A similar issue was discussed here: http://lammps.sandia.gov/threads/msg62178.html
It sounds like what you are looking for are hard reflective walls (i.e. elastic collisions), which can be implemented with fix wall/reflect (http://lammps.sandia.gov/doc/fix_wall_reflect.html). However, since this only displaces atoms, you would probably be better served using a soft fix wall/harmonic.
Andrew
A similar issue was discussed here:
http://lammps.sandia.gov/threads/msg62178.htmlIt sounds like what you are looking for are hard reflective walls (i.e.
elastic collisions), which can be implemented with fix wall/reflect
(http://lammps.sandia.gov/doc/fix_wall_reflect.html). However, since this
only displaces atoms, you would probably be better served using a soft fix
wall/harmonic.
FYI, a reflective wall *still* has an indirect impact for anything
that is not modeled with a hard sphere interaction: as atoms will
experience a dense interacting system (for water with long-ranging
coulomb interactions) on one side and a vacuum on the other. a
harmonic wall will promote the formation of a surface dipole as well.
axel.
Dear Andrew and Axel ,
Thanks for your reply .
What about using potential like Morse potential ?
Can i use it at up and down side of water molecules by LAMMPS ?
Saeed.
Dear Andrew and Axel ,
Thanks for your reply .
What about using potential like Morse potential ?
Can i use it at up and down side of water molecules by LAMMPS ?
this is not helping and not getting anywhere. you obviously have still
not understood what the _principal_ problem is in what you are asking.
please explain in more detail, what your setup is, what you want to
achieve and what the motivation and purpose of that is supposed to be.
without this information, it will not be possible to help you. what
you are asking for currently is equivalent to asking for having a
periodic boundary and a non-periodic boundary in the same dimension at
the same time.
axel.
The simplest possible choice is wall/reflect, which effectively treats the wall-fluid interaction energy as E=0 zero on one side of the wall and E=Infinity on the other side. As Axel mentioned multiple times, this is still an interaction, simply a highly repulsive one. It could be argued that “no interaction” would require balancing repulsion with attraction, which could be accomplished with a LJ wall and suitable values of epsilon and sigma. It all depends on what you mean by “inert”. Gold is inert, but also attractive to many fluids, and people too.
Aidan