Preparation, structure, and properties of solutionpolymerized styrenebutadiene rubber with functionalized endgroups and its silicafilled composites

Polymer xxx (2014) 1e13

Contents lists available at ScienceDirect

Polymer
journal homepage: www.elsevier.com/locate/polymer

Preparation, structure, and properties of solution-polymerized
styrene-butadiene rubber with functionalized end-groups and its
silica-filled composites
Xiao Liu a, Suhe Zhao b, c, *, Xingying Zhang b, c, Xiaolin Li b, Yu Bai b
a
b
c

College of Materials Science and Engineering, Beijing University of Technology, Beijing 100124, China
Key Laboratory of Beijing City on Preparation and Processing of Novel Polymer Materials, Beijing University of Chemical Technology, Beijing 100029, China
Key Laboratory for Nanomaterials, Ministry of Education, Beijing University of Chemical Technology, Beijing 100029, China

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 30 October 2013
Received in revised form
10 January 2014
Accepted 28 February 2014
Available online xxx

With anionic polymerization, the solution-polymerized styrene-butadiene rubber (SSBR) and solutionpolymerized styrene-butadiene rubber with alkoxysilane-functionalization at two ends of macromolecular chains (A-SSBR) were synthesized by dilithium as initiator. The occurrences of end-group functionalization and condensation reaction were confirmed, but also the molecular structure parameters

and end-functionalized efficiency of A-SSBR grafted alkoxysilane groups onto the ends of its macromolecular chains were calculated through the characterizations. By this novel structural modification,
there were chemical bondings rather than conventional physical adsorption between silica and rubber
matrix. This novel technology was beneficial to not only immobilizing the free chain ends to decrease the
amount of macromolecular chains’ free terminals, but also chemically bonding the rubber chains on the
surfaces of silica particles to enhance the filler-polymer interaction significantly. Furthermore, the
covering layer of end-functionalized macromolecular chains around the silica particles was conducive to
reducing the silica agglomeration and improving the silica dispersion. The structures, morphologies, and
properties of SiO2/SSBR and SiO2/A-SSBR composites prepared by co-coagulation and mechanical
blending, were investigated. The results showed that SiO2/A-SSBR composites behaved better comprehensive performances including higher wet skid resistance and lower rolling resistance than SiO2/SSBR
composites. Consequently, A-SSBR was an ideal material for the green tire treads.
Ó 2014 Elsevier Ltd. All rights reserved.

Keywords:
Silica
Solution-polymerized styrene-butadiene
rubber (SSBR)
End-group functionalization

1. Introduction
Recently, with high attention to environmental protection and
saving resources, the reduction of fuel consumption of automobile
will play an effective role in protecting environment. As an
important part of automobile, tire plays an important role in
energy-saving and emission-reduction. In the process of vehicle
driving, the rolling resistance of tire is 20e30% of the total energy
consumption of automobile. Furthermore, the rolling loss of tread is
about 50% of the total energy consumption of tire. Accordingly, it is

* Corresponding author. Key Laboratory of Beijing City on Preparation and Processing of Novel Polymer Materials, Beijing University of Chemical Technology,
Beijing 100029, China. Tel.: þ86 10 6445 6158; fax: þ86 10 6443 3964.

E-mail addresses: , (X. Liu), zhaosh@mail.
buct.edu.cn (S. Zhao).

urgent for researchers to develop and produce the highperformance and energy-saving “green tire tread material” with
low rolling resistance and high wet skid resistance [1].
At the end of twentieth century, researchers found that the
hysteresis loss of rubber mainly originated from unconstrained “free
terminals” of macromolecular chains in the three-dimensional
crosslinked network of vulcanizate. Although they contribute to
tire tread with excellent wet skid resistance, lots of friction heat
produced by their random motion greatly increases the rolling
resistance. In order to lower the hysteresis loss, introducing the
functional groups which can either “passivate” free terminals or
react with reinforcing filler into the ends of macromolecular chains
becomes a research focus in recent years [2]. For instance, due to the
restriction of SneC bond on a part of free terminals of macromolecular chains, the properties of SSBR with end groups coupled by
SnCl4 such as high wet skid resistance, high wear resistance, and low
rolling resistance are significantly improved [3,4].

/>0032-3861/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

2

X. Liu et al. / Polymer xxx (2014) 1e13

With the excellent performances such as low hysteresis loss and
high reinforcement, nanosilica is a widely used filler for preparing

green tire treads in rubber industry [5e9]. However, due to a large
number of hydroxyl groups on the surfaces of silica particles, they
usually show high surface energy, easy self-aggregation, and poor
affinity with non-polar rubber macromolecules [10e13]. Therefore,
the improvements of silica dispersion and filler-polymer interfacial
interaction, as well as the reduction of filler-polymer interfacial
friction are the problems which many researchers have made great
efforts to resolve.
The common method to improve filler-polymer affinity is the
silica particles’ organic modification by a silane coupling agent [14e
18]. However, it is difficult to ensure each silica particle is organically modified by the molecules of silane coupling agent, and thus
the dispersion of silica particles in polymer matrix cannot achieve
the primary particle scale, i.e. 15e30 nm. Besides, the small-size
effect and quantum size effect of silica particle cannot be exhibited at all. To overcome these difficulties, the method of grafting
functional groups [19] which are able to react with silica particles
onto the free terminals of macromolecular chains can be considered. According to this assumption, silica particles will be adsorbed
or bonded on the terminals of rubber macromolecular chains. This
leads to the strengthened filler-polymer interfaces and “passivated”
free terminals of macromolecular chains, which can reduce the
contribution of random thermal motion to hysteresis loss. There are
many reported literatures about the end-group functionalization
such as ethylene-propylene-diene terpolymer (EPDM), polybutadiene, polyisoprene, polydimethylsiloxane, and polystyrene
terminated by glycidyl methacrylate [20], chlorophosphine [21], 1[2-(4-chlorobutoxy)ethyl] aziridine [22], (aminopropyl) dimethylsiloxy [23], and (tridecafluouo-1,1,2,2-tetrahydrooctyl) dimethylchlorosilane [24], respectively. As for the SSBR used for tread
material, the published achievements about its end-group functionalization are divided into several categories according to endfunctionalized reagent, including amide-type (N-phenyl-2pyrrolidone)
[25]
to
improve
storage
stability,
aminobenzophenone-type (4,40 -bis(diethylamino)-benzophenone)

[26,27] to improve rebound resilience or physical properties,
nitrile-type (chloroacetonitrile [28] or benzonitrile [29]) to improve
its affinity for carbon black or lower rolling resistance, fused-ring
polynuclear aromatic compound (benzanthracenes) [30] and
Schiff bases (dimethylaminobenzylidenemethylamine) [31] to
reduce hysteresis properties, and carbodiimide-type (dialkylcarbodiimides) [32] to improve impact resilience. All these endgroup functionalizations involve the carbon black-filled vulcanizates, in which the end-groups physically or chemically react with
carbon black to improve their properties, but for silica-filled system, it is a technical problem in the field of organiceinorganic
interaction which needs to be investigated. There have few related
studies on this.
In this study, two kinds of rubbers were synthesized by anionic
polymerization with initiator of dilithium. The one was normal
SSBR, the other was A-SSBR prepared by adding an endfunctionalized reagent, which can graft onto the end of macromolecular chain at one end and can react with silica particle at the
other end, to SSBR solution in the last stage of polymerization. One
part of A-SSBR solution was coagulated directly through removing
solvent to get the solid rubber, and the other part of A-SSBR solution was treated through the successive steps which include adding
a small amount of silica particles, condensation reaction, and cocoagulation to obtain an A-SSBR/SiO2 co-coagulated rubber. Their
molecular structure parameters were characterized and the endfunctionalized efficiency values of A-SSBR were calculated. The
morphological structure, bound rubber content, crosslink density,
glass-transition characteristics, rheological properties, mechanical

properties and other dynamic properties of three rubbers
(including one SSBR and two A-SSBRs) filled with silica were
investigated respectively. Besides, the mechanism and physics of
the structure formation and the relationship with the properties
were analyzed in detail. It is expected that these experimental results can provide the theoretical basis and novel design for preparing a new nanocomposite with excellent performances and
potential superiorities as green tire tread material.
2. Experimental
2.1. Materials
Styrene (analytical reagent) was from Beijing Chemical Reagents
Company (Beijing, China). Butadiene (industrial grade) and cyclohexane (analytical reagent) were supplied by Beijing Yanshan

