3
$\begingroup$

I am seeking advice regarding a molecular dynamics simulation in LAMMPS of Gallium Phosphide (GaP). I came across a paper by Powell et al. which parameterized a tersoff potential for Ga-P interaction, but it did not include the Ga-Ga and P-P interaction parameters. However, the author did find various elastic constants of GaP using that potential. On the other hand, I was able to successfully recreate the tersoff potential for GaN (included in the LAMMPS example directory) from a paper by Nord et al., which included the Ga-Ga and N-N interaction parameters.

My question is, can I use the parameters from Powell et al. without including the Ga-Ga and P-P interaction parameters, or do I need to use some hybrid potential? I am not sure if this is possible or not. I would appreciate any guidance or suggestions on how to proceed with this issue.

Thank you.enter image description here

$\endgroup$

1 Answer 1

4
$\begingroup$

This is a great example of how MD potentials are meant to fit substances for particular purposes.

The Tersoff potential is meant to fit the cohesive energy of solid-state structures by modifying the energies of bonds in terms of the geometries of other nearby bonds. Yes, it's implemented in LAMMPS as a non-bonded potential because that's how LAMMPS represents materials whose bond topologies can change, but you should consider each pair interaction as implicitly a bond.

So why doesn't Powell's paper have Ga-Ga and P-P parameters? Because he studied GaP only in the regime where there was only one type of nearest-neighbour bond, namely the Ga-P bond. And why does Nord's paper have Ga-Ga and N-N parameters? Because they were studying Ga and N in high-pressure phases, where all of those bonds were relevant.

So the question of whether you need the homoatomic parameters depends on whether you expect to see homoatomic bonds at your chosen temperature and density. Note that there is a recent study which used Powell's parameters in LAMMPS (Phys. Rev. B 101 (2020), 024307), so you should try to contact those authors for more information.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .