Evaluation of fatigue cracking performance of asphalt mixtures under heavy static and dynamic aircraft loads

Construction and Building Materials 95 (2015) 813–819

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Evaluation of fatigue cracking performance of asphalt mixtures under heavy static and dynamic aircraft loads Maria Chiara Guercio a, Yusuf Mehta b, Leslie Myers McCarthy c,⇑ a

Department of Civil and Environmental Engineering, Villanova University, 800 E. Lancaster Avenue, Villanova, PA 19085, USA Department of Civil and Environmental Engineering, Rowan University, Rowan Hall Rm 329, 201 Mullica Hill Rd, Glassboro, NJ 08028, USA c Department of Civil and Environmental Engineering, Villanova University, Tolentine Hall Rm 144, 800 E. Lancaster Avenue, Villanova, PA 19085, USA b

h i g h l i g h t s  Asphalt mixes used as surface in airfield taxiways and aprons were analyzed.  The mixes were compared based on their fatigue cracking performance.  The BRIC and HMA PG82-22 mixes performed better than the FAA P-401.  The WMA–RAP and HMA PG70-22 mixes performed similarly to the FAA P-401.

a r t i c l e

i n f o

Article history: Received 11 October 2014 Received in revised form 25 April 2015 Accepted 12 July 2015

Keywords: Airfield pavements Fatigue cracking Finite element analysis Overlay tester

a b s t r a c t This laboratory and analytical study investigated the pavement’s mechanical responses in terms of strains at the top and bottom of the asphalt surface layer for a number of modified asphalt mixtures. The purpose was to identify their potential for use in airfield aprons and taxiways that are subjected to heavy, slow-moving aircraft loads. The effects of these loads on fatigue behavior were evaluated in ABAQUS™ using the material properties determined in the laboratory. The findings indicated that a number of mixtures, including modified mixtures and warm mix asphalt, experience strain levels comparable to the FAA standard P-401 asphalt mixture. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Fatigue cracking is one of the major concerns in airfield flexible pavements especially when these pavements are subjected to heavy aircraft loads. The cracks initiate at the bottom of the asphalt surface where tensile stresses and strains are the highest and mitigate to the surface [22]. Fatigue cracking can also initiate at the surface of flexible airfield pavements, due to excessive tensile strains near the outer edge of the aircraft wheel [1]. Some of the factors associated with this fatigue cracking include structural support, asphalt content, air voids and aggregate characteristics, temperature, stiffness, thickness, and traffic [22,30]. The relationship between load-induced failure and strain in flexible pavements is dependent on the horizontal tensile strain at the bottom of the asphalt layer and the elastic modulus of the asphalt mixture [10]. The asphalt layer in flexible airfield pavements are currently

⇑ Corresponding author. E-mail address: [email protected] (L. Myers McCarthy). http://dx.doi.org/10.1016/j.conbuildmat.2015.07.025 0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.

designed with stiff, dense-graded hot mix asphalt (HMA) mixtures according to the Federal Aviation Administration (FAA) P-401 specifications [12]. These specifications, developed to provide guidance on the production of asphalt concrete for airfield applications, require the production and placement of dense-graded HMA mixtures with a 25-mm maximum aggregate size and state-specific Superpave penetration grade (PG) binders. In recent years, asphalt mixtures used in highway pavements have been modified to improve aging, rutting, and fatigue cracking performance. Among these asphalt mixtures, the most commonly used materials include a variety of polymer-modified mixtures, warm mix asphalt (WMA), and reclaimed asphalt pavement (RAP). Polymer-modified binders have been used in airfield flexible pavements to accommodate the effects of high ambient temperature and heavier aircrafts by increasing the stiffness of the asphalt mixture. Consequently, the increase in stiffness reduced the rutting potential of the asphalt mixture. The potential of polymer-modified asphalts to improve asphalt pavement rutting resistance for highway and airfield pavements is well documented [28,25,15,21].

