Hi, Kahli, few months ago, I needed to run some time-domain and frequency-domain simulations on the same structure, which contained dispersive materials. As MEEP does not accept dispersive model in frequency-domain computations yet, I had to compute the value of complex epsilon for each frequency manually.
The formula turned out to be the simplest one, known from textbooks, something like: eps_complex(omega) = eps_infinity + sigma/(omega_t**2- i*gamma*omega - omega) + ... ... where omega_t, gamma, and sigma are the three parameters that are passed to MEEP to define the resonance. You may add as many lorentzian terms as needed. I verified that the points (of s-parameters calculation) obtained from the frequency-domain simulation match the time-domain curve very well. I am going to put the whole procedure on the web soon. The conclusion is that you may get along without the meep-fields-analytic-chi1 function and use the Lorentz formula to get correct results. Filip 2013/5/15, Kahli Burke <ka...@uoregon.edu>: > Hi, > > I was looking at some metal material properties created by Aaron Webster > (http://falsecolour.com/aw/meep_metals/meep-metals.pdf), he referenced a way > to have MEEP show the complex epsilon for the material using a 1x1x1 pixel > grid and the 'meep-fields-analytic-chi1' function. This would let you > specify a frequency for the field and very quickly print out the complex > dielectric at any frequency. > > This function doesn't exist anymore, it was in 1.1.1 but not in 1.2. I'm > curious if there is a similar low level function that can do something > similar, I'd like to check that these material properties are still lining > up with other analytic models like the Drude and Debye models. > > It looks like the polarization code was rewritten for 1.2, perhaps to > support more complex materials like anisotropic dielectrics. I looked > through the SWIG wrapper code and there are some functions that reference > 'chi1' (e.g. 'meep-material-function-eff-chi1inv-row') but I'm not sure how > to use them or if they would be of any use for my goal. Any insight would be > appreciated. > > Thanks, > > Kahli Burke > University of Oregon > _______________________________________________ > meep-discuss mailing list > meep-discuss@ab-initio.mit.edu > http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss > _______________________________________________ meep-discuss mailing list meep-discuss@ab-initio.mit.edu http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
