Science-Based Evaluation and Visualization of Sake Flavors and Providing this Information to Consumers
Technical notes | 2023 | ShimadzuInstrumentation
Japanese sake offers a vast spectrum of flavor profiles driven by regional ingredients, brewing conditions, and microbial activity. Accurately characterizing these flavors in scientific terms supports consumer choice, strengthens product branding, and enhances quality control in sake production.
This study aimed to quantify the aroma and taste constituents of 11 sake products from three breweries in Hokkaido. By combining gas chromatography-mass spectrometry (GC/MS) for volatile aroma compounds and liquid chromatography-mass spectrometry (LC/MS) for hydrophilic taste metabolites, the research applied principal component analysis (PCA) and analysis of variance (ANOVA) to visualize flavor differences and construct a flavor map for consumer guidance.
Samples included 11 sake types brewed with varying rice polish ratios (40–70 %), two koji strains (yellow and white), and soft to medium-hard water. Aroma analysis involved headspace GC/MS with the Smart Aroma Database, analyzing 24 key volatiles after heating samples at 60 °C. Taste profiling targeted 151 metabolites (sugars, amino acids, organic and nucleic acids) by LC/MS. Multivariate data treatment employed PCA and ANOVA in eMSTAT Solution statistics software.
Aroma PCA revealed two principal components: PC1 (orthonasal vs retronasal notes) and PC2. High concentrations of ethyl hexanoate and isoamyl acetate defined the “ginjo” fruity notes of Junmai Daiginjo and Ginjo sake. Base notes such as isoamyl alcohol, isobutanol, and 2-phenylethanol were elevated in Yamahai and Junmai styles. ANOVA confirmed significant aroma differences (P < 0.05) among sake types.
Taste PCA separated samples by rice polish and koji type. Ginjo and Daiginjo products showed higher mono-, di-, and trisaccharides linked to sweetness, while Junmai and Yamahai contained elevated organic acids (succinic, malic, citric) contributing to body and umami. White koji–fermented Yamahai sake uniquely exhibited high citric acid. Amino acid profiles (e.g., glutamic acid, proline) correlated with depth of taste and richness.
A combined flavor map plotted taste depth (PC1 of taste PCA) against aromatic character (PC1 of aroma PCA), spanning from light-bodied, highly fragrant sakes to full-bodied, less fragrant types. This visualization demonstrated clear product positioning and regional terroir effects.
Future research will integrate texture and mouthfeel metrics, expand to real-time sensory monitoring, and leverage machine learning for predictive flavor modeling. Mobile apps could deliver interactive flavor maps to consumers, while adaptive quality control systems may use these analytics to ensure consistency across vintages.
This study established a robust workflow to dissect sake flavor through GC/MS and LC/MS coupled with multivariate analysis. The resulting flavor map provides a science-based tool for producers and consumers to navigate sake diversity, with adaptability for ongoing changes in raw materials and fermentation parameters.
GC/MSD, Software, LC/MS
IndustriesFood & Agriculture
ManufacturerShimadzu
Summary
Significance of the Topic
Japanese sake offers a vast spectrum of flavor profiles driven by regional ingredients, brewing conditions, and microbial activity. Accurately characterizing these flavors in scientific terms supports consumer choice, strengthens product branding, and enhances quality control in sake production.
Objectives and Overview of the Study
This study aimed to quantify the aroma and taste constituents of 11 sake products from three breweries in Hokkaido. By combining gas chromatography-mass spectrometry (GC/MS) for volatile aroma compounds and liquid chromatography-mass spectrometry (LC/MS) for hydrophilic taste metabolites, the research applied principal component analysis (PCA) and analysis of variance (ANOVA) to visualize flavor differences and construct a flavor map for consumer guidance.
Methodology and Instrumentation
Samples included 11 sake types brewed with varying rice polish ratios (40–70 %), two koji strains (yellow and white), and soft to medium-hard water. Aroma analysis involved headspace GC/MS with the Smart Aroma Database, analyzing 24 key volatiles after heating samples at 60 °C. Taste profiling targeted 151 metabolites (sugars, amino acids, organic and nucleic acids) by LC/MS. Multivariate data treatment employed PCA and ANOVA in eMSTAT Solution statistics software.
Instrumentation Used
- Gas chromatograph-mass spectrometer (GC/MS)
- Liquid chromatograph-mass spectrometer (LC/MS)
- Smart Aroma Database
- eMSTAT Solution statistical analysis software
Main Results and Discussion
Aroma PCA revealed two principal components: PC1 (orthonasal vs retronasal notes) and PC2. High concentrations of ethyl hexanoate and isoamyl acetate defined the “ginjo” fruity notes of Junmai Daiginjo and Ginjo sake. Base notes such as isoamyl alcohol, isobutanol, and 2-phenylethanol were elevated in Yamahai and Junmai styles. ANOVA confirmed significant aroma differences (P < 0.05) among sake types.
