for the exact exciton energy

[Quxiao]

Dear all:
When I calculate the BSE,and run the ypp,I got the o-2D_WR_WC.exc_qpt1_E_sorted.
However, from the o-2D_WR_WC.eps_q1_diago_bse,I find the first absorption peak is located at 1.55(ev).
Then I open the o-2D_WR_WC.exc_qpt1_E_sorted,find the index is the 34，the strength is 1.
Let me confused is that ,before the 34th,also have the local maximum strength,for example,the 14th,which strength is 0.28041,biger than the 13th and the 15th.
So I think that if I use two few points to calculate the BSE?however，I add the points to 32*34*1,get the same results,before the 1.55(ev),there is no peak in the BSE specturm.So,is the results right?please give me some help
Thanks very much!
Yours
Quxiao
BIT

[Daniele Varsano]

Dear Quxiao,
I'm not sure I have understood your problem.

Can you attach the *E_sorted and *eps file you are talking about?

You can use the attachment button in the post renaming the files with an allowed suffix (e.g. .txt or .zip).

Best,
Daniele

[Quxiao]

Dear Daniele :
o-2D_WR_WC.exc_qpt1_E_sorted:
# Maximum Residual Value = .12151E+01
#
# E [ev] Strength Index
#
0.782 0.310E-11 1.00
0.910 0.384E-13 2.00
0.9517 0.1023E-2 3.000
0.9753 0.3680E-8 4.000
1.27 0.108E-10 5.00
1.27 0.286E-10 6.00
1.31 0.527E-13 7.00
1.32 0.581E-12 8.00
1.38 0.121E-13 9.00
1.414 0.4375E-7 10.00
1.43 0.265E-09 11.0
1.43 0.851E-09 12.0
1.444 0.1968E-3 13.00
1.44400 0.03186 14.00000
1.450 0.6970E-6 15.00
1.455 0.2191E-6 16.00
1.50781 0.72065 17.00000
1.514 0.1242E-6 18.00
1.516 0.4749E-7 19.00
1.527 0.6585E-6 20.00
1.548 0.6278E-8 21.00
1.551 0.1923E-7 22.00
1.552 0.3786E-8 23.00
1.552 0.1037E-7 24.00
1.55821 0.03237 25.00000
1.559 0.2045E-6 26.00
1.564 0.1332E-7 27.00
1.56656 0.04821 28.00000
1.570 0.1600E-4 29.00
1.573 0.4594E-7 30.00
1.574 0.2012E-4 31.00
1.575 0.1255E-4 32.00
1.582 0.5989E-8 33.00
1.58707 1.00000 34.00000
1.589 0.7365E-6 35.00
1.590 0.2457E-4 36.00
1.595 0.6825E-5 37.00
1.60437 0.05944 38.00000
1.612 0.3053E-2 39.00
1.614 0.1963E-6 40.00
1.615 0.3630E-8 41.00
1.616 0.2346E-6 42.00
1.618 0.8798E-8 43.00
/////////
o-2D_WR.eps_q1_diago_bse:BSE.png

BSE.png

[Quxiao]

Dear Daniele:
I submit the figure again to explain better for my question.
Thanks very much,have a good day!
Yours Quxiao
BiT

PE-BSE.png

[Daniele Varsano]

Dear Quixiao,
mosto probably the other excitations are hidden beneath the main peak.
If you want to recognise these excitations having enough osc. strength you may calculate the epsilon reducing the damping to lower values (BDmRange) in the input file which is 0.2eV by default.

Best,

Daniele

[Quxiao]

Dear Daniele:
Thanks for your reply!
I think I got it from your help.If I modify a small BSERange,for example 0~5ev,could I got the small peak?
And I also have an another question:Is the amplitude I got from the o-2D_WR_WC.exc_qpt1_amplitude normalized？or means that Is it divided by the same value？
I am interseted about the amplitude,is it means a density?So it could express the electron-hole pairs contribution the exciton aggregate or discrete,is it?
Thanks for your selfless！
Yours
Quxiao

[Daniele Varsano]

Dear Quxiao,

I think I got it from your help.If I modify a small BSERange,for example 0~5ev,could I got the small peak?

No, the range does not affect the broadening, you need to reduce the damping as I wrote you in the previous post.

I am interseted about the amplitude,is it means a density?So it could express the electron-hole pairs contribution the exciton aggregate or discrete,is it?

If you are interested in the elects-hole pair forming the exciton you can have a look to the *weight* file, there you can find the el-hole pairs and their respective weight for a given exciton. The amplitude is a joint density of states, if I'm not wrong, by default, the energy considered are the KS and not the QP.

Best,

Daniele

[Quxiao]

Dear Daniel
Thanks your reply,I know that o-2D_WR_WC.exc_qpt1_weights.The energy is KS energy should be small effect in my question,I use the yambopy to plot the contribution,find the bandstructure is KS energy,however,I just need the distribution.
Could I see the electron-hole pairs contribution the exciton aggregate or discrete just by the exciton contribution figure? When I use more kpoints to calculate the BSE,I find that the weight of one electron-hole pair become smaller,I guess it may because the A(kcv) becomes smaller.
The A(kcv) is the eigenvector of the BSE:http://www.yambo-code.org/wiki/index.ph ... nalization
The macroscopic dielectric function (from which the absorption and EEL spectra can be computed) is obtained from the eigenvalues Eλ (excitonic energies) and eigenvectors Aλcvk (exciton composition in terms of electron-hole pairs) of the two-particle Hamiltonian.
however,http://www.yambo-code.org/wiki/index.ph ... e:Ampl.pnghttp://www.yambo-code.org/wiki/index.ph ... e:Ampl.png[/url],there Acv may be a amplitude(it's modular square is probability）
or maybe the A(kcv) is a matrix？I am confused.
Thanks for your selfless again
Best wishes!
Quxiao

[Daniele Varsano]

Dear Quxiao,

Could I see the electron-hole pairs contribution the exciton aggregate or discrete just by the exciton contribution figure?

Sorry but this is not clear to me, can you explain more explicitly what you need?

When I use more kpoints to calculate the BSE,I find that the weight of one electron-hole pair become smaller,I guess it may because the A(kcv) becomes smaller.

This is expected, as you have finer grid and the Akcv are normalised such that \sum |A_kcv|^2=1

there Acv may be a amplitude(it's modular square is probability）

Yes, the exciton state is defined as:
\Psi_lambda(re,rh)= \sum_cvk Acvk \psi_v(rh)\psi_c(re)

Best,
Daniele

[Quxiao]

Dear Daniele：
Thanks for your reply!
I'm sorry I didn't make it clear,I want to konw how electron hole pairs contribute excitons (centralized or decentralized) in my systerm.
Now,I maybe learn that it could just get from the exciton contribution figure by yambopy,the exciton contribution diagram or the exciton distribution diagram could express what I want to konw.
Thanks very much!
Best Wishes!
Quxiao
BIT