Tracking Localized Degradation of Solid Oxide Cells Via Multi-Physics Modelling of Impedance Spectroscopy

Abstract

The global, macro-scale description of degradation phenomena in solid oxide cells (SOC) reaches limitations when dealing with localized, micro-scale processes. For example, it is established that carbon deposition and transformation of the active area in nickel containing electrodes in electrolysis mode depend strongly on the local overpotential and gas composition. The state-of-the-art long-term characterization of a full SOC however relies on macroscopic measurements of DC polarization and electrochemical impedance spectroscopy (EIS), which are capturing the average response of the entire cell. The deconvolution of impedance spectra allows to separate the electrolyte resistance from electrode electrochemistry, gas conversion, and diffusion contributions. Nonetheless, heterogeneous, and local degradation cannot be described deploying commonly used tools of EIS analysis, such as equivalent electric circuit models, transmission line models, or the analysis of distribution of relaxation times.Multi-physics based numerical models on the other hand open new opportunities in gaining advanced insights into the complex degradation mechanisms in SOCs. In the model proposed here, combining simulations of transient and frequency domain into one framework allows the description of measured long-term durability of an SOC in terms of voltage-current and impedance characteristic. In addition, the framework enables a local, micro-scale description of the degradation process based upon physical properties confirmable in experiments.The modelling framework is validated with the durability measurements of the SOC in Sun et al. [1]. In this study, a fuel electrode supported cell with an active area of 16 cm2 was tested in galvanostatic operating conditions as an electrolysis cell with the constant current of -1 A/cm2 over 4383 h. Simultaneously to the voltage measurements over the test duration, the impedance response of the cell was tracked every eight hours. The evolution of the measured impedance spectra of the cell can be seen in the attached figure.Comparing the outcomes of the model to the experimental data showcases the benefits of the model in describing the degradation induced increase in voltage over time and the evolution of impedance spectra. In particular, the effect of higher activation overpotential at the fuel inlet to the outlet of the cell is analyzed enabling the generation of local insights concerning the observed degradation. The model establishes an improved validation of equations describing degradation of SOCs encompassing overpotential gradients and relating to postmortem observed micro-structural changes. Reference [1] Sun, X., et al. "Degradation in solid oxide electrolysis cells during long term testing." Fuel Cells 19.6 (2019): 740-747. Figure 1