Construction and Building Materials xxx (xxxx) xxx

Contents lists available at ScienceDirect

Construction and Building Materials

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

Evaluation on the mechanical performance of recycled asphalt mixtures

incorporated with high percentage of RAP and self-developed
rejuvenators
Jizhe Zhang a, Chenchen Guo a, Tingting Chen c, Wenwu Zhang b, Kai Yao a, Chao Fan b, Ming Liang a,
Chunhui Guo b, Zhanyong Yao a,⇑
a
School of Qilu Transportation, Shandong University, Jinan 250061, China
b
Shandong HI-SPEED Group, Shandong Expressway Building, 8# Longao North Road, Lixia District, Jinan 250014, China
c
Key Laboratory of Expressway Maintenance Technology Ministry of Transport, Shandong Transportation Research Institute, Jinan 250001, China

h i g h l i g h t s

Lower viscosity rejuvenator can enhance the cracking resistance of recycled asphalt mixtures.

Rejuvenator with a relatively high-viscosity can obtain better moisture durability.

Adding rejuvenators can restore the dynamic response of recycled asphalt mixtures.

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

Article history: Because of the scarcity of irreplaceable natural resources and environmental pollution of demolished
Received 8 September 2020 asphalt pavement, circular utilization of Reclaimed Asphalt Pavement (RAP) to prepare recycled asphalt
Received in revised form 6 October 2020 pavement mixtures was promoted. To restore properties of the aged bituminous binder in RAP, rejuve-
Accepted 13 October 2020
nators designed with diverse viscosity levels and their rejuvenating effects have been investigated in pre-
Available online xxxx
vious studies. However, the influence of self-developed rejuvenators on the mechanical performance of
high percent RAP recycled asphalt mixtures is not clear. In this study, two self-developed rejuvenators
Keywords:
with different viscosities were incorporated in recycled asphalt mixtures and their influences on mechan-
Reclaimed asphalt pavement
Recycle
ical properties were assessed comprehensively. The wheel tracking test and three-point bending test
Ageing were conducted to assess the rutting resistance and cracking behavior of recycled asphalt mixtures,
Rejuvenator respectively. The Hamburg Wheel Tracking (HWT) test and Simple Performance Test (SPT) were applied
to evaluate the moisture durability and dynamic response of recycled asphalt mixtures. Fatigue resis-
tance of recycled asphalt mixtures was assessed by using two-point bending test. The experimental
results indicated that using the self-developed rejuvenators improved the cracking behavior and fatigue
life of recycled asphalt mixtures. Incorporation of RAP improved the rutting resistance and dynamic mod-
ulus of recycled asphalt mixtures. Even though adding rejuvenators caused the recycled asphalt mixtures
prone to permanent deformation, their moisture durability was still better than that of the new asphalt
mixture. It thus suggests that the mechanical behavior of recycled asphalt mixtures can be improved by
using self-developed rejuvenators.
Ó 2020 Elsevier Ltd. All rights reserved.

1. Introduction

Hot mix asphalt (HMA) is the most applied road material in

pavements all over the world. Experienced to environmental
⇑ Corresponding author.
attacks and repeated traffic loading, it may not only results in
E-mail addresses: jizhe.zhang@sdu.edu.cn (J. Zhang), 17865316091@163.com (C.
pavement distresses but also promotes the ageing of bituminous
Guo), chentingting@sdjtky.cn (T. Chen), 544818480@qq.com (W. Zhang), yaokai@s-
du.edu.cn (K. Yao), 844746446@qq.com (C. Fan), ming.liang@sdu.edu.cn (M. Liang), binder [1,2]. As the performance of asphalt pavement is reduced,
qypgjj@126.com (C. Guo), zhanyong-y@sdu.edu.cn (Z. Yao). the asphalt mixture on the pavement surface layer has to be

https://doi.org/10.1016/j.conbuildmat.2020.121337
0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.

Please cite this article as: J. Zhang, C. Guo, T. Chen et al., Evaluation on the mechanical performance of recycled asphalt mixtures incorporated with high
percentage of RAP and self-developed rejuvenators, Construction and Building Materials, https://doi.org/10.1016/j.conbuildmat.2020.121337
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

