Abstract
This study focused on the application of a clustered micromechanical discrete-element modeling (DEM) technique to predict the uniaxial compression dynamic modulus of hot-mix asphalt (HMA) mixtures. Six HMA mixtures were made with two types of asphalts and two gradations for limestone and gravel aggregates. The dynamic modulus of HMA was determined at three temperatures by using the indirect tensile test (IDT). Similarly, a frequency sweep test was used to determine the shear complex modulus (G*) of the asphalt matrix (AM) at various temperatures. The master curves for both AM and HMA mixtures were generated at 25°C, and results were analyzed. A two-dimensional, synthetic, heterogeneous microstructure of HMA was reconstructed with scanned images of IDT specimens. Bulk characteristics were used to determine the microproperties of the AM and aggregate structure in order to conduct virtual compressive test simulations. The results indicated that the DEM simulation accurately predicted the dynamic modulus of AM and aggregates at 25°C through the quasi-elastic approach. However, the HMA moduli were underpredicted by the model for a range of AM moduli and loading frequencies at 25°C. Calibration factors for aggregate volume concentrations and particle point contacts were developed and have been discussed here.