The application of Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) to the determination of total oxygen, nitrogen and sulphur in organic solution has been studied. Near-infrared emission lines of the three nonmetals have been used and detection limits in xylene have been established. Oxygen was determined for the first time by ICP-OES in a nongaseous sample. The characteristics of the inductively coupled plasma when used with an organic aerosol have been studied with respect to the completeness of atomization and potential interference from the complex excitation environment. Spatially resolved atomic and molecular emission profiles have been collected as part of these investigations. The nonmetal emission intensities were found to depend on the boiling point of the compounds containing the nonmetal because a redistribution effect in the nebulizer chamber was leading to an enrichment of volatile solutes in the aerosol stream to the plasma. Response factors for a series of compounds have been determined. The utility of the method was shown by establishing working curves and by testing certified standards for sulphur. / Science, Faculty of / Chemistry, Department of / Graduate

Hyaluronic acid plays an important role in regulating the transport of fluid and solutes in the interstitium. The concentration and molecular weight of hyaluronic acid in different connective tissues are different. These factors influence the hydraulic flow conductivity, K', of connective tissues. An experimental study of the effect of concentration and molecular weight of hyaluronic acid on the hydraulic flow conductivity is the subject of this work. Hyaluronic acid of different molecular weights were obtained by fractionating commercially available hyaluronic acid using ion-exchange column chromatography. The results were not reproducible, partly because of the elution process was not continuous. Nevertheless, three molecular weight fractions (6.99 to 11.1 X 10⁵ ) were obtained. Hyaluronic acid of lower molecular weights (0.454 to 1.65 X 10⁵) were obtained by acid hydrolysing some of the chromatographed fractions for 15 min., 1 hour and 2 hours. A more homogeneous hyaluronic acid fraction (M.W. = 1.96 X10⁵) was obtained by fractionating hyaluronic acid materials acid hydrolysed for 15 min. The hydraulic flow conductivity of solutions of hyaluronic acid can be calculated from the sedimentation coefficient of the solutions at 20°C, S₂₀‚ measured by ultracentrifugation. Centrifugation experiments determining the S₂₀ of the molecular weight fractions of hyaluronic acid at various concentrations were therefore undertaken. The results showed that S₂₀ decreased with increased concentration of hyaluronic acid. Also, the curves of as a function of hyaluronic acid concentration, c, converged at high concentration, indicating that a three dimensional molecular network is formed at high concentration and the extent of entanglement between molecules is the same for the high and low M.W. fractions. At lower concentrations, for the acid hydrolysed fractions, S₂₀ increased with M.W., which is in agreement with past sedimentation data. For the non-acid hydrolysed fractions, the difference in S₂₀ between two higher M.W. fractions is small, and the lowest M.W. fraction has consistently higher S₂₀ than the higher M.W. fractions. This finding does not agree with past literature results, and the difference in results is most probably due to experimental errors. However, when the fractionated non-acid hydrolysed fractions are taken as a high M.W. group (M.W. = 6.99 to 11.1X10⁵) and the acid hydrolysed fractions as a low M.W. group (M.W. = 0.454 to 1.96X10⁵), the curves of S₂₀ as a function of c of the low M.W. group fall below those of the high M.W. group, which is in agreement with past sedimentation data. The hydraulic conductivities (K'), calculated from S₂₀ data, for all the HA fractions varied inversely with concentration. The log-log plots of K' versus c compared well with the results of Ethier (1986). The K' versus c relationships for all the fractions converged at high concentrations. At low concentrations, the HA molecules of the high M.W. group has a higher K' than those of the low M.W. group. / Applied Science, Faculty of / Chemical and Biological Engineering, Department of / Graduate

