[Wien] about how to decide the bacground charge

[Stefaan Cottenier]

> So what is it that you want to do?  Remove a Ti atom and replace it by a
> Nd3+ ion?  In that case a background charge of -3 in case.inm will keep
the
> unit cell neutral.
>
> > usually the Nd ion is +3, so, when Nd substitutes for the Ti, then the
> > total charges in TiO2 is not balanced. I wanna to know how to decide the
> > background charge in this case.

[Let's see whether today I'm more awake ;-) ]  I don't understand the
original question very well, and it could be there is some confusion there.
Let us assume you have a supercell of TiO2, where some Ti are replaced by Nd
(that is what you want, right ?). Automatically, this supercell is neutral.
In nature, Nd is usually in a +3 state, but that does not affect the overall
neutrality: that just means that 3 electrons from Nd are delocalized and
participate in bonding (note: for lanthanides, LDA/GGA are not really
correct, your selfconsistent solution might not be a truethful
representation of a +3 ion. LDA+U can be a help here.)

What the background charge option in case.inm is meant for, is a different
situation. In nonconducting materials, a piece of material can (locally) get
charged (=non-neutral). An example (treated with WIEN) are group V acceptors
in CdTe (S. Lany et al, PRB 62(4) (2000) R2259), where an extra electron is
added to the unit cell to make it non-neutral.

Make sure you know which of both situations you want to calculate. In the
first you don't have to add charges, in the second you have.

Stefaan