La cathode est responsable d’environ 10 % de la chute de potentiel totale d’une cuve Hall- Héroult. Une partie significative de cette perte, sous forme d’effet Joule, provient d’un mauvais contact entre la couche de fonte et le bloc de carbone. De plus, une distribution non uniforme de la densité de courant à l’intérieur de la cuve engendre une érosion prématurée des extrémités des blocs cathodiques, limitant la durée de vie des cuves. Le présent projet vise à réduire la chute de potentiel ainsi qu’à uniformiser la densité de courant de l’assemblage cathodique par l’amélioration du contact entre la barre collectrice et le bloc de carbone. Il explore plus particulièrement l’utilisation de barres collectrices cylindriques comprenant des insertions de cuivre ne nécessitant aucune couche de fonte lors de l’opération de scellage. Différentes configurations de cathode sont explorées à l’aide d’un modèle numérique thermoélectromécanique dans le but d’en comprendre le comportement et d’évaluer leur impact sur la consommation énergétique et sur la durée de vie d’une cuve. Une analyse économique est également réalisée afin de mesurer la rentabilité des concepts. Celle-ci sert finalement à l’optimisation de la géométrie afin de maximiser les performances de nouveaux concepts. Les résultats indiquent que la chute de voltage peut être réduite et que la distribution de courant peut être uniformisée par l’amélioration de la qualité du contact entre le bloc de carbone et les barres collectrices. / The cathode, located at the bottom of a Hall-Héroult cell, is responsible for nearly 10 % of the total cell voltage drop. Poor contact between the carbon block and the cast-iron layer surrounding the collector bars increase energy losses in the form of Joule heating. In addition, a non-uniform current density distribution inside the cell results in premature erosion of the carbon block extremities, limiting the cell’s life expectancy. This project aims to reduce the voltage drop and to improve the current density uniformity in the cathode assembly by improving the contact between the collector bars and the carbon block. To do so, a new cathode design using cylindrical collector bars with copper inserts is investigated using finite element modeling. The thermal expansion during the cell start up is used to generate contact between the collector bars and the carbon block, thus not requiring any cast iron or sealing operations. Different geometry configurations are explored using a thermo-electro-mechanical model to understand its behavior and to determine their effect on energy consumption and cell life expectancy. An economic analysis is also performed to evaluate the cost effectiveness of those configurations. The geometry is finally optimized to maximize the new design’s performance. Results indicate that a significant voltage drop reduction and a more uniform current distribution can be achieved by improving the contact quality between the carbon block and the collector bars.

Le stockage de l'énergie est l'un des enjeux les plus cruciaux du 21e siècle. Le développement de matériaux abordables qui possèdent une grande densité d'énergie et qui affichent une grande stabilité est recherché. Une demande croissante venant du domaine de l'électronique portative fait pression sur la recherche de matériaux toujours plus performants. L'émergence des ordinateurs et téléphones portatifs ainsi que des véhicules électriques est la pièce maitresse de cette révolution. Par ailleurs, le stockage de l'énergie dans des batteries géantes, mais stationnaires, permettra au cours des prochaines années de pallier à la réalité de production d'énergie fluctuante du solaire et de l'éolien au cours d'une journée. La batterie Li-ion est présentement la technologie la plus mature pour mener à ce type de réalisation. L'atome de lithium est pourvu d'une petite masse molaire et l'ion lithium possède un petit rayon ionique. Utilisé à l'anode, le lithium permet d'y avoir une grande densité d'énergie, puis une faible résistance ionique dans l'électrolyte une fois oxydé. Par contre, les batteries Li-ion d'aujourd'hui reposent sur des matériaux de cathode dispendieux comme le cobalt, le nickel et le manganèse, dont l'exploitation soulève de grandes questions environnementales et éthiques. Avec une demande croissante pour des batteries de haute performance, des matériaux de cathode abordables, renouvelables et avec un impact environnemental faible doivent être développés. Dans ce contexte, les molécules organiques qui ont une activité redox ont attiré l'attention avec un faible cout de production, une faible toxicité et une abondance naturelle élevée. Parmi les différents groupements fonctionnels démontrant une activité rédox, les groupements carbonylés se démarquent par leur grande diversité, et leur stabilité à l'état réduit. Les matériaux redox typiques contenant des carbonyles sont les quinones, les 1,2-diones et les imides qui reposent sur un mécanisme d'énolisation lors du processus de réduction. La principale limitation que présentent ces molécules est la dissolution dans l'électrolyte. La formation d'un sel organique ou l'incorporation de la molécule électroactive au sein d'un polymère inerte sont des stratégies qui ont été apportées pour pallier à ce problème. La versatilité des molécules possédant des fonctions imides rend possible l'étude de plusieurs polymères π-conjugués qui ont l'avantage de pouvoir conduire davantage les charges injectées. Dans le cadre de ces travaux de doctorat, l'objectif général était de synthétiser de nouveaux polymères π-conjugués contenant des fonctions imides et d'analyser leurs performances en tant que matériau actif de cathode en batterie Li-ion. Les molécules qui ont été étudiées sont le maléimide, le pyromellitique diimide et le pyrène diimide. Des polymères π-conjugués ont été synthétisés avec ces unités en utilisant les techniques d'Ullmann, de Stille, de Suzuki ou d'arylation directe.

Approved for public release, distribution is unlimited / Present-day high power microwave devices suffer from a lack of reliable, reproducible cathodes for generating the requisite GW-level electron beam in a vacuum. Standard explosive emission cathode pulse durations have been limited to 10's or 100's of ns due to the expansion of cathode-generated plasma and the ensuing impedance collapse that debilitates microwave output. Traditional thermionic cathodes do not suffer from this drawback of plasma generation, but have not yet been able to provide the required emission current densities explosive emission cathodes are capable of. It is expected that if the plasma could be made cooler and less dense, explosive emission would be more stable. Cesium iodide (CsI) has been found to slow the impedance collapse in many explosive emission cathodes. Herein we will experimentally examine diode impedance collapse, gas production, and cathode conditioning in an effort to perform an evaluation of explosive cathode performance in a typical thermionic electron gun environment. These results will then be used to help demarcate the parameter space over which these CsI-coated carbon fiber cathodes are viable candidates for the electron beam source in next-generation high power microwave devices. / Lieutenant, United States Navy

Oxygen gain is the difference in fuel cell performance operating on oxygen-depleted and oxygen-rich cathode fuel streams. Oxygen gain experiments provide insight into the degree of oxygen mass-transport resistance within a fuel cell. By taking these measurements under different operating conditions, or over time, one can determine how oxygen mass transport varies with operating modes and/or aging. This paper provides techniques to differentiate between mass-transport resistance within the catalyst layer and within the gas-diffusion medium for a polymer-electrolyte membrane fuel cell. Two extreme cases are treated in which all mass transfer limitations are located only (i) within the catalyst layer or (ii) outside the catalyst layer in the gas diffusion medium. These two limiting cases are treated using a relatively simple model of the cathode potential and common oxygen gain experimental techniques. This analysis demonstrates decisively different oxygen gain behavior for the two limiting cases. For catalyst layer mass transfer resistance alone, oxygen gain values are limited to a finite range of values. However, for gas diffusion layer mass transfer resistance alone, the oxygen gain is not confined to a finite range of values. This analysis is then extended to evaluate ionic effects within the catalyst layer. / text

Mémoire (M.Sc.A)--Université du Québec à Chicoutimi, 1989. / Document électronique également accessible en format PDF. CaQCU

By replacing the intercalation electrode with a porous electrode and allowing lithium to react directly with O₂ from the air, the new rechargeable Li/O₂ battery system was studied. The porous cathode is comprised of carbon, catalyst and binder. The effect of every component was investigated. The catalyst was believed to play an important role in the performance of the electrode. A number of potential materials have been examined as the catalyst for the O₂ electrode. It suggests that the nature of the catalyst is a key factor controlling the performance of the O₂ electrode. Several catalysts based on first row transition metal oxides each with three different morphologies, bulk, nanoparticulate and mesoporous were studied. The influence of the morphology on the discharge and charge voltage, discharge capacity and cyclability were examined. Among all the catalysts studied, α-MnO₂ nanowires was found to be the best candidate. The reversible capacities of 3000 mAhg⁻¹(normalised by the mass of carbon) or 505 mAhg⁻¹ (based on the total mass of cathode + O₂ ) was obtained. Some of other factors, such as type of carbon, type of binder, type of electrolyte, the construction of cathode and the modification of the catalyst were also investigated, even just in the early stage. Capacity fading during cycling is the main problem in all the cases. A number of experiments were carried out to understand and attempt to avoid the fading problem. After successful synthesis of mesoporous α-Fe₂O₃ with unique properties (by Jiao et al.), the application of these materials in conventional Li battery was studied. Mesoporous α-Fe₂O₃ with ordered walls, mesoporous α-Fe₂O₃ with disordered walls and Fe₂O₃ nanoparticles were examined. It was also applied to examine the different factors that influence the rate of conversion electrodes, i.e., Li⁺ and e⁻ transport to and within the particles, as well as the rate of the two-phase reaction, demonstrating that for this conversion reaction electron transport to and within the particles is paramount.

