APPLIED CHEMISTRY
Some Novel Liquid Partitioning Systems: Water-Ionic Liquids and
Aqueous Biphasic Systems
Michael H. Abraham* and Andreas M. Zissimos
Department of Chemistry, University College London, 20 Gordon Street, London WC1H OAJ, U.K.
Jonathan G. Huddleston, Heather D. Willauer, and Robin D. Rogers
Department of Chemistry and Center for Green Manufacturing, The University of Alabama,
Tuscaloosa, Alabama 35487
William E. Acree, Jr.
Department of Chemistry, University of North Texas, Denton, Texas 76203-068
Partition coefficients, as values of log P, between water and two room-temperature ionic liquids
and between water and an aqueous biphasic system have been correlated with Abraham’s solute
descriptors to yield linear free energy relationships that can be used to predict further values of
log P, to ascertain the solute properties that lead to increased or decreased log P values, and to
characterize the partition systems. It is shown that, in all three of the systems, an increase in
solute hydrogen-bond basicity leads to a decrease in log P and an increase in solute volume
leads to an increase in log P. For the two ionic liquid systems, an increase in solute hydrogen-
bond acidity similarly decreases log P, but for the aqueous biphasic system, solute hydrogen-
bond acidity has no effect on log P. These effects are rather smaller than those observed in
traditional water-solvent systems. However,the ionic liquids appear to have an increased affinity
for polyaromatic hydrocarbons as compared to traditional organic solvents. Principal component
analysis and nonlinear mapping show that the three unconventional partition systems are
considerably different from conventional water-organic solvent systems.
Introduction
A major contemporary industrial challenge is to
continued manufacturing beneficial chemical products
while eliminating or substantially reducing the detri-
mental environmental consequences of the processes
adopted. The Montreal Protocol
1
identified the need to
reevaluate chemical processes to take account of their
environmental impact, especially with regard to the use
of volatile organic solvents. In addition, some 90% of
hazardous waste is aqueous in nature,
2
and thus,
industry is reliant upon efficient separations from liquid
media. To this end, liquid-liquid separations are widely
applied in the chemical process industry. Typically,
because of their immiscibility with water, volatile
organic solvents are often employed in such processes.
3
Taken together, these issues suggest that the elimina-
tion of the use of flammable toxic and volatile organic
solvents in separations processing represents a signifi-
cant step in the creation of a sustainable industrial
technology.
4
A number of different approaches to this problem
have been identified, including solvent-free synthesis,
the use of water as a solvent,
5
the use of supercritical
fluids,
6
and the use of ionic liquids. Recently, room-
temperature ionic liquids (RTILs) have received world-
wide attention
7,8
as replacements for organic solvents
in catalysis,
9
synthesis,
10,11
and separations processes.
12,13
Room-temperature ionic liquids, in contrast to conven-
tional ionic liquids such as molten sodium chloride,
which are only liquids at temperatures above 800 °C,
represent ionic salts that are liquid at room tempera-
ture. Many RTILs are liquids over a wide temperature
range, and RTILs with melting points as low as -96 °C
are known. The constituents of many RTILs (being
ionic) are constrained by high Coulombic forces and thus
exert practically no vapor pressure above the liquid
surface. These ionic liquids offer a highly solvating, yet
noncoordinating medium in which a number of organic
and inorganic solutes can be dissolved. Many RTILs are
nonvolatile and nonflammable and have high thermal
stabilities. Importantly, RTILs can be relatively unde-
manding and inexpensive to manufacture. These prop-
erties have been the subject of recent reviews.
14,15
Several studies of the solvatochromic properties of
RTILs have also been published.
16-19
More recently,
studies have been made of a group of 1-alkyl-3-meth-
ylimidazolium ionic liquids, using a linear free energy
relationship (LFER) to characterize the solute distribu-
tion between water and the RTILs.
