Electronic noses combine an array of cross-selective gas sensors with a pattern recognition engine to identify odors. Pattern recognition of multivariate gas sensor response is usually performed using existing statistical and chemometric techniques. An alternative solution involves developing novel algorithms inspired by information processing in the biological olfactory system. The objective of this dissertation is to develop a neuromorphic architecture for pattern recognition for a chemosensor array inspired by key signal processing mechanisms in the olfactory system. Our approach can be summarized as follows. First, a high-dimensional odor signal is generated from a chemical sensor array. Three approaches have been proposed to generate this combinatorial and high dimensional odor signal: temperature-modulation of a metal-oxide chemoresistor, a large population of optical microbead sensors, and infrared spectroscopy. The resulting high-dimensional odor signals are subject to dimensionality reduction using a self-organizing model of chemotopic convergence. This convergence transforms the initial combinatorial high-dimensional code into an organized spatial pattern (i.e., an odor image), which decouples odor identity from intensity. Two lateral inhibitory circuits subsequently process the highly overlapping odor images obtained after convergence. The first shunting lateral inhibition circuits perform gain control enabling identification of the odorant across a wide range of concentration. This shunting lateral inhibition is followed by an additive lateral inhibition circuit with center-surround connections. These circuits improve contrast between odor images leading to more sparse and orthogonal patterns than the one available at the input. The sharpened odor image is stored in a neurodynamic model of a cortex. Finally, anti-Hebbian/ Hebbian inhibitory feedback from the cortical circuits to the contrast enhancement circuits performs mixture segmentation and weaker odor/background suppression, respectively. We validate the models using experimental datasets and show our results are consistent with recent neurobiological findings.

No description available.

Bi-annual report on the activities at the ROssendorf BeamLine (ROBL) at the ESRF in Grenoble. The report contains selected contributions on actual research topics, a list of all scheduled experiments, and short experimental reports.

Thesis (MTech. degree in Chemistry)--Tshwane University of Technology, 2010.

published_or_final_version / abstract / toc / Chemistry / Master / Master of Philosophy

HudBay Minerals (formerly the Hudson Bay Mining and Smelting Co., Limited) has operated a Zn and Cu processing facility in Flin Flon, MB since the 1930’s. Located in the Boreal Shield, the area surrounding the mine complex has been severely impacted by both natural (forest fires) and the anthropogenic disturbance, which has adversely affected recovery of the local forest ecosystem. Zinc is one of the most prevalent smelter-derived metals in the soils and has been identified as a key factor limiting natural revegetation of the landscape. Because metal toxicity is related more to speciation than to total concentration, Zn speciation in soils from the impacted landscape was characterized using X-ray absorption fine structure, X-ray fluorescence mapping and µ-X-ray absorption near edge structure. Beginning with speciation at a micro-scale and transitioning to bulk speciation was able to determine Zn speciation and link it to two distinct landform characteristics: (1) soils stabilized by metal tolerant grass species—in which secondary adsorption species of Zn (i.e., sorbed to Mn and Si oxides, and as outer-sphere adsorbed Zn) were found to be more abundant; and (2) eroded, sparsely vegetated soils in mid to upper slope positions that were dominated almost entirely by smelter derived Zn minerals, specifically Franklinite (ZnFe2O4). The long-term effect of liming on pH and Zn speciation was examined using field sites limed by a community led organization over a ten year period. Upon liming to a pH of 4 to 4.5, the eroded, sparsely vegetated soils where found to form a Zn-Al-Hydroxy Interlayer Material (HIM) co-precipitate, reducing the phytotoxicity of both Zn and Al and allowed for boreal forest vegetation to recovery quickly in these areas. The grass stabilized soils experienced a steady pH increase, as compared to a sporadic pH increase in the heavily eroded soils, as the buffering capacity was overcome allowing for a transition between multiple adsorption species based upon the point of zero charge of reactive soil elements. Ultimately reaching a near neutral pH after ten years, this allowed for the formation of stable Zn-Al-layered double hydroxide (LDH) soil precipitates and significantly reduced concentrations of plant available Zn.

