Kinetics of the enzymatic hydrolysis of palm oil by lipase

Process Biochemistry 38 (2003) 1155
/1163 www.elsevier.com/locate/procbio Kinetics of the enzymatic hydrolysis of palm oil by lipase Sulaiman Al-Zuhair, Masitah Hasan, K.B. Ramachandran * Department of Chemical Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia Received 28 March 2002; received in revised form 23 August 2002; accepted 18 September 2002 Abstract The kinetics of the enzymic hydrolysis of palm oil using lipase in a batch reactor has been investigated. The lipase enzyme used was not ester bond position selective and its activity at the interface was higher compared to that in the bulk. A mathematical model taking into account the mechanism of the hydrolysis reaction and the effect of interfacial area between the oil phase and the aqueous phase containing the enzyme was developed. A correlation between the interfacial area and the operating conditions including agitation speed and oil volume fraction was established experimentally. The kinetic parameters were estimated by fitting the data to the model and comparing with previously reported values. The kinetic model represented the experimental data accurately. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Lipase hydrolysis; Palm oil; Kinetic model; Interfacial area 1. Introduction Hydrolysis of oil and fat is an important industrial operation. 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 agents, plastisizers, and emulsifiers and as ingredients in the manufacture of soaps, detergents, and animal feed. 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 [1]. 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. 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 * Corresponding author. Tel.: /60-3-7967-5293; fax: /60-3-79675319. E-mail address: kbram@um.edu.my (K.B. Ramachandran). [1]. Hence, it would be advantageous to develop a lowenergy process that produces a colourless product. Recently, enzymic splitting of fats has gained increasing attention, as lipase (triacylglycerol acylhydrolase) is now available at reasonable cost. 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 or heat sensitive fatty acids [1]. However, a reliable kinetic model to predict the hydrolysis rate is still lacking. Lipase catalysed reactions take place at the interface between the aqueous phase containing the enzyme and the oil phase [2,3]. Hence, the interfacial area, which is affected by mixing and substrate concentration, influences the rate of reaction. All previous studies to establish a rate equation for the enzymic hydrolysis of lipids in batch reactors have assumed that the total interfacial area between oil phase and the aqueous phase remains constant, even when the agitation speed or substrate concentrations are varied. This assumption is valid only if the substrate (oil) is dissolved in an organic solvent (such as hexane), its concentration is changed within that organic phase and the volume fractions of the organic phase containing the substrate and the aqueous phase containing the enzyme are kept constant. Although the method of dissolving the oil in the organic phase and its subsequent hydrolysis using lipase enzyme 0032-9592/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0032-9592(02)00279-0 1156 Nomenclature a at Am C C* Dmean Do E E* E *S Et (Et)m k kcat kd kp k k1 Ke ? Km LU m n P* P S T Greek letters a f u v n S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 Specific free interfacial area (m 1) Specific total interfacial area (m 1) Enzyme molar area (m2 mol1) Proportionality constant defined by Eq. (8) Constant defined by Eq. (13) Surface mean diameter (mm) Overall mean diameter (mm) Free enzyme (mole/total reactor volume) (mol m 3) Penetrated enzyme (mole/total interfacial area) (mol m2) Enzyme
/substrate complex (mole/total interfacial area) (mol m 2) Total active enzyme (mol m 3) Total enzyme mass concentration (g m 3) Constant defined by Eq. (19) Catalytic rate constant (min 1) Desorption rate constant (min1) Adsorption rate constant (m2 min 1) Reaction rate constants (m3 mol 1 min 1) Reaction rate constant (min 1) Equilibrium constant of E *S (mol m 3) Apparent Michaelis constant (mol m 3) Lipase unit Constant defined by Eq. (19) Constant defined by Eq. (19) Interface product concentration (mole/total interfacial area) (mol m 2) Bulk product concentration (mole/total reactor volume) (mol m 3) Bulk substrate concentration (mole/total reactor volume) (mol m 3) Temperature (K) Constant defined by Eq. (19) Volume fraction of oil in the reaction mixture Area fraction Agitation Speed (rpm) Reaction rate (mol m 3 min1) has several advantages, it