Multi-scale Computer Simulations of Materials

Prelegent i afiliacja: 
Prof. Fei Gao Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, Michigan 48109, USA
wt., 2022-06-21 15:00 do 16:30

The recent developments of multiscale computer modeling to simulate materials behavior in an extreme environment, including ab initio calculations, self-consistent accelerated molecular dynamics, advanced kinetic Monte Carlo (KMC) methods, and phase field modeling, will be reviewed. I will firstly focus on the application of multiscale modeling to study microstructural evolution in multicomponent equiatomic alloys under irradiation and some interesting results will be highlighted. These simulations provide the detailed mechanisms on the generation and evolution of irradiation-induced defects in face-centered cubic MEAs and HEAs to understand the new mechanisms of their irradiation tolerance. In addition, an object kinetic Monte Carlo based long-time dynamics is developed to simulate void formation, stacking fault tetrahedron (SFT) nucleation and defect accumulation at experimentally relevant time-scale for the first time in multicomponent equiatomic alloys.

Secondly, a self-adaptive accelerated molecular dynamics (SAAMD) method, which has been developed previously, is used to model infrequent atomic-scale events. This approach is self-evolving and can be applied to the coupled motion of fast and slow dynamics. We applied SAAMD to study evolution of nanosize clusters and dislocation loops in Fe, including (1) a helium-rich He-V cluster migration by an interstitial-assisted mechanism; (2) helium bubble formation and growth. Finally, I will discuss how the atomic-level results can be used to inform mesoscale methods, such as phase field model, to study the growth of gas bubbles and their effects on the mechanical properties of nuclear fuels.