Different smoking processes with the special fuel rods: Impart a smoky aroma to Souchong black tea

Abstract

The smoky scent is the most distinctive feature for Souchong black tea. To reduce the dependence on pinewood in the smoking process of Souchong black tea, it is crucial to find an effective alternative smoking material. Five black tea samples were prepared via using specially designed fuel rods as the smoking material in this study. Sensory analysis showed that DS (smoking at the drying stage) had the most favorable aroma, featuring a pleasant smoky aroma with floral and fruity notes. 69 volatile compounds were detected in five tested samples. Key volatiles such as β-caryophyllene, nerolidol, guaiacol, and α-terpineol, known for their woody or smoky aroma, were prominent in both DS and TS (the traditional Lapsang Souchong process) samples (OAV > 1, VIP > 1 and P < 0.05). However, DS exhibited significantly lower concentration of these volatiles than TS, giving it a more pleasant aroma. Additionally, phenylethyl alcohol and α-farnesene were characteristic volatiles in FS (smoking at the fermentation stage) and DS, imparting a sweet, mildly smoky aroma. Therefore, using these specialized fuel rods to smoking process at drying stage is an optimal method for processing Souchong black tea. These findings provide a theoretical foundation for stabilizing Souchong black tea quality, promoting green and low-carbon tea production methods.

Graphical abstract

Hightlights

• •The number and content of phenols increased significantly after smoking.
• •β-caryophyllene, nerolidol, guaiacol, and α-terpineol were sources of smoky aroma.
• •Fuel rods used at drying stage made smoky aroma of black tea more pleasant.

1. Introduction

Black tea is the most widely consumed and traded tea worldwide, encompassing varieties like Souchong black tea, Congou black tea and broken black tea (Chen et al., 2022; Fang, Liu, Xiao, Ma, & Huang, 2023). Lapsang Souchong, a renowned type of tea, is considered the ancestor of black tea and is produced in Tongmu Village, Xingcun Town, in the Wuyi Mountain National Nature Reserve of Fujian, China. This tea is characterized by its unique smoky aroma, achieved by smoking the leaves over burning pinewood that were extensively cut down to provide a source of smoke in a smokehouse. Because of this distinct flavor, Lapsang Souchong has become highly favored among tea consumers. Notably, its processing method was included in the fifth batch of Fujian Province's provincial intangible cultural heritage projects in 2017 (Wu, 2019; Yao, Guo, Lu, & Jiang, 2005). Aroma is a crucial factor in evaluating tea's sensory quality (Huang et al., 2023; Xu et al., 2022; Yener et al., 2016). It results from the interaction of various volatiles at specific concentrations, which stimulate the olfactory nerves. The processing method significantly impacts tea's aroma formation (Shi, Zhu, Ma, Lin, & Lv, 2022; Yao et al., 2023). For instance, the addition of withering light quality, especially blue light, is conducive to the formation of tea floral aroma (Hua et al., 2024). Shaking during the withering has a strong effect on the aroma of Congou black tea (Wu et al., 2024), and the shaking could facilitate the formation of floral and fruity in black tea and improve its quality (Wang et al., 2023). Traditionally, smoking has been widely applied in meat products, where foods absorb the aroma from both the smoke compounds generated by burning wood and from the volatiles already present in the food. Phenols, formed in this process, are primary contributors to the smoky aroma (Hu et al., 2020; Merlo et al., 2021; Yang et al., 2022; Yin et al., 2021).

The smoking process is a distinctive feature of Souchong black tea, responsible for its unique smoky aroma. This aroma is largely influenced by the type of smoking material used. Studies have indicated that Lapsang Souchong black tea smoked using Huangshan pinewood and Masson pinewood from the Wuyi Mountain National Nature Reserve has a richer, smoky aroma, similar to that of dried longan, than tea smoked with materials sourced from outside this area (Shen & Yang, 1989; Yao et al., 2005). Due to the traditional smoking process and origin of Souchong black tea, there is a scarcity of suitable pinewood materials, making the source of its smoky aroma increasingly limited. This shortage creates a demand for sustainable alternatives that could provide a comparable smoky aroma while supporting ecological conservation. For address this issue, Tang et al.(Tang et al., 2021) developed special fuel rods designed specifically for smoking Lapsang Souchong black tea. This innovation not only helps reconcile the economic interests of tea farmers with ecological preservation but also protects the intangible cultural heritage associated with Souchong black tea production. However, it remains unclear how these fuel rods affect the aroma feature of Souchong black tea, necessitating exploration to confirm which stages of smoking are most beneficial for enhancing the smoky aroma characteristics of the tea.

Molecular sensory technology is increasingly employed in tea flavor analysis, utilizing advanced methods to feature volatile and non-volatile substances. For instance, gas chromatography–mass spectrometry (GC–MS) and liquid chromatography–tandem mass spectrometry (LC–MS/MS) metabolomics have been used to study the changes in these substances during the withering stage of black tea processing (Fang et al., 2023). Gas chromatography–electronic nose (GC–E–Nose), gas chromatography–ion migration spectrometry (GC–IMS), and GC–MS have been used to study the evolving volatile profiles and main aroma components in sweet and floral black tea (Yang et al., 2024). Additionally, stir bar sorptive extraction–gas chromatography–olfactory–mass spectrometry (SBSE–GC–O–MS) was applied to examine flavor formation of Jinmudan black tea (Wu et al., 2023), while comprehensive two–dimensional gas chromatography time of flight mass spectrometry (GC × GC–TOF-–MS) helped identify the changing aroma features in Congou black tea (Yang et al., 2020). Techniques such as headspace solid phase microextraction-gas chromatography–mass spectrometry (HS–SPME–GC–MS) have also been employed to detect key aroma compounds in Lapsang Souchong black tea (Li et al., 2024).

