Dynamic Changes of Volatile Compounds during the Xinyang Maojian Green Tea Manufacturing at an Industrial Scale

Xinyang Maojian (XYMJ) is one of the premium green teas and originates from Xinyang, which is the northernmost green tea production area in China. The special geographic location, environmental conditions, and manufacturing process contribute to the unique flavor and rich nutrition of XYMJ green tea. Aroma is an important quality indicator in XYMJ green tea. In order to illustrate the aroma of XYMJ green tea, the key odorants in XYMJ green tea and their dynamic changes during the manufacturing processes were analyzed by headspace solid-phase microextraction (HS-SPME) combined with gas chromatography-mass spectrometry (GC-MS). A total of 73 volatile compounds of six different chemical classes were identified in the processed XYMJ green tea samples, and the manufacturing processes resulted in the losses of total volatile compounds. Among the identified volatile compounds, twenty-four aroma-active compounds, such as trans-nerolidol, geranylacetone, nonanal, (+)-δ-cadinene, linalool, (Z)-jasmone, cis-3-hexenyl butyrate, cis-3-hexenyl hexanoate, methyl jasmonate, and β-ocimene, were identified as the key odorants of XYMJ green tea based on odor activity value (OAV). The key odorants are mainly volatile terpenes (VTs) and fatty acid-derived volatiles (FADVs). Except for (+)-δ-cadinene, copaene, cis-β-farnesene, (Z,E)-α-farnesene and phytol acetate, the key odorants significantly decreased after fixing. The principal coordinate analysis (PCoA) and the hierarchical cluster analysis (HCA) analyses suggested that fixing was the most important manufacturing process for the aroma formation of XYMJ green tea. These findings of this study provide meaningful information for the manufacturing and quality control of XYMJ green tea.


Introduction
Tea is one of the most consumed non-alcoholic beverages worldwide after water. Due to its pleasant taste, attractive aroma, and health-promoting effects [1][2][3], more than two billion people in approximately 170 countries and regions drink tea. According to the manufacturing techniques and sensory qualities, tea can be classified into six different types, including green, white, yellow, oolong, black and dark tea [4]. Green tea is a nonfermented tea and is usually produced from small-leaf tea cultivars (Camellia sinensis var. sinensis) [5]. The primary manufacturing processes for green tea include spreading, fixing, rolling, and drying, which maintain the color of the infusion and leaves green. Fixing is the feature of green tea manufacturing and is crucial for high-quality green tea, which must follow the

Experimental Materials
The manufacturing of XYMJ green tea was performed by Xinyang Yunzhen Tea Co., Ltd. in Xinyang, China. Slightly different from the traditional manufacturing processes of XYMJ green tea [13,21], the scattering and shaping processes were newly added during the XYMJ green tea manufacturing. Currently, the whole manufacturing process of XYMJ green tea is shown in Figure 1a  a 6CR-45 type rolling machine (Xinyang Yiding Tea Technology Co., Ltd., Xinyang, China). (5) Scattering. The rolled tea leaves were scattered for 15 min at 86 • C in a 6CCGK-7D type scattering machine (Xinyang Yiding Tea Technology Co., Ltd., Xinyang, China). (6) Shaping. The scattered tea leaves were shaped for 6 min at 115 • C in a 6CL-80/16D type shaping machine (Quzhou Hualin Machinery Co., Ltd., Quzhou, China). (7) Drying. The shaped tea leaves were dried for 20 min at 80 • C in a CS-6CHZ-9 type baking machine (Quanzhou Changsheng Tea Machinery Co., Ltd., Quanzhou, China) until the moisture content of tea was approximately up to 6.0%. All the samples were collected after each manufacturing stage, then frozen by liquid nitrogen immediately, vacuum freeze-dried, and packed into the aluminum foil bags and stored at −40 • C for further analysis. temperature (about 20 °C). (3) Fixing. The spread tea leaves were fixed for 120 s in a 6CST-80 type roller-hot air fixation machine (Xinyang Yiding Tea Technology Co., Ltd., Xinyang, China) with rolling temperature of 200 °C. (4) Rolling. The fixed tea leaves were rolled for 40 min in a 6CR-45 type rolling machine (Xinyang Yiding Tea Technology Co., Ltd., Xinyang, China). (5) Scattering. The rolled tea leaves were scattered for 15 min at 86 °C in a 6CCGK-7D type scattering machine (Xinyang Yiding Tea Technology Co., Ltd., Xinyang, China). (6) Shaping. The scattered tea leaves were shaped for 6 min at 115 °C in a 6CL-80/16D type shaping machine (Quzhou Hualin Machinery Co., Ltd., Quzhou, China). (7) Drying. The shaped tea leaves were dried for 20 min at 80 °C in a CS-6CHZ-9 type baking machine (Quanzhou Changsheng Tea Machinery Co., Ltd., Quanzhou, China) until the moisture content of tea was approximately up to 6.0%. All the samples were collected after each manufacturing stage, then frozen by liquid nitrogen immediately, vacuum freezedried, and packed into the aluminum foil bags and stored at −40 °C for further analysis.

