Waste Biomass Valor (2013) 4:467–474 DOI 10.1007/s12649-013-9203-5
ORIGINAL PAPER
Optimization of Biodiesel Production from Iranian Bitter Almond Oil Using Statistical Approach Mehdi Atapour • Hamid-Reza Kariminia
Received: 24 July 2012 / Accepted: 24 January 2013 / Published online: 9 February 2013 Ó Springer Science+Business Media Dordrecht 2013
Abstract Response surface methodology (RSM) was applied to optimize the process of biodiesel production from Iranian bitter almond oil. Design of experiments was performed by application of a 5-level-3-factor central composite design in order to study the effect of different factors on the product yield, biodiesel yield and biodiesel purity. These factors were reaction temperature (30–70 °C), catalyst concentration (0.3–1.7 % w/w) and methanol to oil molar ratio (4.4–13.6 mol/mol). A quadratic model was suggested for the prediction of the biodiesel yield. Analysis of variance revealed that the factors were significant on the production process of biodiesel. For each factor, the optimum value was determined as: reaction temperature of 35 °C, catalyst (NaOH) concentration of 1.4 wt% and methanol to oil molar ratio of 9.7 mol/mol. At the optimal condition, the actual values of the product yield, biodiesel yield and biodiesel purity were 96.7, 94.7 and 97.9 wt%, respectively. At these conditions, the predicted values of the product yield, biodiesel yield and biodiesel purity were 98.1, 96.3 and 98.2 wt%, respectively. The fuel properties of the biodiesel were measured and compared with those of petroleum diesel, ASTM 6751 and EN 14214 biodiesel standards and a reasonable compatibility was observed. Keywords Biodiesel Bitter almond oil Statistical optimization Fuel properties
M. Atapour H.-R. Kariminia (&) Department of Chemical and Petroleum Engineering, Sharif University of Technology, P.O. Box 11155-9465, Azadi Ave., Tehran, Iran e-mail:
[email protected]
Introduction Biodiesel is an attractive substitute or extender for fossil diesel fuels mainly because of being produced from renewable biological sources [1], reducing air pollutants emissions [2], diminishing the greenhouse gas emissions intensity, being non-toxic, environmentally safe and biodegradable [3, 4]. Furthermore, it has fuel properties similar to fossil diesel [5]. There are several methods for production of biodiesel, wherein transesterification is the most common technique that includes the reaction of a lipid with an alcohol to form glycerol and fatty acid alkyl esters [6, 7]. The reaction can be categorized into non-catalytic and catalytic method in which either homogeneous or heterogeneous catalysis reactions occurs. However, the homogeneous catalysts are the most commonly used [8] and amongst these, alkaline catalysts such as sodium hydroxide have been used as a more effective type [9]. On the other hand, methanol is the most commonly used alcohol because of its physical properties and lower cost [10]. The transesterification reaction is influenced by many factors such as alcohol type, molar ratio of alcohol to oil, catalyst type, catalyst concentration, reaction temperature, reaction time, mixing rate, and properties of the feedstock [11, 12]. Various feedstocks have been used for biodiesel production. Virgin vegetable oils including edible oils and non-edible oils, waste vegetable oils, restaurant greases, animal fats, oil from algae, bacteria and fungi are among these feed stocks [9, 13, 14]. However, the main barrier for commercialization of biodiesel is the high production cost, which is mainly because of the high cost of the raw material [15]. Utilizing a less expensive feedstock such as non-edible oils can reduce the biodiesel production costs. Numbers of research studies have been conducted on the
123
468
utilization of non-edible oils for biodiesel production. This includes the application of Indesia polycarpa fruit oil [16], Jatropha curcas oil [17, 18], karanja (Pongamia pinnata) seed oil [19, 20], rubber seed oil [21], mahua oil [22] and okra seed oil [23], rapeseed oil deodorizer distillate [24], stillingia oil [25], Taramira (Eruca sativa L) seed oil [26] and so on. In these studies, the effects of operational parameters on the transesterification reaction have been studied and the properties of the biodiesel produced as fuel have been measured. These oils have been found to be potential sources for biodiesel production. There are also other non-edible oils such as bitter almond oil (BAO) that should be verified as a source for biodiesel production. The Almond tree (Prunus dulcis) is a family of Rosaceae and is classified in the subgenus Amygdalus of the genus Prunus. It is a species of native trees of the warmer parts of western Asia and of North Africa; however it has been extensively distributed over the warm temperate region of the old world. It is cultivated in different places including Northern Africa, Asia and Europe and US as well. Furthermore, Iran, Spain, Morocco, France, Greece, and Italy are the main almond producing countries [27]. Almond with different varieties is divided into two types that includes sweet almond, Prunus dulcis var. Dulcis and bitter almond, Prunus dulcis var. Amara. The bitter almond is slightly wider, shorter and less regular than the sweet almond and the bitter taste is its characteristic. The kernal of the bitter almond contains a colorless and crystalline glucoside, i.e. amygdalin (C20H27NO11). In addition, it contains an enzyme that after chewing, crushing or any other injury to the seed can decompose the soluble amygdalin into glucose, hydrocyanic acid (prussic) and benzaldehyde, in the presence of water. The yield of BAO obtained from the almond tree seed by pressing is approximately 40–45 % [27]. The BAO, which has a yellowish color and acrid taste, is non-edible and toxic due to the presence of prussic acid. Thus, it might be seriously considered as a candidate amongst the raw material for biodiesel production. In our previous study, we studied the feasibility of using BAO for biodiesel production as an alternative feedstock, for the first time [28]. Physical and chemical properties of the BAO were determined and the transesterification of BAO for the production of biodiesel was conducted using KOH as catalyst. We also investigated the effect of three factors including reaction temperature, catalyst concentration and methanol to oil molar ratio, on the biodiesel production from BAO. In the present work, we focused on obtaining optimal values of each above mentioned parameter involved in the improvement of the biodiesel yield using a statistical optimization approach. Effect of different factors on the product yield, biodiesel yield and biodiesel purity were investigated using central composite
123
Waste Biomass Valor (2013) 4:467–474
design (CCD). Furthermore, the fuel properties of the biodiesel produced from the BAO were determined.
Materials and Methods Materials The BAO was purchased from Kimia Pack Co, Tehran, Iran. The physical and chemical properties of the BOA are given somewhere else [28]. Methyl esters of analytical grade that were used as standards in the chromatographic analysis were obtained from Wako Pure Chemicals, Japan. Other solvents and chemicals including sodium hydroxide pellets (98 % purity), n-hexane (99 % purity), methanol (99.5 % purity), manganese sulfate powder (98 % purity) and hydrochloric acid (37 % purity) were purchased from Merck, Germany. Experimental Procedure The reactor used for biodiesel production was a 500 ml three-necked flask, equipped with a thermometer, condenser and magnetic stirrer. The temperature of this reactor containing 50 g of the BAO was controlled by inserting in a water bath that was heated by a hot plate. The reaction mixture was heated to a determined temperature under agitation condition. The catalyst that was previously dissolved in methanol, was added into the reactor, and the reaction began. After 60 min, stirring and heating were stopped and the reaction mixture was transferred into a separatory funnel, left for 1 h. Two distinct phases, i.e. ester phase (biodiesel) and glycerol phase were formed. This phase separation completed in approximately 10 min while the biodiesel layer appeared to be translucent. After 1 h, the transparent ester phase appeared and the separation was complete. There is a possibility of further reaction during the phase separation, but the reaction rate at this stage will be slow due to the lowered temperature, small amounts of catalyst, lack of stirring and reduced amount of the remained methanol. Nevertheless, a longer settling time is favorable for the finer separation [29, 30]. The glycerol phase that appeared in the lower phase was decanted. The biodiesel was washed with 35 ml distilled water (50 °C); subsequently, by 35 ml of 0.5 % hydrochloric acid, where the remaining catalyst was neutralized and the soaps formed during the transesterification reaction were decomposed. Finally, the resulting solution was washed with 35 ml of distilled water three times in order to remove impurities including methanol, residual catalyst, soap and glycerol. Thereafter, manganese sulfate powder was used to dry the biodiesel product. Then, the solution was filtered under
Waste Biomass Valor (2013) 4:467–474
469
vacuum conditions to remove any remained manganese sulfate crystals traces. The final product was weighted and analyzed afterwards by GC to determine biodiesel purity. It is worth mentioning that all the stages including production, settling, washing and purification steps were carried out under a fume cupboard for the sake of safety and environmental considerations. Analytical Methods All analytical methods for the measurement of acid and iodine values of the BAO, fatty acid composition of the BAO, densities of the BAO and the biodiesel, mean molecular weight of the oil, kinematic viscosity of the BAO and the biodiesel, biodiesel properties including pour point, flash point and cloud point, and fatty acid methyl ester composition of the purified biodiesel samples by gas chromatography were measured according to the procedure described in our previous work [28]. To measure the composition of fatty acid methyl esters in the purified biodiesel, a GC PERICHROM PR2100 gas chromatograph was used. It was equipped with a flame ionization detector and a capillary column (50 m 9 0.201 mm 9 0.50 lm) with HP-PONA stationary phase using helium as the carrier gas. Both the injector and detector temperatures were set at 250 °C. The column temperature was kept for 5 min at 120 °C, then, increased to 225 °C with a rate of 4 °C/min and held for 20 min at 225 °C. As much as 1 ll of the purified biodiesel samples that were previously dissolved in n-hexane were injected into the GC. The product yield and the biodiesel yield were calculated as follows: mproduct Product yield ¼ 100 ð1Þ mBAO Biodiesel yield ¼
mproduct purity of biodiesel mBAO
ð2Þ
where, mproduct is the mass of the purified product after washing and purification steps; mBAO is the mass of the BAO (50 g); and the purity of biodiesel was obtained by gas chromatography analysis. It is noteworthy that the purified product mainly includes fatty acid methyl esters. However, it may contain small amount of impurities such as mono-, di- and triglycerides, methanol, residual catalyst,
soap, glycerol and other impurities in the feedstock which may have not been eliminated completely despite washing and purification steps. Design of Experiments Design of experiments (DOE) was applied to investigate the influence of operating factors including the reaction temperature (x1), catalyst concentration (x2) and methanol to oil molar ratio (x3) on the product yield, biodiesel yield and biodiesel purity to obtain optimum conditions. These factors were selected according to preliminary experiments, our experience and previous literature [28] and other researchers’ suggestion [31]. Result of our previous work, imply the importance and necessity of adopting a more effective approach, i.e. the design of experiment to improve the yield, if possible [28]. Here, we utilized response surface methodology (RSM) coupled with central composite design (CCD) as a tool for optimization. RSM is a powerful method which applies a collection of mathematical and statistical techniques to study effects of independent variables on the value of dependent variable(s). By the use of RSM and designing of experiments an empirical model will be established [4, 32]. The CCD is a standard RSM design tool to achieve appropriate data without performing a lot of experiments. Design-Expert 7.1.4 (Stat-Ease, Inc.) software was employed for statistical analysis of experimental data. The DOE led to 20 runs with 6 axial, 8 factorial points and 6 center points. Table 1 illustrates the coded and real levels of the selected factors. The levels were chosen according to the properties of the materials and chemicals, experimental limitations, preliminary experiments and our previous experience. The experiments were performed based on the design matrix indicated in Table 2. All the experimental runs were carried out randomly to minimize systematic errors. The data obtained from the experiments were analyzed to develop a mathematical model; where, a quadratic model was suggested to relate the biodiesel yield as a response to the selected factors. General form of the model is given by Eq. (3). y ¼ a þ
3 X
a i xi þ
i¼1
2 X 3 X
aij xi xj þ
3 X
i¼1 j¼iþ1
aii x2i
ð3Þ
i¼1
Table 1 Coded and real levels of the selected factors in DOE Factors
Unit
Levels -a (-1.32)
-1
0
30
35
50
(x1)—Reaction temperature
°C
(x2)—Catalyst concentration
wt%
0.3
0.5
1.0
(x3)—Methanol to oil molar ratio
mol/mol
4.4
5.5
9
?1
?a (?1.32)
65
70
1.5
1.7
12.5
13.6
123
470
Waste Biomass Valor (2013) 4:467–474
Table 2 Central composite design matrix and experimental results Run
Factors
Product yield
Biodiesel purity
Biodiesel yield
E
P
E
P 91.2
x1
x2
x3
Ea
1
35
0.5
12.5
97.9
98.2
91.3
92.9
89.4
2
35
0.5
5.5
95.5
93.3
87.8
86.9
83.8
80.8
3 4
50 50
1.0 1.0
13.6 4.4
94.0 87.6
92.2 90.9
98.9 92.3
98.4 92.2
92.9 80.9
90.9 83.8
5
50
1.0
9.0
96.3
95.6
99.3
97.2
95.6
92.9
6
50
1.0
9.0
96.3
95.6
96.5
97.2
93.0
92.9
7
50
1.0
9.0
95.6
95.6
93.3
97.2
89.3
92.9
8
35
1.5
5.5
95.5
94.6
97.4
96.5
92.2
92.3
9
65
0.5
12.5
97.0
97.3
98.9
98.2
96.0
95.5
10
65
1.5
12.5
74.9
76.4
98.8
99.9
74.0
76.6
11
50
9.0
0.3
98.6
100.1
92.6
91.5
91.3
91.7
12
50
1.0
9.0
95.7
95.6
99.9
97.2
95.7
92.9
13
30
1.0
9.0
97.6
99.5
95.4
95.3
93.1
94.9
14
50
9.0
1.0
96.3
92.6
97.2
96.1
95.6
92.9
15
50
1.0
9.0
95.2
95.6
97.2
97.2
92.6
92.9
16
65
0.5
5.5
96.7
96.1
87.8
89.4
84.9
85.9
17
65
5.5
1.5
80.0
79.2
97.9
96.5
78.4
76.2
18 19
70 50
1.0 1.7
9.0 9.0
89.3 87.3
88.8 87.2
98.6 99.0
98.2 99.6
88.1 86.4
87.2 86.9
20
35
1.5
12.5
95.5
95.4
99.4
98.0
94.8
93.5
a
Experimental value
b
Predicted value
Pb
where, y represents the predicted biodiesel yield; xi and xj are the independent factors and a0, ai, aij, and aii stand for intercept, linear, interaction and quadratic constant coefficients, respectively.
Results and Discussion Statistical Analysis Various models including linear, two factor interactions, quadratic and cubic were examined sequentially and the quadratic model was suggested because of its low p value
(0.0075). The comparison of the different models according to their p value is given in Table 3. Analysis of variance (ANOVA) for the selected quadratic model is summarized in Table 4. As shown in the table, the model was significant at 1 % level. The ‘‘lack of fit test’’ that compares the residual error to the pure error obtained from replicated design points was not significant for the quadratic model (p value: 0.3235). The accuracy of fitting was evaluated by the R2 value. The values for R2 and the adjusted R2 were 0.8983 and 0.8067, respectively. In addition, it was observed that x1 (reaction temperature), x3 (methanol to oil molar ratio) and x1x2 were significant at 1 % level, while x2 (catalyst concentration), x2 x3 and x23 were significant at 5 % level. Therefore, the methanol to oil molar ratio and the reaction temperature were more influential parameters compared to the catalyst concentration at the given range for these values. The predicted values versus the actual (experimental) values of the biodiesel yield are illustrated in Fig. 1.The horizontal (y) and vertical (x) axes show the predicted and experimental values, respectively. In this graph the line (y = x) represents perfect fitness which the nearer experimental points to the line represent less difference with the predicted values. According to this figure, an acceptable agreement can be observed between experimental results and quadratic model predictions. The final model expressed in terms of actual factors is: y ¼ 8:581 þ 1:037x1 þ 60:259x2 þ 7:023x3 0:706x1 x2 0:004x1 x3 1:31x2 x3 0:005x21 8:433x22 0:264x23 ð4Þ Equation (4) can be represented in three-dimensional response surface plots. Figures 2, 3 and 4 illustrate the response surface of the product yield, biodiesel yield and biodiesel purity as a function of the methanol to oil molar ratio and the reaction temperature, respectively; while the catalyst concentration was held at zero coded level (1 wt%). Figures 2, 3 and 4 indicate that the product yield and biodiesel yield followed similar trends. At a given temperature, the product yield and biodiesel yield initially increased against the methanol to oil ratio until a maximum
Table 3 Comparison of different models Source
Sum of squares
Degree of freedom
Mean squares
F value
p value (Prob. [ F)
Linear versus mean
220.30
3
73.43
2.37
0.1090
2FI versus linear
266.66
3
88.89
5.04
0.0156*
Quadratic versus 2FI
156.36
3
52.12
7.15
0.0075**
42.57
4
10.64
2.11
0.1980
Cubic versus quadratic * Significant at 5 % level ** Significant at 1 % level
123
Waste Biomass Valor (2013) 4:467–474 Table 4 The ANOVA for the selected quadratic model
Source
Sum of squares
Model
* Significant at 5 % level ** Significant at 1 % level
Degree of freedom
Mean square
F value
p value Prob. [ F
643.33
9
71.48
9.81
0.0007**
x1—Reaction temperature
98.63
1
98.63
13.54
0.0043**
x2—Catalyst concentration
39.18
1
39.18
5.38
0.0429*
x3—Methanol to oil molar ratio
R2 = 0.8983, Adj. R2 = 0.8067
471
82.49
1
82.49
11.32
0.0072**
x1x2
224.30
1
224.30
30.78
0.0002**
x1x3
0.32
1
0.32
0.04
0.8382
x2x3
42.04
1
42.04
5.77
0.0372*
x21
8.53
1
8.53
1.17
0.3046
x22 x23
30.97
1
30.97
4.25
0.0662
73.01
1
73.01
10.02
0.0101*
Residual Lack of fit
72.87 44.18
10 5
7.29 8.84
1.54
0.3235
Pure error
28.68
5
5.74
Cor total
716.20
19
level was reached and the yields started to fall down. These results are consistent with our previous findings [28] and compatible with those of other researchers [6, 7, 33, 34]. Figure 3 shows, at low temperatures (30–40 °C), the effects of the methanol to oil ratio on the biodiesel purity were similar to the product yield and biodiesel yield. However, at higher temperatures, there was an increase in the biodiesel purity. It is important to note that the above mentioned trends of the product yield, biodiesel yield and biodiesel purity were sensitive to the catalyst concentration changes. For example, at 1.5 wt% of catalyst concentration and for the whole range of temperature tested, the trend of the biodiesel purity was similar to the product yield and biodiesel yield. In other
words, with increase in the methanol to oil ratio, in the first instance, the biodiesel purity increased, but then it decreased. We observed the same trend in our previous study [28]. Since the transesterification is a reversible reaction, alcohol to oil molar ratios higher than 3:1 (the stoichiometric molar ratio) will increase the product yield, biodiesel yield and biodiesel purity as the reaction equilibrium shifts to the right-hand side. However, using very large amounts of methanol increases the solubility of glycerol and causes difficulties to separate the glycerol from FAME layer. The remaining glycerol in ester layer would cause the biodiesel yield to decrease as the equilibrium backs to the left-hand side [33, 34]. Also, Figures 2, 3 and 4 show the effect of the reaction
Fig. 1 Predicted value versus experimental value of biodiesel yield
Fig. 2 Response surface plot of product yield versus methanol to oil molar ratio and reaction temperature (at 1 wt% for catalyst concentration)
123
472
Fig. 3 Response surface plot of biodiesel purity versus methanol to oil molar ratio and reaction temperature (at 1 wt% for catalyst concentration)
temperature on the biodiesel yield, product yield and biodiesel purity, respectively. From Figs. 2–4, it was observed that the product yield and biodiesel yield decreased with increasing the temperature for any methanol to oil ratio. The response surface in Fig. 3 shows that the biodiesel purity remained nearly constant at low methanol to oil ratios. However, at higher methanol to oil ratios the biodiesel purity enhanced when the reaction temperature increased. Figure 5 shows the response surface of the biodiesel yield for the selected range of the methanol to oil molar ratio and the catalyst concentration while the temperature was held at zero coded level (50 °C). As shown in the figure, increasing the catalyst concentration, at a given methanol to oil ratio, causes the biodiesel yield to increase to a maximum level. However, for higher catalyst concentration, the biodiesel yield decreases. The results are in accordance with those of other studies [6, 7, 34, 35]. In fact, an excessive amount of catalyst leads to an increase in the amount of soaps produced through saponification reaction. The soaps due to their polarity give rise to the solution of the methyl ester in glycerol. Also, the soaps produce an emulsion that makes the glycerol separation difficult during the washing stage. The ester loss increases and the biodiesel yield decreases as a consequence [36]. The saponification is a side reaction that occurs when an alkali-catalyzed transesterification is used for the biodiesel production. During this undesirable reaction, the alkali catalyst reacts with the free fatty acids (FFA) and produces soap [37]. In this study, due to low FFA content (0.24 mg KOH/g), the amount of produced soaps was not considerable
123
Waste Biomass Valor (2013) 4:467–474
Fig. 4 Response surface plot of biodiesel yield versus methanol to oil molar ratio and reaction temperature (at 1 wt% for catalyst concentration)
at low NaOH concentrations. However, as mentioned above, high catalyst concentrations gave rise to the formation of soaps. Optimization of the Reaction Conditions and Model Validation The optimal conditions for the selected variables including the reaction temperature, catalyst concentration and methanol to oil molar ratio were obtained utilizing numerical
Fig. 5 Response surface plot of biodiesel yield versus methanol to oil molar ratio and catalyst concentration (at temperature of 50 °C)
Waste Biomass Valor (2013) 4:467–474
473
Table 5 Physical and chemical properties of the BAO biodiesel and petroleum diesel Property
This study
Mahua biodiesel [22]
Jatropha biodiesel [38]
Karanja biodiesel [39]
Rubber seed biodiesel [40]
Peanut biodiesel [41]
Petroleum diesel [29]
EN 14214
ASTM 6751–02
Density @ 15 °C, g/cm3
0.884
0.880
0.880
0.876
0.874
0.918
0.820–0.860
0.86–0.90
0.87–0.90
Viscosity @ 40 °C, mm2/s
4.51
3.98
4.84
9.6
5.81
5.908
2.5–3.5
3.5–5.0
1.9–6.0
Flash point, °C Cloud point, °C
170 11
208 NA
192 NA
187 NA
130 4
192 6
[55 -16
[120 NA
[130 -3–12
Pour point, °C
-2
6
6
7
-8
3
-33
NA
-15–10
NA not available
optimization method provided in the Design-Expert software. Optimization criteria were set for all variables including independent variables and the response. The goal of the optimization for the response was to maximize biodiesel yield. Its lower and upper limits were set to the lowest observed value (74.0) and the highest theoretical value (100), respectively. Independent variables (x1, x2, x3) were set in a range given in Table 1 and coded as -1 and ?1. The optimum value for each variable was found as: the reaction temperature of 35 °C, the NaOH concentration of 1.4 wt%, the methanol to oil molar ratio of 9.7 mol/mol. For evaluating the accuracy of the selected variables, an experiment was conducted for three times under the optimal conditions. At these conditions, the mean actual values of the product yield, biodiesel yield and biodiesel purity were 96.71, 94.7 and 97.92 wt%, respectively. Also, the predicted values of the product yield, biodiesel yield and biodiesel purity were 98.09, 96.27 and 98.22 wt%, respectively. The difference in the product yield, biodiesel yield and biodiesel purity between the experimental and predicted values was 1.38, 0.3 and 1.57 wt%, respectively. Therefore, the actual values obtained from the experiments were in acceptable agreement with the predicted values, which the errors between two values were small; for example, \1.7 % error for the biodiesel yield. Biodiesel Properties Properties of the BAO biodiesel including kinematic viscosity, flash point, pour point and cloud point are exhibited in Table 5. As shown in the table, the properties are comparable with those from several other studies conducted for production of biodiesel using edible and nonedible oils [22, 38–41]. Properties of the BOA biodiesel were compared with petroleum diesel [29], ASTM 6751 and EN 14214 biodiesel standards. Comparing the properties of the produced biodiesel with petroleum diesel shows that the values for kinematic viscosity and density are relatively close. The value of flash point for the BAO biodiesel was 170 °C. This is a relatively high value in
favor of higher safety or transportability purpose against petroleum diesel. On the other hand, the cloud point and pour point of the biodiesel were considerably higher than petroleum diesel. As a result, this biodiesel is less suitable to be used during winter time. Generally, the properties of the biodiesel produced from BAO are compatible with those of ASTM 6751 and EN standards.
Conclusion Optimal conditions for biodiesel production from bitter almond oil by the application of design of experiments using response surface methodology were investigated. A quadratic model was suggested for the prediction of biodiesel yield. The R2 value was 0.8983 that indicated an acceptable fitting to the experimental data. The variance analysis of the model proved that the reaction temperature, the methanol to oil molar ratio and the NaOH concentration were significant factors. The optimum value for each factor was found as: reaction temperature of 35 °C, NaOH concentration of 1.4 wt%, methanol to oil molar ratio of 9.7 mol/mol. At the optimal conditions, the actual values of the product yield, biodiesel yield and biodiesel purity were 96.71, 94.7 and 97.92 wt%, respectively. At these conditions, the predicted values of the product yield, biodiesel yield and biodiesel purity were 98.09, 96.27 and 98.22 wt%, respectively. The fuel properties of the BAO biodiesel produced were comparable with those of other studies and conformed to EN 14214 and ASTM 6751 standards. Acknowledgments Authors wish to thank the research office of Sharif University of Technology for the financial support.
