Volatile compounds of 7 commercial fermentation starters, including 4 traditional , 2, and 1 crude amylolytic enzyme, were extracted using solvent-assisted flavor evaporation (SAFE), and extracts were analyzed using GC-MS. A total of 70 volatile compounds, including 4 esters, 10 acids, 16 alcohols, 16 hydrocarbons, 9 ketones, 5 aldehydes, 3 volatiles phenols, and 7 miscellaneous components were identified. Concentrations of volatile compounds, including 4,6-dimethylundecane, methyl 2-hydroxybenzoate, 1-hexanol, 2-phenylethanol, 3-methylbutanoic acid, and hexanoic acid were higher in 4 traditional starters than in others. The major volatile components in 2 starters were the hydrocarbones dodecane, heptadecane, octadecane, and nonadecane. Concentrations of 2-methoxy-4-vinylphenol and 3-hydroxy-2-butanone were highest in the crude amylolytic enzyme starter. Compositions and amounts of volatile compounds were different among the 7 commercial fermentation starters.

In this study, four types of Takju (rice wine) were developed using different fermentation starters; two traditional nuruk (SS and SJ), one ipguk (JE), and one crude amylolytic enzyme (HK), respectively. Sample rice wines were analyzed for ethanol, titratable acidity, pH, soluable solids, and Hunter colorimeter values. The Takju made with HK showed the highest ethanol level, while the titratable acidity of Takju made with JE was twice that of the other samples. The volatile compounds in the four types of Takju were isolated by headspace-solid phase microextraction (HS-SPME) and analyzed by gas chromatography-mass spectrometry (GC-MS). A total of 49 volatile components were identified, including 27 esters, 3 acids, 10 alcohols, and 9 miscellaneous compounds. Ethyl esters and alcohols such as ethyl hexanoate, ethyl octanoate, ethyl decanoate, 2-methyl-1-pentanol, 1-pentanol, isoamyl alcohol, and 2-phenyl ethanol were the largest groups among the quantified volatiles. Although the numbers of detected volatile compounds were similar in the four types of Takju, the amount of quantified volatiles showed a great difference among samples.