Correlation Between Sensory Descriptive Analysis and Volatile Composition of Beer Using Multivariate Analysis: The Effect of the Nonvolatile Matrix on the Sensory Perception and Volatile Fraction Behavior

The effect of carbohydrate and protein content on the volatile compound release and sensory perception of four selected volatile compounds (isoamyl acetate [banana], ethyl hexanoate [apple], myrcene [dryhop], benzaldehyde [almond]) from a model beer solution was investigated. Sensory attributes were evaluated using a trained sensory panel, while the concentration of volatile compounds were determined by 1) solid phase microextraction (SPME), 2) solid phase dynamic extraction (SPDE) coupled to gas chromatography mass spectrometry (GC/MS), and 3) stir bar sorptive extraction (SBSE) coupled to gas chromatography flame ionization (GC/FID). In addition, sensory and analytical results were correlated by multivariate analysis. Model beer samples with low levels of proteins and high levels of carbohydrates possessed the highest concentrations of isoamyl acetate, benzaldehyde, and ethyl hexanoate. Myrcene was only influenced by carbohydrate content, with a higher concentration recovered at a lower level of carbohydrate. From sensory analysis, apple and banana aroma attributes were perceived as less intense at the low protein content (P <= 0.05). Dry-hop aroma was perceived as more intense in samples with low carbohydrate content (P <= 0.05). Multivariate analysis showed a weak correlation between the sensory and instrumental data. The results illustrate that sensory perception and volatile fraction behaviors are influenced by the nonvolatile components and also demonstrate the need for caution when correlating instrumental to sensory analysis.