Influence of the roughness of sandblasted surface on the measurement of hardness by nanoindentation

Nanoindentation tests are often used to characterize the surface mechanical properties of materials. However, this technique lacks repeatability: when different zones of the same specimen are tested, there can be large differences between the nanoindentation curves. One of the main causes is the incorrect determination of the point of origin of the load versus indentation depth curves. In this study, a methodology aiming at solving this issue is applied to AU4G specimens that are blasted using different conditions. The results presented hereafter confirm the method robustness. They show that roughness can prevent the determination of a unique mechanical response and that the determination of the origin is strongly linked with the surface roughness of the blasted specimens. The best correlation between the standard deviation of the error made on the origin determination and the root-mean-square roughness is obtained for an evaluation length equal to 15 mu m i.e. at the scale of the indentation imprints.