Construction and Building Materials 150 (2017) 860–871

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effects of polymerized sulfur on rheological properties, morphology and

stability of SBS modified asphalt
Ming Liang a,b, Xue Xin a, Weiyu Fan a,⇑, Hao Wang b,⇑, Shisong Ren a, Jingtao Shi c
a
State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266580, PR China
b
Department of Civil and Environmental Engineering, Rutgers University, Piscataway, NJ 08854, USA
c
PetroChina Fuel Oil Co., Ltd. Research Institute, Beijing 100195, PR China

h i g h l i g h t s

This work revealed polymerized sulfur had delayed vulcanization effect on SBS modified asphalt.

The morphology indicated polymer domains tend to be smaller as the elongation of mixing time.

Viscosity of SBS modified asphalt decreases as polymerized sulfur content increases.

SBS modified asphalt with high level of polymerized sulfur shows the improved storage stability.

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

Article history: The objective of this study was to investigate effects of polymerized sulfur on viscosity functions, mor-
Received 17 April 2017 phology and storage stability of SBS modified asphalt, as compared to elemental sulfur. Viscosity func-
Received in revised form 11 June 2017 tions including complex viscosity, steady flow viscosity and dynamic viscosity were measured by
Accepted 13 June 2017
dynamic shear rheometer (DSR) and Brookfield viscosity rheometer. The morphology and microstructure
Available online 19 June 2017
of modified asphalt were observed using fluorescence microscopy and Fourier Transform Infrared spec-
troscopy (FTIR). The results revealed that the increment in viscosity of modified asphalt containing poly-
Keywords:
merized sulfur was slower than that of elemental sulfur. Vulcanization rate decreases with the increase of
Vulcanization
Sulfur
polymerized sulfur, i.e. delayed vulcanization effect. In morphology, large polymer domains progressively
SBS transform to the fine dispersed SBS phase with smaller dimensions as the elongation of mixing time;
Modified asphalt whereas a fine polymer phase was found for elemental sulfur at the initial stage of mixing, which is sig-
Rheological properties nificant evidence of delayed vulcanization. Moreover, the FTIR results verify the decrease of unsaturation
of poly-butadiene block and the formation of CAS bond. Viscosities of asphalt with polymerized sulfur
were always lower than those of elemental sulfur and decreased with the increase of polymerized sulfur
content. SBS phase domain vulcanized by polymerized sulfur is larger than that of elemental sulfur and
tends to become larger with the increase of polymerized sulfur content. Lastly, SBS modified asphalt with
higher level of polymerized sulfur shows the improved storage stability.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction causing pavement distress such as permanent deformation, fatigue

The vast majority of road pavements in the world are made of
asphalt binders due to its adhesive, impermeable, and viscoelastic
properties [1]. However, increased traffic factors (e.g. heavier loads,
higher traffic volume, and higher tire pressure) as well as extreme
weather events are the great threats to pavement performance,
⇑ Corresponding authors.
E-mail addresses: ming.liang@rutgers.edu (M. Liang), xinxue19910103@163.
com (X. Xin), fanwyu@upc.edu.cn (W. Fan), hwang.cee@rutgers.edu (H. Wang),
1014869187@qq.com (S. Ren), shijingtaoupc@163.com (J. Shi).
cracking, and thermal cracking [2,3]. These undesirable distresses
shorten pavement service life and increase maintenance cost of
pavement. In order to increase longevity of pavement, high-
performance asphalt that has the stronger resistance to high-
temperature rutting and low-temperature cracking is desired in
paving construction. Polymer modification for asphalt has been
an effective method to achieve this goal. To date, the most popular
polymer for asphalt modification is styrene-butadiene styrene
(SBS) tri-block copolymers, which provides asphalt better mechan-
ical behavior at high temperatures and elastic properties at low
temperatures [4,5]. The SBS triblock chains involve a starting

