Annite is the Fe rich end-member of the phlogopite-biotite-annite series of micas. The ideal structural formula of annite is: $\{K\sp+\}\lbrack Fe\sbsp{3}{2+}\rbrack\langle Al\sp{3+}Si\sbsp{3}{4+}\rangle O\sbsp {2-}(OH\sp-)\sb2.$ Two samples of annite have been studied, a synthetic, true end-member (in the form of a fine powder) and a natural, near end-member (in the form of single crystal). The samples have been studied using SQUID magnetometry and Mossbauer spectroscopy at temperatures ranging from 4.2 K to 295 K using a purpose-built Mossbauer cryostat and insert. In disagreement with earlier predictions it has been determined that the magnetic ground state of annite consists of planes of predominantly ferromagnetic spins stacked antiferromagnetically along the c* axis. In further contrast to earlier predictions, it has been found that, for annite, the ordering temperature, T$\sb{\rm N}$, is greater than 10 K, with the synthetic sample having T$\sb{\rm N}$ = 58 K. It is proposed that the planar antiferromagnetic structure is part of the intrinsic zero-field magnetic domain structure and that this structure is stabilized by inter-layer dipole-dipole forces. Mossbauer spectroscopy measurements indicate some striking features of the temperature development of magnetic state of both annite samples. In particular, the results indicate that there is a persistent paramagnetic Fe$\sp{2+}$ contribution at temperatures as low as 0.1 T$\sb{\rm N}$; and that the Fe$\sp{3+}$ moments disorder at temperatures significantly below T$\sb{\rm N}.$ Plausible causes for these features are discussed. The first documented attempt to fit the low temperature spectrum of annite is reported. A method for stripping the Fe$\sp{3+}$ contribution from the liquid helium spectrum of synthetic annite is discussed and a hierarchy of simple models is constructed and applied in attempt to fit the remaining Fe$\sp{2+}$ subspectrum. An acceptable fit is not obtained. The consequences of the failure of the simple models are examined.

The photovoltaic effect was used to study the excitonic transitions in $\rm In\sb{x}Ga\sb{1-x}As/GaAs$ single quantum wells. This work describes at first, the evolution of the exciton observed in photocurrent spectra as a function of the applied electric field. The effect shows a shift of the transition towards longer wavelengths and forbidden transitions ($\rm n \ne n\sp\prime$) are detected. A decrease of the integrated intensity and a broadening of the 11H transition are also observed. Some measurements were taken by simultaneously changing two parameters, the temperature and the electric field. We observed a convolution between the two external influences, a thermal and electric ionization of the excitons that requires a strong electric field at low temperature or a weak field at high temperature. This study then describes the evolution of the 11H exciton (observed in photovoltage and photocurrent spectra) as a function of temperature. The spectra obtained show a large increase of the integrated intensity of the excitonic line between 6K and 87K, and no increase at higher temperature. This is due to the thermal ionization of excitons in the well. In the second chapter, the photovoltaic signal attributed to high density excitons dissociating at a metal-semiconductor interface in a cuprite (Cu$\sb2$O) crystal is measured. The results indicate possible condensation of the excitons and superfluid transport through the three dimensional crystal. The process was repeated for a two dimensional configuration, a $\rm In\sb{x}Ga\sb{1-x}As/GaAs$ single quantum well. We observed a variation in the carrier velocity and intensity as a function of the bi-dimensional excitonic particles density, the sample temperature and the excitation distance. From the results there is no evidence of a phase transition. However, some diffusive transport properties such as the carrier velocity was calculated and compared to previous measurements obtained using different methods.

