A numerical model of the steady state and alternating current behaviour of a solid-oxide fuel cell is presented to explore the possibilities to diagnose and identify degradation mechanisms in a minimally invasive way using impedance spectroscopy. This is the first report of an SOFC impedance model to incorporate degradation, as well as the first one to include the ribbed interconnect geometry, using a 2-D approximation. Simulated degradation modes include: electrode/electrolyte delamination, interconnect oxidation, interconnect/electrode interface detachment, and anode sulfur poisoning. Detailed electrode-level simulation replaces the traditional equivalent circuit approach, allowing the simulation of degradation mechanisms that alter the shape of the current path. The SOFC impedance results from calculating the cell response to a small oscillatory perturbation in potential. Starting from the general equations for mass and charge transport, and assuming isothermal and isobaric conditions, the system variables are decomposed into a steady-state component and a small perturbation around the operating point. On account of the small size of the imposed perturbation, the time dependence is eliminated, and the original equations are converted to a new linear, time independent, complex-valued system, which is very convenient from a numerical viewpoint. Geometrical and physical modifications of the model simulate the aforementioned degradation modes, causing variations in the impedance. The possibility to detect unique impedance signatures is discussed, along with a study of the impact of input parameter inaccuracies and parameter interaction on the presented results. Finally, a study of pairs of concurrent degradation modes reveals the method’s strengths and limitations in terms of its diagnosis capabilities.

Research was performed on the development of relationships between the microstructure of nickel and yttria-stabilized zirconia (YSZ) coatings and the processing parameters used for their deposition by atmospheric plasma spraying (APS). Research was also performed on the development of relationships between the microstructure of plasma sprayed Ni-YSZ coatings and the electrochemical performance of metal-supported solid oxide fuel cells (SOFCs) incorporating these coatings as anodes. Three APS processes were used to deposit Ni-YSZ coatings: dry-powder plasma spraying (DPPS), suspension plasma spraying (SPS), and solution precursor plasma spraying (SPPS). These processes differ in the form of the feedstock injected into the plasma. The composition of the Ni-YSZ coatings deposited with each spray process could be controlled through adjustment of the plasma gas composition and stand-off distance, as well as adjustment of feedstock properties including agglomerate size fraction for DPPS, NiO particle size and suspension feed rate in SPS, and the enthalpy of decomposition of the precursors used in SPPS. The porosity of the Ni-YSZ coatings could be controlled through the addition of a sacrificial pore forming material to each feedstock, with coating porosities up to approximately 35% being achieved for each coating type. Metal-supported SOFCs were fabricated to each have anodes deposited with a different plasma spray process, where all anodes had nominally identical composition. The microstructures obtained for each anode type were distinctly different. SPPS led to the most uniform mixing of the smallest Ni and YSZ particles. These anodes most resembled typical structures from anodes fabricated using conventional methods. It was found that the polarization resistance, Rp, associated with the high frequency (> 1 kHz) range of the impedance spectrum correlated to the three phase boundary length (TPBL) density of each anode, with lower Rp values corresponding to higher TPBL densities. It was also found that the Knudsen diffusion coefficient and effective ordinary diffusion coefficient of the porous anodes correlated with the Rp associated with the low frequency (< 1 kHz) range of the impedance spectrum. Therefore, the impedance spectrum can be used to compare microstructural differences among plasma sprayed Ni-YSZ anodes.

This study investigated the overpotential performance enhancement of cathodes in low temperature solid oxide fuel cells (LT-SOFCs) due to the addition of carbonates to traditional Ce0.9Gd0.1O2 solid oxide fuel cell (SOFC) electrolytes. It was postulated in this study that this enhancement was due to the protonic conductivity of the carbonates. This provided an electrolyte with a dual conduction mechanism which improves the catalytic performance of the cathode. The cathode systems investigated were characterised for overpotential loss, conductivity and thermal expansion matching with the electrolyte. This produced results which predicted power outputs for a standard SOFC configuration as high as 970, 524 and 357 mW/cm2 at operational temperatures of 650oC, 600oC and 550oC. The benefits of these high power outputs and their potential to further reduce SOFC operational temperature was discussed. This study developed a cost-effective, reliable and commercially scalable manufacturing process for carbonate/Ce0.9Gd0.1O2 electrolytes. This pressureless sintering method is the first reported in literature, and is a promising replacement for the current hot-pressing technique currently used for these electrolytes. The electrolyte composition examined was 70 wt% Ce0.9Gd0.1O2 with 30 wt% carbonates (67 mol% Li2CO3 / 33 mol% Na2CO3). The cathode examined in this study was a composite cathode consisting of 50-90 wt% functional cathode material (Gd1-xSrxCoO3 with 10 to 30 mol% Sr doping on the Gd site) with a balance of electrolyte. It was determined that the composite cathode system with 10 wt% electrolyte and 20-30 mol% Sr doping was the optimal composition when operating at 600oC and above, with predicted power densities of 524 and 510 mW/cm2 at 600oC. At operational temperatures between 550oC and 600oC (and potentially lower), it was determined that a composite cathode system with 30 wt% electrolyte and 10-30 mol% Sr doping was the optimal composition. It was found that the presence of carbonates in the electrolyte decreased the overpotential losses of the cathode by 50-70% at 600oC for system studied; indicating that an improvement in cathodic performance coupled with the high conductivities of the electrolyte is most likely responsible for the high power outputs seen in literature. / Thesis (Ph.D, Mechanical and Materials Engineering) -- Queen's University, 2009-04-25 15:53:37.928

