[Wien] magnetism calculation in Wien2K

[Gerhard Fecher]

Just another example:
Indeed the problem is that the paramagnetic calculation (including a non-ferromagnetic solution of a spin-polarized calculation)
will give a zero magnetic moment. From suszeptibility measurements you will find the magnetic moment in the paramagnetic state.
Now you may have the problem how to compare that to the electronic structure calculation (just forget a moment about ground or excited state)
For rare earth elements, the FM or AFM solution may be fine, but for others ?

The general problem with the FM or AFM solutions will be that you imply
a very particular symmetry of the spin arrangement that may lead to unrealistic results.
Therefore, spin spirals using the non-collinear spin code might be a better approach to find some reliable solutions.
(there should be some work published by Leonid Savatzky).

The last remark of Stefaan has to be taken very seriously, as it reminds one
that the A in LDA stands for approximation.
There are some nice examples in literature how to find from the total energy that the ground state of Fe is fcc and anti-ferromagnetic
(unfortunately (or luckily) I did not yet succeed with Wien2k, seems it is too good).
On the other side one finds just recently a lot of strange ground states published, and thats much less funny.

Ciao
Gerhard


________________________________________
Von: wien-bounces at zeus.theochem.tuwien.ac.at [wien-bounces at zeus.theochem.tuwien.ac.at] im Auftrag von Stefaan Cottenier [Stefaan.Cottenier at fys.kuleuven.be]
Gesendet: Montag, 10. Dezember 2007 19:20
An: wien at zeus.theochem.tuwien.ac.at
Betreff: Re: [Wien] magnetism calculation in Wien2K

This will be far from the definitive answer, but if you ask which (NM
or FM) would be the better crude approximation for a paramagnetic
state, I would guess it depends on the quantity you are interested in.

What I am sure about: very local quantities (such as interatomic
distances, and the EFG which can be sometimes very much dependent on
them) in a a PM are better described by the FM solution.

What I am just guessing: properties that are more of an averaged
nature (can we consider lattice constants as an example?) could be
better described by the NM solution.

I vaguely remember to have had a case which was definitely
(experimentally) a paramagnet, and the FM total energy was lower than
the NM one. There the disorder of the moments introduced only a minor
effect on the total energy. But I'm not sure how general this is.

Stefaan


Quoting villesuzanne <ville at icmcb-bordeaux.cnrs.fr>:

> Dear Stefaan,
>
> The paramagnetic case is an interesting, very basic issue to discuss here.
>
> Non spin-polarized calculations are often referred as "non-magnetic" in
> literature, since obviously there is no resulting magnetic moment nor spin
> density (both spin channels have the same electron density).
>
> If I am right (not an expert in magnetism), a paramagnetic phase has
> well-defined local magnetic moments, for example on transition metal ions,
> and these moments are disordered.
>
> Therefore, although the macroscopic magnetization is zero in a paramagnet
> (without an external field), the spin-polarized (FM) solution is maybe a
> better approximation for a paramagnet: Only spin-polarized calculations will
> give a net magnetic moment on the magnetic ion.
>
> The non-magnetic approximation of a paramagnet is equivalent to define it as
> a metallic system (with not even local magnetic moments), that becomes
> magnetic under application of a magnetic field.
>
> Definitely, a paramagnet cannot be described exactly by a periodic
> calculation. But what is the better crude approximation, NM or FM ?
>
> Regards,
> Antoine
>
>
>
> In other words,
>>      How to distinguish the paramagnetism (PM),
>> ferromagnetism (FM) and
>> antiferromagnetism (AFM) of materials, besides the
>> comparing the
>> their total energies?
>
> Comparing total energies is usually the way to go. You might play with
> the non-collinear code to find the orientations of the moment
> 'automatically', always within what your (super)cell allows. The
> Stoner criterium applied to a paramagnetic DOS might give information
> too.
>
>>      If the calculated result shows that the system
>> have magnetic
>>  moment,
>> must the material be FM? Do the material have the
>> possibility of PM ?
>
> Sure it can be paramagnetic, it all depends on the total energy. It
> can be AF as well... A truly paramagnetic state cannot be calculated
> in wien2k, and is approximated by a 'non-magnetic state' (which
> Gerhard Fecher called a non-word once in this list).
>
> Stefaan
>
>
> Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>



--
Stefaan Cottenier
Instituut voor Kern- en Stralingsfysica
K.U.Leuven
Celestijnenlaan 200 D
B-3001 Leuven (Belgium)

tel: + 32 16 32 71 45
fax: + 32 16 32 79 85
e-mail: stefaan.cottenier at fys.kuleuven.be


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien