised for concentrations. Simplifying, it can be said that the refractive index of a solid is directly related to its density. When light strikes a surface, some will be reflected with the amount reflected being determined by the density of the constituent atoms and is thus affected by the refractive index, n. Reflection (r) and transmission (t) coefficients are given by: r=

1− n 1+ n

t=

2 1+ n

(1.1)

To a first approximation this leads to the fraction of incident light being reflected as simply (n – 1)2/(n + 1)2 and approaches zero as the refractive index approaches unity. The response of a plane surface to an incident beam of light is complicated, as briefly described previously. Mineral fillers, however, complicate the subject even more deeply in that they are a priori made up of particles that usually have sizes in the order of the wavelengths of visible light. Thus large surfaces will reflect and transmit the light waves but particles will scatter them. Therefore, light entering a typical powdered filler, comprising many small particles, will experience reflection, refraction, diffraction and interference at each air - particle interface. Refraction into the filler particles can lead to absorption of the light (the radiation absorbed will be converted into another form such as heat) and a measure of this is the absorption coefficient. The remainder of the light is diffused by scattering and the measure of this is its scattering coefficient. The colour (sometimes referred to as brightness, reflectance or whiteness) is related to absorption and scattering coefficients by Kubelka-Munk equations [9]. The reader is referred to the original paper and to some of the speciality books on the subject [10] for more details. The level of scattering is very dependent on particle size, so that finer fillers appear whiter than coarser ones: as an approximation, scattering in air is at a maximum when the particle size of powder is one-third to one-half the wavelength of the radiation. Levels of reflectance and refraction, and hence scattering, are directly related to the refractive index of the mineral. When incorporated into a polymer matrix, the ratio of its refractive index to that of the polymer will strongly affect the optical properties of the composite. As the ratio approaches unity (which is the case for most common polymers and filler): reflection at the interfaces between filler and polymer approaches zero. This means that light scattering is reduced dramatically and light absorption becomes obvious. Mie’s theory [11] now widely used for particle size determinations from light-scattering data, predicts that as differences in refractive indices become small, not only does scattering become less but particle size effects become less and maximum scattering moves to larger sizes, often several times greater than the wavelength of the light.

12

General Principles Guiding Selection and Use of Particulate Materials A filler such as titanium dioxide, having a refractive index much higher than that of the polymer, will therefore exhibit considerable scattering and will give white products, unless its absorption coefficient is high in which case dark, opaque composites will be produced (carbon black or black iron oxide are the ultimate in this instance). Most fillers are white or off-white powders due to the reflection of visible light. This reflection consists of two components, absorption and scattering, and, as discussed previously, most fillers have low absorption but high scattering coefficients. However, the amount of light scattered depends on the refractive index of the mineral relative to the medium in which it is measured. In air the differences in refractive indices mean that most minerals scatter light and appear white. Most plastics, on the other hand, have refractive indices close to those for minerals (1.5-1.6) and much of the scattering is lost (the light, in fact cannot ‘see’ the boundary between filler and polymer). There are a few cases in which mineral and polymer have almost identical refractive indices. In these instances, scattering virtually vanishes and the composite becomes almost transparent until loadings are increased to a level at which particle-particle interaction occurs [12]. This transparency is of commercial importance in several applications such as silica-filled styrene-butadiene rubber (SBR) shoe soles, and calcined clay- and glass fibre-filled unsaturated polyester roof panels. On the other hand, for example, calcite has a refractive index of 1.6, which is sufficiently different from most of the common plastics for some scattering and hence pigmentation to occur, but whiteness and opacity will only become very noticeable at high loadings and with thick products. With the loss of light scattering, light absorption becomes the dominant optical property of a mineral-filled plastic. Light is absorbed by transition metal ions, with different wavelengths being absorbed by different species of ion, and by naturally occurring organics, which are poly-aromatic species with strong absorption across the whole spectrum (that is they appear brown or black). The impurities in mineral fillers are mainly transition metal ions, particularly iron, and humates or lignates. Colours due to these become frequently more marked in a plastic. Some minerals exhibit birefringence with refractive indices different in differing directions, due to an asymmetric crystal structure. Plastics filled with these are dichroic and exhibit unusual visual appearances. As well as being a direct cause of the colour in a filled plastic, a filler may indirectly generate colour by either degrading the polymer (forming radicals and unsaturated species) or by deactivating antioxidants/stabilisers, etc., used in the polymer. In some cases, breakdown of filler surface treatments can also generate colour problems. Filler effects can be observed in most thermoplastics but are particularly relevant in PVC where dehydrochlorination, giving a polyene conjugated structure, is readily promoted by Lewis acid sites on a mineral surface.

