Acid Esterification-Alkaline Transesterification Process For

Acid Esterifi cation-Alkaline Transesterifi cation Process for Methyl Ester Production from Crude Rubber Seed Oil Prachasanti Thaiyasuit, Kulachate Pianthong＊ and Ittipon Worapun Department of Mechanical Engineering, Faculty of Engineering, Ubon Ratchathani University (Ubonratchathani, 34190, THAILAND) Abstract: This study aims to examine methods and the most suitable conditions for producing methyl ester from crude rubber seed oil. An acid esterifi cation-alkaline transesterifi cation process is proposed. In the experiment, the 20% FFA of crude rubber seed oil could be reduced to 3% FFA by acid esterifi cation. The product after esterifi ed was then tranesterifi ed by alkaline transesterifi cation process. By this method, the maximum yield of methyl ester was 90% by mass. The overall consumption of methanol was 10.5:1 by molar ratio. The yielded methyl ester was tested for its fuel properties and met required standards. The major fatty acid methyl ester compositions were analyzed and constituted of methyl linoleate 41.57%, methyl oleate 24.87%, and methyl lonolenate 15.16%. Therefore, the cetane number of methyl ester could be estimated as 47.85, while the tested result of motor cetane number was 51.20.

Key words: methyl ester, crude rubber seed oil, esteriﬁ cation, transesteriﬁ cation

1 INTRODUCTION madhas et al.5）had found that the FFA content in the crude Nowadays, the shortage or crisis of fossil fuel lately rubber seed oil was about 17％ and 19％ of FFA in mahua becomes one of the most concerns in automotive industry oil was found by Ghadge and Raheman6）. This FFA value is and world energy sector. This infl uences researchers and significantly high and obstructs the transesterification industries to search for a renewable energy to compensate process in the methyl ester production. It has been con- the fossil fuel. Methyl ester fuel is one of the possible firmed that the ester yield decreases with increasing in choices, because it is renewable and free of sulfur. Methyl FFA significantly. Usually, the alkaline-catalyzed trans- ester contains about 10-11％ oxygen by weight. This char- esterifi cation takes place well only with refi ned oil having acteristic reduces the emissions of CO, HC, and particulate FFA value of less than 2％5－7）. Canakci and Van Gerpen8） matter in the exhaust gas compared with diesel fuel. As it had found that transesterification would not occur if the is obtained mainly from plantation resources, it is able to FFA content in the oil were above 3％. Usually, free car- reduce the lifecycle of carbon dioxide by almost 78％ com- boxylic acids form soaps with alkaline-catalyzed trans- pared to conventional diesel fuel1）, while its fuel properties esterification, hence they impede the separation of the are very closed to diesel. Also, methyl ester can be used in glycerol phase due to the emulsifying effects of soaps and diesel engines with few or no engine modifi cations2）. lower their catalytic activity9, 10）. In extreme cases, of more Usually, methyl ester has been produced from the edible than 5％ FFA oil, Canakci and Van Gerpen11）reported that plant oils such as soybean oil3）and palm oil4）. However, it is the reacted mixture might completely gel after the addition always controversial on using edible oil or food resource as of KOH or NaOH, so that the charge has to be discarded. fuels. Therefore, in recent years, considerable research Therefore, the single step alkaline-catalyzed tranesterifi ca- effort has been directed towards replacing food oils with tion process is not suitable to produce methyl esters from various non-edible, wasted, or low-cost oils. The unrefi ned high FFA crude oil. In order to reduce the FFA, the crude non-edible oils have been considered, however they have oil should be refi ned, but the overall production cost of the high impurities such as free fatty acid（FFA）and phospha- methyl ester will be increased. The single step acid trans- tide. These impurities impede the transesterification esterifi cation is a typical method of producing methyl ester process in the methyl ester production. For instance, Ra- from high FFA oil, however, it requires more methanol,

＊Correspondence to: Kulachate Pianthong, Department of Mechanical Engineering, Faculty of Engineering, Ubon Ratchathani University, Warinchamrab, Ubonratchathani, 34190, THAILAND. E-mail: [email protected] Accepted September 26, 2011 (received for review August 19, 2011) Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 online http://www.jstage.jst.go.jp/browse/jos/ http://mc.manusriptcentral.com/jjocs

81 P. Thaiyasuit, K. Pianthong and I. Worapun

high temperature, and is also time consuming. 600 which is widely growth in Thailand. The rubber seeds Recently, researchers searched for means to produce were cracked and the kernels（52.5％ of seed weight）were methyl ester from high FFA oil such as non-edible crude dried in the oven at 100℃ for 20 h. The CRSO was extract- plant oils, animal oils, and fi sh oils. The reduction of FFA in ed from kernels by hydraulic press machine and was about crude oil before tranesterifi cation process has been exam- 10％ of seed weight as shown in Fig. 1. The extracted ined. This method is also effi cient to produce methyl ester crude rubber seed oil usually contains sediment of kernel from high FFA oils. However, it consumes more methanol, and moisture. The CRSO should be cleared from adulter- as Ghadge and Raheman6）studied the production of methyl ants before the acid esterifi cation process in order to avoid ester from mahua oil having 19％ FFA by pretreatment the imperfection of the process. The dregs of the kernel process and transesterifi cation process. The pretreatment can be removed by the filter fabric. The moisture in the process comprised of double acid-esterifi cation. The 19％ CRSO is in water form. It is found that existence the addi- high FFA level of mahua oil was reduced to less than 1％ tion of 0.5％ water to a mixture of oil, methanol, and sulfu- by this pretreatment process. This process consumed 0.7 ric acid could reduce ester conversion from 95％ to below v/v（16.8:1 molar ratio）methanol-to-oil ratio. Then, the 90％. Also at a water content of 5％, ester conversion de- product of pretreatment process was proceed to trans- creased to only 5.6％11）. Thereby, the fi ltered oil is heated esterifi cation process and consumed methanol 0.25 v/v（6:1 at 120℃ for 5 min to remove the moisture. Then, the free molar ratio）, thus the overall of methanol consumption was fatty acid（FFA）content of CRSO was determined. It was 22.8:1 in molar ratio. Ramadhas et al.5）investigated on the found that the FFA content was increased associate with production of methyl ester from 17％ FFA rubber seed oil the times keeping the rubber seed before extracting crude by two steps transesterifi cation process. The fi rst step was oil as shown in Fig. 2. The CRSO using in this study has acid-esterification and consumed methanol 6:1 in molar FFA content of 20％（Oleic）. The physical properties of ratio. The second step consumed methanol 9:1, thus the CRSO tested by the Research and Technology Institute of overall of methanol consumption was 15:1 in molar ratio. the Petroleum Authority of Thailand（PTT Public Company By this method, over usage of methanol, compare to theo- Limited）are shown in Table 1. The fatty acid compositions retical requirement, is unavoidable. Ei-Mashad et al.12） analysis carried out by the Thailand Institute of Scientifi c found the two steps transesterification was effective and Technological Research（TISTR）with gas chromato- method for producing methyl ester from the acidified graphic method are shown in Table 2. Knowing the fatty salmon oil having 6％ FFA. The 6％ FFA salmon oil was acid composition, the molecular mass of the CRSO can be reduce to 1.5％ by esterifi cation and consumed methanol then estimated. In this research, the molecular mass of about 9:1 in molar ratio. In the second step, the methanol CRSO was determined as 872.4 g/mole. This molecular was used at molar ratio about 5:1 for transesterification. mass is very useful for the calculation of the amount of the The overall methanol consumption of this studied was 14:1. catalyst and methyl alcohol in the esterifi cation and tranes- Even thought the two steps transesterifi cation can be suitable for produce methyl ester from high FFA crude oils, however it consume too much methanol. Therefore, the aim of this study is to examine the method to reduce the consumption of methanol in methyl ester production from non-edible high FFA crude oil. The non-edible high FFA crude oil in this study was crude rubber seed oil（CRSO）. It was extracted from the rubber seed kernels and then carried out to produce methyl ester by the acid esterification-alkaline transesterification process. The acid esterification process was for reducing FFA content of the CRSO. Then the product of acid esterifi cation is transesterifi ed using alkaline catalyst in the alkaline transesterification process. Then fuel properties of methyl ester from rubber seed oil were examined and compared with diesel and methyl ester standards.

