Discussion Forum

In am simulating plane waves incident upon a stack of thin films (at oblique angle of incidence), each layer with different (real part of) refractive index, e.g. different k-vectors. Now the question is how i should define the k-vector to be used with a periodic FLOQUET boundary condition in COMSOL. I have come up with a few ideas:

1) Just use the k-vector associated with the upper layer, e.g. air.
kx | 0
ky | cos(theta_air)
kz | sin(theta_air)
2) Use a k-vector that is different in each layer.
kx | 0
ky | cos(theta_air)*(z =< z_air) + cos(theta_glass)*(z > z_air)*...
kz | sin(theta_air)*(z =< z_air) + sin(theta_glass)*(z > z_air)*...
3) Define multiple sets of periodic conditions, one for each layer. The first should be like in (1), the next should be
kx | 0
ky | cos(theta_glass)
kz | sin(theta_glass)

where i have assumed that the wave propagates in the z-direction with the electric field polarized in the x-direction and that the two first layers are "air" and "glass". I am aware that COMSOL can do the assignment of k-vectors automatically when periodic ports are used, but i would like to use scattered field formulation if possible, and in this case the k-vectors must be calculated manually AFAIK.

Any comments and/or good references on FLOQUET theory are much appreciated :)