Electromagnetic and Light Scattering by Particles
N e w s l e t t e r
October 2022
Issue 108
20th International Conference on
Electromagnetic and Light Scattering
We are pleased to
announce the 20th International Conference on Electromagnetic and Light
Scattering which will take place
14-19 May 2023
in Almunecar, Spain
The scientific
scope of this conference along with relevant background information can be
found at:
https://www.granadacongresos.com/els2023
Important dates:
- Registration
opening: Friday October 14th,, 2022.
- Abstracts
submission deadline: Friday, March 3rd, 2023.
- Early-bird
registration: Friday March 17th, 2023.
For the ELS
conference we have secured availability of a limited number of rooms at the
conference venue (Hotel Bahia Tropical). Almunecar is
a popular touristic destination. We strongly recommend you to
book your accommodation as soon as possible. The hotel reservation at
the agreed ELS fair must be done via the conference website.
We hope to meet
you all in Almunecar!
Olga Munoz on behalf of the ELS LOC
Postdoc
position in electromagnetic scattering.
The planetary science
research group at the University of Helsinki is seeking a postdoctoral
researcher starting from February 2023 for a fixed-term employment for three
(3) years.
The postdoctoral
researcher will work in a research project entitled Asteroids in the Deep:
Characterization of Sub-Surface of Near-Earth Asteroids the Moon and Mercury
using radar observations.
Do you have
experience in numerical modeling of electromagnetic scattering (or related
processes)? Can you program using Python and/or FORTRAN?
If yes, apply
at https://jobs.helsinki.fi/job/Postdoctoral-researcher%2C-Department-of-Physics/754994602/?feedId=350602&utm_source=CareerSite_UniversityOfHelsinki or
email to anne.virkki@helsinki.fi for more
information.
==================================================================================
SOFTWARE
UPDATES
Dear ADDA users,
We have uncovered
several bugs in calculation of Sommerfeld integrals, which is relevant for the
surface mode (particles
near a substrate).
They are related to standalone somnec.c from the
independent package NEC2C that we used for these
calculations.
While this routine was considered reliable, it happened to be not suitable for
all possible input
parameters,
especially with respect to substrate refractive index. All the details are
given in
https://github.com/adda-team/adda/issues/326.
Most of the several uncovered issues lead only to minor errors (around
1%), but we have
observed 10% and larger errors for very oblate particles lying on plasmonic
(metallic) substrates. So
all previous ADDA
calculations for such configurations must be reconsidered.
To fix these
issues, we had to modify the somnec.c itself. The
main round of the fixes is complete and tested, it is
already present in
the master branch of ADDA. It should work for any passive substrate and in wide
range of particle
sizes with
relative accuracies of Sommerfeld-integral calculation on the order of 0.0001
(much smaller than the expected
accuracy of
computed scattering quantities). So it is highly
recommended to use this version for simulations in surface
mode instead of
the latest release. Moreover, we would appreciate any comparisons with other
simulation methods,
especially for
metallic substrates, similarly as was done for
dielectric substrates in
https://github.com/adda-team/adda/wiki/ComparisonOtherCodes#particles-near-surface .
If you perform such a comparison,
please share the
results with me or at adda-discuss@googlegroups.com .
Since we already
had to understand the inner algorithms in somnec.c,
we further plan to polish these routines and make
it fully
conforming with the ADDA style. Hopefully, we can tune it to make even better
accuracies possible (say, 1e-10),
although that is
not really needed in practice. And maybe also accelerate the computations
somewhat when the default
accuracy is
desired. But these should not affect the ADDA results in any visible way. And
we will keep this routine
standalone, so it
can be easily used in any other code. If you want to play with it, we also
provide the separate
testing program (somnec_test.c), which compares different ways to calculate
the Sommerfeld integrals, that are
implemented in somnec.c, both in terms of results and timing.
Best regards,
Maxim Yurkin.