1718

Energy & Fuels 2009, 23, 17181722

Kinetic Study of Alcoholysis of the Fatty Acids Catalyzed by Tin Chloride(II): An Alternative Catalyst for Biodiesel Production
Abiney L. Cardoso, Soraia C. G. Neves, and Marcio J. da Silva*
Chemistry Departament, Federal UniVersity of Vicosa, Vicosa, Minas Gerais 36570-000, Brazil ReceiVed August 4, 2008. ReVised Manuscript ReceiVed October 20, 2008

A valuable alternative for making the biodiesel price more competitive than petroleum-derived fuel is the use of raw materials of low cost. However, these feedstocks contain a high amount of free fatty acids (FFAs). Currently, the biodiesel production from this kind of resource is comprised in a two-step process, where an initial acid-catalyzed esterication of the FFA followed by a base-catalyzed transesterication of the triglycerides occurs. Commonly, the sulfuric acid is the catalyst on the acid step. However, some serious drawbacks, such as substantial reactor corrosion and the great generation of wastes, even of the salts formed by mineral acid neutralization, are negative aspects of these processes. In this work, the SnCl2-catalyzed esterication of saturated and unsaturated FFA was evaluated. High yields and selectivities were achieved on the ethanolysis of FFA, under mild conditions of reaction. The SnCl2 catalyst has shown to be as active as the mineral acid H2SO4. Their use has relevant advantages in comparison to mineral acid catalysts, such as lower corrosiveness and as well as the unnecessary neutralization after the reaction nishes. Herein, the effects of the principal parameters of the reaction have been investigated. Kinetic investigations revealed a rst-order dependence upon both the oleic acid and the catalyst concentration. The energy of activation achieved for this process was approximately similar to those reported for H2SO4-catalyzed FFA esterication.

Introduction Biodiesel is a suitable substitute for petroleum-derived diesel. It is a nonpolluting fuel, which consists of methyl or ethyl esters, as a result of either transesterication of triacylglycerides (TGs) or esterication of free fatty acids (FFAs).1 The use of biodiesel fuel is environmentally more attractive because it is obtained from vegetable oils and animal fats, which are renewable biomass sources.2 Currently, most of the biodiesel comes from the transesterication of edible resources, such as animal fats, vegetable oils, and even waste cooking oils, in conditions of alkaline catalysis.3-5 However, the high consumption of catalysts, the formation of soaps, and the low yields make biodiesel more expensive than the petroleum-derived fuel.6 The common acid-catalyzed processes of biodiesel production from raw materials of low cost involve the use of mineral acids, owing to the high amount of FFA-edible resources, which are not compatible with the use of base-catalyzed methods.7 Two alternative approaches can be normally used: rst, a two-step process, which requires an initial acid-catalyzed esterication of the FFA, followed by a base-catalyzed transesterication of glycerides present into oil and, second, a one-step process that only uses an acid catalyst that simultaneously promotes both
* To whom correspondence should be addressed. Telephone: 55-3138993210. Fax: 55-31-389923270. E-mail: silvamj2003@ufv.br. (1) Maa, F.; Hannab, M. A. Bioresour. Technol. 1999, 70, 115. (2) Zheng, S.; Kates, M.; Dube, M. A.; Mclean, D. D. Biomass Bioenergy 2006, 30, 267272. (3) Chhetri, A. B.; Watts, K. C.; Islam, M. R. Energies 2008, 1, 318, doi:, 10.3390/en1010003. (4) Kima, H. J.; Kang, B. S.; Kim, M. J.; Park, Y. M.; Kimb, D. K.; Lee, J. S.; Lee, K. Y. Catal. Today 2004, 93-95, 315320. (5) Haas, M. J.; Karen, M. S.; Marmer, W. N.; Foglia, T. A. J. Am. Oil Chem. Soc. 2004, 81, 8389. (6) Haas, M. J. Fuel Process. Technol. 2005, 86, 10871096. (7) Lotero, E.; Liu, Y.; Lopez, D. E.; Suwannakarn, K.; Bruce, D. A.; Goodwin, J. G. Ind. Eng. Chem. Res. 2005, 44, 2535325363.

