COMPUTATIONAL AND EXPERIMENTAL CHARACTERIZATION OF BITUMINOUS COMPOSITES BASED ON EXPERIMENTALLY DETERMINED PROPERTIES OF CONSTITUENTS

The stiffness of asphalt concrete mixtures is characterized in terms of the dynamic modulus for designing the thickness of flexible pavements. The dynamic modulus value of asphalt concrete is either determined experimentally or predicted by using empirical, semi-empirical, analytical or computational micromechanics models. This study proposes to use a computational micromechanics model to predict the dynamic modulus of asphalt concrete mixtures based on the experimentally determined properties of the constituents in the heterogeneous microstructure. The model defines asphalt concrete mixtures as the composites of two different homogeneous isotropic components – the viscoelastic fine aggregate matrix phase and the elastic aggregate phase. Mechanical properties are determined by oscillatory torsional shear tests of cylindrical bars of fine aggregate matrix mixtures, and quasi-static nanoindentation tests of aggregates. A protocol is developed to mix-design and fabricate the Superpave gyratory compacted fine aggregate matrix mixture as a replicate of fine aggregate matrix phase of asphalt concrete mixtures in terms of binder content, air void content and specific gravity. The cyclic uniaxial compressive tests are computationally simulated based on the finite element method (FEM). The model uses the material properties of the two-dimensional microstructure that are obtained from digitally processed images of asphalt concrete mixtures. The results are compared with the experimentally determined dynamic modulus tests of the same cylindrical samples of asphalt concrete mixtures. FEM simulations of the dynamic modulus of rectangular microstructure agreed with the laboratory tests of cylindrical samples