Identification and Aroma Impact of Volatile Terpenes in Moutai Liquor

Abstract

Terpenes are very important impact aroma compounds in wines, and we discovered some trace volatile terpenes in Chinese liquors in this study. The volatile compounds in Kweichow Moutai liquor (Maotai) were enriched and isolated by liquid–liquid extraction and further separated by silica gel normal phase chromatography, then investigated by gas chromatography–olfactometry and gas chromatography–mass spectrometry. A total of 55 terpenes were finally identified in Moutai liquor by mass spectrum data from NIST11 a.L and wiley7n.L mass-spectral library, retention indices, and authentic standards. A majority of terpenes were reported in Chinese liquors for the first time. A headspace–solid phase microextraction coupled with gas chromatography–mass spectrometry was developed to determine terpenes in Moutai liquor and other aroma-type Chinese liquors. According to the gas chromatography–olfactometry and odor activity values analysis, terpenes were the important aroma compounds in Moutai liquor which led the liquor more elegant and delicate.

Keywords:

• Moutai liquor
• Terpenes
• Chinese liquors
• GC–O
• GC-MS
• HS–SPME
• OAVs

Introduction

Kweichow Moutai liquor is one of the most famous distilled spirits in the world, achieving the same international status as Scottish Whisky and French Cognac. It is produced in Maotai town, Renhuai city, in the northwest of the Guizhou province. Benefiting its unique flavor, Moutai liquor is one of the most popular in China. It is regarded as the country’s national liquor.

Moutai liquor belongs to Maotai-flavor of Chinese liquor. The production process of Moutai liquor is unique of Chinese liquor. Because the liquor-making processes, raw material, and the region of Moutai liquor are different from other aroma-type Chinese liquors. The sensory characteristics of Moutai liquor are prominent soy sauce aroma, refined and delicate, pleasantly strong and full, with enduring aftertaste and lingering aroma in empty glasses.^[1] The volatile compounds of Chinese liquor are quite complex. In recent years, some papers reported the volatile compounds of Chinese liquor were main esters, alcohols, acids, phenolic compounds, acetals, ketones, aldehydes, sulfur compounds, and heterocyclic compounds.^[2] There is little literature report aroma compounds in Moutai liquor. Most of literatures about Moutai liquor are focused on manufacturing technique.^[3] Recently, Zhu and co-workers identified 528 volatile compounds in Moutai liquor by comprehensive two-dimensional gas chromatography (GC)-time of flight mass spectrometry (MS).^[4] The aroma of volatile compounds in Moutai liquor were not evaluated.

Terpenes is one of the most important classes of characterized aroma compounds and bioactive components which were mainly determined in essential oils^[5] and Muscat wines.^[6] They were seldom been reported in Chinese liquors. More recently, Hu and co-workers identified terpenes in Dongjiu liquor which is belonged to Chinese herbaceous aroma-type liquor.^[7] It was related with adding Chinese herbaceous plants to the Daqu and Xiaoqu production process of Dongjiu liquor. Compared with Dongjiu liquor, other aroma-type Chinese liquors are directly fermented from grains which are not added Chinese herbaceous plants to production process. As a new category of compounds, terpenes have not been discovered in Chinese liquors except for Dongjiu liquor.

The aim of this research was evaluated the odor of terpenes in Moutai liquor. Liquid–liquid extraction (LLE) and silica gel normal phase chromatography coupled with gas chromatography–olfactometry (GC–O) and GC–MS were used. A headspace–solid phase microextraction (HS–SPME) method was developed to quantify trace terpenes in different aroma-type Chinese liquors.

Materials and Methods

Chemicals

Theaspirane (≥90%), d-limonene (99%), p-cymene (97%), α-longipinene (≥99%), trans-linalool oxide (≥95%), α-gurjunene (≥97%), linalool (≥97%), α-cedrene (99%), β-elemene (≥98%), terpinen-4-ol (95%), β-caryophyllene (98.5%), pulegone (98%), (+)-aromadendrene (≥97%), isophorone (97%), β-guaiene (98%), α-humulene (≥96%), alloaromadendrene (≥98%), β-cedrene (≥97%), α-terpineol (96%), (-)-borneol (99%), valencene (≥65%), β-chamigrene (90%), β-citronellol (≥95%), nerol (97%), β-damascenone (11–13 g/kg in 800 mL/L ethanol solution), E-geranylacetone (97%), geraniol (98%), cis-thujopsene (≥97%), citral (95%), α-ionone (≥90%), globulol (≥98.5%), dihydro-β-ionone (≥90%), β-ionone (97%), α-cedrene epoxide (95%), E-nerolidol (≥98%), α-cedrol (99%), α-bisabolol (≥95%), β-eudesmol (≥90%), farnesol (≥95%), (E,E)-farnesylacetone (≥90%), geranylgeraniol (≥85%), and squalene (98%) were purchased from Sigma-Aldrich (Missouri, USA). L-menthyl acetate (98%) and L-menthol (99%) were internal standard (IS) which were obtained from Sigma-Aldrich (Missouri, USA). An n-alkane mixture (Sigma-Aldrich, Missouri, USA) was used to determine retention indices.

Methanol, diethyl ether, and pentane were purchased from Tedia (Ohio, USA). Ethanol (99.99%), anhydrous sodium sulfate, sodium chloride, and sulfuric acid were purchased from China National Pharmaceutical Group Corp. (Shanghai, China). Pure water was obtained from Aquapro equipment (Aquapro Co, Bedford, MA, USA).

Liquor Samples

Moutai liquor was supplied by Moutai Co. Ltd. Other Chinese liquor samples were purchased from local stores, including sauce-aroma-type liquor: sauce-1# (500 mL, 53% vol) and sauce-2# (500 mL, 53% vol); strong-aroma-type liquor: strong-1# (500 mL, 52% vol), strong-2# (500 mL, 52% vol), strong-3# (500 mL, 46% vol) and strong-4# (500 mL, 52% vol); mixed-aroma-type liquor: mixed-1# (500 mL, 53% vol) and mixed-2# (500 mL, 46% vol); light-aroma-type liquor: light-1# (500 mL, 53% vol); Laobaigan aroma-type liquor: LBG (500 mL, 67% vol). All of these Chinese liquors were bottled in 2013, and all samples were stored at 20°C until analysis.

LLE of Minor Volatiles

About 400 mL Moutai liquor was diluted to 12% vol ethanol content with ultrapure water which was boiled 3 min, and then cooled to 10°C. The diluted liquor sample was saturated with 280 g analytical-grade sodium chloride, and then extracted three times with 120 mL aliquots of redistilled diethyl ether:pentane (v/v = 1:1) in a separatory funnel. All extracts were combined and labelled as “extract 1.” To convenient GC–O and GC–MS analysis. The “extract 1” of Moutai liquor was separated into “neutral/basic” and “acidic/water-soluble” fractions, using a modified method of Fan and Qian.^[2] A total of 100 mL ultrapure water and 40 g analytical-grade sodium chloride were added to “extract 1” and mixed well. The aqueous phase was adjusted to pH 10 with sodium bicarbonate solution (10%). The aqueous phase was released from the bottom of the separating funnel and labelled as “aqueous phase 1.” The organic phase was released from the top of the separating funnel and labelled as “extract 2.” “Extract 2” was washed with 10 mL ultrapure water in a separating funnel. The organic phase was collected and labelled as “extract 3.” Combined the rest aqueous phase and “aqueous phase 1.” The mixed aqueous phase was adjusted to pH 2 with 2 N H₂SO₄ in a separatory funnel, and then extracted three times with 90 mL freshly redistilled diethyl ether. The diethyl ether extracts were combined and dried overnight with 10 g anhydrous sodium sulphate. The extract was slowly concentrated to 200 µL with a gentle stream of nitrogen. This concentrate was labelled as the “acidic/water-soluble fraction” for further GC–MS analysis. “Extract 3” was dried overnight with 10 g anhydrous sodium sulphate, and then concentrated to 1 mL with a gentle stream of nitrogen. This fraction was labelled as the “neutral/basic fraction.”