20
LFERs have been
used to characterize ionic liquids as gas chromato-
graphic (GC) stationary phases, but at 120 °C only,
21-26
and in a recent report,
27
gas-RTIL partition coefficients
were determined. It should be noted that physicochem-
ical properties of pure ionic liquids, as determined by
413Ind. Eng. Chem. Res. 2003, 42, 413-418
10.1021/ie020520y CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/11/2003
GC for example, will generally not be the same as those
of ionic liquids in water partition systems, because, in
the latter case, the ionic liquids are saturated with
water.
Aqueous biphasic systems (ABSs) represent a quite
different type of alternative solvent system that have
mainly been applied to the fractionation of labile
biological materials.
28-31
However, recent work shows
that they also appear to have some utility for solvent
replacement in the extraction of metal ions
32-35
and
small organic molecules,
36,37
as a solvent for pharma-
ceutical compounds,
38
and as a reaction medium.
39,40
Aqueous biphasic systems (ABSs) represent critical
phenomena
41
that occur in aqueous solution when two
or more polymers, or a polymer and a salt, are added to
water above critical concentrations or temperatures.
36,37
Two immiscible liquid phases are formed as a result
without the involvement of any organic solvent but still
with the potential for use in the differential partition
and extraction of a wide variety of solutes. Solutes as
varied in molecular size as inorganic ions, small organic
molecules, biological macromolecules, colloidal inorganic
particles, viruses, and even cells
28-37
can all be success-
fully partitioned between the phases with the correct
choice of ABS.
The solubilizing properties of solutions of polymer
molecules in aqueous solution have been used to effect
the extraction and fractionation of a wide variety of
target solutes. The apparent similarities between many
of these aqueous polymeric systems, including cloud-
point extraction (CPE), micellar extraction (ME), aque-
ous biphasic systems (ABSs), and extractions using
thermoseparating polymers, have recently been re-
viewed.
42
Some studies of the solvent properties of some
of these systems have been made, including the con-
struction of a number of LFERs. For the most part,
these investigations are confined to micellar sys-
tems,
43,44
and few comparable studies are available
relating to the many polymeric systems showing bulk
phase separation. A recent paper illustrates the ap-
plication of LFERs to poly(ethylene glycol) (PEG)-salt
ABSs.
37
None of the above work related to ionic liquids or ABS
systems addresses the important practical problem of
how partition in these systems compares with partition
in conventional water-solvent systems, and the ques-
tion of whether RTIL and ABS systems offer any
enhanced selectivity, in the partition of particular types
of solutes. It is the aim of this work to compare three
nonconventional partitioning systems with a number of
conventional systems. The RTILs are
20
1-butyl-3-meth-
ylimidazolium hexafluorophosphate ([bmim][PF6]), 1-hex-
yl-3-methylimidazolium hexafluorophosphate ([hmim]-
[PF6]), and an ABS
37
composed of poly(ethylene glycol),
MW 2000 Da, and ammonium sulfate (P2000).
Methodology
We base our analysis on the general LFER of Abra-
ham et al.
45-48
In the dependent variable, P is the partition coefficient
for a series of solutes in the same water-solvent system.
The independent variables in eq 1 are solute descriptors
as follows:
45-48
E is the solute excess molar refractivity
in units of mol dm
-3
/10; S is the solute dipolarity/
polarizability; A and B are the overall or summation
hydrogen-bond acidity and basicity, respectively; and V
is the McGowan characteristic volume in units of (mol
dm
-3
)/100. The coefficients in eq 1 are obtained by
multiple linear regression analysis and serve to char-
acterize the system under consideration.
However, the coefficients are not simply fitting values
but represent the difference in chemical properties
between water and the particular solvent in the parti-
tioning system; that is, they can be taken as solvent
properties relative to water. The e coefficient gives the
tendency of the phase to interact with solutes through
polarizability-type interactions, mostly via electron
pairs. The s coefficient is a measure of the solvent
dipolarity/polarizability. The a coefficient represents the
complementary property to solute hydrogen-bond acidity
and so is a measure of the phase hydrogen-bond basicity.
Likewise, the b coefficient is a measure of the phase
hydrogen-bond acidity.
