Queries about input and output files, running specific calculations, etc.
Moderators: Global Moderator, Moderator
-
yuanchang_li
- Newbie
- Posts: 30
- Joined: Tue Mar 17, 2020 2:01 am
#1
Post
by yuanchang_li » Wed Jan 12, 2022 1:21 pm
Dear all,
I have a question that when I do a BSE calculation including the SOC effect, if I want to get the spin-triplet exciton, should I uncomment the "LTRIPLET = T" tag?
I check my results and it seems that when I consider the soc in a non-polarized system, the lowest exciton is spin-triplet directly.
But I'm not so sure
Thank you so much !
Best,
Yuanchang Li
-
alexey.tal
- Global Moderator
- Posts: 374
- Joined: Mon Sep 13, 2021 12:45 pm
#2
Post
by alexey.tal » Thu Jan 13, 2022 3:52 pm
Dear Yuanchang Li,
If you include spin-orbit coupling in your BSE calculation, you should not set LTRIPLET = .TRUE. as, in principle, there is no more distinction between spin channels and the singlet and triplet configurations are mixed. You get the full set of solutions by running the calculation with LHARTREE = .TRUE. and LTRIPLET = .FALSE..
I check my results and it seems that when I consider the soc in a non-polarized system, the lowest exciton is spin-triplet directly.
The lowest exciton should be a triplet state as the repulsive exchange interaction in singlet configurations reduces the binding energy of excitons compared to the triplet states.