Recycled asphalt shingle modified asphalt mixture design and performance evaluation

Abstract

The inclusion of recycled asphalt shingles (RAS) in asphalt mixtures has become increasingly common; however, the underlying design principles vary significantly by agency. The primary objectives of this study included: (1) evaluating the ‘binder availability’ concept for RAS mixtures through a carefully designed laboratory experiment; (2) demonstrating a balanced mixture performance testing approach for the design of RAS mixtures; and (3) evaluating the field data of RAS mixtures placed in the Midwest region of the U.S. Three asphalt mixture designs with RAS contents of 0, 2.5% and 5.0%, which were designed to have nearly identical volumetric characteristics, were investigated. The binder availability was determined to be approximately 100% in the two RAS mixtures considered. In addition, Hamburg wheel tracking and disk-shaped compact tension tests were conducted to evaluate the high- and low-temperature mixture performance. As expected, the addition of RAS significantly improved the rutting resistance. Disk-shaped compact tension (DC(T)) test results demonstrated that a soft base binder effectively permitted the design of thermal-crack-resistant RAS mixtures. Field investigations indicated that the performance of pavement surfaces containing RAS was similar to that of surfaces containing only reclaimed asphalt pavement or virgin materials. This study also highlights a performance-engineered mix design approach, which is currently being adopted by several agencies in the Midwest (e.g., Illinois Tollway, Missouri DOT, etc.) and can provide mix designers a reliable approach for designing innovative asphalt mixtures with higher recycling levels and a modern, heterogeneous composition. Furthermore, the proposed approach may prove to be a simpler, more mixture-centric alternative to the primary method suggested in AASHTO PP78-17, which recommends arbitrary VMA bumping plus binder extraction, recovery, and advanced binder testing. • Study aims at determining binder availability from RAS material in asphalt mixtures. • Virgin and RAS mixtures were designed with identical volumetric parameters and asphalt contents were compared. • Binder availability was determined to be approximately 100% for RAS mixtures. • Performance-based tests were used to characterize RAS mixtures. • Field observations of mixtures with RAS were reported.