Supercritical CO₂ Extraction of Phoenix Dancong Tea Oil: Process Optimization and Fragrance Retention on Textiles

Abstract

Phoenix Dancong tea essential oil possesses unique aroma characteristics and bioactivities, offering broad application potential in the food, pharmaceutical, and daily chemical fields. To achieve efficient extraction and expand its use in functional textiles, supercritical CO₂ (SC-CO₂) extraction was employed to optimize the extraction process of Phoenix Dancong tea essential oil. Based on single-factor experiments, the optimal extraction conditions were determined as follows: pressure of 25 MPa, temperature of 50 °C, CO₂ flow rate of 8 L/h, and extraction time of 3 h, resulting in an essential oil yield of 1.12%. Response surface methodology (RSM) revealed that the experimental data fit the regression model well (R² = 95.49%, R²Adj = 89.69%). Furthermore, the extracted essential oil was blade-coating to cotton, nylon, polyester, and wool fabrics to evaluate its aroma retention performance. Results indicated that cotton fibers exhibited the best absorption and sustained fragrance retention, maintaining a high odor grade even after 8 weeks. This study provides a theoretical basis and practical reference for the green extraction of Phoenix Dancong tea essential oil and its application in smart aromatic textiles.

1. Introduction

Phoenix (Fenghuang) Dancong tea (Camellia sinensis (L.) O. Kuntze), a celebrated oolong variety from Chaozhou, Guangdong Province, China, is renowned for its complex and persistent floral–fruity aroma. Chemical profiling has identified over 80 volatile compounds, primarily monoterpenes (linalool, geraniol, nerol), esters (linalyl acetate, geranyl acetate), and aromatic aldehydes, which, together, define its distinctive sensory character and bioactive potential. Recent analyses report that these volatiles contribute significantly to antioxidant, antimicrobial, and anti-inflammatory activities, supporting applications in cosmetics, functional foods, and smart textiles [1,2,3,4].

Supercritical CO₂ extraction was selected because it provides an environmentally benign, low-temperature, and solvent-free approach that preserves the integrity of delicate aroma compounds. In contrast to steam distillation or Clevenger hydrodistillation, which operate at 100–120 °C and often induce hydrolysis, oxidation, or polymerization of key volatiles such as linalool and geraniol, the SC-CO₂ process runs efficiently at 40–55 °C and 20–30 MPa, maintaining high extract purity without leaving residual solvents. CO₂ is non-toxic, non-flammable, recyclable, and approved by the U.S. FDA as GRAS (Generally Recognized as Safe), making it ideal for food, pharmaceutical, and cosmetic applications. Moreover, its tunable density and diffusivity under supercritical conditions allow selective solubilization of target compounds, enabling both higher yield and improved aroma fidelity compared with traditional extraction methods. Hence, SC-CO₂ extraction aligns with green chemistry principles while meeting industrial requirements for safety, sustainability, and high sensory quality.

Traditional extraction methods—such as steam distillation, hydrodistillation (Clevenger apparatus), and solvent extraction—are widely used for isolating tea essential oils. However, these processes involve high thermal loads (≥100 °C) or residual organic solvents that cause thermal degradation, hydrolysis of glycosides, and loss of low-boiling volatiles [5,6,7,8,9,10,11]. Consequently, they compromise aroma fidelity, reduce bioactivity, and restrict applications where green, residue-free extracts are required, such as food-grade flavorings and aromatic coatings on wearable fabrics.

Therefore, the present study employs supercritical CO₂ as a clean and controllable extraction medium to ensure the recovery of thermolabile volatiles while minimizing environmental and product safety concerns.

Supercritical carbon dioxide (SC-CO₂) extraction has emerged as a sustainable and efficient technology for recovering high-value essential oils from delicate plant matrices. When CO₂ exceeds its critical point (31.1 °C and 7.38 MPa), it attains liquid-like solvating power and gas-like diffusivity, enabling the selective dissolution of volatile compounds while preserving thermolabile constituents. This method offers several advantages over conventional distillation:

Low-temperature operation (35–55 °C) minimizes oxidation and hydrolysis of sensitive monoterpenes.

No solvent residues, as CO₂ reverts to gas during depressurization, leaving a pure extract.

Tunable selectivity, as solvating strength depends on pressure and temperature.

Shorter extraction times and higher yields of aroma-rich fractions.

Previous studies demonstrated the superiority of SC-CO₂ in tea matrices: Icen and Guru (2009) [12] extracted caffeine and polyphenols at 25 MPa, achieving > 95% purity with reduced degradation. Zhang et al. (2020) [13] reported that SC-CO₂-extracted tea polyphenols retained 92% antioxidant activity relative to fresh leaves. For Phoenix Dancong tea, the same approach promises high aroma integrity and greener processing suitable for cosmetic and textile-grade essential oils.

Extraction efficiency in SC-CO₂ systems depends nonlinearly on pressure, temperature, flow rate, and extraction time. Empirical optimization through single-variable trials is inefficient and may overlook synergistic effects. Response Surface Methodology (RSM), particularly the Box–Behnken Design (BBD), provides a statistically robust framework to model interactive parameters, quantify curvature effects, and locate the global optimum with fewer experiments. RSM has been widely applied to optimize bioactive compound recovery in tea and herbal extracts, achieving R² values above 0.90 in predictive models Jirarattanarangsri & Muangrat [14]. Thus, integrating RSM ensures precision in identifying the most productive and energy-efficient extraction window for Phoenix Dancong essential oil.

