Incorrect handling of lossy materials

[kitchenknif]

@MK8J
Let's move this over to the relevant project :)

BTW - The files did not get attached for some reason.
Pavel

Hey Mikhail,

Everything looks great. Thanks for the efforts!

On the scripts you have attached to the github, I think,
https://github.com/kitchenknif/PyTMM, might not properly take into account
the extinction coefficient. There is an implementation that does here
https://pypi.python.org/pypi/tmm.

Mattias

Hi Mattias,
Hmm.
I wasn't sure if the way I had implemented complex refractive indices in PyTMM was correct, but couldn't think of any quick way of checking... If you have an example that definitely shows incorrect behaviour, I would really appreciate it.

That being said,
https://github.com/kitchenknif/PyATMM
Should handle complex refractive indices correctly, and can also work with anisotropic (uniaxial) materials, but is a bit more complex.

Hey Pavel,

I wouldn't rule at user error yet . See below is my attempt at comparing 3
TMM implementations.

I compared a stack of a-Si [272 nm] and Si [infinite], optical constants
shown below. So layers without k, I got good agreement between all
models.
Zip below containing everything required.

i did simulations at 0 deg, so polarization doesn't matter.

pyTMM seems to do some where stuff, giving over 100% transmission.

Yeah I has a look at pyATMM, and put it in the to hard basket for now. I'm
only dealing with isotropic materials.

Mattias

[MK8J]

Files take 2

Transferm Martix method.zip

[kitchenknif]

Hmm.
Thanks for pointing this out, will look at it.