Comprehensive Characterization of Aroma-Active Components in Three Grades of Raw Tea Leaves and Their Jasmine Tea Products of Wuyutai Using GC×GC-O-MS and Chemometrics

Abstract

This study investigated the aroma characteristics of three grades of raw tea leaves and their corresponding jasmine tea products from Guangxi, China. Aromatic profiles of jasmine tea varieties were analysed using two-dimensional gas chromatography-olfactory-mass spectrometry (GC×GC-O-MS), stir bar sorptive extraction (SBSE), and descriptive sensory evaluation. Chemometric methods were applied to compare sensory scores with instrumental data. Volatile compound concentrations and relative odour activity values (r-OAVs) were calculated. The results indicated significant differences in base tea leaf quality: high-grade tea leaf G1 exhibited pure, sweet characteristics, serving as an excellent aroma-absorbing carrier. The scenting process significantly imparted jasmine fragrance to the finished product, although its efficacy was constrained by tea leaf grade. GH1 finished tea exhibited a fresh, vibrant, and rich aroma with a sweet, mellow fragrance and high floral integration. In contrast, GH3, due to its inferior base material quality, yielded a weak aroma after scenting with limited quality improvement. The initial quality of the tea base is the fundamental determinant of the upper limit of the finished jasmine tea’s sensory quality, while the scenting process is the core means of shaping its signature floral aroma. The combination of high-quality tea leaves and precise scenting techniques is essential for developing the fresh, vibrant, and rich flavour profile of premium jasmine tea. This study reveals that the flavour formation of jasmine tea originates from the foundational quality of the tea leaves, providing a theoretical basis for monitoring the aroma quality of jasmine tea produced from different grades of tea leaves.

1. Introduction

Jasmine tea is a reprocessed tea made by scenting green tea with jasmine flowers, with a history dating back to the Song Dynasty [1]. In recent years, jasmine tea has gained increasing popularity due to its unique flavour profile and various health benefits, including effects on hypoglycaemic activity, immune regulation, and antioxidant properties [2]. The quality of jasmine tea is profoundly influenced by the base tea. Among these floral teas, jasmine green tea stands as a significant branch of Chinese floral teas and is the most popular both domestically and internationally [3], and thus was selected as the focus of this study.

Fragrance, as a key factor in evaluating the quality of jasmine tea, is determined by its key aromatic components [4]. Stir bar sorptive extraction (SBSE) is a volatile compound extraction technique that directly contacts samples via immersion to adsorb aroma substances [5]. Furthermore, compared to the low recovery rate of Solvent-Assisted Flavour Evaporation (SAFE), compound degradation in Simultaneous Distillation Extraction (SDE), the high application cost of Supercritical Fluid Extraction (SFE), the low recovery rate and high sample volume in Vacuum Distillation Extraction (VDE), the SBSE method offers technical advantages, such as higher sensitivity, lower detection limits, and good reproducibility. It also avoids the use of organic solvents, contributing to environmental protection.

The GC×GC-O-MS (Comprehensive Two-Dimensional Gas Chromatography-Olfactory Analysis-Mass Spectrometry System), a coupled analytical technique integrating comprehensive two-dimensional gas chromatography (GC×GC) with simultaneous olfactory detection (O) and mass spectrometry (MS) identification, with its high peak capacity, sensitivity, and resolution, enables near-complete separation of volatiles in complex matrices. When coupled with an olfactory detection port (O), this technique simultaneously yields chemical structural and aroma property information, enabling precise identification of key odorants contributing to the overall fragrance. Xia et al. detected 68 aroma compounds in 8 varieties of blue honeysuckle berries through GC×GC-O-MS coupled with sensory-guided analysis, identifying 12 compounds—including linalool, hexanal, and eucalyptol—as primary aroma contributors [6]. Zou et al. employed Solid-Phase Microextraction (SPME) coupled with GC×GC-O-MS to identify 34 odour compounds in three roasted green teas—Huangshan Maofeng, Taiping Houkui, and Shucheng Xiaolanhua—confirming D-limonene and linalool as odour markers for Grade A roasted green tea [7]. However, applications of GC×GC-O-MS to investigate characteristic aromas in different grades of tea leaves and finished jasmine tea products remain scarce. Moreover, its application in exploring characteristic aroma compounds across different jasmine tea varieties remains relatively limited. Therefore, this study employed a combined GC×GC-O-MS and SBSE approach to systematically examine the volatile components of three distinct grades of tea base material from Hengxian, Guangxi, and their corresponding finished jasmine tea products. Further analysis employed multivariate statistical methods, including principal component analysis (PCA) and orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA). The findings elucidate key aromatic differences between tea leaves of varying grades and their corresponding jasmine tea products. This research provides scientific evidence for studying characteristic flavour compounds in jasmine tea, offering theoretical support and practical guidance for cultivar selection and processing optimization within the jasmine tea industry.

2. Materials and Methods

2.1. Chemicals and Reagents

The chemicals and reagents used in this study were procured from several suppliers. The internal standard 2-methyl-3-heptanone (GC grade) and N-alkanes (C₇–C₂₅) were purchased from Sigma-Aldrich Co., Ltd. (Beijing, China), and n-hexane (GC grade) was purchased from Fisher Chemicals (Shanghai, China). Ultra-high purity helium gas (99.9992% purity) and nitrogen gas (99.999% purity) were purchased from AP BAIF Gases Industry Co., Ltd. (Beijing, China). Sodium chloride (NaCl),with 99.5% purity, was purchased from Shanghai Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

2.2. Sample Information

Six samples of jasmine tea I are shown in

Table 1. Sample Information.

Tea Sample Number	Jasmine Tea Sample Number
G1	GH1
G2	GH2
G3	GH3

.

G1–G3 denote three grades of tea base material. G1 represents a premium-grade tea base, G2 denotes a medium-grade tea base, and G3 signifies a low-grade tea base. GH1–GH3 are finished jasmine teas produced from these three tea bases using the same scenting process. All samples were provided by Beijing Wuyutai Tea Co., Ltd. (Beijing, China).

2.3. Sensory Evaluation of Six Types of Jasmine Tea

For the sensory evaluation of jasmine tea samples, 16 trained panellists (8 female and 8 male) with an average age of 26 were recruited. Tea samples were prepared in accordance with the Chinese National Standard GB/T 23776-2018 [8] ‘Methods for Sensory Evaluation of Tea’: 3 g of jasmine tea was placed in a standard evaluation cup and steeped with 150 mL of boiling water. After 3 min, the tea liquor was transferred to a standard evaluation bowl, and an additional 150 mL of boiling purified water was added. After 5 min, the tea liquor was poured into the evaluation bowl. Group members conducted a comprehensive evaluation of the aroma of the first and second infusions of jasmine tea. Aroma evaluation was based on 10 criteria: floral aroma, fruity aroma, grassy/fresh leaf aroma, herbal aroma, sweet aroma, intensity, purity, persistence, freshness, and overall flavour balance. A 0–8 point scale was used to score aroma attributes, with the following criteria: 0 = no odour, 2 = faint, 4 = weak aroma, 6 = strong aroma, and 8 = extremely strong aroma. The aroma quality of the jasmine tea samples was characterized by the average score of each attribute. Prior to the experiment, all participants were informed of the study content and provided informed consent. The study was conducted in strict accordance with relevant regulations protecting participants’ rights and privacy.

