Open Access
Issue
OCL
Volume 26, 2019
Article Number 16
Number of page(s) 9
Section Technology
DOI https://doi.org/10.1051/ocl/2019009
Published online 05 April 2019

© F. Mas’ud et al., Published by EDP Sciences, 2019

Licence Creative Commons
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1 Introduction

RBO is edible oil extracted from rice bran which is a by-product of rice mills. RBO is superior among the other edible oils because it contains unique antioxidants and nutraceutical complexes present in its composition (Oliveira et al., 2012). In the unsaponifiable fraction of RBO contains γ-oryzanol and tocols. These compounds have been reported in the scientific literature as powerful antioxidant agents that are effective for preventing degenerative diseases (Lerma-García et al., 2009). Seema (2015) suggests that bran antioxidants are mainly γ-oryzanol and vitamin E, as well as unsaturated fats capable of lowering cholesterol. γ-oryzanol component of RBO was first presumed to be a single component, but later it was determined to be a fraction containing ferulate (4-hydroxy-3-methoxy cinnamic acid) esters of triterpene alcohols and plant sterols (Rogers et al., 1993). Cycloartenyl ferulate, 24-methylenecycloartanyl ferulate, and campesteryl ferulate are the three major components of γ-oryzanol (Xu et al., 2001).

On the other hand, the most important contribution of vegetable oils is their tocopherol content, which is generally collectively referred to as “vitamin E”. Vegetable oils contain high concentrations of vitamin E (Bauernfeind and Desai, 1977) and can provide the most daily requirement of vitamin E (Desai et al., 1980). High vitamin E levels in rice bran oils are reported to have antihypocholesterolemic, anticancer, and neuroprotective properties. Tocols are capable of reducing lipid peroxidation and lipid risk factors, in this case increasing LDL cholesterol and platelet aggregation, exhibiting anti-inflammatory properties, demonstrating anti-carcinogenic and cardiovascular protection effects (Tiwari and Cummins, 2009).

Some scientific reports on the benefits of RBO for health have led this study to evaluate the optimization of the extraction process of RBO. The direct solvent extraction method, which does not require specific extraction instrumentation, has been most commonly used (Chen and Bergman, 2005), and ethanol has gained attention as a potential solvent for vegetable oils (Rodrigues and Oliveira, 2010). Study on sesame oil revealed that the polar solvent such as ethanol was a good solvent compared to non-polar solvents. According to Péres et al. (2006), these results could be explained by the interaction between the unsaturated fatty acids with a polar solvent, compared with non-polar solvents. In fact, the oils extracted with ethanol presented the typical composition of RBO (Firestone, 1999). Ethanol has attractive advantages to use as a solvent, including low toxicity, good operational security, as well as being obtained from a bio-renewable source (Bessa et al., 2017). The choice of ethanol as a solvent is deemed necessary to implement the concepts and principles of green extraction.

According to Chemat et al. (2012), green extraction is a new concept to protect both the environment and consumers and in the enhanced competition of industries to be more ecologic, economic, and innovative. Within the green extraction approach, the concept of the green extract is an extract obtained in such a way to have the lowest possible impact on the environment. Because of environmental concerns, a suitable solvent which gives less impact to the environment is more preferred nowadays. Further, according to Tekin et al. (2018), ethanol has been widely applied as a viable solvent due to their ease of recovery and low cost in an application and is classified as an environmentally friendly green solvent. Although alcohols such as ethanol, methanol and isopropyl alcohol have similar solvent properties, ethanol has become the foremost among others because of its non-toxic nature. Li et al. (2016) stated that the green solvents have several benefits such as biodegradability, low toxicity, non-flammability, and renewability making them potential candidates in separation science.

Many studies have proven the technical feasibility of employing ethanol in the process of extraction oils (Saxena et al., 2011). On oil extraction from sunflower collets, ethanol gave a higher yield of extracted material, the content of oil phase was similar to that obtained when n-hexane is used. When ethanol was used, about 70% less crystallizable waxes and about 38% more tocopherols and phospholipids were extracted, this shows the feasibility of using ethanol as an alternative solvent to hexane in extracting oil (Baümler et al., 2016), including RBO (Oliveira et al., 2012; Rodrigues and Oliveira, 2010). Previously, Imsanguan et al. (2008) have reported that ethanol was a better solvent for γ-oryzanol extraction compared to hexane. It can be explained by the relatively high polarity of the γ-oryzanol molecule (consisting of triterpene alcohols and phytosterols esterified with ferulic acid), where the polarity of the solvent may significantly affect the extractability of γ-oryzanol (Xu and Godber, 2001).

Based on the results of the existing study on the RBO, and that the optimization of the extraction process is very much related to production costs, as well as considering the needs of the RBO extraction industry for the pharmaceutical and food industries related to the optimization of the extraction process, then this research was conducted. To the best of the author’s knowledge, this is the first study, RBO extraction uses ethanol and monitors the extraction time, ethanol concentration, and volume of ethanol on oil yield, γ-oryzanol, and vitamin E of RBO in the maceration method accompanied by stirring, and applies RSM to optimize the RBO extraction process, standardize, and analyze the resulting model.

2 Materials and methods

Ciliwung rice, local rice of Indonesia, as samples were obtained from milling rice grain in a local grinding mill Makassar Indonesia during March to April in 2017. Ethanol (wt.%) purchased from a local chemical shop, γ-oryzanol standard from Sigma-Aldrich Co and α-tocopherol standard from Sigma, St. Louis, USA. Ethyl acetate, methanol, and chloroform from Merck, Germany.

2.1 Preparation of rice bran and extraction oil

Freshly milled bran samples were directly collected from the milling system in polyethylene bags, the rice bran was screened through a 60-mesh sieve to have a uniform particle size and stabilized at autoclave (Hiclave HV-85 Hirayama) at 100 °C for 15 min for inactivating endogenous lipase. For extraction oil, each experimental unit weighed 50 g of rice bran in the Erlenmeyer 1000 mL, RBO was extracted using maceration method with stirring under the 30 °C of room temperature, and pulp was separated by centrifugation (refrigerated AX-521 centrifuge) at a speed of 3500 rpm for 20 min. The liquid part was accommodated in the evaporator flask. Then, the solvent was removed in a Buchi R-215 rotary evaporator equipped V-700 vacuum Pomp speed of 60 rpm, the heating temperature was of 35 °C, and the evaporation temperature was of 21 °C. RBO was packaged in a dark glass bottle and stored in a freezer before analysis. The percentage of oil yield was calculated as follows (Sani, 2014): (1)

2.2 Analysis of γ-oryzanol and vitamin E

Preparation of γ-oryzanol standard: 0.05 g of the pure γ-oryzanol was dissolved in ethyl acetate in a 100 mL flask, and diluted to respectively 100, 200, 300, 400, and 500 ppm at a 50 mL flask, homogenized by vortex, inserted in the vial GC-MS (gas chromatography-mass spectrometry). Preparation of vitamin E standard: 0.05 g of the pure vitamin E was dissolved in methanol-chloroform (1:1) in a 100 mL flasks, and diluted to respectively 100, 200, 300, 400, and 500 mg.L−1 at a 50 mL flask, homogenized by vortex, inserted in the vial GC-MS. Preparation of sample of RBO for γ-oryzanol analysis: 0.012 g RBO dissolved in 2 mL ethyl acetate, homogenized by vortex and inserted into the vial bottle GC-MS. Preparation of sample of RBO for vitamin E analysis: 0.012 g RBO dissolved in 2 mL methanol-chloroform (1:1), homogenized by vortex and inserted into the vial bottle GC-MS.

Quantification of γ-oryzanol and vitamin E of RBO: γ-oryzanol and vitamin E of RBO were performed on a GC-MS QP2010 by Shimadzu equipped with a split/splitless injector. Separations were achieved using a Rxi SH-5Sil MS capillary column (30 m, 0.25 mm ID, 0.25 μm film thickness). Helium was used as the carrier gas at flow rates of 14.0 mL/min and a split ratio of 1:10. The oven temperature was programmed at 110 °C for a hold of 2 min and increased to 200 °C at a rate of 10 °C/min and hold at the final temperature for 9 min. LabSolution software was used to control the operation of GC-MS. MS spectra were obtained at range width m/z 40–450, interface temperature 280 °C, and ion source temperature 200 °C. γ-oryzanol and vitamin E of RBO peaks were identified by comparing their retention time and equivalent chain length with respect to the standards.

2.3 Experimental design for optimization and statistical analysis

RSM was employed to optimize the parameters of time, ethanol concentration, and ethanol volume for oil yield, γ-oryzanol, and vitamin E maximum of RBO on extraction process by maceration method accompanied by stirring. RSM provides the optimal conditions of a process, using multivariate statistical techniques to obtain responses from the observed process variables. RSM consists of mathematical and statistical techniques based on polynomial equations that best suit experimental data to describe the behavior of data sets with the aim of obtaining statistical predictive values (Rafi et al., 2015). It is a rapid analytical approach and more economical (Banga and Tripathi, 2009). Nowadays, the procedures based on the statistical evaluation using RSM are widely used to determine the interaction between the factors that influence reactions for optimization.

The CCD is one of the most effective RSM designs, which are widely used in the study and optimization of the linear, quadratic, and interaction effects of variables on observed responses (Soundararajan et al., 2016). The CCD is still the symmetrical second order experimental design most utilized for the development of analytical procedures (Bezerra et al., 2008). Three independent variables examined in this study were time (h) (x1), ethanol concentration (%) (x2), and ethanol volume (mL) (x3), while the oil yield, γ-oryzanol, and vitamin E concentration were considered as response variables. The value of independent variables was determined based on the results of the preliminary study. The middle values for the independent variables were 5 h, 86%, and 600 mL, respectively. The lower and upper limits for each treatment were 4 and 6 h for time, 76 and 96% for ethanol concentrations, 500 and 700 mL for ethanol volume (Tab. 1). The CCD consists of 14 experimental points and 6 replications of the center point. Replication of the center point was aims to evaluate the pure error variance as the experimental error and to control the adequacy of the model. To estimate the coefficients of the response function and predict the system’s responses, analysis of the experimental results of CCD was realized using empirical second-order polynomial equations as follows: (Amiri et al., 2018; Keshtegar et al., 2018). (2) where, Y is a response, β0 denotes the constant coefficient; Xi and Xj the independent factors, βi, βii, and βij the regression coefficients for the linear, quadratic, and interaction effects, respectively; k the number of variables; and e stands for the statistical error occurring to response Y (Awad et al., 2017).

Statistic software Design-Expert 10 was used to design, analyze, and optimize experimental models. Analysis of variance (ANOVA) was used to validate the statistical significance of the parameters that influence the responses, and the quality of the predicted model (Chan et al., 2017; Mohammed et al., 2018). The coefficient of determination (R2) shows the total predictive performance of the model, it represented the validity and fit quality of the model’s quadratic polynomials. R2 values are close to 1, indicating a reasonable adjustment of the model to experimental data. A p-value of ≤ 0.05 at a confidence level of 95% and a F-value of the lack-of-fit test were used for statistical analysis to evaluate the significance of model statistics, at a confidence level of 95% and a F-value of the lack-of-fit test were used for statistical analysis to evaluate the significance of model statistics (Tan et al., 2017).

Table 1

Five levels of independent variables of central composite design.

3 Results and discussion

Extraction of edible oil is generally a multi-parameter process so that the optimization of the process conditions represents a critical step in the development of the model. In the present study, RSM has been used as a tool to develop the model of RBO extraction apply maceration method with stirring to see the effect of x1, x2, and x3 on the oil yield (%), γ-oryzanol (mg.L−1), and vitamin E (mg.L−1) of RBO. The effects of x1, x2, and x3 on the yield of oil (%), γ-oryzanol (mg.L−1), and vitamin E (mg.L−1) of RBO have been studied during experimentation. The data results of 20 runs using a CCD showed that oil yield, γ-oryzanol, and vitamin E concentration of RBO ranged from 10.02 to 15.29%, 450.27 to 900.013 mg.L−1, and 120.51 to 129.83 mg.L−1, respectively (Tab. 2). The maximum of the oil yield, γ-oryzanol, and vitamin E concentration of RBO has been achieved at the optimum conditions of x1 = 5.30 h, x2 = 89.21% and x3 = 686.66 mL (50 g rice bran), respectively. Maximum oil yield, γ-oryzanol, and vitamin E concentration of RBO under these conditions were 14.47%, 783.65 mg.L−1, and 127.01 mg.L−1, respectively. The experimental data were fitted with the second-order equation model suggested by Design Expert 10 software.

Table 2

Central composite design, showing coded values of independent variables, with observed of yield, γ-oryzanol, and vitamin E of RBO.

3.1 Model fitting

The quality of the model developed was evaluated based on the correlation coefficient value. According to the result of ANOVA for oil yield, γ-oryzanol, and vitamin E of RBO, the R2 values of oil yield, γ-oryzanol, and vitamin E were 0.89, 0.83, and 0.84, respectively, indicating that the models adequately represented the real relationship between the parameters chosen. According to Singh et al. (2018), when R2 is more than 80%, the regression model shows good fit. The significance of different terms of each coefficient was determined using the F-value and p-value. According to Yolmeh et al. (2014), a large F-value and a small p-value would imply a more significant effect on the corresponding response variable. According to Li et al. (2014), the model is important and can be used to navigate the design domain. The model will be significant at a 95% confidence interval if the F test has a p-value of less than 0.05. In the case of lack-of-fit (p > F), the p-value is greater than 0.05 which shows the failure of the model in finding data points in the experimental domain. The reduced quadratic model equation developed from the experimental data to predict the oil yield, γ-oryzanol, and vitamin E of RBO in terms of coded factors are given in equations (35), respectively. (3) (4) (5) where: x1 = time; x2 = ethanol concentration; x3 = ethanol volume.

3.2 interpretation result of ANOVA

3.2.1 Extraction yield of RBO

The result of ANOVA for response surface reduced quadratic model of the oil yield showed that the model F-value of 9.99 implies the model is significant. Values of Prob > F of 0.0003 (less than 0.05) indicate model terms are significant. Values of Prob > F less than 0.05 indicate variable terms are significant, in this case, the linear term of ethanol concentration (x2), the linear term of ethanol volume (x3), the quadratic terms of extraction time (x12), the interaction between extraction time and ethanol concentration (x1x2), and the interaction between ethanol concentration and ethanol volume (x2x3) are significant variables terms. On the contrary, values Prob > F greater than 0.1000 indicate the variable terms are not significant, in this case, the linear term of extraction time (x1), the quadratic terms of ethanol concentration (x22), the quadratic terms of ethanol volume (x32), and interaction between extraction time and ethanol volume (x1x3) did not give any significant contribution on oil yield.

The lack-of-fit F-value of 1.69 implies the lack-of-fit is not significant relative to the pure error. Non-significant lack-of-fit is good, which indicates that the model is suitable to describe the effect of variables for the oil yield and that the developed model was adequate for predicting the response. According to Bas and Boyaci, (2007), the model will be considered appropriate if the lack-of-fit value model is not significantly different at the level of specific α. Further, according to Bezerra et al. (2008), a model will be well fitted to the experimental data if it presents a significant regression and a non-significant lack-of-fit. These values would give a relatively good fit to the mathematic model in terms of coded factors are given in equation (3).

In this study obtained a yield of RBO was 14.47% in agreement with the previous study. Anwar et al. (2005) has reported that the rice bran contained 15–20% oil, and 10–26% according to Pourali et al. (2009). The simplest method of solvent extraction is the single contact batch operation, where the solid to be leached and the solvent is mixed and the extract solution and raffinate solid phases are separated (Bessa et al., 2017). Considering the time effect, it greatly affects the yield, especially on the value of the mass transfer. The longer the contact time between the solvent with the solute during the extraction process, the more the number of elements extracted chemical content. The more time is given for contact of the sample with the solvent, the higher the extraction yield percentage (Elkhaleefa and Shigidi, 2015). The effect of solvent concentration on the yield was also explained by Chen et al. (2016), that the extraction yield was increased with the increase of ethanol concentration in the extraction solvent.

The effect of volume of solvent is consistent with mass transfer principles. The driving force during mass transfer is the concentration gradient between the solid and the bulk of the liquid, which is greater when a higher solvent to solid ratio is used. Distribution of solvent to solids will greatly affect oil yield, the ratio solids with a solvent will affect the oil yield. The amount of solvent affects the wide contact between solids and solvents. The more solvent, the greater the contact area, so the distribution of the solvent will be even greater. Equitable distribution of solvent to solids increases the oil yields, the amount of solvent will reduce the saturation level of the solvent so that the desired component will be extracted perfectly. In general, the yield increased with increasing volume of the solvent used. The more the volume of solvent used, the greater the ability of a solvent to extract oil contained in the material. The increasing volume of solvent also leads to increasing contact time, which occurs between materials with a solvent.

