[Wien] Question about "Bond-Valence Sum" in case.outputnn file.

[Chang-Jong Kang]

Dear Prof. Blaha and Wien2k users,

When I saw case.outputnn file, there were two Bond-Valence Sum (BVS)
printed like,

Atom   5 equiv  1 ATOM1        Bond-Valence Sum     2.62    2.77
Atom   5 equiv  2 ATOM1        Bond-Valence Sum     2.62    2.77
Atom   6 equiv  1 ATOM2        Bond-Valence Sum     2.48    2.62
Atom   6 equiv  2 ATOM2        Bond-Valence Sum     2.48    2.62
Atom   6 equiv  3 ATOM2        Bond-Valence Sum     2.48    2.62
Atom   6 equiv  4 ATOM2        Bond-Valence Sum     2.48    2.62
Atom   7 equiv  1 ATOM3        Bond-Valence Sum     2.16    2.28
Atom   7 equiv  2 ATOM3        Bond-Valence Sum     2.16    2.28
Atom   8 equiv  1 ATOM4        Bond-Valence Sum     2.30    2.43
Atom   8 equiv  2 ATOM4        Bond-Valence Sum     2.30    2.43
Atom   8 equiv  3 ATOM4        Bond-Valence Sum     2.30    2.43
Atom   8 equiv  4 ATOM4        Bond-Valence Sum     2.30    2.43

In nn.f source code, first and second value of BVS are named as "bva" and
"bvaerr", respectively.

My question is
1. Which one is bond-valence sum?"
2. What is "bvaerr" parameter used in nn.f source file? Is it error of
bond-valence sum?
    Only difference between "bva" and "bvaerr" parameters is that
    "bvaerr" use "DFTERR" parameter but "bva" does not use it.
    However, when PBE funtional is used, "DFTERR" parameter is set to 1, so
that
     "bva" and "bvaerr" values are exactly same in my thought since I used
PBE functional.

I am looking forward to your kind explanation.

Thanks.


Best wishes,
C.-J. Kang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20130416/19c11eb0/attachment.htm>