técnico

1. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
Performance of the FHWA’s ALF Modified-Binder Pavements
Xicheng Qi, Ghazi Al-Khateeb, and Aroon Shenoy
Turner-Fairbank Highway Research Center / SaLUT
6300 Georgetown Pike, McLean, VA 22101-2296, USA
Terry Mitchell, Nelson Gibson, Jack Youtcheff, Tom Harman
Federal Highway Administration
6300 Georgetown Pike, McLean, VA 22101-2296, USA
Theme: Pavement Performance/topic number: 3.2 Full-scale and accelerated pavement
testing
ABSTRACT: In the summer of 2002, twelve (12) full-scale lanes of pavements with various
modified asphalts were constructed at the Federal Highway Administration’s (FHWA)
Pavement Testing Facility (PTF) in Virginia. The primary goal of the study is to use FHWA’s
two Accelerated Loading Facility (ALF) machines to validate and refine changes being
proposed in the Superpave binder specification to properly grade modified binders. In
December 2002, ALF loading began on the pavements to induce two primary modes of
failure: rutting and fatigue cracking. Currently, rut testing has been completed at 64 °C on all
12 lanes. Fatigue testing at 19 °C is underway and scheduled to be complete by early 2006.
The pavement performance is being compared to both binder and mixture parameters. This
paper presents an overview of the pooled fund study, rutting and fatigue cracking
performance and correlation analyses with the laboratory testing results obtained to date.
KEY WORDS: Superpave, modified-binders, APT, rutting, fatigue.
1. INTRODUCTION
With the growing use of polymer-modified asphalts, improvement in Superpave’s asphalt
binder specification is needed to better predict the relative performance of modified and
unmodified binder pavements. Under current specifications, the rankings of many polymer-
modified asphalts by standardized laboratory test procedures do not agree with those by
pavement performance. The FHWA recent multiyear effort is to study the performance of
Superpave mixtures containing polymer-modified asphalt binders. This research is supported
by 16 State highway agencies and more than 30 industry sponsors through a transportation
pooled fund study TPF-5 (019) titled, “Full-Scale Accelerated Performance Testing for
Superpave and Structural Validation.” In the summer of 2002, twelve (12) full-scale lanes of
pavements with various modified asphalts were constructed at the FHWA’s PTF in Virginia.
In December 2002, loading began on the pavements using two ALF machines to induce two
primary modes of failure: rutting and fatigue cracking. The field rutting performance data has
been correlated with binder parameters and performance of the asphalt mixtures in laboratory
tests, e.g., the Superpave shear tester, simple performance tests (SPT), and the French and
Hamburg Wheel Tracking Devices. The field fatigue cracking data are being compared to
both binder parameters and the mixture lab testing data from the Beam fatigue test, the SPT,
and the indirect tensile test (IDT). This paper presents an overview of the pooled fund study,

2. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
2
the pavement rutting and fatigue cracking performance and correlation analyses with the
laboratory testing results obtained to date.
2. FHWA PAVEMENT TESTING FACILITY
2.1 ALF
FHWA’s PTF consists of two ALF machines (Figure 1), to simulate traffic loading at
controlled loading and pavement temperatures, and about 3420 m2
(0.83 acres) of grounds
that provide space for 12 pavement test lanes.
The ALF machines are 29-m (95-ft) long frames with rails to direct rolling wheel loads.
Each ALF machine is capable of applying an average of 35,000 wheel passes per week from
a half-axle load ranging from 33 to 84 kN (7,500 to 19,000 lbf). The load is applied
unidirectionally at 18 km/hr (11 mi/hr) to a 14-m (45-ft) length of pavement. The machines
allow testing with conventional dual truck tires or wide-based, “super-single” tires and
simulation of the real-world, lateral distribution of truck loadings using programmed
transverse wheel wander. In the current experiment, both machines are equipped with super-
single (425/65R22.5 wide base) tires.
Figure 1: The FHWA two ALF machines loading pavements at PTF site.