In the extraction process, an optimum point will be reached. Mas’ud et al. (2017) conducted a study of the effects of extraction time, temperature, and volume of solvent in mango seed kernel oil extraction above room temperature, they reported that the variables had a significant effect on oil yield. Effect of the combination of reactions between variable time and temperature at high-level experimental process will obviously result in a decrease in oil yield. Similar phenomena have also been reported by Oniya et al. (2017) and Pichai and Krit (2015). A decrease in oil yield at high temperatures and long periods of time in oil extraction is thought to be a result of degradations of some oil components such as phenolic compounds degradations as reported by Chew et al. (2011). They conducted a study of the effects of ethanol concentration, extraction time, and extraction temperature on the recovery of phenolic compounds and antioxidant capacity, they were reported that the extraction time of 120 min was achieved the maximum concentration of phenolic compounds, and after this point, total phenolic content and thick tannin content were decreased. According to Chirinos et al. (2007), oxidation of phenolic compounds due to excessive oxygen exposure can occur if the extraction time is longer.

3.2.2 γ-Oryzanol concentration of RBO

The Model F-value of 4.62 implies the model is significant. Values of Prob > F of 0.0106 (less than 0.05) indicate model terms are significant. Variables that have a significant effect on γ-oryzanol concentration were the quadratic terms of ethanol concentration (x22) and the quadratic terms of ethanol volume (x32). Values greater than 0.1000 indicate the model terms are not significant so that the linear term of extraction time (x1), ethanol concentration (x2), and ethanol volume (x3), the quadratic term of extraction time (x12), the interaction between time and ethanol concentration (x1x2), the interaction between time and ethanol volume (x1x3), and the interaction between ethanol concentration and ethanol volume (x2x3) did not give any significant contribution on γ-oryzanol concentration. The reduced quadratic model equation in terms of coded factors was developed from the experimental data to predict the γ-oryzanol of RBO produced from ANOVA as given in equation (4). The coefficient values of variables x1, x2, and x3 still appear in the model because the Design Expert 10 software work system used was hierarchical, so that if the quadratic effect is significant then the linear effect will also appear in the model even though the linear effect is not significant based on the ANOVA results. The lack-of-fit F-value of 2.90 implies the lack-of-fit is not significant relative to the pure error. This indicates that the model is suitable to describe the effect of a parameter observed on γ-oryzanol and that the developed model is adequate for predicting the response.

In this study obtained a γ-oryzanol of 783.65 mg.L−1. According to Arab et al. (2011), RBO contains about 0.9–2.9% of γ-oryzanol, 1.5 to 2.9% according to Krishna et al. (2001), 119.75–281.95 mg.g−1 oil according to Sukanya et al. (2017), 3.33 g.100g−1 according to Al-Okbi et al. (2014), even up to 3000 mg.kg−1 according to Shin et al. (1997). γ-oryzanol extracted with 3:2 chloroform: methanol mixture yielding 23.7–43.0 mg.g−1 in the crude RBO (Azrina et al., 2008). According to Iqbal et al. (2005), the exact composition of γ-oryzanol depends on the rice cultivars. Furthermore, according to Butsat and Siriamornpun, (2010), that the content of γ-oryzanol in rice affected by the variety and growing conditions, as the antioxidant component will respond differently to environmental changes.

The effect of different solvent concentrations can produce different yields have been also explained by Cacace and Mazza (2003), that the change in concentration in the solvent will modify the physical properties of the solvent such as density, dynamic viscosity, and dielectric constant. The solubility of the compound will also be modified by changes in solvent concentration, and this can affect yield. According to Japon-Lujan et al. (2006), 80% aqueous ethanol (v/v) was the optimum solvent for extraction of the targeted phenolics from olive leaf and it can be used as a replacement of toxic solvents (methanol, diethyl ether, chloroform) to obtain bioactive phenols for human use. Further, according to Malik and Bradford (2008), extraction with 80% methanol (v/v) was reported as the most effective method for olive leaves polyphenols. Related to the amount of solvent, the ratio of solvents to solids used by researchers for extraction varies greatly from 4 to 100, but the ratio between 10 and 50 is mostly reported in the literature (Kiritsakis et al., 2010).

3.2.3 Vitamin E concentration of RBO

The result of ANOVA for response surface reduced quadratic model of vitamin E showed that the model F-value of 10.86 implies the model is significant. Values of Prob > F of 0.0002 (less than 0.05) indicate model terms are significant. In this case, the linear terms of extraction time (x1), the linear terms of ethanol concentration (x2), the quadratic term of extraction time (x12), and the quadratic term of ethanol concentration (x22). Values greater than 0.1000 indicate the model terms are not significant, so that the linear term of ethanol volume (x3), the quadratic term of ethanol volume (x32), the interaction between extraction time and ethanol concentration (x1x2), the interaction between extraction time and ethanol volume (x1x3), and the interaction between ethanol concentration and ethanol volume (x2x3) did not give any significant contribution on vitamin E concentration. The lack-of-fit F-value of 2.26 implies the lack-of-fit is not significant relative to the pure error, this indicates that the model is suitable to describe the effect of variable observed for the vitamin E and that the developed model is adequate for predicting the response. 

In this study obtained a vitamin E of 127.01 mg.L−1. According to Xu et al. (2007), RBO contains 0.37–1.84 mg.g−1 oil, 170–218 µg.g−1 according to Schramm et al. (2007) and 665 μg.g−1 according to Al-Okbi et al. (2014). Short-chain alcohols, especially ethanol and isopropanol, have been proposed as alternative extraction solvents due to their greater safety. Alcohols tend to extract more non-glyceride materials than n-hexane, due to their greater polarity. Typically, alcohol-extracted oils contain more phosphatides and unsaponifiable compounds (Lusas et al., 1991). Further, Hu et al. (1996) reported that the average amount of vitamin E of RBO with isopropanol extraction was greater than that with hexane extraction and related to the amount of solvent, an increase in isopropanol-to-bran ratio (w/w) from 2:1 to 3:1 extracted 9.4% more crude RBO that contained 10% more vitamin E which explains that the amount of solvent affects the extraction of vitamin E. They also reported that the extraction time did not have a significant effect on the amount of RBO or vitamin E extracted.

