We developed a technique for measuring the focal properties and the magnification behaviour oi electrostatic lenses with cylindrical geometry. We applied the method to the study of three-element lenses of various proportions, and we found excellent agreement between our experimental results and a simple theoretical model of lenses with centre elements longer than 1.25 internal lens diameters. We extended our study to lenses with centre element 1 and 0.5 lens diameters where the simple theoretical model is inappropriate but there are detailed calculations on the focal properties and the magnification of such lenses. Again we found very good agreement. Having established a sound measuring technique we studied the properties of lenses with controlled magnification. We constructed a five-element lens whose linear magnification can be varied over a range of 9:1 at a given overall voltage ratio. The same lens can also be operated to give a constant magnification over an extremely wide range (10000:1) of overall voltage ratios. Such a lens can prove a powerful tool in electron spectroscopy for uses in electron guns and as the output lens of monochromators. The potential distribution required for a specific purpose can be obtained simply from design curves. By imposing certain constraints on the focusing voltages of the five-element lens we can form a system with a magnification that depends only on the overall voltage ratio as (Vs/V1)-0.2S and with the property of forming an image at a fixed distance from an object. We have made an experimental study of such a lens and found that a prediction of a simple scaling law for the potentials was valid. We developed two independent methods for measuring the spherical aberration of some of the above lenses. We have measured the third order coefficients Cs, Cso, Cs4 for two and three-element lenses of various proportions and the coefficient Cs for a five-element afocal lens. We have observed fifth order effects for large launching angles. The object of this thesis was to develop and use a technique to investigate the focal properties, magnification and spherical aberration of electrostatic lenses.

Using a computer model of a two cylinder electrostatic lens, some novel relationships have been found to exist between the input and output parameters of meridional rays. These relationships have been developed and used to show that, for a wide range of practical lens geometries, it is possible to represent all the third and fifth order aberrations in terms of just two of the normal parameters. Formulae have been derived to describe some of the quantities associated with this type of lens defect and the problems of minimising the aberrations are discussed.

The development of the study of light scattering, from the early work of Stokes and Rayleigh to the modern investigations of non-linear stimulated scattering processes has been reviewed. The conventional theory of stimulated scattering has been extended to describe the scattering of an independent probe laser beam. This scattering is the result of the effect on the refractive index of a medium of the electric field due to two oppositely directed, high intensity light beams. The standing wave of electric field, generated by these beams, spatially modulates the refractive index of the medium by electrostriction, the Kerr effect and absorptive heating, all of which arc proportional to the mean square of the local field. When the beams are of slightly different frequency the standing wave, and the resulting modulation, travel through the medium with a velocity proportional to that difference. Under these circumstances the modulation and the standing wave travel through the medium with the same velocity but with different phase as a result of the inertia associated with each of the above mechanisms. The usual stimulated scattering is determined entirely by the out of phase component of the refractive index modulation while the probe scattering described in this thesis is determined by the whole amplitude of that modulation. A Q-switchd ruby laser has been used to generate such modulations in a liquid and these have been detected by the scattering of a frequency doubled portion of the ruby laser beam and of an independent argon laser beam. The refractive index modulation associated with the temperature modulation induced by absorption of the ruby laser light has Leen investigated in some deta.il and the lifetimes of such modulations in a number of media have been measured. Other related non-linear scattering processes have been investigated using a variety of experimental techniques, particularly interesting results being obtained in the study of non-linear fluorescence from saturable absorbers.

