Stillinger-Weber potential for wurtzite InN

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

I am trying to simulate wurtzite InN system with Stillinger-Weber potential (parameters taken from Lei et. al., “Core structures of the a-edge dislocation in InN”, Applied Physics Letters 90, 111901 (2007)) using LAMMPS May 14, 2012 release. I created the SW table following that of GaN that comes with LAMMPS, but somehow I can not match the lattice parameter, let alone other properties. There is huge compressive stress (~70 GPa) in the system with the reported lattice parameter.

I used my potential table creator script to reproduce GaN.sw and it did very well.

GaN and InN are both wurtzite in my case (different lattice parameter and c/a). Since I could reproduce GaN properties, I think my crystal maker code is also doing fine (so is my lammps input script which just does a static minimization of the structure).

Now, I am somewhat confused as to what is going wrong.

The SW table for InN looks like:

InN potential: Lei et. al., APL 90, 111901(2007)

note that the parameters for this literature potential are pairwise

so that there are some flexibility in the way the

parameters can be entered. As one way, we assume that

lambda_ijk is equal to lambda_ik and eps_ijk is

equal to sqrt(lambda_ijeps_ijlambda_ik*eps_ik)/lambda_ik,

and all other parameters in the ijk line are for ik.

These entries are in LAMMPS “metal” units:

epsilon = eV; sigma = Angstroms

other quantities are unitless

2

Can anyone try this table out?

No one is likely to do that for you.
I would create a simple unit cell where you can calculate
the energy analytically, start with pure N or In, and see
if it agrees with the code.

Steve