Micro-scale moisture distribution and hydrologically active pores in partially saturated asphalt mixtures by X-ray computed tomography

Moisture distribution is the cause of premature deterioration in asphalt mixtures. However, quantifying of complex moisture distributions and their relationships to hydraulic loading patterns and pore structure remain challenging. This study is aimed to (1) quantify 2D moisture distribution within the pore structure of asphalt mixtures using X-ray computed tomography, and to (2) investigate the hydrologically active pore characteristics related to various hydraulic loadings in asphalt mixtures. A dense-graded asphalt mixture with a maximum aggregate size of 13.2 mm was prepared in the laboratory. X-ray computed tomography technology was used to capture moisture distribution within samples under hydrostatic and hydrodynamic pressures. The characteristics of moisture distribution, including volumetric moisture content, degree of saturation, and void number were quantified by comparing pore structure before and after hydraulic action. Analysis of variance indicated that hydraulic loading pattern, hydraulic gradient, vehicular velocity, and vehicular loading significantly influenced moisture distribution characteristics. Within the same sample, the asphalt mixtures under hydrodynamic pressure had lower void number and greater volumetric moisture content/saturation degrees than those under hydrostatic pressure, especially with higher loading frequency and amplitude, indicating that vehicular loading action aggravates moisture infiltration into pavement and increases the possibility of moisture damage. The hydrologically active pore structure was also analyzed using moisture distribution characteristics under varied levels of hydrostatic/hydrodynamic pressure. Interestingly, the hydrologically active pore sizes of asphalt mixtures under various hydraulic pressures differ from each other and provide various flow pathways for moisture transport, leading to the diversity of moisture distributions. These results provide a quantitative evaluation of microscale moisture distribution characteristics in asphalt mixtures, and these characteristics have significant implications for moisture transport modeling and moisture-induced damage prediction in field asphalt pavement.