Laboratory and Field Performance Evaluation of Cracking of Airfield Warm Mix Asphalt with Reclaimed Asphalt Pavement at the National Airport Pavement and Materials Research Center

Abstract

The U.S. Federal Aviation Administration is exploring initiatives to decrease transportation greenhouse gas emissions by developing carbon reduction strategies, including using low-embodied carbon materials, such as reclaimed asphalt pavement (RAP), in airport pavement construction. However, verifying RAP’s viability for airfield pavements with laboratory and field performance measures is essential. This study compared the cracking susceptibility of a P-401 hot mix asphalt and warm mix asphalt (WMA)+RAP airfield mixes using laboratory performance tests and conducted a preliminary evaluation of pavement responses from accelerated pavement testing (APT) fatigue tests. The observations and outcomes presented in this paper found that adding 20% of recycled materials with WMA additives had a decreasing impact in mix performance but did not overly affect the asphalt mix properties, as cracking properties from the different test procedures were found with comparable outcomes. Additional observations from strain gauges and temperature probes in the National Airport Pavement and Materials Research Center test cycle 2 APT fatigue test sections showed that the maximum tensile strains in the WMA surface lane 5 south section were consistently higher than those in the WMA+RAP lane 6 south section, which may be because of the higher load level that it was exposed. Additionally, the hardened or aged RAP binder in the entire lane 6 south section may have increased its stiffness, resulting in lower strain levels than in the lane 5 south section. Notably, both sections showed an increase in tensile strains with an increase in asphalt concrete temperature, which confirmed the temperature dependency behavior of asphalt concrete.

Keywords

aviation airport pavement

The U.S. Federal Aviation Administration (FAA) Advisory Circular AC 150/5320-10H outlines the pavement materials and construction specifications for airport pavement projects funded by the Airport Improvement Program ( 1 ). For flexible airfield pavements subjected to aircraft weighing over 30,000 lb (13,600 kg), the P-401 specification for hot mix asphalt (HMA) should be used. However, using reclaimed asphalt pavement (RAP) in airfield pavements has been limited to shoulders and intermediate courses ( 1 ). Additionally, the current airfield material specifications do not allow warm mix asphalt (WMA) in airport pavement mixes (1, 2).

As the costs of virgin materials such as asphalt binder and aggregates continue to rise and environmentally friendly initiatives gain momentum, RAP has become popular for constructing and rehabilitating flexible pavements. In November 2021, the U.S. Department of Transportation launched a program to reduce transportation greenhouse gas (GHG) emissions by developing carbon reduction strategies ( 3 ). The goal is to cut GHG emissions in half by 2030 by supporting efforts, identifying projects, and implementing strategies to reduce transportation emissions. Incorporating recycled materials in flexible pavement construction is a green technology that has the potential to reduce the carbon footprint significantly (4, 5). RAP can replace virgin aggregates and asphalt binders, which mitigates the emissions associated with producing these materials while also creating cost savings ( 5 ). As a result, the asphalt pavement industry is continuously exploring materials and strategies and developing new techniques for incorporating recycled and environmentally friendly materials into pavements.

The FAA’s research and development program is exploring the use of low-embodied carbon materials in airport pavement construction. Although using recycled materials in airfield pavements can reduce transportation time and costs, it is essential to assess the viability of using RAP in HMA for airfield pavements based on performance measures such as cracking and rutting resistance under repetitive traffic and environment-induced loading. Research must ensure that pavements with RAP and other green technologies perform at least as well as pavements with conventional or standard materials. By ensuring equivalent or better pavement performance, the benefits of recycling can be achieved without sacrificing pavement life or performance in airfield pavements (1, 2, 4).

In 2015, the FAA established the National Airport Pavement and Materials Research Center (NAPMRC) at the William J. Hughes Technical Center in Atlantic City, New Jersey. NAPMRC is home to a heavy vehicle simulator for airports (HVS-A) used for accelerated pavement testing (APT) on airfield surface layers. HVS-A can apply bi-directional or uni-directional wheel loads using a single wheel with a maximum wheel load of 100,000 lb (444.8 kN), or dual wheel gear with a maximum wheel load of 50,000 lb (222.4 kN) for each of the dual wheels ( 2 ). HVS-A also has heaters and insulation panels that can raise the pavement surface temperature to 150°F (67°C) and maintain it for the duration of the tests. Additionally, an air conditioning system can be implemented to maintain low temperatures during the summer months, allowing for low and intermediate temperatures for fatigue traffic tests even during summer months.

In 2016, test cycle 1 (TC-1) was conducted at NAPMRC to examine the impact of high aircraft tire pressure on high-temperature pavement rutting and the possibility of using WMA on airport pavement ( 2 ). The test lanes were divided into three sections, and each section was subjected to full-scale APT with various combinations of tire pressure, temperature, and surface rut accumulation. The results of APT and laboratory characterization tests showed that WMA performed similarly to P-401 HMA ( 2 ). Following TC-1, test cycle 2 (TC-2) test sections were built in 2019 to study the rutting and fatigue performances of chemical, organic, and hybrid additive-based WMAs and compare their performance with the FAA specification P-401 HMA. The TC-2 test sections were also constructed to investigate the fatigue cracking potential of HMA, WMA, and WMA+RAP under heavy aircraft loads. The study involved constructing four outdoor test lanes and two test lanes containing RAP in the indoor facility. Each test lane comprised of three different test sections. Like TC-1, TC-2 (which is still ongoing) used the single-wheel assembly for traffic testing.

Fatigue tests were performed on the south test sections in each lane. The fatigue tests were performed at a temperature of 68°F (20°C) at a depth of 2 in. (5.1 cm) below pavement surface. (More details about fatigue tests are provided in a later section.) A detailed laboratory characterization plan was also undertaken to evaluate the cracking potential of the HMA and recycled material sections (lane 5 south and lane 6 south) and relate to the observed performances during the APT traffic tests. Laboratory performance tests and a preliminary assessment of the RAP+WMA test sections under APT traffic tests are presented in this paper. The fatigue life and damage propagation were assessed using the visco-elastic continuum damage (cyclic fatigue) tests in the laboratory. The cracking potential of the mixes at low and intermediate temperatures was evaluated with disk-shaped compact tension (DCT) and semi-circular bending tests. Viscoelastic properties were also evaluated using the dynamic modulus test. TC-2 APT results will provide the necessary information for considering “green” material technologies in airfield asphalt surface courses. Detailed results from laboratory performance tests and a preliminary assessment of the test sections under APT traffic tests under fatigue tests are presented in this paper.

