Capillary zone electrophoresis (CZE), a powerful separation method based on the differential migration of charged species under the influence of an electric field, has been widely used for separations covering from small ions to big biomolecules. Chapter 1 describes the method, then discusses detection of the separated analytes by laser induced fluorescence and by chemical derivatization, and the use of O-phthaldialdehyde (OPA) as a post-column reagent. Chapter 2 describes a post-column reactor which uses two narrow bore capillaries connected coaxially. This reactor differs from other coaxial reactors in terms of capillary dimensions, reagent flow control, ease of construction and most importantly, better limits of detection. The derivatization reagent is electroosmotically driven into the reaction capillary and the reagent flow rate is independently controlled by a high voltage power supply. Amino acids, amines and proteins, derivatized by OPA/2-mercaptoethanol using this post-column reactor coupled with LIF detection, show low attomole mass limits of detection, and for the first time, the authors demonstrate single cell capability with a post-column derivatization scheme. The single cell capability shows that this reactor could find applications in assaying non-fluorescent or electrochemically inactive components in individual biological cells in the future.

In the first part of the thesis, the electric resistivity, Seebeck coefficient, and Hall effect were measured in X{sub y}(Y{sub 2}S{sub 3}){sub 1-y} (X = Cu, B, or Al), for y = 0.05 (Cu, B) or 0.025-0.075 for Al, in order to determine their potential as high- temperature (HT)(300-1000 C) thermoelectrics. Results indicate that Cu, B, Al- doped Y{sub 2}S{sub 3} are not useful as HT thermoelectrics. In the second part, phase stability of {gamma}-cubic LaSe{sub 1.47-1.48} and NdSe{sub 1.47} was measured periodically during annealing at 800 or 1000 C for the same purpose. In the Nd selenide, {beta} phase increased with time, while the Nd selenide showed no sign of this second phase. It is concluded that the La selenide is not promising for use as HT thermoelectric due to the {gamma}-to-{beta} transformation, whereas the Nd selenide is promising.

This thesis showed that it is possible to achieve well ordered growth at low temperatures when chaing fluxes during the course of the deposition. It was also demonstrated that nucleation theory fails to predict or explain at least part of the results, in particular when deposition takes place at an initially low rate, with presumably a relatively low nucleation density, followed by a change to a high flux rate. This points to an inherent lack of nucleation theory; alternative explanations are presented based on flux-independent growth as reported by Roos (Surf. Sci. 302 (1994) 37).