[Wien] SPIN ORBIT COUPLING and Band structure picture

[Peter Blaha]

LDA+U or EECE calculations require a spin-polarized setup.

You have to use  runsp_lapw; or when you want a non-magnetic solution,  runsp_c_lapw.

> 1) When I am choosing the non-spin polarized case, and run the -dm and -orb command with -so, it gives me an error stating
>
> ....
> LAPW1 END
> LAPW0 END
> ERROR: option -orb does not exist !
> ERROR: option -dm does not exist !
>
> In the meantime, when I run spin-polarized mode with inclusion of all of them, it runs smoothly without giving any error.
>
> ......
>   LAPW2 END
>   LAPW2 END
> LAPWSO END
>   LAPW1 END
>   LAPW1 END
>   LAPW0 END
>
> I observed many a times in non-spin polarized case that we can include the U and SO in lower version of wien2k (08).
> *Could anyone state what the exact reason can be?


NO. For sure LDA+U always required runsp_lapw.


> 2) Once the scf cycles complete successfully without error for LSDA+SO+U mode, we intend to perform DOS and Bandstructure calculations.
> While performing the Bandstructure calculations through w2web,
> you will observe both the spin-up and spin-down mode. *In that context, I suppose that when SOC is included, the energy bandstructure should also be coupled into one
> (single band structure)*. But it does not happen so. We are supposed to calculate xlapw1 -band up, xlapw1 -band dn, xlapwso -up/dn.
> *In this cases, could anyone tell me which bandstructure plot should be used (spin up or spin down) for interpreting the results because from literatures, authors show
> single band structure picture. My understanding may be wrong as I am new to SOC calculations. *

You have to run     x lapw1 -band up, xlapw1 -band dn, x lapwso -up     ONLY ! (no lapwso -dn)


-- 
-----------------------------------------
Peter Blaha
Inst. Materials Chemistry, TU Vienna
Getreidemarkt 9, A-1060 Vienna, Austria
Tel: +43-1-5880115671
Fax: +43-1-5880115698
email: pblaha at theochem.tuwien.ac.at
-----------------------------------------