Yambo Community Forum

Dear Xiangtian,

1) Yes, they can be compared. The exciton wavefunction is a conditional probability amplitude, ie the probability to find an electron in space having the hole fixed in space.
2) Yes correct, depending on the version you are using by default the max if the exc_wf is normalized to 1. This can be controlled by input by using the keyword "NormToOne".

Best,

Daniele

Dear Daniele，

Thank you for the reply. I got the point!

Hello,

I have performed calculations to compute the exciton oscillator strength, amplitude, and spatial distribution. However, I am struggling to interpret the output file in terms of identifying the strongest exciton and its corresponding "state" in the ypp_AMPL.in file.

I have attached the output and report files, along with the strength/energy figure.

Thank you,

Reza

Dear Reza,

in the E_sorted file you have exciton index, energy and strength.
The strength is normalized to the largest one, in your case is the first one (index 1). The normalization factor is also reported.
Please note that you find at negative excitation energies, so it is possible that your calculation is not converged or you do not have taken into account qp correction correctly.
Once you have identified the index of the exciton you want to analyze, you can use ypp (ypp -a), select the index and look at its composition looking at the *weight output.

Please note that degenerate excitons are summed up, if you want to avoid this and analyse a single exciton you need to set:
Degen_Step=0 eV

Best,
Daniele

Dear Daniele,

I calculated the exciton wave function based on the instruction. I have some problems analyzing it. 1. I cannot understand how I should determine the number of cells. Since I increased it, I observed the spreading of the exciton wave function (I attached the image), and I do not know if there is an optimum value for the number of cells.

2. Regarding the exciton index state (for instance, No. 1 in the E_stored output file), how is it possible to compute its contribution in different k-states?

3. In which input file can "ypp -a" be used?
excitons # [R] Excitonic properties
amplitude # [R] Amplitude
States= "1 - 1" # Index of the BS state(s)
BSQindex= 1 # Q-Index of the BS state(s)
Degen_Step= 0.010000 eV # Maximum energy separation of two degenerate states
and how can the contribution be analyzed?

I attached the related output file.

Best.

Reza

Dear Reza,

1) Increasing the cell you span more space, so if you want to look at the localization of the excitonic wavefunction you need to add a number of cells up to see its localization.

2) In the weight file, you have the contribution from the different k points participating in the exciton. Please note that you have in your input file Degen_Step= 0.010, and it seems you are considering degenerate excitons (the sum of weights is larger than one). I do not know if this is intentional.

3) I'm not sure, I understand your question. In the weight file you have the transitions (bands, k points) contributing to the first exciton. In the first line you have the contribution where bands have been summed up.

Best,
Daniele

Dear Daniele,

I have a very particular behavior when plotting the exciton wave function. The point is that when I fixed the position of the hole and then plotted the exciton wave function, it looked like the HOMO wave function. What could be the reason?

Best,

Reza

Dear Reza,

the exciton wfs is defined as:
\psi^exc(r_e)=\sum_vck Acvk \psiv(r_h)\psic^*(r_e)

In case the exciton comes prevalently from homo-> lumo transition, i.e Acv~1 for the homo-> lumo transition and nearly zero for the other transitions, the exciton wave function will be nearly the same as the lumo wavefunction.

Best,
Daniele

Dear Daniele,

I redo my calculations to be sure about them. I have the same behavior in the exciton wave function (HOMO-like).
I attached the report of my BSE and exciton calculations for your review if there is a problem.

Best,

Dear Daniele,

I redo my calculations to be sure about them. I have the same behavior in the exciton wave function (HOMO-like).
I attached the report of my BSE and exciton calculations for your review if there is a problem.

Best,