The coefficients for the water-solvent systems are
collected in Table 1. A feature of the two water-RTIL
systems and the ABS system is the small s coefficient,
which indicates that the RTILs and the less aqueous
phase in the ABS system have the same dipolarity/
polarizability as water. Systems in which the solvent
is only poorly dipolar/polarizable, such as hexane, have
very negative s coefficients. The RTIL systems have
rather average negative a coefficients, which indicates
that the RTILs are less basic than water and about as
basic as olive oil, a typical ester. The basicity of the
RTILs is probably mainly due to the counteranion. It is
known
26
that the basicity of the ionic liquid GC station-
ary phases depends on the anionsthe more the negative
charge is dispersed in the anion, the less basic is the
ionic salt. The P2000 ABS is different to the extent that,
as regards basicity, the system resembles the octanol-
water system.
The b coefficients for the RTIL systems lie between
those for the ethylene glycol (EG) and 2,2,2-trifluoro-
ethanol (TFE) systems, which shows the strong hydrogen-
bond acidity of the RTILs. This is in marked contrast
to the ionic liquid GC phases, which have almost zero
hydrogen-bond acidity. However, these species are
invariably derivatives of quaternary ammonium cations
that contain no acidic hydrogen. In contrast, the 1-alkyl-
3-methylimidazolinium cation contains three acidic
hydrogen atoms, as shown in Figure 1. Thus, the
observed acidity of the three RTILs is chemically
log P ) c + eE + sS + aA + bB + vV (1)
Table 1. Coefficients for Water-Solvent Partitioning
Systems
solvent cesabv
[bmim][PF6] -0.17 0.45 0.23 -1.76 -1.830 2.150
[hmim][PF6] -0.13 0.050 0.40 -1.48 -2.110 2.300
P2000 -0.050 0.650 -0.210 0.210 -1.310 1.710
octanol 0.088 0.562 -1.054 0.034 -3.460 3.814
chloroform 0.327 0.157 -0.391 -3.191 -3.437 4.191
cyclohexane 0.159 0.784 -1.678 -3.740 -4.929 4.577
toluene 0.143 0.527 -0.720 -3.010 -4.824 4.525
ether 0.251 0.588 -1.019 -0.238 -4.523 4.043
chlorobenzene 0.040 0.246 -0.462 -3.038 -4.769 4.640
olive oil -0.035 0.574 -0.798 -1.422 -4.984 4.210
acetone/dry 0.335 0.349 -0.231 -0.411 -4.793 3.963
DMF/dry
a
0.136 0.305 0.431 0.469 -4.833 3.735
DMSO/dry
a
-0.250 0.184 0.905 1.921 -4.739 3.509
EG/dry
a
-0.269 0.586 -0.522 0.712 -2.492 2.708
TFE/dry
a
0.395 -0.094 -0.594 -1.280 -1.274 3.088
a
Abbreviations are as follows: DMF, dimethylformamide; DMSO,
dimethyl sulfoxide; EG, ethylene glycol; TFE, trifluoroethanol.
414 Ind. Eng. Chem. Res., Vol. 42, No. 3, 2003
reasonable. A considerable difference between the RTIL
and ABS systems and the conventional systems is the
small v coefficients of the former three systems. In
chemical terms, the RTILs are less “hydrophobic”, which
again makes chemical sense.
Although it is possible to compare LFERs term by
term, as above, it would be very helpful to have an
overall comparison. A useful method of comparison is
to carry out a principal component analysis (PCA) on
the coefficients e, s, a, b, and v for a number of LFERs,
the c constant being irrelevant in this respect. The
procedure is exactly the same as that used for the
analysis of HPLC retention data from several systems.
49
The 15 partitioning systems we have used are listed in
Table 1. In addition to the three unconventional sys-
tems, we include coefficients for seven typical water-
solvent systems. To expand the range of the coefficients,
we also provide data for five systems, denoted as
solvent/dry, where the coefficients refer to hypothetical
partition between water and the pure dry solvents. In
our PCA, the five columns of coefficients in Table 1, e-v,
are transformed into five principal components (PCs)
that contain the same information as the original
coefficients. However, the first two PCs account for 72%
of the total information, and so, if we restrict our
analysis to this 72%, we can reduce a five-component
system to two components. A plot of the scores of PC2
vs the scores of PC1 is shown in Figure 2. It is very clear
that solvents 14 (ethylene glycol, EG) and 15 (2,2,2-
trifluoroethanol, TFE) are closest to the two ionic
solvents and the ABS as regards the information content
of the coefficients. We can quantify this “nearness” by
calculating the distance in the xy plane between the
point for a standard system, no. 1, and the point for any
other system; we denote this distance asδ-PCA. We put
these distances on a scale such that the distance
between the point for the standard system and the point
for the water-cyclohexane partition is 10, and we list
the distances in Table 2.
