G Model

ARTICLE IN PRESS

CHROMA-356998; No. of Pages 8

Journal of Chromatography A, xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media Kyung Min Jeong a , Min Sang Lee a , Min Woo Nam a , Jing Zhao a , Yan Jin a , Dong-Kyu Lee b , Sung Won Kwon b , Ji Hoon Jeong a , Jeongmi Lee a,∗ a b

School of Pharmacy Sungkyunkwan University, 2066 Seoburo, Suwon 16419, Republic of Korea College of Pharmacy Seoul National University, 1 Gwanak-ro, Seoul 08826, Republic of Korea

a r t i c l e

i n f o

Article history: Received 17 August 2015 Received in revised form 25 October 2015 Accepted 27 October 2015 Available online xxx Keywords: Deep eutectic solvent Solvent design Solvent recycling Green extraction Ginsenoside

a b s t r a c t The present study demonstrates that deep eutectic solvents (DESs) with the highest extractability can be designed by combining effective DES components from screening diverse DESs. The extraction of polar ginseng saponins from white ginseng was used as a way to demonstrate the tuneability as well as recyclability of DESs. A newly designed ternary DES (GPS-5) composed of glycerol, l-proline, and sucrose at 9:4:1 was used as a sustainable and efficient extraction medium. Based on the anti-tumor activity on HCT-116 cancer cells, it was confirmed that GPS-5 was merely an extraction solvent with no influence of the bioactivity of the ginsenosides extracted. Excellent recovery of the extracted saponins was easily achieved through solid-phase extraction (SPE). Recycling of the DES was accomplished by simple freezedrying of the washed solutions from the SPE. The extraction efficiencies of the DESs recycled once, twice, and thrice were 92%, 85%, and 83% of that of the freshly synthesized solvent. © 2015 Published by Elsevier B.V.

1. Introduction Over the past two decades, much attention has been paid to ionic liquids (ILs) as sustainable alternatives to hazardous organic solvents [1]. ILs are a class of fluid that is formed from the combination of anions and cations with melting points below 100 ◦ C [2]. ILs can be synthesized either from eutectic mixtures of metal halides and organic salts, or from those of discrete ions [2]. More recently, deep eutectic solvents (DESs), have been recognized as a novel class of sustainable solvents to replace common organic solvents or even ILs [1–3]. DESs are fluid systems formed from a eutectic mixture of two or more components that are naturally occurring, safe, and inexpensive components. While melting points of DESs are much lower than the individual components, most DESs are liquid between ambient temperature and 70 ◦ C [4]. Although DES components can contain a variety of anionic and/or cationic species, they can be associated with each other through intermolecular hydrogen bonding [4]. DESs are considered superior to ILs due to their biodegradability, non-toxicity, and low costs for synthesis, in addition to the tuneability, negligible volatility, and wide polarity range that are shared by ILs [3,5,6]. DESs have been

∗ Corresponding author. E-mail address: [email protected] (J. Lee).

used in various research fields such as catalysis, organic synthesis, electrochemistry, material chemistry, and extraction processes [1–4]. Extraction of natural products from herbal medicines using green, safe, and efficient solvents is important in pharmaceutical and biochemical research fields [7–10]. Because bioactive natural products vary greatly in polarity, maximized extractability can be achieved by tuning the polarity of the extraction solvent. Although it is known that the properties of DESs can be tailored by changing the components and their molar ratios [3], studies showing the true tuneability of DESs as extraction solvents are very limited. In our previous study, we demonstrated the tuneability of DESs as designer solvents for selective and efficient extraction of bioactive natural products. Using Flos sophorae as a model system, we demonstrated that flavonoids could be efficiently extracted using a tailored DES that was newly synthesized by combining two effective DES components, glycerol and l-proline. In this study, we tested a hypothesis that our previous strategy was generally applicable to a wide variety of classes of bioactive natural products. The previous study was focused on the extraction of one common class of natural products, flavonoids, which are relatively nonpolar. Applicability to a completely different class of natural products would support the expandability of our previous strategy. For this purpose, we employed ginseng as a model system, which is one of the most popular and valuable traditional

http://dx.doi.org/10.1016/j.chroma.2015.10.083 0021-9673/© 2015 Published by Elsevier B.V.

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model CHROMA-356998; No. of Pages 8

ARTICLE IN PRESS K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

2

herbal medicines. We designed a DES with the highest extraction efficiencies for the major bioactive compounds in ginseng, called ginsenosides, which are saponin-based and exist in a wide range of polarities. Ginseng is known to possess many active components with useful effects including anti-aging, anti-diabetic, anti-tumor, and tranquilizing activities, most of which are attributable to ginsenosides [11–14]. The composition of ginsenosides varies depending on the species, processing method, growth environment, etc. [15] In this study, white Korean ginseng (Panax ginseng C.A. Meyer) was selected for experiments due to its easy procurement and because it is reported to contain mostly ginsenosides that are relatively polar such Rg1 , Re, Rf, Rb1 , Rc, and Rb2 [16–23]. A large number of methods have been reported for the extraction of ginsenosides from white ginseng [24–32]. In general, organic solvents such as methanol, ethanol, and their aqueous solutions have been commonly used in combination with various extraction methods including heat reflux extraction (HRE), pressurized liquid extraction (PLE), ultrasound-assisted extraction (UAE), and supercritical fluid extraction (SFE). Although some of these reported methods have displayed high extraction efficiencies for ginsenosides, these methods generally require large quantities of organic solvents, long extraction times, and large amounts of energy. The aim of this study was to tailor DESs and optimize extraction conditions to maximize ginsenoside extraction efficiencies. 2. Materials and methods 2.1. Chemicals, reagents, and equipment Compounds used for DES preparation included choline chloride (≥98.0%), glycerol (≥99.5%), l-proline (≥99.0%), xylitol (≥99.0%), citric acid (≥99.5%), adonitol (≥99.0%), betaine (≥99.0%), d(+)-galactose (≥99.0%), d-(−)-fructose (≥99.0%), d-(+)-glucose (≥99.5%), dl-malic acid (≥99.0%), and sucrose (≥99.5%), all of which were obtained from Sigma–Aldrich (St. Louis, MO, USA). Analytical standards of ginsenosides including Rg1 , Re, Rf, Rb1 , Rd, and Rc (for chemical structures, see Supplementary Fig. S1), all of which had a purity of 95% or higher based on HPLC, and finely pulverized white ginseng powder (diameter < 355 ␮m) produced from authenticated 6-year old Panax ginseng C.A. Meyer, were kindly provided by Prof. Jeong Hill Park (College of Pharmacy, Seoul National University, Seoul, Korea). HPLC-grade acetonitrile, water and methanol were purchased from J.T. Baker (Phillipsburg, NJ, USA). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and formic acid (≥98.0%) were obtained from Sigma–Aldrich. Doubly deionized water was prepared using a Milli-Q water purification system (Millipore, Bedford, MA, USA). Analytical-grade ethanol and pH 0–14 universal pH-indicator strips were purchased from Merck KGaA (Darmstadt, Germany). SPE cartridges (OASIS HLB 6 cc, 200 mg) were purchased from Waters (Milford, MA, USA). Centrifuges, models 1580MGR and Gyrospin, were obtained from Gyrozen (Incheon, Korea), while the Eppendorf 1524 was from Eppendorf (Hamburg, Germany). An ultrasonic bath (Powersonic 410) and a freeze dryer (model FD8508) were obtained from Hwashin Technology (Seoul, Korea) and Ilshin Biobase (Yangju, Korea), respectively. 2.2. Preparation of analytical standard solutions Each ginsenoside stock solution of Rg1 , Re, Rf, Rb1 , and Rc was prepared in methanol at 1 mg mL−1 and stored at −20 ◦ C. Standard working solutions were prepared by diluting the stock solutions