Normal-Phase Liquid Chromatography

The “neutral/basic fraction” was further separated into nine fractions by silica gel normal phase chromatography, using a modified method of Wang and Qian.^[8] A total of 20 g silica gel was soaked with 50 mL methanol (high-performance liquid chromatography [HPLC]) overnight, showing the purity of methanol is Chromatography grade purity. The soaked silica gel was rapidly poured into a glass column (40 cm × 1.5 cm i.d.), and washed with 50 mL methanol, then washed with 50 mL redistilled diethyl ether, and conditioned with 50 mL redistilled pentane. The “neutral/basic fraction” (1 mL) was added to the silica column quickly, and was eluted with 50 mL redistilled pentane. The outflow of pentane fraction was collected and named as fraction 1, 40 mL redistilled pentane:diethyl ether (95:5, fraction 2). The rest of fractions were collected according to Table 1. The flow rate of elution was 1 mL/min. All fractions were dried overnight with anhydrous sodium sulphate and then were slowly concentrated to 200 μL with a gentle stream of nitrogen for GC–O and GC–MS analysis.

TABLE 1 The neutral/basic fraction of Moutai liquor was separated into nine fractions by normal-phase liquid chromatography

Fraction^a	Solvent^b	Volume^c	Aroma descriptors^d
F1	100% pentane	0–50 mL	slight popcorn, herbal
F2	pen:die^e (95:5)	0–40 mL	fruity, floral
F3	pen:die (95:5)	40–80 mL	floral
F4	pen:die (95:5)	80–120 mL	cooked vegetables, slight floral
F5	pen:die (90:10)	0–70 mL	herbal
F6	pen:die (80:20)	0–50 mL	grass, sweet cheese, nutty
F7	pen:die (70:30)	0–50 mL	sweet
F8	diethyl ether	0–50 mL	acidic
F9	100% methanol	0–50 mL	ND^f

^aFraction: the fraction was labelled as F and an arabic number;

^bSolvent: the neutral/basic fraction was eluted by different volume ratio redistilled solvent;

^cVolume: the volume of solvent;

^dAroma descriptors: the characteristic aroma was described by panelists with filter paper strip sniffing;

^epen:die, pentane:diethyl ether(v/v);

^fND: odorless.

GC–O Analysis

GC–O analysis was performed on an Agilent 7890A GC coupled with an Agilent 5975C mass selective detector (MSD) and an olfactometer. The column carrier gas was helium with a constant pressure. Half of the chromatographic column separation flow was directed to the MSD, while the rest was directed to a heated sniffing port for GC–O analysis. Samples were analyzed on a DB-FFAP polyethylene glycol polar column (60 m length, 0.25 mm i.d., 0.25 μm film thickness, J&W Scientific, Folsom, CA, USA). Each fraction (1 μL) was injected in GC injection port with a splitless mode. The oven temperature was held at 40°C for 2 min, then raised to 230°C at a rate of 4°C/min, and held at 230°C for 20 min on a DB–FFAP column. GC injection port was 250°C. Four panelists (two females and two males) were selected for the GC–O study. They were trained with GC–O technique more than 200 h and had more than 5 years of sensory analysis experience in Moutai liquor. A 16-point scale ranging from 0 to 15 was used to respond the aroma intensity by panelists. “0” was none, “7” was moderate, while “15” was extreme. The retention time, aroma descriptor, and intensity value were recorded. Each fraction was sniffed three times by each panelist. The intensity values of aroma compounds were calculated by the averaged for 12 times analysis.

GC–MS Analysis

An Agilent 7890A GC coupled with an Agilent 5975C MSD was used for GC–MS analysis. The injection port temperature was 250°C with a splitless mode. A DB-FFAP column (60 m length, 0.25 mm i.d., 0.25 μm film thickness, J&W Scientific, Folsom, CA, USA) and a DB-5 column (30 m length, 0.25 mm i.d., 0.25 μm film thicknesses, Agilent Technology, Santa Clara, CA, USA) were used, respectively. The oven temperature started at temperature 40°C (holding 2 min), then raised to 230°C at a rate of 5°C/min, and held at 230°C for 20 min on the DB-wax column, while the final temperature was 280°C for 10 min on the DB-5 column. The helium (with the purity 99.999%) was as carrier gas at a flow rate of 1.2 mL/min. The ion source temperature was at 230°C, and the electron impact (EI) energy was 70 eV. Full-scan acquisition mode was used to masses at range from 33 to 500 amu.

Mass spectra of unknown compounds were compared with those in the NIST11 a.L Database (Agilent Technologies, Inc.) and wiley7n.L mass-spectral library. Identification was achieved by comparing mass spectra, aroma, and retention indices of standards on DB-FFAP and DB-5 capillary columns. RIs of unknown terpenes were calculated by the retention time of a standard mixture of n-alkanes (C₅–C₂₅).

SPME Parameter

An automatic SPME sampling system (GERSTEL Inc., Baltimore, MD, USA) with a fiber was used for analyte extraction. Various parameters affecting the extraction efficiency were investigated to find suitable SPME sampling conditions. Different fibers were used to test their extraction efficiency, including 50/30 μm Divinylbenzene (DVB)/Carboxen (CAR)/Polydimethylsiloxane (PDMS) fiber (the length of fiber was 1 and 2 cm), PDMS, CAR/PDMS, and PDMS/DVB (Supelco Inc., Bellefonte, PA, USA). A comparison between direct immersion and headspace techniques was carried out evaluating their efficiency. In order to evaluate the effect of ethanol content (v/v) on extraction efficiency, Moutai liquor was diluted with ultrapure water to obtain different ethanol content (5, 10, 20, and 30% vol). In order to improve the extraction efficiency, various extraction temperatures (40, 50, 60, and 70°C) and extraction time (20, 30, 40, 50, and 60 min) were tested.

Calibration of Standard Curve and Method Assessment

Different concentrations of various representative standard terpenes were added to 10 mL synthetic liquor, then the synthetic liquor sample was saturated with NaCl. A mixed IS of L-menthyl acetate and L-menthol was added with 5 μL (final concentration of 57.43 and 98.79 µg/L, respectively). The vial was tightly fitted with a Teflon septum. The samples were treated at an optimized method. Calibration curves was drew according to concentration and peak area ratios against ISs. All of the analyses were performed in triplicate under the same conditions described for samples. Triple concentration of signal/noise (S/N) was calculated as the limit of detection (LOD), and decuple concentration of S/N was calculated as the limit of quantification (LOQ). Certain concentrations of terpenes were added to different aroma-type Chinese liquor for calculating recovery rate.

