The Hall coefficient and conductivity of silicon films vacuum deposited on 0° and 60° sapphire at 850°C to 1050°C were measured from 100°K to 550°K. Films made from 0.094 Ω-cm p type and 1 Ω-cm n type silicon sources were prepared by electron bombardment heating of the source in a vacuum of 5 x 10⁻⁷ to 10⁻⁶ torr. The orientation and crystallinity of the films were investigated using electron diffraction. It was found that defects in the films introduced both donor and acceptor levels. The heavy compensation thus produced in films deposited at lower temperatures lead to a very low hole concentration. All films were p type at room temperature showing that the acceptor levels slightly dominated the donor levels. The films deposited on 0° sapphire indicated fewer defects than those deposited on 60° sapphire. At high temperature (> 950°C) doping of the silicon by aluminum atoms from the substrate was appreciable. The Hall mobility of the films made from the p type source material decreased with increasing temperature of deposition. This apparent anomaly is explained by the use of the polycrystalline film model suggested by Volger (1950). / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

The technique of ellipsometry was applied to the study of nonuniform anodic oxide films resulting from electrolyte incorporation into the oxide on growth. Ellipsometry results obtained in air and in situ on oxides formed on Ta in phosphoric acid are consistent with the results of tracer studies in which the oxide consists of two layers which grow simultaneously due to metal and oxygen ion transport during anodization, and further that electrolyte incorporation into the outer layer on growth modifies its properties with respect to the inner layer. The ellipsometry results were not consistent with a single homogeneous layer film or with a film possessing an index of refraction changing linearly with thickness. Thus ellipsometry provides a new, nondestructive method of determining ion transport numbers. The index of refraction and thickness of each layer were obtained by curve fitting ellipsometry results obtained as a function of increasing oxide thickness. Computer methods for solving the ellipsometry equation and curve fitting are given. Computed results are given and discussed for cases of one and two layer films growing on a metal. Finally, an error analysis of ellipsometry is made. Ellipsometry results were obtained and curve-fitted for oxides grown in various solutions, at different rates and for anodization in a sequence of electrolytes. The major findings of this part of the study are as follows: Electrolyte incorporation decreases ionic conductivity, dielectric constant and index of refraction. At constant current formation, the fraction of oxide made up by the outer layer increases with current density, and electrolyte concentration, and depends on previous formation of the oxide. The log J-E characteristics at constant voltage in dilute phosphoric acid are curved and occur at higher fields than those for dilute sulphuric acid. Analysis of the above results indicates that: a) the conduction process is bulk controlled b) ionic conduction and dielectric properties arise from the same process and c) that electrolyte incorporation is responsible for part if not all the curvature in the logJ-E characteristics of ionic conduction. Photo-stimulated growth at low electric fields was investigated by ellipsometry. The effect of radiation is to first modify the properties of the existing oxide after which photo-stimulated growth occurs accompanied by a build up of secondary current. The secondary photocurrent is ionic in nature and the radiation rather than the applied field is responsible for the generation of ions to sustain this current. The photo-grown oxide consists of two layer with the outer layer having a much lower index of refraction than normally grown oxide. The thermal recrystallization of stripped anodic oxide films was studied using transmission electron microscopy. Various diffraction patterns were obtained and analyzed. The major result in terms of ionic conduction is that electrolyte incorporation inhibits recrystallization, again, consistent with a decreased ionic mobility with incroporation. A critical test has been devised and applied to a recently proposed theory of ionic conduction, the dielectric polarization theory. This theory postulates that the autocatalytic build up of ionic current on applying a constant high field to the oxide is due to an internal field controlled process and that the rate of build up of polarization (P) towards its equilibrium value (P₀) is enhanced by the passage of current, J, given by dp/dt = AJ(P₀-P). It is shown that this theory predicts an increase in small signal capacitance during the passage of the transient. However, measurements indicate that the capacitance decreases. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

The dielectric polarization processes, conduction mechanisms and space charge effects occurring in tantalum / tantalum pentoxide / metal devices are investigated. The dielectric properties are analyzed on the basis of an ionic relaxation process with a nearly flat distribution of activation energies. This distribution leads to step response polarization currents following an inverse time law. The effect of an injected electronic space charge on the response of the device due to the removal of a step voltage is analyzed and results are given demonstrating this effect. The devices used exhibit a rectification behaviour. For tantalum positive the currents follow a Schottky law and for tantalum negative, the bulk Poole-Frenkel law. Hysteresis effects are observed as well as the effects of a space charge on the Schottky law currents. On one sample, sufficiently high fields causes an increase in the conductance by a factor of 10³ to 10⁵. For this deformed sample, no hysteresis or rectification is observed and the currents follow a Schottky law for both polarities. The validity of the ionic relaxation model is discussed in light of the observed dielectric losses at low temperatures. The experimental results indicate that an electronic rather than an ionic process could be responsible for the dielectric losses. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

