Short communication Volatile Compounds of Raw Spirits from Different Distilling Stages of Luzhouflavor Spirit

Transcription

1 Food Science and Technology Research, 20 (2), 283 _ 293, 2014 Copyright 2014, Japanese Society for Food Science and Technology doi: /fstr Short communication Volatile Compounds of Raw Spirits from Different Distilling Stages of Luzhouflavor Spirit Jia Zheng 1, Ru Liang 1, Jun Huang 1, Rui-Ping Zhou 4, Zhe-Jun Chen 4, Chong-De Wu 1, Rong-Qing Zhou 1,2,3* and Xue-Pin Liao 1,2 1 Key Laboratory of Leather Chemistry and Engineering, Ministry of Education, and College of Light Industry, Textile & Food Engineering, Sichuan University, Chengdu , China 2 National Engineering Laboratory for Clean Technology of Leather Manufacture, Sichuan University, Chengdu , China 3 National Engineering Research Center of Solid-State Brewing, Luzhou , China 4 Xufu distillery company, Yibin , China Received August 6, 2013 ; Accepted November 9, 2013 Distillation process is closely correlated with the quality and yield of the raw spirit. The volatile properties of raw spirits from three distilling stages (head, heart and tail) of Luzhou-flavor spirit were investigated based on GC and GC-MS analyses. A total of 71 compounds were identified and ester derivatives attributed the largest number and concentration of volatile compounds in three raw spirits. The total concentration of raw spirit was decreased from the head raw spirit (5105 mg/l) to the tail raw spirit (1843 mg/l). Many volatile compounds with high concentrations such as acetaldehyde, 1-hexanol, 2-methylbutanoic acid, hexanoic acid, and ethyl hexanoate were found to decrease during distillation. Odor active value (OAV) was used to evaluate the contribution of special compounds to the whole odor of the raw spirit. Ethyl hexanoate followed by ethyl butyrate, ethyl heptanoate, ethyl octanoate, and 3-methylbutanal were found to be the most potent odor-active compounds. The liquid-liquid extraction together with GC-MS and GC analyses could be a useful tool to characterize the volatile composition of different distilling cuts of Chinese spirit. The decreasing of total content of volatile compound was determined by the ester derivate because of its high volatility and content. Keywords: volatile compound, raw spirit, distilling process, odor active value Introduction The manufacturing procedures of spirits in the world such as whisky, brandy, rum, vodka, cachaça, and Chinese spirits are constituted of three consecutive processes: fermentation, distillation and ageing. For these processes, distillation process is closely correlated with the quality and yield of the raw spirit. Alembic stills and column stills are applied to the distillation of brandy, whisky, and cachaça. The former has no trays or appreciable reflux and requires multiple distillation to achieve high-proof raw spirits, while the latter has a reflux column to decreases the concentration of the congeners in raw spirits (Claus and Berglund, 2005). A distinct device called as Zeng-tong (Li et al., 2012) has been widely applied to the distillation of Chinese spirits for a long history. The Zeng-tong is composed of reboiler, taper barrel, condenser and other attachments, such as stream pipes, lid and tube type heat exchanger (shown in Figure 1A). Similar to fruit brandy (Claus and Berglund, 2005), the operation mode of Chinese spirits belongs to a batch distillation. The main production process of Chinese spirits is as follows (Figure 1B). Prior to the spirit distillation, the fresh grains and grain hull are mixed with the fermented grains to adjust *To whom correspondence should be addressed. zhourqing@scu.edu.cn

2 284 J. Zheng et al. Fig. 1. (A) Diagram of the distillation system (Zeng-tong) for Luzhou-flavor spirit. (a): reboiler-equipped with a secondary stream distributor; (b): taper barrel-filled with fermented grains; (c): lid; (d): stream pipes; (e): tube type heat exchanger. (B) Brief manufacturing procedure of Chinese spirits. their titer acidity, moisture and bulk density. Then the mixture is gradually filled into the barrel with the vapor spilling out from the surface of layer and forms a specific packed bed column. The distilling process begins after the mixture fully fills the Zeng-tong. After the distillation completed, the distilled mixture is cooled and mixed with Daqu-starter for fermentation in the pit. At last the fermented grains are sequentially treated with the way described above. Therefore, the distillation process involves not only the condensing of alcohol and fractionating of the congeners, but also the gelatinization and liquification of starch in grains. The quality of raw spirit is generally controlled by dividing the distillates into several appropriate fractions or cuts (i.e discarding the head fraction, collecting the heart fraction and redistilling/discarding the tail fraction) (Scanavini et al., 2010). However, the aromatic stability and consistence of the raw spirit completely depend on the operator s smell and taste. Till now, little information available concerning the effect of distillation process on the quality of raw spirit was reported. Chinese spirits can be classified into several fragrance types including strong aroma (Luzhou-flavor), light aroma (Fen-flavor), sauce aroma (Maotai-flavor), sweet honey and miscellaneous type according to their flavor characteristics and brewing technology. The volatile compositions of them have been previously studied, and more than seven hundreds components have been identified (Fan and Qian, 2006b; Zhu et al., 2007). Whereas, few attention was paid on the characteristics of volatile compounds in different distilling stages (head, heart and tail fraction). Furthermore, confirming the fingerprint of flavor compounds in these distillates may contribute to optimize the distillation process and increase the yield and distilling efficiency, and evaluate the influence of distillation process on the formation of harmful compounds. From the latest literatures, Rogelio et al. analyzed the role of batch rectification on the quality of Mexican tequila, and compared the copper and stainless steel alembics on the formation of volatile compounds (Rogelio et al., 2005). Bruno et al. investigated the influence of the configuration of the distilling system and procedure on the level of ethyl carbamate in cachaça (Bruno et al., 2007). Lukić et al. analyzed the change of concentrations of various volatile constitutions during the distillation process of fermented Muscat Blanc and Muškat ruža porečki grape marcs (Lukic et al., 2011). However, till now, little is known about the variation of the volatile compounds and their odor active values (OAV) in the

3 Volatile Compounds of Luzhou-flavor Raw Spirits 285 distillation process of Luzhou-flavor spirit. The aim of this study was to investigate the volatile compounds in raw spirits from three distilling stages, and analyze the influence of distillation on the quality of raw spirit. The OAV was used to evaluate the contribution of the volatile compounds to the whole odor. Methods and Materials Samples and chemicals Ethyl acetate (99.9%), ethyl lactate (99.0%), ethyl butyrate (99.5%), ethyl hexanoate (99.0%), ethyl palmitate (99.0%), ethyl linoeate (99.0%), ethyl phenylacetate (98.0%), ethyl oleate (98.0%), methyl caprylate (99.0%), phenethyl alcohol (99.0%), furfural (98.0%), benzenacetaldehyde (99.0%), 4-ethylphenol (97.0%), 4-ethyl-2-methoxylphenol (98.0%), acetic acid (99.5%), butanoic acid (99.0%), hexanoic acid (98.0%), caprylic acid (98.0%), methanol (99.8%) and acetaldehyde (99.5%) were all purchased from Sigma-Aldrich (St. Louis, MO, USA), and all standards used were of GC purity. Other reagents were of analytical purity. A total of three raw spirits belong to three distilling stages in the same batch were collected from Xufu distillery Co., Ltd., which is one of the typical Luzhou-flavor spirit manufacturing company in China. The distilling stages and related raw spirits were as follows: (1) head stage: about 1.5 kg raw spirit is collected from the pipe of tube type heat exchanger (e, Fig. 1) at the beginning of distillation, and this spirit is named as head spirit ; (2) heart stage: along with the distillation, the heart raw spirit is collect within 30 min with special operation parameters (the spirit temperature: more than 30, outflow of spirit: approximately 2.5 kg/min), and this part of spirit is named as heart spirit ; (3) tail stage: when the collection of heart spirit completed, the operators usually increase the steam reaching the maximum level to gelatinize and liquify the starch in grains and this stage will cost 45 min to 1 h, and the spirit collected at this stage is named as tail spirit. Before the analyses, a total of 500 ml samples of each distilling stage were putted into reagent bottles with grinding stopper, sealed and stored at room temperature until analysis. Extraction of volatile compounds The volatile compounds in raw spirits were extracted based on the liquid-liquid extraction (LLE) method reported previously (Qian and Reineccius, 2002) with some modifications. Briefly, 10 ml raw spirit was transferred into roundflask and internal standard (methyl caprylate and caprylic acid) was added. Prior to the extraction of volatile compounds, the concentration of ethanol in raw spirit was adjusted to approximately 14% by distilled water. The mixture was saturated with NaCl and adjusted to ph 11 with 10% NaOH. Then 50 ml anhydrous diethyl ether was added into the pretreated raw spirit mixture to extract volatile compounds. The organic phase was transferred into the clean glass tube and labeled as neutral fraction. The aqueous phase was adjusted to ph 1.7 with 2 M H 2 SO 4 and extracted by 50 ml anhydrous diethyl ether, and the extracted organic phase was labeled as acidic fraction. All fractions were dried with 5 g of anhydrous Na 2 SO 4, and concentrated the filtrate to 0.5 ml under the soft nitrogen. Each raw spirit was extracted and tested in three duplicate. GC-MS analysis The neutral and acidic fraction were analyzed on a Trace GC Ultra gas chromatograph-dsq ΙΙ mass spectrometer (Thermo Electron Corporation, Waltham, MA, USA) equipped with a HP-5MS capillary column (30.0 m 0.25 mm i.d., 0.25 μm film thickness, Agilent, Santa Clara, CA, USA), respectively. GC analyses were performed under the following conditions: an inlet temperature of 250, split ratio of 10:1, and Helium (purity: %) carrier gas flow of 1 ml/min. The oven temperature was kept at 40 for 5 min, followed by an increase of 5 / min to 200, and then programmed to 220 at 10 /min, and held for 5 min. For mass spectrometer, the temperatures of the transfer line, quadruple and ionization source were of 250, 150, and 230, respectively. The mass spectrum was generated in the electron impact (EI) mode at 70 ev. Detection was carried out in the full scan mode in the range of m/z 35 ~ 400. Each volatile compound was identified by comparison of their mass spectrum with the NIST05 spectrum database. Kováts retention indices (RI) of each compound were calculated by using C 8 ~C 20 n-alkanes (Sigma-Aldrich, St. Louis, MO, USA) (Cates and Meloan, 1963). The identification of each volatile compound was additionally confirmed by comparison of their RI with the RI reported in previous literatures (RIL). The relative concentration of volatile compound (mg/l) to the internal standards (methyl octanoate and octanoic acid) was semi-quantified on the basis of comparing their peak areas to that of the internal standard on GC total ion chromatograms. GC analysis The specific low molecular volatile compounds (methanol and acetaldehyde) were quantified by GC analysis. Concentrations of methanol and acetaldehyde were analyzed using GC-FID (Agilent 6890A, Santa Clara, CA, USA) equipped