SILICA-REINFORCED NATURAL RUBBER

USE OF NATURAL RUBBER GRAFTED WITH CHEMICAL

FUNCTIONALITIES AS COMPATIBILIZER
 This research is a joint project between the University of Twente and Prince
of Songkla University, sponsored by the Netherlands Natural Rubber
Foundation.

Graduation committee
Chairman: Prof. Dr. G.P.M.R. Dewulf University of Twente, CTW
Secretary: Prof. Dr. G.P.M.R. Dewulf University of Twente, CTW
Promotor: Prof. Dr. Ir. J.W.M. Noordermeer University of Twente, CTW
Asst. Promotor: Dr. K. Sahakaro University of Twente,CTW and
Prince of Songkla University,
Science and Technology
Members: Prof. Dr. Ir. D.J. Schipper University of Twente, CTW
Dr. Ir. P.J. Dijkstra University of Twente, TNW
Prof. Dr. U. Giese University of Hanover, DIK
Hanover, Germany
Prof. Dr. J. Vourinen Tempere University of
Technology, Finland
Referees: Dr. Ir. L.A.E.M. Reuvekamp Apollo Tyres Global R&D,
Enschede
Dr. A.V. Chapman Tun Abdul Razak Research
Centre, UK

Silica-reinforced natural rubber: use of natural rubber grafted with chemical functionalities
as compatibilizer

By Karnda Sengloyluan

Ph.D. Thesis, University of Twente, Enschede, the Netherlands, and Prince of Songkla
University, Pattani Campus, Thailand, 2015.

Copyright Karnda Sengloyluan, 2015.

All rights reserved.
Cover design by Karnda Sengloyluan
Printed at Wöhrmann Print Service, PO Box 92, 7200 AB Zutphen, the Netherlands.

ISBN: 978-90-365-3898-5
DOI: 10.3990/1.9789036538985
USL: http://dx.doi.org/10.3990/1.9789036538985
SILICA-REINFORCED NATURAL RUBBER

USE OF NATURAL RUBBER GRAFTED WITH CHEMICAL

FUNCTIONALITIES AS COMPATIBILIZER

DISSERTATION

to obtain
the degree of doctor at the University of Twente,
on the authority of the rector magnificus,
Prof. Dr. H. Brinksma,
on account of the decision of the graduation committee,
to be publicly defended
on Thursday, June 25th, 2015 at 14.45

by
Karnda Sengloyluan
born on November 29th, 1985
in Satun, Thailand
This dissertation has been approved by:

Promotor : Prof. Dr. Ir. J.W.M. Noordermeer

Assistant Promotor : Dr. K. Sahakaro
“Defeat is not the worst of failures. Not to have tired is the true failure.”

----- “George E. Woodberry” -----

To my beloved family
 Table of Contents

Chapter 1 Introduction 1
Chapter 2 Literature overview: reinforcement efficiency of silica-filled 7
rubber with different compatibilizing techniques

Chapter 3 Silica-reinforced tire tread compounds compatibilized by 45

using epoxidized natural rubber

Chapter 4 Silica-reinforced natural rubber compounds 73

compatibilized by ENR in combination with TESPT and
sulfur compensation

Chapter 5 Silica-reinforced natural rubber compounds 97

compatibilized by ENR in combination with different silane
coupling agent types

Chapter 6 Influence of types of silane coupling agents on the 127

reinforcement of silica in natural rubber compounds

Chapter 7 Preparation and characterization of silane-grafted natural 151

rubber

Chapter 8 Silane-grafted natural rubber as compatibilizer in silica- 177

reinforced natural rubber

Chapter 9 Compatibilization of silica-filled natural rubber by using 203

silane-grafted-natural rubber with compensation of sulfur

Summary 219

Samenvatting 225

Symbols and abbreviations 231

Bibliography 235

Acknowledgements 237

Curriculum vitae 239

 CHAPTER 1

Introduction

1.1 Development of silica technology for tires

A major development of tires was first reported in 1846 when R.W. Thomson
invented an elastomeric air tube or “pneumatic tire” to be fixed onto a wheel to reduce the
power to haul a carriage and wheeling noise. The pneumatic tire was successfully used
for tricycles and bicycles when this concept was reinvented by J.B. Dunlop in 1888.[1,2]
The early development of the pneumatic tire was meant only for bicycles, later on it was
mainly used for automobiles or motorcars. For decades, the chemicals and processing
technology involved in rubber compounds were continuously improved, and both World
Wars were important factors to accelerate tire design and development including the
introduction of synthetic rubbers and reinforcement materials.[3]
The discovery of the vulcanization reaction by Charles Goodyear in 1839[2] and
the development of rubber tires lead to a continuous consumption increase of rubber and
reinforcement materials. The technology of reinforcing fillers was one of the important
factors to accelerate the growth of the rubber industry. Fillers such as chalk and china
clay, were originally used to reduce production costs, but they also improved the
processing behavior of rubber compounds such as reduced die swell and smoothening of
the extruded and calendered products.[4] Carbon black was primarily used as a coloring
pigment until the reinforcement effect of carbon black was discovered by S.C. Mote in
1904[2] in line with the development of rubber tire technology.
Carbon black has a graphite structure with a small amount of functional groups,
e.g. phenol, ketone, lactone, carboxyl, etc, on the surface. It can easily be incorporated
into rubber and forms filler-rubber interactions.[5] The use of carbon black in rubber
compounds leads to an improvement of mechanical properties such as abrasion
resistance and tensile and tear strengths.[6] The properties of the filled rubber are mainly
affected by filler dispersion and distribution, carbon black characteristics and filler-rubber
interactions.[7] Carbon black has a good compatibility with hydrocarbon rubbers, but is
limited in use for only black applications. In 1951, the first commercial silica product was
marketed under the trade name of Ultrasil VN3 in Europe. Silica was initially used in light-
Chapter 1

colored or transparent articles, e.g., shoe soles. For tire compounds, precipitated silica
was first used in small amounts in combination with carbon black to improve adhesion in
tire cord-rubber, and to improve cutting and chipping resistances.[8] After a patent of
Michelin on the so-called “Green Tire”[9], silica-silane technology has been increasingly
used for passenger car tire treads. Silica is an inorganic filler with hydroxyl, also called
silanol groups on its surface that can form hydrogen bonds and thus strong filler-filler
interactions to generate silica aggregates and agglomerates. It is therefore difficult to
disperse and does not interact with non-polar rubber matrices like natural rubber (NR) and
styrene-butadiene rubber (SBR), which are the types of rubbers used in tire compounds.
To improve the properties of silica-filled compounds, silane coupling agents are generally
added to improve silica-rubber interaction and to decrease the silica-silica interaction.
The use of silane coupling agents leads to reduction of compound viscosity, an
improvement in cure characteristics and a significant increase of modulus, tensile
strength, and abrasion resistance. The first silane coupling agent introduced by Union
Carbide, was mercaptopropyl-trimethoxysilane (MPS) under the trade name of A-189.
Afterwards, Bis-(triethoxysilylpropyl)tetrasulfide (TESPT) was introduced under the trade
name Si-69 by Degussa.[10] The use of the silica/TESPT combination in tire tread
compounds improves wet traction and rolling resistance without negative affects on
abrasion resistance. Silica technology has been further developed to improve the
properties of silica-filled compounds, either in aspects of the silica characteristics or
development of new silane coupling agents or compatibilizer systems.

1.2 Background of the thesis

Natural rubber (NR) is an important material to produce rubber tires, especially

heavy duty truck tires. NR has superior mechanical properties over other synthetic
rubbers, due to its high molecular weight, and regular structure of cis-1,4-polyisoprene.
NR can crystallize when it is stretched by external force, so called „strain-induced
crystallization‟, which leads to high tensile strength and elongation at break. Despite the
good strength of NR, reinforcing fillers such as carbon black and silica are commonly used
to enhance the properties of NR further for high performance products such as tires. Due
to the strong filler-filler interactions between silica aggregates caused by the large number
of silanol groups on the silica surface, the use of silica in NR or other hydrocarbon rubbers
requires a silane coupling agent or compatibilizer. The most commonly used silane
coupling agent is TESPT, but some drawbacks of the use of this silane in silica-filled
compounds have been reported. These are an increase of Mooney viscosity upon storage

2
 Introduction

because of self-crosslinking between sulfur in the silane structure and rubber molecules,
leading to less processability of the compound,[11] and low scorch safety due to breaking
of the sulfur-sulfur bonds in the TESPT molecules during mixing at high temperature. Bis-
(triethoxysilylpropyl)disulfide (TESPD) can be used as alternative to obtain a better scorch
safety of silica-filled compounds while providing similar final properties when compared to
those of TESPT treated compounds.[12]
Silica contains silanol groups on its surface which can interact or react with polar
functional groups of some polar rubbers and enhance compatibilization. Acrylonitrile
butadiene rubber (NBR)[13] and polychloroprene rubber[14] have been used as
compatibilizers in silica-filled compounds to improve silica-rubber interaction, silica
dispersion and hence the vulcanizate properties. Chemically modified NRs with polar
functional groups such as epoxidized natural rubber (ENR)[15], maleated natural rubber
(MNR)[16] have also been used as compatibilizers in silica-filled compounds. The epoxide
groups of ENR can interact with silanol groups on the silica surface, leading to a decrease
of filler-filler interaction and an increase of the properties of the filled compounds.

1.3 Aim of the project

The research project covered in this thesis aimed to increase the compatibility
between silica and natural rubber by using chemically functionalized rubber as a
compatibilizer, so as to replace or minimize the use of silane coupling agent. The
objectives for the research are listed below;
1. To prepare modified natural rubbers by using different types and levels of
functional groups which can potentially interact or react with silanol groups on the silica
surface.
2. To investigate the effect of modified natural rubber when added as
compatibilizer in silica-filled natural rubber on cure characteristics, Mooney viscosity,
bound rubber content, mechanical and dynamic properties.
3. To select a suitable type and level of functionality of the modified natural
rubber to be used as compatibilizer in silica-filled natural rubber. The mixing parameters
are to be optimized to obtain optimum properties of silica-filled natural rubber.
4. To characterize the interaction, level of compatibility and reinforcement of
silica-filled natural rubber when the modified rubber is used as compatibilizer, compared to
use of a conventional silane coupling agent.

3
Chapter 1

1.4 Concept of the thesis

This project investigated the use of chemically functionalized natural rubbers as

compatibilizers in silica-reinforced natural rubber with emphasis on the improvement of
silica-rubber interaction, mechanical properties and tire performance, i.e. wet skid and
rolling resistance, in comparison with the use of a conventional silane coupling agent. The
thesis is composed of the following chapters;

Chapter 1 Introduction: This first chapter shortly reveals the development of

tires and filler technology with emphasis on tire compounds. It is then followed by a
description of the background and aims of this research.

Chapter 2 Literature overview: This part focuses on the use of reinforcing fillers
mainly carbon black and silica in rubber compounds. The reinforcing efficiency of silica,
factors influencing silica reinforcement and development of silica technology for tire
compounds are reviewed. The use of silane coupling agents, some polar materials and
chemically modified rubbers to improve silica-rubber interactions and the properties of
silica-filled compounds are discussed. This chapter ends by providing motivation and
scope of the project.

Chapter 3 Silica-reinforced tire tread compounds compatibilized by using

epoxidized natural rubber: This chapter first describes the preparation and
characterization of epoxidized natural rubber (ENR) with various mole% of epoxide. ENRs
with different mole% of epoxide groups are used as compatibilizers to optimize the
properties of silica-filled natural rubber compounds compared to compounds with Bis-
(triethoxysilylpropyl)tetrasulfide (TESPT) and without compatibilizer.

Chapter 4 Silica-reinforced natural rubber compounds compatibilized by

ENR in combination with TESPT and sulfur compensation: While a silica-filled natural
rubber compound with only ENR as compatibilizer shows overall lower properties than
with TESPT silane coupling agent, this section studies the use of an optimized ENR type
and content (i.e. 7.5 phr of ENR-51) in combination with TESPT to further enhance the
properties of the compounds. The TESPT contents are varied and the effect of extra sulfur
to compensate for the sulfur contents in TESPT molecules on the properties of silica-
reinforced natural rubber is investigated.

Chapter 5 Silica-reinforced natural rubber compounds compatibilized by

ENR in combination with different silane coupling agent types: The reinforcing

4
 Introduction

efficiency of silica-filled NR compounds compatibilized with three different silane types, i.e.
TESPT, 3-mercaptopropyl-di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-
363), and 3-octanoyl-thio-propyltriethoxysilane (NXT), is discussed. Then, these different
silane types together with ENR-51 at 7.5 phr are used as compatibilizers in silica-filled NR
compounds. The properties of such compounds are studied in comparison with the
reference compound containing the optimized conventional TESPT silane coupling agent.

Chapter 6 Influence of types of silane coupling agents on the reinforcement

of silica in natural rubber compounds: The effect of different types of silane coupling
agents on the properties of silica-filled NR compounds is studied by using four types of
silane; TESPT, NXT, VP Si-363 and vinyltriethoxysilane (VTES) that are used based on
molar and ethoxy functional group equivalents. The properties of the silica-filled NR
compounds and vulcanizates are comparatively investigated in order to find a grafting
silane candidate to be used later for preparation of silane-grafted NR.

Chapter 7 Preparation and characterization of silane-grafted-natural rubber:

This section focuses on the preparation and characterization of silane-grafted-natural
rubber. Silanes with mercapto function: VP Si-363, and with blocked mercapto function:
NXT are used to graft onto the natural rubber molecules in the melt mixing state in an
internal mixer. The preparation conditions are optimized and the products are
characterized by Fourier-transform infrared spectroscopy (FT-IR), proton nuclear magnetic
resonance spectroscopy (1H-NMR) and elemental analysis.

Chapter 8 Silane-grafted-natural rubber as compatibilizer in silica-

reinforced natural rubber: The use of NXT- and VP Si-363-grafted natural rubbers as
compatibilizers in the silica-reinforced natural rubber compounds are studied in
comparison with the straight use of silane. The properties of the silica-reinforced natural
rubber are studied in terms of filler-filler and filler-rubber interactions, mechanical and
dynamic mechanical properties by taking the compound with optimum amount of TESPT
as a reference.

Chapter 9 Compatibilization of silica-filled natural rubber by using silane-

grafted-rubber with compensation of sulfur: Due to the difference in total sulfur content
in the compounds, this chapter focuses on the further improvement of the use of silane-
grafted-natural rubber as compatibilizer by using extra sulfur to compensate for sulfur
atoms released from the TESPT molecules. Overall properties are assessed and
discussed in comparison with the use of the conventional TESPT silane coupling agent.

5
Chapter 1

Summary: This chapter summarizes all the findings and knowledge derived from
the experimental studies.

1.5 References

1. C.M. Blow, C. Hepburn, “Rubber Technology and Manufacture”, Butterworths,

London, second edition, 1982.
2. J.E. Mark, B. Erman, F.R. Eirich, “Science and Technology of Rubber”, Academic
Press, San Diego, second edition, 1994.
3. T. French, “Tire Technology”, IOP Publishing, 1988.
4. J.A. Brydson, “Rubber Materials and Their Compounds”, Elsevier Applied Science
Publishers, England, 1988.
5. S. Woff, Rubber Chem. Technol., 69, 325 (1996).
6. Z. Rigbi, Adv. Polym. Sci., 36, 21 (1980).
7. L. Kasasek, M. Sumita, J. Mater. Sci., 31, 281 (1996).
8. B. Schwaiger, A. Blume, Rubber World, 222, 32 (2000).
9. R. Rauline, EP Patent 0501227A1, to Michelin & Cie, February 9, 1992.
10. F.W. Barlow, “Rubber Compounding: Principles, Materials, and Techniques”, Marcel
Dekker, New York, 1988.
11. C.J. Lin, W.L. Hergenrother, A.S. Hilson, Rubber Chem. Technol., 75, 215 (2002).
12. J.W. ten Brinke, P.J. van Swaaij, L.A.E.M. Reuvekamp, J.W.M. Noordermeer,
Rubber Chem. Technol., 75, 12 (2003).
13. S.-S. Choi, J. Appl. Polym. Sci., 79, 1127 (2001).
14. A. Das, S.C. Debnath, D. De, D.K. Basu, J. Appl. Polym. Sci., 93, 196 (2004).
15. K.M. George, J.K. Varkey, K.T. Thomas, N.M. Mathew, J. Appl. Polym. Sci., 85, 292
(2002).
16. K. Sahakaro, S. Beraheng, J. Appl. Polym. Sci., 109, 3839 (2008).

6
 CHAPTER 2

Literature Overview: Reinforcement Efficiency of Silica-Filled

Rubber with Different Compatibilizing Techniques

This Chapter gives an overview of reinforcing fillers which have been used in the
rubber industry with particular attention to fillers used in tire compounds, e.g. carbon black
and silica. Filler characteristics, such as specific surface area, filler structure and surface
activity, that influence the properties of filled rubber are discussed. Filler reinforcement of
rubbers in general, and dynamic properties of filled rubber in relation to tire performance
are reviewed. Silica-reinforced rubber compounding and its associated difficulties coming
from silica-rubber incompatibility are addressed. This leads to the development of silica-
silane technology which involves a silanization reaction between alkoxy groups of a silane
and silanol groups on the silica surface. Due to the greatly increased interest in silica
usage especially for low rolling resistance tire treads, several alternative approaches have
been adopted in order to improve silica dispersion and filler-rubber interaction. Some of
those approaches are reviewed in this section. Among them the use of polar polymers,
such as polychloroprene rubber, acrylonitrile-butadiene rubber, epoxidized rubbers, as
compatibilizers for silica-reinforced rubber compounds. Particular attention is given to
chemically modified natural rubber which can potentially be used for silica-reinforced
natural rubber compounds, as investigated in this thesis. The chapter ends by providing
the motivation for this thesis.
Chapter 2

2.1 Introduction

During the past decades the rubber industry has increasing concerns about its
dependence on raw materials which are derived from petroleum. Synthetic rubbers,
carbon black, processing and extender oils, and most of the additives for rubber are
derived from petrochemical products which are affected by the increasing costs and
decreasing supply of crude oil. This situation has stimulated interests in products which
are less dependent on petroleum and require less energy to produce. Fillers for rubber
compounds that are made from virtually inexhaustible natural sources such as clay,
limestone, and talc require less energy to produce than the synthetic fillers like carbon
black, silicates and precipitated silica, but these are classified as non- to low-reinforcing
and have high specific gravities compared to carbon black and silica. Commercial
applications of elastomers often require the use of active particulate fillers to obtain a
certain level of reinforcement and product performance. In the rubber industry, beside
carbon black, silica is an important white reinforcing filler often used to impart specific
properties to the rubber compounds. It is well known that the filled-rubber compounds are
multiphase composites, in which several factors play a role on the reinforcement
efficiency.
In 1951, the first commercial silica product was marketed under the trade-name
of Ultrasil VN3 in Europe. Silica was initially used in light-colored or transparent articles,
e.g., shoe soles. Precipitated silica was also used in small amounts in tire treads for
[1]
commercial vehicles to improve tear propagation resistance. In 1990s, the European tire
manufacturer Michelin introduced passenger car tires with treads formulated by
incorporating silica as reinforcing filler instead of conventional carbon black. The tires with
silica-filled tire tread compounds were claimed as “Green Tires” due to their lower rolling
resistance and heat build-up, when compared with conventional tires with treads filled with
carbon black.[2] However, the surface functional environment of silica particles is different
from that of carbon black due to the existence of the hydrophilic silanol groups on the
surface. Thus, hydrocarbon rubbers like natural rubber (NR) and styrene butadiene
rubber (SBR) are not compatible with silica, and their compounds without compatibilizer or
coupling agent usually have inferior mechanical and physical properties as a result of poor
interfacial adhesion.

2.2 Reinforcing fillers

Several types of fillers are in use in rubber compounds for different purposes,
such as for reduction of production costs or to improve some properties like processing or

8
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

vulcanizate properties, or both. The fillers can be classified by their particle sizes as
shown in Figure 2.1.[3]

Figure 2.1 Classification of fillers on basis of particle sizes.[3]

2.2.1 Filler characteristics

Reinforcement basically relates to composites built from two or more

components which have different mechanical characteristics. The strength of one of these
components is imparted to the composite combined with the set of favorable properties of
another component. Reinforcement of rubber by fillers relates to the improvement of
modulus and failure properties, such as tensile strength, tear and abrasion resistances of
the vulcanizates. The energy at rupture which can be obtained from a stress-strain curve
may be regarded as the best single criterion for reinforcement.[4,5]
The reinforcement of rubber with fillers is dependent on their specific characters.
The important factors that affect the reinforcing efficiency are;[5,6]

1. Specific surface area and particle size - The primary particle size of a filler is
related to its specific surface area and can be determined directly by using electron
microscopy. Alternatively, it can be characterized by different adsorption methods. The
nitrogen adsorption method according to ASTM D6556, the so-called BET (Brunauer-
Emmett-Teller) method measures the total surface area including micro-porosity of the
particles. In this case, the nitrogen surface area (NSA) and the statistical thickness
surface area (STSA) can be determined. Another method is based on the cetyltrimethyl
ammonium bromide (CTAB) adsorption which analyzes only the external surface area of

9
Chapter 2

filler particles or aggregates which can be related to the contact area between filler and
rubber molecules, excluding the micro-pores in which they cannot penetrate. The CTAB-
adsorption on silica surface is measured according to ASTM D6845. It should be noted
that the CTAB-method for the measurement of carbon black surface area according to
ASTM D3765 has been withdrawn due to its poor testing precision and labor
intensiveness.[7] For the specific surface area of carbon black which is accessible for
rubber chains, the STSA technique is applied.[7] Decrease of filler particle size leads to an
increase of surface area which normally has a positive effect on rubber reinforcement.

2. Filler structure - Generally, fillers do not appear in individual primary particles

form, but preferably in cluster forms of aggregates or agglomerates, depending on how
strong the filler-filler interactions are. Carbon black can form filler structure through Van
der Waals forces which are a weaker interaction than hydrogen bonding between silica
particles. The empty space between aggregates (or agglomerates) of a filler can be
determined by the volume of dibutylphthalate (DBP) absorption according to ASTM
D6845-03, which number is used to indicate filler structure. The “structure” of a filler
relates to the aggregate structure size and density. Due to toxicity of DBP, the oil
adsorption number of carbon black was implemented according to ASTM D2414-02a.
Paraffinic oil is preferentially used offering the advantage of being non-hazardous. Filler
structure has an influence on reinforcing efficiency and the addition of high-structure
carbon black into rubber leads also to an increase of rubber vulcanizate properties.

3. Surface activity – The carbon black surface consists of a small amount of

different chemical functional groups such as carboxyl, quinone, lactone, phenol and
hydroxyl groups, causing a difference in capacity and absorption energy. The limited
number of polar functional groups does not essentially contribute to the reinforcement of
carbon black in non-polar rubbers. The surface activity of carbon black mainly refers to the
strength of interactions between the carbon black surface and rubber via either physical or
chemical adsorptions or mechanical interlocking which have an important effect on
modulus, hysteresis, abrasion resistance and other mechanical/physical properties.[8] The
surface activity of carbon black can be influenced by heat treatment at a high temperature
of 1600 to 3000°C, causing a rearrangement of nano-crystallites to be more ordered in
building primary particles of carbon black.[3] In case of inorganic silica with a large number
of hydroxyl groups on its surface, causing strong filler-filler interactions, it is essential to
use a silane coupling agent to create a bridge between silica and rubber.

10
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

2.2.2 Carbon black

Carbon black is composed of aggregated particles of elemental carbon which are

partly graphitic in structure. The carbon atoms in the particles are oriented in layered
planes which, by parallel alignment and overlapping, give the particles their semi-graphitic
nature. The outer layers are more graphitic than those in the center. The particle size
ranges from 10 to 400 nm in diameter, wherein the smaller ones are less graphitic.
Carbon blacks are produced by converting either liquid or gaseous hydrocarbon to carbon
and hydrogen by combustion or thermal decomposition.[9] The most important
characteristic of carbon black is its external specific surface area. High area is associated
with a high level of reinforcement, but at the expense of high cost, high hysteresis and
more difficult processing. The second most important property is its “structure”, which
refers to the bulkiness of the carbon black aggregates. In general, high structure or high
bulkiness carbon black is associated with a large number of carbon black primary particles
per aggregate.[10] It is well-known that filler-rubber interactions depend upon the
compatibility between filler and rubber matrix. Carbon black is a filler that can easily be
incorporated into rubber, generates bound rubber and is compatible with hydrocarbon
rubbers including styrene butadiene rubber (SBR), natural rubber (NR), polybutadiene
rubber (BR) and isobutylene isoprene rubber (IIR). The carbon black surface contains only
a small number of functional groups as shown in Figure 2.2, making it relatively non-polar
and easy to bound to rubber. The adsorption of rubber on carbon black forming bound
[8]
rubber readily occurs during mixing and also after mixing.

Figure 2.2 Functional groups on carbon black surfaces.[3]

11
Chapter 2

The functional groups on carbon black surfaces can react with ethoxy groups of
Bis-(triethoxysilylpropyl)tetrasulfide (TESPT) and become an activated carbon black
surface. Those activated carbon black surface can then form covalent bonds with rubber
and results in a reduction of compound hysteresis.[11]

2.2.3 Silica

In 1939, the first reinforcing siliceous filler was introduced.[9] A calcium silicate
was prepared by wet precipitation from a sodium silicate solution with calcium chloride. In
further development of the process, the calcium was leached out by hydrochloric acid to
yield a reinforcing silica pigment. About ten years later, a direct precipitated silica from
sodium silicate solution was developed to a commercial process and this became the
major process of today. In 1950, different types of anhydrous silica which were made by
reacting silicium-tetrachloride or “silicon chloroform” with water vapor in a hydrogen-
oxygen flame were produced. These pyrogenic silicas formed at high temperature (about
1400°C) have a lower concentration of hydroxyl groups on the surface than the
precipitated silica. The precipitated silica contains about 85-90% silica and has ignition
losses of 10-14%, whereas the pyrogenic silica contains 99.8% silica.[4]
Amorphous silica consists of silicium and oxygen tetrahedrally bonded into an
imperfect three dimensional structure. Silica contains a large number of silanol groups on
its surface, and these polar silanol groups lead to a hydrophilic surface. Thus, silica is not
compatible with hydrophobic rubbers. Silica particles form strong filler-filler interactions
with other particles to generate aggregates and agglomerates. The silanol concentration
on a silica surface depends on the number of silicium atoms per area at the surface and
the number of hydroxyl groups present on each silicium atom. The surface silanol-group
content of Zeosil 1165MP, a typical easy-dispersion silica type used for silica-filled
passenger tires, was reported to be 4.90 OH·nm-2[12] Three types of surface hydroxyl
groups; isolated, vicinal (on adjacent silicium atoms) and geminal (two hydroxyl groups on
the same silicium atom),[11,13] have been identified, as shown in Figure 2.3.

Figure 2.3 Typical silanol groups on a silica surface.[8]

12
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

The surface free energy of fillers, 𝛾𝑠 , can be split in two components which are
𝑠𝑝
the dispersive component, 𝛾𝑠𝑑 , and the specific component, 𝛾𝑠 , as follows;[14,15]

𝑠𝑝
𝛾 = 𝛾𝑠𝑑 + 𝛾𝑠 (2.1)

The dispersive component indicates the tendency of adhesion to organic

molecules such as polymers, and the specific component indicates the tendency of
interaction with itself and polar components. When compared to carbon black, silica has a
low dispersive component but a high specific component of free energy. This results in a
poor interaction between silica and hydrocarbon rubbers but a high degree of silica
agglomeration.
The surface energy of silica has been characterized by techniques, such as the
inverse gas chromatography (IGC) technique.[14,15] In the IGC-experiment the filler is used
as the stationary phase, and the solute probe is injected. Based on the adsorption of
several solutes on the silica surface and filler-probe interactions, the surface energy of
silica and its estimated interactions with rubbers are reported. The specific interactions of
the probes, analogs of elastomers, with silica decrease in the order: nitriles > aromatic
hydrocarbon > olefins > n-alkanes > isoalkanes. Thus, the interactions between
elastomers with silicas, which indicates their compatibility, can be ranked in order as;[14]

NBR > SBR > NR ≥ BR > HV-BR > EPM > IIR

where NBR is acrylonitrile butadiene rubber, SBR is styrene butadiene rubber, NR is

natural rubber, BR is polybutadiene rubber, HV-BR is high-vinyl polybutadiene rubber,
EPM is ethylene propylene rubber and IIR is isobutylene isoprene rubber or commonly
known as butyl-rubber.

2.2.4 Carbon-silica dual phase fillers

Lately, filler technology for tire tread compounds has been focused on new filler
systems and silane coupling agents or functionalized fillers to improve tire rolling
resistance, abrasion resistance and wet traction. A new filler system, so called carbon-
silica dual phase filler (CSDPF) has been presented in the market under the trade name of
ECOBLACKTM by Cabot Corporation. From the results of electron spectroscopy for
chemical analysis (ESCA) and infrared spectroscopy (IR), CSDPF is in the form of
individual composite aggregates of carbon and silica phases. The silica and carbon
phases which differ from simple mixes of carbon black and silica aggregates, are finely
dispersed within primary particles to make up the dual phase aggregates. Silica particle

13
Chapter 2

sizes in the range of 0.4 to 2 nm were reported after treating the dual phase filler with
hydrofluoric acid.[16,17] It has been reported that CSDPF-filled rubber exhibits a lower
Payne effect when compared to carbon black and silica/silane, which is related to filler
[18]
networking in the matrix and a higher dispersive component of the surface energies of
the CSDPF leading to a higher polymer-filler interaction in low polar or non-polar
rubbers.[19] The decrease of elastic modulus (G′) with strain amplitude, or Payne effect[20]
of NR compounds filled with CSDPF in comparison with other fillers is shown in Figure
2.4. This filler type has higher active carbon and hydrogen groups, and is more acidic than
pure carbon black leading to significantly more reactive functional groups available to
react with a silane coupling agent.[21] CSDPF-filled rubber tire tread compounds with
TESPT silane coupling agent are claimed to show superior performance over conventional
fillers, i.e. carbon black and silica, due to higher rubber-filler interaction and less filler-filler
interaction. Wang et al.[22] demonstrated that the use of CSDPF with TESPT shows better
abrasion resistance and more than 40% reduction in tan δ at 70°C which correlates with
rolling resistance of tire treads made thereof when compared to carbon black-filled
compounds. In addition, the CSDPF-filled compounds show superior abrasion resistance
and tear strength over silica-filled compounds, and better wet skid resistance of tire treads
than conventional fillers.

Figure 2.4 Elastic modulus as a function of strain amplitude of truck tire tread compounds
with various fillers.[22]

14
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

2.3 Filler reinforcement of rubbers

2.3.1 Hydrodynamic effect

The viscosity of liquids or modulus of elastomeric matrices is increased when

rigid particles are added. At low filler loadings, the modulus of filled rubber increases
linearly with the volume fraction of filler (ϕ). This can be explained by the first equation
proposed by Einstein.[23]

𝐸 = 𝐸0 (1 + 2.5𝜙) (2.2)

Where 𝐸0 is the Young‟s modulus of unfilled rubber, and 𝐸 is the Young‟s modulus of filled
rubber.

At high filler loading, the linear relation with ϕ does no longer hold. The Guth-
Gold equation is then used to explain the modulus of filled rubber by adding a second
order term ϕ2 which accounts for the interactions between particles in a denser state. The
equation reads as follows;

𝐸 = 𝐸0 (1 + 2.5𝜙 + 14.1𝜙 2 ) (2.3)

Smallwood demonstrated the equivalence of the concentration dependence of

viscosity and modulus for filled rubber. Therefore, this equation is also often called the
[24]
Einstein-Guth-Smallwood model.

2.3.2 Payne effect

The addition of filler to rubber compounds has a strong impact on the static and
dynamic behavior of such rubbers. Figure 2.6 shows the typical elastic modulus of filled
rubber-samples versus dynamic strain. Besides the strain-independent contributions of the
hydrodynamic effect, filler-to-rubber interaction and crosslinks or the network in the rubber
matrix, the elastic modulus shows a strong strain-dependence at low strains. This stress
softening at low deformation is known as the Payne effect.[20]
Therefore, the following effects contribute to the storage modulus;[6]
a) Rubber network: the network depends on the nature of the rubber and
crosslink density in the rubber matrix, involving either physical or chemical
linkages or both.

15
Chapter 2

b) Hydrodynamic effect: this effect results from the fact that filler is a rigid
phase, which cannot be deformed. It can be described by the Guth, Gold
and Smallwood equation.[24]
c) Filler-rubber interaction: or „in-rubber structure‟ depends on a combination of
the structure of the filler in the in-rubber state and the extent of filler-rubber
interactions which can be attributed to physical forces as well as to chemical
linkages or a combination of both. In the case of a silica-silane system, this
effect arises from chemical linkages between rubber and silica by silane
bridging.
d) Filler-filler interaction: the stress softening at low strain is attributed to a
breakdown of filler-filler bonds.[25] This behavior is called the Payne effect,
which plays an important role for understanding the reinforcement
mechanism in filled rubber. The strong filler-filler interaction in silica-filled
rubber is mainly caused by hydrogen bonds between silica particles. The
filler network can be easily destroyed at low strain or low deformation,
leading to a decrease of elastic modulus of the filled rubber.

Figure 2.5 Payne effect of carbon black and silica-reinforced rubber.[24]

It is clearly seen in Figure 2.5 that silica-filled rubber has a weaker filler-rubber
interaction but much stronger filler-filler interaction when compared with carbon black-filled
compounds for reason of their different surface free energies.

16
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

2.3.3 Dynamic properties in relation to tire performance

The viscoelastic behavior of elastomers can be monitored by applying a

sinusoidal stress of frequency ω, as shown in Figure 2.6. When such sinusoidal shear
stress is applied to the viscoelastic material, the strain will show a sinusoidal response but
out of phase.

Figure 2.6 Sinusoidal response of viscoelastic materials.[26]

The correlation between stress, σ, and strain, γ, can be written as;

   0 sin t (2.4)

and    0 sin(t   ) (2.5)

where 𝑡 is time, 𝛿 is the phase angle between stress and strain, 𝛾0 and 𝜎0 are the
maximum amplitudes of strain and stress, respectively. Then, the stress can be
decomposed into two components, i.e. one in-phase and one out-of-phase with strain;

   0 sin t cos    0 cos t sin  (2.6)

The dynamic stress-strain behavior of the elastic material is then expressed in

the storage modulus, 𝐺ʹ, in phase with strain, and the loss modulus, 𝐺ʺ, out of phase;

   0G  sin t   0G cos t (2.7)

with G  ( 0 /  0 ) cos  (2.8)

17
Chapter 2

and G   ( 0 /  0 ) sin  (2.9)

Thus tan   G / G (2.10)

Alternatively, in a cyclic strain test the shear modulus can be expressed in the
complex modulus 𝐺 as;
*
G  ( 0  )  G   iG  (2.11)
0

The addition of fillers into polymers causes a change of the dynamic mechanical
properties of the materials. The energy loss of rubbery materials during dynamic strain
correlates with heat generation and fatigue life, which have an influence on tire rolling
resistance, wet traction and abrasion resistance. The rolling resistance is related to the
response of rubber at a frequency of around 10 Hz and a temperature range of 50 to
80°C. In case of wet grip or wet traction, it is the stress which is generated by the skid-
resistance between the surface of the tire tread and the road surface. The frequency of
this movement depends on the roughness of the road surface. For good wet grip, the
energy loss should be high around 104 to 107 Hz in the same temperature range of 50 to
80°C. By virtue of the time-temperature superposition principle the viscoelastic behavior at
104-107 Hz and 50-80oC, is equivalent to approximately 10 Hz at -20-+20oC. The loss
tangent as a function of temperature at a frequency of 1-10 Hz can therefore be used to
imply the viscoelastic properties of tire compounds as shown in Figure 2.7.[18,26]

Figure 2.7 Tan δ versus temperature at 1-10 Hz related to different dynamic properties of
tire performances.[26]

18
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

From the viscoelastic properties, the ideal tire material should have a low tan δ at
50-80°C in order to reduce rolling resistance and save driving energy, while the tan δ
value should be high at low temperature to provide good wet skid resistance, wet grip or
wet traction and ice grip.
The addition of fillers into rubber compounds has a strong effect on the static and
dynamic properties of the rubbers. The combined effects of the different polymer network,
filler-filler interactions, the hydrodynamic effect and in-rubber structure or filler-polymer
interaction show a unique influence on the dynamic properties. The complex modulus of
filled rubber at low and high strain deformation can be related to two different types of
interactions;[6]

a) Low strain deformation (<< 5%)

The complex modulus at low strain indicates the filler-filler network. High filler
loading and increasing specific surface area, meaning decreasing primary particle size of
filler, lead to a significant increase of filler-filler interaction. This effect can be explained by
the decrease of the inter-aggregates distance of the filler while its surface area is
increased, leading to filler-filler network formation. The filler network has a major effect on
tire rolling resistance, and an increase of filler surface activity or affinity to the rubber
polymer leads to a decrease of filler networking because of better filler-rubber interaction.

b) High strain deformation (> 30%)

The complex modulus at high strain indicates the in-rubber structure or filler-
polymer interaction. In case of carbon black-filled rubber, the increase of filler structure
and surface activity enhances the in-rubber structure. The incorporation of high structure
filler increases occluded rubber in the void spaces within the filler aggregates that leads to
a high effective filler volume in the matrix. Meanwhile, the surface activity contributes to
the physical and chemical interactions between filler and rubber.

2.4 Silica-reinforced rubber

2.4.1 Silane compatibilized silica-filled compounds

The addition of silica into rubber compounds offers at least two advantages that
are reduction of heat build-up and improvement in tear strength, cut, chip and chunking
resistance, when compared to the use of carbon black. Silica itself gives a lower degree of
reinforcement when compared to carbon black of the same primary particle size due to the
different nature of the surface chemistry of the fillers. In general, silica can reinforce better

19
Chapter 2

in more polar rubbers when compared to non-polar rubbers due to the higher silica-to-
rubber interactions. The poor reinforcement efficiency of silica-filled non-polar rubbers can
be improved by using silane coupling agents. The silica-silane coupling agent
reinforcement mechanism involves two key reactions: (1) the silanization reaction in which
the silane coupling agent reacts with the silica; and (2) the formation of crosslinks between
the silane modified silica and rubber.[27,28]

2.4.1.1) Silanization reaction

The hydrophilic surface of silica is incompatible with a hydrophobic rubber. Silane
coupling agents are most widely used to be added into silica-filled compounds to improve
compatibility and increase the interactions between silica and rubber. The process
involves adsorption of the silane coupling agent onto the silica surface and subsequently a
reaction between hydroxyl groups on the silica surface with methoxy- or ethoxy-groups of
the silane coupling agent. The chemical reaction between the silica and alkoxy-silyl
groups of the silane coupling agent is the so-called “silanization”.
The silanization reaction takes place in two steps. The primary step is the
reaction of silanol-groups on the silica surface with alkoxy-groups of a silane molecule.
There are two possible reaction mechanism involved in this primary step: 1) a direct
reaction of silanol groups with the alkoxy group, and 2) hydrolysis of the alkoxy groups
followed by a condensation reaction with the silanol groups. The secondary step is a
condensation reaction between adjacent molecules of silane coupling agent on the filler
[27]
surface. Both steps of the silanization reaction are shown in Figure 2.9.

Figure 2.9 Proposed reaction mechanism of the silanization reaction.[27]

20
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

A silane coupling agent that contains sulfur atoms in its structure such as
TESPT, can act as a sulfur donor to cause premature crosslinking in the rubber compound
during mixing and later-on during the vulcanization reaction.[28] The proposed reaction of
TESPT and the silanol groups on the silica surface is shown in Figure 2.9. The silanization
reaction of a silica-filled natural rubber compound is optimal at a mixing discharge
temperature between 135-150oC.[29] This reaction can be accelerated by increasing the
mixing temperature[28,29], presence of moisture on the silica surface[27] and use of a basic
secondary accelerator like diphenylguanidine (DPG).[30]

2.4.1.2) Kinetics of the silanization reaction

The silanization reaction comprises a primary and a secondary reaction, as
shown in Figure 2.9. The reaction between ethoxy-groups on TESPT molecules and
silanol groups on the silica surface leads to the release of ethanol molecules.[27]
Primary reaction – The evolution of ethanol at 30 to 60°C is considered to
correspond to the primary reaction by assuming that 1 mole of TESPT can react with 2
moles of ethoxy groups. The primary reaction is then described by the following equation;

d [TESPT ] 1 d [ EtOH ]
  k a [TESPT ]  (2.12)
dt 2 dt

EA
and ln k a  ln A  (2.13)
RT

where [TESPT] and [EtOH] are the TESPT and ethanol concentrations in mol/kg
compound, respectively, t is time in minutes, ka is the reaction rate constant, A is the
Arrhenius factor, EA is the activation energy in kJ/mol, R is the gas constant and T is
temperature in K. From a plot of ln ka versus 1/𝑅𝑇, the activation energy of the primary
reaction is derived to be equal to 47 kJ/mol.

Secondary reaction – There are three possible reaction paths in this step.
However, the overall reaction rate constants are assumed to have similar values. Based
on the study at 120 to 160°C, the overall reactions can be described as follows.

d [ EtOH ]
 2k a [TESPT ]  kb [ Z1]  kb [ Z 2]  kb [ Z 3] (2.14)
dt

21
Chapter 2

E A'
and ln k b  ln A  (2.15)
RT

where [Z1], [Z2] and [Z3] are intermediate products concentrations in mol/kg, 𝐸𝐴ʹ is the
activation energy of the secondary reaction in kJ/mol.

From the plot of ln 𝑘𝑏 against 1/𝑅𝑇, the activation energy of the secondary reaction was
derived to be 28 kJ/mol. [27]

2.4.2 Silane coupling agents

Silane coupling agents are commonly used in the rubber industry to enhance the
degree of reinforcement of silica. In the 1970s, silane coupling agents that contain sulfur
atoms were introduced to improve the bonding between silica and rubber during the
mixing and curing stages. TESPT was used in the “Green Tire” due to its bifunctional
character. Ethoxy-groups of the silane react with silanol groups on the silica surface and
the tetrasulfide group bonds with the rubber matrix during the curing stage.
Jesionowski and Krysztafkiewicz[31] investigated the influence of silane coupling
agents on the surface properties of precipitated silica. Three types of silane coupling
agents; 3-mercaptopropyl trimethoxysilane, 3-aminopropyltriethoxysilane and vinyltris(2-
methoxyethoxy) silane, were studied. The results showed that increasing amounts of
mercaptosilane used for the silica modification gradually decreased the tendency of silica
to form agglomerates, but above five parts by weight of this silane, an excess of silane
promoted adhesion between silica particles/aggregates and re-agglomeration again
occurred. The use of aminosilane showed different results, as the tendency of silica
agglomeration was increased due to hydrogen bonds between neighboring modified silica
particles. Meanwhile, the use of vinylsilane to modify the silica surface could break down
the agglomerate structure because vinyl groups are electrostatically inert and do not
interact with other particles.
However, two disadvantages of mercaptosilane coupling agents were observed.
These included an unpleasant odor during mixing especially at elevated temperatures,
and reduction of the scorch time. Thus, this silane is not suitable to be used for products
with long processing procedures.[8] Bis(triethoxysilylpropyl)tetrasulfide (TESPT), the
bifunctional silane, shows better scorch safety than 3-mercaptopropyl-trimethoxy silane.
For a vulcanization temperature at 120oC and below, the increase in scorch time with an
increase of TESPT content may be attributed to the longer chain of TESPT that consists
of triethoxy groups on both ends of the structure. These bulky groups cause steric

22
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

hindrance for the reaction. However, as the vulcanization temperature increases, the
scorch time decreases with increasing TESPT concentration due to the fact that the steric
hindrance of TESPT becomes less significant,[32] and S-S bonds break down giving active
sulfur for the crosslinking reaction. TESPT has a bifunctional structure with ethoxy-groups
for reacting with hydroxyl groups on the silica surface during mixing, and tetrasulfide for
crosslinking with rubber molecules during vulcanization. Therefore, TESPT also acts as a
sulfur donor in which the sulfur-sulfur bonds in the tetrasulfide structure can easily be
broken down under high shear and mixing temperature. The reactive sulfur then reacts
with rubber molecules, leading to a decrease of scorch time and increase of compound
viscosity. The lower sulfur rank in a sulfur-containing silane such as Bis-
(triethoxysilylpropyl)disulfide (TESPD) provides better scorch safety compared to TESPT,
but shows lower tensile modulus of the vulcanizates when compared to the ones with
TESPT.[33]
In the case of silica-filled Nitrile-butadiene rubber (NBR), the effect of curing
systems and silane coupling agent types at a fixed content of 2 phr were studied. It was
found that the addition of 3-thiocyanatopropyl triethoxysilane in a silica-filled compound
could reduce the filler-filler interaction to a greater extent than TESPT, due to its less bulky
structure and lower viscosity leading to improved compound processability and enhanced
mechanical properties of the rubber vulcanizates. For conventional vulcanization (CV) and
semi-efficient vulcanization (semi-EV), 3-thiocyanatopropyl triethoxysilane gave a higher
degree of silica reinforcement than TESPT, attributed to the effect of better filler dispersion
and stronger filler-rubber interaction. However, for efficient vulcanization (EV), TESPT
yielded a greater degree of silica reinforcement than 3-thiocyanatopropyl triethoxysilane
due to the sulfur contribution effect when a large amount of accelerators was used.[34] The
optimum loading levels of TESPT and 3-thiocyanatopropyl triethoxysilane in the EV
system were at 3.0 and 1.5 phr, respectively. The silica-filled compounds with 3-
thiocyanatopropyl triethoxysilane showed better compound processability than the silica-
filled compounds with TESPT, but TEPST exhibited greater cure characteristics due to its
ability to contribute sulfur during the vulcanization process. TESPT-containing
vulcanizates showed better dynamic properties than 3-thiocyanatopropyl triethoxysilane,
but poorer aging resistance.[35]
Due to some drawbacks of the 3-mercaptosilane and TESPT coupling agents as
mentioned above, a new silane type was invented. A mercaptosilane blocked with an
isocyanate-functional group in solid form was introduced to be used as a silane coupling
agent for tires. The use of this isocyanate-blocked mercaptosilane in a silica-filled
compound could avoid sulfur smell during mixing, but no improvement in processability

23
Chapter 2

was observed. Moreover, there was some toxicity released from the isocyanate during
rubber processing.[36]
In recent years, mercaptosilane-derivatives with blocking groups such as a
carboxylate ester or carboxylic acid functional groups are proposed. The use of these
blocked mercaptosilanes in silica-filled rubber compounds needs a deblocking agent to
obtain the optimum properties of the compound.[37,38] The deblocking agent is a material
that is capable to unblock the mercaptosilane to enable the mercapto-group to react with
rubber molecules. The suitability of the deblocking agent depends on the blocking group
which was used to block the reactivity of the mercaptosilane. The quantities of deblocking
agent used are in the range of 0.1 to 5.0 phr, preferably in the range of 0.5 to 3.0 phr.
Otherwise, the deblocking agent should be a nucleophilic molecule containing a hydrogen
atom which can easily transfer to the blocking group in the blocked mercaptosilane, such
as primary, secondary and tertiary amines, Lewis acids, or thiols.[39,40] Other examples of
deblocking agents are accelerators that are commonly used in sulfur cure systems such
as thiuram, thiazole and sulphenamide.[41] The addition of blocked mercaptosilane
improves the properties of silica-filled rubber compounds, i.e. increased reinforcing
efficiency, lowered compound viscosity and enhanced processability. The new silanes are
designed to be used for several rubber products, e.g. tire treads, transmission belts,
conveyor belts, roller coating, damping elements, etc.[42]
3-Octanoylthio-1-propyltriethoxysilane or NXT-silane is an octanoyl-group
blocked mercaptosilane which provides a lower reactivity of the silane during processing.
Silica-filled rubber compounds with NXT-silane should be prepared at high mixing
temperature above 130°C to promote the reactions between the silane and silica. This
blocked silane shows good scorch safety during mixing, even if the silica-NXT-filled
compound is mixed for 10 min at 150°C to ensure a complete reaction between silica and
silane.[43] The hardness, modulus, tensile and tear strength are increased when the NXT-
silane is present in the silica-filled compound. Furthermore, the wet traction and rolling
resistance are improved by incorporating of the NXT-silane due to an increase of crosslink
density in the compound.[44] The scorch time of a silica-filled compound with NXT is longer
than that of one with TESPT. The activation energies of the vulcanization reaction
decrease with increasing NXT and TESPT concentrations in the silica-filled compounds,
but the values of the compounds with NXT are lower than those of the ones with TESPT.
This is due to steric hindrance of TESPT bulky end groups that retard the curing
process.[45] The silica particles modified with NXT-silane have a lower capacity to form
aggregates and agglomerates. However, the only one sulfur atom in the NXT molecule is
not able to form as strong an interaction and bonding to rubber molecules like the case of

24
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

the tetrasulfide in TESPT, leading to lower mechanical properties of the silica/NXT-filled

compound.[46]

2.4.3 Flocculation of silica

The re-agglomeration of filler aggregates or the reformation of a filler network, so

called filler-flocculation, has an influence on the final properties of the filled-compounds.
When the polarity of rubber is increased, polar fillers exhibit a decrease of flocculation
tendency, whereas non-polar fillers tend to provide more flocculation. For example, it has
been reported that EPDM filled with methylated fumed silica showed a lower flocculation
tendency when compared to EPDM filled with unmodified silica.[47] The flocculation
process of the silica-silica network in the silica-filled compounds strongly increased during
heat treatment. This phenomenon could also be observed after silanization but to a lesser
degree due to the shielding of the silica surface. [26]

The degree of silica distribution or dispersion, x, can be expressed by the ratio of

the Payne effect at time, t and at infinite time;[30]

s ' (t )  s ' (1)

x (2.16)
s ' ()  s ' (1)

where sʹ(1), sʹ(t) and sʹ(∞) are the in-phase part of the dynamic torques measured in a
Rubber Process Analyzer (RPA 2000, Alpha Technologies) after preheating time for 1
minute and after heating times for t and 12 minutes, respectively. Herein, heating time at
12 minutes is taken as infinite time for practical purposes. Thus, the silica flocculation can
be described by a simple kinetic parameter through the reaction rate constant, 𝑘𝑎 and the
activation energy, 𝐸𝑎 , as follows;

ln(1  x1 )  ln(1  x2 ) 1
ka  [min ] (2.17)
t 2  t1

E
ln k a   a [ kJ / mol ] (2.18)
RT

where t is the heating time, x1 and x2 are the degrees of silica dispersion at times t1

and t2 , respectively, T is the absolute temperature and R is the gas constant. The

25
Chapter 2

activation energy of flocculation can be obtained from the slope of the straight line of a plot
of 𝑘𝑎 on a logarithmic scale against the inverse temperature.

(a)
G′ (MPa)

Annealing Time (minutes)

(b)
G′ (MPa)

Annealing Time (minutes)

Figure 2.8 Shear storage modulus (G′) at 1% cyclic strain of filled-compounds: (a)
comparison of silica and carbon black-filled compounds, and (b) silica-filled compounds
with various silane loadings.[48]

Böhm et al.[48] studied the flocculation process in carbon black and silica-filled
compounds after annealing at 125-170°C. As shown in Figure 2.8, the storage modulus
(G′) of the compounds increased due to filler flocculation, and silica shows a much
stronger filler flocculation than carbon black due to strong hydrogen bonding with other
particles/aggregates. However, silica flocculation can be decreased by using a silane
coupling agent, as observed in Figure 2.8(b). Lin et al.[49] reported that the degree of silica
flocculation depends on both type and concentration of the silane used in the compounds.

26
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

Alkyl mono-(triethoxy silane) such as n-octyltriethoxysilane (OTES) and 3-mercaptopropyl

triethoxysilane (MPS) could reduce silica flocculation during thermal annealing more
effectively, when compared to bis-(trialkoxysilanes) such as bis-(triethoxysilylpropyl)
tetrasulfide (TESPT), bis-(triethoxysilylpropyl)disulfide TESPD and bis-
(triethoxysilyl)octane (TESO). On the other side, silica dispersing agents such as sorbitan
monooleate (SMO) and N,N-dimethyl-1-octadecylamine (DMOD) were not effective to
prevent flocculation in the silica-filled compounds.

2.4.4 Other approaches towards better dispersion of silica in a rubber matrix

2.4.4.1) In situ silica

Recently, in situ silica synthesized by a sol-gel method has been widely studied
as an alternative method to improve silica dispersion in a rubber matrix, in addition to a
more common dry mixing or compounding method. This process consists of hydrolysis
and condensation reactions of an alkoxysilane such as tetraethoxysilane (TEOS). The in
situ silica generated is expected to be well-dispersed within the polymer matrix. The
structure and morphology of the filler depend on the reaction conditions and nature of the
catalyst. The in situ silica synthesis can be carried out in two different ways for the swollen
state of a rubber matrix; either 1) the previously cross-linked polymer is swollen in TEOS
and then hydrolyzed in situ, or 2) silica particles are precipitated in the polymer matrix
before the cross-linking process.[50] In addition to the swollen state, the in situ technique
can be carried out in solution[51] and in the latex states.[52,53]
Various factors affect the silica content and silica particle sizes obtained from the
sol-gel method, such as the silica precursor and catalyst. Tetraethoxysilane (TEOS) has
been used as silica precursor in the in situ silica reaction. The proportion of TEOS and
water for the hydrolysis reaction was found to have an influence on the silica content
obtained in the system. Increasing the TEOS/water molar ratio resulted in increasing of
the silica content in the matrix.[54] Other types of alkoxysilanes have been also applied to
produce in situ silica such as vinylethoxysilane (VTOS), ethyltriethoxysilane (ETOS), and
i-butyltriethoxysilane (BTOS). The results have shown that TEOS and VTOS are the most
promising silica precursors. It was demonstrated that in situ generated silica can enhance
the tensile modulus and tear resistance of rubber vulcanizates.[55]
Ikeda et al.[56] investigated the effect of hydrochloric acid and n-butylamine as
catalysts for the in situ silica process in a SBR matrix. The results show that the silica
formed by the reaction with n-butylamine was homogeneously dispersed in the SBR
matrix and the diameter of the particles was ca. 25-30 nm. The primary alkylamines with

27
Chapter 2

long hydrocarbon segments were estimated to form reverse micelles like surfactant in the
TEOS-swollen NR matrix[57] as shown in Figure 2.10. Other primary alkylamines with
different hydrocarbon segments, i.e., n-butylamine, n-hexylamine, n-octylamine,
dipropylamine, and triethylamine were also investigated as catalysts for preparation of the
in situ silica. The n-hexylamine catalyst showed a higher in situ silica content and good
silica dispersion than the other catalysts. In addition, the amount of in situ silica increased
with increasing n-hexylamine content. This is due to the fact that n-hexylamine contains a
long chain alkyl group and its solubility in water is higher than that of n-octylamine. Thus, it
can easily penetrate into the NR matrix and account for a more homogeneous reaction
system.[58] For the latex state reaction, an ammonia solution was used as catalyst.[50]

Figure 2.10 Speculated formation of in situ silica in a TEOS-swollen NR-matrix by primary

amine with a long hydrocarbon segment.[58]

2.4.4.2) Self-assembly process

A novel self-assembly nanocomposite was developed to prepare a bulk poly(vinyl
alcohol)/silica nanocomposite.[59] It was found that the chemical and physical properties of
such nanocomposite were significantly increased, due to the uniform distribution of silica
particles and strong interaction between silica and polymer matrix. This novel process has
also been applied to prepare NR/silica nanocomposites by combining the latex
compounding and self-assembly technique. First, negatively charged silica nanoparticles
act as templates to adsorb positively charged poly(diallyldimethylammonium chloride)

28
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

(PDDA) molecules through electrostatic adsorption. Negatively charged natural rubber

latex (NRL) particles are then assembled onto the surface of the silica/PDDA
nanoparticles. Finally, the silica nanoparticles are uniformly distributed in the NR matrix,
as shown in Figure 2.11. The key procedure of this process is the encapsulation of the
silica nanoparicles with PDDA molecules and NRL matrix. The aim of the self-assembly
process is to suppress the silica self-agglomeration reaction caused by strong silica-silica
interaction.[60] When the nanosilica content is less than 4.0 wt%, the silica nanoparticles
are assembled within NRL matrix as core-shell structure and the average size of the
nanosilica clusters range between 65-80 nm. It has been reported that the mechanical
properties of the nanocomposites were markedly improved with increasing nanosilica
loadings from 2.5 to 4.0 wt%. Thermal and thermo-oxidative decomposition temperatures
of the nanocomposites were also increased upon increasing silica loadings.[61]

Figure 2.11 Scheme of the self-assembly process.[61]

2.5 Compatibilizers for silica-reinforced rubbers

Silica shows a high polarity and hydrophilic surface due to the silanol groups on
its surface. Thus, it is less compatible with non-polar rubbers such as natural rubber (NR),
styrene-butadiene rubber (SBR) and butadiene rubber (BR), and more compatible with
polar rubbers such as chloroprene rubber (CR) and acrylonitrile butadiene rubber (NBR).

