Hydrolisis of palm olein to oleic acid using immobilized lipase in a recirculated packed bed batch reactor

Journal ofChemical and Natural Resources Engineering, Special Edition: 10-20 © FKKKSA, Universiti Teknologi Malaysia HYDROLYSIS OF PALM OLEIN TO OLEIC ACID USING IMMOBILIZED LIPASE IN A RECIRCULATED PACKED BED BATCH REACTOR ABSTRACT Hydrolysis of palm olein to oleic acid using immobilized lipase from Asperigillus niger with the trade name Lipozyme TL 1M was studied in a recirculated packed bed batch reactor (RPBBR). This study was conducted to investigate the kinetics of the enzymatic hydrolysis of palm olein and the potential of substrate inhibition using initial velocity analysis. The temperature and pH were set at 37°C and pH 7 while the stirrer speed was set at 250 rpm and the flow rate used was 0.5 mL. min-i. Hexane was selected as the solvent and gas chromatography was used to analyze the product samples. The range of substrate concentration being investigated ranged from 0.3155 moLL- 1 to 0.8412 mol. L- I. The kinetic model of Michaelis-Menten was used to analyze the kinetic data such as the maximum rate of reaction, Vmax and the Michaelis-Menten constant, K m • LineweaverBurk plot, Eadie-Hofstee plot and Hanes-Woolf plot were used to determine the Vmax and Km values and the average values of Vmax and K m obtained from these three plots were 0.00122 moLL-1.min- 1 and 0.167 moLL- 1 respectively. No substrate inhibition was observed for up to the palm olein concentration of 0.8412 moLL-I. Key Words: Hydrolysis, Immobilized lipase, Palm olein, Kinetics, Recirculated packed bed batch reactor 1.0 INTRODUCTION The first Industrial Master Plan of 1985 identified oleochemicals as future growth sector ofthe palm oil industry. Malaysia's oleochemical production was projected to reach 20% of total world production in 2004-2007. Currently, 90% of Malaysia's oleochemical production is exported. With the rapid development of enzyme technology, considerable attention has been focused on the biotechnological oflipase in the fat and oil industry [I, 2]. Lipases are a special type of enzymes that catalyse the hydrolysis of oils and fats [3]. Recently, enzymic splitting of fats has gained increasing attention, as lipase (triacylglycerol acylhydrolase) is now available at reasonable cost. Further reduction in the cost of the enzyme by genetic manipulation of the microbe producing the enzyme is expected. This would make the enzymic hydrolysis of oils and fats highly attractive. The industrial use of lipase for splitting lipids as an energy-saving process has been addressed in the literature, especially for producing high value-added products. The products, fatty acids and glycerol are basic raw materials for a wide range of applications. Fatty acids are used as a feedstock for the production of oleochemicals such as fatty alcohols, fatty amines and fatty esters. These oleochemicals are used as lubricant greases, anti-block IDepartment ofBioprocess Engineering, Faculty of Chemical and Natural Resources Engineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor Bahru, Malaysia. 2Chemical Engineering Pilot Plant, Universiti Teknologi Malaysia, 81310 Skudai, Johor Bahru, Malaysia. Correspondence to : Chew Tin Lee (ctlee@fkkksa.utm.my) HYDROLYSIS OF PALM OLEIN TO OLEIC ACID USING IMMOBILIZED LIPASE agents, plastisizers, and emulsifiers and as ingredients in the manufacture of soaps, detergents, and animal feed [4]. The present method of hydrolysis of crude palm oil to fatty acids and glycerol involves high temperature and pressure operation for about 2 h to achieve the desired 96% - 99% conversion. When these extreme conditions are employed, polymerisation of fat and by-product formation takes place resulting in dark fatty acids and discoloured aqueous glycerol solution [5].To remove the colour and the by-products, further purification by distillation is required. Both