This action might not be possible to undo. Are you sure you want to continue?
INSTITUTE OF TECHNOLOGY, NIRMA UNIVERSITY, AHMEDABAD-382 481, 08-10 DECEMBER 2011
A comparative study on predictions of vapor liquid equilibrium of biodiesel systems
K J Suthar, M H Joshipura
Abstract--In process simulation, reliable and accurate property estimation methods play an important role in the solution of various simulation problems where convergence is often traced to failures in the reliable predictions of physical and thermodynamic properties. Convergence and reliability of the simulation is heavily dependent on the fluid package selected. Predictions of phase equilibrium using various thermodynamic models provide an idea about which model will be able to represent the entire process in better way. No such systematic studies are done in the literature for biodiesel. Paper presents a review on various simulation studies done for biodiesel production to highlight the fact that the thermodynamic models were used with little knowledge or by intuitions. A comparative study is presented for the proper selection of fluid package by determining vapor liquid equilibrium data for various (fatty acid-alcohol, fatty acid-glycerol, alcohol-glycerol, fatty acid ester- alcohol etc) biodiesel systems. The comparison was done with the help of making a model of experimental VLE data and simulated VLE data (by various thermodynamic models). Key word-- Aspen HYSYS, Biodiesel Simulation, VLE
thermodynamic models available for the simulation of process using HYSYS and select the best model for the simulation of biodiesel process production.
Biodiesel (mono-alkyl esters of long chain fatty acids) is a promising alternative or extender to conventional petroleum based diesel fuel. Biodiesel has a number of advantages  – it is derived from a renewable domestic resource (vegetable oil), reduces carbon dioxide emissions by 78% when compared to diesel fuel on a life cycle basis, and is non-toxic and biodegradable, making it a more environmentally benign fuel. Biodiesel system includes fatty acid, alcohol, fatty acid ester and glycerol. But fatty acid is not directly represented as oil, it consists of different fatty acid chains, which vary in carbon chain length and in the number of unsaturated bonds (double bonds). To simplify the analysis of kinetics and thermodynamics, oil can be regarded as pure compound like triolein, trilinolein, tripalmitin, tristearin etc . Different oils and there composition used as biodiesel system have been reviewed in Table-1. For simulation of biodiesel production process, detailed information about oil and its composition is needed. Also thermodynamic models have to be studied with great care. 2) Binary VLE phase diagram
igh cost of biodiesel is mainly because of the feed stock. To face the biodiesel process simulation and the prediction of the normalized final properties using a process simulator, it is necessary to take into account several issues which include the quantity and quality of the constituents of the departing oils, the topology of the process, the detailed kinetic models and the thermodynamic packages. Also triolein, tripalmitin, trilinolein and tristearin etc. should be considered precisely as they are the major triglycerides in most commonly used oils for biodiesel production. The fatty acids alkyl esters resulting from its alcoholysis account for more than 90 wt.% of palm, soybean, rapeseed and sunflower oils. In process simulation, reliable and accurate property estimation methods play an important role in the solution of various simulation problems where convergence is often traced to failures in the reliable predictions of physical and thermodynamic properties. We can partly describe oils by the original source, like peanut oil, canola oil, soy oil, beef tallow, and on and on. There are drying oils and non drying oils. Waste oils need more information, like FFA content, and water content. "Organic oils are three long strings of fatty acids attached to a glycerin molecule. The Fatty Acids can vary in length and in how they are bonded or put together. The way we describe Fatty Acids is by the number of carbon atoms in it and the number of double bonds. We typically see between 12 and 26 carbon molecules in our oils and oils with zero to three double bonds. The purpose of this study is to compare different
The knowledge and the capability to describe the phase equilibria of binary systems formed by a fatty acid, alcohol, FEMA and glycerol are of great interest in biodiesel production. The knowledge about the vapor-liquid equilibrium (VLE) conditions of these systems is also essential for a correct sizing of the operation units of biodiesel production. The miscibility of triglycerides and methanol are rather poor due to their dissimilarity in size and polarity, and they form two liquid phases upon their initial introduction into reactors. One factor of particular importance in the alcoholysis process is the degree of mixing between the alcohol and triglyceride phases . In other words, the phase behavior of the reaction mixture is crucial for the reaction process. The appropriate selection of thermodynamic models has a strong influence on calculations. The appropriate selection of thermodynamic models has a strong influence on VLE and LLE calculations. This reason makes this comparision of model using VLE, especially important.
