kch3782
commented
1 year ago Hi.
First, if the input is a constant rather than a grid, such as add_layer (thickness, eps=4.0), it is judged as a homogeneous layer. And, the function for faster calculation is used only for homogeneous thin film.
When testing both the dielectric and metallic thin film, there was no difference in the result when a constant or grid type was used as input. And it was confirmed that the result values are smoothly changed even when the width are slightly increased to lattice size. (We did not compare with analytic cases.)
Could you explain more specific detail about the simulation or used analytic formulas?
Also, if the example you mentioned is 'Example0.ipynb', in this case, it is not two thin flims with thickness, but the simulation and analytic formula for two different homogeneous layers with half-infinite thickness.
Thank you.
MahmutRuzi
Hello dear Changhyun Kim, Thanks for making the code available. I tried to run a simulation where the size of a rectangular strucrue slows increases to the size of the cell lattice size, which is then essentially a homogenous thin layer. However, the calculated result is different from analytical formulas. I noticed in the example that the setup for homogenous thin film is different. Can you comment why the code is designed that way? It will be great if the setup is the same where one can do simulations on structures that smoothly transition from patterned layers to homogenous layer.