Defect Thermodynamic Modeling of Triple Conducting Perovskites (La,Ba)Fe1-xMxO3-δ for Proton-Conducting Solid-Oxide Cells

Abstract

Density functional theory (DFT) based modeling was performed to investigate the defect energetics of the proton conducting oxides BaFe1-xMxO3-δ (with M=Zr, Y, and Zn at the B-site). High-throughput defect modeling was carried out to obtain the defect energetics ensembles as a function of B-site x (0 – 0.25) and oxygen δ (0 – 0.5) compositions for assessment of oxygen vacancy formation (Evac) and hydration reaction (Ehyd) energies in complex defect-dopant configurations. The modeling effort aims to elucidate the complex coupling among oxygen vacancies, the protons and metal dopants in various concentrations and their combined effect upon defect energetics in BaFe1-xMxO3-δ perovskite. This information complements the current experimental measurements with essential data related to the energetics of various defect configurations that can play a role in solid-oxide-cell (SOC) applications. Preliminary results among more than 3,000 configurations investigated are summarized in Figure 1. All calculations were performed with the Perdew-Burke-Ernzerhof (PBE) functional in charge neutral 2 × 2 × 2 perovskite supercells. An effective Hubbard Ueff correction of 4 eV was applied to the Fe 3d shell, with all the Fe atoms being in a high-spin state and having a ferromagnetic ordering. Stability analysis of each of the BaFe1-xMxO3-δ systems investigated at specified x and δ composition values reveals that the oxygen vacancy (VO) is energetically unfavorable when placed adjacent to the M dopant. Overall, the Fe-VO-Fe configurations are identified to be more stable than the Fe-VO-M and M-VO-M configurations, while one with a M-VO-M cluster arrangement is the least stable. Such a short-range ordering leads to the blocking of the oxygen vacancy migration toward the oxygen sites around the M dopant, whereas the proton hopping between the oxygens around the M dopants is energetically not hindered. Based on the stability of various defect-dopant configurations of the BaFe0.75M0.25O2.875 systems with or without hydration, it is revealed that the stability range of analogous configurations with varying dopant types can be an indicator of the defect-dopant ordering strength and can be correlated with the ionic radii of the dopants in the BaFe1-xMxO3-δ systems.Taking the energy of the most stable BaFe1-xMxO3-δ structures at various x and δ values analyzed as representative for the respective compositions, the corresponding derivations of the Evac and Ehyd energies as a function of δ (and x) are summarized in Figures 1(a) and 1(b), respectively. Overall, the results indicate a stronger oxygen non-stoichiometry (δ) dependence in the Evac values of the BaFeO3-δ and BaFe1-xMxO3-δ systems (M=Zr, Y, and Zn at x=0.125 and x=0.25), whereas the Ehyd data exhibit a much weaker δ dependence. This contrasting behavior in the δ dependences between Evac and Ehyd values may be attributed to the charge transfer characteristics of the oxygen vacancy formation reaction, which is associated with the Fe 3d and the O 2p energy levels of the materials at various δ values while the charge transfer involved in the hydration reaction is relatively minimal. The effect of dopant concentration (x) further shows an opposite trend in the Evac values vs. x with the dopant types, with an increase of Evac upon increasing the Zr doping concentration and a decrease of Evac upon increasing the Y and Zn doping content. Due to the strong oxygen non-stoichiometry dependence in the Evac values, the results indicate an increased proton affinity of the BaFe1-xMxO3-δ oxides. Further assessments of the Evac and Ehyd data on δ values and their impact upon the defect equilibria of the BaFe1-xMxO3-δ systems under the SOC operating conditions will be discussed. Disclaimer This project was funded by the U.S. Department of Energy, National Energy Technology Laboratory, in part, through a site support contract. Neither the United States Government nor any agency thereof, nor any of their employees, nor the support contractor, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. Figure 1