The reaction of YbI2(THF)2 with diphenylmethyldipyrrolide leads to a complex reaction from which is obtained an octameric divalent Yb macrocycle (Diphenyldipyrromethanedi-yl-Yb)8 2.1 as a major product, but from which is also isolated a tetrameric cyclic Yb(II)-oxo complex [(diphenyldipyrrolyl-methanedi-yl)Yb]4[(K2)(THF) 6](mu-O)•2(THF) 2.2•2(THF) and a monomeric Yb(III) complex, Yb(diphenyldipyrrolylmethane-di-yl)3(K 3THF3) 2.3, as minor products. Complex 2.1 is analogous to the Sm(II) dipyrrolide octameric clusters obtained under an argon atmosphere, which are thought to represent the intermediate species in the formation of dinitrogen complexes. The reactions of SmI2(THF)2 and YbI2(THF) 2 with the alkali salts of 2,5-dimethylpyrrole, or the reaction of SMCl 3(THF)3 and YbCl3(TBF)3 with the same ligands followed by reduction with the appropriate alkali metals, led to the formation of divalent mono- and polynuclear complexes. (Abstract shortened by UMI.)

Trichoderma reesei is a filamentous fungus heavily used in the biotechnology industry due to its efficient secretion of cellulases. The enzymatic system of T. reesei consists primarily of four glycoproteins referred to as cellobiohydrolases (CBH I, CBH II) and endoglucanases (EG I, EG II). They exhibit microheterogeneity both in the N- and O-linked glycans. This thesis focuses on the method development to characterize the glycosylation profile and post-translational modifications present in these glycoproteins. Crude cellulase fermentation extracts RUT-C30 and its two mutant strains Iogen-M4, Iogen-B13 were initially analyzed by capillary isoelectric focusing (CIEF) to determine the cellulase composition in them. The major cellulase CBH I was purified from each strain and electrospray mass spectrometry (ESMS) was used to reveal the extent of overall glycosylation. To characterize the N-linked glycans and their attachment sites, CBH I from these strains were subjected to tryptic digest with and without PNGase F incubations followed by mass spectrometric detection. The O-linked glycans were released chemically by hydrazinolysis and were analyzed by high performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). The majority of O-linked glycans was di- and tri-saccharides. Two unusual posttranslational modification from strain RUT-C30 were observed: (1) both high mannose (predominantly Man 8GlcNAc2) and single GlcNAc in putative N-linked sites and (2) mannosylphosphorylation in a O-linked di-saccharide. Heterogeneity in putative N-linked sites was found consistently in CBH II, EG I and EG II from RUT-C30, however, no mannosylphosphorylation was observed at least on the proteins in the purified fractions. These results have led to the proposal of endogenous endoglycosidase H as well as mannosylphosphorylation activities possibly induced during fermentation.

The systematic study of the kinetic method has been carried out by observing a wide variety of proton-bound pairs consisting of amines, nitriles and alkanols using tandem mass spectrometry. The majority of the compounds studied were homologous alkyl series. The results were obtained for other selected amines such as pyridine, allylamine, benzylamine and phenylethylamine. The proton affinities of many of these compounds are well established in the NIST database, although some needed either to be revised or to be given their first experimental values. For the competitive dissociations of proton-bound dimers, which were studied under MI and CID conditions, the relative product yields, [B 1H+]/[B2H+], have been related to the proton affinities (PA) of B1 and B2 by the kinetic method. In its simplest form, the method assumes that there are no entropy effects and a zero reverse energy barrier for the competing dissociations. The general effects expected from non-zero entropies of activation are described in terms of how they influence log k(rate constant) vs. E(internal energy) plots for such competing dissociations. For these homologous series such effects were observed to be minimal and the kinetic method well reproduces most of the reference PA values. However, in view of the very close agreement between the present results and many reference PA values obtained from equilibrium studies, it is possible that small reverse energy barriers (ca. 5 kJ mol -1) may in some cases be identifiable by the kinetic method, in particular for t-C4H9NH2, the homologous di-n-alkylamines and possibly larger barriers for di-sec-alkylamines. The method is quite sensitive to small discrepancies in PA and new values have been proposed for C6H5CH2NH2 (923 kJ mol-1), C6H5CH2CH 2NH2 (930 kJ mol-1), (CH3)(n-C 4H9)NH (951 kJ mol-1), (i-C4H 9)2NH (969 kJ mol-1) and for homologous nitriles and alkanols, namely from C5 to C8. The PA value for n-C4H9OH was revised from 789.2 to 793.7 kJ mol-1. Second part of the thesis regards the following aspects: (A) Investigation of the a-cleavage mechanism in simple amine molecular ions and (B) Mass spectrometric studies of immonium ions, CH3CH2CH +NH2 and (CH3)2C+NH 2 and their oxygen analogues. (Abstract shortened by UMI.)

