Response surface optimization of hemp seed (Cannabis sativa L.) oil yield and oxidation stability by supercritical carbon dioxide extraction

Response surface optimization of hemp seed (Cannabis sativa L.) oil yield and oxidation stability by supercritical carbon dioxide extraction

J. of Supercritical Fluids 68 (2012) 45–51 Contents lists available at SciVerse ScienceDirect The Journal of Supercritical Fluids journal homepage: ...

963KB Sizes 268 Downloads 726 Views

J. of Supercritical Fluids 68 (2012) 45–51

Contents lists available at SciVerse ScienceDirect

The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu

Response surface optimization of hemp seed (Cannabis sativa L.) oil yield and oxidation stability by supercritical carbon dioxide extraction C. Da Porto a,∗ , D. Voinovich b , D. Decorti a , A. Natolino a a b

Department of Food Science, University of Udine, Italy Department of Chemical and Pharmaceutical Sciences, University of Trieste, Italy

a r t i c l e

i n f o

Article history: Received 25 February 2012 Received in revised form 12 April 2012 Accepted 13 April 2012 Keywords: Cannabis sativa L. Supercritical CO2 (SC-CO2 ) Oil yield Oxidation stability RSM

a b s t r a c t Hemp seed oil is considered one of the best nutritional oil for health. The present work is focused on the optimization of the hemp seed oil extractive process at laboratory level using supercritical carbon dioxide (SC-CO2 ) as solvent. Response surface methodology (RSM) was used to optimize hemp seed oil extraction yield and oxidation stability. Independent variables were operating temperature (40, 50 and 60 ◦ C), pressure (250, 300 and 350 bar) and particle diameter (0.59, 0.71 and 0.83 mm). A secondorder polynomial equation was used to express both the oil yield and the oil oxidation stability as a function of independent variables. The responses and variables were fitted well to each other by multiple regressions. The maximum oil yield, 21.50% w/w, was obtained when SC-CO2 extraction was carried out at 40 ◦ C, 300 bar and 0.71 mm of particle size. The maximum oil oxidation stability, 2.35 Eq ␣ toc/ml oil, was obtained at 60 ◦ C, 250 bar and 0.83 mm of particle size. A comparison between hemp seed oil composition extracted by SC-CO2 under the optimum operating conditions determined by RSM for oil yield and by organic solvent was reported. © 2012 Elsevier B.V. All rights reserved.

1. Introduction In hemp (Cannabis sativa L.) seed, the amount of oil is about 30% (w/w) depending on the variety, the year of cultivation, the climatic conditions and the location [1]. The fatty acid composition of hemp seed oil shows that it contains 70–80% polyunsaturated fatty acids (PUFA) with ∼10% saturated fatty acids. The two polyunsaturated essential fatty acids (EFAs), linoleic acid (LA, 18:2n−6) and ␣-linolenic acid (ALA, 18:3n−3), usually account for approximately 50–70% and 15–25% respectively, of the total seed fatty acid content. The ratio between n−6 and n−3 fatty acids is 3:1 [2–5]. Such balance has been claimed optimal for human nutrition and is apparently unique among the common plant oils [6]. The potential health benefits of these two polyunsaturated fatty acids (PUFA) are interesting owing to their anti-inflammatory, antithrombotic, antiarrhythmic and hypolipidemic properties [7]. Unlike most vegetable oils, hemp seed oil contains significant amounts of ␥-linolenic acid (GLA,18:3n−6) typically about 4%. Higher amounts of this fatty acid are synthesized in evening primrose (Oenothera macrocarpa) (∼8%) and borage (Borago officinalis) (∼20%) [8,9]. Positive effects of ␥-linolenic acid have been observed

∗ Corresponding author. Tel.: +39 0432 558141; fax: +39 0432 558120. E-mail address: [email protected] (C. Da Porto). 0896-8446/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.supflu.2012.04.008