29
Chapter 2

The polar functional groups on the silica surface form hydrogen bonds with polar
components in the rubber compound such as N-tert-butyl-2-benzothiazole sulfenamide
(TBBS) and N-cyclohexyl benzothiazole sulfenamide (CBS) which are generally used as
vulcanization accelerators for rubber compounds. These accelerators have alkaline
functional groups in their structures, so they can be easily adsorbed on the acidic surface
of silica. Even though silane coupling agents have been widely used to improve silica
dispersion and to prevent adsorption of polar chemicals on the silica surface, the influence
of polar rubber as an alternative to silane for the silica-filled compounds has also been
studied.
Epoxidized natural rubber (ENR) shows a higher polarity than NR due to the
epoxide groups in its structure. The chemical and physical properties of ENR change
according to the epoxide group contents. The increase of epoxide contents decreases
double bonds in the NR molecule leading to an increase of heat resistance and swelling
resistance in non-polar solvents. The viscosity is also increased when the degree of
epoxidation is increased.[62] From the effect of polar functional groups, ENR interacts with
hydroxyl groups on the silica surfaces. It has been reported that the mechanical properties
of silica-filled ENR without silane coupling agent were higher than those of silica-filled NR
without silane coupling agent. This was due to the improved interaction between the ENR
and silica surface via hydrogen bonds.[63] Reinforcing efficiency of silica-filled ENR was
significantly improved by the good combination of ENR with the range of epoxide groups
of 20-30 mol% and TESPT in the range of 2-4 wt% relative to silica content.[64] The silica-
filled ENR in a presence of N-3-(N-vinyl-benzyl-amino)ethyl-γ-amino-propyl-
trimethoxysilane-monochloride as a silane coupling agent led to a reaction between the -
NH2+- groups of the silane and epoxide groups of ENR resulting in a covalent C-N bond
between the silane and ENR, as shown in Figure 2.12. The extent of interaction between
silica and ENR increased with increasing the volume fraction of filler and silane loading.[65]
Furthermore, the addition of a multifunctional additive (MFA) which is a diamine salt of a
fatty acid ([RNH2+(CH2)3NH3+][R′COO-]2) could improve the mechanical properties of the
silica-filled ENR. Scorch and cure times were decreased with increasing loading level of
MFA.[66]

30
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

(a) (b)

Figure 2.12 Proposed mechanism of the reaction between ENR and silica without silane
(a) and in presence of N-3-(N-vinyl-benzyl-amino)ethyl-γ-amino-propyl-trimethoxysilane-
monochloride (b).[66]

George et al.[67] investigated the addition of ENR as a compatibilizer in silica-filled

NBR vulcanizates. Incorporation of ENR in silica-filled NBR showed improved properties
of the vulcanizates. It was indicated that NBR and ENR can form self-crosslinked

31
Chapter 2

structures and the epoxy groups of ENR could react with hydroxyl groups on the silica
surface, as shown in Figure 2.13, leading to an increase of bound rubber and mechanical
properties. The properties of silica-filled NBR with ENR as a compatibilizer were higher
than those of silica-filled NBR without ENR, and similar to those of silica-filled NBR with
TESPT. The overall property enhancement indicated that ENR could be used as a
reinforcement modifier in silica-filled NBR.

Figure 2.13 Proposed mechanism of reaction between ENR and silica.[67]

The effect of ENRs with 25 mol% (ENR-25) and 50 mol% epoxidation (ENR-50)
as compatibilizers in NR/organoclay nanocomposites was studied.[68] The results showed
that dispersion of organoclay in the matrix, tensile strength and tear strength of ENR-50
containing nanocomposites were superior to those of ENR-25 compatibilized and
uncompatibilized nanocomposites, respectively. Due to the ENR polar character, a higher
interaction with organoclay was observed, leading to the improvement of dispersion of the
organoclay in the rubber matrix.[69] ENR/organoclay nanocomposites were used to
improve the properties of carbon black filled-NR/high styrene rubber blends.[70] The blends
containing the nanocomposites at 10 phr showed high crosslink density, maximum cure
torque and cure rate index. The overall mechanical and thermal properties were improved
by adding the ENR/organoclay in the blends. The interaction between ENR and clay that
was formed during high-temperature (180°C) molding could be observed by the infrared
spectroscopic technique.[71] A decrease of peak intensities at 1568, 1348 and 3696 cm-1,
which were assigned to O-H bending of H-O-H and Si-OH, and O-H stretching of Si-OH,
respectively, as well as a decrease of the peak intensities at 870 and 840 cm-1 assigned to
characteristic peaks of the epoxy group, indicated strong interaction between ENR and
clay. The possible reaction is shown in Figure 2.14.

32
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

Figure 2.14 Proposed mechanism of bonding between ENR and clay.[71]

Maleic anhydride is another polar substance which can potentially react with
hydroxyl groups on the silica surface so that silica agglomerates can be deminished or
agglomerate size is reduced. Bikiaris and coworkers[72] studied the effect of
poly(propylene-g-maleic anhydride) copolymer (PP-g-MA) as a compatibilizer in isotactic
polypropylene (iPP)/ silica nanocomposites. PP-g-MA containing 0.6 wt% of maleic
anhydride content was added to the nanocomposites at various concentrations, i.e. 1, 2.5
and 5 wt% relative to the silica content. The tensile strength, elongation at break and
modulus were increased with increasing nanosilica content as well as the copolymer
content due to the enhancement of interfacial adhesion between PP and silica.
Maleated natural rubber (MNR) is a modified natural rubber prepared by grafting
of maleic anhydride onto natural rubber molecules so that the MNR shows higher polarity
than NR and better compatibility with silica. The interaction between MNR and silica is
shown in Figure 2.15. The silica-silica interaction of silica-filled MNR was decreased as
the succinic anhydride groups of the MNR could interact with the hydroxyl groups on the
silica surface. The silica-filled MNR compound which was obtained from modified NR with
6 wt% of maleic anhydride relative to the NR showed the lowest silica-silica interaction
and optimum mechanical properties. The addition of a silane coupling agent in the
compound did further increase silica-rubber interaction and decrease the silica-silica
bonding, leading to improved mechanical and dynamic properties.[73] A study based on
sludge paper which has hydroxyl groups on its surface similar to that of silica surfaces
also found that the addition of MNR at 5 phr as a compatibilizer into a sludge paper-filled
NR composite improved the mechanical properties and enhanced adhesion between

33
Chapter 2

rubber and sludge.[74] In addition, fatigue life of the composites was improved, while filler
agglomerates were decreased.[75]

Figure 2.15 Postulated interaction between the maleic anhydride groups of MNR and the
hydroxyl groups on the silica surface.[73]

Acrylonitrile butadiene rubber (NBR) is a polar rubber containing basic nitrile

groups (-CN) which can form hydrogen bonds with silanol groups on the silica surfaces.
Therefore, bound rubber between silica and NBR can be easily generated through
hydrogen bonding in the silica-filled NBR.[76]

Figure 2.16 Propose of bound rubber between NBR and silica surface.[76]

NBR has been claimed to act as a compatibilizer in silica-filled non-polar rubbers.

The addition of NBR into a silica-filled styrene-butadiene rubber (SBR) compound
improved the silica dispersion and physical properties of the vulcanizates whereas the
scorch and cure times of the compounds decreased.[77] The properties of the vulcanizates
increased with increasing the acrylonitrile content of NBR. The optimum NBR content was
at 5 phr where the compound showed low viscosity and the vulcanizates had higher
elongation at break, tensile strength and modulus. The addition of the silane coupling
agent, Bis-(3-triethoxysilyl-propyl)tetrasulfide (TESPT) together with NBR into the silica-

34
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

filled SBR compound further improved compatability between the silica and rubber
phases.[78] The effect of NBR on the properties of silica-filled NR was studied in the
presence of the silane coupling agent, 3-octanoylthio-1-propyltriethoxysilane (NXT). The
mechanical properties of the vulcanizates increased with increasing NBR and silane
coupling agent contents. However, the addition of NBR more than 5 phr led to a
continuous decrease of the mechanical properties of the vulcanizates.[79]
Polychloroprene rubber (CR) is widely used to produce rubber products with
requirements for oxidation, oil and heat resistance. CR is a homopolymer consisting of
chloroprene units in which chlorine atoms induce its high polarity.[80] CR could be self-
crosslinked within a silica compound through reaction between allylic chlorine atoms and
silanol groups.[81] Similar to the silica-filled non-polar rubbers, properties of silica-filled CR
can be improved by the addition of a silane coupling agent. The use of Bis-(3-
triethoxysilylpropyl)tetrasulfide (TESPT) showed better property improvement of filled-CR,
compared to 3-thiocyanatopropyl triethoxy silane (Si-264).[82] Metal oxide (ZnO)
vulcanized silica-filled CR in presence of ethylene thiourea (ETU) showed an increase of
crosslink density with increasing ETU loading levels, due to the nitrogen atom in ETU
which could interact with silanol groups on the silica surface and form hydrogen bonds.
The results also showed an increase of crosslink density with increasing silica loading
levels because of a chemical reaction between allylic chlorine atoms and silanol groups.[83]

-HCl

heat

Figure 2.17 Probable reaction between the silanol groups and allylic chlorine atoms of
CR.[83]

The addition of CR into a silica-filled NR compound improved silica dispersion

and increased bound rubber as well as physical properties such as tensile strength,
elongation at break, modulus, abrasion and crack resistance.[84] Increase of the CR
content increased the properties of the filled-rubber, which could also be described by the
fact that the electronegative chlorine atoms of CR might interact with hydroxyl groups on
[85]
the silica surface to form hydrogen bonds, as shown in Figure 2.18, instead of a full
[83]
chemical reaction.

35
Chapter 2

Figure 2.18 Interaction between chloroprene rubber and silica to form hydrogen bonds.[85]

Bertora et al.[86] investigated the use of 3-mercaptopropyltrimethoxysilane (MPS)-

grafted polybutadiene oligomeric chains as compatibilizer in silica-filled SBR. The
determination of the thermodynamic surface properties of the MPS-grafted polybutadiene
modified silica showed a decrease of hydrophilic character of the silica. However, the
silica-filled compound with the MPS-grafted polybutadiene gave lower mechanical
properties than the silica-filled compound with TESPT.

2.6 Silane modified polymers

Silane-grafted polymers have been developed to improve the properties of some

materials, for example, thermoplastic polyolefins, i.e. polyethylene (PE) and polypropylene
(PP) and elastomers, e.g. ethylene-propylene rubber (EPM) and ethylene-propylene-diene
terpolymer (EPDM). The silane-grafted polymers can be crosslinked in presence of
moisture. Blends of polyolefins and silane-grafted polymers show advantages including
improved extrudability of cable or tube jacket products, and high melting viscosity that
allow accurate adjustment to specific needs. Also, strength, flexibility and electrical
properties can be controlled through the level and proportion of the grafted polyolefins.[87]
The silane-grafted polyolefins are normally prepared in the melt in a twin-screw
extruder[88,89] or internal mixer.[90] The grafting efficiency of a vinyl-silane onto polymer
molecules can be increased by using dicumyl peroxide (DCP) as initiator.[88]
Vinyloxyaminosilane grafted ethylene-propylene-diene terpolymer (EPDM-g-
VOS) had been synthesized in toluene using DCP as initiator.[91] The grafting efficiency
increased with increasing EPDM content and reaction temperature when the grafting
reaction was studied at 150-180oC. The optimum VOS and DCP concentrations were 2
and 0.2 mole%, respectively. By using the conditions at which the optimum grafting
efficiency of VOS onto EPDM was obtained in the solution technique, the EPDM-g-VOS
was also prepared in the melt at 160oC and rotor speed of 30 rpm for 6 min in an internal

36
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

mixer. The EPDM was mixed with 0.2 wt% of DCP initiator dissolved in 2 wt% of VOS and
0.1 wt% of dibutyltindilaurate catalyst. The VOS-grafted EPDM showed low mechanical
properties but its dielectric characteristics were improved with increasing VOS
concentrations. The EPDM-g-VOS was later used to blend with linear low density
polyethylene (LLDPE). Both phases were compatible at all blend compositions. The
values of storage modulus (E′), loss modulus (E′′) and mechanical properties were
increased with increase of LLDPE in the blends.[92]
Grafting of vinyltriethoxysilane (VTES) onto ethylene-propylene-diene terpolymer
(EPDM) by using DCP as initiator was investigated by Alagar et al..[93] The EPDM-g-VTES
was prepared by melt mixing in a twin-screw extruder. The effects of various parameters,
i.e. EPDM and VTES contents, reaction time, reaction temperature and initiator
concentration, on the grafting efficiency of VTES onto EPDM were investigated. The
grafting efficiency linearly increased with increasing EPDM contents due to the fact that
more active sites were present. The highest grafting efficiency of 18.7% was observed
when the reaction was carried out using VTES at 5 wt%. The grafting efficiency decreased
with further increase of VTES concentration, due to the formation of more active centers of
VTES which tended to result in homopolymerization rather than the grafting reaction.
Optimized reaction temperature and reaction time were at 180°C and 5 min, respectively.
The grafting reaction of EPDM with VTES is shown in Figure 2.19.

DCP
CH2 CH Si (OC2H5)3
n n
Vinyltriethoxysilane (VTES)

CH2 CH Si (OC2H5)3

Ethylene-propylene-diene terpolymer (EPDM)

P H

CH2 CH2 Si (OC2H5)3

Figure 2.19 Schematic representation of the VTES-grafted EPDM.[93]

The EPDM-g-VTES was later blended with LLDPE. The mechanical properties
such as tensile strength, modulus and hardness of EPDM-g-VTES/LLDPE blends were
higher than those of EPDM/LLDPE blends due to the presence of efficient interaction
between the blend components and the increase of crosslinking in EPDM-g-VTES.

37
Chapter 2

Changjie et al.[94] recently studied the graft copolymerization of VTES onto SBR
molecules in latex by using benzoic peroxide as initiator. The increase of VTES
concentration reduced the grafting efficiency. The VTES-grafted SBR (SBR-g-VTES)
vulcanizates showed an increase of tensile strength, thermal decomposition temperature
and glass transition temperature, but a decrease of cure rate and elongation at break with
increasing concentration of VTES.
3-Mercaptopropyltrimethoxysilane and 3-mercaptopropyltriethoxysilane were
grafted onto a liquid low molar mass polybutadiene (PB) in solution state in order to obtain
a product which is crosslinkable with moisture. The grafting reaction was carried out at
75°C for 3 h under nitrogen atmosphere and azobisisobutylonitrile (AIBN) was used as
initiator.[95,96] It was found that the triethoxysilane groups were attached onto the low molar
mass polybutadiene in the range of 1.05 to 7.9 functions per chain. The reaction between
3-mercaptopropyltriethoxysilane (MPTS) and double bonds of polybutadiene is shown in
Figure 2.20. The reaction occurred faster with double bonds in 1,2-polybutadiene than in
1,4-polybutadiene and the kinetics are very dependent on the chemical structure of the
thiol.

H H
EtOH
C C AIBN CH2 CH CH2 CH2
m n
HS CH2 Si EtOH
CH2 CH2 3 75oC EtOH
n
EtOH
S CH2 Si EtOH
3
Polybutadiene Mercaptopropyltriethoxysilane EtOH

Figure 2.20 Mechanism of 3-mercaptopropyltriethoxysilane modification of

[96]
polybutadiene.

2.7 Motivation of this thesis

Silica has conventionally been used to produce light-colored products and

mechanical goods, but is increasingly used in tire tread compounds to reduce rolling
resistance and thus automotive fuel consumption. However, the large number of hydroxyl-
groups on the silica surface leads to strong filler-filler interactions and poor interaction with
non-polar rubbers. Silane coupling agents have been practically used in silica-filled
compounds to improve silica dispersion and interactions between silica and rubber. The
mixing conditions must be optimized to achieve a good level of silanization and thus
properties. The mixing of a silica-silane filled-compound is more complicated than that of
carbon black-filled ones. A whole or partial replacement of silane coupling agent with
chemically functionalized NR as a compatibilizer should reduce the complications arising

38
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

from the silanization reaction. This will lead to use of more silica compounds for tires and
consequently more efficient use of energy. A new market for modified NR can also be
promoted. Silica is a natural oil-independent filler and NR is a renewable resource. The
increasing use of silica and NR will drive the rubber industry towards greener technology.
In this thesis, modified natural rubbers are used as compatibilizers in silica-filled
natural rubber compounds to achieve optimal properties of such compounds in
comparison with the use of TESPT silane coupling agent. Natural rubber is first modified
to achieve polar functional groups attached onto the natural rubber molecules, such as
epoxide functional groups and silane coupling agent moieties. The silanes under
investigation are vinyltriethoxysilane, 3-mercaptopropyltriethoxysilane and 3-octanoylthio-
1-propyltriethoxysilane (NXT). The modified natural rubbers are subsequently used as
compatibilizers in silica-reinforced NR compounds, and their optimum loadings are
investigated. The modified rubbers are also used in combination with silane coupling
agents to enhance the properties of silica-filled natural rubber, by taking the compound
with optimum TESPT silane coupling agent as reference. All of the silica-filled natural
rubber compounds are characterized for their Mooney viscosities, cure behavior, filler-filler
interaction or Payne effect, filler-rubber interaction by means of bound rubber
measurements, mechanical properties, dynamic mechanical properties and silica
dispersion.

2.11 References

1. B. Schwaiger, A. Blume, Rubber World, 4, 32 (2000).

2. R. Rauline, EP Patent 0 501227A1, to Michelin & Cie, February 9, 1992.
3. J.L. Leblanc, Prog. Polym. Sci., 27, 627 (2002).
4. M. Morton, “Rubber Technology”, Robert E. Krieger Publishing, Florida, second
edition, 1981.
5. C.M. Blow, C. Hepburn, “Rubber Technology and Manufacture”, Butterworths,
London, second edition, 1982.
6. J. Fröhlich, W. Niedermeier, H.-D. Luginsland, Composites: Part A, 36, 449 (2005).
7. R.W. Magee, Rubber Chem. Technol., 68, 590 (1995).
8. S. Wolff, Rubber Chem. Technol., 69, 325 (1996).
9. G. Kraus, “Science and Technology of Rubber”, Academic Press, London, 1978.
10. A.I. Medalia, R.R. Juengel, J.M. Collins, Developments in Rubber Technology-1,
Applied Science, London, 1979.

39
Chapter 2

11. W.W. Barbin, M.B. Rodgers, “Science and Technology of Rubber”, Academic Press,
San Diego, second edition, 1994.
12. L.T. Zhuravlev, Colloids Surf. A: Physicochem. Eng. Aspects, 173, 1 (2000).
13. M.P. Wagner, Rubber Chem. Technol., 49, 703 (1976).
14. M.J. Wang, S. Wolff, J.-B. Donnet, Rubber Chem. Technol., 64, 559 (1991).
15. M.J Wang, S. Wolff, Rubber Chem. Technol., 65, 715 (1992).
16. L.J. Murphy, M.-J. Wang, K. Mahmud, Rubber Chem. Technol., 71, 998 (1998).
17. L.J. Murphy, M.-J. Wang, K. Mahmud, Rubber Chem. Technol., 73, 25 (2000).
18. J.-H. Ma, S.-H. Zhao, L.Q. Zhang, Y.-P. Wu, Rubber Chem. Technol., 86, 664
(2013).
19. M.-J. Wang, H. Tu, L.J. Murphy, K. Mahmud, Rubber Chem. Technol., 73, 666
(2000).
20. A.R. Payne, Rubber Chem. Technol., 44, 440 (1971).
21. L.J. Murphy, E. Khmelnitskaia, M.J. Wang, K. Mahmud, Rubber Chem. Technol., 71,
1015 (1998).
22. M.-J. Wang, P. Zhang, K. Mahmud, Rubber Chem. Technol., 74, 124 (2001).
23. A. Einstein, Ann. der Physik, 19, 289 (1906).
24. J.-B. Donnet, Rubber Chem. Technol., 71, 323 (1998).
25. H.-D., Luginsland, J. Fröhlich, A. Wehmeier, Rubber Chem. Technol., 75, 563
(2002).
26. M.-J. Wang, Rubber Chem. Technol., 71, 521 (1998).
27. U. Görl, A. Hunsche, A. Mueller, H.G. Koban, Rubber Chem. Technol., 70, 608
(1997).
28. J.W.T. ten Brinke, S.C. Debnath, L.A.E.M. Reuvekamp, J.W.M. Noordermeer, Comp.
Sci. Technol., 63, 1165 (2003).
29. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 85, 277 (2012).
30. S.Mihara, R.N. Datta, J.W.M. Noordermeer, Rubber Chem. Technol., 82, 524 (2009).
31. T. Jesionowski, A. Krysztafkiewicz, Appl. Surf. Sci., 172, 18 (2001).
32. B.T. Poh, C.C. Ng, Eur. Polym. J., 34, 975 (1997).
33. K.-J. Kim, J. Vanderkooi, Kautsch. Gummi Kunstst., 55, 518 (2002).
34. P. Sae-oui, U. Thepsuwan, K. Hatthapanit, Polym. Test., 23, 397 (2004).
35. P. Sae-oui, C. Sirisinha, K. Hatthapanit, U. Thepsuwan, Polym. Test., 24, 439
(2005).
36. S. Ahmad, R.J. Schaefer, US Patent 4519430, to The B.F. Goodrich Company, May
28, 1985.

40
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

37. R.W. Cruse, R.J. Pickwell, K.J. Weller, E.R. Pohl, US Patent 6528673 B2, to
Crompton Corporation, March 4, 2003.
38. R.W. Cruse, R.J. Pickwell, K.J. Weller, E.R. Pohl, US Patent 6608125 B2, to
Crompton Corporation, August 19, 2003.
39. M.P. Cohen, US Patent 2005/0009955 A1, to The GoodYear Tire & Rubber
Company, January 13, 2005.
40. R.W. Cruse, R.J. Pickwell, K.J. Weller, E.R. Pohl, US Patent 7122590 B2, to General
Electric Company, October 17, 2006.
41. C.J. Lin, W.L. Hergenrother, US Patent 7256231 B2, to Bridgestone Corporation,
August 14, 2007.
42. H.-D. Luginsland, R. Krafczyk, F. Forster, US Patent 2003/0130388 A1, to Degussa
AG, July 10, 2003.
43. H. Yan, K. Sun, Y. Zhang, Y. Zhang, J. Appl. Polym. Sci., 94, 2295 (2004).
44. H. Yan, G. Tian, K. Sun, Y. Zhang, Y. Zhang, J. Appl. Sci. Part B: Polym. Phys., 43,
573 (2005).
45. H. Yan, K. Sun, Y. Zhang, Y. Zhang Y. Fan, J. Appl. Polym. Sci., 94, 1511 (2004).
46. J.L. Valentín, I. Mora-Barrontes, A. Rodríguez, L. Ibarra, L. Gonzalez, J. Appl.
Polym. Sci., 103, 1806 (2007).
47. K. Stockelhuber, A. Das, R. Jurk, G. Heinrich, Polymer, 50, 1954 (2010).
48. G.A. Böhm, W. Tomaszewski, W. Cole, T. Hogan, Polymer, 51, 2057 (2010).
49. C.J. Lin, W.L. Hergenrother, E. Alexanian, G.G.A. Böhm, Rubber Chem. Technol.,
75, 865 (2002).
50. L. Bokobza, J.-P. Chauvin, Polymer, 46, 4144 (2005).
51. B. Chaichua, P. Prasassarakich, S. Poompradub, J. Sol-Gel Sci. Technol., 52, 219
(2009).
52. K. Yoshikai, T. Ohsaki, M. Furukawa, J. Appl. Polym. Sci., 85, 2053 (2002).
53. V. Tangpasuthadol, A. Intasiri, D. Nuntivanich, N. Niyompanich, S. Kiatkamjornwong,
J. Appl. Polym. Sci., 109, 424 (2008).
54. S. Poompradub, B. Chaichua, C. Kanchanaamporn, T. Boosalee, P. Prasassarakich,
Kautsch. Gummi Kunstst., 4, 152 (2008).
55. J. Siramanont, V. Tangpasuthadol, A. Intasiri, N. Na-Ranong, S. Kiatkamjornwong,
Polym. Eng. Sci., 49, 1099 (2009).
56. Y. Ikeda, A. Tanaka, A. Kohjiya, J. Mater. Chem., 7, 1497 (1997).
57. F. Debuigne, L. Jeunieau, M. Wiame, J.B. Nagy, Langmuir, 16, 7605 (2000).
58. Y. Ikeda, S. Poompradub, Y. Morita, S. Kohjiya, J. Sol-Gel Sci. Technol., 45, 299
(2008).

41
Chapter 2

59. Z. Peng, L.X. Kong, S.-D. Li, J. Appl. Polym. Sci., 96, 1436 (2005).
60. Z. Peng, L.X. Kong, S.-D. Li, J. Nanosci. Nanotechnol., 6, 541 (2006).
61. Z. Peng, L.X. Kong, S.-D. Li, Y. Chen, M.F. Huang, Comp. Sci. Technol., 67, 3130
(2007).
62. I.R. Gelling, M. Porter, “Natural Rubber Science and Technology”, Oxford University
Press, Oxford, 1988.
63. F. Cataldo, Macromol. Mater. Eng., 287, 348 (2002).
64. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Adv. Mater. Res.,
844, 243 (2014).
65. A.K. Manna, A.K. Bhattacharyya, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer,
Polymer, 39, 7113 (1998).
66. H. Ismail, H.H. Chia, Eur. Polym. J., 34, 1857 (1998).
67. K.M. George, J.K. Varkey, K.T. Thomas, N.M. Mathew, J. Appl. Polym. Sci., 85, 292
(2002).
68. P.L. Teh, Z.A.M. Ishak, A.S. Hashim, J. Karger-Kocsis, U.S. Ishiaku, Eur. Polym. J.,
40, 2513 (2004).
69. M. Arroyo, M.A. Lopez-Manchado, J.L. Valentin, J. Carretero, Comp. Sci. Technol.,
67, 1330 (2007).
70. K. Pale, R. Rajasekar, D.J. Kang, Z.X. Zhang, J.K. Kim, C.K. Das, Mater. Design.,
30, 4035 (2009).
71. A.K. Manna, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer, J. Appl. Polym. Sci., 74,
389 (1999).
72. D.N. Bikiaris, A. Vssiliou, E. Pavlidou, G.P. Karayannidis, Eur. Polym. J., 41, 1965
(2005).
73. K. Sahakaro, S. Beraheng, J. Appl. Polym. Sci., 109, 3839 (2008).
74. H. Ismail, A. Rusli, A.A. Rashid, Polym. Test., 24, 856 (2005).
75. H. Ismail, A. Rusli, A.R. Azura, J. Polym. Environ., 25, 67 (2007).
76. N. Suzuki, M.Ito, S. Ono, J. Appl. Polym. Sci., 95, 74 (2005).
77. S.-S. Choi, J. Appl. Polym. Sci., 79, 1127 (2001).
78. S.-S. Choi, J. Appl. Polym. Sci., 85, 385 (2002).
79. H. Yan, K. Sun, Y. Zhang, Y. Zhang, Polym. Test., 25, 32 (2005).
80. J.A. Brydson, “Rubber Chemistry”, Applied Science Publishers, London, 1978.
81. G. Wang, M. Li, X. Chen, J. Appl. Polym. Sci., 72, 577 (1999).
82. P. Sae-oui, C. Sirisinha, U. Thepsuwan, K. Hatthapanit, Eur. Polym. J., 42, 479
(2006).

42
 Reinforcement efficiency of silica-filled rubber with different compatibilizing techniques

83. P. Sae-oui, C. Sirisinha, U. Thepsuwan, K. Hatthapanit, Eur. Polym. J., 43, 1853
(2007).
84. S.-S. Choi, J. Appl. Polym. Sci., 83, 2609 (2002).
85. A. Das, S.C. Debnath, D. De, D.K. Basu, J. Appl. Polym. Sci., 93, 196 (2004).
86. A. Bertora, M. Castellano, E. Marsano, M. Alessi, L. Conzatti, P. Stagnaro,
Macromol. Mater. Eng., 296, 455 (2011).
87. H.-M. Schmidtchen, H.U. Voigt, US Patent 4618654, to Kabel-und Metallwerke
Gutehoffnungshuette, October 21, 1986.
88. Y.-T. Shieh, C.-M. Liu, J. Appl. Polym. Sci., 74, 3404 (1999).
89. J.H. Shim, J.H. Joo, S.H.Jung, J.-S. Yoon, J. Polym. Sci. Part B: Polym. Phys., 45,
607 (2007).
90. J.S.Parent, R. Parodi, W. Wu, Polym. Eng. Sci., 46, 1754 (2006).
91. M.S.C. Kumar, M. Alagar, Eur. Polym. J., 38, 2023 (2002).
92. M.S.C. Kumar, M. Alagar, A.A. Pradu, Eur. Polym. J., 39, 805 (2003).
93. M. Alagar, S.M.A. Majeed, S. Nagendiran, Polym. Adv. Technol., 16, 582 (2005).
94. Y. Changjie, Z. Qiuyu, Y. Dezhong, G. Junwei, G. Guangbi, L. Tao, J. Appl. Polym.
Sci., 119, 2808 (2011).
95. G. Colucci, A.D. Gianni, R. Bongiovanni, A. Priola, F. Torello, S. Bracco, P. Sozzani,
Macromol. Symp., 296, 550 (2010).
96. F. Schapman, J.P. Couvercelle, C. Brunel, Polymer, 39, 4955 (1998).

43
Chapter 2

44
 CHAPTER 3

Silica-Reinforced Tire Tread Compounds Compatibilized by

Using Epoxidized Natural Rubber

Epoxidized natural rubber (ENR) is a modified form of NR bearing polar epoxy

groups. This chapter investigates a silica-reinforced NR tire tread compounds with ENR as
a compatibilizer. The ENRs consisting of 10, 38 and 51 mol% epoxide are used in a range
of 2.5 to 15.0 parts per hundred parts of rubber (phr). The addition of ENRs, especially
ENR-38 and ENR-51, decreases the Mooney viscosity, Payne effect, flocculation rate
constant and filler networking factor, which implies an improvement of silica dispersion in
the compounds. Chemically bound rubber contents and interaction parameters of the
compounds also increase with higher epoxide-contents of the ENRs, indicating more
interactions and/or reaction between the epoxide-groups of the ENR and silanol groups on
the silica surface. The different level of compatibility in the filled-NR vulcanizates that
influences silica dispersion is also evidenced by SEM analysis. Tensile strength of the
vulcanizates is improved with increasing mole% epoxide, and the optimum value is
observed at 7.5 phr of ENR-51. The presence of ENR as compatibilizer increases tan δ at
both 0oC and 60oC which implies improved wet grip but also increased tire rolling
resistance. The overall results show that silica-reinforced NR can be substantially
improved by adding ENR as a compatibilizer, when compared to a compound without
ENR, but somewhat less than with using a silane coupling agent.

Part of this chapter was published in European Polymer Journal, 51, 69 (2014).
Chapter 3

3.1 Introduction

Silica and carbon black are widely used reinforcing fillers in rubber. Because of
their different surface characteristics, both fillers can form agglomerates but the cause of
such agglomeration is not the same, leading to a difference in dispersability. The filler-filler
interaction of carbon black is mainly generated through relatively weak Van der Waals
forces, which can be readily broken during mixing. In contrast, silica agglomeration is due
to hydrogen bonding in addition to Van der Waals forces and other physical interactions,
leading to a much stronger filler-filler interaction.[1] Silica has a high polarity and a
hydrophilic surface due to silanol groups on its surface. Consequently, silica is
incompatible with non-polar rubbers such as natural rubber (NR), styrene-butadiene
rubber (SBR) and polybutadiene rubber (BR), but better compatible with polar rubbers
such as polychloroprene rubber (CR)[2] and acrylonitrile butadiene rubber (NBR).[3] Use of
silica in non-polar rubbers without compatibilizers results in poor dispersion and
consequently inferior properties as a result of these strong filler-filler interactions.[4] In
addition, polar functional groups on the silica surface may form hydrogen bonds with other
polar components in rubber compounds such as accelerators. So, commonly basic
accelerators are adsorbed on the acidic surface of silica which negatively affects cure
properties.[5] Successful use of silica for rubber reinforcement therefore requires
commonly silane coupling agents to enhance silica-rubber interaction and silica dispersion
as well as to prevent accelerator adsorption on the silica surface.
Some polar rubbers bearing functional groups which can interact with the silica
surface have been studied as alternatives to silane coupling agents to enhance the
interaction in silica-filled compounds.[6-9] A non polar rubber such as polybutadiene was
chemically modified to have epoxy groups on the main chain and the interaction of such
epoxidized rubber with silica was consequently improved, as observed by a decrease of
the Payne effect[10] and a change of damping behavior in the glass transition region. The
improvement of silica-rubber interaction was also demonstrated by a higher bound rubber
content and better silica dispersion in the matrix.[11] Incorporation of silica into an
epoxidized SBR showed a strong interaction between the silica and the epoxidized
functional groups due to the highly polar character of the epoxidized rubber.[12]
Epoxidized natural rubber (ENR) is also higher polar than virgin NR due to the
epoxide groups in its structure. The chemical and physical properties of ENR change
according to the epoxide contents, such as heat and swelling resistances.[13] By its polar
functional groups, ENR interacts with hydroxyl groups on the silica surfaces.[14] It has been
reported already that the mechanical properties of silica-filled ENR without coupling agent

46
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

are higher than those of silica-filled virgin NR.[15,16] This was claimed to be due to
improved interaction between ENR and the silica surface via hydrogen bonds. ENR as
the rubber component in tire compounds has been reported to improve the silica
dispersion leading to improved processability, better storage stability and tire
[14,17-19]
performance. ENR was used as a compatibilizer in silica-filled NR/NBR blends to
enhance the mechanical properties.[20] In case of silica-filled NBR vulcanizates, the use of
ENR as a compatibilizer improved tensile strength, tear strength and abrasion loss,[21] as
well as lowered the loss tangent,[22] as the ENR acts as a coupling agent because it can
self-crosslink with NBR and also link with silica at the vulcanization temperature.[21,22]
Furthermore, ENR was used as a compatibilizer in organoclay-filled NR composites
wherein cure characteristics, tensile and dynamic mechanical properties were
[23,24]
improved. Even though ENR has been used to compatibilize silica and various rubber
matrices, the use of ENR as compatbilizer for silica-filled virgin NR has not yet been
extensively studied.
In the present study, ENRs with different mole% of epoxide groups are used as
compatibilizers in a range of 2.5 to 15 phr for virgin NR/silica compounds. The effects of
ENR contents and mole% of epoxide are investigated. The epoxide groups are supposed
to interact with the silanol groups of the silica and so to improve silica-rubber
compatibilization, while the unmodified parts of the ENR structure are compatible with the
NR matrix.

3.2 Experimental

3.2.1 Materials

The rubbers used were Natural Rubber (Ribbed Smoked Sheet (RSS) #3),
locally produced in Thailand, and Epoxidized Natural Rubbers with 10, 38 and 51 mol% of
epoxide, denoted as ENR-10, ENR-38 and ENR-51, respectively, that were in-house
prepared as detailed in 3.2.2. The compounding ingredients were highly dispersible silica
(Zeosil 1165MP, Rhodia, France), bis-(triethoxysilylpropyl)tetrasulfide (TESPT) (Evonik,
Germany), treated distillate aromatic extract oil (TDAE oil) (Hansen & Rosenthal,
Germany), N-cyclohexyl-2-benzothiazole sulfenamide (CBS), diphenyl guanidine (DPG)
and 2,2,4-trimethyl-1,2-dihydroquinoline (TMQ) (all from Flexys, Belgium), ZnO, stearic
acid and sulfur (all from Sigma-Aldrich Chemie, Germany).

47
Chapter 3

3.2.2 Preparation of epoxidized natural rubber

High ammonia (HA) natural rubber latex with 60 wt% of dry rubber content was
used to prepare ENR via an in-situ performic epoxidation reaction.[14] The reaction
between C=C of the NR molecule and performic acid, arising from a reaction between
formic acid and hydrogen peroxide, was carried out in a continuously stirred reactor at a
temperature of 40oC, where alkylphenol ethoxylate non-ionic surfactant (Teric N30,
Huntsman Corp., Australia) was used as a stabilizer.[25,26] The ENRs with 10, 38, and 51
mole% of epoxide groups were obtained when the reaction times were varied at 2, 10 and
12h, respectively. The ENR latex was subsequently coagulated with methanol. The ENR
coagulum was sheeted, washed thoroughly with water and dried in an oven at 50oC for
approximately 4 days. The reaction time of epoxidation was set according to the required
level of epoxide groups in the ENR product. The 1H NMR spectroscopic technique was
used to analyze the molecular structure of the ENR and a typical 1H NMR spectrum of
ENR is shown in Figure 3.1 with details of proton types assigned for each peak.

2
2 6
3 7
1,4

1 4 5 8
6

3
5,8

6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0

Figure 3.1 Typical 1H NMR spectrum of ENR and its proton type assignment.

The mole% of epoxide groups was calculated using Equation (3.1):

b
mol % of epoxide groups   100 (3.1)
ab

48
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

where a is the integrated peak area of the olefinic proton of NR at 5.1 ppm and b is the
integrated peak area of H attached to the oxirane rings of ENR at 2.7 ppm, respectively,
as shown in Figure 3.2. The NMR spectra of NR and ENRs with different levels of epoxide
groups are shown in Figure 3.2.

Figure 3.2 1H NMR spectra of natural rubber (NR) and epoxidized natural rubbers (ENRs)
with 10, 38 and 51 mol% epoxide, denoted as ENR-10, ENR-38 and ENR-51.

3.2.3 Compound preparation

Rubber compounds were prepared using the formulations as shown in Table 3.1.
The ENR content was varied in a range of 2.5 to 15.0 phr. Amounts of TESPT and DPG
were calculated relating to the silica CTAB specific surface area, as suggested by Guy et
al..[27] Mixing was carried out using an internal mixer (Brabender Plasticorder 350s) with
an initial mixer temperature setting of 110oC, rotor speed of 60 rpm, according to the
mixing procedure as shown in Table 3.2. The silica-filled NR compounds with and without
TESPT and without ENR were prepared and treated as reference.

49
Chapter 3

Table 3.1 Compound formulations

Ingredients Amounts (phr)
RSS 3 100.0 100.0 97.5-85.0
ENRs - - 2.5-15.0
Zeosil 1165MP 55.0 55.0 55.0
TESPT* - 4.5 -
TDAE oil 8.0 8.0 8.0
ZnO 3.0 3.0 3.0
Stearic acid 1.0 1.0 1.0
TMQ 1.0 1.0 1.0
DPG* 1.0 1.0 1.0
CBS 1.5 1.5 1.5
Sulfur 1.5 1.5 1.5
[27]
*Amounts of TESPT and DPG were calculated according to the following equations:
TESPT (phr) = 0.00053×Q ×A and DPG (phr) = 0.00012 ×Q×A
2
where Q is the amount of silica (phr) and A is the CTAB surface area of the silica (155 m /g).

Table 3.2 Two-step mixing procedure

Cumulative time
Mixing procedure
(mins.)
Step 1 : Internal mixer
- NR and ENR mastication 2
- Addition of a first half of (silica+TESPT) 7
- Addition of a second half of (silica+TESPT) and 12
TDAE oil
- Addition of ZnO, stearic acid, and TMQ 15
Step 2 : Two roll mill
- Addition of DPG, CBS, and sulfur 5

3.2.4 Mooney viscosity, Payne effect and flocculation rate constant of

unvulcanized compounds

Mooney viscosity [ML(1+4), 100oC] was tested using a Mooney viscometer (MV
2000VS, Alpha Technologies) according to ASTM D1646. The Payne effect[10] or filler-
filler interaction of the uncured silica-filled compounds were studied by using a Rubber
Process Analyzer (RPA 2000, Alpha Technologies) at 100oC, frequency 0.5 Hz and
varying strains in the range of 0.56 to 100%. The difference of storage moduli at low strain
(i.e. 0.56%) and high strain (i.e. 100%) is reported.
The flocculation rate constant (ka) of the uncured silica-filled compounds was
studied by using the RPA 2000 at 100oC, strain 0.56%, frequency 1.00 Hz, and test time
for 12 minutes. The ka was calculated following Equations (3.2) and (3.3):[28]

50
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

s ' (t )  s ' (1)

x (3.2)
s ' ()  s ' (1)

where x is the degree of flocculation, s (t ) is the storage modulus at 0.56% strain at

test time t, s (1) is the storage modulus after preheating for 1 minute, and s() is the

storage modulus after heating for 12 minutes.

ln(1  x1 )  ln(1  x2 ) 1
ka  [min ] (3.3)
t 2  t1

where x1 and x2 are the degrees of flocculation at different heating times, i.e. t1 and t2,

respectively.

3.2.5 Networking factor and interaction parameter

Silica-filled NR compounds without curatives obtained after the first mixing step
were compressed at 150oC for 30 minutes to obtain rubber sheets with a thickness of 2
mm, which were then cut into specimens of 40×7×2 mm3 dimensions. The strain-
dependent dynamic mechanical properties were then analyzed by using a dynamic
mechanical analyzer (DMA) (Viscoanalyzer, VA2000, Metravib, France) at a constant
o
frequency of 3.5 Hz at 25 C and varying strain amplitudes in the range of 0.083% to 5.0%.
A filler-filler networking factor (η) was calculated from the ratio of storage moduli at
0.083% and 5.0% strain of the compounds without curatives.
As proposed by Ayala et al.[29] originally for determination of the carbon black-
rubber interaction, a polymer-filler interaction parameter (I) can be calculated from the
static and dynamic moduli of a compound according to Equation (3.4):


I  (3.4)


Herein, σ is the slope of the stress-strain curve of a cured vulcanizate taken at a relatively
linear region at low elongation (i.e. at 10% elongation in this work).

51
Chapter 3

3.2.6 Bound rubber measurement

0.25 g of uncured compound (without curatives) was cut into small pieces, put
into a metal cage and immersed in toluene at room temperature for 72 h (renewed every
24 h). The sample was removed from the toluene, dried at 50oC for 24 h, then immersed
in toluene again for 72 h at room temperature in either a normal or an ammonia
atmosphere. The ammonia treatment was done to cleave the physical linkages between
rubber and silica, in order to determine the chemically bound rubber versus bound rubber
physical of nature. The sample was finally dried at 50oC for 24 h. The bound rubber
content was then calculated using the following equation:[30]

( m  ms )
Bound rubber (%)   100 (3.5)
mr

Where m is the weight of sample after extraction, ms is the weight of silica in the sample
and mr is the original weight of rubber in the sample.

3.2.7 Cure characteristics and tensile properties

Cure properties of the compounds were studied by using the RPA at 150oC,
frequency 0.833 Hz and 2.79% strain for 30 minutes. Then, the compounds were
vulcanized to their optimum cure time (tc90) by using a Wickert WLP 1600 laboratory
o
compression press at 150 C and 100 bars into 2 mm thick sheets. Type 2 dumb-bell test
specimens were die-cut from the press-cured sheets and tensile tests were carried out
with a Zwick tensile tester Model Z1.0/TH1S at a crosshead speed of 500 mm/min
according to ASTM D412.

3.2.8 Loss tangent at 60oC by RPA and dynamic mechanical properties by DMA

The loss tangent of the vulcanizates was determined using the RPA at 60oC,
strain 3.49% with varying frequency sweeps in the range of 0.05-33.00 Hz. The samples
were cured in the RPA chamber at 150oC to reach their optimum cure times before being
tested.
Dynamic mechanical properties of silica-filled NR vulcanizates, i.e. storage
modulus, loss modulus and tan δ, were characterized with a Metravib Viscoanalyzer VA
2000. The samples were cut into 6 × 4 × 2 mm3 specimen and tested in tension mode
with a temperature sweep from -80oC to 80oC at a frequency of 10 Hz and 0.1% strain.

52
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

3.2.9 FTIR study of silica-filled NR vulcanizates

The silica-filled NR compounds with ENRs containing different mol% epoxide as

compatibilizer were vulcanized to their optimum cure times before being characterized
with ATR-FTIR spectroscopy to investigate the silica-rubber interaction in the compounds.
The ENR content in the samples was fixed at 7.5 phr, while the mol% of epoxide groups
was varied. The silica-filled NR compounds with TESPT and without compatibilizer were
also analyzed for comparison.

3.2.10 SEM analysis of silica-filled NR vulcanizates

The dispersion of silica in the NR matrix with ENR as compatibilizers at 7.5 phr
was analyzed by Scanning Electron Microscopy (SEM). The silica-filled NR compounds
were prepared by reducing silica contents from 55 phr in the normal formation to 30 phr in
order to be able to observe the silica dispersion in the NR matrix. The compounds were
vulcanized to their optimum cure times, and a newly fractured surface of vulcanizate was
created by cryogenic cracking after immersing in liquid nitrogen. The fractured surfaces of
the specimens were coated with gold before being analyzed.

3.3 Results and discussion

3.3.1 Mooney viscosity, Payne effect and flocculation tendency

80

70

60
ML1+4 (100oC)

50

40

30

20
ENR-10 ENR-38

10 ENR-51 TESPT

0
0 5 10 15 20
ENR contents (phr)
Figure 3.3 Mooney viscosity of silica-filled natural rubber with ENR as compatibilizer.

53
Chapter 3

The Mooney viscosities of the compounds with ENR-10 as compatibilizer show

almost no change when the amount of ENR-10 is increased from 2.5 to 15 phr: Figure 3.3.
Incorporation of ENR-38 and ENR-51 decreases the Mooney viscosities of the
compounds to match the level of that of the TESPT-containing compound when the
amount of ENRs used is ≥ 10 phr. The decrease of compound viscosity when ENR is
used as compatibilizer implies an improvement of silica dispersion due to interaction
between epoxide groups and silanol groups of the silica resulting in better compatibility.
However, the compound viscosities can also be affected by various factors including the
NR/ENR blend incompatibility. The presence of polar epoxy groups in the ENRs leads to
a difference between the solubility parameters of NR and ENR, and increases interfacial
tension. It has been reported that the NR/ENR blends exhibit two distinct glass transition
temperatures (Tg) corresponding to the original polymers, as analyzed by differential
scanning calorimetry (DSC)[31], and dynamic mechanical analysis (DMA).[32]

Figure 3.4 Payne effect (a) and flocculation rate constant (ka) (b) of silica-filled natural
rubber with ENR as compatibilizer.

Filler-filler interaction, as indicated by the Payne effect in Figure 3.4(a), of the

filled compounds also decreases when the ENR content is increased. The compound
without any compatibilizer shows the highest filler-filler interaction. ENR-10 as
compatibilizer already shows a strong reduction of the Payne effect with increasing
content. The compounds with ENR-38 and ENR-51 show similar values of the Payne

54
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

effect which is much lower compared to the compounds containing ENR-10. The lower
Payne effect indicates less silica-silica interactions, usually indicating better dispersion of
silica in the matrix. The incorporation of ENR-38 and ENR-51 in the range of 7.5 to 15 phr
results in compounds that have similar Payne effects compared to the compound with
TESPT. The reduction of filler-filler interaction in the presence of ENR in silica-filled NR
compounds can again be attributed to interactions between the silanol groups of silica and
epoxide groups of ENR through hydrogen bonding and/or chemical bonding that occur
after possible ring opening of the epoxy ring either during mixing or compression molding,
as previously proposed by Manna et al.[33] and Rocha et al..[12] A possible mechanism for
bonding between the ENR and silica is shown in Scheme 3.1, and the model of
interactions and/or reactions between the silanol groups of silica and ENR molecules in
the NR matrix is proposed in Scheme 3.2.

HO OH O
H3C H
HO Si O Si C C
O O OH
CH2 CH2
Si Si
silica surface epoxidized natural rubber (ENR)
Mixing

H2C CH2 H2C CH2

C C C C
H CH3 H CH3
OH O HO
O OH
Si O H
H Si O
O H
Si Si O
O and/or
O
Si O Si O
O H H
Si O
H
O Si O OH
HO H CH3
HO H
C C C C CH3
CH2 CH2 CH2 CH2
Molding

H2C CH2
HO C C
H CH3
Si O OH
O
Si
O
Si
O
O OH
Si H
C C CH3
OH
H2C CH2

Scheme 3.1 Possible mechanism for interactions and bonding between ENR and silica.

55
Chapter 3

OH
HO Si
Si O

O
NR matrix
H

ENR molecules

Silica aggregates
H
O
Interaction/reaction types

OH
Si OH
H
O

Scheme 3.2 Proposed model of interactions and bonding between silanol groups on the
silica surface and ENR molecules in NR matrix.

Due to the presence of silanol groups on the silica surface that can form
hydrogen bonds between silica aggregates, these aggregates can again form
agglomerates after mixing. This reformation of filler network in the compound is also called
filler flocculation. The flocculation process of the silica-silica network strongly increases
with heat treatment.[28, 34] The results in Figure 3.4(b) show that the silica-filled compound
with neither ENR nor TESPT has the highest flocculation rate constant (ka) compared to
the other compounds, because of the strong hydrogen bonding between the silica
aggregates.[35] After the ENRs are added into the silica-filled compounds, the flocculation
rate constant decreases, especially when ENR-51 is increased to 7.5 phr, which
compound shows almost the same level of ka compared to the silica-filled compound with
TESPT. The results confirm that the use of ENR as compatibilizers can effectively reduce
filler-filler interaction in the compounds leading to a reduction of filler flocculation of the
compounds.

56
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

3.3.2 Networking factor and interaction parameter

Table 3.3 Storage modulus (E) at low strain (0.083%) and high strain (5.0%) and ΔE at
25oC
E at 0.083% strain E at 5.0% strain ΔE
Sample
(MPa) (MPa) (MPa)
Without compatibilizer 42.40 7.96 34.44
ENR-10 7.5 phr 27.00 5.24 21.76
ENR-38 7.5 phr 26.70 6.73 19.97
ENR-51 7.5 phr 20.60 5.75 14.85
TESPT 4.5 phr 8.37 2.89 5.48

6 0.025
(a) (b)

5
0.020
Interaction parameter, I (MPa)
Networking factor, η

4
0.015

3
0.010
2
ENR at 7.5 phr
0.005
1 TESPT ENR at 7.5 phr

TESPT
0 0.000
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Mole% of epoxide Mole% of epoxide

Figure 3.5 Filler networking factor (a) and rubber-filler interaction parameter (b) of silica-
filled natural rubber with ENR at 7.5 phr as compatibilizer in comparison with the
vulcanizates containing TESPT silane coupling agent.

The variation of storage modulus with % strain of silica-filled vulcanizates has

been related to the networking factor and interaction parameter of the filled samples.[29, 36]
The decrease of storage modulus with increasing strain amplitude infers that the filler-filler
network in the rubber matrix is broken down. The storage modulus at low strain is
strongly affected by the filler-filler interaction in the system, while at high strain amplitude
after the break down of filler-filler network such property is only affected by filler-rubber

57
Chapter 3

interaction, the filler volume fraction and rubber network in the sample.[10,37] As shown in
Table 3.3, the difference of storage moduli at 5.0% and 0.083% strain (ΔE′) of the silica-
filled NR vulcanizates decreases with increase of mole% of epoxide groups, compared to
the silica-filled NR without any compatibilizer. Herein, the rubber with TESPT silane
coupling agent shows the lowest ΔE′. Figure 3.5(a) shows the networking factor (η) of
silica-filled NR compounds with ENRs as compatibilizer. The results show a slight
decrease of filler-filler networking factor (η) when ENR-10 was added into the compounds,
but a strong decrease of η when 7.5 phr of ENR-38 and ENR-51 was used. This again
indicates that the epoxide groups of ENRs interact and/or react with silanol groups on the
silica surface leading to a reduction of silica-silica interaction or silica agglomeration in the
NR matrix and an improvement of silica dispersion. When considering the rubber-filler
interaction parameter (I) which is defined by the ratio of static to dynamic moduli of the
filled-rubber[29], as shown in Figure 3.5(b), the values of I increase with mole% of epoxide
groups of ENRs, but the ultimate effect is lower than that of the silica-filled compound with
TESPT. In the silica-silane system, TESPT can form covalent bonds with the silica surface
via the silanization reaction between ethoxy groups of TESPT and Si-OH groups of silica,
and generate crosslinks with the rubber phase via sulfur atoms in its structure.[38-39] Even
though the use of TESPT provides the lowest filler-filler interaction as indicated by the
lowest Payne effect: Figure 3.4(a), the flocculation rate constant: Figure 3.4(b), the filler
networking factor: Figure 3.5(a) and highest filler-rubber interaction parameter: Figure
3.5(b), the use of ENRs as compatibilizer clearly demonstrates that the epoxy groups on
the main chain of ENR can effectively reduce the filler-filler interactions. By increasing the
epoxide groups in the system, the polymer-filler interaction parameter (I) increases whilst
the Payne effect, flocculation rate constant and filler-filler networking factor (η) decrease.

58
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

3.3.3 Filler-rubber interaction

90
ENR-10 ENR-38 ENR-51
80 ENR-10 ENR-38 ENR-51

70 TESPT TESPT

Bound rubber content (%)

60

50

40
Chemically bound rubber
30

20
Physically bound rubber
10

0
0 5 10 15 20
ENR contents (phr)

Figure 3.6 Physically and chemically bound rubber contents of silica-filled NR with ENR as
compatibilizer.

The chemically bound rubber contents of the silica-filled NR significantly increase

while the physically bound rubber contents slightly decrease by increasing the ENR
contents and mole% of epoxide on ENR, as shown in Figure 3.6. When only NR is used,
the addition of TESPT coupling agent leads to a remarkably high chemically bound rubber
content, compared to the ones without. The addition of ENR into the NR compounds
without TESPT clearly improves the bound rubber content, i.e. filler-rubber interaction.
These results prove that ENR can generate chemical interaction and/or reaction in silica-
filled NR, according to the mechanism as proposed in Scheme 3.1. The use of ENR-51 in
the range of 7.5-15 phr results in an optimum chemically bound rubber content. This is in
accordance with the Payne effect: Figure 3.4(a) and flocculation rate constant results:
Figure 3.4(b), attributed to the compatibilization effect of ENR. For the NR compound
without ENR or TESPT, there is still chemically bound rubber seen in Figure 3.6. This can
be mainly attributed to the nature of NR itself which has the terminal groups associated
with proteins and phospholipids[40] that can possibly interact with the silanol groups of the
silica surface via hydrogen bonding. The interactions between silica and proteins present
in NR has been clearly demonstrated recently.[41]

59
Chapter 3

3.3.4 Cure properties

12 20
(a) (b)
18
10
16

14

Cure time (min)

Scorch time (min)

8
12

6 10

8
4
6

4
2 ENR-10 ENR-38
ENR-10 ENR-38
2 ENR-51 TESPT
ENR-51 TESPT
0 0
0 5 10 15 20 0 5 10 15 20
ENR contents (phr) ENR contents (phr)

Figure 3.7 Scorch (a) and optimum cure times (b) of silica-filled NR with ENR as
compatibilizer.

