Discrete element analysis of recycled concrete aggregate responses during repeated load triaxial testing

Abstract

Abstract More effective and sustainable use of natural resources besides the mitigation of environmental impacts induced by their extraction could be achieved by reusing construction and demolition (C&D) materials. C&D materials are increasingly being used in pavement bases/subbases as a reliable alternative to virgin quarry materials. The behaviour of flexible pavements under dynamic loads conveyed from traffic load can be evaluated in the laboratory by the repeated load triaxial test (RLT). This research study aims to evaluate the macro- and micro-mechanical interaction and underlying mechanisms of the behaviour of unbound recycled concrete aggregate (RCA) materials under RLT test using discrete element method (DEM) and to compare the simulated results with experimental results. In this research a DEM numerical software was used to develop a novel stress-controlled test method for modelling the RLT test, using a flexible membrane boundary with spherical particles linked through linear contact model. During the simulation, the axial recoverable deformations were monitored to determine the resilient modulus (MR) at each of the 16 loading-unloading stress level sequences. Furthermore, the micro-mechanics of the RCA samples were investigated using anisotropy tensor of fabric, contact orientation, contact force network and coordination number (CN). Laboratory RLT tests were conducted to verify the robustness of the DEM model. In general, both quantitative and qualitative agreements were attained between the physical tests and the numerical DEM simulations. The results of the micro-scale analyses revealed that the response of the RCA sample during the cyclic test was almost elastic. In other words, the almost-elastic behaviour was not accompanied by a significant change in the material fabric, rather was accompanied by a variation in the magnitude of the contact forces without variation in the contact force network and CN.