and mixtures. The use of reclaimed asphalt pavement(RAP) in virgin mixtures is increasing in quantity, and so
of recycled mixtures containing a rejuvenator. Researcheshave showed that aged asphalt binders containing a rejuve-nator can reach target PG grades if the amount of the reju-venator is appropriately determined and added .However, the rejuvenator percentage may be crucial tothe properties of the blended aged asphalt. The rejuvenator
* Corresponding author. Tel.: +1 912 681 0084.E-mail addresses: firstname.lastname@example.org (J. Shen), kcdoc@clemson.
edu (S. Amirkhanian), email@example.com (B. Tang).
Construction and Building Materi
Construction1. Background and objectives of the study
1.1. Background and literature
The implementation of SHRP binder specications hasbeen carried out in most of the United States since thenew specications were developed in 1993 [1,2]. Even incountries that the new specications have not been imme-diately introduced; however, researches related with thespecications are still in progress in both asphalt binders
far the use of RAP has not followed the specicationsbecause these specications were not initiated for the recy-cling. Given these facts, it is practically meaningful todevelop new technologies for a better use of the RAP inpavement engineering.
Rejuvenator is one of the recycling agents, suitable foreither highly oxidized or for mixtures containing a largepercentage of RAP. However, rejuvenator is not encour-aged or even not allowed for the recycling in some statesin USA because of the uncertainty of the rutting propertiesAbstract
Firstly, the performance-based properties of rejuvenated aged asphalt binders, i.e., the blends of aged binders containing a rejuvenatorat various percentages, were investigated under high, intermediate and low temperatures. The tests were conducted on the blends at threestages as follows: no aging, rolling thin lm oven (RTFO) residuals and as well RTFO + pressure aging vessel (PAV) residuals throughdynamic shear rheometer (DSR) and bending beam rheometer (BBR) tests. Optimum concentrations of the rejuvenator needed for theblends to reach a target PG grade were obtained from the blending charts of the rejuvenated aged binders in terms of performance prop-erties. The rejuvenator is a soft binder containing a low asphaltene content of 2 wt%. Secondly, selected performance-based propertieswere conducted on hot mix asphalt (HMA) using the rejuvenated aged binder and a virgin HMA as a control mixture. Results showedthat the rejuvenator aected signicantly the performance-based properties of both the rejuvenated aged binders and the mixtures con-taining the rejuvenated aged binders. It was possible to get optimum concentrations of the rejuvenator using the blending charts so thatthe rejuvenated binders reach a target PG grade. The mean value of the concentrations was proved to be more reliable through the per-formance-based properties of the mixtures if it is used for a design value for recycling. The properties of the asphalt paving mixtures withthe rejuvenated binders were even improved or in the same level as the properties of the virgin mixtures. 2006 Elsevier Ltd. All rights reserved.
Keywords: Asphalt binder; Hot mix asphalt; Rheological characteristics; Rejuvenator; Performance-based propertiesEects of rejuvenator on perrejuvenated asphalt
Junan Shen a,*, Serji Amia Chongqing Jiaotong Universi
b Civil Engineering Department, Clemson
Received 16 March 2006; received in revisedAvailable onli0950-0618/$ - see front matter 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.conbuildmat.2006.03.006rmance-based properties ofinder and mixtures
hanian b, Boming Tang a
Chongqing 400074, PR China
iversity, Clemson, SC 29634-0911, USA
m 27 March 2006; accepted 27 March 200615 June 2006
als 21 (2007) 958964
should be properly added in terms of the properties of theblends so that the properties under low temperature areimproved while the properties under high temperature arenot adversely aected. Traditional methods to determinethe percentage of the rejuvenator are predominantly basedon blending charts of either a penetration or a viscosity cri-terion [3,4]. In addition, a more integrated approach for thedetermination of the rejuvenator percentage was proposedby considering not only the penetration or viscosity crite-rion, but the composition requirements for recycled asphaltas well . However, some of these methods do not con-sider the performance-based properties of the rejuvenatedaged asphalt binders using a rejuvenator as required bySHRP specications.
rejuvenated aged binder or a virgin binder (as a controlsample) were tested and compared.
2. Materials and testing procedures
Three types of asphalt binders were used in the study.The properties of the three binders are shown in Table 1.The aged asphalt binders used for producing the blendswere not extracted directly from RAP, but were obtainedby using an accelerated aging process in the laboratoryon a virgin straight-run asphalt binder, Pen 6080, repre-senting those most popularly used in Japan. The virgin bin-
J. Shen et al. / Construction and Building Materials 21 (2007) 958964 9591.2. Objectives and scope
The objectives of the study were as follows:
(1) To investigate the eect of a rejuvenator on the per-formance-based properties of the rejuvenated agedbinder, i.e., the blends of the aged binder containingthe rejuvenator, using a series of dynamic shear rhe-ometer (DSR) and bending beam rheometer (BBR)tests.