Objectives

The primary objectives of this study are:

1) To understand the cracking susceptibility of the TC-2 P-401 HMA, and WMA+RAP mixtures using laboratory cracking performance tests

2) To conduct a preliminary evaluation of pavement responses from APT fatigue traffic tests

Background

Several laboratory studies have determined the feasibility of using WMA+RAP. The presence of aged binder from RAP is expected to increase the performance grade (PG), which results in higher rut resistance. However, the brittleness of the hardened binder, coupled with the low production temperature of WMA, could make these mixes susceptible to low-temperature cracking and moisture damage. Behbahani et al. investigated the rutting resistance of organic- and chemical-additive-treated warm mixes with 25%, 50%, and 75% RAP with regard to dynamic creep and wheel tracking tests ( 6 ). The inclusion of RAP led to increased rutting resistance and higher resilient modulus. The organic-additive-based WMA performed better than the chemical-additive-treated WMA with regard to resilient modulus and permanent deformation. Oliveira et al. studied the stiffness, rutting, and fatigue performance of hot and warm mixes with 50% RAP ( 7 ). The warm mix was treated with a surfactant-based chemical additive. The recycled mixture showed comparable or better performance than the conventional hot mix by all metrics ( 7 ). Mejías-Santiago et al. conducted a laboratory characterization study to evaluate the mechanical performance of airfield-specific warm mixes with RAP ( 8 ). Chemical-, foam-, and organic-additive-based warm mixes with 0%, 25%, and 50% RAP were tested for permanent deformation, workability, durability, and moisture susceptibility ( 8 ). The inclusion of RAP contributed to increased rut resistance in the airfield mixes. Several other studies have also concluded similar findings of increased rut resistance with the inclusion of RAP in WMAs with different additives (8 –12).

Several studies also evaluated the cracking potential and fatigue life of WMA+RAP, showing mixed results. For example, Das et al. found that wax modification had a minimal negative impact on low-temperature cracking performance ( 13 ). Ghabchi et al. found that WMA+RAP mixes had lower temperature cracking potential than HMA counterparts ( 14 ). Hill et al. showed that chemical additives improved fracture resistance compared with HMA mix ( 15 ). Kim et al. found that warm mix additives negatively affected the fatigue performance of mixes ( 10 ). In contrast, Singh et al. showed that wax-treated WMAs had better fracture properties than their chemical counterparts ( 16 ).

Research has also been conducted to evaluate the use of warm mixes with RAP in laboratory studies, accelerated pavement testing (APT), and field trials. In 2017, the Virginia Department of Transportation (VDOT) initiated a study to explore if using warm mixes with RAP by balanced mix design (BMD) was feasible ( 17 ). Diefenderfer et al. researched nine different mixes from two field trials that included varying percentages of RAP, two warm mix additives, two recycling agents, and two binder grades ( 17 ). The study found that WMAs with softer binder grades, 40% RAP, and recycling agents could meet VDOT BMD performance thresholds. In a subsequent study, VDOT conducted APT on full-scale test sections with 30%, 45%, and 60% RAP WMA ( 18 ). The study found that mixes with high RAP content exhibited higher rut depths than the control mix ( 18 ). The Federal Highway Administration (FHWA) also initiated an APT study to assess the fatigue cracking performance of eco-friendly asphalt materials. Ten full-scale test lanes were constructed with chemical- and foam-based warm mixes with 20% and 40% replacement binder ratios from RAP at the FHWA accelerated loading facility in 2013 ( 19 ). A softer binder grade was conducive to lower stiffness and longer fatigue life with 40% RAP recycled binder ratio (RBR) mixes. The lower level of short-term aging with WMA led to a slight improvement in the 20% RAP RBR mixes, but a conclusion could not be drawn in the case of the 40% RBR mix ( 19 ). The fatigue life of the recycled asphalt shingles mixtures decreased with increased aging. Most of the scientific literature on WMA+RAP is related to highway applications. This study aims to fill the gap by verifying the findings from a full-scale APT study considering extensive laboratory characterization for cracking performance of airfield WMA+RAP mixes under high tire pressure.

Full-Scale Test Sections and Materials

NAPMRC APT Test Sections

The TC-2 section lanes replicate an airfield pavement structure that could withstand the high tire pressures and heavy loads from heavy aircrafts (i.e., Boeing 777, Airbus A380). For example, as shown in Figure 1, the asphalt layer in the test sections is 9.0 in. (22.9 cm) thick. The crushed stone P-209 layer is 8 in. (20.3 cm) thick, while the P-154 subbase is 12 in. (30.5 cm) thick. The sandy subgrade remained unchanged from the previous NAPMRC APT study (TC-1) and has a California Bearing Ratio (CBR) of 15 ( 2 ). Figure 1 shows the pavement cross-section used in lanes 1, 5, and 6 in TC-2. Lane 1, with a conventional P-401, is a standard polymer-modified HMA with a PG of 76-22, which serves as the control mix (mix A) for the TC-2 study. Because the inclusion of latex and 20% RAP were anticipated to increase the high temperature PG, the mixes used in the indoor lanes 5 and 6 (under a canopy) both have a high temperature PG of 64. Mix E is a WMA Sasobit with lime and latex used as modifier and 20% RAP. Furthermore, lane 5 includes mix E on the bottom 6 in. (15.2 cm) of the asphalt layer, and the top 3 in. (7.6 cm) contain mix F, which is a WMA PG-64-22 mix with Sasobit, latex, and lime, and no RAP. Lane 6 consists of the 20% RAP mix E throughout the entire 9 in. (22.9 cm) layer, as shown in Figure 1. Figure 2, a to c, displays photos of the TC-2 test section lanes discussed in this paper. The customized cooling system, shown in Figure 2d, made it possible to test asphalt pavements during warmer seasons for fatigue evaluations.

Figure 1.

Test cycle 2 full-scale pavement cross-sections: test lane 1 (left), test lane 5 (middle), test lane 6 (right).

Figure 2.

National Airport Pavement and Materials Research Center (NAPMRC) outdoor and indoor accelerated pavement test sections: (a) lane 1 hot mix asphalt PG76-22, (b) lane 5 warm mix asphalt (WMA) PG64-22, (c) lane 6 WMA PG64-22+reclaimed asphalt pavement, and (d) heavy vehicle simulator cooling system.

A single job mix formula (JMF) was developed for the control mix and the WMA mixes to ensure consistent performance of the asphalt mix in the field, with additional requirements. Once the JMF was approved for the control mix, the different WMA additives were added with the dosage specified by the manufacturer. However, during test production, an increase in air voids (AV) was observed when using the selected asphalt content of 5%. Therefore, the asphalt content was slightly increased to meet the 3.5% AV P-401 specifications requirement (1, 20). More information about the design and construction of the TC-2 test lanes is available elsewhere ( 20 ). Table 1 summarizes some of the asphalt mixes’ essential characteristics and volumetric properties from quality control data. As shown in Figure 3, all mixes were designed to use the same aggregate gradation. The asphalt contents of mix F and mix E were 0.1% and 0.2% higher than that of mix A, respectively. The dust-to-binder ratio, voids in mineral aggregate, and voids filled with asphalt (VFA)—three critical volumetric properties related to mix performance—are presented in Table 1. Mix F has a higher AV with similar asphalt content, producing a lower VFA than the other two mixes.

Table 1.

Selected Warm Mix Technologies for Test Cycle 2 Lanes 1, 5, and 6

Test section	1	5		6
Mix type and binder grade	HMA PG76-22	WMA PG64-22		WMA PG64-22
Mix designation	Mix A	Mix F	Mix E	Mix E
Asphalt layer thickness	9 in. PG76-22 HMA	3 in. PG64-22 WMA (surface)	6 in. PG64-22 WMA+RAP (bottom)	9 in. PG 64-22 WMA+RAP
Additive type and additive content	None	WMA organic (1.5%), lime (1%), and latex (4%)
Number of gyrations	75
RAP content	0%	0%	20%
Mix design air voids	4.3%	4.6%	4.5%
Asphalt content	5.1%	5.1%	5.2%
VMA	17.2	17.0	18.2
VFA	75.0	72.9	81.9
D/B	84.8	92.1	82.9

Note: D/B = Dust/Binder Ratio; HMA = Hot Mix asphalt; PG = performance grade; RAP = Recycled asphalt pavement; VMA = Voids in mineral aggregate; VFA = Voids filled with asphalt; WMA = warm mix asphalt.

Figure 3.

Surface mixes’ aggregate gradations from quality control data.

The test lanes in TC-2 are divided into three sections: the north, center, and south. Ongoing fatigue testing is being conducted in the south sections only. To simulate long-term aging, before APT trafficking, all south sections were artificially aged by heating the pavement at a controlled temperature of 120°F (48.9°C) for 336 h, measured at 2 in. (5.1 cm) below the surface of the asphalt concrete layer. A single wheel with a tire inflation pressure of 254 psi (1,751.3 kPa) applying 61.3 kip (272.7 kN) was moved bi-directionally at a constant speed of 3 mph (5 km/h) to traffic the test sections described in this paper following a predesigned wander pattern. The HVS-A tire load and inflation pressure generated a tire contact area of 288 in.² (0.1858 m²), as shown in Figure 4a. Figure 4b also illustrates a tire contact area of 330 in.² (0.2129 m²), corresponding to 70 kip (311.4 kN). Wander was introduced to simulate the lateral spread of loading associated with various aircraft gear geometries typically observed in airfield flexible pavements. The pavement temperature was maintained at 68°F (20°C) throughout the duration of the traffic test at the same depth mentioned above. Additional details on the traffic test parameters and wander sequence can be found elsewhere (2, 20, 21).

Figure 4.

Heavy vehicle simulator for airports tire contact area: (a) 61.3 kip tire load level and (b) 70 kip tire load level.

Materials

This study evaluated plant-mixed laboratory compacted (LC) and plant-mixed field compacted (FC) test specimens using a set of laboratory performance tests. The LC specimens were prepared from plant-produced loose mixes sampled from the top 3 in. (7.6 cm) surface lifts during construction. The loose mix samples were reheated to the compaction temperature, and specimens were compacted using a gyratory compactor to the target AV content. All LC tests were conducted on test specimens that experience short-term conditioning only, plus the additional aging that happens during the reheating of the loose mix during laboratory compaction. (Note that a new FAA long-term conditioning protocol for asphalt mixes has recently been developed but was not available during the laboratory characterization.) In addition, corresponding FC specimens were prepared from field cores extracted soon after the construction of the TC-2 test lanes and taken for evaluation to the FAA NextGen Pavement Materials Laboratory, located at the National Airport Pavement Test Facility in Atlantic City, NJ. Consequently, the FC specimens were not subjected to long-term conditioning either, as they were obtained before the test lanes were subjected to any artificial conditioning.

Figure 5 shows the AV content distributions of the LC and FC specimens corresponding to all test specimens. As shown in Figure 5, the LC specimens were compacted in the laboratory to 5% target AV content to match average field conditions (field density) in the three test lanes. The mean AV content of all FC specimens is just below 5%, with specimens having AV contents in the range of 5%–8%, especially in mix E and mix F, and as low as 2%–3% for specimens from mix A. Because LC mixes were compacted to the same AV content in the laboratory, they were used to compare mix characteristics more appropriately. On the other hand, FC specimens were used to explore corresponding expected outcomes of the field performances in the APT section. Both sets of test specimens were prepared and tested at the FAA NextGen Pavement Materials Laboratory.

Figure 5.

Air voids content distribution in test specimens.

Laboratory Test Methods

This study aimed to assess the cracking properties of asphalt mixes at low and intermediate temperatures, as well as the linear viscoelastic properties, using LC and FC test specimens. Mixes A, E, and F were used in the study, and various test procedures were employed. Further information on the different testing methods is presented next.

Intermediate Temperature Cracking Test

The fracture potential of asphalt mixes was evaluated using the Illinois flexibility index test (IFIT) AASHTO T-393 procedure. IFIT is conducted at an intermediate temperature using a semi-circular specimen with a vertical notch along its symmetric axis. The test standard requires 1.97 in. (50 mm) thick, 5.9 in. (150 mm) diameter semi-circular specimens to be tested using a three-point bending apparatus at the constant displacement rate of 1.97 in./min (50 mm/min). A 0.6 in. (15 mm) deep, 0.06 in. (1.5 mm) wide notch is cut along the specimen’s axis of symmetry to force the failure location. The flexibility index (FI) is derived from the mixture fracture parameters of fracture energy (G_f) Gf_DCT refers to the fracture energy determined from the DCT test. Other instances (Gf) are used to refer to the fracture energy determined from the IDIT test. and post-peak slope (m), as shown in Equation 1. FI indicates the likelihood of asphalt mixture field cracking performance. However, different researchers have found that G_f, m, and FI in IFIT heavily depended on the AV content of the test specimens, showing higher FI values with higher test specimen AV (22, 23). Therefore, correction factors have been recommended to adjust the FI values to standard conditions (22 –24). In this study, 5% AV content was taken as the standard condition for FC specimens. Furthermore, a simple linear relationship for adjustment, as shown in Equation 2, will be used for AV normalization of FC test specimens, which present different AV content distribution (21, 22). It should be noted that in Equation 2, AV is expressed in percent.

FI = (\frac{g_{f}}{m}) x A,

(1)

where

FI = flexibility index,

$g_{f}$ = fracture energy (J/m²),

$m$ = post-peak slope (kN/mm), and

$A$ = unit conversion and scaling coefficient, taken as 0.01.

