Development of a New Methodology to Effectively Predict the Fracture Properties of RAP Mixtures

Abstract

The existing design method for RAP mixtures assumes virgin and RAP binders fully blend. However, full blending may not occur and the impact of partial blending on mixture cracking performance is still unclear. A previous study revealed that RAP gradation, which is not currently considered from the standpoint of binder blending, controls the distribution of RAP binder within a mixture and consequently affects cracking performance. The objective of this study was to develop a methodology that overcomes the uncertainty of blending and effectively predicts fracture properties of RAP mixtures. This methodology is based on the evaluation of the interstitial component (IC) of a mixture (i.e., the fine portion that governs cracking performance) by means of a direct tension test named ICDT test. Two RAP sources and four RAP contents were considered. ICDT specimens were produced by blending the fine portion of RAP and virgin aggregate with virgin binder in the same way and corresponding proportions as in RAP mixtures. Binder and mixture fracture properties from the previous study were used for comparison. Mixture and IC exhibited almost similar reduction in fracture energy density (FED) with increasing RAP content, whereas fully blended binder exhibited a less pronounced reduction. This indicated that IC better simulated the actual blending that occurred in mixtures. Mixtures with coarsely graded RAP (less RAP content in the IC) exhibited better fracture properties; thus, the key to satisfactory cracking performance appears to be minimizing the amount of RAP in IC. Consequently, the stiffening effect of RAP on the fine portion that controls cracking performance should be directly evaluated, instead of placing focus on the fully blended binder or the whole mixture. The ICDT test was proven to be a valuable tool to predict fracture properties of RAP mixtures.