[Wien] error in spin-orbit coupling

[Peter Blaha]

The generated inso file is clearly wrong. Probably a bug of the new initso_lapw script.
(Please send the case.struct file to my personal email)

The first guess is always to follow the "defaults", and the default would
say: "NO RLOs should be added". This is in any case better then when you
use thew in a wrong way.

However, when you want to improve the description of the SO splitting of p-SEMICORE states,
then, without knowing the struct file, it seems rather clear, that
"number of atoms with RLO" should be 5  (only these atoms have semicore p-states and
"hopefully" these are "heavy" atoms).
Then add the corresponding 5 lines for atoms 1-5 and finish with

"0 0         number of atoms without SO, atomnumbers"


Am 12.03.2012 06:47, schrieb Madhav Ghimire:
> Dear Wien users,
>     I have one system with eight atoms. For this, I intend to run spin-orbit coupling to the total atoms present in it. When I chose the default values in *case.inso* for
> selecting all atoms then it gives the data as appended below
> WFFIL
> 4  0  0                 llmax,ipr,kpot
> -10  1.5                Emin, Emax
>      0 0 1                           h,k,l (direction of magnetization)
>   14                       number of atoms with RLO
> 1 -1.80 0.002 CONT             atom-number, E-param for RLO
> 2 -1.80 0.002 CONT             atom-number, E-param for RLO
> 3 -1.80 0.002 CONT             atom-number, E-param for RLO
> 4 -3.45 0.001 STOP             atom-number, E-param for RLO
> 5 -3.45 0.001 STOP             atom-number, E-param for RLO
> 6 0.30 0.000 CONT             atom-number, E-param for RLO
> 6 0.30 0.000 CONT             atom-number, E-param for RLO
> 6 0.30 0.000 CONT             atom-number, E-param for RLO
> 0  7 8                     number of atoms without SO, atomnumbers
> *Now instead of total atom=8, it shows 14 with RLO and rather than having 7 and 8 atoms **in 12 and 13th row**it shows 6 atoms.
> Similarly, for the 2nd and 3rd column of 14th row, 7 and 8 is shown which in my opinion should be zero because all the atoms are selected with SO.
> After this, (i mean with and without editing to the above mentioned terms in case.inso),when I use this case.inso and run it gives me an error as shown below.*
>
>>    stop error
>
> error: command   /misc/home/ghimire/WIEN2k_11/lapw2c lapw2.def   failed
> 0.194u 0.059s 0:00.98 24.4%	0+0k 17280+0io 23pf+0w
>>    lapw2 -c -so   	(12:56:29) Abort
> 2813.817u 38.733s 41:43.64 113.9%	0+0k 3377456+0io 41pf+0w
>>    lapwso 	(12:14:45)    gamma not equal 90
>>    lapw1    	(11:42:49) 3144.324u 44.979s 31:53.92 166.6%	0+0k 29544+0io 96pf+0w
> 20.171u 0.217s 0:21.03 96.9%	0+0k 28032+0io 110pf+0w
>>    lapw0 	(11:42:28)  gamma not equal 90
>
>      cycle 1 	(Mon Mar 12 11:42:28 JST 2012) 	(40/99 to go)
>
>      start 	(Mon Mar 12 11:42:28 JST 2012) with lapw0 (40/99 to go)
>
>
> *Could anyone help me to sort out this error caused in scf cycle. Is there any way to edit the atom numbers in RLO.
> PLease also note that I did this for both non-spin polarized and spin-polarized case. I observe the similar type of errors.
> *
> Thank you in advance.
> Best wishes
>
> --
> M. P. Ghimire
> NIMS
>
>
> _______________________________________________
> 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: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------