http://dx.doi.org/10.1016/j.conbuildmat.2017.06.069
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871
polystyrene block, followed by an intermediate polybutadiene
block, ending with another polystyrene block, which presents a
two-phase morphology of glassy polystyrene domain dispersed
in the flexible polybutadiene matrix. Within the service tempera-
ture range of pavement, the glassy polystyrene domains contribute
to the stiffness of SBS while the polybutadiene blocks offer the
elasticity [6,7]. This special behavior of physically crosslinking
elastic network makes SBS to be a popular polymer modifier for
asphalt.
However, SBS in asphalt matrix is prone to separate out from
bulk phase at high temperatures because of their thermodynamic
instability [8,9]. It is common and inevitable for SBS modified
asphalt to be stored at high temperatures without stirring in pav-
ing field for several days or even more than weeks at a time [10].
The storage instability of SBS modified asphalt is regarded as one
of the most critical issues that significantly impairs the thermo-
mechanical properties and rheological properties of modified
asphalt. Thus, many efforts have been done to prevent the occur-
rence of phase separation [11–15]. It is commonly believed that
the most effective and practical approach is to use sulfur, usually
elemental sulfur, to improve the stability of SBS modified asphalt
through sulfur vulcanization.
As early as 1958, Welborn and Babashak [16] reported that the
addition of sulfur to the blends of rubber or latex and asphalt can
greatly improve their storage stability. Maldonado et al. [17] devel-
oped a method for preparation of the storage-stable SBS modified
asphalt by adding sulfur. Unfortunately, these attempts are not
put into field practice owing to the excessive viscosity of modified
asphalt. Until 1990, the storage-stable polymer modified asphalt
mixed with sulfur was prepared and applied in the real paving pro-
ject [18]. After that, a number of studies have focused on optimiz-
ing the preparation process, improving the performance,
understanding the reaction mechanism, and reducing the cost for
SBS modified asphalt with sulfur [19–22].
As a matter of fact, it is not an easy task to produce the storage-
stable SBS modified asphalt because the complexity and variation
of asphalt compositions and the poor compatibility among poly-
mer and asphalt. However, the exact mechanism of sulfur reacting
with SBS and asphalt is still unclear. It is commonly agreed that
sulfur chemically bond the polymer chains and couple polymer
and asphalt by means of sulfides and/or polysulfide bonds. In terms
of reaction mechanism, researchers [23–25] argued that the addi-
tional reaction of double bond and substitution of allylic hydrogen
atoms mainly account for the linking between sulfur and SBS
chains, giving rise to the decrease of unsaturation, the shifting of
the double bonds, and the molecular isomerization. While the
dehydrogenation of asphalt components and sulfur radicals may
occur in the linking reaction of sulfur and asphalt [2,26–28]. On
the other hand, sulfur vulcanization for SBS is an anti-
plasticization process and leads to the formation of more complex
higher-molecular weight networks, which easily results in gel for-
mation. SBS is susceptible to vulcanization and the dosage of sulfur
should be precisely tuned to prevent gelation. However, the prop-
erties of SBS modified asphalt varies greatly even using the same
dosage of sulfur because the vulcanization reaction process as well
as the degree of vulcanization is not easily controllable and tun-
able. Wen et al. [29,30] considered sulfur vulcanization as a pro-
cess of dynamic vulcanization and this dynamic process will
sustain until cooling down.
Based on above considerations, technical methods should be
developed to slow down and finely control sulfur vulcanization
process and thus the degree of sulfur vulcanization, thus reducing
the susceptibility of SBS to vulcanization and preventing the gel
formation. Although previous researches have studied optimizing
mixing conditions and sulfur content, little work has considered
the vulcanization process. Besides, vulcanization of SBS in asphalt
861
by polymerized sulfur is not available from the literature. Polymer-
ized sulfur, an amorphous allotrope of sulfur, is a metastable linear
polymer of sulfur, which has an average molecular weight from
100,000 to 300,000 [31,32].
The objective of the present study is to investigate the effect of
polymerized sulfur on the viscoelastic properties and storage sta-
bility of SBS modified asphalt. This will achieve the goal of control-
ling vulcanization process and reaction degree, thus decreasing
susceptibility of SBS to vulcanization, reducing potential gelation,
and improving storage stability. Therefore, elemental sulfur, vari-
ous polymerized sulfur, SBS, and asphalts were used to prepare
SBS modified asphalt containing different polymerized sulfur con-
tents. The effect of polymerized sulfur content on the performance
of SBS modified asphalt was evaluated by viscosity functions mea-
sured from dynamic oscillation shear, steady state shear, and rotat-
ing shear modes. The variations of viscosity functions with time
were also investigated. Stabilities of SBS modified asphalt mixed
with different level of polymerized sulfur were monitored by vis-
coelastic function variations.
2. Experimental testing
2.1. Materials
Two types of paving asphalt with 80/100 penetration grade
were used as base asphalts, which are abbreviated as ZH and K
in this study. Conventional properties such as penetration [33],
softening point [34], and viscosity [35] are presented in Table 1.
The SBS is a linear polymer with 30 wt% styrene and its weight-
average molecular weight is around 110,000. The elemental sulfur
and polymerized sulfur are commercial products. The elemental
sulfur was noted as S0. Three types of polymerized sulfur used in
this study are noted as S40, S60 and S90, respectively. The S40 con-
tains 40% weight of polymerized sulfur and the remaining 60% is
elemental sulfur. The contents of polymerized sulfur in S60 and
S90 are 60% and 90% accordingly.
2.2. Preparation of SBS modified asphalt with polymerized sulfur
SBS modified asphalt was prepared by melt blending method
[37,38] using a high speed disintegrating mixer at 4000 rpm and
170 °C for one hour. The operating conditions such as temperature
control, rotating speed, shearing rate, as well as stirring period
were optimized in order to maximize the rheological performance
and to minimize degradation of asphalt and polymer [39]. The con-
tent of SBS was 3% by weight of the blend for all samples. After
shear blending process, a propeller with clovers was applied to stir
the blends. At the meantime, sulfur was added by weight of 0.2% of
blends. Lastly, the blends were divided into small samples into
containers, cooled down to room temperature, and sealed by alu-
minum foil for further characterization. For the sake of brevity,
the SBS modified asphalt (ZH, K) mixed with elemental sulfur,
Table 1
Physical properties and chemical composition of the base asphalts.
Test specifications Results
Asphalt ZH Asphalt K
Physical properties
Penetration (25 °C, dmm) ASTM D 5 [33] 89 90
Softening point (R&B, °C) ASTM D 36 [34] 42.2 41.6
Viscosity (135 °C, Pas) ASTM D 4402 [35] 0.36 0.38
Chemical composition (wt%)
Saturates ASTM D4124 [36] 22.8 30.9
Aromatics 32.1 27.5
Resins 39.1 41
Asphaltene 6 0.6
862 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871
