A novel microfluidic metal grating integrated terahertz sensor has been designed, which is composed of a metal microstructure array-dielectric layer-metal layer, where the dielectric layer is the microfluidic channel carrying the analyte. By adjusting the structural parameters of the metal grating sensor, a highly confined electromagnetic field can be obtained in the microfluidic channel, thereby significantly enhancing the interaction between the analyte and the terahertz wave and improving the terahertz detection sensitivity. The metal grating described in this paper is composed of an array of square holes, which is manufactured using laser micromachining technology, so that the measurement method is simplified and improved. The results show that the addition of different concentrations of carbendazim solution resulted in a redshift of the overall spectrum, with the highest sensitivity reaching 8.773 GHz/mg L−1, which is about eight times more sensitive than the traditional terahertz transmission sensor. The relative error of using this method to determine carbendazim levels in orange juice samples was less than 5.3%. The terahertz time-domain spectroscopy technology combined with the metal grating integrated microfluidic sensor can improve the sensitivity of sample detection and realize the rapid detection and analysis of trace elements.
The presence of pesticide residues in agricultural products, such as tea, poses significant health risks to consumers and has become a major concern for the food industry. Carbendazim, a widely used fungicide, has been identified as one of the harmful contaminants in tea. Rapid and accurate detection of carbendazim residues is therefore essential for ensuring food safety and quality. In this study, we propose a novel machine learning-assisted electrochemical sensor based on a CuO/reduced graphene oxide (rGO)/poly N-phenylglycine (NPAN) composite electrode for the rapid detection of carbendazim residues in tea. The CuO/rGO/NPAN electrode offers enhanced selectivity and sensitivity, enabling reliable detection of carbendazim even in the presence of other pesticides. By systematically evaluating various machine learning models and preprocessing techniques, we optimize the sensor’s performance and achieve accurate classification and identification of carbendazim residues. Our research contributes to the field of food safety assessment by introducing a novel electrode design and demonstrating the effectiveness of machine learning-assisted electrochemical sensors for the detection of carbendazim residues in tea.
危险品化学品经营许可证（带存储）