2.2 Pavement Test Lanes
The current layout of the 12 as-built pavement lanes is presented in Figure 2. Each pavement
lane is 4 m (13 ft) wide and 50 m (165 ft) long, and is divided into four test sites. All lanes
consist of a hot-mix asphalt (HMA) layer and a dense-graded, crushed aggregate base (CAB)
course over a uniformly prepared, AASHTO A-4 subgrade soil. The total thickness of the
HMA and CAB layers is 660 mm (26 in.). Lanes 1 through 7 were constructed with a 100-
mm (4-in.) thick layer of HMA, while lanes 8 though 12 were constructed with 150 mm (6
in.) of HMA. The binders used in each lane are also listed in Figure 2. Note that the control
binder (PG 70-22) and three modified binders (Air-Blown, SBS-LG, and Terpolymer) are
used in both 100-mm and 150-mm thick lanes to see the effect of thickness of HMA layer.
2.3 Binder Selection
Most of the binders chosen for this field study had the same base asphalt (a Venezuelan
blend) and were modified to have the same high temperature Superpave performance grade
(PG 74-xx) so that the observed performance could be attributed only to the mode of

3. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
3
modification. Testing of binder samples collected during construction showed that the binders
had the following continuous PG values: PG 70-22 control (continuous PG 72-23); air-blown
(PG 74-28); SBS LG (PG 74-28); CR-TB (PG 79-28); Terpolymer (PG 74-31); and SBS 64-
40 (PG 71-28). The CR-TB missed the PG target of 74-xx, and the SBS 64-40 was purposely
designed have a PG different from the rest in order to check out whether the performance of
binders with high polymer content and soft bases can be captured by the Superpave
specification. The intermediate grade temperatures for |G*|sinδ = 5MPa, which are shown in
the legend of Figure 2, differ significantly and should provide a good test for checking the
ability of the current intermediate binder specification to rate asphalt binders according to the
fatigue cracking performance.
Lane Number 1 2 3 4 5 6 7 8 9 10 11 12
AR-AZ
PG
70-22
PG
70-22
Air-
Blown
SBS
LG
CR-
TB
Ter-
polymer
Fiber
Removed 100
mm
PG
70-22
SBS
64-40
Air-
Blown
SBS
LG
Ter-
polymer
Of Existing
CAB
100 mm of New No. 21A CAB Under All 12 Lanes
Removed 50 mm of Existing CAB
Existing VDOT No. 21A Crushed Aggregate Base (CAB)
(25-mm Nominal Maximum Aggregate Size)
Bottom of CAB to Pavement Surface is 660 mm
Re-compacted AASHTO A-4 Subgrade Soil
PG 70-22 = Unmodified Asphalt Binder Control (Intermediate Grade Temperature TIS = 26.1°C)
CR-AZ = Crumb Rubber Asphalt Binder, Arizona DOT Wet Process
CR-TB = Crumb Rubber Asphalt Binder, Terminal Blend (TIS =17.9°C)
Terpolymer = Ethylene Terpolymer Modified Asphalt Binder (TIS = 14.3°C)
SBS LG = Styrene-Butadiene-Styrene Modified Asphalt Binder with Linear Grafting (TIS = 18.1°C)
SBS 64-40 = Styrene-Butadiene-Styrene Modified Asphalt Binder Graded PG 64-40 (TIS =8.6°C)
Air-Blown = Air-Blown Asphalt Binder (TIS = 22.6°C)
Fiber = Unmodified PG 70-22 Asphalt Binder with 0.2 Percent Polyester Fiber by Mass of the
Aggregate.
Figure 2: Layout of the 12 as-built pavement lanes (not to scale)
2.4 Pavement Construction
The test lanes were constructed in the summer and fall of 2002. The mixtures were produced
in a counter flow drum plant located in Sterling, Virginia, 27 km (17 mi) from the PTF site.
After transport, trucks unloaded the HMA into a material transfer device (MTD), which fed a
Blaw-Knox PF3200 rubber tire paver. Use of an infrared camera during construction
indicated the MTD was very effective in eliminating temperature and aggregate segregation.
All of the test lanes were constructed in two lifts, each 50-mm (2-in.) or 75-mm (3-in.)
thick, as appropriate. A 12.3-Mg (13.5-ton) vibratory roller was used for the breakdown,
followed by a 9.1-Mg (10-ton) static steel roller for the finish rolling.
