Construction and Building Materials xxx (xxxx) xxx

Contents lists available at ScienceDirect

Construction and Building Materials

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

Chemical and rheological characteristics of rejuvenated bitumen with

typical rejuvenators
Wentong Huang a, Yunfei Guo b, Yu Zheng a, Qile Ding a, Can Sun a, Jiangmiao Yu b,c, Minye Zhu b,
Huayang Yu b,c,⇑
a
School of Environment and Civil Engineering, Dongguan University of Technology, Dongguan, China
b
School of Civil Engineering and Transportation, South China University of Technology, Guangzhou, China
c
State Key Laboratory of Subtropical Building Science, South China University of Technology, Guangzhou, China

h i g h l i g h t s

The selected rejuvenators make the aged bitumen softer.

FTIR and GPC are reliable tests for rejuvenating effect.

Sulfoxide index, Mw, PD, and LMS correlate well with rheological parameters.

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

Article history: In recent years, reclaimed asphalt pavement (RAP) is widely used in pavement construction because of its
Received 17 May 2020 merits on both economic and environmental aspects. The use of RAP contributes to significant reduction
Received in revised form 16 September on resource consumption, construction emission and the overall construction costs. To better utilize RAP,
2020
different rejuvenators are applied to soften the aged bitumen and enhance the blending efficiency
Accepted 27 October 2020
Available online xxxx
between new and aged bitumen. Although the engineering performance of RAP mixtures has been well
studied, the interaction among new bitumen, aged bitumen and rejuvenators still remains unclear. This
study is conducted for rejuvenating mechanism investigation by chemical and rheological analyses. To
Keywords:
Typical rejuvenators
achieve this goal, three typical rejuvenators (soft bitumen, warm mix surfactant, and bio-rejuvenator)
Rejuvenating effect were mixed with aged bitumen. Various tests including Dynamic Shear Rheometers (DSR), Fourier
Functional group indices Transform Infrared (FTIR) spectroscopy and Gel Permeation Chromatography (GPC) were performed. It
Molecular distribution is found that all three rejuvenators exhibited certain effect on restoring rheological performance of aged
bitumen. The softening effect of DAT and bio-rejuvenator were strongest when aged bitumen percentages
were 30% and 80% respectively. The incorporation of appropriate percentage of aged bitumen improved
the fatigue lives of rejuvenated bitumen. Soft bitumen exhibited the optimal performance on improving
fatigue resistance of rejuvenated bitumen. Additionally, the alterations of functional group indices and
molecular distribution can be considered as explanations for the regenerating effect of rejuvenators.
Moreover, the sulfoxide index of FTIR and three GPC parameters can be used as reliable indicators to pre-
dict the rheological properties of rejuvenated bitumen.
Ó 2020 Elsevier Ltd. All rights reserved.

1. Introduction because of the significant social environmental and economic mer-

its [1,2]. Especially, the use of reclaimed asphalt pavement (RAP)
In recent decades, incorporating reclaiming or waste materials has been a common practice in pavement construction. For exam-
into infrastructure establishment has gained increasing attention ple, hot in-place recycling technology helps to save 5% cost and
reduce 16% overall environmental pollution compared to conven-
⇑ Corresponding author at: School of Civil Engineering and Transportation, South tional new asphalt pavement construction [3]. The combination
China University of Technology, Guangzhou, China. of RAP and warm mix asphalt (WMA) leads to a reduction of 12%
E-mail addresses: [email protected] (W. Huang), [email protected]. CO2, 15% energy and 15% water consumption [4]. It is generally
edu.cn (Y. Guo), [email protected] (Y. Zheng), [email protected] (Q. Ding), acknowledged that the larger amount of RAP is used, the better
[email protected] (C. Sun), [email protected] (J. Yu), [email protected]. resource conservation can be achieved. However, since the bitu-
edu.cn (M. Zhu), [email protected] (H. Yu).

https://doi.org/10.1016/j.conbuildmat.2020.121525
0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.

Please cite this article as: W. Huang, Y. Guo, Y. Zheng et al., Chemical and rheological characteristics of rejuvenated bitumen with typical rejuvenators,
Construction and Building Materials, https://doi.org/10.1016/j.conbuildmat.2020.121525
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

