<p>Research in the area of improving the efficiency and manufacturability of alternative energy technologies has been of high interest due to the growing environmental concerns of energy resources. Group III-antimonide-based compound semiconductors have been sought after as excellent candidates for photovoltaic conversion of infrared radiation, outside the spectral range absorbed by the currently available crystalline Si solar cells. The major challenge is the GaSb/Si interface is highly lattice mismatched, and inherently heterovalent. This leads to a high density of structural defects, many of which have not been investigated fully. Both optical and electrical properties of such heteroepitaxy thin films are strongly dependent on the periodicity of the crystal lattice, and the presence of extended defects cause perturbations in the lattice periodicity. Therefore the nature of such extended defects must be understood, in order to better manipulate the growth process to minimize their presence. This thesis demonstrates that through the use of conventional transmission electron microscopy, further insight can be gained into understanding the origin, distribution, propagation, and interaction of various extended defects. From this, a couple of ways to systematically suppress some of the defects have also been implemented, and the mechanism by which they induce such a suppression is also discussed.</p> / Master of Applied Science (MASc)

<p>Proton exchange membrane fuel cells (PEMFC) are a technology of high interest for the automotive and power generation industry. The catalyst layer plays a critical role in fuel cells as it is responsible for catalyzing hydrogen oxidation and oxygen reduction to generate electricity. The current challenge in catalyst development is to produce highly active and economical catalysts. This challenge cannot be overcome without an accurate understanding of catalyst surfaces and morphology since the catalytic reactions occur on the surface active sites. Transmission electron microscopy (TEM) is an excellent tool to understand the structures of the nanoparticles down to the atomic level in determining the relationship with the catalyst’s performance in fuel cell applications. Platinum (Pt) is one of the best commercially available catalysts for PEMFC due to its highly active, inert, and relatively stable properties. However, Pt is a rare precious metal due to its low abundance and high demand. Further research is aimed at developing highly active and more economical catalysts in order to mass produce PEMFC. A strategic approach is to use platinum bimetallic alloys, which greatly reduce the platinum loading as they enhance the oxygen reduction reaction. A detailed understanding of the nanoparticle surface is critical as the catalyst surface strongly determines its catalytic activity. Furthermore, another challenge in utilizing fuel cells is the life-time of the catalysts. It is known that electrochemical cycling affects Pt alloys. As a result, the understanding of the effect of electrochemical treatments on the catalyst’s v morphology and composition is key to improving the fuel cell’s performance and durability. This thesis demonstrates that through the use of TEM, useful insights regarding the morphology, surfaces, and compositions of the catalysts can be gained and contribute to the improvement in catalyst development for next generation fuel cells.</p> / Master of Applied Science (MASc)

<p>InP nanowires with short InAs segments were grown on InP (111)B substrates by Au assisted vapour-liquid-solid growth in a gas source molecular beam epitaxy system. Nanowire crystal structure and morphology were investigated by transmission electron microscopy as a function of temperature, growth rate, and V/III flux ratio. At 370C predominantly kinked nanowires with random morphology and low areal density were observed with a rough parasitic 2D film. At 440C, nanowire density was also reduced but the 2D film growth was smoother and nanowires grew straight without kinking. An optimum temperature of 400C maximized areal density with uniform nanowire morphology. At the optimum temperature of 400C, an increase in V/III flux ratio changed the nanowire morphology from rod-shaped to pencil like indicating increased radial growth. Growth rate did not affect the crystal structure of InP nanowires. For InAs nanowires, changing the growth rate from 1 to 0.5 μm/hr reduced the presence of stacking faults to as low as one per nanowire. Short InAs segments in InP nanowires were found to grow through two mechanisms for nanowires of length L and diameter D. The first mechanism described the supply of In to the growth front via purging of In from the Au droplet where L was proportional to D. The second mechanism involved direct deposition of adatoms on the nanowire sidewall and subsequent diffusion to the growth front where L was proportional to 1/D. For intermediate growth durations, a transition between these two mechanisms was observed. For InP and InAs nanowires, the growth mode was varied from axial to radial through the inclusion of Al to form a core shell structure. Al_xIn_1-xAs(P) shells were grown on InAs cores with Al alloy fractions between 0.53 and 0.2. These nanowires were examined by transmission electron microscopy and it was found, for all values of x in InAs-Al_xIn_1-xP structures, that relaxation had occurred through the introduction of dislocations. For InAs-Al_xIn_1-xAs structures, all values except x=0.2 had relaxed through dislocation formation. A critical thickness model was developed to determine the core-shell coherency limits which confirmed the experimental observation of strain relaxation. The effects of passivation on the electronic transport and the optical properties were examined as a function of structural core-shell passivation and chemical passivation. The mechanisms for the observed improvement in mobility for core-shell versus bare InAs nanowires was due to the reduction in ionized impurity scattering from surface states. Similarly an increase in photoluminescence intensity after ammonium sulfide passivation was explained by the reduction of donor type surface states.</p> / Doctor of Philosophy (PhD)