814

M.C. Guercio et al. / Construction and Building Materials 95 (2015) 813–819

However, the fatigue life of asphalt mixtures might either increase or decrease with the use of modified binders [31,17]. Studies have shown that modified asphalt mixtures may reduce the number of strain cycles to failure. However, it was found that the same modifier used with different asphalt increased the fatigue life [19]. The selection of a polymer-modifier may have significant effects on the ultimate asphalt field performance [28]. The highway industry has implemented RAP in the aggregate portion of the asphalt mixture as a mean for achieving stiffer mixtures. Flexible pavements containing RAP have been evaluated by the U.S. Army Corps of Engineers (U.S. ACE) in airfield applications [18], and the addition of RAP has been reported to increase the dynamic modulus of airfield asphalt mixtures [17]. However, with the increase in stiffness, the concern of the mixture resistance to long-term fatigue cracking arises. Generally, the use of RAP increases the mixture laboratory resistance to fatigue cracking [29]. WMA is a relatively new technology that allows the reduction of the production and compaction temperatures of asphalt mixture by as much as 75 °C. This is achieved by altering the HMA with water-, organic-, or chemical-based additives. The purpose with WMA is to produce asphalt mixtures with comparable strength, durability, and performance characteristics as a conventional HMA, but at significantly lower production and compaction temperatures [8]. This technology was first introduced in airfield pavements in 2006 at the Boston Logan International Airport (runway 4R/22L) [8]. WMA was also paved at the Stevens Anchorage International Cargo Airport (on the taxiways) and at the Chicago O’Hare International Airport (on the runways and taxiways). Lower production temperatures can potentially improve pavement performance by reducing binder aging, providing added time for mixture compaction, and allowing improved compaction during cold weather paving [8]. Another emerging asphalt technology is the performance-based mixture known as bottom rich intermediate course (BRIC). This mixture is usually placed in flexible highway perpetual pavements between a concrete structural layer and an HMA overlay to minimize the development of cracking due to joint movement in composite pavements [27]. The implementation of BRIC allows for a thinner flexible pavement section compared to pavements using conventional HMA. This type of mixture reduces the potential for fatigue cracking and structural rutting and limits any pavement distress to the surface lift [26]. Due to the different loading configuration and frequency applied in an airfield pavement, the BRIC was analyzed as a surface lift in this study. This mixture was considered for this study because the load repetitions over the design fatigue life in airfield pavements are much lower, up to twenty times less, than those experienced in highways [9]. 1.1. Objectives This study investigated how the mechanical responses of a broad range of asphalt mixtures compare under heavy static and dynamic airfield loads using laboratory-measured fatigue cracking properties of the mixtures. The main objectives of this study are as follows:  to measure the cycles to failure of a broad range of asphalt mixtures according to the Texas Department of Transportation (TxDOT) overlay tester specifications;  to determine mechanical responses in terms of tensile strains and stresses at the bottom and top of the asphalt surface layer under these static and dynamic aircraft loading through a finite element analysis (FEA);  to determine the relative pavement fatigue life of the mixtures using the FAA Rigid and Flexible Iterative Elastic Layered Design tool, FAARFIELD (FAA, 2009).

Accomplishing the objectives will allow for the comparison of the mechanical responses and predicted fatigue life of various asphalt airfield pavement subjected to heavy, standing and slow moving aircrafts typical of airfield taxiways and aprons. The findings from this study will assist in discerning whether or not other types of asphalt mixtures can provide fatigue resistance properties comparable to those of the FAA P-401 mixture, making them viable options for use in airfield aprons and taxiways. As the highway pavement industry continues to move forward with the implementation of more innovative asphalt mixtures, it is essential that research be conducted in order to determine whether and how best the airfield sector can benefit from the characteristics of these same mixtures. 1.2. Scope This study is an expansion of a previous study [16] that focused on evaluating mechanical responses (stresses and deflections) of a broad range of mixtures used as surface course in airfield pavements especially in taxiways and aprons where aircrafts are standing or slow moving (4.8 km/h.). This study focuses on evaluating mechanical responses in terms of strains at the bottom of the same asphalt mixtures presented in the previous study [16]. Mixtures analyzed in this study include those modified by addition of polymers, lower production and compaction temperatures, addition of reclaimed asphalt pavement to the aggregate portion, or by implementing alternative aggregate gradation. Laboratory testing of laboratory-compacted specimens was conducted to determine the cycles to failure according to the TxDOT specifications. Tensile strains at the top and bottom of the surface course were obtained using the three-dimensional (3-D) FEA software, ABAQUS™ [32]. The FAARFIELD software was used to compare the design fatigue life of the different mixes for an airfield pavement. 2. Methods and materials 2.1. Materials A total of five modified asphalt mixtures were tested and analyzed in this study. The baseline case was a dense-graded asphalt mixture that met all the FAA P-401 mixture specifications [12]. The other four modified asphalt mixtures were obtained from recent highway field projects and consisted of WMA with 35% RAP added; two HMA mixtures with two different modified binder grades (PG82-22 and PG70-22); and, a BRIC mixture. All of the mixtures, except for the BRIC, have similar aggregate gradations and asphalt contents. The BRIC mixture is typically used in flexible highway perpetual pavements as an intermediate layer placed between a concrete structural layer and an HMA overlay [27]. The BRIC mixture features finer aggregates than to the other mixes. As a result, the BRIC mixture has the highest binder content, but a portion of this binder content is absorbed by the mineral aggregates. The asphalt absorbed may improve the asphalt mixture’s strength due to the increased mechanical interlocking [23]. Table 1 summarizes the features of the mixtures and their volumetric properties, while Fig. 1 presents the corresponding aggregate gradations. 2.2. Laboratory testing There have been numerous efforts on the development of asphalt performance tests to relate laboratory-measured parameters to predicted distresses in highway pavements. The most recent test that can be conducted in the Asphalt Mixture Performance Tester (AMPT) is the Overlay Test [34], which originated with the purpose of simulating the opening and closing of joints or cracks in an asphalt pavement (Fig. 2). This test is performed in accordance with the TxDOT test procedure Tex-248-F [33]. The overlay test assesses the mixture’s resistance to crack propagation and correlates well with the field cracking performance for both composite pavements and flexible pavements [35,7]. The overlay test is performed at 25 °C with a minimum opening width of 0.625 mm. The test measures how many loading cycles it takes for a specimen to fail. Each cycle consists of five seconds of loading and five seconds of unloading. In this test failure is defined as 93% reduction of initial load or 1200 cycles, whichever comes first. The specimens can be prepared from either field cores or from Superpave Gyratory Compactor (SGC) molded specimens.

815

M.C. Guercio et al. / Construction and Building Materials 95 (2015) 813–819 Table 1 Range of asphalt mixtures analyzed in study.

Table 2 Summary of pavement structural and material properties.

Mixture design properties

Asphalt mixtures tested FAA P-401 (baseline)

WMA–RAP

HMA PG82-22

HMA PG70-22

BRIC

PG Grade Asphalt content (%) RAP (%) NMAS (mm)

76-22 5.02 0 19.0

64-28 5.25 35 9.5

82-22 5.41 0 12.5

70-22 4.83 0 12.5

70-28 8.40 0 4.75

Aggregate gradation 25.4 mm 19 mm 12.5 mm 9.5 mm 4.75 mm 2.36 mm 1.18 mm 0.6 mm 0.3 mm 0.15 mm 0.075 mm

(% passing) 100 98 86 77 58 40 30 19 12 5 6

100 100 100 96 66 36 24 18 12 9 6.5

100 100 96 89 58 38 26 18 11 6 5

100 100 96 89 58 38 26 18 11 6 5

100 100 100 100 99 80 54 39 27 16 7

Material

Thickness (mm)

Density (kg/m3)

Poisson’s ratio

Elastic modulus (MPa)

Asphalt surface Stabilized asphalt-treated base P-401 Subbase P-209 Medium strength subgrade

127 127

– 2403

– 0.35

– 2758

216 2438

2162 93

0.35 0.4

261 72

density specification of 93 ± 1% (7 ± 1% in situ air voids) in accordance with AASHTO T-209 [4] and AASHTO T-166 [5], after being trimmed to test sample size as shown in the schematic in Fig. 3. The BRIC mixture was compacted to achieve 3% air voids in order to reflect the typical field density at which this mixture is placed. It should be noted that the BRIC is not used as a surface lift in highway pavements. Typically, failure criteria are up to 300 load cycles for dense-graded mixtures and 750 load cycles for fine-graded crack-attenuated mixtures [33]. Therefore, the main parameter for comparison among mixtures is their number of cycles to failure. 2.3. Airfield pavement analysis

100

FAA P-401 (Baseline) WMA-RAP

90

Percent Passing (%)

80 70

BRIC

60

HMA PG82-22

50

HMA PG70-22

40 30 20 10 0 100

10

1 0.1 Aggregate Particle diameter (mm)

0.01

Fig. 1. Aggregate gradation portion for asphalt mixtures investigated.

The FEA software, ABAQUS™, was used to model the airfield flexible pavement and to assess the mechanical responses of the asphalt surface course. The asphalt mixtures analyzed in this study were compared based on the tensile strains at the top and bottom of the surface course. The airfield flexible pavement section constructed and tested at the FAA’s National Airport Pavement Test Facility (NAPTF) Construction Cycle – 1 [14] was modeled for this study. The pavement structure materials and material properties used in this study were the same as the structure tested at the FAA NAPTF (Table 2). The flexible pavement structure modeled in ABAQUS™ consists of the four layers with each layer assumed to be perfectly bonded. The pavement cross-section is comprised of a 127-mm asphalt surface placed over a 127-mm P-401 asphalt treated base layer. These bituminous layers are then modeled over 216 mm of P-209 crushed aggregate subbase layer and 2438 mm of a medium-strength subgrade. For the surface course, the material properties used as inputs to the FEA were the mixtures’ dynamic modulus, Poisson’s ratio, and density which were all obtained through the laboratory testing. The loading footprint of a typical A340 aircraft with a 1448 kPa tire pressure was applied to the flexible pavement model. 2.4. Finite element model

Fig. 2. Overlay tester in the AMPT and test configuration.

The finite element model (FEM) mesh developed for this study had the following dimensions: 14-m length (x-direction), 3-m height (y-direction), and 3-m width (z-direction), modeling the properties of the NAPTF test section. The model featured 3D-reduced integration elements (C3D8R) and was reduced in size by the use of symmetry and asymmetric boundary conditions. A finer mesh was created near the loads to capture the most significant stress and strain gradients. The boundary conditions also have a significant influence on response predictions and the model was constrained along the bottom in all directions and on the sides to restrain its movement in the x- and z-directions. The pavement layers were modeled as linear elastic. A standing aircraft load was modeled by applying pressure loads to four rectangular contact areas with uniform tire pressure of 1448 kPa (representative of an A340 aircraft). A typical aircraft wheel imprint is of elliptical shape; however, according to Huang [20], creating a rectangular element with equivalent contact area is a valid assumption and thus, the wheel imprint was modeled to be 0.3 m by 0.5 m. The FEM was also run simulating a dynamic wheel load. The dynamic load was simulated by moving the four wheel loads across the surface using the frequency and amplitude features within the ABAQUS™ analysis tool to represent a slow moving (4.8 km/h) aircraft on an apron or taxiway [16]. The model was validated by comparing the mechanical responses from a simplified version of the flexible airfield pavement analyzed in this study to those from a closed form solution, accomplished by conducting an elastic layer analysis using KENLAYER [16,20]. The validation resulted in stresses, strains, and deflections generated by both analysis tools that were within 4% error. This error could have been decreased by applying a finer mesh to the FEM. However, the computational time would have increased significantly.

