Page 1 of 3

GW vs GGA band gap

Posted: Sat Jul 11, 2020 9:45 pm
by lamia
Dear Yambo Developers,

I am running GW calculations and obtaining the band structure. However, when I look at the band structure, I see that the band gap (1 eV) of GW is equal to the band gap (1 eV) of DFT-GGA . How is this possible?

Best,

Re: GW vs GGA band gap

Posted: Sun Jul 12, 2020 4:32 pm
by Palummo Maurizia
Dear Taouil

without further detail it is impossible to help you.
Please provide at least the input and report files of your GW and ypp post-processing runs.

Best
Maurizia

Re: GW vs GGA band gap

Posted: Mon Jul 13, 2020 4:33 pm
by lamia
Dear Maurizia,
Thank you for your reply.
You'll find the input of GW and ypp post-processing runs.

Best,

Re: GW vs GGA band gap

Posted: Tue Jul 14, 2020 8:04 am
by Daniele Varsano
Dear Taouil,
in order to include GW correction in the band structure, you need to provide the QP database in the input file:
GfnQPdb = " E < ./SAVE/ndb.QP"
if the QP database is in the SAVE directory, otherwise you need to indicate the path where the database is.

Best,
Daniele

Re: GW vs GGA band gap

Posted: Wed Jul 15, 2020 7:43 pm
by lamia
Dear Daniele,

Should I add GfnQPdb = " E < ./SAVE/ndb.QP" in yambo.in input file or just in ypp.in or both ? I've tried to add it in ypp.in but I still get same band structure.

best,

Lamia

Re: GW vs GGA band gap

Posted: Thu Jul 16, 2020 6:57 am
by Daniele Varsano
Dear Lamia,
the QP correction in G has to be added in the ypp.in. If the bands are the same it means that the QP corrections are not read correctly. To understand what went wrong can you post the report and output?

Best,
Daniele

Re: GW vs GGA band gap

Posted: Thu Jul 16, 2020 11:53 am
by lamia
Dear Daniele,

You'll find the report and output files in the attachments.

Best,

Lamia

Re: GW vs GGA band gap

Posted: Thu Jul 16, 2020 12:11 pm
by Daniele Varsano
Dear Taouil Lamia,
from the report it seems the QP corrections are read correctly.
Can you try to repeat the calculation by commenting the BANDS_path variable, and using
GfnQPdb= "E < ./SAVE/ndb.QP"
instead of:
GfnQPdb= "E < SAVE/ndb.QP"
if the correction is not applied please post also the output file.

Best,

Daniele

Re: GW vs GGA band gap

Posted: Thu Jul 16, 2020 12:27 pm
by lamia
Dear Daniele,

I've tried with your last suggestion .

PS: I should get an indirect band gap with 2.43 eV
best,
lamia

Re: GW vs GGA band gap

Posted: Thu Jul 16, 2020 1:36 pm
by Daniele Varsano
Dear Taouil Lamia,
I was meaning the bands outputs you calculated with and w/o the GfnQPdb input variable.

Best,

Daniele