High temperature oxygen exchange reaction on dense and porous La0.6Sr0.4CoO3-δ electrodes: An overview of the experimental evidence for modeling

Oxygen exchange kinetics was investigated to model the current-potential relationship of mixed conducting oxide electrodes used in SOFC and SOEC. Focusing on La_0.6Sr_0.4CoO₃ as a model material, experimental evidence so far obtained in our group were summarized and reanalyzed. The reaction order analysis suggested a complex reaction mechanism, for which we came to think of two series kinetics, surface process and subsurface process. The former refers to an exchange process between gas-phase oxygen molecules and some sort of surface oxygen species. The latter refers to the exchange of surface oxygen with bulk oxide ions, and the reaction barrier is not necessarily oxygen transport, but may be electron transport/transfer for oxygen in/ex-corporation This hypothesis appeared to resolve some of our remaining questions regarding the experimental results, such as scattered p_O2 dependence in high partial pressure range, the higher isotope exchange rates than electrochemical impedance, and the reaction rate enhancement in the presence of the LaSrCoO₄ phase. While a single piece of such experimental evidence is insufficient to prove the hypothesis, considering all the results together provides strong support. We then tried to separate the contributions of surface and subsurface processes by measuring the surface oxygen potential using a porous oxygen sensor. It revealed that the surface process is written as $J_s=J_{s,0}\bullet \delta \bullet \left(a_{O,s}^2-p_{O_2,g}\right)$ and the subsurface process as $J_{\mathrm{ss}}=J_{\mathrm{ss},0}\bullet \left(a_{O,e}{a}_{O,s}^{-1}-a_{O,e}^{-1}{a}_{O,s}\right)=J_{\mathrm{ss},0}\left(\exp \left(\frac{\beta ∆{\mu }_{O_2}}{\mathrm{RT}}\right)-\exp \left(-\frac{\left(1-\beta \right)∆{\mu }_{O_2}}{\mathrm{RT}}\right)\right)$, which are in good agreement with the experimental data even for films with different oxygen vacancy formation energies. For modeling porous electrodes, we considered it unnecessary to include the subsurface process based on experimental evidences, such as the higher area-specific reaction rate on the particle surface in a porous electrode than on the film electrode, and the lack of enhancement effect by LaSrCoO₄ phase. The current-voltage relationship estimated by applying J_s to the transmission line model was in good agreement with the experimental results.