Uncovering the deformation mechanisms in nanocrystalline metals through simulation

The numerical modeling of nanocrystalline metals provides a unique challenge. Due to the length and time scales associated with the problem, modeling must take place at the interface between the continuum and atomistic domains. Accordingly, an intertwining of atomistic, discrete dislocation, and continuum simulations were used to gain a better understanding of how these materials deform. Because grain boundaries play a key role in the behavior of these materials, a large portion of this work was focused on exploring the mechanical properties of individual boundaries using molecular dynamics simulations. More specifically, the mechanisms of grain boundary sliding, migration, and partial dislocation emission were examined over a range of mixed loading conditions, temperatures, and strain rates. As a first approximation, the strain rate sensitivity exponent measured in the molecular dynamics simulations was used to predict the stresses necessary to activate these mechanisms at experimental strain rates. In most cases, the values of the strain rate sensitivity exponent and the activation stress appear reasonable compared to the macroscopic values measured in experiments. At a coarser scale, continuum simulations were performed using the finite element method in an attempt to illuminate the relationship between grain boundary deformation mechanisms and the macroscopic response of the material. Calculations combining both intragranular crystal plasticity and grain boundary sliding did not reveal an inverse Hall-Petch effect when the grain boundary volume was considered negligible. The effect of grain size distribution was investigated on larger microstructures using parallel finite element simulations. When the average (volume weighted) grain size was held constant, the width of the grain size distribution was not found to significantly effect the macroscopic response. Finally, two separate discrete plasticity models were developed to more accurately represent deformation in sub 100nm grains. The first model was formulated within the finite element methodology and captured the non-local and discrete nature of deformation at this length scale by explicitly representing the deformation from each dislocation. Without sacrificing accuracy, the second model demonstrates the potential to perform million grain simulations while maintaining individual dislocation resolution.