Chem. Eng. Technol. 2008, 31, No.

7, 993999

993

Jibrail Kansedo1 Keat Teong Lee1 Subhash Bhatia1

1

Research Article

School of Chemical Engineering, Universiti Sains Malaysia, Pulau Pinang, Malaysia.

Feasibility of Palm Oil as the Feedstock for Biodiesel Production via Heterogeneous Transesterification
The production of biodiesel has become popular recently as a result of increasing demand for a clean, safe and renewable energy. Biodiesel is made from natural renewable sources such as vegetable oils and animal fats. The conventional method of producing biodiesel is by reacting vegetable oil with alcohol in the presence of a homogenous catalyst (NaOH). However, this conventional method has some limitations such as the formation of soap, usage of significant quantities of wash water and complicated separation processes. Heterogeneous processes using solid catalysts have significant advantages over homogenous methods. Currently, more than 90 % of world biodiesel is produced using rapeseed oil. The production of biodiesel from rapeseed oil is considered uneconomical, considering the fact that palm oil is currently the worlds cheapest vegetable oil. Therefore, the focus of this study is to show the feasibility of producing biodiesel from palm oil using montmorillonite KSF as a heterogeneous catalyst. The heterogeneous transesterification process was studied using design of experiment (DOE), specifically response surface methodology (RSM) based on a four-variable central composite design (CCD) with a = 2. The transesterification process variables were reaction temperature, x1 (50190 C), reaction period, x2 (60300 min), methanol/oil ratio, x3 (412 mol mol1) and the amount of catalyst, x4 (15 wt %). It was found that the conversion of palm oil to biodiesel can reach up to 78.7 % using the following reaction conditions: reaction temperature of 155 C, reaction period of 120 min, ratio of methanol/oil at 10:1 mol mol1 and amount of catalyst at 4 wt %. From this study, it was shown that montmorillonite KSF catalyst can be used as a solid catalyst for biodiesel production from palm oil.
Keywords: Biodiesel, Heterogeneous catalysts, Transesterification Received: December 06, 2007; accepted: January 30, 2008 DOI: 10.1002/ceat.200700473

Introduction

Due to escalating crude petroleum oil prices and the effects of global warming, the search for alternative fuels has gained significant attention over the years. Among the various sources of alternative energy, renewable energy from biomass presents a promising alternative to replace some of the conventional fossil fuels. In recent years, a number of studies have shown that vegetable oils such as rapeseed [1], soybean [29] and sun-

Correspondence: Dr. K. T. Lee (chktlee@eng.usm.my), School of Chemical Engineering, Universiti Sains Malaysia, Engineering Campus, Seri Ampangan, 14300 Nibong Tebal, Pulau Pinang, Malaysia.

flower seed [10, 11] oil can be used as sources of biodiesel replacing conventional petroleum-derived diesel fuels. The biodiesel is produced through a transesterification reaction where the triglycerides present in vegetable oil are converted to fatty acid alkyl esters. Transesterification, also called alcoholysis, is the reaction of the triglycerides in the vegetable oil with alcohol to form a mixture of fatty acid alkyl esters and glycerol [1216]. The fatty acid alkyl ester produced from this process is known as biodiesel, which has become more attractive recently because it has almost the same properties and characteristics of petroleum-derived diesel. Several studies have shown that biodiesel produced from vegetable oils have a viscosity close to petroleum-derived diesel fuel [17, 18]. Their volumetric heating val-