Petrochemical Co., Ltd. (Beijing, China). Tetrahydrofuran (THF,
analytical reagent) and ethanol (analytical reagent) were purchased
from Beijing Chemical Works (Beijing, China). Butyl dilithium
initiator was self-made in laboratory. g-chloropropyl trimethoxy
silane (CPTMO, industrial grade) and nitrogen (!99.999%) were
provided by Qufu Wanda Chemical Co., Ltd. (Shandong, China) and
Beijing Shunanqite Gas Company (Beijing, China), respectively.
Precipitated silica (Tixosil 383) with an average particle diameter of
20e40 nm and specific surface area of 100e200 m2/g came from
Qingdao Rhodia Co., Ltd. (Shandong, China). The other rubber additives, such as zinc oxide, stearic acid, and sulfur, were commercial
grades.
2.2. Formula
The formula of all the vulcanizates was as follows: 100 parts
SSBR, 30 parts precipitated silica, 4 parts zinc oxide, 1 part stearic
acid, 1.5 parts polymerized 2,2,4-trimethyl-1,2-dihydroquinoline,
1.2 parts benzothiazyl disulfide, 1.2 parts diphenyl guanidine, 1 part
triethanolamine, and 1.8 parts sulfur.
2.3. Synthesis and preparation
2.3.1. Purification
The reaction conditions of anionic polymerizations were so severe that only a few of impurities such as hydrogen and water in the
system can terminate the reaction; thus it was necessary to purify
the monomers, solvents, and other chemicals. In this experiment,
the styrene as a monomer was bathed in calcium hydride for 24 h
followed by reduced pressure distillation, and then was stored with
seal to avoid light under the environment of high purity nitrogen
at À15  C. The THF, a structure regulator, was purified by using the
same methods as the styrene except for the atmospheric distillation. The cyclohexane as a solvent was bathed in calcium hydride
for 24 h followed by atmospheric distillation to collect the fraction
at 65e70  C, and then the sodium wire was put into it to remove the
micro-water before storing with seal under the environment of

high purity nitrogen, which was bubbled into the solvent for 15 min
before experiment to remove a small amount of hydrogen. The
CPTMO as an end-functionalized reagent was bathed in calcium
hydride for an hour followed by atmospheric distillation to collect
the fraction at 78e80  C, and then was stored with seal to avoid
light under the environment of high purity nitrogen.
2.3.2. Synthesis and numbering
First, the entire polymerization plant was cleaned by both
high purity nitrogen and reactive polymer to ensure the reaction
conditions of anionic polymerization. Second, the butyl dilithium
initiator was synthesized through a small quantity of butadiene

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

X. Liu et al. / Polymer xxx (2014) 1e13

initiated by the ready-made naphthalene-lithium at 25  C for 2 h
in cyclohexane solution. Next, the cyclohexane, styrene, butadiene, and THF were added successively in a 2L reaction vessel
which was purified by butyl dilithium as a purifying agent at
room temperature, followed by adding the initiator (butyl
dilithium). The polymerization lasted 3 h at 50  C with a stirring
speed of 250r/min. Finally, SSBR was prepared by adding alcohol
to SSBR solution to terminate the reaction. The two solutionpolymerized styrene-butadiene rubbers with alkoxysilane functionalized end-groups, i.e. A-SSBR-1 and A-SSBR-2, were prepared by adding CPTMO to SSBR solution in the last stage of
polymerization and then reacting for half an hour at 65  C. The
amounts or concentration of all the used chemicals are listed in
Table 1.
To examine the kinetics of anionic polymerization of SSBR, ASSBR-1, and A-SSBR-2 in detail, polymerizations were monitored at
50  C. The conversionetime relationships (the total conversion,
styrene conversion, and butadiene conversion as a function of time

with an interval of 30 min) were obtained by determining the
amount of unconsumed monomers and polymerization product.
For SSBR, the total, styrene, butadiene conversion respectively were
66.3%, 63.6%, 67.2% (t ¼ 30 min), 81.2%, 79.4%, 81.8% (t ¼ 60 min),
91.7%, 90.5%, 92.1% (t ¼ 90 min), 95.4%, 95.1%, 95.5% (t ¼ 120 min),
and 98.9%, 98.6%, 99% (t ¼ 150 min). For A-SSBR-1, the total, styrene,
butadiene conversion respectively were 63.7%, 60.4%, 64.8%
(t ¼ 30 min), 79.7%, 77%, 80.6% (t ¼ 60 min), 90.9%, 90.3%, 91.1%
(t ¼ 90 min), 95.2%, 94.9%, 95.3% (t ¼ 120 min), and 98.8%, 98.5%,
98.9% (t ¼ 150 min). For A-SSBR-2, the total, styrene, butadiene
conversion respectively were 65.8%, 62.5%, 66.9% (t ¼ 30 min),
82.5%, 80.1%, 83.3% (t ¼ 60 min), 93.1%, 92.5%, 93.3% (t ¼ 90 min),
96.2%, 95.9%, 96.3% (t ¼ 120 min), and 98.9%, 98.9%, 98.9%
(t ¼ 150 min).
1
H NMR (CDCl3): d ¼ 6.85e7.40 (aromatic proton in each
random-copolymerized styrene unit), 6.20e6.85 (aromatic proton
in each block-copolymerized styrene unit), 5.50e5.60 (aCHe proton in each 1, 2-butadiene structural unit), 5.37e5.50 (eCHa and
aCHe proton in each 1, 4-butadiene structural unit), 4.79e4.99
(aCH2 proton in each 1, 2-butadiene structural unit), and 3.40e
3.60 ppm (proton in eSie(OCH3)3) [33e36].
FTIR (KBr): 3080e3020 (CeH stretching vibration peak of benzene), 1495e1453 (skeleton vibration peak of benzene ring), 968
(CeH bending vibration peaks of polybutadiene’s trans-1,4 structures), 910 (CeH bending vibration peaks of polybutadiene’s vinyl
structures), 728 (CeH bending vibration peaks of polybutadiene’s
cis-1,4 structures), 1178 and 1090e1020 cmÀ1 (eSieOeC stretching
vibration peaks) [37].
Thereafter, co-coagulated SiO2/A-SSBR-1 was prepared through
a successive process of adding 5 hr (parts per hundred of rubber)
silica powder to rubber solution, stirring and reflux at 85  C for 3 h,

Table 1
Amounts of all the used chemicals.
Chemical

Cyclohexane (g)
Styrene (g)
Butadiene (g)
THF (ml)
Butyl dilithium (purifying agent) (ml)
Butyl dilithium (initiator) (ml)
CPTMO (end-functionalized reagent) (ml)
Monomer concentration (%)
Molar ratio of THF to active center
Molar ratio of and CPTMO to active center

Compound no.
SSBR

A-SSBR-1

A-SSBR-2

1161.5
23
89.3
7
1.12
3.45
0
9.67

68.18:1
0

1183.9
31.1
122.8
7
1.87
4.85
0.33
13
48.5:1
1:1

1816.4
37.3
149.7
8.5
1.49
6
0.41
10.3
47.6:1
1:1

3

and removing solvent. A-SSBR-2 solid rubber was obtained by a
direct co-coagulation and then removing solvent.
The samples are identified as follows:

1#-Adding 30 phr silica powder to SSBR by mechanical
blending;
2#- Adding the rest silica powder to SiO2/A-SSBR-1 cocoagulated rubber by mechanical blending (the total amount
of silica was kept constant at 30 phr);
3#- Adding 30 phr silica powder to A-SSBR-2 by mechanical
blending.
2.3.3. Mixing and vulcanization
The mixing of silica with rubber was carried out on a KGSA11
Haake internal mixer (Xiamen Rectifier Co., Ltd, Fujian, China) with
a volume of 55 ml and a rotating speed of 10r/min. The torque value
as a function of time was recorded to investigate the condensation
reaction. The other rubber additives were added to rubber in a 6inch open mill (Zhanjiang Machinery Plant, Guangdong, China) by
the conventional mixing technique.
A XLB-D350 Â 350 plate vulcanization machine (Huzhou
Dongfang Machinery Co., Ltd, Zhejiang, China) was used to prepare vulcanizates, and the curing condition was 150  C Â t90(cure
time). The hydraulic pressure was 15 MPa on the mould and each
vulcanizate had a thickness of about 2 mm. The cure times of
rubber compounds were determined at 150  C with a P3555B2
oscillating disk rheometer (Huanfeng Chemical Technology and
Experiment Machine Plant, Beijing, China). About 8 g of rubber
compound was used for each test and a 1 arc oscillating angle
was applied.
2.4. Characterization of structure and properties
2.4.1. Gel permeation chromatography (GPC)
The number-average molecular weight (Mn), weight-average
molecular weight (Mw), and polydispersity index (Mw/Mn) of
the synthesized copolymers were measured by using a Waters150C gel permeation chromatograph (Waters Corporation, United
States) with three Waters Styragel columns (pore size 102, 103, and
104 Å, respectively) in series calibrated by narrow polystyrene
standard with molecular weight ranging from 2.2 Â 103 to

5.15 Â 105 g/mol. THF was used as the eluent at a flow rate of 1.0 mL/
min at 40  C.
2.4.2. 1H nuclear magnetic resonance (1H NMR)
The characteristic groups including functionalized end-groups
of polymers were tested by 1H NMR measurement carried out on
a Bruker AV600 high-resolution NMR spectrometer (Bruker Corporation, Bremen, Germany) with a frequency of 600 MHz at room
temperature (25  C). The polymer samples were dissolved in CDCl3
in a 5 mm NMR tube. Chemical shifts were reported in ppm and
referenced to tetramethylsilane (TMS) as an internal standard and
calculated by using the residual isotopic impurities of the deuterated solvents.
2.4.3. Fourier transform infrared (FTIR) spectrometry
FTIR spectra were recorded on a Tensor-37 FTIR spectrometer
(Bruker Optik Gmbh, Germany) at room temperature. The rubber
samples were extracted by boiling ethanol for 72 h, and then the
extraction products were dried to a constant weight in a vacuum
drying oven followed by dissolving in organic solvent at a concentration of about 10%. The sample films were prepared by
spreading a small amount of rubber solution on a KBr pellet uniformly after the evaporation of solvent. In all cases, 64 scans in a

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

4

X. Liu et al. / Polymer xxx (2014) 1e13

wavenumber range of 400e4000 cmÀ1 at a resolution of 0.6 cmÀ1
were used to record the spectra.
2.4.4. Energy dispersive X-ray spectroscopy (EDS)
The elemental distributions on the surfaces of sample residues
were analyzed by a Hitachi S-4300 field emission scanning electron

microscope (FE-SEM) equipped with a Genesis-60 energy dispersive spectrometer (EDAX Inc., United States).
2.4.5. Transmission electron microscopy (TEM) observation
The micrographs of vulcanizates were observed by a Hitachi H800-1 transmission electron microscope (Hitachi Corporation,
Tokyo, Japan) with an acceleration voltage of 200 kV and a
magnification of 5 Â 104. The samples were ultramicrotomed
at À100  C under liquid nitrogen cooling to give the ultrathin
section with a thickness of 70e90 nm, and then were placed onto a
200 mesh cooper grid coated with carbon film.
2.4.6. Mechanical properties
The tensile strength, tear strength, Shore A hardness, and dynamic compression properties of vulcanizates were measured according to ASTM D412 (dumbbell shaped), ASTM D624 (right-angle
shaped), ASTM D2240, and ASTM D395, respectively. The tensile
and tear strengths of the samples prepared from hot pressed sheets
were clamped at the both ends and pulled in uniaxial elongation
with a CMT4104 electrical tensile tester (Shenzhen SANS, Guangdong, China) at 25 Æ 2  C, with a constant crosshead speed of
500 mm/min and an initial gauge length of 25 mm. The Shore A
hardness and dynamic compression properties of vulcanizate were
measured by an XY-1 rubber hardness apparatus (4th Chemical
Industry Machine Factory, Shanghai, China) and a YS-25 compression fatigue testing machine (Shanghai Chemical Machinery No.4
Factory, Shanghai, China), respectively. The dynamic compression
measurement lasted 25 min at 55  C with a load of 1.01 MPa, a
compression stroke of 4.45 mm, and a compression frequency of
1800 minÀ1. During tensile, tear, and dynamic compression test,
five, five, and three specimens were tested to give the average
value, respectively, and during the hardness test, the hardness
values of three different sample (over 6 mm in thickness) spots
were measured to give the average value.
2.4.7. Differential scanning calorimetry (DSC)
The determination of glass-transition temperature (Tg) to assess
the interfacial bonding was carried out on a STARe system differential scanning calorimeter (Mettler-Toledo, Switzerland). The
curves for samples (3e6 mg) were obtained by heating sample

from À80 to 40  C at a rate of 10  C/min under nitrogen atmosphere. Appearing as a step in the baseline or heat capacity (Cp), the
Tg could be calculated by either the half height of the Cp step, the
onset of the transition obtained by extrapolating the tangent of the
inflection point to the initial baseline, the inflection point of the
step, or the 1/2 DCp between the baselines. In our case, Tg was
estimated by the inflection point of the step.
2.4.8. Dynamic mechanical analysis (DMA)-temperature sweep
The storage modulus (G0 ) and internal friction loss (tand) as a
function of temperature were measured by a DMTA V dynamic
mechanical thermal analyzer (Rheometrics Scientific Inc., Piscataway, New Jersey, United States) with rectangular tension mode of
deformation. The measurements were carried out at a frequency of
10 Hz, a heating rate of 3  C/min, and a double strain amplitude of
0.1% over a temperature range of À100 to 100  C. Each sample was
30 mm in length, 6 mm in width, and 2 mm in thickness. The Tg
value was taken to be the maximum of the tand versus temperature
curve.

2.4.9. Rubber process analysis (RPA)-strain sweep
Strain sweep experiments (G0 and tand as a function of scanning
strain) were performed on vulcanizates by a RPA2000 rubber process analyzer (Alpha Technologies Corporation, Akron, Ohio, United
States) at 60  C. The strain amplitude (ε%) was varied from 0.28 to
100% and the frequency was 1 Hz.
2.4.10. Bound rubber content
About 2 g rubber compound was cut into small pieces followed
by being placed in a steel wire mesh with an average pore diameter
of 75 mm and then was dissolved in toluene solvent. Bound rubber
content was determined by extracting the unbound materials such
as ingredients and free rubbers with toluene for 3 days and acetone
for 1 day followed by drying for 2 days at room temperature until a
constant mass value. The toluene was changed every 24 h. The

weights of samples before and after the extraction were measured
and the bound rubber contents were calculated according to the
equation [38]:

h
ii. h
i
h


Wt mr = mf þ mr
Rb ð%Þ ¼ 100 Â Wfg À Wt mf = mf þ mr
(1)
where Rb was the bound rubber content, Wfg was the weight of filler
and gel, Wt was the weight of sample, mf was the fraction of filler in
the compound, and mr was the fraction of rubber in the compound.
2.4.11. Rheological properties
The viscosity (h) and non-Newtonian index (n) of rubber compounds at various shear rates (g) were determined by an Instron3211 capillary rheometer (Instron Corporation, UK) at 100  C
under a shear rate ranging from 1 to 104 sÀ1, and the samples were
preheated for 10 min before the measurement. The capillary die
had a diameter of 0.1595 cm as well as a length of 2.5557 cm, and
the plunger speeds varied at 0.06, 0.2, 0.6, 2.0, 6.0, 20.0 cm/min.
2.4.12. Crosslink density (XLD)
XLD measurements were carried out on a XLDS-15 crosslink
density analyzer and NMR spectrometer (IIC Innovative Imaging
Corporation, Blieskastel, Germany) with a magnetic field intensity
of 15 MHz at 80  C. Rubber sample with a length of 8 mm and a
diameter of approximately 5 mm was placed into a glass tube for
the measurement. Totally 64 measurements at different values
were carried out for determining the relaxation time. Data analysis

was performed according to the IIC Analysis Software package,
using a non-linear MarquardteLevenberg algorithm.
3. Results and discussion
3.1. Structure and characterizations of A-SSBR
3.1.1. Mechanism and physics of structure formation of A-SSBR/SiO2
composite
In this experiment, each polymer was synthesized through
anionic polymerization with monomers of styrene and butadiene
and initiator of dilithium. As far as A-SSBR was concerned, the eSie
(OCH3)3 groups were grafted onto the two ends of polymer
macromolecular chains after adding CPTMO to polymer solution in
the last stage of polymerization. After end-group functionalization,
A-SSBR can react with silica particles by the condensation reaction
between A-SSBR’s eSie(OCH3)3 groups and silica’s eSieOH groups
at 85  C for 3 h. The structural sketch of synthesis process for ASSBR and condensation reaction between A-SSBR and silica are
shown in Fig. 1.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

X. Liu et al. / Polymer xxx (2014) 1e13

5

Fig. 1. Sketch of the synthesis process for A-SSBR and condensation reaction between A-SSBR and silica.

