Simulation of focused beam incident on substrate


Hi there,

I am trying to model a focused azimuthally polarized beam incident on a SiO2 substrate. I previously modeled a focused azimuthally-polarized beam propagating in air in Lumerical FDTD using the “Import Source” feature. Now I am trying to model the same beam propagating in air, but impinging on a SiO2 substrate. When I introduce the SiO2 substrate into the simulation, instead of forming a nice focused magnetic field at the focus of the beam, the beam reflects off the SiO2 substrate and produces a standing wave pattern. I have tried to remove this effect by drastically increasing the number of PML layers on the boundary conditions and moving the source very far from the SiO2 substrate, but neither of these has given rise to a nice focused beam. Could you please help me in figuring out how to model a focused azimuthally polarized beam impinging on a SiO2 substrate?

I am attaching the fsp file for the beam without the substrate (FocusedAPbeam-2000nm.fsp) and the fsp file for the beam with the substrate (FocusedAPbeam-2000nm-SiO2substrate.fsp).

FocusedAPbeam-2000nm.fsp (6.3 MB)

FocusedAPbeam-2000nm-SiO2substrate.fsp (3.8 MB)

Thank you for any help you can provide.



Hi Tanya,

Thanks for posting.

I checked the simulation files that you uploaded and the setup used in the simulation file without the substrate looked good. Since the source is closer to the monitor positions the field profile of the source at the injection position is small enough that the fields aren’t getting truncated at the edges of the source area.

In the simulation with the substrate it looks like the field profile is getting truncated a bit which could introduce some errors in the injected source.

Other than this, I believe that the interference pattern once the SiO2 substrate is added is a physical effect of the constructive and destructive interference between the incident light from the source and and the reflected light from the substrate and it is not due to any artificial reflections from the PML boundaries, so I don’t think you can avoid the interference pattern for this particular setup.

Is there a particular reason why you want to see the focused beam shape without the interference pattern?


Hi Nancy,

I appreciate you taking the time to look over my simulation files. Thank you for confirming that my simulation setup is correct, that is very helpful to know. Ultimately, my goal is to model the interaction between a focused beam and a resonator on a SiO2 substrate, so the interference pattern may not be an issue for this application (and in fact may produce interesting results!). Thank you for your help!