Dear MEEP Users,
I want to do some simulatons which include the linear dispersive
properties of silica. I converted a published set of Sellmeier coefficients
for use in MEEP with a characteristic length of 1 micrometer. The problem is
that when I increase the resolution, the simulation blows up. Here is the
model I am having problems with:
(reset-meep)
(set! eps-averaging? false)
(set! geometry-lattice (make lattice (size .05 .05 12))); define unit cell
(set! geometry (list (make block (center 0 0 3) (size .05 .05 6)
(material (make dielectric (epsilon 1)
(polarizations
(make polarizability
(omega 14.61896)
(gamma 0)
(delta-epsilon 0.69617))
(make polarizability
(omega 8.60279)
(gamma 0)
(delta-epsilon 0.40794))
(make polarizability
(omega 0.10105)
(gamma 0)
(delta-epsilon .89748)))))))); silica half space
(set! k-point (vector3 0 0 0)); enable periodic boundary condition
(set! pml-layers (list (make pml (thickness 1.0) (direction Z)))); set pml
(set! sources (list
(make source
(src(make gaussian-src (wavelength 2) (width .5)))
(component Ey)
(center 0 0 -5.5)
(size .05 .05 0))))
(set! resolution 40)
(set! Courant .25)
(run-sources+ (stop-when-fields-decayed 10 Ey (vector3 0 0 -2) 1e-4)
(to-appended "refhist"
(in-volume (volume (center 0 0 -5 ) (size 0 0 0 )) output-efield-y))
(at-beginning output-epsilon))
I have a half space filled with linear, dispersive silica. I illuminate
this with a short, plane wave pulse from the free space side. In
*.refhist.h5, I record the observed electric field on the free space side.
This single observation point should let me record the incident and
reflected pulse so I can compute a reflection spectrum with a single
simulation. This model runs fine when resolution is set to 20. Also note
that I lowered the Courant number which should have improved stability, but
to no avail. Why does this model blow up with resolution set to 40, but not
at 20?
Thank you for your help,
Neal Skinner, Ph.D., P.E.
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
