Discussion Forum

Hi everyone,

I'm trying to simulate a plane EM-wave hitting a sub-wavelength aperture in a thin metal film that's perpendicular to the incoming wave. I'd like to model the full electric field not just inside the aperture, but on both sides of that metal film.

So far I've created a geometry consisting of three stacked cylinders (the two outermost are air, and the middle is metal with an air cylinder cut out of its center). I've then wrapped those in another hollow cylinder that consists of PML, since I want the simulation to consider the metal plane to extend out to infinity.

(I'm also setting the "metal" film to glass at the moment; I'd like to use gold but will need to interpolate the permittivity from empirical values as described in the gold nanosphere scattering example on the COMSOL website. I'd like to get the simulation working before making that change).

I've capped the ends of my outermost air cylinders with ports, with one of them having wave excitation set to "on" and the other set to "off." I tried using the circular ports, but they require me to send in non-plane wave TE modes. (In addition, I'm not sure they will work with PML rather than perfect conductor boundaries?)

I then tried to user-define my ports, with the input port "electric mode field" having an x-component of exp(-j*emw.k0*z) and y and z components of 0 (as well as beta of 0). For my output port, the wave excitation was still set to off, but it still asked me to define the electric mode field; I put in the same expression for the x-component as I did for the input port.


The simulation was able to run, but gave me an E-field norm of 0 everywhere.

I apologize if this is a really basic question, but any help would be greatly appreciated.

Thanks!