Abstracts for the March 15-18, 2009 Annual Meeting
EVALUATION OF AASHTO RUT TEST PROCEDURE USING THE ASPHALT PAVEMENT ANALYZER
Nam Tran, Randy West, Buzz Powell, Andrea Kvasnak
Abstract
High truck traffic, heavy axle loads, and high tire pressures have contributed to the demand for rut-resistant hot mix asphalt (HMA) on major highways. Therefore, there is a need for a test that can indicate the rutting resistance of mixes and that can be conducted in a timely manner during mix design processes or for quality control/quality assurance (QA/QC) purposes. For these reasons, the Asphalt Pavement Analyzer (APA) has been widely adopted for checking the rutting resistance of HMA during the mix design process in many states. Most state procedures for APA testing are based on AASHTO TP 63. This AASHTO procedure was updated in 2007. However, questions still remain related to the sample preparation, testing parameters, and rut depth measurement methods in the AASHTO procedure, as well as the suitability of the APA for screening mix designs and QA/QC purposes. A study was conducted at the National Center for Asphalt Technology (NCAT) to address some of these questions. The objectives of this study were to (1) determine the differences in APA rut depth measurements for two combinations of hose pressure and wheel load; (2) investigate the possibility of using automated rut depth measurements interchangeably with manual rut depths measured according to the revised AASHTO T 63 procedure; and (3) evaluate the suitability of the APA for screening mix designs and QA/QC purposes. This study utilized field cores and laboratory specimens compacted to Ndes and the target air voids of 7 percent using HMA mixtures sampled during the construction of 12 test sections at the NCAT Pavement Test Track in 2006. The APA tests were conducted at two combinations of hose pressure and wheel load. The laboratory measured rut depths were then analyzed and compared with performance data collected at the NCAT Pavement Test Track. This study found that it was difficult to achieve 7 +/- 0.5 percent air voids for APA specimens prepared in the field without prior experience with the mix. The difference between manual and automated rut depths measured according to the revised AASHTO TP 63 procedure was statistically significant in this study. For both combinations of wheel load and hose pressure, the APA rut depths for specimens compacted to the target air voids of 7 percent had reasonably good correlations with the field rutting performance. The revised AASHTO TP 63 procedure was the most suitable but still far from a perfect procedure for screening mix designs and QA/QC purposes.
MULTIAXIAL STRAIN RESPONSE OF ASPHALT CONCRETE MEASURED DURING FLOW NUMBER SIMPLE PERFORMANCE TEST
Nelson Gibson, Muhammed Kutay, Dan Keramat, Jack Youtcheff
Abstract
The objective of this research is to develop a more complete understanding of the mechanics of permanent deformation in asphalt concrete such that future methodologies for materials characterization and rutting performance prediction are guided by these principles. Currently, the Simple Performance Tester (SPT) and corresponding Flow Number test are well poised to impact the state-of-the-practice. Salient aspects of the test and phenomena associated with rutting were reviewed. A logical mechanistic methodology was followed to determine relevant multiaxial stresses associated with rutting. In addition, a non-contact image-based strain measurement technique was developed to provide the radial and axial strain occurring during the SPT Flow Number test.
Multiaxial strains measured under the application of two complementary stress states provided additional insight into the three dimensional mechanics of permanent strain growth in asphalt concrete. A single parameter like the Flow Number cannot provide this information. There are no guarantees that the Flow Number will be encountered in all mixtures and tests. The multiaxial characterization clearly shows that asphalt concrete mixtures can and do volumetrically compress or volumetrically dilate depending on stress state and density, sometimes before and after tertiary flow is achieved.
A means to more clearly separate the contribution of binder apart from the contribution of aggregate on permanent deformations was identified. Concepts of permanent strain path and permanent strain trajectory are utilized to evaluate two distinctive sets of mixtures; one having measured rutting from an accelerated pavement test and the other having a broader variation in air void content and aggregate gradation.