<p>Heteroepitaxial films of group III-antimonide-based semiconductor compounds on Si are amongst the most appealing candidates for solar applications because of the well-established Si platform and also for offering band-gap energies beyond the silicon road map. Nonetheless, high lattice mismatch between GaSb and Si as well as ambiguous nucleation of GaSb on Si are major drawbacks in manufacturing of heteroepitaxial GaSb/Si films because they can generate various defects in films. Atomic-level detection of these defects and delving into their origin, orientation, distribution, propagation, and interaction with each other will therefore provide an insight into inhibiting their formation or reducing their severity. State-of-the-art aberration-corrected transmission electron microscopes have marked a new era in the investigation of interfaces and defects. With sub-angstrom electron probes in scanning transmission electron microscopes, it is possible to pinpoint the individual atomic columns at interfaces and defects.</p> <p>In this thesis, GaSb epilayers grown with molecular beam epitaxy on Si substrates were studied through aberration-corrected scanning transmission electron microscopy. The strain-relief mechanism of the epitaxial GaSb through formation of interfacial misfit dislocations was investigated and the strain distribution in the vicinity of dislocation cores as well as epitaxial layer was analyzed. The specific atomic-number dependent contrast mechanism of the high-angle annular dark-field technique enabled the unprecedented direct observation of anti-phase boundaries, the extended defects of highest interest in polar-on-nonpolar growths. This observation unraveled the ambiguity of nucleation of GaSb at interface regardless of preferential deposition of atomic species during growth procedure. The growth of GaSb at the initial stage of deposition was further investigated to understand the role of an AlSb buffer layer and growth mechanism of GaSb precisely. This investigation showed that AlSb and GaSb epilayers occur by Volmer-Weber growth mode and AlSb islands provide energetically favorable nucleation sites for GaSb film. Furthermore, taking advantage of atomic-resolution detection capability of high-angle annular dark-field in scanning transmission electron microscopy a novel mechanism of strain relief through multiple twining resulting in a lattice-registered growth of GaSb on Si(211) was elucidated. This contribution demonstrates that aberration-corrected scanning transmission electron microscopy provides profound insight into the polar-on-nonpolar growth which can be exploited to suppress the formation of structural defects.</p> / Doctor of Philosophy (PhD)

The work presented in this thesis centers on the application of atomic-resolution transmission electron microscopy to study Platinum-alloy nanoparticles. In particular, the thesis focusses on the platinum-iron and platinum-gold systems. Additionally, few other complementary structures based on Pt thin films and nanowires are also characterized. These materials are studied in the context of their catalytic application towards the oxygen reduction reaction in polymer electrolyte membrane fuel cells (PEMFCs). Here we report on the detailed investigation of many structural and compositional aspects of these catalyst nanoparticles including lattice strain, the surface and bulk atomic-structure, the surface/bulk chemical composition, surface segregation, and atomic ordering. In some cases we have even looked beyond the traditional characterization approaches. For instance, instead of observing the particle structures before and after a particular treatment (e.g., heating and degradation tests), we have captured the dynamics of structural evolution over the entire course of such treatments. These investigations were useful in interpreting their catalytic performances, which opened new perspectives towards further optimization of their material structure on the atomic-level. / Thesis / Doctor of Philosophy (PhD)

Ternary InGaN and AlGaN alloys have been sought after for the application of various optoelectronic devices spanning a large spectral range between the deep ultraviolet and infrared, including light-emitting diodes, and laser diodes. Their non-ideal alloy mixing, and differences in bond energy and in adatom diffusion are established as the cause for various types of nanoscale compositional inhomogeneity commonly observed in nitride thin films. Growth in a nanowire geometry can overcome the phase separation, surface segregation, and chemical ordering by providing enhanced strain relaxation of the large lattice mismatch at the free surfaces. In this dissertation, the spectral and spatial luminescence distributions of ternary III-N alloy nanowire heterostructures are investigated and correlated to structural and chemical properties with scanning transmission electron microscopy. Quantitative elemental mapping of InGaN/GaN dot-in-a-wire structures using electron energy-loss spectroscopy revealed compositional non-uniformity between successive quantum dots. Local strain mapping of the heterostructure showed a dependence of the incorporation of indium on the magnitude of the out-of-plane compressive strain within the underlying GaN barrier layer. Cathodoluminescence spectroscopy on individual nanowires presented diverse emission properties, nevertheless, the In-content variability could be directly correlated to the broad range of peak emission energies. Atomic-level chemical ordering within the InGaN was then reported, and attributed to the faceted growth surface in nanowires that promotes preferential site incorporation by In-atoms that allows for better strain relaxation. Distinct atomic-scale alloy inhomogeneities were also investigated in AlGaN nanowires, which evidenced spatial localization of carriers taking place at the resulting energy band fluctuations. A high spectral density of narrow emission lines arose from such compositional modulations, whose luminescence behaviours exhibit a dependence on the nature of the compositional fluctuations from which they originate. / Thesis / Doctor of Philosophy (PhD)

Layered lithium transition metal oxides represent a major type of cathode materials that are widely used in commercial lithium-ion batteries. Nevertheless, these layered cathode materials suffer structural changes during electrochemical cycling that could adversely affect the battery performance. Clear explanations of the cathode degradation process and its initiation, however, are still under debate and are not yet fully understood. In this thesis, the cycling-induced chemical and structural evolution of LiNi1/3Mn1/3Co1/3O2 (NMC) and high-energy Li1.2Ni0.13Mn0.54Co0.13O2 (HENMC) cathodes are investigated in details using state-of-the-art electron microscopy techniques combined with other bulk measurements to uncover the mechanisms at the source of cell deterioration. / Thesis / Doctor of Philosophy (PhD)

Graphene and its functionalized derivatives, such as nitrogen-doped graphene, have recently become a popular substrate material for the proton exchange membrane fuel cell (PEMFC) due to its enhanced electrical conductivity, electrochemical stability, and increased surface area when compared to the conventional, carbon black. In order to further develop the alternative fuel industry, the Pt catalyst within the PEMFC must also be considered. Single Pt atoms have a higher surface area to volume ratio when compared to nanoparticles, thus offering the potential to create a more affordable and efficient PEMFC. In this thesis, electrode materials comprising single Pt atoms and clusters, produced using atomic layer deposition (ALD) on various C derivatives, including graphene, N-doped graphene, carbon nanotubes (CNTs), and N-doped CNTs (NCNTs) are investigated through the utilization of aberration corrected transmission electron microscopy. Structural and chemical analysis was performed on thermally exfoliated N-doped graphene and CVD-produced graphene that was exposed to N+ ion sources. It was determined that the thermally exfoliated N-doped graphene maintained the short-range order of the graphene lattice; however, local inhomogeneities existed for the total N concentration, and the specific N-dopants within and between graphene sheets. More importantly, Pt atoms and clusters were observed and determined to be primarily stabilized at the edge of the N-doped graphene sheets. The stabilization of the Pt atoms and clusters resulted in a significantly higher mass and specific activity for the hydrogen evolution reaction, when compared to the use of a graphene substrate and Pt nanoparticles on C black. The N+ ion implantation in the CVD graphene showed the incorporation of N-dopants; however, electron energy loss spectroscopy revealed structural damage to thin sheets. NCNTs were also characterized in this thesis as possible gas containers, and as a substrate material to examine the effects of varying ALD conditions. It was determined that the NCNTs were an effective N2 gas conduit, wherein a decreasing pressure was observed with an increase to the inner diameter of the nanotubes. Using similar NCNTs, the effect of dosing time, temperature, and substrate on the Pt size were analyzed using ALD. While no singular condition resulted in the sole production of single Pt atoms, modifying both the substrate and dosing time were shown to provide the greatest potential for producing individual Pt atom catalysts. / Thesis / Doctor of Philosophy (PhD)

