A statistical understanding of oxygen vacancies in distorted high-entropy perovskite oxides
Adam Potter 1 , Yifan Wang 1,2 , Kiran Hamkins 1 , Dongjae Kong 1 , Yuzhe Li 2 , Jian Qin 3 , Xiaolin Zheng 1,4,* · 2025
Read A statistical understanding of oxygen vacancies in distorted high-entropy perovskite oxides by Adam Potter 1 , Yifan Wang 1,2 , Kiran Hamkins 1 , Dongjae Kong 1 , Yuzhe Li 2 , Jian Qin 3 , Xiaolin Zheng 1,4,* (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-11587.txt.
Introduction
Potter and six colleagues measure oxygen gaps in mixed metal oxides with a statistical eye in twenty twenty-five. These high entropy perovskites are eyed as electrodes for splitting water with heat. Missing oxygen atoms steer their traits, yet mix complexity hid the rules. Fourteen oxides with varied A site mixes were tested.
Gap counts hinged on cation size spread as well as charge mix. Simulations showed size mismatch widens the spread of gap forming costs. That spread shifts bulk behavior in a way plain models miss. Bigger size spread dulled heat sensitivity of gap counts. The rule gives makers a tuning knob, they say.
Catalog record
| CREATOR | Adam Potter 1 , Yifan Wang 1,2 , Kiran Hamkins 1 , Dongjae Kong 1 , Yuzhe Li 2 , Jian Qin 3 , Xiaolin Zheng 1,4,* |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-11587
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A statistical understanding of oxygen vacancies in distorted high-entropy perovskite oxides Adam Potter 1 , Yifan Wang 1,2 , Kiran Hamkins 1 , Dongjae Kong 1 , Yuzhe Li 2 , Jian Qin 3 , Xiaolin Zheng 1,4,* Abstract High-entropy perovskite oxides have emerged as promising electrode materials for solid oxide electrolyzers. However, their compositional complexity makes the formation of oxygen vacancies, which influence properties such as oxygen ionic conductivity and thermal expansion, challenging to predict. Here, we experimentally measure changes in oxygen vacancy concentration for fourteen perovskite oxides with high and low-entropy A-site compositions, finding a dependence on cation size variance in addition to divalent cation fraction. Atomistic simulations using a machine-learned universal interatomic potential reveal cation size mismatches broaden a distribution of vacancy formation energies, shown through statistical thermodynamics to shift bulk formation thermodynamics. Treating oxygen vacancies statistically enables accurate predictions of oxygen vacancy formation compared to traditional models. Practically, increasing the size variance between A-site cations reduces the temperature sensitivity of oxygen vacancy concentrations, making it key for tuning critical properties. More broadly, this study demonstrates statistical treatment of oxygen vacancies is essential for understanding high-entropy perovskite oxides. I Introduction Perovskite oxides (ABO 3-δ ), such as LSCF ( \rm La_{0.6}Sr_{0.4}Co_{0.2}Fe_{0.8}O_{3-\delta} ), are widely utilized as anodes for high-temperature solid oxide electrolyzer cells (SOECs), which offer the potential for efficient hydrogen production using renewable energy and heat. However, the performance of SOECs is often limited by anode properties, including low oxygen ion conductivity and high thermal expansion [ 1 ] . The oxygen ion conductivity of perovskites relies on a high concentration of oxygen vacancies ( \delta ) to enable efficient ion conduction via vacancy hopping [ 2 ] . The concentration of oxygen vacancies also affects other important properties, such as the thermal expansion coefficient, thermochemical stability, and electrical conductivity [ 3 , 4 ] . As a result, understanding and careful control of oxygen vacancy behavior is critical for optimizing the performance of perovskite oxides in high-temperature SOECs. Recently, high entropy perovskite oxides (HEPOs) (ABO 3-δ ), with five or more cations with nearly equal molar amounts in either A- or B-site [ 5 ] , have attracted great attention for SOECs as they theoretically could have good high-temperature chemical and structural stability due to the high mixing entropy [ 2 ] . In addition, high-entropy perovskites were shown to introduce bond distortion [ 6 ] and local ordering [ 7 ] , which could potentially influence the formation of oxygen vacancies. It has been demonstrated that substituting metal cations can change the oxygen vacancy concentration [ 8 , 9 , 3 , 9 ] . However, exactly how the oxygen vacancy concentration in HEPOs varies with metal cation composition and temperature remains elusive as the net energy needed to remove a lattice oxygen is significantly complicated by variations from different metal cations and configurational entropy. Previous research on simple perovskites has provided valuable methodologies and insights for studying oxygen vacancy. For example, experimental studies vary concentrations of one or two metal cations systematically and measure oxygen vacancies using techniques, such as thermo-gravimetry analysis (TGA) [ 10 , 11 , 4 , 12 , 13 , 14 ] , Columbic titration [ 15 , 16 , 17 , 9 ] , X-ray photoelectron spectroscopy (XPS) [ 8 ] , or iodometric titration [ 17 ] . Then, the vacancies’ formation thermodynamics can be determined by applying concentration data to established defect models [ 3 , 16 ] . Those studies consistently reveal the importance of the concentration of 2+ cations in the A-site ( X_{2+} ) for forming high oxygen vacancy concentrations [ 12 , 16 ] . However, experimentally sweeping every metal cation concentration is impractical for high-entropy materials with five or more dopants, and no such experimental studies have been reported in the literature. In comparison, there are a few computational studies that aim to predict high-entropy doping effects on the oxygen vacancy concentration by calculating the vacancy formation energy ( E_{\rm v} ). Density functional theory (DFT) studies calculated E_{\rm v} for simple [ 11 ] and high-entropy oxides [ 13 , 7 , 8 ] and found cation substitutions could tune E_{\rm v} as well as where vacancies preferred to form. An empirical study found such DFT-calculated E_{\rm v} values could be predicted from the oxide’s formation enthalpy, electronegativity, and band gap [ 18 ] . However, while it is possible to predict E_{\rm v} for many high-entropy oxide compositions computationally, these E_{\rm v} predictions have yet to consistently align with experimentally measured oxygen vacancy concentrations. A recent study by Park et al. [ 13 ] calculated E_{\rm v} with DFT across many oxygen sites for one HEPO and three simpler oxides and found all four had significant deviations in E_{\rm v} depending on their cation neighbors. To address this variability, Park proposed E_{\rm v} values exist in a distribution, noting the potential implications of the mean and variance on equilibrium vacancy concentration. Xu et al. [ 7 ] also used a statistical approach in their DFT study of complex perovskites to predict oxygen vacancy formation, finding that vacancies preferentially form near cobalt sites. Both studies suggest the statistics of E_{\rm v} may be key to predicting and tuning oxygen vacancies, however, the importance of such statistical effects for predictive models is unclear. The mechanisms of E_{\rm v} statistics has also remained largely unexplored for a wide range of high-entropy oxides compositions, both experimentally and computationally. This gap needs to be filled to provide principles or guiding models for the selection of cation dopants to tune oxygen vacancies, and the critical properties dependent on them, in complex perovskite oxides. In this work, we establish a combined experimental, computational, and theoretical workflow to study the effect of high-entropy mixed A-sites in perovskite oxides on the concentration of oxygen vacancies. Experimentally, we measured the oxygen vacancy concentration of 11 high-entropy (HE) and 3 low-entropy (LE) perovskites as a function of temperature using TGA, and found an enthalpy-entropy compensation relation where HE samples exhibit both lower enthalpy and entropy of oxygen vacancy formation. We found that the oxygen vacancy concentrations in LSCF-based perovskite oxides are directly affected by two primary variables: the conventional 2+ cation molar ratio in the A-site ( X_{2+} ), and the newly identified A-site ionic radius variance ( \sigma_{A} ). Computationally, our high-throughput simulations with a machine-learned interatomic potential show that increased \sigma_{A} distorts oxygen–B-site bonds, increasing the variance in oxygen bonding energy ( \sigma_{\rm Ev} ). Our statistical thermodynamic analysis shows that this broadening in \sigma_{\rm Ev} reduces the vacancy enthalpy ( \Delta H^{f} ) and entropy ( \Delta S^{f} ) of formation, matching experimental trends. We further propose a statistical model of vacancy formation that accurately predicts the vacancy concentrations of LE and HE materials from simulation. Overall, our results highlight A-site size mismatches in the A-site ( \sigma_{A} ) as a practical lever for tuning oxygen vacancies and their temperature dependence in LSCF-based perovskite oxides. Theoretically, tuning \sigma_{\rm Ev} via structural distortion or other means could have similar effects in other oxide systems, high-entropy or not. Label A-Site Composition X_{2+} t \sigma_{A} (%) \Delta S_{mix} (J mol