Nuclear magnetic resonance lock-in absorption mode and dispersion mode spectra of polycrystalline samples of CaF₂ potassium caproate (KC₆), and lithium stearate (LiC₁₈) have been obtained at various levels of saturation. The line widths narrow and the line shapes change in both the absorption and dispersion mode spectra on saturation. This behaviour is not predicted by previous theories of saturation, but is predicted by the new magnetic resonance saturation theory of Provotorov. The effects of modulation saturation have also been demonstrated. They are in agreement with Goldman's extension of Provotorov theory to include the audio modulation field. An Important prediction of Provotorov-Goldman theory is that saturation narrowing and modulation saturation do not affect the signal at the centre of resonance (within certain limiting conditions) and so the signals at this point are expected to saturate with the normal saturation factor: Z(O) = [ 1 + ɤ²H₁²T₁f(O)/2]⁻¹, where H₁ is the rf field amplitude, f(Δ) is the absorption line shape function normalized to 2π, and Δ is in rad/sec. Therefore the progressive saturation of the lock-in dispersion signal, u₁(0), has been studied in the CaF₂ KC₆, and LiC₁₈ samples at room temperature. The results verify the above prediction and yield the spin lattice relaxation times (T₁) of the samples. The CaF₂ result of 0.385 ±0.03 sec compares well with 0.45 ± 0.05 sec, the value found by adiabatic rapid passage. A modified Linder signal decay technique has also been used to measure T₁ values in KC₆ and LiC₁₈. The innovation being that the signal u₁(0) was used instead of the lock-in maximum absorption signal. The results are in good agreement with the progressive saturation results. It is concluded that one is finally in a position to measure correct T₁ values in solids by CW techniques. A technique for recording the true shapes of rapid passage signals has been developed. Using the shape of the rapid passage signal as a criterion of whether or not the passage was also adiabatic, it was found that the Bloch adiabatic condition, dH₀/dt « ɤH₁² , is also applicable to solids. The inequality, however, must be larger for solids than for liquids. The width at half its peak height of an adiabatic rapid passage (Arp) signal in a solid was shown to be [12(H₁²+HL²)]⅛, where HL² = <ΔH²>/3, is called the local magnetic field, and <ΔH²> is the Van Vleck second moment. ARP signals were used to find local field and second moment values in polycrystalline and single crystal forms of CaF₂ and also in polycrystalline LiC₁₈, all at room temperature. The results are in excellent agreement with theory and CW measurements. It is believed that this is the first time this method has been used. The ARP technique was also used to measure T₁ values. A symmetric sweep method was used for the above samples and a two pass method (equivalent to the π- π/2 sequence used in pulse spectrometry) was used for a very pure crystal of maleic anhydride. A value of 76 min was found for this sample at room temperature. This is a particularly good example of the usefulness of the ARP technique since it is difficult to measure such a long T₁ by the usual pulse method. Normal and saturation narrowed lock-in absorption spectra of LiC₁₈ have been obtained over the temperature range 25° to 193°C. There are two phase transitions in this region. They were revealed by both the normal and saturation narrowed spectra. / Science, Faculty of / Chemistry, Department of / Graduate

The purpose of this study was: (1) to find the effect of temperature on the wet oxidation of chalcopyrite, pyrite, and copper sulfide over the biological and nonbiological range; (2) to determine the influence of ferrous and ferric iron and free oxygen(air) on the oxidation. The sulfide minerals studied were oxidized over a range of temperatures between 25° to 75° C. The optimum biological oxidation occurred near 35° C. The nonbiological oxidation was very slow at low temperatures but increased with increasing temperature. The effect of solutions used in leaching the various sulfides over the temperature range differred. A synthetic nutrient solution, containing only a source of phosphate and nitrogen, was more effective on the biological oxidation of pyrite than on copper sulfide at the optimum temperature. Tailings water, obtained from Bingham Canyon, Utah, was more effective on chalcopyrite over a long period of time than the synthetic nutrient. Ten times more chalcopyrite was oxidized at 70° C. than at 55° C. in a dilute solution of sulfuric acid. A dilute solution of ferrous sulfate had very little effect on the oxidation of the sulfides. Oxygen and ferric iron in oxidized tailings water were shown to affect the oxidation of chalcopyrite at 65° C. The ferric iron was reduced. In an atmosphere of nitrogen, ferrous iron in fresh tailings water was inactive on the oxidation of chalcopyrite. Oxygen is the primary oxidizing agent in the biological and nonbiological oxidation of sulfide minerals in the leaching process.

