Machine Learning-Guided Design of Catalysts with SO ₂ Resistance for Low-Temperature NH ₃ –SCR Reaction

The design of efficient catalysts to mitigate SO₂ poisoning in low-temperature Selective Catalytic Reduction of NO_<i>x</i> (LT-SCR) is challenging. Herein, machine learning (ML) was employed to design catalysts with SO₂ resistance. A multidimensional data set containing 242 data points was constructed, including elemental descriptors, catalyst structures, reaction conditions, and SO₂ poisoning conditions. Regression models such as XGBoost (XGB) were trained to predict NO_<i>x</i> conversions and SO₂ resistance over different catalysts. It was found that the electronegativity descriptor (EN.) was the critical factor influencing NO_<i>x</i> conversions in the presence of SO₂, with approximately a monotonic positive correlation trend with NO_<i>x</i> conversions within the range of 0.7-0.8. Based on these findings, quaternary CeMoFe/Ti catalysts were further synthesized via an inverse design. It was observed that this catalyst could still maintain 60% NO_<i>x</i> conversion in the presence of SO₂ after 6 h at 250 °C, wherein the SO₂ resistance was significantly improved compared to ternary CeMo/Ti (∼40%) and binary Ce/Ti (∼4%) catalysts. In addition, the ML revealed the core roles of EN. in the design of SO₂ resistance catalysts, breaking through the limitations of traditional ternary catalyst systems. This study provided a data-driven paradigm for precise design of SO₂ resistance catalysts for the LT-SCR reaction, holding promise to accelerate the research and development of effective catalysts.