[Pw_forum] Small U for Co in ZnO for LDA+U (update)

[Matteo Cococcioni]

Dear Nektarios,

I was going to write but Nicola has been faster than me. Nicola is right: 
probably the reason why "ortho-atomic" and "norm-atomic" produce similar 
results is that the atomic wavefunctions in the Co PP are not normalized. 
Try to check the occupations in the "atomic" case. are they much bigger 
than 1?

Ortho-atomic: atomic wfcs are orthogonalized with those belonging to 
different atoms. atomic: wfcs are orthogonal (by construction) only to 
those belonging to the same atom (there are overlap regions in between 
atoms) norm-atomic: quick and dirty fix added to use PPs whose wfcs are not 
normalized. when these wfcs are normalized it gives the same results as 
"atomic".

In general I would suggest to not look at the value of U itself but rather 
to the results it produces. I just found a paper: PRL 100, 256401 (2008) 
where they use U=2 for Co and 5eV for Zn (with vasp). Why don't you 
actually try to compute the U (PRB 71, 35105 (2005))?

hope this helps.

Matteo 


On Jun 30 2008, Nicola Marzari wrote:

>
>Dear Nektarios,
>
>
>thanks for the update. Maybe Matteo Cococcioni or other could chip in -
>could it be that the atomic orbital in the Co psp was not normalized,
>so only when ortho-atomic or norm-atomic are applied the normalization
>constant is taken into account ?
>
>If the case, is it correct to say that norm-atomic and atomic are doing
>the same thing, but norm-atomic automatically normalizes the projection
>(something that should have been there in the Co psp to begin with) ?
>
>
>Best,
>
>					Nicola
>
>
>Nektarios Lathiotakis wrote:
>> Hi again,
>> 
>> As I wrote previously I have a Co impurity atom in ZnO and
>> trying to find the right Hubbard-U for Co by fitting to experimental Co-d
>> projected band  from J. Appl. Phys. 103 07D130 (2008). A typical input 
>> file is in my
>> previous e-mail.
>> 
>> I played with U_projection_type parameter and:
>> 
>> 1) 'ortho-atomic' and 'norm-atomic' agree and give the same optimal U 
>> for Co that is around 4eV. Very reasonable value and no referee would 
>> ever object.
>> 
>> 2) 'atomic' (the default) gives as optimal U 0.5eV which is much smaller.
>> 
>> In other words the Co-d projected band you get with 
>> U_projection_type='ortho-atomic'
>> and U_Co = 4 eV is the very close to the band you get with 
>> U_projection_type='atomic'
>> and U_Co = 0.5 eV.
>> 
>> Why is that difference? The problem with changing the U_projection_type 
>> is that only
>> atomic supports forces.
>> 
>> Thanks in advance and also to Nicola for his previous reply.
>> Nektarios
>
>

-- 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Matteo Cococcioni
Department of Chemical Engineering and Materials Science
University of Minnesota
151 Amundson Hall
421 Washington Av. SE
Minneapolis, MN 55455
Tel. +1 612 624 9056    Fax +1 612 626 7246
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%