hydrolysis and subsequent distillation of fatty acids are energy intensive processes. Hence, it would be advantageous to develop a lower energy process that produces a colourless product [6]. On the other hand, lipases obtained from plants and microbes, which catalyze hydrolysis of oils and fats at the oil-water interface, yielding free fatty acids and glycerol can be effectively and economically used to conduct the same reaction, under mild conditions [7]. However, a reliable kinetic model to predict the hydrolysis rate is still lacking [5]. Though several reports have appeared on the use of lipase for hydrolysis of fats and oils, very few are available on the hydrolysis of palm olein [8]. Also, most hydrolysis studies have been carried out in Erlenmeyer flasks under poorly defined agitation conditions. Hence, only limited information is available on the hydrolysis of palm olein by lipase in a packed bed reactor and there is still a need to further optimize the yield of fatty acid from palm olein. Information from these studies will help to establish the hydrolysis kinetics of palm olein by lipase and for designing a large-scale hydrolysis reactor. 2.0 MATERIALS & METHODOLOGIES 2.1 ｳｬ｡ｩｲ･ｴ｡ｾ The immobilized enzyme used, known as Lipozyme TL 1M, was obtained commercially from Novozymes Co. Lipozyme TL 1M is an immobilized triacylglycerol lipase from Thermonyces lanuginosus produced by submerged fermentation of a genetically modified Aspergillus oryzae microorganism. Palm olein was obtained from Chemical Engineering Pilot Plant (CEPP), Universiti Teknologi Malaysia. Potassium dihydrogen phosphate, sodium hydroxide, tributyrin and hexane were obtained from Fluka Chemie AG, Switzerland. Oleic acid, stearic acid, and methyl oleate were obtained from Merck KGaA, Germany. Borontrifluoride -methanol was obtained from BDH Laboratory Supplies. All chemicals used were of analytical grade. 11 t\ ｾ tl C. T. LEE, Y. L. YAK" F. RAZALI, 1. I. MUHAMAD, M. R. SARMIDI 2.2 Determination of Enzyme Activity The method is based on the hydrolysis of tributyrin by the enzyme and titrating the butyric acids produced with 0.1 M NaOH. 48.5 mL of 10mM potassium dihydrogen phosphate buffer (KH2P04) at pH 7.0 was incubated in a thermostated vessel equipped with a magnetic stirrer at 25°C. After addition of tributyrin (1.47 mL), the pH-meter was started to keep the pH at 7.0. When the pH stabilized, the enzymes (3 mg) were added. The consumption of sodium hydroxide (100 mM) was monitored for 30 minutes. The specific activity of the enzymes was calculated from the base consumption at the linear part of the graph using Equation 1. The enzyme activity is expressed as one unit of enzyme activity corresponds to one micromole of butyric acid liberated from tributyrin per minute per milligram of enzyme at 25°C. . .. S(ml/ s) II 60s 10 6 JOrlol SpeCific activity = x x--x x[NaOH]M WE (g) 1000mI Imin Imol 2.3 (1) Experimental Method The main components of RPBBR are peristaltic pump (Masterflex,Cole-Parmer), water jacketed vessel with the maximum capacity of 50 mL and thermostat jacketed column (Pharmacia Biotech, Sweden) with solvent resistant fittings. Immobilised Lipase was packed in a XK16/20 (16 mm ill x 20 cm length) column. Substrate was fed and recirculated through the column using the peristaltic pump. The reaction mixture was stirred with magnetic stirrer in order to thoroughly mix the substrate and the solvent. A water bath (Grant Intruments, Cambridge, England) was used to maintain the temperature of reaction solution. The reactions were carried out by varying the palm olein concentrations. The experimental setup of RPBBR and its schematic diagram are illustrated in Figure 1. thermometer solution out sampling LJ water bath • stirrer pump Figure 1 Schematic diagram of the experimental setup 12 HYDROLYSIS OF PALM OLEIN TO OLEIC ACID USING IMMOBILIZED LIPASE The concentrations of palm olein were based on the ratio of volume of palm olein to volume of hexane. In order to determine the kinetics parameters, the substrate concentration was increased until a constant rate of product formation was achieved. The selected range of oil to hexane ratio started from OJ to 0.8 with a total volume of38 mL consisted of palm olein and hexane only. A constant volume of 2 mL water was also added into each reaction mixture to make the total volume of the reaction mixture 40mL. The temperature and pH were being set at 37°C and pH 7. The stirrer speed was set at 250 rpm and the flow rate used was 0.5mL.min- l . Samples were withdrawn from the reactor at regular intervals which were 15, 30, 60 and 90 minutes for analyses and the results were used to determine the initial rates of reaction. The initial rates of reaction were calculated using the slope of the hydrolysis profiles' linear portion. All experiments were conducted in triplicate and the values were averaged. 2.4 Determination of Oleic Acid Concentration Gas chromatography was selected to provide data on the production progressive of each fatty acid. In order to determine the oleic acids using a gas chromatography, esterfication of the fatty acids to their respective methyl ester was carried out first. Samples of 0.5 mL were withdrawn from the reaction mixture each time and placed into reagent bottles. One mL of BF3 -methanol was added and nitrogen gas was used to purge the air from the bottles. The bottles were tightly covered and heated at 100°C for 30 min for the reaction to take place. After 30 minutes, the mixtures were cooled and I mL of hexane was added to each bottle to extract the methyl esters. This was followed by adding I mL of distilled water to stop the esterification reaction and to allow phase separation. The mixture was then stirred well to separate the hexane and aqueous phases. Sample of 0.5 mL from the top portion was coIlected and used for GC analysis. The sampel was then injected into the gas chromatograph ( Shimadzu GC-17A ), equipped with a flame ionization detector, and area of each peak was determined. The temperature of injector and detector were maintained at 260°C and the column temperature was kept at 50°C for 2 minutes and then programmed to rise 4°C per minute to reach 220°C within 15 minutes. The retention time for methyl oleate was in the range of 13.7-13.8 minutes. The peaks areas obtained with samples were compared with the calibration curve which was first developed. 2.5 Determination of Kinetics Parameter Concentrations of oleic acid produced against sample time were plotted for each substrate concentration and the initial rate of reaction were obtained from the slope of the graph's linear portion. The initial reaction rates obtained for each substrate concentrations were rearranged and several plots including the lineweaver-Burk equation, Eadie-Hofstee equation and Hanes-Woolf equation were plotted in order to determine the kinetic parameters of Vmax and Km • The equation of lineweaver-Burk, the Eadie-Hofstee equation and the Hanes-Woolf equation are shown in Equation (2), (3) and (4), respectively. 1 K 1 1 [S] Vmax m -=-x-+-Vo Vmax (2) 13 t) C. T. LEE, Y. L. YAK, , F. RAZALJ, I. I. MUHAMAD, M. R. SARMIDI v vmax -K m [S] (3) [S] = K m +_l_[S] v vmax vmax (4) V= 3.0 RESULTS AND DISCUSSIONS 3.1 Enzyme Assay A short chain of triglyceride, tributyrin, was used in the hydrolysis assay of lipases. Tributyrin was selected because it is able to fonn stable emulsions in aqueous media without the addition of stabilizer. The emulsions of tributyrin were hydrolyzed by lipases to fonn dibutyrin, monobutyrin, glycerol and free butyric acid. All of these water-soluble products of tributyrin hydrolysis did not inhibit the reaction by accumulating at the oilwater interface (Wu and Tsai, 2004). The activity assay measured the liberation of butyric acid from tributyrin using titration method. It was