Dear LAMMPS users,
I woud like to ask if there is a possibility to mimic something like
a Kern-Frenkel patchy particle model used in Monte Carlo simulations by a
simple overlap of two atoms to create a “Janus dumbbell”?
By the overlap I mean the slight shift of one atom with respect to another.
I know that there is a possibility to do that with standard LJ potential using special_bonds command, but by doing that we lose the repulsion on one side (we attain one hard sphere) and we can get only one attractive part.
However, to be closer to, for instance, electrostatics,
yukawa potential would be better, but as far as I know this feature is not implemented. This in consequence results in high intramolecular energy contributions, which are not desired.
Is there any other possibility to get rid of those high intramolecular energies contributions?
Sincerely,
Lukasz Baran
PhD Candidate
Department of Theoretical Chemistry
Institute of Chemical Sciences
Faculty of Chemistry
Maria Curie Skłodowska University in Lublin
Dear LAMMPS users,
I woud like to ask if there is a possibility to mimic something like
a Kern-Frenkel patchy particle model used in Monte Carlo simulations by a
simple overlap of two atoms to create a “Janus dumbbell”?
people seem to be doing this by assigning a molecule id to the constituent atoms of each patchy particle and then use fix rigid/small for time integration.
By the overlap I mean the slight shift of one atom with respect to another.
I know that there is a possibility to do that with standard LJ potential using special_bonds command, but by doing that we lose the repulsion on one side (we attain one hard sphere) and we can get only one attractive part.
you are not making sense here. the special_bonds exclusion only happens, if there is a bond between those two particles (which is doable). there is no hard sphere resulting from that or a lost interaction from one side.
since interactions inside a rigid particle have no impact, they should be removed, simply to avoid adding numerical noise to the force calculation and spurious contributions to the virial.
However, to be closer to, for instance, electrostatics,
yukawa potential would be better, but as far as I know this feature is not implemented. This in consequence results in high intramolecular energy contributions, which are not desired.
Is there any other possibility to get rid of those high intramolecular energies contributions?
if you assign molecule ids as mentioned above, you can use “neigh_modify exclude molecule/intra”
axel.