A series of cermet powders of composition NixCo(1-x)O-YSZ were synthesized for testing as cermet anode materials for SOFCs. The Co is found by powder XRD to become incorporated into the crystal lattice of the NiO, thus forming a true alloyed material. SEM and EDS results show two types of particles upon sintering to 1380oC: small, amorphous particles of YSZ and large, crystalline particles of nickel. The electrochemical oxidation of hydrogen on a cermet anode composed of Ni0.7Co0.3O-YSZ was investigated using a series of many button cells. Through EIS data, cyclic voltammetry data, the exchange current densities for these button cells were determined. Although a relatively large variation was found (expected to be due to microstructural variation) the average values for both methods of measurement is in good agreement in hydrogen. Following reduction in pure hydrogen, the fuel was changed to a mixture with high concentration of H2S. It was found that a concentration of 10 % H2S/H2 produced a sudden change in anode microstructure and resulted in loss of exchange current density. Lowering the amount of H2S in the initial fuel feed, which allowed for a more gradual microstructural change, allowed the cell to eventually function at concentrations in excess of 10 % H2S/H2. It was determined by OCV values in various concentrations of H2S/H2 that hydrogen is the predominant fuel of choice, even if H2S is available. Following electrochemical testing, slow cooling in a 10 % H2S/H2 mixture following produced metal sulfide spheres, as determined by SEM and EDS. Investigation in hydrocarbon, alcohol and biodiesel fuels was then undertaken to test the fuel variability of the given cermet anode material. Methane containing 10 % H2S was found to have increased exchange current density relative to poisoned hydrogen. Ethane and biodiesel experienced no increase in exchange current density, but a lengthening of the functional lifetime of the cell was observed, indicating reduced carbon poisoning. Methanol is a promising oxygen-containing SOFC fuel since it produced exchange current density values larger than hydrogen, and showed no evidence of coke formation by post-mortem SEM. Since oxygen-containing fuels are known to decompose in the gas phase at typical SOFC operating temperatures, the performance in a mixture of various CO/H2 fuels was then investigated. The Ni0.7Co0.3O-YSZ cermet anode gave higher exchange current density values for low ratio of CO/H2 fuels in the range 20/80 and 30/70 compared to pure H2. This is the first example of a Ni-based anode providing higher performance with a CO/H2 mixed fuel than for a pure H2 fuel. Finally, continuous running of a cell with fuel ratio 25/75 CO/H2 for 7 days produced exchange current density values, which were observed to increase significantly above the values for pure H2 during days 1-4 followed by deterioration below the value for hydrogen on subsequent days.

京都大学 / 0048 / 新制・課程博士 / 博士(人間・環境学) / 甲第21433号 / 人博第871号 / 2018||人博||871(吉田南総合図書館) / 京都大学大学院人間・環境学研究科相関環境学専攻 / (主査)教授 内本 喜晴, 教授 吉田 寿雄, 准教授 戸﨑 充男 / 学位規則第4条第1項該当 / Doctor of Human and Environmental Studies / Kyoto University / DFAM

This thesis focuses on an experimental study of the anode area of the hybrid water- gas DC arc plasma torch that is used in many industrial applications, including plasma spraying, hydrocarbon reforming, pyrolysis, and organic waste gasification. The effects of ambient pressure and plasma generation conditions on the torch's plasma jet were studied, with particular focus on the torch's anode area. Movement of the anode arc attachment is described in detail, including its speed, range of its motion on the anode surface, restrike periods, and the frequency of its many sudden decelerations and re-accelerations. It was found that the anode erosion can be compared relatively simply by quick processing of high-speed camera videos. The anode erosion was also measured directly. Many electric probe measurements were made in the anode area of this plasma torch for the first time. By using these electric probes, shock waves, turbulent vortices, and plasma potential fluctuations were studied directly. It was found that a mean plasma electric field and a mean plasma electrical conductivity in the anode area can be satisfactorily estimated also non-intrusively by quick processing of high-speed camera videos. Moreover, schlieren videos of the plasma jet in the anode area were created. 1

Lithium-ion batteries (LIBs) are receiving significant attention from both academia and industry as one of the most promising energy storage and conservation devices due to their high energy density and excellent safety. Graphite, the most widely used anode material, with limitations on energy density, can no longer satisfy the requirements proposed by new applications. Therefore, further improvement on the electrochemical performance of anodes has been long pursued, along with the development of new anode materials. Among potential candidates, Si and Sn based anodes are believed to be the most promising. However, the dramatic volume expansion upon Li-intercalation and contraction upon Li de-intercalation cause mechanical instability, and thus cracking of the electrodes. To overcome this issue, many strategies have been explored. Among them the most efficient strategies include introduction of a nanostructure, coupled with a buffering matrix and coating with a protective film. However, although cycling life has been significantly increased using these three strategies, the capacity retention still needs improvement, especially over extensive charge-discharge cycles. In addition, more efforts are still needed to develop new fabrication methods with low costs and high efficiency. To further improve mechanical stability of electrodes, understanding of the failure mechanisms, particularly, the failure mechanisms of Si and Sn nanomaterials is essential. Therefore, some of the key factors including materials fabrication and microstructural changes during cycling are studied in this work. Hollow Si nanospheres have proved to be have a superior electrochemical performance when applied as anode materials. However, most of fabrication methods either involve use of processing methods with low throughput, or expensive temporary templates, which severely prohibits large-scale use of hollow Si spheres. This work designed a new template-free chemical synthesis method with high throughput and simple procedures to fabricate Si hollow spheres with a nanoporous surface. The characterization results showed good crystallinity and a uniform hollow sphere structure. The substructure of pores on the surface provides pathways for electrolyte diffusion and can alleviate the damage by the volume expansion during lithiation. The success of this synthesis method provides valuable inspiration for developing industrial manufacturing method of hollow Si spheres.3D graphene is the most promising matrix that can provide the necessary mechanical support to Sn and Si nanoparticles during lithiation. 2D graphene, however, results in Sn/graphene nanocomposites with a continuous capacity fade during cycling. It is anticipated that this is due to microstructural changes of Sn, however, no studies have been performed to examine the morphology of such cycled anodes. Hence, a new Sn/2D graphene nanocomposite was fabricated via a simple chemical synthesis, in which Sn nanoparticles (20-200 nm) were attached onto the graphene surface. The content of Sn was 10 wt.% and 20 wt.%. These nanopowders were cycled against pure Li-metal and, as in previous studies, a significant capacity decrease occurred during the first several cycles. Transmission and scanning electron microscopy revealed that during long term cycling electrochemical coarsening took place, which resulted in an increased Sn particle size of over 200 nm, which could form clusters that were 1 m. Such clusters result in a poor electrochemical performance since it is difficult for complete lithiation of the Sn to occur. It is hence concluded that the inability of Sn/2D graphene anodes to retain high capacities is due to coarsening that occurs during cycling. In addition to using forms of carbon to buffer the Sn expansion, it has been proposed to alloy Sn with S, which has a low redox potential vs Li⁰/Li⁺. Therefore, another new anode proposed here is that of SnS attached to graphite. The as prepared powders had a flower-like structure of the SnS alloy. Electrochemical cycling and subsequent microstructural analysis showed that after electrochemical cycling this pattern was destroyed and replaced by Sn and SnS nanoparticles. Based on the electron microscopy and XRD analysis, it was concluded that selective leaching of S occurs during lithiation of SnS particles, which results into nano SnS and Sn particles to be distributed throughout the electrolyte or SEI layer, without being able to take part in the electrochemical reactions. This mechanism has not been noted before for SnS anodes and indicates that it may not be possible to retain the initial morphology of SnS alloy during cycling, or the ability of SnS to be active throughout long term cycling. To conclude it should be stated that the goal and novelty of this thesis was (i) the fabrication of new Si, Sn/graphene and SnS/C nanostructures that can be used as anodes in Li-ion batteries and (ii) the documentation of the mechanisms that disrupt the initial structural stability of Sn/2D graphene and SnS/C anodes and result in severe capacity loss during long term cycling (over 100 cycles). These systems are of high interest to the electrochemistry community and battery developers.

