Issues with MEAM file format for Bismuth in LAMMPS

LAMMPS LAMMPS General Discussion
,
1

Hello,

I am currently trying to study the Bismuth system using MEAM potential. This can be found in the paper at the doi link:

https://doi.org/10.1088/1361-651X/ac095c

When looking into the meam file format, it has sc for the lattice type which is simple cubic and is apparently not supported in LAMMPS. I tried changing to rock salt using ‘b1’ as it also only has 6 neighbors, but lammps says it is an unrecognizable lattice type as well. I tried other lattice types but could not reproduce good results for the base bismuth structure doing this.

Also, I would like to analyze a large bismuth cell, so i took a simple cubic cell and deformed using lammps fix/deform option. When I tried to do a large cell with more than 2 atoms, it required a large change in neigh_modify one, but when I did, the error changed to Neighbor page size must be >= 10x the one atom setting. I am not sure how to adjust this, but I feel as though something is wrong here. It could be that the deformation is too large as the angles between axes are 57.4 degrees which I know may cause an issue for LAMMPS and the number of ghost atoms increased dramatically?

The current lammps version that I am using is LAMMPS (3 Mar 2020).

Thank you for reading.

2

This uses a modified version of LAMMPS (that should have been explained in the publication).
You need to contact the authors. They had announced to submit their modifications to be included into the official LAMMPS distribution, but as of yet have failed to do so.