FTIR and Classification Study on the Chinese Compound Recipe

Wang Jing-juan,Zhang Gui-jun,Zhou Qun,Sun Su-qin

DOI: https://doi.org/10.3964/j.issn.1000-0593.2008.02.022

2008-01-01

Abstract:On the basis of building a model of main components and with the fingerprint infrared spectrum as the object-Xiaoerchoufeng powders were rapidly classified and closely studied by using the method of clustering analysis. Because of the tiny difference between the spectra of different compound recipes, including the compound recipe without scolopendra, the compound recipe without scorpio, the compound recipe without bombyx batryticatus, the compound recipe without eupolyphaga seu steleophaga and the compound recipe without periostracum cryptotympanae, the outcome of the clustering is ideal. When tested with the blind samples (which were picked out from the chosen samples) the accuracy of clustering reaches up to 90%. In order to increase the clustering accuracy, the number of the samples should be increased and also these samples should be more characteristic. On the whole, with the combination of the fingerprint spectrum and the clustering analysis, the compound recipe can be distinguished rapidly.