3. Results Fig. 3. Typical specimen dimensions. Specimens were prepared in accordance with AASHTO T-312 [2]. For this study, loose mixtures were re-heated at 145 °C (and 135 °C for the WMA–RAP mixture) and compacted using the SGC. Specimens used for overlay testing meet the relative

3.1. Laboratory testing results Fig. 4 was generated to illustrate a typical overlay test curve for the FAA P-401 mix. The plot is divided into three basic phases or

816

M.C. Guercio et al. / Construction and Building Materials 95 (2015) 813–819

Fig. 4. Typical AMPT overlay test results for FAA P-401 asphalt mix. Fig. 5. Dynamic modulus test (AMPT) and specimen dimensions. Table 3 Overlay test results. Asphalt mixture

Sample no.

Initial load (kN)

Final load (kN)

Reduction (%)

Load cycles to failure

FAA P-401

1 2

3.614 3.408

0.923 0.915

83 81

1200 1200

WMA–RAP

1 2

2.412 2.414

0.756 0.444

93 93

728 640

BRIC

1 2

2.172 2.129

0.773 0.719

77 78

1200 1200

stages. The first phase is where the crack initiates and early crack propagation is observed. In this phase the load decreases rapidly as the crack starts to propagate through the specimen. At this stage the displacement increases to the minimum amount of 0.635 mm. The second phase is where late crack propagation occurs. This phase is monitored as a slow decrease in maximum load. This phase occurs before and up to the cycle when 93% of load (initial) reduction is reached. The third phase is when failure is detected. In this phase the crack has propagated completely through the specimen or 93% of load reduction has occurred or the specimen has reached the 1200th cycle. The overlay test correlates well with field fatigue cracking performance of asphalt mixtures used as a surface course. Mixtures with higher cycles to failure are considered desirable for fatigue cracking resistance [35,7]. Table 3 summarizes the results obtained from the overlay testing of each mixture. Overlay testing was performed for the FAA P-401, WMA–RAP, and BRIC mixtures. Insufficient material samples were available to run the overlay test on the HMA PG70-22 and HMA PG82-22 mixtures. The data collected through the test include the initial and final loads; percent initial load reduction; and, cycles to failure. As mentioned previously, the main parameter for comparison among mixtures is their number of cycles to failure. As specified in the test specifications, the test was stopped at either a 93% load reduction or at 1200 cycles, depending which condition was met first. As shown in Table 2, the FAA P-401 and BRIC mixtures exceeded the threshold limits and reached 1200 cycles before the initial load was reduced by 93%. All mixtures performed adequately by achieving more than 300 cycles to failure. Some variation was noted among the samples for the WMA–RAP mixture which prompted investigation into the repeatability and variability of the overlay test. These variations were attributed to the inability to trim the test specimens exactly the same (within ±2 mm), as well as slightly different air voids (within ±0.5%) among the laboratory-compacted samples. A coefficient of variation (COV) was calculated for the set of data associated with the WMA–RAP (9.1%) mixture. As reported in [24], the coefficient of variation of asphalt mixtures tested with the Overlay Tester typically falls in the range of 10–25%, which

Table 4 Dynamic modulus data used as inputs to FAARFIELD. Asphalt mixtures analyzed

Dynamic modulus, |E⁄| (MPa)

Baseline (FAA P-401)