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com

994

J. Kansedo et al.

Chem. Eng. Technol. 2008, 31, No. 7, 993999

ues are a little lower compared to the petroleum-derived diesel 2 Materials and Methods fuel, but they have high cetane and flash point values. The transesterification process is catalyzed by catalysts with 2.1 Materials both, acidic and basic sites. Currently, all the transesterification processes operate via a homogenous transesterification process Purified palm oil was purchased from Yee Lee Edible Oils Sdn mainly using potassium hydroxide as the catalyst [1216]. In Bhd, Malaysia and methanol from R & M Chemicals, UK. the homogeneous transesterification process, the catalysts used Montmorillonite KSF, methyl heptadecanoate (internal stanare in the same phase as the reactants/products. However, dard) and standard references for methyl esters analysis, i.e., homogeneous transesterification processes using potassium methyl myristate, methyl palmitate, methyl stearate, methyl hydroxide present certain disadvantages. These include the foroleate and methyl linoleate were obtained from Fluka Chemie, mation of soap in the product mixture leading to additional Germany. The catalyst, montmorillonite KSF, has a water concosts for the separation of soap from the biodiesel. In addition, tent of 12 %, free acid content of 812 %, and has its surface the formation of soap has also led to the loss of triglycerides partially loaded with H2SO4. The properties of palm oil are molecules that can be used to form biodiesel, resulting in a loss summarized in the Tab. 1 [28]. in yield. Since the catalyst and the reactants/products are in the same phase, the separation of the biodiesel product from Table 1. Properties of palm oil. the catalyst becomes complex. These disadvantages make the current homogeneous transFatty acids Percentage (%) esterification process expensive, complex and uneconomical. Lauric, C-12:0 0.1 In order to overcome these limitations, researchers are now looking for alternative routes to produce biodiesel from vegeMyristic, C-14:0 1.0 table oils. One of the possible solutions is by using a heterogePalmitic, C-16:0 42.8 neous transesterification process whereby solid catalyst is used. Recent research has shown that there are various types of cataStearic, C-18:0 4.5 lyst that can be used in the heterogeneous transesterification Oleic, C-18:1 40.5 process. Jitputti et al. [19] studied the transesterification of coconut oil and palm kernel oil using zirconium oxide-based and Linoleic, C-18:2 10.1 zinc oxide-based catalysts, while Suppes et al. [2] used zeolite Linolenic, C-18:3 0.2 and metal catalysts for the conversion of soybean oil to biodieOther/Unknown 0.8 sel. In other research, heterogeneous base [20], Na/NaOH/ Al2O3 [21] and metal complexes [3] have been used as the catalyst. Currently more than 90 % of world biodiesel is produced 2.2 Design of Experiments from rapeseed oil. However, due to the high price of rapeseed oil, the use of biodiesel for replacing conventional petroleumThe experimental design selected for this study is a Central derived diesel oil can only be applied in developed countries, Composite Design (CCD) that helps in investigating linear, e.g., in Europe and the USA. In order to overcome this probquadratic, cubic and cross-product effects of the four transeslem, researchers have been trying to source other types of vegeterification process variables (independent) on the conversion table oils that are cheaper, e.g., palm oil [22], jatropha oil of palm oil to biodiesel (response). The four transesterification [23, 24], soybean oil [39], canola oil [25] and even waste process variables studied are reaction temperature, reaction cooking oil [26, 27]. Since palm oil is the worlds cheapest veperiod, ratio of oil/methanol and amount of catalyst. Tab. 2 getable oil, it presents a promising alternative as a feedstock lists the range and levels of the four independent variables for biodiesel production as a possible replacement for rapeseed studied. The CCD comprises a two-level small factorial design oil. Nevertheless, the research knowledge on converting palm (12 24 = 8), eight axial or star points and five center points. oil to biodiesel using heterogeneous reactions is still very limThe value of a for this CCD is fixed at 2. The complete design ited. Thus, in this study, the feasibility of producing biodiesel from palm oil using montmorillonite Table 2. Levels of the transesterification process variables chosen for this study. KSF as the heterogeneous catalyst will be presented. Statistical design of experiments will be Variable Coding Unit Levels used to accumulate and analyze information on 2 1 0 +1 +2 the effect of process variables on the rapid and efficient conversion of palm oil to biodiesel using the Reaction temperature x1 C 50 80 120 155 190 minimum number of experiments. It is illustrated Reaction period x2 min 60 120 180 240 300 that this method is found to be far superior to the 1 conventional research methods involving the study Ratio of oil/methanol x3 mol mol 1:4 1:6 1:8 1:10 1:12 of one variable at one time while keeping the re% 1 2 3 4 5 Amount of catalyst x4 mainder constant.

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com

Chem. Eng. Technol. 2008, 31, No. 7, 993999

Heterogeneous transesterification

995

matrix of the experiments employed and results are given in Tab. 3. All variables at zero level constitute the center points and the combination of each of the variables at either their lowest (2.0) level or highest (+2.0) level with the other variables at zero level constitute the axial points. The experimental sequence was randomized in order to minimize the effects of the uncontrolled factors. Each response of the transesterification process was used to develop a mathematical model that correlates the conversion of palm oil to the transesterification process variables studied through first-order, second-order and interaction terms, according to the following second-order polynomial expression, Eq. (1): Y = bo +
4 P j1

the offset term, bj is the linear effect, bij is the first-order interaction effect and bjj is the squared effect.