Scorch time (ts1) and optimum cure time (tc90) of the compounds are shown in
Figures 3.7(a) and 3.7(b), respectively. Both scorch and optimum cure times of the
compounds are prolonged when ENR contents and mole% of epoxide are increased. This
might be due to the high polarities of ENR and silica which interfere with the vulcanization.
Even though part of the silanol groups are assumed to have interacted with epoxide
groups, the remaining free silanol groups as well as free epoxide groups can possibly form
hydrogen bonds with polar accelerators, causing accelerator adsorption on the polar
surface. In addition, due to the polarity difference between NR and ENR, the more polar
curatives can migrate into the ENR phase, and so cause a delay of the vulcanization
reaction in the NR matrix. The silica filled-compound with TESPT displays shorter scorch
and cure times, because TESPT itself can act as sulfur donor and optimized mixing
conditions were employed to ensure a good silanization reaction.

60
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

3.3.5 Tensile properties

Reinforcememnt index (M300/M100)

5

ENR-10 ENR-38
1
ENR-51 TESPT

0
0 5 10 15 20
ENR contents (phr)

Figure 3.8 Reinforcement index of silica-filled natural rubber with ENR as compatibilizer.

The silica-filled NR with TESPT shows a higher reinforcement index

(M300/M100) when compared to the compounds with ENRs as compatibilizers: Figure
3.8, indicating a greater extent of interactions between the silica and rubber phases when
the silane coupling agent is used, as also previously observed by the chemically bound
rubber content: Figure 3.6. When compared among the ENR types, the use of ENR-51
gives the highest reinforcement index. An increase of ENR content tends to decrease the
reinforcement index slightly. This may be caused by the presence of two different rubber
phases in the system due to the incompatibility between NR and ENR[31, 32], as previously
mentioned in Section 3.1.

61
Chapter 3

30 700
(a) (b)

25 600

Elongation at break (%)

Tensile strength (MPa)

500
20
400
15
300
10
200
ENR-10 ENR-38 ENR-10 ENR-38
5
100 ENR-51 TESPT
ENR-51 TESPT

0 0
0 5 10 15 20 0 5 10 15 20
ENR contents (phr) ENR contents (phr)

Figure 3.9 Tensile strength (a) and elongation at break (b) of silica-filled natural rubber
with ENR as compatibilizer.

The use of ENR-10 shows no positive effect on the tensile strength of the silica-
filled NR, but the addition of ENR-38 and ENR-51 enhances the tensile strength,
compared to the compound without compatibilizer: Figure 3.9(a). ENR-51 gives a higher
tensile strength than ENR-38, and increasing the ENR content increases tensile strength
to a maximum whereafter the property drops. The optimum value is observed at 7.5 phr
for ENR-51 and 12.5 phr for ENR-38. The results support the proposition that the epoxide
functional groups improve silica-rubber interaction in the compounds. This is in
accordance with the increased bound rubber content in Figure 3.6 and the decreased
elongation at break as shown in Figure 3.9(b).

62
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

3.3.6 Dynamic mechanical properties

0.25

0.20

Tan δ at 60oC
0.15

0.10

0.05 ENR-10 ENR-38

ENR-51 TESPT

0.00
0 5 10 15 20
ENR contents (phr)

Figure 3.10 Tan δ at 60oC of silica-filled NR with ENR as compatibilizer.

The loss tangent or tan δ at 60oC of the vulcanizates is an indication of tire rolling
resistance in a laboratory scale test. The compound with lower tan δ at 60oC implies a
lower rolling resistance of tires made thereof. The tan δ of silica-filled NR vulcanizates at
10 Hz, as shown in Figure 3.10, reveal that the compound with TESPT shows an
outstanding low rolling resistance compared to the other compounds because of good
silica dispersion and silica-to-rubber bonding via the silane molecule. On the other hand,
the silica-filled NR compounds with ENRs as compatibilizers show an increase of tan δ
with mole% of epoxide content. The addition of ENR-10 results in a similar tan δ,
compared to the compound without coupling agent, and shows no strong dependence on
ENR-10 loading. But the use of ENR-38 and ENR-51 as compatibilizer clearly increases
tan δ which negatively affects the tire rolling resistance. This increase of tan δ is
associated with the damping properties of ENRs as the glass transition temperature of this
material is shifted to higher values with increasing mol% epoxide.[13, 26]
The glass
transition temperatures (Tg) of ENR-10, ENR-38 and ENR-51 are approximately -60 oC, -
30 oC and -25oC, respectively.[26] However, the Tg of reinforcing ENR-50 with silica occurs
at -3oC and either the increase of molding time or addition of silane coupling agent
influences on shifting the Tg of silica-filled ENR-50 to higher temperature.[36] It is due to the
movement of rubber segments is restricted by the present of chemically bonding of rubber

63
Chapter 3

chains to silica surfaces.[36,42] In addition, the incorporation of a second polymer phase

with different polarity compared to the matrix normally leads to a phase separation and
has an influence on dynamic properties.[43] The weaker interfacial adhesion leads to more
energy losses under dynamic deformation and so a higher loss tangent.
The dynamic mechanical properties as a function of temperature of silica-filled
NR with 7.5 phr of various ENRs as compatibilizers are shown in Figures 3.11-3.12. The
temperature dependence of dynamic storage modulus (Eʹ) and loss modulus (Eʺ) of silica-
filled NR vulcanizates is shown in Figure 3.11. The silica-filled NR vulcanizate without
compatibilizer shows the lowest storage modulus due to the poorest interfacial adhesion
between silica and rubber. Its strongest filler-filler interactions as indicated by Payne effect
(Figure 3.4(a)) suggest a formation of silica agglomerates which introduce voids and pores
that result in a decrease of storage modulus of the materials.[44] The storage moduli of the
vulcanizates with TESPT and ENRs in the rubbery region are clearly higher than that of
the rubber without compatibilizer. The vulcanizates with ENR-38 and ENR-51 as
compatibilizer show a two-step change of E′ and Eʺ due to a second glass transition
temperature introduced by the ENRs. The interactions between silica and rubber phases
as well as between ENR molecules result in the higher E′, but due to the damping
behavior of the ENRs[13,26], the NR vulcanizates with ENRs as compatibilizer show higher
Eʺ when compared to the one with TESPT. Moreover, the immiscibility between NR and
ENR,[45] contributes to a higher loss of energy and so a loss modulus Eʺ as shown in
Figure 3.11(b).

64
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

10000 10000
(a) (b)

1000
1000

E" (MPa)
E' (MPa)

ENR-51
100 ENR-51
ENR-38
ENR-38
100 ENR-10 ENR-10
TESPT 10
TESPT

Withoutcompatibilizer Withoutcompatibilizer
10 1
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Temperature (oC) Temperature (oC)

Figure 3.11 Storage modulus (a) and loss modulus (b) of silica-filled NR vulcanizates with
7.5 phr of ENRs as compatibilizer.

1.2
TESPT
1.0
Withoutcompatibilizer
ENR-10
0.8
ENR-38
Tan δ

0.6 ENR-51

0.4

0.2

0.0
-80 -60 -40 -20 0 20 40 60 80
Temperature (oC)

Figure 3.12 Tan δ as a function of temperature of silica-filled NR vulcanizates with 7.5 phr
of ENRs as compatibilizer.

65
Chapter 3

Figure 3.12 shows the effect of ENR compatibilizers in silica-filled NR

vulcanizates on the loss tangent (tan δ) as a function of temperature. The addition of 7.5
phr of ENRs with different levels of mole% epoxide results in a different response of the
tan δ curves. The height of the tan δ peaks is decreased when the epoxide group level is
increased. This is caused by the interactions between silica and ENRs in the NR matrix
and between ENR molecules themselves, leading to a restriction of the molecular chain
mobility and relaxation. Furthermore, the higher epoxide group content leads to a possible
self-crosslinking between the ENR itself and introduces an immobilization of the rubber
chains. Second transitions of tan δ are observed when ENR-38 and ENR-51 were added
into the silica-filled NR compounds, in accordance with the second step changes of E′ and
Eʺ as observed in Figure 3.11. These are again attributed to the glass transition
temperatures of ENR-38 and ENR-51. The highest tan δ peak is observed for the
silica/TESPT-filled vulcanizate, which can be described by its better silica dispersion
causing a decrease of rubber trapped within the filler network and more mobility of rubber
segments in the matrix that involve in the relaxation process. The change of tan δ peak
height with respect to compatibilizer types, i.e. TESPT and epoxidized rubbers, are in
accordance with the work previously reported by Saramolee et al.[46]

Table 3.4 Transition temperatures at loss modulus and tan δ peaks, and values of tan δ at
different temperatures of silica-filled NR vulcanizates with various compatibilizers.
Glass transition
Values of tan δ
Compatibilizer types temperature (oC)
at Eʺ peak at tan δ peak at 0oC at 5oC at 60oC
Without compatibilizer -56 -50 0.11 0.11 0.12
ENR-10 7.5 phr -55 -50 0.09 0.09 0.09
ENR-38 7.5 phr -55 -50 0.14 0.13 0.10
ENR-51 7.5 phr -54 -50 0.13 0.13 0.11
TESPT -53 -48 0.08 0.07 0.06

The temperatures at peak of loss modulus (Eʺ) and tan δ can be taken as glass
transition temperatures (Tg) of the materials. As reviewed by Robertson and Roland[47], Tg
based on the tan δ peak can be problematic because tan δ in the glass-to-rubber
softening region is influenced not only by the local segmental polymer dynamics but also
by filler induced changes in both storage and loss moduli at higher temperatures. Due to
some discrepancies in Tg position, this work reports the Tg values of silica-filled NR
vulcanizates with and without compatibilizers taken from both Eʺ and tan δ peaks, as
shown in Table 3.4. It can be seen that the Tg’s taken from the Eʺ peaks show a small but

66
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

gradual change of values with compatibilizer types. The silica-filled NR without

compatibilizer shows the lowest Tg and the presence of compatibilizers tends to shift the
Tg slightly to higher temperature in accordance with the degree of interactions as
observed by bound rubber content in Figure 3.6. The silica/TESPT-filled NR vulcanizate
has the highest Tg, while the ENR-compatibilized vulcanizates show similar Tg values. The
shift of Tg to higher temperature is due to the strong interactions between filler and rubber
phases leading to restricted movement of the rubber molecules at the interface.[48]
Tan δ in the low temperature range e.g. in the range of 0-20oC can be used to
indicate wet grip or resistance between tire tread and road surface. A higher tan δ at low
temperature means a better wet grip of a tire. The use of ENR-38 and ENR-51 as
compatibilizers in silica-filled NR shows an increase of tan δ values at low temperature
(Table 3.4) due to the effect of the glass transition temperatures and damping properties
of ENR-38 and ENR-51 in the NR matrix. The incorporation of ENR-38 and ENR-51
positively affects tire wet grip but negatively influences tire rolling resistance, as observed
by the higher tan δ values at 60oC when compared to the silica-filled NR vulcanizate with
TESPT. The results at 60oC are in good agreement with the tan δ values tested by RPA
(Figure 3.10), as previously discussed.

3.3.7 FTIR spectra of silica-filled NR vulcanizates

As reflected in the changes of properties of both compounds and vulcanizates:

Figures 3.3-3.10, the epoxide groups of ENR interact with the silanol groups of silica and
promote compatibilization. In order to gain more evidence of such interactions between
the epoxide groups of ENR and the silanol groups of silica, vulcanized sheets of silica-
filled NR containing either no compatibilizer, TESPT or 7.5 phr of ENR-10, ENR-38 and
ENR-51 were subjected to ATR-FTIR analysis to look for a change of absorption peak
intensities of the functional groups of silica. The spectra of the vulcanizates in the region
of absorption bands that are associated with -OH and Si-O-Si stretching vibrations are
shown in Figure 3.13. Herein, Figure 3.13(a) shows the -OH stretching region at 3500-
3000 cm-1 and Figure 3.13(b) displays the absorption band of Si-O-Si asymmetric stretch
in the range of 1300-900 cm-1.[8, 49]

67
Chapter 3

0.02 0.8
Without compatibilizer (a) Without compatibilizer (b)
0.7
ENR-10 ENR-10
ENR-38
0.6
ENR-51
ENR-38
Absorbance

TESPT

Absorbance
0.5

0.01 TESPT 0.4

ENR-51
0.3

0.2

0.1

0 0
3650 3450 3250 3050 1300 1200 1100 1000 900
Wave number (cm-1) Wave number (cm-1)

Figure 3.13 ATR-FTIR spectra of silica-filled NR vulcanizates with 7.5 phr of ENR as a
compatibilizer compared to that of the vulcanizates with TESPT and without any
compatibilizer.

The ATR-FTIR spectra in Figure 3.13(a) show a reduction of -OH groups of silica
with increase of mol% epoxide and use of TESPT. In addition to hydrogen bonding
between silanol groups and oxirane oxygen of ENR, as proposed in Scheme 3.1 and
Scheme 3.2, -OH groups of silica can react with -OH groups of ENR after ring opening,
and form bonding during molding. Therefore, the free -OH of silanol groups decrease.
According to Bertora et al.[50] and Zhuravlev et al.[51], the surface silanol groups content of
Zeosil 1165MP is 4.90 OH•nm-2. As only part of these silanol groups are able to react or
interact with epoxide groups of ENR for reason of steric hindrance, providing the same
content of ENR compatibilizers with varying mol% epoxide leads to more epoxide groups
available to interact, and less free silanol groups remaining in the matrix. As a result, a
lower intensity of the -OH peak with increasing mol% of epoxide is observed in Figure
3.13(a).
The peak intensity of Si-O-Si asymmetric stretching of silica in the NR matrix at
1080 cm-1 decreases with the addition of ENR as compatiblizer and TESPT as coupling
agent. The higher the mol% epoxide of ENR, the more the decrease of peak intensity,
while TESPT shows the lowest peak intensity. In Figure 3.13(b), the intensities of the peak
shoulder at 965 cm-1 arising from the Si-O of the Si-OH group also change in the same

68
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

order as observed for the peak at 1080 cm-1 when ENR and TESPT are added. This is
also due to the reduction of the amount of Si-O-Si and Si-O bonds in the compounds.
Thus the use of ENR and TESPT in the silica-filled NR compounds reduces silica-silica
interaction leading to a lower number of silica agglomerates in the compounds.

3.3.8 SEM images of silica-filled NR with ENR as compatibilizers

Filler dispersion and distribution is an important factor that influences the

properties of filled-compounds and vulcanizates. Silica primary particles themselves have
a dimension in the nanometer range, but due to the highly polar silanol groups on its
surface, silica can easily form aggregates and agglomerates.[48] The hydrophilic character
of silica makes it difficult to uniformly disperse and distribute within hydrocarbon rubbers in
the absence of any surface modifiers or compatibilizers. The silica dispersion in the NR
matrix with a variation of ENR types as compatibilizer is shown in Figure 3.14. In this
study, the silica content in the compounds was reduced from 55 phr in the normal
formulation to 30 phr for a better clarification of the silica in the matrix.

(a) (b)

(c) (d) (e)

Figure 3.14 SEM micrographs of silica-filled NR vulcanizates with various types of

compatibilizers at 15000x magnification; without compatibilizer (a), ENR-10 7.5 phr (b),
ENR-38 7.5 phr (c), ENR-51 7.5 phr (d) and TESPT (e).

The silica dispersion level of filled-NR vulcanizates without any compatibilizer

and with ENR-10 7.5 phr looks similar and poorer than that of the other compounds, as

69
Chapter 3

aggregates/agglomerates can clearly be observed. The addition of ENR-38 and ENR-51

at 7.5 phr into the silica-filled compounds visibly shows an improvement of silica
dispersion as shown in Figures 3.14c-d, respectively. This improved silica dispersion
again supports the occurrence of interactions and/or reaction between silanol groups on
the silica surface and epoxy groups of ENR, according to the mechanism proposed in
Scheme 3.1 and 3.2, leading to reduction of polar Si-OH groups, as also evidenced in the
FTIR spectra Figure 3.13(a). The silica-filled NR with TESPT silane coupling agent (Figure
3.14(e)) shows the best silica distribution and dispersion which is in accordance with its
superior properties.

3.4 Conclusions

Epoxidized natural rubber (ENR) can be used as compatibilizer in silica-filled

natural rubber, as it results in improved filler-rubber interaction, silica dispersion and
tensile strength. Increase of ENR content and mole% of epoxide groups increases the
interactions between silica and rubber, and decreases filler-filler interaction, as reflected
by the lower Mooney viscosity, Payne effect, flocculation rate constant, filler networking
factor, and the higher filler-rubber interaction parameter, chemically bound rubber, tensile
strength and silica dispersion level. With respect to tire performance, the presence of
ENR-38 and ENR-51 as compatibilizer improves wet grip but negatively affects rolling
resistance. The addition of ENR-51 at 7.5 phr shows the best overall properties, however
still somewhat lower when compared to a compound with TESPT silane as coupling
agent.

3.5 References

1. O. Seyvet, P. Nevard, J. Appl. Polym. Sci., 78, 1130 (2000).

2. A. Das, S.C. Debnath, D. De, D.K. Basu, J. Appl. Polym. Sci., 93, 196 (2004).
3. S.-S. Choi, J. Appl. Polym. Sci., 79, 1127 (2001).
4. J.-B. Donnet, Rubber Chem. Technol., 71, 323 (1998).
5. J.H. Bachmann, J.W. Sellers, M.P. Wagner, R.F. Wolf, Rubber Chem. Technol., 32,
1286 (1959).
6. H. Ismail, H.H. Chia, Eur. Polym. J., 34, 1857 (1998).
7. H. Ismail, A. Rusli, A.A. Rashid, Polym. Test., 24, 856 (2005).
8. A.K. Manna, A.K. Bhattacharyya, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer,
Polymer, 39, 7113 (1998).

70
 Silica-reinforced tire tread compounds compatibilized by using epoxidized natural rubber

9. N. Suzuki, M Ito, S. Ono, J. Appl. Polym. Sci., 95, 74, (2005).

10. A.R. Payne, Rubber Plast. Age, 42, 963 (1961).
11. M.M. Jacobi, M.V. Braum, T.L.A.C. Rocha, R.H. Schuster, Kautsch. Gummi Kunstst.,
60, 460 (2007).
12. T.L.A.C. Rocha, R.H. Schuster, M.M. Jacobi, D. Samios, Kautsch. Gummi Kunstst.,
57, 656 (2004).
13. I.R. Gelling, M. Porter, “Natural Rubber Science and Technology”, Oxford University
Press, Oxford, 1988.
14. A.V. Chapman, In: 24th International H.F. Mark-Symposium on Advances in the Field
of Elastomers & Thermoplastic Elastomers, Vienna, Austria, 2007. p. 15.
15. F. Cataldo, Macromol. Mater. Eng., 287, 348 (2002).
16. Y.Y. Luo, Y.Q. Wang, J.P. Zhong, C.Z. He, Y.Z. Li, Z. Peng, J. Inorg. Organomet.
Polym., 21, 777, (2011).
17. S.H. Sandstrom, T.J. Segatta, J.J.A. Verthe, US Patent 5396940, to The Goodyear
Tire & Rubber Company, March 14, 1995.
18. S. Schaal, A.Y. Coran, S.K. Mowdood, US Patent 6482884 B1, to Pirelli Pneumatici
S.p.A., November 19, 2002.
19. C.J. Lin, W.L. Hergenrother, US Patent 6845797 B2, to Bridgestone Corporation,
January 25, 2005.
20. C. Kantala, E. Wimolmala, C. Sirisinha, N. Sombatsompop, Polym. Adv. Technol.,
20, 448 (2009).
21. K.M. George, J.K. Varkey, K.T. Thomas, N.M. Mathew, J. Appl. Polym. Sci., 85, 292
(2002).
22. K.M. George, J.K. Varkey, B. George, S. Joseph, K.T. Thomas, N.M. Mathew,
Kautsch. Gummi Kunstst., 59, 544 (2006).
23. P.L. Teh, Z.A. Mohd Ishak, A.S. Hashim, J. Karger-Kocsis, U.S. Ishiaku, J. Appl.
Polym. Sci., 94, 2438 (2004).
24. P.L. Teh, Z.A. Mohd Ishak, A.S. Hashim, J. Karger-Kocsis, U.S. Ishiaku, Eur. Polym.
J., 40, 2513 (2004).
25. C. Nakason, A. Kaesaman, T. Wongkul, S. Kiatkamjornwong, Plast. Rubb. Compos.,
30, 154 (2001).
26. I.R. Gelling, Rubber Chem. Technol., 58, 86 (1985).
27. L. Guy, S. Daudey, P. Cochet, Y. Bomal, Kautsch. Gummi Kunstst., 62, 383 (2009).
28. S. Mihara, R.N. Datta, J.W.M. Noordermeer, Rubber Chem. Technol., 82, 524
(2009).

71
Chapter 3

29. J.A. Ayala, W.M. Hess, G.A. Joyce, F.D. Kistler, Rubber Chem. Technol., 66, 772
(1993).
30. S. Wolff, M.J. Wang, E.H. Tan, Rubber Chem. Technol., 66, 163 (1993).
31. S.C. Ng, K.K. Chee, Rubber Chem. Technol., 62, 585 (1989).
32. T.L.A.C. Rocha, C. Rosca, J. Ziegler, R.H. Schuster, Kautsch. Gummi Kunstst., 58,
22 (2005).
33. A.K. Manna, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer, J. Appl. Polym. Sci., 74,
389 (1999).
34. H.-D. Luginsland, J. Frohlich, A. Wehmeier, Rubber Chem. Technol., 75, 563 (2002).
35. G.A. Böhm, W. Tomaszewski, W. Cole, T. Hogan, Polymer, 51, 2057 (2010).
36. A.J. Manna, P.P De, D.K. Tripathy, J. Appl. Polym. Sci., 84, 2171 (2002).
37. S. Bandyopadhyay, P.P. De, D.K. Tripathy, S.K. De, Rubber Chem. Technol., 69,
637 (1996).
38. J.W. ten Brinke, S.C. Debnath, L.A.E.M. Reuvekamp, J.W.M. Noordermeer,
Compos. Sci. Technol., 63, 1165 (2003).
39. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 85, 277 (2012).
40. J.T. Sakdapipanich, J. Biosci. Bioeng., 103, 287 (2007).
41. S.S. Sarkawi, W.K. Dierkes, J.W.M. Noordermeer, Eur. Polym. J., 49, 3199 (2013).
42. S. Wolff, M.J. Wang, E.H. Tan, Kautsch. Gummi Kunstst., 47, 873 (1994).
43. M.M. Jayasuriya, D.J. Hourston, J. Appl. Polym. Sci., 112, 3217 (2009).
44. R. Rajasekar, K. Pal, G. Heinrich, A. Das, C.K. Das, Mater. Design, 30, 3839 (2009).
45. S. Maiti, S.K. De, A.K. Bhowmick, Rubber Chem. Technol., 65, 293 (1992).
46. P. Saramolee, K. Sahakaro, N. Lopattananon, W.K. Dierkes, J.W.M. Noordermeer,
Rubber Chem. Technol., in press (2014). doi: http://dx.doi.org/10.5254/rct.13.86970.
47. C.G. Robertson, C.M. Roland, Rubber Chem. Technol., 81, 506 (2008).
48. Y. Li, M.J. Wang, T. Zhang, Rubber Chem. Technol., 67, 693 (1994).
49. M.K. Kralevich, J.L. Koenig, Rubber Chem. Technol., 71, 300 (1997).
50. A. Bertora, M. Castellano, E. Marsano, M. Alessi, L. Conzatti, P. Stagnaro, G.
Colucci, A. Priola, A. Turturro, Macromol. Mater. Eng., 296, 455 (2011).
51. L.T. Zhuravlev, Colloids Surf. A: Physicochem. Eng. Aspects, 173, 1 (2000).

72
 CHAPTER 4

Silica-Reinforced Natural Rubber Compounds Compatibilized

by ENR in combination with TESPT and Sulfur Compensation

In the previous study as described in Chapter 3, the use of ENR containing 51

mol% epoxide groups (ENR-51) as a compatibilizer without bis-(triethoxysilylpropyl)
tetrasulfide (TESPT) coupling agent was optimized at 7.5 phr, giving properties which
were still somewhat interior to the compound with TESPT. In this chapter, ENR and
TESPT are cooperatively used to improve the properties of silica-filled NR. By using 7.5
phr of ENR-51 with varying amounts of TESPT in a range of 2 to 5 wt% relative to the
silica, the properties of compounds are compared to those with optimum TESPT content,
i.e. 8.6 wt% relative to the silica, and without TESPT. The addition of TESPT to the ENR-
51 compatibilized silica-filled NR compound has no effect on Mooney viscosity but lowers
the Payne effect to the same level as that of the silica/TESPT compound. It significantly
decreases both scorch and optimum cure times, and increases the tensile strength to
match the property of the optimized silica/TESPT system while maintains the elongation at
break. Furthermore, the tan δ at 60oC is lowered by the addition of TESPT and extra sulfur
when compared to the use of ENR-51 alone as compatibilizer, while the tan δ at 0oC is
maintained. Silica dispersion in the NR matrix and increased interactions within the
material by use of the compatibilizer are evidenced by SEM analysis. This work
demonstrates that the use of ENR as compatibilizer already enhances the properties of
silica-filled NR compounds, and that such properties can further be improved by adding
TESPT at half or lower amount of TESPT than normally needed for silica-filled
compounds, and can be further improved by sulfur compensation relative to the reference
compound with TESPT alone. The combination of ENR, TESPT and sulfur compensation
presents itself as a better option than TESPT alone, for tires where skid-resistance is of
prime importance; e.g. winter tires.

Part of this chapter was published in Advanced Materials Research, 844, 272(2014).
Chapter 4

4.1 Introduction

Silica-filled compounds offer a better wet traction and lower rolling resistance for
passenger tire tread compounds, compared to carbon black-filled counterparts.[1,2]
However, silica mixing requires longer mixing time and higher mixing temperatures than
carbon black. Moreover, silica-filled hydrocarbon rubber compounds without any coupling
agent as compatibilizer suffer from poor silica dispersion due to strong filler-filler
interactions and thus result in inferior mechanical properties. To enhance the compatibility
between silica and non-polar rubbers, a silane coupling agent or polar rubber such as
acrylonitrile-butadiene rubber (NBR)[3], polychloroprene rubber (CR)[4] and epoxidized
natural rubber (ENR)[5] can be used as a compatibilizer. The most widely used silane
coupling agent in silica-filled compounds cured with a sulfur system is bis-
(triethoxysilylpropyl)tetrasulfide (TESPT). It has long been known that this sulfur-
containing silane coupling agent also acts as a sulfur donor to promote crosslinking in the
compound. Under optimal mixing conditions, the use of TESPT effectively reduces filler-
filler interaction and enhances silica-rubber interaction.[6,7] The silica/silane combination
leads to a good silica dispersion and significant improvement in mechanical and
dynamical properties due to the rubber-silica bonds bridged by TESPT.[8]
Epoxidized natural rubber shows a higher polarity than original natural rubber
(NR) due to the epoxy groups in its structure. Its viscosity and polarity increases with
increasing epoxy content.[9] ENR has been used as a compatibilizer in filled systems such
[10,11]
as NR-organoclay nanocomposites in which the use of ENR improved filler-rubber
interaction and filler dispersion in the matrix. The improved compatibility between filler and
rubber subsequently improved physical and mechanical properties of the
[10,11]
nanocomposites. Silica-filled ENR shows significantly improved mechanical
properties when compared to a silica-filled NR counterpart without silane, because of an
interaction of ENR and silica through hydrogen bonds.[12] In addition, chemical bonding
between epoxy groups of ENR and silanol groups of silica can be promoted during
molding at high temperatures[3] or mixing at 150-170oC.[13] At high temperature, epoxy-
groups of ENR could break down to become hydroxyl groups and to bond to the silica
surface through Si-O-C bonds.[5,13]
As demonstrated in Chapter 3, the addition of ENR as compatibilizer in silica-
filled NR compounds enhanced both processing and vulcanizate properties. The Payne
effect was reduced and filler-rubber interaction was promoted leading to the improvement
of mechanical properties. However, the properties were still lower compared to the
reference compound that employed TESPT as coupling agent. One of the reasons can be

74
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

attributed to the influence of network formation within the material. The TESPT containing
compounds obtain a contribution from sulfur that is released from TESPT and so have a
higher network density. A cooperative effect of epoxide groups in ENR and TESPT on
reinforcing efficiency of silica-filled ENR compounds was recently reported.[14] The use of
epoxidized low molecular weight natural rubber (ELMWNR) as compatibilizer in silica-filled
NR compounds reduced the Payne effect and improved the mechanical properties when
compared to an uncompatibilized system.[15] In addition, the effect of extra sulfur added
into the silica-filled NR compounds compatibilized with ELMWNR was reported to
enhance modulus and tensile strength as well as to reduce the loss tangent at 60C.[16]
Herein, the amount of extra sulfur was calculated to compensate for the sulfur in the
TESPT molecules.
In Chapter 3, the use of 7.5 phr of ENR-51 as compatibilizer showed the best
overall properties of the silica-filled NR compounds, but still inferior to those of the
reference compound with TESPT. This chapter is therefore to investigate the use of 7.5
phr of ENR-51 in combination with TESPT at varying amounts from 2 to 5 wt% relative to
silica, without and with sulfur compensation. The compensation is applied by adding
elemental sulfur into the compound to adjust the total sulfur content, equal to the sulfur
contained in the reference compound with 4.7 phr of TESPT.

4.2 Experimental

4.2.1 Materials and compound preparation

In this chapter, the materials used for compounding are as described in Chapter
3. The compound recipes are shown in Table 4.1, and the same two-step mixing
procedure, as described in Table 3.2, is employed for compound preparation. The first
mixing step was carried out using an internal mixer (Brabender Plasticorder 350s) with an
initial mixer temperature setting of 110oC, rotor speed of 60 rpm to obtain the optimized
dump temperature in the range of 135-150C.[6] The second step of curative mixing was
carried out using a two-roll mill for 5 mins. All elemental sulfur was added in this second
mixing step.

75
Chapter 4

Table 4.1 Compound formulations.

Parts per hundred parts of rubber (phr)
Ingredients
None TESPT ENR+TESPT ENR+TESPT+S
RSS3, Natural Rubber 100.0 100.0 92.5 92.5
ENR-51 - - 7.5 7.5
TESPT - 4.7* 1.1-2.7** 1.1-2.7
Zeosil 1165MP 55.0 55.0 55.0 55.0
TDAE oil 8.0 8.0 8.0 8.0
ZnO 3.0 3.0 3.0 3.0
TMQ 1.0 1.0 1.0 1.0
Stearic acid 1.0 1.0 1.0 1.0
DPG 1.0 1.0 1.0 1.0
CBS 1.5 1.5 1.5 1.5
Sulfur 1.5 1.5 1.5 2.3-1.9***
Remarks: * TESPT 4.7 phr equal to 8.6 wt% relative to silica; ** TESPT 1.1-2.7 phr are equal to 2-5
wt% relative to silica; *** sulfur contents were adjusted to compensate for the sulfur in the TESPT
molecules by taking the compound with 4.7 phr of TESPT as reference.

4.2.2 Testing of compound and vulcanizate properties

Mooney viscosity, Payne effect, flocculation rate constant, bound rubber content,
cure characteristics, tensile and dynamic mechanical properties of the silica-filled NR
compounds were investigated following the methods and using the equipments as
described in Chapter 3. For dynamic mechanical properties, the DMA (Metravib
Viscoanalyzer VA 2000, Metravib R.D.S.) was operated in tension mode under the same
conditions as applied in Chapter 3.

4.2.3 SEM analysis of silica-filled NR vulcanizates

SEM analysis was performed by using two different sets of specimens. First,
compounds containing 30 phr of silica, without compatibilizer, with ENR-51 7.5 phr, with
ENR-51 7.5 phr and TESPT 4 wt% relative to silica, and with TESPT 8.6 wt% relative to
silica, were prepared using the same formulation as shown in Table 4.1, except the lower
amount of silica. The compounds were vulcanized to their optimum cure times to produce
a thin sheet of 1-2 mm thickness, immersed in liquid nitrogen, and cryogenically cracked.
The newly cracked surface was gold-coated before being analyzed by SEM (Quanta 400,
FEI). The second set of vulcanized specimens was taken from tensile-fractured test
pieces which consisted of 55 phr of silica, i.e. at normal silica content used in this study.
The tensile-fractured surface was also coated with gold prior to being analyzed.

76
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

4.3 Results and discussion

4.3.1 Mooney viscosity, Payne effect and flocculation tendency

Figure 4.1 shows the Mooney viscosities of the silica-reinforced NR compounds

with ENR-51 as compatibilizer in combination with varying contents of TESPT, with and
without extra sulfur, in comparison with that of the compounds with no compatibilizer at all
and with only 4.7 phr of TESPT as reference.

70

60

50
ML1+4 (100oC)

40

30

ENR+TESPT
20
ENR+TESPT+S

10 TESPT

None
0
0 2 4 6 8
TESPT contents
(wt% rel. to silica)

Figure 4.1 Mooney viscosity of silica-filled NR compounds with 7.5 phr of ENR-51 as
compatibilizer in combination with TESPT with and without sulfur compensation.

Without extra sulfur, the incorporation of 2 wt% TESPT relative to silica into the
silica-reinforced NR compounds leads to a decrease of Mooney viscosity, suggesting a
further improvement of silica dispersion in the NR matrix. But a further increase of TESPT
contents shows no additional effect on this property. The compounds with extra sulfur
show no change in Mooney viscosity with increasing silane content and the values are
marginally higher than that of the silica-filled NR compounds with no extra sulfur. All of the
compounds with ENR-51/TESPT combinations, either with or without extra sulfur, show
lower Mooney viscosities when compared to the reference compound with 4.7 phr TESPT.
The lower Mooney viscosity may be attributed to a better filler distribution causing smaller
silica aggregates and less resistance to flow. As the sulfur was added in the second step

77
Chapter 4

of mixing on a two-roll mill, a premature crosslinking reaction is unlikely to take place like
in the case of addition of extra sulfur into the internal mixer at high temperature.[7]
Nevertheless, the compounds with TESPT and extra sulfur show a slightly higher Mooney
viscosity compared to the mixes with lower sulfur content. It should be noted that, in the
absence of TESPT, the ENR-compatibilized compound with higher sulfur content shows a
lower Mooney viscosity.

500 0.35
(a) (b)
450
Payne effect [G′(0.56%)-G′(100%)]

ENR+TESPT 0.30

Flocculation rate conatant (k a)

400 ENR+TESPT+S

350 TESPT 0.25

None
300
0.20
250
0.15
200
150 0.10 ENR+TESPT

100 ENR+TESPT+S
0.05 TESPT
50 None
0 0.00
0 2 4 6 8 0 2 4 6 8
TESPT contents TESPT contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 4.2 Payne effect (a) and flocculation rate constant (b) of silica-filled NR compounds
with 7.5 phr of ENR-51 as compatibilizer in combination with TESPT with and without
sulfur compensation.

The filler-filler interaction or Payne effect of the silica-filled NR compounds is

shown in Figure 4.2 (a). The addition of 7.5 phr of ENR-51 already remarkably decreases
the filler-filler interactions in the silica-filled NR compounds when compared to the non-
compatibilized one. This is due to interaction between silanol groups on the silica surface
and epoxy groups of ENR, as previously described in Chapter 3. The use of extra sulfur in
the silica-filled NR compounds with ENR-51 as compatibilizer results in a somewhat
higher Payne effect compared to the compound with normal sulfur content. This may be
due to their lower Mooney viscosities: Figure 4.1, that cause a tendency of silica re-
agglomeration, i.e. flocculation, which is in agreement with the flocculation rate constant in
Figure 4.2 (b). The flocculation process is the reformation of silica-silica interaction in the
compound under heat treatment to regenerate a too rigid silica network.[17] The

78
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

flocculation behavior of silica-filled NR compounds containing ENR-51 as compatiblilizer in

combination with TESPT and extra sulfur shows the same trend as the Payne effect.
The Payne effect: Figure 4.2 (a) and flocculation rate constant: Figure 4.2 (b) of
the silica-filled NR compounds with 2 wt% TESPT relative to silica and extra sulfur
decrease to the same level as that of the reference compound with TESPT only. This
further reduction of filler-filler interaction by the addition of TESPT suggests that even
though epoxy-groups of ENR interact with the silanol-groups on the silica surfaces, there
are still free silanol-groups present in the system. So, the use of just small amounts of
TESPT can further decrease the filler-filler interactions through the silanization reaction of
ethoxy-groups and silanol-groups as shown in Scheme 4.1. Without TESPT, the
compound with extra sulfur which has a lower Mooney viscosity shows a higher Payne
effect and higher flocculation rate constant. But in the presence of both TESPT and extra
sulfur, the compounds that have higher Mooney viscosities exhibit lower Payne effects as
well as lower flocculation rate constants, and the values are similar to that of the reference
silica/TESPT system. The results suggest an influence of Mooney viscosity on flocculation
rate constant which in turn relates to the Payne effect.

H3C OH H
5
C C
Si O CH
O Si
1 H O OH C CH3
H
H 4
H3C O O
H
C C C C

O H
2
H 3 OEt
O Si
Si O Si CH2 S S
3
OEt

Scheme 4.1 Various possible interactions and bonding in the silica-filled NR compounds
compatibilizied with ENR in combination with TESPT.

There are several possible interactions and/ or reactions within the system
including: (1) interaction between hydroxyl-groups of ring-opened ENR; (2) interaction
between hydroxyl-group of ring-opened ENR with silanol group of silica[18]; (3) bonding

79
Chapter 4

between silica and rubber molecules via silane molecules; (4) interaction between epoxy
and silanol groups; and (5) bonding between silanol and hydroxyl groups, as shown and
numbered in Scheme 4.1. The interactions/reactions between epoxy- or hydroxyl-groups
of the ENR and silanol groups of the silica contribute to the decreased silica-silica
interactions in the system.

4.3.2 Filler-rubber interactions

Silica particles and aggregates commonly form a filler-filler network to their

neighbors due to the polar silanol groups on the surface[19] causing a poor filler-rubber
interaction which is a key parameter for reinforcement. Bound rubber content can
[20]
optionally be used as an indication for filler-rubber interactions in the silica-reinforced
rubber compounds. The results of silica-filled NR compounds compatibilized with ENRs
and TESPT are shown in Figure 4.3. The test was performed using the non-productive
compounds, i.e. the compounds without curatives, collected after the first mixing step.
Ammonia treatment was employed during the bound rubber test to destroy weak physical
[20]
interactions, so that the chemically bound rubber content could be determined. The
silica-filled compound without compatibilizer shows the lowest chemically bound rubber
content because of high filler-filler interactions through hydrogen bonding between silica
aggregates and the poor interactions between silica and rubber due to their difference in
polarity. The reference silica-filled NR compound with 4.7 phr TESPT gives the highest
chemically bound rubber content which is distinctively higher than the value for the non-
compatibilized compound, due to a good interfacial interaction via the silanization reaction
of silanol-groups of silica and the ethoxy-groups of the silane. The ENR-compatibilized
silica-filled NR compounds also show an improvement in bound rubber content compared
to the compound without compatibilizer, due to the interactions/reactions between epoxy-
and silanol-groups. The chemically bound rubber contents increase slightly with the
addition of TESPT and with increasing TESPT contents, as a result of an increase of
chemical bonds between silica and the rubber matrix through the silane bridge.[21]
However, the level of chemically bound rubber does not reach the value observed for 4.7
phr TESPT.

80
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

80

Chemically bound rubber contents (%)

70

60

50

40

30

20 ENR+TESPT
TESPT
10 None

0
0 2 4 6 8
TESPT contents
(wt% of silica)

Figure 4.3 Chemically bound rubber content of silica-filled NR masterbatches with 7.5 phr
of ENR-51 as compatibilizer in combination with TESPT.

4.3.3 Cure characteristics

16 16
(a) (b)
14 14

12 12
Torque (dN.m)
Torque (dN.m)

10 10

8 8
ENR+S
ENR
6 6 TESPT 2 wt%+S
TESPT 2 wt%
TESPT 3 wt%+S
ENR+TESPT 3 wt%
4 4 TESPT 4 wt%+S
ENR+TESPT 4 wt%
ENR+TESPT 5 wt% TESPT 5 wt%+S
2 2 TESPT
TESPT
None None
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (min) Time (min)

Figure 4.4 Cure curves of silica-filled NR compounds with 7.5 phr of ENR-51 as
compatibilizer and varying contents of TESPT without sulfur compensation (a) and with
sulfur compensation (b).

Cure curves of the silica-filled compounds with different types of compatibilizer

are shown in Figures 4.4(a) and (b) for the compounds without and with extra sulfur,

81
Chapter 4

respectively. The silica-filled NR compound without any compatibilizers shows a clear

increase of rheometer torque at the beginning of the cure curve, prior to the vulcanization
reaction, which can be attributed to flocculation of silica aggregates that is accelerated by
heat, resulting in the highest minimum and maximum torques as seen in Figure 4.4(a).
The use of the optimal TESPT content of 4.7 phr diminishes the flocculation phenomenon
in the silica-filled compound due to strong interactions between silica and rubber via the
silane so that the silica surface become less polar. The addition of only ENR-51 at 7.5 phr
as compatibilizer significantly decreases filler flocculation, but the scorch and cure times
are largely prolonged due to the presence of highly polar epoxy groups that may interfere
with the curatives and furthermore, some residual acid in the ENR may retard the cure.
This compound also shows the lowest maximum torque indicating the least overall
network crosslink density.
Silica has a hydrophilic character[22] and can easily form a strong interaction with
polar functional groups or polar rubbers like ENR.[5] The addition of ENR-51 to a silica-
filled NR compound introduces interactions between the epoxy-groups of the ENR and
silanol-groups of the silica to some extent, but there will remain some free polar functional
groups that may interact with polar ingredients generated during the vulcanization
reaction, such as sulfurating intermediates, and result in longer scorch and cure
times.[23,24] The use of TESPT in combination with ENR-51 for compatibilizing the silica-
filled NR compounds effectively decreases silica flocculation and the cure behavior is
improved with increasing TESPT concentrations due to a higher degree of silica
hydrophobation and increase of silica-rubber interaction.[25] However, the maximum cure
torques and cure rate indices are still lower when compared to the use of the optimal
content of TESPT of 4.7 phr.
The addition of extra sulfur into the compounds definitely has a positive effect on
cure behavior due to the presence of a higher content of curatives. The compound that is
compatibilized by ENR alone in the presence of extra sulfur shows still lower flocculation
compared with the non-compatibilized one, but a remarkably higher flocculation tendency
than the compound with TESPT, as seen in Figure 4.4(b). The addition of extra sulfur into
the ENR/TESPT-compatibilized silica-filled compounds increases the maximum cure
torques of the compounds compared to the results in Figure 4.4(a) due to the increase of
crosslink density in the vulcanizates. The flocculation phenomenon of the ENR-
compatibilized compound with extra sulfur is in accordance with the higher Payne effect:
Figure 4.2 (a) and higher flocculation rate constant: Figure 4.2 (b). The presence of ENR-
51 alone is apparently not sufficient to interact with the silanol groups of the silica, to
suppress the flocculation and enhance the cure characteristics. The use of TESPT in

82
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

smaller amounts than in the referent formulation can overcome the problems. In the case
of the compounds with ENR-51 and extra sulfur, due to a reduction of compound viscosity
when sulfur was additionally added (Figure 4.1), an easier condition for filler flocculation
was generated as the silica aggregates could migrate faster in the lower viscosity matrix.

12 20
(a) (b)
18
10
16

Optimum cure time (min)

14
Scorch time (min)

8
12
6 10
8
4
ENR+TESPT 6 ENR+TESPT
ENR+TESPT+S 4 ENR+TESPT+S
2
TESPT TESPT
2
None None
0 0
0 2 4 6 8 0 2 4 6 8
TESPT contents TESPT contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 4.5 Scorch and cure times of silica-filled NR compounds with ENR-51 at 7.5 phr, in
combination with TESPT silane coupling agent as compatibilizers, without and with sulfur
compensation.

Scorch (ts1) and optimum cure times (tc(90)) of the silica-filled NR compounds
with 7.5 phr of ENR-51 in combination with TESPT without and with extra sulfur are shown
in Figures 4.5(a) and 4.5(b), respectively. These figures help to give a clear picture of the
changes in scorch and cure times between the various compounds, according to the cure
curves as shown in Figures 4.4(a) and 4.4(b). The use of ENR-51 alone as compatibilizer
in the silica-filled NR compound shows the longest ts1 and tc(90) due to the remaining free
silanol groups and polar epoxide groups that can adsorb polar curatives and so retard the
vulcanization reaction as described previously. Residual acid left in the ENR may also
take part in the cure retardation. The incorporation of TESPT in the silica-filled compound
helps to reduce the remaining polar moieties through the silanization reaction. In addition,
TESPT acts as a sulfur donor for the vulcanization reaction [7], leading to a faster cure time
compared to the silica-filled NR with ENR alone as compatibilizer. The addition of extra
sulfur in the systems results in shorter scorch times with increasing TESPT

83
Chapter 4

concentrations, compared to the compounds with normal sulfur content, due to the
presence of a higher concentration of curatives that can generate crosslinks. But the
compounds with different sulfur contents show mutually more or less the same optimum
cure times, which decrease with increasing TESPT loadings. Interestingly, the non-
compatibilized compound show similar scorch and cure times to that of the silica/TESPT
system despite the polar and acidic silanol groups. This might be attributed to the
occurrence of the filler-filler network, as seen by its highest Payne effect and flocculation
rate constant in Figure 4.2, that results in a smaller surface area and less available silanol
groups. On the side, the TESPT-system has a good filler dispersion, the aggregate size is
smaller, providing more surface area and free silanol groups to pay a role in the
vulcanization reaction. With this competitive balance, the compounds with no
compatibilizer and with optimal TESPT content show equal optimum cure times as shown
in Figure 4.5(b).

4.3.4 Tensile properties

16 6
(a) (b)
Reinforcement index (M300/M100)

14
5
12
300% Modulus (MPa)

4
10

8 3

6
2 ENR+TESPT
ENR+TESPT
4 ENR+TESPT+S
ENR+TESPT+S
TESPT
1 TESPT
2
None None

0 0
0 2 4 6 8 0 2 4 6 8
TESPT contents TESPT contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 4.6 300% Modulus and reinforcement index (M300/M100) of silica-filled NR

compounds with 7.5 phr of ENR-51 in combination with TESPT without and with sulfur
compensation.

The incorporation of TESPT without sulfur compensation clearly enhances the

modulus at 300% elongation but the values of the reinforcement index of the silica-filled

84
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

NR vulcanizates do not change much, as respectively shown in Figures 4.6(a) and (b).
This is due to an increase of silica-rubber interactions/reactions in the system. The
introduction of more sulfur to match the level of sulfur available in the reference
TESPT/silica system significantly increases the tensile modulus of the vulcanizates
because of an increased amount of crosslinking agent and consequently a higher
crosslink density and more stiffness. With ENR/TESPT and extra sulfur, the vulcanizates
still show a lower 300% modulus when compared to the compound with the optimal
TESPT content of 4.7 phr. The reinforcement index (M300/M100) of the silica-filled NR
vulcanizates is improved when TESPT and extra sulfur are added, however an increasing
amount of TESPT has no clear effect on the reinforcement index of the vulcanizates. All of
the compatibilized compounds show a remarkable improvement in modulus compared
with the one without, but with little dependence on the TESPT-content.

35 800
(a) (b)
700
30
600
Elongation at break (%)
Tensile strength (MPa)

25
500
20
400
15
300
ENR+TESPT ENR+TESPT
10
ENR+TESPT+S 200 ENR+TESPT+S

TESPT TESPT
5 100
None None
0 0
0 2 4 6 8 0 2 4 6 8
TESPT contents TESPT contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 4.7 Tensile strength (a) and elongation at break (b) of silica-filled NR compounds
with 7.5 phr of ENR-51 in combination with TESPT without and with sulfur compensation.

The tensile strength: Figure 4.7(a), of the vulcanizates is improved by adding

TESPT on top of the ENR-51 as compatibilizer, and further enhanced by the addition of
extra sulfur. The use of TESPT at 4 wt% relative to the silica with ENR-51 and extra sulfur
gives the same level of tensile strength as that of the silica/TESPT filled reference
compound, while the elongation at break remains more or less unchanged: Figure 4.7(b).
In combination with the epoxy-groups that can interact and/or react with the silanol-groups

85
Chapter 4

of silica, the amount of TESPT needed to optimize the silica-filled compound properties
can be reduced, compared to a conventional TESPT-silanized silica filled system. The
different levels of modulus and tensile strength caused by the difference in sulfur contents
because of the sulfur donation of TESPT, can be improved by introducing more crosslinks
generated by extra elemental sulfur added.

4.3.5 Dynamic mechanical properties

Silica-filled NR-vulcanizates with various types of compatibilizers have different

dynamic mechanical properties as shown in Figures 4.8-4.9. These properties are related
to the viscoelastic behavior of the materials, which depend on several factors including
polymer characteristics, reinforcing filler types and contents, process oil, crosslink density,
etc. In tire technology, the viscoelastic properties of the rubbers are intrinsic
characteristics that are related to tire performance, i.e. wet traction and rolling
resistance.[26]

10000 10000
(a) (b)
None None
TESPT TESPT
ENR 1000
ENR
1000 ENR+TESPT ENR+TESPT
E″ (MPa)
E′ (MPa)

ENR+TESPT+S ENR+TESPT+S

100

100
10

10 1
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Temperature (oC) Temperature (oC)

Figure 4.8 Storage modulus (a) and loss modulus (b) of silica-filled NR vulcanizates
containing different compatibilizer types and combinations.

The storage modulus, loss modulus and tan  as functions of temperature of the
silica-filled NR vulcanizates are shown in Figures 4.8(a), 4.8(b) and 4.9. In these figures,
the results of the vulcanizates with 8.6 wt% TESPT relative to silica, with 7.5 phr ENR-51,
with 7.5 phr ENR-51 plus 4 wt% TESPT relative to silica, and with 7.5 phr ENR-51 plus 4

86
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

wt% TESPT relative to silica plus extra sulfur, as compatibilizers, are comparatively
displayed.

1.0
None
TESPT

0.8 ENR
ENR+TESPT
ENR+TESPT+S
Tan δ 0.6

0.4

0.2

0.0
-80 -60 -40 -20 0 20 40 60 80
Temperature (oC)

Figure 4.9 Tan δ of silica-filled NR vulcanizates containing different compatibilizer types

and combinations.

Figures 4.8(a) and 4.8(b) clearly show that the addition of the different types of
compatibilizers into the silica-filled NR vulcanizates has little effect on the storage (E′) and
loss (Eʺ) moduli in the glassy state and in the glass-to-rubber softening region, but
significantly affects the material properties in the rubbery region. The different extent of
interactions between the various components, the presence of ENR blended with NR and
filler dispersion all play a role in the ability of the rubber chains to move and respond to
deformation. After passing through the glass transition, the vulcanizates that contain ENR-
51 as compatibilizer, either without or with TESPT, show higher storage and loss moduli
than the ones with TESPT and without any compatibilizer. The presence of 7.5 phr ENR-
51 with high epoxide-content and a glass transition temperature of  0oC causes a certain
chain stiffness. Therefore, the NR with ENR-51 as compatibilizer has a higher elastic
modulus, and in the mean time also shows a higher viscous or loss moludus because of
less chain flexibility, as seen in Figure 4.8.
The use of the optimal content of 4.7 phr TESPT gives vulcanizates with the
highest tan δ peak due to improvement of silica dispersion, leading to less trapped rubber
in the silica aggregates and so more rubber segments to respond to the dynamic

87
Chapter 4

deformation in this region. On the other hand, the use of ENR as compatibilizer results in
a decrease of the tan δ peak intensity, even in combination with TESPT. This can partially
be caused by the presence of the ENR which at this temperature is still in the glassy state,
and may also result from interactions between the ENR and silica and self-crosslinked
ENR that restrict motions of the rubber segments. In the rubbery region, the silica/TESPT
system provides the lowest tan δ value due to the good chemical bonding between silica
and rubber via silane bridges which help to reduce energy loss under dynamic conditions.
The higher network density, as indicated by the rheometer maximum torque in Figures
4.4(a) and 4.4(b), enhances the elasticity further, resulting in lower loss tangent as well.
The presence of ENR-51 that possesses higher damping properties results in the higher
loss tangent in the temperature range of 0 to 80C above its glass transition temperature
of around 0oC. The compound without any compatibilizer also shows an increasing trend
of the loss tangent with rise of the test temperature.
The glass transition temperatures (Tg) is the viscoelastic glass-to-rubber
softening transition which is an important property of amorphous and semi-crystalline
polymers. By dynamic mechanical analysis, the Tg of polymers can be determined by
using the peak temperatures of either tan  or loss modulus (E). Which point is to be
used to at best represent Tg is still a subject of debate. Robertson et al.[27] discussed that
tan  in the glass-to-rubber softening transition is determined not only by the local
segmental motions of the polymer, as reflected in the loss modulus towards lower
temperature, but also by the filler reinforcement effect on both the storage and loss moduli
at higher temperature. They reported that the loss modulus peak which corresponds to the
segmental relaxation process is not significantly affected by the particle surface area of
[27]
the filler and degree of filler-polymer interactions , whereas the shape and magnitude of
the tan  peak are influenced by these parameters associated with the nature of the filler.
The present work reports the Tg’s of the silica-filled NR vulcanizates containing
none and different contents of TESPT, in combination with 7.5 phr of ENR-51, with and
without sulfur compensation, as determined by using the peaks of both tan  and loss
modulus, as shown in Figure 4.10. The compounds with different degrees of filler-rubber
interactions, as indicated by their bound rubber contents in Figure 4.3, show the same
shapes and magnitudes of E peak: Figure 4.8(b), but different shapes and magnitudes of
the tan  vs. temperature peaks: Figure 4.10, in accordance with the work of Robertson et
al.[27] The Tg’s of the silica-filled NR vulcanizates from both Eʺ and (tan δ)max show the
same trend, merely shifted over a few degrees, where the Tg of the vulcanizate with
optimal content of 4.7 phr TESPT which has the highest chemical bound rubber content

88
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

as well (Figure 4.3), is the highest. The vulcanizates with extra sulfur that have a higher
crosslink density compared to the ones with normal sulfur content show a slightly higher
Tg and increasing TESPT content in the ENR-compatibilized NR shows an insignificant
effect on Tg.

-48 (Tan δ)max

-50

-52
T g (oC)

Loss modulus (Eʺ)

-54

-56
ENR+TESPT ENR+TESPT
ENR+TESPT+S ENR+TESPT+S
-58
TESPT TESPT
None None
-60
0 2 4 6 8
TESPT contents
(wt% rel. to silica)

Figure 4.10 Glass transition temperatures (Tg) at the peaks of tan  (solid symbols) and
loss moduli (open symbols) of silica-filled NR vulcanizates with different compatibilizer
combinations.

A study on the viscoelastic properties of crosslinked NR by DMA[28] showed that

the glass transition temperature depended somewhate on both the crosslink density and
the crosslink types. A higher crosslink density increased the Tg because of a more
restrictions on molecular chain motions, as is commonly observed. Furthermore, the use
of higher sulfur contents lead to a higher proportion of polysulfidic crosslinks and cyclic
sulfidic structures which had a greater influence on Tg than monosulfidic or carbon-carbon
crosslink types.[28] The chemically bonded rubber to the silica surface via silane bridges
further restricts the movement of polymer chains and increases the Tg. This is in
accordance with the work that reported the shift of Tg towards higher temperature by
polymer-filler coupling bonds.[5]

89
Chapter 4

0.16 0.14
(a) (b)
0.14 0.12

0.12
0.10
Tan δ at 5oC

Tan δ at 60oC
0.10
0.08
0.08
0.06
0.06
ENR+TESPT
0.04 ENR+TESPT
0.04 ENR+TESPT+S ENR+TESPT+S

0.02
TESPT 0.02 TESPT
None None
0.00 0.00
0 2 4 6 8 0 2 4 6 8
TESPT contents TESPT contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 4.11 Tan δ at 5oC (a) and 60oC (b) of silica-filled NR vulcanizates without and with
different compatibilizer combinations.

