Construction and Building Materials 53 (2014) 596–603

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Impacts of rejuvenators on performance and engineering propertiesof asphalt mixtures containing recycled materials

0950-0618/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.conbuildmat.2013.12.025

⇑ Corresponding author. Tel.: +1 979 845 7799.E-mail addresses: s-im@tamu.edu (S. Im), f-zhou@ttimail.tamu.edu (F. Zhou),

Robert.lee@txdot.gov (R. Lee), t-scullion@tamu.edu (T. Scullion).1 Tel./fax: +1 512 506 5938.

Soohyok Im a,⇑, Fujie Zhou b, Robert Lee c,1, Tom Scullion b

a Texas A&M Transportation Institute, Flexible Pavements – CE/TTI, Room 309G, 3135 TAMU, College Station, TX 77843-3135, United Statesb Texas A&M Transportation Institute, College Station, TX 77843-3135, United Statesc Texas Department of Transportation, Austin, TX 21915, United States

h i g h l i g h t s

� Performance comparison of asphalt mixtures by performing various laboratory tests.� Simple cost analysis to investigate the cost benefits of using rejuvenators.� Cost effective way to enhance the overall performance of asphalt mixtures containing recycled materials.

a r t i c l e i n f o

Article history:Received 18 November 2013Received in revised form 2 December 2013Accepted 6 December 2013Available online 8 January 2014

Keywords:RejuvenatorAsphalt mixtureReclaimed asphalt pavement (RAP)Recycled asphalt shingles (RAS)

a b s t r a c t

Recycled materials, such as recycled asphalt shingles (RAS) and reclaimed asphalt pavement (RAP), havebeen widely used in asphalt paving industry, and the trend seems to use more and more, which can savetaxpayer dollars, preserve energy and protect the environment. However, these recycled materials areoften highly aged and cause potential durability issue for asphaltic layers. To balance out the impact ofstiff binder of recycled materials, rejuvenators have been recently evaluated. This study investigatedthe impacts of various rejuvenators on the performance and engineering properties of hot-mix asphalt(HMA) mixtures containing recycling materials (i.e., RAP and RAS). Various laboratory tests, includingHamburg test, overlay test, dynamic modulus test, and repeated load test, were performed to comparethe performance and engineering properties of HMA mixtures without rejuvenators to those of mixturesincorporated with rejuvenators. In addition, a simple cost analysis was performed to investigate the costbenefits of using rejuvenators. The laboratory test results and the cost analysis were presented and dis-cussed in this paper.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

In the last several years, both reclaimed asphalt pavement (RAP)and recycled asphalt shingles (RAS) have been widely used in as-phalt mixtures and the trend seems to use more and more becausethey can significantly reduce the cost of asphalt mixtures, conserveenergy, and protect the environment. However, the use of moreRAP and RAS often makes asphalt mixtures too stiff, and conse-quently less workable and difficult to compact in the field, whichmay ultimately lead to premature field failure [1]. In attempt to re-duce the stiffness of RAP and/or RAS mixtures, one option is to userejuvenators. Recently, rejuvenating agents have been receivingattention from the pavement research community because theycan improve the engineering properties of asphalt mixtures

containing high content recycled materials. Generally, rejuvenatoris a kind of asphalt additives to soften the stiffness of the oxidizedasphalt mixtures. Typically, rejuvenators contain a high proportionof maltenes constituents that help re-balance the composition ofthe aged binders that lost its maltenes during construction andfield service [2]. According to Carpenter and Wolosick [3], theworking mechanism (or diffusion process) of a rejuvenator consistsof the following four steps:

I. The rejuvenator forms a very low viscosity layer that sur-rounds the asphalt-coated aggregate which is highly agedbinder layer.

II. The rejuvenator begins to penetrate into the aged binderlayer, decreasing the amount of raw rejuvenator that coatsthe particles and softening the aged binder.

III. No raw rejuvenator remains, and the penetration continues,decreasing the viscosity of the inner layer and graduallyincreasing the viscosity of the outer layer.

