Ionic Liquid-Based Ultrasonic-Assisted Extraction of Secoisolariciresinol Diglucoside from Flaxseed (Linum usitatissimum L.) with Further Purification by an Aqueous Two-Phase System

In this work, a two-step extraction methodology of ionic liquid-based ultrasonic-assisted extraction (IL-UAE) and ionic liquid-based aqueous two-phase system (IL-ATPS) was developed for the extraction and purification of secoisolariciresinol diglucoside (SDG) from flaxseed. In the IL-UAE step, several kinds of ILs were investigated as the extractants, to identify the IL that affords the optimum extraction yield. The extraction conditions such as IL concentration, ultrasonic irradiation time, and liquid–solid ratio were optimized using response surface methodology (RSM). In the IL-ATPS step, ATPS formed by adding kosmotropic salts to the IL extract was used for further separation and purification of SDG. The most influential parameters (type and concentration of salt, temperature, and pH) were investigated to obtain the optimum extraction efficiency. The maximum extraction efficiency was 93.35% under the optimal conditions of 45.86% (w/w) IL and 8.27% (w/w) Na2SO4 at 22 °C and pH 11.0. Thus, the combination of IL-UAE and IL-ATPS makes up a simple and effective methodology for the extraction and purification of SDG. This process is also expected to be highly useful for the extraction and purification of bioactive compounds from other important medicinal plants.


Selection of IL
In order to find the optimal extractant, five types of ILs were investigated for the extraction of SDG from flaxseed powder. Although the pure ILs selected for extraction have low viscosity, some studies reported that despite that ILs normally possesses high viscosity, the presence of water reduces it considerably [35,36]. Therefore, the viscosity of pure ILs is less important when the ILs aqueous solutions are used in the solid-liquid extraction. The results in Figure 1 show that the IL OTf, because of different solutions have different multi-interactions including π-π, n-π, ionic/charge-charge and hydrogen bonding with SDG of interest, which results in the decrease of extraction yield for [C4mim]OTf [37,38]; as to [C4mim]N(CN)2 and [C6mim]N(CN)2, increasing the alkyl chain length from butyl to hexyl decreases the extraction efficiency, which could be attributed to increased steric hindrance with the longer alkyl chain [39]. Therefore, this IL [C4mim]N(CN)2 has the optimal extraction ability and was chosen as the extractant for further study.

Single Factor Experiments
A number of factors, including IL concentration, liquid-solid ratio, and ultrasonic irradiation time, could affect the UAE extraction yield. Therefore, it is necessary to identify the most influential factors for obtaining the maximum extraction yield.

Effect of IL Concentration
The effect of [C4mim]N(CN)2 aqueous solutions with mass concentrations of 20%-60% (w/w) was calculated to determine the optimal concentration of IL-UAE. The liquid-solid ratio and ultrasonic irradiation time were set at constant values of 20:1 and 30 min, respectively. The maximum extraction yield was obtained at IL concentration of 50% (w/w) ( Figure 2). No obvious increase in the extraction yield was observed with further increase in the IL concentration; thus, the IL concentration of 50% (w/w) was selected for subsequent experiments.

Effect of Liquid-Solid Ratio
The liquid-solid ratio is an important parameter influencing the extraction yield. If the liquid-solid ratio is too high, the result would be waste of IL and increased cost, while a too-low ratio would result in incomplete extraction. The effect of liquid-solid ratios of 10:1, 20:1, 30:1, 40:1, and 50:1 was studied to evaluate the influence of this ratio on extraction yield. The IL concentration and ultrasonic irradiation time were set at constant values of 50% (w/w) and 30 min, respectively. The results in Figure 2 showed that the maximal extraction yield was obtained when the liquid-solid ratio was 20:1.

Effect of Ultrasonic Irradiation Time
To some extent, the ultrasonic irradiation time also plays an important role in the IL-UAE procedure, because sonication significantly shortens the extraction time. The effect of ultrasonic irradiation time from 20 to 60 min was studied by maintaining other experimental conditions constant such as IL concentration of 50% (w/w) and liquid-solid ratio of 20:1. Higher extraction yields were obtained when the ultrasonic irradiation time was longer than 40 min; no obvious increase was observed with further increase in time ( Figure 2).

Optimization of IL-UAE Conditions by the Response Surface Method (RSM)
To further study the interaction between these factors, the conditions were optimized using RSM in Design-Expert 7.0 software (Stat-Ease, Minneapolis, MN, USA). Three factors: IL concentration (40%, 50%, and 60% (w/w)), liquid-solid ratio (10:1, 20:1, and 30:1), and ultrasonic irradiation time (40,50, and 60 min), were applied using a Box-Behnken design. The experimental conditions are shown in Table 1. Experimental results of the extraction yield of SDG were analyzed by multiple regressions to fit the second-order regression equation. The coded factors of the regression model were predicted to be: where Y is the extraction yield in mg/g, A is the IL concentration (%, w/w), B is the liquid-solid ratio, and C is ultrasonic irradiation time (min). It can be seen in Table 2 that the value for the coefficient of determination (R 2 ) was 0.9883, which implies that over 98.83% of the variation in the process efficiency could be explained by the model. The model F-value of 65.53 and the model p-value of <0.0001 indicate that the model is significant. There is only less than 0.01% chance that such a large "Model F-Value" could occur due to noise, implying that the model was acceptable. A "Prob > F" of value less than 0.0500 indicates that the model terms are significant. In this case, A, B, AB, A 2 , B 2 , and C 2 are significant model terms. The response surfaces for the effects of independent variables on the average extraction yield of SDG are shown in Figure 3. Based on the quadratic model, the optimum conditions for the extraction of SDG were calculated to be IL concentration of 55.49% (w/w), liquid-solid ratio of 24.50:1, and ultrasonic irradiation time of 40 min. Triplicate runs were carried out at the optimum conditions, and the average yield was 16.3374 mg/g, which is very close to the predicted value of 16.5934 mg/g. This demonstrated that the model was adequate for reflecting the expected optimization.

Comparison with Other Extraction Methods
The heat reflux extraction (HRE) and microwave-assisted extraction (MAE) were compared with UAE to extract SDG using [C4mim]N(CN)2 as extractant. The operating temperature of the three methods was set at approximate 40 °C. The results are presented in Table 3, which shows that the IL-UAE and IL-MAE have higher extraction yield and consume a shorter extraction time than IL-HRE. The IL-MAE give similar yield with IL-UAE, however, MAE gives microwave leakage that is potentially harmful to humans compared with UAE.

Ionic Liquid-Based Aqueous Two-Phase System (IL-ATPS)
In the IL-ATPS section, a kosmotropic salt was directly added into the IL extract to construct ATPS for further purification of SDG. The influential parameters: type and concentration of salt, temperature, and pH, were investigated. by Na2SO4 had the highest extraction efficiency. The extraction ability of IL-ATPSs follows the phase-forming order of three types of salt: Na2SO4 > (NH4)2SO4 > K2HPO4. The maximum extraction efficiency was obtained when the ATPS was composed of 45.86% (w/w) IL and 8.27% (w/w) Na2SO4.

Effect of Temperature
The partitioning of SDG in IL-ATPS within the temperature range 12-32 °C was investigated, with unadjusted pH. The results are shown in Figure 4b. When the temperature was 22 °C, the extraction efficiency was maximum. The higher temperature will be not suitable for the stability for SDG. Moreover, a higher temperature will cause consumption of more energy; therefore, it is better to operate the IL-ATPS closing to the room temperature.

Effect of pH
A buffer solution (Na2HPO4-H3PO4) was used to adjust the pH of the ATPS. Na2SO4 was added into IL-UAE extract to form ATPS. The effect of pH within the range 7.0-12.0 on the extraction efficiency was investigated. After IL-ATPS, SDG was quantified by HPLC after alkaline hydrolysis of the samples in IL-rich phase. As shown in Figure 4c, it can be observed that the highest extraction efficiency was obtained at pH 11.0.
To conclude, the maximum IL-ATPS extraction efficiency of 93.35% was obtained under the following conditions: ATPS 45.86% (w/w) [C4mim]N(CN)2 and 8.27% (w/w) Na2SO4, extraction temperature 22 °C, and pH 11.0. Under the optimal IL-UAE and IL-ATPS conditions, the HPLC chromatograms for the analysis of SDG samples after IL-UAE and IL-ATPS are shown in Figure 5.

SDG Isolation and IL Recycling
After IL-ATPS, SDG is extracted into IL-rich phase, and how to recovery SDG and IL from the IL-based raw extract is becoming a difficult task. Some literature reported that liquid-liquid back-extraction with organic solvents is a reasonable approach for back-extraction of compounds and recovery of ILs [31,40]. In order to achieve this aim, several non-miscible with water organic solvents of ethyl acetate, n-butanol, chloroform, and dichloromethane were tested, to extract SDG into organic phase with IL staying in aqueous phase. SDG can be isolated and IL can be recycled with further treatment after separation of two phases. The results in Figure 6 shows that the maximal back-extraction efficiency of SDG reaches to 98.84% using n-butanol as solvent. Figure 6. Back-extraction of SDG using organic solvents.