Determination of Odor Thresholds

According to the method previously reported,^[9] the perception threshold of terpenes was determined by a forced-choice test at seven concentration steps. A certain amount of terpenes was added in 46 vol% hydroalcoholic solution, and stepwise diluted (a step factor of “3” was used). A triangle test including one glass of the dilution and two glasses of hydroalcoholic solution was prepared. All of the samples were encoded with random three-digit numbers. Sensory analyses were performed in a sensory room with 20 ± 1°C. A sensory panel with six panelists were asked to sniff each triangular sample and select the one was different from the other two. All of the panelists had prior sensory training and experience in Chinese liquor evaluation. The odor threshold of each terpenes was calculated by using the formulas described in the literature.^[9]

Statistical Analysis

Analysis of variance (ANOVA) was performed by SPSS 17.0 (SPSS, Chicago, IL). Significant difference was estimated at 0.05 levels. A heat map was constructed by means of R-program version 3.0.2.

Results and Discussion

Different Odor Series in Moutai Liquor

The aroma compounds of Moutai liquor were separated into “neutral/basic” and “acidic/water-soluble” fractions by LLE. The compounds in “neutral/basic” fraction were complex, and it was difficult to perform GC–O analysis. So the “neutral/basic” fraction was fractionated into nine fractions by normal-phase liquid chromatography depending on the aroma characteristics of each fraction (Table 1). Then all the fractions were detected and identified by GC–O and GC–MS.

According to the GC–O results, the descriptors of aroma (odor series) in Moutai liquor were classified into floral, fruity, green grass, acidic, herbal (dry plant), nut, sweet, and other aroma (data not shown). The aroma spider plot of Moutai liquor was drew according to the summation of odor intensity values with the same descriptor (see Fig. 1). Floral, nut, fruity odor are the key aroma in Moutai liquor. Most of terpenes and aromatic compounds contributed to floral odor. Pyrazines contributed to nut and green grass odor. Most of esters and alcohols contributed to fruity odor. Some of alcohols, aldehydes, and ketones contributed to green grass odor. A few of terpenes, phenolic, and unknown compounds contributed to herbal odor. Acidic compounds contributed to acidic odor, and some of ketones and esters contributed to sweet odor.

FIGURE 1 The spider plot of the aroma in Moutai liquor analyzed by GC–O and GC–MS.

Fruity was the prime odor in strong aroma-type liquor. Compared with strong aroma-type liquor, Moutai liquor was nice-smelling but not excessive. The odor of Moutai liquor was more complex, with elegant and delicate odor. Floral was an important part of elegant and delicate odor. Terpenes and aromatic compounds were mainly contributed floral of Moutai liquor which smelled more elegant and delicate. Aromatic compounds have been reported in Chinese liquors,^[2] so the analysis of terpenes in Chinese liquors was necessary.

Identification of Unknown Terpenes in Moutai Liquor

According to the pretreatment methods described above, unknown terpenes were identified by GC–O and GC–MS. A total of 55 terpenes were first identified in Moutai liquor by normal-phase liquid chromatography coupled with GC–MS in this work (Table 2), including 13 monoterpenes, 36 sesquiterpenes, 1 diterpene, 1 triterpene, and 4 C₁₃-norisoprenoids. Of these terpenes, d-limonene, p-cymene, α-gurjunene, α-cedrene, β-caryophyllene, terpinen-4-ol, (+)-aromadendrene, isophorone, alloaromadendrene, linalool, γ-muurolene, α-terpineol, (-)-borneol, α-muurolene, pulegone, α-humulene, β-citronellol, nerol, geraniol, δ-cadinene, dihydro-β-ionone, β-damascenone, E-geranylacetone, E-nerolidol, β-ionone, α-cedrol, α-cadinol, ledol, β-eudesmol, and farnesol have been identified by GC–O and GC–MS in Moutai liquor. According to the odor intensity value, β-damascenone (odor intensity value = 15, floral), dihydro-β-ionone (odor intensity value = 13, floral), E-geranylacetone (odor intensity value = 12, floral), E-nerolidol (odor intensity value = 11, floral), β-ionone (odor intensity value = 11, violet), linalool (odor intensity value = 11, floral), α-terpineol (odor intensity value = 10, floral), pulegone (odor intensity value = 10, floral), β-citronellol (odor intensity value = 9, floral), β-caryophyllene (odor intensity value = 8, woody), geraniol (odor intensity value =8, floral), α-cedrol (odor intensity value = 8, herbal), ledol (odor intensity value = 8, sweet), and terpinen-4-ol (odor intensity value = 7, herbal) were significant odor active terpenes in Moutai liquor. About 16 terpenes were not detected in Dongjiu liquor, including trans-linalool oxide, β-cedrene, piperitone, γ-elemene, α-cedrene epoxide, pulegone, α-chamigrene, β-citronellol, nerol, geraniol, dihydro-β-ionone, α-bisabolol, germacrene D, (E,E)-farnesylacetone, geranylgeraniol, and squalene.^[7]

TABLE 2 Identification of terpenes in Moutai liquor by normal-phase liquid chromatography coupled with GC–MS on both DB-FFAP and DB–5 columns

Number	Terpenes	Osme val^a	Aroma des^b	Fra^c	RIFFAP^d	RIDB-5^e	Identification^f
1	d-limonene	5	lemon	F1	1173	1020	MS, RI, aroma
2	p-cymene	4	lemon	F1	1264	1008	MS, RI, aroma
3	α-longipinene			F1	1463	1364	MS, RI
4	trans-linalool oxide			F8	1472	ND^g	MS, RI
5	α-gurjunene	6	woody	F1	1512	1413	MS, RI, aroma
6	linalool	11	floral	F6	1555	1087	MS, RI, aroma
7	α-cedrene	5	woody	F1	1561	1389	MS, RI, aroma
8	longifolene^h			F1	1570	1393	MS, RIL
9	β-elemene			F1	1585	1377	MS, RI
10	α-bergamotene^h			F1	1588	1420	MS, RIL
11	calarene^h			F1	1592	1424	MS, RIL
12	terpinen-4-ol	7	herbal	F7	1594	1160	MS, RI, aroma
13	β-caryophyllene	8	woody	F1	1600	1407	MS, RI, aroma
14	pulegone	10	floral	F6	1600	ND	MS, RI, aroma
15	(+)-aromadendrene	3	woody	F1	1604	1432	MS, RI, aroma
16	isophorone	5	woody	F6	1609	1114	MS, RI, aroma
17	γ-elemene^h			F1	1617	1378	MS, RIL
18	β-guaiene			F1	1620	1477	MS, RI
19	α-humulene	5	woody	F1	1651	1440	MS, RI, aroma
20	alloaromadendrene	4	woody	F1	1657	1469	MS, RI, aroma
21	β-cedrene			F1	ND	1404	MS, RI
22	γ-muurolene^h	5	herbal	F1	1694	1469	MS, RIL,aroma
23	α-chamigrene^h			F1	1701	1521	MS, RIL
24	α-terpineol	10	floral	F7	1701	1177	MS, RI, aroma
25	(-)-borneol	3	camphor	F7	1704	1159	MS, RI, aroma
26	α-muurolene^h	4	herbal	F1	1709	1497	MS, RIL, aroma
27	α-selinene^h			F1	1712	1496	MS, RIL
28	valencene			F1	1719	1498	MS, RI
29	β-bisabolene^h			F1	1725	1512	MS, RIL
30	β-chamigrene			F1	1728	1461	MS, RI
31	piperitone^h			F7	1741	1352	MS, RIL
32	germacrene D^h			F3	1742	1471	MS, RIL
33	γ-cadinene^h			F1	1751	1501	MS, RIL
34	δ-cadinene	3	woody	F1	1757	1509	MS, RIL, aroma
35	β-citronellol	9	floral	F8	1767	1429	MS, RI, aroma
36	nerol	6	floral	F5	1769	ND	MS, RI, aroma
37	α-curcumene^h			F1	1789	1467	MS, RIL
38	β-damascenone	15	floral	F3	1820	1369	MS, RI, aroma
39	dihydro-β-ionone	13	floral	F3	1833	1379	MS, RI, aroma
40	calamenene			F1	1848	1525	MS, RI
41	E-geranylacetone	12	floral	F3	1866	1461	MS, RI, aroma
42	geraniol	8	floral	F8	1871	ND	MS, RI, aroma
43	α-calacorene^h			F1	1904	1552	MS, RIL, aroma
44	β-ionone	11	violet	F3	1971	1491	MS, RI, aroma
45	α-cedrene epoxide			F5	2000	1586	MS, RI
46	ledol^h	8	sweet	F7	2013	ND	MS, RIL, aroma
47	E-nerolidol	11	floral	F7	2020	1568	MS, RI, aroma
48	α-cedrol	8	herbal	F7	2117	1594	MS, RI, aroma
49	α-cadinol^h			F8	2209	1652	MS, RIL, aroma
50	α-bisabolol			F7	2212	1692	MS, RI
51	β-eudesmol	6	woody	F7	2238	1657	MS, RI, aroma
52	farnesol	5	herbal	F7	2361	1730	MS, RI, aroma
53	(E,E)-farnesylacetone			F5	2376	1922	MS, RI
54	geranylgeraniol			F5	2785	2045	MS, RI
55	squalene			F1	3034	2834	MS, RI

^aOsme val: the aroma intensity values were the average values of 12 analyses (four panelists, three times);

^bAroma des: aroma descriptor was sniffed by each panelist;

^cFraction: terpenes were identified in major fraction;

^dRI_FFAP: retention indices were calculated on DB–FFAP column;

^eRI_DB-5: retention indices were calculated on DB–5 column;

RIL: retention indices were the literature;

^fIdentification: terpenes identified by MS (mass spectra), authentic standards, RI (retention index) and aroma descriptors;

^gND: not determined;

^hTentatively identified.

Optimization of SPME Parameters

In view of trace compounds analysis, simple and automation. SPME could be used as a quick method to determine important compounds in complex matrix samples. SPME coupled with GC–MS had been applied to detect terpenes in wines.^[10,11] In this study, SPME combined with GC–MS was employed to determine terpenes in Chinese liquors.

Different Fiber Types

Five fibers, coating PDMS, PDMS/DVB, carboxen/PDMS (CAR/PDMS), DVB/CAR/PDMS (the length of fiber was 1 and 2 cm) were used to compare their extraction efficiency of terpenes in Chinese liquors. The results showed that the DVB/CAR/PDMS fiber (with 2 cm length) had the best enrichment capacity when compared to CAR/PDMS, PDMS, and PDMS/DVB (Fig. 2a). Therefore, a DVB/CAR/PDMS fiber (with 2 cm length) was selected for further experiments.

FIGURE 2 The peak areas of terpenes classes in Moutai liquor as a function of distinct SPME extraction efficiency: Peak areas of terpenes classes were extracted by different fiber types (2-a); peak areas of terpenes classes for the different analytes according to the extraction patterns (2-b); sample ethanol content (2-c); sampling extraction temperature (2-d); extraction time (2-e); desorption temperature (2-f).

Different Extraction Patterns

Headspace and direct immersion are two extraction patterns of SPME. They have different extraction efficiency to volatile compounds. Volatile compounds are more likely to be extracted by headspace, while semi-volatile compounds are more likely to be extracted by direct immersion. The results showed that the extraction efficiency of terpenes by headspace was better than direct immersion (Fig. 2b). Consequently, further studies were performed with HS–SPME.

Ethanol Content Effect

The sample ethanol content is an important factor for volatile compounds extraction.^[12] Ethanol content with 5, 10, 20, and 30%vol were tested. It was apparent that the terpenes areas were the largest with 10% vol ethanol content (Fig. 2c). In addition, the data showed that the higher ethanol content (from 10 to 30%), the smaller the peak areas of these terpenes. It was estimated that the volatility of analytes decreased with the increment of ethanol content. On the other hand, there was greater dilution of the volatile compounds at 5% ethanol content. It was unfavorable for the extraction of minor terpenes. Consequently, further studies were performed with 10% ethanol content.

Different Extraction Temperature

It is well known that the extraction temperature is an important factor for HS–SPME analysis. Extraction temperatures of 40, 50, 60, and 70°C were tested. The results showed that the peak areas of terpenes increased with extraction temperature up to 60°C except for monoterpenes (Fig. 2d). Extraction efficiencies for terpenes were dramatically decreased at 70°C. In addition, the data showed that a lot of monoterpenes areas were largest with extraction temperature of 50°C. The amount of monoterpenes was less than sesquiterpenes and C₁₃-norisoprenoids. In general, the temperature of 60°C was selected to terpenes extraction.

Different Extraction Time

In order to obtain the best extraction efficiency, extraction time of 20, 30, 40, 50, and 60 min were tested. The results showed that the peak areas of terpenes reached a maximum value when extracted for 40 min. More extraction time was not increased the total peak areas or numbers of terpenes (Fig. 2e). An equilibration time of 5 min and extraction for 40 min was selected to terpenes detection.

Different Desorption Temperature

Desorption temperature is another important influence on sensitivity and precision. Injection port temperature must also be optimized for the analytes involved. Three different temperatures (210, 230, and 250°C) were investigated (Fig. 2f). The peak areas of terpenes desorpted from the fiber were increased with the desorption temperature. A blank run was performed before each run, in order to evaluate the efficiency of desorption. The results showed that terpenes were completely desorpted from the fiber at 250°C.

On the basis of these observations, the optimized HS–SPME parameters were as follows. Liquor sample was diluted to 10% vol with ultrapure water. The diluted samples were saturated with NaCl and equilibrated at 60°C for 5 min, and then extracted for 40 min at the same temperature by DVB/CAR/PDMS fiber (2 cm length). The extracted fiber was directly inserted injection port temperature (250°C) for desorption.

Quantification Method of Terpenes

For most terpenes, the linear responses were obtained with R² ranging from 0.9940 (nerol) to 0.9999 (theaspirane), and the LOQ were all below 1.3483 µg/L, the LOD of β-ionone was particular low at 1.61 ng/L (Table 3). Different concentrations of terpenes were added to the synthetic liquor. The recovery rate of different terpenes in the Moutai liquor was at ranged from 82.03 to 118.07%, and the recovery rate of terpenes in other Chinese liquors were 78.71 to 124.68% (data not shown). In conclusion, method of this study was reliable and powerful, able to quantify the most important terpenes in different aroma-type Chinese liquors. Except for 26 terpenes, the rest of terpenes were only semi-quantified. The slope and intercept of α-cedrene were used for the rest of terpenes which were no standard products, these terpenes included calarene, γ-muurolene, α-selinene, α-muurolene, eremophilene, γ-elemene, β-bisabolene, α-chamigrene, α-cadinol, longifolene, isoledene, β-guaiene, δ-cadinene, calamenene, α-calacorene, ledol, γ-cadinene, α-bergamotene, isolongifolol, and isocaryophillene.

