Hello every one,
I need your help to solve the problem for silicon simulation using comsol multiphysics.
In the plot, the reflection spectrum from 400 nm - 1000 nm are almost matched, but 1000 nm -1800 nm there in some difference in the result. I am using port condition to calculate the reflection and transmittance spectrum.
T = abs(emw.S21)^2 and R = abs(emw.S11)^2
again I am using the dispersive nature refractive index of the Si by interpolation the function.
In my point all thing are correct but what wrong with simulation not getting the right result in the whole band of my interest.
Could you suggest me the right answer to minimize this error.
Thank you All
Dear Wajid,
welcome,
What is the thickness of the silicon layer assumed in the simulation with COMSOL?
The transmittance of the silicon dependence on it thickness. As the thickness increases the cut off wavelength increases. And this is most probable which i see from your curves.
Solution: Make simulation at different thickness of silicon
Best wishes
Thank you very much Sir for your kind response. you are right the transmittance of silicon is thickness dependent and to get the most suitable result at least 500 micrometer thick size is required but my computer capacity is not that much to run it with study dependent mesh model. here the free space wavelength i am taking is 500nm so if i take all other parameter in micro size, does it necessary to change wavelength scale (e.g 500um).
Dear Wajid,
Since the optical attenuation is proportional to exp - alpha x, with alpha is the absorption coefficient and x is the propagation distance in the material. Then you can increase the mesh size and make it smaller = 1/ 10 alpha alpha can be made varying with the wavelength. In this way you can overcome the problems of intensive computations. You can also use adaptive mesh where the mesh is made fine in the side of light incidence and course away from light incidence surface.
Best wishes
- Badges
- Science topic