This study aims to explore how special fuel rods influence quality on aroma of Souchong black tea and to pinpoint the smoking process stage that yields optimal aroma characteristics. Five black tea samples, each subjected to different smoking processes with these specialized fuel rods, were collected. The sensory quality evaluation, headspace solid phase microextraction gas chromatography-time-of-flight-mass spectrometry (HS-SPME-GC-TOF-MS), along with and odor activity value (OAV), were utilized to compare and analyze sensory evaluation, volatile compounds, and characteristic compounds of these samples. Findings from this study provide a theoretical foundation for stabilizing the quality of Souchong black tea while supporting the green and low-carbon tea production practices.

2. Materials and methods

2.1. Tea samples preparation

Five black tea samples were processing at Fujian Agriculture and Forestry University, Fuzhou City, Fujian, China. Fresh tea shoots from Jinmudan cultivar (Camellia sisnensis var. sinensis cv. Jinmudan) were used, with a picking standard of one bud and two or three leaves. The smoking material consisted of special fuel rods for Lapsang Souchong black tea (Tang et al., 2021). The processing procedure for the sample is shown in Fig.1. They were processed following the conventional Souchong black tea method (Xia, 2014), which included special equipment for smoking the tea at various processing stages. Notably, no open flame was used at the smoking process. Fresh tea leaves were first placed in a withering tank for withering (26 °C for 6 h). The withered leaves were then rolled for 50 min using a 6CRT-35B Rolling Machine (New Fangchun Tea Machinery Co., Ltd., Quanzhou, China) in a sequence of pressures: no pressure for 5 min, light pressure for 15 min, heavy pressure for 15 min, light pressure for 10 min, and no pressure again for 5 min. Traditional fermentation was carried out at 24–27 °C for 4 h with 85–95 % relative humidity, while fermentation with smoke was conducted at 29–34 °C for 2.5 h. Following fermentation, an initial drying step, termed “underwent red pot,” was performed at 200 °C for 2–3 min, followed by rolling for 3–4 min, and final drying at 85 °C until the water content reached 5 %. To assess the effect of smoking at different stages (Fig. 1),

Fig. 1.

The preparation process of five black tea samples with different smoking processes.

CK: black tea without smoking process; WS: black tea that smoking at the withering stage; FS: black tea that smoking at the fermentation stage; DS: black tea that smoking at the drying stage; TS: traditional Souchong black tea.

five samples were prepared: (1) a control without smoking (CK), (2) smoking at the withering stage (WS), (3) smoking at the fermentation stage (FS), (4) smoking at the drying stage (DS), and (5) the traditional Lapsang Souchong process (TS). The finished tea samples were stored at −80 °C for future analysis.

2.2. Chemicals

This study mainly utilized the following chemicals: Benzaldehyde, α-ionone, methyl salicylate, nerolidol, and β-cyclocitral from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China); octanal, nonanal, phenylacetaldehyde, hexanoic acid, and methyl hexanoate from Shanghai Yi'en Chemical Technology Co., Ltd. (Shanghai, China); heptanal and 1-hexanol from Shanghai Zhenzhun Bio-Technology Co., Ltd. (Shanghai, China); and decanoic acid ethyl ester (chromatographically pure, purity≥99.8 %) along with a standard mixture of normal alkanes (C₈-C₃₀) from Sigma (Shanghai, China).

2.3. Sensory evaluation combined with quantitative descriptive analysis

The sensory evaluations were conducted independently by five expert evaluators (two men and three women, average age of 37 years). These evaluations adhered to the black tea method outlined in GB/T 23776–2018 “Tea Sensory Review Methods”, and referenced the aroma evaluation terms for black tea in GB/T 14487–2017, “Tea Sensory Review Terms” and GB/T 13738.3–2012, “Black Tea Part 3: Souchong Black Tea”. For each tea sample (3.0 g), the infusion was prepared by immersing it in the 0.15 L column cup filled with boiled water, and then transferring the brewed tea into a tea bowl after 5 min. The panel assessed the aroma type, concentration, and persistence, providing comments and scores. Each trial was performed three times for consistency.

Quantitative descriptive analysis (QDA) was conducted based on prior studies (Hao et al., 2023). The brewed tea samples were anonymous and appraised by expert panel members, who rated the aroma attributes with a scale from 0 to 5: 0 indicated no aroma, 1 represented identifiable, 2 indicated weak, 3 represented moderate, 4 represented strong, and 5 expressed extremely strong. Each sample was evaluated three times at different intervals. Informed consent was obtained from all participants, and protocols for protecting participants' rights and privacy were strictly followed throughout the research process.

2.4. Color difference analysis

To measure color, a colorimeter (NR200, Sanenshi Technology Co., Ltd., Shenzhen, China) was used to obtain color parameters. Both the finished tea samples and their infusions were scanned three times, with average values recorded (Wu et al., 2023). The color was displayed digitally using the Hunter color degree values: L*, a*, and b*. Here, L* indicates brightness, ranging from dark (0) to white (100); a* indicates the red-green degree, with positive values showing the degree of red and negative values indicating the degree of green; and b* denotes the yellow-blue degree, with positive values for yellow and negative for blue.

2.5. HS-SPME

Each black tea sample (2 g) was powdered and precisely weighed (± 0.001 g) into a 20 mL headspace bottle. A 0.5 μL solution of decanoic acid ethyl ester (internal standard concentration: 2 mg/mL) was added. The bottle was sealed immediately with a threaded cap and silicone spacer, then mixed on a G560E scroll oscillator, following the method of Zeng et al. (Zeng et al., 2022). Each sample was prepared in triplicate. For instrument stability testing, equal of tea samples were mixed and analyzed on an ACQUITY UPLC I-Class liquid chromatography system. Each test was also repeated three times. Headspace solid phase microextraction (HS-SPME) was performed using a DVB/CAR/PDMS extraction needle (1 cm, 65 μm, Supelco, Bellefonte, PA, USA). The headspace flasks were incubated at 80 °C for 31 min, followed by continuous extraction for 60 min, and subsequently desorbed at 250 °C for 3.5 min. The volatile substances were then analyzed using GC-TOF-MS (Agilent 7890B gas chromatograph, Agilent Technologies, USA; Pegasus HT time-of-flight mass spectrometer, LECO Corporation, USA).

2.6. GC-TOF-MS

The volatiles from the fiber were analyzed via GC-TOF-MS (Zeng et al., 2022). Volatile substances were separated on a Rxi®-5silMS column (30 m × 0.25 mm × 0.25 μm film thickness, Restek, Bellefonte, PA, USA). The injector temperature reached 250 °C, and the sample was injected in without diverging mode. The carrier gas (He) had a purity of 99.99 %, and the flow rate was 1 mL/min. The GC temperature program was as follows: hold at 50 °C for 5 min, increase to 210 °C at 3 °C/min and hold for 3 min, then ramp to 230 °C at 15 °C/min and hold for 3 min. For MS analysis, the electron energy of electron ionization (EI) was 70 eV, with the transmission line and ion source temperatures set at 275 °C and 250 °C, respectively. The acquisition rate was 10 spectra/s, with a mass scanning range of 30–500 m/z.

The volatiles detected by GC-TOF-MS were analyzed as follows: First, the peaks corresponding to these compounds were matched with the mass spectrometry database of the National Institute of Standards and Technology (NIST). Subsequently, the retention index (RI) of the volatiles was confirmed based on the retention times (RT) of normal alkanes C₈-C₃₀. The retention index (recorded as RI₀) from the database, which was determined on the same chromatographic column, was then qualitatively compared to identify the volatile compounds present in the samples (Xiao, Cao, Zhu, Chen, & Niu, 2022). Additionally, the chemical structures, names, CAS numbers, and scent descriptions of the volatile compounds were obtained from online resources for further identification (https://pubchem.ncbi.nlm.nih.gov; https://www.chemicalbook.com; https://www.femaflavor.org/flavorlibrary). The relative concentration of volatiles in five black tea samples was confirmed based on the concentration (μg/kg) and peak area of decanoic acid ethyl ester, along with the peak area of other volatiles. The calculation of RI value followed the method outlined by Van et al.(Van Den Dool and Dec. Kratz, P., 1963), and the relevant equation is provided as follows (1):

$${\mathrm{RI}}_{\mathrm{x}}=100\mathrm{n}+100\times \left({\mathrm{RT}}_{\mathrm{x}}-{\mathrm{RT}}_{\mathrm{n}}\right)/\left({\mathrm{RT}}_{\mathrm{n}+1}-{\mathrm{RT}}_{\mathrm{n}}\right)$$ (1)

In this equation, ‘RI_x’ represents the RI of volatile compounds; ‘RTx’ represents RT of volatile compounds; ‘RT_n’ represents RT of the normal alkane C_n; ‘RT_n+1’ represents the RT of the normal alkane C_n+1.

2.7. OAV

OAV was employed to evaluate the flavor contribution of each aroma compound. An OAV of ≥1 is generally regarded as indicative of a key aroma compound, directly influencing the overall flavor of the tea samples (Xu et al., 2022). The calculation equation is presented below (2):

$${\mathrm{OAV}}_{\mathrm{i}}={\mathrm{C}}_{\mathrm{i}}/{\mathrm{OT}}_{\mathrm{i}}$$ (2)

In this equation, ‘C_i’ represents the relative content of volatile compounds(μg/kg); ‘OT_i’ was the odor threshold in water of the volatiles (μg/kg).

2.8. Statistical analysis

These data were analyzed using Microsoft Excel 2019, and it was also applied to create the radar figure. Bar charts were generated using GraphPad Prism 9, while heatmaps were drawn with TBtools. The orthogonal partial least squares-discriminant analysis (OPLS-DA) diagram was produced using SIMCA 14.1 software. The Duncan method for significance analysis in one-way analysis of variance (ANOVA) was adopted, and SPSS 26.0 software was used.

3. Results and discussion

3.1. Sensory quality accessment

The sensory aroma characteristics of five black tea samples were significantly influenced by different smoking processes (Fig. 2A, B). Among these, the DS Souchong black tea obtained the highest score (Fig. 2C). Specifically, the CK exhibited the most prominent floral and fruity fragrances. The WS presented floral, fruity, and smoky aromas, although they were less pronounced. The FS primarily featured a sweet aroma, accompanied by a faint smoky note. In contrast, the DS displayed a strong and pleasant combination of smoky, woody, floral, and fruity aromas, while the TS highlighted smoky and woody notes, with other aromas being less apparent.

Fig. 2.

(A)The appearance of dry tea and tea infusion. (B) Radar plot of aroma quality characteristics of five black tea samples with different smoking processes. (C) Radial bar chart of the overall score of sensory evaluation. (D) Color difference analysis of dry tea. (E) Color difference analysis of tea infusions.

The intensity of the smoky aroma varied across samples: WS had a slight smoky note, FS a strong pleasant aroma, DS a very strong smoky aroma, and TS a pronounced smoky aroma. These differences in aroma may be attributed to the structural state of the tea samples resulting from the various smoking processes. Tea, being a loose porous solid, contains numerous capillary tubes in its dry state, which facilitate the absorption of volatile substances from the environment (Shi, 1984). Additionally, previous studies have indicated that volatile compounds can alter the quantity and concentration of volatile compounds during processing (Hu et al., 2018). Compared to CK, the samples from WS, FS, DS, and TS exhibited varying degrees of smoky aroma, ranging from slight to strong. This suggests that incorporating a smoking process during the drying stage can enhance the smoky aroma and improve quality of Souchong black tea.

3.2. Color difference analysis of the five black tea samples and their infusions

The color of tea is a critical indicator of quality. A colorimeter was employed to measure the color of five finished tea samples and their respective infusions (Fig. 2A). As shown in Fig. 2D, the L*, a*, and b* values for WS, FS, and CK were significantly higher than those of DS and TS (P < 0.05). This indicates which DS and TS had a reduced degree of red and yellow compared to the deeper hues observed in CK, WS, and FS. This observation aligns with the sensory characteristics of DS and TS, which appeared black-brown, whereas CK, WS, and FS exhibited auburn-brown colors (Table S1). The samples exhibited a range of red degree from high to low: WS, CK, FS, DS, and TS.

