Backside and front side multilayer absorption

[toutsec]

Hi,
I may be wrong, but the example of the front/back multilayer doesn't seem logical to me.
In both cases, you have differential absorption. And yet, for the backside calculation, the result gives 0% absorption.
This is not consistent. I am able to reproduce this bug. I would guess that there is an issue in the get_multilayers_absorption_profile function (obviously) where you introduced the backside boolean.

Thanks for the software!

udkm1Dsim-2.0.0

[dschick]

Hi @toutsec ,

Thanks a lot for finding this mistake.

I think the problem must be related to this part here:

udkm1Dsim/udkm1Dsim/simulations/heat.py

Line 530 in 59b015f

opt_ref_indices[0] = 1+0.0j

For most samples you do not care about the last interface, typically from the substrate to air or something else.
So in the code, I only introduce an air/vacuum layer as the first medium but not as a last medium.
For the backside excitation with transparent substrates, this become relevant I guess.

Would you be willing to help testing some new versions solving this issue?

[toutsec]

Hi Daniel,
Yeah sure I can test new version, no problem.
I have also a personal code to backtest yours (or backtest mine).
Best

[dschick]

Great, actually, I have already started to implement a more general matrix-formalism for optical simulations, based on pyGTM, see #97

This formalism could not only be used for scatting simulations but also for calculating the laser absorption in the heat class.
However, the main issue was also to transparently handle super- and substrates.
So maybe this is a good point to also work on that part.

May I ask what your background is and what you want to use the simulations for?

[toutsec]

I am one of the author of Gu and al (Science advances, 2023 https://www.science.org/doi/10.1126/sciadv.adi1160 ) where I did the simulation of the phonons propagation in BiFeO3 with the help of your software.
I'm doing experiments that require a comparison between backward and forward pumping.