[QE-users] Ynt: Zero occupations of Hubbard levels

[Mustafa Özgür]

Hi Iurii,

I used ultra soft pps from quantum espresso website. Scalar relativistic for 'vc-relax' - 'relax' and full relativistic 'scf' - 'nscf'.
Problem is solved by removing 'lforcet' tag from nscf input. However It is not giving  "eband, Ef (eV) = ** **" energy output anymore in nscf output file.

Now I added J value too since U+J supports noncollinear calculation. But vc-relax calculation gives 'forces in the DFT+U+J scheme are not implemented' error. (Here I'm trying to apply self-consistent procedure for Hubbard and J parameters)

How can I overcome this problem to compute U and J parameter with hp.x code if it is possible? (like U and V)
https://drive.google.com/file/d/1fLyD3OvXiM3w1i2dMfJDpd6_BUBRHlqp/view?usp=share_link

Regards,
Mustafa.

Mustafa Ozgur
PhD Student
ESOGU, Turkey
________________________________
Gönderen: Iurii TIMROV <iurii.timrov at epfl.ch>
Gönderildi: 8 Şubat 2023 Çarşamba 20:04
Kime: Mustafa Özgür <MustafaOzgr at hotmail.com>; Iurii TIMROV <iurii.timrov at epfl.ch>; Quantum ESPRESSO users Forum <users at lists.quantum-espresso.org>
Konu: Re: Zero occupations of Hubbard levels


Dear Mustafa,


I do not mange to reproduce your problem.


I have used the latest development version of Quantum ESPRESSO (QE) from Gitlab (downloaded on 8 February 2023). Also, I changed somewhat your input file to make it run faster and also I used norm-conserving fully-relativistic pseudopotentials from the PseudoDojo library (because I do not know which ones you used). Please try to reproduce my results using the latest development version of QE. Here are all the files:

https://drive.google.com/drive/folders/1vWriLf_IZge___Zs9Oh_eQsEECpwy7f7?usp=share_link


If this works, then try once again for your input file using the same development version of QE. Please note that there are some inaccuracies in the syntax in your input file, so please pay attention to this and fix it.


HTH


Iurii


--
Dr. Iurii TIMROV
Senior Research Scientist
Theory and Simulation of Materials (THEOS)
Swiss Federal Institute of Technology Lausanne (EPFL)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334
________________________________
From: users <users-bounces at lists.quantum-espresso.org> on behalf of Iurii TIMROV via users <users at lists.quantum-espresso.org>
Sent: Tuesday, February 7, 2023 12:45:02 PM
To: Mustafa Özgür; Quantum ESPRESSO users Forum
Subject: Re: [QE-users] Zero occupations of Hubbard levels


Indeed there seems to be a problem. We will investigate this issue and will come back to you ASAP.


Iurii


--
Dr. Iurii TIMROV
Senior Research Scientist
Theory and Simulation of Materials (THEOS)
Swiss Federal Institute of Technology Lausanne (EPFL)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334
________________________________
From: Mustafa Özgür <mustafaozgr at hotmail.com>
Sent: Tuesday, February 7, 2023 11:34:51 AM
To: Iurii TIMROV; Quantum ESPRESSO users Forum
Subject: Ynt: Zero occupations of Hubbard levels

Hi Iurii,

Thanks for the fast reply. Here is the Google Drive link.
https://drive.google.com/drive/folders/1JjcbGerQrt7bvPqanKwXT29m7jIGKmId?usp=sharing

Best Regards,
Mustafa.

Mustafa Ozgur
PhD Student
ESOGU, Turkey
________________________________
Gönderen: Iurii TIMROV via users <users at lists.quantum-espresso.org> adına users <users-bounces at lists.quantum-espresso.org>
Gönderildi: 7 Şubat 2023 Salı 13:14
Kime: users at lists.quantum-espresso.org <users at lists.quantum-espresso.org>
Konu: Re: [QE-users] Zero occupations of Hubbard levels


Dear Mustafa,


> Is it possible to calculate DFT+U with noncollinear?


Yes


Currently, in Quantum ESPRESSO this is implemented in the framework of the Lichtenstein formulation.


Please provide all the input and output files (using public shared folders, e.g.Google Drive) that show the problem.


HTH


Iurii


--
Dr. Iurii TIMROV
Senior Research Scientist
Theory and Simulation of Materials (THEOS)
Swiss Federal Institute of Technology Lausanne (EPFL)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334
________________________________
From: users <users-bounces at lists.quantum-espresso.org> on behalf of Mustafa Özgür <MustafaOzgr at hotmail.com>
Sent: Tuesday, February 7, 2023 8:50:06 AM
To: users at lists.quantum-espresso.org
Subject: [QE-users] Zero occupations of Hubbard levels

Hi,

I'm trying DFT+U calculation with spin orbit and noncollinear. However the current working version is giving the "Hubbard manifold with zero occupations is not allowed" error.

Then I downloaded the developed version. It is not giving any error anymore but occupation of hubbard levels are still zero for nscf calculation but not for scf calculation.

     Number of occupied Hubbard levels =    3.6094 for last iteration of scf
     Number of occupied Hubbard levels =    0.0000 for nscf

Is it possible to calculate DFT+U with noncollinear? I mean is this the problem? or something else. Here I attached scf and nscf inputs and outputs.

Best Regards,
Mustafa.

Mustafa Ozgur
PhD Student
ESOGU, Turkey
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20230209/e1c8ba0c/attachment.html>