Study On Classification And Quality Analysis Of Oolong

Oolong, originated in Fujian, is unique to China which mainly produced in Fujian, Guangdong and Taiwan. In recent years, with the rapid development of it, oolong industry brings more and more consumers and benefits to its producing area. However, diversity and similarity of oolong, as the reasons for adulteration, have seriously influenced the supervision effect and the sustainable development. In this thesis, the methods for classification and quality analysis of oolong had been established by near-infrared spectrum (NIR), comprehensive two-dimensional gas chromatography-time-of-flight mass spectrometry (GC×GC-TOFMS) and chemometrics. The thesis includes four chapters.In the first chapter, the characteristics, problems and their causes of oolong were described. Recent advances on classification and quality analysis of tea were also reviewed. In addition, the data analysis methods in chemometrics involved with this thesis were introduced. At last, the purpose and contents of this research had been pointed out.Second chapter presents a fast and non-destructive analytical method of identifying oolong teas by NIR. Firstly, principal component analysis (PCA) method was used to classify oolong samples from different plantations, areas and varieties. Then,210 samples including Tieguanyin, Huangjingui, Benshan, Maoxie and Meizhan were collected. Prediction model was built with principal component analysis (PCA) combined with multiplicative scatter correction (MSC) using spectra region of 1100 nm-1300 nm and 1640 nm-2498 nm. The classification accuracy of the models to the prediction samples was 90%.This study provides a theoretical reference for designing NIR Analyzer for oolong.In the third chapter, a detection method of aroma components in oolong based on GC×GC-TOFMS and headspace solid phase micro extraction (HS-SPME) had been developed. The relative peak areas of common compounds were used for discrimination of five oolong varieties. The result showed PCA provided a way to visualize the difference among these varieties. And Fisher’s discriminant analysis (FDA) was used for creating four discriminant functions and showed 97.9% of accuracy. Besides, nine compounds had significant impact on the results of classification. The study will actively promote the research on aroma components and classification of oolong.In chapter four, Tieguanyin samples were extracted by acetone and analyzed by GCxGC-TOFMS. The result was used for quality analysis of Tieguanyin. This study showed that there were differences between actual value and the clustering result of ward method. But step discriminant analysis (SDA) and FDA can give a much more accurate result, including 95.8% accuracy and five compounds with significantly difference.