Correlation analysis of the age of brandy and volatiles in brandy by gas chromatography-mass spectrometry and gas chromatography-ion mobility spectrometry

Abstract In this study, brandy samples of different ages (from 2 to 13 years old) collected directly from the barrel were analyzed by gas chromatography-mass spectrometry (GC-MS) and gas chromatography-ion mobility spectrometry (GC-IMS) to study the correlations between brandy age and its volatiles. Partial least squares regression (PLSR) analysis was carried out with the detection results of the two techniques, and the quality of the model established by GC-IMS results was superior to that by GC-MS. Moreover, we found that a large number of esters and alcohols played an important role in the two models. As to the same compounds—isoamyl acetate, ethyl octanoate, and 1-pentanol—detected by both techniques, GC-IMS results showed a better correlation with the age of brandy. The above phenomena further confirmed the high predictive ability of the model established by GC-IMS results for the age of brandy. After further optimization, the model based on the volatiles with VIP >1 achieved good fit and predictability, with the prediction ability parameter (Q2), the goodness of fit value (R2), the root mean square error estimation value (RMSEE) and the root mean square error estimated using the cross-validation (RMSECV) values of 0.957, 0.984, 0.490 and 0.673, respectively. GC-IMS combined with PLSR analysis may serve as an effective method for brandy age identification in industrial applications.