摘要
利用自动化静态顶空-气相色谱-质谱联用技术对江西三大产烟区(赣州、抚州、吉安)共120个烤烟样品进行了检测,匹配并定量了54个挥发性化合物。采用遗传算法(GA,Genetic algorithm)选择其中18种代表性化合物,并利用主成分分析法(PCA,Principal component analysis)对这120个烤烟样品进行产地分类,发现这种模式识别(Pattern recognition)方法可将烤烟样品基本分为3类:A类主体为赣州样品(兴国、宁都除外);B类主体为抚州样品(外加兴国、宁都,宜黄、崇仁除外);C类主体为吉安样品(外加宜黄、崇仁)。随后,我们将样品按照2:1的比例分为验证集和预报集,用有监督模式识别方法径向基函数-神经网络(Radial basis function-neural network,RBF-NN)对样品产地进行预报,正确率达到92.5%。据此,我们建立了用于江西烤烟样品的产地鉴别模型,该模型可用于研究不同产地间挥发性化合物含量的分布规律,为划分不同产地烤烟香型风格提供依据,实现对烟叶的质量控制。
In this paper,54 volatile compounds in a total of 120 flue-cured tobacco samples from three major tobacco-producing areas(Ganzhou,Fuzhou and Ji'an) in Jiangxi province were identified and quantified by automatic static headspace-gas chromatography-mass spectrometry(HS-GC-MS).18 compounds were selected by genetic algorithm(GA) for performing principal component analysis(PCA) on these 120 flue-cured tobacco samples,and all the samples could be classified into three classes:Class A are samples from Ganzhou excluding Xingguo and Ningdu counties;Class B are samples from Fuzhou plus Xingguo and Ningdu counties,minis Yihuang and Chongren counties;Class C are samples from Ji'an in addition of Yihuang and Chongren counties.Subsequently,we divided the samples into a calibration set and a prediction set according to the ratio of 2:1,Radial basis function-neural network(RBF-NN) was used to predict the origin of samples in prediction set with a correct rate of 92.5%.Thus we established a model for the identification of the origin of the flue-cured tobacco samples in Jiangxi,with can be used to study the content regulation of volatile compounds and provide a basis for dividing the flavor pattern of flue-cured tobacco from different areas,which can achieve the quality control of tobacco.
引文
[1] 赖燕华,肖明礼,卢嘉健,等.基于烟丝动态顶空香气成分的卷烟品牌分析[J].中国烟草科学,2017,38(6):48-54.
[2] WANG H M,LIU H Y,GUO B,et al.A novel poly(NMA-co-DEA-co-EDMA) monolithic column as a sorbent for online solid-phaseextraction and its application in the determination of beta-sitosterol in plant oil samples[J].Food Chemistry,2019,278:594-600.
[3] JIANG H,HU X R,LI Y,et al.Large-pore ordered mesoporous carbon as solid-phase microextraction coating for analysis of polycyclic aromatic hydrocarbons from aqueous media[J].Talanta,2019,195:647-654.
[4] BAJER T,SURMOVA S,EISNER A,et al.Use of simultaneous distillation-extraction,supercritical fluid extraction and solid-phase microextraction for characterisation of the volatile profile of Dipteryx odorata(Aubl.) Willd[J].Industrial Crops and Products.2018,119:313-321.
[5] DARAEE A,GHOREISHI S M,HEDAYATI A,et al.Supercritical CO2 extraction of chlorogenic acid from sunflower(Helianthus annuus) seed kernels:modeling and optimization by response surface methodology[J].Industrial Crops and Products,2019,144:19-27.
[6] 杨承清,雷凌华,夏更寿等.利用气质联用技术研究不同方法提取的黄心夜合鲜叶挥发油成分[J].南昌大学学报(理科版),2018,42(1):51-58.
[7] TSACHAKI M,LINFORTH R S T,TAYLOR A J.Dynamic headspace analysis of the release of volatile organic compounds from ethanolic systems by direct APCI-MS[J].Journal of Agricultural and Food Chemistry,2005,53(21):8328-8333.
[8] BICCHI C,CORDERO C,LIBERTO E,et al.Automated headspace solid-phase dynamic extraction to analyse the volatile fraction of food matrices[J].Journal of Chromatography A,2004,1024(1-2):217-226.
[9] CAMELEYRE M,LYTRA G,BARBE J C.Static Headspace analysis using low-pressure gas chromatography and mass spectrometry,application to determining multiple partition coefficients:a practical tool for understanding red wine fruity volatile perception and the sensory impact of higher alcohols[J].Analytical Chemistry,2018,90(18):10812-10818.
[10] VARONA-TORRES E,CARLTON D D,HILDENBRAND Z L,et al.Matrix-effect-free determination of BTEX in variable soil compositions using room temperature ionic liquid co-solvents in static headspace gas chromatography mass spectrometry[J].Analytica Chimica Acta.2018,1021:41-50.
[11] 江勇,袁美兰,赵利等.啤酒中挥发性成分的测定及产地鉴别[J].南昌大学学报(理科版),2014,38(5):472-476.
[12] 周秀锦,杨会成,邵宏宏等.HS/GC-MS/MS法检测海产品中二甲苯[J].分析测试学报,2018,37(7):844-848.
[13] JIA W,WANG H,SHI L,et al.High-throughput foodomics strategy for screening flavor components in dairy products using multiple mass spectrometry[J].Food Chemistry,2019,279:1-11.
[14] CRUSE C A,GOODPASTER J V.Generating highly specific spectra and identifying thermal decomposition products via Gas Chromatography / Vacuum Ultraviolet Spectroscopy(GC/VUV):Application to nitrate ester explosives[J].Talanta,2019,195:580-586.
[15] 倪永年.化学计量学在分析化学中的应用[M].北京:科学出版社,2004:2-3.
[16] YC/T 31-1996 烟草及烟草制品试样的制备和水分测定烘箱法[S].
[17] 谢剑平.烟草与烟气化学成分[M].北京:化学工业出版社,2010.
[18] NI Y N,SONG R M,KOKOT S.Discrimination of Radix Isatidis and Rhizoma et Radix Baphicacanthis Cusia samples by near infrared spectroscopy with the aid of chemometrics[J].Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy,2012,96:252-258.
[19] NEKOEI M.Genetic algorithm based wavelengths selection coupled with partial least squares for simultaneous spectrophotometric determination of phosphate and silicate in detergent products[J].Current Analytical Chemistry,2018,14(2):151-158.
[20] 杜一平,潘铁英,张玉兰.化学计量学应用[M].北京:化学工业出版社,2008.