12 x 4 m = 48 m
50 m
4 Test Sites for
Each Lane

4. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
4
An extensive quality control/quality assurance (QA/QC) test program was conducted
during the construction of both the crushed aggregate base and the HMA. The detailed testing
results have been reported elsewhere (Mitchell et al 2004).
3. ALF AND LABORATORY TESTS
3.1 ALF tests
Since each pavement lane has four test sites available, the full-scale pavement testing is being
conducted at two failure modes, rutting tests (sites 1 and 2) at 64 and 74 o
C (selected lanes),
and fatigue cracking tests (sites 3 and 4) at 19 and 28 o
C. According to the results of
“shakedown” rutting and fatigue cracking tests early in 2003, it was decided that all rutting
tests use a wheel load of 44 kN (10,000 lb) without transverse wander while all fatigue tests
use a wheel load of 74 kN (16,600 lb) with transverse wander. This will fit the project
schedule and provides results in a reasonable length of time. An infrared heating system and
thermocouples in the pavements provide the required pavement temperature.
During loading, pavement layer rutting data are collected through differential rod and level
surveys on eight sets of reference plates installed at the time of construction along the
centerline of the test section. The plates are located at the surface of the pavement and on top
of the aggregate base in order to measure permanent displacement at these two locations at
predetermined ALF loading passes. The difference between these two measurements yields
the permanent vertical deformation (rutting) in the asphalt layer.
For fatigue test sections, cracks were manually traced onto clear plastic Mylar sheets as
they formed at the surface of the pavements. Different color pens were used to correspond to
the number of load repetitions. Two approaches were used to process the data. One was to
measure the total crack length and the other was to measure the percentage of area cracked in
the loaded area.
3.2 Mixture Tests
Laboratory performance rutting tests were conducted in the bituminous mixture laboratory
(BML) on specimens fabricated from plant-produced loose mix sampled from trucks and then
lab-compacted or field cores taken from the ALF pavements. Table 1 summarizes the
different laboratory rutting tests conducted in the BML.
Laboratory performance fatigue tests were also conducted in the BML. Table 2 describes
the different laboratory fatigue tests performed in the BML.
3.3 Binder Tests
Binders were sampled from the plant tanks on the day of construction. The type of binder
tests and the testing conditions in the Binder Rheology Laboratory (BRL) used in this study
are shown in Table 3.
4. ALF TESTING RESULTS
4.1 Rutting Tests
The ALF rutting tests conducted at 64 o
C and 44 kN load were complete for all 12 lanes.
Figure 3 graphically presents the fairly wide range of rutting results in the HMA layer in two

5. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
5
groups for the two levels of HMA thickness: 100-mm for Lanes 1 to 7 and 150-mm for Lanes
8 to 12. A statistical analysis was conducted to identify any significant differences among the
mean values of rutting at 25,000 ALF passes for each thickness, respectively. This specific
Table 1: Summary of Laboratory Mixture Rutting Tests.
Type of
Rutting Tests
Source of
Specimens
Test
Temperature
(°C)
Testing Mode
Loading
Frequency (Hz)
Material
Property
Measured
Standard(1)
French
Permanent
Rut
Plant-Produced
Lab-Compacted
74
Reciprocating
pneumatic tire
1.1 Hz (speed =
1.1 m/s)
Rut Depth NA(2)
Hamburg
Wheel
Tracking
Plant-Produced
Lab-Compacted
64
Vertical moving
load and steel
wheel
0.5 Hz (speed =
0.3 m/s)
Rut Depth NA
Plant-Produced
Lab-Compacted
74Repeated
Shear at
Constant
Height Field Cores 64
Haversine shear
stress with rest
period applied
1.4 Hz (0.1-sec
loading and
0.6-sec rest
period)
Cycles to 2
percent Strain
AASHTO
TP7
Frequency
Sweep at
Constant
Height
Plant-Produced
Lab-Compacted
74
Sinusoidal shear
strain applied
Sweep of 10, 5,
2, 1, 0.5, 0.2,
0.1, 0.05, 0.02,
and 0.01 Hz
Shear
Modulus
AASHTO
TP7
Flow
Number
Plant-Produced
Lab-Compacted
64
Haversine load
with rest period
applied
1 Hz (0.1-sec
loading
followed by a
0.9-sec rest
period)
Flow Number
(Cycles at
Tertiary
Flow), and
Cycles to 2%
Strain
NCHRP
Report 465
Plant-Produced
Lab-CompactedDynamic
Modulus
Field Cores
58
Haversine load
applied
Sweep of 20,
10, 5, 1, 0.5,
and 0.1 Hz
Dynamic
Modulus and
Phase Angle
NCHRP
Report 465
Resilient
Modulus
Field Cores 40
Haversine load
with rest period
applied
1 Hz (0.1-sec
loading and
0.9-sec rest
period)
Resilient
Modulus
AASHTO
TP9
(1) Slight deviations from standards such as temperature were followed.
(2) NA = Not Applicable.
Table 2: Summary of Laboratory Mixture Fatigue Tests.
Type of
Fatigue Test
Source of
Specimens
Test
Temperature
(°C)
Testing Mode
Loading
Frequency (Hz)
Material
Property
Measured
Standard
Plant-
Produced Lab-
Compacted
Dynamic
Modulus
Field Cores
19
Haversine
load applied
Sweep of 20, 10,
5, 1, 0.5, and 0.1
Hz
Dynamic
Modulus and
Phase Angle
NCHRP
Report 465
Tensile
Strength
Field Cores 19
Constant ram
rate applied
NA
Tensile
Strength and
Strain at Failure
AASHTO
TP9
Bending Beam
Fatigue
Lab-Produced
Lab-
Compacted
19
Sinusoidal
Strain applied
10 Hz
Cycles to 50
percent
Stiffness
AASHTO
TP8
Table 3 Summary of Laboratory Binder Tests
Type of Test Experimental Conditions Standard
Frequency Sweep, 7°C, 19°C, 25°C, 64°C, 70°C, 76°C AASHTO T315-02Dynamic Shear
Rheometer (DSR)
Multi-Stress Creep Recovery Test, 64°C NA

6. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
6
number of ALF passes was selected because the rutting measurements at this pass number
were available for most of test lanes. The analysis results showed that four significant
different groups of rutting exist within 100-mm HMA pavements while only two significantly
different groups of rutting within 150-mm HMA pavements. Note that replicate rutting tests
were conducted in Lane 9, Site1 and Site 2, with binder SBS 64-40 to see the variation
between the replicate tests. Statistical tests on means of rutting from replicate tests showed no
significant difference at all ALF load passes. Another interesting finding is that the rutting
rankings of Terpolymer binder are just opposite in the two thickness pavements (lanes 6 and
12). These conflicting data are still under investigation. The current hypothesis attributes the
poorer performance of Lane 6 to an adverse chemical interaction between the Terpolymer
and the hydrated lime added during the mix production to reduce the moisture damage.
0.0
5.0
10.0
15.0
20.0
0 10,000 20,000 30,000 40,000 50,000 60,000
ALF Wheel Loading Pass
HMALayerRutting,mm
L6S1(Terpolymer)
L3S1(Air-Blown)
L2S1(70-22)
L1S1(AZ-CRM)
L4S1(SBS LG)
L7S1(Fibers)
L5S1(CR-TB)
0.0
5.0
10.0
15.0
20.0
25.0
0 10,000 20,000 30,000 40,000 50,000 60,000
ALF Wheel Loading Pass
HMALayerRutting,mm
L9S2(SBS 64-40)
L9Avg(SBS 64-40)
L9S1(SBS 64-40)
L8S1(PG 70-22)
L10S1(Air-Blown)
L11S1(SBS LG)
L12S1(Terpolymer)
(a) Lanes 1 to 7 with 100-mm Pavements (b) Lanes 9 to 12 with 150-mm Pavements
Figure 3: HMA layer rutting data at 64 o
C and 44 kN load
4.2 Fatigue Tests
The ALF fatigue tests at 19 o
C and 74 kN have been completed for all 100-mm HMA
pavements and are being conducted for 150-mm HMA pavements. The cumulative crack
lengths at various ALF passes are presented in Figure 4 for all 100-mm HMA pavements.
