Skip to main content
NCBI home page
Food Chemistry: X logoLink to Food Chemistry: X
. 2023 Apr 25;18:100693. doi: 10.1016/j.fochx.2023.100693

Characterization of volatile metabolites in Pu-erh teas with different storage years by combining GC-E-Nose, GC–MS, and GC-IMS

Yuting Rong a,1, Jialing Xie b,1, Haibo Yuan b, Lilei Wang b, Fuqiao Liu a, Yuliang Deng b, Yongwen Jiang b,⁎, Yanqin Yang b,⁎
PMCID: PMC10314134  PMID: 37397226

Graphical abstract

graphic file with name ga1.jpg

Open in a new tab

Keywords: Storage time, Pu-erh tea, GC-E-Nose, GC-MS, GC-IMS

Highlights

  • •

    The fingerprints of Pu-erh tea were constructed by GC-E-Nose, GC–MS and GC-IMS.

  • •

    Significant changes in the volatile components occurred over 3-year-storage time.

  • •

    A clear discrimination was achieved based on the volatile metabolites with GC-IMS.

  • •

    Nine volatile biomarkers were identified based on VIP＞1.2 and p＜0.05.

Abstract

Storage time is one of the important factors affecting the aroma quality of Pu-erh tea. In this study, the dynamic changes of volatile profiles of Pu-erh teas stored for different years were investigated by combining gas chromatography electronic nose (GC-E-Nose), gas chromatography-mass spectrometry (GC–MS), and gas chromatography-ion mobility spectrometry (GC-IMS). GC-E-Nose combined with partial least squares-discriminant analysis (PLS-DA) realized the rapid discrimination of Pu-erh tea with different storage time (R2Y = 0.992, Q2 = 0.968). There were 43 and 91 volatile compounds identified by GC–MS and GC-IMS, respectively. A satisfactory discrimination (R2Y = 0.991, and Q2 = 0.966) was achieved by using PLS-DA based on the volatile fingerprints of GC-IMS. Moreover, according to the multivariate analysis of VIP > 1.2 and univariate analysis of p < 0.05, 9 volatile components such as linalool and (E)-2-hexenal were selected as key variables to distinguish Pu-erh teas with different storage years. The results provide theoretical support for the quality control of Pu-erh tea.

1. Introduction

Tea is highly popular among the world's three major beverages, attracting many consumers with its pleasant aroma and taste. As a special microbial post-fermented tea variety, Pu-erh tea originates from Yunnan Province of China, and is famous for its unique stale flavor and mellow taste, as well as multiple health benefits such as anti-tumor, anti-oxidation, or lipid reduction (Yang, Wang, & Sheridan, 2018). The reported bioactive components of Pu-erh tea mainly include tea polyphenols, tea polysaccharides, gallic acid, caffeine and so on (Liu et al., 2021). Using leaves from Camellia sinensis var. assamica as raw material, Pu-erh tea is made through a series of special processes, including spreading, fixation, post-fermentation, shaping, and drying. And the post-fermentation is the key process to influence the quality of Pu-erh tea (Ma et al., 2022). In this process, the special environment of high humidity and temperature promotes the growth, propagation and fermentation of microorganisms, forming the special flavor of Pu-erh tea.

Aroma is one of the importance factors in the evaluation of Pu-erh tea quality, although it only accounts for 0.005–0.02% of the chemical substances in the tea. Tea aroma is a combination of various volatile components, and the differences in the types and contents of these substances may render different aroma types. Nowadays, instrumental analysis technologies have been increasingly used to characterize and quantify the volatile components. So far, more than 1000 volatile compounds have been found in Pu-erh tea (Wang et al., 2022). Electronic nose (E-nose) is one of the common intelligent sensory technologies, consisting of odor sensors and a pattern recognition algorithm. It can partly eliminate the subjectivity of human sensory evaluation and serve as its crucial supplement (Xu et al., 2021, Yu et al., 2008). As a new type of odor analyzer, gas chromatography electronic nose (GC-E-nose) exhibits the advantages of rapid separation and pattern recognition of the overall odor characteristics, making it an efficient and sensitive odor analysis technology (He et al., 2021, Li et al., 2022, Yang et al., 2020). Gas chromatography-mass spectrometry (GC–MS) has relatively matured over the years and is currently the most commonly used analysis technique for odorants in tea. It has the largest variety of columns for selection, and features the benefits of rapid separation with both qualitative and quantitative performances (Yin et al., 2022, Yun et al., 2021). As a new technology for the analysis of food flavors, gas chromatography-ion migration spectrometry (GC-IMS) combines the benefits of gas chromatography and ion migration spectrometry (IMS), and capitalizes on rapid, sensitive and simple pretreatment. Moreover, it supports detection and analysis in atmospheric pressure environments (Fan et al., 2020, Feng et al., 2021). At present, using a single analysis technique may lead to failure to detect some important compounds. Meanwhile, the combination of multiple analysis technologies can provide more comprehensive, reliable and scientific information, and is becoming a popular research trend (Feng et al., 2022, Qi et al., 2020, Wang et al., 2019).

In Pu-erh tea market, there is an old saying that “the older the better”, which means that Pu-erh tea with a longer storage time has a higher market value compared with newer tea (Wang, et al., 2018). Related studies also show that the quality of Pu-erh tea is positively correlated with the storage life (Lv et al., 2015), and long-term storage has a significant impact on the aroma. However, in order to pursue profits, fake and shoddy phenomena exist in the Pu-erh tea market, which seriously damages the interests of consumers. Therefore, it is necessary to understand the dynamic changes of volatile substances in Pu-erh tea during storage. In this study, the dynamic changes of volatile metabolites of Pu-erh tea with different storage years were comprehensively explored by combining GC-E-Nose, GC–MS and GC-IMS. Moreover, the key differential compounds and the key storage nodes responsible for Pu-erh tea with different storage periods were elucidated through multivariate statistical analysis. The results provide valuable information for distinguishing Pu-erh tea and also provide technical support for quality control of Pu-erh tea.

2. Materials and methods

2.1. Tea samples and chemicals

The experimental samples were provided by Yunnan Shuangjiang Mengku Tea Co., Ltd, mainly including one-storage-year sample (S1Y), two-storage-year sample (S2Y), three-storage-year sample (S3Y), and four-storage-year sample (S4Y). 20-mL headspace vials covered with 18 mm magnetic PTFE/silicone caps were purchased from Agilent Technologies Inc. (Palo Alto, CA, USA). A manual SPME fiber holder and fibers of divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS, 50/30 μm, 2 cm) were purchased from Supelco (Bellefonte, PA, USA). Ethyl decanoate was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd (Shanghai, China). The purified water was obtained by Hangzhou Wahaha Group Co., Ltd (Hangzhou, China).

2.2. Gas chromatography electronic nose analysis

The volatile fingerprints of Pu-erh tea samples with a different number of storage years were analyzed by GC-E-Nose (Α M.O.S., Toulouse, France), as described in our previous work (Yang et al., 2020). In brief, 0.5 g of tea sample was weighed into a 20-mL sealed glass vial. High-purity helium was used as the carrier gas at a flow rate of 1 mL/min. 5000 μL of headspace gas was injected into the system at 200 °C, with an injection pressure of 10 kPa. The volatile compounds were absorbed by an embedded odor concentrator (Tenax TA) at 20 °C for 27 s, with a split mode of 10 mL/min, and thermal desorption was performed at 240 °C for 35 s. The temperature was programmed as follows: the initial temperature was 50 °C for 5 s, increased to 80 °C at 0.1 °C/s, then raised to 250 °C at 0.4 °C/s and held for 10 s. The acquisition time was 740 s. The detector temperature was 260 °C, and the gain factor of two flame ionization detectors (FID) was 12.

