TOWARD REPLACEMENT of PETROLEUM PLASTICIZER by MODIFIED SOYBEAN OIL in RUBBERS a Dissertation Presented to the Graduate Faculty

Home , Precipitated silica

TOWARD REPLACEMENT OF PETROLEUM PLASTICIZER BY MODIFIED

SOYBEAN OIL IN RUBBERS

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Jiaxi Li

April, 2017

TOWARD REPLACEMENT OF PETROLEUM PLASTICIZER BY MODIFIED

SOYBEAN OIL IN RUBBERS

Jiaxi Li

Dissertation

Approved: Accepted:

Advisor Department Chair Dr. Avraam I. Isayev Dr. Sadhan C. Jana

Advisor Dean of the College Dr. Mark D. Soucek Dr. Eric J. Amis

Committee Member Dean of the Graduate School Dr. Younjin Min Dr. Chand Midha

Committee Member Date Dr. Yu Zhu

Committee Member Dr. Lu-Kwang Ju

ABSTRACT

Petroleum-based plasticizers are widely used to lower the viscosity of rubber compounds, increase processability and decrease cost. However, some popular petroleum-based plasticizers such as aromatic oils are considered as carcinogenic due to the high polycyclic aromatic hydrocarbon (PAH) content. Bio-based oil such as soybean oil (SO) has recently emerged as sustainable replacement for petroleum-based plasticizers in rubbers. However, such replacement often comes with tradeoffs and deficiency in certain properties. Modified SO includes norbornylized SO (NSO) and isoprene modified

SO (ISO) prepared through the reaction of SO with diene including dicyclopentadiene

(DCPD) or isoprene, is a green and low cost solution. After the reaction, the C=C double bonds in the SO can be converted into cycloaliphatic groups, which have higher reactivity.

Different modification levels can be achieved.

The present research studied the effect of modified SO in different rubber matrices including uncoupled and tin-coupled styrene-butadiene rubber (SBR), chloroprene rubber

(CR) and butyl rubber (IIR). Different filler systems including carbon black (CB), precipitated silica and their mixture (hybrid filler) and also different curing systems including sulfur, metal oxide and phenolic resin were used. Rheological, thermal, dynamic, mechanical, abrasion and aging properties of rubbers containing modified SO were compared with those containing virgin SO and a petroleum-based naphthenic oil

(NO). Results showed that SO and modified SO react with various additives such as

iii

curatives and silane coupling agent during the curing process and change various properties. SBR and CR containing NSO showed improved elongation at break, tensile strength, cure rate and aging resistance than those containing NO or SO. IIR containing

SO and NSO showed improvement of all the above properties except some decrease of tensile strength. However, all rubbers containing SO and NSO showed a decrease of modulus in compassion with those containing NO. Both SO and NSO provided rubbers with better thermal stability and lower glass transition temperature than NO. With proper adjustment to the curing recipe, tin-coupled SBR containing NSO was found to be suitable for manufacturing tires, since it showed desirable dynamic properties predicting better wet traction, lower rolling resistance and also better abrasion resistance simultaneously, compared with those containing NO. Uncoupled SBR containing NSO and ISO showed similar trend. The systematic study showed that NSO and ISO could be good replacements for petroleum-based rubber plasticizers in various rubber compounds along with improved performances and safety.

ACKNOWLEDGEMENTS

The author would like to express sincere gratitude to his advisor, Dr. Avraam I.

Isayev, for his excellent instructions on research and courses. The author also would like to thank Dr. Mark D. Soucek and his group for the long and successful collaboration. The author is grateful for the help from Dr. Keyuan Huang, Dr. Jing Zhong, Mr. Tian Liang,

Ms. Jieruo Liu, Mr. Xiang Gao, Ms. Yingmin Guo, Mrs. Xiaoping Zhang, Mr. Sandeep

Pole, Dr. Ivan Mangili and Dr. Shifeng Wang.

Finally, the author would like to express his thanks to his families for their lifelong support.

TABLE OF CONTENTS

Page

LIST OF FIGURES ...... xi

LIST OF TABLES ...... xvii

CHAPTER

I. INTRODUCTION ...... 1

II. LITERATURE SURVEY ...... 4

2.1 Petroleum rubber plasticizer ...... 4

2.2 Soybean oil and its modification...... 7

2.2.1 Soybean oil...... 7

2.2.2 Modification and other reactions of soybean oil...... 9

2.3 Rubbers ...... 12

2.3.1 Styrene-butadiene rubber ...... 12

2.3.2 Chloroprene rubber ...... 14

2.3.3 Butyl rubber ...... 16

2.4 Filler and filler network ...... 18

2.4.1 Carbon black ...... 18

2.4.2 Precipitated silica ...... 20

2.4.3 Thermodynamics of filler network formation ...... 22

2.4.4 Kinetics of filler network formation ...... 23

2.4.5 Flocculation of silica and silane coupling agents ...... 24

2.4.6 CB/silica hybrid filler in rubbers ...... 26

2.5 Curing systems ...... 27

2.5.1 Sulfur curing system ...... 27

2.5.2 Metal oxide curing system ...... 28

2.5.3 Phenolic resin curing system ...... 29

III. EXPERIMENTAL ...... 32

3.1. Preparation of modified soybean oils ...... 32

3.2 SBR system ...... 36

3.2.1 Materials ...... 36

3.2.2 Compounding and curing ...... 37

3.2.3 Characterizations...... 41

3.3 CR system ...... 44

3.3.1 Materials ...... 44

3.3.2 Compounding and curing ...... 45

3.3.3 Characterizations...... 47

3.4 Butyl rubber system ...... 49

3.4.1 Materials ...... 49

3.4.2 Compounding and curing ...... 50

3.4.3 Characterizations...... 52

IV. COMPARISON OF CB- AND SILICA-FILLED TIN-COUPLED SBR WITH

VARIOUS OILS ...... 54

4.1 Introduction ...... 54

vii

4.2 Results and Discussion ...... 55

4.2.1 Reaction of SO/NSO with Silane ...... 55

4.2.2 Gel Fraction, Crosslink Density and Bound Rubber ...... 57

4.2.3 Rheological Properties ...... 63

4.2.4 Thermal Properties ...... 67

4.2.5 Curing Behaviors ...... 71

4.2.6 Mechanical Properties ...... 74

4.2.7 DMA Test and Performance Predictors ...... 80

4.3 Conclusions ...... 84

V. COMPARISON OF CB/SILICA HYBRID FILLER-FILLED TIN-COUPLED

SBR WITH VARIOUS OILS...... 86

5.1 Introduction ...... 86

5.2 Results and discussion ...... 87

5.2.1 Bound rubber, gel fraction and crosslink density ...... 87

5.2.2 Rheological properties ...... 91

5.2.3 Thermal properties ...... 95

5.3.4 Mooney viscosity and curing behaviors ...... 99

5.3.5 Mechanical properties and aging resistance ...... 103

5.3.6 DMA and performance predictors ...... 111

5.3 Conclusions ...... 113

VI. CB-FILLED CR WITH VARIOUS OILS CURED BY METAL OXIDE ...... 117

6.1 Introduction ...... 117

6.2 Results and discussion ...... 118

viii

6.2.1 Rheological properties ...... 118

6.2.2 Gel fraction, crosslink density and bound rubber ...... 122

6.2.3 Thermal properties ...... 122

6.2.4 Curing behaviors ...... 126

6.2.5 Mechanical properties ...... 130

6.3 Conclusions ...... 136

VII. CB-FILLED IIR WITH VARIOUS OILS CURED BY PHENOLIC RESIN

...... 138

7.1 Introduction ...... 138

7.2 Results and discussion ...... 138

7.2.1 Gel fraction, crosslink density and bound rubber fraction ...... 138

7.2.2 Rheological properties ...... 144

7.2.3 Thermal properties ...... 147

7.2.4 Curing behaviors ...... 147

7.2.5 Mechanical properties and aging resistance ...... 151

7.2.6 Oil migration ...... 155

7.3 Conclusions ...... 155

VIII. COMPARISON OF CB-FILLED TIN-COUPLED AND UNCOUPLED

LINEAR SBR WITH VARIOUS OILS ...... 158

8.1 Introduction ...... 158

8.2 Results and discussion ...... 159

8.2.1 Gel fraction, crosslink density and bound rubber ...... 159

8.2.2 Rheological properties ...... 161

8.2.3 Thermal properties ...... 167

8.2.4 Curing behaviors ...... 170

8.2.5 Mechanical properties ...... 174

8.2.6 DMA and performance predictors ...... 179

8.3 Conclusion ...... 183

IX. SUMMARY ...... 184

REFERENCES ...... 188

LIST OF FIGURES

Figure Page

Figure 2.1. Chemical structures of aromatic (a), naphthenic (b) and paraffinic oil (c).

...... 5

Figure 2.2. Chemical structure of SO...... 7

Figure 2.3. Structure of SBR...... 12

Figure 2.4. Chemical structure of CR...... 14

Figure 2.5. Chemical structure of IIR...... 16

Figure 2.6. Surface structure of CB...... 19

Figure 2.7. Surface structure of precipitated silica...... 21

Figure 2.8. Structure of octylphenol formaldehyde resin...... 30

Figure 2.9. Reaction of phenolic curing resin with C=C double bonds...... 30

Figure 3.1 Preparation of NSO...... 34

Figure 3.2 Preparation of ISO...... 34

Figure 3.3. 1H NMR spectra of SO (a), 11NSO (b), and 11ISO (c)...... 35

Figure 4.1. FT-IR spectra of unreacted and reacted SO/silane (a), 5NSO/silane (b),

and 11NSO/silane (c)...... 56

Figure 4.2. Viscosity as a function of shear rate of unreacted and reacted SO/silane,

5NSO/silane, and 11NSO/silane at 35°C...... 58

Figure 4.3. Gel fraction (a), crosslink density (b) of various silica- and CB-filled SBR

vulcanizates and bound rubber fraction (c) of various silica- and CB-filled

SBR compounds...... 59

Figure 4.4. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) as a

function of frequency of various silica- and CB-filled SBR compounds at

90°C...... 64

Figure 4.5. Storage modulus as a function of strain amplitude of various silica- and

CB-filled SBR compounds at 90°C in logarithm (a) and linear (b) scale.

...... 66

Figure 4.6. TGA curves of various silica-filled (a), CB-filled (b) SBR compounds,

silica-filled (c) and CB-filled (d) SBR vulcanizates...... 68

Figure 4.7. DSC curves of various silica-filled (a), CB-filled (b) SBR compounds,

silica-filled (c) and CB-filled (d) SBR vulcanizates...... 69

Figure 4.8. Curing curves of various silica-filled (a) and CB-filled (b) SBR

compounds at 160°C...... 72

Figure 4.9. Stress-strain curves of various silica-filled (a) and CB-filled (b) SBR

vulcanizates...... 75

Figure 4.10. Abrasion loss of various silica- and CB-filled SBR vulcanizates...... 79

Figure 4.11. Storage modulus of silica-filled (a) and CB-filled (b) SBR vulcanizates,

loss modulus of silica-filled (c) and CB-filled (d) SBR vulcanizates, and

tan δ of silica-filled (e) and CB-filled (f) SBR vulcanizates as a function

of temperature...... 81

xii

Figure 4.12. Tan δ values of silica- and CB-filled SBR vulcanizates at 10 °C (a) and

60 °C (b)...... 83

Figure 5.1. Bound rubber fraction (a) of SBR compounds, gel fraction (b) and

crosslink density (c) of SBR vulcanizates...... 88

Figure 5.2. Relative crosslink density of various SBR/SO and SBR/NSO vulcanizates.

...... 90

Figure 5.3. Storage modulus vs. strain amplitude of various SBR compounds in linear

(a) and logarithm (b) scale...... 92

Figure 5.4. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various

SBR compounds...... 94

Figure 5.5. TGA curves of various hybrid filler-filled SBR compounds (a) and

vulcanizates (b)...... 97

Figure 5.6. DSC curves of various hybrid filler-filled SBR compounds (a) and

vulcanizates (b)...... 97

Figure 5.7. Curing curves of various hybrid filler-filled SBR compounds...... 100

Figure 5.8. Minimum (ML) and Maximum (MH) torque from curing curves of

various SBR compounds...... 101

Figure 5.9. Stress-strain curves of various hybrid filler-filled SBR vulcanizates before

(a) and after (b) aging...... 104

Figure 5.10. Comparison of M100 (a), M300 (b), elongation at break (c) and tensile

strength (d) of various CB-, silica- and CB/silica hybrid filler-filled SBR

vulcanizates...... 107

Figure 5.11. Mooney-Rivlin curves of various SBR vulcanizates...... 108

xiii

Figure 5.12. Abrasion loss of various hybrid filler-filled SBR vulcanizates...... 112

Figure 5.13. Storage modulus (a), loss modulus and tan δ of various hybrid filler-filled

SBR vulcanizates as a function of temperature...... 114

Figure 5.14. Tan δ values at 10 °C (a) and 60 °C (b) of various SBR vulcanizates

from DMA test...... 115

Figure 6.1. Storage (a), loss (b) moduli and tan δ (c) as a function of strain amplitude

of various CB-filled CR compounds at 90 °C...... 119

Figure 6.2. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) as a

function of frequency of various CB-filled CR compounds at 90 °C. . 121

Figure 6.3. Gel fraction (a), crosslink density (b) of various CB-filled CR vulcanizates

and bound rubber fraction (c) of various CB-filled CR compounds. .. 123

Figure 6.4. TGA curves of CR gum (a), various CB-filled CR compounds (b) and

vulcanizates (c)...... 124

Figure 6.5. DSC curves of various CB-filled CR compounds (a) and vulcanizates (b).

...... 127

Figure 6.6. Curing curves of various CB-filled CR compounds with (a) and without (b)

curatives...... 129

Figure 6.7. Stress-strain curves of various CB-filled CR vulcanizates before aging (a)

and after aging (b)...... 131

Figure 6.8. Abrasion loss of various CB-filled CR vulcanizates...... 135

Figure 7.1. Gel fraction (a) and crosslink density (b) of various CB-filled IIR

vulcanizates and bound rubber fraction of various IIR compounds (c).

...... 140

xiv

Figure 7.2. DSC curves of various oils (a) and oil/resin mixtures (b) at 160 °C. ... 142

Figure 7.3. Shear viscosity of various oil/resin mixtures at room temperature...... 143

Figure 7.4. Storage modulus as a function of strain amplitude of various CB-filled IIR

compounds at 90°C...... 145

Figure 7.5. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various

CB-filled IIR compounds and vulcanizates as a function of frequency at

90 °C...... 146

Figure 7.6. TGA curves of CB-filled IIR vulcanizates with 15 phr of oil (a) and 10 phr

of oil (b)...... 148

Figure 7.7. Curing curves of various CB-filled CR compounds...... 150

Figure 7.8. Stress-strain curves of various CB-filled IIR vulcanizates before aging (a)

and after aging (b)...... 152

Figure 7.9. FT-IR spectrum for phenolic resin, NO, SO, 11NSO, CB-filled IIR/NO,

IIR/SO and IIR/11NSO vulcanizates...... 156

Figure 8.1. Gel fraction (a), crosslink density (b) of various CB-filled SBR

vulcanizates, and bound rubber fraction (c) of various CB-filled SBR

compounds...... 160

Figure 8.2. Storage modulus as a function of strain amplitude of various CB-filled

linear and tin-coupled SBR containing different oils in linear (a) and

logarithm (b) scale...... 162

Figure 8.3. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various

unfilled linear and tin-coupled SBR gums and with different oils...... 164

Figure 8.4. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various

CB-filled linear and tin-coupled SBR compounds with different oils. 166

Figure 8.5. TGA curves of various CB-filled linear SBR vulcanizates (a) and

tin-coupled SBR vulcanizates (b)...... 171

Figure 8.6. Curing curves of various CB-filled linear SBR compounds (a) and

tin-coupled SBR compounds (b) with different oils...... 172

Figure 8.7. Stress-strain curves of various CB-filled linear SBR (a) and tin-coupled

SBR (b) vulcanizates...... 175

Figure 8.8. Abrasion loss of various CB-filled linear and tin-coupled SBR

vulcanizates...... 178

Figure 8.9. Storage modulus (a), loss modulus and tan δ of various CB-filled linear

SBR vulcanizates as a function of temperature...... 180

Figure 8.10. Tan δ values at 10 °C (a) and 60 °C (b) of various CB-filled linear and

tin-coupled SBR vulcanizates from DMA test...... 181

Figure 9.1. Relative crosslink density vs. modification level of various rubber

vulcanizates...... 186

xvi

LIST OF TABLES

Table Page

Table 2.1. Carbon structure of common rubber plasticizers...... 5

Table 3.1. Compounding recipes for various tin-coupled SBR...... 38

Table 3.2. Compounding procedures for various tin-coupled SBR compounds...... 38

Table 3.3. Compounding recipe for linear SBR...... 40

Table 3.4. Compounding procedure for CB-filled linear SBR...... 40

Table 3.5. Compounding recipe for CB-filled CR compounds...... 46

Table 3.6. Compounding procedure for CB-filled CR compounds...... 46

Table 3.7. Compounding recipes for CB-filled IIR compounds...... 51

Table 3.8. Compounding procedure for CB-filled IIR compounds...... 51

Table 4.1. Glass transition temperature of various silica- and CB-filled SBR

compounds and vulcanizates...... 70

Table 4.2 Curing time T95 of various CB-filled SBR compounds...... 72

Table 4.3. Tensile properties and hardness of various silica- and CB-filled SBR

vulcanizates...... 76

Table 5.1. Glass transition temperature of various SBR compounds and vulcanizates.

...... 98

Table 5.2. Mooney viscosity and curing characteristics of various hybrid filler-filled

SBR compounds...... 102

xvii

Table 5.3. Tensile properties, hardness and tear strength of various hybrid filler-filled

SBR vulcanizates before and after aging and their change in percentage.

...... 105

Table 6.1. Glass transition temperature of various CB-filled CR compounds and

vulcanizates and curing time of various CB-filled CR compounds...... 127

Table 6.2. Ts1, T95 and CRI of various CB-filled CR compounds...... 129

Table 6.3. Tensile properties, tear strength and hardness of various CB-filled CR

vulcanizates before and after aging and their relative change...... 132

Table 7.1. Glass transition temperatures of various CB-filled IIR vulcanizates. .... 149

Table 7.2. Ts1, T95 and CRI of various CB-filled IIR compounds...... 149

Table 7.3. Mechanical properties before and after aging of various CB-filled IIR

vulcanizates...... 153

Table 8.1. Glass transition temperature of CB-filled linear and tin-coupled SBR

compounds and vulcanizates...... 169

Table 8.2. Ts1, T95 and CRI of various CB-filled linear and tin-coupled SBR

compounds with different oils...... 173

Table 8.3. M100, M300, elongation at break, tensile strength and hardness of various

CB-filled uncoupled and tin-coupled SBR vulcanizates...... 176

xviii

CHAPTER I

INTRODUCTION

Various types of rubber plasticizers are used in rubber compounds in order to reduce viscosity, provide better processability, increase low temperature flexibilities and also decrease the cost. Most rubber plasticizers are petroleum-based, such as aromatic oil (AO), naphthenic oil (NO) and paraffinic oil (PO). AO is the most widely used plasticizer in tires due to its good compatibility with tire rubbers. However, Swedish National Chemicals

Inspectorate reported in 1994 that AO containing high polycyclic aromatic hydrocarbon

(PAH) level is highly carcinogenic.1 Therefore, safer rubber plasticizers are investigated and used in tires and other rubber products to replace high PAH AO. However, all these rubber plasticizers are derived from petroleum oil and they are not able to reduce the carbon footprint of rubber products. The petroleum resources on earth will possibly be depleted in the next century.2 Accordingly, polymer industry is facing a challenge of the petroleum reserves and green resources are desperately needed.

Bio-based oil such as soybean oil (SO) is a renewable, abundant and inexpensive oil resource. SO can be used as a plasticizer in rubber matrix. Previous studies already reported that the rubber product containing SO could have similar properties compared to the rubber product containing petroleum-based plasticizers.3-5 However, some tradeoffs were observed. It is very necessary to further investigate the application of SO in rubbers to obtain improved properties than those with petroleum-based plasticizers.

The C=C double bonds in the SO molecules can undergo a Diel-Alder reaction with conjugated dienes. One common diene resource is dicyclopentadiene (DCPD), it has been widely used to synthesis various polymers. When DCPD is heated above 150 °C, it undergoes a Retro-Diels-Alder reaction yielding two cyclopentadiene molecules. Another accessible and abundant diene resource is isoprene. About 800,000 tons of isoprene are produced each year, and about 95% are used to manufacture polyisoprene rubbers.6 In this research, SO is modified with DCPD or isoprene at different levels, a ring structure is introduced to the SO. Consequentially, the reactivity of C=C double bonds in modified

SO is increased.7 It is anticipated that the virgin and modified SO will react with curatives during the curing and change the properties of rubber vulcanizates.

The presented research is inspired by automotive rubber products such as tires, automotive rubber parts and curing bladders. Therefore, the objective of this research is to study the effect of SO modified via different methods with aim to use in different rubbers including tin-coupled SBR, uncoupled SBR, CR and IIR, different filler systems including carbon black (CB), precipitated silica and their hybrid, and also different curing systems including sulfur, metal oxides and phenolic resins. The effect of modified SO will be studied and compared with virgin SO and a petroleum-based plasticizer NO.

Rheological, thermal, dynamic, mechanical, abrasion and aging properties of various rubber compounds will be studied.

In this dissertation, Chapter II provides literature survey including petroleum-based rubber plasticizers, SO and its modification, different rubbers, filler systems and curing systems. Chapter III presents the experimental methods including materials, sample preparations and characterization methods. Chapter IV presents the study of CB- and

silica-filled tin-coupled SBR compounds and vulcanizates cured by sulfur containing various oils. Chapter V presents the study of CB/silica hybrid filler-filled tin-coupled

SBR compounds and vulcanizates cured by sulfur containing different oils. Chapter VI presents the study of CB-filled CR compounds and vulcanizates cured by metal oxide containing different oils. Chapter VII presents the study of CB-filled IIR compounds and vulcanizates cured by phenolic resin containing different oils. Chapter VIII presents the study of CB-filled uncoupled SBR containing different oils including SO modified by

DCPD and isoprene. Chapter IX provides a summary of this dissertation.

CHAPTER II

LITERATURE SURVEY

2.1 Petroleum rubber plasticizer

Most rubber plasticizers are hydrocarbon oils derived from petroleum, usually used as an extender or process aid. Petroleum rubber plasticizers can be roughly classified into paraffin, naphthenic and aromatic oils based on their carbon structure distribution, which includes paraffinic carbon (Cp), naphthenic carbon (Cn) and aromatic carbon (Ca). Carbon type composition of petroleum-based plasticizers can be determined according to ASTM

D2140. Table 2.1 shows the typical distribution of three types of petroleum plasticizers as given in ASTM D2140. These values are obtained from the correlation chart of viscosity-gravity constant (VGC) and refractivity intercept of the oil in ASTM D2140.

Figure 2.1 shows the chemical structures of aromatic (a), naphthenic (b) and paraffinic oil

(c).

Plasticizers have been used in rubber compounds for over century. The usage of petroleum plasticizer can reduce the compound viscosity and elasticity, decrease the glass transition temperature, improve the dispersion of fillers and also decrease overall cost. The wetting and lubricity of rubber compounds are influenced by the addition of plasticizers, and they are directly related to the processability of rubber compounds. The addition of plasticizers will have impact on the mixing and dispersion of ingredients in the compounds, and also subsequent processes such as extrusion, calendaring, sheeting, molding, etc. 4

Table 2.1. Carbon structure of common rubber plasticizers.

Type Cp% Cn% Ca% Paraffinic oil 60-75 20-35 0-10 Naphthenic oil 35-45 30-45 10-30 Aromatic oil 20-35 20-40 35-50

Figure 2.1. Chemical structures of aromatic (a), naphthenic (b) and paraffinic oil (c).

Different petroleum plasticizers have different compatibility with rubbers, therefore it is necessary to use a suitable plasticizer in a specific formulation. Too much oil or wrong compatibility will often lead to blooming of oils, and provide poor mechanical properties and mold fouling. AO such as distillate aromatic extracts (DAE) have been widely used in natural or synthetic rubbers and also tire compounds due to their good compatibility.

However, in 1994 a report by Swedish National Chemicals Inspectorate showed that polycyclic aromatic hydrocarbon (PAH), which is the main constituent of AO, is highly carcinogenic.3 According to European Union’s REACH Regulation (EC No. 1907/2009,

Annex XVII), extender oil used in tires must contain PAH below 3.0 wt% with the sum of individual PAH must be below 10 mg/kg and Benzo(a)pyrene lower than 1 mg/kg. High

PAH AO have been banned since 2010 in rubber products. In this case, various safe petroleum rubber plasticizers were introduced by manufacturers including mild extraction solvate (MES), treated distillate aromatic extract (TDAE), treated residual aromatic extract

(TRAE) and also heavy naphthenic oils. Delpech et al. studied these safe petroleum plasticizers and the results showed that these oils could be alternatives for DAE in rubber products.8 The treatment of these safe petroleum plasticizers increases the cost of the product, treated DAE is around $1000-1500 per ton.

Petroleum plasticizers are heavily used in tires. Each tire contains approximately 2 lbs of petroleum plasticizers. In 2014, the number of tire replacement and original equipment shipments was around 296 million in the US according to Rubber

Manufacturers Association’s 2014 Tire Industry Factbook. It is clearly seen that the manufacture of tires and other rubber products heavily relies on petroleum resource.

Accordingly, polymer industry is facing a huge challenge of the oil resources, therefore it is imperative to find alternatives.

2.2 Soybean oil and its modification

2.2.1 Soybean oil

SO is extracted from the seed of soybean, it is a renewable and inexpensive oil resource. SO is a triglyceride-based ester of glycerol and three fatty acids, typically contains around 23.4% of oleic, 53.3% of linoleic, 7.8% of linolenic, and around 15% of palmitic and stearic acids. Figure 2.2 shows the chemical structure of a SO molecule. The distribution of the fatty acids may vary from different soybeans and processing methods.9

According to the report from the United States Department of Agriculture, the domestic soybean production in 2015/16 is 106.8 million tons, and 313.3 million tons in the global market. About 11 million tons of soybean oil is produced in the U.S in 2015. and more than

50 million tons in the world.10 Abundant supply of the soybean leads to a low price of soybean oil about $700/ton.11

Figure 2.2. Chemical structure of SO.

SO could be an excellent resource to replace petroleum oils in various fields. Efforts of replacing the conventional petroleum oils with SO were made by many researchers. In rubber industry, SO can be added directly as a plasticizer in the rubber matrix. Dasgupta et al.3 investigated various bio-based oils including SO in natural rubber compounds. The effect of SO on rheological properties and processability was compared with aromatic oils.

Results showed that bio-based oil exhibited similar processing characteristics as aromatic oils. Flanigan et al. evaluated 10 phr of various bio-based oils including SO replacing part of the petroleum oil such as TDAE in the silica-filled SBR/NR 80/20 blend formulations.4

The study indicated that the product containing bio-based oils exhibited similar mechanical properties compared to the product with conventional petroleum oils. Wang et al. compared the properties of EPDM with various amount of transgenic soybean oil (TSO) and paraffin oil (PO).5 The results showed that the TSO had higher plasticization effect than that of PO. When the plasticizer amount was over 15 phr, the mechanical properties of the TSO-plasticized EPDM were decreased. Kundu et al.12-14 studied the coupling effect of vegetable oil between carbon black and rubber. Results showed that the monolayers of the oil on carbon black had a microplasticization effect and enhanced the interaction between carbon black and rubber when the oil content was below 3 phr. Petrović et al. studied the polymerized SO of different molecular weights, used as plasticizers, in 50/50 of NR/SBR compositions.15 It was found that the polymerized SO-extended rubber gave minimal differences in mechanical and dynamic properties compared to those of the naphthenic process oil-extended rubber.

Multiple patents were claimed on the use of SO in rubbers. Recker et al. claimed a rubber compound containing oils from non-fossil sources such as SO for tires, belts and

hoses.16 The included examples showed that the use of SO had minimum influence on the final properties. Thompson and Saintigny claimed a tire tread rubber composition containing SO.17 Experimental results showed that the use of SO in tire treads lead to a significant increase in rolling resistance. Wilson claimed a rubber composition used for footwear with non-petroleum based oils including SO.18 Various non-petroleum oils were used with content being less than 5 phr. The examples showed that the SO could replace petroleum-based oils in the silica-filled rubber compositions. Herberger et al.19 claimed a rubber composition with SO and SO/starch blends. Results showed that the samples containing SO and SO/starch exhibited a similar physical property compared to those with petroleum-based oil.

Based on previous studies, it is seen that the usage of virgin SO in rubber can provide comparable properties than those of the conventional petroleum plasticizers. However, some deficiencies in certain properties were observed. There is still huge space of improvement that needs to be further investigated.

2.2.2 Modification and other reactions of soybean oil

SO contains C=C double bonds and other function groups, it is very versatile for preparing other materials through different reactions. SO can be hydrogenated and used in food industries,20 or can be epoxidized and used in polyvinyl chloride (PVC) and coatings.21,22 SO can also be used to produce bio-fuel,23 polyols and polyurethanes,24 and other novel SO-based elastomers such as poly(epoxidized soybean oil-co-decamethylene diamine).25

SO can be thermally polymerized at 300 °C through a Diels-Alder addition, but with severe thermal degradation. SO can also be polymerized through sulfur vulcanization to prepare factice or vulcanized oil. Erhan and Kleiman studied the preparation of factice by various vegetable oils.26 Rapeseed and meadowfoam oils were able to produce brown factice with high quality. SO could produce low quality but less expensive factice. The low cost factice could benefit the rubber industry. Cowan et al.27 used SO and ethylene glycol to prepare polyester using heat, and then cured with curatives such as sulfur to obtain a rubber like material. Li and Larock investigated the copolymerization of SO, styrene and divinylbenzene.28 Dynamic properties of the copolymers were characterized. The mechanical properties and fracture surfaces were studied.29 Results showed that the copolymer exhibited a good damping property. The shape memory effect, and curing behavior were also studied.30,31 Li and Larock also found that the copolymer made from

SO with styrene and dicyclopentadiene had good mechanical properties and thermal stability.32 Kammann Jr. and Phillips discovered that the vulcanized SO could also be used as lubricant additives.33 Extreme pressure, friction and wear performances were tested and the vulcanized SO exhibited potential to replace conventional additives.

Xia et al. investigated bio-based thermosets through the ring-opening metathesis polymerization (ROMP) of norbornyl-functionalized fatty alcohols derived from SO and other commercialized bi-based oils.34 The mechanical and thermal properties of the obtained thermosets were studied. It was found that the material properties were directly influenced by the viscosity and structure of oil monomers. Valverde et al.35 studied

SO-based rubbery thermosets synthesized by cationic copolymerization of conjugated SO,

styrene, and 1,5-hexadiene or isoprene. The structure-property relationship was investigated.

