<p>The intermetallic compound Al₁₃Fe₄ was rapidly solidified at a velocity in excess of 2 to 5m/s by a pulsed laser melting technique. Resolidification after laser melting occurred epitaxially by a stepwise growth process. The phase resulting from this growth process was determined by electron microscopy, to be a faulted slightly disordered variant of the stable compound. The chemical long-range order parameter of this phase was estimated to be 0.6 to 0.7 from the relative intensities of HOLZ spots in an electron diffraction pattern. A model relating order parameter to solidification velocity was developed to explain these results. The measured order parameter of the rapidly solidified Al₁₃Fe₄ was in agreement with the model predictions. A modification of the same model successfully explains the observed velocity at which the intermetallic compound ϒNi₃Al solidified without long-range order. An untested prediction of the general model is that above a critical velocity, estimated to be equal to between 0.5 and 3 times the liquid diffusion coefficient divided by the spacing of close-packed planes in the compound, a compound with long-range order will not form. This model is expected to apply to all types of intermetallic compounds except those in which size factor effects dominate the structure. The results of this work show that rapid solidification experiments on intermetallic compounds can be used to examine the melt-crystal interface kinetics.</p> <p>Experimental problems associated with pulsed laser melting experiments on intermetallic compounds were also examined in detail. To do this, artifacts from the starting material and from the TEM sample preparation processes were identified and characterized. From this characterization, three important things were learned. Firstly, laser melting experiments on thin films of intermetallic compounds are not equivalent to laser melting experiments on the bulk compound. This was determined by doing laser melting experiments on thin films of Al₁₃Fe₄ on rock salt, which resolidified by nucleation and growth from the melt, in contrast with the bulk compound, which resolidified epitaxially after laser melting. Secondly, the mechanism of laser induced topography developrrent on intermetallic compounds was examined, and found to be a spalling process caused by the thermal contraction on solidification. Thirdly, artifacts from ion thinning were characterized. The most important of these, 10nm grains of redeposited sputtered material on the sample, is difficult to separate from the solidification microstructure without the aid of cross section samples, and thus, may be misinterpreted as nucleation and growth in the melt.</p> / Doctor of Philosophy (PhD)

<p>The drive to produce materials with novel or beneficial combinations of properties has prompted research into a range of new materials and processing routes. In many applications one of the important design variables is the mechanical strength. Exceptional strengths can be achieved in certain materials consisting of two deformable phases when they are drawn into fine wires or rolled into thin sheets, the common example being pearlitic steel wire which can achieve strengths in excess of 5 GPa. The mechanisms that permit co-deformation and result in the observed strengthening are, however, not well understood. In this thesis an approach was adopted whereby co-deformation of a well characterized model material has been studied primarily using uniaxial tensile tests. Directional solidification of a Cu-1.56at%Cr eutectic alloy has been used to produce material consisting of submicron diameter single crystals of Cr embedded within a polycrystalline Cu matrix. It has been shown that these two phases exhibit preferred crystallographic orientation relationships, habit planes and growth directions the same as those found for solid state precipitates of Cr in Cu. On deforming this material it is found that the Cr fibres yield at stresses close to the theoretical limit. However, their are able to continue to co-deform with the Cu matrix to large plastic strains. This process of co-deformation is observed to cause a rate of nearly constant work hardening that results in both high strength and high ductility. This behaviour has been attributed to the fact that the Cr fibres continue to carry increasing elastic strain beyond their yield thereby contributing to an increasing level of internal stress in the material. It is suggested that this mechanism may play an important role in other co-deformed two phase materials. In particular, it is suggested that this may provide one mechanism for the continued high rate of work hardening in heavily co-deformed two phase materials.</p> / Doctor of Philosophy (PhD)

