Supercritical Carbon Dioxide Extraction of Seed Oil from Winter Melon (Benincasa hispida) and Its Antioxidant Activity and Fatty Acid Composition

In the present study, supercritical carbon dioxide (SC-CO2) extraction of seed oil from winter melon (Benincasa hispida) was investigated. The effects of process variables namely pressure (150–300 bar), temperature (40–50 °C) and dynamic extraction time (60–120 min) on crude extraction yield (CEY) were studied through response surface methodology (RSM). The SC-CO2 extraction process was modified using ethanol (99.9%) as co-solvent. Perturbation plot revealed the significant effect of all process variables on the CEY. A central composite design (CCD) was used to optimize the process conditions to achieve maximum CEY. The optimum conditions were 244 bar pressure, 46 °C temperature and 97 min dynamic extraction time. Under these optimal conditions, the CEY was predicted to be 176.30 mg-extract/g-dried sample. The validation experiment results agreed with the predicted value. The antioxidant activity and fatty acid composition of crude oil obtained under optimized conditions were determined and compared with published results using Soxhlet extraction (SE) and ultrasound assisted extraction (UAE). It was found that the antioxidant activity of the extract obtained by SC-CO2 extraction was strongly higher than those obtained by SE and UAE. Identification of fatty acid composition using gas chromatography (GC) showed that all the extracts were rich in unsaturated fatty acids with the most being linoleic acid. In contrast, the amount of saturated fatty acids extracted by SE was higher than that extracted under optimized SC-CO2 extraction conditions.

allows extraction at low temperatures and relatively low pressures. In addition, CO 2 can be immediately evaporated when exposed to atmospheric conditions, thus, the extract obtained is free from chemicals and thermal degradation compounds [16]. In fact, SC-CO 2 extracts are generally recognized as safe (GRAS) to be used in food products [17]. Furthermore, CO 2 is a non polar solvent therefore, adding a small amount of polar solvents as co-solvent can enhance greatly the extraction efficiency of polar compounds. Among different solvents ethanol (EtOH) is the most used solvent due to its good miscibility with CO 2 , non-toxicity and allowed use in the food and pharmaceutical industries [16]. Previously, SC-CO 2 has been used to extract valuable compounds from different raw materials such as hazelnut [18], grape seed [19], watermelon [20], orange pomace [21], peach (Prunus persica) seed oil [22], Bidens pilosa Linné [23] and Mitragyna speciosa leaves [24]. Herrero et al. [25] presented an overview on the recent advances and applications of SFE. Furthermore, SFE has been proposed for separation of antioxidant compounds from sage, rosemary leaves and some Brazilian plants [26][27][28], but to the best of our knowledge no extraction of Benincasa hispida by SC-CO 2 has been reported.
Optimization of the experimental conditions is a critical step to develop a successful process due to the effect of various variables including pressure, temperature and extraction time on the efficiency of SC-CO 2 extraction. Over the past several years, response surface methodology (RSM) has been used in many areas such as foods, chemicals, or biological processes. This experimental methodology combines mathematics with statistics for generating a mathematical model to describe process, examine the relationship between one or more response variables and a set of quantitative experimental variables and optimizing the process, while at the same time reducing the number of experimental trials [14,29]. Nowadays, RSM is widely used to optimize the conditions of SFE method [11,14]. The objectives of this study were to evaluate the effect process variables including pressure, temperature and dynamic extraction time on crude extract yield (CEY). RSM was employed to optimize extraction conditions in order to obtain the maximum CEY. Furthermore, the antioxidant activity and fatty acid composition of the extract obtained under optimized conditions were determined and then compared with those obtained by the Soxhlet and UAE method.

Effect of the Process Variables on Crude Extraction Yield
The effects of pressure, temperature and dynamic extraction time on the CEY were evaluated. The individual effect of independent variables including pressure (A), temperature (B) and dynamic extraction time (C) was found by perturbation plot. A perturbation plot does not show the effect of interactions and it is like one factor-at-a-time experimentation. The perturbation plot helps to compare the effect of all independent variables at a particular point in the design space. The response is plotted by changing only one factor over its range while holding the other factors constant. A steep slope or curvature in a factor shows that the response is sensitive to that factor. A relatively flat line shows insensitivity to change in that particular factor [30]. Therefore, it was revealed that CEY is more sensitive to pressure and temperature than dynamic extraction time. The perturbation plot for the CEY is shown in Figure 1.