The structure of the final composite with excellent performances was formed through three steps which were illustrated in
this schematic representation in detail. The first step is conventional anionic polymerization of SSBR. The only difference is the
initiator (butyl dilithium), which reacted with suitable monomers
including butadiene and styrene to form a polymer chain with two

anionic sites. The second step is end-group functionalization of
SSBR. The selected end-functionalized agent (CPTMO) was added to
SSBR solution in the last stage of anionic polymerization. Its role is
to obtain the siloxane-functionalized SSBR through chemically

reacting with the active center at the end of macromolecular chains
after the monomers are consumed, providing the chemical basis of
subsequent condensation reaction with silica particles. The third
step is condensation reaction between A-SSBR and silica particles.
The mechanism of this condensation reaction is essentially the
same as that for organic modification of silica particles by CPTMO as
silane coupling agent. The function of siloxane groups at the end of
macromolecular chains is similar to silane coupling agent. The
condition of reacting at 85  C for 3 h is chosen to ensure the sufficient condensation reaction. For A-SSBR/SiO2 compound, the end-

Fig. 2. 1H NMR spectra of rubbers (a) SSBR (b) A-SSBR-1 (c) A-SSBR-2 (d) SiO2/A-SSBR-1.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

6

X. Liu et al. / Polymer xxx (2014) 1e13

functionalized polymer macromolecular chains are chemically
bounded to the silica particles through eSieOeSie bonds, which in
turn decrease the amount of macromolecular chains’ free ends. In
this way, there are chemical bondings rather than conventional
physical adsorption between A-SSBR and silica after adding silica to
rubber matrix and condensation reaction at high temperature. The

eSieOeSie(CH2)3e bonds herein play a role in linking silica and
rubber matrix. This particular kind of filler-polymer interaction can
be beneficial to the improvements of A-SSBR/SiO2 composites’
comprehensive performances.
3.1.2. 1H NMR characterization and end-functionalized efficiency of
A-SSBR
After extracted by boiling ethanol solvent for 72 h, SSBR, ASSBR-1, and A-SSBR-2 rubber samples were measured by 1H NMR,
which spectra are shown in Fig. 2 (a)e(c) respectively.
According to the reported equations [33], styrene content and
vinyl content can be calculated. Moreover, the molecular-weight
parameters can be determined by GPC measurement. All the obtained structural parameters of the synthesized SSBR, A-SSBR-1,
and A-SSBR-2 are listed in Table 2.
All Fig. 2 (a)e(c) spectra exhibit no signal peak in the chemical
shift of 6.20e6.85 ppm, indicating that the styrene units are
randomly distributed rather than block-copolymerization for these
three rubbers [39]. Also from Fig. 2 (a)e(c) spectra, in the chemical
shift of about 3.562 ppm, i.e. the position of hydrogen atoms of e
Sie(OCH3)3, both Fig. 2 (b) and (c) exhibit one peak, but Fig. 2 (a)
not, indicating that the alkoxysilane groups were successfully
grafted onto the ends of macromolecular chains of A-SSBR.
For each group in 1H NMR spectra, the ratio of its signal peak
area to its hydrogen atom number is equivalent. According to this
principle, the ratio of the number of chain ends containing
trimethoxyl-silylpropyl to total chain ends namely the endfunctionalized efficiency therefore can be calculated by the ratio
of the peak areas of benzene-H to alkoxysilane-H from the 1H NMR
spectra, and the related equation is listed as follows:

SBenzeneÀH =SAlkoxysilaneÀH ¼ ð5 Â St% Â Mn=MS Þ=ð9 Â 2 Â EÞ

(2)

SBenzene-H, SAlkoxysilane-H, St%, Mn, MS, and E represent the peak area
of hydrogen atoms of benzene, peak area of hydrogen atoms of
alkoxysilane, styrene content, number-average molecular weight of
polymer, molecular weight of styrene, and end-functionalized efficiency, respectively. Moreover, “5” and “9” are respectively the
hydrogen atom numbers of benzene and alkoxysilane, i.e., eSie
(OCH3)3, and “2” implies the two functionalized ends of macromolecular chains. The calculated values of end-functionalized efficiency are shown in Table 3.
In addition, 1H NMR spectrum of SiO2/A-SSBR-1 co-coagulated
rubber is depicted in Fig. 2 (d) to investigate whether the

Table 3
End-functionalized efficiency values of SSBR and A-SSBR.
Sample no.

SSBR

A-SSBR-1

A-SSBR-2

End-functionalized
efficiency (%)

0

75.5

71.0

condensation reaction between A-SSBR-1 and silica is carried out.

Compared NMR spectrum of A-SSBR-1 (Fig. 2 (b)) with that of SiO2/
A-SSBR-1 (Fig. 2 (d)), it can be seen that the two curves are similar
except for the peak at 3.562 ppm in Fig. 2 (b), indicating that the e
Sie(OCH3)3 groups of A-SSBR-1 disappear after the addition of silica
powder. This only is caused by the condensation reaction between
eSie(OCH3)3 groups of A-SSBR-1 and eSieOH groups of silica under the condition as described in Section 2.3.2, similarly according
to the reported mechanism [40]. Thus, it demonstrates that the
condensation reaction occurs and the silica-rubber chemical
bondings are achieved.
3.1.3. FTIR spectrometry characterization
FTIR spectra of SSBR and A-SSBR are displayed in Fig. 3. It can be
seen that both SSBR and A-SSBR spectra exhibit all of the characteristic peaks mentioned in Section 2.3.2, but only A-SSBR spectrum
has eSieOeC stretching vibration peaks appeared at 1090e1020
and around 1178 cmÀ1, indicating the occurrence of end-group
functionalization reaction.
3.1.4. EDS characterization
SSBR, A-SSBR-1 and A-SSBR-2 samples were heated in an
alumina crucible at 600  C for 6 h, and then the residues were
measured by EDS characterization. The data for Si element content
are listed in Table 4. From Table 4, the Si element contents of the
two A-SSBR samples are much higher than that of SSBR sample; the
Si element contents of A-SSBR-1 sample is slightly higher than that
of A-SSBR-2 sample. This result is in accordance with the endfunctionalized efficiency calculated by 1H NMR. The high Si
element content of A-SSBR sample implies that it can only be
derived from CPTMO-functionalized end groups, but low Si
element content of SSBR sample can only be from impurities,
confirming the grafting reaction of CPTMO onto the ends of
macromolecular chains of A-SSBR.

