Composites Science and Technology 88 (2013) 69–75

Contents lists available at ScienceDirect

Composites Science and Technology

journal homepage: www.elsevier.com/locate/compscitech

Surface modification of silica by two-step method and properties of

solution styrene butadiene rubber (SSBR) nanocomposites filled with
modified silica
Yan Li b, Bingyong Han c, Li Liu a,b,⇑, Fazhong Zhang b, Liqun Zhang a,b, Shipeng Wen a,b, Yonglai Lu b,
Haibo Yang b, Jing Shen b
a
Key Laboratory of Beijing City for Preparation and Processing of Novel Polymer Materials, Beijing University of Chemical Technology, Beijing 100029, PR China
b
State Key Laboratory of Organic–Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, PR China
c
State Key Laboratory Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, PR China

a r t i c l e i n f o a b s t r a c t

Article history: The traditional method for directly compounding rubber, silica, and bis(3-triethoxysilylpropyl)tetrasul-
Received 24 June 2013 fide (TESPT) by shear force, which we called it One-Step Method (OSM), results in a modification of uncer-
Received in revised form 17 August 2013 tain mechanism: it is not clear whether chemisorption or physisorption is involved. In addition, OSM
Accepted 19 August 2013
leads to large processing energy consumption (PEC). In view of these issues, we used a novel method
Available online 5 September 2013
(Two-Step Method, TSM) to investigate the modification process in detail. The TSM modification indi-
cated that the TESPT hydrolyzed firstly to generate the silanol (Si–OH), and the silanol reacted with
Keywords:
the hydroxyl groups on the surface of silica, which characterized by FTIR. The properties of modified silica
A. Nanocomposites
B. Surface treatment
were studied. Furthermore, the SSBR nanocomposites filled with modified silica by TSM and OSM were
A. Nanoparticles prepared and the properties comparisons were carried out. The obtained results exhibited the advantages
A. Coupling agents of TSM, and also revealed that 8% TESPT amount was suitable than 12% and 15% TESPT amount.
D. Dynamic mechanical thermal analysis Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction process. We call this method the one-step method (OSM). Sae-oui
et al. [15] used this method to prepare a silica/TESPT/polychloro-
Silica [1–3] is widely used to improve the static and dynamic prene nanocomposite in a Haake Rheomix mixer. They suggested
mechanical properties [4,5] of rubber, such as wet skid resistance that TESPT improved the processability of the compound, mechan-
and rolling resistance. Furthermore, silica is also independent of ical properties of the vulcanizates, and dispersion of silica in the
oil resources and environmentally friendly filler used as a replace- vulcanizates. Dohi and Horiuchi [16] used a similar way to prepare
ment for carbon black [6,7]. Because of the poor compatibility be- a silica/TESPT/styrene butadiene rubber nanocomposite in a Ban-
tween silica and the rubber matrix and the severe agglomeration bury mixer. They found that a TESPT coupling layer on the surface
resulting from the surface hydroxyl groups of silica, the applica- of silica and affected by different temperature, which could im-
tions of silica are limited [8,9]. So the surface modification [10– prove the dispersion of silica in the nanocomposite. Even though
12] must be introduced in application of silica and silane coupling such in situ reactions during processing result in a uniform disper-
agents [13,14], such as bis(3-triethoxysilylpropyl)tetrasulfide (TES- sion of silica in the rubber, the process energy consumption (PEC)
PT), are most commonly used to modify the silica. here is always very high. Another disadvantage of this method is
There are two methods for the surface modification of the silica. that the large silica aggregates cannot be broken to a large extent
In the first method, the rubber matrix, silica, and TESPT are com- by mechanical shear force to have a good dispersion in the rubber
pounded in a two-roll mill or an internal mixer. TESPT reacts matrix. In the second method, the silica is pre-modified by TESPT
in situ with the surface hydroxyl of the silica by shear force in the before mixing with rubber to prepare the nanocomposite. For
compounding process, and then the sulfur of TESPT reacts with example, Castellano et al. [17] studied the pre-modification of silica
the unsaturated bonds of the rubber matrix during the curing in a CH2Cl2 solution of TESPT to obtain modified silica, which has a
lower surface energy than that of pure silica because of the exis-
tence of TESPT. They further concluded that the pre-modified silica
⇑ Corresponding author at: Key Laboratory of Beijing City for Preparation and improved the mechanical properties and filler dispersion of the vul-
Processing of Novel Polymer Materials, Beijing University of Chemical Technology, canizates. However, it is not clear whether the interaction between
Beijing 100029, PR China. Tel.: +86 10 64443413; fax: +86 10 64433964.
silica and TESPT is simple physical absorption or the chemical
E-mail address: [email protected] (L. Liu).