A series of chelating amino siloxide and alkoxide ligands have been prepared. Barium, lanthanide and zirconium complexes bearing these new ligands have been synthesized and characterized by NMR or X-ray crystallography. Reactivity of the zirconium complexes has been studied. The tris(siloxide) lanthanide(III) complexes are all monomeric in both solution and solid-state. A yttrium tris(siloxide) and a zwitterionic ytterbium tetrakis(siloxide) complexes have been characterized by X-ray crystallography. The yttrium complexes have shown high volatility. Mono(siloxide) barium(II) silamide complexes are dimeric as determined by X-ray crystallography, and barium bis(siloxide) exists as a monomer-dimer equilibrium in hydrocarbon solvents. Although highly soluble in hydrocarbon solvents, the barium complexes are non-volatile. A new aryl(siloxide) ligand has been synthesized and used as an ancillary ligand for the preparation of lanthanide dialkyl complexes. Although ligand redistribution was not observed, “ate”-complexes were isolated instead of neutral dialkyl complexes. Zirconium bis(aminodiolate) complexes were synthesized by reaction of tetrabenzyl zirconium and two equivalents of the ligands. The substituents on nitrogen were found to have a great effect on the structure. Mono(aminodiolate) zirconium dialkyl complexes were successfully synthesized by three different methods: protonolysis, ligand redistribution and metathesis. Thermal decomposition of these zirconium dialkyl complexes was found to show a marked dependence on the substituents at nitrogen. The α-methyl benzyl derivative decomposed by ortho-metallation of phenyl group exclusively and resulted in clean formation of a metallacyclic complex. The reaction of primary amines with the metallacycle allowed isolation of amide intermediates protonolysis of the benzyl group, and eventually resulted in formation of bridging imide complexes. The insertion reaction of carbonyl groups (C=O) into the metallacycle Zr-carbon bond proceeds regio- and stereoselectively. The first insertion products of carbonyl insertion were isolated, and the β-naphthaldehyde insertion product was characterized by X-ray crystallography. The metallacycle exhibited catalytic activity towards alkyne cyclotrimerization without preactivation. Zirconium cationic complexes were generated by alkyl abstraction using [special characters omitted]. The cationic complexes generated from the zirconium dibenzyl derivatives showed catalytic activity towards both ethylene and 1-hexene polymerization. A new cyclopentadienyl ligand bearing pendant fluorinated alkoxide functionality was synthesized. The zirconium complexes bearing this ligand exhibited remarkably high Lewis acidity, and the dichloride complex was shown to be catalytically active towards vinylether polymerization. / Graduate

La formulation des principes actifs (PA) pharmaceutiques à l'état amorphe est une préoccupation majeure actuelle du génie pharmaceutique car elle permet d'améliorer la solubilité des (PA) peu solubles. Son optimisation requiert une parfaite connaissance de la courbe de solubilité du (PA) dans le polymère, de manière à définir le taux maximum de (PA) pouvant être chargé dans le polymère sans risque de recristallisation ultérieure. La détermination de cette courbe de solubilité est cependant extrêmement difficile en raison de la forte viscosité des polymères qui rend les états saturés très difficiles à atteindre. Dans cette thèse, nous présentons une méthode efficace de détermination des courbes de solubilité qui permet de s'affranchir des limitations cinétiques liées à la viscosité. L'idée générale est d'utiliser des dispersions moléculaires et cristallines (DMC) dans lesquelles le (PA) est dispersé dans le polymère, en partie au niveau moléculaire et en partie sous la forme de minuscules cristallites. Cette microstructure particulière augmente considérablement la vitesse de dissolution des cristallites puisque : (1) La dispersion moléculaire a un effet plastifiant qui augmente considérablement la mobilité moléculaire dans la matrice amorphe. (2) La fine dispersion des cristallites réduit fortement les distances sur lesquelles les molécules de PA doivent diffuser pour envahir le polymère. La méthode a été mise au point et testée sur deux (PA) (le sulindac et le paracétamol) et un polymère (le polyvinylpyrrolidon). L'étude des courbes de solubilité nous a par ailleurs permis de préciser le polymorphisme du sulindac et du paracétamol, et en particulier les relations de stabilité des polymorphes étudiés. / The formulation of pharmaceutical active principles (PA) in the amorphous state is a major current concern of pharmaceutical engineering because it improves the solubility of poorly soluble (PA). Its optimization requires a perfect knowledge of the solubility curve of (PA) in the polymer, so as to define the maximum level of (PA) that can be loaded into the polymer without the risk of subsequent recrystallization. The determination of this solubility curve is, however, extremely difficult because of the high viscosity of the polymers which makes the saturated states very difficult to achieve. In this thesis, we present an efficient method for determining solubility curves that allows us to overcome the kinetic limitations related to viscosity. The general idea is to use molecular and crystalline dispersions (MCDs) in which (PA) is dispersed in the polymer, partly at the molecular level and partly in the form of tiny crystallites. This particular microstructure greatly increases the dissolution rate of the crystallites since: (1) Molecular dispersion has a plasticizing effect which greatly increases the molecular mobility in the amorphous matrix. (2) The fine crystallite dispersion strongly reduces the distances over which the API molecules must diffuse to invade the polymer. The method was developed and tested on two (PA) (sulindac and paracetamol) and a polymer (polyvinylpyrrolidon). The study of solubility curves also allowed us to specify the polymorphism of sulindac and paracetamol, and in particular the stability relationships of the polymorphs studied.

