28.03.2013 Views

Rutting Prediction of Asphalt Mixtures from Asphalt Cement

Rutting Prediction of Asphalt Mixtures from Asphalt Cement

Rutting Prediction of Asphalt Mixtures from Asphalt Cement

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Transportation Research Institute<br />

Iran University <strong>of</strong> Science and Technology<br />

Ministry <strong>of</strong> Science, Research and Technology<br />

Transportation Research Journal 1 (2012) 25-36<br />

<strong>Rutting</strong> <strong>Prediction</strong> <strong>of</strong> <strong>Asphalt</strong> <strong>Mixtures</strong> <strong>from</strong> <strong>Asphalt</strong> <strong>Cement</strong><br />

Imran Hafeez a , Mumtaz Ahmad Kamal b , Mohammad Reza Ahadi c*<br />

a. Assistant Pr<strong>of</strong>essor, Department <strong>of</strong> Civil Engineering, University <strong>of</strong> Eng. & Tech. Taxila, Pakistan.<br />

b. Pr<strong>of</strong>essor, Department <strong>of</strong> Civil Engineering, University <strong>of</strong> Eng. & Tech. Taxila, Pakistan.<br />

c. Assistant Pr<strong>of</strong>essor, Transportation Research Institute, Iran university <strong>of</strong> Science and Technology, Tehran, Iran.<br />

Received: 6 June 2011 – Accepted: 16 October 2011<br />

ABSTRACT<br />

<strong>Asphalt</strong> cement is a visco-elastic material used for cementing materials in preparation <strong>of</strong> hot mix asphalt. Its<br />

performance in the field mainly depends on the size and proportions <strong>of</strong> aggregates <strong>of</strong> hot mix asphalt being used. The<br />

main objective <strong>of</strong> this study was to characterize asphalt cement and to investigate the influence <strong>of</strong> asphalt cement on<br />

the rutting <strong>of</strong> asphalt mixture. Two bituminous binders (PG76-22 and PG 58-16) and two aggregate gradations<br />

(coarse and fine) were selected to study their effect on the mixtures rutting behavior. A wheel tracking test and a<br />

dynamic modulus test were selected to determine the rutting potential <strong>of</strong> asphalt mixtures at different temperature<br />

levels. The study revealed that high performance grade bitumen with fine aggregate gradation may <strong>of</strong>fer more<br />

resistance to rutting than low grade bitumen with coarse aggregate gradation. The master curve technique can be used<br />

to predict asphalt mixtures’ rutting performance <strong>from</strong> asphalt binders. Complex shear modulus <strong>of</strong> asphalt binder and<br />

mixtures can fairly be correlated with each other.<br />

Keywords: Hot Mix <strong>Asphalt</strong>, Performance Testing, Dynamic Modulus, Wheel Tracker Machine, Master Curves<br />

1- Introduction<br />

Bitumen commonly known as asphalt cement<br />

(AC) is a viscoelastic material used for binding<br />

materials in road pavements (Soleimani, 2009). Its<br />

deformation and flow measurement in the<br />

laboratory, when subjected to stress, explains the<br />

elastic and viscous behavior. Complex modulus (G*)<br />

and phase angle (δ) are considered to be the<br />

principal rheological parameters, normally measured<br />

<strong>from</strong> a device known as a Dynamic Shear<br />

Rheometer (DSR) (Huang, Shin-Che et.al 2007).A<br />

number <strong>of</strong> studies had been conducted to investigate<br />

asphalt binder rheology (Tarefder Zaman 2003 &<br />

Kanitpong, Bahia, 2005).<br />

TRANSPORTATION<br />

RESEARCH<br />

JOURNAL<br />

www.trijournal.ir<br />

<strong>Asphalt</strong> concrete is a generally a mixture <strong>of</strong><br />

bitumen and aggregates used for the construction <strong>of</strong><br />

road pavements. An accumulation <strong>of</strong> small strain<br />

values under repeated cycles <strong>of</strong> loading is commonly<br />

known as rutting (Faheem and Bahia, 2004).<br />

Performance <strong>of</strong> asphalt mixtures in terms <strong>of</strong> linear<br />

viscoelastic behavior can be predicted using stressstrain<br />

behavior which can be defined by a dynamic<br />

or complex shear modulus test or wheel tracker<br />

machine (Loulizi, Flintsch, Al-Qadi and Mokarem,<br />

2006). Dynamic modulus (E*) is the ratio <strong>of</strong> the<br />

absolute value <strong>of</strong> the peak to peak shear stress by the<br />

absolute value <strong>of</strong> the peak to peak shear strain under<br />

Corresponding Author, Tel: +989121713297 Transportation Research Journal, Vol. 2, No. 1, 2012/ 25<br />

