透過您的圖書館登入
IP:34.200.248.66
  • 學位論文

台灣特色茶感官特性與電子舌及電子鼻分析之相關性

Relationship between Sensory Characteristics and Electronic Tongue / Electronic Nose Analyses of Taiwan Special Tea

指導教授 : 區少梅

摘要


在台灣有許多的以茶知名的產地且茶葉種類繁多並各具風味,依發酵程度可分為不發酵茶、部份發酵茶、全發酵茶,ㄧ般常用感官品評的方法評估不同程度發酵茶之感官特性及品質。在品評員在感官品評上面對大量受測樣品時,易產生疲倦感之困擾,且訓練品評員較為耗時。自從電子鼻與電子舌相繼問世後,期待利用這些仿生科技儀器取代人之感官品評,可克服人在感官品評上易之困擾,亦可提高再現性及靈敏度。因此本研究目的以電子舌與電子鼻測定台灣特色茶 (綠茶、松柏長青、凍頂烏龍茶及紅茶),觀察電子舌與電子鼻區分不同特色茶之區分能力,並將其結果與感官品評與物化分析結果進行相關性分析,以評估電子舌與電子鼻取代人的感官品評之可行性。 由物化分析結果得知,綠茶之總兒茶素含量、總多元酚含量、總游離胺基酸含量、個別兒茶素含量都為最高;紅茶之pH值最低、茶湯水色最暗,咖啡因含量為最高;屬於部份發酵茶之凍頂烏龍茶與松柏長青茶所有物化學成分都較無明顯差異性。 探討台灣特色茶之定量描述分析感官特性,結果顯示,綠茶之乾果香、熟果香、甜香及焙炒香較弱;而紅茶之紅色、酸味較強,松柏長青及凍頂烏龍茶之回甘感、回香感、生津感、順口感、甘味及花香較強。 電子舌感測分析結果由主成分分析 (Principal Component Analysis, PCA) 後,電子舌可區分四種特色茶解釋能力達95%,綠茶、松柏長青茶、凍頂烏龍茶與紅茶可以明顯被區分開。同樣地,將電子鼻感測分析結果由主成分分析後,區分四種特色茶解釋能力為97%,綠茶與紅茶可以明顯被區分出來,其餘松柏長青茶、凍頂烏龍茶香氣較為相似,因此有部份重疊的情況。 以定量描述分析與物化分析進行最小平方迴歸分析 (Partiial Least Squares, PLS2) ,茶樣中總多元酚、總兒茶素、EGCG、EC含量與感官的苦澀味、收斂感呈正相關,而與生津感、回香感、順口感、甘味呈負相關;Hunter a、b與黃色、明亮度呈負相關。至於電子舌之數據與感官品評數據以PLS2分析之結果,BA sensor 與酸味以及BB、GA、JB sensors則與甘味、回香感、順口感、生津感及回甘感具正相關。將電子鼻與感官品評數據以PLS2分析,六個sensors (T70/2、PA2、SY/LG、P10/2、T30/1、SY/AA) 與甜香、焙炒香、熟果香、乾果香相關性較高,而P40/1、SY/gCT1、SY/gCT、SY/Gh與花香較有相關性。 最後,以線性迴歸分析探討感官特性與電子舌、電子鼻之相關性,電子舌確實對於感官特性之酸味、順口感、回甘感、收斂感具有當不錯的預測能力;電子鼻可預測感官特性之甜香、焙炒香、熟果香。由此可知,電子舌與電子鼻對於台灣特色茶之感官特性具有很好的預測能力。