Optimization of SC-CO 2 Extraction Conditions and Verification of the Model
The experimental design and related results are presented in Table 1. The CEY ranged from 119.11 to 173.33 mg-extract/g-dried sample. A second-order polynomial model for predicting CEY was rendered by multiple linear regression analysis of the experimental data. Estimated regression equation coefficients and statistical significance of linear, quadratic and interaction of process variables along with the corresponding R 2 , adjusted-R 2 and lack of fit test for the final reduced model are presented in Table 2. Probability values (p-value) revealed that all linear, quadratic and interaction terms had significant (p < 0.05) effect on the CEY. The p-value of final reduced model was less that 0.05 which confirmed that model fitness was significant (Table 2). Furthermore, the lack of fit was non-significant (p > 0.05), indicating that the model could adequately fit the experiment. The developed final reduced model in term of CEY was obtained as follows: where X 1 , X 2 and X 3 are the coded variables for pressure, temperature and dynamic extraction time, respectively.   Figure 2 indicates that the predicted values were very close to the actual values. Indeed, the high value of coefficient of determination, R 2 (0.998) and adjusted-R 2 (0.997) confirmed the adequate accuracy of the polynomial reduced model. The small E value (0.35%) suggested that the obtained model was acceptable. Moreover, the coefficient of variation (C.V.% = 0.630) indicated that the final reduced model was reproducible.
The best way to visualize the influence of the process variables on the response is to generate three dimensional surface response plots [30]. Response surface plots are illustrated in Figure 3a-c. Figure 3a illustrates the effect of pressure and temperature on the CEY at a fixed dynamic extraction time of 90 min. It is clear that there is an optimal value for pressure to obtain the highest CEY. At a fixed temperature level, the CEY increased with increasing pressure up to a certain value (around 250 bar), which was due to the increased SC-CO 2 density at higher pressure [15]. It was known that as pressure increased the extraction rate also increased due to the changes in solubility of solutes in solvent [18][19][20][21][22][23][24]. However, in the current study, further increase in the pressure level (above 250 bar) led to an unexpected reduction in the CEY. Same unexpected behavior was also observed by Rezaei and Temelli [31] which they stated it could be related to the reduced diffusion rates of the solutes from the plant matrix to the SC-CO 2 medium. Another possible explanation for this unexpected reduction could be compressing of seeds at higher pressure. Luengthanaphol et al. [32] reported that increasing pressure compresses the tamarind seed coats and caused reducing ability for solvent used to diffuse inside the seed particles. The solubility of the solute is likely to decrease, keep constant, or increase with the temperature rise at constant pressure, which depends on the solvent density and the solute vapor pressure [33][34][35][36][37]. As presented in Figure 3a,c, an increase in CEY was observed with enhancing of the temperature in an early stage of extraction, because the change of solvent density is less effective than that of solute vapor pressure, but the trend was reversed when the temperature reached a certain value under the given range of pressure range for this work.
Similar results were also reported for the extraction of tea (Camellia sinensis L.) seed oil [34] and extraction of lipids by SC-CO 2 extraction from grain sorghum [38]. Since the solubility of the solutes depends on the balance between fluid density and solute vapor pressure, both controlled by fluid pressure and temperature, an extraction temperature and pressure in the range from 40 to 47 °C, 150 bar to 270 bar, respectively, is practical during the SC-CO 2 extraction of Benincasa hispida seed oil.
Dynamic extraction time had a positive effect on the CEY of Benincasa hispida seed oil. As shown in Figure 3b,c the CEY was increased with dynamic extraction time. It was clear that there is an optimal level for the dynamic extraction time. It should be mentioned that due to the physical structure of the seeds, the penetration of the solvent and the diffusion of the unreleased oil in the particles are very slow. Therefore, dynamic extraction time is limited to the fast extraction period since the amount of oil recovered in the slow extraction period is negligible. However, the duration of fast extraction period is affected by temperature and pressure. In this study, above the moderated extraction pressure ■ Experimental points (225 bar) and temperature (45 °C), by increasing dynamic extraction time beyond 100 min the CEY did not changes significantly (p > 0.05). Same result was found by Xu et al. [30] investigating the effect of extraction time on SC-CO 2 extraction of sea buckthorn (Hippophae thamnoides L.) oil. The regression equation obtained in this study was used to maximize the CEY from Benincasa hispida seed as the desired optimum condition of the extraction process. In this regard, a numerical optimization was performed through desirability function method. The optimal conditions for the desired goal were predicted as 244 bar pressure, 46 °C temperature and 97 min dynamic extraction time. The SC-CO 2 extraction of Benincasa hispida seed under the optimal conditions was performed in triplicate as a rechecking experiment and then the obtained value were compared with the predicted result. The mean value of 175.60 mg-extract/g-dried sample obtained from the real experiment was very close to the predicted value (176.30 mg-extract/g-dried sample). This finding indicates the validity and adequacy of the rendered model to reflect the expected optimization.

Antioxidant Activity of Extracted Crude Oil
The antioxidant activity of Benincasa hispida seed oil was determined through the DPPH˙ and ABTS˙+ radicals scavenging assays. The results obtained under optimized condition of SC-CO 2 extraction compared with the antioxidant activity of catechin and BHT as a synthetic antioxidant Figure 4a,b. The antioxidant activity of Benincasa hispida seeds extract was lower compared with those of synthetic antioxidants at the same concentration (0.1 mg/mL). As reported by Liu et al. [34] the antioxidant activity values for seeds oil from Opuntia dillenii Haw ranged from 90 to 17% when the concentrations varied from 4 to 0.1 mg/mL, accordingly. Therefore, it was expected that this weak point could be addressed by using higher concentration of extracts which is of great interest and need to be further investigated.

Identification and Quantification of Fatty Acid Composition
The fatty acid composition of a kind of extracts is its most useful chemical feature. Many of the chemical tests for extracts identity or purity can be related to their fatty acid content [34]. The fatty acid composition of extract obtained using optimized SC-CO 2 extraction is presented in Table 3. The total saturated fatty acid (SFA) content (myristic acid, palmitic acid and stearic acid) was 15.97% of total fatty acid, and the palmitic acid (9.83 ± 0.33% of total fatty acid) content was higher comparatively. However, the total unsaturated fatty acid (UFA) (palmitoleic acid, oleic acid, linoleic acid and linolenic acid) content was 83.53% of total fatty acids. The major UFA was linoleic acid (18:2) which constitutes 67.17% of total fatty acid. α-Linolenic acid, which is a kind of omega-3 fatty acid, was detected in extract obtained by the optimized SC-CO 2 extraction method. The presence of essential fatty acids (EFAs), namely linoleic acid (C18:2) and α-linolenic acid (C18: 3) in Benincasa hispida seed oil demonstrated its importance in the human body. These two EFAs are necessary to support normal cell functions, promote disease prevention and may even be useful for disease treatment [39].