The integration of botanical essential oils into fabrics represents a growing frontier in functional textile design. “Aromatic textiles” are engineered to deliver controlled fragrance release, antimicrobial protection, and stress-reducing sensory effects during wear or storage. Among various finishing techniques—microencapsulation, sol–gel embedding, and spraying—blade-coating offers superior film uniformity, mechanical durability, and scalability. In this study, a waterborne polyurethane (WPU) matrix and polydimethylsiloxane (PDMS) additive were used to embed Phoenix Dancong tea oil onto cotton, nylon, polyester, and wool fabrics, enabling comparative analysis of aroma retention over eight weeks [12,15,16,17,18,19].

Aromatic textiles represent a class of functional fabrics engineered to release pleasant natural fragrances or bioactive volatile compounds during wear, storage, or contact with air. They combine textile science, surface chemistry, and sensory technology to enhance user comfort, mood, and antimicrobial protection. Among several finishing techniques—such as microencapsulation, padding, or sol–gel embedding—blade-coating offers high uniformity, controllable film thickness, and industrial scalability. In this process, an emulsion containing the essential oil and a polymeric binder is spread over the fabric at a fixed blade gap, forming a thin, adherent layer after drying [20,21,22,23,24,25]. This technique ensures reproducible fragrance loading and gradual release, making it ideal for evaluating aroma retention on different fibers such as cotton, nylon, polyester, and wool.

This study aims to:

(1)

Optimize the supercritical CO₂ extraction parameters (pressure, temperature, flow rate, and time) for Phoenix Dancong tea essential oil using single-factor experiments followed by RSM–BBD modeling.

(2)

Characterize the extracted oil composition and purity through GC–MS analysis, verifying the preservation of key monoterpenes and esters.

(3)

Evaluate the fragrance retention and sensory persistence of coated fabrics through standardized olfactory scoring (GB/T 14454) and correlate findings with fiber structure.

(4)

Assess the industrial scalability, cost–energy trade-offs, and sustainability prospects of the optimized SC-CO₂ process for green aromatic textile manufacturing.

In this study, we conducted a systematic optimization of the extraction parameters (pressure, temperature, CO₂ flow rate, and time) using single-factor and RSM methods. The extracted oil was then blade-coated to various textile fibers (cotton, nylon, polyester, wool) to assess its fragrance-retention performance. Its extraction process and textile applications can be optimized with the help of this study, which offers both a scientific foundation and a practical reference.

2. Materials and Methods

2.1. Materials and Equipments

The germplasm resources of the tea plants were characterized in accordance with the Descriptors and Data Standard for Tea Germplasm Resources (NY/T 2943-2016; Ministry of Agriculture of the People’s Republic of China: Beijing, China, 2016).

Carbon dioxide (CO₂, purity ≥ 99.9%) for supercritical extraction was supplied by Shanghai Silong Gas Supply Co., Ltd. (Shanghai, China). The extraction was carried out using a supercritical fluid extraction system (Model: SFE220-50-06), manufactured by Nantong Ruizhi Supercritical Technology Development Co., Ltd. (Nantong, China).

An analytical balance (Model: XS205, Mettler Toledo, precision: 0.01 mg) was purchased from Changzhou Wantai Balance Instrument Co., Ltd. (Changzhou, China). A high-shear homogenizer (Model: FSH-2A, FLUKO) was provided by Shanghai Fluko Technology Development Co., Ltd. (Shanghai, China).

Four types of textile fabrics were selected for essential oil application, as follows:

Cotton fabric: 100% cotton, plain weave, model SHMM, weight 350 g/m², obtained from Shaoxing Yingxin Textile Co., Ltd. (Shaoxing, China);

Polyester fabric: 100% polyester, model PKFMTX, weight 380 g/m², obtained from Shaoxing Yingxin Textile Co., Ltd. (Shaoxing, China);

Wool fabric: 100% wool, weight 430 g/m², provided by Nanxun Zihao Textile Co., Ltd. (Huzhou, China);

Nylon fabric: 100% polyamide, model F10040D, weight 85 g/m², supplied by Shaoxing Lvtian Textile Co., Ltd. (Shaoxing, China).

Waterborne polyurethane (WPU), used as the primary film-forming agent for essential oil coating, was supplied by Dongguan Kesixin Adhesive Co., Ltd. (Dongguan, Guangdong, China). To enhance aroma retention and prolong fragrance release, polydimethylsiloxane (PDMS) was incorporated, which was bought from Hubei Candis Chemical Co., Ltd. (Wuhan, Hubei, China).

2.2. Extraction Processes of Phoenix Dancong Tea Essential Oil

The dried Dancong tea leaves were crushed and filtered using the 40–60 mesh stainless steels screens, which are equivalent to the size of particles with estimated diameter of 250–420 µm. Mesh number refers to the number of holes per linear inch, and the sieving was performed to ensure that only particles within this area were extracted.

A total of 100 g of pulverized Phoenix Dancong tea leaves was weighed and loaded into a 1 L stainless-steel extraction vessel. Carbon dioxide (CO₂), sourced from a high-pressure gas cylinder, was first passed through a purification unit and subsequently cooled to −5 °C in a liquefaction chamber.

Once the extraction vessel and the separation vessels reached their respective set temperatures, their heating systems were activated. Upon thermal stabilization, the high-pressure plunger pump was switched on to deliver CO₂ into the system. A fixed number for the CO₂ flow rate was used to initiate the extraction process.

The supercritical fluid, containing dissolved extractables, was depressurized through a throttle valve before entering the separation vessels, where the CO₂ returned to its gaseous state. The extracted tea essential oil precipitated out in the separation chambers and was collected for subsequent analysis:

$$C={\displaystyle \frac{m}{M}}$$

where $C$ is tea essential oil extraction rate (%); $m$ is tea essential oil mass (g); $M$ is tea mass (g).

