The Effectiveness of Immobilized Lipase Thermomyces lanuginosa in Catalyzing Interesterification of Palm Olein in Batch Reaction

Transcription

1 Journal of Oleo Science Copyright 2014 by Japan Oil Chemists Society doi : /jos.ess13176 The Effectiveness of Immobilized Lipase Thermomyces lanuginosa in Catalyzing Interesterification of Palm Olein in Batch Reaction Mei Huey Saw * and Wai Lin Siew Malaysian Palm Oil Board (No. 6, Persiaran Institusi, Bandar Baru Bangi, Kajang, Selangor Malaysia) Abstract: Lipase Thermomyces lanuginosa has shown potential in modifying oils and fats through interesterification. Analyzing the physicochemical properties of the modified oils is important to determine the effectiveness of lipase in catalyzing interesterification. In this study, the effectiveness of the immobilized lipase (Lipozyme TL IM) in catalyzing interesterification of palm olein in pilot-scale batch reactor was determined. The evaluation was done by analyzing the changes of triacylglycerol (TAGs) composition, sn-2 position fatty acids composition and the physical properties of the palm olein after the interesterifications. The pilot-scale batch reaction was conducted for 8 hours with 5 %w/w enzyme dosage based on the results of TAGs composition of the laboratory-scale interesterified products. The pilot-scale results showed that Lipozyme TL IM act as an effective enzyme in converting TAGs, in which 4.5% of trisaturated TAGs (PPP and PPS) were produced in the batch reaction. The formation of these new TAGs had also altered the thermal and physical properties of the palm olein. These interesterified products showed a broad peak and shoulder at high temperature, ranging from 10 to 40, indicating the formation of some new TAGs with high melting points. However, the enzyme did not perform perfectly as a 1,3-specific enzyme in the reaction as a significant reduction of oleic acid and an increment of palmitic acid at the sn-2 position was observed. Key words: enzymatic interesterification, lipase specificity, Thermomyces lanuginose, palm olein 1 INTRODUCTION Interesterification is one of the most important processes commonly used for oil modification that allows the modification of the physical properties crystallization and melting behaviors, as well as the chemical and nutritional properties of the oils and fats 1 3. Interesterification can be defined as the rearrangement of acyl groups between esters at specific or non-specific positions of the glycerol backbone to form new TAGs species, without altering its fatty acids composition 4, 5. Interesterification can be achieved by mean of chemical or enzyme catalysis. Chemical interesterification process involves a complete positional randomization of the acyl groups in the TAG non-specifically, whereas enzymatic process permits more specificity 6. The specificity behavior of lipases in catalyzing interesterification can be categorized into three main classes that are positional specificity regiospecificity, fatty acids selectivity, and stereospecificity 6, 7. The sn-1,3 positional specificity in lipase-catalysed interesterification is due to steric hindrance of the sn-2 position fatty acids in TAG that prevents the fatty acid at the sn-2 position from entering the lipase active site 6. However, lipases can only behave as a high specificity biocatalyst under specific reaction conditions. There are several factors that can affect the performance of enzyme such as the reaction system, reaction temperature, water content, enzyme dosage etc Over the years, interest in interesterification has increased significantly due to its ability of manufacturing food products free from transand unnatural cis-fatty acids 5. Palm olein of IV 62 is a product from double fractionation of palm oil and appears as liquid oil in tropical countries. Interesterification of palm olein may lead to the formation of some high melting TAGs, thus hardening the oil without altering its fatty acid composition. Therefore, interesterified palm olein is a potential hard stock for formulation of low saturation solid fats such as soft margarine, vanaspati, vegetable ghee etc. Besides, it can also be used * Correspondence to: Mei Huey Saw, Malaysian Palm Oil Board (No. 6, Persiaran Institusi, Bandar Baru Bangi, Kajang, Selangor Malaysia) meihuey@mpob.gov.my Accepted November 27, 2013 (recieved for review October 24, 2013) Journal of Oleo Science ISSN print / ISSN online