esterication and transesterication reactions.8 It must be said, however, that major drawbacks. such as reactor corrosion and substantial generation of byproduct and wastes, including the salts formed as a result of the mineral acid neutralization, which must be disposed off into the environment, comprise negative and virtually impassable aspects for the mineral acid-catalyzed process.9 The biodiesel manufacturing technique has to accomplish the reduction of waste pollutants provoked by conventional catalytic processes.10 Thus, the extensive demand for cleaner methodologies has forced the chemical industry to search for environmentally friendly acidic catalysts. A highlighted example is the solid acids.11 These solid catalysts, which normally present Lewis acidity, are easily separated from the reaction medium and are potentially less corrosive for the reactors.12-15 However, the use them usually require drastic reaction conditions, i.e., high temperature and elevated pressure. Thus, the development of alternative catalysts for the FFA esterication based on the Lewis acids over traditional Brnsted acids, which operate under mild conditions of reaction and are less corrosive, is one of the main challenges to be overcome. Moreover, this technology
(8) Zullaikah, S.; Lai, C. C.; Vali, S. R.; Ju, Y. H. Bioresour. Technol. 2005, 96, 18891896. (9) Corma, A.; Garca, H. Chem. Ver. 2003, 103, 43074365. (10) Marchetti, J. M.; Miguel, V. U.; Errazu, A. F. Fuel 2007, 86, 906 910. (11) Timofeeva, M. N. Appl. Catal. 2003, 256, 1935. (12) Reis, S. C. M.; Lachter, E. R. R.; Nascimento, R. S. V.; Rodrigues, J. A., Jr.; Reid, M. G. J. Am. Oil Chem. Soc. 2005, 82, 661665. (13) Ferreira, D. A. C.; Meneghetti, M. R.; Meneghetti, S. M. P.; Wolf, C. R. Appl. Catal. 2007, 317, 5861. (14) Alizadeh, M. H.; Kermanil, T.; Tayebee, R. Monatsh. Chem. 2007, 138, 165170. (15) Narasimharao, K.; Brown, D. R.; Lee, A. F.; Newman, A. D.; Siril, P. F.; Tavener, S. J.; Wilson, K. J. Catal. 2007, 248, 226234.

10.1021/ef800639h CCC: $40.75 2009 American Chemical Society Published on Web 01/20/2009

AlternatiVe Catalyst for Biodiesel Production

Energy & Fuels, Vol. 23, 2009 1719

Scheme 1. Reversible Ethanolysis of Fatty Acid

could thus allow for the production of biodiesel in more competitive costs, in processes of lower environmental impact.16 For the catalysts based on metal complexes or metal salts. Lewis acids are widely used in organic synthesis and are potentially useful on the production of biodiesel.17 For instance, Di Serio and co-workers have reported the use of carboxylic salts of metals, such as Cd, Mn, Pb, and Zn, for the production of biodiesel from vegetable oil containing high amount of FFAs.18 However, among the metals, Lewis acids with more applicability to esterication reactions containing tin metal have been more extensively explored.19-21 As an example, recently, Otera and co-workers have reported successfully the use of distannoxanes as catalysts in esterication reactions of several organic substrates.22 In addition, other organotin compounds, such as BuSn(O)OH, have shown high catalytic activity.23 Conversely, although the catalytic activity of some tributyltin alkoxides is recognized, the nal yield of the process proved to be disappointing.23 Recently, the use of metal complexes with the pyrone ligand (3-hydroxy-2-methyl-4-pyrone) as a catalyst in the soybean oil methanolysis under homogeneous conditions was described by Ziani and co-workers and results in a high yield of fatty acid methyl esters (FAMEs).24,25 Despite the successful use of these catalysts, their application to biodiesel manufacturing is restricted because of their cost and delicate conditions of synthesis. Moreover, these catalysts are active only for methanolysis reactions. In view of that, replacing the corrosive liquid catalysts by alternative Lewis acid catalysts might lead to the development of an innovative technology for biodiesel production.25 In this work, the use of SnCl2 2H2O as a catalyst on the alcoholysis of FFA was evaluated. The investigation herein involves the correlation between key parameters of the reaction, such as the catalyst and fatty acid concentrations, as well as the inuence of the molar ratio FFA/ethanol and the temperature upon the ethyl oleate production. Experimental Section
Materials. All of the chemicals were purchased from known chemical companies. They were used without prior purication, unless otherwise stated. Ethanol was puried twice by distillation over phosphorus pentoxide. The purchasing of SnCl2 2H2O was from Sigma-Aldrich (Milwaukee, WI), and sulfuric acid (H2SO4, 98 wt %) was from Qumica Moderna (Sao Paulo, SP, Brazil). Hexane and chloroform were purchased from Vetec (Sao Paulo, SP, Brazil). The oleic acid and their ethyl ester were purchased from Qumica Moderna (Sao Paulo, SP, Brazil). Methods. General Procedure. The reactions were carried out in a 50 mL three-necked glass ask, equipped with a sampling
(16) Einloft, S.; Magalhaes, T. O.; Donato, A.; Dullius, J.; Ligabue, R. Energy Fuels 2008, 22, 671675. (17) Macedo, C. C. S.; Abreu, F. R.; Tavares, A. P.; Alves, M. B.; Zara, L. F.; Rubin, J. C.; Suares, P. A. Z. J. Braz. Chem. Soc. 2006, 17, 1291 1296. (18) Di Serio, M.; Tesser, R.; Dimiccoli, M.; Cammarota, F.; Nastasi, M.; Santacesaria, E. J. Mol. Catal. A: Chem. 2005, 239, 111117. (19) Jousseaume, B.; Laporte, C.; Rascle, M.-C.; Toupance, T. Chem. Commun. 2003, 14281429. (20) Baumhof, P.; Mazitschek, R.; Giannis, A. Angew. Chem., Int. Ed. 2001, 40, 36723678. (21) Mascaretti, O. A.; Furlan, L. E. Aldrich Acta 1997, 30, 5568. (22) Otera, J.; Dan-oh, N.; Nozaki, H. J. Org. Chem. 1991, 56, 5307 5311. (23) Hoydonckx, H. E.; De Vos, D. E.; Chavan, S. A.; Jacobs, P. A. Top. Catal. 2004, 27, 8388. (24) Abreu, F. R.; Lima, D. G.; Hamu, E. H.; Einloft, S.; Rubim, J.; Suarez, P. A. Z. J. Am. Oil Chem. Soc. 2003, 80, 601604. (25) Abreu, F. R.; Lima, D. G.; Hamu, E. H.; Wolf, C. R.; Suarez, P. A. Z. J. Mol. Catal. A: Chem. 2005, 227, 263268.

system, with a reux condenser, magnetic stirrer, and a thermometer. The experiments were carried out at reux temperature (ca. 78 C). All catalytic tests were performed in triplicate. In the catalytic runs, excess ethanol was added to the reactor containing the FFAs to drive the equilibrium toward the formation of ester and for the reaction rate not becoming dependent upon the concentration of this substrate. In a typical run, 10 mL of an ethanol solution (120 mmol) and oleic acid (1 mmol) was heated to the reux temperature followed by the addition of SnCl2 2H2O (0.01-0.4 mmol). SnCl2 Catalyst Versus H2SO4. Catalytic runs with sulfuric acid were also carried out for comparison purposes with the latter because this is a common homogeneous catalyst in a number of related investigations. The sulfuric acid and tin chloride concentrations used were 0.1 and 0.4 mmol, respectively. The reaction was continuously monitored by gas chromatography (GC) analysis of aliquots taken at regular intervals of 30 min. The yields for those catalytic processes were obtained by means of the GC peak areas of the ethyl ester in comparison to the corresponding calibrating curve. Alternatively, aliquots from the reactions were titrated against an alcoholic solution of KOH, aiming at monitoring at short intervals of time (ca. 10 min). The catalytic runs monitored by titration with alcoholic solution KOH were carry out in a scale 3+ higher that the runs followed by GC analyses. This was necessary because of a higher amount of aliquots and a higher sample volume. The amount of KOH to neutralize the residual fatty acid and to react with the acid catalyst (SnCl2 or H2SO4) was taking into account to calculate the concentration of the ethyl oleate in solution. Kinetic Measurements. Two procedures have been used herein to evaluate the effect of the catalyst and oleic acid concentrations on the reaction rate. For the fatty acid, the molar ratio between ethanol, oleic acid, and SnCl2 has been preserved to 120:1:0.1. For the catalyst, only the molar ratio ethanol/oleic acid was maintained constant, i.e., 120:1. The catalyst concentration ranged from 0.01 to 0.4 mmol. In both procedures, the residual acid was titrated against an alcoholic solution of 0.01 mol L-1 KOH to follow the progress of the reaction. Identication of the Reaction Products. The reaction products were isolated by a column chromatography (silica) using a solution of CHCl3/hexane (1:1) as an eluent. The analyses by GC-mass spectrometry (MS) were carried out in a Shimadzu GC 17A gas chromatograph coupled with a MS-QP 5050A Shimadzu mass spectrometer (Tokyo, Japan) with a DB5 capillary column (30 m length, 0.25 mm inner diameter, and 0.25 mm lm thickness) and helium as the carrier gas at 2 mL/min. The temperature program was as follows: 80 C for 1 min, 10 C/min up to 280 C, and holding time of 5 min. The GC injector and MS ion source temperatures ranged from 250 to 260 C. The MS detector operated in the EI mode at 70 eV, with a scanning range of m/z 50-500.

Results and Discussion General Aspects. The ethanolysis of FFA is a typical reversible acid-catalyzed reaction that produces ester and water as byproducts (Scheme 1). In the majority of the catalytic tests, the oleic acid and the ethanol were the substrates selected. First, the esterication of oleic acid with ethanol conducted in the absence of the acidic catalysts (SnCl2 2H2O and H2SO4) produced no signicant yields of ethyl oleate, despite the high molar ratio of ethanol/ fatty acid used, i.e., 120:1 (see the Experimental Section for full details). Very low conversion of oleic acid into ethyl oleate (ca. 12%; see Figure 1) has occurred even after more than 12 h of reaction. Conversely, in the presence of SnCl2, a much greater yield (>90%) with a high selectivity (>93%; see the Experimental

1720 Energy & Fuels, Vol. 23, 2009

Cardoso et al.
Table 1. FAEE Yields and Selectivities Obtained from Ethanolysis of FFA Catalyzed by SnCl2 2H2Oa
run 1 2 3 4 5 fatty acid myristic palmitic stearic oleic linoleic CN/DBb 14:0 16:0 18:0 18:1 18:2 ethyl ester yield (%)c,d 90 86 87 90 92 selectivity (%)c,d,e 97 95 97 95 93

a Reaction conditions: ethanol, 120 mmol; fatty acid, 1 mmol; SnCl2 2H2O, 0.1 mmol; reux temperature; reaction time, 12 h. b CN and DB refer to the number of carbons and double bonds in the carbonic chain of the FFA, respectively. c Calculated from the areas of the ethyl ester GC peaks. d Average of the results of three assays. e The differences observed between the selectivity results were small in relation to the experimental error.

Figure 1. Ethanolysis reaction of oleic acid with and without SnCl2 catalyst.

Figure 3. Linear plot of the range of the oleic acid concentration versus time for the SnCl2-catalyzed ethanolysis reaction. Figure 2. Ethanolysis of oleic acid catalyzed by H2SO4 and SnCl2.