F I_{av corrected} = FI x \frac{5.0}{AV},

(2)

Low-Temperature Cracking Test

A DCT test was conducted following ASTM D7313 to evaluate the fracture properties of the mixes at low temperatures. The DCT requires a circular specimen of 5.9 in. (150 mm) diameter by 1.97 in. (50 mm) height with a 0.06 in. (1.5 mm) wide single edge notch and two loading holes parallel to the faces by 90°. Before testing, the test specimen is conditioned for at least 2 h in an environmental chamber at a suggested test temperature of 10°C greater than the continuous low temperature PG of the asphalt binder. The specimen is loaded in tension, and the test is performed with a constant crack mouth opening displacement rate of 0.0007 in./s (0.017 mm/s). The fracture energy (G_f-_DCT) is used to describe the fracture resistance of asphalt mixtures with higher G_f-_DCT, indicating better cracking resistance. Peak loads associated to crack initiation in the DCT test were also evaluated.

Dynamic Modulus Test

Dynamic modulus testing was conducted following AASHTO T342 to characterize the viscoelastic behavior of asphalt mixtures. A sinusoidal compression load is applied to asphalt mixture specimens at different temperatures and frequencies in unconfined conditions. Tests were performed at five temperatures: 14°F, 39°F, 50°F, 70°F, and 99°F (−10°C, 4°C, 10°C, 21°C, and 37°C) using loading frequencies of 25, 10, 5, 1, 0.5, and 0.1 Hz. Dynamic modulus tests were conducted on three replicate specimens, with a diameter of 4 in. (100 mm) and trimmed to 6 in. (150 mm) height. The dynamic modulus and phase angle data were shifted accordingly using the time-temperature superposition principle to generate master curves for a straightforward evaluation of the stiffness characteristics of the mixes.

Researchers have investigated the correlation between dynamic modulus test properties and cracking performance (25, 26). For example, the Glover–Rowe parameters are obtained from asphalt binder shear modulus from dynamic shear rheometer data and have been proposed to assess cracking resistance characteristics ( 27 ). Similarly, in recent years, the mixture Glover–Rowe parameter (G-R_m), as determined from asphalt mixture dynamic modulus and phase angle data, has been evaluated, showing acceptable correlations with some cracking parameters from the fracture tests ( 26 ). G-R_m is calculated at the temperature-frequency combination of 68°F (20°C) and 5 Hz, as shown in Equation 3, with a higher G-R_m suggesting higher cracking potential ( 26 ).

G - R_{m} = \frac{| E^{*} | {(\cos (δ))}^{2}}{Sin (δ)},

(3)

where

$| E^{*} |$ = dynamic modulus, and

$δ$ = the phase angle of the asphalt mixture at the corresponding test temperature and frequency.

Cyclic Fatigue

The direct tension cyclic fatigue test AASHTO T411 (formerly AASHTO TP 133) was conducted to determine damage characteristics and fatigue analysis parameters of asphalt mixes. During testing, the actuator applies a displacement-controlled repeated cyclic loading to the asphalt mix tests specimen until it fractures ( 28 ). Small 1.5 in. (38 mm) diameter by 4.3 in. (110 mm) length test specimens were subjected to direct cyclic fatigue testing to evaluate their fatigue cracking characteristics. The simplified viscoelastic continuum damage (S-VECD) model is used to analyze the cyclic fatigue data and relates the material’s integrity, represented by the pseudo stiffness (C), to the amount of cumulative damage (S) (28 –30). The S-VECD model applies the pseudo stiffness concept to indicate material integrity under loading. The C-value starts with the unity when the material is undamaged and decreases as S accumulates ( 29 ). The model defines a unique relationship (the so-called damage characteristic curve) between C and S for a given mixture (28, 29). The C versus S relationship is a constitutive material property independent of strain level, mode of loading, temperature, and loading amplitude, and it is a vital material input for estimating long-term performance using the FlexPave software (29 –31). The characteristic damage curve and different performance-based fatigue indices from the cyclic fatigue tests were used for fatigue cracking evaluation. These inputs include the mix toughness or the average reduction in material integrity up to the failure point (D_r) index and the toughness and the total pseudo strain of the mix under loading (S_app) index (29, 30, 32).

AASHTO T411 (formerly AASHTO TP 133 calls for small specimens to evaluate a mix cracking resistance. Before testing, test specimens need to be outfitted with platens on both ends that are used to transfer the load from the testing machine to the specimens. The end platens are attached to the top and bottom of the specimens using epoxy glue. The fatigue testing was performed using an asphalt mixture performance tester by applying a sinusoidal load at the frequency of 10 Hz and using different strain rates. Testing temperatures were determined based on the binder PG. The test results were imported to FlexMat, an Excel-based software tool for characterizing dynamic modulus and cracking of asphalt mixes.

Results and Discussions

IFIT Test Results

Figure 6 displays the IFIT results, highlighting the G_f, m, and FI values. The error bars represent the variability among the test replicates. G_f is the energy required to create a unit surface area of a crack. It represents a mix’s ability to resist cracking and is calculated by dividing the area under the load-displacement curve by the crack’s area propagating during testing. Figure 6a shows that the LC mixes F and E have similar G_f values, especially when considering the error bars indicating plus and minus one standard deviation. However, it is observed that mix E test specimens have a significantly higher average m-value than the other mixes, most likely because of the presence of stiffer recycled binder in the mix. A higher m-value indicates faster crack propagation once the specimen has been cracked. Therefore, the results from Figure 6a indicate that the mixes required similar energy to crack, however, once cracking has been initiated, the development of the crack in mix E may growth faster than mix A and F. Figure 6b displays the FC test specimen results, showing that the mixes have similar G_f. Yet, the m-value of mix A is higher than the other mixes, probably because of the higher density of mix A. Since mixes with similar G_f values have shown opposite behavior, the FI has been recommended to rank cracking resistance and differentiate a brittle mix prone to early cracking.

Figure 6.

Illinois flexibility index test results: (a) fracture energy (G_f) and post-peak slope value of laboratory compacted (LC) mixes, (b) G_f and post-peak slope value of field compacted (FC) mixes, and (c) flexibility indexes (FI) of LC and FC mixes.

In Figure 6c, the LC specimens of mix A performed similarly to those from mix F, while mix E had the lowest FI among all LC specimens, most likely because of the presence of RAP. To evaluate field conditions fairly in relation to FI values, the FC specimens’ FI values have been normalized for AV using Equation 2. After normalization, similar values of FI are shown for FC mixes in Figure 5. When comparing LC and FC specimens test results, there’ is a significant difference in results. LC test specimens show higher m-values and lower FI values than the corresponding FC. The LC test specimens were transported into the laboratory and reheated to compaction temperature for a few hours, which could have led to additional aging that could have influenced the test results.

DCT Test Results

Figure 7 shows the average peak loads and G_f-DCT values. Figure 7a displays the peak loads. Commonly, higher mix stiffness results in higher peak loads in the DCT test; therefore, they are typically used as mix stiffness indicators. Considering the LC and FC test specimens’ results, there is no significant difference between peak loads in the three mixes, indicating comparable stiffness among the mixes at the testing temperature of 10.4°F (−12°C).

Figure 7.

Disk-shaped compact tension test results: (a) peak load and (b) fracture energy.

The G_f-DCT values of LC and FC mix test specimens are displayed in Figure 7b. G_f-DCT is determined from the work of fracture in DCT testing and is linked to thermal cracking in asphalt pavements ( 33 ). Many transportation agencies use G_f-DCT value and different thresholds as a thermal cracking performance parameter or index for pavement projects ( 33 ). Again, the LC specimens performed similarly at the testing temperature, indicating similar low-cracking susceptibility among the mixes. The trend observed among the FC test specimens is comparable with that from the LC mixes.

Dynamic Modulus Test Results

Dynamic modulus master curves were constructed at 68°F (20°C) using the time-temperature superposition principle, as shown in Figure 8, a and b. The master curves overlap in a small area across the frequency range. However, mix E appears to have a higher stiffness than the other two, especially in Figure 8a (LC). Phase angle master curves on the other hand, are shown in Figure 8, c and d. The peak phase angle moves up and left with lower frequencies and higher temperatures. Minimal variation in phase angle values is observed in both LC and FC mixes. LC mix E and mix F have comparable phase angles and dynamic modulus values at high frequencies, indicating similar stiffness and relation characteristics. However, RAP mixes showing high dynamic modulus values at high frequencies, and low phase angle values may be more susceptible to cracking than those without RAP under the same conditions. In Figure 8d, mix A has lower phase angle values than the other mixes, which may be because of the higher density (lower AV content) in FC test specimens.

Figure 8.

Dynamic modulus test results: (a) dynamic modulus values in laboratory compacted (LC) mixes, (b) dynamic modulus values in field compacted (FC) mixes, (c) phase angles of LC mixes, (d) phase angles values in FC mixes, and (e) mixture Glover–Rowe (G-R_m) parameter.

Figure 8e displays the G-R_m parameter values for LC and FC mixes. It is worth noting that G-R_m from the LC mixes captures the influence of RAP addition in mix E, exhibiting a significant increase compared with mix A and mix F. Meanwhile, FC mixes have comparable values, especially between mix A and mix E. This suggests that a mix with a softer PG and RAP may perform similarly to a mix with a stiffer binder, to a certain extent. The softer PG could balance out the RAP’s contribution to the mix’s overall stiffness. FC mix E has a higher G-R_m compared with the FC mix F. Generally, mixes with the same PG display lower G-R_m (better cracking resistance) without RAP. Thus, according to the results, mix F will likely have better cracking resistance in the field than mix E under similar testing conditions.

Cyclic Fatigue Test Results

Figure 9, a and b, shows the damage characteristic curves of LC and FC mixes, respectively. The curves illustrate how C, or material integrity, decreases as internal micro-cracking damage increases in the test specimen ( 19 ). The last point in the C-S curves shown in the figures (the C_f value), represents the C-value at the maximum phase angle point (failure). Higher C_f values indicate mixtures with a lower tolerance to resist damage before cracking. Figure 9a shows that all three LC mixes have a similar tendency to resist damage, with low values (higher tolerance to damage). However, as noted by several researchers, C_f values alone cannot fully characterize the fatigue resistance or susceptibility of mixes (34, 35). For instance, mix F shows inferior internal material integrity with increasing damage, suggesting inferior performance among the LC mixes. Mix E displayed the best performance among the curves, surpassing others because of its higher stiffness from RAP. In Figure 9b, mix F demonstrated the least capacity to maintain its integrity with increasing damage, indicated by a higher C_f. This observation suggests that the WMA additives may not significantly enhance the fatigue resistance of mixes, as per other studies (31, 36, 37). In contrast, mix A in Figure 9b showed the best ability to maintain material integrity with increasing damage, probably because of the significantly higher stiffness from lower AV content (see Figure 1) in test specimens, higher binder PG, or both.

Figure 9.

Cyclic fatigue test results: (a) DCC for laboratory compacted (LC) mixes, (b) DCC field compacted (FC) mixes, (c) mix toughness or the average reduction in material integrity up to the failure point (D_R) parameter LC mixes, (d) D_R parameter FC mixes, (e) the toughness and the total pseudo strain of the mix under loading (S_app) parameter LC mixes, and (f) S_app parameter FC mixes.

Figure 9c indicates similar D_R values for mix A and Mix F, while mix E had the lowest D_R value. The D_R values of the FC mixes in Figure 9d followed the same trend as the damage characteristics curve. The control mix in Figure 9d showed better performance compared with the WMA mixes. It should be noted that D_R values only measure material toughness and ignore stiffness or modulus values, therefore they cannot fully explain fatigue cracking resistance. The S_app values may be better fatigue resistance indicators as they account for both the material’s modulus and toughness (30, 19, 34). In Figure 9e, LC mixes showed comparable S_app values. Figure 9f presented FC S_app’s results, mirroring the observations made from the damage characteristic curves (Figure 9b), showing mix A with better performance than mix F, probably capturing the higher density. The FC mix E in Figure 9f also showed higher damage capacity than mix F, which may seem counterintuitive considering the RAP content, but can be explained by the higher mix stiffness, as the S_app parameter consider both the effects of material toughness and stiffness at the same time.

Analysis of Variance and Tukey’s Significant Difference Test

Visual observation of test replicates has been used to assess trends in the data. However, a more formal approach is used in this section to identify the statistical significance or differences among mix test properties. The analysis of variance (ANOVA) was used to determine if the cracking performance properties and indices of different mixes from the IFIT, DCT, and cyclic fatigue tests are significantly different. The analysis was conducted for both LC and FC mix test specimens and p-values were tabulated from the analyses as shown in Table 2. The analysis was not performed for G-R_m values, as the analysis provided a single value. The null hypothesis (H_o) in ANOVA assumes that all means are equal and can be rejected if the p-value is less than the significance level (0.05 in this study). Although ANOVA can determine if one or more group means are different, it cannot determine which particular differences between pairs of means are significant. Tukey’s significant difference test can be employed to identify the specific group means that differ. As shown in Table 2, out of all the LC test specimen properties, the only one with a p-value lower than the significance level is the G_f from the IFIT, indicating rejection of the H_o. In Figure 10, the LC IFIT G_f Tukey’s interval shows the statistical difference between mix A and mix E. Interestingly, zero is barely outside the Tukey interval between the mix A and mix F G_f mean values. It is important to note that the corresponding means are significantly different if the interval does not contain zero. When analyzing the FC mixes, only the m p-value from IFIT shows significance. All the other p-values are above the significance levels, indicating insufficient evidence to reject H_o, therefore accepting the alternative hypothesis that the mean properties are similar. Figure 10 also shows Tukey’s interval of the m-value, which leads to a similar outcome as the LC case. However, in this case the difference in m-values from mix A and mix F was also found significant.