Control Mixtures 1. RAS 5%2. RAP 13% and RAS 5%3. RAP 19%

Virgin BinderPG 64-22

Control143/121 °C (2hr)

R1to Virgin Binder

R2to Virgin Binder

Hamburg TestMixture Tests

Overlay Test

Dynamic Modulus Test

Repeated Load Test

R3to RAS

Fig. 1. Experimental test plan.

Table 1Gradation of aggregates and asphalt contents used in this study.

Combination of Materials Sieve analysis (sieve number and size in mm)

Aggregate sources % 19 (mm) 9.5 (mm) #4 (4.75) #8 (2.36) #30 (0.60) #50 (0.30) #200 (0.075)

5% RAS MixtureLimestone (type C) 26 100 56.4 10.9 4.7 3.3 2.6 2.2Limestone (type D) 19 100 70.7 14.3 6.3 3.7 3.3 2.7Limestone (type F) 21 100 100 76.4 20.6 6.2 4.8 3.9Manufactured sand 22 100 100 99.9 89.8 40.3 24.2 7.6Field sand 7.8 100 100 99.8 98.1 90.5 66.9 3.7RAS 5 100 100 99.7 98.9 62.8 53.7 23.4Combined gradation 100 100 83.1 55.5 38.3 21.4 15.1 4.8PG 64-22 5.2 (%)

13% RAP plus 5% RAS Mixture12.7 (mm) 9.5 (mm)

Limestone (type D) 51 100 96.7 39.1 8.6 3.5 3.0 2.6Screenings 25 100 100 99.0 78.5 27.5 16.0 4.3Sand 6.7 100 100 100 99.7 99.3 85.3 8.7RAP 13 100 98.7 69.3 41.0 27.2 20.9 7.0RAS 5 100 100 100 100 67.5 51.1 14.5Combined gradation 100 100 98.1 64.7 40.3 21.8 16.2 4.5PG 64-22 5.1 (%)

19% RAP MixtureLimestone (type D) 46 100 99.2 40.1 10.1 6.2 3.1 1.5Manufactured sand 29 100 100 99.3 83.6 39.1 19.9 3.0Field sand 6 100 100 100 99.0 96.0 73.0 3.0RAP 19 100 96.5 66.3 43.8 27.7 22.8 7.0Combined gradation 100 100 99.0 65.8 43.2 25.2 15.9 3.1PG 64-22 4.8 (%)

S. Im et al. / Construction and Building Materials 53 (2014) 596–603 597

IV. After a certain time, equilibrium is approached over themajority of the recycled binder film.

Recent studies on evaluating the effect of rejuvenators on engi-neering and performance properties of mixtures and/or binderscan be also found in the literature [4–7]. Shen et al. [8] investigatedthe effects of a rejuvenator on properties of rejuvenated asphaltbinders and mixtures by adding varying dosages. They found thatthe rejuvenator percentage significantly affected the properties ofboth rejuvenated aged binders and the mixtures. They also notedthat the optimum percentages of the rejuvenator could be obtainedby satisfying SHRP specifications (strategic highway research

program) through the blending charts. Similar studies conductedby Booshehrian et al. [9] and Tran et al. [10] reported that rejuve-nators mitigated the stiffness of the resultant binder and improvedthe cracking resistance of the mixtures. However, most of the stud-ies focused on one specific rejuvenator. Indeed, there are differenttypes of rejuvenators available in the market. Therefore, it is neces-sary to evaluate them and compare the cost-benefit of theserejuvenators.

This study evaluated the impacts of three commercial rejuvena-tors on performance and engineering properties of mixtures con-taining recycled materials (i.e., RAP and RAS) in terms ofmoisture resistance, cracking resistance, dynamic modulus, and

Table 2Testing matrix for each mixture.

Mix type Tests Control mix R1 R2 R3

5% RAS Hamburg wheel tracking test 2 2 2 2Overlay test 5 5 5 5Dynamic modulus test 2 2 2 2Repeated load test 2 2 2 2

13% RAP plus 5% RAS Hamburg wheel tracking test 2 2 2 2Overlay test 5 5 5 5Dynamic modulus test 2 2 2 2Repeated load test 2 2 2 2

19% RAP Hamburg wheel tracking test 2 2 2 2Overlay test 5 5 5 5Dynamic modulus test 2 2 2 2Repeated load test 2 2 2 2

Table 3Hamburg testing results (mm).