with DB-WAX (30.0 m 0.25 mm i.d., 0.25 μm film thickness, Agilent, Santa Clara, CA, USA). The GC analysis procedure was according to the procedure reported in Lukić et al. (2011). Statistical analysis The content of volatile compound was expressed as mg/l. Analysis of variance (ANOVA) with Turkey s test was performed to evaluate significant differences in volatile compounds from different distilling stages. Significance of difference was defined at p < 0.05 (n = 3). One-way ANOVA was conducted using SPSS software (version 17.0; SPSS Inc., Chicago, IL, USA). Results and Discussion Accuracy of quantitative analysis based on GC-MS The accuracy and reliability of final analyses results are tightly correlated with the sample preparation procedure. In some previous literatures, numerous sample preparation methods, such as LLE, solid-phase microextraction (SPME) and stir bar sorptive extraction

4 286 (SBSE), coupled with GC-MS were used to determine volatile compounds in Chinese spirits and wines (Hernandez-Gomez et al.; 2005, Sanchez-Palomo et al., 2009; Fan and Qian, 2006a). LLE represented a better applicability for sample preparation in previous experiments because all volatile compounds (low, medium and high volatility) can be analyzed in one or more extraction steps (Hernanz et al., 2008; Caldeira et al., 2007). Prior to GC-MS detection, the main volatile compounds determined in previous literature (Fan and Qian, 2006a) were used to validate the accuracy of the detection method in the present research. The peak area ratios (neutral fraction/methyl caprylate and acidic fraction/caprylic acid) were used for the quantification of each compound. As shown in Table 1, good linearity of calibration curve (r > 0.9) was obtained, and extraction recoveries were higher than 90% for all compounds tested. It was suggested that methyl caprylate and caprylic acid used as the internal standards was suitable to semiquantitatively or quantitatively evaluate the relative level of volatile compounds in distillates from different distilling stages. Volatile compositions of different raw spirits In this study, the volatile compounds from the head, heart and tail raw spirit were extracted by LLE method and then detected using GC-MS and GC analyses. Typical total ion chromatograms of neutral and acidic fraction in GC-MS system are displayed in Figure 2. Figure 3 represents the total concentration of the volatile compounds. It is clear that the total concentration of volatile compounds significantly decreased with the distillation ( p < 0.05). The highest concentration was detected in head raw spirit (5105 mg/l), which was 1.4- and 2.8-fold higher than that in the heart (3611 mg/l) and tail raw spirits (1843 mg/l), respectively. The levels of esters and other compounds were sharply decreased from the head to tail raw spirit, whereas high levels of acids, alcohols, aldehydes and ketones were observed in the heart raw spirit ( p < 0.05). J. Zheng et al. The compounds identified, quantified and grouped by their affiliation with different chemical classes are revealed in Table 2. A total of 71 compounds composed of 10 alcohols, 34 esters, 10 acids, 5 aldehydes, 4 ketones, 2 phenols, 3 hydrocarbon compounds and 3 oxygen-containing compounds were identified and quantified. The raw spirit exhibited the ethanol contents ranging from 44% in the tail spirit to 75% (v/v) in the head spirit. Besides the ethanol, esters which accounted for 14% _ 61% of the total concentration were the largest group among these volatiles. The most abundant compounds were methanol, 2-pentanol, 3-methylbutanol, 1-hexanol, propanoic acid, hexanoic acid, ethyl butyrate, ethyl pentanoate, ethyl hexanoate, ethyl heptanoate, ethyl octanoate, acetaldehyde, 3-methylbutanal, 1,1-diethoxyethane and 1,1-diethoxyl-3-methylbutane. Esters: Esters, the most numerous volatiles, are formed mostly through the esterification of alcohols with fatty acids during the fermentation, thermal distillation and maturing processes (Fan and Qian, 2005). A total of 34 esters including 24 ethyl esters and 10 other esters were quantified with the total concentration ranging from 260 mg/l to 3133 mg/l (Table 2). Esters especially ethyl esters represented the largest group in the number and concentration of volatile compounds identified. This result was coincided with the previous study (Fan and Qian, 2006a) which also reported that ethyl ester was the main group in esters. Ethyl hexanoate was considered to be the dominant compound in Luzhou-flavor spirit (Xu et al., 2010). In this study, ethyl hexanoate accounting for 36.7 _ 70.0% of the total esters was the most important ethyl ester and its concentration obviously decreased during the distillation process (P < 0.05). Meanwhile, the concentration of most esters was also significantly decreased from the head to tail raw spirit (P < 0.05), such as ethyl butyrate, pentanoate, hexanoate, nonanoate, palmitate, oleate and butyl Table 1. Characteristics of calibration curves. Standard material a Standard curve R 2 Validation rage (mg/l) Recovery (%) Detection limit (LOD, μg/l) d Ethyl butyrate y=0.0309x b _ Ethyl lactate y=0.093x b _ Ethyl hexanoate y=1.8196x b _ Ethyl phenylacetate y=0.0286x b _ Ethyl hexadecanoate y=0.0063x b _ Ethyl oleate y=0.095x b _ Ethyl phenylacetate y=0.0711x b _ Phenylacetaldehyde y=0.093x b _ Butanoic acid y=0.0789x c _ Hexanoic acid y= _ 0.001x c _ a Standard materials were all purchased from Sigma-Aldrich (St. Louis, MO, USA), and five levels of concentration for each volatile compound, covering the concentration ranges expected, were tested in triplicate. b Standard curves were fitted with the ratio of volatile compound to methyl caprylate. c Standard curves were fitted with the ratio of volatile compound to caprylic acid. d Detection limit of each compounds was calculated under the condition of the standard signal/noise of base line = 3.