on patients with rheumatoid arthritis, atopic dermatitis and allergies [10–12]. Due to the high content of polyunsaturated fatty acids, hemp seed oil is very susceptible to oxidative degradation. The oxidation stability of an oil is determined by its fatty acid composition in addition to several minor components that have antioxidant properties. In vegetable oils, tocopherols are the most important natural antioxidants present [13]. Oomah et al. [14] found in hemp seed about 800 mg/kg of oil tocopherols, mostly in the form of ␥-tocopherol (about 85%). ␥-Tocopherol has higher antioxidant activity than the form ␣ and ␤ and less than ␦ [15]. Currently hemp seed oil is obtained by solvent extraction or by cold-pressing. Solvent extraction is efficient and relatively inexpensive, however, it involves longer extraction time and the chemicals employed are generally hazardous to both workers and the environment. Additionally, the oil obtained by solvent extraction generally requires significant refining and solvent residues can remain in the final product. Screw pressing is relatively inexpensive, although it is not as efficient as solvent extraction. Matthäus and Brühl [5] reported that 60–80% of available oil can be extracted by a screw press from the hemp seeds, depending on the settings of the screw press. Latif and Anwar [16] reported that the oil contents (28.4–32.8%) obtained by enzyme-assisted cold-pressing of hemp seeds were found to be significantly higher than that determined for the control (26.7%). The problem of hemp seeds cold-processing is the high amounts of chlorophyll coextracted with the oil. This is due to the unripe seeds which are very present, since the most

46

C. Da Porto et al. / J. of Supercritical Fluids 68 (2012) 45–51

hemp cultivars are developed for the production of straw and fibre and are not optimized for the production of oil. Chlorophyll content requires to be reduced, however, this adds an additional step and cost. Moreover, unripe seeds contain higher amounts of moisture, which influences the moisture content of the whole material stored together and at least the taste and smell of the resulting oil [5]. Supercritical fluid extraction (SFE) has been widely employed as alternative of organic solvent extraction for vegetable oils extraction. The application of SFE has grown continuously because it showed several advantages over classical extraction processes with organic solvents. CO2 is probably the most widely used supercritical fluid because it is non-toxic, recyclable, cheap, relatively inert, non-flammable and easily separated from the extract Due to low critical temperature (31.1 ◦ C), CO2 is applied in SFE processes at near-environmental temperatures thus minimizing heat requirement and thermal damage to bioactive compounds [17–19]. Although SC-CO2 extraction of vegetable oils has been extensively studied in the laboratory, no studies have been reported on total oil yield and oxidation stability of hemp seed oil in SC-CO2 by response surface methodology to our knowledge. In a previous paper we established a preliminary set of supercritical CO2 extraction conditions to obtain high quality hemp seed oil [20]. Thus, the aim of this research was to ascertain how pressure, temperature and particle size influenced oil yield and quality of the extract (expressed here in terms of oxidation stability). 2. Materials and methods 2.1. Materials Hemp seeds were collected from experimental cultivation of hemp Felina cultivar (THC ≤ 0.2%) (Regulation EC n. 2860/2000) carried out at Prato Carnico (Udine, Italy). They were harvested during September 2011. Subsequently after drying, hemp seeds were stored at 4 ◦ C prior to extractions. The finale moisture content of the seeds was 9.8 ± 0.3%. 2.2. Solvents and reagents The carbon dioxide used (purity > 99.99%) was supplied by Rivoira Spa (Udine, Italy). All other solvents and reagents used in analytical determinations were Sigma–Aldrich Co. (Milan, Italy), pro analysis type. The chemicals used were of analytical reagent grade that include 1,1-diphenyl-2-picrylhydrazyl (DPPH – 90% purity, Sigma–Aldrich Co., Milano, Italy) and (+)-␣-tocopherol (Sigma–Aldrich Co., Milano, Italy). 2.3. Methods 2.3.1. Soxhlet extraction Thirty grams of hemp seeds were ground in a stainless steel blender, transferred into a filter paper extraction thimble and extracted with 240 ml n-hexane for 8 h at a maximum temperature of 70 ◦ C in a Soxhlet apparatus. After extraction was completed, nhexane was removed at 50 ◦ C under reduced pressure using a rotary evaporator (Rotavapor R210, Buchi, Flawil, Switzerland). Subsequently, the flask was placed into a desiccator chamber for 1 h. The oil obtained was weighed and the yield was calculated. Determination was done in triplicate. 2.3.2. Supercritical CO2 extraction Supercritical CO2 extractions were performed using a Lab Scale supercritical fluid system (M-LAB-SFE100; Tecnoprocess srl, Roma, Italy) equipped with a 100 cm3 extraction vessel (Fig. 1). For SC-CO2 extractions hemp seeds were ground in a stainless steel blender for 10, 30 and 60 s. The amount of ground hemp seeds placed in

