Near-field Tomography (Task 2)
The effort at UIUC headed by Carney has three principal components.
First, an experimental thrust to develop power-extinction based
imaging methods (OPET) produced its first images in accordance
with the year three milestone. Second, theoretical advances were
made on a number of fronts: two results related to the OPET experiment
were published, a novel design and solution of the inverse problem
was presented for probe-free subwavelength microscopy, an analysis
of the information content of near-field experiments for three-dimensional
samples was made, and a major advance was made in the modeling
of the forward problem for apertureless near-field probe microscopy
and incorporated in to the solution of the inverse problem. Third,
collaboration between the BU group and the UIUC group led to advances
in evanescent wave microscopy for fluorescent molecules the modeling
of the diffraction of evanescent fields.
Optical Power Extinction Tomography (OPET)
In OPET, two coherent waves are used to interrogate a sample while
the total power extinguished from the fields by the sample is
investigated. The method yields tomographic (subsurface) images
and obviated the need for phase measurements usually required
for diffraction tomography. We overcame a phase stability problem
through temperature
stabilization and improved mechanical stability in the instrument.
Our first images were obtained this summer and presented at the
International Workshop on Nanophotonics and Nanobiotechnology
in Istanbul.
Theoretical advances
In the OPET experiment, the sample resides on a glass substrate.
It was necessary to account for the substrate in the expressions
for the extinguished power. We developed a new generalization
of the optical theorem to account for the substrate and properly
describe the unperturbed modes. Preliminary to this work, we had
to develop a new form of the Stokes reciprocity relations and
have recently presented this work separately. The solutions of
the inverse problem developed in the course of this work naturally
suggested a method to eliminate the conventional probe used in
methods such as NSOM and replace the probe with a distributed
diffractive element. The resultant data must be processed
making use of the algorithms developed here to produce images.
We showed that such a method is significantly more photon efficient
and noise tolerant than conventional NSOM. A disclosure has been
filed with the UIUC patent department (the Office of Technology
Management).
A very general approach to the analysis of the achievable resolution
and information content of the data for a broad class of modalities
in near-field optics was developed and presented at a conference
this summer. A publication is in preparation. The solutions of
the inverse problem for near-field optics by Prof. Carney in collaboration
with John Schotland of U Penn and demonstrated experimentally
in year two of the MURI were not amenable to the apertureless
instrument currently under development at the Rochester (Novotny)
lab. The problem related to the strong enhancement and multiple
interactions induced by an apertureless probe. This year we (Schotland
and Carney) were able to develop a self-consistent solution of
the problem for these strongly interacting probes. The results
of the forward problem appear to well model the results in the
literature (previously without theoretical description) and were
presented in part at an OSA conference this summer. Results of
simulation of the forward problem are shown in Fig. 5. A publication
is currently in preparation and experimental implementation of
the inverse problem will be conducted with the Rochester group.
Figure 5: Simulation
of the optical near field and the normal derivative of the optical
near-field above a
sample with three buried point-like objects.
Methods in modeling and inverse scattering in collaboration
with the Boston group
The UIUC and BU groups collaborated on two important
problems this year. First, solution of the inverse problem for
the self-interference fluorescence microscope (SFM) built at BU
(Unlu and Goldberg) is being developed jointly by the BU and UIUC
teams. SIM relies on evanescent wave excitation of fluorophores
attached to long molecules. The data available may be fit to curves
describing the exponential decay of the excitation field to determine
the height of the fluorophor above the interface. By carefully
modeling the physics of the forward problem and then solving the
inverse problem, the sensitivity and stability of the method has
been improved and improvements in the lateral resolution are being
pursued. Results were presented at the Instanbul conference this
summer. The student at BU, Bryn Davis, involved in this work will
be graduating this winter. It is hoped that he can be brought
to UIUC as a post-doc in year 4 of the MURI to continue this fruitful
collaboration. The BU and UIUC teams have been collaborating on
simulation and design of an
instrument using focused evanescent fields. This work is of both
practical importance in 100 Å ?z = 0.30 Å instruments
design and fundamental interest as the first calculation of a
diffracted evanescent field with self-consistent boundary conditions.