CONSTRUCTING THE DYNAMIC MODULUS MASTERCURVE USING IMPACT RESONANCE TESTING
Andrew LaCroix, Y. Richard Kim, Maryam Sakhaei Far
Abstract
This paper presents a laboratory study to refine the impact resonance (IR) test to measure the dynamic modulus (|E*|) and phase angle of hot-mix asphalt (HMA). Kweon and Kim (1) presented analytical solutions for the determination of the |E*| and phase angle from the IR test and verified the solutions at high reduced frequencies (i.e., low temperatures) using typical North Carolina mixtures. They recommended investigating a wider range of temperatures to further define the |E*| mastercurve. Also, they stated that shift factors must be known a priori to produce the mastercurve. The goals of this study are to investigate the IR test's ability to determine the |E*| and phase angle at a wider range of temperatures and verify the use of binder shift factors to shift the IR |E*| values. The experiments show that IR tests can be performed from 5 to 60 degrees Celcius to produce the higher reduced frequency portion of the mastercurve. The results show that using averaged or measured binder shift factors to shift |E*| data obtained from IR tests produces reasonable |E*| values at higher reduced frequencies. Also, the IR test performs well with different nominal maximum sized aggregates, binders, specimen dimensions, and operators. However, it is found that the accurate construction of the entire mastercurve requires another means of estimating |E*| values at low reduced frequencies (i.e., high temperatures). Adoption of Witczak's |E*| predictive equation for the |E*| values at 54.4 degrees C reveals that the accuracy of the predictive equation affects the mastercurve construction quite significantly. Finally, recommendations are made to improve the IR test procedure so that it is suitable for quality control (QC) and quality assurance (QA) of HMA mixtures.
PERFORMANCE EVALUATION OF CHIP SEALS WITH POLYMER-MODIFIED EMULSIONS
Youngsoo Kim, Jaejun Lee
Abstract
Chip seals provide a durable and functional pavement surface and serve as a highly economical highway maintenance option when constructed properly. Data and literature suggest that chip seal sections constructed with polymer-modified binders provide better initial and long- term performance and also extend the overall service life of pavements. This paper compares the performance of chip seals constructed with polymer-modified emulsion (PME) versus those constructed with non-modified emulsion. The overall performance evaluation is based on aggregate retention performance. Three kinds of emulsion (CRS-2, CRS-2P, and CRS-2L) are used to fabricate samples in the laboratory and in the field. Adhesion of the emulsions is examined using the Vialit test with various curing times and temperature conditions in the laboratory. To evaluate the aggregate retention performance of the chip seals, the Vialit test, flip- over test (FOT), and the third-scale Model Mobile Loading Simulator (MMLS3) are employed. The results from these tests indicate that the PMEs (CRS-2P and CRS-2L) enhance the aggregate retention performance, and do so more significantly in the early stages and at low temperatures. This improvement is due to the fast and improved adhesion of the PME and its ability to enhance the aggregate retention performance at low temperatures.
A NEW LABORATORY TEST FOR PREDICTING VERY EARLY CHIP SEAL PERFORMANCE
Scott Shuler, Anthony Stock
Abstract
The adhesive and cohesive strength of the emulsion residue used as the binder in a chip seal is directly related to when the chip seal can be opened to traffic after construction. This strength is usually judged subjectively during construction by experienced personnel. Unfortunately, this experience is often gained by trial and error, the error leading to vehicle damage when residues that have not gained sufficient strength release chips under traffic loads. This research was conducted to help eliminate the subjectivity involved in determining when a chip seal can be safely opened to traffic without undue chip loss. The study began with the hypothesis that the exisiing ASTM D7000 sweep test could be used for this purpose. The ASTM procedure was evaluated and it was determined the test would be adequate for relative comparison of different emulsions for the same aggregates, but not sufficient to evaluate different aggregates for the same or different emulsions. The reason for this is because D7000 uses a template to establish residue depth. Chip gradation is roughly controlled so embedment depth is relatively equal for different chips but variability remains high so predicting actual times to traffic are not feasible. Therefore, a revision of the D7000 test was done to eliminate these variables so that actual project materials could be used and time to trafficking could be estimated. Results indicate the amount of water remaining in the emulsion is directly related to residue strength, as expected. Therefore, by establishing the relationship between water content and chip loss in the revised sweep test the time required in the field before traffic is allowed on the fresh chip seal can be estimated in advance. Results indicate little correlation between emulsion particle charge and aggregate type and also that chip loss is directly related to aggregate moisture content upon embedment.
EFFECTS OF TEMPERATURE AND COMPACTION EFFORT ON FIELD AND LAB DENSIFICATION OF HMA
Robert Schmitt, Carl Johnson, Hussain Bahia, Andrew Hanz
Abstract
The objectives of this study are to evaluate the effect of compaction effort and temperature on densification of HMA in the filed and in the lab. The field study investigated the minimum limiting temperatures at which 92% Gmm field density can be achieved with commonly used compaction effort. The lab study was conducted to determine if a relationship similar to the field can be found for lab compaction using varying temperatures and compaction pressures. To accomplish these objectives, field testing and loose-mix sampling occurred on 22 unique construction projects, totaling 30 unique layers of HMA during the 2007 paving season. Field data recorded included nuclear density, core density, temperature, roller passes, roller type, and vibratory setting. Loose-mix samples from the construction projects were compacted in the Superpave Gyratory compactor at two pressure settings, 300 kPa and 600 kPa; and at three temperatures, 120, 90, and 60 degrees C. Analysis of field data found that factors affecting density gain in importance rank order were temperature of mat surface, number of roller passes, roller type, vibratory setting, and PG binder grade. The results from field data indicate that a density of 92% Gmm can always be achieved, however at lower temperatures, more roller passes are necessary. For lab compaction, pressure and temperature showed significant main effects and significant interactive effects. Using 300 kPa pressure yields a density (@ Ndes) about 1.8 % less than 600 kPa at a baseline temperature of 248 degrees F. The density is reduced by about 0.4% when compacting at 194 degrees F, and 2.4% at 140 degrees F. The results, in general, point out the possibility of optimizing the compaction process by understanding the role of temperature and pressure, which are mixture- type specific.
