[QE-users] Wrong electron number in superposition of atomic charge densities

[Paolo Giannozzi]

 (Not-so-)Quick and dirty fix: run the atomic code for the Ti
pseudopotential on the ground-state electronic configuration; replace the
(pseudo-)atomic charge in the file with the one you get for the ground
state. Only the starting charge will be affected.

Paolo

On Thu, Aug 27, 2020 at 5:28 PM Colonna Nicola (PSI) <nicola.colonna at psi.ch>
wrote:

> Dear QE experts,
>
>
>  I recently perfomed a scf PBEsol calculation for Tb2Ti2O7 (attached the
> input file).
>
>  I would like to plot the difference between the scf charge density and
> the superposition
>
>  of atomic charge. For this purpose I used the PostProcessing tool
> (plut_num = 9).
>
>
>  I expected the integral of this difference to be ~0 , but I found it to
> be 4. After a quick check
>
>  I realized that the Ti pseudopotetial (GBRV-1.4,
> ti_pbesol_v1.4.uspp.F.UPF) was generated
>
>  in an ionized configuration with just 1 electron (instead of 2 ) in the
> 3d orbital. Since there
>
>  are 4 Ti in the primitive cell, I think this is the reason for the odd
> result described above.
>
>
>  Indeed during the initialization of a scf calculation for an isolated Ti
> atom I got:
>
>
> Initial potential from superposition of free atoms
>      Check: negative starting charge=   -0.001319
>      starting charge   10.99996, renormalised to   12.00000
>
>  However this renormalization is not done in the PP, and in any case I do
> not think it would make sense there.
>
>
>  That said, my feeling is that I cannot completely trust the charge
> density difference.
>
>  Do you have any advise on how to solve this problem (beside changing
> pseudo)?
>
>
>  Thank you in advance and best regards,
>
>
> Nicola Colonna
>
>
>
> -----------------------------------------------------
> PAUL SCHERRER INSTITUT
> Nicola S. Colonna
> WHGA/150
> CH-5232 Villigen-PSI
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users



-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200827/c2704d88/attachment.html>