2 MATERIALS AND METHODS Fresh rubber seeds were collected from the local area called Ubonratchathani, Thailand. The rubber tree is RRIM Fig. 1 Rubber seed and crude rubber seed oil. 82 J. Oleo Sci. 61, (2) 81-88 (2012) Acid Esteriﬁ cation-Alkaline Transesteriﬁ cation Process

by the acid esterification-alkaline transesterification process. Note that, the study is performed in the laboratory and in a small batch process（0.5 kg）.

2.1 Acid esterifi cation process Acid esterifi cation process offers the advantage of partial esterifying FFA contained in the CRSO as shown in Fig. 3. This is to reduce FFA value of the CRSO to about 3％ or less. In addition, this step is the removal of phosphatides, which is also known as acid degumming. Phosphatides promote the accumulation of water in the ester product. Moreover, they increased catalyst consumption during alkaline transesterifi cation13）. Although a variety of alcohols can be used to produce methyl ester such as methanol, ethanol or butanol. The methanol is normally used as the Fig. 2 Relation between FFA and collecting times of reactor, because it is a short chain alcohol（provide simpler rubber seed. and faster reaction）and low cost. Sulfuric acid is used in this stage because of its low price, and higher catalytic ac- terification reactions. This study is to determine the tivity. Sulfuric acid is also hygroscopicity, which is impor- optimum condition for producing methyl ester from CRSO tant for the esterifi cation of free fatty acids, removing re-

Table 1 Properties of crude rubber seed oil. Property Test Method Unit This worka AB Speciﬁ c gravity ASTM D4052 kg/m3 918.2 910 922 Viscosity ASTM D445cSt 31.84 66.2 41.24b Flash point ASTM D93 ℃ 240 198 294 Water content ASTM D6308 %wt 0.108 NA NA Heating value ASTM D240 MJ/kg 39.223 37.5 39.25 NA: not available. a: Tested by the Petroleum Authority of Thailand (PTT Public Company Limited). b: Tested at 30℃. A: from Ramadhas et al.5). B:from Ikwuagwu et al.16).

Table 2 Fatty acids composition of crude rubber seed oil. Property Test Method This worka (%wt) A C Myristic acid C14:0 GC 0.11 NA 0.1 Palmitic acid C16:0 GC 9.89 10.2 12.0 Plamitoleic acid C16:1 GC 0.24 NA NA Stearic acid C18:0 GC 7.96 8.7 10.7 Oleic acid C18:1 GC 24.43 24.6 20.0 Linoleic acid C18:2 GC 41.38 39.6 36.0 Linolenic acid C18:3 GC 15.55 16.3 23.5 Arachidic acid C20:0 GC 0.27 NA NA Cis-11-Eicosenoic acid C20:1 GC 0.17 NA NA NA: not available. a: Tested by the Thailand Institute of Scientiﬁ c and Technological Research (TISTR). A: from Ramadhas et al.5). C: from Okieimen et al.21).

83 J. Oleo Sci. 61, (2) 81-88 (2012) P. Thaiyasuit, K. Pianthong and I. Worapun

Fig. 3 Esteriﬁ cation of free fatty acids.

Fig. 4 Transesterifi cation reaction between triglyceride leased water from the reaction mixture13）. This step is and methanol. sometimes called“ pre-treatment” step. The important pa- rameters affecting the acid esterification step such as molar ratio between CRSO and methanol, catalyst（sulfuric ratio of alcohol to triglyceride interferes with the separa- acid）amount and reaction duration are investigated. The tion of glycerol18）, because there is an increase in solubility. reaction temperature at 60℃ is chosen, even though an in- The separation of glycerol is difficult and the apparent crease in the transesterification rate was found with in- yield of esters decreased, because a part of the glycerol creasing reaction temperature. However, the maximum remains in the methyl ester phase. When glycerol remains temperature should not exceed the boiling point of the re- in solution, it drives the equilibrium backward to the left, actant; e.g. 64.4℃ for methanol14）. If the reaction tempera- lowering the yield of the esters19）. The alkaline catalyst ture is above this point, some portion of methanol will be used in this study is the Potassium hydroxide（KOH）. loss during reaction process. Because it decreases the tendency for soap formation The amount of the FFA in CRSO affects the appropriate when using KOH as a catalyst compared to NaOH. Also it molar ratio of methanol:oil. Therefore, this experiment is reduces the amount of methyl esters dissolved in the glyc- to fi nd the optimum molar ratio of methanol:oil. The molar erol phase after reaction and thus reduces ester losses20）. ratio at 3:1, 4.5:1, 6:1, 7.5:1 and 9:1 are varied in this ex- In addition, the glycerol of the KOH catalyst is in liquid periment. Sulfuric acid（H2SO4）is the catalyst to reduce the form, which is easier to separate from methyl ester com- FFA value in the crude rubber seed oil. The appropriate pared to the muddy gel glycerol from NaOH catalyst. amount of sulfuric acid will shorten the reacting time to The amount of KOH used in the alkaline transesterifi ca- reduce FFA to 3％. Also, the sulfuric acid may affect dark tion process, g KOH/g methanol, can be calculated as coloring in the methyl ester product15）and its corrosive- follows; ness, if it is abundantly added. So this experimental is also KOH A B /100 to find the optimum amount of the sulfuric acid. The Transesterifi cation＝（ × ） amount of sulfuric acid is varied in the range of 1.5-5.5％ KOH B AV /1000 by mass of the crude rubber seed oil. The mixture of crude Neutralization＝（ × ） rubber seed oil, methanol, and sulfuric acid has reacted by KOH KOH KOH high speed mixing machine at 14,000 rpm in a closed glass Total＝ Transesterifi cation＋ Neutralization jar. The optimum reaction time（mixing）is also determined where A is the KOH/CRSO for catalysis（％w/w）and B is at 15, 30, 45, 60, 75, 90 min. This is to save the energy in the molar ratio（oil/methanol）. AV is the acid value of the production process and to protect the reversion of the product from acid esterifi cation process（mg KOH/g oil）. chemical reaction process. Then product is poured in to a The amount of A is varied in the range of 0.5-2.5％ by separating funnel and at each 6 hours, the product was mass of the crude rubber seed oil. In order to find the taken to measure the FFA content. But from previous ex- optimum ratio between methanol and crude rubber seed perimented5）, product was continued to transesterifi cation oil, the molar ratio at 3:1, 4.5:1, 6:1, 7.5:1 and 9:1 are varied process without measuring the FFA content. However, ac- in this experiment. The required amount of KOH was dis- id-catalyzed esterifi cation is usually far slower than alkali- solved into the required methanol amount. The KOH- catalyzed reaction16）. methanol mixture was added to the product obtained from the acid esterifi cation step. Similar to the acid esterifi cation 2.2 Alkaline transesterifi cation process step, the reaction was performed at 60℃. After the reac- Alkaline transesterification process is the process that tion, resting the mixture for more than 4 h, there will be one mole of triglyceride reacts with three moles of alcohol two layers of the products. The upper layer is the alkyl to from one mole of glycerol and three moles of the respec- ester and the lower layer is the glycerol. The glycerol is in tive fatty acid alkyl esters as shown in Fig. 4. However, the liquid form and can be drained from the reacting tank transesterification is an equilibrium reaction in which a easily. large excess of alcohol is required to drive the reaction to the right direction. The molar ratio of alcohol to triglyceride has no effect on acid, peroxide, saponification and iodine value of methyl esters17）. However, the high molar

84 J. Oleo Sci. 61, (2) 81-88 (2012) Acid Esteriﬁ cation-Alkaline Transesteriﬁ cation Process

3 RESULTS AND DISCUSSIONS 3.1 FFA reduction by acid esterifi cation From the experiment, on completion of the acid esterifi - cation reaction（high speed mixing）, product is poured in to a separating funnel. Initially, the reaction was very fast and could notice the separation easily. The products of this process separated into two layers as show in Fig. 5. The upper layer was the products of this process. The lower layer was a solution which contains water（from esterifi cation of FFA）, hydrate gum, sulfuric acid and excess methanol. They were dissolved to a solution（lower layer）thus the density of this solution was higher than the upper layer. The lower layer was then drawn off. The upper layer was the emulsion between triglycerides and monoesters. It became the low FFA product which is suitable to use in the alkaline transesterification. The effectiveness of the acid Fig. 6 Effect of molar ratio of methanol to CRSO to esterifi cation was evaluated by measuring the FFA of this FFA reduction. intermediate product. From FFA measurement, at each 6 h interval between 0-48 h, the FFA content in product became lower, but it had not changed afterwards. From Fig. 6, it is found that the proper molar ratio, which can reduce the FFA from 20％ to 3％, is 6:1. With further increase in molar ratio the effectiveness was remained con- stant and some excess methanol moves over the product layer, which also found by Ramadhas et al.5）and Ghadge and Raheman6）studied. In their studied, the reason that excess methanol is required, because there are not suffi-