Table 2

Structural parameters of SSBR and A-SSBR.
Parameter

a

Mn g/mol
Mw g/molb
Polydispersity index ac
Styrene content (%)
Vinyl contentd (%)

Compound no.
SSBR

A-SSBR-1

A-SSBR-2

140,852
199,189
1.41
19.4
48.5

180,685
255,700
1.42
20.7
49.0

173,457
235,843
1.36
18.5
50.7

a

Number-average molecular weight.
Weight-average molecular weight.
c
Ratio of weight-average molecular weight to number-average molecular
weight.
d
Content of 1, 2-butadiene structure.
b

Fig. 3. FTIR spectra of SSBR and A-SSBR.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

X. Liu et al. / Polymer xxx (2014) 1e13
Table 4
Silicon element contents of SSBR and A-SSBR.
Sample no.

SSBR

A-SSBR-1

A-SSBR-2

Initial mass of sample (g)
Mass of sample after heated (g)
Mass fraction of silicon element
displayed in EDS (%)
Mass of silicon element in
sample (g)

0.1500
0.0001
6.41

0.1500
0.0006
48.08

0.1500
0.0006
46.58

6.41 Â 10À6

2.88 Â 10À4

2.79 Â 10À4

From the above three characterization results, the accurate
structural information can be obtained. The conclusion that A-SSBR

is indeed terminated by eSie(OCH3)3 groups is drawn, demonstrating that alkoxysilane functionalizations on the ends of
macromolecular chains are achieved.
3.2. Processability
3.2.1. Reaction characteristics in mixing process
The torqueetime curves of mixing A-SSBR with silica powder in
Haake internal mixer at 50  C and 90  C are shown in Fig. 4. A
significant fluctuation of the two torque curves can be seen in the
first 10 min, which is the time of adding silica powder to rubber
matrix. In Fig. 4(a), the torque value directly keeps constant after
fluctuation, but in Fig. 4(b) the torque value shows a peak value in
the 20e27th minute and then remains constant, probably owing to
the condensation reaction between alkoxysilane groups on the
ends of macromolecular chains and hydroxyl groups of silica particles. This indicates that mixing at 90  C is helpful to condensation
reaction between silica and alkoxysilane groups of A-SSBR. Besides,
it is also possible to display good rheological characteristics.
3.2.2. Rheological properties
The rheological curves of 1#, 2#, and 3# rubber compounds are
shown in Fig. 5. As is clearly displayed in Fig. 5, all these rubbers
belong to shear-thinning non-Newtonian fluid. In heg curves, all
the viscosity values decrease as the shear rate increases, and the
three curves are approximately parallel to each other, indicating the
similar shear sensitivities. The highest viscosity of 1# rubber is
exhibited, namely the poorest flowability and highest energy consumption in processing. It is probably because the strong interaction among silica particles [41] of 1# rubber results in the formation
of filler-aggregates which occlude a part of rubbers [42] and increase the effective volume of filler [43], leading to slip and relaxation difficulties in the flow process of macromolecular chains. On
the contrary, due to the low occluded rubber content and the weak

7

interaction among silica particles induced by the fewer silanol
groups [6] after end-group functionalization, the lower viscosity

values of 2# and 3# rubbers are exhibited, which manifest the
better processability than 1# rubber.
In neg curves, the non-Newtonian index values of 2# and 3#
rubbers decrease significantly when the shear rate increases,
manifested as the non-Newtonian behavior. This may be caused by
2# and 3# rubbers’ strong filler-polymer interaction, which also
results in good filler dispersion [44], besides may be caused by the
diversified conformation of rubber macromolecules in the flow
process.
3.3. Filler-polymer interaction
3.3.1. Effect of macromolecular chain terminals passivated by silica
particles on Tg
DSC curves of SiO2/A-SSBR-1 co-coagulated rubber and A-SSBR1 pure rubber are shown in Fig. 6. A single transition in the temperature range from À50  C to À30  C with Tg at À38.5  C for ASSBR-1 and that with Tg at À36.7  C for SiO2/A-SSBR-1 are
observed, i.e. an increment of nearly 2  C after adding silica to
rubber. This may be partly caused by SiO2/A-SSBR-1’s strengthened
interfacial bonding, and partly by an extended crosslinked network
among reactive eSi(OCH3)3 groups [45]. It enables the increased
resistance to the slippage and motion of polymer segments [46],
the decrease of chain mobility [47], and the enhancement of Tg.
3.3.2. Bound rubber content
The bound rubber content is affected by the filler-polymer
interaction [48]. The bound rubber contents of 1#, 2#, and 3#
rubber compounds are displayed in Table 5. In Table 5, the higher
bound rubber contents of 2# and 3# rubber compounds are
exhibited, indicating more chemical bondings between macromolecular chains and silica particles. This performance is the result of
alkoxysilane functionalization on the ends of macromolecular
chains, but also can contribute good mechanical properties to the
corresponding vulcanizates discussed in Section 3.5. Also by
contrast in Table 5, the bound rubber content of 2# rubber compound is the highest, which is close to that of 3# and that of 1# is
the lowest. This result is directly proportional to the result of endfunctionalized efficiency; further illustrating the end-group functionalization is an effective way to strengthen filler-polymer

interaction. A similar literature reported by Mélé [15] showed
that the amount of bound rubber increased with the addition of the
silane coupling agent in silica/SBR compounds and this could result
from the increase of the specific surface induced by the better
dispersion of fillers.

Fig. 4. Torqueetime curves of mixing A-SSBR with silica powder in Haake internal mixer (a) 50  C (b) 90  C.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

8

X. Liu et al. / Polymer xxx (2014) 1e13

Fig. 5. Rheological curves of rubber compounds.

3.3.3. Crosslink density
To further study the filler-polymer interaction, the data of 1H
NMR relaxation parameters measured by a NMR crosslink density
analyzer are listed in Table 6. 1H NMR relaxation is produced by
intermolecular and intramolecular magnetized-dipole interaction
[49]. When the testing temperature is above Tg, the intermolecular
dipole interaction can be neglected and hence the main part is
intramolecular dipole interaction, which is affected by surroundings and can reflect molecular activity ability. Signal decay data and
molecular weight between crosslinking points (Mc) are analyzed
according to a Gaussian-exponential function [50] and a formula
reported by Kuhn [51], respectively.
From Table 6, the lowest physical crosslink density and the
highest chemical crosslink density of 2# rubber are exhibited,

indicating that the crosslinking bond formed by condensation reaction between functionalized macromolecules and hydroxyl
groups of silica particles can increase the amount of chemical
bonding and can reduce the physical filler-polymer adsorption and
the physical entanglement among macromolecular chains. Also in
Table 6, the Mc, T2, A(T2) values of all the vulcanizates are lower and
the A(Mc) values of all the vulcanizates are higher than those of the
corresponding rubber compounds. This result shows that the
vulcanization can bring on more crosslinking points, fewer activity
units, and lower activity ability.

Also from Table 6, the highest A(Mc) values and lowest A(T2)
values of 2# rubber are exhibited; the second is 3# and the last is
1#. It demonstrates that the larger amount of chemical bonding
derived from condensation reaction between silica and functionalized polymer corresponds to the larger crosslinking point amount
and the lower mobile fraction. It is proved that the free movement
of the molecular chain ends is restrained after the macromolecular
chains’ end-group functionalization and condensation reaction
with silica. Furthermore, the result of the lowest Mc values for 2#
rubber also implies its densest chemical crosslinking and strongest
interfacial bonding between silica and A-SSBR. This is in accordance
with the data of bound rubber content and S. J. Park [52] ’s investigation reporting that the organic functional groups on the silica
surface make an increase of the adhesion at interfaces between
silica and rubber matrix, resulting in improved crosslink density.
3.4. Dynamic mechanical properties
3.4.1. Temperature sweep
The curves of both G0 and tand as a function of temperature with
a constant frequency for all the vulcanizates are shown in Fig. 7. The
approximate G0 values in glassy state for three vulcanizates are
observed, but in high-elastic state there are two distinct differences. First, the highest value of 1# vulcanizate is attributed to the
low mobility of matrix inside its silica aggregates [53] induced by

the strong fillerefiller interaction [54]. Next, G0 value of 2# is close
to that of 3#, and both are lower than that of 1#, indicating that the
better filler dispersion of 2# and 3# vulcanizates can decrease the
effects of filler aggregates on G0 value. The reason, according to
some researchers’ viewpoints [53,55], is likely that the presence of
CPTMO favors the filler dispersion or reduces the strengthening
effect of rigid inorganic particles thus, decreasing the storage shear
modulus. The similar results were also reported in the literature
[56].
The glass-transition characteristic data obtained from Fig. 7 are
shown in Table 7. Generally, the temperature associated with the
peak magnitude of the tand plot is defined as Tg [57]. It can be seen

Table 5
Bound rubber contents of rubber compounds.