The primary objective of this project was the development of affinity selection techniques that can be used with mass spectral detection for the direct identification of trace level target antigens from complex matrices. Mouse monoclonal antibody IgG1 specific for the human c-myc epitope---EQKLISEEDLN---was covalently immobilized on recombinant protein G beads using chemical crosslinking. This immunomatrix was then used to selectively enrich both the wild-type and mutant single chain (scFv) antibodies (containing the c-myc epitope) from the periplasmic extract of Escherichia coli (E. coli). To identify the point mutation within the scFv mutant, a tryptic digest was performed followed by LC-MS of the tryptic peptides. This novel approach was applied to a number of scFv proteins containing single point mutations. In particular, mass spectrometry was used to confirm the site of mutation L46N on one specific mutant using LC-MS/MS. Immunoaffinity enrichment was also used to select the c-myc peptide from a spiked sample of human serum at a concentration of 20 ng/ml. LC-MS/MS was performed to confirm the identity of the selected peptide. (Abstract shortened by UMI.)

The techniques of electrophoresis have undergone rapid development in recent years. The problem of predicting and explaining results of electrophoretic experiments requires a well designed dynamical model that reproduces the proper dynamics under well-understood conditions. The challenge is to design a model which can predict complex dynamics in a variety of situations of interest and yet, be efficient enough to run on common workstations. We believe that we have been successful in achieving these goals. Our Brownian Dynamics model is sufficiently general for the investigation of a wide variety of problems (as we will demonstrate), moreover, it implements a simulation innovation, a dynamically self-regulating time increment which varies in accordance with the stresses in the system. This permits for reasonably long simulation times with good averaging statistics. After benchmarking the model with the well understood Rouse regime in isotropic and confined media, the model was applied to two case studies of particular interest: (i) electrophoretic collisions and stacking (chain pinning), and (ii), the onset of the entropic trapping regimes for low fields. We have found that, as far as simple isolated collisions are concerned, our model predicts a widening of electrophoretic molecular bands with an accompanying loss in spatial resolution. The onset of entropic regimes for low fields is demonstratedly counter-productive due to the wide variance of trapping times. Both these results are in accord with recent experimental observations. A more complete study is planned for a supercomputer to elaborate on these findings as well as to investigate kink-formations in pulsed-field gel electrophoresis (PFGE).