0266-3538/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.compscitech.2013.08.029
70 Y. Li et al. / Composites Science and Technology 88 (2013) 69–75

bonding during the condensation reaction of the hydrolyzed hydro- Table 1

xyl of TESPT and surface hydroxyl of silica. As is well known the The formulation of OSM and TSM modified silica filled SSBR
compounds.
deep chemical bonding of silica and silane favors the dispersion
of silica and interfacial interaction between the rubber and silica Ingredient Loading/phra
during mixing. However, it is not clear how to control the hydroly- SSBR 137.5
sis of TESPT in this method. Modified Silica 75
In this paper, we put forward a two-step method (TSM) for the Zinc oxide 3.5
Stearic acid 2.0
modification of the silica. We first controlled the conditions such as Antioxidant 4010NAb 1.3
pH value for the hydrolysis of TESPT to obtain a TESPT hydrolysate Paraffin wax 1.5
solution and then mixed it with silica for a certain time. Subse- CBSc 1.5
quently, the mixture of the hydrolysate and silica was spray-dried DPGd 2.0
TMTDe 0.4
to obtain modified silica. The hydrolysis of TESPT and formation of
Sulfur 1.4
the chemical bond between silica and TESPT were confirmed by
a
Fourier transform infrared spectroscopy (FT-IR). The size distribu- Parts-per-hundred rubber.
b
N-Isopropyl-N0 -phenyl-1,4-phenylenediamine.
tion and agglomeration of silica nanoparticles before and after c
N-Cyclohexyl-2-beozothiazole sulfonamide.
modification were characterized by dynamic light scattering d
1,3-Diphenyl guanidine.
(DLS). Finally, the effect of this novel silica on the modification of e
Tetramethylthiuramdisulfide.
rubber was investigated by dynamic mechanical thermal analysis
(DMTA) and transmission electron microscopy (TEM). Meanwhile,
the processing energy consumptions of OSM and TSM were inves- the silica modified by TSM was added directly, but the silica mod-
tigated in detail. ified by OSM was alternately added with TESPT. Reaction blending
with silica was carried out for 15 min by the same equipment and
2. Experimental under same conditions. In the second stage, the compounds were
taken out, cut into pieces, and put back into the same equipment
2.1. Materials to be heat treated mixing for 5 min at a rotational speed of
100 rpm and temperature of 120 °C. Finally, the remainder of other
Solution styrene butadiene rubber (Buna VSL 5025-2HM) com- ingredients were added together and mixed with the compounds
posed of 75% butadiene (50% vinyl content and 25% styrene) and prepared before uniformly on a 6-inch two-roll mill (Shanghai
extended with 37.5 phr of treated distillate aromatic extracts Rubber Machinery Works No. 1, Shanghai, China). For the two
(TDAE) oil was purchased from Lanxess Chemical Industry Co., kinds of silica compounds, the mixing procedure was omitted.
Ltd. (Germany). Precipitated silica of Tixosil 383 (median diameter The optimum cure time (t90) of the compounds was determined
is 13.7 nm, DOP oil absorption is 2.66 mL/g, CTAB specific surface by a P3555B2 Disc Vulkameter (Beijing Huanfeng Chemical
area is 163 m2/g) was produced by Rhodia France (Qingdao, China). Machinery Trial Plant, Beijing, China), and then the nanocompos-
Absolute ethanol (analysis grade, AR) and acetic acid (mass fraction ites were prepared by a XLB-D350  350-type Automatic Opera-
of 37.5%) were purchased from Beijing Chemical Works (China). tion Vulcanizing Press (Huzhou Dongfang Machinery Co., Ltd.,
Bis(3-triethoxysilylpropyl)tetrasulfide (TESPT) was obtained from Zhejiang, China). The compounds were vulcanized at 160 °C for
Nanjing Shuguang Chemical Group Co., Ltd. (China). The other t90 in a standard mold to form the nanocomposites.
materials are all commercially available.

