Seperate pe/atom for atoms interaction with multiple potentials

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I have a quick question. currently I am computing the total potential energy for a group of atoms using the following.
compute 1 stickers pe/atom pair

compute 2 stickers pe/atom bond

compute 3 stickers pe/atom angle

however these group of atoms interact with a a hybrid potential as the following

pair_style hybrid/overlay table linear 1836025

pair_coeff * * table Lj_cal.table Lj_cal

pair_coeff 8401678 8401678 table stickers_soft2.table stickers_cal

My question can I compute pe/atom and the contribution of each potential alone.

Thanks

2

You cannot decompose pe/atom from different pair styles in a hybrid set up. Instead, you can use compute pair (https://lammps.sandia.gov/doc/compute_pair.html) to get the total potential energy of each sub styles. In your hybrid set up there are no bond, angle, dihedral, etc anyway.

3

You can also do a rerun command on your dump snapshots and redefine the
potetential to only be one of your hybrid sub-styles. They you can invoke
compute pe/atom and dump out those partial forces if you wish.

Steve