Epoxidized SO (ESO) is produced by the epoxidation of SO. It is an excellent HCl scavenger for PVC at high temperature, and also a good plasticizer. The epoxidation process can be achieved by using peroxide or peroxyacetic acid. Yang et al.36 Investigated the structure-property relationship of ESO with various levels of epoxidation. The thermal stability, melting point and molecular weight were characterized and correlated to the epoxidation level. Sahakaro and Beraheng investigated the use of ESO in CB-filled SBR compounds.37 Results showed that compared with DAE, ESO provided lower state of cure and inferior tensile strength, modulus and abrasion resistance. ESO can also be used in rubber-modified asphalt. Yin et al.38 investigated asphalt modified by crumb rubber containing ESO. The softening point, penetration, thermal stability, ductility, viscosity and anti-aging mechanism of the modified asphalt were studied. The results showed that the asphalt modified by crumb rubber containing ESO showed higher thermal stability, lower temperature ductility and better anti-aging performance than the control sample. Wang et al. studied a novel ESO-based elastomer.25 Poly(epoxidized soybean oil-co-decamethylene diamine) (PESD) was synthesized by ring-opening polymerization from ESO and decamethylene diamine at various ratios. The ESO-based elastomer exhibited excellent damping property, low water absorption, and low degradation rate in phosphate buffer solution.

2.3 Rubbers

2.3.1 Styrene-butadiene rubber

Styrene-butadiene rubber, also known as SBR, is a synthetic rubber being a random copolymer of styrene and butadiene.39 Figure 2.3 shows the structure of SBR. SBR was first developed by German scientists prior to World War II. The mass production of SBR in the US began during the war time to replace natural rubber in various rubber products, such as tires, wires, cables, hoses and gaskets, etc. SBR is the most widely used rubber, it constitutes about 40% of the total synthetic rubber consumed in the US.40 SBR has good wear resistance and traction performance, it is the primary rubber for passenger car tire tread compounds.

Figure 2.3. Structure of SBR.

SBR can be manufactured by emulsion or solution polymerization. For emulsion polymerized SBR (ESBR), the styrene and butadiene monomers are emulsified in water using soaps as emulsifier. At around 60% of conversion rate, stop agents are added. The emulsion is then coagulated, and unreacted monomers are removed. The heat of polymerization can be dissipated by the emulsion water and also cooling water. High

(~50 °C) or low (~5 °C) temperatures can be used to produce “Hot SBR” and “Cold SBR”.

Cold SBR provide better abrasion resistance and improved dynamic properties.41 12

Solution polymerized SBR (SSBR) has higher purity due to the absence of soap and other emulsifiers. Typically, alkyl-lithium-based catalysts are used, and the polymerization takes place in hydrocarbon solvents. SSBR has a narrower molecular weight distribution and less branching than ESBR. SSBR can also have various styrene content and butadiene configuration to meet different needs. In SSBR, a higher vinyl content will decrease the hysteresis and cure rate. SSBR has been widely used in tires for decades due to its lower rolling resistance and better wet traction than ESBR.

It is known that common non-functionalized SSBR has a linear molecular structure.42

In order to increase the processability, branched SSBR was introduced since the 1980s, and the technique of modifying the chain ends of SBR was developed.43 Among various grades of SSBR, tin-coupled SBR has a unique star-shape molecular structure with tin-butadienyl bonds, and shows lower hysteresis than uncoupled linear SBR. The tin-carbon bond can break up during compounding and generate chemical bonding with organic groups at the surface of CB.44 Therefore, the rubber-CB interaction is improved. Previous studies reported that the tin-coupled SBR exhibited better filler dispersion and rubber-filler interaction than uncoupled SBR.44-47 The tin-coupled SBR has a good combination of low rolling resistance and high wet traction.

Regardless of different types of SBR, it must be reinforced to provide satisfied mechanical properties. Various reinforcing fillers such as CB and silica have been widely used. Other ingredients including plasticizers, antioxidants, curatives such as sulfur, accelerators and activators have been intensively used in a typical SBR compound. More discussion on fillers and curatives will be given in Section 2.4 and 2.5.

2.3.2 Chloroprene rubber

Chloroprene rubber or CR, also known as “Neoprene”, was first introduced by

DuPont in 1931. CR is polymerized mainly from chloroprene monomers using emulsion polymerization. Figure 2.4 shows the chemical structure of CR. Various modifiers and stabilizers are used during the polymerization of CR to meet different needs. Crystallinity is an important property of CR, the degree of trans configuration and copolymerization with different monomers can adjust the crystallinity to meet various needs. CR has outstanding physical properties with resistance to hydrocarbon oils, heat, ozone and weather. It is also more flame retardant than most general purpose hydrocarbon-based elastomers. CR has a variety of applications such as automotive parts, construction parts, cable jackets, coated fabrics, conveyor belts, hoses, etc.

Figure 2.4. Chemical structure of CR.

According to DuPont, CR can be classified into three types. G types of CR are manufactured by the copolymerization of chloroprene and sulfur, stabilized with thiuram disulfide to control the molecular weight.48 Most G type CR have limited shelf life due to the crystallization. They are ideal for dynamic applications such as belts. W types of CR are manufactured by the copolymerization of chloroprene and alkyl mercaptans.49 They have longer shelf life than G type and they have excellent heat resistance and compression set. T types of CR are similar to the W types. A higher gel fraction is introduced to facilitate

material processability such as extrusion and calendaring. In addition to these three types,

CR also comes with various grades for adhesives.

Although it was reported by previous study that the CR could be crosslinked by sulfur and also thermovulcanization,50 the conventional sulfur curing systems for diene rubbers can’t provide satisfactory performance of the vulcanizates. This is due to the electronegative chlorine atoms, which can inhibit the electrophilic substitution that is common for hydrocarbon-based diene rubbers. Metal oxides such as zinc oxide (ZnO) and magnesium oxide (MgO) are widely used for curing CR, and they are sufficient for curing

G types CR in the absence of other organic accelerators. Differently, W and T type of CR need organic accelerators for curing. However, ethylene thiourea (ETU) is widely used to achieve a better performance and a faster curing of CR compounds. The curing mechanism of CR compounds were studied by many researchers.51-54 It is generally accepted that the major curing site of CR is the tertiary allylic chlorine atom. The ZnO can generate crosslinks between two chloroprene units via a cationic mechanism, and the ETU can crosslink the CR alone and also in combination with ZnO.

CR can also be crosslinked with other ingredients such as amino acid derivatives,55 thiophosphoryl disulfides56 and fillers.57,58 CR can also be blended with other elastomers to improve properties such as low temperature flexibility and mechanical properties. Zheng et al. studied the blend of CR with butadiene rubber (BR).59 It was found that the addition of

BR increases the low temperature resistance and electrical insulation but decreases the mechanical properties of the blend. Das et al. investigated the physical and electrical properties of CR and natural rubber (NR) blends.60 It was observed that the blend exhibited significant improvement on curing behavior and physical property.

2.3.3 Butyl rubber

Butyl rubber or isobutylene isoprene rubber, abbreviated as IIR, is a copolymer of isobutylene and generally less than 3% of isoprene. Figure 2.5 shows the chemical structure of IIR. It was first commercialized in 1942. IIR is typically prepared via cationic

61 polymerization with AlCl3 as a catalyst and methyl chloride as a solvent. The polymerization takes place at -100 °C within a second.

Figure 2.5. Chemical structure of IIR.

Butyl rubber has low gas permeability, good thermal and oxidative stability, and excellent moisture and chemical resistance. Van Amerongen62 investigated the gas permeability of different rubbers. Results showed that IIR vulcanizates exhibit 8 times lower air permeability than natural rubber. Therefore, it is widely used in tire inner tubes and inner liners. IIR also exhibits excellent thermal stability due to its mainly saturated backbone. However, IIR cured by sulfur will soften at elevated temperatures caused by the breakage of sulfur crosslinks. Certain curing systems such as phenolic resin can provide outstanding heat resistant crosslinks. This has found successful applications in manufacturing of tire-curing bladders with life of 300-700 curing cycles. The saturated structure of IIR gives it moisture and chemical resistance. While IIR can be extensively swollen by hydrocarbon solvents and oils, they are only slightly affected by other polar

liquids or oxygenated solvents.63 Thus IIR can be used in electrical insulation or rubber sheets exposed to weather.

Another important application for IIR is vibration damping. The two methyl side groups on the isobutylene results in delayed elastic response to deformation.64 The damping and shock absorption of IIR give it wide application in automotive parts.

Halogenated IIR was developed in the 1950s, they have similar structures to that of the butyl rubber but with faster cure rate and better co-curing compatibility with other general purpose rubbers, therefore they are commonly used in tubeless tire inner liners.65-67

For tire sidewalls, ozone and crack growth resistance are crucial. Blends of halogenated

IIR with natural rubber and ethylene propylene terpolymers can yield excellent side walls in tires.68 Using up to 30 phr of halogenated IIR in tire treads can improve traction and lower the rolling resistance.

Butyl rubber can be cured using a variety of cure systems such as sulfur,69 quinone,70 phenolic resin71 or metal oxide72 curing system via reactions involving either allylic hydrogen or allylic halogen in butyl rubber.

Typically, IIR compounds require reinforcing fillers such as CB to achieve excellent mechanical properties of vulcanizates. Numerous studies have investigated the effect of

CB in IIR compounds.73,74 Feng and Isayev also thoroughly studied the recycling of IIR vulcanizates.75,76 It was well demonstrated that the ultrasonic devulcanization technology was very effective to decrosslink IIR vulcanizates and the recycled materials could be reused in new IIR products.

2.4 Filler and filler network

2.4.1 Carbon black

In the rubber industry, fillers are widely used to enhance the performance while reducing the cost of the rubber products. Each year, about 2.5 million tons of different fillers are consumed. CB comprises about 75% of the market, silica and silicates comprise about 4%, others such as mineral fillers comprise about 21%.77 Among various fillers used in rubber, CB and silica are the most important ones. CB was used by human beings since antiquity until it was discovered in the early 1900s that it could improve the properties of the rubbers.78 The CB is manufactured via various methods to obtain furnace blacks, channel blacks, and thermal blacks. Furnace black is one of the most widely used types, more than 95% of the CB used for rubber are produced by this method. Hydrocarbon oil is usually used as a feedstock and injected into a furnace which is heated up to 1000-2000 °C, the incomplete combustion will yield very fine CB particles. The graphitic layers of carbon atoms form first, then they will combine and grow into spherical particles while spinning in the gas stream inside the furnace. The particles will form structural aggregates. Vital properties such as surface area and structure can be controlled via adjusting the processing parameters such as temperature, flow rates, feedstock injection method, additives, cooling methods, etc.77 The surface of the CB is covered with various groups such as aromatic rings, phenol, carboxyl, quinone, lactone, ketone, lactol, and pyrone groups.79 Figure 2.6 shows the surface structure of CB.

Figure 2.6. Surface structure of CB.

Classification of rubber-grade CB is described in ASTM D1765. An example is CB

N330. Typically, the first letter in the nomenclature system for rubber grade CB indicates the influence of CB on the cure rate of a typical rubber compound. Letter “N” here stands for “Normal” curing rate which is for furnace black without special modification to alter the influence on cure rate. Letter “S” is also used for furnace or channel blacks that have been modified to reduce the cure rate. The second character is a number that indicates the average surface area measured using N2. Here the number 3 indicates that the N2 surface area is around 70-79 m2/g according to ASTM D1765-16. The third and fourth numbers are usually arbitrarily assigned. The characterization of CB is extremely important for the selection of CB in rubbers. Specific surface area and structure are two important properties.

Specific surface area is usually measured using Brunaner-Emmet-Teller (BET) method80 or using hexadecyltrimethylammonium bromide (CTAB) method according to ASTM

D3765-99. Structure of CB indicates the void space between CB aggregates or agglomerates. ASTM D2414-97 and D3493-97 described the tests using dibutyl-phthalate.

(DBP) to measure the structure. The specific surface area and structure are directly related to the rubber-CB interaction, filler dispersion and dynamic properties. Wang did a systematic study on the effect of CB with different specific surface area and structure.81 It was found that the specific area can affect various properties of CB-filled rubbers. At the same level of filler loading, the higher the surface area, the higher the reinforcement. A higher surface area also increased the difficulty of dispersion and created a stronger tendency of filler networking.82-84 A higher structure can facilitate a higher bound rubber fraction, a smaller aggregate size and ease in processing.85-87

2.4.2 Precipitated silica

Silica including precipitated silica, fumed silica and surface treated silica is another important filler. Precipitated silica is by far the most commonly used silica, it is a form of synthetic amorphous silicon dioxide derived from quartz sand. Precipitated silica was first used in the 1950s in shoe soles and then was used in tires in the 1970s.88 It is produced by adding different types of acid to a sodium silicate solution to form a precipitate of SiO2 particle aggregates. The particles of precipitated silica are filtered, washed and dried. Spin flash drying, rotary drum drying and spray drying can be used. Various particle size and specific surface area (40 to > 200 m2/g) can be controlled. The term “silica” in this study stands for precipitated silica.

As described above, the surface of CB is relatively non-polar, which is very compatible with hydrocarbon-based rubbers. However, the surface of silica particles is covered by siloxane and various types of silanol groups including isolated, germinal and vicinal silanol groups.89 Figure 2.7 shows the surface structure of silica. Thus the surface of silica is more polar than that of the CB, and less compatible with most hydrocarbon-based rubbers. Besides weaker interaction with rubbers, silica particles can also generate strong hydrogen bonding due to the presence of silanol groups, leading to a strong filler-filler interaction. Precipitated silica also can absorb moisture at the surface. Moisture levels are typically 4%-7%. Kim and VanderKooi studied the effect of moisture content on the properties of CB/silica/SBR compounds.90 Results showed that the addition of proper amount of water molecules to the surface of silica improved the silanization reaction via improved hydrolysis, increasing the level of crosslinking. However, too much moisture will form a shell outside the silica particle and reduces the chemical activity during the silanization reaction. Moisture can be driven off during the compounding, thus the mixing temperature can influence the final properties of silica-filled rubber compounds.91

Figure 2.7. Surface structure of precipitated silica.

2.4.3 Thermodynamics of filler network formation

The filler agglomerates need to be broken down and dispersed into rubber matrix during the compounding. Rubber fillers are usually used at high loadings and filler network is very likely to be formed. The formation of filler network in the rubber matrix is determined by the attractive force between filler particles or aggregates, the interaction between rubber molecules, as well as the interaction between filler and rubber.92 The attractive forces are generally expressed by the potential of intermolecular interactions.

These interactions are classified into two groups, a dispersive or non-polar interaction and a specific or polar interaction, including dipole–dipole interaction, induced dipole–dipole interaction, hydrogen bonding, and acid–base interaction.81 The following equation presents the filler network formation:

1 1 1 1 푑 2 푑 2 2 푝 2 푝 2 2 ℎ ℎ ℎ 푎푏 푎푏 푎푏 ΔW = 2[(γ푓 ) − (γ푝) ] + 2[(γ푓) − (γ푝) ] + 2[W푓 + W푝 − 2W푓푝] + 2[W푓 + W푝 − 2W푓푝 ]

d Here ΔW is the total change in adhesive energy in the filler agglomeration, γf and

p d p γf are the dispersive and polar components of filler surface energy, γp and γp are the dispersive and polar components of polymer surface energy, Wh and Wab are the work of

h ab hydrogen bonding and acid-base interaction between filler-filler (Wf , Wf ),

h ab h ab polymer-polymer (Wp , Wp ) and filler-polymer (Wfp , Wfp ). When the surface energy of fillers and polymers are the same, and the work of hydrogen bonding and acid-base interactions between different phases are the same, ΔW=0. Which means a zero attractive potential between the fillers, the filler network will be stable. This equation suggests that the driving force of filler networking is the difference in surface energies, both in the intensity and nature, between filler and polymer.81 The larger the difference in

the surface energies and the lower the interaction between filler and polymer, the higher is the tendency for the filler networking in the polymer.

2.4.4 Kinetics of filler network formation

Due to the different structure of rubbers and fillers, there is always a surface energy difference in a filled rubber compound, even for a well dispersed system. The filler aggregates will always flocculate during storage or vulcanization of the filled compound.

Besides the attractive potential, the filler flocculation process is also determined by the diffusion of filler aggregates due to Brownian motion to form thermodynamically stable agglomerates. The diffusion constant ∆ is related to the temperature T, and resistance coefficient f:

∆ = kT/f

Here k is the Boltzmann constant. The resistance coefficient f is determined by the medium viscosity, η, and the size and shape of the particle. For a spherical particle with a radius a, according to the Stokes law, the resistance coefficient f is:

f = 6πηa so that the diffusion constant is:

∆ = kT /6πηa

This equation indicates that at certain T, the diffusion rate of filler aggregates is controlled by the rubber matrix viscosity and the size of the aggregates.

Based on the thermodynamics and kinetics of filler networking described above, there are some approaches to achieve less filler network in rubber compounds. On the thermodynamic approaches, one can reduce the surface energy difference between rubber

and filler; increase rubber-filler interaction and compatibility; or blend fillers with different surface characteristics. On the kinetic approaches, one can improve the initial filler dispersion or smaller agglomerate size in the compound; increase the average particle-particle distance by changing filler morphology; increase bound rubber to increase the effective aggregate size and also increase the viscosity of rubber matrix. Also, an earlier and faster cure for the compound can lock filler aggregates in place the fillers they start to flocculate.

2.4.5 Flocculation of silica and silane coupling agents

Flocculation has been widely observed in both CB-filled and silica-filled rubber compounds. Fillers have tendency to re-agglomerate to form larger structure of agglomerates in rubber compounds. Böhm et al. studied the flocculation in CB-filled polybutadiene rubbers (BR).93 The research indicated that the flocculation was observed during the storage and also curing of the CB-filled BR compounds. The rate of flocculation was affected by the loading and grades of CB, rubber molecular structure, compounding process and annealing temperature. Lin et al. investigated the flocculation in both CB- and silica-filled rubber compounds.94,95 It was shown that the flocculation was influenced by the polymer-filler interaction at various stages of processing including mixing, annealing, storage, and curing. Particularly, a strong flocculation was seen in the silica-filled rubber compounds, and it had negative effect of the dispersion of fillers. Wang and his coworkers investigated the surface difference between silica and CB. 81, 96-98 Results showed that silica had a lower dispersive component and a higher polar component of surface energy than CB and common hydrocarbon rubbers, indicating a less rubber-filler interaction for silica than

CB. In addition, the hydrogen bonding between the silanol groups at the surface of the silica also contributed to the filler flocculation. Mihara et al. studied the flocculation in silica-filled SBR/BR blends.99 The activation energy and rate constant of the flocculation process were calculated. Results showed that the silica flocculation was a pure physical process. It was known that the existence of the silanol groups at the surface of the silica could not only lead to a poor dispersion, but also inferior curing in sulfur-accelerator curing systems.100,101 It is believed that commonly used CB has limited side effects on the curing of the rubber, sometimes can even speed up the curing.102 However, the silica has a different surface chemistry and therefore can have a different effect on the curing of the silica-filled rubber compounds. The silanol groups can attach to activators or amine accelerators, thus decreasing cure. Laning et al. found that the silanol groups could react with the zinc oxide (ZnO).103 Reuvekamp et al. studied the effect of ZnO on the silane coupling effect in a silica-filled SBR/BR system.104 The results showed that adding ZnO at early stage of mixing could lead to a less coupling effect. Previous studies also found that the diphenyl guanidine (DPG) could act as a secondary accelerator and also a catalyst of the silanization reaction.99, 105, 106

With the use of silane coupling agents or other chemicals such as polyethylene glycol, the silanol groups can be tied up and also increase the polymer-filler interaction. Numerous researchers investigated various types of silane coupling agents in silica-filled rubber compounds.105, 107-109 It was found that the silane coupling agents could bond with the silanol groups of the silica, and the sulfur containing groups could react with rubbers, forming a strong rubber-filler interaction. At the same time, the dispersion of silica was improved and the flocculation was suppressed. Sarkawi et al. investigated the effect of

processing temperature on the properties of the silica-filled natural rubber compounds.91 It was shown that dumping temperature had obvious effect on the vulcanizate properties. An elevated dumping temperature could benefit the silanization reaction and improve the filler dispersion. The optimum dumping temperature was found in the range of 140 to 160 °C.

The usage of precipitated silica in tire treads provides a lower rolling resistance at equal wear resistance and wet traction compared to the CB.81, 110-112 Therefore, the application of silica in tire treads is being well accepted.

2.4.6 CB/silica hybrid filler in rubbers

The blending of CB and silica is a common method to achieve balanced performance and costs for tires.113 The use of CB/silica hybrid filler in rubber has been studied intensively. Wang pointed out that the filler networking was suppressed by using CB/silica hybrid filler, both thermodynamically and kinetically.81 Liu et al. studied the morphology of SBR filled with CB/silica hybrid filler.114 Results showed that hybrid filler had better dispersion in SBR than either type of filler. Wang and his coworker found that the CB/silica hybrid filler-filled SBR vulcanizates exhibited balanced rolling resistance, wet traction and abrasion resistance.115 Rattanasom et al. studied the effect of blending ratio of CB/silica hybrid filler on mechanical properties in natural rubber.116 It was found that the best overall mechanical properties were achieved when the blending ratio of CB/silica was between

20/30 to 30/20. Dong et al. studied the fracture and fatigue of CB/silica hybrid filler-filled natural rubber composites.117 Results showed that with increasing silica content, natural rubber vulcanizates had better fatigue resistance. Wang et al. also introduced the carbon– silica dual-phase filler (CSDPF).118 It combined the advantages of both types of fillers,

leading to desired mechanical and dynamic properties.119 In general, the use of CB/silica hybrid filler will lower the filler-filler interaction and improve dispersion of fillers.

2.5 Curing systems

2.5.1 Sulfur curing system

Sulfur is the most widely used curing agent for rubbers and comprises more than 90% of all vulcanizations in the rubber industry. It was first discovered by Goodyear in 1839 in the US. Sulfur curing system is easy to process, low cost and easily available in the market.

It can adapt with various other ingredients to give diverse properties for rubber vulcanizates.

Sulfur is in a cyclic octatomic form, or S8, in its crystalline state. It is thermally stable with a melting point of 115.2 °C. The ring opening activation energy is reported as 270 kJ/mol.120 High temperature and long time period are required to open the sulfur ring. Thus the vulcanization of rubber by sulfur alone is slow and inefficient. In the 1910s, aniline was found by Oenslanger to be able to increase the speed of sulfur curing. Since then, a variety of rubber curing chemicals have been discovered. A typical sulfur curing system may include sulfur or sulfur donor, accelerators, activators and others. The mechanism of accelerated sulfur curing system is still in dispute. Radical and ionic process are two well accepted mechanisms. The reactive sites are usually allylic hydrogen atoms on the C=C carbon double bond groups. Activators and accelerator can form a complex that can open the sulfur ring and increase its reactivity. Then the sulfur links can subtract allylic hydrogen atoms and crosslink the rubbers.121

It has been known that the sulfur crosslinks can be mono-, di- or polysulfide crosslinks. Depending on the numbers of sulfur atoms between crosslinks, there are three types of sulfur curing systems: conventional vulcanization (CV), semi-efficient vulcanization (SEV) and efficient vulcanization (EV). With the increasing ratio of accelerator/sulfur, more mono- and disulfidic crosslinks are formed instead of polysulfidic crosslinks. A higher portion of polysulfidic crosslinks can provide better flexibility but also lability to heat. A typical EV system may have 80% of the crosslinks as monosulfidic and a

CV system may have 95% of crosslinks as polysulfidic.122 Most general purpose diene-based rubber can be vulcanized using sulfur.

2.5.2 Metal oxide curing system

Metal oxide curing system is commonly used for halogenated elastomers such as CR, halogenated IIR and chlorosulfonated polyethylene. Three most used metal oxides are ZnO,

MgO and PbO. ZnO and MgO are widely used to cure CR. ZnO and MgO both can cure

CR alone, ZnO can provide a faster cure and strong scorch tendency, yet MgO will provide a slower cure. Therefore, they are usually used together at a weight ratio of 5:4. In this case,

ZnO can act as a major curative and provide excellent thermal resistance for the CR vulcanizates. MgO can increase the scorch safety and also scavenge the HCl released during the curing. In rubber products that require water resistance, PbO can be used instead of ZnO and MgO.

The curing mechanism of CR compounds by metal oxides were studied by many researchers.51-54 It was widely accepted that the major curing site of CR is the tertiary allylic chlorine atom. G type of CR typically does not require an organic accelerator. W and

T types of CR usually require organic accelerators such as sulfenamides and thiazoles. In addition, ETU is also widely used to increase the cure rate. The ZnO can generate crosslinks between two chloroprene units via a cationic mechanism, and the ETU can crosslink the CR alone and also in combination with ZnO. Metal oxides are also highly effective in curing rubbers containing pendant carboxyl groups.

2.5.3 Phenolic resin curing system

Phenolic resins such as octylphenol formaldehyde resin are widely used to cure butyl rubbers. Structure of octylphenol formaldehyde resin is shown in Figure 2.8. The reactive phenolic resin can form a cyclic ether via the reaction between the phenol-methylol groups and the C=C bond of the butyl rubber backbone.123 Reaction of phenolic resin with C=C double bonds is shown in Figure 2.9. Typically, phenolic resin curing system requires ZnO and a halogen donor such as polychloroprene or stannous chloride (SnCl2). It was proved that the zinc halide reacts with the terminal hydroxyl groups and increase the reactivity of the phenolic resin. Brominated octylphenol formaldehyde resin has a similar structure but has a bromine atom in place of the terminal hydroxyl group and it does not require a halogen donor. It can provide a faster cure to the butyl rubber.

Phenolic resin curing system will generate C-C crosslinks which are relatively stable at elevated temperatures, therefore, it is widely used in curing bladders124 and aerospace materials.125

Figure 2.8. Structure of octylphenol formaldehyde resin.

Figure 2.9. Reaction of phenolic curing resin with C=C double bonds.

The use of petroleum-based plasticizers in rubbers has been intensively studied by previous researchers. The current trend is to replace petroleum-based plasticizers by renewable bio-based plasticizers. A lot of efforts have been made to use bio-based oils such as SO in rubbers, especially in tires. However, there are limited studies on the use of modified SO in rubber compounds. Therefore, it is necessary to investigate the effect of modified SO in various rubbers, not only in tire rubbers, to further expand the application of bio-based plasticizers in the rubber industry.

CHAPTER III

EXPERIMENTAL

3.1. Preparation of modified soybean oils

A technical grade RBD Soybean Oil by Cargill Industrial Oils & Lubricants

(Minneapolis, MN) was used and modified using two methods.

The first modification was through the reaction of SO with DCPD. Upon heating,

DCPD was cracked into two cyclopentadienes having a Diel-Alder reaction with the C=C double bonds on the fatty acid chains of SO. After the reaction, the C=C double bonds on the fatty acid chains were converted into norbornyl groups. The norbornyl groups have a ring structure and thus the C=C double bonds on the norbornyl group are more reactive than those on the fatty acid chains.7 This product was named as norbornylized soybean oil

(NSO).

The preparation of NSO is shown in Figure 3.1. 300 g of SO, DCPD (14, 26 or 155 g), and 4 wt% butylated hydroxytoluene (BHT, based on the weight of DCPD) were charged into a flask and stirred until all the BHT dissolved. The mixture then was transferred to a

1000 ml Parr reactor and heated to 240 °C. The pressure was increased with the increasing temperature and reached a stable value between 0.14 to 0.21 MPa. The temperature was maintained until the pressure dropped to atmosphere pressure, and then the reaction mixture was slowly cooled to room temperature. Hexane was used to dilute the mixture, and the mixture was stirred for 0.5 h. This solution was filtered, and the solvent was 32

removed with a rotary evaporator at 55-60 °C. About 5%, 11% and 33% of the C=C bonds in the virgin SO were converted into norbornyl groups. They were named as 5NSO,

11NSO and 33NSO, respectively. Detailed characterizations of NSO can be found in previous study.126

The second modification was through the reaction of SO and isoprene. It is also a typical Diel-Alder reaction. The isoprene modified soybean oil was named as ISO. Figure

3.2 shows the preparation of the ISO. 300 g of SO, 12 g or 23 g of isoprene, 4 wt% of BHT

(based on the weight of SO) were used. The preparation procedure and parameters were similar to that of the NSO. Similarly, the methyl cyclopentene group affords higher reactivity of the C=C bonds than those on the fatty acid chains. It is expected that the ISO will have similar property and effect in rubber compounds as NSO. An advantage of isoprene to DCPD is its lower boiling point, thus it is easier to remove the residual unreacted isoprene. Two different levels of ISO were used, 5% (5ISO) and 11% (11ISO).

Their effects were compared with 5NSO and 11NSO in rubber compounds.

Figure 3.3 shows the 1H NMR spectra of SO, 11NSO and 11ISO. In Figure 3.3 (a), the resonance at δ = 5.34 ppm (H9 and H10) is observed. In 11NSO (b) and 11ISO (c), the intensity of this resonance is decreased and new resonances at δ = 6.08-5.95 ppm (H2 and

H3), δ = 2.50 ppm (H1 and H4), δ = 1.82 ppm (H5 and H6), δ = 1.43 ppm (H8) and δ =

1.05-1.20 ppm (H7syn and H7anti) are observed.

The resonance at δ = 5.4 ppm is attributed to the olefinic C=C double bonds from the fatty acid chain of SO. The 11NSO spectrum (b) and the 11ISO spectrum (c) afforded new resonances at δ = 6.08-5.95 ppm and δ = 1.43 ppm, respectively, due to the formation of the six-member ring structure from the Diels-Alder reaction.

Figure 3.1 Preparation of NSO.

Figure 3.2 Preparation of ISO.

Figure 3.3. 1H NMR spectra of SO (a), 11NSO (b), and 11ISO (c).

The cyclic olefinic resonances (δ = 5.34) overlap with the unreacted acyclic olefinic resonances at δ = 5.4 ppm, indicating that not all acyclic double bonds in the fatty acid chain were converted into six-member ring structure.

In general, the NMR spectra confirmed the successful modification of SO into NSO and ISO.

3.2 SBR system

3.2.1 Materials

A tin-coupled SBR Duradene HX739 from Firestone Polymers (Akron, OH) with 20 % bound styrene and 60 % vinyl content were used. It has a Mooney viscosity of 92. Three different filler systems including 60 phr of CB, 60 phr of silica and CB/silica hybrid filler at a ratio of 30/30 were used. CB N330 from Sid Richardson (Fort Worth, TX) with a

Brunauer-Emmet-Teller (BET) surface area of 78 m2/g, and precipitated silica Hi-Sil HDP

320G from PPG Industries (Pittsburgh, PA) with a BET surface area of 160 m2/g were used.

Curatives including sulfur under the trade name of Rubber Maker (RM) Sulfur, accelerator

N-cyclohexyl-2-benzothiazole sulfenamide (CBTS) and diphenyl guanidine (DPG), ZnO under the trade name of Zinc Oxide RGT-M, N-octadecanoic acid (stearic acid) under the trade name of Stearic Acid Rubber Grade, and antiozonant N-(1,3-dimethyl butyl)

N'-phenyl-p-phenylene diamine under the trade name of PD-2 were used. These ingredients were kindly provided by Akrochem Corporation (Akron, OH). For silica- and hybrid filler-filled compounds, a silane coupling agent Bis(3-Triethoxysilylpropyl) tetrasulfide under the trade name of SCA 98, provided by Struktol Co. of America (Stow,

OH), was used. For different oils, Plasticizer LN by Akrochem Corporation (Akron, OH) 36

was used as a petroleum-based naphthenic plasticizer. A technical grade RBD Soybean Oil by Cargill Industrial Oils & Lubricants (Minneapolis, MN) was used as received and also modified into 5NSO and 11NSO. Detailed recipes are tabulated in the next section.