Oleic. An example of methyl oleate-methanol system is demonstrated further.5 ASPEN Engineering SuiteTM 11. Linoleic. Methyl or ethyl oleate represent biodiesel Triolein represents oil and ethyl oleate represent biodiesel Simulator used ASPEN Plus® Aspen Plus 2006. Triolein.2 HYSYS plant NetVers 2. linoleic acid 32.1  3) Thermodynamic Properties Packages Various thermodynamic property packages used for the simulation of biodiesel production process along with the simulator is reviewed in Table-1 which showed most of the process have used NRTL or UNIQUAC. Biodiesel systems includes both oil and alcohol and it ultimately the extent to which both react forming the product biodiesel. II.alcohol etc.3 ASPEN Plus®    Rapeseed oil Virgin vegetable oil and waste oil Palm oil ASPEN Plus Aspen plus 2006 NRTL and UNIFAC -LLE SRK and Electrolyte NRTL NRTL and UNIFAC   Aspen Plus v.0%.7%. UNIQUAC and UNIFAC-LL.Trilinolein. alcohol-glycerol. Hence. COMPARING FLUID PACKAGES BY DETERMINING VLE DATA OF VARIOUS BIODIESEL SYSTEMS Various fluid packages were compared on the basis of available VLE data from various literatures [19-24]. fatty acid-glycerol. temperature (˚C) was plotted to obtain base equation from trend line using MS excel® for the particular system. methyl oleate and methyl linoleate are biodiesel. Such plots were formed for every different fluid package and every different system.2 INTERNATIONAL CONFERENCE ON CURRENT TREND IN TRCHNOLOGY.8%. palmitic acid 14. were plotted using HYSYS.1. estimating thermodynamic model suitable for biodiesel system is essentially a primary work for the simulation of biodiesel production process. methyl oleate (C19H36O2) is used to represent biodiesel Oleic acid is used as the key component representing triglycerides in this study (triolein) and methyl oleate (C19H36O2) represents the final product (FAME or biodiesel) Triolein represents oil and methyl oleate as biodiesel Triolein represents oil and methyl oleate represents the resulting biodiesel oleic acid 44. NRTL for polar compound RK Aspen EOS. A base equation was used to calculate temperature/pressure at common composition points for different fluid packages for a particular system.0 software Thermodynamic model used NRTL and UNIQUAC A UNIFAC-DMD (Dortmund) Wilson Equation Reference    NRTL and UNIFAC NRTL and UNIQUAC for polar compound and UNIFAC UNIFAC LLE model. Consequently. electrolyte NRTL NRTL NRTL and UNIQUAC NRTL and UNIQUAC   Waster vegetable oils Renewable lipids feedstock Aspen Hysys ASPEN Plus® V2006   Vegetable oil Conola Oil Jatropha Oil HYSYS and plant NetVer 3. „NUiCONE 2011‟ TABLE-1 LITERATURE SURVEY: OIL. steric acid 7. base equation was generated for all the fluid packages for further calculations. . fatty acid ester. M-palmitate and Mstearate represent biodiesel 94 wt% Triolein and 6 wt% oleic acid (C18H34O2).2%. Triolein is used to represent and methyl oleate represents biodiesel Triolein represents the virgin vegetable oils and triolein and oleic acid in the case of waste cooking oils. VaporLiquid equilibrium data of various system including fatty acid-alcohol. The presence of polar compounds such as methanol and glycerol makes necessary the use of the non-random two liquids (NRTL) model or Universal Quasi Chemical (UNIQUAC) model  On other hand.Tripalmitin. fatty acid like triolein are highly non polar organic phase.1 HYSYS (Hyprotec system) Aspen HYSYS V7. Methyl Oleate represent biodiesel Oleic acid represent as the overall fatty acid and biodiesel as methyl oleate Triolein represents Oil and Methyl Oleate represent fatty acid ester Triricinolein represents castor oil and ethyl ricinoleate to represent biodiesel. Data tables were generated from these plots and from the generated data.3% represent oil and methyl palmitate. and others 1.Tristearin. 11. a plot of liquid composition (x) vs. SIMULATOR AND THERMODYNAMIC MODEL USED FOR BIODIESEL PRODUCTON PROCESS Oil Used Vegetable Oil: rapeseed and palm oil Rapeseed Oil High quality Rapeseed Oil Castor Oil Vegetable oil (sunflower Oil) Oil composition Triolein and Oleic acid represent Oil. Palmitic and Stearic acid represent oil and methyloleate. Similarly. M-linoleate.
