Single-atom collaboration with cluster for accelerated nitrate electroreduction: Synergy revelation via machine learning and DFT calculations

Exploring high-performance electrocatalysts for the nitrate reduction reaction (NO₃RR) is crucial for environmental nitrate removal and ammonia synthesis. Single-atom collaboration with cluster can provide sufficient active sites for catalysts to promote NO₃RR, yet the unclear synergistic effect between the two hinders their rational design. Herein, a series of Ir₃ clusters and metal single atoms co-embedded in graphitic carbon nitride (g-CN) catalysts (Ir₃M₁) were constructed, and the synergistic effects of Ir₃ clusters and M₁ single atoms on the NO₃RR catalytic mechanism and activity were systematically explored using density functional theory (DFT) calculations combined with machine learning. Comprehensive evaluations of structural stability and catalytic activity demonstrate that the synergy between single atoms and clusters effectively balances the adsorption energies of key intermediates, yielding exceptional catalytic performance (the limiting potential of Ir₃Ti₁ can reach −0.22 V). Machine learning models further clarify the synergistic mechanism, where the geometric configurations of clusters serve as critical features for modulating the catalytic activity of single-atom sites, whereas the electronic structures of single atoms directly govern the reactivity of cluster sites. This DFT-machine learning approach provides theoretical guidelines for catalyst design and a predictive framework for efficient NO₃RR electrocatalysts.