13

Particulate-Filled Polymer Composites During the processing of filled thermoplastics, but particularly during their injection moulding, white markings are observed at points where high melt strain rates are encountered. These are especially apparent around the sprue and are very much more noticeable than for unfilled plastics. Some may be due to water or gas bubbles (splash markings), some to poor dispersion or filler separation, but most seem to result from flow patterns. These have been frozen in a strained state before they have had time to relax due to the high thermal conductivity of filled polymers [13]. When filled polymers are strained, they whiten to some extent. This is partly due to orientation of polymer chains around filler particles but also to light scattering from vacuoles formed as the polymer is pulled away from the particle surfaces. Optical properties can also be affected by the fact that filler particles can affect the morphology of the plastic by nucleating its crystallisation at the interface. Thus a composite with different light scattering properties may be obtained.

1.2.6.1 Spectral Absorption The ultraviolet and visible spectra of the common fillers are largely featureless, unless there are significant levels of transition metal ions in the mineral, when strong absorptions will occur. However, the most important practical aspect is the infra-red (IR) spectrum of a filler. This arises from the nature of the chemical bonds present and will be unique for each filler type. A common use of IR spectroscopy is as an analytical tool to identify a filler. An important practical consequence of a filler’s IR absorption is the use of certain fillers in horticultural and agricultural films (see Chapter 8). Spectra of some of the more common minerals incorporated in a lowdensity polyethylene (PE) film are shown in Figure 1.1. Calcium carbonate shows only two strong absorption bands because the only energy absorptions occur in the CO3; ion (bending, stretching and deformation modes). Kaolin, a complex aluminosilicate, shows a much more complicated IR spectrum: after calcining, the —Al—O—Si— bonds rearrange and the resultant product has a very complex IR pattern with several broad, overlapping bands.

1.2.7 Thermal Properties Leaving aside the special cases of phase change and decomposition, the principal thermal properties of interest are specific heat, thermal conductivity and coefficient of expansion.

14

General Principles Guiding Selection and Use of Particulate Materials

Figure 1.1 Infra-red spectra of mineral filled polyethylene films

1.2.7.1 Specific Heat This is usually defined as the energy required to raise 1 gram of material by 1 K with SI units of J.kg-1.K-1. Specific heats are easily measurable and available for most particulates [14]. Where they are not, rules exist for estimating the value [7]. The volume specific heats of most inorganic fillers are similar to the common polymers and the rule of mixtures gives a good approximation to the values found in composites.

1.2.7.2 Thermal Conductivity This is the rate at which heat energy is transmitted through a substance, SI units are W m-1 K-1 but it is usually quoted as cal cm-1K-1. Mineral fillers are about an order of magnitude more conducting than most polymers, while metallic fillers are even more so. Filled composites are therefore more conductive in general than the base polymer. The simple law of mixtures is only applicable at low concentrations because, as with electrical conductivity, particle-particle contact will occur at high loadings leading to a sudden

15

Particulate-Filled Polymer Composites sharp increase in conductivity, the so-called percolation threshold [15], which is dependent on particle shape and packing. This increased conductivity is of considerable importance in some processing operations, especially injection moulding.

1.2.7.3 Coefficient of Thermal Expansion As the name implies, this is the rate at which a material changes volume on heating or cooling. SI units are K-1. Most polymers have coefficients of expansion at least an order of magnitude greater than mineral fillers: thus mineral-filled composites have lower coefficients of expansion than unfilled polymers and, in well-annealed systems, the rule of mixtures applies unless strong bonding occurs between filler and polymer. Some fillers have a negative coefficient of expansion and are used to achieve composites with zero expansion. Other fillers show anisotropy, that is they have coefficients that are direction dependent. The differences in expansion between particle and matrix can result in intimate contact between them in a composite and consequently considerable stresses occur. These can have the same properties as chemical bonding at low strains and can mask any effects of changes in polymer-filler interactions in some tests.

1.2.8 Particle Shape and Size The importance of these factors is felt at all stages of composite production and use. They affect powder flow, compounding behaviour, composite viscosity, and mechanical, thermal and optical properties. Indeed, most of the current predictive equations for the properties of filled composites use shape and size factors, often determined using model particles such as glass spheres and flakes. Unfortunately such particles are rare in the real world and it is important in applying equations to appreciate the limitations imposed by problems of adequately measuring and describing the morphology of fine particles. One of the principal difficulties is due to the ability of many fillers to exhibit a variety of particle shapes and sizes depending on the work done in dispersing them. Their ‘effective’ shape and size can therefore vary at any stage of composite formation and use. In principle one would like to characterise them in situ. This is, however, far less easy than characterising the initial particulate material itself and one is usually reduced to trying to carry out measurements under conditions that will represent as near as possible those encountered in use. In this context, the concept of ‘effective’ particle, which is the size and shape achieved in the actual application, is a very useful one to keep in mind and is returned to later in this chapter.

