On Aug 4, 2008, at 7:49 PM, Zhu, Lei wrote:
I guess the field enhancement that you mentioned is the Raman
scattered field enhancement.
Therefore, the E0 is the incident field and the E is the scattered
field.
You can do the simulation twice. Take a sphere for an example:
first, you can just do a calculation without the sphere. Then you
can do the calculation with the sphere. You can use the fields
gotten with object subtract the fields without the object. I think
this is the scattered field which is the E. Then the field without
sphere is E0.
Actually, for surface-enhanced Raman scattering you really need to do
(at least) four simulations, I think, because the field is enhanced
twice.
1) First, you have some incident wave propagating towards the surface
with your sphere (or whatever). By having a small metal object like a
sphere you will tend to increase the resulting field near the object.
This calculation gives you the field near the object.
2) Second, you suppose you have some molecule (analyte) on the surface
of your object somewhere, with some (known, experimental) Raman
polarizability coefficient. You multiply the field you computed in (1)
by this polarizability to get the induced dipole moment of the Raman
scatterer. Then you have to do a new simulation with a dipole source
at this point, with the given dipole moment determining the current
amplitude/direction, to get the Raman-scattered field back at your
detector (wherever that is). Note that this simulation is at a
slightly different frequency than the first (depending on the process,
Stokes or anti-Stokes).
Then, you want to repeat steps (1) and (2) for some reference
calculation (e.g. a flat surface) to get the enhancement factor. The
increased local density of states near the small object enhances both
steps (1) and (2), which is why I said the field is enhanced twice.
Hence, four simulations.
You also may want to repeat step (2) multiple times -- in general,
there are molecules all over your objects, with some distribution, so
you will need to integrate the detected field from step (2) over all
positions of the analytes.
Regards,
Steven G. Johnson
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