Fig. 7 Effect of sulfuric/CRSO to FFA reduction.

cient of sulfuric acid and high water content in crude oil. These yield the lower esterification reaction efficiency. Figure 7 shows the relationship between the concentra- tions of sulfuric/crude oil used to the FFA reduction. It reveals that the amount of the sulfuric should be more than 2.5％ by weight of the crude rubber seed oil. The appropriate reaction period（mixing）is around 30 min or more as shown in Fig. 8.

3.2 Methyl ester production by alkaline transesterifi cation The product after esterifi ed from acid esterifi cation was then transesterifi ed by alkaline transesterifi cation process. The product from the alkaline transesterifi cation is separated into two layers as shown in Fig. 9. The lower layer is glycerol and the upper layer is methyl esters. The glycerol in lower layer is drawn off to only remain the methyl ester. Methyl ester is then washed to remove the left over impuri- Fig. 5 Product from step 1 (acid esterifi cation) during ties and glycerol by warm water temperature of 50℃. The the separation process. warm water used is about 50％ of methyl ester by volume.

85 J. Oleo Sci. 61, (2) 81-88 (2012) P. Thaiyasuit, K. Pianthong and I. Worapun

Fig. 8 Effect of reaction period (high speed mixing) to Fig. 10 Effect of alkali (KOH)/CRSO to yield of meth- FFA reduction. yl ester.

Fig. 11 Effect of molar ratio of methanol to CRSO to Fig. 9 Methyl ester and glycerol during the separation yield of methyl ester. process. linoleate 41.57％, methyl oleate 24.87％, and methyl lono- The washing is carried out for three to four times or until lenate 15.16％. They are unsaturated and middle fatty acid the pH of the methyl ester is neutral（pH of around 7-8）. chain length and have 2, 1, and 3 double bonds, respective- After that, the methyl ester is heated at 120℃ for 15 min ly. Types of fatty acid methyl ester influence the cetane to remove the moisture. The fi nal product of methyl ester number, which affects to ignition quality in CI engine. The was used to calculate the yield of each condition. The yield increasing of cetane number related to length of fatty acid is defined by the obtained methyl ester divided by the chain, whereas the decreasing of cetane number related to initial CRSO（by mass）. From the results of this step, the the number of double bond, as detailed in Table 3. Approx- KOH（A）concentration of 1.5％ and the methanol to oil imately, the cetane number can be estimated from the pro- molar ratio of 4.5:1 give the highest yield of methyl ester portion of the fatty acid composed in the methyl ester. which is around 90％. These results are plotted in Fig. 10 Therefore, this rubber seed oil methyl ester has cetane and Fig. 11. Figure 11 shows that the molar ratio, in this number of 47.85 omitting the value of C16:1, C20:0 and study, can be significantly lower than previous study5）. C20:1. Knowing the fatty acid methyl ester composition, Because the most FFA content in CRSO（20％）were pre- the chemical formula of methyl ester from rubber seed oil treated to methyl ester in the fi rst step（acid esterifi cation） is calculated as C18.8H34.5O2. This formula of methyl ester is and the gum was separated before transesterifi cation. useful for the calculation of the methyl ester combustion The optimum final product of methyl ester was tested characteristics. The fuel properties of this optimum final for its composition as shown in Table 3. The major fatty product were tested and compared with the methyl ester acid methyl ester compositions of methyl ester are methyl from previous study5）and methyl ester standards of Europe

86 J. Oleo Sci. 61, (2) 81-88 (2012) Acid Esteriﬁ cation-Alkaline Transesteriﬁ cation Process

Table 3 Composition of methyl esters from crude rubber seed oil and their properties. Molecular Fatty acid Resulta Melting Cetane Formula Weight Methyl ester (%wt) pointb (℃) numberb (g/mole)

