基于近红外光谱的维生素C片无损鉴别分析

基于近红外光谱技术与化学计量学方法,建立了一种国内外不同品牌维生素C片的无损鉴别方法。采集了国内外8个品牌的维生素C片共计40个样本的近红外光谱数据,比较了完整样品以及粉末样品的近红外光谱,采用连续小波变换技术消除背景干扰和基线漂移,基于标准偏差与相对标准偏差的变量筛选方法筛选出具有代表性的波数点,结合主成分分析方法对国内外不同品牌维生素C片进行鉴别分析。结果表明:原始光谱存在着明显的背景干扰和基线漂移现象,且粉末样品的重现性要优于完整样品;单纯使用原始光谱无法辨别来自不同品牌的维生素C片;连续小波变换可以有效消除背景干扰,提高模型鉴别能力;完整样品的鉴别准确率优于粉末样品,说明国内外不同品牌维生素C片主要成分基本一致,可能是辅剂和工艺上存在细微差异。通过结合近红外光谱分析技术与化学计量学方法,可实现对国产以及进口不同品牌维生素C片的鉴别分析。

Nondestructive identification of vitamin C tablets from different domestic and foreign brands by near infrared spectroscopy

Based on near infrared spectroscopy and chemometrics methods,a nondestructive identification method for different brands of vitamin C tablets was established. Firstly,the near infrared spectra of 40 vitamin C samples from 8 domestic and overseas brands were collected,and the spectra of the complete and powder samples were compared. Secondly,the background interference and baseline drift were eliminated by continuous wavelet transform technology. The representative wave-numbers were screened by variable selection method based on the standard deviation and relative standard deviation. Finally,the principal component analysis was applied for the identification of vitamin C samples. The results showed the background interference and baseline drift in the original spectra were obvious,while the reproducibility of powder samples was better than that of the complete samples. The vitamin C samples from different brands are unable to be distinguished by using the original spectra. Continuous wavelet transform can effectively eliminate background interference,and improve the model identification ability. Besides,the accuracy of identification of complete samples is better than that of powder samples,indicating that the main components of different brands samples are similar,and the subtle differences may remain in the adjuvants and processes. By combining the near infrared spectroscopy with chemometric methods,the vitamin C from different brands and different manufacturers can be accurately identified.