This work focused on the study of three independent projects involving perovskite oxide materials and their applications as solid oxide fuel cell (SOFC) anodes and electrochemical oxygen sensors. The underlying theme is the versatility and tune-ability of the perovskite structure. Reactivity and conductivity (ionic as well as electronic) are modified to optimize performance in a specific application. The effect of Ce doping on the structure and the conductivity of BaFeO3 perovskite materials is investigated and the resulting materials are applied as oxygen sensors. The new perovskite family, Ba1-xCexFeO3-δ (x=0, 0.01, 0.03, and 0.05), was prepared via a sol-gel method. Powder XRD indicates a hexagonal structure for BaFeO3 with a change to a cubic perovskite upon Cerium doping at the A site. The solubility limit of Ce at the A site was experimentally determined to be between 5-7 mol %. Bulk, electronic and ionic conductivities of BaFeO3-δ and Ba0.95Ce0.05FeO3-δ were measured in air at temperatures up to 1000˚C. Cerium doping increases the conductivity throughout the entire temperature range with a more pronounced effect at higher temperatures. At 800˚C the conductivity of Ba0.95Ce.05FeO3-δ reaches 3.3 S/cm. Pellets of Ba0.95Ce.05FeO3-δ were tested as gas sensors at 500 and 700˚C and show a linear, reproducible response to O2. Promising perovskite anodes have been tested in high sulfur fuel feeds. A series of perovskite solid oxide fuel cell (SOFC) anode materials: Sm0.95Ce0.05FeO3-δ, Sm0.95Ce0.05Fe0.97Ni0.03O3-δ and Sm0.95Ce0.05Fe0.97Co0.03O3-δ have been tested for sulfur tolerance at 500°C. The introduction of the extreme 5% H2S enhances the performance of these anodes, verified by EIS and CA experiments. Post mortem analyses indicate that the performance XII enhancement arises from the partial sulfidation of the anode, leading to the formation of FeS2, Sm3S4 and S on the perovskite surface. Testing in lower concentrations of sulfur, more common in sour fuels, 0.5% H2S, also enhances the performance of these materials. The SCF-Co anode shows promising stability and an increase in exchange current density, io, from 13.72 to 127.02 mA/cm2 when switching from H2 to 0.5% H2S/99.5% H2 fuel composition. Recovery tests performed on the SCF-Co anode conclude that the open cell voltage (OCV) and power density of these cells recover within 4 hours of H2S removal. We conclude that the formation of metal sulfide species is only partially reversible, yielding an anode material with an overall lower Rct upon switching back to pure H2. Combining their performance in sulfur containing fuels with their previously reported coke tolerance makes these perovskites especially attractive as low temperature SOFC anodes in sour fuels. A new perovskite family Ba1-xYxMoO3 (x=0-0.05) has been investigated in regards to electrical conductivity and performance as IT-SOFC anode materials for the oxidation of H2. Refinement of p-XRD spectra as well as SEM imaging conclude that the solubility limit of Y doping at the A site is 5 mol%, beyond which Y2O3 segregation occurs. The undoped BaMoO3 sample has a colossal room temperature conductivity of 2500 S/cm in dry H2. All materials maintain metallic conductivity in the temperature range of 25-1000°C with resistance increasing with Y doping. The Ba1-xYxMoO3 (x=0, 0.05) materials exhibit good performance as SOFC anode materials between 500-800°C, with Rct values at 500°C in dry H2 of 3.15 and 6.33 ohm*cm2 respectively. The catalytic performance of these perovskite anodes is directly related to electronic conductivity, as concluded from composite anode performance.

