Simulating Charges in VASP

[stijn_devos]

Dear VASP team, I have a question regarding the simulation of charged systems with VASP. Lets say for example that I want to study the behavior of Na+ which has 10 electrons, if I want to tell VASP that I have this net charge present in my system I do this in the INCAR under the command NELECT=10. Wont this create an infinite charged system under periodic boundery conditions? And therefore introduce a lot of artifacts in the simulation?
And if the VASP software would automatically add a charge of -1 to the system, am I then not just simulating Na, a radical with completely different behavior than Na+? Thank you for your time!

[alexey.tal]

Dear stijn_devos,

We have a very detailed guide on how to deal with charged cells on our wiki.
But in short, if you change NELECT from default and your cell becomes charged, VASP will add a homogeneous background charge to compensate it.

Best,
Alexey

[stijn_devos]

Hi Alexey, thank you for your explanation and in the meantime i have been reading more about the background charge added. i do have one more question:
Members of our team say the best way to simulate a charged system is to add a counter ion of opposite charge. In the example of Na+ this could be for example Cl- that you place far away so that it doesnt interact with your desired Na+ simulation. The downside of this technique is that you have to enlarge your unit cell. But i personally think that you need to set NELECT=10 so that a background charge of -1 is added. Which technique do you think is best? Regards, Stijn

[alexey.tal]

By placing an ion of the opposite charge you would still get an interaction between the charges themselves for any practically possible cell sizes. Furthermore, we will get a dipole-dipole interaction between periodic images, which needs to be corrected for. Thus, I don't see what would be the advantage of this approach.