requires the addition of an organic solvent in the reaction mixture, which needs to be separated later for reuse. The preferable method for the enzymic hydrolysis of the oil is to bring directly in contact the aqueous phase containing the enzyme and the oil phase. For this system, it is not correct to assume that the total interfacial area is constant, irrespective of the intensity of agitation and the volume fraction of oil and water. Hence, to verify the kinetics, a model taking into account the change in interfacial area with agitation speed and substrate concentration is needed. Such a model will be useful in optimal design of a batch or a continuous hydrolysis process. The enzymic lipolytic reaction is an important example of an heterogeneous catalytic reaction. The watersoluble enzymes act at the interfaces of the insoluble lipid substrate. The X-ray crystallography technique has been used to show the existence of a preferable config- uration change of the enzyme at the water
/lipid interface [4]. This interfacial activation phenomenon is thought to be due to the unfolding of an amphiphilic peptidic loop, covering the active sites of the enzyme when the enzyme is attached to the lipid at the interface. When contact occurs with a lipid
/water interface, the enzyme undergoes a conformational rearrangement, rending the active site accessible to the substrate. It is suggested in this study that determining the activity of the enzyme at the interface, and comparing it to that in the bulk, would help to strengthen this understanding. Determination of the fatty acids produced from the enzymic hydrolysis of oil as a function of time is widely addressed in the literature, being the direct way to determine the reaction rate. This can be done by two methods, namely: (1) titration of the products after extraction using a auto-titrator and (2) gas chromatographic determination of the fatty acids produced. Unlike the titration method that gives the overall fatty acids concentration, the gas chromatographic method 1157 S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 provides data on the progressive production of each fatty acid. The results of the gas chromatography would help in determining the selectivity of the enzyme with respect to the ester-bonds in the glycerol chain of the palm oil, to hydrolyse. It is known that some lipases favour the terminal ester bonds of the oil glycerol chain, and are called 1,3 lipases [5]. In most oils, the saturated fatty acids are naturally situated on the terminal positions of glycerol, while the unsaturated ones are in the centre. Examining the progressive production of different fatty acids, saturated such as palmitic acid, and unsaturated such as oleic acid, would help to determine whether or not the lipase used prefers to attack the terminal-ester bonds. The rate of product formation can be expressed as: y dP  dt kp E a?E kd k1 ES ? ES k1 kcat ES 0 EP (1) (2) (3) The concentration of the enzyme
/substrate complex and the adsorbed enzyme are both assumed constant (quasi-steady state) [2,7], and the interfacial product concentration, P *, is assumed to be proportional to the free product concentration, P [2]. It is also assumed that the interfacial product concentration, P *, is low and hence it occupies negligible fraction of the total interfacial area. With the above mechanism and assumptions, the model equations can be written as: kp E × a(kd k1 S)E(k1 kcat )ES 0 k1 E × S (k1 kcat )ES 0 at aAm (EES)at Et E at (EES) Since, (4) (5) (6) (7)  at C kcat ES (9) Solving Eqs. (4)
/(9) simultaneously, the final equation for the rate of reaction can be expressed as: kcat at (G1  G2 ) CAm y kcat  k1 k1 G1  The mechanistic model proposed to describe the action of lipase on palm oil hydrolysis is similar to that proposed by Tsai and Chang [2]. The first step is the reversible adsorption of a water-soluble enzyme at the interface to produce a penetrated enzyme, E *. In order to develop the model equations, the absorption rate is assumed to be proportional to the free enzyme concentration, E and the specific free interfacial area, a. The substrate, S , then binds to the adsorbed enzyme giving an interfacial enzyme
/substrate complex, E *S [2,6]. This complex then generates the product, P * at the interface, while regenerating the enzyme in the form of E *. The product, P *, then desorbs from the interface into the organic phase to give rise to product P . The steps up to the production of product P * are illustrated in Eqs. (1)
/(3) at dP C dt where
2. The kinetic model (8) PCP=at S (10) S kcat  k1 k 1 at 

 kd at  Am Et  at  Am Et  S at at 
kcat  k1 S k1 (11)