Building on the recent advances in computer vision with convolutional neural networks, we have built SpectralNet, a spectroscopy-optimized convolutional neural network architecture capable of classifying spectra despite large temporal (i.e. translational, chemical, calibration) shifts. Present methods of measuring the local chemical environment of atoms at the nano-scale involve manual feature extraction and dimensionality reduction of the original signal such as: using the peak onset, the ratio of peaks, or the full-width half maximum of peaks. Convolutional neural networks like SpectralNet are able to automatically find parts of the spectra (i.e. features) of the spectra which maximally discriminate between the classes without requiring manual feature extraction. The advantage of such a process is to remove bias and qualitative interpretation in spectroscopy analysis which occurs during manual feature extraction. Because of this automated feature extraction process, this method of spectroscopy analysis is also immune to instrument calibration differences since it performs classification based on the shape of the spectra. Convolutional neural networks are an ideal statistical classifier for spectroscopy data (i.e. time-series data) due to its shared weighting scheme in neural network weights which is ideal for identifying local correlations between adjacent dimensions of the time-series data. Over 2000 electron energy loss spectra were collected using a scanning transmission electron microscope of three oxidation states of Mn. SpectralNet was trained to learn the differences between them. We prove generalizability by training SpectralNet on electron energy loss spectroscopy data from one instrument, and test it on a variety of reference spectra found in the literature with perfect accuracy. We also test SpectralNet against a wide variety of high noise samples which a trained human spectroscopist would find incomprehensible. We also compare other neural network architectures used in the literature and determine that SpectralNet, a dense-layer free neural network, is immune to calibration differences whereas other styles of network are not. / Thesis / Master of Applied Science (MASc) / Spectroscopy is the study of the interaction between photons or electrons and a material to determine what that material is made of. One advanced way to make accurate measurements down to the atomic scale is to use high energy electrons in a transmission electron microscope. Using this instrument, a special type of photograph can be taken of the material (a spectrograph or spectrum) which is detailed enough to identify which kinds of atoms are in the material. The spectrographs are very complicated to interpret and the human eye struggles to find patterns in noisy and low resolution data. Depending on which instrument that the spectrographs are taken on, the resulting spectrograph will also change which adds extra difficulty. In this study, advanced algorithms are used to identify which types of atoms can be identified in the noisy signal from the spectrograph regardless of which instrument is used. These algorithms (convolutional neural networks) are also used in self-driving cars for a similar task of identifying objects whereas in this study we use it for identifying atoms.

Human bone tissue has an inherent hierarchical structure, which is integral to its material properties. It is primarily composed of a collagen fiber matrix that is mineralized with hydroxyapatite. A comprehensive understanding of bone and the linkages between structural and cellular organization is imperative to developing fundamental knowledge that can be applied to better our understanding of bone disease manifestations and its interaction with implant devices. Herein, this thesis investigated non-traditional methods for evaluating bone structure across the nano- and meso-length scales. Firstly, due to the inhomogeneous organization of collagen fibrils and mineral platelets of bone ultrastructure, a suitable methodology for the investigation of both phases needed to be generated. In this work, focused ion beam (FIB) microscopy was employed to create site-specific scanning transmission electron microscopy (STEM) lift-outs of human osteonal bone that could be visualized with correlatively with STEM and small angle X-ray scattering (SAXS). Samples were successfully characterized using both techniques, and minimal visual damage was induced during data acquisition. This work is the first to demonstrate the potential for bone to be investigated correlatively using both STEM and SAXS. Secondly, this work is the first to employ a dual-beam plasma FIB (PFIB) equipped with a scanning electron microscope (SEM), to investigate bone tissue across the mesoscale. This equipment enables large volume three-dimensional (3D) imaging at nanoscale resolution across larger mesoscale volumes. This thesis aimed to reduce ion beam-based artifacts, which presents as curtain-like features by adjusting the composition of protective capping layers. Subsequently, large volume tomograms of bone tissue were acquired, demonstrating the effectiveness of the PFIB to reveal mesoscale features including the cellular network of bone tissue. Overall, this thesis has developed methods that allow for the application of advanced microscopy techniques to enhance the understanding of bone tissue across the nanoscale and mesoscale. / Thesis / Master of Applied Science (MASc) / Bone tissue has a unique structure that perplexes both biologists and materials scientists. The hierarchical structure of bone has garnered the interest of materials scientists since the body’s skeletal strength and toughness are governed by the nanoscale (millionth of centimetres) to macroscale (centimeters) organization of bone. In this work, the intricate organization of bone is investigated using advanced electron and ion beam microscopy techniques, which achieve high-resolution imaging of bone structure. Firstly, this work developed a sample preparation workflow to correlate electron and X-ray imaging of the same bone tissue. Secondly, this work was the first to apply serial-sectioning plasma focused ion beam tomography to human bone tissue to investigate its structure at high resolution across micron-sized volumes. Here, previously unexplored methodologies to image bone are demonstrated with the hopes of applying such techniques to investigate healthy and pathological bone tissue in the future.