To compare the tin-coupled and uncoupled linear SBR, a linear SBR Duradene

HX263 with 23.5% bound styrene manufactured by Firestone Polymers (Akron, OH) was also used. It has a Mooney viscosity of 55. CB N330 from Sid Richardson (Fort Worth, TX) was used. Additives including antiozonant, sulfur, CBTS, ZnO and stearic acid, and plasticizers including NO, SO, 5NSO and 11NSO were same as described above. In addition, 5ISO and 11ISO were also used.

3.2.2 Compounding and curing

Table 3.1 shows the compounding recipes for various tin-coupled SBR compounds filled with different filler systems including CB, silica and CB/silica hybrid fillers. The

SBR gum was first masticated for 2 min, then mixed with 30 phr of NO, SO, 5NSO and

11NSO for another 4 min using a Banbury mixer (1.2 L Model 86EM9804, Banbury USM

Corp., Ansonia, CT). A rotor speed of 60 rpm, a fill factor of 0.7, and a cooling water temperature of 50 °C were used. The oil-extended SBR masterbatches was then mixed with different fillers and other ingredients, except sulfur and CBTS, for 6 min (CB) and 8 min

(Hybrid filler and silica) using an 85 mL Brabender mixer (Model Plasti-corder, C.W

Brabender Instruments, Inc. South Hackensack, NJ). A rotor speed of 55 rpm and a setup temperature of 80 °C were used. Detailed compounding procedures are shown in Table 3.2.

2 phr of sulfur and 1.3 phr of CBTS were then mixed with filled compounds on a laboratory size two roll mill (Reliable Mill Supply Co., Ukiah, CA) for 20 passes. A rotor speed of 15

Table 3.1. Compounding recipes for various tin-coupled SBR.

CB-filled (phr)* Hybrid-filled (phr) Silica-filled (phr)* SBR 100 100 100 Oil 30 30 30 CB 60 30 - Silica - 30 60 Silane - 2.4 4.8 Sulfur 2 or 3 2 or 3 2 or 3 CBTS 1.3 or 1.95 1.3 or 1.95 1.3 or 1.95 ZnO 5 5 5 Stearic Acid 1 1 1 DPG - 0.75 1.5 Antiozonant 2 2 2 Total 201.3 or 202.95 204.45 or 206.1 207.6 or 209.25

Table 3.2. Compounding procedures for various tin-coupled SBR compounds.

CB-filled Hybrid-filled Silica-filled Time Action Time Action Time Action Oil-extended Oil-extended Oil-extended 0' 0' 0' rubber rubber rubber ZnO, Stearic 1'30" 30" Stearic Acid, PD-2 30" Stearic Acid, PD-2 Acid, PD-2 2' CB 1' Silica, Silane, CB 1' Silica, Silane 6' Dump 5' ZnO, DPG 5' ZnO, DPG

8' Dump 8' Dump

rpm, a gap of 3 mm and a cooling water temperature of 40 °C were used. In addition, a recipe with 3 phr of sulfur and 1.95 phr of CBTS was also used to increase the crosslink density of various tin-coupled SBR compounds. These samples are named as SBR/NO-3S,

SBR/SO-3S, SBR/5NSO-3S and SBR/11NSO-3S, respectively. Here “3S” stands for 3 phr of sulfur.

The compounding recipe for the uncoupled linear SBR is shown in Table 3.3. It was measured that linear SBR gum containing 15 phr of NO had similar Mooney viscosity as tin-coupled SBR gum containing 30 phr of NO with the Mooney viscosity being 32.6 and

31.8, respectively. Thus 15 phr of different oils including NO, SO, 5NSO, 11NSO, 5ISO and 11ISO were used. The compounding procedure is shown in Table 3.4. Various ingredients, except for sulfur and CBTS, were compounded using the same Brabender mixer, as described above. A rotor speed of 55 rpm and a setup temperature of 80 °C were used. Then 2 phr of sulfur and 1.3 phr of CBTS were mixed with CB-filled linear SBR compounds using the two-roll mill for 20 passes. A rotor speed of 15 rpm, a gap of 3 mm and a cooling water temperature of 40 °C were used.

The curing behaviors of various tin-coupled and uncoupled linear SBR compounds were measured using an Advanced Polymer Analyzer (APA 2000, Alpha Technologies,

Akron, OH) at a temperature of 160 °C, a frequency of 10.5 rad/s and a strain amplitude of

4.2%. The scorching time TS1, and curing time T95 of various SBR compounds were obtained from the curing curves. Cure rate index (CRI) was also calculated based on the equation CRI=100/(T95-TS1). For silica-filled SBR compounds, a curing time of 30 min was applied to all the silica-filled SBR compounds due to their marching cure. An electrically heated compression-molding press (Carver, Wabash, IN) was used for

vulcanization at the temperature of 160 °C and a pressure of 15 MPa. The sheeted SBR compounds were vulcanized into slabs with a dimension of 150 mm × 150 mm and a thickness range of 1.9–2.2 mm. Cylindrical vulcanized samples with a diameter of 16 mm and a length of 12 mm for DIN abrasion tests were also prepared.

Particularly, for various hybrid filler-filled tin-coupled SBR compounds, the Mooney viscosity ML(1+4) at 100 °C was measured using a Mooney viscometer (MV2000, Alpha

Technologies, Akron, OH).

Table 3.3. Compounding recipe for linear SBR.

Recipe (phr)

SBR 100 Oil 15 CB 60 Sulfur 2 CBTS 1.3 ZnO 5 Stearic Acid 1 Antiozonant 2 Total 186.3

Table 3.4. Compounding procedure for CB-filled linear SBR.

Time Action 0' Add SBR gum Add Stearic Acid, ZnO, 30" PD-2 1' Add CB + Oil 7' Dump

3.2.3 Characterizations

The bound rubber fraction of various filled SBR compounds, the gel fraction and crosslink density of various SBR vulcanizates were measured using Soxhlet extraction method, benzene was used as the solvent. Prior to the extraction, SBR compounds and vulcanizates of 0.7-0.8 g were weighed, and then they were placed inside Whatman cellulose extraction thimbles in Soxhlet extractors which were heated on electronic heaters.

The weights of the swollen vulcanizates were measured after an extraction time of 24 hours.

Then the samples were dried in a heated vacuum oven at 65 °C for another 24 hours.

Finally, the weights of dried rubbers were measured. The gel fraction was determined as the ratio of the final weight of dried vulcanizate over the initial weight of the vulcanizate.

The apparent crosslink density of SBR vulcanizates was calculated using the Flory-Rehner equation.127 Flory-Rehner equation is as follows:

ln(1 − 푉 ) + 푉 + 휒푉2 υ = − 푟 푟 푟 1 푉 푉 (푉3 − 푟) 1 푟 2

푤푒𝑖푔ℎ푡 표푓 푑푟푦 푟푢푏푏푒푟

푑푒푛푠𝑖푡푦 표푓 푟푢푏푏푒푟 with 푉푟 = 푤푒𝑖푔ℎ푡 표푓 푑푟푦 푟푢푏푏푒푟 푤푒𝑖푔ℎ푡 표푓 푠표푙푣푒푛푡 + 푑푒푛푠𝑖푡푦 표푓 푟푢푏푏푒푟 푑푒푛푠𝑖푡푦 표푓 푠표푙푣푒푛푡

푉1 is the molar volume of solvent. 휒 is the polymer-solvent interaction parameter.

Benzene has a value of 푉1= 88.838 cc/mole and its density is 0.874 g/mL. The density of

SBR is 0.93 g/cm3 and an interaction parameter (χ) of 0.4 was used for SBR and benzene.

The bound rubber fraction of compounds was calculated as a ratio of the weight of the rubber after the extraction over the initial weight of rubber within the filled compounds. All

SBR compounds were kept for 3 weeks before the bound rubber test.

Rheological properties of various SBR compounds were studied using the APA 2000.

A frequency sweep was performed in a frequency range from 0.06 to 200 rad/s (logarithmic mode) at a temperature of 90 °C and a strain amplitude of 4.2%. The frequency dependence of the storage (G’), loss (G”) moduli, tan δ and complex viscosity of various SBR compounds were measured. A strain sweep was also performed in a strain amplitude range from 0.3% to 200%, at a frequency of 0.6 rad/s and a temperature of 90 °C. The G’ of compounds as a function of strain amplitude of various SBR compounds was measured.

Differential Scanning Calorimeter (DSC Q200, TA Instruments, New Castle, DE) and

Thermogravimetric Analysis (TGA Q50, TA Instruments, New Castle, DE) were used to study the thermal properties of various SBR compounds and vulcanizates. The DSC tests at a heating rate of 10 °C/min were conducted under nitrogen atmosphere. Glass transition temperature (Tg) of various SBR compounds and vulcanizates was determined. The TGA curves of various SBR compounds and vulcanizates in a range from room temperature to

600°C at a heating rate of 20 °C/min were obtained under nitrogen atmosphere to report their thermal stability.

Mechanical properties including tensile, hardness, and abrasion resistance of various

SBR vulcanizates were studied. Dumb-bell shape specimens (ASTM D412, Die C) for tensile test were cut from the vulcanized slabs. The tensile tests were conducted at room temperature using an Instron tensile tester (Model 5567, Instron, Canton, MA) equipped with an extensometer at a crosshead speed of 500 mm/min. In addition, tear strength of various hybrid filler-filled tin-coupled SBR vulcanizates was measured using the Instron tensile tester according to ASTM D624 at a crosshead speed of 500 mm/min. Hardness of various SBR vulcanizates was measured at room temperature using a Durometer Type A

according to ASTM 2240. The abrasion test was conducted using a Deutsches Institut für

Normung Abrader (Zwick Abrasion Tester 6102, Ulm, Germany) according to DIN 53516.

The weight losses of the abraded samples were measured. The densities of various SBR vulcanizates were measured using an Accupyc 1340 Helium Pycnometer (Micromeritics

Instrument Corp., Norcross, GA). Then the volume losses of the abraded samples were calculated to report the abrasion resistance in volume. Specifically, to study the aging behavior of various hybrid filler-filled SBR vulcanizates, a hot air accelerated aging test was performed according to ASTM D573 at 100 °C for 48 hours. The tensile properties, tear strength and hardness of the aged hybrid filler-filled SBR vulcanizates were measured using the same conditions, as described above. The percentage change of properties of various hybrid filler-filled SBR vulcanizates was evaluated.

The dynamic mechanical properties of various SBR vulcanizates were evaluated using a Dynamic Mechanical Analysis (DMA Q800, TA Instruments, New Castle, DE), at a frequency of 1 Hz, a strain amplitude of 0.5% in tensile mode, a temperature sweep ranging from -90 °C to 90 °C, and a heating rate of 3°C/min. The storage (E’), loss (E’’) moduli and tan δ of various SBR vulcanizates were obtained. The values of tan δ at 10 °C and 60 °C were used as predictors of the wet traction and rolling resistance for tires.128

In order to investigate the possible reaction between the SO or NSO with the sulfur-containing silane coupling agent, SO, 5NSO and 11NSO were mixed with silane coupling agent and heated at 160 °C in a vacuum oven to simulate the curing process. 10 g of oils was mixed with 1.6 g of silane coupling agent and heated under vacuum for 30 min.

In addition, 10 g of oils and 1.6 g of silane coupling agent were heated at the same condition, but separately, and they were mixed afterwards to rule out other factors. The

oil/silane mixture undergoes reactions with each other during the heating, and the samples heated separately do not undergo such reactions. Therefore, the changes caused by the possible reactions can be investigated. Fourier transform infrared (FT-IR) spectroscopy was performed for the reacted and unreacted oil/silane mixtures after heating using a FT-IR spectrometer (Nicolet 380, Thermo Scientific, MA) by casting thin liquid film onto KBr crystal. Resolution of 4 cm-1 and number of scans of 32 were used. Steady-state shear viscosities of reacted and unreacted oil/silane mixtures after heating were measured by a

Discovery Hybrid Rheometer (DHR-2, TA Instruments, New Castle, DE) equipped with a

25 mm cone and plate fixtures with a cone angle of 2°. The steady-state shear viscosities of the oil/silane mixtures as a function of shear rate ranged from 10 s-1 to 100 s-1 were measured at a temperature of 35 °C.

3.3 CR system

3.3.1 Materials

A CR under the trade name of Neoprene GW manufactured by DuPont (Wilmington,

DE) was used. Since it is a G type CR, sulfur was copolymerized with chloroprene monomer during the production. N550 is the most widely used CB in CR. Therefore, CB

N550 from Evonik Corporation (Parsippany, NJ) with a BET surface area of 40 m2/g was used as a filler. ZnO under the trade name of Zinc Oxide RGT-M, MgO under the trade name of Elastomag 170, N-octadecanoic acid (stearic acid) under the trade name of Stearic

Acid Rubber Grade, and ETU Akroform ETU-75/EPR/P from Akrochem Corporation

(Akron, OH) were used as compounding additives. For conventional petroleum plasticizer, a NO under the trade name Plasticizer LN by Akrochem Corporation (Akron, OH) was 44

used. A technical grade RBD soybean oil manufactured by Cargill Industrial Oils &

Lubricants (Minneapolis, MN) was used. NSO with modification levels of 11% (11NSO) and 33% (33NSO) was used.

3.3.2 Compounding and curing

The compounding recipe is shown in Table 3.5 and the compounding procedure is shown in Table 3.6. The CR gum and other ingredients were compounded using an 85 mL

Brabender mixer (Model Plasti-corder, C.W Brabender Instruments, Inc. South

Hackensack, NJ). A rotor speed of 60 rpm and a setup temperature of 60 °C were used.

Then the ZnO was mixed with compounds on a laboratory size two roll mill (Dependable

Rubber Machinery Co., Cleveland, OH) for 20 passes. A rotor speed of 20 rpm, a gap size of 3 mm and a cooling water temperature of 40 °C were used. The final compounds were stored overnight before the vulcanization.

The curing curves of various CB-filled CR compounds with curatives were obtained from an Advanced Polymer Analyzer (APA 2000, Alpha Technologies, Akron, OH) at a temperature of 160 °C, a frequency of 10.5 rad/s and a strain amplitude of 4.2%. The scorch time Ts1 and curing time T95 were obtained based on the curing curves. CRI was also calculated based on the equation CRI=100/(T95-TS1). Same curing test was also applied to

CB-filled CR compounds without any curatives. An electrically heated compression-molding press (Carver, Wabash, IN) was used for vulcanization at a temperature of 160 °C and a pressure of 15 MPa. Vulcanized slabs with a dimension of 150 mm × 150 mm and a thickness range of 2.0-2.2 mm were obtained. Cylindrical samples with a diameter of 16 mm and a length of 12 mm for abrasion tests were also prepared.

Table 3.5. Compounding recipe for CB-filled CR compounds.

Amount Ingredients (phr) CR 100 Oil 15 CB 50 MgO 4 ZnO 5 Stearic Acid 1 ETU 0.75

Table 3.6. Compounding procedure for CB-filled CR compounds.

Time, min Action 0 Add CR 1 Add MgO+Stearic Acid 2 Add Oil 2.5 Add CB 11 Add ETU 12 Dump

3.3.3 Characterizations

The Soxhlet extraction method was applied to calculate the gel fraction and crosslink density of various CB-filled CR vulcanizates, and also the bound rubber fraction of various uncured CB-filled CR compounds. Benzene was used as the solvent. CR vulcanizate around 0.8 g was weighed first, then the sample was placed inside a Whatman cellulose extraction thimble in the Soxhlet extractor heated on a heater. The weight of the swollen vulcanizate was measured after an extraction time of 48 hours. Then the sample was dried in a heated vacuum oven at 65 °C for another 24 hours. Finally, the weight of dry rubber was measured. The gel fraction was determined as the ratio of the final weight of dried CR vulcanizate over the initial weight of the CR vulcanizate. The crosslink density was calculated using the Flory-Rehner equation127 and the Kraus correction.129

Kraus correction was applied using the following equation:

1 휙 [3퐶 (1 − 푉3 ) + 푉 − 1] 푉 푟표 푟표 푟표 = 1 − 푉푟 1 − 휙

Here 푉푟표 is the volume fraction of CR in the unfilled swollen sample, 푉푟 is the volume fraction of rubber in the CB-filled swollen sample. 휙 is the volume fraction of filler in the filled CR vulcanizate after drying, and C is a constant depending on the type of

CB. Here C is 1.16 for CB N550.129 The bound rubber fraction was calculated as the ratio of the weight of the rubber after the extraction over its initial weight within the compound.

Various CB-filled CR compounds were stored for 3 weeks at room temperature prior to the bound rubber test.

Rheological properties of various CB-filled CR compounds were measured by the

APA 2000. A frequency sweep was performed in a frequency range from 0.06 to 200 rad/s 47

(logarithmic mode) at a temperature of 90 °C and a strain amplitude of 4.2%. The frequency dependence of the G’, G”, tan δ and complex viscosity of various CB-filled CR compounds was measured. A strain sweep was performed in a strain amplitude range from

0.3% to 200% at a frequency of 0.6 rad/s and a temperature of 90 °C. The G’, G” and tan δ of CB-filled CR compounds as a function of strain amplitude were measured.

Thermogravimetric Analysis (TGA Q50, TA Instruments, New Castle, DE) and

Differential Scanning Calorimeter (DSC Q200, TA Instruments, New Castle, DE) were used to evaluate the thermal properties of various CB-filled CR compounds and vulcanizates. The TGA curves in a range from room temperature to 700°C at a heating rate of 20 °C/min were obtained under nitrogen atmosphere. The DSC curves in a temperature range from -90 °C to 30 °C at a heating rate of 10 °C/min were obtained under a nitrogen atmosphere.

Mechanical tests including tensile, tear, hardness and abrasion tests were performed for various CB-filled CR vulcanizates. The tensile tests were conducted at room temperature according to ASTM D412 using an Instron tensile tester (Model 5567, Instron,

Canton, MA) equipped with an extensometer. At least five samples cut with an ASTM

D412 Die C were tested and a crosshead speed of 500 mm/min was used. Tear strength of various CB-filled CR vulcanizates was measured using the Instron tensile tester according to ASTM D624. Hardness of various CB-filled CR vulcanizates was measured at room temperature using a Durometer Shore A according to ASTM 2240. The abrasion test was conducted using a Zwick Abrasion Tester 6102 according to DIN 53516. A load of 10 N and a pathway of 40 meters was applied to the specimens without specimen rotation. The weight loss of the sample after the abrasion was measured. The densities of various

CB-filled CR vulcanizates were measured using an Accupyc 1340 Helium Pycnometer

(Micromeritics Instrument Corp., Norcross, GA). Then the volume loss of the abraded sample was calculated to report the abrasion resistance in volume. At least four specimens of each type of vulcanizate were used in the test.

To study the aging effect of various CB-filled CR vulcanizates, a hot air aging test was performed according to ASTM D573 at 100 °C for 48 hours. The tensile properties, tear strength and hardness of the aged CB-filled CR vulcanizates were measured using the same conditions, as described above. The volume of the same sample before and after the aging was also measured using a pycnometer. The changes of properties in percentages of various CB-filled CR vulcanizates were obtained.

3.4 Butyl rubber system

3.4.1 Materials

Exxon Butyl 268 from ExxonMobil Chemical (Spring, TX) was used. CB N330 from

Sid Richardson (Fort Worth, TX) was used as a filler. ZnO (Zinc Oxide RGT-M), stearic acid (Stearic Acid Rubber Grade) and brominated octylphenol-formaldehyde resin P-124 from Akrochem Corporation (Akron, OH) were used as compounding additives. For different plasticizers, NO (Plasticizer LN) by Akrochem Corporation (Akron, OH) and SO

(RBD Soybean Oil) by Cargill Industrial Oils & Lubricants (Minneapolis, MN) were used.

NSO with modification levels of 5% (5NSO) and 11% (11NSO) were used.

3.4.2 Compounding and curing

The compounding recipes are shown in Table 3.7. Two recipes with different oil loadings were used. It was found that by reducing the loading of SO and NSO, the crosslink density of IIR vulcanizates was increased. Recipe-1 contained 15 phr of oil. In recipe-2 a content of oil was reduced to 10 phr. The IIR gum was compounded with ingredients using an 85 mL Brabender mixer (Model Plasti-corder, C.W Brabender Instruments, Inc. South

Hackensack, NJ). Compounding procedure is shown in Table 3.8. A rotor speed of 55 rpm and a setup temperature of 80 °C were used. The CB-filled IIR compounds were then compounded with curing resin in the second stage mixing using the same set of parameters.

The final compounds were dumped at 100 °C. Then the IIR compounds were sheeted on a laboratory size two roll mill (Reliable Mill Supply Co., Ukiah, CA) for 15 passes. A rotor speed of 15 rpm, a gap size of 3 mm and a cooling water temperature of 45 °C were used.

Advanced Polymer Analyzer (APA 2000, Alpha Technologies, Akron, OH) was used to obtain curing curves of various CB-filled IIR compounds at a temperature of 180 °C. A frequency of 10.5 rad/s and a strain amplitude of 4.2% were used. The scorch time Ts1, curing time T95 and CRI of various CB-filled IIR compounds was then calculated. An electrically heated compression-molding press (Carver, Wabash, IN) was used for vulcanization at a temperature of 180 °C and a pressure of 15 MPa. Vulcanized slabs of a dimension of 150 mm × 150 mm with a thickness range of 2.0-2.2 mm were obtained.

Table 3.7. Compounding recipes for CB-filled IIR compounds.

Recipe-1 (phr) Recipe-2 (phr)

IIR 100 100 Oil 15 10 CB 60 60 ZnO 5 5 Stearic Acid 1 1 Resin 10 10 Total 191 186

Table 3.8. Compounding procedure for CB-filled IIR compounds.

Time Action 0' Add IIR gum 30" Add Stearic Acid, ZnO 1' Add CB + Oil 8' Dump

3.4.3 Characterizations

The Soxhlet extraction method was applied to measure the gel fraction and crosslink density of various CB-filled IIR vulcanizates. Samples were extracted by cyclohexane in

Soxhlet extractors for 24 hours. Then the sample was dried in a heated vacuum oven at

65 °C for 24 hours. The gel fraction was determined as the ratio of the final weight of dried vulcanizate over the initial weight of the vulcanizate. The crosslink density was calculated using the Flory-Rehner equation127 and the Kraus correction.129 Cyclohexane has a value of

3 푉1= 103.5 cc/mole and its density is 0.778 g/mL. The density of IIR is 0.92 g/cm and an interaction parameter (χ) of 0.43 were used for IIR and cyclohexane. For the Kraus correction, the constant C is 1.17 for CB N330.129

To study the reaction of SO and NSO with curing resin, oil/resin mixtures were prepared. 10.0 g of oil (SO, 5NSO or 11NSO), 1.0 g of P-124 resin and 0.1 g of zinc stearate were mixed. DSC (Q200, TA Instruments, New Castle, DE) was used to characterize the thermal behavior of oil/resin mixture. Samples were kept at 160 °C under

N2 atmosphere for 5 min, and DSC curves were obtained. The reaction is very rapid and it happened before the samples reach a higher temperature, such as curing temperature

180 °C, thus a lower temperature of 160 °C was used to have a proper reaction speed. To study the viscosity change of the oil/resin mixture before and after the reaction, the mixture was heated in a vacuum oven at 180 °C for 60 min. The pure oils were also heated at the same condition, and then mixed with resin to rule out other factors influencing the viscosity change. The shear viscosity was measured by an ARES-G2 Rheometer (TA Instruments,

New Castle, DE). The steady-state shear viscosities of the oil/resin mixtures as a function of shear rate ranged from 10 s-1 to 100 s-1 were measured at room temperature.

APA 2000 was used to perform rheological tests at a temperature of 90 °C. A strain sweep was performed in a strain amplitude range from 0.3% to 200% at a frequency of 0.6 rad/s. G’ of uncured CB-filled IIR compounds as a function of strain amplitude was measured. A frequency sweep was performed in a frequency range from 0.06 to 200 rad/s

(logarithmic mode) and a strain amplitude of 4.2% for CB-filled IIR compounds and vulcanizates. For CB-filled IIR vulcanizates, samples were first cured at 180 °C and then cooled down to 90 °C in the APA 2000. The G’, G”, tan δ and complex viscosity of various

CB-filled IIR vulcanizates as a function of frequency were measured.

TGA (Q50, TA Instruments, New Castle, DE) was used to report the thermal stability of various CB-filled IIR vulcanizates. The TGA curves in a range from room temperature to 600°C at a heating rate of 20 °C/min were obtained under N2 atmosphere. DSC (Q200,

TA Instruments, New Castle, DE) was used to report the glass transition temperature (Tg) of various CB-filled IIR vulcanizates. The DSC curves were obtained in a temperature range from -90 °C to 40 °C at a heating rate of 10 °C/min under N2 atmosphere.

Tensile tests were performed according to ASTM D412 at room temperature using an

Instron tensile tester (Model 5567, Instron, Canton, MA) equipped with an extensometer. A crosshead speed of 500 mm/min was used. Hardness of various CB-filled IIR vulcanizates was measured at room temperature using a Durometer Shore A according to ASTM 2240.

A hot air aging test was conducted according to ASTM D573 at 120 °C for 100 hours. The tensile properties and hardness of the aged CB-filled IIR vulcanizates were measured.

FT-IR spectroscopy was obtained for the surface of CB-filled IIR vulcanizates to characterize the surface oil migration. Resolution of 4 cm-1 and number of scans of 32 were used.

CHAPTER IV

COMPARISON OF CB- AND SILICA-FILLED TIN-COUPLED SBR WITH VARIOUS OILS

4.1 Introduction

The purpose of this study is to compare the behavior of silica- and CB-filled tin-coupled SBR compounds and vulcanizates containing NO, SO, 5NSO and 11NSO. The behavior of NSO in silica-filled SBR is expected to be different due to the different properties of silica and CB. By comparing the effect of NSO in silica- and CB-filled SBR systems will lead to a better understanding of the behavior of the NSO. It has great importance for the applications of the NSO in rubber products such as tires. The reaction between the SO, NSO and silane is investigated. Bound rubber fraction, rheological and thermal properties of various silica- and CB-filled SBR compounds are studied. For vulcanizates, mechanical, thermal and dynamic properties are studied. The importance of this study is related to the fact that silica and CB fillers are widely used in tires and investigation on addition of novel environmental friendly NSO, instead of petroleum oil, to rubbers will provide novel approach to improve tires performance along with safety in their manufacturing.

This chapter is based on a paper published by the author of this dissertation.130 Data of various CB-filled tin-coupled SBR are from author’s previous study.126

4.2 Results and Discussion

4.2.1 Reaction of SO/NSO with Silane

It was reported that when the SO was compounded into the rubber matrix, it could consume the curatives during the vulcanization resulting in a decrease of the state of cure.5,

13, 15, 126 In previous study,126 it was found that with the increase of the modification level, the NSO was more reactive during the vulcanization and could consume more curatives such as sulfur compared to that of the virgin SO. The silane coupling agent used in the silica-filled SBR compounds also contains sulfur groups and can create chemical bonds with rubber chains upon heating.105 Since there is similarity between the sulfur crosslinking and the coupling reaction of silane, the silane can possibly react with SO or

NSO. Therefore, it is necessary to investigate the possible reaction between the silane and

SO or NSO.

Figure 4.1 shows the FT-IR spectra of unreacted and reacted SO/silane (a),

5NSO/silane (b) and 11NSO/silane (c). In Figure 4.1 (a), the band at 3007 cm-1 is attributed to the C-H stretching of the C=C-H groups in the fatty acid chains of the SO molecule. The intensity of the 3007 cm-1 band decreases after reacting with the silane, indicating the silane reacts with the double bonds in the SO. In Figure 4.1 (b) and (c), the band at 3052 cm-1 is attributed to C-H stretching of norbornyl C=C-H groups. It is observed that after the reaction with silane, the band of norbornyl C=C-H groups of the NSO disappears and the intensity at 3007 cm-1 also decreases tremendously. In addition, the change of C=C absorption band of 1745 cm-1 is also observed due to the reaction. The results from the

FT-IR analysis indicate that the silane coupling agent can react with the double bonds on the fatty acid chains of the SO and the norbornyl groups of the NSO. 55

Figure 4.1. FT-IR spectra of unreacted and reacted SO/silane (a), 5NSO/silane (b), and 11NSO/silane (c).

After the reaction, the SO and NSO may be crosslinked by the silane, therefore it is anticipated an increase of the viscosity of the mixture. In order to prove this hypothesis, the shear viscosities of the mixtures were measured. Figure 4.2 shows the viscosity as a function of shear rate of various oil/silane mixtures with and without reaction after heating at 160 °C. All the samples show a Newtonian behavior in the tested range of shear rates. It is clearly seen that after the reaction, the viscosity of SO/silane, 5NSO/silane and

11NSO/silane mixture is increased by 298.7%, 351.4% and 456.8%, respectively. The result demonstrates that the oils can react with silane resulting in an obvious increase in viscosity. With an increase of the modification level, more norbornyl C=C double bonds could react with the silane leading to a higher increase in the viscosity.

The FT-IR and viscosity results demonstrate that the SO and NSOs can react with the silane upon heating, and that with the increase of the modification level, the NSOs can react more with the silane. This reaction may consume certain amount of silane in the rubber compounds, possibly leading to less rubber-filler coupling effect. At the same time, part of the oil will be cured onto the surface of the silica via silane. These effects brought about by the reaction between the oil and silane will be further discussed in the next part of study.

4.2.2 Gel Fraction, Crosslink Density and Bound Rubber

Figure 4.3 shows gel fraction (a), crosslink density (b) of various silica- and CB-filled

SBR vulcanizates and bound rubber fraction (c) of various silica- and CB-filled SBR compounds. Figure 4.3 (a) shows that both silica- and CB-filled SBR/SO and SBR/NSO vulcanizates exhibit higher gel fractions than those of the SBR/NO vulcanizates due to the

Figure 4.2. Viscosity as a function of shear rate of unreacted and reacted SO/silane, 5NSO/silane, and 11NSO/silane at 35°C.

Figure 4.3. Gel fraction (a), crosslink density (b) of various silica- and CB-filled SBR vulcanizates and bound rubber fraction (c) of various silica- and CB-filled SBR compounds.

covulcanization of the oils with rubber.126 It is noticed that the increase in gel fraction of the silica-filled SBR/SO and SBR/NSO vulcanizates compared to the silica-filled SBR/NO vulcanizate is much more obvious than that of the CB-filled case. This is due to the fact that the sulfur-containing silane coupling agent in the silica-filled SBR vulcanizates can react with the SO or NSO and bound onto the silica surface, which will effectively increase the gel fraction. With an increase of the modification level, the gel fractions of the silica- and

CB-filled SBR/NSO vulcanizates are slightly increased due to the increasing reactivity of the NSO. The gel fractions of the silica- and CB-filled SBR/NO-3S, SBR/SO-3S,

SBR/5NSO-3S and SBR/11NSO-3S vulcanizates are increased due to the extra amount of curatives in the recipe. It is shown that the increase in gel fraction of the silica- and

CB-filled SBR/SO-3S and SBR/NSO-3S vulcanizates is much larger than those of the silica- and CB-filled SBR/NO-3S vulcanizates. This is also due to the reactivity of NSO which can covulcanize with the rubber matrix at the presence of extra amount of sulfur, increasing the gel fraction.