E-mail address: ahadi@rahiran.ir


sinusoidal loading conditions (Anderson and<br />

Christensen, 1992). The master curve enables<br />

fundamental characterization <strong>of</strong> asphalt concrete in<br />

which the time and temperature dependencies can be<br />

fully described. Various functional forms like the<br />

sigmoidal fitting function can be used to<br />

mathematically model the response <strong>of</strong> asphalt mixes<br />

(Pellinen and Witczak, 2002).<br />

(Colbert and Zhanping 2012) characterized the<br />

rheological properties <strong>of</strong> asphalt binders extracted<br />

<strong>from</strong> a recycled asphalt pavement (RAP) mixture<br />

and observed significant differences in dynamic<br />

shear moduli master curve performance for high<br />

percentage RAP binder blends versus virgin binders<br />

at the three aging states (Colbert and<br />

Zhanping,2012.). Nur et al. (2011). Reports based on<br />

existing literature stated that the use <strong>of</strong> reliable<br />

models can in general be considered as a valuable<br />

alternative tool for estimating the Linear<br />

viscoelastic rheological properties <strong>of</strong> bitumen.<br />

(NurIzzi, et al. (2011) recommended the significance<br />

<strong>of</strong> creep compliance on G*/sinδ in predicting the<br />

rutting behavior <strong>of</strong> asphalt mixtures. (Wasage et al,<br />

2011) Kumar et al (2011) studied the effect <strong>of</strong><br />

styrene butadiene styrene (SBS) polymer and crumb<br />

rubber modified asphalt binders on asphalt mixtures’<br />

dynamic mechanical behavior using dynamic<br />

modulus, dynamic and static creep tests at varying<br />

temperatures and frequency levels. It was revealed<br />

that the mechanical response <strong>of</strong> the SBS polymer<br />

modified asphalt binders were significantly<br />

correlated with the rutting resistance <strong>of</strong> asphalt<br />

concrete mixes (Kumar et al, 2011).<br />

The relationship between the dynamic modulus<br />

in compression |E*| <strong>of</strong> the asphalt mixer to the G*,<br />

and the complex shear modulus <strong>of</strong> the binder<br />

developed through engineering mechanics were<br />

generated by equation 1 (Charles et al, 2003);<br />

E 21µG (1)<br />

The effectiveness <strong>of</strong> binder’s stiffness can be<br />

used to predict the mixture’s stiffness and can be<br />

studied using a different mode <strong>of</strong> testing. Interaction<br />

<strong>of</strong> low and high performance grade bitumen with<br />

coarser and finer aggregate gradation in rut<br />

prediction would be an interesting phase. The wheel<br />

26/ Transportation Research Journal, Vol. 2, No. 1, 2012<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

tracking machine and dynamic modulus testing can<br />

be used to compare with the results <strong>of</strong> frequency<br />

sweep testing on asphalt binders.<br />

2- Objectives<br />

To characterize the asphalt mixtures using a high<br />

and a low grade asphalt binder and a coarse and a<br />

fine aggregate gradations.<br />

To find out the possible correlation between<br />

asphalt binders and asphalt mixtures using the<br />

master curve techniques.<br />

To develop any possible relationship between<br />

wheel tracker rutting with binder rut factor<br />

3- Experimental program<br />

The experimental program comprises <strong>of</strong> testing<br />

individual binders and aggregates followed by their<br />

combined mixture’s testing on two different testing<br />

protocols.<br />

3-1-Binders<br />

PG76-22 and PG58-16 were collected <strong>from</strong><br />

single source. These binders were used on highways<br />

and motorways which have been carrying<br />

approximately 75% <strong>of</strong> the total road freight in<br />

Pakistan. Rolling Thin Film Aging Oven aged<br />

specimens were used in binder testing (AASHTO T<br />

240). Sinusoidal, oscillatory stress over a range <strong>of</strong><br />

temperatures and loading frequencies were applied<br />

to a 25 mm diameter and 2mm thin disc <strong>of</strong> asphalt<br />

binder using Dynamic Shear Rheometer (AASHTO<br />

T 240, 2004). Dynamic Shear Rheometer was<br />

applied to determine the phase angle and Complex<br />

Shear Modulus at different temperatures and<br />

frequency ranges (AASHTO, 1995).<br />

Master curves were constructed and accordingly,<br />

shift factor and sigmoidal function were computed to<br />

check the accuracy <strong>of</strong> the data.<br />

3-2- <strong>Asphalt</strong> mixes<br />

Four asphalt mixtures were prepared using two<br />

binders and two aggregate gradations namely Class-<br />

A (Coarser) and Class-B (Finer) using the National<br />

Highway Authority <strong>from</strong> Pakistan. The<br />

specifications are shown in Figure 1 (National<br />

Highway Authority, 1998).