Mix type Passes Control mix R1 R2 R3

5% RAS 5000 3.40 5.08 2.59 3.4810,000 6.23 9.73 3.14 4.2915,000 12.33 12.82 (11,700) 3.56 5.4220,000 N/A N/A 4.02 9.88

13% RAP plus 5% RAS 5000 1.97 5.12 3.81 3.0710,000 11.22 12.65 (8000) 12.71 (9150) 11.9115,000 12.60 (10,950) N/A N/A 12.61 (10,350)20,000 N/A N/A N/A N/A

19% RAP 5000 12.55 (4800) 9.54 6.05 12.54 (3800)10,000 N/A 12.71 (5700) 13.00 (8550) N/A15,000 N/A N/A N/A N/A20,000 N/A N/A N/A N/A

Number in parenthesis indicates the actual failure passes.

598 S. Im et al. / Construction and Building Materials 53 (2014) 596–603

rutting resistance. Additionally, a simple cost analysis was per-formed. At the end of this paper, a summary and conclusion ispresented.

2. Study objectives

The primary objective of this study is to evaluate the effect ofthree rejuvenators on engineering properties of asphalt mixturescontaining RAP/RAS in terms of dynamic modulus, moisture dam-age, rutting resistance, and cracking resistance.

3. Experimental test plan

In order to achieve the above objective, a lab test plan wasdeveloped, as shown in Fig. 1. Three different control mixtures con-taining RAS only, RAP and RAS, and RAP only, respectively, wereproduced to compare the mixture performance and engineeringproperties to those of mixtures incorporated with three differentcommercial rejuvenators (i.e., designated as R1–R3 in this paper).

4. Material selection

In this study, local aggregates used at three different field pro-jects were collected to produce asphalt mixtures in the laboratory.Table 1 illustrates gradation of the aggregates and asphalt contentsused in the mix design for each field project. As shown in Table 1,each mixture contains different contents of RAP and/or RAS (i.e., 5%RAS, 13% RAP plus 5% RAS, and 19% RAP, respectively). Note thatthese three mixtures were designed following the Texas Depart-ment of Transportation (TxDOT)’s standard mix design procedure:Tex-204-F [11]. A PG 64-22 asphalt binder was used to produceeach mixture, and binder contents for different mixtures, 5.2%,

5.1%, and 4.8%, were determined as appropriate values that satisfyall key volumetric characteristics of asphalt mixtures.

As mentioned earlier, three different commercial rejuvenatorswere chosen in this study. The dosage of each rejuvenator was rec-ommended by each specific manufacturer. Two of them (R1 andR2) were directly added to the virgin binder, while the other one(R3) was blended into the recycled materials. The following is ashort summary of the information on the rejuvenators used in thisstudy:

� R1: directly added to virgin binder (0.6% of total asphaltbinder by weight).

� R2: directly added to virgin binder (1.5% of total asphaltbinder by weight).

� R3: dry recycled materials were mixed with a moisturecontent of 1% by weight and then 2% agent (by dry weight)heated at 65 �C directly was blended with wet recycledmaterials.

5. Laboratory tests, results, and discussion

As mentioned earlier, various laboratory tests were performedto evaluate the impact of rejuvenators on engineering propertiesof RAP/RAS mixtures. A total of 132 specimens were prepared tocomplete multiple testing for the whole study. Table 2 shows thematrix for the whole laboratory tests performed in this study. Norejuvenator was added to the control mixtures. Following TxDOT’sspecification, all HMA mixtures were mixed at 143 �C, cured for 2 hat 121 �C, and then compacted at 121 �C.

5.1. Hamburg wheel tracking test and associated results

Hamburg Wheel Tracking Test (HWTT) was conducted at atemperature of 50 �C in accordance with TEX-242-F [12], Test

(a) 5% RAS Control (b) R1 (c) R2 (d) R3

(e) 13%RAP/5% RAS Control

(f) R1 (g) R2 (h) R3

(i) 19%RAP Control (j) R1 (k) R2 (l) R3

Image not available

Fig. 2. Pictures of each mixture after HWTT.

S. Im et al. / Construction and Building Materials 53 (2014) 596–603 599

Procedure for Hamburg Wheel-Tracking Test (HWTT). A Superpavegyratory compactor was used to mold cylindrical specimens witha diameter of 150 mm and a height of 62 mm. A masonry sawwas used to cut along the edge of the cylindrical specimens.The target air void of specimens was 7 ± 1%. To evaluate the rut-ting susceptibility and moisture resistance, specimens were sub-merged under water at a temperature of 50 �C during the test,and a linear variable differential transducer (LVDT) device mea-sured deformations of specimens. The stop criterion was rutdepth of 12.5 mm or 20,000 passes.

Table 3 summarizes the rut depths from each test at differentpasses. Fig. 2 shows images of specimens after testing. The obser-vations are discussed below:

� For the 5% RAS mixtures, as shown in Table 3 and Fig. 2, R2and R3 rejuvenating agents significantly improved therutting/moisture resistance of the mixtures, while theperformance of the mixture with R1 rejuvenating agentwas similar to the 5% RAS control mixture.

� For the 13% RAP/5% RAS mixtures, all rejuvenated mixtureswere not better than the control mixture.

� For the 19% RAP mixtures, R1 and R2 rejuvenating agentsimproved HWTT results but not significantly.

In summary, based on the test results obtained from this study,the incorporation of rejuvenators with RAP and/or RAS improvedthe rutting resistance and moisture susceptibility of HMA mix-tures, although a clear trend from all three mixtures was notobserved.

5.2. Overlay test and associated results

The overlay test was used to evaluate the cracking resistance ofthe asphalt mixtures. This test procedure is described in TEX-248-F[13], Test Procedure for Overlay Test (OT). Five trimmed specimensfrom each mixture targeting air void of 7 ± 1% were prepared. Be-fore testing, individual OT specimens were conditioned in an envi-ronmental chamber with a target temperature of 25 �C. The slidingblock applied tension in a cyclic triangular waveform to a constantmaximum displacement of 0.06 cm. The sliding block reached themaximum displacement and then returned to its initial positionin 10 s. The time, displacement, and load corresponding to a certainnumber of loading cycles were recorded during the tests. The num-ber of cycles to failure is reported at the end of the test. The largerthe OT cycles, the better cracking resistance is.

The average OT cycles of the five specimens from eachmixture tested is presented in Fig. 3. For the 5% RAS mixtures, R1

0

5

10

15

20

25

30

OT

Cyc

les

to F

ailu

re

5% RAS ControlMixture

5% RAS Mixturew / R1

5% RAS Mixturew / R2

5% RAS Mixturew / R3

0

5

10

15

20

25

30

OT

Cyc

les

to F

ailu

re

13% RAP/ 5% RASControlMixture

13% RAP/ 5% RAS

w / R1

13% RAP/ 5% RAS

w / R2

13% RAP/ 5% RAS

w / R3

0

20

40

60

80

100

OT

Cyc

les

to F

ailu

re

19% RAPControlMixture

19% RAPMixturew / R1

19% RAPMixturew / R2

19% RAPMixturew / R3

Fig. 3. Overlay test results for each asphalt type.

(a) 5% RAS mixtures

(b) 13% RAP / 5% RAS mixtures

(c) 19% RAP mixtures

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

Dyn

amic

Mod

ulus

(MPa

)

Reduced Frequency (Hz)

5 % RAS Control

w / R1

w / R2

w / R3

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

Dyn

amic

Mod

ulus

(MPa

)

Reduced Frequency (Hz)

13% RAP+5% RASControl

w / R1

w / R2

w / R3

1.0E+01

1.0E+02

1.0E+03

1.0E+04

1.0E+05

1.0E-08 1.0E-05 1.0E-02 1.0E+01 1.0E+04 1.0E+07

1.0E-08 1.0E-05 1.0E-02 1.0E+01 1.0E+04 1.0E+07

1.0E-08 1.0E-05 1.0E-02 1.0E+01 1.0E+04 1.0E+07

Dyn

amic

Mod

ulus

(MPa

)

Reduced Frequency (Hz)

19 % RAP Control

w / R1

w / R2

w / R3

Fig. 4. Dynamic modulus test results for each asphalt type.