Comparison with Other Extraction Techniques
Different extraction processes have different extraction yield and efficiencies. Higher concentration of valuable compounds in the extracts is an important aspect in the production of natural products while a primary task in the industries is lowering economic cost which can be achieved by better extraction yield [11]. As shown in Table 3, the results of this study were compared with those obtained and published with UAE [40] and SE [41]. According to the results, SE resulted in the highest value of CEY (250.00 ± 1.30 mg-extract/g-dried sample) comparing to the UAE and optimized SC-CO 2 extraction. The efficiency of UAE and optimized SC-CO 2 extraction techniques for recovery of CEY from Benincasa hispida seeds was 43% and 70% of those found with SE, respectively, but SC-CO 2 extracts were found to possess better quality in terms of antioxidant activity compared to those obtained with UAE and SE processes.
Natural bioactive compounds which are contributed to the antioxidant activity of extracts are thermo sensitive and may decompose at higher temperature (78 °C) for long extraction time (6 h) which was applied for SE. UAE is one of the simplest extraction techniques because it is easy to operate in common laboratory equipment but its efficiency in terms of CEY was lower than SE and optimized SC-CO 2 extraction. SC-CO 2 extraction has different advantages over SE such as low operating temperature, thus no thermal degradation of most of the labile compounds, shorter extraction duration and high selectivity in the extraction of target compounds. SC-CO 2 extraction also, seems to be a cost-effective process at laboratory scale, but a precise economic evaluation will need additional experiments for establishing large-scale units [42]. As shown in Table 3 the fatty acid contents of the extracts from different extraction methods (SE, UAE and SC-CO 2 ) were different, however, the difference between UAE and SC-CO 2 extracts was minor. The SFAs extracted by SE (24.3% of total fatty acid) was higher than those extracted by optimized SC-CO 2 extraction (15.97% of total fatty acids) and UAE (17.80% of total fatty acids). In a study which conducted by Rajaei et al. [43] they obtained same results about the SC-CO 2 extraction of tea seeds. They found that the UFAs extracted by SC-CO 2 extraction were higher than SE. They pointed out that higher temperatures appear to be unfavorable to the extraction of UFAs. It was obtained that in all extracts obtained with different extraction process the linoleic acid (C18:2) was dominant (around 60.60-67.17% of total fatty acids). Furthermore, essential fatty acids (including 18:2 and 18:3) contributed to high percentage (around 68%) of total fatty acids which confirmed the importance of the extracts.

Material and Reagents
Whole winter melons (Benincasa hispida L.) were purchased from a local market in Serdang, Selangor, Malaysia. Fruits were chosen at commercial maturity according to their similarity of color, size and absence of surface defects. The fruits were cut, and the seeds were separated manually and washed under tap water. Seeds were dried at 40 °C in a ventilated oven (1350FX, Cornelius, OR, USA) for 24 h and then stored at an ambient temperature in the dark. The seeds were ground in a grinder mill (MX-335, Panasonic, Shah Alam, Malaysia) for 10 s to produce a powder with an approximate size of 1.5-2.5 mm.

Supercritical Carbon Dioxide Extraction
Supercritical carbon dioxide (SC-CO 2 ) extraction was carried out in a batch system using a supercritical fluid apparatus (ABRP200, Pittsburgh, PA, USA) including 500 mL extraction vessel, automated back pressure regulator (model BPR-A-200B), high pressure pump (model P-50) . Figure 5 displayed a scheme of the supercritical fluid extractor (500 mL sample capacity). The flow rate of CO 2 , the extraction temperature and pressure were adjusted by the ICE software, and the extraction time was measured by a stopwatch. Samples of ground Benincasa hispida seeds (40 g) and glass beads (120 g with 2.0 mm in diameter) were mixed and then put into an extraction vessel. The application of some rigid materials such as glass bead with the ground sample is useful for maintaining a proper CO 2 flow rate in the extraction vessel and maintained the desired permissibility of the particles during extraction process [15]. Before pressurization, the SFE system was allowed to reach the desire operating temperature. Then the liquefied CO 2 was transferred to the extraction vessel by a high pressure pump (P-50, Thar Designs, Inc., Pittsburgh, PA, USA). In this work, the pressure, temperature and dynamic extraction time of the SC-CO 2 extraction were varied from 150 to 300 bar, from 40-50 °C and from 60 to 120 min, respectively. The supercritical CO 2 flow rate was maintained at 10 g/min based on preliminary experiments. A higher yield can be obtained by adding small amount of co-solvent. In all experimental conditions, absolute ethanol (EtOH) was applied as the co-solvent due to its good miscibility with CO 2 , non-toxicity and allowed use in the food and pharmaceutical industries. The co-solvent flow rate was maintained at 1 g/min. After reaching the desired pressure and temperature, the Benincasa hispida seeds soaked in the solvent and co-solvent for 30 min (static extraction) in order to equilibrate the mixture at desire pressure and temperature. The static extraction time was applied at desire pressure and temperature for each conducted run. During the dynamic extraction time, CO 2 carrying the released solutes flowed out of the unit and the extract was collected into a pre-weighed collection flask. The procedure was performed in duplicate.