In the test method for wet grip grading of C1 tires according to UNECE
Regulation R117[29], the wet surface temperature range for normal or summer tires is 5–
35C, and for winter tires 2–20C. Figures 4.11(a) and (b) show tan δ values at 5oC and
60oC for the silica-filled NR-vulcanizates with different compatibilizers. Higher tan δ at low
temperature indicates a better tire wet grip. As shown in Figure 4.11(a), the tan δ at 5oC of
the silica-filled NR vulcanizates is substantially increased with ENR added as
compatibilizer, due to the damping behavior of epoxidized rubber influenced by its Tg as
exhibited in Figure 4.10. The use of TESPT in the range of 2–5 wt% relative to the silica
content in combination with ENR has only little effect on tan δ. The silica-filled NR
vulcanizate with TESPT shows the lowest tan δ at 5oC implying the lowest tire wet skid
resistance. The chemical bonds between silica and rubber via silane molecules give an
additional contribution to the crosslink network that is normally generated by the
vulcanization reaction. This results in a better elastic response and lower energy loss. This
double network also causes the lowest tan δ at 60oC which indicates a low rolling
resistance for tires, as shown in Figure 4.11(b). The presence of ENR-51 leads to a
significantly higher tan δ at 60oC compared to the use of TESPT alone, but is still lower
than for the non-compatibilized compound. The addition of TESPT of just only 2 wt%
relative to silica shows an improvement of tan δ at 60oC due to the improved filler
dispersion and cure behavior. The correction of sulfur-contents in the ENR-compatibilized

90
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

compounds decreases tan δ at both 0oC and 60oC of the vulcanizates due to the increase
of crosslink density, leading to better rubber elasticity.

4.3.6 SEM images of silica-filled NR with ENR, TESPT and extra sulfur as
compatibilizers, with and without sulfur compensation

Filler dispersion level is one of the parameters that affect the properties of filled
compounds and vulcanizates. The silica dispersion in the NR matrix containing different
types of compatibilizers was studied by the SEM technique using two different sets of
specimens, as described in Section 4.2.3. For the cryogenically cracked surfaces as
shown in Figure 4.12, the compounds contain only 30 phr of silica in order to be able to
better observe the silica dispersion. For the regular formulation with 55 phr of silica, tensile
fractured surfaces were used and the SEM micrographs are shown in Figure 4.13.

(a) (b)

(c) (d)

Figure 4.12 SEM micrographs of filled NR vulcanizates with 30 phr of silica at 15000x
magnification: (a) without compatibilizer; (b) with 7.5 phr ENR-51; (c) with 7.5 phr ENR-51
plus 4 wt% TESPT rel. to silica; and (d) with 8.6 wt% TESPT rel. to silica.

Due to a large number of silanol groups on the silica surface, breaking of silica
agglomerates and dispersion of aggregates into the rubber matrix are difficult as a result
of hydrogen bonding. The silica dispersion of the filled NR vulcanizate without
compatibilizer: Figure 4.12(a), is clearly poorer than the vulcanizates that contain

91
Chapter 4

compatibilizers, as the silica is less homogeneously dispersed in the matrix and

aggregates/agglomerate size is larger than what can be observed in the other
compounds. The incorporation of ENR-51 into the silica-filled NR improves the silica
dispersion in the matrix as visually observed in Figure 4.12(b) due to interactions between
the silanol groups of the silica and epoxide groups of ENR leading to hydrophobation of
the silica. The dispersion of silica is further improved by the use of ENR in combination
with TESPT at 4 wt% relative to the silica, as seen in Figure 4.12(c), in which the silica is
finely dispersed to more or less the same level as that of the silica-filled NR vulcanizate
with optimal content of 4.7 phr TESPT: Figure 4.12(d). The SEM results are in
accordance with the Payne effect: Figure 4.2(a), and the resulting vulcanizate properties
in which the non-compatibilized system is the worst. The addition of TESPT to react with
free silanol groups through the silanization reaction to form siloxane bonds and the bound
silane on the silica surface suppress the filler-filler interactions and prevents the silica from
re-agglomeration. The better silica dispersion in NR with TESPT silane coupling agent
compared with vulcanizates without silane was also clearly demonstrated by SEM images
of cut-off surfaces by Choi et al.[30]

(a) (b)

(c) (d) (e)

Figure 4.13 SEM micrographs of tensile fractured surfaces of filled NR vulcanizates with
55 phr of silica at 800x magnification: (a) without compatibilizer; (b) with 8.6 wt% TESPT
rel. to silica; (c) with 7.5 phr ENR-51; (d) with 7.5 phr ENR-51 plus 4 wt% TESPT rel. to
silica; and (e) with 7.5 phr ENR-51, 4 wt% TESPT rel. to silica and extra sulfur.

92
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

SEM micrographs of tensile fractured surfaces of the silica-filled NR vulcanizates

containing different compatibilizers are shown in Figure 4.13. The silica-filled NR
vulcanizate without compatibilizer that has the lowest tensile strength because of the poor
filler dispersion and filler-rubber interactions, shows a smooth surface with some vacuoles
on the failure surface indicating de-wetting of silica by the NR matrix, as shown in Figure
4.13(a). The observation of silica agglomerates by SEM in silica-filled NR without
compatibilizer was previously demonstrated by Pal and De.[31] The effective use of TESPT
silane coupling agent in the silica-filled NR that leads to the highest modulus: Figure
4.6(a), and tensile strength: Figure 4.7(a), gives a rough fractured surface with many tear
lines from ductile failure[32] as a result of good interfacial interactions between rubber and
silica by silane coupling. All of the fractured surfaces of silica-filled NR with ENR-51, ENR-
51 in combination with TESPT and ENR-51 in combination with TESPT and extra sulfur as
compatibilizers as shown in Figure 4.13(c-e) respectively, show a rough surface and tear
lines pattern but at different degrees compared with the silica/TESPT system. The ENR-
compatibilized vulcanizate with improved filler-rubber interactions via the
interaction/reaction between the epoxide groups of ENR and the silanol groups on the
silica surface, has a considerably less complicated fractured surface when compared to
the one with optimal content of TESPT. The silica-filled vulcanizates with TESPT and
ENR/TESPT/extra sulfur show the same level of tensile strength as shown in Figure 4.7(a)
show very similar tensile fractured surfaces, as shown in Figure 4.13(c) and 4.13(e),
respectively. SEM micrographs of tensile fractured surfaces of filled NR vulcanizates have
also been used to demonstrate dispersion and wetting of filler by the matrix, such as in the
work of Arayapranee et al.[32] and Ismail et al.[33] in which the rough surfaces with tear
lines indicated higher tensile strength.

4.4 Conclusions

The use of ENR-51 at 7.5 phr in combination with TESPT silane coupling agent
at 2 wt% relative to silica further decreases Mooney viscosity, Payne effect, flocculation
rate constant and increases bound rubber contents of silica-filled NR compounds, when
compared to a silica-filled compound with ENR-51 alone. A further increase of TESPT-
content has only a small additional effect on the compound properties. Scorch and cure
times of the compounds decrease by increasing TESPT-contents and the addition of extra
sulfur decreases the scorch time further. Tensile strength and modulus of the vulcanizates
increase with increasing TESPT-contents and further improve by sulfur compensation
relative to the reference compound with TESPT. The use of ENR-51 at 7.5 phr with

93
Chapter 4

TESPT at 4 wt% relative to silica and extra sulfur gives vulcanizates with the same levels
of tensile strength compared to the vulcanizates with TESPT at 8.6 wt% relative to silica.
The dynamic mechanical properties of the silica-filled NR vulcanizates are significantly
affected by the presence of ENR-51 that has a higher damping than NR. Above the glass-
to-rubber region, the ENR-containing vulcanizates show higher storage and loss modulus
and higher tan  compared to the silica/TESPT system. The addition of TESPT to the
ENR-compatibilized compound cause only a small change in Tg and tan  at 5C, but
lower tan δ at 60oC. The addition of extra sulfur leads to an increase of Tg and lower tan 
at both 5C and 60oC as a result of increased crosslink density. The SEM micrographs
clearly show good silica dispersion in all compatiblized vulcanizates, and the tensile
fracture surfaces of the materials show different tear line patterns dependent on their
tensile strengths.
Overall, the combination of ENR-51, TESPT and sulfur compensation does not
fully match the properties of a compound with TESPT alone at optimal quantity. However,
if in the perspective of the “Magic Triangle of Tire Technology” the wet skid resistance
needs to be boosted versus rolling resistance, e.g. for “Winter Tires”, then the combination
of ENR-51, TESPT and sulfur compensation presents itself as a better option: which has
to be confirmed in real tire tests.

4.5 References

1. R. Rauline, EP Patent 0501227A1, to Michelin & Cie, February 9, 1992.

2. S. Wolff, Kautsch. Gummi Kunstst., 41, 674 (1988).
3. N. Suzuki, M. Ito, S. Ono, J. Appl. Polym. Sci., 95, 74 (2005).
4. P. Sae-oui, C. Sirisinha, U. Thepsuwan, K. Hatthapanit, Eur. Polym. J., 43, 185
(2007).
5. A.K. Manna, P.P. De, D.K. Tripathy, J. Appl. Polym. Sci., 84, 2171(2002).
6. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 85, 277 (2012).
7. J.W. ten Brinke, S.C. Debnath, L.A.E.M. Reuvekamp, J.W.M. Noordermeer, Comp.
Sci. Technol., 63, 1165 (2003).
8. A.S. Hasham, B. Azahari, Y. Ikeda, S. Kohjiya, Rubber Chem. Technol., 71, 289
(1998).
9. A.D. Roberts, “Natural Rubber Science and Technology”, Oxford University Press,
New York, 1988.

94
 Silica-reinforced NR compatibilized by ENR in combination with TESPT and sulfur compensation

10. M. Arroyo, M.A. López-Manchado, J.L. Valentín, J. Carretero, Comp. Sci. Technol.,
67, 1330 (2007).
11. P.L. Teh, Z.A. Mohd Ishak, A.S. Hashim, J. Karger-Kocsis, U.S. Ishiaku, Eur. Polym.
J., 40, 2513 (2004).
12. F. Cataldo, Macromol. Mater. Eng., 287, 348 (2002).
13. A.K. Manna, A.K. Bhattacharyya, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer,
Polymer, 39, 7113 (1998).
14. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 87, 291 (2014).
15. P. Saramolee, K. Sahakaro, N. Lopattananon, W.K. Dierkes, J.W.M. Noordermeer,
Adv. Mat. Res., 747, 522 (2013).
16. P. Saramolee, K. Sahakaro, N. Lopattananon, W.K. Dierkes, J.W.M. Noordermeer,
Adv. Mat. Res., 844, 235 (2014).
17. H.-D. Luginslands, J. Fröhlich, A. Wehmeier, Rubber Chem. Technol., 75, 563
(2002).
18. J. Zhi-xin, X. Ti-wen, L. Yuan-fang, J. De-min, International Rubber Conference,
Beijing, September 16-18, 2014.
19. M.-J. Wang, S. Woff, J.-B. Donnet, J.-B. Donnet, Rubber Chem. Technol., 64, 559
(1991).
20. S. Wolff, M.-J. Wang, E.-H. Tan, Rubber Chem. Technol., 66, 163 (1993).
21. S.-S. Choi, Polym. Test., 21, 201 (2002).
22. S. Wolff, Rubber Chem. Technol., 69, 325 (1996).
23. P.L. Teh, Z.A. Mohd Ishak, A.S. Hashim, J. Karger-Kocsis, U.S. Ishiaku, J. Appl.
Polym. Sci., 94, 2438 (2004).
24. B.T. Poh, H. Ismail, E.H. Quah, P.L. Chin, J. Appl. Polym. Sci., 81, 47 (2001).
25. C.J. Lin, W.L. Hergenrother, E. Alexanian, G.G.A. Böhm, Rubber Chem. Technol.,
75, 865 (2002).
26. L.J. Murphy, E. Khmelnitskaia, M.J. Wang, K. Mahmud, Rubber Chem. Technol., 71,
1015 (1998).
27. C.G. Robertson, C.J. Lin, M. Rackaitis, C.M. Roland, Macromolecules, 41, 2727
(2008).
28. R. Hagen, L. Salmén, B. Stenberg, J. Polym. Sci. Part B: Polym. Phys., 34, 1997
(1996).
29. UNECE Regulation No 117, GRRF 71st session, September 13-15, 2011.
30. S.-S. Choi, C. Nah, B.-W. Jo, Polym. Int., 52, 1382 (2003).
31. P.K. Pal, S.K. De, Rubber Chem. Technol., 55, 1370 (1982).

95
Chapter 4

32. W. Arayapranee, N. Na-Ranong, G.L. Rempel, J. Appl. Polym. Sci., 98, 34 (2005).
33. H. Ismail, R. Nordin, A.M. Noor, Polym. Test., 21, 565 (2002).

96
 CHAPTER 5

Silica-Reinforced Natural Rubber Compounds Compatibilized

by ENR in combination with Different Silane Coupling
Agent Types

Bis-(triethoxysilylpropyl)tetrasulfide (TESPT), 3-mercaptopropyl-di(tridecan-1-

oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-363) and 3-octanoyl-thio-
propyltriethoxysilane (NXT) are comparatively studied at varying amounts from 2 to 5
wt% relative to the silica, in combination with 7.5 phr of epoxidized natural rubber with 51
mol% of epoxide (ENR-51), as compatibilizers for silica-filled NR compounds. The use of
different types of silane coupling agents together with ENR gives silica-filled NR
compounds with different Mooney viscosities, bound rubber contents, cure properties and
tensile properties. Over the range of silane contents used with ENR, the compounds with
VP Si-363 and NXT show a lower Mooney viscosity, bound rubber content and moduli at
100% and 300% elongation when compared to the one with TESPT. The ENR/NXT
combinations give the best scorch safety. As the presence of ENR-51 at 7.5 phr already
effectively reduces the filler-filler interactions, the addition of silane on top of the ENR has
only little influence on Mooney viscosity and Payne effect, but clearly enhances filler-
rubber interactions as indicated by an increased chemically bound rubber content. The
tensile properties of the silica-filled NR vulcanizates are increased with increasing silane
contents, in which state-of-the-art TESPT still provides better properties than the other
silanes, while VP Si-363 and NXT give more or less the same properties. At a fixed silane
content of 4 wt% relative to silica and 7.5 phr of ENR-51, sulfur compensation added into
the compounds by taking the mix with optimal content of TESPT as reference, enhances
the tensile modulus and strength. The differences in tensile properties of the vulcanizates
containing different compatibilizer combinations are also reflected in their different tear line
patterns on tensile fractured surfaces, as analyzed by the SEM technique. In the presence
of ENR, the dynamic mechanical properties of the silica-filled NR compounds either with
or without silane are influenced by the glass transition temperature of the ENR. With
respect to tire rolling resistance, the use of TESPT still provides the lowest loss tangent at
60C.
Chapter 5

5.1 Introduction

Incorporation of fillers into rubber compounds has an effect on the physical

properties of the rubber vulcanizates that results from the development of stress
concentration, which in turn depends on several factors such as shape of the filler, the
filler surface, particle size, particle size distribution, and nature of the boundary layer
between filler and matrix.[1-3] Silica is a non-black filler which consists of a great number of
silanol groups on its surface, leading to high polarity, acidity and low adhesion between
silica and elastomers.[2] Use of silica as a reinforcing filler in rubber compounds needs
some chemical functional groups or silane coupling agent to increase the reinforcing
efficiency by means of increased silica-rubber interactions and improved silica dispersion.
The use of epoxidized natural rubber (ENR) as compatibilizer in silica-filled NR
compounds, as shown in Chapter 3, demonstrated that it could enhance the reinforcing
efficiency of silica to some extent but the final properties were still somewhat inferior to the
use of state-of-the-art TESPT silane coupling agent. It is known that the properties of ENR
itself depend on the epoxide-group content and that there is a decrease of strain-induced
crystallization at high levels of epoxidation.[4,5] Various studies on silica-filled ENR
compounds[6-8] reported good reinforcement and improved properties compared to silica-
filled NR without silane coupling agent, due to chemical interactions between the polar
epoxide-groups of the ENR and the silanol-groups on the silica surface.[7,8]
The most widely used method to optimize the silica reinforcing efficiency in
rubber compounds is the use of a silane coupling agent. The silane modified silica surface
has a reduced specific surface energy, leading to improvement of the compatibility
between silica and hydrocarbon rubber, better silica dispersion, improved compound
processability and vulcanization properties.[9] Sulfur-containing silane coupling agents
have more influence on the overall properties of silica-filled rubber than a sulfur-free
silane.[10] Bis-(triethoxysilylpropyl)tetrasulfide (TESPT) gives better scorch safety and
lower compound viscosity when compared to γ-mercaptopropyltriehoxysilane (MPS) due
to the reactive thiol end-group of the MPS which has a catalytic effect on accelerated
sulfur vulcanization, whereas TESPT has steric hindrance by bulky ethoxy end-groups.[11]
TESPT with its tetrasulfide moiety is known to donate free sulfur to the compounds, and
the use of a silane with lower sulfur rank or a sulfur-free silane needs correction of the
sulfur-content relative to the sulfur contained in the reference compound with TESPT. It
has been reported that the extra sulfur added either in the first step of mixing in an internal
mixer or in the second step on a two-roll mill can react with the coupling agent to some
extent, leaving a large part in the system to participate in the vulcanization reaction.[10]

98
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

Now silica is constantly gaining importance for energy-saving tires, there is also a
steady development of alternative silanes. 3-Octanoyl-thio-propyltriethoxysilane (NXT) has
been introduced as a silane coupling agent offering a better scorch safety than TESPT,
because of its bulky blocking end-group on the mercapto function.[12-14] The blocked silane
[13,14]
will couple with NR only during the curing period and it has been reported that the
o
NXT-silane needs a mixing temperature above 130 C to react with the silanol groups on
silica. The longer scorch time of NXT compound is in accordance with the higher
activation energy for premature scorch compared to that of TESPT.[13] A study on SBR
compounds filled with silica that were modified by NXT in comparison with TESPT showed
that the NXT compounds showed lower Payne effects and lower viscosities due to better
filler dispersion, but poorer modulus because the silane contains only one sulfur atom in
its structure.[15] Another new silane that was proposed for tire compounds is VP Si-363
which is a bifunctional organosilane having long dialkoxy-ethoxy-silyl groups and a
mercapto function.[16] The “scorchy” nature of the mercapto function of this silane is
suppressed by steric hindrance by the large side-arms.[16] As the silica-silane reaction
proceeds via the ethoxy-group, this silane type produces less ethanol emission during
processing compared to the conventional TESPT or Si-69. A kinetic study by Blume[17]
using a model system showed that the presence of –SH and long alkoxy groups enhances
the rate of reaction. Furthermore, the oxygen atoms inside the long alkoxy chain of VP Si-
363 may increase the possibility of silane adsorption onto the silanol groups of the silica
via hydrogen bonding and shield the free silanol groups.
Based on the results in Chapter 3, that dealt with the effect of epoxidized natural
rubbers (ENRs) as compatibilizers in silica-filled NR, the use of ENR containing 51 mol%
of epoxide (ENR-51) at 7.5 phr gave the best overall improvement in the Payne effect,
bound rubber content, mechanical and dynamic mechanical properties. With the aim to
reduce the use of silane coupling agent and to reduce ethanol emission, this chapter
explores combinations of ENR-51 at 7.5 phr with three different types of silane: TESPT,
VP Si-363 and NXT in silica-filled NR-compounds. The amount of ENR-51 is fixed at 7.5
phr based on the optimum properties as reported in Chapter 3, and the silane contents are
varied from 2.0 to 5.0 wt% relative to the silica amounts. Later on, with fixed amounts of
both ENR-51 and silane, compensations are applied in the compound with elemental
sulfur relative to the total sulfur contained in the reference TESPT-compound, and the
resulting properties are investigated in comparison with the normal formulation.

99
Chapter 5

5.2 Experimental

5.2.1 Materials

All of the compounding ingredients used in this chapter are the same as
described in Chapter 3. Three different types of silane coupling agents were used
including bis-(triethoxysilylpropyl)tetrasulfide (TESPT) (Evonik, Germany), 3-
mercaptopropyl-di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-363)
(Evonik, Germany) and 3-octanoyl-thio-propyltriethoxysilane (NXT) (Momentive, USA).
Their details are given in Table 5.1.

Table 5.1 Silane coupling agent information.

Silane MW
Chemical names Structures
types (g/mol)
CH3CH2O OCH2CH3
Bis-(triethoxysilylpropyl)
TESPT CH3CH2O Si CH2 Sx CH2 Si OCH2CH3 532.5
tetrasulfide CH3CH2O
3 3
OCH2CH3
with x = 3.70

3-Mercaptopropyl-di(tridecan- CH3(CH2)12(OC2H4)5O
VP Si-
1-oxy-13-penta CH3CH2O Si CH2 SH 987.5
363 3
(ethyleneoxide))ethoxysilane CH3(CH2)12(OC2H4)5O

CH3CH2O
O
3-Octanoyl-thio-
NXT CH3CH2O Si CH2 S C CH2 CH3 365
propyltriethoxysilane 3 6
CH3CH2O

5.2.2 Compound formulations and mixing procedures

The first mixing step of silica-filled NR compounds in this chapter was performed
using an internal mixer, Brabender® 50EHT, (Brabender® GmbH & Co.KG, Germany)
with a mixing chamber volume of 70 cm3. The initial mixer temperature setting was 100oC
in order to reach a dump temperature in the range of 135-150oC, as reported by
Kaewsakul et al.[15], that is the temperature range providing optimal overall properties of
the silica/silane-filled NR compounds. The compound formulations are shown in Tables
5.2 and 5.3, and the mixing procedures are described in Table 5.4. The second step of
curative addition was carried out on a two-roll mill.

100
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

Table 5.2 Compound formulations without extra sulfur.

Parts per hundred parts of rubber (phr)
Ingredients
References No extra sulfur
RSS3 100.0 100. 92.5 92.5 92.5 92.5
0
ENR-51 - - 7.5 7.5 7.5 7.5
TESPT - 4.7* - 1.1-2.7** - -
VP Si-363 - - - - 1.1-2.7** -
NXT - - - - - 1.1-2.7**
Zeosil 55.0 55.0 55.0 55.0 55.0 55.0
1165MP
TDAE oil 8.0 8.0 8.0 8.0 8.0 8.0
ZnO 3.0 3.0 3.0 3.0 3.0 3.0
TMQ 1.0 1.0 1.0 1.0 1.0 1.0
Stearic acid 1.0 1.0 1.0 1.0 1.0 1.0
DPG 1.0 1.0 1.0 1.0 1.0 1.0
CBS 1.5 1.5 1.5 1.5 1.5 1.5
Sulfur 1.5 1.5 1.5 1.5 1.5 1.5
Remarks: *TESPT 4.7 phr equals to 8.6 wt% rel. to the silica; **Amounts of silane were varied at 2, 3,
4, 5 wt% rel. to silica, which are respectively equal to 1.1, 1.65, 2.2, 2.7 phr;

Table 5.3 Compound formulations with sulfur compensation.

Parts per hundred parts of rubber
Ingredients (phr)
Extra sulfur*
RSS3 92.5 92.5 92.5 92.5
ENR-51 7.5 7.5 7.5 7.5
TESPT - 2.2 - -
VP Si-363 - - 2.2 -
NXT - - - 2.2
Zeosil 1165MP 55.0 55.0 55.0 55.0
TDAE oil 8.0 8.0 8.0 8.0
ZnO 3.0 3.0 3.0 3.0
TMQ 1.0 1.0 1.0 1.0
Stearic acid 1.0 1.0 1.0 1.0
DPG 1.0 1.0 1.0 1.0
CBS 1.5 1.5 1.5 1.5
Sulfur 2.6 2.1 2.5 2.4
Remarks: *sulfur content was compensated relative to the sulfur contained in the reference
compound with 4.7 phr TESPT.

101
Chapter 5

Table 5.4 Mixing procedures.

Cumulative
Mixing procedures
time (mins.)
Step 1 : Internal mixer
- NR and ENR mastication 2
- Addition of a first half of silica (and ½ of silane) 7
- Addition of a second half of silica( and ½ silane) and TDAE oil 12
- Addition of ZnO, stearic acid, and TMQ 15
Step 2 : Two roll mill
- Addition of DPG, CBS, and sulfur (+ extra sulfur) 5

5.2.3 Mooney viscosity, Payne effect and bound rubber content

Mooney viscosity [ML(1+4), 100oC] was tested by using a Visc Tech+ (Tech-Pro
Inc., USA) according to ASTM D1646. The Payne effect or filler-filler interaction of the final
compounds (with curatives) was analyzed by using a Rubber Process Analyzer
(RPA2000, Alpha Technologies, USA) at 100oC, frequency 0.5 Hz and varying strains
from 0.56% to 100%. Bound rubber content of the silica-filled NR non-productive
compounds, i.e. the compounds without curatives, was analyzed following the procedures
as described in Chapter 3.

5.2.4 Cure characteristics, vulcanization and testing of tensile properties

Cure characteristics of the silica-filled NR compounds were analyzed by using a

Moving Die Rheometer (MDR) (rheoTech MD+, Tech-Pro, Inc., USA) at 150oC for 30
minutes at a frequency of 1.67 Hz and 13.95% strain. The compounds were later cured to
their respective cure times (tc90) with a compression mold (Chaicharoen Karnchang Ltd.,
Thailand). The vulcanized sheets of 2 mm thickness were cut into dumbbell specimens
using die type C, and tensile testing was carried out using a Hounsfield Tensile Tester
(H10KS, Hounsfield Test Equipment, England) at a crosshead speed of 500 mm/min
according to ASTM D412.

5.2.5 Dynamic mechanical properties

Storage modulus, loss modulus and tan δ of the silica-filled NR vulcanizates

containing different compatibilizers were determined using a dynamic mechanical thermal
analyzer, DMTA V (Rheometrics Scientific, USA). The samples were tested in the tension
mode in a temperature range from -80oC to 80oC, at a frequency of 10 Hz under two strain
deformations: 0.001% strain at -80oC to -30oC and 0.01% strain from -30oC to 80oC.

102
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

5.2.6 Scanning electron microscopy

Tensile fractured surfaces of the silica-filled NR vulcanizates were gold sputter

coated and characterized by Scanning Electron Microscopy (SEM) as described in
Chapter 4.

5.3 Results and discussion

5.3.1 Effect of silane coupling agents on the properties of ENR-compatibilized

silica-filled NR

5.3.1.1 Mooney viscosity and Payne effect

90 500
(a) ENR+TESPT (b)
80
Payne effect [G′(0.56%)-G′(100%)] 450 ENR+VP Si-363
400 ENR+NXT
70
TESPT 8.6 wt%
350 None
60
ML1+4(100oC)

300
50
250
40
200
30 ENR+TESPT 150
ENR+VP Si-363
20 100
ENR+NXT
10 TESPT 8.6 wt% 50
None
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 5.1 Mooney viscosity (a) and Payne effect (b) of silica-filled NR compounds with
7.5 phr of ENR-51 in combination with varying amounts of TESPT, VP Si-363 and NXT.

Mooney viscosities of the silica-filled NR compounds with different types and

amounts of silane coupling agents added on top of 7.5 phr of ENR-51 as compatibilizer
are shown in Figure 5.1(a). The silica-filled compound without compatibilizer shows by far
the highest compound viscosity as previously observed in Figure 4.1, due to hindrance of
rubber flow by the filler network and silica agglomerates. The presence of ENR-51 as
compatibilizer significantly reduces the Mooney viscosity as a result of interactions
between the epoxy groups of ENR and silanol groups on the silica surface, as discussed

103
Chapter 5

in Chapter 3 and proposed in Scheme 3.2. The addition of increasing amounts of silane
further decreases the Mooney viscosity slightly, indicating an improvement of silica
dispersion, as a small amount of free silanol groups may still exist in the system after the
silica-ENR interactions. An excess amount of silane coupling agents could also act as
plasticizer for the rubber compounds and reduce the compound viscosity.
The compounds with VP Si-363 and NXT silane coupling agents show lower
Mooney viscosities than those with TESPT. The long alkoxy groups with polyglycol units
of VP Si-363 enhance the hydrophobation and so improve silica dispersion as well as
promote the rubber chain mobility around the silica surface[16], whereas the blocking group
in the NXT molecules can act as a plasticizing fragment that may result in a reduction of
compound viscosity. Furthermore, the bulky side-groups of VP Si-363 and the blocking
group of NXT prevent a reaction between the mercapto group and rubber polymer during
processing. This is different from TESPT that has polysulfidic linkages which can
disproportionate and donate sulfur atoms into the system to generate premature
crosslinking during processing and may cause a higher compound viscosity.[10,18] The
Payne effect of the silica-filled NR compounds with different compatibilizers is shown in
Figure 5.1(b). All the compatibilized compounds show similar Payne effects close to that
of the reference TESPT compound, indicating that the filler-filler interactions are efficiently
replaced by interactions between the epoxy groups of ENR and alkoxy groups of the
silanes with the silanol groups on the silica surface. The increase of silane coupling
agents to 4-5 wt% relative to silica results in the same level of Payne effect as the
compound with optimal TESPT dosage, and all the three types of silanes show no
different effect on this property.

5.3.1.2 Filler-rubber interaction

Silica is an inorganic filler with a large number of polar silanol groups on its
surface leading to a poor interaction with hydrocarbon rubbers. To improve the interfacial
interaction between silica and rubber, silane coupling agents or other types of
compatibilizers are generally added into the silica-filled rubber compounds, so that an
enhancement of mechanical and dynamic mechanical properties can be achieved. Bound
rubber content is a general parameter used to determine the degree of filler-rubber
interaction in the compounds. The effects of silane types and concentrations on the bound
rubber content of the ENR-compatibilized silica-filled NR compounds are shown in Figure
5.2. Therein, Figure 5.2 (a) shows the chemically bound rubber contents after the

104
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

treatment with ammonia for cleaving physical interactions, and Figure 5.2 (b) displays the
results of physically bound rubber content.

60 16
Chemically bound rubber content (%)

(b)

Physically bound rubber content (%)

(a)
14
50
12
40
10

30 8

6
20 ENR+TESPT
ENR+TESPT
ENR+VP Si-363
4 ENR+VP Si-363
ENR+NXT
10 ENR+NXT
TESPT 8.6 wt% 2 TESPT 8.6 wt%
None
None
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 5.2 Chemically (a) and physically (b) bound rubber contents of silica-filled NR
compounds with 7.5 phr of ENR-51 as compatibilizer in combination with varying amounts
of TESPT, VP Si-363 and NXT.

The addition of ENR-51 as compatibiizer in the silica-filled NR compounds

increases the chemically bound rubber but decreases the physically bound rubber,
attributed to generation of interactions via hydrogen bonding and/or chemical bonding
during mixing at high temperature between the epoxy groups of ENR and the silanol
groups of silica. The chemically bound rubber contents of the compounds clearly increase
with increasing amount of TESPT, NXT and VP Si-363, respectively, indicating that
remaining silanol groups further interact with the additional functional groups of the
silanes, leading to a further increase of silica-rubber interactions.
The use of TESPT gives the highest chemically bound rubber content and the
lowest physically bound rubber content, while the VP Si-363 gives the lowest chemically
bound rubber and the highest physically bound rubber of the three silanes. TESPT has a
bis-(triethoxy-)-structure in which there are more ethoxy groups available to react with the
silanol groups of silica. Moreover, the tetrasulfide part may introduce premature scorch
during mixing at high temperature.[18] NXT has a triethoxy group and the lowest molecular
weight among the three types of silane used in this work, and so will have more ethoxy

105
Chapter 5

groups available for silanization compared with VP Si-363 which has only 1 ethoxy group
in a molecule and the highest molecular weight due to the bulky long alkoxy substituents,
as shown in Table 5.1. As reported by Klockmann[16], for VP Si-363 the silica-silane
reaction proceeds via the ethoxy group, while the long dialkoxy groups with polyglycol
parts leads to fast adsorption on the silica surface and increases hydrophobation. This
results in the lowest chemically bound rubber content, but the highest physically bound
rubber content of the three silanes.

5.3.1.3 Cure properties

From the cure curves of the silica-filled NR compounds with different types of
silane coupling agent in Figures 5.3(a-c), the compound without any compatibilizer shows
the highest torque prior to real vulcanization due to filler flocculation, causing the re-
formation of silica agglomerates that hinder chain movement. It is well-known that re-
agglomeration of silica or filler flocculation in the filled compounds may take place in the
beginning of the vulcanization reaction when the compounds are heated and less
viscous.[19-21] The rate of the filler flocculation in the rubber matrix depends on filler
loading, polymer matrix type, viscosity, mixing history and temperature.[19] Increasing the
temperature for vulcanization leads initially to a decrease of polymer matrix viscosity and
increase of the filler diffusion coefficient causing an easier movement of filler
particles/aggregates to quickly form filler-filler networks under the shearing condition.[20]
The silica-filled NR compounds with optimal TESPT content, i.e. 8.6 wt% relative to silica,
shows no occurrence of flocculation due to a good interfacial interaction between silica
and the rubber phase through the silanization reaction that shields the free silanol groups
on the silica surface.[21] The eventual premature crosslinks introduced by sulfur out-of
TESPT also help to suppress the flocculation in silica-filled NR compounds.[18] The use of
ENR-51 at 7.5 phr as compatibilizer substantially reduces the silica flocculation in the
silica-filled NR compound but the phenomenon is still clearly observed in the induction
period of the cure curves, indicating that free silanol groups are still present in the
compound at a larger amount compared with TESPT. The ENR compatibilized
compounds also show the poorest cure rate, as observed by a low slope of the cure curve
in Figure 5.3, the longest optimum cure time as seen in Figure 5.4 (a), and the lowest
torque difference in Figure 5.4 (b). This indicates poor cure behavior in the presence of
ENR-51 alone as compatibilizer.

106
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

40
(a)
35

30

Torque (dN.m)
25

20 None
TESPT 8.6 wt%
15 ENR-51 7.5 phr
ENR+TESPT 2 wt%
10
ENR+TESPT 3 wt%
ENR+TESPT 4 wt%
5
ENR+TESPT 5 wt%

0
0 5 10 15 20
Time (min)

40 40
(b) (c)
35 35

30 30
Torque (dN.m)

Torque (dN.m)

25 25

20 20
None None
15 TESPT 8.6 wt% 15 TESPT 8.6 wt%
ENR-51 7.5 phr ENR-51 7.5 phr
10 ENR+VP Si-363 2 wt% 10 ENR+NXT 2 wt%
ENR+VP Si-363 3 wt% ENR+NXT 3 wt%
5 ENR+VP Si-363 4 wt% 5 ENR+NXT 4 wt%
ENR+VP Si-363 5 wt% ENR+NXT 5 wt%
0 0
0 5 10 15 20 0 5 10 15 20
Time (min) Time (min)

Figure 5.3 Cure curves of silica-filled NR compounds with 7.5 phr of ENR-51 as
compatibilizer and varying amounts of different silanes; TESPT (a), VP Si-363 (b) and
NXT (c), in comparison with that of the compounds without compatibilizer and with
reference TESPT.

107
Chapter 5

10 35
(a) (b)
9
30

Torque difference (dN.m)

8
Optimum cure time (min)

7 25

6
20
5
4 15
ENR+TESPT
ENR+TESPT
3 10 ENR+VP Si-363
ENR+VP Si-363
2 ENR+NXT ENR+NXT

TESPT 8.6 wt% 5 TESPT 8.6 wt%

1
None None
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 5.4 Optimum cure time (a) and torque difference (b) of silica-filled NR compounds
with 7.5 phr of ENR-51 as compatibilizer in combination with varying amounts of TESPT,
VP Si-363 and NXT.

The addition of silane coupling agents in combination with ENR-51 into the silica-
filled NR compounds reduces the silica flocculation by an additional increase of chemical
interfacial interactions between the silica and rubber phases, as seen in the increased
bound rubber content in Figure 5.2. In Figure 5.3(a), the ENR/TESPT compounds show a
decrease of minimum torque with increasing TESPT concentrations, to finally reach the
same level as that of the compound with reference TESPT. Shorter scorch and optimum
cure times (Figure 5.4(a)) are also observed when TESPT was added to the compound
with ENR as compatibilizer. This is due to a further decrease of free silanol groups by the
silanization reaction on the silica. In the presence of TESPT, the vulcanization reaction is
also accelerated by its sulfur donor effect. The results for mercaptosilane VP Si-363, as
shown in Figure 5.3(b), display a reduction of silica flocculation with increasing VP Si-363
content, but the minimum torques of the compounds are still higher than for the reference
TESPT. This indicates a lesser extent of filler-rubber interactions due to the fact that the
VP Si-363 molecule has two bulky alkoxyl groups with only one ethoxy group to react with
the silanol groups of silica leading to less silanization when compared with TESPT. On an
equal weight basis, the higher molecular weight of VP Si-363 also results in lower number
of molecules and functional groups for the reaction. The bulky substituents prevent the
“scorchiness” of the mercapto group by steric hindrance[16], but this mercapto group

108
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

provides a more efficient coupling with rubber during the vulcanization reaction when
compared with the compounds with ENR and TESPT. This is also observed by the shorter
scorch and optimum cure times of the compounds containing VP Si-363 in Figure 5.3 (a-b)
and Figure 5.4(a). Generally, compounds with TESPT have a much better scorch safety
than compounds with a normal mercaptosilane such as -mercapto propyltriethoxysilane
(MPS) due to the steric hindrance of the bifunctional groups of tri-ethoxysilyl-propyl that
surround the tetrasulfide part in TESPT, leading to a less active sulfur moiety for coupling
and vulcanization.[11]
Increasing NXT content shows a gradual decrease of silica flocculation and
minimum cure torques as shown in Figure 5.3 (c), because of improvements in silica-
rubber interactions and the increase of plasticization fragments (i.e. octanoyl part) of the
NXT molecule. This aliphatic tail also provides hydrophobicity and improves the silica
dispersion causing a lowering in the minimum torque, corresponding to the compound
viscosity. The compounds with NXT show outstanding long scorch times, i.e. better scorch
safety compared to the ones with TESPT. This is due to the mercaptan group blocked by
the long aliphatic blocking group in NXT structure, as shown in Table 5.1, that reduces the
silane-NR reactivity during processing, so that the silane-NR coupling occurs only during
vulcanization.[14]
Figure 5.4(b) shows the cure torque difference (MH-ML) which is closely related to
crosslink density in the materials, including both chemical bonds and interactions. It is
clearly seen that the use of all compatibilizer types strongly increases the cure torque
difference compared to the non-compatibilized one, and the use of the silane containing a
higher sulfur rank, i.e. TESPT, gives the highest torque differences, indicating more
crosslinks in the vulcanizates.

109
Chapter 5

5.3.1.4 Tensile properties

2.5 12
(a) (b)

10
2.0

300% Modulus (MPa)

100% Modulus (MPa)

8
1.5

6
1.0
ENR+TESPT
4 ENR+TESPT
ENR+VP Si-363
ENR+VP Si-363
0.5 ENR+NXT
ENR+NXT
TESPT 8.6 wt% 2
TESPT 8.6 wt%
None
None
0.0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 5.5 Moduli at 100% (a) and 300% (b) elongations of silica-filled NR vulcanizates
with 7.5 phr of ENR-51 as compatibilizer in combination with varying amounts of TESPT,
VP Si-363 and NXT.

The silica-filled NR vulcanizates with ENR/TESPT combinations as

compatibilizers show a remarkable increase of the 100% and 300% moduli with increasing
TESPT content, as shown in Figures 5.5(a) and (b), whereas the ENR/VP Si-363 and
ENR/NXT combinations show only a small change in tensile modulus with increasing
silane contents. The substantially higher modulus of the vulcanizates with TESPT is
caused by the highest filler-rubber interactions in the silica-filled compounds as
determined by the bound rubber content: Figure 5.2, and the higher crosslink density
resulting from sulfur released from the TESPT molecule. The VP Si-363 and NXT have
only 1 sulfur atom in the molecule and cannot donate sulfur to the system. Furthermore,
when considering the molecular structures and molecular weights of the silanes, VP Si-
363 and NXT provide lower amounts of ethoxy groups to bond to the silanol groups of
silica when compared to TESPT.

110
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

Reinforcement index (M300/M100)

4

2 ENR+TESPT

ENR+VP Si-363

ENR+NXT
1
TESPT 8.6 wt%

None
0
0 2 4 6 8
Silane contents
(wt% rel. to silica)

Figure 5.6 Reinforcement index of silica-filled NR vulcanizates with 7.5 phr of ENR-51 as
compatibilizer in combination with varying amounts of TESPT, VP Si-363 and NXT.

The ratio of tensile moduli at 300% to 100% elongation is often used for
determining the reinforcement index of silica-filled rubber vulcanizates. As shown in Figure
5.6, the reinforcement index of the silica-filled NR vulcanizate with reference TESPT only
is remarkably higher than that of all other vulcanizates, and the compatibilization by using
ENR results only in a slightly higher reinforcement index than the fully non-compatibilized
one. The introduction of silanes to combine with ENR increases the reinforcement index
somewhat, but not to the level of TESPT alone. The reinforcement index relates to filler-
rubber interactions which are indicated by the chemically bound rubber content as shown
in Figure 5.2(a). A higher silica-rubber interaction increases the modulus of vulcanizates
due to more restriction of chain movement. For the ENR/silane combination, the use of
TESPT silane gives the highest reinforcement index among the three types of silanes
studied and an optimum value is observed at 4.0 wt% of TESPT relative to silica.

111
Chapter 5

30 800
(a) (b)
700
25

Elongation at break (%)

Tensile strength (MPa)

600
20
500

15 400

ENR+TESPT
300 ENR+TESPT
10 ENR+VP Si-363
ENR+VP Si-363
ENR+NXT 200
ENR+NXT
5 TESPT 8.6 wt%
TESPT 8.6 wt%
None 100
None
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 5.7 Tensile strength (a) and elongation at break (b) of silica-filled NR vulcanizates
with 7.5 phr of ENR-51 as compatibilizer in combination with varying amounts of TESPT,
VP Si-363 and NXT.

Tensile strength and elongation at break of the silica-filled NR vulcanizates with

ENR-51/silane combinations as compatibilizer are increased slightly with increasing silane
concentrations, as shown in Figure 5.7. The use of 7.5 phr of ENR-51 as compatibilizer
apparently leaves some free silanol groups in the system and so the addition of just a
small amount of silane coupling agent can further enhance the interaction between the
silica and rubber phases through the silanization reaction. Tensile strength of the silica-
filled vulcanizates with optimal content of TESPT is the highest, and the ENR/TESPT
combination gives a higher tensile strength than the use of the other two types of silane.
The ENR/TESPT system shows an optimum tensile strength when 4.0 wt% of TESPT
relative to silica was used, which is still inferior to that of the reference compound with
TESPT alone. Overall, the use of different silanes in combination with 7.5 phr of ENR-51
results in only small differences in tensile strength and elongation at break, where both
properties are only slightly increased with increasing silane loadings.

112
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

5.3.2 Effect of extra sulfur on the properties of silica-filled NR compatibilized

with the ENR51/silane combination

Due to the structural differences between the three silane types as shown in
Table 5.1, the amount of total sulfur in the compounds which affects the network formation
is therefore different. By taking the compound with TESPT of 8.6 wt% relative to silica as a
reference, corrections are applied for the other compounds with extra elemental sulfur
added together with other curatives on a two-roll mill in the second step of mixing. The
silica-filled NR compounds for this study are compatibilized with ENR-51 at 7.5 phr and
silane at 4.0 wt% relative to silica, i.e. 2.2 phr in the formulation: Table 5.3. The three
types of silane coupling agents, i.e. TESPT, VP Si-363 and NXT, are used with a sulfur
correction to compensate for the sulfur content in the reference compound with TESPT
alone. The properties are reported in comparison with those without compatibilizer, with
reference TESPT (i.e. 8.6 wt% rel. to silica), and with 7.5 phr of ENR-51 only.

5.3.2.1 Mooney viscosity

90

80 No sulf ur correction
With sulf ur correction
70

60
ML1+4 (100oC)

50

40

30

20

10

0
None TESPT ENR ENR/ ENR/ ENR/
TESPT VP Si-363 NXT
Compatibilizer Types

Figure 5.8 Mooney viscosities of silica-filled NR compounds with different combinations of

ENR-51 (7.5 phr)/silanes (2.2 phr) without and with extra sulfur, compared to the ones
without compatibilizer, resp. with reference TESPT only and with ENR-51 alone.

As previously seen in Figure 5.1, and again in Figure 5.8, the silica-filled NR
compound without compatibilizer shows the highest Mooney viscosity, and the
incorporation of compatibilizers at their optimum amount drastically reduces the Mooney

113
Chapter 5

viscosity to similar levels as results of enhanced filler-rubber interactions and improved

silica dispersion. With 7.5 phr of ENR-51 and 4 wt% of VP Si-363 and NXT relative to
silica content, the compounds show slightly lower Mooney viscosities compared with the
ones with TESPT 8.6 wt% relative to silica, with ENR and with the ENR/TESPT
combination. The long blocking group of NXT and the polymeric alkoxy substituents of VP
Si-363 should help in shielding the silica surface, promote more hydrophobicity, aid silica
dispersion, and improve processability. The compounds with and without sulfur correction
show more or less the same level of Mooney viscosities as the extra sulfur was only
introduced in the second step of mixing on the two-roll mill. This extra sulfur is therefore
expected to take part in the vulcanization only and play no role in the mixing stage.

5.3.2.2 Cure properties

40 40
(a) (b)
35 35

30 30
Torque (dN.m)

Torque (dN.m)

25 25

20 20
None None
15 TESPT 8.6 wt% 15 TESPT 8.6 wt%
ENR-51 7.5 phr ENR-51 7.5 phr
10 10 ENR+TESPT
ENR+TESPT
ENR+VP Si-363 ENR+VP Si-363
5 5
ENR+NXT ENR+NXT

0 0
0 5 10 15 20 0 5 10 15 20
Time (min) Time (min)

Figure 5.9 Cure characteristics of silica-filled NR compounds with different combinations

of ENR-51 (7.5 phr)/silanes (2.2 phr) without (a) and with (b) extra sulfur, compared to the
ones without compatibilizer, resp. with reference TESPT only and with ENR-51 alone.

114
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

10 35
No sulf ur correction (a) No sulf ur correction (b)
9 With sulf ur correction With sulf ur correction
30

Torque differences (dN.m)

8
25
Cure time (min)

6
20
5
15
4

3 10
2
5
1

0 0
None TESPT ENR ENR/ ENR/ ENR/ None TESPT ENR ENR/ ENR/ ENR/
TESPT VP Si-363 NXT TESPT VP Si-363 NXT
Compatibilizer types Compatibilizer types

Figure 5.10 Cure time (a) and torque differences (b) of silica-filled NR compounds with
different combinations of ENR-51 (7.5 phr)/silanes (2.2 phr) without and with extra sulfur,
compared to the ones without compatiblizer, resp. with reference TESPT only and with
ENR-51 alone.

When extra sulfur was added into the silica-filled NR compounds, all of the
compounds show an increase of maximum cure torques and faster cure times as seen in
Figures 5.9 (a, b) and Figures 5.10 (a, b). The correction of the compounds with elemental
sulfur relative to the sulfur contained in the reference TESPT system leads to an increase
of crosslink density in the rubber matrix and consequently to the increase of maximum
cure torques and torque differences, as shown in Figure 5.10(b). It was reported that
increasing sulfur content in the rubber compounds increases the maximum torque due to
the increase in total crosslink density, increased glass transition temperature (Tg) and
decreased fractional free volume of the material.[22] With sulfur correction, the minimum
cure torques show only a little change, in accordance with the Mooney viscosity results in
Figure 5.8. On comparing between the three silane types that were used in combination
with ENR-51, the use of VP Si-363 gives the fastest scorch and cure times due to the
reactive mercapto function, and the cure curves show a strong reversion after the
optimum cure. The use of NXT gives the best scorch safety, while the use of TESPT
shows results somewhere in between.

115
Chapter 5

5.3.2.3 Tensile properties

3.0 5.0
No sulf ur correction (a) No sulf ur correction (b)
4.5

Reinforcement index (M300/M100)

With sulf ur correction
With sulf ur correction
2.5
4.0
100% Modulus (MPa)

3.5
2.0
3.0

1.5 2.5

2.0
1.0
1.5

1.0
0.5
0.5

0.0 0.0
None TESPT ENR ENR/ ENR/ ENR/ None TESPT ENR ENR/ ENR/ ENR/
TESPT VP Si-363 NXT TESPT VP Si-363 NXT
Compatibilizer types Compatibilizer Types

Figure 5.11 100% Modulus (a) and reinforcement index (b) of silica-filled NR vulcanizates
with different combinations of ENR-51 (7.5 phr)/silanes (2.2 phr) without and with extra
sulfur, compared to the ones without compatiblizer, resp. with reference TESPT only and
with ENR-51 alone.

The tensile modulus at 100% elongation and reinforcement index of the silica-
filled NR vulcanizates with different types of compatibilizers are shown in Figures 5.11(a)
and (b). For the combination system, the use of ENR-51/TESPT shows the highest 100%
modulus and reinforcement index, while the ENR-51/VP Si-363 and ENR-51/NXT show
similar properties. All the compatibilized compounds show very much better moduli than
the compounds without compatibilizer and the benchmark TESPT gives outstanding
properties. With the correction of sulfur content in the silica-filled NR compounds relative
to the reference TESPT compound, a significant increase of tensile modulus of the
vulcanizates is obtained due to an increase in crosslink density. The reinforcement index
is also slightly increased by sulfur correction.

116
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

35 800
No sulf ur correction (a) No sulf ur correction (b)
Tensile strength (MPa)
30 With sulf ur correction 700 With sulf ur correction

600

Elongation at break (%)

25
500
20
400
15
300
10
200

5 100

0 0
None TESPT ENR ENR/ ENR/ ENR/ None TESPT ENR ENR/ ENR/ ENR/
TESPT VP Si-363 NXT TESPT VP Si-363 NXT
Compatibilizer types Compatibilizer types

Figure 5.12 Tensile strength (a) and elongation at break (b) of silica-filled NR vulcanizates
with different ENR-51 (7.5 phr)/silanes (2.2 phr) combinations without and with extra
sulfur, compared to the ones without compatiblizer, with reference TESPT only and with
ENR-51 alone.

Tensile strength of the vulcanizates is also improved somewhat by applying the

sulfur correction relative to the reference TESPT compound as shown in Figure 5.12(a). In
accordance with the modulus, the use of TESPT together with ENR-51 gives the highest
tensile strength among the three different ENR/silane combinations due to the number of
ethoxy groups for interaction with the silanol groups on the silica surface to increase the
silica-filler interactions, and the sulfur donated from TESPT molecules to contribute to
network formation in the vulcanizates. When compared to the compounds that contain
TESPT and the same total sulfur content, the use of TESPT 8.6 wt% relative to silica
gives a higher 300% modulus and tensile strength than the use of ENR/TESPT with sulfur
correction. This may indicate a more effective use of sulfur released from the TESPT
molecules during the mixing at high temperature, when compared to the corrected sulfur
that was added on the cold two-roll mill. When the modulus of a vulcanizate is increased
due to higher network density, causing a higher stiffness, elongation at break decreases
as shown in Figure 5.12(b).

117
Chapter 5

5.3.2.4 Dynamic mechanical properties

100000 10000
(a) (b)
None
TESPT None
10000 1000 TESPT
ENR
ENR/TESPT ENR
ENR/VP Si-363 ENR/TESPT

Eʺ (MPa)
Eʹ (MPa)

ENR/NXT ENR/VP Si-363

1000 100 ENR/NXT

100 10

10 1
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Temperature (oC) Temperature (oC)

Figure 5.13 Storage (a) and loss (b) moduli of silica-filled NR vulcanizates compatibilized
with 7.5 phr of ENR-51 and 4.0 wt% of silanes relative to silica, with sulfur compensation,
in comparison with those vulcanizates without compatibilizer, resp. with reference TESPT
and with ENR-51 alone.

Storage moduli (E′) of the silica-filled NR vulcanizates containing different

compatibilizers show only small differences in the glassy region, where the molecular
motions are largely restricted. However, the vulcanizates do show their different behaviors
in the rubbery region, which is determined by the segmental mobility of the rubber chains.
When compared the silica-filled NR with only TESPT and without compatibilizer, the
silica/TESPT system shows higher storage modulus (E′) but lower loss modulus (Eʺ) after
the glass-to-rubber transition as shown in Figures 5.13(a) and (b). As previously
discussed, the silica/TESPT compound has the strongest filler-rubber interactions by its
highest chemically bound rubber content in Figure 5.2(a) and an increase of crosslink
points through silane bridges. This vulcanizate therefore shows higher elastic modulus
and better elasticity that leads to lower energy loss under dynamic conditions. The silica-
filled NR without compatibilizer that has the highest filler-filler interaction or Payne effect;
Figure 5.1(a), leads to less elasticity of the material as the rubber being trapped in the
filler-filler network and low interfacial interaction between the silica and rubber phases
causes loss of energy as dissipated in heat.[23] The incorporation of ENR-51 and ENR-

118
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

51/silanes combinations introduce the secondary transition in the range of -10 to 10oC as
observed in both the storage and loss moduli curves. This second transition is due to the
presence of the second polymer ENR-51, which has a higher Tg than NR and the damping
behavior of the ENR in the Tg-transition range. The presence of two Tg’s confirms that NR
and ENR-51 are basically immiscible.

1.2

None
1.0 TESPT
ENR
ENR/TESPT
0.8
ENR/VP Si-363
ENR/NXT
Tan δ

0.6

0.4

0.2

0.0
-80 -60 -40 -20 0 20 40 60 80
Temperature (oC)

Figure 5.14 Loss tangent (tan δ) of silica-filled NR vulcanizates compatibilized with 7.5 phr
of ENR-51 and 4.0 wt% of silanes relative to silica with sulfur compensation, in
comparison with those vulcanizates without compatibilizer, resp. with reference TESPT
only and with ENR-51 alone.

The results of tan δ in Figure 5.14 show that the silica-filled NR with TESPT has
the highest intensity of tan δ peak due to the improved of silica dispersion and less rubber
trapped in the silica network, leading to more rubber segments involved in the polymer
segmental relaxation in the glassy region.[24] The addition of ENR as compatibilizer in the
silica-filled NR decreases the intensity of tan δ at peak, as also shown in the values in
Table 5.4. This reduction is caused by the silica-ENR interactions and intermolecular
interactions between epoxy groups of adjacent ENR molecules, which results in a
restriction of polymer chain relaxations and an increase in the material stiffness. These
results are in accordance with the tan δ curves as previously shown in Figures 3.12 and
4.10. The addition of silane coupling agents together with ENR-51 increases the (tan δ)max
values slightly when compared to the vulcanizate with only ENR-51, as summarized in

119
Chapter 5

Table 5.5, because of a further improvement of filler dispersion and less occluded rubber.
Among the different compatibilizer types, the tan δ peak of the TESPT-compound is
broader towards the rubbery region compared to that of the others as a result of a greater
extent of filler-rubber interactions in this material.
The second glass transition peak corresponding to ENR-51 is clearly observed
for the compounds that are compatibilized with either ENR only or with ENR/silane.
According to Manna et al. [25], epoxidized natural rubber with 50 mol% of epoxy groups
and 60 phr of silica has a Tg of about -3 to 1oC. It is observed in Figure 5.14 that in the
presence of silane coupling agents, the second tan δ peak intensity is increased and the
peak position shifted toward lower temperature in comparison with ENR-51 alone. In the
systems with ENR/silane combinations, the silica-silane interactions are promoted
together with epoxide-silica interactions. In this case, there is competition of more free
epoxide groups in the ENR compared to the silica-filled NR with only ENR as
compatibilizer. In the combination system, ENR therefore exhibits its characteristic peak in
a more prominent manner. For the compound with only ENR as compatibilizer, the second
tan δ peak is broader and less pronounced because of more interactions between epoxide
groups of ENR and silanol groups on the silica, leading to a weaker characteristic peak of
ENR. This observation of the change of tan δ peak of epoxidized natural rubber as
compatibilizer in silica-filled NR due to the level of filler-rubber interactions was also
reported by Saramolee et al.[26]
The Tg values taken at both tan  and loss modulus peaks are summarized in
Table 5.5 together with the tan  values at different temperatures. On comparing with the
non-compatibilized material, the Tg’s of the vulcanizates with compatibilizers shift to higher
temperature slightly, due to an increase of silica-rubber interactions that restrict the rubber
segmental relaxations in the transition region.

