| The fingerprint has been increasingly utilized as practical method for the identification and quality evaluation of the Traditional Chinese Medicine (TCM), since it can reflect the difference of samples, such as the species, the origin, sampling data, processing. Pyrolysis-gas chromatography (Py-GC) has been used extensively in the analysis of various polymers and natural products including TCM because the obtained pyrogram can provid more utility information on TCM sample. In this thesis, a method based on the Py-GC fingerprint with pattern recognition is developed for discriminating the species and the origin of Dendrobium.The analytical conditions such as pyrolysis temperature, sample weight, GC capillary column and program temperature were optimized using a Dendrobium candidum Wall, ex Lindl. sample. At the optimized conditions, the 51 peak components on the obtained pyrogram were assigned by Py-GC/MS. The fingerprints of different species for 10 samples of Dendrobium obtained using 0.4 mg at 450℃showed a passably similarity and a good reproducibility. Based on the peak area of 51 pesks in each Py-GC profile, the 10×7 data matrix for principal component analysis (PCA) was obtained. As a result, discrimination of the sample from 3 different species was achieved by Py-GC fingerprints with PCA. Furthermore, the fingerprints of 10 Dendrobium candidum Wall, ex Lindl. samples collected from different origin obtained using 0.4 mg at 450℃showed a highly similarity and a good reproducibility with the RSD of t_R less than 0.12% and the RSD of S% less than 4.91%. Based on the peak area of 31 pesks in each Py-GC profile, the 10×31 data matrix for hierarchical cluster analysis (HCA) was obtained. As a result, discrimination of the sample from 3 different origins was achieved by Py-GC fingerprints with HCA.Thus, the results proved that Py-GC fingerprint with pattern recognition was a simple, rapid method showing suitable for the quality control of TCM. |
PDF Full Text Request