Thesis (Ph. D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2007. / Smith, Glenn, Committee Member ; Trebino, Rick, Committee Member ; Adibi, Ali, Committee Chair ; Gaylord, Thomas, Committee Member ; Buck, John, Committee Member.

The structure and conformation of lithium phenyl stearate (and to a lesser extent, for comparative purposes, cadmium stearate) was investigated using Fourier transform infrared spectroscopy, and various modelling techniques. The infrared results for LiPS show that the aliphatic portion of the soap molecule is much more ordered at room temperature than had been expected, having only 0.62 and 0.60 gtg and gg defects per molecule respectively, where an isotropic chain would have 1.35 and 1.21 gtg and gg defects per chain respectively. As the temperature is increased the number of conformational defects increases continuously, until at < 130°C the chain reaches an isotropic degree of disorder. At this point the phase transition begins, so the chain reaches liquid like disorder before the phase transition begins. Modelling of the phenyl stearic acid showed that the phenyl group was restricted to certain angle of rotation values, and that the bonds close to the phenyl group were prevented from attaining true rotational isomeric state conformations, gtg defects near the phenyl group were distorted only slightly from their usual angular position, and an additional band in the infrared spectrum of LiPS at 1363 cm-1 has been assigned to this distorted gtg/gtg' defect. The gg defects near the phenyl group have a much greater distortion (and energy) resulting in a much reduced probability of occurrence. The number of gg defects present at the phase transition (< 130°C) was only 75% of that expected for an isotropic n-alkane of equivalent chain length, indicating that the four bonds nearest to the phenyl group have a reduced probability of forming a gg defect. The modelling of the ionic core of LiPS gives a reasonable estimate of between 5.6 to 7.1 A for the core radius. When this is used to calculate the hexagonal cylinder diameter, at room temperature, along with the average chain extension, it gives a value for the cylinder diameter of between 33.9 to 36.8A. The hexagonal lattice parameter determined by X-ray diffraction has a value 35.9A. Also after the LiPS sample has gone through the phase transition beginning at >130&deg;C, the hexagonal lattice parameter is 31.4A while the cylinder diameter lies between 30.2 and 33.2A.Crystalline cadmium stearate was found to contain two crystal forms, orthorhombic which has lattice dimensions of a0=5.05A, b0=7.35A and c0=48.6A and the other eithermonoclinic or triclinic. In the reverse hexagonal phase, the cadmium stearate molecule behaves like an isotropic n-alkane of equivalent chain length. The model used to predict the core radius of divalent metal soaps gives rise to some inconsistencies: the cylinder diameter thus determined gives a result between 28.8A to 31.7A, while the lattice parameter determined by X-ray diffraction gives a value of 36.9A. The assumption that the n-carboxylate ions in a divalent metal soap behave like two independent monovalent metal ion soaps appears to be incorrect.

