Sensory lexicon and aroma volatiles analysis of brewing malt

Craft beer's growth has increased the focus on brewing malt, a crucial ingredient impacting beer's color and flavor. While terms like "malty" and "worty" are used, they're insufficient to capture the complexity of malt's contribution. Base malts undergo a standard process, providing basic flavor, while specialty malts (High aroma, Biscuit, Caramel, Crystal, Chocolate, Coffee, Black, etc.) are further processed for specific color and flavor characteristics. Traditional quality control methods (visual inspection, smelling, chewing) are inadequate for precise flavor description, as some flavor compounds are only detectable after extraction (as in brewing). The Brewing Association highlighted malt flavor as a major knowledge gap in the industry, prompting the development of new tools like the American Society of Brewing Chemists' Hot Steep Method for malt flavor extraction. Malt flavor's complexity stems from its chemical composition, influenced by the malting process (moisture, temperature, airflow, pH). Volatile compounds arise from reactions like the Maillard reaction, caramelization, Strecker degradation, and lipid peroxidation, with Maillard Reaction Products (MRPs) significantly contributing to flavor and color. Specific compounds like pyrazines, furfural, and maltol contribute to aromas such as toasted, nutty, bread, and caramel. Previous research identified benzaldehyde, benzeneacetaldehyde, hexanol, and ethyl acetate as key volatile compounds during malting. This study aimed to develop a sensory lexicon and wheel for brewing malt, analyze aroma-active compounds, and explore the relationships between aroma compounds and flavor attributes.

The literature review covers existing methods for sensory evaluation of food and beverages, highlighting descriptive analysis (DA) as a widely used technique for quantitative description of sensory features. The study also mentions the increasing use of novel sensory profiling methods like check-all-that-apply (CATA) and rate-all-that-apply (RATA), which allow for sensory profiling using untrained panels or consumers. The importance of establishing appropriate sensory lexicons and sensory wheels for standardization and communication within the industry is also emphasized. Previous research on the sensory properties of brewing malt is lacking, underscoring the need for this study.

The study involved two sensory panels: an Industry Panel (8 assessors with extensive experience) and a Laboratory Panel (15 assessors with sensory analysis experience). **Lexicon and Sensory Wheel Development:** 22 malt samples representing diverse sensory attributes were used. The panels generated 214 initial terms, which were consolidated to 53 attributes and definitions for the lexicon (grouped into appearance, flavor, taste, and mouthfeel). 46 attributes were selected to create a three-circle sensory wheel (appearance, flavor, taste/mouthfeel). **RATA Analysis:** Six malt samples representing six categories were evaluated using RATA by the Laboratory Panel (15 attributes of flavor and 3 of taste). Principal Component Analysis (PCA) was used to analyze the data. **Aroma Compound Analysis:** 18 malt samples representing different colors and types were analyzed using headspace solid-phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS). Gas chromatography-olfactometry (GC-O) was used to identify aroma-active compounds. **Correlation Analysis:** Partial least squares (PLS) regression was used to analyze the correlation between the seven main flavor attributes (baking, smoky, nutty, fruity, caramel, grain, and green) and the identified aroma compounds in the 18 malt samples. **Statistical Analysis:** Analysis of variance (ANOVA), PCA, and PLS regression were performed using XLSTAT software. Panel performance was monitored using PanelCheck Software.

The study resulted in a novel sensory lexicon comprising 53 attributes and a sensory wheel showcasing 46 attributes for brewing malt. RATA analysis and PCA successfully discriminated between different malt types based on sensory characteristics. HS-SPME-GC-MS identified 76 volatile compounds in 18 malt samples, and GC-O analysis identified 34 aroma-active compounds. PLS regression revealed strong correlations (R²Y = 0.728, Q² = 0.620) between the seven main flavor attributes and the identified aroma compounds, demonstrating the chemical basis of the sensory perceptions. Specific findings include the identification of key aroma compounds associated with different malt types and flavors. For instance, furfural was found in all specialty malts and its concentration increased with malt color depth. Nitrogen-containing heterocyclic compounds like pyrazine and pyrrole contributed to the aroma of dark malt. Aldehydes, alcohols, ketones, and esters were linked to lipid oxidation. Compounds like geranylacetone (green), maltols (caramellic), and 2-methylbutyraldehyde (baked) were identified and linked to specific malt types.

This study successfully addressed the need for a comprehensive sensory description of brewing malt, providing a valuable tool for industry and research. The developed lexicon and sensory wheel offer a standardized vocabulary for communication and quality control. The strong correlations found between aroma compounds and sensory attributes using PLS regression provide insights into the chemical basis of malt flavor, paving the way for targeted manipulation of flavor profiles during malting and brewing. The findings support the use of the Hot Steep Method for malt flavor extraction and highlight the efficacy of RATA and PLS regression in sensory analysis.

This research provides the first comprehensive sensory lexicon and wheel for brewing malt, enabling standardized communication and quality control. The strong correlation between sensory attributes and aroma compounds emphasizes the value of this approach in guiding malt production and beer brewing. Future research could explore the impact of specific malting parameters on aroma compound profiles and sensory properties, focusing on optimization of individual flavor notes.

The study used commercial malt samples, limiting the control over malting parameters. The number of malt types and samples could be expanded in future research to enhance the lexicon's comprehensiveness. While PLS regression identified correlations, it cannot definitively prove causality between aroma compounds and sensory perceptions. Further research employing controlled malting experiments and larger sample sizes could validate and extend the findings.