[Wien] QTL Projection f-7/2

[D.A. Tompsett]

Dear Peter,
          thank you for the help on the f-7/2 projections. I was able to 
get the f-7/2 projection of the DOS to work well for the simple Hf example 
that you suggested.

However, when I attempt to use QTL on my f-system which is Yb based I 
receive strange errors. With QSPLIT=0,-1 I receive: *** glibc detected *** 
malloc(): memory corruption: 0000000000071eb00 *** forrtl: error (76): IOT 
trap signal

and the process hangs in the shell. If I remove the Yb atoms from 
consideration in case.inq the problem disappears. Also, I attempted 
removing some orbitals from case.inq for the Yb atom, but the problem still 
occurs even if I only consider the s(L=0) projection.

With QSPLIT=-2 I receive: forrtl: severe (64): input conversion error, unit 
29, file /home/dat36/Wien2k/Calcs/YbSOSp/BackupSO/DOS/YbSOSp/fort.29 Image 
PC Routine Line Source qtl 00000000004ED992 Unknown Unknown Unknown qtl 
00000000004EBE76 Unknown Unknown Unknown qtl 00000000004C28A0 Unknown 
Unknown Unknown qtl 0000000000497BB2 Unknown Unknown Unknown qtl 
0000000000497813 Unknown Unknown Unknown qtl 00000000004AA977 Unknown 
Unknown Unknown qtl 000000000041C6A2 Unknown Unknown Unknown qtl 
000000000040E748 Unknown Unknown Unknown qtl 00000000004131EF Unknown 
Unknown Unknown qtl 0000000000403BEA Unknown Unknown Unknown libc.so.6 
0000003179B1C40B Unknown Unknown Unknown qtl 0000000000403B2A Unknown 
Unknown Unknown 234.150u 48.660s 4:43.31 99.8% 0+0k 0+0io 0pf+0w error: 
command /home/dat36/Wien2k/qtl upqtl.def failed

upqtl.error simply reads 'Error in QTL' I looked through the other output 
files and can not find any information about the error.

Thank you for any help,
David.

On Aug 26 2008, Peter Blaha wrote:

> At present (this constraint will be relaxed in the next version) any 
> QTL-calculations including SO must be done with a spin-polarized setup 
> (even in a non-magnetic case, you have to initialize "spin-polarized", 
> but then can use "runsp_c -so ", and x qtl -up -so
>
> Yes, you have to use -1 or 0 for QSPLIT. Look into the header of case.qtl 
> or case.qtltext, where it is indicated (there might be a slight 
> inconsistency in the present release?).
>
> I suggest you do simple Hf first, where the semicore 4f states split and 
> it should be rather obvious what is 7/2 and ...
>
>D.A. Tompsett schrieb:
>> Dear All,
>>         I am using the latest WIEN2k_08.2.
>> 
>> I am working on a 4f metallic system and am interested in using QTL to 
>> investigate the occupation of the J=7/2 f-states. I would like to get 
>> the charge projection of the J=7/2 states to be able to make some 
>> comment about the valence of my system.
>> 
>>>From the documentation I can not see how to set the parameters in the 
>> case.inq file for QTL. The UG refers to 'QTL - technical report', but I 
>> have been unable to find it on the Wien2k site or the web.
>> 
>> I suspect that I will need QSPLIT=-1 or 0, but do not know how to refer 
>> to the states in this basis.
>> 
>> Thank you for any help,
>> David A. Tompsett
>> Quantum Matter Group
>> Cavendish Laboratory
>> 
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>

-- 
David A. Tompsett
Quantum Matter Group
Cavendish Laboratory
J. J. Thomson Avenue
Cambridge CB3 0HE
U.K.
Tel: +44 7907 566351 (mobile)
Fax: +44 1223 337351
http://www-qm.phy.cam.ac.uk/