Complete analysis of castor oil methanolysis to obtain biodiesel

Fuel 147 (2015) 95–99

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Complete analysis of castor oil methanolysis to obtain biodiesel Nuria Sánchez a, Ramiro Sánchez b, José M. Encinar a,⇑, Juan F. González c, Gloria Martínez a a

Departamento de Ingeniería Química y Química Física, Universidad de Extremadura, Avda. de Elvas s/n., 06006 Badajoz, Spain Centro de Investigaciones Científicas y Tecnológicas de Extremadura, Instituto de Investigaciones Agrarias Finca La Orden-Valdesequera, Autovía A-V, km 372, 06187 Guadajira, Badajoz, Spain c Departamento de Física Aplicada, Universidad de Extremadura, Avda. de Elvas s/n., 06006 Badajoz, Spain b

h i g h l i g h t s  Response Surface Methodology was suitable to optimize castor oil transesterification.  The most influential variables were methanol:oil ratio and catalyst concentration.  High ester content could be reached in a wide range of experimental conditions.  The optimum conditions led to biodiesel with 97 wt% ester content.

a r t i c l e

i n f o

Article history: Received 23 June 2014 Received in revised form 19 January 2015 Accepted 20 January 2015 Available online 31 January 2015 Keywords: Biodiesel Castor oil Methanolysis Response Surface Methodology

a b s t r a c t Biodiesel production provides an alternative non-fossil fuel without the need to redesign current direct injection engine technology. In this work biodiesel production from castor oil was analyzed studying all of the main variables of the process. Experimental design was used to evaluate the influence of catalyst concentration, methanol:oil molar ratio, reaction temperature and reaction time in the methyl ester content reached by castor oil transesterification. Results were analyzed by Response Surface Methodology and a quadratic polynomial model was achieved. The model fitted properly the data, as was shown by the validation experiments. The most influential variables were catalyst concentration and methanol:oil molar ratio and the optimum conditions were 0.064 mol L1 of CH3OK, 18.8:1 as methanol:oil molar ratio, 45 °C and 10 min of reaction. In these conditions, 97 wt% methyl ester content biodiesel was achieved. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Biodiesel is a promising diesel fuel substitute because it is a clean renewable fuel which can be used in any direct injection engine without the need to redesign the current technology. Biodiesel is derived from renewable and domestic feedstock and shows higher biodegradability than fossil fuels excellent lubricity and negligible sulfur content [1,2]. For biodiesel–diesel blends, comparable engine efficiency was showed. From an environmental point of view, in spite of higher emission level of NOx, the emission from biodiesel combustion contained lower amounts of CO, CO2, HC and smoke [3]. The most common method to obtain biodiesel is the transesterification of vegetable oils or animal fats. In the reaction, triglycerides are reacted in presence of a catalyst with an alcohol with short-chain [2,4]. Methanol is the most used alcohol because it is ⇑ Corresponding author. Tel.: +34 924289672. E-mail address: [email protected] (J.M. Encinar). http://dx.doi.org/10.1016/j.fuel.2015.01.062 0016-2361/Ó 2015 Elsevier Ltd. All rights reserved.

the least expensive alcohol and it shows chemical advantages such as its shorter chain and its polar nature [1,5]. The most employed catalysts are homogeneous alkaline catalysts such as NaOH, KOH, CH3ONa and CH3OK and methoxides are the most suitable due to their ability to dissociate into the methoxide and the metal ion without the production of water during transesterification reaction [6]. Biodiesel feedstock can be categorized into three groups: vegetable oils (edible or non-edible oils), animal fats and used waste cooking oil. Biodiesel has been mainly produced from edible vegetable oils all over the world. More than 95% of global biodiesel production is made from edible vegetable oils and this fact has an influence on the global imbalance to the market demand and the food supply [5]. In addition, the price of this kind of feedstock makes 70–80% of the total biodiesel cost. Non-edible oils, which are not used in human nutrition and whose plants could grow in barren lands, should be increasingly used. Non-edible oil plants usually can be cultivated in lands unsuitable for human crops with much lower cost and no influence in food market [7].

96

N. Sánchez et al. / Fuel 147 (2015) 95–99

Castor oil is one of the most often used non-edible oil in biodiesel synthesis [7]. Castor plant is originally a tree or shrub and there are different varieties that can be cultivated. Castor oil seeds usually contain 40–55% oil and the average yield of castor oil seed in the world is about 1.1 t ha1, although it may be possible to obtain a maximum of 4.2 t ha1. Therefore, castor is amongst the plants with the highest oil yield potential [8]. Biodiesel from castor oil has been obtained and its properties has been studied, as well as, the properties of blends with biodiesel from different origin and diesel fuel [9,10]. The study of biodiesel production is usually carried out by the analysis of the influence of the reaction parameters in the transesterification of triglycerides. To carry out this analysis Response Surface Methodology (RSM) is one of the most suitable methodologies. This is a useful statistical technique to optimize a process because it allows the simultaneous consideration of many variables at different levels and the interactions between those variables. Moreover this methodology only requires a reduced number of experimental runs to generate statistically acceptable results [11]. There are a lot of works where ethanolysis of castor oil with basic homogeneous catalyst has been studied; the most relevant ones are listed below. De Oliveira et al. [12] assumed a Taguchi experimental design to study the influence of NaOH concentration (0.5–1.5 wt%), ethanol:oil molar ratio (3:1–9:1) and reaction temperature (30–70 °C) and time (1–3 h). All main variables of reaction were studied and 96.2% of conversion was reached under the optimal conditions (0.5 wt% catalyst, 3:1 ethanol:oil, 70 °C and 3 h). De Lima da Silva et al. [13] used RSM to optimize the transesterification reaction of castor oil using ethanol as alcohol and sodium ethoxide as catalyst. The studied parameters were reaction temperature (30–80 °C), catalyst concentration (0.5–1.5 wt%) and ethanol:oil molar ratio (12:1–20:1). The best results (93.78 wt% ester content) were reached at 30 °C with large catalyst content and lower ethanol:oil ratio or with lower catalyst content and large alcohol percentage. Cavalcante et al. [14] kept 30 °C as a constant parameter and studied the influence of ethanol:oil molar ratio, catalyst content and reaction time in the yield of biodiesel after the transesterification of castor oil with KOH as catalyst. In this work, the highest yield (close to 86%) was obtained using 11:1 as ethanol:oil molar ratio, 1.75 wt% KOH and 90 min as reaction time. Barbosa et al. [15] studied the production of biodiesel from castor oil and mixed castor and soybean oils. In the used conditions, they managed only 30% ethyl esters content after 30 h of reaction. Montoya et al. [16] used NaOH as catalyst and adopted RSM as methodology to optimize catalyst amount, ethanol:oil ratio and reaction temperature. The maximum ethyl ester content was 93.64 wt% using 1.2 wt% catalyst with respect to oil weight, 9.86:1 ethanol:oil molar ratio, 30 °C and 1 h. On the other hand, the number of works where methanol was used is more limited. Meneguetti et al. [17] started with a comparison of ethanolysis versus methanolysis. In this work they concluded that similar yields could be obtained using both alcohols but reaction time for methanolysis was shorter. Canoira et al. [18] achieved total transesterification conversion using 1 wt% CH3ONa, 5:1 methanol:oil molar ratio and 40 °C, although the study of variables was carried out one by one. Dias et al. [19] grew castor plants, extracted its oil and produced biodiesel. Moreover, they established models to predict biodiesel yield and methyl ester content among other properties. The models depended on reaction temperature and reaction time. The highest reached ester content was 83.41 wt%, carrying out the reaction at 65 °C for 8 h. Considering the background, methanolysis has been less studied than ethanolysis as route to obtain biodiesel from castor oil; although this has been showed as a faster route. In addition, it is important a global study where optimal conditions were

