Liquid-Liquid Phase Equilibrium Of Single Fatty Acid Methyl Ester Systme

As an emerging renewable energy, biodiesel has attracted extensive research for its economy and eco-friendly. At present, researches on biodiesel focus on development of new catalysts such as enzyme catalyst and solid acid catalyst and optimization of reaction conditions.The phase equilibrium data related to biodiesel are very important to the design and optimization of reactors and separation processes. Separation limits are determined by liquid-liquid phase equilibrium (LLE) data, based on which proper extractants are chosen. LLE data is also the basis of chemical process simulation for new systems. The lack of LLE data for biodiesel systems limits the development of separator design and process simulation. However, there are still few reports on relevant phase equilibrium. Present high quality experimental data and effective interaction parameters are needed. Because of the great number of the systems and design condition of potential practical interest, the amount of experimental data obtained will never amount to more than a small fraction of that required. Therefore it is necessary to make predictions from a minimum amount of experimental data.Using self-designed constant temperature oscillator and GC, with suitable experimental conditions, LLE data of three binary systems (methanol+soybean oil, soybean oil+glycerol, glycerol+biodiesel), ternary systems(methyl palmitate+methanol+water, methyl palmitate+methanol+glycerol. methyl stearatem+ethanol+water, methyl stearatem+ethanol+glycerol. mixture+ethanol+glycerol) were determined experimentally at different temperature. And the LLE data obtained were presented in the phase diagrams.This paper has a discussion on the selection of αij. The results show that αij is relevant with the components and temperature. And it has impact on the regression. The value should be reselected when the condition is changed.For the widely use of LLE data, NRTL and UNIQUAC models are used to regress the experiment data through chemical process simulation software Aspen Plus. Binary interaction parameters are obtained and RMSDs are calculated to check the precision of regression. The results show that these parameters can be used to predict phase behavior of ternary systems. And these parameters expand the use scope of LLE data for biodiesel systems.Based on the calculated parameters, phase behavior of ternary systems are forecasted using Aspen Plus. Results show that binary interaction parameters of components in binary systems can be used in ternary systems especially for the industrial application. According to this, more realistic prediction will be made.