Methyl palmitate C16:0 C17H34O2 270.46 9.90 30.5 74.3

Methyl palmitoleate C16:1 C17H32O2 268.45 0.26 －42.0 NA

Methyl stearate C18:0 C19H38O2 298.51 7.73 39.1 75.5

Methyl oleate C18:1 C19H36O2 296.49 24.87 －20.0 55.0

Methyl linoleate C18:2 C19H34O2 294.48 41.57 －35.0 42.2

Methyl linolenate C18:3 C19H32O2 292.46 15.16 －55.0 22.7

Methyl arachidate C20:0 C21H42O2 326.57 0.30 54.5 NA

Methyl eicosenoate C20:1 C21H40O2 324.56 0.21 －15.0 NA NA: not available. a: Tested by the Thailand Institute of Scientiﬁ c and Technological Research (TISTR). b: from Mittebach and Remschmidt13).

Table 4 Fuel properties of methyl ester from rubber seed oil compared to others. Methyl ester standard Diesel Testing Fuel properties This work A EN ASTM EN method 14214:2003 D6751 590:1999 Density (g/cm3) ASTM D4052 0.8874a 0.874 0.86-0.90 NA 0.82-0.845 Viscosity (mm2/s) at 40℃ ASTM D445 4.456a 5.81 3.5-5.0 1.96-6.0 3.8 Cloud point (℃) ASTM D2500 3.4a 4 NA report NA Flash point (℃) ASTM D93 187a 130 120 min 130 min 55 min Gross heating value (MJ/kg) ASTM D240 39.63a 36.50 NA NA NA Methyl ester content (%wt.) EN 14103 96.4b NA 96.5 min NA NA Total glycerin (%wt.) EN 14105 0.23a NA 0.25 max 0.24 max NA Carbon residue (%wt.) ASTM D4530 0.22a NA 0.3 max NA 2.0-4.5 Acid value (mg KOH/g) ASTM D664 0.18a 0.118 0.5 max 0.8 max NA a Iodine value (g I2/100 g oil) EN 14111 82.9 NA 120 max NA NA Oxidation stability (hour) EN 14112 7.82a NA 6 min NA NA Cetane number ASTM D613 51.2c NA 51 min 47 min 51 min NA: not available. a: Tested by the National Metal and Materials Technology Center of Thailand (MTEC). b:Tested by the Department of Chemical Technology, Chulalongkorn University. c: Tested by the Petroleum Authority of Thailand (PTT Public Company Limited). A: from Ramadhas et al.5).

（EN 14214:2003）, USA（ASTM D6751）as detailed in Table. has been successfully performed. The acid esterification- 4. Most of methyl ester fuel properties met required stan- alkaline transesterifi cation reaction was adopted. The fi rst dards except the methyl ester content being nearby lower. step is the acid esterifi cation purposed to reduce the FFA Note that the tested cetane number is 51.20 which is rea- from the CRSO. At the same time, it also extracts the gum sonably higher than the approximated value. from the CRSO and converts some of FFA molecules to be mono methyl ester and water at this stage. The benefi t of using the acid esterification is to reduce the chance to waste CRSO which may occur in the soap form, in stead of 4 CONCLUDING REMARKS using the alkaline catalyst in the FFA reduction process. In this study, the production of methyl ester from CRSO The optimum conditions in the acid esterifi cation to reduce