Currently, automotive batteries use intercalation cathodes such as lithium iron phosphate (LiFePO4) which provide high levels of safety while sacrificing cell voltage and therefore energy density. Lithium transition metal oxide (LiMO2) batteries achieve higher cell voltages at the risk of releasing oxygen gas during charging, which can lead to ignition of the liquid electrolyte. To achieve both safety and high energy density, oxide cathodes must be well characterized under operating conditions. In any intercalation cathode material, the loss of positive lithium ions during charge must be balanced by the loss of negative electrons from the host material. Ideally, the TM ions oxidize to compensate this charge. Alarmingly, the stoichiometry of the latest LiMO2 cathode materials includes more lithium ions than the TM ions can compensate for. Inevitably, peroxide ions or dioxygen gas must form. The former mechanism is vital for lithium-air batteries, while the latter must be avoided. Battery researchers have long sought to completely characterize the intercalation reaction in working batteries. However, the volatile electrolytes employed in batteries are not compatible with vacuum-based characterization techniques, nor are the packaging materials required to contain the liquid. For the first time, a solid state battery (using exposed particles of Li1.17Ni0.25Mn0.58O2) was charged while using soft X-ray absorption spectroscopy to observe the redox trends in nickel, manganese and oxygen. This was combined with innovative hard X-ray absorption spectroscopic studies on the same material to create the most complete picture yet possible of charge compensation.

This study has been mainly focused on high temperature solid oxide electrolysis cells (HT-SOECs) for steam electrolysis. The compositions, microstructures and metal catalysts for SOEC cathodes based on (La₀.₇₅Sr₀.₂₅)₀.₉₅Mn₀.₅Cr₀.₅O₃ (LSCM) have been investigated. Hydrogen production amounts from SOECs with LSCM cathodes have been detected and current-to-hydrogen efficiencies have been calculated. The effect of humidity on electrochemical performances from SOECs with cathodes based on LSCM has also been studied. LSCM has been applied as the main composite in HT-SOEC cathodes in this study. Cells were measured at temperatures up to 920°C with 3%steam/Ar/4%H₂ or 3%steam/Ar supplied to the steam/hydrogen electrode. SOECs with LSCM cathodes presented better stability and electrochemical performances in both atmospheres compared to cells with traditional Ni cermet cathodes. By mixing materials with higher ionic conductivity such as YSZ(Y₂O₃-stabilized ZrO₂ ) and CGO(Ce₀.₉Gd₀.₁O₁.₉₅ ) into LSCM cathodes, the cell performances have been improved due to the enlarged triple phase boundary (TPB). Metal catalysts such as Pd, Fe, Rh, Ni have been impregnated to LSCM/CGO cathodes in order to improve cell performances. Cells were measured at 900°C using 3%steam/Ar/4%H₂ or 3%steam/Ar and AC impedance data and I-V curves were collected. The addition of metal catalysts has successfully improved electrochemical performances from cells with LSCM/CGO cathodes. Improving SOEC microstructures is an alternative to improve cell performances. Cells with thinner electrolytes and/or better electrode microstructures were fabricated using techniques such as cutting, polishing, tape casting, impregnation, co-pressing and screen printing. Thinner electrolytes gave reduced ohmic resistances, while better electrode microstructures were observed to facilitate electrode processes. Hydrogen production amounts under external potentials from SOECs with LSCM/CGO cathodes were detected by gas chromatograph and current-to-hydrogen efficiencies were calculated according to the law of conservation of charge. Current-to-hydrogen efficiencies from these cells at 900°C were up to 80% in 3%steam/Ar and were close to 100% in 3%steam/Ar/4%H₂. The effect of humidity on SOEC performances with LSCM/CGO cathodes has been studied by testing the cell in cathode atmospheres with different steam contents (3%, 10%, 20% and 50% steam). There was no large influence on cell performances when steam content was increased, indicating that steam diffusion to cathode was not the main limiting process.

