Influence of specimen size on fracture energy of asphalt concrete

The development of mechanistic-based design approaches has driven efforts to develop models that describe the cracking phenomena in asphalt concrete. The models require material properties to describe how a crack nucleates, initiates, and propagate in asphalt concrete. Asphalt concrete falls within a category of materials that are defined as quasi-brittle, which requires careful analysis of test data to ensure that the obtained properties are relevant. The fracturing of asphalt concrete involves complex phenomena that occur within a fracture process zone (FPZ) ahead of the crack tip. Within the FPZ, aggregates can slide along the crack face, bridge the crack face, blunt the crack tip, and the asphalt mastic can yield under high strain. This study seeks to extend the current understanding of fracture in asphalt concrete by: providing new experimental fracture data from specimens with a range of in-plane dimensions and thicknesses; studying the size effect of asphalt concrete fracture in the context of the classic size effect law; applying a boundary effect model to obtain the size-independent fracture energy parameter, and; demonstrating how the current results can be used in simulations of pavement fracture using non-linear fracture mechanics. Experimentally-determined fracture energy using routine data interpretation methods produce values which are sensitive to specimen dimensions, particularly the length of the crack ligament and specimen thickness. A statistical analysis of data demonstrated significant differences in fracture energy for changes in fracture ligament lengths as little as 56 mm and for thickness variations as small as 25 mm. The size effect in asphalt concrete is found to adhere to a size effect law. The boundary effect model, originally developed for Portland cement concrete, is applied to asphalt concrete to obtain a size-independent estimate of material fracture energy.