3.3 Interpretation of response surface and contour plots

Based on the fitted model, the response surface and contour plots were generated by the model for extraction of oil yield, γ-oryzanol, and vitamin E of RBO as a response. In order to gain a better understanding of this study, related to the effect of observed variables on the response, then the predicted models are presented as the 3D plot (Figs. 13, respectively). Response surface plot is a representation of the surface plot in 3-D space as the plot determining optimum operating conditions reaching maximum from the best-fitted model. These plots are obtained depicting two variables within experimental range and keeping the third variable at a constant level. According to Bezerra et al. (2008), a two-dimensional representation of a three-dimensional plot can be explained. If there are three or more variables, the plot visualization is possible only if one or more variables are set to a constant value.

Figure 1 shows a 3D plot corresponding to the effect of x2 and x3 on oil yield at the fixed of x1 (5 h). The effect of x3 is stronger than the effect of x2 on increasing the oil yield. It is evident in the coefficient variable of x2 and x3, where coefficient variable of x3 > x2, and it is evident in the coefficient estimate of the ANOVA, where the coefficient variable in term of actual factor of x3 (0.98) is higher than x2 (0.52), while the effect of x1 is not significant in increasing the yield of the RBO, the estimated coefficient of x1 is 0.14.

The plot that discloses the effects of x1 and x2 on increasing γ-oryzanol of RBO at a fixed of x3 (600 mL) is shown in Figure 2. It can be seen that the γ-oryzanol had a maximum point. The combination of variables observed is an effect on increasing the γ-oryzanol of RBO. At x3 constant (600 mL), the influence of x2 is stronger than the effect of x1 on increased of the γ-oryzanol. It is evident in the 3D plot that the curve at x2 is more convex than the curve at x1, meaning that a small change in x2 has greatly affected the acquisition of γ-oryzanol.

The 3D plot corresponding to explain the effect of x1 and x2 on the vitamin E at a fixed of x3 (600 mL) is shown in Figure 3. It is showed that the vitamin E had a maximum point. x1 and x2 were effects on increasing the vitamin E. At x3 constant (600 mL), the effect of x1 is very strong compared to the effect of x2 on the acquisition of vitamin E. It can also be proved from the coefficients of x1 and x2 in equation (5), where the value of coefficient x1 > x2.

thumbnail Fig. 1

Response surface plot of oil yield RBO at a fixed extraction time of 5 h.

thumbnail Fig. 2

Response surface plot of γ-oryzanol RBO at a fixed ethanol volume of 600 mL.

thumbnail Fig. 3

Response surface plot of vitamin E RBO at a fixed ethanol volume of 600 mL.

3.4 Interpretation of the optimum conditions

Based on the optimization solution generated from the Design Expert 10 software that the optimum of x1, x2, and x3 to obtain the maximum of oil yield, γ-oryzanol, and vitamin E of RBO can be achieved at x1 of 5.30 h, x2 of 89.21%, and x3 of 686.66 mL. It can be explained that an increase x1 of 4 h to 5.30 h and x2 of 76% to 89.21% (at fixed x3 of 600 mL) cause an increase in the oil yield, γ-oryzanol, and vitamin E of RBO. The addition of x1 of 5.30 h up to 6 h and addition x2 of 89.21% up to 98% did not cause an increase in oil yield, γ-oryzanol, and vitamin E of RBO.

The same explanation at fixed x1 of 5 h, an increase of x2 of 76% to 89.21%, and x3 of 500 mL to 686.66 mL cause an increase in the oil yield, γ-oryzanol, and vitamin E of RBO. Further, increasing x2 of 89.21% up to 96%, and addition x3 of 686.66 mL up to 700 mL did not cause an increase in the oil yield, γ-oryzanol, and vitamin E of RBO. The same phenomenon at fixed x2 of 86%, an addition of x1 of 4 h to 5.30 h, and x3 of 500 mL up to 686.66 mL cause an increase in the oil yield, γ-oryzanol, and vitamin E of RBO. The addition of x1 of 5.30 h up to 6 h and addition x2 of 89.21% up to 98% did not cause an increase in the oil yield, γ-oryzanol, and vitamin E of RBO.

3.5 Verification model

An optimization model of RBO extraction by maceration method with stirring has been developed. The laboratory scale for verification of the model has been carried out by conducting triplicate experiments using the recommended value of variables from the software. The result showed that the average oil yield, γ-oryzanol, and vitamin E were 13.91% ± 0.44, 775.36 ± 31.38, and 126.71 ± 0.61 mg.L−1, respectively. The RBO produced in the verification process showed that the average value of the oil yield, γ-oryzanol, and vitamin E was close to the predicted values (14.47%, 783.65 mg.L−1, and 127.01 mg.L−1, respectively). This indicates that the model developed is quite valid in their predictions.

4 Conclusions

RBO extraction use ethanol by maceration method accompanied by stirring is very promising to be applied in the industry, as an effort to implement the concept of green solvents that are safe for consumers and the environment. The effect of extraction time, ethanol concentration, and ethanol volume on the oil yield, γ-oryzanol, and vitamin E of RBO has been evaluated. Ethanol can be used to extract RBO without involving heating and proven to be able to provide satisfactory results. The optimization of the extraction process has been developed and verified at the laboratory scale with sufficient results. Predictive values produced by a model that approaches the actual value have proven that the model developed is quite valid and feasible to be applied in the RBO extraction process.

Conflicts of interest

The authors declare that they have no conflicts of interest in relation to this article.

Acknowledgments

The authors would like to thank the Higher Education Ministry of Indonesia (DIKTI) and Department of Chemical Engineering, Politeknik Negeri Ujung Pandang, Makassar, Indonesia.

References

  • Al-Okbi SY, Ammar NM, Mohamed DA, et al. 2014. Egyptian rice bran oil: Chemical analysis of the main phytochemicals. Riv Ital Sostanze Gr 91: 47–58. [Google Scholar]
  • Amiri H, Nabizadeh R, Martinez SS, et al. 2018. Response surface methodology modeling to improve degradation of Chlorpyrifos in agriculture runoff using TiO2 solar photocatalytic in a raceway pond reactor. Ecotoxicol Environ Saf 147: 919–925. DOI: 10.1016/j.ecoenv.2017.09.062. [CrossRef] [PubMed] [Google Scholar]
  • Anwar F, Tabreez A, Zahid M. 2005. Methodical characterization of rice (Oryza sativa) bran oil from Pakistan. Grasas Aceites 56 Fasc. 2: 125–134. [CrossRef] [Google Scholar]
  • Arab F, Alemzadeh I, Maghsoudi V. 2011. Determination of antioxidant component and activity of rice bran extract. Scientia Iranica 18(6): 1402–1406. DOI: 10.1016/j.scient.2011.09.014. [CrossRef] [Google Scholar]
  • Awad OI, Mamat R, Ali OM, et al. 2017. Response surface methodology (RSM) based multi-objective optimization of fusel oil-gasoline blends at different water content in SI engine. Energy Convers Manag 150: 222–241. DOI: 10.1016/j.enconman.2017.07.047. [Google Scholar]
  • Azrina A, Maznah I, Azizah AH. 2008. Extraction and determination of oryzanol in rice bran of mixed herbarium UKMB; AZ 6807: MR 185, AZ 6808: MR 211, A Z6809: MR 29. ASEAN Food J 15(1): 89–96. [Google Scholar]
  • Banga J, Tripathi CKM. 2009. Response surface methodology for optimization of medium components in submerged culture of Aspergillus flavus for enhanced heparinase production. Lett Appl Microbiol 49: 204–209. DOI: 10.1111/j.1472-765X.2009.02640.x. [CrossRef] [PubMed] [Google Scholar]
  • Bas D, Boyaci IH. 2007. Modelling and optimization I: Usability of response surface methodology. J Food Eng 78: 836–845. DOI: 10.1016/j.jfoodeng.2005.11.024. [Google Scholar]
  • Bauernfeind JC, Desai ID. 1977. The tocopherol content of food and influencing factors. Crit Rev Food Sci Nutr 8: 337–382. DOI: 10.1080/10408397709527226. [Google Scholar]
  • Baümler ER, Carrín ME, Carelli AA. 2016. Extraction of sunflower oil using ethanol as solvent. J Food Eng 178: 190–197. DOI: 10.1016/j.jfoodeng.2016.01.020. [Google Scholar]
  • Bessa LCBA, Ferreira MC, Rodrigues CEC, Batista EAC, Meirelles AJA. 2017. Simulation and process design of continuous countercurrent ethanolic extraction of rice bran oil. J Food Eng DOI: 10.1016/j.jfoodeng.2017.01.019. [Google Scholar]
  • Bezerra MA, Santelli RE, Oliveira EP, Villar LS, Escaleira LA. 2008. Response surface methodology (RSM) as a tool for optimization in analytical chemistry. Talanta 76(5): 965–977. DOI: 10.1016/j.talanta.2008.05.019. [CrossRef] [PubMed] [Google Scholar]
  • Butsat S, Siriamornpun S. 2010. Antioxidant capacities and phenolic compounds of the husk, bran and endosperm of Thai rice. Food Chem 119: 606–613. DOI: 10.1016/j.foodchem.2009.07.001. [Google Scholar]
  • Cacace JE, Mazza G. 2003. Optimization of extraction of anthocyanins from black currants with aqueous ethanol. J Food Sci 68(1): 240–248. [Google Scholar]
  • Chan YH, Yusup S, Quitain AT, Uemura Y, Loh SK. 2017. Fractionation of pyrolysis oil via supercritical carbon dioxide extraction: Optimization study using response surface methodology (RSM). Biomass Bioenergy 107: 155–163. DOI: 10.1016/j.biombioe.2017.10.005. [Google Scholar]
  • Chemat F, Vian MA, Cravotto G. 2012. Green extraction of natural products: Concept and principles. Int J Mol Sci 13: 8615–8627. DOI: 10.3390/ijms13078615. [Google Scholar]
  • Chen MH, Bergman CJ. 2005. A rapid procedure for analyzing rice bran tocopherol, tocotrienol and gamma oryzanol contents. J Food Compos Anal 18: 139–151. DOI: 10.1016/j.jfca.2003.09.004. [CrossRef] [Google Scholar]
  • Chen Qi, Fung Ka Y, Lau Yeuk T, Ng Ka M, Lau David TW. 2016. Relationship between maceration and extraction yield in the production of Chinese herbal medicine. Food Bioprod Process. DOI: 10.1016/j.fbp.2016.02.005. [Google Scholar]
  • Chew KK, Ng SY, Thoo YY, Khoo MZ, Wan Aida WM, Ho CW. 2011. Effect of ethanol concentration, extraction time and extraction temperature on the recovery of phenolic compounds and antioxidant capacity of Centella asiatica extracts. Int Food Res J 18: 571–578. [Google Scholar]
  • Chirinos R, Rogez H, Campos D, Pedreschi R, Larondelle Y. 2007. Optimization of extraction conditions of antioxidant phenolic compounds from mashua (Tropaeolum tuberosum Ruíz and Pavón) tubers. Sep Purif Technol 55(2): 217–225. [Google Scholar]
  • Desai ID, Swann MA, Garcia Tavares ML, Dutra de Oliveira BS, Duarte FAM, Dutra de Oliveira JE. 1980. Vitamin E status of agricultural migrant workers in Southern Brazil. Amer J Clin Nutr 33: 2669–2673. DOI: 10.1093/ajcn/33.12.2669. [CrossRef] [PubMed] [Google Scholar]