Fig. 6. DSC curves of A-SSBR-1 pure rubber and SiO2/A-SSBR-1 co-coagulated rubber.

Sample no.

1# rubber
compound

2# rubber
compound

3# rubber
compound

Bound rubber

content (%)

14.69

59.52

48.95

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

X. Liu et al. / Polymer xxx (2014) 1e13

9

Table 6
Relaxation parameters of rubbers.
Sample no

1#
1#
2#
2#
3#
3#
a
b
c
d
e

f
g

Parameter

rubber compound
vulcanizate
rubber compound
vulcanizate
rubber compound
vulcanizate

Physical XLD Â 10À5a
(mol/cm3)

Chemical XLD Â 10À5b
(mol/cm3)

Mcc
(kg/mol)

A(Mc)d
(%)

T2e
(ms)

A(T2)f
(%)

qM2 Â 104g
(s2)

7.64
4.57
6.54
3.18
7.43
4.03

e
4.11
e
5.68
e
4.84

12.43
10.95
14.53
10.72
12.78
10.97

35.49
57.78
46.51
60.08
44.39
58.24

2.12
1.60
2.00
1.44
2.02
1.52

64.50
41.70
53.30
39.55
55.30
41.78

62.43
80.49
45.70
84.06
59.12
80.25

Physical crosslink density.
Chemical crosslink density.
Molecular weight between crosslinking points.
Percentage of crosslinking fractions.
Relaxation time.
Percentage of high-mobile fractions.
Residual dipolar interaction.

that Tg values of 1#, 2#, and 3# vulcanizates are 4.1  C, 5.9  C, and
5.5  C higher than those of corresponding pure rubbers respectively, and there is a similar feature between this result and DSC
data. It may be in relation with the constrained dynamics of
segmental motions of the rubber molecules interacting with filler
surface, as supported by NMR experiments [58]. This result presents the larger increments for 2# and 3# vulcanizates, which

shows that the end-group functionalization helps to achieve
stronger constraints to macromolecular motions and stronger fillerpolymer interactions.
In tire industry, tand values at 0  C and 60  C usually are used as
the indexes to evaluate wet skid resistance and rolling resistance
respectively [59,60]. Also from Table 7, 2# vulcanizate exhibits the
highest 0  C tand value and lowest 60  C tand value, implying a
balance between high wet skid resistance and low rolling resistance can be achieved through end-group functionalization. This
result indicates that the firm eSieOeSie bonds formed by
condensation reaction between silica and functionalized end-group
can enhance the filler-polymer interaction. Besides, the constrained
terminals also can greatly reduce the irregular thermal motion
among macromolecular chains, thus resulting in the low tand value
at 60  C. 3# vulcanizate is not better than 2# vulcanizate in above
aspects, for its relatively lower degree of end-group
functionalization.
3.4.2. Strain sweep
G0 -ε% and tand-ε% curves of all the vulcanizates are shown in
Fig. 8. Payne reported [61] that G0 decreased with the increase of
strain. This result was explained by the breakdown of aggregated
secondary network of filler particles or agglomerates formed by van
der Waals-London attraction forces. This consideration was supported by Kraus [62]. Therefore, Payne effect [61,63,64] usually is
used as an evaluation of the three-dimensional filler network for
fillerefiller and fillerepolymer interaction. The lower Payne effect
implies the stronger filler-polymer interaction and better filler

dispersion [65]. From G0 -ε% curve, the DG’ values (G0 difference
between ε ¼ 0.28% and ε ¼ 100%) and Payne effect of both 2# and
3# vulcanizates are lower than those of 1# vulcanizate, i.e., fewer
agglomerates and stronger filler-polymer interaction. This reveals
that functionalized groups at the end of macromolecular chains can

Table 7
Glass-transition parameters of rubbers.
Sample no.
Glasstransition
SSBR
A-SSBR-1
A-SSBR-2
parameter
1# pure 1#
2# pure 2#
3# pure 3#
rubber vulcanizate rubber vulcanizate rubber vulcanizate

Fig. 7. G0 -T and tand-T curves of vulcanizates.

Tg ( C)
À26.5
Tand value e
at 0  C
Tand value e
at 60  C

À22.4
0.269

0.076

À25.8
e
e

À19.9
0.417
0.069

À25.1
e
e

À19.6
0.400
0.071

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

10

X. Liu et al. / Polymer xxx (2014) 1e13

Fig. 8. G0 -ε% and tand-ε% curves of vulcanizates (60  C).

improve the silica-rubber linkage and silica dispersion in rubber
matrix.
From tand-ε% curve, tand value of 1# vulcanizate is the highest

and increases rapidly with the increase of strain. This result shows
that its filler aggregates are gradually split with the increase of
strain, and thus the released occluded rubber can enhance the internal friction loss [56]. The tand values of 2# and 3# vulcanizates
are significantly lower than that of 1# vulcanizate, and rise slightly
with the increase of strain. This indicates that there are few large
aggregates in 2# and 3# vulcanizates, and the generated friction
heats are derived from the split of small aggregates. This further
indicates that the functionalized groups at the end of macromolecular chains immobilize the mobility of the free chain terminals
and reduce the friction loss of the chain terminals. These G0 -ε% and
tand-ε% results are basically consistent with the other published
research finding [56].
The above statements show that the filler-polymer interaction
and interfacial bonding are strengthened as well as the internal
friction loss and heat are reduced after the condensation between
hydroxyl groups of silica and functionalized end-groups of A-SSBR.

occurs at the filler-rubber interface during the stretching [70,71].
The modulus at 300% elongation, tensile strength, tear strength,
and lowest permanent set of 2# vulcanizate are the highest in
Table 8. Polmanteer [72] and Pal [73] discovered that some properties of sulfur-cured rubbers, such as tensile strength and tear
strength, were improved as the quality of silica dispersion
increased. This result therefore indicates that the filler dispersion
[74], filler-polymer interaction, and external force resistance are
improved and the propagation paths of tear crack are lengthened
after the reactive blending of A-SSBR-1/silica by two steps.
In Table 8, the order from low to high for the three vulcanizates’
compression heat build-up values is 2#, 3#, and 1#, exhibiting the
same feature as their tand values at 60  C. This result indicates the
interfacial bonding between rubber and silica for 2# vulcanizate
can be enhanced by its strong filler-polymer interaction [75]. Due to

the formation of effective interface which avoids severe friction
[19], it leads to the reduction of friction heat loss in the process of
dynamic compression.

3.5. Mechanical properties

TEM photographs of all the vulcanizates are shown in Fig. 9. In
these photographs, the dark color part is silica and the light color
part is SSBR or A-SSBR matrix. In Fig. 9(a), there are the evident
filler-aggregation and poor dispersion, which can weaken the
rubber by creating structural flaws and damage to properties [76],
whereas in Fig. 9(b) and (c), the only tiny aggregates which hinder
the formation of local stress concentrations where fracture is easy
to start [53], and the isolated single silica particles which are uniformly distributed in rubber matrix in a scale of less than 50 nm
with good filler dispersion are observed. This result can serve as the
direct proof of good filler dispersion used for analyzing the improvements in such as bound rubber content, dynamic, rheological,
and mechanical properties, and further indicates that silica
dispersion in nano-scale can be effectively achieved by condensation reaction between alkoxysilane groups on the ends of macromolecular chains and hydroxyl groups of silica particles.

