[lammps-users] Delete molecules at end of calculation box

Dear all,

If anyone know how to delete molecules at edges of calculation box in
LAMMPS, would you tell me how to do it?
I don't use periodic boundary condition, and I would like to delete
molecules which reach to the end of calculation box.
My system is under strong electric field in x direction, so certain
positive and negative ions try to escape from the box at the yz plane at
"xlo" and "xhi" of calculation box.
But without any command for it, it cause errors in terms of Bonds.
I tried to use delete_atoms and delete_bonds as following, but it seems
it doesn't work in my case...

I might be wrong, but probably because it is for the initial condition?
I would be appreciate any suggestions.
Thank you very much.

Best regards,
Nanako Takahashi

If you use the boundary command with a fixed style 'f', then
atoms that move beyond the box boundary will be deleted.


Dear Steve and lammps users,

Thank you very much for answering my question.
I use boundary command with 'f'.
And I guess that one of atom in molecules is missing because it is away from calculation box.
I also guess that that is the reason why I get the following error which I mentioned in previous mail titled "external efield and temperature control".
Would you tell me whether it is possible to delete "molecules" (not "atoms") which reach to the end of calculation box?

"ERROR on proc 0: Bond atoms 20 23 missing on proc 0 at step 184356"

Thank you very much for your help.

Nanako Takahashi

Steve Plimpton wrote

If you want to delete a molecule you'd need to write code yourself
to do that. I suggest a new fix.


Dear Steve and LAMMPS users,

Thank you so much for your reply. Whenever I ask questions to this mailing list, I feel I gradually understand about LAMMPS and also about my research's concept. I always appreciate you so much.

The reason why I would like to delete molecules is that I need to observe how molecules extracted by strong electric field from droplet which includes a lot of molecules with many iterations, but flying away molecules stops the simulation because one of atoms is beyond calculation box.
I used very big calculation box, but I could not get enough size because of memory issue.
So I tried boundary command with "m" or "s".
When I tried "m", + ions could fly away, but - ions couldn't.
But with "s", it seems both ions can move smoothly so far.
I might not understand shrink-wrap very well, but do you think the command help my problem?
Why do we need calculation box for non-periodic boundary condition?
Any comments help me a lot.
Thank you very much.

Best regards,
Nanako Takahashi

Steve Plimpton wrote: