Optimization of Oil Extraction from Nigella sativa Seeds by Pressing Using Response Surface Methodology

MATERIALS AND METHODS

Materials: The Nigella sativa seeds were purchased from local market. The seeds have the following characteristics: Small, black, flat, funnel-shaped, 0.2 cm long and 0.1 cm wide.

Analytical methods: All analytical determinations were performed in triplicate for each sample with the standard deviation.

Sample preparation for oil extraction by hydraulic press: Oil extraction is carried out from Nigella sativa seeds using hydraulic press described by Acheheb et al. (2012). The ground sample, having a particle size less than 1 mm, is compressed, during 60 min, at various pressures (50, 90 and 120 bars), temperatures (25, 40 and 60°C) and thickness cake (1.30±0.047, 2.80±0.190 and 4.04±0.28 cm). The oil yield is the ratio between the mass of the oil extracted and mass of the sample (g 100 g^-1 dry weight).

Determination of chemical and botanical properties of Nigella sativa seeds: The moisture content was determined using Infra-humid meter (Sartorius). Crude proteins, lipids and ash were determined by the method described by AFNOR (2000). Carbohydrates were evaluated by difference using the following equation:

100-(weight in gram (proteins+fat+water+ash) in 100 g of food

(FAO, 1998). The observation of transverse section of Nigella sativa seed was carried out using the photonic microscope.

Determination of physicochemical characteristics of Nigella sativa seeds oil
Oil content: Oil is extracted from 10 g of ground seeds in a Soxhlet extractor using hexane as a solvent. The result is expressed as the percentage of lipids in the dry matter of seeds.

Determination of fatty acids composition: The methyl esters were prepared according to AFNOR (2000) and analysed using a Thermo Finnigan Gas Chromatograph equipped with flame ionisation detector. The separation of fatty acid methyl ester was carried out at 190°C on a 30x0.32 mm capillary packed with the polar stationary phase Nona Fluoro Pentanoic Acid (NFPA), with nitrogen as a carrier gas. The temperature of detector was 230°C and the injection block was recorded as 210°C. Peaks were identified by comparing the retention times with those of a mixture of standard methyl esters. Peroxidizability Index (PI) was evaluated using the equation of Song et al. (2000):

(1)

Experimental design: Response surface methodology was used to optimize the oil extraction from Nigella sativa seeds by hydraulic press. The effects of independent variables, temperature (25-60°C), pressure (50-120 bars) and thickness cake (1.30-4.04 cm) on oil extraction were studied using the Box-Behnken design. The coded and uncoded levels of different process variables are indicated in Table 1.

The second response surface model used to fit the experimental data has the following form:

(2)

where, Y is the response (oil yield in%), β₀, β_i, β_ii and β_ij are constant coefficients of intercept, linear, quadratic and interaction terms, respectively xi and xj are coded independent variables (Montgomery, 2005; Bradley, 2007). Analysis was conducted using Design-Expert^® v. 8.3.1. The quality of the fitted model was evaluated by the analysis of variance (ANOVA).

RESULTS AND DISCUSSIONS

Botanical characteristics of Nigella sativa seeds: Figure 1 shows a transverse section of Nigella sativa seed. The seed is covered externally by an outer integument. The outer integument is followed by 2-4 layers of parenchymatous cells with a thick wall and by a pigmented layer (inner integuments) filled with brown compounds (oil). Below the inner integument, there is a cuticle followed by an endosperm consisting of rectangular to polygonal cells with thick wall filled with globules oil. These observations agree with those reported by Esau (1965). The layer of globules oil of the seed coat is separated from the endosperm by a cuticle. This observation allows concluding that oil of Nigella sativa seed is located in the cotyledons and integuments.

Table 1:	Coded and uncoded levels of different process variables used in Box-Behnken design

Fig. 1:

Transverse section of Nigella sativa seed, A: Oil of seed coat, B: Endosperm and C: Globules oil

Table 2:	Physicochemical properties of Nigella sativa seeds (g 100 g-1 dry weight)

Table 3:	Fatty-acids composition of the oil Nigella sativa seeds (Unit: relative peak area%)

Chemical proprieties of Nigella sativa seeds: Table 2 shows the chemical composition of Nigella Sativa seeds. Moisture content of whole seeds was 9.61±0.14 (g 100 g^-1 on dry weight). Oil contents were 40.37±1.70 g 100 g^-1 on dry weight, this reflects the importance of using such seeds for oil production. Crude protein level was 20.69±0.84 g 100 g^-1 on dry weight, show that the Nigella sativa seeds have an important nutritional value. The other components were in the following decreased order, Carbohydrates total 25.29 g 100 g^-1 on dry weight and ash content 4.04±0.29 g 100 g^-1 on dry weight. The carbohydrate content indicates that Nigella sativa seeds can be a source of energy.