Fig. 1. Scheme of the SC-CO2 laboratory unit for supercritical fluid extraction: (1) solvent cooler; (2) pump; (3) heater; (4) extractor; (5) separator. Table 1 Range and variables used for the experimental designs. Experimental variables Particle diameter, X1 (mm) Pressure, X2 (bar) Temperature, X3 (◦ C)

Experimental levels 0.59 250 40

0.71 300 50

0.83 350 60

the extractor was 15 g. Glass beads were placed on the bottom of the extractor, the ground hemp seeds were placed above them and another layer of glass beads was put at the top. After the extraction vessel was tightly sealed, the desired extraction temperature was set. Pressure within the extraction vessel was built up with a constant dioxide flow rate at 8 × 10−5 kg/s. The SFE extraction was initiated after the desirable temperature and pressure were achieved. The entire extraction process lasted 60 min and the collected oil was weighted. After extraction the exhausted matrix was characterized by size classification in a standard sifter with several mesh sizes (<0.25, 0.25–0.5, 0.8–1.0, 1.0–1.25, 1.25–1.50, 1.50–1.75, 1.75–2.0, >2.0 mm) [21]. Corresponding to the different grinding times 10, 30 and 60 s, the average particle diameters resulted of 0.59, 0.71 and 0.83 mm. These values were calculated by Sauter’s equation [22] applied to each set of fractions, within the previous mesh sized: mt dp =  k m /dpi i=1 i where mi is the mass of particles retained below mesh size dpi , mt is the total mass of milled seeds and k is the number of mesh sized. 2.3.3. GC analysis of fatty acids The fatty acid methyl esters (FAME) were prepared by transesterification of oil with 2 N KOH in methanol and n-hexane. Gas chromatographic (GC) analysis of FAME were performed in a Varian 3400 gas chromatograph equipped with a SP-2380 fused-silica column (Supelco, Bellafonte, PA) (30 m × 0.32 mm i.d., film thickness 0.20 ␮m), a split injector at 250 ◦ C; flame ionization detector at 260 ◦ C. Helium was used as carrier gas and the split ratio was used 1:50. The programmed temperature was: 2 min at 50 ◦ C, 50–250 ◦ C at 4 ◦ C/min. The identification of FAME was based on external standards using commercial reference compounds (Sigma Aldrich, Milan, Italy). Each result presents the mean and the standard deviation for a minimum of three analyses. 2.3.4. Oxidation stability The oxidation stability of the hemp seed oil samples was evaluated by the total free radical scavenger capacity (RSC) following the methodology described by Espín et al. [23] with slight modification. In brief, 10 ␮l of ethyl acetate sample solution at different concentrations was added with 1990 ␮l of fresh ethyl acetate DPPH solution (93 ␮M). Then the mixture was shaken vigorously and left in darkness for 60 min. Finally, the absorbance of the mixture was measured against pure ethyl acetate (blank) at 515 nm using

C. Da Porto et al. / J. of Supercritical Fluids 68 (2012) 45–51

47

Table 2 Analytical results of the extracts obtained from the 15 SFE experiments following the experimental designs. Exp. no.

X1 Dp (mm)

X2 P (bar)

X3 T (◦ C)

Total oil on seed weight (%)

Oxidation stability (RSC, ␮M)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

0.83 0.59 0.83 0.59 0.83 0.59 0.83 0.59 0.83 0.59 0.71 0.71 0.71 0.71 0.71

250 250 350 350 250 250 350 350 300 300 250 350 300 300 300

40 40 40 40 60 60 60 60 50 50 50 50 40 60 50

15.04 16.05 16.04 19.05 10.04 12.02 15.07 20.09 17.01 19.05 14.07 17.01 21.05 19.07 20.02

58.81 37.58 28.40 29.98 46.94 46.56 42.10 47.34 39.47 38.86 40.21 38.83 32.07 36.51 34.04

a UV–visible spectrophotometer (Shimadzu UV 1650, Italia srl). The RSC is the variation of the concentration of DPPH• free radical (CDPPH• ,i ) previously dissolved in ethyl acetate, after 60 min of reaction with the samples (CDPPH• ,f ): RSC = CDPPH• ,i − CDPPH• ,f The oxidation stability of the hemp seed oil samples was expressed both as RSC (␮M) and as ␣-tocopherol equivalents, the concentration of ␣-tocopherol solution which gives rise to the same RSC. A calibration curve was built using a series of tocopherols standard solutions in ethyl acetate. All determinations were done in triplicate.

