Abstract
Flexible surface-enhanced Raman scattering (SERS) sensors offer a promising solution for the rapid in situ monitoring of food safety. The sensor's capability to furnish quantitative detection and retain recyclability is crucial in practical applications. This study proposes a self-cleaning flexible SERS sensor, augmented with an intelligent algorithm designed for expeditious in situ and non-destructive thiram detection on apples. Flexible carriers were prepared via electrostatic spinning, while cuprous oxide spheres decorated with silver (Cu2O@Ag) were synthesized through surfactant-mediated in situ reduction of silver spheres. Then, PAN/Cu2O@Ag/Au@AgNPs flexible sensors with both SERS enhancement and photocatalytic degradation effects were generated by self-assembling core-shell Au@Ag nanoparticles on the flexible carriers. Convolutional neural network (CNN) and competitive adaptive reweighted sampling-partial least squares (CARS-PLS) algorithms were applied for the quantitative prediction of thiram. The results showed that the CNN algorithm has better performance, with correlation coefficient of 0.9963 and detection limit of 0.020mg/L, respectively. Notably, the flexible SERS sensor could be recycled at least 5 times, with thiram detection recovery ranging from 88.32% to 111.80%. This self-cleaning flexible sensor combined with deep learning algorithm has shown significant potential for applications in food safety monitoring.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call