M.Ing. / This thesis covers the theory and development of fused, biconically tapered, fibre-optic couplers and their application in sensors. The optical coupling between fibres is described by means of the coupled-mode theory in the case of a weakly guiding coupler. Firstly, simplified expressions for the guided modes in an optical fibre are derived. This is followed by the derivation of the coupled-mode equations for a perturbed waveguide. The coupled-mode equations are then used, together with the expressions for the guided modes, to determine the coupling coefficient for a weakly guiding coupler. In the case where the coupler is not weakly guiding, a model based on a rectangular dielectric waveguide is presented. A process for the manufacture of fibre-optic couplers, based on the fusion and elongation technique, was developed. This process is described and special attention is paid to the influence which different process parameters has on the properties of the couplers. During the development, the coupling theory was utilized to optimize the process. Results obtained with couplers manufactured by this process, are presented. A specific application of couplers, namely a new type of fibre-optic coupler sensor, was developed. The operation of the sensor relies on the change of the coupling ratio with an axially applied mechanical strain. Analog signal processing of the two outputs yields a signal which is directly proportional to the strain and which is independent of fluctuations of the light source power. The sensor was investigated experimentally by manufacturing temperature and elongation sensors which work on this principle. The operation and features of these sensors are presented, together with very promising experimental results.

The multi-configuration time-dependent Hartree-Fock (MCDTHF) method is a promising new method that can solve the time-dependent Schrodinger equation (TDSE) for a correlated multielectron system. We have written a computer code to solve the MCTDHF equations of motion. This code is highly optimized for speed and makes efficient use of memory. As a first test of the code, we will investigate the ionization dynamics of N2 in a strong laser-electric field using two active electrons. By examining the final momenta, parallel to the field polarization direction, of the two ionized electrons we are able to determine: (i) which laser dressed excited state the second electron was in at the time of ionization and (ii) the birth times of the electrons in the field. The results of these calculations provide a means for characterizing the attosecond dynamics of atoms and molecules without using attosecond pulses.

In a distributed Brillouin sensor system, it is crucial to keep the pulse energy uniform for constant signal to noise ratio. This means that the variable DC leakage (pulse base) through the electro-optic modulator must be locked. In this thesis I examine two different methods of locking the pulse base level and look at the advantages and disadvantages of each locking method. It is found that the two locking methods, one based on a lock-in amplifier and the other using proportional-integral-derivative control, both have applications in which they excel at locking the pulse base. Also, a technique to simultaneously lock the pulse base, top, and width is developed and tested. In the field of structural health monitoring, it is often advantageous to monitor the dynamic behaviour of a structure in real-time. The traditional distributed Brillouin sensor does not allow for this dynamic measurement due to the need to sweep the frequency difference between the two lasers and subsequent averaging of waveforms. For the first time to our knowledge, a real-time vibration sensor based on polarization-state perturbations in stimulated Brillouin scattering instead of resonant frequency mismatching monitoring of the Brillouin spectrum has been proposed. The long measurement time of traditional distributed Brillouin sensors is avoided by eliminating the frequency sweep of the pump and Stokes lasers and instead locking them at a single beat frequency corresponding to the static strain of the structure in which the fiber is embedded. This unique sensor allows measurement of vibration frequencies along a sensing fiber as shown in laboratory experiments and also the detection of impact waves from passing vehicles in field tests in which the sensor was embedded in the concrete pavement of a highway. A 20 ns pulse width with potential spatial resolution of 2 m was used over a sensing length of 300 m. Also, studies of the Brillouin linewidth under cw pump and Stokes waves are done in order to confirm the validity of a new definition of the threshold power in Brillouin fiber amplifiers which involve both input pump and input Stokes waves - all previous threshold definitions took into account only the input pump power. Finally, some interesting lineshape characteristics such as spectral hole burning and side-lobes on the Brillouin spectrum are observed for high power 2 ns Stokes pulses and their origins explained qualitatively. The evolution of these features with increasing pump power is investigated. The effect of laser linewidth and fiber length on these features is also considered.

This thesis explores the production, characterization, and water oxidation efficiency of wide bandgap semiconductors made through femtosecond-laser irradiation of various materials. Our investigation focuses on three main aspects: 1) producing titanium dioxide (TiO2) from titanium metal, 2) using our laser-made materials in a photoelectrochemical cell for water oxidation, and 3) utilizing the femtosecond laser to create a variety of other mixed metal oxides for further water oxidation studies and biological applications. We first discuss producing TiO2 and titanium nitride. We report that there is chemical selectivity at play in the femtosecond laser doping process so not all dopants in the surrounding atmosphere will necessarily be incorporated. We then show that the material made from laser-irradiation of titanium metal, when annealed, has a three-fold enhancement in overall water oxidation when irradiated with UV light. We attribute this enhancement through various material characterization methods to the creation of a more pure form of rutile TiO2 with less defects. We then present a variety of studies done with doping both TiO2 and other oxides with broadband photoelectrochemistry and offer that the dopant incorporation hurts the overall water oxidation rate. Lastly, we use the laser-treated titanium to test cell adhesion and viability. Our results demonstrate an ability to femtosecond-laser process semiconductors to produce materials that no one has made previously and study their properties using collaborations across chemistry and biology, yielding true interdisciplinary research.

Gallium nitride (GaN) microcavities with embedded optical emitters have long been sought after as visible light sources as well as platforms for cavity quantum electrodynamics (cavity QED) experiments. Specifically, materials containing indium gallium nitride (InGaN) quantum dots (QDs) offer an outstanding platform to study light matter interactions and realize practical devices, such as on-chip light emitting diodes and nanolasers. Inherent advantages of nitride-based microcavities include low surface recombination velocities, enhanced room-temperature performance (due to their high exciton binding energy, as high as 67 meV for InGaN QDs), and emission wavelengths in the blue region of the visible spectrum. In spite of these advantages, several challenges must be overcome in order to capitalize on the potential of this material system. Such diffculties include the processing of GaN into high-quality devices due to the chemical inertness of the material, low material quality as a result of strain-induced defects, reduced carrier recombination effciencies due to internal fields, and a lack of characterization of the InGaN QDs themselves due to the diffculty of their growth and therefore lack of development relative to other semiconductor QDs. In this thesis we seek to understand and address such issues by investigating the interaction of light coupled to InGaN QDs via a GaN microcavity resonator. Such coupling led us to the demonstration of the first InGaN QD microcavity laser, whose performance offers insights into the properties and current limitations of the nitride materials and their emitters. This work is organized into three main sections. Part I outlines the key advantages and challenges regarding indium gallium nitride (InGaN) emitters embedded within gallium nitride (GaN) optical microcavities. Previous work is also discussed which establishes context for the work presented here. Part II includes the fundamentals related to laser operation, including the derivation and analysis of the laser rate equations. A thorough examination of the rate equations serves as a natural motivation for QDs and high-quality factor low-modal volume resonators as an optimal laser gain medium and cavity, respectively. The combination of the two theoretically yields the most efficient semiconductor laser device possible. Part III describes in detail the design, growth, fabrication and characterization of the first InGaN QD microcavity laser. Additional experiments are also conducted in order to conclusively prove that the InGaN QDs serve as the gain medium and facilitate laser oscillation within the microdisk cavities. Part III continues with work related towards the development of the next generation of nitride light emitting devices. This includes the realization of photonic crystal cavity (PCC) fragmented quantum well (FQW) lasers that exhibit record low lasing thresholds of 9.1 uJ/cm2, comparable to the best devices in other III-V material systems. Part III also discusses cavity QED experiments on InGaN QDs embedded within GaN PCCs in order to quantify the degree of light-matter interaction. The lack of experimental evidence for weak or strong coupling, in the form of the Purcell Effect or cavity-mode anti-crossing respectively, naturally motivates the question of what mechanism is limiting the device performance. Part III concludes with cathodoluminesence and tapered fiber measurements in order to identify the limiting factor towards achieving strong coupling between InGaN QDs and GaN microcavities.

