Feasibility Research on Rapid Detection of Prochloraz in Green Tea Soft Drink by Near-Infrared Spectroscopy

Abstract

A rapid and non-destructive method, based on near-infrared spectroscopy (NIRS) was established for screening prochloraz in green tea soft drink. Two chemometric methods, including partial least-squares discriminant analysis (PLS-DA) and least-squares support vector machines (LS-SVM), were used to establish the calibration model for full-spectrum classification. The results of LS-SVM outperformed those of PLS-DA, with the classification accuracy of 100 % for the calibration set and the prediction set when the two parameters, γ and σ ², were 454.994 and 1,057.77, respectively, while the classification accuracy of PLS-DA for the calibration set and the prediction set is 99.1 and 94.8 %, respectively. In order to avoid model overfitting, the procedure was applied to the analysis of prochloraz residue in other four kinds of green tea soft drinks. Good results were also received. Due to the small difference between the blank samples and the contaminated ones, an interval, not a threshold, was set. The samples in the interval are suspicious which should be detected by chromatographic method. Thus, after screening by NIRS and the further chemometric analysis, the appearance of false negative samples was avoided, and workload could be greatly reduced.