2.4. Characterizations
2.2. Preparation of TESPT hydrolysates and modified silica
The hydrolysis of TESPT and surface modification of modified
Absolute ethanol 2000 mL and deionized water 60 mL were silica was confirmed by fourier transform infrared spectroscopy
added to a beaker, and acetic acid was used to adjust the pH value (FTIR; Tensor 27, Bruker Optik GmbH Co., Ltd., Germany). The
to approximately 4.5. TESPT was added to the solution and hydro- hydrolysate of TESPT was dropped on a block of potassium bro-
lysis of TESPT was allowed to proceed for 10 h, resulting in the mide (KBr) and dried absolutely under the IR LED to remove the
transformation of ethyoxyl to hydroxyl, so called ‘‘TESPT hydroly- distraction of the ethanol in the hydrolysate for measurement.
sates’’. Afterwards, silica was added to the hydrolysate and stirred The pure and the modified silica powder with KBr were made into
for 16 h at a rate of 300 rpm. The mass ratio of TESPT and silica is ultra-thin disk specimens for characterization. The organic degree
8%, 12%, and 15%, respectively. The slurry was spray-dried to obtain [18,19] of modified silica was calculated from the grafting degree
the modified silica. A part of the modified silica was extracted with (K) according to the following equation:
toluene in a Soxhlet extractor for 48 h and then dried completely in
a vacuum oven. The dried silica was used to prepare samples for K ¼ ðR  rÞ=R  100%; ð1Þ
characterization. The rest of the modified silica was compounded
with rubber to prepare nanocomposites. where K is the grafting degree of the hydroxyl, R is the ratio of the
hydroxyl peak at 3447.35 cm1 to that at 1108.32 cm1 of pure sil-
2.3. Preparation of rubber nanocomposites ica in the FTIR spectra, and r is the ratio of the hydroxyl peak at
3447.35 cm1 to that at 1108.32 cm1 of modified silica in the FTIR
The formulation of OSM and TSM modified silica filled SSBR spectra.
compounds is shown in Table 1. The rubber compounds were ob- The nanoparticle size distribution of silica was analyzed by
tained in two mixing stages by using a Haake PolyLab torque rhe- dynamic light scattering (DLS; Malvern ZEN3600, Malvern Instru-
ometer (Thermo Electron Corporation, USA). In the first stage, ments Co., Ltd., England). The silica powder was dispersed in
SSBR, zinc oxide, stearic acid, 4010NA, and paraffin wax were ethanol for 2 h by ultrasonic technology.
mixed in a mixing chamber for 5 min (SSBR for 1 min, zinc oxide The specific surface area of Brunauer–Emmett–Teller (SSA-BET)
and stearic acid for 2 min, 4010NA and paraffin wax for 2 min) at of silica was determined by using an automated surface area and
a rotational speed of 20 rpm and temperature of 70 °C. Afterwards pore size analyzer (ASAP; Quadrasorb SI, Quantachrome
 Y. Li et al. / Composites Science and Technology 88 (2013) 69–75 71

Fig. 1. FTIR spectra of (A) TESPT and (B) hydrolysates with different amount TESPT.