established taking into account all the main variables. In this way, the aim of this work was to achieve the optimization of the parameters which affect methanolysis of castor oil, using CH3OK as catalyst. Catalyst concentration, methanol:oil molar ratio, reaction temperature and reaction time were studied in this work by the RSM. In addition, catalyst concentration was considered as molar concentration taking into account the total reaction medium, instead of as percentage based on the oil weight, because the effect of the alcohol:oil molar ratio could be mask by the increase or decrease of the real catalyst concentration in the medium. 2. Materials and methods 2.1. Materials Refined castor oil, supplied by INTERFAT (Barcelona, Spain), was transesterified using methanol (99.6%) as alcohol and potassium methoxide (90%) as catalyst, both were purchased from Panreac and Alfa Aesar, respectively. Sulfuric acid (95–98%) to neutralize the catalyst was also purchased from Panreac. The reagents used for oil characterization were of analytical grade. Methyl esters (employed as standards in the chromatographic determination) were purchased from Sigma–Aldrich. 2.2. Transesterification reaction The reactions were carried out under atmospheric pressure in a spherical glass reactor (500 mL) provided with a condensation system, sampling outlet, magnetic stirring, heating and temperature control system. A water batch was used to maintain the reaction temperature. The amount of oil and alcohol was fixed to reach 300 mL in every reaction and the catalyst concentration was measured as molar concentration. To carry out the experiments, the oil was preheated up to desired temperature, and then the alcoholcatalyst mixture was added to the reactor. Reaction time was established for every reaction and sulfuric acid was used to stop the reaction by catalyst neutralization. The final mixture was washed with distilled water until glycerol, methanol and the salts of the neutralization were removed. The remaining water was removed by heating at 110 °C. 2.3. Experimental design and statistical analysis A central composite design (CCD) was applied to find out the influence of the operational conditions of the transesterification reaction, such as catalyst concentration, methanol:oil molar ratio, temperature and reaction time, on the methyl ester concentration. A three-level-four-factors CCD was adopted, requiring 29 experimental runs whose operational conditions are shown in Table 1. The variables were coded in the range of 1 to +1 to allow a direct comparison between variables according to Eq. (1):

xi ¼

2ðX i  X min Þ 1 ðX max  X min Þ

ð1Þ

where xi is the normalized value of the variable X at condition i; Xi is the actual value; Xmin and Xmax are the lower and upper limit, respectively. The limits for each variable were chosen by considering preliminary tests: catalyst amount: 0.028–0.064 mol L1, methanol:oil molar ratio: 9:1–22:1, reaction temperature: 33–57 °C and reaction time: 10–30 min. Experimental reactions were run at random to minimize errors due to possible systematic trends in the variables. Results were analyzed via RSM in order to fit a second-order polynomial model (Eq. (2)). Model fit quality was evaluated by variance analysis (ANOVA) for the model and a confidence level of a = 5% was used

97

N. Sánchez et al. / Fuel 147 (2015) 95–99 Table 1 CCD and experimental data. Runs

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

Table 2 Castor oil fatty acid profile and properties.

Catalyst concentration (mol L1) (A)

Methanol:oil molar ratio (B)

Temp. (°C) (C)

Time (min) (D)

Ester content (wt%)

0.028 0.064 0.046 0.046 0.046 0.046 0.046 0.046 0.028 0.064 0.028 0.064 0.028 0.064 0.028 0.064 0.028 0.064 0.028 0.064 0.028 0.064 0.028 0.064 0.046 0.046 0.046 0.046 0.046

15.5 (0) 15.5 (0) 9.0 (1) 22.0 (1) 15.5 (0) 15.5 (0) 15.5 (0) 15.5 (0) 9.0 (1) 9.0 (1) 22.0 (1) 22.0 (1) 9.0 (1) 9.0 (1) 22.0 (1) 22.0 (1) 9.0 (1) 9.0 (1) 22.0 (1) 22.0 (1) 9.0 (1) 9.0 (1) 22.0 (1) 22.0 (1) 15.5 (0) 15.5 (0) 15.5 (0) 15.5 (0) 15.5 (0)

45 45 45 45 33 57 45 45 33 33 33 33 57 57 57 57 33 33 33 33 57 57 57 57 45 45 45 45 45

20 20 20 20 20 20 10 30 10 10 10 10 10 10 10 10 30 30 30 30 30 30 30 30 20 20 20 20 20

94.6 94.9 91.8 94.5 93.6 93.7 94.6 96.3 73.0 89.8 89.7 93.8 84.8 89.4 92.2 94.3 84.9 90.3 89.5 94.3 91.2 89.7 93.1 94.0 94.9 94.0 93.7 93.3 96.0

(1) (1) (0) (0) (0) (0) (0) (0) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (0) (0) (0) (0) (0)

(0) (0) (0) (0) (1) (1) (0) (0) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (0) (0) (0) (0) (0)

(0) (0) (0) (0) (0) (0) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (1) (0) (0) (0) (0) (0)

Fatty acid profile C16:0 palmitic C18:0 stearic C18:1 oleic C18:2 linoleic C18:3 linolenic C18:1OH ricinoleic Density at 15 °C Viscosity at 40 °C Water content Saponification value Iodine value Acid value Molecular weight

similar to the composition of a generic castor oil: 90% ricinoleic acid, 4.5% linoleic acid and 3.6% oleic acid [19]. Ricinoleic acid, in which there is a hydroxyl group, shows special physical and chemical properties such as high density and viscosity, high hygroscopicity, low iodine value and high solubility in alcohols. The last property is the most interesting for the use of this oil to obtain biodiesel because it promotes transesterification reaction even at low temperature [8,21]. The acid value of the oil allows the use of basic catalysis in the transesterification reaction [22], therefore potassium methoxide will be an appropriate catalyst for this study; as far as basic catalysts are preferable in the case of oils with low acid value [6,21]. 3.2. Regression model development

to examine the statistical significance of the fitted polynomial model.

y ¼ b0 þ

X X XX bi xi þ bii x2i þ bij xi xj þ e i

i

1.3% 1.2% 3.6% 4.6% 0.4% 88.9% 961 kg m3 262 cSt 0.31% 179.3 mgKOH g1 80.5 mI2 g1 1.19 mgKOH g1 926 g mol1

ð2Þ

i
where y is the response factor (% methyl ester); xi the ith independent factor; b0 the intercept; bi the first order coefficient of the model; bii the quadratic coefficient of i factor; bij the lineal coefficients of the model for interaction between i and j factors; and e the experimental error attributed to y. 2.4. Analytical procedure Castor oil was characterized by the measurement of fatty acid profile, density at 15 °C, viscosity at 40 °C, water content and saponification, iodine and acid value. The used methods were described in previous works [20]. Biodiesel samples were analyzed by gas chromatography. VARIAN 3900 gas chromatograph with flame ionization detector was used. The column was a polyethylene glycol column (Zebron ZBWAX PLUS, Phenomenex, length: 30 m, film thickness: 0.5 lm and i.d.: 0.32 mm), the carrier gas was helium at a flow rate of 1.4 mL min1, and the analysis was carried out at 220 °C for 34 min and at 245 °C for 29 min with a ramp of 20 °C min1. Injector and detector temperatures were 270 and 300 °C, respectively. The internal standard method was used with methyl heptadecanoate as standard for minority esters and methyl erucate for methyl ricinoleate. The solvent was ethyl acetate and calibration curves were carried out for all the esters. 3. Results and discussion 3.1. Raw material Raw material properties and its fatty acid profile are shown in Table 2. The oil composition of feedstock used in this work is

The experimental design for transesterification of castor oil can be observed in Table 1. Three levels of the factors were studied in this work; however, a central composite rotatable design (fivelevel-four-factors) was previously applied. In the previous work, some of the axial points of the design corresponded with very low catalyst concentration and reaction time, so the value of the achieved ester content was quite low. The sharp change of the response factor for a short change in the value of the parameters of the model led to the impossibility of a simple mathematical equation fit the data. Then, the experimental design shown in this work has three levels for each factor instead of five. Table 1 lists the experimental conditions of the runs; they were carried out according to the design. The reached ester content as response factor by each run is also shown in Table 1. This planning is based on a CCD with four variables: catalyst concentration, methanol:oil molar ratio, reaction temperature and reaction time, and three levels for each factor. The results were analyzed via multiple regression, testing several models as linear, two factor interaction, three factor interaction, two and three factor interaction, cubic, quadratic and cubic plus quadratic, and, among all of them, the quadratic one fitted better the real data. Thus, the methyl ester content model, which includes linear, quadratic and cross-product terms, is shown in the following equation:

%Methyl ester ¼ 95:046 þ 2:094  A þ 2:794  B þ 1:315  C þ 1:201  D  0:840  AB  1:560  AC  1:117  AD  0:674  BC  1:132  BD  0:348  CD  0:829  AA  2:444  BB  1:958  CC  0:140  DD

ð3Þ

where linear terms show positive coefficient values and the rest ones have negative coefficient values, this allows to draw a response surface where a maximum of ester content could be found. The ANOVA for the response surface quadratic model is provided in Table 3. The values shown in the table allowed us to

determine the model significance, the effect of each term and the goodness of fit. In addition to the ANOVA table, the achieved determination coefficient was 0.898, therefore, 89.8% of the variability in the ester content can be predicted from the relation with the studied variables via Eq. (3). As can be seen in Table 3, the p-value of the model is lower than 0.05 (p-valueModel = 0.0001), hence there is a statistical relation between the response and the selected variables at a confidence level of 95%. An additional parameter to check the goodness of fit is the lack of fit. The lack of fit test assesses whether the model fits the data well, the test relies on the ability to estimate the variance of the response using an estimate that is independent of the model and a significant value of the lack of fit (p-value < 0.05) leads to reject the model. In this case p-value of lack of fit is 0.0699, therefore the selected model is suitable to fit the experimental data. On the other hand, the p-value of each term considered in the model is listed in the ANOVA table. These values claim that all linear term of analyzed variables had significant influence in the ester content of the biodiesel in the following order, as their p-values decrease: methanol:oil molar ratio, catalyst concentration, reaction temperature and reaction time. Meanwhile the two first variables were much more influential than the rest. Most of quadratic and cross-product terms showed p-values greater than 0.05, thereby they were not significant in the response factor. However, all the coefficients were kept in the equation of the model (Eq. (3)) in order to achieve more accurate fitting of the experimental data.

Methyl ester content (wt %)

N. Sánchez et al. / Fuel 147 (2015) 95–99

96 100 98 96 94 92 90 88 86

95 94 93 92 91

57

51 39 33

Temperature (ºC)

0.037 0.028

95

100 98 96 94 92 90 88 86 22

94 93 92 91 90 89

18.75

0.064 0.055

Methanol:oil molar ratio

Mean square

F-value

P-value

Model A B C D AB AC AD BC BD CD AA BB CC DD Error Lack of fit Pure error Total error

536.745 78.935 140.471 31.129 25.968 11.295 38.941 19.970 7.272 20.514 1.948 1.780 15.441 9.915 0.05044 60.809 56.211 4.598 60.809

14 1 1 1 1 1 1 1 1 1 1 1 1 1 1 14 10 4 14

38.339 78.935 140.471 31.129 25.968 11.295 38.941 19.970 7.272 20.514 1.948 1.780 15.441 9.915 0.05044 4.343 5.621 1.149 4.343

8.827 18.173 32.341 7.167 5.979 2.600 8.965 4.598 1.674 4.723 0.449 0.410 3.555 2.283 0.01161

0.0001 0.0008 0.0001 0.0180 0.0283 0.1291 0.0097 0.0501 0.2166 0.0474 0.5139 0.5324 0.0803 0.1531 0.9157

4.890

0.0699

87

0.046 9

88

0.037 0.028

-1

Catalyst Concentration (mol·L )

Methyl ester content (wt %)

Fig. 2. Response surface of methyl ester content vs. catalyst concentration and MeOH:oil molar ratio. Reaction conditions: reaction temperature: 45 °C, reaction time: 20 min.

97

100 98 96 94 92 90 88 86 30

96.5 96 95.5 95 94.5

25 94

20

10

DF

88

96

Time (min) 15

Sum of squares

89 -1

12.25

Source

0.046

Catalyst Concentration (mol·L )

15.5

Table 3 ANOVA for transesterification of castor oil.

0.055

Fig. 1. Response surface of methyl ester content vs. catalyst concentration and reaction temperature. Reaction conditions: MeOH:oil molar ratio: 18.8:1, reaction time: 20 min.

3.3. Response surface graphs The 3D response surface graph is one of the most common forms to show the effect of the different factors in the response variable. When more than two variables are considered in the model, two of them can be plotted, keeping in a constant value the others. According to the ANOVA table, the cross-product term with most significant effect is the term AC, i.e., catalyst concentration times temperature. The response surface graph considering these variables is shown in Fig. 1. As can be seen, the needed catalyst concentration is strongly dependent on the reaction temperature since biodiesel with high ester content could be reached by a reaction at 51 °C with just 0.04 mol L1 CH3OK. However if 33 °C is the reaction temperature, a catalyst concentration of 0.064 mol L1 would be needed, to achieve the same value of ester concentration. The high interaction between catalyst concentration and temperature is natural since catalysts increase the rate of the chemical

90

0.064

45

Methyl ester content (wt %)

98

33

39

45

51

57

93.5

Temperature (ºC)

Fig. 3. Response surface of methyl ester content vs. reaction time and temperature. Reaction conditions: catalyst concentration: 0.064 mol L1, MeOH:oil molar ratio: 18.8:1.