A LABORATORY STUDY ON REDUCTION OF THE HEAT ISLAND EFFECT OF ASPHALT PAVEMENTS
Rajib Mallick, Bao-Liang Chen, Sankha Bhowmick
Abstract
Heat islands are formed as a result of construction that replaces vegetation with absorptive surfaces. Air temperature rises as a result of formation of heat islands. One suggested method to reduce the emitted heat from asphalt pavement surfaces is to reduce the temperature of the surface by flowing a suitable fluid through the pavement. The heated fluid could then be used for different end applications. Laboratory experiments were carried out using compacted hot mix asphalt samples with quartzite and metagranodiorite aggregates. Pipes with different surface area were used to flow water through the samples, and the processes were modeled using finite element method. The results clearly show the feasibility of the proposed method, and indicate the beneficial effects of higher thermal conductivity of aggregates and larger surface area of pipes. Velocity and thermal profiles of water in the pipe inside asphalt pavement are analyzed, and the necessity of good contact between asphalt mix and fluid carrying pipe is illustrated.
INTERFACE SHEAR STRENGTH CHARACTERISTICS OF EMULSIFIED TACK COATS
Louay Mohammad, Seok Bae, Mostafa Elseifi, Joe Button, James Scherocman
Abstract
The objective of this study was to evaluate interface shear strength of emulsified tack coats under a wide range of testing conditions commonly encountered in field applications. Three types of emulsified tack coats, CRS-1, SS-1h, and Trackless, were considered at three application rates, 0.14, 0.28, and 0.70 l/m2. In addition, a "no tack coat" condition was included in the analysis. The effects of construction conditions such as wet (rainfall) and dusty conditions were also evaluated. Laboratory direct shear tests were performed at 25C under two confinement pressures, 0- and 138-kPa. To simulate these test conditions, cores were extracted from a full- scale test site that was designed and constructed using conventional tack coat application and paving equipment. Results of this analysis showed that the trackless tack coat produced the highest shear strength at the three application rates, and SS-1 and CRS-1 resulted in the medium and the lowest strength, respectively. Within the considered application rate range, it was difficult to determine the optimum residual application rate. This may be attributed to the highly- oxidized and coarse HMA surface at the selected field site that required greater tack coat rates than expected. While higher application rates may increase interface shear strength, excessive tack coat may migrate into the HMA mat during compaction causing a decrease in the air void content in the mix. The majority of the cases showed statistically significant difference between clean and dusty conditions. However, no significant difference was found between dry and wet conditions. The results presented herein were part of the National Cooperative Highway Research Program (NCHRP) Project 9-40 on the "Optimization of Tack Coat for HMA Placement."
(FOR PUBLICATION ONLY)
MICROSTRUCTURAL AND MICROMECHANICAL PROPERTIES OF FIELD AND LAB-COMPACTED ASPHALT MIXTURES
Zhanping You
Abstract
The main objectives of this paper are: 1) to study the microstructural of the air void effect under laboratory and field compaction patterns on asphalt mixture; 2) to investigate the air void effect of the mechanical properties of the asphalt mixtures under the two compactions using distinct element models (DEM). The laboratory and field compacted specimen was simulated using two-dimensional (2D) and three-dimensional (3D) DEM. The laboratory specimens and field cores asphalt mixture samples were scanned with X-ray computed tomography techniques to obtain the images of the microstructure of the air void levels. The laboratory specimens and field cores were used to investigate the air void effect. The laboratory compaction was used by Superpave gyratory compactor. Field cores were produced from the pavements compacted with the asphalt mixtures used in the laboratory compaction. In the simulation of mechanical properties of the asphalt, 2D and 3D distinct element models were prepared in order to evaluate the performance of the mixtures. In the 2D models, both vertical and horizontal-cut images of the 3D specimen were used to evaluate the different air void levels. In addition, the relationship between the mixture modulus versus the air void levels was investigated.