<p>Ceramics have material properties that make them useful for many industrial applications. They are strong, hard, and chemically inert. Their refractoriness gives them an advantage over metals and polymers for use at high temperature. Unfortunately, the inherent brittleness of ceramics limits their use in structural applications.</p> <p>One way to improve the toughness of ceramics is to combine them with other materials to make composites. The correct combination of materials can lead to synergism, and a significant improvement in properties. In this work, brittle laminates that contain weak interlayers are considered. The weak interlayers lead to crack deflection, and can result in non-catastrophic failure of the material. The requirements for consistent crack deflection and non-catastrophic failure are not fully understood.</p> <p>This work is an attempt to explain the observed fracture behaviour in brittle laminar composites that contain weak interlayers. A combination of experimental work, fracture mechanics modeling and finite element modeling has been used to predict the requirements necessary for non-catastrophic failure.</p> <p>The work shows the size of flaws in the surface of the composite, in the weak interlayer, and in subsequent strong layers in the material, all play an important role in the fracture behaviour. Control and understanding of the effect of the various flaw sizes can be used to achieve non-catastrophic failure and increased work of fracture in these composites.</p> / Doctor of Philosophy (PhD)

<p>Organic electroluminescent devices (OLED's) have been intensely researched in recent years. Poly(p-phenylene vinylene) (PPV), due to its ease of processing and potential high quantum yield, is considered a good candidate for future display applications. By synthesizing PPV films and fabricating PPV-based organic ELD's, optoelectronic properties of this conjugated polymer were investigated. The PPV synthesis was modified to improve material quality which affected device stability, lifetime, and luminescent yield.</p> <p>By utilizing transmission electron microscopy, the presence of PPV crystals which lowered the quantum yield was thoroughly investigated. Electron micrographs revealed that aggregates of crystals were embedded in an amorphous matrix. Analysis of diffraction patterns established that NaCI was a major impurity and that its crystals acted as nucleation sites. Steps were accordingly taken to exclude NaCI and prepare a more amorphous material.</p> <p>Impedance spectroscopy was used to measure conductivity and the activation energy. The low activation energy suggested carrier transport via hopping between localized states. Furthermore, the presence of a metal/polymer interface, which limited charge injection into the sample and lowered the efficiency, was identified. By controlling the parameters affecting device/material preparation, this interface was to a great extent improved.</p> <p>Electroluminescent devices with areas of 40 mm² were fabricated. These devices could draw currents of up to 0.2 mA/cm² and emit a yellowish green light, visible under laboratory lighting at various viewing angles. An external efficiency of 0.4 LumlWatt was attained. A conservative internal efficiency of 0.59%, significantly higher than the ones previously reported for single layer organic devices, was calculated. The threshold voltage was reduced to 2-5 volts and stable OLED's with a lifetime of over 160 hours were fabricated.</p> <p>The photovoltaic effect in PPV was explored by employing delayed-collection-field and field-induced fluorescent quenching techniques. A carrier generation efficiency of 42% was obtained. The strong field dependence of the carrier generation efficiency was noticed which supported the formation of bound electron-hole pairs.</p> / Doctor of Philosophy (PhD)