16

General Principles Guiding Selection and Use of Particulate Materials

1.2.9 Shape 1.2.9.1 Introduction Particle shape (as will be discussed in following chapters) is very important in determining the stiffness, or rigidity, of a composite, the flow and rheology of a melt or liquid, tensile and impact strength, and the surface smoothness of a component, i.e., many of the important properties of a composite. Shape is determined by the genesis of the filler, by its chemistry, its crystal structure and by the processing it has undergone. Unfortunately, it is usually poorly defined, the literature abounding with vague terms such as roughly spherical, blocky, irregular, platy, acicular, etc. Some typical particles likely to be found in fillers are shown in Figure 1.2. All the fillers commonly used are microscopic in size imposing major difficulties both in how to measure, and then how to describe and quantify shape in any simple yet meaningful way. In addition, there are problems of distinguishing between primary particles, agglomerates and aggregates that occur in a filler, especially in synthetic materials, in which a ‘structure’ is sometimes deliberately designed. The subject is complicated further because aggregates and even particles may break down during processing (see Section 1.2.10.3 for a definition of these various terms).

Figure 1.2 Some particle types likely to be found in common fillers

17

Particulate-Filled Polymer Composites Despite the complex problems outlined previously, fillers are used because of their shape in a very wide range of polymers to give specific properties. For example, conventional clays are used in hose and chemical lining because their shape reduces permeability to fluids; platy talcs give rigidity to PP; the complex aggregate structure of precipitated calcium carbonates contributes to the structure of liquid polysulfides; and the special structured shape of many carbon blacks and synthetic silicas is important to their performance in elastomers. A further example is the emerging use of very high aspect ratio, nano-clays as reinforcing, fire retarding and gas and fluid barrier fillers.

1.2.9.2 Origins of Particle Shape Synthetic products will have their shape determined by their chemical composition and by the production conditions. For example, precipitated calcium carbonate (CaCO3) can be produced with different shapes by changing precipitation conditions to produce aragonite, calcite or vaterite. These conditions can be chosen to produce single crystals or complicated aggregates, which can be modified during drying and milling. Another example is the very complex shape of most carbon black particles arising from the partial fusing and solidification of pyrolysing droplets into three-dimensional chain-like aggregates during manufacture. The external shape of a natural mineral is a manifestation of its crystal structure, which will be briefly discussed in Chapter 2. It is also dependent on the environmental conditions in which the mineral was formed. If allowed to grow without constraint, then the particles are bounded by crystal faces, which are disposed in a regular way such that there is a particular relationship between them in any one mineral species, which is derived from regular atomic arrangement. However, under pressure, temperature or the effects of impurities, the crystal may adopt different shapes or habits. These include cubic, fibrous (fine, long, needles), acicular (needle-like), lamellar (plate-like) and prismatic. It is very unusual for perfect crystals to be found, but even poorly formed ones will always show evidence of their intrinsic symmetry. Many fillers (as will be discussed later) being extracted from the earth or rocks and processed by fairly simple methods will exhibit the same basic original shapes. Others, however, may be extracted by complicated routes or involve intensive comminution, thus altering particle shape. This is determined by the chemical nature and strength of the bonds between the atoms and groups in the mineral. These depend on electronegative differences between atoms and the bonding can range from covalent through to ionic. Two other types of interaction or bonding are important. Hydrogen forms co-ordinate bonds with atoms with very high electro-negativities, such as oxygen. This is very important in hydrated minerals such as kaolinite and gibbsite,

18

General Principles Guiding Selection and Use of Particulate Materials with layers being bonded together by these relatively weak hydrogen bonds. In ‘neutral units’ another weak force, known as Van der Waals bonding, due to non-uniform charge distributions, bonds group together. The types and strengths of bonds present in the crystal determine the methods by which the minerals break down on processing. Thus in the calcite crystal, the weakest bonds exist between the calcium ion and the carbonate group and these cleave invariably, giving rise to a regular rhombohedral shape. Kaolin platelets, held by hydrogen bonds, separate more readily from each other rather than fracture. In talc, magnesium silicate layers are held together by weak Van der Waals forces and very readily undergo cleavage into progressively thinner plates.

1.2.9.3 Assessment and Measurement of Shape The shapes outlined previously are very difficult to specify precisely using Euclidean geometry, but various approaches have been attempted to describe shape, anisotropy and ‘structure’ as numbers. Structure is particularly interesting and, while only really recognised currently in the elastomer field, it probably has significance, as yet overlooked, in other areas. In synthetic fillers it is sometimes difficult to separate fundamental from aggregate shape but, where there is sufficient incentive, then ways will usually be found to overcome such difficulties. Such incentives arose in carbon blacks and more recently in precipitated silica, where shapes are very complex, but an understanding is critical to their high value usage in the tyre industry. Much work has been done, especially by Medalia and Heckman [16], and by Hess and co-workers [17], to develop automatic image analysis procedures. Using such procedures, all the aspects described previously have been investigated. This work has much to teach us about other filler particles. Fractal geometry is now showing great promise for describing the shape of various complex, irregular particles. As usual, the carbon black industry is leading the way in exploring the potential of fractals for describing fillers and promising results are already emerging. A good introduction to fractal geometry and its applications to particles can be found in the excellent work by Kaye [18], and a number of recent papers have explored the measurement and significance of fractal dimensions of filler particles including carbon black [19-21]. The importance of measuring ‘structure’ results from its role in describing the ability of a particle to trap and partly shield a portion of the polymer matrix from deformation. This is generally known as an occluded polymer and its presence is important, especially in many of the unique properties of carbon and silica-filled elastomers. A simplified picture is given in Figure 1.3 and structure can be thought of as increasing the effective filler volume. While the carbon black industry is leading the way in using sophisticated techniques to measure shape and assess structure, they are also prepared to use fairly simple measurements, and in particular the oil absorption procedure with certain refinements, to give a quantification of this parameter [22].

