|
|
Study on Identification of Confusable Chinese Herbs by FTIR Combined with Multivariate Statistical Analysis
ZHANG Chuanyun, LI Desheng, LIANG Siyue, LI Lun, LIU Ruifan, ZHANG Deqing
2025, 40(3):
50-57.
In this paper, Fourier transform infrared spectroscopy ( FTIR ) combined with multivariate statistical analysis was used to identify and analyze the two kinds of confusable Chinese herbs, Sedum emarginatum and Portulaca oleracea. The analysis results of infrared spectra showed that the chemical components in these two kinds of confusable Chinese herbs were similar, and both of them contained esters, proteins, alkalis, saccharide and flavonoids. However, there were obvious differences in their infrared spectra and second derivative infrared spectra in the 1800-700cm-1 band. In addition, hierarchical cluster analysis (HCA), principal component analysis (PCA), and discriminant analysis (DA) were sequentially conducted with the second derivative infrared spectra of 32 random herbal samples selected from them in the 1800-700cm-1 range. The HCA results indicated that when the inter-group distance was 6, samples of the 32 Chinese herbs were clustered into two categories. The PCA revealed that the cumulative variance contribution of the first three principal components had exceeded 96%, and on the 3D distribution figure of principal components, the 32 samples were aggregated in two different regions with a significant classification effect, and the samples of the same Chinese herbs had good aggregation. Then, the first three principal components were utilized as the variables for discriminant analysis. The 32 herbal samples were accurately classified into two distinct groups–Sedum emarginatum and Portulaca oleracea–based on the discriminant function. Furthermore, through discriminant verification, 100% correct discrimination of the two newly collected unknown samples was achieved. The above research results can provide theoretical basis and technical reference for the development of spectral identification technology of confusable Chinese herbs in the field of rapid detection and classification of Chinese herbs.
References |
Related Articles |
Metrics
|