Figure 4.3 (b) depicts the apparent crosslink densities of various silica- and CB-filled

SBR vulcanizates. The silica-filled SBR vulcanizates show overall higher apparent crosslink densities than those of the CB-filled SBR vulcanizates, with a similar trend in terms of different oils. At the same level of curatives, the silica- and CB-filled SBR/NO vulcanizates exhibit higher apparent crosslink densities than those of the silica- and

CB-filled SBR/SO and SBR/NSO vulcanizates. With an increase of the modification level, the apparent crosslink densities of silica- and CB-filled SBR/NSO vulcanizates are decreased. Previous studies reported that the double-bonds-containing triglyceride-based oils can consume the curatives resulting in a decrease of crosslink density.13, 15, 131 With an

increase of the modification level, the reactivity of the NSO increases, leading to more consumption of the curatives. Due to the extra amount of curatives, the silica- and

CB-filled SBR/NO-3S, SBR/SO-3S, SBR/5NSO-3S and SBR/11NSO-3S vulcanizates exhibit higher apparent crosslink densities compared to those of the SBR vulcanizates with less curatives.

Figure 4.3 (c) shows the results of bound rubber test of silica- and CB-filled SBR compounds. It is observed that the CB-filled SBR/SO and SBR/5NSO compounds exhibit higher bound rubber fractions than that of the CB-filled SBR/NO compound. Due to the increase of the modification level, the bound rubber fraction of CB-filled SBR/11NSO compound is decreased. However, the bound rubber fractions of various silica-filled SBR compounds exhibit a different trend. Compared to the CB-filled SBR compounds, the silica-filled compounds show an overall lower bound rubber fraction due to the weak rubber-silica interaction. The silica-filled SBR/SO compound exhibits lower bound rubber fraction than that of the silica-filled SBR/NO compound, and with an increase of the modification level, the bound rubber fractions of silica-filled SBR/NSO compounds are further decreased. This is due to the fact that the silane coupling agent, bonded on the surface of the silica particles, can react with SO and NSO instead of bonding with rubber, leading to less coupling effects and less bound rubber, as shown in Scheme 1. With the increasing reactivity of the NSO, this reaction will further decrease the bound rubber fraction of the compound.

Scheme 1. Proposed structure at the surface of silane treated silica.

4.2.3 Rheological Properties

Figure 4.4 shows the frequency dependences of the G’ (a), G” (b), tan δ (c) and complex viscosity (d) of various silica- and CB-filled SBR compounds in the linear region of their behavior. From Figure 4.4 (a) and (b), it is observed that the G’ and G” of silica-filled SBR compounds are higher than those of the CB-filled SBR compounds due to the higher surface area of the silica and also a stronger filler-filler interaction. It should be noted that a higher BET surface area of filler provides a higher specific surface area that is accessible by rubber molecules.81 For both types of fillers, different oils have minimal effects on the values of G’ and G”. It is also noticed that at lower frequency region, the silica-filled SBR compounds exhibit less frequency dependence of G’ than those of the

CB-filled SBR compounds. This is due to the stronger filler-filler interaction of the silica forming a stronger filler-filler network. The network restrains the relaxation of the polymer chains, leading to a higher G’ at low frequency region.132 Figure 4.4 (c) depicts the tan δ as a function of frequency of various silica- and CB-filled SBR compounds. The tan δ values of the silica-filled SBR compounds are lower than those of the CB-filled SBR compounds.

At high frequency region, the tan δ values of various silica- and CB-filled SBR compounds converge, showing less difference in values. In both silica- and CB-filled SBR compounds, the SBR/NO compounds exhibit lower tan δ values than those of the SBR/SO and

SBR/NSO compounds. For the same type of filler, SO and NSO give negligible differences in the tan δ values. Figure 4.4 (d) shows complex viscosities of various silica- and

CB-filled SBR compounds. It is observed that the complex viscosities of both silica- and

CB-filled SBR compounds exhibit a strong decrease with the increase of frequency. Due to the higher surface area of silica, the silica-filled SBR compounds have overall higher

Figure 4.4. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) as a function of frequency of various silica- and CB-filled SBR compounds at 90°C.

complex viscosities than those of the CB-filled SBR compounds. Similar to the G’, the complex viscosities of silica-filled SBR compounds are more frequency dependent in the low frequency region, showing a tendency to exhibit yielding behavior which indicates the presence of a stronger filler-filler interaction.

Figure 4.5 shows the dependence of G’ on the strain amplitude of various silica- and

CB-filled SBR compounds plotted in logarithm (a) and linear (b) scale. It is clearly observed that the silica-filled SBR compounds show an overall higher G’ at lower strain amplitudes and showing higher reduction of G’ with the strain amplitude than that of the

CB-filled SBR compounds. This reduction in G’ is also known as the Payne effect.133 The stronger Payne effect of the silica-filled SBR compounds than CB-filled SBR compounds is due to the larger surface energy difference between precipitated silica and SBR.81 In addition, the hydrogen bonding between the silanol groups on the surface of the silica particles creates a stronger filler-filler interaction compared to that of the CB. This filler-filler interaction gives a higher G’ of silica-filled SBR compounds at low strain amplitudes. Similar phenomenon was also observed in previous studies on silica- and

CB-filled rubbers.116 It is found that the silica- and CB-filled SBR/NO compounds exhibit a slightly stronger rubber-filler interaction than those of the SBR/SO and SBR/NSO compounds. This is indicated by higher values of G’ at high strain amplitudes in these compounds. Both silica- and CB-filled SBR/SO and SBR/NSO compounds exhibit a similar rubber-filler interaction.

Figure 4.5. Storage modulus as a function of strain amplitude of various silica- and CB-filled SBR compounds at 90°C in logarithm (a) and linear (b) scale.

4.2.4 Thermal Properties

Figure 4.6 shows the TGA curves of various silica-filled SBR compounds (a),

CB-filled SBR compounds (b), silica-filled SBR vulcanizates (c) and CB-filled SBR vulcanizates (d). It is clearly observed that for all the SBR compounds and vulcanizates, the samples containing NO exhibit the most inferior thermal stability. A two-stage decomposition process was observed for all the SBR/NO compounds and vulcanizates.

They start to lose weight around 130 °C due to the evaporation of the NO. On the other hand, the SBR compounds and vulcanizates containing SO and NSO exhibit similar and excellent thermal stability compared to those of the samples containing NO. The thermal stability difference is due to the differences in decomposition temperature and flash point of oils.126 Crosslinking does not provide any additional improvement in thermal stability.

Silica- and CB-filled SBR/SO and SBR/NSO samples exhibit minimal weight loss until the rubber matrix starts to decompose. It is noticed that the type of filler has negligible influence on the thermal stability of rubber compounds and vulcanizates. The usage of

SO and NSO can enhance processing safety during the compounding.

Figure 4.7 shows the DSC curves of various silica-filled SBR compounds (a),

CB-filled SBR compounds (b), silica-filled SBR vulcanizates (c) and CB-filled SBR vulcanizates (d). Table 4.1 shows the glass transition temperature Tg determined from these DSC curves. From Figure 4.7 (a) and (b), the silica- and CB-filled SBR compounds exhibit similar thermal behavior. For both silica- and CB-filled SBR compounds, the

SBR/NO compounds have the highest Tg and the SBR/SO compounds have the lowest Tg.

With the increase of the modification level, the Tg of SBR/NSO compounds are seen to increase. This is due to the increasing numbers of cycloaliphatic norbornyl groups on

Figure 4.6. TGA curves of various silica-filled (a), CB-filled (b) SBR compounds, silica-filled (c) and CB-filled (d) SBR vulcanizates.

Figure 4.7. DSC curves of various silica-filled (a), CB-filled (b) SBR compounds, silica-filled (c) and CB-filled (d) SBR vulcanizates.

Table 4.1. Glass transition temperature of various silica- and CB-filled SBR compounds and vulcanizates.

Silica-filled CB-filled Compound T Vulcanizate T Compound T Vulcanizate T g g g g (°C) (°C) (°C) (°C) SBR/NO -44.0 -40.0 -45.8 -38.8 SBR/SO -48.6 -45.1 -52.2 -42.3 SBR/5NSO -47.2 -43.0 -48.8 -41.2 SBR/11NSO -46.3 -42.8 -48.1 -40.5 SBR/NO-3S -44.0 -39.1 -45.8 -37.6 SBR/SO-3S -48.6 -43.9 -52.2 -41.6 SBR/5NSO-3S -47.2 -41.6 -48.8 -40.4 SBR/11NSO-3S -46.3 -40.7 -48.1 -39.9

126 the NSO leading to a higher Tg of compounds. The Tg values of NO, SO, 5NSO and

11NSO are -80.6 °C, -105.2 °C, -96.4 °C and -89.1 °C, respectively. These differences in

Tg values of various oils correlate with differences observed in Tg values of compounds and vulcanizates. Figure 4.7 (c) and (d) show the DSC curves of silica- and CB-filled

SBR vulcanizates. After the vulcanization, the Tg of both silica- and CB-filled SBR vulcanizates are increased compared to those of the uncured SBR compounds.

Silica-filled SBR vulcanizates exhibit a slightly lower Tg compared to those of the

CB-filled SBR vulcanizates containing same type of oil. This is possibly due to the reaction between oils, silane and silica leading to reduced mobility of rubber chains. With the increase of the modification level, the Tg of the SBR/NSO vulcanizates is increased.

After increasing the amount of curatives, silica- and CB-filled SBR/NO-3S, SBR/SO-3S,

SBR/5NSO-3S, and SBR/11NSO-3S vulcanizates show an increase in Tg compared to those of the SBR vulcanizates with less curatives. This is due to the increasing crosslink density shown in Figure 3 (b). It is found that the usage of SO and NSO can achieve a lower Tg of the SBR product than the conventional NO, providing a better low temperature flexibility which is crucial for tire rubbers.

4.2.5 Curing Behaviors

Figure 4.8 shows the curing curves of various silica-filled SBR compounds (a) and

CB-filled SBR compounds (b). For silica-filled SBR compounds, all the compounds show a marching torque. The silica-filled SBR/NO-3S and SBR/SO-3S compounds exhibit the highest and second highest state of cure, respectively, due to their higher content of curatives, followed by the silica-filled SBR/NO compound which has a higher maximum

Figure 4.8. Curing curves of various silica-filled (a) and CB-filled (b) SBR compounds at 160°C.

Table 4.2 Curing time T95 of various CB-filled SBR compounds.

Compound T CB-filled 95 (min) SBR/NO 18.2 SBR/SO 17.8 SBR/5NSO 14.4 SBR/11NSO 11.5 SBR/NO-3S 12.9 SBR/SO-3S 9.7 SBR/5NSO-3S 7.6 SBR/11NSO-3S 7.6

torque than those of the silica-filled SBR/SO and SBR/NSO compounds. With the increase of modification level, the silica-filled SBR/5NSO and SBR/11NSO compounds exhibit a decreasing state of cure due to the consumption of the curatives by the reactive NSOs. By adjusting the curative recipe, the silica-filled SBR/5NSO-3S and SBR/11NSO-3S compounds exhibit a similar maximum torque compared to that of the silica-filled SBR/SO compound. For CB-filled SBR compounds shown in Figure 4.8 (b), a torque plateau is observed. The CB-filled SBR/NO-3S compound has the highest maximum torque, followed by CB-filled SBR/SO-3S, SBR/NO, SBR/5NSO-3S and SBR/11NSO-3S compounds. The CB-filled SBR/SO, SBR/5NSO and SBR/11NSO compounds have a decreasing order of state of cure due to the curative consumption. By comparing the curing curves of various silica- and CB-filled SBR compounds, it is noticed that the silica-filled

SBR compounds have an overall higher minimum and maximum torque compared to those of the CB-filled SBR compounds. This is due to the higher surface area of this silica and a stronger Payne effect. This effect was also seen in the crosslink density data shown in

Figure 4.3 (b) and the rheological data shown in Figure 4.4. The curing times of various silica-filled SBR compounds are longer (30 min) than those of the CB-filled SBR compounds (Table 4.2). This is due to the different surface chemistry between the silica and CB. The silanol groups can absorb accelerators and ZnO leading to a slower curing.88,

103

4.2.6 Mechanical Properties

Figure 4.9 shows the stress-strain curves of the various silica-filled SBR vulcanizates

(a) and CB-filled SBR vulcanizates (b). The comparison of M100, M300, elongation at break and tensile strength of these vulcanizates are shown in Table 4.3. Both silica- and

CB-filled SBR/SO vulcanizates show an increase in elongation at break and tensile strength and decrease in M100 and M300 compared to those of the silica- and CB-filled

SBR/NO vulcanizates. Similar results were also observed in previous studies on silica and

CB-filled SBR/BR/NR blend compounded with SO4, 15and CB-filled SBR compounded with crosslinked SO. With the increase of the modification level, the elongation at break of the silica- and CB-filled SBR/NSO vulcanizates is further increased. The elongation at break of the silica- and CB-filled SBR/11NSO vulcanizates are more than two times of the silica- and CB-filled SBR/NO vulcanizates, respectively. It is also observed that the silica- and CB-filled SBR/5NSO and SBR/11NSO vulcanizates exhibit higher tensile strength than those of the silica- and CB-filled SBR/SO and SBR/NO vulcanizates. The silica- and

CB-filled SBR/5NSO exhibit the highest tensile strength among the SBR samples with the same recipe. While the usage of NSO can tremendously improve the ultimate tensile properties, the moduli are decreased with the increase of the modification level due to the increasing consumption of the curatives by the NSOs. In order to increase the moduli of the silica- and CB-filled SBR vulcanizates, 3 phr of sulfur and 1.95 phr of CBTS were used.

After adjusting the recipe, it is observed that the moduli of the silica-filled SBR/5NSO-3S and SBR/11NSO-3S are tremendously increased and higher than that of the silica-filled

SBR/SO vulcanizate. At the same time their tensile strength is maintained and the elongation at break is similar to that of the silica-filled SBR/SO vulcanizates. It is also seen

Figure 4.9. Stress-strain curves of various silica-filled (a) and CB-filled (b) SBR vulcanizates.

Table 4.3. Tensile properties and hardness of various silica- and CB-filled SBR vulcanizates.

Elongation at M100 M300 Tensile Strength Hardness Break (MPa) (MPa) (%) (MPa) (Shore A)

Silica-filled

SBR/NO 3.15 ± 0.07 9.96 ± 0.28 351.8 ± 8.5 12.3 ± 0.3 67 SBR/SO 2.26 ± 0.04 6.92 ± 0.23 484.2 ± 10.3 13.4 ± 0.3 64 SBR/5NSO 1.66 ± 0.04 4.76 ± 0.13 683.0 ± 5.8 15.0 ± 0.5 64 SBR/11NSO 1.50 ± 0.02 4.13 ± 0.12 731.2 ± 15.3 14.0 ± 0.7 64 SBR/NO-3S 4.93 ± 0.12 - 195.2 ± 15.2 10.0 ± 0.8 74 SBR/SO-3S 3.66 ± 0.04 - 272.8 ± 5.8 11.6 ± 0.2 69 SBR/5NSO-3S 2.56 ± 0.04 8.09 ± 0.22 459.1 ± 12.1 14.7 ± 0.7 68 SBR/11NSO-3S 2.39 ± 0.05 7.39 ± 0.16 489.2 ± 9.6 14.4 ± 0.8 67 CB-filled

SBR/NO 2.74 ± 0.04 12.36 ± 0.02 321.3 ± 4.2 13.6 ± 0.3 65 SBR/SO 2.22 ± 0.02 9.87 ± 0.18 427.3 ± 3.1 15.8 ± 0.4 61 SBR/5NSO 1.41 ± 0.04 5.48 ± 0.13 671.8 ± 10.9 17.1 ± 0.4 59 SBR/11NSO 1.28 ± 0.02 4.51 ± 0.05 755.2 ± 6.4 16.6 ± 0.3 59 SBR/NO-3S 3.12 ± 0.07 - 267.1 ± 11.5 13.0 ± 0.9 67 SBR/SO-3S 2.54 ± 0.04 11.66 ± 0.12 342.7 ± 12.5 15.1 ± 0.6 62 SBR/5NSO-3S 2.03 ± 0.03 9.10 ± 0.08 512.5 ± 2.1 18.6 ± 0.2 60 SBR/11NSO-3S 2.19 ± 0.01 8.63 ± 0.10 560.0 ± 8.5 19.3 ± 0.4 61

the silica-filled SBR/SO-3S vulcanizate exhibits a higher modulus but a lower elongation at break and tensile strength than the silica-filled SBR/NO vulcanizate. For the CB-filled

SBR/5NSO-3S and SBR/11NSO-3S, their moduli are slightly lower than that of the

CB-filled SBR/SO vulcanizate and exhibit much higher tensile strength and elongation at break. The CB-filled SBR/SO-3S vulcanizate exhibits a similar property to that of the

CB-filled SBR/NO vulcanizate. For both silica- and CB-filled SBR/NO-3S vulcanizates, the tensile properties are inferior due to the higher amount of curatives. This result shows that the mechanical properties of the silica- and CB-filled SBR/NSO vulcanizates can be easily improved by controlling the curative recipes, achieving superior tensile properties than those of the silica- and CB-filled SBR/NO vulcanizates. It is also noticed that for the same type of oil, the silica-filled SBR vulcanizates exhibit a higher M100 and a slightly lower tensile strength compared to those of the CB-filled SBR vulcanizates. This is due to the higher surface area of the silica, a difference in rubber-filler interaction and also a higher crosslink density of the silica-filled SBR vulcanizates.

Table 4.3 also shows the hardness (Shore A) of various silica- and CB-filled SBR vulcanizates. Both silica- and CB-filled SBR vulcanizates exhibit a similar trend in hardness in terms of different oils. The silica- and CB-filled SBR/SO vulcanizates exhibit a lower hardness than those of the silica- and CB-filled SBR/NO vulcanizates. With the increase of the modification level, the hardness of the silica-filled SBR vulcanizates remains at a similar value of 64, while the hardness of CB-filled SBR/NSO vulcanizates is slightly decreased compared to that of the CB-filled SBR/SO vulcanizate. After adjusting the recipe, the silica-filled SBR/NO-3S vulcanizate exhibits a tremendous increased hardness of 74, while the silica-filled SBR/SO-3S, SBR/5NSO-3S, and SBR/11NSO-3S

vulcanizates exhibit a hardness of 69, 68 and 67, respectively, similar to that of the silica-filled SBR/NO vulcanizate. The CB-filled SBR/NO-3S vulcanizate exhibits a slightly increased hardness of 67, and the CB-filled SBR/SO-3S, SBR/5NSO-3S, and

SBR/11NSO-3S vulcanizates exhibit a hardness of 62, 60, and 61, respectively. These increases in hardness can be attributed to the increased crosslink density as shown in

Figure 4.3 (b). In general, the various silica-filled SBR vulcanizates show a higher hardness than those of the CB-filled SBR vulcanizates. This is due to the higher surface area of the silica and stronger filler-filler interaction than the CB and also a higher apparent crosslink density of the silica-filled SBR vulcanizates.

Figure 4.10 depicts the abrasion loss of various silica- and CB-filled SBR vulcanizates. In general, the silica-filled SBR vulcanizates show better abrasion resistance compared to those of the CB-filled SBR vulcanizates, except for the silica- and CB-filled

SBR/5NSO and SBR/11NSO vulcanizates exhibiting similar abrasion resistance. Results show that the silica- and CB-filled SBR/SO vulcanizates have better abrasion resistance than those of the silica- and CB-filled SBR/NO vulcanizates. This result is in agreement with previous studies.4, 14-15 With an increase of the modification level, the abrasion resistance of silica- and CB-filled and SBR/NSO vulcanizates is decreased due to their decreased crosslink densities. After adjusting the recipe, the silica- and CB-filled

SBR/NO-3S vulcanizates show little improvement in abrasion resistance. The silica- and

CB-filled SBR/SO-3S vulcanizates exhibit a slightly better abrasion resistance than the silica- and CB-filled SBR/SO vulcanizates due to the increase of the crosslink density.

However, the abrasion resistance of various silica- and CB-filled SBR/5NSO-3S and

SBR/11NSO-3S vulcanizates is tremendously improved compared to those of the

Figure 4.10. Abrasion loss of various silica- and CB-filled SBR vulcanizates.

SBR/NSO vulcanizates with original recipes. The CB-filled SBR/5NSO-3S vulcanizate exhibits a slightly better abrasion resistance than that of the CB-filled SBR/NO vulcanizate.

The silica-filled SBR/5NSO-3S and SBR/11NSO-3S all exhibit a better abrasion resistance than that of the silica-filled SBR/NO vulcanizates. Therefore, with the simple adjustment of recipe, the silica- and CB-filled SBR/NSO vulcanizates can have a better abrasion resistance along with improved tensile properties than the SBR/NO vulcanizates.

4.2.7 DMA Test and Performance Predictors

Figure 4.11 shows the temperature dependencies of E’ of silica-filled SBR vulcanizates (a) and CB-filled SBR vulcanizates (b), E” of silica-filled SBR vulcanizates (c) and CB-filled SBR vulcanizates (d), and also tan δ of silica-filled SBR vulcanizates (e) and

CB-filled SBR vulcanizates (f). It is observed that the silica-filled SBR vulcanizates show similar dynamic mechanical properties compared to those of the CB-filled SBR vulcanizates. The silica- and CB-filled SBR vulcanizates exhibit similar values of E’ in the glassy state, while the silica-filled SBR vulcanizates exhibit a slightly higher E’ at rubbery state than those of the CB-filled SBR vulcanizates due to the higher reinforcement of the silica. The E” of the CB-filled SBR vulcanizates in the glassy state is slightly higher than that of the silica-filled SBR vulcanizates, the silica- and CB-filled SBR vulcanizates show similar values of E” in the rubbery state. Both silica- and CB-filled SBR vulcanizates have a single tan δ peak. The silica-filled SBR vulcanizates show lower peak values of tan δ than those of the CB-filled SBR vulcanizates in the glass transition region. At higher temperature region, the silica-filled SBR vulcanizates exhibit an overall lower value of tan

δ values than those of the CB-filled SBR vulcanizates.

Figure 4.11. Storage modulus of silica-filled (a) and CB-filled (b) SBR vulcanizates, loss modulus of silica-filled (c) and CB-filled (d) SBR vulcanizates, and tan δ of silica-filled (e) and CB-filled (f) SBR vulcanizates as a function of temperature.

The tan δ values of the vulcanizates at certain temperatures could be used as a performance predictor such as wet traction and rolling resistance for tire rubbers.128 In this study, the tan δ values at 10 °C are used to predict the wet traction performance and the tan δ values at 60 °C are used to evaluate the rolling resistance of the rubber vulcanizates. Figure 4.12 shows the tan δ values of various silica- and CB-filled SBR vulcanizates at 10 °C (a) and 60 °C (b). From Figure 4.12 (a), the CB-filled SBR/NO vulcanizate has a higher tan δ value at 10 °C than that of the silica-filled SBR/NO vulcanizate, indicating a better traction performance. On the other hand, the silica-filled

SBR/SO and SBR/NSO show higher tan δ values at 10 °C than those of the CB-filled

SBR/SO and SBR/NSO vulcanizates. The CB-filled SBR/5NSO and SBR/11NSO vulcanizates show lower tan δ values at 10 °C than that of the CB-filled SBR/NO vulcanizates while the silica-filled SBR/5NSO and SBR/11NSO show higher tan δ values at 10 °C than that of the silica-filled SBR/NO vulcanizates. After adjusting the curing recipe, the silica- and CB-filled SBR/NO-3S vulcanizates show decreased tan δ values at

10 °C, while the CB-filled SBR/SO-3S, SBR/5NSO-3S, and SBR/11NSO-3S vulcanizates exhibit increased tan δ values at 10 °C. For the silica-filled SBR/SO-3S, SBR/5NSO-3S, and SBR/11NSO-3S, the tan δ values exhibit a slight decrease. The silica-filled

SBR/SO-3S vulcanizate shows the lowest wet traction predictor. Figure 4.12 (b) shows the tan δ values at 60 °C, a lower tan δ value at 60 °C indicates a lower rolling resistance for tire rubbers which will benefit the fuel economy. It is observed that all the silica-filled SBR vulcanizates exhibit lower tan δ values at 60 °C than those of the CB-filled SBR vulcanizates due to the stronger filler-filler interaction of silica-filled SBR vulcanizates which is the dominating effect at small strain amplitudes. The difference in gel fraction

Figure 4.12. Tan δ values of silica- and CB-filled SBR vulcanizates at 10 °C (a) and 60 °C (b).

between silica- and CB-filled SBR vulcanizates is insignificant to affect the tan δ values.

For both silica- and CB-filled vulcanizates, the usage of SO will increase the tan δ values at

60 °C, with an increase of the modification level, the SBR/NSO vulcanizates exhibit increasing tan δ values at 60 °C due to the decreased crosslink density. After adjusting the curing recipe, the tan δ values at 60 °C of both silica- and CB-filled SBR/NO-3S vulcanizates are decreased tremendously compared to those of the SBR/NO vulcanizates with less curatives. Moreover, the silica- and CB-filled SBR vulcanizates with 3 phr sulfur exhibit lower tan δ values at 60 °C than those of the SBR/NO vulcanizates. These results show that by adjusting the curing recipe, the SBR/NSO-3S vulcanizates can achieve a better wet traction and a lower rolling resistance simultaneously than the conventional petroleum oil plasticized SBR vulcanizates. It should be noticed that the silica-filled SBR/NSO-3S vulcanizates show an overall improvement in properties satisfying the “magic triangle” requirements (wet traction, rolling resistance and abrasion resistance), which can benefit the performance of the tires while being environmental friendly. The tan δ peak values for silica-filled SBR vulcanizates are lower than those of the CB-filled SBR vulcanizates due to their higher apparent crosslink density. For SBR vulcanizates containing the same type of filler, the increase in crosslink density leads to a decrease in the tan δ peak value due to reduced chain mobility by the presence of crosslink.

4.3 Conclusions

NSO is an environmental friendly, low cost and innovative rubber plasticizer. Its reactivity can improve the performance of the rubber product in various aspects. The SO and NSO can react with curatives and also silane coupling agent in the silica- and CB-filled

SBR systems. With an increase of the modification level, both silica- and CB-filled

SBR/NSO vulcanizates exhibit a slight increase in gel fraction and a significant decrease in crosslink density. The reaction between the SO, NSO and silane decreases the bound rubber fraction of the silica-filled SBR/SO and SBR/NSO compounds. Due to the higher surface area of the silica used in this study, the reinforcement of the silica is higher than that of the CB, resulting in a higher G’, G” and also complex viscosity of the silica-filled SBR compounds. Different types of oils have little effect on the rheological properties of the

SBR compounds. The filler-filler interaction of the silica is much stronger than that of the

CB, therefore the silica-filled SBR compounds exhibit a stronger Payne effect than those of the CB-filled SBR compounds. The silica- and CB-filled SBR compounds and vulcanizates containing SO and NSO exhibit much better thermal stability and lower the Tg compared to those containing NO. It was found that the silica- and CB-filled SBR compounds exhibit different curing behaviors due to the different surface chemistry. The silica- and CB-filled SBR/NSO vulcanizates exhibit improved tensile properties compared to those containing NO. After adjusting the recipe, the silica- and CB-filled SBR/NSO-3S vulcanizates exhibit excellent abrasion resistance. The DMA test shows that the wet traction and rolling resistance can be balanced by adjusting the recipes. The silica-filled

SBR/NSO-3S vulcanizates have better abrasion resistance, higher wet traction and lower rolling resistance simultaneously than that of the silica-filled SBR/NO vulcanizate. In conclusion, the usage of NSO in rubber products is an inexpensive and effective way to improve the performance of the product. Its unique features make it possible to replace the conventional petroleum rubber plasticizers.

CHAPTER V

COMPARISON OF CB/SILICA HYBRID FILLER-FILLED TIN-COUPLED SBR WITH VARIOUS OILS

5.1 Introduction

It is known that the usage of silica in rubber can enhance the tear strength, abrasion resistance, aging resistance and also adhesion properties. In tire treads, silica provides a lower rolling resistance and better wet traction than CB, silica-based tire tread is the current trend for tire manufacture.100 In practice, silica is usually used together with CB. CB/silica hybrid filler has been widely used in tire treads, conveyor belts, hoses, automotive parts and cable jackets.134 So far, there is no study reported the comparative behavior of different plasticizers in hybrid filler systems. Investigating the effect of NSO in hybrid filler-filled tin-coupled SBR is of tremendous importance for use in tires and can help to further develop NSO’s applications.

In this study, various characterizations of hybrid filler-filled tin-coupled SBR compounds and vulcanizates are conducted, including bound rubber fraction, gel fraction, crosslink density, rheological properties, thermal properties, curing behavior, tensile property, hardness, tear strength, abrasion resistance, aging properties and dynamic properties. Effect of different oils are investigated. The conducted research will help the application of green and sustainable technology in rubber compounds, especially for tires.

This Chapter is based on a paper published by the author of this Dissertation.135 Data of various CB-filled tin-coupled SBR are from author’s previous study.126

5.2 Results and discussion

5.2.1 Bound rubber, gel fraction and crosslink density

Figure 5.1 shows bound rubber fraction (a) of various SBR compounds, gel fraction

(b), and crosslink density (c) of various SBR vulcanizates. The data are compared with previous data on CB-filled and silica-filled SBR vulcanizates.126, 130 From Figure 5.1 (a), it is observed that the bound rubber fraction of CB-filled SBR compounds is higher than that of the silica-filled SBR compounds. The hybrid filler-filled SBR compounds exhibit a similar bound rubber fraction as the CB-filled SBR compounds. CB has better interaction with SBR than silica, the blend of CB/silica can also reduce filler networking, thus more rubber molecules can be adsorbed at the surface of fillers.81 Therefore, the bound rubber fraction of hybrid filler-filled SBR compounds exhibit higher values than those of the silica-filled SBR compounds. It is observed from Figure 5.1 (b) that various hybrid filler filled SBR vulcanizates exhibit gel fraction values between those of CB- and silica-filled

SBR vulcanizates. The hybrid filler-filled SBR/SO and two SBR/NSO vulcanizates exhibit higher gel fractions than that of the SBR/NO vulcanizate due to the covulcanization of the

SO and NSO with rubber at the presence of sulfur.126, 130 After increasing the sulfur level from 2 phr to 3 phr, the gel fractions of hybrid filler-filled SBR/SO-3S and two

SBR/NSO-3S vulcanizates increase to a higher value compared with SBR/SO and two

Figure 5.1. Bound rubber fraction (a) of SBR compounds, gel fraction (b) and crosslink density (c) of SBR vulcanizates.

SBR/NSO vulcanizates. On the contrary, the gel fraction of hybrid filler-filled SBR/NO-3S only shows a slight increase compared with that of the hybrid filler-filled SBR/NO vulcanizate. It can be concluded that more NSO molecules are effectively cured within the rubber network at the presence of higher amount of curatives. However, NO has minimal reactivity with curatives resulting in a little change in gel fraction. The covulcanization of oil with rubber network will reduce the potential migration of the oil, which can greatly enhance the storage stability. Figure 5.1 (c) shows the apparent crosslink densities of various hybrid SBR vulcanizates. Similarly, various hybrid filler-filled SBR vulcanizates exhibit intermediate values compared with CB- and silica-filled SBR vulcanizates. The hybrid filler-filled SBR/NO vulcanizates exhibit higher apparent crosslink densities than those of the hybrid filler-filled SBR/SO and two SBR/NSO vulcanizates. With an increase of the norbornylization level, the apparent crosslink densities of two hybrid filler-filled

SBR/NSO vulcanizates are further decreased. The decrease in crosslink density is due to the fact that the bio-based oils can consume the curatives.5, 13, 15, 126, 130 With an increase of the norbornylization level, the reactivity of the NSO increases, leading to more consumption of the curatives.126, 130 It is clearly seen that after adjusting the recipe, the hybrid filler-filled SBR/NO-3S, SBR/SO-3S and two SBR/NSO-3S vulcanizates exhibit higher apparent crosslink densities compared to those of the SBR vulcanizates with less curatives.

