converging nband?

GW, Bethe-Salpeter …

Moderators: maryam.azizi, bruneval

Locked
hammouri
Posts: 57
Joined: Fri Feb 13, 2015 6:44 am

converging nband?

Post by hammouri » Sun Mar 06, 2016 12:53 am

Hi all,
I need to calculate the optical absorption of SrTiO3. I found when I use "bs_loband=1" and "nband=20", the results looks good compared with experiment. But when I increase # of band (e.g. nband=30)which the curve should get converge I noticed results go bad compared with experiment. My question is what parameters should I also change in order to get good results as when "nband=20"?? FYI see the attached inputs.

Thank you all!
Attachments
tbs_1.in
(2.01 KiB) Downloaded 346 times
tbs_2.in
(1.78 KiB) Downloaded 346 times
Top
hammouri
Posts: 57
Joined: Fri Feb 13, 2015 6:44 am

Re: converging nband?

Post by hammouri » Mon Mar 07, 2016 5:36 pm

Any help?
Top
Locked

Return to “Many-body perturbation theory”