S40, S60 and S90 were noted as SZH1, SZH2, SZH3, SZH4 and SK1,
SK2, SK3, SK4, respectively. It is noteworthy that the SBS/asphalt
blends without sulfur were also prepared by the same methods
as reference samples (SZH0, SK0) for comparison. The samples
(SZH0 to SZH4 and SK0 to SK4) used to investigate the effect of
polymerized sulfur were subjected 4 h mixing, while the samples
(SZH0, SZH1, SZH2, SZH4 and SK0, SK1, SK2, SK4) used to study
the time effect underwent 2, 4, 6, 8 and 10 h mixing respectively.
2.3. Test methods and measurements
The measurement of viscoelastic properties were conducted
using dynamic shear rheometer with parallel plate geometry
(25 mm diameter and 1 mm gap, 8 mm diameter and 2 mm gap)
[11,40]. Dynamic viscosity functions were obtained through fre-
quency sweep from 0.1 to 100 rad/s (in the range of linear vis-
coelasticity) at 50 °C and 75 °C, respectively. Flow viscosity in
steady state shear in the scale of large deformation was measured
from 103 s1 to 102 s1 at 60 °C. The oscillation temperature
sweeps, with a continuous heating ramp of 1 °C/min and frequency
10 rad/s, were carried out from 10 to 100 °C.
The high-temperature rotating viscosity was evaluated by a
rotary viscometer at 135 °C. The appropriate rotator was selected
to ensure the torque within the range of 10% to 90% of the
viscometer.
Fluorescence microscopy has been used to investigate the state
of dispersion of SBS within the asphalt matrix and to evaluate the
nature of microscopic interactions among SBS, asphalt, and sulfur.
The sample was sandwiched in two glass slides. After that, the slide
was observed in microscopy with a magnification ratio of 400 at
room temperature and the images were analyzed by cellSens
software.
Storage tests were designed for evaluating the stability of poly-
mer modified for a period of storage at high temperatures [41].
Standardized aluminum tubes (25 mm in diameter and 140 mm
in height) with loaded samples were sealed and stored vertically
in an oven at 163 °C for 48 h. After completion, the tube was chilled
and divided equally into three parts and stored for further mea-
surement. In this study, the stability was measured by means of
viscoelastic functions of samples derived from the top and bottom
of the tube. The difference in viscoelastic functions between the
upper and bottom samples was used to quantify the stability of
polymerized sulfur.
Fourier transforms infrared (FTIR) spectroscopy is a simple ana-
lytical technology to characterize chemical bonds in different sam-
ples [42]. In this study, FTIR was used to determine the
microstructure of SBS modified asphalt with various sulfur in the
range of wave number from 4000 cm1 to 400 cm1.
3. Results and discussions
3.1. Viscosity results of asphalt with different sulfur materials
The complex viscosity (g⁄) obtained from frequency sweep
tests was used to determine their difference in performance. Dif-
ferences in complex viscosity for SBS modified asphalt mixed with
various sulfur materials can be observed in Fig. 1. The testing
results of reference samples are also displayed for comparison. It
was found that the reference sample without sulfur exhibited vis-
cous behavior at low frequencies and intermediate temperatures,
which could be observed from the plateau of viscosity plots. This
viscous behavior is more evident at 75 °C and for asphalt K. From
the view of colloidal model for asphalt, this finding is reasonable
for asphalt K, which has the lower asphaltene content and shows
more sol character. The addition of elemental sulfur to SBS modi-
fied asphalt gave rise to a great increase in complex viscosity that
is more obvious for asphalt K. Nevertheless, the viscosity of asphalt
ZH with elemental sulfur was found greater than that of asphalt K
at the low frequency range.
On the other hand, the incorporation of polymerized sulfur
resulted in different viscosity values in comparison to elemental
sulfur. The viscosities of all samples with polymerized sulfur are
smaller than those of asphalt with elemental sulfur. With the
increase of polymerized sulfur content, a decrease in viscosity pre-
sented. Such decreasing tendencies in viscosity became more evi-
dent at 75 °C and for asphalt K. It is believed that sulfur may act
as a vulcanization agent connecting SBS molecules with asphalt
together [24]. The greatest value in viscosity for SBS modified
asphalt with elemental sulfur corresponds to the strongest resis-
tance to deformation, which benefits from the cross-linked poly-
mer elastic network. It was found that the increase of
polymerized sulfur content weakened the deformation resistance,
which indicated inferior cross-linked network by polymerized sul-
fur in asphalt. In other words, elemental sulfur is more effective
and efficient than polymerized sulfur. This is why elemental sulfur
as a cross-linking agent for SBS may easily lead to over cross-
linking and gel formation. Since polymerized sulfur is a linear poly-
mer of sulfur, it is unstable under high temperatures and will
change into S8 rings [31]. Therefore, polymer network cross-
linked by polymerized sulfur is inferior in terms of density and
strength, generating the lower viscosity with the increase of poly-
merized sulfur content.
The steady state viscosity values at different shear rates for SBS
modified asphalt at 60 °C is shown in Fig. 2. The SBS modified
asphalt without sulfur presented Newtonian behavior in a wide
range where the steady state shear viscosity remained constant
as a function of shear rate. Since the SBS/asphalt blend is a hetero-
geneous mixture where polymer-rich phase disperse in
asphaltene-rich phase, both phases move together when subjected
to the very small shear stress. At this situation, the SBS-rich phase
will shift with asphalt phase in synchronized step. Thus, the obvi-
ous Newtonian behavior was showed for SBS modified asphalt
without sulfur. However, a great change of flow behavior occurred
after the addition of sulfur. For SBS modified asphalt ZH with sul-
fur, the shear-thinning behavior was found as the increase of shear
rate. The viscosity in an initial low rate region appeared to
approach a constant value. The elemental sulfur causes the greater
viscosity while viscosity drops with the increase of polymerized
sulfur content within the low shear range. It is noteworthy that a
slight rise in viscosity was observed at the end of curves especially
for the higher polymerized sulfur content, which indicated a larger
polymer phase domain. The samples made with asphalt K with sul-
fur showed a slight shear-thinning behavior and the difference in
viscosity caused by various sulfur materials was not as significant
as the different observed for asphalt ZH. The results from flow
curves indicate that elemental sulfur facilitate a stronger cross-
linked polymer network. However, polymerized sulfur shows an
inferior effectiveness of vulcanization on SBS in comparison to ele-
mental sulfur and the higher content of polymerized sulfur
degrades the efficiency of vulcanization. It is reasonable to con-
clude that the cross-linked polymer network in asphalt ZH is stron-
ger than the one in asphalt K.
The difference in steady state viscosity can be identified by fit-
ting Williamson model for the flow curve in Fig. 2. The equation of
Williamson model [43] is shown in Eq. (1).
g0
g¼ ð1Þ
1 þ ðkc_ Þ
s
where, g0 is zero shear-rate limiting viscosity (ZSV), k is an intrinsic
characteristic time for a certain sample, which correlates with the
consistency of asphalt, and s is flow index.
 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871 863

