Modeling of 2D and 3D resonant structures : application to the near-field Raman Spectroscopy.

Objectives
We aim to model the near-field raman spectroscopy experiments developed in our laboratory, as well as optimize the structures and substrate used for these experiments.

Methods
Development of simulation tools based on the Finite Element Method, the Finite-Difference Time-Domain method and the Coupled-Wave Method.

Results and prospects

Near-field Raman spectroscopy is a technique to study the structure of matter at the nano-metric scale. This kind of spectroscopy allows us to find informations about the composition of the studied samples. Nevertheless, this is only possible if the signal over noise ratio is great enough. For this reason, the electric field seen by the sample has to be enhanced by mean of surface plasmons generated through the excitation of resonant substrates.

Two approaches were developed in order to model this kind of resonant structures. The first one is based on the Finite Element Method (currently 2D), the second one on the Finite-Difference Time-Domain method (2D and 3D). The FEM allows the fine description of geometries with a high level of complexity, and can solve harmonic or stationary problems (i.e. independent of time). The FDTD method is currently applied to the study of plasmon resonance in nano-structures, and numerical results in good agreement with those obtained experimentally were observed.

Comparisons between both methods are performed, and the choice of the method is directed by the time requirement, the memory requirement, and the precision of the mesh needed.

A collaboration with Brahim Guizal (Laboratoire P.M. Duffieux, Besançon) will enable us to perform further comparisons with spectral method (mainly the Coupled-Wave Method).