displace_box command - impact on molecules

Hi there,

I was wondering what the impact of a command along the lines of:

“displace_box all x scale A y scale B z scale C”

was on the relative molecular dimensions within a box? I can’t see any reference in the documentation as to whether this:
a) scales the box and all atomic positions, or
b) scales the box and all molecular centres of mass, retaining all relative atomic positions within a linked molecule (thus bond lengths and bond angles remain constant)?

As a little more background, I’m using LAMMPS as a library to run a hybrid MC-MD experiment. Although the code seems to work sensibly for LJ particles, it fails once I try to put molecules (in this case water) into the simulation. I appreciate that my code may well be at fault, but thought I should clarify whether this might be an issue first.

Many thanks,

Mark
University of Cambridge

Hi there,

I was wondering what the impact of a command along the lines of:

"displace_box all x scale A y scale B z scale C"

was on the relative molecular dimensions within a box? I can't see any
reference in the documentation as to whether this:
a) scales the box and all atomic positions, or
b) scales the box and all molecular centres of mass, retaining all relative
atomic positions within a linked molecule (thus bond lengths and bond angles
remain constant)?

please note that there is no more "displace_box" command
as of the 9Feb2012 patch.
http://lammps.sandia.gov/bug.html

have a look at the "remap" option of the new consolidated "change_box"
command to select the behavior that you desire.

As a little more background, I'm using LAMMPS as a library to run a hybrid
MC-MD experiment. Although the code seems to work sensibly for LJ particles,
it fails once I try to put molecules (in this case water) into the
simulation. I appreciate that my code may well be at fault, but thought I
should clarify whether this might be an issue first.

adding/removing molecules with bonded interaction require a lot
of additional bookkeeping to be handled and also preallocating
suitable storage that cannot be increased later.

cheers,
   axel.

As Axel says, see the doc page for the change_box command.

a) scales the box and all atomic positions, or
b) scales the box and all molecular centres of mass, retaining all relative atomic positions within a linked molecule (thus bond lengths >and bond angles remain constant)?

If you use the remap option it does (a) and not (b).

Steve