近红外光谱技术无损检测克咳片中麻黄类生物碱和吗啡含量研究

Nondestructive Determination of Ephedrine Alkaloids and Morphine in Keke Tablets by Near Infrared Spectroscopy

麻黄类生物碱以及吗啡含量对评价克咳片质量具有重要意义,该文基于近红外光谱(NIRS)技术建立了快速、无损、高效的克咳片质量评价方法,采用NIRS和偏最小二乘回归(PLSR)法建立了麻黄类生物碱以及吗啡的定量分析模型,并采用不同预处理方法和不同变量选择方法优化了模型性能。将校正集决定系数(R_(c)^(2))、验证集决定系数(R_(p)^(2))和相对分析误差(RPD)作为模型的评价指标。结果表明,标准正态变量变换(SNV)预处理方法结合竞争自适应重加权采样(CARS)变量选择方法所建立的模型性能最优。克咳片中麻黄类生物碱及吗啡近红外光谱模型的R_(c)^(2)、R_(p)^(2)及RPD分别为0.91、0.93、4.00和0.92、0.94、4.24,说明模型性能较好。该研究基于NIRS和CARS实现了克咳片指标成分的快速、无损分析,为克咳片的质量快检提供了参考。

The content of ephedrine alkaloids and morphine is of great significance for the quality evaluation of Keke tablets.In view of the time and effort of traditional methods for the quality analysis of Keke tablets,this study was based on near infrared spectroscopy(NIRS)to establish a rapid,nondestructive and efficient method for the quality evaluation of Keke tablets.The quantitative analysis model of ephedrine alkaloids and morphine was established by NIRS and partial least squares regres⁃sion(PLSR),and the performance of the model was optimized by different pretreatment methods and variable selection methods.R-square of calibration set(R_(c)^(2)),R-square of prediction set(R_(p)^(2))and re⁃sidual predictive deviation(RPD)were calculated as the evaluation index of the correction model.The results showed that the pretreatment method is standard normal variable transformation(SNV)combined with variable selection for competitive adaptive reweighted sampling(CARS)model has the better performance.The R_(c)^(2),R_(p)^(2) and RPD of the NIR models of ephedrine alkaloids and morphine in Keke tablets were 0.91,0.93,4.00 and 0.92,0.94 and 4.24,respectively,indicating that the model performance was good and the quality detection of Keke tablets was realized.The results show that NIRS and CARS can be used for rapid and non-destructive analysis of Keke tablets,which can provide a good reference for quality detection of Keke tablets.