Figure 5.2 shows the relative crosslink density of various SBR/SO and SBR/NSO vulcanizates with different fillers. The relative crosslink density values of various

SBR/NSO vulcanizates are defined as the ratio of crosslink density of SBR/NSO vulcanizates to the crosslink density of SBR/SO vulcanizate containing same filler and

Figure 5.2. Relative crosslink density of various SBR/SO and SBR/NSO vulcanizates.

curing recipe. It is seen that the crosslink density of SBR/NSO vulcanizates with different fillers exhibits a similar exponentially decreasing trend. The data can be fitted to an empirical equation:

N = A ∗ exp(−B ∗ M) + C

Here, N is the relative crosslink density normalized to the value of the SBR/SO vulcanizate,

M is the modification level of NSO in percentage. Coefficient A, B, and C are related to the amount of sulfur and reactivity of the NSO. Here the values of A, B and C are 0.5744,

0.0712 and 0.4159, respectively. The value of C can be used to predict the crosslink density of SBR with NSO at the theoretically highest modification level. The fitting result is shown in Figure 5.2. From the fitting, it can be concluded that the types of fillers have little influence on the reaction between the NSO and curatives. The empirical equation can be used to predict the crosslink density of SBR/NSO vulcanizates at various modification levels and adjust the curatives accordingly.

5.2.2 Rheological properties

Figure 5.3 shows the dependence of G’ on the strain amplitude of SBR compounds with different fillers plotted in linear (a) and logarithm (b) scale. It is clearly observed that all SBR compounds show a Payne effect.133 Compared with CB- and silica-filled SBR compounds data from previous studies,126, 130 the hybrid filler-filled SBR compounds exhibit lower G’ values at low strain amplitudes than those of various silica-filled SBR compounds and CB-filled SBR/NO compound, similar to those of the CB-filled SBR/NSO compounds. The results demonstrate that the use of hybrid filler greatly reduce the filler-filler interaction. Similar effects were also reported by previous studies, indicating

Figure 5.3. Storage modulus vs. strain amplitude of various SBR compounds in linear (a) and logarithm (b) scale.

that the filler networking is suppressed by the use of hybrid filler.81, 116, 136-139 It is observed that for the same type of filler, SBR compounds containing NO exhibited the higher G’ values at large strain amplitudes than those with SO and NSO, indicating a higher rubber-filler interaction.

Figure 5.4 shows the frequency dependences of the G’ (a), G” (b), tan δ (c) and complex viscosity (d) of various SBR compounds. The data of hybrid filler-filled SBR compounds are compared with our previous data on CB- and silica-filled SBR compounds.126, 130 From Figure 5.4 (a) and (b), it is observed that the G’ and G” of silica-filled SBR compounds are much higher and also have less frequency dependence at lower frequency region. The G’ and G” values of hybrid filler-filled SBR compounds are more close to those of the CB-filled SBR compounds. Active fillers such as CB and silica form a filler network at high loading due to the strong van der Waals force or hydrogen bonds, particularly, silica forms a much stronger filler network than CB due to its surface structure.81,100 Since the data were obtained at a strain amplitude of 4.2%, the network can be broken-down and reform under strain rather slowly.140 A stronger filler network will result in higher G’ and G” values at low frequency region. Therefore, the G’ and G” values of silica-filled SBR compounds have less frequency dependence. It should be noted that the hybrid filler-filled SBR/NO compound exhibits a slightly higher G’ value at low frequency region than other hybrid filler-filled SBR compounds, which indicates that the use of SO and NSO further reduces the filler networking. The results can be correlated with the strain sweep data shown in Figure 5.3. Figure 5.4 (c) shows that all the compounds show a plateau in tan δ value at lower frequency and then decrease at higher frequency.

This is attributed to the star-shape molecular structure of the SBR used in this study. As

Figure 5.4. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various SBR compounds.

discussed above, the silica-filled SBR compounds have strong filler networks, leading to the lowest tan δ values. Hybrid filler-filled SBR compounds exhibit tan δ values between

CB- and silica-filled SBR compounds. Particularly, the hybrid filler-filled SBR/NO compound exhibits the lower tan δ value than other hybrid filler-filled SBR compounds at low frequency region, due to the presence of a stronger filler network. At higher frequency region, the tan δ values of various hybrid filler-filled SBR compounds start to converge, showing less difference in values. Similarly, Figure 5.4 (d) shows that the silica-filled SBR compounds exhibit a stronger tendency of yielding behavior. The values of hybrid filler-filled SBR compounds are close to those of the CB-filled SBR compounds.

In conclusion, the use of hybrid filler greatly reduces the filler networking compared with silica only, leading to less energy dissipation upon cyclic strains which can be beneficial in many applications. The use of SO and NSO only has slight effect on reducing filler networking in comparison with NO.

5.2.3 Thermal properties

Figure 5.5 shows the TGA curves of various hybrid filler-filled SBR compounds (a) and vulcanizates (b). It is observed that the SBR/NO compound and vulcanizate exhibit an inferior thermal stability compared with SBR/SO and SBR/NSO compounds and vulcanizates. SO and NSO are more thermally stable than the NO,126 thus the SBR compounds and vulcanizates containing SO and NSO exhibit similar and excellent thermal stability. Similar effect was also seen in previous studies.126, 130, 141 The usage of

SO and NSO can tremendously enhance processing safety.

Figure 5.6 shows the DSC curves of various hybrid filler-filled SBR compounds (a) and vulcanizates (b). All hybrid filler-filled SBR compounds and vulcanizates containing different oils exhibited smooth DSC curves and single Tg. Table 5.1 shows the Tg of various SBR compounds and vulcanizates obtained from DSC tests. The Tg of hybrid filler-filled SBR compounds and vulcanizates are compared with previous results on

CB-filled and silica-filled samples.126, 130

In general, hybrid filler-filled SBR compounds and vulcanizates exhibit Tg values between CB- and silica-filled SBR compounds and vulcanizates. The vulcanizates have higher Tg than uncured compounds due to the formation of crosslinked network. For both compounds and vulcanizates, SBR/NO have the highest Tg while SBR/SO have the lowest

126 Tg. This is due to the fact that NO has a higher Tg than SO. With the increase of the norbornylization level, the Tg of NSO increases, leading to an increasing Tg of SBR/NSO compounds and vulcanizates.126 After increasing the amount of curatives, various SBR vulcanizates show an increase in Tg compared to those of the SBR vulcanizates with less curatives. This is due to the increasing crosslink density as shown in Figure 5.1 (c). It is found that the usage of SO and NSO provides a lower Tg for SBR vulcanizates than the conventional NO, providing a better low temperature flexibility.

Figure 5.5. TGA curves of various hybrid filler-filled SBR compounds (a) and vulcanizates (b).

Figure 5.6. DSC curves of various hybrid filler-filled SBR compounds (a) and vulcanizates (b).

Table 5.1. Glass transition temperature of various SBR compounds and vulcanizates.

CB-filled Hybrid filler-filled Silica-filled

Compound Vulcanizate Compound Vulcanizate Compound Vulcanizate Tg Tg Tg Tg Tg Tg

(°C) (°C) (°C) (°C) (°C) (°C) SBR/NO -45.8 -38.8 -44.2 -39.5 -44.0 -40.0 SBR/SO -52.2 -42.3 -48.8 -44.1 -48.6 -45.1 SBR/5NSO -48.8 -41.2 -47.9 -42.1 -47.2 -43.0 SBR/11NSO -48.1 -40.5 -46.8 -41.0 -46.3 -42.8 SBR/NO-3S -45.8 -37.6 -44.2 -38.5 -44.0 -39.1 SBR/SO-3S -52.2 -41.6 -48.8 -43.2 -48.6 -43.9 SBR/5NSO-3S -48.8 -40.4 -48.8 -40.7 -47.2 -41.6 SBR/11NSO-3S -48.1 -39.9 -47.9 -40.1 -46.3 -40.7

5.3.4 Mooney viscosity and curing behaviors

Figure 5.7 shows the curing curves of various hybrid filler-filled SBR compounds.

The maximum torque (MH) values of various compounds reflect their crosslink density, the order of the MH values is the same as the crosslink density shown in Figure 5.1 (c). The usage of SO and NSO consumes the sulfur, leading to a lower state of cure and therefore lower MH value. With the increase of norbornylization level, more sulfur is consumed.

After increasing the sulfur to 3 phr, the state of cure of SBR compounds is largely increased.

All compounds exhibit a marching torque at post cure region due to the flocculation of silica, which was also seen in previous studies.99, 130 The MH and minimum torque (ML) values of CB-, silica- and CB/silica hybrid filler-filled SBR compounds are compared in

Figure 5.8.126, 130 It is clearly seen that the CB- and hybrid filler-filled SBR compounds exhibit similar ML values but lower than that of the silica-filled SBR compounds. The hybrid filler-filled SBR compounds exhibit intermediate MH values than those of the CB- and silica-filled SBR compounds, which is also seen in the crosslink density data shown in

Figure 5.1 (c).

Table 5.2 shows the Mooney viscosity, scorch time TS1, curing time T95 and CRI of various hybrid filler-filled SBR compounds. Results show that the hybrid filler-filled

SBR/NO compound exhibit the highest Mooney viscosity. At the same usage level of oil,

SO and NSO exhibit a better plasticization effect than NO, which will benefit rubber processing. However, the usage of SO decreases the scorch safety compared with NO.

After the modification, the scorch times of the hybrid filler-filled SBR/NSO compounds are extended, achieving a better scorch safety than that of the hybrid filler-filled SBR/NO compound. For curing time, the hybrid filler-filled SBR/SO compound exhibits a longer

Figure 5.7. Curing curves of various hybrid filler-filled SBR compounds.

100

Figure 5.8. Minimum (ML) and Maximum (MH) torque from curing curves of various SBR compounds.

101

Table 5.2. Mooney viscosity and curing characteristics of various hybrid filler-filled SBR compounds.

Compound Compound Compound Mooney CRI Ts1 T Viscosity 95 ML(1+4) 100°C (min) (min)

SBR/NO 50 4.3 22.9 5.49 SBR/SO 42 3.7 23.7 5.00 SBR/5NSO 42 4.5 23.5 5.26 SBR/11NSO 43 4.6 23.3 5.35 SBR/NO-3S 50 3.7 20.9 5.81 SBR/SO-3S 42 3.0 20.9 5.59 SBR/5NSO-3S 42 3.8 20.7 5.92 SBR/11NSO-3S 43 4.0 19.8 6.33

102

T95 while having a lower CRI. After the modification, the T95 of hybrid filler-filled

SBR/NSO compounds are seen to be shortened, while having a higher CRI. For the compounds with 3 phr of sulfur, the scorch time TS1 exhibits a similar trend as compounds with less sulfur. hybrid filler-filled SBR/11NSO-3S exhibits the best scorch safety, the shortest curing time and also the highest CRI among various hybrid filler-filled SBR-3S compounds. Therefore, the usage of NSO overcomes the negative effect of SO on curing, improves scorch safety and curing speed.

5.3.5 Mechanical properties and aging resistance

Figure 5.9 shows the stress-strain curves of various hybrid filler-filled SBR vulcanizates before (a) and after hot air aging (b). The addition of SO in hybrid filler-filled

SBR vulcanizate slightly increases the tensile strength and elongation at break than NO.

Similar effect was also observed by other researchers.4-5, 15 With the increase of norbornylization level, the hybrid filler-filled SBR/NSO vulcanizates exhibit tremendous increase in both elongation at break and tensile strength. The tensile test data are listed in

Table 5.3. The results demonstrate that the hybrid filler-filled SBR/11NSO vulcanizate exhibits a 92% increase in elongation at break and a 17% increase in tensile strength compared with SBR/NO vulcanizate. However, the moduli of various hybrid filler-filled

SBR/SO and SBR/NSO vulcanizates are lower than that of the SBR/NO vulcanizate due to the decreased crosslink density. After adjusting the curing recipe, various SBR-3S vulcanizates exhibit increased moduli and decreased elongations at break. However, the hybrid filler-filled SBR/NO-3S and SBR/SO-3S exhibit a large decrease in tensile strength compared with their vulcanizates containing less sulfur. Differently, two SBR/NSO-3S

103

Figure 5.9. Stress-strain curves of various hybrid filler-filled SBR vulcanizates before (a) and after (b) aging.

104

Table 5.3. Tensile properties, hardness and tear strength of various hybrid filler-filled SBR vulcanizates before and after aging and their change in percentage.

Original Properties M100 M300 E@B Tensile Strength Hardness Tear Strength

(MPa) (MPa) (%) (MPa) (Shore A) (kN/m)

SBR/NO 2.82 ± 0.02 11.38 ± 0.10 345.3 ± 9.7 13.4 ± 0.3 66 31.27 ± 0.59 SBR/SO 2.26 ± 0.04 8.40 ± 0.12 433.5 ± 15.9 14.3 ± 0.3 62 34.81 ± 1.27 SBR/5NSO 1.72 ± 0.01 5.74 ± 0.07 575.9 ± 7.8 15.6 ± 0.2 62 37.09 ± 0.55 SBR/11NSO 1.57 ± 0.03 5.03 ± 0.09 662.7 ± 10.8 15.7 ± 0.3 62 38.76 ± 0.69 SBR/NO-3S 3.92 ± 0.04 - 233.6 ± 10.5 11.8 ± 0.2 69 24.76 ± 1.85 SBR/SO-3S 2.96 ± 0.03 10.98 ± 0.04 320.1 ± 14.7 11.9 ± 0.8 66 27.15 ± 1.12 SBR/5NSO-3S 2.59 ± 0.06 9.05± 0.14 414.2 ± 8.2 15.5 ± 0.4 64 36.01 ± 0.91 SBR/11NSO-3S 2.42 ± 0.07 8.61 ± 0.11 461.7 ± 10.9 15.2 ± 0.5 64 38.30 ± 0.76 Aged 100°C 48h M100 M300 E@B Tensile Strength Hardness Tear Strength

(MPa) (MPa) (%) (MPa) (Shore A) (kN/m)

SBR/NO 4.31 ± 0.03 - 285.0 ± 9.3 15.1 ± 0.3 76 28.01 ± 1.65 SBR/SO 2.71 ± 0.04 10.27 ± 0.12 408.8 ± 5.5 13.9 ± 0.2 67 36.25 ± 0.73 SBR/5NSO 2.06 ± 0.07 7.04 ± 0.08 559.6 ± 3.8 15.8 ± 0.2 63 37.92 ± 1.54 SBR/11NSO 1.88 ± 0.02 6.04 ± 0.04 641.2 ± 8.9 15.4 ± 0.4 63 39.24 ± 0.65 SBR/NO-3S 6.03 ± 0.05 - 196.4 ± 8.6 13.6 ± 0.9 79 22.30 ± 1.39 SBR/SO-3S 3.80 ± 0.09 - 271.1 ± 10.5 11.2 ± 0.8 70 27.74 ± 1.46 SBR/5NSO-3S 3.06 ± 0.06 10.72 ± 0.09 409.2 ± 12.8 15.6 ± 0.5 67 36.38 ± 1.58 SBR/11NSO-3S 2.77 ± 0.07 9.82 ± 0.05 446.1 ± 13.4 15.0 ± 0.6 67 39.16 ± 1.54 Change in Percentage M100 M300 E@B Tensile Strength Hardness Tear Strength

(%) (%) (%) (%) (%) (%)

SBR/NO +52.8 - -17.5 +12.7 +15.2 -10.4 SBR/SO +19.9 +22.3 -5.7 -2.8 +8.1 +4.1 SBR/5NSO +19.6 +22.6 -2.8 +1.3 +1.6 +2.2 SBR/11NSO +19.7 +20.1 -3.2 -1.9 +1.6 +1.2 SBR/NO-3S +53.8 - -15.9 +15.2 +14.5 -9.9 SBR/SO-3S +28.4 - -15.3 -5.9 +6.1 +2.2 SBR/5NSO-3S +21.9 +19.4 -1.2 +0.6 +4.7 +1.0 SBR/11NSO-3S +19.4 +18.2 -3.4 -1.2 +4.7 +2.2

105

vulcanizates exhibit minimum changes in tensile strength. It shows that by adjusting the curing recipe, hybrid filler-filled SBR/NSO vulcanizates are able to achieve superior tensile performance than that of the SBR/NO and SBR/SO vulcanizates. Similar behavior was also seen in our previous studies on CB- and silica-filled SBR vulcanizates.126, 130

Figure 5.10 shows the comparison of M100 (a), M300 (b), elongation at break (c) and tensile strength (d) of various SBR vulcanizates with different fillers. Hybrid filler-filled

SBR vulcanizates with NO and SO exhibit intermediate tensile properties compared with those of the CB- and silica-filled SBR vulcanizates. However, two hybrid filler-filled

SBR/NSO vulcanizates exhibit higher M100, M300 and lower elongation at break than those of the CB- and silica-filled SBR/NSO vulcanizates. Tensile property can be affected by several factors such as crosslink density, filler dispersion and rubber-filler interaction, etc. Figure 5.1 (c) shows that the hybrid filler-filled SBR/NSO vulcanizates have crosslink densities in between the CB- and silica-filled SBR/NSO vulcanizates. Figure 5.3 shows that the hybrid filler-filled SBR compounds exhibit filler-filler interaction similar to the

CB-filled SBR compounds, which can be an indication of similar dispersion.116 Therefore, it is possible that the hybrid filler-filled SBR/NSO vulcanizates have a better rubber-filler interaction than the CB- and silica-filled SBR/NSO vulcanizates.

To verify this assumption, Mooney-Rivlin curves were used to evaluate the rubber-filler interaction:142

휎 2퐶 σ∗ = = 2퐶 + 2 (휆 − 휆−2) 1 휆

Here σ* is the reduced stress, λ is the extension ratio, C1 and C2 are parameters not functions of the tensile deformation. Figure 5.11 shows the σ* vs. λ-1 of various SBR vulcanizates. At small strain where λ-1 is close to 1, an obvious increase in σ* is observed

106

Figure 5.10. Comparison of M100 (a), M300 (b), elongation at break (c) and tensile strength (d) of various CB-, silica- and CB/silica hybrid filler-filled SBR vulcanizates.

107

Figure 5.11. Mooney-Rivlin curves of various SBR vulcanizates.

108

for various vulcanizates, this is due to the breakdown of filler network (Payne effect). It can be seen that silica-filled SBR vulcanizates always have a larger Payne effect than CB- and hybrid filler-filled SBR vulcanizates, which is in agreement with Figure 5.3. At larger strain where λ-1 < 0.6, with the decrease of λ-1, an upturn of σ* is observed. This is due to the rubber-filler interaction and limited extensibility of rubber chains attached between neighboring fillers. Similar behavior was also observed in previous researches.136, 143 A stronger rubber-filler interaction will restrict the rubber chains leading to an “earlier” increase of σ* (at smaller strain), a weaker rubber-filler interaction will result in slipping of rubber chains leading to a “later” increase of σ* (at larger strain).144 Therefore the upturn point of σ* can be used to compare rubber-filler interaction. For all vulcanizates, CB-filled vulcanizates have turning point earlier than silica-filled vulcanizates indicating a stronger rubber-filler interaction for SBR filled with CB. For hybrid filler-filled SBR vulcanizates,

SBR/NO and SBR/SO have an intermediate behavior between CB- and silica-filled vulcanizates. However, hybrid filler-filled SBR/NSO vulcanizates exhibit earlier and more obvious upturn than the CB- and silica-filled vulcanizates, indicating a stronger rubber-filler interaction. Therefore, the use of hybrid filler has synergistic effect on the moduli of SBR/NSO vulcanizates. After adjusting the recipe, similar effect is also seen on hybrid filler-filled SBR/NSO-3S vulcanizates, but the increase of crosslink density makes the synergistic effect less obvious.

Table 5.3 also shows the hardness (Shore A) and tear strength of various hybrid filler-filled SBR vulcanizates. The hybrid filler-filled SBR/NO vulcanizates exhibit a higher hardness than other vulcanizates. Hybrid filler-filled SBR/SO and two SBR/NSO vulcanizates have similar hardness. After adjusting the curing recipe, hybrid filler-filled

109

SBR/NSO-3S vulcanizates exhibit less change in hardness than those of the SBR/NO-3S and SBR/SO-3S vulcanizates. For tear strength, hybrid filler-filled SBR/SO vulcanizate shows higher tear strength than that of the SBR/NO vulcanizate. With the increase of norbornylization level, the tear strength of hybrid filler-filled SBR/NSO vulcanizates is increased. Similar effect was observed and discussed in our previous study,141 the increase in tear strength is related to a lower crosslink density and more polysulfidic crosslinks.145,146

Figure 5.9 (b) shows the stress-strain curves of various hybrid filler-filled SBR vulcanizates after hot air aging. The change in tensile properties, hardness and tear strength of various hybrid filler-filled SBR vulcanizates after hot air aging is listed in Table 5.3.

After aging, all vulcanizates exhibit an increase in M100, M300, hardness, tear strength, and a decrease in elongation at break. The tensile strength of aged hybrid filler-filled SBR vulcanizates containing NO exhibits an increase around 15%. But the aged hybrid filler-filled SBR vulcanizates with SO and NSO exhibit much less change in tensile strength. It can be clearly seen that SBR vulcanizates with NSO have less change in tensile properties, hardness and tear strength than those of the SBR vulcanizates with NO and SO.

The hot air aging of SBR is related to the oxidation of SBR molecules.147 Free radicals are generated further increasing the crosslink density of SBR vulcanizates.148 The latter leads to an increase of modulus and a decrease of elongation at break.149 During the oxidation, unsaturated groups in SBR such as C=C double bonds are vulnerable and can react with free radicals.147 However, the reactive C=C double bonds in NSO can act as effective free radical scavengers. As consequence, the use of NSO improves aging resistance than NO and also SO in SBR. Similar effect was also seen in our previous study.141

110

Figure 5.12 depicts the abrasion loss of various SBR vulcanizates. In general, hybrid filler-filled SBR vulcanizates have the best abrasion resistance, showing a synergistic effect. Similar effect was also seen by other researchers.114,115 For different oils, SO provides the best abrasion resistance which is in agreement with previous studies.4, 130 With an increase of the norbornylization level, the abrasion resistance of hybrid filler-filled

SBR/NSO vulcanizates is decreased due to their decreased crosslink densities. After adjusting the curing recipe, the abrasion resistance of two hybrid filler-filled SBR/NSO-3S vulcanizates is tremendously improved. However, the hybrid filler-filled SBR/NO-3S vulcanizates show negligible improvement in abrasion resistance. It can be concluded that with the adjustment of curing recipe, the hybrid filler-filled SBR/NSO vulcanizates are able to achieve a better abrasion resistance than the SBR/NO vulcanizates.

5.3.6 DMA and performance predictors

Figure 5.13 shows the DMA curves of various hybrid filler-filled SBR vulcanizates.

Hybrid filler-filled SBR vulcanizates containing different oils show similar behavior but show different glass transition temperature. The relationship is in agreement with Tg data shown in Table 5.1 and based on the DSC data. Figure 5.14 shows the tan δ values of various SBR vulcanizates at 10 °C (a) and 60 °C (b) from DMA tests. The tan δ values at

10 °C and 60 °C are used to predict wet traction and rolling resistance of tire made of

SBR vulcanizates, respectively.128 For tan δ values at 10 °C, SBR vulcanizates with SO always show the lowest values, and with the increase of the norbornylization level, the tan

δ values of SBR/NSO vulcanizates are increased, indicating an improved wet traction.

After adjusting the curing recipe, hybrid filler-filled SBR/NSO vulcanizates are able to

111

Figure 5.12. Abrasion loss of various hybrid filler-filled SBR vulcanizates.

112

achieve a better wet traction than SBR vulcanizates with NO and SO. In terms of different filler systems, no clear trend can be observed. For tan δ values at 60 °C, most hybrid filler-filled SBR vulcanizates exhibit intermediate values lying between CB- and silica-filled SBR vulcanizates, except hybrid filler-filled SBR/5NSO and SBR/11NSO vulcanizates. Hybrid filler-filled SBR/5NSO and SBR/11NSO vulcanizates show lower tan δ values at 60 °C compared with the other two filler systems, indicating a lower rolling resistance. It is possibly due to the better rubber-filler interaction which is also seen in Figure 5.10. After adjusting the curing recipe, all vulcanizates exhibit decreased tan δ values at 60 °C due to the increased crosslink density. The SBR/NO-3S vulcanizates show the lowest tan δ values at 60 °C. It should be noticed that hybrid filler-filled

SBR/5NSO-3S and SBR/11NSO-3S vulcanizates exhibit lower tan δ values at 60 °C than that of the hybrid filler-filled SBR/NO vulcanizate with original recipe and higher tan δ values at 10 °C. With adjustment of curing recipe, hybrid filler-filled SBR/NSO vulcanizates are able to achieve lower rolling resistance, higher wet traction, and also better abrasion resistance simultaneously.

5.3 Conclusions

NSO is a bio-based, green and sustainable rubber plasticizer. Besides excellent plasticization effect, its reactivity can improve the performance of rubber vulcanizates in various aspects. The use of CB/silica hybrid filler in SBR compounds can lower the filler-filler interaction and improve the abrasion resistance than using single type of filler.

Compared with NO, NSO can provide superior mechanical properties, faster curing, desired thermal properties and better aging resistance and processing safety for SBR

113

Figure 5.13. Storage modulus (a), loss modulus and tan δ of various hybrid filler-filled SBR vulcanizates as a function of temperature.

114

Figure 5.14. Tan δ values at 10 °C (a) and 60 °C (b) of various SBR vulcanizates from DMA test.

115

compounds. Particularly, the addition of NSO in CB/silica hybrid filler-filled SBR vulcanizates also shows synergistic effect in tensile modulus and rolling resistance. With the adjustment of curing recipe, hybrid filler-filled SBR/NSO vulcanizates can achieve lower rolling resistance, better traction and abrasion resistance simultaneously than

SBR/NO vulcanizate.

116

CHAPTER VI

CB-FILLED CR WITH VARIOUS OILS CURED BY METAL OXIDE

6.1 Introduction

It is known from previous studies that the reactive SO or NSO could consume sulfur during the vulcanization of rubbers.5, 15, 126 However, the CR can be cured using a metal oxide system, which has a very different mechanism from the sulfur curing system. The reaction of SO or NSO in a metal oxide curing system has not yet been reported and the effect of SO and NSO in such curing system could be very interesting.

In the present study, 11NSO and 33NSO are used in CR compounds to extend a further utilization of the renewable and environmentally friendly oil by rubber industry.

The aim is to investigate and compare the effect of NO, SO and two level modifications

NSO on CB-filled CR compounds and vulcanizates. The rheological properties, gel fraction, crosslink density, bound rubber fraction, thermal properties, curing behavior, tensile property, hardness, abrasion resistance, tear strength and aging properties are studied.

This Chapter is based on a paper published by the author of this Dissertation.141

117

6.2 Results and discussion

6.2.1 Rheological properties

Figure 6.1 shows the dependence of G’ (a), G” (b), and tan δ (c) on the strain amplitude of various CB-filled CR compounds. From Figure 6.1 (a), it is observed that all

CB-filled CR compounds exhibit a reduction of G’ with the increase of strain amplitude, which is also known as the Payne effect.133 The CB-filled CR/SO and both CR/NSO compounds show similar G’ values at low strain amplitudes. However, the CB-filled

CR/NO compound exhibits a slightly higher G’ at low strain amplitudes compared to those of the CB-filled CR/SO and CR/NSO compounds, indicating a stronger filler-filler interaction. At high strain amplitudes, all compounds exhibit similar values of G’, which indicates a similar rubber-filler interaction. Figure 6.1 (b) shows that the G” values of various CB-filled CR compounds exhibit a small peak at a strain amplitude around 0.5%, and then the G” values decrease rapidly with further increase in strain amplitudes. The value of G” is dominated by the energy loss during dynamic strain, it is controlled by the breakdown and reformation of the filler network.46 The CB-filled CR/NO compound exhibits a higher G” value than other compounds at low strain amplitudes, indicating a more developed filler network breaking down and reforming during the dynamic strain.

Figure 6.1 (c) shows the tan δ of various CB-filled CR compounds. The value of tan δ is dominated by the state of filler network.81 It can be observed that all compounds exhibit similar behavior. A peak of tan δ at a strain amplitude around 1.25% is observed, which corresponds to the breakdown of CB network. At high strain amplitudes, the tan δ values increase rapidly with the increase of strain amplitude, indicating a further breakdown of rubber structure, such as rubber-filler interaction. It is possible that the usage of SO and 118

Figure 6.1. Storage (a), loss (b) moduli and tan δ (c) as a function of strain amplitude of various CB-filled CR compounds at 90 °C.

119

NSO can help the dispersion of CB in the CR matrix, which reduces the filler-filler interaction.

Figure 6.2 shows the frequency dependences of the G’ (a), G” (b), tan δ (c) and complex viscosity (d) of various CB-filled CR compounds. From Figure 6.2 (a) and (b), it is observed that the CB-filled CR compounds containing different oils exhibit similar frequency dependence of G’ and G”. In particular, the CB-filled CR/NO compound exhibit the highest G’ at low frequency region. This is due to the strong filler-filler network which can restrain the relaxation of the polymer chains, leading to a higher G’ at low frequency region.132 This result is in agreement with the strain sweep data shown in Figure 6.1. Figure

6.2 (c) depicts the tan δ as a function of frequency of various CB-filled CR compounds. It is observed that various CB-filled CR compounds show similar tan δ values. All the compounds exhibit a plateau at lower frequency region, and a decrease at higher frequency region, indicating that the materials are approaching the rubbery plateau. Figure 6.2 (d) shows the complex viscosity as a function of frequency of various CB-filled CR compounds. Various CR compounds exhibit a significant decrease of the complex viscosity with the increase of frequency. The addition of different oils gives negligible differences in the complex viscosity values. It can be concluded that the usage of SO and

NSO has little influence on the rheological properties of the CB-filled CR compounds, indicating the addition of SO and NSO will not change the processability of CB-filled CR compounds during the processing.

120

Figure 6.2. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) as a function of frequency of various CB-filled CR compounds at 90 °C.

121

6.2.2 Gel fraction, crosslink density and bound rubber

Figure 6.3 shows gel fraction (a), crosslink density (b) of various CB-filled CR vulcanizates and bound rubber fraction (c) of various CB-filled CR compounds. It is observed from Figure 6.3 (a) that the CB-filled CR/NO, CR/SO and CR/11NSO vulcanizates have similar gel fraction, and with the increase of the modification level, the

CR/33NSO vulcanizate exhibits a slightly higher gel fraction. Figure 6.3 (b) shows that the

CB-filled CR/NO vulcanizate exhibits the highest crosslink density. The CB-filled CR/SO and CR/11NSO vulcanizate have a similar crosslink density which is slightly lower than that of the CR/NO vulcanizate. The CB-filled CR/33NSO vulcanizate exhibit the lowest crosslink density. It is observed from Figure 6.3 (c) that the CB-filled CR/NO exhibit the highest bound rubber fraction, followed by the CB-filled CR/SO compound. With the increase of the modification level, the CB-filled CR/NSO compounds exhibit slightly decreased bound rubber fractions. The previous studies showed that the SO and NSO could react with curatives, slightly increase the gel fraction and decrease the crosslink density of diene rubber vulcanizates compared to rubber vulcanizates containing conventional petroleum oils.5, 15, 126 The increase of the gel fraction and decrease of the crosslink density of CB-filled CR/NSO vulcanizates is possibly due to the reaction of the highly reactive

NSO with the sulfur and also the thiuram disulfide in the CR gum. The possible reaction will be discussed later.