conducted to check for the apparent degradation of enzyme l activity. The hydrolytic activity of immobilised Lipase was found to be 8 kU.g- . This activity was assumed constant throughout the whole study. The consumption rate of 100 mM NaOH for the first 30 minutes is shown by the slope of Figure 2, the slope was plotted using Microsoft Excel through linear regression. Volume of NaOH consumed overTime 0.12 0.1 '8 ow a -- = Q > ::c 0 ;z '" 0.08 y = O.004x - 0.0067 0.06 2 R = 0.9813 0.04 0.02 • 0 0 5 10 15 20 25 30 35 TlDle (s) Figure 2 Consumption rate of NaOH in the activity assay 14 HYDROLYSIS OF PALM OLEIN TO OLEIC ACID USING IMMOBILIZED LIPASE 3.2 Hydrolysis Profile In a typical enzymatic reaction, the reaction can be monitored by tracing either the loss of substrate or the formation of product. In this study, the formation of oleic acid as product was followed. Figure 3 shows the hydrolysis profile of palm olein at an initial palm olein concentration of 0.4206 moI.L- I . The reaction conditions for all the runs were set at 37°C and an agitation speed of250 rpm, without the addition of surfactant. Based on Figure 3, the progress curve goes through 3 stages, in the first 15 minutes, there was a lag phase where the presence of product in the reaction medium were not significant (0.00004 mol.L- 1 to 0.0005 mol.L- I ), the substrates was yet to bind with the desired conformation of the enzymes active site and the enzymes also required certain period of initial time to contact with the droplets of oil. From 15 to 30 minutes, the concentration of oleic acid increased exponentially and the rate of reaction was maximum. This robust rate of reaction was achieved as the active sites on the enzyme were saturated with the substrate molecules. After 30 minutes, the concentration of fatty acid did not show further increment, indicating that the reaction had reached its equilibrium state. Thus, three parts of the kinetic curve can be distinguished, the lag phase, the initial increasing part and the plateau region. Time course of reaction for the hydrolysis of 0.4428 mol.VI palm olein. -- 0.02 !, 0.015 ｾ c; 'l:l '<;j ｾ ... 'il ....c;c I: c ..... = ... c = U 0.01 y = 0.0007x - 0.0 I R2= I 0.005 '';: ｾ Q< 0 40 60 80 100 -0.005 Time (min) Figure 3 Concentration of oleic acid produced versus time at an intial substrate concentration of 0.4206 moI.L- I . Similar hydrolysis profiles were obtained for other experiments with different initial substrate concentrations ranging from 0.3155 mol.L- 1 to 0.8412 mol.L- 1 as shown in Figure 4. 15 ｾ I C. T. LEE, Y. L. YAK" F. RAZALI, I. 1. MUHAMAD, M. R. SARMIDI Time Course of Reaction for the Hydrolysis of 0.3-0.8 v/v palm olein 0.Q7 1 0.06 'tl ｉｾ 0.05 I]"" ! 0.04 0 .-, 'Q S = => '1.51 E \ ｾ :] - I L '-' 0.03 IS] =04206mollL fS) =O.5258mol/l -e-- [S) =O.6309molll 0.02 ｾｏＺ 0 01 ｾｩＺ］ ｾ Ａ Ｚ ｌ ｾ - - f S ) =0.736 Imolll S 20 40 W Time (min) '--1O-:_f _]=_O_.8_41_2m-O_III 80 . J _ Figure 4 Concentrations of oleic acid produced versus time for the hydrolysis of 0.3155 mol.L- l to 0.8412 moLL-I palm olein Each time course of reactions was followed by gas chromatography. Reactions were stopped at 90 minutes in order to save time because many repetitive experiments at various initial concentrations of substrate had to be performed in order to determine the Km and Vmax• The real enzymatic reaction mechanism is complex anddifficult to decipher. This is mainly due to the interactive nature of biological processes. According to Figure 4, it is observed that for concentration ranging from 0.5258 moLL-) to 0.8412 moLL-I, the rates of reactions were increased for up to 90 minutes while for substrates concentration ranging from 0.3155 mol.L-) to 0.4206 moLL-I, the plateau had been reached at 30 minutes indicating that the reaction had reached its equilibrium state. 3.3 Data Analysis for Kinetics Studies The kinetics studies were conducted based on the initial velocity analysis. From the hydrolysis profiles, the initial rate of hydrolysis for each substrate concentration was determined using the slope of the linear portion of their progress curves. The initial reaction rates, V o at different palm olein concentration, [S] and all the rearrangement values of V o and [S], namely Ilv o, I/[S], vJ[S] and [S]/v o are summarized in Table 1. The initial rate of hydrolysis was expressed in moles (oleic acid produced) per L per min. The data in Table I were used to determine the value of Vmax and K m by using the Lineweaver-Burk plot, the Eadie-Hofstee plot and the Hanes-Woolfplot. Initially, six data points (option 1) from Table I were used to plot the three plots. However, in order to get the more precise values of Vmax and Km , two data points have been eliminated. In other words, only four data points (option 2) were considered and shown in Figure 5, 6 and 7. 16 HYDROLYSIS OF PALM OLEIN TO OLEIC ACID USING IMMOBILIZED LIPASE Table 1 Initial reaction rates and other rearrangement values for the hydrolysis of 0.3155 moLL-I to 0.8412 mol.L- 1 palm olein. [S] (mol.L- I ) 1/vo (L.s.mol: l ) 1/[S] (L.mor l ) [S]/vo (s) vJ[S] (mol.L-1. S-1) 0.3155 0.4206 0.5258 0.6309 0.7361 0.8412 0.0005 0.0007 0.0009 0.001 0.001 0.001 2000 1429 1111 1000 1000 1000 3.17 2.38 1.90 1.58 1.36 1.19 0.0016 0.0017 0.0017 0.0016 0.0014 0.0012 631 601 584 631 736 841 Vo (S-I) Lineweaver-Burk Plot 1200 1000 y= 153.16x+796.73 R2 =07231 400 200 o II1S1 (mollL)-1 -I -2 2 3 Figure 5 Lineweaver-Burk Plot Eadie Hofstee Plot 0.0014 ｾ .5 .E :sa... . " .!:! Q Q e '1 0.0012 .... . 0.0010 00008 00006 y =-0.1540 x +0.0012 R2 =0.5148 , 0.0004 0.0002 0.0000 0.0000 0.0005 0.0010 0.0015 0.0020 v"lS (min)"1 Figure 6 Eadie-Hofstee plot 17 C. T. LEE, Y. L. YAK" F. RAZALI, J. 1. MUHAMAD, M. R. SARMIDI Hanes-W oolf plot 1000 y = 833.33x + J 28.52 Rl =0.974 800 --. 600 ｾ 400 ｾ ｾ 200 0 0.2 0 0.4 0.8 0.6 Substract concentration, [S\ (moI/L) Figure 7 Hanes-Woolf plot The apparent kinetic parameters including the maximum reaction rate (V max) and the Michaelis constants (K m) of immobilized lipases obtained from the three types of plots are summarised in Table 2. Table 2 Summary of the kinetic parameters Type of Plots (mol.L-I.min- l ) Vmn Km (moJ.L- I) Lineweaver-Burk plot 0.00126 0.192 Eadie-Hofstee plot 0.00120 0.154 Hanes-Woolf plot 0.00120 0.154 Average 0.00122 0.167 An average value of V max and K m was obtained from Lineweaver-Burk Plot, Eadie-Hofstee Plot and Hanes-Woolf plot. The value was 0.00122 mol.L-I.min- 1 for V max and 0.167 mol.L- 1 for Km • It is quite difficult to determine the best method of data plotting to be used in order to determine the value of V max and Km • In any case it should be emphasized that the data should be as good as possible. The lineweaver-Burk plot is still the most common method being used and it has the advantages that the variables v and [S] are plotted on separate axes. However, an analysis of the errors involved in the collection of the data (and hence in the determination of the parameters K m and V max) showsx that there was a highly non-uniform distribution of errors over the range of values of I Iv and I/[S] in the lineweaver-Burk plot. For this reason, the plots from Eadie-Hofstee and Hanes-Woolf have been recommended as the distribution of errors in these plots were more uniform. According to Steven and Lee (2005), who used the same substrate and the same loading of the same enzyme but different reactor which was a batch stirred tank reactor and without the addition of water. The values of V max and Km they reported were 0.00114 18 HYDROLYSIS OF PALM OLEIN TO OLEIC ACID USING IMMOBILIZED LIPASE mol.L-I.min- 1 and 2.53 mol.L- 1 ,respectively. Their Vmax compared reasonably well with the values obtained in this study whereas their Km was significantly larger compared to 0.167 moI.L-) obtained in this study. It is known that a small value of K m indicating a high affinity of enzyme towards the substrate, hence it was likely that the addition of small amount of water in this study (2 mL water over 40 mL total reaction mixture) was favorable to improve the affinity between substrate-enzyme due. 3.4 Effect of Substrate Concentration on Rate of Reaction Based on the data from Table 1, a Michaelis-Menten Plot is plotted as Figure 8 which shows the effect of palm olein concentration on the initial rate of reaction. Michaelis-MeDten Plot 0.0012 C .... -..5 ... .JE .