The team’s breakthrough centers around an innovative deep learning framework coined the Crossbreeding Neural Network (CBNN). Unlike conventional AI models trained within narrowly defined material boundaries, CBNN simultaneously assimilates data from two chemically distinct catalyst families: carbon-supported single-atom catalysts and perovskite oxide catalysts. This dual learning strategy enables CBNN to capture complementary insights—single-atom catalysts elucidate atomic-scale surface activity while perovskite oxides provide critical information on bulk crystal structure effects. The result is an unprecedented ability to predict properties of a hybrid catalyst class previously unexplored: single-atom catalysts anchored upon perovskite oxide supports.

Mechanistically, this hybrid system exploits the atomic precision of single metal atoms dispersed on catalytically active substrates, merged seamlessly with the structural and electronic versatility inherent in perovskite frameworks. Surface atomic arrangement characteristics are encoded in the model as image data, allowing the AI to interpret local atomic environments visually. Simultaneously, the bulk crystal structure of the oxide, represented by its graph information, informs the network about extended lattice periodicity and connectivity. By integrating these multidimensional data modalities, CBNN constructs a holistic representation of catalyst behavior anchored in both micro- and macro-scale structural features.

To enhance the model’s predictive rigor and interpretability, the research group implemented an automated descriptor-selection pipeline. This pipeline leverages classical statistical techniques coupled with advanced natural language processing (NLP) methods to identify physicochemical descriptors that robustly correlate with catalytic performance across both catalyst families. Key factors singled out include oxidation state, ionic radius, valence d-electron count, electronegativity, and coordination number. Such descriptors succinctly capture the electronic and geometric environment influencing OER activity and allow the AI to generalize learned relationships to novel material domains.

Experimental validation of the CBNN predictions involved synthesizing catalysts within the targeted hybrid family of single-atom-perovskite composites. Twenty-seven distinct catalysts spanning various elemental combinations were produced and rigorously tested under alkaline OER conditions. Impressively, the AI’s activity rankings for twelve of these synthesized materials precisely matched experimental measurements, confirming that the model did not simply interpolate within its training data but extrapolated genuine chemical insights to unknown material classes.

The research did not stop at single-element models; it further expanded into the complex realm of multimetallic catalysts. By computationally screening a vast candidate space of 8,008 permutations containing tungsten (W), molybdenum (Mo), ruthenium (Ru), and rhodium (Rh) single atoms embedded on a calcium–praseodymium cobalt iron oxide perovskite scaffold (Ca0.8Pr0.2Co0.8Fe0.2O3−δ, or CPCF), the AI identified optimal multimetallic compositions. These predictions were experimentally verified, with the top-performing multimetallic catalyst demonstrating superior oxygen evolution rates compared to all mono- and bi-metallic counterparts tested, as well as outperforming traditional perovskite oxides and carbon-supported single-atom catalysts.

Beyond raw catalytic activity predictions, the CBNN framework integrates explainable AI techniques to unpack the atomic-scale design principles responsible for enhanced performance. Visualization of feature importance within the model highlighted synergistic effects arising from specific neighboring metal atom configurations, revealing how electronic interactions at the atomic interface promote catalytic turnover. This level of interpretability is crucial for guiding rational catalyst design and advancing mechanistic understanding beyond black-box computational models.

Director HYEON Taeghwan eloquently summarizes the significance of this work: the AI system transcended conventional boundaries by leveraging heterogeneous datasets to explore uncharted catalyst territories rather than simply optimizing within known categories. This achievement opens the door to a paradigm wherein artificial intelligence holistically synthesizes fragmented experimental knowledge and catalyzes innovation at the interface of multiple material classes.

Looking ahead, the implications of this cross-material AI-driven discovery framework extend far beyond the oxygen evolution reaction or even catalysis. Similar integrative approaches could revolutionize materials development arenas such as battery electrode formulation, complex energy storage device design, and pharmaceutical drug discovery, where disparate heterogeneous datasets remain challenging to unify. By enabling AI to “speak” the common scientific language across domains, the potential for unveiling unanticipated material architectures and performance regimes dramatically expands.

In a landscape where the grand challenge is to design materials that reconcile often competing factors such as activity, stability, and cost, the fusion of AI with cross-class learning exemplified by CBNN offers a visionary roadmap. This study, published in the esteemed journal Nature Materials, marks a watershed moment in the journey toward generalized materials artificial intelligence capable of uncovering entirely new classes of functional materials through hybridized data-driven insight.

As the clean hydrogen economy accelerates, breakthroughs like this underscore the indispensable role of machine learning in pushing the frontiers of catalyst innovation beyond current scientific constraints. The ability to rationally engineer catalysts by bridging atomic to bulk scale phenomena unlocks unprecedented efficiency pathways essential for a sustainable energy future.

Subject of Research: Not applicable

Article Title: Cross-material catalyst discovery via deep learning

News Publication Date: 28-May-2026

Web References: 10.1038/s41563-026-02622-6

Image Credits: Institute for Basic Science

Keywords: Catalysis, Oxygen evolution reaction, Green hydrogen production, Single-atom catalysts, Perovskite oxides, Deep learning, Artificial intelligence, Multimetallic catalysts, Water electrolysis, Catalyst design, Machine learning, Nanomaterials

Among the various candidates for SOEC anode materials, perovskite oxides have garnered significant attention. These materials possess a unique combination of mixed ionic and electronic conductivity, enabling effective charge transport, and their electronic structures can be finely tuned through compositional modifications. The tunability of the perovskite structure translates into a rich platform for exploring how electronic configurations impact catalytic behavior. In alkaline solutions, prior studies have elucidated a volcano-shaped relationship between the occupancy of the 3d electron in the e_g orbital of transition metals within perovskites and the intrinsic OER activity. This correlation suggests an optimal electronic state where the oxygen evolution reaction can proceed most efficiently. However, translating these findings to the extreme environments of high-temperature SOEC operation has remained an unresolved challenge. The direct connection between e_g electron occupancy and OER activity under such thermally demanding conditions has yet to be fully established.

A breakthrough was recently reported by a collaboration between researchers led by Associate Professor SONG Yuefeng at the Dalian Institute of Chemical Physics (DICP) and Professor WANG Guoxiong at Fudan University. Their study centered on a novel series of alkaline-earth-metal-doped perovskites, specifically Pr_0.5Ae_0.5FeO_3−δ (where Ae represents calcium, strontium, and barium—denoted as PCF, PSF, and PBF respectively). By systematically varying the size of the dopant cation, the team sought to unravel how subtle shifts in electronic structure influenced the OER performance at elevated temperatures relevant to SOEC applications. This innovative approach allowed them to engineer the material’s electronic environment with unparalleled precision.

The experimental findings were striking: an increase in the ionic radius of the dopant corresponded to a pronounced enhancement in OER catalytic activity. Among the variants tested, the barium-doped PBF material demonstrated remarkable performance, achieving a current density of 3.33 A cm^-2 at an applied potential of 2.0 V and a temperature of 800 °C. This record signifies a substantial advancement in high-temperature oxygen evolution catalysis, marking PBF as a promising candidate for next-generation SOEC anodes. The superior activity is directly attributed to electronic and structural modifications induced by the alkaline-earth doping strategy.

Delving deeper into the mechanistic origins of this performance gain, the researchers employed an array of advanced analytical techniques. They revealed that doping with larger alkaline-earth cations enhanced the hybridization between Fe 3d and O 2p orbitals. This increased orbital overlap effectively lowered the charge-transfer energy, a critical parameter determining the ease of electron flow during the OER cycle. In addition, the presence of larger cations facilitated the migration of oxygen ions within the lattice and supported surface oxygen spillover processes. These dynamic oxygen behaviors are integral to accelerating the multi-step oxygen evolution reaction, thereby boosting overall catalytic rates.

The research team’s magnetic measurements unveiled another pivotal aspect of the doping effect. Ba doping precipitated a spin-state transition in the iron ions from a high-spin Fe^3+ configuration (t_2g^3 e_g^2) to a low-spin Fe^4+ state (t_2g^4 e_g^0). This transformation diminished the occupancy of the e_g orbital, a factor previously correlated with OER activity at room temperature but whose role in high-temperature contexts was ambiguous until now. The iron ion’s low-spin state streamlined oxygen movement and reaction kinetics, underscoring the importance of spin-state tuning as a novel lever for enhancing catalytic functionality in harsh environments.

These insights collectively establish that electronic structure engineering, particularly via controlled spin-state manipulation, holds immense potential for optimizing SOEC anode materials. The findings highlight that beyond mere electron count or doping concentration, the spin configuration of transition metal centers critically modulates catalytic behavior. Such knowledge paves the way for rational design strategies that transcend trial-and-error approaches, enabling the creation of bespoke perovskite catalysts tailored for high-performance oxygen evolution at elevated temperatures.

The practical implications of this study extend beyond the laboratory. SOECs equipped with such finely tuned perovskite anodes could catalyze a paradigm shift in renewable energy storage, facilitating the large-scale production of synthetic fuels like syngas and hydrogen. These fuels are pivotal for decarbonizing sectors that are challenging to electrify directly. By enhancing the durability and efficiency of oxygen evolution catalysts, researchers are addressing a key obstacle that has long limited the commercial viability of SOEC technology.

Moreover, the approach undertaken by SONG, WANG, and colleagues opens broader avenues for exploring the fundamental interplay between spin states, electronic structure, and catalytic function in complex oxides. The ability to manipulate spin states through chemical doping offers a powerful tool for tuning activity in other crucial energy conversion reactions, such as oxygen reduction, hydrogen evolution, and CO_2 reduction. The insights gleaned here might thus reverberate across electrocatalysis and materials science disciplines.

The research was published in the highly regarded Journal of the American Chemical Society on August 26, 2025, underscoring its significance within the scientific community. The work represents a culmination of meticulous experimentation, insightful theoretical interpretation, and collaborative scientific effort, typifying the interdisciplinary nature of cutting-edge energy research.

Ultimately, this advancement exemplifies how nuanced control of atomic and electronic structures within perovskite oxides can surmount long-standing catalytic challenges. It reinforces the promise of SOECs as keystones in a sustainable energy future and exemplifies the power of fundamental science to unlock transformative technologies. As the global demand for clean energy accelerates, breakthroughs such as these will be instrumental in redefining how we generate, store, and utilize energy on a planetary scale.

Subject of Research: Not applicable

Article Title: Spin-State Tuning in PrFeO3-δ Perovskite for High-Temperature Oxygen Evolution Reaction

News Publication Date: 26-Aug-2025

Web References: https://pubs.acs.org/doi/10.1021/jacs.5c10937

References: 10.1021/jacs.5c10937

Keywords

Electrolysis