No description available.

by Man Wah-Kit. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1996. / Includes bibliographical references (leaves 177-180). / LIST OF FIGURES / LIST OF TABLES / abstract --- p.1 / Chapter 1. --- introduction --- p.3 / Chapter 2. --- fabrication process / Chapter 2-1 --- INTRODUCTION --- p.7 / Chapter 2-2 --- PROCESS DEVELOPMENT --- p.8 / Chapter 2-3 --- FABRICATION PROCEDURES FOR TIN FILMS --- p.10 / Chapter 2-4 --- FABRICATION PROCEDURES FOR TIN OXIDE FILMS --- p.14 / Chapter 2-5 --- FABRICATION THEORY --- p.21 / Chapter 2-6 --- OXYGEN ION IMPLANTATION OF TIN FILMS --- p.24 / Chapter 3. --- structural characterization / Chapter 3-1 --- INTRODUCTION --- p.30 / Chapter 3-2 --- MICROSTRUCTURE / Chapter 3-2-1 --- SOME RELATED THEORIES OF GRAIN GROWTH / Chapter (1) --- Classical Theory of Grain Growth --- p.30 / Chapter (2) --- Hillock Growth --- p.31 / Chapter (3) --- Dislocation Creep Theory --- p.33 / Chapter (4) --- Biaxial Stress in Thin Films --- p.35 / Chapter (5) --- Surface Cluster Growth --- p.37 / Chapter 3-3 --- EXPERIMENTATION AND RESULTS / Chapter 3-3-1 --- MICROSTRUCTURAL ANALYSIS UNDER OPTICAL MICROSCOPE --- p.39 / Chapter 3-3-2 --- THE STRESS AND HILLOCK HEIGHT ANALYSIS OF TIN OXIDE FILMS --- p.48 / Chapter 3-3-3 --- MICROSTRUCTURAL ANALYSIS BY MEANS OF ATOMIC FORCE MICROSCOPE (AFM) --- p.52 / Chapter 3-3-4 --- MICROSTRUCTURAL ANALYSIS BY X-RAY DIFFRACTION --- p.69 / Chapter 3-3-5 --- SURFACE ANALYSIS BY MEANS OF X-RAY PHOTOELECTRON SPECTROSCOPY / Chapter (1) --- Introduction --- p.73 / Chapter (2) --- Basic Theory --- p.73 / Chapter (3) --- Experimentation And Results --- p.75 / Chapter 3-3-6 --- SURFACE STUDY OF ION IMPLANTED TIN OXIDE FILMS / Chapter (1) --- Experimental Results --- p.82 / Chapter 3-4 --- DISCUSSION / Chapter 3-4-1 --- QUALITATIVE ANALYSIS OF MICROSTRUCTURE WITH THE OPTICAL MICROSCOPE --- p.88 / Chapter 3-4-2 --- QUALITATIVE ANALYSIS OF MICROSTRUCTURE WITH SEM AND AFM / Chapter (1) --- Grain Growth of Tin Oxide Films --- p.89 / Chapter (2) --- Dependence of Grain Size on Deposition Rate --- p.91 / Chapter (3) --- Dependence of Grain Size on Film Thickness --- p.92 / Chapter (4) --- Dependence of Grain Size on Substrate Temperature --- p.92 / Chapter (5) --- Origin of Hillock Growth of Tin Oxide Films --- p.93 / Chapter 3-4-3 --- FILM COMPOSITIONAL ANALYSIS WITH X-RAY DIFFRACTION --- p.95 / Chapter 3-4-4 --- SURFACE ANALYSIS WITH X-RAY PHOTOELECTRON SPECTROSCOPY …… --- p.95 / Chapter 3-4-5 --- SURFACE ANALYSIS OF OXYGEN IMPLANTED TIN FILMS --- p.96 / Chapter 4. --- OPTICAL CHARACTERIZATION / Chapter 4-1 --- INTRODUCTION --- p.98 / Chapter 4-2 --- THEORY / Chapter (1) --- Free Electron Model --- p.99 / Chapter (2) --- Effect of Film Thickness --- p.100 / Chapter (3) --- Effect of Oxygen Contents --- p.101 / Chapter (4) --- Electron-Lattice Interaction and Bandgap Studies --- p.102 / Chapter 4-3 --- EXPERIMENTATION AND RESULTS --- p.105 / Chapter 4-4 --- DISCUSSION / Chapter 4-4-1 --- BANDGAP STUDIES FOR TIN OXIDE FILMS WITH DIFFERENT DEPOSITION CONDITIONS / Chapter (1) --- Variation of Film Thickness --- p.122 / Chapter (2) --- Film Appearance --- p.123 / Chapter (3) --- Variation of Substrate Temperature --- p.123 / Chapter (4) --- Variation of Oxidation Conditions --- p.123 / Chapter 5. --- ELECTRICAL CHARACTERIZATION / Chapter 5-1 --- INTRODUCTION --- p.126 / Chapter 5-2 --- RELATED THEORY / Chapter 5-2-1 --- CURRENT-VOLTAGE (I-V) CHARACTERISTICS --- p.127 / Chapter 5-2-2 --- CAPACITANCE-VOLTAGE (C-V) CHARACTERISTICS --- p.131 / Chapter 5-2-3 --- RELATION OF ELECTRICAL TO STRUCTURAL PROPERTIES / Chapter (A) --- Effects of Deposition Conditions --- p.133 / Chapter (B) --- Effects of Grain Boundaries --- p.133 / Chapter (C) --- Effects of Ionic Impurities --- p.134 / Chapter (D) --- Effects of The Interface Properties --- p.134 / Chapter 5-2-4 --- MEASURING TECHNIQUES / Chapter (A) --- I-V Measurment of Tin Oxide on a Silicon Substrate --- p.136 / Chapter (B) --- C-V Measurement of Tin Oxide Films on Silicon Substrates --- p.137 / Chapter (C) --- Electrical Measurement of Tin Oxide Films on a Quartz Substrate --- p.137 / Chapter 5-3 --- EXPERIMENTATION --- p.138 / Chapter 5-4 --- RESULTS --- p.141 / Chapter 5-5 --- DISCUSSION / Chapter 5-5-1 --- Analysis of the Conduction Mechanism for Sn02/Si n-p Heterojunctions --- p.161 / Chapter 5-5-2 --- Analysis of the Conduction Mechanism for Sn02/Si n-n Heterojunctions --- p.162 / Chapter 5-5-3 --- Effect on the Conduction Mechanisms of Film Thickness --- p.164 / Chapter 5-5-4 --- Effect on the Conduction Mechanisms of Oxidation Time --- p.166 / Chapter 5-5-5 --- Interfacial Properties of SnOx/Si Heterojunctions --- p.166 / Chapter 5-5-6 --- Electrical Properties of SnOx Films on Quartz / Chapter (1) --- Dependence of Film Conductivity on Measuring Temperatures --- p.168 / Chapter (2) --- Dependence of Film Conductivity on Oxidation Time --- p.168 / Chapter (3) --- Dependence of Film Conductivity on Oxidation Temperature --- p.169 / Chapter (4) --- Invariance of Film Conductivity at Some Certain Measuring Temperatures --- p.170 / Chapter (5) --- Activation Energy of Sn02 Films on Quartz --- p.170 / Chapter 6. --- CONCLUSIONS --- p.172 / Chapter 7. --- FUTURE WORKS --- p.175 / Chapter 8. --- REFERENCES --- p.177 / Chapter 9. --- APPENDICES / Chapter 9-1 --- APPENDIX A List of photos --- p.181 / Chapter 9-2 --- APPENDIX B (1) ED AX results for some selected regions on samples with hillocks --- p.182 / Chapter (2) --- Relations between mean surface roughness and oxidation conditions --- p.185 / Chapter (3) --- XPS original data and typical XPS spectra for vacuum- evaporated SnO2 thin film --- p.186 / Chapter 9-3 --- "APPENDIX C Variations of optical parameters, refractive index n and extinction coefficient k in visible region with different oxidation conditions" --- p.189 / Chapter 9-4 --- APPENDIX D Electrical results for Sn02/Si heterojunction s --- p.191 / Chapter 9-5 --- APPENDIX E Calculations of band diagram for Sn02/Si heterojunctions --- p.194 / Chapter 9-6 --- APPENDIX F Resistivity versus impurity concentration for silicon at 300K --- p.196

