Hi Harish,
To the best of my knowledge, it is not yet possible to run the Feuston/Garofalini potential in LAMMPS because it has a three body interaction that depends not only on the angle between the atoms involved but also on the radii of the relevant bonded atoms. Thus, one needs to maintain a dynamically updated list of three body interactions (and, of course, apply the proper form to each triplet to compute relevant energy/forces). Again to the best of my knowledge, there is nothing preventing one from implementing this potential in LAMMPS; in fact, the modified BMH part may be easily implemented with modifications to code that currently implements:
pair_style born/coul/long
The F/G potential is attractive because it is a computationally simple - yet reactive - potential and I believe it has been parameterized to allow one to address water reacting with amorphous silica. If you pursue implementing this potential in LAMMPS, please keep me abreast of your progress.
That said, an alternate potential to consider employing is the one due to van Beest, Kramer, and van Santen – the so called, BKS potential. You can look at the following for details:
Phys Rev B, vol 43, p. 5068 (1991)
also see references therein. Put simply, I think you can implement this using either the buck/coul/cut or buck/coul/long pair style. I’m uncertain which but, if you read the relevant refs and LAMMPS doc pages, you should be able to figure this out.
You may also want to take a look at the following from Goddard’s group:
http://www.wag.caltech.edu/home-pages/tahir/glassff.htm
This is a straightforward pair potential and should be “easy” to implement. Alternatively, you can investigate the REAXX force field for silica – I believe one exists but I am a bit clueless beyond that.
Best of luck,
Ed