气相色谱-质谱技术结合化学计量学对5种动物油进行判别分析  

Discriminant Analysis of Five Kinds of Animal Oils by Gas Chromatography-Mass Spectrometry Combined with Chemometrics

作　　者：王同珍 陈孝建[1,2] 安爱 邱思聪 王修伟 姜春宝[3] 陈鹏飞 曹维强[1,2] 

机构地区：[1]深圳中检联检测有限公司,广东深圳518052 [2]广东工业大学轻工化工学院,广东广州510006 [3]惠州出入境检验检疫局,广东惠州516006

出　　处：《分析测试学报》2016年第5期557-562,共6页Journal of Instrumental Analysis

摘　　要：测定了5种不同种类动物油(鸡油、牛油、鸭油、羊油和猪油)的脂肪酸组成和含量,探讨了利用动物油脂肪酸的指标对不同种类的动物油进行分类和判别的可能性。采用气相色谱-质谱联用技术(GC-MS)对5种动物油脂肪酸的组成和含量进行测定,利用SPSS 19.0统计软件进行主成分分析、聚类分析和判别分析。对5种动物油的脂肪酸进行统计、分析和对比,得出动物油的主要组成为C16∶0,C18∶0,C18∶1c-9,C18∶2c-9,12和C14∶0,这5种脂肪酸在鸡油、牛油、鸭油、羊油和猪油中的总含量分别为92.1%,93.5%,93.5%,90.6%和95.6%。5种脂肪酸的不饱和度均小于1。主成分分析降维得到3个主成分,利用主成分分析数据,依次进行聚类分析和判别分析,建立了3个动物油典则判别函数,相关系数均大于0.995,对于鸡油、牛油、鸭油、羊油和猪油的初始分类正确率和交叉验证正确率均为100%。Gas chromatography -mass spectrometry was used to determine the fatty acid compositions in five different animal oils, and chemometric methods （such as principal component analysis （PCA） , cluster analysis （CA） and discriminant analysis （DA） ） were applied to the obtained data. The fatty acids in five animal oils samples were investigated, and the results showed that the main compositions of animal oils were C16 ： 0, C18 ： 0, C18 ： lc-9, C18 ： 2c-9, 12 and C14 ： 0. Total con- tents of the five major fatty acids were 92. 1% , 93.5% , 93.5% , 90.6% and 95.6% corresponding to the chanterelle, butter, duck fat, suet and lard, respectively. The ratios of unsaturated and sat- urated fatty acids（U/S） were all less than i for five animal oils. Three PCs were sequentially used for CA and DA. Three discriminant functions were established and the correlation coefficients were larger than 0. 995. The initial classification accuracy and cross-validation accuracy of chanterelle, butter, duck fat, suet and lard all reached up to 100%.

关 键 词：动物油 气相色谱-质谱联用仪(GC—MS) 主成分分析 聚类分析 判别分析 

分 类 号：O657.63[理学—分析化学] TQ645.3[理学—化学]