Radiation Damage Simulations

Lammps users,

I created a splined BKS/ZBL potential for O-Si and O-O interactions for radiation damage in silica. I am also aware that Dr. Crozier has created a tabulated file where he added a repulsive core to the BKS potential but he mentioned that it might not be suitable for radiation damage simulations.

I am curious as to whether I also should add an Si-Si pair interaction as well. Note that the original BKS potential neglects the buckingham Si-Si interactions. I initially equilibrated quartz with the 2 aforementioned splined potentials and regular ZBL potential for Si-Si. However, this (predictably) produced a much stronger core due to Si-Si interactions that were not previously there, and thus the density of my system decreased by about 10%.

If I remove Si-Si ZBL potential, but still use splined potentials for O-Si and O-O, I get statistically the same results as just BKS at ambient conditions, which is what I want.

Once I start irradiating the material, can anyone speculate on the consequences of not having ZBL potential for Si-Si interactions? I am worried that the Si atoms may collapse onto one another. However, when I add the ZBL potential for Si-Si interactions, the structural properties of quartz suffer considerably.

Ultimately I have this potential so far:

O-Si = splined BKS and ZBL with Coulomb
O-O = splined BKS and ZBL with Coulomb
Si-Si = Coulomb

Any suggestions on a good way to proceed? Perhaps I should modify the ZBL potential for Si-Si such that it has no effect unless at higher temperature? I could spline BKS with ZBL for Si-Si pairs, but the BKS is 0. This would create a potential that is 0 until a certain point, in which there would be an increase and ultimately join ZBL potential.

Thanks for reading,


You should be able to find some clues in the Literature. If not, this is a question for your advisor not for the LAMMPS mailing list.