EVALUATION OF RAP IMPACT ON HOT-MIX-ASPHALT DESIGN AND PERFORMANCE
Hasan Ozer, Imad Al-Qadi, Samuel Carpenter, Qazi Aurangzeb, Geoffrey Roberts, James Trepanier
Abstract
This study investigates the reclaimed asphalt pavement (RAP) effect on hot-mix asphalt (HMA) volumetric and mechanical properties. An experimental program, including tests for measuring mixture and binder complex moduli, fracture energy, and moisture susceptibility, was conducted. Six different mixture designs were prepared with varying percentages of RAP material (0, 20, and 40%) and two different material sources from Illinois. Because RAP binder is believed to be the only factor contributing to stiffness changes in the mixtures, it is essential to determine RAP's binder contribution; in other words "working RAP binder", which affects the HMA stiffness and the mixing/compaction process. Control specimens and actual practice specimens were also prepared to serve as reference mixes. Control specimens included RAP materials (binder and aggregate) recovered using the Rotovap method and virgin materials. Control specimens were designed to simulate the presence of varying proportions of working RAP binder in an RAP mixture. Actual practice specimens were a combination of RAP and virgin materials (binder and aggregate). Complex modulus test was conducted on HMA to quantify the impact of the change in binder stiffness. Other parameters, such as changes in volumetric properties, aggregate structure, and binder diffusion into aggregate, are also important. The study found that the optimum JMF's asphalt content of the virgin HMA and HMA containing RAP is similar. The current assumption of 100% working binder does not need to be modified from a mix design point of view. The effect of binder diffusion into aggregate between virgin and RAP materials was manifested in the results of the complex modulus tests. In addition, fracture energy and moisture induced damage tests were conducted to investigate the impact of RAP materials on HMA susceptibility to cracking and stripping. The study concluded that using RAP materials improves moisture susceptibility; but increases HMA cracking potential.
SENSITIVITY OF RAP BINDER GRADE ON PERFORMANCE PREDICTION IN THE MEPDG
Jo Daniel, Ghassan Chehab, Dinesh Ayyala
Abstract
The main objective of this study is to assess the sensitivity of assumed binder grade on performance prediction of mixtures containing Reclaimed Asphalt Pavement (RAP). This was achieved by utilizing the Mechanistic-Empirical Design Guide (MEPDG) software to predict the rutting, fatigue, and thermal cracking distresses of a flexible pavement structure with a HMA surface layer containing RAP. Design runs were conducted at all three Levels of analysis for two climactic regions and select Levels for two additional regions. Only material properties were changed across the design runs, with all other design variables held constant. Results from the various runs shed light on the extent and manner that properties of RAP mixtures affect pavement distresses and performance at the three levels of analysis. Although the assumed PG grade for RAP mixtures has small impact on performance predictions for Level 1 analysis, its effect is significant for Level 2 and Level 3 analysis. For the cases examined in this study, design using MEPDG Level 1 analysis yields the least conservative performance compared to the other levels of analysis. Additionally, discrepancies between Level 2 and Level 3 predictions are observed and become more prominent as the difference between the assumed high and low temperature PG grade of the RAP mixtures increases. The study confirms that realistic and reliable performance prediction and estimation of pavement life using the MEPDG require careful selection of analysis level and accurate characterization of PG grade for RAP mixtures. The relative ranking in performance among the various mixtures is consistent for all input levels, conditions tested, and distresses examined. The ultimate merit of this study lies in the determination of critical and sensitive factors that need to be considered when incorporating RAP mixtures in pavement design, and their effect on various pavement distresses.
ANALYSIS OF HMA FATIGUE DATA USING THE CONCEPTS OF REDUCED LOADING CYCLES AND ENDURANCE LIMIT
Donald Christensen, Ray Bonaquist
Abstract
This paper presents two new and very useful concepts for inclusion in the continuum damage analysis of fatigue data on HMA mixtures. The first is the concept of reduced loading cycles. Reduced loading cycles can be used as a much simpler alternative to the continuum damage parameter S in developing damage functions for HMA mixtures. The second concept introduced in this paper is that of effective strain, which is the applied strain minus the endurance limit. Using effective strain allows for better collapse of damage curves generated at different temperatures and strain levels compared to the traditional approach. Furthermore, it appears at this time that using effective strain might very well eliminate the need to vary the continuum damage parameter during analysis of HMA fatigue data. The proposed approach to the analysis of fatigue data has potential applications not only in the laboratory characterization of HMA, but also in pavement design and in the modeling of pavement performance.