2.3. Gas chromatography-mass spectrometry analysis

The volatile compounds of Pu-erh tea samples were analyzed by an Agilent 7890B GC system coupled with an Agilent 7000C series triple quadrupole system (Agilent Technologies, Palo Alto, CA, USA). A non-polar capillary column (DB-5MS, 30 m × 0.25 mm × 0.25 μm; Agilent Technologies, Palo Alto, CA, USA) was applied for separation. In brief, 0.5 g of tea sample was placed in the 20-mL headspace vial, 5 μL of ethyl decanoate (100 mg/L, internal standard) was added, followed by 5 mL of pure water, and the vial was sealed. Then, the DVB/CAR/PDMS fiber was inserted into the headspace vial, and it was incubated at 60 °C. After extraction for 60 min, the fiber was inserted into the GC–MS injector for thermal desorption (about 5 min). The initial column temperature was kept at 40 °C for 5 min, then increased to 160 °C at 4 °C/min, held for 2 min, and finally increased to 270 °C at 10 °C/min, holding for 12 min. High-purity helium (＞99.999%) was used as carrier gas with a flow rate of 1 mL/min. The splitless mode was selected. The mass spectrometry was operated in an electron ionization (EI) mode at 70 eV, with a mass range of 40–450 m/z. The temperature of ionization source and transmission line were 230 °C and 250 °C, respectively.

The volatile compounds were determined by MassHunter Workstation Software Unknowns Analysis, and searched by the NIST 11 library according to the principle of similarity>80%. The retention index (RI) of each compound was calculated according to the linear formula of n-alkanes (C7-C40) by comparing with the values in the literature (https://webbook.nist.gov/chemistry/and https://www.flavornet.org/flavornet.html). The volatile components were quantified using the following formula:

Ci=Cki∗AiAki (1)

where Ci denotes the concentration of volatile components to be measured, μg/L; Ai denotes the peak area of volatile components to be measured; Cki denotes the concentration of ethyl decanoate (internal standard), μg/L; Aki is the peak area of ethyl decanoate.

2.4. Gas chromatography-ion mobility spectrometry analysis

The volatile compounds in Pu-erh tea samples were detected by a GC-IMS instrument (Flavourspec®, G.A.S, Dortmund, Germany). Briefly, 1 g of tea sample was placed in a 20-mL headspace vial, and incubated at 60 °C for 20 min with an agitation speed of 500 rpm. Then, 500 μL of headspace gas was added into the injection port via a heated syringe at 85 °C. A polar capillary column (WAX, 30 m × 53 mm × 1 μm, RESTEK Company, USA) was used for the separation. The column temperature was set at 60 °C, and the ion mobility spectrum temperature was set at 45 °C. High-purity nitrogen was employed as carrier gas at the following programmed flow: 2 mL/min for 2 min, raised to 10 mL/min within 10 min, then raised to 100 mL/min within 20 min and held for 20 min, with a drift gas flow rate of 150 mL/min.

The RI of the volatile compounds in Pu-erh tea samples was calculated using n-ketones C4–C9 as external references. The volatile compounds were qualitatively characterized by comparing the RI and drift time with those of standard substances in the user-built database, as well as the NIST library and IMS database. The preliminary choice and analysis of the GC-IMS data was performed by Laboratory Analytical Viewer (LAV). The fingerprints and differential spectrograms were analyzed by the Gallery Plot plug-in and the Reporter plug-in, respectively.

2.5. Statistical analysis

All tea samples were tested three times in parallel. Partial least squares discriminant analysis (PLS-DA) and orthogonal PLS-DA (OPLS-DA) were performed using SIMCA-P 14.1 software (Umetrics, Sweden). The significance levels among Pu-erh tea samples with different storage years were analyzed by one-way analysis of variance (ANOVA) using SPSS statistics 24.0 software (SPSS Inc., Chicago, IL, USA). The pie chart, histogram and box-plots were plotted by Origin software (Originlab, USA).

3. Results and discussion

3.1. The volatile fingerprints analyzed by gas chromatography electronic nose

Traditional electronic noses are mostly based on metal oxide sensors, which are easily polluted by different samples and subsequently fail to obtain complete and accurate information by each sensor. As a new odor analysis technique, GC-E-Nose combines the advantages of gas chromatography and electronic nose, with the merit of sensitive detection and rapid analysis. The instrument contains two columns (weak-polarity column MXT-5 and medium-polarity column MXT-1701), which can effectively separate volatile components with different polarity (Yang et al., 2021). In this study, the odor information of Pu-erh teas stored for a different number of years was analyzed and collected by GC-E-Nose. The radar maps obtained from MXT-5 and MXT-1701 columns in parallel were shown in Fig. S1. Each peak represents a specific volatile component, and the peak area represents the level of the volatile component. It could be seen that the peak intensities of volatile compounds detected by the MXT-5 column were greater than those by MXT-1701 column in all samples. Moreover, the peak intensities showed different trends with the extension of storage time, indicating that storage time had an important impact on the volatile profiles of Pu-erh teas.

Next, PLS-DA was conducted to further characterize the changes during the storage process. As a supervised discriminant method, PLS-DA can effectively solve the high correlation between variables. As illustrated in Fig. S2A, good model parameters (R2Y = 0.992, Q2 = 0.968) were obtained, indicating that the model had an excellent explanatory ability and predictive ability. The score of PLS-DA analysis revealed that the samples could be divided into four clusters based on the volatile fingerprints resulting from four different storage times. Specifically, S1Y, S2Y and S3Y could be clearly distinguished by the principal component one (PC1), while the difference between S4Y and other samples could be separated by the principal component two (PC2). In order to assess the robustness of the model, 200 permutation tests were conducted. The criteria for model validity are as follows: The regression line of the Q2-points intersects the vertical axis (on the left) at, or below zero. The parameters (R2 = 0.396, Q2 =  − 0.342) indicated that the model was robust and showed no overfitting (see Fig. S2B). Based on the above analysis, GC-E-Nose combined with multivariate statistical analysis could quickly distinguish Pu-erh teas stored for a different number of years.

Fig. 2.

Fig. 2

Open in a new tab

The OPLS-DA results of Pu-erh teas with different storage years by using GC–MS. (A) Scores plots of OPLS-DA (R2Y = 0.729, Q2 = 0.527); (B) Cross-validation plot by 200 permutation tests (R2 = 0.498, Q2 =  − 0.86); (C) The metabolic trajectory plot; (D) The red part represents 15 key compounds with VIP > 1.

3.2. Analysis of Pu-erh teas with different storage years by GC–MS

3.2.1. The volatile components analyzed by GC–MS

A total of 43 volatile components were identified by GC–MS, which were classified into 8 categories, including 10 aldehydes, 8 ketones, 6 alcohols, 6 heterocyclic compounds, 4 hydrocarbons, 3 methoxy-phenolic compounds, 2 esters, and 4 other compounds. The detailed information of volatile components was listed in Table S1. Among them, the four component types with higher proportions were aldehydes (23.26%), ketones (18.60%), alcohols (13.95%) and heterocyclic compounds (13.95%) (see Fig. 1A). In addition, the contents of volatile component categories varied between different storage stages (see Fig. 1B). The content of heterocyclic compounds in S1Y (145.78 μg/L) was the highest, followed by S2Y (123.68 μg/L), S4Y (89.21 μg/L), and S3Y (55.88 μg/L). Heterocyclic compounds such as furan, pyrrole and their derivatives have been reported to be generated by Maillard reaction during the manufacturing process of fixation and drying (Yang, Xie, et al., 2022). It is worth mentioning that 2-ethyl-pyridine, which has been detected only in Pu-erh tea, provided green and grassy odors (Ma et al., 2021).

Fig. 1.

Fig. 1

Open in a new tab

The volatile compounds in Pu-erh teas with different storage years obtained from GC–MS. (A) Proportion of different classes of volatile compounds; (B) Content comparison of different classes of volatile compounds. Values with different letters are significantly different (p < 0.05).

Four hydrocarbons were identified in four storage years of Pu-erh tea, one of which was unsaturated hydrocarbon, which played an important role in tea aroma, while saturated hydrocarbons had little effect on tea aroma. The contents of hydrocarbon components were in the order of S2Y (110.90 μg/L) > S1Y (97.67 μg/L) > S3Y (93.94 μg/L) > S4Y (48.48 μg/L), and those in S4Y were significantly lower than other groups (p < 0.05).

The contents of alcohols in S2Y were the highest with up to 98.51 μg/L, followed by S1Y (89.35 μg/L), S4Y (43.58 μg/L) and S3Y (35.54 μg/L). Linalool was reported to contribute significantly to tea aroma (Mao et al., 2018, Pang et al., 2019). The concentration of linalool was the highest in S1Y (4.59 μg/L), significantly higher than S3Y (0.63 μg/L) and S4Y (1.32 μg/L) (p＜0.05). Phenylethyl alcohol was an important volatile component released from the hydrolysis reaction of glycosides or shikimate synthesis pathway, and was linked to the aromas of sweet flowers, fresh bread and rose (Su, Xia, Gao, Dai, & Zhang, 2010). Its content reached the maximum level in S2Y (37.00 μg/L) and dropped to its lowest level in S3Y (3.65 μg/L). L-α-terpineol, contributing to the tree, lilac and woody odors, decreased gradually with the number of storage years.

Methoxy-phenolic compounds played an important role in the unique “stale” aroma of Pu-erh tea, which were the special compounds different from green tea, black tea, and others (Lv, Zhang, Yang, Shi & Lin, 2015). These volatile components originated from microbial and thermal degradation during the pile fermentation process. In this study, the contents of methoxy-phenolic compounds in Pu-erh tea gradually decreased with the rising number of storage years, but reached the highest level in S2Y (75.93 μg/L). Among them, 1,2,3-trimethoxybenzene and 1,2,4-trimethoxybenzene have been reported to be the crucial odorants (Lv et al., 2012, Pang et al., 2019).

As for aldehydes, their contents reached the maximum level in S2Y (30.32 μg/L), while they reached the minimum level in S3Y (15.94 μg/L). For example, benzeneacetaldehyde with sweet and rose aromas was reported to be an essential product of the shikimate pathway (Chen et al., 2019). Hexanal and (E)-2-hexenal was reported to contribute to green and grass fragrances, which were usually formed by the degradation of fatty acids (Yang, Baldermann, & Watanabe, 2013).

Usually, ketones make an essential contribution to the aroma of tea because of their low threshold values. For example, 3-methyl-2(5H)-furanone has a strong caramel flavor in tea (Lv et al., 2014). α-Ionone with a relatively low odor threshold (0.4 μg/L in water) was described as presenting sweet and floral odors (Zhai, Zhang, Granvogl, Ho, & Wan, 2022). Among the four storage years of Pu-erh tea, S2Y possessed the highest content of ketones (15.63 μg/L). The others followed the order of S1Y (12.75 μg/L) > S4Y (6.85 μg/L) > S3Y (5.82 μg/L).

The content of esters in S2Y (10.06 μg/L) was the highest, followed by S1Y (8.27 μg/L), S4Y (4.20 μg/L) and S3Y (2.59 μg/L). Methyl salicylate is recognized as a vital volatile compound by providing sweet and floral fragrances, and is derived from the oxidation of α-linolenic acid or hydrolysis reaction of β-Primeveroside (Wang et al., 2011). (R)-5,6,7,7a-tetrahydro-4,4,7a-trimethyl-2(4H)-benzofuranone with sweet, coumaric and coconut fragrance, reached the highest level in S2Y (9.00 μg/L) and dropped to the lowest in S3Y (2.27 μg/L).

The above results indicated that the contents of volatile components of Pu-erh samples stored for a different number of years varied considerably, that is, storage time had a great influence on the flavor of Pu-erh tea.

3.2.2. Multivariate statistical analysis

OPLS-DA analysis was conducted to determine the volatile components causing the aroma differences of Pu-erh tea among the four storage years. As shown in Fig. 2A, a clear separation could be observed through a dependable OPLS-DA model. Specifically, S1Y and S2Y were located in the fourth and first quadrants, respectively while S3Y and S4Y were located in the second and third quadrants, respectively. And the S3Y and S4Y samples almost coincided, indicating that their volatile metabolites were relatively similar. The model parameters (R2Y = 0.729, Q2 = 0.527) showed that it had good interpretation and prediction performance. In addition, 200 iterations of permutation test showed no over-fitting (R2 = 0.498, Q2 =  − 0.86) (see Fig. 2B). The metabolic trajectory plot can reflect the changing trend of metabolites. Generally, the greater the distance between the two, the more drastic the change of metabolites. As shown in Fig. 2C, the distance between S3Y and S2Y was the furthest, indicating that the storage time of 3 years was a critical transition point. Variable Importance in Projection (VIP) was employed to evaluate the influence strength and explanatory ability of each variable on the classification and discrimination. The variable was considered to play an important role when VIP exceeded 1. In this study, a total of 15 compounds with VIP > 1 were screened out, which were mainly aldehydes and esters (see Fig. 2D). The representative volatile compounds mainly included 4-ethyl-phenol, 1-methyl-1H-pyrrole-2-carboxaldehyde, 1-ethyl-1H-Pyrrole-2-carboxaldehyde, methyl salicylate, nonanal, 1,2,4-trimethoxybenzene, (R)-2(4H)-5,6,7,7a-tetrahydro-4,4,7a-trimethyl-benzofuranone, α-ionone, dimethyl ether, (E)-2-nonenal, linalool, dehydromevalonic lactone, 2-methyl-butanal, hexanal, and phenylethyl alcohol.

3.3. Analysis of Pu-erh teas with different storage years by GC-IMS

3.3.1. The topographic plots in Pu-erh tea with different storage years

As a new analytical technique, GC-IMS was applied to obtain the global IMS information of Pu-erh tea samples, with the advantages of extraordinary sensitivity, high separation ability, easy operation, and visualization of flavor substances (Guo et al., 2021, Yang et al., 2022). This technique could separate and identify ionized compounds at ambient pressures based on the difference of migration rates in an electric field. The topographic plots obtained from GC-IMS analysis in this study were shown in Fig. 3A. The vertical coordinate represents the retention time of gas chromatography, and the horizontal coordinate represents the ion migration time. The red vertical line at abscissa 1.0 represents the reactive ion peak (RIP) after normalization. Each point on both sides of the RIP peak represents a volatile component, and the color represents the content of a volatile component. White color represents a lower content while red represents a higher content. It could be found that the types of volatile components of the different storage samples were similar, mainly reflected in the content difference.

Fig. 3.

Fig. 3

Open in a new tab

Fingerprints of Pu-erh teas with different storage years obtained from GC-IMS. (A) Topographic plots; (B) Difference comparison plots (S1Y as the reference).

In order to observe the differences more clearly, the spectral diagram of S1Y was chosen as the reference, and the other samples were deducted as the reference. Red signified that the content of a volatile component was higher than the reference, whereas blue signified that the content of a volatile component was lower than the reference. As can be seen in Fig. 3B, most of the signal points were shown as a retention time range of 0 to 1500 s and a drift time range of 1.0 to 2.0. The concentrations of some volatile compounds increased significantly with the increase in storage years and reached the highest level in S3Y. This phenomenon indicates that the contents of volatile components of Pu-erh tea were significantly different based on the storage time.

3.3.2. Volatile compounds in Pu-erh teas stored for different periods

In order to further observe the variation regularity in specific volatile compounds throughout the storage process, the volatile components were qualitatively characterized by comparing the retention time and drift time with those of the authentic reference compounds. A total of 91 known volatile substances (corresponding to 109 peak signals) were identified, which were divided into 8 categories, including 19 alcohols, 21 aldehydes, 18 ketones, 9 heterocyclic compounds, 14 esters, 4 hydrocarbons, 3 sulfur compounds, and 3 other compounds (Table 1). Aldehydes (23.08%), alcohols (20.88%) and ketones (19.78%) were the main volatile components, consistent with the results of GC–MS (Fig. 4A). It is worth mentioning that some single volatile compounds might produce multiple signals of their dimers in the ionization region due to the concentrations of volatile components.

Table 1.