Doctor of Philosophy / Department of Chemical Engineering / James H. Edgar / Recently, the rare earth nitrides have received a considerable attention from theorists and experimentalists due to their potential applications in spintronic, piezoelectric, and thermoelectric devices. In this work, erbium nitride (ErN) and scandium nitride (ScN) crystals were grown and characterized experimentally, and the growth process was modeled and simulated. Erbium nitride (ErN) is a rare earth nitride notable for its magnetic and optical properties. Here we report on its growth on a non-native substrate, tungsten foil, via physical vapor transport, and its characterization. The source material was erbrium metal that was converted to ErN by heating in nitrogen. Subsequently, it was sublimed to form the ErN crystals. The operating conditions were 1620-1770 ⁰C and 150-330 Torr in pure nitrogen. The growth rate increased exponentially with temperature with an activation energy of 508 kJ/mol, and inversely with pressure. X-ray diffraction revealed the ErN preferentially adopted a (100) orientation, the same as the dominant orientation of the tungsten sheet. The lattice constant was 4.853 Å. The crystal shapes and sizes were dependent on the temperature, as revealed by SEM and optical microscopy. The ErN crystals were highly faceted, bound by (100) and (111) crystal planes. The ErN compound deviated from stoichiometry: the Er:N atomic ratio ranged from 1:1.15 to 1:1.2 according to EDX and XPS elemental analysis. Raman spectra was in good agreement with theoretical predictions. Scandium nitride single crystals (14–90 µm thick) were grown on tungsten (100) single crystal substrate by physical vapor transport in the temperature range of 1850-2000 ⁰C and pressure of 15-35 Torr. Epitaxial growth was confirmed using in-plane ɸ scan and out-of-plane x-ray diffraction techniques which revealed that ScN exhibited cube-on-cube growth with a plane relationship ScN (001) || W (001) and normal direction ScN [100] || W [110]. Atomic force microscopy revealed the surface roughness decreased from 83 nm to 18 nm as the growth temperature was increased. X-ray diffraction (XRD) rocking curves widths decreased indicating the crystal quality improved with increasing growth temperature. The lowest XRC FWHM was 821 arcsec, which is so far the lowest value reported for ScN. Scanning electron microscopy (SEM) exhibited the formation of macrosteps and cracks on the crystal surface with latter due to the mismatch of ScN’s and tungsten’s coefficients of thermal expansion . In general for crystal growth, material should deposit on the seed crystal and not on any adjacent supporting structures. This efficiently uses the source material and avoids the possibility of spurious polycrystals encroaching on, and interfering with the single crystal growth. To achieve this goal, a new crucible design with a cooling fin in contact with the seed was simulated and experimentally demonstrated on the physical vapor transport (PVT) crystal growth of scandium nitride. The heat transfer of the growth cavity for a conventional crucible and a modified crucible with the cooling fin were modeled theoretically via computational fluid dynamics (CFD) with FLUENT. The CFD results showed that the seed in the modified crucible was approximately 10 °C cooler than the crucible lid, while in the conventional crucible the temperature of the seed and lid were uniform. The experimental results showed that increasing the temperature gradient between the source and the seed by employing the cooling fin led to a dramatic increase in the growth rate of ScN on the seed and reduced growth on the lid. The relative growth rates were 80 % and 20 % on the seed and lid respectively, in the modified crucible, compared to 25% and 75% with the conventional crucible. Thus, the modified crucible improved the process by increasing the species transporting to the seed by sublimation.

No description available.

In the last few years research has identified Photonic Crystals (PhCs) as promising material that exhibits strong capability of controlling light propagation in a manner not previously possible with conventional optical devices. PhCs, otherwise known as Photonic Bandgap (PBG) material, have one or more frequency bands in which no electromagnetic wave is allowed to propagate inside the PhC. Creating defects into such a periodic structure makes it possible to manipulate the flow of selected light waves within the PhC devices outperforming conventional optical devices. As the fabrication of PhC devices needs a high degree of precision, we have to rely on accurate numerical modelling to characterise these devices. There are several numerical modelling techniques proposed in literature for the purpose of simulating optical devices. Such techniques include the Finite Difference Time Domain (FDTD), the Finite Volume Time Domain (FVTD), and the Multi-Resolution Time Domain (MRTD), and the Finite Element (FE) method among many others. Such numerical techniques vary in their advantages, disadvantages, and trade-offs. Generally, with lower complexity comes lower accuracy, while higher accuracy demands more complexity and resources. The Complex Envelope Alternating Direction Implicit Finite Difference Time Domain (CE-ADI-FDTD) method was further developed and used throughout this thesis as the main numerical modelling technique. The truncating layers used to surround the computational domain were Uniaxial Perfectly Matched Layers (UPML). This thesis also presents a new and robust kind of the UPML by presenting an accurate physical model of discretisation error. iv This thesis has focused on enhancing and developing the performance of PhC devices in order to improve their output. An improved and new design of PhC based Multiplexer/Demultiplexer (MUX/DEMUX) devices is presented. This is achieved using careful geometrical design of microcavities with respect to the coupling length of the propagating wave. The nature of the design means that a microcavity embedded between two waveguides selects a particular wavelength to couple from one waveguide into the adjacent waveguide showing high selectivity. Also, the Terahertz (THz) frequency gap, which suffers from a lack of switching devices, has been thoroughly investigated for the purpose of designing and simulating potential PhC based switching devices that operate in the THz region. The THz PhC based switching devices presented in this thesis are newly designed to function according to the variation of the resonant frequency of a ring resonator embedded between two parallel waveguides. The holes of the structures are filled with polyaniline electrorheological fluids that cause the refractive index of the holes to vary with applied external electric field. Significant improvements on the power efficiency and wavelength directionality have been achieved by introducing defects into the system.