$ co ｾ .::co= .. ;S .. c: 0.001 0.0008 <:> 00006 ! 0.0004 -I ｾ 0.0002 0 0 0.2 0.4 06 08 Concentration of substract, IS I (mollL) Figure 8 Michaelis-Menten plot Initially, the oil concentration was varied from 0.3155 mol.L-1 to 0.6309 mol.L- I. In this range of substrate concentration, the initial rate of hydrolysis varied linearly with substrate concentration. Noor et al. (2003) also observed that the rate of hydrolysis of tallow, coconut oil and olive oil by lipase from Candida rugosa varied linearly with oil concentration. Experiments were also conducted at high oil concentrations, up to 0.8412 mol.L- 1 to check the substrate inhibition. However, these experiments showed that no substrate inhibition was observed for oil concentration up to 8412 mol.L -I. 4.0 CONCLUSIONS The hydrolytic activity of immobilised Lipase was found to be 8 kU.g- 1 and this activity was assumed constant throughout the whole study. The kinetics studies were carried out using initial velocity analysis. Vmax, the maximum rate of reaction and Km , the MichaelisMenten constant were detennined based on Michaelis-Menten model and their values were derived from the Lineweaver-Burk plot, Eadie-Hofstee plot and Hanes-Woolf plot. The average values of Vmax and Km obtained from these three plots were 0.00122 mol. The investigated range of substrate L-1.min-1 and 0.167 mol.L- 1 respectively. 19 C. T. LEE, Y. L. YAK, , F. RAZALI, I. I. MUHAMAD, M. R. SARMIDI concentration was varied from 0.3155 mol.L- 1 to 0.8412 mol.L- 1• No substrate inhibition was observed for palm olein concentration of up to 0.8412 mol.L- I . ACKNOWLEDGEMENTS The authors are grateful to Chemical Engineering Pilot Plant UTM for supply of raw materials and the Faculty of Chemical & Natural Resources Engineering, UTM for support funding of this study. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] Halling, PJ., A.E. Janssen, A.M.Vaidya. 1996. Substrate Specificity and Kinetics of Candida rugosa Lipase in Organic Media. Enzyme Microb. Technol. 18:340346. Chang, J.H., S.c. Lee, W.K. Lee. 1999. Effects of Preparation Variables of Enzyme-Encapsulating Water-in-Oil Emulsion on Enzymatic Reaction Conversion and Emulsion Stability in an Enzyme-Emulsion-Liquid-Membrane Reactor. Chem. Eng. J. 73:43-51. Bilyk, A., R.G. Bistline, R.G. 1991. Lipase-Catalyzed Triglyceride Hydrolysis in Organic Solvent. J Am. Oil Chem. Soc. 68:320-323. Long, W.S., A. Harun, Kamaruddin, S. Bhatia.2003. A Comparative Study of Palm Oil Hydrolysis by C. Rugosa Lipase in Packed Bed Reactor: Covalent Bound VS. Adsorbed to Amberlite MB-l. Journal ofIndustrial Technology. 12: 37-55. AI-Zuhair, S., M. Hasan, K.B. Ramachandran. 2003. Kinetics of the Enzymatic Hydrolysis of Palm Oil by Lipase. Process Biochemistry. 38: 1155-1163. Arbidge, M.V., W.H. Pitcher. 1989. Industrial Enzymology: A Look Towards the Future. Trends Biotech. 7:330-335. Sharon, c., M. Nakazato, H.I. Ogawa, Y. Kato. 1998. Lipase-Induced Hydrolysis of Castor Oil: Effect of Various Metals. Journal of Industrial Microbiology & Biotechnology. 21 :292-295. Noar, M., M.Hasan, K.B. Ramachandran. 2003. Effect of Operating Variables on the Hydrolysis Rate of Palm Oil by Lipase. Process Biochem. 39: 13-20. Steven, A.T. C.T. Lee. 2005. Hydrolysis of Palm Olein to Fatty Acid using Immobilized Lipase. Universiti Teknologi Malaysia: Thesis for Bachelor of Chemical Engineering (Bioproses). 20