Effect of Radial Inhomogeneity on Shear Properties of Asphalt Mixtures

Abstract

Radial inhomogeneity is often observed in superpave gyratory compacted specimens because of the rotational movement of the gyratory and the boundary condition imposed by the gyratory mold. The effect of the inhomogeneity on mechanical properties of asphalt mixtures is not known. To study this effect, laboratory specimens of asphalt mixtures with known levels of aggregate inhomogeneity were fabricated. Nondestructive X-ray computed tomography was used to scan cross-sectional images of the specimens. A statistical index was developed to evaluate the scanned images for radial inhomogeneity. The results clearly show that the index can distinguish between levels of inhomogeneity. By means of frequency sweep at constant height (FSCH) test, the shear stiffness and the fatigue susceptibility at test temperature of 25°C and the shear stiffness and rutting susceptibility at test temperature of 50°C were evaluated. The resistance of the material to cumulative shear deformation was measured using the repeated shear at constant height (RSCH) test at 50°C. The effects of radial inhomogeneity on the shear properties of the specimens were investigated at the two test temperatures. As indicated from the FSCH test, the shear modulus increases, the rutting susceptibility, and the fatigue susceptibility in thick layers decreases, and the fatigue susceptibility in thin layers increases with increasing radial inhomogeneity. The correlation between the shear properties from the FSCH test and the inhomogeneity index is greater at the test temperature of 25°C than at the test temperature of 50°C. Although the RSCH test is conducted at 50°C, the permanent deformations resulted from the test shows the highest correlation with the level of inhomogeneity. The permanent strain decreases significantly with the increase in the level of inhomogeneity. The results of this study indicate that predicting the performance of the same material in the field based on the properties of laboratory-made specimens will lead to overprediction of the field performance and as a result underdesign of pavements.