(2) To study the inuence of the rejuvenator on perfor-mance-based properties of hot mix asphalt (HMA)containing the rejuvenated aged binders, and to com-pare the properties of the mixtures containing therejuvenated aged binders with those of virginmixtures.
In this study, a series of DSR and BBR tests were con-ducted in the laboratory in order to establish the blendingcharts of aged binder containing a rejuvenator with dier-ent percentages. The tested blends were in three agingstages of no-aging, RTFO and RTFO + PAV aging. Thepercentages of the rejuvenator needed for the blends ofaged binder and rejuvenator to reach a target PG gradewere determined from blending charts. Selected perfor-mance-based properties of the mixtures using either the
Table 1Properties of straight-run base asphalts and aged asphalt binder
Sources Penetration (25 C, 1/10 mm)
Straight-run Pen 6080 66Straight-run Pen 4060 50Aged asphalta 30
a Aged asphalt binder was produced articially with straight-run Pen 60
Table 2Properties and composition of the rejuvenator
Dynamic viscosity 60 C (Pa s) Flash point (C)
Asphaltene Saturate2.0 wt% 51.9 wt%der used for virgin mixtures was a Japanese standardquality of penetration level of 4060 (1/10 mm) . ThePen 4060 virgin binder was selected as for virgin mixturesbecause the binder was graded as PG64-22, a target gradefor the blends used for all the mixtures.
The rejuvenator used for the blends of rejuvenated agedbinder and the mixtures is a soft binder with low asphalteneof 2 wt%, as listed in Table 2. The viscosity of the rejuvena-tor at 60 C is 202 Pa s. The percentages of the rejuvenatorfor establishing blending charts of aged binder were 0, 6.0,9.0 and 14.0 wt% of the aged binders.
Based on these blending charts, optimum percentages ofthe rejuvenator used for the rejuvenated aged binders toachieve PG64-22, a Performance Grade at which DSR willfail at 64 C, and BBR will fail at 22 C, were determined.Finally, both the low and high limits of the rejuvenator per-centages, i.e., 2% and 7.4%, respectively, were obtained,Table 3.
A virgin mix design with HMA 13 conducted using vir-gin binder Pen 4060 came out an optimum asphalt per-centage (OAC) of 5.4 wt% of the dry aggregate. TheOAC was obtained by Marshall Test at a compaction con-dition of 50 blows on each size of the samples. The compac-tion condition corresponds to a middle trac. This valuewas used for all the asphalt mixture samples with the typesand gradations of the aggregates being identical. Totally,three recycled mixture samples and one virgin mixture
Softening point (C) Ductility (cm)
48 100+50 100+Not available Not available
by RTFO and PAV process.
Ratio of viscosity Unit weight (kN/m3)
Aromatic Resin33.2 wt% 12.7 wt%
sample were produced. The three recycled were producedusing rejuvenated aged binders at 0%, 2.0% and 7.4% ofrejuvenate percentages, respectively, and the virgin samplesusing a virgin binder Pen 4060.
2.2. Test methods
DSR (AASHTO T315-2) and BBR (AASHTO T313-2)tests were used for determining the rheological and creepstiness properties of the blends in dierent temperatures.
Wheel tracking test was used to determine rutting resis-
rate in the bath. The specimen is broken when the stressdeveloped is greater than the strength of the specimen.The temperature and the strength at which the mixture isbroken are called fracture temperature and fracturestrength; respectively. The fracture strength measured inthe study is the fracture propagation study.
In this study, two processes are carried out for theTSRS. One is a direct single cooling test, called singleTSRS. The other is a repeated TSRS test. The initial tem-perature of the test is 10 C, and the cooling rate is 10 C/hfor all the tests. The repeated TSRS test was also carriedout between a temperature range of 0 C and 10 C for20 cycles, and then followed by the single TSRS test.
The test program in the study was shown in Fig. 1. Therst part is binder test owchart, and the second is mixturetest owchart.
3. Results and discussions
3.1. Performance-properties of the blends of aged binders
containing a rejuvenator
Table 3Rejuvenator percentage (%) to meet PG64-22 for Pen 30
SHRP tests required Percentage
DSR (original), T = 64 C, G*/sind > 1.00 kPa 2.20 kPa 0BBR(RTFO + PAV), T = 12 C, Stiness < 300 MPa >0BBR (RTFO + PAV)@ T = 12 C, m > 0.30 >2.0Common region of the rejuvenator percentage 2.07.4Average value of the rejuvenator percentage 4.7
960 J. Shen et al. / Construction and Building Materials 21 (2007) 958964tance properties of mixture samples under high tempera-tures. In this method, a solid wheel moves back and forthin the middle of the sample compacted by a roller compac-tor. Dynamic stability (DS), dened as the passes to get1 mm rutting, is calculated from the test result.
