[Wien] Problems in generating case.struct using shell script and overlapping of spheres with error in lapw1

[Madhav Ghimire]

Dear Peter Blaha and wien users,
     I am trying to run a very simple compound Bi2Se3 with the given
lattice parameters and atomic positions for space group R3m (166).
The atomic positions along (x, y, z) are as follows:
Bi = 0, 0. 0.40046
Se (1) = 0, 0, 0.2097
Se (2) = 0, 0, 0
With the above given parameters, I initiated the calculations, but when I
select l start =-6 or -9 or any value between -1 to -9.9 for energy to
separate core and valence states, it shows that core charge leaks out of
the spheres. Then the calculations stuck there itself. The error to this is
as listed below:
Commandline: *x lstart *
Program input is: *"13 -6.0 "*

  SELECT XCPOT:
  recommended: 13: GGA (Perdew-Burke-Ernzerhof 96)
                5: LSDA
               11: GGA (Wu-Cohen 2006)
  SELECT ENERGY to separate core and valence states:
  recommended: -6.0 Ry (check how much core charge leaks out of MT-sphere)
 WARNING:  0.328380476966373       CORE electrons leak out of MT-sphere !!!!
 WARNING:   1.30946564796119       CORE electrons leak out of MT-sphere !!!!
 WARNING:   1.30946564796119       CORE electrons leak out of MT-sphere !!!!
LSTART ENDS
0.537u 0.023s 0:00.59 93.2%	0+0k 8+0io 0pf+0w


 When I perform for other compounds complexer than Bi2Se3, I resolve the
problem easily by changing its RMT value or lattice parameters or the
energy to separate core/valence states. Here I do not understand why this
case arises. This case occurs especially for wien2k 05 to 09 version. I did
the same calculation using wien2k 11 version, here I got through this
problem by selecting 1.0 as the alternative option given as ALTERNATIVELY:
specify charge localization (between 0.97 and 1.0) to select core state.
Then once I complete initialization part and give a run scf cycles, I
observe error in show dayfile as:

start 	(Fri Feb 17 12:32:41 JST 2012) with lapw0 (40/99 to go)

    cycle 1 	(Fri Feb 17 12:32:41 JST 2012) 	(40/99 to go)

>   lapw0 	(12:32:41) 11.846u 0.116s 0:12.02 99.4%	0+0k 0+4712io 0pf+0w
>   lapw1  -up    	(12:32:53) 150.227u 0.014s 2:30.80 99.6%	0+0k 0+1088io 0pf+0w

>   stop error

To clarify this, I checked the error file and error is observed in lapw1 as
listed below:
Error in LAPW1
 'SELECT' - no energy limits found for L=
2
 'SELECT' - E-bottom  -10.46100   E-top
-200.00000
I do not understand why this case is observed in this particular compound
although there had been a work on the same material using wien2k code.
Apart from this, I have a question regarding *"How to generate a
case.struct without using the w2web". I hope the experts in wien2k
generally use shell script rather than the interface so hopefully you can
instruct the step by step commands for generating case. struct.
*Thank you in advance
Best regards
-- 
M. P. Ghimire
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20120217/118cf2e2/attachment.htm>