analysis of excitons

Anything regarding the post-processing utility (e.g. excitonic wavefunction analysis) is dealt with in this forum.

Moderators: Davide Sangalli, andrea marini, Daniele Varsano

Post Reply
8813204602
Posts: 58
Joined: Tue Jun 15, 2021 5:53 pm

analysis of excitons

Post by 8813204602 » Tue Mar 14, 2023 8:20 am

Dear developers,

I have a number of questions about the analysis of excitons with ypp:
1) When I calculate the excitonic weights using the input file obtained with ypp -e a command, the output file obtained as seen in the attachment is different from what is mentioned in the tutorial on the site. Is this normal or is there a mistake in the calculations? The slepc solver in Yambo-5.0.3 was used for calculations.
2) In my calculation, the first absorption peak corresponding to the first bright exciton consists of the following two excitons:

Code: Select all

      E [ev]                   Strength                 Index
2.02250910        0.946212888        17.0000000   
2.04752350         1.00000000         18.0000000
in this case, in order to obtain the excitonic spatial distribution related to this peak, How should the "States" parameter be set in the input file?
3) Is it correct to calculate the average of the amplitudes of these two excitons to calculate the amplitude of this peak?
4) Considering that the material under investigation is a 2D monolayer, is it enough to use the cartesian coordinates of the atom in o-BSE60.exc_qpt1_E_sorted file for "Hole" parameter in input file?

Code: Select all

Atom 1 with Z 28 [cc]: 0.210328158E-1  0.359024890E-1   5.66934252   
Atom 1 with Z 17 [cc]: 0.212262236E-1   3.82510161      3.17067671   
Atom 2 with Z 17 [cc]:  3.30249858      1.93065667      8.16750908   
or should the vertical position of the hole be shifted from 5.66 to 6.66 like 3D-hBN tutorial.

Best Regards,
Mitra
You do not have the required permissions to view the files attached to this post.
Mitra Helmi,Ph.D student
Ferdowsi university of Mashhad
Mashhad,Iran

User avatar
palful
Posts: 56
Joined: Tue Jan 26, 2016 11:23 am
Location: Modena and Milan

Re: analysis of excitons

Post by palful » Tue Mar 14, 2023 11:32 am

Dear Mitra,

1) Indeed, there is some problem with the output you posted.
At line 462 you have:

Code: Select all

#    Band_V             Band_C             Kv-q ibz           Symm_kv            Kc ibz             Symm_kc            Weight             Energy
#    
# 
however there are no actual numbers here, as there should be. (Here you have some more columns with respect to the tutorial since the possibility of finite-q excitons is taken into account in this yambo version)
This points to a problem in the ypp run, typically forgetting to include the BS, dipoles and em1s databases with "-J", but could be something else.

2) In this case, if you are sure the two excitons are degenerate (and not separate non-degenerate bright states - you can check this by plotting them separately or looking at the system symmetries), you must set the DegenStep parameter to a value larger than their energy difference. Then, you can either put States = "17 - 17" or States = "17 - 18", in any case the code will account for both states.

3) The strength of a degenerate peak is simply the sum of its components, as in the absorption spectrum each state appears with its own strength and if two states are at the same energy the resulting peak is summed.

4) Here the "hole" must be put in the most "physical" position in order to get a meaningful resulting electron distribution. For example, in hBN (bulk or 2D), the hole forming the exciton comes fro valence states which have the shape of a p_z orbital and are localised on the nitrogen atom. So in that case, the hole position must be put a little bit above or below the N position so as to "hit" the lobe of the p_z orbital, representing the maximum likelihood where the hole can be found. In your system, this position can be different, e.g. it could be above or below a bond. In order to find this out, you should first identify (for example using ypp) the valence state \phi_vk which matters most in the exciton you are analysing. Then, you can project that state onto atomic orbitals (normally a postprocessing tool of many DFT codes), and see where \phi_vk(r) is mostly localised. Then, r_hole will be that position. Note that ypp will not place the hole exactly at the position you specify in the input because it uses a discrete space grid: the accuracy of the placement will improve the more you increase FFTGvecs.

Hope this helps!

Cheers,
Fulvio
Dr. Fulvio Paleari
S3-CNR Institute of Nanoscience and MaX Center
Modena, Italy

Post Reply