[Wien] spin-orbit coupling and partial charges

[Ulrich Wedig]

Using WIEN2k 9.1, I tried to investigate partial DOS's of a system
including spin-orbit coupling. After some problems, I generated a more
simple case, NaTl, in order to look for their cause. I have attached the
 original case.struct file and the one generated by symmetso.

I performed calculations with and without spin-polarization, with and
without spin-orbit coupling, sequential and parallel. The partial
charges were computed either with lapw2 -qtl or with the qtl program.
As a result I want to summarize three features:

- lapw -qtl considers the multiplicity when calculating the sum of the
partial charges for a given atom (-> 'atomic DOS'). Thus (1 - sum of
'atomic DOS's') gives the correct interstitial-DOS (from the
(nat+1)-rows in the case.qtl file) in cases without spin-orbit coupling.
In contrast, the qtl program does not consider the multiplicity. The
interstitial-DOS is not correct in any case. The total DOS is not
affected thereby in non-spin-polarized systems, even with spin-orbit
coupling. However in the case of both, spin-polarization and spin-orbit
coupling the total DOS, which is then equal to the (nat+1)-values varies
with the QSPLIT parameter (-2 vs. 0 e.g.).

- In a non-spin-polarized. spin-orbit case, being executed in parallel,
lapw1 writes the files case.vectorso_1_dn_1, case.vectorso_2_dn_2.....
If I calculate the partial charges with the qtl program, though only if
QSPLIT is set to 0 or -1, the program stops with the message given in
the attached file error_qtl. If I rename case.vectorso_1_dn_1 to
case.vectorsodn_1 and so on, the qtl program works properly. By the way,
in a spin-polarized case, lapw1 writes to case.vectorsodn_1,
case.vectorsodn_2..... directly.

- If I use the original case.struct file in a non-spin-polarized
calculations, several messages 'WRONG SYMMETRY' appear in the
case.outputso file, although there is no need for a symmetry lowering.
With the case.struct file generated by symmetso, these messages
disappear. The results however are not affected by these messages.

Best regards

Ulrich Wedig

-- 
-----------------------------------------------------------------
Dr. Ulrich Wedig                              Tel. 0711/6891535
Max-Planck-Institut fuer Festkoerperforschung FAX  0711/6891502
Heisenbergstr. 1
70569 Stuttgart                               U.Wedig at fkf.mpg.de
-----------------------------------------------------------------

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: error_qtl
Url: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20090318/8a71d357/error_qtl.ksh
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: NaTl_2.struct
Url: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20090318/8a71d357/NaTl_2.ksh
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: NaTl_2.struct_so
Url: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20090318/8a71d357/NaTl_2-0001.ksh