Table 5.5 Tg and tan δ values at different positions taken from Figure 5.14.

Tg(oC) Values of Tan δ

Compatibilizer types
at tan  peak at E peak at peak at 5oC at 60oC
Without compatibilizer -47 -58 0.94 0.09 0.11
TESPT -45 -54 0.97 0.10 0.07
ENR-51 -44 -52 0.77 0.09 0.11
ENR-51 +TESPT+S -45 -57 0.79 0.12 0.10
ENR-51 +VP Si-363+S -45 -57 0.80 0.13 0.09
ENR-51 +NXT+S -45 -57 0.82 0.11 0.10

120
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

0.16 0.12
(a) (b)
0.14
0.10
0.12

Tan δ at 60oC
0.08
Tan δ at 5oC

0.10

0.08 0.06

0.06
0.04
0.04
0.02
0.02

0.00 0.00
None TESPT ENR ENR/ ENR/ ENR/ None TESPT ENR ENR/ ENR/ ENR/
TESPT VP Si-363 NXT TESPT VP Si-363 NXT

Compatibilizer types Compatibilizer types

Figure 5.15 Tan δ values at 5oC (a) and 60oC (b) of silica-filled NR vulcanizates with
different combinations of ENR-51 (7.5 phr)/silanes (2.2 phr) with sulfur correction,
compared to the ones without compatiblizer, resp. with reference TESPT only and with
ENR-51 alone.

For the tan δ values at 5oC and 60oC, the results are plotted in Figure 5.15.
According to UNECE regulation no. 117[27], wet grip grading of normal tires should be
tested in the temperature range of 5 – 35C. Figure 5.15(a) shows the tan  values at 5C
of the silica-filled vulcanizates containing different compatibilizers and without. The use of
ENR-51 together with silane coupling agents gives the highest tan δ values in the
temperature range of 5-35C which is in the range of the second glass transition
associated with the ENR, when compared to the use of TESPT, ENR only and no
compatibilizer, respectively. The different levels of interactions between the epoxide
functions and silanol groups on the silica when ENR is used alone or with silane, may
cause differences in the segmental relaxation behavior of the NR and affect the peak
intensity. The increased tan  at low temperature indicates a better wet grip of tire
compounds.
On the other hand, the tan δ value at high temperature, commonly in the range of
40-80C indicates tire rolling resistance. As shown in Figure 5.15(b), the incorporation of
ENR either with or without silane results in high tan δ values at 60C which are just slightly
lower than that of the compound without compatibilizer. This damping property is again
related to the Tg of ENR-51. The high mol% of epoxide groups on the ENR introduces
chain stiffness and lower elasticity. Moreover, the different polarity of the two rubber
phases leads to blend incompatibility which causes an energy loss at their interface under
dynamic deformation. Among the different compatibilizers studied, the use of conventional

121
Chapter 5

TESPT in silica-filled NR compounds provides a remarkably low tan δ at 60C, indicating

superior low rolling resistance for tire compounds.

5.3.2.5 Surface topography

Tensile fractured surfaces of silica-filled NR vulcanizates without compatibilizer,

with TESPT, with only ENR-51 and with ENR/silane combinations are shown in Figures
5.16(a) – (i). As previously seen in Figure 4.13 and discussed in Chapter 4, the lowest
tensile strength due to poor filler-rubber interactions of the vulcanizate without
compatibilizer results in a rather smooth fracture surface, whereas the fracture surface of
the silica-filled rubber vulcanizate with TESPT shows a rough surface with many tear lines
because of the strong silica-rubber interactions and bonding, leading to the high
mechanical strength of the material.[28] The fracture surface of the vulcanizate with ENR
as compatibilizer also shows tear lines, but less than that of the reference TESPT
compound. Therefore, the use of silane in addition to ENR as compatibilizer may reduce
silica dewetting in the rubber matrix, and improve interfacial adhesion between filler and
rubber.

122
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

(a) (b) (c)

(d) (e) (f)

(g) (h) (i)

Figure 5.16 Tensile fractured surfaces of silica-filled NR vulcanizates with different types
of compatibilizers; upper row: reference compounds without compatibilizer (a), with
TESPT (b), with ENR-51 (c); middle row: compounds without extra sulfur: ENR/TESPT
(d), ENR/VP Si-363 (e), ENR/NXT (f); lower row: compounds with extra sulfur:
ENR/TESPT/S (g), ENR/VP Si-363/S (h), ENR/NXT/S (i).

Figures 5.16(d)-(f) show the failure surfaces of the silica-filled NR with

ENR/silane combinations without sulfur correction and Figures 5.16(g)-(i) with sulfur
correction. All of them show more tear lines on the fracture surface compared to the silica-
filled NR without compatibilizer in Figure 5.16(a), but a less complicated tear line pattern
compared with that of the vulcanizate with TESPT; Figure 5.16(b). The use of ENR-51 in
combination with silane coupling agents leads to an improvement of silica-rubber
interaction, and the addition of extra sulfur that increases the crosslink density in the
rubber vulcanizate further increases the tensile strength of the materials. The increase of
tensile strength creates more surface roughness due to ductile failure. Among the three

123
Chapter 5

types of silane coupling agents used together with ENR-51 as compatiblizer, there is no
clear trend seen between the two sets of vulcanizates without and with extra sulfur.

5.4 Conclusions

The properties of silica-filled NR compounds compatibilized with 7.5 phr of ENR-

51 can be further enhanced by the addition of silane coupling agents at a smaller amount
compared to the normal use of TESPT. The study on the use of three types of silane;
TESPT, NXT and VP Si-363, at varying amounts in the range of 2 to 5 wt% relative to
silica together with 7.5 phr of ENR-51 shows, that the addition of silane reduces the
compound viscosity slightly but clearly enhances the chemically bound rubber content and
cure properties. Among the 3 different ENR-51/silane combinations, the use of
ENR/TESPT gives the highest chemically bound rubber content and tensile properties,
whereas the ENR/NXT combination gives higher chemically bound rubber content but
similar tensile properties compared to the ENR/VP Si-363 system. The optimum tensile
strength is obtained when 7.5 phr of ENR-51 with TESPT at 4.0 wt% relative to silica are
used. The better improvement of the overall properties by the use of TESPT over the
other two types of silanes is due to the bis-(triethoxysilyl)- and tetrasulfide-functional
groups that provide more ethoxy groups relatively to react with the silanol groups on the
silica surface, and the reactive free sulfur released to involve in network formation. When
the amount of silane coupling agent is fixed at 4.0 wt% relative to silica in combination
with 7.5 phr of ENR-51 as compatibilizers in the silica-filled NR compound, the addition of
extra sulfur to correct the total sulfur content relative to the reference compound with
TESPT leads to increases of modulus and tensile strength while elongation at break
decreases. The enhanced interactions and bonding between the silica and rubber phases
by means of compatibilizers, either without or with sulfur correction, shift the Tg of the
silica-filled NR to higher temperature when compared to the rubber without any
compatibilizer. The use of an ENR/silane combination increases the tan δ at 5oC and
reduces the tan δ at 60oC compared to the use of ENR-51 alone, indicating an
improvement in wet traction and rolling resistance of tires. However, overall the use of
TESPT alone at optimum loading gives the best mechanical and dynamic mechanical
properties compared to the presence of ENR as compatibilizer. The tensile rupture
surfaces, as observed by SEM, of the silica-filled NR vulcanizates that are compatibilized
using different compatibilizers correspond to their tensile strength, that is the materials of
higher tensile strength show more tear lines and fracture surface roughness.

124
 Silica-reinforced NR compounds compatibilized by ENR in combination with different silanes

Among the three types of silanes studied in combination with ENR-51 and sulfur
compensation, the use of 2.2 phr TESPT together with 7.5 phr ENR-51 provides the best
overall properties, but it does not fully match the properties of a compound with TESPT
alone at optimal quantity, i.e. 4.7 phr. However, in the perspective of the “Magic Triangle
of Tire Technology” where wet skid resistance, rolling resistance and wear are to be
optimized, the combination of ENR-51, TESPT and sulfur compensation may become a
better option for “Winter Tires” that demand good wet grip, providing that the real tire test
has yet to be confirmed.

5.5 References

1. W.H. Waddell, L.R. Evans, Rubber Chem. Technol., 69, 377 (1996).
2. A. Voet, J.C. Morawski, J.B. Donnet, Rubber Chem. Technol., 50, 342 (1977).
3. S. Wolff, Rubber Chem. Technol., 69, 325 (1996).
4. I.R. Gelling, Rubber Chem. Technol., 58, 86 (1985).
5. A. Bandyopadhyay, M.D. Sarkar, A.K. Bhowmick, Rubber Chem. Technol., 77, 830
(2004).
6. C.S.L. Baker, I.R. Gelling, R. Newell, Rubber Chem. Technol., 58, 67 (1985).
7. A.K. Manna, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer, J. Appl. Polym. Sci., 74,
389 (1999).
8. T.L.A.C. Rocha, C. Rosca, J. Ziegler, R.H. Schuster, Kautsch. Gummi Kunstst., 58,
22 (2005).
9. M.-J. Wang, S. Wolff, Rubber Chem. Technol., 65, 715 (1992).
10. J.W. ten Brinke, S.C. Debnath, L.A.E.M. Reuvekamp, J.W.M. Noordermeer, Comp.
Sci. Technol., 63, 1165 (2003).
11. B.T. Poh, C.C. NG, Eur. Polym. J., 34, 975 (1998).
12. H. Yan, G. Tian, K. Sun, Y. Zhang, Y. Zhang, J. Appl. Polym Sci., 43, 573 (2005).
13. H. Yan, K. Sun, Y. Zhang, Y. Zhang, Y. Fan, J. Appl. Polym. Sci., 94, 1511 (2004).
14. H. Yan, K. Sun, Y. Zhang, Y. Zhang, J. Appl. Polym. Sci., 94, 2295 (2004).
15. J.L. Valentín, I. Mora-Barrantes, A. Rodríguez, L. Ibarra, L. Gonzalez, J. Appl.
Polym. Sci., 103, 1806 (2007).
16 O. Klockmann, the International Tire Exbibition & Conference (ITEC) 2006,
September 12-14, 2006, Akron, OH, paper no.29B.
17. A. Blume, Kautsch. Gummi Kunstst., 64, 38 (2011).
18. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 85, 277 (2012).

125
Chapter 5

19. G.G.A. Böhm, M.N. Nguyen, J. Appl. Polym. Sci., 55, 1041 (1995).
20. C.J. Lin, W.L. Hergenrother, E. Alexanian, G.G.A. Böhm, Rubber Chem. Technol.,
75, 865 (2002).
21. S. Mihara, R.N. Datta, J.W.M. Noordermeer, Rubber Chem. Technol., 82, 524
(2009).
22. W. Salgueiro, A. Marzocca, A. Somoza, G. Consolati, S. Cerveny, F. Quasso, S.
Goyanes, Polymer, 45, 6037 (2004).
23. H.M. Da Costa, L.L.Y. Viconte, R.C.R. Nunes, C.R.G. Furtado, J. Appl. Polym. Sci.,
83, 2331 (2002).
24. L. Qu, G. Yu, L. Wang, C. Li, Q. Zhao, J. Li., J. Appl. Polym. Sci., 126, 116 (2012).
25. A.J. Manna, P.P. De, D.K. Tripathy, J. Appl. Polym. Sci., 84, 2171 (2002).
26. P. Saramolee, K. Sahakaro, N. Lopattananon, W.K. Dierkes, J.W.M. Noordermeer,
Rubber Chem. Technol., 87, 320 (2014).
27. UNECE Regulation No 117, GRRF 71st session, September 13-15, 2011.
28. W. Arayapranee, N. Na-Ranong, G.L. Rempel, J. Appl. Polym. Sci., 98, 34 (2005).

126
 CHAPTER 6

Influence of Types of Silane Coupling Agents on the

Reinforcement of Silica in Natural Rubber Compounds

In this chapter, the effect of different types of silane coupling agents on the
properties of silica-filled natural rubber (NR) compounds is studied in order to gain basic
information for selecting a silane to be used later for preparation of silane-grafted NR.
Three silane types: 3-Octanoylthio-1-propyltriethoxysilane (NXT), 3-mercaptopropyl-
di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-363) and
vinyltriethoxysilane (VTES) are used in the silica-filled NR compounds, based on molar
and alkoxy functional group equivalents in comparison with conventional bis-
(triethoxysilylpropyl)tetrasulfide (TESPT). The addition of silane coupling agents in the
silica-filled NR compounds reduces mixing energy, Mooney viscosity and Payne effect,
while it improves the chemically bound rubber content and tensile properties. Increase of
the silane coupling agent content to the same level of alkoxy functional groups as that of
TESPT, results in improvement of both the compound and vulcanizate properties. Among
the three types of silane, the use of VTES in silica-filled NR gives the poorest properties.
With equivalent amounts of ethoxy groups, the use of NXT and reference TESPT results
in the same level of tensile strength. The silica-filled NR compounds with NXT, VP Si-363
and VTES as coupling agents show inferior wet grip and rolling resistance, as far as can
be derived from a lower tan δ at 5C and higher tan δ at 60C, when compared to the
reference compound with TESPT. With respect to the tire performance, NXT offers still
better properties than VP Si-363 and VTES, respectively.
Chapter 6

6.1 Introduction

Based on the results in Chapters 3-5, even though epoxidized natural rubber
(ENR) could promote filler-rubber interactions and improve both compound and
vulcanizate properties to some extent, the presence of ENR as compatibilizer has a
negative effect on the dynamic mechanical properties, especially with respect to tire rolling
resistance. The combination of ENR and a small amount of silane coupling agent
enhances the properties but the results remain inferior to the use of a silica/TESPT
system in which the rubber matrix is homogeneous in absence of ENR. With the objective
to modify NR by means of grafting silanes onto the NR molecules, in order to overcome
the problems raised by the NR/ENR blend, several potential silanes are selected for this
study. NXT with shielded mercapto function and VP Si-363 with a sterically hindered
mercapto functional group, and VTES with a vinyl function are of interest, all commercially
available. The chemical structures of the silanes are given in Table 6.1. Prior to
performing the grafting reaction, the influence of these silanes on the properties of silica-
filled NR compounds is investigated and reported in this chapter, in comparison with the
use of TESPT as reference. This in order to know the reinforcing efficiency of silica in NR
compounds when these silanes are applied as coupling agents.
Bis-(triethoxysilylpropyl)tetrasulfide (TESPT) has become the state-of-the-art in
silica-silane technology and is most widely used in silica-filled rubber compounds. The
silanization reaction of the ethoxy-groups in TESPT molecules with silanol groups of silica
during mixing, and the coupling reaction between silane and rubber molecules during the
vulcanization reaction lead to generation of chemical bonds between silica and rubber[1,2],
introducing a double network structure in the filled rubber; i.e. a network between filler and
rubber via silane bridges and a rubber network generated by the curatives. The effective
use of TESPT provides a good silica dispersion, low Payne effect, good abrasion
resistance and mechanical properties, and low tire rolling resistance.[2-4] TESPT itself has
some drawbacks however, as the tetrasulfide structure can lead to “scorch” problems and
the silanization reaction emits ethanol during mixing. Alternative silanes such as 3-
octanoylthio-1-propyltriethoxysilane (NXT) and 3-mercaptopropyl-di(tridecan-1-oxy-13-
penta(ethyleneoxide))ethoxysilane (VP Si-363) have been introduced, as previously
mentioned in Chapter 5. NXT is an octanoyl-blocked mercaptosilane developed to provide
a better scorch safety than TESPT for the silica-filled rubber compounds.[5] However, the
apparent activation energy of the vulcanization reaction of the NXT compound is lower
than that of the TESPT compound, and the apparent activation energy of both compounds
decreases with increasing silane concentration.[6] VP Si-363 or 3-mercaptopropyl-

128
 Influence of silane types on the reinforcement of silica in natural rubber compounds

di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane is a mercaptosilane containing

one ethoxy-group and two long alkoxy-groups. It has been reported that the presence of
the thiol-group or the long alkoxy groups in the VP Si-363 structure significantly increases
the reaction rate between silica and silane. Furthermore, the long alkoxy chains with
oxygen atoms in the structure can increase the possibility of adsorption of this silane on
the silica leading to a faster silane-silica reaction when compared to the mercaptosilane
with short chain alkoxy groups.[7] Compared with the use of TESPT, the incorporation of
VP Si-363 into silica-filled styrene-butadiene rubber (SBR)/butadiene rubber (BR) blends
for tire compounds improved the rolling resistance of a tire by 13% and reduced volatile
organic compounds (VOC) emission by 80%.[8,9] A study on nanoclay-filled rubber showed
that VP Si-363 has a smaller effect on tensile modulus than TESPT, which was possibly
due to the bulky alkoxy group of VP Si-363 that took some space when bonded to the
clay surface and restricted the number of bonds between silane and filler.[10]
In addition to the mercapto-silanes, there are still other types of silane coupling
agents that have been tested for silica-filled polymers, and also for some other purposes,
to include amino-silane, cyano-silane, vinyl silane, etc. Vinyl silanes such as
vinyltrimethoxysilane (VTMS) and vinyltriethoxysilane (VTES) are applied for moisture
crosslinking of polyolefins in order to improve their properties and expand the applications
of these materials.[11-14] In the rubber industry, vinyl silanes are commonly used to improve
the properties of inorganic fillers-filled ethylene-propylene rubber (EPM) and ethylene-
propylene-diene rubber (EPDM) with a peroxide vulcanization system. Vinyl-silane
containing formulations show a remarkable improvement in modulus, tensile and tear
properties, filler dispersion and dynamic mechanical properties of these materials.[15,16]
Besides a good interaction between filler and rubber as a result of the silanization
reaction, the vinyl-group in the silane structure can also react with free radicals on rubber
molecules during peroxide vulcanization[16] leading to a strong bonding between filler and
rubber, as reflected in good static and dynamic properties of the filled rubber. The
presence of a C=C structure in the vinyl silane offers the possibility for a grafting reaction
of this silane onto the polymer chains if radicals are generated in the process.
In this chapter, the effect of different types of silane coupling agents: NXT, VP
Si-363 and VTES, on the properties of silica-filled NR is studied by applying two sets of
compounds based on equal molar and on alkoxy-functional groups relative to the
reference compound with TESPT. The reinforcing efficiency of silica in combination with
the different silanes in NR compounds is compared. Herein, the three types of silanes are
studied with respect to their potential to be later used for preparation of silane grafted NR.

129
Chapter 6

6.2 Experimental

6.2.1 Materials

The compounding ingredients: natural rubber (Ribbed Smoked Sheet (RSS) #3),
highly dispersible silica (Zeosil 1165MP), treated distillate aromatic extract oil (TDAE-oil),
N-cyclohexyl-2-benzothiazole sulfenamide (CBS), diphenyl guanidine (DPG) and 2,2,4-
trimethyl-1,2-dihydroquinoline (TMQ), ZnO, stearic acid and sulfur, are as detailed in
Chapter 3. Bis-(triethoxysilylpropyl)tetrasulfide (TESPT), 3-Octanoylthio-1-
propyltriethoxysilane (NXT) and 3-Mercaptopropyl-di(tridecan-1-oxy-13-
penta(ethyleneoxide))ethoxysilane (VP Si-363) are described with their producers in
Chapter 5. Vinyltriethoxysilane (VTES) was supplied by Evonik (Germany). All ingredients
were used as received. The chemical names, structures and molecular weights of all the
silanes used in this chapter are shown in Table 6.1.

Table 6.1 Silane coupling agents information.

Silane MW
Chemical names Structures
types (g/mol)
CH3CH2O OCH2CH3
Bis-(triethoxysilylpropyl)
TESPT CH3CH2O Si CH2 Sx CH2 Si OCH2CH3 532.5
tetrasulfide 3 3
CH3CH2O with x = 3.70 OCH2CH3

3-Mercaptopropyl- CH3(CH2)12(OC2H4)5O
VP Si-
di(tridecan-1-oxy-13-penta CH3CH2O Si CH2 SH 987.5
363 3
(ethyleneoxide))ethoxysilane CH3(CH2)12(OC2H4)5O

CH3CH2O
O
3-Octanoyl-thio-
NXT CH3CH2O Si CH2 S C CH2 CH3 365
propyltriethoxysilane 3 6
CH3CH2O

CH3CH2O

VTES Vinyltriethoxysilane CH3CH2O Si CH CH2 190

CH3CH2O

6.2.2 Compound formulations and mixing procedures

Compound formulations are shown in Table 6.2 and the amounts of NXT, VP Si-
363 and VTES silanes were calculated to either molar or alkoxy groups equivalents to that
of the reference compound with TESPT. All the compounds were mixed following the
mixing procedures as previously described in Table 5.3, but in the present case, only NR

130
 Influence of silane types on the reinforcement of silica in natural rubber compounds

was used without ENR. The mixer and mixing conditions are as described in Section
5.2.2.

Table 6.2 Compound formulations.

Parts per hundred parts of rubber (phr)
Ingredients References Molar equivalents Alkoxy groups
equivalents
RSS3 100.0 100.0 100.0 100. 100.0 100.0 100.0 100.
0 0
TESPT - 4.7* - - - - - -
NXT - - 3.2 - - 6.4 - -
VP Si-363 - - - 8.7 - - 17.4 -
VTES - - - - 1.7 - - 3.4
Zeosil 55.0 55.0 55.0 55.0 55.0 55.0 55.0 55.0
1165MP
TDAE oil 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0
ZnO 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
TMQ 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Stearic acid 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
DPG 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
CBS 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
Sulfur 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
Remarks: * TESPT 4.7 phr equals 8.8 mmol.

6.2.3 Testing of compound and vulcanizate properties

Mooney viscosity, Payne effect, bound rubber content and cure characteristics of the
compounds were tested using the methods and equipments as described in Chapter 5.
The compounds were cured to their respective optimum cure times, and the resulting
vulcanizates were tested for their tensile and dynamic mechanical properties, also as
described in Chapter 5.

6.3 Results and discussion

6.3.1 Mixing behaviors

The processability of the rubber compounds can initially be determined by their

mixing behavior and the torques generated during mixing, which are associated with the
compound viscosities. The plots of mixing torques versus mixing time of the compounds
with equal molar and alkoxy-functional groups of NXT, VP Si-363 and VTES compared to
TESPT respectively, are shown in Figures 6.1–6.3. The mixing torque of the filled
compounds largely depends on the level of filler dispersion. By using a bifunctional silane

131
Chapter 6

with alkoxy groups, the silanization reaction between the alkoxy groups of the silane and
the silanol groups of the silica takes place in the mixer. A TESPT-molecule contains an
average 3.7 sulfur atoms and two sides of triethoxy groups that can effectively react with
the silanol groups on the silica surface under optimum mixing conditions. It has been
reported that the premature crosslinks created by the sulfur in TESPT in silica-filled NR
compounds showed a positive effect on silica dispersion as it helped to suppress filler
flocculation.[17] The NXT, VP Si-363 and VTES are also bifunctional silanes with different
chemical structures, which may have an influence on the reactivity towards the silanization
and later coupling reaction with the rubber.

50

40
Mixing torque (N.m)

Without silane

30
NXT 1x

TESPT
20
NXT 2x

10
NXT 1x = molar equivalent to TESPT
NXT 2x = alkoxy-f unctional equivalent to TESPT

0
0 5 10 15 20
Time (min)

Figure 6.1 Mixing torques of silica-filled NR with NXT in comparison with TESPT and
without silane coupling agent.

Figure 6.1 shows the mixing torques of silica-filled NR compounds without silane
coupling agent, with TESPT, and with NXT at two different contents based on equal molar
and ethoxy-groups relative to TESPT. As expected, the silica-filled NR compound without
silane coupling agent clearly shows the highest mixing torque, related to the highest
compound viscosity due to strong filler-filler interaction that causes formation of a filler-
filler network of silica aggregates and restricts the mobility of the rubber. Moreover, some
parts of rubber are trapped inside the silica network, i.e. occluded rubber. It has been
demonstrated that the silica-silica network and a high amount of occluded rubber
contributes to the shear modulus of the uncured compounds[18], that shall reflect in the
mixing torque. The incorporation of silane coupling agents in the silica-filled compounds

132
 Influence of silane types on the reinforcement of silica in natural rubber compounds

clearly improves the compound processability as seen in the reduction of mixing torques
especially after the addition of the 2nd half of the filler plus silane and oil, as a result of
improved silica dispersion and presence of the process oil. At the beginning, the silane
addition is comparable to a plasticizer: it reduces the viscosity. At the end of the mixing
cycle, the compound with NXT at equal ethoxy-functional groups to TESPT shows a
slightly lower mixing torque compared to the reference compound with TESPT.

50

40
Mixing torque (N.m)

Without silane

30

TESPT

20 VP Si-363 1x
VP Si-363 2x

10
VP Si-363 1x = molar equivalent to TESPT
VP Si-363 2x = alkoxy-f unctional equivalent to TESPT
0
0 5 10 15 20
Time (min)

Figure 6.2 Mixing torques of silica-filled NR with VP Si-363 in comparison with TESPT and
without silane coupling agent.

The use of VP Si-363 at both loadings in the silica-filled NR compounds results in

a remarkably lower mixing torque than the compound with TESPT as seen in Figure 6.2.
The mixing torque decreases with increasing amount of VP Si-363 and the difference in
mixing torques can be clearly observed in the second interval of silica addition already,
indicating a very fast dispersion of silica in the rubber matrix. As reported by Blume [7], the
presence of the –SH and two long alkoxy groups in VP Si-363 increases the rate of
reaction with the silica due to the increase of adsorption possibilities to the silanol-groups
which can occur via the ethoxy-group, thiol and different oxygen atoms inside the alkoxy
chain through hydrogen bonding, as proposed in Scheme 6.1. Besides the reaction
between the ethoxy-group and the silanol-group, the long alkoxy chains that shield the
silica surface promote the silica dispersion and improve the processability of the silica-
filled rubber compound.

133
Chapter 6

SH

O O O O O O Si O O O O O O
H H H H H O H H H H H
O O O O O O O O O O
Si Si Si Si Si Si Si Si Si Si Si Si Si
O O O O O O O O O O O O O O

Scheme 6.1 Proposed interaction mechanism between VP Si-363 silane and silanol
groups on the silica surface via hydrogen bonds.[7]

50

40
Mixing torque (N.m)

Without silane

30 VTES 1x
VTES 2x

TESPT
20

10
VTES 1x = molar equivalent to TESPT
VTES 2x = alkoxy-f unctional equivalent to TESPT

0
0 5 10 15 20
Time (min)

Figure 6.3 Mixing torques of silica-filled NR with VTES in comparison with TESPT and
without silane coupling agent.

In contrast to the other two silanes, the addition of vinyl-silane VTES results in
final mixing torques which are higher than for the compound containing TESPT, but still
lower than without silane, as shown in Figure 6.3. At equal amount of ethoxy-functional
groups of TESPT and VTES, the VTES-containing compound still shows a higher mixing
torque compared to the use of TESPT. As the silane molecule reacts with the silica via the
ethoxy groups and the chemical structure of the whole silane influences its kinetic
[7]
behavior , the VTES with the short vinyl function that gives less steric hindrance
compared to NXT and VP Si-363, should provide a better reactivity towards the silanol
groups. According to the study by Blume[7] with a model compound, the rate of the
beginning reaction of VTES is slightly higher than for VP Si-363 and clearly higher than for

134
 Influence of silane types on the reinforcement of silica in natural rubber compounds

TESPT. Despite the higher rate of the beginning reaction of VTES over TESPT, the
compound with VTES shows a higher mixing torque, i.e. higher viscosity. The smaller
steric effect of VTES molecules leads to a lesser shielding effect on the silica surface.
Moreover, there may be a loss of some ethoxy-groups of VTES through the hydrolysis-
reaction with moisture and a consequent condensation reaction with neighboring ethoxy-
groups due to their high reactivity. For the case of TESPT, the bulky molecule introduces
steric hindrance and reduces the rate of condensation between its own molecules, and in
addition TESPT has a larger shielding effect on the silica surface compared to VTES,
resulting in the possibly better silanization efficiency and filler dispersion.

Table 6.3 Characteristics of mixing behavior of silica-filled NR compounds with different

silane types.
Silane content Specific energy Dump temperature
Compounds
(mmol) (kJ/g) (oC)
Without silane - 4.37 156
TESPT 8.8 3.43 144
NXT 1x 8.8 3.48 144
VP Si-363 1x 8.8 3.13 138
VTES 1x 8.8 3.89 148
NXT 2x 17.6 3.32 142
VP Si-363 2x 17.6 2.94 134
VTES 2x 17.6 3.80 146

The incorporation of rigid filler particles into rubber compounds increases the
viscosity and impedes the flow of rubber molecules, causing a higher mixing energy and
an increase of compound temperature due to generated heat. The silanization reaction
between the alkoxy-groups of the silanes and the polar silanol groups of silica leads to
less filler-filler interaction and improvement of silica dispersion in the rubber matrix, and
consequently affects compound viscosity and mixing torque.[3] Based on the mixing torque
data, the mixing energy (kN.m/kg) was obtained and the specific energy in units of (kJ/g)
for each compound was calculated, and the values summarized in Table 6.3. As already
observed in the mixing torque curves in Figures 6.1-6.3, the use of silane coupling agents
reduces the mixing torques and the specific energy is in accordance with the mixing
torques or compound viscosity. Lower viscosity generates less shearing heat during
mixing and results in a lower dump temperature, as also shown in Table 6.3. Among the
four types of silanes, VP Si-363 gives the lowest specific mixing energies and dump

135
Chapter 6

temperatures whereas VTES shows the highest values. The use of higher silane contents
in the compounds in case of functional groups-equivalence to TESPT leads to less mixing
energy and lower dump temperature. In addition to the improvement of silica dispersion
and filler-rubber interactions, a silane itself can also act as a plasticizer that helps the
mobility of the rubber chains and therefore reduces viscosity.

6.3.2 Mooney viscosity, Payne effect and bound rubber content

The properties of filled uncured compounds generally correspond to the level of

filler dispersion, filler-rubber interactions and the occurrence of a filler-filler network.
Mixing of silica into NR or other hydrocarbon rubbers is known to face difficulties arising
from the strong silica-silica interactions and poor silica-rubber compatibility. Different
compatibilizers or silane coupling agents will have a different efficiency in silica-filled NR
compounds which affects the properties of the compounds. The use of NXT, VP Si-363
and VTES in comparison with TESPT affects Mooney viscosity, Payne effect and bound
rubber content of the compounds as follows.

90
molar equivalent to TESPT
80
alkoxy-f unctional equivalent to TESPT

70

60
ML1+4 (100oC)

50

40

30

20

10

0
Without silane TESPT NXT VP Si-363 VTES

Silane coupling agent types

Figure 6.4 Mooney viscosity of silica-filled NR compounds with different types and
amounts of silane coupling agents.

Mooney viscosities of the silica-filled NR compounds silanized with different

types of silane coupling agent are shown in Figure 6.4. The silica-filled NR compound
without silane shows a significantly higher Mooney viscosity than all other compounds in
accordance with the mixing torque result, as discussed previously. At the same molar

136
 Influence of silane types on the reinforcement of silica in natural rubber compounds

content, the compound with TESPT shows the lowest Mooney viscosity indicating its
superior flow due to the good silica shielding by the six ethoxy-groups. Among the
compounds with silanes, the use of VTES gives the highest compound Mooney viscosity,
independent whether it was used at either molar or functional equivalents to TESPT. This
poorest efficiency of VTES to enhance filler dispersion, by means of silanization and
shielding off the silanol groups, despite its high reactivity due to less steric effects, may be
attributed to the possibility of self-reaction between the VTES molecules as proposed in
Scheme 6.2. At the high temperature during mixing, the reactive ethoxy groups of VTES
may be hydrolyzed or directly condensed with neighboring ethoxy groups leaving less
functional groups to react with the silanol moieties on the silica surface and a minor
shielding-effect. Moreover, after the silanization of the silica the short vinyl-group on the
other side of the VTES molecule cannot promote molecular chain movement like in the
case of the long fragments of NXT and VP Si-363 molecules.

OC2H5 partially OC2H5

hydrolyzed
H2C CH Si OC2H5 H2C CH Si OH
OC2H5 OH

OC2H5 OC2H5 OC2H5 OC2H5

H2C CH Si OC2H5 HO Si CH CH2 H2C CH Si O Si CH CH2
OC2H5 OH OC2H5 OH

Scheme 6.2 Possible self-condensation reaction of VTES molecules.

At equal molar basis, the Mooney viscosity of the compound with TESPT is lower
than with VP Si-363 and NXT, but at the equal functional groups content the compound
with VP Si-363 shows the lowest Mooney viscosity of all. The addition of VP Si-363 results
in a lower Mooney viscosity than the use of NXT, because besides the silanization
reaction between the alkoxy groups of the silane and the silanol groups of silica, two long
alkoxy groups with different oxygen atoms promote the adsorption of silane on silica via
hydrogen bonding, as shown in Scheme 6.1.[7] Furthermore, the efficient shielding of the
silica surface and the long alkyl chains enhance the mobility of rubber chains around silica
aggregates leading to the low compound viscosity.

137
Chapter 6

3000 3000
(a) (b)

Payne effect [G′(0.56%)-G′(100%)] (kPa)

molar equivalent to TESPT
2500 Without compatibilizer 2500 alkoxy-f unctional equivalent to TESPT
TESPT
NXT 1x
2000 VP Si-363 1x 2000
Gʹ (kPa)

VTES 1x
NXT 2x
1500 1500
VP Si-363 2x
VTES 2x
1000 1000

500
500

0
0 Without TESPT NXT VP-Si-363 VTES
0.5 5 50 500 compatibilizer

Strain (%) Silane coupling agent types

Figure 6.5 Storage modulus versus strain (a) and Payne effect [G′(0.56%)-G′(100%)] (b)
of silica-filled NR compounds with different types and amounts of silane coupling agents.

Figures 6.5 (a) and (b) show the Payne effect or filler-filler interaction of the
silica-filled NR compounds. The plots of storage shear modulus G′ versus strain amplitude
in Figure 6.5 (a) clearly show a strong filler-filler interaction of the compound without silane
and only a small reduction of the Payne effect by the use of VTES. Increasing the VTES-
content reduces the Payne effect, but it is still very much higher than the use of TESPT,
VP Si-363 and NXT which can effectively diminish the filler-filler interactions in the
compounds. This is clearly seen in Figure 6.7 (b) in which the difference of G at 0.56%
and 100% strain is presented. The lower filler-filler interaction, i.e. less filler-filler
networking and better dispersion of the silanized silica-filled compounds with TESPT, NXT
and VP Si-363 is in good agreement with the compound viscosities: Figure 6.4, and
mixing torques: Figures 6.1-6.3, as previously discussed. With the same alkoxy-group
content, NXT gives even a lower Payne effect compared to TESPT. This difference may
be caused by the smaller size and less steric hindrance of the NXT molecules compared
to TESPT, which leads to an easier reaction between the alkoxy-group of NXT-silane and
the silanol groups of silica. Another effect is that TESPT can couple two silica-aggregates
which slightly increases the viscosity.

138
 Influence of silane types on the reinforcement of silica in natural rubber compounds

70 14
molar equivalent to TESPT (a) molar equivalent to TESPT (b)
Chemically bound rubber content (%)

Physically bound rubber content (%)

60 alkoxy-f unctional equivalent to TESPT 12 alkoxy-f unctional equivalent to TESPT

50 10

40 8

30 6

20 4

10 2

0 0
Without TESPT NXT VP Si-363 VTES Without TESPT NXT VP Si-363 VTES
silane silane
Silane coupling agent types Silane coupling agent types

Figure 6.6 Chemically bound rubber (a) and physically bound rubber content (b) of silica-
filled NR compounds with different types and amounts of silane coupling agents.

The interactions between silica and the rubber phase promoted by silane can be
[19]
proved by the chemically bound rubber content in the silica-filled NR compounds as
shown in Figure 6.6 (a). Among the various types of silane studied, the silica-filled NR
compound with TESPT shows the highest chemically bound rubber content, whereas the
compounds with NXT and VP Si-363 display similar values lower than for TESPT but still
higher than for the mixes with VTES and without silane. All the compounds show similar
physically bound rubber contents as shown in Figure 6.6 (b). The somewhat low value for
VP Si-363 is probably due to the large measurement error for the low values of the
physically bound rubber, being the difference of two large numbers: total-chemically
bound rubber. The good filler-rubber interactions in the compounds with the sulfur
containing silanes, i.e. TESPT, NXT and VP Si-363, is not only the result of an efficient
silanization reaction but also because of an increase of the adsorption possibility on the
silica surface by thiols. The exceptionally high chemically bound rubber content in the
TESPT compound can be attributed to the tetrasulfide structure that can donate free sulfur
into the system to generate some crosslinks in the compound, as previously demonstrated
by Kaewsakul et al.[17] The lowest chemically bound rubber of the compound with VTES
confirms that it has the lowest efficiency due to its minor shielding effect for use with silica
in NR-compounds compared to the other silane coupling agents. The decrease of the
physically and chemically bound rubber contents in the VP Si-363 silanized silica-filled NR
compounds when the amount of silane was increased could be due to its molecular
structure that has a long alkoxy-group. The ethoxy-groups will react with silanol-groups of
the silica, although the oligomeric poly(ethyleneoxide) side-arms will also react to some

139
Chapter 6

extent, in the same manner as poly(ethyleneoxide) itself can act as a shielding agent for
silica. The two long alkoxy chains per molecule should also improve the rubber chain
mobility around the silica aggregates and cause more loose chains to be dissolved in
toluene during the bound rubber content measurement.
The different structures of silanes on the silica surface after silanization and the
linkages between silane and rubber chains caused by either premature scorch in the
TESPT compound or after vulcanization in the compounds with TESPT, NXT and VP Si-
363, are proposed in Scheme 6.3.

140
 Influence of silane types on the reinforcement of silica in natural rubber compounds

Before vulcanization After vulcanization

S S
Si O Si (CH2)3 S Si O Si (CH2)3 S S
S
O O O O
Si O Si (CH2)3 S S Si O Si (CH2)3 S S
O O O O
Si O Si (CH2)3 S S Si O Si (CH2)3 S S
S
TESPT silanized silica

Si O Si Si O Si
O O O O O
S
Si O Si (CH2)3 S C (CH2)6 CH3 Si O Si (CH2)3 S
O O O O O
Si O Si (CH2)3 S C (CH2)6 CH3 Si O Si (CH2)3 S
S S
NXT silanized silica

O O O O
O O O
OH OH O
O OH OH
Si O OH O S
Si Si O OH
O Si
O O
O
Si O Si SH Si O Si S
O O
Si OH O O
Si OH
HO O OH
O S S
O O O
O O O O O

VP Si-363 silanized silica

Si O Si CH CH2 Si O Si CH CH2
O O O O S
Si O Si CH CH2 Si O Si CH CH2
O O O O
Si O Si CH CH2 Si O Si CH CH2 S S

VTES silanized silica

Scheme 6.3 Different structures of silanes on the silica surface after silanization and
coupling, before and after vulcanization.


The lateral condensation of silanes attached to adjacent isolated silanol-sites as depicted in this
Scheme 6.3 has often been quoted in literature. After the present project was completed it came to
our attention, that based on molecular modeling for steric reasons this reaction can only take place
for a minimum distance >4 Å, which surpasses the distance between adjacent silanol-sites by a factor
th
of appr. 2. (A.Blume, M. El-Roz, F.Thibault-Starzyk. Presented at KHK 11 Fall Rubber Colloquim,
Hanover, Germany, November 26-28, 2014).

141
Chapter 6

6.3.3 Cure characteristics

40 2.5 40
TESPT (a) (b)
35 Scorch time Torque dif f erence
35

Torque difference (dN.m)

2.0
30 NXT 1x 30
Torque (dN.m)

Without silane

Scorch time (min)

25 25
1.5
20 NXT 2x 20

15 1.0
15

10 10
0.5
NXT 1X = molar equivalent to TESPT
5 5
NXT 2x = alkoxy-f unctional equivalent to TESPT

0 0.0 0
0 5 10 15 20 25 30 35 Without silane TESPT NXT 1x NXT 2x

Time (min) Silane coupling agent types

Figure 6.7 Cure curves (a) and scorch time and torque difference (b) of silica-filled NR
compounds with NXT silane compared to the compounds with TESPT and without silane.

NXT silane is a mercapto silane with a carboxylic blocking group to retard the
reactivity of the mercapto-group. Basically, the NXT-compound is more stable at high
temperature than TESPT, providing a better scorch safety. As reported by Yan et al.[6], the
scorch time of a NR compound with NXT is longer than that with TESPT due to the
bonding dissociation energy of NXT, which is higher than that of TESPT. However, the
apparent activation energy of vulcanization of the compound with NXT is lower than that
with TESPT at a given concentration due to the steric hindrance of NXT which is less than
for TESPT. From the cure curves of silica-filled NR compounds in Figure 6.7 (a) the
compound without silane coupling agent clearly shows silica flocculation which means re-
agglomeration of silica under shear conditions at high temperature.[20] The addition of
TESPT and NXT suppresses silica flocculation due to increased filler-rubber interaction
through the silanization reaction and so decrease the filler-filler interaction. The use of
NXT and TESPT silanes at the same molar content results in a similar rheometer
minimum cure torque (ML) in which the NXT compound shows a little silica flocculation,
but after vulcanization, the maximum cure torque (MH) and torque difference (MH-ML) of
the NXT compound is lower than that of the TESPT compound, as shown in Figure 6.7(b).
Increase of the NXT silane loading in order to have the same amount of alkoxy-groups as
TESPT leads to a significant decrease of ML without a sign of silica flocculation and longer
induction period prior to vulcanization, as well as a lower MH due to its plasticization- in
combination with silanization-effects. The scorch time of the silica-filled NR compound

142
 Influence of silane types on the reinforcement of silica in natural rubber compounds

without silane is longer than that of the compound with TESPT due to the polar character
of the silica surface that adsorbs the polar curatives and results in cure retardation as well
as less vulcanization efficiency. In contrast, the addition of NXT-silane gives a longer
scorch time than the compounds without silane and with TESPT, respectively, due to the
carboxylic blocking group in the NXT-structure that results in more thermal stability than
the TESPT silane. The higher sulfur rank of TESPT with lower bonding energy of S-S in
TESPT compared to the C-S in NXT also leads to a higher reactivity of TESPT towards
coupling with rubber during the vulcanization reaction. TESPT may “donate” some of its
sulfur to the compound, to implicitly raise the amount of free sulfur and consequently give
extra crosslinking. NXT cannot do that.

40 2.5 40
(a) Scorch time Torque dif f erence
(b)
35 35
TESPT

Torque difference (dN.m)

2.0
30 30
Torque (dN.m)

Scorch time (min)

Without silane
25 25
1.5
VP Si-363 1x
20 20

1.0
15 15
VP Si-363 2x
10 10
0.5
5 VP Si-363 1x = molar equivalent to TESPT 5
VP Si-363 2x = alkoxy-f unctional equivalent to TESPT
0 0.0 0
0 5 10 15 20 25 30 35 Without silane TESPT VP Si-363 1x VP Si-363 2x

Time (min) Silane coupling agent types

Figure 6.8 Cure curves (a) and scorch time and torque difference (b) of silica-filled NR
compounds with VP Si-363 silane compared to the compounds with TESPT and without
silane.

Figure 6.8 (a) shows the cure curves of silica-filled NR compounds with two
different amounts of VP Si-363 in comparison with that of the compounds with TESPT and
without silane. VP Si-363 is a mercapto-silane with one ethoxy-group and two bulky long
alkoxy-groups. The reactive mercapto-group in the VP Si-363 molecule leads to a very
fast curing reaction as seen by the shorter scorch and cure times compared to the
compound with TESPT. The very short scorch time due to the highly reactive thiol end-
group of VP Si-363 can be clearly seen in Figure 6.8 (b). The compounds with VP Si-363
show a lower minimum torque than that with TESPT and no silica flocculation. Increasing
the amount of VP Si-363 drastically decreases the cure torque maximum and torque
difference: Figure 6.8 (b), which may be caused by a plasticizing effect of the long alkoxy-

143
Chapter 6

groups, leading to a reduction of stiffness of the final vulcanizate. Again, TESPT

outperforms VP Si-363 in maximum torque due to its sulfur donating effect.

40 2.5 40
TESPT (a) (b)
Scorch time Torque dif f erence
35 35

Torque difference (dN.m)

2.0
30 30
VTES 1x Without silane
Torque (dN.m)

Scorch time (min)

25 25
1.5
VTES 2x
20 20

15 1.0
15

10 10
VTES 1x = molar equivalent to TESPT 0.5
5 VTES 2x = alkoxy-f unctional equivalent to TESPT 5

0 0.0 0
0 5 10 15 20 25 30 35 Without silane TESPT VTES 1x VTES 2x

Time (min) Silane coupling agent types

Figure 6.9 Cure curves (a) and scorch time and torque difference (b) of silica-filled NR
compounds with VTES silane compared to the compounds with TESPT and without
silane.

Among the four types of silanes used in this work, VTES is a sulfur-free silane
with tri-ethoxy and vinyl-functional groups. The silica-filled NR compounds with VTES
silane show a reduction of silica flocculation with increasing VTES content, but the
flocculation phenomenon still clearly exists even in the compound with ethoxy-groups
equivalent to TESPT in accordance with the high Payne effect: Figure 6.5, and low
chemically bound rubber content: Figure 6.6 (a). As discussed previously, the small
molecule of VTES can easily react with the silanol groups of silica but not properly shield
the silica surface leading to a lower silanization efficiency between silica and silane.
Furthermore, the vinyl-structure does not enhance the adsorption on the silica surface like
with the other silanes. As observed in Figures 6.9 (a) and (b), increasing the amount of
VTES has only a little effect on the cure behavior of the compounds. The absence of
sulfur in the VTES/silica compound leads to a lack of coupling reactions via the sulfur
atoms and consequently lower network density, resulting in a lower maximum cure torque
and torque difference compared to the TESPT compound. The compound without silane
has a filler-filler network that results in a higher maximum torque MH compared to the one
with VTES, but due to its higher minimum torque ML, the torque differences (MH-ML) of the
compounds without silane and with VTES are similar. During vulcanization via a radical

144
 Influence of silane types on the reinforcement of silica in natural rubber compounds

mechanism, the vinyl-part may be bound to the rubber molecules, but this apparently did
not happen to a significant extent as the sulfur vulcanization is not a radical reaction.

6.3.4 Tensile and dynamic mechanical properties

3.0 6
molar equivalent to TESPT (a) molar equivalent to TESPT (b)

Reinforcement index (M300/M100)

alkoxy-f unctional equivalent to TESPT
2.5 5 alkoxy-f unctional equivalent to TESPT
100% Modulus (MPa)

2.0 4

1.5 3

1.0 2

0.5 1

0.0 0
Without silane TESPT NXT VP Si-363 VTES Without silane TESPT NXT VP Si-363 VTES

Silane coupling agent types Silane coupling agent types

Figure 6.10 Modulus at 100% strain (a) and reinforcement index (b) of silica-filled NR
compounds with different types and amounts of silane coupling agents compared to that
of the compounds with TESPT and without silane.

Figure 6.10 shows the 100%-modulus and reinforcement index of the silica-filled
NR vulcanizates containing different types of silanes in which the one with TESPT shows
the highest modulus, in correspondence with its highest chemically bound rubber: Figure
6.6, and maximum cure torques: Figures 6.7-6.9. Compared with NXT and VP Si-363
silanes, either at equal molar or number of alkoxy-groups, TESPT that consists of a higher
sulfur rank gives a remarkably higher tensile modulus contributed by a higher crosslink
density because of its higher sulfur content and sulfur donating ability. The lower bonding
dissociation energy of TESPT compared to NXT and VP Si-363 leads to a faster and
easier reaction with rubber. The bonding dissociation energy in sulfur-containing silanes is
as follows: C-S-C 285 kJ/mol; C-Sx-C (where x ≥ 2) < 268 kJ/mol; S-S  134 kJ/mol.[6,21]
Consequently, there are more chemical bonds between TESPT and rubber and a higher
crosslink density because of the extra sulfur donated by TESPT leading to higher
stiffness, i.e. higher modulus. At equal moles of silanes used, the vulcanizates with NXT
and VP Si-363 silanes show significantly lower moduli compared to that with TESPT, but
clearly higher than that with VTES. The differences in tensile modulus of the vulcanizates
containing different silane types resemble the differences in chemically bound rubber
content, as shown in Figure 6.6. The inferior modulus of the vulcanizate with VTES

145
Chapter 6

demonstrates that it suffers from absence of coupling reaction even though it shows some
improvement in the properties of uncured compounds such as the Payne effect compared
to the compound without silane.

35 900
molar equivalent to TESPT
alkoxy-f unctional equivalent to TESPT
(a) molar equivalent to TESPT (b)
800 alkoxy-f unctional equivalent to TESPT
30
700

Elongation at break (%)

Tensile strenght (MPa)

25
600

20 500

15 400

300
10
200
5
100

0 0
Without silane TESPT NXT VP Si-363 VTES Without silane TESPT NXT VP Si-363 VTES

Silane coupling agent types Silane coupling agent types

Figure 6.11 Tensile strength (a) and elongation at break (b) of silica-filled NR compounds
with different types and amounts of silane coupling agents compared to that of the
compounds with TESPT and without silane.

The tensile strength of the silica-filled NR vulcanizates having different types of

silane coupling agents as shown in Figure 6.11(a) shows the same trend as that of the
100% modulus: Figure 6.10(a) is in accordance with the chemically bound rubber content:
Figure 6.6(a). The state-of-the-art TESPT gives superior modulus and tensile strength
compared to the other silanes. However, the addition of NXT silane at equal amount of
alkoxy groups also results in a tensile strength at the same level as the use of TESPT. At
the same molar equivalents to TESPT, NXT and VP Si-363 give vulcanizates with the
same level of tensile strength, slightly lower than TESPT. But an increasing amount of VP
Si-363 decreases the tensile strength like for the 100% modulus, while the use of VTES
shows no improvement in tensile properties compared to the compound without silane.
The elongation at break of the vulcanizates without and with different silanes show only
small differences, where the vulcanizates with higher 100% modulus tend to have a lower
elongation at break. The maximum extent of deformation also correlates with the
maximum cure torque (MH) which may be taken as an indication for the crosslink or
network density in the materials. As seen in Figure 6.8, the NR with VP Si-363 has a
significantly lower MH compared to the vulcanizates without silane and with TESPT; it can
sustain larger deformation before breaking.

146
 Influence of silane types on the reinforcement of silica in natural rubber compounds

Tensile strength and modulus at 300% elongation show a slight correlation with
the chemically bound rubber content as shown in Figures 6.12(a) and (b), respectively.
Different types of silane show a different reinforcing efficiency depending on their
functional groups and structure, that influence the level of interaction and/or reaction
between silica and rubber. The use of TESPT, NXT and VP Si-363 significantly increases
the chemically bound rubber content compared to the non-compatibilized compound, and
consequently there is a large increase seen in tensile strength and 300%-modulus. The
use of TESPT gives the highest chemically bound rubber content and the highest tensile
strength and 300%-modulus of the silica-filled NR vulcanizates. The addition of VTES
causes no change in the chemically bound rubber content and even slightly lowers the
tensile strength and 300%-modulus compared to the compound without silane. It is worth
to note that the use of NXT and VP Si-363 gives comparable tensile strengths but lower
moduli compared to the use of TESPT.

35 12
(a) TESPT
(b)
30 NXT 2x
TESPT 10
NXT 1x VP Si-363 1x
300% Modulus (MPa)

25
Tensile strength (MPa)

8
VP Si-363 2x NXT 2x
20 Without compatibilizer
VP Si-363 1x
6
NXT 1x
15 VTES 2x
VTES 1x Without compatibilizer
4 VP Si-363 2x
10 VTES 2x
VTES 1x
5 2

0 0
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
Chemically bound rubber content (%) Chemically bound rubber content (%)

Figure 6.12 Tensile strength (a) and 300%-modulus (b) of silica-filled NR compounds with
different types of silane coupling agents as a function of chemically bound rubber content.

147
Chapter 6

0.15 0.12
molar equivalent to TESPT (a) molar equivalent to TESPT (b)
alkoxy-f unctional equivalent to TESPT alkoxy-f unctional equivalent to TESPT
0.10

0.10 0.08

Tan δ at 60oC
Tan δ at 5oC

0.06

0.05 0.04

0.02

0.00 0.00
Without TESPT NXT VP Si-363 VTES Without TESPT NXT VP Si-363 VTES
silane silane
Silane types Silane types

Figure 6.13 Tan δ values at 5C (a) and 60C (b) of silica-filled NR with different types and
amounts of silane coupling agents.

For tire applications or mechanical goods for use under dynamic conditions,
dynamic mechanical properties are of great importance. The key tire performances, i.e.
wet grip, rolling resistance and wear, are all related to the dynamic mechanical properties
of the rubber vulcanizates. Basically, wet grip refers to breaking ability for normal rubber
tires[22], and rolling resistance is a measure of the amount of energy consumed over
distance travelled. Herein, tan δ values at 5oC and 60oC may respectively be used to
indicate wet grip and rolling resistance, as shown in Figure 6.13. Figure 6.13(a) shows
o
the tan δ at 5 C of the silica-filled NR compounds with different types of silane coupling
agents in which the use of TESPT gives the highest tan δ at 5oC whereas the addition of
VTES results in the lowest tan δ at 5oC, close to the value of the compound without silane.
The vulcanizates with NXT and VP Si-363 show a higher tan δ at 5oC compared to the
one without silane, and the values are increased with increasing concentration of the
silanes. The results give a first indication that silica-filled NR vulcanizates with TESPT,
NXT and VP Si-363 should provide a better wet grip than that without silane.
The tan δ values at 60oC of the silica-filled NR vulcanizates are shown in Figure
6.13(b). For energy saving tires, this value should be low for reduction of heat build-up or
decrease of energy loss of tires, affected by friction and heat developed between the tire


In another project “Safe Tires-Save Energy” within the department of Elastomer Technology and
o
Engineering at the University of Twente, the correlation between tan δ at 0 C and factual friction
performance of tire compounds has been studied: E.M. Cichomski, “Silica-silane reinforced
passenger car tire treads: Effect of silica morphology, silica-polymer interface structure and rubber
matrix network on tire-performance indicators”: PhD-thesis, University of Twente, 29 May 2015. In
o
spite of the commonly accepted assumption that the tan δ at 0-20 C is indicative of wet grip, it has
been shown that this has to be taken with great reserve and does not generally hold.

148
 Influence of silane types on the reinforcement of silica in natural rubber compounds

tread and road surface. The results in Figure 6.13(b) clearly show that the silica-filled NR
without silane coupling agent shows the highest tan δ value at 60oC due to poor filler-
rubber interactions and thus inferior energy transfer as well as poor reinforcing efficiency.
The use of TESPT gives the best low tan δ at 60oC implying the best low rolling resistance
of a tire compound. Several factors contribute to this. The best filler-rubber interactions as
indicated by the chemically bound rubber content: Figure 6.6(a); and the highest crosslink
density as implied by the highest cure torque difference; Figures 6.7-6.9, likely due to the
earlier mentioned sulfur “donation” of TESPT. The use of NXT and VP Si-363 lowers the
tan δ value at 60oC compared to the filled rubber with no silane, where both types of
silanes give similar tan δ values. Increasing the NXT and VP Si-363 contents in order to
provide equal amounts of alkoxy groups to the reference TESPT slightly decrease the tan
δ at 60oC, which may be caused by a better filler dispersion.
Based on the overall results, TESPT is the best silane coupling agent in this
study for silica-filled NR compounds providing outstanding filler-rubber interaction,
mechanical and dynamic properties. The mercaptosilanes NXT and VP Si-363 efficiently
reduce the filler-filler interaction, enhance the compound properties, i.e. lower Mooney
viscosity and improve cure behavior, but give somewhat inferior mechanical and dynamic
properties compared to TESPT, most likely due to a lower degree of crosslinking than for
TESPT, which “donates” extra sulfur to the curing package. The difference could probably
partly or totally be overcome by adding some extra sulfur to the pertinent formulation.
VTES is not an effective silane coupling agent for this sulfur-crosslinked system as it
results in no improvement in properties compared to the compound without silane

6.4 Conclusions

The comparative study of the compound and vulcanizate properties of the silica-
filled NR compounds silanized by four types of silane coupling agents, i.e. TESPT, NXT,
VP Si-363 and VTES, shows their different effects on the reinforcement of silica in NR.
Among the four silane types, VTES gives the poorest overall properties due to its poor
shielding efficiency and the vinyl-group that is difficult to couple with NR rubber chains in
the presence of sulfur. The use of NXT and VP Si-363 shows a good improvement in
compound properties, i.e. mixing energy, Mooney viscosity, cure behaviors, bound rubber
content, and an increase of mechanical and dynamic mechanical properties compared to
that without silane, but the properties are inferior to the use of TESPT, mainly due to lower
crosslink densities compared with TESPT which has an intrinsic sulfur “donating” effect.
This was not further pursued in the context of this chapter. The incorporation of NXT

149
Chapter 6

silane provides better scorch safety, and use at the same amount of ethoxy-functional
groups with reference to TESPT gives vulcanizates with the same level of tensile strength,
but poorer modulus and tan δ at both 5oC and 60oC.

6.5 References

1. L.A.E.M. Reuvekamp, J.W. ten Brinke, P.J. Van Swaaij, J.W.M. Noordermeer,
Rubber Chem. Technol., 75, 187 (2002).
2. J.W. ten Brinke, S.C. Debnath, L.A.E.M. Reuvekamp, J.W.M. Noordermeer, Comp.
Sci. Technol., 63, 1165 (2003).
3. J.W. ten Brinke, P.J. Van Swaaij, L.A.E.M. Reuvekamp, J.W.M. Noordermeer,
Kautsch. Gummi Kunstst., 55, 244 (2002).
4. L. Ladouce-Stelandre, Y. Bomal, L. Flandin, D. Labarre, Rubber Chem. Technol., 76,
145 (2003).
5. H. Yan, K. Sun, Y. Zhang, Y. Zhang, J. Appl. Polym. Sci., 94, 2295 (2004).
6. H. Yan, K. Sun, Y. Zhang, Y. Zhang, Y. Fan, J. Appl. Polym. Sci., 94, 1511 (2004).
7. A. Blume, Kautsch. Gummi. Kunstst., 64 (4), 38 (2011).
8. O. Klockmann, The International Tire Exhibition & Conference 2006, 12-14
September 2006, Akron, United State of America.
9. W. Niedermeier, B. Schwaiger, Kautsch. Gummi. Kunstst., 60, 184 (2007).
10. D.J. Lowe, A.V. Chapman, S. Cook, J.J.C. Busfield, Rubber Chem. Technol., 86, 538
(2003).
11. Y.-T. Shieh, C.-M. Liu, J. Appl. Polym. Sci., 74, 3404 (1999).
12. M. Beltran, C. Mijangos, Polym. Eng. Sci., 40, 1534 (2000).
13. C. Jiao, Z. Wang, Z. Gui, Y. Hu, Eur. Polym. J., 41, 1204 (2005).
14. K. Sirisinha, M. Boonkongkaew, S. Kositchaiyong, Polym. Test., 29, 958 (2010).
15. M.W. Ranny, C.A. Pagano, Rubber Chem Technol., 44, 1080 (1971).
16. P.K. Pal, S.K. De, Polymer, 25, 855 (1984).
17. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 85, 277 (2012).
18. H.D. Luginsland, J. Fröhlich, A. Wehmeier, Rubber Chem. Technol., 75, 563 (2002).
19. S. Wolff, M.J. Wang, E.H. Tan, Rubber Chem. Technol., 66, 163 (1993).
20. G.A. Bohm, W. Tomasawski, W. Cole, T. Hogan, Polymer, 51, 2057 (2010).
21. S. Oae. Organic Sulfur Chemistry: Structure and Mechanism, CRC Press, Boca
Raton, USA, p.3 (1991).
22. UNECE Regulation No 117, GRRF 71st session, September 13-15, 2011.

150
 CHAPTER 7

Preparation and Characterization of Silane-Grafted

Natural Rubber

Natural rubber (NR) grafted with silane was prepared by melt mixing using 1,1′-
di(tert- butylperoxy)-3,3,5-trimethylcyclohexane (Luperox® 231XL40) as initiator, and two
types of silane: 3-Octanoylthio-1-propyltriethoxysilane (NXT) and 3-mercaptopropyl-
di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-363). First, VP Si-363
silane was used to optimize the grafting conditions for preparing the silane-grafted NR.
Grafting temperatures were varied in the range of 100-160oC at a fixed initiator content of
0.1 phr and of silane 20 phr. The VP Si-363 grafted NR at 140oC shows infrared
absorption peaks at wave numbers of 3270 and 1075 cm-1 assigned respectively to the
deformation of -OH and Si-O-C functional groups. Increase of the grafting temperature to
160C introduces a new absorption peak at 1710 cm-1 of C=O bonds in the grafted NR,
indicating occurrence of NR degradation at this high temperature. Increase of the initiator
concentration in the range of 0.1-1.0 phr gives no positive effect on the grafted amount but
leads to a significant increase of gel content: crosslinking. The conditions for grafting the
silanes onto NR in an internal mixer were therefore fixed at 140oC and initiator
concentration of 0.1 phr. NXT- and VP Si-363-grafted NRs were prepared with different
silane concentrations at 10 and 20 phr. The purified silane-grafted NRs show infrared
absorption peaks at 1075 and 1035 cm-1 which can be assigned respectively to the
deformation of Si-O-C and Si-O-Si bonds, and the peak at 3270 cm-1 of O-H deformation.
The 1H-NMR spectra of both NXT- and VP Si-363-grafted NRs confirm the presence of
methylene protons of the alkoxy groups (-O-CH2-C-) from the fragments of the silanes that
were attached to the NR molecules. The use of NXT silane at 10 and 20 phr for the
grafting reaction results in 0.66 and 1.32 mol% of grafted NXT, whereas the use of VP Si-
363 at 10 and 20 phr gives 0.85 and 1.64 mol% of grafted VP Si-363.
Chapter 7

7.1 Introduction

Silica has a hydrophilic surface by its specific character of a large number of

surface silanol groups, i.e. 4.9 OH.nm-2[1], leading to strong hydrogen bonding between
silica particles and poor filler-rubber interactions in non-polar rubbers. In general, a silane
coupling agent is directly added together with silica into a rubber during the mixing
process to enhance the silica-rubber compatibility. Under suitable mixing conditions, the
silane coupling agent undergoes a silanization reaction with the silanol groups on the
silica involving primary and secondary reactions.[2] To complete the reaction between silica
and bis-(triethoxysilylpropyl)tetrasulfide (TESPT) silane during mixing, the dump
temperature of the silica-filled SBR compound should be above 130oC, but at a
temperature above 160oC the silane starts to react with rubber to form premature
crosslinks.[3] For silica-filled NR compounds with TESPT as coupling agent, a dump
temperature is recommended in a range of 135-150oC and a higher temperature leads to
degradation of the NR.[4] Some sulfur-free silanes with alkoxy functional groups may react
with the silanol groups on the silica and so shield the silica surface to make it more
hydrophobic.[5] Various silanes show a different effect on the properties of silica-filled
rubber. A sulfur-free silane shows a good improvement in the Payne effect, but gives
poorer mechanical and dynamic mechanical properties of the silica-filled rubber compared
to sulfur-containing silanes such as TESPT.[6]
Surface modification of silica is an alternative method to improve silica
reinforcement in the rubber matrix. Modification of silica with a silane coupling agent can
change the characteristic surface from hydrophilic to hydrophobic, which is achieved by
partial blocking of hydroxyl groups on the silica surface. Surface modification with vinyl-
and mercapto-silanes reduces the silica aggregate size without re-agglomeration of the
silica, but modification with an amino-silane tends to increase the silica agglomerate size
because of its hydrophilicity that leads to formation of hydrogen bonds between
particles/aggregates.[7,8] Organic monomers such as styrene, isoprene and butadiene
have been used to modify the silica surface to obtain a hydrophobic character through in
situ polymerization of the monomers adsorbed on the surface via bi-layers of
surfactants.[9-11] The use of organic monomer surface-modified silica in a rubber matrix
leads to the improvement of cure behaviors, and tensile and tear properties.[10,11] The
plasma polymerization technique has been employed to modify silica by deposition of a
thin polymer film on its surface to improve the adhesion between the silica and rubber
phases, in which the polymer film can be tailored by choosing the monomer types and
reaction conditions. Polyacetylene (PA)-, polypyrrole (PPy)- and polythiophene (PTh)-

152
 Preparation and characterization of silane-grafted NR

coated silica surfaces showed a less hydrophilic character and less filler-filler interaction
compared to untreated silica.[12] The plasma polymerization of acetylene onto the silica
surface could improve the silica dispersion in styrene–butadiene rubber (SBR) and
enhance the tensile strength and modulus.[13,14] A PA-coated silica-filled SBR/ethylene-
propylene-diene rubber (EPDM) blend showed more homogeneous distribution of the filler
compared to the untreated one.[15]
Silane-modified polymer is another route employed for improvement of silica
reinforcement in polymeric materials. Polypropylene (PP) grafted with vinyltriethoxysilane
(VTES) was used in PP/silica nanocomposites in which the grafted PP could attach to the
silica surface through VTES molecules, leading to a decrease of chain mobility and
diffusion as proven by a shift of crystallization and melting behaviors to a higher
temperature.[16] A liquid low molecular weight polybutadiene (PB) was grafted with
mercaptopropyltrimethoxysilane (MPS) through radical addition of the thiol group to the
double bond on polymer molecule at 75oC and the MPS-grafted PB was silanized onto
silica surfaces at 135oC, allowing the alkoxy-groups of the silane fragment to react with
the silanol groups of the silica.[17] Modified-silica with the grafted PB showed a decrease of
the number of OH-groups on the surface to the same level as that of silica with TESPT,
due to the influence of the shielding effect of the grafted PB. The decrease of hydrophilic
character of the surface modified silica further lead to a reduction of filler-filler interaction
and an improvement of filler dispersion in the rubber matrix.[18] The use of MPS silane-
modified silica enhanced the final properties like tensile strength, but to a lesser extent
than the conventional use of TESPT as coupling agent during rubber mixing and
silanization.
In Chapter 6, the use of 3-octanoylthio-1-propyltriethoxysilane (NXT) and 3-
mercaptopropyl-di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-363) in
silica-filled NR improved both compound and vulcanizate properties, although the use of
TESPT still provided the best overall properties. However, the grafting of NXT and VP Si-
363 through their thiol groups onto the NR molecule is still of interest, and this Chapter 7
reports a study on such silane-grafted NRs. The effects of grafting temperature and
initiator concentration are investigated in order to find a suitable grafting condition. Then,
the optimized condition is applied to prepare NR grafted with silanes using 10 and 20 phr
of NXT and VP Si-363 silanes. The silane-grafted NRs are characterized by Fourier-
Transform Infrared (FTIR) and Proton Nuclear Magnetic Resonance (1H NMR)
spectroscopic techniques.

153
Chapter 7

7.2 Experimental

7.2.1 Materials

Natural rubber was Ribbed Smoked Sheet (RSS) #3, locally produced in
Thailand. Two types of silane coupling agents; 3-Octanoylthio-1-propyltriethoxysilane
(NXT) (Momentive, USA) and 3-Mercaptopropyl-di(tridecan-1-oxy-13-
penta(ethyleneoxide))ethoxysilane (VP Si-363) (Evonik, Germany), and 1,1′-di(tert-
butylperoxy)-3,3,5-trimethylcyclohexane (Luperox® 231XL40) (Arkema, USA) 40%
extended on calcium carbonate and silica, were used for the melt grafting reactions. The
structure of the initiator is shown in Figure 7.1 and the silane coupling agent details are as
previously given in Table 6.1 in Chapter 6.

CH3 CH3
CH3 C O O O O C CH3
CH3 CH3

CH3
CH3
CH3

Figure 7.1 Structure of 1,1′-di(tert-butylperoxy)-3,3,5-trimethylcyclohexane (Luperox®

231XL40).

7.2.2 Preparation of silane-grafted NR

NR was mixed with the silane coupling agent and the initiator in an internal mixer,
Brabender® 50EHT (Brabender® GmbH & Co.KG, Germany), for 12 minutes with a rotor
speed of 60 rpm using the formulations and mixing steps as shown in Tables 7.1 and 7.2,
respectively. For the grafting conditions study, VP Si-363 was used to investigate the
effect of grafting temperatures in the range of 100-160oC and of the initiator
concentrations in the range of 0.1-1.0 phr. The most suitable grafting temperature and
initiator concentration were then selected for preparing the VP Si-363- and NXT-grafted
NRs by using silane contents at 10 and 20 phr.

154
 Preparation and characterization of silane-grafted NR

Table 7.1 Formulations for preparing the silane-grafted NRs.

Amount (phr)
Chemicals Varying temp. Varying initiator Varying types & amounts of
100-160C* contents silane
RSS3 100.0 100.0 100.0 100.0
VP Si-363 20.0 20.0 10.0, 20.0 -
NXT - - - 10.0-20.0
Luperox® 231XL40 0.1 0.1-1.0 0.1 0.1
* The initial temperature setting of the mixer.

Table 7.2 Mixing procedures for preparing the silane-grafted NRs.

Cumulative time
Grafting procedures
(mins.)
- NR mastication 0
- Addition of silane coupling agent (NXT or VP Si-363) 1
- Addition of Luperox® 231XL40 (initiator) 2
- Discharge 12

7.2.3 Characterization of silane-grafted NRs

7.2.3.1 Gel content

0.5 g of the resulting materials of the melt grafting reactions was cut into small
pieces and dissolved in 50 ml of toluene for 72 hr under continuous stirring. The mixture
was then filtered through a filter paper to separate the gel component out of the solution,
and the gel part was dried in an oven for 24 hr at 107oC. The gel content was calculated
by using Equation (7.1):

Wg
Gel content (%)   100 (7.1)
Ws

where Wg is the weight of the dried gel component, and Ws is a weight of grafted

rubber sample.

7.2.3.2 Structural characterization by Fourier-Transform Infrared

Spectroscopy (FTIR)
An Attenuated Total Reflection (ATR) - FTIR spectrometer (Tensor 27, Bruker)
was used to characterize the purified grafted NR samples. Purification of the silane-

155
Chapter 7

grafted NR samples was performed by dissolving the materials in toluene under

continuous stirring for 72 hr, then filtering to remove any insoluble part, and subsequently
precipitating the soluble part in ethanol to remove any free or ungrafted silane. The
purified sample was finally dried in an oven at 60oC for 24 hr and kept in a descicator
before the analysis. The extent of silane grafting (R) onto the NR molecules was
calculated by using the peak height ratios according to Equation (7.2):[19]

A1075 A1035
R1075  and R1035  (7.2)
A1375 A1375

where R1075 and R1035 are the extents of silane-grafted NR at 1075 cm-1 and 1035 cm-1,
repectively, A1075 and A1035 are the peak heights of Si-O-C formation at 1075 cm-1 and

1035 cm-1, respectively, and A1375 is the peak height of -CH3 in NR at 1375 cm-1.

7.2.3.3 Structural characterization by Proton Nuclear Magnetic Resonance

1
( H-NMR) Spectroscopy

Purified grafted NR samples were prepared using the same method as described
in Section 7.2.3.2 for FTIR. The purified and dried grafted NR samples were dissolved in
deuterated chloroform (CDCl3) and characterized by NMR spectroscopy (Varian Unity
Inova 500 MHz, Varian). The amount of VP Si-363 grafted onto NR molecules in mol%
can be quantitatively analyzed by using Equation (7.3):

A4.03.5 / 46
mol % of grafted VP Si  363 on NR   100 (7.3)
( A4.03.5 / 46)  A5.1

where A4.03.5 is the integrated peak area of methylene protons of the alkoxy groups (–Si-

O-CH2-CH3 and -Si-O-(C2H4O)5-CH2-C12H25) of the VP Si-363 in the range 4.0-3.5 ppm,

CH3
and A5.1 is the integrated peak area of alkene protons ( C CH ) of NR at 5.1 ppm.

There are a total of 46 alkoxymethylene protons in each VP Si-363 molecule (as the
structure shown in Table 6.1), and 1 alkene proton in each isoprene unit.

The amount of NXT grafted onto NR molecules in mol% can be quantitatively

analyzed by using Equation (7.4):

156
 Preparation and characterization of silane-grafted NR

A3.8 / 6
mol % of grafted NXT on NR   100 (7.4)
( A3.8 / 6)  A5.1

where A3.8 is the integrated peak area of methylene protons of the alkoxy group (-Si-O-

CH2-CH3) of the NXT at 3.8 ppm, and A5.1 is the integrated peak area of alkene protons
CH3
( C CH ) of NR at 5.1 ppm. There are 6 alkoxy-methylene protons in each NXT
molecule and 1 alkene proton in each isoprene unit.

7.2.3.4 Grafting efficiency of silane-grafted NR

Based on mol% of grafted silane on the NR obtained from the calculation

according to Equations (7.3) and (7.4), the weight% of grafted silane can be calculated by
using Equation (7.5):

mol% silane  MWsilane

W t% of grafted silane on NR   100
( mol% silane  MWsilane)  ( mol% NR  MWNR )

(7.5)

where mol%silane is the mol% of grafted silane on NR, MWsilane is the molecular weight of
the silane, mol%NR is the mol% of NR and MWNR is the molecular weight of the NR
repeating unit (68 g/mol).

Then, the grafting efficiency can be calculated according to Equation (7.6):

wt % of grafted silane on NR
Grafting efficiency (%)   100 (7.6)
wt % of silane used for the reaction

7.2.3.5 Elemental analysis by Scanning Electron Microscopy - Energy

Dispersive X-Ray Spectroscopy (SEM-EDX)

The purified silane-grafted NRs were analyzed for the chemical elements carbon
(C), oxygen (O) and silicon (Si) by using SEM-EDX spectroscopy (Quanta 400, FEI).

157
Chapter 7

7.3 Results and discussion

7.3.1 Characterization of NR, NXT and VP Si-363 as starting materials

Before the grafting reaction, the starting materials: rubber and silanes were
characterized for their chemical structures by FTIR and 1H-NMR spectroscopy in order to
have the reference spectra, as respectively shown in Figures 7.2 and 7.3.

Virgin NR

NXT

VP Si-363

838
Transmittance (a.u.)

467
1467

2930
2875 1695

788
960

1107
1105

4000 3600 3200 2800 2400 2000 1600 1200 800 400
Wave number (cm-1)

Figure 7.2 ATR-FTIR spectra of NR, NXT and VP Si-363.

Figure 7.2 shows the FTIR spectra of the starting materials: unmodified NR, NXT
and VP Si-363 silane coupling agents, in the wave number range of 4000-400 cm-1. The
assignments for the main absorption bands of NR and the two types of silane are listed in
Table 7.3.[20-23]

Table 7.3 Wave number and assignments for NR, NXT and VP Si-363.
Wave number (cm-1) Functional groups
2930 C-H asymmetric stretching
2875 C-H symmetric stretching
1695 C=O stretching
1670 C=C stretching
1467 C-H bending
~1100 Si-O-C stretching
960 Si-O-C stretching
838 C-H out of plane deformation of NR
788 Si-O-C vibration of alkoxy-silane
467 O-Si-O deformation

158
 Preparation and characterization of silane-grafted NR

The characteristic peaks of NR are displayed at the wave numbers of 1670 and
-1
838 cm which are assigned respectively to the C=C stretching vibration and =C-H out of
plane deformation of the isoprene unit in NR molecules. For the NXT mercaptosilane, in
addition to the C-H stretching vibrations at 2930 and 2875 cm-1 and C-H bending
-1
deformations at 1467 cm , strong characteristic peaks of NXT are observed at the wave
-1
number of 1695 cm due to C=O deformation of the blocking group, and at ~1100, 960
and 788 cm-1 due to the vibrations of the Si-O-C functional group of the silane.
Furthermore, the spectrum of NXT shows an absorption peak at 467 cm-1 assigned to O-
Si-O deformation.[21] VP Si-363 is a mercapto-silane with one ethoxy group and two long
bulky groups comprising of long aliphatic chains and ethylene oxide units. The spectrum
of VP Si-363 shows strong absorption peaks of the Si-O-C functional group at the wave
numbers of ~1100, 960 and 788 cm-1.

a b
CH3 CH2O
c d e f g h i
CH3 (CH2)12 (OC2H4)5O Si CH2 CH2 CH2 SH
e g a
CH3 (CH2)12 (OC2H4)5O

d c
b
h i f
VP Si-363
a b a
CH3 CH2O O
c d e f g h h
CH3 CH2O Si CH2 CH2 CH2 S C CH2 CH2 CH3 b
5
CH3 CH2O
f e
g c
d
NXT
b c
H3C H a,d b
a C C d
H2C CH2
n

c
NR
9 8 7 6 5 4 3 2 1 0
ppm

Figure 7.3 1H-NMR spectra of NR, NXT and VP Si-363.

The 1H-NMR spectra of pure NR, NXT and VP Si-363 silanes are comparatively
shown in Figure 7.3. NR has specific absorption bands at chemical shifts of 5.1, 2.0 and
1.6 ppm assigned to the alkene proton (=CH), the methylene proton (-CH2-) and methyl
protons (-CH3) of polyisoprene, respectively. The NXT and VP Si-363 silanes have 1H-
NMR spectra showing more characteristics peaks assigned to different proton types in

159
Chapter 7

their structures as indicated in Figure 7.3. The main peaks of methylene- and methyl-
protons of the ethoxy groups appear at 3.8 and 1.2 ppm, respectively.[24,25] The 1H-NMR
spectrum of NXT shows the absorption bands of the methylene proton adjacent to the
sulfur atom at 2.45 ppm and to the carboxyl-group at 2.8 ppm. The methylene-proton
adjacent to –Si is displayed at 0.7 ppm. The long alkoxy-groups of VP Si-363 silane
displays a strong specific absorption band at a chemical shift of 3.6 ppm of the methylene-
proton of the ethylene oxide unit (-OCH2CH2-). Assignments for each peak are detailed in
Figure 7.3. Due to the presence of several methylene protons (-CH2-) attached to oxygen
atoms in the structure of VP Si-363, its absorption bands in the region of 4.0-3.5 ppm are
more complicated compared to the spectrum of NXT. A small change in the chemical
environment around -O-CH2- causes a slight shift of the peak and so more peaks
associated with the methylene protons of the ethoxy and ethylene oxide groups are
observed for the case of VP Si-363.

7.3.2 Preparation and characterization of silane-grafted NR

Grafting of silane onto polymers has been reported employing different reaction
conditions. Silane grafted onto polybutadiene was done in solution under nitrogen
atmosphere.[17,18,26]. The solution state is known to provide a better control of the reaction
compared to the melt state, but it is environmentally unfriendly and provides only a small
amount of the grafted-polymer per batch. The reaction in solution is therefore not
preferred for industrial scale. VTES grafted onto styrene butadiene rubber (SBR) had
been studied in the latex state using benzoic peroxide as initiator.[27] Furthermore, silane
grafting in the melt state has been reported for vinyl-silane-grafted polyolefins.[28,29] For
practical purposes, the present work therefore prepared silane-grafted NR by melt mixing
in an internal mixer using 1,1′-di(tert-butylperoxy)-3,3,5-trimethylcyclohexane or Luperox®
231XL40 as initiator. The half-life time of Luperox® 231 at 140oC is about 5 minutes which
is shorter than that of dicumyl peroxide (DCP) with its half-life time of about 30 minutes at
the same temperature. The shorter half-life time of Luperox® 231 allows the grafting
process to be conducted at lower temperature and shorter reaction time compared to the
use of DCP to prevent thermal degradation of the NR.

7.3.2.1 Effect of grafting temperature and initiator concentration

Grafting conditions, i.e. the grafting temperature and initiator concentration, were
investigated by using VP Si-363 silane at 20 phr. First, the grafting temperature was
varied in the range of 100 to 160oC while the initiator content was fixed at 0.1 phr. Then,

160
 Preparation and characterization of silane-grafted NR

the selected grafting temperature was fixed and the initiator contents were varied from 0.1
to 1.0 phr. The resulting materials were characterized for gel content and structure to
confirm the presence of grafted silane on the NR.

4
Gel content (%)

0
80 100 120 140 160 180
Grafting temperature (oC)

Figure 7.4 Gel content of VP Si-363-grafted NR vs. different grafting temperatures.

Gel contents of VP Si-363-grafted NRs as shown in Figure 7.4 increase with

increasing grafting temperature, caused by the increases of the reaction efficiency of the
initiator as well as the possibility of the macroradicals to form crosslinks. As mentioned
previously, Luperox® 231 has the half-life time of 5 min at 140oC. Increasing the reaction
temperature therefore generates more free radicals in shorter time, and causes a
crosslinking reaction between the rubber molecules in competition with the grafting
reaction, leading to the increase of gel content in the material.

161
Chapter 7

Virgin NR

NR-g-VP Si-363 at 100oC

NR-g-VP Si-363 at 120oC

Transmittance (a.u.)

NR-g-VP Si-363 at 140oC

NR-g-VP Si-363 at 160oC

3270 1710
1670 1125 1035
1075
565

2848 1375
2908 838
2956 1467

3900 3400 2900 2400 1900 1400 900 400

Wave number (cm-1)

Figure 7.5 ATR-FTIR spectra of VP Si-363-grafted NR with different grafting temperatures.

0.6
(a) (b)
NR

100oC
0.5
Peak height ratio

120oC 0.4
Transmittance (a.u.)

140oC
0.3

160oC
0.2
1125
1035 A1075/A1375
1075
0.1 A1075/A838

1375
838
0.0
1300 1100 900 700 80 100 120 140 160 180
Wave number (cm-1) Temperature (oC)

Figure 7.6 FTIR spectra in the range of 1400-700 cm-1 (a) and peak height ratios (b) of VP
Si-363-grafted NR with different grafting temperatures.

From the FTIR spectra in Figure 7.5, the peak intensity at the wave number of
-1
1075 cm associated with the vibration of Si-O-C bonds is increased when the grafting

162
 Preparation and characterization of silane-grafted NR

temperature was increased. Jiao et al.[23] studied silane grafted ethylene-octene

copolymer and reported that the peaks at 1167, 1105, 1082 and 958 cm-1 could be
assigned to the deformation of Si-O-CH2CH3 of the silane coupling agent. In this finger
print region as seen in Figures 7.5 and 7.6(a), virgin NR also shows absorption bands in
-1
the range of 1400-700 cm , so it is not easy to identify the Si-O-C deformations of the
silane grafted NR. However, the peak intensity at 1075 cm-1 relative to the neighboring
peaks at 1125 and 1035 cm-1 has changed indicating that some reactions have taken
place between silane and rubber. It can be noticed that the intensity of the peak at 1075
cm-1 is becoming stronger when the reaction temperature is increased especially at
140C. This peak is due to the Si-O-C deformation which confirms the presence of silane
attached to the NR molecules. A weak peak at 3270 cm-1 is also observed in the silane-
grafted NR that can be assigned to Si-O-H deformations due to hydrolysis of the alkoxy-
group in the silane structure during the grafting reaction or during the post-treatment. The
VP Si-363-grafted NR prepared at 160oC additionally shows an absorption peak at 1710
cm-1 due to the vibration of a carbonyl group (C=O), suggesting the occurrence of thermal
degradation of the NR. The peak height ratios at A1075/A1375 and A1075/A838 as shown in
Figure 7.6(b) are used to indicate the extent of the grafting reaction. The increase of
grafting temperature from 100oC to 140oC increases the peak height ratios, but the further
increase of temperature to 160oC decreases that ratios due to the competitive crosslinking
and degradation reactions. Based on the peak height ratios and the absence of
degradation, the grafting temperature at 140oC seems the most suitable condition, and
this temperature was therefore selected for a further study on the effect of initiator
concentration.

163
Chapter 7

60

50

40

Gel content (%) 30

20

10

0
0.0 0.3 0.6 0.9 1.2
Initiator concentration (phr)

Figure 7.7 Gel content of VP Si-363-grafted NR with different initiator concentrations.

Increasing of initiator concentrations for the melt mixing of silane and NR leads to
an increase of gel content in the grafted NR as shown in Figure 7.7. The gel content
sharpy increases when 0.3 phr of initiator was used and progressively rises with a further
increase of peroxide content. When a high amount of initiator is used, a large number of
reactive free radicals is generated and leads to a faster rate of termination as well as a
crosslinking reaction without enhancing the grafting efficiency. The effect of increasing
initiator concentration on the grafting reaction has also been reported to not only lead to
an increase of the termination rate, but also to a slight decrease of grafting efficiency.[30,31]
With a high amount of reactive free radicals in the system, rubber crosslinks can easily be
formed through C-C linkages generating gel or an insoluble part of the rubber network,
and causing more difficulties for the grafting reaction to take place.

164
 Preparation and characterization of silane-grafted NR

Virgin NR

Without initiator

Initiator 0.1 phr

Transmittance (a.u.)

Initiator 0.3 phr

Initiator 0.5 phr

3270
1670 1125
1035
1075 565

2848 1375
2956 2908 838
1467

3900 3400 2900 2400 1900 1400 900 400

Wave number (cm-1)

Figure 7.8 ATR-FTIR spectra of VP Si-363-grafted NR with different initiator

concentrations at 140C.

0.6
(a) (b)
NR

0.5
Without initiator
Peak height ratio

0.1 phr 0.4

Transmittance (a.u.)

0.3 phr
0.3
0.5 phr
0.2
1125
A1075/A1375
1035
1075 A1075/A838
0.1
1375
838
0.0
1300 1100 900 700 0 0.2 0.4 0.6
Wave number (cm-1) Initiator concentration (phr)

Figure 7.9 FTIR spectra in the range of 1400-700 cm-1 (a) and peak height ratios (b) of VP
Si-363-grafted NR with different initiator concentrations.

165
Chapter 7

The purified silane-grafted NRs that were prepared either without or with initiator
show a change of the peak intensity at 1075 cm-1 of Si-O-C bonds relative to the
neighboring peaks at 1127 and 1035 cm-1, compared to that of the virgin NR. A slight
increase of the peak height ratios of A1075/A1375 and A1075/A838 when 0.1 phr of initiator was
added into the grafting system compared to that without initiator is shown in Figures 7.8
and 7.9, indicating a slightly larger extent of grafting reactions. The grafted materials also
show a weak peak of -OH at a wave number of 3270 cm-1 suggesting the presence of
hydrolyzed alkoxy groups. However, the further increase of initiator concentration
decreases the peak height ratio to more or less the same level as that of the material
without initiator as seen in Figure 7.9(b), due to the fact that the high amount of generated
radicals tends to generate more crosslinking between the rubber chains than grafting of
silane onto the rubber molecules, as seen in the high gel contents in Figure 7.7. Moreover,
the absorption intensity at 1670 cm-1 due to the C=C deformation of the NR molecules
tends to decrease with increasing initiator concentrations. The decreased C=C double
bonds are most likely due to the crosslinking reaction as well as the grafting reaction in
NR via the addition at the unsaturated C=C bonds in the NR structure. In this case, the
decrease of C=C peak intensity at higher initiator content shall be mainly because of the
crosslinking reaction, in correspondence with the increasing gel content in Figure 7.7.
Based on the varied temperatures and initiator contents, a grafting temperature
o
at 140 C with 0.1 phr of initiator was selected to prepare the silane-grafted NR’s by using
10 and 20 phr of VP Si-363 and NXT. The resulting products will be additionally
characterized by 1H-NMR and elemental analysis.

7.3.2.2 VP Si-363-grafted NR

The grafting of NR with VP Si-363 using two different silane loadings; 10 and 20
phr, results in the materials having the FTIR and 1H-NMR spectra as shown in Figures
7.10 and 7.11, respectively. The spectrum of virgin NR is also included in the figures for
comparison.

166
 Preparation and characterization of silane-grafted NR

Virgin NR

VP Si-363 10 phr

VP Si-363 20 phr
Transmittance (a.u.)

3270

1125
1670 1035
1075 565

2848 1375

2908 838
2956 1467

3900 3400 2900 2400 1900 1400 900 400

Wave number (cm-1)

Figure 7.10 ATR-FTIR spectra of virgin NR and NR-grafted with VP Si-363 at 10 and 20
phr.

Similar to Figure 7.5, the NR grafted with VP Si-363 shows the characteristic
peaks of silane on NR at the wave number of 1075 cm-1 which is assigned to the Si-O-C
bond, and at 1035 cm-1 due to the Si-O-Si vibration. The Si-O-Si bond is the result of a
condensation reaction between the alkoxy groups. For comparison, the extent of silane
grafting (R) can be calculated by using the peak height ratios taking the peak of the -CH3
vibration in the NR structure at 1375 cm-1 as an internal standard, as also applied for other
silane grafted polymers.[19] The calculated results are given in Table 7.4. With increasing
silane content used for the reaction, the peak height ratios at the wave numbers of 1075
cm-1 and 1035 cm-1 relative to 1375 cm-1 increase. On comparing with virgin NR, the
extent of silane grafting or R-values of the grafted NR clearly increase with increasing
silane content used for the melt grafting, providing the proof that the silane is successfully
grafted onto the NR molecules with, to some extent, the alkoxy-groups reacted with each
other to form siloxane linkages (Si-O-Si) during the grafting reaction. Moreover, the
grafted NRs also show a weak intensity peak of -OH at 3270 cm-1, that may indicate the
presence of Si-OH due to hydrolysis of the alkoxy-group of the silane by moisture.

167
Chapter 7

H3C H a
C C
H2C CH
n

S b
OCH2CH3
C3H6 Si O(CH2CH2O)5 CH2 C12H25
O(CH2CH2O)5 CH2 C12H25

a
b
VP Si-363 20 phr

VP Si-363 10 phr

Virgin NR

9 8 7 6 5 4 3 2 1 0
ppm

Figure 7.11 1H-NMR spectra of virgin NR and NR-grafted with VP Si-363 at 10 and 20 phr.

Table 7.4 The extent of VP Si-363 grafting (R) onto NR molecules from ATR-FTIR and
mol% of VP Si-363 on NR molecules from 1H-NMR.
R value from ATR-FTIR Mol% of VP Si-
Amount of VP Si-
R1075=A1075/A1375 R1035=A1035/A1375 363
363 (phr)
from 1H NMR
0 phr (Virgin NR) 0.31 0.28 0.00%
10 phr 0.41 0.31 0.11%
20 phr 0.53 0.41 0.22%

Table 7.5 Grafting efficiency of VP Si-363 grafted onto NR molecules.

Amount of VP Si- Amount of VP Si- Amount of grafted Grafting efficiency
363 (phr) 363 used (wt%) VP Si-363 (wt%) (%)
10 9.09 1.57 17.25
20 16.67 3.10 18.60

The 1H-NMR spectra of the purified VP Si-363-grafted NRs in Figure 7.11 show
the characteristic peak of silane at the chemical shift at 3.6 ppm assigned to the
methylene proton of the alkoxy-group and the ethylene oxide repeating unit in the silane
structure. The peak intensity increases with rising silane content indicating the increase of
grafted silane on the NR chains. By taking the integrated peak area of the –Si-O-CH2-CH3

168
 Preparation and characterization of silane-grafted NR

and -Si-O-(C2H4O)5-CH2-C12H25 of the grafted silane in the range of 4.0-3.5 ppm and of
CH3
C CH of NR at 5.1 ppm, the mole% and wt% of VP Si-363 grafted on NR were
calculated according to Equations (7.3) and (7.5), and the results are shown in Tables 7.4
and 7.5, respectively. With 10 and 20 phr of VP Si-363 used for the reaction, grafted
silanes of 0.11 and 0.22 mol% or 1.57 and 3.10 wt% on the NR molecules are obtained
giving grafting efficiencies of 17.25 and 18.60%, respectively.
The SEM-EDX technique was additionally used to confirm the presence of Si-
atoms in the grafted rubber and the analysis results of the purified samples are shown in
Table 7.6.

Table 7.6 Chemical elements of VP Si-363-grafted NRs from SEM-EDX.

Chemical elements (wt%)
Sample
C O Si
Virgin NR 98.2 1.8 -
NR grafted with VP Si-363 10 phr 94.1 5.8 0.1
NR grafted with VP Si-363 20 phr 88.6 11.2 0.2

The elemental analysis data of the purified VP Si-363-grafted NRs reveals an

increase of oxygen and silicon contents with increasing silane concentrations used in the
reaction as shown in Table 7.6. Virgin NR contains no Si-atoms and a small amount of
oxygen that is naturally present in the natural rubber. The detectable Si-component in the
NR grafted with VP Si-363 confirms the presence of the silane fragment bonded to NR
after all unreacted silane was removed in the purification step.
Under the grafting conditions, the peroxide initiator is decomposed giving the
reactive radicals as shown in Scheme 7.1, and a possible reaction mechanism of the VP
Si-363 grafting onto NR molecules in the presence of radicals is shown in Scheme 7.2.

CH3 CH3
CH3 C O O O O C CH3 O O
CH3 CH3
CH3
T 2 O C CH3 (I)
CH3 CH3 CH3
CH3 CH3
CH3 CH3
Fragmentation

O
2 CH3 C CH3 2 CH3 (II)

Scheme 7.1 Decomposition of 1,1′-di(tert-butylperoxy)-3,3,5-trimethylcyclohexane

(Luperox® 231XL40).

169
Chapter 7

CH3 OCH2CH3
O C CH3 or CH3 HS C3H6 Si O(CH2CH2O)5C13H27
CH3 O(CH2CH2O)5C13H27
(I) (II)

OCH2CH3 CH3
S C3H6 Si O(CH2CH2O)5C13H27 CH4 or HO C CH3
O(CH2CH2O)5C13H27 CH3

H3C
C CH
H2C CH2
n

H3C H3C
C CH C CH2
H2C CH CH2 CH2
n n
S S
OCH2CH3 OCH2CH3
C3H6 Si O(CH2CH2O)5C13H27 C3H6 Si O(CH2CH2O)5C13H27
O(CH2CH2O)5C13H27 O(CH2CH2O)5C13H27

Scheme 7.2 Postulated reaction mechanism of VP Si-363 grafting onto NR molecules in

the presence of radical species.

The Luperox® 231XL40 is a diperoxide initiator that can be thermally

decomposed to give two free radicals or radical species I (OC(CH3)3) per molecule. The
tert-butoxy radicals can further spit to create more reactive methyl radicals (CH3) under
the high grafting temperature, as shown in Scheme 7.1. It is probable that both types of
free radicals formed in the system[32], can initiate the grafting reaction on the NR which
may proceed through two different pathways via H-abstraction and addition[33], as shown
in Scheme 7.2. The reactive initiator transfers its radical to the sulfur on the coupling agent
which subsequently attacks a double bond of the polymer molecule. The grafting reaction
tends to proceed then through an abstraction reaction at the allylic position of the NR
structure, more than through the addition reaction like with common sulfur
[34]
vulcanization. Upon successful grafting a covalent bond C-S-C bond between rubber
and silane is generated with the pendant alkoxy groups.

170
 Preparation and characterization of silane-grafted NR

7.3.2.3 NXT-grafted NR

The FTIR spectra of the purified NR grafted in presence of 10 and 20 phr of NXT
are shown in comparison with that of virgin NR in Figure 7.12.

Virgin NR

NXT 10 phr
Transmittance (a.u.)

NXT 20 phr

3270
1125
1670 1075
1035
1010 565

2848
1375
838
2908
2956 1467

4000 3600 3200 2800 2400 2000 1600 1200 800 400
Wave number (cm-1)

Figure 7.12 ATR-FTIR spectra of virgin NR and NR grafted with NXT at 10 and 20 phr.

The grafting of NXT silane onto the NR molecules was performed using the same
conditions as applied to prepare VP Si-363-grafted NR. The blocked mercapto-group of
NXT should be more thermally stable than –SH of VP Si-363, but because the steric
effects are also different in NXT and VP Si-363, the reactivity towards the grafting reaction
cannot be simply compared. Once the mercapto function is activated, the NXT with three
ethoxy groups should be more reactive than VP Si-363 with two bulky long alkoxy-groups.
The FTIR spectra in Figure 7.12 clearly show absorption bands at 3270 cm-1 due to O-H
stretching, and at 1075 and 1035 cm-1 assigned to Si-O-C and Si-O-Si deformations,
respectively, as previously observed in the spectra of the VP Si-363-grafted NR in Figure
7.10. For NR-g-NXT, there is an additional strong band at 1010 cm-1 which may be
assigned to the Si-OH group[35] due to hydrolysis of the ethoxy groups. Compared with the
spectra of NR-g-VP Si-363 in Figure 7.10, the strong increase of peak intensity at 1035
and 1010 cm-1 may be attributed to the faster reaction of the tri-ethoxy groups of NXT
compared to the two bulky alkoxy groups with only one ethoxy group of VP Si-363. The
hydrolysis of ethoxy groups by moisture during the grafting reaction leads to hydroxyl

171
Chapter 7

groups Si-OH which can further form a crosslink Si-O-Si in the grafted rubber. By taking
the peak height at 1075 and 1035 cm-1 relative to the peak at 1375 cm-1 as internal
standard, the increased amount of the silane used for the grafting reaction with NR results
in the increase of peak height ratios or R values as shown in Table 7.7, indicating the
presence of more grafted silane fragments.

H3C H a
C C
H2C CH
n
S
OCH2CH3
C3H6 Si OCH2CH3
OCH2CH3

a b

b
NXT 20 phr

NXT 10 phr

Virgin NR

9 8 7 6 5 4 3 2 1 0
ppm
1
Figure 7.13 H-NMR spectra of virgin NR and NR grafted with NXT at 10 and 20 phr.

Table 7.7 The extent of NXT grafting (R) onto NR molecules from ATR-FTIR and mol% of
NXT on NR molecules from 1H-NMR.
Amount of NXT R value from ATR-FTIR Mol% of NXT
(phr) R1075=A1075/A1375 R1035=A1035/A1375 from 1H NMR
0 phr (Virgin NR) 0.31 0.22 0.00%
10 phr 0.49 0.42 0.66%
20 phr 0.55 0.49 1.32%

Table 7.8 Grafting efficiency of NXT grafted onto NR molecules.

Amount of NXT Amount of NXT Amount of grafted Grafting efficiency
(phr) used (wt%) NXT (wt%) (%)
10 9.09 3.43 37.69
20 16.67 6.68 40.07

172
 Preparation and characterization of silane-grafted NR

Figure 7.13 shows 1H-NMR spectra of the purified NXT-grafted NRs prepared by
using two different NXT amounts of 10 and 20 phr in comparison with that of virgin NR.
The grafted rubber shows a characteristic peak of NXT-silane at a chemical shift of 3.8
ppm assigned to the proton of the ethoxy-groups of the silane. This peak is intensified with
increasing silane content and the mol% of NXT-silane in the grafted NR as calculated
according to Equation (7.4) is shown in Table 7.7. The grafted silane on the NR molecules
increases with increasing amount of silane used for the reaction, and the mol% as well as
the grafting efficiency of NXT grafted on NR are twice that of VP Si-363 grafted on NR
(Table 7.8 vs. 7.5). Even thought the blocked mercapto-group of NXT may be more
difficult to activate than the –SH group of VP Si-363, once the –S is formed the higher
steric effect of the long alkoxy-groups in the VP Si-363 molecule may lead to more
difficulties to react with the NR molecules. The lower steric effect of the ethoxy-groups of
NXT makes it easier for the grafting reaction but may also lead to a loss of ethoxy
methylene-protons via hydrolysis and self-condensation, resulting in Si-OH and Si-O-Si
bonds, as seen by the strong increase of the FTIR peak intensities at 1035 and 1010 cm-1
in Figure 7.12.
The presence of the silane grafted on the NR molecules is confirmed by
elemental analysis, as reported in Table 7.9. The amounts of silane used for the grafting
reaction were at 10 and 20 phr, and the materials were purified to eliminate any unreacted
silane prior to analysis.

Table 7.9 Chemical elements of NXT-grafted NRs from SEM-EDX.

Chemical elements (wt%)
Sample
C O Si
Virgin NR 98.2 1.8 -
NR grafted with NXT 10 phr 97.5 2.4 0.1
NR grafted with NXT 20 phr 96.8 3.0 0.2

The grafting of silane fragments on the NR molecules is proved by the presence

of silicon atoms in the modified NR, and the increase of silane content used for the
reaction results in increase of Si and O components in the grafted materials. When
compared to the NR-g-VP Si-363 elemental analysis results in Table 7.6, the NR-g-NXT
shows the same Si-contents but lower O contents in accordance with their molecular
structures. The equal Si- contents could be due to experimental uncertainties at these low
levels.

173
Chapter 7

CH3 O OCH2CH3
O C CH3 or CH3 H3C H2C C S CH2 Si OCH2CH3
6 3
CH3 OCH2CH3
(I) (II)

OCH2CH3 O O CH3
S CH2 Si OCH2CH3 H3C H2C C CH3 or H3C H2C C O C CH3
3 6 6
OCH2CH3 CH3

H3C
C CH
H2C CH2
n

H3C H3C

C CH C CH2
H2C CH CH2 CH2
n S n
S OCH2CH3
OCH2CH3
C3H6 Si OCH2CH3
C3H6 Si OCH2CH3
OCH2CH3
OCH2CH3
Scheme 7.3 Postulated reaction mechanism of NXT grafting onto NR molecules in the
presence of radical species.

The possible reaction mechanism of NXT-grafting of NR as displayed in Scheme

7.3 involves two reaction pathways: abstraction and addition, like the reaction
mechanisms proposed for the grafting of VP Si-363 onto NR molecules in Scheme 7.2.
After the grafting reaction, the deblocked carboxylic fragment of NXT is left in the grafted
rubber as by-product either in the form of an aldehyde or carboxylic acid. The presence of
this by-product is not visible in the FTIR and 1H-NMR spectra because the materials were
purified prior to the analysis.

7.4 Conclusions
Silane-grafted natural rubber is successfully prepared under melt mixing
conditions in an internal mixer at 140oC using 0.1 phr of 1,1′-di(tert-butylperoxy)-3,3,5-
trimethylcyclohexane (Luperox® 231XL40) as initiator. The NXT-grafting and VP Si-363-
grafting of NR are comfirmed by elemental analysis via the SEM-EDX technique in which
the modified NRs have silicon (Si) atoms in their structures, and the amounts of oxygen
and silicon in the grafted NRs increase with increasing amount of silane used for the

174
 Preparation and characterization of silane-grafted NR

reaction. Both VP Si-363-grafted NR and NXT-grafted NR show infrared absorption bands

at 3270, 1075 and 1035 cm-1 which can be assigned to the deformations of -OH, Si-O-C
and Si-O-Si, respectively, and the 1H-NMR absorption band at 3.8-3.6 ppm assigned to
the methylene protons of alkoxy groups. By calculation using the integrated peak areas of
–OCH2- of alkoxy-groups of silane and =CH of NR, the use of 10 and 20 phr of silane
results in grafted VP Si-363 on NR of 0.11 and 0.22 mol% or grafting efficiency of ≈ 18%,
and grafted NXT of 0.66 and 1.32 mol% or grafting efficiency of ≈ 38%, respectively. The
differences in grafting efficiency may tentatively be related to the respective molecular
structures of the silanes.

7.5 References

1. L.T. Zhuravlev. Colloids Surf. A. 173, 1 (2000).

2. A. Hunsche, U. Görl, H.G. Koban, T. Lehmann, Kautsch. Gummi Kunstst., 51, 525
(1998).
3. L.A.E.M. Reuvekamp, J.W. ten Brinke, P.J. van Swaaij, J.W.M. Noordermeer,
Rubber Chem. Technol., 75, 187 (2002).
4. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 85, 277 (2012).
5. J.W. ten Brinke, S.C. Debnath, L.A.E.M. Reuvekamp, J.W.M. Noordermeer, Comp.
Sci. Technol., 63, 1165 (2003).
6. J.W. ten Brinke, P.J. van Swaaij, L.A.E.M. Reuvekamp, J.W.M. Noordermeer,
Kautsch. Gummi Kunstst., 55, 244 (2002).
7. T. Jesionowski, A. Krysztafkiewicz, Appl. Surf. Sci., 172, 18 (2001).
8. C.H. Lee, S.H. Park, W. Chung, J.Y. Kim, S.H. Kim, Colloids Surf. A: Physicochem.
Eng. Aspects, 384, 318 (2011).
9. B. Kitiyanan, J.H. O’Haver, J.H. Harwell, S. Osuwan, Langmuir, 12, 2162 (1996).
10. W.H. Waddell, J.H. O’Haver, L.R. Evans, J.H. Harwell, J. Appl. Polym. Sci., 55, 1627
(1995).
11. V. Thammathadanukul, J.H. O’Haver, J.H Harwell, S. Osuwan, N. Na-ranong, W.H.
Waddell, J. Appl. Polym. Sci., 59, 1741 (1996).
12. M. Tiwari, “Plasma Coating of Silica: A Key to Improved Dispersion and Properties of
Reinforced Elastomer Blends”, Ph.D. Thesis: 2010, Dept. of Elastomer Technology
and Engineering, Univ. of Twente, Enschede, the Netherlands.
13. G. Mathew, M.-Y. Huh, J.M. Rhee, M.-H. Lee, C. Nah, Polym. Adv. Technol., 15, 400
(2004).

175
Chapter 7

14. C. Nah, M.-Y. Huh, J.M. Rhee, T.-H. Yoon, Polym. Int., 51, 510 (2002).
15. M. Tiwari, J.W.M. Noordermeer, W.J.V. Ooij, W.K. Dierkes, Polym. Adv. Technol.,
19, 1672 (2008).
16. S. Jain, J.G.P. Goossens, M. Van Duin, Macromol. Symp., 233, 225 (2006).
17. G. Colucci, A.D. Gianni, R. Bongiovanni, A. Priola, F. Torello, S. Bracco, P. Sozzani,
Macromol. Symp., 296, 550 (2010).
18. A. Bertora, M. Castellano, E. Marsano, A. Turturro, M. Alessi, L. Conzatti, P.
Stagnaro, G. Colucci, A. Priola, Macromol. Mater. Eng., 296, 455 (2011).
19. Y.-T. Shieh, T.-H. Tsai, J. Appl. Polym. Sci., 69, 255 (1998).
20. K. Nakashima, P.H. Solomon, Infrared Absorption Spectroscopy, 2nd Edition,
Holdenday, San Francisco (1997).
21. J. Lin, J.A. Siddiqui, R.M. Ottenbrite. Polym. Adv. Technol., 12, 285 (2001).
22. K. Sirisinha, S. Chimdist, Polym. Test., 25, 518 (2006).
23. C. Jiao, Z. Wang, Z. Gui, Y. Hu, Eur. Polym. J., 41, 1204 (2005).
24. M.-C. Brochier Salon, P.-A. Mayle, M. Abdelmouleh, S. Boufi, M. N. Belgacem.
Colloid Surface A., 312, 83 (2008).
25. T. Ogasawara, A. Yoshino, H. Okabayashi., C. J. O’Connor. Colloid Surface A., 180,
317 (2001).
26. F. Schapman, J.P. Couvercelle, C. Bunel, Polymer, 39, 4955 (1998).
27. C. Yin, Q. Zhang, J. Gu, J. Zheng, G. Gong, T. Liang, H. Zhang, J. Appl. Polym. Sci.,
128, 2262 (2013).
28. K. Sirisinha, M. Boonkongkaew, S. Kositchaiyong, Polym. Test., 29, 958 (2010).
29. M. Beltrán, C. Mijangos, Polym. Eng. Sci., 40, 1534 (2000).
30. S. Lenka, P.L. Nayak, I.B. Mohanty, S.N. Mishra, J. Appl. Polym. Sci., 30, 2711
(1985).
31. L. Thiraphattaraphun, S. Kiatkamjornwong P. Prasassarakich, S. Damronglerd, J.
Appl. Polym. Sci., 81, 428 (2001).
32. S.S. Pesetskii, B. Jurkowski, Y.M. Krivoguz, K. Kelar, Polymer, 42, 469 (2001).
33. R.S. Lehrle, S.L. Willist, Polymer, 38, 5937 (1997).
34. P.C. de Oliveira, A.M. de Oliveira, A. Garcia, J. C. de Souza Barboza, C.A. de
Carvalha Zavaglia, A.M. dos Santos, Eur. Polym. J., 41, 1883 (2005).
35. M. Abdelmouleh, S. Boufi, M.N. Belgacem, A. Dufresne, Compos. Sci. Technol., 67,
1627 (2007).