Table 2.

P-Values from Analysis of Variance

Test property	Laboratory compacted	Field compacted
IFIT
G_f	0.02	0.59
m	0.06	0.01
FI	0.09	0.21
DCT
Peak load	0.12	0.38
G_f	0.81	0.10
Cyclic fatigue
D_r	0.62	0.20
S_app	0.14	0.08

Note: IFIT = Illinois flexibility index test; Gf = Fracture energy for IFIT test; m = post peak slope; FI = Flexibility Index; DCT = Disc Shaped Compact Tension Test; Gf = Fracture energy from DCT test; Dr = Index representing average loss of integrity per cycle throughout an asphalt mixture s fatigue life; Saap = Fatigue cracking index parameter.

Figure 10.

Tukey’s intervals of pairwise differences between cracking properties: (a) laboratory compacted (LC) fracture energy (Gf) from Illinois flexibility index test (IFIT) and (b) field compacted (FC) post-peak slope (m) from IFIT.

Field Performance from APT

APT studies are an essential tool that helps us understand how pavements perform in real-life situations. While laboratory tests can provide helpful information about the engineering properties of pavement materials, combining data from both APT and laboratory characterization can provide a better overall understanding of how the pavement behaves and how it will perform over time. The combination of both types of analysis allows for a more comprehensive examination of the pavement’s lifespan and performance.

As mentioned before, the south sections were designated for the assessment of fatigue performance in TC-2. Followed by the initiation of traffic test in lane 6 south, no cracking was visible even after the application of 91,264 passes with the 61.3 kip (272.7 kN) moving wheel load. Loading was elevated to 72 kip (320.3 kN) for the remainder of trafficking in that section. The remaining south sections were trafficked with the 72 kip (320.3 kN) wheel load for 75,000 passes. None of the sections discussed in this paper exhibited any visible cracking, regardless of load level or mix type. Pavement sensor (asphalt strain gages, pressure cells) responses also showed no indication of crack initiation.

The fatigue damage algorithm in the FAARFIELD design procedure relies on the tensile strain at the bottom of asphalt concrete for the prediction of service life ( 38 ). Considering this, three asphalt strain gauges were installed in both parallel and perpendicular directions of traffic in each test section. A thermocouple tree consisting of five temperature probes was installed in the asphalt concrete layer to monitor the temperature of the asphalt concrete at five different depths. Compressive stresses on top of each layer were measured with a single pressure cell over the duration of trafficking. Additional details on instrumentation are beyond the scope of this paper and have been discussed elsewhere ( 21 ).

Key observations from the embedded strain gauges and temperature probes in lane 5 and 6 south sections as are summarized in Figure 11 and can be used together with the materials properties from the different test procedures to verify the outcomes from the APT test. None of the strain gauges survived in the controlled section paved with the PG76-22 polymer-modified mix (mix A). Figure 11a shows sample strain responses from the transverse asphalt strain gages (ASGs) in lane 5 and 6 south sections. Residual noises recorded at the beginning of data acquisition were subtracted from the subsequent strain measurements for an objective comparison of the measured strain values. The letter “P” in the legend of Figure 11a stands for number of passes, whereas the letter “O” designates the transverse offset of the moving wheel in inches. The maximum tensile strains presented in this paper were selected on the premise that the wheel was directly on top of the embedded ASG, at which the sensor registered the highest response.

Figure 11.

Pavement responses from instrumentation during accelerated pavement testing: (a) sample strain responses from the transverse ASGs in Test Cycle 2 (TC-2) lane 5 and 6 south sections, (b) maximum tensile strains versus respective number of passes, and (c) maximum tensile strain versus temperature at the bottom of asphalt layer.

Observed maximum tensile strains are plotted against the respective number of passes and temperature at the bottom of the asphalt concrete in Figure 11, b and c, respectively. L6S-TSG-1 exhibited highly sporadic tensile strains at the beginning of the traffic test and failed below around 13,000 passes. Erratic responses from that gauge led to the wide variation in tensile strains in that section. Overall, maximum tensile strains recorded in the lane 5 south section were consistently higher than those recorded in the lane 6 south section. Both sections exhibited an increase in tensile strains with an increase in asphalt concrete temperature, validating the viscoelastic behavior of asphalt concrete. The relatively higher magnitude of strains in lane 5 south section might be attributed to a higher load level, as mentioned above, and the 3 in. softer surface lift in the lane 5 south section paved with mix F, unlike the bottom two lifts paved with the WMA+RAP mix. The three lifts of asphalt concrete in lane 6’s south section had the same mix E. The presence of hardened or aged binder in RAP might have contributed to increased stiffness in the surface lift of lane 6 south section and therefore resulted in lower strain levels at the bottom compared with its lane 5 counterpart.

Conclusions

The main objectives of this study were to understand the cracking susceptibility of the TC-2 P-401 and WMA+RAP mixtures using laboratory cracking performance tests and to conduct a preliminary evaluation of pavement responses from APT fatigue traffic tests. The observations and outcomes presented in this paper suggest that the addition of low amounts of recycled materials (i.e., 20%) in conjunction with WMA additives had a decreasing impact in mix performance but did not overly affect the asphalt mix properties as cracking properties from the different test procedures were found with comparable outcomes. Therefore, the results may indicate the absence of an excessively brittle and stiff mix with significantly inferior cracking performance when mixes were evaluated in the laboratory using LC test specimens with the same AV content. Field test specimens’ (field cores) evaluations, on the other hand, mainly captured the effect of density generally providing superior performance in the laboratory to the control mix which showed higher density (lower AV content). However, findings should be validated with further data, especially with regard to their practical application. This is because there may be different variables or factors that cannot be adequately isolated during laboratory testing, especially when using FC specimens. The results and observations of this study suggest the following conclusions.

1) The findings from IFIT suggest that RAP decreased the performance of the airfield TC-2 mixes. Test results indicate that the mixes required similar energy to crack; however, once cracking initiated, the m-value indicates that the development of the crack in mix E grew faster than mix A and mix F. The use of additives such as WMA and softer binders may have mitigated the impact of RAP, leading to comparable performance results, as evidenced by the findings from FC mix testing.

2) Overall DCT test results showed that the three evaluated mixes presented comparable performance, indicating similar thermal cracking susceptibility when tested at a temperature of 10.4°F (−12°C).