The percentage of area cracked was also counted for each lane; the fatigue performance
ranking is identical by both crack length and crack area. Since the crack length is more
sensitive to the ALF loading passes, it is selected for the following correlation analyses. As
expected, a wide range of fatigue performance can be observed from Figure 4. To normalize
the data, both the ALF pass at 20 m crack length and the crack length at 100,000 ALF passes
were used in correlation analyses.
5. MIXTURE TEST RESULTS AND CORRELATIONS
The laboratory rutting and fatigue test results were compared with the ALF test results and
are shown in Figure 5. Correlation between the ALF pavement rut depth and the different lab
rutting parameters were established. The correlation analysis between ALF rutting and lab
French / Hamburg Wheel Tracking devices rutting was also performed although it is not

7. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
7
shown in the figure. The French PRT rutting only showed a R2
of 0.47 with ALF rutting for
100-mm HMA pavements and no correlation with 150-mm pavements while Hamburg
rutting showed no correlation at all with ALF rutting for each thickness of pavements. Also
the ALF wheel passes at 20 m crack length (corresponding to some 20 percent area cracked)
as well as the crack length at 100,000 wheel passes were correlated with the lab fatigue
0
20
40
60
80
100
120
0 50,000 100,000 150,000 200,000 250,000 300,000
ALF Wheel Loading Pass
CumulativeCrackLength(m)
L3S3 (Air Blown)
L2S3 (Control)
L5S3 (CR-TB)
L6S3 (Terpolymer)
L4S3 (SBS LG)
L7S3 (Fibers)
L1S2 (CR-AZ)
Figure 4: Fatigue cracking data at 19 o
C and 74 kN load for 100-mm HMA pavements
parameters. The best lab rutting and fatigue parameters that provide the highest coefficient of
determination (R2
) were selected.
6. BINDER TEST RESULTS AND CORRELATIONS
6.1 Rutting
In seeking a correlation of ALF rutting performance to binder data, four parameters were
evaluated – (1) The conventional |G*|/sinδ (Bahia and Anderson, 1995), which is the inverse
of the loss compliance obtained from the time or frequency sweep, (2) |G*|/(1-1/tanδsinδ)
(Shenoy, 2004), which is the inverse of the non-recovered compliance obtained from DSR
time or frequency sweep, (3) JNR (D’Angelo and Dongré, 2004), which is the non-recovered
compliance obtained from DSR multi-stress creep recovery tests (MSCR), and (4) MVR
(Shenoy, 2001), which is the material’s volumetric-flow rate obtained from the flow
measuring device (FMD).
The comparisons were made with rut depth after 20,000 ALF passes, and the results of the
coefficient of determination R2
obtained for a linear fit are shown in Table 4.
Table 4 Coefficients of correlation between ALF rutting and binder parameters
(a)
R2
values
|G*|/sinδ at 64°C
& ω=10 rads/
|G*|/(1-1/tanδsinδ) at
64°C & ω=0.25 rads/s
JNR at 64°C & MSCR
1s/9s, for 25-3200Pa
MVR at 64°C &
L=1.225kg
All Pavements 0.28 0.06 0.03 0.30
Thin Pavements 0.52 0.33 0.09 0.05
Thick Pavements 0.003 0.19 0.0004 0.81
(b)

8. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
8
R2
values
THS when |G*|/sinδ = 2200
Pa at ω=10 rads/s
THS when |G*|/(1-1/tanδsinδ) = 50
Pa at ω=0.25 rads/s
THS when MVR = 50 cc/10min
at L=1.225kg
All Pavements 0.55 0.05 0.04
Thin Pavements 0.47 0.01 0.01
Thick Pavements 0.84 0.17 0.40
y = -0.346 x + 59.2
R2
= 0.86
0.0
5.0
10.0
15.0
20.0
0 50 100 150 200
E*/sinδδδδ at 0.1 Hz and 58°C (MPa)
PavementRutDepthat
40,000Passesand64°C(mm)
y = - 0.027 x + 19.5
R2
= 0.81
0.0
5.0
10.0
15.0
20.0
0 100 200 300 400 500
FN Cycles to 2% Strain at 64°C
PavementRutDepthat
40,000Passesand64°C
(mm)
y = - 0.001 x + 18.6
R2
= 0.74
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
0 2500 5000
Flow Number at 64°C
PavementRutDepthat
40,000Passesand64°C
(mm)
y = - 0.001 x + 51.0
R2
= 0.84
0.0
5.0
10.0