2 M. H. Saw and W. L. Siew to produce low saturation liquid products by removal of the saturated components via fractionation process 12. In this study, the specificity of an immobilized enzyme - Lipozyme TL IM Thermomyces lanuginosa in catalyzing the interesterification of palm olein in batch reaction was studied. Laboratory-scale reaction was conducted prior to the pilot-scale reaction in order to determine the optimum reaction conditions to be used. The specificity of the immobilized enzymes was determined using the Grignard degradation method. The TAG composition was determined after each reaction to study its effectiveness in catalyzing the reaction. 2 MATERIALS AND METHODS 2.1 Materials Refined, bleached and deodorized palm olein of IV 62 was purchased from Golden Jomalina Food Industries Sdn. Bhd Sime Darby Plantation. The 1,3-specific immobilized lipase Lipozyme TL IM was obtained from Novozymes A/ S. Chromatography grade solvents; acetone, acetonitrile, petroleum ether, and heptanes, and analytical grade diethyl ether were purchased from Fischer Scientific. The sn-2 analysis reagent, allyl magnesium bromide solution AMB was obtained from Sigma-Aldrich. The laboratory reagents used were silica gel 60G Merck, Germany, sodium chloride; saturated aqueous solution Prolabo France, sodium sulphate anhydrous Systerm, borontrifluoride; methanolic solution Merck, Germany, hydrochloric acid, 36 and acetic acid, 98 Systerm. Standard material used for the TAG analysis by high-performance liquid chromatography HPLC was a secondary standard R.B.D palm oil purchased from Golden Jomalina Sdn. Bhd., while gas chromatography standard used in determining the fatty acid composition at sn-2 position was FAME mix RM6 obtained from Supelco. 2.2 Methods Laboratory-scale Batch Interesterification Batch laboratory-scale interesterification was conducted to determine the optimum reaction time and enzyme dosage to be applied in the pilot-scale reaction. In this study, 2, 5 and 10 w/w of Lipozyme TL IM were used, and the reactions were conducted for 24 hours at 65. Pre-conditioning of the enzyme was carried out to get rid of the extra moisture before conducting the interesterification reaction. According to the Novozyme Interesterification Manual 2004, enzymes contain approximately 5 w/ w of water which needs to be removed 13. This is to prevent hydrolysis of the oil substrate and the formation of free fatty acid and partial glycerides such as diacylglycerols DAGs and monoacyglycerols MAGs. Enzyme conditioning process involves hydrolysis of the substrate with the water in the granules of the immobilized enzyme. 0.6 g of Lipozyme TL IM 2 was weighed into six conical flasks 100 ml. Twenty grams of palm olein IV 62 were added into each flask. The flasks were closed tightly with stopper and wrapped with a few layers of parafilm and alumimium foil to avoid presence of water in the reaction system. The flasks were then placed into a 65 water bath shaker filled with clean distilled water. The granules with oil were treated for 30 min at a shaking rate of 200 rpm. The oil was discarded via vacuum filtration after the conditioning process. Each conditioned enzymes were rinsed thoroughly with some IV 62 palm olein. The laboratory-scale lab-scale batch interesterification was conducted by using the same water bath shaker. The treated/conditioned enzymes were recovered into the 100 ml conical flask. Thirty grams of palm olein IV 62 was added into each conical flask. The conical flasks were closed tightly with the stopper and wrapped with a few layers of parafilm and aluminum foil. The six cornical flasks were fitted simultaneously into the 65 water bath and shaken at 200 rpm. Each conical flask was taken out and filtered immediately at 1, 2, 3, 4, 8 and 24 hours of reaction. The experiment was conducted in duplicate. The same procedures were carried out for the 5 w/w 1.50 g and 10 w/w 3.00 g of Lipozyme TL IM runs. The lab-scale interesterified products were analyzed for their TAG compositions Pilot-scale Batch Interesterification Pilot-scale batch interesterification reaction was conducted using the optimum reaction conditions obtained from the lab-scale interesterification. The 90 kg of premelted palm olein of IV 62 was loaded into a batch reactor followed by addition of 4.5 kg of pre-conditioned Lipozyme TL IM. The temperature of the reactor was adjusted to 65 2 using steam and cooling water. The reactor was blanketed with nitrogen gas to avoid the presence of moisture during the reaction. The mixture was stirred at a rate of 300 rpm. The reaction was conducted for 8 hours which was the optimum reaction time. The products were collected and analyzed for their TAG profiles, sn-2 fatty acids by Grignard degradation method, thermal properties and solid fat content SFC. 2.3 Analysis Procedures HPLC Separation of Triacylglycerols TAG compositions of the samples were determined by reversed-phase high performance liquid chromatography Gilson HPLC. The instrument was equipped with a refractive index detector from Waters 2410, USA, and Lichrospher 100 RP-18 column 250 mm with 5 μm particle size. About g of oil sample was weighed and diluted in 296

3 BATCH EIE OF PALM OLEIN acetone for injection into the HPLC column. The mobile phase used was a mixture of equal volumes of acetone and acetonitrile at a flow rate of 1 ml/min. The TAG peaks identification was based on a reference RBD palm oil and comparison with the literature 14, 15. The most abundant TAGs in RBD palm oil are POO and POP, in which POP refers to mixture of POP with the small amount of PPO included. Isomers of TAGs cannot be separated using this method Analysis of sn-2 Position Fatty Acids: Grignard Degradation reaction The regiospecific analysis was conducted by the Grignard degradation method 16, 17. Four drops of purified oil sample approximately mg was weighed into a clean 50 ml round bottom flask. Diethyl ether 10 ml was added to dissolve the oil sample. The solution was stirred using a magnetic stirrer. 0.3 ml of allyl magnesium bromide AMB solution was withdrawn with a 1 ml glass syringe and added to the solution. The syringe must be flushed with nitrogen gas before use to eliminate the presence of air. The diethyl ether solution became opaque, indicating a spontaneous reaction. After exactly one minute, 8 ml of acidic buffer 1 ml of 37 hydrochloric acid in 36 ml of 0.4 M boric acid was added to neutralize the magnesium hydroxide formed in the reaction mixture upon addition of water. The mixture was then transferred into a test tube. The aqueous phase was removed using a dropper, while the organic phase diethyl ether phase was washed twice with 8 ml of 0.4 M boric acid solution to remove the excess hydrochloric acid. The ether phase was then dried with some sodium sulphate anhydrous before transference to a small vial. The ether phase was dried by blowing with nitrogen gas. The acyglycerols mixture was re-dissolved in about 300 μl of diethyl ether Preparation of thin layer chromatography Thin layer chromatography TLC was used for separation of the acyglycerols. For preparation of TLC, five and a half clean glass plates with the same thickness were placed in the TLC spreader with the half plate on the left end. The plates were cleaned by wiping with acetone a few times. Forty five grams of silica gel G Fluka or Fischer was weighed into a 500 ml conical flask. Distilled water 90 ml was added into the flask and the flask was shook vigorously for exactly one minute. The silica gel mixture was spread smoothly from left to right onto the plates with a layer thickness of 0.5 mm. The TLC plates were taken apart from each another immediately after spreading. After leaving at room temperature for two hours, the plates were placed into the TLC plate holder and dried overnight at room temperature. The TLC plates were activated by heating at 110 for an hour. After cooling to room temperature, the plates were sprayed with 0.4 M boric acid solution until fully saturated. The plates were then dried overnight in a fume hood and stored in desiccators before use TLC separation of acylglycerols The sample was applied onto the prepared boric acid impregnated TLC plate by using 0.1 ml glass syringe. The spot must be as small as possible to prevent tailing effect of the TLC bands. The TLC plate was developed in a saturated chamber with a developing solvent chloroform/ acetone; 90:10. The saturation of the TLC chamber was indicated by the use of chromatography paper. The plate was taken out after 50 min and dried in the fume hood for 5 min. The plate was then placed inside the saturated chamber again for another 50 min. The plate was then dried in the fume hood, and sprayed with 2,7-dichlorofluorescein. Visualization of the acylglycerols bands was carried out by using a UV-lamp. The 2-monoacylglycerols 2-MAG band second band from the baseline was scrapped into a