with water and were used for analytical method validation including linearity and assay precision and accuracy. 2.3. Procedures for DES preparation DESs were synthesized using a freeze-drying method as previously described [7]. Water added in the mixture was then removed by lyophilization for 24 h or longer until the mixture reached a constant weight. 2.4. LC–UV analysis for the quantification of extracted ginsenosides Liquid chromatography coupled to ultraviolet detection (LCUV) was performed using a PerkinElmer LC system (Norwalk, CT, USA) equipped with a PerkinElmer interface 600 series link, a quaternary pump (series 200), an auto-sampler (series 200), and a UV-visible detector (series 200). TotalChrom Workstation software was used for system operation and data management. The detection wavelength was 203 nm. Standard compounds and extracts were chromatographed on a Waters Xbridge phenyl column (5 ␮m, 4.6 mm × 150 mm) from Waters (Milford, MA, USA) at a flow rate of 1.0 mL min−1 at 25 ◦ C. The mobile phase consisted of water (A) and acetonitrile (B), and the binary linear gradient elution was as follows: 0–20 min, 20–22% B; 20–23 min, 22–28% B; 23–45 min, 28–35% B; 45–55 min, maintained at 35% B for 10 min. The system was returned to the initial conditions within 1 min and equilibrated for 20 min before subsequent injections. The standards and extracts diluted in water were filtered through a 0.45 ␮m membrane filter (Whatman, Piscataway, NJ, USA) prior to injection. The established LC-UV method was validated in terms of linearity, precision, and accuracy, and the validation results are summarized in Supplementary Table S1. The calibration curve for each ginsenoside was plotted as peak area versus concentration of each ginsenoside standard. Assay precisions and accuracies were determined using quality control (QC) samples prepared at three different concentrations (low, middle, and high) for intra-day (n = 3) and inter-day (n = 3 × 3) assays within one day and on three separate days, respectively. 2.5. UHPLC–Q-TOF–MS analysis for the qualitative analysis of extracted ginsenosides An Acquity UPLC system (Waters Co., Milford, MA, USA) was composed of a binary solvent delivery system and a cooling autosampler maintained at 4 ◦ C. Ginsenoside standards and ginseng extracts were chromatographed on an Acquity UPLC BEH C18 column (50 mm × 2.1 mm, 1.7 ␮m) from Waters at a flow-rate of 0.35 mL min−1 at 40 ◦ C. A linear gradient system was employed for elution using a mobile phase consisting of (A) 0.1% formic acid in water and (B) 0.1% formic acid in acetonitrile: 0–20 min, 10–90% B; 20–21 min, 90–100% B, followed by washing with 100% B for 1 min. The extract was diluted with water to produce a four-fold dilution of the original extract and was filtered through a 0.2 ␮m membrane filter (Whatman) prior to injection. Mass spectrometric analysis was conducted using a Waters Acquity Xevo G2 Q-TOF tandem mass spectrometer (Waters Co., Manchester, UK) equipped with an electrospray ionization interface in positive and negative ion mode, which was controlled by Masslynx software (version 4.1, Waters Co., Milford, MA, USA). The experimental conditions were described elsewhere [7]. Peak identification of the ginseng extract was performed based on accurate mass measurements in comparison to standard compounds and values reported in the literature [23,33–35].

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model CHROMA-356998; No. of Pages 8

ARTICLE IN PRESS K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

2.6. Evaluation of extraction efficiencies among different extraction methods The extraction efficiency of each extraction method was assessed based on the levels of total ginsenosides extracted. Each extract was analyzed using the LC-UV method, and the levels of five ginsenosides, Rg1 , Re, Rf, Rb1 , and Rc were determined. For the initial DES screening, an accurately weighed 100 mg sample of powder was mixed with 1 mL of extraction solvent (water, methanol, ethanol, aqueous methanol, aqueous ethanol, and various types of DESs) in a 2 mL microfuge tube. After brief vortexing, the mixture was irradiated with ultrasound starting at ambient temperature under the maximum power (∼500 W) for 45 min and then centrifuged at 12,300 × g for 25 min. Temperature of the ultrasonic bath was not controlled. The clear supernatant was diluted with water for chromatographic analysis. During the optimization processes and after the extraction condition optimization, the extraction procedures for UAE were performed in the same way as described above except for irradiation time and DES volume. Stirring, heating, and stirring + heating methods were compared to the UAE method for the extraction involving the final designed DES (GPS-5). One hundred milligrams of ginseng powder was extracted in 1 mL of GPS-5 for 45 min by stirring (600 rpm, room temperature), heating (60 ◦ C), or stirring + heating (600 rpm, 60 ◦ C). 2.7. Conventional extraction methods used for ginsenoside extraction–heating, UAE, and HRE Extraction methods based on the literature were reproduced with modifications: the liquid-to-solid ratio was fixed at 1.9 mL per 100 mg of ginseng powder. In the UAE method (Ref-1), 70% v/v aqueous methanol was used for ultrasonic radiation for 30 min [27]. The heating method (Ref-2) included the extraction of 100 mg of ginseng in 80% v/v aqueous methanol at 70 ◦ C for 1 h [30]. HRE (Ref3) was performed on 10 g of ginseng powder using 190 mL of 100% methanol at 60 ◦ C for 1 h [29]. 2.8. Recovery of extracted ginsenosides from DES and recycling of DES Recovery of the ginsenosides from the DES extracts was conducted based on the SPE method using HLB cartridges [36]. Briefly, a cartridge was placed in a vacuum manifold and equilibrated with 5 mL of ethanol, followed by 5 mL of water. After loading the DES extract three-fold diluted in water, the cartridge was rinsed with 6 mL of water and dried under vacuum. The aqueous wash was collected and lyophilized until the mixture reached a constant weight, indicating the regeneration of the DES (GPS-5). For ginsenoside elution, 6 mL of ethanol was slowly added and the collected eluate was evaporated with a gentle stream of nitrogen at room temperature. The residues were reconstituted in 40% aqueous methanol and subjected to LC-UV analysis to evaluate the recovery efficiency. The regenerated DESs were tested for the second round of ginsenoside extraction. 2.9. Experimental design and statistical analysis RSM was carried out using the Design-Expert Ver, 8.0 (Statease Inc., Minneapolis, MN, USA). Central composite design (CCD) was performed to determine the optimal values for the three variables [37], ultrasonic irradiation time (A, min), DES content in the extractant (B, w/w %), and extractant volume per 100 mg of sample powder (C, mL). A total of 20 experiments including six replicates of center points, eight factorial points, and six axial points, were randomly conducted in blocks.