WMA– RAP

HMA PG82-22

HMA PG70-22

BRIC

1996

1935

2128

1822

4901

indicates that the variations observed in this study are within the reportable ranges. The time–temperature dependent dynamic modulus (|E⁄|) is a fundamental bituminous material property necessary to predict pavement distresses such as rutting and fatigue cracking. The |E⁄| is the ratio of the dynamic stress over the recoverable axial strain [34]. The asphalt mixtures included in this study were tested according to the AASHTO TP 79-11 [6] for |E⁄| at 25 °C to determine their behavior at the intermediate temperature. Loose mixtures were re-heated at 145 °C (135 °C for the WMA–RAP mixture) and compacted using the SGC to achieve 7 ± 1% air voids content. The BRIC mixture was compacted to achieve 97% density. Samples were then trimmed to the test sample size shown in Fig. 5. A total of three samples were prepared and tested for each asphalt mixture. The average |E⁄| values for each mixture obtained from laboratory testing are summarized in Table 4 and were used as the material property input to FAARFIELD. The laboratory |E⁄| value for the FAA P-401 mixture was higher than the default stiffness value of 1379 MPa assumed in FAARFIELD. The default stiffness value assumed in FAARFIELD was selected conservatively and corresponds to a pavement temperature of approximately 32 °C [13], while the laboratory-measured stiffness corresponds to a temperature of 25 °C. 3.2. Mechanical responses under static and dynamic aircraft loads The mechanical responses evaluated using ABAQUS™ are summarized in Table 5. A relative comparison of the tensile strains at the top (near the wheel path), etop, and bottom, ebottom, of the asphalt surface course was made for the FAA P-401 and the other mixtures. Table 5 shows that the HMA 82-22 produced lower static and dynamic ebottom (3% and 4%, respectively) and etop (3% for both static and dynamic aircraft loading) as compared to the FAA P-401 mix. The BRIC mixture exhibited the lowest tensile strains. For the BRIC mixture, the FEA predicted a reduction in ebottom of 41% for the static analysis and 49% for the dynamic analysis compared to the baseline FAA P-401 mixture. The BRIC mixture also produced lower etop (37% for static loading and 43% for dynamic loading) compared to the P-401 mix. This is indicative of better long-term fatigue

817

M.C. Guercio et al. / Construction and Building Materials 95 (2015) 813–819 Table 5 Mechanical responses from finite element analysis. Asphalt mixtures

3D FEA (ABAQUS™) Static load

ebottom (le) FAA P-401 (Baseline) WMA–RAP HMA PG82-22 HMA PG70-22 BRIC

1165 1183 1128 1218 684

Dynamic load

etop (le) 291 295 283 302 182

% variation from P-401

ebottom

etop

Baseline 2% 3% 5% 41%

Baseline 1% 3% 4% 37%

(a) Static Aircraft Load

(b) Dynamic Aircraft Load Fig. 6. Distribution of tensile stresses at the top of the surface asphalt layer from center to edge of wheel subjected to a (a) static aircraft load and (b) dynamic aircraft load.

cracking resistance to heavy aircraft loading. A slight increase in ebottom of 2% (for both static and dynamic loading) was produced by the WMA–RAP mixture. The HMA PG70-22 mixture produced a slight increase in ebottom of 5 and 7% for the static load and dynamic loading, respectively. The etop of the WMA–RAP and the HMA PG70-22 mixtures were estimated to increase approximately 1–2% (static-dynamic) and 4–6% (static–dynamic), respectively. Overall, the findings are consistent with the overlay test results obtained through laboratory testing of the same mixtures. Tensile stresses induced by the aircraft wheel loading at the top of the asphalt surface course are also significant contributors to top-down fatigue cracking [1]. The distribution of such stresses from the center to the edge of the wheel was estimated for each mixture through the FEA and is represented in Fig. 6. As shown in Fig. 6a, the static aircraft load produced horizontal compressive stresses under the wheel imprint. The compressive stress then changed to tensile at the edge of the wheel. This behavior is indicative of a possible development of top-down cracking at the edge of the wheel when a static aircraft load is applied to the flexible pavement [1]. Stresses obtained using the FEA for the P-401 mixture were compared to the tensile strength measured in the indirect tension testing (IDT) using the AASHTO T322 [3], in order to validate the model. The IDT test was performed at a temperature of 25 °C to determine the tensile stress at first failure. For the P-401 mixture, a tensile stress of approximately 0.35 MPa was estimated from the FEA, while the tensile strength of the same mixture was approximately 0.32 MPa in the IDT. This indicates that the FEA model estimated a tensile strength occurring at first failure

ebottom (le)

etop (le)

