Headspace gas chromatography-ion mobility spectrometry (HS-GC-IMS) and similarity analysis were used to evaluate the differences in the volatile flavor compounds of five different colored unpolished rices (red, yellow, green, purple and black) from Yangxian county, Hanzhong city after cooking. The results showed that a total of 61 volatile flavor compounds were identified, including 35 aldehydes (49.83%–57.06%), 13 ketones (accounting for 34.40%–41.45%), 5 alcohols (accounting for 1.42%–1.96%), 2 pyrazines (accounting for 0.02%–0.07%), 2 acids (accounting for 0.19%–0.49%), 1 ester (accounting for 0.08%–0.67%), 1 furan (accounting for 5.61%–8.23%), 1 ether (accounting for 0.02%–0.10%), and 1 phenol (accounting for 0.04%–0.22%). The content of aldehydes in cooked unpolished red rice was relatively higher, the content of acids in cooked unpolished yellow rice was relatively higher; the contents of alcohols and ethers in cooked unpolished green rice were relatively higher; the content of furans in cooked unpolished purple rice was relatively higher; and the contents of ketones, esters, pyrazines and phenols in cooked unpolished black rice were relatively higher. Principal component analysis (PCA) showed that the cumulative contribution rate of the first two principal components was 74.1%, which could better explain the characteristics of the original sample data. HS-GC-IMS spectral data could be used to distinguish the volatile flavor components of cooked unpolished rices of different colors. A volatile component fingerprint of cooked unpolished rice of different colors from Yangxian county was established in this study, which can visually presents the contour information of volatile flavor components and enrich the information about the eating quality of the five colored rices from Yangxian.

The volatile components in giant salamander oil adulterated with different amounts of peanut oil (0%, 5%, 10%, 20%, 30%, and 100%) were studied by gas chromatography-ion mobility spectroscopy (GC-IMS) combined with chemometrics. The results showed that a total of 41 volatile compounds were identified in all samples, including 21 aldehydes, 6 ketones, 4 alcohols, 4 heterocyclic compounds, 3 esters, 2 sulfur-containing compounds and 1 acid. With increasing adulteration level, the contents of aldehydes, heterocycles, acids and esters increased, while the contents of ketones, alcohols and sulfur compounds decreased. A partial least squares regression (PLSR) model between volatile components and adulteration level was established. The correlation coefficient (R²) values for the calibration and verification sets were 0.9924 and 0.9882, respectively, indicating that the reliability of the model. Principal component analysis (PCA) showed that the cumulative contribution rate of the first two principal components (PC) was 94.3%, indicating that the different adulteration levels could be well distinguished by volatile components. Thirteen differential volatile compounds with variable importance for the projection (VIP) scores greater than one, including seven aldehydes, three ketones, one alcohol, one sulfur compound and one ester, were selected by partial least squares-discriminant analysis (PLS-DA). PCA and cluster analysis showed that these differential volatile components could also be used to distinguish the different adulterated salamander oil samples. This study can provide technical support for the nondestructive rapid identification of adulterated giant salamander oil.