New Approaches To Chemometric Methods In Process Analysis And Quality Control Strategy On Chinese Herbal Medicine

There are six parts in this article. The applications of chemometric method in process analysis and fingerprint of Chinese herbal medicine are reviewed in the first two chapters. In the first chapter, the development and application of the EFA-MCR method and the study of chemometric applied in standardization of capillary electrophoresis (CE) are introduced. In the second chapter, the analysis method of fingerprint of Chinese herbal medicine and the application of chemometric used in fingerprint of Chinese herbal medicine are introduced.In the third chapter, the application of evolving factor analysis (EFA) and multivariate curve resolution (MCR) in the enzyme reaction mechanism is discussed. Syringic acid is used as the model compound of lignin to study the reaction mechanism of laccase catalysed decomposition process of syringic acid and the reaction mechanism of laccase catalysed decomposing lignin. Farther more, based on the EFA-MCR method, the concentration of the reactant and the intermediate product in reaction process can be confirmed. Meanwhile, spectrum of reactant and intermediate product can be obtained by means of MCR. The important information of reaction process is provided, and the reasonable mechanism is deduced.In the forth chapter, the study of chemometric appliced in CE is also discussed. Many investigators complained that the repeatability of analyte migration times is poor in CE. In this study, we developed the new standardization methods to eliminate the influence. In this paper, the markers are used to determine the basis position and then correct the data of sample by the migration time of standard sample, and make migration time of samples consistent with the standard sample by the criterion of the marker. The problem of time transition was corrected in this way. Then according to the peak height or peak area of the marker in the sample (here what we use is peak high) compared with the standard sample, the sample data was zoomed appropriately. The absorbance error wasmade to get correctly. The wavelet de-noise method is also used to make the data11smooth and get a good baseline.The Similarity algorithm on fingerprint of Chinese Herbal medicine is discussed in the fifth chapter. This includes data standardization and similarity comparison of complex system. For the characteristic of Chinese herbal medicine, the fuzzy similarity algorithm is developed.The sixth chapter is the simple introduction of the similarity evaluation system for the chromatographic fingerprint of Chinese herbal medicine developed in our laboratory.