Author: Shawn Shiangfeng Hung

Publisher:

Published: 2018

Total Pages:

ISBN-13: 9781085585194

The pavement community, including both agencies and industries, is moving toward more sustainable pavement designs and pavement network management. Increasing amounts of recycled materials, both reclaimed asphalt pavement (RAP) and recycled tire rubber, are expected to be used in new pavement construction projects in the future to reduce the use of virgin binder and aggregates. The main concern of using recycled materials in new asphalt pavement is the potential negative effect on the performance. Thus, the primary objective of this dissertation is to improve the current laboratory testing technologies and performance assessment approaches for characterizing the performance-related properties of asphalt mixes containing recycled materials and to improve understanding of how these properties affect the performance of asphalt pavements so that they can be designed and constructed better. A major challenge regarding the use of high RAP content mixes is the differences in the rheological properties of the virgin binder (mixes without RAP) and the blended binder (mixes with RAP). Traditionally, binder blending charts are used to determine the appropriate RAP content in asphalt mixes and the selection of virgin binder grade as part of the Superpave volumetric mix design procedures when RAP is incorporated in the mix. However, producing mixes based on blending charts that require testing of extracted and recovered RAP binders is expensive and hazardous. An alternative test approach for binder blending charts using fine aggregate matrix (FAM) mix testing is presented in this dissertation. The results demonstrated that the proposed approach could estimate the blended binder intermediate and low performance grading temperatures within ±3°C of the measured blended binder performance grading temperatures. Even though the proposed approach is not as accurate as the blending chart method (within ±2°C), it provides both cost and environmental benefits. Currently, the Superpave Performance Grading (PG) system cannot not be used to evaluate the performance-related properties of asphalt rubber binders produced using larger crumb rubber particles (maximum particle size passing 2.36 mm sieve) due to the limitations of parallel plate geometry. With the consideration of more open-graded or gap-graded rubberized hot mix asphalt (RHMA-O and RHMA-G) projects in the future, it is important to be able to perform Superpave PG testing on asphalt rubber binder and to establish performance-based contract acceptance criteria for the production of asphalt rubber binders. The test results indicated that the concentric cylinder geometry is an appropriate alternative geometry to parallel plates for quantifying the properties of asphalt rubber binders and specifically for assessing the high-temperature performance properties of binders containing crumb rubber particles larger than 250 [mu]m. Concerns have been raised with regard to incorporating reclaimed rubberized asphalt pavement (RRAP) into dense-graded new hot mix asphalt (HMA-DG) and RAP into new RHMA-G since the interactions between the virgin binder, age-hardened binder, and recycled tire rubber could considerably affect the rutting, fatigue cracking, and thermal cracking performances of new HMA-DG and RHMA-G. The fundamental differences between RAP and RRAP were identified and the performance of new mixes that contain these recycled materials were evaluated in this study. The experimental results showed that adding RRAP to HMA-DG mixes is ideal to resist rutting and low-temperature cracking based on the changes in mix stiffness. The HMA-DG mixes containing RRAP are better at resisting high tensile strain loadings than mixes containing RAP. In addition, adding RAP to RHMA-G mixes improves the rutting performance but diminishes the cracking performance, and potentially negating the benefits of selecting RHMA-G as an overlay to retard the rate of reflection cracking. Lastly, the effects of rest periods on asphalt fatigue performance considering asphalt thixotropy, non-linearity, self-heating, self-cooling, and steric hardening were also investigated in this research. The experimental test results showed that asphalt thixotropic softening and other biasing effects control the first 10 to 15 percent decrease in stiffness for unmodified binders and 15 to 35 percent decrease in stiffness for modified binders under cyclic loading, and this decrease in stiffness can be recovered with the introduction of rest periods. This means that most of the repeated loadings applied to test specimens within the thixotropic softening range do not caused any fatigue damage but only softening of the materials. Thus, by providing sufficient rest periods within the thixotropic softening range can effectively improve asphalt fatigue performance. Both the thixotropic softening range and the required time for thixotropic recovery (i.e., rest periods) need to be considered in asphalt fatigue test and mechanistic-empirical (ME) design for better evaluation of the true fatigue performance.