Table 3 Estimation and significance degree of coefficients for oil extraction yield (%, w/w). Coefficient ˇ0 ˇ1 ˇ2 ˇ3 ˇ11 ˇ22 ˇ33 ˇ12 ˇ13 ˇ23 * **

2.3.5. Experimental design for response surface methodology (RSM) The effects of SC-CO2 extraction parameters particle diameter (Dp ), pressure (P) and temperature (◦ C) on the extraction oil yield (wt%) and the oil oxidation stability (RSC, ␮M) were determined by 33 full factorial designed. The levels of independent parameters (Table 1) were determined based on preliminary experiments. All experiments were carried out in a randomized order to minimize the effect of unexpected variability in the observed response due to extraneous factors. A second-order polynomial equation was used to express both the oil yield and the oil oxidation stability as a function of independent variables: Y = ˇ0 + ˇ1 X1 + ˇ2 X2 + ˇ3 X3 + ˇ11 X1 2 + ˇ22 X2 2 + ˇ33 X3 2 + ˇ12 X1 X2 + ˇ13 X1 X3 + ˇ23 X2 X3

(1)

where Y represents the response variable, ˇ0 is a constant, ˇi , ˇii and ˇij are the linear, quadratic and interactive coefficients, respectively. The coefficients of the response surface equation were determined by using Nemrodw software (LPRAI, Marseille, France). The goodness of fit of the model was evaluated by the coefficient of determination R2 and the analysis of variance (ANOVA). 3. Results and discussion 3.1. Response surface methodology analysis: hemp seed oil yield The results of the RSM analysis carried out as shown in Section 2.3.5 are shown in Table 2. Experimental hemp seed oil yields were used to determine the coefficients of the response surface equation (Eq. (1)). Estimated coefficients are given in Table 3.

***

Estimate 19.68 −1.36 2.04 −1.04 −1.24 −3.64 1.06 −0.67 −0.35 1.25

Standard error

Significance

0.38 0.22 0.22 0.22 0.44 0.44 0.44 0.25 0.25 0.25

<0.01*** 0.175** 0.0267*** 0.562** 3.72* 0.0426*** 6.02 4.30* 22.01 0.414**

˛ < 0.05. ˛ < 0.01. ˛ < 0.001.

Obtained second-order polynomial equation (Eq. (2)) was found well to represent the experimental data (R2 = 0.982). Yield (g oil/100 g seed) = 19.68 − 1.36X1 + 2.04X2 − 1.04X3 − 1.24X1 2 − 3.64X2 2 + 1.06X3 2 − 0.67X1 X2 − 0.35X1 X3 + 1.25X2 X3

(2)

In Eq. (2), X1 is the particle size, X2 is the pressure, X3 is the temperature. The best way of expressing the effect of any parameter on the yield within the experimental space under investigation was to generate response surface plots of the equation. Fig. 2 illustrates the response surface (a) and contour plots (b) for the influence of particle diameter and pressure on the yield of oil for a fixed 50 ◦ C temperature. As shown in Fig. 2, the results graphically revealed the nature of the fitted surface as a maximum oil yield (∼20%, w/w) obtained with increasing pressure from 300 to 350 bar and decreasing particle size from 0.71 to 0.59 mm. At high pressures the solubility of the oil increased due to the increase in density of CO2 , leading to greater oil solubility in CO2 [17,24]. However, a negative quadratic effect at high pressure was highlighted (Table 3) probably due to the fact that the highly compressed CO2 facilitates solute–solvent repulsion [25]. For this reason, high pressure is not always recommended [26]. As expected, the extraction yield of hemp seed oil increased with decreasing particle size because grinding process not only increased the interfacial area but also released oil from the broken cells [27,28]. Fig. 3 illustrates the response surface (a) and contour plots (b) depicting the influence of temperature and pressure on the oil extraction yield at fixed 0.71 mm particle diameter. As shown in Fig. 3, the results revealed

48

C. Da Porto et al. / J. of Supercritical Fluids 68 (2012) 45–51

Fig. 2. Response surface (a) and contour plots (b) for the oil yield as related to particle diameter (Dp ) and pressure at a fixed 50 ◦ C temperature.