Instruments Co., Ltd., USA) by nitrogen adsorption at 196 °C. The FTIR spectra of which are shown in Fig. 2A is used together with
silica powder was degassed at 70 °C for 18 h before adsorption. Eq. (1) for calculating the grafting degree (K) at different TESPT
The morphological structure such as filler dispersion and filler amount, and the results are shown in Table 2. Fig. 2A shows that
network structure in the vulcanizates was observed by transmis- there is a strong absorbance at 1117 cm1 attributed to the
sion electron microscopy (TEM; Tecnai G2 20 S-TWIN, FEI Co., Si–O–Si groups and a broad stretching vibration peak at
Ltd., USA) using an acceleration voltage of 200 kV. Thin sections 3400 cm1 attributed to the surface hydroxyl groups [24]. As
of the vulcanizates were cut by a microtome at 100 °C and col- shown in Table 2, the values of K at TESPT amount of 8 wt.%,
lected on copper grids. 12 wt.%, and 15 wt.% are 45.2%, 41.6% and 32.4%, respectively, indi-
The processing energy consumption (PEC) was transformed cating a decrease in grafting degree with increasing TESPT amount.
from the processing torque as measured by the Haake PolyLab tor- After high temperature spray drying, a large TESPT amount may
que rheometer. The transformation was carried out according to produce an oligomer by inter-reaction or hydrogen bonds by hy-
the following equation: droxyl of TESPT hydrolysates, emerging steric hindrance to make
the other TESPT difficultly graft on the surface of silica. After Soxh-
P ¼ Te  ð2p  n=60Þ=1000 ð2Þ let extraction, the oligomer is removed, resulting in much hydroxyl
where P is processing power, Te is the processing torque, and n is exposing outside the surface of silica and low grafting degree. We
the rotational speed of the roller rotors. The power versus time conclude that a TESPT amount of 8 wt.% is enough to modify silica
curves was integrated in all regions to obtain the PEC. by introducing organic groups. Fig. 2B gives the detailed view of
The dynamic mechanical properties [20–22] (temperature FTIR spectra of silica from 2700 cm1 to 3000 cm1 to show the dif-
sweep) of a vulcanizates were measured by a dynamic mechanical ferences before and after the modification of silica. In Fig. 2B, the
thermal analyzer (DMTA; VA3000, 01 dB-Metravib Co., Ltd., strong absorbance at 2850 cm1 and 2930 cm1 [25] attributed
France) in the rectangular tension mode. The temperature depen- to the –CH2 vibration indicates that TESPT is successfully grafted
dence of the loss factor (tan d) and storage modulus (G0 ) was mea- on the surface of silica, resulting in the formation of chemical bond
sured in the temperature range 60 °C to 100 °C at a heating rate between silica and TESPT.
of 3 °C/min, frequency of 10 Hz, and strain amplitude of 0.1%. The nanoparticle size distributions of the modified silica and the
pure silica were characterized by DLS, and the results are shown in
Fig. 2C. For the pure silica, the distribution is wide and the nano-
3. Results and discussion particle size is large, and obviously the silica nanoparticles tend
to aggregates to a large extent. The primary peak of aggregates is
3.1. Chemical structure of hydrolysate of TESPT at about 270 nm and the secondary peak is at about 700 nm. In
addition, there exists some large silica aggregates in the range
The FTIR spectra of TESPT and hydrolysates with different 4–7 lm. After modification of the silica by TSM method, the size
amount of TESPT are shown in Fig. 1A shows that the broad absorp- distribution of the silica narrows and the nanoparticle size de-
tion associated with the hydroxyl stretching vibration peaks at creases. For example, the particle size peak of silica modified with
approximately 3400 cm1 in the spectrum of TESPT hydrolysate, 8 wt.% of TESPT appears at 50 nm, showing a good modification ef-
but is not found in the spectrum of TESPT. In addition, the spec- fect. With the increase of the TESPT amount from 8 wt.% to 15 wt.%,
trum of TESPT hydrolysate shows the characteristic absorption the distribution curves shift right. It is considered that more TESPT
peaks at 1640 cm1 and 1740 cm1, corresponding to the stretch- grafted on the surface of silica forms hydrogen bond with silica,
ing and deforming vibration modes, respectively, of the H–O–H leading to an increase of thickness of the coupling layer bonded
bond of the absorbed water [23]. Fig. 1B shows that the density on the silica surface.
of hydrolysates increases obviously with increasing TESPT amount Because of the formation of the coupling layer on the silica sur-
based on the characteristic absorption peaks at 3400 cm1 and face, the specific surface area of Brunauer–Emmett–Teller [26]
1640 cm1. It illustrates that the end group ethyoxyl transforms (SSA-BET) of modified silica shows a decrease tendency, as shown
into hydroxyl after hydrolysis, which offers the necessary condi- in Fig. 2D. At 8 wt.% of TESPT, the SSA-BET value decreases sharply
tion for the grafting reaction between silica and TESPT. from 159 m2/g to 133 m2/g, but hardly changes with further in-
creases of TESPT amount. It is considered that the porous silica sur-
3.2. Chemical characterization of modified silica by TSM face have been mostly filled by TESPT molecules as the TESPT
amount reaches 8 wt.%, as a result of the condensation reaction be-
The slurry of TESPT hydrolysate and silica was dried by spray tween the hydroxyl of hydrolyzed TESPT and the surface hydroxyl
drying under high temperature to obtain the modified silica, the of silica (inset in Fig. 2D).
72 Y. Li et al. / Composites Science and Technology 88 (2013) 69–75