% Passing<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

0<br />

0.075<br />

Fineer<br />

Gradation<br />

3-2-1- Dynnamic<br />

moddulus<br />

testingg<br />

<strong>Mixtures</strong>s<br />

were testeed<br />

on repeatted<br />

load uniiaxial<br />

loading testing<br />

machhine<br />

commoonly<br />

knownn<br />

as<br />

dynamic mmodulus<br />

testing<br />

(Assocciation<br />

<strong>of</strong> State<br />

Highway aand<br />

Transpportation<br />

OOfficials<br />

(TPP-62)<br />

2004).<br />

A sinussoidal<br />

commpressive<br />

strress<br />

wave was<br />

applied to ttest<br />

specimeens<br />

at loadinng<br />

frequenciees<br />

<strong>of</strong><br />

25, 10, 5, 1,<br />

0.5, and 0. 1 Hz in a temmperature<br />

coontrol<br />

chamber (Figure<br />

2). Thhese<br />

frequenccies<br />

were buuilt<br />

in<br />

by default iin<br />

the s<strong>of</strong>twaare<br />

set up. TThe<br />

main purrpose<br />

for choosinng<br />

controlleed<br />

stress waas<br />

to get lower<br />

numbers <strong>of</strong>f<br />

cycles to failure at kknown<br />

stresss<br />

as<br />

compared too<br />

controlled strain test.<br />

<strong>Asphalt</strong>iic<br />

specimenss<br />

which are1180mm<br />

in height<br />

and 150 mmm<br />

in diammeter<br />

were prepared oon<br />

a<br />

superpave ggyratory<br />

commpactor<br />

at aaverage<br />

air vvoids<br />

<strong>of</strong> 7±0.2%. These speccimens<br />

were then cored to a<br />

150mm height<br />

by 100mmm<br />

diameter with averagge<br />

air<br />

voids <strong>of</strong> 3.5%.<br />

Three stress leveels<br />

at a uniiform<br />

interval withhin<br />

the rangees<br />

specified in AASHTOO<br />

TP-<br />

62, at each temperaturee<br />

were seleccted<br />

and repoorted<br />

in Table 1.<br />

Hafeez et al. / <strong>Rutting</strong> Produ uction <strong>of</strong> Asphallt<br />

…<br />

0.6 2.36 44.75<br />

9.5<br />

Sieve Size e, mm<br />

12.5 19.0<br />

CCoarser<br />

Gradatioon<br />

25.4 37.55<br />

Figure<br />

1. Aggrregate<br />

Grada ations for Aspphalt<br />

Mixturees<br />

Two parammeters,<br />

dynaamic<br />

modulu us E* and phhase<br />

angle a φ, weree<br />

obtained frrom<br />

dynamic c modulus teests.<br />

The T rutting ffactor,<br />

a parrameter<br />

to measure m ruttting<br />

characteristic<br />

c cs <strong>of</strong> E*/sinφφ<br />

for mixes, , at a particuular<br />

frequency f and<br />

temperaturre<br />

was also computed. c<br />

3-2-2- 3 Wheel<br />

Tracker Test<br />

Slab speciimens<br />

<strong>of</strong> a 400.3cm<br />

x 40.3 3cmx7.5cm ssize<br />

were w prepared<br />

on a rollerr<br />

compactor and tested oon<br />

a<br />

Wheel W Trackker<br />

(WT) at a 720 Newt ton wheel looad.<br />

The T lift thickknesses<br />

<strong>of</strong> slaab<br />

specimen ns were takenn<br />

as<br />

7.5cm 7 and 5.0<br />

cm to covver<br />

the Nom minal maximmum<br />

size s <strong>of</strong> 19mmm<br />

and 12.5mmm<br />

as per EN 1269-22, 20003.<br />

Specimens S wwere<br />

compaccted<br />

to targ get density aand<br />

were w tested uunder<br />

the WWheel<br />

Tracker<br />

in replicaates<br />

<strong>of</strong> o three at each temperature e level. TThe<br />

moving m tablee<br />

<strong>of</strong> the WT mmachine<br />

as shown s in Figgure<br />

3 reciprocates<br />

at a frequenncy<br />

<strong>of</strong> 53 pa asses per minnute<br />

and a the formmation<br />

<strong>of</strong> rut was measur red using linnear<br />

variable v ddifferential<br />

transducer rs (European<br />

Standard, S 20002).<br />

Transporrtation<br />

Researcch<br />

Journal, Vol.<br />

2, No. 1, 20122<br />

/27


Description<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

Table 1. Stress levels at each temperature in Dynamic Modulus Test<br />

Temperatures ( o C)<br />

25 40 55<br />

Stress Level (kPa) 700, 500 300 250, 200 150 70, 50, 30<br />

Figure 2. Sinusoidal loading pulse pattern in dynamic modulus testing (Ge<strong>of</strong>frey, Rowe and Sharrock, 2000)<br />

Figure 3. Wheel Tracker Machine (Cooper Technology, 2006)<br />

Corresponding Author, Tel: +989121713297 Transportation Research Journal, Vol. 2, No. 1, 2012/ 28<br />

E-mail 28/ Transportation address: ahadi@rahiran.ir<br />

Research Journal, Vol. 2, No. 1, 2012


Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

Table 2. Experimental Program<br />

Step No. Descriptions Details <strong>of</strong> activities in the experimental design<br />

