The ABINIT Users Mailing List ( CLOSED )

[Vincenzo Fiorentini <vincenzo.fiorentini@dsf.unica.it>]

Hi Nicola and all:

I have no real experience, but I found that calculating a GGA spin- 
polarized O atom with partially filled shell in a supercell *with  
symmetry removed* (as Andy suggests), does give a spin-polarized  
ground state with integer occupation for some p orbitals and zero for  
others. Of course it's no conceptual proof (and may also depend on  
which smering is used and the starting wf/density), but it may be a  
practical way to mine for the correct occupations.

About the other point,

> letting the electrons spread equally in a partially-filled  
subshell seems unstisfactory
> to calculate ionization energies of a transition metal, while you  
are emptying the d
> shell (e.g. you would have have first 5*1.0, then 5*0.8, then  
5*0.6, etc...).

I suspect ionization energies may be flawed anyway at non-integer  
occupations mostly because of selfinteraction.

best, Vincenzo

---
Anche il progresso, diventato vecchio e saggio, voto' contro. (Ennio  
Flaiano)
---
vincenzo.fiorentini@dsf.unica.it
mobile +39 347 1410906
phone  +39 070 6754912
fax    +39 070 510171
fax    +39 1782248846
http://www.slacs.it
http://www.dsf.unica.it/~fiore
http://vincenzofiorentini.blogspot.com