-
电子鼻是一种新兴的,能在短时间内分析、识别和检测复杂气味和大多数挥发性成分的智能感官仪器[1],具有重复性好、不需要复杂的样品预处理过程、不发生感官疲劳和检测结果客观可靠等特点.与色谱仪、光谱仪等普通的化学分析仪器不同,电子鼻得到的不是被测样品中某种或某几种成分的定性与定量结果,而是给予样品中挥发性成分的整体信息,也称“指纹”数据[2].基于电子鼻的特点及其方便快捷的优越性,近几年来,电子鼻技术被广泛应用在绿茶[3-10]、红茶[11-13]、乌龙茶[14]和花茶[15]等茶叶香气成分的研究领域中,显示出很好的应用前景.
普洱熟茶是以云南大叶茶的晒青毛茶为原料,在高温高湿环境中以及微生物的参与下,制成的具有独特风味品质特征的后发酵茶[16-17],茶叶香气是决定茶叶品质最重要的因素之一,而“陈香”是普洱熟茶香气品质最主要的特征,也是决定其销售价格的一个最重要因子.在适当储藏条件下,普洱熟茶随着储存年份的增加,茶叶内含物质转化就会越充分,“陈香”也更加纯正.根据前人对普洱茶的研究发现,其独特的香气是在渥堆工序中形成的,而普洱茶内含物质成分则是在一定的湿热、酶促、微生物作用下形成的[18-20].但是,目前对于普洱茶“陈香”与年份之间的推断大多基于感官审评的经验性来判断,仍缺乏科学研究的理论基础.因此,验证电子鼻技术在不同年份普洱熟茶香气判别方面的可行性迫在眉睫.基于此,本文选取同一厂家,且茶叶等级、加工、仓储等条件一致的不同年份(2006-2015年)的普洱熟茶标准样为研究材料,利用德国Airsense公司的PEN3型便携式电子鼻和电子鼻Winmuster软件自带的模型识别方法进行主成分分析(Principal Component Analysis,PCA)、线性判别分析(Linear discriminant analysis,LDA)和传感器区分贡献率分析(Loadings)对干茶、茶汤和叶底香气进行分析和识别,以期为普洱熟茶香气的品质判别和年份鉴定确立实用的分析方法,同时验证电子鼻技术在普洱熟茶领域的可行性.
Discriminant Research for Identifying Aromas of Post-Fermented Pu-Erh Tea from Different Storage Years Using an Electronic Nose
-
摘要: 采用电子鼻技术对10个不同贮藏年份(2006-2015年)普洱熟茶的香气进行了研究.通过电子鼻Winmuster软件自带的模型识别方法进行主成分分析(PCA)、线性判别分析(LDA)和传感器区分贡献率分析(Loadings),对不同年份普洱熟茶的干茶、茶汤和叶底的香气进行分析和识别.结果表明:10个传感器中,W1W(对硫化物敏感),W2W(对芳香成分、有机硫化物敏感)、W1S(对甲烷敏感)和W2S(对乙醇敏感)对干茶香气灵敏度较高;W1W,W2W,W1S,W5S(对氮氧化合物敏感)对茶汤和叶底香气灵敏度较高,对普洱熟茶香气测定的贡献率较大;PCA和LDA能很好地区分不同贮藏年份普洱熟茶干茶、茶汤和叶底的香气,且年份差别越大,分离效果越好.Loadings结果分析显示:传感器中W1S,W1W,W2W和W2S在普洱熟茶香气判别中起主要作用,表明普洱熟茶储藏期间其挥发性有机硫化物、甲烷、部分芳香型化合物和乙醇类物质可能发生了较大变化.研究结果表明,电子鼻技术在普洱茶香气的检测领域具有较强的可操作性,同时为普洱熟茶的香气品质判别以及年份鉴定提供了理论依据.Abstract: In a study reported in this paper, an electronic nose was used to evaluate the aromas of post-fermented pu-erh tea samples from 10 different storage years (from 2006 to 2015). The samples were analyzed with principal component analysis (PCA), linear discriminant analysis (LDA) and loadings analysis (Loadings) using the Winmuster software system. Of the 10 sensors used, W1W (sensitive to sulfides), W2W (sensitive to aromatic components and organic sulfides), W1S (sensitive to methane) and W2S (sensitive to ethanol) were the most sensitive for dry tea leaves, the sensitivity of W1W, W2W, W1S and W5S (sensitive to nitrogen oxides) to tea infusion and infused leaf aroma was higher, and their contribution rates for tea aromas of post-fermented pu-erh tea were found to be significant. PCA and LDA could well differentiate aromas between dry teas, tea infusions and infused leaves and were better able to differentiate between tea samples if their storage years were far apart. The results of loadings analysis showed that W1S, W1W, W2W and W2S played a substantial role in discriminating the aromas of post-fermented pu-erh tea and large changes in the contents of volatile organic sulfides, methane, some aromatic compounds, and ethanol material occurred during its storage. The results of this study showed that the electronic nose technology is effective in detecting the aroma components of pu-erh tea.
-
Key words:
- electronic nose /
- post fermented pu-erh tea /
- aroma /
- electrical conductivity .
-
表 1 传感器性能
阵列序号 传感器名称 性能描述 备注 1 W1C 芳香成分aromatic 甲苯10 (mL·m-3) 2 W5S 对氮氧化合物很敏感broadrange NO2 1 (mL·m-3) 3 W3C 对芳香成分敏感aromatic 苯10 (mL·m-3) 4 W6S 对氢气有选择性hydrogen H2 100 (mL·m-3) 5 W5C 对烷烃芳香成分敏感arom-aliph 丙烷1 (mL·m-3) 6 W1S 对甲烷敏感broad-methane CH4 100 (mL·m-3) 7 W1W 对硫化物敏感sulphur-organic H2S 1 (mL·m-3) 8 W2S 对乙醇敏感broad-alcohol CO 100 (mL·m-3) 9 W2W 对芳香成分、有机硫化物敏感sulph-chlor H2S 1 (mL·m-3) 10 W3S 对烷烃敏感methane-aliph CH4 10 (mL·m-3) 表 2 不同年份普洱熟茶干茶样电导率G/G0的值
年份 W1C W5S W3C W6S W5C W1S W1W W2S W2W W3S 2015 1.077±0.047ab 1.381±0.249ab 1.165±0.093ab 1.005±0.017ab 1.021±0.016ab 2.463±1.030a 3.269±1.295b 1.956±0.640a 2.050±0.424b 1.056±0.042b 2014 1.041±0.002c 1.225±0.020b 1.104±0.002b 0.984±0.006d 1.010±0.002bc 1.738±0.098b 2.551±0.095bc 1.510±0.057bc 1.799±0.040bc 1.008±0.021b 2013 1.037±0.004c 1.224±0.019b 1.101±0.016b 0.990±0.009bcd 1.010±0.002bc 1.592±0.058b 2.483±0.262bc 1.418±0.039bc 1.755±0.104bc 1.020±0.023e 2012 1.034±0.005c 1.245±0.020b 1.114±0.017b 0.986±0.003cd 1.008±0.001c 1.545±0.068b 2.558±0.117bc 1.392±0.053d 1.776±0.056bc 1.012±0.011b 2011 1.084±0.023a 1.514±0.240a 1.219±0.086a 1.008±0.014ab 1.024±0.008a 2.176±0.500ab 4.480±1.307a 1.840±0.300ab 2.429±0.372a 1.071±0.031a 2010 1.049±0.006bc 1.281±0.019b 1.128±0.008b 1.004±0.013abc 1.014±0.003abc 1.648±0.063b 2.736±0.045bc 1.476±0.034bc 1.850±0.013bc 1.044±0.037ab 2009 1.043±0.008c 1.215±0.025b 1.106±0.010b 0.997±0.006abcd 1.011±0.003bc 1.644±0.129b 2.134±0.104c 1.452±0.083bc 1.607±0.062c 1.029±0.014ab 2008 1.037±0.011c 1.186±0.025b 1.091±0.021b 0.996±0.001abcd 1.011±0.003bc 1.609±0.199b 1.922±0.190c 1.417±0.123bc 1.549±0.116c 1.020±0.008b 2007 1.037±0.004c 1.183±0.111b 1.110±0.021b 1.007±0.006ab 1.009±0.001bc 1.509±0.024b 1.902±0.063c 1.365±0.009c 1.525±0.040c 1.029±0.017ab 2006 1.041±0.015c 1.169±0.029b 1.107±0.020b 1.014±0.011a 1.010±0.004bc 1.859±0.219b 1.859±0.092c 1.431±0.120bc 1.556±0.093c 1.040±0.032ab 注:表中同列小写字母不同表示Duncan's新复极差测验SSR法在p<0.05水平下差异有统计学意义(n=3). 表 3 不同年份普洱熟茶茶汤电导率G/G0值