The objective of this thesis is to study the fracture behavior of bulk metallic glasses. For this purpose, detailed finite element investigation of the mode I and mixed mode (I and II) stationary crack tip fields under plane strain, small scale yielding conditions is carried out. An implicit backward Euler finite element implementation of the Anand and Su constitutive model [Anand, L. and Su, C., 2005, J. Mech. Phys. Solids 53, 1362] is used in the simulations. The effects of internal friction (μ), strain softening, Poisson's ratio (ν) and elastic mode mixity (Me) on the near-tip stress and deformation fields are examined. The results show that under mode I loading, a higher μ leads to a larger normalized plastic zone size and higher plastic strain level near the notch tip, but causes a substantial decrease in the opening stress. The brittle crack trajectories and shear band patterns around the notch are also simulated. An increase in ν reduces the extent of plastic zone and plastic strain levels in front of the notch tip. The results from mixed mode simulations show that increase in the mode II component of loading dramatically increases the maximum plastic zone extent, lowers the stresses and significantly enhances the plastic strain levels near the notch tip. Higher μ causes the peak magnitudes of tensile tangential stress to decrease. The implications of the above results on the fracture response of bulk metallic glasses are discussed. The possible variations of fracture toughness with mode mixity predicted by employing two simple fracture criteria are examined. Finally, mixed mode (I and II) fracture experiments on a Zr-based bulk metallic glass are performed. It is found that the fracture toughness increases with Me and Jc under mode I is higher than that under mode II loading by a factor of 4. The operative failure mechanism and fracture process zone size are discerned based on observations of incipient crack growth and fractographs. Lastly, a fracture criterion is proposed which predicts the experimentally observed variation of fracture toughness with mode mixity.

No description available.

Bulk metallic glasses and multi-principal element alloys represent relatively new classes of multi-component engineering materials designed for satisfying multiple functionalities simultaneously. Correlating the microstructure with mechanical behavior (at the microstructural length-scales) in these materials is key to understanding their performance. In this study, the structure evolution and nano-mechanical behavior of these two classes of materials was investigated with the objective of fundamental scientific understanding of their properties. The structure evolution, high temperature nano-mechanical behavior, and creep of two Zr-based alloys was studied: Zr41.2Ti13.8Cu12.5Ni10.0Be22 (Vitreloy1) and Zr52.5Ti5Cu17.9Ni14.6All0 (Vitreloy105). Devitrification was found to proceed via the formation of a metastable icosahedral phase with five-fold symmetry. The deformation mechanism changes from inhomogeneous or serrated flow to homogenous flow near 0.9Tg, where Tg is the glass transition temperature. The creep activation energy for Vitreloy1 and Vitreloy105 were 144 kJ/mol and 125 kJ/mol, respectively in the range of room temperature to 0.75Tg. The apparent activation energy increased drastically to 192 kJ/mol for Vitreloy1 and 215 kJ/mol for Vitreloy105 in the range of 0.9Tg to Tg, indicating a change in creep mechanism. Structure evolution in catalytic amorphous alloys, Pt57.5Cu14.7Ni5.3P22.5 and Pd43Cu27Ni10P20, was studied using 3D atom probe tomography and elemental segregation between different phases and the interface characteristics were identified. The structure evolution of three multi-principal element alloys were investigated namely CoCrNi, CoCrFeMnNi, and Al0.1CoCrFeNi. All three alloys formed a single-phase FCC structure in as-cast, cold worked and recrystallized state. No secondary phases precipitated after prolonged heat treatment or mechanical working. The multi-principal element alloys showed less strain gradient plasticity compared to pure metals like Ni during nano-indentation. This was attributed to the highly distorted lattice which resulted in lesser density of geometrically necessary dislocations (GNDs). Dislocation nucleation was studied by low load indentation along with the evaluation of activation volume and activation energy. This was done using a statistical approach of analyzing the "pop-in" load marking incipient plasticity. The strain rate sensitivity of nanocrystalline Al0.1CoCrFeNi alloy was determined by in situ compression of nano-pillars in a Pico-indenter. The nanocrystalline alloy demonstrated a yield strength of ~ 2.4 GPa, ten times greater than its coarse grained counterpart. The nanocrystalline alloy exhibited high strain rate sensitivity index of 0.043 and activation volume of 5b3 suggesting grain boundary dislocation nucleation.