211 216 202 225 541

56 58 55 60 32

% variation from P-401

ebottom

etop

Baseline 2% 4% 7% 49%

Baseline 2% 3% 6% 43%

approximately 9% higher than the strength obtained through laboratory testing. Fig. 6b illustrates the distribution of tensile stress from the center to the edge of the wheel when the FE model was subjected to the dynamic aircraft loading condition. In this case, tensile stresses were produced throughout the wheel width and were propagated past the edge of the wheel. This translates to a lower top-down fatigue cracking potential under dynamic loading conditions for the mixtures analyzed in this study. It was also observed that the dynamic aircraft load produced lower compressive stresses at the top of the asphalt surface course compared to the static aircraft load and there were no tensile stresses at the edge. These results appear to indicate that the top-down cracking would be critical for standing or slow-moving loads as compared to fast moving loads. Based on the results from the FEA analysis, it was observed that the trends between mixtures for static and dynamic loading conditions were similar. For both loading conditions, the best performance in terms of tensile strain at the top (etop), tensile strains at the bottom (ebottom), and tensile stresses at the top of the surface course was observed in the BRIC mixture. Furthermore, the strains obtained from the dynamic analysis were lower compared to the strains obtained from the static analysis, even though the tensile stresses produced from the dynamic load were higher compared to the tensile stresses produced from the static aircraft load. 3.3. Analysis of relative pavement life The FAARFIELD version 1.305 program [11] was used in this study to compare the mixtures based on the predicted fatigue life. This program uses layered elastic theory to predict airfield pavement life. FAARFIELD failure models relate a computed structural response to the number of coverages (repetition of maximum strain) that a pavement structure can carry. For flexible pavement design, FAARFIELD uses the maximum vertical strain at the top of the subgrade and the maximum horizontal strain at the bottom of the asphalt surface layer to predict pavement structural life. The |E⁄| for each mixture obtained from laboratory testing (Table 4) was used as the material property inputs to FAARFIELD. Since the FAA P-401 mixture was considered to be the baseline, it was assigned a pavement life factor of 1.0 in order to scale the results. The performance associated with the modified mixtures was compared to the performance of the baseline mixture and the predicted results are shown in Fig. 7. Fig. 7 illustrates that the BRIC mixture was approximately predicted to have four times more fatigue life than the FAA P-401 mixture. The lower tensile strains predicted for the BRIC mixture (41% and 49% reduction for the static and dynamic analyses, respectively) resulted in a marked increase in pavement life as compared to the FAA P-401 mixture. The predicted lives of the WMA–RAP, HMA 82-22, and HMA 70-22 were comparable to that of the standard FAA P-401 mixture. The data showed that the relatively slight increases in tensile strains at the bottom of the asphalt surface course (up to 7%) in the static and dynamic analyses for the

818

M.C. Guercio et al. / Construction and Building Materials 95 (2015) 813–819 5.0

Predi cted Pave ment Life (facto r)

5. Conclusions and recommendations

4.5 4.0 3.5 3.0

Baseline (FAA P-401) Mixture

2.5 2.0 1.5 1.0 0.5 0.0 WMA-RAP

HMA PG82-22

HMA PG70-22

BRIC

Fig. 7. Relative predicted pavement fatigue life from FAARFIELD analysis.

WMA–RAP and HMA PG70-22 mixtures resulted in pavement fatigue life equivalent to that of the FAA P-401 mixture. According to a previous study, Guercio et al. [16] performed on the same mixtures, it was found that BRIC performed better than the FAA P-401 under high temperature conditions (52.5 °C) as well. At this high temperature, the WMA–RAP and HMA PG82-22 mixtures also performed similarly to the FAA P-401, in terms of stresses and deflections within the asphalt surface course. The HMA PG70-22 underperformed as compared to the FAA P-401 mixture at the higher temperature. Based on these analyses, a number of mixtures including the BRIC, WMA–RAP, and HMA PG82-22 exhibited a better or similar rutting and fatigue cracking performance compared to the standard P-401 mixture. 4. Summary of findings This study was conducted to evaluate and compare the mechanical responses of a number of modified asphalt mixtures subjected to heavy static and dynamic aircraft loads typical of aprons and taxiways. The responses were defined in terms of tensile strains at the top (for top-down fatigue cracking) and bottom (for bottom-up fatigue cracking) of the asphalt surface course. The findings from the FEA and FAARFIELD analyses are summarized as follows: 1. All of the mixtures used in this study exceeded the threshold limits for cycles to failure according to the TxDOT specifications. 2. Higher cycles to failure obtained through the overlay test resulted in lower tensile strains at the bottom of the surface course. The results from the overlay test were found to be closely related to the fatigue performance of the modified asphalt mixtures. 3. The BRIC and HMA PG 82-22 mixtures were predicted to have a better fatigue cracking performance than to the FAA P-401. Both mixtures had lower strains and better fatigue life performance than the FAA P-401. 4. The WMA–RAP and HMA PG70-22 mixtures were predicted to produce a slight increase in strains and to have relatively similar performance lives to that of the FAA P-401. 5. For all the mixtures analyzed in this study, the strains obtained from the dynamic analysis were lower compared to the strains obtained from the static analysis. 6. For all the mixtures analyzed in this study, the static aircraft load produced tensile stresses at the edge of the aircraft wheel. The dynamic aircraft load produced only compressive horizontal stresses near the wheel path. The dynamic aircraft load produced lower compressive horizontal stresses at the top of the asphalt surface course compared to the static aircraft load. 7. For both loading conditions, the best performance in terms of tensile stresses was produced by the BRIC mixture followed by the HMA PG82-22, FAA P-401, WMA–RAP, and HMA PG70-22 mixtures.

