Assessment of Existing Micro-mechanical Models for Asphalt Mastics Considering Viscoelastic Effects

Micromechanical models have been directly used to predict the effective complex modulus of asphalt mastics from the mechanical properties of their constituents. Because the micromechanics models traditionally employed have been based on elastic theory, the viscoelastic effects of binders have not been considered. Moreover, due to the unique features of asphalt mastics such as high concentration and irregular shape of filler particles, some micromechanical models may not be suitable. A comprehensive investigation of four existing micromechanical methods is conducted considering viscoelastic effects. It is observed that the self-consistent model well predicts the experimental results without introducing any calibration; whereas the Mori-Tanaka model and the generalized self-consistent model, which have been widely used for asphalt materials, significantly underestimate the complex Young's modulus. Assuming binders to be incompressible and fillers to be rigid, the dilute model and the self-consistent model provide the same prediction, but they considerably overestimate the complex Young's modulus. The analyses suggest that these conventional assumptions are invalid for asphalt mastics at low temperatures and high frequencies. In addition, contradictory to the assumption of the previous elastic model, it is found that the phase angle of binders produces considerable effects on the absolute value of the complex modulus of mastics.