5 Volatile Compounds of Luzhou-flavor Raw Spirits 287 Fig. 2. Total ion chromatogram of volatile compounds in (A) neutral fraction and (B) acidic fraction using GC-MS. Compounds series: internal standard (IS), alcohols (AL), aldehydes (AD), acids (AC), esters (ES), ketones (KT), phenols (PH), hydrocarbon compounds (HC) and oxygencontaining compounds (OC). Fig. 3. Total concentration and relative abundance of each kind of volatile compounds in the head, heart and tail raw spirits. Error bars indicated standard deviations (n = 3). Different letters indicate significant differences ( p < 0.05, ANOVA, Turkey s test). hexanoate. The similar pattern was also found in Muškat Ruža Porečki grape marcs and melon wine (Lukic et al., 2011; Hernandez-Gomez et al., 2005). Whereas ethyl lactate significantly decreased during the distilling process. It is possible that high flow rate of vapor in the initial stage of distillation process promote the volatilization of these esters with high boiling point. Traditionally, the content sequence of ethyl esters in commercial Luzhou-flavor spirit is ethyl hexanoate > acetate

6 288 J. Zheng et al. Table 2. Volume fraction of ethanol (v/v) and quantitative concentration (mg/l) of volatile compounds in different raw spirits. Number Compounds RI HP-5MS Identification 1 Threshold 2 (mg/l) Concentration 3 (mg/l) OAV Head Heart Tail Head Heart Tail Volume fraction of ethanol Alcohols (10) AL1 Methanol - Std 1368 a ± a ± a ±7 AL2 1-Propanol <800 MS, RIL ± ± ±0.4 <1 <1 <1 AL3 2-Methylpropanol <800 MS, RIL 16 (Qian and Wang, 2005) 0.53 a ± a ± b ±0.94 <1 <1 <1 AL4 2-Pentanol <800 MS, RIL 8.1 (Qian and Wang, 2005) 21.9 c ± b ± a ± <1 AL5 3-Methylbutanol <800 MS, RIL 30 (Guth, 1997) 117 b ±6 136 a ± c ± <1 AL6 1-Pentanol <800 MS, RIL 1.5 (Giri et al., 2010b) 0.10±0.04 nd 1.58±0.14 <1 - <1 AL7 1-Hexanol 875 MS, RIL 2.5 (Qian and Wang, 2005) 16.9 ± ± ± AL8 2-Furaldehyde diethyl acetal 1080 MS 0.07± ±0.01 nd AL9 1-Nonanol 1106 MS, RIL 1 (Qian and Wang, 2005) 1.44 b ± b ± a ± <1 AL10 Phenethyl alcohol 1116 MS, RIL, Std 1 (Qian and Wang, 2005) 1.80 a ± a ± b ± Acids (10) AC1 Acetic acid <800 MS, Std 200 (Guth, 1997) 1.73 ab ± b ± a ±0.02 <1 <1 <1 AC2 Propanoic acid <800 MS 8.1 (Ferreira et al., 2000) 44.2 b ± b ± a ± AC3 2-Methylpropanoic acid <800 MS 200 (Guth, 1997) 0.65± ± ±0.01 <1 <1 <1 AC4 Butanoic acid 871 MS,Std 1 (Qian and Wang, 2005) 8.03 b ± b ± a ± <1 AC5 3-Methylbutanoic acid 839 MS 3 (Guth, 1997) 5.65± ± ± AC6 2-Methylbutanoic acid 841 MS 3 (Guth, 1997) 19.9 ± ± ± AC7 Pentanoic acid 916 MS 4.58 a ± b ± c ±2.2 AC8 Hexanoic acid 1037 MS, Std 3 (Guth, 1997) 115 b ± c ± a ± AC9 Heptanoic acid 1115 MS 17.9 b ± b ± a ±1.19 AC10 Hexadecanoic cid 1963 MS, Std 5.22 b ± a ± a ±0.02 Esters (34) ES1 Ethyl acetate <800 MS, RIL, Std 7.5 (Guth, 1997) 13.0 b ± b ± a ± <1 ES2 Ethyl butyrate 806 MS, RIL, Std 0.02 (Guth, 1997) 123 c ± b ± a ± ES3 Ethyl lactate 823 MS, RIL, Std 14 (Ferreira et al., 2000) 30.3 b ± a ± a ± <1 <1 ES4 Ethyl 2-methylbutanoate 853 MS, RIL (Guth, 1997) 0.28 b ± c ± a ± ES5 Ethyl 3-methylbutanoate 857 MS, RIL (Guth, 1997) 0.81 b ± b ± a ± ES6 Ethyl pentanoate 904 MS, RIL (Takeoka et al., 1990) 90.8 b ± b ± a ± ES7 Ethyl 3-methylpentanoate 929 MS,RIL 1.07 b ±0.13 b 0.49 a ± a ±0.02 ES8 Ethyl 4-methylpentanoate 969 MS 1.82 b ± ab ± a ±0.04 ES9 Ethyl 2-methylpentanoate 972 MS, RIL 0.07 ± ±0.09 nd ES10 Ethyl hexanoate 1005 MS, RIL, Std (Guth, 1997) 2578 c ± b ±3 191 a ± ES11 Ethyl 2-hydroxy-4-methylpentanoate 1061 MS 3.07 b ± a ± a ±0.17

7 Volatile Compounds of Luzhou-flavor Raw Spirits 289 Table 2. Volume fraction of ethanol (v/v) and quantitative concentration (mg/l) of volatile compounds in different raw spirits. Number Compounds RI HP-5MS Identification 1 Threshold 2 (mg/l) Concentration 3 (mg/l) OAV Head Heart Tail Head Heart Tail ES12 Ethyl 5-methylhexanoate 1065 MS 0.63± ± ±0.14 ES13 Ethyl heptanoate 1100 MS, RIL (Giri et al., 2010b) 84.3 c ± b ± a ± ES14 Ethyl benzoate 1173 MS, RIL nd 0.15±0.01 nd ES15 Ethyl octanoate 1199 MS, RIL (Guth, 1997) 81.2 c ± b ± a ± ES16 Ethyl benzeneacetate 1248 MS, RIL, Std (Giri et al., 2010b) 0.45± ± ± ES17 Ethyl nonanoate 1297 MS, RIL 1.61 b ± b ± a ±0.02 ES18 Ethyl benzenpropanoate 1352 MS 1.18± ± ±0.26 ES19 Ethyl decanoate 1397 MS, RIL 0.2 (Tao and Zhang, 2010) 3.00 b ± b ± a ± ES20 Ethyl dodecanoate 1597 MS 2.12 b ± b ± a ±0.01 ES21 Ethyl tetradecanoate 1796 MS 4.19 b ± a ±0.25 a 1.04 a ±0.25 ES22 Ethyl palmitate 1996 MS, Std 2 (Qian and Wang, 2005) 32.8 b ± b ± a ±0.20 ES23 Ethyl linoleate >2000 MS, Std 17.1 c ± b ± a ±0.69 ES24 Ethyl oleate >2000 MS, Std 9.67 c ± b ± a ±0.48 ES25 Hexyl acetate 1018 MS, RIL 1.5 (Campo et al., 2006) 3.11 b ± a ± a ± <1 <1 ES26 3-Methylbutyl butanoate 1059 MS 1.71 b ± a ± a ±0.03 ES27 3-Methylbutyl 2-methoxyacetate 1072 MS 0.48 ± ± ±0.01 ES28 Propyl hexanoate 1096 MS, RIL nd 5.75 b ±0.81 b 0.67 a ±0.04 ES29 Butyl hexanoate 1134 MS 0.7 (Takeoka et al., 1990) 0.52 b ± b ± a ±0.01 <1 <1 <1 ES30 Isobutyl hexanoate 1153 MS 2.97 b ± b ± a ±0.04 ES31 Diethyl succinate 1184 MS 0.70 ± ± ±0.01 ES32 Hexyl butanoate 1193 MS 20.2 b ± b ± a ±0.41 ES33 Isopentyl hexanoate 1252 MS 1.52 a ± b ± a ±0.05 ES34 Pentyl hexanoate 1289 MS,RIL 1.97 b ± b ± a ±0.02 Ketones (4) KT1 2-Pentanone <800 MS 0.01 (Qian and Wang, 2005) nd nd 0.21± KT2 2-Butanone <800 MS 80 (Qian and Wang, 2005) 0.05 b ± a ± c ±0.00 <1 <1 <1 KT3 2-Heptanone 894 MS (Qian and Wang, 2005) 0.57 c ± b ± a ± KT4 4-Methyl-3-heptanone 934 MS 0.02±0.01 nd 0.03 ±0.02 Aldehydes (5) AD1 Acetaldehyde - Std b ± a ± a ± AD2 3-Methylbutanal <800 MS (Qian and Wang, 2005) 37.9 c ± b ± a ± AD3 1,1-Diethoxyethane <800 MS, RIL 0.05(Guth, 1997) 48.3 b ± b ± a ± AD4 Furfural 838 MS, RIL,Std 14.1(Ferreira et al., 2000) 4.06 b ± a ± a ±0.00 <1 <1 <1 AD5 Benzeneacetaldehyde 1044 MS, RIL 1(Campo et al., 2006) 0.38 a ± a ± b ±0.21 <1 <1 <1

