Gap-Plasmon Physics – PGP

This project is entirely numerical and theoretical. Its goal is to provide a thorough understanding of how light can propagate close to metals (by taking into account phenomenon that are instrinsically of a quantum origin) and to devise numerical methods that would be able to simulate this propagation. The numerical methods we currently use are unable to describe these new phenomenon.

This is important because, nowadays, it is very common for simulations to predict which experiment will have a chance to give new results. Simulations almost always precede experiments. Thanks to simulations, it is easy to optimize: without even doing an experiment, it is possible to predict which structure will be the best (the more efficient to scatter or absorb light for instance).

But a goal of the project is to set up collaborations with experimentalists worldwide (for instance with Duke University). They could do the experiments that we have imagined (and we have already many ideas !).

We have devised very efficient numerical methods to simulation the propagation of light in structures that are multilayered (composed of many different layers, that can be metallic or dielectric). This methods are able to take nonlocality (the effect of interactions between electrons) into account. We have shown that there are mainly two cases when these effects really kick in: when trying to design super-lenses that can beat the diffraction limit (while classical lenses can't), and when gap-plasmons are excited. This work has already suggested experimental setups to actually measure the funcamental parameters of this nonlocality.

We are already working with experimentalists in order to see how hard it can be to realize the experiments we have in mind. After the theoretical work is done, we will turn to more pragmatic problems: how sensitive will our structures be, how efficient, and where and how exactly can they be useful.

The project is ongoing right now. We have already submitted an article and we are on our way to free the codes we have written, to make them public and available for anybody who would be interested in using them.

A gap-plasmon is the fundamental mode supported by a metallic waveguide when the thickness of between the two metallic interfaces becomes very small. In this regime, the mode presents a very high effective index. This allows to understand why nanocubes that are only 75 nm wide, coupled to a metallic film constitute a optical resonator : the gap between the cubes and the film behaves as an optical cavity for the gap-plasmon.

The interpretation in terms of gap-plasmons has been largely ignored by the community of metamaterials, who was relying more on an electronic circuit vision of meta-atoms (the basic constituent of metamaterials). We think there are many structures that are effective cavities for gap-plasmons that have not been considered that way. Such a vision has the huge advantage to allow for a easy modelization of the structure, easing the optimization of the structure for a given purpose.

The first goal of the project is to study “gap-plasmon optics” : how gap-plasmons are reflected when the waveguide comes to an end, how they can be guided by multiple reflections which will finally lead to an analysis of gap-plasmon cavities based on semi-analytical models. Using these models, we will study phenomena like the interferometric control of the absorption by meta-atoms and its impact on the cross-section of coupled film nanocubes or of nanocube dimers. We want this kind of study to help imagine new bottom-up processes for the fabrication of these optical metamaterials.

On another hand, recent progress in the fabrication and assembly of deeply subwavelength structures have allowed to reach the limits of our current electromagnetic description of metals (in which the fundamental nonlocality of their response is neglected). It turns out that, due to their high effective index, gap-plasmons are very sensitive to nonlocality. Some of us have shown recently that for very small gaps (below 5 nanometers) nonlocality is beginning to have an impact on the dispersion relation of the gap-plasmon. This is precisely the thickness of the spacer for which the absorption cross-section of film-coupled nanocubes is maximum.

When gap-plasmons are responsible for the resonance of a given structure, this means that the chances are high the nonlocality cannot be neglected when trying to shrink the size of the resonators by increasing the effective index of the gap-plasmon. In order to thoroughly assess the impact of nonlocality on such structures, adapted numerical tools are required, like modal methods. These are well known to facilitate the physical analysis of optical devices. The Fourier Modal Methods are however not suitable here because of their inability to take the peculiar boundary conditions imposed by nonlocality. New approaches, like the ones that some of us have proposed in the framework of modal methods, are a clear path to simulations that can account for nonlocality. This is the second goal of the project.

Many other structures are likely to be sensitive to nonlocality, as some preliminary results suggest, like metallic gratings with very narrow and shallow grooves, or hyperbolic metamaterials either because they present very high effective index, or because they are used to build flat lenses with super-resolution. We can finally mention modes guided in grooves or along corners. The last goal of the project is to assess the impact of nonlocality on all these structures.