Sensory Evaluations
The sensory evaluation of the XYMJ green tea samples was performed by five experienced tea experts (three women and two men; aged 28-35 years; nonsmokers) [18], and the aroma description and quality scores of the XYMJ green tea samples were assessed according to the national standards Methodology of sensory evaluation of tea (GB/T 23376-2018) [22] and Tea vocabulary for sensory evaluation (GB/T 14487-2017) [23]. To meet the requirements of this study, we only focused on the aroma quality. Accurately, 3.0 g of the tea sample was infused with 150 mL freshly boiled distilled water (100 °C), and filtered after brewing for 4.0 min. The brewed tea leaves were sniffed thrice to evaluate the intensity and persistence of their aroma. The aroma qualities of the XYMJ green tea samples were evaluated using a 100-point scoring scale from 0 to 100 and were determined by calculating the averages of the scores from the five panelists. Each tea sample was assessed three times through blind evaluation.

Extraction of XYMJ Green Tea Volatiles by HS-SPME
All the processed XYMJ green tea samples were ground into a powder in liquid ni-

Sensory Evaluations
The sensory evaluation of the XYMJ green tea samples was performed by five experienced tea experts (three women and two men; aged 28-35 years; nonsmokers) [18], and the aroma description and quality scores of the XYMJ green tea samples were assessed according to the national standards Methodology of sensory evaluation of tea (GB/T 23376-2018) [22] and Tea vocabulary for sensory evaluation (GB/T 14487-2017) [23]. To meet the requirements of this study, we only focused on the aroma quality. Accurately, 3.0 g of the tea sample was infused with 150 mL freshly boiled distilled water (100 • C), and filtered after brewing for 4.0 min. The brewed tea leaves were sniffed thrice to evaluate the intensity and persistence of their aroma. The aroma qualities of the XYMJ green tea samples were evaluated using a 100-point scoring scale from 0 to 100 and were determined by calculating the averages of the scores from the five panelists. Each tea sample was assessed three times through blind evaluation.

Extraction of XYMJ Green Tea Volatiles by HS-SPME
All the processed XYMJ green tea samples were ground into a powder in liquid nitrogen, and 0.5 g of the powder was transferred to a 20 mL headspace vial (Agilent, Santa Clara, CA, USA) containing 2.0 mL NaCl saturated solution to inhibit enzyme reactions. Then, 10 µL internal standard solution (3-hexanone-2,2,4,4-D4, 50 µg/mL in anhydrous ethanol) was added immediately. At the time of SPME analysis, each vial was shaken at 100 • C for 5 min, and then a 120 µm DVB/CAR/PDMS microextraction fiber (Supelco, Bellefonte, PA, USA) was exposed to the headspace of the sample for 15 min at 100 • C. The fiber was preconditioned for 5 min in the injection port of the gas chromatograph at 250 • C before analysis. Desorption of the volatile compounds from the fiber coating was carried out in the injection port of the GC apparatus at 250 • C for 5 min in splitless mode.

GC-MS Analysis
The identification and quantification of volatiles were performed by MetWare (http: //www.metware.cn/) (accessed on 14 December 2021) using the 8890 GC and 5977B mass spectrometer (Agilent, Santa Clara, CA, USA) equipped with a 30 m × 0.25 mm × 0.25 µm DB-5 MS (5% phenylpolymethylsiloxane) capillary column. Helium was used as the carrier gas at a linear velocity of 1.2 mL/min. The oven temperature was programmed from 40 • C and held for 3.5 min, firstly increased at 10 • C/min to 100 • C, then increased at 7 • C/min to 180 • C, and finally increased at 25 • C/min to 280 • C and held for 5 min. Mass spectra were recorded in electron impact (EI) ionization mode at 70 eV. The quadrupole mass detector, ion source and transfer line temperatures were set at 150, 230, and 280 • C, respectively. Mass spectra were scanned in the range m/z 50-500 amu at 1 s intervals. Volatile compounds were identified by comparing the mass spectra with the data system library (MWGC) and linear retention index [24,25]. The concentrations of the volatiles were calculated in µg/L based on the internal standard solution.

Odor Activity Value (OAV) Calculation
Odor activity value (OAV) is often applied to evaluate the contributions of aroma compounds. OAV was calculated using the equation OAV = C/OT, where C was the concentration of the volatile compound and OT was its odor threshold in water obtained from references.

Statistical Analysis
All the results were carried out in triplicate for the analytical determination. The analysis of significant differences between the samples was determined by one-way ANOVA (Duncan's multiple range tests) using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). The principal coordinate analysis (PCoA) and the hierarchical cluster analysis (HCA) were based on Bray-Curtis dissimilarities using vegan package in R (version 4.1, https://www.r-project.org/) (accessed on 16 May 2022).

Sensory Evaluation Analysis
As shown in Figure 1b, the processed XYMJ green tea samples presented various aroma characteristics, including green and floral aroma (tea shoots and spreading), clean aroma (fixing, rolling, scattering, and shaping), and long-lasting chestnut aroma (drying). The chestnut-like aroma is the typical aroma characteristic of some Chinese green tea and is referred to as an important indicator of an excellent-quality green tea [26,27]. Seventeen volatiles were identified as the key odorants responsible for the chestnut-like aroma of green tea, including trans-nerolidol, linalool, nonanal, cis-3-hexenyl hexanoate, 3-methylbutanal, (E)-3-penten-2-one, ethylbenzene and so on [19].

Identification and Quantification of the Volatile Compounds in XYMJ Green Tea
The volatile compounds in each manufacturing process of XYMJ green tea were tentatively identified by HS-SPME/GC-MS. A total of 73 volatile compounds were identified as common compounds for all the tea samples during the XYMJ green tea manufacturing, which were listed in Table 1. According to their chemical structures, these compounds were divided into six different chemical classes, including alcohols, esters, terpenes, aldehydes, ketones, and hydrocarbons. As presented in Figure 2, alcohols were present in the highest concentration (8797.9 µg/L-56,586.2 µg/L), followed by terpenes (11,540.3 µg/L-56,835.6 µg/L), esters (9312.5 µg/L-25,332.2 µg/L), and aldehydes (2835.3 µg/L-24,390.7 µg/L), indicating that they were the four major volatile compound groups. Compared to the other chemical classes, the ketones (1263.8 µg/L-3794.2 µg/L) had the lowest concentration. The concentrations of total volatile compounds in tea shoots and drying stage were 168,218.8 µg/L and 44,851.9 µg/L, respectively, which decreased by 73.3%, indicating that the manufacturing processes resulted in the losses of volatile compounds.

Alcohols
The alcoholic compounds varied during the XYMJ green tea manufacturing, and the changes in the presentative alcoholic compounds were shown in Figure 3. Spreading significantly increased the concentrations of trans-furan linalool oxide, trans-nerolidol, phytol, isophytol, dehydroisophytol, and cis-cubenol. The highest concentrations of trans-furan linalool oxide and trans-nerolidol were 6638.6 ug/L and 4422.4 ug/L at the spreading stage, respectively. Spreading is an indispensable process in the aroma formation of premium green tea. Volatile metabolomics and transcriptomics revealed that the concentration of trans-furan linalool oxide significantly increased after spreading [28]. Except for phytol, isophytol and dehydroisophytol, the fixing significantly decreased the concentrations of other alcoholic compounds. The lowest concentration of linalool was 4265.1 ug/L at the fixing stage. After fixing, remarkable losses were observed for linalool, trans-furan linalool oxide, and trans-nerolidol during green tea manufacturing [10,11,20]. From fixing to drying, linalool and trans-furan linalool oxide, trans-nerolidol, and cis-cubenol, presented the same trends, respectively. The concentrations of phytol, isophytol, and dehydroisophytol gradually increased from spreading to shaping and then decreased at drying The aroma profiles of green tea made with fresh tea leaves plucked in summer were studied by the analysis of volatile compounds during the manufacturing, which showed that the concentration of phytol in tea shoots was significantly higher than the summer green tea [29], which is inconsistent to the present study. We speculate that the reason might be due to the different manufacturing techniques and the harvested season of tea shoots.