References 1. Naik, S.N., Goud, V.V., Rout, P.K., Dala, A.K.: Production of first and second generation biofuels: A comprehensive review. Renew. Sust. Energ. Rev. 14, 578–597 (2010)
123
474 2. Devan, P.K., Mahalakshmi, N.V.: A study of the performance, emission and combustion characteristics of a compression ignition engine using methyl ester of paradise oil–eucalyptus oil blends. Appl. Energy 86, 675–680 (2009) 3. Basha, S.A., Raja, G.K., Jebaraj, S.: A review on biodiesel production, combustion, emissions and performance. Renew. Sust. Energy Rev. 13, 1628–1634 (2009) 4. Vicente, G., Coteron, A., Martinez, M., Aracil, J.: Application of the factorial design of experiments and response surface methodology to optimize biodiesel production. Ind. Crops Prod. 8, 29–35 (1998) 5. Leduc, S., Natarajan, K., Dotzauer, E., McCallum, I., Obersteiner, M.: Optimizing biodiesel production in India. Appl. Energy 86, S125–S131 (2009) 6. Kafuku, G., Mbarawa, M.: Alkaline catalyzed biodiesel production from Moringaoleifera oil with optimized production parameters. Appl. Energy 87, 2561–2565 (2010) 7. Lin, L., Ying, D., Chaitep, S., Vittayapadung, S.: Biodiesel production from crude rice bran oil and properties as fuel. Appl. Energy 86, 681–688 (2009) 8. Kansedo, J., Lee, K.T., Bhatia, S.: Biodiesel production from palm oil via heterogeneous transesterification. Biomass Bioenerg 33, 271–276 (2009) 9. Ma, F., Hanna, M.A.: Biodiesel production: a review. Bioresour. Technol. 70, 1–15 (1999) 10. Berrios, M., Gutie´rrez, M.C., Martı´n, M.A., Martı´n, A.: Application of the factorial design of experiments to biodiesel production from lard. Fuel Process. Technol. 90, 1447–1451 (2009) 11. Dias, J.M., Alvim-Ferraz, M.C.M., Almeida, M.F.: Comparison of the performance of different homogeneous alkali catalysts during transesterification of waste and virgin oils and evaluation of biodiesel quality. Fuel 87, 3572–3578 (2008) 12. Gui, M.M., Lee, K.T., Bhatia, S.: Supercritical ethanol technology for the production of biodiesel: Process optimization studies. J. Supercrit. Fluids 49, 286–292 (2009) 13. Balat, M., Balat, H.: Progress in biodiesel processing. Appl. Energy 87, 1815–1835 (2010) 14. Diya’uddeen, B.H., Abdul Aziz, A.R., Daud, W.M.A.W., Chakrabarti, M.H.: Performance evaluation of biodiesel from used domestic waste oils: A review. Process Saf. Environ. 90, 164–179 (2012) 15. Demirbas, A.: Political, economic and environmental impacts of biodiesels: A review. Appl. Energy 86, S108–S117 (2009) 16. Yanga, F.X., Sua, Y.Q., Li, X.H., Zhanga, Q., Sun, R.C.: Preparation of biodiesel from Idesia polycarpa var. vestita fruit oil. Ind. Crops Prod. 29, 622–628 (2009) 17. Berchmans, H.J., Hirata, S.: Biodiesel production from crude Jatropha curcas L. seed oil with a high content of free fatty acids. Bioresour. Technol. 99, 1716–1721 (2008) 18. Rashid, U., Anwar, F., Jamil, A., Bhatti, H.N.: Jatropha curcas seed oil as a viable source for biodiesel. Pak. J. Bot. 42, 575–582 (2010) 19. Meher, L.C., Dharmagadda, V.S.S., Naik, S.N.: Optimization of alkali-catalyzed transesterification of Pongamia pinnata oil for production of biodiesel. Bioresour. Technol. 97, 1392–1397 (2006) 20. Ahmad, M., Zafar, M., Khan, M.A., Sultana, S.: Pongamia pinnataas a biodiesel resource in Pakistan. Energy Sour Part A 31, 1436–1442 (2009) 21. Ikwuagwu, O.E., Ononogbu, I.C., Njoku, O.U.: Production of biodiesel using rubber [Heveabra siliensis (Kunth. Muell)] seed oil. Ind. Crops Prod. 12, 57–62 (2000)
123
Waste Biomass Valor (2013) 4:467–474 22. Ghadge, S.V., Raheman, H.: Process optimization for biodiesel production from mahua (Madhuca indica) oil using response surface methodology. Bioresour. Technol. 97, 379–384 (2006) 23. Anwar, F., Rashid, U., Ashraf, M., Nadeem, M.: Okra (Hibiscus esculentus) seed oil for biodiesel production. Appl. Energ. 87, 779–785 (2010) 24. Liu, Y., Wang, L., Yan, Y.: Biodiesel synthesis combining preesterification with alkali catalyzed process from rapeseed oil deodorizer distillate. Fuel Process. Technol. 90, 857–862 (2009) 25. Liu, Y., Xin, H.L., Yan, Y.J.: Physicochemical properties of stillingia oil: Feasibility for biodiesel production by enzyme transesterification. Ind. Crops Prod. 30, 431–436 (2009) 26. Chakrabarti, M.H., Ali, M., Baroutian, S., Saleem, M.: Technoeconomic comparison between B10 of Eruca sativa L. and other indigenous seed oils in Pakistan. Process Saf. Environ. 89, 165–171 (2011) 27. Dorado, M.P.: Raw materials to produce low-cost biodiesel. In: Nag A. (ed.) Biofuels refining and Performance, McGraw-Hill, New York (2008) 28. Atapour, M., Kariminia, H.-R.: Characterization and transesterification of Iranian bitter almond oil for biodiesel production. Appl. Energy 88, 2377–2381 (2011) 29. Kaya, C., Hamamci, C., Baysal, A., Akba, O., Erdogan, S., Saydut, A.: Methyl ester of peanut (Arachis hypogea L.) seed oil as a potential feedstock for biodiesel production. Renew. Energy 34, 1257–1260 (2009) 30. Leung, D.Y.C., Guo, Y.: Transesterification of neat and used frying oil: Optimization for biodiesel production. Fuel Process. Technol. 87, 883–890 (2006) 31. Banerjee, A., Chakraborty, R.: Parametric sensitivity in transesterification of waste cooking oil for biodiesel production-A review. Resour. Conserv. Recy. 53, 490–597 (2009) 32. Liu, Y., Yan, Y., Hu, F., Yao, A., Wang, Z., Wei, F.: Transesterification for Biodiesel Production Catalyzed by Combined Lipases: Optimization and Kinetics. AIChE J. 56, 1659–1665 (2010) 33. Murugesan, A., Umarani, C., Chinnusamy, T.R., Krishnan, M., Subramanian, R., Neduzchezhain, N.: Production and analysis of bio-diesel from non-edible oils-A review. Renew. Sust. Energy Rev. 13, 825–834 (2009) 34. Phan, A.N., Phan, T.M.: Biodiesel production from waste cooking oils. Fuel 87, 3490–3496 (2008) 35. Bhatti, H.N., Hanif, M.A., Qasim, M.: Ata-ur-Rehman.: Biodiesel production from waste tallow. Fuel 87, 2961–2966 (2008) 36. Encinar, J.M., Gonzalez, J.F., Rodriguez-Reinares, A.: Biodiesel from used frying oil. Variables affecting the yields and characteristics of the biodiesel. Ind. Eng. Chem. Res. 44, 5491–5499 (2005) 37. Leung, D.Y.C., Wu, X., Leung, M.K.H.: A review on biodiesel production using catalyzed transesterification. Appl. Energy 87, 1083–1095 (2010) 38. Becker, K., Francis, G.: Biodiesel from Jatropha plantations on degraded land. A brief technical report, University of Hohenheim (2002) 39. Raheman, H., Phadatare, A.G.: Diesel engine emissions and performance from blends of Karanja methyl ester and diesel. Biomass Bioenerg. 27, 393–397 (2004) 40. Ramadhas, A.S., Jayaraj, S., Muraleedharan, C.: Biodiesel production from high FFA rubber seed oil. Fuel 84, 335–340 (2005) 41. Ahmad, M., Rashid, S., Khan, M.A., Zafar, M., Sultana, S., Gulzar, S.: Optimization of base catalyzed transesterification of peanut oil biodiesel. Afr. J. Biotech. 8, 441–446 (2009)