THE EFFECT OF LONG TERM LABORATORY AGING ON ASPHALT CONCRETE FRACTURE ENERGY
Andrew Braham, William Buttlar, Timothy Clyne, Mihai Marasteanu, Mugurel Turos
Abstract
A pavement ages over time from the effects of the environment and traffic loading. This study was conducted to examine asphalt mixture laboratory aging protocols from the standpoint of both mixture and binder physical properties that are believed to relate to various forms of pavement cracking. The new protocol, aging uncompacted mix at 135 C for 24 hours appeared to give more realistic results compared to the AASHTO R30 long-term aging procedure in regards to compliance and tensile strength for the unmodified MnROAD Cell 33 mixture. The results of this study suggest that the R30 procedure may not be severe enough on the basis of mixture fracture properties (using the Disk-Shaped Compact Tension Test), but appears to be conservative with respect to the aging of the binders on the basis of testing of the recovered binder through the Double-Edge Notched Tension Test, the Direct Tension Test, and the Bending Beam Rheometer. This study also found that the proposed 24-hour, 135 C loose mix aging procedure may be slightly conservative with respect to mixture fracture behavior, but overly conservative (too much aging) on the basis of testing of the recovered binders. This study highlights the potential danger in selecting an aging procedure or conditioning parameters based upon binder or mixture testing alone. It also recommends that more polymer modified mixtures be investigated, as it may be necessary to develop a separate aging protocol for unmodified and modified asphalt concrete mixtures.
CHEMISTRY AND EFFECTS OF POLYPHOSPHORIC ACID ON THE MICROSTRUCTURE, MOLECULAR MASS AND GLASS TRANSITION TEMPERATURES OF ASPHALTS
J-F Masson, Peter Collins, John Woods, Sladjana Bundalo-Perc, Jim Margeson
Abstract
Polyphosphoric acid (PPA) is used to modify asphalt in efforts to raise pavement upper service temperatures and to reduce or eliminate rutting. Given the expanding use of PPA, its chemical reaction with asphalt and its effect on binder properties are the subject of great interest. In this perspective, the goal of this work was to acquire an effective understanding of the chemical reactions and physico-chemical effects of PPA onto asphalt. To this end, the morphology, the molecular weight, and the glass transition temperatures of four SHRP asphalts modified with 0%, 0.5% and 1.0% PPA were measured. The results were interpreted based on known reactions of PPA with asphalt model compounds. The findings indicated that upon the reaction of asphalt with PPA, the natural segregation of maltenes and asphaltenes upon annealing is amplified and this affects their glass transition temperatures, which in turn affect both the true low and the high performance grades. The basis of this phenomenon resides in great part in the reaction of asphalt pyrrole groups and the disruption of the hydrogen bond network. After rheological testing, it was also noteworthy that the true high temperature PG correlated with the PPA concentration and the temperature at which tan d = 2.
CHARACTERIZATION OF ASPHALT BINDER RESISTANCE TO PERMANENT DEFORMATION BASED ON NONLINEAR VISCOELASTIC ANALYSIS OF MULTIPLE STRESS CREEP RECOVERY (MSCR) TEST
Eyad Masad, Chien-Wei Huang
Abstract
A significant emphasis has been placed in the asphalt community on development of a method to characterize the resistance of asphalt binders to permanent deformation. The multiple stress creep recovery (MSCR) test has been proposed as a means of accomplishing this objective. In this test, an asphalt binder is subjected to creep loading at different stress levels with recovery (unloading) periods between stresses. The current analysis method of the MSCR test uses the strain accumulated at the end of the test to derive an index describing the resistance of asphalt binders to permanent deformation. However, the accumulated strain is not due only to permanent strain; some of this accumulated strain is viscoelastic strain that might not fully recover depending on the duration of the unloading period. In order to ensure that asphalt binders are characterized based on the actual permanent strain at the end of the test, a method to separate the actual permanent strain (irrecoverable) from the viscoelastic strain (recoverable with time) is needed.
The challenge in separating the recoverable and irrecoverable components is that these two components occur simultaneously during loading, and the recoverable component can exhibit nonlinear behavior. This paper presents an analytical method to analyze the MSCR test results and determine the actual irrecoverable and nonlinear recoverable response. Subsequently, the irrecoverable strain is used to develop an index by which to evaluate the resistance of asphalt binders to permanent deformation. The analytical approach is corroborated by analyzing asphalt binders that have been used as part of the Accelerated Loading Facility (ALF) experiment of the Federal Highway Administration (FHWA). The new permanent deformation index shows excellent correlation with the performance of the asphalt binders in the ALF experiment.
RHEOLOGICAL AND CHEMICAL INVESTIGATION ON THE DAMAGE AND HEALING PROPERTIES OF BITUMINOUS BINDERS
Ezio Santagata, Orazio Baglieri, Davide Dalmazzo, Lucia Tsantilis
Abstract
In order to optimize the performance of flexible pavements, one of the key issues which has to be addressed is the characterization and selection of bituminous binders with respect to their capability of contributing to fatigue resistance. In such a context, there is currently the need of developing and validating test protocols and analysis procedures which may enhance the understanding and quantitative description of the two phenomena which underlie fatigue: microcrack damage and healing.