Concentrations of Terpenes in Moutai Liquor and Other Aroma-Type Chinese Liquors

Terpenes in Moutai liquor and other aroma-type Chinese liquors were quantified by optimized HS–SPME coupled with GC–MS. A total of 46 terpenes were detected in Chinese liquors, including eight monoterpenes, 32 sesquiterpenes, and six C₁₃-norisoprenoids. They were quantified in Chinese liquors for the first time.

Different aroma-type Chinese liquors had different numbers and concentrations of terpenes. The concentrations of terpenes in Moutai liquor and other aroma-type Chinese liquors were listed in Table 4. According to the quantification date, E-geranylacetone (77.99 µg/L), E-nerolidol (50.66 µg/L), farnesol (33.67 µg/L), α-cedrene (32.71 µg/L), and linalool (32.03 µg/L) were the higher concentration of terpenes in Moutai liquor. Moutai liquor, sauce-1# and sauce-2# liquor belong to the soy sauce aroma-style liquor. The high concentration of terpenes in soy sauce aroma style liquor was different. By the way, linalool was the highest concentration of monoterpenes in young white wines.^[13] LBG aroma-type liquor and light-1# liquor had the highest concentrations of β-caryophyllene in all Chinese liquors (see Fig. 3). It might be related with their similar manufacturing technique. d-Limonene, citral, terpinen-4-ol, β-citronellol, geraniol, eremophilene, globulol, β-eudesmol, longifolene, isoledene, α-ionone, and theaspirane were not detected in four strong aroma-type liquors. Citral, terpinen-4-ol, β-citronellol, geraniol, α-gurjunene, isoledene, theaspirane, and isophorone were not detected in two mixed aroma-type liquors. The manufacturing technique of mixed aroma-type liquor was inherited from strong aroma-type liquor. Eremophilene was the lowest content at the range of ng/L in Moutai liquor and other aroma-type Chinese liquors. Theaspirane was only detected in light-1# liquor. It was the characteristic compound in light-1# liquor.

TABLE 3 Calibration of standard curve datas of quantification method, R², LOD, LOQ, linearity range, and recovery rate of terpenes in Moutai liquor

Terpenes	IS^a	Slo^b (×10)	Int^c (×10^4)	R^2	n^d	Quant^e (m/z)	LOD^f (µg/L)	LOQ^g (µg/L)	Linearity range (µg/L)	Recovery rate (%)
Theaspirane	L-m	0.19	0.01	0.9999	6	138	0.0030	0.0101	0.73–23.42	99.74
linalool	L-m	3.98	3.00	0.9990	6	93	0.0412	0.1374	3.63–116.01	96.08
β-caryophyllene	L-ma	4.05	–2.00	0.9977	8	133	0.0408	0.1359	2.75–352.26	105.42
cis-thujopsene	L-ma	0.29	1.81	0.9985	8	119	0.0137	0.0457	1.89–242.03	94.89
α-gurjunene	L-ma	0.51	1.96	0.9962	6	204	0.0243	0.0811	2.13–68.18	108.41
alloaromadendrene	L-ma	0.77	2.41	0.9994	7	91	0.0377	0.1256	1.72–110.11	107.35
(+)-aromadendrene	L-ma	0.35	1.84	0.9969	6	133	0.0517	0.1723	2.39–76.48	100.98
β-cedrene	L-ma	0.92	4.08	0.9991	6	161	0.0123	0.0411	1.07–34.29	115.94
α-cedrene	L-ma	0.60	–2.00	0.9972	8	119	0.0029	0.0095	0.93–118.97	106.79
d-limonene	L-ma	5.25	15	0.9965	10	93	0.3429	1.1429	0.69–352.77	86.75
isophorone	L-m	6.49	4.10	0.9954	7	81	0.0429	0.1431	3.92–251.01	90.79
terpinen-4-ol	L-m	4.18	–32.00	0.9991	6	111	0.1104	0.3680	1.85–59.29	89.47
citral	L-m	2.88	–17.00	0.9982	5	69	0.0577	0.1923	1.67–26.75	93.45
β-citronellol	L-m	1.14	1.90	0.9994	7	69	0.0623	0.2075	1.17–74.62	97.90
nerol	L-m	3.11	17.00	0.9940	6	69	0.2783	0.9278	3.38–108.08	94.02
β-damascenone	L-m	0.64	1.20	0.9991	8	121	0.0120	0.0402	1.01–128.90	95.84
geraniol	L-m	4.08	13.90	0.9976	8	69	0.0241	0.0804	2.39–305.28	94.17
E-geranylacetone	L-m	0.39	0.80	0.9975	7	69	0.0220	0.0734	1.94–124.28	106.8
β-ionone	L-m	0.31	2.00	0.9980	8	177	0.0016	0.0054	0.11–14.08	101.87
E-nerolidol	L-m	0.50	–5.00	0.9944	8	69	0.0063	0.0209	2.28–291.67	91.48
α-ionone	L-m	0.41	–1.00	0.9981	7	121	0.0076	0.0253	2.04–193.30	102.47
α-cedrol	L-m	0.54	33.00	0.9994	6	150	0.0055	0.0185	3.59–114.89	99.82
farnesol	L-m	0.84	19.00	0.9997	9	69	0.0148	0.0493	4.10–1049.79	82.03
β-eudesmol	L-m	2.50	–1.00	0.9964	8	59	0.4045	1.3483	2.04–261.12	118.07
globulol	L-m	2.01	8.43	0.9971	7	95	0.0991	0.3302	2.71–173.44	93.46
α-terpineol	L-m	3.49	6.76	0.9991	6	95	0.1038	0.3460	3.37–107.81	95.47

^aIS: internal standard, L-m: L-menthol, L-ma: L-menthyl acetate;

^bSlo: slope;

^cInt: intercept;

^dNumber of points in the calibration curve;

^eQuant: quantitative of m/z;

^fLOD: limit of detection;

^gLOQ: limit of quantification.

