Nuclear magnetic resonance experiments were performed on a series of selenium model compounds, tellurium model compounds, a tin model compound, a selenium metabolite that is an excretory product, and a selenoprotein. ¹H-{⁷⁷Se}, ¹H-{¹¹⁹Sn},¹H-{¹²⁵Te}, and ¹H-{¹²³Te} heteronuclear multiple quantum coherence experiments were performed for the first time on these compounds. In all cases the use of indirect detection substantially increased the sensitivity of observing these nuclei. Coupling constants between 9.4 and 54.2 Hz were successful on selenium compounds and coupling constants between 14.3 and 102.5 Hz were successful on model tellurium compounds. The increase in sensitivity for the observation of selenium compounds was 68 which is close to the theoretical value of 73 and for the observation of tellurium the increase in sensitivity was 46 which is close to the theoretical value of 50.7. These large gains in sensitivity allowed the detection of selenium present in trimethylselenonium iodide at levels below 1 mM. This could conceivably allow this methodology to be adopted as a non-invasive method for the detection of this selenium metabolite and as a means of measuring selenium toxicity levels in blood or urine. Indirect ¹H-{⁷⁷Se} HMQC experiments were also successful on protein A from the glycine reductase complex of Clostridium sticklandii. In addition to enhancing the sensitivity of detecting selenium in this protein, these indirect experiments greatly simplify the spectrum so that only protons that are scalar coupled to selenium are seen in the NMR spectrum. Work on the tin compound had the same aim but it involves using tin as a filter and then performing a NOESY experiment. In the appendices it is shown how selenium NMR was used to identify the presence of a selenite ester in a long chain fatty acid. A lanthanide shift experiment aided the assignment of both the major and minor diastereomer of the selenite ester. The design of a triple resonance box is shown. Finally, electrochemical oxidation data on selected dithiins, thiatellurins, selenothiins, dithiiranes, and thiophene derivatives are presented.

The factors which determine the growth mode and molecular architecture of vacuum deposited organic thin films on single crystalline substrates were investigated. Specifically, the relative importance of layer planes in the bulk structure, lattice matching between the overlayer and substrate, topographic direction by the substrate, and specific molecule-substrate interactions, in determining the growth mode were examined. The majority of the molecules studied here (ClAlPc, F₁₆ZnPc, PTCDA, C4-PTCDI, and C5-PTCDI) exhibited layer planes in their bulk structures, however, the molecular plane is coincident with the layer plane only for PTCDA and ClAlPc. ClAlPc and F₁₆ZnPc were found to adopt different flat-lying commensurate square lattices on the Cu(100) surface. In both cases, the flat-lying orientation of the molecules was dictated by specific molecule-substrate interactions, while the orientation of the lattice was dictated by lattice matching with the substrate. ClAlPc was also able to adopt an incommensurate centered rectangular lattice whose orientation was directed by alignment along step edges. Fluorescence investigation of submonolayer PTCDA and PTCDI films on alkali halide substrates demonstrated the great potential of fluorescence spectroscopy as a means of monitoring film growth. PTCDA was found to adopt a flat-lying orientation on NaCl, KCl, and KBr, while a flat-lying orientation of the PTCDI molecules was determined by the strength of the molecule-substrate interactions. From these measurements, the relative interaction strengths of the substrates were determined to be KCl > KBr > NaCl. IR dichroism showed that the expected growth along the layer planes was found only to occur for PTCDA, due to the coincidence of the layer and molecular planes. IR spectroscopy also revealed that a new polymorph of C5-PTCDI had been formed on these surfaces. These studies showed that the relative importance of the factors in determining the molecular architecture adopted within the first 1-2 MLE of a film are: (1) molecule-substrate interaction, (2) lattice matching, (3) topographic direction, (4) layer planes in the bulk structure. In addition the use of fluorescence spectroscopy to probe the evolution of vacuum deposited films was significantly advanced.

The research projects described in this Dissertation were designed to investigate the interfacial chemistry of alkylsilanes on metal and oxide surfaces, specifically Ag and silica, respectively. A variety of surface analysis tools, including FTIR and Raman spectroscopies, electrochemistry, ellipsometry, and x-ray photoelectron spectroscopy, have been applied to study important aspects of bonded alkylsilane structures. Two model silica surfaces were designed and fabricated as substrates for the attachment of alkylsilanes. The first model silica surface employs self-assembled monolayers of hydrolyzed and condensed (3-mercaptopropyl)trimethoxysilane (3MPT) on Ag surfaces. The Ag substrate is used to provide a small enhancement to the Raman scattering at the interface. Octadecyltrichlorosilane (OTS) and dimethylchlorooctadecylsilane (DOS) are covalently bonded to these surfaces. Monolayers of OTS on 3MPT-modified Ag surfaces are ordered, while submonolayers of DOS, the maximum coverage achievable, are disordered. Results obtained from this study demonstrate the importance of van der Waals interactions and siloxane cross-linking in promoting an ordered alkylsilane structure. The second model silica surface studied is based on thin silica films prepared through sol-gel technology. These are prepared by spin-coating prehydrolyzed solutions of tetratmethoxysilane (TMOS) onto 3MPT-modified Ag surfaces. These surfaces are designed to contribute to an understanding of the partitioning process associated with alkylsilane stationary phases in reversed-phase liquid chromatography (RPLC). OTS layers covalently attached to such surfaces were studied previously in this laboratory. In this Dissertation, the sol-gel methodology used is improved to enable silica films in the ultrathin (< 100 Å) regime to be fabricated. These substrates allow study of model stationary phases of DOS. Both FTIR and Raman spectroscopies indicate that the DOS alkyl chains on silica are disordered, consistent with previous notions about monomeric alkylsilane stationary phases in RPLC. Further characterization of these thin silica films reveals them to be non-porous, dielectric, and homogeneous. Their dielectric strengths are found to be equivalent to or better than those from device-quality thermally grown silica. This research expands application of the well-defined sol-gel technology to the fabrication of ultrathin silica films that may be useful as insulating layers in the microelectronics and semiconductor industries.

Two major lines of investigation, both involving collisions of ions with organic thin films on metal, are described in this dissertation. The first topic involves studies of low-energy (e.g., 10-250 eV) ion-surface interactions with Langmuir-Blodgett (L-B) films labeled at the outermost surface carbon atom, either by isotopes (deuterium or 13C) or by fluorine. The L-B films are prepared from the labeled fatty acids. The ion-surface collision results suggest that the outermost surface atoms/groups are the main determinant of energy transfer, electron transfer, and ion-surface reactions for polyatomic projectile ions (e.g., benzene, pyrazine). The results presented in this highly interdisciplinary area could be of interest to ion chemists, surface scientists, molecular physicists, physical chemists and others. In addition, the results promise a novel surface characterization technique using ion-surface interactions in the future. The second research topic involves studies that utilize surface-induced dissociation (SID) for mechanistic investigations of peptide fragmentation. Easily prepared self-assembled monolayers (SAMs) of alkanethiols or fluoroalkanethiols on gold are used in SID of peptides. A long-term goal associated with the work on peptide fragmentation is to provide improved and additional predictive rules of peptide dissociation for the computer-aided interpretation of MS/MS spectra in MS-based high throughput peptide/protein sequencing. For example, MS/MS results of fixed-charge derivatized peptides unequivocally demonstrate that in the absence of an available mobile proton, selective cleavages at the peptide bond immediately C-terminal to an Asp residue (Asp-Xxx), initiated by the Asp side chain acidic hydrogen, dominate the MS/MS spectra. SID on a series of dendrimers was also performed to investigate the effect of different charge states on the ion fragmentation. The dendrimers serves as model compounds that have a number of protonation sites with similar gas-phase basicities compared to peptides that have a more heterogeneous population of basic sites. In contrast to previously reported results for multiply protonated peptides of comparable size and charge states, no dependence of SID characteristic collision energy on the charge state of the dendrimers is observed. This supports the idea that it is a mobile proton available to the amide group moiety, instead of simply an additional positive charge, that promotes the lower energy cleavage of peptide bonds.

As the amount of data generated by today's advanced analytical instrumentation grows, the need for more intelligent and strategic manipulation of that data increases correspondingly. In the case of echelle ICP-AES employing array detectors, the large amount of the data available in a given experiment is often under-utilized. Efficient utilization of the vast information content available in these spectra holds promise for significant improvements in elemental analysis. Many samples, especially those that are completely unknown, require the expertise and time of a highly trained user in order to develop a robust and reliable method that will also be appropriate for the further analysis of similar samples. More intelligent use of the available data and implementation of multivariate chemometric techniques can afford improvement and even automatic generation of analytical methods for atomic emission spectroscopy. An intelligent multivariate calibration protocol has been developed, which is suitable for use with data from any of the state-of-the-art echelle-CID based atomic emission instruments, including those employing ICP, DCP, Arc, or spark sources. Protocols are based on custom approaches to direct and iterative classical least squares (CLS) fitting of signal vectors to stored calibration matrices. Subsequent to calibration, rapid semi-quantitation is made possible, allowing for more strategic choices of standards, more systematic selection of lines for quantitation, improved identification, better interference correction, and ultimately more precise and accurate quantitation and higher sample throughput. Practical demonstration is presented with both validation and unknown sample sets, using ICP-AES. The calibration protocol developed demonstrates the feasibility of a stored, stable, and correctable multivariate calibration matrix. State-of-the-art echelle-CID based ICP-AES instrumentation has been under-utilized as a method of simultaneous multielement-specific detection in chromatographic applications due to shortcomings in the acquisition and manipulation of data. New software to facilitate export, viewing, and calculations with time-resolved ICP-AES data has been developed. Applications are presented for the speciation of compounds of toxicological importance, including potential arsenic metabolites, a proposed biological model arsenic-selenium complex of glutathione, and a novel mercury selenium glutathione complex.

A novel instrument for simultaneous multielement graphite furnace atomic absorption is described. This instrument employs a continuum light source and an echelle-charge injection device array detector system to achieve both simultaneous analysis of multiple wavelengths and true simultaneous background correction. The charge injection device detector also allows continuous wavelength coverage from 400-190 nm. Multichannel GFAAS analysis improves not only analytical speed but enables the use of elements resistant to matrix effects to assist in the identification of matrix effects in other elements and confirmation of calibration accuracy. The research project described discusses not only basic instrument design but also practical application of the new design to current analytical problems. Analysis of Ag, Pb, Tl, Cr, and Mn in drinking water, and red table wine allowed not only evaluation and investigation of instrumental problems but also allowed investigation of the various chemical problems associated with simultaneous multielement determinations in general. Multielement analysis involves the use of compromise furnace conditions, and matrix effects can further complicate multielement determinations. The use of chemical modifiers such as a mixed Pd(NO₃)₂ and Mg(NO₃)₂ modifier provide a partial solution to some of the problems associated with multielement determinations. For the two applications, the analytes were split into two groups. One group analyzed Ag, Pb, and Tl and the other analyzed Cr and Mn. The modifier system was applied to the Ag, Pb, and Tl analysis. For drinking water, analytical results were obtained with errors and RSD's less than 10%. The modifier system was not effective when water solutions contained high levels of Na and Ca. For red wines, the modifier system was effective for Ag, Pb and Tl. However, maximum pretreatment temperatures were approximately 100-200°C lower than that of high purity water. No detectable levels of Ag or Tl were found in any of the wine samples analyzed. Trace levels of approximately 20 ppb Pb and Cr were detected in several red wines. Levels of Mn approached 1 ppm. Matrix effects were found to suppress the Pb signal and enhance the Tl signal. Sample recoveries for both elements typically ranged from 80-120%.

Formation and characterization of well-organized protein film assemblies are of high interest due to potential applications in biomolecular devices. The hypothesis that a macroscopically ordered protein film can be formed by site-directed, covalent binding of a protein to an appropriately derivatized surface formed the basis for the reported studies. The molecular architecture chosen to address this hypothesis consisted of yeast cytochrome c, a heme protein containing a unique binding site, immobilized on a surface derivatized with a silane self-assembled monolayer, (SAM). A combination of two techniques: total internal reflectance fluorescence (TIRF), to measure fluorescence anisotropy, and integrated optical waveguide-attenuated total reflectance (IOW-ATR) spectroscopy, to measure absorbance linear dichroism, was used to probe the macroscopic order of the heme groups in the film assemblies. Epifluorescence microscopy and absorbance (in an IOW-ATR geometry) were used to probe the nature of the surface-protein interactions and to determine relative protein affinities for different SAM-derivatized surfaces. The molecular orientation distribution for yeast cytochrome c immobilized on a 100% thiol-terminated SAM was 67° ± 39°. Partial protein removal was observed under a variety of rinse conditions, indicating that multiple protein-surface interactions may have contributed to the wide distribution value. Relative binding affinity constants and protein-surface interactions are compared for yeast cytochrome c and horse heart cytochrome c on 100% thiol-capped, 100% hydroxyl-capped, and mixed SAM-modified surfaces. These studies were also extended to include a variant of yeast cytochrome c, Thr8Cys/Cys102Thr. Similar adsorption and removal trends were observed for all the protein-SAM combinations. The adsorption isotherms indicated that at least two binding processes occur during formation of each protein film, high and low affinity binding. Removal studies indicated that the adsorption process is only partially reversible. It was concluded that employing a site-directed immobilization strategy does not necessarily produce a well-ordered protein film.