Part A. The preparation of a fully functionalized ring A of Taxol$\sp\circler$ (1), its further elaboration to the intramolecular Diels-Alder precursor 272 and the attempted cyclization of this compound are described. Allylic oxidation of the cyclohexene acetal 138 was accomplished in 65% yield using buffered selenium dioxide in toluene. Ketone 139 was converted in two steps to the vinyl ketone 197 in 33% yield. Reduction of ketone 197 occurred with excellent diastereoselectivity to furnish the allylic alcohol 198 in 96% yield. Protection and subsequent ozonolytic cleavage of 198 furnished the key intermediate aldehyde 200. Aldehyde 200 was converted through a three step sequence to the fully functionalized TMS cyanohydrin A ring compound 203. Addition of the vinyl anion derived from the bromo-diene 233 to the aldehyde 225 gave an approximately 5:1 mixture of diastereomeric alcohols 238/239 in 80% yield. Protection and separation of the major diastereomer followed by deprotection of the acetal gave the aldehyde 265 in 65% yield. Several methods were evaluated for the preparation of the dienophile component by the addition of the anions derived from dithianes 244 and 260 and (phenylselenomethyl)trimethylsilane 253. Addition of TBSCN to aldehyde 265 gave a 5:1 mixture of silyl-cyanohydrin diastereomers 268/269 with the major diastereomer 268 being isolated in 67% yield. X-ray crystallographic analysis of 268 was performed. Reduction of nitrile 268 to the aldehyde 267, addition of ethynylmagnesium chloride and Dess-Martin periodinane oxidation of the alcohols 270/271 gave the diene-dienophile ketone 272. Heating of this material in toluene in a sealed tube using "microwave assistance" failed to produce any of the desired cyclized product.* Part B. Preparation of the diene 400 and the attempted Diels-Alder cyclization of this diene to the adduct 398 is described. $\beta$-Keto-sulfone 371 was prepared in 89% yield in two steps from the vinyl-sulfide 369. Alkylation of 371 with the allylic bromide 374 followed by hydrolysis of the enol ether functionality furnished the diketone 378 in 55% yield for two steps. Intramolecular aldol-condensation and subsequent reduction gave the diol 380. Protection of the primary alcohol of 380 as a TBS ether afforded the secondary allylic alcohol 381. A variety of methods were investigated for the dehydration of alcohol 381 to the diene 383, with Grieco's procedure giving a yield of about 65%. Heating of diene 383 at $35\sp\circ$C was found to give the rearranged product 91/392. The cycloaddition of diene 383 with a number of dienophiles was investigated. Nitroethylene was found to give a mixture of unstable compounds from which no adduct could be isolated. Diene 383 underwent rearrangement to the exocyclic methylene compound 401 when Lewis acid catalysts were employed. Heating diene 400 with nitroethylene in benzene at reflux followed by treatment of the crude with potassium hydride and dimethyldioxirane gave a mixture of compounds of correct mass for structure 404 but of unproven structure.* ftn*Please refer to the dissertation for diagrams.

The goal of this project was to study the reactivity of thin pyrolitic carbon films on quartz, formed from propylene, methane and butadiene, using the methane conversion reaction as a test. The reactivity was compared to that of quartz. The rate of dissociation of methane on the carbon surface was as much as forty times greater than on quartz. The actual rate increase depended on the precursor of the carbon film. When the reproducibility of the carbon films was tested, the rate was observed to decrease with consecutive experiments. Subsequent analysis showed that small amounts of carbon were deposited during the course of each reaction. Since small amounts of carbon caused inhibition, but carbon films caused catalysis, the rate of formation of ethane, a measure of the rate of dissociation of methane, was determined as a function of the amount of carbon. The rate was also determined as a function of methane reactant pressure and temperature. All the results indicated that the methane was dissociated on the carbon surface to form methyl and hydrogen radicals. The increase in formation of these radicals led to the increased rate. The subsequent mechanism was similar to the homogeneous mechanism of methane decomposition. In order to confirm this result, the rate was determined in a reaction vessel with a different surface to volume ratio. The surface mechanism was confirmed. The carbon film was characterized using different techniques. The hydrogen content of the films was determined and related to the reactivity of the carbon formed from different precursors. The film was studied by electron microscopy, XRD, FTIR and mass spectrometry. An overall mechanism was developed for the methane decomposition reaction in the presence of carbon which was extended over the different temperature and pressure conditions studied. Finally, a new technique of synthesizing a catalyst containing carbon and iron, and results obtained in the study of water vapour gasification and methane conversion in the presence of this catalyst are discussed.