TABLE 4 Concentrations of terpenes in Moutai liquor and other aroma-type Chinese liquors (µg/L; n = 3)

No.	Terpenes	Ave. ± SD
		Moutai	Sauce-1#	Sauce-2#	Light-1#	Strong-1#	Strong-2#	Strong-3#	Strong-4#	Mixed-1#	Mixed-2#	LBG
C1	d-limonene	7.04 ± 1.23^b	6.92 ± 1.71^b	3.30 ± 1.01^d	4.83 ± 0.30^c	ND^e	ND^e	ND^e	ND^e	ND^e	2.28 ± 0.14^d	9.03±0.23^a
C2	citral	24.97 ± 1.46^a	16.47 ± 1.57^c	21.14 ± 2.29^b	25.18 ± 1.58^a	ND^d	ND^d	ND^d	ND^d	ND^d	ND^d	19.36±1.17^b
C3	linalool	32.03 ± 2.03^a	26.51 ± 2.37^b	26.04 ± 1.87^b	14.14 ± 1.26^d	15.32 ± 2.52^d	12.41 ± 2.81^d	6.09 ± 0.19^e	22.56 ± 0.03^c	15.38 ± 1.71^d	15.26 ± 1.71^d	16.91±1.56^d
C4	nerol	6.81 ± 0.15^b	3.88 ± 0.31^d	8.61 ± 0.18^a	4.37 ± 0.27^c	4.75 ± 0.47^c	1.87 ± 0.06^f	ND^g	ND^g	3.07 ± 0.33^e	ND^g	ND^g
C5	terpinen-4-ol	<QL	<QL	ND	<QL	ND	ND	ND	ND	ND	ND	ND
C6	α-terpineol	26.19 ± 2.5^a	15.75 ± 1.77^e	12.74 ± 0.37^f	24.09 ± 1.54^ab	17.21 ± 1.12^de	19.47 ± 1.08^cd	12.18 ± 1.16^f	18.12 ± 1.24^d	23.57 ± 2.02^ab	22.08 ± 2.72^bc	14.97±1.46^ef
C7	β-citronellol	15.96 ± 1.47^a	7.11 ± 0.11^c	10.21 ± 2.06^b	4.15 ± 0.37^d	ND^e	ND^e	ND^e	ND^e	ND^e	ND^e	ND^e
C8	geraniol	7.77 ± 1.09^b	5.08 ± 0.20^d	9.56 ± 0.72^a	6.41 ± 0.16^c	ND^e	ND^e	ND^e	ND^e	ND^e	ND^e	ND^e
C9	α-cedrene	32.71 ± 2.15^a	15.81 ± 1.54^b	10.72 ± 1.48^c	8.24 ± 1.06^cd	9.03 ± 1.10^c	10.54 ± 0.71^c	6.60 ± 0.99^d	9.31 ± 1.24^c	9.12 ± 0.69^c	9.39 ± 1.94^c	10.87±1.61^c
C10	β-caryophyllene	21.68 ± 2.73^d	6.06 ± 1.08^h	28.59 ± 1.39^c	71.11 ± 3.38^b	9.82 ± 0.62^g	18.59 ± 1.44^de	7.93 ± 1.04^gh	9.43 ± 0.74^g	17.75 ± 0.88^e	13.41 ± 1.52^f	140.11±4.82^a
C11	α-cedrol	13.14 ± 1.13^a	6.15 ± 1.01^cd	10.42 ± 1.22^b	4.26 ± 0.30^ef	4.28 ± 0.41^ef	5.37 ± 0.50^de	3.95 ± 0.87^f	5.21 ± 0.38^de	4.67 ± 0.18^ef	6.96 ± 0.96^c	4.77±0.45^e
C12	calarene*	3.52 ± 0.06^b	3.43 ± 0.07^bc	1.33 ± 0.07^g	3.26 ± 0.30^cd	3.19 ± 0.08^de	3.92 ± 0.17^a	2.39 ± 0.09^f	3.53 ± 0.06^b	2.63 ± 0.11^f	2.99 ± 0.14^e	4.05±0.17^a
C13	γ-muurolene*	3.82 ± 0.01^e	6.09 ± 0.28^b	4.35 ± 0.16^d	3.88 ± 0.23^e	3.55 ± 0.20^ef	4.96 ± 0.07^c	3.34 ± 0.04^f	3.59 ± 0.49^ef	3.55 ± 0.03^e	5.84 ± 0.23^b	8.27±0.13^a
C14	α-selinene*	6.81 ± 0.12^a	ND^c	3.11 ± 0.14^b	3.04 ± 0.16^b	2.96 ± 0.55^b	3.05 ± 0.13^b	ND^c	2.94 ± 0.16^b	ND^c	ND^c	ND^c
C15	α-muurolene*	3.85 ± 0.57^def	4.27 ± 0.45^d	3.97 ± 0.44^de	3.57 ± 0.04^ef	4.29 ± 0.13^d	4.94 ± 0.55^c	2.59 ± 0.08^h	5.76 ± 0.23^b	3.32 ± 0.33^f	3.44 ± 0.24^ef	8.07±0.30^a
C16	eremophilene*	1.02 ± 0.01^a	0.95 ± 0.06^a	0.97 ± 0.04^a	0.93 ± 0.01^b	ND^d	ND^d	ND^d	ND^d	0.79 ± 0.02^c	ND^d	ND^d
C17	β-cedrene	4.03 ± 0.30^ab	4.27 ± 0.33^a	3.78 ± 0.51^b	ND^f	ND^f	3.91 ± 0.16^ab	2.57 ± 0.27^e	3.56 ± 0.38^bc	2.97 ± 0.06^de	3.32 ± 0.33^cd	ND^f
C18	γ-elemene*	10.33 ± 0.98^c	10.17 ± 1.82^c	9.03 ± 0.95^c	ND^g	12.04 ± 1.10^b	6.72 ± 0.61^d	2.49 ± 0.11^f	13.60 ± 0.14^a	4.17 ± 0.07^e	3.23 ± 0.10^ef	ND^g
C19	alloaromadendrene	3.86 ± 0.14^d	5.58 ± 0.13^b	4.53 ± 0.16^c	3.24 ± 0.20^e	3.33 ± 0.24^e	3.85 ± 0.07^d	2.55 ± 0.01^f	3.26 ± 0.17^e	2.77 ± 0.13^f	3.31 ± 0.21^e	6.66±0.08^a
C20	β-bisabolene*	3.91 ± 0.21^d	4.88 ± 0.07^b	4.42 ± 0.44^c	ND^g	3.70 ± 0.08^d	3.83 ± 0.20^d	2.49 ± 0.48^f	3.86 ± 0.28^d	2.82 ± 0.30^ef	3.19 ± 0.14^e	6.73±0.20^a
C21	α-chamigrene*	14.87 ± 1.73^b	13.66 ± 2.45^b	26.87 ± 1.03^a	3.71 ± 0.51^de	5.31 ± 0.42^cd	6.01 ± 0.45^c	2.35 ± 0.99^e	3.98 ± 0.10^de	4.14 ± 0.34^d	4.08 ± 0.18^de	7.16±0.08^c
C22	globulol	4.67 ± 0.37^b	18.25 ± 0.20^a	3.88 ± 0.16^c	ND^e	ND^e	ND^e	ND^e	ND^e	3.16 ± 0.04^d	4.41 ± 0.38^b	ND^e
C23	α-cadinol*	4.32 ± 0.33^bc	4.62 ± 0.13^b	3.97 ± 0.54^c	4.33 ± 0.17^b	3.74 ± 0.78^cd	3.86 ± 0.20^c	3.18 ± 0.48^d	3.80 ± 0.01^c	6.12 ± 0.24^a	4.30 ± 0.17^bc	6.12±0.06^a
C24	β-eudesmol	3.73 ± 0.38^a	<QL^d	<QL^d	<QL^d	ND^d	ND^d	<QL^d	ND^d	2.12 ± 0.17^c	<QL^d	2.70±0.28^b
