# Molecular dynamics to reproduce dispersion interactions

I'm trying to simulate the behavior of chromophore molecules incorporated into a polymer under an external electric field.

I've tried to use classical molecular dynamics (GROMACS) but it's unable to reproduce some phenomena such as molecule aggregation due to its dispersion nature.

So I'm looking for a dynamic method/software that is able to reproduce dispersion interactions on the scale of thousands of atoms. I'm personally thinking about CP2K or CPMD but I'm not sure.

• First of all, are you sure the molecules do aggregate in reality? Force fields are coarse methods indeed, but they are not very likely to be qualitatively wrong for dispersion. Jul 3 at 14:42
• I believe yes, they do aggregate. At least I've seen a couple of papers where some phenomena were attributed to aggregation. Jul 3 at 15:29
• roma, @wzkchem5: this may be a naive question, but dispersion interactions are typically pairwise (e.g. between 2 atoms), so what is meant by "on the scale of thousands of atoms"? Jul 3 at 16:08
• What forcefiels are you using? I would blame the brand of FF, not the use of a FF Jul 3 at 22:47
• I believe gromacs has an option to turn on long range dispersion corrections (DispCorr=EnerPress). However, according the manual it should only be used for force fields which were parameterized with dispersion corrections (e.g. Amber99). For CHARMM the dispersion should not be used. You can also try the software xTB which has fast semi-empirical methods and a polarizable force-field. Jul 4 at 3:57