The significance of effects of pressure on the kinetics and equilibria of chemical processes in the condensed phase has been known since the time of van't Hoff. The use of pressure as an extra variable has become an integral part of kinetic and related studies on reactions over the past 25 years. However, interest in the behaviour of electrochemical processes as a function of pressure is of more recent origin (<10 years); consequently, experimental data are relatively scarce. This thesis describes the pressure dependent behaviour of a variety of electrochemical reactions of fundamental and practical interest. The pressure coefficient of EMF of several electrochemical cells comprised of two reversible electrodes is evaluated in order to determine their suitability for use as reference electrodes at elevated pressures. The electrochemical cells examined are: (1) Pt,H2|0.5 M H2SO4 |Pd-H,Pd; (2) Pt,H2|1.0 M HCl|Pd-H,Pd and (3) Pd,Pd-H|HCl|AgCl,Ag. The pressure coefficient of EMF of each electrode cell is converted to the partial molar volume change for the corresponding overall cell reaction. The partial molar volumes of Pd-H,H in Pd and of H2 itself in solution are also evaluated. The reproducibility of the measured pressure coefficients indicates that the Ag,AgCl and Pd-H,H+ reference electrodes are the most suitable of those evaluated for use at elevated pressures. The effects of high hydraulic pressures on the kinetics of electrode reactions are complex because of (a) the variation of reference electrode potentials with pressure so that only an apparent volume of activation can be directly measured experimentally and (b) dependence of coverage by adsorbed intermediates, such as H, with pressure. Methods for dealing with these problems are treated and the significance of measured apparent volumes of activation for the hydrogen evolution reaction is discussed in terms of the nature of the transition state for proton transfer and neutralization. The negative true volumes of activation found for the cathodic H2 evolution reaction under some conditions are attributed to increasing electrostriction of the proton in the H9O+4 complex as the transition state is formed. New information on the nature of the solvent reorganization process involved in formation of the transition state in an electrochemical redox reaction [the Fe&parl0;CN&parr0;3-6/Fe&parl0;CN&parr0; 4-6 couple] is given by studies of the kinetics of this process at high pressures in aqueous solutions at Au. The apparent activation volume is evaluated together with the true volumes of activation for the forward and reverse directions of the redox reaction. By means of calculations of the electrostriction associated with the long-range dielectric polarization in comparison with that associated with ion-solvent interaction in the primary hydration shell, it is shown that the activation process must be mainly (75%) due to short-range solvent reorganization in the primary shell. This is contrary to what has often been assumed in the interpretation of the energy of activation of ionic redox reactions. The experiments also allow some deduction to be made about the "symmetry" of the transition states in electrochemical reactions in comparison with that in the corresponding homogeneous reactions, in so far as solvent reorganization is concerned. The multiple states of electrosorbed H and oxygen species at Pt, which arise below monolayer coverage, are of current interest in electrochemical surface science. Experiments on the effects of high hydraulic pressure on H chemisorption and surface oxide formation at Pt electrodes are described. "Clean" electrochemical surface studies can be performed in a shrinkable teflon vessel, under oil, up to several thousand bars. The results of these experiments enable the equilibrium volume changes for electrodeposition of H and surface oxidation of Pt to be evaluated, allowing for the effect of pressure on the potential of the reversible reference electrode used. The volume of electrodeposited H is found to be ca. 5 cm3mol -1 at Pt, a value substantially larger than that for sorption of H into (alpha + beta) Pd-H but similar to that of H covalently bound in aliphatic hydrocarbon methylene groups. (Abstract shortened by UMI.)

The deamination reaction of 2-aminotriphenylcarbinol in 97% formic acid has been studied in order to investigate the possibility of benzyne formation via a carbonium ion precursor. Product analysis showed that a small but significant amount of benzophenone was formed, which indicated some fragmentation. While this evidence strongly suggests that the other fragment was benzyne, no direct proof of this has been obtained.

The analysis of the N.M.R. spectra of a series of substituted 1,4-dioxanes showed the following configurations and conformations for the compounds: 2,3-dichloro-1,4-dioxane m.p. 52°, 2,3-diphenyl-1,4-dioxane m.p. 132° and naphtodioxane have the cis configuration and at room temperature are rapidly interconverting between chair conformations; 4a,8a-dimethoxy-4a,8a-dihydro-1,4-benzodioxan and 4a,8a-dimethoxy-4a,5,6,7,8,8a-hexahydro-1,4-benzodioxan have the trans configuration with a rigid chair conformation. Non-equivalence in the vicinal coupling constants JAX and JBX were found for 2,2,2-trifluoroethyl-2-tetrahydrofuranyl ether and 2,2,2-trifluoroethyl-2-tetrahydrofuranyl ether and 2,2,2-trifluoroethyl-2-tetrahydrothiapyranyl ether as found previously for 2,2,2-trifluoroethyl-2-tetrahydrofuranyl ether by Dr. R.R. Fraser and P. Hanbury. No definite conclusion regarding the origin of this non-equivalence could be drawn from the results. Linear Hammett plots were obtained in a study of the effect of substituents on the chemical shift of the methylene protons in para-substituted benzyl phenyl sulfides, sulfoxides and sulfones. It has been found that the protons in the methylene group are affected very differently by the substituent in each series. The causes for this variation have been examined and it is proposed that the effect is conformational. In the conformation in which a C-H bond is paralled to the adjacent p-orbital of the benzene ring, that C-H proton will experience the maximum substituent effect.