men in RAP has experienced significant aging and stiffening during is expected to provide guidance on the selection of different reju-
service period, using high RAP contents may cause fatigue and venators and promote their application in RAP.
thermal cracking problems [5]. In addition, the degree of blending
or blending efficiency of the virgin bitumen and aged bitumen is a 2. Experimental program
recurring concern which affects the design and performance of RAP
mixtures. 2.1. Materials and sample preparation
Different types of rejuvenators have been developed to enhance
the engineering performance of asphalt mixtures containing RAP. 2.1.1. Virgin and aged bitumen
According to Moghaddam and Baaj [6], the main types of rejuvena- Bitumen with penetration grading from 60 to 70, designated as
tors include plant oils, waste-derived oils, engineered products, as virgin bitumen (shortly labelled as VB), was used to prepare artifi-
well as traditional and non-traditional refinery base oils. Good cial aged bitumen (shortly labelled as AB). Given the uncertainty of
rejuvenators should efficiently soften aged bitumen and provide the controllability and homogeneity of extracted RAP binders from
sufficient blending between new and aged bitumen. Hugener previous pavement, the artificial aged bitumen obtained through
et al. [7] concluded that bio-based rejuvenators exhibited satisfy- laboratory aging was used. The rolling thin film oven (RTFO) test
ing performance for in-place or in-plant cold mix recycling. Zau- and the pressure-aging-vessel (PAV) test were conducted to
manis et al. [8] found that waste vegetable oil was able to achieve the standard laboratory aging [17,18]. The RTFO test
rejuvenate aged bitumen by reducing the performance grade and required the bitumen samples to be heated at 163℃ and subjected
fatigue parameters to a higher degree. Yu et al. [9] compared the to 4 L/min of a continuous airflow for 85 min. In the PAV test, the
rejuvenating effect of soft bitumen, liquid surfactant, and one RTFO aged samples were exposed to a temperature of 100℃ and a
bio-rejuvenator on artificial aged bitumen by rheological charac- pressure of 2.1 MPa for 20 h. The properties of virgin bitumen and
terization. They found that three rejuvenators showed efficient aged bitumen are shown in the Table 1.
effects on rejuvenating bitumen with low aged bitumen content.
And the bio-rejuvenator exhibited the most effective rejuvenating
2.1.2. Rejuvenators
effect when more aged bitumen was used.
In this study, three typical rejuvenators were used. One is the
Despite the fact that the engineering performance of RAP mix-
soft bitumen (80/100 penetration grade) which has been consid-
tures has been well investigated, the interaction among new bitu-
ered as an effective recycling agent to rejuvenate the aged bitumen
men, aged bitumen and different rejuvenators still remains
with high stiffness [9,14,19]. The information about the soft bitu-
unclear. Recently, many researchers are keen on the investigation
men is listed in Table 1. The warm mix surfactant, Evotherm-DAT
of the chemical characteristics of rejuvenated bitumen as well as
(shortly labelled as DAT), was selected as another rejuvenator
the correlation between rheological and chemical characteristics
among three types of WMA technologies (including foaming tech-
[10–13]. Evaluation methods including saturates, aromatics, resins,
nology, chemical additives and organic additives) and used in this
and asphaltenes (SARA) fractions, Fourier Transform Infrared (FTIR)
study. DAT as a liquid surfactant mainly consisting of fatty amine
spectroscopy and Gel permeation chromatography (GPC) have
derivatives and alkylamines, can not only compensate the light
been extensively used to investigate the chemical characteristics
fractions of aged bitumen but also facilitate the coating of bitumen
of bitumen. The incorporation of rejuvenators can contribute to
on aggregates by reducing the friction between them, thereby
the variation of constituents in the bitumen [10,13], functional
showing an advantage in incorporating higher amount of RAP into
groups indices [10,14–16], molecular weight and size distribution
asphalt mixture [20–23]. The dosage of 5% by the total bitumen
[13,14]. However, the relationships between chemical and rheolog-
weight was selected for DAT [9,24]. The third one is a bio-
ical recovery are complicated and still remain uncertainty due to
rejuvenator, containing 80% epoxidized vegetable oil-derived reju-
the variety of rejuvenators. Liu et al. [14] studied the effects of
venator, 15% Evotherm-3G and 5% Styrene Butadiene Rubber (SBR)
softer bitumen on the chemical and rheological properties of
latex. It should be noted that 5 wt% dosage of Dibutyl phthalate
reclaimed asphalt binders containing polymer modifiers. The
(DBP) was used to modify epoxidized vegetable oil-derived rejuve-
results showed the chemical changes obtained from GPC and FTIR
nator and can be beneficial to improve low-temperature perfor-
tests were not well consistent with the rheological properties ana-
mance of bitumen [25]. Evotherm-3G and SBR latex both show
lyzed by master curves, although the significant impacts of softer
promising potential to enhance the performance of aged bitumen
bitumen can be seen in terms of the two aspects. Bio-based rejuve-
[23,26]. 10% bio-rejuvenator by mass of aged bitumen was used
nators were investigated by Cavalli et al. [15], who found that the
in this study [9,25]. Table 2 presents the properties of bio-
mechanical changes resulted from the addition of bio-based reju-
rejuvenator.
venators were not attributed to the changes of function group
level. In addition, according to Mansourkhaki et al. [10], the chem-
ical characteristics of different contents of RAP binders containing 2.1.3. Sample preparation
three different modifiers were analyzed through FTIR, SARA and Eleven bitumen samples including two non-blended samples
Atomic Force Microscopy tests and have been proved to have good (virgin bitumen and aged bitumen) and nine blended bitumen
correlations with physical-rheological parameters, especially the samples with different rejuvenators and aged bitumen contents
correlations between colloidal index and Superpave fatigue factors. (30%, 50% and 80% by weight of total bitumen) were prepared in
Based on the previous studies, it has been found that chemical total. Table 3 shows the detailed description of composition of all
analysis can be an effective method to investigate the rejuvenating prepared samples. To meet the demand of homogeneous state,
effect. However, very few studies have been conducted to study the each blended sample was stirred at 135℃ with a mixing speed of
rejuvenating mechanism and effect of different rejuvenators. To fill 200 rpm for 10 min, and then reheated for 10 min to eliminate
this gap, this study utilized chemical and rheological analyses to the bubbles in the bitumen samples.
gain a better understanding of rejuvenating mechanism. Three typ-
ical rejuvenators (soft bitumen, warm mix surfactant, and bio- 2.2. Testing program
rejuvenator) were selected and mixed with bitumen containing
different contents of aged bitumen. The rheological properties Multiple stress creep compliance recover (MSCR) test and linear
and chemical characteristics of different rejuvenated bitumen amplitude sweep (LAS) test were conducted using Dynamic Shear
and their correlations were comprehensively analyzed. This study Rheometers (DSR, British, Malvern) in order to evaluate the rutting
2
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

Table 1
Properties of virgin bitumen, aged bitumen and soft bitumen.

Property Virgin bitumen Aged bitumen Soft bitumen Specification

unaged Penetration at 25 ℃(dmm) 68 23 91 ASTM D5
Softening point (℃) 49 30 46 ASTM D36
Ductility at 10℃(cm) 34 / >100 ASTM D113
Viscosity at 135 ℃ (mPa.s) 416 959 346 AASHTO T316
G*/sind at 64 ℃ (kPa) 1.543 9.751 0.9745 AASHTO M320
RTFO aged G*/sind at 64 ℃(kPa) 2.682 13.071 1.623 AASHTO M320
PAV aged G*sind at 25 ℃(kPa) 3705 6048 1901 AASHTO M320

