M. Kagan

Doped sol-gel glasses as pH sensors

Abstract A series of pH indicators were trapped in sol-gel porous glasses by polymerization of tetramethoxysilane in the presence of a surface active agent. The properties of these novel sensing materials including spectral shifts, shifts in the pH-sensing range, cycle repeatability, leachability, rates of response and isosbestic points are described. A prototype of a pH meter based on a pH-sensing glass was constructed.

Jet-images: computer vision inspired techniques for jet tagging

A bstractWe introduce a novel approach to jet tagging and classification through the use of techniques inspired by computer vision. Drawing parallels to the problem of facial recognition in images, we define a jet-image using calorimeter towers as the elements of the image and establish jet-image preprocessing methods. For the jet-image processing step, we develop a discriminant for classifying the jet-images derived using Fisher discriminant analysis. The effectiveness of the technique is shown within the context of identifying boosted hadronic W boson decays with respect to a background of quark- and gluoninitiated jets. Using Monte Carlo simulation, we demonstrate that the performance of this technique introduces additional discriminating power over other substructure approaches, and gives significant insight into the internal structure of jets.

New systems for pattern formation studies

Abstract Recent years have witnessed remarkable developments in the study of spatial pattern formation in reaction-diffusion systems. One of the most notable achievements has been the discovery of Turing patterns in the chlorite-iodide-malonic acid (CIMA) system. We have developed a mechanism for the chemistry of that system and from that mechanism have derived: (a) an understanding of how the Turing patterns arise; (b) a simple two-variable model, amenable to analytic study, that reproduces both the homogeneous behavior of the system and the Turing patterns; and (c) a general approach to the design of new systems that will show Turing patterns. We present experimental results of Turing patterns in the chlorine dioxide-iodine-malonic acid system, which our mechanistic analysis suggests lies at the heart of the CIMA Turing patterns. We also discuss, using the example of traveling waves in the Belousov-Zhabotinsky reaction, another experimental configuration, a sol-gel glass impregnated with key reagents, that shows great promise for the study of pattern formation and wave behavior.

Formation of dissipative spatial structures during photolysis of halogen compounds

der photischen Zone fiihrt zu einer teilweisen Freisetzung der Metalle. DaB die Korrelation zwischen Metallund Nfihrstoffkonzentration nicht so ausgeprggt ist wie in den Meeren [8], ist darauf zuriickzuftihren, dab der Metalleintrag an der Wasseroberflfiche, insbesondere aus der AtmosphOxe, recht grog ist. Der EinfluB des atmosph~rischen Eintrags ist fiir Pb besonders deutlich. Ferner besteht in den sedimentnahen Wasserschichten ein zusfitzliches partikul/ires F6rderband aus Eisenund Manganoxiden; diese entstehen vor allem im sp/iteren Teil der Stagnationsperiode durch die Oxidation des aus teilweise anoxischen Sedimenten riickdiffundierenden Fe(II) und Mn(II).

The evolution of chemical patterns in reactive liquids, driven by hydrodynamic instabilities

We summarize our activity in unveiling a very wide phenomenon: When a chemical reaction takes place at a liquid interface, spectacular patterns of product form (see Plate 1). The pattern formation phenomenon is general, and is observed in reactions between liquids separated by a membrane, in liquids subjected to gaseous reactants, and in photoreactive liquids. We have demonstrated the phenomenon on over 100 different reactions of all types, thus discovering what we believe to be one of the widest macroscopic pattern formation processes known to chemistry. As can be seen in the accompanying pictures, the richness, beauty, and variations in types of patterns can be breathtaking. Two important aspects of these patterns are noted: First, the patterns are true far-from-equilibrium structures, which are maintained only as long as reactants are available, or only as long as light energy is supplied to the system; and second, the chemical products that form the patterns are not precipitates, but are entirely soluble in the liquid in which they form. Thus, if the containers in which the patterns form are shaken or stirred, a homogeneous solution results. Our research of this phenomenon concentrated on three main aspects. The first one was phenomenological. Here we explored the scope and generality of the phenomenon, motivated both by the aesthetic appeal of the phenomenon, and by the puzzle of how is it that such a wide-scope, experimentally simple phenomenon, has by and large, escaped the attention of the scientific community.The second aspect was devoted to the understanding of the underlying general mechanism. Of the many mechanisms we analyzed and tested, some very complex, others quite trivial, the one that fits the majority of the physical and chemical observations is the following: By performing a reaction through a liquid interface, a concentration gradient of the product forms near the interface. We have shown that in many cases, these gradients lead to hydrodynamic instabilities, which then break nonlinearly into a pattern which onsets slow convections. In other words, we found that these patterns mark the route along which a chemical instability relaxes. The third aspect of our research was theoretical. Here we concentrated in depth on one of the reactions (the Fe(+2)/Fe(+3) photoredox reaction), determined all its important physical parameters, and modeled its behavior theoretically. Our model, which was based on the instability buildup described above, was solved numerically, and its results compared with computerized image analysis of the evolving patterns; very good agreement between theory and experiment, was obtained. (c) 1995 American Institute of Physics.