3

Extraction yields for ginsenosides were statistically compared using a two-tailed t-test and one-way analysis of variance (ANOVA) in GraphPad Prism 5.01 for Windows (GraphPad Software, San Diego, CA, USA). 2.10. Evaluation of the potential influence of DES on anti-tumor activity of ginsenosides Human colorectal cancer cell lines HCT-116 (ATCC, Manassas, VA, USA) were grown in McCoy’s 5A medium supplemented with 10% fetal bovine serum and 50 IU penicillin/streptomycin. Cells were maintained at 37 ◦ C and 5% CO2 in a humidified incubator. Ginseng extracts prepared in 70% ethanol and GPS-5 aqueous solution using the UAE method were diluted in each extraction solvent to produce a series of ginseng extracts at different concentrations (0.5–10 mg mL−1 based on the extracted amounts of five major ginsenosides). In order to eliminate the potential influences of these solvents, the extracts were diluted 10-fold with water before being applied to the cells. Cytotoxic effects of the solvents free of ginsenoside extracts were also tested after 10fold dilution with water. Cell viability was determined with MTT assay as described previously [38]. The formed MTT formazan crystals were measured at 490 nm using a MultiskanTM GO microplate spectrophotometer (Thermo Fisher Scientific, MA, USA). The final ginsenoside concentrations in the cell culture were in the range of 5–100 ␮g mL−1 . 3. Results and discussion 3.1. Selection of bioactive compounds for comparison of extraction efficiency Five extraction solvents commonly used for ginsenoside extraction were tested as reference solvents, including water, 70% aqueous ethanol, 70% aqueous methanol, 100% ethanol, and 100% methanol. During the screening processes to choose effective DES components, simple and effective extraction methods that can also enable simultaneous extraction of a large number of samples are preferable. UAE, which has been shown to be efficient extraction method having good compatibility with ILs and DESs [39–42], was used as the extraction method for screening, using the same amounts of ginseng powder and extraction solvent volume, 100 mg and 1.0 mL, respectively, for every extraction solvent. Extraction time was maintained at 45 min, which appeared long enough to prevent any potential variances due to insufficient extraction time. Extraction efficiencies of the reference solvents varied significantly (Supplementary Fig. S2). Nonetheless, the overall extraction patterns were similar. Specifically, Rg1 , Rb1 , and Re were commonly the major ginsenosides extracted, while Rf and Rc were extracted in smaller amounts (Supplementary Fig. S2). In addition to these five ginsenosides, other less polar ginsenosides such as Rb2 and Rh2 and different classes of natural products such as linoleic acid were also detected using ultra-high performance liquid chromatography–quadrupole-time of flight–mass spectrometry (UHPLC–Q-TOF–MS; Supplementary Table S2). However, the levels of these compounds, if detected at all, were so low that reliable quantification by LC-UV analysis was not feasible, and they had little effect on the total extraction yields. Accordingly, the five main ginsenosides, Rg1 , Re, Rf, Rb1 , and Rc were quantified in the subsequent analyses. Because the extraction amounts of the individual ginsenosides and their summed amounts were changed in similar patterns depending on the extraction solvents and methods, the totals of these five ginsenosides were compared to evaluate extraction efficiencies for simplicity.

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model

ARTICLE IN PRESS

CHROMA-356998; No. of Pages 8

K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

4

3.2. Preparation of various types of DESs DESs can be prepared from combinations of hydrogen bond donors (HBDs) and hydrogen bond acceptors (HBAs) at various molar ratios using different methods including heating, evaporating, and freeze-drying [2,3,8,43]. In this study, a freeze-drying method was employed because it allowed for simultaneous synthesis of a number of DESs, and the low production temperature prevented potential thermal degradation of heat-labile components such as sugars and amino acids [2,7,44,45]. The main criteria for DES component selection in the present study were easy procurement, low cost, safety, and good biodegradability. Based on previous reports and our own experiences [7,8], a number of HBAs and HBDs were combined at various ratios. As a result, 18 different types of DESs, i.e., five citric acid-based, six choline chloride-based, and seven glycerol-based DESs, were successfully produced as clear and stable liquids at room temperature without solid precipitation (column I in Table 1). As previously discussed [7], only the contents of major components were taken into consideration for convenience, although water was contained as one component of the DESs produced. During the study, water content of the produced DESs was controlled through tight capping of tubes containing DESs [7]. 3.3. Screening of DESs for the initial solvent design For easy handling of DESs, a portion of water is often added, thereby reducing their viscosities [8,46]. In this study, all produced DESs were mixed with water at 7:3 (w/w) and used for screening [7]. Initial screening results for the 18 DESs in comparison with the five reference solvents are displayed in Fig. 1. Among the five reference solvents, water, the most preferable green solvent, exhibited an extraction yield comparable to that of 100% methanol and was almost two times more efficient than 100% ethanol. However, 70% ethanol displayed the highest extraction efficiency (5.8 mg g−1 ), which was similar to that of 70% methanol Table 1 A total list of all DESs synthesized and tested in this study. I

II a

Cit:Ado (1:1)

III

IV

b

Gly:Pro:Suc (5:4:1)

b

Gly:Pro:Suc (1:4:1; GPS-1) Gly:Pro:Suc (3:4:1; GPS-2) Gly:Pro:Suc (5:4:1; GPS-3) Gly:Pro:Suc (7:4:1; GPS-4) Gly:Pro:Suc (9:4:1; GPS-5) Gly:Pro:Suc (11:4:1; GPS-6) Gly:Pro:Suc (13:4:1; GPS-7)

Pro:Suc (4:1)

Cit:Glu (2:1)

Bet:Suc (4:1)

Gly:Bet:Suc (5:4:1)

Cit:Gal (1:1)

