Hello,
Meep still does some averaging, even when you set eps-averaging? to
false. There's a description of how to disable this averaging here:
http://article.gmane.org/gmane.comp.science.electromagnetism.meep.general/726
Best,
Matt
On Sun, 4 Nov 2007, huoliangcl2 wrote:
> Dear Steven and meep users,
> I have been using meep to simulate spp of gold+dielectric structure for
> a while, but the simulated result cann't conform to the analytical
> result,the meep calculated dispersion curve of spp is lower than
> the curve calculated by the equation ksp=k0*sqrt(epsm*epsd/epsm+epsd) .
> If I increase the resolution the simulated result get closer to the
> analytical result, see the attached figure, but I think the resolution does
> not necessarily to be so high to get the correct result.
> I think perhaps it's the problem of eps averaging, I have used (set!
> eps-averaging? false) in ctl file, and turned off basic eps averageing in
> anisotropic_averaging.cpp, but when I use h5totxt to list the epsilon , I got
> the following data
>
> 2.0164
> 2.0164
> 2.0164
> 2.0164
> 2.741150936030191
> 9.75
> 9.75
> 9.75
> 9.75
> 9.75
> there still exist some kind of averaging, It is equal to introduce a 2nm
> layer of low eps-inf dispersive material if I set resulotion to 500 and a to
> 1000 nm, which will apperently drop the neff vs. omega curve. Is there any
> clue to fix this problem?
>
>
> (define-param sz 3) ; size of cell in z direction
> (define-param dpml 1) ; PML thickness
> (set! geometry-lattice (make lattice (size no-size no-size sz)))
> (define d-au 0.5)
> (define die1 (make dielectric ( epsilon (* 1.42 1.42))))
> (set! eps-averaging? false)
> (define omega_d (/ 1 0.138)) ; plasma frequency of Au in Drude model
>
>
> (define Au ; definition of material dispersion
> (make dielectric (epsilon 9.75)
> (polarizations
> (make polarizability
> (omega 1e-20) (gamma (/ 1 13))
> (delta-epsilon (* (* omega_d omega_d) 1e+40)))
> )))
>
>
> (set! geometry
> (list (make block (center 0 0 0)
> (size infinity infinity sz )
> (material die1))
> (make block (center 0 0 (- (/ sz 2) (/ d-au 2)))
> (size infinity infinity d-au)
> (material Au))
>
> ))
>
> (set! pml-layers (list (make pml (direction Z) (side Low) (thickness dpml))))
> (set-param! resolution 500)
>
> (define-param fcen (/ 1 0.63))
> (define-param df 1)
> (set! sources (list
> (make source
> (src (make gaussian-src (frequency fcen) (fwidth df)))
> (component Hy) (center 0 0 (- (/ sz 2) d-au )) )) )
>
> (define-param k-interp 19)
> (define-param k-points (list (vector3 2 0 0)
> (vector3 2.6 0 0)
> ))
> (set! k-points (interpolate k-interp k-points))
>
> (run-k-points 120 k-points)
>
>
> Sincerely!
>
> Liang Huo
>
>
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