87 J. Oleo Sci. 61, (2) 81-88 (2012) P. Thaiyasuit, K. Pianthong and I. Worapun

the FFA from 20％ in CRSO to less than 3％ were using 7） Sahoo, P. K.; Das, L. M.; Babu, M. K. G.; Naik, S. N. 2.5％ by mass of sulfuric acid, and molar ratio of methanol Biodiesel development from high acid value polanga to oil is 6:1. The alkaline transesterifi cation is the core re- seed oil and performance evaluation in a CI engine. action process which is KOH transesterification. The Fuel 86, 448-454（2007）. optimum conditions in the alkaline transesterification 8） Canakci, M.; Van Gerpen, J. Biodiesel production from process for the production were 1.5％ of KOH for catalysis oils and fats with high free fatty acids. Trans ASAE （％by mass）, while the amount of KOH for neutralization 44, 1429-1436（2001）. depends on the FFA value, and molar ratio of methanol to 9） Toyama, Y.; Tsuchiya, T.; Ishikawa, T. Alcoholysis of oil is 4.5:1. Therefore, the overall consumption of methanol fats. I. ethanolysis of olive oil by sodium hydroxide in was 10.5:1 and it is much less than previous study5, 6, 12）. ethyl alcohol. J. Soc. Chem. Ind. Jpn. N36, The yielded methyl ester was tested for it fuel properties 230B-231B（1933）. and met requirement standard except methyl ester content 10） Wright, H. J.; Segur, J. B.; Clark, H. V.; Coburn, S. K.; met nearby lower. Langdon, E. E.; Dupuis, R. N. A report on ester inter- change. Oil Soap 21, 145-148（1944）. 11） Canakci, M.; Van Gerpen, J. Biodiesel production via acid catalysis. Trans ASAE 42, 1203-1210（1999）. ACKNOWLEDGEMENT 12） Ei-Mashad, H. M.; Zhang, R.; Avena-Bustillos, R. J. A Authors would like to express their gratitude to Ubon two-step process for biodiesel production from salmon Ratchathani University and National research council of oil. Biosyst. Eng. 99, 220-227（2008）. Thailand（NRCT）for the financial support on this study. 13） Mittelbach, M.; Remschmidt, C. Biodiesel the Com- The second author thanks the Energy Policy & Planning prehensive Handbook. 3rd ed. Martin Mittelbach Graz. Offi ce（EPPO）, Ministry of Energy, Thailand, for the fi nan- （2006）. cial grant in his PhD course. We wish to express our cordial 14） Vicente, G.; Coteron, A.; Martinez, M.; Aracil, J. Appli- thanks to Professor Brian Milton（School of Mechanical and cation of the factorial design of experiments and re- Manufacturing Engineering, University of New South sponse surface methodology to optimize biodiesel pro- Wales, Australia）for his useful discussion and suggestion. duction. Ind. Crop. Prod. 8, 29-35（1998）. 15） Bondioli, P. The preparation of fatty acid esters by means of catalytic reaction. Top. Catal. 27, 77-82 （2004）. Reference 16） Ikwuagwu, O. E., Ononogbu, I. C.; Njoku, O. U. Pro- 1） Van Gerpen, J. Biodiesel processing and production. duction of biodiesel using rubber［Hevea brasiliensis Fuel Process. Technol. 86, 1097-1107（2005）. （Kunth. Muell.）］seed oil. Ind. Crop. Prod. 12, 57-62 2） Benjumea, P.; Agudelo, J.; Agudejo, A. Basic proper- （2000）. ties of palm oil biodiesel-diesel blends. Fuel 87, 17） Tomasevic, A. V.; Siler-Marinkovic, S. S. Methanolysis 2069-2075（2008）. of used frying oil. Fuel Process Technol. 81, 1-6 3） Noureddini, H.; Gao, X.; Philkana, R.S. Immobilized （2003）. pseudomonas cepacia lipase for biodiesel fuel produc- 18） Srivastava, A.; Prassad, R. Triglycerides-based diesel tion from soybean oil. Bioresource Technol. 96, fuels. Renew. Sust. Energ. Rev. 4, 111-133（2000）. 769-777（2005）. 19） Meher, L. C.; Sagar, D. V.; Naik, S. N. Technical aspect 4） Al-Widyan, M. I.; Al-Shyoukh, A.O. Experimental eval- of biodiesel production by transesterification - a re- uation of the transesterifi cation of waste palm oil into view. Renew. Sust. Energ. Rev. 10, 248-268（2006）. biodiesel. Bioresource Technol. 85, 253-256（2002）. 20） Vicente, G.; Martinez, M.; Aracil, J. Integrated biodie- 5） Ramadhas, A. S.; Jayaraj, S.; Muraleedharan, C. Biodie- sel production: a comparison of different homoge- sel production from high FFA rubber seed oil. Fuel 84, neous catalysts systems. Bioresource Technol. 92, 335-340（2005）. 297-305（2004）. 6） Ghadge, S. V.; Raheman, H. Biodiesel production from 21） Okieimen, F. E.; Bakare, O. I.; Okieimen, C. O. Studies mahua（Madhuca indica）oil having high free fatty ac- on the epoxidation of rubber seed oil. Ind. Crop Prod. ids. Biomass Bioenerg. 28, 601-605（2005）. 15, 139-144（2002）.

88 J. Oleo Sci. 61, (2) 81-88 (2012)