基于近红外光谱的五味子提取过程在线检测方法研究

目的 利用近红外(NIR)光谱法,建立定量模型,实现五味子提取过程中关键指标的快速检测。方法 在线采集五味子提取液的近红外光谱,以高效液相色谱法和差重法获取参照,采用最小二乘回归法(PLSR)分别建立五味子醇甲、果糖、葡萄糖和可溶性固形物的定量模型。结果 定量模型的校正集相关系数R、校正均方根误差RMSEC、预测均方根误差RMSEP分别为：五味子醇甲0.99792、1.41、8.76,果糖0.99246、0.0919、0.333,葡萄糖0.99371、0.0954、0.448,可溶性固形物0.98723、3.07、11.2。结论 该方法快速、便捷,结果准确,适用于五味子提取过程的快速检测。

On-line monitoring of extraction process for Schisandra chinensis using near-infrared spectroscopy

It is the objective of this study to establish quantitative models to detect quality indicators rapidly during the extraction process of Schisandra chinensis using NIR spectroscopy. NIR spectroscopy was collected on-line and PLSR models were developed for Schisandrin,fructose, glucose and soluble solids content, which were measured with HPLC and weight loss methods. The correlation coefficients of cross validation for Schisandrin,fructose, glucose and soluble solids content were 0.99792, 0.99246, 0.99371, 0.98723, respectively. RMSEC for Schisandrin,fructose, glucose and soluble solids content were 1.41, 0.0919, 0.0954, 3.07, respectively. RMSEP for Schisandrin,fructose, glucose and soluble solids content were 8.76, 0.333, 0.448, 11.2, respectively. In conclusion, NIR models were developed successfully to determine the quality indicators including Schisandrin,fructose, glucose and soluble solids content rapidly and accurately.