[Pw_forum] lowdin charge

Gabriele Sclauzero sclauzer at sissa.it
Wed Nov 18 08:54:25 CET 2009



yaldaa kh wrote:
>  dear all
> I performed pdos calculation(scf ,nscf pdos) for Cu slab ,but in lowdin 
> charge, the Cu-p had some charge while p orbitals in Cu atom 
> are unoccupied. if this result is wrong?

Not necessarily. It depends on how much is "some charge". You are right when you say that 
in the Cu atom p orbitals are unoccupied (strictly), but in a solid the situation is 
different. Also empty orbitals may be involved in the binding, for instance when some 
charge transfer takes place (remember that in Cu the 3d shell is full and there is place 
for only 1e- in the 4s).
Because of this, pseudopotentials usually include among the projectors (or in the local 
channel) electronic states which can be empty in the atom, but not too much higher in 
energy wrt the occupied ones (e.g., in Cu, which is [Ar] 3d10 4s1, also 4p are included 
since they are only 3-4 eV higher than 4s).
So, if you have the 4p channel in your PP, you will get also the 4p projection file from 
projwfc.x, but these states should be restricted to a high energy region. Have you checked 
how much 4p charge (if any) is found BELOW the Fermi energy?


GS

> thanks a lot
> yaldaa Kh.
> chem. student
> Azad university Iran
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-- 


o ------------------------------------------------ o
| Gabriele Sclauzero, PhD Student                  |
| c/o:   SISSA & CNR-INFM Democritos,              |
|        via Beirut 2-4, 34014 Trieste (Italy)     |
| email: sclauzer at sissa.it                         |
| phone: +39 040 3787 511                          |
| skype: gurlonotturno                             |
o ------------------------------------------------ o



More information about the users mailing list