600 S. Im et al. / Construction and Building Materials 53 (2014) 596–603

rejuvenating agent exhibited the best performance, followed bythe R2 rejuvenating, and then R3 rejuvenating agent. The controlmixture showed the lowest value of cracking life. Similar resultswere observed for both the 13% RAP/5% mixtures and the 19%RAP mixtures, but the performance ranking of the rejuvenatorswere changed as shown in Fig. 3. In summary, all the mixtures withrejuvenators exhibited higher OT cycles (approximately from 110%to 300% improvements) than the control mixtures. This observa-tion implies that the rejuvenators reduced the stiffness of the agedbinder from the recycled materials so that they improved crackingresistant of mixtures. Similar test results can be found in the studyconducted by Booshehrian et al. [9].

5.3. Dynamic modulus test and associated results

The dynamic modulus test was conducted to measure changesof the viscoelastic stiffness of the asphalt mixtures due to theincorporation of rejuvenators. The test was conducted followingthe standard, AASHTO TP79-11 [14], Determining the Dynamic Mod-ulus and Flow Number for Hot Mix Asphalt (HMA) Using the AsphaltMixture Performance Tester (AMPT). Again, the Superpave gyratorycompactor was used to mold cylindrical samples with a diameterof 150 mm and a height of 170 mm. The samples were cored and

Table 4Temperature shift factors of each test case.

Mix type Log aT Control mix R1 R2 R3

5% RAS Log a(4) 2.152 2.224 1.922 2.092Log a(20) 0 0 0 0Log a(40) �2.213 �2.182 �2.122 �2.103

13% RAP plus 5% RAS Log a(4) 2.190 2.254 2.225 2.279Log a(20) 0 0 0 0Log a(40) �2.217 �2.111 �2.295 �2.071

19% RAP Log a(4) 2.144 2.010 2.134 2.087Log a(20) 0 0 0 0Log a(40) �2.303 �2.513 �2.227 �2.141

S. Im et al. / Construction and Building Materials 53 (2014) 596–603 601

then cut to produce cylindrical specimens with a diameter of100 mm and a height of 150 mm. The target air void of the coredand cut specimens was 7 ± 1%. To measure the axial displacementof the testing specimens, mounting studs were glued to the surfaceof the specimens so that three linear variable differential trans-formers (LVDTs) could be installed on the surface of the specimensthrough the studs at 120� radial intervals with a 70 mm gaugelength. Three temperatures of 4, 20, and 40 �C and six and/or sevenloading frequencies of 25, 10, 5, 1, 0.5, and 0.1 Hz, and 0.01 Hz(40 �C only) were used, and the frequency-temperature superposi-tion concept was applied to obtain the linear viscoelastic mastercurves at a target reference temperature of 20 �C. Two replicateswere tested and average values of dynamic modulus at each differ-ent testing temperature over the range of loading frequencies wereobtained.

Fig. 4 presents the master curves at the reference temperatureof 20 �C. Table 4 summarizes the shift factors obtained from eachmaster curve. For all mixtures tested, regardless of the rejuvenatortype, it is interesting to note that there were no significant differ-ence between the control mixtures and their counterparts for mostof the master curves (4 �C and 20 �C zone). However, most counter-parts showed the decreased stiffness characteristics at the low fre-quency (or high temperature 40 �C zone) except the two cases: the13% RAP/5% RAS mixture with R1 and R3 rejuvenating agents. Forexample, the 19% RAP control mixture exhibited the highest dy-namic modulus values, R1 mixture was second, and followed bythe R3 and R2 mixtures, respectively.

Based on the dynamic modulus test results obtained from thisstudy, it can be concluded that rejuvenators may affect the stiff-ness characteristic of mixtures only over lower loading frequencylevels (or hot temperature ranges).

