Dear researchers,
When modelling a simulation cell to study the adsorption of Phosphate ion (PO4-3, HPO4-2, and H2PO4-) on a crystal surface (Goethite), there is a charge imbalance due to the phosphate ion.
Is it necessary to neutralize this extra charge?
If required, What would be the appropriate method to neutralize the charge?
If you really need to neutralize the surface you can "protonate" it at some point. Since Goethite is FeO(OH) the surface will be a wild distribution of protonated and unprotonated sites, anyway.
Please be aware that this may generate some sort of "unit cell dependency" as well as convergency issues, so please check at an early stage that there is really an extra charge on your anion, otherwise your calculation will just converge to "weirdness". Eventually you will have to reassign occupation numbers manually. I never did that in VASP, but it should be possible in any quantum chemistry package. If your SCF still moves away from it, you might have to play with the SCF optimization parameters. If all of that has worked, you still have to check how big your unit cell needs to be so that the extra proton does not directly affect the local geometry at the adsorption site.
That's quite extensive, but surfaces are the hardest object to calculate in DFT. If someone has an easier idea, folllow that, however.
It petty much depends on your problem. If you are interested in studying the chemical reaction of charged ions with an initially charge-neutral surface, there is no need to add a counter-charge to neutralize the whole system. On the other hand, if you consider only a certain portion of a molecule with a particular ionic charge, you then have to add extra charges to null the whole entity. In this case, you can define the position of the extra charges and add additional monopole, dipole and quadruple corrections to get your ideal results and performance.
Dear Mohammad Saeed Bahramy , thank you very much for your suggestion.
Dear Jürgen Weippert Thank you for your suggestion. I agree with you, that the goethite surface is already having wild number of protonated and unprotonated sites after cleaving at particular crystal plane. As you mentioned, the convergence is the main problem when we do structural relaxation along with a charged molecule. It's hard to find the stable ground state, even for a small supercell. Anyway, thank you for your suggestion.
Dear Nam Van Tran, please read the following research articles for your references. I hope it might help.
Dear Nam Van Tran Yes, the energy calculation wouldn't be a reasonable one. Since the charged molecules' potential energy is generally infinite. But, I was trying to figure out a possible way to study the adsorption of charged molecules which still has lots of challenges.
Let me know if you have any suggestions.
Thanks in advance.
Dear Nam Van Tran
Thanks for your suggestions. I agree with you and I have also read some articles where they have added some water molecules directly into the simulation cell instead of using the solvation model and run MD relaxation using VASP. And finally corrected the energy of charged molecules as well. Meantime it is quite complicated as well as time-consuming.