Decoding and tuning the surface stability of perovskite oxides at the atomic level for faster oxygen exchange kinetics in energy conversion devices

点击翻译按钮获取中文摘要

Doped perovskite oxides serve as functional electrocatalyst layers in solid oxide fuel cells (SOFCs) because they can attain high electronic and oxide ion conductivity as well as good compatibility with common electrolyte materials. It has recently been identified that aliovalently doped perovskites, such as La1-xSrxMnO3 (LSM), form segregation layers capping the electrode surface under operating conditions, and this process is detrimental for electrode surface activity, thus degrading the material and SOFC performance over time. Reduction of strain energy due to cation size mismatch and electrostatic interaction with a space charge zone in the near-surface region have been proposed as the key mechanisms leading to cation segregation. The segregation is affected by external parameters, such as p(O2), T and applied potential. Because the interplay between the external conditions and material properties (i.e. cation size mismatch, and oxygen vacancy concentration) are not quantitatively understood on the atomistic level, it has not been possible to stabilize these surfaces using knowledge-based approaches. This project aims to quantitatively understand the relations of surface structure with environmental conditions, and material properties, and to predict conditions where stability against surface segregation and electrochemical activity are improved. The proposed computational framework unites the material properties and thermodynamic factors into a single Monte-Carlo model for predicting the evolution of the near-surface region. The expected outcome of the model is the ability to predict the distribution of dopant and host cations at the near-surface region of perovskite oxides, by accounting explicitly for the distribution of oxygen vacancies (giving rise to the proposed space charge zone). As a technologically important model perovskite electrocatalyst system, this research takes La0.8Sr0.2MnO3 as a starting point. The formation energies of oxygen vacancies and SrLa' defects close to the surface will be computed by density functional theory calculations and analyzed in terms of a cluster expansion in order to obtain a lattice Hamiltonian. Monte Carlo simulations of the near-surface region will yield the distribution of oxygen vacancies, as well as the cation segregation profile as a function of T and p(O2). Secondary phase formation will be studied by ab-initio thermodynamics considering formation and interface energies of likely candidates (e.g., SrO, Ruddlesden-Popper phases). Stabilization of the surface by modification with transition metal cations will be studied by considering the effect of surface substitution of Mn by Hf on the Sr segregation profile. This model will make it possible to understand accurately the factors leading to cation segregation and thus provide a first-principles basis for the optimization of surface properties on the atomic scale.