Geometric phases in dissipative systems.

It is shown that a phenomenon analogous to the geometric phase shifts of Berry and Hannay occurs for dissipative oscillatory systems and can be detected in numerical simulations of chemical oscillators. The approach herein to the theory of geometric phases begins with a study of simple first-order differential equations on the circle (circle dynamics). It is shown how more complicated systems exhibit geometric phases through reduction to a circle dynamics. In this way, the various manifestations of the phenomenon are seen from a single unified perspective. The results are illustrated in numerical experiments on several model systems ranging from analytically solvable, but contrived, to realistic models of chemical oscillators.

Spatial dissipative structures formed by spontaneous molecular aggregation at interfaces

Interfacial processes as well as formation of dissipative structures have been suggested to play a key role in early pre-biotic evolutionary stages, mainly due to the ability of such processes to induce aggregation and spatial structuring. In this context we would like to draw attention to our recent findings regarding a remarkably wide collection of interfacial chemical reactions which form dissipative spatial structures. Three types of interfacial processes were found to yield this phenomenon: photochemical oxidations at liquid/air and liquid/liquid interfaces; gas/solution reactions; and reactions at membrane surfaces. The phenomenon we describe is the first major example of a network of chemical reactions that develop into macroscopic far-from-equilibrium concentration patterns.

Spatial Structures Induced by Chemical Reactions at Interfaces: Survey of some Possible Models and Computerized Pattern Analysis

According to the theories of far-from-equilibrium thermodynamics three types of homogeneity breaking are possible: spatial, temporal, and spatio-temporal.1 The latter two are well known in purely chemical systems, e.g. bromate oscillators2 and wave oxidations3, both having their parallels in biochemistry, e.g. the glycolytic cycle4, and slime-mold aggregation5. However the most abundant class of biological structures, namely spatial structures, are relatively unexplored in chemistry6,7 (Liesegang precipitation phenomenon being an exception8). We present the following experimental results for a system that couples a chemical reaction to physical parameters to produce spontaneous pattern formation from initially homogeneous states. A complete model of the mechanism is still under investigation but it is clear at this preliminary stage that one is dealing with a system that is open to an analysis involving instabilities, perturbations and bifurcations. Some partial models are discussed below, some of which were also experimentally tested.

Image Processing, Computer Vision, and Deep Learning: new approaches to the analysis and physics interpretation of LHC events

This review introduces recent developments in the application of image processing, computer vision, and deep neural networks to the analysis and interpretation of particle collision events at the Large Hadron Collider (LHC). The link between LHC data analysis and computer vision techniques relies on the concept of jet-images, building on the notion of a particle physics detector as a digital camera and the particles it measures as images. We show that state-of-the-art image classification techniques based on deep neural network architectures significantly improve the identification of highly boosted electroweak particles with respect to existing methods. Furthermore, we introduce new methods to visualize and interpret the high level features learned by deep neural networks that provide discrimination beyond physics- derived variables, adding a new capability to understand physics and to design more powerful classification methods at the LHC.

Image Analysis in the Study of Dissipative Spatial Patterns

The rapidly growing science of image analysis (IA) by artificial intelligence methods has been applied to many diverse fields such as: biology, geology, earth scanning by satellites, photography, and a variety of military uses. In chemistry IA has been used mainly for rapid analysis of spectral output [1] in, for instance mass spectroscopy. Recently we have applied this powerful tool for the analysis of real spatial structures in chemistry. One example is in fractal surface analysis [2] and the other is in dissipative spatial patterns. In the latter case, the discovery of the general phenomenon of spatial structure formation by chemical reactions at liquid interfaces [3] has led to questions previously unasked by chemists, for instance, ‘How to measure the kinetic growth of a product not distributed evenly in space ?’; ‘What is the change in entropy as a pattern develops ?’; ‘How to qualitatively differentiate one pattern from another ?’.