Characterisation of the Roughness of Sand Particles

The surface of soil grains is not smooth especially when examined at growing smaller scales. In geotechnical engineering there is increasing evidence of the significant role of surface roughness on the micro-mechanical behaviour of particulate samples. Characterizing the particles roughness is therefore an essential and pivotal step in the investigation on its role. Previous studies on the surface roughness generally used the optical interferometer and the roughness was calculated from the roughness profile which was separated from the measured surface profile so that the features at larger scale were not taken into account for the roughness. However, the approach used in this separation procedure is not clear, in particular that to what scale should features that contribute to the roughness be considered, and it may influence the roughness measured. Most research has been performed on engineered surfaces, but when applied to natural sand surfaces which are complex and mostly curved in nature, more uncertainties arise. In this paper, an alternative method is proposed which characterizes roughness with the aid of the power spectrum of the whole surface profile of sand particles. Results from natural quartzitic sand (Leighton Buzzard sand) tested using a high-resolution optical interferometer are presented. Fractal analysis was involved in the characterization. It was found that the sand roughness could be adequately characterized by the power spectrum and the fractal dimension calculated from a cut-off length scale that was inferred from surface area estimation