As summarized in

Table 1. Comparison between SC-CO₂ and Clevenger Hydrodistillation Methods for Tea Essential Oil Extraction.

Parameter	Supercritical CO2 Extraction (SC-CO2)	Clevenger Hydrodistillation	Remarks/Implications
Operating Temperature	35–55 °C (sub-to-moderate heat)	100–120 °C (boiling point of water)	SC-CO2 preserves thermolabile compounds; Clevenger causes thermal degradation.
Operating Pressure	20–30 MPa (high pressure)	Atmospheric pressure	SC-CO2 requires specialized pressure vessels but enhances solvation and selectivity.
Solvent Used	CO2 (green, GRAS, recyclable)	Water or hydro-steam	SC-CO2 leaves no solvent residues; Clevenger co-extracts water-soluble volatiles.
Extraction Time	2–4 h	5–6 h	Faster mass transfer in SC-CO2 due to gas-like diffusivity.
Yield (typical for teas)	1.0–1.2% (w/w)	0.6–0.8% (w/w)	~30–50% higher yield with SC-CO2 under optimized conditions.
Aroma Fidelity	High; preserves top-note monoterpenes (linalool, geraniol)	Moderate; partial oxidation/hydrolysis	SC-CO2 retains > 90% of monoterpenes compared with ~60% in hydrodistillation (Kim et al., 2008 [22]).
Energy Consumption	Moderate; electrical for compressor/pump	High; continuous heating	SC-CO2 can be energy-recovered via CO2 recirculation loops.
Environmental Impact	Green, closed-loop CO2 system	Generates wastewater and requires disposal	SC-CO2 adheres to green-chemistry principles.
Equipment Cost	High initial CAPEX; long-term solvent savings	Low CAPEX; higher running cost due to energy	SC-CO2 favored for high-value aroma extracts.
Scalability & Industrial Use	Widely used for coffee, hops, essential oils	Limited to lab-scale aroma recovery	SC-CO2 easily scaled with automated control.

, the supercritical CO₂ system provides a low-temperature, residue-free, and selective extraction environment, whereas Clevenger hydrodistillation relies on prolonged high-temperature exposure that can alter or degrade volatile components. Therefore, SC-CO₂ was chosen as the primary extraction method for this study.

2.3. Single-Factor Experimental Test

To evaluate the individual effects of key process variables on the extraction yield of Phoenix Dancong tea essential oil, single-factor experiments were performed using analysis of variance (ANOVA) method. Based on preliminary experiments and relevant literature [14,26], four primary parameters: extraction pressure, extraction temperature, CO₂ flow rate, and extraction time were selected. Each parameter was investigated at five different levels, while the remaining conditions were held constant. The detailed tested ranges were as follows:

• Extraction pressure: 10, 15, 20, 25, and 30 MPa;
• Extraction temperature: 35, 40, 45, 50, and 55 °C;
• CO₂ flow rate: 5, 6, 7, 8, and 9 L/h.

Table 2. Experimental design for evaluating the effects of supercritical CO₂ extraction parameters on the yield of Phoenix Dancong tea essential oil.

No.	Extraction Pressure/MPa	Extraction Temperature/°C	CO2 Flow/(L/h)	Extraction Time/h
1	10	50	6	3
2	15	50	6	3
3	20	50	6	3
4	25	50	6	3
5	30	50	6	3
6	20	35	6	3
7	20	40	6	3
8	20	45	6	3
9	20	55	6	3
10	20	50	5	3
11	20	50	7	3
12	20	50	8	3
13	20	50	9	3
14	20	50	6	2
15	20	50	6	4
16	20	50	6	5
17	20	50	6	6

provides a detailed summary of the experimental settings, including the times of extraction (2, 3, 4, 5, and 6 h).

2.4. Response Surface Optimization Design

The SC-CO₂ extraction technique was improved and the link between essential elements was better understood by using RSM after the single-factor tests. The RSM analysis took into account three independent variables: extraction pressure, extraction temperature, and CO₂ flow rate. These three factors had a substantial impact on essential oil production. The software Design-Expert 13.0 was also used to apply a three-level Box–Behnken design (BBD) with essential oil yield as the response variable. Finding the best conditions for extracting and ranking the variables and their interactions required a study with a second-order polynomial model. Each experiment was performed three times, and the findings are displayed as an average. In

Table 3. Factor Levels for RSM.

Level	A: Extraction Pressure (MPa)	Factor 2 B: Extraction Temp (°C)	Factor 3 C: Carbon Dioxide Flow (L/h)
−1	20	45	7
0	25	50	8
1	30	55	9

, you can see all of the experiments’ parameters [13,27,28,29].

All experimental runs, including single-factor and RSM trials, were performed in triplicate (n = 3) to ensure reproducibility. Data are expressed as mean ± standard deviation (SD). The SD values were calculated from replicate measurements, and error bars in all figures represent one SD from the mean. Statistical significance of model terms was evaluated using ANOVA (α = 0.05) as implemented in Design-Expert 13.0 software.

Table 3 shows the three extraction factors—pressure, temperature, and CO₂ flow rate—used in the RSM design, each tested at three levels (low, medium, high). The central point (25 MPa, 50 °C, 8 L/h) represents the optimal midpoint for analyzing their combined effects on extraction yield.

2.5. Preparation of Phoenix Dancong Tea Essential Oil-Coated Fabrics and Subjective Odor Retention Evaluation

To investigate the fragrance-retention performance of Phoenix Dancong tea essential oil on textile substrates and explore its potential applications in functional textiles, a standardized blade-coating technique was employed. The essential oil was incorporated into various textile fabrics, and its adsorption behavior and aroma release properties were systematically evaluated.

The essential oil used in this study was obtained under optimized SC-CO₂ extraction conditions. Based on our previous work [30], a coating formulation was prepared using WPU as the film-forming agent and PDMS as an auxiliary additive. Textile substrates were cut into A4-sized specimens (210 mm × 297 mm) for coating.

Specifically, a coating paste with a 30% (w/w) Phoenix Dancong Tea essential oil concentration was prepared by mixing WPU (30 g), PDMS (20 g), Phoenix Dancong Tea essential oil (30 g), and distilled water (20 g). PDMS and Phoenix Dancong Tea essential oil were premixed to form a uniform oil phase, while WPU and distilled water were blended as the aqueous phase. Under continuous stirring, the oil phase was slowly added dropwise into the aqueous phase and emulsified using a high-shear homogenizer at 10,000 rpm for 5 min, yielding a stable, milky-white emulsion with moderate viscosity and no phase separation [31,32]. The resulting paste was then uniformly coated onto the fabric surface using a blade to achieve a wet film thickness of 0.10 mm. The coated samples were dried in a forced-air oven at 40 °C for 30 min to complete thermal curing and were subsequently sealed for storage.

To assess the aroma release performance, a subjective olfactory evaluation method based on human perception was adopted. In accordance with GB/T 14454 “Methods for Odor Evaluation of Fragrances,” a sensory panel of 10 certified senior tea tasters (5 males and 5 females, average age 46 years) was assembled to conduct the evaluation. A five-point scale was used, with Level 1 indicating extremely faint aroma and Level 5 indicating intense and overwhelming fragrance.

Prior to testing, the coated fabric samples were placed in a cool, ventilated environment for 12 h to reach volatilization equilibrium. Evaluations were performed independently in a controlled sensory room free of external disturbances. Each panelist rated the aroma intensity and provided descriptive feedback. The final aroma grade was determined as the average score of all 10 panelists, ensuring objectivity and reproducibility of the results.

Table 4. Formulation and Processing Parameters for Blade-Coated Fabrics.

Component/Parameter	Description/Value	Remarks
Essential oil concentration	30% (w/w of total emulsion)	Optimized for strong yet stable fragrance release
Binder	Waterborne polyurethane (WPU, 30 g)	Primary film-forming agent
Additive	Polydimethylsiloxane (PDMS, 20 g)	Enhances hydrophobicity and slow release
Solvent	Distilled water (20 g)	Adjusts viscosity
Mixing method	High-shear homogenization at 10,000 rpm for 5 min	Produces uniform emulsion
Fabric types	Cotton, polyester, wool, nylon (A4 size; 210 × 297 mm)	Equal coating area
Blade gap (wet film thickness)	0.10 mm	Controls coating thickness
Drying conditions	40 °C for 30 min in forced-air oven	Ensures full film curing
Storage	Sealed, 25 °C, 50% RH before testing	Prevents premature volatilization

outlines the blade-coating formulation, showing a 30% tea oil emulsion with WPU and PDMS applied on fabrics using a 0.10 mm gap and dried at 40 °C for 30 min to ensure uniform coating and lasting fragrance.

Tea leaves are dried, ground, and extracted by SC-CO₂ to obtain pure essential oil, which is then emulsified and blade-coated onto fabric. The coated fabric is dried at 40 °C to produce a stable aromatic textile as shown in Figure 1.

2.6. GC–MS Profiling (Objective Verification)

To objectively verify that the optimized supercritical CO₂ extraction preserved the structural integrity of volatile compounds, the essential oil obtained under optimal conditions (25 MPa, 50 °C, 8 L h⁻¹, 3 h) was analyzed by gas chromatography–mass spectrometry (GC–MS).

• Instrument and operating parameters:

Analyses were performed using an Agilent 7890B GC system coupled with a 5977B mass selective detector (Agilent Technologies, USA) and an HP-5MS column (30 m × 0.25 mm × 0.25 µm). The carrier gas, helium, was constantly supplied at a flow rate of 1.0 mL min⁻¹. I programmed the oven to start at 50 °C and keep it there for 2 min. After that, I was to bring it up to 250 °C and keep it there for 5 min, increasing the temperature by 4 degrees Celsius per minute. The injector and detector were kept at a ten-to-one ratio at 250 °C. Mass spectra were obtained in EI mode at 70 eV over m/z 35–400.

• Identification of compounds:

Peaks were identified by matching the mass spectra with the NIST 2020 Library and verified by comparing retention indices (RI) with literature data. Relative composition (%) was calculated by peak area normalization without correction factors. All analyses were performed in triplicate, and data are presented as mean ± SD.

In addition to sensory evaluation, gas chromatography–mass spectrometry (GC–MS) was employed to objectively verify the chemical integrity and compositional stability of volatile compounds extracted under optimized SC-CO₂ conditions.

3. Results and Discussion

3.1. Single-Factor Experimental

3.1.1. Effect of Extraction Pressure

As shown in Figure 2, under extraction conditions of 50 °C, a CO₂ flow rate of 6 L/h, and an extraction time of 3 h, the yield of tea essential oil increased rapidly with rising pressure, then gradually plateaued. This trend indicates that at lower pressures, increasing pressure significantly enhances the density of CO₂, thereby improving its solvent power and increasing the solubility of extractable compounds in the supercritical fluid. Moreover, higher pressure promotes more efficient contact between tea leaves and CO₂, accelerates mass transfer, and facilitates the release of volatile components, ultimately contributing to higher oil yields.

Extraction yield increased rapidly during the first 3–4 h, after which it approached a plateau. Beyond 6 h, only marginal increases (<2%) were observed, accompanied by higher energy consumption and reduced cost-efficiency. Therefore, 6 h was selected as the practical upper limit to ensure optimal balance between recovery and processing time. Longer extractions may be explored for denser matrices, but within this study’s scope, extending duration yielded no significant benefit.

Additionally, elevated pressure helps overcome physical barriers within plant cell structures, promoting the desorption and diffusion of active constituents. However, when the pressure exceeds a certain threshold, the increase in CO₂ density slows, meaning that further pressure increments provide diminishing returns in terms of solvent capacity. Excessive pressure also significantly increases equipment energy consumption and operating costs, while imposing greater demands on equipment safety and structural integrity. Furthermore, overly high pressures may lead to the overextraction or thermal degradation of heat-sensitive or volatile compounds, adversely affecting essential oil quality.

Therefore, the optimal extraction pressure for SC-CO₂ extraction of tea essential oil in this study was determined as 25 MPa.

3.1.2. Effect of Extraction Temperature

As shown in Figure 3, under the fixed conditions of 20 MPa extraction pressure, 6 L/h CO₂ flow rate, and 3 h extraction time, the yield of tea essential oil initially increased and then decreased with rising extraction temperature. This indicates that temperature is determined as a critical parameter in the supercritical CO₂ extraction process, exerting a dual effect on extraction yield.

At lower temperatures (<45 °C), increasing temperature enhances the diffusion coefficient and penetration ability of CO₂, thereby accelerating mass transfer of extractable compounds into the solvent. Moreover, elevated temperatures facilitate the desorption and release of aromatic volatile components, resulting in increased essential oil yield.

However, when the temperature exceeds a certain threshold (approximately 45 °C), the density of CO₂ decreases significantly with further temperature rise, indicating a reduction in its solvent power. Consequently, the solubility of the extractables and mass transfer efficiency decline, leading to a decrease in oil yield. In addition, excessively high temperatures may cause thermal degradation or excessive volatilization of aromatic constituents, further compromising both the quantity and quality of the extracted oil.

However, the positive effects of temperature on mass transfer and desorption efficiency predominate when the temperature is below 45 °C. In contrast, above 45 °C, the negative impact of reduced CO₂ density becomes more significant, leading to a noticeable drop in yield. Thus, the optimal extraction temperature for SC-CO₂ extraction of Phoenix Dancong tea essential oil in this study was determined to be 45 °C.

3.1.3. Effect of CO₂ Flow Rate

Considering the extraction temperature. Maintaining the following conditions: an extraction pressure of 20 MPa, an extraction temperature of 50 °C, an extraction period of 3 h, and a CO₂ flow rate of 5–9%, as shown in Figure 3, would provide the following amounts of tea essential oil. As the CO₂ flow rate was increased, a plateau was reached in the creation of tea essential oil. The reason behind this is that there are other factors besides CO₂ flow rate that influence the process of supercritical CO₂ extraction. On one hand, increasing the CO₂ flow rate enhances the mixing and contact between the extractable compounds and supercritical CO₂ through an entrainment effect, thereby improving the solvent power and accelerating the extraction process, which ultimately increases the yield. On the other hand, excessive flow rates shorten the contact time between the supercritical fluid and the plant matrix, resulting in insufficient mass transfer, which limits the extraction efficiency and may even reduce the yield [33].

As shown in Figure 4, the highest yield was obtained at a CO₂ flow rate of 8 L/h. This indicates that at this flow rate, a balance was achieved between the positive effect of improved solute–solvent interaction and the negative effect of reduced contact time, leading to an optimized extraction outcome. Overall, although increasing CO₂ flow rate improves yield to a certain extent, the marginal benefit diminishes beyond a threshold, resulting in a plateauing trend in extraction efficiency.

3.1.4. Effect of Extraction Time

As shown in Figure 5, under the fixed conditions of 20 MPa extraction pressure, 50 °C extraction temperature, and 6 L/h CO₂ flow rate, the yield of tea essential oil increased significantly during the initial 1–3 h of extraction. However, after 3 h, the rate of increase slowed markedly, and little to no further yield improvement was observed beyond 4 h. This suggests that the influence of extraction time on yield is time-dependent, with significant effects only observed within a specific time window.

During the initial stage of extraction, prolonged contact between the CO₂ and the tea matrix enhances solubilization and mass transfer, leading to increased yield. In the later stage, once sufficient interaction between the solute and solvent is achieved, the yield approaches its maximum, and further extension of the extraction time has negligible impact. Moreover, unnecessarily long extraction durations lead to increased energy consumption and operational costs without proportional benefits.

Based on the data in Figure 5, a 3 h extraction period is considered optimal for supercritical CO₂ removal of tea essential oil, as it maximizes yield while maintaining cost-efficiency.

The comparison in

Table 5. Comparative performance of different extraction methods for tea essential oils.

