[lammps-users] About tabulated potential

Dear LAMMPS users,

My question is about tabulated potential file.

My inner cutoff distance is 0.01Angstrom. This short atomic separation is required for modeling energetic recoils. I tabulated the potential in a file called “Yakub.table”, which is attached with this mail.

As advised, I used “pair_write” to verify how LAMMPS interpolate the tabulated file. Unfortunately, the output potential from LAMMPS is different from my input tabulated file at short distance. But interesting thing is they are the same at long distance.

I attached the LAMMPS input file and the figure comparing tabulated potential versus LAMMPS “pair_write” potential along with this mail.

I wonder how I could fix this problem.

Any suggestion or comment would be highly appreciated.

Thank you.

Bin.

Compare.jpg

in.TT (1.79 KB)

Yakub.table (105 KB)

It sounds like you are doing something rather unusual for which the table style was not intended. I am going to pass this one on to Paul Crozier. Perhaps it would be more appropriate to use keyword cubic in this case.

Aidan

Dear LAMMPS users,

My question is about tabulated potential file.

My inner cutoff distance is 0.01Angstrom. This short atomic separation is
required for modeling energetic recoils. I tabulated the potential in a
file called “Yakub.table”, which is attached with this mail.

As advised, I used “pair_write” to verify how LAMMPS interpolate the
tabulated file. Unfortunately, the output potential from LAMMPS is different
from my input tabulated file at short distance. But interesting thing is
they are the same at long distance.

this is a typical effect of cubic splines at boundaries.
it is grossly enhanced by the awkward way that you are plotting it.

I attached the LAMMPS input file and the figure comparing tabulated
potential versus LAMMPS “pair_write” potential along with this mail.

I wonder how I could fix this problem.

there are a number of things that you should do.
a) tabulate at a higher resolution, e.g 0.01 instead of 0.1.
b) tabulate with rsq instead of r. you want the high resolution
  for small r, but don't care so much at longer r
c) tabulate even lower r or rsq than what you need.

keep in mind that this deviation is only for _very_ short distances
< 0.3 angstrom. do your atoms ever get _this_ close?

if you just make sure that your tabulated potential extends beyond
the range that you expect to ever reach, those spline related
ondulations should not be an issue.

cheers,
   axel.

Bin,

I agree with Axel's suggestions. Also, please try to follow these guidelines from the documentation:

Here are some guidelines for using the pair_style table command to best effect:

Vary the number of table points; you may need to use more than you think to get good resolution.
Always use the pair_write command to produce a plot of what the final interpolated potential looks like. This can show up interpolation "features" you may not like.
Start with the linear style; it's the style least likely to have problems.
Use N in the pair_style command equal to the "N" in the tabulation file, and use the "RSQ" or "BITMAP" parameter, so additional interpolation is not needed. See discussion below.
Use as large an inner cutoff as possible. This avoids fitting splines to very steep parts of the potential.

Paul