Fig. 2. Chemical characterization of modified silica by TSM: FTIR spectra (A and B), nanoparticle size distributions (C) and SSA-BET (D) of pure silica and modified silica.

force is dominant in dispersing silica nanoparticles. During this

Table 2 process, the TESPT has limited opportunity to contact with the sur-
Values of grafting degree (K) of modified silica with different amount of TESPT.
face of the silica aggregates because parts of them also dissolve in
Sample 8% 12% 15% the rubber matrix. Thus, a high loading of TESPT is needed to have
K (%) 45.2 41.6 32.4 any chance of interacting with the silica aggregates. Without a high
loading of TESPT, it is not clear what extent TESPT hydrolyzats oc-
curs and how much interfacial interaction originates from the
chemical bonding between the hydrolyzed TESPT and silica.
3.3. Morphology of silica/SSBR nanocomposites and mechanisms of
TSM and OSM
3.4. Processing energy consumption (PEC) during blending
We prepared silica/SSBR nanocomposites with silica modified
separately by OSM and TSM. The dispersion of the silica was char- The two stages of blending are shown in Section 2.3 (prepara-
acterized by TEM, and the micrographs are shown in Fig. 3. In the tion of rubber nanocomposites). The first stage is compounding
TEM micrographs of the nanocomposites prepared by OSM the SSBR and the silica (or modified silica) and the second stage
(Fig. 3A1–C1), the dark phase represents the dispersed silica nano- is thermal treated which is good for the silica dispersion after first
particles. The TEM micrographs of the nanocomposites obtained by stage of preparation of rubber compounds. We calculated the pro-
TSM (Fig. 3A2–C2) reveal that there are a large number of extre- cessing energy consumption (PEC) of samples with different TESPT
mely fine nanoparticles because some large local-areas which contents prepared by OSM and TSM, and the results are calculated
represented rubber matrix in OSM samples obviously decrease. by Fig. 5 and shown in Table 3. The PEC values of TSM are lower
Therefore, the dispersion of silica in the rubber matrix is more than those of OSM because the TSM modification grafting of TESPT
uniform in nanocomposites prepared by TSM method than in on the silica surface improves the compatibility between silica and
nanocomposites prepared by the OSM method. the rubber matrix. The formation of chemical bond between silica
A comparison of the above results obtained from TSM with that and TESPT makes the silica closely combine with TESPT; mean-
obtained from OSM suggests that the surface modification mecha- while, the TSM modification decreases the hydroxyl groups outside
nisms of silica are different at the same nanocomposite composi- of the surface of silica. But in OSM modification, the hydroxyl
tion. In our TSM modification as shown in Fig. 4, the TESPT is groups outside of the surface of pure silica make the weak compat-
first hydrolyzed under certain conditions, and then the hydrolyzed ibility with the rubber matrix, resulting from the stronger internal
TESPT is grafted on the surface of silica through spray drying, fraction of silica and high PEC. From Table 3, we can find that when
resulting in the formation of chemical bonds at the interface be- the TESPT amount is 8%, 12% and 15%, the energy saving ratio is
tween silica and hydrolyzed TESPT. This procedure not only fa- 25.7%. 21.1% and 22.1%, respectively.
vored a small-sized pre-distribution of silica nanoparticles but
also remarkably improved the compatibility between silica and 3.5. Dynamic mechanical thermal analysis of silica/SSBR vulcanizates
the rubber matrix, making it possible to homogeneously disperse
silica with small nanoparticle size into the rubber matrix. How- The temperature dependence of viscoelasticity of the loss factor
ever, in the traditional OSM modification as shown in Fig. 4, shear tan d and storage modulus G0 for the silica/SSBR vulcanizates is
 Y. Li et al. / Composites Science and Technology 88 (2013) 69–75 73

