This Is AuburnElectronic Theses and Dissertations

A Performance-based Evaluation of Asphalt Mixtures with High Recycled Asphalt Material Contents and Recycling Agents




Fortunatus, Mawazo

Type of Degree

PhD Dissertation


Civil and Environmental Engineering


It is agreed that the stiffness and rutting resistance of asphalt mixtures containing recycled asphalt material (RAM) increase with increasing their content; however, their cracking resistance and durability decrease due to the heavily aged asphalt binders in RAM. To mitigate these drawbacks, asphalt researchers and practitioners have explored the incorporation of recycling agents (RAs) in asphalt mixtures with high RAM content. The purpose of this study was to develop a fundamental understanding of the effect of high RAM content on the performance of asphalt binder and mixtures, while evaluating the effectiveness of RAs in restoring the properties of RAM binder and thus, improving the performance of high RAM content mixtures. Finally, a methodology to guide the usage of RAs to produce high RAM content mixtures with satisfactory performance was recommended. To accomplish the above objectives, three base binders, four RAM sources, three aggregate sources, four-bio-based RAs, and one petroleum-based RA were utilized. Recycled binder blends with different content of RAM binder alone or with RAs were evaluated using the Dynamic Shear Rheometer (DSR) to conduct Superpave performance grading (PG), multiple stress creep recovery (MSCR) test, linear amplitude sweep (LAS) test, and frequency sweep test to determine the Glover-Rowe (G-R) parameter. Bending Beam Rheometer (BBR) test was used to determine the low-temperature performance grade and Delta Tc (ΔTc) parameter, while the Fourier-Transform Infrared Spectroscopy with the Attenuated Total Reflectance (FTIR-ATR) was used for evaluation of oxidation products. For short-term aging, the Rolling Thin Film Oven (RTFO) was performed, and the Pressure Oven Vessel (PAV) was followed for long-term aging simulation. The mixtures were designed to meet the volumetric criteria and were subjected to short-term aging (STOA) and long-term aging (LTOA) of 4 and 6 hours, respectively at 135 °C. The mixtures were characterized for rutting resistance using the Hamburg Wheel Tracking Test (HWTT) and the Asphalt Pavement Analyzer (APA) test and moisture susceptibility using the HWTT. Both rutting resistance and moisture susceptibility were performed after STOA. The intermediate temperature cracking resistance and low temperature cracking resistance were evaluated after LTOA using the Indirect Tensile Asphalt Cracking Test (IDEAL-CT) and the Disk-Shaped Compact Tension Test (DCT), respectively. The dynamic modulus (|E*|) was conducted to analyze the viscoelasticity and aging properties of the mixtures after both STOA and LTOA. Binder testing results showed that the performance of the recycled binder blends was influenced by the RAM type, quantity, and RA type. Based on the DSR testing, the addition of RAM to the base binders improved their rutting and stiffening properties but had a negative impact on fatigue resistance and thermal cracking properties. The addition of RAs improved the fatigue and thermal cracking properties without being detrimental to the rutting properties of the recycled binder blends. Based on the G-R parameter, stiffness and embrittlement properties of the recycled binder blends increased with the addition of RAM binder, but they decreased when RAs were incorporated. Moreover, the ∆Tc showed that recycled binder blends with up 40% RAP binder met the minimum threshold of -5 °C after RTFO plus 40 hours of PAV that has been recommended to minimize the risk of age-related block cracking. According to the HWTT, all high RAM content mixtures had improved rutting resistance and were less prone to moisture damage. The APA suggested that RAs did not have a significant impact on the rutting resistance of the high RAM content mixtures. The fracture energy parameter from the DCT test showed that the control mixtures were not statistically significant different from the high RAM content mixtures with or without RAs. The IDEAL-CT results showed that the CTindex values of the control mixtures were significantly higher than all high RAM contents mixtures (with or without RAs). This indicates that mixtures with high RAM contents were expected to have lower intermediate temperature cracking resistance than the control mixtures. The G-Rm aging index showed that high RAM content mixtures with RAs had low or equivalent aging susceptibility as the control mixture. The results from this study were used to develop a methodology to guide the usage of RAs for production of high RAM content mixtures and materials from Wisconsin were used to demonstrate it. The first part of the methodology involves determination of optimum RA dosage using blending charts. Mixtures are prepared at desired proportions of RAM and aged accordingly, then performance testing of the mixtures in terms of rutting, intermediate and low temperature cracking resistance is conducted to evaluate if they meet the requirements. Further, analysis of IDEAL-CT parameters, suggested that the CTindex had moderate correlation with the post peak slope whereas the relationship with the fracture energy was very poor. In addition, preliminary thresholds for the post peak slope and fracture energy were suggested to facilitate compliance with the CTindex criterion. A limited evaluation using the IDEAL-CT test showed that a combination of RAs with a discount factor to account for the inactive RAM binder could improve the intermediate temperature cracking resistance of high RAM content mixtures. However, this finding needs to be further evaluated using more mixtures with different materials components.