[lammps-users] gibbs ensemble


It seems like it's not possible to make gibbs ensemble molecular dynamics
simulations using lammps. How hard do you think it would be to implement it in
lammps? And is there any plans of doing it in the near future?
Thank you,


Hi Sezen. I think it would be fairly simple to add since LAMMPS can already do multiple replicas (see the temper command — does replica exchange MD), and the machinery is there to do adding and deleting particles. The problem would be the insertions, which wouldn’t be too hard to add for simple stuff I think. We have no plans to do this. If you do end up writing a fix or something to do this, please send it along. Thanks, Paul

I would write a command modeled after temper.cpp to
do a Gibbs ensemble (i.e. run 2 simulations with swaps
between them). It's really not a fix, b/c a fix would work
within one of the simulations.