Interaction forces between bonded and non-bonded atoms

Dear Everyone,

I have a simple question about the interaction forces between bonded and non-bonded atoms. Please see the figure attached for the two statements, I am not very sure if they are true for the force calculation in Lammps.

Thank you very much for any comments,

Lili Zhang

Question about interaction forces in Lammps.jpg

Dear Everyone,

I have a simple question about the interaction forces between bonded and
non-bonded atoms. Please see the figure attached for the two statements, I
am not very sure if they are true for the force calculation in Lammps.

the answer, as usual, is: it depends.

whether the non-bonded interaction between i and j is computed depends
on the parameters used with the special_bonds command. the default
settings are so that those interactions are indeed excluded (bonds and
angles). mind you, this is true for the *direct* interaction. if you
have a periodic system with small dimensions and a large cutoff or use
long-range electrostatics, you'll still have contributions to the
forces between i and j due to their respective mutual periodic images.

is that confusing enough? :wink:

ciao,
     axel.

Dear Dr. Kohlmeyer,

Yes, the interaction force calculation is more complex than I thought. :slight_smile:

Let me explain why I concern this:

I am doing my own calculation for different interaction forces for class2 force field.
After comparing my results with Lammps force calculation, they don’t match exactly. So I doubt that I might make a mistake considering bond and non-bond interactions.

I am comparing each term for : bond, angle, lj, dihedral, improper, electrostatics, respectively.

Thanks a lot,

Lili