A technique that allows all of the information content
in the coefficients to be used is that of nonlinear
mapping (NLM),
50
as applied by Valko et al.
51,52
Two
components are obtained from the five coefficients, and
a nonlinear map of the two components is shown in
Figure 3. We calculate the distance in the xy plane
between the point for a given system and the point for
system no. 1 as before. This distance, δ-NLM, is also
scaled as before, and values of the scaled distance are
given in Table 2.
According to both the PCA and NLM methods, the
hypothetical water-solvent systems with solvents TFE
and EG are the closest to the standard ionic liquid
system, no. 1. This is largely due to the rather small
numerical values for the b and v coefficients in the five
solvent group including nos. 1, 2, 3, 14, and 15. In
chemical terms, all five solvents are nearer water in
hydrogen-bond acidity, because a numerically small b
coefficient corresponds to a high solvent acidity. The v
coefficient in water-solvent systems can be thought of
as a measure of solvent hydrophobicity with respect to
water as the standard. The small v coefficients in the
two RTIL systems are probably due to the small
intrinsic hydrophobicity of the ionic liquids themselves,
together with the fact that the organic layer will contain
a substantial amount of water. In the case of the ABS
system, both phases are aqueous, and so the vcoefficient
is almost bound to be rather small. For systems 14 and
15, the small v coefficient is entirely due to the small
intrinsic hydrophobicity of organic solvent.
Figure 1. Ionic liquids bmim, R ) Me, and hmin, R ) Pr.
Figure 2. Plot of PC2 vs PC1.
Figure 3. Plot of NL2 vs NL1.
Table 2. Nearness Parameters for the Water-Solvent
Partition Systems
solvent no. δ-PCA δ-NLM
[bmim][PF6] 1 0.0 0.0
[hmim][PF6] 2 2.9 1.0
P2000 3 2.3 3.4
octanol 4 4.7 5.2
chloroform 5 5.6 6.2
cyclohexane 6 10.0 10.0
toluene 7 7.7 8.5
ether 8 6.0 6.6
chlorobenzene 9 7.6 8.0
olive oil 10 6.8 7.8
acetone/dry 11 5.6 7.8
DMF/dry 12 5.5 8.3
DMSO/dry 13 6.2 9.6
EG/dry 14 1.6 4.5
TFE/dry 15 2.0 1.6
Table 3. Descriptors for Some Solutes
solute ESAB V
iodomethane 0.676 0.43 0.00 0.13 0.5077
4-chloroaniline 1.060 1.13 0.30 0.31 0.9390
benzoic acid 0.730 0.90 0.59 0.40 0.9317
4-hydroxybenzoic acid 0.930 0.90 0.81 0.56 0.9904
benzamide 0.990 1.50 0.49 0.67 0.9728
1,2,4-trichlorobenzene 0.980 0.81 0.00 0.00 1.0836
bromohexane 0.349 0.40 0.00 0.12 1.1290
fluoranthene 2.377 1.55 0.00 0.24 1.5846
bromooctane 0.339 0.40 0.00 0.12 1.4108
Ind. Eng. Chem. Res., Vol. 42, No. 3, 2003 415
It is not obvious how the coefficients dictate the
partitioning behavior of solutes, because any two solutes
will usually differ in several descriptors, not just one.