A method is reported for determining the species formed in solution by ligand replacement reactions. Reactions in dilute aqueous solution were studied in which the Cl^- on HgCl_2(aq) was replaced by hydroxide ion, ethylenediamine, glycinate ion, and methylamine. The species formed by the replacement reactions were established through computer analysis of pH titration data. Various reaction sequences were proposed for each system and log K values were calculated for each sequence. The selection of the correct reaction sequence for each system was based on the premise that only when the correct sequence is proposed will consistent log K values be obtained as a function of both n (average number of ligands bound per metal ion) and total metal concentration. Omitting ionic charges the reactions were found to be of the general type HgCl_2 + Y = HgClY + Cl (1) HgClY + Y = HgY_2 + Cl (2) where Y is one of the four ligands used in the study. Log K values valid at 25° C and zero ionic strength were determined. The reactions were carried out at ionic strength values of about 10^-3, and activity coefficients were used to correct the calculated concentration quotients to thermodynamic equilibrium constants. Values of ΔH° were determined calorimetrically at 25° C for each of the replacement reactions using a thermometric titration procedure. Values of ΔS° for the reactions were calculated from the corresponding ΔG° and ΔH° values. The thermodynamic data obtained indicate that the ethylenediamine and glycinate ion act as bidentate ligands in the HgClY species (equation (1)).

The results of an investigation of the solubilities of silver chloride, silver bromide, silver iodide, and silver thiocyanate in water, in approximately 10, 20, 30, 40, and 50% by weight ethanol-water, methanol-water, acetone-water, and dioxane-water mixtures, and in an approximately 60% by weight dioxane-water mixture at 25°C are presented. The concentration of the anion of each of these four silver salts was varied between about 10^-5 and 10^-1 formal in each solvent. A radiotracer technique was used to determine the total silver content of these various solutions saturated with the appropriate slightly soluble silver salt. The results of "silver ion specific" electrode potential difference measurements in these saturated solutions were used in some of the calculations. Least-squares methods of data treatment were used to calculate the values of the thermodynamic equilibrium constants that describe the relationships among the various species in solution. A comparison of the results with the various theories of ionic interactions in solution is made and some of the inadequacies of these theories are discussed.

The heat of mixing of nonelectrolytic liquids is important in predicting the type of molecular interactions which exist in solutions of nonelectrolyted. The problem was to construct a calorimeter in which these heats of mixing could be measured accurately. Once constructed, the calorimeter was used to measure the heats of mixing of some binary and a ternary solutions. The systems chosen for study were the three liquid pairs formed between benzane, carbon tetrachloride, and cyclohexane, and the ternary mixture of the three. The results of the experimentation are summarized by the following equations. (1) H_12^M = X_1x_2 [103.2 + 13.1(x_1 - x_2) - 13.3(x_1 - x_2)^2] (2) H_23^M = x_2x_3 [729.6 + 102.4(x_2 - x_3)^2] (3) H_13^M = x_1x_3 [160.8 + 18.9(x_1 - x_3) - 20.2(x_1 - x_3)^2] (4) H_123^M = H_12^M + H_23^M + H_13^M + x_1x_2x_3 [128 + 161(x_1 - x_3) - 1636(x_1 - x_3)^2] where H^M stands for heat of mixing and x for mole fraction. Here, (1) is carbon tetrachloride, (2) is benzene, and (3) is cyclohexane. The value of the heat of mixing as calculated from the equation has the dimensions of caloried per mole. The measurements were all made at 25.0°C. The plots of heat of mixing against mole fraction for the three binaries are approximately symmetrical with respect to the line x + 0.500. The plot for the ternary mixture is symmetrical with respect to the line x_3 = 0.44. Furthermore, the heats of mixing of the ternary can be predicted from the binaries to within 5%.

The Reflected Quasipotential V(x) is the solution to a variational problem that arises in the study of reflective Brownian motion. Specifically, the stationary distributions of reflected Brownian motion satisfy a large deviation principle (with respect to a spatial scaling parameter) with V(x) as the rate function. The Skorokhod Problem is an essential device in the construction and analysis of reflected Brownian motion and our value function V(x). Here we characterize V(x) as a solution to a partial differential equation H(DV(x))=0 in the positive n-dimensional orthant with appropriate boundary conditions.  H(p) is the Hamiltonian and DV(x) is the gradient of V(x). V(x) is continuous but not differentiable in general. The characterization  will need to be in terms of viscosity solutions. Solutions are not unique, thus additional qualifications will be needed for uniqueness. In order to prove our uniqueness result we consider a discounted version of V(x) in a truncated region and pass to the limit. In addition to this characterization of V(x) we explore the possibility of cyclic optimal paths in 3 dimensions. / Ph. D.

No description available.

No description available.