19

Particulate-Filled Polymer Composites

Figure 1.3 Schematic illustration of the effect of particle structure and occluded polymer

The shape of simple particulate fillers is usually expressed as its aspect ratio (AR), which is the ratio of the particle’s diameter to its thickness. Measurement of aspect ratio is relatively simple for large particles, although time consuming. But, for the micrometer-sized particles that are normally encountered in fillers used in polymers, it is difficult, needs expensive equipment (such as scanning electron microscopes), and is very time consuming, depending on the spread of aspect ratios normally found. Other techniques are being sought to give an easier, more rapid measurement. Fourty reports that a comparison of particle-size distribution curves measured by sedimentation and laser diffraction techniques can give a shape factor [23]. Electrical conductivity of an aqueous suspension of a mineral under flow compared with it at rest (after the particles have randomised) gives a mean value for AR or a shape factor. Reasonable agreement with electron microscopy has been found [24]. This complex situation is further complicated by the fact that not only is shape often dependent on size, but measurement of size is affected by shape. This interdependence is discussed next and also in the descriptions of individual minerals given in Chapter 2.

1.2.10 Particle Size 1.2.10.1 Introduction Particle size is a very important, property of a filler, but is a particularly complex area where great confusion still occurs. For an individual, naturally occurring, filler it will have been determined by the origin and mineralogy of the deposit from which it has been extracted, by the method used in mining, and by separation procedures used during processing. For synthetic fillers, size will be determined by the conditions used

20

General Principles Guiding Selection and Use of Particulate Materials in its synthesis such as precipitation and possibly by the drying and any coating procedures. Size is one variable that can be controlled (in theory at least), and its importance is felt at all stages of composite production and use. Hence, there is considerable interest in its measurement. Particle size distributions are more useful than single average values, although the latter have the merit of simplicity. Size is an easy property to measure reproducibly using a variety of techniques including sieving, sedimentation, optical scattering and diffraction from particulate suspensions. Each, however, has a different dependence upon particle dimension and varies with particle shape. Thus correlation between different minerals measured by different techniques, in different laboratories is difficult. Again the reader is referred to one of the specialised books on the subject [25]. The problems arise from the fact that most filler particles are irregular in shape and contain a wide range of sizes, some of which will be individual particles and others agglomerates (also with a different shape from the individual particles). In such instances it is not possible to describe fully the particle size by a few numbers, as is often attempted. Commercially, most producers will quote two or three values to indicate the size of their fillers (see Section 1.2.10.2 and Chapter 2). Thus the data used in published studies are often a poor approximation of the real situation and can be very misleading. Wherever possible, the reader should try to determine and to use the full particle size distribution curve and be very cautious in interpreting literature studies in which only limited data such as average particle sizes or surface areas are given. In particular very small amounts of particles above or below a critical size can sometimes dominate certain properties, (e.g., viscosity, fatigue, tensile and impact strength) but will not be apparent in many particle size determinations. An interesting example of the effect of small amounts of oversize particles on the properties of carbon black-filled elastomers has been given by Gent [26].

1.2.10.2 Particle Size Measurement As already stated, filler suppliers usually quote three or four properties as a measure of the particle size: 1. Top cut. This is the size below which the majority (usually 99% or 99.9% by weight) of the particles are finer. 2. Coarse particles may also be reported as that percentage above a certain size. 3. Average particle size is obtained from particle size distribution curves and is the size that 50% of the particles are below. This can be defined in a variety of ways, e.g., by weight, volume or number, and care must be taken to understand the definition being used.

21

Particulate-Filled Polymer Composites 4. Specific surface area. This is determined by a number of adsorption techniques and is affected by size, shape and structure of the filler particles. Coarse particles, i.e., particles above 40 μm in size are usually quoted separately and are measured by screening the filler through standard mesh sieves. This may be carried out in aqueous suspension or as a dry powder. If the filler is heavily agglomerated then erroneous results may be obtained especially in dry screening. For accuracy, aqueous measurements are preferred with the filler being completely dispersed both by mechanical means and also using a dispersing agent. An idea of the structure and ease of dispersion of a filler may be obtained by comparing screen residues obtained by both wet and dry screening. If the filler is hydrophobic, then difficulties will be encountered in wet screening. These may be overcome by dispersing the filler in an organic liquid such as butanol, but a wetting agent and a dispersing agent are often both required. For some applications it is important to know the amount of particles that exceed 10 μm. This may be obtained by sieving but practical difficulties due to sieve screen blinding can occur. The most widely used techniques in this case involve sedimentation in a fluid, usually water and applying Stokes’ law. This states that, under standard equilibrium conditions, the time, t, taken for a particle to settle to a fixed depth is inversely proportional to the square root of its spherical diameter, d. This is given in Equation 1.2: ⎡