176
 CHAPTER 8

Silane-Grafted-Natural Rubber as Compatibilizer in Silica-

Reinforced Natural Rubber

The effect of silane-grafted-natural rubber (NR) as compatibilizer in silica-filled

NR compounds is studied by using NXT- and VP Si-363-grafted-NRs prepared by using
10 and 20 phr of the silanes. The silane-grafted-NRs were used in a range of 5-20 phr, so
as to give 0.8-6.1 wt% silane functionality relative to the silica. The properties of the silica-
filled NR compounds compatibilized with the silane-grafted-NRs are studied in comparison
with those compatibilized by straight use of a silane TESPT at varying amounts in the
same range, and with 8.6 wt% relative to the silica. The use of the silane-grafted-NRs as
compatibilizers decreases the Mooney viscosity, Payne effect and silica flocculation,
improves filler-rubber interaction, and significantly increases the tensile modulus,
reinforcement index and tensile strength of the silica-filled NR-compounds compared to
the non-compatibilized one. At the same silane-content, the use of silane-grafted-NRs
gives slightly better properties than the straight use of the same silane, and the use of
NXT-grafted-NR as compatibilizer shows a better improvement of overall properties than
the use of VP Si-363-grafted NR. In the presence of silane-grafted-NRs as
compatibilizers, the glass transition temperatures (Tg) of the silica-filled NR vulcanizates
slightly shift to a higher temperature, while the tan δ values at 5oC remain more or less the
same and the tan δ values at 60oC decrease compared to that with TESPT and without
compatibilizer.
Chapter 8

8.1 Introduction

Chemical modification of natural rubber (NR) by epoxidation[1] has been widely

used to increase the polarity, improve some properties of the rubber and enhance the
compatibility with polar substances including silica. The silanol-groups on the silica
surface interact with the epoxide groups of ENR through hydrogen bonds during
processing and ring opening of the epoxide groups can form chemical bonds with silica via
a condensation reaction with the silanol groups.[2,3] Silica-filled epoxidized polybutadiene
(EpBR) and epoxidized styrene-butadiene rubber (EpSBR) show improvement in the
chemical interaction between the rubbers and silica as well as improved filler
dispersion.[4,5] A restriction on polymer chain movement by presence of the chemical
bonds between the epoxide-groups and silica resulted in a shift of the glass transition
temperature (Tg) of the silica-filled epoxidized rubber.[4-6]
An alternative pathway to enhance the compatibility between polymer and silica
is to use silane modified polymers instead of straight use of a silane. Grafting of silane
coupling agents onto polymer molecules has been reported by using different types of
silanes such as vinyltrimethoxysilane (VTMS)[7-9], vinyltriethoxysilane (VTES)[9-12] and
mercaptopropyl-trimethoxysilane.[12-13] Vinylsilanes such as VTMS and VTES are widely
used to modify polyolefins to improve the strength properties of the pure polymers via a
self-crosslinking reaction. The grafting of vinylsilanes onto the polyolefins was carried out
in melt condition using dicumyl peroxide (DCP) as initiator[7-10], and under inert conditions
[11]
at low temperature with azobisisobutyronitrile (AIBN) as initiator. Increased crosslink
density in the silane-grafted polyolefins leads to improvement in thermal stability and
tensile properties.[8,9] Polypropylene (PP) grafted with VTES and PP-graft-VTES/silica
nanocomposites were investigated in which the grafted silane affects the crystallization
temperature of PP and the impact strength of the composites.[11] VTES was grafted onto
styrene-butadiene rubber at a temperature of about 60-70oC by using potassium
persulfate (KPS)[12] or benzoyl peroxide (BPO)[13] as initiators under nitrogen atmosphere.
The use of SBR-g-VTES for a silica-filled compound led to improved cure and tensile
properties,[12,13] and a shift of the glass transition temperature[12] (Tg) to higher temperature
due to the good interactions between silica and SBR-g-VTES, that restrict movement of
the SBR-chains. Further, liquid polybutadiene (PB) has been modified with
mercaptopropyltrimethoxysilane and mercaptopropyltriethoxysilane to improve the
[14-15]
reinforcing efficiency of silica in this rubber. The mercapto-silane-modified PB was
o
prepared in solution at 75 C under inert atmosphere with AIBN as initiator, and the
grafting reaction proceeded through radical addition of the thiol group of the silane to the

178
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

double bond of PB, introducing alkoxy groups on the rubber chains. After a silanization
reaction of silica with the silane-grafted PB, the silica characteristics showed a more
hydrophobic surface, which was proved by an increase of contact angle[14] and a decrease
of hydroxyl groups on the silica surface.[15] The covalent bond Si-O-C of the silanized PB
was analyzed by proton-, carbon- and silicon nuclear magnetic resonance spectroscopy
(1H-NMR, 13 29
C-NMR and Si-NMR).[14,16] The better interaction of silica and rubber via
silane-grafted PB improved the silica dispersion and tensile strength.[15] Therefore, the use
of silane-modified polymer chains is an alternative method to improve bonding between
hydrophilic silica and hydrophobic hydrocarbon rubber.
This chapter reports the influence of NXT-grafted-NRs and VP Si-363-grafted-
NRs, i.e. NR-g-NXT and NR-g-VP Si-363, respectively, as compatibilizers in silica-filled
NR. The non-purified silane-grafted-NRs were used in a range of 5-20 phr and the
respective amounts of silanes were also straightly used in another set of compounds, for
comparison. The properties of both compounds and vulcanizates compatibilized with
silane and silane-grafted-NR are discussed in comparison with those without
compatibilizer and with reference TESPT.

8.2 Experimental

8.2.1 Materials and compound preparation

The compounding ingredients, i.e. NR (RSS3), TESPT, silica (Zeosil 1165MP),

TDAE oil, ZnO, TMQ, stearic acid, DPG, CBS and sulfur, used for preparation of the silica-
filled NR compounds in this chapter were as described in Chapter 3. NXT and VP Si-363
are as detailed in Chapter 5 and the information shown in Table 5.1. The NXT-grafted-NR
and VP Si-363-grafted-NR were prepared according to the methods described in Chapter
7.
The NXT- and VP Si-363-grafted-NRs, pre-prepared by using 10 and 20 phr of
silane, were used without purification in order to simplify the process. The amounts of
grafted-NXT and grafted-VP Si-363 on the NR as well as the grafting efficiencies are as
previously shown in Tables 7.8 and 7.5. To study the effect of NXT- and VP Si-363-
grafted-NRs as compatibilizers in silica-filled NR, the silane-grafted-NRs were varied at 5,
10, 15 and 20 phr as part of the rubber matrix. Total silane contents based on calculation,
either in the form of free silane and of grafted silane in the compatibilizers are summarized
in Table 8.1.

179
Chapter 8

Table 8.1 Total NXT and VP Si-363 contents including both reacted and unreacted parts
in different amounts of NXT- and VP Si-363-grafted-NR being used as compatibilizers in
the silica-filled NR compounds.
Amount of NXT- and VP Si-363-grafted-NR Amount of total silane
(wt% rel. to silica)
Silane-grafted-NR with 10 phr of silane
 5 phr 0.8
 10 phr 1.7
 15 phr 2.5
 20 phr 3.4
Silane-grafted-NR with 20 phr of silane
 5 phr 1.5
 10 phr 3.0
 15 phr 4.6
 20 phr 6.1

The silica-filled NR compound formulations with the various amounts of silane-

grafted-NRs and the straight silanes are shown in Table 8.2 together with the references.
The properties of the silica-filled NR compounds with the silane-grafted-NRs as
compatibilizers were studied in comparison with the results of silica-filled NR with straight
silanes at different amounts, as calculated based on the silane contents used in the
preparation of the silane-grafted rubber, as shown in Table 8.1. The mixing procedures
are given in Table 8.3. The mixer type and temperature setting are as described in
Chapter 5.

180
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

Table 8.2 Compound formulations.

Parts per hundred parts of rubber (phr)
Ingredients
References Silane-grafted-NRs Straight use of silane
RSS3 100.0 100.0 95.0-80.0 95.0-80.0 100.0 100.0
TESPT - 4.7* - - - -
NXT-grafted-NR** - - 5.0-20.0 - - -
VP Si-363-grafted-NR** - - - 5.0-20.0 - -
NXT - - - - 0.8-6.1*** -
VP Si-363 - - - - - 0.8-6.1***
Zeosil 1165MP 55.0 55.0 55.0 55.0 55.0 55.0
TDAE oil 8.0 8.0 8.0 8.0 8.0 8.0
ZnO 3.0 3.0 3.0 3.0 3.0 3.0
TMQ 1.0 1.0 1.0 1.0 1.0 1.0
Stearic acid 1.0 1.0 1.0 1.0 1.0 1.0
DPG 1.0 1.0 1.0 1.0 1.0 1.0
CBS 1.5 1.5 1.5 1.5 1.5 1.5
Sulfur 1.5 1.5 1.5 1.5 1.5 1.5
Remarks: * TESPT 4.7 phr equals 8.6 wt% rel. to silica; ** NXT- and VP Si-363-grafted-NRs were
prepared by using silane contents at 10 and 20 phr; *** Silane contents for straight use were
calculated based on silane loadings in the non-purified silane-grafted-NR as shown in Table 8.1.

Table 8.3 Mixing procedures.

Cumulative
Mixing procedures
time (mins.)
Step 1 : Internal mixer
- NR and silane-grafted-NR (or NR only) mastication 2
- Addition of first half of silica (and ½ of silane if any) 7
- Addition of second half of silica (and ½ silane if any) and TDAE oil 12
- Addition of ZnO, stearic acid and TMQ 15
Step 2 : Two roll mill
- Addition of DPG, CBS and sulfur 5

8.2.2 Property testing and characterization

Mooney viscosity of the compounds after step 2 and cure characteristics were
tested according to the methods as described in Chapter 5. Payne effect and flocculation
rate constant of the final compounds (with curatives) were analyzed according to the
methods as detailed in Chapter 3. Bound rubber content of the non-productive compounds
after step 1 was analyzed following the procedures as described in Chapter 3.
The compounds were press-cured to their respective optimum cure times (tc90)
and tensile as well as dynamic mechanical properties were tested following the methods

181
Chapter 8

as described in Chapter 5. Tensile fractured surfaces were analyzed by using the

scanning electron microscopy (SEM) technique, as described in Chapter 4.
The properties are reported as function of silane contents in the compounds
compatibilized by both silane-grafted-NRs and with straight use of silane.

8.3 Results and discussion

8.3.1 Effect of NXT-grafted-NR as compatibilizer and use of straight NXT silane

on the properties of silica-filled NR compounds

8.3.1.1 Mooney viscosity, bound rubber content, Payne effect and

flocculation rate constant
For filled rubber compounds, the viscosity increment depends on several factors,
to include mastication time[17], mixing time[18], filler types[19,20], filler contents[20,21],
compatibilizer or coupling agent[22,23] and storage time.[19,24] The change of compound
viscosity is strongly influenced by the levels of filler-filler and filler-rubber interactions
which consequently have an influence on the final properties of the filled rubber. The
decrease of compound viscosity results in a lower mixing torque and consequently less
heat generation or lower compound temperature.

90 60
(a) (b)
Chemically bound rubber contents (%)

80
50
70

60 40
ML1+4 (100oC)

50
30
40

30 20

20 None TESPT
None TESPT
10
10 NXT NR-g-NXT
NXT NR-g-NXT
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.1 Mooney viscosity (a) and bound rubber contents (b) of silica-filled NR
compounds with use of straight NXT (open symbols) and NXT-grafted-NR (closed
symbols).

182
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

Mooney viscosities of the silica-filled NR compounds decrease with increasing

amounts of silane either in the form of NR-g-NXT or straight used NXT, and both sets of
compounds show similar properties, as shown in Figure 8.1(a). The Mooney viscosity of
the filled-NR with NXT and NXT-grafted-NR with 6 wt% of silane relative to the silica,
decreases to almost the same level as that of the reference compound with TESPT at 8.6
wt% relative to the silica. This level of 8.6 wt% of TESPT relative to the silica was found to
be the optimum, as calculated based on the work described by Guy et al.[25] For silica-
filled compounds, the viscosity reduces when filler dispersion is improved.[23] Thus, the
incorporation of NXT or NXT-grafted-NR improves the silica dispersion and consequently
reduces the Mooney viscosity due to a diminished hydrophilic character of the silica after
the silanization reaction. As expected, the silica-filled NR compound without compatibilizer
shows the highest Mooney viscosity because of the strong filler-filler interactions via
hydrogen bonding of the silanol groups on the silica surfaces that lead to poor dispersion
and large agglomerates to obstruct flow.
Figure 8.1(b) shows the chemically bound rubber contents of silica-filled NR
compounds with different compatibilizers in which the use of either NR-g-NXT or straight
NXT significantly increases the bound rubber, i.e. increased silica-rubber interaction
compared to the non-compatibilized one, but still to a significantly lower level compared
with the reference compound with TESPT. The chemically bound rubber contents
increase with rising amount of NXT silane up to 3.4 wt% relative to the silica, thereafter
the values level off. The use of NXT-grafted-NR tends to show a marginally higher
chemically bound rubber content than the straight use of NXT, in accordance with the
Payne effect in Figure 8.2(a). The partially pre-bonded silane to NR may enhance the
compatibility between rubber and silica and allow the ethoxy-groups of NXT to react more
effectively with the silica during the silanization reaction. However, the use of TESPT in
the silica-filled NR compound still gives an outstanding chemically bound rubber content,
which is due to its tetrasulfide structure that results in pre-mature crosslinking with the NR
molecules[26] and the bis-triethoxy-groups to react with the silanol groups during
silanization.

183
Chapter 8

1800 0.35
Payne effect [G′(0.56%)-G′(100%)] (kPa) (a) (b)

Flocculation rate constant (ka) (s-1)

1600
None 0.30
1400 TESPT
NXT 0.25
1200
NR-g-NXT
1000 0.20

800 0.15
600
0.10
400 None TESPT
0.05
200 NXT NR-g-NXT

0 0.00
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.2 Payne effect (a) and flocculation rate constant (b) of silica-filled NR compounds
with straight use of NXT (open symbols) and NXT-grafted-NR (closed symbols).

The filler-filler interaction or Payne effect of the silica-filled NR compounds with

both NXT and NXT-grafted-NR is remarkably decreased with increasing silane content to
finally meet the same level of Payne effect of the compound with TESPT, as shown in
Figure 8.2(a). The ethoxy-groups of the NXT-silane undergo a silanization reaction with
the silanol-groups of the silica and reduce the filler-filler interactions. Increasing amounts
of NXT-silane further decrease the Payne effect of the compounds and the use of NXT-
grafted-NR gives even slightly lower Payne effects than the straight use of NXT.
The flocculation behavior of the silica in the rubber matrix is due to the different
polarity between silica and rubber causing re-agglomeration of silica aggregates or de-
mixing, which will influence the properties of the material.[27] The flocculation rate constant
(ka) is used to determine the degree of re-agglomeration of the silica to form a stable
silica-silica network under heat treatment.[28] At higher temperatures, silica aggregates can
migrate easier in the rubber matrix to reform agglomerates and re-create a filler-filler
network. The silica-filled compound without silane shows a strong silica flocculation
tendency as shown in Figure 8.2(b) due to hydrogen bonding between the silanol groups
on the silica surface. The addition of both NXT and NXT-grafted-NR at only a small
amount already reduces the flocculation rate of the silica-filled compounds, like in the case
of TESPT, due to reduction of the silica polarity by silanization of the silane onto the silica

184
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

surface. However, further increase of NXT-contents has no additional effect on the silica
flocculation rate constant.

8.3.1.2 Tensile properties

12 5
(a) (b)

Reinforcement index (M300/M100)

10
4
300% Modulus (MPa)

8
3
6
2
4

None TESPT
None TESPT 1
2
NXT NR-g-NXT NXT NR-g-NXT

0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.3 300% Modulus (a) and reinforcement index (b) of silica-filled NR with straight
use of NXT (open symbols) and NXT-grafted-NR (closed symbols).

The modulus at 300% elongation and reinforcement index of the silica-filled NR

vulcanizates compatibilized by straight use of NXT in comparison with the use of NXT-
grafted-NR are shown in Figures 8.3(a) and (b). The modulus of the vulcanized silica-filled
NRs significantly increases with rising NXT-contents in both cases due to increase of the
silica-rubber interactions and the use of NXT-grafted-NR as compatibilizer gives a higher
modulus than the straight use of NXT, in accordance with the higher chemically bound
rubber content: Figure 8.1(b), and lower Payne effect: Figure 8.2(a). The reinforcement
index of the silica-filled NR vulcanizates also increases as a function of NXT-content:
Figure 8.3(b), to reach an optimum at about 4.5 wt% of NXT relative to the silica, above
which the addition of either more NXT or NXT-grafted-NR results in more or less the same
values. In both cases the property levels of the TESPT-filled compound are not,
respectively nearly reached.

185
Chapter 8

30 800
(a) (b)
700
25

Elongation at break (%)

600
Tensile strength (MPa)

20
500

15 400

300
10
200
5 None TESPT None TESPT
100
NXT NR-g-NXT NXT NR-g-NXT
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.4 Tensile strength (a) and elongation at break (b) of silica-filled NR compounds
with straight use of NXT (open symbols) and NXT-grafted-NR (closed symbols).

The tensile strength and elongation at break of the silica-filled NR vulcanizates

with NXT-grafted-NR and straight NXT as compatibilizers show only small differences,
where the tensile strength reaches a level slightly lower than that of the vulcanizate with
8.6 wt% of TESPT relative to the silica. The use of either NXT or NXT-grafted-NR clearly
improves the tensile strength of the silica-filled NR, as a result of the increased silica-
rubber interactions by the coupling reaction of the NXT-silane. The tensile strength rises
progressively when the NXT-content is increased up to about 3.5 wt% relative to the silica,
and thereafter tends to level off. Like the 300% modulus in Figure 8.3(a), the use of NXT-
grafted-NR as compatibilizer gives a higher tensile strength than the straight use of NXT-
silane. The coupling between silica and rubber that leads to the increased modulus and
tensile strength affects the elongation at break slightly, decreasing a little as shown in
Figure 8.4(b).

186
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

8.3.1.3 Surface topography

(a) (b)

(c) (d)

(e) (f)

Figure 8.5 SEM micrographs of tensile fractured surfaces of silica-filled NR vulcanizates at

800x magnification: (a) without compatibilizer; (b) with TESPT at 8.6 wt% rel. to silica; (c)
with NXT at 3.4 wt% rel. to silica; (d) with NXT-grafted-NR containing total NXT of 3.4 wt%
rel. to silica; (e) with NXT at 6.1 wt% rel. to silica; and (f) with NXT-grafted-NR containing
total NXT of 6.1 wt% rel. to silica.

SEM micrographs of tensile fractured surfaces of silica-filled NR vulcanizates

without compatibilizer, with TESPT, with NXT and NXT-grafted-NR are compared in
Figures 8.5(a)-(f). As described in Chapters 3 and 4, a smooth failure surface of a silica-
filled vulcanizate is due to a weak filler-rubber interaction and poor silica dispersion,
whereas a rough failure surface with many tear lines from ductile failure indicates a good
silica-rubber interaction and high tensile strength. The SEM micrographs of the

187
Chapter 8

vulcanizates with no compatibilizer and with reference TESPT, as shown in Figure 8.5(a)
and (b) respectively, clearly correspond to their different tensile strengths: Figure 8.4(a).
The addition of NXT improves silica-rubber interactions to lead to more surface roughness
and tear lines on the failure surfaces when compared to the filled vulcanizate without
compatibilizer. The increasing amount of NXT silane from 3.4 to 6.1 wt% relative to the
silica results in ever more tear lines on the failure surfaces of the filled vulcanizates, as
shown in Figure 8.5(c)-(f), due to more filler-rubber interaction and the higher strength,
and the surface topography becomes more to resemble that of the reference compound
with TESPT. However, there is no clear difference in the failure surface patterns of the
filled vulcanizates prepared by the straight use of NXT and by the use of NXT-grafted-NR.

8.3.2 Effect of VP Si-363-grafted-NR as compatibilizer and straight use of VP Si-

363 on the properties of silica-filled NR compounds

8.3.2.1 Mooney viscosity, bound rubber contents, Payne effect and

flocculation rate constant

90 60
(a) (b)
Chemically bound rubber contents (%)

80
50
70
ML1+4 (100oC)

60 40

50
30
40

30 None 20
TESPT None
20 TESPT
VP Si-363 10
10 VP Si-363
NR-g-VP Si-363 NR-g-VP Si-363
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.6 Mooney viscosity (a) and chemically bound rubber contents (b) of silica-filled
NR with straight use of VP Si-363 (open symbols) and VP Si-363-grafted-NR (closed
symbols).

188
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

The Mooney viscosities of the silica-filled NR compounds are different depending

on the types and amounts of compatibilizers; as shown in Figure 8.6(a). The incorporation
of compatibilizers either in the form of straight VP Si-363 silane or VP Si-363-grafted-NR
dramatically decreases the Mooney viscosity because of the improvement in silica-rubber
interactions after the silanization reaction that lead to the improved silica dispersion, but
there is no difference in Mooney viscosities between the two sets of compounds.
Figure 8.6(b) shows the chemically bound rubber contents after treatment in
ammonia atmosphere to cleave all weak physical filler-rubber interactions.[29] The results
are similar to what has been observed in Figure 8.1(b) for NXT-silane: the increasing
amount of VP Si-363 increases the chemically bound rubber content in the compounds to
just a slightly lower level-off value than for NXT. As only some of the silanol groups will
react with the ethoxy groups of silane, the adsorbed silane on the silica surface will also
act as a shielding agent that only forms physical interactions between filler and rubber.
The chemically bound rubber content therefore reaches a saturated point, as observed in
Figures 8.1(b) and 8.6(b). The use of VP Si-363-grafted-NR shows a slightly higher
chemically bound rubber content than the straight use of VP Si-363 in the range of high
silane contents, i.e. > 3.5 wt% relative to the silica, due to some thiol-groups of the VP Si-
363 molecules already bonded to NR molecules prior to use, causing the higher extent of
silica-rubber interactions.

1800 0.35
(a) (b)
Payne effect [(G′(0.56%)-G′(100%)] (kPa)

Flocculation rate constant (k a) (s-1)

1600
0.30
1400
0.25
1200

1000 0.20

800 0.15
600 None
None 0.10
400 TESPT TESPT
VP Si-363 0.05 VP Si-363
200
NR-g-VP Si-363 NR-g-VP Si-363
0 0.00
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.7 Payne effect (a) and flocculation rate constant (b) of silica-filled NR with straight
use of VP Si-363 (open symbols) and VP Si-363-grafted-NR (closed symbols).

189
Chapter 8

The filler-filler interaction or Payne effect of the silica-filled NR compounds with

various amounts of VP Si-363 silane is shown in Figure 8.7(a). Increasing VP Si-363
contents progressively reduce the Payne effect, and the use of silane either in its original
form or in the silane-grafted-NR form results in similar properties. The flocculation rate
constant (ka) or re-agglomeration rate of the silica under heat treatment is reduced after
applying the compatibilizers, as shown in Figure 8.7(b), but the scatter in the data due to
the measurement error is too large to significantly discriminate between the two different
forms of silane. During the silanization reaction, while mixing the ethoxy groups of the
silane are condensed with the silanol groups of silica to form a covalent bond between
silica and silane if straight silane is used, or to possibly form a link between silica and
rubber if silane-grafted-rubber is used. Moreover, the bulky alkoxy-groups with oxygen
atoms of VP Si-363 can adsorb on the silica surface giving a shielding effect on the polar
surface. Therefore, the filler-filler interactions as well as flocculation rate constant are
decreased.

8.3.2.2 Tensile properties

12 5
(a) (b)
Reinforcement index (M300/M100)

10
4
300% Modulus (MPa)

8
3
6

2
4 None None

TESPT TESPT
1
2 VP Si-363 VP Si-363
NR-g-VP Si-363 NR-g-VP Si-363
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.8 300% Modulus (a) and reinforcement index (b) of silica-filled NR with straight
use of VP Si-363 (open symbols) and VP Si-363-grafted-NR (closed symbols).

The 300% modulus and reinforcement index (M300/M100) of the silica-filled NR

vulcanizates containing different amounts of VP Si-363 in the form of straight silane and

190
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

silane grafted NR, are shown in Figures 8.8(a) and (b) in comparison with those
containing TESPT and no compatibilizer. The two sets of compounds that contain VP Si-
363 and VP Si-363-grafted-NR show no significant difference in 300% modulus and
reinforcement index. The values increase with increasing VP Si-363 contents but to a
lower level compared to the use of TESPT. The increasing filler-rubber interactions as
derived form the chemically bound rubber contents: Figure 8.6(b), lead to a higher
resistance to deformation, but because there are no additional crosslinks generated by
sulfur donated to the compounds by the silane molecules like in the case of TESPT, the
VP Si-363 sillanized compounds therefore have a significantly lower modulus.

30 800
(a) (b)
700
25
Elongation at break (%) 600
Tensile strength (MPa)

20
500

15 400

300
10 None None

TESPT 200 TESPT

5 VP Si-363 VP Si-363
100
NR-g-VP Si-363 NR-g-VP Si-363
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.9 Tensile strength (a) and elongation at break (b) of silica-filled NR with straight
use of VP Si-363 (open symbols) and VP Si-363-grafted-NR (closed symbols).

The tensile strength of the silica-filled vulcanizates also increases as a function of

VP Si-363 content due to the effect-of the coupling reaction, where the use of VP Si-363-
grafted-NR as compatibilizer shows a better improvement in tensile strength especially at
high VP Si-363-grafted-NR contents compared to the use of straight silane, as seen in
Figure 8.9(a). The elongation at break of the silica-filled NR vulcanizates marginally
decreases with increasing amount of VP Si-363, as shown in Figure 8.9(b), due to
deformation restriction by silica-to-rubber linkages but the vulcanizates show slightly
higher elongation at break than the reference TESPT vulcanizate that has more crosslinks
by the sulfur released from its tetrasulfide structure.

191
Chapter 8

8.3.2.3 Surface topography

(a) (b)

(c) (d)

Figure 8.10 SEM micrographs of tensile fractured surfaces of silica-filled NR vulcanizates

at 800x magnification: (a) without compatibilizer; (b) with TESPT at 8.6 wt% rel. to silica;
(c) with VP Si-363 at 6.1 wt% rel. to silica; (d) with VP Si-363-grafted-NR containing total
silane of 6.1 wt% rel. to silica.

SEM images of the tensile fractured surface of silica-filled NR vulcanizates with

the use of straight VP Si-363 and VP Si-363 that was partially bonded to NR are shown in
Figures 8.10(c) and (d) in comparison with that of reference vulcanizates with no
compatibilizer and with TESPT in Figures 8.12(a) and (b). The vulcanizates with TESPT
and VP Si-363-grafted-NR show similar failure surfaces with many tear lines in
accordance with their similar tensile strength, whereas the one with the use of VP Si-363
shows a slightly less complicated fracture surface indicating a somewhat less ductile
behavior. The weakest silica-filled NR vulcanizate without compatibilizer in Figure 8.10(a)
shows a smooth fracture surface due to its poor filler-rubber interactions that lead to easy
sample rupture.

192
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

8.3.3 Comparative properties of silica-filled NR containing NXT-grafted-NR and

VP Si-363-grafted-NR as compatibilizers

The results in Sections 8.3.1 and 8.3.2 have shown that the use of silane-grafted
NRs, i.e. NXT-grafted-NR and VP Si-363-grafted-NR, as compatibilizers give somewhat
better overall properties of the silica-filled NR compounds compared to the straight use of
free silane. In the silane-grafted NRs, part of the silanes is already grafted on the NR-
molecules as proved in Chapter 7, and another part has remained free in the material. The
improvement in the filled-rubber properties comes from a good filler-rubber interaction via
the silanization reaction between the ethoxy-groups of the silanes and the silanol-groups
of silica during mixing, where the pre-bonded silane and rubber via the grafting reaction
may lead to a better coupling reaction during vulcanization that involves several chemicals
and a complex reaction path. This part provides a comparison of the influence of NXT-
and VP Si-363-grafted-NRs on the properties of silica-filled NR compounds. Mooney
viscosity, chemically bound rubber content, Payne effect, flocculation rate constant, 300%
modulus, reinforcement index, tensile strength, elongation at break and dynamic
mechanical properties are presented in comparison with those of the reference
compounds with TESPT and without compatibilizer.

90 60
(a) (b)
Chemically bound rubber contents (%)

80
50
70
ML1+4 (100oC)

60 40

50
30
40

30 None 20
TESPT None
20 TESPT
NR-g-NXT 10
10 NR-g-NXT
NR-g-VP Si-363 NR-g-VP Si-363
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.11 Mooney viscosity (a) and chemically bound rubber content (b) of silica-filled
NR with NXT-grafted-NR (open symbols) and VP Si-363-grafted-NR (closed symbols).

193
Chapter 8

Mooney viscosities of the silica-filled NR compounds with NXT- and VP Si-363-

grafted-NRs decrease with increasing amounts of the compatibilizers, and both types of
the silane-grafted NRs give very similar viscosity levels as shown in Figure 8.11(a). Figure
8.11(b) shows the chemically bound rubber contents of the silica-filled NR compounds in
which the compounds with NXT-grafted-NR show a higher chemically bound rubber
content than those with VP Si-363-grafted-NR.

2000 0.35
None (a) (b)
Payne effect [(G′(0.56%)-G′(100%)] (kPa)

Flocculation rate constant (k a) (s-1)

1800 TESPT
0.30
1600 NR-g-NXT
NR-g-VP Si-363 0.25
1400
1200
0.20
1000
800 0.15

600 None
0.10
TESPT
400
0.05 NR-g-NXT
200
NR-g-VP Si-363
0 0.00
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.12 Payne effect (a) and flocculation rate constant (b) of silica-filled NR with NXT-
grafted-NR (open symbols) and VP Si-363-grafted-NR (closed symbols).

The filler-filler interaction or Payne effect of the silica-filled NR compounds

remarkably decreases with increasing silane contents in the NXT- and VP Si-363-grafted-
NRs, as shown in Figure 8.12(a). The use of NXT-grafted-NR results in a clearly lower
Payne effect of the filled compounds compared to the use of VP Si-363-grafted-NR. At 6.1
wt% of NXT, the compound shows a slightly lower Payne effect compared to the reference
compound with 8.6 wt% of TESPT. Different silanes have different efficiencies for silica-
reinforced compounds. At equal weight, NXT that has a lower molecular weight and three
ethoxy-groups per molecule, shall have more functional groups available on a molar basis
to react with the silanol groups of silica. As reported in Chapter 6, even at the same
amount of functional groups, the use of NXT silane still provides a better improvement in
the Payne effect than VP Si-363 due to its smaller molecule and less steric hindrance for
the silanization reaction. Figure 8.12(b) shows the flocculation rate constants of the silica-

194
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

filled compounds with different types of compatibilizer. The addition of both types of
compatibilizer at only a small amount reduces the flocculation rate of the compounds to a
similar level as that of the TESPT-compound. However, the scatter in the data is too large
to discriminate between the NR-g-NXT and NR-g-VP Si-363.

12 5
(a) (b)

Reinforcement index (M300/M100)

10
4
300% Modulus (MPa)

8
3
6

2
4 None
None
TESPT
TESPT
1
2 NR-g-NXT
NR-g-NXT

NR-g-VP Si-363 NR-g-VP Si-363

0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.13 300% Modulus (a) and reinforcement index (b) of silica-filled NR with NXT-
grafted-NR (open symbols) and VP Si-363-grafted-NR (closed symbols).

The 300% modulus and reinforcement index of the silica-filled NR vulcanizates

increase with the addition of increasing amounts of silane in the grafted NRs as displayed
in Figures 8.13(a) and (b), respectively. This corresponds with the increased filler-rubber
interaction and decreased Payne effect. The incorporation of NXT-grafted-NR leads to a
clearly higher 300% modulus than the use of VP Si-363-grafted-NR, indicating a higher
reinforcement efficiency of NXT for silica-reinforced NR. The NXT silane consists of more
reactive ethoxy-groups to form a bond with the silanol-groups on the silica surface
compared to the VP Si-363 silane. Moreover, at equal total silane content in the
compounds, the higher grafting efficiency of the grafted-NXT on NR should result in a
higher level of crosslinks or chemical bonds between rubber and filler, leading to a higher
300% modulus and reinforcement index.

195
Chapter 8

30 800
(a) (b)
700
25

Elongation at break (%)

600
Tensile strength (MPa)

20
500

15 400

300
10 None None

TESPT 200 TESPT

5 NR-g-NXT NR-g-NXT
100
NR-g-VP Si-363 NR-g-VP Si-363
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 8.14 Tensile strength (a) and elongation at break (b) of silica-filled NR with NXT-
grafted-NR (open symbols) and VP Si-363-grafted-NR (closed symbols).

The tensile strength of the silica-filled NR vulcanizates is enhanced with

increasing total silane content in the grafted-NRs in correspondence with the increased
modulus and reinforcement index as shown in Figure 8.14(a). However, among the silane
types studied, TESPT still provides the best silica reinforcement for the NR compounds.
The use of NR-g-NXT and NR-g-VP Si-363 as compatibilizers results in more or less the
same elongation at break and increasing the total silane content decreases the
extensibility of the vulcanizate to the level of the reference TESPT system as shown in
Figure 8.14(b), due to the increase of linkages within the materials.
For the vulcanizates containing a total silane content of 6.1 wt% relative to the
silica, with NR grafted by using 20 phr of silane, the dynamic mechanical properties were
analyzed and the results are shown in Figures 8.15 and 8.16 in comparison with the ones
with TESPT at 8.6 wt% relative to the silica and without compatibilizer.

196
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

100000 10000
None (a) None (b)
TESPT TESPT

NR-g-NXT NR-g-NXT
10000 1000
NR-g-VP Si-363 NR-g-VP Si-363

Eʺ (MPa)
Eʹ (MPa)

1000 100

100 10

10 1
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Temperature (oC) Temperature (oC)

Figure 8.15 Storage modulus (E′) (a) and loss modulus (Eʺ) (b) of silica-filled NR with
NXT- and VP Si-363-grafted-NRs, both with total silane of 6.1 wt% relative to the silica;
TESPT at 8.6 wt% relative to silica.

The storage (E′) and loss (Eʺ) moduli as function of temperature of the silica-filled
NR vulcanizates are shown in Figures 8.15(a) and (b), respectively. In the glassy region,
the silica-filled vulcanizate with TESPT shows only small difference in storage modulus
compared to the mixes with NXT- and VP Si-363-grafted NRs. But in the rubbery region,
the silica/TESPT system shows the highest storage modulus while the other vulcanizates
show similar levels of the moduli. For filled rubber, the modulus of the materials depends
on filler content and filler-filler interactions that affect the segmental mobility of the rubber
chains, and the modulus normally increases with increasing amount of filler[30] or degree of
filler-rubber interaction.[31] The silica-filled NR with TESPT that has the highest filler-rubber
interactions therefore shows the highest storage modulus. On the contrary, the silica-filled
NR without compatibilizer that has the highest filler-filler interaction or Payne effect; Figure
8.12, and a filler-filler network with trapped and less flexibility of the chains to relocate
under deformation[31] shows the highest loss modulus in the rubbery region, i.e. the
highest loss of energy as dissipated in heat. The incorporation of TESPT, NXT- or VP Si-
363-grafted-NRs results in less filler-filler interaction; Figure 8.12(a), and higher chemically
bound rubber content; Figure 8.11(a), leads to a better elasticity and lower loss modulus.

197
Chapter 8

1.2
None

1.0 TESPT
NR-g-NXT
NR-g-VP Si-363
0.8

Tan δ 0.6

0.4

0.2

0.0
-80 -60 -40 -20 0 20 40 60 80
Temperature (oC)

Figure 8.16 Tan δ of silica-filled NR with NXT- and VP Si-363-grafted-NRs, both with total
silane of 6.1 wt% relative to the silica; TESPT at 8.6 wt% relative to silica.

The loss tangent or tan δ as a function of temperature of the silica-filled NR

vulcanizates with different compatibilizers are shown in Figure 8.16, wherein the rubber
with TESPT shows a slightly higher tan δ peak intensity over the other vulcanizates, which
indicates that there is less trapped rubber in the silica network and so more flexible rubber
chains to respond to the dynamic deformation in the transition region. In the rubbery
region where the rubber chains are in motion, the compatibilized systems with the three
types of compatibilizers show lower tan δ values compared to the non-compatibilized one,
due to their better filler-rubber interactions and so less energy loss during deformation.
The values of tan δ at different positions are summarized in Table 8.5 together with the
values of Tg taken from the tan δ and Eʺ peaks.

Table 8.4 Tg and tan δ values of silica-filled NR vulcanizates at 6.1 wt% relative to the
silica for NR-g-NXT and NR-g-VP Si-363; and 8.6 wt% for TESPT.

Tg(oC) Values of Tan δ

Compatibilizer types
at tan  peak at E peak at peak at 5oC at 60oC
Without compatibilizer -47 -58 0.94 0.09 0.11
TESPT -45 -54 0.97 0.10 0.07
NR-g-NXT -44 -52 0.95 0.08 0.05
NR-g-VP Si-363 -44 -52 0.95 0.08 0.05

198
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

The glass transition temperature (Tg) is the temperature of the glass-to-rubber

softening transition which can be observed as a large drop in modulus. Even though the
Tg is often taken at the peak of tan δ, the Tg of a polymer can be reported by using the
temperature at either the loss modulus (Eʺ) peak or at the tan δ peak, in which the value at
the Eʺ peak corresponds to segmental relaxation processes only and is not affected by
different levels of filler-rubber interaction.[32] This present work compares the Tg’s of the
silica-filled NRs with different compatibilizers taken from both the Eʺ and tan δ peaks,
which show the same trend, as shown in Table 8.4. That is, the Tg’s of the silica-filled
vulcanizates with TESPT, NXT- and VP Si-363-grafted-NRs are shifted to a higher
temperature compared to the one without compatibilizer, due to an increased degree of
filler-rubber interactions which restrict the mobility of the polymer chains. Although the
compounds with NXT- and VP Si-363-grafted-NRs as compatibilizers show lower
chemically bound rubber content and inferior mechanical properties than for TESPT, their
Tg’s are slightly higher than the one with TESPT. The use of silane grafted NR with some
silane molecules already bonded to rubber may result in a higher extent of the coupling
reaction between rubber and silica. In the transition region, the free volume between
polymer molecules rapidly increases leading to an increase of polymer chain mobility to
respond to dynamic deformation. The better chain flexibility in the compatibilized
compounds and good filler-rubber interactions lead to lower energy losses and so lower
tan δ in the rubbery region.
As mentioned in the previous chapters, the tan δ values at low temperature, i.e.
5-35 C for summer tires and 2-20oC for winter tires[33], and at higher temperatures
o

commonly at 60oC can be used to indicate tire wet grip and rolling resistance,
respectively. The results in Table 8.4 shows that the use of NXT- and VP Si-363-grafted-
NRs as compatibilizers give slightly lower tan δ values at 5oC, i.e. a small decrease in wet
grip, but lower tan δ at 60oC implying an improvement in rolling resistance compared to
the reference compound with TESPT. The lower tan δ at 60oC in the vulcanizates
containing the silane grafted NR again may be attributed to efficient coupling or bridging
between rubber and silica via the pre-grafted silane fragments and some additional
linkages created by self-crosslinking between the grafted moieties.
Overall, the use of silane-grafted-NRs as compatibilizers in the silica-filled NR
compounds without sulfur compensation can effectively enhance the properties compared
to the non-compatibilized system. By taking the compound with 8.6 wt% TESPT relative to
the silica as reference, the use of NXT-grafted-NR that gives better performance than VP
Si-363 with 4-6 wt% of silane relative to the silica, gives similar Mooney viscosity, Payne
effect and flocculation rate constant, but inferior chemically bound rubber content,

199
Chapter 8

modulus and tensile strength. However, a slight shift in the Tg of the vulcanizate in the
presence of NXT-grafted-NR gives a positive effect on tire rolling resistance as indicated
by the lower tan δ at 60oC.

8.4 Conclusions

Increasing amounts of NXT- and VP Si-363-grafted-NRs as compatibilizers in

silica-filled NR-compounds decrease the Mooney viscosity, filler-filler interaction or Payne
effect, flocculation rate constant and increase the chemically bound rubber contents,
300% modulus, reinforcement index and tensile strength. The reinforcement index and
tensile strength of the silica-filled NR vulcanizates almost level off when the silane content
in the silane-grafted-NRs, which includes free and grafted silane, exceeds 3.0 wt% relative
to the silica. Tensile fractured surfaces of the silica-filled NR vulcanizates indicate ductile
failure with surface roughness and many tear lines when the silane-grafted-NRs are used,
similar to the compound with TESPT. At the same silane loading, the use of silane-
grafted-NR as compatibilizer gives a better improvement in Payne effect, chemically
bound rubber content, 300% modulus and tensile strength compared to the straight use of
a silane. Comparing the two types of silane-grafted-NRs, the use of NXT provides a better
improvement of overall properties than VP Si-363, but the properties are still somewhat
lower than for the use of TESPT at its optimum content, i.e. at 8.6 wt% relative to the
silica. In the presence of the silane-grafted-NR as compatibilizer, the Tg of the silica-filled
o
NR vulcanizates is shifted to higher temperature. Tan δ at 5 C that relates to wet grip is
slightly decreased but the tan δ at 60oC is clearly reduced indicating a lower tire rolling
resistance for tire treads made thereof, compared to the reference compound with TESPT.

8.5 References

1. I.R. Gelling, Rubber Chem. Technol., 58, 86 (1985).

2. A.K. Manna, A.K. Bhattacharyya, P.P De, D.K. Tripathy, S.K. De, D.G. Peiffer,
Polymer, 39, 7113 (1998).
3. A.K. Manna, P.P. De, D.K. Tripathy, S.K. De, D.G. Peiffer, J. Appl. Polym. Sci., 74,
389 (1999).
4. M.M. Jacobi, M.V. Braum, T.L.A.C. Rocha, R.H. Schuster, Kautsch. Gummi Kunstst.,
60, 460 (2007).
5. T.L.A.C. Rocha, R.H. Schuster, M.M. Jacobi, D. Samios, Kautsch. Gummi Kunstst.,
57, 656 (2004).
6. A.K. Manna, P.P. De, D.K. Tripathy, J. Appl. Polym. Sci., 84, 2171 (2002).

200
 Silane-grafted-NR as compatibilizer in silica-reinforced NR

7. Y.-T. Shieh, T.-H. Tsai, J. Appl. Polym. Sci., 69, 255 (1998).
8. K. Sirisinha, M. Boonkongkaew, S. Kositchaiyong, Polym. Test., 29, 958 (2010).
9. M. Beltrán, C. Mijangos, Polym. Eng. Sci., 40, 1534 (2000).
10. Y.-T. Shieh, C.-M. Liu, J. Appl. Polym. Sci., 74, 3404 (1999).
11. S. Jain, J.G.P. Goossens, M. van Duin, Macromol. Symp., 233, 225 (2006).
12. C. Yin, Q. Zhang, J. Gu, J. Zheng, G. Gong, T. Liang, H. Zhang, J. Appl. Polym. Sci.,
128, 2262 (2013).
13. C. Yin, Q. Zhang, D. Yin, J. Gu, G. Gu, T. Liang, J. Appl. Polym. Sci., 119, 2808
(2011).
14. G. Colucci, A.D. Gianni, R. Bongiovanni, A. Priola, F. Torello, S. Bracco, P. Sozzani,
Macromol. Symp., 296, 550 (2010).
15. A. Bertora, M. Castellano, E. Marsano, M. Alessi, L. Conzatti, P. Stagnaro, G.
Colucci, A. Priola, A. Turturro, Macromol. Mater. Eng., 296, 455 (2011).
16. F. Schapman, J.P. Couvercelle, C. Bunel, Polymer, 39, 4955 (1998).
17. S.-S. Choi, Korea Polym. J., 8, 73 (2000).
18. A. Ansarifar, S.F. Shiah, M. Bennett, Int. J. Adhes. Adhes., 26, 454 (2006).
19. S. Schaal, A.Y. Coran, Rubber Chem. Technol., 73, 225 (2000).
20. S. Choi, B.-H. Park, H. Song, Polym. Adv. Technol., 15, 122 (2004).
21. S. Wolff, M.-J. Wang, Rubber Chem. Technol., 65, 329 (1992).
22. S.-S. Choi, C. Nah, B.-W. Jo, Polym. Int., 52, 1382 (2003).
23. S.-S. Choi, J. Appl. Polym. Sci., 79, 1127 (2001).
24. C.Jr. Lin, W.L. Hergenrother, A.S. Hilton, Rubber Chem. Technol., 75, 215 (2002).
25. L. Guy, S. Daudey, P. Cochet, Y. Bomal, Kautsch. Gummi Kunstst., 62, 383 (2009).
26. W. Kaewsakul, K. Sahakaro, W.K. Dierkes, J.W.M. Noordermeer, Rubber Chem.
Technol., 86, 313 (2013).
27. S. Mihara, R.N. Datta, J.W.M. Noordermeer, Rubber Chem. Technol., 82, 524
(2009).
28. H.-D. Luginsland, J. Fröhlich, A. Wehmeier, Rubber Chem. Technol., 75, 563 (2002).
29. S. Wolff, M.-J. Wang, E.-H. Tan, Rubber Chem. Technol., 66, 163 (1993).
30. C. Gauthier, E. Reynaud, R. Vassoille, L. Ladouce-Stelandre, Polymer, 45, 2761
(2004).
31. H.M. Da Costa, L.L.Y. Visconte, R.C.R. Nunes, C.R.G. Furtado, J. Appl. Polym. Sci.,
83, 2331 (2002).
32. C.G.Robertson, C.J. Lin, M. Rackaitis, C.M. Roland, Macromolecules, 41, 2727
(2008).
33. UNECE Regulation No 117, GRRF 71st Session, September 13-15, 2011.

201
Chapter 8

202
 CHAPTER 9

Compatibilization of Silica-Filled Natural Rubber by Using

Silane-Grafted-Rubber with Compensation of Sulfur

Sulfur compensation is applied to the silica-filled NR compounds with the NXT-

grafted-NR as compatibilizer and the properties are investigated in comparison with those
of the silica-filled NR with TESPT and without compatibilizer. The non-purified NXT-
grafted-NRs were used as part of the rubber matrix in a range of 5-20 phr, so as to give
total NXT contents of 0.8-6.1 wt% relative to the silica. The compounds with sulfur
compensation show slightly lower Mooney viscosity, similar Payne effect and flocculation
rate constant but higher cure torque difference compared to the compounds without extra
sulfur. After vulcanization, the rubber with NR-g-NXT at 6 wt% NXT relative to the silica
and with sulfur compensation shows almost the same levels of 300% modulus,
reinforcement index, tensile strength and elongation at break, compared to the reference
compound with 8.6 wt% TESPT relative to the silica. Tensile fractured surfaces of the
vulcanizates with sulfur compensation show more tear lines, which correlates with their
tensile strength. The use of NR-g-NXT as compatibilizer increases Tg of the vulcanizates
slightly and sulfur compensation causes a slight further shift of Tg to a higher temperature.
Chapter 9

9.1 Introduction

The results in Chapter 8 have shown that the use of NXT-grafted-NR as

compatibilizer gives better properties than the straight use of NXT and the use of VP Si-
363-grafted-NR. However, due to the difference in sulfur amounts in the compounds, the
properties of the NXT compatibilized compounds are still inferior to those of the reference
TESPT compound. In Chapter 4, it has been demonstrated for the silica-filled NR
compounds compatibilized with ENR and a small amount of TESPT that sulfur
compensation relative to the reference compound with TESPT alone could further improve
the properties such as crosslink density, modulus and tensile strength. The use of
epoxidized low molecular weight NR as compatibilizer for silica-filled NR also showed that,
even though the addition of compatibilizer could improve both compound and vulcanizate
properties, the addition of extra sulfur to compensate for the sulfur released from silane
molecules remarkably enhanced the 300% modulus, tensile strength and loss tangent at
60oC due to the better network formation.[1]
A number of previous studies reported the same with regard to the influence of
sulfur content in the silica-reinforced rubber compounds. Ten Brinke et al.[2] showed that,
with sulfur correction, all sulfur containing silanes behaved more like TESPT; such as bis-
(triethoxysilylpropyl)disulfide (TESPD) with sulfur correction could give final properties
similar to those of TESPT. A mechanistic study of the role of coupling agents in silica-
rubber compounds[3] reported that the carbon and sulfur chain lengths within the coupling
agents prominently influence the tensile properties. Sulfur in the silanes plays a dual role:
one is as part of the coupling agent to attach to the rubber, another role is as direct
curative. The addition of extra sulfur to the TESPT-silica-NR significantly changes the
overall properties of both compounds and vulcanizates. The correction of sulfur deficiency
compared to the TESPT-based system leads to enhanced properties which are related to
extra network formation.[4]
This Chapter reports the effect of sulfur compensation relative to the sulfur
contained in the reference TESPT compound in order to further improve the properties of
the silica-filled NR compounds with NXT-grafted-NR as compatibilizer.

9.2 Experimental

9.2.1 Materials and compound preparation

All of the materials used for compound preparation in this chapter, i.e. natural
rubber, silica, TESPT, TDAE oil, TMQ, zinc oxide, stearic acid, CBS, DPG and sulfur,

204
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

were the same as described in Chapter 3. The NXT-grafted-NR was prepared and
characterized as detailed in Chapter 7.
The silica-filled NR compounds were prepared using the formulations as shown
in Table 9.1, and the mixing procedures as described in Table 9.2. The pre-prepared NXT-
grafted-NRs were used without purification at the varied amounts at 5, 10, 15 and 20 phr
to give the different total silane contents as shown in Table 8.1 in Chapter 8. The
compounds were mixed in an internal mixer with the initial mixer temperature setting of
100oC to complete the silanization of silica and silane, as explained in Chapter 5. The
reference compounds with 8.6 wt% TESPT relative to the silica and without compatibilizer
were also prepared according to the formulations as shown in Table 8.2 and the mixing
procedures in Table 8.3.

Table 9.1 Compound formulations with the use of NR-g-NXT and sulfur compensation.
Parts per hundred parts of rubber (phr)
Ingredients NXT-grafted-NR with silane NXT-grafted-NR with silane
10 phr 20 phr
RSS3 95.0 90.0 85.0 80.0 95.0 90.0 85.0 80.0
TESPT - - - - - - - -
NXT-grafted-NR 5.0 10.0 15.0 20.0 5.0 10.0 15.0 20.0
Zeosil 1165MP 55.0 55.0 55.0 55.0 55.0 55.0 55.0 55.0
TDAE oil 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0
ZnO 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
TMQ 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Stearic acid 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
DPG 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
CBS 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
Sulfur* 2.59 2.55 2.51 2.47 2.56 2.48 2.41 2.33
Remarks: *Sulfur content was compensated for the amount of sulfur contained in TESPT in the
reference compound.

Table 9.2 Mixing procedures.

Mixing procedures Cumulative time (mins.)
Step 1 : Internal mixer
- NR and silane-grafted-NR mastication 2
- Addition of first half of silica 7
- Addition of second half of silica and TDAE oil 12
- Addition of ZnO, stearic acid, and TMQ 15
Step 2 : Two roll mill
- Addition of DPG, CBS, and sulfur (+ extra sulfur) 5

205
Chapter 9

9.2.2 Property testing and characterization

Mooney viscosity of the compounds after step 2 and cure characteristics were
tested according to the methods as described in Chapter 5. Payne effect and flocculation
rate constant of the final compounds (with curatives) were analyzed according to the
methods as detailed in Chapter 3. Bound rubber content of the non-productive compounds
after step 1 was analyzed following the procedures as described in Chapter 3.
The compounds were press-cured to their respective optimum cure times (tc90)
and tensile as well as dynamic mechanical properties were tested following the methods
as described in Chapter 5. Tensile fractured surfaces were analyzed by using the
scanning electron microscopy (SEM) technique, as described in Chapter 4.

9.3 Results and discussion

9.3.1 Mooney viscosity, bound rubber content, Payne effect and flocculation
rate constant

The Mooney viscosity of the silica-filled compound is affected by the degree of

silica dispersion and silica-rubber interactions in a rubber matrix which in turn may be
correlated to other properties. The presence of a compatibilizer which is adsorbed onto
the silica surface to increase its hydrophobic character results in a suppression of the
silica-silica network and improves the silica dispersion in the rubber matrix, leading to a
decrease of Mooney viscosity of the silica-filled compounds.[5] On the other hand, an
increase of the Mooney viscosity may also be observed if the filled-compound shows a
significant improvement in filler-rubber interactions by impeded rubber segmental mobility.

206
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

90 60
(a) (b)
None
80
50 TESPT

Bound rubber contents (%)

70 NR-g-NXT

60 40
ML1+4 (100oC)

50
30 Chemically bound rubber
40

30 20
Physically bound rubber
20
None TESPT 10
10 NR-g-NXT NR-g-NXT+S

0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 9.1 Mooney viscosity (a) and bound rubber contents (b) of silica-filled NR
compounds with NXT-grafted-NR as compatibilizer without and with sulfur compensation,
in comparison with the reference compounds with TESPT and without compatibilizer.

Figure 9.1(a) shows the Mooney viscosities of the silica-filled NR compounds

with NXT-grafted-NR as compatibilizer without and with sulfur compensation in
comparison with the reference compounds. Increasing amounts of NXT-silane in the
grafted-NR significantly decrease the Mooney viscosities of both sets of silica-filled
compounds without and with sulfur compensation, in which the sulfur-compensated ones
show slightly lower viscosities. The compounds containing NR-g-NXT at 6.1 wt% silane
relative to the silica show more or less the same level of the Mooney viscosity to that of
the reference compound with 8.6 wt% TESPT relative to the silica.
The bound rubber content indicates the degree of silica-rubber interaction in the
silica-filled rubber compounds. This measurement was done by using a non-productive
compound (without curative) as obtained after the 1st mixing step. The extra sulfur was
added in the 2nd step of mixing on the two roll mill and so there is only one set of data
available. Chemically and physically bound rubber contents of the silica-filled NR
compounds with different amounts of NXT-grafted-NR as compatibilizer are shown in
Figure 9.1(b) in comparison with those of the reference compounds. As expected, the
silica-filled NR compound with TESPT shows the highest chemically bound rubber content
due to the strong filler-rubber interactions and some premature crosslinks generated by
the sulfur atoms in the TESPT tetrasulfide structure. The silica-filled NR compound without

207
Chapter 9

any compatibilizer shows the lowest chemically bound rubber, which is somewhat higher
than the value of its physically bound rubber content. In the non-compatibilized compound,
the chemically bound rubber can be attributed to the trapped rubber within the tight filler-
filler network and some filler-rubber interactions enhanced by the naturally occurring polar
functional groups such as aldehydes, proteins and phospholipids in the NR. The addition
of NXT-grafted-NR as compatibilizer significantly increases the chemically bound rubber
contents with increasing NXT amount to about 3 wt% in relation to the silica and thereafter
the values level off. The compound with NXT-grafted-NR as compatibilizer contains less
active groups to react and/or interact with the silanol groups on the silica surface.
Moreover, it has no contribution from partial network formation like for the TESPT
compound. The chemically bound rubber content of the system with NR-g-NXT is
therefore still significantly lower than the reference compound with TESPT. However, the
physically bound rubber contents of all the silica-filled NR compounds are comparable.

2000 0.35
None (a) (b)
Payne effect [(G′(0.56%)-G′(100%)] (kPa)

Flocculation rate constant (ka) (s-1)

1800 TESPT 0.30

1600 NR-g-NXT
NR-g-NXT+S 0.25
1400
1200
0.20
1000
0.15
800
600 None
0.10
TESPT
400
0.05 NR-g-NXT
200 NR-g-NXT+S
0 0.00
0 2 4 6 8 0 2 4 6 8
Silane contents NXT contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 9.2 Payne effect (a) and flocculation rate constant (b) of silica-filled-NR compounds
with NXT-grafted-NR as compatibilizer without and with sulfur compensation, in
comparison with reference compounds with TESPT and without compatibilizer.

The Payne effect of the silica-filled NR compounds dramatically decreases with

the addition of NXT-grafted-NR as compatibilizer as shown in Figure 9.2(a), however,
there is no significant difference between the silica-filled compounds without and with
sulfur compensation. The increasing silane contents in the silica-filled compounds

208
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

increase the amount of reactive groups to interact or react with the silanol groups on the
silica surface which leads to a decrease of silica surface polarity and silica agglomeration.
The Payne effects of the silica-filled compounds decrease to a similar level as that of the
silica-filled compounds with TESPT when NR-g-NXT at 4-6 wt% silane relative to the silica
was used.
The flocculation rate constant (ka) of the silica-filled NR compounds with NXT-
grafted-NR with and without sulfur compensation is shown in Figure 9.2(b). As previously
discussed in Chapter 8, the addition of NXT-grafted-NR reduces the flocculation rate due
to a reduction of the silica polarity after the silanization. The compounds without sulfur
compensation show a fluctuation of data that may be caused by experimental error, but
the ones with sulfur compensation show a better trend with slightly lower values. The
addition of just a small amount of NXT-grafted-NR decreases the flocculation rate
constant to the same level of TESPT. The effective suppression of the silica re-
agglomeration is derived from the silanization of the grafted silane onto the silica surface
as well as the shielding effect.

9.3.2 Cure characteristics

14 25
(a) (b)
12
Optimum cure time (min)

20
Torque difference (dN.m)

10

15
8

6
10
None None
4
TESPT TESPT
5
2 NR-g-NXT NR-g-NXT
NR-g-NXT+S NR-g-NXT+S
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 9.3 Optimum cure time (a) and torque difference (b) of silica-filled NR compounds
with NXT-grafted-NR as compatibilizer without and with sulfur compensation, in
comparison with the reference compounds with TESPT and without compatibilizer.