ChCl:Sucb (4:1)

Gly:ChCl:Suc (5:4:1)

Cit:Pro (2:1) Cit:Bet (1:1) ChCl:Mal (1:1) ChCl:Cit (2:1) ChCl:Xyl (5:2) ChCl:Ado (5:2) ChCl:Glu (1:1) ChCl:Fru (1:1) Gly:ChClb (1:1) Gly:Xyl (1:1) Gly:Prob (1:1) Gly:Betb (1:1) Gly:Mal (1:1) Gly:Fru (1:1) Gly:Sucb (3:1) a

Molar ratio. Tested for extraction efficiency in Fig. 2. Abbreviations: Cit, citric acid; Ado, adonitol; Glu, d-(+)-glucose; Gal, d-(+)-galactose; Pro, l-proline; Bet, betaine; Mal, dl-malic acid; ChCl, choline chloride; Xyl, xylitol; Fru, D-(−)-fructose; Gly, glycerol; Suc, sucrose. b

Fig. 1. Extraction yields (milligrams of five ginsenosides per gram of white ginseng powder) of the 18 initially synthesized DESs in comparison to the five reference solvents. Error bars indicate the SEM (n = 3).

(p = 0.5210; Fig. 1). These results demonstrated that the solubility of ginsenosides differed depending on the polarity of extraction solvents used. None of the 18 DESs tested exhibited an extraction yield higher than 70% ethanol. However, a series of choline chloride-based (ChCl:Xyl, ChCl:Ado, ChCl:Glu, and ChCl:Fru) and glycerol-based (Gly:ChCl, Gly:Xyl, Gly:Pro, Gly:Bet, Gly:Fru, and Gly:Suc) solvents were similarly effective; none of their extraction yields were significantly different from that of 70% ethanol (p > 0.05). On the other hand, the DESs containing organic acids, ChCl:Mal, ChCl:Cit, and Gly:Mal, yielded significantly lower extraction efficiencies than other choline chloride- or glycerol-based solvents. Consistent with these findings, the extraction efficiencies of the citric acid-based DESs, especially Cit:Ado, Cit:Glu, and Cit:Gal, were generally low compared to those of choline-chloride- and glycerol-based DESs. It is known that the acidity of a DES is strongly affected by the native properties of the HBDs and that DESs containing sugarderived HBDs exhibit neutral pHs [4,47]. Based on the pH measured, moderate levels of acidity (pH < 4) were observed in the organic acid-containing DESs such as ChCl:Mal and ChCl:Cit, while the pH values of other DESs were found to be above 4. Thus, it can be concluded that the organic acid-based DESs were too acidic to effectively extract neutral ginsenosides from ginseng. Taken together, these results show that choline chloride and glycerol were generally effective components for ginsenoside extraction, and their extraction properties were modified by their counterpart components. In particular, betaine and l-proline effectively improved the extraction efficiencies of glycerol-based DESs, as well as those of citric acid-based solvents. Sucrose, a disaccharide, appeared to be a potentially better enhancing component compared to monosaccharide (fructose) or sugar alcohol component (xylitol) when it was combined with glycerol, although the differences were not significant. 3.4. DES tailoring – synthesis of new binary and ternary DESs Based on the results above, we attempted to synthesize a new series of DESs with enhanced extraction efficiency by combining two of the selected effective DES components, choline chloride, glycerol, betaine, l-proline, and sucrose. These five components can theoretically produce 10 different combinations. However, the combination of choline chloride and betaine, both of which function only as HBAs, was not feasible, while glycerol, l-proline,

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model CHROMA-356998; No. of Pages 8

ARTICLE IN PRESS K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

Fig. 2. Extraction yields (milligrams of five ginsenosides per gram of white ginseng powder) of the binary DESs composed of effective components. Extraction efficiencies that were significantly lower than that of Pro:Suc were indicated with *p < 0.05, **p < 0.01, and ***p < 0.001. Error bars indicate the SEM (n = 3).

and sucrose can act as HBAs or HBDs. Among the possible nine combinations, the four glycerol-based solvents, Gly:ChCl, Gly:Pro, Gly:Bet, and Gly:Suc, had been previously synthesized (column I in Table 1), and three new sucrose-based solvents, Pro:Suc, Bet:Suc, and ChCl:Suc were successfully synthesized at the molar ratios previously reported (column II in Table 1) [8,46]. Measurement of the extraction abilities showed that the new combinations of sucrose with betaine (Bet:Suc) or choline chloride (ChCl:Suc) were not as effective as the glycerol-based solvents (Fig. 2). In contrast, Pro:Suc exhibited the highest ginsenoside extraction efficiency among the seven tailored solvents, with an its efficiency much higher than that of 70% ethanol (p < 0.05). DESs can be synthesized from two or more components [8,48]. In this study, glycerol was a generally effective component in the screening tests. Maugeri and de Maria showed that glycerol added to binary DESs as a third component resulted in reduced viscosity of ternary DESs [47]. These findings led us to hypothesize that we could further tailor ternary DESs by adding glycerol to the sucrosebased DESs. We first added an equimolar amount of glycerol to the other two components. To our surprise, ternary DESs containing glycerol were produced relatively easily, and the resulting solvents were Gly:Pro:Suc, Gly:Bet:Suc, and Gly:ChCl:Suc (column III in Table 1). More excitingly, as intended, the extraction efficiencies of ternary DESs were significantly improved compared to their corresponding binary DESs (p < 0.05 for all three cases; Fig. 3). The extraction yield of Gly:Pro:Suc was 24% higher than that of 70% ethanol (p < 0.01). Continuing on the path of our previous study [7], the current results show that DESs can be tuned to achieve the desired extractability against bioactive compounds of diverse structures and properties. Because molar ratios of the components were found to affect the extraction efficiencies in the case of flavonoid extraction from F. sophorae [7], the effects of molar ratios of glycerol, l-proline, and sucrose were investigated in the new ternary DES. For simplicity, only the glycerol ratio was varied from 1 to 13, while the ratio of l-proline and sucrose was fixed at 4:1 (column IV in Table 1). As a result, GPS-4 and GPS-5, which are the DESs containing glycerol, lproline, and sucrose at 7:4:1 and 9:4:1, respectively, displayed the highest extraction yields (Supplementary Fig. S3). GPS-5, which had an extraction yield slightly higher than that of GPS-4 but without reaching statistical significance (p = 0.10), was selected as the final

5

Fig. 3. Improved extraction efficiencies (milligrams of five ginsenosides per gram of white ginseng powder) in the ternary DESs due to the addition of glycerol. Significantly different extraction efficiencies due to the addition of glycerol are indicated with *p < 0.05 and **p < 0.01. Error bars indicate the SEM (n = 3).

