Hi all,
I trying to simulate a charge surface and I would like to use fix wall/lj93 command.
My question is:
is in lammps any potential which I can use to describe the electrostatic interaction between that wall and atoms?
Thanks
Pawel
I assume that because you are using fix wall, you are not interested
in the atomic details near the wall surface. (...because fix wall uses
a flat smooth featureless barrier, with no atomistic details). If my
understanding is correct, fix wall, is only used to create repulsive
barriers. It's not intended to represent a realistic surface.
If you want to add some kind of continuum theory to approximate the
electrostatic interactions near the surface, eg:
http://en.wikipedia.org/wiki/Double_layer_(interfacial)
...then you can currently do that using fix addforce to add the
relevant formula to the force acting on every particle:
http://lammps.sandia.gov/doc/fix_addforce.html
To do this, particles with different charges must be put in
different groups. Parsing the text for the mathematical expression
for the force must be done on every charged particle, (and for every
verlet iteration), and could be slow. (Try it and see.)
Alternatively, you could write your own fix, which would be much faster.
------- atomistic detail --------
If you want to represent the surface with atomistic detail, you can do that too.
If you want to turn off periodicity in the direction perpendicular to
the surface, then see the recent posts on the mailing list which
discuss how to use "kspace_modify slab":
http://lammps.sandia.gov/doc/kspace_modify.html
If you want to prevent the slab from moving, you can omit it from the
group of atoms that you send to "fix nve", "fix nvt", or "fix npt".
Cheers
Andrew
yes, I’m not interested in the atomic details near the wall surface.
I thought that I would have to use fix addforce,but I was not sure - now I’m.
Best regards,
Pawel