maintained by removing the old one. The milling machine is nor- be effectively enhanced by adding a proper dosage of rejuvenator,
mally used to mill and crush the old pavement and obtain but without causing obvious negative impact [21]. Zhang et al.
Reclaimed Asphalt Pavement (RAP) [3]. It has been estimated that developed several composite rejuvenators with different viscosi-
the annual generation of RAP in China is about 220 million tons ties and characterized their influence on the properties of the aged
and this value is still ramping [4]. Different from the new asphalt binders [22,23]. Although the rheological behavior and stiffness of
mixture, the RAP material has been seriously aged because of the the aged bituminous binder can be restored by rejuvenators with
long serving time subjected to air and solar radiation. The ageing different viscosities, the cracking resistance and moisture resis-
process results in the evaporation and oxidative condensation of tance of the rejuvenated binder seemed sensitive to the rejuvena-
light components, which in turn causes the change of chemical tor viscosity. It was demonstrated that the moisture and cracking
composition and mechanical performance of the bituminous bin- resistance of recycled asphalt mixtures can theoretically be guar-
der [5,6]. anteed if rejuvenators with appropriate viscosity incorporated.
Because of the scarcity and of non-renewable natural resources However, the influence of these rejuvenators on the mechanical
and environmental pollution of demolished asphalt pavement, cir- performance of recycled asphalt mixtures with a high percentage
cular utilization of RAP to produce recycled asphalt mixture has of RAP has not been investigated systemically.
been promoted and encouraged. The use of RAP offers two main In this research, the rejuvenators developed in the previous
advantages, which are a reduction of pavement construction cost researches by authors were incorporated and used for preparing
and environmental benefits [7,8]. It has been calculated that using recycled asphalt mixtures with a high percentage of RAP. Their
RAP with the dosage of 50% can reduce the production cost of related mechanical properties were tested and compared with
asphalt mixture from 70$ per ton to 50$ per ton. The production the original asphalt mixture. The rutting resistance and thermal
cost of the asphalt mixture could be reduced by 50%-70% as the cracking behavior of recycled asphalt mixtures were evaluated by
RAP dosage increased to 100% [9]. With respect to environmental using wheel tracking test and three-point bending test, respec-
benefits, the incorporation of RAP in asphalt mixtures cuts down tively. The Simple Performance Test (SPT) and the Hamburg Wheel
the demand on non-renewable resources, such as bitumen and Tracking (HWT) test were applied to determine the dynamic
aggregate [10,11]. In addition, using RAP materials in asphalt pave- behavior and moisture resistance of recycled asphalt mixtures.
ment construction reduces the fuel consumption and greenhouse Finally, the fatigue life of recycled asphalt mixtures was character-
gas emission [12]. Moreover, the consumption of RAP can also ized by using the two-point bending test.
reduce the landfill space for disposal of RAP [13].
However, the bituminous binder in RAP is aged and has a high
viscosity that behaves stiffer than that of the fresh bitumen [5]. If 2. Materials and experimental methods
the recycled asphalt mixture is produced with a high RAP percent-
age, the aged bituminous binder is prone to cause adhesion loss 2.1. Materials
and cracking susceptibility of the constructed pavement [14].
Research conducted by Arshad et al. found that the performance The bituminous binder applied in this study is a type of original
of recycled asphalt mixtures with the incorporation of RAP showed bitumen with a penetration grade of Pen 60/80. This bitumen was
significant variability in the increasing of RAP percentage [15]. So, produced in Huarui Asphalt Company. The conventional properties
it is of great importance to recover the rheological behavior of the of this base bitumen were assessed based on the Chinese standards
aged bitumen by adding proper additives to rejuvenate and guar- JTG E20-2011, and the parameters of various properties are listed
antee the long-term performance of the recycled HMA. in Table 1.
In recent years, rejuvenators have been investigated by many The rejuvenators used in this research were optimized from an
researchers to characterize the feasibility of increasing the RAP authors’ previous research that the R2 and R3 designed with an
percent in the production of recycled asphalt mixtures. From the appropriate viscosity were proved to perform well in rejuvenating
chemical composition point of view, the rejuvenator is incorpo- the aged binders [22]. The primary formulation of the rejuvenator
rated to supplement the loss of light components during the ageing was first prepared by using three original materials with the mass
process, which in turn recovers the colloidal behavior and rheolog- ratio of aromatic oil: plasticizer: anti-stripping agent = 100: 10: 2.
ical properties of the aged bitumen [16]. Normally, the rejuvenator The R2 and R3 were finally assembled by adding soft bitumen (110
should contain high aromatic dosage to restore the original ratio of pen) into the primary formulation with the dosage of 20% and 30%,
maltenes to asphaltenes, which in turn improves the temperature respectively. In this research, these two rejuvenators were
susceptibility and hardening susceptibility of the aged bitumen employed to produce recycled asphalt mixtures and compared
[17]. Nahar et al. studied the influence of rejuvenators with differ- with the original asphalt mixture. The optimum dosages of these
ent viscosities on the rheological and microstructural performance two rejuvenators were determined by the penetration test and
of rejuvenated bituminous binders [18]. The results indicated that Brookfield viscosity test at 90℃ and the detailed explanation was
two types of rejuvenators (emulsion and liquid) performed differ- shown in [22]. The optimal dosages of these two rejuvenators
ent mechanisms for rejuvenating the aged binder, which in turn based on the aged bitumen dosage in RAP materials were 10% for
obtained distinct chemo-physical properties. Shen et al. investi- R2 and 12% for R3, respectively.
gated the properties of recycled asphalt mixtures produced by The new limestone aggregates with four size groups (0–3 mm,
incorporating rejuvenators and the softer binders [19]. The results 3–5 mm, 5–10 mm, 10–20 mm) and the limestone powder were
indicated that recycled asphalt mixtures prepared by adding reju- collected from Wenzu stone quarry in Jinan, China. The RAP
venators obtained higher strength and lower rutting depth than
those prepared with the softer binders. Walaa et al. studied the
influence of three rejuvenators on the performance of recycled Table 1
Conventional parameters of Pen 60/80 base bitumen.
mixtures with high percentage of RAP [20]. It was revealed that
rejuvenators diffused into the aged bitumen film, which in turn Items Results Technical requirements
reduced the air voids of the recycled asphalt mixture. In addition, Softening point (℃) 46.2 46
rejuvenators also contributed to the resistance of recycled asphalt Penetration (25℃, 0.1 mm) 68.2 60–80
mixtures to thermal cracking. Le et al. also found that the Ductility (15℃, cm) >150 100
Dynamic viscosity (60℃, Pas) 208 180
properties of recycled asphalt mixtures at low temperature can
2
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

materials as recycled aggregates were obtained from the pavement

surface of an old expressway named Bin-Lai Expressway. The
milling machine was used to remove the surface layer of asphalt
pavement and obtain old asphalt mixture. The collected RAP mate-
rials were then crushed by using a counterattack crusher to sepa-
rate some agglomerated aggregates and followed by sieving to
obtain three size groups (0–5 mm, 5–10 mm, 10–20 mm). The
burning furnace was used to measure the binder contents of RAP
materials and the binder content of these three size groups are
0–5 mm: 6.93%, 5–10 mm: 4.38% and 10–20 mm: 3.45% by mass.
The passing rate of the new aggregates, RAP materials and lime-
stone powder are presented in Table 2.
In order to evaluate the mechanical performance of recycled
asphalt mixtures, AC-20 dense asphalt mixtures were designed.
Firstly, an original AC-20 aggregate gradation was designed with
Fig. 1. Aggregate grading curves of new mixture and recycled mixtures.
new aggregates and limestone powder, which was considered as
a benchmark to compare with recycled asphalt mixtures. In the
recycled mixes, the aggregate grading curves of two recycled Table 3
asphalt mixtures were designed by using new aggregates, lime- Asphalt mixtures investigated in the research and their sample IDs.
stone powder and RAP materials, with the RAP dosages of 40%
Sample composition Sample ID
and 60%, respectively. The grading curves of these three asphalt
New asphalt mixture New
mixtures are shown in Fig. 1. The optimal binder dosage of the
Recycled mixture with 40% RAP RM40
new asphalt mixture was determined to be 4.6% based on the Mar- Recycled mixture with 40% RAP and R2 RM40-R2
shall stability test and volumetric parameters. In addition, the opti- Recycled mixture with 40% RAP and R3 RM40-R3
mal bitumen content of the recycled asphalt mixtures with the RAP Recycled mixture with 60% RAP RM60
dosage of 40% and 60% were finally determined as 4.3% and 4.4%, Recycled mixture with 60% RAP and R2 RM60-R2
Recycled mixture with 60% RAP and R3 RM60-R3
respectively. In which, the recycled binder ratio of these two recy-
cled asphalt mixtures are RM40 for 2.21% and RM60 for 2.73%.
With respect to investigate the performance development of recy-
cled asphalt mixtures with the incorporation of rejuvenators, dif- cutting procedure to obtain prismatic beams with dimensions of
ferent recycled asphalt mixtures were prepared, and the 250 mm  35 mm  30 mm. These prepared samples were first
specimen IDs are shown in Table 3. conditioned at 10℃ for at least 1.5 h before testing. The condi-
tioned samples were then assembled onto the test fixture with a
span length of 200 mm. Finally, the upper loading head was
2.2. Experimental methods induced on the mid-span of the prismatic beam at a constant load-
ing rate of 50 mm/min. In order to reduce the experimental error,
Several testing methods, including Three-point Bending Test, four replicated tests were performed for each asphalt mixture.
Wheel Tracking Test, Hamburg Wheel Tacking Test, Dynamic During testing, the deflection in the mid-cross section was used
Modulus Test and Two-point Bending Test, were employed to eval- for calculating the maximum bending strain as shown in Equation
uate the performance of recycled asphalt mixtures with high RAP (1):
dosage. The experimental images of each test were shown in Fig. 2.
6hd
eB ¼ ð1Þ
2.2.1. Three-point bending test (3 PB) L2
Based on previous literatures, the addition of RAP materials
Where, eB is the maximum bending strain, le; h is the height of the
seriously affect the cracking resistance of recycled asphalt
beam, mm; d is the deflection of the beam in the mid-cross section;
mixtures at low temperature [24,25]. Because of this, the cracking
L is the span of the testing fixture, 200 mm.
resistance becomes an important parameter to reflect the
durability of recycled asphalt mixtures at low-temperature. The
Three-point Bending test (3 PB) was thus employed to measure 2.2.2. Wheel tracking test
the cracking resistance of recycled asphalt mixtures, based on Due to the incorporation of RAP materials, the recycled asphalt
the Chinese standard JTG E20-2011 (T0715). According to the mixtures normally obtain higher stiffness which in turn results in
standard, the asphalt mixture specimens with dimensions of an enhancement on the rutting resistance of recycled asphalt mix-
300 mm  300 mm  50 mm were compacted and followed by a tures at a high temperature [26]. However, the addition of rejuve-

Table 2
Grading curves of different aggregate materials.

Materials Passing percentage (%)

26.5 19 16 13.2 9.5 4.75 2.36 1.18 0.6 0.3 0.15 0.075
RAP 10–20 mm 100 97.32 91.83 78.67 41.86 24.35 16.55 11.55 7.82 5.77 4.22 2.99
RAP 5–10 mm 100 100 100 99.87 96.00 15.80 1.07 0.63 0.55 0.51 0.45 0.15
RAP 0–5 mm 100 100 100 100 100 98.47 64.87 44.09 28.63 18.26 13.31 2.93
New 10–20 mm 100 85.47 59.75 33.67 8.41 0.03 0.00 0.00 0.00 0.00 0.00 0.00
New 5–10 mm 100 100 100 99.87 96.00 15.80 1.07 0.63 0.55 0.51 0.45 0.15
New 3–5 mm 100 100 100 100 100 96.07 16.98 0.20 0.17 0.17 0.13 0.02
New 0–3 mm 100 100 100 100 100 100 99.93 73.24 50.33 33.39 24.29 13.19
Powder 100 100 100 100 100 100 100 100 99.61 99.20 96.35 75.28

3
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

Fig. 2. Experimental images of asphalt mixture tests used in this research: (a) Three-point Bending Test, (b) Wheel Tracking Test, (c) Hamburg Wheel Tacking Test, (d)
Dynamic Modulus Test and (e) Two-point Bending Test.

