[QE-developers] QE 7.1 Hubbard card - Mismatch between the requested and available manifolds

[Iurii TIMROV]

Dear Will,


You did everything correctly.


Since QE v7.1 there will be a new input syntax for DFT+U and other extensions (e.g. DFT+U+V). More info can be found in the development version on Gitlab at q-e/Doc/Hubbard_input.tex.


The problem is that the pseudopotentials from this library

http://www.mineralscloud.com/resources/repaw/Gd/index.shtml

do not have labels for the atomic orbitals (e.g. label="4F", etc.). For this reason the new Hubbard input syntax does not manage to recognize the character of each orbital. I have already contacted Renata Wentzcovitch asking to add atomic orbitals labels to the pseudos. Meanwhile, you can add this info by hand in the pseudos based on the available info about the orbital quantum number "l" and the occupations.


For example, for Gd.GGA-PBE-paw-v1.0.UPF you have this info in the pseudo:

<PP_PSWFC>
 <PP_CHI.1 type="real" size="  1110" l="0" occupation=" 2.0000" columns="3">
You can modify it by hand:
<PP_CHI.1 type="real" size="  1110" l="0" occupation=" 2.0000" columns="3" label="6S">
And you need to do it for each orbital.

I discussed this issue with Paolo Giannozzi. He said that in principle we can try to implement an automatic algorithm that tries to guess what is the label of each atomic orbital based on "l" and occupation. But it is unlikely to work in all cases (because sometimes the guess might be wrong). That is why I asked Renata to add this info to the pseudos in order to avoid possible mistakes. Note that pseudos from other libraries have this info (this is why you do not have problems for N.pbe-kjpaw.UPF).

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 William Hewett <williamholmeshewett1 at gmail.com>
Sent: Wednesday, April 13, 2022 3:32:23 AM
To: developers at lists.quantum-espresso.org
Subject: [QE-developers] QE 7.1 Hubbard card - Mismatch between the requested and available manifolds

Hi QE Developers,

Apologises if this is not the appropriate channel for this message. I am running the developer version 7-1 and am getting the following error regarding the new Hubbard input:


Gd pseudopotential contains the orbitals:

     Requested Hubbard manifold from the input: 4F


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

     Error in routine offset_atom_wfc (1):

     Mismatch between the requested and available manifolds

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


This is found when applying U to a 4f orbital in the Gd pseudopotential, this calculation worked fine in the previous version of QE I was running (6-7). Note that applying U to the N2p orbitals works fine in 7-1.

7-1 input:

...


ATOMIC_SPECIES



 N  14.0067 N.pbe-kjpaw.UPF

 Gd 157.250 Gd.GGA-PBE-paw-v1.0.UPF


ATOMIC_POSITIONS {crystal}


Gd 0.500000000000000 0.500000000000000 0.500000000000000

N  0.000000000000000 0.000000000000000 0.000000000000000


K_POINTS {automatic}

 10 10 10 0 0 0


HUBBARD {atomic}

U Gd-4f 5.0


6-7 input:
...


&SYSTEM

 lda_plus_u=.true.

 lda_plus_u_kind=0

 Hubbard_U(1) = 5

...


ATOMIC_SPECIES



 Gd 157.250 Gd.GGA-PBE-paw-v1.0.UPF

 N  14.0067 N.pbe-kjpaw.UPF

...

I notice that the code prints:

'Gd pseudopotential contains the orbitals:       '

i.e. there are no orbitals in the Gd PP? I'll note that in QE 6-7 the parameter hubbard_l = 3 is set in set_hubbard_l.f90 I think should be equivalent to U Gd-4f 5.0 in the new Hubbard input card in 7-1.

The PP I'm using for Gd can be found here:

http://www.mineralscloud.com/resources/repaw/Gd/index.shtml

Perhaps this is just a problem of misunderstanding on my end? Any help is appreciated.

Kind regards,

Will Hewett
Postdoctoral researcher
Victoria University of Wellington
New Zealand
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20220413/a6a29c68/attachment-0001.html>