  • Elkhaleefa A, Shigidi I. 2015. Optimization of sesame oil extraction process conditions. Adv Chem Eng Sci 5: 305–310. DOI: 10.4236/aces.2015.53031. [CrossRef] [Google Scholar]
  • Firestone D. (Ed.). 1999. Physical and chemical characteristics of oils, fats, and waxes. Washington, DC: AOCS Press, pp. 86–87. [Google Scholar]
  • Hu W, Wells JH, Tai-Sun S, Godber JS. 1996. Comparison of isopropanol and hexane for extraction of vitamin E and oryzanois from stabilized rice bran. JAOCS 73(12): 1653–1656. DOI: 10.1007/BF02517967. [CrossRef] [Google Scholar]
  • Iqbal S, Bhanger MI, Anwar F. 2005. Antioxidant properties and components of some commercially available varieties of rice bran in Pakistan. Food Chem 93: 265–272. DOI: 10.1016/j.foodchem.2004.09.024. [Google Scholar]
  • Imsanguan P, Roaysubtawee A, Borirak R, Pongamphai S, Douglas S, Douglas PL. 2008. Extraction of α-tocopherol and γ-oryzanol from rice bran. Food Sci Technol 41: 1417–1424. [Google Scholar]
  • Japon-Lujan R, Luque-Rodriguez JM, DeCastro MDL. 2006. Multivariate optimisation of the microwave-assisted extraction of oleuropein and related biophenols from olive leaves. Anal Bioanal Chem 385(4): 753–759. [CrossRef] [PubMed] [Google Scholar]
  • Keshtegar B, Mert C, Kisi O. 2018. Comparison of four heuristic regression techniques in solar radiation modeling: Kriging method vs. RSM, MARS and M5 model tree. Renew Sustain Energy Rev 81: 330–341. DOI: 10.1016/j.rser.2017.07.054. [CrossRef] [Google Scholar]
  • Kiritsakis K, Kontominas MG, Kontogiorgis C, Hadjipavlou-Litina D, Moustakas A, Kiritsakis A. 2010. Composition and antioxidant activity of olive leaf extracts from greek olive cultivars. J Am Oil Chem Soc 87(4): 369–376. [Google Scholar]
  • Krishna AGG, Khatoon S, Shielaa PM, Sarmandala CV, Indirab TN, Mishrac A. 2001. Effect of refining of crude rice bran oil on the retention of oryzanol in the refined oil. JAOCS 78(2): 127–131. [CrossRef] [Google Scholar]
  • Lerma-García MJ, Herrero-Martínez JM, Simó-Alfonso EF, Mendonça CRB, Ramis-Ramos G. 2009. Composition, industrial processing and applications of rice bran γ-oryzanol. Food Chem 115(2): 389–404. DOI: 10.1016/j.foodchem.2009.01.063. [Google Scholar]
  • Li L, Li X, Yan C, et al. 2014. Optimization of methyl orange removal from aqueous solution by response surface methodology using spent tea leaves as an adsorbent. Front Environ Sci Eng 8: 496–502. DOI: 10.1007/s11783-013-0578-0. [Google Scholar]
  • Li Z, Smith KH, Stevens GW. 2016. The use of environmentally sustainable bio-derived solvents in solvent extraction applications – A review, Chin J Chem Eng 24: 215–220. DOI: 10.1016/j.cjche.2015.07.021. [CrossRef] [Google Scholar]
  • Lusas EW, Watkins LR, Koseoglu SS. 1991. Isopropyl alcohol to be tested as solvent. Inform 2: 970–976. [Google Scholar]
  • Malik NSA, Bradford JM. 2008. Recovery and stability of oleuropein and other phenolic compounds during extraction and processing of olive (Olea europaea L.) leaves. J Food Agric Environ 6(2): 8–13. [Google Scholar]
  • Mas’ud F, Mahendradatta M, Amran L, Zainal Z. 2017. Optimization of mango seed kernel oil extraction using response surface methodology. Oilseeds Fats Crops Lipids. Available from www.ocl-journal.org. DOI: 10.1051/ocl/2017041. [Google Scholar]
  • Mohammed BS, Khed VC, Nuruddin MF. 2018. Rubbercrete mixture optimization using response surface methodology. J Clean Prod 171: 1605–1621. DOI: 10.1016/j.jclepro.2017.10.102. [Google Scholar]
  • Oliveira R, Oliveira V, Aracava KK, Rodrigues CEdC. 2012. Effects of the extraction conditions on the yield and composition of rice bran oil extracted with ethanol – A response surface approach. Food Bioprod Process 90(C1): 22–31. DOI: 10.1016/j.fbp.2011.01.004. [CrossRef] [Google Scholar]
  • Oniya OO, Oyelade JO, Ogunkunle O, Idowu DO. 2017. Optimization of solvent extraction of oil from sandbox kernels (Hura Crepitans L.) by a response surface method. Energy Policy Res 4(1): 36–43, DOI: 10.1080/23815639.2017.1324332. [CrossRef] [Google Scholar]
  • Pichai E, Krit S. 2015. Optimization of solid-to-solvent ratio and time for oil extraction process from spent coffee grounds using response surface methodology. ARPN J Eng App Sci 10(16): 7049–7052. [Google Scholar]
  • Péres VF, Maria JS, Melecchi IS, et al. 2006. Comparison of soxhlet, ultrasound-assisted and pressurized liquid extraction of terpenes, fatty acids and Vitamin E from Piper gaudichaudianum Kunth. J Chromatogr A 1105(1-2 SPEC. ISS.): 115–118. DOI: 10.1016/j.chroma.2005.07.113. [CrossRef] [PubMed] [Google Scholar]
  • Pourali O, Feridoun SA, Hiroyuki Y. 2009. Simultaneous rice bran oil stabilization and extraction using sub-critical water medium. J of Food Eng 95: 510–516. DOI: 10.1016/j.jfoodeng.2009.06.014. [CrossRef] [Google Scholar]
  • Rafi NM, Halim NRA, Amin AM, Sarbon NM. 2015. Response surface optimization of enzymatic hydrolysis conditions of lead tree (Leucaena leucocephala) seed hydrolyzate. Int Food Res J 22(3): 1015–1023. [Google Scholar]