According to Fig. 2E, the a* value for the tea infusions of WS and CK was significantly higher than that of the other samples. The L* value for DS and TS was also significantly higher (P < 0.05), indicating greater brightness, while their a* value was notably lower than that of the other three samples, showing a lower degree of red. Additionally, the b* values for WS and FS were significantly lower than those of the other samples (P < 0.05), indicating that the yellow hue was more pronounced in CK, DS, and TS, with DS showing the highest b* value. Within a certain range, the L* value of black tea infusion is negatively correlated with the content of theaflavins and thearubins, while it is positively correlated with the a* and b* values (Lu et al., 2002). This analysis suggests that the color of the DS tea infusion is orange-red and brighter, consistent with sensory evaluation results (Table S1).

3.3. Volatile compounds profiling of the five black tea samples

A total of 69 volatile compounds were identified (Table 1), comprising 18 hydrocarbons, 14 alcohols, 14 phenols, 9 aldehydes, 8 esters, 4 ketones, and 2 other classes. The total relative content of volatiles varied significantly in the samples, with the relative content in DS reaching 6275.05 μg/kg. Notably, the total content of volatiles in DS was notably higher than in the other samples (P < 0.05).

Table 1.

Volatile compounds and their relative content in five black tea samples.

No.	Volatile Compounds	CAS Number	RI	Relative Content/(μg/kg)
				CK	WS	FS	DS	TS
Hydrocarbons				150.83 ± 29.75d	289.52 ± 65.2d	453.28 ± 47.23c	848.93 ± 217.2b	1625.02 ± 73.54a
1	Junipene	3387-41-5	977.46	–	–	–	2.9 ± 1.12a	3.22 ± 0.33a
2	α-Ocimene	502–99-8	1046.61	4.58 ± 1.07a	2.95 ± 0.69b	3.27 ± 0.32b	2.11 ± 0.68bc	1.17 ± 0.28c
3	1-Methyl-4-(1-methylethenyl)-benzene	1195-32-0	1089.98	–	0.14 ± 0.14b	1.62 ± 0.14b	16.02 ± 3.65a	13.51 ± 0.43a
4	3,4-Dimethoxytoluene	494–99-5	1233.97	3.25 ± 0.68b	6.32 ± 1.25b	8.92 ± 1.03b	20.11 ± 5.14a	14.73 ± 0.25a
5	1-Methyl-naphthalene	90–12-0	1292.11	0.33 ± 0.04b	1.57 ± 0.39b	7.63 ± 0.79a	9.81 ± 2.74a	9.9 ± 0.5a
6	Longifolene	475–20-7	1404.07	9.16 ± 1.91d	105.45 ± 23.05c	136.74 ± 13.99c	270.78 ± 69.86b	737.82 ± 29.75a
7	β-Caryophyllene	87–44-5	1415.65	63.18 ± 12.66d	83.34 ± 19.48d	137.26 ± 14.98c	211.74 ± 51.98b	429.13 ± 23.1a
8	7,11-Dimethyl-3-methylene-1,6,10-dodecatriene	77,129–48-7	1424.21	–	1.68 ± 0.4d	3.38 ± 0.45c	5.46 ± 1.47b	16.05 ± 0.55a
9	2-(1-methylethyl)-Naphthalene	2027-17-0	1450.27	–	3.07 ± 0.6c	10.25 ± 1.01b	12.93 ± 2.91b	22.51 ± 0.87a
10	α-Farnesene	502–61-4	1502.72	41.19 ± 7.35b	37.51 ± 8.83b	88.84 ± 8.44a	86.61 ± 21.97a	72.82 ± 3.59a
11	Germacrene D	23,986–74-5	1509.28	1.34 ± 0.25c	1.34 ± 0.27c	1.98 ± 0.22c	3.46 ± 0.99b	5.82 ± 0.25a
12	δ-Cadinene	483–76-1	1515.98	22.92 ± 4.72cd	17.69 ± 3.82d	28.15 ± 3.07bc	33.29 ± 8.45b	43.39 ± 1.45a
13	α-Calacorene	21,391–99-1	1538.28	2.46 ± 0.45c	2.55 ± 0.53c	4.57 ± 0.48c	11.07 ± 2.36b	16.3 ± 0.25a
14	Caryophyllene oxide	1139-30-6	1580.66	1.84 ± 0.27d	3.48 ± 0.71cd	10.2 ± 0.92c	32.55 ± 8.28b	66.86 ± 2.34a
15	1,6-Dimethyl-4-(1-methylethyl)-naphthalene	483–78-3	1659.90	–	1.21 ± 0.31c	1.65 ± 0.28c	9.57 ± 2.25b	16.85 ± 0.8a
16	Phenanthrene	85–01–8	1771.55	0.39 ± 0.34c	18.94 ± 4.2c	7.7 ± 0.85c	85.73 ± 24.22b	121.9 ± 6.56a
17	2,6-Diisopropylnaphthalene	24,157–81-1	1796.63	0.17 ± 0.03b	1.06 ± 0.33b	0.87 ± 0.04b	26.72 ± 6.81a	24.75 ± 1.64a
18	3,6-Dimethyl-phenanthrene	1576-67-6	2033.00	0 ± 0b	1.22 ± 0.19b	0.24 ± 0.24b	8.08 ± 2.33a	8.3 ± 0.58a
Alcohols				2755.23 ± 573.72a	1886.07 ± 434.25a	3035.68 ± 331.18a	3287.25 ± 864.03a	2626.29 ± 92.48a
19	(Z)-2-Penten-1-ol	1576-95-0	782.95	10.05 ± 2.03a	6.66 ± 1.38b	11.48 ± 1.84a	4.05 ± 1.93bc	2.33 ± 0.25c
20	Furfuryl alcohol	98–00-0	861.10	–	–	2.35 ± 0.27b	11.22 ± 3.73a	13.84 ± 0.27a
21	(Z)-3-Hexen-1-ol	928–96-1	863.30	34.56 ± 6.93a	22.22 ± 4.96b	17.78 ± 2.02b	10.08 ± 2.53c	2.91 ± 0.42c
22	Benzyl alcohol	100–51-6	1031.65	60.46 ± 12.58ab	43.51 ± 10.24b	47.84 ± 5.69b	81.21 ± 24.52a	41.39 ± 0.69b
23	trans-Linalool oxide (furanoid)	34,995–77-2	1088.57	621.4 ± 131.11a	402.62 ± 93.98b	488.48 ± 53.95ab	361.38 ± 93.06b	140.14 ± 0.34c
24	Linalool	78–70-6	1096.59	69.51 ± 16.43a	45.24 ± 11.06b	39.24 ± 4.68b	20.83 ± 5.62c	6.81 ± 0.31c
25	Phenylethyl alcohol	60–12-8	1108.05	301.06 ± 66.48ab	207.47 ± 49.2c	314.36 ± 35.35ab	360.22 ± 101.88a	205.94 ± 0.57c
26	Terpinen-4-ol	562–74-3	1179.66	0.58 ± 0.15c	1.45 ± 0.3c	18.5 ± 2.02b	47.56 ± 12.2a	53.93 ± 1.12a
27	α-Terpineol	98–55-5	1195.14	21.52 ± 5.61d	54.88 ± 12.44d	447.3 ± 50.18c	942.07 ± 243.57b	1159.68 ± 39.8a