Ce travail a concerné les anodes tournantes légères utilisées pour la production de rayons X dans les scanners médicaux. Ces anodes sont constituées de graphite revêtu de tungstène par projection plasma à basse pression. Nous nous sommes intéressés au comportement mécanique des matériaux constituants (particulièrement le tungstène) dans les conditions de sollicitations thermomécaniques intenses rencontrées durant la vie en service des anodes. La première étape de ce travail a été une phase d'identification du problème posé. Nous avons ainsi cerné les conditions de sollicitation en service en termes de fréquence d'excitation, de température, de déformation et de vitesse de déformation. Des expertises d'anodes usagées ont permis de dégager les principaux modes d'endommagement en service. Pour caractériser le comportement mécanique des matériaux, un dispositif d'essai existant a été adapté pour pouvoir travailler sur des éprouvettes de flexion de petite taille. Les procédures d'essai ont, quant à elles, été choisies pour générer des sollicitations proches de celles associées au fonctionnement. Dans toute la plage de température investiguée (ambiante à 1800°C), notre graphite s'est avéré posséder un comportement élastique fragile. Le comportement du tungstène s'est révélé plus complexe: élastique-fragile jusqu'à 400°C, il devient ensuite plastique pour devenir sensible au fluage à partir de 1200°C. Des trajets de chargement originaux ont permis de mettre en évidence l'existence d'une contrainte interne de rappel, ainsi qu'un couplage entre déformations plastiques et déformations visqueuses. Une loi phénoménologique originale de comportement, à double déformation inélastique, plastique et viscoplastique avec un terme d'interaction entre les deux mécanismes d'écoulement, a donc été nécessaire pour décrire le comportement mécanique du tungstène. L'écrouissage par chacun des écoulements est traduit par des variables cinématiques. L'identification numérique des paramètres a été réalisée par un optimiseur, couplé à un code de calcul par éléments finis simulant l'essai de flexion. La loi ainsi établie a été finalement validée en confrontant ses prévisions au comportement observé expérimentalement pour des trajets de chargement complexes. Cette loi de comportement a finalement été utilisée pour simuler les conditions d'utilisation d'une anode réelle. On a ainsi calculé, à l'aide d'un maillage bidimensionnel axisymétrique, les contraintes générées par le refroidissement post-recuit, une puis plusieurs séries de clichés radiographiques, et enfin un refroidissement total après utilisation. La répétition de séries de clichés conduit rapidement à des cycles stabilisés. Contrairement aux simples simulations thermoélastiques réalisées préalablement à ce travail, les niveaux de contraintes calculés sont réalistes et demeurent inférieurs à la résistance à la rupture des matériaux. Cette simulation peut d'ores et déjà être utilisée industriellement pour évaluer l'influence d'une modification de la géométrie de l'anode et/ou des conditions de sollicitation sur les contraintes en service.

In this study we looked at several different silicon nanostructures grown for the purpose of optimizing anodes for lithium ion batteries. We primarily focused on two distinct types of structures, nanospirals, and Rugate structures. The samples were designed to have the mechanical robustness to endure the massive expansion caused by lithiation of silicon. All of the samples were grown using an electron beam evaporator. Scanning electron microscope images show that we have achieved the desired structural growth. The spirals were shown to have an average diameter of 343 nm on polished copper, and 366 nm on unpolished copper. The Rugate structures had two distinct sample sets. The first mimicked the design of a thin film. The other formed distinct pillars that grouped into islands. The tops of the islands had an average diameter of 362 nm, while the pillars had an average width varying between 167 nm and 140 nm.

Thesis advisor: Dunwei Wang / When their electrodes are made of nanomaterials or materials with nanoscale features, devices for energy conversion and energy storage often exhibit new and improved properties. One of the main challenges in material science, however, is to synthesize these nanomaterials with designed functionality in a predictable way. This thesis presents our successes in synthesizing TiSi₂ nanostructures with various complexities using a chemical vapor deposition (CVD) method. Attention has been given to understanding the chemistry guiding the growth. The governing factor was found to be the surface energy differences between various crystal planes of orthorhombic TiSi₂ (C54 and C49). This understanding has allowed us to control the growth morphologies and to obtain one-dimensional (1D) nanowires, two-dimensional (2D) nanonets and three-dimensional (3D) complexes with rational designs by tuning the chemical reactions between precursors. Among all these morphologies, the 2D nanonet, which is micrometers wide and long but only approximately 15 nm thick, has attracted great interest because it is connected by simple nanostructures with single-crystalline junctions. It offers better mechanical strength and superior charge transport while preserving unique properties associated with the small-dimension nanostructure, which opens up the opportunity to use it for various energy related applications. In this thesis we focus on its applications in lithium ion batteries. With a unique heteronanostructure consisting of 2D TiSi₂ nanonets and active material coating, we demonstrate the performances of both anode and cathode of lithium ion batteries can be highly improved. For anode, Si nanoparticles are deposited as the coating and at a charge/discharge rate of 8400 mA/g, we measure specific capacities >1000 mAh/g with only an average of 0.1% decay per cycle over 100 cycles. For cathode, V₂O₅ is employed as an example. The TiSi₂/V₂O₅ nanostructures exhibit a specific capacityof 350 mAh/g, a power rate up to 14.5 kW/kg, and 78.7% capacity retention after 9800 cycles. In addition, TiSi₂ nanonet itself is found to be a good anode material due to the special layer-structure of C49 crystals. / Thesis (PhD) — Boston College, 2012. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Chemistry.

