I am trying to simulate bulk water using various models based on the Lenard Jones potential. This also includes a Coulomb potential between atoms belonging to different molecules. When using a cutoff for the coulomb part of the interaction should this be applied at a per-molecule basis? i.e. I cannot just use a hydrogen atom that happens to be within the cutoff radius but not the rest of its molecular partners. I find that this can make a big difference. Is this right? Thinking about charge conservation it makes sense. However, I have not simulated such systems before and I want to make sure.
Hi Theodoros,
This is a good question. I have to admit, although I have experience with implementing Ewald sums for Coulomb interactions, I have never had to apply it to molecular systems. However, if I were faced with such a problem I would probably define the molecule with respect to its center of mass and associate the total charge of the molecule with that center of mass. Therefore, when considering the cutoff radius, you would need to examine whether or not the center of mass of the molecule in question falls within the cutoff radius.
I hope that helps.
Thanks for your answers. Dear Adib what you are saying makes sense. I guess I cannot only consider a part of a molecular charge, especially for a neutral molecule, so it makes sense to assign the charge to the center of mass.
In the papers people usually give the cutoff (~9AA for water). Eventually what Achraf suggests is the correct way for validation. However, this is a subtle point that is not clear from the literature.
Again thanks for your commends.
- Badges
- Science topic