For the most part of this work, we study the effects of focused ion beam implantation in InGaAs/GaAs quantum well structures. The technique of steady-state photoluminescence spectroscopy is utilized to study the compositional disordering occurring in InGaAs/GaAs quantum well structures having been implanted at different doses and subsequently annealed. Photoluminescence results of uniformly implanted regions of InGaAs/GaAs samples are along with a simple formalism permitting the calculation of the interdiffusion lengths. These results, along with SIMS measurements, show that channeling effects play a most important role in disordering quantum wells that lie deep beneath the sample's surface. The observations also show a photoluminescence shift saturation with dose at approximately 2 $\times$ 10$\sp $ Ga$\sp{+}$/cm$\sp2$. Results of pattern-implanted samples give rise to the question of the minimum dose required to induce uniform compositional disordering in InGaAs/GaAs structures. Studies of this "critical" dose, are presented for both AlGaAs/GaAs and InGaAs/GaAs quantum well structures and show that the minimum dose required to induce intermixing in InGaAs/GaAs is approximately two orders of magnitude lower than that observed for the AlGaAs/GaAs structures which was 2 $\times$ 10$\sp $ Bi$\sp{+}$/cm$\sp2$. As another major part of this work, we describe a novel, low light level optical detection system that can be easily configured for various modes of operation. These include (1) time-integrated (steady-state) photoluminescence spectroscopy, (2) transient, spectrally gated photoluminescence decay, (3) time-windowed photoluminescence spectroscopy, (4) two-dimensional, time-integrated photoluminescence mapping, and (5) time-resolved, two-dimensional photoluminescence mapping with time resolution of ${\sim}$100 ps. We use the system in mode (5) to study electron-hole pair expansion in GaAs quantum wells as a function of temperature for a fixed excitation intensity. The results show that the diffusion coefficient rises from an initial value of 26 cm$\sp2$/s at 80 K to 38 cm$\sp2$/s at ${\sim}$100 K and then monotonically decreases to a value of ${\sim}$10 cm$\sp2$/s at room temperature. The rise in the diffusion coefficient is attributable to the gradual decrease of interface roughness scattering with increasing temperature while the decrease of the diffusion coefficient is attributable to the increase of phonon scattering mechanisms with increasing temperature. Results of diffusion at fixed temperature, but as a function of excitation intensity, are also presented and discussed.

Many workers visualize that, when magnetic flux first penetrates into type II superconductors, the flux line lattice exhibits a discontinuity of magnitude comparable to the lower critical field $H\sb{c1}$ over a dimension of the order of the penetration depth $\lambda$ along the front of the invading flux line lattice. This discontinuity is referred to as the dB/dH effect in the literature. Such a discontinuity is also thought to exist in the magnetic flux configuration (B profile) at the periphery of the specimen when the applied magnetic field $H\sb{a}$, which caused the flux lines to enter the specimen, is then removed. We have compared $\langle B\rangle\sb{rem}$, the amount of magnetic flux trapped in a sample after it has been subjected to an excursion (half cycle) of $H\sb{a}$ with $\langle B\rangle\sb{in}$, the amount of flux permeating the specimen at the peak of a half cycle of $H\sb{a}$. Such measurements were performed on six conventional type II superconductors in the form of long solid cylinders and ribbons. These observations are compared with critical state model predictions which incorporate discontinuities in the B profile of various magnitude ranging from zero to $B\sb{c1}$ and beyond. We show that the presence of any such discontinuities will be dramatically evident in graphs of $\langle B\rangle\sb{rem}$/ $\langle B\rangle\sb{in}$ versus $\langle B\rangle\sb{in}$. Plots of the data in this format indicate quite clearly that discontinuities are not present.

$\rm Mn\sb2SiS\sb4$ is a synthetic Olivine in which we have discovered an exceptional temperature dependent magnetic behavior that may lead to interesting applications. A state of spontaneous magnetization was found to exist in this material, but only between 83 and 86.5 K. In order to explain this unusual magnetic phenomena, polycrystalline samples of $\rm Mn\sb2SiS\sb4$ were first prepared from the elements and studied by SQUID magnetometry in low fields, by X-ray and neutron diffraction, by electron spin resonance (ESR) and by specific heat calorimetry. This research, in which the author was a member of a team, is described. His specific contributions include among others; the debugging and extension of some of the computer programs used to analyze the neutron diffraction results, the interfacing of the Bruker ESR spectrometer to a IBM compatible computer and the development of the computer progams to acquire the resonance curves and analyze the line spectra, the analysis of the ESR results, the assembly of parts of an apparatus and the development of a procedure to measure the specific heat of small samples in the 77 to 150 K range as well as some preliminary measurements on a sample of $\rm Mn\sb2SiS\sb4.$ These contributions are emphasized in this thesis. (Abstract shortened by UMI.)

