# inverse-power potential

Dear LAMMPS Users,

I want to use a purely repulsive A/r^12 potential.

LAMMPS does not seem to have an inverse-power potential implemented and there is no version of the LJ potential where the r^-12 and r^-6 coefficient can given independently (E=A/r^12-B/r^6, like in gromacs for instance)

I am wondering what would happen if I would use a Mie potential with gamma_rep=12 and gamma_att=0 ? Would C be undetermined or could it return a value of 1 ? Could someone suggest a work-around possibly other than a tabulated potential to get this inverse-power potential ?

Thank you in advance for helping

Dear LAMMPS Users,

I want to use a purely repulsive A/r^12 potential.

LAMMPS does not seem to have an inverse-power potential implemented
and there is no version of the LJ potential where the r^-12 and r^-6
coefficient can given independently (E=A/r^12-B/r^6, like in gromacs
for instance)

I am wondering what would happen if I would use a Mie potential with
gamma_rep=12 and gamma_att=0 ? Would C be undetermined or could it

the computation of C would cause a division by zero and thus make this
path unattainable.

return a value of 1 ? Could someone suggest a work-around possibly
other than a tabulated potential to get this inverse-power potential ?

what is your objection against tabulation?

if people want a purely repulsive r^12 term, they usually take a 12-6
LJ and put the cutoff in the minimum of the potential.

otherwise, you have to grab a text editor and do some programming.
LAMMPS is designed so that it is fairly easy to add another pair
potential.

axel

This should be a quick fix. Copy and rename the 6-12 LJ to whatever you like and take out all the occurances where the r^6 piece is calculated.

if people want a purely repulsive r^12 term, they usually take a 12-6
LJ and put the cutoff in the minimum of the potential.

And that cutoff is 2^(1/6) sigma.

Steve