nators (softening agent) soften the RAP materials and lower the 2.2.3. Hamburg Wheel Tracking (HWT) test
rutting resistance of recycled asphalt mixtures. The rutting resis- The Hamburg Wheel Tracking (HWT) test was considered as a
tance of asphalt mixtures was therefore assessed through the repeatable and credible method for characterizing the premature
Wheel Tracking Test in accordance with the Chinese standard failure of asphalt mixtures [28]. This test was conducted according
JTG E20-2011 (T0719). This test was considered as a reliable to the standard AASHTO T 324–11. Before testing, Superpave Gyra-
approach to evaluate the permanent deformation of asphalt mix- tory Compactor (SGC) was employed to prepare cylinder-shaped
tures because the induced loading is close to the actual axial load- asphalt mixture samples with the dimensions of 62 mm in height
ing [27]. Specimens used for the Wheel Tracking Test were and 150 mm in diameter. Then, two samples were trimmed on
compacted asphalt specimens with dimensions of edges and assembled into one testing specimen [29]. The assem-
300 mm  300 mm  50 mm. Before testing, asphalt mixture slabs bled specimens were immersed into the water bath at 50℃ and
were conditioned in a testing chamber with a temperature con- the specimen needed to be pre-conditioned with the period of
trolled at 60℃ for at least 5 h to make sure the temperature dis- 5 h. During testing, the steel wheel was loaded on the specimen
tributed homogeneously inside the specimen. A rubber tire was surface and applied a repeated loading with a speed of 52 ± 2
then loaded on the specimen surface with a pressure of 0.7 MPa passes/minute. The rutting depth development of the specimen
and a speed of 42 rpm. During testing, the Linear Variable Differen- was monitored with a LVDT and the testing procedure was stopped
tial Transformer (LVDT) assembled in the equipment record the automatically when a failure criterion of either 12.5 mm rutting
rutting depth development automatically. In order to reduce the depth or 20,000 passes was reached [30]. After testing, a deforma-
experimental error, three replicated tests were performed for each tion development curve was generated and the rutting resistance
asphalt mixture. The dynamic stability was obtained according to as well as the moisture sensitivity of each specimen can be evalu-
the Equation (2), by which a higher value indicates a better resis- ated from it. Fig. 3 presents a standard rutting development curve
tance to the high-temperature rutting. of the WHT test. In the creep region, the creep slope represents rate
of rutting deformation, and higher value indicating much prone to
rutting destruction [31]. The stripping slop represents the defor-
15N
DS ¼ ð2Þ mation speed of asphalt mixtures in the stage of moisture damage
RD60  RD45
occurred. So, a large stripping slope value of asphalt mixture indi-
cating more sensitive to moisture damage [32]. The stripping
Where, DS is the dynamic stability (cycles/mm), N is the loading inflection point (SIP) is the point where the deformation curve
speed (42 cycles/min); RD60 is the rutting depth at 60 min; RD45 transformed from the creep slope to the stripping slope, and this
is the rutting depth at 45 min. indicates the bitumen film starts to strip from the aggregate [33].

4
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

Nf/50 was defined as the fatigue failure criterion when the modulus
was reduced by 50%.

3. Results and discussion

3.1. Low-temperature performance of recycled asphalt mixtures

The 3-point bending test was conducted to assess the influence

of RAP on the thermal cracking of recycled asphalt mixtures, and
their maximum bending strains are presented in Fig. 4. The maxi-
mum bending strain is a parameter to indicate the flexibility of
asphalt mixtures at low temperature, with a high value implies a
better resistance to thermal cracking at low temperatures. In
accordance with the Chinese Standard (JTG D50-2017), the maxi-
mum bending strain value should be larger than 2000 me to guar-
antee the service reliability of asphalt mixtures. For the reference
of the new mixture, its maximum bending strain was 2921 me,
which was the highest value among these mixtures. However,
Fig. 3. Schematic diagram of a standard WHT rutting development curve [31]. directly adding RAP without rejuvenators seriously lowered the
maximum bending strain of recycled asphalt mixtures, and its
value was 1981 me for RM40 and 1846 me for RM 60, respectively.
2.2.4. Dynamic modulus test This implies that the recycled asphalt mixture with the RAP per-
The dynamic modulus of asphalt mixture is the ratio of the cent over 40% was difficult to fulfill the low-temperature cracking
applied axial stress to the generated axial strain. This is considered requirement if no rejuvenator was incorporated. Meanwhile, the
as an important parameter to evaluate the performance of asphalt addition of rejuvenators increased the maximum bending strain,
mixture and provide guidance for structural design. The dynamic indicating its contribution to the cracking resistance of related
modulus was assessed using a Simple Performance Test (SPT) asphalt mixtures. By comparing recycled asphalt mixtures after
according to the AASHTO:TP62-03 standard [34]. The SPT equip- incorporating rejuvenators, those with 40% RAP obtained a higher
ment can induce a haversine compressive stress to the asphalt maximum bending strain. It is indicated that a higher RAP dosage
mixture and obtain an axial strain with the amplitude of 50–150 deteriorated the cracking resistance of recycled asphalt mixtures,
micro-strains. Cylindrical asphalt mixture specimens (150 mm in even though rejuvenators was incorporated. It was noticed that
diameter and 170 mm in height) used for dynamic modulus eval- the addition of rejuvenator R3 obtained the highest maximum
uation were prepared by Superpave Gyratory Compactor (SGC). bending strain for the recycled asphalt mixture prepared with
Then, these cylindrical samples were cored from the center to 40% RAP, while the rejuvenator R2 was useful to obtain the highest
obtain diameter of 100 mm, followed by trimming both ends to maximum bending strain for the recycled asphalt mixture pre-
obtain 150 mm in height. Before testing, the cylindrical specimens pared with 60% RAP. It is thus suggested that the rejuvenator with
were needed to be conditioned in an environmental chamber for at relatively lower viscosity is more suitable for recycled asphalt mix-
least 4 h. In this research, the SPT test was conducted at 20℃ with tures with a high dosage of RAP.
the loading frequencies from 0.1 Hz to 25 Hz. Finally, the dynamic
modulus jE j can be calculated by Equation (3):
3.2. High-temperature performance of recycled asphalt mixtures
r0
jE j ¼ ð3Þ
e0 The wheel tracking test was performed to investigate the
Where, jE j is the dynamic modulus (MPa), r0 is the applied maxi- rutting resistance development of asphalt mixtures with the
mum stress (MPa), and e0 is the measured peak strain.

2.2.5. Two-point bending fatigue test

Fatigue cracking is one of common distresses of asphalt pave-
ment due to the repeated loading of vehicles. Because of the incor-
poration of RAP, the fatigue resistance of recycled asphalt mixtures
is prone to be impacted [35]. As such, it is of great importance to
reveal the fatigue behavior of recycled asphalt mixtures with dif-
ferent RAP dosages and the addition of rejuvenators. The two-
point bending test was employed to assess the fatigue behavior
of recycled asphalt mixtures. The specimens used for the 2-point
bending test were prepared by first compacting asphalt mixtures
with dimensions of 300 mm  300 mm  50 mm. Then, these
asphalt mixture slabs were cut into trapezoidal specimens with
dimensions: 250 mm (height)  25 mm (thickness)  70 mm (lar-
ger base)  25 mm (smaller base). Before testing, the prepared
trapezoidal specimens were glued between the top and the bottom
plates. The test was performed at 10℃ with the sinusoidal load fre-
quency controlled at 25 Hz and the displacement was 100 micro-
deformations (mdef). This test was performed according to the
European Standard BS EN 12697–24. The initial E* modulus was
captured at the 100th load application and the number of cycled Fig. 4. Maximum bending strain of different asphalt mixtures at low temperature.

5
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

incorporation of RAP and rejuvenators. Fig. 5 presents the dynamic

stability of asphalt mixtures before and after the addition of RAP
materials at high temperature. It was found that the new asphalt
mixture without the addition of RAP resulted in the lowest
dynamic stability of 1065 cycles/mm. However, the addition of
RAP in the asphalt mixtures seriously enhanced the dynamic sta-
bility of recycled asphalt mixtures, with the values 2173 cycles/
mm and 2687 cycles/mm for RM40 and RM60, respectively. This
is because RAP containing the aged bituminous binders improved
the stiffness of recycled asphalt mixtures, and lowered the defor-
mation of asphalt slabs under repeated loading [36]. The incorpo-
ration of rejuvenators obtained reduction in the dynamic stability
with the value declined to 1424 cycles/mm for RM40-R2 and
1467 cycles/mm for RM40-R3, respectively. This demonstrated
that the rejuvenators R2 and R3 can both soften the aged binder
of RAP in recycled asphalt mixtures to obtain similar dynamic sta-
bility values. However, the recycled asphalt mixtures after adding
rejuvenators still showed higher dynamic stability than that of the
new asphalt mixture. With respect to the recycled asphalt mix-
tures prepared with 60% RAP, adding R2 resulted in lower dynamic
Fig. 6. Deformation development of different asphalt mixtures obtained by HWT in
stability (1596 cycles/mm) than that of the R3 (1732 cycles/mm). the water bath.
This phenomenon suggests that the rejuvenator with a lower vis-
cosity improved the rejuvenating efficiency in comparison with a
relatively higher viscosity. With respect to the recycled asphalt is due to the coupled impact of rutting and moisture attack on
mixtures prepared with the same rejuvenator, a higher RAP dosage recycled asphalt mixtures. As 40% of RAP incorporated, the
resulted in a higher dynamic stability. In summary, the incorpora- improvement effect of RAP on the rutting resistance seems greater
tion of RAP has a positive influence on the high-temperature rut- than its attenuation effect on the moisture susceptibility. That is
ting resistance of recycled asphalt mixtures. Even though why the RM40 obtained the lowest permanent deformation of
incorporating R2 and R3 reduced the dynamic stability of recycled 2.94 mm. When the RAP dosage increased to 60%, the permanent
asphalt mixtures, but the rutting resistance of these mixtures were deformation of RM 60 increased to 3.24 mm because of the atten-
higher than that of the new asphalt mixture. uation effect of RAP on the moisture susceptibility. This phe-
nomenon also proved that increasing RAP percentage has
detrimental effect on the moisture resistance of recycled asphalt
3.3. Moisture sensitivity of recycled asphalt mixtures in the water bath
mixtures. By incorporating rejuvenators, the permanent deforma-
tion of recycled asphalt mixtures experienced various increasing
The HWT test was conducted to characterize the moisture sen-
degrees. This is because the rejuvenator softened the aged bitumen
sitivity of asphalt mixtures before and after the addition of RAP as
and reduced the stiffness of recycled asphalt mixtures. By compar-
listed in Table 3. The deformation development of various asphalt
ing two rejuvenators, the recycled asphalt mixture prepared with
mixtures with the increase of loading cycles are presented in Fig. 6.
R2 resulted in a higher permanent deformation than that with
The deformation of asphalt mixture slabs developed with more
R3, even the value of RM40-R2 is higher than that of the new
loading cycles, but all these mixtures did not still reach to the strip-
asphalt mixture. It suggests that the recycled asphalt mixture pre-
ping inflection point. The new asphalt mixture obtained relatively
pared with a relatively high-viscos rejuvenator obtained a better
higher permanent deformation with a value being 5.21 mm after
durability at high temperature in water bath.
20,000 cycles loading. It is known that the deformation of HWT

3.4. Dynamic response of recycled asphalt mixtures

The SPT was conducted to investigate the influence of RAP and

rejuvenators on the dynamic response of recycled asphalt mix-
tures. During testing, the controlled sinusoidal axial strain was
applied to the cylindrical specimen and the respond stress was
recorded by the built-in software automatically. Fig. 7 shows the
dynamic modulus and phase angle of different asphalt mixtures.
The dynamic modulus (E*) is defined as the maximum stress to
the amplitude of applied strain, which reflects the stiffness of the
tested specimen. The phase angle (d) is the time lag between the
applied stress and the respond strain. This parameter represents
the visco-elasticity of asphalt mixtures. As shown in Fig. 7 (a),
the dynamic modulus increased with the growth of testing fre-
quency. However, the phase angle presented in Fig. 7(b) experi-
enced the opposite trend with the phase angle value declining
with the increase of frequency. The new asphalt mixture obtained
the lowest dynamic modulus value, while RM60 and RM40 had a
high value. It indicated that the stiffness of recycled asphalt mix-
tures increased with the incorporation of RAP and the dynamic
Fig. 5. Dynamic stability of different asphalt mixtures at high temperature. modulus improved as well. Adding rejuvenators decreased the
6
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

Fig. 7. Dynamic modulus and phase angle of different asphalt mixtures obtained by SPT at 20℃.

