LAUSR

BACKGROUND Chlorpyrifos is a commonly used organophosphorus pesticide in agriculture. However, its neurotoxicity poses a huge threat to human health. In this study, a chitosan-modified filter paper-based SERS active substrate (Ch/AgNPs/paper) was fabricated and used to detect trace amounts of chlorpyrifos in 120 treated wheat samples. RESULTS Results showed that the Ch/AgNPs/paper substrate could be employed to enhance the chlorpyrifos spectral fingerprint only up to a concentration of 0.000558 mgL-1 . Following Raman spectra acquisition, three pre-processing methods including Savitzky-Golay (Savitsky-Golay filter with a second order polynomial) smoothing with first derivative and second derivative and normalization were used to reduce baseline variation and increase resolutions of spectral peak features of the original spectra dataset. Then, prediction models based on Partial Least Squares (PLS) were established for detecting chlorpyrifos pesticide residue in wheat. The PLS model with normalization yielded optimal result, with correlation coefficient of 0.9764, root mean square error of prediction of 1.22 mg L-1 in the prediction, and relative analysis deviation of 4.12. Five unknown samples were prepared to verify the accuracy of the prediction model. The predicted recoveries were calculated to be between 97.25 and 119.38% with absolute t value of 0.598. The value of a t-test shows that the prediction model is accurate and reliable. CONCLUSION This study demonstrates that the proposed method can achieve a rapid detection of chlorpyrifos in wheat. This article is protected by copyright. All rights reserved.