Objective To establish the rapid detection method for imperatorin in Angelicae dahuricae Radix and the amounts of prim-O-glucosylcimifugin and 5-O-methylvisammioside in Saposhnikoviae Radix by near infrared spectroscopy (NIRS).Methods 100 batches of samples were collected from different regions,the content of label components were determined by high performance liquid chromatography (HPLC),and near infrared spectra of samples were also determined.Due to smaller differences between the spectral,it was needed to deal with spectrum by regression methods,pretreatment methods,bands as well as the numbers of principal components.The near infrared spectroscopy quantitative model were established for Angelicae dahuricae Radix and Saposhnikoviae Radix.Results In Angelicae dahuricae Radix,the calibration model of R2 was 0.975 5,RMSEC and RMSEP were respectively 0.011 2 and 0.011 6,the principal component number was 11.In Saposhnikoviae Radix,the calibration model of R2 was 0.977 3,RMSEC and RMSEP were respectively 0.054 5 and 0.051 4,the principal component number was 6.Conclusion The quantitative analysis models established in this research were more accurate and can be used to predict imperatorin in Angelicae dahuricae Radix and the amounts of prim-O-glucosylcimifugin and 5-O-methylvisammioside in Saposhnikoviae Radix,which have good application prospects.%目的 应用近红外光谱技术建立对白芷中欧前胡素,防风中升麻素苷及5-O-甲基维斯阿米醇苷总量的快速检测方法.方法 分别采集自不同产地100批样品,通过高效液相色谱法测定指标成分含量,并测定样品的近红外光谱图.由于光谱图间差异较小,需要对光谱进行处理,通过比较不同回归方法,预处理方法,波段以及主成分数,选择最优模型参数,分别建立白芷中欧前胡素,防风中升麻素苷及5-O-甲基维斯阿米醇苷总量的最优化定量分析模型.结果 白芷校正模型中R2为0.975 5,校正误差均方根(RMSEC)为0.011 2,预测误差均方根(RMSEP)为0.011 6,主成分数为11;防风校正模型中R2为0.977 3,校正误差均方根(RMSEC)为0.054 5,预测误差均方根(RMSEP)为0.051 4,主成分数为6.结论 本研究所建立的近红外定量分析模型预测结果较准确,可用于对白芷中欧前胡素,防风中升麻素苷及5-O-甲基维斯阿米醇苷总量进行预测,具有良好的应用前景.
展开▼
