Ultrasound-assisted temperature-responsive deep eutectic solvents for simultaneous extraction and separation of the alkaloid components from the white pepper

Graphical abstract

Abstract

In this study, eighteen LCST-type temperature-responsive deep eutectic solvents (TRDESs) were developed for the ultrasound-assisted extraction of alkaloids from white pepper, using piperine (PIP) as the indicator. After optimization, the best conditions were identified as lidocaine/valeric acid (1:1) as the solvent, a solid‑to‑liquid ratio of 1:40  mg/mL, 75°C, and 30 min of ultrasound, achieving a PIP yield of 33.7502 mg/g. By utilizing the temperature‑responsive behavior of TRDES, PIP was separated solely by adjusting temperature and water content, without additional reagents, resulting in a crude extract containing 122.3  μg/mg of PIP. Subsequent analysis confirmed that the TRDES‑based ultrasound‑assisted extraction did not affect the antioxidant or anti‑diabetic (α‑glucosidase and α‑amylase inhibitory) activities of the extract. The TRDES system also showed good reusability, retaining over 85% of its initial extraction efficiency after five cycles. Density functional theory calculations revealed that ultrasound‑assisted extraction proceeds via hydrogen bonding between lidocaine/valeric acid and the five‑membered ring of PIP. With a high GAPI score of 82, this method offers a clearly greener and more sustainable alternative to conventional extraction techniques. This work integrates the novel green solvent TRDES with ultrasound‑assisted extraction, laying a key foundation for developing environmentally friendly processes in natural product preparation.

1. Introduction

White pepper (Piper nigrum L.) is a woody climbing vine of the Piperaceae family. Its fruit, known for its pungent aroma, is widely used as a seasoning to enhance the flavor of various dishes and is valued as a versatile ingredient formedicinal purposes. White pepper possesses a diverse phytochemical composition, encompassing alkaloids, volatile oils, phenolics, and organic acids [1], [2]. The bioactivity is primarily attributed to the alkaloid fraction, where the amide alkaloid piperine (PIP) stands out as the chief contributor and most versatile active compound [3], [4]. Modern pharmacological studies have demonstrated that the amide alkaloids in white pepper exhibit a range of biological activities, including anti-inflammatory, antioxidant, anti-obesity, and hypoglycemic effects, as well as the ability to modulate and protect the central nervous system [5], [6], [7], [8]. Thus, the efficient extraction of alkaloids, particularly PIP, is essential for unlocking the full potential of pepper in product development.

Currently, organic solvents remain the primary method for extracting PIP [4]. Although they offer high extraction yields, their use is associated with inherent drawbacks [9], [10]. The volatilization of organic solvents during extraction not only causes air pollution but also poses irreversible health risks to humans [11], [12]. Moreover, the disposal of post-extraction waste residues and liquids imposes an economic burden as well as a potential risk of soil contamination [13]. Solvent residue is a major concern, resulting not only in potential environmental pollution but also in health-related problems [14]. Given these concerns, the use of green solvents as an extraction solution has become a major focus. Deep Eutectic Solvents (DESs) offer a promising solution as a novel, eco-friendly alternative [15], [16], [17]. These solvents, formed by the complexation of a hydrogen bond donor (HBD) and hydrogen bond acceptor (HBA), are attractive due to their low toxicity, biodegradable nature, and straightforward synthesis [18], [19]. Due to these beneficial properties, DESs are now applied extensively for the ultrasound‑assisted extraction of key bioactive constituents from natural sources, exemplified by flavonoids, alkaloids, and polyphenols [20], [21], [22], [23], [24]. The application of conventional evaporation for DES removal is limited due to its incompatibility with their non-volatile character [25]. Subsequent processing of DES extracts typically relies on methods like macroporous resin adsorption to separate the target compounds [26]. However, this process is operationally complex and requires substantial amounts of organic solvents for the adsorption and desorption steps, which can, in turn, cause environmental and health concerns. The development of responsive DESs elegantly resolves this challenge [27].

A key feature of responsive DESs is their ability to change hydrophilicity/ hydrophobicity under external triggers like pH, temperature, or CO₂/N₂, enabling straightforward separation of the solvent from the target solutes and streamlining the overall extraction process [28], [29], [30]. Notably, temperature-responsive DESs (TRDESs) achieve this separation purely through thermal modulation, eliminating the need for external additives and thereby offering a more straightforward operational approach. TRDESs are categorized into lower critical solution temperature (LCST) and upper critical solution temperature (UCST) types [31]. For the LCST-type, the system is hydrophilic below its critical temperature and undergoes a transition to a hydrophobic state when the temperature rises above this point [32]. The UCST-type exhibits the opposite behaviour [33]. Benefiting from their responsive properties, TRDESs have been successfully applied to the ultrasound‑assisted extraction and separation of bioactive plant components [33], [34], [35], [36], [37]. In the experiment by Cai et al., a UCST-type TRDES composed of ethanolamine and o-cresol was utilized to extract polysaccharides from Ganoderma lucidum [35]. The separation of the polysaccharides from the TRDES was achieved by adding water at room temperature, which induced biphasic separation. The operation of an LCST-type TRDES system is demonstrated in the work of Shen et al., who extracted β-carotene and lycopene from tomatoes using a TRDES synthesized from MBPCA and a fatty acid [34]. Upon mixing the TRDES with water at room temperature (a temperature below its LCST), a homogeneous phase was established, facilitating the integration of the solvent with the aqueous medium. The subsequent precipitation of the hydrophobic β-carotene and lycopene from this homogeneous mixture effectively achieved their separation, and the TRDES was concurrently regenerated for further use. These results establish a valuable foundation for the employment of green and recyclable solvents in the ultrasound‑assisted extraction of natural product active ingredients, thereby underscoring the considerable promise of the TRDES for the future of natural product separation.

