Home	Research	CV	Academia	Publications	Links

Unexploded Ordnances

--

Optical binding

--

Recent LHM

--

Detection and classification of unexploded ordnances (UXOs) [top]

Research topics: Modeling clutter using the ellispoidal coordinate system in order to reduce the false alarm rate. Detection and classification of UXOs using support vector machine. Comparison with real field measurements (example of UXO in the figure). Summary: Our purpose is to accurately model the reponse of buried UXOs in the presence of clutter, in order to improve their detection and classification, and to lower the extremely high (and costly) false alarm rate. The modeling is performed using advanced mathematical models, based on the ellipsoidal coordinate system: the clutter surrounding UXOs can be of arbitrary shape and therefore needs to be modeled using a very general coordinate system. The UXOs, on the other hand, are modeled in the spheroidal coordinate system since they often exhibit body-of-revolution shapes. The secondary fields produced by buried objects (UXOs or clutter) are modeled in the electromagnetic induction regime (tens of Hertz to hundreds of kilo-Hertz), where displacement currents are negligible compared to conduction currents. Thus, the conduction currents in the soil are negligible and the fields are computed in the magneto-quasistatic regime based on the Laplace equation. Inside the clutter objects, where the wave equation governs the field distribution, the currents are assumed to have a small penetration depth, allowing for the analytical simplification of the field components which become decoupled at the surface. This approximation, valid across the entire frequency spectrum because of the high permeability and conductivity of typical clutter objects, avoids the necessity of using ellipsoidal wave functions and results in a considerable saving of computational time. Numerical results compare favorably with numerical and experimental data (see an example on the right), which proves the usefulness of our method to model unexploded ordnances in clutter contaminated soils. Finally, the optimization approach used to match our numerical predictions with experimental data demonstrates the possibility of remotely inferring the material properties of objects. Sponsor: Army Research Office (ARO) Cold Regions Research and Engineering Laboratories (CRREL)

Optical trapping and binding [top]

Current research topics: Binding phenomena in systems of multiple particles (see animation on the right) Optical matter and particle organization. Brownian motion. Laser Trapped Mirror: maybe the new generation of telescopes. Summary: Optical manipulation of small particles has been pioneered by A. Ashkin in the 1970s. Since then, various groups throughout the world perform experiments of optical manipulation of dielectric objects. In 1989-1990, a group at the Rowland institute for Science discovered that when multiple particles are immersed in an electromagnetic radiation, binding forces appear than disturb the expected equilibrium state. Although there is in fact only one force (the Lorentz force) acting on the particles, the self-consistent interactions between particles within a system yield a redistribution of the force. Despite the large number of experimental verifications of this phenomenon, its theoretical understanding is still lacking solid grounds. This is one of the question we are addressing in this work by using Maxwell's equations in conjunction with the Foldy-Lax multiple scattering equations. To our knowledge, this is the first time that the dynamics of a system of many Mie particles is predicted exactly, and is shown to match experimental data. In addition, this ability of computing the exact force distribution also allows us to work the problem backward and optimize the positions of a series of fixed particles in order to achieve a customizable force distribution in space. This work finds applications in many area where manipulation of small objects is needed. One particular such area we are actively exploring is the ralization of a Laser Trapped Mirror, as introduced by A. Labeyrie in 1979. More details on this project as well as on optical binding in general can be found at [link to come] Sponsors: MIT Lincoln Laboratory NASA Institute for Advanced Concepts (NIAC)

Left-Handed Metamaterials (LHM) [top]

Current research topics: Retrieval process for isotropic, anitrotropic, bianisotropic media Negative refraction from standard anisotropic media, photonic crystals, LHM, and moving media. Experiments: microstrip filter enhancement, antenna isolation. Summary: Left-handed media (LHM) are commonly refering to media in which the permittivity and the permeability are negative (in standard media these parameters are positive). LHM have been postulated in 1968 by V. Veselago from a purely theoretical point of view. It is only at the end of the XXth century that J. B. Pendry imagined a way to realize a negative permittivity and a negative permeability in the GHz regime. His work was experimentally confirmed by R.A. Shelby, D. R. Smith, and S. Schultz in 2001, and marked the beginning of the research in this field. I am interested in the theory and simulation of propagation of electromagnetic (EM) waves in complex media. Left-handed media (LHM) have given us access to negative values of the constitutive parameters, which was not conceivable only a few years ago. In addition, LHM are often intrinsically anisotropic, possibly bianisotropic. The propagation of electromagnetic waves in such media has revealed a series of new and interesting physical phenomena, such as negative refraction, focusing, hyperbolic dispersion relations, inversion of critical angle, inversion of Brewster angle, backward wave, etc. It is also interesting to look at the inverse problem: given a series of measurements illustrating the way EM waves propagate as function of frequency and incident angle, we can infer the material properties. These can be as simple as an isotropic constant permittivity and permeability, and as complicated as a full bianisotropic description requiring the inversion of 72 paramters. An optimization scheme is typically used to handle these highly non-linear problems. A last word to say that contrary to what is often believed apparently (reading recent journal papers), negative refraction is not a new phenomenon. It is for example well-known that negative refraction can happen in standard (i.e. where all the constitutive paramters take positive values) anisotropic media when the principle axis is properly rotated with respect to the incident plane. Another instance where negative refraction is known to happen is with moving media at high velocities (higher than the Cerenkov velocities). Recently, negative refraction has been obtained with photonic crystals, and finally with LHM. However, LHM is probably the only material than can yield a negative refraction of both the phase and the power, which is what yields some of its unique properties (refraction laws for biaxial media are available in one of our paper). Sponsors: MIT Lincoln Laboratory Office of Naval Research Naval Research Laboratory