[Wien] Charged supercells

[Peter Blaha]

For charged supercells:

1) Setup your supercell, put impurities, vacancies,.. (all with neutral atoms!)

2) Run a standard initialization (init_lapw) with NEUTRAL atoms.
Don't fiddle around with case.inst (positively charged atoms may work (and
one can neglect the warnings "NUMBER OF ELECTRONS NE IZ "), but most
negative ions are unstable in DFT and will not work.

3) run_lapw  (again for neutral atoms !) . Save the results "save case_neutral"

4) Edit case.in2: In the second line you see the number of electrons NE.
   Add or remove an electron to get the desired charged state.

5) Edit case.inm: The first line must read:

BROYD  1.0   YES        (when you added an electron at step 4     or
BROYD -1.0   YES        (when you removed an electron at step 4

The position of   "YES" is important!!! When reading characters in Fortran
one must use a fixed format (or put characters into quotes).
(In your mail the YES was misplaced)

6) run_lapw

You see, I NEVER run   init_lapw again!!!  Usually  init_lapw should be run
only ONCE in a directory, because any changes which would require rerunning
init_lapw (symmetry changes,....) should be done in a fresh "case" (maybe
except changing RMT, but for this purpose I'll provide another tool with the
next update).

It is NOT necessary to start an scf cycle with a density which has the
"correct" number of electrons for these charged cells.

> I have a couple of doubts about charged supercells.

> 2 My second doubt deals with case of a charged vacancy or an impurity
> complex. I have not found information about this in the mailing list. The
> point is ... where do I add/remove electrons in the case.inm file in these
> cases? Say I study the N vacancy in AlN. With all the neutral atoms I find
> the expected additional eigenvalues: a a1 (s-like) state just above the
> valence band (occupied by two electrons) and three degenerate t2 (p-like)
> states, which are resonances in the conduction band, and are occupied by
> just one electron. Now, I should like to remove this last electron,
> creating a VN1+, i.e., the vacancy in a 1+ charge state. How do I do it?
> Should I remove one electron of, say, one of the four neighboring Al atoms?
> The same question raises for impurity clusters: If I work with, say, a MgO
> cluster in AlN, and I find that occupied states appear in the band gap, how
> do I give the cluster a 1+ charge state? Is there a rule for deciding from
> which one of the atoms in the cluster should I remove 1 electron?

I guess with my explanations above the answer to your point 2) is clear.
You cannot (and need not) remove an electron from a particular atom.
All you can specify is the total number of electrons in your supercell!!!
During the scf cycle the system will decide itself if an electron comes/goes
from/to a particular site.

                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------