[QE-users] Hubbard+U+V and Error in routine offset_atom_wfc

[Christoph Wolf]

Dear all,

I am running into some problem when using the same species twice in a
calculation (eg. O1, O2 for two inequivalent oxygen atoms) and then
applying "standard U" on both species, e.g.

ATOMIC_POSITIONS (angstrom)
V             0.0 0.0 0.0
O1            0.0 0.0 2.0
O2            0.0 0.0 -2.0


lda_plus_u = .true.
lda_plus_u_kind = 2
U_projection_type = 'ortho-atomic',
Hubbard_V(1,1,1)=1
Hubbard_V(2,2,1)=2

will work, but adding

Hubbard_V(3,3,1)=3

will throw the "offset error";

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine offset_atom_wfc (2):
     wrong offset: your pseudopotential file for atomic species  3 likely
does not contain the needed atomic wavefunctions
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I guess that is because the offset is already set from the previous call to
offset_atom which is checked at the end of that routine:

      IF ( ( is_hubbard(nt) .OR. is_hubbard_back(nt) ) .AND. offset(na) < 0
) THEN
        CALL errore('offset_atom_wfc', 'wrong offset: your pseudopotential &
                &file for atomic species ' // s // new_line('a') // 'likely
&
                &does not contain the needed atomic wavefunctions', l_back)
      ENDIF

Is there a way to safely detect if the wave-functions are present in the PP
without running into this problem?

Best,
Chris
--
IBS Center for Quantum Nanoscience
Seoul, South Korea
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20220525/22d96aa4/attachment.html>