Mechanical properties of all the vulcanizates are shown in
Table 8. From Table 8, the Shore A hardness value of 1# vulcanizate
is the highest, which is related to the poor silica dispersion in
rubber matrix and is in accordance with its highest modulus value
in G0 -T curve. It is well known that the tensile properties are
affected by the size of agglomerates formed by the silica [66,67] and
rubber/silica interaction [68,69]. For the samples with weak or
without filler-rubber chemical bonding, the dewetting firstly
Table 8
Mechanical properties of vulcanizates.
Mechanical parameter

Sample no.
1# vulcanizate

Shore A hardness
Modulus at 100%
elongation (MPa)
Modulus at 300%
elongation (MPa)
Tensile strength (MPa)
Elongation at break (%)
Permanent set (%)
Tear strength (KN$mÀ1)
Compression heat
build-up ( C)

2# vulcanizate

3# vulcanizate

68
1.7

60
1.6

60
1.7

4.2

7.3

6.7

14.3
593
18
31.9
14.9

17.6
440
10
38.0
8.5

16.0
459
10
37.4
10.2

3.6. Microstructure

3.7. Relationship between novel chemical structure and properties
According to the theory of network elasticity, the hysteresis loss
of rubber material is mainly from its free terminals which have no
contribution to elasticity and increase the internal resistance.
Nagata [77] considered the modification of the molecular chain

ends was the principal means endowing SSBR with energy-saving
properties. This implies both hysteresis loss and heat build-up are
decreased via reducing the amount of free chain terminals. In this

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

X. Liu et al. / Polymer xxx (2014) 1e13

11

Fig. 9. TEM photographs of vulcanizates (a) 1# (b) 2# (c) 3#.

study, controlled/“living” anionic polymerization, which makes the
methods of molecular structure modification diversified, has been
adopted to perform end modification of SSBR. Besides, one special
silane coupling agent with one end which can terminate the active
center of anionic polymerization and the other end of siloxane
groups which can react with silica particles, was selected as endfunctionalized agent. As a result, it can not only reduce significantly the free ends of macromolecular chains, but also chemically
anchor these rubber chains on the surfaces of silica particles to
further strengthen the silica-rubber interaction. The covering layer
of end-functionalized macromolecular chains around the silica
particles is beneficial to reducing the silica agglomeration and
improving the silica dispersion.
Some characterizations and measurements were carried out to
investigate the relationship between this novel structure and
properties. There are three aspects about this relationship by
summarizing all the experimental results. The first is the degree of
structural modification. This is represented by end-functionalized
efficiency, which is an important parameter and calculated from

1
H NMR spectra. The higher the end-functionalized efficiency is,
the greater the number of chain restrainedly bonded on silica
particle is. The end-functionalized efficiency is directly proportional to the excellence of most properties, including EDS, viscosity,
bound rubber content, chemical crosslink density, G0 value in highelastic state or at lower strain, internal friction heat, tensile and tear
strengths, wet skid resistance, and rolling resistance. This further
implies the final properties can be improved by increasing the
degree of this structural modification. The second is filler-polymer
interaction. The advantages of silica’s high reinforcement and low
heat build-up can be largely taken only when the interfacial
adhesion between added silica and rubber chain (i.e., filler-polymer
interaction) is strengthened. This aim can be achieved by condensation reaction between end-functionalized macromolecular
chains and silica particles. The silica particles and rubber chains are

chemically linked by the resultant eSieOeSie(CH2)3e bonds
instead of conventional physical adsorption. This action can
significantly strengthen the filler-polymer interaction so that the
firm eSieOeSie bond can well resist the external force or dynamic
strain. As a macroscopic result of filler-rubber interactions, the
bound rubber content is consequently considered, and its importance is stressed as the obvious link towards excellent properties.
The bound rubber content of A-SSBR is significantly higher than
that of SSBR, testifying the improvement of filler-rubber interaction
by this end-group functionalization method. The third is filler
dispersion. This has relevance to the filler-rubber interaction. The
increased filler-rubber interaction results in the decreased fillere
filler interaction, and thus silica particles’ well-known agglomeration is weaken to improve the silica dispersion. The situation is
totally different with silica whose surface is occupied by sizable
quantities of siloxane and silanol groups, giving rise to strong interparticles interactions through hydrogen bonding. Fillerefiller
interaction leads to the formation of silica networks and the associated difficulties in dispersing such a material in non-polar elastomers [78]. Generally, silica particles are organically modified by
using silane coupling agent to loosen their networks. Nevertheless

this cannot immobilize the free movement of the molecular chain
ends. On the contrary, the more positive experimental results,
which are consistent with our expected results from original molecular design, can be obtained by our end-group functionalization
method. The most intuitionistic results are shown by TEM photographs, illustrating the improved silica dispersion in nanoscale.
Furthermore, good silica dispersion effectively reduces the stress
concentration and molecular friction heat, manifesting other
excellent properties such as low Payne effect, low tand value under
temperature sweep or strain sweep, high wet skid resistance, low
rolling resistance, and good mechanical properties.
In a word, the filler dispersion in rubber matrix and interfacial
bonding between silica particles and macromolecules are improved

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

12

X. Liu et al. / Polymer xxx (2014) 1e13

significantly after the end-group functionalization. An effective
technique (i.e., end-group functionalization and then condensation
reaction with silica) is provided in this study. It widens the way of
thinking to achieve the reduction of free macromolecular chain
ends and the chemical bonds of silica particles on rubber chains,
since silica only naturally interact with certain specialty elastomers
such as polydimethylsiloxane, thank to their similar chemistry.
Therefore, A-SSBR/SiO2 composite exhibiting excellent rheological,
mechanical and dynamic properties, high wet skid resistance, and
low rolling resistance will be as an ideal material for preparing the
green tire tread.

4. Conclusion
SSBRs with alkoxysilane-functionalization at two ends of
macromolecular chains (A-SSBRs) were successfully synthesized,
and were characterized by 1H NMR, FTIR, and EDS, respectively. The
end-functionalized efficiency reaches 71e76%. The filler-polymer
interaction and silica dispersion can be significantly improved by
using A-SSBR as matrix. As a result, SiO2/A-SSBR composites exhibit
low viscosities, good flowabilities, low energy consumptions in
process, high bound rubber contents, good mechanical properties,
short relaxation times, low internal friction loss and heat build-up,
as well as high wet skid resistance and low rolling resistance.
Particularly, the comprehensive properties of 2# vulcanizate prepared by reactive blending in two steps are the most excellent. Its
novel process of end-group functionalization and then condensation reaction with silica can not only reduce significantly the free
ends of macromolecular chains to restrain their irregular thermal
motions, but also promote these rubber chains chemically anchor
on the surfaces of silica particles. In brief, A-SSBR is considered to
be appropriate as green tire tread material.
Acknowledgments
The authors gratefully acknowledge the financial supports of the
National Tenth-five Year Plan (Grant number: 2004BA310A41),
National Natural Science Foundation of China (Grant number:
50573005), National Natural Science Foundation of China (Grant
number: 51208012) and Research Fund of New Teachers for the
Doctoral Program of Higher Education of China (Grant number:
3c009011201301).
References
[1] Castellano M, Conzatti L, Costa G, Falqui L, Turturro A, Valenti B, et al. Polymer
2005;46(3):695e703.
[2] Tsutsumi F, Sakakibara M, Oshima N. Rubber Chem Technol 1990;63(1):8e22.
[3] Manley K, Barry M. Benefits of solution polymerized SBR. In: Alan B, editor.

Tire technology international. Dorking: UK & International Press; 1998.
pp. 78e82.
[4] Sierra CA, Fatou JMG, Quiteria VRS, Galan C. Rubber Chem Technol 1995;68(2):
259e66.
[5] Wolff S. Rubber Chem Technol 1996;69(3):325e46.
[6] Wolff S, Görl U, Wang MJ, Wolff W. Eur Rubber J 1994;16(1):16e9.
[7] Tse MF. J Appl Polym Sci 2006;100(6):4943e56.
[8] Venter SAS, Kunita MH, Matos R, Nery RC, Radovanovic E, Muniz EC, et al.
J Appl Polym Sci 2005;96(6):2273e9.
[9] Sae-oui P, Sirisinha C, Thepsuwan U, Hatthapanit K. Polym Test 2004;23(8):
871e9.
[10] Wang MJ. Rubber Chem Technol 1998;71(3):520e89.
[11] Sheng E, Sutherland I, Bradley RH, Freakley PK. Eur Polym J 1996;32(1):
35e41.
[12] Manna AK, De PP, Tripathy DK, De SK, Peiffer DG. J Appl Polym Sci 1999;74(2):
389e98.
[13] Ray S, Bhowmick AK. Polym Eng Sci 2004;44(1):163e78.
[14] Leblanc JL, Cartault M. J Appl Polym Sci 2001;80(11):2093e104.
[15] Mélé P, Marceau S, Brown D, De Puydt Y, Albérola ND. Polymer 2002;43(20):
5577e86.
[16] Yan HX, Sun K, Zhang Y, Zhang YX, Fan YZ. J Appl Polym Sci 2004;94(4):
1511e8.

