[Pw_forum] xspectra - spin polarization and core-hole delocalization

[maxradin at umich.edu]

Hi, 

I have two questions regarding XAS calculations. 

1. For a system whose ground state is non-magnetic, should the core-hole
SCF calculation be spin-polarized? My understanding is that the excited
state must be non-magnetic to conserve spin. Thus I would expect that the
non-spin-polarized calculation is more physical, even though in my case it
is higher in energy than the spin-polarized calculation.  

2. From what I've read in the literature, in some systems core-holes are
delocalized over two neighboring symmetry equivalent sites. Has anyone
tried modeling this situation with xspectra? I could imagine that one might
use a half core-hole PP at both sites instead of a full core-hole at one
site. 

Thanks in advance. 

Max Radin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20111007/93b6c445/attachment.htm