Three-dimensional confinement of carriers eliminates the problem of thermal spreading of carriers observed in higher-dimensional systems. Uniform self-assembled quantum dots (QDs) are obtained using the spontaneous islanding of highly strained III-V semiconductors grown with standard epitaxy. Visible stimulated emission has been obtained with red-emitting AlInAs QDs in AlGaAs barriers. Continuous (CW) threshold current densities below 100A/cm2 have been measured at low temperatures and QD material gain larger than 1.7 x 104 cm-1 demonstrate good material quality. Room temperature lasing has also been observed for higher threshold current densities. For longer wavelengths where the thermionic emission problem is less important, InAs/GaAs lasers can operate at room temperature for current densities below ∼100A/cm2 for wavelengths around 950 nm. The zero-dimensional transitions between confined electrons and holes in artificial atoms allow the observation of state-filling at relatively low level of material excitation. Lasing is observed in the upper QD shells for small gain media, and progress towards the QD ground states for longer cavity lengths. Gain may also be increased by including multiple layers of QDs in the active region. To understand the shell structure of AlInAs/AlGaAs QDs, we present results of interband spectroscopy of single Al0.36In0.64As/Al 0.33Ga0.67As self-assembled QDs. The single dot spectroscopy has been carried out at low temperature as a function of the excitation power and magnetic field up to 8 T. The emission spectra as a function of excitation power show two distinct groups of transitions which we associate with the recombination from ground and excited QD levels with a spacing of ∼70 meV. The application of magnetic field allows to identify the exciton emission as well as the emission from the bi-exciton, and charged exciton complexes with binding energies of ∼5 meV. The binding energies compare favorably with results of calculations. Artificial molecules are studied using coupled QD ensembles and single QD spectroscopy. The coupling between the zero-dimensional states is varied by changing the distance between two layers of stacked InAs/GaAs QDs. Energy level splitting larger than 30 meV of the symmetric and anti-symmetric states of the lowest confined shell are measured and are compared to theory.

The thesis consists of two research directions: Optical fiber applications in gas concentration and radiation dose measurements; and polarization effects in fiber optic communication systems. Part I of the thesis presents two optical fiber applications. (1) An infrared (IR) fiber bundle has been designed and fabricated to measure gas concentrations in a chemical vapor deposition (CVD) chamber using Fourier transform infrared spectroscopy. This fiber bundle covers the IR range from 0.5 to 20 mum and reduces the light beam divergence in the CVD chamber, which makes it possible to measure gas concentrations in a region near the substrate surface. Semi-ellipsoid mirrors have been designed and used to increase the collection efficiency of infrared radiation and to compensate the loss introduced by the fiber bundle. (2) A fiber optic radiation sensor based on radiation-induced fiber loss is reported. The gamma radiation-induced loss spectra in various fibers have been studied. Among all the fibers tested, 5% P-doped fiber shows the highest sensitivity to gamma radiation. The wavelength and dose rate dependence of radiation-induced loss in 5% P-doped fiber are investigated and the possibility of using this fiber as a radiation sensor for radiation therapy is discussed. Part II of the thesis examines two polarization effects, polarization mode dispersion (PMD) and polarization dependent loss (PDL), in fiber optic communication systems based on the waveplate models. A new waveplate model, capable of generating any PMD and PDL values, is proposed to overcome the limitations of the conventional waveplate model. Using both models the statistical distributions of PDL and differential group delay (DGD) have been studied considering the presence of biased elliptical birefringence. The principal state of polarization (PSP) of an optical pulse is proposed for a fiber having both PMD and PDL. PMD and PDL of a pulse for a fiber consisting of two polarization maintaining fiber segments are calculated, and the pulse distortions due to PMD and PDL are analyzed. PMD and PDL impact on digital communication systems have been studied in terms of system Q factor, bit error rate (BER) and system outage probability. The acceptable PMD values in a system with PDL are discussed for various power margins. The Q factor distributions due to PDL and its combination with PMD have been measured experimentally and compared to numerical simulations.