C25	longifolene*	7.14 ± 0.20^b	4.07 ± 0.17^e	4.29 ± 0.73^de	6.75 ± 0.21^b	4.01 ± 0.76^e	4.76 ± 0.07^cd	3.26 ± 0.24^f	3.99 ± 0.20^e	4.76 ± 0.07^cd	5.10 ± 0.04^c	9.54±0.48^a
C26	α-gurjunene	6.50 ± 0.30^a	ND^d	ND^d	3.80 ± 0.72^b	ND^d	2.93 ± 0.52^c	ND^d	ND^d	ND^d	ND^d	6.60±0.03^a
C27	isoledene*	3.41 ± 0.44^b	1.49 ± 0.72^d	2.73 ± 0.38^c	2.71 ± 0.41^c	ND^e	ND^e	ND^e	ND^e	ND^e	ND^e	5.01±0.01^a
C28	(+)-aromadendrene	3.14 ± 0.23^b	3.21 ± 0.65^b	3.02 ± 0.64^b	3.30 ± 0.23^b	ND^c	2.97 ± 0.13^b	ND^c	3.01 ± 0.23^b	ND^c	3.06 ± 0.18^b	4.74±0.16^a
C29	cis-thujopsene	3.86 ± 0.75^cd	7.54 ± 0.01^a	4.16 ± 0.67^c	3.24 ± 0.49^de	3.30 ± 0.42^de	3.85 ± 0.17^cd	2.45 ± 0.18^f	3.22 ± 0.27^de	2.66 ± 0.14^ef	ND^g	6.66±0.71^b
C30	β-guaiene*	3.60 ± 0.33^b	6.89 ± 0.30^a	3.69 ± 0.78^b	3.38 ± 0.55^bc	3.34 ± 0.04^bc	3.37 ± 0.69^bc	ND^e	2.60 ± 0.40^d	2.75 ± 0.04^cd	3.19 ± 0.37^bcd	6.23±0.31^a
C31	δ-cadinene*	10.43 ± 0.30^b	3.23 ± 0.27^fg	3.50 ± 0.23^ef	4.34 ± 0.24^cd	4.05 ± 0.08^de	4.91 ± 0.07^c	2.64 ± 0.28^g	3.43 ± 0.41e^f	3.79 ± 0.04^def	3.16 ± 0.21^fg	13.53±1.21^a
C32	calamenene*	4.47 ± 0.98^e	6.58 ± 0.03^c	5.29 ± 0.62^de	4.51 ± 1.19^e	8.47 ± 0.16^b	9.12 ± 0.41^b	4.87 ± 0.57^e	6.02 ± 0.14^cd	6.06 ± 0.25^cd	4.75 ± 0.45^e	14.01±0.42^a
C33	α-calacorene^e	7.60 ± 0.17^d	6.99 ± 0.04^e	9.05 ± 0.47^c	8.53 ± 0.23^c	11.44 ± 0.25^b	11.27 ± 0.10^b	5.18 ± 0.20^f	7.96 ± 0.47^d	8.42 ± 0.14^c	6.94 ± 0.24^e	26.55±0.24^a
C34	ledol*	4.42 ± 0.21^d	4.50 ± 0.17^cd	5.18 ± 0.18^bc	5.64 ± 0.41^b	4.20 ± 0.52^d	5.75 ± 0.27^b	2.98 ± 0.24^e	4.45 ± 0.31^d	4.24 ± 0.59^d	3.92 ± 0.14^d	15.65±0.82^a
C35	γ-cadinene*	3.23 ± 0.10^bc	3.52 ± 1.02^bc	3.20 ± 0.64^bc	3.31 ± 0.06^bc	3.52 ± 0.45^bc	3.96 ± 0.23^b	2.41 ± 0.34^d	3.22 ± 0.69^bc	2.78 ± 0.03^d	3.16 ± 0.25^c	5.27±0.31^a
C36	α-bergamotene*	11.96 ± 0.06^b	17.52 ± 0.40^a	17.68 ± 1.2^a	3.90 ± 0.25^e	5.82 ± 0.30^d	5.21 ± 1.03^de	2.99 ± 0.24^e	4.38 ± 1.07^de	3.80 ± 1.02^e	3.98 ± 1.40^e	9.87±0.91^c
C37	E-nerolidol	50.66 ± 3.72^a	11.05 ± 1.74^d	24.50 ± 0.07^b	4.35 ± 0.86^e	2.92 ± 0.04^ef	4.46 ± 1.11^e	1.43 ± 0.44^f	10.82 ± 0.68^d	10.19 ± 0.30^d	21.11 ± 1.10^c	20.97±0.20^c
C38	α-ionone	4.06 ± 0.35^ab	3.82 ± 0.58^b	4.32 ± 0.31^a	3.85 ± 0.03^ab	<QL^d	<QL^d	ND^d	ND^d	3.72 ± 0.47^b	4.06 ± 0.28^ab	3.31±0.27^c
C39	isolongifolol*	4.44 ± 0.08^cd	4.07 ± 0.37^de	4.29 ± 0.52^cd	6.75 ± 0.23^b	4.01 ± 0.23^de	4.76 ± 0.17^cd	3.26 ± 0.12^e	3.99 ± 0.33^de	4.76 ± 0.21^cd	5.10 ± 0.87^c	9.54±1.48^a
C40	isocaryophillene*	5.58 ± 0.82^bc	5.02 ± 0.90^cd	4.88 ± 0.43^cd	6.31 ± 0.41^b	3.61 ± 0.18^f	4.45 ± 0.27^de	3.33 ± 0.03^f	3.89 ± 0.44^ef	4.06 ± 0.04^ef	4.39 ± 0.20^de	13.07±0.08^a
C41	farnesol	33.67 ± 1.95^a	11.26 ± 1.56^e	20.00 ± 1.15^c	10.36 ± 0.27^ef	5.48 ± 1.23^hi	6.84 ± 0.10^gh	4.37 ± 1.24^i	7.45 ± 1.92^gh	8.32 ± 0.91^fg	15.61 ± 1.40^d	27.27±1.32^b
C42	theaspirane	<QL	<QL	<QL	0.82 ± 0.01	ND	ND	ND	ND	ND	ND	<QL
C43	isophorone	13.99 ± 1.43^a	5.73 ± 1.47^d	12.09 ± 1.06^b	8.33 ± 0.98^c	ND^e	6.17 ± 0.13^d	ND^e	ND^e	ND^e	ND^e	6.24±0.57^d
C44	E-geranylacetone	77.99 ± 2.93^a	20.47 ± 1.53^d	34.90 ± 2.47^b	7.97 ± 1.16^f	16.72 ± 1.00^e	13.70 ± 1.32^e	5.25 ± 0.55^f	21.15 ± 2.60^d	18.25 ± 0.64^de	20.85 ± 1.58^d	28.74±2.76^c
C45	β-ionone	1.73 ± 0.18^b	0.49 ± 0.13^ef	1.47 ± 0.16^c	1.27 ± 0.01^d	0.51 ± 0.14^ef	0.40 ± 0.01^fg	0.30 ± 0.01^g	0.64 ± 0.04^e	2.20 ± 0.03^a	1.54 ± 0.05^c	1.38±0.13^cd
C46	β-damascenone	11.83 ± 0.34^ab	4.59 ± 0.20^f	12.32 ± 0.81^a	9.92 ± 0.25^c	7.64 ± 0.54^d	6.64 ± 0.34^e	4.24 ± 0.55^f	8.05 ± 0.45^d	7.95 ± 0.01^d	7.78 ± 0.52^d	10.51±0.99^bc
monoterpenes		120.77	81.72	91.60	83.17	37.28	33.75	18.27	40.68	42.02	39.62	60.27
sesquiterpenes		304.44	204.95	239.72	194.60	129.41	158.16	81.60	140.27	140.41	155.40	405.54
C13-norisoprenoids		105.54	31.28	60.78	28.31	24.87	26.91	9.79	29.84	28.40	30.17	46.87
total concentrations		530.75	317.95	392.10	306.08	191.56	218.82	109.66	210.79	210.83	225.19	512.68