vial, and 2-MAG was extracted into a 25 ml round bottom flask by adding 3 ml of diethyl ether three times. The filter paper was flushed with diethyl ether for a few times to ensure complete extraction Preparation of Fatty acid methyl ester FAME : Boron trifluoride methods The solvent of the extract was evaporated by rotary evaporation. The methylation of the product 2-MAG was conducted using boron trifluoride according to MPOB Test Methods P3.4-Part FAME Analysis by Gas-chromatography GC The fatty acid methyl esters was filtered into a 1.5 ml vial by syringe filter pore size 0.45 μm; diameter 17mm. The concentrated test portion was then transferred into a GC-vial for GC analysis. 1 μl of the sample was injected into the SGE BPX70 column with a dimension of 60 m 0.25 μm 0.25 μm. The instrument was fitted with a Flame ionization detector FID. The carrier gases used were hydrogen with a ratio of 1: 100 with air and a flow rate of 0.8 ml/min. Detector and injector temperature were both set at 240. The separation of FAME was performed under an isothermal condition with temperature of 185. Standard material used for the fatty acid identification was FAME mix RM6 purchased from Supelco. 2.4 Determination of Solid Fat Content SFC Determination of SFC was carried out by using a Bruker Minispec pulsed nuclear magnetic resonance PNMR analyser MQ20, using the indirect MPOB Test Method Thermal Analysis by Differential Scanning Calorimeter DSC The thermal properties of the samples were measured using a Perkin Elmer Diamond DSC. The instrument was calibrated using indium at a rate of 5 /min. Volatile sample pans made of aluminium were used. The sample weights used were from 5-10 mg with an empty sample pan as the reference. The sample was heated from 55 to

4 M. H. Saw and W. L. Siew and held at this temperature for 10 min to destroy the entire crystal structures. The sample was then cooled to 55 at a cooling rate of 5 /min for recording of the cooling thermogram. The sample was held at this temperature for another 10 min. The melting properties of the samples were then investigated by heating the sample from 55 to 80 at a heating rate of 5 /min. 3 RESULTS AND DISCUSSION 3.1 Laboratory Trials: Optimization of batch interesterification Enzymatic interesterification involves the rearrangement of the acyl group between the palm olein TAG molecules to form new TAG species, until equilibrium is achieved. In this study, batch laboratory-scale interesterification was conducted with three dosages of Lipozyme TL IM: 2 w/w, 5 w/w and 10 w/w to determine the optimum enzyme dosage to apply in the pilot-scale reaction. TAGs composition of the oil during each interesterification reaction was determined to obtain the equilibrium stage for the reaction. As reaction time progresses, the TAG composition changed, in which trisaturated TAGs PPP and PPS and triunsaturated TAGs OLL, OLO, and OOO were increased, whereas diunsaturated TAGs PLL, PLO, POO and SOO and disaturated TAGs PLP were decreased. Figure 1 a and b shows the two major TAG species PPP and OOO that increased during enzymatic interesterification with the use of different dosages of Lipozyme TL IM. It was found that the rate of increment of both PPP and OOO for 2 w/w reaction was much lower than that of the 5 w/w and 10 w/w reactions. The rate of increment for 5 w/w and 10 w/w reactions were similar and both reactions achieved equilibrium PPP and OOO percentages at 8 hours of reaction. Similar trend was observed in Fig. 1 c and d as the reactions with 5 w/w and 10 w/w Lipozyme TL IM showed similar rate of reduction of POO and PLP. The rate of reduction for 2 w/w was significantly lower compared to both the 5 w/w and 10 w/w reactions. For industrial processes, cost of production, yield Fig. 1 The major changes of TAG species (a) PPP and (b) OOO (c) POO and (d) PLP during laboratory-scale enzymatic interesterification ( 2 %w/w, 5%w/w, 10 %w/w Lipozyme TL IM). 298

5 BATCH EIE OF PALM OLEIN and quality of the products are the important parameters for profit optimization 20. Therefore 5 w/w was chosen in this lab-scale batch interesterification study, as it promised adequately high conversion rate which was comparable to the 10 w/w enzyme dosage. 3.2 Batch Pilot-scale Enzymatic Interesterification Triacylglycerols Composition TAG compositions of the pilot-scale interesterified product are shown in Table 1. The pilot-scale interesterification increased the percentage of FFA, MAG, DAG and some TAG species. The increase of the FFA and MAG content is due to the hydrolysis activity in the enzyme system that also leads to slightly decrease of the TAG yield 21. Overall, the total triunsaturated and trisaturated TAGs were increased, while total diunsaturated and disaturated TAGs were reduced at the end of the enzymatic interesterification reactions. Initially, palm olein contained 8.9 of triunsaturated TAGs with 5.7 of OOO content. After the reaction, the amount of OOO was increased from 5.7 to 8.8 for the reaction using Lipozyme TL IM in the batch reaction. Palm olein of IV 62 only contained trace amount of trisaturated TAGs. After the enzymatic interesterification, the batch reaction produced a greater amount of total trisaturated TAGs, which was 4.5, with 3.5 of PPP and 1.0 of PPS. Total diunsaturated TAGs of palm olein IV 62 was reduced to 15.5 in the pilot-scale interesterification of the batch reactions. The most significant reduction in diunsaturated TAGs was POO, where 11.4 reduction of POO was observed in the batch reaction. For disaturated TAGs, only PLP was significantly reduced, the rest were maintained at about the same level as the feed olein Positional Analysis Lipozyme TL IM Thermomyces lanuginose is always known as a 1,3-specific lipase in which the enzyme may retain the sn-2 position fatty acids in the glycerol backbone after the interesterification 9. Positional analysis was conducted to evaluate the specificity of the enzyme in catalyzing interesterification in batch reactor. Figure 2 showed the sn-2 position fatty acids of the feed oil and the pilotscale interesterified product. Theoretically, the sn-2 posi- Table 1 Chemical properties: acylglycerols compositions (%) of the feed olein (POo IV62) and the pilotscale enzymatic interesterified products Triacylglycerols POo IV 62 Batch TL IM IV FFA & MAG DAG Total triunsaturated OLL OLO OOO Total diunsaturated PLL PLO POO SOO Total disaturated MLP PLP POP POS SOS Total trisaturated PPP PPS Total TAG

6 M. H. Saw and W. L. Siew Fig. 2 sn-2 fatty acid composition of palm olein IV 62 and the pilot-scale interesterified palm olein (Batch EIE). Fig. 3 Solid fat content of palm olein IV 62 and the pilot-scale interesterified palm olein (Batch EIE). tion fatty acids should remain unchanged after interesterification. However, some positional randomization of fatty acid residues in TAGs occurred during the enzymatic interesterification. The most obvious changes were the percentage of oleic acid C18:1 and palmitic acid C16:0 at the sn-2 position; percentage of C18:1 was decreased while percentage of C16:0 was increased significantly in the pilotscale interesterification. The change of fatty acids in the sn-2 position is due to acyl migration. At first, 1,3-specific lipases produced mixture of TAGs, 1,2-DAGs, and 2,3-DAGs without interfering the sn-2 position fatty acids. However after prolonged reaction, with the formation of 1,3-DAG, acyl migration occurred that permit some randomization of the sn-2 position fatty acids in the TAG backbone 6. The factors of acyl migration were reaction time, reaction temperature, presence of water, reactor type and reaction system 21, 22. In other words, the specificity performance of enzyme is very dependent on the reaction system and conditions, as they can only perform their specificity under particular conditions 10, 11. Although the enzymatic interesterifications did not perfectly retain the fatty acids at the sn-2 position, the positional specificities were much better compared to the products from chemical interesterification where full randomization of the fatty acids occur 23. In environmental and customer perceptions, enzymatic interesterification is a greener and safer technology because it does not involve the use of any chemicals and can be carried out under mild reaction conditions 20. The use of mild reaction conditions pressure, temperature is advantageous due to the lower cost of production and energy consumption and may also lead to lower degradation of minor components such as carotene compounds, tocotrienol etc Physical Properties of Interesterified Products Solid Fat Content The SFC of the palm olein of IV 62 and the pilot-scale interesterified product was measured as a function of temperature as shown in Fig. 3. From the results, it can be seen that the enzymatic interesterification had much higher SFC than the feed palm olein. At 0, the