CPC G01N 21/658 (2013.01) [B01L 3/502 (2013.01); G16C 20/20 (2019.02); B01L 2200/028 (2013.01); B01L 2200/16 (2013.01); B01L 2300/0663 (2013.01); G01N 2333/415 (2013.01)] | 8 Claims |
1. A method for detecting pesticide residues in tea based on a surface-enhanced Raman scattering (SERS) sensor, comprising:
(S1) preparing silver on a surface of an octahedral cuprous oxide template by reduction, adding an acid to dissolve the octahedral cuprous oxide template to obtain an octahedral silver hollow cage, and preparing gold on a surface of the octahedral silver hollow cage by reduction to obtain an octahedral gold-silver hollow cage sensor;
wherein the step (S1) comprises:
(S1.1) reacting copper chloride dihydrate as a copper source with sodium hydroxide in the presence of polyvinylpyrrolidone as a stabilizer to obtain a copper hydroxide precipitate; reducing the copper hydroxide precipitate into cuprous oxide with ascorbic acid; and subjecting the cuprous oxide to centrifugal washing with an aqueous ethanol solution three times, and drying to obtain the octahedral cuprous oxide template;
(S1.2) preparing a silver layer on the surface of the octahedral cuprous oxide template from silver nitrate in the presence of sodium citrate and sodium borohydride, so as to obtain a silver-cuprous oxide octahedron; and adding a 1 vol. % acetic acid solution to a solution of the silver-cuprous oxide octahedron followed by stirring for 2 h and centrifugal washing to obtain the octahedral silver hollow cage; and
(S1.3) reacting tetrachloroauric acid (HAuCl4) with sodium borohydride in the presence of sodium citrate as a stabilizer on the surface of the octahedral silver hollow cage followed by washing multiple times with the aqueous ethanol solution and concentration to obtain the octahedral gold-silver hollow cage sensor;
(S2) mixing the octahedral gold-silver hollow cage sensor with a pesticide standard solution followed by analysis using a Raman spectrometer to obtain SERS spectral data; processing the SERS spectral data by a Baseline Removal method and a Savitzky-Golay smoothing algorithm of a Python third-party library to eliminate a background interference and a fluorescence noise; and based on processed SERS spectral data, building a one-dimensional convolutional neural network by using a Keras framework to construct a quantitative model; and
(S3) mixing the octahedral gold-silver hollow cage sensor with a pretreated tea sample followed by analysis using the Raman spectrometer to collect sample SERS spectral data, processing the sample SERS spectral data by the Baseline Removal method and the Savitzky-Golay smoothing algorithm, and inputting processed sample SERS spectral data into the quantitative model, so as to obtain a predicted result; and then calculating a spiked recovery rate by comparing the predicted result with a spiked concentration of the tea sample.
|