The specific procedure for splining EAM with ZBL potential

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

I’m try to spline the pairwise part in a EAM/FS potential with the ZBL for close particle separation, after checked the related issues in the mailing list, I think it can be done like : first, write a code myself to truncate both EAM (write out by pair_write) and ZBL potential; then connect them use a splining function, and keep the energy and force continuous by choosing proper splining constants. Finally, this modify pairwise potential need to be transformed back into DYNAMO format and replace that part in original EAM/FS file.

I’m not sure if this procedure is proper since I have never done this before, so please correct me if these is anything wrong.

2

The EAM file has a tabulated pairwise term (different

than embedding terms). That term needs to be

replaced by something that splines smoothly

to a ZBL potential. Then use pair hybrid with ZBL

and your modified EAM file.

Steve

3

Hi, Steve. I don’t understand Why the modified EAM file is needed to be hybrid with ZBL again, after its pairwise term has already replaced by the splined curves.

4

If the new pair term in the EAM file includes ZBL (all the way to r=0)

then you just need the modified EAM. But you can also truncate

the EAM (or any other) pairwise at r = R0 and use pair zbl for r < R0.

So then you run a pair hybrid with EAM + ZBL.

Steve