G2  G12  4Am Et at  (12) and C2C (13) Eqs. (10)
/(13) are the corrected version of the equations derived by Tsai and Chang [2] for the hydrolysis of lipids by lipase enzyme. This equation is applicable for predicting the hydrolysis rate for any enzyme concentration. Hydrolysis reactions are usually carried out at low enzyme concentration and it is useful to get a simplified rate expression, applicable for such a condition. At low enzyme concentrations, it has been shown experimentally that the area occupied by the enzyme that has penetrated the interface is negligible in comparison to the total interfacial area [2,7]. In this case, the free specific interfacial area, a , will be equal to the total specific interfacial area, at. Replacing a with at in Eqs. (4)
/(10), a simplified form of rate equation at low enzyme concentration can be derived as given below: kcat Et S C y
 (kcat  k1 )(kd  kp a2t ) k1 kp a2t (14) S The above rate equation can be further simplified to: y Ke + kcat Et S  kd 1 S kp a2t
+ where Ke (kcat k1 )=k1 and kcat kcat =C/ (15) 1158 S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 The above rate expressions agree in the basic form with the previous models [2,6,7]. However, the rate constants in the previous models were based on proportionality constants that had no link to the reaction mechanism and this led to lumped rate constants. 3. Materials and methods 3.1. Materials Lipase (Type-VII) from Candida rugosa was obtained from Sigma Chemical Co., Japan. Refined palm oil used in this study was obtained from Lam Soon (M) Berhad, Malaysia. Analytical grade isopropanol was obtained from Scharlau Chemicals Co., Spain. Gum Arabic, oleic acid methyl ester, linoleic acid methyl ester, stearic acid methyl ester and palmitic acid methyl ester were obtained from Sigma Chemicals Co., Germany. All other chemicals used were of analytical grade. 3.2. Hydrolysis reactor The reactor consisted of a glass flask with a capacity of 600 ml and an inside diameter of 9 cm. A four-bladed paddle impeller (5 cm in diameter) immersed in the solution at one-third-depth level was used for agitation. No baffle plates were provided. The total volume of the reaction mixture was 400 ml at the beginning of each run, consisting of palm oil and distilled water. The reactor was placed in a temperature controlled water bath and a cover was used to prevent evaporation of water during the progress of the experiment. After the desired conditions were reached in the bioreactor, the required amount of enzyme solution was added to initiate the reaction. Samples were withdrawn from the reactor at regular intervals for analyses. using the method proposed by Rooney and Weatherley [9]. One gram of enzyme was dissolved in 100 ml of distilled water and coated with a thin layer of palm oil. The mixture was kept at room temperature for about 2 h in a 100-ml beaker. The temperature was then reduced to /20 8C by placing the beaker in a freezer and left overnight. The solid upper oil layer was then peeled off and the solid ice surface wiped clear of any remaining oil. A sharp knife was used to scrap off small samples from the interface and its activity was then determined as described in the previous paragraph. 3.4. Determination of fatty acids concentration A gas chromatograph was used to determine the fatty acid composition of the hydrolysed samples. In order to determine the fatty acids using a gas chromatograph, esterification of the fatty acids to their respective methyl esters was carried out first, following the AOCS Official Method Ce 1
/62 [10]. After esterification, a sample of 1 ml of the fatty acids methyl ester was withdrawn into a 10-ml syringe. The sample was then injected into the gas chromatograph (Chemito GC 8610), equipped with a flame ionisation detector, and area of each peak was determined. The column was conditioned prior to use and the operating conditions used for the analysis were, oven temperature 200 8C, injector and detector temperatures 240 8C. The peaks areas obtained with samples were compared with the peaks areas obtained with three different dilutions of the standards, prepared by dissolving different weights of oleic, linoleic, stearic and palmitic acid methyl esters in heptane. These are the main fatty acids produced by the hydrolysis of palm oil. A straight-line method was used to determine the amount of fatty acids in the reactor sample. 3.5. Determination of oil drop size 3.3. Determination of enzyme activity The method is based on the hydrolysis of tributyrin by the enzyme and titrating the butyric acids produced with 0.05 NaOH in distilled water [8]. The alkali consumption is registered as a function of time under standard conditions of 30.0 8C and pH 7.0, using an auto-titrator (Metrohm 702 SM titrino). From the amount of alkali consumed, the equivalent amount of butyric acid in the samples was calculated and the enzyme activity determined. The enzyme activity is expressed in lipase unit (LU), where the 1 LU is defined as the amount of enzyme which liberates 1 mmol titrable butyric acid per min at 30 8C. In addition to determining the activity of the enzyme used for the kinetic studies, additional analyses were carried out to compare the activity of the enzyme in the bulk to that at the interface. The interface was created The method proposed by Mukataka et al. [7] was used to determine the specific interfacial area at various operating conditions. Five minutes after mixing the oil and the aqueous phase, a sample was withdrawn and placed on a slide for observation under microscope. A microscope, model Leica DMLS, linked to a computer software image analyser, VIDEO TEST
MASTER 4.0, was used to determine the droplets size of palm oil in water. Around 200
/400 photographs of drops, which were enlarged 4
/20 times were taken. The drops were divided into five categories: drops of diameter smaller than 100 mm, 100
/200 mm, 200
/300 mm, 300
/400 mm, and larger than 400 mm. The mean diameter of each category was calculated using the following equation: X X Dmean; j  dj3 dj2 (16) / = S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 The overall surface mean diameter, which takes into account the respective portions of each size category in the oil/water mixture, was calculated using the equation: X (uj =Dmean;j ) (17) Do 1 = where the subscript j stands for each of the five different size categories. From the values of the overall mean diameter, interfacial area per unit volume of the oil/water mixture in the bi-phasic system was calculated using the following Eq. (18). at 6f=Do (18) where f is the oil volume fraction. For each pre-specified operating conditions, the above procedure was repeated three times; after 5, 10 and 15 min from the beginning of each experimental run and a statistical average of the three runs was determined. It has been found that the results were reproducible and the standard deviation of the different runs under the same operating conditions was in the range of 2.1
/57.8 mm, which means the errors have not exceeded 10%. Many models have been previously proposed to correlate the specific interfacial area as a function of agitator speed and volume fraction at constant temperature [11
/13]. Calderbank [13] also included in his correlation the effect of temperature on the average drop diameter. After considering the equipment parameters and the physical properties of the fluids in the experiment, the most suitable correlation for these studies is that of Calderbank [13], given below: at  6f Do avm fT k =( 1nf) (19) The above equation was used to correlate the total specific interfacial area to various operating conditions. 3.6. Determination of hydrolysis rate The enzyme mixture was prepared by dissolving different weights of solid lipase powder in 100 ml of distilled water and 10 ml of this solution was added to the reaction mixture. Samples of 10 ml volume were withdrawn from the reaction mixture at desired time intervals. The water in the samples was evaporated under a vacuum of 300 mbar using Buchi Rota Vapor (R-144) and the remaining fatty acids were extracted in 50 ml of isopropanol. The extracted fatty acids were titrated with 0.05 N NaOH solution in isopropanol to determine the concentration of fatty acids, using an auto-titrator. A plot of fatty acid concentration (mol m 3) against sample time was made and the slope of the plot at the origin gave the initial rate of reaction. 1159 4. Results and discussions 4.1. Interfacial enzyme concentration To determine the lipase activity, at the interface, five repetitive experiments were carried out and a statistical average was determined. The results showed that the activity of the enzyme at the interface was 15.7% higher than that in the bulk. This observation agrees with the result reported by Rooney and Weatherley [9] who found that the activity at the interface was higher than that in the bulk, however, they did not determine the percentage increase. The increase in the activity at the interface was explained using X-ray studies of the 3D structure of lipase [4] and the conformation change of the enzyme at the interface. This encourages the enzyme to migrate to the interface from the bulk and expose the active sites by folding the protein chains in a preferable manner. 