Fig. 4. Comparison of the extraction efficiency of UAE with other extraction methods using GPS-5. Experimental conditions are described in Section 2.7. Extraction efficiencies that were significantly lower than that of GPS-5 are indicated with **p < 0.01 and ***p < 0.001. Error bars indicate the SEM (n = 3).

extraction solvent and the subsequent condition optimization was conducted. 3.5. Optimization of the final operation conditions 3.5.1. Selection of the extraction method UAE was employed as the extraction method during the initial screening and solvent tailoring processes above because it has been shown to be very efficient and convenient for viscous solvents including ILs and DESs [49–52]. Prior to optimization of the final operation conditions, other types of extraction methods that are compatible with the selected DES, GPS-5, including stirring, heating, and heating + stirring methods [7,10,30,53], were compared with UAE. As displayed in Fig. 4, stirring, heating, and heating + stirring methods exhibited no discernable differences in extraction yield. In contrast, the level of extracted ginsenosides using the UAE method was at least 11% higher than those of the other three methods (p < 0.01). Accordingly, further extraction

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model CHROMA-356998; No. of Pages 8

ARTICLE IN PRESS K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

6

conditions were optimized using UAE as the extraction method, and the optimization procedure is described below. 3.5.2. Optimization of the final extraction conditions using RSM Several numerical variables that could affect the extraction efficiencies were optimized by experimental design based on the RSM approach in order to yield the highest efficiency in ginsenoside extraction. Similar to previous studies [7,46], three variables of extraction time, DES content in the extraction solvent, and liquidto-sample powder ratio, were evaluated for optimization. For the RSM, a central composite design was employed because it is the one most commonly used for building a quadratic response surface model [37]. The ultrasonic irradiation time (A), content of GPS-5 in the extraction solvent (B), and volume of the extraction solvent per 100 mg of sample powder (C) were varied at five levels (−˛, −1, 0, +1, +˛) as follows: A, 20.0–70.0 min; B, 50.0–100.0% w/w; C, 0.7–2.3 mL. For the response, the extraction yield of each ginsenoside or the total extraction yield of all ginsenosides was used as a measure of extraction efficiency. Since the patterns of extraction yield generally did not differ between these two (data not shown), the sum amount of the five ginsenosides extracted was used as the response. Experiments conducted according to the design resulted in a second-order polynomial equation for extraction yield (Y) expressed using coded variables as follows: Y = 7.26 − 0.074A − 0.89B + 0.21C + 0.0098AB − 0.38AC + 0.47BC + 0.17A2 − 0.92B2 + 0.042C 2 The model was evaluated based on the ANOVA results at the 95% confidence level (Supplementary Table S3). The model ‘Fvalue’ of 5.96 and ‘Prob > F’ less than 0.05 indicated that this model is significant. The insignificant lack-of-fit value (p = 0.1888) and high correlation coefficient value (R2 = 0.8701) helped validate the model quality. According to the model, only the terms related to GPS-5 content in the extraction solvent (B and B2 ) significantly affected extraction yield (p < 0.05), whereas extractant volume and extraction time had no significant influences, similar to the results of a study by Dai et al. (Supplementary Table S3 and Supplementary Fig. S4) [46]. A solution was sought for the highest desirability with minimized extraction time. The resulting conditions were A = 21.5 min, B = 66.1% w/w, and C = 1.9 mL, with a predicted extraction yield was 8.40 mg g−1 . Experimental values performed under these conditions were determined to be 8.16 ± 0.12 mg g−1 , n = 5, which fell within the 95% prediction interval (PI), 6.18–10.63 mg g−1 , supporting the reproducibility of the model. Water contents in the DES-based extraction solvents have been found to be important for tuning extractability of solvents [7,10,46,54,55]. Our model suggests that GPS-5 content negatively influences extraction yield. In fact, the yields were lowest at 100% GPS-5 without water addition (Supplementary Figs. S4a and S4c). Improved extractability from the increased water content might be attributed to reduced viscosity and/or increased polarity of the extraction solvents [8,10,54,55]. However, the reduced viscosity is unlikely to be responsible for the altered extractability, because the extraction yields at any GPS-5 content remained unchanged regardless of the extraction time (Supplementary Fig. S4a). This finding indicates that the ultrasonic radiation enabled very rapid mass transfer between the ginsenosides and solvent. Previously, it was suggested that DESs with high water content are suitable for the extraction of polar compounds [8,10,46]. In this study, the optimal composition of the extractant contained a relatively high water content, implying that enhanced extractability for ginsenosides was mainly achieved by the increased solvent polarity caused

Table 2 Comparison of the optimized GPS-5-based method with conventional extraction methods. Method Ref-1 Ref-2 Ref-3 Our method a b

Extraction yield (mg g−1 ) a

7.48 (±0.13) 6.46 (±0.03) 7.26 (±0.02) 8.24 (±0.13)

t-testb

Reference

p < 0.01 p < 0.001 p < 0.001 –

[27] [30] [29]

Mean ± S.D. (n = 3). Statistical difference in comparison to the optimized GPS-5-based method.

by the addition of water. These results are consistent with our observations that aqueous methanol with a higher polarity was more efficient than pure methanol (Fig. 1). Taken together, our results suggest that the extraction properties of DESs can be tailored by combining individual effective components, followed by fine tuning by adjustment of water content. 3.6. Evaluation of extraction efficiency of the optimized method in comparison to conventional methods The extraction yields of the optimized conditions above were compared with those from three previously reported methods with slight modifications [27,29,30]. The optimized liquid-to-sample powder ratio (1.9 mL of extraction solvent to 100 mg of ginseng powder) was equally applied to the three reference methods to enable the evaluation of only the effects of extraction solvent and extraction method. Among the three reference conditions, Ref-1 exhibited the highest extraction yield despite the short extraction time (30 min vs. 1 h; Table 2). It is noteworthy that increasing the volume of 70% methanol effectively enhanced the extraction yields of UAE; the yield of Ref-1 obtained using 1.9 mL of 70% methanol was 25% larger than that obtained with the initial reference conditions using 1 mL of 70% methanol (Fig. 1). The other two reference methods were much less efficient than Ref-1 in spite of long extraction times and high extraction temperatures. These results demonstrate that UAE can be applied as a very efficient, energy-saving extraction method regardless of the type of extraction solvent. As compared in Table 2, the extraction yield of our optimized method using GPS-5 was significantly higher than that of any of the reference methods. More importantly, this highest extraction yield was achieved in only 36–72% of the time required for the other extraction methods. Conventional methods for the extraction of ginsenosides usually involve a large volume of volatile organic solvent and a long extraction time. Moreover, they are often conducted at a high temperature, which requires more energy. In fact, based on the electric power of each instrument used and the extraction time, the calculated electric energy consumption of our method (179 W h) was lower than the reference methods (250–800 W h). Therefore, our newly optimized method using the tailored GPS-5 and UAE is clearly efficient, eco-friendly, non-toxic, and energysaving. 3.7. Investigation of potential influence of the tailored DES on the bioactivity of the ginsenosides extracted The tailored DES, GPS-5 as the extraction solvent might influence bioactivities of the ginsenosides extracted. Among a variety of bioactivities of ginsenosides, anti-tumor activity was selected to examine whether GPS-5 would alter the native activities of ginsenosides extracted. Extraction solvents themselves, 70% ethanol and GPS-5 aqueous solution, had no cytotoxic effects. The cell proliferation inhibitory effects expressed as an IC50 were acquired from dose–response curves and determined to be 58 ␮g mL−1 and 61 ␮g mL−1 for the extracts prepared in GPS-5 and 70% ethanol,

