wdm2 example: Is it correct to use effective index method with topology optimization?

Question

wdm2 example: Is it correct to use effective index method with topology optimization?

Closed this issue 4 years ago · 1 comments

Hi,

I have some doubts about using the effective index method (EIM) to simulate a structure that is not slowly varying in the x-y plane. As I understand it, to use the EIM approximation the mode profile should be slowly varying in the y-direction which I believe is not the case here?

Also, I have been trying to validate the example with Lumerical 3D FDTD and 2.5D (EIM-based) FDTD and the results are totally different from spins-b's. I have used the same configurations as in the spins-b code. Any advice would be appreciated.

Thanks in advance.

Update: I should have used Lumerical 2D FDTD not 2.5D or 3D for validation. My question is, should this output be the input to the 3D optimization?

Answer 1 · 2021-04-05T04:22:30.000Z

Hi! In general, you cannot automatically extract a 2D optimization/structure and expect it to work in 3D. EIM would only work well for slowly varying structures as you mention.