Enhancing Biochar-Based Nonradical Persulfate Activation Using Data-Driven Techniques.

Area of Science:

• Environmental Chemistry
• Materials Science
• Catalysis

Background:

• Biochar (BC) shows promise as a catalyst for persulfate activation in water remediation.
• Understanding the structure-activity relationship in BC is crucial for optimizing its catalytic performance.
• Machine learning (ML) offers a powerful tool for accelerating material design and property prediction.

Purpose of the Study:

• To apply ML techniques to rationally design biochar for enhanced nonradical pathways in persulfate activation.
• To identify key biochar properties that promote nonradical degradation mechanisms.
• To demonstrate the utility of ML in tailoring biochar catalysts for water treatment.

Main Methods:

• Utilized ML models to analyze the relationship between BC properties and persulfate activation.
• Identified specific surface area and oxygen content as critical features for nonradical pathways.
• Optimized BC synthesis by tuning temperature and biomass precursors based on ML predictions.

Main Results:

• ML analysis revealed that high specific surface area and oxygen content significantly enhance nonradical contribution.
• Tuning synthesis parameters (temperature, precursors) effectively regulated these key features.
• Two novel biochar catalysts with distinct active sites, optimized for nonradical pathways, were successfully prepared.

Conclusions:

• ML is a valuable tool for the rational design and synthesis of tailored biochar catalysts.
• This study provides a proof of concept for accelerating bio-based catalyst development using ML.
• The developed ML-guided approach facilitates efficient persulfate activation for water remediation via nonradical pathways.