2.3

Model Fitting and Statistical Analysis

Design Expert software version 6.0.6 (STAT-EASE Inc., Minneapolis, USA) was used for regression analysis of the experimental data to fit the second-order polynomial equation and also for evaluation of the statistical significance of the equation developed.

bj xj +

4 P

Transesterification reactions were carried out in a low pressure batch reactor with a magnetic stirrer, as shown in Fig. 1. A mixture of palm oil, catalyst and methanol was charged into the reactor. The mixing intensity of the magnetic stirrer was set at 190 200 rpm. The reaction temperature, Table 3. Experimental design matrix and results. duration, ratio of oil/methanol and amount of catalyst (montmorillonite Catalyst Code Experimental variables Conversion % KSF) were set according to the values as proposed in the DOE listed in Reaction tempe- Reaction period Ratio of oil/ Amount of rature methanol catalyst Tab. 3. The excess methanol from the samples was removed using a rotary C 1 ( ) (min) (mol mol ) (% g) evaporator. The upper layer of the K01 +1 +1 +1 1 71.63 sample was separated from the bottom layer and was analyzed to detect bioK02 +1 +1 1 1 59.10 diesel content. where Y is the predicted conversion of palm oil to biodiesel in mol mol1, xi and xj represent the variables or parameters, bo is
K03 K04 K05 K06 K07 K08 K09 K10 K11 K12 K13 K14 K15 K16 K17 K18 K19 K20 K21 +1 1 +1 1 1 1 2 +2 0 0 0 0 0 0 0 0 0 0 0 1 +1 1 1 +1 1 0 0 2 +2 0 0 0 0 0 0 0 0 0 +1 1 1 +1 +1 1 0 0 0 0 2 +2 0 0 0 0 0 0 0 +1 +1 +1 1 +1 1 0 0 0 0 0 0 2 +2 0 0 0 0 0 78.69 5.07 55.10 3.92 5.43 2.90 0.79 74.34 14.49 60.05 36.52 51.39 32.54 61.98 55.07 50.88 55.75 55.07 55.75

ij1

bij xi xj +

4 P

j1

bjj xj2

(1)

2.4

Activity Study

2.5

Analysis

The resulting products from the transesterification processes were analyzed using Gas Chromatography (GC) by means of an Inert Cap WAX capillary column (30 mm 0.25 mm, I.D. 0.25 lm) to identify the presence of methyl esters (biodiesel) in the sample. Helium was used as the carrier gas. An oven temperature of 120 C was initially held for 1 min and then increased to 220 C (held for 15 min) at a rate of 4 C min1. The temperatures of the injector and detector were set at 220 C and 250 C, respectively. A quantity of 1 lL from each sample was injected into the column. Methyl heptadecanoate was used as an internal standard for the calculation of the conversion. The conversions of the transesterification processes were calculated as the mol. ratio of methyl esters produced to palm oil used divided by 3. The formula is given as Eq. (2):

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com

996

J. Kansedo et al.

Chem. Eng. Technol. 2008, 31, No. 7, 993999

antagonistic effect. Eq. (3) shows that the conversion of palm oil to biodiesel has a linear and quadratic effect on the four transesterification process variables studied. In addition, interactions between the variables also effect the conversion of palm oil to biodiesel.

3.2

Model Adequacy Check

The quality of the model developed can be evaluated from the coefficients of correlation. The value of R for Eq. (3) is 0.9846. The high value of R (very close to 1) shows that there is very good agreement between the experimental and predicted values from the model. On the other hand the value of R2 for Eq. (3) is 0.9515. This implies that 95.15 % of Figure 1. Schematic diagram of the experimental setup for the transesterification the total variation in the conversion of palm oil to process. biodiesel responses can be attributed to the experimental variables studied. The adequacy of the model was further checked with analysis of variance (ANOVA) Conversion % = [(Total mol. of methyl esters)/(3 Total mol. as shown in Tab. 4. Based on a 95 % confidence level, the of oil)] 100 % (2) model was tested to be significant since the computed F value (42.67) is much higher than the theoretical F0.05 (12,8) value (3.28), indicating that the regression model is reliable for the 3 Results and Discussion prediction of the conversion of palm oil to biodiesel. Apart from that, each term in the model was also shown to be significant at a 95 % confidence level, since the computed F values 3.1 Development of Regression Model Equation for the respective terms are higher than the theoretical F0.05 A central composite design (CCD) was used to develop a cor(1,8) value (5.32). From these statistical tests, it was found that relation between the transesterification reaction variables to the model is adequate for predicting the conversion of palm the conversion of palm oil to biodiesel. The complete design oil to biodiesel within the ranges of variables studied. matrix and conversion of palm oil to biodiesel (responses) are Fig. 2 shows the experimental values versus predicted values listed in Tab. 3. It can be seen from Tab. 3 that the conversion using the model equation developed. A line of unit slope, i.e., of palm oil to biodiesel ranges from 0.7978.69 %. Runs K17 the line of perfect fit with points corresponding to zero error to K21 at the center point of the design are to determine the between experimental and predicted values, is also shown in experimental error. As a result of the conversion, these five Fig. 2. Therefore, this plot visualizes the performance of the runs were quite consistent, and only single replicate experimodel in an obvious way. The results in Fig. 2 demonstrates mental runs are essential in this study. Apart from the results that the regression model equation provides a very accurate of these five runs, the experimental error for the conversion of description of the experimental data, indicating that it was palm oil to biodiesel was calculated as 1.2 %. By using multiple regression analysis, the response (conversion of palm oil to biodiesel) obtained in Tab. 3 was correlated with the four transesterification reaction variables or parameters studied using the polynomial expression in Eq. (1). The coefficients of the full regression model equation and their statistical significance were determined and evaluated using Design-Expert 6.0.6 software. The final equation in terms of the actual value after excluding the insignificant terms (identified using Fishers Test) is outlined in Eq. (3): Y = 54.50 + 18.39 x1 + 11.39 x2 + 4.20 x3 + 7.36 x3 5.02 x12 5.09 x22 3.42 x32 2.60 x42 + 6.52 x1 x2 + 4.34 x1 x3 + 11.31 x1 x4 12.51 x2 x4 (3) A positive sign in front of the terms indicates a synergistic effect, while a negative sign indicates an

Figure 2. Experimental versus predicted conversion of palm oil to biodiesel.

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com

Chem. Eng. Technol. 2008, 31, No. 7, 993999

Heterogeneous transesterification

997

successful in capturing the correlation between the four transesterification process variables to the conversion of palm oil to biodiesel.

3.3

Effects of Transesterification Process Variables

From Tab. 4, it can be observed that among the four individual variables studied, reaction temperature (x1) has the largest effect on the conversion of palm oil to biodiesel (it has the highest F value) followed by reaction period (x2). The ratio of oil/ methanol (x3) and amount of catalyst (x4) have similar least significant effects on the conversion of palm oil to biodiesel. The quadratic terms of (x1) and (x2) were fairly significant for the conversion of palm oil to biodiesel compared to those of (x3) and (x4). The other significant terms include the effect of interaction between the variables, particularly between variables (x2) and (x4) and between variables (x1) and (x4). Longer reaction periods allow sufficient time for the transesterification process to occur completely, or at least, until it reaches its maximum level. A higher reaction temperature allows transesterification reactions to occur at an accelerated rate since there is a greater supply of energy to the reactions. On the other hand, a larger amount of catalyst used will result in higher conversion of triglycerides to biodiesel because of higher availability of active sites. Apart from that, a higher ratio of oil/ methanol of 1:10 gives higher conversion since excess methanol is required to push the transesterification reaction forward. All of these results are in agreement with those reported in the literature [13]. The results in Tab. 3 show that the transesterification process variables have a significant effect on the conversion of

Table 4. Analysis of variance (ANOVA) for the regression model equation and coefficients.
Source Model x1 x2 x3 x4 x1 x2 x3 x4
2 2 2 2

Sum of squares 13390.31 2704.80 1037.86 282.58 433.36 632.50 651.22 294.00 169.18 169.91 150.86 511.90 626.25 209.21

Degrees of freedom 12 1 1 1 1 1 1 1 1 1 1 1 1 8

Mean of squares 1115.86 2704.80 1037.86 282.58 433.36 632.50 651.22 294.00 169.18 169.91 150.86 511.89 626.25 26.15

F-test 42.67 103.43 39.69 10.81 16.57 24.19 24.90 11.24 6.47 6.50 5.77 19.57 23.95

x1 x2 x1 x3 x1 x4 x2 x4 Residual

palm oil to biodiesel. The highest conversion was achieved at a reaction temperature of 155 C, reaction period of 240 min or 4 h, oil/methanol ratio of 1:10 and an amount of catalyst of 4 wt %. In contrast, the lowest conversion was obtained at a reaction temperature of 50 C, reaction period of 3 h, oil/methanol ratio of 1:8 and an amount of catalyst of 3 wt %. These results illustrate that there is a significant possibility for improving the conversion of palm oil to biodiesel with proper selection of transesterification process variables using montmorillonite KSF as the catalyst. The ability of montmorillonite KSF to catalyze the transesterification reaction, at any rate, is due to the acidic properties of the catalyst itself. Montmorillonite KSF has a free acid content of 812 %, and its surface is partially loaded with H2SO4. H2SO4 is known to be a typical strong acid, and therefore, since the transesterification reactions can be catalyzed with acid or base, the strong acidity of this montmorillonite KSF will definitely contribute to the high conversion of palm oil to biodiesel. However, since the maximum conversion occurs at the border of the experimental variables selected for this study, the optimum experimental conditions to obtain the highest conversion cannot be derived (except for the experimental conditions at the border of the design variables). This finding was also supported by the model developed using the DOE software, whereby no maximum conversion was located within the experimental domain examined. Therefore, the model developed by the DOE software was used to facilitate a straightforward examination of the effects of the variables and their interactions towards the conversion of the transesterification process. Among the four interaction terms that were found to have significant effects on the conversion of palm oil to biodiesel, emphasis was placed on the interactions between (x2) and (x4) and between variables (x1) and (x4). Fig. 3 shows the changes in conversion with varying reaction temperature at 2 wt % and 4 wt % catalyst. The other two process variables, i.e., reaction period and ratio of oil/methanol were kept constant at 180 min and 1:8, respectively. It is generally perceived that a higher percentage of catalyst used will result in higher conversion of triglycerides to biodiesel because of the higher availability of active sites. However, this perception may not always be true. By referring to Fig. 3, the conversion of oil to biodiesel was found to be lower at 4 wt % catalyst as compared to 2 wt % catalyst when the reaction temperature was lower than 95 C. This may suggest the existence of real interactions between the supplied energy (heat) and the active sites on the catalyst itself. The active sites on the catalyst may require a certain amount of energy to be activated. At lower temperatures, with a smaller percentage of catalyst, i.e., 2 wt %, the activation energy will be reached more quickly than for the 4 wt % catalyst, since the ratio of active sites to energy (heat) is smaller. In other words, the total energy required to activate all of the active sites on the 2 wt % catalyst is much smaller than for the 4 wt % catalyst. Thus, lower temperatures (smaller amounts of energy) may be just enough to activate the entire number of active sites on the 2 wt % catalyst but not for the 4 wt % catalyst, due to the assumption of a fair distribution of energy to all active sites. Thus, the transesterification reactions by 2 wt % catalysts will commence sooner, while the 4 wt % catalyst will show delays in transesterification

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com

998
a)

J. Kansedo et al.

Chem. Eng. Technol. 2008, 31, No. 7, 993999

Amount of catalyst (%)

b)

Reaction temperature (C)

4%

2% Amount of catalyst

ceed that of the 2 wt % catalyst. Thus, it is only when the reaction temperature is sufficient to activate all of the active sites of the catalyst that the perception that higher amounts of catalyst will result in higher conversion applies. These results demonstrate the capability of using systematical statistical design, whereby the effect of the interaction between variables can be identified easily as compared to the conventional method of studying one variable at a time. Fig. 4 shows the changes in conversion on varying reaction period at 2 wt % and 4 wt % catalyst. The other two process variables, i.e., reaction temperature and ratio of oil/methanol were kept constant at 120 C and 1:8, respectively. As in Fig. 4, at a shorter reaction period, the 2 wt % catalyst showed a significantly lower conversion of palm oil to biodiesel compared to that of the 4 wt % catalyst. However, the conversion by the 2 wt % catalyst continues to steadily increase until it subsequently exceeds that of the 4 wt % catalyst after 210 min of reaction. On the other hand, the 4 wt % catalyst showed a fairly constant conversion for most of the time and slightly dropped when the reaction period went beyond 210 min. From this behavior, it can be concluded that the reaction equilibrium is reached rapidly with the 4 wt % catalyst, and after reaching equilibrium, the transesterification reactions may become stagnant or start to reverse. The equilibrium of transesterification reactions by the 2 wt % catalyst is only reached after a longer reaction time due to lower active site availability, and thus, this defines the staged increments of conversion by 2 wt % catalyst from 120 min to 240 min. The total rate of reaction, i.e., both transesterification and reversed reactions, may be directly influenced by the total number of active sites of the catalyst, whereby higher quantities of active sites will accelerate both reactions.

