Mixing AIREBO/MORSE (metal) with GEAFF (real) in hybrid simulations

LAMMPS
, ,
1

Hello,

I am simulating a “polymer+solvent system interacting with CNT” using hybrid pair styles in LAMMPS.

*CNT interactions: Using airebo/morse, as π-π stacking interactions are important.
*Polymer & Solvent interactions: Using GEAFF, which consists of Lennard-Jones (LJ) + Coulomb interactions.

The issue is that airebo/morse is typically defined in metal units, whereas GEAFF uses real units. Since LAMMPS does not support multiple units within a single simulation, I am wondering:

  1. Is it possible to use airebo/morse with units real?
  • If so, are there any recommended ways to modify the potential file properly?
  1. Alternatively, can GEAFF be used with units metal instead of real?
  2. If neither is feasible, what would be the best workaround to simulate this hybrid system?

Any suggestions or past experiences on handling mixed-unit hybrid potentials would be greatly appreciated.

Thank you!

2

Technically it is possible, but you have to understand all the details of the potential file and convert its values. You can easily test this, since both a real units version and a metal units version have to produce the same trajectory when starting from the same geometry at least for the first few 100s of steps.

Again, the challenge is on you to correctly convert all potential parameters.

Mixing potentials in general is only a suboptimal solution. It would be better to find a force field that can describe everything you want to model consistently. Mixed potentials always have inconsistencies.

1 Like
3

Thanks to your advice, I was able to consider alternative solutions.
I appreciate your guidance.