1 Materials<br />

2<br />

Testing machine/<br />

Procedures<br />

3 Testing Conditions<br />

4 Output parameters<br />

5 Characterization<br />

6<br />

7<br />

Generating the<br />

Correlations<br />

Wheel Tracker<br />

Testing<br />

<strong>Asphalt</strong> Binders<br />

PG58-16<br />

(neat binder 60/70 pen. grade)<br />

PG 76-22<br />

(modified binder prepared<br />

using 60/70 pen. grade with<br />

1.65% Elvaloy)<br />

Frequency sweep test at Dynamic<br />

Shear Rheometer<br />

Frequency:100 to 1Hz<br />

Temperature ; 25, 40, 55 o C<br />

A summary <strong>of</strong> the experimental program has<br />

been reported in Table 2 in order to understand the<br />

volume <strong>of</strong> work and the testing methodology.<br />

4- Results and Discussions<br />

4-1-Test results<br />

Two binders and four mixes were studied under<br />

different times and temperature conditions. Master<br />

curves were developed to evaluate the relationship<br />

between the complex modulus and the phase angle<br />

at different frequency levels. Shift factor curves<br />

were also plotted to determine temperature<br />

sensitivity <strong>of</strong> binders and mixes. Results showed that<br />

the temperature and frequency <strong>of</strong> loading had a<br />

significant influence on the behavior <strong>of</strong> asphalt. The<br />

complex modulus decreases with an increase in test<br />

<strong>Asphalt</strong> <strong>Mixtures</strong><br />

PG76-22 +Class-A<br />

PG76-22 +Class-B<br />

PG58-16 +Class-A<br />

PG58-16 +Class-B<br />

i) Dynamic modulus testing at<br />

NU-14 (AASHTO TP 62)<br />

ii) Wheel Tracker Test (EN,<br />

2003)<br />

Frequency: 25 to 0.1Hz<br />

Temperature ; 25, 40, 55 o C<br />

53 Passes per minute on Wheel<br />

Tracker<br />

Dynamic modulus (E*)<br />

Phase angle (φ)<br />

Master curves development<br />

Shift factor computation,<br />

Sigmoidal parameters <strong>of</strong> best fit<br />

Complex shear modulus (G*)<br />

Phase angle (δ)<br />

Master curves development<br />

Shift factor computation,<br />

Sigmoidal parameters <strong>of</strong> best fit<br />

Comparison <strong>of</strong> G* and E* master curves on single plots and development <strong>of</strong><br />

relationships at highest degree <strong>of</strong> determinacy.<br />

Correlation between Log G* and E*.<br />

Rut development history <strong>of</strong> asphalt mixtures using wheel tracker machine.<br />

temperature under a specific loading frequency,<br />

whereas it increases with an increase in frequency at<br />

a specified temperature. This shows that the elastic<br />

portion <strong>of</strong> viscoelastic property decreases or is<br />

reduced over a range <strong>of</strong> temperatures <strong>from</strong> 25 o C to<br />

55 o C. The complex shear modulus and phase angle<br />

values <strong>of</strong> binders have been reported in Table 3. The<br />

phase angle values <strong>of</strong> the binders mostly ranged<br />

between 70 o and 87 o , whereas for mixes it ranged<br />

<strong>from</strong> 12 o to 33 o . These values <strong>of</strong> phase angles, for<br />

both asphalt binders and mixtures, confirm the<br />

results <strong>of</strong> previous research and are in accordance<br />

with their limits. Also, The phase angle <strong>of</strong> the above<br />

asphalt mixtures are three times less than the<br />

corresponding asphalt binders at any temperature<br />

and frequency level. The reduction is mainly<br />

contributed by the aggregates in the mixtures.<br />

Transportation Research Journal, Vol. 2, No. 1, 2012 /29


E*/Sinφ<br />

Temperature<br />

25 o C<br />

40 o C<br />

55 o C<br />

Temperature<br />

14000<br />

12000<br />

10000<br />

8000<br />

6000<br />

4000<br />

2000<br />

25 o C<br />

40 o C<br />

55 o C<br />

0<br />

30/ Transportation Research Journal, Vol. 2, No. 1, 2012<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