Fig. 3. Response surface (a) and contour plots (b) depicting the influence of temperature on the oil extraction yield at fixed 0.71 mm particle diameter.

the nature of the fitted surface as a saddle displaying two maximum oil yield (20% w/w) between 300 and 350 bar pressure and temperature ranging from 40 to 60 ◦ C. Instead, at pressure between 250 and 280 bar, at constant pressure, oil yield decreased significantly with increasing temperature. Such behaviour suggests that in the experimental range of temperature analysed (40–60 ◦ C), at higher pressures (300–350 bar) the effect of the increased vapour pressure of oil compounds on oil solubility is more important than that of the drop in CO2 density whilst at lower pressures (250–280 bar) the drop of CO2 density is dominant. Özkal et al. [28] reported similar results for SC-CO2 extraction of apricot kernel oil.

Table 4 Estimation and significance degree of coefficients for oil oxidation stability (RSC, ␮M).

3.2. Response surface methodology analysis: hemp seed oil oxidation stability Experimental hemp seed oil oxidation stability was used to determine the coefficients of the response surface equation. Estimated coefficients are given in Table 4. Obtained second-order polynomial equation (Eq. (3)) was found well to represent the experimental data (R2 = 0.909). Yield (RSC, ␮M) = 34.993 − 1.54X1 − 4.345X2 + 3.261X3 + 3.933X1 2 +4.288X2 2 − 0.942X3 2 − 3.554X1 X2 − 3.064X1 X3 + 4.244X2 X3

In Eq. (2), X1 is the particle size, X2 is the pressure, X3 is the temperature. It seems that the pressure and the temperature of experiments are the most decisive working conditions in the oil oxidation stability. Fig. 4 shows the response surface (a) and contour plots (b) depicting the influence of pressure and temperature on the oil oxidation stability at fixed 0.71 mm particle diameter. It can be observed that at a fixed particle diameter the oxidation stability of hemp seed oil peaked at 250 bar to 44 ␮M RSC (1.76 Eq ␣ toc/ml oil) and a further increase of pressure led to

(3)

Coefficient

Estimate

Standard error

Significance

ˇ0 ˇ1 ˇ2 ˇ3 ˇ11 ˇ22 ˇ33 ˇ12 ˇ13 ˇ23

34.993 1.540 −4.345 3.261 3.933 4.288 −0.942 −3.554 −3.064 4.244

2.125 1.250 1.250 1.250 2.465 2.465 2.465 1.398 1.398 1.398

<0.01*** 27.3 1.77* 4.77* 17.01 14.02 71.8 5.02 8.00 2.89*

* ˛ < 0.05. ** ˛ < 0.01. *** ˛ < 0.001.

C. Da Porto et al. / J. of Supercritical Fluids 68 (2012) 45–51

49

Fig. 4. Response surface (a) and contour plots (b) depicting the influence of pressure and temperature on the oil extraction yield at fixed 0.71 mm particle diameter.

Fig. 5. Response surface (a) and contour plots (b) depicting the influence of pressure and particle diameter on the oil extraction yield at fixed 50 ◦ C temperature.

a decrease in oxidation stability to 26 ␮M RSC (1.04 Eq ␣ toc/ml oil). These results can be explained by considering the extraction of hemp seed oil antioxidants (mostly tocopherols) related to the extraction capabilities of other matrix compounds. With high extraction pressures antioxidants concentration decreased probably because of higher solubility for other diluting materials. Others authors observed similar behaviour in the supercritical extraction of tocopherols from different matrices [29–31]. At pressure lower than 300 bar, an increase in temperature does not increase the oxidation stability of oil significantly. This fact indicates that antioxidants are easily extractable compounds, even at low temperature. Fig. 5 illustrates the response surface (a) and contour plots (b) for the influence of particle diameter and pressure on the oxidation stability of oil for at fixed 50 ◦ C temperature. As shown in Fig. 4, the results graphically revealed the nature of the fitted surface as a minimum of oxidation stability at 34 ␮M RSC (1.36 Eq ␣ toc/ml oil) between 300 and 350 bar pressure, confirming the results previously reported for pressure. The oxidation stability increased with increasing particle diameter. This is probably due to antioxidant compounds degradation during sample preparation. de Lucas et al. [31] observed a similar behaviour in supercritical extraction of tocopherols from olive tree leaves. Fig. 6 shows the response surface (a) and contour plots (b) for the influence of temperature and particle diameter on the oxidation stability of oil for at fixed 300 bar pressure. At constant pressure, an increase in temperature slightly increase the oxidation stability of

oil. This could indicate that antioxidants in hemp seed oil are easily extractable compounds, even at low temperature. 3.3. Extraction kinetic, fatty acid composition and oxidation stability of hemp seed oil obtained under the optimum SC-CO2 operating conditions determined by RSM for oil yield For the optimum SC-CO2 operating conditions determined by means of experimental design to obtain high oil yield 300 bar pressure, 40 ◦ C temperature and at 0.71 mm particle size, the extraction kinetic of hemp seed oil was studied and its fatty acid composition, oxidation stability and yield compared with Soxhlet data. Fig. 7 shows the extraction kinetic of hemp seed oil recovered by SC-CO2 . Almost 73% of total oil was recovered in the first 60 min of extraction whilst the remaining 27%, subsequently. As reported by Sovovà [32], the first part of extraction kinetic represents a linear period where the slope of this line as it is related to the extract solubility, only depends on pressure and temperature. In the following stage of extraction, intra-particle diffusion controlled oil transfer from the interior to the surface of the particles. In Table 5 the fatty acid profiles and yields of hemp seed oils obtained by SC-CO2 extraction and by n-hexane extraction are reported. As regard the fatty acid profiles, it can be observed that no significant difference was highlighted. Others authors obtained similar results from different vegetable oils [33,34]. Hemp seed oil is characterized by an interesting fatty acid composition with

50

C. Da Porto et al. / J. of Supercritical Fluids 68 (2012) 45–51

Fig. 6. Response surface (a) and contour plots (b) for the influence of temperature and particle diameter on the oxidation stability of oil for at fixed 300 bar pressure. Table 6 Comparison of hemp seed oil oxidation stability between SC-CO2 extraction and Soxhlet using n-hexane. Sample

RSC (␮M)

Oxidation stability (Eq ␣ toc/ml oil)

SC-CO2 Soxhlet (n-hexane)

46.7 ± 6.1 aa 21.0 ± 4.2 c

1.87 ± 5.6 a 0.84 ± 1.2 c

Vergin olive oil

20.4 ± 5.6

0.82 ± 2.2

Each data represents the mean of three replicates ± standard deviation. a Values with different letter within columns indicate significant differences (p < 0.05).

Fig. 7. Extraction kinetic of hemp seed oil recovered by SC-CO2 at 300 bar pressure, 40 ◦ C temperature and at 0.71 mm particle size.

a high content of polyunsaturated fatty acids (81%). Linoleic acid is the predominant fatty acid (59%), which comes, together with ␣linolenic acid (18%), to approximately 77% of the total fatty acids. The high ratio of polyunsaturated to saturated fatty acids and the Table 5 Fatty acid composition of extract obtained by SC-CO2 at 300 bar pressure, 40 ◦ C temperature and at 0.71 mm particle size and by Soxhlet using n-hexane. SC-CO2 extraction

Soxhlet

Fatty acid composition (%) Palmitic acid (C16:0) Stearic acid (C18:0) Oleic acid (C18:1) Linoleic acid (C18:2ω6) ␥-Linolenic acid (C18:3ω6) ␣-Linolenic acid (C18:3ω3) Eicosenoic acid (C20:1) Behenic acid (C22:0)

5.85 ± 0.06 1.45 ± 0.04 ba 10.67 ± 0.14 b 59.21 ± 0.70 3.40 ± 0.09 18.47 ± 0.63 0.12 ± 0.06 b 0.84 ± 0.01

5.37 ± 0.13 1.56 ± 0.05 a 11.51 ± 0.15 a 59.16 ± 0.55 3.48 ± 0.05 17.96 ± 0.23 0.18 ± 0.03 a 0.80 ± 0.01

EFAs sum ω−6/ω−3 ratio

77.68 3.21

77.12 3.29

PUFAs sum Monounsaturated Saturated Polyunsaturated/saturated ratio

81.08 10.78 8.14 9.96

80.60 11.69 7.73 10.42

Each data represents the mean of three SFE extraction replicates ± standard deviation. a Values with different letter within rows indicate significant differences (p < 0.05).

optimal ratio of ω−6 to ω−3 of oil confirmed that it is very interesting from a healthy point of view [9–11]. The fatty acid composition of hemp seed oil from Felina cultivated at Prato Carnico (Udine, Italy) was similar to that reported by Callaway et al. [3] for the French cultivar Futura-77. Although the extraction with n-hexane gave the highest oil yield (30% (w/w)), this organic solvent is usually non-selective and causes the simultaneous removal of non-volatile pigments and waxes contaminated with solvent residues. For this reason, the oil obtained by solvent extraction generally required refining, which, however, decreases the health value of the product. Using SC-CO2 extraction the oil yield is lower (21.50%(w/w)) but its quality is very high. Hemp seed oil for its potential health benefits is a special oil and its corresponding higher value make its extraction using SC-CO2 an economically viable option. Table 6 shows the oxidation stability, expressed both in terms of RSC (␮M) and ␣-tocopherol equivalents for ml of oil, of the hemp seed oil samples obtained by SC-CO2 extraction and Soxhlet. As shown in Table 6 hemp seed oil extracted by supercritical CO2 exhibits the highest value of RSC (46.7 ± 3.1 ␮M) corresponding to 1.87 ␣-tocopherol equivalents/ml oil, about two-fold higher than vergin olive oil (20.4 ± 5.6).

