[Wien] experimental vs relaxed parameters using wien2k

[Fecher, Gerhard]

Dear Tristan,
don't you think that this is a rather strange advice,
it might suggest that the optimized lattice parameters are bad, whereas any arbitrary choosen lattice parameters are good.

I  do not think it is a good idea to vary the lattice parameters until an arbitrarily choosen property fits an experiment.
I guess if you do that for the band gap, the hyperfine field, the conductivity, the magnetic moment, the optical spectrum and whatever else
you will have a plenty lot of fitting lattice parameters that all do not agree with the optimized ones.

However, one should point on an important fact: there are probably more than one minima in the total energy (or forces, or both)
and by optimization one may not alwys find the absolute minimum, so care has to be taken what is called "optimized" lattice parameter.
(Maybe different start parameters help to go out of a local minimum, but the next found ... ==> be careful with your interpretation !!)

Hmmm wasn't it that the A in LDA, GGA, etc.  stands for approximation ;-)

Final remark: In the case under discussion one would need to calculate the electronic structure for about 100 "experimental" sets of lattice parameters
found in Pearsson's crystallographic database and then decide which ones fit the magnetic moment of Fe best 
(or at least has the lowest energy in the antiferromagnetic state, in which collinear or non-collinear, etc.). 
If one is not satisfied  with PBE then one still may use some other functionals or some U with size of choice of the day ;-)

Ciao
Gerhard

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
and
Max Planck Institute for Chemical Physics of Solids
01187 Dresden
________________________________________
Von: Wien [wien-bounces at zeus.theochem.tuwien.ac.at] im Auftrag von Tristan de Boer [tristan.deboer at usask.ca]
Gesendet: Dienstag, 25. Juli 2017 17:55
An: wien at zeus.theochem.tuwien.ac.at
Betreff: Re: [Wien] experimental vs relaxed parameters using wien2k

You can find many papers that use either approach. I considered this
somewhat tangentially in a paper on several semiconductor polymorphs
that were synthesized under different pressures. [1] We collected x-ray
absorption and emission spectra on thee polymorphs, and compared our
spectra to those calculated using the PBE GGA functional for both
relaxed and unrelaxed atom positions. We found that there was worse
agreement with experiment using relaxed parameters compared to the
experimental parameters, so at least for this case, using the
experimental structure yields more correct results. I'm not sure how
universal this is, and it may not apply as you move higher up the
'Jacob's ladder' of density functional approximations.
Best Regards,
Tristan

[1] doi:10.1103/PhysRevB.93.155205


On 2017-07-24 6:27 AM, Chouaib AHMANI FERDI wrote:
> Greatings Wien users,
>
> I just want to ask if the experimental parameters (lattice parameters,
> internal parameter, atomic positions, ...) are already available for a
> certain compound, would it be better to run SCF cycle with them or to
> make preliminary relaxation and volume optimization and then perform SCF
> cycle using the calculated parameters.
>
> Faithfully,
>
>
>
> --
> AHMANI FERDI Chouaïb
> Ph.D Student in Material Science
> Ecole Normale Supérieure
> Mohammed V University, Rabat.
> Tel : +212 6 94 59 57 60
>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>
_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html