The ABINIT Users Mailing List ( CLOSED )

[Matthieu Verstraete <mjv500@york.ac.uk>]

Ok, I see what you are looking at better now. It depends on what you want 
to study and quantify. The difference almost certainly comes from the 
pseudopotential approximation, and possibly could be lesser for other 
psps, e.g. PAW (I would definitely give this a try). Otherwise, if you 
need the splitting (which could be quite bad in DFT anyway without LDA+U 
or something), you'll have to stick to all electron. How much slower is 
wien2k?

A final note, it looks like the up channel is shifted down in energy while 
the down is shifted up. This feels like a difference in Fermi level or 
broadening btw wien and abinit (especially as the down has 0 DOS at Ef, so 
it's chemical potential could be somewhat arbitrary, although the lowest 
states align nicely), or a difference in the way the spin-dep-GGA is 
implemented. Possible?

Matthieu

--
================================================================
Dr. Matthieu Verstraete                 mailto:mjv500@york.ac.uk
Dept. of Physics, University of York,     tel: +44 1904 43 22 08
Heslington, YO10 5DD York, United Kingdom fax: +44 1904 43 22 14