stretching of the bands

Concerns issues with computing quasiparticle corrections to the DFT eigenvalues - i.e., the self-energy within the GW approximation (-g n), or considering the Hartree-Fock exchange only (-x)

Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano

Post Reply
8813204602
Posts: 58
Joined: Tue Jun 15, 2021 5:53 pm

stretching of the bands

Post by 8813204602 » Sat Mar 04, 2023 6:06 am

Dear users and developers,

I have a number of questions about the stretching of the conduction (Mc) and valence (Mv) bands obtained from the linear fit of the quasiparticle energies in terms of Kohn-Sham energies:
1) Does Mc & Mv being closer to one indicate that more accurate calculations have been done? For example, as can be seen in the attached figures, when I consider only the highest 5 valence bands in GW calculations, Mv becoms 0.5, while for 6 valence bands, Mv is approximately one.
Dose this mean that the second calculation is correct?
2) We know that in GW calculations, the amount of energy correction of the conduction band is higher than that of the valence band. Does this also apply to the stretching of the bands, that is, the stretching of the conduction band should be greater than the stretching of the valence band? Because in my calculations Mv=1.028 is obtained more than Mc=0.908.

Thanks and Regards,
Mitra Helmi
You do not have the required permissions to view the files attached to this post.
Mitra Helmi,Ph.D student
Ferdowsi university of Mashhad
Mashhad,Iran

User avatar
Daniele Varsano
Posts: 3773
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: stretching of the bands

Post by Daniele Varsano » Sat Mar 04, 2023 11:45 am

Dear Mitra,

this is really system dependent and it is not related with the accuracy of the calculation. Scissor/stretching it is an approximation that in many cases it applies, but not in all cases. If your bands have different characters, you will have different corrections that cannot be caught by a simple scissor/stretching behaviour.

Looking at your plot it seems you can apply it only if you are interested in a small energy range E > -0.2 eV for the valence and E < 0.5 eV for the conduction.
It seems in your case it is not safe to apply it and you need to resort to the QP energies calculated at GW level reading the databases:
KfnQPdb= "E < ./PATH/ndb.QP"
PATH is the directory where the QP database is present, SAVE by default.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

Post Reply