The information of identified volatile compounds by GC-IMS.

No. Compounds MW RI Rt (sec) Dt (RIP relative) Peak intensities
 VIP
S1Y S2Y S3Y S4Y
1 α-Terpineol 154.3 1732.7 2106.033 1.20441 916.79±86.87a 787.49±43.13a 693.19±157.71a 826.66±186.80a 0.513
2 Terpinen-4-ol 154.3 1640.5 1724.156 1.22738 677.44±125.11b 512.10±21.80b 536.33±193.81b 1714.64±128.26a 1.528
3 Linalool 154.3 1560.3 1448.81 1.23197 626.81±86.80b 357.45±25.29c 377.93±93.12c 841.12±125.22a 1.247
4 1-Octen-3-ol 128.2 1483.8 1227.125 1.16532 146.80±10.35a 120.85±22.63a 108.86±34.08a 122.69±34.00a 0.514
5 1-Hexanol-M 102.2 1367.4 953.192 1.33114 135.22±21.90b 129.86±14.59b 417.49±16.48a 136.04±25.25b 0.902
6 Cis-2-Penten-1-ol 86.1 1335.2 888.778 0.94607 1193.26±41.73a 634.45±37.19d 951.52±41.92c 1062.05±61.99b 1.068
7 Pentanol-M 88.1 1264 763.941 1.25615 481.23±18.44a 260.16±6.87c 502.74±21.88a 396.10±8.02b 1.080
8 Pentanol-D 88.1 1265 765.514 1.50954 38.63±5.91a 30.37±0.64a 45.18±13.68a 34.08±8.37a 0.642
9 Linalool oxide-M 170.3 1488.8 1240.392 1.26991 384.63±12.12a 229.51±39.05b 234.45±27.10b 286.24±38.83b 0.858
10 3-Methyl-1-butanol-M 88.1 1218.7 695.497 1.23944 1399.08±61.66b 1160.16±117.05c 1848.55±42.53a 1359.35±105.71b 0.968
11 3-Methyl-1-butanol-D 88.1 1220.3 697.857 1.49005 270.52±21.28b 181.69±21.56c 456.28±19.55a 263.87±42.19b 0.983
12 1-Penten-3-ol-M 86.1 1175.4 622.845 0.94022 4314.71±28.71a 2897.27±72.23d 4075.46±78.39b 3789.26±62.27c 1.081
13 1-Penten-3-ol-D 86.1 1175.4 622.845 1.37749 925.03±53.89a 228.76±5.40d 586.31±47.24b 435.03±32.79c 1.071
14 1-Butanol-M 74.1 1158.8 589.405 1.18231 1971.69±19.70c 2468.66±62.70b 2641.35±10.39a 1647.28±22.06d 1.186
15 1-Butanol-D 74.1 1159.4 590.599 1.38052 726.20±36.47c 1188.76±66.29b 1507.61±9.44a 459.17±18.22d 1.096
16 2-Methyl-1-propanol-M 74.1 1107.6 497.36 1.17483 685.80±4.17b 716.84±27.41b 1186.50±24.16a 623.00±22.79c 0.912
17 2-Methyl-1-propanol-D 74.1 1106.9 496.136 1.37214 71.56±0.10b 55.60±4.99c 146.90±7.87a 55.98±5.90c 0.941
18 1-Propanol-M 60.1 1051.5 419.614 1.11146 992.23±22.46b 716.07±31.25c 2709.60±8.61a 593.69±6.23d 0.947
19 1-Propanol-D 60.1 1051 419.002 1.25548 140.85±1.77b 116.28±7.97c 878.15±23.87a 97.59±5.75d 0.913
20 2-Butanol-M 74.1 1036.4 401.196 1.14949 354.42±8.78b 301.43±18.01c 481.32±14.39a 373.26±32.58b 1.022
21 2-Butanol-D 74.1 1035.6 400.236 1.32496 556.31±21.35a 305.37±19.83c 314.58±18.19c 429.54±14.20b 0.889
22 Ethanol 46.1 940.8 314.289 1.13231 1849.63±135.44b 1513.90±66.66c 4030.40±128.67a 1617.91±96.26c 0.930
23 tert-Butanol 74.1 920.5 300.274 1.32561 3267.28±47.26a 2213.70±43.10b 2236.94±103.04b 2340.16±109.88b 1.048
24 3-Methyl-3-buten-1-ol 86.1 1271.2 775.495 1.19675 102.32±10.49a 82.97±4.46ab 78.87±6.88b 87.38±15.00ab 0.689
25 2-Octanol-M 130.2 1444.2 1126.068 1.42754 5971.70±218.94c 6459.46±182.54b 7028.30±49.89a 7163.88±88.84a 1.398
26 2-Octanol-D 130.2 1444.8 1127.445 1.82345 2232.43±196.08c 2502.23±156.57b 3119.17±55.23a 3293.79±66.69a 1.465
27 (Z)-3-Hexenol-M 100.2 1400.7 1024.621 1.23628 158.11±15.42ab 146.34±11.35ab 138.72±19.77b 205.88±57.08a 1.043
28 Benzaldehyde-M 106.1 1547.6 1409.475 1.15618 805.31±44.09a 611.61±16.46b 337.58±66.66c 602.55±84.46b 0.883
29 (E)-2-Hexenal-M 98.1 1230 712.018 1.18096 2675.08±73.38b 2070.46±97.25d 2221.99±12.14c 2856.72±50.21a 1.222
30 (E)-2-Hexenal-D 98.1 1229.5 711.231 1.51929 1664.03±134.33b 830.08±59.34d 1006.26±21.76c 1963.53±74.65a 1.236
31 Octanal 128.2 1294.4 813.603 1.4117 398.76±10.09b 464.70±3.75a 408.61±28.69b 423.13±16.48b 0.947
32 2-Hexenal-M 98.1 1212.7 686.98 1.18122 160.16±6.85b 125.28±5.57d 136.87±0.81c 176.39±6.92a 1.284
33 3-Methyl-2-butenal 84.1 1211.8 685.666 1.0961 254.46±4.45a 210.13±20.36b 196.12±9.33b 235.50±14.18a 0.765
34 Heptanal-M 114.2 1194.9 662.154 1.33514 1038.49±48.05b 944.42±40.27c 615.05±9.68d 1149.87±10.27a 0.971
35 Heptanal-D 114.2 1194.9 662.154 1.6966 129.72±17.51b 101.01±11.49c 50.42±7.64d 159.97±10.50a 1.032
36 Hexanal-M 100.2 1097.4 480.831 1.26412 3966.58±32.58a 3410.04±25.56c 3185.27±83.70d 3857.74±36.20b 0.885
37 Hexanal-D 100.2 1097.4 480.831 1.56224 5269.79±199.40a 2602.96±78.30c 2012.88±181.70d 4496.21±106.80b 0.881
38 Pentanal-M 86.1 1002 362.304 1.18998 881.65±29.01b 1003.66±4.13a 682.49±19.83c 1002.14±11.39a 1.010
39 Pentanal-D 86.1 1002 362.304 1.42435 1337.48±79.50a 867.26±96.35c 227.50±17.13d 1040.09±11.26b 0.836
40 3-Methylbutanal 86.1 924.9 303.246 1.40472 7051.17±225.79c 10244.57±13.56a 7523.77±292.31b 10120.71±169.31a 1.189
41 Butanal 72.1 885.2 277.318 1.28078 1443.48±64.23a 932.78±78.56c 672.93±38.07d 1131.46±21.32b 0.838
42 Propanal 58.1 812.2 235.229 1.14226 6943.95±180.98a 5142.64±254.44c 4118.84±167.61d 5930.79±72.08b 0.832
43 2-Methylpropanal 72.1 825.5 242.395 1.28209 974.80±69.79b 1313.53±39.92a 840.02±86.11b 1292.87±100.41a 1.067
44 (Z)-2-Pentenal 84.1 1126 528.659 1.11217 106.12±6.09a 65.81±5.06c 69.79±5.04c 90.22±2.10b 0.890
45 cis-4-Heptenal 112.2 1253.9 748.091 1.15001 646.98±39.65a 269.03±17.04c 285.42±5.75c 496.75±4.65b 0.898
46 2-Methylbutanal 86.1 953.2 323.183 1.40723 271.42±50.41a 271.19±46.82a 49.29±6.39b 273.98±43.84a 0.856
47 (E)-2-Pentenal 84.1 1147.6 568.039 1.36016 1194.84±61.48a 520.60±37.51d 637.84±7.46c 959.88±40.35b 0.935
48 (E, E)-2,4-Heptadienal 110.2 1515.9 1315.616 1.20425 281.87±13.10a 128.98±27.16c 93.85±9.50d 170.06±17.17b 0.896
49 Nonanal 142.2 1401.2 1025.804 1.48396 417.54±15.94b 524.57±29.35a 345.36±43.82c 565.34±14.71a 1.134
50 Acrolein 56.1 876.1 271.653 1.06297 553.02±18.99b 595.75±37.97ab 648.71±34.81a 601.99±49.83ab 0.720
51 Phenylacetaldehyde 120.2 1763.9 2253.641 1.25786 422.38±37.21a 435.22±120.94a 328.95±100.89a 398.85±137.05a 0.438
52 (E)-2-Nonenal 140.2 1536.7 1376.348 1.42632 311.67±40.46b 296.25±24.27b 466.36±53.55a 191.92±47.18c 1.055
53 6-Methyl-5-hepten-2-one 126.2 1347.3 912.554 1.17885 445.64±13.15b 409.21±33.16c 592.57±39.88a 495.55±46.21b 1.022
54 Acetoin 88.1 1296.9 818.043 1.06398 222.39±12.57a 156.13±8.54b 171.50±13.30b 164.36±8.17b 1.054