batch interesterified product contained 47.1 of SFC compared to 26.0 in the palm olein. The interesterified product appeared as a semi-solid fat at ambient temperature 25 but the feed olein was fully melted at 15. The changes of SFC depend on the TAG composition of the oil. The SFC of the feed olein was the lowest at all tested temperatures because it does not contain high melting TAGs such as PPP and PPS. The presence of trisaturated TAGs, which has high melting points, promotes crystallization, thus increasing the SFC 23. The solids present in the feed olein at temperatures lower than 15 were mainly due to disaturated TAGs which consists mainly of POP Thermal Properties by Differential Scanning Calorimeter DSC DSC curves reveal the transition temperatures and heat of fusion or crystallization 24. Figure 4 showed the thermal properties of the pilot-scale interesterified products as compared to the feed palm olein. In general, the melting and crystallization properties of the oil are dependent on the chemical structure and polymorphic behavior of TAGs mixture 23. From the melting thermograms Fig. 4b, palm olein showed a sharp endothermic peak at about 3, while the batch interesterified product displayed broad peaks and shoulders at high temperatures ranging from 20 to 40. The thermogram of the batch reaction showed that three major components were present in the oil. These could be categorized by the melting points ranging from 20 to 0, 0 to 10, and 10 to 40 ; the left and the right shoulders indicated the formation of triunsaturated TAGs OLL, OLO and OOO and trisaturated TAGs PPP and 300

7 BATCH EIE OF PALM OLEIN feed olein were crystallized mostly at 0. The peak value of crystallization thermogram of the interesterified product from the batch reaction was about 20. This sharp exothermic peak was not present in the thermogram of the feed palm olein. This was due to the presence of significant amounts of high melting TAGs PPP and PPS in the interesterified product, which was absent in the feed olein Table 1. This study shows that enzymatic interesterification of palm olein has altered the physical properties of the oil from liquid to semi-solid without changing the FAC and IV. This interesterified product could be useful for production of low saturation solid products such as margarine, vanaspati etc. Besides, this product can also be used as an intermediate product for production of low saturation palm olein by fractionation process. Fractionation of the interesterified oil is much easier compared to fractionation of palm olein. The presence of high melting TAGs in the interesterified olein can facilitate crystallization of the oil and the high melting TAGs can be easily removed as stearin during fractionation process. Fig. 4 (a) Crystallization thermogram and (b) Melting thermogram of the pilot scale interesterified palm olein (Batch EIE). PPS, respectively, in the interesterification reaction. The intensity of the middle peak from 0 to 10 was reduced compared to the feed olein. This is mainly due to the reduction of the percentage of the diunsaturated and disaturated TAGs such as PLO, POO, PLP, POP etc. Table 1. In addition, the thermogram also indicated that the interesterified palm olein was fully melted at much higher temperature 40 compared to the feed olein 16. This is also due to the presence of the high melting trisaturated TAGs in the interesterified olein which are not present in the feed olein. From the crystallization thermograms Fig. 4a, it was observed that the reaction products started to crystallize at higher temperature as compared to the feed olein. Only a sharp peak was observed in the crystallization thermogram of palm olein at 0, meaning that the TAGs in the 4 CONCLUSIONS Lipozyme TL IM is very effective in term of TAGs conversion, in which 4.5 total trisaturated TAGs were formed in addition to other transformation. The formation of the high melting TAGs altered the thermal and physical properties of the interesterified oil, in which SFC was increased after the reaction. However, the enzyme did not perform perfectly as 1,3-specific enzyme in the batch interesterification as the reaction had resulted in a reduction of oleic acid and an increment of palmitic acid at the sn-2 position. Hence, it can be concluded that Lipozyme TL IM is an effective interesterification enzyme with great TAGs conversion capability but less efficient in retaining the sn-2 fatty acids. ACKNOWLEDGEMENT: The authors would like to thank the Director General of MPOB for permission to publish this paper. References 1 Ucciani, E.; Debal, A. Chemical properties of fats. in Oils and Fats Manual Karleskind, A. ed. Vol. 1, Lavoisier Publishing, Paris, pp Lampert, D. Process and products of interesterification. in Introduction to Fats and Oils Technology 301