2.4. Stir Bar Sorptive Extraction (SBSE)

A 3.0 g tea sample was accurately weighed using an electronic balance, 150 mL of boiling water was added, and the mixture was brewed for 3 min. Then, 10 mL of the tea infusion was transferred into a 25 mL headspace vial, and 3.0 g of sodium chloride was added. After thorough mixing, 1 µL of the internal standard solution (2-methyl-3-heptanone, 0.816 µg/µL) was added. The vial was immediately sealed and mixed thoroughly again. Each mixing step was performed using a rotary mixer, taking care to avoid splashing the sample onto the inner wall of the vial. The headspace vial was then placed on the plate of a heating magnetic stirrer preheated to 40 °C and equilibrated for 10 min under agitation, and then the polydimethylsiloxane (PDMS)-coated stir bar was quickly placed into the headspace vial and stirred and adsorbed for 60 min at a rate of 1250 r/min. The stir bar was then removed, rinsed with ultrapure water, dried with lint-free microscope paper, and placed into an empty injection tube in the thermal desorption unit. The stir bar was thermally desorbed in a thermal desorption unit (TDU) in an empty inlet tube and repeated three times for each sample.

2.5. Comprehensive Two-Dimensional Gas Chromatography-Olfactometry-Mass Spectrometry (GC×GC-O-MS)

The aroma compounds were analysed using a GC × GC-O-MS system, which consisted of an Agilent 8890 A GC coupled with a 5977B mass spectrometer (Agilent Technologies, Santa Clara, CA, USA) and an olfactory detection port ODP4 (Gerstel, Ruer, Germany). The gas chromatography consisted of two columns: a polar DB-Wax capillary column (30 m × 0.25 mm × 0.25 μm) and a medium polar DB-17 ms capillary column (1.85 m × 0.18 mm × 0.18 μm). The two columns were combined with a solid modulator SSM1800 (J&X Technology Co., Ltd., Shanghai, China) for the heating and cooling stage. For the two-dimensional separation, the modulator cold zone was set to −50 °C with a modulation period of 5 s. The GC oven temperature program was initialized at 40 °C (held for 2 min), then increased to 230 °C at a rate of 4 °C/min (final hold time: 5 min). The mass spectrometer operated in electron ionization (EI) mode at 70 eV, with the ion source and quadrupole temperatures maintained at 230 °C and 150 °C, respectively. Full-scan data were acquired over a mass range of m/z 33–350. For the two-dimensional separation, the modulator cold zone was set to −50 °C with a modulation period of 4 s. Ultra-high purity helium was used as the carrier gas with a constant flow rate of 1.2 mL/min, and the sample was injected without splitting. To allow the participants to sniff aroma compounds through the instrument, an olfactory detection port (ODP) was installed between the GC system and the MSD system. The operational parameters of the ODP were adjusted. The temperature of the ODP transmission line was set to 280 °C, and the temperature of the sniffing port was set to 200 °C. The split ratio between the olfactory detection port and the mass spectrometry detector (MSD) was 1:1. The ODP sniffing port was continuously ventilated with ultra-high-purity nitrogen gas. Moisture was continuously supplied through the ODP port to keep the sniffer’s nasal cavity moist and avoid the drying of nose mucus during the sniffing experiment. Instrument control, data acquisition, and chromatographic processing were performed using a MassHunter Workstation (Version 10.0, Agilent Technologies). Full two-dimensional data reconstruction and visualization were handled with Canvas software (Version 1.0, J&X Technology, Shanghai, China). Sniffing data recording and aroma feature annotation were synchronously completed via the ODP4 accompanying recording software (Gerstel, Mülheim an der Ruhr, Germany).

2.6. Characterization of Odour-Active Compounds

The odour-active compounds were identified using mass spectrometry (MS), a retention index (RI), odour description (O) and STD PV. Volatile compounds were tentatively identified by matching mass spectra with the NIST17 library (Version 2.3), applying a minimum similarity threshold of 800 (match factor scale: 0–1000). The retention index (RI) serves as an auxiliary qualitative reference. The RI value is calculated using the Kovats method based on the retention time on the first dimension column (DB-Wax), obtained by analysing the retention times of the C7–C25 n-alkane series under identical GC conditions. It is provided solely as a supplementary reference; final identification relies on mass spectrometry matching, sensory description, and standard sample verification.

The odour profile of an odour-active compound was determined by an assessor familiar with the odour descriptors. Standard verification means that a dilution of a standard of the target compound is injected into the instrument for analysis, and the results of this analysis are compared with the results of the analysis of the target compound in the original sample to determine the qualitative results of the target compound. To ensure consistency and accuracy, the recorded odour descriptions were cross-validated with publicly available aroma databases, including Flavornet and The Pherobase, as well as the Compendium of Odour Thresholds in Air, Water and Other Media (Van Gemert, 2011) [9].

2.7. Semi-Quantification of Odour-Active Compounds

The semi-quantification of the odour-active compounds in the sample was carried out by the internal standard method, using the added 2-methyl-3-heptanone as the internal standard reference, and the semi-quantitative concentration of the target compounds was calculated by the peak area of the target compounds proportional to the concentration of the internal standard compounds, which is shown in Formula (1):

$${\mathrm{C}}_{\mathrm{x}}={\displaystyle \frac{{\mathrm{A}}_{\mathrm{x}}}{{\mathrm{A}}_{\mathrm{n}}}}\times {\mathrm{C}}_{\mathrm{n}}$$

(1)

In Formula (1), C_x represents the concentration of the target compound, C_n represents the concentration of the internal standard compound, A_x represents the peak area of the target compound, and A_n represents the peak area of the internal standard compound.

2.8. Determination of Relative Odour Activity Value (r-OAV)

The relative odour activity values (r-OAV) of the aroma compounds were calculated for a total of 10 samples from 5 jasmine tea varieties using the ratio of the relative concentration of each identified aroma compound to its odour threshold in water. The odour thresholds of the aroma compounds were collected from previous studies, databases and a book (Compendium of Odour Thresholds in Air, Water and Other Media). The r-OAV of the aroma compounds was calculated by the following formula:

r-OAV = C/t

(2)

where C denotes the concentration of the aroma compound and t denotes the odour threshold of the odorant in water.

2.9. Chemometric Data Analysis (CDA)

Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) was performed using SIMCA (version 13.0) software (SIMCA Umetrics, Umeå, Sweden). For OPLS-DA model analyses, one type of scaling (Pareto (Par)) was used for all variables in the dataset. The data were analysed by principal component analysis (PCA). Hierarchical clustering heat map analysis (HCA) was used to analyse the content of volatile compounds in jasmine tea samples. Heat maps were generated by TB Tools software (version v2.142). Sensory scores were analysed by spidergrams generated by Origin Pro, version 2024, Origin Lab Corporation, Northampton, MA, USA. The content of volatile compounds was used to cluster the different jasmine tea samples. One-way data were analysed using one-way ANOVA (one-way analysis of variance). Significant (p ≤ 0.05) differences between jasmine tea samples were analysed using post hoc analysis combined with Duncan’s test. Each jasmine tea sample underwent three parallel experiments (three independent preparations and analyses), and all results of this study were expressed as mean (taking the mean of replicates) and standard deviation (SD) values.

