Gw+bse optical properties

Concerns any physical issues arising during the setup step (-i option). This includes problems with symmetries, k/q-point sets, and so on. For technical problems (running in parallel, etc), refer to the Technical forum.

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

Post Reply
mjakhar7665
Posts: 7
Joined: Tue Jan 19, 2021 11:12 am

Gw+bse optical properties

Post by mjakhar7665 » Mon Feb 08, 2021 10:19 am

Hello Sir,
I calculate the gw band gap and then i want to calculate the optical spectra with BSE+Gw but i get an error.
I search about it many times but i din't get any solution to that.
Can you check my parameters for gw and bse calculation is right ? and the result is good.

Can you help me in this regard
Mukesh
CEntral University of Punjab
India
Here is my attached file for gw and Bse
Mukesh Jakhar
Central University of Punjab
Bathinda India

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

Re: Gw+bse optical properties

Post by Daniele Varsano » Mon Feb 08, 2021 12:31 pm

Dear Mukesh,
note you can insert your signature in your user profile in order to avoid to repeat it in each post.
Coming to your questions, there are 2 issues:
1)

Code: Select all

% BSEBands
   1 | 200 |                   # [BSK] Bands range
%
This is an extremely high number of bands for BSE, this will end up in enormous matrix nc*nv*nk dimension, that probably will not fit the memory of your machine. Usually BSE converges considering few valence and few conduction bands across the gap. You can start with few bands and increase them until convergence.

2)
Looking at your log files:

Code: Select all

P1: [ERROR] STOP signal received while in :[06] Bethe Salpeter Equation @q1
P1: [ERROR]USER parallel structure does not fit the current run parameters. 10 CPU for 6 elements (ROLE is 'eh'
I suggest you to use a parallelisation scheme assigning CPU to k points as much as possibile, something like:

Code: Select all

BS_CPU= "20 1 1"             # [PARALLEL] CPUs for each role
BS_ROLEs= "k eh t"           # [PARALLEL] CPUs roles (k,eh,t)
Finally note that if you have already calculated the screening for the GW, yambo can read it so it is unnecessary to recalculate it in the BSE, you can remove form your input the em1s keyword and all the Xs variables.

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