4.2. Fatty acids profile The percentages of the main saturated fatty acid, palmitic acid, and the main unsaturated one, oleic acid produced with time is shown in Fig. 1. The figure shows that both percentages remained almost constant with time and equalled to that found in the palm oil itself, i.e., 45.5% palmitic acid and 39.8% oleic acid [5]. This indicates that the lipase used was not a position selective enzyme, otherwise, the production of palmitic acid would be much faster than the production of oleic acid. In Fig. 2, the results obtained by the gas chromatograph are compared to the results found by the auto-titrator method. The average difference in readings between these two methods is about 10%, which is relatively low. Since, the auto-titrator method is less time consuming, it was used to determine the reaction rate in subsequent experiments. Fig. 1. Percentages of palmitic acid and oleic acid produced with time. 1160 S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 Fig. 2. Comparison between the gas chromatograph and the autotitrator results (f/0.2, T /45 8C, v /1300 rpm, and (Et)m /25 g m 3). 4.3. Determination of interfacial area The effect of different stirrer speeds and different oil volume fractions on the total interfacial area was measured at 45 8C and the results are shown in Fig. 3. It can be clearly seen that increasing the agitation speed and/or the volume fraction resulted in increased total interfacial area. This observed effect of agitation speed is due to increase in shear rate on the oil droplets with increasing agitation speed that caused the breakage of the larger oil droplets into smaller ones. These results agree with the earlier reported results [11,12] and the basic empirical model proposed by Calderbank [13]. However, at high volume fraction, the increase in interfacial area tended to decrease with increasing agitation speed as suggested by the model of Tavlarides and Bepat [12]. The effect of temperature on the total interfacial area was not reported either by Albasi et al. [11] or by Tavlarides and Bepat [12], but was reported by Tsai and Chang [2]. The effect of temperature on the droplet size was determined and is shown in Fig. 4. It shows that the mean drop diameter decreases as the temperature increases. This is due to the reduction in viscosity of Fig. 3. Effect of stirrer speed and oil volume fractions on the specific interfacial area (T/45 8C). Fig. 4. Effect of temperature and agitation speed on the mean drop diameter (f/0.5). the oil and the surface tension at the interface between water and oil, with increase in temperature [13]. The data shown in Figs. 3 and 4 were fitted to Eq. (19) and the model parameters in the equation were estimated by applying a stepwise numerical method using EXCEL. The resultant correlation for the total specific interfacial area is as follows: at  0:024v0:6 T 1:7 f=(13:0f) (20) 4.4. Hydrolysis reaction Experiments were run at different oil volume fractions, temperatures, enzyme-concentrations and stirrer speeds to determine their effect on the initial rate of palm oil hydrolysis. Fig. 5 shows the effect of oil volume fraction, which reflects the substrate concentration, on the initial rate of reaction. It can be noted that, as the volume fraction of oil increased the initial rate of reaction also increased. The increase in reaction rate with volume fraction at a given agitation speed is due to an increase in substrate concentration and the increased interfacial area, as reported in the previous section. However, above 30
/40% oil (v/v), a slight decrease in Fig. 5. Effect of oil volume fraction on the initial rate of reaction at various stirrer speeds (T /45 8C and (Et)m /25 g m 3). S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 the initial rate of reaction was observed. This result agrees with the results previously presented in the literature [7,14], which showed that the initial rate of reaction decreased above an oil volume fraction of 43%, when the unfavourable phase inversion begins to take place (i.e. the aqueous phase becomes the dispersed one). When phase inversion occurs, dispersion of the enzyme at the interface will be restricted, as it would be trapped in the water droplets, where the agitation effect is limited. It can also be observed from Fig. 5 that at a given volume fraction of oil, the initial rate of reaction increased as the agitation speed increased. This again is due to the increase in mean interfacial area with increase in agitation speed as reported in the previous section. These results also agree with previous literature reports [11,13,14]. The effect of increasing enzyme concentration on the initial rate of reaction is shown in Fig. 6. The