6.2.3 Thermal properties

Figure 6.4 shows the TGA curves of CR gum (a), CB-filled CR compounds (b) and

CB-filled CR vulcanizates (c). From Figure 6.4 (a), it is observed that the CR gum starts to 122

Figure 6.3. Gel fraction (a), crosslink density (b) of various CB-filled CR vulcanizates and bound rubber fraction (c) of various CB-filled CR compounds.

123

Figure 6.4. TGA curves of CR gum (a), various CB-filled CR compounds (b) and vulcanizates (c).

124

slowly lose weight at 200 °C and then starts to degrade rapidly at about 350 °C. Above

500 °C, the CR gum has a second stage of slow degradation, and at 700 °C the CR gum has a residue of 16%. The first stage of weight loss is due to the dehydrochlorination and then the autocatalytic chloroprene chain pyrolysis of CR, which was well studied by

Kaiersberger et al..150 The second stage of weight loss at 500 °C is due to the high pyrolysis shoot of CR. Similar effect was also observed in the previous study.151 Figure 6.4 (b) shows the TGA curves of CB-filled CR compounds containing various oils. The CB-filled CR compounds exhibit the thermal behavior similar to the CR gum. However, the CB-filled

CR/NO compound exhibits a significant weight loss at a lower temperature due to the low thermal stability of the NO. On the contrary, the CB-filled CR/SO and CR/NSO compounds exhibit much better thermal stability than that of the CR/NO compound before the rapid degradation of the rubber matrix at high temperatures. The better thermal stability of the SO and NSO was also reported in previous studies.5, 126 It is observed from Figure

6.4 (c) that the CB-filled CR/SO and CR/NSO vulcanizates exhibit a slightly better thermal stability below 350 °C than the corresponding CB-filled CR compounds. This is probably due to the addition of SO, NSO, steric acid, ZnO and MgO that can slow down the autocatalytic decomposition of the rubber matrix. Similar to that of the CB-filled CR/NO compound, the CB-filled CR/NO vulcanizate starts to lose weight at about 170 °C due to the evaporation of the NO. After the temperature reaches 350 °C, all the CB-filled CR vulcanizates exhibit similar weight loss process. In general, the addition of SO and NSO can enhance the thermal stability of CB-filled CR compounds and vulcanizates compared with conventional NO plasticizer. This advantage will improve the safety of the processing and application of rubber.

125

Figure 6.5 shows the DSC curves of various CB-filled CR compounds (a) and vulcanizates (b). Table 6.1 shows the glass transition temperature Tg determined from the

DSC curves. It is observed from the DSC curves that various CB-filled CR compounds and vulcanizates exhibit similar thermal behavior upon heating. From Table 3, it is indicated that the CB-filled CR compound and vulcanizate containing NO exhibit the highest Tg compared with the CR compounds and vulcanizates containing SO and NSO.

The CB-filled CR/SO compound and vulcanizate exhibit the lowest Tg. Similar effect was also observed in the previous study on EPDM/SO,5 which showed that SO had better plasticization effect than conventional petroleum-based plasticizer. With the increase of the modification level, the Tg of CB-filled CR/NSO compounds and vulcanizates is increased. This effect is due to the modification of the SO which introduces norbornyl groups on the fatty acid chain. Similar effect was also observed in the previous study.126 It is also noticed that the Tg of the CB-filled CR vulcanizates slightly increases compared to those of the CB-filled CR compounds due to the vulcanization. In general, the addition of

SO and NSO to the CB-filled CR compounds and vulcanizates can achieve a lower Tg value than NO, which can benefit the low temperature flexibility of the CR product.

6.2.4 Curing behaviors

Figure 6.6 shows the curing curves of various CB-filled CR compounds with curatives (a) and CB-filled CR compounds and pure CR gum without curatives (b). Table

6.2 shows the scorch time Ts1, curing time T95 and CRI calculated from the curing curves shown in Figure 6.6 (a). It is observed from this figure that all the CB-filled CR compounds exhibit a similar scorching time and minimum torque (ML). The CB-filled CR/NO

126

Figure 6.5. DSC curves of various CB-filled CR compounds (a) and vulcanizates (b).

Table 6.1. Glass transition temperature of various CB-filled CR compounds and vulcanizates and curing time of various CB-filled CR compounds.

Compounds Tg (°C) Vulcanizates Tg (°C) CR/NO -44.3 -43.9 CR/SO -48.6 -48.0 CR/11NSO -47.7 -47.1 CR/33NSO -45.2 -44.3

127

compound exhibits the highest maximum torque (MH), while the CB-filled CR/SO and

CR/NSO compounds exhibit similar values of MH which are lower than that of the

CB-filled CR/NO compound. Various CB-filled CR compounds exhibit similar Ts1. The

CB-filled CR/SO compound has a T95 of 19.2 min, slightly shorter than that of the

CB-filled CR/NO compound (19.7 min). With the increase of the modification level, the

CB-filled CR/11NSO and CR/33NSO compounds have a shorter T95, with times being 17.3 min and 12.0 min, respectively. The CRI of CB-filled CR compounds containing NSO is significantly higher than those with SO and NO.

There are two possible reasons why the CB-filled CR/NO compound exhibits a higher

MH than those of the CB-filled CR/SO and CR/NSO compounds. First, the filler-filler interaction of CB-filled CR/NO compound is slightly higher than those of the CR/SO and

CR/NSO compounds and gives a higher torque, which was reflected by a higher value of G’ in Figure 6.1 (a). Second, the reactive SO and NSO can react with sulfur and thiuram disulfide in the CR gum, leading to a lower crosslink density, which was shown in Figure

6.3 (b). In order to prove this hypothesis, CR gum and CR compounds containing different oils were heated to a temperature of 160 °C corresponding to the curing test. Similar to the previous study,50 the present study shows that the crosslinking of CR occurs by sulfur copolymerized into CR during its manufacturing, as shown in the curves in Figure 6.6 (b).

In the absence of ZnO/MgO in curing system, the CB-filled CR compounds still can be partially cured by the sulfur and thiuram disulfide in the CR gum. The CB-filled CR/NO compound exhibit the highest torque, followed by the CR/SO compound. With an increase of the modification level, the CR/11NSO and CR/33NSO exhibit a lower maximum torque.

This is due to the increasing consumption of sulfur with

128

Figure 6.6. Curing curves of various CB-filled CR compounds with (a) and without (b) curatives.

Table 6.2. Ts1, T95 and CRI of various CB-filled CR compounds.

T (min) T (min) CRI s1 95 CR/NO 1.2 19.7 5.4 CR/SO 1.1 19.2 5.5 CR/11NSO 1.1 17.3 6.2 CR/33NSO 1.1 12.0 9.2

129

the increasing modification level of NSO.126 The observation is in agreement with the gel fraction and crosslink density data shown in Figure 6.3.

6.2.5 Mechanical properties

Figure 6.7 shows the stress-strain curves of various CB-filled CR vulcanizates before aging (a) and after aging (b). The data of M100, M300, elongation at break, tensile strength, tear strength and hardness before and after the aging of various CB-filled CR vulcanizates are shown in Table 6.3. Also the relative changes of properties after aging are shown in this table in percentages. From Figure 6.7 (a) and Table 6.3, it is observed that the CB-filled

CR/NO vulcanizate has the highest modulus, the CB-filled CR/SO and CR/11NSO vulcanizates have similar values of modulus, but being lower than that of the vulcanizate containing NO. The CB-filled CR/33NSO vulcanizate has the lowest modulus. Meanwhile, the CB-filled CR/SO vulcanizate exhibits higher elongation at break than that of the

CR/NO vulcanizate, and with the increase of the modification level, the CB-filled CR/NSO vulcanizates exhibit higher elongation at break. The CB-filled CR/11NSO and CR/33NSO vulcanizates exhibit a slightly higher tensile strength than those of the CB-filled CR/NO and CR/SO vulcanizates. These effects are due to the reactions of the SO and NSO with sulfur during the vulcanization, which slightly decrease the crosslink density, as shown in

Figure 6.3 (b). Similar effects of SO were also seen in previous studies on SBR and

SBR/BR/NR blends.4, 15, 126

130

Figure 6.7. Stress-strain curves of various CB-filled CR vulcanizates before aging (a) and after aging (b).

131

Table 6.3. Tensile properties, tear strength and hardness of various CB-filled CR vulcanizates before and after aging and their relative change.

Original Properties M100 M300 E @ B Tensile Strength Tear Strength Hardness (MPa) (MPa) (%) (MPa) (kN/m) (Shore A) CR/NO 5.69±0.13 - 288.0±1.8 21.3±0.3 38.52±0.79 74 CR/SO 4.96±0.09 19.99±0.20 321.7±7.0 21.0±0.3 40.45±1.04 71 CR/11NSO 4.99±0.07 19.59±0.14 355.8±8.4 22.2±0.4 43.13±1.10 72 CR/33NSO 4.79±0.14 19.53±0.26 362.0±9.6 22.2±0.5 45.69±0.73 73 Aged Properties (100 °C 48h) M100 M300 E @ B Tensile Strength Tear Strength Hardness (MPa) (MPa) (%) (MPa) (kN/m) (Shore A) CR/NO 6.42±0.09 - 281.5±6.7 20.2±0.3 37.29±0.92 78 CR/SO 5.33±0.03 19.81±0.10 313.3±1.6 20.5±0.1 40.00±1.23 74 CR/11NSO 5.20±0.07 19.49±0.15 344.1±11.6 21.6±0.5 42.55±0.77 73 CR/33NSO 5.05±0.03 19.38±0.14 350.6±8.6 21.6±0.6 44.58±0.34 74 Percentage of Change M100 M300 E @ B Tensile Strength Tear Strength Hardness (%) (%) (%) (%) (%) (%) CR/NO +12.8 - -2.3 -5.2 -3.2 +5.4 CR/SO +8.1 -0.9 -2.6 -2.4 -1.1 +4.2 CR/11NSO +4.2 -0.5 -3.3 -2.7 -1.3 +1.4 CR/33NSO +5.4 -0.8 -3.1 -2.7 -2.4 +1.4

132

It is found that the CB-filled CR/SO vulcanizate exhibit a higher tear strength than that of the CB-filled CR/NO vulcanizate. With the increase of the modification level, the

CB-filled CR/NSO vulcanizates exhibit progressively higher tear strength. The CB-filled

CR/33NSO vulcanizate has a tear strength of 45.7 kN/m and is 18.6% higher than that of the CB-filled CR/NO vulcanizate. The tear strength of rubber vulcanizates is related to the filler dispersion, crosslink density, and crosslink type. Direct information about CB dispersion in various samples of CR is not available. However, the study145 showed the effect of crosslink density and crosslink type on tear strength. Results indicated that with the decrease of the crosslink density, the tear strength of rubber vulcanizates first increased and then decreased. In addition, the study146 showed that the inherent flaw size increased with the decrease of crosslink density increasing ability to blunt the crack tip. The

CB-filled CR/SO vulcanizate has a lower crosslink density than that of the CB-filled

CR/NO vulcanizate, and with the increase of modification level, the crosslink density is further decreased. The slightly decreased crosslink density of CB-filled CR/NSO vulcanizates can enhance the tear strength of vulcanizates. The study145 also found that rubber vulcanizates containing more polysulfidic crosslinks exhibited higher tear strength than those with shorter crosslinks. Our previous study126 suggested that the use of NSO in

SBR may alter the crosslink type, creating more flexible crosslinks. Therefore, the use of

NSO in CB-filled CR vulcanizates can also increase their tear strength.

For the hardness of the various CR vulcanizates, the CB-filled CR/SO vulcanizate exhibits the lowest hardness with the value being 71 Shore A. With the increase of the modification level, the hardness of various CB-filled CR/NSO vulcanizates is increased gradually, but it is still lower than that of the CB-filled CR/NO vulcanizate.

133

To study the aging of various CB-filled CR vulcanizates, the properties described above before and after the hot air aging are compared. It is seen that after the aging, the

M100 of various CB-filled CR vulcanizates is increased and the elongation at break and tensile strength are decreased. This is a typical behavior due to the post curing effect, similar behavior was also observed in previous studies on CR/NR blends58 and CR152. It is observed from Table 4 that the CB-filled CR/NO vulcanizate exhibits the largest change in all the properties after aging, except the elongation at break. The CB-filled CR/NSO vulcanizates exhibit less change in M100 than that of the CR/SO vulcanizate. The change of M300 of various CB-filled CR/SO and CR/NSO is negligible. The CB-filled CR/NSO vulcanizates exhibit a slightly larger change in elongation at break than that of the CR/NO vulcanizate, but they have a smaller change in tensile strength. With the increase of the modification level, the CB-filled CR/NSO vulcanizates exhibit a slightly increased change in tear strength, but they still outperform the CR/NO vulcanizate. The CB-filled

CR/11NSO and CR/33NSO vulcanizates exhibit the best hardness consistency during the aging among all the samples. In addition, the volume change after the aging was measured.

The volume of CB-filled CR/NO vulcanizate decreases by about 4.8%, while the volume of CR/SO, CR/11NSO and CR/33NSO vulcanizates only slightly increases, by about 0.3%,

0.3% and 0.2%, respectively. It is obvious that the CB-filled CR/SO and CR/NSO vulcanizates have a much better volume consistency during the aging. Clearly, the addition of SO and NSO can enhance the aging resistance.

134

Figure 6.8. Abrasion loss of various CB-filled CR vulcanizates.

135

Figure 6.8 depicts the abrasion loss of various CB-filled CR vulcanizates. It is noticed that the CB-filled CR/NO vulcanizate has the highest abrasion loss among all the samples.

The CB-filled CR/SO and CR/NSO vulcanizates exhibit a similar level of abrasion resistance being better than that of the CB-filled CR/NO vulcanizate. In general, the addition of NSO to CR can increase the tensile properties, tear strength, aging resistance and also abrasion resistance compared with CR with conventional NO, without changing other compounding ingredients.

6.3 Conclusions

NSO is a novel rubber plasticizer. It has the advantages of being the environmental friendly and low cost. It enhances the CR rubber performance without an additional cost.

The addition of SO and NSO in CB-filled CR compounds can provide lower filler-filler interaction without changing the processability of the compounds. The SO and NSO can react with the CR modifiers, such as sulfur and thiuram disulfide, during the vulcanization, slightly increase the gel fraction and decrease the crosslink density. The use of SO and

NSO also slightly decreases the bound rubber fraction of the CB-filled CR compounds.

The CB-filled CR compounds and vulcanizates containing SO and NSO exhibit much better thermal stability compared to those containing conventional NO, improving the safety during processing and application. Also the SO and NSO contribute to a lower Tg of the CR vulcanizates, which can improve the low temperature performance. With the increase of the modification level of the NSO, the curing time of the CB-filled CR/NSO compounds is shortened compared with the CB-filled CR/NO compound. The CB-filled

CR/NSO vulcanizates exhibit better tensile properties and tear strength compared with the

136

CB-filled CR/NO and CR/SO vulcanizates. The hardness of CB-filled CR/NSO vulcanizates is slightly increased with the increase of the modification level. The CB-filled

CR vulcanizates containing SO and NSO exhibit better aging resistance compared with the

CR vulcanizate containing NO. Also, the addition of SO and NSO to the CB-filled CR provides vulcanizates with a better abrasion resistance than that of the NO. In conclusion, the usage of NSO in CR rubber products is an inexpensive and effective way to improve their performance. Its unique features make it an excellent potential candidate to replace the conventional petroleum rubber plasticizers.

137

CHAPTER VII

CB-FILLED IIR WITH VARIOUS OILS CURED BY PHENOLIC RESIN

7.1 Introduction

Since IIR compounds are widely used in tire industry and their wastes could be a potential environmental problem, the use of NSO in butyl rubber could be an excellent solution to have a lower carbon footprint. Besides, the phenolic resin curing system has a different curing mechanism than sulfur or metal oxide curing system, it is necessary to expand the research of NSO in the resin curing system.

In this study, IIR compounds and vulcanizates containing NO, SO, 5NSO and 11NSO are investigated. Two recipes with different oil levels are used. Properties including gel fraction, crosslink density, thermal and rheological properties, curing behavior, tensile properties, hardness, and aging resistance are compared. The research helps to expand the application of green and sustainable technology to the rubber used in high temperature environment.

7.2 Results and discussion

7.2.1 Gel fraction, crosslink density and bound rubber fraction

Figure 7.1 shows gel fraction (a) and crosslink density (b) of various CB-filled IIR vulcanizates and bound rubber fraction of various CB-filled IIR compounds. It is observed

138

from Figure 7.1 (a) that IIR vulcanizates with 10 phr of oil exhibit overall higher gel fraction than those with 15 phr of oil. Incorporation of different oils does not strongly affect the gel fraction of vulcanizates, while it has a strong effect on the crosslink density. At the same oil level, the CB-filled IIR/SO vulcanizates exhibit a slightly lower gel fraction than that of the CB-filled IIR/NO vulcanizate. With the increase of the modification level, the

CB-filled IIR/NSO vulcanizates exhibit a slightly increased gel fraction. Figure 7.1 (b) shows that the usage of SO dramatically decreases the crosslink density of vulcanizate in comparison with IIR vulcanizate containing NO. With the increase of modification level, the crosslink density of CB-filled IIR/NSO vulcanizates decreases even further. It is very possible that the SO and NSO react with curatives, which is phenolic resin in this case, leading to a lower crosslink density. Similar effect was also seen for the sulfur-cured SBR vulcanizates in the previous study.126 After changing the oil level from 15 phr to 10 phr, the crosslink density of CB-filled IIR/SO-2 and two IIR/NSO-2 vulcanizates is increased. This is due to the less consumption of curatives by the reactive SO and NSO at lower concentration of oil. However, the crosslink density of CB-filled IIR/NO-2 vulcanizate exhibits only a slight change, indicating that the NO level has a less influence on the crosslink density of IIR vulcanizates than the SO and NSO have. Figure 7.1 (c) shows the bound rubber fraction of various CB-filled IIR compounds. It is seen that CB-filled IIR compounds with SO and NSO have

139

Figure 7.1. Gel fraction (a) and crosslink density (b) of various CB-filled IIR vulcanizates and bound rubber fraction of various IIR compounds (c).

140

less bound rubber fraction than those with NO. Decreasing oil level from 15 to 10 phr exhibits a limited influence on the bound rubber fraction.

To further study the reaction of SO/NSO with phenolic resin, oil/resin mixtures are prepared and investigated. Figure 7.2 shows the DSC curves of different oils (a) and oil/resin mixtures (b) at 160 °C. It can be seen from Figure 7.2 (a) that various oils exhibit no significant change in thermal behavior after 30 s at 160 °C. The initial “bumps” are due to the initial change of heating rate. However, Figure 7.2 (b) shows that the oil/resin mixtures exhibit obvious exothermic behavior indicating that a curing reaction occurs between the phenolic resin and oils. With the increase of the modification level, the exothermic peak becomes more distinct. It is very likely that the reactive oils are crosslinked by the resin and this reaction can be detected by measurements of the viscosity of oil/resin mixtures. Figure 7.3 shows shear viscosity vs. shear rate of various oil/resin mixtures. All samples exhibit Newtonian behavior. The unreacted mixtures have overall lower viscosities than those of the reacted ones. After the reaction, the viscosity of

SO/resin, 5NSO/resin and 11NSO/resin mixture increases by 290%, 419% and 440%, respectively. It can be concluded that the SO and NSO react with phenolic resin and the increase of modification level enabled the NSO to have more reaction with the phenolic resin. During the curing process of IIR compounds, the curative is partially consumed by the oils, leading to a lower crosslink density as shown in Figure 7.1 (b). Similar effect was also seen in our previous study on oil/sulfur and oil/silane mixtures.126, 130

141

Figure 7.2. DSC curves of various oils (a) and oil/resin mixtures (b) at 160 °C.

142

Figure 7.3. Shear viscosity of various oil/resin mixtures at room temperature.

143

7.2.2 Rheological properties

Figure 7.4 shows the dependence of G’ of various CB-filled IIR compounds on the strain amplitude plotted in logarithm (a) and linear (b) scale. As expected, CB-filled IIR compounds containing 15 phr of oils exhibit overall lower G’ values than those with 10 phr oils, due to more plasticizers. Different types of oils have little influence on the Payne effect133 of IIR compounds indicating similar filler-filler and filler-rubber interactions.

Figure 7.5 shows the G’ (a), G” (b), tan δ (c) and complex viscosity (d) vs. frequency of various CB-filled IIR compounds and vulcanizates. CB-filled IIR compounds with 15 phr of oils exhibit lower G’ and G” values than those with 10 phr of oils. Different types of oils have a limited influence on the dynamic properties. It is also seen that the CB-filled IIR vulcanizates containing NO exhibit higher G’ and G” values, and a lower tan δ value than those of the vulcanizates containing SO and NSO. This is due to higher crosslink density, as was shown earlier in Figure 7.1 (b). It should be noticed that with the increase of the modification level, the CB-filled IIR/NSO vulcanizates exhibit a decreasing value of G’ at low frequency range and an increasing tan δ value. This is due to the decreasing crosslink density of IIR/NSO vulcanizates. After decreasing the oil level to 10 phr, the G’ and G” values of various CB-filled IIR vulcanizates are increased. The tan δ values of CB-filled

IIR/SO-2 and IIR/NSO-2 vulcanizates are decreased due to the crosslink density increase, except for the CB-filled IIR/NO-2 vulcanizate. Different oils show a limited influence on the complex viscosity of CB-filled IIR vulcanizates. IIR vulcanizates containing 10 phr oil exhibit a slightly higher complex viscosity value than those with 15 phr oils. These changes are well correlated to the change in crosslink density and also related to the less oil amount.

144

Figure 7.4. Storage modulus as a function of strain amplitude of various CB-filled IIR compounds at 90°C.

145

Figure 7.5. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various CB-filled IIR compounds and vulcanizates as a function of frequency at 90 °C.

146

7.2.3 Thermal properties

Figure 7.6 shows the TGA curves of CB-filled IIR vulcanizates with 15 phr of oil (a) and 10 phr of oil (b). In general, IIR/SO and IIR/NSO vulcanizates exhibit similar thermal stability and better than those of IIR/NO vulcanizates. This is due to the better thermal stability of the SO and NSO than that of the NO.126 Similar behavior was also seen in many previous studies.5, 130, 141 The better thermal stability of CB-filled IIR/SO and IIR/NSO vulcanizates is beneficial for their applications in high temperature environment, such as curing bladders.

Table 7.1 shows the Tg of various CB-filled IIR vulcanizates. CB-filled IIR/SO vulcanizates exhibit the lowest Tg. With the increase of the modification level, the Tg of the

CB-filled IIR/NSO vulcanizates is slightly increased, but still lower than that of the

CB-filled IIR/NO vulcanizates. This effect is due to the introduction of norbornyl groups to the SO. Similar effect was also observed in previous studies on the sulfur-cured

EPDM5 and SBR15 vulcanizates. After decreasing the oil level to 10 phr, various IIR vulcanizates exhibit a slightly increased Tg, but still in the same order with respect of various oils. In general, the addition of SO and NSO to the CB-filled IIR vulcanizates can achieve a slightly lower Tg than NO.

7.2.4 Curing behaviors

Figure 7.7 shows the curing curves of various CB-filled IIR compounds. Table 7.2 shows the scorch time Ts1, curing time T95 and CRI. It can be observed that the CB-filled

IIR/NO compounds exhibit higher maximum torques (MH) than those of the CB-filled

IIR/SO and IIR/NSO compounds. With the increase of modification level, the CB-filled 147

Figure 7.6. TGA curves of CB-filled IIR vulcanizates with 15 phr of oil (a) and 10 phr of oil (b).

148

Table 7.1. Glass transition temperatures of various CB-filled IIR vulcanizates.

Material Tg (°C) IIR/NO-1 -62.4 IIR/SO-1 -64.9 IIR/5NSO-1 -64.2 IIR/11NSO-1 -63.6 IIR/NO-2 -61.6 IIR/SO-2 -63.1 IIR/5NSO-2 -63.2 IIR/11NSO-2 -62.8

Table 7.2. Ts1, T95 and CRI of various CB-filled IIR compounds.

Ts1 (min) T95 (min) CRI IIR/NO-1 2.1 22.9 4.80 IIR/SO-1 2.0 16.9 6.71 IIR/5NSO-1 2.2 14.2 8.33 IIR/11NSO-1 2.2 13.9 8.55 IIR/NO-2 2.0 23.3 4.69 IIR/SO-2 1.8 19.6 5.62 IIR/5NSO-2 1.9 17.6 6.37 IIR/11NSO-2 2.0 17.2 6.58

149

Figure 7.7. Curing curves of various CB-filled CR compounds.

150

IIR/NSO compounds exhibit decreasing state of cure, which corresponds to different crosslink densities, as shown in Figure 7.1 (b). By decreasing the oil level from 15 phr to 10 phr, the MH of various IIR-2 compounds exhibit an obvious increase. Also, the minimum torque (ML) of various CB-filled IIR-2 compounds with 10 phr of oil is higher than those with 15 phr of oil. These effects are due to the increase in crosslink densities and also less plasticization effect and correlated with the G’ data as shown in Figure 7.4 (a). From Table

7.2, it can be seen that different CB-filled IIR compounds show similar scorch time. The use of SO and NSO decreases the T95 and increases the CRI. The effect of different amounts of NO on curing time and CRI is less obvious than that of the SO and NSO.

CB-filled IIR-2 compounds with 10 phr of SO or NSO exhibit longer T95 and lower CRI than those with 15 phr of SO or NSO. Previous studies also showed that the use of bio-based oils could increase the cure rate and shorten the curing time due to the extra fatty acids and covulcanization with SBR/NR with rubber seed oil,131 and CR with SO141. Thus the various CB-filled IIR-2 compounds with 10 phr of SO or NSO exhibit a slower cure than those with 15 phr of SO or NSO, respectively.

7.2.5 Mechanical properties and aging resistance

Figure 7.8 shows the stress-strain curves of various CB-filled IIR vulcanizates containing 15 and 10 phr of different oils before (a) and after (b) hot air aging. Table 7.3 shows detailed data of M100, M300, elongation at break, tensile strength and hardness before and after the hot air aging. From Figure 7.8 (a) and Table 7.3, it can be observed that the CB-filled IIR/SO vulcanizates exhibit a higher elongation at break, but lower tensile strength and modulus than those of the CB-filled IIR/NO vulcanizate. With the increase of

151

Figure 7.8. Stress-strain curves of various CB-filled IIR vulcanizates before aging (a) and after aging (b).

152

Table 7.3. Mechanical properties before and after aging of various CB-filled IIR vulcanizates.

Original Properties M100 M300 E @ B Tensile Strength Hardness (MPa) (MPa) (%) (MPa) (Shore A) IIR/NO -1 1.77±0.05 6.97±0.24 602.6±1.7 15.2±0.1 60 IIR/SO-1 1.17±0.01 4.04±0.10 734.4±13.5 13.6±0.3 56 IIR/5NSO-1 1.10±0.05 3.50±0.26 759.7±6.0 13.2±0.2 57 IIR/11NSO-1 1.05±0.02 3.39±0.10 756.3±13.3 13.1±0.4 57 IIR/NO-2 1.86±0.06 7.92±0.08 594.3±8.3 16.4±0.4 66 IIR/SO-2 1.47±0.04 5.58±0.24 667.1±18.9 14.6±0.7 63 IIR/5NSO-2 1.42±0.05 5.11±0.21 742.7±12.2 15.2±0.1 64 IIR/11NSO-2 1.35±0.02 4.47±0.15 763.7±19.8 15.1±0.5 64 Aged Properties (120 °C 100 h) M100 M300 E @ B Tensile Strength Hardness (MPa) (MPa) (%) (MPa) (Shore A) IIR/NO -1 2.83±0.04 10.54±0.23 450.7±18.6 15.1±0.3 70 IIR/SO-1 1.74±0.05 5.90±0.24 610.4±22.3 13.3±0.4 65 IIR/5NSO-1 1.62±0.03 5.14±0.21 698.3±13.5 13.4±0.1 66 IIR/11NSO-1 1.48±0.09 4.84±0.35 703.8±7.4 13.5±0.1 65 IIR/NO-2 3.06±0.10 11.71±0.54 423.9±14.1 15.7±0.3 74 IIR/SO-2 2.01±0.03 7.47±0.02 575.5±11.4 14.4±0.2 70 IIR/5NSO-2 1.87±0.04 6.82±0.21 640.4±11.5 14.8±0.4 70 IIR/11NSO-2 1.80±0.10 6.03±0.38 661.2±9.9 15.0±0.2 70 Percentage of Change M100 M300 E @ B Tensile Strength Hardness (%) (%) (%) (%) (%) IIR/NO-1 +59.9% +51.2% -25.2% -0.7% +16.7% IIR/SO-1 +48.7% +46.1% -16.9% -2.2% +16.1% IIR/5NSO-1 +47.2% +46.8% -8.1% +1.5% +15.7% IIR/11NSO-1 +41.0% +42.8% -6.9% +3.1% +14.3% IIR/NO-2 +64.5% +47.9% -28.7% -4.3% +12.1% IIR/SO-2 +36.7% +33.9% -13.7% -1.4% +11.1% IIR/5NSO-2 +31.7% +33.5% -13.8% -2.6% +9.4% IIR/11NSO-2 +33.3% +34.9% -13.4% -0.7% +9.4%

153

modification level, the tensile strength and modulus of IIR/NSO vulcanizates are further decreased. This is primarily due to the consumption of curative by the oils, leading to low crosslink density, as discussed above. After adjusting the oil level, various CB-filled IIR-2 vulcanizates with 10 phr of oils exhibit increased moduli and tensile strength. Similarly, as seen in Figure 7.4 and Figure 7.7, there is an increase in G’ and curing torque values. These are due to the less plasticization effect and also increase in crosslink density. It should be pointed out that the CB-filled IIR/NSO-2 vulcanizates with 10 phr of oils exhibit a similar tensile strength and much higher elongation at break compared with the CB-filled

IIR/NO-2 vulcanizate with 10 phr of oil. Table 7.3 also shows the hardness of various

CB-filled IIR vulcanizates. The use of SO and NSO tends to decrease the hardness of IIR vulcanizates in comparison with NO, which was also seen in previous studies on SBR and

CR rubbers.141,126,130 Decreasing the oil level to 10 phr increased the hardness of various

CB-filled IIR-2 vulcanizates and followed a trend similar to an addition of 15 phr oil.

Figure 7.8 (b) and Table 7.3 also depicts the change of properties after hot air aging.

CB-filled IIR/NO-1 and IIR/NO-2 vulcanizates exhibit larger change in modulus, elongation at break and hardness than other vulcanizates. The aging resistance of CB-filled

IIR/NSO-1 and IIR/NSO-2 vulcanizates is slightly better in terms of elongation at break and hardness than the CB-filled IIR/SO-1 and IIR/SO-2 vulcanizates. In general, the use of

NSO improves the overall aging resistance for CB-filled IIR vulcanizates. Similar effect was also seen in our previous study on CR and SBR.135,141 The big change in mechanical properties for CB-filled IIR/NO-1 and IIR/NO-2 vulcanizates is due to the significant oil loss during the hot air aging test. 5.0 % and 3.7 % total weight loss are observed for

CB-filled IIR/NO-1 and IIR/NO-2 vulcanizates, respectively. However, the IIR

154

vulcanizates containing SO and NSO only show less than 1% change in weight after the aging. This is due to the good thermal stability of SO and NSO, as shown in Figure 7.6.