209
Chapter 9

Optimum cure time and torque difference of silica-filled NR compounds with

NXT-grafted-NR as compatibilizer without and with sulfur compensation are shown in
Figures 9.3(a) and (b) in comparison with those containing TESPT and without any
compatibilizer. In Figure 9.3(a), the silica-filled NR-compound with TESPT shows the
shortest optimum cure time as the silane has efficiently reacted with the silanol groups on
the silica surface, leading to more hydrophobicity and less curative adsoption on the silica
surface. Moreover, TESPT is a sulfur donor.[6] The TESPT-compound therefore has a
faster cure time compared to the other compounds. The optimum cure times of the silica-
filled NR compounds with NXT-grafted-NR as compatibilizer decrease with increasing
NXT-contents due to more reactive ethoxy-groups available to react with the silanol
groups resulting in less interference with the curatives in the vulcanization reaction. The
addition of more elemental sulfur to correct the total sulfur content towards the sulfur
contained in the TESPT compound results in longer cure times, indicating that the change
of sulfur-to-accelerator ratio has an influence on the crosslinking reaction in the rubber. In
this case, the increase of sulfur content while the accelerator and activator contents
remain the same leads to a longer time required to complete the vulcanization reaction.
The silica-filled NR-compounds with NXT-grafted-NR as compatibilizer show a
decrease of torque difference with increasing amounts of NXT-silane and the compounds
with sulfur correction show a higher torque difference compared to the compounds with
normal sulfur content, as seen in Figure 9.3(b). When the amount of sulfur is increased,
the crosslink density in the rubber matrix is increased. However, at the same sulfur
content, the compound with NXT-grafted-NR has a lower torque difference compared to
the compound with the use of straight TESPT, and the torque difference decreases with
increasing NXT-content. This decreasing torque difference might be caused by the
presence of deblocked NXT-fragments which can act as a plasticizer, as previously
discussed in Chapter 6, but was not further studied in the present context.

210
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

9.3.3 Tensile properties and fractured surface topography

9.3.3.1 Tensile properties

12 5.0
(a) (b)

Reinforcement index (M300/M100)

10
4.0
300% Modulus (MPa)

8
3.0
6

2.0
4

None TESPT 1.0 None TESPT

2
NR-g-NXT NR-g-NXT+S NR-g-NXT NR-g-NXT+S
0 0.0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. to silica)

Figure 9.4 Modulus at 300% elongation (a) and reinforcement index (b) of silica-filled NR
compounds with NXT-grafted-NR as compatibilizer without and with sulfur compensation,
in comparison with the reference compounds with TESPT and without compatibilizer.

The 300% Modulus and reinforcement index of the silica-filled NR-vulcanizates

with NXT-grafted-NR as compatibilizer, with and without sulfur correction, increase with
rising silane contents as shown in Figures 9.4(a) and (b). Sulfur compensation increases
the modulus of the vulcanizates due to a resulting increase of the crosslink density via
sulfur crosslinks as indicated by the increase of torque difference in Figure 9.3(b).
However, the sulfur correction in the NXT-grafted-NR compatibilized vulcanizates has no
effect on the reinforcement index, as seen in Figure 9.4(b). The decrease in torque
difference with increasing NXT-grafted-NR contents in Figure 9.3(b) does not reflect in the
300% modulus, as would have been expected: on the contrary, the 300% modulus
increases, as does the reinforcement index. The ultimate 300% modulus and
reinforcement index for the sulfur compensated compounds almost reach the level
obtained for straight TESPT.

211
Chapter 9

30 800
(a) (b)
700
25

Elongation at break (%)

600
Tensile strength (MPa)

20
500

15 400

300
10
200
5 None TESPT None TESPT
100
NR-g-NXT NR-g-NXT+S NR-g-NXT NR-g-NXT+S
0 0
0 2 4 6 8 0 2 4 6 8
Silane contents Silane contents
(wt% rel. to silica) (wt% rel. silica)

Figure 9.5 Tensile strength (a) and elongation at break (b) of silica-filled NR compounds
with NXT-grafted-NR as compatibilizer without and with sulfur compensation, in
comparison with the reference compounds with TESPT and without compatibilizer.

Tensile strength of the vulcanizates increases with increasing NXT-contents in

the NXT-grafted-NR due to an increase of silica-rubber interactions in the compounds,
and the sulfur correction increases the tensile strength of the vulcanizates relative to no
sulfur-correction. The presence of 4.6 wt% NXT relative to the silica in the NXT-grafted-
NR used as compatibilizer combined with sulfur correction increases the tensile strength
to the same level as that of the use of optimum TESPT content. With sulfur correction and
increasing NXT content, the elongation at break of the vulcanizates tends to decrease
slightly as a result of more filler-rubber interactions and a higher crosslink density.

212
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

9.3.3.2 Surface topography

(a) (b) (c)

Figure 9.6 Tensile fractured surfaces of silica-filled NR vulcanizates: (a) with NR-g-NXT
containing total 1.7 wt% NXT rel. to silica; (b) with NR-g-NXT containing total 1.7 wt%
NXT rel. to silica and sulfur correction; (c) with 8.6 wt% TESPT rel. to silica.

With elemental sulfur added into the compounds to compensate for the sulfur in
the TESPT molecules, the fractured surfaces of the vulcanizate show more tear lines
which correlates with their higher tensile strength: Figure 9.5.

9.3.4 Dynamic mechanical properties

Dynamic mechanical analysis has been applied to study the relaxation behavior
of the filled polymer systems which correlates with the mobility of the polymer chains, that
depends on the specific character of the filler and the polymer matrix, the amount of filler
and degree of filler-polymer interactions.[7-9]

213
Chapter 9

100000 10000
(a) (b)

None None

10000 TESPT 1000 TESPT

NR-g-NXT
NR-g-NXT
E′ (MPa)

NR-g-NXT+S

Eʺ (MPa)
NR-g-NXT+S
1000 100

100 10

10 1
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Temperature (o C) Temperature (oC)

Figure 9.7 Storage (a) and loss (b) moduli of silica-filled NR compounds with NXT-grafted-
NR with 6.1 wt% NXT rel. to silica as compatibilizer without and with sulfur compensation,
in comparison with the reference compounds with TESPT and without compatibilizer.

Storage (E′) and loss (Eʺ) moduli of the silica-filled NR with NXT-grafted-NR, with
and without sulfur correction, with TESPT and without compatibilizer are compared in
Figures 9.7(a) and (b). In the glassy state, all the filled vulcanizates show a similar level of
E′ even though they have different degrees of filler-rubber interaction as reflected in the
bound rubber contents in Figure 9.1(a). However, the Eʺ-peak of the silica-filled NR with
NXT-grafted-NR is shifted to a higher temperature compared to the silica-filled NR with
TESPT and without compatibilizer, respectively. The increase of sulfur content due to the
sulfur compensation relative to the TESPT compound results in a slight further shift of the
Eʺ-peak due to the increase of crosslink density, which restricts the segmental motion of
the rubber chains. In the rubbery region, the silica-filled NR vulcanizate with TESPT
shows the highest E′ due to the greatest extent of filler-rubber interactions and an increase
of crosslink points through silane bridges, but lower Eʺ compared to the non-
compatibilized system. The pronounced silica-silica network and trapped rubber chains in
the silica-filled NR vulcanizate without compatibilizer results in the highest loss of energy.

214
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

1.2

1 None

TESPT

0.8 NR-g-NXT

Tan δ
NR-g-NXT+S

0.6

0.4

0.2

0
-80 -60 -40 -20 0 20 40 60 80
Temperature (oC)

Figure 9.8 Tan δ of silica-filled NR compounds with NXT-grafted-NR with 6.1 wt% NXT rel.
to silica as compatibilizer without and with sulfur compensation, in comparison with the
reference compounds with TESPT and without compatibilizer.

Tan δ of silica-filled NR vulcanizates with NXT-grafted-NR as compatibilizer with

and without sulfur compensation is shown in Figure 9.8 in comparison with the
vulcanizates without any compatibilizer and with TESPT. As previously observed in Figure
8.16, the addition of NXT-grafted-NR as compatibilizer into the silica-filled NR slightly
shifts the Tg’s of the vulcanizates to higher temperature. With sulfur compensation the Tg
further shifts as a result of the increased crosslink density. The tan δ value at peak is also
reduced after the sulfur compensation as more crosslink points restrict the relaxation and
reduce chain flexibility. The tan δ values at different positions are summarized in Table
9.3.

215
Chapter 9

Table 9.3 Tg and tan δ values of silica-filled NR vulcanizates with different compatibilizers.

Tg(oC) Values of Tan δ

Compatibilizer types at tan  at E
at peak at 5oC at 60oC
peak peak
Without compatibilizer -47 -58 0.94 0.09 0.11
TESPT -45 -54 0.97 0.10 0.07
NR-g-NXT with NXT 1.7 wt% -44 -52 0.96 0.07 0.05
NR-g-NXT with NXT 3.4 wt% -44 -52 0.96 0.07 0.05
NR-g-NXT with NXT 6.1 wt% -44 -52 0.95 0.08 0.05
NR-g-NXT with NXT 1.7 wt%+S -42 -50 0.91 0.07 0.05
NR-g-NXT with NXT 3.4 wt%+S -42 -50 0.92 0.08 0.05
NR-g-NXT with NXT 6.1 wt%+S -42 -50 0.93 0.08 0.06

The use of TESPT in the silica-filled NR vulcanizate gives the highest tan δ at
o
5 C, that suggests the best wet grip for a tire-tread made thereof; and significantly lower
tan δ at 60oC compared to the non-compatibilized vulcanizate. However, the tan δ at 60oC
of the silica/TESPT system is slightly higher when compared to the vulcanizate containing
NR-g-NXT. The use of NR-g-NXT as compatibilizer suggests therefore somewhat inferior
tire wet grip compared to the use of TESPT, but the lowest tan δ values at 60oC imply the
lowest tire rolling resistance. The addition of sulfur compensation relative to the reference
compound with TESPT has no influence on tan δ at both 5oC and 60oC.

9.4 Conclusions

The properties of silica-filled NR compounds with NXT-grafted-NR as

compatibilizer without and with sulfur compensation change according to the NXT
amounts. The sulfur-compensated-compounds show slightly lower Mooney viscosities,
comparable Payne effect and flocculation rate constant, and significantly higher cure
torque difference compared to the counterparts without sulfur compensation. Compared to
the straight use of 8.6 wt% TESPT, the addition of NR-g-NXT with 4.5-6 wt% NXT relative
to the silica, either with or without sulfur compensation, gives more or less the same levels
of Mooney viscosity and Payne effect while with about 1wt% NXT relative to the silica
already gives a similar level of the flocculation rate constant. Sulfur compensation
increases the modulus and tensile strength of the vulcanizates due to the increased
crosslink density, but has no effect on the reinforcement index. The 300% modulus,
reinforcement index, tensile strength and elongation at break for the sulfur compensated
compounds reach the level obtained for straight TESPT. The presence of NR-g-NXT as

216
 Compatibilization of silica-filled NR by using silane-grafted-rubber with compensation of sulfur

compatibilizer results in a shift of the glass transition temperature (Tg) to a higher

temperature compared to the silica-filled NR with TESPT and without compatibilizer. With
sulfur compensation, the Tg is slightly shifted further due to the increase in crosslink
density. As far as can be derived from the tan δ values at 5oC and 60oC, the use of NXT-
grafted-NR as compatibilizer suggests somewhat inferior tire wet grip but superior tire
rolling resistance compared to the use of TESPT, and the sulfur compensation shows no
influence on the tan δ values at both temperatures.
Overall the use of NXT-grafted-NR with about 6 wt% NXT relative to the silica
and with sulfur compensation gives the properties of compounds and vulcanizates; i.e.
Mooney viscosity, Payne effect, 300% modulus, reinforcement index, tensile strength,
elongation at break, storage modulus and tan δ at 60oC that reach the levels obtained for
straight use of TESPT at 8.6 wt% relative to the silica. Therefore, the NXT-grafted-NR with
sulfur compensation is an alternative to the conventional TESPT for the silica-reinforced
NR compounds, providing better environmental friendliness and safety due to less emitted
ethanol. The amounts of 8.6 wt% TESPT and 6 wt% NXT relative to the silica are more or
less equal in moles, and so the compounds with NXT emit half the ethanol compared to
the use of TESPT. Moreover, the NR-g-NXT can be prepared by melt mixing and the pre-
prepared grafted material is easier to use than the straight silane in liquid form.

9.5 References

1. P. Saramolee, K. Sahakaro, N. Lopattananon, W.K. Dierkes, J.W.M. Noordermeer,

Adv. Mat. Res., 844, 235 (2014).
2. J.W. ten Brinke, P.J. van Swaaij, L.A.E.M. Reuvekamp, J.W.M. Noordermeer,
Rubber Chem. Technol., 76, 12 (2003).
3. J.W. ten Brinke, P.J. van Swaaij, L.A.E.M. Reuvekamp, J.W.M. Noordermeer, Comp.
Sci. Technol., 63, 1165 (2003).
4. K. Kaewsakul, W.K. Dierkes, J.W.M. Noordermeer, S. Sahakaro, Kautsch. Gummi
Kunstst., 65, 33 (2014).
5. S. Wolff, M.-J. Wang, Rubber Chem. Technol., 65, 329 (1992).
6. S. Wolff, Kautsch. Gummi Kunstst., 30, 516 (1977).
7. C.G. Robertson, C.J. Lin, M. Rackaitis, C.M. Roland, Macromolecules, 41, 2727
(2008).
8. L. Ladouce-Stelandre, Y. Bomal, L. Flandin, D. Labarre, Rubber Chem. Technol., 76,
145 (2003).
9. V. Arrighi, I.J. McEwen, H. Qian, M.B. Serrano Prieto, Polymer, 44, 6259 (2003).

217
Chapter 9

218
 Summary

Silica has conventionally been used to produce light-colored rubber products and
mechanical goods, but is increasingly used in tire tread compounds to reduce rolling
resistance and thus automotive fuel consumption. However, its use is hampered by a
large number of silanol groups on the silica surface that lead to strong filler-filler bonds
and poor interaction with non-polar rubbers. Silane coupling agents have been practically
used in silica-filled compounds to improve silica dispersion and the interactions between
silica and rubber. The mixing conditions must be optimized to achieve a good level of
silanization and thus properties. A whole or partial replacement of silane coupling agents
with chemically functionalized natural rubber (NR) as a compatibilizer should reduce the
complications arising from the silanization reaction. This will lead to more use of silica
compounds for tires and consequently more efficient use of automotive energy. After an
introduction in Chapter 1, a literature overview with emphasis on the reinforcement
efficiency of silica-filled rubber with different compatibilizing systems is given in Chapter 2.
Filler characteristics, filler reinforcement in general, the development of silica-silane
technology and alternative approaches to replace the silane are addressed. Particular
attention in the present thesis is given to the use of chemically modified NR as
compatibilizer for silica-reinforced NR compounds. The experimental part starts with
investigating the effect of epoxidized natural rubber (ENR) with different epoxy contents
on the reinforcement of silica-filled NR as detailed in Chapter 3. The use of ENR as
compatibilizer improves the filler-rubber interaction, silica dispersion and tensile strength.
Increase of the ENR content and mole% of epoxide groups decreases the filler-filler
interaction and improves silica dispersion, as reflected in a lower Mooney viscosity, Payne
effect, flocculation rate constant and filler networking factor. The addition of ENR-38 and
ENR-51 (with 38 resp. 51 mole% epoxide) leads to higher filler-rubber interaction
parameters, chemically bound rubber, tensile strength and silica dispersion levels. With
respect to tire performance, the presence of ENR-38 and ENR-51 as compatibilizer
improves wet grip but negatively affects rolling resistance due to an increased glass
transition temperature of the epoxidized NR. The addition of ENR-51 at 7.5 phr shows the
best overall properties, however still somewhat lower when compared to a compound with
the commonly employed bis-(triethoxysilylpropyl)tetrasulfide (TESPT) silane as coupling
agent.
Summary

The use of ENR-51 was optimized at 7.5 phr, but without additional silane
coupling agent the properties of the silica-filled NR are still inferior to the use of TESPT
alone. For further property improvement, the ENR is used in combination with TESPT and
the results are discussed in Chapter 4. The use of ENR-51 at 7.5 phr in combination with
TESPT at 2 wt% relative to the silica further decreases the Mooney viscosity, Payne
effect, flocculation rate constant and increases bound rubber contents of the silica-filled
NR compounds, when compared to a silica-filled compound with ENR-51 alone. A further
increase of TESPT-content has only a small additional effect on the compound properties.
Tensile strength and modulus of the vulcanizates increase with increasing TESPT-
contents and further improve by sulfur compensation relative to the reference compound
with TESPT, to equilibrate the level of crosslink density. The use of ENR-51 at 7.5 phr with
TESPT at 4 wt% relative to silica and sulfur compensation gives vulcanizates with the
same levels of tensile strength compared to vulcanizates with TESPT at 8.6 wt% relative
to silica. The presence of ENR in the silica-filled NR affects the dynamic mechanical
properties. Above the glass-to-rubber transition, the ENR-containing vulcanizates show a
higher tan  compared to the silica/TESPT system. The addition of TESPT to the ENR-
compatibilized compounds causes only a small change in Tg and tan  at 5C, but lowers
tan δ at 60oC. Furthermore, the addition of extra sulfur leads to an increase of Tg and
lower tan  at both 5C and 60oC as a result of increased crosslink density. SEM
micrographs clearly show good silica dispersion in all compatiblized vulcanizates, and the
tensile fracture surfaces of the materials show different tear line patterns dependent on
their tensile strengths.
The use of three compounds with ENR-51 at 7.5 phr in combination with three
types of silane: TESPT, 3-octanoylthio-1-propyltriethoxysilane (NXT) and 3-
mercaptopropyl-di(tridecan-1-oxy-13-penta(ethyleneoxide))ethoxysilane (VP Si-363) in the
range of 2 to 5 wt% relative to the silica, is comparatively discussed in Chapter 5. Addition
of silane reduces the compound viscosity slightly but clearly enhances the chemically
bound rubber content and cure properties. Among the 3 different ENR-51/silane
combinations, the use of ENR/TESPT gives the highest chemically bound rubber content
and tensile properties, whereas the ENR/NXT combination gives higher chemically bound
rubber content but similar tensile properties compared to the ENR/VP Si-363 system. The
better improvement of the overall properties by the use of TESPT over the other two types
of silanes is due to the bis-(triethoxysilyl)- and tetrasulfide-functional groups that provide
more ethoxy groups relatively to react with the silanol groups on the silica surface, and the
reactive free sulfur released by TESPT to involve in network formation. When the amount

220
 Summary

of silane coupling agent is fixed at 4.0 wt% relative to silica in combination with 7.5 phr of
ENR-51 as compatibilizers in the silica-filled NR compound, the addition of extra sulfur to
correct the total sulfur content relative to the reference compound with TESPT leads to
increases of modulus and tensile strength while elongation at break decreases. The use of
o o
an ENR/silane combination increases the tan δ at 5 C and reduces the tan δ at 60 C
compared to the use of ENR-51 alone, indicating an improvement in wet traction and
rolling resistance of tires made thereof. However, overall the use of TESPT alone at
optimum loading still gives the best mechanical and dynamic mechanical properties
compared to the presence of ENR as compatibilizer.
Among the three types of silanes studied in combination with ENR-51 and sulfur
compensation, the use of 2.2 phr TESPT together with 7.5 phr ENR-51 provides the best
overall properties, but it does still not fully match the properties of a compound with
TESPT alone at optimal quantity, i.e. 4.7 phr as previously observed in Chapter 4.
However, in the perspective of the “Magic Triangle of Tire Technology” where wet skid
resistance, rolling resistance and wear are all to be optimized, the combination of ENR-51,
TESPT and sulfur compensation may become a better option for “Winter Tires” that
demand good wet grip, providing that real tire tests have yet to confirm this.
By excluding the ENR as compatibilizer, the properties of silica-fillled NR with
four types of silane coupling agents, i.e. TESPT, NXT, VP Si-363 and vinyltriethoxysilane
(VTES) is investigated in order to screen for a potential silane coupling agent to be used
later for grafting onto NR molecules, as detailed in Chapter 6. Among the four silane
types, VTES gives the poorest overall properties due to its poor shielding effect, and the
vinyl-group that is difficult to couple with NR rubber chains in the presence of sulfur.
Hence, VTES is not an effective silane coupling agent for this silica-filled NR system. The
use of NXT and VP Si-363 shows a good improvement in compound properties, i.e.
mixing energy, Mooney viscosity, cure behaviors, bound rubber contents, and an increase
of mechanical and dynamic mechanical properties compared to that without silane, but the
properties are inferior to the use of TESPT, mainly due to lower crosslink densities
compared with TESPT which has an intrinsic “sulfur-donating” effect. This was not further
pursued in the context of this chapter. The incorporation of NXT silane provides better
scorch safety, and use at the same amount of ethoxy-functional groups with reference to
TESPT gives vulcanizates with the same level of tensile strength. With respect to tire
performance, NXT offers better properties than VP Si-363 but still somewhat inferior when
compared to the reference compound with TESPT.
Later on, NXT and VP Si-363 silane coupling agents are employed to graft onto
NR in order to introduce reactive side-groups for interacting and/or reacting with the

221
Summary

silanol groups on the silica surface. Chapter 7 shows that the silane-grafted-natural
rubbers are successfully prepared under melt mixing conditions in an internal mixer at
140oC using 0.1 phr of 1,1′-di(tert-butylperoxy)-3,3,5-trimethylcyclohexane or Luperox®
231XL40 as initiator. By elemental analysis via the SEM-EDX technique, the modified NRs
contain silicon (Si) atoms in their structures, and the amounts of oxygen and silicon in the
grafted NRs increase with increasing amounts of silane used for the reaction. Both VP Si-
363-grafted-NR and NXT-grafted-NR show infrared absorption bands at 3270, 1075 and
1035 cm-1 assigned to the deformations of -OH, Si-O-C and Si-O-Si, respectively, and the
1
H NMR absorption band at 3.8 ppm assigned to the methylene proton of the ethoxy
groups.
The use of silane-grafted-NRs as compatibilizers in silica-filled NR compounds is
discussed in Chapter 8 in comparison with the straight use of silane in the compound
formulation and the use of TESPT at its optimal amount. The increasing amount of NXT-
and VP Si-363-grafted-NRs as compatibilizers in silica-filled NR compounds decreases
the Mooney viscosity, filler-filler interaction or Payne effect, flocculation rate constant and
increases the chemically bound rubber contents, 300% modulus, reinforcement index and
tensile strength. The reinforcement index and tensile strength of the silica-filled NR
vulcanizates almost level off when the silane content in the silane-grafted-NRs, which
includes free and grafted silane, exceeds 3.0 wt% relative to the silica. Tensile fractured
surfaces of the silica-filled NR vulcanizates indicate ductile failure with surface roughness
and many tear lines when the silane-grafted-NRs are used, like the case for the
compound with TESPT. At the same silane loading, the use of silane-grafted-NR as
compatibilizer gives a better improvement in Payne effect, chemically bound rubber
content, 300% modulus and tensile strength compared to the straight use of that silane.
The presence of NXT provides better improvement of overall properties than VP Si-363,
but both are still lower than the use of TESPT at its optimum content. In the presence of
the silane-grafted-NR as compatibilizer, the Tg of the silica-filled NR vulcanizates is shifted
to a higher temperature. The tan δ at 5oC that relates to wet grip is slightly decreased but
the tan δ at 60oC is clearly reduced indicating a lower tire rolling resistance, compared to
the reference compound with TESPT. Sulfur compensation for the compounds having
NXT-grafted-NR as compatibilizer is also investigated to enhance the properties of the
silica-filled NR as discussed in Chapter 9. The sulfur-compensated-compounds show
slightly lower Mooney viscosity but significantly higher cure torque difference compared to
the compound without extra sulfur. With sulfur compensation relative to the sulfur
contained in the reference TESPT compound, the use of NR-g-NXT with 6 wt% NXT
relative to the silica reduces Mooney viscosity and Payne effect as well as enhances the

222
 Summary

300% modulus and tensile strength of the silica-filled NR to almost the same level as
obtained from the straight use of TESPT at its optimum content, but without significant
effects on the tan δ values at 5 and 60oC. As the amounts of 8.6 wt% TESPT and 6 wt%
NXT relative to the silica are more or less equal in molar amounts, the use of NXT-grafted-
NR therefore emits only a half the ethanol commonly produced by TESPT, leading to a
better environmentally friendly and safe system.
A growing importance of silica-reinforced tire compounds for reduction of
automotive fuel consumption motivates research and development of the relevant rubber
technology. In addition to the commonly used silica/silane technology, alternative
compatibilizers to totally or partially replace the silane for more “green” products are of
interest. This present thesis provides an in-depth study on the application of ENRs as
alternative compatibilizer. The use of ENRs alone as compatibilizers clearly shows an
enhancement already of the silica-NR compatibility, silica dispersion and some
mechanical properties. The properties can be further enhanced by using ENR in
combination with only small amounts of silane coupling agents and sulfur compensation
relative to a reference compound with TESPT. In addition to the more general use of
ENRs, silane-grafted-NR offers another alternative. The preparation of silane-grafted-NR
in the melt mixing state is practical but complicated. The non-purified product contains
both free and grafted silane. Nevertheless, the use of non-purified silane-grafted NRs as
compatibilizers shows an improvement in filler-rubber interactions and the properties of
silica-reinforced NR. At the same total silane content, the use of silane-grafted-NR gives
overall better properties than the straight use of silane and the properties can be further
enhanced by sulfur compensation, taking the compound with optimal content of TESPT as
reference. Overall, even though the application of chemically modified NRs in the forms of
ENRs and silane-grafted-NR as compatibilizers without further silanes results in inferior
overall properties compared to the use of the state-of-the-art TESPT, the right
combination of modified NR with only small amounts of TESPT and sulfur compensation
offers an alternative system with the advantages of NR-based product usage and less
ethanol emission from the silanization reaction. Moreover, the use of ENR brings in a
good wet grip in the perspective of the “Magic Triangle of Tire Technology”.

223
Summary

224
 Samenvatting

Silica wordt van oudsher gebruikt om licht-gekleurde rubber producten en

mechanische artikelen te maken, maar wordt in toenemende mate gebruikt in autobanden
loopvlak-mengsels om de rolweerstand te verlagen en daardoor het brandstofverbruik.
Evenwel, het gebruik daarvan wordt bemoeilijkt door het grote aantal silanol-groepen op
het silica-oppervlak, hetgeen leidt tot sterke vulstof-vulstof bindingen en slechte interactie
met niet-polaire rubbers. Silaancouplingagents worden in de praktijk ingezet in silica-
gevulde mengsels om de silica-dispersie en de interacties tussen silica en rubber te
verbeteren. De mengcondities moeten worden geoptimaliseerd om een goed silanisatie-
niveau te bereiken en daarmee goede eigenschappen. Een algehele of gedeeltelijke
vervanging van silaancouplingagents met chemisch gefunctionaliseerde Natuur Rubber
(NR) als compatibilizer zou de complicaties van de silaniserings-reactie moeten
verminderen. Dit zal dan leiden tot een hoger gebruik van silica mengsels voor banden en
dientengevolge een efficiënter gebruik van de energie benodigd voor wegtransport. Na
een inleiding in Hoofdstuk 1, wordt in Hoofdstuk 2 een literatuur overzicht gegeven met
nadruk op de efficientie van de versterkende werking van silica-gevulde rubber met
verschillende compatibiliserende systemen. Vulstof karakteristieken, vulstof-versterking in
het algemeen, de ontwikkeling van de silica-silaan technologie en alternatieven ter
vervanging van silaan worden belicht. Bijzondere aandacht in het huidige proefschrift ligt
op het gebruik van chemisch gemodificeerde NR als compatibilizer voor silica-versterkte
NR mengsels. Het experimentele gedeelte begint met een onderzoek naar het effect van
geëpoxideerde natuurrubber (ENR) met verschillende epoxy-gehaltes, op de versterking
van silica-gevulde NR, zoals gegeven in Hoofdstuk 3. Het gebruik van ENR als
compatibilizer verbetert de vulstof-rubber interactie, de silica dispersie en de treksterkte.
Verhogen van het ENR-gehalte en mol% epoxide-groepen verlaagt de vulstof-vulstof
interactie en verbetert de silica-dispersie, zoals blijkt uit de Mooney viscositeit, het Payne
effect, de snelheidsconstante van uitvlokken van de silica en de vulstof netwerk-factor.
Toevoeging van ENR-38 en ENR-51 ( met 38, resp. 51 mol% epoxide) leidt tot hogere
vulstof-rubber interactie parameters, chemische bound rubber, treksterkte en silica-
dispersie niveaus. Met betrekking tot banden-gedrag verbetert de aanwezigheid van ENR-
38 en ENR-51 als compatibilizers de natte grip, maar verslechtert de rolweerstand
vanwege een verhoogde glasovergangs-temperatuur van de geëpoxideerde NR. De
toevoeging van 7,5 phr ENR-51 geeft over het geheel genomen de beste eigenschappen,
Samenvatting

maar nog wel enigszins lager in vergelijking met een mengsel met het gewoonlijk
gebruikte bis-(triethoxysilylpropyl)tetrasulfide (TESPT) silaan als coupling agent.
Het gebruik van ENR-51 werd geoptimaliseerd op 7,5 phr, maar zonder extra
silaancoupling agent blijven de eigenschappen van silica-gevulde NR toch achter bij het
gebruik van TESPT alleen. Voor verder eigenschappen-verbetering wordt het gebruik van
ENR in combinatie met TESPT en de resultaten daarvan in Hoofdstuk 4 besproken.
Gebruik van 7,5 phr ENR-51 in combinatie met 2 gew.% TESPTbetrokken op de silica,
verlaagt de Mooney viscositeit verder, evenals het Payne effect en de uitvlokkings
snelheidsconstante, en verhoogt de bound rubber gehaltes van de silica-gevulde NR
mengsels, in vergelijking met een silica-gevuld mengsel met alleen ENR-51. Een verdere
verhoging van het TESPT-gehalte heeft slechts een klein additioneel effect op de
mengsel-eigenschappen. De treksterkte en modulus van de vulcanisaten nemen toe met
hogere TESPT-gehaltes en nog verder door zwavel-compensatie t.o.v. het referentie-
mengsel met TESPT, om de crosslink-dichtheid gelijk te trekken. Het gebruik van 7,5 phr
ENR-51 samen met 4 gew% TESPT betrokken op de silica, en zwavel compensatie geeft
vulcanisaten met dezelfde niveaus treksterkte vergeleken met vulcanisaten met 8,6 gew%
TESPT betrokken op de silica. De aanwezigheid van ENR in de silica-gevulde NR
beïnvloedt de dynamisch mechanische eigenschappen. Boven de glas-rubber overgang
vertonen de ENR-bevattende vulcanisaten een hogere tan δvergeleken met het
silica/TESPT systeem. Toevoeging van TESPT aan de ENR-gecompatibiliseerde
mengsels veroorzaakt slechts een geringe verandering in Tg en tan δ bij zowel 5 oC en 60
o
Ctengevolge van een verhoogde crosslink-dichtheid. SEM microscopische opnamen
tonen duidelijk een goede silica dispersie in alle gecompatibiliseerdevulcanisaten, en de
breukvlakken bij trekken van de materialen tonen verschillende patronen van scheur-lijnen
afhankelijk van hun treksterktes.
Drie mengsels met ENR-51 in combinatie met drie types silaan: TESPT, 3-
octanoylthio-1-propyltriethoxysilaan (NXT) en 3-mercaptopropyl-di(tridecaan-1-oxy-13-
penta(ethyleenoxide))ethoxysilaan (VP Si-363) in de range van 2 – 5 gew% betrokken op
de silica, worden vergeleken in Hoofdstuk 5. Toevoegen van silaan reduceert de
mengsel viscositeit enigszins, maar verbetert duidelijk het chemische bound rubber
gehalte en de crosslinking/vernettings eigenschappen. Van de 3 verschillende ENR-
51/silaan combinaties geeft ENR/TESPT het hoogste chemischebound rubber gehalte en
trek-rek eigenschappen, terwijl de ENR/NXT combinatie weliswaar een hoger chemisch
bound rubber gehalte geeft maar vergelijkbare trek-rek eigenschappen als het ENR/VP Si-
363 systeem. De betere eigenschappen door gebruik van TESPT vergeleken met de
andere twee types silaan valt terug te voeren op de bis-(triethoxysilyl)- en tetrasulfide-

226
 Samenvatting

functionele groepen, die relatief meer ethoxy-groepen aanbieden voor reactie met de
silanol-groepen op het silica-oppervlak, en de reactieve vrije zwavel vrijgemaakt uit de
TESPT die deelneemt in de netwerk-vorming. Bij een vaste hoeveelheid silaancoupling
agent van 4,0 gew% betrokken op de silica, in combinatie met 7,5 phr ENR-51 als
compatibilizers in het silica-gevulde NR mengsel, leidt toevoegen van extra zwavel om te
corrigeren voor de totale zwavel-hoeveelheid ten opzichte van het referentie-mengsel met
TESPT tot verhoging van de molulus en trek-sterkte, terwijl de rek-bij-breuk afneemt. Het
gebruik van een ENR/silaan verhoogt de tan δ bij 5 oC en verlaagt de tan δ bij 60 oC
vergeleken met gebruik van alleen ENR-51, hetgeen een verbetering in natte tractie en
rolweerstand indiceert van banden gebaseerd op deze combinatie. Echter over het geheel
genomen geeft gebruik van een optimale dosering van alleen TESPT nog de beste
dynamisch mechanische eigenschappen vergeleken met de aanwezigheid van ENR als
compatibilizer.
Van de drie types bestudeerde silanen in combinatie met ENR-51 en zwavel
compensatie, geeft gebruik van 2,2 phr TESPT samen met 7,5 phr ENR-51 de beste
overall eigenschappen, maar evenaart nog niet volledig de eigenschappen van een
mengsel met de optimale hoeveelheid van alleen TESPT: 4,7 phr, zoals eerder gezien in
hoofdstuk 4. Evenwel, in het licht van de “Magische Driehoek van de Banden
Technologie”, waarin natte slipweerstand, rolweerstand en slijtage-weerstand alle moeten
worden geoptimaliseerd, kan de combinatie van ENR-51, TESPT en zwavel compensatie
een beter optie zijn voor “Winterbanden”, die een goede natte grip vereisen, ofschoon
echte banden-testen dat nog wel moeten bevestigen.
Met uitsluiting van ENR als compatibilizer worden de eigenschappen van silica-
gevulde NR met vier types silaancouplingagents, t.w. TESPT, NXT, VP Si-363 en
vinyltriethoxysilaan (VTES) onderzocht teneinde een mogelijke silaancoupling agent uit te
kiezen om later te worden gebruikt voor enting op NR moleculen: Hoofdstuk 6. Van de
vier silaan types geeft VTES over het geheel genomen de slechtste eigenschappen
vanwege z’n geringe afschermings-effect, en de vinyl-groep die moeilijk te koppelen is
met NR rubber ketens in aanwezigheid van zwavel. Dientengevolge is VTES geen
effectieve silaancoupling agent voor dit silica-gevulde NR systeem. Gebruik van NXT en
VP Si-363 vertoont een goede verbetering in mengsel-eigenschappen, t.w. meng-energie,
Mooney viscositeit, vulcanisatie-gedrag, bound rubber gehaltes, en een toename van
mechanische en dynamisch mechanische eigenschappen vergeleken met helemaal geen
silaan, maar de eigenschappen zijn inferieur t.o.v. gebruik van TESPT, voornamelijk
vanwege lagere crosslink-dichtheden vergeleken met TESPT, dat een intrinsiek “zwavel-
donerend” effect vertoont. Dit is verder niet nader onderzocht in de context van dit

227
Samenvatting

hoofdstuk. De incorporatie van NXT-silaan geeft betere scorch-safety, en toepassing bij

een gelijke hoeveelheid ethoxy-functionele groepen t.o.v. TESPT geeft vulcanizaten met
hetzelfde niveau treksterkte. Met betrekking tot banden-gedrag biedt NXT betere
eigenschappen dan VP Si-363, maar nog wel iets inferieur t.o.v. het referentie-mengsel
met TESPT.
Verderop worden NXT een VP Si-363 silaancouplingagents gebruikt om te enten
op NR, teneinde reactieve zij-groepen in te brengen voor interactie en/of reactie met de
silanol-groepen op het silica oppervlak. Hoofdstuk 7 laat zien dat de silaan-enting van de
natuur rubbers succesvol verloopt onder condities van smelt-mengenin een interne
o
menger bij 140 C, onder gebruikmaking van 0,1 phr 1,1’-di(tert-butylperoxy)-3,3,5-
trimethylcyclohexaan of Luperox® 231XL40als initiator. Op basis van element-analyse via
de SEM-EDX techniek bevatten de gemodificeerde NRs silicium (Si) atomen in hun
structuren, en de hoeveelheden zuurstof en silicium in de geënte NRs nemen toe met
hogere hoeveelheden silaan gebruikt voor de ent-reactie. Zowel VP Si-363-geënte NR als
NXT-geënte NR laten infrarood absorptie-banden zien bij 3270, 1075 en 1035 cm-1,
toegeschreven aan respectievelijkde deformaties van de –OH, Si-O-C en Si-O-Si
groepen, en de 1H NMR absorptie band bij 3,8 ppm, behorende bij het methyleen-proton
van de ethoxy-groepen.
Het gebruik van silaan-geënte NRs als compatibilizers in silica-gevulde mengsels
wordt besproken in Hoofdstuk 8 in vergelijking met het regelrechte gebruik van silaan in
de mengsel-formulering en gebruik van de optimale hoeveelheid TESPT. Toenemende
hoeveelheden van NXT- en VP Si-363-geënte NRs als compatibilizers in silica-gevulde
NR mengsels verlaagt de Mooney viscositeit, de vulstof-vulstof interactie of Payne effect,
de snelheidsconstante voor flocculatie en verhoogt de chemische bound rubber gehaltes,
300% modulus, versterkings index en treksterkte. De versterkingsindex en treksterkte van
de silica-gevulde NR vulcanizaten bereiken nagenoeg een plateau wanneer het silaan-
gehalte in de silaan-geënte NRs, met inbegrip van vrije en geënte silaan, meer bedraagt
dan 3,0 gew% betrokken op de silica. Breukvlakken onder rek van de silica-gevulde NR
vulcanizaten laten ductiele breuken zien met oppervlakte ruwheid en vele scheur-
lijnenwanneer de silaan-geënte NRs worden gebruikt, zoals in het geval van het mengsel
met TESPT.Bij dezelfde silaan hoeveelheid geeft gebruik van silaan-geënte NR als
compatibilizer een hogere verbetering in Payne effect, chemischbound rubber gehalte,
300% modulus en treksterkte in vergelijking met betere regelrecht gebruik van dezesilaan.
De aanwezigheid van NXT geeft een hogere verbetering van de overall eigenschappen
dan VP Si-363, maar beide liggen toch nog wel lager dan bij gebruik van de optimale
hoeveelheid TESPT. In aanwezigheid van silaan-geënte NR als compatibilizerverschuift

228
 Samenvatting

de Tg van de silica-gevulde NR vulcanizaten naar een hogere temperatuur. De tan δ bij 5

o
C, welke relateert aan de natte grip, neemt iets af, maar de tan δ bij 60 oC wordt duidelijk
verlaagd, hetgeen duidt op een lagere banden rolweerstand vergeleken met het
referentie-mengsel met TESPT. Zwavel compensatie voor demengsels met NXT-geënte
NR als compatibilizer wordt ook onderzocht om de eigenschappen van silica-gevulde NR
te verbeteren, zoals besproken in Hoofdstuk 9. De zwavel-gecompenseerde mengsels
vertonen een iets lagereMooney viscositeit, maar significant hoger vulcanisatiekoppel
vergeleken met het mengsel zonder extra zwavel. Met zwavel-compensatie t.o.v. het
zwavel-gehalte in het referentie TESPT mengsel, geeft het gebruik van NR-g-NXT met 6
gew% NXT t.o.v. de silica een verlaging van de Mooney viscositeit en het Payne effect,
evenals een verhoging van de 300% modulus en treksterkte van silica-gevulde NR tot
nagenoeg hetzelfde niveau als bereikt met regelrecht gebruik van de optimale
hoeveelheid TESPT, maar zonder significante effecten op de tan δ waardes bij 5 en 60
o
C. Aangezien de hoeveelheden van 8,6 gew% TESPT en 6 gew% NXT betrokken op de
silica min of meer gelijke molaire hoeveelheden vertegenwoordigen, geeft het gebruik van
NXT-geënte NR derhalve slechts de halve hoeveelheid ethanol-emissie t.o.v. van TESPT:
milieu-technisch een vriendelijker en veiliger systeem.
Het toenemende belang van silica-versterkte banden mengsels ten behoeve van
verlaging van het brandstof-verbruik van automobielen motiveert research en ontwikkeling
van de relevante rubber technologie. Naast de gewoonlijk gebruikte silica/silaan
technologie zijn alternatieve compatibilizers van belang om gedeeltelijk of volledig de
silaan te vervangen door “groenere” producten. Het huidige proefschrift presenteert een
diepgaande studie inde toepassing van ENRs als alternatieve compatibilizer. Gebruik van
ENRs alleen als compatibilizers laat reeds een verbetering zien van de silica-NR
compatibiliteit, de silica-dispersie en enige mechanische eigenschappen. Deze
eigenschappen kunnen verder worden verbeterd door het gebruik van ENR te combineren
met slechts kleine hoeveelheden silaancouplingagents en zwavel compensatie t.o.v. een
referentie mengsel met TESPT. Naast het meer algemene gebruik van ENRs biedt silaan-
geënte NR een ander alternatief. De bereiding van silaan-geënte NR in de gesmolten
meng-toestand is uitvoerbaar maar gecompliceerd. Het niet-gezuiverde product bevat
zowel vrije als geënte silaan. Desalniettemin geeft het gebruik van niet-gezuiverde silaan-
geënte NR als compatibilizers een verbetering in vulstof-rubber interacties en de
eigenschappen van silica-versterkte NR. Bij dezelfde totale hoeveelheid silaan geeft
gebruik van silaan-geënte NR over het geheel genomen betere eigenschappen dan het
regelrechte gebruik van silaan, en de eigenschappen kunnen nog verder verbeterd
worden door zwavel compensatie, gebruikmakend van het mengsel metoptimale

229
Samenvatting

hoeveelheid TESPT als referentie. Over het geheel genomen, zelfs ofschoon de
toepassing van chemisch gemodificeerde NRs in de vormen van ENRs en silaan-geënte
NR als compatibilizers zonder verdere silaan resulteert in mindere eigenschappen
vergeleken met het gebruik van state-of-the-art TESPT, biedt de juiste combinatie van
gemodificeerde NR samen met slechts geringe hoeveelheden TESPT en zwavel
compensatie een alternatief systeem met als voordelen gebruik van een op NR gebaseerd
product en minder ethanol-emissie afkomstig van de silanizerings-reactie. Bovendien
levert het gebruik van ENR een goede natte wet grip in het perspectief van de “Magische
Driehoek van de Banden Technologie”.

230
 Symbols and Abbreviations

Symbol Description

γ Strain
𝛾0 Maximum amplitudes of strain
𝛾𝑠 Surface free energy of fillers
𝛾𝑠𝑑 Surface free energy of dispersive component
𝑠𝑝
𝛾𝑠 Surface free energy of specific component
σ Stress
𝜎0 Maximum amplitudes of stress
𝛿 Phase angle between stress and strain
ϕ Volume fraction of filler
A Arrhenius factor
E0 Young’s modulus of unfilled rubber
E Young’s modulus of filled rubber
EA Activation energy
E′ Tension storage modulus
E′′ Tension loss modulus
G′ Shear storage modulus
Gʺ Shear loss modulus
Η Filler-filler networking factor
I Polymer-filler interaction parameter
ka Flocculation rate constant
M100 Tensile modulus at 100% strain
M300 Tensile modulus at 300% strain
M300/M100 Reinforcement index
o
ML1+4(100 C) Mooney viscosity at 4 min Of test time, after preheating for 1
min measured at 100oC using a large rotor
R Gas constant
T Temperature
tc90 Optimum cure time
Tg Glass transition temperatures
t Time
Symbols and abbreviations

ts1 Scorch time

x Degree of flocculation

Abbreviation Description

AIBN Azobisisobutylonitrile
ASTM American Society of Testing Materials
ATR-FTIR Attenuated Total Reflection-Fourier Transform Infrared
spectrometer
BET Brunauer-Emmett-Teller
BR Polybutadiene rubber
BTOS i-Butyltriethoxysilane
CBS N-cyclohexyl benzothiazole sulfenamide
CDCl3 Deuterated chloroform
CR Polychloroprene rubber
CSDPF Carbon-silica dual phase filler
CTAB Cetyltrimethyl ammonium bromide
CV Conventional vulcanization
DBP Dibutylphthalate
DCP Dicumyl peroxide
DMA Dynamic mechanical analysis
DMOD N,N-dimethyl-1-octadecylamine
DPG Diphenylguanidine
DSC Differential scanning calorimetry
ENR Epoxidized natural rubber
ENR-xx Epoxidized natural rubber with xx mol% epoxide content
EpBR Epoxidized polybutadiene
EPDM Ethylene-propylene-diene rubber
EPDM-g-VOS Vinyloxyaminosilane grafted ethylene-propylene-diene rubber
EPM Ethylene propylene rubber
EpSBR Epoxidized styrene-butadiene rubber
ESCA Electron spectroscopy for chemical analysis
ETU Ethylene thiourea
FT-IR Fourier-transform infrared spectroscopy
1
H-NMR Proton nuclear magnetic resonance spectroscopy
HA-NR High ammonia natural rubber latex

232
 Symbols and abbreviations

HV-BR High-vinyl polybutadiene rubber

IGC Inverse gas chromatography
IIR Isobutylene isoprene rubber
KPS Potassium persulfate
LLDPE Linear low density polyethylene
MFA Multifunctional additive
MNR Maleated natural rubber
MPS Mercaptopropyl-trimethoxysilane
NBR Acrylonitrile butadiene rubber
NR Natural rubber
NR-g-NXT Natural rubber grafted with NXT
NR-g-VP Si-363 Natural rubber grafted with VP Si-363
NRL Natural rubber latex
NSA Nitrogen surface area
NXT 3-Octanoyl-thio-propyltriethoxysilane
OTES N-octyltriethoxysilane
PDDA Poly(diallyldimethylammonium chloride)
PE Polyethylene
PP Polypropylene
PP-g-MA Poly(propylene-g-maleic anhydride) copolymer
RPA Rubber process analyzer
RSS Ribbed smoked sheet
SBR Styrene-butadiene rubber
SEM Scanning electron microscopy
SEM-EDX Scanning Electron Microscopy - Energy Dispersive X-Ray
Spectroscopy
Semi-EV Semi-efficient vulcanization
SMO Sorbitan monooleate
STSA Statistical thickness surface area
TBBS N-tert-butyl-2-benzothiazole sulfenamide
TDAE Treated distillate aromatic extract oil
TEOS Tetraethoxysilane
TESPT Bis-(triethoxysilylpropyl)tetrasulfide
TESPD Bis-(triethoxysilylpropyl)disulfide
TMQ 2,2,4-trimethyl-1,2-dihydroquinoline
VOS Vinyloxyaminosilane

233
Symbols and abbreviations

VP Si-363 3-Mercaptopropyl-di(tridecan-1-oxy-13-
penta(ethyleneoxide))ethoxysilane
VTES Vinyltriethoxysilane
VTMS Vinyltrimethoxysilane
VTOS Vinylethoxysilane
ZnO Zinc oxide

234
 Bibliography

Journal Articles

1. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer, Silica-

reinforced tire tread compounds compatibilized by using epoxidized natural
rubber, European Polymer Journal, 51, 69-79 (2014).
2. K. Sengloyluan, K. Sahakaro and J.W.M. Noordermeer, Silica-reinforced natural
rubber compounds compatibilized through the use of epoxide functional groups
and TESPT combination, Advanced Materials Research, 844, 272-275 (2014).
3. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer, Enhanced
properties by a combination of epoxidized natural rubber and silane coupling
agent of silica-reinforced natural rubber-based tire treads, European Polymer
Journal, submitted.
4. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer, Silica-
reinforced natural rubber compounds compatibilized by ENR in combination with
different silane coupling agent types, Kautschuk Gummi Kunststoffe, in
prepration.
5. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer,
Comparative reinforcement efficiency of silica in combination with different silane
coupling agents in natural rubber-based tire compounds, Rubber Chemistry and
Technology, in preparation.
6. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer,
Functionalized natural rubber by grafting reaction with silanes, Express Polymer
Letters, in preparation.
7. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer,
Compatibilization of silica-filled natural rubber by using silane-grafted-NR,
Rubber Chemistry and Technology, in preparation.

Presentations

1. K. Sengloyluan, K. Sahakaro, W. K. Dierkes and J. W.M. Noordermeer, Effect of

epoxidized natural rubber used as a compatibilizer in silica-filled natural rubber,
Dutch Polymer Days – 2012, Lunteren, the Netherlands, March 12-13, 2012.
Bibliography

2. K. Sengloyluan, K. Sahakaro and J.W.M. Noordermeer, Use of epoxidized

natural rubber as a compatibilizer in silica-filled natural rubber tire tread
compounds, Deutsche Kautschuk-Tagung (DKT) 2012, Nuremberg, Germany,
July 2-5, 2012.
3. K. Sengloyluan, K. Sahakaro and J.W.M. Noordermeer, Use of epoxidized
natural rubber as a compatibilizer in silica-filled natural rubber compounds, The
28th International Conference of Polymer Processing Society (PPS-28), Pattaya,
Thailand, December 11-15, 2012.
4. K. Sengloyluan, K. Sahakaro and J.W.M. Noordermeer, Use of epoxidized
natural rubber as a compatibilizer in silica-filled natural rubber compounds, The
3rd Sci-tech Grad Symposium, Prince of Songkla University, Pattani, Thailand,
March 20, 2013.
5. K. Sengloyluan, K. Sahakaro and J.W.M. Noordermeer, Silica-reinforced natural
rubber compounds compatibilized through the use of epoxide functional groups
and TESPT combination, The 1st Asia Pacific Rubber Conference (APRC) 2013,
Suratthani, Thailand, September 5-6, 2013.
6. K. Sengloyluan, K. Sahakaro, W.K. Dierkes and J.W.M. Noordermeer, Effect of
TESPT and extra sulfur on the properties of silica-reinforced natural rubber with
epoxidized natural rubber as compatibilizer, The 4th Polymer Conference of
Thailand (PCT-4), Bangkok, Thailand, March 20-21, 2014.

236
 Acknowledgement

I started my role as a Ph.D. student in September 2010 under a double degree

Ph.D. program between University of Twente in the Netherlands and Prince of Songkla
University, Pattani campus in Thailand with the financial supported from the Netherlands
Natural Rubber Foundation. During the past 4 years, I studied and worked for 2.5 years at
Department of Rubber Technology and Polymer Science (Rubber Tech.), Prince of
Songkla University (PSU), Thailand and for 1.5 years at Department of Elastomer
Technology and Engineering (ETE), University of Twente (UT), the Netherlands. From the
started, I have got lots of help and support from many people. I would like to thank and
express my sincere gratitude to all of them and institutions involved in this co-operation
program.
I may have not been succeeded as a Ph.D. without Prof. Dr. J.W.M.
Noordermeer and Dr. Kannika Sahakaro, who gave me a good opportunity to join in this
co-operation program. I would like to appreciate and express my deep sense of gratitude
to both of them for teaching, coaching and helping me throughout this period. I have
always appreciated in your high standard and quality of work. I have gained lots of
experiences, not only about academic knowledge but also work and lifestyle especially
when I was abroad. It is the most precious time for me to work under your supervision.
Thank you very much for being the promotors for my thesis. Dr. Kannika, I know you have
a hard time to train me to prepare and write articles and chapters in the proper way. Even
I discouraged to reach my point sometime, you always encouraged me with your active
suggestions and constant support, thank you very much.
I would like to appreciate and express my respect to Prof. Dr. A.Blume and Dr.
W.K. Dierkes for your help and support when I was a part of ETE group. Thank you very
much for investing your valuable time for suggesting and correcting my articles and
chapters. Even though we had only a few times for discussion, but every time I learned a
lot of things from your knowledge and experiences that made my thesis possible.
It is my pleasure to express my sincerely thank to all of staff from Rubber Tech.
and ETE groups for training, teaching and supporting me. Dear P-Somkid, P-Siradanai, P-
Apichart, P-Inyas P-Absorn and Vidchapon, thank you very much for your co-operation
and everything that you have done for me. Dear Jacob, you are absolutely incredible
person. I’m glad to have opportunity to know you. I always appreciate and remember what
Acknowledgement

you have done, including a correction of my English skill. Dear Yvonne and Ceciel, thank
you for helping and taking care of me.
I would like to appreciate and express my regards to Prof. Dr. G.P.M.R. Dewulf,
Prof. Dr. Ir. D.J. Schipper, Dr. Ir. P.J. Dijkstra, Prof. Dr. U. Giese, Prof. Dr. J. Vourinen, Dr.
Ir. L.A.E.M. Reuvekamp and Dr. A.V. Chapman for being my graduation committee.
I would like to convey my thankful to all of the former Ph.D. researchers in this
field. I learned a lot of things through all sentences in their theses. What they have done
before is really useful for my work.
Dear Dr. Sitisyidah Saiwaree and family, thank you very much for teaching,
supporting and taking care of me when I was in the Netherlands for the first time. P-
Prachid, P-Anee, P-Methakarn, P-Wisut and P-Ekwipoo, thank you very much for
supporting and encouraging me to pass a hard time during my education period. You are
a good example for me in both studying and working.
To all of my colleagues in ETE group; Salina, Ernest, Morteza, Somayeh, Andre,
Nadia, Gerda, Vignesh Kumar, Akansha, Monika, Katazyna and Indria, I had a good time
when I spent my time in the group with all of you. This is my first time to study and live
oversea, it is a challenge and adventure in the same time. Thank you very much for your
sincerely friendship and cheerful that you all giving me all the time. I appreciated to all of
activities that we had done together.
Thank you very much to Thai people in Enschede; P-May, P-Tom, P-Waew, P-
Pipe, P-Rex, P-Neung, Si, Gift, N-Meaw, N-Ham, N-May, N-Golf, N-Wa, you made my
days in oversea more colorful. Spending sometimes together for cooking and traveling
made me feel like we were family and helped me ease my loneliness. My special thanks
to André and Vignesh Kumar for being my paranymfen.
My sincere thanks to my colleagues in Thailand; P-Tuk, P-Pui, P-Ball, P-Alif, P-
Dee, Ang, Chalao, Hakim, Suradech, N-Dream, N-Peung and others who have become
my good friends. Special thank to my lovely and close friends; Nah and Da, we have
shared our happiness and sadness together. We have chosen to walk in the same fate, I
wish we will reach the goal together.
My utmost thanks go to my beloved family; my father: Nauy, my mother: Pin, my
sisters: Keaw and Koi and also my beloved relatives. Thank you for giving me your
unconditional love, being by my side and supporting me.

Karnda Sengloyluan
June, 2015

238
 Curriculum Vitae

th
Karnda Sengloylan was born on November 29 , 1985, in Satun, Thailand. She
received her B.Sc. degree in Rubber Technology in 2008 and Master degree in Polymer
Technology in 2010, from Prince of Songkla University, Pattani Campus, Thailand. She
finished her bachelor project under the topic of “Mechanical and dynamic mechanical
properties of NR/EPDM blends”. After she got her bachelor degree, she continued to
study her master in Polymer Technology program. The research title of her master project
was “Effect of silica and clay on rheological, mechanical and thermal properties of
thermoplastic vulcanizates based on epoxidized natural rubber and polypropylene blends”.
In September 2010, she started her Ph.D. under the supervision of Prof. Dr. J.W.M.
Noordermeer and Dr. Kannika Sahakaro in the co-operation program of University of
Twente in the Netherlands and Prince of Songkla University in Thailand. Her project was
focused on “Silica-reinforced natural rubber: use of natural rubber grafted with chemical
functionalities as compatibilizer”. This research was sponsored by the Netherlands Natural
Rubber Foundation.
Curriculum Vitae

240
“You can fool all the people some of the time, and some of the people all the time,
but you cannot fool all the people all the time.”

-----“Abraham Lincoln”-----

241