Problems with simulations using tilted wave

[Luis Rodriguez]

Dear Scott,

I have another question. I have tried to calculate the reflectance and transmittance of a semi-infinite layer (a substrate) of a homogeneous and continuous material with a constant refractive index at an angle of incidence (AOI) of 45 deg, so the results can be easily verified with the Fresnel equations. Consistently I was getting incompatible results when I tilted the incident wave. Actually, everything indicates that when I am using a wave tilted 45deg, the software does not take it into account, and the result I get is like my wave were tilted 0-degrees (normal incidence). For instance, you can check that my incident wave is RH-polarized, but when I check the intensity of the reflected wave with an observation area I get the same values for both polarizations, x and y, which is incorrect for an AOI=45 deg. Similar happens when you tilt the wave 60 degrees or even 80 degrees. Besides, I have ploted the R vs wavelength at AOI=45 deg and I compared it with Fresnel equations, and the reflectivity result is compatible with normal incidence again.

I have attached a copy of the OptiFDTD layout I am using, and a plot summarizing the main results. The materials layer is longer than the wafer size to get rid of the backside reflectance. I use 3D simulation.

Eventually I found a explanation in the manual (OptiFDTD Technical Background and Tutorials, version 8.1 for Windows 32/64 bits). It says in pag. 41 that: “…Currently, plane wave does not support the tilting angle. Currently, the plane wave simulations is only effective for symmetrical structures”

So, my question is, I am doing something wrong, or the OptiFDTD indeed does not support tilted waves? The weird thing is that I have seen posts in this forum and also papers using OptiFDTD to correctly simulate reflectances or transmittances at different angles of incidence….

About the version, I am using the free 32-bits version. By the way, another question is, if the free version does not support tilted waves as indicated in the manual, does the full version support it?

Thanks in advance!
Luis

[Luis Rodriguez]

And this is the result of the simulation using a tilted wave at 45degrees compared with what you obtain using Fresnel equations. All reflectances are averaged for both polarizations.

[Scott Newman]

Hello Luis,

While it should support it, I cannot comment on the 32-bit support as it is a dated product that has not been maintained. The current product does support tilt and we are currently updating the input plane for improved support. Let me take a look at your design and see if I can determine what issue you are running into.

Scott

[Scott Newman]

Luis,

I was unable to open your fdt file as there is an issue with the material definitions when opening in the newest products. Unfortunately the 32-bit product is quite out-dated and not supported.

Without seeing your design I can speculate on the issue you are encountering and the confusion regarding what you are reading. The limitation at this time with tilt is not in the input plane but instead in the boundary conditions. Currently the periodic boundary conditions require symmetry both in structure and in field which is not the case when you tilt. The measurements you are looking for can be done without the use of PBC with careful setup of your design.

In order to move forward I would recommend you get an evaluation copy of the latest product.

Scott

[Author]

Posts