Table 3a. Complex shear modulus (G*) <strong>of</strong> asphalt binders<br />

Description<br />

<strong>Asphalt</strong><br />

Binders<br />

Description <strong>of</strong><br />

<strong>Mixtures</strong><br />

Class-A<br />

Class-B<br />

Class-A<br />

Class-B<br />

Class-A<br />

Class-B<br />

Class-A, PG76<br />

Class-A, PG58<br />

G*at<br />

0.1Hz(MPa)<br />

20.72<br />

5.65<br />

1.40<br />

PG76-22<br />

(PMB)<br />

Phase Angle (δ)<br />

72.44<br />

81.80<br />

87.41<br />

Table 3b. Dynamic modulus (E*) <strong>of</strong> asphalt mixtures<br />

E* at 0.1Hz<br />

(MPa)<br />

3897<br />

2978<br />

1652<br />

1245<br />

1157<br />

1007<br />

0 1 10 100 1000<br />

Log(G*/Sinδ)<br />

PG76-22<br />

(PMB)<br />

Phase Angle (φ)<br />

17.46<br />

21.09<br />

23.27<br />

27.96<br />

28.59<br />

32.54<br />

PG58-16<br />

(60-70Grade Bitumen)<br />

G*at 0.1Hz<br />

(MPa)<br />

12.479<br />

3.97<br />

1.00<br />

Phase Angle (δ)<br />

77.56<br />

82.18<br />

87.33<br />

PG58-16<br />

(60-70 Grade Bitumen)<br />

E* at 0.1Hz<br />

(MPa)<br />

2811<br />

1754<br />

956<br />

743<br />

453<br />

348<br />

Figure 4. Relationship between asphalt binder and mixture’s rut factors<br />

The elastic modulus (G*) <strong>of</strong> binder mostly range<br />

between 20 MPa and 1 MPa and that <strong>of</strong> mixtures<br />

between 400 and 4000MPa. A significant change<br />

was observed within the selected temperature range<br />

which depicts the changed behavior <strong>of</strong> asphalt mixes<br />

over the said range <strong>of</strong> temperatures in the field.<br />

Furthermore, rut depth factors for asphalt binder<br />

(G*/Sinδ) and mixture (E*/Sinφ) were computed<br />

E*/Sinφ<br />

9000<br />

8000<br />

7000<br />

6000<br />

5000<br />

4000<br />

3000<br />

2000<br />

1000<br />

0<br />

Class-B, PG76<br />

Class-B, PG58<br />

Phase Angle<br />

(φ)<br />

19.46<br />

22.34<br />

25.56<br />

30.02<br />

31.84<br />

33.57<br />

0 1 10 100 1000<br />

Log(G*/Sinδ)<br />

and compared to ascertain possible relationship.<br />

Logarithmic scale was used to cover the entire range<br />

<strong>of</strong> data set.<br />

Figure 4 reveals that prediction <strong>of</strong> asphalt<br />

mixture’s rut factor can be made <strong>from</strong> asphalt<br />

binder rut factor with a degree <strong>of</strong> determinacy <strong>of</strong><br />

over 0.8.


4-2- Development <strong>of</strong> Master Curve<br />

The time and temperature super positioning<br />

(TTS) method was adopted to develop master<br />

curves. This principle describes the same<br />

viscoelastic properties <strong>of</strong> asphalt binder and<br />

mixtures measured <strong>from</strong> different temperatures<br />

over the same period <strong>of</strong> time. Viscoelastic<br />

properties measured at 40 and 55 o C were shifted to<br />

25 o C by a horizontal shift (aT) and they developed<br />

Sigmoidal fit master curves which have been shown<br />

in Figures 5 and 6 for asphalt binders and mixtures<br />

respectively. These curves actually describe the<br />

time and temperature dependent shear properties <strong>of</strong><br />

Log (G*)<br />

Log (E*)<br />

4<br />

3<br />

2<br />

1<br />

0<br />

‐1<br />

‐2<br />

4<br />

3.8<br />

3.6<br />

3.4<br />

3.2<br />

3<br />

2.8<br />

2.6<br />

2.4<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

PG 58‐16<br />

Figure 5. Development <strong>of</strong> master curves for asphalt binders<br />

Figure 6. Development <strong>of</strong> master curves for asphalt mixtures<br />

asphalt binders. One can observe the effect<br />

<strong>of</strong> PG grading and aggregate gradation on the<br />

viscoelastic behavior <strong>of</strong> asphalt mixtures.<br />

<strong>Asphalt</strong> mixtures with PG 76 and coarser aggregate<br />

gradation showed higher E* values at all frequency<br />

levels and PG 58 with finer gradation showed<br />

minimum values at all levels <strong>of</strong> frequency and<br />

temperature.<br />

Also, <strong>Asphalt</strong> mixture with finer aggregate<br />

gradation and PG 76 showed relatively good results<br />

than PG58 and coarser gradation. This means that<br />

the effect <strong>of</strong> aggregate gradation can be improved by<br />

binder stiffness effects.<br />

‐2.0 ‐1.0 0.0 1.0 2.0 3.0<br />

Log (Tr)<br />

Class‐A, PG76<br />

Class‐A, PG58<br />

‐2.0 ‐1.0 0.0 1.0 2.0 3.0<br />

Log (Tr)<br />

Transportation Research Journal, Vol. 2, No. 1, 2012 /31


Master curves <strong>of</strong> asphalt binders (log G*) and<br />

mixtures (log E*) were compared on single space as<br />

presented in Figures 7 and 8. The purpose <strong>of</strong><br />

comparing both the materials’ curve (s) was to<br />

estimate the influence <strong>of</strong> the binder’s stiffness on the<br />

mixture’s stiffness.<br />

Log (Dynamic Modulus)<br />

Log (Dynamic Modulus)<br />

4.0<br />

3.0<br />

2.0<br />

32/ Transportation Research Journal, Vol. 2, No. 1, 2012<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