Fig. 3. TEM micrographs of silica/SSBR nanocomposites.

Fig. 4. Surface modification mechanism of silica with silane coupling agent TESPT.
74 Y. Li et al. / Composites Science and Technology 88 (2013) 69–75

Fig. 5. PEC curves of compounds filled with silica modified by OSM and TSM in Haake Rheomix mixer.

Table 3 higher than those of the OSM vulcanizates at 0 °C, an indication

PEC of compounds filled with silica modified by OSM and TSM in Haake Rheomix that the TSM vulcanizates have better wet skid resistance than
mixer.
the OSM vulcanizates. Meanwhile, the tan d values of the TSM vul-
TESPT amount (%) OSM (kJ) TSM (kJ) Energy saving ratio (%) canizates are similar to those of the OSM vulcanizates at 60 °C, but
8 138.067 102.584 25.7 continuously decrease with temperature above the 60 °C, suggest-
12 134.227 105.905 21.1 ing an improvement of rolling resistance at high temperatures. The
15 122.467 95.381 22.1 tan d results of three different TSM vulcanizates also can be seen in
Fig. 6A. We can see that the heights of the tan d peak are in the or-
der 8% TSM > 12% TSM > 15% TSM. The higher tan d peak represents
shown in Fig. 6. It is well-known that the weaker filler network the reduction of the secondary filler network structure, so the
structure leads to the higher tan d peak [27]. In the glass-transition lower TESPT amount leads to relatively weaker filler network
zone, the main source of energy dissipation is from the polymer structure, in other words 8% TSM sample have better filler disper-
matrix and the filler network cannot be easily broken. The hyster- sion than the other two. However, we also find that the tan d values
esis loss will be significantly attenuated by filler agglomeration be- of 12% TSM and 15% TSM are both lower than that of 8% TSM in the
cause of the reduction of the effective volume of the polymer temperature range 0–100 °C, indicating complicated influences on
matrix [28]. In Fig. 6A, the tan d peaks of the TSM vulcanizates the tan d values from the filler network, the coupling layer on silica
are higher than those of the OSM vulcanizates, indicating that surface, and the change of the crosslink network as increase of
the filler network structure of TSM vulcanizates is weaker than TESPT amount.
that of OSM vulcanizates. The two temperatures 0 °C and 60 °C The storage modulus (G0 ) of the vulcanizates in the rubbery
have special significance for tread tires [29,30]. A higher hysteresis state can be used to evaluate filler–filler interaction. The higher
at 0 °C and lower hysteresis at 60 °C will be favorable for higher value of G0 indicates the more severe agglomeration of filler
wet skid resistance and lower rolling resistance, respectively. From nanoparticles and the stronger filler network structure [31,32].
Fig. 6A, the tan d values of the TSM vulcanizates are comparatively As shown in Fig. 6B, the OSM vulcanizates have a much higher G0

Fig. 6. Temperature dependence of tan d (A) and G0 (B) of vulcanizates filled with modified silica.
 Y. Li et al. / Composites Science and Technology 88 (2013) 69–75 75