<p>A two part study was conducted on the nitridation behaviour of dilute Ni-Ti alloys, and on the orientation relationships (OR)s and interfacial structures between TiN and Ni. Beginning with the Fickian relationships first proposed by Kirkaldy (1969) to describe internal sulphidation and assuming a solubility product of the form K=BC, these equations were combined to yield for the metal solute:</p> <p>(δB/δt)=[Dʙ+(KDc/B²)](δ²B/δx²)</p> <p>This second order differential equation shows that the concentration profile of the metal solute, B, is dependent not only on the diffusivity of the metal, Dʙ, but also on the solubility product K and on the gas diffusivity Dc. Two limiting cases were identified from the previous equation. The first which depends only on the metal diffusivity, Dʙ, corresponds to Wagner's (1952) error function solution for the superficial oxidation of alloys dilute in B. The second limiting case (Dʙ=0) corresponds to Wagner's (1959) oxygen controlled internal oxidation as proven analytically by Ohriner and Morral (1979). A criterion describing the transition from internal to external nitridation (oxidation) was deduced from the previous equation as:</p> <p>(KDc/B₁²Dʙ)<0.1</p> <p>where B₁ is the interfacial composition of the metal solute.</p> <p>A finite difference algorithm was presented based on these ideas. A comprehensive test of this approach showed that it could generate the well known analytical solutions of Wagner's two limiting conditions. It was also shown that provided the solubility product K is small, Ohriner and Morral's (1979) analytical solution is valid even for an alloy not saturated in the gas phase (i.e. the composition need not be confined to the solvus of the ternary isotherm). As well as reproducing these analytical solutions, the model was successfully tested against the experimental work of Swisher (1968) and Kirkaldy (1969). A detailed investigation was conducted on the concentration profiles of Ti in Ni as a consequence of either internlal nitridation or superficial nitridation using Energy Dispersive X-Ray (EDX) analysis. It was found that the program always generated the correct shapes of the diffusion profiles as well the transition to external nitridation in the Ni-Ti-N system. Based on the measured interfacial composition of the interface, first time determinations of both KDN and DTi values were obtained between 800°C and 1020°C. The diffusivity of Ti in Ni is similar to other substitution solutes in Ni and can be expressed as</p> <p>Dn=0.07 exp(-320000/RT) m²/s (900°C-1020°C)</p> <p>where R is the universal gas constant 8.314 J/mole-K</p> <p>A solubility product calculation of TiN in Ni showed values between 1.4 x 10ˉ⁶ and 5x10ˉ⁸ [w/o]² (between 1020°C and 800°C respectively). Using this data a diffusivity of N in Ni was obtained which compares favourably to other interstitial diffusivity data in Ni and ϒFe. The upper bound value for the diffusivity of N in dilute Ni-Ti alloys can be expressed as:</p> <p>DN= 0.0003 exp(-17000/RT) m²/s (900°C-1020°C)</p> <p>Stoichiometric TiN was the only precipitated phase found between 800°C-1020°C in this work. A first time investigation of the interfacial structure between TiN and Ni was conducted using transmission electron microscopy (TEM). It was determined that at least five different orientation relationships between TiN and Ni exist. All these ORs had relatively small near coincidence site lattice (CSL) relationships. The planar Σ value of Ni (δɴᵢ) with respect to the Ti sublattice were all found to be under 50. No correlation was found between the size of the CSL unit cell and the frequency of observation of a particular OR. In only two cases were misfit dislocations observed. Both O-Iattice and plane matching theory made correct predictions as to the misfit dislocation configurations. In both cases one predominate set of dislocations was calculated. It was not possible, however, to unambiguously identify the Burgers vector. The habit planes were dominated by {010}TiN indicating that this was an important factor in selecting the morphology and orientation relationships of these particles. This plane coincided both with a primary O-lattice plane and, as already noted, with a planar CSL relationship. These factors probably all play a role in the selection of these orientation relationships and habit planes in the Ni/TiN system.</p> / Doctor of Philosophy (PhD)

<p>Heteroepitaxial growth of In₁_ᵪGaᵪAsᵧP₁_ᵧ alloy thin films on InP(001) substrates using molecular beam epitaxy (MBE) is technologically important for the fabrication of optoelectronic devices. Recent progress in MBE, for example the use of hydride gas sources such as AsH₃ and PH₃ has improved the control of the P to As ratio and added more versatility to this thin film growth technique. In this thesis, transmission electron microscopy (TEM) studies of some unique growth-related phenomena occurring in In₁_ᵪGaᵪAsᵧP₁_ᵧ films grown with gas-source MBE are described. The experiments were intended to understand the growth mechanism of the films and the characteristics of structural "defects", rather than to obtain practical knowledge which can directly be applied to the manufacturing of optoelectronic devices.</p> <p>The initial stage of strain relaxation in compressively-strained InAsᵧP₁_ᵧ ternary alloy films was studied. The compressive misfit strain in 190Å thick films ranged from -0.96% to -2.43% (0.30≤y≤0.77). Both TEM and cathodoluminescence imaging revealed the existence of new types of misfit dislocations aligned close to [100] and [010] directions generated from heterogeneous dislocation sources at the InAsᵧP₁_ᵧ/ InP (001) interfaces. The new types of misfit dislocations were found to be pure edge in character, formed probably by slip on {011} planes. Contrast profiles were computed for misfit dislocations parallel and adjacent to one of the free surfaces of a plan-view TEM foil by taking into account the significantly modified elastic fields associated with such misfit dislocations. The calculated profiles were useful when Burgers vectors of misfit dislocations generated from dislocation source were analyzed.</p> <p>A composition modulation occurring only along the [110] direction with a small wavelength (100Å) was repeatedly observed in 3000Å thick In₁_ᵪGaᵪAsᵧP₁_ᵧ alloy films having a wide range of nominal composition being lattice-matched to InP. This modulation is different from the "classical" spinodal decomposition along elastically soft [100] and [010] directions with long-wavelength (1000-1500Å), commonly observed in In₁_ᵪGaᵪAsᵧP₁_ᵧ films grown by liquid phase epitaxy. A simple free energy argument including the effect of constraint from the InP substrate suggests that the observed modulation is not directly caused by the minimization of free energy but is driven by kinetic processes occurring at the surface of growing films.</p> <p>The morphological instability of a strained film surface was also studied. Cross-sectional TEM investigation revealed the coexistence of prominent composition modulation and surface undulation in tensile-strained In₁_ᵪGaᵪAsᵧP₁_ᵧ films. The surface morphology of the strained films showed a clear correlation with the degree of the composition modulation, occurring predominantly along the [110] direction. When the film was under tension, the surface undulation developed at relatively low strains (as low as 0.5%); on the other hand, films with the same magnitude of compressive strain did not exhibit any surface undulation. In addition, the surface undulation was composition dependent; it was enhanced as the degree of composition modulation increased. These features distinguish the present morphological instability induced by the inherent composition modulation from the "classical" three-dimensional island growth mode of compressively strained films having strains larger than 2%.</p> / Doctor of Philosophy (PhD)