28	Nerol	106–25-2	1221.91	30.4 ± 6.45a	20.72 ± 5.01a	26.42 ± 3.37a	29.79 ± 8.35a	21.52 ± 0.92a
29	Geraniol	106–24-1	1254.08	1496.18 ± 307.45a	932.04 ± 212.33bc	1466.04 ± 156.88a	1149.05 ± 290.87ab	630.84 ± 30.53c
30	p-Cymen-7-ol	536–60-7	1291.30	0.43 ± 0.12c	0.2 ± 0.2c	0.89 ± 0.15c	3.52 ± 0.73a	2.43 ± 0.68b
31	Nerolidol	7212-44-4	1562.17	106.75 ± 17.94c	146.55 ± 32.5c	151.36 ± 14.77c	244.7 ± 61.27b	324.33 ± 15.41a
32	α-Cadinol	481–34-5	1653.11	2.32 ± 0.44b	2.5 ± 0.63b	3.64 ± 0.02b	21.58 ± 13.77a	20.22 ± 1.17a
Phenols				6.11 ± 1.43b	18.39 ± 6.54b	272.55 ± 36.22b	1512.06 ± 406.37a	1234.27 ± 24.95a
33	Phenol	108–95-2	980.63	–	1.41 ± 0.37b	17.72 ± 2.15b	111.47 ± 32a	105.8 ± 0.33a
34	2-Methyl-phenol	95–48-7	1052.61	–	0.35 ± 0.1b	5.55 ± 0.6b	19.48 ± 5.51a	16.12 ± 0.83a
35	p-Cresol	106–44-5	1077.81	0 ± 0c	1.02 ± 1.02c	14.19 ± 1.71b	46.03 ± 12.68a	38.76 ± 1.01a
36	Guaiacol	‘90–05-1	1084.02	–	1.41 ± 0.35b	30.63 ± 3.37b	172.99 ± 48.45a	157.03 ± 2.05a
37	2-Ethyl-phenol	90–00-6	1134.18	–	0.26 ± 0.26b	2.75 ± 0.41b	10.62 ± 3.09a	8.29 ± 0.25a
38	4-Ethyl-phenol	123–07-9	1165.92	–	–	4.93 ± 4.93b	55.26 ± 15.61a	43.55 ± 0.39a
39	3-Ethyl-phenol	620–17-7	1175.97	–	–	1.31 ± 1.31b	10.72 ± 2.66a	7.42 ± 0.36a
40	4-Methyl-guaiacol	93–51-6	1187.90	0.75 ± 0.18b	1.39 ± 0.36b	39.52 ± 4.6b	182.84 ± 49.46a	156.71 ± 2.41a
41	4-Ethyl-guaiacol	2785-89-9	1272.65	3 ± 0.63c	4.43 ± 0.9c	89.72 ± 9.91c	480.18 ± 128.12a	353.96 ± 8.24b
42	Thymol	89–83-8	1293.05	–	–	2.76 ± 0.33b	12.91 ± 3.42a	12.43 ± 0.48a
43	Carvacrol	499–75-2	1294.92	–	0.91 ± 0.29b	4.38 ± 0.43b	24.23 ± 6.38a	28.69 ± 0.34a
44	Eugenol	97–53-0	1349.45	0.1 ± 0.09b	2.26 ± 0.19b	3.52 ± 0.43b	21.03 ± 5.41a	22.07 ± 0.47a
45	4-Propyl-guaiacol	2785-87-7	1360.37	2.26 ± 0.53c	2.88 ± 0.63c	54.14 ± 5.87c	342.6 ± 88.52a	265.72 ± 7.5b
46	(Z)-2-Methoxy-4-(1-propenyl)-phenol	5912–86-7	1398.01	–	–	1.44 ± 0.18b	21.7 ± 5.07a	17.7 ± 0.28a
Aldehydes				41.45 ± 9.19d	40.99 ± 9.83d	53.28 ± 5.55bc	80.68 ± 20.36a	72.47 ± 3.67ab
47	Hexanal	66–25-1	825.76	3.75 ± 0.83c	9.81 ± 2.34b	7.97 ± 0.92b	11.47 ± 3.15b	15.55 ± 0.86a
48	2-Hexenal	505–57-7	861.42	1 ± 0.24b	1.28 ± 0.3ab	1.73 ± 0a	1.52 ± 0.58ab	1.58 ± 0.3ab
49	Heptanal	111–71-7	907.76	1.17 ± 0.36ab	1.16 ± 0.17ab	1.41 ± 0.1ab	0.9 ± 0.78b	1.79 ± 0.26a
50	5-Methyl-2-furancarboxaldehyde	620–02-0	960.43	–	–	0.23 ± 0.01b	2.62 ± 0.43a	2.13 ± 0.26a
51	Benzaldehyde	100–52-7	961.08	6.02 ± 1.32b	6.25 ± 1.48b	5.78 ± 0.61b	11.42 ± 2.72a	9.95 ± 0.38a
52	Benzeneacetaldehyde	122–78-1	1040.36	14.39 ± 2.99bc	10.83 ± 2.56c	17.2 ± 1.61b	26.39 ± 5.72a	15.65 ± 0.28bc
53	Nonanal	124–19-6	1101.08	6.31 ± 1.76cd	4.25 ± 1.18d	7.99 ± 1.29bc	10.25 ± 2.66ab	12.97 ± 0.62a
54	(E)-2-Nonenal	18,829–56-6	1158.73	2.55 ± 0.33b	2.62 ± 0.75b	4.88 ± 0.44b	13.2 ± 3.65a	11.29 ± 0.35a
55	(E)-3,7-Dimethyl-2,6-octadienal	141–27-5	1267.88	6.26 ± 1.35a	4.8 ± 1.04a	6.09 ± 0.57a	2.9 ± 0.67b	1.57 ± 0.35b
Esters				259.66 ± 32.51a	118.85 ± 26.99b	149.4 ± 11.24b	270.97 ± 74.27a	131.06 ± 16.7b
56	Acetic acid, butyl ester	123–86-4	833.86	6.42 ± 1.28c	4.73 ± 1.4c	10.65 ± 0.85b	16.31 ± 3.92a	7.6 ± 0.37bc
57	γ-Butanolide	96–48-0	912.16	0.32 ± 0.28b	0.46 ± 0.06b	7.43 ± 0.81b	71.93 ± 22.66a	61.86 ± 4.6a
58	Hexanoic acid, ethyl ester	123–66-0	997.82	5.68 ± 1.43a	1.05 ± 0.07c	0.3 ± 0.3d	3.96 ± 0.29b	0.77 ± 0.71d
59	Benzoic acid, methyl ester	93–58-3	1093.08	–	–	–	12 ± 3.32a	11.23 ± 0.54a
60	Methyl salicylate	119–36-8	1190.89	100.09 ± 22.27a	56.43 ± 13.15b	48.63 ± 5.41b	37.21 ± 10.33b	8.39 ± 7.27c
61	Hexanoic acid, hexyl ester	6378-65-0	1387.97	3.78 ± 0.9bc	2.68 ± 0.65c	4.55 ± 0.33bc	9.42 ± 2.65a	5.77 ± 0.27b
62	Dihydroactinidiolide	17,092–92-1	1520.10	3.89 ± 0.55b	3.59 ± 0.66b	7.68 ± 0.84a	9.97 ± 3.49a	8.23 ± 0.28a
63	(Z)-3-Hexen-1-ol, benzoate	25,152–85-6	1570.22	139.48 ± 5.8a	49.91 ± 11bc	70.16 ± 2.68b	110.17 ± 27.61a	27.22 ± 2.67c
Ketones				59.54 ± 9.63c	54.24 ± 12.13c	89.88 ± 7.7ab	143.11 ± 66.93a	137.89 ± 9.82a
64	Cyclopentanone	120–92-3	796.47	8.21 ± 0.42a	4 ± 0.87b	3.9 ± 0.47b	2.07 ± 0.54c	1.36 ± 0.31c
65	Acetophenone	98–86-2	1062.55	0.41 ± 0.17b	0.3 ± 0.14b	1.55 ± 0.33b	6.26 ± 1.42a	5.69 ± 0.47a
66	Geranylacetone	3796-70-1	1446.46	45.05 ± 7.62b	43.68 ± 9.83b	72.78 ± 5.51ab	124.68 ± 62.2a	122.14 ± 8.77a
67	trans-β-Ionone	79–77-6	1478.08	5.88 ± 1.42b	6.26 ± 1.3b	11.65 ± 1.39a	10.1 ± 2.77a	8.7 ± 0.28b
Others				1.05 ± 0.15c	5.11 ± 1.03c	7.19 ± 0.69c	132.05 ± 34.73a	85.2 ± 3.21b
68	Benzonitrile	100–47-0	982.26	–	–	–	22.44 ± 6.32a	14.31 ± 0.52a
69	Dibenzofuran	132–64-9	1507.41	1.05 ± 0.15c	5.11 ± 1.03c	7.19 ± 0.69c	109.61 ± 28.42a	70.89 ± 2.69b
Total				3273.86 ± 656.38d	2413.17 ± 555.97e	4061.27 ± 439.82c	6275.05 ± 1683.89a	5912.21 ± 224.36b