Alcohols
The alcoholic compounds varied during the XYMJ green tea manufacturing, and the changes in the presentative alcoholic compounds were shown in Figure 3. Spreading significantly increased the concentrations of trans-furan linalool oxide, trans-nerolidol, phytol, isophytol, dehydroisophytol, and cis-cubenol. The highest concentrations of trans-furan linalool oxide and trans-nerolidol were 6638.6 µg/L and 4422.4 µg/L at the spreading stage, respectively. Spreading is an indispensable process in the aroma formation of premium green tea. Volatile metabolomics and transcriptomics revealed that the concentration of trans-furan linalool oxide significantly increased after spreading [28]. Except for phytol, isophytol and dehydroisophytol, the fixing significantly decreased the concentrations of other alcoholic compounds. The lowest concentration of linalool was 4265.1 µg/L at the fixing stage. After fixing, remarkable losses were observed for linalool, trans-furan linalool oxide, and trans-nerolidol during green tea manufacturing [10,11,20]. From fixing to drying, linalool and trans-furan linalool oxide, trans-nerolidol, and cis-cubenol, presented the same trends, respectively. The concentrations of phytol, isophytol, and dehydroisophytol gradually increased from spreading to shaping and then decreased at drying. The aroma profiles of green tea made with fresh tea leaves plucked in summer were studied by the analysis of volatile compounds during the manufacturing, which showed that the concentration of phytol in tea shoots was significantly higher than the summer green tea [29], which is inconsistent to the present study. We speculate that the reason might be due to the different manufacturing techniques and the harvested season of tea shoots.

Esters
A total of seventeen esters were identified (Table 1), among which were six hexen esters, including cis-3-hexenyl hexanoate, (Z)-3-hexenyl octanoate, cis-3-hexenyl butyra cis-3-hexenyl valerate, (E)-2-hexenyl hexanoate, and cis-3-hexenyl crotonate. From t shoots to drying, cis-3-hexenyl butyrate gradually decreased and reached its minimu level (154.4 ug/L) at drying (Figure 4a), which decreased by 88.5%. cis-3-Hexenyl hexan ate ( Figure 4b) and cis-3-hexenyl valerate (Figure 4c) presented a similar trend during t manufacturing, and reached their minimum levels at drying, which decreased by 56.1 and 76.9%, respectively. The concentrations of the six hexenyl esters were the lowest drying, and the hexenyl esters losses might be due to their high volatility. As shown Figure 4d and Figure 4f, the variations of methyl jasmonate and dihydroactinolide a peared a similar trend, and spreading significantly increased their concentrations. In grated volatile metabolomics and transcriptomics confirmed that spreading could crease the concentration of methyl jasmonate [28]. The high-temperature fixing sign cantly resulted in the losses of methyl jasmonate and dihydroactinolide, which decreas by 45.5% and 50.5%, respectively. After scattering, the high temperature (115℃) shapi promoted the increase in the concentrations of methyl jasmonate and dihydroactinolid Being similar to the above esters mentioned, the concentration of methyl salicylate sign icantly decreased from tea shoots (11,159.7 ug/L) to fixing (712.9 ug/L). Although the m thyl salicylate fluctuated from fixing to drying, there were no significant differences b tween them. The changing trend of methyl salicylate in this study was consistent with t previous findings [20,30].

Esters
A total of seventeen esters were identified (Table 1), among which were six hexenyl esters, including cis-3-hexenyl hexanoate, (Z)-3-hexenyl octanoate, cis-3-hexenyl butyrate, cis-3-hexenyl valerate, (E)-2-hexenyl hexanoate, and cis-3-hexenyl crotonate. From tea shoots to drying, cis-3-hexenyl butyrate gradually decreased and reached its minimum level (154.4 µg/L) at drying (Figure 4a), which decreased by 88.5%. cis-3-Hexenyl hexanoate ( Figure 4b) and cis-3-hexenyl valerate (Figure 4c) presented a similar trend during the manufacturing, and reached their minimum levels at drying, which decreased by 56.1% and 76.9%, respectively. The concentrations of the six hexenyl esters were the lowest at drying, and the hexenyl esters losses might be due to their high volatility. As shown in Figure 4d,f, the variations of methyl jasmonate and dihydroactinolide appeared a similar trend, and spreading significantly increased their concentrations. Integrated volatile metabolomics and transcriptomics confirmed that spreading could increase the concentration of methyl jasmonate [28]. The high-temperature fixing significantly resulted in the losses of methyl jasmonate and dihydroactinolide, which decreased by 45.5% and 50.5%, respectively. After scattering, the high temperature (115 • C) shaping promoted the increase in the concentrations of methyl jasmonate and dihydroactinolide. Being similar to the above esters mentioned, the concentration of methyl salicylate significantly decreased from tea shoots (11,159.7 µg/L) to fixing (712.9 µg/L). Although the methyl salicylate fluctuated from fixing to drying, there were no significant differences between them. The changing trend of methyl salicylate in this study was consistent with the previous findings [20,30].

