Very high QP eigenvalue for lowest orbital

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
subhayan
Posts: 32
Joined: Thu Sep 26, 2019 7:48 am

Very high QP eigenvalue for lowest orbital

Post by subhayan » Mon Dec 06, 2021 10:49 pm

Dear developer,

I am trying to run a BSE calculation on a slightly distorted (actually from a molecular-dynamics snapshot) methanol molecule. However, I am getting negative excitation energies for the first two excitations. I guess this has to do with the fact that, in the output of the GW calculation, I am getting E-Eo= 21.49 (Eo=-20.33 , E=1.16) for the first quasiparticle orbital. Could you kindly take a look at the attached files and let me know your thoughts?

For the GW calculation: the input is GW_input.in and the output is r_em1d_ppa_HF_and_locXC_gw0_rim_cut
For the BSE calculation: the input is 02_BSE_construct_solve.in and the output is r-3D_BSE_optics_bse_bss_dipoles_em1s_rim_cut

Kindly let me know if you need any other files.

Best regards,
Subhayan
MethanolYambo.tar.bz2
You do not have the required permissions to view the files attached to this post.
Subhayan Roychoudhury
Postdoctoral researcher
Lawrence Berkeley National Laboratory

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

Re: Very high QP eigenvalue for lowest orbital

Post by Daniele Varsano » Tue Dec 07, 2021 8:12 am

Dear Subhayan,
yes, the problem is that the QP energy of the first deep state is not calculated properly. The problem is the Z factor:

Code: Select all

B=1 Eo=-20.33 E=  1.16 E-Eo= 21.49 Re(Z)=**** Im(Z)= -2.05918 nlXC=-31.45032 lXC=-19.44300 So=  7.05307
It would be possible to investigate it, e.g. some instability in the self-energy derivative for this deep state, but it is not straightforward, you could see for instance the effect
of reducing the dScStep variable.

Anyway, I think that it is not relevant to include such a deep state in the BSE and I would remove the first states from the BSEBands.

Please also check your CUTRadius that seems a bit small.

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/

subhayan
Posts: 32
Joined: Thu Sep 26, 2019 7:48 am

Re: Very high QP eigenvalue for lowest orbital

Post by subhayan » Tue Dec 07, 2021 10:19 am

Perfect. Thanks a lot :-)
Subhayan Roychoudhury
Postdoctoral researcher
Lawrence Berkeley National Laboratory

Post Reply