[lammps-users] asymmetric dihedral angle potential

LAMMPS LAMMPS Mailing List Mirror
1

Dear,

I need to define a dihedral angle potential which is not symmetric about 0 degree,
Something like the dihedral_style charmm and class2 but maybe more parameters.
For example, in dihedral_style charmm, a non-zero phase shifting parameter d can
make the potential is not symmetric about zero. But there are only one cosine term
in that function which is insufficient for the potential I want. But I think I could
manage to compose a complex function by using more terms.

Now what I would like to know is
1. How the torsional angle is defined in lammps.? [0,pi] or [-pi, pi] or [0,2pi]?
2. Under what circumstance, one should/could use phase shift parameter?
    In CFF91, I see it is always 0. I don't know how it is used in Charmm.
3. Is there possibility to use tabulated potential for dihedrals? (Most probably is not)

Regards,
li

2

Most of your questions are asking for things LAMMPS doesn't
do - you'd have to write these new dihedral potentials yourself,
using a current LAMMPS one as a starting point.

I'd look in the code to see about 0,2*pi vs -pi,pi - it probably doesn't
matter, but you could make sure that way.

No one has written a dihedral_style table, but it could be
done, so long as you had just one free param (the torsion
angle).

Steve