Hi,
Please find attached my LAMMPS input script and the log file generated.
1: My objective is to create a single Si atom atop a substrate with fixed and moving atoms. The single atom needs to be atop the moving atoms. I used the following create_atoms command create_atoms 1 single 10.87 11.16 10.87 according to the syntax for creating a single atom but the log file shows that no atom is created. A lattice is defined as expected.
2: I’ve even tried using the fix deposit command but the syntax is too tricky. Following is what I used:
fix 1 all deposit 1 1 500000 12345 region myreg global 10.875 11.16 10.875 vz 0 -11.7 units lattice . To avoid the single atom created from zipping away, I have useda negative Z-direction velocity range((0 – (-11.7)). This too throws an error ERROR: Replacing a fix, but new style != old style.
3: I am using periodic boundary conditions in my input script. Does usage of the fix wall/reflect command help me retaining the atom in the created box?
Could anybody please help me with these queries…
Regards,
Saketh.
University of Toledo.
3.15.11log.lammps (1.65 KB)
NEWTADSI.txt (1.87 KB)
Hi,
Please find attached my LAMMPS input script and the log file generated.
1: My objective is to create a single Si atom atop a substrate with fixed
and moving atoms. The single atom needs to be atop the moving atoms. I used
the following create_atoms command create_atoms 1 single 10.87 11.16 10.87
according to the syntax for creating a single atom but the log file shows
that no atom is created. A lattice is defined as expected.
would the created atom be inside the simulation box?
it doesn't look like it, so lammps doesn't do it.
if you want "absolute" coordinates, you have to add the "units box" flag.
2: I’ve even tried using the fix deposit command but the syntax is too
tricky. Following is what I used:
fix 1 all deposit 1 1 500000 12345 region myreg global 10.875 11.16 10.875
vz 0 -11.7 units lattice . To avoid the single atom created from zipping
away, I have useda negative Z-direction velocity range((0 – (-11.7)). This
too throws an error ERROR: Replacing a fix, but new style != old style.
remember that you read a restart which may create a fix for you, e.g. fix nvt,
so you have to make sure that the fix labels match or that you unfix the old
one first.
3: I am using periodic boundary conditions in my input script. Does usage of
the fix wall/reflect command help me retaining the atom in the created box?
yes. but you should be careful which walls to define and what group of atoms
you want to apply them to.
cheers,
axel.
3: I am using periodic boundary conditions in my input script. Does usage of
the fix wall/reflect command help me retaining the atom in the created box?
yes. but you should be careful which walls to define and what group of atoms
you want to apply them to.
If you have walls, you shouldn't be using a periodic boundary in that dimension.
Steve
Hi people,
I have a couple of questions:
1. When I create vacuum in a simulations box, let's say to simulate a molecule absorption in a surface, sometimes my simulation crashes with no apparent reason. I realized that decreasing the number of processors usually solves the problem. Is there any problem if at some point the space assigned to a processor is empty?
2. Has anyone implemented Monte Carlo in LAMMPS at some point? If not, I would like to know if there is some low-level reason in the code that makes it unpractical.
Best
Answers below.
Steve
Hi people,
I have a couple of questions:
1. When I create vacuum in a simulations box, let's say to simulate a
molecule absorption in a surface, sometimes my simulation crashes with
no apparent reason. I realized that decreasing the number of processors
usually solves the problem. Is there any problem if at some point the
space assigned to a processor is empty?
This shouldn't happen, and it might be a bug. But there have been
several bug fixes related to this recently, so you might try the most
current version.
2. Has anyone implemented Monte Carlo in LAMMPS at some point? If not, I
would like to know if there is some low-level reason in the code that
makes it unpractical.
MC means many different things. There is a MC fix bond/swap in LAMMPS.
Paul has been working on a GCMC, and has a preliminary version (not
yet released). But many kinds of MC are not very parallel and would not
be compatible with LAMMPS. So it depends on what you want to do.
Hi,
Thanks for the answers. I will try a new version if my problems with the processors became too annoying.
Regarding MonteCarlo, I was basically thinking on Gran Canonical MC. It would be great to combine MD with GCMC in some studies. Or even MC movements of atoms instead MD. It is a good new that you are working on that. Is there a tentative release date for the version with GCMC?
Best
Talk to Paul about GCMC ... (CCd above).
Steve
Anyone who would like to be a "beta tester" of the new not-yet-released "fix GCMC" code I've written, please e-mail me, and I'll send you the code.
Paul