基于近红外光谱技术的当归提取物混合工艺研究

目的采用近红外光谱技术(NIR)建立当归提取物混合物中当归挥发油β-环糊精包合物的定量模型,并用于优化当归提取物混合工艺。方法以当归挥发油β-环糊精包合物含量为真实值,运用偏最小二乘法(PLS)建立近红外光谱与当归挥发油β-环糊精包合物含量之间的定量校正模型,并对混合工艺进行优选。结果所建近红外光谱定量模型的相关系数(R2)为99.33%,交互验证误差均方根(RMSECV)为0.195,验证集平均预测回收率为98.53%,所建立的当归提取物测定模型预测能力较好;当归提取物混合物不同部位中挥发油β-环糊精包合物含量测定的RSD值分别为1.88%、1.15%、2.71%,均小于5.0%,当归提取物的混合工艺为:混合转速5 r·min^-1,混合时间25 min。结论近红外光谱技术可用于当归提取物混合工艺过程中产品的检测与评价。

The Reserch of the Mixing Process of Angelica Sinensis Extract Based on Near-Infrared Spectroscopy Technique

OBJECTIVE To establish a quantitative model of Angelica sinensis volatile oilβ-cyclodextrinβ-CD inclusion compound in Angelica sinensis extract mixture by near-infrared spectroscopy(NIR)and optimize the mixing process of Angelica sinensis extract.METHODS Taking the actual content of Angelica sinensis volatile oil inclusion compound as the true value,the quantitative correction model between NIR and Angelica sinensis volatile oilβ-CD inclusion compound content was established by partial least squares(PLS)method,and the mixing process was optimized.RESULTS The correlation coefficient(R2)of the established NIR quantitative model was 99.33%,the root mean square(RMSECV)of the interactive verification error was 0.195,and the average prediction recovery rate of the verification set was 98.53%,indicating that the model had good prediction ability.The RSD values of volatile oilβ-CD inclusion compounds in different parts of Angelica sinensis extract mixture were 1.88%,1.15%,and 2.71%respectively,all less than 5.0%,indicating that when the Angelica sinensis extract was mixed at 5 r·min^-1,the mixture was almost evenly mixed at 25 min.CONCLUSION NIR technology can be used to detect and evaluate the mixing process of Angelica sinensis extract.