[lammps-users] Setting 1-4 coefficients

Hi all,

Is there any way i can set the 1-4 weighting coefficients for individual dihedrals? I have both electrostatic and vanderwals interactions in my system. The special bonds command sets it for all the dihedrals at once but what I am looking for is the ability to set it for each particular dihedral. Please let me know if anyone knows a way around it.

thanks.

Hi all,

Is there any way i can set the 1-4 weighting coefficients for individual
dihedrals? I have both electrostatic and vanderwals interactions in my
system. The special bonds command sets it for all the dihedrals at once but
what I am looking for is the ability to set it for each particular dihedral.
Please let me know if anyone knows a way around it.

i would exclude 1-4 interactions completely in special_bonds
and then use a dihedral style that includes the charges in the
calculation. this way you can turn this into a per dihedral property.
if this kind of dihedral style doesn't exists, you need to write one.

cheers,
   axel.

The only force field I know that is defined similar to this way
is CHARMM. LAMMPS has pair lj/charmm/coul/charmm (or long)
that lets you define the sigma_14 and epsilon_14 for individual
1-4 interactions. This in is conjunction with dihedral_style charmm.
If that doesn't do what you want, you'd have to add your own customized
pair style.

Steve

Thanks, Steve and Axel for your suggestions.

I had couple more questions regarding dihedrals and impropers.

  1. As i understand it, the special bonds command sets the weighting for non-bonded interactions between atoms connected through an improper also. If an improper dihedral is ordered i-j-k-l and i-l end up being bonded (as 1-2 neighbors) then non-bonded interactions between i-l is weighted by the 1-2 weighting coefficient in special bonds command. Is this correct?

  2. For improper style harmonic (assuming the improper is ordered i-j-k-l), the angle computed is the Wilson-out-of-plane angle between the plane of i-j-k and the plane of j-k-l. This is same as the angle that is computed in improper style cvff. Is this correct?

thanks in advance for the help.

Comments below.

Steve

Thanks, Steve and Axel for your suggestions.

I had couple more questions regarding dihedrals and impropers.

1. As i understand it, the special bonds command sets the weighting for
non-bonded interactions between atoms connected through an improper also. If
an improper dihedral is ordered i-j-k-l and i-l end up being bonded (as 1-2
neighbors) then non-bonded interactions between i-l is weighted by the 1-2
weighting coefficient in special bonds command. Is this correct?

The non-bond weightings pay no attention to diherdarl/improper topology.
Only to bond topology. So if I-J are bonded directly, that's a 1-2, no
matter what diherdarl/improper it might appear in.

2. For improper style harmonic (assuming the improper is ordered i-j-k-l),
the angle computed is the Wilson-out-of-plane angle between the plane of
i-j-k and the plane of j-k-l. This is same as the angle that is computed in
improper style cvff. Is this correct?

You'll have to look at the formulas on the doc page for these 2 improper
styles. if that isn't clear, I'd suggest a CVFF paper. Or the source code.