Thanks for your suggestion. Actually I also need to extend it to minimize other
parameters of a group of atoms, not only energy. The first step I plan to do is
to make a small change so that I can do energy minimization only to certain
group of atoms.
Quoting L Konzern <[email protected]...>:
The suggetion by L Konzern to just use fix setforce
to freeze the atoms you don't want to move is a good
way to do it.
I don't see how you would modify minimize to work on a group
of atoms. The energy and force calculation is
done by the usual LAMMPS pair, bond, etc classes,
and they don't work on groups of atoms, just the
entire system. The "all" group ID to fix minimize is
ignored. That fix (which is hidden from the user) stores
the gradient vector for minimization and has to be over
all atoms since its communicated as atoms move.
Steve