8 290 J. Zheng et al. lactate (or lactate > acetate) > butanoate > pentanoate, and the Table 2. Volume fraction of ethanol (v/v) and quantitative concentration (mg/l) of volatile compounds in different raw spirits. Concentration 3 (mg/l) OAV Number Compounds RI HP-5MS Identification 1 Threshold 2 (mg/l) Head Heart Tail Head Heart Tail Phenols (2) PH1 4-Ethylphenol 1175 MS, RIL, Std 1.01(Giri et al., 2010b) 0.52 b ± a ± c ± PH2 4-Ethyl-2-methoxyphenol 1282 MS, RIL, Std 0.033(Ferreira et al., 2000) 1.27 a ± a ± b ± Hydrocarbon compounds (3) HC1 n-propylbenzene 954 MS, RIL 0.09± ±0.04 nd HC2 2-Methylnaphthalene 1295 MS, Std 0.30 ± ± ±0.07 HC3 1-Methylnaphthalene 1313 MS, Std 0.18 a ± b ± ab ±0.10 Oxygen-containing compound (3) OC1 1,1-Diethoxy-3-methylbutane 958 MS, RIL 47.8 c ± b ± a ±0.03 OC2 (2,2-Diethoxyethyl)benzene 1327 MS 1.02± ± ±0.09 OC3 Hexanoic acid, anhydride 1370 MS, RIL 0.47 b ± a ± a ±0.01 MS, compounds were identified by MS spectra; RIL, compounds were identified by a comparison with the retention index from the literatures; Std, compounds were identified by a comparison with the retention index from the pure standard. The reference from which the odor threshold had been taken was given in parentheses. The concentrations of volatile compounds were represented as mean value of triplicate samples ± standard deviates (mean ± SD). Different letters indicate significant differences ( p < 0.05, ANOVA, Turkey s test). nd means not determined appropriate content range of ethyl hexanoate in typical Luzhouflavor spirit range from 1200 mg/l to 2800 mg/l according to the China National Standard (GB , Strong flavor Chinese spirit). Result of this study showed that the highest level of ethyl hexanoate was monitored in the head spirit (2578 mg/l), and ethyl pentanoate, butanoate and palmitate also represent higher content in the head raw spirit than these in other samples. Furthermore, the tail spirit represented the lowest content of ethyl esters especially ethyl hexanoate (191 mg/l). In this way, only the heart raw spirit showed the most acceptable content of ethyl hexanoate (1453 mg/ L) because the blending process could increase the total content of ethyl hexanoate. Acids: Acids were the second largest group detected in this study. In the spirit samples, 10 acid derivates were identified (Table 2), and 9 of them were also indentified in previous studies (Fan and Qian, 2006a). The total concentration decreased from 239 mg/l in the heart raw spirit to 167 mg/l in the tail raw spirit. Based upon their concentration percentage, hexanoic acid was the major acid in all spirits (accounting for approximately 50% of total acids). A sharp decrease of the concentration was noted in propanoic acid and palmitic acid from the head to tail raw spirit ( p < 0.05), which was deviated with that obtained by Lukić (Lukic et al., 2011). Li et al., (2012) found that high flow rate of vapor could increase the volatility of hexanoic acid. Combined with the decreasing tendency of ethyl lactate, it assumed that the excessive high flow rate of vapor at the head stage lead to the acids streamed out too early. Alcohols: Alcohols are major products of fermentation of sugars and amino acid (Silva et al., 1996; Wondra and Berovic, 2001). Among 10 alcohols detected, methanol, 2-methylpropanol, 2-pentanol, 3-methylbutanol and 1-hexanol were considered to be the predominant alcohols on account of their high content. Approximately equal concentrations of methanol were detected from the head raw spirit to tail spirit, which showed a similar pattern in the Muškat ruža porečki grape marc (Lukic et al., 2011). It was reported that methanol is produced by the hydrolysis of pectic substances (Mangas et al., 1995). High level of methanol in raw spirits maybe explained by the intense liquefaction of grains especially the grain hull during the distilling process. The concentration of 2-pentanol significantly decreased from the head to tail spirit ( p < 0.05). High concentrations of 1-propanol, 3-methylbutanol and 1-hexanol were monitored in the head and heart raw spirits, and this was agreement with the previous finding (Leaute, 1989). Probably high volatility of these low boiling point compounds such as 1-hexanol and 1-propanol leaded to high content in the heart raw spirit. As it documented that 1-hexanol is considered to be a positive affection on the aroma of the distillate when the content of which up to 20 mg/l (Apostolopoulou et al., 2005). It suggested that the quality of the heart raw spirit may superior to others. The aromatic higher alcohol-phenethyl alcohol (8.70 mg/l) exhibited an increasing trend from the head to tail raw