At low frequency the asphalt binder response<br />

becomes viscous and the slopes <strong>of</strong> the curve become<br />

steeper. One can conclude <strong>from</strong> these Figures that by<br />

using the PG 76 in asphalt mixtures the effect <strong>of</strong><br />

aggregate gradation is not distinguished when<br />

compared to PG 58.<br />

yClas‐A = ‐0.033x<br />

Figure 7. Comparison <strong>of</strong> G* and E* master curves using PG 76-22<br />

2 ‐ 0.182x + 3.686<br />

R² = 0.998<br />

yPG76= ‐0.153x2 yClass‐B = ‐0.011x<br />

‐ 0.746x + 1.978<br />

R² = 0.997<br />

2 1.0<br />

0.0<br />

‐1.0<br />

‐2.0<br />

‐ 0.192x + 3.551<br />

R² = 0.997<br />

Class‐A, PG 76<br />

PG 76‐22<br />

Class‐B, PG76<br />

‐3.0 ‐2.0 ‐1.0 0.0 1.0 2.0 3.0 4.0<br />

Log (Tr)<br />

4.0<br />

3.0<br />

2.0<br />

1.0<br />

0.0<br />

‐1.0<br />

‐2.0<br />

y Class‐A = ‐0.039x 2 ‐ 0.194x + 3.550<br />

R² = 0.999<br />

y Class‐B = ‐0.045x 2 ‐ 0.189x + 3.368<br />

R² = 0.999<br />

y PG58 = ‐0.081x 2 ‐ 0.956x + 1.286<br />

R² = 0.997<br />

‐3.0 ‐2.0 ‐1.0 0.0 1.0 2.0 3.0 4.0<br />

Log (Tr)<br />

Figure 8. Comparison <strong>of</strong> G* and E* master curves using PG58-16<br />

Class‐A, PG58<br />

PG 58‐16<br />

Class‐B, PG58


4-3- Sigmoidal parameters <strong>of</strong> Master Curves<br />

Sigmoidal parameters best describes the shape<br />

and the location <strong>of</strong> the master curve by<br />

mathematical modelling, using the following<br />

relationship (State Highway and Transportation<br />

Officials (TP-62) 2004).<br />

<br />

Log E<br />

<br />

1 e<br />

* <br />

(log Tr )<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

(2)<br />

Where δ, α are known to be the fitting parameters<br />

that depend on aggregate gradation, binder contents<br />

and air void, while β, γ depend on characteristics <strong>of</strong><br />

Sr. No.<br />

1<br />

Parameters<br />

Table 4. Sigmoidal parameters for binders and mixtures<br />

Binder<br />

PG76-22 (PMB)<br />

Mix. Class<br />

A<br />

asphalt binder and describes the shape <strong>of</strong><br />

the sigmoidal functions. Sigmoidal functions (δ, α,<br />

β, & γ) for binders and mixtures were determined<br />

simultaneously and reported in Table 4 for<br />

comparison purposes. Sigmoidal functions help to<br />

characterize the master curves in terms <strong>of</strong> shape and<br />

location. Values <strong>of</strong> sigmoidal functions at different<br />

test conditions as reported by different researcher in<br />

the past have been given in Table 5 for comparison.<br />

One can observe <strong>from</strong> this Table that values <strong>of</strong><br />

signmoidal function depend upon mix properties and<br />

testing features that mainly involve types <strong>of</strong> testing,<br />

material types and test conditions.<br />

Mix. Class<br />

B<br />

PG58-16 (60-70Grade Bitumen)<br />

Binder<br />

Mix. Class<br />

A<br />

Mix. Class<br />

B<br />

δ -2.55 2.36 2.50 -0.22 1.67 1.79<br />

α 9.11 1.62 1.54 6.21 1.96 2.08<br />

β -1.14 -1.58 -0.81 -1.70 -1.90 -1.73<br />

γ 0.58 0.79 0.62 0.89 0.80 0.71<br />

References <strong>of</strong><br />

previous studies<br />

Ge<strong>of</strong>frey et al., 2009<br />

2 Kim and Partl., 2009<br />

3 Amara et al., 2006<br />

Table 5. Sigmoidal functions reported in other studies<br />

Sigmoidal Functions<br />

Witczak parameters computed with RHEA S<strong>of</strong>tware: For mix at 3.8% air<br />

voids;<br />

δ = -0.954, α = 4.613, β = -1.64, γ = -0.428,<br />

Richard parameters for dynamic modulus master curve: For mix at 3.8% air<br />

voids;<br />

δ = 0.301, α = 4.094, β = -1.513, γ = -0.395,<br />

For Mastic asphalt tested under uniaxial compression test:<br />

δ = 1.350,α = 1.860, β = -2.820, γ = 0.680,<br />

For in place hot mix asphalt cores taken <strong>from</strong> 18 sites; Average values<br />

δ = 4.2837, α = 2.1672, β = -0.6373, γ = 0.5853,<br />

Transportation Research Journal, Vol. 2, No. 1, 2012 /33


Rut Depth (mm)<br />

34/ Transportation Research Journal, Vol. 2, No. 1, 2012<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

Figure 9. Typical Rut depth formation <strong>of</strong> <strong>Asphalt</strong> <strong>Mixtures</strong> at 55oC<br />

