Experimental investigation and finite element simulation of the effect of surface roughness on nanoscratch testing

Nanoscratch testing is a highly reliable method used to extract a variety of film properties. It is proven that many of the experimental factors can influence the obtained results, such as the probe tilt, the scratch depth, etc. On the other hand, the surface roughness of the samples is an important parameter in nanoscratch and other similar tests, including the nanoindentation test. Thus, the effect of surface roughness on both the nanoscratch experiments and finite element simulations has been investigated. By performing scratch tests on gold and copper films and carrying out the finite element (FE) simulations on the rough and smooth surfaces, the importance of surface morphology was studied. The results indicate that the FE models consisting of the films with rough surfaces can lead to more accurate results, compatible with the experimental data. It was also revealed that the impact of surface morphology on the extracted friction coefficient tends to lose its significance in higher scratch depths. However, comparing the experimental results and simulation data, at lower scratch depths due to the presence of van der Waals forces the FE simulations will not lead to precise results.