Evaluation of Oxygen-Ionic Transport in Mixed-Conducting Electrode Materials by Electrical Conductivity Relaxation Technique

Abstract

Oxide materials with mixed ionic and electronic conductivity (MIECs) are important components of high-temperature electrochemical systems, including electrodes of solid oxide fuel and electrolysis cells (SOFC/SOEC) and dense ceramic membranes for partial oxidation of hydrocarbons and oxygen separation. Characterization of ionic transport in MIECs faces two challenges: the deconvolution of the ionic and electronic transport, and the deconvolution of bulk diffusion and kinetics of the surface electrochemical reaction. The electrical conductivity relaxation (ECR) technique [1-3] is based on assumptions that (i) the total conductivity of the MIEC has a monotonic dependence on oxygen partial pressure in a given p(O2) range with sufficient sensitivity; (ii) the nature of the ionic transport and surface exchange processes allows reasonable linearization in their numerical description; (iii) there are no unaccounted by-processes and phase transitions. In practice, the conductivity relaxation measurements are organized as a continuous recording of the changes in the sample resistivity after abrupt changes in oxygen partial pressure (or oxygen chemical potential). The collected ECR data can be fitted using numerical models that originate from the solutions described in [4]. In general, these models require well-defined and simple geometry of the sample, negligible porosity, and a known state of the surface (polished, rough, activated…). Of key importance in this technique is the precise fixation of the moment of change of p(O2) in the gas phase and the initial resistivity (conductivity) of the sample. In the ECR setup designed in the present study, oxygen partial pressure in the heated sample chamber is changed by a fast short-term connection to a gas container with different oxygen pressure. This produces a very abrupt change in the p(O2), and the time of this event is well-defined due to the high operation speed of vacuum magnet valves and also is confirmed by the response of the electrolytic oxygen sensor. All data acquisition is performed in automatic mode, enabling long-term data collection for subsequent non-linear least square analysis. The ECR technique was employed for the characterization of a series of mixed-conducting ceramics typically used as oxygen electrode materials for SOFC/SOEC, including La0.3Sr0.7CoO3-δ (LSC) and (La,Sr)(Co,Fe)O3-δ (LSCF) – based perovskite-type oxides. The ECR measurements were performed in the p(O2) range of 0.010-0.210 atm with steps of ~ 0.03-0.0.05 atm. The properties of each particular material determined the corresponding temperature range. In the case of LSC, the highest temperature was limited to 850°C by the time resolution of the data acquisition system, and the low-T limit corresponded to 450°C. For LSCF-based compositions, the studied temperature range was 650-875°C. The chemical diffusion and surface exchange coefficients obtained by the ECR technique were analyzed in combination with the data on oxygen nonstoichiometry δ and ionic transport parameters measured by the oxygen permeability technique.