The unit of the absorption spectrum calculated by BSE

Deals with issues related to computation of optical spectra, solving the Bethe-Salpeter equation.

Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano

Post Reply
Guo_BIT
Posts: 28
Joined: Tue Jun 06, 2023 2:55 am

The unit of the absorption spectrum calculated by BSE

Post by Guo_BIT » Wed Mar 20, 2024 1:44 pm

Dear developers:
I am confused about alpha, the unit of the absorption spectrum. :?:
I obtained the absorption spectrum by solving the BSE with Coulomb cutoff technique(o-2D_WR_WC.alpha_q1_diago_bse).
By modifying the value of BDmRange, the absorption spectrum values have changed significantly. Therefore, I don't know how to convert this result to other units, such as those shown in Figure 3 of this article:
https://journals.aps.org/prb/abstract/1 ... 102.155119

Best,

Jingda Guo
Jingda Guo
Beijing Institute of Technology

User avatar
Daniele Varsano
Posts: 3926
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: The unit of the absorption spectrum calculated by BSE

Post by Daniele Varsano » Wed Mar 20, 2024 2:03 pm

Dear Jingda,
BDmRange governs the width of the Lorentzian functions used to simulate the spectrum. Note that as results, BSE provides excitation energies and strengths, so the meaningful values are the oscillator strengths and not the values of the peaks that of course increase when reducing the damping values.

In the paper you refer, it is reported the imaginary part of the dielectric function (which is unitless), but for a non-periodic system (e.g. 2D) this quantity is not well-defined as it depends on the volume of the used supercell and goes to (1,0) for large vacuum (i.e. isolated 2D system). This is explained e.g. in Cudazzo et al. PRB 84, 085406 (2011) (see Eq. 16 and discussion).

Anyway, you can compare your results with the one published in that paper as the shape and relative weight of the peaks are not affected as alpha is proportional to epsilon, and it is well-defined.

Alternatively you can avoid using the coulomb cutoff, and look at the epsilon, using the same cell used by the authors of the paper (if reported), note that convergence with vacuum will be slower.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

Guo_BIT
Posts: 28
Joined: Tue Jun 06, 2023 2:55 am

Re: The unit of the absorption spectrum calculated by BSE

Post by Guo_BIT » Wed Mar 20, 2024 3:03 pm

Dear Daniele:
Thank you for patiently explaining :D
Jingda Guo
Beijing Institute of Technology

Post Reply