Fig. 1. Complex viscosity g* as a function of frequency for SBS modified asphalt mixed with different sulfur materials.

Fig. 2. Steady state shear viscosity for SBS modified asphalt mixed with various sulfur materials and reference sample.

The fitted Williamson model parameters are listed in Table 2. As
is shown, the zero shear-rate viscosity of SBS modified asphalt
underwent a significant increase after addition of elemental sulfur,
which showed the greatest value among all samples. However, the
zero shear-rate viscosity values for all samples mixed with poly-
merized sulfur were lower than elemental sulfur. Meanwhile, the
zero shear-rate viscosity decreased with the increase of polymer-
ized sulfur content.
During mixing and compaction of asphalt, viscosity plays a sig-
nificant role in the potential of the resulting paving mixtures.
864 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871
Table 2
The Williamson model parameters for SBS modified asphalt with different sulfur materials and reference sample.
Blank sample Elemental Sulfur
Asphalt ZH
g0 (Pas) 595.23 1860.85
k (s) 0.0356 0.724
s 0.905 0.3264
Asphalt K
g0 (Pas) 479.6 590.6
k (s) 0.0219 0.0365
s 0.891 0.0628
Therefore, high-temperature viscosity was measured to investigate
the effect of different sulfur materials on SBS modified asphalt. The
viscosities at 135 °C for SBS modified asphalt are presented in
Fig. 3. It was found that when asphalt was mixed with small
amounts of sulfur, the viscosity at 135 °C increased for SBS/asphalt
blends. The sample with elemental sulfur showed the highest vis-
cosity at 135 °C and the viscosity of samples with polymerized sul-
fur decreased as the increasing content of polymerized sulfur. The
increase in viscosity of asphalt K was not as significant as the one
for asphalt ZH after incorporating sulfur, because asphalt K con-
tains less asphaltene. Consequently, all viscosity results indicated
that modified asphalt ZH had the stronger resistance to deforma-
tion which was provided by sulfur vulcanization. Thus it is reason-
able to deduce that adding sulfur to SBS modified asphalt with the
lower penetration can improve mechanical property at high
temperatures.
3.2. Comparison of morphology of asphalt with different sulfur
materials
To explain these rheological results, fluorescence microscopy
was used to shed light on the morphology and interaction of poly-
mer domain within asphalt matrix. Fluorescence microscopy
images for SBS modified asphalt ZH without and with different sul-
fur materials are presented in Fig. 4. Meanwhile, it is very helpful
to explain the rheological results from the view of the molecular
structure of SBS. Fig. 5 displays schematic illustration of micro-
scopy for SBS modified asphalt from micron and nano scale.
As can be seen in Fig. 4, a heterogeneous system was formed for
all SBS modified asphalt, including polymer-rich domain (highlight
area) and asphaltene-rich phase (dark area). For SBS modified
asphalt without sulfur (Fig. 4A), polymer-rich phase was dispersed
Fig. 3. Dynamic viscosity for SBS modified asphalt with different sulfur and blank
sample at 135 °C.
S40 S60 S90
1666.5 1421.0 1399.7
0.5244 0.249 0.178
0.290 0.328 0.309
543.7 533.9 531.9
0.0203 0.0269 0.0169
0.699 0.835 0.707
as a discrete phase in asphalt when adding 3% SBS. Also, polymer-
rich phase was formed by SBS particles swelling light components,
which showed circular shapes with the diameters less than 20 lm.
Within the polymer-rich phase (Fig. 5a), polystyrene domains were
surrounded by polybutadiene matrix. At the ambient temperature,
a rubbery supporting network is constructed by crystalline poly-
styrene domain connecting with flexible polybutadiene block,
which brings about enhanced elastic response in the high-
frequency region and improves viscosity compared to base asphalt.
But such polystyrene domain will dissociate when temperature
exceeding the Tg of polystyrene and SBS molecules diffuse in the
asphalt matrix. The physical cross-linked network will disappear
and cannot contribute to elastic behavior. Therefore, the SBS/
asphalt blends showed viscous behavior in low-frequency region
and Newtonian behavior at high temperatures that is similar to
base asphalt.
Contrary to this, when elemental sulfur was incorporated into
SBS/asphalt blends, chemical reaction occurred, i.e. vulcanization.
Vulcanization of SBS by elemental sulfur generated smaller SBS-
rich phases and promoted dispersion of polymer, forming a smaller
and less marked highlight area as shown in Fig. 4B. It is commonly
agreed that elemental sulfur can react by two approaches: chemi-
cally bonding between sulfur and the polymer chains and the cou-
pling of polymer and asphalt by means of sulfides and/or
polysulfide bonds [2]. Vulcanization takes place through polybuta-
diene block, resulting in a chemical cross-linked domain and creat-
ing a more complex macromolecule. At the ambient temperature,
both physical and chemical cross-linked network (Fig. 5b) can
work together to resist the stress imposed by traffic or temperature
cycles. As a result, SBS modified asphalt with elemental sulfur
exhibits the much greater complex viscosity and zero shear-rate
viscosity than reference sample at low to intermediate tempera-
ture ranges. Moreover, these chemical linking bonds are much
stronger than the physical aggregation of polystyrene domains
and will not disappear at high temperatures. In other words, the
permanent chemical cross-linked domain is very beneficial to
improve resistance to rutting at high temperatures. This explains
why SBS modified asphalt with elemental sulfur presents the most
significant shear-thinning behavior and the highest viscosity at
135 °C.
An interesting phenomenon was observed for polymerized sul-
fur in Fig. 4(C–E), regarding to the microscopy of polymer phase.
The size of SBS phase domain vulcanized by polymerized sulfur
was found larger than that of elemental sulfur. Besides, polymer
phase in Fig. 4(C–E) seems to form a larger cluster and to be more
distinct as the increase of polymerized sulfur content, indicating
the inferior compatibility.
As mentioned previously, polymerized sulfur is an amorphous
allotrope of sulfur and a metastable linear polymer of sulfur with
an average molecular weight from 100,000 to 300,000. When poly-
merized sulfur was mixed into blends, it kept the polymer macro-
molecule and easily formed aggregation domain during the
preparation of modified asphalt. Previous study also argued that
 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871
Fig. 4. Fluorescence microscopy images for SBS modified asphalt ZH without sulfur
(A) and with elemental sulfur (B), S40 (C), S60 (D) as well as S90 (E).
the dispersibility of polymerized sulfur was poor [44]. Therefore,
vulcanization may merely occur at the both ends of sulfur polymer
chain and the border of aggregation domain of polymerized sulfur.
Consequently, the density of cross-linking network, the degree of
865
vulcanization, and the complexity of high-molecular-weight net-
work are lower than those of elemental sulfur. Because the disso-
ciative sulfur molecules to blends after addition of elemental
sulfur are much more than polymerized sulfur, the amount of sul-
fur involved in chemical vulcanization is very vast, giving rise to a
greater cross-linking density, deeper vulcanization, and more com-
plex network. That’s why asphalt samples with elemental sulfur
have the highest complex viscosity, zero shear-rate viscosity and
viscosity at 135 °C. Furthermore, with the increase of polymerized
sulfur content, the aggregation domain of sulfur polymer become
larger and thus vulcanization become more difficult as well as
poorer, which weakens the strength of chemical cross-linked
domain (Fig. 5c). This illustrates why complex viscosity, zero
shear-rate viscosity and viscosity at 135 °C decrease with the
increase of polymerized sulfur content.
3.3. Variations of viscosity with mixing time
It is a common experience that the dosage of elemental sulfur
and operating condition need be strictly designed to obtain the
desirable properties. Otherwise, it easy to cause gel formation.
Generally, it is believed that gelation is caused by over-
crosslinking of polymer where a super high-weight molecular net-
work is created through sulfur/polysulfide bonding. With respect
to vulcanization reaction, previous studies considered the prepara-
tion of SBS modified asphalt containing sulfur as a process of
dynamic vulcanization [29,30,45]. They observed that the torque
necessary to keep a constant rotating speed of mixer progressively
increased for the asphalt/SBS/sulfur ternary blend whereas the tor-
que always remained unchanged for asphalt/SBS binary blend,
which was an indirect evidence of chemical reaction. In this study,
the changes of complex viscosity at 50 °C, rotating viscosity at
135 °C with mixing time were investigated using dynamic shear
rheometer and Brookfield viscometer and the results are shown
in Figs. 6 and 7, respectively.
It was found that overall the complex viscosity decreased with
the increase of frequency, which approached the similar value in
high frequency region (low temperature). There was a great differ-
ence in viscosity at low frequency (high temperature) as the
change of time for samples with various sulfur materials. With
the extension of mixing time, the complex viscosity for reference
sample almost kept constant within the period of first four hours
and then increased when mixing time exceeded six hours. More
increments were observed in complex viscosity for the longer time.
The hardening phenomenon of asphalt matrix mainly accounts for
this trend for reference sample in addition to the physical entan-
glement among polymer and asphalt. Hardening of asphalt is
caused by volatilization of light components and the asphaltene
concentration resulting from partial oxidation of the light fractions,
particularly a transformation of resins into asphaltenes. Compared
with the reference sample without sulfur, the greater increment in
complex viscosity was observed within the interval of two hours
for SK1. The complex viscosity of sample with elemental sulfur
was always higher than that of reference sample. It is noted that
the complex viscosity underwent a significant increase for mixing
with 10 h, increasing the risk of gel formation. On the other hand,
the increase in complex viscosity at low frequency for SK2 was not
as fast as that of SK1. Also, the increase in complex viscosity for
SK4 is slower than that of SK2. The viscosity of SK2 after 10-h mix-
ing is lower than that of SK1. As for SK4, such trend of drastic
increase in viscosity after 10-h mixing disappeared.
Fig. 7 displays the evolution of viscosity at 135 °C as a function
of mixing time for SBS modified asphalt with different sulfur. As
can be seen, the viscosity of all sample increased with the exten-
sion of mixing time, but the rate of such increase differed. The vis-
cosity of sample with elemental sulfur increased faster than the
866 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871

