Help regarding hybrid MC/MD simulation in lammps

LAMMPS LAMMPS General Discussion
1

Hi,

I created an input file to perform hybrid MC/MD simulation. I am using atom/swap command to swap the position of different types. The simulation ran smoothly, however I do not see decrease in potential energy which i should expect. Could anyone please have a look at my input file and suggest me any issue with my input file? I am attaching my input file in this post.

Regards
Anas Manzoor
input-file.txt (2.1 KB)

2

I cannot comment on the specific question, since the input is incomplete and missing the data file.

However, from looking at the input, I have two questions:

  • why are you using the same random number seed for all fix swap instances?
  • why are you using multiple fix atom/swap instances in the first place? why not a single one?
3

Thank you for responding to my post. Yes, you are right I should use different random number for each swap. In my system there are 5 elements, so I am trying to swap all different pairs which seems reasonable to me.

4

But you should be able to do this with just one instance of fix atom/swap. Please re-read the documentation carefully.

5

Ok, I got it. Thank you for pointing out to me.

6

Were you able to do that in one instance of fix atom/swap as axel (@akohlmey) suggested?
According to the documentation as seen here “The types keyword is required. At least two atom types must be specified. If not using semi-grand , exactly two atom types are required.”

But if you used “semi-grand yes” that would mean your sample might not end up being equimolar.

What am I missing here?

One could use loop here with “one” instance fix atom/swap.