5814 2 0.40 80.4964 4 0.054 0.22 84.2793 10 0.3KPa using various fluid packages Once base equation is generated.055 0.055 0.72 80.85 80.048 0.3KPa for various FP 2) Computing the base equation VLE date collected from various literatures have temperature/pressure range at varying composition points from that of the data available from HYSYS.38 Where N is number of points 88.052 0.3791 8 0.38 0. a base equation is generated using the trend line as shown in figure2. Table-2 shows %AAD for methyl oleate-Ethanol at 101.10 85. VLE plot for Methyl oleate-Ethanol system at 1atm Fig 3 Composition(x) vs.37 80. Similarly. Hence to obtain the data at common point for their comparision.13 83.31 81. Various VLE for the biodiesel systems were plotted as shown in an example of methyl-oleate and ethanol at 101.36 R² = 0. The y term represents the vapor composition. T(˚C) 1 0.36 86.3184 9 0.7x3-323.43 80. But an equilibrium plot unit operation extension should be registered. (Series1) 1 Temperature (C) 200 150 100 50 0 0 0.3 INSTITUTE OF TECHNOLOGY.02 85.54 0.35 80.055 0. Similarly. obtaining it for various fluid packages and tabulating it.70 83.51 80.5 X Fig 2 Composition(x) vs. temperature/pressure at common composition can be obtained.4653 5 0.04 0. These types of diagrams are generated at a constant pressure.30 79.4279 6 0. usually expressed in terms of mole fraction.57 81.55 83.09 83.76 85.74 81.7x3 .65 81. 08-10 DECEMBER 2011 1) VLE Plot using HYSYS Two types of vapor–liquid equilibrium diagrams are widely used to represent data for two-component (binary) systems in the study.30 81. The x term represents the liquid composition.32 N TABLE.4025 7 0. Composition(x) vs.32KPa pressure in figure-1. AHMEDABAD-382 481.46 80.63 85.78 78.12 84.3KPa USING NRTL X X exp Deviation T ˚C Base case value.46 0. Temperature (˚C) plot for base equation for methyl oleate-ethanol system at 101. T(˚C) NRTL.6x2 100 + 126.53 83.4x + 64.34 86.016 0.323.2125 %AAD for NRTL is 4.28 80.036 0.015 0. Temperature (˚C) is potted for various fluid packages as shown in figure-3.5305 3 0. The values obtained from the literature are considered as the basis for the comparative study with the one obtained from HYSYS. The diagram is easily generated using Aspen HYSYS by going to flowsheet-add operationextension-equilibrium plots.04 85. And the other type is „pressure versus x and y‟ which are generated at a constant temperature .2369 11 0. NIRMA UNIVERSITY.77 84.5 X 1 NRTL UNIQUAC Wilson Chein Null PSRV PR Fig 1.82 85.5 0.42 0.4 0. Temperature (˚C) plot for methyl oleateethanol at 101.054 0.58 0.09 84.32 0.24 0.2 CALCULATION OF DEVIATION FOR METHYL OLEATE-ETHANOL SYSTEM AT 101.31KPa Series1 Poly. The first is a „temperature versus x and y‟ diagram (Txy).22 0.4x+64.991 50 0 0 0.051 0. temperature (C) 150 y = 269.2 y= 269.10 77.28 0.0438 .6x2+126. 3) Calculating %AAD for Methyl oleate-Ethanol at 101.