  • Rodrigues CEC, Oliveira R. 2010. Response surface methodology applied to the analysis of rice bran oil extraction process with ethanol. Int J Food Sci Technol 45(4): 813–820. DOI: 10.1111/j.1365-2621.2010.02202.x. [Google Scholar]
  • Rogers EJ, Rice SM, Nicolosi RJ, Carpenter DR, McClelland CA, Romanczyk LJ Jr. 1993. Identification and quantitation of γ-oryzanol components and simultaneous assessment of tocols in rice bran oil. JAOCS 70(3): 301–307. [CrossRef] [Google Scholar]
  • Sani I. 2014. Soxhlet extraction and physicochemical characterization of Mangifera indica L. Seed kernel oil. Res Rev: J Food Dairy Tech 2(1): 20–24. [Google Scholar]
  • Saxena DK, Sharma SK, Sambi SS. 2011. Comparative extraction of cottonseed oil by n-hexane and ethanol. ARPN J Eng App Sci 6(1): 84–89. [Google Scholar]
  • Schramm RC, Alicia M, Na H, Marybeth L. 2007. Fractionation of the rice bran layer and quantification of vitamin E, oryzanol, protein, and rice bran saccharide. J Biol Eng 1(1): 9. DOI: 10.1186/1754-1611-1-9. [CrossRef] [PubMed] [Google Scholar]
  • Seema P. 2015. Cereal bran fortified-functional foods for obesity and diabetes management: Triumphs, hurdles and possibilities. J Funct Foods 14: 255–269. [Google Scholar]
  • Shin T, Godber JS, Martin DE, Wells JH. 1997. Hydrolitic stability and changes in E vitamers and oryzanol of extruded rice bran during storage. J Food Sci 62: 704–708. [Google Scholar]
  • Singh V, Belova L, Singh B, Sharma YC. 2018. Biodiesel production using a novel heterogeneous catalyst, magnesium zirconate (Mg2Zr5O12): Process optimization through response surface methodology (RSM). Energy Convers Manag 174: 198–207. DOI: 10.1016/j.enconman.2018.08.029. [Google Scholar]
  • Soundararajan R, Ramesh A, Mohanraj N, Parthasarathi N. 2016. An investigation of material removal rate and surface roughness of squeeze casted A413 alloy on WEDM by multi response optimization using RSM. J Alloys Compd 685: 533–545. DOI: 10.1016/j.jallcom.2016.05.292. [Google Scholar]
  • Sukanya M, Aikkarach K, Khongsak S, Riantong S. 2017. Physicochemical and antioxidant properties of rice bran oils produced from colored rice using different extraction methods. J Oleo Sci 66(6): 565–572. DOI: 10.5650/jos.ess17014. [CrossRef] [PubMed] [Google Scholar]
  • Tan YH, Abdullah MO, Hipolito CN, Syuhada N, Zauzi A. 2017. Application of RSM and Taguchi methods for optimizing the transesterification of waste cooking oil catalyzed by solid ostrich and chicken-eggshell derived CaO. Renewable Energy 114: 437–447. DOI: 10.1016/j.renene.2017.07.024. [CrossRef] [Google Scholar]
  • Tekin K, Hao N, Karagoz S, Ragauskas AJ. 2018. Ethanol: A promising green solvent for the deconstruction of lignocellulose. ChemSusChem. DOI: 10.1002/cssc.201801291. [Google Scholar]
  • Tiwari U, Cummins E. 2009. Nutritional importance and effect of processing on tocols in cereals. Trends Food Sci Technol 20: 511–520. DOI: 10.1016/j.tifs.2009.06.001. [CrossRef] [Google Scholar]
  • Xu Z, Godber JS. 2001. Comparison of supercritical fluid and solvent extraction methods in extracting gamma-oryzanol from rice bran. J Am Oil Chem Soc 77: 1127–1131. DOI: 10.1007/s11746-000-0087-4. [Google Scholar]
  • Xu Z, Hua N, Godber JS. 2001. Antioxidant activity of tocopherols, tocotrienols, and gamma oryzanol components from rice bran against cholesterol oxidation accelerated by 2, 2’-Azobis (2-methylpropionamidine) dihydrochloride. J Agr Food Chem 49: 2077–2081. DOI: 10.1021/jf0012852. [CrossRef] [Google Scholar]
  • Xu YX, Hanna MA, Josiah SJ. 2007. Hybrid hazelnut oil characteristics and its potential oleochemical application. Ind Crops Prod 26(1): 69–76. DOI: 10.1016/j.indcrop.2007.01.009. [CrossRef] [Google Scholar]
  • Yolmeh M, Habibi Najafi MB, Farhoosh R. 2014. Optimisation of ultrasound-assisted extraction of natural pigment from annatto seeds by response surface methodology (RSM). J Food Chem 155: 319–324. DOI: 10.1016/j.foodchem.2014.01.059. [CrossRef] [Google Scholar]

Cite this article as: Mas’ud F, Fajar, Bangngalino H, Indriati S, Todingbua A, Suhardi, Sayuti M. 2019. Model development to enhance the solvent extraction of rice bran oil. OCL 26: 16.

All Tables

Table 1

Five levels of independent variables of central composite design.

Table 2

Central composite design, showing coded values of independent variables, with observed of yield, γ-oryzanol, and vitamin E of RBO.

All Figures

thumbnail Fig. 1

Response surface plot of oil yield RBO at a fixed extraction time of 5 h.

In the text
thumbnail Fig. 2

Response surface plot of γ-oryzanol RBO at a fixed ethanol volume of 600 mL.

In the text
thumbnail Fig. 3

Response surface plot of vitamin E RBO at a fixed ethanol volume of 600 mL.

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.