Table 2 included in the LAS test are described as follows: the first step, a
Properties of bio-rejuvenator. frequency sweep test at 0.1% strain level with a range of 0.2–
Property Values 30 Hz was performed to provide an estimation of undamaged char-
Density at 25 ℃ (g/cm3) 1.10
acteristic of bitumen; the second step, the amplitude test with the
Flash Point (℃) >220 linearly increasing loading stain from 0% to 30% was performed at
Viscosity at 60 ℃ (Pa.s) 0.405 10 Hz. Each bitumen sample with the thick of 2 mm and the diam-
State Viscous liquid eter of 8 mm after long-term aging was prepared and tested at
25℃. The fatigue life (Nf, number of cycles to failure) can be mea-
sured by employing the following Eq. (3).
Table 3
Descriptions of all prepared samples.
Nf ¼ AcB ð3Þ
Bitumen sample Components
VB (control) 100% virgin bitumen where c refers to the expected maximum strain level, the coeffi-
S30 70% soft bitumen + 30% aged bitumen cients A and B in this equation are determined in accordance to vis-
S50 50% soft bitumen + 50% aged bitumen coelastic continuum damage (VECD) theory [33].
S80 20% soft bitumen + 80% aged bitumen
D30 DAT + 70% virgin bitumen + 30% aged bitumen
D50 DAT + 50% virgin bitumen + 50% aged bitumen 2.2.3. FTIR test
D80 DAT + 20% virgin bitumen + 80% aged bitumen
B30 Bio-rejuvenator + 70% virgin bitumen + 30% aged bitumen
FTIR as an analysis technique can be used to detect the chemical
B50 Bio-rejuvenator + 50% virgin bitumen + 50% aged bitumen functional groups within solid, liquid or gas. Many researchers
B80 Bio-rejuvenator + 20% virgin bitumen + 80% aged bitumen have successfully applied FTIR spectroscopy to characterize the
AB 100% aged bitumen aging behavior and polymer content of bitumen [5,21,34]. The
identification of functional groups of all prepared samples as well
as three rejuvenators was achieved by using Vertex 70 FTIR Spec-
and fatigue performance respectively. For chemical characteriza- trometer (German, Bruker). The testing specimens were prepared
tion, Fourier Transform Infrared spectroscopy (FTIR) and Gel per- using a liquid membrane method with a thickness of approxi-
meation chromatography (GPC) tests were conducted. mately 1 mm and infrared spectroscopy scanning range was
4000–400 cm1. The functional groups (C = O and S = O) indices
2.2.1. MSCR test were used to evaluate the effect of aged bitumen on aging and
The permanent deformation resistance of all bitumen samples the rejuvenating effect of three rejuvenators. According to Dony
at high temperature is measured by the non-recoverable creep et al. [35], the functional groups indices can be calculated via
compliance (Jnr) and the percent strain recovery (R%). The MSCR Eqs. (4) and (5).
test was conducted to obtain these two parameters using DSR in

accordance to AASHTO TP 70–10 [27]. This test is composed of Area of peak limits 1635 to 1735 cm1
IC¼O ðCarbonyl indexÞ ¼
10 loading and unloading cycles with the duration of 1 s for creep Area of peak limits 1349 to 1538 ðcm1 Þ
loading followed by 9 s for recovery. Two levels of stress namely ð4Þ
0.1 and 3.2 kPa were applied during the test. The selected plate
diameter was 25 mm and the gap between the plates was 1 mm.
The temperature 64℃ was selected to perform the MSCR test for Area of peak limits 980 to 1082 ðcm1 Þ
IS¼O ðSulfoxide indexÞ ¼
all RTFO aged samples. The calculations of Jnr and R% are shown Area of peak limits 1349 to 1538 ðcm1 Þ
as in Eqs. (1) and (2) respectively [27]. ð5Þ
Nonrecov erable Strain
J nr ¼ ð1Þ
Stress Lev el
2.2.4. GPC test
The GPC (USA, Agilent Technologies Inc.) test was performed to
recov erable Strain
Rð%Þ ¼ ð2Þ obtain the molecular weight and molecular size distribution of test
Maximum Strain
samples. Two chromatographic columns, PLgel 3 lm MIXED-E and
PLgel 5 lm 103Å, were used to separate constituents with different
2.2.2. LAS test molecular size. The specimens were dissolved in Tetrahydrofuran
The LAS test was conducted for fatigue performance evaluation. (THF) to create a solution of specific concentration (the concentra-
It is an updated fatigue measuring method which can address some tion in this study was 1 mg/mL) and then were filtered using a
drawbacks of conventional Superpave fatigue test. Compared to 0.45 lm Polytetrafluoroethylene (PTFE) syringe filter to remove
stiff-based G*sind test, LAS calculates the accumulated damage the insoluble particulates. A pump flow rate of 1 mL/min with
by repeated loading, which can better simulate the real traffic con- THF as the carrier solvent was used and the columns used for the
dition [28–31]. According to AASTHO TP 101–14 [32], two steps separation of multi-component blend were maintained at 35℃.
3
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

3. Results and discussion R value in comparison with other two types of blended bitumen
samples at the same aged bitumen content except D30. It also
3.1. Rheological properties can be observed that bitumen with bio-rejuvenator exhibited sim-
ilar or even better elastic properties compared to control at high
3.1.1. Rutting resistance temperature. This phenomenon indicated that the incorporation
The results of two main MSCR parameters, non-recoverable of bio-rejuvenator had a negative impact on the elastic behavior
compliance (Jnr) and percent recovery (R%) for the bitumen at of bitumen, whereas aged bitumen contributed to the increase in
two stress levels of 0.1 kPa and 3.2 kPa, are plotted in Fig. 1. The R value.
lower Jnr implies the superior rutting resistance while the percent-
age of recovery is corresponding to the elastic behavior of bitumen 3.1.2. Fatigue resistance
[36]. For each bitumen sample, the R value at 0.1 kPa was higher Table 4 provides the results of parameters A and -B obtained
than the counterpart at 3.2 kPa, meaning a lower stress level from LAS tests, referring to the Nf value of the testing sample at
related to a better elastic recovery due to the decreasing traffic strain level of 1% and the dependence of bitumen on the varying
loading. As expected, Jnr values of three types of rejuvenated bitu- strain amplitude respectively. The Nf values at two different stain
men at both stress levels decreased with the increasing content of levels (2.5% and 5.0%) were used to predict the fatigue lives of bitu-
aged bitumen. This is because that stiffer and more brittle aged men samples, as shown in Fig. 2 (a) and (b) respectively. It can be
bitumen can enhance the resistance to permanent deformation. seen from Fig. 2 (a), at the strain level of 2.5%, the increasing aged
And aged bitumen with the lowest Jnr values (0.6 kPa1 at bitumen content can lead to the growth of fatigue lives of rejuve-
0.1 kPa and 0.7 kPa1 at 3.2 kPa) showed the greatest resistance nated bitumen. But there is not always a positive correlation
to the deformation at high temperature. When aged bitumen con- between aged bitumen content and its impact on the Nf value of
tent was 30%, the negative effect of DAT on the rutting perfor- rejuvenated bitumen. In Fig. 2 (a), Nf values of bitumen containing
mance was extremely strong. The Jnr values of D30 were soft bitumen and bio-rejuvenator decreased by 5.7% and 8.5%
4.6 kPa1 at 0.1 kPa (Fig. 1 (a)) and 5.1 kPa1 at 3.2 kPa (Fig. 1 respectively with the content of aged bitumen rising from 50% to
(b)), significantly higher than those of S30 and B30. As aged bitu- 80%. The impact of aged bitumen on fatigue performance is consis-
men content increased to 50%, D50 and B50 had similar Jnr values, tent with the findings in previous studies [37,38]. It also should be
and both showed poorer rutting resistance compared with S50. noted that, as observed in Fig. 2 (b), the improvement in fatigue life
However, with the content up to 80%, the bitumen containing arising from the increasing aged bitumen content was not as nota-
bio-rejuvenator showed the worst rutting performance, but the bly significant as that shown in Fig. 2 (a). This indicates the
Jnr value was still lower than that of control. The influence of aged increased strain might limit the positive effect of aged bitumen
bitumen content on the R value was inverse compared with Jnr on the fatigue life. Notably, the Nf value of aged bitumen at 5.0%
value. The bitumen incorporating bio-rejuvenator showed lowest strain level showed a sharp decease of 45,562 compared with the
value at 2.5% strain level, falling to 2779 (the lowest fatigue life
among all bitumen samples). This means the fatigue performance
of aged bitumen with brittle nature showed higher dependence
on the varying strain amplitude, conforming to the highest –B
value of aged bitumen presented in Table 4. Besides, the incorpora-
tion of higher content aged bitumen led to the decrease of Nf value.
As the content increased from 50% to 80%, the reductions in Nf val-
ues at 5% strain level were 864, 89 and 376 for samples containing
soft bitumen, DAT and bio-rejuvenator respectively. The fatigue life
at 5.0% strain level is more appropriate for the evaluation of fatigue
resistance of asphalt pavement in the field [39]. The Nf value at this
strain level was used to compare the fatigue performance of virgin
bitumen and rejuvenated bitumen. Obviously, all rejuvenated bitu-
men samples showed better fatigue resistance than control. It also
can be found that bitumen rejuvenated by soft bitumen had the
greatest potential to enhance the fatigue life when aged bitumen
content was not more than 50%, while the improving effect of
bio-rejuvenator on this value was less obvious than other two
rejuvenators.

