@article{Xu2025, 
author = {Lintao Xu and Yuhong Huang and Haiping Lin and Xiumei Wei and Fei Ma},
title = {Machine learning driven high-throughput screening of S and N-coordinated SACs for eNRR},
year = {2025},
journal = {Nano Research},
volume = {18},
number = {4},
pages = {94907289},
keywords = {machine learning, nitrogen reduction reaction (NRR) process, catalytic descriptors, SHapley Additive exPlanations (SHAP) analysis},
url = {https://www.sciopen.com/article/10.26599/NR.2025.94907289},
doi = {10.26599/NR.2025.94907289},
abstract = {This study constructs 196 transition metals (TM)@SxNy single-atom catalysts (SACs) (x = 0–4 and y = 0–4) and employs the eXtreme Gradient Boosting (XGBoost) classification model in machine learning (ML) for effectively distinguishing qualified and unqualified catalysts. The prediction accuracy rate is high, up to 95%. The SHapley Additive exPlanations (SHAP) analysis reveals that the N≡N bond length and the number of outermost d electrons (Nd) can well describe the nitrogen (N2) reduction reaction (NRR) activity. The relationships between N≡N, Nd, the adsorption energies of different intermediates ( ΔE∗N2,  ΔE∗N2H, and  ΔE∗NH2), the general descriptor (φ), and the Gibbs free energy of key steps ( ΔG∗N2,  ΔG∗N2−∗N2H, and  ΔG∗NH2−∗NH3) indicate that moderate nitrogen activation can enhance the reaction activity. Among the 17 screened SACs, Mo@S3N1, and W@S3N1 demonstrate the best catalytic performance, with limiting potential (UL) values of only −0.26 and −0.25 V under implicit solvation conditions. The electronic properties and variations in N≡N and TM–N bond lengths are investigated to reveal the origin of NRR activity. This study provides the decisive features and NRR dataset for ML research, as well as a feasible strategy for rational design of NRR SACs.}
}