Energy conservation problem regrading REBO potential in LAMMPS

The person I was trying to remember is Markus Hohnerbach (CCd).

Maybe he can comment on the possible “bug” you are seeing.

Ouyang - you can point him to the thread on this …


Hey Ouyang,

Can you send me your input files/data? I am subscribed to the list, but the initial mail of the thread has vanished for me...

I can certainly run it against a few independent implementations (namely OpenKIM, LAMMPS, and USER-INTEL (mine)), and see if I find discrepancies.

In addition, it would be useful if you could provide me with at least the forces produced by your own code (e.g. the equivalent of dump x y z fx fy fz in LAMMPS).

Cheers, Markus