Thermal stress restrained specimen (TSRS) was used toevaluate the fracture properties of the mixtures under lowtemperature. The specimen of 5 cm 5 cm 30 cm wasmade by cutting a compacted mixture slab, and was furtherxed in a steel test frame at its two ends and then put into acooling bath. Thermal stress in the mixture specimen isdeveloped with the decreasing of temperature at a cooling
DSR on Original, 64C DSR on RTFO, 64C D
Hot MiVirgin asphalt binder PG64-22 Re
Same as rejuvenated asphalt 0.0 wt%
Same as 2.0 wt
Fig. 1. Test program for: (a) blend of aged bindeFigs. 2 and 3 show the relationship between ruttingresistance parameter, G*/sin(d), obtained using DSR test-ing procedures at 64 C, and the percentage of the rejuve-nator on original binders and RTFO residual;respectively. The G*/sin(d) of the blend, both in originalstate and RTFO residual, decreased linearly with theincrease of the rejuvenator. As we expected, this fact indi-cated that rejuvenator can soften the hard aged asphaltbinder eciently. It should be noted that an over dose con-centration of the rejuvenator will increase the ruttingpotential of the blend. Considering this fact, a maximum
9, 14 wt%
on PAV 25C BBR on PAV -12C
added (i.e., 0%), the parameter of G*sin(d) is still less thanthe standard value.
Figs. 5 and 6 show the relationship between low temper-ature properties parameters, stiness and m-value of theblend in RTFO + PAV residual, and the percentage ofthe rejuvenator under BBR test temperatures of 15 and10 C. The stiness decreased with the increase of therejuvenator linearly, whereas the m-value increases withthe increase of the rejuvenator. Both of the trends indicatedthat rejuvenator can soften the hard aged asphalt binder toa great extent, leading to the improvement of the shrinkageproperties under low temperatures. In order to get a targetPG64-22, the low temperature grade must be 22 C, thestiness of the blend under test temperature 12 C shouldbe less than 300 MPa, and the m-value larger than 0.3according to SHRP binder specications. The minimumpercentages of the rejuvenator to satisfy the stiness andm-value requirements are 0% and 2.0%; respectively, whichare obtained by interpolation.
The required rejuvenator percentages needed for theaged binder with Pen 30 to reach PG64-22 can be summa-rized as shown in Table 3.
3.2. Rutting resistance wheel tracking test
The rutting potential of an asphalt paving mixture under
0 5 10 15
Original binder, DSR test
Fig. 5. Stiness versus rejuvenator percentage of the blends at 12 C(BBR, RTFO + PAV residual) (interpolated using temperatures 15 and10 C).
0 5 10 15Rejuvenator content (%)
-15 C -10 C2.0%
RTFO+PAV, residue, BBR
Fig. 6. m-Value versus rejuvenator percentage of the blends at 12 C
(BBR, RTFO + PAV residual) (interpolated using temperatures 15 and10 C).
ture. The DS is dened as the number of passes requiredfor the loaded wheel to produce a vertical plastic deforma-tion of 1 mm on the surface of the test sample. The resultsof the wheel tracking test are listed in Table 4.
Fig. 7 shows the relationship between the DS of the pav-ing mixtures and the rejuvenator percentage in the binder.The DS of the mixtures decreased as the amount of the
the thermal stress restrained specimen (TSRS) test has beenused to characterize the properties at low temperature forsome years [7,8,10].
The results from single TSRS tests of both the recycledand the virgin mixtures are shown in Fig. 8. Generally, thecurves showing the relationships between the stress and thecooling temperature have a similar tendency for all the
rejuvenated binder rejuvenator (%) Code (JRA)
1.26 1.332100 1310 10001500
y = -171.68x + 2551.3R2 = 0.9788
0 2 4 6 8Rejuvenator content (%)
Fig. 7. Dynamic stability (DS) as a function of rejuvenator percentage inrejuvenated mixtures.
0.0-40 -30 -20 -10 0 10 20
Cooling temperature (oC)
Fig. 8. Tensile stress developed as a function of cooling temperature.
962 J. Shen et al. / Construction and Building Materials 21 (2007) 958964rejuvenator percentage increased, which is similar to thending for the blends, namely, the parameters, G*/sin(d),of rejuvenated aged binders decreased as the rejuvenatorpercentage increased.
The DS of the mixtures containing rejuvenated agedbinders decreased signicantly from 2630 to 1310 passes/m...