Extraction Method	Operating Temperature (°C)	Pressure (MPa)	Solvent Type	Typical Yield (% w/w)	Aroma Fidelity (Qualitative)	Processing Time (h)	Residue/Safety	Reference
SC-CO2 (this study)	50	25	CO2 (>99.9%)	1.12	High (>90% monoterpene retention)	3	Residue-free, GRAS solvent	[12,13,20,21,22]
Steam distillation	100–120	1 atm	Water	0.6–0.8	Moderate (loss of thermolabile notes)	5–6	None (but oxidation risk)	[16,17]
Solvent extraction (hexane, ethanol)	60–80	1 atm	Organic solvent	0.9–1.0	Good (occasionally residual odor)	4–5	Solvent traces possible	[18,19]
Microwave-assisted extraction (MAE)	70–100	1 atm	Water/ethanol	1.0–1.3	Good but risk of localized heating	1–2	Requires special equipment	[23,24]
Enzyme-assisted extraction (EAE)	45–55	1 atm	Buffer + enzyme	1.1	High purity, moderate cost	8–10	Eco-friendly, slow	[25,26]

, shows that SC-CO₂ achieves higher yield and superior aroma preservation compared with steam or solvent extraction, while maintaining a shorter processing time and no solvent residues. Although MAE and EAE can enhance efficiency, they often require higher equipment cost or longer enzymatic reactions. Overall, the SC-CO₂ approach offers the best balance between productivity, quality, and environmental performance, supporting its adoption for high-value aromatic textile applications

3.2. GC–MS Profiling and Structural Integrity of Volatiles

The GC–MS chromatogram revealed that the optimized SC-CO₂ extract retained the characteristic monoterpene profile of Phoenix Dancong tea essential oil. Thirty-two major volatile constituents were detected, representing >96% of the total ion current, dominated by linalool (21.4%), geraniol (14.7%), nerol (8.9%), linalyl acetate (6.8%), geranyl acetate (5.2%), and trace sesquiterpenes such as β-caryophyllene (3.6%) and farnesol (2.8%). The composition pattern closely corresponds to previous reports for Dancong oolong essential oil [10,11,15], confirming that supercritical CO₂ extraction maintained the structural stability of key aroma compounds.

Table 6. Representative volatile components of Phoenix Dancong tea essential oil extracted by SC-CO₂ (optimized conditions).

Compound	Chemical Class	RI (Calc.)	Relative Area (%) ± SD	Aroma Descriptor
Linalool	Monoterpene alcohol	1096	21.4 ± 0.6	Floral, citrus
Geraniol	Monoterpene alcohol	1253	14.7 ± 0.4	Rose-like
Nerol	Monoterpene alcohol	1232	8.9 ± 0.5	Fresh floral
Linalyl acetate	Monoterpene ester	1268	6.8 ± 0.3	Sweet, fruity
Geranyl acetate	Monoterpene ester	1384	5.2 ± 0.2	Citrus, waxy
β-Caryophyllene	Sesquiterpene	1421	3.6 ± 0.2	Woody, spicy
Farnesol	Sesquiterpene alcohol	1703	2.8 ± 0.1	Balsamic
Others (25 minor constituents)	—	—	32.6 ± 1.4	—

lists the main volatile compounds identified in Phoenix Dancong tea essential oil extracted under optimized SC-CO₂ conditions. Linalool and geraniol were the dominant components, giving floral and citrus notes, while esters and sesquiterpenes contributed to sweet and woody aromas, confirming high aroma integrity and chemical stability.

Total identified fraction: 96.0 ± 1.0%

Compared with hydrodistillation data (Li et al., 2023 [10]), monoterpenes with low boiling points (≤200 °C) were retained at >90% of their native proportion, while oxygenated species (linalool, geraniol) showed minimal oxidation or rearrangement. No evidence of thermal degradation products (e.g., campholenic or α-terpineol oxide) was observed, confirming excellent structural preservation under mild SC-CO₂ conditions.

3.3. Response Surface Optimization Design

3.3.1. Analysis of Variance and Significance Testing

To minimize the number of experimental trials and optimize the extraction time of essential oil, the parameters selected for the RSM were determined based on preliminary single-factor experiments, as summarized in

Table 7. Experimental data for the oil yield obtained from the central composite experimental design.

NO.	Factor A: Extraction Pressure (MPa)	Factor B: Extraction Temperature (°C)	Factor C: Carbon Dioxide Flow (L/h)	Response: Essential Oil Yield (%)
1	25	50	8	1.01
2	20	45	8	0.59
3	20	50	7	0.66
4	25	55	9	0.91
5	25	50	8	1.05
6	30	45	8	0.85
7	20	50	9	0.93
8	25	50	8	1.05
9	25	50	8	1.12
10	25	55	8	0.78
11	20	55	7	0.76
12	25	45	7	0.81
13	30	55	8`	0.84
14	25	45	9	0.79
15	30	50	9	0.98
16	25	50	8	0.99
17	30	50	7	0.92

.

In RSM, the F-value is also used as an indicator of the influence strength of each factor. By jointly considering the F- and p-values, it was found that the variables significantly affected the supercritical CO₂ extraction yield of black tea essential oil. The detailed ANOVA results are presented in

Table 8. Results of the analysis of variance to the response surface quadratic model.

Source	Sum of Squares	df	Mean Square	F-Value	p-Value
Model	0.3165	9	0.0352	16.47	0.0006
A-Extraction Pressure	0.0496	1	0.0496	23.24	0.0019
B-Extraction Temperature	0.0153	1	0.0153	7.17	0.0316
C-Carbon Dioxide Flow Rate	0.0264	1	0.0264	12.39	0.0097
AB	0.01	1	0.01	4.68	0.0672
AC	0.011	1	0.011	5.16	0.0573
BC	0.0012	1	0.0012	0.5738	0.4735
A2	0.0528	1	0.0528	24.74	0.0016
B2	0.1174	1	0.1174	55	0.0001
C2	0.0149	1	0.0149	6.98	0.0333
Residual	0.0149	7	0.0021
Lack of Fit	0.005	3	0.0017	0.6754	0.6109
Pure Error	0.0099	4	0.0025
Cor Total	0.3314	16

p < 0.01 highly significant; 0.01 ≤ p < 0.05 significant; p ≥ 0.05 not significant.