reaction and the increase of temperature leads to greater kinetic constant in the transesterification of castor oil [23]. The effect of catalyst concentration and methanol:oil molar ratio on ester content is shown in Fig. 2. The surface is drawn taking the average value for temperature and reaction time, 45 °C and 20 min, respectively. The highest ester content can be reached using 0.064 mol L1 of catalyst concentration (A = 1) and 18.8:1 of methanol:oil molar ratio (B = 0.5). Because of the highest ester concentration was obtained in the limit of the range of catalyst concentration, we could think about doing a new design increasing the limits of the factor. Nevertheless the reached ester content is high and the shape of the response surface plot makes understood

99

N. Sánchez et al. / Fuel 147 (2015) 95–99 Table 4 Experiments to validate the accuracy of the model. Catalyst concentration (mol L1) (A)

Methanol:oil molar ratio (B)

Temp. (°C) (C)

Time (min) (D)

Ester content from the model (wt%)

Experimental ester content (wt%)

Relative Error (%)

0.055 (0.5) 0.037 (0.5)

18.8 (0.5) 12.3 (0.5)

39 (0.5) 51 (0.5)

15 (0.5) 25 (0.5)

96.3 92.3

95.4 92.9

0.9 0.4

that the results would not change a lot if the catalyst limit were increased. In Fig. 3 catalyst concentration of 0.064 mol L1 and methanol:oil molar ratio of 18.8:1 are kept constant in the response surface equation and the effect of reaction temperature and reaction time on the ester content is shown. A wide area of high ester content can be observed in this graph, as in Figs. 1 and 2. The maximum response value was at 45 °C (C = 0) and 10 min (D = 1). Although castor oil transesterification is a endothermic reaction [23], the increase of reaction temperature have no positive effect in biodiesel production, at all times. As other authors reported, the optimum temperature to produce biodiesel from castor oil is about 30–45 °C. It is probably due to the solubility of the alcohols in castor oil, what also promotes the reaction between the reagents at moderate temperatures [13,18,24]. The most remarkable aspect of these response surface graphs is their wide area of high ester content and their slight changes of slope. This fact is a sign of stability of the system and it is a desired aspect because high ester content could be reached with different experimental conditions. Therefore, quite precise conditions are not required. 3.4. Optimization of methyl ester content and model validation Two extra experimental runs were carried out in order to evaluate the accuracy of the model. In Table 4 are shown the operational conditions as well as the obtained results. The aim of the model is to be able to predict the ester content which would be reached in some specific experimental conditions. Therefore, the best way to check if it works reasonably well is carrying out some runs in different conditions from the ones used in the design, but within the analyzed region. As shown in Table 4, the experimental ester contents were 95.4 and 92.9 wt%, very close to 96.3 and 92.3 wt%, the values estimated by the model. Thus, the obtained model predicts very well the experimental ester content, based on the reaction variables. To conclude, the specific conditions which led to the optimum of ester content were 0.064 mol L1 of catalyst concentration, 18.8:1 of methanol:oil molar ratio, 45 °C and 10 min of reaction temperature and time, respectively. The estimated and the experimental ester content, in these conditions, showed the same value, 97.0 wt%. In addition, final biodiesel yield was determined based on the initial oil amount for this reaction, which was 93.6 wt%. The similarity between experimental and predicted ester content showed the goodness of the model again.

4. Conclusions Response Surface Methodology was a suitable method for the optimization of castor oil transesterification, yielding a model which describes the ester content of the biodiesel depending on the most influential variables. The most influential factors were methanol:oil molar ratio and catalyst concentration. Reaction temperature and reaction time showed less effect. High ester content could be reached in a wide range of experimental conditions, although the optimum conditions were 0.064 mol L1 as catalyst concentration, 18.8:1 methanol:oil molar ratio, 45 °C and 10 min

