[lammps-users] The NPT parameters setting for cylinder wall and particle inside

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

What I want to simulate is liquid crystal confined in a z-axis cylinder wall. The model just built by two kinds of atom. One is for the wall which will be fixed with fix freeze command. The other is for the liquid crystal. I am not sure about two parameters setting as bellow. Any comments will be appreciated.

  1. boundary. We can set boundary with “s s p” or “p p p” but the box size of xy is much bigger than the cylinder diameter. In theoretic, will the two method give us similar result? Whether the former will save computation time?

  2. fix npt groupid npt/asphere. I do not know whether the groupid here should be set as the whole system ( wall and liquid crystal) or just liquid crystal. As the introduction in manual, only the groups whose ID is set in fix command will be updated the velocity, position and other properties. Because I used fix freeze for wall, I think the position of wall should be not change or just scale with the whole box length in z axis. But there are some differences in the two dump file for the wall if I set the groupid for whole system and liquid crystal respectively. How should I set the group ID then?

Thanks in advance.

2

Kitty,

1. boundary. We can set boundary with "s s p" or "p p p" but the box
size of xy is much bigger than the cylinder diameter. In theoretic, will the
two method give us similar result? Whether the former will save computation
time?

Yes and yes. If there are no long-range electrostatics (which I assume
is true in your case), the results should be identical once the
nearest cross-boundary atoms are farther than your cutoff. If you do
have long-range electrostatics, the answers will converge as your box
size in xy approaches infinity. "s s p" will in general save
computation time since there will be the same number or fewer pair
interactions.

2. fix npt groupid npt/asphere. I do not know whether the
groupid here should be set as the whole system ( wall and liquid crystal) or
just liquid crystal. As the introduction in manual, only the groups whose ID
is set in fix command will be updated the velocity, position and other
properties. Because I used fix freeze for wall, I think the position of wall
should be not change or just scale with the whole box length in z axis. But
there are some differences in the two dump file for the wall if I set the
groupid for whole system and liquid crystal respectively. How should I set
the group ID then?

If you're not using an aspherical potential such as Gay-Berne or
Resquared, you shouldn't use an asphere fix. See the docs, for
example:

http://lammps.sandia.gov/doc/fix_npt_asphere.html

"Perform constant NPT integration to update positions, velocities, and
angular velocity each timestep for aspherical or ellipsoidal particles
in the group using a Nose/Hoover temperature thermostat and
Nose/Hoover pressure barostat."

You probably want fix npt:

http://lammps.sandia.gov/doc/fix_npt.html

"Regardless of what atoms are in the fix group, a global pressure is
computed for all atoms. Similarly, when the size of the simulation box
is changed, all atoms are re-scaled to new positions, unless the
keyword dilate is specified with a value of partial, in which case
only the atoms in the fix group are re-scaled. The latter can be
useful for leaving the coordinates of atoms in a solid substrate
unchanged and controlling the pressure of a surrounding fluid."

Looks to me like you want to set your group ID to the fluid atoms and
use "dilate partial".

Paul