.

All of the parameter ranges investigated were well-fit by the regression model, and the model’s statistical significance (F = 16.47, p = 0.0006) is shown in Table 8. The lack-of-fit test yielded a non-significant result (p = 0.6109), and there was no statistically significant unexplained variation, thus it appears that the model fits the data well.

The coefficient of determination (R²) and adjusted R² were 0.9549 and 0.8969, respectively (Table 3), further confirming the model’s high explanatory power. The order of significance for the main factors is as follows: extraction pressure > CO₂ flow rate > extraction temperature. Specifically, extraction pressure and CO₂ flow rate had highly significant effects on yield (p < 0.01), while extraction temperature had a significant effect (p < 0.05). Moreover, the interaction between extraction pressure and CO₂ flow rate, as well as between extraction pressure and temperature, was significant, whereas the interaction between CO₂ flow rate and temperature was not.

These results were used to build a second-order polynomial regression model, which is represented by Equation (1):

$$Y=1.04+0.0787A+0.0438B+0.0575C-0.0500AB-0.0525AC+0.0175BC-0.1120A2-0.1670B2-0.0595C2$$

(1)

where $Y$ represents the essential oil extraction yield (%); A, B, and C represent the coded values of extraction pressure (MPa), extraction temperature (°C), and CO₂ flow rate (L/h), respectively.

3.3.2. Interaction Effects of Factors Analyzed by RSM

The influence of the independent parameters on the SC-CO₂ extraction of Camellia sinensis (Dancong tea) essential oil was better understood with the help of contour plots and three-dimensional response surfaces. Two variables varied within the experimental range, and one was maintained at its mean value so that the influence of each component could be observed. Keeping the CO₂ flow rate, extraction pressure, and temperature constant, we created surface response and contour plots, respectively.

Keeping the CO₂ flow rate constant at 8 L/h, Figure 6 displays the impact of temperature and pressure on oil yield. The prominent tilt of the contour lines and the spherical form indicate a substantial interaction between temperature and extraction pressure. As predicted by the one-factor experiment, oil production rose continuously with increasing pressure while temperature remained unchanged. Oil yield was higher at lower pressures but dropped with rising temperature. One possible explanation is that at high temperatures, the solvating power of CO₂ decreases with increasing density. Under the same thermal conditions as the solute vapor pressure increases with rising temperature, CO₂ becomes more soluble, and yields increase at higher pressures. It appears that extraction pressure has a greater influence on oil yield than extraction temperature, as indicated by the closer spacing of the contour lines along the pressure axis compared to the temperature axis.

Figure 7 illustrates the effects of CO₂ flow rate and extraction pressure on oil yield at a constant extraction temperature of 50 °C. The elliptical contour shape again indicates a significant interaction between pressure and CO₂ flow rate. The variation in contour density is more pronounced along the pressure axis, confirming that pressure plays a more dominant role than CO₂ flow rate in influencing extraction yield.

Figure 8 shows the combined effect of extraction temperature and CO₂ flow rate on the extraction rate. The response surface indicates an optimal region near 50 °C and 8 L/h where the extraction yield peaks. Elliptical contour lines confirm a moderate interaction between the two factors, supporting the RSM model prediction.

The multi-factor SC-CO₂ extraction procedure for essential oil of Dancong tea was fine-tuned with the use of RSM and multiple regression analysis. Temperature was set to 50 °C, pressure to 25 MPa, CO₂ flow rate to 8 L/h, and duration to 3 h for the final extraction. Two variables stood up as being highly significant in the regression analysis: extraction pressure and CO₂ flow rate. Increasing the extraction time mostly raised cost rather than yield, and the correlation between temperature and CO₂ flow rate had no effect on yield. This study’s objective is to identify the optimal balance between oil yield, process efficiency, and production costs.

3.3.3. Sensory Evaluation of Aroma Retention on Various Fabrics Coated with Dancong Tea Essential Oil

To evaluate the aroma retention performance of Phoenix Dancong tea essential oil on different textile substrates, four commonly used fabrics—cotton, nylon, polyester, and wool—were selected. A five-point subjective olfactory scoring method was employed to assess the aroma intensity of coated samples at different time intervals. The detailed results are listed below (

Table 9. Results of the aroma retention performance of Phoenix Dancong tea essential oil on different textile substrates.

Fabric Placement Time/Week	Cotton	Nylon	Polyester	Wool
0	+++++	+++++	+++++	+++++
1	+++++	++++	+++++	+++++
2	+++++	+++	+++++	++++
4	+++++	+++	++++	+++++
6	+++++	+++	++++	+++
8	++++	++	++++	+++

Note: ++ indicates a relatively mild fragrance. +++ indicates a pleasant fragrance. ++++ indicates a more fragrant smell; +++++ indicates overly fragrant.

). The aroma intensity was categorized as: very faint, faint, moderate, strong, and overpowering.

As shown in Table 7, all fabrics were rated as overpowering at the initial stage (immediately after application), indicating a strong initial aroma release across all materials. However, as time progressed, variations in aroma retention became apparent among the different fabric types.

After one week, both cotton and polyester maintained an overpowering aroma intensity, while nylon dropped slightly to strong, and wool showed a slight decline but still retained a high level of aroma. This suggests varying adsorption and retention capacities among the fabrics. By the second week, cotton continued to retain an overpowering aroma, while the others showed further decline, with wool dropping to moderate, indicating a relatively weaker adsorption capacity. After four weeks, cotton showed a slight decline to strong, whereas nylon and polyester exhibited noticeable aroma loss, likely due to the low polarity and porosity of synthetic fibers. By the sixth week, cotton still maintained a strong aroma, while other materials further weakened. At eight weeks, cotton retained a perceptible aroma (approaching moderate), while polyester dropped to faint, and nylon and wool lost nearly all perceptible scent.