Currently, one study has reported the application of DES composed of choline chloride and 1,2-propanediol for extracting active ingredients like PIP from black pepper, achieving an extraction yield of 39.075 mg/g [38]. However, the subsequent removal of the DES relied on macroporous adsorption resin, which not only reintroduced organic solvents such as ethanol but also made the recovery and reuse of the DES unfeasible. Therefore, this study presents a novel strategy for PIP ultrasound‑assisted extraction and separation from white pepper using an LCST-type TRDESs. 18 types of TRDESs using lidocaine, procaine, and tetracaine as HBAs and fatty acids with different carbon chain lengths as HBDs, and determined their viscosity values at different temperatures. Combined with ultrasound-assisted technology, these TRDESs were applied to the extraction of PIP from white pepper. The extraction conditions, including the type of TRDESs, solid-to-liquid ratio, extraction temperature, and extraction time, were optimized using single-factor experiments and response surface methodology (RSM). Furthermore, the separation and recovery of PIP were achieved by utilizing the temperature-dependent hydrophilic-hydrophobic transition of TRDESs. The precipitated crude extract was subsequently subjected to composition identification and in vitro pharmacological activity tests. The reusability of the TRDES was also investigated. Finally, the greenness of the extraction process was evaluated using the Green Analytical Procedure Index (GAPI), and the underlying mechanism was explored by Density Functional Theory (DFT) calculations. Overall, the entire extraction process operates exclusively through controlled changes in temperature and water content, enabling direct product isolation without any chemical additives. This approach successfully yielded the crude extract, underscoring the method's green and efficient nature.

2. Method

2.1. Materials and reagents

PIP was purchased from Shanghai Yuanye Bio-Technology Co., Ltd. Lidocaine (Lido), Procaine (Pro), and Tetracaine (TET) were procured from Shanghai Rhawn Chemical Technology Co., Ltd. Valeric Acid (VA, C₅) and Heptanoic Acid (Hep A, C₇) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. The information on the above related compounds is listed in Table 1.

Table 1.

Physical-chemical properties and structure of HBAs and PIP.

Compound	CAS number	Molecular weight	Chemical formula	LogP	Structure
Piperine	94–62-2	285.34	C17H19NO3	2.94
Lidocaine	137–58-6	234.34	C14H22N2O	2.44
Procaine	59–46-1	236.31	C13H20N2O2	1.87
Tetracaine	94–24-6	264.36	C15H24N2O2	3.51

Octanoic acid (OA, C₈), Decanoic acid (DA, C₁₀), Lauric acid (LA, C₁₂), and hydrochloric acid were obtained from Sinopharm Chemical Reagent Co., Ltd. Hexanoic Acid (HA, C₆), Nonanoic Acid (NA, C₉) and Myristic Acid (MA, C₁₄) were procured from Shanghai Macklin Biochemical Co., Ltd. Methanol (HPLC-grade) was obtained from Merck (Darmstadt, Germany). ABTS Free Radical Scavenging Capacity Assay Kit, DPPH Free Radical Scavenging Capacity Assay Kit, α-Gucosidase Inhibitor Activity Assay Kit, and α-amylase(α-AL) Activity Assay Kit were procured from Beijing Solarbio Science & Technology Co., Ltd.

White pepper (Production Batch No. A220301), sourced from Hainan Province, China, was ground into powder using a commercial grinder (Zhejiang Yili Industry & Trade Co., Ltd., China) and then passed through a No. 4 sieve before use.

2.2. Instrumentation

Quantitative analysis was performed using a high-performance liquid chromatography (HPLC) system (CDD-10AVP, Waters, USA) equipped with a diode array detector (DAD). Separations were achieved on a ThermoFisher C18 analytical column (250 mm × 4.6 mm, 5 µm particle size; ThermoFisher, China) maintained at 25°C. The mobile phase consisted of ultrapure water and methanol (25:75, v/v), delivered isocratically at a flow rate of 1.0 mL·min⁻¹. Samples were injected at a volume of 5 μL, and analytes were monitored at a detection wavelength of 343 nm.

2.3. Preparation and characterization of TRDESs

According to the literature [39], the HBA and HBD were combined and stirred magnetically at 80°C for 20 min, forming a homogeneous, transparent mixture. The resulting solution remained stable after cooling to room temperature. Viscosity measurements for all synthesized DESs were conducted using a viscometer (Shanghai Fangrui Instrument Co., Ltd., China) at 25°C and 80°C.

Fourier transform infrared spectrometer analysis (Nicolet 6700, USA) was employed to characterize the bonding interactions between the HBA, HBD, and their resulting TRDES. The wavenumber range is set to 400–4000 cm⁻¹, the spectrometer resolution is 4 cm⁻¹, and the signal-to-noise ratio is 50,000:1. All samples were scanned 32 times. Differential scanning calorimetry analysis (Basis unit DSC214 Intracooler, Germany) was employed to determine the melting points of HBA, HBD, and the TRDES synthesized from them. Accurately weighted amounts of samples were placed in aluminum pans and heated at a scanning rate of 10˚C min⁻¹ from −150 to 150˚C, under a nitrogen purge gas flow rate of 25 mL min⁻¹. The HBA, HBD, and the resulting TRDES were characterized using a fully digital 600 MHz NMR spectrometer (AVANCE NEO 600, Bruker, Switzerland). All samples were dissolved in deuterated DMSO for analysis. The measurements were performed on an Ascend SB (54 mm) 14.1 T magnet with a ¹H frequency of 600 MHz and an RF channel of H-100 W.

2.4. Ultrasonic extraction process

Firstly, TRDES is synthesized according to section 2.3, and then the TRDES ultrasound-assisted extraction procedure was performed under the optimized conditions as follows: approximately 0.1 g of white pepper powder was mixed with 4 mL of TRDES and subjected to ultrasound-assisted extraction (250 W, 20 kHz) at 75°C for 30 min. For the ethanol extraction, the process followed the method outlined in the Pharmacopoeia of the People's Republic of China. Briefly, approximately 0.1 g of white pepper powder was weighed and mixed with 40 mL of anhydrous ethanol. The mixture was then subjected to ultrasonication (250 W, 20 kHz) for 30 min.

2.4.1. Screening of the ultrasound‑assisted extraction process by single-factor experiments

To evaluate their impact on the extraction yield, single-factor experiments were designed by varying the TRDES type (DES 1–18), solid-to-liquid ratio (1:10–1:60 g/mL), extraction temperature (25-80°C), and ultrasound‑assisted extraction time (15–75 min). In each test, a measured quantity of white pepper powder was combined with TRDES in a conical flask and processed using an ultrasonic apparatus.

Upon completion of the extraction, the samples were centrifuged at 3000 rpm for 5 min. The subsequent supernatant was subjected to HPLC analysis. The extraction yield (Y, mg/g) of PIP was defined by Equation (1).

$$Y=\frac{\mathit{Mass}\mathit{of}\mathit{measured}\mathit{PIP}\left(mg\right)}{\mathit{Mass}\mathit{of}\mathit{white}\mathit{pepper}\left(g\right)}$$ (1)

2.4.2. Optimization of ultrasound‑assisted extraction parameters by RSM

Preliminary ultrasound‑assisted extraction conditions were determined through single-factor experiments. Subsequently, the extraction conditions were optimized employing a Box-Behnken design (BBD, Design-Expert Software Version 13). Independent variables, including solid-to-liquid ratio (A), extraction temperature (B), and extraction time (C), were investigated for process improvement. The extraction yield (Y) of PIP served as the response to determine the optimal conditions. The corresponding levels for all designed factors are presented in Table 2.

