[lammps-users] negative pressure comes up when proceeding an minimize process

Hi,
I have create an GPTMS model with Amorphorous cell modelMaterial studio and modify it to fit the format used in LAMMPS. When I try to minimize the system with the ReaxFF, I found that the pressure in the system was negative and, to make it worse, after some steps of minimize the pressure is “nan”. Why would this comes up? Any reply will be appreciated.
Sincerely,
Shuping Huang

There's nothing unphysical about a negative pressure. Nans are typically
due to bad forces leading to large atom movements. For a minimization
I would check the various min options and do something simple and robust
to start with (e.g. backtracking) and see if you can make progress.

Steve

There's nothing unphysical about a negative pressure. Nans are typically
due to bad forces leading to large atom movements. For a minimization
I would check the various min options and do something simple and robust
to start with (e.g. backtracking) and see if you can make progress.

Steve

Thanks, Steve. Yet, I have another question. According to the doc page of the min_style modify, the default line searching method is backtrack and the default maximum of distance is 0.1 Angstrom. I did not use the min_modify command in my script and I suspect that if such an operation would work. Can there be some other factors that led to such a result? Can it be a matter of model or force field? Looking forward to your reply.Thanks in advance.
Sincerely,
Shuping Huang

I haven't done minimization with ReaxFF problems - maybe Aidan wants
to comment. My guess is that you have a badly formulated initial geometry
and the code and FF are having severe problems dealing with it.

Steve

There are many possible reasons for this. The force field is one
possibility. I would recommend that whatever force field you are using, you
should test it out on a simple example of your system of interest, before
throwing a large complex problem at it.

Aidan