NPT and NVT simulation coupling

LAMMPS LAMMPS Mailing List Mirror
1

Hi lammps user and developers,

I am trying couple two simulation system. First, I simulated a NPT ensemble to find the equilibrium volume, and then simulate a NVT ensemble starting from NPT simulation results.

For this purpose, I am using the restart file to convert in data file to NVT simulation, but the molecules are out of the box, for this reason i cannot simulate the system with NPT ensemble.

I will be grateful for any guidance.

2

Hi lammps user and developers,

I am trying couple two simulation system. First, I simulated a NPT ensemble
to find the equilibrium volume, and then simulate a NVT ensemble starting
from NPT simulation results.

For this purpose, I am using the restart file to convert in data file to NVT
simulation, but the molecules are out of the box, for this reason i cannot
simulate the system with NPT ensemble.

sorry, but this makes next to no sense. for periodic boundaries, it
doesn't matter in which periodic image a molecule is.

axel.

3

Sorry, i am not explaining well. The nvt ensemble has not periodic boundary in the z axis. For this reason, i need all molecules within box simulation

4

Sorry, i am not explaining well. The nvt ensemble has not periodic boundary
in the z axis. For this reason, i need all molecules within box simulation

please note, that with non-periodic boundaries it is *not* an NVT ensemble!

on top of that, doing an NPT equilibration for that is a waste of
time. you need re-equilibrate the boundary.
better to first set up the system as non-periodic from the beginning
and then relax only the periodic boundaries.

to continue from here you have two options.

1) load your data file into VMD via topotools and use the PBCtools
plugin to wrap your molecules back into the principal cell, extend the
box for the non-periodic dimension a little bit to include the atoms
that "stick out" and write it back as a data file.

2) use two instances of fix oneway with opposite direction to create a
little layer of vacuum in the non-periodic direction and then reset
the image flags for that direction to zero.

axel.

5

Thanks very muchs