黄芪提取过程总皂苷质量浓度的在线监测

目的 利用近红外光谱分析技术对黄芪提取过程进行在线监测,实时反映药效成分的溶出信息.方法 利用远程光纤流通池在线采集黄芪提取过程中提取液的近红外光谱,同时采集提取液样品,利用HPLC-MS方法测得提取液中黄芪皂苷Ⅱ、异黄芪皂苷Ⅱ、黄芪皂苷Ⅰ和黄芪甲苷4种成分的质量浓度作为对照值,利用偏最小二乘法建立4种成分及黄芪总皂苷的定量分析模型,将所建的黄芪总皂苷定量分析模型用于黄芪提取过程的在线分析,实时反映提取液中黄芪总皂苷的质量浓度信息.结果 黄芪提取液中4种皂苷的近红外定量分析模型相关系数分别达到0.9876、0.9640、0.8571、0.9816,黄芪总皂苷定量分析模型的相关系数为0.9932,校正均方差(RMSEC)为5.94,内部交叉验证的相关系数为0.9766,交叉验证均方差(RMSECV)为11.1.将黄芪总皂苷定量分析模型用于3个批次提取过程的在线监测,能够及时准确地反映提取液中黄芪总皂苷的质量浓度信息.结论 利用本实验所建立的方法,可以实现以黄芪总皂苷为检测指标,对黄芪提取过程的在线监测,为黄芪提取过程的终点判断和工艺优化提供参考.

On-line monitoring of total saponins in extracting process of Astr ti Radix

Objective To investigate the effective components dissolution using near-infrared(NIR) spectroscopy for the on-line monitoring of total saponins in the extracting process of Astragali Radix.Methods The extracts and their NIR spectra were collected using a long-distance fibre-optics flow cell in the extracting process of Astragali Radix,and the contents of four compounds,astrgaloside II,isoastragaloside II,astragaloside I,and astragaloside IV,were determined as control values with HPLC-MS method.The quantitative analysis models of the above four compounds and total saponins in Astragali Radix were established using partial least square method.The established quantitative analysis model of total saponins in Astragali Radix was applied to the on-line monitoring in the extracting process of Astragali Radix to reveal the concentration of total saponins in the extracting process.Results The correlation coefficients of the four compounds reached 0.987 6,0.964 0,0.857 1,and 0.981 6,respectively,and 0.993 2 for the total saponins.The root mean square error of calibration(RMSEC) was 5.94,the correlation coefficient of inner cross-validation was 0.976 6,and root mean square error of cross validation(RMSECV) was 11.1.Using the calibration model of the total saponins for the on-line monitoring of Astragali Radix extracting process,the dissolving information of the total saponins was obtained synchronously and exactly.Conclusion This method is beneficial to the estimating of the end point and the optimizing of the process parameters.