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model CHROMA-356998; No. of Pages 8

ARTICLE IN PRESS K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

respectively. These values were not significantly different (p > 0.05). Therefore, it was concluded that GPS-5 was merely an extraction solvent with no influence on the anti-tumor activity of the ginsenosides extracted. 3.8. Recovery of ginsenosides from the extracted phase and recycling of the used DES Inherent negligible vapor pressure and the generally high water miscibility of DESs make it challenging to isolate and recover extracted compounds from DESs [7,46]. In recent years, several strategies for recovery of extracted compounds have been reported, including the applications of anti-solvents, supercritical carbon dioxide, and adsorption chromatography [4,7,39]. A SPE method was found to be very simple and efficient for the recovery of flavonoids from a DES solution [7]. Thus, the SPE strategy was applied to recover the extracted ginsenosides from the GPS-5-based extracted phase in this study. To this end, a cartridge of a hydrophilic-lipophilic-balanced sorbent (Oasis® HLB) was employed to selectively enrich the ginsenosides possessing both lipophilic dammarane moieties and hydrophilic sugars. For the cartridge conditioning and elution, methanol was replaced by ethanol due to the methanol toxicity. Using the simple procedure, which included the loading of diluted extracts and washing with water, followed by elution with ethanol, the recovery was determined to be 102.6% ± 4.1% (n = 5). Recyclability of green solvent is also an important issue to be addressed. Successful recycling of DESs has been limited to those used as solvent media for chemical reactions or extraction solvents for fuels and alcohol–ester mixtures [56–58]. DES recycling for natural product extraction has never been reported. In the SPE procedure for ginsenoside recovery, the loaded DES extracts were simply washed with 6 mL of water, which produced a diluted aqueous solution of GPS-5. These washes were lyophilized to remove excess water until a constant weight was reached with the same appearance of viscous liquid phase. The regenerated solvent was then reused for ginsenoside extraction under the optimized conditions. These recycling experiments were repeated up to three times (n = 3 for each round of recycling). As a result, all the regenerated DESs were stable and produced almost the same chromatographic patterns of the extracts except for the increased peaks that were eluted very early, probably corresponding to sugars extracted from ginseng. The extraction efficiencies of the DESs recycled once, twice, and thrice were 91.9% (±2.9%), 85.4% (±2.3%), and 82.6% (±4.7%), respectively, of that for the original solvent. These results indicate that the DES can be recycled at least three times to achieve a reasonably high level of ginsenoside extraction yields. For the first time, our study demonstrated that DES can be recycled by simple lyophilization of the aqueous solution of DES produced during the recovery of extracted compounds. 4. Conclusions We designed a ternary DES, GPS-5, with the highest extractability for ginsenosides by combining three effective DES components of glycerol, l-proline, and sucrose at a ratio of 9:4:1. These results imply that DESs can be tailored for maximized extractability for any given classes of compounds from various kinds of herbal medicines or even other types of matrices. Glycerol was verified as a DES component that could be added to binary DESs to produce ternary DESs with enhanced extraction efficiencies and reduced viscosities. In combination with the efficient condition optimization based on the RSM and extraction using the convenient and efficient UAE method, the developed extraction method using GPS-5 was shown to be

7

significantly more efficient than previous time-consuming methods requiring the consumption of organic solvents. Employment of SPE to recover the extracted ginsenosides from extracts allowed for complete recovery through a very simple and fast procedure using ethanol as the elution solvent, and it also enabled easy, yet effective recycling of GPS-5 through simple freeze-drying of the washes from the SPE procedure. This study demonstrated that DESs are tuneable and are a green extraction media without potential toxicity. In addition, our study suggested a clever approach to solve the problems of both recovery of extracted compounds and recycling of DESs in extraction through the freezedrying method.

Acknowledgment This study was supported by a grant (no. 2011-0024225) from the Basic Science Research Program of the National Research Foundation of Korea funded by the Ministry of Education, Science, and Technology, Republic of Korea.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma.2015.10. 083.