Ionic Liquid-Based Ultrasound-Assisted Extraction
The flaxseeds were smashed to powder of about 60 meshes (250 μm) particle size. Then, n-hexane was added to the powder (5:1, v/m) and stirred for about 6 h for degreasing. After evaporating out the n-hexane and drying, degreased flaxseed powder was obtained. The IL aqueous solution was added to the degreased flaxseed powder for ultrasonic extraction of SDG in an ultrasonic bath (KQ-5200DE, Kunshan Ultrasound Co. Ltd., Kunshan, China).The electric power is 200 W, generators frequency is 40 kHz, and operating temperature of 25 °C were set for this bath. The IL extract was centrifuged, and then hydrolyzed using NaOH aqueous solutions (0.25-0.4 mol/L) for 2 h. The pH value of the extract solution was adjusted to 4.0-6.0 by HCl aqueous solution before analysis. Actually, SDG is a product of alkaline hydrolysis of polymeric lignans, the IL-UAE extract polymeric compounds without defined structure, which are subsequently hydrolyzed in presence of NaOH. The latter gives a mixture of phenolic compounds, containing a certain amount of SDG. The SDG concentration was quantified using HPLC. The extraction yield of SDG by IL-UAE was calculated according to Equation (1):

Ionic Liquids-Based Aqueous Two-Phase System
SDG was extracted from the IL aqueous solution, and a certain amount of the IL-extract and a certain amount of salt were added to a tube. The tube was shaken well to completely dissolve the salt, and then it was centrifuged for complete phase formation. Two clear phases formed, and the volume of each phase was noted. SDG sample in IL-rich phase was withdrawn using a syringe, and then hydrolyzed using NaOH aqueous solutions (0.25-0.4 mol/L) for 2 h. The pH value of the sample was adjusted to 4.0-6.0 by HCl aqueous solution before HPLC analysis. The SDG concentration in the salt-rich phase was determined by mass balance.
The phase ratio (R) was defined as in Equation (2): where Vt and Vb are the volumes of IL-rich phase and salt-rich phase, respectively. The partition coefficient (K) was defined as in Equation (3): where Ct and Cb are the SDG concentrations in the IL-rich phase and salt-rich phase, respectively. The extraction efficiency (E) of SDG in the IL-rich phase was determined from Equation (4):

HPLC Conditions
SDG was quantified using Dionex UltiMate 3000 LC Modules (Sunnyvale, CA, USA), and equipped with an LPG-3400 pump (Sunnyvale, CA, USA) and a UV-Vis detector (model: VWD-3400, Sunnyvale, CA, USA). A personal computer equipped with Chameleon software was used to collect and process the chromatographic data. The analyses were performed with a Promosil C18 column (250 × 4.6 mm I.D.; 5 μm, Bonna-Agela Technologies, Tianjin, China). The mobile phase of SDG analysis was composed of methanol and H2O (33:67; v/v) with isocratic elution. The flowrate was 1.0 mL/min, and the effluent was monitored at a wavelength of 290 nm. The column temperature was maintained 30 °C. The samples were analyzed by injecting 20 μL of the diluted sample into the HPLC. The mobile phase and samples before injection were filtered through a 0.45 μm membrane. The linear regression equation is Y = 63.504X + 0.5651 (R 2 = 0.9995), where Y is the peak area and X is the SDG concentration. Standard solutions of SDG were diluted at concentrations ranging from 0.25 to 4 mg/mL.

Conclusions
An IL-UAE coupled with IL-ATPS methodology was developed for the extraction and purification of SDG from flaxseed. The IL [C4mim]N(CN)2 was screened as the extractant due to its exceptional extraction ability. The optimized conditions determined using RSM for IL-UAE were IL concentration 55.49% (w/w), liquid-solid ratio 24.50:1, and ultrasonic irradiation time 40 min. The maximum extraction efficiency (93.35%) of SDG in the IL-rich phase resulted when the ATPS was composed of 45.86% (w/w) [C4mim]N(CN)2 and 8.27% (w/w) Na2SO4 at 22 °C and pH 11.0. This combination of IL-UAE and IL-ATPS provides a simple, efficient, and environmentally friendly methodology for the extraction and purification of SDG from flaxseeds. This process is also expected to be useful for the extraction of other bioactive compounds.