55 1-Hydroxy-2-propanone-M 74.1 1311.3 844.024 1.06166 1030.97±105.36a 461.91±6.60c 467.06±25.38c 794.54±69.05b 0.874
56 1-Hydroxy-2-propanone-D 74.1 1312.4 845.956 1.23458 181.43±13.39a 69.12±10.26d 114.11±4.40c 135.82±7.50b 0.977
57 2-Heptanone-M 114.2 1190.7 655.402 1.26445 751.27±8.88a 740.56±40.21a 772.17±26.28a 751.88±23.30a 0.475
58 2-Heptanone-D 114.2 1191.4 656.818 1.63184 189.21±8.15a 185.11±21.54a 197.15±13.58a 220.49±7.71a 1.276
59 4-Methyl-3-penten-2-one-M 98.1 1149.8 572.088 1.12027 1972.90±33.28a 1398.77±42.22c 1964.99±42.75a 1788.93±33.87b 1.095
60 4-Methyl-3-penten-2-one-D 98.1 1149.5 571.569 1.44526 105.19±7.07a 53.25±5.82b 69.06±10.42b 63.52±13.32b 1.008
61 Cyclopentanone 84.1 1143.8 560.766 1.33168 4054.91±266.69b 1650.08±236.75d 4607.57±49.98a 3471.27±15.12c 1.107
62 2-Hexanone 100.2 1120 518.174 1.20363 47.55±3.93a 34.62±7.21b 32.15±1.36b 28.96±4.77b 1.171
63 1-Penten-3-one 84.1 1040.4 405.998 1.07955 842.23±17.81a 545.77±39.12c 479.72±10.97d 698.50±11.31b 0.850
64 3-Pentanone 86.1 1026.6 389.673 1.11268 48.69±2.25b 30.75±5.55c 62.17±3.38a 28.75±3.83c 1.078
65 2-Pentanone 86.1 996.1 356.062 1.35809 1633.97±64.07b 987.37±108.53d 1783.70±80.02a 1390.23±38.67c 1.061
66 Diacetyl 86.1 990.8 351.741 1.17035 515.56±4.17a 522.79±33.59a 258.52±5.78b 526.02±16.94a 0.903
67 2-Butanone 72.1 911.4 294.123 1.24643 10524.34±128.99a 4082.73±198.99c 5227.79±365.65b 4747.29±394.37b 1.110
68 Acetone 58.1 835.7 248.029 1.11513 12345.51±191.31a 9996.29±150.45b 10125.02±324.65b 10478.78±315.39b 0.990
69 2-Octanone-M 128.2 1292.8 810.848 1.33785 2122.51±73.56d 2272.53±59.58c 2956.14±54.33a 2636.06±34.20b 1.162
70 2-Octanone-D 128.2 1292.9 811.056 1.75355 606.67±23.90d 715.42±28.82c 1314.67±33.39a 995.73±3.26b 1.124
71 Acetophenone 120.2 1688.4 1912.864 1.18817 1835.94±89.17a 594.15±98.55c 453.61±35.23c 780.75±73.20b 0.997
72 3-Methyl-2-pentanone 100.2 1066.6 438.779 1.1815 149.12±6.92a 158.76±11.10a 130.71±2.26b 149.31±5.76a 0.885
73 4-Methyl-2-pentanone 100.2 1023.3 385.955 1.17961 197.75±7.12b 187.63±3.47b 141.34±5.12c 223.90±12.66a 1.036
74 Cyclohexanone 98.1 1293.7 812.495 1.15813 151.05±1.34b 212.93±24.99a 143.67±12.07b 111.56±17.29c 1.377
75 2-Acetylpyridine 121.1 1636.1 1707.766 1.11714 2905.60±157.15a 2044.02±222.31b 1087.18±60.31c 2090.82±79.38b 0.892
76 2-Pentylfuran 138.2 1239.5 726.179 1.25197 199.76±14.48a 118.85±6.70c 200.05±7.91a 180.86±7.07b 1.102
77 2-Isopropyl-3-methoxypyrazine 152.2 1453.2 1148.289 1.26683 477.11±16.63b 404.89±29.97c 471.64±17.71b 540.69±32.55a 1.388
78 2,3-Dimethylpyrazine 108.1 1349.4 916.75 1.10501 195.63±13.70a 177.66±10.44a 181.01±17.83a 181.83±17.20a 0.504
79 Pyrazine 80.1 1194.3 661.31 1.04571 2514.41±129.77a 1126.42±75.24c 1287.32±3.46bc 1418.29±120.03b 1.018
80 2-Ethylfuran 96.1 966.5 333.015 1.04519 1107.34±30.78b 717.73±20.40d 1332.35±27.88a 847.17±31.81c 1.050
81 2-Methylpyrazine 94.1 1274.1 780.036 1.10196 169.30±4.92a 113.31±19.37b 154.27±5.80a 167.24±9.99a 1.113
82 3-Ethylpyridine 107.2 1384.1 988.417 1.09814 134.44±10.41a 117.85±9.98a 107.14±19.02a 132.59±18.92a 0.601
83 Tetrahydrofuran 72.1 876.1 271.653 1.22321 235.62±13.74a 175.30±19.23bc 150.89±11.73c 191.79±19.95b 0.814
84 γ-Butyrolactone 86.1 1705.8 1986.389 1.08958 850.14±21.59a 544.97±29.36c 432.93±93.62c 706.90±74.51b 0.805
85 Butyl acetate-M 116.2 1084.3 462.466 1.24108 1062.30±13.91a 735.66±46.24b 92.23±8.93c 752.19±48.65b 0.879
86 Ethyl butanoate 116.2 1060.9 431.445 1.19857 200.09±21.21a 103.01±2.75c 77.37±4.50d 138.85±8.94b 0.856
87 Methyl butanoate-M 102.1 1013 374.308 1.16299 540.00±36.75a 379.12±42.70b 351.44±27.66b 494.58±20.86a 0.844
88 Methyl butanoate-D 102.1 1013 374.308 1.45625 65.58±5.17a 44.91±2.97c 52.24±9.25bc 58.38±2.03ab 0.852
89 Diethyl acetal 118.2 899.6 286.441 1.02924 1029.63±43.25c 1628.08±70.54b 2491.08±68.86a 1551.06±28.69b 0.873
90 Ethyl Acetate 88.1 893.6 282.6 1.34091 7195.51±115.37a 6830.01±273.88b 860.52±44.66d 6475.90±147.24c 0.902
91 Methyl acetate 74.1 848.4 255.234 1.19997 444.96±32.32b 551.25±10.97a 280.33±27.08c 543.01±33.28a 0.994
92 Propyl acetate 102.1 989 350.357 1.47598 124.47±5.61a 105.65±13.29a 18.57±2.81b 107.79±13.95a 0.862
93 Ethyl formate 74.1 826.2 242.791 1.19639 1285.79±22.45b 1355.00±21.15ab 1072.92±59.81c 1419.77±76.40a 0.991
94 Dihydro-5-methyl-2(3H)-furanone 100.1 1735.3 2117.807 1.12836 1125.28±102.29a 811.94±45.26b 879.04±59.49b 845.06±99.08b 0.991
95 1-Methoxy-2-propyl acetate 132.2 1237.2 722.756 1.14162 276.83±22.58b 196.56±19.55c 315.56±12.63a 214.67±11.11c 1.032
96 Butyl acetate-D 116.2 1086.2 465.09 1.61886 155.82±6.25a 62.98±7.12b 11.41±3.15c 71.65±8.94b 0.942
97 Ethyl propanoate 102.1 965.8 332.52 1.46303 122.03±1.75a 69.65±1.52b 46.66±6.38c 64.16±4.47b 1.075
98 2-Methylbutyl acetate 130.2 1153.3 578.852 1.28631 195.13±6.82a 108.41±5.11b 61.41±6.60d 74.54±6.75c 1.213
99 β-Pinene 136.2 1132.5 540.212 1.21515 54.10±5.53b 54.10±7.16b 105.40±3.18a 58.82±8.13b 0.891
100 α-Pinene 136.2 1030.7 394.474 1.18876 352.84±14.90b 391.60±8.87a 319.14±6.47c 391.30±8.77a 1.034
101 α-Phellandrene 136.2 1166.8 605.393 1.2376 304.40±8.35b 357.29±27.02a 295.79±16.93b 300.27±27.12b 1.029
102 Camphene 136.2 1093.4 475.029 1.18993 329.80±5.09d 458.30±10.61a 438.27±10.82b 374.15±10.34c 0.928
103 Acetic acid-M 60.1 1501.5 1275.22 1.05656 19181.60±204.69a 14537.82±527.90b 14055.54±197.53b 18929.03±32.50a 0.986
104 Acetic acid-D 60.1 1501.5 1275.22 1.15696 11114.35±772.25a 3845.19±422.38c 3401.38±165.49c 10261.99±45.16b 0.968
105 p-Xylene 106.2 1143 559.416 1.07488 331.37±10.68a 316.96±5.87b 239.45±5.03d 296.83±1.55c 0.976
106 2-Butoxyethanol 118.2 1413.8 1054.134 1.23657 176.47±31.23b 462.00±24.13a 99.29±17.51c 117.95±26.90c 1.284
107 2-Methylthiophene 98.2 1060.9 431.445 1.03783 1068.94±44.87a 785.84±16.33c 704.49±11.54d 929.95±21.80b 0.841
108 Thiophene 84.1 1027.4 390.633 1.03905 1830.97±24.50b 1396.30±54.68c 2202.56±22.24a 1343.00±23.44c 1.094
109 Dimethyl sulfide-M 62.1 788.5 222.988 0.95914 510.05±9.89a 307.37±15.73b 302.63±15.81b 323.94±11.72b 1.065
Open in a new tab

Note: MW: Molecular weight; RI: Retention index; Rt: Retention time; Dt: Drift time; RIP: Reactive ion peak; VIP: Variable importance in projection; Suffix M represented the monomer of volatile compound and suffix D represented the dimer of volatile compound; S1Y, S2Y, S3Y and S4Y represented four Pu-erh samples with different storage years, respectively; Values with different letters in a row indicated significant differences using Duncan’s multiple comparison tests (p < 0.05).

Fig. 4.

Fig. 4

Open in a new tab

The information of volatile compounds in Pu-erh teas with different storage years by using GC-IMS. (A) Proportion of different classes of volatile compounds; (B) Volatile fingerprints generated by Gallery Plot.

In order to directly reflect the content difference of volatile components in Pu-erh teas with different storage years, the volatile fingerprints were constructed and analyzed. Each column was presented on behalf of a volatile compound in different samples, and each row represented the signal peak of one sample. The color was on behalf of the content of a volatile component, and the brighter the color, the higher content. As shown in Fig. 4B, the fingerprint could be divided into four regions of A, B, C and D. The volatile components in region A reached the maximum concentration in S1Y, and were dominated by heterocyclic compounds and esters (such as pyrazine, 2-methylpyrazine, 2-mthylthiophene, 2-ethylfuran, 3-ethylpyridine, ethyl acetate, propyl acetate and ethyl propanoate, etc). Alkenes (such as α-pinene, camphene and α-phellandrene) dominated in B region, and reached the highest level in S2Y. In region C, ketones and alcohols were the main compounds, reaching the highest content in S3Y. The typical compounds included 6-methyl-5-hepten-2-one, cyclopentanone, 2-pentanone, pentanol monomer and dimer, and 1-butanol monomer and dimer, etc. Aldehydes were the main odorants in D region. The representative compounds were 2-methylbutanal, nonanal, phenylacetaldehyde and heptanal monomer and dimer, which increased slowly with the passage of storage time and reached their highest concentration in S4Y.

3.3.3. Multivariate statistical analysis

The fingerprint only roughly distinguished the volatile components in Pu-erh teas with different storage years. It was hard to precisely establish which volatile compounds made the difference. PLS-DA could effectively distinguish between the observed values in different groups and found the importance variables that led to the differences among all groups. The similarities and differences between samples could be visualized by the score plot of the model. As shown in Fig. 5A, the four groups (S1Y, S2Y, S3Y and S4Y) achieved a good separation. The fitting parameters of the PLS-DA model (R2Y = 0.991, Q2 = 0.966) indicated that it had strong explanatory and predictive ability. In addition, 200 replications of permutation test were performed to evaluate the robustness of the PLS-DA model. As presented in Fig. 5B, the intercept between the Q2 regression line was less than 0, proving that the model was reliable and there was no overfitting (R2 = 0.207, Q2 =  − 0.483). The key compounds responsible for the aroma profile differences between Pu-erh tea samples with different storage years were further analyzed by a load diagram (Fig. 5C). For example, tert-butanol (No.23), acetone (No.68) and dimethyl sulfide monomer (No.109) that appeared in higher concentrations in S1Y than other tea samples, have been reported to produce camphor, apple, pear, and creamy or vegetable odors. Octanal (No.31), with a citrus orange and green odor, presented high content in S2Y. Some volatiles such as 1-hexanol monomer (No.5), 1-propanol dimer (No.19), and β-pinene (No.99) had higher concentrations in S3Y than other groups, with a contribution to fruity, green, musty, and woody aromas. (Z)-3-Hexenol monomer (No.27) and phenylacetaldehyde (No.51), which occurred in higher concentrations in S4Y than others, provided green and floral aromas. To characterize the key differential compounds obtained from different storage years, the VIP was investigated. In this model, a total of 49 variables with VIP > 1 were screened out, and alcohols (10), ketones (12) and aldehydes (8) were more abundant (Fig. 5D). In addition, based on double variable criterion of one-way ANOVA (p < 0.05) and VIP > 1.2, nine key substances were picked out to distinguish the tea samples with different storage durations. The box-plots of the above nine key volatile components were shown in Fig. S3. The contents of nine key compounds could be divided into three groups. Group I was dominated by 2-octanol monomer and dimer (No.25 and No.26), with fresh grassy and earthen aroma. Its content increased gradually with the extension of storage time. The volatile compounds including cyclohexanone (No.74) and 2-butoxyethanol (No.106) in group II reached the highest level in S2Y, and declined sharply after S3Y. The volatile components in group III decreased gradually and reached the lowest level during storage for two or three years (S2Y or S3Y), and then increased sharply after S4Y. These representative compounds included linalool (No.3), (E)-2-hexenal (monomer and dimer) (No.29 and No.30), 2-hexenal monomer (No.32), 2-methylbutyl acetate (No.98), terpinen-4-ol (No.2) and 2-isopropyl-3-methoxypyrazine (No.77). Linalool was an important monoterpene alcohol which was widely exists in teas. It was mainly derived from the hydrolysis reaction of β-glucoside and β-primrose glucoside in tea (Ho, Zheng, & Li, 2015). As stereoisomers, (E)-2-hexenal and 2-hexenal were important volatile compounds in fresh leaves, contributing to the green and grassy aromas (Guo, Ho, Wan, Zhu, Liu, & Wen, 2021). Terpinen-4-ol was reported to present woody and earthy odor and was detected in oolong tea (Guo, Schwab, Ho, Song, & Wan, 2021). 2-Isopropyl-3-methoxypyrazine had a lower threshold (0.0039 μg/L in water), and was reported as an important heterocyclic compound with earthy and pea-like odor (Flaig, Qi, Wei, Yang, & Schieberle, 2020).

