[QE-developers] DFT+U+SOC band structure

[Iurii TIMROV]

Dear Ignacio Martin Alliati,


I think your email should go into the QE-users mailing list and not in the QE-developers mailing list.


Please read carefully the instructions on how to ask questions on the QE forums:

https://www.quantum-espresso.org/forum

In particular, you should provide input and output files.


> However, Run3 and Run4 (GGA+U) show a huge impact of SOC.


Isn't it the effect of the +U correction? Did you check the PDOS with and without +U? Try to apply +U without SOC and see what you obtain.

I am not familiar with DFT+U in the noncollinear case, so maybe other QE developers can help you with this.


> Also, in Run4, with a different number of cores, I get NaN instead of 0.000 when reading the Hubbard data. Do you know what may be causing this?


This seems to be a problem. In this case, you should open an issue on Gitlab and provide all necessary information that the QE developers can reproduce your problem.

https://gitlab.com/QEF/q-e/-/issues


Greetings,

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: developers <developers-bounces at lists.quantum-espresso.org> on behalf of Ignacio Alliati <ialliati01 at qub.ac.uk>
Sent: Tuesday, April 20, 2021 12:27:48 PM
To: developers at lists.quantum-espresso.org
Subject: [QE-developers] DFT+U+SOC band structure

Dear QE developers,

I’m using PWSCF v.6.5 to address the effect of SOC in the band structure of bulk MnPS3. I’m comparing 4 runs, namely

Fig1_GGA.pdf: PBEsol without U
Run1 (Orange curve): Ncl without SOC
Run2 (Blue curve): Ncl with SOC

Fig2_GGA+U.pdf: PBEsol with U=2.7 eV
Run3 (Orange curve): Ncl without SOC
Run4 (Blue curve): Ncl with SOC

Run1 and Run2 (GGA) show a small impact of SOC, as expected. However, Run3 and Run4 (GGA+U) show a huge impact of SOC. Do you think this is physical?

I tend to think it is not, and that Run4 is not being shifted as expected by the Hubbard correction. I believe the reason may have to do with the following.

In Run3:
[Run3]$ grep 'Hub. E' stdoutscf | tail -n 1
Hub. E (dc, noflip, flip, total)    5.0002   5.0996  -0.0000   0.0994
[Run3]$ grep 'Hub. E' stdoutnscf
Hub. E (dc, noflip, flip, total)    5.0002   5.0996  -0.0000   0.0994

But in Run4:
[Run4]$ grep 'Hub. E' stdoutscf | tail -n 1
Hub. E (dc, noflip, flip, total)    5.0003   5.0999  -0.0002   0.0995
[Run4]$ grep 'Hub. E' stdoutnscf
Hub. E (dc, noflip, flip, total)    0.0000   0.0000   0.0000   0.0000

Also, in Run4, with a different number of cores, I get NaN instead of 0.000 when reading the Hubbard data. Do you know what may be causing this?

Thanks in advance,
Ignacio Martin Alliati
-----------------------------
PhD student
Maths and Physics
Queen’s University Belfast.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20210420/40e7d926/attachment.html>