Machine learning assisted multi-signal nanozyme sensor array for the antioxidant phenolic compounds intelligent recognition

Abstract

Identifying antioxidant phenolic compounds (APs) in food plays a crucial role in understanding their biological functions and associated health benefits. Here, a bifunctional Cu-1,3,5-benzenetricarboxylic acid (Cu-BTC) nanozyme was successfully prepared. Due to the excellent laccase-like behavior of Cu-BTC, it can catalyze the oxidation of various APs to produce colored quinone imines. In addition, Cu-BTC also exhibits excellent peroxidase-like behavior, which can catalyze the oxidation of colorless 3,3′,5,5′-tetramethylbenzidine (TMB) to form blue oxidized TMB and exhibits higher photothermal properties under near-infrared laser irradiation. Due to the strong reducibility of APs, this process can be inhibited. A dual-mode colorimetric/ photothermal sensor array was constructed, successfully achieving discriminant analysis of APs. Moreover, by integrating artificial neural network (ANN) algorithms with sensor arrays, precise identification and prediction of APs in black tea, coffee, and wine have been successfully accomplished. Finally, with the assistance of smartphones, a portable detection method for APs was developed.