AIREBO potential for nonreactive systems

LAMMPS LAMMPS Mailing List Mirror
1

Dear LAMMPS users,

I was searching the mailing list for information about the simulation of graphitic systems. I noticed that a lot of search results indicate that people simulate such systems (graphene and graphite) using the AIREBO potential. I was wondering if there is a particular reason this is used so much in LAMMPS for non-reactive systems. For the research I’ve been doing, I have thus far used the OPLS-AA carbon potentials. It is my understanding from reading the AIREBO paper that the method is primarily for reactive systems. I was wondering if perhaps the method has been fine-tuned such that it is faster or more accurate than conventional methods even for non-reactive systems or if there is another reason for its popular use in LAMMPS?

(I did notice that Dr. Kohlmeyer used LJ, harmonic, and CHARMM for these systems https://sites.google.com/site/akohlmey/software/vrpn-icms so now I’m not really sure when AIREBO is used).

Thank you,
Nick

2

On the contrary, AIREBO is much more expensive and slower
than OPLS. You can look at the "potentials" section of the
LAMMPS benchmark web page. It's probably more
accurate than OPLS for some properties, but you are
correct that the chief motivation to use it is when
there is bond breaking/formation.

Steve