4. Conclusions It was not possible to determine optimum co-extraction conditions for hemp seed oil yield and oxidation stability, under the specific conditions used. In conclusion, the use of supercritical CO2 in hemp seed oil extraction resulted optimized for oil yield at a temperature of 40 ◦ C, a pressure of 300 bar and particle size of 0.71 mm. Instead, the optimum oxidation stability of oil could be achieved at a temperature of 60◦ C, a pressure of 250 bar and 0.83 mm particle size.

C. Da Porto et al. / J. of Supercritical Fluids 68 (2012) 45–51

Acknowledgement Financial support of this work by Friuli Venezia Giulia Region through grants (art. 17/LR 26/2005) is gratefully acknowledged. References [1] C. Leizer, D. Ribnicky, A. Poulev, S. Dushenkov, I. Raskin, The composition of hemp seed oil and its potential as an important source of nutrition, J. Nutraceuticals, Functional & Medical Foods 4 (2000) 35–53. [2] J.L. Deferne, D.W. Pate, Hemp seed oil: a source of valuable essential fatty acids, J. International Hemp Association 3 (1996) 4–7. [3] J.C. Callaway, T. Tennilä, D.W. Pate, Occurrence of omega−3 stearidonic acid (cis-6,9,12,15-octadecatetraenoic acid) in hemp (Cannabis sativa L.) seed, J. International Hemp Association 3 (1996) 61–63. [4] U. Kriese, E. Schumann, W.E. Weber, M. Beyer, L. Brühl, B. Matthäus, Oil content, tocopherol composition and fatty acid patterns of the seeds of 51 Cannabis sativa L. genotypes, Euphytica 137 (2004) 339–351. [5] B. Matthäus, L. Brühl, Virgin hemp seed oil: an interesting niche product, European J. Lipid Science and Technology 110 (2008) 655–661. [6] A.P. Simopoulos, The importance of the ratio of omega−6/omega−3 essential fatty acids, Biomedicine & Pharmacotheray 56 (2002) 365–379. [7] K. Jones, Nutritional and Medicinal Guide to Hemp Seed, Rainforest Botanical Laboratory, Gibsons, BC, Canada, 1995. [8] M. Pina, J. Graille, P. Grignac, A. Lacombe, O. Quenot, P. Garnier, Recherche d’oenothères riches en acide gammalinolénique, Oléagineux 39 (1984) 594–596. [9] S.P.J.N. Senanayake, F. Shahidi, Lipid components of borage (Borago officinalis L.) seeds and their changes during germination, J. American Oil Chemists’ Society 77 (2000) 55–61. [10] P.H Remans, J.K. Sont, L.W. Wagenaar, W. Wouters-Wesseling, W.M. Zuijderduin, A. Jongma, F.C. Breedveld, J.M. Van Laar, Nutrient supplementation with polyunsaturated fatty acids and micronutrients in rheumatoid arthritis: clinical and biochemical effects, European J. Clinical Nutrition 58 (2004) 839–845. [11] S. Senapati, S. Banerjee, D.N. Gangopadhyay, Evening primrose oil is effective in atopic dermatitis: a randomized placebo-controlled trial, Indian J. Dermatology Venereology and Leprology 74 (2008) 447–452. [12] K. Wakai, K. Okamoto, A. Tamakoshi, Y. Lin, T. Nakayama, Y. Ohno, Seasonal allergic rhinoconjunctivitis and fatty acid intake: a cross-sectional study in Japan, Annals of Epidemiology 11 (2001) 59–64. [13] A. Kamal-Eldin, L.A. Appelqvis, The chemistry and antioxidant properties of tocopherols and tocotrienols, Lipids 31 (1996) 671–701. [14] B.D. Oomah, M. Busson, V. David, D.V. Godfrey, C.G. John, J.C.G. Drover, Characteristics of hemp (Cannabis sativa L.) seed oil, Food Chemistry 76 (2002) 33–43. [15] A. Lampi, A. Hopia, V. Piironen, Antioxidant activity of minor amounts of ␥tocopherol in natural triacylglycerols, J. American Oil Chemists Society 74 (1997) 549–555.

