The fix command fails to immobilize graphene in PIMD.

GPUMD
1

Hello all,

I am trying to immobilize graphene (freeze the position of all C atoms) in a PIMD simulation (GPUMD version 4.5). The specificsettings are carbon atoms of the graphene are assigned to group 0 in model.xyz (only one grouping method), and fix 0 is set in run.in. Such setting successfully fixes the graphene in CMD(ensemble nvt_ber 100 100 20), but fails in PlMD (ensemble pimd 32 100 100 20) that, all C atoms move randomly in all directions. What causes this issue, and are there any methods to resolve it?

2

Yes, this is the expected behavior as the fix keyword has not been implemented for PIMD runs. Could you explain why you need this? Or if you really believe this is needed, at present you need to mofify the code. I am not sure if this makes sense for PIMD runs but you can try to modify the condition in line 198 of the [source file](GPUMD/src/integrate/ensemble_pimd.cu at fe2c880eee8a6b991f959971fe1f839b6a944cec · brucefan1983/GPUMD · GitHub)

   if (n < number_of_atoms) {  // modify this line such that the code below only applies to the mobile atoms