The research into materials for use as cathode materials for solid oxide fuel cells (SOFC) is ongoing, with many different avenues being investigated. Copper based delafossites were studied for cathode side applications in SOFCs, as a novel and comparatively cheap material. The aim was to identify suitable materials with appropriate electrical conductivity, thermal, chemical and mechanical stability in air. Furthermore, understanding the behaviour of the delafossites during the thermal oxidation to spinel and copper oxide would be beneficial to further development of the materials. The structure and properties of the copper based delafossites CuFeO₂, CuAlO₂ and CuCrO₂ were studied, alongside several doped compositions for each parent composition. The electronic conductivity of the CuFeO₂ family was improved by doping fluorine into the structure, with 1 atomic % doping producing ~3.8 S cm⁻¹ at 800 °C. However, as reported in literature the structure is vulnerable to oxidation at higher temperatures. In contrast, CuAlO₂ was stable over the SOFC temperature range, and therefore had appropriate thermal expansion coefficients (TEC) of ~11 x 10⁻⁶ K⁻¹, but relatively low electronic conductivity. CuCrO₂ compositions had good overall TECs, but aliovalent doping of Mg²⁺ improved the conductivity to ~17.1 S cm⁻¹ at 800°C for 2.5 atomic % doped CuCrO₂. Neutron diffraction was utilised to study members of the solid solution CuFe₁₋ₓCrₓO₂ (x = 0, 0.25 and 0.5) during in-situ oxidation at high temperature. Points of positive scattering density were identified within the CuFeO₂ structure, which were attributed to the location of the intercalated oxygen ions before the transformation proceeded. Additionally, the cation distribution between the tetrahedral and octahedral sites within the developing spinel were characterised for x = 0, and partially for the x = 0.25 and 0.5 compositions using complimentary XRD patterns. Finally, magnesium doped CuCrO₂ delafossites were used in several different preliminary symmetrical cells for study using electrochemical impedance spectroscopy (EIS). Pure delafossite inks gave relatively large area specific resistance (ASR) values, 1.29 - 2.69 Ω cm² at 800 °C. It was attempted to improve upon these values through infiltration of CeO₂ and through change in microstructure using composite type inks, without much success. Inks using CuCr₀.₈Fe₀.₂O₂ were also tested as both a single phase electrode and as a composite type electrode. The pure delafossite electrode still had a large ASR value, (~33.4 Ω cm² at 800 °C) while composite electrodes obtained much more respectable ASR values ~0.75 Ω cm² at 800 °C.

Scandate cathodes, where scandia is added to the tungsten cathode pellets, have recently received substantial and renewed research interest owing to significantly improved electron emission capabilities at lower temperatures, as compared with conventional dispenser cathodes. However, there are several persistent issues including non-uniform electron emission, lack of understanding regarding scandium’s role in the emission mechanism, and unreliable reproducibility in terms of scandate cathode fabrication. As a result, scandate cathodes have not yet been widely implemented in actual vacuum electron devices (VEDs). The surface structure and chemical composition of multiple scandate cathodes – prepared with the powder using the liquid-solid (L-S) technique – and exhibiting excellent emission behavior were characterized to give insight into the fundamental mechanism(s) of operation. This was achieved with high-resolution electron microscopy techniques that include high-precision specimen lift-out. These studies showed that the micron-sized tungsten particles that compose the largest fraction of the cathode body are highly faceted and decorated with nanoscale Ba/BaO (~10 nm), as well as larger (~150 nm) Sc2O3 and BaAl2O4 particles. The experimentally identified facets were confirmed through Wulff analysis of the tungsten crystal shape and were determined to consist of {110}, {100}, and {112} facets, in increasing order of surface area prevalence. Furthermore, it is estimated that Ba atoms decorating the tungsten crystal surfaces are present in quantities such that monolayer coverage is possible at elevated temperatures. The high-resolution electron microscopy techniques used to investigate the cross section (near-surface) of the L-S scandate cathodes also revealed that the BaAl2O4 particles (100-500 nm) that attach to the larger tungsten particles are either adjacent to the smaller Sc2O3 nanoparticles or encompass them. Furthermore, high-resolution chemical analysis and 3D elemental tomography show that the two oxides always appear to be physically distinct from each other, despite their close proximity. 3D elemental tomography also showed that the Sc2O3 particles can sometimes appear inside the larger tungsten particles, but are inhomogeneously distributed. Nanobeam electron diffraction confirmed that the crystal structure of the tungsten particles are body-centered cubic, and imply that the structure remains unchanged despite the numerous complex chemical reactions that take place throughout the impregnation and activation procedures. The role of Sc and the emission mechanism for scandate cathodes are discussed. Based on characterization results and materials computation, the role of Sc in scandate cathodes is possibly related to tuning the partial pressure of oxygen in order to establish an oxygen-poor atmosphere around the cathode surface, which is a necessary condition for the formation of the (near) equilibrium tungsten shape. A thin Ba-Sc-O surface layer (~8 nm) was detected near the surface of tungsten particles, using electron energy loss spectroscopy in the scanning transmission electron microscope. This stands in stark contrast to models invoking a ~100 nm Ba-Sc-O semiconducting surface layer, which are broadly discussed in the literature. These results provide new insights into understanding the emission mechanism of scandate cathodes.