3) Dynamic modulus master curves presented comparable values, indicating that the WMA mixes with softer PG and with and without low RAP content could perform similarly to the control mix with a stiffer binder, to a certain extent, when considering stiffness. The softer PG used in WMA mixes may have counterbalanced the contribution of RAP to the overall stiffness of the mix. However, the G-R_m index captured the stiffening effect of RAP more significantly in LC mix E. Thus, according to those results, mix E presented lower cracking resistance when compared with the other two mixes under similar conditions.

4) The D_r and S_app indexes can be used as toughness and fatigue resistance indicators, respectively. The indexes found no significant differences between the mixes when considering LC test specimens. On the other hand, FC mix E and F, were found to have inferior performance compared with the control mix, probably because of both having higher AV content (lower density) than mix A.

5) Only IFIT was able to discriminate significant differences between the mixes when using ANOVA. In general, there was not enough statistical evidence to reject H_o, that the test properties and indexes means were equal for the other test evaluations.

6) As the asphalt concrete temperature increased, both sections experienced an increase in tensile strains. The higher magnitude of strains in the lane 5 south section may be attributed to a heavier load level and the use of a 3 in. softer surface lift paved with mix F.

Future research efforts will aim to characterize and differentiate the extent and effects of field aging on the different TC-2 mixes by evaluating properties from the mix and extracted binder using the post-APT field cores.

Footnotes

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: N. Garg; data collection: D. Batioja-Alvarez, H. Kazmee; analysis and interpretation of results: D. Batioja-Alvarez, H. Kazmee; draft manuscript preparation: D. Batioja-Alvarez, N. Garg, H. Kazmee. All authors reviewed the results and approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work described in this paper was supported by the FAA Airport Technology Research and Development Branch.

ORCID iDs

Dario Batioja Alvarez

Hasan Kazmee

Navneet Garg

The work described in this paper reflect the views of the authors, who are responsible for the facts and accuracy of the data presented within. The contents do not necessarily reflect the official views and policies of the FAA. The paper does not constitute a standard, specification, or regulation.

References

FAA. Item P-401 Asphalt Mix Pavement. In Advisory Circular 150/5370-10H - Standards for Specifying Construction of Airports. Office of Airport Safety & Standards - Airport Engineering Division, Federal Aviation Administration, Washington, D.C., pp. 263–292.

Garg

Kazmee

Ricalde

Parsons

Rutting Evaluation of Hot and Warm Mix Asphalt Concrete under High Aircraft Tire Pressure and Temperature at National Airport Pavement and Materials Research Center. Transportation Research Record: Journal of the Transportation Research Board, 2018. 2672: 117–127.

US Department of Transportation. President Biden, USDOT Announce New Guidance and $6.4 Billion to Help States Reduce Carbon Emissions under the Bipartisan Infrastructure Law. https://www.transportation.gov/briefing-room/president-biden-usdot-announce-new-guidance-and-64-billion-help-states-reduce-carbon. Accessed July 15, 2022.

Bizarro

D. E. G.

Steinmann

Nieuwenhuijse

Keijzer

Hauck

Potential Carbon Footprint Reduction for Reclaimed Asphalt Pavement Innovations: LCA Methodology, Best Available Technology, and Near-Future Reduction Potential. Sustainability, Vol. 13, 2021, p. 1382. https://doi.org/10.3390/su13031382.

National Asphalt Pavement Association (NAPA). The Role of Asphalt Pavements in a Low-Carbon Transportation Network. https://www.asphaltpavement.org/uploads/documents/GovAffairs/NAPA%20Carbon%20Footprint%20Priorities.pdf. Accessed July 15, 2022.

Behbahani

Ayazi

M. J.

Moniri

Laboratory Investigation of Rutting Performance of Warm Mix Asphalt Containing High Content of Reclaimed Asphalt Pavement. Petroleum Science and Technology, Vol. 35, No. 15, 2017, pp. 1556–1561. https://doi.org/10.1080/10916466.2017.1316738.

Oliveira

J. R. M.

Silva

H. M. R. D.

Abreu

L. P. F.

Gonzalez-Leon

J. A.

The Role of a Surfactant Based Additive on the Production of Recycled Warm Mix Asphalts – Less Is More. Construction and Building Materials, Vol. 35, 2012, pp. 693–700. https://doi.org/10.1016/j.conbuildmat.2012.04.141.

Mejías-Santiago

Doyle

J. D.

Howard

I. L.

Brown

E. R.

Evaluation of Warm-Mix Asphalt Technologies for Use on Airfield Pavements. US Army Corps of Engineers, Washington, D.C., 2013.

Guo

You

Tan

Zhao

Performance Evaluation of Warm Mix Asphalt Containing Reclaimed Asphalt Mixtures. International Journal of Pavement Engineering, Vol. 18, No. 11, 2017, pp. 981–989. https://doi.org/10.1080/10298436.2016.1138114.

10.

Kim

Norouzi

Kass

Liske

Kim

Y. R.

Mechanistic Performance Evaluation of Pavement Sections Containing RAP and WMA Additives in Manitoba. Construction and Building Materials, Vol. 133, 2017, pp. 39–50. https://doi.org/10.1016/j.conbuildmat.2016.12.035.

11.

D. X.

Saleh

Laboratory Evaluation of Warm Mix Asphalt Incorporating High RAP Proportion by Using Evotherm and Sylvaroad Additives. Construction and Building Materials, Vol. 114, 2016, pp. 580–587. https://doi.org/10.1016/j.conbuildmat.2016.03.200.

12.

Valdes-Vidal

Calabi-Floody

Sanchez-Alonso

Performance Evaluation of Warm Mix Asphalt Involving Natural Zeolite and Reclaimed Asphalt Pavement (RAP) for Sustainable Pavement Construction. Construction and Building Materials, Vol. 174, 2018, pp. 576–585. https://doi.org/10.1016/j.conbuildmat.2018.04.149.

13.

Das

P. K.

Tasdemir

Birgisson

Low Temperature Cracking Performance of WMA with the Use of the Superpave Indirect Tensile Test. Construction and Building Materials, Vol. 30, 2012, pp. 643–649. https://doi.org/10.1016/j.conbuildmat.2011.12.013.

14.

Ghabchi

Singh

Zaman

Laboratory Evaluation of Stiffness, Low-Temperature Cracking, Rutting, Moisture Damage, and Fatigue Performance of WMA Mixes. Road Materials and Pavement Design, Vol. 16, No. 2, 2015, pp. 334–357. https://doi.org/10.1080/14680629.2014.1000943.

15.

Hill

Behnia

Buttlar

W. G.

Reis

Evaluation of Warm Mix Asphalt Mixtures Containing Reclaimed Asphalt Pavement through Mechanical Performance Tests and an Acoustic Emission Approach. Journal of Materials in Civil Engineering, Vol. 25, No. 12, 2012, pp. 1887–1897. https://doi.org/10.1061/(ASCE)MT.1943-5533.0000757.