No description available.

No description available.

Nonlinear optical phenomena associated with high-order soliton breakup in photonic crystal fibres and squeezed state generation in three dimensional photonic crystal microcavities are investigated. In both cases, the properties of periodically patterned, high-index contrast dielectric structures are engineered to control the dispersion and local field enhancements of the electromagnetic field. Ultra-short pulse propagation in a polarization-maintaining microstructured fibre (with 1 um core diameter and 1.1 m length) is investigated experimentally and theoretically. For an 80 MHz train of 130 fs pulses with average propagating powers in the fibre up to 13.8 mW, the output spectra consist of multiple discrete solitons that shift continuously to lower energies as they propagate in the lowest transverse mode of the fibre. The number of solitons and the amount that they shift both increase with the launched power. All of the data is quantitatively consistent with solutions of the nonlinear Schrodinger equation, but only when the Raman nonlinearity is treated without approximation, and self-steepening is included. The feasibility of using a parametric down-conversion process to generate squeezed electromagnetic states in 3D photonic crystal microcavity structures is investigated for the first time. The spectrum of the squeezed light is theoretically calculated by using an open cavity quantum mechanical formalism. The cavity communicates with two main channels, which model vertical radiation losses and coupling into a single-mode waveguide respectively. The amount of squeezing is determined by the correlation functions relating the field quadratures of light coupled into the waveguide. All of the relevant model parameters are realistically estimated using 3D finite-difference time-domain (FDTD) simulations. Squeezing up to ~20% below the shot noise level is predicted for reasonable optical excitation levels. To preserve the squeezed nature of the light generated in the microcavity, a unidirectional coupling geometry from the microcavity to a ridge waveguide in a slab photonic crystal structure is studied. The structure was successfully fabricated in a silicon membrane, and experimental measurements of the efficiency for the signal coupled out of the structure are in good agreement with the result of FDTD simulations. The coupling efficiency of the cavity mode to the output channel is ~60%. / Science, Faculty of / Physics and Astronomy, Department of / Graduate

The area of quantum information promises to deliver a range of new technologies in the fields of quantum computing and quantum communication. Devices based on semiconductor quantum dots hold great potential for the practical realisation of many of the components required in the proposed schemes. This thesis describes the development of several quantum dot devices. By integrating a quantum dot into a p-i-n diode, it was possible to control the dominant emission lines in its photoluminescence spectrum and to maximise the degree of polarisation correlation between the two photons emitted in the biexciton decay. With the same device under a magnetic field, a digital memory was demonstrated. The polarisation information of a single photon was stored as the spin of an electron inside the quantum dot, and was deterministically recovered some time later by the application of an electrical trigger. A fabrication process was developed in order to produce high quality two dimensional slab photonic crystals operating with a photonic band gap at ~ 900 nm. By placing a quantum dot into an appropriately designed H1 photonic crystal cavity, strong coupling was achieved between the dot and the monopole mode of the cavity. The vacuum Rabi splitting was found to be constant for all linear polarisations due to the unpolarised nature of the far-field of the mode. Finally, a new kind of cavity based on photonic crystal waveguides was developed. A Purcell enhancement of the in-plane spontaneous emission from a quantum dot coupled to a unidirectional photonic crystal waveguide was demonstrated.