3.2. Functional groups

The OMNIC software was used for conducting the baseline cor-
rection to provide a common baseline for spectra comparison. The
corrected spectra of bitumen samples and three rejuvenators at the
wavenumber ranging from 2000 cm1 to 500 cm1 are shown in
Fig. 3. It can be seen that the soft bitumen and bio-rejuvenator dis-
played nearly the same functional group as virgin bitumen. All
three samples had lower peak at 1030 cm1 (S = O stretching),
but no distinct peak at 1700 cm1 (C = O stretching) was found
in them. When the virgin bitumen experienced the long-term
aging process, the distinct peak at 1700 cm1 was observed and
the peak at 1030 cm1 became more obvious, which can be seen
Fig. 1. Jnr and recovery for bitumen samples at two stress levels: (a) 0.1 kPa; (b) 3.2 from the spectra of aged bitumen. The significant increase in the
kP. peak at 1700 cm1 was the result of the break of the C = C bonds
4
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

Table 4
Parameters A and –B obtained from LAS tests.

VB S30 S50 S80 D30 D50 D80 B30 B50 B80 AB

A 5.38E + 05 9.82E + 05 1.21E + 06 1.38E + 06 3.53E + 05 6.93E + 05 1.62E + 06 5.95E + 05 8.74E + 05 8.20E + 05 2.28E + 06
-B 3.231 3.295 3.446 3.652 2.882 3.175 3.714 3.279 3.397 3.424 4.121

spectra among virgin bitumen, aged bitumen and three rejuvena-

tors were related to the changes of functional groups indices.
Fig. 4 shows the carbonyl and sulfoxide indices of all bitumen
samples. Obviously, the rising trend in both two indices can be
observed with the increasing aged bitumen content, and aged bitu-
men showed the highest values. This change in both indices was
consistent with findings about the spectra of aged bitumen dis-
cussed before. It also can be found that the two indices of rejuve-
nated bitumen were lower in comparison with the corresponding
value of aged bitumen, but these bitumen samples were unable
to be restored to the original level. This can be recognized as an
explanation for differences between virgin bitumen and rejuve-
nated bitumen in rheological properties obtained in previous tests.
As a result of the significant peak area at 1662 cm1 of DAT, the
carbonyl index of adding DAT bitumen samples was higher than
those of other rejuvenated bitumen samples containing the same
content aged bitumen. Therefore, the sulfoxide index was more
appropriate for comparing rejuvenating effect of three rejuvenated
bitumen. The comparison of sulfoxide index among three rejuve-
nated bitumen showed that adding soft bitumen sample and add-
ing DAT sample had the lowest value when the aged bitumen
content was 30% and 50% respectively. However, the use of bio-
rejuvenator showed the strongest effect on the reduction of the
sulfoxide index when 80% aged bitumen was used. Overall,
although the incorporation of three rejuvenators had the counter-
acting effect against the oxidation of the sulfoxide groups, it did
not mean they were capable of helping the bitumen recover to
the original intensity of functional groups.
Fig. 2. Fatigue lives for bitumen samples at two strain levels: (a) 2.5%; (b) 5.0%

3.3. Molecular distribution

3.3.1. Molecular weight and polydispersity

Table 5 presents GPC parameters including Mn (number-
average molecular weight), Mw (weight-average molecular weight)
and PD (polydispersity, is equal to Mw/Mn) of three rejuvenators
and all prepared bitumen samples. It is obvious that DAT and
bio-rejuvenator both had lower Mn, Mw and PD values than other
testing samples. With the aged bitumen content increasing, Mn,
Mw, and PD values of blended bitumen samples containing the
same rejuvenator increased while the increment in Mn was not
so prominent. This implied chemical reactions existing in the bitu-
men during the aging led to the formation of larger molecules. By
observing the changes in the Mw and PD values caused by the addi-

Fig. 3. FTIR spectra of bitumen samples and three rejuvenators.

(peak at 1603 cm1) and the generation of carbonyl during the

long-term exposure to the oxygen [40]. And the presence of larger
peak area at 1030 cm1 indicated that the impact of oxygen
absorption on the sulfoxide. Besides, unlike virgin bitumen and
other two rejuvenators, DAT showed the especially prominent
peak at 1662 cm1 (C = O stretching). The exiting differences of Fig. 4. Carbonyl and sulfoxide indices for bitumen samples.

5
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

Table 5 Fig. 5, the three integrated region included LMS (large molecular
GPC parameters for three rejuvenators and all prepared samples. size, slices 1–5), MMS (medium molecular size, slices 6–9) and
Materials Mn Mw PD SMS (small molecular size, slices 10–13) [43]. The molecular size
Soft bitumen 889 ± 40 1590 ± 79 1.7894 ± 0.0080 distributions of all testing bitumen samples are exhibited in the
DAT 508 ± 3 510 ± 3 1.0079 ± 0.0001 Fig. 6. As can be seen, with the increase in the aged bitumen con-
Bio-rejuvenator 493 ± 36 505 ± 35 1.0245 ± 0.0047 tent, each type of rejuvenated bitumen exhibited higher LMS and
VB 985 ± 8 1869 ± 83 1.8972 ± 0.0684 lower SMS except the comparison of SMS between B30 and B50.
S30 1028 ± 39 2297 ± 123 2.2349 ± 0.0352
S50 1099 ± 11 2663 ± 4 2.4239 ± 0.0266
This conformed to the growth in Mw and PD values due to the
S80 1109 ± 17 2731 ± 21 2.4626 ± 0.0562 increasing aged bitumen content. It has been found that the LMS
D30 1009 ± 6 1822 ± 238 1.8054 ± 0.2249 of bitumen showed better correlations with performance of bitu-
D50 1018 ± 14 1962 ± 134 1.9261 ± 0.1045 men compared with other sizes [41]. Therefore, the comparison
D80 1025 ± 9 2139 ± 167 2.0887 ± 0.1816
of reversing aging effects among three rejuvenated bitumen was
B30 1069 ± 13 2280 ± 23 2.1334 ± 0.0050
B50 1063 ± 3 2487 ± 117 2.3395 ± 0.1042 focused on the LMS. It was apparent that the addition of DAT con-
B80 1119 ± 64 2625 ± 18 2.3492 ± 0.1172 tributed to the lowest LMS, which is corresponding to the lower
AB 1159 ± 31 3190 ± 24 2.7531 ± 0.0532 Mw and PD values of adding DAT bitumen found before. And when
The numbers after ‘‘±” are standard deviations.
30% aged bitumen was used, bitumen with bio-rejuvenator
showed approximately similar LMS with bitumen containing soft
bitumen, whereas bio-rejuvenator led to lower LMS with aged
tion of three different rejuvenators, it can be found that the use of bitumen content increasing to 50% or more. Moreover, though
DAT had the most significant reduction effect in these two values the utilization of three rejuvenators can replenish the loss of light
of bitumen, and the PD value of D50 was extremely close to that components and thus decrease LMS, it was still hard to restore the
of the control. The reason may be that light molecular weight com- LMS to the level of the control and rebalance the distributions of
ponent in DAT played the major role in the rejuvenated bitumen. each size. This can explain why the rutting and fatigue perfor-
The incorporation of bio-rejuvenator, however, cannot restore the mance of every rejuvenated sample were different from virgin
Mw and PD values to the level of the control. This is probably bitumen counterparts.
because large molecular weight component in aged bitumen was
still in a dominant position in the blended bitumen although the 3.4. Correlation between rheological and chemical characteristics
light component in bio-rejuvenator led to the decrease in Mw
and PD values. In addition, soft bitumen made some contribution Another significant objective of this study is to explore the cor-
to complementing the light molecular weight component in the relation between rheological and chemical parameters obtained
blended bitumen, but it did not amount to a great deal. from the conducted tests, apart from the investigation concerning
the rheological and chemical characteristics of rejuvenated bitu-
3.3.2. Molecular size distribution men. The linear regression method was employed to establish this
Fig. 5 illustrates the typical chromatograms of the virgin bitu- correlation and detect the possible relationships between these
men and aged bitumen. Apparently, the elution time of the aged parameters. The selected parameters from rheological tests were
bitumen appeared earlier than the virgin one did. This can be Jnr and R value at 3.2 kPa (Jnr3.2 and R3.2), Nf at strain level of 5%
attributed to the higher molecular weights in the aged bitumen (Nf5) and –B, while the selected chemical parameters included car-
compared to virgin bitumen counterparts. The previous study con- bonyl index, sulfoxide index, Mw, PD, and LMS. Table 6 shows the
cluded that there were better relationships between molecular size linear correlation coefficients (r) between each individual pair vari-
and performance of bitumen. The worse the asphalt pavement was
with regard to the deterioration, the higher the number of large
molecules exists in the bitumen [41]. A number of approaches have
been conducted to obtain the large molecular size (LMS), and each
of them demonstrates that the chromatogram should be parti-
tioned into several slices and subsequently integrated [42]. In this
study, the chromatogram obtained from GPC test was divided into
13 slices and subsequently varying quantity of slices were inte-
grated to represent different molecular size. As presented in

Fig. 5. Typical chromatograms of the virgin bitumen and aged bitumen. Fig. 6. Molecular size distributions of bitumen samples.

6
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

Table 6
Correlations between rheological and chemical characteristics for all bitumen
samples.

Parameters Jnr,3.2 R3.2 -B Nf5

Carbonyl index 0.57 0.72 0.51 0.31
Sulfoxide index 0.89 0.81 0.80 0.19
Mw 0.69 0.71 0.81 0.10
PD 0.69 0.70 0.83 0.03
LMS 0.69 0.67 0.80 0.03

ables evaluated. The coefficient r varying from 1 to 1 can help

determine the correlation level between two sets of data, and the
larger absolute value of r implies the higher correlation level. Fur-
thermore, the detailed description of the correlation between each
pair of rheological and chemical test data, as well as the corre-
sponding information including the regression equations, R2 and
significant value (sig.) are displayed in Figs. 7–10.
Obviously, Nf5 had the poorest correlation to chemical parame-
ters compared to other rheological parameters, which is attributed
to the impact of aged bitumen on the fatigue life of bitumen. As
discussed before, the incorporation of appropriate aged bitumen
content helped improve Nf of bitumen, whereas excessive amounts
of aged bitumen may have a negative effect on this value. However,
the –B value obtained from LAS tests, referring to the sensitivity of
the bitumen to the loss of fatigue life with the increasing strain
amplitude, had the better correlation with chemical parameters
except carbonyl index. Carbonyl index was also not well correlated
to rutting resistance parameters. The reasonable explanation for
this poor correlation is that the bitumen containing DAT showed
higher carbonyl indices due to the existence of significant peak

Fig. 8. Correlation between MSCR results (Jnr, R) and GPC results: (a) Mw; (b) PD and
(c) LMS.

area at 1662 cm1 of DAT, thus making this parameter not suitable
for comparing the rejuvenating effect. And the established linear
regression models involving the use of these two parameters, Nf5
and carbonyl index, gave a higher value of sig. (larger than 0.05)
except the correlation between carbonyl index and R3.2. Normally,
a value of sig. less than 0.05 can provide a valid correlation on the
basis of the data assessed [44,45]. Thus, the relatively higher corre-
lation coefficients can be observed when these two parameters
were not involved. As shown in Table 6, the best correlation can
be observed between sulfoxide index and Jnr3.2 (r = 0.89), and sul-
foxide index also agreed better with R3.2 compared with other
chemical parameters. The GPC parameters (Mw, PD, and LMS) were
moderately well correlated to MSCR parameters, all showing the
Fig. 7. Correlation between MSCR results (Jnr, R) and FTIR results: (a) Carbonyl similar R2 value. It can be conclude from the negative correlation
index and (b) Sulfoxide index. between chemical parameters and Jnr3.2 that the lower sulfoxide
7
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

Fig. 9. Correlation between LAS results (-B, Nf5) and FTIR results: (a) Carbonyl index
and (b) Sulfoxide index.

index (i.e. the lower absorbance of oxygen) and the lower values of
Mw, PD and LMS (i.e. the lower component of large molecules) indi-
cate the higher Jnr3.2 value, implying the ability of rejuvenators to
counteract the stiffening effect resulted from the incorporation of
aged bitumen and the poor rutting resistance of bitumen.
The relationships between rheological and chemical character-
istics of rejuvenated bitumen have been investigated in some ear-
lier studies, from which some similar findings can be observed. The
good correlation between high temperature performance parame-
ters obtained from MSCR tests and ISO (sulfoxide index) has been
found [16,38]. Besides, it has also been found that carbonyl and
sulfoxide indices are well correlated with other properties includ-
ing the penetration, softening point and viscosity when RAP binder
Fig. 10. Correlation between LAS results (-B, Nf5) and FTIR results: (a) Mw; (b) PD
with one rejuvenator called cereclor was studied [46]. Although and (c) LMS.
carbonyl index is not appropriate in this study, the sulfoxide index
obtained from FTIR tests can provide a good relationship with rhe-
ological properties. Whether a certain functional group index can parameters can be used to indicate the dependency of the fatigue
show a good correlation to rheological properties or not may be life on the amplitude of strain.
dependent on the type of rejuvenator used. The high correlation
between LMS and the high-failure temperature obtained from 4. Conclusions
DSR tests also can be found in this literature [47]. Apart from the
LMS, the finding in this study showed that Mw and PD in the GPC In this study, soft bitumen, Evotherm-DAT, and bio-rejuvenator,
tests also had a good correlation with high temperature perfor- were used to rejuvenate varying percentages of aged bitumen. A
mance. However, there are limited studies involving the relation- series of laboratory tests were performed to investigate the rheo-
ship between the fatigue performance parameters from LAS tests logical and chemical characteristics of rejuvenated bitumen. Addi-
and chemical parameters. By the perusal of the study concerning tionally, the correlation between rheological and chemical
the correlation between fatigue parameters obtained from other parameters was investigated. Based on analyses and results, the
tests and chemical characteristics, it has been found that the col- following conclusions are drawn:
loidal index obtained from SARA tests can be considered as a effec-
tive indicator to evaluate Superpave fatigue factors [10]. But in this
The incorporation of aged bitumen led to a lower Jnr value and a
study, the fatigue lives of bitumen obtained from LAS tests were higher R value, while rejuvenated bitumen showed poorer rut-
not well correlated with chemical parameters, while the chemical ting resistance than the aged bitumen. The negative effects of

8
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

Evortherm-DAT and bio-rejuvenator on the rutting resistance References

were strongest when aged bitumen percentages were 30% and
[1] M. Ameri, A. Mansourkhaki, D. Daryaee, Evaluation of fatigue behavior of high
80% respectively.
reclaimed asphalt binder mixes modified with rejuvenator and softer bitumen,

Incorporating appropriate content of aged bitumen improved Constr. Build. Mater. 191 (2018) 702–712, https://doi.org/10.1016/
the fatigue performance while Nf of bitumen samples contain- j.conbuildmat.2018.09.182.
ing a larger amount of aged bitumen showed higher depen- [2] H. Yu, Z. Zhu, Z. Zhang, J. Yu, M. Oeser, D. Wang, Recycling waste packaging
tape into bituminous mixtures towards enhanced mechanical properties and
dency on the strain amplitude. And soft bitumen showed the environmental benefits, J. Cleaner Prod. 229 (2019) 22–31, https://doi.org/
best improvement in fatigue performance with using no more 10.1016/j.jclepro.2019.04.409.
than 50% aged bitumen. [3] R. Cao, Z. Leng, M.-S.-C. Hsu, Eco-efficiency analysis on asphalt pavement
rehabilitation alternatives: hot in-place recycling (HIPR) and milling and filling

Aged bitumen showed higher carbonyl and sulfoxide indices. (M&F), DEStech Transactions on Engineering and Technology Research (ictim)
The sulfoxide index was more appropriate for restoration eval- (2016).
uation. Soft bitumen, Evotherm-DAT, and bio-rejuvenator [4] M.I. Giani, G. Dotelli, N. Brandini, L. Zampori, Comparative life cycle assessment
of asphalt pavements using reclaimed asphalt, warm mix technology and cold
showed the greatest effect on decreasing the sulfoxide index in-place recycling, Resour. Conserv. Recycl. 104 (2015) 224–238, https://doi.
when 30%, 50% and 80% aged bitumen was used, respectively. org/10.1016/j.resconrec.2015.08.006.
But the intensity of functional groups cannot be restored to [5] Z. Yang, X. Zhang, Z. Zhang, B. Zou, Z. Zhu, G. Lu, W. Xu, J. Yu, H. Yu, Effect of
aging on chemical and rheological properties of bitumen, Polymers 10(12)
the level of virgin bitumen. (2018) 1345

Three rejuvenators all possessed the ability to decrease the Mw, [6] T.B. Moghaddam, H. Baaj, The use of rejuvenating agents in production of
PD, and LMS values that had increased due to the incorporation recycled hot mix asphalt: A systematic review, Constr Build Mater 114 (2016)
805-816
of aged bitumen, and the significant reduction appeared in the
[7] M. Hugener, M.N. Partl, M. Morant, Cold asphalt recycling with 100% reclaimed
adding Evotherm-DAT samples. asphalt pavement and vegetable oil-based rejuvenators, Road Materials and

Sulfoxide index from FTIR tests and GPC parameters (Mw, PD, Pavement Design 15 (2) (2014) 239–258, https://doi.org/10.1080/
and LMS) were well correlated with MSCR parameters (Jnr3.2, 14680629.2013.860910.
[8] M. Zaumanis, R.B. Mallick, L. Poulikakos, R. Frank, Influence of six rejuvenators
R3.2) and –B obtained from the LAS test, indicating the possibil- on the performance properties of Reclaimed Asphalt Pavement (RAP) binder
ity of using chemical properties to evaluate the certain rheolog- and 100% recycled asphalt mixtures, Constr. Build. Mater. 71 (2014) 538–550,
ical properties of bitumen. https://doi.org/10.1016/j.conbuildmat.2014.08.073.
[9] J. Yu, Y. Guo, L. Peng, F. Guo, H. Yu, Rejuvenating effect of soft bitumen, liquid
surfactant, and bio-rejuvenator on artificial aged asphalt, Constr Build Mater
It should be noted that this study mainly focuses on the rheo- 254 (2020) 119336.
logical and chemical characteristics of bitumen mixed with differ- [10] A. Mansourkhaki, M. Ameri, D. Daryaee, Application of different modifiers for
improvement of chemical characterization and physical-rheological
ent typical rejuvenators, as well as their correlations. Future parameters of reclaimed asphalt binder, Constr. Build. Mater. 203 (2019)
studies will focus on investigating the rejuvenating mechanism 83–94, https://doi.org/10.1016/j.conbuildmat.2019.01.086.
of rejuvenated bitumen through Atomic Force Microscopy (AFM) [11] A. Mansourkhaki, M. Ameri, M. Habibpour, B. Shane Underwood, Chemical
Composition and Rheological Characteristics of Binders Containing RAP and
test and exploring the correlation among the different fatigue Rejuvenator, J. Mater. Civ. Eng. 32 (4) (2020) 04020026, https://doi.org/
parameters from other fatigue tests and chemical parameters. 10.1061/(ASCE)MT.1943-5533.0003016.
The comprehensive analysis on the field performance of RAP with [12] A. Mansourkhaki, M. Ameri, M. Habibpour, B.S. Underwood, Relations between
colloidal indices and low-temperature properties of reclaimed binder modified
using these typical rejuvenators is also worthwhile to be studied.
with softer binder, oil-rejuvenator and polybutadiene rubber, Constr Build
Mater 239 (2020) 117800.
[13] P.H. Osmari, F.T.S. Aragão, L.F.M. Leite, R.A. Simão, L.M.G. da Motta, Y.-R. Kim,
Chemical, Microstructural, and Rheological Characterizations of Binders to
CRediT authorship contribution statement Evaluate Aging and Rejuvenation, Transp. Res. Rec. 2632 (1) (2017) 14–24,
https://doi.org/10.3141/2632-02.
Wentong Huang: Methodology, Writing - original draft. Yunfei [14] G. Liu, E. Nielsen, J. Komacka, G. Leegwater, M. van de Ven, Influence of soft
bitumens on the chemical and rheological properties of reclaimed polymer-
Guo: Data curation, Writing - original draft. Yu Zheng: Software,
modified binders from the ‘‘old” surface-layer asphalt, Constr. Build. Mater. 79
Validation. Qile Ding: Data curation. Can Sun: Writing - original (2015) 129–135, https://doi.org/10.1016/j.conbuildmat.2015.01.002.
draft. Jiangmiao Yu: Investigation, Supervision. Minye Zhu: Inves- [15] M.C. Cavalli, M. Zaumanis, E. Mazza, M.N. Partl, L.D. Poulikakos, Aging effect on
rheology and cracking behaviour of reclaimed binder with bio-based
tigation. Huayang Yu: Conceptualization, Supervision.
rejuvenators, J. Cleaner Prod. 189 (2018) 88–97, https://doi.org/10.1016/j.
jclepro.2018.03.305.
[16] R.S. Mullapudi, K.G. Deepika, K.S. Reddy, Relationship between Chemistry and
Declaration of Competing Interest Mechanical Properties of RAP Binder Blends, J. Mater. Civ. Eng. 31 (7) (2019)
04019124, https://doi.org/10.1061/(ASCE)MT.1943-5533.0002769.
[17] AASHTO T240, Standard Method of Test for Effect of Heat and Air on a Moving
The authors declare that they have no known competing finan- Film of Asphalt Binder (Rolling Thin-Film Oven Test), American Association of
cial interests or personal relationships that could have appeared State and Highway Transportation Officials 2013 Washington, DC, USA
[18] AASHTO R 28–12, Standard Practice for Accelerated Aging of Asphalt Binder
to influence the work reported in this paper. Using a Pressurized Aging Vessel (PAV), American Association of State and
Highway Transportation Officials 2012 Washington, DC, USA
[19] J. Ekblad, R. Lundström, Soft bitumen asphalt produced using RAP, Mater
Acknowledgement Struct 50 (1) (2017), https://doi.org/10.1617/s11527-016-0871-z.
[20] D.X. Lu, M. Saleh, Laboratory evaluation of warm mix asphalt incorporating
high RAP proportion by using evotherm and sylvaroad additives, Constr. Build.
The authors sincerely acknowledge the funding support from Mater. 114 (2016) 580–587, https://doi.org/10.1016/
National Natural Science Foundation of China (Grant No. j.conbuildmat.2016.03.200.
[21] H. Yu, Z. Leng, Z. Zhang, D. Li, J. Zhang, Selective absorption of swelling rubber in
51809046), The Key Platforms and Major Scientific Research Pro- hot and warm asphalt binder fractions, Constr Build Mater 238 (2020) 117727.
jects in Universities in Guangdong Province (Grant No. [22] C. Hettiarachchi, X. Hou, J. Wang, F. Xiao, A comprehensive review on the
2016KZDXM053), The Key Research Project by Department of Edu- utilization of reclaimed asphalt material with warm mix asphalt technology,
Constr Build Mater 227 (2019) 117096.
cation of Guangdong Province (Grant No. 2018KZDXM068), Key [23] S. Wu, W. Zhang, S. Shen, X. Li, B. Muhunthan, L.N. Mohammad, Field-aged
Programs of Dongguan Social Science and Technology Develop- asphalt binder performance evaluation for Evotherm warm mix asphalt:
ment Fund (Grant No. 2020507140155), The open funding of Key Comparisons with hot mix asphalt, Constr. Build. Mater. 156 (2017) 574–583,
https://doi.org/10.1016/j.conbuildmat.2017.09.016.
Laboratory for Special Area Highway Engineering [24] H. Yu, Z. Zhu, Z. Leng, C. Wu, Z. Zhang, D. Wang, M. Oeser, Effect of mixing
[300102219517], and Key Senior Administration of Dongguan Phi- sequence on asphalt mixtures containing waste tire rubber and warm mix
losophy and Social Science Planning Project (Grant No. 2020ZD02). surfactants, J Clean Prod 246 (2020) 119008.

