Unified electronic-geometric descriptor deciphers peroxymonosulfate activation using Fe-based dual-atom catalysts

Area of Science:

• Materials Science
• Environmental Chemistry
• Computational Chemistry

Background:

• Rational catalyst design for Fenton-like reactions is limited by understanding electronic-geometric synergy in peroxymonosulfate (PMS) activation.
• Classical d-band theory does not fully capture the complexities of PMS activation across diverse coordination environments.

Purpose of the Study:

• To develop a machine learning-decoded binary descriptor (BD) unifying orbital electronic structure (I_OES) and orbital geometric structure (I_OGS) indices.
• To predict and screen efficient Fe-based dual-atom catalysts (DACs) for PMS activation.

Main Methods:

• Proposed a novel binary descriptor (BD) integrating I_OES and I_OGS.
• Quantified antibonding orbital occupancy via d-p hybridization and geometric constraints.
• Screened various FeM DACs (M = Ti, V, Cr, Mn, Fe, Co, Ni, Cu) using the BD framework.

Main Results:

• Identified FeMn DACs with optimal BD values (I_OES=0.86, I_OGS=0.40) as highly efficient catalysts.
• Achieved 94.2% singlet oxygen (¹O₂) yield and fast kinetics (k_obs = 1.2 min⁻¹) for sulfadiazine degradation.
• Demonstrated >90% pollutant removal in a flow-through reactor over 30 days at industrial flux.

Conclusions:

• Established universal orbital-level design principles for sustainable water remediation.
• Bridged atomic-scale insights with engineering-scale implementation for practical applications.
• The BD framework offers a powerful tool for designing high-performance catalysts for environmental cleanup.