<p>This work concerns the effect of manipulating the mesoscopic spatial arrangement of coarse, hard particles in a ductile metallic matrix on the overall macroscopic deformation behaviour of the bulk solid. The hypothesis that the spatial distribution of the harder phase influences the onset of yielding and strain hardening in a particle hardened ductile solid is examined by way of experiment on a well characterized metallic system containing two phases. Rapidly solidified hypoeutectic binary Al-Cu granules, with nominal Cu compositions of 5%, 10%, 17% and 24% by weight were chosen as a model system. At room temperature, the binary system consists of two terminal equilibrium phases: the intermetallic compound CuAl2 , and the ductile Al solid solution. At room temperature the intermetallic is approximately seven times harder than the matrix and intrinsically brittle. Materials possessing either a uniform spatial distribution or a bimodal spatial distribution of the CuAl 2 particles are generated through a combination of hot-pressing and high temperature forging. Compression tests and complimentary experiments were performed on the materials in order to determine the magnitude and distribution of plasticity and damage in the materials as a function of the local heterogeneity and applied strain. The experimental flow curves were then compared to simulations obtained from two non-linear self-consistent continuum models of particle hardened, power law solids developed from the Eshelby "Equivalent Inclusion" Method. The flow curves obtained experimentally for the spatially uniform materials are in good agreement with a self-consistent method in which the matrix is assumed to uniformly coat the elastic particles to form a continuous network. In contrast, a model which assumes a random disordered morphology of both the particles and the matrix underestimates the plastic compliance of the uniform materials when the concentration of the particles is non-dilute. At small strains, the hardening rate observed experimentally is enhanced by inhomogeneous spatial distribution of the second phase when the contrast between the properties of the hard and soft regions of the microstructure is strong and the volume occupied by the hard regions is high. A simple continuum deformation model which accounts for clustering is in good agreement with the flow curves of the clustered materials.</p> / Doctor of Philosophy (PhD)

<p>The proposed use of molybdenum disilicide (MoSi2) in gas turbines can only be properly evaluated when mechanical property information is available for all temperatures that are encompassed in this application. Because molybdenum disilicide is brittle at room temperature special testing techniques are required to obtain deformation information at this temperature. An indentation test is one method of obtaining room temperature deformation information of brittle materials. Indentations were made on two surfaces of molybdenum disilicide single crystals at room temperature. The indentations were made with a range of applied loads and the indenter was aligned in two different orientations on each surface. The deformation and fracture produced by the indentations was studied with Scanning Electron Microscopy (SEM), Atomic Force Microscopy (AFM) and Transmission Electron Microscopy (TEM). The results from these analyses are discussed in terms of the micromechanisms of deformation and fracture and related to both the hardness and changes in applied load.</p> / Doctor of Philosophy (PhD)