Crude Extraction Yield Measurement
The extracts were weighed gravimetrically using a Mettler Toledo analytical balance (±0.0001 g) (Mettler Toledo GmbH, Greinfensee, Switzerland) and then the CEY was calculated according to the following equation:

1000
(2) where m e is the crude extract mass (g) and m s is the extracted sample mass (g). The measurement was performed in triplicate and the mean values of CEY were expressed as mg-extract/g-dried sample.

Determination of Radical Scavenging Activity
The extracts of Benincasa hispida seeds were subjected to antioxidant activity analysis using DPPH˙ and ABTS˙+ free radical scavenging assays. All determinations were done in triplicate and expressed as means ± Standard Deviation.

Determination of DPPH˙ Radical Scavenging Activity
This assay was carried out as described by Zengin et al. [44] with some modifications. A total of 0.1 mg/mL of the extracts and synthetic antioxidant (catechin) in the ethanol were added into an ethanolic solution of DPPH˙ (3 mL, 6 × 10 −5 M). The mixture was vortexed for 20 s at room temperature. Absorbance measurements at 515 nm commenced immediately in a 1 cm quartz cell after 1 min up to 60 min with 10 min intervals using a UV-260 visible recording spectrophotometer (Thermo 4001/4 UV-Vis Spectrophotometer, Thermo Fisher Scientific, West Palm Beach, FL, USA). The blank test was conducted with 0.1 mL ethanol instead of extracts and the absorbance was recorded as A blank . The inhibition percent of DPPH˙ which was scavenged (% DPPH sc ) was calculated according to the following equation: where A blank and A sample are the absorbance values of the blank and of the tested samples, respectively.

Determination of ABTS˙+ Radical Scavenging Activity
The 2,2'-azinobis(3-ethylbenzothiazoline-6-sulphonic acid) diammonium salt (ABTS˙+) assay was carried out according to the method of Cai et al. [45]. The ABTS˙+ radical solution was prepared by mixing 7 mM ABTS and 2.45 mM potassium persulphate, and incubating the mixture in the dark at room temperature for 16 h. The ABTS˙+ solution was then diluted with 80% (v/v) ethanol to obtain an absorbance of 0.70 at 734 nm. ABTS˙+ solution (3.9 mL) was added to sample (0.1 mg/mL) and mixed vigorously. The absorbance of the mixtures at room temperature was recorded immediately using UV-260 visible recording spectrophotometer (Thermo 4001/4 UV-Vis Spectrophotometer) at 734 nm for 10 min at 2 min intervals. The blank test was conducted with ethanol instead of extracts and the absorbance was recorded as A blank . The inhibition percent of ABTS˙+ which was scavenged (%ABTS sc ) was calculated using the following equation: %ABTS sc = (A blank -A sample ) × 100/A blank (4) where A blank and A sample are the absorbance values of the blank and of the tested samples, respectively.

Preparation of Fatty Acid Methyl Esters
Samples were brought to a temperature of 50-60 °C and homogenized thoroughly before taking a test sample in order to obtain the fatty acid methyl esters (FAMEs). An aliquot of the test sample (100 µL) was mixed with hexane (1 mL) in a 2 mL vial. An aliquot of sodium methoxide (1 µL, 1% w/v) was added to the vial which was mixed vigorously using a vortex mixer. The mixture first became clear and then turbid as sodium glyceroxide was precipitated. After a few minutes, the clear upper layer of methyl ester was pipetted off and injected into the gas chromatograph (GC) for further analysis.

Gas Chromatography Analysis
Fatty acids composition analysis was performed in a Hewlett-Packard 6890 gas chromatograph (Wilmington, DE, USA), equipped with a flame ionization detector (FID) and a BPX70 (30 m × 0.25 mm × 0.25 µm, Victoria, Australia) GC column. Oven temperature was programmed isothermally to 115 °C during 2 min, then was raised at 4 °C/min to 163 °C and then at 1 °C/min to 170 °C. Finally, temperature increased to 200 °C at 10 °C/min and held at this temperature for 2 min. Helium was used as a carrier gas which flowed at a rate of 1 mL/min. The injection volume was 1 µL. Standard methyl esters of fatty acids were used as authentic samples. The fatty acids determination was accomplished by comparing it with standards and was valued by the area percentage of each fatty acid. The fatty acid determination was performed in triplicate for each sample and expressed as means ± Standard Deviation.

