[Pw_forum] Dear Dr. Sclauzero,

[Gabriele Sclauzero]

Dear Dr. Chan-Woo,

   you should post such question on the Forum. Also other QE developers or users could give a useful reply and take part to the discussion, or just be interested in it because experiencing a similar problem. Last but not least, me, like many other users and developers, have chosen to give their contribution to the whole community of QE users through the Forum. In my opinion, it's not fair to address these people with private messages asking for help, as if they were "help-desks".

Briefly, I think that your problem stems exactly from that fix which I proposed and is now included in the latest PWscf version.
That fix will actually introduce a problem if the atomic wave function which you want to use for making projections have zero occupations. 
I thought that this would never be the case, but probably one should not exclude it a priori.
A further problem arises if you generate such a pseudopotential (Ce with zero 4f occupation) and want to use it for making the LDA+U projections on the empty states, since the LD1 code usually saves in the PP file only the wavefunctions with nonzero occupations.
First, you should check if you PP has the 4f wavefunctions. If it does, you can remove the check
upf(nt)%oc(n) > 0.D0
as a quick fix. If not, then you need to regenerate the PP.


Regards,

GS

Il giorno 13/dic/2011, alle ore 20.42, Chan-Woo Lee ha scritto:

> Dear Dr. Sclauzero:
>  
> May be I sent email to your old email address.
>  
> Please find forwarded email and thank you in advance!
>  
> Sincerely,
>  
> Chan-Woo
>  
> P.S. One quick question. Did you happen to get the answer about http://www.democritos.it/pipermail/pw_forum/2010-July/017622.html ? As you see from my Ce.upf, in initial configuration, 4f is empty. I did it intentionally as I generated the psp for oxide applications (CeO2, Ce2O3, etc) … Is it safe to make modification from the link?
>  
>  
>  
> From: Chan-Woo Lee [mailto:leechanw at sas.upenn.edu] 
> Sent: Tuesday, December 13, 2011 11:10 AM
> To: 'sclauzer at sissa.it'
> Cc: leechanw at sas.upenn.edu
> Subject: Dear Dr. Sclauzero,
>  
> Dear Dr. Gabriele Sclauzero,
>  
> Hello, my name is Chan-Woo Lee, working at Chemistry department of UPenn as a postoc. I email you as I saw a lot of your posts on PW_Forum about DFT+U. It looks like I have similar issue with Jia Chen, http://www.democritos.it/pipermail/pw_forum/2010-July/017602.html .
>  
> My problem is that in applying U to Ce (in CeO2), I got somewhat weird occupation matrix (both in Cerium oxide and single Ce atom in a box).
>  
> Anyway, if you don’t mind, may I ask your help on my problem? If yes, I will email you with Ce.upf, output file, and input file.
>  
> Thank you in advance and it will be VERY helpful for me to fix my problem if you can provide me some advice on the problem.
>  
>  
>  
> Sincerely,
>  
> Chan-Woo
>  
>  
>  
>  
>  
>  
>  
> -------
> Chan-Woo Lee, Ph.D.
> Postdoctoral Research Associate
>  
> Room 265
> Department of Chemistry
> University of Pennsylvania
> 231 South 34th Street
> Philadelphia, PA 19104-6323 
> Phone: 1-215-898-3564 (Office)
> Email: leechanw at sas.upenn.edu
>  
>  


§ Gabriele Sclauzero, EPFL SB ITP CSEA
   PH H2 462, Station 3, CH-1015 Lausanne







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20111214/eb9f822d/attachment.html>