Hello Kenny,
the spectral position of 0.5µm sounds like what happens around the
on-set of an additional diffraction order (you have a period of 1, which
means that the second diffraction order in air exists for wavelengths
below 0.5µm and vanishes for wavelengths above 0.5µm).
You have a resolution of 40 and a period of 1, which means that your
entire structure should fit exactly into your grid, which should be 40
times 40 times 320 pixel. However, I found that in some of my
calculations the permittivity values saved into the h5-files exhibited
different permittivity values at the boundaries [in your case, e.g.,
(-sx/2, -sy/2, -1.5)] when compared to values away from the boundary [in
your case, e.g., (0, 0, -1.5)]. Try to look at your permittivity data.
Is your gold film continuos across your boundaries, or do you find
different permittivity values at the boundary when compared to values
from the center? If there are fifferences, than MEEP is treating a
grating which explains your unrealistic features (Maybe I am mistaken
here and these are normal artefacts due to the interpolation during the
h5-saving process. There have been some questions concenring this in the
mailing list).
In order to prevent this you can
a) go to much smaller periods (smaller than your smallest considered
wavelength)
b) make your film box broader than your period (e.g. sx+4/resolution)
Method b) woks well for me. When using RCWA/FMM I use method a)
Best wishes,
Stephan.
Am 21.09.2010 11:47, schrieb [email protected]:
Hi All,
I am having some issues getting the expected reflectivity spectra from a
simple gold film in air. Below is my ctl file. Essentially, even for a
normalizing run with just air I get unrealistic featues in my specta,
notably a large peak at a frequency of 2 (corresponding to 0.5um in my
base units). I have attempted to increas ethe resolution (upto 100) and
have enlarged the cell size in the Z direction to increase the pml size
(to the level now). This also increased the distance between my source and
structure, but to no avail. It appears from the from the field profile
that I may be getting some form of standing wave formed accros the
periodic boundries?
Any advice would be gratefully recieved.
Thanks in advance,
Kenny
; Meep simulation to compare reflectivity of Gold
(define-param sx 1)
(define-param sy 1)
(define-param sz 8)
(define-param pml-t 1)
(set! ensure-periodicity true)
(set! geometry-lattice (make lattice (size sx sy sz)))
(set! k-point (vector3 0 0 0))
(define-param gold? false)
(define omega_d 7.045)
(define Au
(make dielectric (epsilon 5.967)
(polarizations
(make polarizability
(omega 1e-20) (gamma 5.310E-02)
(sigma (* (* omega_d omega_d) 1e+40)))
(make polarizability
(omega 2.168) (gamma 3.498E-01)
(sigma 1.09)))))
(set! geometry (if gold?
(list (make block (center 0 0 -2) (size sx sy 1)
(material Au)))
(list)))
(set! pml-layers (list (make pml (direction Z) (thickness pml-t))))
(define-param fcen 1.75)
(define-param df 1.5)
(set! sources (list
(make source (src (make gaussian-src (frequency fcen) (fwidth df)))
(component Ex)
(center 0 0 (- (* sz 0.5) (+ pml-t 1))
(size sx sy 0)))))
(set! resolution 40)
(define-param nfreq 100)
(define refl
(add-flux fcen df nfreq
(make flux-region
(center 0 0 0)
(size sx sy 0)
(direction Z))))
(if gold? (load-minus-flux "refl-flux" refl))
(run-sources+
(stop-when-fields-decayed 50 Ex
(vector3 0 0 (* 0.25 sz))1e-3)
(at-beginning output-epsilon)
(to-appended "ex" (at-every 5 output-efield-x)))
(if (not gold?)
(save-flux "refl-flux" refl))
(display-fluxes refl)
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