Experimental Design and Statistical Analysis
Response surface methodology (RSM) was applied to optimize the process variables including pressure (150-300 bar), temperature (40-50 °C) and dynamic extraction time (60-120 min) to achieve the highest amount of crude oil from Benincasa hispida seeds. A central composite design (CCD) with axial points was used for designing the experimental data. This generated 20 treatments with six replications at the centre points to estimate the repeatability of the method ( Table 1). The effect of unexplained variability induced by extraneous factors on the observed responses was minimized by randomizing the order of experiments. Blocks are assumed to have no impact on the nature and shape of the response surface. The following second-order polynomial model was fitted to the data: Y i = β 0 + β 1 X 1 + β 2 X 2 + β 3 X 3 + β 11 X 1 2 + β 22 X 2 2 + β 33 X 3 2 + β 12 X 1 X 2 + β 13 X 1 X 3 + β 23 X 2 X 3 (5) where Y i is predicted response, β 0 is offset term, β 1 , β 2 and β 3 are the regression coefficients for linear effect terms, β 11 , β 22 and β 33 are quadratic effects and β 12 , β 13 and β 23 are interaction effects. In this model, X 1 , X 2 and X 3 represent power level, temperature and sonication time, respectively. The significant terms (p < 0.05) in the model were found by analysis of variance (ANOVA) based on p-value. The terms statistically found non-significant (p > 0.05) were dropped from the initial model and the experimental data was refitted only to significant (p < 0.05) variables in order to obtain the final reduced model [46]. The three-dimensional response surface plot was generated for the graphical interpretation of the interaction effect of independent variables on the response. Numerical optimization was carried out to predict the exact optimum level of independent variables leading to the desirable response goal. The model adequacy was determined using model analysis, lack of fit test, coefficient of determination (R 2 ) and adjusted-R 2 . Furthermore, experimental data were compared with predicted values (method validation) in order to verify the adequacy of final reduced model. In addition, the quality of fit between the experimental and predicted data was determined according to value of the mean relative deviation modulus (E). The criteria can be calculated as follows: E (%) = ∑ (6) where V exp and V pre are the experimental and predicted values, respectively, n is the number of experimental data. A model is considered acceptable if E value is less than 10% [47]. The experimental design matrix, data analysis, regression coefficients, generation of 3D graph and numerical optimization procedure were created using Design Expert Version 8.0.7 software (Stat-Ease Inc., Minneapolis, MA, USA, trial version).

Conclusions
In this study, Benincasa hispida seed oil has been extracted using SC-CO 2 under various operating conditions. The effects of pressure, temperature and dynamic extraction time on the crude extraction yield were studied. It was found that all process variables had positive significant effect on the response follows the order of pressure > temperature > dynamic extraction time. Response surface methodology with a central composite design (CCD) was successfully applied for optimization of SC-CO 2 extraction process. The empirical polynomial equation has been proposed to predict optimum operating conditions of process. The highest crude extraction yield of 175.60 (mg-extract/g-dried sample) was obtained when the process was carried out at 244 bar of pressure, 46 °C of temperature and dynamic extraction time of 97 min. Under these optimized conditions, the experimental crude extraction yield agreed closely with the predicted yield. The antioxidant activities of seed oil obtained by optimized SC-CO 2 extraction was compared with those obtained and published using SE and UEA techniques. It was found that SC-CO 2 extract was substantially better than extract obtained by SE and UEA in term of reducing ability. Moreover, the crude extraction yield obtained with SC-CO 2 extraction was significantly higher than that reported by using UAE. It was found that the extract obtained by SC-CO 2 extraction is a rich source of linoleic acid (18:2) compared with that obtained by SE, so it can provide new natural products to the food industry. Thus, it can be concluded that Benincasa hispida seed is a potential source of valuable compounds and the green SC-CO 2 extraction technique is an effective and promising process to obtain Benincasa hispida seeds oil with strong antioxidant activity and also rich in essential fatty acids.