51

[16] S. Latif, F. Anwar, Physicochemical studies of hemp (Cannabis sativa) seed oil using enzyme-assisted cold-pressing, European Journal of Lipid Science and Technology 111 (2009) 1042–1048. [17] G. Brunner, Gas extraction: an introduction to fundamentals of supercritical fluids and the application to separation processes, in: H. Baumgurtel, E.U. Franck (Eds.), Topic in Physical Chemistry, Springer, New York, 1994. [18] Q. Lang, C.M. Wai, Supercritical fluid extraction in herbal and natural product studies – a practical review, Talanta 53 (2001) 771–782. [19] E. Reverchon, I. De Marco, Supercritical fluid extraction and fractionation of natural matter, J. Supercritical Fluids 38 (2006) 146–166. [20] C. Da Porto, D. Decorti, D.F. Tubaro, Fatty acid composition and oxidation stability of hemp (Cannabis sativa L.) seed oil extracted by supercritical carbon dioxide, Industrial Crops and Products 36 (2012) 401–404. [21] L. Fiori, Grape seed oil supercritical extraction kinetic and solubility data: critical approach and modeling, Journal of Supercritical Fluids 43 (2007) 43–54. [22] N.P Povh, M.O.M. Marques, M.A.A. Meireles, Supercritical CO2 extraction of essential oil and oleoresin from chamomile (Chamomilla recutetia L. Rauschert), J. Supercritica Fluids 21 (2001) 245–256. [23] J.C. Espín, C. Soler-Rivas, H.J. Wichers, Characterization of the total free radical scavenger capacity of vegetable oils and oil fractions using 2,2-diphenyl-1picrylhydrazyl radical, J. Agricultural Food Chemistry 48 (2000) 648–656. [24] A. Molero Gomez, E. Martinez de la Ossa, Quality of borage seed oil extracted by liquid and supercritical carbon dioxide, Chemical Engineering J. 88 (2002) 103–109. [25] S. Liu, F. Yang, C. Zhang, H. Ji, P. Hong, C. Deng, Optimization of process parameters for supercritical carbon dioxide extraction of Passiflora seed oil by response surface methodology, J. Supercritical Fluids 48 (2009) 9–14. [26] Y. Yamini, M. Khajeh, E. Ghasemi, M. Mirza, K. Javidnia, Comparison of essential oil compositions of Salvia mirzayanii obtained by supercritical carbon dioxide extraction and hydrodistillation methods, Food Chemistry 108 (2008) 341–346. [27] C. Marrone, M. Paletto, E. Revenchon, A. Stassi, Almond oil extraction by supercritical CO2 : experiments and modeling, Chemical Engineering Science 53 (1998) 3711–3718. [28] S.G. Özkal, M.E. Yener, L. Bayind, Mass transfer modeling of apricot kernel oil extraction with supercritical carbon dioxide, J. Supercritical Fluids 35 (2005) 119–127. [29] A. Birtigh, M. Johannsen, G. Brunner, N. Nair, Supercritical-fluid extraction of oil-palm components, J. Supercritical Fluids 8 (1995) 46–50. [30] J.W. King, O.F. Favati, S.L. Taylor, Tocopherol concentrates production by supercritical fluid extraction and chromatography, Separation Science and Technology 31 (1996) 1843–1857. [31] A. de Lucas, E. Martinez de la Ossa, J. Rincón, M.A. Blanco, I.I. Gracia, Supercritical fluid extraction of tocopherol concentrates from olive tree leaves, J. Supercritical Fluids 22 (2002) 221–228. [32] H. Sovovà, Rate of the vegetable oil extraction with supercritical CO2 . Modelling of extraction curves, Chemical Engineering Science 49 (1994) 409–415. [33] M.G. Bernardo-Gil, L.M.C. Lopes, Supercritical fluid extraction of Cucurbita ficifolia seed oil, European Food Research Technology 219 (2004) 593–597. [34] C. Devittori, D. Gumy, A. Kusy, L. Colarow, C. Bertoli, P.P. Lambelet, Supercritical fluid extraction of oil from millet bran, J. American Oil Chemists Society 77 (2000) 573–579.