The Study On Aroma Quality Evaluation Of Longjing Tea Assisted With E-Nose

The quality of tea aroma is one of the most important factor of the tea quality. Quality evaluation of tea generally relies on sensory evaluation of human. However, the evaluation accuracy will be affected by physical health status, psychological status, work experience and other factors. In order to improve the accuracy and repeatability of quality discrimination, using auxiliary equipment is necessary. In this work, four different methods for collection of tea aroma were investigated, such as tea infusion method, tea leaves waste method, spray method and tea powder method. And the tea powder method was selected to analysis aroma quality of 59 Longjing tea samples from six geographical origins such as Meijiawu, Longjingcun, Shangmanjuelong, Longwu, Jingling and Fine breeds demonstration farm of Zhejiang province, three breeds such as Longjing 43 breed, Group breed and Yingshuang breed, and five grade levels from grade superfine to grade four. Pattern recognitions PCA and DFA were used to analysis Longjing tea aroma quality in different regions, species and grades. Combined with sensory evaluation results, the feasibility of using electronic nose technology to identify different aroma quality of Longjing tea was analysed. Then, representative tea samples of different aroma quality were analysed by GC-MS to discuss main components of aroma causing the difference.Electronic nose technology has high sensitivity. DFA method can get better distinction results than PCA method. For the three factors of geographical origin, variety and grade, the electronic nose can well distinguish the quality of Longjing tea aroma under different single factor conditions when the other two factors were fixed. While the difference of aroma quality could not be distinguished in the condition of three mixture factors.Using electronic nose to do correlation and PLS analysis for sensory scores of aroma quality and aroma quality sub-properties, such as fire degree, Lixiang smell, fragrance, richness, tenderness. In addition to the sensor LY2/G with lower correlation with all the aroma characteres, the majority of sensor correlation coefficients of sub-properties are close. And sensors P30/1, P30/2, PA/2, T40/1, T30/1, P10/2 have higher correlation coefficients with most sub-properties and could be used as main sensors. Electronic nose can be used to predict Lixiang smell score, the richness of tea aroma and sensory scores of aroma quality of tea samples in similar grades, and the correlation coefficients of PLS model were 0.9478,0.9275 and 0.9398.Major aroma components of different Longjing tea are a-Pinene,2-Pentyl furan, Ocimene, Tetradecane, Nonanal, (E)-Butanoic acid-3-hexenyl ester, cis-3-Hexenyl isovalerate, Benzaldehyde, Linalool and its oxides, (Z)-Hexanoic acid-3-hexenyl ester, Cadinene, Octanol, 2-Ethyl butyrate, Hexanoic acid anhydride, Methyl salicylate, Geraniol, Cis-Jasmone etc. The content of different aroma components in different tea samples is quite different, which may cause the difference of aroma quality:Tea samples of Shangmanjuelong contain more 2-Pentyl Furan, Linalool and Ocatanol. Tea samples of fine breeds demonstration farm contain more three major esters as (E)-Butanoic acid-3-hexenyl ester, cis-3-Hexenyl isovalerate and (Z)-Hexanoic acid-3-hexenyl ester. And tea samples of Meijiawu contain moreδ-Cadinene.