than the TSM vulcanizates above the glass-transition temperature [3] Peng Z, Kong LX, Li SD, Chen Y, Huang MF. Self-assembled natural rubber/silica
nanocomposites: its preparation and characterization. Compos Sci Technol
(Tg) of the vulcanizates, especially at higher temperatures (at
2007;67(15–16):3130–9.
which the polymer matrix is at the rubbery state). Because of the [4] Pan XD. Impact of reinforcing filler on the dynamic moduli of elastomer
lower G0 (at 0.1% strain amplitude), the TSM vulcanizates show a compounds under shear deformation in relation to wet sliding friction. Rheol
weaker filler network structure and fewer filler aggregates than Acta 2005;44(4):379–95.
[5] Gal AL, Yang X, Kluppel M. Evaluation of sliding friction and contact mechanics
the OSM vulcanizates, indicating better dispersion of TSM-modi- of elastomers based on dynamic-mechanical analysis. J Chem Phys
fied silica in the rubber matrix, in agreement with TEM observa- 2005;123(1):014704.
tions. Meanwhile, Fig. 6B shows that the G0 of TSM vulcanizates [6] Gerspacher M, O’Farrell CP, Nikiel L, Yang HH. High frequency viscoelasticity of
carbon black filled compounds. Rubber Chem Technol 1996;69(5):786–800.
changes with the TESPT amount, and the G0 values are in the order [7] Wang MJ. New developments in carbon black dispersion. Kautsch Gummi
8% TSM < 12% TSM < 15% TSM at the high temperature, also be- Kunstst 2005;58(12):626–37.
cause of the complicated effect of the changes in filler network [8] Byers JT. Silane coupling agents for enhanced silica performance. Rubber
World 1998;218(6):38–44.
structure as TESPT amount varies. We conclude that the dynamic [9] Li YB, Wang MJ, Zhang T, Zhang F, Fu X. Study on dispersion morphology of
mechanical properties of vulcanizates decrease with increasing silica in rubber. Rubber Chem Technol 1994;67(4):693–9.
TESPT amount. [10] An DM, Wang ZC, Zhao X, Liu Y, Guo YP, Ren SX. A new route to synthesis of
surface hydrophobic silica with long-chain alcohols in water phase. Colloids
Surf A: Physicochem Eng Aspects 2010;369(1–3):218–22.
[11] Lee YL, Du ZC, Lin WX, Yang YM. Monolayer behavior of silica particles at air/
4. Conclusion water interface. A comparison between chemical and physical modifications of
surface. J Colloid Interface Sci 2006;296(1):233–41.
[12] Ma XK, Lee NH, Oh HJ, Kim JW, Rhee CK, Park KSS, et al. Surface modification
In this work, a novel two-step method (TSM) modification was
and characterization of highly dispersed silica nanoparticles by a cationic
used to successfully prepare the modified silica. The chemical bond surfactant. Colloids Surf A: Physicochem Eng Aspects 2010;358(1–3):172–6.
formed after grafting reaction between silica and TESPT was gener- [13] Suzuki N, Ito M, Yatsuyanagi F. Effects of rubber/filler interactions on
ated, which was confirmed by FTIR experiments. DLS results re- deformation behavior of silica filled SBR systems. Polymer
2005;46(1):193–201.
vealed that the nanoparticle size decreased, the nanoparticle size [14] ten Brinke JW, Debnath SC, Reuvekamp LAEM, Noordermeer JWM. Mechanistic
distribution narrowed, and the agglomeration trend of silica weak- aspects of the role of coupling agents in silica–rubber composites. Compos Sci
ened after TSM modification. Such behaviors were due to the cou- Technol 2003;63(8):1165–74.
[15] Sae-oui P, Sirisinha C, Thepsuwan U, Hatthapanit K. Roles of silane coupling
pling layer which existed between the surface of silica and TESPT, agents on properties of silica-filled polychloroprene. Eur Polym J
resulting in decreasing silica–silica interaction. The SSBR nano- 2006;42(3):479–86.
composites filled with silica modified by TSM and OSM were also [16] Dohi H, Horiuchi S. Locating a silane coupling agent in silica-filled rubber
composites by EFTEM. Langmuir 2007;23(24):12344–9.
prepared. TEM observations indicated that the dispersion of TSM [17] Castellano M, Conzatti L, Costa G, Falqui L, Turturro A, Valenti B, et al. Surface