High-pressure homogenization using a new flat valve homogenizer in combination with enzymatic digestion with a crude protease was investigated as a means of releasing Se compounds from zoological and botanical matrices prior to slurry introduction GF-AAS. Timed trials with four zoological certified reference materials (CRMs), three botanical reference materials (RMs), and a food crop indicated that Se release was quantitative after homogenization or became quantitative within 1 h of digestion at 60°C. / The same technique was employed on five animal feed samples. / A new model of homogenizer equipped with ceramic homogenizing valve was evaluated in terms of analyte metal contamination levels within the final sample dispersion. / Five animal feed samples and four wood pulp samples, were investigated for their content of Cu, Fe and Mn using high-pressure homogenization as the sample preparation technique prior to GF-AAS. Dispersions of dried animal feeds were sub-sampled reliably after 7 days of storage at 4°C. Trials on pulp samples indicated that pulps could be sub-sampled reliably after 1 day of storage. (Abstract shortened by UMI.)

X-ray absorption is not only element specific, but atom specific: two atoms of the same element in different states or in different neighbourhoods will have slightly different absorption characteristics. These energy dependent atomic form factors are carried over to the diffraction intensities. The atomic form factors are sensitive not only to the the energy of the X-ray but also the diffraction criteria; providing individual local physical data at different ratios in various diffractions. This process is referred to as site selectivity, it is unique to Diffraction Spectroscopy, and is achieved only when the sample is in crystal form. Through this work, a technique has been devised to site-separate two atoms of iron from within a protein, that builds on prior small unit cell Diffraction Anomalous Fine Structure experiments and harnesses the collection and processing software commonly used in large unit cell crystallography. A technique (dev + PCA) has been developed to retrieve the small signals from individual atom-labels out of the large and noisy background of real diffraction taken across a spectrum. The intensity of the diffractions are calculated by integrating over multiple images, profiling spots, merging datasets, and scaling across the whole spectrum. This thesis explores how Diffraction Spectroscopy can be used effectively on large unit cells, namely those of proteins. Site-selective absorption experiments were conducted on large unit cell crystals at a 3rd generation beamline, exclusively using existing equipment. The spectra generated were limited in scope but are an adequate proof of concept.

Diluted magnetic semiconductor oxides (DMSOs) have received great attention recently due to their outstanding applications in optoelectronic and spintronic devices. Ever since the initial observation of ferromagnetism at room temperature in cobalt-doped titania, extensive effort is concentrated on preparation of transition metal doped wide band gap semiconductors, especially Mn- doped ZnO. Compared to Mn-doped ZnO, magnetic interactions in SnO! and In!O! semiconductors have been underexplored. SnO! and In!O! semiconductors have many applications, owing to their high charge carrier density and mobility as well as high optical transparency. Investigation on electronic structure changes induced by dopants during the synthesis procedure can effectively influence magnetic interactions between charge carriers. In this work, a combination of structural and spectroscopic methods was used to probe as-synthesized SnO! and In!O! nanocrystals doped with Mn!! and Mn!! as precursors. X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectroscopy are powerful techniques to explore formal oxidation state of manganese dopant, electronic environment, number of nearest neighbors around the absorbent, and bond lengths to the neighboring atoms. Analysis reveals the presence of multiple oxidation states in the doped nanocrystals, and establishes a relation between !"!! ratio and expansion or contraction of lattice parameters. !"!! Although doping semiconductors are crucial for manipulating the functional properties, the influence of dopants on nanocrystals structure is not well understood. Nanocrystalline films prepared from colloidal Mn-doped SnO! and In!O! nanocrystals through spin coating process exhibit ferromagnetic behavior in temperatures ranging from 5 K to 300 K. Magnetic transformation from paramagnetic in free-standing Mn-doped nanocrystals to strong ferromagnetic ordering in nanocrystalline films is attributed to the formation of extended structural defects, e.g., oxygen vacancies at the nanocrystals interface. Magnetic circular dichroism (MCD) studies clearly show that Mn!! occupies different symmetry sites in indium oxide, when bixbyite and rhombohedral In!O! nanocrystals (NCs) are compared.

No description available.