Hyperuricemia and local temperature changes in the joints of the extremities are known to be responsible, in part, for the development of gouty arthritis. No satisfactory explanation is yet available for (1) the selective deposition of monosodium urate monohydrate (MSUM) crystals in connective tissues (2) the increased incidence of gout in the later years of life and (3) the increased incidence of MSUM crystal deposition in connective tissues after trauma and in joints with preexisting disease. It is possible that the alterations in composition of the non-fibrillar matrix of cartilage and synovial fluid which are thought to occur with ageing, trauma or preexisting disease, may predispose these tissues to crystal deposition. The objectives of this study were to determine the effect of the cartilage and synovial fluid components, chondroitin sulfate, hyaluronic acid, proteoglycan monomer, proteoglycan aggregate, phospholipids and albumin on the growth of MSUM. The degradation of MSUM solutions was studied under sterile and non-sterile conditions to determine the possible causes of degradation and to define the time span of crystal growth experiments. The rate of degradation of MSUM solutions increased with an increase in temperature. The concentration of MSUM in solution fell sharply after autoclaving and solutions stored in containers with rubber closures showed greater degradation of MSUM than autoclaved solutions stored in all-glass containers. Rubber stoppers apparently absorbed MSUM from solution. The degradation of MSUM solutions was thought to be due to both bacterial consumption and chemical decomposition in non-sterile solutions but was due only to chemical decomposition in sterile solutions. The aqueous solubility of MSUM was determined at different temperatures and in the presence of varying concentrations of sodium chloride. Sodium chloride suppressed MSUM solubility. The aqueous solubility of MSUM was also determined in the presence of several connective tissue components at 37°. Chondroitin sulfate (CS) decreased the saturation solubility of MSUM probably due to the sodium present in the CS samples. Proteoglycan aggregate, proteoglycan monomer, hyaluronic acid an albumin resulted in very slight increases in the solubility of MSUM. The growth kinetics of MSUM was studied using the seeded growth technique. An equation of the general form: [formula omitted] was used to determine the overall growth rate constant, [formula omitted]. Linear plots of the integrated form of the second order growth equation gave the best fit between the points and gave reasonably constant values for [formula omitted] determined at a given initial supersaturation concentration and varying seed amounts. An induction period or a period of slow growth was observed at both the initial supersaturation concentrations studied. The length of the induction period was inversely proportional to the added seed amount. Differing concentrations of additives were included in the growth medium and K' determined. Chondroitin sulfate (CS) significantly increased the growth rate constant for MSUM growth. However, the proportion of CS decreases in aged and osteoarthritic cartilage and thus a decreasing proportion of a growth accelerator is unlikely to be a factor in the deposition of MSUM in cartilage. CS has been found in the synovial fluid of arthritic joints and may act as an MSUM growth accelerator in this medium. Hyaluronic acid (HA) and albumin caused significant inhibition of the growth of MSUM crystals. This effect may be due to the adsorption of these molecules onto the MSUM seed crystals resulting in the poisoning of the active growth sites on the crystal surface. Cartilage HA and synovial fluid albumin levels are increased in aged and/or diseased joints. Increased proportions of growth inhibitors do not offer likely explanations of crystal deposition in joint tissues. At concentrations of 0.1-1.0 mg mL⁻¹ proteoglycan monomer (PGM) and proteoglycan aggregate (PGA) slightly increased the MSUM growth rate constant but this increase was statistically insignificant. The two phospholipids, phosphatidylcholine and phospha-tidylserine increased the growth rate constant of MSUM. Phosphatidylserine, however, did not significantly increase the growth rate constant at the concentrations studied. It is possible that the raised levels of phospholipids in aged or diseased cartilage and synovial fluid could accelerate the growth of MSUM crystals resulting in MSUM deposition in these tissues. / Pharmaceutical Sciences, Faculty of / Graduate