Fig. 5. Schematic illustration of microscopy for SBS modified asphalt without sulfur (a) and with elemental sulfur (b), polymerized sulfur (c) from different dimensions.

sample with polymerized sulfur and this rate became lower with
the increase of polymerized sulfur content. These results indicate
that vulcanization rate decreases for polymerized sulfur.
Fig. 8 shows the changes of morphology with mixing time for
SBS modified asphalt ZH containing S90. For the sample taken after
four-hour mixing, the domain of SBS phase tended to aggregate
each other to form a very large cluster, the dimensions of which
exceeded 100 lm (Fig. 8a). The distinct binary character of SBS
modified asphalt indicates the poor compatibility between asphalt
and polymer. With the elongation of mixing time, the cluster of
polymer domain became smaller and the size of them was less
than 100 lm (Fig. 8b). Meanwhile, the average distance between
the clusters increased which indicated that the dispersion became
better over time. Furthermore, a markedly change was observed
after 12-h mixing and a fine dispersion of SBS phase instead of
large clusters was observed in Fig. 8c. The smaller polymer-rich
domain and the smooth boundary between two phases indicated
that extension of mixing time facilitated dispersion of polymer
and improved the compatibility.
It is interesting to compare the morphology of polymerized sul-
fur and elemental sulfur. At the same mixing time (for example,
four hours), the size of SBS phase with polymerized sulfur was
much larger than that of elemental sulfur (Fig. 4A and E). The pro-
longed mixing time renders smaller polymer domain in size. The
morphology of sample with polymerized sulfur after 12-h mixing
approached that of sample with elemental sulfur after four-hour
mixing. Molecular structures of polymerized sulfur and its changes
in system can give a reasonable explanation on these results. Poly-
merized sulfur is a metastable high polymer of sulfur, which is
unstable at high temperatures and will convert into smaller sulfur
molecules. At the initial period of mixing, polymerized sulfur
maintains a character of polymer macromolecule and form aggre-
gation due to poor dispersibility of polymer macromolecules. Con-
sequently, vulcanization occurs around the aggregation of
 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871 867

Fig. 6. Variations of complex viscosity with mixing time for SBS modified asphalt with different sulfur materials at 50 °C.