Table 2.

Design factors and levels of response surface.

Variables	Units	Symbol	Variable levels
			−1	0	1
Solid-liquid ratio	mL/g	A	20	40	60
Extraction temperature	◦C	B	60	70	80
Extraction time	min	C	15	30	45

2.5. Recovery of TRDES and PIP

The TRDES extract containing PIP was transferred to a test tube and cooled to room temperature. A defined volume of water (1–20 times the volume) was introduced, and the mixture was subsequently maintained at 4℃ for 24 h. This addition resulted in the precipitation of water-insoluble PIP from the previously homogeneous TRDES-aqueous solution. After the precipitation was complete, the sample was centrifuged at 3000 rpm for 5 min. The supernatant was transferred and collected. Following this, the pellet was subjected to three washes with 1 mL of ultrapure water, with subsequent drying in a vacuum oven at 60°C for 24 h. Subsequently, a phase transition was induced by heating the solution above its critical temperature, followed by the collection of the upper TRDES phase for recycling. The content of residual PIP in the recovered TRDES layer was determined, and the recovery (R%) was calculated using the formula (2) reported in the literature [38].

$$R\%=\frac{C_0{V}_0-C_1{V}_1}{C_0{V}_0}\times 100\%$$ (2)

where C₀ and C₁ represent the concentration of PIP in the TRDES before and after recovery, respectively, and V₀ and V₁ denote the volume of the TRDES before and after recovery, respectively.

2.6. Reuse of the TRDES

Considering the need to reduce processing time and minimize TRDES retention in the aqueous phase, a rotary evaporator (RE-2000A, Shanghai Yarong Biochemical Instrument Factory, China) was employed for the recovery of TRDES. To evaluate the reusability of the TRDES, it was subjected to five consecutive extraction cycles, and its extraction efficiency for PIP was assessed after each cycle.

2.7. Analysis of extract

The sample extracts were analyzed using a UPLC-ESI-MS/MS system (UPLC, ExionLC™ AD, https://sciex.com.cn/) and a Tandem mass spectrometry system (https://sciex.com.cn/). The analytical conditions were as follows: UPLC: column, Agilent SB-C18 (1.8 µm, 2.1 mm × 100 mm); The mobile phase consisted of solvent A, pure water with 0.1% formic acid, and solvent B, acetonitrile with 0.1% formic acid. Sample measurements were performed with a gradient program that employed the starting conditions of 95% A, 5% B. Within 9 min, a linear gradient to 5% A, 95% B was programmed, and a composition of 5% A, 95% B was kept for 1 min. Subsequently, a composition of 95% A, 5.0% B was adjusted within 1.1 min and kept for 2.9 min. The flow velocity was set as 0.35 mL per minute; The column oven was set to 40°C; The injection volume was 2 μL. The effluent was alternatively connected to an ESI-triple quadrupole-linear ion trap (QTRAP)-MS.

The ESI source operation parameters were as follows: source temperature 500°C; ion spray voltage (IS) 5500 V (positive ion mode)/-4500 V (negative ion mode); ion source gas I (GSI), gas II(GSII), curtain gas (CUR) were set at 50, 60, and 25 psi, respectively; the collision-activated dissociation (CAD) was high. QQQ scans were acquired as MRM experiments with collision gas (nitrogen) set to medium. DP (declustering potential) and CE (collision energy) for individual MRM transitions were determined with further DP and CE optimization. A specific set of MRM transitions was monitored for each period according to the metabolites eluted within this period.

2.8. In vitro pharmacodynamic evaluation of pepper extract

2.8.1. Antioxidant activity

The free radical scavenging abilities of the crude extracts were tested by the 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) methods. In accordance with the kit manuals, vitamin C (Vc) was selected as the positive control.

For the assay, equal quantities of the TRDES and ethanol crude extracts were dissolved in the kit's proprietary extraction solvent before analysis. For the DPPH assay, 5 μL of crude extract was reacted with 195 μL of DPPH working solution. Control experiments included: 1) a Blank Control (5 μL extraction solvent + 195 μL DPPH working solution), and 2) a Sample Blank (5 μL crude extract + 195 μL ethanol). Following mixing, all samples were incubated in the dark at room temperature for 30 min, and the absorbance was subsequently measured at 515 nm.

The ABTS assay was carried out following the same sample and control addition procedure. The mixtures were incubated under the same conditions for 6 min, after which the absorbance was recorded at 405 nm. All experiments were conducted in triplicate.

The free radical scavenging activity (D%) of the extracts against DPPH and ABTS was determined according to the equation below:

$$D\%=\frac{A_0-\left(A_2-A_1\right)}{A_0}\times 100\%$$ (3)

where A₀ is the absorbance of the blank control, A₁ is the absorbance of the sample blank, and A₂ is the absorbance of the test sample.

2.8.2. Hypoglycemic activity

The α-glucosidase (α-GC) inhibition assay was conducted as per the kit instructions, using ethanol to dissolve equal amounts of TRDES and ethanol crude extracts and acarbose as a positive control. To correct for background interference, four setups were used: test sample (α-GC + sample), sample control (sample + H₂O), blank test (α-GC + solvent), and blank control (solvent + H₂O). After mixing and incubating for 2 min at room temperature, absorbance was measured at 400 nm. All tests were run in triplicate.

The α-GC inhibition rate (I_α-GC%) for each sample was calculated using the following formula:

$$I_{\alpha -\mathrm{G}\mathrm{C}}\%=\frac{\left(A_0-A_1\right)-\left(A_2-A_3\right)}{A_0-A_1}\times 100\%$$ (4)

where A₀ is the absorbance of the blank test group, A₁ is the absorbance of the blank control group, A₂₂ is the absorbance of the test sample group, and A₃ is the absorbance of the sample control group.