Section) was obtained after 12 h of reaction. Thus, these results are evidence that the excess of ethanol has not been given a signicant effect on the yield nor in the reaction rate. In all catalytic runs, the ethyl oleate yield increases with the increasing reaction time. However, when the reaction time goes further than 12 h, the ester concentration remains almost invariable. The catalytic tests were performed in triplicate, with a molar ratio of fatty acid/catalyst (100:1). It is worth mentioning that these conditions were not optimized to achieve high yields at short-time reactions. However, they provide an adequate methodology to evaluate the catalytic activity of SnCl2 2H2O on these reactions. SnCl2 Catalyst versus H2SO4: A Comparative Study. The performance of both acid catalysts was outstanding, as shown in Figure 2, where the kinetic curves are presented. Although negligible hydrolysis of fatty acid ethyl esters (FAEEs) may take place in these reactions, the results were satisfactorily reproducible and were in accordance with those obtained by GC. Note that the reaction yields increase steadily, reach the maximum values (ca. 90%) after a reaction time of ca. 120 min, and thus, stay almost invariable afterward. The two acidic catalysts have different structures and acid character and, certainly, different mechanisms of action. Despite that, they displayed similar behavior as can be conrmed by the attainment of comparable ethyl oleate yields at a given reaction time (Figure 2). Note that reaction yields have increased steadily to a maximum value of ca. 90%, in approximately 120 min after setting up the reaction. The monitoring of the reaction for periods higher than 120 min reveals that the yields of both remained invariable after this time. Ethanolysis of Saturated and Unsaturated Fatty Acids Catalyzed by SnCl2. The results of esterication with ethanol

Table 2. Effect of the Concentration of Tin Chloride on the Initial Rate of Ethanolysis of Oleic Acida
run 1 2 3 4 SnCl2 concentration (mmol) 0.01 0.1 0.2 0.4 rate initial (mmol FA min-1) 0.12 0.23 0.39 0.56

a Reaction conditions: [oleic acid], 1 mmol; [ethanol], 120 mmol; reux temperature.

of some FFA normally present in vegetable oil and fats are concisely showed in Table 1. It must be noted that, in all catalytic runs, the FAEE yields increase with the augment of the reaction time. However, from obtained results monitoring the reaction at longer times (>12 h, not showed herein), it was possible to observe that the FAEE concentration tends to stay invariable. Reaching 12 h of reaction, with a catalytic amount of SnCl2 (0.1 mmol), FAEEs were produced in the range of 86-92% yields for all of the fatty substrates. Effect of the Oleic Acid Concentration. Literature data have been attributed to a rst-order dependence in relation to the fatty acid concentration in the majority of several esterication reactions.26 From this standpoint, in the present work, we can assume that this is also true for the oleic acid esterication. Thus, to obtain kinetic information, i.e., the reaction order in relation to the oleic acid concentration, plots of ln[oleic acid] versus time were thus built for each acid-catalyzed process (Figure 3 presents only the data of ethanolysis of the tincatalyzed oleic acid). The resulting data were adjusted up to a kinetic behavior of rst order. It is important mentioning that a high molar excess
(26) Silva, M. J.; Augusti, R.; Cardoso, A. L. J. Am. Oil Chem. Soc. 2008, 85, 555560.

AlternatiVe Catalyst for Biodiesel Production

Energy & Fuels, Vol. 23, 2009 1721

Figure 4. Effect of the temperature on the ethanolysis of oleic acid catalyzed by SnCl2.

Figure 6. Effect of the chain carbonic of alcohol on the conversion of oleic acid into a respective ester.

Figure 5. Linear plot of ln K versus 1/T resulting from the esterication of oleic acid catalyzed by SnCl2.

of ethanol in relation to oleic acid (120:1) was used to ensure that the ethanol concentration would remain essentially constant during the reaction course (see the Experimental Section). Effect of the Tin Chloride Concentration. The effectiveness of the catalyst SnCl2 2H2O has been investigated in a broad range of concentrations, but only the signicant results are showed on Table 2. To control the accuracy of the kinetic data related to the residual concentration of oleic acid, it has been essential to maintain the acid-catalyzed process distant from the equilibrium position. In view of that, the kinetic measurements have been carried out within 2 h of reaction. The concentration of the residual oleic acid was determined directly from the reaction medium via titration with an alcoholic solution of KOH (more details are described in the Experimental Section).27 To calculate the initial rate, only the rst interval of 30 min was taken into account, because in this period, the dependence in relation to the oleic acid concentration is of pseudo-zero order. Thus, we assume that only the tin chloride concentration is ranging in each catalytic run. The initial SnCl2 concentration in each reaction varied from 0.01 to 0.40 mmol. The oleic acid load employed was of 1 mmol, and a large excess of ethanol was used. As a general tendency, an increase in the catalyst concentration has caused an improvement in the ethyl oleate
(27) Berrios, M.; Siles, J.; Martn, M. A.; Martin, A. Fuel 2007, 86, 23832388.

yields at any reaction time. A low molar ratio between the oleic acid and SnCl2 (1:0.01) has afforded approximately ca. 12% of ethyl oleate after 30 min of reaction. However, the use of a high molar ratio (1:0.4) for the same interval of reaction has revealed a signicant rise in the yield (up to ca. 56%). The results are approximately concomitant with a rst-order dependence in relation to the catalyst concentration. The high value of R2 is suggestive that the data t to rst-order kinetics. In addition, the angular coefcient value (ca. 0.87) gives the kinetic constant value. The angular coefcient of the high curve obtained with this data is suggestive of this rst-order dependence. Effect of the Temperature on the Ethanolysis of SnCl2Catalyzed Oleic Acid. It was veried that the temperature noticeably affects both the reaction rate and conversion of oleic acid into ethyl oleate (see Figure 4). For instance, at room temperature, even for longer time reactions, only low conversions were obtained. However, at temperatures up of 45 C, a signicance increase of the initial rate of reaction occurs, and consequently, higher conversions were reached. Expectedly, an increase in the reaction temperature has caused a corresponding improvement on the reaction rate, especially at a range of 45-75 C. From the data obtained at each temperature of the reaction, it was possible to determine the specic values for the rate constants. Thus, from the resulting data constructing the curve shown in Figure 5 and employing a linear regression method, the angular coefcient (-E/R) of the curve obtained allow us to calculate the activation energy of this reaction. We determined a value of activation energy equal to 46.69 kJ mol-1. This value is approximately similar to those found by Berrios and co-workers on the kinetic study of the esterication of FFA in sunower oil catalyzed by H2SO4.28 These researchers found values of 50.74 and 42.76 kJ mol-1 in reactions where the oleic acid concentration was 5 and 10% (w/w), respectively. Note that, despite these relatively similar results, the ratio between the oleic acid and catalyst in our experiments has been different compared to those described by them (10:1), even of the two acidic catalysts bearing different structures, i.e., H2SO4 and SnCl2. Effect of the Carbonic Chain of Alcohol Substrate on the Reaction Rate of Tin-Chloride-Catalyzed Esterication.
(28) Kulkarni, M. G.; Sawant, S. B. J. Am. Oil. Chem. Soc. 2003, 80, 10331038.

1722 Energy & Fuels, Vol. 23, 2009

Cardoso et al.

A drastic lowering on the conversion rate of oleic acid into a respective ester has been observed when the carbon chain of alcohol is higher (see Figure 6). The action of the catalyst probably consists of polarizing the carbonyl of the substrate fatty acid. However, when the carbonic chain of alcohol increases, a high bulk hindrance occurs on the hydroxyl of the alcohol, and this fact probably reduces the efcient attack of them to the polarized carbonyl of oleic acid. Finally, SnCl2 2H2O efciently promotes the esterication of saturated and unsaturated FFAs in ethanol solutions under mild reaction conditions. The high achievable yields under mild reaction conditions are comparable to other obtained with a common acid catalyst, such as sulfuric acid (H2SO4). Therefore, SnCl2 is a potential catalyst for the production of biodiesel from

low-cost raw materials. Kinetic measurements revealed that the acid-catalyzed esterication is of rst order relative to the concentration of both the fatty acid and SnCl2. The advantages of this protocol are the use of an available low-cost catalyst, which is easy to manipulate and potentially less corrosive. In addition, studies of the recovered and reused SnCl2 in consecutive catalytic processes are being developed in our laboratories. The results of this work put forward SnCl2 as a promising acid catalyst for the production of biodiesel in processes of lower environmental impact.
Acknowledgment. We thank the Brazilian research agencies, FUNARBE, FAPEMIG, and CNPq, for the nancial support.
EF800639H