Fatty acids: The fatty acids composition of oil Nigella sativa seeds and their percentages are presented, in order of their elution in the column, in Table 3. Palmitic, stearic and behenic acids were the major satured acids, whereas linoleic and oleic were the major unsatured acids. The content of linoleic acid was much higher than that of oleic acid. The presence of fairly high amounts of the essential fatty acid linoleic, suggests that the Nigella sativa seeds oil is highly nutritious. Linoleic acid is also a precursor of arachidonic acid, another essential fatty acid (Ali et al., 2012). However, (Cater and Denke, 2001) have reported that behenic acid, despite its low bioavailability, is considered as a factor in the increase of cholesterol in humans.

The results of this study revealed also that the oil Nigella sativa seeds contains 1.722% of eicosenoic acid (gadoleic acid, C20:1) but not eicosadienoic acid (C20:2). These results are in accordance with those of Ustun et al. (1990). On the other hand, Aitzetmuller et al. (1997) and Matthaus and Ozcan (2011) have reported that these fatty acids (C20:1 and C20:2) could be considered as chematoxonomic characteristics of Nigella and as an identity criterium of Nigella sativa seeds oil. The values of peroxidizability index and of unsaturated/saturated ratio which are, respectively 55.45% and 4.33, show that the crude oil extracted is stable to auto-oxidation rancidity during storage.

Response surface analysis: The seventeen generated experiments with the values of various responses to different experimental combination for coded variables are given in Table 4. A large variation in the oil yield, from 3.81-22.64 g 100 g^-1 on dry weight was observed for different experimental combinations. The experiments were conducted in accordance with the Box-Behnken design to find the optimal combination of pressure, temperature and thickness of cake for maximum oil yield.

The results of analysis of variance carried out to estimate the quality of the fitted second order response surface model are shown in Table 5. The sign and magnitude of the coefficients allows interpreting the effect of the variable on the response. The negative sign of coefficient indicates that when the level of the variable increases the response decreases while the positive sign indicates an increase in the response. The model F-value of 61.36 reveals that the model is significant. Values of "Prob>F" less than 0.0500 indicate that model terms are significant. In this study A, B, C, AB, A², C² are significant model terms.The high coefficient of determination (R²) which is of 0.987 shows that the fit of the model is good. The value 8.88% of coefficient of variance (CV) indicates that the deviations between experimental and predicted values are low. Figure 2 confirms the good agreement between the experimental and predicted values for oil yield from Nigella sativa seeds. In conclusion the final model for the response variable oil yield is as following:

(3)

Response surfaces and contour plots: Results of analysis of variance have shown that the linear term of temperature (A), pressure (B) and thickness cake (C) had a significant effect on the oil extraction yield (Table 5). The effects of linear terms are followed by those of quadratic term of temperature (A²), thickness cake (C²) and interaction between temperature and pressure (AB).

Table 4:	Experimental conditions and observed response values of BBD
T (A): Temperature, P (B): Pressure, TC (C): Thickness cake

Fig. 2:

Predicted vs. observed values for extraction yield of oil from Nigella sativa seed

Fig. 3(a-c):

Surface plot and contour of oil yield from Nigella sativa seeds as a function of (a) Temperature and pressure, (b) Contour plot of temperature and pressure and (c) Thickness cake and pressure

The effect of independent variables (temperature, pressure, cake thickness) on the dependent variable (oil yield) is indicated by the response surfaces plots developed from Eq. 3 (Fig. 3a-c).

Table 5:	Analysis of variance (ANOVA) for the response surface quadratic model for oil yield from Nigella sativa seeds using hydraulic press

Figure 3a shows that the oil yield increased with increase in temperature and in pressure. Figure 3c shows that the oil yields decrease with increase of thickness cake. The contour plots can spot the optimum response approximately. The interactions of temperature-pressure have positive effect on oil yield. The optimum value response (>20%) was observed for the high level of independent variables and visualized in Fig. 3b at the corner of the top and right side.

CONCLUSION

This study has shown that the Response Surface Methodology is a effective way to determine the optimal conditions for oil extraction from Nigella Sativa seeds by pressing. Analysis of variance has shown that the effects of all the process variables including temperature, pressure and thickness cake were statistically significant. Polynomial model was obtained for predicting oil yield. The optimal conditions for maximum oil yield (22.27 g 100 g^-1 on dry weight) correspond to temperature of 60°C, pressure of 120 bars and thickness cake of 1. 30 cm.