It is not always so simple to find the charge state of an atom. It really depends on the material you are looking at.

But for Fe 2+ or 3+ in high-spin configuration of some Fe-oxide it is rather simple:

Fe3+ in high-spin conf. has 3d5-up and no 3d-dn electrons. This makes a spin-moment of 5 per Fe atom (if you have a ferromagnet, you may see :MMTOT a multiple of 5, but if it is a ferri or antiferromagnet MMTOT may be zero or different). The MMIxxx value of this atom will be large (around 4).

If it is Fe 2+, we have 3d5-up/3d1-dn; i.e. only a moment of 4.
The MMIxxx of Fe 2+ is usually around 3.6 muB.

Total charges within an atomic sphere are usually not suitable to decide on that.

An alternative is Bades AIM theory and the AIM program can calculate these Bader charges. However, also for these Bades charges, the differences between Fe 2+ or 3+ might be samll (maybe 0.1 e at most).

So the spin-analysis is best, if you have a high-spin state. Low spin-states and/or different symmetries (tetrahedral vs. octahedral,...) have different laws and you need some knowledge of crystal field theory to get an idea what can happen.

On 08/28/2017 05:11 PM, Fecher, Gerhard wrote:
Hallo Abderrahmane,
you find the occupation of the states (INSIDE THE SPERES !) in the 
case.scf2up/dn files.
See the lines with e.g.:
:PCS002: PARTIAL CHARGES SPHERE = 2  .........
:QTL002: ........
the content should be selfexplaining.

Take care, you are talking about a model, not about reality.
Therefore, do not wonder if you do not find integer occupations that you might 
expect from your chemical model.

You assume that only a certain, integer (!) number of d electrons contribute to 
the problem (here magnetization)
but why should this be the absolute and only truth in a solid ?
Just as an example: Why has bcc Fe a magnetic moment of about 2.2 muB and not 
exactly 0, 2, or 4 ?


DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:
"I think the problem, to be quite honest with you,
is that you have never actually known what the question is."

Dr. Gerhard H. Fecher
Institut of Inorganic and Analytical Chemistry
Johannes Gutenberg - University
55099 Mainz
Max Planck Institute for Chemical Physics of Solids
01187 Dresden
Von: Wien [] im Auftrag von pieper 
Gesendet: Montag, 28. August 2017 11:10
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] How to know if the M+2 has a high spin from calculation?

No, Wien2k probably won't 'determine alone which spin state is
energetically more stable'.

States with different spin configurations (PM, FM, AF, ...) frequently
have small differences in total energy but large differences in their
electronic configuration. Due to this the scf cycle tends to converge in
local minima associated with the starting configuration. Thats why you
should set different spin configurations in lstart and see where the scf
cycle leads you. The solution with lowest total energy is what you seek.

For your low/high spin configuration this might be enough. There are
more complicated cases where you may have to guide the scf to the
correct symmetry by populating the density matrices accordingly - this
has been discussed in the mailing list a lot - see also option -orbc in
the UG.

Good luck

Martin Pieper

Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Tel.: +43-(0)316-380-8564

Am 27.08.2017 13:06, schrieb Abderrahmane Reggad:
Hello again

I have found in the literature that the spin state configuration is
like the magnetic configuration. So we have to make 2 calculations:
one for the high spin configuration and another one for the low spin
configuration and we look after for the configuration more
energetically stable.

I want to know if the things are so in the wien2k code or it behaves
differently and can determine alone which spin state is the more
energetically stable.

Best regards
Wien mailing list
Wien mailing list
Wien mailing list


Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email:    WIEN2k:
Wien mailing list

Reply via email to