85* 2.2 * % Deviation SRK 3.12 Where. From all isobaric systems considered.70* 27.97* 18.48* 13.4642 41.03* 18.84 * Chein Null 9.44* 11. N is number of points and *Binary interaction parameters are missing and Systems for which one component is hypothetically created .59 52.32KPa 101.52 10. PRSV and Chein Null had 17.18* 4.32KPa 101.43* 11.32KPa 92KPa 101.78* 18.25* 7.78 * 109.57* 84.14 79.12% and 13.5* 9.42* 4.24 4.5001 6.51 95.89 55.88 * NRTL 3.32KPa 101.66 24.32 25.67* 37.35 168.92 301.28* 27.005%.18* 17.32KPa N PR 15 16 15 15 11 9 9 8 9 9 3. isothermal and isobaric were considered for the comparative study on vapor liquid equilibrium prediction.51 37.44* 25.058* 15. deviation have increased for Wilson activity model.63 a Wilson 15.66 35.801 56 UNIQUAC 17.1 25.7* 37.88 5.32KPa 101.99 PRSV 3.91* 12* 10.93 15.43* 53.63 UNIQUAC 11.275* 9.91* 7.2* 14.54 3.934* 7.11 19. Similarly. FEMA were not recommended for PR and SRK equation of states as there use would invalidate the simulation result.34* 22.59 55.94 18.04* 13.95 44.508 20. Wilson model gave minimum percentage absolute average deviation of 12.07* 24.37* 0. The absolute average deviation for PR and SRK is 13.451* 15.005 9.89 4.15 41.39 17.11* 40.17* 33.44 53.22* 15.32 51.36 3.99 48.9* 9.17 15.54% of absolute average deviation respectively.36 42.48 3. N is number of points and *Binary interaction parameters are missing.24* 099* 19.2* 21.3329KPa 101. Also compounds like Glycol and ethanol were not recommended for PR and SRK.32KPa 101.38* 19.4 INTERNATIONAL CONFERENCE ON CURRENT TREND IN TRCHNOLOGY.4* 5.96 26.08* 10.16* 17.1 291.72* 9.24 13.96 * % Deviation SRK 10. „NUiCONE 2011‟ III. aSystems for which one component is hypothetically created and bSystem for wich both components are hypothetically created TABLE-4 % DEVIATION FOR THE ISOTHERMAL SYSTEM System Methanol-Glycol Triolein-Methanola Ethanol-Ethyl Lauratea Ethanol-Ethyl Lauratea Methanol-Methyl Myristate Methanol-Methyl Myristate Methanol-Methyl Lauraate Methanol-Methyl Lauraate Methanol-Glycol Ethanol-Glycol %AAD Isothermal 493K 493K 523K 493K 523k 493K 523K 493K 493K 493K N PR 9 10 8 8 6 6 9 9 9 8 13.38 125.5 56.35 * 26.65 6. The results for the isothermal systems are tabulated in Table-4.94% and11. methyl oleate-ethanol at 101.93% respectively but most of their values were found for the missing binary coefficient.78 * 17.43 14.318 26.75 82.32KPa Wilson activity model showed minimum deviation of 3.65 25.967 3.07 1. various biodiesel systems were compared using different property packages.01* 53.8321 43.98* 20.95* 55.45* 4.77* 2.57 PRSV 18.494* 1.98 26.67 4. RESULTS AND DISCUSSION Two types of systems i.3* 53.99.65 26.81* 2.84 52.06* 1.9 15.33* 3.71 26.88 14.e.04 * NRTL 9. For the example considered.13 27.55* 15.99 26.6* 6.16 Wilson 1.27 67.33* 4.3 1.42* 9.59* 6.85* 1.07* 11.71* 3.166 37.69 3.32KPa 5.68* 18.51 10. Similarly. TABLE-3 % DEVIATION FOR THE ISOBARIC SYSTEM System Methyl Oleate-Methanol Methyl Oleate-Ethanol Ethnol-Glycol Methyl Laurate-Methanol Ethanol-Ethyl Stearatea Methyl Myristate-Methanol Methyl Laurate-Ethanol Methyl Myristate-Ethanol Ethyl palmitate-ethyl stearateb Glycol-Water %AAD Isobaric 101.334 26.98* 20.32 3.69* 19.885 1.9507 * Chein Null 2.46 212.45 35. Results for the isobaric systems are tabulated in Table-3.16* 13.2 Where.0363 7.67 * 19.404 1. Considering isothermal systems.98 * 128.26 12.
