The accuracy and precision of catalytic converter analysis using conventional analytical methodology such as fire assay, x-ray fluorescence, atomic absorption and ICP-AES are typically in the range of ±7-10% RSD. Due to the high cost of noble metals, methods of analysis with increased accuracy and precision are desired to evaluate the loading of noble metals onto converter bricks. The investigations described in this work have resulted in a better understanding of many of the inherent problems and have contributed new approaches for sample dissolution and analysis using array detector based Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES) and Inductively Coupled Plasma Mass Spectrometry (ICP-MS). These methods are shown to be accurate and precise for the analysis of Pt, Pd, and Rh in catalytic converters. Catalytic converters are difficult to dissolve by conventional acid methodology. While carius, tubes have previously been employed to dissolve small weights of sample, complete dissolution of increased amounts of sample, as needed for high precision ICP-AES analysis, has been hindered by the insufficient oxidation potential of the acids in the carius tube. In this work, the addition of ferric chloride is shown to increase the dissolving power of the carius tube method and specifically targets Pt, Pd and Rh for dissolution. Simultaneous collection of analyte wavelengths and simultaneous background correction, as performed with multichannel array detector ICP-AES instrumentation, have enhanced sensitivity and precision in catalytic converter analysis when compared to single channel instrumentation. The studies described within this dissertation demonstrate that flicker noise has been effectively eliminated through the use of multichannel array based ICP-AES instrumentation. With proper line selection and the use of the high-resolution system, Pt, Pd and Rh in catalytic converters can be analyzed with precision of 1-1.5%. ICP-AES accuracy has been confirmed through isotope dilution ICP-MS employing new methodology to avoid Zr isobaric interferences.

The research which will be presented on the combination of Raman spectroscopy with an ultrahigh vacuum (UHV) environment. The data show that UHV Raman spectroscopy is a useful analytical technique for modeling electrochemical interfaces. Raman spectroscopy and x-ray photoelectron spectroscopy (XPS) were utilized to study the surface chemistry of 1-butanol and thiophenol. These molecules were studied on coldly-deposited and annealed Ag films grown in the UHV environment. Surface Raman studies of molecules adsorbed on coldly-deposited Ag films are abundant, because these films support significant surface Raman enhancement. However, recent advances in instrumentation make studies at relatively unenhancing annealed Ag films possible. Roughness and surface enhancement factors for coldly-deposited and annealed Ag surfaces were investigated using thiophenol. In addition, a correction factor was determined which allows data acquired in the UHV environment to be directly compared to data acquired in the laboratory ambient. Butanol orientation was studied at these Ag surfaces using Raman surface selection rules. The orientation of butanol is dictated by the solid-vacuum interface and the unique surface morphology present at these surfaces. Br and Na ions were used to model the interaction of butanol with electrolyte species found in normal electrochemical systems. The coverages of Br and Na were calibrated using XPS. The orientation of butanol was determined for varying coverages of these ions and compared the bare Ag surfaces. My research represents the first stages of modeling the electrochemical double-layer in the UHV environment using Raman spectroscopy.

The most commonly used chromatographic and solid phase extraction sorbents are silica-based. Usually trifunctional silanes are used to modify silica sorbents because they can form polymer networks that provide increased stability as compared to sorbents prepared with monofunctional silanes. Unfortunately, the reactions of trifunctional silanes with a silica surface, and with each other, are difficult to control and the resulting polymerized bonded phase is very complex. Although these sorbents have been used extensively for various applications, studies of bonded phase chemistry continue to this day. The importance of such studies lies in the fact that an understanding of chemically modified silica sorbents will lead to the development of better chromatographic and solid phase extraction methods. This study focuses on the influence of silica topography on the structure and surface coverage of bonded phases, as well as the differences between hydrocarbon and ion exchange bonded phases. Solid-state NMR spectra, recorded for alkyl sorbents, showed that the porosity of the base silica determined the surface coverages of the different bonded phases. Large silanes and siloxane oligomers were excluded from micropores, limiting their access to a significant fraction of the silica surface area. These conclusions were made after analyzing the spectra of endcapped sorbents, which contained peaks that represented polymer and surface bound endcapping reagent. These experiments were supplemented with normal and reversed phase chromatographic data that indicated that shorter alkyl silanes were not dispersed evenly on the heterogeneous silica surface, as opposed to octyl and octadecyl silanes. In addition, bonded phases that contained excessive vertical polymerization were found to exhibit non-uniform retention behavior. Spectroscopic characterization of anion exchange sorbents supported the hypothesis that ionic silanes affect the structure of a bonded phase. It was discovered that the post-modification of cation exchange precursor sorbents also produced bonded phases without vertical polymerization. The uniformity of both anion and cation exchange sorbents was confirmed by studying the effects of endcapping. However, it was shown that an octyl thioacetate phase hindered bond breakage during postmodification oxidation. The resulting octyl sulfonic acid sorbent was shown to exhibit increased selectivity for analytes with both hydrophobic and ionic character.