Fig. 7. Viscosity at 135 °C as a function of mixing time for SBS modified asphalt with different sulfur.

polymerized sulfur, which acts as a cross-linking point and gives
rise to the enrichment of SBS. The chemical cross-linked domain
formed by polymerized sulfur is also large because of high macro-
molecules of sulfur (Fig. 8a). Thus, the dimension of SBS phase is
very large for the sample after four-hour mixing. At this situation,
the lower density of cross-linking network renders the low viscos-
ity. As the extension of mixing time, large polymerized sulfur
aggregations become unstable at 170 °C and some convert into
small molecule sulfur or even elemental sulfur. The small molecule
sulfur departs from the main part of the large polymerized sulfur
and diffuses into asphalt matrix, which serves as a new cross-
linking point. In other words, aggregation of polymerized sulfur
becomes small and its dispersibility improves under effects of time
and temperature (Fig. 8b and c). The more dissociative sulfur, the
better dispersion, and the smaller size of polymerized sulfur make
cross-linking deeper, denser, and stronger. As a result, viscosity
keeps growing except for attribution of aging.
Besides, the compatibility between asphalt and polymer
improve progressively and the polymer-rich phase becomes not
so highlighted. It is slower conversion of polymerized sulfur to
868 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871

Fig. 8. Changes of morphology as a function of mixing time for SZH4 and their schematic illustrations.

Fig. 9. FTIR spectra of SBS modified asphalt without and with elemental sulfur and polymerized sulfur (S90).

small molecule sulfur and progressive improved dispersibility that
made vulcanization rate slower than elemental sulfur. Once ele-
mental sulfur is added into SBS/asphalt blends, a vast amount of
small S molecule acting as cross linking point makes vulcanization
fast and the cross-linked network is created in a short time. At this
situation, if vulcanization is not terminated in time, over cross link-
ing will occur which lead to gel formation. That is why the dosage
of elemental sulfur and operating condition must be strictly
designed. Compared to elemental sulfur, polymerized sulfur has
an effect of delayed dynamic vulcanization.
In order to verify the above-mentioned schematic illustration
and clarify the reaction of sulfur vulcanization, FTIR spectra of
SBS modified asphalt without and with elemental sulfur as well
as polymerized sulfur (S90) were measured as depicted in Fig. 9.
The strong peaks in 2923 cm1 and 2852 cm1 were caused by ali-
phatic CAH stretching vibrations of asphalt fractions. The peak at
1610.1 cm1 and the small peaks at 750–900 cm1 were C@C and
CAH vibrations in aromatics. It is noteworthy that a slight decrease
in intensity of peak at 1459.3 cm1 and 1376.2 cm1 was observed
for the sample with sulfur, which indicates the asymmetric
deforming in ACH2A and ACH3 and the symmetric vibrations in
ACH3. Some researchers argued that the peak intensity of ACH2A
and ACH3 decreased after vulcanization [22]. Another significant
difference is the peak at 966.6 cm1, which corresponds to the
bending vibration of ACH@CHA in polybutadiene block. The
decrease in 966.6 cm1 means the reduction of unsaturation of
PB block, which indicates double bonds in polybutadiene partici-
pate in cross-linking reaction. More importantly, the peak at
 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871 869

Fig. 10. Evolution of elastic modulus and viscous modulus with frequency at 50 °C for the top and bottom samples in the tube after storage.

668.2 cm1 is attributed by CAS stretching vibrations and the (G0 ) between the top and bottom samples narrowed as the increase
increased intensity after vulcanization, which is a significant evi- of polymerized sulfur content, which means the improved storage
dence of formation of CAS bond and also verifies the reasonability stability. The results show that the sample with S90 has the best
for schematic illustrations as shown in Figs. 5 and 8. storage stability, followed by the sample with elemental sulfur,
S60, and S40.
3.4. Thermal stability of asphalt with different sulfur materials The distribution of discrete relaxation times used in the gener-
alized Maxwell model is shown in Fig. 10, from which the zero
The most important objective of sulfur vulcanization for SBS shear-rate viscosity (g0, ZSV) can be calculated (Table 3). The zero
modified asphalt is to improve the storage stability under quies- shear-rate viscosity of the top sample was much greater than that
cent condition for a period of time, which is the major concern of the bottom sample and it decreased as the increase of polymer-
for polymer modified asphalt during application at high tempera-
tures. In this study, storage stability was distinguished by the dif-
ference in viscoelastic modulus of samples from the top as well as Table 3
ZSV and stability index for SBS modified asphalt with different sulfur materials.
the bottom of tube after storage for 48 h at 163 °C. Not only that,
the generalized Maxwell model was used to fit the viscoelastic Element Sulfur S40 S60 S90
modulus and to quantify the stability. The results were presented Asphalt ZH
in Fig. 10. As can be seen, viscous modulus (G00 ) for the top and bot- g0,top  103 (Pas) 30.3 3360 222 9.83
tom samples was almost the same; whereas a big difference in g0,bottom  103 (Pas) 4.25 4.28 4.56 4.33
SI 7.1 785.0 48.7 2.3
elastic modulus (G0 ) existed among two samples. Additionally,
elastic modulus (G0 ) of the top sample was greater than that of Asphalt K
g0,top  103 (Pas) 5.04 10.2 9.63 4.18
the bottom sample in the tested frequency region, indicating shift-
g0,bottom  103 (Pas) 1.80 1.82 1.83 1.98
ing of separated polymer phase to the upside of the tube. For the SI 2.8 5.6 5.2 2.1
sample containing polymerized sulfur, the gap in elastic modulus
870 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871
ized sulfur content. A stability index (SI) was defined as ratio of
zero shear-rate viscosity between the top and bottom samples
and thus the higher value in SI indicated the poorer storage stabil-
ity. As depicted in Table 3, the sample with S90 has the best storage
stability, followed by the sample with elemental sulfur, S60, and
S40. Besides, the zero shear-rate viscosity and stability index of
modified asphalt K were found much lower than those of asphalt
ZH, implying the better storage stability for asphalt K.
The delayed vulcanization of polymerized sulfur accounts for
the improved storage stability. The smaller polymer-rich domain
and the smooth boundary between SBS phase and asphalt phase
is caused by the delayed vulcanization of polymerized sulfur in
the late period of storage, which improves the compatibility and
results in the denser cross-linking network and the higher degree
of vulcanization. As a result, the better thermal stability corre-
sponds to higher level of polymerized sulfur.
4. Conclusions
Effects of polymerized sulfur on rheological properties, mor-
phology and stability of SBS modified asphalt were studied. Time
effect of polymerized sulfur on SBS modified asphalt was evaluated
by dynamic oscillation shear and dynamic rotating viscosity. The
rheological results were explained by the findings from fluores-
cence microscopy and FTIR. Moreover, storage stability was evalu-
ated using dynamic shear modulus and viscosity. The following
conclusions can be concluded:
1) Polymerized sulfur leads to significant increase in complex
viscosity, steady viscosity, and dynamic viscosity at 135 °C
of SBS modified asphalt compared to reference sample with-
out sulfur. These viscosity functions decrease with the
increase of polymerized sulfur content. However, viscosity
functions of samples with elemental sulfur are greater than
that of polymerized sulfur. Thus, polymerized sulfur shows
the inferior effectiveness of vulcanization on SBS compared
to elemental sulfur and the higher content of polymerized
sulfur corresponds to the poorer vulcanization efficiency.
2) The microscopy observations show that SBS phase domain
vulcanized by polymerized sulfur is larger than that of ele-
mental sulfur. The domains tend to become larger as the
increase of polymerized sulfur, implying the lower degree
of vulcanization and density of cross-linking network.
3) With the extension of mixing time, complex viscosity of SBS
modified asphalt with elemental sulfur increases faster com-
pared to reference sample. On the contrary, increase in com-
plex viscosity for polymerized sulfur is not as fast as
elemental sulfur and further become slower as the increase
of polymerized sulfur. The similar trend in viscosity at
135 °C with time was also observed, which indicates vulcan-
ization rate decreases for samples with polymerized sulfur.
4) Large SBS domain progressively transform to the fine dis-
persed SBS phase with the elongation of mixing time, indi-
cating the finer dispersion of polymer. The FTIR results
indicate the decrease of unsaturation of polybutadiene block
and formation of CAS bond, which is a significant evidence
of vulcanization. The storage test results revealed that the
better storage stability was achieved for samples with
higher content of polymerized sulfur.
Acknowledgements
This work is part of a research project supported by the Funda-
mental Research Funds for the Central Universities (Research Pro-
ject No. 24720156046A) and China University of Petroleum
Postgraduate Innovation Project (Research Project No.
15CX06046A). The authors gratefully acknowledge their financial
support.
References
[1] J. Zhu, B. Birgisson, N. Kringos, Polymer modification of bitumen: advances and
challenges, Eur. Polym. J. 54 (2014) 18–38.
[2] H. Wang, I.L. Al-Qadi, Evaluation of surface-related pavement damage due to
tire braking, Road Mater. Pavement Des. 11 (1) (2010) 101–122.
[3] J. Baek, H. Ozer, H. Wang, I.L. Al-Qadi, Effects of interface conditions on
reflective cracking development in hot-mix asphalt overlays, Road Mater.
Pavement Des. 11 (2) (2010) 307–334.
[4] M. Liang, P. Liang, W. Fan, C. Qian, X. Xin, J. Shi, G. Nan, Thermo-rheological
behavior and compatibility of modified asphalt with various styrene-
butadiene structures in SBS copolymers, Mater. Des. 88 (2015) 177–185.
[5] G. Hernández, E.M. Medina, R. Sánchez, A.M. Mendoza, Thermomechanical and
rheological asphalt modification using styrene-butadiene triblock copolymers