3. Results and Discussion

3.1. Sensory Quality Evaluation and Flavour Classification of Jasmine Tea

Aroma is the most crucial indicator determining jasmine tea quality, with the distinctiveness of its refreshing floral fragrance being the primary characteristic consumers evaluate when assessing its quality [10]. To investigate the combined influence of raw tea leaf quality and scenting process on the flavour profile of finished floral tea, this study conducted systematic sensory evaluation analyses on three grades of raw tea leaves (G1, G2, G3) and their corresponding finished jasmine teas (GH1, GH2, GH3) processed using a standardized jasmine scenting technique. Within the sensory evaluation framework for jasmine tea, the assessors conducted comprehensive appraisals based on 10 key dimensions: floral notes (jasmine or rose-like aromas), fruity notes (fresh fruit or date-like scents), grassy/fresh leaf notes (aroma of freshly cut grass), herbal notes (fresh medicinal herb scent), sweetness (honey-like sweetness), intensity (gradation of aroma strength), purity (degree of purity without off-flavours), persistence (duration of aroma retention), freshness (fresh, transparent floral quality), and overall flavour balance (harmonious integration of floral and tea aromas). All dimensions were scored on a 0–8 point scale: 0 denotes ‘undetectable’, 2 denotes ‘faint’, 4 denotes ‘moderate’, 6 denotes ‘rich’, and 8 denotes ‘extremely rich’. As illustrated in Figure 1, the sensory profiles of the three tea bases exhibited significant differences, reflecting their inherent quality hierarchy. The sensory qualities of G1 premium tea leaves—notably pronounced sweetness, exceptional purity and balanced aroma—are fundamentally the result of synergistic interactions between their specific chemical composition and microscopic physical structure. Chemically, elevated theanine content and a harmonious phenolic-to-amino acid ratio establish the foundation for a mellow flavour profile with a pronounced sweet aftertaste. Concurrently, minimal levels of undesirable volatile compounds (such as hexanal) within the base aroma create a pristine aromatic carrier. Physically, the highly developed pore structure combined with an optimized moisture content (3–5%) endows the tea leaves with superior adsorption potential, ensuring efficient loading and stable fixation of jasmine fragrance molecules. Thus, its outstanding performance in sensory evaluation scientifically reflects its intrinsic physicochemical qualities, confirming the suitability of G1 tea leaves as an ideal raw material for jasmine tea scenting. The medium-grade tea base G2 exhibited prominence in fruity notes, although grassy and herbal characteristics were also relatively pronounced. Its overall harmony and purity fell short of G1. The low-grade tea base G3 scored lowest across all sensory dimensions, particularly in purity, indicating weak aroma intensity with off-flavours and a fundamentally poor raw material quality. The standardized scenting process profoundly transformed all three tea bases, notably enhancing flavour dimensions, such as jasmine and sweetness. However, the scenting effect exhibited a clear interactive relationship with tea base grade. Following scenting, the aroma radar chart for jasmine tea samples made from three different grades of tea base is shown in Figure 2, GH1’s sensory profile became balanced and full-bodied. Not only did its floral notes become more intense and vibrant, but its inherent fresh sweetness was enhanced and transformed into a mellow sweet aroma. This demonstrates that superior tea base material can synergize with the scenting process, achieving a perfect fusion of floral fragrance and tea flavour. GH2 scored lower than GH1 across all metrics, with particularly noticeable gaps in purity and persistence. This indicates that the inherent flavour characteristics of G2 tea leaves somewhat constrained the freshness and purity of the floral aroma in the finished product. GH3 achieved the lowest overall score, exhibiting a thin aroma and poor harmony. This outcome may be attributed to inherent deficiencies in G3 tea leaves as an adsorbent carrier: their physical structure exhibits poorly developed porosity and a limited specific surface area, significantly reducing the adsorption capacity and efficiency for aromatic molecules. Chemically, their composition contains relatively high levels of baseline undesirable aromatic compounds, which not only generate off-flavours but also competitively occupy effective adsorption sites, further diminishing the capacity to adsorb and fix the characteristic jasmine fragrance. Ultimately, the dual constraints of physical and chemical properties collectively result in a finished tea exhibiting a weak aroma and poor integration between the tea and floral scents.

3.2. Analysis of Three Different Grade Tea Blanks and Their Finished Jasmine Tea Samples Using GC×GC-O-MS

The aroma of tea represents the sensory integration of its aromatic constituents, which form the material basis for flavour development. Employing GC×GC-O-MS, a total of 121 aromatic compounds were detected across three grades of tea base material and their corresponding finished jasmine tea samples (

Table 2. A list of aroma compounds identified by GC × GC-O-MS in three grades of tea base material and their corresponding finished jasmine tea samples.

