<p>In this thesis, we consider the log-gamma distribution and discuss some of its properties. We then study the order statistics from this distribution. We derive the explicit expressions for the means, variances and covariances of order statistics from the distribution for various choices of the shape parameter.</p> <p>Next, we discuss the following methods of estimation of the unknown location and scale parameters (i) bext linear unbiased estimation based on Type II censored samples, (ii) best linear unbiased estimation based on optimal selected order statistics, (iii) maximum likelihood estimation based on Type II censored samples, (iv) approximate maximum likelihood estimation based on Type II censored samples. We illustrate these estimation procedures through a real-life data set by Lieblein and Zelen (1956).</p> <p>We also study the construction of the confidence intervals of these parameters. Both conditional and unconditional approaches are discussed and comparison between these two approaches is made. We also discuss the determination of the tolerance limits and confidence limits for the reliability.</p> <p>After discussing the estimation methods of the location and scale parameters of the log-gamma distribution, we study the estimation method of shape parameter under Type II censoring. We derive the log-likelihood function of the shape parameter and its derivative and hence Newton-Raphson algorithm can be used to obtain the maximum likelihood estimate of the parameter. Asymptotic Fisher information matrix of all three parameters is also derived and so are the asymptotic variances and covariances of the maximum likelihood estimators.</p> <p>Finally, we study the relation between the log-gamma distribution and record value theory. A bivariate log-gamma model is proposed because of this relation. We also study the record values which come from populations other than log-gamma distribution. Statistical inference based on the record values is also studied.</p> / Doctor of Philosophy (PhD)

<p>The hyperfine splitting (hfs) of deuterium shows a substantial deviation from the prediction based on Fermi's formula. This reflects the effect of the deuteron structure which was not taken account of in Fermi's formula. Following Bohr's original work on the deuteron structure effect, several calculations were done prior to the 1960's, using nucleon-nucleon (NN) potentials which are from today's standard, rather primitive. In this thesis we reexamine this problem by using several modern realistic NN potentials which reproduce the deuteron properties and the NN scattering data very well. In addition to the correction of the type which was examined by Bohr and Low, we examine a number of other corrections. The Bohr-Low correction, which is the most important one, turns out to be remarkably insensitive to the choice of the potential, and this correction significantly over-estimates the experimentally observed anomaly. However, the correction arising through the angular momentum dependent terms in the NN potential is sensitive to the choice of the potential. When this effect is included the theory and experiment can be reconciled for some of the potentials. In this sense the long standing anomaly of the deuterium hfs can be explained.</p>

<p>The seasonal response of net photosynthesis to temperature, light and moisture has been examined in two populations of the lichen Peltigera rufescens. The sub-arctic population had a temperature optimum of 35°C and no seasonal changed in the photosynthetic response were observed. In contrast, the gas exchange rate did change on a seasonal basis in the second population studied (from a temperate climatic zone). In this population, the temperature optimum for the net photosynthesis shifted from 25°C in the winder to 35°C in the summer. This seasonal change can be induced in the laboratory storing the lichen under appropriate temperature and photoperiod conditions. This change was shown, with the use of photosynthetic-illumination curves, to be restricted to alterations in the light saturated rates of net photosynthesis only, indicating that the acclimation process resulted from a change in the dark reactions of photosynthesis. This hypothesis was further investigated with studies on an important rate controlling enzyme of the Calvin cycle, fructose-1, 6-bisohosphatase. This enzyme was purified from summer and winter collections of the temperate population of P. rufescens and the total activity, thermal stability, substrate affinity and activation energy was compared in these isolates. It was shown that the temperature response of the Km for this enzyme closely paralleled the net photosynthetic response and that the activation energy was lower in the summer form of the enzyme. These kinetic differences largely explain the seasonal change in net photosynthetic capacity observed in the temperate population and support the enzymatic theory of acclimation suggested by the photosynthetic-illumination curves.</p> / Doctor of Philosophy (PhD)

