Detection of difenoconazole pesticides in pak choi by surface-enhanced Raman scattering spectroscopy coupled with gold nanoparticles

Abstract

Difenoconazole is a highly effective and broad-spectrum triazole bactericide pesticide and was generally applied to protect and cure foods such as vegetables and fruits. Because pesticide residue may pose a threat to humans for its contamination of foodstuffs, development of the detection and identification of trace pesticides is essential. In this study, we present a surface-enhanced Raman scattering (SERS) spectroscopy method for detecting difenoconazole in pak choi using a portable Raman analyzer. The entire experiment for each sample, including sample preparation, solvent extraction and SERS spectra collection, was completed in about 15 min. Density functional theory (DFT) calculations were executed with a Gaussian 03 at B3LYP/6-311G basis sets. Solid, theoretical and SERS spectroscopy techniques of difenoconazole were compared to analyze the assignments. Magnesium sulfate, PSA, graphitized carbon and C18 were used to decrease the distractions of chlorophyll, protein and other substances in pak choi. The original spectra were preprocessed by the methods of MSC, SNV, first derivative, second derivative, smoothing and normalization and then used to establish prediction models by the method of partial least squares (PLS); the prediction model property of standard normal variate (SNV) was optimal. The correlation coefficient of the prediction model (R_p) was 0.9458; the root mean square error of prediction (RMSEP) was 3.27 mg L⁻¹. The higher R_p value and lower RMSEP show that the established model of SNV can precisely detect difenoconazole residues in pak choi. Five unknown pak choi samples containing difenoconazole pesticide were used to verify the accuracy of the prediction model, the values of relative deviation were calculated to be between 2.42% and 9.95%, and the predicted recovery rates were calculated to be between 94.64% and 109.95%. The t value was 0.475, less than t_{0.05, 4} = 2.776, which indicates the difference between the predicted and measured values is not obvious. This study demonstrates that the SERS technique serves as an effective approach for quick and stable detection of difenoconazole in pak choi.