Following a previously reported protocol [40], the inhibitory activity against α-amylase (α-AL) was evaluated. The TRDES and ethanol crude extracts were dissolved in ethanol at equal concentrations, whereas acarbose (used as the positive control) was dissolved in ultrapure water. The α-AL solution was diluted with PBS to a final activity of 150 U/L before use. 30 μL aliquot of the sample solution was mixed with 30 μL of the α-AL solution. After thorough mixing, the mixture was incubated in a water bath at 37°C for 10 min. Then, 30 μL of starch solution was added to each tube, followed by a second incubation in a boiling water bath (100°C) for 15 min. After the solution cooled to room temperature, 50 μL of 3,5-dinitrosalicylic acid (DNS) reagent was added. The mixture was vortexed and heated in a boiling water bath for 5 min. Once cooled to room temperature again, 50 μL of ultrapure water was added, and the absorbance was measured at 540 nm. All samples were analyzed in triplicate.

The α-AL inhibition rate (I_α-AL%) for each sample was calculated using the following formula:

$$I_{\alpha -\mathrm{A}\mathrm{L}}\%=1-\frac{\left(A_1-A_2\right)}{A_3}\times 100\%$$ (5)

where A₁ is the absorbance of the mixture containing the sample solution and α-AL solution, A₂ is the absorbance of the mixture where the α-AL solution is replaced with PBS, and A₃ is the absorbance of the sample solvent.

2.9. Green assessment

The Green Analytical Procedure Index (GAPI) software (Płotka-Wasylka,2018) provides a tool for greenness evaluation and visualization, which facilitates the comparison of analytical procedures [41], [42]. The assessment results are presented in an intuitive diagram using a color code (red, yellow, green) to signify high, medium, and low environmental impact. Thus, the greenness of the TRDES-based extraction method was systematically compared with that of the traditional ethanol-based method using GAPI.

2.10. The extraction principle of TRDES

2.10.1. Scanning electron Microscope (SEM)

The morphological characteristics of the white pepper powder, the drug residue after TRDES and ethanol extraction, were observed using a Scanning Electron Microscope (SEM, FEI CZECH REPUBLIC S.R.O, USA) at magnifications of 2000× and 10000×. The three powder samples were separately placed on specimen stubs with conductive adhesive, subjected to a 3 min spray-gold treatment, and then imaged.

2.10.2. Density functional theory (DFT)

To gain insight into the intermolecular interactions, DFT calculations were performed to determine the binding energy of the synthesized TRDES system. Furthermore, the binding energies between different TRDES and PIP were evaluated using the same computational method. All DFT simulations were performed using the Materials Studio software package (version 2020, BIOVIA). The molecular structures of the TRDES and PIP were constructed using Materials Studio (MS) software. Geometry optimization and binding energy calculations were subsequently performed using the DMol3 module. The relative strength of different TRDES was compared based on the absolute value of their interaction energy (Eint). A larger absolute Eint value indicates a stronger hydrogen-bonding network between the HBA and HBD, resulting in a more stable TRDES. Similarly, a larger absolute binding energy between TRDES and PIP suggests stronger intermolecular interactions, which is more favorable for the extraction of PIP.

3. Results and discussion

3.1. Preparation and characteristics of TRDESs

Eighteen types of TRDES were successfully synthesized and subsequently evaluated for their viscosity profiles under different conditions (Table 3). For detailed characterization, TRDES-1 (lidocaine/valeric acid) was chosen as a model system and analyzed by FT-IR, DSC, and ¹H NMR.

Table 3.

List of proposed TRDESs.

NO.	HBA	HBD	Molar ratio(HBA: HBD)	Viscosity (mPa·s)
				25℃	80℃
TRDES-1	Lidocaine (Lido)	Valeric acid (VA)	1:1	131.2	25.81
TRDES-2		Hexanoic Acid (HA)		166.1	26.61
TRDES-3		Heptanoic Acid (Hep A)		190.8	32.06
TRDES-4		Octanoic Acid (OA)		200.8	33.78
TRDES-5		Nonanoic Acid (NA)		190.2	26.84
TRDES-6		Decanoic Acid (DA)		192.0	41.07
TRDES-7		Lauric Acid (LA)		207.8	41.74
TRDES-8		Myristic Acid (MA)		223.8	46.70
TRDES-9	Procaine (Pro)	Valeric acid (VA)	1:1	1934	183.3
TRDES-10		Hexanoic Acid (HA)		2002	192.5
TRDES-11		Octanoic Acid (OA)		2122	200.4
TRDES-12		Decanoic Acid (DA)		2416	191.5
TRDES-13		Lauric Acid (LA)		2658	195.4
TRDES-14	Tetracaine (TET)	Valeric acid (VA)	1:1	483.5	70.19
TRDES-15		Hexanoic Acid (HA)		490.8	71.04
TRDES-16		Octanoic Acid (OA)		522.2	72.59
TRDES-17		Decanoic Acid (DA)		529.1	79.83
TRDES-18		Lauric Acid (LA)		556.3	79.55

In the FT-IR spectrum of valeric acid (Fig. 1A), a broad absorption band appearing from 2500 to 3300 cm⁻¹ signifies the O-H stretching vibration of the carboxyl group. Meanwhile, the strong, sharp peak at 1711.84 cm⁻¹ is characteristic of the C=O stretching vibration from the same functional group. The appearance of these distinctive peaks collectively confirms the molecular structure of valeric acid. Two primary features are evident in the FT-IR spectrum of lidocaine (Fig. 1B). A broad peak at 3255.18 cm⁻¹ is characteristic of the secondary amine N-H stretch, while the strong signal at 1664.64 cm⁻¹ originates from the amide C=O stretch. These diagnostic peaks collectively confirm the identity of lidocaine. The FT-IR analysis of the lidocaine/valeric acid TRDES provides clear evidence of hydrogen bonding. The amide N-H stretching vibration shifted to 3263.35 cm⁻¹, broadened considerably, and lost sharpness. This was accompanied by the broadening of both the carbonyl (C=O) and the carboxylic acid (O-H) absorption bands. These spectral modifications provide clear evidence for the formation of hydrogen bonds between the amide group of lidocaine and the carboxyl group of valeric acid.

Fig. 1.

FT-IR spectra (A), ¹H NMR spectra (B), DSC spectra (C).