5.4. Repeated load test and associated results

The unconfined, repeated load test was performed under a devi-ator stress level of 138 kPa at 40 �C. A loading stress level of138 kPa was selected based on the studies performed by Zhouet al. [15,16]. The 138 kPa loading stress is applied in the form ofa haversine curve with a loading time of 0.1 s with a rest periodof 0.9 s in one cycle. Loading stress is repeatedly applied on thespecimens until it exhibits a tertiary flow and reaches 5% perma-nent strain level or the number of loading cycles reaches 10,000.Two replicates from each mixture were prepared as with the dy-namic modulus test specimens.

Fig. 5 presents plots of the measured accumulative permanentstrain against the number of loading cycles. The result of the re-peated load test is strongly related to the stiffness of that mixture,which is the dynamic modulus; the stiffer mixture, the more rutresistance and less accumulative permanent strain. For example,the ranking of the rut resistance of mixtures is very similar tothe dynamic modulus test results of the same mixtures as

presented in the later section of this paper. Some mixtures withrejuvenators (i.e., 5% RAS mixture-R2 and -R3, 13% RAP/5% RASmixture-R3) exhibited similar rut resistance characteristics, whilethe others showed less or better rut resistance characteristics com-pared to their control mixtures.

5.5. Discussion of test results

Table 5 summarizes the performance ranking of the mixturesfrom each test. A lower numerical number in Table 5 is indicativeof the better performance result from the HWTT, OT, dynamicmodulus test, and repeated load test. In addition, the lower numer-ical number used for the dynamic modulus test indicates the rank-ing of the mixtures’ stiffness. The results shown in the tableindicate that the rejuvenators improve the mixture performancein terms of rutting resistance and cracking resistance. However,it should be noted that the ranking of the rejuvenators, based onthe test results obtained from this study, depends on the engineer-ing properties and mixture types.

6. Cost analysis

A simple cost analysis was conducted to investigate the costbenefits of HMA mixtures containing recycled materials incorpo-rated with and without rejuvenators. According to Copeland [17],there are four cost categories for asphalt production: Material,Plant production, Trucking, and Lay down. Among them, the mostexpensive production cost category is materials, comprising 70% ofthe cost to produce HMA. In this paper, the cost related only tomaterials was considered, including asphalt binder, rejuvenator,recycled materials, and virgin aggregates. The cost of each materialwas simply assumed based on the literature and the actual cost ofthe rejuvenator (R1) was used for the calculation. Based on the mixdesign information on the 19% RAP control mixture, the cost toproduce a HMA batch of 1000 kg (1 ton) was calculated as followsand presented in Table 6:

� Virgin aggregates = 0.015 ($/kg)� Virgin binder = 0.685 ($/kg)� RAP = 0.005 ($/kg)� R1 = 1.67 ($/kg)

As shown in Table 6, using RAP can significantly reduce the costof asphalt mixtures up to 18% (comparison of case 1 and case 2) asexpected. Even though the price of the rejuvenator (R1) is veryhigh compared to other materials, the total cost associated withthe rejuvenator (case 3) is actually not even higher than that ofthe case 2. As shown in the table, it is because the same amountof virgin binder is back out as the rejuvenator is added, since it be-comes the malthenes phase of the binder and remains there assuch as a Rejuvenator. The cost analysis results presented in the

(a) 5% RAS mixtures

(b) 13% RAP / 5% RAS mixtures

(c) 19% RAP mixtures

0

500

1000

1500

2000

2500

3000

3500

4000

Accu

mul

ated

Mic

ro-s

train

Cycle Numbers

5 % RAS Control

w / R1

w / R2

w / R3

0

500

1000

1500

2000

2500

3000

3500

4000

Accu

mul

ated

Mic

ro-s

train

13% RAP+5% RAS Control

w / R1

w / R2

w / R3

0

1000

2000

3000

4000

5000

6000

7000

0 2000 4000 6000 8000 10000 12000

Accu

mul

ated

Mic

ro-s

train

19 % RAP Control

w / R1

w / R2

w / R3

Cycle Numbers0 2000 4000 6000 8000 10000 12000

Cycle Numbers0 2000 4000 6000 8000 10000 12000

Fig. 5. Repeated load test results for each asphalt type.

Table 5Summary of performance ranking.

Mix type Tests Control mix R1 R2 R3

5% RAS HWTT 3 4 1 2OT 4 1 2 3Dynamic modulus 2 1 3 4Repeated load 3 1 2 4

13% RAP plus 5% RAS HWTT 1 4 3 2OT 4 1 3 2Dynamic modulus 3 2 4 1Repeated load 1 3 4 2

19% RAP HWTT 3 2 1 4OT 4 3 2 1Dynamic modulus 1 2 4 3Repeated load 3 1 4 2

Table 6Example of cost saving using rejuvenator.

Case 1. Virgin aggregates and virgin binderTotal weight of asphalt mixture 1000 (kg)4.8% asphalt by total weight (virgin binder) 48 (kg)Total weight of aggregate (virgin aggregates) 952 (kg)Total cost = (952 � 0.015) + (48 � 0.685) 47.2 ($/ton)

Case 2. Virgin aggregates plus19% RAP and virgin binder

Total weight of asphalt mixture 1000 (kg)19% RAP 190 (kg)Asphalt content of RAP (5%) = (190 � 0.05) 9.5 (kg)4.8% Asphalt by total weight (virgin binder) 48 (kg)Actual virgin binder needed = (48–9.5) 38.5 (kg)Total weight of aggregate (virgin aggregates) 771.5 (kg)Total cost = (771.5 � 0.015) + (190 � 0.005)

+ (38.5 � 0.685)38.9 ($/ton)

Case 3. Virgin aggregates plus 19%RAP and virgin binder plus 0.6% R1

Total weight of asphalt mixture 1000 (kg)19% RAP 190 (kg)Asphalt content of RAP (5%) = (190 � 0.05) 9.5 (kg)4.8% Asphalt by total weight (virgin binder) 48 (kg)0.6% R1 on total binder 0.288 (kg)Actual virgin binder needed = (48–9.5–0.288) 38.2 (kg)Total weight of aggregate (virgin aggregates) 771.5 (kg)Total cost = (771.5 � 0.015) + (190 � 0.005)

+ (38.2 � 0.685) + (0.288 � 1.67)39.2 ($/ton)

602 S. Im et al. / Construction and Building Materials 53 (2014) 596–603

table do not show significant cost saving using the rejuvenator.However, it can be concluded that the incorporation of rejuvena-tors is a cost effective way to enhance the overall performance of theasphalt mixtures as this paper presented through the entire sections.

7. Summary and conclusions

This study evaluated various rejuvenators and their influenceon the performance and engineering properties of HMA mixturescontaining recycled materials. Various laboratory tests were em-ployed to compare the performance and engineering properties

of control asphalt mixtures (without rejuvenators) with those ofasphalt mixtures incorporated with rejuvenators. In addition, asimple cost analysis was performed to investigate the cost benefitsof using rejuvenators when HMA mixtures are produced alongwith recycled materials. Based on the test results, the followingconclusions can be made.

� With respect to cracking resistance, all of the mixturesusing rejuvenators exhibited improved cracking resistance,compared to the control mixtures. This clearly indicatesthat rejuvenators can reduce the stiffness of the aged bin-der from the recycled materials.

� Similarly, the incorporation of rejuvenators improved themoisture susceptibility and rutting resistance of the mix-tures containing recycled materials.

� The performance ranking of the rejuvenators depends onmixture types and engineering properties evaluated. Twopossible reasons for this can be speculated. The first onemay be because of the degree of blending between thebinder of recycled materials and virgin binder. The otherreason may be due to different contents of recycled mate-rials, different source of the aggregates, and the rejuvenatordosage. This study used the dosage recommended by each

S. Im et al. / Construction and Building Materials 53 (2014) 596–603 603

manufacturer. As mentioned in the introduction, the reju-venator percentage significantly affects the properties ofthe mixtures [8]. Thus, it may be necessary to determinethe optimum dosage of each rejuvenator for each mixtureif it is desired to properly improve the performance of themixture.

� The simple cost analysis results showed that using rejuve-nators may be a cost effective way to enhance the overallperformance of the asphalt mixtures containing recycledmaterials.

The authors are currently working on binder tests to investigatethe impacts of rejuvenators on rheology and chemical properties ofblended binders which were extracted from the asphalt mixtures.Any significant findings relevant to this study will be presented.Specifically, field test sections with different types of rejuvenatorsshould be constructed for further evaluation.

8. Disclaimer

The contents and opinions of this paper reflect the views of theauthors, who are solely responsible for the facts and the accuracyof the data presented herein. The contents of this paper do not nec-essarily reflect the official views or the policies of any agencies.

Acknowledgment

The authors would like to thank the Texas Department of Trans-portation (TxDOT) for their financial support to complete thisstudy.

References

[1] Mogawer W, Bennert T, Daniel JS, Bonaquist R, Austerman A, Booshehrian A.Performance characteristics of plant-produced high RAP mixtures. J AssocAsphalt Paving Technol 2012;81:403–39.

[2] Terrel R, Epps J. Using additives and modifiers in hot-mix asphalt. Lanham,MD: Quality Improvement Series (QIP 114 A), NAPA; 1989.

[3] Carpenter S, Wolosick J. Modifier influence in the characterization of hot-mixrecycled material. Washington, DC: Transportation Research Record 777, TRB;1980. p. 15–22.

[4] Elseifi MA, Mohammad LN, Cooper III SB. Laboratory evaluation of asphaltmixtures containing sustainable technologies. J Assoc Asphalt Paving Technol2011;80:227–54.

[5] Zaumanis M, Mallick RB, Frank R. Evaluation of rejuvenator’s effectivenesswith conventional mix testing for 100% RAP mixtures. Trans Res Board CD-ROM 2013.

[6] Hajj EY, Souliman MI, Alavi MZ, Salazar LGL. Influence of hydrogreen bio-asphalt on viscoelastic properties of reclaimed asphalt mixtures. Trans ResBoard CD-ROM 2013.

[7] Hill B, Oldham D, Behnia B, Fini EH, Buttlar WG, Reis H. Low temperatureperformance characterization of bio-modified asphalt mixtures containingreclaimed asphalt pavement. Trans Res Board CD-ROM 2013.

[8] Shen J, Amirkhanian S, Tang B. Effects of rejuvenator on performance-basedproperties of rejuvenated asphalt binder and mixtures. Constr Build Mater2007;21:958–64.

[9] Booshehrian A, Mogawer WS, Vahidi S. Evaluating the effect of rejuvenators onthe degree of blending and performance of high RAP, RAS, RAP/RAS mixtures. JAssoc Asphalt Paving Technol 2013;82.

[10] Tran NH, Taylor A, Willis R. Effect of rejuvenator on performance properties ofhma mixtures with high RAP and RAS contents. Final Report, NCAT Report 12-05, 2012.

[11] TxDOT Designation: TEX-204-F. Design of bituminous mixtures. Constr Div,Texas Department Trans, 2008.

[12] TxDOT Designation: TEX-242-F. Hamburg wheel-tracking test. Constr Div,Texas Department Trans, 2009.

[13] TxDOT Designation: TEX-248-F. Overlay test. Constr Div, Texas DepartmentTrans, 2009.

[14] AASHTO TP 79-11. Determining the dynamic modulus and flow number forhot mix asphalt (HMA) using the asphalt mixture performance tester (AMPT).Am Assoc State Highway Transport Officials, 2011.

[15] Zhou F, Scullion T, Chen D. Laboratory characterization of asphalt mixes of SPS-1 sections on US281. Int J Pavement Mater Des 2002;3(4):439–53.

[16] Zhou F, Chen D, Scullion T, Bilyeu J. Case study: evaluation of laboratory testmethods to characterize permanent deformation properties of asphalt mixes.Int J Pavement Eng 2003;4(3):155–64.

[17] Copeland A. Reclaimed asphalt pavement in asphalt mixtures: state of thepractice. Final Report, No. FHWA-HRT-11-021, Office of InfrastructureResearch and Development Federal Highway Administration, 2011.