[Wien] Continue from my previous mail on Ghostband error

[Laurence Marks]

Several other points:
1) It appears that you did not to follow the standard initiation. Your atom
positions have errors, for instance atom 2 is
ATOM  -2: X=0.77850183 Y=0.75000002 Z=0.15072562  and should be
ATOM  -2: X=0.77850183 Y=0.75000000 Z=0.15072562
This can matter. With this corrected Wien will (properly) switch axes to
put the mirror down z.

2) It looks like you used a non-standard energy in lstart, to avoid core
leaking. These are warnings, and you should not adjust too much
particularly when you have a complex structure with (perhaps) inaccurate
atomic positions. I suggest just using -7, your use of -10 (?) was not
needed.

3) Beware that Vasp can reduce symmetry inappropriately. I would start of
with the published Pnma symmetry and occupy one of the Fe sites and one of
the Sn sites.

4) You will need to look into the magnetism. I don't know if this is
ferromagnetic or anti-ferromagnetic. If the later you will need to break
the symmetry appropriately (carefully).

5) I doubt that you will get meaningful results unless you use either -eece
or +U for the Fe sites.

On Tue, Feb 4, 2020 at 9:25 AM Peter Blaha <pblaha at theochem.tuwien.ac.at>
wrote:

> Does the problem happen in the first scf cycle or later on ???
>
> If it is later, make sure :NEC01 is ok in the scf fle.
>
> Check case.scf1 to see where the actual E-parameters for atom 5 l=2 are.
> If the E-parameters for APW and for the LO are too close ---> ghostbands
>
>
> Are you sure you need the Sn-s semicore states ??
>
>
>
> --
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
Corrosion in 4D: www.numis.northwestern.edu/MURI
Co-Editor, Acta Cryst A
"Research is to see what everybody else has seen, and to think what nobody
else has thought"
Albert Szent-Gyorgi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20200204/83aaf096/attachment.html>