7.2.6 Oil migration

It is observed that the CB-filled IIR vulcanizates containing 15 phr of SO and NSO exhibit oil migration at the surface after a storage time around 1 month, resulting a “sticky” and “shining” surface. It indicates limited solubility of SO or NSO in the CB-filled IIR cured by phenolic resin. Oil migration is not seen in CB-filled IIR/NO vulcanizates, indicating better compatibility between IIR and NO compared with SO.

Figure 7.9 shows the FT-IR spectrum for P-124 curing resin, NO, SO, 11NSO,

CB-filled IIR/NO, IIR/SO and IIR/11NSO vulcanizates containing 15 phr of various oils.

With no surprise, the surfaces of various CB-filled IIR vulcanizates exhibit absorbance peaks of various oils, indicating the presence of different oils. However, the migrated oils are extremely hard to separate from the surfaces, thus the conducted FT-IR spectrum can’t directly show the content of the migrated materials. By decreasing the oil level from 15 phr to 10 phr, the oil migration is significantly less pronounced.

Oil migration for SO and NSO has not been seen in other rubbers including SBR and

CR. The amount of SO and NSO used in CB-filled IIR compounds should be limited.

7.3 Conclusions

NSO is a bio-based, low cost and reactive rubber plasticizer. It can be used in

CB-filled IIR compounds cured by phenolic resins. The use of NSO in CB-filled IIR compounds improves the thermal stability, cure rate, elongation at break and hot air aging

155

Figure 7.9. FT-IR spectrum for phenolic resin, NO, SO, 11NSO, CB-filled IIR/NO, IIR/SO and IIR/11NSO vulcanizates.

156

resistance. CB-filled IIR vulcanizates containing SO and NSO show slightly lower Tg than those containing NO. SO and NSO are found to react with the phenolic resin during the vulcanization process. This causes a reduction in the crosslink density of the vulcanizates.

By decreasing the oil level, crosslink density of CB-filled IIR/NSO-1 vulcanizates is increased, while reducing amount of oil in IIR/NO-1 vulcanizates does not affect the crosslink density, providing CB-filled IIR/NSO-2 vulcanizates with comparable tensile strength as that of the CB-filled IIR/NO vulcanizate. It should be noticed that oil migration is observed for IIR containing 15 phr of SO and NSO.

157

CHAPTER VIII

COMPARISON OF CB-FILLED TIN-COUPLED AND UNCOUPLED LINEAR SBR WITH VARIOUS OILS

8.1 Introduction

SO has C=C bonds on its fatty acid chains. DCPD can be used to norbornylize the SO via a Diels-Alder reaction. Similarly, other dienes can also be used to modify the SO.

Isoprene is an accessible and abundant diene resource. About 800,000 tons of isoprene are produced each year, and about 95% are used to manufacture polyisoprene rubbers.6

Isoprene modified soybean oil (ISO) has a similar property to NSO, it could be another alternative rubber plasticizer. An advantage of isoprene to DCPD is its lower boiling point, thus it is easier to remove the residual unreacted isoprene. The comparative study of NSO and ISO can provide information on the effect of diene-modified SO in SBR compounds and vulcanizates.

Although tin-coupled SBR has several advantages than common uncoupled linear

SBR, the latter still takes a majority share of SBR market. The different molecular structure of the SBR can result in a different rheological behavior of the compounds, and the functionalization may create a difference in rubber-filler interaction at the presence of different oils. Therefore, it is of interest to investigate the behavior of NSO in functionalized and uncoupled SBR. The research will also make a comparison between the properties of CB-filled linear and tin-coupled SBR containing different oils. The gel

158

fraction, crosslink density, bound rubber fraction, rheological properties, thermal properties, curing behaviors, mechanical and dynamic properties of various CB-filled linear and tin-coupled SBR are compared. 15 phr of various oils are used in uncoupled

SBR compounds in order to match the Mooney viscosity to tin-coupled SBR with 30 phr of

NO. Data of various CB-filled tin-coupled SBR compounds and vulcanizates are from author’s previous study.126

8.2 Results and discussion

8.2.1 Gel fraction, crosslink density and bound rubber

Figure 8.1 shows the gel fraction (a), crosslink density (b) of various CB-filled linear and tin-coupled SBR vulcanizates, and bound rubber fraction (c) of various

CB-filled linear and tin-coupled SBR compounds. In Figure 8.1 (a), it can be seen that, in general, various CB-filled linear SBR vulcanizates exhibit higher gel fraction than those of the CB-filled tin-coupled SBR vulcanizates. This is primarily due to the different oil content in two recipes. The CB-filled linear SBR vulcanizates have 15 phr of different oils instead of 30 phr in various CB-filled tin-coupled SBR vulcanizates, thus the

CB-filled linear SBR vulcanizates exhibit higher gel fraction. For both linear and tin-coupled of SBR, different oils provide similar gel fraction values.

Figure 8.1 (b) shows that the CB-filled tin-coupled and linear SBR vulcanizates containing NO exhibit similar crosslink density, indicating that at the same level of curatives, the linear or star-shaped structure of SBR has little effect on the crosslink density. For both CB-filled tin-coupled and linear SBR vulcanizates, the use of SO

159

Figure 8.1. Gel fraction (a), crosslink density (b) of various CB-filled SBR vulcanizates, and bound rubber fraction (c) of various CB-filled SBR compounds.

160

decreases the crosslink density. With the increase of modification level, the crosslink density is further decreased. This is due to the consumption of sulfur by reactive SO and modified SO. Similar behavior was also seen in Chapter IV and Chapter V. Two CB-filled linear SBR/ISO vulcanizates show similar values of crosslink density as those with NSO, indicating a similar reactivity between NSO and ISO.

Figure 8.1 (c) shows the bound rubber fraction of various CB-filled SBR compounds.

Due to the less oil level in linear SBR compounds, the bound rubber fraction is higher than those of tin-coupled SBR compounds. But they have a similar trend in terms of different oils. When using SO in CB-filled linear and tin-coupled SBR compounds, the bound rubber fraction is slightly increased compared with those with NO. With the increase of the norbornylization level, the bound rubber fraction of CB-filled linear and tin-coupled SBR is slightly decreased. This behavior is also seen in Chapter IV and

Chapter V on tin-coupled SBR compounds with different fillers and also discussed in our previous study.126 The use of ISO in CB-filled linear SBR shows similar effect on bound rubber fraction as those with NSO. CB-filled linear SBR/ISO compounds exhibit similar levels of bound rubber fraction as those of CB-filled SBR/NSO compounds.

8.2.2 Rheological properties

Figure 8.2 shows the G’ as a function of strain amplitude of various CB-filled linear and tin-coupled SBR compounds with different oils plotted in linear (a) and logarithm (b) scale. It is clearly observed that all SBR compounds show a decrease in G’ with the increase of strain amplitude, which is known as Payne effect.133 Generally, CB-filled linear

SBR compounds exhibit higher G’ values at low strain amplitude than those of the

161

Figure 8.2. Storage modulus as a function of strain amplitude of various CB-filled linear and tin-coupled SBR containing different oils in linear (a) and logarithm (b) scale.

162

CB-filled tin-coupled SBR compounds. This effect is firstly due to the lower level of plasticizer in CB-filled linear SBR compounds, 15 phr in linear SBR compounds instead of 30 phr in the tin-coupled SBR compounds. Secondly, it is due to the unique property of tin-coupled SBR. The carbon-tin bond in tin-coupled SBR can break and attach to the surface of CB,44 which can benefit the dispersion of CB. Thus decrease the filler-filler interaction. At higher strain amplitude region, CB-filled linear SBR compounds exhibit a slightly higher G’ value than those of CB-filled tin-coupled SBR compounds, indicating a higher rubber-filler interaction. The results are in agreement with the bound rubber test shown in Figure 8.1 (c). Theoretically, tin-coupled SBR should have a better rubber-CB interaction than the linear SBR,44-47 but in this case, the effect of tin-coupling is diminished by the high level of plasticizer in CB-filled tin-coupled SBR compounds.

For both CB-filled linear and tin-coupled SBR compounds, SBR/NO compounds show higher filler-filler interaction than those with SO and modified SO. This behavior is due to the microplasticization effect of SO on CB12 and is also seen in Chapter IV and

Chapter V. ISO and NSO exhibit similar effect in CB-filled linear SBR compounds, indicating limited influence on the dispersion of fillers.

Figure 8.3 shows the G’ (a), G” (b), tan δ (c) and complex viscosity (d) of various unfilled linear and tin-coupled SBR gums containing different oils. From Figure 8.3 (a), it is seen that the tin-coupled SBR gum without any oil exhibits the highest G’, higher than that of the linear SBR gum without any oil. This is primarily due to the higher molecular weight of tin-coupled SBR gum. The Mooney viscosity of tin-coupled SBR gum is 92, much higher than the linear SBR gum, which has a Mooney viscosity of 55. In consequence, various tin-coupled SBR gums with different oils exhibit higher value of G’

163

Figure 8.3. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various unfilled linear and tin-coupled SBR gums and with different oils.

164

than those of linear SBR gums with oils. Figure 8.2 (b) shows that the tin-coupled SBR gum without oil exhibit a peak in G”, indicating transition to the rubbery plateau. However, the linear SBR gum without oil exhibits an increasing G”. This is due to the Tg difference between the tin-coupled SBR (-32 °C) and linear SBR (-62 °C) gums. Consequently, unfilled SBR gums containing different oils show similar behavior as their gums without oils. From Figure 8.3 (c), it is observed that all linear SBR gums exhibit overall higher tan

δ values than the tin-coupled SBR gum. The tin-coupled SBR gum without any oil exhibits the lowest tan δ. It is also noticed that various tin-coupled SBR gums show a more obvious drop in tan δ with the increase of the frequency. This is due to the star-shaped highly branched structure of tin-coupled SBR. The relaxation of its rubber chains is more restricted than that of the linear SBR. Figure 8.3 (d) shows the complex viscosity of various unfilled SBR gums. Similar to the G’, tin-coupled SBR gum without any oil exhibits the highest value, followed by linear SBR gum without any oil. After adding oils, the trend is still the same. At high frequency region, the complex viscosity of various oil-extended

SBR gums starts to converge to a similar value. Overall, the molecular structure has obvious influence on the dynamic behavior of SBR compounds, and different oils show limited effect on the dynamic properties of both linear and tin-coupled SBR gums. The different dynamic behavior between linear and star-shaped polymer is also seen in previous studies on polybutadiene153 and polystyrene154.

Figure 8.4 shows the G’ (a), G” (b), tan δ (c) and complex viscosity (d) of various

CB-filled linear and tin-coupled SBR compounds containing different oils. From Figure

8.4 (a), it is seen that the CB-filled linear SBR/NO compound exhibits the highest G’, followed by various CB-filled SBR compounds with SO and modified SO. The CB-filled

165

Figure 8.4. Storage (a), loss (b) moduli, tan δ (c) and complex viscosity (d) of various CB-filled linear and tin-coupled SBR compounds with different oils.

166

tin-coupled SBR compounds also exhibit significantly lower G’ values at low frequency region than those of the CB-filled linear SBR compounds. These effects are due to the lower plasticizer content of CB-filled linear SBR compounds, breaking down of tin-coupled SBR chain during the compounding, and also a more developed filler network in the CB-filled linear SBR compounds. A stronger filler network will increase the G’ values at low frequency region.140 Since the frequency sweep is performed at a strain amplitude of 4.2%, such a tendency can also be seen in the strain sweep test shown in

Figure 8.2. The G” values shown in Figure 8.4 (b) follow the same behavior as G’. As a result of stronger filler network in the CB-filled linear SBR, and also less plasticizer content, the CB-filled linear SBR compounds exhibit a lower tan δ value than those of the

CB-filled tin-coupled SBR compounds. However, a crossover in tan δ value is observed.

For complex viscosity in Figure 8.4 (d), the relationship between complex viscosity values of various SBR compounds is similar to the G’ and G”. A less filler-filler interaction gives

CB-filled tin-coupled SBR compounds a lower complex viscosity than the CB-filled linear

SBR compounds at low frequency region. Similar behavior was also seen in Chapter IV and Chapter V.

8.2.3 Thermal properties

Table 8.1 shows the Tg of various CB-filled SBR compounds and vulcanizates. The

Tg of pure linear SBR gum is -62 °C, much lower than that of the tin-coupled SBR gum with Tg being -32 °C. Therefore, the CB-filled linear SBR compounds and vulcanizates exhibit a lower Tg than those of the CB-filled tin-coupled SBR compounds and vulcanizates. Compared with NO, SO provides the lowest Tg for both CB-filled linear

167

and tin-coupled SBR compounds and vulcanizates. With the increase of the modification level, the CB-filled linear and tin-coupled SBR/NSO, and CB-filled linear SBR/ISO compounds and vulcanizates all exhibit an increase in Tg. CB-filled linear SBR compounds and vulcanizates with ISO and NSO exhibit similar values of Tg. It should be noticed that after vulcanization, the Tg of various SBR vulcanizates are higher than the

SBR compounds due to the introduction of crosslinks. In general, the lower Tg brought by the SO and modified SO can provide better low temperature flexibility for the SBR.

Figure 8.5 shows the TGA curves of various CB-filled linear SBR vulcanizates (a) and tin-coupled SBR vulcanizates (b). In terms of different oils, the NO provides the worst thermal stability among all oils. SBR vulcanizates containing NO show weight loss starting from around 150 °C, yet SBR vulcanizates containing SO and modified SO do not show significant weight loss till temperature higher than 300 °C. This is due to the lower flash point of NO (171 °C) than SO (>270 °C), which makes NO easier to become volatile. Thus SO and various modified SO provide similar and better thermal stability than NO for SBR vulcanizates. Due to the less oil content in CB-filled linear SBR vulcanizates, the oil loss in Figure 8.5 (a) is less pronounced than the CB-filled tin-coupled SBR vulcanizates, shown in Figure 8.5 (b). Similar behavior was seen in various rubber systems in Chapter IV, V, VI and VII. A better thermal stability brought by

SO and modified SO can provide better safety for processing and applications.

168

Table 8.1. Glass transition temperature of CB-filled linear and tin-coupled SBR compounds and vulcanizates.

Linear SBR Tin-Coupled SBR

Compound Tg Vulcanizate Tg Compound Tg Vulcanizate Tg (°C) (°C) (°C) (°C) SBR/NO -67.8 -61.4 -45.8 -38.8 SBR/SO -69.9 -64.5 -52.2 -42.3 SBR/5NSO -69.6 -64.4 -48.8 -41.2 SBR/11NSO -69.5 -63.1 -48.1 -40.5 SBR/5ISO -69.8 -64.4 - - SBR/11ISO -69.3 -64.3 - -

169

8.2.4 Curing behaviors

Figure 8.6 shows the curing curves of various CB-filled linear SBR compounds (a) and tin-coupled SBR compounds (b). The scorch time Ts1, curing time T95 and CRI are shown in Table 8.2. Compared with CB-filled tin-coupled SBR compounds, the CB-filled linear SBR compounds exhibit higher state of cure, shorter Ts1 and T95, and also a higher

CRI. It is possibly due to the less plasticization by the oils and also the different molecular structure between these linear and tin-coupled SBR. Similar behavior was also seen in previous study on linear and tin-coupled SBR with petroleum based plasticizer.155 The molecular structure of SBR can have influence on the curing speed.

In CB-filled tin-coupled SBR compounds, the use of SO can decrease the MH and Ts1 compared with NO. With the increase of norbornylization level, the CB-filled tin-coupled

SBR/NSO compounds show further decreased MH, shortened T95 and higher CRI. Similar behavior was also seen in CB-filled CR and IIR compounds shown in Chapter VI and VII.

However, in CB-filled linear SBR compounds, the use of SO, NSO and ISO all show similar behavior. CB-filled linear SBR compounds containing SO, NSO and ISO exhibit lower MH, shorter T95 and higher CRI compared with CB-filled linear SBR/NO compound.

In general, compared with NO, the use of SO and modified SO increase the cure rate and reduce curing time for both CB-filled linear and tin-coupled SBR compounds. The addition of ISO to SBR shows similar effect on curing behaviors as NSO.

170

Figure 8.5. TGA curves of various CB-filled linear SBR vulcanizates (a) and tin-coupled SBR vulcanizates (b).

171

Figure 8.6. Curing curves of various CB-filled linear SBR compounds (a) and tin-coupled SBR compounds (b) with different oils.

172

Table 8.2. Ts1, T95 and CRI of various CB-filled linear and tin-coupled SBR compounds with different oils.

Linear SBR Tin-Coupled SBR

Ts1 (min) T95 (min) CRI Ts1 (min) T95 (min) CRI SBR/NO 3.8 8.8 20.0 5.2 18.2 7.7 SBR/SO 3.4 6.5 32.3 4.6 17.8 7.6 SBR/5NSO 3.6 6.5 34.5 4.7 14.4 10.3 SBR/10NSO 3.6 6.5 34.5 4.7 11.5 14.7 SBR/5ISO 3.4 6.4 33.3 - - - SBR/10ISO 3.3 6.2 34.5 - - -

173

8.2.5 Mechanical properties

Figure 8.7 compares the stress-strain curves of various CB-filled linear and tin-coupled SBR vulcanizates containing different oils. Detailed data on M100, M300, elongation at break and tensile strength are shown in Table 8.3. In general, the CB-filled linear SBR vulcanizates exhibit higher modulus and tensile strength compared to those of the CB-filled tin-coupled SBR. This is due to the less plasticizer content in the CB-filled linear SBR vulcanizates. CB-filled linear SBR/NO and SBR/SO vulcanizates exhibit higher elongation at break, while the CB-filled linear SBR/NSO vulcanizates exhibit lower elongation at break than the CB-filled tin-coupled SBR/NSO vulcanizates.

For CB-filled tin-coupled SBR vulcanizates, the use of SO provides increased elongation at break and tensile strength compared with NO. With the increase of norbornylization level, the elongation at break and tensile strength of CB-filled tin-coupled SBR/NSO vulcanizates are further increased. However, the moduli are decreased. This is due to the decrease in crosslink density caused by the consumption of sulfur by oils. The modulus of vulcanizate correlates well to the crosslink density data shown in Figure 8.1 (b). Similar effect was also seen in our previous study126 and Chapter

IV to VII.

CB-filled linear SBR vulcanizates containing SO, NSO and ISO show lower modulus, increased elongation at break and tensile strength than CB-filled linear SBR containing NO. Table 8.3 shows that the CB-filled linear SBR vulcanizates with NSO and

ISO exhibit around 45% increase in elongation at break, and 9% in tensile strength compared with CB-filled linear SBR/NO vulcanizate. However, the addition of SO, NSO and ISO to CB-filled linear SBR does not show big difference in tensile properties as

174

Figure 8.7. Stress-strain curves of various CB-filled linear SBR (a) and tin-coupled SBR (b) vulcanizates.

175

Table 8.3. M100, M300, elongation at break, tensile strength and hardness of various CB-filled uncoupled and tin-coupled SBR vulcanizates.

M100 M300 E@B Tensile Strength Hardness

(MPa) (MPa) (%) (MPa) (Shore A)

Linear SBR

SBR/NO 2.88 ± 0.08 13.10 ± 0.18 397.4 ± 10.5 18.7 ± 0.2 65 SBR/SO 2.37 ± 0.05 9.67 ± 0.14 543.7 ± 11.9 19.5 ± 0.8 63 SBR/5NSO 2.36 ± 0.04 9.57 ± 0.09 559.1 ± 10.8 20.0 ± 0.6 62 SBR/11NSO 2.31 ± 0.09 9.40 ± 0.17 583.9 ± 12.2 20.6 ± 0.2 63 SBR/5ISO 2.11 ± 0.04 9.05 ± 0.12 562.1 ± 3.3 20.0 ± 0.5 62 SBR/11ISO 2.05 ± 0.04 8.52 ± 0.15 587.2 ± 8.5 20.4 ± 0.4 63 Tin-Coupled SBR

176

in the CB-filled tin-coupled SBR vulcanizates. The use of NSO and ISO in linear SBR only show slight increase in elongation at break and tensile strength, with slight decrease in modulus compared with SO. This is due to the low amount of plasticizer in CB-filled linear SBR vulcanizates. The consumption of sulfur is limited thus the changes in crosslink density and mechanical properties are less pronounced. It also can be seen that the effect of ISO is similar to that of the NSO on tensile properties.

Table 8.3 also shows the hardness of various CB-filled SBR vulcanizates. The

CB-filled linear and tin-coupled SBR/NO vulcanizates all have a hardness of 65. Due to the decreased crosslink density, CB-filled linear and tin-coupled SBR vulcanizates containing SO and modified SO all exhibit a decrease in hardness. The CB-filled linear

SBR/SO, SBR/NSO and SBR/ISO have a hardness around 62-63. The CB-filled tin-coupled SBR/SO and SBR/NSO vulcanizates exhibit hardness of 61 and 59, respectively, due to the further decreased crosslink density. Data can be correlated to the crosslink density shown in Figure 8.1 (b) and also modulus shown in Table 8.3.

Figure 8.8 shows the abrasion loss of various CB-filled linear and tin-coupled SBR vulcanizates containing different oils. In general, the CB-filled linear SBR vulcanizates exhibit lower abrasion loss due to the less plasticizer compared with those of the

CB-filled tin-coupled SBR vulcanizates. For CB-filled tin-coupled SBR vulcanizates, the

SBR/SO vulcanizate exhibits the lowest abrasion loss due to its lowest Tg. With the increase of the norbornylization level, the CB-filled tin-coupled SBR/NSO vulcanizates exhibit an increasing abrasion loss. For various CB-filled linear SBR vulcanizates, the differences between SBR containing different oils are less pronounced. CB-filled linear

SBR/SO vulcanizate has a lower abrasion loss than the CB-filled linear SBR/NO

177

Figure 8.8. Abrasion loss of various CB-filled linear and tin-coupled SBR vulcanizates.

178

vulcanizate. The CB-filled linear SBR/5NSO and SBR/11NSO vulcanizates exhibit a similar abrasion loss and slightly higher than the CB-filled linear SBR/SO vulcanizate.

The CB-filled linear SBR/5ISO and SBR/11ISO exhibit similar and lower abrasion loss compared with that of the CB-filled linear SBR/SO vulcanizate, respectively. The addition of SO, NO and ISO can improve the abrasion resistance for CB-filled linear

SBR vulcanizates.

8.2.6 DMA and performance predictors

Figure 8.9 shows the DMA curves including E’ (a), E” (b) and tan δ (c) as a function of temperature of various CB-filled linear SBR vulcanizates. Compared with DMA curves of

CB-filled tin-coupled SBR vulcanizates shown in Figure 4.11, it can be seen that the

CB-filled linear SBR vulcanizates show lower Tg, which correlates well with the DSC data shown in Table 8.1. It can be observed from Figure 8.9 (a) that the CB-filled linear

SBR/NO vulcanizate exhibits a slightly higher E’ value than other vulcanizates in the rubbery region. This is due to the higher crosslink density, as shown in Figure 8.1 (b) and also in agreement with data shown in Table 8.3. Various CB-filled linear SBR vulcanizates containing different oils show very similar behavior on E” and tan δ vs. temperature.

Figure 8.10 shows the tan δ values of various CB-filled linear and tin-coupled SBR vulcanizates at 10 °C (a) and 60 °C (b) from DMA tests. The tan δ values at 10 °C and

60 °C are used to predict wet traction and rolling resistance of tire made of SBR vulcanizates, respectively.128 In general, the CB-filled tin-coupled SBR exhibit higher tan δ at 10 °C than the various CB-filled linear SBR vulcanizates, indicating a better wet traction. For CB-filled tin-coupled SBR vulcanizates, the SBR/NO vulcanizate exhibits

179

Figure 8.9. Storage modulus (a), loss modulus and tan δ of various CB-filled linear SBR vulcanizates as a function of temperature.

180

Figure 8.10. Tan δ values at 10 °C (a) and 60 °C (b) of various CB-filled linear and tin-coupled SBR vulcanizates from DMA test.

181

the highest tan δ at 10 °C, while the SBR/SO exhibits the lowest. With the increase of the norbornylization level, the tan δ value at 10 °C is increased but still lower than that of the

SBR/NO vulcanizate. For various CB-filled linear SBR vulcanizates, SBR/NO vulcanizate exhibit the lowest tan δ value at 10 °C, while SBR/SO and SBR/NSO vulcanizate exhibit increasing tan δ value at 10 °C, indicating an improvement of wet traction. The CB-filled linear SBR/ISO vulcanizates exhibit very similar tan δ values at

10 °C as CB-filled linear SBR/NSO vulcanizates. It shows that ISO behaves similarly compared with NSO in CB-filled linear SBR vulcanizates.

The CB-filled tin-coupled SBR/NO vulcanizate exhibit a slightly lower tan δ at 60 °C than the CB-filled linear SBR/NO vulcanizate, indicating a slightly improved rolling resistance. The use of SO and NSO can increase the tan δ value at 60 °C for both CB-filled linear and tin-coupled SBR vulcanizates. Both CB-filled linear and tin-coupled SBR/SO vulcanizates exhibit similar tan δ value at 60 °C, but with the increase of the norbornylization level, the CB-filled tin-coupled SBR/NSO vulcanizates show more increase than those of the CB-filled linear SBR/NSO vulcanizates. This is due to the further decreased crosslink density of CB-filled tin-coupled SBR/NSO vulcanizates, as shown in

Figure 8.1 (b). The use of ISO in CB-filled linear SBR vulcanizates show similar behavior as NSO since the tan δ values at 60 °C are very close to each other. In general, CB-filled tin-coupled SBR vulcanizates show better wet traction and similar rolling resistance than

CB-filled linear SBR vulcanizates. Similar effect was also observed by previous studies on

CB- tin-coupled SBR without plasticizers44 and with petroleum based plasticizer45. The use of modified SO can improve the wet traction but also increase rolling resistance compared with SO. NSO and ISO behave similarly in CB-filled linear SBR vulcanizates.

182

8.3 Conclusion

The study compared the CB-filled linear and tin-coupled SBR compounds and vulcanizates containing different oils. The molecular structure has obvious influence on the rheological and dynamic properties of SBR compounds and vulcanizates. Tin-coupled

SBR has higher Tg, molecular weight and wet traction than linear SBR. The effect of SO and modified SO is similar in both CB-filled linear and tin-coupled SBR compounds and vulcanizates. SO modified through different methods, including NSO and ISO, exhibits very similar effect on rubber properties. In general, the use of SO and modified SO will improve the thermal stability and lower the Tg of SBR compounds and vulcanizates compared with NO. SO and modified SO also decrease the crosslink density of SBR vulcanizates. Mechanical test results show that SO and NSO provide increased elongation at break and tensile strength than NO. The use of SO, NSO and ISO exhibits very similar effect on mechanical properties in CB-filled linear SBR vulcanizate, yet in CB-filled tin-coupled SBR vulcanizates, difference in mechanical properties between NO, SO and

NSO is more pronounced. With the increase of modification level, SBR with NSO and

ISO exhibit increased wet traction and also rolling resistance.

In conclusion, SO or NSO has similar influence on properties in CB-filled linear and tin-coupled SBR compounds and vulcanizates. ISO exhibits similar influence of properties as NSO. The amount of oil has important influences on rubber properties. SO and modified SO can be used in both CB-filled linear and tin-coupled SBR compounds and vulcanizates to replace conventional petroleum-based plasticizer, and gain improvements in several properties.

183

CHAPTER IX

SUMMARY

Modified SO, such as NSO and ISO, is a low cost, environmental friendly and reactive rubber plasticizer. It is derived from renewable SO resource and has shown different properties from conventional petroleum-based rubber plasticizers. Modified SO can be used in various rubber compounds to replace petroleum-based rubber plasticizers such as NO, including different rubbers such as linear and tin-coupled SBR, CR and IIR, different filler systems including CB, silica and CB/silica hybrid fillers, and also different curing systems including sulfur, metal oxide and phenolic resin systems. Rubbers containing modified SO show improvement in several aspects.

SO and modified SO have higher flash point and lower Tg than NO. It is shown to provide better thermal stability and lower Tg for various SBR, CR and IIR compounds and vulcanizates than conventional NO. Better thermal stability brings better safety to the processing and also application of rubbers. A lower Tg may benefit low temperature flexibility of rubbers and is particularly important for certain applications. The use of SO and modified SO is seen to decrease the filler-filler interactions in CB-filled rubber compounds, and can benefit the dispersion of fillers in rubbers.

Compared with NO and SO, modified SO can provide better scorch safety, shorter curing time, higher cure rate and lower state of cure in different curing systems. The unique structure of bio-based oils provides extra fatty acids during the curing influencing

184

the curing process. The most important feature of SO and modified SO is their reactivity.

The reactive C=C double bonds react with different curing agents such as sulfur, sulfur-containing coupling agent and phenolic resin during the curing process. The reaction consumes certain amount of curatives and silane coupling agent to affect the rubber-filler interaction, gel fraction, crosslink density and also crosslink structure. With the increase of modification level, the C=C double bonds in the modified SO are more reactive. Therefore, more reaction takes place tremendously influencing the properties of rubbers. Different modifications to the SO, which are NSO and ISO, show very similar reactivity and overall effect on rubbers.

Since SO and modified SO consume curatives and decrease the crosslink density, the change in crosslink density vs. modification level of SO is particularly important.

Figure 9.1 shows a summary of relative crosslink density vs. modification levels of SO of various rubber vulcanizates in this study. The relative crosslink density values of various rubber vulcanizates are normalized to the rubber vulcanizates containing SO with the same rubber type, filler type and curative level. It can be seen that the type of filler and amount of sulfur have little effect on the change of the relative crosslink density of the tin-coupled

SBR vulcanizates. CB-filled linear SBR vulcanizates with 15 phr of oils show different behavior due to the lower oil level. The CB-filled CR vulcanizates exhibit the least decrease in crosslink density, the oils react with sulfur which is copolymerized with CR monomer. Different reaction mechanisms result in different behavior of change in crosslink density. The IIR vulcanizate containing 5NSO exhibit similar change as the tin-coupled SBR with same modification level of oil, and show less change in the 11NSO.

185

Figure 9.1. Relative crosslink density vs. modification level of various rubber vulcanizates.

186

By decreasing the oil content to 10 phr in the CB-filled IIR vulcanizates, less decrease in relative crosslink density is observed compared with 15 phr of oils. In general, the use of modified SO will decrease the crosslink density of various rubber vulcanizates, proper adjustment of curing recipe should be made to increase crosslink density and achieve required rubber properties.

In most rubber systems, with proper adjustment to the recipe, SO and modified SO are shown to provide higher elongation at break, tensile strength, abrasion resistance, and tear strength compared with NO. For CB-filled tin-coupled SBR system, an increase in elongation at break and tensile strength can be up to 208% and 42%, respectively.

Abrasion resistance and tear strength improvement is seen in both SBR and CR systems.

Both SO and modified SO improve aging resistance for SBR, CR and IIR vulcanizates than NO. The reactive oils suppress the generation of free radicals and thus improve the aging resistance. Also, SO and modified SO exhibit much better thermal stability and show less oil loss during the aging of rubbers compared with NO, providing better volume and mechanical property consistency.

In rubber compounds containing modified SO, a proper adjustment to the curing recipe can improve the wet traction, lower the rolling resistance and increase abrasion resistance simultaneously. This phenomenon is extremely beneficial for tire performance.

In conclusion, modified SO can be used in various rubber products to replace conventional petroleum-based rubber plasticizers. Meanwhile, it can improve several important properties of the rubber products. These green, low cost and reactive plasticizers demonstrate a great potential for application in various rubber products.

187

REFERENCES

1. Ahlbom, J.; Duus, U. Nya hjulspår—en produktstudie av gummidäck (New Wheel Tracks—a product study of rubber tyres). ; KEMI National Chemicals Inspectorate: June, 1994; p 78. 2. Weisz, P. B., Basic choices and constraints on long‐term energy supplies. Phys. Today 2004, 57, 47. 3. Dasgupta, S.; Agrawal, S. L.; Bandyopadhyay, S.; Chakraborty, S.; Mukhopadhyay, R.; Malkani, R. K.; Ameta, S. C., Characterization of eco-friendly processing aids for rubber compound. Polym. Test. 2007, 26 (4), 489-500. 4. Flanigan, C.; Beyer, L.; Klekamp, D.; Rohweder, D., Using bio-based plasticizers, alternative rubber. Rubber & Plastic News Feb 11, 2013. 5. Wang, Z.; Han, Y.; Zhang, X.; Huang, Z.; Zhang, L., Plasticization effect of transgenic soybean oil. I. On ethylene propylene diene monomer (EPDM), as substitute for paraffin oil. J. Appl. Polym. Sci. 2013, 130 (6), 4457-4463. 6. Greve, H. H., Rubber, 2. Natural. In Ullmann's Encyclopedia of Industrial Chemistry, John Wiley and Sons, Inc.: Hoboken, 2000. 7. Zong, Z.; Soucek, M. D.; Liu, Y.; Hu, J., Cationic photopolymerization of epoxynorbornane linseed oils: The effect of diluents. J. Polym. Sci., Part A: Polym. Chem. 2003, 41 (21), 3440-3456. 8. Delpech, M. C.; Mello, I. L.; Delgado, F. C. S.; Sousa, J. M., Evaluation of thermal and mechanical properties of rubber compositions based on SBR extended with safe oils. J. Appl. Polym. Sci. 2012, 125 (5), 4074-4081. 9. Keshun, L., Soybeans: chemistry, technology, and utilization. Chapman & Hall: London, 1997. 10. USDA World Agricultural Supply and Demand Estimates, ; Jan. 12th, 2017; p 3. 11. Soybean Oil Futures End of Day Settlement Price. http://www.indexmundi.com/commodities/?commodity=soybean-oil (accessed Feb 5th 2017). 12. Kundu, P., Improvement of filler–rubber interaction by the coupling action of vegetable oil in carbon black reinforced rubber. J. Appl. Polym. Sci. 2000, 75 (6), 735-739. 13. Kundu, P.; Kukreja, T., Surface modification of carbon black by vegetable oil—its effect on the rheometric, hardness, abrasion, rebound resilience, tensile, tear, and adhesion properties. J. Appl. Polym. Sci. 2002, 84 (2), 256-260. 14. Kukreja, T.; Chauhan, R.; Choe, S.; Kundu, P., Effect of the doses and nature of vegetable oil on carbon black/rubber interactions: Studies on castor oil and other vegetable oils. J. Appl. Polym. Sci. 2003, 87 (10), 1574-1578. 15. Petrović, Z. S.; Ionescu, M.; Milić, J.; Halladay, J. R., Soybean oil plasticizers as replacement of petroleum oil in rubber. Rubber Chem. Technol. 2013, 86 (2), 233-249. 16. Recker, C.; Reese, T.; Schoning, J. Rubber mixture comprising environmentally friendly softeners. US 8,236,875 B2 2012. 17. Thompson, W. M.; Saintigny, X. Tire tread for high performance tires. US 2013/0096248 A1, 2013. 18. Thomas W Wilson, I. Rubber compositions with non-petroleum oils. US 7,211,611 B2, 2007.

188

19. James Robert Herberger SR.; Bina Patel Botts; Anthony James Houser J R. Tire component, and tire with such component, of rubber composition which contains combination of soybean oil and starch/plasticizer composite. US 2005/0145312 A1, 2005. 20. Emken, E., Nutrition and biochemistry of trans and positional fatty acid isomers in hydrogenated oils. Annu. Rev. Nutr. 1984, 4 (1), 339-376. 21. Fankhauser-Noti, A.; Fiselier, K.; Biedermann, S.; Biedermann, M.; Grob, K.; Armellini, F.; Rieger, K.; Skjevrak, I., Epoxidized Soy Bean Oil (ESBO) migrating from the gaskets of lids into food packed in glass jars. Eur. Food Res. Technol. 2005, 221 (3-4), 416-422. 22. Teng, G.; Soucek, M., Epoxidized soybean oil-based ceramer coatings. J. Am. Oil Chem. Soc. 2000, 77 (4), 381-387. 23. Balat, M.; Balat, H., A critical review of bio-diesel as a vehicular fuel. Energy Convers. Manage. 2008, 49 (10), 2727-2741. 24. Zlatanić, A.; Lava, C.; Zhang, W.; Petrović, Z. S., Effect of structure on properties of polyols and polyurethanes based on different vegetable oils. J. Polym. Sci., Part B: Polym. Phys. 2004, 42 (5), 809-819. 25. Wang, Z.; Zhang, X.; Wang, R.; Kang, H.; Qiao, B.; Ma, J.; Zhang, L.; Wang, H., Synthesis and characterization of novel soybean-oil-based elastomers with favorable processability and tunable properties. Macromolecules 2012, 45 (22), 9010-9019. 26. Erhan, S. M.; Kleiman, R., Factice from oil mixtures. Journal of the American Oil Chemists’ Society 1993, 70 (3), 309-311. 27. Cowan, J.; Ault, W.; Teeter, H., Polymerization of drying oils. Industrial & Engineering Chemistry 1946, 38 (11), 1138-1144. 28. Li, F.; Larock, R. C., New soybean oil–styrene–divinylbenzene thermosetting copolymers. II. Dynamic mechanical properties. J. Polym. Sci., Part B: Polym. Phys. 2000, 38 (21), 2721-2738. 29. Li, F.; Larock, R. C., New soybean oil‐styrene‐divinylbenzene thermosetting copolymers. III. Tensile stress–strain behavior. J. Polym. Sci., Part B: Polym. Phys. 2001, 39 (1), 60-77. 30. Li, F.; Larock, R. C., New soybean oil–styrene–divinylbenzene thermosetting copolymers. v. shape memory effect. J. Appl. Polym. Sci. 2002, 84 (8), 1533-1543. 31. Li, F.; Larock, R. C., New soybean oil–styrene–divinylbenzene thermosetting copolymers. VI. Time–temperature–transformation cure diagram and the effect of curing conditions on the thermoset properties. Polym. Int. 2003, 52 (1), 126-132. 32. Li, F.; Larock, R. C., New soybean oil–styrene–divinylbenzene thermosetting copolymers. I. Synthesis and characterization. J. Appl. Polym. Sci. 2001, 80 (4), 658-670. 33. Kammann Jr, K. P.; Phillips, A. I., Sulfurized vegetable oil products as lubricant additives. Journal of the American Oil Chemists’ Society 1985, 62 (5), 917-923. 34. Xia, Y.; Lu, Y.; Larock, R. C., Ring-opening metathesis polymerization (ROMP) of norbornenyl-functionalized fatty alcohols. Polymer 2010, 51 (1), 53-61. 35. Valverde, M.; Yoon, S.; Bhuyan, S.; Larock, R. C.; Kessler, M. R.; Sundararajan, S., Rubbers based on conjugated soybean oil: Synthesis and characterization. Macromolecular Materials and Engineering 2011, 296 (5), 444-454. 36. L. Yang, H. D., A. Yi, B. Lin and G. Li, Structure and properties of partially epoxidized soybean oil. J. Therm. Anal. Calorim. 2008, 93 (3), 875-879.

189

37. Sahakaro, K.; Beraheng, A., Epoxidized natural oils as the alternative safe process oils in rubber compounds. Rubber Chem. Technol. 2011, 84 (2), 200-214. 38. Yin, J.; Wang, S.; Lv, F., Improving the short-term aging resistance of asphalt by addition of crumb rubber radiated by microwave and impregnated in epoxidized soybean oil. Construction and Building Materials 2013, 49 (0), 712-719. 39. Parker, S. P., Dictionary of Scientific and Technical Terms. 6 ed.; McGraw-Hill: London, 2002; p 2380. 40. Sheridan, M. F., The Vanderbilt Rubber Handbook. 14th ed.; R.T. Vanderbilt Company, Inc.: Norwalk, 2010; p 68. 41. Sheridan, M. F., The Vanderbilt Rubber Handbook. 14th ed.; R.T. Vanderbilt Company, Inc.: Norwalk, 2010; p 57. 42. Rader, C. P., Basic Elastomer Technology. The Rubber Division American Chemical Society: Akron, 2001; p 264. 43. Nagata, N.; Kobatake, T.; Watanabe, H.; Ueda, A.; Yoshioka, A., Effect of chemical modification of solution-polymerized rubber on dynamic mechanical properties in carbon-black-filled vulcanizates. Rubber Chem. Technol. 1987, 60 (5), 837-855. 44. Tsutsumi, F.; Sakakibara, M.; Oshima, N., Structure and dynamic properties of solution SBR coupled with tin compounds. Rubber Chem. Technol. 1990, 63 (1), 8-22. 45. Sierra, C. A.; Galán, C.; Gómez Fatou, J. M.; Quiteria, V. R. S., Dynamic-mechanical properties of tin-coupled SBRs. Rubber Chem. Technol. 1995, 68 (2), 259-266. 46. Feng, H.; Zhang, X.; Zhao, S., Tin-coupled star-shaped block copolymer of styrene and butadiene (II) properties and application. J. Appl. Polym. Sci. 2009, 111 (2), 602-611. 47. Quiteria, V. R.-S.; Sierra, C. A.; Gómez-Fatou, J. M.; Galán, C.; Fraga, L. M., Tin-coupled styrene-butadiene rubbers (SBRs). Relationship between coupling type and properties. Die Angewandte Makromolekulare Chemie 1997, 246 (1), 85-96. 48. Mochel, W.; Nichols, J.; Mighton, C., The structure of neoprene. I. The molecular weight distribution of neoprene type GN. J. Am. Chem. Soc. 1948, 70 (6), 2185-2190. 49. Colbert, G. P., Basic Elastomer Technology. In Basic Elastomer Technology, Baranwal, K. C.; Stephens, H. L., Eds. The Rubber Division American Chemical Society: Akron, 2001; p 387. 50. Zhovner, N. A.; Zakharov, N. D.; Orekhov, S. V., Sulphur vulcanization of chloroprene rubber: Comparison with other diene elastomers. Polymer Science U.S.S.R. 1971, 13 (1), 85-92. 51. Pariser, R., NeuereErgebnisse zur Elastomeren-Vernetzung. English: German Plastics Digest Kunststoffe 1960, 50, 623. 52. Vukov, R., Zinc oxide crosslinking chemistry of halobutyl elastomers-a model compound approach. Rubber Chem. Technol. 1984, 57 (2), 284-290. 53. Miyata, Y.; Atsumi, M., Zinc oxide crosslinking reaction of polychloroprene rubber. Rubber Chem. Technol. 1989, 62 (1), 1-12. 54.Berry, K.; Liu, M.; Chakraborty, K.; Pullan, N.; West, A.; Sammon, C.; Topham, P. D., Mechanism for cross-linking polychloroprene with ethylene thiourea and zinc oxide. Rubber Chem. Technol. 2015, 88 (1), 80-97.

190

55. Das, A.; Naskar, N.; Datta, R. N.; Bose, P. P.; Debnath, S. C., Naturally occurring amino acid: Novel curatives for chloroprene rubber. J. Appl. Polym. Sci. 2006, 100 (5), 3981-3986. 56. Das, A.; Naskar, N.; Basu, D. K., Thiophosphoryl disulfides as crosslinking agents for chloroprene rubber. J. Appl. Polym. Sci. 2004, 91 (3), 1913-1919. 57. Rajeev, R. S.; De, S. K., Crosslinking of rubbers by fillers. Rubber Chem. Technol. 2002, 75 (3), 475-510. 58. Sae-Oui, P.; Sirisinha, C.; Hatthapanit, K., Effect of blends ratio on aging, oil and ozone resistance of silica-filled chloroprene rubber/natural rubber (CR/NR) blends. Express Polymer Letters 2007, 1 (1), 8-12. 59. Zheng, J.; Tan, J.; Gao, H.; Wang, C.; Dong, Z., Preparation of low temperature resistant and high electrical insulation chloroprene rubber-butadiene rubber blends. Rubber Chem. Technol. 2014, 87 (2), 360-369. 60. Das, A.; Ghosh, A.; Basu, D., Evaluation of physical and electrical properties of chloroprene rubber and natural rubber blends. KGK-Kautschuk Gummi Kunststoffe 2005, 58 (5), 230-238. 61. Fusco, J. V., Basic Elastomer Technology. In Basic Elastomer Technology, Baranwal, K. C.; Stephens, H. L., Eds. The Rubber Division American Chemical Society: Akron, 2001; p 313. 62. Van Amerongen, G., Influence of structure of elastomers on their permeability to gases. J. Polym. Sci. 1950, 5 (3), 307-332. 63. Fusco, J. V., Basic Elastomer Technology. In Basic Elastomer Technology, Baranwal, K. C.; Stephens, H. L., Eds. The Rubber Division American Chemical Society: Akron, 2001; p 324. 64.Fusco, J. V.; Hous, P., Butyl and halobutyl rubbers. In Rubber Technology, Morton, M., Ed. Springer: Linden, New Jersey, 1999; pp 284-321. 65. ExxonMobil, Butyl Rubber Compounding and Applications Manual. 66. ExxonMobil, Bromoutyl Rubber Compounding and Applications Manual. 67. Liang, Y.; Cao, W.; Li, Z.; Wang, Y.; Wu, Y.; Zhang, L., A new strategy to improve the gas barrier property of isobutylene–isoprene rubber/clay nanocomposites. Polym. Test. 2008, 27 (3), 270-276. 68. McDonel, E. T.; Baranwal, K. C.; Andries, J. C., Elastomer blends in tires. In Polymer Blends, Paul, D. R.; Newman, S., Eds. Academic Press, Inc.: New York, 1978; Vol. 2, p 266. 69. Rehner, J.; Holowchak, J., Determination of total and combined sulfur in butyl rubber. Rubber Chem. Technol. 1944, 17 (3), 738-744. 70. Gan, L. M.; Soh, G. B.; Ong, K. L., Vulcanization of butyl rubber by p-quinone dioxime. Rubber Chem. Technol. 1978, 51 (2), 267-277. 71. Tawney, P. O.; Little, J. R.; Viohl, P., The vulcanization of butyl rubber with phenol formaldehyde derivatives. Rubber Chem. Technol. 1960, 33 (1), 229-236. 72.Morrissey, R. T., Butyl-type polymers containing bromine. Rubber Chem. Technol. 1955, 28 (4), 952-967. 73. Dutta, N.; Khastgir, D.; Tripathy, D. K., The effect of carbon black concentration on the dynamic mechanical properties of bromobutyl rubber. Journal of Materials Science 1991, 26 (1), 177-188.

191

74. Lewis, C.; Buanpa, R.; Kiatkamjornwong, S., Effect of rubber ratio, carbon black level, and accelerator level on natural rubber/bromobutyl rubber blend properties. J. Appl. Polym. Sci. 2003, 90 (11), 3059-3068. 75. Feng, W.; Isayev, A. I., Continuous ultrasonic devulcanization of unfilled butyl rubber. J. Appl. Polym. Sci. 2004, 94 (3), 1316-1325. 76. Feng, W.; Isayev, A. I., Recycling of tire-curing bladder by ultrasonic devulcanization. Polymer Engineering & Science 2006, 46 (1), 8-18. 77. Sheridan, M. F., The Vanderbilt Rubber Handbook. 14th ed.; R.T. Vanderbilt Company, Inc.: Norwalk, 2010; p 492. 78. Stern, H. J., Rubber: Natural and Synthetic. In Rubber: Natural and Synthetic, Maclaren: New York, 1967; p 277. 79. Donnet, J.-B., Carbon black: Science and technology. CRC Press: Boca Raton, 1993. 80. Brunauer, S.; Emmett, P. H.; Teller, E., Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 1938, 60 (2), 309-319. 81. Wang, M.-J., Effect of polymer-filler and filler-filler interactions on dynamic properties of filled vulcanizates. Rubber Chem. Technol. 1998, 71 (3), 520-589. 82. Hess, W., Characterization of dispersions. Rubber Chem. Technol. 1991, 64 (3), 386-449. 83. Ayala, J.; Hess, W.; Dotson, A.; Joyce, G., New studies on the surface properties of carbon blacks. Rubber Chem. Technol. 1990, 63 (5), 747-778. 84. Ulmer, J.; Chirico, V.; Scott, C., The effect of carbon black type on the dynamic properties of natural rubber. Rubber Chem. Technol. 1973, 46 (4), 897-926. 85. Wolff, S.; Wang, M.-J.; Tan, E.-H., Filler-elastomer interactions. Part VII. Study on bound rubber. Rubber Chem. Technol. 1993, 66 (2), 163-177. 86. Gessler, A., Effect of mechanical shear on the structure of carbon black in reinforced elastomers. Rubber Chem. Technol. 1970, 43 (5), 943-959. 87. Medalia, A. I., Morphology of aggregates: VI. Effective volume of aggregates of carbon black from electron microscopy; Application to vehicle absorption and to die swell of filled rubber. J. Colloid Interface Sci. 1970, 32 (1), 115-131. 88. Wagner, M. P., Reinforcing silicas and silicates. Rubber Chem. Technol. 1976, 49 (3), 703-774. 89. Iler, R. K., The chemistry of silica. Wiley, New York: 1979. 90. Kim, K.-J.; VanderKooi, J., Moisture effects on TESPD-Silica/CB/SBR compounds. Rubber Chem. Technol. 2005, 78 (1), 84-104. 91. Sarkawi, S. S.; Dierkes, W. K.; Noordermeer, J. W. M., Reinforcement of natural rubber by precipitated silica: The influence of processing temperature. Rubber Chem. Technol. 2013, 87 (1), 103-119. 92. Wang, M.-J., The role of filler networking in dynamic properties of filled rubber. Rubber Chem. Technol. 1999, 72 (2), 430-448. 93. Böhm, G. G.; Nguyen, M. N., Flocculation of carbon black in filled rubber compounds. I. Flocculation occurring in unvulcanized compounds during annealing at elevated temperatures. J. Appl. Polym. Sci. 1995, 55 (7), 1041-1050. 94. Lin, C. J.; Hergenrother, W. L.; Alexanian, E.; Böhm, G. G. A., On the filler flocculation in silica-filled rubbers Part I. Quantifying and tracking the filler flocculation

192

and polymer-filler interactions in the unvulcanized rubber compounds. Rubber Chem. Technol. 2002, 75 (5), 865-890. 95. Lin, C. J.; Hogan, T. E.; Hergenrother, W. L., On the filler flocculation in silica and carbon black filled rubbers: Part II. Filler flocculation and polymer-filler interaction. Rubber Chem. Technol. 2004, 77 (1), 90-114. 96. Wolff, S.; Wang, M.-J., Filler—elastomer interactions. Part IV. The effect of the surface energies of fillers on elastomer reinforcement. Rubber Chem. Technol. 1992, 65 (2), 329-342. 97. Wang, M.-J.; Wolff, S.; Donnet, J.-B., Filler-elastomer interactions. Part I: Silica surface energies and interactions with model compounds. Rubber Chem. Technol. 1991, 64 (4), 559-576. 98. Wang, M.-J.; Wolff, S., Filler-elastomer interactions. Part V. Investigation of the surface energies of silane-modified silicas. Rubber Chem. Technol. 1992, 65 (4), 715-735. 99. Mihara, S.; Datta, R. N.; Noordermeer, J. W. M., Flocculation in silica reinforced rubber compounds. Rubber Chem. Technol. 2009, 82 (5), 524-540. 100. Wolff, S.; Gorl, U.; Wang, M.; Wolff, W., Silica-based tread compounds. European rubber journal 1994, 176 (1), 16-19. 101. Wolff, S., Chemical aspects of rubber reinforcement by fillers. Rubber Chem. Technol. 1996, 69 (3), 325-346. 102. Rader, C. P., Basic Elastomer Technology. In Basic Elastomer Technology, Baranwal, K. C.; Stephens, H. L., Eds. The Rubber Division American Chemical Society: Akron, 2001; p 175. 103. Laning, S. H.; Wagner, M. P.; Sellers, J. W., The determination of zinc oxide in rubber vulcanizates by x‐ray diffraction. J. Appl. Polym. Sci. 1959, 2 (5), 225-230. 104. Reuvekamp, L. A. E. M.; Debnath, S. C.; Ten Brinke, J. W.; Van Swaaij, P. J.; Noordermeer, J. W. M., Effect of zinc oxide on the reaction of TESPT silane coupling agent with silica and rubber. Rubber Chem. Technol. 2004, 77 (1), 34-49. 105. Brinke, J. W. t.; Debnath, S. C.; Reuvekamp, L. A. E. M.; Noordermeer, J. W. M., Mechanistic aspects of the role of coupling agents in silica–rubber composites. Compos. Sci. Technol. 2003, 63 (8), 1165-1174. 106. Kaewsakul, W.; Sahakaro, K.; Dierkes, W. K.; Noordermeer, J. W. M., Optimization of rubber formulation for silica-reinforced natural rubber compounds. Rubber Chem. Technol. 2012, 86 (2), 313-329. 107. Robertson, C. G.; Lin, C. J.; Bogoslovov, R. B.; Rackaitis, M.; Sadhukhan, P.; Quinn, J. D.; Roland, C. M., Flocculation, reinforcement, and glass transition effects in silica-filled styrene-butadiene rubber. Rubber Chem. Technol. 2011, 84 (4), 507-519. 108. Luginsland, H.-D.; Fröhlich, J.; Wehmeier, A., Influence of different silanes on the reinforcement of silica-filled rubber compounds. Rubber Chem. Technol. 2002, 75 (4), 563-579. 109. Voet, A.; Morawski, J. C.; Donnet, J. B., Reinforcement of elastomers by silica. Rubber Chem. Technol. 1977, 50 (2), 342-355. 110. Wolff, S., Silanes in tire compounding after ten years-a review. Tire Science and Technology 1987, 15 (4), 276-294. 111. Luginsland, H.-D.; Niedermeier, W., New reinforcing materials for rising tire performance demands. Rubber world 2003, 228 (6), 34-45.

193

112. Wang, M.-J.; Kutsovsky, Y., Effect of fillers on wet skid resistance of tires. Part II: Experimental observations on effect of filler-elastomer interactions on water lubrication. Rubber Chem. Technol. 2008, 81 (4), 576-599. 113. Byers, J. T., Fillers for balancing passenger tire tread properties. Rubber Chem. Technol. 2002, 75 (3), 527-548. 114. Liu, X.; Zhao, S.; Yang, Y.; Zhang, X.; Wu, Y., Structure and properties of star‐shaped solution‐polymerized styrene‐butadiene rubber and its co‐coagulated rubber filled with silica/carbon black‐I: Morphological structure and mechanical properties. Polym. Adv. Technol. 2009, 20 (11), 818-825. 115. Wang, L.; Zhao, S., Study on the structure‐mechanical properties relationship and antistatic characteristics of SSBR composites filled with SiO2/CB. J. Appl. Polym. Sci. 2010, 118 (1), 338-345. 116. Rattanasom, N.; Saowapark, T.; Deeprasertkul, C., Reinforcement of natural rubber with silica/carbon black hybrid filler. Polym. Test. 2007, 26 (3), 369-377. 117. Dong, B.; Liu, C.; Wu, Y.-P., Fracture and fatigue of silica/carbon black/natural rubber composites. Polym. Test. 2014, 38, 40-45. 118. Wang, M.-J.; Kutsovsky, Y.; Zhang, P.; Murphy, L. J.; Laube, S.; Mahmud, K., New generation carbon-silica dual phase filler Part I. Characterization and application to passenger tire. Rubber Chem. Technol. 2002, 75 (2), 247-263. 119. Ma, J.-H.; Zhang, L.-Q.; Wu, Y.-P., Characterization of filler-rubber interaction, filler network structure, and their effects on viscoelasticity for styrene-butadiene rubber filled with different fillers. Journal of Macromolecular Science, Part B 2013, 52 (8), 1128-1141. 120. Benson, S. W., Thermochemistry and kinetics of sulfur-containing molecules and radicals. Chem. Rev. 1978, 78 (1), 23-43. 121. Layer, R. W., Recuring vulcanizates. I. A novel way to study the mechanism of vulcanization. Rubber Chem. Technol. 1992, 65 (1), 211-222. 122. Datta, R. N., Rubber Curing Systems. Rapra Technology: Shrewsbury, 2001; Vol. 12. 123. Lattimer, R. P.; Kinsey, R. A.; Layer, R. W.; Rhee, C., The mechanism of phenolic resin vulcanization of unsaturated elastomers. Rubber Chem. Technol. 1989, 62 (1), 107-123. 124. Feng, W.; Isayev, A. I., Blends of ultrasonically devulcanized tire-curing bladder and butyl rubber. Journal of Materials Science 2005, 40 (11), 2883-2889. 125. Sieron, J., High temperature elastomers for extreme aerospace environments. Rubber Chem. Technol. 1966, 39 (4), 1141-1160. 126. Li, J.; Isayev, A. I.; Ren, X.; Soucek, M. D., Modified soybean oil-extended sbr compounds and vulcanizates filled with carbon black. Polymer 2015, 60, 144-156. 127. Flory, P. J.; Rehner Jr, J., Statistical mechanics of cross‐linked polymer networks II. Swelling. The Journal of Chemical Physics 1943, 11, 521. 128. Schuring, D. J.; Futamura, S., Rolling loss of pneumatic highway tires in the eighties. Rubber Chem. Technol. 1990, 63 (3), 315-367. 129. Kraus, G., Swelling of filler‐reinforced vulcanizates. J. Appl. Polym. Sci. 1963, 7 (3), 861-871.

194

130. Li, J.; Isayev, A. I.; Ren, X.; Soucek, M. D., Toward replacement of petroleum oils by modified soybean oils in elastomers. Rubber Chem. Technol. 2016, 89 (4), 608-630. 131. Nandanan, V.; Joseph, R.; George, K., Rubber seed oil: A multipurpose additive in NR and SBR compounds. J. Appl. Polym. Sci. 1999, 72 (4), 487-492. 132. Du, F.; Scogna, R. C.; Zhou, W.; Brand, S.; Fischer, J. E.; Winey, K. I., Nanotube networks in polymer nanocomposites: Rheology and electrical conductivity. Macromolecules 2004, 37 (24), 9048-9055. 133. Payne, A. R., The dynamic properties of carbon black‐loaded natural rubber vulcanizates. Part I. J. Appl. Polym. Sci. 1962, 6 (19), 57-63. 134. Okel, T. A.; Waddell, W. H., Silica properties/rubber performance correlation. Carbon black-filled rubber compounds. Rubber Chem. Technol. 1994, 67 (2), 217-236. 135. Li, J.; Isayev, A. I.; Ren, X.; Soucek, M. D., Norbornylized soybean oil as sustainable new plasticizer for rubbers with hybrid fillers. Polym. Int. 2017. 136. Feng, W.; Tang, Z.; Weng, P.; Guo, B., Correlation of filler networking with reinforcement and dynamic properties of SSBR/carbon black/silica composites. Rubber Chem. Technol. 2015, 88 (4), 676-689. 137. Mora-Barrantes, I.; Ibarra, L.; Rodríguez, A.; González, L.; Valentín, J., Elastomer composites based on improved fumed silica and carbon black. Advantages of mixed reinforcing systems. J. Mater. Chem. 2011, 21 (43), 17526-17533. 138. Choi, S.-S.; Nah, C.; Lee, S. G.; Joo, C. W., Effect of filler–filler interaction on rheological behaviour of natural rubber compounds filled with both carbon black and silica. Polym. Int. 2003, 52 (1), 23-28. 139. Noriman, N. Z.; Ismail, H., Properties of styrene butadiene rubber (SBR)/recycled acrylonitrile butadiene rubber (NBR) blends: The effects of carbon black/silica (CB/Sil) hybrid filler and silane coupling agent, Si69. J. Appl. Polym. Sci. 2012, 124 (1), 19-27. 140. Eggers, H.; Schummer, P., Reinforcement mechanisms in carbon black and silica loaded rubber melts at low stresses. Rubber Chem. Technol. 1996, 69 (2), 253-265. 141. Li, J.; Isayev, A. I.; Ren, X.; Soucek, M. D., Effect of norbornyl modified soybean oil on CB‐filled chloroprene rubber. J. Appl. Polym. Sci. 2016, 133 (33). 142. Kraus, G., Interactions of elastomers and reinforcing fillers. Rubber Chem. Technol. 1965, 38 (5), 1070-1114. 143. Chen, Y.; Xu, C., Stress softening of NR reinforced by in situ prepared zinc dimethacrylate. J. Appl. Polym. Sci. 2012, 123 (2), 833-841. 144. Wang, Z.; Liu, J.; Wu, S.; Wang, W.; Zhang, L., Novel percolation phenomena and mechanism of strengthening elastomers by nanofillers. PCCP 2010, 12 (12), 3014-3030. 145. Kok, C. M.; Yee, V. H., The effects of crosslink density and crosslink type on the tensile and tear strengths of NR, SBR and EPDM gum vulcanizates. Eur. Polym. J. 1986, 22 (4), 341-345. 146. Hamed, G. R., Effect of crosslink density on the critical flaw size of a simple elastomer. Rubber Chem. Technol. 1983, 56 (1), 244-251. 147. Studebaker, M. L.; Beatty, J. R., The oxidative hardening of SBR. Rubber Chem. Technol. 1972, 45 (2), 450-466.

195

148. Keller, R. W., Oxidation and ozonation of rubber. Rubber Chem. Technol. 1985, 58 (3), 637-652. 149. Hamed, G. R.; Zhao, J., Tensile behavior after oxidative aging of gum and black-filled vulcanizates of SBR and NR. Rubber Chem. Technol. 1999, 72 (4), 721-730. 150. Kaisersberger, E.; Knappe, S.; Möhler, H.; Rahner, S., NETZSCH annual for science and industry NETZSCH-Geratebau GmbH: Selb, Germany, 1993; Vol. 2. 151. Das, A.; Costa, F. R.; Wagenknecht, U.; Heinrich, G., Nanocomposites based on chloroprene rubber: Effect of chemical nature and organic modification of nanoclay on the vulcanizate properties. Eur. Polym. J. 2008, 44 (11), 3456-3465. 152. Sae-Oui, P.; Sirisinha, C.; Thepsuwan, U.; Hatthapanit, K., Dependence of mechanical and aging properties of chloroprene rubber on silica and ethylene thiourea loadings. Eur. Polym. J. 2007, 43 (1), 185-193. 153. Fetters, L. J.; Kiss, A. D.; Pearson, D. S.; Quack, G. F.; Vitus, F. J., Rheological behavior of star-shaped polymers. Macromolecules 1993, 26 (4), 647-654. 154. Graessley, W. W.; Roovers, J., Melt rheology of four-arm and six-arm star polystyrenes. Macromolecules 1979, 12 (5), 959-965. 155. Wang, X.; Jia, H.; Ji, Q.; Jin, Z.; Wen, W.; Zhang, S., Effects of microstructure of SBR on the properties of SBR vulcanizates. Journal of Nanjing University of Science and Technology 2000, 3, 004.

196