A comparative analysis of the ¹H NMR spectra (Fig. 1B) provides direct evidence for the interactions in the TRDES. The spectrum of pure lidocaine displays characteristic signals at: δ 9.16 ppm (amide –NH), 7.07 ppm (aromatic protons), 3.14 ppm (−N-CH₂-), 2.14 ppm (acetyl –CH₃), and 1.07 ppm (diethyl –CH₃), which are in full agreement with its molecular structure. In the ¹H NMR spectrum of valeric acid (Fig. 1B), a characteristically broad and low-intensity signal for the carboxylic acid proton (–COOH) was observed at 11.93 ppm. The methylene protons were assigned as follows: the α-CH₂ signal appeared at 2.18 ppm, the β-CH₂ at 1.47 ppm, and the γ-CH₂ at 1.28 ppm. A terminal methyl signal (–CH₃) was observed at 0.86 ppm. All signals are consistent with the structure of valeric acid. Upon the formation of the TRDES, the intensity of the proton signals was significantly reduced, and several weak signals from lidocaine and valeric acid disappeared, notably the –COOH signal of valeric acid. This is attributed to the complex chemical environment around the hydrogen species of the carboxyl and amide groups, resulting from strong hydrogen-bonding interactions between lidocaine and valeric acid. Furthermore, the absence of new resonance signals suggests that the formation of the TRDES is a physical process.

The DSC measurements, limited to a minimum of −150°C by the instrument (Fig. 1C), recorded the melting points of the individual components at −58.54°C for valeric acid and 10.59°C for lidocaine. In marked contrast, the TRDES (lidocaine/valeric acid) showed no detectable melting transition over the entire tested range (−150°C to 150°C), indicating a melting point below the detection limit. This profound depression of the melting point serves as confirmation of the successful TRDES synthesis.

3.2. Screening and optimization of the ultrasound‑assisted extraction method

3.2.1. Selection of the optimal TRDES

Given that the physicochemical properties of DESs vary with their composition, they exert a significant influence on the extraction efficiency of PIP. Therefore, this study screened three categories of TRDESs (TRDES 1–18), constructed from lidocaine, procaine, or tetracaine as HBAs and fatty acids of varying carbon chain lengths as HBDs. When the total molar ratio of HBA to HBD deviated from 1:1, the resulting TRDES found it hard to exhibit a temperature-responsive phase transition. Therefore, the molar ratio for all TRDES used in this study was fixed at 1:1 [34].

The data presented in Fig. 2A reveal a clear trend: TRDES formulated with lidocaine as the HBA consistently achieved higher PIP extraction efficiencies, highlighting the significance of HBA selection. This correlation, where the closer match between the log P of PIP (2.94) and lidocaine (2.44) results in superior compatibility within the lidocaine-based TRDES, as predicted by the “like dissolves like” principle. Concurrently, as can be seen from Table 3, the lidocaine-based TRDES generally possessed substantially lower viscosity compared to the other two systems. This lower viscosity is likely another contributing factor to their superior performance, as it facilitates more efficient mass transfer during extraction. Moreover, a notable inverse correlation was observed between the carbon chain length of the HBD and the PIP extraction efficiency across the lidocaine-based series (TRDES 1–8). This phenomenon occurs because extending the carbon chain directly impacts critical physicochemical parameters of the TRDES, including viscosity and acidity, which are one of the key determinants of extraction efficiency. Three key factors influenced the selection of the HBD: viscosity, lipophilicity, and acidity. With increasing carbon chain length of the HBD, the viscosity of the TRDES increased, which impedes mass transfer during extraction. Similarly, the log P value also rose with chain length (e.g., from 3.39 for C₅ to 7.90 for C₁₄). According to the “like dissolves like” principle, a lower log P is more compatible with the extraction of PIP. Furthermore, the acidity of the HBD decreases with longer carbon chains; shorter-chain carboxylic acids possess higher acidity, which can more effectively disrupt cell walls and facilitate the release of the target compound. To sum up, TRDES-1 was selected as the optimal extraction solvent due to its high extraction efficiency.

Fig. 2.

Effect of the type of TRDESs (A), Solid-liquid ratio (B), Extraction temperature (C), and Extraction time (D) on the extraction yield of PIP.

3.2.2. Effect of solid–liquid ratio

The impact of the solid-to-liquid ratio (1:10–1:60 mg/mL) on PIP extraction efficiency was carried out at a fixed temperature (80°C) and time (30 min), and the results were displayed in Fig. 2B. The extraction yield showed a mountain-type fluctuation, achieving an optimal value at 1:40 mg/mL before decreasing with further solvent addition. The extraction yield showed an initial increase followed by a decrease beyond a 1:40 mg/mL solid-to-liquid ratio. This phenomenon can be explained by two opposing effects: while a larger solvent volume enhances contact and mass transfer, an excess requires disproportionate energy for phase transition, reducing the efficiency of cell wall disruption and PIP release. Furthermore, excessive solvent use is economically and environmentally unfavorable. Thus, the ratio of 1:40 mg/mL was identified as the optimal condition, representing a practical compromise between maximizing extraction yield and minimizing solvent consumption.

3.2.3. Effect of extraction temperature

Using the optimal solid-to-liquid ratio (1:40 mg/mL) and a fixed extraction time of 30 min, the extraction temperature was optimized between 25 and 80°C. As shown in Fig. 2C, the extraction efficiency of PIP displayed an increasing trend with rising temperature. Below its critical temperature, the LCST-type TRDES exhibits hydrophilic characteristics, while PIP is a hydrophobic compound. This discrepancy, coupled with the higher viscosity of the TRDES at 25°C, which impedes mass transfer, results in low extraction efficiency at this temperature. When the temperature exceeds the critical point, the hydrophobicity of the TRDES intensifies with further heating, and its viscosity decreases significantly compared to that at 25°C, thereby favoring mass transfer during extraction. Therefore, 80°C was preliminarily selected as the optimal temperature, which yielded the highest extraction rate.

3.2.4. Effect of extraction time

Using the previously optimized solid-to-liquid ratio (1:40 mg/mL) and extraction temperature (80°C), the extraction time was optimized over a range of 15 to 75 min. As shown in Fig. 2D, the extraction efficiency of PIP increased with time from 15 to 30 min, reaching a maximum at 30 min. An extraction time of less than 30 min resulted in insufficient extraction, leading to lower PIP yields. Conversely, extending the time beyond 30 min up to 75 min did not lead to a significant further increase in efficiency. Since prolonged extraction is costlier and contradicts the principles of economy and environmental friendliness, 30 min was preliminarily selected as the optimal extraction time.

3.2.5. Optimization by RSM

Guided by the results of the single-factor experiments, which helped define the preliminary optimal ranges, the ultrasound‑assisted extraction process was further optimized using a Box-Behnken Design (BBD) with three factors and three levels. The factors and their respective levels were as follows: A (Solid-liquid ratio: 20–60 mg/mL), B (Extraction temperature: 60-80°C), and C (Extraction time: 15–45 min). The experimental design, detailed in Table 2, required a total of 17 runs, the results of which are provided in Table S1.

