Comparative analysis of aroma compounds in Chinese traditional dry-rendered fat by HS/GC-IMS, SPME/GC-MS, and SPME/GC-O

Aroma compositions of Chinese traditional dry-rendered subcutaneous chicken, pork, beef, and sheep fats were analyzed by HS/GC–IMS, SPME/GC–MS, and SPME/GC–O. HS/GC–IMS and SPME/GC–MS identified 53 and 86 volatile compounds, respectively. SPME/GC–O analysis identified 35 odor-active compounds. The identifications included fatty aldehydes, ketones, alcohols, acids, and esters from lipid degradation and compounds from the Maillard reaction (e.g., trimethylpyrazine) or other sources (e.g., p-cresol from animal feeds). PCA analysis showed all three approaches could discriminate aroma compositions of the four fats. PCA loading plot suggested the aroma compounds related to each of the four fats and PLS-DA model analysis screened the marker compounds differentiating aroma of the four fats. Notably, SPME/GC–O approach behaved the best in differing aromas of the four fats by PCA. The HS/GC–IMS analysis was biased for the chicken fat and pork fat, and the SPME/GC–MS analysis was biased towards the beef fat and sheep fat in differing the aroma compositions by PCA. PLS–DA model analysis showed that for the SPME/GC–O analysis, the marker compounds (VIP scores >1) discriminating aromas of the four fats were hexanal, (E,E)-2,4-decadienal, 2-butanone, 4,5-epoxy-2(E)-decenal, γ-octalactone, γ-undecalactone, heptanal, octanal, 4-ethyloctanoic acid, etc.