[17] Hashim AS, Azahari B, Ikeda Y, Kohjiya S. Rubber Chem Technol 1998;71(2):
289e99.
[18] Choi SS. J Appl Polym Sci 2006;99(3):691e6.
[19] Ramesan MT, Mathew G, Kuriakose B, Alex R. Eur Polym J 2001;37(4):719e28.
[20] Malaika SA, Kong W. Polym Degrad Stab 2005;90(2):197e210.
[21] Lindsell WE, Radha K, Soutar I, Stewart MJ. Polymer 1990;31(7):1374e8.
[22] Kobayashi M, Uchino K, Ishizone T. J Polym Sci Part A: Polym Chem

2005;43(18):4126e35.
[23] Jalbert C, Koberstein JT, Yilgor I, Gallagher P, Krukonis V. Macromolecules
1993;26(12):3069e74.
[24] Tanaka K, Kawaguchi D, Yokoe Y, Kajiyama T, Takahara A, Tasaki S. Polymer
2003;44(15):4171e7.
[25] Kawanaka T, Watanabe H, Noguchi K, Akita S. Process for preparation of diene
type polymer rubber. US Patent 5,001,196; 1991.
[26] Ueda A, Akita S, Namizuka T. Rubber composition for tire treads. US Patent
4,555,547; 1985.
[27] Kiiyon R, Bamujae L. Production of solution-polymerized styrene-butadiene
rubber or butadiene rubber. Jpn Kokai Tokkyo Koho JP 11,284,998; 1999.
[28] Halasa AF, Matrana BA, Robertson WSE. Capped polydienes. US Patent
4,935,471; 1990.
[29] Lawson DF, Stayer JML, Antkowiak TA. Diene polymers and copolymers
terminated by reaction with aromatic nitriles. US Patent 5,153,271;
1992.
[30] Lawson DF, Stayer JML, Antkowiak TA. Diene polymers having reduced hysteresis and improved raw viscosity incorporating primary partial coupling and
fused-ring polynuclear aromatic terminators. US Patent 5,210,144; 1993.
[31] Antkowiak TA, Lawson DF, Koch RW, Stayer JML. Elastomers and products
having reduced hysteresis. US Patent 5,153,159; 1992.
[32] Kitagawa Y, Yamada H, Saito A. Rubber for tire treads and compositions
thereof. US Patent 4,835,209; 1989.
[33] Feng HD, Zhang XY, Zhao SH. J Appl Polym Sci 2008;110(1):228e36.
[34] Wang S, Luo XL, Ma DZ, Huang YP. Chin J Chem Phys 2004;17(5):652e6.
[35] Dutta NK, Choudhury NR, Haidar B, Vidal A, Donnet JB, Delmotte L, et al.
Rubber Chem Technol 2001;74(2):260e80.
[36] Huang YP, Chen GM, Wang S, Li HW, Ma DZ. Chin J Appl Chem 2005;22(4):
431e4.
[37] Ke YK, Dong HR. Analytical chemistry manual III: spectral analysis. 2nd ed.
Beijing: Chemical Industry Press; 1998. pp. 975e7.

[38] Choi SS, Kim IS. Eur Polym J 2002;38(6):1265e9.
[39] Mochel VD. Rubber Chem Technol 1967;40(4):1200e11.
[40] Bauer F, Gläsel HJ, Decker U, Ernst H, Freyer A, Hartmann E, et al. Prog Org
Coat 2003;47(2):147e53.
[41] Tan EH, Wolff S, Haddeman M, Grewatta HP, Wang MJ. Rubber Chem Technol
1993;66(4):594e604.
[42] Sawanobori J, Ono S, Ito M. Jpn J Polym Sci Technol 2000;57(6):356e62.
[43] Medalia AI. J Colloid Interf Sci 1970;32(1):115e31.
[44] Medalia AI. Rubber Chem Technol 1978;51(3):437e523.
[45] Shea KJ, Loy DA. Chem Mater 2001;13(10):3306e19.
[46] Zhang Z, Yang JL, Friedrich K. Polymer 2004;45(10):3481e5.
[47] Fröhlich J, Niedermeier W, Luginsland HD. Compos Part A Appl Sci
2005;36(4):449e60.
[48] Mohsen M, Abdel-El Salam MH, Ashry A, Ismail A, Ismail H. Polym Degrad Stab
2005;87(3):381e8.
[49] Beck K, Voda A, Bescher M, Peterseim V, Viol M. Kautsch Gummi Kunstst
2007;60(3):100e5.
[50] Zhao F, Ping Z, Zhao S, Jian Y, Kuhn W. Kautsch Gummi Kunstst 2007;60(12):
685e8.
[51] Kuhn W, Barth P, Denner P, Müller R. Solid State Nucl Magn Reson 1996;6(4):
295e308.
[52] Park SJ, Cho KS. J Colloid Interf Sci 2003;267(1):86e91.
[53] Aso O, Eguiazábal JI, Nazábal J. Compos Sci Technol 2007;67(13):2854e63.
[54] Evans LR, Fultz WC, Huber JM. Rubber World 1998;219(2):38e44.
[55] Albérola ND, Mélé P. Polym Compos 1996;17(5):751e9.
[56] Wu YP, Zhao QS, Zhao SH, Zhang LQ. J Appl Polym Sci 2008;108(1):112e8.
[57] Yang ZL, Wicks DA, Hoyle CE, Pu HT, Yuan JJ, Wan DC, et al. Polymer
2009;50(7):1717e22.
[58] Haidar B, Salah DH, Vidal A, Papirer E. Macromol Symp 1996;108(1):147e61.
[59] Zeng ZQ, Yu HP, Wang QF, Lu G. J Appl Polym Sci 2008;109(3):1944e9.

[60] Takino H, Nakayama R, Yamada Y, Kohjiya S, Matsuo T. Rubber Chem Technol
1997;70(4):584e94.
[61] Payne AR, Whittaker RE. Rubber Chem Technol 1971;44(2):440e78.
[62] Kraus GJ. Appl Polym Sci Appl Polym Symp 1984;39:75e92.
[63] Payne AR. J Appl Polym Sci 1962;6(19):57e63.
[64] Harwood JAC, Mullins L, Payne AR. J Appl Polym Sci 1965;9(9):3011e21.
[65] Reuvekamp LAEM, Ten BJW, Van SPJ, Noordermeer JWM. Rubber Chem
Technol 2002;75(2):187e98.
[66] Suzuki N, Yatsuyanagi F, Ito M, Kaidou H. J Appl Polym Sci 2002;86(7):
1622e9.
[67] Yatsuyanagi F, Suzuki N, Ito M, Kaidou H. Polymer 2001;42(23):9523e9.
[68] Wagner MP. Rubber Chem Technol 1976;49(3):703e74.
[69] Yatsuyanagi F, Suzuki N, Ito M, Kaidou H. Polym J 2002;34(5):332e9.
[70] Gent AN, Park B. J Mater Sci 1984;19(6):1947e56.
[71] Wolff S, Wang MJ, Tan EH. Kautsch Gummi Kunstst 1994;47(2):102e7.
[72] Polmanteer KE, Lentz CW. Rubber Chem Technol 1975;48(5):795e809.
[73] Pal PK, De SK. Rubber Chem Technol 1982;55(5):1370e88.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />

X. Liu et al. / Polymer xxx (2014) 1e13
[74] Peng Z, Kong LX, Li SD, Chen Y, Huang MF. Compos Sci Technol 2007;67(16):
3130e9.
[75] Rattanasom N, Saowapark T, Deeprasertkul C. Polym Test 2007;26(3):369e77.
[76] Stumpe NA, Railsback HE. Rubber World 1964;151(3):41e3.

13

[77] Nagata N, Kobatake T, Watanabe H, Ueda A, Yoshioka A. Rubber Chem Technol

1987;60(5):837e55.
[78] Leblanc JL. Prog Polym Sci 2002;27(4):627e87.

Please cite this article in press as: Liu X, et al., Preparation, structure, and properties of solution-polymerized styrene-butadiene rubber with
functionalized end-groups and its silica-filled composites, Polymer (2014), />