Site-directed mutagenesis and gene replacement were used to probe the pathway of electron transfer in nitrogenase by substituting single or groups of amino acid residues that, within the current view of component protein docking and nitrogenase catalysis, are likely to be involved in inter- or intra-molecular electron transfer. Intermolecular electron transfer was probed by substituting charged residues that, within the model for component protein docking proposed by Rees and Howard (Kim and Rees. 1992. Nature 360:553-60; Howard. 1993. In Molybdenum Enzymes, Cofactors and Model Systems, eds. EI Stiefel, D Coucouvanis, and WE Newton, pp.271-89. Washington DC: Am. Chem. Soc. 387 pp), are likely to be involved in electrostatic interactions that facilitate component protein association or dissociation. Intramolecular electron transfer was probed by substituting residues which are located in the polypeptide matrix that separates the P cluster and the iron-molybdenum cofactor based on the generally accepted view that the P cluster is an intermediate in electron transfer from the Fe protein to the iron-molybdenum cofactor at the substrate reduction site. The results of the biochemical characterization of a hybrid Azotobacter vinelandii-Clostridium pasteurianum Fe protein indicate that the region of the A. vinelandii Fe protein defined by residues 59 through 67 is involved in Fe protein-MoFe protein interaction. The rationale for construction of this hybrid Fe protein was based partially on the observation that the Fe protein from C. pasteurianum forms a tight-binding inactive complex with the MoFe protein from A. vinelandii. Detailed studies involving NaCl sensitivity and component protein ratio titrations suggest that this region may have a specific role in component protein dissociation. Further studies involving substitution of individual residues of the MoFe protein indicate that α-Asp¹⁶¹ is involved in component protein interaction. MoFe protein intramolecular electron transfer was probed by placing amino acid substitutions at β-Tyr⁹⁸, which is located directly between the P cluster and the iron-molybdenum cofactor. The results of the biochemical characterization of an altered MoFe with β-Tyr⁹⁸ substituted by His, support the generally accepted view that electron transfer from the Fe protein to the substrate reduction site involves the P cluster as an intermediate electron acceptor. It was also indicated that the P cluster may be able to accept more than one electron, which is consistent with the mechanism of P cluster reduction suggested by Rees (Rees DC, Chan MK, Kim J. 1993. Adv. Inorg. Chem. 40:89-119). / Ph. D.

This dissertation consists of two independent but related chapters on exchange rate theory and practice. The first chapter analyzes target zone exchange rate systems and provides an extension to the standard model of Krugman (1991). The standard model does not allow for imperfect credibility nor for interventions within the band, thence, it cannot explain the empirical evidence from the target zones of the European Monetary System and the Nordic countries. The model in this chapter allows for a realignment in the central parity of the exchange rate while the fundamentals follow a mean-reverting Brownian motion which models the intramarginal interventions of the central banks. We argue that, with these extensions, the model accounts well for the empirical facts. The second chapter develops a model of asymmetrical currency substitution. The friction for exchange rate determinacy is provided by costly spot market transactions. A stationary equilibrium exists if and only if the cost of spot market transactions is greater than either country's money growth rates, and country 1 money growth rate is lower than that of country 2, given that country 2 has a stronger currency. When the countries cooperate, there are two optimal money growth rates that maximize the steady state utility of the consumers. The optimal money growth rate strictly decreases in currency substitutability and the cost of income taxation. When they do not cooperate, there is a unique positive Nash money growth for country 1 which increases in response to increases in country 2 money growth if the transaction costs are not too high. / Ph. D.