15.0
20.0
0 20000 40000 60000
FSCH G* (MPa) at 74o
C
PavementRutDepthat
40,000Passesand64°C
(mm)
y = -0.004 x + 18.99
R2
= 0.76
0.0
5.0
10.0
15.0
20.0
0 500 1000 1500 2000
RSCH Cycles to Failure at 74°C
PavementRutDepthat
40,000Passesand64°C
(mm)
y = 528 x - 356123
R2
= 0.92
0
100,000
200,000
300,000
400,000
0 500 1000 1500
IDT Tensile Strength at 19°C (kPa)
ALFPassesat20m
CrackLengthand19°C

9. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
9
y = 0.185 x - 84.20
R2
= 0.87
0.0
50.0
100.0
150.0
0 500 1000 1500
E* sinδδδδ at 0.1 Hz and 19°C
CrackLengthat100,000ALF
Passesand19°C
y = -223.47 x + 255539
R2
= 0.72
0
100,000
200,000
300,000
400,000
0 500 1000 1500
E* sinδδδδ at 0.1 Hz and 19°C
ALFPassesat20mCrack
Lengthand19°C
Figure 5: Mixture Rutting and Fatigue Correlations to ALF Rut Depth and Fatigue Cracking
6.2 Fatigue Cracking
In seeking a correlation of ALF fatigue cracking performance to binder data, three parameters
were evaluated – (1) The conventional |G*|sinδ (Bahia and Anderson, 1995) obtained from
the time or frequency sweep at low strains, (2) |G*S|sinδS (Shenoy, 2002) obtained from DSR
strain sweep at high strains, (3) EWF (Andriescu et al., 2004), which is the essential work of
fracture. EWF is still under investigation as how the tests were performed on PAV binder
while RTFOT is more appropriate given the young age of the ALF sections. The comparisons
were made with crack lengths (m) at 100K loads and ALF passes at 50m and 20m crack
lengths, and the results of the coefficient of determination R2
obtained for a linear fit are
shown in Table 5.
Table 5 Coefficients of correlation between ALF fatigue cracking and binder parameters
(a)
R2
values
Thin Pavements |G*|sinδ at 19°C, ω=10
rads/s, 0.4% strain, PAV
|G*S|sinδ S at 19°C, ω=10
rads/s, 25% strain, RTFOT
EWF (kJ/m2
) at
25°C, RTFOT
Crack Length (m) at 100K Loads 0.56 0.61 0.01
ALF Passes at 50m Crack Length 0.47 0.50 0.06
ALF Passes at 20m Crack Length 0.43 0.44 0.09
(b)
R2
values
Thin Pavements TIS when |G*|sinδ = 5 MPa at
ω=10 rads/s, 0.4% strain, PAV
TIS = TEsinδ where TE is when |G*S| = 1
MPa at ω=10 rads/s, 25% strain, RTFOT
Crack Length (m) at 100K Loads 0.71 0.78
ALF Passes at 50m Crack Length 0.50 0.59
ALF Passes at 20m Crack Length 0.47 0.54
7. SUMMARY AND CONCLUSIONS
● ALF loading on 12 full-scale lanes of pavements has generated a bank of field
performance data on a series of modified asphalt binders. Although the as-built PG
grades at high temperature are very close for all binders (except for the CR-TB
binder), a wide range of rutting performance has been found for the 100-mm HMA
pavements. The intermediate temperature binder properties were significantly
different and the pavement fatigue performance also showed wide variations. These

10. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
10
performance data provide a good test for checking the ability of the laboratory mix
testing and binder specifications to rate the binders according to the rutting and
fatigue performances.
● ALF performance and laboratory mixture test results were compared. An analysis
reveals that several laboratory mix tests are very promising to highly correlate to the
ALF pavement performance. The shear modulus from the SST frequency sweep at
constant height test provides a high correlation with the ALF pavement rutting. The
strength from indirect tensile tests was highly correlated to the ALF pavement fatigue
cracking. The mixture parameters from the dynamic modulus tests were correlated to
both the ALF rutting and fatigue performance. From these findings, both the SST
shear modulus test and dynamic modulus test have been selected to further evaluate a
much wider range of high temperature PG grade of modified binders, which will
provide extensive data to develop the refined binder specifications.