1/ 2

⎤ ⎥ d = 10 ⎢ ρ s − ρ f gt ⎥ ⎣ ⎦ 4⎢

(

18ηl

)

(1.2)

where d is the particle diameter, η is the viscosity of the fluid, ρs and ρf are the densities of the solid and the fluid, respectively, g is the gravitational constant and t is the time for the particle to settle a distance, l. Diameters obtained in this manner are precise descriptions of the particles but will compare with the results of other techniques only when the particles are perfectly spherical. In practice, most fillers have irregularly shaped particles and it is common to interpret experimental data in terms of theories applicable to spherical particles. Dimensions obtained are then for ‘equivalent spherical diameters’ (esd), which are the diameters of spheres that would give the same behaviour as that obtained from the sample by the method in question. Specific surface area is usually measured by the quantitative adsorption of nitrogen following the procedure originally described by Brunauer, Emmett and Teller, and known as the BET method (Equation 1.3) [27]: 22

General Principles Guiding Selection and Use of Particulate Materials P V (Po − P )

=

C −1 P 1 + VmC VmC Po

(1.3)

where V is the volume of gas, under standard conditions, adsorbed at equilibrium pressure, P; Vm is the volume of gas necessary for monolayer coverage; Po is the saturation pressure; and C is the BET constant. Often only a single point measurement is made to obtain surface area, but it is much more reliable to measure adsorptions at several pressures and then plot P/ V(Po – P) against P/Po. Permeability of a bed of packed particles to a fluid can also be used to give a surface area [28]. However, there are major sources of difficulty and confusion in describing particle size and shape. These include adequately defining what subdivision of material is to be described as the particle, in relying on particle size measurements carried out on the powdered particulate as an adequate description of it in a composite, and how shape and size interact.

1.2.10.3 Primary Particles, Agglomerates and Aggregates It surprises many people that a variety of ‘particles’ can exist in a powder, each of which can be determined as the particle size under certain conditions of dispersion and measurement. Moreover, the appropriate particle size for consideration can itself vary according to whether one is dealing with powder flow, behaviour during compounding and dispersion, or the properties of the final composite. These different types of particles are generally described as primary or ultimate particles, agglomerates and aggregates. In some instances these particle types are readily distinguishable but in others there can be appreciable overlap. Before discussing the types of particle in detail it is necessary to clarify the terminology as two contradicting conventions are widely used, and this in itself is a cause of considerable confusion. The need is to distinguish between collections of particles that are weakly and strongly bonded together. In this book we shall use the term agglomerate for weakly bonded particle collections and aggregate for strongly bonded ones. This is opposed to the views of Kaye [18] based on the derivations of the two words. The chosen terminology is, however, at least as widely used and is especially prevalent in the carbon black industry. The reader should always ascertain which terminology is in use when reading articles in the literature. An idealised view of particle type and breakdown with work during composite formation is presented in Figure 1.4. This goes beyond primary particles and considers the case where fragmentation of these may occur, e.g., hollow glass beads during thermoplastic compounding. With such simple systems it is usually fairly easy to match the particle sizing method to the degree of dispersion expected in the composite and obtain realistic answers. 23

Particulate-Filled Polymer Composites

Figure 1.4 Idealised view of the way filler particles disperse and of the different forms of particle types that might be encountered

Unfortunately, many filler systems do not exhibit such simple profiles, and the steps shown are often less sharp and overlap. Some possibilities are shown in Figure 1.5. A further complication arises when agglomerates form from initially well-dispersed systems. These agglomerates are sometimes referred to as flocs and can arise due to loss of colloidal stability in polymerising systems, or to reticulation (filler network formation) above the glass transition, especially in cured elastomers, an effect often observed with carbon blacks. The most difficult situation to deal with is quite frequently met with synthetic products, especially those formed by precipitation. This is where quite strong, complex aggregates are present, in addition to agglomerates. These aggregates often break down slowly, leading to a drawn-out step in the effective size profile. Frequently, they do not fully break down to primary particles under normal processing conditions. The effective particle size will then be critically dependent on the exact processing conditions and will be very difficult to predict in advance. Precipitated calcium carbonates are a good example where this type of situation is encountered. Most particle sizing techniques attempt to break down powders to their primary sizes by use of intensive mixing, ultrasonics and dispersants, but even these can be insufficient. Curves of the type shown in Figures 1.4 and 1.5 would be very useful for predicting filler behaviour, especially if the work input could be related to different compounding procedures. In principle, this should be possible, at least in a semi-quantitative way, by varying the energy used in dispersing the powder. Such procedures have been found