Table 6. Summary <strong>of</strong> Rut Depth <strong>of</strong> different asphalt <strong>Mixtures</strong> tested under wheel tracker<br />

NMAS Size<br />

(mm)<br />

16<br />

14<br />

12<br />

10<br />

8<br />

6<br />

4<br />

2<br />

0<br />

Class-A,PG76<br />

Class-B,PG76<br />

Class-A,PG58<br />

Class-B,PG58<br />

Rut depth <strong>of</strong> SMA mixtures<br />

(mm)<br />

25 o C 40 o C 55 o C<br />

Class-A,PG 76-22 2.23 4.16 6.59<br />

Class-B,PG 76-22 2.57 6.39 9.12<br />

Class-A,PG 58-16 2.62 9.41 11.02<br />

Class-B,PG 58-16 3.31 11.12 14.15<br />

4-4- Wheel Tracking Test<br />

Rut development history <strong>of</strong> asphalt mixtures used<br />

at different temperature levels have been shown in<br />

Figure 9. One can observe that rut depth formation<br />

at any number <strong>of</strong> load repetitions depends mainly on<br />

mixture compositions, especially the asphalt binder<br />

grade. <strong>Asphalt</strong> mixture with finer gradation and PG<br />

76 showed more resistance to rutting than a coarser<br />

gradation and PG 58 asphalt binder. A detailed<br />

comparison <strong>of</strong> rut development in different mixtures<br />

at different temperature levels under the wheel<br />

tracker machine have been reported in Table 6.<br />

The above table shows that the rut depth <strong>of</strong><br />

mixtures increases with an increase in the<br />

temperature levels. Also, one can easily compare the<br />

rut resistance <strong>of</strong> different mixtures at different<br />

temperatures. <strong>Asphalt</strong> mixture with finer aggregate<br />

gradation and PG 58 showed minimum resistance to<br />

rutting at any temperature level.<br />

0 2000 4000 6000 8000 10000 12000<br />

Load Cycles (N)<br />

5-iConclusion<br />

This study characterizes two asphalt binders and<br />

four mixtures based on their stiffness modulus and<br />

phase angles and determine some possible<br />

correlations between the set performance’s criteria<br />

for both. The following conclusions have been<br />

drawn <strong>from</strong> this study:<br />

Dynamic modulus and shear complex modulus<br />

are sensitive to test temperature and frequency <strong>of</strong><br />

loading. The elastic modulus (G*) <strong>of</strong> the binder<br />

mostly ranges between 20 to 0.01 MPa and that<br />

<strong>of</strong> mixtures between 400 and 4000 MPa in a<br />

temperature range <strong>of</strong> 25 to 55 o C.<br />

Dynamic modulus and wheel tracker testing<br />

showed that asphalt mixtures with finer gradation<br />

and PG76 resist more against rutting than PG 58<br />

and finer aggregate gradation.<br />

Performance <strong>of</strong> asphalt mixtures under a domain<br />

<strong>of</strong> frequency and temperatures can easily be


predicted <strong>from</strong> that <strong>of</strong> asphalt binders using the<br />

master curve techniques.<br />

References<br />

AASHTO T 240, (2004) "Standard Method <strong>of</strong> Test for<br />

Effect <strong>of</strong> Heat and Air on a Moving Film <strong>of</strong> <strong>Asphalt</strong><br />

Binder (Rolling Thin-Film Oven Test)".<br />

Amara Loulizi, Gerardo Flintsch, and Kevin McGhee<br />

(2006) "Determination <strong>of</strong> the In-Place Hot-Mix<br />

<strong>Asphalt</strong> Layer Modulus for Rehabilitation Projects<br />

Using a Mechanistic-Empirical Procedure"<br />

FHWA/VTRC 07-CR1, Virginia Transportation<br />

Research Council, Charlottesville, Virginia.<br />

American Association <strong>of</strong> State Highway and<br />

Transportation Officials (1995) "AASHTO Provisional<br />

Standard TP5-93: standard test method for<br />

determining the rheological properties <strong>of</strong> asphalt<br />

binder using a dynamic shear rheometer". Washington<br />

D.C.<br />

American Association <strong>of</strong> State Highway and<br />

Transportation Officials (TP-62) (2004) "Determining<br />

Dynamic Modulus <strong>of</strong> Hot-Mix <strong>Asphalt</strong> Concrete<br />

<strong>Mixtures</strong>".<br />

Anderson, D. A. and Christensen D.W. (1992)<br />

"Interpretation <strong>of</strong> Dynamic Mechanical Test Data for<br />

Paving Grade <strong>Asphalt</strong> <strong>Cement</strong>s," Proceedings <strong>of</strong> the<br />

Association <strong>of</strong> <strong>Asphalt</strong> Paving Technologists, volume<br />

