Custom Index Perturbation Model in FDTD



Hi. I’d like to create a custom index perturbation material model. I’d like to write my own equations to compute delta_n and delta_alpha from the np density grid attribute. I need to use a more general Drude model to capture the effects of a dense electron-hole plasma (up to 10^22 cm-3). The model is similar to what you see in the paper linked below. To be clear, I am not interested in the generation aspect. I already have the necessary np density object instantiated in FDTD.

See equations 4 and 5 in the paper linked above. I wrote a script that calculates the delta_n and delta_alpha from the EHP density following this model. How do I implement this in FDTD? I have read a lot about the different material models and there doesn’t appear to be any clear way to do this. Does FDTD allow custom material models? Any help would be very greatly appreciated.




Hi @patrick.goley,

Unfortunately, the np Density index perturbation material model can only use the Drude model or the Soref and Bennet model (for silicon), as shown in the “Charge to index conversion” page.

That said, you can use another model to calculate the resulting spatial index distribution and import that distribution using the “Import Object - Spatial (n,k) data”. The downside is the index distribution data can only be single frequency/wavelength. So if you would be interested in broadband simulation, you would have to do a sweep of single wavelength simulations, each having the corresponding material data. This was discussed in this topic:


Got it. Thanks @gbaethge. I will use the Import Object - Spatial (n,k) data method you suggested. My simulation is narrow band so it should work for me. In my opinion the ability to define a custom index perturbation model would be a great feature to add in a future update. Something to consider. Thanks again. - Patrick


You’re most welcome :slight_smile: That would sure be an interesting feature, I will check if we can add it to our feature request list.