These observations can be attributed to the structural and chemical properties of the fabric substrates. Cotton, as a natural fiber, possesses high porosity and a complex surface morphology that enhances essential oil adsorption and sustained release [33]. In contrast, synthetic fibers such as nylon and polyester have smoother surfaces and lower porosity, making it more difficult for oil molecules to adhere and reducing retention time [34]. Although wool is also a natural fiber, its adsorption capacity is lower than that of cotton, leading to faster aroma dissipation [35,36].

Furthermore, the high hydrophilicity and surface polarity of cotton facilitate the formation of hydrogen bonds between its hydroxyl groups and the polar components of tea essential oil (e.g., alcohols and aldehydes), thereby delaying fragrance evaporation. On the other hand, the low surface energy and hydrophobicity of synthetic fibers limit their interaction with essential oil molecules. The aromatic components in the oil form hydrogen bonds and van der Waals interactions with cotton’s hydroxyl-rich surface, enhancing stability and fixation. Conversely, the smooth and chemically inert surface of synthetic fibers lacks such interactions, allowing essential oil molecules to desorb and evaporate more easily. In addition, the higher surface energy and wettability of cotton enable better penetration and anchoring of the oil, while nylon and polyester retain the oil mostly on the surface, leading to rapid volatilization.

The high retention of oxygenated monoterpenes, which are key contributors to antioxidant and antimicrobial efficacy, suggests that the extract maintains its bioactive potential. Previous studies have linked linalool and geraniol to >90% inhibition of Staphylococcus aureus and strong free-radical scavenging activity (IC₅₀ ≈ 0.35 mg mL⁻¹) [5,6,7]. Although this study did not include direct bioassays, the compositional integrity strongly indicates that functional bioactivity is preserved.

No in vitro or in vivo bioassays were performed in the present work. Future studies will include DPPH antioxidant testing, agar-diffusion antimicrobial assays, and cell-viability assessments to quantitatively correlate chemical composition with biological performance and substantiate the functional claims of SC-CO₂-derived Phoenix Dancong tea oil.

To address natural variability in tea harvests, future extraction setups should adopt an adjustable SC-CO₂ platform equipped with inline moisture monitoring, automatic pressure–temperature control, and feedback sensors. Such a flexible system would allow operators to fine-tune pressure (20–30 MPa) and temperature (40–55 °C) windows based on real-time leaf conditions—such as moisture content, maturity, or cell-wall rigidity—rather than applying fixed parameters. This adaptive control can ensure consistent yield and aroma quality across seasonal and raw-material variations.

Scale-Up and Sustainability Considerations

From a process-engineering standpoint, SC-CO₂ extraction at 25 MPa and 50 °C involves moderate energy input (≈1.2–1.5 kWh kg⁻¹ of CO₂ circulated) and requires a high-pressure vessel rated to 40 MPa. The principal cost factors are compressor power and CO₂ recycling systems, which constitute ~60% of total CAPEX/OPEX for small-scale units. However, these can be mitigated by heat-integration and closed-loop CO₂ recovery, reducing energy demand by up to 25%. Life-cycle assessments (LCA) of comparable botanical SC-CO₂ facilities show a 30–40% lower carbon footprint than solvent extraction due to the absence of hazardous waste. For industrial translation, scale-up should focus on modular extractors (10–50 L capacity) with automated control of pressure and temperature, ensuring consistent yield and quality while maintaining sustainability compliance. Future work should include a techno-economic analysis (TEA) and LCA benchmarking to quantify long-term environmental and economic benefits.

In conclusion, cotton demonstrated superior aroma retention performance for Phoenix Dancong tea essential oil, due to its porous structure, high surface polarity, and hydrophilic nature. These properties make it an ideal substrate for essential oil delivery. In contrast, synthetic fibers such as nylon and polyester showed poor fragrance retention, mainly due to unfavorable surface characteristics. These findings provide a theoretical foundation for the development of long-lasting aromatic textiles, positioning cotton as the preferred material for such applications.

4. Conclusions

Optimized extraction conditions in this paper are only applicable to dried, spring-harvested Phoenix Dancong tea leaves ground and sieved through 40–60 mesh stainless steel screens (which represents a particle of about (250–420 µm) having a low moisture level, less than 8%. Results may differ for fresh or older leaves due to changes in moisture, cell-wall structure, and volatile composition. Future studies should validate and adjust the model for different harvest seasons and leaf maturities to broaden its practical applicability.

This study optimized supercritical CO₂ extraction of Phoenix Dancong tea essential oil using response surface methodology, identifying 25 MPa, 50 °C, 8 L h⁻¹, and 3 h as optimal conditions, yielding 1.12% (w/w) oil. GC–MS confirmed preservation of key aroma compounds—linalool, geraniol, and their acetates—demonstrating excellent structural integrity and compositional stability. Among tested fabrics, cotton exhibited the highest fragrance retention, attributed to hydrogen bonding between cellulose hydroxyl groups and polar volatiles. The process offers green, residue-free extraction suitable for aromatic textile applications.

Merits: environmentally safe process, strong aroma fidelity, and scalable potential.

Limitations: evaluation restricted to dried, spring-harvested leaves and sensory-based retention tests.

Future work: extend optimization to varied leaf states, integrate inline moisture control for adaptive SC-CO₂ systems, and conduct bioassays and life-cycle assessments to validate industrial sustainability.