[Wien] In and spin orbit interaction

[Peter Blaha]

You must do something terrible wrong, but of course I can't say what.

It does not happen by default. The SO splitting of the d-states is rather
small and of course they remain at the lowest part of the valence bands and
are fully occupied. There are no qtl-b warnings either.
Starting from a scf non-so calculation inclusion of SO converges in 3 iterations.

Don't use relativistic LOs! (there's no p semicore in In).

Gerhard Fecher schrieb:
> In calculations without SO, the In 4d states are well below the valence band.
> 
> If switching on spin orbit interaction the 4d states move to the Fermi energy and the accompanied band about half filled and occupied by only about 5 
> of the d electrons. This appears for element In (tetragonal) as well as in cubic ternary compunds.
> 
> Any idea how to prevent that (may also effect Ga 3d electrons), so far I had no problems with other semi-corelevel that are close to
> the valence band like p states in alkali metals when calculating with SO interaction.
> 
> Is there any reason that the QTL-B errors appear heaped when using SO ?
> (mainly at the beginning of scf calculations, unfortunately I could not figure any clear systematic behind it up to now.)
> 
> Ciao
> Gerhard
> 
> ====================================
> Dr. Gerhard H. Fecher
> Institut of Inorganic and Analytical Chemistry
> Johannes Gutenberg - University
> 55099 Mainz
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

-- 

                                       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/
--------------------------------------------------------------------------