<p>RNA blot analysis, cell-free translation and protein immunoblot analyses were used to characterize mRNA purified from pituitary glands of rapidly growing rainbow trout. Cell-free translation products of total pituitary RNA, analyzed by direct immunoprecipitation or protein immunoblot analysis with an antiserum to chum salmon GH, indicated that the rainbow trout pre-GH is a polypeptide of 25 kDa. RNA blot analysis competitor, revealed the presence of at least four size classes of pituitary-specific RNA sequences. The abundant class of mRNA had a size immunoblot analysis, it was found that the GH mRNA size class. Furthermore, the size of the GH mRNA of rainbow trout is similar to that of mammalian GH mRNA.</p> <p>The immunochemical reactivity of rainbow trout GH with antiserum to other GH was also determined. Results of this study indicate that antisera to mammalian GH crossreact poorly with rainbow trout GH. This observation correlates with the inability of any of the mammalian GH nucleic acid probe sequences to hybridize with a specific mRNA in the pituitary gland of rainbow trout. On the other hand, antiserum to chum salmon GH shows good crossreactivity with rainbow trout GH present in cell-free translation products programmed by rainbow trout pituitary RNA or in extracts of rainbow trout pituitary glands. Several attempts were made to construct a cDNA library using total pituitary RNA of rapidly growing rainbow trout. The early attempts were unsuccessful and reasons for the failures are discussed. Subsequently, a simplified method for the synthesis and cloning of cDNA was developed. A library containing approximately 30,000 recombinant clones was obtained following this procedure. A simple immuno-screening procedure using an antiserum to chum salmon GH as probe was also developed and serveral recombinant clones carrying the complimentary DNA sequence of rainbow trout GH mRNA were isolated. One clone, designated pAF51, was found to program the synthesis of an immunoreactive polypeptide of 24kDa and contains a cDNA insert of approximately 900bp. The cDNA sequence in pAF51 was determined by direct super-coiled plasmid sequencing. The cloned sequence encode, a hybrid polypeptide that includes the first 9 amino acid residues of the E. coli B-galactosidase a portion of the predicted signal peptide of the rainbow trout pre-GH and the entire sequence of the mature GH polypeptide. The primary structure of the mature GH polypeptide predicted from the rainbow trout GH cDNA sequence shows complete homology with the chum salmon GH and weak homology to the primary structures of the mammalian GH. The tertiary structure of the rainbow trout GH was also compared with other GH polypeptides. The pairwise comparisons of the hydropathy profiles of bovine, human, rat and rainbow trout GH polypeptides indicate that regions of similarity exist between the rainbow trout GH and the mammalian GH. In particular, there are two major regions of similarity found near the amino terminal and at the carboxyl terminal region. These regions correspond to hydrophilic domains of the GH molecules. The significance of these domains is discussed.</p> <p>The biological activity of the rainbow trout GH synthesized in bacteria was tested in vivo. The chimeric hormone was partially purified from extracts of E. coli harboring plasmid pAF51. The hormone was administered to intact fingerling and yearling rainbow trout by intra-peritoneal injection, and to fry by the dip methoid. The results from the in vivo test indicate that the chimeric GH is biologically active with respect to its growth promoting activity; a marked enhancement of growth was seen in GH-treated fish. A dose of hormone as low as 0.2 ug/g of body weight when administered by injection of 50 ug/l GH when administered by dipping is sufficient to promote enchanced growth. However, high dosages used in this study consistently exhibited the least induction of growth, althought significantly higher than the controls. Certain morphological and behavioral effects were also observed. Analysis of compositiion of muscle tissue obtained from GH-treated and untreated fish indicated no detectable difference in tissue quality.</p> / Doctor of Philosophy (PhD)

