[Pw_forum] A question about tot_charge flag with DFT+U calculations

[kosuke.nakano at agc.com]

Dear Stefano,

Thank you for your prompt reply.

> the total charge option acts on deciding the number of states used and how to fix the fermi energy/chemical potential.
> the occupation matrix return the projection of the occupied manifold (determined by the previous recipe)
> on the localized states so it should be what one expects it to be.
> so yes it looks like it's Ni ~ +2 (d8)

Thank you. This is what I expected.

> but with partial occupation of the eg orbitals and NOT spin polarized...
> Is this what you expect/desire ?.

The system is a test one.  I'm sorry I didn't make it clear enough …

Best regards.

Kosuke Nakano
Asahi Glass Co., Japan.

From: pw_forum-bounces at pwscf.org [mailto:pw_forum-bounces at pwscf.org] On Behalf Of stefano de gironcoli
Sent: Friday, January 13, 2017 5:04 PM
To: pw_forum at pwscf.org
Subject: Re: [Pw_forum] A question about tot_charge flag with DFT+U calculations

Dear Kosuke Nakano,
   the total charge option acts on deciding the number of states used and how to fix the fermi energy/chemical potential.
   the occupation matrix return the projection of the occupied manifold (determined by the previous recipe) on the localized states so it should be what one expects it to be.
   so yes it looks like it's Ni ~ +2 (d8) but with partial occupation of the eg orbitals and NOT spin polarized... Is this what you expect/desire ?.
Did you suggest a starting magnetization to see it the system does not prefer being d5_up d3_dw ?

stefano

On 13/01/2017 05:26, kosuke.nakano at agc.com<mailto:kosuke.nakano at agc.com> wrote:
Hi, all:

I have a question about tot_charge flag with DFT+U calculations

I would like to calculate d state of transition metal with tot_charge flag.

I know that tot_charge flag with DFT+U calculations is valid as written in
<http://qe-forge.org/pipermail/pw_forum/2015-August/107486.html><http://qe-forge.org/pipermail/pw_forum/2015-August/107486.html>

I am wondering if occupation matrices are also valid in tot_charge
flag with DFT+U calculations.

For example, I obtained the following occupation matrices with a finite tot_charge(+4)
The system has 24 atoms and 4 of them are Ni atoms.

--The final result--
atom   1   Tr[ns(na)] (up, down, total) =   4.26150  4.08319  8.34469 <- Ni (+2, d8?)
   spin  1
    eigenvalues:
  0.658  0.658  0.981  0.982  0.982
    eigenvectors:
  0.000  0.000  0.996  0.003  0.002
  0.588  0.074  0.000  0.291  0.046
  0.074  0.588  0.001  0.046  0.291
  0.038  0.300  0.002  0.089  0.571
  0.300  0.038  0.001  0.571  0.090
    occupations:
  0.981 -0.000 -0.000 -0.000 -0.000
-0.000  0.768  0.000 -0.000  0.153
-0.000  0.000  0.768  0.153 -0.000
-0.000 -0.000  0.153  0.873 -0.000
-0.000  0.153 -0.000 -0.000  0.873
   spin  2
    eigenvalues:
  0.649  0.649  0.923  0.931  0.931
    eigenvectors:
  0.000  0.000  1.000  0.000  0.000
  0.649  0.018  0.000  0.078  0.255
  0.018  0.649  0.000  0.255  0.078
  0.009  0.324  0.000  0.510  0.157
  0.324  0.009  0.000  0.157  0.510
    occupations:
  0.923  0.000  0.000  0.000  0.000
  0.000  0.743  0.000  0.000  0.133
  0.000  0.000  0.743  0.133  0.000
  0.000  0.000  0.133  0.837  0.000
  0.000  0.133  0.000  0.000  0.837
atomic mag. moment =     0.178309

I am wondering if it means Ni(+2, d8) or not.

Best regards

Kosuke Nakano
Asahi Glass Co., Japan.




_______________________________________________

Pw_forum mailing list

Pw_forum at pwscf.org<mailto:Pw_forum at pwscf.org>

http://pwscf.org/mailman/listinfo/pw_forum


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170113/9ab97b93/attachment.html>