Hydrogen migration/diffusion on graphene surface using Airebo

Hello,

����I am following the procedure outlined in the paper "Hydrogen transport on graphene: Competition of mobility and desorption" (Phys Rev B 84 075486 2011), by V.A. Borodin et al., to approximate the diffusion activation energy for a single hydrogen atom on a graphene surface. In this paper the authors use the first generation Brenner interatomic potential, (D. W. Brenner, Phys. Rev. B 42, 9458 (1990)). Molecular dynamics is used to determine the temperature dependence of the average on site residence time, which can be used in approximating the diffusion activation energy. The authors cite that at a temperature of 2000K and a simulation time of 2ns, a total of 88 diffusion jumps were recorded. I have tried to follow the exact method outlined in this paper, the only difference being the interatomic potential I am using is the LAMMPS Airebo potential, and I have not recorded a single diffusion or desorption event at 2000K. The hydrogen atom is observed to oscillate around its initial equilibrium position for the entire duration of the simulation. I assumed that since the Airebo potential contains the 2nd generation REBO term, which is supposedly an improvement of the original Brenner potential, it could also model this system and provide similar results. Perhaps I am wrong in this assumption or maybe there is an error in my LAMMPS input script that results in the hydrogen not diffusing or desorbing. Below is a copy of my LAMMPS input script, I thank you for your time and would appreciate any input on this topic.

I am using the 30 Aug 2012 version of LAMMPS.

Hi Justin,

Indeed 2nd gen. REBO is an improvement over the 1st, but you are using
the parameters given in the 2002 Stuart paper of AIREBO, not
parameters from the 1st nor 2nd gen. REBO paper by Brenner et. al.
Formalism of the potential can give a difference, so do the
parameters. Parameters given in CH.airebo may not reproduce the
properties given by 1st. gen. REBO. Perhaps you can ask the authors
of the 2011 paper why did they chose 1st gen. REBO for their work.

Ray