Este estudo visa a definição de condições de síntese e processamento cerâmico que possibilitem a obtenção do componente anódico com características adequadas para a operação de uma SOFC (Solid Oxide Fuel Cell), ou seja, boa distribuição microestrutural do NiO na matriz de YSZ e porosidade cerca de 30% após redução de NiO. As rotas de síntese selecionadas englobaram a coprecipitação em meio amoniacal, mistura mecânica dos pós e combustão a partir de sais de nitrato. As técnicas de caracterização de pós empregadas incluíram a difração de raios X, microscopia eletrônica de varredura, microscopia eletrônica de transmissão, difração a laser, adsorção gasosa (BET) e picnometria de hélio. Os resultados obtidos indicaram que empregando-se a técnica de coprecipitação, a perda de Ni2+, na forma de complexo [Ni(NH3)n]2+, pode ser minimizada pelo controle do pH em 9,3, mantendo-se a concentração de Ni2+ na solução inicial em 0,1M. No método de mistura mecânica a melhor condição de dispersão dos pós, sem a sedimentação diferencial, foi obtida para valores de potencial zeta em pH 8,0, fixando-se a concentração de dispersante em 0,8% em massa. Na síntese por combustão observou-se que para composições pobres em combustível, os produtos finais apresentaram-se amorfos e com alta área superficial (120,2 m2.g-1). Para as composições ricas em combustível, uréia, os pós obtidos apresentaram-se cristalinos sendo que a intensidade das reflexões do padrão de DRX aumenta com o aumento do excesso de combustível, devido ao aumento da temperatura de reação. No estudo de sinterabilidade dos compactados preparados a partir de pós preparados pelos três métodos determinou-se a temperatura ao redor de 1300 ºC para máxima taxa de densificação e porosidade entre 6,0 e 14%. Os resultados da redução em atmosfera de H2 dos compósitos confirmam que a cinética de reação ocorre em duas etapas, sendo que a primeira etapa com comportamento linear é controlada por reação química na superfície. Na segunda etapa a redução passa a ser controlada pela difusão do gás nos micros poros, gerados pela redução do NiO, diminuindo a taxa de redução. / This study aim the definition of synthesis and ceramic processing conditions of the anodic component suitable for operation of SOFC, i.e, homogeneous distribution of NiO in YSZ matrix and porosity after reduction above 30%. The selected synthesis routes included the co-precipitation in ammonia media, mechanical mixing of powders and combustion reaction from nitrate salts. The characterization techniques of powders included the X-ray diffraction, scanning and transmission electron microscopy, laser diffraction, nitrogen gas adsorption technique (BET) and Helium pycnometry. The obtained results indicated that the loss of Ni2+ in co-precipitation process, due to the formation of complex [Ni(NH3)n]2+, can be minimized by controlling the pH around 9.3, keeping the concentration of nickel cation in the solution to be precipitated around 0.1M. In the mechanical mixing method the best condition of powder dispersion, without differential sedimentation, was obtained for zeta potential values at pH around 8.0, fixing the dispersant concentration at 0.8%. For the combustion synthesis it was observed that when stoichiometric and twofold stoichiometric urea was used, amorphous phase was formed and a higher surface area was attained in the final products. Employing the fuel-rich solution condition, crystallization of the powder was observed and the relative intensity of reflections of XRD patterns increased with excess of fuel, due to increasing the reaction temperature. Sinterability studies of pellets prepared from powder synthesized by the three routes described above showed the temperature around 1300 º C for maximum rate densification and porosity between 6.0 and 14%. Reduction results of the composites confirmed that the reduction kinetics occurs in two steps. The first one with a linear behavior and controlled by chemical reaction on the surface. The second reduction step is the reduction that is controlled by gas diffusion in micro pores, generated by reduction of nickel oxide, decreasing the rate of reduction.

