论文部分内容阅读
目的:建立近红外漫反射光谱法快速测定夏枯草中迷迭香酸的含量方法。方法:采用高效液相色谱法测定不同产地的170批次夏枯草中迷迭香酸的含量,同时采集近红外漫反射光谱,对原始光谱进行多元散射校正(MSC)、一阶导数(First derivative)和S-G(Savitzky-Golay filter)平滑等光谱预处理方法,采用偏最小二乘法(PLS)建立近红外定量分析模型,实现夏枯草中指标性成分迷迭香酸含量的快速测定。结果:所建立的迷迭香酸近红外定量分析模型,模型R~2为0.9768,RMSEC和RMSEP分别为0.0387和0.0441,表明所建近红外模型预测准确度高。交叉验证均方差RMSECV为0.0706,表明所建模型稳健性好。结论:本研究所建迷迭香酸近红外定量分析模型具有很好的预测准确度和稳健性,为市场上夏枯草质量的快速评价提供新的方法。
Objective: To establish a rapid method for determination of rosmarinic acid in Prunella vulgaris by near infrared diffuse reflectance spectroscopy. Methods: The content of rosmarinic acid in 170 batches of Prunella vulgaris L. from different areas was determined by high performance liquid chromatography (HPLC). At the same time, the near-infrared diffuse reflectance spectra were acquired. The original spectra were subjected to multiple scattering calibration (MSC), first derivative ) And SG (Savitzky-Golay filter) smoothing and other spectral pretreatment methods, using partial least squares (PLS) to establish a quantitative model of near infrared spectroscopy to achieve rapid determination of prunella rosemary acid content of the indicator component. Results: The established model of rosmarinic acid near-infrared quantitative analysis model R ~ 2 was 0.9768, RMSEC and RMSEP were 0.0387 and 0.0441, respectively, indicating that the proposed near infrared model has high prediction accuracy. The RMSECV of cross-validation was 0.0706, which showed that the model was robust. Conclusion: The rosmarinic acid NIR quantitative analysis model established in this study has good predictive accuracy and robustness, and provides a new method for rapid assessment of Prunella vulgaris quality in the market.