of reaction time; the achieved biodiesel showed 97 wt% ester content. Acknowledgment The authors express their gratitude to the ‘‘MICINN’’ from Spain and the ‘‘Gobierno de Extremadura’’ for the financial support received to perform this study by means of Projects ENE200913881, PRI09B102 and ‘‘ayuda a grupos GR10159’’ respectively. Nuria Sánchez thanks Ministry of Education from Spain for FPU Grant received. References [1] Balat M, Balat H. Progress in biodiesel processing. Appl Energy 2010;87:1815–35. [2] Knothe G, Krahl J, Van Gerpen J. The biodiesel handbook. Champaign (IL): AOCS; 2005. [3] Chattopadhyay S, Sen R. Fuel properties, engine performance and environmental benefits of biodiesel produced by a green process. Appl Energy 2013;105:319–26. [4] Santori G, Di Nicola G, Moglie M, Polonara F. A review analyzing the industrial biodiesel production practice starting from vegetable oil refining. Appl Energy 2012;92:109–32. [5] Leung DYC, Wu X, Leung MKH. A review on biodiesel production using catalyzed transesterification. Appl Energy 2010;87:1083–95. [6] Atadashi IM, Aroua MK, Abdul Aziz AR, Sulaiman NMN. The effects of catalysts in biodiesel production: a review. J Ind Eng Chem 2013;19:14–26. [7] Gui MM, Lee KT, Bhatia S. Feasibility of edible oil vs. non-edible oil vs. waste edible oil as biodiesel feedstock. Energy 2008;33:1646–53. [8] Scholz V, da Silva JN. Prospects and risks of the use of castor oil as a fuel. Biomass Bioenergy 2008;32:95–100. [9] Berman P, Nizri S, Wiesman Z. Castor oil biodiesel and its blends as alternative fuel. Biomass Bioenergy 2011;35:2861–6. [10] Meneghetti SMP, Meneghetti MR, Serra TM, Barbosa DC, Wolf CR. Biodiesel production from vegetable oil mixtures: cottonseed, soybean, and castor oils. Energy Fuel 2007;21:3746–7. [11] Box G, Hunter S, Hunter W. Statistics for experimenters: design, innovation, and discovery. 2nd ed. Wiley-Interscience; 2005. [12] De Oliveira D, Di Luccio M, Faccio C, Dalla Rosa C, Bender JP, Lipke N, et al. Optimization of alkaline transesterification of soybean oil and castor oil for biodiesel production. Appl Biochem Biotech – Part A Enzyme Eng Biotechnol 2005;122:553–60. [13] de Lima da Silva N, Maciel M, Batistella C, Filho R. Optimization of biodiesel production from castor oil. Appl Biochem Biotechnol 2006;130:405–14. [14] Cavalcante KSB, Penha MNC, Mendonça KKM, Louzeiro HC, Vasconcelos ACS, Maciel AP, et al. Optimization of transesterification of castor oil with ethanol using a central composite rotatable design (CCRD). Fuel 2010;89:1172–6. [15] Barbosa DDC, Serra TM, Meneghetti SMP, Meneghetti MR. Biodiesel production by ethanolysis of mixed castor and soybean oils. Fuel 2010;89:3791–4. [16] Montoya J, Benjumea P, Pashova V. Optimization of the basic ethanolysis of ricin oil using the response surface methodology. Dyna 2011;168:90–7. [17] Meneghetti SMP, Meneghetti MR, Wolf CR, Silva EC, Lima GES, Silva LdL, et al. Biodiesel from castor oil: a comparison of ethanolysis versus methanolysis. Energy Fuel 2006;20:2262–5. [18] Canoira L, García Galeán J, Alcántara R, Lapuerta M, García-Contreras R. Fatty acid methyl esters (FAMEs) from castor oil: Production process assessment and synergistic effects in its properties. Renew Energy 2010;35:208–17. [19] Dias JM, Araújo JM, Costa JF, Alvim-Ferraz MCM, Almeida MF. Biodiesel production from raw castor oil. Energy 2013;53:58–66. [20] Encinar JM, González JF, Martínez G, Sánchez N, Pardal A. Soybean oil transesterification by the use of a microwave flow system. Fuel 2012;95: 386–93. [21] Bankovic´-Ilic´ IB, Stamenkovic´ OS, Veljkovic´ VB. Biodiesel production from nonedible plant oils. Renew Sust Energy Rev 2012;16:3621–47. [22] Canakci M, Van Gerpen J. Biodiesel production from oils and fats with high free fatty acids. Trans Am Soc Agric Eng 2001;44:1429–36. [23] Montoya J. Modelamiento y simulación de la cinética de transesterificación del aceite de ricino con alcohol etílico, catalizada con NaOH. Thesis. Universidad Nacional de Colombia; 2009. [24] Ma F, Hanna MA. Biodiesel production: a review. Bioresource Technol 1999;70:1–15.