In this paper the Authors present the results obtained during an experimental investigation which addressed the abovementioned issues in the case of several unmodified bituminous binders sampled from different refineries. The objective of the study was to highlight the relationships which may exist between the composition and structure of the binders and their fatigue and healing properties. This required the combined use of thin-layer chromatographic analyses and specifically-developed rheological test protocols, which may be run in the oscillatory mode by using a standard dynamic shear rheometer.
It was observed that clear correlations can be found between the two sets of test results. Binder fatigue life is directly linked to the colloidal instability index, while the relative healing index, introduced to quantify the stiffness gain which occurs during rest periods, can be expressed as a function of the ratio between saturates and aromatics. Both relationships, to be validated by further investigations, may be especially useful in the preliminary screening of binders and in the consequent prediction of their expected fatigue properties.
PERFORMANCE BASED SPECIFICATION GUIDELINES FOR THE SELECTION OF BITUMINOUS- BASED HOT-POURED CRACK SEALANTS
Imad Al-Qadi, Shih-Hsien Yang, Elham Fini, J-F Masson, Kevin McGhee
Abstract
The long-term performance of pavements depends greatly on the quality and frequency of maintenance. Appropriate maintenance protects the pavement from deterioration, corrects deficiencies, and ensures safe and smooth riding. Crack sealing is practiced on a routine basis as preventive maintenance and as part of corrective maintenance prior to an overlay or a greater rehabilitation project. A timely and properly installed sealant adds several years of service life to the pavement at a relatively low cost. Therefore, the selection of an appropriate sealant in a maintenance project becomes an important issue. Current sealant selection is based on ASTM standards that consist of quality control tests, not of performance indicators. These standards do not consider the changes in mechanical properties due to aging or the differences in local service temperatures. Therefore, the main purpose of this study was to develop a systematic process to help users to select appropriate bituminous hot-poured sealants for pavement cracks and joints. This paper summarizes a four-year research project that included several tests: an accelerated aging test, a viscosity test performed at installation temperatures, dynamic shear rheometer (DSR) tests to assess tracking resistance in summer temperature, a crack sealant bending beam rheometer (CSBBR) and a crack sealant direct tension test (CSDTT) for cohesive properties at sub-zero temperature, and a blister test for adhesive properties.
PRACTICAL APPLICATION OF VISCOELASTIC CONTINUUM DAMAGE THEORY TO ASPHALT BINDER FATIGUE CHARACTERIZATION
Carl Johnson, Hussain Bahia, Haifang Wen
Abstract
The ability of the binder phase of an asphalt mixture to resist fatigue damage can have a profound effect on the service life of an asphalt pavement. An accurate and efficient test method for evaluating the fatigue characteristics of asphalt binder has thus far been elusive due to excessive time requirements or equipment limitations. This paper presents a new attempt to use existing testing procedures (the Dynamic Shear Rheometer) to estimate fatigue resistance in a relatively short period of time. The test involves subjecting binder specimens to a monotonic constant strain-rate shear testing at intermediate temperatures. To evaluate fatigue performance a parameter is derived from the test results by integrating the area under the stress-strain curve to the maximum stress value. Initial validation is carried out by testing binders used in the FHWA Accelerated Loading Facility (ALF). The results collected at the same temperature used in the ALF are shown to accurately rank the fatigue resistance of pavements tested in full scale. The data collected in the new test can also be used for a fundamental analysis procedure based on the Viscoelastic Continuum Damage (VECD) theory to estimate fatigue of binders under varying levels of traffic and pavement conditions. The procedure is modeled after the extensive work already published on mixtures. There are, however, challenges in applying the theory to modified binders that are believed related to the non-linear behavior of some modified binders. These challenges can prove to be critical in accurately applying the fundamental approach to modified binders.
MODELING OF ASPHALT MIXTURE LABORATORY AND FIELD COMPACTION USING A THERMODYNAMICS FRAMEWORK
Eyad Masad, Saradhi Koneru, Kumbanokam Rajagopal, Tom Scarpas, Cor Kasbergen
Abstract
The objective of this paper is to present a model, developed within the context of a thermomechanical framework, for the compaction of asphalt mixtures. The asphalt mixture is modeled as a nonlinear compressible material exhibiting time-dependent properties.
