Abstract: A rapid method to detect mutton adulterated with duck by electronic nose combined with gas chromatgraphy-mass spectrometry (GC-MS) was proposed and validated by applying it on adulterated samples of Ningxia small-tailed han sheep meat with different amounts of cooked meat added. The electronic nose responses to the volatile components of meat samples were used to establish qalitative and quantitative models through Fisher linear discriminant analysis (FLDA) and linear regression analysis. Meanwhile, solid phase microextraction coupled with GC-MS was used to detect the volatile compounds. The results showed that different proportions of adulteration could be distinguished by FLDA with an accuracy rate of 98.2%. The correlation coefficient from linear regression analysis was as high as 0.965. A total of 26 volatile compounds were detected by GC-MS, including 3-cyclohepten-1-one and 3-methyl butyraldehyde, which played a significant role in the discrimination of different adulterated samples. The partial least squares regression models established based on the electronic nose and GC-MS data exhibited a determination coefficient of greater than 0.950. There was a high correlation between the response values of sensors S1, S3, S5, S6, S8 and S9 and the GC-MS data.

Key words: mutton, duck, electronic nose, gas chromatography-mass spectrometry (GC-MS), multivariate analysis method