silica in nanocomposites was better than that of OSM silica nano- modification of silica: 1. Thermodynamic aspects and effect on elastomer
composites. Especially, TSM modification formed a coupling layer reinforcement. Polymer 2005;46(3):695–703.
[18] Krysztafkiewicz A, Jesionowski T, Binkowski S. Precipitated silicas modified
to improve the compatibility with rubber matrix before com-
with 3-aminopropyltriethoxysilane. Colloids Surf A: Physicochem Eng Aspects
pounding, the processing energy consumption decreased 25.7%, 2000;173(1–3):73–84.
21.1% and 22.1% when the TESPT amount was 8 wt.%, 12 wt.%, [19] Jesionowski T, Krysztafkiewicz A. Preparation of the hydrophilic/hydrophobic
silica particles. Colloids Surf A: Physicochem Eng Aspects 2002;207(1–
and 15 wt.% respectively. Furthermore, compared with OSM mod-
3):49–58.
ification, TSM modification could decrease the filler network struc- [20] Heinrich G. The dynamics of tire tread compounds and their relationship to
ture and improve the wet skid resistance of nanocomposites filled wet skid behavior. Prog Colloid Polym Sci 1992;90:16–26.
with the modified silica, which was investigated by DMTA. These [21] Heinrich G, Dumler HB. Wet skid properties of filled rubbers and the rubber–
glass transition. Rubber Chem Technol 1998;71(1):53–61.
results are of great importance to develop new approaches in sur- [22] Grosch KA. The rolling resistance, wear and traction properties of tread
face modification of silica and promote the practical application of compounds. Rubber Chem Technol 1996;69(3):495–568.
TSM modification in rubber industry. [23] Chen L, Zhang LQ, Tang RF, Lu YL. J Appl Polym Sci 2012;124(1):689–98.
[24] Park JT, Seo JA, Ahn SH, Kim JH, Kang SW. Surface modification of silica
nanoparticles with hydrophilic polymers. J Ind Eng Chem 2010;16(4):517–22.
[25] Hua DB, Tang J, Jiang JL, Gu ZQ, Dai LL, Zhu XL. Controlled grafting modification
Acknowledgements of silica gel via RAFT polymerization under ultrasonic irradiation. Mater Chem
Phys 2009;114(1):402–6.
The research was supported by the Program for Changjiang [26] Brunauer S, Emmett PH, Teller E. Adsorption of gases in multimolecular layers.
J Am Chem Soc 1938;60(2):309–19.
Scholars and Innovative Research Team in University (PCSIRT, [27] Robertson CG, Lin CJ, Rackaitis M, Roland CM. Influence of particle size and
IRT0807), the National Natural Science Foundation of China polymer-filler coupling on viscoelastic glass transition of particle-reinforced
(51073008 and 51103005) and the Fundamental Research Funds polymers. Macromolecules 2008;41(7):2727–31.
[28] Wang MJ, Mahmud K, Murphy LJ, Patterson WJ. Carbon–silica dual phase filler,
for the Central Universities in China. a new generation reinforcing agent for rubber Part I. Characterization. Kaut
Gummi Kunst 1998;51(5):348–60.
[29] Zeng ZQ, Yu HP, Wang QF, Lu G. Effects of coagulation processes on properties
References of epoxidized natural rubber. J Appl Polym Sci 2008;109(3):1944–9.
[30] Takino H, Nakayama R, Yamada Y, Kohjiya S, Matsuo T. Viscoelastic properties
[1] Zou H, Wu SS, Shen J. Polymer/silica nanocomposites: preparation, of elastomers and tire wet skid resistance. Rubber Chem Technol
characterization, properties, and applications. Chem Rev 1997;70(4):584–94.
2008;108(9):3893–957. [31] Aso O, Eguiazabal JI, Nazabal J. The influence of surface modification on the
[2] Wu CL, Zhang MQ, Rong MZ, Friedrich K. Silica nanoparticles filled structure and properties of a nanosilica filled thermoplastic elastomer.
polypropylene: effects of particle surface treatment, matrix ductility and Compos Sci Technol 2007;67(13):2854–63.
particle species on mechanical performance of the composites. Compos Sci [32] Evans LR, Fultz WC. Tread compounds with highly-dispersible silica. Rubber
Technol 2005;65(3–4):635–45. World 1998;219(2):38–44.