The quadratic polynomial equation was used to replace the experimental points, the experimental results were explained, and the mathematical model was given:

Y = -13.86125 + 0.223462 A + 1.12055B + 0.217283C- 0.000975 AB + 0.000092 AC + 0.000067 BC-0.002002 A²- 0.007582 B²- 0.003670 C².

The statistical analysis of the model (Table 4 and S2) yielded a significant p-value (0.0091), and the high R2 value (0.8993) indicated excellent predictive accuracy. The close agreement between Adjusted R2 and Predicted R2, a low C.V. (1.19%), and a high Adeq Precision (6.442) collectively demonstrate the model's robustness and reliability for guiding the design space. Analysis of the F-values revealed the hierarchy of factor effects on the response as: B (Extraction temperature) > C (Extraction time) > A (Solid-liquid ratio).

Table 4.

ANOVA for quadratic model.

Source	Sum of Squares	df	Mean Square	F-value	p-value
Model	9.54	9	1.06	6.95	0.0091	significant
A-Solid-liquid ratio	0.0153	1	0.0153	0.1003	0.7607
B-Extraction temperature	0.3872	1	0.3872	2.54	0.1552
C-Extraction time	0.0528	1	0.0528	0.346	0.5749
AB	0.1521	1	0.1521	0.9965	0.3514
AC	0.003	1	0.003	0.0198	0.892
BC	0.0004	1	0.0004	0.0026	0.9606
A2	2.7	1	2.7	17.69	0.004	significant
B2	2.42	1	2.42	15.86	0.0053	significant
C2	2.87	1	2.87	18.81	0.0034	significant
Residual	1.07	7	0.1526
Lack of Fit	0.0603	3	0.0201	0.0798	0.9676	not significant
Pure Error	1.01	4	0.252
Cor Total	10.61	16

The shape of the contour map reflects the strength of the interaction between the two factors. The circle indicates that the interaction between the two factors is not significant, and the ellipse or saddle indicates that the interaction is significant. Combined with the contour map of Fig. 3, it can be seen that the shape of Fig. 3E and Fig. 3F is nearly circular, indicating that the interaction between A and C, B and C is not significant. The shape in Fig. 3D is nearly elliptical, indicating that the interaction between A and B is better than that between A and C, B and C. Similarly, the three-dimensional surface response diagram can also reflect the interaction between the two factors. The surfaces in Fig. 3A, B, and C all show peaks, but the surface does not show obvious distortion, indicating that there is no significant interaction between the three factors.

Fig. 3.

3D surface and contour plots of solid–liquid ratio, extraction temperature and extraction time.

The optimum conditions generated by the BBD were a solid-to-liquid ratio of 40.6576 mL/g, a temperature of 75.3135°C, and an extraction time of 28.5475 min. Considering operational feasibility, the parameters were rounded to 40 mL/g, 75°C, and 30 min. To validate the model's predictive accuracy, triplicate experiments were performed, yielding an average extraction yield of 33.7502 mg/g with a low RSD of 1.6%. This result shows excellent agreement with the model's prediction (33.7188 mg/g), thereby verifying the feasibility of the BBD optimization approach for PIP extraction.

To evaluate the viability of TRDES extraction as a substitute for conventional organic solvents (e.g., ethanol), white pepper was extracted following the method described in Section 2.4. The extraction yield of PIP was 35.8774 mg/g, which is slightly higher than that achieved with the TRDES method. However, when TRDES is used for extraction, its temperature-responsive behavior allows direct recovery of the crude pepper extract. Separation of PIP from TRDES can be accomplished without complex procedures. Therefore, from the perspectives of green chemistry and sustainable development, TRDES can serve as a viable alternative to ethanol for the extraction of PIP from white pepper, thereby reducing potential harm to human health and the environment.

3.3. Recovery of TRDES and PIP

The recycling process leverages the temperature-responsive nature of TRDES and the hydrophobicity of PIP. The TRDES-water system can form hydrogen bonds of varying number and strength, which govern the solution's state and properties, ultimately enabling phase transition. Below the critical temperature, water molecules cluster around the hydrogen-bonding sites of the TRDES, forming a robust network that promotes miscibility, resulting in a single homogeneous phase. As the temperature increases, the number and strength of these hydrogen bonds decrease, reducing the overall solution ionicity and causing the TRDES to transition into a hydrophobic molecular form. This shift triggers the system to separate into two distinct phases. Consequently, when the ambient temperature is below the critical point, the TRDES exhibits hydrophilic behavior. This property allows the TRDES extract to be miscible with water, causing the hydrophobic PIP to precipitate out of the TRDES-water system, thereby achieving its recovery.

The effect of different water addition volumes (ranging from 1 to 20 times) on the hydrogen bond network within the system was investigated. As shown in Fig. 4A, the optimal PIP recovery efficiency was achieved at a 10 times water volume. An insufficient amount of water led to an excessively high concentration of TRDES in the system, which increased the retention of PIP in the TRDES phase. Conversely, an excessive amount of water caused dilution of PIP, thereby hindering its effective precipitation. Therefore, a 10 times water volume was ultimately selected for the precipitation step.

Fig. 4.

Effect of water addition on PIP recovery% (A), Extraction performance of recycled TRDES (B).

A variety of alternative solvents have been reported for the efficient extraction of alkaloids from natural products. For instance, Alula Yohannes et al. employed a tropine-type ionic liquid for the extraction of tropane alkaloids from Radix physochlainae [43]. In another study, Wei Dai et al. utilized a conventional DES composed of lactic acid and fructose to extract hypaphorine from Nanhaia speciosa, achieving an extraction yield of 3.15 mg/g under optimized conditions [44]. However, the subsequent separation of the target alkaloid from the DES remains challenging and is still largely dependent on macroporous adsorption resins. In contrast, the TRDES-based approach presented in this work enables a more convenient and efficient extraction and separation process, thereby aligning more closely with the principles of green chemistry and sustainability.

3.4. Reuse of the TRDES

To achieve efficient solvent recycling after PIP recovery, the TRDES was recovered by rotary evaporation. This method was chosen to streamline the process and mitigate TRDES loss in the aqueous phase, thereby enhancing practicality. The reusability of the TRDES was then assessed over five extraction cycles.

Although a marginal decrease was observed over five cycles, the extraction efficiency of PIP using TRDES consistently remained above 85%, confirming its robust recycling capability (Fig. 4B). The observed decline is postulated to be caused by the incremental retention of PIP in the TRDES matrix during repeated use, which could moderately impede further extraction efficiency.