● A comprehensive correlation analysis was also conducted between the ALF testing
results and the selected binder parameters. The correlations for the high temperature
binder parameters were strongly dependent on the HMA thickness. Moderate
correlations were found between ALF fatigue cracking and the intermediate
temperature binder parameters. These binder parameters will be fully evaluated when
the extensive laboratory mixture characterizations are completed.
ACKNOWLEDGMENTS
The work presented here was conducted as part of national pooled fund study TPF-5(019),
which is partially funded by 15 State highway agencies (CT, FL, IL, IN, KS, MD, MI, MS,
MT, NE, NV, NJ, NM, NY, and TX) with materials provided by the asphalt industry; funds
are also provided by the FHWA, and the study is staffed by FHWA and its contractors.
REFERENCES
AASHTO TP7-01, 2001. Standard Test Method for Determining the Permanent Shear Strain
and Stiffness of Asphalt Mixtures Using the Superpave Shear Tester (SST). AASHTO
Provisional Standards, Interim Edition, American Association of State Highway and
Transportation Officials, Washington D. C.
AASHTO TP8-94, 2002. Method for Determining the Fatigue Life of Compacted Hot-Mix
Asphalt (HMA) Subjected to Repeated Flexural Bending, AASHTO Provisional Standards,
American Association of State Highway and Transportation Officials, Washington D. C.
AASHTO TP9-96, 2002. Method for Determining the Creep Compliance and Strength of Hot
Mix Asphalt (HMA) Using the Indirect Tensile Test Device, AASHTO Provisional
Standards, American Association of State Highway and Transportation Officials,
Washington D. C.
AASHTO T315-02, 2002. Determining the Rheological Properties of Asphalt Binder Using
A Dynamic Shear Rheometer (DSR). American Association of State Highway and
Transportation Officials, Washington, D. C.
Andriescu, A., Hesp, S. A. M. and Youtcheff, J. S., 2004. Essential and Plastic Works of
Ductile Fracture in Asphalt Binders. Transportation Research Record 1875, pp. 1-8.
Bahia, H. U. and Anderson, D. A., 1995. The New Proposed Rheological Properties: Why
Are They Required and How Do They Compare to Conventional Properties, Physical
Properties of Asphalt Cement Binder, American Society for Testing and Materials,
Hardin, J. C. (Ed.), pp. 1-27.

11. Qi, Al-Khateeb, and Shenoy
Proceedings of the 10th
International Conference on Asphalt Pavements (ICAP), Quebec, Canada,
August 12-17, 2006
11
D’Angelo, John and Dongré, Raj, 2004. Multi-Stress Creep Recovery: A New Approach to
the Refinement of High Temperature Binder Specification. 41st
Annual Meeting of the
Petersen Asphalt Research Conference, Cheyenne, WY.
Mitchell, T., Stuart, K., Qi, X., Al-Khateeb, G., Youtcheff, J., and Harman, T., 2004. ALF
Testing for Development of Improved Superpave Binder Specification. Proceedings of the
2nd
International Conference on Accelerated Pavement Testing, Minnesota, USA
Shenoy, Aroon, 2001. High Temperature Performance Grade Specification of Asphalt
Binder from the Material’s Volumetric-flow Rate. Materials and Structures, Vol. 34, pp.
629-635.
Shenoy, Aroon, 2002. Fatigue Testing and Evaluation of Asphalt Binders using the Dynamic
Shear Rheometer. ASTM Journal of Testing and Evaluation, Vol. 30(4), pp. 303-312.
Shenoy, Aroon, 2004. A Comprehensive Treatise of The High Temperature Specification
Parameter |G*|/(1-(1/tanδsinδ)) for Performance Grading of Asphalts, Applied Rheology,
Vol. 14(6), pp. 303-314.
Witczak, M., Kaloush, K., Pellinen, T., El-Basyouny, M. and Quintus, H. V., 2002. Simple
Performance Test for Superpave Mix Design, NCHRP Report 465, National Research
Council, Washington, D.C.