Demetri Psaltis

Developing optofluidic technology through the fusion of microfluidics and optics

We describe devices in which optics and fluidics are used synergistically to synthesize novel functionalities. Fluidic replacement or modification leads to reconfigurable optical systems, whereas the implementation of optics through the microfluidic toolkit gives highly compact and integrated devices. We categorize optofluidics according to three broad categories of interactions: fluid–solid interfaces, purely fluidic interfaces and colloidal suspensions. We describe examples of optofluidic devices in each category.

A multilayered neural network controller

A multilayered neural network processor is used to control a given plant. Several learning architectures are proposed for training the neural controller to provide the appropriate inputs to the plant so that a desired response is obtained. A modified error-back propagation algorithm, based on propagation of the output error through the plant, is introduced. The properties of the proposed architectures are studied through a simulation example.<<ETX>>

Position, rotation, and scale invariant optical correlation

A new optical transformation that combines geometrical coordinate transformations with the conventional optical Fourier transform is described. The resultant transformations are invariant to both scale and rotational changes in the input object or function. Extensions of these operations to optical pattern recognition and initial experimental demonstrations are also presented.

Optical Implementation Of The Hopfield Model

Optical implementation of content addressable associative memory based on the Hopfield model for neural networks and on the addition of nonlinear iterative feedback to a vector-matrix multiplier is described. Numerical and experimental results presented show that the approach is capable of introducing accuracy and robustness to optical processing while maintaining the traditional advantages of optics, namely, parallelism and massive interconnection capability. Moreover a potentially useful link between neural processing and optics that can be of interest in pattern recognition and machine vision is established.

Adaptive optical networks using photorefractive crystals

The capabilities of photorefractive crystals as media for holographic interconnections in neural networks are examined. Limitations on the density of interconnections and the number of holographic associations which can be stored in photorefractive crystals are derived. Optical architectures for implementing various neural schemes are described. Experimental results are presented for one of these architectures.

Non-volatile holographic storage in doubly doped lithium niobate crystals

Photorefractive materials are being widely investigated for applications in holographic data storage. Inhomogeneous illumination of these materials with an optical interference pattern redistributes charge, builds up internal electric fields and so changes the refractive index. Subsequent homogeneous illumination results in light diffraction and reconstructs the information encoded in the original interference pattern. A range of inorganic and organic photorefractive materials are known, in which thousands of holograms of high fidelity can be efficiently stored, reconstructed and erased. But there remains a problem with volatility: the read-out process usually erases the stored information and amplifies the scattered light. Several techniques for ‘fixing’ holograms have been developed, but they have practical disadvantages and only laboratory demonstrators have been built. Here we describe a resolution to the problem of volatility that should lead to the realization of a more practical system. We use crystals of lithium niobate — available both in large size and with excellent homogeneity — that have been doped with two different deep electron traps (iron and manganese). Illumination of the crystals with incoherent ultraviolet light during the recording process permits the storage of data (a red-light interference pattern) that can be subsequently read, in the absence of ultraviolet light, without erasure. Our crystals show up to 32 per cent diffraction efficiency, rapid optical erasure of the stored data is possible using ultraviolet light, and light scattering is effectively prevented.

Recognitive Aspects of Moment Invariants

Moment invariants are evaluated as a feature space for pattern recognition in terms of discrimination power and noise tolerance. The notion of complex moments is introduced as a simple and straightforward way to derive moment invariants. Through this relation, properties of complex moments are used to characterize moment invariants. Aspects of information loss, suppression, and redundancy encountered in moment invariants are investigated and significant results are derived. The behavior of moment invariants in the presence of additive noise is also described.

Optical Information-Processing Based On An Associative-Memory Model Of Neural Nets With Thresholding And Feedback

The remarkable collective computational properties of the Hopfield model for neural networks [Proc. Nat. Acad. Sci. USA 79, 2554 (1982)] are reviewed. These include recognition from partial input, robustness, and error-correction capability. Features of the model that make its optical implementation attractive are discussed, and specific optical implementation schemes are given.

System metric for holographic memory systems

We introduce M/# as a metric for characterizing holographic memory systems. M/# is the constant of proportionality between diffraction efficiency and the number of holograms squared. Although M/# is a function of many variables in a holographic recording system, it can be measured from the recording and erasure of a single hologram. We verify experimentally that the diffraction efficiency of multiple holograms follows the prediction of M/# measured from a single hologram.

Heterogenous Catalysis Mediated by Plasmon Heating

We introduce a new method for performing and miniaturizing many types of heterogeneous catalysis involving nanoparticles. The method makes use of the plasmon resonance present in nanoscale metal catalysts to provide the necessary heat of reaction when illuminated with a low-power laser. We demonstrate our approach by reforming a flowing, liquid mixture of ethanol and water over gold nanoparticle catalysts in a microfluidic channel. Plasmon heating of the nanoparticles provides not only the heat of reaction but the means to generate both water and ethanol vapor locally over the catalysts, which in turn allows the chip and the fluid lines to remain at room temperature. The measured products of the reaction, CO(2), CO, and H(2), are consistent with catalytic steam reforming of ethanol. The approach, which we refer to as plasmon-assisted catalysis, is general and can be used with a variety of endothermic catalytic processes involving nanoparticles.