No.	RT	RI	Compounds	CAS	Odour	Formula
			Esters
T1	29.063	1754	Acetic acid, phenylmethyl ester	140-11-4	sweet floral fruity jasmine fresh	C9H10O2
T2	31.529	1791	Acetic acid, 2-phenylethyl ester	103-45-7	floral rose sweet honey fruity tropical	C10H12O2
T3	30.733	1744	Geranyl acetate	105-87-3	floral rose lavender green waxy	C12H20O2
T4	25.199	1555	Linalyl acetate	115-95-7	sweet green citrus bergamot lavender woody	C12H20O2
T5	9.599	1049	Butyl acetate	123-86-4	ethereal solvent fruity banana	C6H12O2
T6	17.196	1315	3-Hexen-1-ol, acetate	3681-71-8	fresh green sweet fruity banana apple grassy	C8H14O2
T7	18.999	1258	(Z)-3-Hexen-1-ol, formate	33467-73-1	fresh green waxy sweet vegetable grassy sharp	C7H12O2
T8	43.329	2202	Hexadecanoic acid, methyl ester	112-39-0	oily waxy fatty orris	C17H34O2
T9	39.596	2041	Isopropyl myristate	110-27-0	faint oily fatty	C17H34O2
T10	44.463	2232	Isopropyl palmitate	142-91-6	bland oily	C19H38O2
T11	30.663	1755	Methyl salicylate	119-36-8	taste of holly, herbal	C8H8O3
T12	31.796	-	2-hydroxy-Benzoic acid, ethyl ester	118-61-6	sweet wintergreen mint floral spicy balsam	C9H10O3
T13	40.996	2188	Methyl anthranilate	134-20-3	fruity grape orange flower neroli	C8H9NO2
T14	33.329	1863	Butanoic acid, phenylmethyl ester	103-37-7	fresh fruity jasmine apricot loganberry	C11H14O2
T15	14.791	1248	(E)-pent-2-en-3-yl acetate	1191-16-8	sweet fresh banana fruity jasmine ripe heliotrope balsam	C7H12O2
T16	26.129	1605	Benzoic acid, methyl ester	93-58-3	floral, cherry scent	C8H8O2
T17	27.729	1644	Benzoic acid, ethyl ester	93-89-0	fruity dry musty sweet wintergreen	C9H10O2
T18	38.129	2033	2-(methylamino)-Benzoic acid, methyl ester	85-91-6	fruity musty sweet neroli powdery phenolic wine	C9H11NO2
T19	37.063	1991	dihydro-5-pentyl-2(3H)-Furanone	104-61-0	coconut creamy waxy sweet buttery oily	C9H16O2
			Alcohols
T20	19.53	1321	(E)-2-Hexen-1-ol	540-18-1	sweet fruity banana pineapple cherry tropical	C6H12O
T21	33.396	1901	Phenylethyl Alcohol	60-12-8	floral rose dried rose flower rose water	C8H10O
T22	32.329	1877	Benzyl alcohol	100-51-6	floral rose phenolic balsamic	C7H8O
T23	32.529	1822	Geraniol	106-24-1	sweet floral fruity rose waxy citrus	C10H18O
T24	31.262	1826	(Z)-3,7-dimethyl-2,6-Octadien-1-ol	106-25-2	sweet natural neroli citrus magnolia	C10H18O
T25	24.395	1525	Linalool	78-70-6	citrus floral sweet bois de rose woody green blueberry	C10H18O
T26	18.796	1390	(Z)-3-Hexen-1-ol	928-96-1	fresh green cut grass foliage vegetable herbal oily	C6H12O
T27	18.863	1394	(E)-3-Hexen-1-ol	928-97-2	green cortex privet leafy floral petal oily earthy	C6H12O
T28	19.399	1420	trans-2-Hexenol	928-95-0	fresh green leafy fruity unripe banana	C6H12O
T29	18.933	1410	3-Cyclopentyl-1-propanol	928-92-7	green herbal musty tomato metallic	C8H16O
T30	21.934	1430	1-Octen-3-ol	3391-86-4	mushroom earthy green oily fungal raw chicken	C8H16O
T31	11.658	1158	1-Penten-3-ol	616-25-1	ethereal horseradish green radish chrysanthemum vegetable tropical fruity	C5H10O
T32	21.798	1468	6-methyl-5-Hepten-2-ol	1569-60-4	sweet oily green coriander	C8H16O
T33	30.533	1755	(E)-3,7-Dimethyl-2-octen-1-ol	106-22-9	floral leather waxy rose bud citrus	C10H20O
T34	30.399	1766	(Z)-3,7-Dimethyl-2-octen-1-ol	6812-78-8	floral red rose fruity metallic	C10H20O
T35	17.999	1334	3-Methylpentan-1-ol	589-35-5	fusel cognac wine cocoa green fruity	C6H14O2
T36	28.929	1618	dl-Menthol	89-78-1	peppermint cool woody	C10H20O
No.	RT	RI	Compounds	CAS	Odour	Formula
T37	42.596	2228	α-Cadinol	481-34-5	herb wood	C15H26O
T38	39.999	2088	(-)-Spathulenol	6750-60-3	earthy herbal fruity	C15H24O
T39	18.129	1339	1-Hexanol	111-27-3	ethereal fusel oil fruity alcoholic sweet green	C6H14O
T40	10.929	1125	1-Butanol	71-36-3	fusel oil sweet balsam whisky	C4H10O
T41	25.801	1537	1-Octanol	111-87-5	waxy green orange aldehydic rose mushroom	C8H18O
T42	46.329	2363	1-Hexadecanol	36653-82-4	waxy clean greasy floral oily	C16H34O
T43	36.465	1414	1-Dodecanol	112-53-8	earthy spicy herbal sweet mushroom hay blueberry	C12H26O
T44	14.468	1217	1-Pentanol	71-41-0	fusel oil sweet balsam	C5H12O
T45	9.868	1218	3-Methyl-1-butanol	123-51-3	fusel oil alcoholic whisky fruity banana	C5H12O
T46	21.333	1447	1-Heptanol	111-70-6	musty leafy violet herbal green sweet woody peony	C8H16O2
T47	5.929	925	Ethanol	64-17-5	strong alcoholic ethereal medical	C2H6O
T48	11.796	1104	Allyl Alcohol	107-18-6	pungent mustard	C3H6O
			Aldehydes
T49	10.129	1078	Hexanal	66-25-1	fresh green fatty aldehydic grass leafy fruity sweaty	C6H12O
T50	22.996	1494	Benzaldehyde	100-52-7	strong sharp sweet bitter almond cherry	C7H6O
T51	23.996	1498	Decanal	112-31-2	sweet aldehydic waxy orange peel citrus floral	C10H20O
T52	28.601	1598	Undecanal	112-44-7	waxy soapy floral aldehydic citrus green fatty fresh laundry	C11H22O
T53	22.329	1700	Dodecanal	112-54-9	soapy waxy aldehydic citrus green floral	C10H20O
T54	35.668	1822	Tridecanal	10486-19-8	fresh clean aldehydic soapy citrus petal waxy grapefruit peel	C13H26O
T55	20.462	1398	Nonanal	124-19-6	waxy aldehydic rose fresh orris orange peel fatty peel	C9H18O
T56	16.996	1293	Octanal	124-13-0	aldehydic waxy citrus orange peel green fatty	C8H16O
T57	29.799	1696	Neral	106-26-3	sweet citral lemon peel	C10H16O2
T58	25.602	1532	(E)-2-nonenal	18829-56-6	fatty green cucumber aldehydic citrus	C9H16O
T59	27.866	1630	(E)-2-decenal	3913-81-3	waxy fatty earthy coriander green mushroom aldehydic	C10H18O
T60	21.735	1424	(E)-2-octenal	2548-87-0	fresh cucumber fatty green herbal banana waxy green leaf	C8H14O
T61	13.668	1205	(E)-2-hexenal	6728-26-3	green banana aldehydic fatty cheesy	C6H10O
No.	RT	RI	Compounds	CAS	Odour	Formula
T62	13.729	1193	2-hexenal	505-57-7	sweet almond fruity green leafy apple plum vegetable	C6H10O
T63	14.196	1236	(Z)-4-Heptenal	6728-31-0	oily fatty green dairy milky creamy	C7H12O
T64	6.535	984	Pentanal	110-62-3	fermented bready fruity nutty berry	C5H10O
T65	13.463	1181	Heptanal	111-71-7	fresh aldehydic fatty green herbal wine-lee ozone	C7H14O
T66	33.065	1755	2-Undecenal	2463-77-6	fresh fruity orange peel	C11H20O
T67	23.268	1454	(E,E)-2,4-Heptadienal	4313-03-5	fatty green oily aldehydic vegetable cake cinnamon	C7H10O
			Ketones
T68	35.459	1917	trans-β-Ionone	79-77-6	dry powdery floral woody orris	C13H20O
T69	17.995	1323	6-methyl-5-Hepten-2-one	110-93-0	citrus green musty lemongrass apple	C8H14O
T70	35.335	1805	2-Tridecanone	593-08-8	fatty waxy dairy milky coconut nutty herbal earthy	C13H26O
T71	24.602	1495	2-Decanone	693-54-9	orange floral fatty peach	C10H20O
T72	41.735	2021	2-Pentadecanone	2345-28-0	fresh jasmine celery	C15H30O
T73	35.329	1972	(Z)-3-methyl-2-(2-pentenyl)-2-Cyclopenten-1-one	488-10-8	woody herbal floral spicy jasmine celery	C11H16O
T74	16.001	1281	Cyclohexanone	108-94-1	minty acetone	C6H10O
T75	7.329	918	3-methyl-2-Butanone	563-80-4	camphor	C5H10O
T76	10.996	1132	3-Penten-2-one	625-33-2	fruity acetone phenolic fishy	C5H8O
T77	6.396	932	Methyl vinyl ketone	78-94-4	sweet	C4H6O
T78	26.862	1627	Acetophenone	98-86-2	sweet pungent hawthorn mimosa almond acacia chemical	C8H8O
T79	6.929	955	2,3-Butanedione	431-03-8	strong butter sweet creamy pungent caramel	C4H6O2
T80	10.193	1091	1-Hexen-3-one	1629-60-3	cooked vegetable metallic	C6H10O
T81	29.933	1708	2-Dodecanone	6175-49-1	fruity citrus floral orange	C12H22O
T82	15.529	1290	1-hydroxy-2-Propanone	116-09-6	pungent sweet caramellic ethereal	C3H6O2
T83	12.065	1150	Cyclopentanone	120-92-3	mity	C5H8O
			Alkenes
T84	31.729	1745	α-Farnesene	502-61-4	citrus herbal lavender bergamot myrrh neroli green	C15H24
No.	RT	RI	Compounds	CAS	Odour	Formula
T85	13.395	1164	β-Myrcene	123-35-3	peppery terpene spicy balsam plastic	C10H16
T86	15.991	1233	(Z)-β-Ocimene	13877-91-3	citrus tropical green terpene woody green	C10H16
T87	25.258	1475	α-Copaene	3856-25-5	woody spicy honey	C15H24
T88	11.199	1096	β-Pinene	127-91-3	dry woody resinous pine hay green	C10H16
T89	11.866	1113	Sabinene	3387-41-5	woody terpene citrus pine spice	C10H16
T90	13.459	1157	α-Phellandrene	99-83-2	citrus herbal terpene green woody peppery	C10H16
T91	13.866	1172	α-Terpinene	99-86-5	woody terpene lemon herbal medicinal citrus	C10H16
T92	17.399	1274	Terpinolene	586-62-9	fresh woody sweet pine citrus	C10H16
T93	23.993	1459	α-Cubebene	17699-14-8	herbal waxy	C15H24
T94	31.259	1727	α-Muurolene	10208-80-7	woody	C15H24
T95	16.066	1142	3-Carene	13466-78-9	citrus terpenic herbal pine solvent resinous phenolic cypress medicinal woody	C10H16
T96	14.532	1189	Limonene	138-86-3	citrus herbal terpene camphor	C10H16
T97	15.063	1195	β-Phellandrene	555-10-2	mint terpentine	C10H16
T98	15.662	1234	(Z)-3,7-dimethyl-1,3,6-Octatriene	3338-55-4	warm floral herb flower sweet	C10H16
T99	15.458	1250	(E)-β-Ocimene	3779-61-1	sweet herbal	C10H16
T100	15.061	1236	Styrene	100-42-5	sweet balsam floral plastic	C8H8
T101	8.863	1043	Toluene	108-88-3	sweet	C7H8
			Nitrogen-containing compounds
T102	45.129	2426	Indole	120-72-9	animal floral moth ball faecal naphthelene	C8H7N
T103	22.668	1470	Pyrrole	109-97-7	sweet warm nutty ethereal	C4H5N
T104	12.266	1196	Pyridine	110-86-1	sickening sour putrid fishy amine	C5H5N
T105	14.868	1256	methyl-Pyrazine	109-08-0	nutty cocoa roasted chocolate peanut green	C5H6N2
T106	19.329	1390	2,4,5-trimethyl-Thiazole	13623-11-5	musty nutty vegetable cocoa hazelnut chocolate coffee	C6H9NS
T107	34.796	1937	Benzothiazole	95-16-9	sulphury rubbery vegetable cooked nutty coffee meat	C7H5NS
No.	RT	RI	Compounds	CAS	Odour	Formula
			Others
T108	35.329	1956	Phenol	108-95-2	phenolic plastic rubber	C6H6O
T109	37.133	2057	p-Cresol	106-44-5	phenolic narcissus animal mimosa	C8H8O2
T110	39.729	2167	Eugenol	97-53-0	sweet spicy clove woody	C10H12O2
T111	34.399	1959	2-Methoxy-5-methylphenol	93-51-6	spicy clove vanilla phenolic medicinal leathery woody smoky burnt	C8H10O2
T112	36.662	2030	Methyleugenol	93-15-2	sweet fresh warm spicy clove carnation cinnamon	C11H14O2
T113	31.733	1862	Guaiacol	90-05-1	phenolic smoke spice vanilla woody	C7H8O2
T114	22.062	1501	1-(2-furanyl)-Ethanone	1192-62-7	sweet balsam almond cocoa caramel coffee	C6H6O2
T115	14.868	1200	2-Pentylfuran	3777-69-3	fruity green earthy beany vegetable metallic	C9H14O
T116	20.596	1432	Furfural	98-01-1	sweet woody almond fragrant baked bread	C5H4O2
T117	26.063	1613	2-Furanmethanol	98-00-0	alcoholic chemical musty sweet caramel bread coffee	C5H6O2
T118	11.529	1176	o-Xylene	95-47-6	geranium	C8H10
T119	11.665	1140	1,3-dimethyl-Benzene	108-38-3	plastic	C8H10
T120	50.863	2456	Coumarin	91-64-5	sweet hay tonka new mown hay	C9H6O2
T121	11.398	1131	4-methyl-3-Penten-2-one	141-79-7	pungent earthy vegetable acrylic	C6H10O