24

General Principles Guiding Selection and Use of Particulate Materials

Figure 1.5 Complex particle dispersion behaviour, as often encountered with fine, precipitated fillers

valuable by one of the authors and some very useful preliminary work of this type has been reported by Thoma and co-workers [29]. Some typical results obtained by the author (RR) with a coated precipitated calcium carbonate are given in Table 1.3 (Note only average particle size values are given for illustrative purposes, the earlier comments about

Table 1.3 Effect of measuring conditions on the apparent particle size of a coated precipitated calcium carbonate Measuring Condition

Average particle size (μm)

Comments

Laser diffraction of organic dispersion using weak ultrasonics

20

Detecting agglomerates

Laser diffraction of organic dispersion using medium ultrasonics

4

Detecting basic aggregate size

Laser diffraction of organic dispersion using strong ultrasonics

0.2

Detecting some form of sub-aggregate structure

Electron microscopy

0.07

Detecting primary crystallite size

X-ray line broadening

0.07

Detecting primary crystallite size

25

Particulate-Filled Polymer Composites the dangers of using a single value for a distribution still apply). Furnace-type carbon blacks provide a good example of the problems involved with concepts and terminology in this area. Such carbon black particles are formed by partial fusion and solidification of very small spherical units. These spherical units are generally referred to as the primary particles. Under no compounding or particle-sizing conditions, however, is it possible to break down the actual particles completely into these ultimate units. Unfortunately, some breakdown of the fused structures does occur, depending on the processing conditions, so that it is not appropriate to consider these fused structures as the ultimate particle either. In view of these complexities, the best rule is to keep the basic principles as outlined previously firmly in mind when trying to relate particle size as determined on a powder to what happens in an actual composite. As mentioned before, many of the previous problems would be removed if there were simple and reliable methods for assessing the ‘effective’ particle size in the polymer matrix. Various methods can be used including microscopy and contact radiology, both coupled with image analysis, and recovery from the polymer followed by conventional sizing techniques, although none are completely satisfactory. A more detailed discussion of some of these aspects will be found in Chapter 5. In the authors’ opinions the understanding of filled composites would be greatly improved by advances in this area.

1.3 Surface Modification Surface modification of fillers to give improved properties to a polymer composite is a topic that has received enormous attention over the last 30 years. Improvements in mechanical properties, dispersion of the filler (which leads to improved properties), improved rheology and higher filler loading have all been reported to accrue from rendering the surface more hydrophobic and hence compatible with the polymer or by enabling the filler to bond covalently, through hydrogen or ionic bonds to the polymer; or by changing the physical nature of the interface so that energy absorption can occur. This section will deal mainly with mineral surface interactions. The use of surface modifiers is dealt with at length in Chapter 4 and specific examples are described in Chapters 2, 7, 8 and 9.

1.3.1 Stearic Acid and Stearates The most widely used surface modification is treatment with stearic acid. This is believed to result in a stearate salt coating on most fillers and metal stearates are also used. Stearic acid will react with basic minerals to give a surface that is covered with strongly bonded long organic ions (this is discussed in more detail in Chapter 4). Stearic acidmodified silicates are commercially available but in these cases the stearic acid is almost

26

General Principles Guiding Selection and Use of Particulate Materials certainly weakly adsorbed and probably desorbs during melt compounding. Similarly, metal stearates will form weak bonds with mineral surfaces and desorb from them. Organo-amines have been used to render silicate surfaces hydrophobic, bonding probably through strong coordinate bonds. These relatively labile coatings will help improve dispersion in the initial stages of a compounding operation, before they desorb. They also give a protective coating to the filler, minimising any polymer degradation that may occur before stabilisers, antioxidants, etc., are fully dispersed.

1.3.2 Coupling Agents Surface treatments with bi-functional additives, which form very strong covalent bonds to the filler and then bond to a polymer by a variety of mechanisms, are widely available. They are based on organo-metallic compounds with the general formula: (RI)a M(RII)b

(1.4)

where M is a metal ion with valency a + b; RI is an organic group, the choice of which depends on the polymer in which it is to be used and RII is a group designed to react with a mineral surface. The most commonly encountered compounds are organo-silanes [30] and organotitanates [31], but others, including organo-zirconates [32], organo-borates [33] and complexes of chromium [34] have been proposed. R II in most of the coupling agents is an alkoxy group (methoxy, ethoxy or 2-methoxyethoxy are common), which is reactive to hydroxyl groups and to water. The preferred reaction mechanism is for hydrolysis to occur, initially by reaction with environmental or adsorbed water, with the kinetics being affected by the nature of both the alkoxy and other reactive groups on the molecule. The organo-metallic hydroxide then condenses with hydroxyls on the mineral surface or it can form oligomers followed by polymers by self-condensation reactions [35]. Multiple layers of many coupling agents can be adsorbed onto a mineral surface, with the packing of the coupling reagent on the surface being determined by the size of the RI organic group [36]. The structure of these layers can be very complex, with both strongly and weakly adsorbed species being present. This is discussed in detail in Chapter 4.

