Cloud machine learning-enhanced peroxidase-like enzyme visual sensor for rapid detection of sulfur-containing pesticides

Sulfur-containing pesticides are widely used in agricultural products because of their high insecticidal and herbicidal properties, but the pesticide residues caused by them pose a serious threat to public safety. In this study, a nanometer peroxidase-like enzyme was prepared as a colorimetric probe, constructing a highly sensitive visual colorimetric sensor for detection of 4 sulfur-containing pesticides (thiophanate-methyl: TM; cartap: CT; dimethoate: DT; temephos: TP). Detailedly，the outer electrons of Au-Ag NCs can catalyze the hydrogen peroxide to oxidize TMB to blue, while the positive sulfur-containing pesticides can interact electrostatic with the negative charge of Au-Ag NCs to expose the silver element in the inner layer. Moreover, sulfur in pesticides binds to silver atoms, resulting in the looser structure and disintegration of Au-Ag NCs, ultimately blocking the pseudo-peroxidase activity to produce a distinct color signal. The results demonstrated that all four pesticides exhibited a good linearity (R 2 > 0.99) over a wide concentration range with a low detection limit of 0.03 mg/L. In real samples such as green tea, cabbage, goji berry, apple etc., the recovery rate for added standards of visual sensor based on smartphone APP ranged between 90 % and 115 %, which is significantly better than that obtained by UV-Vis spectrum, while RSD remained below 5 %. Compared with the recovery results of LC-MS, the deviation between these two methods is less than 10 %. The visual sensor based on Au-Ag NCs nano peroxidase-like enzyme offers a novel approach for highly specific and sensitive detection of sulfur-containing pesticides in food safety monitoring.