年份 W1C W5S W3C W6S W5C W1S W1W W2S W2W W3S 2015 1.034±0.002a 1.219±0.022bcd 1.082±0.015a 1.049±0.004a 1.019±0.001a 1.356±0.061a 1.895±0.113bc 1.106±0.032a 1.503±0.059bc 1.189±0.001a 2014 1.031±0.003ab 1.225±0.036bc 1.269±0.354a 1.041±0.005b 1.018±0.001ab 1.310±0.059a 1.852±0.160bcd 1.075±0.031ab 1.463±0.068bcde 1.181±0.005ab 2013 1.029±0.001bc 1.242±0.010ab 1.082±0.005a 1.040±0.002bc 1.017±0.001bc 1.285±0.020ab 1.965±0.036b 1.063±0.002bc 1.517±0.013b 1.177±0.004b 2012 1.027±0.001cd 1.278±0.030a 1.099±0.006a 1.048±0.001a 1.015±0.001de 1.163±0.020c 2.158±0.123a 1.048±0.009bcd 1.605±0.041a 1.169±0.006c 2011 1.025±0.003de 1.186±0.051cd 1.058±0.014a 1.039±0.003bc 1.016±0.001cde 1.185±0.097c 1.739±0.187cd 1.033±0.037cde 1.403±0.079de 1.160±0.006d 2010 1.025±0.002de 1.193±0.022bcd 1.056±0.010a 1.031±0.001e 1.015±0.001de 1.213±0.052bc 1.736±0.089cd 1.033±0.024cde 1.398±0.041de 1.154±0.004de 2009 1.025±0.001de 1.172±0.014d 1.059±0.007a 1.035±0.004cde 1.016±0.001cde 1.166±0.019c 1.652±0.050d 1.009±0.011de 1.370±0.019e 1.160±0.008d 2008 1.024±0.000de 1.174±0.012cd 1.060±0.008a 1.033±0.003de 1.015±0.001de 1.173±0.034c 1.735±0.073cd 1.008±0.009de 1.414±0.038cde 1.154±0.001de 2007 1.024±0.006de 1.189±0.014cd 1.067±0.007a 1.037±0.001bcd 1.016±0.000cd 1.163±0.020c 1.820±0.074bcd 1.017±0.008de 1.456±0.039bcde 1.148±0.003ef 2006 1.022±0.002e 1.182±0.024cd 1.075±0.007a 1.036±0.000cde 1.014±0.001e 1.143±0.041c 1.829±0.110bcd 1.006±0.016e 1.466±0.053bcd 1.144±0.005f 注:表中同列小写字母不同表示Duncan's新复极差测验SSR法在p<0.05水平下差异有统计学意义(n=3). 表 4 不同年份普洱熟茶茶叶底电导率G/G0的比值
年份 W1C W5S W3C W6S W5C W1S W1W W2S W2W W3S 2015 1.032±0.002a 1.232±0.004bc 1.071±0.005abc 1.048±0.004a 1.018±0.001a 1.286±0.027a 1.880±0.038cd 1.068±0.006b 1.480±0.020bc 1.193±0.004a 2014 1.032±0.001a 1.245±0.023bc 1.061±0.003bcd 1.044±0.007a 1.018±0.001ab 1.295±0.034a 1.867±0.092cd 1.068±0.014b 1.456±0.036cd 1.186±0.001ab 2013 1.032±0.002a 1.284±0.018a 1.075±0.007a 1.061±0.012a 1.018±0.001a 1.428±0.111a 2.024±0.060ab 1.108±0.036ab 1.516±0.029ab 1.180±0.009bc 2012 1.030±0.003ab 1.246±0.041bc 1.072±0.008ab 1.063±0.025a 1.016±0.002bc 1.346±0.210a 1.907±0.135bc 1.099±0.062ab 1.477±0.034bc 1.169±0.010de 2011 1.028±0.002b 1.210±0.003c 1.060±0.010cd 1.068±0.019a 1.015±0.002c 1.312±0.121a 1.770±0.028d 1.079±0.040ab 1.412±0.018d 1.166±0.009de 2010 1.030±0.002ab 1.247±0.030b 1.067±0.011abcd 1.063±0.003a 1.016±0.000bc 1.354±0.013a 1.896±0.119bcd 1.087±0.009ab 1.470±0.053bc 1.175±0.004cd 2009 1.029±0.002ab 1.210±0.010c 1.058±0.003d 1.074±0.024a 