Detailed Mechanistic Studies of Electrochemical Reactions on Pt and Au Electrodes in Solid Oxide Cells Via EIS Data Analysis

Abstract

In order to reduce the environmental impact of fossil fuels, a range of sustainable technologies for power generation and storage are currently being developed. In addition, efficient and low-cost methods for CO2 capture and conversion to form useful products are critically needed. One of the promising technologies towards achieving these goals involves the use of solid oxide cells (SOC), which are unique devices that can be employed in both the fuel cell mode for clean power generation and in the electrolysis mode to achieve CO2 conversion through its electrochemical reduction (‘CO2RR’).One of the burning problems that is impeding SOC implementation is catalyst selection, with many promising materials having been reported over the last 50 years, including metals and mixed ionic and electronic conducting (MIEC) metal oxide materials. However, the reasons for better or worse activity, as well as the mechanism of the CO2RR at these catalysts remains unknown and it is often unclear what factors limit the catalyst activity during cell operation and how this can be improved. This is due, in part, to the limited number of techniques that can be used to evaluate these performance metrics, especially at high operating temperatures. One of the most useful techniques is electrochemical impedance spectroscopy (EIS). However, data interpretation is often quite complex and thus other methods, such as CNLS fitting or the determination of the distribution of relaxation times (DRT), are needed.While CNLS fitting allows the fit parameters to be correlated with specific physical properties, a precise model of the system is required. Since SOC systems are relatively unexplored and good models are not yet available, this could lead to the mis-interpretation of the EIS data. DRT, on the other hand, can be a useful tool since it does not require any pre-existing models of the system. Unlike CNLS fitting, DRT can provide the correct number of time constants present (from the number of peaks obtained), as well as the associated resistance and capacitance values. According to the literature, each peak is typically related to a specific reaction step with a unique capacitance value, identified by changing conditions such as temperature, polarization or gas composition.In the present work, DRT analysis of Pt and Au electrodes on YSZ electrolytes was carried out in order to determine the origin of each resistance and to determine the effect of temperature, electrode polarization and gas composition on both the resistance and capacitance values. Here, both porous and point metal electrodes were investigated as no double phase boundary activity or chemical capacitance effects should be present, making data interpretation less complex. Furthermore, to avoid the challenges of porous electrodes, point electrodes, produced by pressing the metal wire to a flat YSZ surface by applying pressure with a spring, followed by softening at 1100 °C to achieve better contact, were used to achieve a controllable triple phase boundary (TPB) length. This allowed the true activity of the metal/YSZ TPB to be measured as a function of reactive interface length, rather than just per geometric surface area. Porous electrodes were also studied, made by depositing Pt or Au paste onto YSZ, followed by sintering at 600 °C for 20 min. Polymeric YSZ precursors were then infiltrated into the metallic backbone, then sintered at 750 °C for 2 hours. Then, a thin layer of the same metal paste was applied in order to provide better conductivity. The point electrodes were made by attaching 0.3 mm thick Pt and Au wires onto dense and flat YSZ electrolyte by a spring-loaded cell holder, then held at 1050 °C for 1 hour to soften and produce a fixed TPB length. Electrochemical testing of the porous electrodes was conducted at 750 °C in both half- full cell modes in air, CO2 and CO2/CO mixtures, while the wires were tested at 650-750°C in the same gases, but in half-cell mode.8 individual process were seen for both the porous and point electrodes, including both electrodes in each cell. The time constant with the smaller capacitors are likely related to the electrode-electrolyte interface, while the mid-range capacitors are associated with electrode surface processes, such as surface diffusion, molecular species dissociation, and adsorption. The high capacitance steps are likely related to gas-phase processes, such as gas diffusion and the possible presence of a conversion layer. More detailed analysis of polarization and gas atmosphere dependence is ongoing in order to confirm the nature of each process.