A numerical scheme, based on finite elements, is employed to solve the equations governing compaction mechanisms. Due to the difficulty of conducting tests on the mixture at the compaction temperature, a procedure was developed to determine the model's parameters from the analysis of the Superpave gyratory compaction (SGC) curves. A number of mixtures were compacted in the SGC using an angle of 1.25 degrees in order to determine the model's parameters. Consequently, the model was used to predict the compaction curves of mixtures compacted using a 2 degree angle of gyration. The model compared reasonably well with the SGC compaction curves.
Finite element simulations of the compaction of a pavement section using a roller compactor were conducted in this study. The results demonstrated the potential of the material model to represent asphalt mixture field compaction.
The developed model is a useful tool for simulating the compaction of asphalt mixtures under laboratory and field conditions. In addition, it can be used to determine the influence of various material properties and mixture designs on model's parameters and mixture compactability.
CONCEPTUAL PHENOMENOLOGICAL MODEL FOR INTERACTION OF ASPHALT BINDERS WITH MINERAL FILLERS
Ahmed Faheem
Abstract
Many studies focused on modeling the stiffening effect of filler on asphalt binder. However, the interaction between both constituents was always a challenge to address and include in such efforts. In this study, the authors provide a conceptual model for understanding the mechanism by which the filler stiffens the asphalt mastic. The model hypothesizes that the influence of the filler takes two modes; one within the diluted region and the other within the concentrated region. This model is capable of determining the influence of the filler in a quantitative manner using new measurements of cohesive strength. The tackiness (cohesion) of the mastic is used as a measure to determine the volume fraction of influenced asphalt. Examining 5 different fillers mixed with one asphalt binder, the conceptual model is validated. The measurement of the volume fraction of the influenced asphalt shows strong correlation with the filler interaction measure provided by the conceptual model. In this study the concept of fractional voids using the Rigden test is shown to provide good correlation with the average filler particle size as well as the stiffening rate of fillers, but only within the diluted region. However, within the concentrated region it was incapable of predicting the interaction of the filler with the asphalt binder.
PERPETUAL PAVEMENT RESPONSES UNDER VARIOUS TIRE CONFIGURATIONS: ACCELERATED PAVEMENT TESTING AND FINITE ELEMENT MODELING
Imad Al-Qadi, Hao Wang
Abstract
A challenge associated with the use of wide-base tires is the accurate quantification of pavement damage induced by these tires. In this paper, an accelerated pavement testing program was performed to compare the strain responses of full-depth pavement caused by conventional dual-tire assembly, the first generation of wide-base tires, and the new generation of wide-base tires. Results of the experimental program indicate that the new generation of wide-base tire (455/50R22.5) causes less fatigue damage potential than the first generation of wide-base tire (425/60R22.5). Considering the significant effect of tire-pavement interaction on pavement damage quantification caused by different tire configurations, a three-dimensional (3D) finite element (FE) model was developed using the measured 3D tire contact stress distribution. The FE model incorporates linear viscoelasticity of hot-mix asphalt (HMA) and continuous moving load utilizing implicit dynamic analysis. After the model was validated against field strain measurements, the calculated octahedral shear stresses were used to compare the surface cracking and HMA rutting potential under different tire configurations. Results showed that the wide-base 455 tire causes less octahedral shear stress at the upper part of HMA layer, probably due to the more uniform vertical contact stress distribution and less transverse tangential stresses. However, the wide-base 455 tire causes greater responses at deeper pavement depth, including tensile strain at the bottom of HMA and compressive strain at top of subgrade. These response differences diminish as pavement thickness increases. In addition, the relative fatigue damage potential caused by the wide-base tire in thin pavements could be reduced when considering the wandering effect and possible pressure differential in dual tires.
