肉桂中4种成分近红外定量分析模型的建立

目的:采用近红外光谱法(NIR,near infrared spectroscopy)结合高效液相色谱法(HPLC,high performance liquid chromatography)建立肉桂饮片的快速评价方法。方法:收集不同产地的肉桂86批,采用HPLC法测定不同批次肉桂饮片中香豆素、肉桂醇、肉桂酸、肉桂醛的含量,同时采集不同批次肉桂的NIR光谱。以NIR光谱为自变量,以香豆素、肉桂醇、肉桂酸、肉桂醛的含量为因变量,采用偏最小二乘法(PLS,partial least squares)建立肉桂中4种成分的定量分析模型,并以校正集进行验证。结果:肉桂饮片中香豆素、肉桂醇、肉桂酸、肉桂醛近红外定量分析模型的相关系数(r)依次为0.952 8,0.977 7,0.961 9,0.992 2,校正集预测误差均方根(RMSEC)依次为0.012 2,0.006 1,0.004 3,0.82 g·g^-1,均方根误差(RMSECV)依次为0.015 8,0.011 2,0.002 0,1.481 1 g·g^-1,测试集预测误差均方根(RMSEP)依次为0.017 8,0.010 3,0.005 6,1.63 g·g^-1。结论:该实验所建立的肉桂饮片中香豆素、肉桂醇、肉桂酸、肉桂醛4种有效成分的近红外定量分析模型具有较好的准确性,可用于这4种成分的快速同步测定,为实现肉桂饮片质量的快速评价提供依据。

Establishment of Near Infrared Quantitative Analysis Model for Four Components in Cinnamomi Cortex

Objective: To establish a rapid evaluation method for Cinnamomi Cortex decoction pieces by near infrared spectroscopy. Method: The contents of coumarin,cinnamalol,cinnamic acid and cinnamaldehyde in 86 batches of Cinnamomi Cortex of different origins were determined by HPLC. And the NIR spectra of different batches of Cinnamomi Cortex were also collected. With NIR spectrum as independent variable and coumarin,cinnamalol,cinnamic acid and cinnamaldehyde as dependent variables,a quantitative analysis model of four components in cinnamon was established by partial least squares method. Result: The correlation coefficients (r) of coumarin,cinnamic alcohol,cinnamic acid and cinnamaldehyde near infrared quantitative analysis models were 0.952 8,0.977 7,0.961 9,0.992 2,root mean square error of cross(RMSEC) were 0.012 2,0.006 1,0.004 3,0.82 g·g^-1,root mean square errorof cross-validation(RMSECV) were 0.015 8,0.011 2,0.002 0,1.481 1 g·g^-1,and root mean square error of prediction(RMSEP) were 0.017 8,0.010 3,0.010 3,0.005 5,1.63 g·g^-1. Conclusion: The established NIR quantitative analysis model of four active ingredients in Cinnamomi Cortex slices has a good accuracy, and provides a basis for rapid evaluation of the quality of Cinnamomi Cortex slices.