1.3.3 Polymer Modifications To try to improve energy absorption at the interface, and hence improve composite toughness, modification of the filler with a polymeric, low modulus, interlayer has

27

Particulate-Filled Polymer Composites been reported [37]. There have been many papers published on this technique. Aivazyan and co-workers [38] review the surface modification of mineral fillers by polymer deposition, using polymer adsorption from solution, mixing of polymer with dispersed filler and hetero-coagulation of latex polymer particles on the filler. Good bonds between coating and filler are often achieved by grafting a reactive group on to the interfacial polymer before or during filler treatment. Thus, kaolin can be coated with a styrenediacetone acrylamide copolymer by dispersing the two monomers with the clay in an aqueous medium and polymerising with a persulfate catalyst [39]. Unsaturated monomers have been adsorbed on to fillers and then polymerised to give encapsulated products. The modulus of the polymer could be modified by selecting the monomer [40]. Acrylic acid-vinyl chloride (1:99) has been polymerised on to calcium carbonate [41]; reports of the use of 3,5-triacryloxyhexahydro-S-triazine [42], bis-phenol A and epichlorohydrin [43], methyl methacrylate [44], and acrylic acid [45] have also been published. Graft copolymers between unsaturated acids, especially acrylic acid and maleic anhydride (MA), and polyolefins (PE and PP) are widely used as surface modifiers and compatibilisers, sometimes in combination with bi-functional coupling agents [46], for talc, calcium carbonate and calcined clays. Such polymer coatings include polypropylene-maleic anhydride [47], polypropylene cis-4-cyclohexene-1,2 dicarboxylic acid [48], polystearyl or polylauryl acrylate [49], polypropylene-acrylic acid, partially oxidised poly(butane diol) [50] and ethylene-vinyl acetate copolymers [51]. Acidcontaining products can react with basic fillers. With most other types, they will simply adsorb on to the mineral surface, but they can form esters with some non-basic metal hydroxyls, notably silanols.

1.3.4 Direct Bonding Polymers themselves can be grafted on to a mineral surface and the resulting composites have been reported to be significantly better than simple mixes because of strong covalent polymer-filler bonds. Composites of clay with ethylene-MA graft copolymer, in which the anhydride groups interact with the hydroxyl groups on the surface of the clay, give increased impact strength and flexural modulus compared with a physical blend of clay and ethylene-MA random copolymers [52]. Polymerisation of monomers can be induced on the mineral by natural catalytic sites on its surface [53] or by adsorbing polymerisation catalysts on to the mineral prior to treatment with monomers [54, 55].

28

General Principles Guiding Selection and Use of Particulate Materials

1.4 Particle Packing and the Maximum Packing Fraction 1.4.1 Introduction The packing behaviour of particles in a polymer matrix determines at what loading particle/particle effects become important and is a critical factor in the understanding and design of polymer composites, especially when highly filled systems are involved. The maximum packing fraction, Pf, is a particularly useful concept. This is the maximum volume fraction of particulate that can be incorporated before a continuous network is developed and voids begin to appear in the composite. In addition to indicating the maximum practical loadings obtainable, Pf is also useful in understanding and describing the effect of filler loading on composite properties. Many properties change rapidly as one approaches Pf, almost in a percolation way, and this parameter is now incorporated into many mathematical treatments of property-loading dependence [56]. Indeed, it often goes a long way to explaining why different physical forms of the same filler material can give markedly differing results at the same loading.

1.4.2 Determination of Maximum Packing Fraction (Pf) by Oil Absorption Procedures Despite its utility, Pf is difficult to measure or predict with great precision. This is due to complex interactions between the particles and the polymer matrix, and the influence of the fabrication method used. Both of these factors affect the way the particles pack and make conventional measurements of packing, such as the tap density of the powder, of limited quantitative value. This is especially true for many typical particulate fillers, which, due to their fine size and irregular shapes, exhibit particle agglomeration and do not pack well in the dry state. Such powders often pack considerably better when wetted by suitable liquids, especially if dispersants are present. These wetting and dispersing effects lie behind the pragmatic approach of using oil absorption procedures to determine an effective Pf. Originally introduced for application in putty and paint technology, oil absorption methods rely on titrating a sample of powder with an oil or other liquid, while continually rubbing and mixing the mass. End-points are readily detected in these procedures, often defined as the point at which a putty of a certain consistency is obtained. It is generally believed that at this point all the particles are dispersed and wetted, and all the gaps between the particles are just filled with liquid. In practice, only two liquids are used to any extent, linseed oil and phthalate esters. Both simple manual and instrumental methods are available for determining oil absorption values. It is important that standard procedures are closely followed if reproducible results are to be obtained and a number of standards have been issued. The simplest procedure is the spatula rub-out method, which is embodied in ASTM D281 [57]. A useful description of its application to particulate fillers has been given by Ferrigno [58].