In this thesis, fuel cell systems are analysed thermodynamically and electrochemically. Thermodynamic relations are applied in order to determine the change of first law and second law efficiencies of the cells, and using the electrochemical relations, the irreversibilities occuring inside the cell are investigated. Following this general analysis, two simple solid oxide fuel cell systems are examined. The first system consists of a solid oxide unit cell with external reformer. The second law efficiency calculations for the unit cell are carried out at 1273 K and 1073 K, 1 atm and 5 atm, and by assuming different conversion ratios for methane, hydrogen, and oxygen in order to investigate the effects of temperature, pressure and conversion ratios on the second law efficiency. The irreversibilities inside the cell are also calculated and graphed in order to examine their effects on the actual cell voltage and power density of the cell. Following the analysis of a solid oxide unit cell, a simple fuel cell system is modeled. Exergy balance is applied at every node and component of the system. First law and second law efficiencies, and exergy loss of the system are calculated.

Solid oxide fuel cells with graded electrodes consisting of multiple composite layers yield generally lower polarization resistances than single layer composite electrodes. Optimization of the performance of solid oxide fuel cells with graded electrode composition and/or microstructure requires an evaluation of both the three phase boundary length per unit volume and the effective diffusion coefficient in order to provide insight into how these properties vary over the design space. A numerical methodology for studying the three phase boundary length and effective diffusivity in composite electrode layers with controlled properties is developed. A three dimensional solid model of a sintered composite electrode is generated for which the mean particle diameter, composition, and total porosity may be specified as independent variables. The total three phase boundary length for the modeled electrode is calculated and tomographic methods are used to estimate the fraction of this length over which the electrochemical reactions can theoretically occur. Furthermore, the open porosity of the modeled electrode is identified and the effective diffusion coefficient is extracted from the solution of the concentration of the diffusing species within the open porosity. Selected example electrode models are used to illustrate the application of the methods developed, and the resulting connected three phase boundary length and diffusion coefficients are compared. A significant result is the need for thickness-specific effective diffusivity to be determined, rather than the general volume averaged property, for electrodes with porosity between the upper and lower percolation thresholds. As the demand for current increases, more of the connected three phase boundaries become active, and therefore a greater fraction of the electrode layer is utilized for a given geometry, resulting in a higher apparent effective diffusivity compared to the same electrode geometry operating at a lower current. The methods developed in this work may be used within a macroscopic electrode performance model to investigate optimal designs for solid oxide fuel cell electrodes with stepwise graded composition and/or microstructure. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

A numerical model of the steady state and alternating current behaviour of a solid-oxide fuel cell is presented to explore the possibilities to diagnose and identify degradation mechanisms in a minimally invasive way using impedance spectroscopy. This is the first report of an SOFC impedance model to incorporate degradation, as well as the first one to include the ribbed interconnect geometry, using a 2-D approximation. Simulated degradation modes include: electrode/electrolyte delamination, interconnect oxidation, interconnect/electrode interface detachment, and anode sulfur poisoning. Detailed electrode-level simulation replaces the traditional equivalent circuit approach, allowing the simulation of degradation mechanisms that alter the shape of the current path. The SOFC impedance results from calculating the cell response to a small oscillatory perturbation in potential. Starting from the general equations for mass and charge transport, and assuming isothermal and isobaric conditions, the system variables are decomposed into a steady-state component and a small perturbation around the operating point. On account of the small size of the imposed perturbation, the time dependence is eliminated, and the original equations are converted to a new linear, time independent, complex-valued system, which is very convenient from a numerical viewpoint. Geometrical and physical modifications of the model simulate the aforementioned degradation modes, causing variations in the impedance. The possibility to detect unique impedance signatures is discussed, along with a study of the impact of input parameter inaccuracies and parameter interaction on the presented results. Finally, a study of pairs of concurrent degradation modes reveals the method’s strengths and limitations in terms of its diagnosis capabilities. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