This dissertation starts with a complete survey of the depolarization characteristics of axisymmetric and offset reflector antennas. Single and dual reflector configurations are considered. Cross polarization (XPOL) and beam squint are examined in detail. Although many of the phenomena are known, they are interpreted and explained in new ways and presented in a single coherent treatment. It is shown that single offset reflector antennas are limited in performance by high XPOL. A cost effective way to improve the performance of an existing offset prime-focus reflector is to add a subreflector, forming a dual offset reflector system with very low cross polarization (e.g., -35 dB). The motivation to use a specified main reflector often arises from a desire to use an existing mold which is usually very expensive. Within this context, procedures to upgrade existing reflector configurations are developed and presented. In addition, the influence of XPOL effects caused by low-cost, conventional feeds is analyzed in detail. A model for predicting the total system XPOL due to the reflectors and feed is discussed. Various techniques to reduce feed XPOL effects are introduced. Also, practical manufacturing constraints for large scale production are imposed on low-cost dual offset Gregorian reflector antennas. In particular, a design for adequate clearance between the bottom of the main reflector and feed axis is addressed. These constraints are not taken into account by other design procedures and are not addressed in the open literature. All innovative design algorithms developed in this dissertation were implemented as numerical codes referred to as DORA (Dual Offset Reflector Antenna Synthesis Package). DORA is a complete suite of codes for the synthesis of nonconventional, low-cost dual offset Gregorian reflector antennas with very low cross polarization. Several practical examples are discussed, including a performance assessment of the largest steerable reflector in the world, the Green Bank Radio Telescope located in Green Bank, West Virginia. Finally, an overview of the various analytical and numerical methods employed in the analysis of reflector antennas is presented in the appendices. The philosophical differences between the methods are highlighted. In particular, the physical optics approach and the Jacobi-Bessel series expansion method are described in detail. The combination of these two formulations results in one of the most accurate and efficient numerical tools in the analysis of reflector antennas. This is shown with the developed code PRAC, Parabolic Reflector Analysis Code. The effectiveness of PRAC is confirmed through extensive comparisons with measured data and results obtained from the literature and with the commercial code GRASP7. Most of the reflector antenna configurations obtained from the procedures developed in this dissertation are analyzed with PRAC and/or GRASP7. / Ph. D.

This work describes the development of an adaptive out-of-step relay, from the formulation of its concept to its field implementation and one year testing at the Florida- Georgia interface. This dissertation describes the theory of such a relay, its hardware configuration, the system as it was installed in the field, the major results and improvements obtained after the one year field test, and the adaptive features developed after the analysis of the collected data. Most of the adaptive concepts applied on this relay were used on field application for the first time and proved their value through the one year field test. Synchronized phasor measurements were used for the first time for relaying application, proving their ability to detect and analyze system disturbances through the measurement of angle differences between any two points in a system. It is shown that for a system that behaves primarily as a two-machine power system, the out-of-step relay could be enhanced and made more secure by applying the principle of equal area criterion. The main contribution of this dissertation is the use and application of old and new adaptive concepts as well as new technology to the solution of the out-of-step problem for a system that behaves like a two machine system. This work provides a solution for the basic stability problem with currently available technology and knowledge. In addition, the data collected during this research has been and will be of great help for those studying the power system stability problem and those developing new adaptive relaying techniques. / Ph. D.

Where Mn²⁺ _(aq) is found in water supplies, filter media may naturally develop surface coatings bearing MnO_x(s). These may absorb Mn²⁺ _(aq), and in the presence of oxidant, sorbed Mn²⁺* is oxidized to MnO_x(s), regenerating sorption capacity. The filter accomplishes Mn²⁺ _(aq) removal, a process called the "natural greensand effect". Characterization of naturally coated media showed variation in coating composition and structure. With thicknesses from 1 - 125 μm, primary coating constituents were Al and Mn, with incorporation of minor amounts of Fe, Cu, and Si and trace elements. "Growth ring" features in coating cross-section corresponding to compositional variation were characterized by SEM, electron microprobe, and energy-dispersive x-ray analysis (EDS). Media surface areas of 2 - 135 m² g⁻¹ land microporosity of 15 - 533 cm³ kg⁻¹ were linearly related to extractable Mn content. Diatom remains found in coatings suggest a key role for coating deposition in filtration phenomena. Atomic force microscopy found surface self-similarity over 10 nm - 10 μm. X-ray photoelectron spectroscopy (XPS) confirmed heterogeneous surface composition including C, Al, Si, and Fe. A method to rapidly deposit up to 4 mg g⁻¹ Mn on media was developed, employing sequential batch and recycle reactors. Mn(IV) was the surface species found by XPS analysis. The Freundlich isotherm described Mn²⁺ sorption on this and the naturally coated media; sorption capacity increased between pH 6.0 and 7.5, and was reduced by [Ca²⁺] = 60 mg L⁻¹. The global Mn²⁺ oxidation rates for all coated media at pH 7.5 were 0.008 - 0.11 mg Mn²⁺ g⁻¹ hr⁻¹: rates increased with flow and decreased with pH. A numerical process model for sorption and oxidation of Mn²⁺ _(aq) was calibrated with short bed absorber and differential reactor columns. The Freundlich isotherm, film transport, internal diffusion, and hydrodynamic dispersion were included, with sorption capacity apportioned into kinetically available and unavailable sites. The model performed well in calibration, predicting dynamic system response across a range of flow, pH, [Ca²⁺], and reactant levels. Model performance in validation was less satisfactory, probably due to experimental difficulties and the sensitivity of process performance on recent coating history and media regeneration status. / Ph. D.

An investigation into optical fiber sensing methods for monitoring metallic corrosion 1s presented. A series of sensing techniques are presented and explored, and two techniques are identified as possible candidates for a corrosion sensing system. The first is a technique based on the phenomenon of surface plasmon resonance, a technology useful in the sensing of changes in the index of refraction of a material in a localized area. The second involves the use of a metal-clad fiber sensing region where the standard core/cladding fiber geometry is modified to have a silica core and a metal cladding. A series of improvements to the current surface plasmon devices lead to the choice of a multimode cylindrical geometry for the sensor fiber. Under the correct conditions, energy in the TM modes in the fiber will transfer to a surface plasmon mode coupled to the surface of a metal film on the fiber. Monitoring the spectral output of the sensor provides information on the change in index of refraction of the surrounding material or the remaining thickness of the metal film applied to the fiber. The metal-clad sensor involves replacing the standard dielectric cladding with a metal in the sensing region. The loss of power through this region is directly related to the thickness of the metal cladding, which decreases with corrosion. A theoretical analysis is performed and compared with experimental results which demonstrate the feasibility of the sensing technique. Both the surface plasmon approach and the metal-clad approach are demonstrated as corrosion presence indicators, that is, they cannot currently resolve the level of corrosion, but just indicate when a certain amount of corrosion has occurred. The metal-clad technique is therefore identified as the most practical technique due to its relatively simple operation and easily understandable output response to corrosion. Future improvements are presented for the metal-clad sensor to help resolve the total amount of corrosion and a method is demonstrated for the repeatable fabrication of the metal-clad sensor. / Ph. D.

Numerical schemes are developed to study the impact response of composite structures to rigid (spherical masses) and flexible (uniform and nonuniform bars) projectiles. In the first phase of this study the impact response of imperfect ]aminated cylindrical panels to rigid projectiles is investigated. A 48 degree-of-freedom (DOF) shell finite element based on the classical laminated plate theory, which is capable of modeling geometric imperfections is used to model the shell. Linear and geometrically nonlinear transient responses are obtained using reduction methods based on the use of (i) natural modes and (ii) the Ritz vectors (also called Lanczos’ vectors) as the basis functions. The results obtained from these schemes are compared with those obtained using direct integration schemes, the Newmark-β and the Wilson-σ methods. The effect of number of reduced basis on the response is also studied. The impact loads are obtained using a modified Hertzian contact law by Tan and Sun. Effects of geometric imperfections and shell radius of curvature on the response are also studied. The present results are compared with those obtained experimentally at the NASA Langley Research Center. With one notable exception, a good agreement between the theoretical predictions and experimental results is observed. In the second phase, numerical schemes are developed to incorporate the effect of the projectile flexibility on the impact response of structures. A step by step approach, in which the impact responses of increasingly complex structures, namely, the axial bars, beams, and shear deformable plates subjected to flexible projectiles (uniform and nonuniform bars) are obtained, is used. The target axial bar is-:modeled using two degree-of-freedom axial bar elements. For the projectile, two different finite element models using, an axial bar element and a six-degree-of-freedom axisymmetric solid element with a triangular cross-section, are employed. The axisymmetric element (from the general purpose code MSC/NASTRAN) is used for those cases in which the target axial bar area is smaller than the projectile area and a two dimensional modelling of the projectile is needed. The impact response is obtained using an explicit algorithm based on the central difference scheme. In the algorithm developed, the target is assumed to be at rest and the projectile is assumed to be moving at a constant velocity, the impact velocity. At time t=0, the projectile hits the bar. At each time step, and as long as the two bars, are in contact, we assume that the two impacting bodies have the same velocity. For each time step, an iterative procedure is incorporated to predict the force that will enforce the velocity condition described previously. The results obtained from this approach are compared with other analytical and experimental results available in the literature for the impact response of a Hopkinson's bar. A good agreement is achieved. The algorithm developed here is next applied to study the impact response of beams and generally laminated, skew trapezoidal plates subjected to low velocity impact of a non-uniform linearly elastic composite projectile. The beam is modelled using two different approaches: a four degree-of-freedom beam element and an eight degree-of-freedom plane stress element. For the case of laminated plates, a Ritz method based approach developed by Kapania and Lovejoy is used. The present approach can be easily extended to study the nonlinear impact response of geometrically imperfect plates and shells. / Ph. D.

Using the state agencies that regulate wastewater discharges to surface waters, this quantitative study examined the validity of three theories about the effects of decentralization on organizational outcomes. It also used qualitative analysis to probe the internal adjustments to optimize outcomes. Theoretical advantages to decentralized organizations can be summarized as greater effectiveness or greater efficiency. Effectiveness here is the rate of compliance with legal and permit requirements for dischargers adjusted for enforcement actions. The model used consists of decentralization independent variables, exogenous independent variables (e.g., population, funding), efficiency dependent variables, and compliance (or effectiveness) dependent variables. Model calibration applied principal component analysis and multiple regression analysis to questionnaire and other data for FY93 from 39 states. Key intervening variables -- information costs, innovation, and strategic planning -- characteristic of the theoretical effects of decentralization received special attention. Case studies based on interviews and document studies in five states illustrate key points. The large survey sample (78%) of the 50 states reduced threats to external validity. A significant relationship to decentralization was found for major permit processing time and percentage of expired permits. When controlled for exogenous variables, no compliance common factor related to decentralization, but medium-term compliance unit-cost did. Time is essential to all compliance common factors. Innovation, information cost, and strategic planning were not found to be intervening variables. Highly decentralized states were found to be trending towards increased centralization, and vice versa. All states make internal adjustments to optimize their programs, and some adjustments are powerful enough to make highly dissimilar organizations converge; strong basin planning is one such adjustment. Neither decentralization nor centralization is automatically the best way to organize a complex system. Devolution, where transfer of functions goes with significant delegation of authority to autonomous local units, would probably give superior environmental, economic, and administrative outcomes over a uniform approach. / Ph. D.

"Ethanol oxidation over supported molybdenum oxide was studied by in situ laser Raman spectroscopy, deuterium isotopic substitution, together with steady-state and transient reactivity and kinetic measurements. The combination of techniques allows a complete description of the mechanism of ethanol oxidation on MoO₃ supported on SiO₂, Al₂O₃, and TiO₂, including the quantification of adsorbed ethoxide species. Very interestingly, on the SiO₂-supported catalysts two types of adsorbed ethoxide species were observed. One, associated with Mo=O bonds, was a true reactive intermediate, while the other, associated with Mo-O-Mo bonds, was an unreactive spectator on the surface." Isotopic substitution experiments established that the same mechanism with two stages occurred over MoO₃ on SiO₂, Al₂O₃ and TiO₂: an equilibrated adsorption step to form adsorbed ethoxide intermediates, followed by either of two rate-determining steps, α-hydrogen abstraction from the ethoxide intermediates to form acetaldehyde or β-hydrogen abstraction to form ethylene. The activation energy was found to be the same for the different supports at the similar loading levels. The link to reducibility and the existence of a common ethoxide intermediate indicated that the rate was controlled by a term in the preexponential factor. From the similarity in the isotope effect the controlling factor was deduced to be the electronic partition function associated with the density of electron-accepting levels in the molybdate-support complex. Analysis of the reaction isotherms showed that activation energy decreased with increasing coverage, suggesting a non-uniform surface. Consistent with this, the kinetics could be described by an extended treatment of Temkin’s theory of rates. A full derivation of the theory, which was expanded to accommodate two types of sites, is presented. Propylene oxidation was studied on MoO₃ on SiO₂, Al₂O₃ and TiO₂. Oxygen chemisorption and temperature programmed reduction (TPR) results indicated that the molybdenum oxide - support interaction decreased in the order: TiO₂ > Al₂O₃ > SiO₂. Temperature programmed surface reaction (TPSR) of adsorbed ethanol was used to characterize the acid-base properties of the catalysts, which played an important role in the formation of oxidation products. A comparative study of ethanol oxidation over MoO₃ on SiO₂, Al₂O₃, and TiO₂ by O₃/O₂ and O₂ was carried out in a flow reactor. The effect of ozone appeared below a transition temperature which differed according to the support (523 K on SiO₂, 448 K on Al₂O₃ and 443 K on TiO₂). The reactivity below the transition temperature followed the order SiO₂ > TiO₂ > Al₂O₃, while above it the reactivity followed the order TiO₂ > Al₂O₃ > SiO₂. It appeared that the ozone decomposition activity of the catalysts played an important role in ethanol oxidation by O₃/O₂. / Ph. D.

The primary purpose of the study was to estimate a retention model that predicted graduation and college grade point averages for scholarship football players. Several nontraditional factors, such as each player’s career maturity, their use of academic support services, and institutional characteristics were included in the model. Astin’s Involvement Theory, Spady’s Dropout Prediction Model, and Tinto’s Integration Theory served as anchors for the study. The assigned respondent from each athletic department of the nine institutions of the Atlantic Coast Conference provided the data for the study. Each respondent completed a survey that asked for information concerning the individual players and their institutions. A total of 216 football players were included in the study. The information provided by the institutions was used to quantify ten independent variables that were hypothesized to influence retention. Eight of these variables pertained to characteristics of the individual football players; the other two variables were specific to the institutions. Path analysis was used to estimate the retention model. The results indicated that of the four endogenous variables questioned, mature career attitude, use of academic support services, and atmosphere toward academics had a significant affect on retention. In the reduced model, mature career attitude was the strongest predictor of graduation, with use of academic support services second. Mature career attitude was also the strongest predictor of college grade point average. From the research findings several recommendations are made. It is suggested that athletic departments develop or identify an assessment test that measures career maturity and administer it to the football recruits. A specific counseling program should be established to assist those players who lack career maturity. It is also suggested that the Atlantic Coast Conference prepare a videofilm to address the career maturity issue. This video would include a discussion of the percentage of college players who successfully enter the professional ranks and interviews with college stars who were unsuccessful in pursuing a professional football career. Another recommendation is for each institution to require every player to actively participate in the academic support program. It is of utmost importance that the institutions send a strong message to the players that the main function of the university is for every student to successfully fulfill degree requirements. / Ph. D.