References [1] A. Paiva, R. Craveiro, I. Aroso, M. Martins, R.L. Reis, A.R.C. Duarte, Natural deep eutectic solvents–solvents for the 21st century, ACS Sustain. Chem. Eng. 2 (2014) 1063–1071. [2] E.L. Smith, A.P. Abbott, K.S. Ryder, Deep eutectic solvents (DESs) and their applications, Chem. Rev. 114 (2014) 11060–11082. [3] M. Francisco, A. van den Bruinhorst, M.C. Kroon, Low-Transition-Temperature Mixtures (LTTMs): a new generation of designer solvents, Angew. Chem. Int. Ed. 52 (2013) 3074–3085. [4] Q. Zhang, K.D.O. Vigier, S. Royer, F. Jérôme, Deep eutectic solvents: syntheses, properties and applications, Chem. Soc. Rev. 41 (2012) 7108–7146. [5] D. Yang, M. Hou, H. Ning, J. Zhang, J. Ma, G. Yang, B. Han, Efficient SO2 absorption by renewable choline chloride–glycerol deep eutectic solvents, Green Chem. 15 (2013) 2261–2265. [6] Y. Dai, R. Verpoorte, Y.H. Choi, Natural deep eutectic solvents providing enhanced stability of natural colorants from safflower (Carthamus tinctorius), Food Chem. 159 (2014) 116–121. [7] M.W. Nam, J. Zhao, M.S. Lee, J.H. Jeong, J. Lee, Enhanced extraction of bioactive natural products using tailor-made deep eutectic solvents: application to flavonoid extraction from Flos sophorae, Green Chem. 17 (2015) 1718–1727. [8] Y. Dai, J. van Spronsen, G.J. Witkamp, R. Verpoorte, Y.H. Choi, Natural deep eutectic solvents as new potential media for green technology, Anal. Chim. Acta 766 (2013) 61–68. [9] S. Cull, J. Holbrey, V. Vargas-Mora, K. Seddon, G. Lye, Room-temperature ionic liquids as replacements for organic solvents in multiphase bioprocess operations, Biotechnol. Bioeng. 69 (2000) 227–233. [10] W. Bi, M. Tian, K.H. Row, Evaluation of alcohol-based deep eutectic solvent in extraction and determination of flavonoids with response surface methodology optimization, J. Chromatogr. A 1285 (2013) 22–30. [11] J.-S. Kim, K.-S. Yoon, Y.-S. Lee, Antioxidant activity of main and fine roots of ginseng (Panax ginseng CA Meyer) extracted with various solvents, Food Sci. Biotechnol. 17 (2008) 46–51. [12] G.J. Im, J.W. Chang, J. Choi, S.W. Chae, E.J. Ko, H.H. Jung, Protective effect of Korean red ginseng extract on cisplatin ototoxicity in HEI-OC1 auditory cells, Phytother. Res. 24 (2010) 614–621. [13] M.V.N. López, M.P.G.-S. Cuadrado, O.M.P. Ruiz-Poveda, A.M.V. Del Fresno, M.E.C. Accame, Neuroprotective effect of individual ginsenosides on astrocytes primary culture, Biochim. Biophys. Acta 1770 (2007) 1308–1316. [14] Z. Ji, T. Dong, W. Ye, R. Choi, C. Lo, K. Tsim, Ginsenoside Re attenuate ␤-amyloid and serum-free induced neurotoxicity in PC12 cells, J. Ethnopharmacol. 107 (2006) 48–52. [15] K.S. Kang, T. Yokozawa, H.Y. Kim, J.H. Park, Study on the nitric oxide scavenging effects of ginseng and its compounds, J. Agric. Food Chem. 54 (2006) 2558–2562. [16] Y.W. Ha, S.S. Lim, I.J. Ha, Y.-C. Na, J.-J. Seo, H. Shin, S.H. Son, Y.S. Kim, Preparative isolation of four ginsenosides from Korean red ginseng (steam-treated Panax ginseng CA Meyer), by high-speed counter-current chromatography coupled with evaporative light scattering detection, J. Chromatogr. A 1151 (2007) 37–44.

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083

G Model CHROMA-356998; No. of Pages 8 8

ARTICLE IN PRESS K.M. Jeong et al. / J. Chromatogr. A xxx (2015) xxx–xxx

[17] S.I. Kim, J.H. Park, J.-H. Ryu, J.D. Park, Y.H. Lee, J.-H. Park, T.-H. Kim, J.M. Kim, N.-I. Baek, Ginsenoside Rg5, a genuine dammarane glycoside from Korean red ginseng, Arch. Pharmacal Res. 19 (1996) 551–553. [18] I. Kitagawa, M. Yoshikawa, M. Yoshihara, T. Hayashi, T. Taniyama, Chemical studies of crude drugs (1). Constituents of Ginseng radix rubra, Yakugaku zasshi: J. Pharm. Soc. Jpn. 103 (1983) 612–622. [19] J.-H. Ryu, J.-H. Park, J.-H. Eun, J.-H. Jung, D.H. Sohn, A dammarane glycoside from Korean red ginseng, Phytochemistry 44 (1997) 931–933. [20] I.H. Park, N.Y. Kim, S.B. Han, J.M. Kim, S.W. Kwon, H.J. Kim, M.K. Park, J.H. Park, Three new dammarane glycosides from heat processed ginseng, Arch. Pharmacal Res. 25 (2002) 428–432. [21] M.R. Lee, B.S. Yun, C.K. Sung, Comparative study of white and steamed black Panax ginseng, P. quinquefolium, and P. notoginseng on cholinesterase inhibitory and antioxidative activity, J. Ginseng Res. 36 (2012) 93. [22] H.-J. Kwon, J.-S. Jeong, Y.-M. Lee, S.-P. Hong, A reversed-phase highperformance liquid chromatography method with pulsed amperometric detection for the determination of glycosides, J. Chromatogr. A 1185 (2008) 251–257. [23] H.-S. Jee, K.-H. Chang, S.-H. Park, K.-T. Kim, H.-D. Paik, Morphological characterization, chemical components, and biofunctional activities of Panax ginseng, Panax quinquefolium, and Panax notoginseng roots: A Comparative Study, Food Rev. Int. 30 (2014) 91–111. [24] A.-J. Lau, S.-O. Woo, H.-L. Koh, Analysis of saponins in raw and steamed Panax notoginseng using high-performance liquid chromatography with diode array detection, J. Chromatogr. A 1011 (2003) 77–87. [25] X. Wang, T. Sakuma, E. Asafu-Adjaye, G.K. Shiu, Determination of ginsenosides in plant extracts from Panax ginseng and Panax quinquefolius L. by LC/MS/MS, Anal. Chem. 71 (1999) 1579–1584. [26] J. Wan, F. Yang, S. Li, Y. Wang, X. Cui, Chemical characteristics for different parts of Panax notoginseng using pressurized liquid extraction and HPLC–ELSD, J. Pharm. Biomed. Anal. 41 (2006) 1596–1601. [27] X.Q. Ma, X.M. Liang, Q. Xu, X.Z. Zhang, H.B. Xiao, Identification of ginsenosides in roots of Panax ginseng by HPLC-APCI/MS, Phytochem. Anal. 16 (2005) 181–187. [28] S. Zhang, R. Chen, H. Wu, C. Wang, Ginsenoside extraction from Panax quinquefolium L., Am. Ginseng (2006) 57–63. [29] R.M. Corbit, J.F. Ferreira, S.D. Ebbs, L.L. Murphy, Simplified extraction of ginsenosides from American ginseng (Panax quinquefolius L.) for high-performance liquid chromatography-ultraviolet analysis, J. Agric. Food Chem. 53 (2005) 9867–9873. [30] E.M. Schlag, M.S. McIntosh, Ginsenoside content and variation among and within American ginseng (Panax quinquefolius L.) populations, Phytochemistry 67 (2006) 1510–1519. [31] J. Zhao, J. Deng, Y. Chen, S. Li, Advanced phytochemical analysis of herbal tea in China, J. Chromatogr. A 1313 (2013) 2–23. [32] N. Angelova, H.-W. Kong, R. Van Der Heijden, S.-Y. Yang, Y.H. Choi, H.K. Kim, M. Wang, T. Hankemeier, J. Van Der Greef, G. Xu, Recent methodology in the phytochemical analysis of ginseng, Phytochem. Anal. 19 (2008) 2–16. [33] G.X. Xie, Y. Ni, M.M. Su, Y.Y. Zhang, A.H. Zhao, X.F. Gao, Z. Liu, P.G. Xiao, W. Jia, Application of ultra-performance LC-TOF MS metabolite profiling techniques to the analysis of medicinal Panax herbs, Metabolomics 4 (2008) 248–260. [34] D.-Y. Lee, J.K. Kim, S. Shrestha, K.-H. Seo, Y.-H. Lee, H.-J. Noh, G.-S. Kim, Y.B. Kim, S.-Y. Kim, N.-I. Baek, Quality evaluation of Panax ginseng roots using a rapid resolution LC-QTOF/MS-based metabolomics approach, Molecules 18 (2013) 14849–14861. [35] M. Dan, M. Su, X. Gao, T. Zhao, A. Zhao, G. Xie, Y. Qiu, M. Zhou, Z. Liu, W. Jia, Metabolite profiling of Panax notoginseng using UPLC–ESI-MS, Phytochemistry 69 (2008) 2237–2244. [36] J. Xing, C. Xie, H. Lou, Recent applications of liquid chromatography–mass spectrometry in natural products bioanalysis, J. Pharm. Biomed. Anal. 44 (2007) 368–378.

[37] M.A. Bezerra, R.E. Santelli, E.P. Oliveira, L.S. Villar, L.A. Escaleira, Response surface methodology (RSM) as a tool for optimization in analytical chemistry, Talanta 76 (2008) 965–977. [38] M.S. Lee, J.E. Lee, E. Byun, N.W. Kim, K. Lee, H. Lee, S.J. Sim, D.S. Lee, J.H. Jeong, Target-specific delivery of siRNA by stabilized calcium phosphate nanoparticles using dopa–hyaluronic acid conjugate, J. Control. Release 192 (2014) 122–130. [39] Y. Dai, J. van Spronsen, G.-J. Witkamp, R. Verpoorte, Y.H. Choi, Ionic liquids and deep eutectic solvents in natural products research: mixtures of solids as extraction solvents, J. Nat. Prod. 76 (2013) 2162–2173. [40] F. Pena-Pereira, J. Namie´snik, Ionic liquids and deep eutectic mixtures: sustainable solvents for extraction processes, ChemSusChem 7 (2014) 1784–1800. [41] M. Sharma, C. Mukesh, D. Mondal, K. Prasad, Dissolution of ␣-chitin in deep eutectic solvents, RSC Adv. 3 (2013) 18149–18155. [42] T. Gu, M. Zhang, T. Tan, J. Chen, Z. Li, Q. Zhang, H. Qiu, Deep eutectic solvents as novel extraction media for phenolic compounds from model oil, Chem. Commun. 50 (2014) 11749–11752. [43] M.a.C. Gutiérrez, M.L. Ferrer, C.R. Mateo, F. del Monte, Freeze-drying of aqueous solutions of deep eutectic solvents: a suitable approach to deep eutectic suspensions of self-assembled structures, Langmuir 25 (2009) 5509–5515. [44] A.P. Abbott, D. Boothby, G. Capper, D.L. Davies, R.K. Rasheed, Deep eutectic solvents formed between choline chloride and carboxylic acids: versatile alternatives to ionic liquids, J. Am. Chem. Soc. 126 (2004) 9142–9147. [45] C. Ruß, B. König, Low melting mixtures in organic synthesis – an alternative to ionic liquids? Green Chem. 14 (2012) 2969–2982. [46] Y. Dai, G.-J. Witkamp, R. Verpoorte, Y.H. Choi, Natural deep eutectic solvents as a new extraction media for phenolic metabolites in Carthamus tinctorius L., Anal. Chem. 85 (2013) 6272–6278. [47] Z. Maugeri, P.D. de Maria, Novel choline-chloride-based deep-eutectic-solvents with renewable hydrogen bond donors: levulinic acid and sugar-based polyols, RSC Adv. 2 (2012) 421–425. [48] B. Tang, K.H. Row, Recent developments in deep eutectic solvents in chemical sciences, Monatshefte Chem. – Chem. Mon. 144 (2013) 1427–1454. [49] K. Vilkhu, R. Mawson, L. Simons, D. Bates, Applications and opportunities for ultrasound assisted extraction in the food industry—a review, Innov. Food Sci. Emerg. Technol. 9 (2008) 161–169. [50] X. Cao, X. Ye, Y. Lu, Y. Yu, W. Mo, Ionic liquid-based ultrasonic-assisted extraction of piperine from white pepper, Anal. Chim. Acta 640 (2009) 47–51. [51] J. Azmir, I. Zaidul, M. Rahman, K. Sharif, A. Mohamed, F. Sahena, M. Jahurul, K. Ghafoor, N. Norulaini, A. Omar, Techniques for extraction of bioactive compounds from plant materials: a review, J. Food Eng. 117 (2013) 426–436. [52] D. Pingret, A.-S. Fabiano-Tixier, F. Chemat, Ultrasound-Assisted Extraction. Natural Product Extraction: Principles and Applications, RSC Publishing, Londres, 2013, pp. 89–112. [53] L.P. Christensen, M. Jensen, U. Kidmose, Simultaneous determination of ginsenosides and polyacetylenes in American ginseng root (Panax quinquefolium L.) by high-performance liquid chromatography, J. Agric. Food Chem. 54 (2006) 8995–9003. [54] R.B. Leron, A.N. Soriano, M.-H. Li, Densities and refractive indices of the deep eutectic solvents (choline chloride + ethylene glycol or glycerol) and their aqueous mixtures at the temperature ranging from 298.15 to 333.15 K, J. Taiwan Inst. Chem. Eng. 43 (2012) 551–557. [55] E. Durand, J. Lecomte, P. Villeneuve, Deep eutectic solvents: synthesis, application, and focus on lipase-catalyzed reactions, Eur. J. Lipid Sci. Technol. 115 (2013) 379–385. [56] S.B. Phadtare, G.S. Shankarling, Halogenation reactions in biodegradable solvent: Efficient bromination of substituted 1-aminoanthra-9,10-quinone in deep eutectic solvent (choline chloride: urea), Green Chem. 12 (2010) 458–462. [57] C. Li, D. Li, S. Zou, Z. Li, J. Yin, A. Wang, Y. Cui, Z. Yao, Q. Zhao, Extraction desulfurization process of fuels with ammonium-based deep eutectic solvents, Green Chem. 15 (2013) 2793–2799. [58] Z. Maugeri, W. Leitner, P.D. de María, Practical separation of alcohol–ester mixtures using deep-eutectic-solvents, Tetrahedron Lett. 53 (2012) 6968–6971.

Please cite this article in press as: K.M. Jeong, et al., Tailoring and recycling of deep eutectic solvents as sustainable and efficient extraction media, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.10.083