Modeling a droplet between two silicon structures- issues with reflectance in Si

  • 13 January 2020
  • 4 replies
  • 98 views

Hi!

I am trying to model fluid that is sandwiched by two different silicon structures to gauge optical throughput. Since material data for Silicon in visible wavelength regime is not provided in the default Optic Studio, I am modeling it using a refractive index fit. During the same, I am not sure how to put in the transmission data for an accurate model.

As I understand, when light is incident on silicon, there is non-zero reflectivity and the rest of the light is absorbed within Silicon (600 micron substrate). I don't know how to accurately model the transmission or the silicon material to show me what reflectivity would I expect. Ideally, I expect about 5% reflectivity, as I am deducing from my experiments, but I need a prediction using Zemax.

Any suggestions will be helpful! Thank you!



~Aditi

4 replies

Userlevel 7
Badge +3
Hi Aditi,



Silicon is in the INFRARED catalog, and if you have data you want to enter there are multiple ways of doing that. The one thing you might want to check is if there is multi-beam interference in the fluid layer. You could just make a sandwich of Si/Fluid/Si, and that will give you the correct reflectivity/transmission at the surfaces, but if there is any etaloning going on (multi-beam interference) in the fluid layer then define the fluid layer as a coating instead. Check out these two sample files for how to do this:



Zemax\Samples\Sequential\Polarization\Fabry-Perot.ZMX. This is a classic Fabry-Perot with two pieces of glass as the substrates, and a coating (of air) in between them.



Documents\Zemax\Samples\Sequential\Polarization\Etalon.ZMX. This treats the whole etalon as a coating, substrate and all.



- Mark
Userlevel 7
Badge +3
P.S. See https://my.zemax.com/en-US/Knowledge-Base/kb-article/?ka=KA-01673 for how to add materials data to OpticStudio.
Hi Mark!



Thanks for responding- I have been using your article to model Silicon.

Since I am working in visible region, the existing material Si in INFRARED catalog does not work for me. I did get a range of refractive index and wavelength pairs to input into the glass fitting tool to get a Sellmeir and Conrady fit for the newly defined Si in the visible range. However, I don't have the exact transmission data, which probably makes the model less accurate. Should I be setting the transmission as zero then assuming from one face to the other, there is no light transmitted in the silicon?



Also, if instead of defining the material using glass fitting tool, I model the material using Nd and Abbe Vd, my surface reflectivity is different too. Is that to be expected?



Because both of those Silicon objects are covered with an ARC coating, theoretically speaking, I should not be getting more than 5% reflection from either of the surfaces. If I understand correctly, that should not cause multi-beam interference then?

I will look into modeling the fluid layer as a coating too!



Thanks again!



Best,

Aditi

 
Userlevel 7
Badge +3

Hey Aditi,



Defining index via just nd and Vd is approximate, and the calculations will just use whatever index value they are given. The key thing is to give the calculations the correct index data, whatever that is.


Reply