<p>This thesis describes the phenomenon of chemically induced grain boundary migration, a process in which a grain boundary migrates in response to a chemical driving force. The migration characteristics from various binary alloy systems are documented, especially in the aluminum-zinc system where extensive measurements have been made in both static and dynamic situations. The dynamics of chemically induced grain boundary migration are also studied in conjunction with the growth of a second phase, a situation referred to as discontinuous precipitation. It is suggested that the force which is responsible for moving the boundary reveals different facets of itself for different cases. The facets of itself for different cases. The facets are extremes in the chemical force levels; one is at a low level and is suggested to be derived from coherency strains when solute penetration into the bulk is significant. The other force is a high level force and which originates from large concentration differences which exist across the boundary at low volume penetration levels. It is uncertain whether the transition between the two chemical driving force levels is continuous or discontinuous since the transition ranges are small. Elastic anisotropy is suggested to be responsible for the initiation of migration due to an unbalanced coherent strain energy present across the boundary. The migration process has been modeled macroscopically and computer simulated. At this time, current models of grain boundary structure are suggested to be inadequate to account for the coupling of a chemical force to grain boundary structural components.</p> / Doctor of Philosophy (PhD)

<p>A novel technique of using an in-situ ECR generated H-plasma to produce high quality surfaces on InP substrates, for growth of n- and p-type InP layers by GSMBE, has been investigated. The initial substrate surface quality determines the quality of the subsequently grown layers and therefore, the cleaning procedure is of critical importance. The standard approach entails a thermal desorption of a passivating oxide; however, this technique leaves carbon on the surface and, because oxides may vary in composition depending on the growth conditions, a consistent temperature for desorption may not be obtained. The desorption process is also dependent on the atmosphere in which it is carried out; i.e. whether an overpressure of P₂ or As₂ is used. Thermal desorption of oxides from InP requires the substrate to reach temperatures higher than normal GSMBE growth temperatures which can lead to substrate decomposition and, for regrowth applications, can alter dopant profiles and layer composition in ternary and quaternary layer growths. As an alternative, H-plasmas in separate vacuum chambers have been used to remove oxides from InP but this typically produces highly defective substrates due to a loss of phosphorus from the substrate. In this work the combination of an in-situ H-plasma with a stabilizing atmosphere of P₂ is used for the removal of oxides at temperatures equal to growth temperature and below.</p> <p>The mechanism involved in the thermal desorption of an oxide is first clarified and this procedure is compared with oxide removal by H-plasma etch in a phosphorus atmosphere. The ECR source can produce various plasma modes which have been thoroughly characterized. The effect of the different plasma conditions on clean lnP is determined. These modes have different properties which result in different oxide etch rates. Various modes are compared and the mechanism of oxide removal is documented. As an alternative to oxide growth S-passivation has recently received attention as a surface passivation technique. Application of this surface, with and without H-plasma cleaning, has been investigated. The above treatments and other wet chemical surface treatments are compared to determine the optimal surface cleaning technique.</p> <p>The results indicate that thermal desorptions are driven by reaction with phosphorus from the substrate and therefore require high temperatures to promote significant phosphorus evaporation from the substrate. A H-plasma etch of the oxide is driven by the presence of atomic H in the plasma and results in the formation of water. Oxide removal rates were determined at various temperatures from 250°C - 490°C. The plasma etch was found to remove carbon contamination from the surface whereas, thermal desorptions do not. Defect levels in the underlying lnP are sensitive to the plasma properties but with careful choice of plasma conditions defect states can be minimized. An optimal surface cleaning procedure has been developed which involves a UV-ozone treatment and H-plasma cleaning, resulting in interfaces free of electrically active defects in n-type material and a defect concentration of 8x10¹¹ cm⁻² in p-type material.</p> / Doctor of Philosophy (PhD)