16.

Singh

Ashish

P. K.

Chitragar

S. F.

Laboratory Performance of Recycled Asphalt Mixes Containing Wax and Chemical Based Warm Mix Additives Using Semi Circular Bending and Tensile Strength Ratio Tests. Construction and Building Materials, Vol. 158, 2018, pp. 1003–1014. https://doi.org/10.1016/j.conbuildmat.2017.10.080.

17.

Diefenderfer

S. D.

Boz

Habbouche

Balanced Mix Design for Asphalt Surface Mixtures: 2019 Field Trials. Research Report No. FHWA/VTRC 21-R21. Virginia Transportation Research Council, Charlottesville, pp. 1–82. http://www.virginiadot.org/vtrc/main/online_reports/pdf/21-r21.pdf.

18.

Tong

Habbouche

Flintsch

G. W.

Diefenderfer

B. K.

Rutting Performance Evaluation of BMD Surface Mixtures with Conventional and High RAP Contents under Full-Scale Accelerated Testing. Materials, Vol. 16, No. 24, 2023, p. 7611. https://doi.org/10.3390/MA16247611.

19.

Gibson

Comparison of Laboratory Fatigue Characteristics with Full-Scale Pavement Cracking for Recycled and Warm-Mix Asphalts. Transportation Research Record: Journal of the Transportation Research Board, 2016. 2576: 100–108.

20.

Garg

Kazmee

Ricalde

Comparative Performance of Different Warm Mix Asphalt Technologies under the Influence of High Aircraft Tire Pressure and Temperature. Transportation Research Record: Journal of the Transportation Research Board, 2021. 2675: 657–669.

21.

Kazmee

Garg

Offenbacker

Ricalde

Villafane

Comparing Critical Pavement Responses in HMA and WMA (with RAP) Test Sections Measured Using Embedded Sensors under Accelerated Pavement Tests at NAPMRC. Proc., Airfield and Highway Pavements, Austin, ASCE, New York, 2023, pp. 293–304. https://doi.org/10.1061/9780784484890.027.

22.

Batioja-Alvarez

Lee

Haddock

J. E.

Understanding the Illinois Flexibility Index Test (I-FIT) Using Indiana Asphalt Mixtures. Transportation Research Record: Journal of the Transportation Research Board, 2019. 2673: 337–346.

23.

Kaseer

Yin

Arámbula-Mercado

Eps Martin

Daniel

Salari

Development of an Index to Evaluate the Cracking Potential of Asphalt Mixtures Using the Semi-Circular Bending Test. Construction and Building Materials, Vol. 16, 2018, pp. 286–298.

24.

Rivera-Perez

J. J.

Effect of Specimen Geometry and Test Configuration on the Fracture Process Zone for Asphalt Materials. Master thesis. University of Illinois at Urbana-Champaign, 2017.

25.

Zhang

Sias

J. E.

Dave

E. V.

Evaluation of the Cracking and Aging Susceptibility of Asphalt Mixtures Using Viscoelastic Properties and Master Curve Parameters. Journal of Traffic and Transportation Engineering, Vol. 9, No. 1, 2022, pp. 106–119.

26.

Rahbar-Rastegar

Daniel

J. S.

Reinke

Comparison of Asphalt Binder and Mixture Cracking Parameters. Road Materials and Pavement Design, Vol. 18, Supplement 4, 2017, pp. 211–233.

27.

Glover

C. J.

Davison

R. R.

Domke

C. H.

Ruan

Juristyarini

Knorr

D. B.

Jung

S. H.

Development of a New Method for Assessing Asphalt Binder Durability with Field Validation. Rep. No. FHWA/TX-05/1872-2. Texas State Dept. of Highways and Public Transportation, Austin, 2005.

28.

Underwood

B. S.

Baek

Kim

Y. R.

Simplified Viscoelastic Continuum Damage Model as Platform for Asphalt Concrete Fatigue Analysis. Transportation Research Record: Journal of the Transportation Research Board, 2012. 2296: 36–45.

29.

Norouzi

Kim

Y. R.

Mechanistic Evaluation of Fatigue Cracking in Asphalt Pavements. International Journal of Pavement Engineering, Vol. 18, No. 6, 2017, pp. 530–546.

30.

Federal Highway Administration. Cyclic Fatigue Index Parameter for Asphalt Performance Engineered Mixture Design. https://www.fhwa.dot.gov/pavement/asphalt/analysis/. Accessed July 15, 2022.

31.

Federal Highway Administration. Cyclic Fatigue Index Parameter (S_app) for Asphalt Performance Engineered Mixture Design. Tech Brief. 2019. https://www.fhwa.dot.gov/pavement/pubs/hif19091.pdf. Accessed July 20, 2022.

32.

Ozer

Al-Qadi

I. L.

Singhvi

Khan

Rivera-Perez

El-Khatib

Fracture Characterization of Asphalt Mixtures with High Recycled Content Using Illinois Semicircular Bending Test Method and Flexibility Index. Transportation Research Record: Journal of the Transportation Research Board, 2016. 2575: 130–137.

33.

Zhu

Dave

E. V.

Rahbar-Rastegar

Daniel

J. S.

Zofka

Comprehensive Evaluation of Low-Temperature Fracture Indices for Asphalt Mixtures. Road Materials and Pavement Design, Vol. 18, Supplement 4, 2017, pp. 467–490.

34.

Mensching

D. J.

Rahbar-Rastegar

Underwood

B. S.

Daniel

J. S.

Identifying Indicators for Fatigue Cracking in Hot-Mix Asphalt Pavements Using Viscoelastic Continuum Damage Principles. Transportation Research Record: Journal of the Transportation Research Board, 2016. 2576: 28–39.

35.

Etheridge

R. A.

Wang

Y. D.

Kim

S. S.

Kim

Y. R.

Evaluation of Fatigue Cracking Resistance of Asphalt Mixtures Using Apparent Damage Capacity. Journal of Materials in Civil Engineering, Vol. 31, No. 11, 2019, p. 04019257.

36.

Zhang

Wang

Xing

Predict the Phase Angle Master Curve and Study the Viscoelastic Properties of Warm Mix Crumb Rubber-Modified Asphalt Mixture. Materials, Vol. 13, No. 21, 2020, p. 5051.

37.

Batioja-Alvarez

Lee

Rahbar-Rastegar

Haddock

J. H.

Asphalt Mixture Quality Acceptance Using the Hamburg Wheel-Tracking Test. Transportation Research Record: Journal of the Transportation Research Board, 2020. 2674: 338–349.

38.

FAA. Airport Pavement Design and Evaluations (FAA Advisory Circular AC 150/5320-6G). Office of Airport Safety and Standards, Federal Aviation Administration, Washington, D.C., 2020.