AN EVALUATION OF THE EFFECTS OF NONLINEAR LOAD-STRAIN BEHAVIOR ON MEPDG ANALYSIS OF FLEXIBLE PAVEMENTS
Senthilmurugan Thyagarajan, Nadarajah Sivaneswaran, Balasingam Muhunthan, Katherine Petros
Abstract
The Mechanistic-Empirical Pavement Design Guide (MEPDG) employs the JULEA layered elastic analysis procedure for the structural analysis of flexible pavements. JULEA is utilized at each time increment within the design period, to compute strains within the pavement structure in response to traffic loading and the computed strains are used to accumulate damage and distresses over the design period. To minimize computing time, the MEPDG makes the assumption that the computed strains are linearly proportional to the applied load and exploits this assumption to extrapolate strains from an 18-kip single axle load and the specified tire pressure to the entire load spectrum in the traffic composition. The load-strain linear proportionality assumption, however, holds true only if the contact area remains the same as load varies, resulting in similar variation in the contact pressure. However, for truck loading of interest, the contact (tire inflation) pressure remains within a narrow range and the contact area changes with axle load. This reality is violated in the load-strain linear proportionality assumption made in the current MEPDG procedure and this paper analyzes the effect of this assumption on MEPDG rutting predictions. This study quantifies the effect of the load-strain linearity assumption and the expected deviation in MEPDG distress predictions in flexible pavements. The MEPDG rutting models were coded and incorporated along with JULEA in to a 'stand-alone application' to conduct independent analyses for comparison with MEPDG predictions. The stand-alone application utilizes the MEPDG computer climatic and material properties. The stand-alone application was used to analyze pavement sections with consideration of constant tire pressure, representing the field conditions more closely, and constant tire contact area, assumed in the MEPDG, for strain computations. The distress values computed using the two criteria (constant tire pressure and constant contact area) showed higher deviations within top several inches (up to 6 inches deep). As expected, the deviation in calculated strains diminishes as the evaluation depth increases. The study revealed that the load- strain linear proportionality assumption has significant impact on the prediction of permanent deformation. A viable alternative has been identified by extrapolating the strain values from three axle loads (3Point extrapolation) instead of a single load as currently implemented in the MEPDG. The distress predicted from 3Point extrapolation matches well with those computed with constant tire pressure.
REFLECTIVE CRACKING: MODELING FRACTURE BEHAVIOR OF HOT-MIX ASPHALT OVERLAYS WITH INTERLAYER SYSTEMS
Jongeun Baek, Imad Al-Qadi
Abstract
This study uses three-dimensional finite element modeling to investigate the fracture behavior of hot-mix asphalt (HMA) overlays that resulted in reflective cracking. The investigated pavement consists of a 57-mm-thick HMA overlay over a 200-mm-thick jointed plain concrete pavement (JPCP). Using a bilinear cohesive zone model (CZM), the crack behavior is characterized by traction-separation law. Cohesive elements are inserted directly over the transverse joints, which are potential reflective cracking locations in HMA overlay. The viscoelastic and fracture properties of the HMA are obtained through laboratory tests. A transient moving load is applied over a transverse joint, and implicit dynamic analysis is conducted to obtain time-dependent responses of the HMA overlay pavement. To examine the performance of interlayer systems, sand mix and steel netting interlayer are added in the overlay as substitutive and supplemental layers, respectively. The mechanism of the interlayer systems on delaying reflective cracking is inspected by means of a degradation scalar to represent a degree of damage in the cohesive elements. These results are compared to a control HMA overlay section that has no interlayer system. In the control section, reflective cracking initiates in the middle of the leveling binder under a wheel path and propagates vertically to the surface as well as laterally to the outside of the wheel paths. The sand mix interlayer efficiently reduces reflective cracking in the leveling binder. Shear as well as tensile fracture energy of the sand mix affects the performance. Finally, the steel netting interlayer reduces reflective cracking significantly when properly installed. When steel netting was used, reflective cracking significantly reduced in the leveling binder due to a strong shear deformation support as well as high tensile strain compensation in the overlay. This also diminishes the potential of reflective cracking in the wearing surface.
(FOR PUBLICATION ONLY)
RESPONSE AND FATIGUE PERFORMANCE MODELING OF ALF PAVEMENTS USING 3-D FINITE ELEMENT ANALYSIS AND A SIMPLIFIED VISCOELASTIC CONTINUUM DAMAGE MODEL
B. Shane Underwood, Youngsoo Kim, Murthy Guddati, Senganal Thirunavukkarasu, Siddarth Savadatti
Abstract
This paper presents results from a study that uses an in-house developed finite element analysis program, FEP++. Finite element analysis yields ultimate flexibility for including viscoelasticity, stress-state dependence of unbound paving layers, damage or any other processes or mechanisms that are known to affect the behavior of asphalt concrete pavements. The advantage of using this in-house developed software is that expensive commercial packages, such as ABAQUS or ANSYS, are not necessary. In this work, the asphalt concrete layers are considered as linear viscoelastic and the unbound layers as linear elastic. The advantage of using this level of complexity is that it offers an improved representation of asphalt concrete pavements while using the same inputs that are required for the NCHRP 1-37A Mechanistic Empirical Pavement Design Guide. After using the finite element package to assess the impacts of wheel speed, temperature gradient, and material type on pavement response, attention turns towards an advanced mechanistic material model for predicting the fatigue response of asphalt concrete, i.e., the viscoelastic continuum damage (VECD) model. This model is characterized with mixtures from the Federal Highway Administration's Accelerated Load Facility (FHWA ALF) and is found to capture an underlying material property, the damage characteristic relationship. Finally, results from FEP++ simulations of the FHWA ALF pavements are combined, in a simplified modeling scheme, with the VECD model to predict the fatigue performance of these pavements.