Fig. 5.

Fig. 5

Open in a new tab

The PLS-DA results of Pu-erh teas with different storage years by using GC-IMS. (A) Scores plots of PLS-DA (R2Y = 0.991, Q2 = 0.966); (B) Cross-validation plot by 200 permutation tests (R2 = 0.207, Q2 =  − 0.483); (C) Loading plot of PLS-DA; (D) The red part represents key differential compounds with VIP > 1.2, and compound numbers corresponded to Table 1.

4. Conclusions

In this study, the aroma profiles of Pu-erh teas stored for a different number of years were comprehensively characterized by combining GC-E-nose, GC–MS and GC-IMS. GC-E-nose achieved the rapid differentiation of Pu-erh teas with different storage years (R2Y = 0.992, Q2 = 0.968). A total of 43 volatile components were identified by GC–MS, while 91 volatile substances were detected by GC-IMS. A satisfactory discrimination was achieved by using PLS-DA based on the volatile fingerprints obtained from GC-IMS, with robust model parameters (R2Y = 0.991, and Q2 = 0.966). In addition, 9 flavor compounds were considered as important variables that caused the aroma differences of Pu-erh teas with different storage years, including linalool, (E)-2-hexenal, 2-hexenal, 2-methylbutyl acetate and terpinen-4-ol, cyclohexanone, 2-butoxyethanol, 2-octanol, and 2-isopropyl-3-methoxypyrazine. The overall information of volatile components of Pu-erh tea was retained to the greatest extent by the combination of GC-E-nose, GC–MS and GC-IMS, and the use of multivariate statistical analysis realized the rapid distinction of Pu-erh teas with different storage periods.

CRediT authorship contribution statement

Yuting Rong: Resources, Conceptualization, Methodology, Investigation. Jialing Xie: Methodology, Software, Validation, Writing – original draft. Haibo Yuan: Formal analysis, Visualization. Lilei Wang: Software. Fuqiao Liu: Resources. Yuliang Deng: Data curation. Yongwen Jiang: Funding acquisition, Supervision, Project administration. Yanqin Yang: Conceptualization, Methodology, Supervision, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the Science and Technology Innovation Project of the Chinese Academy of Agricultural Sciences (CAAS-ASTIP-TRICAAS).

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2023.100693.

Contributor Information

Yongwen Jiang, Email: jiangyw@tricaas.com.

Yanqin Yang, Email: yangyq@tricaas.com.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary data 1
mmc1.docx (1.3MB, docx)

Data availability

Data will be made available on request.

References

  1. Chen X., Chen D., Jiang H., Sun H., Zhang C., Zhao H.…Xu Z. Aroma characterization of Hanzhong black tea (Camellia sinensis) using solid phase extraction coupled with gas chromatography–mass spectrometry and olfactometry and sensory analysis. Food Chemistry. 2019;274:130–136. doi: 10.1016/j.foodchem.2018.08.124. [DOI] [PubMed] [Google Scholar]
  2. Fan X., Jiao X., Liu J., Jia M., Zhou Z. Characterizing the volatile compounds of different sorghum cultivars by both GC-MS and HS-GC-IMS. Food Research International. 2020;140(4) doi: 10.1016/j.foodres.2020.109975. [DOI] [PubMed] [Google Scholar]
  3. Feng D., Wang J., Ji X.-J., Min W.-X., Yan W.-J. HS-GC-IMS detection of volatile organic compounds in yak milk powder processed by different drying methods. LWT. 2021;141 doi: 10.1016/j.lwt.2021.110855. [DOI] [Google Scholar]
  4. Feng X., Wang H., Wang Z., Huang P., Kan J. Discrimination and characterization of the volatile organic compounds in eight kinds of huajiao with geographical indication of China using electronic nose HS-GC-IMS and HS-SPME-GC-MS. Food Chemistry. 2022;375 doi: 10.1016/j.foodchem.2021.131671. [DOI] [PubMed] [Google Scholar]
  5. Flaig M., Qi S.C., Wei G., Yang X., Schieberle P. Characterisation of the key aroma compounds in a Longjing green tea infusion (Camellia sinensis) by the sensomics approach and their quantitative changes during processing of the tea leaves. European Food Research and Technology. 2020;246(12):2411–2425. doi: 10.1007/s00217-020-03584-y. [DOI] [Google Scholar]
  6. Guo X., Ho C.T., Wan X., Zhu H., Liu Q., Wen Z. Changes of volatile compounds and odor profiles in Wuyi rock tea during processing. Food Chemistry. 2021;341(Pt 1) doi: 10.1016/j.foodchem.2020.128230. [DOI] [PubMed] [Google Scholar]
  7. Guo X., Schwab W., Ho C.T., Song C., Wan X. Characterization of the aroma profiles of oolong tea made from three tea cultivars by both GC-MS and GC-IMS. Food Chemistry. 2021;376 doi: 10.1016/j.foodchem.2021.131933. [DOI] [PubMed] [Google Scholar]
  8. He X., Yangming H., Gorska-Horczyczak E., Wierzbicka A., Jelen H.H. Rapid analysis of Baijiu volatile compounds fingerprint for their aroma and regional origin authenticity assessment. Food Chemistry. 2021;337 doi: 10.1016/j.foodchem.2020.128002. [DOI] [PubMed] [Google Scholar]
  9. Ho C.-T., Zheng X., Li S. Tea aroma formation. Food Science and Human Wellness. 2015;4(1):9–27. doi: 10.1016/j.fshw.2015.04.001. [DOI] [Google Scholar]
  10. Li Y., Fei C., Mao C., Ji D., Gong J., Qin Y.…Lu T. Physicochemical parameters combined flash GC e-nose and artificial neural network for quality and volatile characterization of vinegar with different brewing techniques. Food Chemistry. 2022;374 doi: 10.1016/j.foodchem.2021.131658. [DOI] [PubMed] [Google Scholar]
  11. Liu J.Y., He D., Xing Y.F., Zeng W., Ren K., Zhang C.…Xing X.H. Effects of bioactive components of Pu-erh tea on gut microbiomes and health: A review. Food Chemistry. 2021;353 doi: 10.1016/j.foodchem.2021.129439. [DOI] [PubMed] [Google Scholar]
  12. Lv H.P., Zhang Y., Yang T., Shi J., Lin Z. Key aroma constituents affecting the quality of ripened Pu-erh tea. Modern Food Science and Technology. 2015;31:394–399. doi: 10.13982/j.mfst.1673-9078.2015.12.058. [DOI] [Google Scholar]
  13. Lv H.P., Zhong Q.S., Lin Z., Wang L., Tan J.F., Guo L. Aroma characterisation of Pu-erh tea using headspace-solid phase microextraction combined with GC/MS and GC–olfactometry. Food Chemistry. 2012;130(4):1074–1081. [Google Scholar]
  14. Lv S., Wu Y., Li C., Xu Y., Liu L., Meng Q. Comparative analysis of Pu-erh and fuzhuan teas by fully automatic headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry and chemometric methods. Journal of Agricultural and Food Chemistry. 2014;62(8):1810–1818. doi: 10.1021/jf405237u. [DOI] [PubMed] [Google Scholar]
  15. Lv S., Wu Y., Wei J., Lian M., Wang C., Gao X., Meng Q. Application of gas chromatography-mass spectrometry and chemometrics methods for assessing volatile profiles of Pu-erh tea with different processing methods and aging years. RSC Advances. 2015;5(107):87806–87817. doi: 10.1039/c5ra15381f. [DOI] [Google Scholar]
  16. Ma W., Shi Y., Yang G., Shi J., Ji J., Zhang Y.…Lv H. Hypolipidaemic and antioxidant effects of various Chinese dark tea extracts obtained from the same raw material and their main chemical components. Food Chemistry. 2022;375 doi: 10.1016/j.foodchem.2021.131877. [DOI] [PubMed] [Google Scholar]
  17. Ma W., Zhu Y., Shi J., Wang J., Wang M., Shao C.…Lv H. Insight into the volatile profiles of four types of dark teas obtained from the same dark raw tea material. Food Chemistry. 2021;346 doi: 10.1016/j.foodchem.2020.128906. [DOI] [PubMed] [Google Scholar]
  18. Mao S., Lu C., Li M., Ye Y., Wei X., Tong H. Identification of key aromatic compounds in Congou black tea by partial least-square regression with variable importance of projection scores and gas chromatography-mass spectrometry/gas chromatography-olfactometry. Journal of the Science of Food and Agriculture. 2018;98(14):5278–5286. doi: 10.1002/jsfa.9066. [DOI] [PubMed] [Google Scholar]
  19. Pang X., Yu W., Cao C., Yuan X., Qiu J., Kong F., Wu J. Comparison of potent odorants in raw and ripened Pu-Erh tea infusions based on odor activity value calculation and multivariate analysis: understanding the role of pile fermentation. Journal of Agricultural and Food Chemistry. 2019;67(47):13139–13149. doi: 10.1021/acs.jafc.9b05321. [DOI] [PubMed] [Google Scholar]
  20. Qi H., Ding S., Pan Z., Li X., Fu F. Characteristic volatile fingerprints and odor activity values in different citrus-tea by HS-GC-IMS and HS-SPME-GC-MS. Molecules. 2020;25(24):6027. doi: 10.3390/molecules25246027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Su E., Xia T., Gao L., Dai Q., Zhang Z. Immobilization of beta-glucosidase and its aroma-increasing effect on tea beverage. Food and Bioproducts Processing. Transactions of the Institution of Chemical Engineers, Part C. 2010;88(2/3) [Google Scholar]
  22. Wang C., Li J., Wu X., Zhang Y., He Z., Zhang Y.…Liu Z. Pu-erh tea unique aroma: Volatile components, evaluation methods and metabolic mechanism of key odor-active compounds. Trends in Food Science & Technology. 2022;124:25–37. doi: 10.1016/j.tifs.2022.03.031. [DOI] [Google Scholar]
  23. Wang K., Liu F., Liu Z., Huang J., Xu Z., Li Y.…Yang X. Comparison of catechins and volatile compounds among different types of tea using high performance liquid chromatograph and gas chromatograph mass spectrometer, International. Journal of Food Science & Technology. 2011;46(7):1406–1412. doi: 10.1111/j.1365-2621.2011.02629.x. [DOI] [Google Scholar]
  24. Wang T., Li X., Yang H., Wang F., Kong J., Qiu D., Li Z. Mass spectrometry-based metabolomics and chemometric analysis of Pu-erh teas of various origins. Food Chemistry. 2018;268:271–278. doi: 10.1016/j.foodchem.2018.06.041. [DOI] [PubMed] [Google Scholar]
  25. Wang X., Rogers K.M., Li Y., Yang S., Chen L., Zhou J. Untargeted and targeted discrimination of honey collected by Apis cerana and Apis mellifera based on volatiles using HS-GC-IMS and HS-SPME-GC-MS. Journal of Agricultural and Food Chemistry. 2019;67(43):12144–12152. doi: 10.1021/acs.jafc.9b04438. [DOI] [PubMed] [Google Scholar]
  26. Xu M., Wang J., Zhu L. Tea quality evaluation by applying E-nose combined with chemometrics methods. Journal of Food Science and Technology. 2021;58(4):1549–1561. doi: 10.1007/s13197-020-04667-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Yang C.S., Wang H., Sheridan Z.P. Studies on prevention of obesity, metabolic syndrome, diabetes, cardiovascular diseases and cancer by tea. Journal of Food and Drug Analysis. 2018;26(1):1–13. doi: 10.1016/j.jfda.2017.10.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Yang Y., Hua J., Deng Y., Jiang Y., Qian M.C., Wang J.…Yuan H. Aroma dynamic characteristics during the process of variable-temperature final firing of Congou black tea by electronic nose and comprehensive two-dimensional gas chromatography coupled to time-of-flight mass spectrometry. Food Research International. 2020;137 doi: 10.1016/j.foodres.2020.109656. [DOI] [PubMed] [Google Scholar]
  29. Yang Y., Xie J., Chen J., Deng Y., Shen S., Hua J.…Jiang Y. Characterization of N, O-heterocycles in green tea during the drying process and unraveling the formation mechanism. Food Control. 2022;139 doi: 10.1016/j.foodcont.2022.109079. [DOI] [Google Scholar]
  30. Yang Y., Zhu H., Chen J., Xie J., Shen S., Deng Y.…Jiang Y. Characterization of the key aroma compounds in black teas with different aroma types by using gas chromatography electronic nose, gas chromatography-ion mobility spectrometry, and odor activity value analysis. LWT. 2022;163 doi: 10.1016/j.lwt.2022.113492. [DOI] [Google Scholar]
  31. Yang Y.Q., Rong Y.T., Liu F.Q., Jiang Y.W., Deng Y.L., Dong C.W., Yuan H.B. Rapid characterization of the volatile profiles in Pu-erh tea by gas phase electronic nose and microchamber/thermal extractor combined with TD-GC-MS. Journal of Food Science. 2021;86(6):2358–2373. doi: 10.1111/1750-3841.15723. [DOI] [PubMed] [Google Scholar]
  32. Yang Z., Baldermann S., Watanabe N. Recent studies of the volatile compounds in tea. Food Research International. 2013;53(2):585–599. doi: 10.1016/j.foodres.2013.02.011. [DOI] [Google Scholar]
  33. Yin, X., Huang, J. a., Huang, J., Wu, W., Tong, T., Liu, S., . . . Zhang, S. (2022). Identification of volatile and odor-active compounds in Hunan black tea by SPME/GC-MS and multivariate analysis. LWT, 164, 113656. doi: 10.1016/j.lwt.2022.113656.
  34. Yu H., Wang J., Zhang H., Yong Y., Cong Y. Identification of green tea grade using different feature of response signal from E-nose sensors. Sensors and Actuators B Chemical. 2008;128(2):455–461. [Google Scholar]
  35. Yun J., Cui C., Zhang S., Zhu J., Peng C., Cai H.…Hou R. Use of headspace GC/MS combined with chemometric analysis to identify the geographic origins of black tea. Food Chemistry. 2021;360 doi: 10.1016/j.foodchem.2021.130033. [DOI] [PubMed] [Google Scholar]
  36. Zhai X., Zhang L., Granvogl M., Ho C.T., Wan X. Flavor of tea (Camellia sinensis): A review on odorants and analytical techniques. Comprehensive Reviews in Food Science and Food Safety. 2022;21(5):3867–3909. doi: 10.1111/1541-4337.12999. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary data 1
mmc1.docx (1.3MB, docx)

Data Availability Statement

Data will be made available on request.

Articles from Food Chemistry: X are provided here courtesy of Elsevier

RESOURCES