initial rate of reaction increased linearly with enzyme concentration at low enzyme concentrations. At high enzyme concentrations, this increase tends to fall. This result agrees with that found by Albasi et al. [11] for the hydrolysis of sunflower oil. This phenomenon is explained by hypothesising that at high concentrations of enzyme the interfacial area is totally saturated with enzyme molecules. Hence, any further increase in enzyme concentration in the bulk would not enhance the reaction rate. It can also be seen, from Fig. 6 that the point where the effect of enzyme concentration tended to fade, shifts to the right (higher enzyme concentration) as the agitation speed increased. This is due to increase in interfacial area available for the enzyme to occupy at higher agitation speeds and hence, a higher enzyme concentration is needed to saturate the available interfacial area. A multiple regression method, using a MATLAB computer package was applied to the data shown in Figs. 5 and 6, to determine the parameters of the model (Eq. (15)). Under these conditions, the enzyme concentration used was low enough to assume that the area of coverage of the enzyme is negligible. The model Fig. 6. Effect of enzyme concentration on the initial rate of reaction at different stirrer speeds (f/0.20 and T/45 8C). 1161 equation with the estimated rate constants can be represented as given below: y 1:8  103 (Et )m S 90:018
 7:7  107 5:65 1 S a2t (21) 4.5. Effect of temperature Fig. 7 shows the effect of temperature on the initial rate of reaction. It shows, initially as the temperature increased, the reaction rate increased. This is mainly due to increase in rate constant with temperature and partly due to increase in interfacial area with temperature as discussed in the previous section. However, the initial rate decreased sharply after 50 8C, which was due to the deactivation of the enzyme. It is known that most proteins tend to decompose at temperatures above 50 8C [15]. In addition to the deactivation of the enzyme, the presence of the inactive enzyme at the interface would block the active enzyme from penetrating the interface, which would further decrease the reaction rate. The experimental results shown in Fig. 7 were used to determine the effect of temperature on the rate constant, kcat. + kcat  9:6103 exp(1:2=RT) (22) The activation energy of 1.2 kcal gmole 1 is lower than the reported activation energies for most enzyme reactions [15]. Kim and Chung [16] reported a value of 7.0 kcal gmole 1 for the hydrolysis of palm kernel oil in reversed micelle systems. Desnulle [17] reported in an aqueous emulsion system, the activation energy was 5.3 kcal gmole 1. Our result is lower, but still comparable to those reported in the literature. Fig. 7. Effect of temparature on the initial rate of reaction at various oil volume fraction (v /800 rpm and (Et)m /25 g m 3). 1162 S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 Fig. 8. Comparison between experimental results and the proposed model curve showing the effect of substrate concentration on the initial rate of reaction (v /800 rpm, T /45 8C and (Et)m /25 g m 3). Fig. 10. Comparison between experimental results and the proposed model curve showing the effect of enzyme concentration on the initial rate of reaction (v /1000 and 1300 rpm, T /45 8C and S /660.7 mole m 3). 4.6. Model verification Figs. 8 and 9 show the comparison between the experimental results and the proposed model curve based on Eq. (21), for two agitation speeds. It can be noted that the model predicts fairly well the initial rate of reaction at various substrate concentrations for both agitation speeds. It can be seen from Fig. 10 that the experimental results are well represented by the model at low enzyme concentrations. However, they tend to diverge at high enzyme concentrations, as the model is limited to low enzyme concentrations only. As discussed earlier, at high concentrations of enzyme, the interfacial area is mostly saturated with enzyme molecules and hence the assumption that the area of occupation by the enzyme is only a fraction of the total interfacial area in deriving Eq. (21) breaks down. It is also seen from Fig. 10 that the enzyme concentration at which the experimental results are well represented by the model, is wider at higher agitation speeds. This is due to large interfacial area available for the enzyme to penetrate the interface at higher agitation speeds, and hence, even at higher enzyme concentration the interface is not satu- Fig. 9. Comparison between experimental results and the proposed model curve showing the effect of substrate concentration on the initial rate of reaction (v /1000 rpm, T /45 8C and (Et)m /25 g m 3). rated with the enzyme. Under these conditions a low interfacial area of occupation by the enzyme is valid for a wider range of enzyme concentration. The experimentally determined values of apparent ? by Mukataka et al. [7] Michaelis