9 Volatile Compounds of Luzhou-flavor Raw Spirits 291 spirit, which was in accordance with the result of Lukić and Apostolopoulou (Lukic et al., 2011; Apostolopoulou et al., 2005). Additionally, 2-methylpropanol also exhibited the highest concentration of 37.5 mg/l in the tail raw spirit. Aldehydes and ketones: Most aldehydes are probably metabolites of bacteria (Lachenmeier and Sohnius, 2008), and ketones are formed by the autoxidation of fatty acids, especially unsaturated fatty acids (Grosch, 1982). Furfural is formed during the distillation due to dehydration of fermentable sugars (pentoses) caused by heating in acid conditions and/or Maillard reaction (Mangas et al., 1995). A total of 4 ketones and 5 aldehydes were observed, and the highest levels of aldehyde and ketone were determined in the heart raw spirit, followed by the head and the tail raw spirit. However, the previous literature observed a significant decrease in concentration during the distillation (Apostolopoulou et al., 2005). It assumed that this was also owing to the high flow rate of vapor. Acetaldehyde, 3-methylbutanal and 1,1-diethoxyethane were considered to be dominant aldehydes in raw spirit upon their high concentration, and this agreed with these reported in previous studies (Fan and Qian, 2005; Kim et al., 2009). A significantly decreased pattern of acetaldehyde, 3-methylbutanal, 1,1-diethoxyethane and furfural was represent from the head to tail raw spirit ( p < 0.05), assumedly depending on their relatively low boiling points (103 _ 209 ) and the solubility in alcohol. Acetaldehyde is originated from the spontaneous or microbially mediated oxidation in distillates (Qian and Wang, 2005). In this study, the acetaldehyde in raw spirits may from the spirit fermentation process of raw material because the raw spirits were not stored before sampling. Phenols: Phenols especially 4-ethyl-2-methoxyphenol (4-ethylguaiacol) were the thermal degradation products of ligninrelated phenolic carboxylic acid (Zhang and Tao, 2009). The highest concentrations of 4-ethylphenol and 4-ethyl-2- methoxyphenol were monitored in the tail spirit, with the concentration of 1.92 mg/l and 6.51 mg/l, respectively. Similarly, these two phenols were also identified in Yanghe spirit (Fan and Qian, 2006b), Maotai spirit (Zhu et al., 2007) and whisky (Caldeira et al., 2007). High boiling point (217 and 236 ) could determine the concentration of them only emerging in the tail raw spirit because the last stage with higher flow rate and temperature comparing with other stages is always applied to gelatinize and liquefy the starch of grains. Moreover, a total of 6 volatile compounds were identified. N-propylbenzene, 2-methylnaphthalene, 1-methylnaphthalene belongs to hydrocarbon compound and (2,2-diethoxyethyl) benzene, 1,1-diethoxyl-3-methylbutane, and hexanoic acid, anhydride belongs to oxygen-containing compound. The decreased tendency was monitored in 1,1-diethoxyl-3-methylbutane because of its low boiling point (156 ) and which has also been identified in other Chinese Luzhou-flavor liquor (Fan and Qian, 2005), aronia spirit and sparkling wines (Bosch-Fusté et al., 2007; Balcerek, 2010). Due to their low concentration, the identification and classification of many compounds in these two groups (hydrocarbon and oxygen-containing compound) such as n-propylbenzene, 2-methylnaphthalene, and 1-methylnaphthalene in raw spirit are still required in the further analysis. OAV of volatiles in different raw spirits The quantification data of volatile compounds in raw spirits from three distilling stages is listed in Table 2. It is obvious that ethyl hexanoate had the highest OAV, and its value decreased from the highest in the head spirit (515766) to the lowest in the tail spirit (38286). Previous studies also reported that ethyl hexanoate was the key volatile compound in most Luzhou-flavor spirits, such as Yanghe (Fan and Qian, 2005), Wuliangye and Jiannanchun (Fan and Qian, 2006a) based on its flavor dilution (FD) value, which contributed to the fruity, anise, apple, sweetish and pear odor (Jiang and Zhang, 2010). It suggested that ethyl hexanoate played the most important role in the whole odor feature of the raw spirit. Most esters with high OAV detected had a typical fruit and floral descriptor, and contributed to fruity, sweet, apple, pineapple, and floral odors. For example, ethyl pentanoate (858 _ 18160), ethyl heptanoate (2650 _ 42160), and ethyl octanoate (3315 _ 40590) contributed to the odor of fruity, flora, sweet and pineapple (Fan and Qian, 2006a; Jiang and Zhang, 2010; Zhang et al., 2010). It was worth noting that 3-methylbutanal with low threshold value (0.35 μg/l) (Qian and Wang, 2005) may largely contributed to a characteristic aroma of apple, although the concentration was relatively low (3.18 mg/l-37.9 mg/l). Meanwhile, the important aroma compounds (OAV 10) included 1-hexanol, hexanoic acid, ethyl 2-methylbutanoate, ethyl 3-methylbutanoate, ethyl benzenacetate, ethyl decanoate, 2-pentanone, 2-heptanone, 1,1-diethxoyethane 4-ethylphenol and 4-ethyl-2-methoxylphenol. 