Note:‘-’ indicates that the compound was not detected. Different lowercase letters of superscript in the same line were significant differences in the relative content of compounds (P < 0.05).

Statistical analysis of the volatiles (Fig. 3A, B) uncovered that the relative content of alcohols, hydrocarbons, phenols, esters, and ketones in the five black tea samples was notably elevated, accounting for 96.61 % to 98.70 % of the total volatile concentration. After smoking, the phenols proportion in Souchong black tea increased significantly, with DS showing a notably higher proportion (24.10 %) compared to the other samples (P < 0.05). In comparison to CK, the diversity of phenols in the other samples also significantly increased, particularly in DS and TS. The relative content of phenols in the DS was higher than other samples. Additionally, relative concentration of alcohols across all samples surpassed that of other compounds, making them the mainly constituents responsible for the aroma of black tea. This finding aligns with predecessors' studies (Lian et al., 2015; Lu, Du, & Xiao, 2015). Furthermore, earlier research indicated that hydrocarbons had the highest relative content among the components, followed by alcohols (Li et al., 2022). The discrepancies between these findings and the results of this study may stem from variations in the detection conditions of aroma components, the types of smoking materials used, or the species of tea shoots. Variations in the relative content of volatiles may also contribute to differences in aroma profiles among the five black tea samples. The smoking process can significantly influence types and quantities of volatiles, particularly phenols, in DS and TS samples.

Fig.3.

(A) The types and proportions of volatile compounds in five black tea samples. (B) Number of volatile compounds in all samples. (C) The scatter plots of PLS-DA for finished tea samples. (D) The validation of the OPLS-DA model for all samples. (E) VIP score and corresponding heat maps of volatile compounds relative content in five black tea samples.

3.4. OPLS-DA analysis for the differential volatiles in five black tea samples

OPLS-DA is a supervised analysis mode that reflects the degree of compositional differences between samples. The closer the distance between samples, the more similar their aroma compositions and relative content (Zhang et al., 2018). The model was utilized to analyze the volatiles in five black tea samples, revealing that the inter-group distances were larger than the intra-group distances. This indicates effective separation among the black tea samples (Fig. 3C) and highlights the differences in aroma compounds across the five samples. The reliability of the OPLS-DA model was tested by 200 permutations (Fig. 3D), confirming that the model had good fitting.

The variable importance in projection (VIP) score quantifies the weight of the OPLS-DA model variables. It is generally accepted that substances with a VIP score greater than 1 significantly influence aroma, with higher VIP value indicating a greater contribution of volatiles to the overall aroma of tea (Huang et al., 2022). Based on the normal of VIP > 1 and P < 0.05, 17 different volatiles were identified in black tea samples with varying smoking processes (Fig. 3E). As shown in group a (Fig. 3E), the content of geraniol, trans-linalool oxide (furanoid), (Z)-3-Hexen-1-ol, benzoate, and methyl salicylate in CK were significantly higher than those in Souchong black tea (P < 0.05). As can be seen from group b (Fig. 3E), the content of α-terpineol, longifolene, β-caryophyllene, nerolidol, and several phenolic compounds with woody, spicy, and smoky aromas in DS and TS were significantly higher than in the other samples (P < 0.05). This indicates that the drying process enhances the adsorption of smoky aroma in Souchong black tea. Previous studies have identified longifolene and α-terpineol as key aroma components of Souchong black tea, both derived from pinewood (Su et al., 2024). Additionally, several smoke-derived volatiles, such as guaiacol, 4-methyl guaiacol, and 4-ethyl guaiacol, result from the pyrolysis of pinewood and are characteristic flavor compounds in smoked foods (Li et al., 2024; Yu, Sun, Tian, & Qu, 2008). In summary, volatile compounds known for their strong woody and smoky aromas, including α-terpineol, longifolene, β-caryophyllene, nerolidol, guaiacol, 4-methyl guaiacol, and 4-ethyl guaiacol, showed dramatic increases in both DS and TS.

3.5. OAV analysis for characteristic volatile compounds

OAV is widely used in food flavor research, where it is generally accepted that aroma compounds with an OAV greater than 1 are considered key aroma compounds. Additionally, a higher OAV within a certain range correlates with a greater aroma contribution (Xu et al., 2022). The analysis of the five samples revealed the presence of a total of 30 essential aroma compounds (OAV > 1; Table 2), with the following counts in each sample: 18 in CK, 19 in WS, 25 in FS, 29 in DS, and 26 in TS. Among these, the OAV values for trans-β-ionone (OAV > 830) and β-caryophyllene (OAV > 390), which exhibit violet and woody fragrances, were notably higher than those of other volatiles. Although the relative concentration of these volatiles in the samples was low, their low thresholds resulted in a higher OAV, indicating that trans-β-ionone is the main aroma compound in black tea. This finding is consistent with previous researches (Ho, Zheng, & Li, 2015). Furthermore, benzoic acid methyl ester, characterized by floral and sweet aromas, was detected exclusively in the DS and TS samples, contributing significantly to the aroma profile of DS. Conversely, geranylacetone, which offers fruity, floral, and minty flavors, and α-terpineol, known for its smoky and minty notes, exhibited a decorative effect (OAV < 1) in CK and WS. However, they emerged as the main aroma compounds in DS and TS, indicating a higher degree of compound similarity in these samples.

Table 2.

The key volatile compounds with OAV ≥ 1 in five black tea samples.

Volatile Compounds	Odor Description	OT (μg/kg)	OAV
			CK	WS	FS	DS	TS
trans-β-Ionone	Violet, sweet	0.007	839.65	894.37	1664.38	1442.99	1243.28
β-Caryophyllene	Woody, spicy	0.16	394.90	520.86	857.88	1323.40	2682.08
Terpinen-4-ol	Woody	0.2	2.92	7.27	92.48	237.78	269.63
p-Cresol	Sweet, spicy	0.24	–	4.26	59.11	191.78	161.51
Geraniol	Rose, fruity	7.5	199.49	124.27	195.47	153.21	84.11
trans-Linalool oxide (furanoid)	Floral	6	103.57	67.10	81.41	60.23	23.36
(E)-2-Nonenal	Violet	0.4	6.37	6.54	12.21	33.00	28.23
Benzoic acid, methyl ester	Violet, sweet	0.52	–	–	–	23.08	21.60
δ-Cadinene	Fruity, woody	1.5	15.28	11.79	18.77	22.20	28.93
2-Methoxy-phenol	Smoky -	16	–	<1	1.91	10.81	9.81
Nonanal	Honey, rose, orange	1	6.31	4.25	7.99	10.25	12.97
(Z)-3-Hexen-1-ol	Herb, fresh	1	34.56	22.22	17.78	10.08	2.91
Benzeneacetaldehyde	Sweet, honey-like, floral	4	3.60	2.71	4.30	6.60	3.91
Hexanoic acid, ethyl ester	Floral, fruity	0.61	9.31	1.72	<1	6.50	1.26
Nerolidol	Woody, lily	40	2.67	3.66	3.78	6.12	8.11
2-Methyl-phenol	sweet	3.9	–	<1	1.42	5.00	4.13
Phenylethyl alcohol	Rose, fruity, honey-like	86	3.50	2.41	3.66	4.19	2.39
Benzaldehyde	Almond, floral	3	2.01	2.08	1.93	3.81	3.32
Linalool	Magnolia-like, lemon	6	11.58	7.54	6.54	3.47	1.14
α-Terpineol	Smoky, minty	300	<1	<1	1.49	3.14	3.87
(E)-3,7-Dimethyl-2,6-octadienal	lemon	1	6.26	4.80	6.09	2.90	1.57
Eugenol	Lilac，smoky	10	<1	<1	<1	2.10	2.21
Geranylacetone	Fruity, rose, minty	60	<1	<1	1.21	2.08	2.04
Hexanal	Apple-like, fresh, woody	7	<1	1.40	1.14	1.64	2.22
Acetic acid, butyl ester	Apple-like, banana-like, sweet	10	<1	<1	1.07	1.63	<1
1-Methyl-naphthalene	Aging aroma, camphor	7.5	<1	<1	1.02	1.31	1.32
γ-Butanolide	Caramel	60	<1	<1	<1	1.20	1.03
(Z)-3-Hexen-1-ol, benzoate	Floral	106	1.32	<1	<1	1.04	<1
α-Farnesene	Floral, fresh-sweet	87	<1	<1	1.02	1.00	<1
Methyl salicylate	Almond, caramel, minty	40	2.50	1.41	1.22	<1	<1

Note: “–” indicates that the OAV cannot be calculated. Odor thresholds were obtained from the literature (Yang et al., 2024; Chen et al., 2023; Yang et al., 2022;Yang, Xie, et al., 2024).

Combined with 17 different volatile compounds, 10 characteristic aroma compounds (with OAV > 1, VIP > 1, and P < 0.05) in the samples were further screened (Fig. 4). Based on the type of aroma description analysis, these aroma compounds can be categorized into three clusters: The first cluster has a floral and fruity aromas (Fig. 4I); The second cluster has smoky and woody aromas (Fig. 4. II); The third cluster has sweet aroma (Fig. 4. III). In general, compared to CK, the content of floral and fruity compounds in the four Souchong black tea samples decreased, while the relative content of smoky and woody compounds increased. Additionally, the content of sweet-scented compounds increased in FS and DS. Combined with Table 2, further analysis shows that trans-linalool oxide (furanoid), methyl salicylate, geraniol, and (Z)-3-Hexen-1-ol, benzoate contributed significantly to CK compared to the other samples, and the floral and fruity characteristics in CK were mainly owed to these compounds. The DS and TS primarily exhibit woody and smoky notes, with key compounds including β-caryophyllene, nerolidol, guaiacol, and α-terpineol contributing more to DS and TS than other samples. The aroma compounds, such as phenylethyl alcohol and α-farnesene were particularly notable in FS and DS, where phenylethyl alcohol, known for its fruity and sweet fragrance, was the main characteristic compound of DS. In contrast, α-farnesene, which offers a floral and sweet aromas, was the primary compound in FS. In summary, the pleasant smoky aroma in DS is closely related to its key aroma compounds: β-caryophyllene, nerolidol, guaiacol, α-terpineol, and phenylethyl alcohol. These main components of Souchong black tea align with findings from previous research (Guo, Lv, & Jiang, 2005; Yu et al., 2008).

Fig. 4.

Heat map of characteristic aroma compounds content in five black tea samples.

4. Conclusions

The smoky aroma is a significant characteristic of Souchong black tea. In this study, we employed sensory quality evaluation and HS-SPME-GC-TOF-MS to analyze the volatile compounds in five black tea samples smoked with special fuel rods. We detected 69 volatile compounds. The DS sample had the highest total volatile and phenolic content. Furthermore, ten characteristic volatiles (OAV > 1, VIP > 1, and P < 0.05) were identified. Notable volatiles in the CK, which exhibited floral and fruity aromas, included trans-linalool oxide (furanoid), methyl salicylate, geraniol, and (Z)-3-hexen-1-ol. In contrast, the key volatiles contributing to the woody and smoky aromas in the DS and TS samples were β-caryophyllene, nerolidol, guaiacol, and α-terpineol. The smoky fragrance in the WS and FS samples was weak, while it was excessively strong in the TS. The DS presented a pleasant smoky aroma and superior quality overall. Therefore, special fuel rods promote the aroma quality of Souchong black tea, especially when used at the drying stage. These findings provide a theoretical basis for quality control of Souchong black tea, promoting the green and low-carbon development of tea production. Future research should analyze the quality of experimental samples and Tongmu Village's Lapsang Souchong black tea, and further explore the non-volatile metabolites and quality formation mechanism of Souchong black tea processed using fuel rods.

CRediT authorship contribution statement

Weisu Tian: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Jiao Feng: Writing – review & editing, Supervision, Methodology, Formal analysis, Conceptualization. Jinyuan Wang: Software, Methodology, Investigation, Conceptualization. Hongzheng Lin: Writing – review & editing, Validation, Supervision, Conceptualization. Qianlian Chen: Methodology, Investigation. Jiayun Zhuang: Methodology, Investigation. Guanjun Pan: Investigation, Conceptualization. Jiake Zhao: Investigation, Conceptualization. Lirong Tang: Resources, Project administration, Funding acquisition. Zhilong Hao: Writing – review & editing, Supervision, Project administration, Methodology, Conceptualization.

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.

Appendix A. Supplementary data

Supplementary material

: Table S1. Quality evaluation of the five black tea samples.

Data availability

Data will be made available on request.