Characterization of Atomic Surface Roughness in Nanometric Machining Molecular Dynamics (MD) Simulations

Background: The atomic surface roughness is very important in assessing the quality of high performance nano surfaces, in ultra-precision machining, and in silicon fabrication. The surface roughness of nano devices will invariably affect their quality and performance, so its understanding is very crucial. In this study, multi-pass nanometric atomistic simulations were employed in the characterization of this parameter to gain better insight. Methods: Multipass MD simulations were carried out to create nano surfaces by the nanomachining of copper workpiece with a diamond tool. The effects of machining parameters during the nanometric phenomena were investigated. The copper-copper interactions were modelled by the Embedded Atom Method (EAM) potential and the copper-diamond interactions were modelled by the Lennard- Jones (LJ) potential. The diamond tool was modelled as a deformable body and the Tersoff potential was applied for the carbon-carbon interactions. Results: The simulation results show two peaks and three valleys for all the depth of cut and the cutting velocity range used. These waviness displayed by the surface atoms contributing to the roughness, are due to the overlap caused by the consecutive passes, the tool geometry and the choice of the interatomic potentials used in the simulations. Conclusion: There is a noticeable variation of atomic surface roughness, Sa with machined thickness and machining velocity. The estimated Sa from the study, is in the range 0.189-0.345nm. The Sa increases initially up to 1.5nm depth of cut, then it decreases before showing another upturn. It seems there is no logical relationship between the depth of cut and the surface roughness. Furthermore, Sa increases and decreases for a certain range, as velocity increases. In conventional machining, the Sa should improve as the velocity increases. However, on the nanoscale, the parameters are very sensitive to small variations. This variation may either be due to size effects of the simulation model or some other factors. Also, the importance of adhesion is highlighted in the investigation of friction between the tool atoms and interacting workpiece atoms, as friction is shown to decrease with increase in the depth of cut. The importance of this study of atomic surface roughness and its clearer understanding can be useful in the assessment of atomically flat silicon.