Taste PCA separated samples by rice polish and koji type. Ginjo and Daiginjo products showed higher mono-, di-, and trisaccharides linked to sweetness, while Junmai and Yamahai contained elevated organic acids (succinic, malic, citric) contributing to body and umami. White koji–fermented Yamahai sake uniquely exhibited high citric acid. Amino acid profiles (e.g., glutamic acid, proline) correlated with depth of taste and richness.
A combined flavor map plotted taste depth (PC1 of taste PCA) against aromatic character (PC1 of aroma PCA), spanning from light-bodied, highly fragrant sakes to full-bodied, less fragrant types. This visualization demonstrated clear product positioning and regional terroir effects.
Benefits and Practical Applications
- Objective flavor profiling enhances consumer education and product matching.
- Flavor mapping supports marketing strategies and regional branding.
- Analytical benchmarks aid quality control and recipe optimization.
- Data-driven insights facilitate collaboration between breweries and rice growers.
Future Trends and Applications
Future research will integrate texture and mouthfeel metrics, expand to real-time sensory monitoring, and leverage machine learning for predictive flavor modeling. Mobile apps could deliver interactive flavor maps to consumers, while adaptive quality control systems may use these analytics to ensure consistency across vintages.
Conclusion
This study established a robust workflow to dissect sake flavor through GC/MS and LC/MS coupled with multivariate analysis. The resulting flavor map provides a science-based tool for producers and consumers to navigate sake diversity, with adaptability for ongoing changes in raw materials and fermentation parameters.
References
- Fukushima Prefecture Sake Revitalization Strategy: Enhancing the brand power and name recognition of Fukushima as a sake brewing region (2021)
- National Research Institute of Brewing: Sake aromas and their sources (2011)
- Akita O. et al.: Aroma design of sake, Kagaku to Seibutsu, Vol. 29, No. 9, p. 585–592 (1991)
- Saegusa S. et al.: Analysis of the taste characteristics of sake brewed in Tokyo in 2008–2010, Bulletin of Tokyo Metropolitan Agriculture and Forestry Research Center, Vol. 8, p. 35–48 (2013)
- Hishinuma M. et al.: Organic acids of sake, Journal of the Brewing Society of Japan, Vol. 61, No. 12, p. 1092–1097 (1966)
Content was automatically generated from an orignal PDF document using AI and may contain inaccuracies.
Similar PDF
Investigation of Components that Affect Flavors and Visualizing Differences in Tastes
2022|Shimadzu|Applications
C190-E290 Technical Report Investigation of Components that Affect Flavors and Visualizing Differences in Tastes Keiko Matsumoto1 A b s tra c t: Sensory analysis of food is based on the five senses—taste, smell, touch, sight, and hearing. In recent years,…
Key words
fukurami, fukuramiwithout, withoutdisaccharide, disaccharidesensory, sensoryacid, acidyeast, yeastisobutanol, isobutanolflavor, flavorcytidine, cytidinecomponents, componentsglyoxylic, glyoxylicornithine, ornithinemalic, malicisobutyl, isobutylisocitric
METABOLOMICS: Applications for Food Safety and Quality Control
|Shimadzu|Brochures and specifications
METABOLOMICS: Applications for Food Safety and Quality Control Metabolomics is an array of techniques used to comprehensively detect and analyze various metabolites formed in vivo during biological activity. In the food industry, metabolomics is used to qualitatively and quantitatively analyze…
Key words
acid, acidshu, shumetabolites, metabolitesethyl, ethylkynurenine, kynureninefutsu, futsujunmai, junmaigeographical, geographicalcysteine, cysteinetryptophan, tryptophanaroma, aromadaiginjo, daiginjosake, sakeuric, uricmethionine
Investigating Food Quality Evaluation: Complete Analysis of Aroma Compounds and Metabolites in Food
2016|Shimadzu|Applications
LAAN-A-MS-E037 Application News M271 Gas Chromatography Mass Spectrometry Investigating Food Quality Evaluation: Complete Analysis of Aroma Compounds and Metabolites in Food No. There are a wide variety of methods of ensuring food quality evaluation, and which method is used depends…
Key words
acid, acidshu, shuethyl, ethylfutsu, futsudaiginjo, daiginjojunmai, junmaicompounds, compoundssakes, sakesaroma, aromaacetate, acetatetemperature, temperaturenews, newspressurization, pressurizationequalization, equalizationmode
Analyzing Flavor Scientifically - Analytical and Testing Instruments for Food Development
2018|Shimadzu|Others
C10G-E052 Analyzing Flavor Scientifically Analytical and Testing Instruments for Food Development World Map of Shimadzu Sales, Service, Manufacturing, and R&D Facilities Sales and Service Manufacturing R&D 2 Analyzing Flavor Scientifically At Shimadzu, we are using our technologies to support the…
Key words
texture, texturedeliciousness, deliciousnessfood, foododors, odorscomponents, componentsacids, acidsitems, itemsanalysis, analysisodor, odoramino, aminochromatograph, chromatographflavor, flavorbeer, beerparticle, particleacid