This study was conducted to determine whether or not the fatigue cracking performance of the mixtures analyzed in this study was comparable to the performance of the FAA P-401 mixtures. As the highway asphalt pavement industry is moving forward with the implementation of more durable and sustainable materials, it is of interest to determine whether the airfield sector can benefit from the characteristics of the same mixtures. Based on this study, several conclusions can be drawn regarding the bottom-up and top-down fatigue cracking behavior of the asphalt mixtures analyzed in this study. A combined approach with FEA and layered elastic analysis was effective for discerning the fatigue cracking performance of a number of asphalt mixtures used as the surface course in flexible airfield apron and taxiway pavements. Based on the overlay test results, it was found that all of the mixtures analyzed in this study exceeded the minimum threshold value and provide comparable fatigue performance when compared to the baseline (FAA P-401) mixture. It was also determined that the BRIC mixture outperformed the baseline mixture possibly for bottom-up and top-down fatigue cracking. Overall, the overlay test was found to correlate well with the mixture fatigue cracking performance and can be used to evaluate new mixture applications in flexible taxiway and apron pavements where fatigue cracking is the primary distress. Based on these findings, it was concluded that there is a potential for the airfield industry to consider emerging mixture technologies, such as WMA–RAP. Also, the use of BRIC mixture as surface lift in airfield taxiways and aprons should be further investigated since this mixture appears to exhibit the potential for improving surface fatigue cracking performance. However, the potential for decreased rutting performance may be present in BRIC mixtures and should be verified using a loaded wheel test. It is recommended to use this study as basis of a more in-depth analysis of WMA–RAP and BRIC mixtures used as surface lift to investigate the effects of a larger variety of aircraft wheel configurations. Also, to confirm the results of this study and further validate the analysis, it is recommended that effort be initiated to compare the mechanical responses obtained from the FEA models with the actual field data and performance. Acknowledgements The research reported was performed under Region II University Transportation Research Center (UTRC) by the Department of Civil and Environmental Engineering at Rowan and Villanova Universities. Mixtures were obtained from the Delaware Department of Transportation, New Jersey Department of Transportation, Rhode Island Department of Transportation, and FAA NAPTF. Laboratory testing of the asphalt mixtures was performed at the South Jersey Tech Park. The authors would also like to acknowledge the Federal Aviation Administration and U.S. Army Corps of Engineers for providing information related to this study and the airfield professionals who shared their insight on the state-of-the-practice of WMA–RAP mixtures in airfield pavements. References [1] Al-Qadi, I. and Wang, H. Evaluation of Pavement Damage Due to New Tire Designs. Publication FHWA-ICT-09-048. Illinois Department of Transportation, 2009. [2] American Association of State Highway and Transportation Officials. AASHTO T312: Preparing and Determining the Density of Hot-Mix Asphalt (HMA) Specimens by Means of the Superpave Gyratory Compactor. Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 25th Edition, AASHTO, Washington, D.C., 2005. [3] American Association of State Highway and Transportation Officials. ASSHTO T 322: Determining the Creep Compliance and Strength of Hot-Mix Asphalt

M.C. Guercio et al. / Construction and Building Materials 95 (2015) 813–819

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11] [12]

[13] [14] [15]

[16]

[17]

[18]

(HMA) Using the Indirect Tensile Test Device. Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 30th Edition, AASHTO, Washington, D.C., 2007. American Association of State Highway and Transportation Officials. AASHTO T 209: Theoretical Maximum Specific Gravity and Density of Hot Mix Asphalt (HMA). Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 30th Edition, AASHTO, Washington, D.C., 2010. American Association of State Highway and Transportation Officials. AASHTO T 166: Bulk Specific Gravity of Compacted Hot-Mix Asphalt Using Saturated Surface-Dry Specimens. Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 25th Edition, AASHTO, Washington, D.C., 2005. American Association of State Highway and Transportation Officials. AASHTO TP 79–11: Determining the Dynamic Modulus and Flow Number for Hot Mix Asphalt (HMA) Using the Asphalt Mixture Performance Tester (AMPT). Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 31st Ed. AASHTO, Washington, D.C., 2011. T. Bennert, W. Worden, M. Turo, Field and Laboratory Forensic Analysis of Reflective Cracking on Massachusetts Interstate 495, Transportation Research Record: Journal of the Transportation Research Board, No. 2126, Transportation Research Board of the National Academies, Washington, D.C., 2009, pp. 27–38. R. Bonaquist, NCHRP Report 691: Mix Design Practices for Warm Mix Asphalt, Transportation Research Board of the National Academies, Washington, D.C., 2011. L.A. Cooley, R.C. Ahlrich, R.S. James, Implementation of Superpave Mix Design for Airfield Pavements. FAA Worldwide Airport technology Transfer Conference, Atlantic City, New Jersey, USA, 2007. E.O. Ekwulo, D.B. Eme, Fatigue and rutting strain analysis of flexible pavements designed using CBR methods, African J. Environ. Sci. Technol. 3 (12) (2009) 412–421. Federal Aviation Administration (FAA). Airport Pavement Design Software version 1.305, 2009. Federal Aviation Administration. AC 150/5370-10G – Standard for Specifying Construction of Airports. Item P-401 Plant Mix Bituminous Pavements. http:// www.faa.gov/documentLibrary/media/Advisory_Circular/AC_150-5370-10G.pdf. Federal Aviation Administration. Airport Pavement Design and Evaluation. Advisory Circular AC 150/5320-6E, U.S. Department of Transportation, 2009. Federal Aviation Administration. Database. http://www.airporttech. tc.faa.gov/naptf/databases.asp. Accessed November 10, 2012. R.B. Freeman, J.K. Newman, R.A. Ahlrich, Effect of Polymer Modifiers on DenseGraded, Heavy-Duty Mixtures. Presented at the 8th International Conference on Asphalt Pavements, Seattle, Washington, 1997. M.C. Guercio, Y. Mehta, L. Myers McCarthy, Mechanical responses and viscoelastic properties of asphalt mixtures under heavy static and dynamic aircraft loads, Transportation Research Record: Journal of Transportation Research Board, No. 719, Transportation Research Board of the National Academies, Washington, D.C., 2014, pp. 88–95. E.Y. Hajj, A.E. Sebaaly, P. Kandiah, Use of Reclaimed Asphalt Pavements (RAP) in Airfield HMA Pavements. AAPTP Project No. 05-06, Airfield Asphalt Pavement Technology Program, Auburn, AL, 2008. K.R. Hansen, A. Copeland, Annual Asphalt Pavement Industry Survey on Recycled Materials and Warm-Mix Asphalt Usage: 2009-2012, National Asphalt Pavement Association (NAPA), 2013.

819

[19] J. Harvey, T. Lee, J. Sousa, J. Pak, C.L. Monismith, Evaluation of fatigue and permanent deformation properties of several asphalt-aggregate field mixes using Strategic Highway Research Program A-003A equipment, Transport. Res. Rec. 1454 (1994) 123. [20] Y.H. Huang, Pavement Analysis and Design, Prentice Hall, Englewood Cliffs, New Jersey, 2004. [21] M.M.J. Jacobs, M.J.A. Stet, A.A.A. Molenaar, Decision Model for the Use of Polymer Modified Binders in Asphalt Concrete for Airfields. Presented at the Federal Aviation Administration (FAA) Worldwide Airport Technology Transfer Conference, 2002. [22] V. Jha, Sensitivity Analysis and Calibration of the Alligator Cracking Model in the Mechanistic-Empirical Pavement Design Guide Using Regional Data. Master’s Thesis, Rowan University, Glassboro, N.J., 2009. [23] D.Y. Lee, J.A. Guin, P.S. Kandhal, R.L. Dunning, A Literature Review: Absorption of Asphalt into Porous Aggregates. Prepared for Strategic Highway Research Program Contract A-003B, 1990. [24] G. Loria-Salazar, Reflective Cracking of Flexible Pavements: Literature Review, Analysis Models, and Testing Methods. Thesis, University of Nevada, Reno, 2008. [25] C.L. Monismith, R.G. Hicks, F.N. Finn, J. Sousa, J. Harvey, S. Weissman, J. Deacon, J. Coplantz, G. Paulsen, Permanent deformation response of asphalt-aggregate mixes, Transportation Research Record: Journal of the Transportation Research Board, Report No. SHRP-A-415, Transportation Research Board of the National Academies, Washington, D.C., 1994. [26] D.E. Newcomb, K.R. Hansen, Mix type selection for perpetual pavements, Proceedings of International Conference on Perpetual Pavements. CD-ROM, Ohio University, Columbus, 2006. [27] D. Newcomb, R. Wills, D.H. Timm, Perpetual Pavements A Synthesis. APA 101, National Asphalt Pavement Association, Asphalt Institute, State Asphalt Pavement Associations, MD, 2002. [28] K. Newman, Polymer-Modified Asphalt Mixtures for Heavy-Duty Pavements: Fatigue Characteristics as Measured by Flexural Beam Testing. Presented at the FAA Worldwide Airport Technology Transfer Conference, Atlantic City, New Jersey, 2004. [29] B.D. Powell, B. Hurley, Quality Improvement Series 125: Warm-Mix Asphalt: Best Practices, National Asphalt Pavement Association, Lanham, MD, 2007. [30] F.L. Roberts, P.S. Kandahl, E.R. Brown, D. Lee, T.W. Kennedy, Hot Mix Asphalt Materials, Mixture Design, and Construction, 2nd Edition., National Center for Asphalt Technology, National Asphalt Pavement Association Education Foundation, Maryland, 1996. [31] J.E. Shoenberger, T.A. Demoss, A. Hot-Mix, Recycling of asphalt concrete airfield pavements, Int. J. Pavement Eng. 6 (1) (2005) 17–26. [32] Simulia. ABAQUS™ 6.12. http://www.tu-chemnitz.de/projekt/abq_hilfe/docs/ v6.12/. Accessed December 12, 2012. [33] Texas Department of Transportation. Test Procedure for Overlay Test. Tx-248F, 2014. [34] F. Zhou, T. Scullion, Overlay Tester: A Rapid Performance Related Crack Resistance Test. FHWA/TX-05/0-4467-2, Texas Department of Transportation Research and Technology Implementation Office, 2005. [35] F. Zhou, S. Hu, T. Scullion, Development and Verification of the Overlay Tester Based Fatigue Cracking Prediction Approach. FHWA/TX-07/9-1502-01-8, 2007, pp. 90.