三黄片专属性近红外定量分析模型的建立初探

Construction of quantitative model for determination of Sanhuang Tablet

目的:利用近红外漫反射光谱(NIRDRS)分析技术和化学计量学的方法对三黄片中的盐酸小檗碱进行无损、快速定量分析。方法:从全国不同生产企业中筛选出9家生产的三黄片为分析对象,用HPLC测定三黄片中的盐酸小檗碱含量,采用偏最小二乘法(PLS),光谱数据预处理方法为矢量归一化,谱段范围为6402.9～5596.8cm-1建立NIR光谱与HPLC测定值之间的多元校正模型,预测三黄片中盐酸小檗碱的含量。结果:36个样品经内部交叉验证建立预测模型,纯度范围为0.7133%～2.1880%,内部交叉验证均方差(RMSECV)为0.093%,决定系数为96.24;用16个样品进行外部验证,外部验证均方差(RMSEP)为0.127%,决定系数为92.41,外部验证预测值与HPLC测定值的相关系数为0.9673,预测值平均回收率为100.7%,RSD=5.63%,n=16。结论:该方法快速、简便,具有一定的专属性,可用于药品现场快速分析。

AIM：To develop a near infrared diffuse reflectance spectroscopy（NIRDRS） method for rapidly and nondestructively determining berberine hydrochloride content of Sanhuang Tablet（Radix et Rhizoma Rhei,berberine hydrochloride and Extractum Scutellariae Siccus）.METHODS：Berberine hydrochloride content of Sanhuang Tablet from nine different pharmaceutical factories was measured by HPLC.Vector normalization was used in the pretreatment of spectroscopic data.Multivariate calibration model based on PLC between NIR data and HPLC value fell in the range from 6 402.9 to 5 596.8 cm-1 used for the prediction of berberine hydrochloride content.RESULTS：Thirty-six batches of samples were used as calibration set to establish pretreating model,the purity range was 0.713 3%-2.188 0%,the root mean square error of cross validation was 0.093%,the determination coeficients was 96.24,and 16 batches as validation set.The root mean square error of prediction was 0.127%.The determination coeficients was 92.41.The correlation coeficient of the true value and predictied value was 0.967 3.The average recovery of the prediction set was 100.7%（RSD = 5.63%,n = 16）.CONCLUSION：The method is rapid,simple,specialized and can be applied to fast analysis on testing drug vehicles.