dynamic modulus of recycled asphalt mixtures, but these values 3.5. Fatigue resistance of recycled asphalt mixtures
were still higher than the new asphalt mixture. By comparing recy-
cled asphalt mixtures, those prepared with a lower RAP dosage The two-point bending (2 PB) test was conducted to investigate
obtained a smaller dynamic modulus values. Under the same RAP the fatigue resistance of asphalt mixtures with the addition of RAP
dosage, these two rejuvenators with different viscosities obtained and rejuvenators. Four replicated tests were conducted for each
similar dynamic modulus. As shown in Fig. 7(b), the new asphalt asphalt mixture to minimize the trial error. During testing, the
mixture obtained the highest phase angle and the recycled asphalt flexural stiffness of asphalt mixture beams deteriorated with load-
mixtures without adding rejuvenator (RM60 and RM40) showed ing cycles. As shown in Fig. 9, the development of flexural stiffness
the lowest value. This could be attributed to the addition of RAP reflects the internal structural damage of asphalt mixtures and the
increased the rigidity of recycled asphalt mixtures and makes them destruction process can be briefly classified into three stages. At
behaved more elasticity. The incorporation of rejuvenators the beginning (Stage Ⅰ), the flexural stiffness experienced a rapid
restored the visco-elasticity of recycled asphalt mixtures to some decrease, but the rate of decline gradually slowed down. This stage
extent, but all their values were lower than that of the new asphalt represents the initial development of fatigue damage and defined
mixture. Under the same RAP percent, the phase angle of the recy- as deceleration. In Stage Ⅱ, the decline rate of the flexural stiffness
cled asphalt mixture with the addition of R2 was higher than that gradually came stable and defined as linear. In the linear stage, a
with R3. It indicated that the rejuvenator with a lower viscosity can constant value of dissipated energy at each loading cycle was
better recover the visco-elasticity behavior of recycled asphalt transformed into a low level of damage and led to the generation
mixtures. of microcracks. When the damage developed to a certain extent,
The rheological performance of recycled asphalt mixtures were the decline rate of the flexural stiffness accelerated with macroc-
also measured at 35℃ and 50℃. To have a better comparison, racks and structural failure generated.
data of dynamic modulus and phase angle at 10 Hz are presented As stated above, the fatigue test was conducted with the defor-
in Fig. 8. The dynamic modulus and phase angle seriously influ- mation control mode and the failure criterion was a 50% decline of
enced by the testing temperature. As a temperature sensitive its initial stiffness. So, the initial stiffness is the reference parame-
material, increasing the testing temperature reduced the stiffness ter to characterize the fatigue life of asphalt mixtures. Fig. 10
of the bituminous binder and the dynamic modulus of the related shows the flexural stiffness of different asphalt mixtures at the ini-
asphalt mixture. The contribution of RAP on the dynamic modu- tial stage. The initial flexural stiffness of asphalt mixtures increased
lus of recycled asphalt mixtures was also observed at high testing from 9567 MPa to 13074 MPa after incorporating 40% RAP, and this
temperatures. The recycled asphalt mixtures after incorporating value reached to 14776 MPa as the RAP percent increased to 60%.
rejuvenators (R2 and R3) resulted in decrease of dynamic modu- This is because the aged binder in reclaimed asphalt improved the
lus, but these values were still larger than that of the new asphalt stiffness of recycled asphalt mixtures. However, as the fatigue test
mixture. Phase angle showed the opposite trend with the values conducted in the deformation control mode, more energy was dis-
increasing with the testing temperature. This is because the bitu- sipated under each loading cycle and this may accelerate the fati-
men behaved more viscous at high temperature and increased the gue damage. After adding rejuvenators, the initial flexural stiffness
phase angle of related asphalt mixtures. It can be summarized of recycled asphalt mixtures decreased because of the softening
that the RAP can increase the stiffness and elasticity of recycled effect of light components and this phenomenon was correlated
asphalt mixtures at moderate and high in-service temperatures. well with the dynamic modulus results.
Incorporating the self-developed rejuvenators (R2 and R3) The fatigue life results of different asphalt mixtures are pre-
restored the dynamic response of recycled asphalt mixtures to sented in Fig. 11. It can be found that the new asphalt mixture
some extent, but it can guarantee the performance of the related achieved the highest fatigue life of 1834893, suggesting its supe-
mixtures. rior resistance to fatigue damage. However, incorporating RAP

7
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

Fig. 8. Dynamic modulus and phase angle of different asphalt mixtures at 10 Hz and different testing temperatures.

Fig. 10. Initial stiffness of different asphalt mixtures.

Fig. 9. Schematic of fatigue deterioration of asphalt mixtures with numbers of
loading cycles [37].

materials seriously deteriorated the fatigue resistance of recycled

asphalt mixtures and the fatigue life declined with the increase
of RAP dosage. The reduction in the fatigue life of recycled asphalt
mixtures can be explained in two ways. Firstly, the aged asphalt
mastic film coated on RAP aggregates reduced the adhesion
between new bitumen and RAP aggregates, which affected the
durability of the bonded asphalt mixture under the repeated load-
ing. Secondly, RAP increased the stiffness of recycled asphalt mix-
tures so that the energy dissipated during repeated loading was
higher than normal asphalt mixture. As the rejuvenators incorpo-
rated, the fatigue resistance of recycled asphalt mixtures was
improved in different degrees. For example, RM40-R3 obtained
the highest fatigue life of 1,193,108 among recycled asphalt mix-
tures. The addition of rejuvenator softened the aged bitumen and
improved the adhesion between RAP and new bitumen, which in
turn led to the increase of fatigue life. It was noticed that the Fig. 11. Fatigue life of different asphalt mixtures.

8
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

self-developed rejuvenator R3 with a relatively high viscosity Acknowledgement

exhibited more rejuvenating efficiency for the recycled asphalt
mixture prepared with a low RAP dosage (40%). In contrast, the The authors acknowledge the financial support of the National
R2 with a relatively low viscosity obtained better restoration effect Natural Science Foundation of China (Project No. 51908331), Shan-
for that with a high RAP dosage (60%). dong Natural Science Foundation Committee (Project No.
ZR2019BEE004) and Shandong Transportation Technology Plan
4. Conclusions (2019B47_1).

