[QE-users] Projected DOS, and sumpdos.x for noncollinear calculation

[Lucas Nicolás Lodeiro Moraga]

Dear all,
I have calculated the DOS and projected DOS for a crystal using ONCV
pseudopotential. First I use a collinear calculation using scalar
relativistic PP, the projection runs and the sumpdos.x works fine.
To investigate the effects of Spin-Orbit Coupling, noncollinear calculation
with Full Relativistic ONCV pseudopotentials only for heavy atoms, and for
light atoms the later scalar relativistic pseudopotential were used.
For both calculations, DOS are fine. But for projected DOS, in the case of
SOC calculation, the projection was only done for heavy atoms. Also the
sumpdos.x does not work with the projected files. The error is:

from sumpdos : error #         1
     wrong fmf in the first line

For the first error, I think it is because only the FR-PP have spinors and
the wavefuntions can only be projected in spinors. For the second, the
number of columns is major in SOC calculation, and the sumpdos.x does not
allow it.

My question is: There is any way to project a SOC calculation  with mixed
SR and FR pseudo potentials? There is a simple way to sum the pdos files in
SOC calculations using sumpdos.x?

Regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20190916/9cd1c933/attachment.html>