I am simulating the absorption from a corrugated grating with infinite substrate in cst microwave studio. I put boundary condition along grating vector and the other dimension perpendicular to the incident plane to unit cell and upper and bottom side to open.but there are two problems.
1st one is I don't know how to model the infinite substrate. when I put the size of substrate to ten times of the wavelength, some unwanted oscillations and peaks appear.
the 2nd problem is about the size of direction perpendicular to the incident plane. by changing this size the results are also changing.
A part solution would be to define your background as made out of the substrate. You may need to still have a block of substrate in which to see 2 and 3D results, but there will be no reflections at its edges. You will need to define a block of air/vacuum above the substrate, and the size of this may cause problems instead!
Thank you Malcolm. but do you have any opinion about the size of structure in the direction perpendicular to the incident plane?
I am not sure what you mean by the incident plane - do you mean the plane of incidence, which is perpendicular to the interface between the two materials and contains the incident and reflected directions? I expect you mean the plane separating the two materials. If this is what you mean, then making the background material the same as the bottom material means that the bottom boundary can be very close to the interface. The size of the air-space is a problem. One way to get some results might be to give the air some loss - not enough to change the refractive index much but enough to suppress resonances due to the finite size of the air-space. You could try adding an open (add space) condition, but I haven't tried this with non-vacuum background so I don't know if this will add vacuum or background material. I suspect it won't work!
Because the material layer can be thin you may be able to have quite a large air space without having too many mesh cells. Sorry not to be more helpful - someone else may have better ideas. Try changing the size as a parameter and see how the results vary - you may be able to derive the free-space result by stripping out the varying component.
Thank you Malcolm.
my mean was that if the incident wave vector is in z direction and grating vector in x direction, how to model the y direction in which the structure does not have any change and is infinite. I set unit cell B.C for both x and y direction but the results are very sensitive to the size of thickness of structure in y direction that is the resonance frequency has shift with varying this size but the absorption peaks (I am studying absorption of grating) have a little bit change.
best regards.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Algebra