[lammps-users] RE-squared parameters in pair_coef command

Dear LAMMPS users,

I want to simulate a system consist of two types of ellipsoidal particles. One type is thick and another one is thin. The radii of thick particles are 0.5 and 0.5 and 1 and the radii of thin particles are 0.4 and 0.4 and 1.56.

I want to use RE-squared potential between my particles but I don’t know how I should set the parameters in the Pair_coef command. If I consider thick particles as type 1 and thin particles as type 2, and set sigma to be 1 and Hamaker constant to be 100, then I set the parameters of pair_coef command as follow:

Pair_coef [type i] [type j] [sigma] [Hamaker] [epsilon_1_i] [epsilon_2_i] [epsilon_3_i] [epsilon_1_j] [epsilon_2_j] [epsilon_3_j] [cutoff(optional)]

Pair_coef 1 1 1 100 1 1 4 1 1 4

Pair_coef 1 2 1 100 1 1 4 0.64 0.64 9.75

Pair_coef 2 2 1 100 0.64 0.64 9.75 0.64 0.64 9.75

Are these parameters correct or not? When should I set some of epsilons equal to zero?

Thank you


I will let Mike Brown answer this.


I believe that the Hamaker constant should come before sigma; please see the order of coefficients in the documentation page. There is no need to set epsilons to zero. This is an option to remove redundancy. The epsilons for a particle type should not change. If you have a LJ point particle for RE-squared, epsilons are ignored. - Mike