61, pp. 67-116.<br />

Charles, E. Dougan, Jack, E. Stephens, James<br />

Mahoney and, Gilbert Hansen, (2003) "E*-<br />

DYNAMIC MODULUS" Test Protocol – Problems<br />

and Solutions, Report Number-CT-SPR-0003084-F-<br />

03-3, University <strong>of</strong> Connecticut, USA.<br />

Colbert B. and Zhanping, (2012) "The properties <strong>of</strong><br />

asphalt binder blended with variable quantities <strong>of</strong><br />

recycled <strong>Asphalt</strong> using short term and long term aging<br />

simulations", Journal <strong>of</strong> Construction and Building<br />

Materials, Vol. 26, pp. 552–557.<br />

Cooper Technology, (2006) "Wheel Tracker Small<br />

Device", Product Catalogue, Issue 1, CRT-WTEN1&<br />

CRT-WTEN2 in conformance to E13108.<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

European Standard; (2002) "Bituminous mixtures-Test<br />

methods for hot mix asphalt", EN 12697-22, part-2,<br />

Wheel Tracking.<br />

Faheem, A., and Bahia H.U. (2004) "Using Gyratory<br />

Compactor to Measure Mechanical Stability <strong>of</strong><br />

<strong>Asphalt</strong> <strong>Mixtures</strong>", Wisconsin Highway Research<br />

Program.<br />

Ge<strong>of</strong>fry M. Rowe, Salman HakimzadheKhee,<br />

PhillilBlanenship and Kamyar C. Mahboub (2009)<br />

"Evaluation <strong>of</strong> Aspects <strong>of</strong> E* Test by using Hot-Mix<br />

<strong>Asphalt</strong> Specimens with Varying Void contents",<br />

Transportation Research Record, Journal <strong>of</strong> the<br />

Transportation Research Board, No. 2127,<br />

Transportation Research Board <strong>of</strong> the National<br />

Academies, Washington D.C, pp. 164-172.<br />

Huang, Shin-Che and Zeng Menglan, (2007)<br />

"Characterization <strong>of</strong> aging effect on rheological<br />

properties <strong>of</strong> asphalt-filler systems", International<br />

Journal <strong>of</strong> Pavement Engineering, 8:3, pp. 213-223.<br />

Kanitpong, K., Bahia, H., (2005) "Relating adhesion<br />

and cohesion <strong>of</strong> asphalts to the effect <strong>of</strong> moisture on<br />

laboratory performance <strong>of</strong> asphalt mixtures", T. R. R,<br />

1901, pp. 33-43.<br />

Kim H. AND Partl M.N., (2009) "Stiffness<br />

comparison <strong>of</strong> mastics asphalt in different tests<br />

modes", 2nd workshop on four point bending, pais<br />

(ed.), University <strong>of</strong> Minho, ISBN 978-972-8692-42-1.<br />

Kumar S. A. and Veeraragavan A., (2011) "Dynamic<br />

mechanical characterization <strong>of</strong> asphalt concrete mixes<br />

with modified <strong>Asphalt</strong> binders", J. Materials Science<br />

and Engineering, Vol. 528, pp. 6445– 6454.<br />

Loulizi, Flintsch, Al-Qadi and Mokarem, (2006)<br />

"Comparison between Resilient Modulus and<br />

Dynamic Modulus <strong>of</strong> Hot-Mix <strong>Asphalt</strong> as Material<br />

Properties for Flexible Pavement Design".<br />

National Highway Authority, (1998) "Surface course",<br />

Item No. 305-I, General Specification, Pakistan.<br />

NurIzzi Md. Yus<strong>of</strong>f, Montgomery Shaw T. and<br />

Gordon D. Airey, (2011) "Modeling the linear<br />

viscoelastic rheological properties <strong>of</strong> bituminous<br />

binders", J. Construction and Building Materials,<br />

Vol. 25, pp. 2171–2189.<br />

Transportation Research Journal, Vol. 2, No. 1, 2012 /35


Pellinen, T K and Witczak, M W., (2002) "Stress<br />

dependent master curve construction for dynamic<br />

modulus".<br />

Soleimani A., (2009) "Use <strong>of</strong> dynamic phase angle<br />

and complex modulus for the low temperature<br />

performance grading <strong>of</strong> asphalt cements", Ph.D. Thesis,<br />

Queen’s University, Kingston, Canada, Page 7.<br />

36/ Transportation Research Journal, Vol. 2, No. 1, 2012<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …<br />

Tarefder, R.A., Zaman, M., Hobson, K., (2003) "A<br />

Laboratory and statistical evaluation <strong>of</strong> factors<br />

affecting rutting", International Journal <strong>of</strong> Pavement<br />

Engineering”. Vol. 4 (1), pp. 59-68.<br />

Wasage T.L.J., Jiri Stastna and LudoZanzotto, (2011)<br />

"Rheological Analysis <strong>of</strong> multi-stress creep recovery<br />

test", in Int. J. <strong>of</strong> Pavement Eng., Vol. 12, No. 6,<br />

pp. 561-568.


2/ Transportation Research Journal, Vol. 2, No. 1, 2012<br />

Hafeez et al. / <strong>Rutting</strong> Production <strong>of</strong> <strong>Asphalt</strong> …

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!