A foil mask spectrometer is described which allows for the simultaneous determination of position, energy, and intensity of monochromatic spots in a Laue diffraction pattern. The instrument may also be used for the quantitative determination of the intensity components corresponding to each energy contained in a spot with harmonic overlap. The spectrometer is uniquely suited to Laue diffraction applications and is demonstrated to be useful for determining unit cell dimensions and systematic absences. This dissertation discusses the characterization of a charge injection device camera system and provides a theoretical basis for selecting a charge transfer device detector. The principles, construction, design, and limiting equations for the foil mask spectrometer are described. Equations for limiting resolution are verified and the correspondence between predicted and observed energy is shown for a variety of crystal systems. The foil mask spectrometer is then used to verify the unit cell dimensions of six compounds and to observe systematic absences due to 2₁ screw axes and unit cell centering conditions. These crystals belonged to four different crystal systems including: cubic, orthorhombic, tetragonal, and monoclinic cells. The crystals had cell volumes from 179.4 ų to 4588.3 ų. Comparison of known and re-determined cells showed good agreement (ratio of known to measured cells = 0.987 ± 0.020). A single procedure was suitable for all unit cell determinations. Some of the crystals represent space groups containing systematic absences normally obscured by harmonic overlap when using the Laue method. These include absences due to 2₁ screw axes (h, k, or 1 = 2n + 1) and cell centering (h + k = 2n + 1). All systematic absences were identified using a combination of multiple linear regression with either stepwise elimination or stepwise inclusion and an F-test for assignment of systematic absence. The methods are discussed in detail and simulations are used to evaluate critical tolerances for future systems.

The application of planar integrated optical waveguide (IOW) technology to chemical sensor development results in very sensitive, fast-response devices. The additional dimension of selectivity that is possible in these devices, through the ability to select an analytical wavelength of light particular to only a specific interaction, is also a very attractive feature. The implementation of sol-gel materials processing in the sensor fabrication proves beneficial to the device performance in many ways. Sol-gels provide the high optical quality materials used in both the waveguiding layer and indicator film overlayers of the planar IOW chemical sensors. Since they are chemically prepared, their physical properties and chemical properties are easily tailored through chemical means or processing conditions. The fact that sol-gel bulks and films are prepared at room temperature allows the immobilization of heat labile species such as organic indicators or, in the case of biosensors, proteins or enzymes. The feasibility of the laminate planar IOW sensor geometry is demonstrated through the formation strongly light-absorbing complexes between the analytes and sol-gel immobilized indicators, detected by the attenuation of totally internally reflected light guided in the IOW (Part I). The construction of a gas-phase humidity sensor is shown in Chapter 3, along with a solution-phase sensor for aqueous isopropyl alcohol (IPA) in Chapter 4. An IOW-based humidity sensor utilizing a completely non-sol-gel based transducer layer is also reported (Chapter 5). The humidity sensors were found to exhibit limits of detection of below 1% RH (down to 50 ppm in nitrogen gas) with response and recovery times ranging from over one minute down to several seconds. The IPA sensor had response times under one minute, with an extrapolated detection limit of 0.02% v/v (160 ppm) IPA in water. This work also describes the spectroscopic properties of novel indicator compounds that are useful in chemical sensing (Part II). In particular, Chapter 5 describes the fabrication of unique starch/amylose-iodine composite films which demonstrate dramatic, humidity-sensitive colorimetric changes. Chapter 6 deals with the examination of the spectroscopic properties of a novel class of peripherally substituted porphyrazines that exhibit selective metal ion binding. Finally, Appendix C describes the use of a naphthalocyanine for the detection of tributylphosphate (TBP) in polar solution by visible absorbance spectroscopy.

The objective of this research was to measure orientation distributions in protein films as a function of immobilization chemistry to determine what conditions lead to ordered films. The hypothesis was that an ordered film could be achieved using a site-directed approach to immobilize a protein to a surface via a unique functional group. In order to accomplish this goal, a novel combination of absorbance linear dichroism performed in a integrated optical waveguide total internal reflectance geometry and total internal reflection fluorescence anisotropy (IOW-ATR+TIRF) was used to measure the orientation distribution of protein films in situ. The development of this combined technique included the synthesis of a mathematical theory to relate measured spectroscopic parameters to the orientation distribution. This was followed by testing on model molecular assemblies of Langmuir-Blodgett films doped with fluorescent amphiphiles with known orientations. Past the development phase, the IOW-ATR+TIRF technique was used to measure the first orientation distributions for protein films. This represents a significant advance in the study of protein film assemblies. The results of this research indicate that the hypothesis is correct and that ordered protein films can result from site-directed methodologies. This was demonstrated by the narrow orientation distributions of yeast cytochrome c covalently bound to phospholipid bilayers and biospecifically bound to streptavidin films. This research also indicates that of equal importance to a site-directed approach is the ability to tailor a surface to prevent unwanted non-specific adsorption interactions. This type of behavior was observed for horse heart cytochrome c adsorbed to bare hydrophilic glass surfaces which exhibited a broad orientation distribution. However, it should also be pointed out that adsorption processes can lead to ordered protein films as demonstrated by cytochrome c adsorbed to arachidic acid LB films. It can be broadly stated that ordered protein films result from situations where a single high energy binding mechanism immobilizes a protein to the surface and non-specific interactions are prevented. This conclusion is certainly not novel, but the fact it has been demonstrated to be true by the direct measurement of orientation distributions is revolutionary in the development of protein thin film devices.