[Wien] symmetso failed to initialize in Hematite job after running eece

[Peter Blaha]

You have done the right thing and removed the call to these subroutines,
which work only for LDA+U, but not for EECE.

Alternatively (and simpler) would have been, to remove the case.inorb
file (as this is rewritten anyway by runeece all the time)

and move case.vorbup/dn files temporarily to a different name.

Please note, however, that when the number of inequivalent atoms/ 
multiplicities have changed, you have to adapt case.vorbup/dn and in 
particular case.ineece manually.



On 01/08/2013 10:06 AM, Francisco Garcia wrote:
> Dear Prof. Blaha,
>
> There seems to be a bug in WIEN2k 12 when I try to run an SO
> calculation after first running an EECE calculation on Hematite.
> Basically this is the problem:
>
> Step 1---->run a usual DFT (non-SO) calculation [runsp_lapw –i 100 –ec
> 0.00001 –cc 0.0001]
> Step 2----->continue step 1 with a EECE calculation (25% HYBR)
> [runsp_lapw –i 100 –ec 0.00001 –cc 0.0001 -eece]
> Step 3----->initso_lapw with the intent of continuing step 2 with SO
> correction [runsp_lapw –i 100 –ec 0.00001 –cc 0.0001 –eece -so]
>
> The problem is that initso_lapw in step 3 gives an error when symmetso
> is called. Apparently, the problem arises from the routines inorbch
> and vorbch called in lines 421 and 422 in symso.f in SRC_symmetso.
>
> inorbch.f reads in case.inorb (unit 31) commencing from lines 17--31.
> However, there is a problem reading line 25. Similarly, vorbch.f reads
> case.vorbdn (unit 32) and case.vorbup (unit 33) from lines 18-29 but
> there is a problem reading line 22.
>
> It appears that both subroutines try to read U and J (akin to a DFT+U
> input) but only U (or perhaps only J) is present in the input file.
> Upon commenting out  inorbch and vorbch in symso.f  (lines 421, 422)
> and recompiling, the problem goes away and symmetso runs fine during
> the SO initialization but I want to be certain that it this was done
> right.
>
> I was wondering if you could look into the matter and suggest a
> potential patch which could be used to temporarily alleviate this
> problem.
>
> Thanks!
> _______________________________________________
> 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/
--------------------------------------------------------------------------