No.: number of compounds; Ave: average concentration; SD: standard deviation, three replicates for each of the liquor; ND: not detected; <QL: below quantification limit;

*Terpenes were semi-quantified by using the slope and intercept of α-cedrene;

^{a,b,c,d,e,f,g,h,i}Different letters in the same row denote significant differences (p < 0.05), and same letter in the same row denote no significant differences (p > 0.05).

TABLE 5 OAVs^a of terpenes in Moutai liquor and other aroma-type Chinese liquors

Terpenes	OAVs
	Odor^b	Thr (µg/L)^c	Moutai	Sauce-1#	Sauce-2#	Light-1#	Strong-1#	Strong-2#	Strong-3#	Strong-4#	Mixed-1#	Mixed-2#	LBG
β-damascenone	floral	0.12^d	98	38	103	83	64	55	35	67	66	65	88
β-ionone	violet	1.3^e	1.3	<1	1.1	1.0	<1	<1	<1	<1	1.7	1.2	1.1
α-ionone	violet	1.8^e	2.3	2.1	2.4	2.1	<1	<1	<1	<1	2.1	2.3	1.8
citral	lemon	3^e	8.3	5.5	7.0	8.4	<1	<1	<1	<1	<1	<1	6.5
linalool	floral	13.1^e	2.4	2.0	2.0	1.1	<1	<1	<1	1.7	1.2	1.2	1.3
d-limonene	lemon	41^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
β-citronellol	floral	300^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
geraniol	floral	120^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
E-geranylacetone	floral	267^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
nerol	floral	130 ^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
terpinen-4-ol	herbal	940^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
α-terpineol	floral	680^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
β-caryophyllene	woody	130^e	<1	<1	<1	1	<1	<1	<1	<1	<1	<1	1.0
E-nerolidol	floral	99^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
theaspirane	woody	9.2^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
cedrol	woody	7300^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1
α-cedrene	woody	6500^e	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1	<1

^aOAVs: odor activity values;

^bOdor: aroma descriptors from the standards;

^cThr: odor threshold was selected from literature;

^dThresholds was determined in 46% volume hydroalcoholic solution (Gao et al., 2014);

^eThresholds was determined in 46% volume hydroalcoholic solution.

FIGURE 3 Heat map analysis of terpenes in different aroma type Chinese liquors.

Among all the Chinese liquors, the concentrations of monoterpenes, sesquiterpenes and C₁₃-norisoprenoids in different aroma-type Chinese liquors were at range from 18.27 to 140.77 µg/L, 81.60 to 405.54 µg/L, and 9.79 to 105.54 µg/L, respectively. The total concentrations of terpenes in different aroma-type Chinese liquors were at range from 109.66 to 530.75 µg/L. The concentrations of terpenes in Moutai liquor were different from other aroma-type Chinese liquors. The results showed that the concentrations of monoterpenes, C₁₃-norisoprenoids, and total terpenes in Moutai liquor were higher than other aroma-type Chinese liquors, whereas the LBG aroma-type liquor had a higher concentration of sesquiterpenes. This was mainly caused by the high content of β-caryophyllene. According to the quantification date, the total concentration of terpenes in Moutai liquor was the highest, followed by LBG aroma-type liquor, sauce-1#, sauce-2# liquors, light-aroma-type liquor, finally was the mixed-aroma-type liquors and strong-aroma-type liquors.

Odor Activity Values (OAVs) of Terpenes in Moutai Liquor and Other Aroma-Type Chinese Liquors

It is known that, not all of the compounds contribute its odor to food, OAVs was the key indicators of volatile compounds in food.^[14] On the basis of OAVs, the most important terpenes (OAVs ≥ 1) in Moutai liquors were β-damascenone, citral, linalool, α-ionone, and β-ionone. They were the trace characteristic flavor compounds in Chinese liquors. But their odor threshold was also low in wines and liquors. For example, the odor threshold of β-damascenone was only 0.05 µg/L in wines,^[15] and 0.12 µg/L in Chinese liquor.^[16] It is widely present in wines,^[17] beers, distilled spirits, and other alcoholic brew products.^[18] Although the odor threshold of β-damascenone in Chinese liquor was one times higher than in wine, the concentration of β-damascenone in liquor is much higher than in wines, the OAVs of β-damascenone in Chinese liquors was higher than in wines^[16,19]. Linalool and α-ionone were the free aroma compounds in monovarietal and co-winemaking wines.^[20] Comparing to the other aroma-type Chinese liquors, terpenes had higher OAVs in Moutai liquor, whereas β-caryophyllene was the important terpenes in the LBG aroma-type liquor (OAV = 1). Comparing with the results of GC–O analysis reported previously, most terpenes with higher OAVs in Moutai liquor also had higher odor intensity values.

According to the citations, terpenes are the hydrolysis product of the bound compounds. One way is biological effect which microorganism or enzymes use substrate to generate terpenes,^[21] and another way is chemical cleavage process under acid or hot conditions,^[22] temperature was the most important parameter in the formation of β-damascenone and other C₁₃-norisoprenoids in Chardonnay white wines aging.^[23] Recently, some articles report β-damascenone was generated by some non-Saccharomyces cerevisiae, such as Debaryomyces, Maggi Metschnikowia, abnormal Pichia, and Hansenula in the wine making process.^[24] The origin of β-damascenone and other terpenes in Chinese liquor have been still unknown. The production process of Chinese liquor is different from other alcoholic beverages. It was very complex, including raw material, solid-state fermentation of Daqu, solid-state fermentation of Jiupei, solid-state distillation process, and storage process in a china jar. A variety of precursors would be generated by microorganisms in the solid-state fermentation process of Chinese liquor. Quantification of terpenes in the production process was conducive to understanding the origin of terpenes in Chinese liquor.

Conclusion

In this study, the aroma of Moutai liquor was classified into floral, fruity, green grass, acidic, herbal (dry plant), nut, sweet and other aroma by GC–O analysis. A total of 55 terpenes were identified by GC–MS (MS, RIs, authentic standards, and aroma), and 30 terpenes were analyzed by GC–O in Moutai liquor for the first. According to GC–O analysis, the most intense odorant of terpenes were β-damascenone, pulegone, β-ionone, E-geranylacetone, E-nerolidol, α-terpineol, and linalool. A total of 46 terpenes were quantified in different aroma-type Chinese liquors by HS–SPME coupled with GC–MS. The concentrations of monoterpenes, sesquiterpenes, C₁₃-norisoprenoids and total terpenes in different aroma-type Chinese liquors were at range from 18.27 to 140.77, 81.60 to 405.54, 9.79 to 105.54, and 109.66 to 530.75 µg/L, respectively. The results of OAVs analysis showed that β-damascenone, citral, linalool, α-ionone, and β-ionone had comparatively high OAV values, which mean they were the important terpenes in Moutai liquor and other aroma-type Chinese liquors. According to GC–O and OAVs analysis, terpenes could contribute a more elegant and delicate odor to Moutai liquor.