29

Particulate-Filled Polymer Composites Oil-absorption results are usually quoted as cubic centimetres of oil per 100 grams of particulate at the end point. Simple mathematical treatment using oil and particle densities can then convert this to a volume fraction. As a guide, linseed oil absorptions for calcium carbonate of 20 cm3 100 g-1 and 50 cm3 100 g-1 correspond to filler volume fractions of 0.64 and 0.41, respectively. The attraction of the oil absorption methods lies in their simplicity and their crude approximation to polymer processing. Different liquids are also observed to give markedly different values and this is believed to be due to their respective abilities to wet out and disperse the particles. Thus in theory, by choosing a liquid of similar properties to the polymer in question, one should be able to improve the usefulness of the data, although this does not seem to have been greatly used to date. As mentioned previously, only two liquids are commonly used and, of these, linseed oil generally gives lower values suggesting that it wets and disperses particles better than phthalates. The phthalates are popular because they are widely used as plasticisers in PVC compounds. Liquid paraffins have been used as model systems for olefinic polymers, and squalane could be used as a model system for PP. A detailed study of the oil absorption test and its meaning in terms of particle packing has been reported by Huisman [59]. He found that the agglomerates present in the original powder are frequently not fully dispersed during the test, but may be compressed and reduced in porosity. The degree to which this occurs critically affects the packing and hence the oil-absorption value. Differences in the degree of compaction obtained go a long way to explaining variation in oil-absorption values between different operators, and between manual and instrumental methods, as well as between different liquids. While oil absorption serves a useful purpose as a guide to the effective value of Pf, great care must be taken in applying it too literally. In some polymer systems there may be a good match with the polymer wetting properties and the dispersion procedures. In others this may be very poor. Oil absorption values are widely used for assessing carbon black properties in the rubber industry. In this field, special procedures have been developed to attempt to reproduce the effect of rubber compounding conditions on particle morphology [22].

1.4.3 Particle Packing Theory Much effort has been devoted to studying the effect of particle size and shape on the packing properties of powders. Special attention has been given to identifying those factors that allow high packing fractions to be obtained. This subject is very relevant to the design of filler systems and the general principles, which often cause considerable confusion, are outlined here. For greater detail specialist publications such as that by German should be consulted [60].

30

General Principles Guiding Selection and Use of Particulate Materials The packing behaviour of particulate materials depends largely on their particle size, shape and surface characteristics. The behaviour of model systems with closely defined size and shape distributions is now well understood. Real particulate materials are much harder to treat, largely due to the difficulty in determining and describing their size and shape distributions accurately. Nevertheless, the principles derived for the model systems can be applied in a semi-quantitative way and appear to work reasonably well. The aims of particle packing theory are to predict the maximum volume fraction that can be obtained under a given set of circumstances and the structure of the particle assembly at this point. The first point to appreciate is that there are several ways of packing the same collection of particles, each one resulting in a different maximum volume fraction. The circumstances under which the particles are assembled will determine which of these structures is obtained in practice. In the extreme case consider a three-dimensional jigsaw puzzle. With care this can be assembled to give perfect packing with no free space. This is known as ordered packing. However, if the pieces are put in a container and shaken, it is extremely unlikely that they will ever discover this perfect packed structure. Nevertheless, it will be found that, providing there are enough particles, a certain packed density can be reproducibly obtained implying that some statistically balanced structure is obtainable. This is known as a random packing situation. More than one ordered and random packing fraction may be observed with many systems. Appreciation of the nature of ordered and random packing is important in understanding the application of packing theory. Ordered packing results in long-range structure as in a crystal lattice and, although it can sometimes be observed in suspensions of monodisperse spherical particles, is not of great importance in our context. As its name implies, random packing is a statistical process with no long-range structure and it is more relevant to particulate-filled composites. While some ordered packing configurations can be of very low density, ordered packing will always be capable of giving higher packing densities than random packing. Two classes of random packing are recognised, loose and dense random packing. Loose packing refers to the sort of packing obtained when particles are randomly assembled under conditions where they cannot easily move past each other, while in dense random packing, conditions are such that movement is possible. In powder technology terms, loose random packing corresponds with a pour density and dense random packing to a tapped density. The simplest case to consider first is the packing of smooth, regular, mono-sized particles. This has been well studied for smooth spherical particles, which can readily move past one another. With these particles a maximum ordered packing fraction of 0.74 has been established, although other, less dense packings are feasible. Random loose packing fraction is difficult to predict accurately but is about 0.60. Random dense packing is more readily predicted and is about 0.64.

31

Particulate-Filled Polymer Composites The theory has been extended quantitatively to a number of regular, non-spherical shapes. Many of these such as cubes, rectangles and plates can have ordered packing fractions of unity but most have less dense random packing than spheres, largely due to the difficulty of the particles moving with respect to one another to optimise packing. Cubic particles do, however, seem to have a slightly higher random packing density than spheres (about 0.75). Quantification of the packing behaviour of irregular particles is poorly developed at present but the enhanced interparticle friction soon causes the packing fraction to decrease as the particles depart from sphericity. Values of 0.50 are not uncommon for three-dimensional irregular particles. Very low packing fractions (