[Wien] possible bug in orb

[Peter Blaha]

Thanks for the report.

However, I would call this not a bug, but a "feature" (or maybe the bug 
is that it does not complain when s or p electrons are selected).

It does not make sense to apply LDA+U to s or p electrons. They are 
always delocalized (non-correlated) electrons.


Am 10.10.2018 um 19:50 schrieb Kateryna Foyevtsova:
> Dear wien2k developers,
> 
> I think there is a bug in orb which shows up when a Hubbard U is being
> applied to L=1 (p) electrons.
> 
> In Vcalc.f, no value is assigned for F(2) for the L=1 case. This will give
> that, for example, in a case of case.inorb like the one below:
> 
>    1  3  0                     nmod, natorb, ipr
> PRATT  1.0                    BROYD/PRATT, mixing
>    2 1 2                          iatom nlorb, lorb
>    3 1 1
>    4 1 1
>    1                              nsic 0..AFM, 1..SIC, 2..HFM
>     0.882 0.074
>     0.000 0.000
>     0.000 0.000
> 
> the atoms 3 and 4 will contribute non-zero EORB as well as will have some
> non-zero orbital potential matrices, because for these atoms the code is
> going to use F(2) taken over from atom 2.
> 
> Thank you,
> Kateryna
> 
> 
> 

-- 
--------------------------------------------------------------------------
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    WIEN2k: http://www.wien2k.at
WWW: 
http://www.imc.tuwien.ac.at/tc_blaha-------------------------------------------------------------------------