[QE-users] (1) QE-7.1 supports SR NCPP, not FR NCPP. (2) phonon does not support "ortho-atomic" HUBBARD options

[Iurii TIMROV]

Please check how to make posts on the forum: https://www.quantum-espresso.org/users-forum/

(i.e. add your full name, affiliation, technical details, etc.)


> Specifically, the vc-relax with HUBBARD U does not support fully relativistic  NCPP.


Right


> There might be some bug here for fully relativistic pseudopotential.

In order to check this we would need your input and output files to reproduce the error. But as written above, the vc-relax and FR PPs in DFT+U are not compatible at present.


> So, would it be possible to calculate phonon with "the "ortho-atomic" Hubbard option soon ?


This requires a significant amount of work. Our group does not plan to work on this in the near term. You can perform this type of calculations using a finite-differences approach e.g. using Phonopy.


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 None <west_lifer at 163.com>
Sent: Sunday, September 4, 2022 11:30:24 AM
To: users at lists.quantum-espresso.org
Subject: [QE-users] (1) QE-7.1 supports SR NCPP, not FR NCPP. (2) phonon does not support "ortho-atomic" HUBBARD options

(1)  The same question was reported here "https://www.mail-archive.com/users@lists.quantum-espresso.org/msg42402.html "

Specifically, the vc-relax with HUBBARD U does not support fully relativistic  NCPP.

     %%%%%%%%%%%%%%
     Error in routine offset_atom_wfc (1):
     Mismatch between the requested and available manifolds
     %%%%%%%%%%%%%%

    The mainly input parameters are:

    calculation   = 'vc-relax'

    HUBBARD {atomic}
    U Ru-4d 4.0

    The fully relativistic NCPP is Ru_ONCV_PBE_FR-1.0.upf.

%%%%%%%%%%%%%%%%%%%%%

    However, when I test the SR pseudopotential, the vc-relax with atomic-type Hubbard options works successfully.

There might be some bug here for fully relativistic pseudopotential.


(2) I also tested the GRBV fully relativistic pseudopotential to calculate phonon spectrum: ru_pbesol_v1.2.uspp.F.UPF, it works successfully.

    Specifically, I find that only the "atomic" Hubbard option works when calculating phonon spectrum, that is "HUBBARD {atomic}", just like the LiCoO2 case in ./QE/PH/example18 dir with "atomic" Huubard option.

    Sadly, the "ortho-atomic" Hubbard option does not work when calculating phonon, and the code reports an error, "The phonon code for this Hubbard projectors type is not implemented".

    Acturally, I used HP code to calculate the self-consistent U. The "atomic" type calculation would result in much larger U value than "ortho-atomic", the difference is about 1 eV. Here, http://theossrv1.epfl.ch/Main/DFTHubbard,<http://theossrv1.epfl.ch/Main/DFTHubbard%22,> tell us that "the values of U depend strongly on the technical details of calculations, and Ortho-atomic orbitals provide more accurate results - maybe because if atomic (i.e. non-orthogonal) orbitals are used, the Hubbard correction can be added twice in the regions where orbitals overlap."

    So, would it be possible to calculate phonon with "the "ortho-atomic" Hubbard option soon ?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20220905/2eb1c89b/attachment.html>