Computer simulations of crystal plasticity at different length scales

Abstract

Crystal plasticity has been an active research field for several decades. The crystal plasticity of the bulk materials has its key relevance in the industrial process. Besides, the plasticity of nano-sized materials becomes a topic attracting a lot of interest recently.

In the Part I of the thesis, molecular dynamics (MD) simulations were used to study the plasticity of small nanoparticles. Firstly, the coalescence process of Cu nanoparticles was explored. It was found that a peculiar type of five-fold twins in the sintered products were formed via an unseen before dislocation-free process involving a series of shear waves and rigid-body rotations. Secondly, a similar study on the heating of a single nanoparticle was conducted. The same dislocation-free shear wave mechanism was spotted again. In this mechanism, a cluster of atoms rearranges in a highly coordinated way between different geometrical configurations (e.g. fcc, decahedral, icosahedral) without involving dislocations. Thirdly, simulations on the sintering of many nanoparticles were performed, and the governing processes during the consolidation were discussed. The findings in this part of the thesis can provide some guidance for controlling the motifs of nanoparticles.

In Part II of the thesis, the emphasis was switched to the crystal plasticity at larger spatial and temporal scales. A dislocation density-based model was developed in our research group. This model employs a dynamics formulation in which the force on each group of dislocation density is calculated with the Taylor and mutual elastic interactions taken into account. The motion of the dislocation densities is then predicted using a conservative law, with annihilation and generation considered. The new dislocation density-based model was used in this work to simulate the plastic deformation of single crystals under ultrasonic irradiation. Softening during vibrations as well as enhanced cell formation was predicted. This is the first simulation effort to successfully predict the cell formation phenomenon under vibratory loadings.