Current ethanol oxidation catalysts in direct ethanol fuel cells (typically platinum-based) suffer from low conversion and are susceptible to CO poisoning. Therefore we determined to find viable alternative catalysts for ethanol oxidation based on iridium using density functional theory to model bimetallic alloy (111) surfaces. Iridium was alloyed with another transition metals M in an overlayer (one layer of metal M on top of bulk iridium) or subsurface configuration (M is inserted under the first layer of iridium). Complete oxidation of ethanol is limited by the breaking of strong C-C bonds, so any catalyst must lower the barriers for C-C bond breaking. We modeled the reaction CH+CO →CHCO.Segregation energies were calculated and the subsurface configuration was the most stable configuration in the vast majority of alloy cases. CO adsorption was also studied and a lower CO adsorption energy was found in many alloy cases compared to pure Pt (, providing encouraging results about the possibility of reducing CO poisoning. Activation energies were lowered for the vast majority of the alloys used in an underlayer structure, reinforcing our interest in the underlayer structures or “subsurfaceâ€� alloys. Finally, we found, based on the CO adsorption energies, activation energies of the C-C breakage reaction, and metal cost, three important catalyst descriptors, a number of promising catalysts for the ethanol oxidation reaction. The most interesting alloys all adopted the underlayer structure Ir/M/Ir. With M = Ta, Hf, Nb, V, Zr, they demonstrated enhanced reactivity and high CO tolerance, having the advantage of reducing the cost of the catalyst, potentially substituting expensive platinum group metals by more affordable components.

Este estudo visa a definição de condições de síntese e processamento cerâmico que possibilitem a obtenção do componente anódico com características adequadas para a operação de uma SOFC (Solid Oxide Fuel Cell), ou seja, boa distribuição microestrutural do NiO na matriz de YSZ e porosidade cerca de 30% após redução de NiO. As rotas de síntese selecionadas englobaram a coprecipitação em meio amoniacal, mistura mecânica dos pós e combustão a partir de sais de nitrato. As técnicas de caracterização de pós empregadas incluíram a difração de raios X, microscopia eletrônica de varredura, microscopia eletrônica de transmissão, difração a laser, adsorção gasosa (BET) e picnometria de hélio. Os resultados obtidos indicaram que empregando-se a técnica de coprecipitação, a perda de Ni2+, na forma de complexo [Ni(NH3)n]2+, pode ser minimizada pelo controle do pH em 9,3, mantendo-se a concentração de Ni2+ na solução inicial em 0,1M. No método de mistura mecânica a melhor condição de dispersão dos pós, sem a sedimentação diferencial, foi obtida para valores de potencial zeta em pH 8,0, fixando-se a concentração de dispersante em 0,8% em massa. Na síntese por combustão observou-se que para composições pobres em combustível, os produtos finais apresentaram-se amorfos e com alta área superficial (120,2 m2.g-1). Para as composições ricas em combustível, uréia, os pós obtidos apresentaram-se cristalinos sendo que a intensidade das reflexões do padrão de DRX aumenta com o aumento do excesso de combustível, devido ao aumento da temperatura de reação. No estudo de sinterabilidade dos compactados preparados a partir de pós preparados pelos três métodos determinou-se a temperatura ao redor de 1300 ºC para máxima taxa de densificação e porosidade entre 6,0 e 14%. Os resultados da redução em atmosfera de H2 dos compósitos confirmam que a cinética de reação ocorre em duas etapas, sendo que a primeira etapa com comportamento linear é controlada por reação química na superfície. Na segunda etapa a redução passa a ser controlada pela difusão do gás nos micros poros, gerados pela redução do NiO, diminuindo a taxa de redução. / This study aim the definition of synthesis and ceramic processing conditions of the anodic component suitable for operation of SOFC, i.e, homogeneous distribution of NiO in YSZ matrix and porosity after reduction above 30%. The selected synthesis routes included the co-precipitation in ammonia media, mechanical mixing of powders and combustion reaction from nitrate salts. The characterization techniques of powders included the X-ray diffraction, scanning and transmission electron microscopy, laser diffraction, nitrogen gas adsorption technique (BET) and Helium pycnometry. The obtained results indicated that the loss of Ni2+ in co-precipitation process, due to the formation of complex [Ni(NH3)n]2+, can be minimized by controlling the pH around 9.3, keeping the concentration of nickel cation in the solution to be precipitated around 0.1M. In the mechanical mixing method the best condition of powder dispersion, without differential sedimentation, was obtained for zeta potential values at pH around 8.0, fixing the dispersant concentration at 0.8%. For the combustion synthesis it was observed that when stoichiometric and twofold stoichiometric urea was used, amorphous phase was formed and a higher surface area was attained in the final products. Employing the fuel-rich solution condition, crystallization of the powder was observed and the relative intensity of reflections of XRD patterns increased with excess of fuel, due to increasing the reaction temperature. Sinterability studies of pellets prepared from powder synthesized by the three routes described above showed the temperature around 1300 º C for maximum rate densification and porosity between 6.0 and 14%. Reduction results of the composites confirmed that the reduction kinetics occurs in two steps. The first one with a linear behavior and controlled by chemical reaction on the surface. The second reduction step is the reduction that is controlled by gas diffusion in micro pores, generated by reduction of nickel oxide, decreasing the rate of reduction.