Computational studies of structure and dynamics in amorphous materials

Abstract

This thesis presents a theoretical study of the shear stress relaxation in amorphous silicon network in 3D system and the microscopic origin of shear stress in inherent structure supercooled liquid mixtures in 2D and 3D glass forming alloys. We present the shear stress relaxation in an amorphous silicon network via single bond rotation. To understand the shear stress relaxation we have generated the distribution of shear stress and stress change in inherent structure in both Cartesian and rotated frames. We see the distributions are Gaussian with zero mean and the variance of the stress is weakly dependent on temperature. We have also generated stress autocorrelation function for a range of temperatures and we see shear stress relaxed through the transitions in the inherent structures. We calculate the participation ratio of stress change in the rotated frame. We see almost 20% particles contribute strongly in the small system for stress change. The participation ratio decreases with increasing system sizes. In molecular dynamic simulation of 2D and 3D glass forming mixtures, we generate the plots of distributions of individual particle shear stress in inherent structures to understand the origin of total shear stress. Both distributions appear to have Gaussian with zero mean and demonstrate that the variance of atomic shear stress is very weak dependence on the parent liquid temperatures. To understand how the local shear stress arises in the inherent structures we have analysed local packing of different local compositions in 2D and 3D liquids but we don’t see any significant correlation with shear stress on them. We then analysed shear stress with force network. We have found a strong correlation with maximum shear stress and force fluctuations in the force network. We also found that the shear stress exhibits a long range anisotropic correlation in the inherent structures of both liquids.