Discussion Forum

I am trying to validate the program Comsol 3.5a using analytical solutions for transmittance, reflectance and absorbance for various geometries and properties.

A transversal wave arriving at an interface with an incident angle theta.i is shown in Figure 1. Figure 2a shows the analytical solution for reflectance versus that obtained using Comsol for real refraction indexes (non-absorbing media). In figure 2b, the same comparison is made for imaginary refraction indexes (absorbing media).

As can be seen, for non-absorbing media the solution rendered by Comsol is very satisfactory. But this is not the case for absorbing media (N=n+i*k).

I think that the problem lies on the solver of Comsol because, as shown in Figure 3b, reflectance plus transmittance (P_tot2D) is larger than 1 (W) which is not logical because the port was configured for a power of 1W and because for a non-absorbing medium reflectance plus transmittance should be equal to 1 (which is shown in Figure 3a), and for an absorbing medium it should be less than 1 (because part of the energy is absorbed by the medium).


I have used all the solvers of version 3.5a but none of them renders correct solutions. Then, is it a problem with this version or are there other parameters that could be varied to obtain valid solutions?