Conversion (%)

Conversion (%)

Conclusions

Reaction temperature (C)

Figure 3. Effect of reaction temperature and amount of catalyst on the conversion of palm oil to biodiesel: (a) response surface plot, and (b) two-dimensional drawing.

Based on the experimental result obtained, it can be concluded that montmorillonite KSF could be used as an effective catalyst for the conversion of palm oil to biodiesel. However, future research on this catalyst is still required to investigate a wider range of the parameters, and also possible new parameters that can be included in the reaction process, e.g., the mixing intensity, reaction pressure, etc., to optimize the biodiesel production.

Acknowledgement
reactions. However, since the activation energy for all catalyst in the 4 wt % loaded sample is fully achieved at higher temperatures, the reactions will proceed rapidly and subsequently exThe authors would like to thank the Universiti Sains Malaysia for providing funding for this project.

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com

Chem. Eng. Technol. 2008, 31, No. 7, 993999

a)

Heterogeneous transesterification

999

References
[1] W. Li, W. Du, D. Liu, Process Biochem. 2007, 42, 1481. [2] G. J. Suppes et al., Appl. Catal., A 2004, 257, 213. [3] D. A. C. Ferreira, M. R. Meneghetti, S. M. P. Meneghetti, S. R. Wolf, Appl. Catal., A 2007, 317, 58. [4] L. Guerreiro et al., Catal. Today 2006, 118, 166. [5] W. Xie, H. Li, J. Mol. Catal. A: Chem. 2006, 255, 1. [6] W. Xie, X. Huang, Catal. Lett. 2006, 107, 1. [7] W. Xie, H. Peng, L. Chen, Appl. Catal., A 2006, 300, 67. [8] X. Liu, H. He, Y. Wang, S. Zhu, Catal. Commun. 2007, 8, 1107. [9] Z. Yang, W. Xie, Fuel Process. Technol. 2007, 88, 631. [10] O. S. Stamenkovic et al., Bioresour. Technol. 2007, 98, 2688. [11] M. Lopez Granados et al., Appl. Catal., B 2007, 73, 317. [12] F. Ma, M. A. Hanna, Bioresour. Technol. 1999, 70, 1. [13] H. Fukuda, A. Kondo, H. Noda, J. Biosci. Bioeng. 2001, 92, 405. [14] J. Van Gerpen, Fuel Process. Technol. 2005, 86, 1097. [15] A. Demirbas, Prog. Energy Combust. Sci. 2007, 33, 1. [16] B. K. Barnwal, M. P. Sharma, Renewable Sustainable Energy Rev. 2005, 9, 363. [17] R. J. Crookes, Biomass Bioenergy 2006, 30, 461. [18] C. Carraretto et al., Energy Fuels 2004, 29, 2195. [19] J. Jitputti et al., Joint Int. Conf. on Sustainable Energy and Environment, Hua Hin, Thailand, December 2004. [20] H. J. Kim et al., Catal. Today 2004, 9395, 315. [21] G. Arzamendi, I. Campo, E. Arguinarena, M. Sanchez, M. Montes, L. M. Gandia, Chem. Eng. J. 2007, 134, 123. [22] M. A. Kalam, H. H. Masjuki, Biomass Bioenergy 2002, 317, 58. [23] A. Kumar Tiwari, A. Kumar, H. Raheman, Biomass Bioenergy 2007, 31, 569. [24] R. Sarin, M. Sharma, S. Sinharay, R. K. Malhotra, Fuel 2007, 86, 1365. [25] M. G. Kulkarni, A. K. Dalai, N. N. Bakhshi, Bioresour. Technol. 2007, 98, 2027. [26] M. Canakci, Bioresour. Technol. 2007, 98, 183. [27] Y. Wang et al., J. Mol. Catal. A: Chem. 2006, 252, 107. [28] J. M. Marchetti, V. U. Miguel, A. F. Errazu, Renewable Sustainable Energy Rev. 2007, 11, 1300.

Conversion (%)

Amount of catalyst (%)

b)

Reaction period (min)

4% Conversion (%)

2% Amount of catalyst

Reaction period
Figure 4. Effect of reaction period and amount of catalyst on the conversion of palm oil to biodiesel: (a) response surface plot, and (b) two-dimensional drawing.

2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

http://www.cet-journal.com