/Menten constant, Km at different agitation speeds are compared in Fig. 11, with those determined by the present model, under the same operating conditions. It can be noted that the ? in the present study are lower than that apparent Km reported by Mukataka et al. [7]. In the experiments of Mukataka et al. [7], the organic phase consisted of 10% beef tallow dissolved in isooctane, whereas the organic phase in this study consisted of palm oil only. The difference could be due to the different substrate used in both studies. Fig. 11 also shows that the specific interfacial area determined in the present study, using palm oil, was lower than that obtained by Mukataka et al. [7] for beef tallow dissolved in isooctane, under the same operating conditions. This is due to the high viscosity of palm oil in comparison with isooctane
/ tallow mixture. Fig. 11. Comparison between the present model results and the results of Mukataka et al. [7]. S. Al-Zuhair et al. / Process Biochemistry 38 (2003) 1155
/1163 5. Conclusion A kinetic model based on the mechanism of the reaction of the lipase-catalysed hydrolysis of palm oil in bi-phasic oil
/aqueous system has been proposed, taking into account the variation of interfacial area with agitation speed and substrate volume fraction. This was verified with experimental results at low enzyme concentrations. There was a good agreement between the model prediction and the experimental results. The rate constants in the mathematical model were determined numerically from the experimental results. This model can be used to predict the rate of hydrolysis in a batch reactor and to determine optimal conditions. It has been shown experimentally that the lipase enzyme used was not ester bond position selective. In addition, the activity of the enzyme was shown to increase at the interface compared to that at the bulk, as previous studies have also indicated [9]. References [1] Arbidge MV, Pitcher WH. Industrial enzymology: a look towards the future. Trends Biotech 1989;7:330
/5. [2] Tsai SW, Chang CS. Kinetics of lipase-catalysed hydrolysis of lipids in biphasic organic-aqueous systems. J Chem Tech Biotech 1993;57:147
/54. [3] Knezevic ZD, Siler-Marinkovic SS, Mojovic LV. Kinetics of lipase-catalysed hydrolysis of palm oil in lecithin/isooctane reversed micelles. Appl Microbiol Biotechnol 1998;49:267
/71. [4] Panalotov I, Verger R. Physical chemistry of biological interfaces. New York: Marcel Dekker Inc, 2000. 1163 [5] George S, Arumughan C. Positional distribution of fatty acids in the triacylglycerols of developing oil palm fruit. J Am Oil Chem Soc 1993;70(12):1255
/8. [6] Verger R, Maria CEM, Gerard HD. Action of phospholipase A at interfaces. J Biological Chem 1972;218(11):4028
/34. [7] Mukataka S, Tetsuo K, Joji T. Kinetics of enzymatic hydrolysis of lipids in biphasic organic-aqueous systems. J Ferment Technol 1985;63(5):461
/6. [8] NOVO Industrials. Analytical methods handout, Enzyme Process Division, NOVO industrials, Denmark; 1995. [9] Rooney D, Weatherley LR. The effect of reaction conditions upon lipase catalysed hydrolysis of high oleate sunflower oil in a stirred liquid
/liquid reactor. Process Biochem 2001;36:947
/53. [10] A.O.C.S. Official Method Ce 1
/62. Fatty acid composition by gas chromatography. Sampling and analysis of commercial fats and oils; 1990. pp. 1
/4. [11] Albasi C, Bertrand N, Riba JP. Enzymatic hydrolysis of sunflower oil in a standardised agitated tank reactor. Bioprocess Eng 1999;20:77
/81. [12] Tavlarides LL, Bepat PM. Models for scale-up of dispersed phase liquid
/liquid reactors. Proceedings of ISEC’83, AIChE Symp 1983;80(238):12
/46. [13] Calderbank PH. Physical rate processes in industrial fermentation, Part I: the interfacial area in gas
/liquid contacting with mechanical agitation. Tans Inst Chem Eng 1958;36:443. [14] Tsai SW, Wu GH, Chiang CL. Kinetics of enzymatic hydrolysis of olive oil in biphasic organic-aqueous systems. Biotech Bioeng 1991;38(7):762
/6. [15] Baily JE, Ollis DF. Biochemical engineering fundamentals, second ed.. New York: McGraw Hill Book Company, 1986. [16] Kim T, Chung K. Some characteristics of palm oil kernel olein hydrolysis by Rhizopus arrhizus lipase in reversed micelle of AOT in iso-octane and additive effects. Enzyme Microbial Technol 1989;11:528
/31. [17] Desnulle D. Pancreatic lipase. Adv Enzymology 1961;23:129
/61.