大白菜中马拉硫磷农药的表面增强拉曼光谱快速检测

Rapid detection of malathion residues in Chinese cabbage by surface-enhanced Raman spectroscopy

为了检测大白菜中马拉硫磷农药残留,该文采用表面增强拉曼光谱技术结合化学计量学方法建立马拉硫磷残留的快速检测模型。采用硫酸镁、N-丙基乙二胺、石墨化炭黑和C18去除大白菜中蛋白质、脂肪、碳水化合物等物质的影响。利用不同预处理方法对原始光谱信号进行预处理,建立大白菜中马拉硫磷残留的偏最小二乘模型。研究发现,大白菜中马拉硫磷的检测浓度达到1.082 mg/L以下;归一化预处理后建立的模型预测性能最好。配制5个未知浓度样本验证模型的准确度,预测值与真实值相对误差的绝对值为0.70%～9.84%,预测回收率为99.30%～109.84%;配对t检验的结果表明样本的预测值与真实值之间无明显差异,说明模型是准确可靠的。结果表明,SERS（surface-enhanced Raman spectroscopy）方法可以实现大白菜中马拉硫磷残留的快速检测。

The traditional pesticide residues detection methods had the disadvantages of complex sample preparation,expensive apparatus and high cost. For developing a rapid analysis detection method of pesticide residues, we investigated a surface-enhanced Raman spectroscopy（SERS） method coupled with colloidal gold for detection and characterization malathion residues in Chinese cabbage. Chemometric method was used to establish a rapid detection model of malathion pesticide residues in Chinese cabbage. A 200 mg/L standard solution was prepared by dissolving malathion power in acetonitrile. The standard solution was serially diluted with ultrapure water to prepare working solutions of 100, 50, 20, 15,10, 5, 2, 1 and 0.5 mg/L. Fresh Chinese cabbages were collected from the agronomy experimental base of Jiangxi Agricultural University in June 2015. The Chinese cabbages were used to prepare samples as follows. 50 g Chinese cabbages were weighed and transferred on a plastic wrap. 76 Chinese cabbage samples were manufactured by spraying different concentration standard solution with a sprinkling can, and each concentration has two parallel samples. Then the76 samples were homogenized separately by pulverizer. After that, the sample preparation steps were implemented for both SERS collection and GC-MS measurement as follows. 1） 10 g homogenized chinese cabbage sample, 1 g anhydrous sodium acetate, 5 g sodium chloride and 10 m L acetonitrile were blended in a centrifuge tube of 50 m L, and the centrifuge tube was vibrated for 1 min with a vortex mixer. A homogeneous solution was obtained and then separated for 5 min at a speed of4 200 rpm on the centrifuge, and a yellow supernatant was acquired. 2） 2 m L of the supernatant was injected to a centrifuge tube of 15 m L containing anhydrous Magnesium sulfate, PSA, graphitized carbon and C18 for removing the effect of protein,fat, carbohydrates and other substances in Chinese cabbage. The centrifuge tube was blended for 1 min and then centrifuged for 5 min at a speed of 4 200 r/min. Then, the colourless supernatant was filtered. The filtrate was used directly for SERS measurement in the Optics-Electrics Application of Biomaterials Lab. 3） 1 m L of the filtrate was transferred into a10 m L centrifuge tube and condensed with a termovap sample concentrator at 60 ℃ until the solvent absolutely evaporated.4） The concentrated pesticide was diluted with 1 m L ethyl acetate and shaken for a moment. Then the eluted solution was transferred into a vial and used to measure its actual value by GC-MS in Jiangxi Entry-Exit Inspection and Quarantine technology center. Then three methods as SNV, MSC and Normalization were used to optimize the original Raman spectra signals, and the PLS models of malathion pesticide residues in Chinese cabbage were established.The limit of detection（LOD） can reach the level of 1.082 mg/L by SERS method, and the concentration can meet the tolerance levels for malathion pesticide residues in chinese cabbage. The model predictive performance used normalization preprocessing method was optimal. The correlation coefficient of the calibration samples model（Rc） was 0.983 2, RMSECV was 1.78 mg/L,the correlation coefficient of prediction model（Rp） was 0.973 2, and RMSEP was 2.37 mg/L. The model results of the higher Rp value and the lower RMSEP value indicated that the method of SERS could accurately predict the malathion pesticide residues in Chinese cabbage. The five unknown concentration samples were prepared to verify the accuracy of the prediction models. The absolute values of relative deviation were calculated to be between 0.70%-9.84%. The predict recoveries were calculated to be between 99.30%-109.84%. These indicated that the SERS method was receivable and credible for rapid detection of malathion pesticide residues in Chinese cabbage. The t value was 1.589, less than t0.05,4=2.776. The results of t test demonstrated that the difference between SERS and GC-MS was not significant. This study demonstrates that SERS is capable of detecting and identifying malathion pesticide residues in Chinese cabbage quickly and accurately.