Yambo Community Forum

Dear Reza,

as commented before, have you looked at the weight of the excitons? If it is come from a single transition, the excwf it will be proportional to a bloch wavefuntion. Check also the LUMO wfs as if you fix the hole, it should be similar to the LUMO.
Best,
Daniele

Dear Daniele,

Thanks for the prompt response.
I'd calculated the weight of the corresponding exciton.

# Electron-Hole pairs that contribute to Excitonic State 1 for iq=1 more than 5.000000%
# K-point [iku] Weight
# : 0.460000008 0.00000000 0.00000000 0.783106834E-1
# : 0.480000019 0.00000000 0.00000000 0.617443740
# : -0.500000000 0.00000000 0.00000000 1.00000000
#
#
# Band_V Band_C Kv-q ibz Symm_kv Kc q ibz Symm_kc Weight Energy [eV]
#
66 67 26 1 26 1 0.584478 0.156310
66 67 25 1 25 1 0.180413 0.195523
66 67 25 5 25 5 0.180412 0.195523

How can I do this? "Check also the LUMO wfs." Do you mean i do it with yambo?

Reza,

Best,

Dear Reza,

you have few transitions contributing to the excitons, anyway a small portion of the BZ is contributing resulting that you can have a rather delocalized excitonic wavefunction. Most probably if you want to look at the localization you need to add more cell in the wfs plot.
KS wfs can be plotted either with QE or Yambo, as you did for the HOMO if I have well understood.
Anyway, looking at these results I doubt that the excwf is similar to the HOMO. Also, please note that you should look at the right k point (here ik=26).
Best,
Daniele

Dear Daniele,

I have plotted the LUMO and HOMO wave functions. The excitonic wave function is similar to HOMO!

1) I have a computational question: Is there any way to plot the exciton wave function with the corresponding phase?
I could not find it in published tutorials.

2) Moreover, in one of your replies on top, you said "the excwf it will be proportional to a bloch wavefuntion", Could you please explain, Why?

Best,

Reza

Dear Reza,

The exc wave function is a linear combination of valence/conduction wfs product, in your case different transitions are involved, in particular, wfs at different k points of the BZ. The fact that you obtain something similar to HOMO does not mean that this is the HOMO.

1) It is possible, but you need to modify few lines of the ypp/excitons/excitons_WFs.F code: see here:
viewtopic.php?p=11484&hilit=excitons_WFs.F#p11484

2) Suppose you have Acv=1 for one transition and zero for all the others. Psi_exc then is proportional to \psi_v(r_o)\psic^*(r)


Best,
Daniele