Furthermore, the analysis is made more difficult in that
the small v coefficient for the RTIL and ABS systems
means that the actual range of observed log P values is
considerably condensed. This might be an advantage,
because log Pvalues in the RTIL and ABS systemscould
be determined in cases where log P values in conven-
tional water-solvent systems might be out of practical
range. Descriptors for some typical solutes are listed in
Table 3; these solutes were chosen because they cover
a range of properties and also because a number of
experimental partition coefficients are available for
these particular solutes.
In Table 4, we present 36 pairs of calculated and
experimental log P values, covering a range of 7.08 log
units in the experimental values. For these 36 pairs,
the average deviation in the log P values, defined as
[Σ(calc - exp)]/36, is -0.01 log units; the average
absolute deviation, defined as [Σ|(calc - exp)|]/36, is 0.11
log units; and the standard deviation, defined as
x{[Σ(calc - exp)
2
]/35}, is 0.14 log units. We are there-
fore reasonably certain that we can use calculated
values of log P in cases where there are no available
experimental values in the following discussion.
Iodomethane and the two bromoalkanes were chosen
to show the effect of solute size, the other descriptors
being almost the same. On the other hand, the solutes
from 4-chloroaniline to benzamide have very similar V
descriptors but differ markedly in other properties. The
pairs of solutes 1,2,4-trichlorobenzene-bromohexane
and fluoranthene-bromooctane were included to exam-
ine any effect of the RTIL systems on aromatic solutes.
A comparison of the three haloalkanes shows very
clearly that an increase in the solute volume results in
much larger increases in log P for the three conventional
systems than for the two RTIL systems and the ABS
system. If we take bromohexane as a standard “non-
polar” solute, then a polar hydrogen-bond base such as
4-chloroaniline has very much smaller log P values in
the three conventional systems, smaller values in the
two RTIL systems, but a nearly unchanged one in the
ABS system (-0.26 log units). However, polar hydrogen-
bond acids are very difficult to extract into cyclohexane
or chloroform, moderately extractable into octanol and
the two RTILs, but easily extracted into the ABS
system. An examination of the final two pairs of solutes
shows that the three unconventional systems have a
slight extra affinity for 1,2,4-trichlorobenzene and a
substantial affinity for the polyaromatic hydrocarbon
fluoranthene. None of the conventional systems would
separate fluoranthene from bromooctane, but there are
considerable differences in the log P values in the three
unconventional systems.
Although the above analysis has been carried out on
log P values that were calculated from the solvation
equation and solute descriptors, a comparison of the
observed and experimental log P values in Table 4
indicates that the method has considerable predictive
value.
Conclusion
The use of our general LFER in an analysis of water-
solvent partitioning systems enables RTIL and ABS
systems to be compared with conventional systems in
general chemical terms. In addition, possible separa-
tions can be investigated theoretically, and partition
coefficients can be predicted through the LFERs and
compound descriptors. A comparison of predicted log P
values for two or more solutes then provides a method
of predicting possible separations and of choosing a
particular solvent system for these separations. At the
moment, descriptors are available for over 3000 com-
pounds,
34
and so a huge number of partitions and
separations can be predicted.
Acknowledgment
We thank the EPSRC for support of this work and
for a postdoctoral fellowship (to A.Z.). We are very
grateful to Dr Klara Valko for a copy of her program
for nonlinear mapping. The RTIL research at The
University of Alabama was supported by the U.S.
