[Wien] band strucuture for spin polarized case with spin-orbital coupling

[Andrei Postnikov]

On Thu, 27 Nov 2003, stargmoon wrote:

| Dear Wien2k user,
|
| Have you ever used Wien2k to calculate the band structure
| for spin polarized case with spin-orbital coupling?

Yes

| I did it by following procedure:
| x lapw1 -up
| x lapw1 -dn
| x lapwso -up
| x lapw2 -c -so -qtl -up
| x lapw2 -c -so -qtl -dn
| x spaghetti -up -so
| x spaghetti -dn -so
| Now the problem is, after lapwso, the eigenvalues are stored in file
| "case.outputso", because of spin-orbital coupling, we can't distinguish
| up and down spin. How can I get the band structure for up and down spin
| ( or majority and minority spin)?

Yes the eigenvalues in case of spin-orbit cannot be clearly separated
into spin-up and spin-down. However, the eigenvectors contain
the information on the amount of spin-up and spin-down in each solution.
The most straigtforward thing you can do is to plot band structure with
orbital character, e.g.,
x spaghetti -up -so
will make a plot with emphasizing spin-up contributions of whatever atom
(and l-channel, or summed up over all l-contributions),
and
x spaghetti -dn -so
the same band structure but with spin-down emphasized.
Probably you can only do it for one atom at a time separately and not
summing up over all atoms without changes in
the script, but the most clear separation normally happens just
for certain (localized?) states...

Good luck,

Andrei Postnikov

 +-- Dr. habil. Andrei Postnikov ----- Tel. +49-541-969.2377 -- Fax .2351 ---+
 | Universitaet Osnabrueck - Fachbereich Physik, D-49069 Osnabrueck, Germany |
 +-- apostnik at uos.de --------- http://www.home.uni-osnabrueck.de/apostnik/ --+