9
W. Huang, Y. Guo, Y. Zheng et al. Construction and Building Materials xxx (xxxx) xxx

[25] H. Zhu, G. Xu, M. Gong, J. Yang, Recycling long-term-aged asphalts using bio- Mater. 114 (2016) 665–672, https://doi.org/10.1016/
binder/plasticizer-based rejuvenator, Constr. Build. Mater. 147 (2017) 117– j.conbuildmat.2016.03.182.
129, https://doi.org/10.1016/j.conbuildmat.2017.04.066. [37] Y. Sun, W. Wang, J. Chen, Investigating impacts of warm-mix asphalt
[26] J. Li, F. Ni, Y. Huang, L. Gao, New Additive for Use in Hot In-Place Recycling to technologies and high reclaimed asphalt pavement binder content on rutting
Improve Performance of Reclaimed Asphalt Pavement Mix, Transp. Res. Rec. and fatigue performance of asphalt binder through MSCR and LAS tests, J.
2445 (1) (2014) 39–46, https://doi.org/10.3141/2445-05. Cleaner Prod. 219 (2019) 879–893, https://doi.org/10.1016/j.
[27] AASHTO Designation: TP 70-10. Standard method of test for multiple stress jclepro.2019.02.131.
creep recovery (MSCR) test of asphalt binder using a dynamic shear rheometer [38] D. Singh, D. Sawant, Understanding effects of RAP on rheological performance
(DSR). and chemical composition of SBS modified binder using series of laboratory
[28] D.A. Anderson, Y.M. Le Hir, M.O. Marasteanu, J.-P. Planche, D. Martin, G. tests, Int. J. Pavement Res. Technol. 9 (3) (2016) 178–189, https://doi.org/
Gauthier, Evaluation of Fatigue Criteria for Asphalt Binders, Transp. Res. Rec. 10.1016/j.ijprt.2016.06.002.
1766 (1) (2001) 48–56, https://doi.org/10.3141/1766-07. [39] M. Elkashef, R.C. Williams, Improving fatigue and low temperature
[29] J. Zhang, G.S. Simate, S.I. Lee, S. Hu, L.F. Walubita, Relating asphalt binder performance of 100% RAP mixtures using a soybean-derived rejuvenator,
elastic recovery properties to HMA crack modeling and fatigue life prediction, Constr. Build. Mater. 151 (2017) 345–352, https://doi.org/10.1016/
Constr. Build. Mater. 111 (2016) 644–651, https://doi.org/10.1016/ j.conbuildmat.2017.06.099.
j.conbuildmat.2016.02.175. [40] X. Hou, S. Lv, Z. Chen, F. Xiao, Applications of Fourier transform infrared
[30] A. Kumar, R. Choudhary, P.S. Kandhal, A. Julaganti, O.P. Behera, A. Singh, R. spectroscopy technologies on asphalt materials, Measurement 121 (2018)
Kumar, Fatigue characterisation of modified asphalt binders containing warm 304–316, https://doi.org/10.1016/j.measurement.2018.03.001.
mix asphalt additives, Road Materials and Pavement Design 21 (2) (2020) [41] P. Jennings, P. Pribanic, W. Campbell, K. Dawson, S. Shane, R. Taylor, High
519–541, https://doi.org/10.1080/14680629.2018.1507921. pressure liquid chromatography as a method of measuring asphalt
[31] R. Hajj, A. Bhasin, The search for a measure of fatigue cracking in asphalt composition, Bituminous Mixtures (1980).
binders – a review of different approaches, Int. J. Pavement Eng. 19 (3) (2018) [42] B.F. Bowers, Investigation of asphalt pavement mixture blending utilizing
205–219, https://doi.org/10.1080/10298436.2017.1279490. analytical chemistry techniques Doctoral thesis, The University of Tennessee,
[32] AASTHO TP 101–14, UL, Estimating Damage Tolerance of Asphalt Binders 2013.
Using the LAS, American Association of State and Highway Transportation [43] Y.S. Doh, S.N. Amirkhanian, K.W. Kim, Analysis of unbalanced binder oxidation
Officials 2013 Washington DC, USA. level in recycled asphalt mixture using GPC, Constr. Build. Mater. 22 (6) (2008)
[33] R. Micaelo, A. Pereira, L. Quaresma, M. Cidade, Fatigue resistance of asphalt 1253–1260, https://doi.org/10.1016/j.conbuildmat.2007.01.026.
binders: Assessment of the analysis methods in strain-controlled tests, Constr [44] J. Wang, T. Wang, X. Hou, F. Xiao, Modelling of rheological and chemical
Build Mater 98 (2015) 703–712. properties of asphalt binder considering SARA fraction, Fuel 238 (2019) 320–
[34] H. Yu, Z. Leng, Z. Zhou, K. Shih, F. Xiao, Z. Gao, Optimization of preparation 330, https://doi.org/10.1016/j.fuel.2018.10.126.
procedure of liquid warm mix additive modified asphalt rubber, J. Cleaner [45] M. Vamegh, M. Ameri, Rutting performance of road pavement asphalt binders
Prod. 141 (2017) 336–345, https://doi.org/10.1016/j.jclepro.2016.09.043. modified by polymers, Proceedings of the Institution of Civil Engineers-
[35] A. Dony, L. Zyian, I. Drouadaine, S. Pouget, S. Faucon-Dumont, D. Simard, V. Construction Materials (2019) 1-8.
Mouillet, T. Gabet, L. Boulange, A. Nicolai, MURE National Project: FTIR [46] G. Yaseen, I. Hafeez, Effect of Cereclor as Rejuvenator to Enhance the Aging
spectroscopy study to assess ageing of asphalt mixtures, Proceedings of the Resistance of Reclaimed Asphalt Pavement Binder, Materials 13(7) (2020)
E&E congress, 2016 1582.
[36] L. Gungat, N.I.M. Yusoff, M.O. Hamzah, Effects of RH-WMA additive on [47] J. Shen, S.N. Amirkhanian, S.-J. Lee, HP-GPC characterization of rejuvenated
rheological properties of high amount reclaimed asphalt binders, Constr. Build. aged CRM binders, J Mater Civil Eng 19(6) (2007) 515-522).

10