In this study, two self-developed rejuvenators were applied for

rejuvenating RAP to prepare recycled asphalt mixtures. Mechanical
References
performance of recycled asphalt mixtures, which including thermal
cracking resistance, high-temperature rutting resistance, moisture [1] M. Naskar, K.S. Reddy, T.K. Chaki, M.K. Divya, A.P. Deshpande, Effect of ageing
durability, dynamic response, and fatigue life were assessed to on different modified bituminous binders: comparison between RTFOT and
characterize the rejuvenating effect of rejuvenators. Moreover, the radiation ageing, Mater. Struct. 46 (7) (2013) 1227–1241.
[2] Z. Yao, J. Zhang, F. Gao, S. Liu, T. Yu, Integrated utilization of recycled crumb
influence of RAP dosage on these mechanical properties were also rubber and polyethylene for enhancing the performance of modified bitumen,
assessed. Some major findings can be summarized below: Constr. Build. Mater. 170 (2018) 217–224.
The incorporation of RAP seriously declined the maximum [3] S. Xu, J. Yu, C. Hu, D. Qin, L. Xue, Laboratory evaluation of rejuvenation effect of
reactive rejuvenator on aged SBS modified bitumen, Mater. Struct. 50 (6)
bending strain of recycled asphalt mixtures, with the value can (2017), https://doi.org/10.1617/s11527-017-1106-7.
be lower than the required one in the standard when the RAP [4] H.J. Meng, Experimental Study on Changsha Municipal Hot Mix Plant Recycling
dosage was over 40%. Incorporating rejuvenators enhanced the [D], Changsha University of Science and Technology, 2013.
[5] H. Asli, E. Ahmadinia, M. Zargar, M.R. Karim, Investigation on physical
thermal cracking resistance of recycled asphalt mixtures but it
properties of waste cooking oil – Rejuvenated bitumen binder, Constr. Build.
seems difficult to reached to similar level as that of the new asphalt Mater. 37 (2012) 398–405.
mixture. The rejuvenator with a relatively lower viscosity was [6] A. Stimilli, A. Virgili, F. Canestrari, Warm recycling of flexible pavements:
more suitable for restoring the low-temperature performance of Effectiveness of warm mix asphalt additives on modified bitumen and mixture
performance, J. Cleaner Prod. 156 (2017) 911–922.
recycled asphalt mixtures when high percent of RAP incorporated. [7] R. Izaks, V. Haritonovs, I. Klasa, M. Zaumanis, Hot mix asphalt with high RAP
The addition of RAP improved the rutting resistance of recycled content, Procedia Eng. 114 (2015) 676–684.
asphalt mixtures. The recycled asphalt mixtures after adding the [8] X. Ding, L. Chen, T. Ma, H. Ma, L. Gu, T. Chen, Y. Ma, Laboratory investigation of
the recycled asphalt concrete with stable crumb rubber asphalt binder, Constr.
self-developed rejuvenators R2 and R3 can both deteriorate rutting Build. Mater. 203 (2019) 552–557.
resistance, but these values are still higher than that of the new [9] M. Zaumanis, R.B. Mallick, R. Frank, 100% recycled hot mix asphalt: A review
asphalt mixture. and analysis, Resour. Conserv. Recycl. 92 (2014) 230–245.
[10] Q. Aurangzeb, I.L. Al-Qadi, H. Ozer, R. Yang, Hybrid life cycle assessment for
Because of high stiffness of the aged binder in RAP, its incorpo- asphalt mixtures with high RAP content, Resour. Conserv. Recycl. 83 (2014)
ration in recycled asphalt mixtures reduced the permanent defor- 77–86.
mation in HWT. Even though adding rejuvenators increased the [11] T. Chen, Y. Luan, T. Ma, J. Zhu, X. Huang, S. Ma, Mechanical and microstructural
characteristics of different interfaces in cold recycled mixture containing
permanent deformation of recycled asphalt mixtures, their mois- cement and asphalt emulsion, J. Cleaner Prod. 258 (2020) 120674, https://doi.
ture durability seems still better than that of the new asphalt mix- org/10.1016/j.jclepro.2020.120674.
ture. Using the rejuvenator with a relatively high-viscosity in [12] Y. Huang, R. Bird, O. Heidrich, Development of a life cycle assessment tool for
construction and maintenance of asphalt pavements, J. Cleaner Prod. 17 (2009)
recycled asphalt mixtures can obtain better moisture durability.
283–296.
The addition of RAP improved the stiffness of recycled asphalt [13] R.C. West, Reclaimed Asphalt Pavement Management: Best Practices, National
mixtures, which in turn resulted in a higher dynamic modulus and Center for Asphalt Technology Auburn University, Auburn, Alabama, 2010.
a lower phase angle. Adding rejuvenators can restore the dynamic [14] M.C. Cavalli, M.N. Partl, L.D. Poulikakos, Measuring the binder film residues on
black rock in mixtures with high amounts of reclaimed asphalt, J. Cleaner Prod.
response of recycled asphalt mixtures to some extent, but it was still 149 (2017) 665–672.
difficult to reach to the same level as that of the new asphalt mixture. [15] H. Arshad, Y. Qiu, Evaluation of asphalt mixtures containing reclaimed asphalt
The rejuvenator with a lower viscosity can better recover the visco- pavement, Appl. Mech. Mater. 178–181 (2012) 1522–1525.
[16] N.H. Tran, A. Taylor, R. Willis. Effect of rejuvenator on performance properties
elasticity behavior of recycled asphalt mixtures. of HMA mixtures with high RAP and RAS contents, Publication Report No.
Incorporating RAP obviously lowered the fatigue life of recycled NCAT 12-05, National Center for Asphalt Technology, Alabama, 2012.
asphalt mixtures and the fatigue life declining with the increase of [17] Á. García, E. Schlangen, M. van de Ven, Properties of capsules containing
rejuvenators for their use in asphalt concrete, Fuel 90 (2) (2011) 583–591.
the RAP percent. The fatigue life of recycled asphalt mixtures was [18] S.N. Nahar, J. Qiu, A.J.M. Schmets, E. Schlangen, M. Shirazi, M.F.C. van de Ven, G.
increased after adding self-developed rejuvenators. The rejuvena- Schitter, A. Scarpas, Turning back time: Rheological and microstructural
tor with a relatively high viscosity exhibited more rejuvenating assessment of rejuvenated bitumen, Transp. Res. Rec. 2444 (1) (2014) 52–62.
[19] J. Shen, S. Amirkhanian, J. Aune Miller, Effects of rejuvenating agents on
efficiency at low RAP percent, while the one with low viscosity superpave mixtures containing reclaimed asphalt pavement, J. Mater. Civ. Eng.
seems more suitable for high RAP dosage. 19 (5) (2007) 376–384.
[20] W.S. Mogawer, A. Booshehrian, S. Vahidi, A.J. Austerman, Evaluating the effect
of rejuvenators on the degree of blending and performance of high RAP, RAS,
CRediT authorship contribution statement and RAP/RAS mixtures, Road Mater. Pave. Des. 14 (sup2) (2013) 193–213.
[21] J. Li, F. Ni, Y. Huang. New additive used in hot in-place recycling to improve
performance of RAP mix, Presented at the 93rd Annual Meeting of the
Jizhe Zhang: Writing - original draft. Chenchen Guo: Data Transportation Research Board, Washington, DC, 2014.
curation. Tingting Chen: Methodology. Wenwu Zhang: Methodol- [22] J.Z. Zhang, C.J. Sun, P.Z. Li, et al., Effect of different viscous rejuvenators on
ogy. Kai Yao: Writing - review & editing. Chao Fan: Project chemical and mechanical behavior of aged and recovered bitumen from RAP,
Constr. Build. Mater. 239 (2020) 117755.
administration. Ming Liang: Writing - review & editing. Chunhui [23] H.G. Jiang, J.Z. Zhang, C.J. Sun, et al., Experimental assessment on engineering
Guo: Resources. Zhanyong Yao: Supervision. properties of aged bitumen incorporating a developed rejuvenator, Constr.
Build. Mater. 179 (2018) 1–10.
[24] O. Reyes-Ortiz, E. Berardinelli, A.E. Alvarez, J.S. Carvajal-Muñoz, L.G. Fuentes,
Declaration of Competing Interest Evaluation of hot mix asphalt mixtures with replacement of aggregates by
reclaimed asphalt pavement (RAP) material, Procedia – Soc. Behav. Sci. 53
The authors declare that they have no known competing finan- (2012) 379–388.
[25] M. Zaumanis, R.B. Mallick, L. Poulikakos, R. Frank, Influence of six rejuvenators
cial interests or personal relationships that could have appeared on the performance properties of Reclaimed Asphalt Pavement (RAP) binder
to influence the work reported in this paper. and 100% recycled asphalt mixtures, Constr. Build. Mater. 71 (2014) 538–550.

