Thermally activated dislocation events significantly govern yield characteristics, especially temperature and strain-rate dependence of yield strength in metal plasticity. Generally being difficult to quantify directly, the entropic contributions are often approximated with empirical compensation rules to inform the activation energetics associated with a wide-range of dislocation driven plastic events, e.g. slip nucleation, cross-slips, kink-pair nucleation/migration and twinning, to name a few. Despite their popular use in crystal plasticity models, several recent investigations demonstrate significant anomaly with both empirical as well as harmonic approximations when entropic effects dominate defect pathways. In this talk, I will introduce a computationally simple predictive approach to quantify the rates of dislocation mediated events aided by atomistic simulation data. To exhibit the significance of entropic contributions and the approach’s potential impact, we discuss a key scenario concerning crystal plasticity i.e., the nucleation of dislocations from surface steps, where currently existing approximations dramatically fails to reproduce the nucleation rates observed in direct MD. By accurately quantifying the change in vibrational entropy along the minimum energy pathway, we elucidate the non-trivial effects of anharmonic kinetics which is otherwise intractable with the existing empirical and harmonic approaches ubiquitous in modeling materials.
Back to Workshop I: Increasing the Length, Time, and Accuracy of Materials Modeling Using Exascale Computing