Liquid-liquid equilibrium phase diagrams of polyethyleneglycol + sodium tartrate + water two-phase systems

Article abstract of DOI:10.1021/je8000188

The phase diagrams of aqueous two-phase systems formed by water, sodium tartrate, and polyethyleneglycol of different molecular weights (600, 1000, 2000, 4000, 6000, and 8000) were determined at 298 K. The effect of NaCl on the liquid-liquid equilibrium was also analyzed. The binodal curves were satisfactorily described using a four-parameter sigmoidal equation. The tie lines were determined by combining a colorimetric method and density measurements. The reliability of the measured tie line compositions was ascertained by the Othmer-Tobias and Bancroft correlation equations.

Full text of DOI:10.1021/je8000188

J. Chem. Eng. Data 2008, 53, 1175–1178
1175
Liquid–Liquid Equilibrium Phase Diagrams of Polyethyleneglycol + Sodium
Tartrate + Water Two-Phase Systems
Luciana Pellegrini Malpiedi, Cintia Fernández, Guillermo Picó, and Bibiana Nerli*
Physical Chemistry Department and CONICET, Faculty of Biochemistry and Pharmaceutical Sciences, National University of
Rosario, Suipacha 570, S2002LRK Rosario, Argentina
The phase diagrams of aqueous two-phase systems formed by water, sodium tartrate, and polyethyleneglycol
of different molecular weights (600, 1000, 2000, 4000, 6000, and 8000) were determined at 298 K. The
effect of NaCl on the liquid–liquid equilibrium was also analyzed. The binodal curves were satisfactorily
described using a four-parameter sigmoidal equation. The tie lines were determined by combining a
colorimetric method and density measurements. The reliability of the measured tie line compositions was
ascertained by the Othmer-Tobias and Bancroft correlation equations.
Introduction
Aqueous two-phase systems (ATPSs) are formed when two
mutually incompatible water-soluble polymers (e.g., dextran and
polyethyleneglycol, PEG) or one polymer and one inorganic
salt (e.g., PEG and potassium phosphate) are dissolved in water
above a critical concentration. The unique feature of these
biphasic systems is that both phases have mass fractions of
almost (90 to 95) % water. This biaqueous nature provides a
gentle and protective environment for biological materials and
makes it possible to use aqueous two-phase systems for
partitioning and separation of cells, organelles, enzymes, etc.
Although ATPSs were reported for the first time by Beijernick1
as early as 1896, almost 60 years later, Albertsson2 began
employing ATPSs for bioseparation purposes. Since then, many
research groups have been devoted to the study of these systems
and their applications. Among ATPSs suitable for bioseparation,
those formed by polymer and salt are the most widely used for
large-scale extraction due to their low cost and rapid phase
disengagement. PEG has been predominantly combined with
either phosphate or sulfate salts.3 However, high concentrations
of these salts are not desirable in the effluent streams due to
environmental problems. Recently, several authors4,5 have used
a citrate salt as a substitute due to its biodegradability and
nontoxicity. Later, Tubio et al.6 demonstrated that PEG +
sodium citrate systems had partitioning properties similar to the
traditional PEG + phosphate ATPSs and could be satisfactorily
applied for separating pancreatic enzymes. Ananthapadmanab-
han and Goddard7 have reported that tartrate salts could also
form ATPS with PEG of molecular weight 3350. The tartrate
anion is known to be readily and 100 % biodegradable.8 It is
used to enhance flavors in foods, confectionery, and beverages.
No pollution problems are known to be caused by it. Therefore,
PEG + tartrate salts + water could form environmentally safe
aqueous two-phase systems which might constitute a suitable
alternative for extraction of biological materials. Before applying
these systems for bioseparation purposes, detailed information
on their phase composition and physicochemical properties is
required. However, at present, these ternary PEG + tartrate +
Figure 1. Binodal curve and tie lines for water (1) + PEG2000 (2) + sodium
tartrate (3) systems at 298 K and pH 4.90.
Table 1. Values of Fit Parameters i, j, x0, and y0 of Equation 3 for
Water (1) + PEG (2) + Sodium Tartrate (3) at 298 K and pH 4.90
parameters
PEG
i
j
x0
y0
σa
600
1797.0
63.7
245.2
55.4
81.1
92.7
-13.34
-5.03
-6.14
-3.09
-3.21
-3.27
-45.184
5.90
-7.11
4.88
-6.00
-1.90
-2.80
-1.56
-1.09
-0.86
0.39
0.25
0.13
0.49
0.69
0.55
1000
2000
4000
6000
8000
2.13
1.43
in NaCl 5 wt %/wt presence
1000
4000
8000
1303.0
78.1
74.8
-7.59
-3.21
-2.32
-22.80
-3.40
-1.65
-0.49
0.69
0.81
0.65
1.43
1.62
a σ, Standard deviations.
water systems have not been explored enough, and the available
experimental data of their liquid–liquid equilibrium are limited.
The goal of this work is to obtain a complete set of
measurements on the phase equilibrium for PEGs of increasing
molecular weights and sodium tartrate solutions with pH 4.90
at 298 K. In addition to the equilibrium measurements, the
* Corresponding author. Phone: +54(341) 4804592. Fax: +54(341) 4804598.
E-mail: bnerli@fbioyf.unr.edu.ar.
10.1021/je8000188 CCC: $40.75
2008 American Chemical Society
Published on Web 04/09/2008

1176 Journal of Chemical & Engineering Data, Vol. 53, No. 5, 2008
For the determination of the tie lines, a series of ATPSs of
at least three different known total compositions were prepared
in 5 mL graduated glass tubes (uncertainty ( 0.05 mL) and
placed in a thermostatic bath. Each mixture was prepared in
triplicate. After reaching the phase equilibrium, visual estimates
of top and bottom volumes (VT and VB, respectively) were made.
When phases were separated, they were properly diluted to
determine the polymer and salt equilibrium concentrations.
The concentrations of tartrate were determined according to
a colorimetric assay.10 It was based on the reaction of tartaric
acid with ammonium metavanadate in dilute aqueous acetic acid
solution to form a red dye that was determined spectrophoto-
metrically at 530 nm. The uncertainty in the measurement of
the mass fraction of the salt was estimated to be ( 0.002. The
analysis showed that the presence of PEG did not interfere with
the tartaric determination.
The mass fraction of PEG was determined from density
measurements at 25 °C using an Anton Paar model DMA 35N
densimeter. All density measurements were done with a (
0.0001 g · cm-3 uncertainty. Each mixture was prepared and
analyzed in triplicate. Temperature was maintained within (
0.1 °C. Since the density of the phase samples depends on the
PEG and salt compositions, calibration plots of density versus
both polymer and sodium tartrate concentrations were obtained.
The calibration was given by the following relation between
density (F) and the tartrate (w3) and PEG (w2) mass fractions
Figure 2. Binodal curves for water (1) + PEG of varying molecular weight
(2) + sodium tartrate (3) systems at 298 K and pH 4.90. Experimental
data: 2, PEG600; 9, PEG1000; [, PEG2000; 1, PEG4000; b, PEG6000.
Calculated data using eq 3 (—).
empirical description of the phase equilibrium data with an
adequate equation will be performed to provide the basis for
interpolating and extrapolating experimental data and predicting
phase compositions when such data are not available.
F ) a + bw3 + cw2
(1)
Experimental Section
The estimated values of the coefficients a, b, and c (in g ·cm-3
units) for the present system were 0.9980, 0.9215, and 0.1776,
respectively. We found that these coefficients were practically
independent of the molecular mass of the PEG and comparable
to those coefficients obtained by Taboada et al.11 for Na2CO3
and PEG mixtures. The uncertainty in the measurement of the
mass fraction of PEG was estimated to be 0.003. A mass balance
check was made between the initial mass of each component
and the amounts in the bottom and top phases on the basis of
equilibrium compositions. The relative error in the mass balance
was < 2 %. The tie line lengths (TLL) for the different
compositions were calculated according to
Chemicals. Polyethyleneglycols of the following average
molecular masses, 600 (PEG600), 1000 (PEG1000), 2000 (PEG-
2000), 4000 (PEG4000), 6000 (PEG6000), and 8000
(PEG8000), were purchased from Sigma Chemical Co. and
used without further purification. Tartaric acid was obtained
from Sigma Chemical Co. and used without further purifica-
tion. All the other reagents were of analytical grade with a
minimum purity of 99 %.
Procedures. A phase diagram includes the binodal curve and
the tie lines. The determination of the binodal curve was carried
out by a turbidimetric titration method.9 Stock aqueous solutions
of PEGs of different molecular weights with a mass fraction of
40 % and sodium tartrate with a mass fraction of 25 % at pH
4.90 were employed. Sodium tartrate solution was prepared from
a tartaric solution whose pH was adjusted to 4.90 by the addition
of the appropriate quantities of sodium hydroxide. Small aliquots
of approximately (0.01 to 0.05) g of the polymer stock solution
were added to 1 g of the sodium tartrate stock solution that
was placed in a glass tube. After each aliquot addition, the
system was thoroughly mixed. The first appearance of turbidity
(the cloud point) indicated that the system was about to enter
the two-phase region. With knowledge of the composition of
the starting polymer and salt solutions and of the added masses,
the total system composition, just prior to the two-phase
formation, was calculated and provided a point on the binodal
curve. The starting and added solution masses were measured
on an analytical balance with a precision of ( 0.0001 g.
Additional binodal points were obtained by adding a small
amount of water to clear the system and then enough drops of
the PEG stock solution to produce turbidity again. To obtain
the binodal points corresponding to higher polymer mass
fractions, the above-mentioned procedure was inverted, thus
titrating the stock polymer solution with the stock salt solution.
The system temperature was maintained constant and controlled
to within ( 0.1 °C by immersing the glass tube and the stock
solutions in a thermostatic bath.
TLL ) wT - wB 2 + wT - wB 2 1 ⁄ 2
(2)
[(
)
(
) ]
2
2
3
3
where wT2, w3T, wB2 , and w3B are the top (T) and bottom (B)
equilibrium mass fractions of PEG (2) and sodium tartrate (3).
The tie line lengths are expressed in mass fractions.
Results and Discussion
Figure 1 shows the phase diagram corresponding to the water
(1) + PEG2000 (2) + sodium tartrate (3) at pH 4.90 ATPS at
298 K. It includes the binodal curve and the tie lines. ATPSs
with total initial compositions above the binodal curve separate
into two equilibrium immiscible phases: a PEG-poor, tartrate-
rich bottom layer and a PEG-rich, tartrate-poor upper phase. A
similar behavior was observed for the traditional PEG +
phosphate ATPSs.2,3
Binodal data were fitted with different expressions, previously
used by other authors;5,12,13 however, better results were
obtained with the following empirical equation14
i
w2 ) yo +
(3)
w3-xo
1 + e
j
where w2 and w3 (PEG and Tart mass fraction, respectively)
are the coordinates of the binodal points; and yo, xo, i, and j

Journal of Chemical & Engineering Data, Vol. 53, No. 5, 2008 1177
Table 2. Phase Compositions for the Water (1) + PEG (2) + Sodium Tartrate (3) Systems at 298 K and pH 4.90
total compositions top phase bottom phase
c
100 w3
100 w2
100 w3
100 w2
100 w3
100 w2
STLa
100TLLb
Kp
PEG600
19.98
12.17
12.18
12.85
12.17
19.50
20.00
20.62
19.50
7.24
6.93
5.78
5.23
28.63
29.43
32.53
43.10
7.32
7.42
5.61
4.87
-1.66
-1.70
-1.68
-1.68
24.83
25.53
31.33
42.02
2.76
2.87
3.77
4.34
19.88
21.80
22.68
PEG1000
15.27
12.18
12.85
12.93
12.67
20.00
20.62
21.20
20.67
8.32
6.76
5.36
4.16
25.87
31.51
36.93
41.75
7.85
5.25
2.72
1.33
-2.59
-2.58
-2.51
-2.48
19.31
28.16
36.83
43.59
1.84
2.50
3.54
4.92
16.94
19.00
20.47
PEG2000
14.87
10.87
12.00
13.00
12.50
13.00
15.00
6.05
4.54
3.08
22.93
29.14
36.39
3.85
1.82
0.18
-2.16
-2.16
-2.16
21.02
30.09
39.91
2.45
3.78
6.45
17.16
19.87
PEG4000
13.44
12.97
13.66
13.91
14.57
9.37
9.48
9.75
9.95
5.30
4.86
4.62
4.22
24.24
26.21
27.28
29.07
1.70
1.11
0.52
0.07
-2.77
-2.72
-2.61
-2.53
23.96
26.75
28.66
31.18
2.54
2.90
3.22
3.72
14.10
14.88
15.68
PEG6000
11.47
9.50
8.30
10.00
10.50
11.45
6.01
4.70
4.22
3.51
17.53
24.02
26.71
30.86
3.08
1.24
0.91
0.48
-2.64
-2.64
-2.66
-2.72
15.45
24.36
27.56
32.37
1.91
2.84
3.30
4.18
10.02
10.31
10.65
13.34
13.92
14.68
PEG8000
12.76
9.50
8.79
10.00
10.00
10.00
14.00
12.00
14.00
4.97
4.25
3.97
3.48
22.62
26.53
28.32
31.46
1.18
0.66
0.34
0.03
-2.71
-2.76
-2.71
-2.68
22.81
27.51
29.83
33.53
2.57
3.20
3.60
4.37
13.62
14.31
15.21
a STL, tie line slope. b TLL, tie line length. c Kp, Fraction of sodium tartrate retained in the bottom phase divided by the tartrate in the top phase.
Table 3. Values of the Parameters of Othmer-Tobias and Bancroft
Equations, k, n, k1, and r, for Water (1) + PEG (2) + Sodium
Tartrate (3) Systems at 298 K and pH 4.90
parameters
PEG
k
n
σa
k1
r
σa
600
0.0134
0.0951
0.1284
0.2444
0.0253
0.0479
3.742
1.981
1.866
1.369
2.564
2.218
0.0110
0.0070
0.0011
0.0007
0.0118
0.0017
3.072
3.436
3.137
3.044
4.378
4.072
0.2175
0.4153
0.4929
0.6834
0.3574
0.4346
0.0023
0.0033
0.0011
0.0002
0.0034
0.0009
1000
2000
4000
6000
8000
a σ, Standard deviations.
hydrophobic character of PEGs of higher molecular weight. The
binodal curve corresponding to systems of PEG8000 was not
included in Figure 2 since it superimposed to the binodal curve
of PEG6000 + sodium tartrate ATPSs.
The liquid–liquid equilibrium data of assayed water + PEG
+ sodium tartrate ATPSs are given in Table 2. It also lists the
value of the partition coefficient (Kp) of the salt, defined as the
fraction of salt retained in the bottom phase divided by the salt
in the top phase. High values of Kp were obtained (from 1.84
to 6.45) being an indication of the separation obtained by adding
an amount of PEG to a given brine solution.
For most of the studied systems, the slope of the tie lines
(STL) is practically constant, which implies that tie lines are
parallel to each other, thus allowing us to know the coexisting
phase compositions for any given total polymer phase-forming
composition. ATPSs formed by PEGs of molecular weights
4000, 6000, and 8000 show the higher STL absolute values.
An increase in the STL magnitude indicates an increase in the
difference between the polymer concentrations at a given
difference in the salt concentrations. This implies a decrease in
Figure 3. Effect of the presence of NaCl, mass fraction of 5 %, on the
binodal curve of the water (1) + PEG (2) + sodium tartrate (3) systems at
298 K and pH 4.90. 9, PEG1000; 0, PEG1000/NaCl; 1, PEG4000; 3,
PEG4000/NaCl; `, PEG8000; ", PEG8000/NaCl.
are the fit parameters. The parameters of eq 3 along with the
corresponding standard deviations (σ) are summarized in
Table 1.
The effect of PEG molecular weight on phase separation was
evaluated. Figure 2 summarizes the binodal data corresponding
to ATPSs formed by sodium tartrate and PEGs of different
molecular weights. It can be seen in this figure that binodal
curves became close to the origin with the increase in PEG
molecular weight, thus requiring lower concentrations for phase
separation. This may be caused by the increase in the incompat-
ibility between the system components, due to the more

1178 Journal of Chemical & Engineering Data, Vol. 53, No. 5, 2008
the mutual solubility of the aqueous polymer- and salt-containing
media. A similar behavior was observed for PEG + citrate
ATPSs.14
a promising, versatile, and attractive system in the field of
bioseparation.
Acknowledgment
Several authors found that the addition of NaCl to a PEG-salt
system may increase the system resolution by promoting an
increase in the difference between the partition coefficients of
a target and its contaminant.15 The effect of NaCl presence may
affect protein partition coefficient either by modifying the
interaction between the partitioned proteins with phase com-
ponents or by altering the binodal compositions. Figure 3 shows
that the NaCl presence does not alter the shape of binodal curves
but induces a displacement of them to the left, thus expanding
the biphasic area. This behavior was also observed for other
polymer/salt systems15 and is a consequence of a salting out
effect due to the presence of NaCl.
The authors thank Maria Robson and Marcela Culasso for their
assistance in the translation of this paper.
Literature Cited
(1) Beijernick, M. W. Original mitteilung uber eine eigentumlichkeit der
loslichen starke. Centrabl. Bakteriologie, Parasitenkunde Infektiosk-
rankheiten 1896, 22, 699–701.
(2) Albertsson, P. Partition of Cell Particles and Macromolecules, 2nd
ed.; Wiley: New York, 1971.
(3) Zaslavsky, B. Y. Aqueous Two-Phase Partitioning Physical Chemistry
and Bioanalytical Applications; Marcel Dekker: New York, 1994.
(4) Vernau, J.; Kula, M. R. Extraction of proteins from biological raw
materials using aqueous polyethyleneglycol-citrate phase systems.
Biotechnol. Appl. Biochem 1990, 12, 397–404.
Empirical equations have been proposed to ascertain the
reliability of calculated tie line data in traditional liquid–liquid
extraction, the most used being those of Othmer-Tobias (eq
4) and Bancroft (eq 5)16
(5) Zafarani Moattar, M. T.; Hamidi, A. A. Liquid-liquid equilibria of
aqueous two-phase (polyethyleneglycol)-potassium citrate system.
J. Chem. Eng. Data 2003, 48, 262–265.
(6) Tubío, G.; Nerli, B.; Picó, G. Partitioning features of bovine tripsin
and alpha-chymotrypsin in polyethyleneglycol-sodium citrate aqueous
two-phase systems. J. Chromatogr. B 2007, 852, 244–249.
(7) Ananthapadmanabhan, K. P.; Goddard, E. D. Aqueous Biphase
Formation in Polyehtylene Oxide-Inorganic Salt Systems. Lagmuir
1987, 3, 25–31.
(8) Khan, E.; Sy-Savane, O.; Jittawattanarat, R. Application of commercial
biochemical oxygen demand inocula for biodegradable dissolved
organic carbon determination. Water Res. 2005, 39, 4824–4834.
(9) Hatti-Kaul, R., Ed. Aqueous Two-Phase Systems: Methods and
Protocols; Methods in biotechnology 11; Humana Press: Totowa, NJ,
2000.
(10) Matchett, J. R.; Legault, R. R.; Nimmo, C. C.; Notter, G. K. Tartrates
from grape wastes. Ind. Eng. Chem. 1944, 38, 851–857.
(11) Taboada, M. E.; Asenjo, J. A.; Andrews, B. A. Liquid-liquid and liquid-
liquid-solid equilibrium in Na2CO3-PEG-H2O. Fluid Phase Equilib.
2001, 180, 273–280.
w1B
w1T
r
) k1
(4)
(5)
B
T
( )
( )
w3
w2
1 - wT2
1 - wB3
w3B
n
) k
w2T
(
)
(
)
where wT1 and wB1 are the mass fractions of water in the top and
bottom phases and k, n, k1, and r represent the parameters to
be determined. Linearization of both equations produced ac-
ceptable consistency in the results. Values of the fit parameters
and the corresponding coefficients of determination are given
in Table 3.
(12) Graber, T. A.; Taboada, M. E. Liquid-liquid Equilibrium of the
Poly(ethylene glycol) + Sodium Nitrate + Water System at 298.15
K. J. Chem. Eng. Data 2000, 45, 182–184.
(13) Gonzále Tello, P.; Camacho, F.; Blázquez, G.; Alarcón, F. J. Liquid-
liquid Equilibrium in the System Poly(ethyleneglycol) + MgSO4
H2O at 298 K. J. Chem. Eng. Data 1996, 41, 1333–1336.
+
Conclusions
(14) Tubio, G.; Pellegrini, L.; Nerli, B. B.; Picó, G. A. Liquid-liquid
equilibria of aqueous two-phase systems containing poly(ethylene-
glycols) of different molecular weight and sodium citrate. J. Chem.
Eng. Data 2006, 51, 209–212.
The aqueous two-phase partitioning method of liquid–liquid
extraction is useful for separating material of biological origin.
Water + PEG + sodium tartrate systems seem to be attractive
for protein purification because of the biodegradability of
tartrate. Composition and properties of these systems are
necessary for the design of an extraction process. However, this
information was not available up to the present. In this work,
the phase diagrams of water + PEG + sodium tartrate ATPSs
were determined and reliable, and complete data were obtained.
The phase formation showed to be affected by both the PEG
molecular weight and the NaCl presence. These characteristics
and several additional advantages such as low cost and rapid
phase separation make water + PEG + sodium tartrate systems
(15) Marcos, J. C.; Fonseca, L. P.; Ramalho, M. T.; Cabral, J. M. S. Partial
purification of penicillin acylase from Escherichia coli in poly(ethylene
glycol)-sodium citrate aqueous two-phase systems. J. Chromatogr. B
1999, 734, 15–22.
(16) Othmer, D. F.; Tobias, P. E. Liquid-Liquid Extraction Data-Toluene
and Acetaldehyde Systems. Ind. Eng. Chem. 1942, 34, 690–692.
Received for review January 7, 2008. Accepted March 10, 2008. This
work was supported by grants from CONICET (PIP5053) and ANPCyT
(FonCyT 06-12476/02).
JE8000188