1-Hexanol, with a floral and fruity odor (Jiang and Zhang, 2010), could be the dominant alcohol with the highest OAV. Hexanoic acid, one of main metabolites of Clostridium bacteria (Weimer and Stevenson, 2011), was also identified as the dominating volatile component in various raw spirits. Previously, it was reported that 1,1-diethxoyethane was an important aroma compound in Yanghe spirit with the FD value more than 256 (Fan and Qian, 2005). In this study, 1,1-diethxoyethane also showed high OAV, assuming that a tight relationship existed between the raw spirit and commercial spirit. 4-Ethylphenol and 4-ethyl-2-methoxyphenol were the odor active compounds in many oriental fermentation condiments such as soy sauce and soybean paste, which had the odor of cooked soybean, smokey and phenolic odours (Lee and Ahn, 2009; Giri et al., 2010b; Giri et al., 2010a). In addition, other volatile compounds such as propanoic acid, 2-methylbutanoic acid, ethyl acetate, ethyl butyrate, hexyl acetate and acetaldehyde with the OAV more than 1 may also contribute to the whole odor of the spirit. Ethyl acetate and ethyl butyrate represented the odor descriptions of green apple, strawberry, pineapple and sweet (Zhang et al., 2010; Vilanova et al., 2012).

10 292 Conclusion The volatile compounds of raw spirits collected from three distilling stages of Luzhou-flavor spirit were detected using GC- MS and GC analyses. Results showed that the total concentration of volatile compounds was obviously decreased from the head spirit to tail spirit ( p < 0.05) without any research purely based on the different boiling points or solubility of the compounds (e.g. esters have higher volatility than alcohols). Ethyl hexanoate with the highest concentration and OAV was the most important odor active compound in each spirit. On the basis of high OAV, ethyl butyrate, ethyl heptanoate, ethyl octanoate, and 3-methylbutanal were also elucidated to be main contributors to the overall flavor of raw spirits. Results presented in this article may benefit to understand the changes and odorant contributions of volatile compounds in the distillation process of Luzhou-flavor spirit. Acknowledgments This work was financially supported by the National Science Foundation of China ( ). References Apostolopoulou, A. A., Flouros, A. I., Demertzis, P. G., and Akrida- Demertzi, K. (2005). Differences in concentration of principal volatile constituents in traditional Greek distillates. Food Control, 16, Balcerek, M. (2010). Carbonyl compounds in aronia spirits. Polish J. Food Nutri. Sci., 60, Bosch-Fusté, J., Riu-Aumatell, M., Guadayol, J. M., Caixach, J., López- Tamames, E., and Buxaderas, S. (2007). Volatile profiles of sparkling wines obtained by three extraction methods and gas chromatography mass spectrometry (GC-MS) analysis. Food Chem., 105, Bruno, S. N. F., Vaitsman, D. S., Kunigami, C. N., and Brasil, M. G. (2007). Influence of the distillation processes from Riode Janeiro in the ethyl carbamate formation in Brazilian sugar cane spirits. Food Chem., 104, Caldeira, M., Rodrigues, F., Perestrelo, R., Marques, J. C., and Câmaraj.S. (2007). Comparison of two extraction methods for evaluation of volatile constituents patterns in commercial whiskeys elucidation of the main odour-active compounds. Talanta, 74, Campo, E., Ferreira, V., Escudero, A., Marqués, J. C., and Cacho, J. (2006). Quantitative gas chromatography olfactometry and chemical quantitative study of the aroma of four Madeira wines. Anal. Chim. Acta, 563, Claus, M. J. and Berglund, K. A. (2005). Fruit brandy production by batch column distillation with reflux. J. Food Process Eng., 28, Cates, V. and Meloan, C. (1963). Separation of sulfones by gas chromatography. J. Chromatogr. A, 11, Fan, W. L. and Qian, M. C. (2005). Headspace solid phase microextraction and gas chromatography-olfactometry dilution analysis of young and aged Chinese Yanghe Daqu liquors. J. Agric. Food Chem., 53, Fan, W. L. and Qian, M. C. (2006a). Characterization of aroma compounds of Chinese Wuliangye and Jiannanchun liquors by aroma extract J. Zheng et al. dilution analysis. J. Agric. Food Chem., 54, Fan, W. L. and Qian, M. C. (2006b). Identification of aroma compounds in Chinese Yanghe Daqu liquor by normal phase chromatography fractionation followed by gas chromatography/olfactometry. Flav. Frag. J., 21, Ferreira, V., Lpez, R., and Cacho, J. F. (2000). Quantitative determination of the odorants of young red wines from different grape varieties. J. Sci. Food Agri., 80, Giri, A., Osako, K., and Ohshima, T. (2010a). Identification and characterisation of headspace volatiles of fish miso, a Japanese fish meat based fermented paste, with special emphasis on effect of fish species and meat washing. Food Chem., 