A series of chlorinated low-molecular-weight alkanes and alkenes was transformed electrolytically at metal cathodes at potentials from -0.3 to -1.4V (vs. SHE). Products included nonchlorinated hydrocarbons and less chlorinated intermediates. Product distributions are highly dependent on cathode material and applied cathode potential. Kinetics was first-order in the concentration of the halogenated targets. The specific first-order rate constants are function of cathode potential, cathode material, solution characteristics, and reactant identify. When transformation kinetics was governed by polarization resistance, rate constants were correlated with degree of halogenation and standard reduction potential for the predominant transformation reaction (as indicated by product analysis). Log-transformed reaction rate constants for reduction of chlorinated alkanes, derived via experiments at the same cathode potential (E(c) = -1.0 or -1.2V vs. SHE), were linearly related to carbon-halogen bond dissociation energies. A physical model for the observed correlation was developed from transition-state theory. The chlorinated ethenes reacted much faster than predicted from bond enthalpy calculations, suggesting that alkenes are not transformed via the same mechanism as the chlorinated alkanes. Polarographic study demonstrated that the shift of E₁/₂ of CCl₄ reduction was correlated with water concentration in solvent-predominated mixtures. Successful interpretation of these findings with a physical model suggested that solvents involved the rate-determining step of CCl₄ electrolysis both kinetically and mechanistically. The capture of trichloromethyl radicals with a spin trap (PBN) in an electrochemical system provided direct evidence supporting the free radical mechanism in electrolytic reduction of CCl₄. Gas-phase reductions of chlorinated alkanes and alkenes were studied in a modified fuel cell. Reactor performance was a function of the metal catalyst amended to the reactor cathode, the reactor potential, cathode temperature, the target compound identity, the partial pressure of O₂(g) in the cathode chamber and the condition (time in service) of the cathode. Single-pass CCl₄ conversions could achieve 90 percent with a mean residence time for gases in the porous cathode much less than a second. Reactor performance deteriorated with the presence of oxygen and time in service. Conversion efficiency was restored, however, by temporarily eliminating the halogenated target(s) from the influent stream or by briefly reversing reactor polarity.

New methods of analyzing pharmaceuticals by high performance thin-layer chromatography and microplate imaging were investigated using array imaging technology. Both techniques provide high sample throughput over more traditional analytical techniques for the analysis of pharmaceuticals. HPTLC provides high sample throughput by performing separations in parallel using a planar stationary phase. Imaging the entire plate with a single exposure performs quantitation of the analytes in the chromatographic medium. Fluorescence and fluorescence quenching detection modes are presented by employing tetracyclines, famotidine, and several over the counter drugs as model compounds. Studies conducted include sensitivity, separation efficiency and reproducibility of the system. Microplate imaging allows for the quantitation of numerous analytes in parallel. In this technique, solutions containing the analyte of interest are deposited into numerous self-contained wells on microplates, also known as 96-wellplates or ELISA plates. Light is passed through the wells of the plate allowing for absorption, fluorescence, or fluorescence quenching. Imaging with an array detector enables the researcher to simultaneously quantitate each well in parallel. The reaction between primary amines, ampicillin and amoxicillin, and fluorescamine was investigated. Microplate imaging was also tested for quantitating analytes in the low UV region (254 nm).

A novel line of integrated-optic-based chemical sensors was developed. The sensors are based on modification of the optical cavity of a single-mode semiconductor distributed Bragg reflector (DBR) laser. A sensitive layer changes its refractive index in presence of a specific chemical, thus changing the effective refractive index of the section and the optical length of the cavity. This results in laser frequency shift measured either directly or by heterodyne detection using a reference laser as the second source. It is shown that DBR-laser-based sensors can achieve in principle a much higher sensitivity than passive sensors, such as Mach-Zehnder interferometers, due to the narrow linewidth of DBR lasers. The theory of DBR-laser-based sensors is described. It allows optimizing the sensitive section length and field confinement in the sensitive layer for the lowest detection limit. The optimum parameters depend on cavity losses and absorption of the sensitive material. Numerical modeling shows a wide acceptable range of sensitive section parameters for low-loss materials, while for higher-loss materials this range becomes much narrower. Narrow-linewidth DBR lasers are required for high sensitivity. In this respect, sol-gel waveguides with and without Bragg grating were incorporated in the DBR laser scheme. Single-mode operation of DBR lasers with sol-gel waveguide gratings was demonstrated for the first time, with 34-dB side mode suppression and a short-term linewidth of 150 to 500 kHz. A 3-section configuration with sol-gel waveguides and fiber grating showed 28-dB side mode suppression and a short-term linewidth of 600 kHz. Chemical sensing was performed with fiber grating, sol-gel waveguide grating, and 3-section DBR lasers. The first two types showed frequency shift of over 130 MHz in the presence of acetone vapors, and reversibility within experimental errors. The 3-section scheme showed significant dispersion of response and lack of reversibility due to parasitic reflections and instability of the setup. The effect of reflections from facets on performance of this design was examined and found to reduce the maximum sensitivity.

PIXE--Proton Induced X-ray Emission is a novel and powerful tool for non-destructive multi-element analysis with excellent sensitivity. This technique has been applied to several biological and environmental studies at the University of Arizona. The first application reported here is the study of carrot tissue samples, obtained from plants irrigated with uranium, molybdenum, selenium or manganese contaminated water. The carrots were analyzed with PIXE for the concentrations of these elements taken up by the plants from water. The standard addition method along with an added internal standard was utilized for the study of thick targets made from carrot powder. ICP was also used to analyze the same samples for purposes of comparison. Excellent agreement was found between the PIXE and ICP results. The second study with the PIXE technique was to measure the uptake of mercury by the kidney. Mercury tubule suspensions from the rabbit kidney were incubated with various mercury compounds and enzymes and analyzed with PIXE for their mercury content. Mercury was successfully fixed by exposing the tubule or tissue samples to hydrogen sulfide to convert all forms of mercury to mercury(II) sulfide. Saturated and unsaturated chlorinated aliphatic hydrocarbons are often found as contaminants in ground water. A palladized-iron bimetallic system can be used to completely dechlorinate these compounds. The third application of PIXE was to study the surface of the bimetallic system before and after dechlorination and acid regeneration. There were no significant changes in the surface composition before and after dechlorination. A dilute hydrochloric or oxalic acid solution could be used to regenerate the palladized iron surface.

Initial Raman spectra of 1-alkanethiol self-assembled monolayers (SAMs) at mechanically polished (MP) polycrystalline Ag surfaces could not be accurately interpreted for alkanethiol conformational order, because the spectra indicated these monolayers were contaminated. From XPS and Raman spectra of the unmodified MP Ag surfaces, the primary contaminants were identified as graphitic carbon and alkyl hydrocarbons. As determined by XPS, mechanical polishing procedures were adopted which reduced the quantity of these contaminants; nonetheless, significant contaminant band intensity continued to be observed in Raman spectra of these alkanethiol SAMs. The contaminant was more accurately identified as a polyaromatic hydrocarbon (PAH) and its molecular identity was suggested as phenanthrene (phen). Attempts at its removal from the unmodified MP Ag surface by solvent dissolution and electrochemical reductive desorption were unsuccessful. However, this contaminant study was significant, because it initiated efforts to further reduce the quantity of this contaminant at the MP Ag surface. XPS and Raman spectra of chemically polished (CP) polycrystalline (poly) Ag surfaces indicated them to contain substantially less carbon contamination than the MP Ag surfaces. Thus, Raman spectra of short-chain alkanethiol SAMs at CP Ag (poly) and (chemically polished) Ag (111) were interpreted for alkanethiol conformational order. To better understand the signal intensities from CP Ag (poly), Ag (111), and MP Ag, surface enhancement factors (SEFs) at these surface types were quantified and compared to those measured for electrochemically-roughened (ORC) Ag, coldly-deposited Ag (Cold Ag), "thick" room temperature (RT)-deposited Ag, and MP Au. These SEFs were determined by reference of monolayer signal intensities at these surfaces to that from the unenhancing MP Pt surface (i.e., SEF of 1). Relative surface Raman sensitivities were assessed with use of these SEFs, and limits of detection (LODs) were calculated from the normalized S/N values in these surface spectra. Since the single spectrograph and Triplemate were both available for surface and normal Raman spectral acquisition, their relative performances were characterized to determine the advantages/disadvantages of each spectrograph. Specifically, the S/N and S/B values in spectra acquired on both spectrographs were quantified and compared.