This work focused on the study of three independent projects involving perovskite oxide materials and their applications as solid oxide fuel cell (SOFC) anodes and electrochemical oxygen sensors. The underlying theme is the versatility and tune-ability of the perovskite structure. Reactivity and conductivity (ionic as well as electronic) are modified to optimize performance in a specific application. The effect of Ce doping on the structure and the conductivity of BaFeO3 perovskite materials is investigated and the resulting materials are applied as oxygen sensors. The new perovskite family, Ba1-xCexFeO3-δ (x=0, 0.01, 0.03, and 0.05), was prepared via a sol-gel method. Powder XRD indicates a hexagonal structure for BaFeO3 with a change to a cubic perovskite upon Cerium doping at the A site. The solubility limit of Ce at the A site was experimentally determined to be between 5-7 mol %. Bulk, electronic and ionic conductivities of BaFeO3-δ and Ba0.95Ce0.05FeO3-δ were measured in air at temperatures up to 1000˚C. Cerium doping increases the conductivity throughout the entire temperature range with a more pronounced effect at higher temperatures. At 800˚C the conductivity of Ba0.95Ce.05FeO3-δ reaches 3.3 S/cm. Pellets of Ba0.95Ce.05FeO3-δ were tested as gas sensors at 500 and 700˚C and show a linear, reproducible response to O2. Promising perovskite anodes have been tested in high sulfur fuel feeds. A series of perovskite solid oxide fuel cell (SOFC) anode materials: Sm0.95Ce0.05FeO3-δ, Sm0.95Ce0.05Fe0.97Ni0.03O3-δ and Sm0.95Ce0.05Fe0.97Co0.03O3-δ have been tested for sulfur tolerance at 500°C. The introduction of the extreme 5% H2S enhances the performance of these anodes, verified by EIS and CA experiments. Post mortem analyses indicate that the performance XII enhancement arises from the partial sulfidation of the anode, leading to the formation of FeS2, Sm3S4 and S on the perovskite surface. Testing in lower concentrations of sulfur, more common in sour fuels, 0.5% H2S, also enhances the performance of these materials. The SCF-Co anode shows promising stability and an increase in exchange current density, io, from 13.72 to 127.02 mA/cm2 when switching from H2 to 0.5% H2S/99.5% H2 fuel composition. Recovery tests performed on the SCF-Co anode conclude that the open cell voltage (OCV) and power density of these cells recover within 4 hours of H2S removal. We conclude that the formation of metal sulfide species is only partially reversible, yielding an anode material with an overall lower Rct upon switching back to pure H2. Combining their performance in sulfur containing fuels with their previously reported coke tolerance makes these perovskites especially attractive as low temperature SOFC anodes in sour fuels. A new perovskite family Ba1-xYxMoO3 (x=0-0.05) has been investigated in regards to electrical conductivity and performance as IT-SOFC anode materials for the oxidation of H2. Refinement of p-XRD spectra as well as SEM imaging conclude that the solubility limit of Y doping at the A site is 5 mol%, beyond which Y2O3 segregation occurs. The undoped BaMoO3 sample has a colossal room temperature conductivity of 2500 S/cm in dry H2. All materials maintain metallic conductivity in the temperature range of 25-1000°C with resistance increasing with Y doping. The Ba1-xYxMoO3 (x=0, 0.05) materials exhibit good performance as SOFC anode materials between 500-800°C, with Rct values at 500°C in dry H2 of 3.15 and 6.33 ohm*cm2 respectively. The catalytic performance of these perovskite anodes is directly related to electronic conductivity, as concluded from composite anode performance.

This thesis attempts to test the effects of barium non stoichiometry and varying calcination temperatures on the microstructure and electrical conductivity of BaxZr0.7Pr0.1Y0.2O3- δ (x = 0.9, 1.0, 1.1). BZPY powders were fabricated using a combustion method, with the quantity of barium carefully controlled to create powders with a 10% molar excess or deficiency of barium. Then, portions of the precursor were calcined at 900 ºC, 1000 ºC, 1100 ºC, 1200 ºC and 1300 ºC for 5 h. The resulting calcined powders were pressed into pellets and sintered at 1600 ºC for 10 h, in a powder bath of the same chemical composition. In all, three chemically different powders were synthesized, and each composition was subjected to five different calcination temperatures, resulting in fifteen different samples to characterise. The precursor from the combustion method was characterised by using an STA to perform both TG and DSC simultaneously. The chemical composition of the precursor and calcined samples was analysed using ICP-OES. XRD was used to characterise the phases of both the powders and the sintered pellets. Lattice parameter indexing using Topaz and Scherrer's equation were used to extract the lattice parameters and crystallite sizes respectively. The microstructure of the pellets was examined using an SEM, the grain size measured using a linear intercept method and pore size using ImageJ. Finally, EIS was used to measure the conductivity of the pellets in dry and wet Argon atmospheres, with silver electrodes. Unfortunately, neither changes to barium stoichiometry nor partial calcination could improve the performance of BZPY. Partially calcined samples did not give rise to dense pellets, barium deficient samples showed inferior conductivity and barium excess samples, while showing higher conductivity than the barium deficient pellets at high temperature, were fragile and had to be handled carefully. Ultimately, the attempt to improve the performance of BZPY did not succeed and alternate methods of improving the grain growth need to be sought.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第19729号 / 工博第4184号 / 新制||工||1645(附属図書館) / 32765 / 京都大学大学院工学研究科物質エネルギー化学専攻 / (主査)教授 江口 浩一, 教授 安部 武志, 教授 陰山 洋 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM