Ivan Surin

Transmission electron microscopy (TEM) plays a crucial role in heterogeneous catalysis for assessing the size distribution of supported metal nanoparticles. Typically, nanoparticle size is quantified by measuring the diameter under the assumption of spherical geometry, a simplification that limits the precision needed for advancing synthesis‐structure‐performance relationships. Currently, there is a lack of techniques that can reliably extract more meaningful information from atomically resolved TEM images, like nuclearity or geometry. Here, cycle‐consistent generative adversarial networks (CycleGANs) are explored to bridge experimental and simulated images, directly linking experimental observations with information from their underlying atomic structure. Using the versatile Pt/CeO₂ (Pt particles centered ≈2 nm) catalyst synthesized by impregnation, large datasets of experimental scanning transmission electron micrographs and physical image simulations are created to train a CycleGAN. A subsequent size‐estimation network is developed to determine the nuclearity of imaged nanoparticles, providing plausible estimates for ≈70% of experimentally observed particles. This automatic approach enables precise size determination of supported nanoparticle‐based catalysts overcoming crystal orientation limitations of conventional techniques, promising high accuracy with sufficient training data. Tools like this are envisioned to be of great use in designing and characterizing catalytic materials with improved atomic precision.

Nitrous oxide, N₂O, exhibits unique reactivity in oxidation catalysis, but the high manufacturing costs limit its prospective uses. Direct oxidation of ammonia, NH₃, to N₂O can ameliorate this issue but its implementation is thwarted by suboptimal catalyst selectivity and stability, and the lack of established structure–performance relationships. Systematic and controlled material nanostructuring offers an innovative approach for advancement in catalyst design. Herein low‐valent manganese atoms stabilized on ceria, CeO₂, are discovered as the first stable catalyst for NH₃ oxidation to N₂O, exhibiting two‐fold higher productivity than the state‐of‐the‐art. Detailed mechanistic, computational and kinetic studies reveal CeO₂ as the mediator of oxygen supply, while undercoordinated manganese species activate O₂ and facilitate N₂O evolution via NN bond formation between nitroxyl, HNO, intermediates. Synthesis via simple impregnation of a small metal quantity (1 wt%) predominantly generates isolated manganese sites, while full atomic dispersion is achieved upon redispersion of sporadic oxide nanoparticles during reaction, as confirmed by advanced microscopic analysis and electron paramagnetic resonance spectroscopy. Subsequently, manganese speciation is maintained, and no deactivation is observed over 70 h on stream. CeO₂‐supported isolated transition metals emerge as a novel class of materials for N₂O production, encouraging future studies to evaluate their potential in selective catalytic oxidations at large.