Saturated and unsaturated chlorinated aliphatic hydrocarbons containing one to three carbons are commonly used as industrial solvents. These solvents are often found as contaminants in groundwater through improper disposal and many of them are carcinogenic. Remediation of groundwater containing these chlorinated compounds is of obvious importance. Methods such as air stripping and carbon adsorption involve the physical transfer of these contaminants onto activated carbon, and that does not diminish their toxicity. Other methods such as U.V. irradiation in the presence of ozone, hydrogen peroxide, or TiO₂, are not cost effective for large volume contamination problems such as polluted lakes, rivers and streams. Elemental iron has been shown to dechlorinate low molecular weight chlorinated aliphatic compounds but the observed end products were still partially chlorinated. We have established that the bimetallic system palladized iron (0.05% Pd) is preferable to elemental iron for the rapid and complete hydrodechlorination of 1- and 2-carbon chlorinated compounds with minimal loss of palladium. The major product observed from the hydrodechlorination of the 2-carbon chlorinated compounds was ethane. Methane was the major product observed from the hydrodechlorination of the 1-carbon chlorinated compounds. The palladized iron bimetallic system is also capable of reducing CFC's, nitro-organic compounds and inorganic species such as nitrate ions.

It was discovered in our laboratory that the bimetallic system palladium on iron (palladized-iron, Pd/Fe) rapidly and completely hydrodechlorinates aqueous solutions of ubiquitous chlorinated aliphatic and aromatic environmental contaminants such as TCE (trichloroethylene), chlorophenols and PCBs (polychlorobiphenyls) at room temperature and atmospheric pressure. When TCE comes in contact with the Pd/Fe surface, it forms ethane as the major gaseous product. Chlorophenols are converted to phenol and PCBs are converted to biphenyl. It was also discovered that through a series of redox reactions metallic iron completely reduces perchlorate to chloride. The objectives of this research are: to show that two iron materials can completely dechlorinate aqueous solutions of toxic chlorinated compounds; to begin to understand how these reactions occur, and; to understand the implications of oxidizing iron. In this research, batch and scaled-up systems were used for fundamental and practical studies of the Pd/Fe system. TCE, chlorophenols and PCBs were the model compounds for these studies. The batch reactions were performed in 1 OmL vials containing 2-5g Pd/Fe and 2-5mL of an aqueous solution of a reactant. The scaled-up reactions were performed in a closed-loop column apparatus with 700g of Pd/Fe and approximately 250mL of an aqueous solution of a reactant. The batch reactions were used to: show complete and rapid hydrodechlorination of the reactants to their products; demonstrate the performance of various types of iron; determine the sequential order of the removal of the chlorine substituents from chlorophenols and PCBs; observe the effect of an HCl-treated and an untreated iron surface prior to palladium deposition and the effects these surfaces have on the hydrodechlorination reactions of chlorophenols and PCBs; and determine what implications iron oxide build-up on the Pd/Fe surface may have on scaled-up Pd/Fe systems. The scaled-up column apparatus was used to determine the longevity of Pd/Fe surfaces and to find a suitable regeneration method. Because Pd/Fe was found to be a suitable reductant for chlorinated organic molecules, chlorine-containing ions such as ClO₄⁻ were also investigated. It was discovered that HCl-treated iron is a suitable reductant for the reduction of the perchlorate ion at room temperature and atmospheric pressure. Palladized-iron and zero-valent iron have been found to be suitable materials for the remediation of many ubiquitous environmental contammants. The hydrodechlorination of chlorinated compounds occurs completely and rapidly with Pd/Fe. Iron effectively reduces perchlorate ions to chloride ions. Both Pd/Fe and zero-valent iron surfaces are relatively easy to prepare, and can be used at room temperature and atmospheric pressure for the dechlorination reactions.