identifying transitions from BSE/TDHF output

Queries about input and output files, running specific calculations, etc.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
awang95
Newbie
Newbie
Posts: 41
Joined: Thu May 26, 2022 3:50 pm

identifying transitions from BSE/TDHF output

#1 Post by awang95 » Tue Jun 27, 2023 5:00 pm

Hi,

I have the optical transitions from the vasprun.xml file for a BSE/TDHF calculation. However, in addition to the energies of each transition I'm interested in knowing the bands/orbitals involved in that specific transition (i.e. from which band to which band does the transition correspond to). I can't seem to find the corresponding information in the output files...could you please help me with this? Alternatively, if there are suggestions for any post-processing software that can do this that would be awesome as well

thanks!

alexey.tal
Global Moderator
Global Moderator
Posts: 314
Joined: Mon Sep 13, 2021 12:45 pm

Re: identifying transitions from BSE/TDHF output

#2 Post by alexey.tal » Wed Jun 28, 2023 1:58 pm

Hi awang95,

You can find this information, i.e. the contribution of the individual transitions into the excitonic state, in the BSEFATBAND file.

awang95
Newbie
Newbie
Posts: 41
Joined: Thu May 26, 2022 3:50 pm

Re: identifying transitions from BSE/TDHF output

#3 Post by awang95 » Wed Jun 28, 2023 2:58 pm

thanks!

should the highlighted part of this picture from the wiki be NB_e?
You do not have the required permissions to view the files attached to this post.

awang95
Newbie
Newbie
Posts: 41
Joined: Thu May 26, 2022 3:50 pm

Re: identifying transitions from BSE/TDHF output

#4 Post by awang95 » Wed Jun 28, 2023 3:27 pm

sorry I would like to clarify another thing

from the BSEFATBAND output it seems that the values correspond to the below as noted (it only makes sense if I interpret it this way), which is different from the order posted on the wiki - can I confirm whether if the below would be correct? thanks

alexey.tal
Global Moderator
Global Moderator
Posts: 314
Joined: Mon Sep 13, 2021 12:45 pm

Re: identifying transitions from BSE/TDHF output

#5 Post by alexey.tal » Wed Jun 28, 2023 4:08 pm

It's a typo, thanks for pointing that out. Actually, these are the valence and conduction band energies and numbers, so I changed the notation on the wiki accordingly. Does that answer your question?

awang95
Newbie
Newbie
Posts: 41
Joined: Thu May 26, 2022 3:50 pm

Re: identifying transitions from BSE/TDHF output

#6 Post by awang95 » Wed Jun 28, 2023 4:15 pm

yes thanks!

awang95
Newbie
Newbie
Posts: 41
Joined: Thu May 26, 2022 3:50 pm

Re: identifying transitions from BSE/TDHF output

#7 Post by awang95 » Thu Oct 26, 2023 12:30 am

sorry just like to follow up on this as I'm now interested in the eigenvectors: I'm not quite sure what it means when the wiki says "X_BSE" corresponds to the "component of the eigenvector" - is it the component in the x-direction only or something else? I think I'm not really sure which eigenvector and what component of it the "X_BSE" is referring to here. Also, what physical information can we extract from Abs(X_BSE) and Re(X_BSE)+i*(X_BSE)? sorry if this is too obvious or stupid - I'm still quite new to this and have been quite confused for a while

thanks!

alexey.tal
Global Moderator
Global Moderator
Posts: 314
Joined: Mon Sep 13, 2021 12:45 pm

Re: identifying transitions from BSE/TDHF output

#8 Post by alexey.tal » Thu Oct 26, 2023 9:11 am

The BSEFATBAND file stores the eigenvectors of the BSE Hamiltonian found when we diagonalize the Hamiltonian. X_BSE are the components of the these eigenvectors which are essentially the coupling coefficients or contributions of individual transitions between valence and conduction bands. Using this information you can make a fatband plot as described on the wiki Plot BSE fatbands. Such a plot can give you an idea of what transitions contribute the most to an excitonic state and how localized this state is.

awang95
Newbie
Newbie
Posts: 41
Joined: Thu May 26, 2022 3:50 pm

Re: identifying transitions from BSE/TDHF output

#9 Post by awang95 » Thu Oct 26, 2023 3:42 pm

ok thanks so much!

Post Reply