GW QP Bandgap

[Daniele Varsano]

Dear Reza,

Yambo shows in the report the KS gap and the QP gap only when a QP database is read externally, e.g. in BS calculation.
HF gap should be evaluated by hand, looking at the Eks, Vxc and Sx value reported for the valence band maximum and condution band minimum.

Best,
Daniele

[Reza_Reza]

Thank you for the response.
What is exactly the difference between:
[Hartree-Fock] Direct Gap: 0.156309 [eV]
and
[X] Direct Gap: 0.156309 [eV]
Which are printed in the report file.

Best,

[Daniele Varsano]

Dear Reza,
in this case, it seems you calculated HF corrections in points of the BZ or bands not participating in the evaluation of the gap, but this is a guess as I have a look to the entire report file (that you can attach) in order to provide a more detailed answer.
Best,
Daniele

[Reza_Reza]

Thanks,

I attached the report file.

[Daniele Varsano]

Dear Reza,

indeed: the gap is located at k point with index 76:

Code: Select all

[X] Direct Gap localized at k                     :   76

and HF corrections are calculated up to the first 26 kpoint:

Code: Select all

  QP @ state[ 1 ] K range:   1  26

When plugging these calculations in a BSE kernel Yambo automatically makes an extrapolation in the whole BZ, but in this case this is not done (needed) and the HF report is somehow misleading as the gap is not touched. In any case, you have all the corrections listed.

Best,
Daniele

[Reza_Reza]

For optimization of EXXRLvcs
How should I adjust the input file to have a correct gap?

This is the input file for the same system, as I've already sent the report:

HF_and_locXC # [R] Hartree-Fock
EXXRLvcs= 30 Ry # [XX] Exchange RL components
VXCRLvcs= 2060811 RL # [XC] XCpotential RL components
%QPkrange # [GW] QP generalized Kpoint/Band indices
26|26|56|59|
%

In general, what has to be done to have the correct value of the gap in the calculation?

Thank You

[Daniele Varsano]

Dear Reza,

if you want to look at the gap you need to calculate the correction to the bands around the gap in all the BZ.
In your case:

Code: Select all

%QPkrange # [GW] QP generalized Kpoint/Band indices
1|76|56|59|
%

Best,
Daniele

[Reza_Reza]

Hello,

I calculated the GW band structure using Yambo, but the resulting band structure appears unusual (attached). For the Quantum Espresso calculation, I employed the PBE functional with k-points set to 50 for a 1D system.

The ypp input file is as follows:

electrons # [R] Electronic properties
bnds # [R] Bands
PROJECT_mode= "none" # Instruct ypp how to project the DOS. ATOM, LINE, PLANE.
INTERP_mode= "BOLTZ" # Interpolation mode (NN=nearest point, BOLTZ=boltztrap aproach)
INTERP_Shell_Fac= 20.00000 # The bigger it is a higher number of shells is used
INTERP_NofNN= 1 # Number of Nearest sites in the NN method
OutputAlat= 6.87 # [a.u.] Lattice constant used for "alat" ouput format
cooIn= "rlu" # Points coordinates (in) cc/rlu/iku/alat
cooOut= "rlu" # Points coordinates (out) cc/rlu/iku/alat
% BANDS_bands
65 | 68 | # Number of bands
%
CIRCUIT_E_DB_path= "none" # SAVE obtained from the QE run (alternative to %BANDS_kpts)
BANDS_path= "" # High-Symmetry points labels (G,M,K,L...) also using composed positions (0.5xY+0.5xL).
BANDS_steps= 500 # Number of divisions
#BANDS_built_in # Print the bands of the generating points of the circuit using the nearest internal point
%BANDS_kpts # K points of the bands circuit
0.00000 |0.00000 |0.00000 |
0.50000 |0.00000 |0.00000 |
%
GfnQPdb= "E < ./ndb.QP"
XfnQPdb= "E < ./ndb.QP"

I attached the report file for your review.

I am looking forward to hearing from you.

Best,

[Daniele Varsano]

Dear Reza,

indeed it is weird, and I do not know what went wrong in the interpolation.
Anyway, you have a 1D sampling, and you have already the QP band structure in your QP output without need to interpolate.
As a first step, I would look to the band structure, extracting the GW value directly from the output or the report file.
Next, can you also post the report of the interpolation (ypp) step?


Best,
Daniele

[Reza_Reza]

Dear Dr. Varsano,
Many thanks for your reply.

I attached the qp output file and the ypp report file.

Best,