8 M. H. Saw and W. L. Siew Richard, D. O.; Walter, E. F.; Peter, J. W. ed. Second Edition, AOCS Press, Champaign, pp Pacheco C.; Palla C.; Crapiste G. H.; Carrín M. E. Optimization of reaction conditions in the enzymatic interesterification of soybean oil and fully hydrogenated soybean oil to produce plastic fats. J Am Oil Chem Soc Faur, L.; Transformation of fat for use in food products. in Oils and Fats Manual: A Comprehensive Treatise Karleskind, A. ed. Vol. 2, Lavoisier Publishing, Paris, pp Silva R. C.da; Martini Soares F. A. S. D.; Fernandes T. G.; Castells A. L. D.; da Silva K. C. G.; Goncalves M. I. A.; Ming C. C.; Goncalves L. A. G.; Gioielli L. A. Interesterification of lard and soybean oil blends catalyzed by immobilized lipase in a continuous packed bed reactor. J Am Oil Chem Soc Willis, W. M.; Marangoni, A. G. Enzymatic interesterification. in Food Lipids Chemistry, Nutrition, and Biotechnology Akoh, C. C.; Min, D. B. ed. Second Edition, Revised and Expanded, Marcel Dekker Inc., New York, pp Cheah, S. C.; Augustine, O. S. H. Application of Enzyme Reaction to Oils and Fats. in Proc. of 1987 Int. O.P/P.O Conf.-Technology. Palm Oil Research Institute of Malaysia, Kuala Lumpur, pp Quilan, P.; and Moore, S. Modification of triglycerides by lipases: process technology and its application to the production of nutritionally improved fats. INFORM 4, Yang, T.; Xu, X. Enzymatic modification of palm oils: Useful products with potential processes. In Proceeding of the 2001 PIPOC International Palm Oil Congress Chemistry and Technology, Kuala Lumpur, pp Natália, M. O. M.; Manuela, R. da F.; Suzana, F-D. Operational stability of Thermomyces lanuginose lipase during interesterification of fat in continuous packedbed reactors. Eur. J. Lipid. Sci. Technol. 108, Criado, M.; Hernández-Martín, E.; and Otero, C. Optimized interesterification of virgin olive oil with a fully hydrogenated fat in a batch reactor: Effect of mass transfer limitations. Eur. J. Lipid. Sci. Technol. 109, Saw, M. H.; Chuah, C. H.; Siew, W. L. Characterization of low saturation palm oil products after continuous enzymatic interesterification and dry fractionation. J Food Science. 74, E177-E Novozyme. Enzymatic interesterification. in: Novozyme Interesterification Manual Oils & Fats/ , Version , Available from: www. novozymes.com., Accessed Oct 5, Ghazali, H. M.; Hamidah, S.; and Che Man, Y. B. Enzymatic transesterification of palm olein with nonspecific and 1,3-specific lipases. J Am Oil Chem Soc. 72, Swe, P. Z.; Che Man, Y. B.; Ghazali, H. M. Composition of crystal of palm olein formed at room temperature. J. Am. Oil Chem Soc. 72, Claus, C. B.; Annmette, R.; and Gunhild, H. Regiospecific analysis of triacylglycerols using ally magnesium bromide. Lipids 28, Paul, A.; Édith T., Armand, B.; and Joseph, A. Regiospecific analysis of fractions of bovine milk fat triacylglycerols with the same partition number. Lipids 33, MPOB Test Methods P3.4-Part 1. Methods of test for palm oil and palm oil products: Preparation of methyl esters of fatty acids-part 1: Boron trifluoride method. in MPOB Test Methods, Malaysian Palm Oil Board, Kuala Lumpur, pp MPOB Test Methods P4.9-Section 1. Determination of solid fat content by pulsed nuclear magnetic resonance PNMR -Section 1: Direct Method. in MPOB Test Methods, Malaysian Palm Oil Board, Kuala Lumpur, pp Xu X. Engineering of enzymatic reactions and reactors for lipid modification and synthesis. Eur. J. Lipid Sci. Technol. 105, Svensson J. and Adlercreutz P. Effect of acyl migration in Lipozyme TL IM-catalyzed interesterification using a triacylglycerol model system. Eur. J. Lipid Sci. Technol. 113, Xu X.; Skands A.; Adler-Nissen J. Purification of specific structured lipids by distillation: Effects of acyl migration. J Am Oil Chem Soc. 78, Gunstone F. D. Procedures used for lipid modification. In Gunstone F. D. ed. Structured and Modified Lipids. New York: Marcel Dekker Zaliha, O.; Chong, C. L.; Cheow, C. S.; Norizzah, A. R.; Kellens, M. J. Crystallization properties of palm oil by dry fractionation. Food Chemistry 86,