1.016±0.001bc 1.338±0.132a 1.766±0.020d 1.087±0.044ab 1.412±0.008d 1.171±0.002cde 2008 1.031±0.001ab 1.235±0.004bc 1.057±0.002d 1.061±0.034a 1.017±0.001abc 1.371±0.101a 1.821±0.022cd 1.119±0.022ab 1.416±0.006d 1.169±0.002de 2007 1.032±0.002a 1.284±0.003a 1.066±0.001abcd 1.050±0.016a 1.017±0.000abc 1.378±0.050a 2.024±0.010ab 1.135±0.025a 1.504±0.010abc 1.168±0.001de 2006 1.032±0.001a 1.307±0.013a 1.073±0.003ab 1.038±0.012a 1.017±0.002abc 1.358±0.070a 2.127±0.053a 1.137±0.023a 1.538±0.022a 1.161±0.003e 注:表中同列小写字母不同表示Duncan's新复极差测验SSR法在p<0.05水平下差异有统计学意义(n=3). -
[1] 程绍明, 王俊, 王永维, 等.基于电子鼻信号判别番茄苗机械损伤程度[J].农业工程学报, 2012, 28(15):102-106. doi: http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_nygcxb201215018 [2] 王俊, 胡桂仙, 于勇, 等.电子鼻与电子舌在食品检测中的应用研究进展[J].农业工程学报, 2004, 20(2):292-295. doi: http://cpfd.cnki.com.cn/Article/CPFDTOTAL-ZYPZ200710001012.htm [3] YU Hui-chun, WANG Jun, ZHANG Hong-mei, et al. Identification of Green Tea Grade Using Different Feature of Response Signal from E-Nose Sensors[J]. Sensors and Actuators:B. Chemical, 2008, 128(2):455-461. doi: 10.1016/j.snb.2007.07.048 [4] doi: https://www.deepdyve.com/lp/elsevier/identification-of-coumarin-enriched-japanese-green-teas-and-their-emkhCMBjbm YANG Zi-yin, DONG Fang, SHIMIZU K, et al. Identification of Coumarin-Enriched Japanese Green Teas and Their Particular Flavor Using Electronic Nose[J]. Journal of Food Engineering, 2008, 92(3):312-316. [5] 于慧春, 王俊, 张红梅, 等.龙井茶叶品质的电子鼻检测方法[J].农业机械学报, 2007, 38(7):103-106. doi: http://www.oalib.com/paper/4661507 [6] 刘远方, 李阳, 梁飞, 等.绿茶香气的电子鼻分析方法研究[J].食品科技, 2012, 37(1):58-62. doi: http://www.wanfangdata.com.cn/details/detail.do?_type=degree&id=Y1851911 [7] 甘芝霖, 刘远方, 杨阳, 等.基于电子鼻技术的信阳毛尖茶品质评价[J].食品工业科技, 2013, 34(2):54-57. doi: http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_spgykj201302010 [8] 史波林, 赵镭, 支瑞聪, 等.应用电子鼻判别西湖龙井茶香气品质[J].农业工程学报, 2011, 27(S2):302-306. doi: http://www.cqvip.com/QK/90712X/2011S2/1003578166.html [9] 薛大为, 杨春兰.基于电子鼻技术的黄山毛峰茶品质检测方法[J].