Comparison of molecular dynamics simulation methods for the study of grain boundary migration

In the present study, grain boundary (GB) mobility was determined by molecular dynamics (MD) simulations using two different techniques: the applied strain method and the adapted interface random walk method. The first method involves a driving force while the second method does not. Nevertheless, both methods led to essentially the same values of the GB mobility. This shows that the GB mobility is independent of the nature of the driving force, provided that it is low enough that the linear velocity–driving force relationship is properly sampled. The case studied here can be viewed as a validated reference case that can be used in future studies to test new techniques to determine the GB mobility. For this purpose we provide the full information about the interatomic potential we employed and the initial atomic configurations. Finally, we use the obtained results to discuss whether any existing MD simulation data agree with experimental data on pure metals. (paper)