Learning Forum

[giovanni.piscopo]

Dear all,

I hope you are doing well. I have some issues regarding the illumination of a metasurface with an incident angle greater than 0° with FDTD.

To be more specific, the metasurface under test is a 2D grating that I previously simulated with Lumerical RCWA, with an incident angle of 45° and P polarization.

Of course with RCWA i got the R/T spectra that I expected however I cannot figure out how to set the same simulation with FDTD.

In particular my main doubts are:

1) Which kind of boundary conditions should I use to ensure still the periodicity (along the x direction) but taking into account the illumnation angle ad the P polarization?

2) What kind of plane wave source should I set?

Thank you in advance.

Best Regards

 

Giovanni

[Kirill]

Hello Giovanni,

Obtaining meaningful results from an FDTD simulation can be challenging.

I’d suggest starting with this article: Periodic Boundary Conditions in FDTD and MODE. It includes an “Additional Tips” section that might help you get up to speed with your simulation.

Basically, use Bloch BC in the direction of angled incidence for a single frequency case, or BFAST for broadband simulation. Use Periodic BC in another direction, and PML with 'Steep angle' profile on top and bottom.

If you still have questions, I highly recommend the Ansys Lumerical FDTD course, which provides a solid overview of all the essential steps for FDTD simulation.

You might also want to check the Application Gallery (APP) for relevant examples.

Best regards,
Kirill

[pm796]

I also have what I think is the same or very similar question! If you have a plane wave source angled at say 45deg and at a distance say 'z' below a surface on which you wish it to be incident, do you also have to offset it laterally by 'z' to accommodate the fact that it is angled at 45deg? Or does the software ignore the incidence angle and automatically just take it and apply it to any structures directly in line with its footprint on the propagation axis? (And why is this not discussed or explained in the reference manual or 'FDTD course'?). 

[Kirill]

Hello pm796,

Thank you for being an active participant in our forum. It’s always a pleasure to see your engagement!

To answer your first question — no, you don’t need extra offsets to simulate an infinite plane wave. The wave propagation appearance results from combining the correct source and corresponding boundary conditions. You can think of the injection plane as an aperture for the infinite plane wave, where the wave angle translates into the corresponding phase distribution at the injection plane. To avoid diffraction effects, Bloch boundary conditions should be applied. Also consider using Broadband Fixed Angle Source Technique (BFAST) with broadband source.

For your second question, I’d recommend watching at least two videos from this part of the course: Plane Wave Source in Ansys Lumerical FDTD — Lesson 2.

Best regards,

Kirill