3.5. Analysis of extract

Different extraction solvents exhibit distinct selectivity towards target natural products. Therefore, subsequent analysis and identification of the extracts are necessary to verify the efficacy of the solvent selection. Qualitative analysis of TRDES and ethanol extracts using a UPLC-ESI-MS/MS system led to the identification of various compounds, including alkaloids, flavonoids, amino acids, and their derivatives, among twelve categories of constituents, as detailed in Table S3. In the ethanol extract, a total of 159 alkaloid components were identified, including piperine and piperoleine A, accounting for 35.33% of the total detected compounds. Meanwhile, the TRDES extract yielded 274 alkaloid components, such as piperine, piperlongumine, and piperoleine A, representing 42.02% of the total constituents. It is noteworthy that, as can be seen from Fig. 5, 78 alkaloids were identified as common constituents in both extracts, including amide alkaloids such as piperine. However, the TRDES extract contained a greater number of alkaloids, indicating its superior broad-spectrum selectivity for alkaloids in white pepper.

Fig. 5.

Venn diagram of two extracts.

PIP, the primary active constituent of white pepper, is a key quality indicator of its extracts. To provide a comparative basis for quality assessment, the PIP content in both the TRDES and ethanol extracts was quantified by HPLC. Given that the PIP content from TRDES extraction (122.3 μg/mg) was only marginally lower than that from conventional ethanol extraction (131.9 μg/mg), TRDES can be considered a functional replacement for ethanol in this application.

3.6. In vitro pharmacodynamic evaluation of pepper extract

One of the core pharmacological actions of PIP lies in its potent antioxidant capacity, which enables it to directly scavenge free radicals and enhance the body’s intrinsic defense systems, thereby protecting cells from oxidative damage [45]. At the same time, studies have confirmed that PIP exhibits clear potential in lowering blood glucose. It acts by inhibiting the activities of α-GC and α-AL, preventing the hydrolysis of glycosidic bonds, delaying carbohydrate breakdown, and reducing the postprandial rise in blood sugar—thereby contributing positively to diabetes management [46]. Therefore, we have selected four straightforward in vitro efficacy assays to evaluate the antioxidant and hypoglycemic activities of different pepper extracts.

3.6.1. Antioxidant activity

DPPH and ABTS assays are commonly used to evaluate antioxidant capacity in vitro. As shown in Fig. 6, the TRDES extract demonstrated high scavenging activity in the DPPH assay, while its ABTS radical scavenging capacity was lower than that of the ethanol extract. Overall, the antioxidant capacities of the two extracts were comparable. This indicates that using TRDES for PIP extraction does not negatively impact its bioactivity.

Fig. 6.

Comparison of the in vitro efficacy of two pepper extracts.

3.6.2. Hypoglycemic activity

As shown in Fig. 6, compared with the conventional ethanol extract, the TRDES extract exhibited more potent inhibitory effects on both α-GC and α-AL. This indicates that the TRDES-based extraction method can enhance the hypoglycemic activity of the crude pepper extract to some extent compared to the ethanol-based approach. These findings further support the feasibility and potential advantages of using TRDES for the extraction of active components in natural products.

3.7. Green assessment

GAPI was utilized to conduct a greenness assessment comparing the TRDES extraction process with the traditional ethanol extraction. This metric evaluates environmental impacts across the entire methodology, covering key areas ranging from the procedural steps and solvent choice to the instrumentation used. By integrating these diverse factors, GAPI produces a visual profile that enables a clear comparison of the overall environmental performance.

The GAPI assessment results, displayed in Fig. 7, reveal a significantly higher greenness score for the TRDES method (82) compared to the conventional ethanol extraction (68). This difference is visually apparent in their respective profiles: the TRDES profile is dominated by 9 green zones, whereas the ethanol profile shows a predominance of 7 red zones. The stark contrast primarily stems from the solvent choice and its recyclability, positioning the TRDES method as a more environmentally sustainable and promising alternative.

Fig. 7.

GAPI assessment of TRDES extraction and organic solvent extraction.

3.8. The principle of TRDES extraction

3.8.1. SEM analysis

SEM analysis provided further insights into the principle of TRDESs for extracting alkaloids from white pepper. As shown in Fig. 8, the untreated white pepper powder exhibited a fibrous, non-porous surface. In contrast, the powders extracted with TRDES and ethanol both developed visible pores, with the TRDES-treated sample showing a more pronounced cavitation effect and deeper pores, which are more conducive to the release of PIP. However, compared to ethanol, the higher viscosity of TRDES impedes mass transfer of PIP into the solvent. Consequently, although TRDES causes more substantial microstructural disruption, the final extraction yields of the two methods are not significantly different. It is noteworthy that, based on the results of component identification, the TRDES extract contained a greater variety of alkaloids than the ethanol extract.

Fig. 8.

SEM images of the extracted with TRDES (A) (D), extracted with ethyl alcohol (B) (E), and sample powder before extraction (C) (F).

3.8.2. DFT analysis

The structures and energies of different TRDES types and their corresponding TRDES-PIP complexes are shown in Fig. 9. As illustrated, the hydrogen bonds in lidocaine/valeric acid, lidocaine/HA, and lidocaine/heptanoic acid primarily form between the N and O atoms of the lidocaine amide group (R–CO–NH–R') and the H atom of the fatty acid's carboxyl group (–COOH). In contrast, the hydrogen-bonding network in procaine/valeric acid and tetracaine/valeric acid mainly involves interactions between the O atom of the procaine or tetracaine carbonyl group (–C=O) and the H atom of the fatty acid carboxyl group. Typically, a larger absolute value of Eint indicates stronger intermolecular interactions (such as hydrogen bonding and van der Waals forces) and a greater attractive effect. As shown in Fig. 9, the order of absolute Eint values for these TRDES is lidocaine/valeric acid (10.07) > lidocaine/HA (9.13) > tetracaine/valeric acid (9.40) > procaine/valeric acid (8.47) > lidocaine/ heptanoic acid (2.49). This trend indicates that, when the HBD (valeric acid) is fixed, TRDES with lidocaine as the HBA exhibit higher interaction energies. This may be attributed to the unique amide structure of lidocaine, which differs from the other two HBAs. The nitrogen and oxygen atoms in its amide group can both participate in hydrogen bonding. Moreover, due to the p-π conjugation in the amide group, it exhibits an electron-donating effect, while the –COOH group in the fatty acid has an electron-withdrawing effect. This electronic complementarity likely contributes to the formation of more stable hydrogen bonds. Similarly, when the HBA (lidocaine) is fixed, TRDES formed with shorter-chain fatty acids exhibit stronger and more stable hydrogen-bonding networks. This can be attributed to the reduced steric hindrance of shorter-chain fatty acids, which facilitates easier interaction with the nitrogen and oxygen atoms in lidocaine.

Fig. 9.

The structural model and binding energy of TRDESs and TRDES-PIP.

Given that all these TRDES exhibit certain extraction efficiency for PIP, simulations were also conducted to investigate their interactions with PIP. The order of absolute Eint values for the TRDES-PIP systems is as follows: lidocaine/valeric acid-PIP (12.56) > lidocaine/hexanoic acid-PIP (10.68) > procaine/valeric acid-PIP (11.08) > tetracaine/valeric acid-PIP (9.39) > lidocaine/heptanoic acid −PIP (5.02). The higher Eint values observed in the lidocaine-based systems suggest that the binding sites between TRDES and PIP may vary depending on the TRDES composition. In the lidocaine-based TRDES system, the primary interaction occurs with the hydrogen atoms of the terminal five-membered ring in the PIP structure. In contrast, the procaine-based system mainly binds to the oxygen atom in the amide group of PIP, while the tetracaine-based system primarily interacts with the hydrogen atoms along the carbon chain of PIP. The amide group in PIP features a distinct p–π conjugation system, resulting in a stronger electron-donating effect than an electron-withdrawing effect. Meanwhile, the ether oxygen in the five-membered ring exhibits a strong electron-withdrawing inductive effect, causing the terminal hydrogen atoms of the ring to carry a partial positive charge. This electronic environment enhances the intermolecular interactions between the lidocaine-based system and PIP, making them stronger than those in the procaine and tetracaine systems. The lidocaine-based TRDES yielded a higher extraction rate than the other two types, which is consistent with the findings detailed in Section 3.2.1. Therefore, the lidocaine-based TRDES is more suitable for the extraction of PIP. Within the same lidocaine-based system, shorter fatty acid chain length corresponds to a greater absolute Eint value. This may be attributed to the fact that the intermolecular interactions between lidocaine/valeric acid and PIP primarily occur between the two oxygen atoms of the fatty acid –COOH group and the hydrogen atoms of the terminal five-membered ring in PIP. In contrast, the interactions between lidocaine/hexanoic acid and PIP mainly involve the –C=O group in the lidocaine amide structure and the hydrogen atoms of PIP's terminal five-membered ring. The interaction between lidocaine/heptanoic acid and PIP mainly involves the oxygen atom on the carboxyl group of fatty acid and the hydrogen atom of the five-membered ring at the end of PIP. Moreover, due to the p–π conjugation in the amide structure, its electron-withdrawing effect is weaker than that of the –COOH group in valeric acid. Within the same TRDES system, shorter-chain fatty acids exhibit an enhanced binding affinity for PIP, thus improving their efficacy in PIP extraction processes. In agreement with the experimental results in Section 3.2.1, DFT calculations confirm that the lidocaine/valeric acid pair exhibits a higher extraction rate for PIP. Consequently, lidocaine/valeric acid demonstrated superior performance and was selected as the solvent with the highest extraction efficiency for PIP in this experiment.

4. Conclusion

A method was established for the extraction and purification of piperine from white pepper. Utilizing TRDES as the extraction solvent combined with ultrasound-assisted technology, this approach enables the green and efficient preparation of the active ingredient. Throughout the ultrasound‑assisted extraction and separation process, the use of organic solvents was avoided. By leveraging the temperature-responsive behavior of the TRDES, PIP was directly separated from the system, while simultaneously enabling the recycling and reuse of the TRDES.

We synthesized 18 types of TRDES using lidocaine, procaine, and tetracaine as HBA and fatty acids as HBD, and measured their viscosity values at different temperatures. The optimal TRDES was identified as the 1:1 M ratio formulation of lidocaine and valeric acid, which effectively extracts PIP from pepper. Single-factor experiments and RSM analysis revealed the optimal ultrasound‑assisted extraction conditions as follows: solid-to-liquid ratio of 1:40 mg/mL, extraction temperature of 75°C, and extraction time of 30 min, resulting in a PIP extraction yield of 33.7502 mg/g. The synergistic combination of ultrasound and DESs enables a green and highly efficient extraction approach. SEM imaging demonstrated that TRDES induces substantial structural disruption to plant fibers. Utilizing the temperature-dependent hydrophilic-hydrophobic transition characteristics of TRDES, PIP recovery was evaluated, yielding a recovery rate of 41.76%. Compositional analysis and in vitro efficacy assessments were subsequently performed on the recovered crude extract. While all extracts exhibited comparable antioxidant activity, the TRDES-derived extract showed significantly enhanced hypoglycemic activity compared to that obtained using ethanol extraction.

The TRDES was successfully reused for five consecutive extraction cycles, with the PIP extraction efficiency remaining above 85% of the original value in each cycle. The SEM results indicated that the ultrasound-assisted TRDES method produced a superior cavitation effect, facilitating the dissolution of various alkaloid components. DFT calculations revealed that lidocaine/valeric acid exhibits the highest binding energy (12.56) with PIP, indicating that the hydrogen-bonding network formed between lidocaine/valeric acid and the terminal five-membered ring of PIP is stronger than those formed by other TRDES. From a green chemistry perspective, the TRDES extraction procedure received a higher GAPI score (82) compared to ethanol (68), demonstrating its superior environmental profile. These results indicate that the integrated TRDES and ultrasound-assisted extraction system holds considerable potential for extracting active compounds from Chinese herbal medicines. By virtue of avoiding additional organic solvents and complex processing steps, it enables efficient target compound separation, while the cavitation effect of ultrasound enhances the dissolution of active components. This combined approach thus offers an efficient, straightforward, and green strategy for the extraction and preparation of bioactive ingredients. This study provides an effective guide for green extraction methods of natural products.

CRediT authorship contribution statement

Huixin You: Writing – original draft, Software, Methodology, Formal analysis. Haonan Shen: Software. Hejie Han: Visualization. Yongjing Liu: Writing – review & editing, Funding acquisition, Conceptualization. Hua Li: Supervision.

Funding

This research was supported by the National Natural Science Foundation of China (Grant/Award Number: 81903928) and the Project of the Department of Science and Technology of Fujian Province (Grant/Award Number: 2024J01128).

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

The following are the Supplementary data to this article: