Article pubs.acs.org/jced
Ternary Liquid−Liquid Equilibrium of Biodiesel Compounds for Systems Consisting of a Methyl Ester + Glycerin + Water Joseph C. Bell,* Richard A. Messerly, Ryan Gee, Aaron Harrison, Richard L. Rowley, and W. Vincent Wilding Department of Chemical Engineering, Brigham Young University, Provo, Utah 84601, United States ABSTRACT: Ternary LLE data have been experimentally measured for several systems consisting of biodiesel compounds. Systems measured include mixtures with the methyl esters of lauric, myristic, palmitic, and oleic acids, each with glycerin and water. Data were collected at atmospheric pressure and 60 °C. These ternary systems have been correlated using the NRTL equation. These data and correlation parameters can be used to improve separations efficiency in transesterified biodiesel fuels.
■
Experimental Apparatus. Equilibrium experiments were carried out in a cylindrical glass cell (Figure 1B) with a volume of approximately 250 mL. The top of the cell was threaded with coarse threads to receive a matching PTFE cap fitted with an O-ring to seal the vessel. The cap was machined with five holes passing through, and into each hole was inserted a threaded stainless steel fitting. Two of these fittings contained 1/16 in stainless steel sampling lines with valves and syringe ports (Figure 1F) extending into each of the two liquid phases. A third fitting in the cap allowed for measurement of the actual temperature in the mixture via a platinum RTD (Figure 1G). A fourth fitting connected to tubing permitted the vapor space in the cell to be pressurized with nitrogen, evacuated by vacuum, or vented to the atmosphere. A custom-made compression fitting allowed a stir shaft (Figure 1C) to pass through the center of the cap, sealing the shaft with two compressed O-rings. The shaft was fitted with two propellers to provide stirring in each liquid phase and ensure intimate contact of the two phases. The glass cell was held in place with a pair of ring clamps to prevent it from moving or spinning during stirring. This entire equilibrium cell and support structure was contained within a thoroughly mixed temperature controlled water bath (Figure 1A). Temperature control inside the cell was within ± 0.1 K of the set point. To combat a relatively high evaporation rate, the water level in the bath was maintained by a float valve gravity fed from a secondary tank. Experimental Procedure. The cell was charged with a ternary mixture of a fatty-acid methyl ester (x1), water (x2), and
INTRODUCTION Biodiesel processes are processes where vegetable oils (triglycerides) are extracted from oilseed feedstocks and then catalytically upgraded to biodiesel through transesterification. This reaction produces a mixture of fatty acid esters and glycerin. Separation of the esters (biodiesel) and glycerin can be accomplished through liquid−liquid extraction by water addition. Designing this extraction with water as the solvent requires ternary liquid−liquid equilibrium data for mixtures of fatty acid esters, water, and glycerin.1 This work is focused on experimental liquid−liquid equilibrium measurement for systems of fatty acid methyl esters, water, and glycerin.
■
EXPERIMENTAL SECTION Chemicals. Deionized and distilled water was used for the equilibrium experiments. All other chemicals were obtained from Sigma-Aldrich. Purities of the chemicals used as obtained from lot analyses provided by the manufacturer are shown in Table 1. The purity of each chemical was also verified by GC Table 1. Purity of Chemicals Used component
purity, mass %
methyl caprate (internal standard) methyl laurate methyl myristate methyl palmitate methyl stearate methyl oleate glycerin
99.3 99.1 99.9 99.0 98.7 99.6 99.9
Received: December 21, 2012 Accepted: February 28, 2013 Published: March 11, 2013
analysis. Chemicals were all used as received with no further purification. © 2013 American Chemical Society
1001
dx.doi.org/10.1021/je301348z | J. Chem. Eng. Data 2013, 58, 1001−1004
Journal of Chemical & Engineering Data
Article
Figure 1. Experimental apparatus.
Table 2. Mole Fraction Tie-Line Data at T = 60 °C and P = 0.086 MPa: Methyl Laurate (x1)−Water (x2)−Glycerin (x3). Uncertainties (u) Are Percentages Relative to the Value ester phase x1 (calc)
x2
0.9958 0.9918 0.9891 0.9850 0.9774 0.9667 0.9573 0.9515
0.0003 0.0045 0.0077 0.0122 0.0204 0.0320 0.0423 0.0485
± u2
water/glycerin phase x3
25% 10% 2% 3% 1% 2% 1% 0.3%
± u3
0.0039 0.0037 0.0032 0.0029 0.0023 0.0013 0.0004 0.0000
x1
2% 17% 9% 12% 18% 22% 52% 0.0%
2.4 2.1 1.5 1.1 7.0 2.6 2.1 2.5
Table 3. NRTL Parameters at T = 60 °C: Methyl Laurate (x1)−Water (x2)−Glycerin (x3) i
j
τij
τji
1 1 2
2 3 3
2.49 4.15 1.33
13.45 8.89 −1.09
10−5 10−5 10−5 10−5 10−6 10−6 10−7 10−7
× × × × × × × ×
± u1 10% 32% 16% 51% 13% 29% 46% 53%
x2
± u2
x3 (calc)
0.0025 0.1166 0.2168 0.3162 0.5052 0.7253 0.8892 1.0000
9% 1% 0.4% 1% 2% 1% 1% 0.0%
0.9974 0.8833 0.7832 0.6838 0.4948 0.2747 0.1108 0.0000
evacuating by vacuum, the cell was vented to the atmosphere. Ambient pressure (0.086 MPa) was maintained in the cell throughout the heating, stirring, and settling stages by venting to the atmosphere. Adequate time was allowed for the water bath and cell to reach the set point temperature of 60 °C. When the temperature had equilibrated, the mixture was stirred vigorously for 30 min. By the end of the allotted stirring time the mixture was an even emulsion. In most cases individual bubbles of each phase were not visible; instead the mixture was cloudy throughout. This cloudy mixture was left to settle long enough for the two phases to fully disengage. In most cases two clear phases were attained within 24 h. Prior to extracting samples the sampling line for each phase was purged with approximately 30 times the dead volume of the sampling line and valve. Twelve samples were then extracted from each phase using the sampling lines. Throughout the purge and
glycerin (x3). The initial mixture composition was determined by specifying an overall water content. The amount of each component to add was then calculated such that there were approximately 60 mL of each phase in the cell. After the appropriate amount of each component was added, the vapor space above the liquid mixture was purged by pressurizing the cell with nitrogen followed by evacuating the cell to 1 psia. This cycle was repeated three times. On the third cycle instead of
Table 4. Mole Fraction Tie-Line Data at T = 60 °C and P = 0.086 MPa: Methyl Myristate (x1)−Water (x2)−Glycerin (x3). Uncertainties (u) Are Percentages Relative to the Value ester phase
water/glycerin phase
x1 (calc)
x2
± u2
x3
± u3
0.9960 0.9935 0.9904 0.9869 0.9799 0.9706 0.9635 0.9579
0.0011 0.0041 0.0072 0.0110 0.0184 0.0286 0.0363 0.0421
28% 12% 3% 6% 2% 2% 1% 3%
0.0029 0.0025 0.0024 0.0021 0.0017 0.0008 0.0002 0.0000
10% 10% 13% 8% 16% 34% 87% 0.0%
x1 8.3 6.8 4.7 3.3 9.6 2.9
