Computational and experimental evaluation of hydraulic conductivity anisotropy in hot-mix asphalt

Moisture damage in asphalt pavements is one of the primary distresses that is associated with the disintegration of the pavement surface, excessive cracking and permanent deformation. Moisture damage is a function of the chemical and physical properties of the mix constituents, and the distribution of the pore structure (microstructure), which affects fluid flow within the pavement. This paper deals with the relationship between the hot-mix asphalt (HMA) microstructure and hydraulic conductivity, which has traditionally been used to characterize the fluid flow in asphalt pavements. Conventional laboratory or field measurements of hydraulic conductivity only provide information about the flow in one direction and do not consider flow in other directions. Numerical modeling of fluid flow within the pores of asphalt pavements is a viable method to characterize the directional distribution of hydraulic conductivity. A three-dimensional lattice Boltzmann (LB) fluid flow model was developed for the simulation of fluid flow in the HMA pore structure. Three-dimensional real pore structures of the specimens were generated using X-ray computed tomography (CT) technique and used as an input in the LB models. The model hydraulic conductivity predictions for different HMA mixtures were validated using laboratory measurements. Analysis of the hydraulic conductivity tensor showed that the HMA specimens exhibited transverse anisotropy in which the horizontal hydraulic conductivity was higher than the vertical hydraulic conductivity. Analysis of X-ray CT images was used to establish the link between fluid flow characteristics and the heterogeneous and anisotropic distributions within the pore structure.