9
J. Zhang, C. Guo, T. Chen et al. Construction and Building Materials xxx (xxxx) xxx

[26] X. Yu, M. Zaumanis, S. dos Santos, L.D. Poulikakos, Rheological, microscopic, [32] L.N. Mohammad, M. Elseifi, W. Cao, et al., Evaluation of various Hamburg
and chemical characterization of the rejuvenating effect on asphalt binders, wheel-tracking devices and AASHTO T 324 specification for rutting testing of
Fuel 135 (2014) 162–171. asphalt mixtures, Road Mater. Pave. Des. 18 (2017) 128–143.
[27] B. Javilla, L. Mo, F. Hao, B. Shu, S. Wu, Multi-stress loading effect on rutting [33] S. Schram, R.C. Williams, A. Buss, Reporting results from the hamburg wheel
performance of asphalt mixtures based on wheel tracking testing, Constr. tracking device, Transp. Res. Rec. 2446 (1) (2014) 89–98.
Build. Mater. 148 (2017) 1–9. [34] AASHTO TP 62 (2009): Standard method of test for determining the dynamic
[28] P. Chaturabong, H.U. Bahia, Mechanisms of asphalt mixture rutting in the dry modulus and flow number for hot mix asphalt (HMA) using the asphalt
Hamburg Wheel Tracking test and the potential to be alternative test in mixture performance tester (AMPT). American Association of State Highway
measuring rutting resistance, Constr. Build. Mater. 146 (2017) 175–182. and Transportation Officials, Washington, DC, the United States.
[29] L.F. Walubita, A.N.M. Faruk, J. Zhang, et al., The Hamburg rutting test – Effect of [35] K. Su, Y. Hachiya, R. Maekawa, Study on recycled asphalt concrete for use in
HMA sample sitting time and test temperature variation, Constr. Build. Mater. surface course in airport pavement, Resour. Conserv. Recycl. 54 (1) (2009) 37–
108 (2016) 22–28. 44.
[30] Q. Lv, W. Huang, M. Zheng, H. Sadek, Y. Zhang, C. Yan, Influence of gradation on [36] J. Zhang, H. Sun, H. Jiang, X. Xu, M. Liang, Y. Hou, Z. Yao, Experimental
asphalt mix rutting resistance measured by Hamburg Wheel Tracking test, assessment of reclaimed bitumen and RAP asphalt mixtures incorporating a
Constr. Build. Mater. 238 (2020) 117674, https://doi.org/10.1016/ developed rejuvenator, Constr. Build. Mater. 215 (2019) 660–669.
j.conbuildmat.2019.117674. [37] L.D. Poulikakos, M. Pittet, A.-G. Dumont, M.N. Partl, Comparison of the two
[31] M. Solaimanian, H. John, T. Maghsoud, T. Vivek. Test methods to predict point bending and four point bending test methods for aged asphalt concrete
moisture sensitivity of hot-mix asphalt pavements. Moisture Sensitivity of field samples, Mater. Struct. 48 (9) (2015) 2901–2913.
Asphalt Pavements: A National Seminar. San Diego, CA: Transportation
Research Board 2003.

10