论文部分内容阅读
采用拉曼光谱技术结合QuEChERS(Quick Easy Cheap Effective Rugged and Safe)样本前处理建立了黄瓜上吡虫啉残留量的快速检测方法。以进行了不同前处理步骤(乙腈提取、去水萃取、褪色除杂)的三批黄瓜样本作为实验对象,利用780 nm激光器采集样本的拉曼光谱图,并分别采取偏最小二乘(PLS)和主成分回归(PCR)算法建立了六个黄瓜中吡虫啉含量预测模型。结果表明,仅进行了乙腈(C2H3N)提取一步前处理的样本建模效果最优,校正集及预测集的相关系数均在0.99以上,其中PLS的预测集相对分析误差(RPD)达到5.52,说明模型具有一定的预测精度,此结果可为后续前处理简化研究提供有力依据。
Rapid detection of imidacloprid residue in cucumber was established by Raman spectroscopy combined with QuEChERS (Quick Easy Cheap Effective Rugged and Safe) sample preparation. Three batches of cucumber samples with different pretreatment steps (acetonitrile extraction, dewatering and fading) were used as the experimental subjects. The Raman spectra of the samples were collected with a 780 nm laser and the PLS (partial least squares) And principal component regression (PCR) algorithm to establish the prediction model of imidacloprid content in six cucumbers. The results showed that the model with one-step pretreatment with acetonitrile (C2H3N) was the best model, and the correlation coefficients between the calibration set and the prediction set were all above 0.99. The RPD of PLS was 5.52, indicating that The model has a certain degree of prediction accuracy. This result can provide a strong basis for the subsequent simplification of preprocessing.