Atomistic simulation of plasma interaction with plasma facing components in fusion reactors

Xue Yang, Purdue University


The interaction between plasma and fusion relevant materials is one of the critical issues in successfully using those materials in Tokamak reactors. This research uses molecular dynamics, kinetic Monte Carlo and binary collision approximation methods to model fusion relevant material bombarded by energetic particles to investigate retention, deposition, sputtering, erosion, blistering effects, diffusion, and so on. The deuterium bombardment of monocrystalline tungsten was modeled by LAMMPS code using Tersoff type interatomic potential. The deuterium trapping rate, implantation depth, and stopping time in 600-2000 K tungsten bombarded by 5-100 eV deuterium atoms were simulated. Irradiated monocrystalline tungsten became amorphous prior to deuterium cluster formation, and gas bubbles were observed. The formation of gas bubbles were caused by the near surface deuterium super-saturation region and the subsequent plastic deformation induced by the local high gas pressure. Tungsten irradiated by carbon and deuterium is also modeled by molecular dynamics simulation. The threshold for carbon induced tungsten physical sputtering yield is predicted as ~40 eV. Cumulative carbon irradiation of crystal tungsten reveals that the tungsten erosion is enhanced by high substrate temperature. Cumulative carbon induced tungsten sputtering yield matches both experimental and Monte Carlo results very well. Carbon pre-irradiated tungsten tends to trap more hydrogen and facilitate gas bubble formation. Simultaneous deuterium and carbon bombardment on crystal tungsten indicates that carbon-induced tungsten sputtering yield exhibits a maximum value when carbon ratio is around 20%. Tungsten surface binding energy is calculated by molecular dynamic and many-body potentials. Consistency in tungsten sputtering yield by beryllium bombardment between molecular dynamic and binary collision approximation using the new surface binding energy is achieved. The analysis of the sputtered tungsten angular distributions show that molecular dynamic accurately reproduced the [111] most prominent preferential ejection directions in bcc tungsten, while the distinct shapes by typical Monte Carlo codes is caused by the treatment of amorphous target. A kinetic Monte Carlo (KMC) algorithm based on experiment and first principle calculation have been developed to study the hydrogen diffusion on tungsten reconstructed (001) surface. The predicted hydrogen diffusion coefficients match the experimental values very well, and a diffusion coefficient formula as a function of temperature and hydrogen coverage was derived from KMC simulations.




Hassanein, Purdue University.

Subject Area

Engineering|Plasma physics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server