Environmental Protection Agency’s STAR program
through Grant R-82825701-0. (Although the research
described in this article was funded in part by the EPA,
it has not been subjected to the Agency’s required peer
and policy review and therefore does not necessarily
reflect the views of the Agency and no official endorse-
ment should be inferred.) The ABS work at The Uni-
versity of Alabama was supported by the Division of
Chemical Sciences, Office of Basic Energy Sciences,
Table 4. Calculated Values of log P for Solutes in Water-Solvent Partitioning Systems
a
solute
octanol
(4)
cyclohexane
(6)
CHCl
3
(5)
[bmim][PF6]
(1)
[hmim][PF6]
(2)
P2000
(3)
iodomethane 1.50
(1.51)
1.65 1.95
(2.13)
1.09
(0.93)
0.97
(0.97)
0.99
(0.75)
4-chloroaniline 1.87
(1.88)
0.74
(0.69)
2.01
(2.09)
1.48
(1.25)
1.41
(1.31)
1.61
(1.44)
benzoic acid 1.74
(1.87)
-0.69
(-0.85)
0.74
(0.71)
0.60
(0.57)
0.69
(0.85)
1.43
(1.45)
4-hydroxybenzoic acid 1.53
(1.58)
-1.88
(-1.77)
-0.24
(-0.50)
0.14
(0.06)
0.17
(-0.04)
1.49
(1.63)
benzamide 0.47
(0.64)
-2.26
(-1.92)
0.11
(0.11)
0.63
(0.66)
0.62
(0.66)
1.16
(1.24)
1,2,4-trichlorobenzene 3.92
(4.02)
4.53 4.71 2.79
(2.97)
2.73
(2.74)
2.26
(2.14)
bromohexane 3.75
(3.80)
4.34 4.55 2.29 2.39 1.87
fluoranthene 5.00
(5.16)
5.44 5.91 4.22 3.75 3.57
bromooctane 4.82
(4.89)
5.62 5.72 2.89 3.04 2.34
a
Experimental values are in parentheses.
416 Ind. Eng. Chem. Res., Vol. 42, No. 3, 2003
Office of Energy Research, U.S. Department of Energy
(Grant DE-FG02-96ER14673).
Literature Cited
(1) Bradley, D. R. OrganicReactions Enter a New Phase.Chem.
Ind. 1999, 86.
(2) Lo, T. C. Handbook of Separations Techniques for Chemical
Engineers; Schwietzer, P. A., Ed.; McGraw-Hill: New York, 1996.
(3) Rydberg, J.; Musikas, C.; Choppin, G. R. Principles and
Practices of Solvent Extraction; Marcel Dekker: New York, 1992.
(4) Vision 2020, 1998 Separations Roadmap; Center for Waste
Reduction Technologies of the American Institute of Chemical
Engineers (AIChE): New York, 1998.
(5) Li, C. J.; Chan, T. H. Organic Reactions in Aqueous Media;
John Wiley & Sons: New York, 1997.
(6) Poliakoff, M.; Meehan, N. J.; Ross, S. K. A Supercritical
Success Story. Chem. Ind. 1999, 19, 750-752.
(7) Freemantle, M. Designer solventssIonic liquids may boost
clean technology development. Chem. Eng. News 1998, 76 (Mar
30), 32-37.
(8) Freemantle, M. Ionic Liquids Show Promise for Clean
Separation Technology. Chem. Eng. News 1998, 76 (Aug 24), 12.
(9) Dyson, P. J.; Grossel, M. C.; Srinivasan,N.; Vine, T.; Welton,
T.; Williams, D. J.; White, A. J. P.; Zigras, T. Organometallic
synthesis in ambient temperature chloroaluminate(III) ionic
liquids. Ligand exchange reactions of ferrocene. J. Chem. Soc.,
Dalton Trans. 1997, 3465-3469.
(10) Earle, M. J.; McCormac, P.; Seddon, K. R. Regioselective
Alkylation in Ionic Liquids. Chem. Commun. 1998, 2245-2246.
(11) Earle, M. J.; McCormac, P.; Seddon, K. R. Diels-Alder
Reactions in Ionic Liquids. Green Chem. 1999,23-25.
(12) Dai, S.; Ju, Y. H.; Barnes, C. E. Solvent Extraction of
Strontium Nitrate by a Crown Ether Using Room-Temperature
Ionic Liquids. J. Chem. Soc., Dalton Trans. 1999, 1201-1202.
(13) Huddleston, J. G.; Willauer, H. D.; Swatloski, R. P.; Visser,
A. E.; Rogers, R. D. Room-Temperature Ionic Liquids as Novel
Media for ‘Clean’ Liquid-Liquid Extraction. Chem. Commun.
1998, 1765-1766.
(14) Bonhote, P.; Dias, A.-P.; Papageorgiou, N.; Kalyana-
sundaram, K.; Gratzel, M. Hydrophobic, highly conductive ambi-
ent-temperature molten salts. Inorg. Chem. 1996, 35, 1168-1178.