Aldehydes and Ketones
The concentration of nonanal was the highest (20,634.3 ug/L) in tea shoots and t lowest (2701.0 ug/L) at fixing, which decreased by 86.9%. From fixing to drying, the co centration of nonanal increased by 18.1% due to the thermal degradation of lipids. A shown in Figure 5a, the dynamic change of nonanal during the XYMJ green tea manufa turing was consistent with the previous studies [11,20]. The concentration of geranial i creased to the highest level (4372.1 ug/L) at spreading and sharply decreased to the lowe level (134.3 ug/L) at fixing, which decreased by 96.9%. The concentration of geranial i creased from fixing to shaping and then decreased at drying (Figure 5b). The high tem perature (200 °C) of fixing resulted in the losses of nonanal and geranial. The variations nonanal and geranial were no significant differences from fixing to drying.

Aldehydes and Ketones
The concentration of nonanal was the highest (20,634.3 µg/L) in tea shoots and the lowest (2701.0 µg/L) at fixing, which decreased by 86.9%. From fixing to drying, the concentration of nonanal increased by 18.1% due to the thermal degradation of lipids. As shown in Figure 5a, the dynamic change of nonanal during the XYMJ green tea manufacturing was consistent with the previous studies [11,20]. The concentration of geranial increased to the highest level (4372.1 µg/L) at spreading and sharply decreased to the lowest level (134.3 µg/L) at fixing, which decreased by 96.9%. The concentration of geranial increased from fixing to shaping and then decreased at drying (Figure 5b). The high temperature (200 • C) of fixing resulted in the losses of nonanal and geranial. The variations of nonanal and geranial were no significant differences from fixing to drying. As presented in Figure 5c, the concentration of (Z)-jasmone significantly increased to 3330.3 ug/L at spreading and then remarkably decreased to 1004.2 ug/L at fixing. It has been reported that spreading contributes to the (Z)-jasmone accumulation, regardless of the tea plant cultivars [28]. Compared to tea shoots (1806.1 ug/L), the concentration of (Z)jasmone decreased by 51.0% at the drying. The variations of (Z)-jasmone were no significant differences among the subsequent manufacturing processes (from fixing to drying). The concentration of geranylacetone fluctuated during the manufacturing (Figure 5d), which was the highest at spreading (251.8 ug/L) and lowest at rolling (159.2 ug/L). The dynamic change of geranylacetone in the present study was consistent with the previous research [29].

Principal Component Analysis and Hierarchical Cluster Analysis
The identified 73 volatile compounds were used as variables to perform the princip coordinate analysis (PCoA, Figure 7a) and hierarchical clustering analysis (HCA, Figu  7b). The clustering results of PCoA were evaluated by PERMANOVA in vegan, whi showed that the processed XYMJ green tea samples were divided into two clusters (PE MANOVA R 2 = 0.976, p < 0.001). Cluster-1 consisted of the tea shoots and spreading sa ples, and Cluster-2 consisted of the remained XYMJ green tea samples from fixing to d ing. The results indicated that fixing was the most important manufacturing process the aroma formation of XYMJ green tea. A total of nine volatile compounds with a va ance importance value (VIP) > 1 might contribute to separating the processed XYMJ gre tea samples. Linalool had the highest VIP (4.50), followed by β-myrcene

Principal Component Analysis and Hierarchical Cluster Analysis
The identified 73 volatile compounds were used as variables to perform the principal coordinate analysis (PCoA, Figure 7a) and hierarchical clustering analysis (HCA, Figure 7b). The clustering results of PCoA were evaluated by PERMANOVA in vegan, which showed that the processed XYMJ green tea samples were divided into two clusters (PERMANOVA R 2 = 0.976, p < 0.001). Cluster-1 consisted of the tea shoots and spreading samples, and Cluster-2 consisted of the remained XYMJ green tea samples from fixing to drying. The results indicated that fixing was the most important manufacturing process for the aroma formation of XYMJ green tea. A total of nine volatile compounds with a variance importance value (VIP) > 1 might contribute to separating the processed XYMJ green tea samples. Linalool had the highest VIP (4.50), followed by β-myrcene

Odor Activity Values
Generally, volatile compounds with OAV ≥ 1 are regarded as potential contribut to an aroma profile [31]. Higher OAVs correspond to a greater contribution to the arom Twenty-four volatile compounds had OAVs＞1 in all the processed XYMJ green tea sa ples, except for phytol acetate with OAV (0.9) in tea shoots (Table 2), and most of the k odorants had floral attributes. The top ten key odorants with the highest OAVs in made XYMJ green tea were trans-nerolidol (10,199 Previous studies and the present result revealed that (E)-nerolidol was the key od ant contributing to the aroma profile of XYMJ green tea and other green teas [11,17,19,3 Multiple stresses, such as mechanical damage and low temperature, had a synergistic fect on (E)-nerolidol formation during oolong tea manufacturing [33]. (E)-Nerolidol cou also be derived from the non-enzymatic degradation of phytofluene [7]. Derived from carotenoids, the floral geranylacetone was also considered the aroma-active compound Japanese green tea (Sen-cha) [34], summer green tea [29], and Longjing tea [35]. Based OAV calculation, nonanal, linalool, (Z)-jasmone, and methyl jasmonate were the key od ants in premium green teas (Longjing, XYMJ, Taiping Houkui, Lu'an Guapian, e [11,[35][36][37]. Cis-3-hexenyl hexanoate in XYMJ green tea was determined by both OAV c culation and GC-O, while (+)-δ-cadinene was only detected by OAV calculation, wh might be due to the different sample preparation techniques [11]. Therefore, a comp

Odor Activity Values
Generally, volatile compounds with OAV ≥ 1 are regarded as potential contributors to an aroma profile [31]. Higher OAVs correspond to a greater contribution to the aroma. Twenty-four volatile compounds had OAVs > 1 in all the processed XYMJ green tea samples, except for phytol acetate with OAV (0.9) in tea shoots (Table 2), and most of the key odorants had floral attributes. The top ten key odorants with the highest OAVs in the made XYMJ green tea were trans-nerolidol (10,199.4 Previous studies and the present result revealed that (E)-nerolidol was the key odorant contributing to the aroma profile of XYMJ green tea and other green teas [11,17,19,32]. Multiple stresses, such as mechanical damage and low temperature, had a synergistic effect on (E)-nerolidol formation during oolong tea manufacturing [33]. (E)-Nerolidol could also be derived from the non-enzymatic degradation of phytofluene [7]. Derived from the carotenoids, the floral geranylacetone was also considered the aroma-active compound in Japanese green tea (Sen-cha) [34], summer green tea [29], and Longjing tea [35]. Based on OAV calculation, nonanal, linalool, (Z)-jasmone, and methyl jasmonate were the key odorants in premium green teas (Longjing, XYMJ, Taiping Houkui, Lu'an Guapian, etc.) [11,[35][36][37]. Cis-3-hexenyl hexanoate in XYMJ green tea was determined by both OAV calculation and GC-O, while (+)-δ-cadinene was only detected by OAV calculation, which might be due to the different sample preparation techniques [11]. Therefore, a comprehensive analysis of XYMJ green tea aroma by GC-O technique is required in future studies. As shown in Table 2, the key odorants in the processed XYMJ green tea samples almost belong to endogenous biosynthesis volatiles, including fatty acid-derived volatiles (FADVs), amino acid-derived volatiles (AADVs), volatile terpenes (VTs) and carotenoidderived volatiles (CDVs) [2,41]. There are fourteen key odorants that are VTs and they present fruity and floral odors, which may contribute to the formation of the superior aroma quality of XYMJ green tea. In addition to cis-3-hexenyl hexanoate with a green odor, the FADVs, including nonanal, (Z)-jasmone, cis-3-hexenyl butyrate, methyl jasmonate, and cis-3-hexenyl valerate, could form the floral quality of XYMJ green tea. As the only AADVs with a green odor, the decrease of the methyl salicylate is beneficial to the formation of the attractive aroma quality of XYMJ green tea. Derived from the carotenoids, geranylacetone and dihydroactinolide are necessary for the formation of the aroma of XYMJ green tea. To sum up, the aroma of XYMJ green tea is the comprehensive presentation of the key odorants, which are derived from multiple biosynthetic pathways.