並列摘要


Taiwan, one of most famous tea production area in the world, owns various kinds of tea and all have their unique aroma and taste. In general, depending on the degree of fermentation, all sorts of tea divide into three categories, unfermented, partially fermented and fully fermented. For examining the quality of tea, sensory evaluation is the most common method used. However, its weakness is often pointed out as the unconsistent panelists performance and time-consuming. Since electronic nose and electronic tongue are available in the market, it is reported that they not only overcome the problem from the sensory evaluation but also increase the reproductivity and sensitivity of quality results. Therefore, the purpose of this study is to examine the discrimination ablility of electronic nose and electronic tongue for selected Taiwan special tea including green tea, Song-Bo-Zhang-Qing tea (SBZQ), Dongding oolong tea (DDO), and black tea, as compared with the sensory evaluation and physicochemical analysis in order to see the feasibility of these instruments to replace the sensory evaluation. The results of physicochemical analysis of Taiwan special tea, green tea had the highest contents of total catechins, total polyphenols, total free amino acids, and individual catechins. The pH of black tea had the lowest pH and darkest infusion color while the physicochemical contents of the partially fermented teas, DDO and SBZQ, were not significantly different. To investigate the sensory attributes of Taiwan special tea by quantitatively descriptive analysis, the results showed, geeen tea was weaker in dry-fruity aroma, ripen-fruity aroma, sweet aroma, and roasted aroma. The black tea was stronger in redness and sourness while SBZQ and DDO both were stronger in sweet aftertaste, aroma aftertaste, salivated aftertaste, smoothtaste, sweetness, and floral aroma. Based on the results from the principal component analysis (PCA) of the electronic tongue analysis, the first two components could explain the variation in these four tea samples up to 95% and they were all well separated. However, the PCA results of electronic nose analysis showed the first two components could explain the sample variation up to 97%, but only green tea and black tea were discriminated well from others due to the aroma of SBZQ and DDO was too similar. In terms of relationship between sensory data and physicochemical analysis data by Partial Least Squares Regression (PLSR) analysis, total polyphenols, total catechins, EGCG, EC were highly correlated with bitterness, astringency, and astringency aftertaste. Hunter a and Hunter b were negative related with yellowness and brightness. Based on the results correlation analysis between sensory data and electronic tongue analysis by PLS2, BA sensor was highly correlated with sourness while BB, GA and JB sensors were positively correlated with sweetness, aroma aftertaste, salivated aftertaste and sweetness aftertaste. The results of relationship between sensory data and electronic nose analysis on PLS2, there were six sensors (T70/2, PA2, SY/LG, P10/2, T30/1 and SY/AA) closely correlated with sweet aroma, roasted aroma, ripen-fruity aroma and dry-fruity aroma. The other sensors (P40/1, SY/gCT1, SY/gCT and SY/Gh) were correlated with floral aroma. Based on the results of correlation analysis between sensory attributes and electronic tongue/electronic nose by Linear Regression Analysis, the electronic tongue could predict the sourness, smooth taste, sweet aftertaste, and salivated aftertaste and the electronic nose could predict the sweet aroma, roasted aroma and ripen-fruity aroma. They both have good predicting ability for the sensory quality of these selected Taiwan special tea samples.

參考文獻


24. 陳正輝。2000。台灣茶葉品質分級制度之探討。農業經營管理會訊,31:
35. 劉士綸、蔡永生、區少梅。2008。台灣高山茶之化學特性與等級之鑑別。
40. 蔡永生、劉士綸、區少梅。2008。台灣凍頂烏龍茶之化學特性與等級之
53. Ding, Z., Kuhr, S, & Engelhardt, U.H., 1992, Influence of catechins and thea-
54. Dutta, R., Hines, E.L., & Gardner, J.W., 2003, Tea quality prediction using a

被引用紀錄


許婷琬(2009)。利用電子鼻及電子舌辨別蜂蜜真假之研究〔碩士論文,中臺科技大學〕。華藝線上圖書館。https://doi.org/10.6822%2fCTUST.2009.00023
林哲宇(2016)。乳酸菌發酵紅茶茶湯化學成分組成與品質之評估〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342%2fNTU201603862
施景彬(2012)。上市公司導入IFRS個案實務分析:以X集團為例〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342%2fNTU.2012.10247
林伯昌(2010)。以電子舌探討台灣市售茶飲料之感官品質的研究〔碩士論文,中臺科技大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0099-1901201112022904
林伯昌(2010)。以電子舌探討台灣市售茶飲料之感官品質的研究〔碩士論文,中臺科技大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0099-1901201115490404

延伸閱讀