(15) Huddleston, J. G.; Visser, A. E.; Broker, G. A.; Reichert,
W. M.; Willauer, H. D.; Rogers, R. D. Characterization and
Comparison of Hydrophilic and Hydrophobic Room-Temperature
Ionic Liquids Incorporating the Imidazolium Cation. Green Chem.
2001, 4, 156-164.
(16) Carmichael, A. J.; Seddon, K. R. Polarity study of some
1-alkyl-3-methylimidazolium ambient-temperature ionic liquids
with the solvatochromic dye Nile Red. Phys. Org. Chem. 2000, 13,
591-595.
(17) Huddleston, J. G.; Visser, A. E.; Broker, G. A.; Reichert,
W. M.; Willauer, H. D.; Rogers, R. D. Free Energy Relationships
and SolvatochromicProperties of Room-Temperature Ionic Liquids
Based on Methylimmidazolium Cations. In Abstracts of Papers,
Part 2, 221st National Meeting of the American Chemical Society,
San Diego, CA, Apr 1-5, 2001; American Chemical Society:
Washington, DC, 2001; Abstract IEC 278.
(18) Muldoon, M. J.; Gordon, C. M.;Dunkin, I. R. Investigations
of solvent-solute interactions in room-temperature ionic liquids
using solvatochromic dyes. J. Chem. Soc., Perkin. Trans. 2 2001,
433-435.
(19) Park, S.; Kazlauskas, R. J. Improved Preparation and Use
of Room-Temperature Ionic Liquids in Lipase-Catalyzed Enantio-
and RegioselectiveAcylations. J. Org.Chem. 2001, 66, 8395-8401.
(20) Huddleston, J. G.; Broker, G. A.; Willauer, H. D.; Rogers,
R. D. Free Energy Relationships and Solvatochromic Properties
of 1-Alkyl-3-methylimidazolium. In Ionic Liquids; Industrial Ap-
plications for Green Chemistry; Rogers, R. D., Seddon, K. R., Eds.;
ACS Symposium Series 818; American Chemical Society: Wash-
ington, DC, 2002; pp 270-288.
(21) Abraham, M. H.; Whiting, G. S.; Doherty, R. M.; Shuely,
W. J. Hydrogen-bonding. 17. The characterization of 24 gas-
liquid- chromatographic stationary phases studied by Poole and
co-workers including molten salts, and evaluation of solute
stationary phase interactions. J. Chromatogr. 1991, 587, 229-
236.
(22) Kollie, T. O.; Poole, C. F.; Abraham, M. H.; Whiting, G. S.
Comparison of Two Free Energy of Solvation Models for Charac-
terizing Selectivity of Stationary Phases Used in Gas-Liquid
Chromatography. Anal. Chim. Acta 1992, 259,1-13.
(23) Kollie, T. O.; Poole, C. F. Influence of Fluorine Substitution
on the Solvation Properties of Tetraalkylammonium Alkane-
sulfonate Phasesin Gas Chromatography. Chromatographia 1992,
33, 551-559.
(24) Poole, S. K.; Poole, C. F. Application of Principal Compo-
nent Factor Analysis to the Cavity Model of Solvation to Identify
Factors Important in Characterizing the Solvent Properties of Gas
Chromatographic Stationary Phases. J. Chromatogr. A 1995, 697,
429-440.
(25) Poole, S. K.; Poole, C. F. Chemometric Evaluation of the
Solvent Properties of Liquid Organic Salts. Analyst 1995, 120,
289-294.
(26) Abraham, M. H.; Poole, C. F.; Poole, S. K. Classification
of Stationary Phases and Other Materials by Gas Chromatogra-
phy. J. Chromatogr. A 1999, 842,79-114.
(27) Heintz, A.; Kulikov, D. V.; Verevkin, S. P. Thermodynamic
properties of mixtures containing ionic liquids. 1. Activity coef-
ficients at infinite dilution of alkanes, alkenes, and alkylbenzenes
in 4-methyl-n-butylpyridinium tetrafluoroborate using gas-liquid
chromatography. J. Chem. Eng. Data 2001, 46, 1526-1529.
(28) Albertsson, P.-Å. Partition of Cell Particles and Macro-
molecules, 3rd ed.; Wiley: New York, 1986.
(29) Walter, H., Brooks, D. E., Fisher, D., Eds. Partitioning in
Aqueous Two-Phase Systems: Theory, Methods, Uses and Applica-
tions to Biotechnology; Academic Press: New York, 1985.
(30) Zaslavsky, B. Y. AqueousTwo-Phase Partitioning. Physical
Chemistry and Bioanalytical Applications; Marcel Dekker: New
York, 1994.
(31) Aqueous Two-Phase Systems; Walter, H., Johannsson, G.,
Eds.; Methods in Enzymology Vol. 228; Academic Press: New
York, 1994.
(32) Huddleston, J. G.; Griffin, S. T.; Zhang, J.; Willauer, H.
D.; Rogers, R. D. Metal Ion Separations in Aqueous Biphasic
Systems and with ABEC Resins. In Aqueous Two-Phase Systems;
Kaul, R., Ed.; Methods in Biotechnology Series; Humana Press:
Totowa, NJ, 1999.
(33) Huddleston, J. G.; Griffin, S. T.; Willauer, H. D.; Rogers,
R. D. Metal Ion Separations in Aqueous Biphasic Systems and
Using Aqueous Biphasic Extraction Chromatography. In Metal-
Ion Separation and Preconcentration, Progress and Opportunities;
Dietz, M. L., Bond, A. H., Rogers, R. D., Eds.; ACS Symposium
Series 716; American Chemical Society: Washington, DC, 1998;
pp 79-100.
(34) Rogers, R. D.; Zhang, J. New Technologies for Metal Ion
Separations Polyethylene Glycol Based Aqueous Biphasic Systems
and Aqueous Biphasic Extraction Chromatography. In Ion Ex-
change and Solvent Extraction; Marinsky, J. A., Marcus, Y., Eds.;
Marcell Dekker: New York, 1997; Vol. 13, pp 141-193.
(35) Rogers, R. D.; Bond, A. H.; Griffin, S. T.; Horwitz, E. P.
New Technologies for Metal Ion Separations: Aqueous Biphasic
Extraction Chromatography (ABEC): Part 1. Uptake of Pertech-
netate. Solvent Extr. Ion Exch. 1996, 14, 919-946.
(36) Willauer, H. D.; Huddleston, J. G.; Rogers, R. D. Aspects
of the Partition of Small Organic Molecules in Aqueous Biphasic
Systems. Ind. Eng. Chem. Res. 2002, 41, 1892-1904.
(37) Willauer, H. D.; Huddleston, J. G.; Rogers, R. D. Solvent
Properties ofAqueous Biphasic Systems Composed of Polyethylene
Glycol and Salt Characterized by the Free Energy of Transfer of
a Methylene Group Between the Phases and by a Linear Solvation
Energy Relationship. Ind. Eng. Chem. Res. 2002, 41, 2591-2601.
(38) Leininger, N. F. Polyglycols as alternative solvents for
pharmaceutical reactions. Paper presented at the 5th Green
Chemistry andEngineering Conference, Washington, DC, Jun 25-
28, 2001.
(39) Guo, Z.; Li, M.; Willauer, H. D.; Huddleston, J. G.; April,
G. C.; Rogers, R. D. Polymer-Based Aqueous Biphasic Systems as
Alternatives for the Kraft Pulping Process. Ind. Eng. Chem. Res.
2001, 41, 2535-2542.
(40) Guo, Z.; Li, M.; Willauer, H. D.; Huddleston, J. G.; April,
G. C.; Rogers, R. D. Evaluation of Polymer-Based Aqueous
Biphasic Systems as Green Alternatives for Alkaline Pulping
Process. Chem. Eng. Commun. 2001, in press.
(41) Domb, C. The Critical Point: A Historical Introduction to
the Modern Theory of Critical Phenomena; Taylor & Francis:
London, 1996.
Ind. Eng. Chem. Res., Vol. 42, No. 3, 2003 417