RT: Retention time. RI: The compounds RI from the NIST library. CAS: Chemical Abstracts Service number. All compounds listed in this table were detected in at least one jasmine tea sample. Detailed presence/absence profiles (specific occurrence/absence data for each sample) are available upon reasonable request to the corresponding author.

). Specifically, 49 and 70 volatile compounds were identified in G1 and GH1 respectively, 56 and 77 in G2 and GH2, and 63 and 84 in G3 and GH3.

All compounds were categorized into esters (19), alcohols (29), aldehydes (19), ketones (16), alkenes (18), nitrogen-containing compounds (6), and other compounds (14), as shown in Figure 3A. From a compound category perspective, the chemical composition patterns of all tea leaf samples (G1–G3) were highly similar, indicating that the fundamental aroma compound profiles of different grade tea leaves are largely consistent. Esters and alcohols were the primary contributors to absolute content, while other compound categories accounted for lower proportions. Figure 3A reveals differences in chemical profiles between the samples from the perspective of relative composition. However, following the scenting process, the chemical composition of the finished samples (GH1–GH3) underwent significant restructuring. Although esters and alcohols remained predominant, the relative proportions of other components—such as nitrogenous compounds and alkenes—increased markedly. This rendered the finished tea’s aroma composition more complex and balanced. This transformation constitutes the key chemical manifestation through which the scenting process endows jasmine tea with its distinctive, multi-layered aromatic profile.

Figure 3B presents a cumulative absolute content diagram for various aroma-active compounds. Analysis indicated the scenting process exerts a highly significant influence on total aroma content. All finished jasmine tea samples (GH1, GH2, GH3) exhibited substantially higher total volatile compound content than their corresponding raw tea leaves (G1, G2, G3). The total content in the high-grade finished product GH1 was significantly higher than that in its corresponding raw tea base G1. This phenomenon directly confirms that the scenting process greatly promotes the enrichment of aroma compounds within the tea leaves. Furthermore, within the finished product group, the total aroma content showed a trend positively correlated with the raw material grade (GH1 > GH2 > GH3), indicating that higher-grade tea bases possess superior adsorption potential, forming the material basis for high aroma loading capacity. The findings indicate that the fundamental quality of the tea base constitutes the pivotal factor in determining the flavour characteristics of jasmine tea. Premium tea bases (such as Grade G1), owing to their raw material properties, provide an ideal medium for the adsorption and transformation of aromatic compounds, thereby establishing the foundation for the profound development of flavour.

3.2.1. Alcohols

Twenty-nine types of alcohols were identified across the three different grades of tea base material and the resulting finished jasmine tea samples. In jasmine tea, linalool imparts floral, fruity, and woody notes, while benzyl alcohol contributes a sweet, toasty, mellow, fruity, and citrus-like aroma. Both these scents are present in both the tea liquor and the jasmine flowers [11]. Here, the relative content of 3-hexen-1-ol among the alcohols was lower than that of linalool and benzyl alcohol. Concurrently, these compounds were closely associated with sensory attributes of grassy and lettuce-like aromas [12].Furthermore, among the alcohols, cyclopentanone, 1-hexanol, (Z)-linalool oxide, and (E)-linalool oxide were negatively correlated with jasmine tea grade. These four volatile compounds are also present in green tea, with prior studies indicating negative correlations between cyclopentanone, (Z)-linalool oxide, and (E)-linalool oxide and green tea quality [13]. Phenethyl alcohol, α-pinene, and geraniol have also been reported to originate from jasmine flowers, contributing floral or sweet notes [14].

3.2.2. Esters

Nineteen esters were identified in the three different grades of tea base material and the resulting finished jasmine tea samples. These esters accounted for 53.55% of the identified volatile organic compounds (VOCs), exhibiting the greatest increase in concentration from raw tea leaves to the finished product. Previous studies have reported that esters such as benzyl acetate, methyl salicylate, methyl phthalate, and 3-hexen-1-ol benzoic acid, along with terpenes such as α-farnesene, linalool, and geraniol, are key aroma compounds in jasmine flowers [15]. This indicates that most aroma compounds in jasmine tea are absorbed from the jasmine flowers, consistent with our findings. Among these, benzyl acetate possesses floral and fruity notes [15], while (Z)-3-hexanol benzoate exhibits green, spicy, and woody notes, with the herbaceous fraction displaying prominent jasmine floral characteristics [16]. Methyl o-aminobenzoate is described as resembling the peachy, sweet, fruity, grape-like aroma derived from jasmine [3]. Methyl salicylate is perceived as a sweet, spicy, minty, holly-like odour and is considered a significant aroma compound in black tea [17]. Additionally, through weighted gene co-expression network analysis (WGCNA) [18] of jasmine black tea [13], seven compounds have been identified as key aroma constituents in jasmine tea: benzyl alcohol, linalool, (Z)-3-hexenyl acetate, methyl benzoate, benzyl acetate, indole, and α-farnesene as the principal aroma compounds in jasmine black tea, consistent with our findings. Notably, most volatile ester compounds correlate positively with jasmine tea grade and originate from the jasmine flower.

3.2.3. Aldehydes

Nineteen aldehydes were identified in the three grades of tea leaves and their corresponding jasmine tea samples. Although aldehydes constituted only 4.7% of the identified volatile organic compounds (VOCs), their low odour threshold [14] rendered them highly significant contributors to aroma performance. Among these, benzaldehyde (contributing almond, sugar, and burnt notes) and decanal (imparting herbal, fatty, and citrus aromas) have been demonstrated to play pivotal roles in fragrance development [19]. Compared to finished jasmine tea, green tea leaves exhibit significantly reduced aldehyde content (Figure 3B), including lipid-derived aldehydes, such as nonanal and decanal. These lipid-derived compounds are heat-induced and likely heat-released. Multiple additional heat treatments post-floral scenting can reduce these volatiles in scented teas [20].

3.2.4. Ketone Compounds

Sixteen ketones were identified in the three grades of tea leaves and their corresponding jasmine tea samples. Although ketones constitute a relatively minor proportion of the overall volatile constituents, their rich aromatic properties play a crucial supplementary and modifying role in shaping the complex bouquet of jasmine tea. Methylheptenone, as a representative ketone compound, exhibits pronounced green fruit and fatty aromas, imparting a refreshing green note to the tea’s fragrance [21]. Concurrently, the detected β-ionone possesses characteristic violet floral and woody aromas, significantly enhancing the fragrance’s elegance and complexity [22].

3.2.5. Alkenes

Eighteen alkenes were identified in the three grades of tea leaves and their corresponding finished jasmine tea samples. Despite their abundance, their contribution to tea aroma is limited. Among these, α-farnesene was present in the highest quantity. Possessing floral and herbaceous notes, it is recognized as one of the key aroma components in jasmine tea. In this study, both α-farnesene and linalene exhibited higher concentrations in the finished jasmine tea than in the raw tea leaves. Furthermore, linalene and α-farnesene showed positive correlations with jasmine tea grade, consistent with reports attributing their origin to jasmine flowers [23]. Conversely, α-pinene and limonene exhibited negative correlations with grade.

3.2.6. Nitrogen-Containing Compounds

Six nitrogen-containing compounds were identified across the three grades of tea leaves and their corresponding jasmine tea samples. Indole, common to all five jasmine teas, exhibited significantly higher concentrations in the finished tea than in the raw leaves. It contributes nutty, floral, camphor ball, and burnt aromas [24]. Previous studies have largely identified indole alongside five other compounds—α-farnesene, (Z)-3-hexenyl benzoate, benzyl acetate, linalool, and methyl o-aminobenzoate—as key contributors to jasmine tea’s aromatic quality [3]. These compounds also exhibit a positive correlation with grade.

3.3. PCA OPLS-DA Analysis of Three Different Grade Tea Leaf Samples and Their Jasmine Tea Samples

PCA is one of the most widely applied data dimensionality reduction algorithms. It is an unsupervised clustering method that transforms raw data into two new indicators, Principal Component 1 (PC1) and Principal Component 2 (PC2). Through PC1 and PC2, the differences between samples can be maximized for analysis [25]. This study investigated the aromatic active compounds in three different grades of tea leaf samples. The PCA plot derived from the characterization of these aromatic compounds is shown in Figure 4A. PC1 and PC2 explained 53.6% and 42.9% of the variance, respectively, with a cumulative contribution rate of 96.5%, thus accounting for the vast majority of the original variation. The distribution of the three grade samples exhibits a clear separation trend: G1 grade samples (green) cluster on the right side, G2 grade samples (blue) occupy the upper-middle region, while G3 grade samples (red) gather in the lower-left area. This distinct spatial separation indicates fundamental differences in the baseline metabolite composition among the different grades of tea leaves. This result confirms that the fundamental quality of tea leaves is graded, laying the foundation for their differing adsorption capacities and flavour potentials during subsequent scenting processes. Moreover, this variation can be effectively captured by the PCA model. Furthermore, OPLS-DA stands as one of the most crucial supervised statistical methods in discriminant analysis. This approach enables more precise differentiation among tea leaves of varying grades. Additionally, OPLS-DA remains one of the most significant supervised statistical methods within discriminant analysis. Its application facilitates more precise differentiation among tea leaves of varying grades. As shown in Figure 4B, the OPLS-DA model clearly separates the three grades of tea leaves. Samples G1 (green), G2 (blue), and G3 (red) cluster in the upper-right, middle-left, and lower-right regions, respectively. The Hotelling’s T² ellipse with a 95% confidence interval indicates statistically significant differences between the groups. The model’s validity confirms that chemical characteristics underpin the inherent quality grades of tea leaves. Analysis of the OPLS-DA model yielded the Importance of Projection Variables (VIP), where compounds with VIP > 1 were identified as potential markers distinguishing grade levels. As shown in Figure 4C, multiple aromatic compounds with VIP > 1 were identified, including those with the highest VIP values: benzyl acetate (floral), decanal (fruity), methyl salicylate (minty), and linalool (floral). These compounds serve not only as key chemical indicators for distinguishing tea leaf grades but, more importantly, constitute the fundamental aroma profile of tea leaves. The enrichment of these pleasant aroma components in high-grade tea leaves (e.g., G1) is a prerequisite for their role as premium carriers. Through the scenting process, these leaves ultimately develop the distinctive fresh, vibrant, rich, and mellow flavour profile. Model validation was conducted via 200 replacement tests, with results shown in Figure 4. Both R² and Q² exceeded 0.5, indicating acceptable model fitting. No overfitting was observed, confirming the model’s validity.

Aromatic active compounds were studied in three different grades of tea base (G1, G2, G3) and their corresponding jasmine tea products (GH1, GH2, GH3). The PCA plot derived from the characterization of aroma compounds is shown in Figure 4E. Across all samples, PC1 and PC2 accounted for 70.6% and 14.3% of the variance, respectively, with a cumulative contribution rate of 84.9%, explaining the majority of the original variation. Along the horizontal axis (t[1]), all raw tea samples (G1, G2, G3) clustered in the right region of the plot, while all finished tea samples (GH1, GH2, GH3) aggregated in the left region. The two groups were distinctly separated, indicating significant differences. This clearly demonstrates that the core characteristics of tea leaves, such as metabolite composition, undergo fundamental changes after scenting treatment. This confirms that the scenting process is the key step in imparting the signature floral aroma to jasmine tea. As shown in Figure 4F, the OPLS-DA model exhibits excellent classification parameters (R²_X = 0.999, R²_Y = 0.989, Q² = 0.977), indicating strong predictive capability. The sample distribution clearly separates into two major clusters: the raw tea samples (G1, G2, G3) densely cluster in the right region, while the finished product samples (GH1, GH2, GH3) are entirely grouped in the left region. This strongly corroborates the role of scenting in enhancing and reshaping the flavour profile of raw tea (e.g., significant enrichment of the aroma components). Furthermore, distinct separation along the vertical axis is observed among the finished tea grades (GH1, GH2, GH3), potentially reflecting differences in aroma precursor substances inherent to each raw tea grade. This indicates that the baseline quality of raw tea serves as the foundation for the efficacy of the scenting process. Analysis of the OPLS-DA model yielded the importance of projected variables (VIP), identifying eight active aroma compounds with VIP > 1. The VIP values, in descending order, were: benzyl acetate (floral), methyl anetate (fruity), indole (floral), decanal (sweet aroma), methyl salicylate (minty aroma), methyl benzoate (floral aroma), linalool (floral aroma), and benzyl alcohol (floral aroma). These compounds were the primary drivers of significant separation between the raw tea leaves and the finished products. Their enrichment levels directly reflected the chemical transformations induced by the scenting process. The model was validated through 200 substitution tests, with the results shown in Figure 4H. The intersection point of the Q2 regression line with the vertical axis is well below zero, and both R² and Q2 exceed 0.5. This indicates no overfitting in the model, confirming its validity. The results demonstrate that this OPLS-DA model possesses strong explanatory and predictive capabilities for classifying raw tea leaves and finished products, yielding reliable outcomes.

3.4. Screening Aromatic Active Compounds Contributing to Dominant Flavour Characteristics in Jasmine Tea

Based on Figure 4G, aromatic volatiles with a Variable Importance (VIP) > 1 were collated, identifying eight distinct volatile compounds: benzyl acetate (floral), methyl anetate (fruity), indole (floral), decanal (sweet aroma), methyl salicylate (minty aroma), methyl benzoate (floral aroma), linalool (floral aroma), and benzyl alcohol (floral aroma). However, judging their importance solely by aroma content is insufficient to determine whether these substances contribute to the aroma quality of jasmine tea. Therefore, it is necessary to introduce r-OAV to evaluate the contribution of these aroma components, ultimately identifying the aroma compounds responsible for the distinctive fragrance quality of jasmine tea. Consequently, the primary aroma contributors were further screened based on r-OAVs > 1, calculated according to aroma thresholds and concentrations [26]. The five aroma-active compounds in jasmine tea identified by GC×GC-O-MS) are presented in

Table 3. Aromatic Active Compounds in Three Different Grade Tea Blanks and Their Finished Jasmine Tea Samples Identified by GC×GC-O-MS.

NO.	CAS	Compound	Odour	r-OAV
				G1	GH1	G2	GH2	G3	GH3
1	66-25-1	Hexanal	Grass	1	1	2	3	2	2
2	123-35-3	Myrcene	Grass	40	450	65	247	6	126
3	124-13-0	Cinnamaldehyde	Citrus	2	1	6	5	1	1
4	3681-71-8	Linalyl acetate	Sweet	1	18	1	21	5	6
5	110-93-0	Methylheptenone	Menthol	<1	3	<1	12	<1	2
6	928-96-1	Geraniol	Fresh	2	4	2	10	2	11
7	124-19-6	Nonanal	Fatty	6	2	29	15	3	5
8	112-31-2	Guaiacol	Sweet	1	40	8	26	8	14
9	78-70-6	Linalool	Floral	89	216	57	210	73	124
10	93-58-3	Methyl benzoate	Floral	5	16	2	11	3	8
11	140-11-4	Benzyl acetate	Floral	<1	6	<1	6	<1	4
12	119-36-8	Methyl salicylate	Fruity	4	27	4	16	7	13
13	97-53-0	Eugenol	Sweet	12	43	7	33	2	26
14	106-24-1	Geraniol	Sweet	18	51	17	42	13	41
15	100-51-6	Benzyl alcohol	Floral	<1	<1	<1	<1	<1	<1
16	134-20-3	Methyl Anisate	Fruity	41	524	43	346	44	564
17	120-72-9	Indole	Floral	3	22	3	45	5	27

r-OAV = Concentration (ng/g)/Threshold (mg/kg).

, where linalool, myrcene, and methyl anetate exhibit higher r-OAV values. Combined with the analysis of components exhibiting VIP values > 1, seven compounds—benzyl acetate, methyl anetate, indole, linalool, benzyl alcohol, methyl salicylate, and methyl benzoate—emerged as the primary aroma components distinguishing tea leaves from jasmine tea, constituting the characteristic aroma profile of jasmine tea.

Linalool serves as a core component in numerous aromatic plant essential oils, constituting up to 70% of volatile concentrates [27]. In recent years, it has garnered attention for its demonstrated bioactive potential in alleviating depressive symptoms [28]. Benzyl acetate, a commonly used fragrance ingredient, is widely incorporated into cosmetics, premium perfumes, shampoos, soaps, and other personal care products [29]. Research indicates that this compound possesses a relative odour activity value (r-OAV) greater than 1 in jasmine tea, signifying its significant contribution to the aroma profile, whereas its r-OAV typically falls below 1 in green tea [18]. Benzyl alcohol, possessing both ‘fruity’ and ‘floral’ characteristics, is prevalent in numerous tea varieties [21]. Methyl benzoate, meanwhile, constitutes a common component of floral volatiles released by many flowering plants. Methyl salicylate, a characteristic aroma marker in grapes and wine, also stimulates plants’ resistance defence mechanisms. Methyl o-aminobenzoate is sensorially described as evoking peach, sweet fruit, and grape-like aromas reminiscent of jasmine origins, with its content in jasmine tea positively correlating with quality grades. Indole, a typical aroma component of jasmine flowers characterized by floral notes, significantly influences the overall quality of jasmine [30] and is also an important aroma substance in tea leaves [31]. The aforementioned volatile compounds have all been identified as key aroma components in jasmine flowers and tea leaves, collectively shaping the characteristic aroma profile of jasmine tea. This study systematically analysed the synergistic effects of the aforementioned volatile compounds in the aroma composition of jasmine tea, further confirming their status as key aromatic constituents in both jasmine flowers and tea leaves, collectively shaping the distinctive complex fragrance profile of jasmine tea. Compared to previous studies, this work specifically highlights the variations in each component’s contribution to aroma across different tea types and its correlation with quality grades, providing new scientific evidence for the precise evaluation of jasmine tea’s aromatic quality.

3.5. Hierarchical Cluster Analysis and Heatmap

We conducted hierarchical cluster analysis on the 17 compounds in Table 3 exhibiting VIP > 1 and detectable by olfaction during the experiment. The results, presented in Figure 5, clearly distinguish all samples into two major clusters: the green tea leaf group (G1–G3) and the finished product group (GH1–GH3), with significant inter-cluster distances. This differentiation directly confirms the fundamental restructuring effect of the scenting process on the tea leaf metabolome, constituting a key step in imparting the signature floral aroma to jasmine tea. Furthermore, within the finished product group, the samples clustered strictly according to raw material grade (with distinct sub-clusters for GH1, GH2, and GH3), indicating that initial quality differences in the tea leaves were preserved and amplified after scenting. This outcome underscores that the final flavour profile of jasmine tea is predominantly shaped by the inherent quality gradations of the raw tea leaves, with their initial compositional differences serving as the primary determinant of product differentiation. A hierarchical cluster analysis was conducted on the relative odour activity values (r-OAV) of 17 key aroma compounds, with the results presented in Figure 6A,B. Furthermore, a correlation analysis explored the relationship between the sensory attributes of odour components and their r-OAV values, with the results presented in Figure 6C. Analysis indicated that the three aroma compounds contributing most significantly to the overall flavour profile of the samples were X16 (methyl anetate), X2 (linalool), and X9 (linalool), whose r-OAV values were particularly prominent among all compounds.

Classifying these compounds by their aromatic function yields the following four categories:

• Green & Fresh Notes: X1 Hexanal (grass, green leaf), X6 Geraniol (rich fresh grass scent), X2 Linalool (fresh grass, green leaf), X3 capraldehyde (fatty, citrusy green notes), X7 nonanal (citrus peel, oily aroma), and X8 decanal (intense citrus peel fragrance). These compounds exhibit significantly elevated concentrations in the finished tea, constructing jasmine tea’s ‘freshness’ and preventing an overly sweet aroma. Among these, X8 decanal proved particularly noteworthy, exhibiting a remarkably strong positive correlation with the herbal flavour profile. The radar chart demonstrates that G3 and its corresponding finished tea GH3 perform exceptionally well on the ‘herbal’ dimension. Their chemical counterparts are markedly higher levels of geraniol and decanal, making these compounds particularly prominent in shaping the fresh character.
• Floral Heart Notes: Includes X9 linalool (lily/lily-of-the-valley), X14 geraniol (sweet, mellow rose), X10 methyl benzoate (ylang-ylang, fruity sweetness), and X11 benzyl acetate (core jasmine aroma). This constitutes the core floral unit of jasmine tea. All varieties exhibit exceptionally high concentrations of these compounds in the finished tea. Moreover, these substances all exhibit a strong positive correlation with floral fragrance, freshness, concentration, and purity. Notably, X11 Benzyl acetate, as the characteristic aroma molecule of jasmine, demonstrates exceptionally high levels across all finished teas, directly attesting to the success of the scenting process. The sensory radar chart confirms that GH2 exhibits the most pronounced floral intensity, consistent with its high content of X9 Linalool and X11 Benzyl acetate. These compounds collectively contribute complexity and sweetness to the floral profile.
• Sweet & Fruity Notes: Comprising X12 methyl salicylate (holly-like sweetness), X16 methyl acetate (grape-like sweet fruitiness), and X4 linalyl acetate (fruity character). These compounds furnish the indispensable sweet undertones within jasmine’s bouquet. The sweet aroma exhibits a strong positive correlation with X12 methyl salicylate, whilst sweet, floral and fruity aromas show strong positive correlations with X16 methyl acetate and X4 linalyl acetate. Methyl salicylate (X12) and methyl acetate (X16) exhibit significant increases in concentration within the finished tea. Their synergistic interaction with floral components shapes the characteristic “sweet floral” profile of jasmine tea. This is clearly reflected in the radar chart, where GH1 scores highest in the “sweetness” dimension, aligning with its elevated levels of these sweet and fruity compounds.
• Other significant contributors: X5 Methylheptenone: Exhibits green fruit and fatty aromas, contributing to the green fragrance note. X13 Eugenol: Provides spicy and woody notes, enhancing the fragrance’s “body” and exotic character. Moreover, persistence exhibits an extremely strong positive correlation with X13 eugenol. X15 Benzyl alcohol: Exhibits a faint sweet floral note. Although its aroma intensity is low (high threshold), thermal imaging reveals high concentrations. It typically functions as an aroma solvent and carrier, lending overall fragrance a softer, more enduring quality. X17 Indole: An exceptionally crucial compound. At low concentrations, it imparts jasmine’s “animalic” and “grounding” notes, being key to the complexity of premium jasmine scents. Heat maps reveal moderate levels in finished teas (predominantly pale orange), consistent with expectations as excessive indole yields unpleasant faecal odours. Its precise control reflects masterful craftsmanship. The fragrance of flowers and fruit is positively correlated with X17 indole.

Comparing flavour profiles across the varieties, G3 exhibits greater freshness (high geraniol content) with slightly reduced floral notes, potentially appealing to consumers favouring a crisp profile. G2 exhibits pronounced floral notes (high linalool and benzyl acetate content), with well-balanced flavour and superior quality. G1 resembles G2 but features higher geraniol and benzyl alcohol content, resulting in enhanced sweetness. These findings correlate with the sensory evaluations and aroma component analyses, mutually corroborating the results.

4. Conclusions

This study employed two-dimensional GC×GC-O-MS coupled with chemometric analysis to investigate three grades of tea leaves and their corresponding finished jasmine tea products. Chromatographic analysis of the three tea leaf grades and finished jasmine tea using GC×GC-O-MS identified a total of 121 volatile compounds. Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) was employed to analyse differences in volatile compound content among the three tea leaf grades and their corresponding finished jasmine tea products. Samples from the tea leaf group were positioned at a greater distance from those in the finished product group in the analysis diagram, indicating significant differences. This corroborates that the scenting process is a critical step in imparting the characteristic floral aroma to jasmine tea. Furthermore, distinct separation along the vertical axis was observed among the finished products of varying grades (GH1, GH2, GH3). This separation may correlate with differences in the inherent aroma precursor compounds within the base tea leaves of each grade, indicating that the fundamental quality of the base tea leaves underpins the efficacy of the scenting process. By comparing the concentrations of aroma components and calculating the r-OAV (Relative Aroma Volatility Average) of representative constituents, the contribution of volatile components to aroma quality was evaluated. Among these, benzyl acetate, methyl acetate, indole, decanal, linalool, benzyl alcohol, methyl salicylate, and methyl benzoate emerged as the eight primary aroma compounds distinguishing tea leaves from jasmine tea, constituting the characteristic fragrance profile of jasmine tea. This study confirms that the initial quality of tea leaves constitutes the fundamental basis determining the upper limit of finished jasmine tea’s aroma quality, while the scenting process serves as the core method for shaping its signature floral characteristics. The findings indicate that a high-grade tea base (G1), possessing a pure aroma and a sweet, mellow flavour, exhibited superior fragrance adsorption and carrying capacity. Its finished tea (GH1) demonstrates exceptional performance in jasmine freshness, sweet aroma richness, and integration of tea and floral notes. In contrast, lower-grade tea leaves (G3), possessing weaker inherent aroma qualities and limited adsorption capacity, yielded finished tea (GH3) with a thin fragrance even after identical scenting treatment, demonstrating significantly constrained quality enhancement. This demonstrates that while the scenting process universally imparts jasmine fragrance, its efficacy is markedly constrained by the grade of tea base. The synergistic interaction between premium tea base and precise scenting techniques constitutes a necessary condition for developing the characteristic fresh, vibrant, rich, and mellow flavour profile. In summary, the flavour formation of jasmine tea stems from the foundational role of tea base quality and the decisive influence of the scenting process, together constituting a dual mechanism for shaping its aromatic qualities. This conclusion provides theoretical grounding for raw material selection and process optimization in jasmine tea production, offering guidance for targeted quality control.