湖北工程学院学报, 2014, 34(3):64-67. doi: http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xgxyxb201403013 [10] TUDU B, JANA A, METLA A, et al. Electronic nose for black tea quality evaluation by an incremental RBF network[J]. Sensors and Actuators:B. Chemical, 2009, 138(1):90-95. doi: 10.1016/j.snb.2009.02.025 [11] doi: https://www.researchgate.net/publication/222652330_Preemptive_identification_of_optimum_fermentation_time_for_black_tea_using_electronic_nose BHATTACHRYA N, BIPAN T, ARUN J, et al. Preemptive Identification of Optimum Fermentation Time for Black Tea Using Electronic Nose[J]. Sensors and Actuators:B. Chemical, 2007, 131(1):110-116. [12] 王帅, 李文举, 韦丽华, 等.基于电子鼻的有机正山小种红茶的检测[J].食品科技, 2015, 40(11):292-296. doi: https://www.wenkuxiazai.com/doc/8b81d0eb0912a216157929bf.html [13] 赵超艺, 王秋霜, 卓敏, 等.乌龙茶审评方法研究概述[J].广东农业科学, 2009, 36(12):46-48. doi: 10.3969/j.issn.1004-874X.2009.12.016 [14] 吴亮亮, 张丹丹, 叶小辉, 等.电子鼻在对名优茉莉花茶香气评价中的应用[J].福建茶叶, 2016, 38(6):5-6. doi: http://www.cnki.com.cn/Article/CJFDTotal-FJCA201606004.htm [15] 江昕田, 郭雅玲, 赖凌凌, 等.电子鼻技术在不同厂家特种茉莉花茶香气判别中的应用研究[J].食品安全质量检测学报, 2017, 8(12):4760-4765. doi: 10.3969/j.issn.2095-0381.2017.12.043 [16] LV Hai-peng, ZHANG Ying-jun, LIN Zhi, et al. Processing and Chemical Constituents of Pu-erh Tea:a Review[J]. Food Research International, 2013, 53(2):608-618. doi: 10.1016/j.foodres.2013.02.043 [17] 国家质量监督检验检疫总局, 国家标准化管理委员会. 地理标志产品普洱茶: GB/T22111-2008[S]. 北京: 中国标准出版社, 2008. [18] 赵龙飞, 周红杰, 安文杰.云南普洱茶保健功效的研究[J].食品研究与开发, 2005, 26(2):114-118. doi: http://www.cqvip.com/QK/85425A/201301/45005540.html [19] 陈颖, 周芳梅, 杨秀莲, 等.高压脉冲电场对云南普洱茶总灰分含量的影响研究[J].西南农业学报, 2015, 28(3):1278-1282. doi: http://cpfd.cnki.com.cn/Article/CPFDTOTAL-ZGHY200810002008.htm [20] 任洪涛, 周斌, 夏凯国, 等.不同发酵程度普洱茶香气成分的比较分析[J].食品研究与开发, 2011, 32(11):23-26. doi: 10.3969/j.issn.1005-6521.2011.11.007 [21] 王力, 林智, 吕海鹏, 等.茶叶香气影响因子的研究进展[J].食品科学, 2010, 31(15):293-298. doi: http://www.cykk.cbpt.cnki.net/EditorCN/WebPublication/index.aspx?mid=cykk [22] 吕世懂, 孟庆雄, 徐咏全, 等.普洱茶香气分析方法及香气活性物质研究进展[J].食品科学, 2014, 35(11):292-298. doi: 10.7506/spkx1002-6630-201411058 [23] 吕海鹏, 钟秋生, 施江, 等.普洱茶挥发性成分指纹图谱研究[J].茶叶科学, 2014, 34(1):71-78. doi: http://www.cqvip.com/QK/97861X/201401 [24] LV Hai-peng, ZHANG Ying-jun, LIN Zhi, et al. Aroma Characterization of Pu-erh Tea Using Headspace-Solid Phase Microextraction Combined with GC/MS and GC-Olfactometry[J]. Food Chemistry, 2012, 130(4):1074-1081. doi: 10.1016/j.foodchem.2011.07.135 [25] 张雪寒, 司辉清.腊梅精油不同部分的香气成分鉴定及差异分析[J].西南师范大学学报(自然科学版), 2016, 41(4):87-95. doi: http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xnsfdxxb201604017