<p>The systems Na - C60 , K - C60 and Sr - Al were studied by the Electromotive force (EMF) technique using electrochemical cells which were composed of beta-alumina tubes glass-sealed to α-alumina lids. EMF measurements of Nax C60 at 599 K indicated solid solution regions at 1.7 < x < 3 and 3.3 < x < 12. (A sudden drop in EMF in the range of 3.3 < x < 3.7 allows the possibility of a nearly stoichiometric phase in this range). No compounds were observed in the composition range of 0 < x < 1.7. From EMF measurements, the Gibbs energies of mixing of [Special characters omitted.] Na3 C60 and [Special characters omitted.] Na6 C60 were determined as 85 and 81 kJ/mol, respectively. From EMF measurements of Kx C 60 at 572 K, the Gibbs energies of mixing of [Special characters omitted.] were found to be 83, 117, 120 and 121 kJ/mol, for x = 1, 3, 4 and 6, respectively. The ideal interstitial solution model of the Na - C 60 system indicates that tetrahedral sites are favored by Na. The difference between Nax C60 and Kx C60 seems to be related to the ionic size effects. EMF measurements of the Sr - Al system using a Sr beta-alumina solid electrolyte were unsuccessful. Thermodynamic consideration suggests that Sr beta-alumina is not compatible with pure Al or Sr. A model for the intermolecular interactions between C60 molecules, based on the effective Lennard-Jones interaction centers and local charge distribution, was proposed and tested for the Pa 3¯ structure of pure C60 by both regular solution and cluster variation methods. As the effective Lennard-Jones interaction centers move from carbon atoms to the centers of adjacent double bonds of the same molecule, the Pa 3¯ structure in which the pentagons of a molecule face double bonds of nearest molecules becomes more stable relative to the Pa 3¯ structure in which the hexagons face double bonds. By assigning charges to carbon atoms and to the centers of bonds, we can model the increase in activation energy for the jump between 22 and 82 degree orientations. External vibrational frequencies of simple cubic (sc ) C60 and M3 C60 (M = K, Rb) were computed by the harmonic approximation, using the model for the intermolecular interactions between C60 molecules. The computed frequencies of phonons are quite consistent with experiments. However, for librons there are some differences between computed and observed frequencies, which suggests that the harmonic approximation is not adequate for librons. By using group theory, the external vibrational modes of sc C60 at Γ, Σ, Δ and Λ are labeled and the corresponding symmetry adapted vectors are obtained. Thermodynamic properties of sc C60 were computed using the canonical partition function. The lattice energy and configurational entropy terms are estimated by the regular solution model. The dispersion curves were approximated by the Debye model for acoustic modes and the Einstein model for optical modes. The Debye characteristic temperature and optimum Grüneisen constant were determined as 54.14 K and 7.5, respectively. The contribution of the 82 degree orientation of the molecules to the constant volume heat capacity was found to be significant in the temperature range of 20 to 100 K.</p> / Doctor of Philosophy (PhD)

<p>The creep rates of SiC whisker reinforced Al2 O3 have been found to be one or two orders of magnitude lower than the creep rate of unreinforced alumina. However, whiskers are a serious health hazard due to their asbestos-like geometry, they are expensive (thousands of dollars per kilogram), and they tend to get damaged during processing. Platelets have been proposed as an alternative to whiskers due to their reinforcement potential comparable to that of whiskers, forgiving geometry (with respect to safety), better thermal stability, lower price (hundreds of dollars per kilogram) and ease of processing. Up to now, research in platelet reinforced ceramics has concentrated mainly in room temperature properties and little is known about their high temperature mechanical properties. The aim of this work was to study the way in which different reinforcement network morphologies affect the creep behaviour of SiC-platelet/Al2 O3 composites and to determine the important deformation mechanisms at the studied temperature (1250°C). To this end, composites with different platelet volume fractions (0 to 30%) and orientation distributions were fabricated. The samples were subjected to flexure and compression creep tests and characterized using optical and electron microscopy, dilatometry, and neutron diffraction. The analysis of the creep behaviour was found to be complicated by the differences in impurity content in the samples and the increase in glass content with the platelets volume fraction. However, the results clearly indicate a strong influence of the reinforcement morphology on the creep properties. Special attention was given to an unusual time-dependent transition from high to low creep strain rate in some of the composites. The phenomenon was ascribed to the possible relief of bending strains in the platelets. In addition, some of the possible main mechanisms responsible for the increased creep resistance in SiC-whisker reinforced ceramics were found not be operative in platelet-reinforced ceramics.</p> / Doctor of Philosophy (PhD)