120, Giri, A., Osako, K., Okamoto, A., and Ohshima, T. (2010b). Olfactometric characterization of aroma active compounds in fermented fish paste in comparison with fish sauce, fermented soy paste and sauce products. Food Res. Inter., 43, Grosch, W. (1982). In Food flavors. Part A. Introduction, 1st ed.; Morton, I.D., Macleod, A.J., Eds.; Elsevier: Amsterdam, The Netherlands, pp Guth, H. (1997). Quantitation and sensory studies of character impact odorants of different white wine varieties. J. Agr. Food Chem., 45, Hernandez-Gomez, L. F., Ubeda-Iranzo, J., Garcia-Romero, E., and Briones-Perez, A. (2005). Comparative production of different melon distillates: Chemical and sensory analyses. Food Chem., 90, Hernanz, D., Gallo, V., Recamales, A. F., Melendez-Martinez, A. J., and Heredia, F. J. (2008). Comparison of the effectiveness of solid-phase and ultrasound-mediated liquid-liquid extractions to determine the volatile compounds of wine. Talanta, 76, Jiang, B. and Zhang, Z. (2010). Volatile compounds of young wines from cabernet sauvignon, cabernet gernischet and chardonnay varieties grown in the loess plateau region of china. Molecules, 15, Kim, J. S., Kam, S. F., and Chung, H. Y. (2009). Comparison of the volatile components in two Chinese wines, Moutai and Wuliangye. J. Korean Soc. Appl. Biol. Chem., 52, Lachenmeier, D. W., and Sohnius, E. M. (2008). The role of acetaldehyde outside ethanol metabolism in the carcinogenicity of alcoholic beverages: evidence from a large chemical survey. Food Chem. Toxicol., 46, Leaute, R. (1989). Distillation in Alambic. Am. J. Enol. Viticul., 41, Lee, S. J. and Ahn, B. (2009). Comparison of volatile components in fermented soybean pastes using simultaneous distillation and extraction (SDE) with sensory characterisation. Food Chem., 114, Li, H. L., Huang, W. X., Shen, C. H., and Yi, B. (2012). Optimization of the distillation process of Chinese liquor by comprehensive experimental investigation. Food Bioprod. Process., 90, Lukic, I., Tomas, S., Milicevic, B., Radeka, S., and Persuric, D. (2011). Behaviour of volatile compounds during traditional alembic distillation of fermented Muscat Blanc and Muškat Ruža Porečki grape marcs. J. Inst. Brew., 117, Mangas, J., Rodrıguez, R., Moreno, J., and Blanco, D. (1995). Changes in

11 Volatile Compounds of Luzhou-flavor Raw Spirits 293 the major volatile compounds of cider distillates during maturation. LWT - Food Sci. Technol., 29, Qian, M. and Reineccius, G. (2002). Identification of aroma compounds in Parmigiano-Reggiano cheese by gas chromatography/olfactometry. J. Dairy Sci., 85, Qian, M. C. and Wang, Y. Y. (2005). Seasonal variation of volatile composition and odor activity value of Marion (Rubus spp. hyb) and Thornless Evergreen (R. laciniatus L.) blackberries. J. Food Sci., 70, C13-C20. Rogelio, P. R., Victor, G. A., Carlos, P. O., Norberto, C., Mirna, E., and Héctor E, G. H. (2005). The role of distillation on the quality of tequila. Inter. J. Food Sci. Technol., 40, Sanchez-Palomo, E., Alanon, M. E., Diaz-Maroto, M. C., Gonzalez-Vinas, M. A., and Perez-Coello, M. S. (2009). Comparison of extraction methods for volatile compounds of Muscat grape juice. Talanta, 79, Scanavini, H. F. A., Cheriani,R., Cassini, C. E. B., Souza, E. L. R., Filho, F. M., and Meirelles, A. J. A. (2010). Cachça production in a lab-scale alembic modeling and computational simulation. J. Food Process Eng., 33, Silva, M., Malcata, F., and Revel, G. (1996). Volatile contents of grape Marcs in Portugal. J. Food Comp. Anal., 9, Takeoka, G. R., Flath, R. A., Man, T. R., Teranishi, R., and Guentert, M. (1990). Volatile Constituents of Apricot (Prunus armeniaca) J. Agr. Food Chem., 38, Tao, Y. S. and Zhang, L. (2010). Intensity prediction of typical aroma characters of cabernet sauvignon wine in Changli County (China). LWT - Food Sci. Technol., 43, Vilanova, M., Siebert, T. E., Varela, C., Pretorius, I. S., and Henschke, P. A. (2012). Effect of ammonium nitrogen supplementation of grape juice on wine volatiles and non-volatiles composition of the aromatic grape variety Albariño. Food Chem., 133, Weimer, P. J. and Stevenson, D. M. (2011). Isolation, characterization, and quantification of Clostridium kluyveri from the bovine rumen. Appl. Microbiol. Biotechnol., 94, Wondra, M. and Berovic, M. (2001). Analyses of aroma components of Chardonnay wine fermented by different yeast strains. Food Technol. Biotechnol., 39, Xu, Y., Wang, D., Fan, W. L., Mu, X. Q., and Chen, J. (2010). Traditional chinese biotechnology. Adv. Biochem. Eng. Biotechnol., 122, Zhang, Y., Li, X., Lo, C. K., and Guo, S. T. (2010). Characterization of the volatile substances and aroma components from traditional soypaste. Molecules, 15, Zhang, Y. F. and Tao, W. Y. (2009). Flavor and taste compounds analysis in Chinese solid fermented soy sauce. Afr. J. Biotechnol., 8, Zhu, S., Lu, X., Ji, K., Guo, K., Li, Y., Wu, C., and Xu, G. (2007). Characterization of flavor compounds in Chinese liquor Moutai by comprehensive two-dimensional gas chromatography/time-of-flight mass spectrometry. Anal. Chim. Acta., 597,