with different microstructure, Energy Fuels 20 (2006) 2623–2626.
[6] G. Polacco, J. Stastna, D. Biondi, L. Zanzotto, Relation between polymer
architecture and nonlinear viscoelastic behavior of modified asphalts, Curr.
Opin. Colloid Interface Sci. 11 (2006) 230–245.
[7] J.F. Masson, P. Collins, G. Robertson, J.R. Woods, J. Margeson, Thermodynamics,
phase diagrams, and stability of bitumen-polymer blends, Energy Fuels 17
(2003) 714–724.
[8] R. He, H. Dou, D. Li, Experimental research on compatibility between SBS
modifier and base asphalt, Appl. Mech. Mater. 633 (2014) 1090–1094.
[9] L. Zanzotto, J. Stastna, O. Vacin, Thermomechanical properties of several
polymer modified asphalts, Appl. Rheol. 10 (2000) 185–191.
[10] M. Liang, X. Xin, W. Fan, H. Luo, X. Wang, B. Xing, Investigation of the
rheological properties and storage stability of CR/SBS modified asphalt, Constr.
Build. Mater. 74 (2015) 235–240.
[11] H. Wang, I.L. Al-Qadi, A.F. Faheem, H.U. Bahia, S.H. Yang, G.H. Reinke, Effect of
mineral filler characteristics on asphalt mastic and mixture rutting potential,
Transport. Res. Rec. 2208 (2011) 33–39.
[12] F. Zhang, C. Hu, The research for SBS and SBR compound modified asphalts
with polyphosphoric acid and sulfur, Constr. Build. Mater. 43 (2013) 461–468.
[13] G. Polacco, P. Kříž, S. Filippi, J. Stastna, D. Biondi, L. Zanzotto, Rheological
properties of asphalt/SBS/clay blends, Eur. Polym. J. 44 (2008) 3512–3521.
[14] B. Zhang, M. Xi, D. Zhang, H. Zhang, B. Zhang, The effect of styrene-butadiene-
rubber/montmorill- onite modification on the characteristics and properties of
asphalt, Constr. Build. Mater. 23 (2009) 3112–3117.
[15] F. Zhang, J. Yu, S. Wu, Influence of ageing on rheology of SBR/sulfur-modified
asphalts, Polym. Eng. Sci. 52 (2012) 71–79.
[16] J.T. Welborn, J.F. Babashak, New rubberized asphalt for roads, J. Highway Div.
Proc. Am. Soc. Civil Eng. 1651 (1958).
[17] P. Maldonado, J. Mas, T.K. Phung, Process for preparing polymer-bitumen
compositions, US Patent 4145322, 1979.
[18] J. P. Planche, Method for the preparation of bitumen-polymer compositions,
PCT WO 9002776, 1990.
[19] G. Wen, Y. Zhang, Y. Zhang, K. Sun, Y. Fan, Rheological characterization of
storage-stable SBS-modified asphalts, Polym. Test. 21 (2002) 295–302.
[20] J.S. Chen, C.C. Huang, Fundamental characterization of SBS-modified asphalt
mixed with sulfur, J. Appl. Polym. Sci. 103 (2006) 2817–2825.
[21] F. Zhang, J. Yu, S. Wu, Effect of ageing on rheological properties of storage-
stable SBS/sulfur- modified asphalts, J. Hazard. Mater. 182 (2010) 507–517.
[22] F. Zhang, J. Yu, J. Han, Effects of thermal oxidative ageing on dynamic viscosity,
TG/DTG, DTA and FTIR of SBS-and SBS/sulfur-modified asphalts, Constr. Build.
Mater. 25 (2011) 129–137.
[23] G. Polacco, S. Filippi, Vulcanization accelerators as alternative to elemental
sulfur to produce storage stable SBS modified asphalts, Constr. Build. Mater. 58
(2014) 94–100.
[24] S.H. Chough, D.H. Chang, Kinetics of sulfur vulcanization of NR, BR, SBR, and
their blends using a rheometer and DSC, J. Appl. Polym. Sci. 61 (1998) 449–
454.
[25] P. Versloot, J.G. Haasnoot, P.J. Nieuwenhuizen, J. Reedijk, M. van Duin, J. Put,
Sulfur vulcanization of simple model olefins, Part V: double bond
isomerization during accelerated sulfur vulcanization as studied by model
olefins, Rubber Chem. Technol. 70 (1997) 106–119.
[26] J. Stastna, L. Zanzotto, O. Vacin, Viscosity function in polymer-modified
asphalts, J. Colloid Interface Sci. 259 (2003) 200–207.
[27] E. Martínez, C. Chávez, A. Herrera, N. Herrera, Comparative study of the effect
of sulfur on the morphology and rheological properties of SB- and SBS-
modified asphalt, J. Appl. Polym. Sci. 115 (2010) 3409–3422.
[28] Q. Zhang, T. Wang, W. Fan, Y. Ying, Y. Wu, Evaluation of the properties of
bitumen modified by SBS copolymers with different styrene-butadiene
structure, J. Appl. Polym. Sci. 131 (2014) 403981–403987.
[29] G. Wen, Y. Zhang, Y. Zhang, K. Sun, Y. Fan, Improved properties of SBS modified
asphalt with dynamic vulcanization, Polym. Eng. Sci. 42 (2002) 1070–1081.
[30] G. Wen, Y. Zhang, Y. Zhang, K. Sun, Z. Chen, Vulcanization characteristics of
asphalt/SBS blends in the presence of sulfur, J. Appl. Polym. Sci. 82 (2001) 989–
996.
 M. Liang et al. / Construction and Building Materials 150 (2017) 860–871
[31] F. Fairbrother, G. Gee, G.T. Merrall, The polymerization of sulfur, J. Polym. Sci.
16 (1955) 459–469.
[32] J.J. Griebel, R.S. Glass, K. Char, J. Pyun, Polymerizations with elemental sulfur: a
novel route to high sulfur content polymers for sustainability, energy and
defense, Prog. Polym. Sci. 58 (2016) 90–125.
[33] ASTM Standard D5, Standard Test Method for Penetration of Bituminous
Materials, ASTM International, West Conshohocken, PA, 2002.
[34] ASTM Standard D36, Standard Test Method for Softening Point of Bitumen
(Ring-and- Ball Apparatus), ASTM International, West Conshohocken, PA,
2002.
[35] ASTM Standard D4402, Standard Test Method for Viscosity Determination of
Asphalt at Elevated Temperatures using a Rotational Viscometer, ASTM
International, West Conshohocken, PA, 2002.
[36] ASTM Standard D4124, Standard Test Methods for Separation of Asphalt into
Four Fractions, ASTM International, West Conshohocken, PA, 2002.
[37] A. Adedeji, T. Grünfelder, F.S. Bates, C.W. Macosko, M. Stroup-Gardiner, D.E.
Newcomb, Asphalt modified by SBS triblock copolymer: structures and
properties, Polym. Eng. Sci. 36 (1996) 1707–1723.
[38] B. Golestani, F.M. Nejad, S.S. Galooyak, Performance evaluation of linear and
nonlinear nanocomposite modified asphalts, Constr. Build. Mater. 35 (2012)
197–203.
871
[39] M. Liang, X. Xin, W. Fan, S. Ren, J. Shi, H. Luo, Thermo-stability and aging
performance of modified asphalt with crumb rubber activated by microwave
and TOR, Mater. Des. 127 (2017) 84–96.
[40] M. Jasso, R. Hampl, O. Vacin, D. Bakos, J. Stastna, L. Zanzotto, Rheology of
conventional asphalt modified with SBS, Elvaloy and polyphosphoric acid, Fuel
Process. Technol. 140 (2015) 172–179.
[41] JTJ T 0661-2000, Standard test method for segregation of polymer modified
asphalt.
[42] R. Rasool, S. Wang, Y. Zhang, Y. Li, G. Zhang, Improving the aging resistance of
SBS modified asphalt with the addition of highly reclaimed rubber, Constr.
Build. Mater. 145 (2017) 126–134.
[43] H. Geng, C.S. Clopotel, H.U. Bahia, Effects of high modulus asphalt binders on
performance of typical asphalt pavement structures, Constr. Build. Mater. 44
(2013) 207–213.
[44] J.E. Kearnan, R.W. Newsbaum, Uniformly dispersible insoluble sulfur
vulcanizing compositions: U.S. Patent 3,706,708. 1972-12-19.
[45] H. Jin, G. Gao, Y. Zhang, Y. Zhang, K. Sun, Y. Fan, Improved properties of
polystyrene-modified asphalt through dynamic vulcanization, Polym. Test. 21
(2002) 633–640.