We present strong evidence that helical flux lines can concurrently, enter and leave the wall of hollow cylinders of sintered high superconductors at T$\sb{\rm c}$ 77 K, hence cut and traverse each other. Tubes of two different materials, denoted BiSCCO and YBCO, have been studied. The evidence for flux line cutting and cross-flow is examined in the perspective of basic concepts. The traffic of the flux lines is also described formally. The penetration fields across the wall and into the grains of the sintered ceramic tubes are determined from their magnetic response to an applied axial field H$\sb{//}$ and from their flux trapping behaviour when subjected to two standard procedures, denoted H$\sb{\rm cool}$ and H$\sb{\rm cycle}$. This provides information on j$\sb{\rm c\perp m}$ and j$\sb{\rm c\perp g}$, the intergrain and intragrain critical current densities for depinning of flux lines. The penetration fields are seen to correlate with salient features (peaks and valleys) of the flux line cutting and cross flow phenomena. This enables us to claim that we are witnessing flux line cutting and cross flow in the weak link regime in the BiSCCO tube and in this regime as well as in the interior of the grains in the YBCO tube. The Clem/Perez-Gonzalez phenomenological theory is exploited in a simplified framework to describe the crucial features of the data semi-quantitatively. This analysis confirms the above conclusions and provides estimates of j$\sb{\rm c//m}$ and j$\sb{\rm c//g}$, the critical current densities for intergranular and intragranular flux line cutting and their dependence on the magnetic flux density.

No description available.

A RHEED system was assembled and incorporated in the existing ultra-high-vacuum chamber. A system for the production and transport of low energy ions to the ultra-high-vacuum environment was maintained. Samples cut from wafers of Si(100) were investigated using electron diffraction as well as low energy ion scattering techniques. The surface composition, symmetry and orientation of these specimens could be determined. Attempts to adsorb hydrogen onto the Si(100) surface indicated that the procedure in use needs to be modified. An initial RHEED study of the Si(110) surface revealed a complex arrangement of atoms which does not correspond to the "16 x 2" structure reported for the clean surface.

The constituent layers of a semiconductor quantum well system were studied systematically and sequentially using various spectroscopic and optoelectronic techniques such as photoluminescence, photoreflectance, photovoltage, electroreflectance, or electroluminescence. Complete single quantum well samples with various well parameters were then studied with the appropriate technique. The purpose of the study was to find how the energy levels (the bound states as well as the continuum states) of the quantum well structure were influenced by the surroundings (such as the cap layer, the buffer or the substrate) and the well parameters. To that effect, the well width and the alloy composition (x) of an In$\sb{\rm x}$Ga$\sb{\rm 1-x}$As/GaAs single quantum well were varied, and the results obtained with samples having different cap thicknesses (the other parameters being kept the same) were compared. A theory predicting the influence on the energy levels of the cap layer thickness was developed, and experimental evidence was found in modulated reflectance measurements which showed spectral oscillations for energies above the barrier. The behavior is explained by deriving the wavefunction of the carriers with energies larger than the barrier (continuum states) taking into account the finite size of the cap layer. It was also found in the sequential study that in luminescence, depending on the excitation used, the substrate can contribute to the response obtained with the complete device. The influence of an external perturbation such as an electric field was also studied. Electroreflectance spectra showed that the bound states were Stark shifted, while the peaks of the cap-related oscillations of the continuum states were shifted linearly in energy with the electric field. Study of a multiple quantum well under the same conditions showed that the cap-related oscillations of the continuum states were strengthened by the presence of the additional wells. The theory also predicts that the first well(s), closer to the surface, will be more energy selective in capturing carriers in the continuum. This may result in different capture efficiencies for the various wells of a multiple quantum well structure.

We have studied the oxidation in air of a well characterized biotite sample. Large single crystal wafers were annealed at various temperatures up to 875$\sp\circ$C and for various times up to 94 hours. The oxidation proceeds primarily via the oxyannite reaction: $\rm (Fe\sp{2+} + OH\sp-)\sb{mica} \to (Fe\sp{3+} + O\sp{2-})\sb{mica} + H$ and was therefore conveniently followed by $\sp{57}$Fe Mossbauer spectroscopy, which resolves the 2+ and 3+ valence states of iron. The main features of the annealing time and temperature dependencies of the $\rm Fe\sp{3+}/Fe\sp{2+}$ amounts are understood in terms of a simple model in which: (i) the overall oxidation reaction proceeds homogeneously via a time-wise bottleneck step that follows a classic thermal activation law and (ii) a certain fraction of the original Fe$\sp{2+}$ is inaccessible to the oxidation reaction. The resulting barrier energy, $\rm E\sb{b} = 2.36\sbsp{-.02}{+.01}$ eV is in the range of measured barrier energies for dehydroxylation of layer silicates. This suggests that the bottleneck step may be local dehydroxylation: $\rm (OH\sp-\ \to O\sp{2-}\ + H\sp+).$ The persistent Fe$\sp{2+}$ can be understood from local crystal-chemical considerations. (Abstract shortened by UMI.)