No description available.

No description available.

No description available.

The work described in this thesis is concerned with the study of ph-ysisorbed phases by x-ray diffraction using a conventional sealed x-ray tube source. Diffraction data has been collected for a number of adsorption systems using graphite, a montmorillonite clay (Gel White) and zeolite rho as the substrate. It is well known that phases of unique two-dimensional character can be formed on the surface of graphite, and the structure of adsorbed benzene and hexaflurobenzene on graphite have been studied in this thesis. Contrary to current theoretical predictions the >/7x T/7R19" commensurate structure of submonolayer benzene has been confirmed. Submonolayer hexaflurobenzene appears to form a striped domain structure based upon the commensurate x3 lattice in which the molecules are incommensurate with the surface. ' In contrast to the homogeneous surface of graphite, the surface of a clay is microporous and heterogenous, and this gives rise to broad diffraction lines from an adsorbed phase. Despite this, the surface area of Gel White has been deduced from the evolution of the diffraction pattern of a krypton adlayer as a function of krypton loading. The formation of bulk krypton is readily identified and the small size of the three-dimensional crystallites suggests that they are formed within the micropores of the clay. The structure of adlayers of krypton and xenon within the interlayer spacing of Al-pillared Gel White has been studied in order to determine the mean inter-pillar separation. At all the coverages studied, xenon forms a close packed single layer structure whilst krypton appears to form a more complex bilayer phase. A tentative suggestion as to the mean pillar separation from this work is 30X. The adsorption site of krypton, xenon and CH3CI within zeolite rho has been determined using the method of x-ray Rietveld whole profile refinement. The principal site of adsorbed krypton and xenon is at the centre of the octagonal prism. The chlorine atom of CH3CI sits in the centre of the face of the octagonal prism and the methyl group is slightly displaced from the centre of the prism. 37-5354 Hydrodynamics of liquid encapsulation czochralski crystal growth Hicks, T.W. Bristol Ph.D. 1989 Dig. Certain aspects of crystal growth from a melt are investigated. We begin by describing the methods of producing single crystals. Particular emphasis is placed on the need for a better understanding of the hydrodynamics of the encapsulant region of the Liquid Encapsulation Czochralski (LEG) technique. We also introduce the basic physical processes which govern crystal growth. In Chapter 2 we develop a mathematical model of the encapsulant region of the LEC crystal growth system. The equations and boundary conditions that govern the encapsulant flow are formed using a vorticity-stream-function approach, after which the problem is recast in a dimensionless form. In Chapter 3 the equations of motion are represented in a finite difference form and a numerical method for solving the time-marching problem presented by the parabolic equations is developed. The elliptic stream-function equation is solved at each time level using the successive over-relaxation technique. Solutions of the model equations for the growth of GaAs crystals through B3O3 encapsulant are presented in Chapter 4. In all cases considered the flow field tends towards a steady state. For shallow encapsulants, the heat transfer in the encapsulant is conduction dominated, but for deeper encapsulants, advective heat transfer can be significant. In the last chapter we investigate the effect of Soret diffusion on the morphological stability of a freezing interface using linear stability theory. A Soret flux directed towards the interface has a destabilising effect. Over-stable modes of instability exist for very low crystal growth rates, but we are unable to find conditions under which the overstable mode is the most unstable.

No description available.

This dissertation details the effect of underpotentially deposited (UPD) Pb on the surface enhanced Raman scattering (SERS) ability of roughened polycrystalline Ag electrodes. The deposition of monolayer and submonolayer amounts of Pb results in a quenching of the SERS response for pyridine and Cl⁻ adsorbed at Ag electrodes. Various factors which may contribute to the loss of SERS intensity are investigated. The most significant factors include changes in surface roughness features brought about by Pb UPD, changes in surface electronic properties of Pb-modified Ag and changes in a chemical contribution to surface enhancement. Possible changes in surface roughness properties of the Ag electrode due to Pb deposition are examined using scanning electron microscopy (SEM) and SERS reversibility studies. SEMs of roughened Ag electrodes before and after Pb monolayer deposition show no significant change in the morphology of the larger roughness features. However, the deposition and stripping of 60 - 70% of a Pb monolayer results in a loss of ca. 50% of the original SERS intensity for both adsorbate bands. This irreversible loss of SERS intensity is attributed to the destruction of atomic scale roughness (ASR). These results suggest that ca. 50% of the observed SERS response arises from a mechanism involving ASR. In addition, the destruction of ASR is shown to be largely responsible for the quenching of SERS at higher Pb coverages. The morphology of the SERS quenching profiles at lower Pb coverages for pyridine and Cl⁻ varies as a function of excitation wavelength. Experimental quenching profiles are compared with theoretical quenching profiles based on an electromagnetic contribution to SERS. Theoretical quenching profiles are calculated using a model for electromagnetic enhancement at a overlayer-covered ellipsoids proposed by Murray. The experimental results for both adsorbates are in agreement with the theoretical predictions for laser excitation in the blue. Experimental results in the green and red wavelength regions are best explained in terms of photoassisted charge-transfer mechanisms for surface enhancement.

The properties of surface plasmon polaritons which propagate in optically nonlinear or magnetic media are investigated. Expressions for the electromagnetic fields of a multilayer stack are derived by modification of the Fresnel reflection coefficients. Original research on prism-coupled, nonlinear, long-range surface plasmon polaritons, and on bound and prism-coupled, long-range surface magnetoplasmon polaritons in a transversely-applied magnetic field, is compared to previous research on single-interface plasmon polaritons. The reflectance from prism-coupled, nonlinear surface plasmons is analyzed using the infinite plane-wave approximation and a substrate nonlinearity which depends on the square of the transverse-electric field. Bistable switching requires incident intensities two orders of magnitude smaller for the long-range mode than for the single-interface mode. The regime in which the approximations are valid is shown to extend beyond that of first-order perturbation theory to guided waves that are very near cutoff. The sign and location of the nonlinearity become significant for these waves. For positive nonlinearities, nonlinear wave analysis indicates an additional branch of the reflected intensity curve, due to self-focussing of the guided wave. Positive and negative nonlinearities exhibit different switching intensities. The propagation constant of the long-range surface plasmon of a magnetic metal film is shifted by the application of a transverse magnetic field. The sign and magnitude of the shift are highly dependent on the metal thickness and the refractive indices of the bounding media. The shift is manifested experimentally as a resonant modulation of the reflectance from the prism-coupled surface plasmon due to changes in the angular position and width of the plasmon resonance. Experimental prism-coupling to the long-range surface magnetoplasmon in thin nickel films confirms the theoretical expectations for a wide variety of sample parameters. The phase of the magneto-optic coefficient is determined from the angular profile of the reflectance modulation. Although the shift of the propagation constant may be two orders of magnitude smaller for the long-range mode, the modulation signal is the same order of magnitude for long-range and single-interface magnetoplasmons.

Three alkyldihydrochlorosilanes were synthesized; ethyldihydrochlorosilane, octyldihydrochlorosilane and octadecyldihydrosilane. Ethyldihydrochlorosilane was produced by the reaction of ethylsilane with mercuric chloride and the other two chlorosilanes were produced by the reaction of the alkyl Grignard reagent with dichlorosilane. Each alkyldihydrochlorosilane was reacted with porous silica in an attempt to discover the extent of reaction or the highest surface concentration of bonded groups attainable. The reaction between these alkydihydrochlorosilanes and porous silica was compared to the reaction between silica and the analogous alkyldimethylchlorosilane. The rate of reaction of both type of chlorosilane was found to be essentially the same. The maximum surface concentration of bonded surface groups attainable by alkyldihydrochlorosilanes was found to be approximately 1.3 #moles/m² greater than that attainable by alkyldimethylchlorosilanes. This increased surface coverage seemed to depend very little on the chain length of the alkyl group and was attributed to the decrease in steric hindrance of the bonding silicon atom of the silane. Surface bound silyl hydrides could be oxidized selectively and sequentially to form silane silanols. Surface silanes also appeared to reduce chloroplatinic acid, but were not observed to add efficiently to olefins. The chromatographic properties of silica modified with alkyldihydrochlorosilanes were compared to those of equivalent silicas modified with alkyldimethylchlorosilanes and alkyltrichlorosilanes before and after the surface silanes were oxidized. Both normal and reversed-phase liquid chromatographic studies were conducted. In general, it was found that alkyldihydrochlorosilanes yielded the most polar modified silicas. This greater surface polarity was attributed to an increase in the activity of water in the near surface region of the bonded phase.

The growing technological application of metallic lithium has produced a greater need to understand its fundamental surface chemical properties. The use of lithium as an anode in high-energy density battery systems represents one application where this knowledge is required to optimize system performance. The surface chemistry of lithium will be discussed in terms of oxidants which represent the reductive half-cell components of these batteries, contaminants present during cell fabrication, and solvents used as the electrolytic medium. These systems have been studied in the low pressure limit ( < 1 millitorr) at atomically clean lithium surfaces using X-ray Photoelectron Spectroscopy (XPS). The lithium/sulfur dioxide system has been singled out for detailed study in order to explore the relationship between gas-phase and solution-phase processes. Electrochemical characterization of the lithium anode has been conducted as a function of controlled surface composition within this system. The ability of lithium to induce corrosion at structural components of these batteries (i.e., glass insulators) has also been investigated. A description of the chemical activity of lithium and its consequence has been developed from these results.

Significant advances in the understanding of chemical interactions at the solid-liquid interface have been made in this research through the use of diagnostic separations as a surface analysis technique. Diagnostic liquid chromatography has been employed in a detailed investigation of the thermodynamic and kinetic quantities which describe the interactions associated with a temperature induced conformational change in the octadecyldimethylsilane moieties of two different bonded silica materials. As a result of this study the nature of the structure and interactions of the ∼20Å thick interfacial region which acts as the stationary phase in reversed-phase liquid chromatography (RPLC) has been elucidated. The location and orientation of the average intermolecular interactions in the solvated layer stationary phase for solutes of differing hydrogen bonding ability and geometry has been determined as affected by bonded surface coverage, solvent hydrogen bonding competition and the structure of the solvated layer. These refinements in the model of the stationary phase solvated layer provide a much more detailed and accurate description of the intermolecular interactions responsible for retention and selectivity in RPLC than was previously available. A new modification of the method of measuring column mobile phase volume in RPLC employing retention linearization of an homologous series of compounds has been described from fundamental themodynamic principles and a statistically valid data reduction approach. The added advantage of providing thermodynamic information about the chromatographic system under study is inherent in this new technique. The experimental and theoretical bases for the new separation technique of magnetic field-flow fractionation (magnetic FFF) have been demonstrated. It has been shown that FFF techniques can be used in a diagnostic mode to study the dynamic stability of particle suspensions. The application of an external magnetic field to non-aqueous suspensions of sub-micron sized γFe₂O₃ particles, whose surface character has been modified by the adsorption of water, has been shown to enhance the suspension stability with respect to sedimentation. With the choice of proper operational conditions, magnetic FFF has also been demonstrated to be useful in monitoring particle flocculation as a result of its ability to separate particle flocculates on the basis of size.