pub. All the three systems have component which are missing in HYSYS component library. “Design and control of biodiesel production processes with phase split and recycle in the reactor system”. PRSV and Wilson.17 11. pp 189–192. “Biodiesel as an alternative motor fuel: Production and policies in the European Union”.166 27. Though missing binary interaction parameter have to be calculated for accurately recommending best suitable property package for biodiesel system and it might have an impact on the present result.92 1..68 V. FernAndez.318 26. A.885 15. Jesus Arauzo.03 7. The results for the deviation of the systems for which components were present in HYSYS library and binary interaction parameter were not missing are tabulated in Table-6. of Cleaner production. Costa “Simulation and life cycle assessment of process design alternatives for biodiesel production from waste vegetable oils”.967 3.bsystems for which compounds were hypothetically created and high deviation is noticed for all the systems. pp 136-141 M.71 136. Property packages PR SRK NRTL UNIQUAC Wilson PRSV % AAD 46. Springer sci.. hence binary interaction parameters for PRSV and Chein Null are of main interest for comparative study.e. Suppes.69 3. Mata. CONCLUSION Vapor liquid equilibrium prediction for the biodiesel systems were investigated using different property packages like Peng-Robinson. of the Taiwan Ins. TABLE-5 % DEVIATION FOR THE SYSTEMS FOR WHICH COMPONENTS WERE HYPOTHETICALLY CREATED But PR and SRK property packages are not recommended for glycol. of Chem. Dasari.78 Wilson 1..51 1. volume 18. 101. Jose Angel Pena “Prediction of normalized biodiesel properties by simulation of multiple feedstock blends”. It concludes the use of Wilson activity model over NRTL and UNIQUAC giving minimum deviation of 8. triolein-methanol. Also critical properties have to be carefully studied in the case of hypothetically added compounds for biodiesel production process.50 19. Nirma University for their invaluable guidance and assistance throughout the work.23 Where.V. Manuel Garci. alcoholysis kinetics of soybean oil. G. 2008. ethanol-ethyl laurate and ethanol-ethyl laurate. Jose Luis Sanchez. pp 4431–4439 Monitors of organic chemicals in the environment: semi permeable membrane devices By James N. Jr. Huckins.1 1.80 26.A. SRK.J. TABLE-6 % DEVIATION FOR THE SYSTEMS FOR WHICH THE COMPONENTS ARE PRESENT IN HYSYS LIBRARY AND BINARY INTERACTION PARAMETER ARE NOT MISSING Systems Methyl Oleate-Methanol Methyl Oleate-Ethanol Ethnol-Glycol Methanol-Glycol Methyl Laurate-Methanol Methyl Myristate-Methanol Methyl Laurate-Ethanol Methyl Myristate-Ethanol Glycol-Water %AAD NRTL 3.89 26.52 15.07 1. Ward. September. Jeffrey D. Tech. Bio. issue 13. Kahraman Bozbas. Giberto A.7 91. Rodríguez.66 17.  REFERENCES       Ramos. activity models have strong predictability. Engg.508 20. pp 12511259 . edition 1.Tech. But from the present study.5 INSTITUTE OF TECHNOLOGY. VI. Petty. UNIQUAC. Pinto and A. 2006. Jr. 2010.M. pp 53 Sargio Morais. M. The accurate results could not be drawn as binary interaction parameter for most of the property packages are missing. Teresa M.6%.71 48.40 9.48 3. 2010.33 26.35 UNIQUAC 17.100. C.3 11. Though accurate results would only be revealed once missing binary interaction parameter are calculated. 2009.32 25.of the American Oil Che. Jian Kai Cheng.36 3.10 82. L. Bio. Kees Booij. 2003. Casas.005 9. PR and SRK property packages showed comparatively 1314% deviation but it can only be selected once the missing interaction parameters are obtained and then getting minimum deviation and if it is recommended for the biodiesystem. Jr. NIRMA UNIVERSITY. ethanol and FEMA.J. Cheng Ching Yu. Among those property packages. 08-10 DECEMBER 2011 The deviation increased because of three main systems i. Renewable and Sustainable Energy Reviews 12. Chein Null. Jimmie D. Goff. pp 261–268. M. they can be used for predicting vapor liquid equilibrium producing an accurate result for the selection of suitable property package. AKNOWLEDGEMENT I would like to acknowledge all the faculty members of Chemical department.55 43.. Antonio A. AHMEDABAD-382 481. Table-5 shows absolute average deviation for the a.67 4. NRTL.26 8. pp 542–552 Yin Heng Shen. It showed the high deviation in the result due to critical properties and hence they have to be studied with greater care..J. Pérez. IV. “Influence of fatty acid composition of raw materials on biodiesel properties”. N is number of points and *Binary interaction parameters are missing Most recommended property packages NRTL and UNQUAC gives the deviation of 82% and 91% respectively. 2011. Martin.59 26.. A.11 Chein Null 39. Wilson activity model have better accuracy for the biodiesel system. Once those missing parameter values are calculated. Noncatalytic. Alberto Gonzalo. Soc. 80.
Cat. Fluid Phase Equilib. West. & des. issue 2. Tech. G. volume 100. 26. Marin “Simulation of heterogeneously MgO-catalyzed transesterification for fine-chemical and biodiesel industrial production”. R. Cardona. Oliveira. Process design and technology assessment”. Dube. Chem. Bio. Takeshi Hirayama. Santana.F. the jr. B: Env. “Process integration possibilities for biodiesel production from palm oil using ethanol obtained from lignocellulosic residues of oil palm industry”. of Taiwan Ins. 49. Knud V. “Phase Equilibria of Ester + Alcohol Systems and Their Description with the Cubic-Plus-Association Equation of State”. pp 217–222 Yusuke Shimoyama. & des. Fuel pro. issue 1-2. “Assessment of four biodiesel production processes using HYSYS Plant”.B. „NUiCONE 2011‟ Lene Fjerbaek Sotoft. Coutinho. Pinto and F. Kant Prommeta and Thongchai Srinophakun. I. Tech. pp 1087-1094 Luis F. Berger and Guy B. pp 1-16 Uraiwan Kaewcharoensombat. volume 42. John Wiley & Sons. By William L.S. Jr. A. Chem. Zhang. M. Eng. 2009. Dusko Posarac and Naoko Ellis “Process simulation and economic analysis of biodiesel production using fresh and waste vegetable oil and supercritical methanol”. pp 22–29 Yusuke Shimoyama. Yoshio Iwai. volume 89.. Eng. Issue 1. Oscar J. Kates “Biodiesel production from waste cooking oil: 1. C. Fluid Phase Equilib.K.“Life cycle assessment of biodiesel production from Jatropha” . 2011. 2010. “Life cycle analysis of biodiesel production”. Dossin.C. Tao Jiang. “Techno-economic analysis of a biodiesel production process from vegetable oil”. Bio. 2008. Yasuhiko Arai. Tech. Issue 7-8. volume 67. 2011. 2010. Miguel. 257. Varanda. 2009. pp 8–11  “Fundamentals of vapor liquid phase equilibrium Distillation Design and Control Using AspenTM Simulation”. Maciel Filho. Queimada.A.R. tech. Antonio J. Tech. pp 64–69 Yusuke Shimoyama. Ind...2007.C. Wolf Maciel “Simulation and cost estimate for biodiesel production using castor oil”.A. Ben-Guang-Rong. fluids. Bo Seok Jin. D. Fluid Phase Equilib.R. Eng. Fluid Phase Equilib. Issue 1. 239. Zexue Du. 2010. pp 1227-1237 Mariana B. “Phase equilibria of glycerol containing systems and their description with the Cubic-Plus-Association (CPA) Equation of State”. Batistella. Sofia I. Res..J. Naoko Ellis. M. P. P. “Measurement and correlation of vapor–liquid equilibria for methanol + methyl laurate and methanol + methyl myristate systems near critical temperature of methanol”. Queimada and Joa A. Angelopoulos. pp 791-808 Snadra Gilsic and Dejan Skala “The problem in design and analysis of energy consumption of biodiesel synthesis at supercritical conditions”. Marie Francoise Reyniers. (2006)                . pp 626-632 Soojin Lee. Martins. Luyben. Teles.A. volume 92.R. App. C. “Measurement and correlation of vapor–liquid equilibria for methanol + methyl laurate and methanol + methyl myristate systems near critical temperature of methanol”. 257. Yasuhiko Arai. 2006.. pp 217–222  Zhong Tang. 2006. Yoshio Iwai. pp 136148 G. Chem. 280. tech. Fuel process. Apostolakou. pp 293-301 Alex H. Dusko Posarac. 2010. Sanchez and Carlos A. Bo Seok Jin. Bio. Rob J. of sup. pp 5266-5274 Tanguy F. Oliveira. Issue 3. Liang Gao. of Chem. de Lima da Silva.Marazioti and K. pp 6587–6601 Y. Bio. Enze Min. Fluid Phase Equilib. volume 101. volume 90. “Measurement and calculation of vapor–liquid equilibria for methanol + glycerol and ethanol + glycerol systems at 493–573K”. 284. volume 49. Res. Christensen and Birgin Norddahl “Process simulation and economical evaluation of enzymatic biodiesel production plant”.B.P. pp 454-462 A. J.. A.Kookos. 99. Yoshio Iwai. July. Liang Zhao.. Eng. 2006. 2011. Coutinho. Issue 5.D... Res.Toshio Abeta. pp 3452–3458 M. volume 88. Gutierrez. McLean and M. issue 14. Buxing Han. Inc. “Phase equilibria of methanol–triolein system at elevated temperature and pressure”. pp 10231031 Marta G.2007. 2008. Takeshi Hirayama. issues 5-6. N. Martins.6  INTERNATIONAL CONFERENCE ON CURRENT TREND IN TRCHNOLOGY. 2006.