How to dump out Peridynamics bonding information?

Dear Lammps users,

I am using the Peridynamics package in Lammps, and I need to find out which atoms are bonded to an atom of interest at a time. Is there any Lammps command to dump out that information? Or do I have to modify Lammps to do that?

Thanks,
Quang.

I'll let Mike Parks respond to this. I believe
that info is currently known only
internal to LAMMPS. But maybe he
has an idea about how to get at it,
or whether it would be useful to expose
it in some manner.

Steve

Thank you Dr. Plimpton. I'll look forward to it.

Regards,
Quang.

Quang,

   This feature has been requested fairly frequently. In the next release of the peridynamics implementation within lammps there will be a new compute style allowing you to write the peridynamic bond family to disk via a dump command. I'm hoping to have the next release out soon -- I'm currently performing testing on the new code.

- Mike

Thank you for the reply, that's great to know. I have another question: currently I'm using the compute stress/atom command to estimate local stress with peridynamics pair style. Somehow in the output there's only contribution from kinetic energy but no contribution from either pair energy or bond energy.

Is that something expected with pair_style peri? Or is the command "compute stress/atom" compatible with peridynamic interactions at all?

Regards,
Quang.

Quang,

   I don't believe the stress/atom compute is compatible with the peridynamic interactions. In any case, it's not clear that the functionality in that compute style is appropriate for computing a peridynamic stress.

   For a description of a peridynamic stress tensor, have a look at the paper "Force flux and the peridynamic stress tensor", Journal of the Mechanics and Physics of Solids 56 (2008) 1566-1577. doi:10.1016/j.jmps.2007.08.004. It contains a closed-form formula for a peridynamic stress tensor (a nonlocal analog to the first Piola-Kirchhoff stress tensor). Using this formula, one could write a compute style in LAMMPS to compute the peridynamic stress tensor. I don't have any plans to do this in the immediate future, but if you are interested in developing a compute style for this I'd be happy to include it into the peridynamics module within LAMMPS so that everyone could use it.

Best,

- Mike

All the compute stress/atom is doing is grabbing the virial
terms tallied by the pair style. So if the peri pair style tallies
a virial (and it looks like it does, see the ev_tally calls), then
there should be a virial contribution to the stress (along
with the KE contribution).

Steve

Thank you for the suggestions. I will try it out.

Regards,
Quang.