In this work, an instrument for measuring pulsed streaming potentials was constructed and optimized for analytical and teaching applications. This thesis is divided in three chapters, the first one deals with the construction of the instrument, the second describes a microfluidic experiment designed for undergraduate and high school students using this instrument, and the third one shows an application of pulsed streaming potential measurements in the detection of heparin. Streaming potential is the electric field generated when a liquid is forced to flow by a pressure gradient through a channel or other stationary charged surfaces.1,2 These measurements were done in microfluidic channels built with commodity plastics such polycarbonate (PC) and cyclic olefin copolymer (COC). Microfluidics studies the changing behaviors of fluids within small volumes, (nL, pL, fL), or small sizes, (channel size is about 100 nanometers to several hundred micrometers).3 With low level of complexity in instrumentation, low cost, and easy way to implement, the system is ideally suited as a teaching instrument in high 2 school and undergraduate labs. By creating simple experiments with suitable processing time, our goal is to introduce to students several fundamental concepts related to ionic solutions, electrochemical potentials, and charged surfaces. By doing the experiments, students can improve their analytical skills, and problem solving skills. They can learn many useful techniques, such as measuring pH, measuring conductivity, and calculating zeta potential. For these experiments, Polycarbonate (PC) is chosen as microfluidic platform because it is commercially available and the cost is low enough for a school budget. PC microfluidic channels are modified by different charged species, which are the anionic poly (sodium 4-styrenesulfonate) (PSS), the cationic poly(allylamine) hydrochloride (PAH), and bovine serum albumin (BSA). Since the relative polarity of streaming potential is determined by the surface charge, the signal detected is the reverse of streaming potential with different charged modified surfaces. With the same strategy, heparin is detected by real time monitoring adsorption on COC and PC microfluidic channels modified by protamine. The results on the two kinds of channels are compared. For COC channels, linear correlation of initial adsorption rates is found in the range between 0.00074 units/ml and 0.050 units/ml. For PC channels it is between 0.00074 units/mL and 0.074 units/mL. Streaming potential measurements have been useful for determining the charge of such surfaces as capillaries, 1 membranes, 4 and other porous materials.5 There has been no work done using pulsed streaming potential measurements for sensing purposes in microfluidic channels. With our sensing device, no referent electrode is needed since the signal acquisition is made using pulsed flow, so drifting of the measure voltages can be avoided. In addition, no such fluorescent, electrochemical, or radioactive labeling is required for detection.

A fundamental understanding of the relationship between structure and activity of neurotransmitters in the human brain is of vital importance for the design targeted drugs. Using density functional theory and hybrid exchange correlation energy functionals we have studied the structure-activity-relationships of some important neurotransmitters and selected drugs by calculating their absolute hardness (η) and absolute electronegativity (χ). A plot of the η- χ diagram allowed us to assign them into three distinct groups, namely, (i) Acetylcholine analogs (positively charged structure), (ii) GABA analogs (zwitterionic structures) and (iii) monoamines. The results suggest that brain stem is chemically soft because of distribution of monoaminergic nerve pathways. Prefrontal cortex is also chemically soft due to secretion of dopamine from mesocortical dopaminergic nerve A10, whereas neocortex is chemically hard due to presence of zwitterionic neurotransmitters. Target drugs (agonists/antagonists) can also be predicted by comparing the η- χ diagram of neurotransmitters with those of the drugs.

Since the events of September 11, 2001, the federal government is focused on homeland security and the fight against terrorism. This thesis addresses the idea of terrorist groups smuggling nuclear weapons through the borders of the United States. Security screening decisions are analyzed within maritime and aviation domains using discrete optimization models, specifically knapsack problems. The focus of the maritime chapters involves a risk-based approach for prescreening intelligence classifications for primary and secondary screening decisions given limited budget and resources. Results reveal that screening decisions are dependent on prescreening classification and the efficacy of the screening technologies. The screening decisions in the aviation security chapter highlight different performance measures to quantify the effectiveness of covering flights with the intent of covering targets. Results reveal that given scarce resources, such as screening devices capacities and budget, flights and targets can be covered with minimal expense to the system.

For a graph $G$, let $\alpha (G)$ be the cardinality of a maximum independent set, let $\mu (G)$ be the cardinality of a maximum matching and let $\xi (G)$ be the number of vertices belonging to all maximum independent sets. Boros, Golumbic and Levit showed that in connected graphs where the independence number $\alpha (G)$ is greater than the matching number $\mu (G)$, $\xi (G) \geq 1 + \alpha(G) - \mu (G)$. For any graph $G$, we will show there is a distinguished induced subgraph $G[X]$ such that, under weaker assumptions, $\xi (G) \geq 1 + \alpha (G[X]) - \mu (G[X])$. Furthermore $1 + \alpha (G[X]) - \mu (G[X]) \geq 1 + \alpha (G) - \mu (G)$ and the difference between these bounds can be arbitrarily large. Lastly some results toward a characterization of graphs with equal independence and matching numbers is given.