Highly Efficient Photo-Fenton Reaction at Full Ph Range Instructed by Machine Learning

Abstract

The study of photo-Fenton technology with a full pH range is significant and challenging. In this paper, Ferrocene nanoparticles modified Co-MOF were constructed for efficient photo-Fenton degradation of tetracycline (TC) at full pH range (0-14). The systematic mechanism study indicates that the nitrilotriacetic acid organic ligands in Co-MOF offers more ·OH that can achieve high activity under alkaline conditions, while the synergistic effect between Co-MOF and Ferrocene promotes the activation of potassium persulfate and improves the stability of the catalyst under extreme pH conditions. The π-π interaction between TC and Fc-Co-MOF can also improve electron injection capability and accelerates the production of ·OH/SO4•-, so as to accelerate the redox cycles of Fe(II)/Fe(III) and Co(II)/Co(III), bringing about prolonged carrier lifetime and better photocatalytic activity. The possible degradation pathway of TC was investigated by Fukui function and LC-MS. A machine learning model is used to optimize the synthesis and photocatalytic parameters of Fc-Co-MOF. This study provides a new design idea to prepare highly efficient and stable photo-Fenton advanced oxidation technology at full pH range.

Keywords: Co-MOF, Ferrocene, machine learning, Photo-Fenton, Full pH range