Addition of suitably-designed zwitterions improves the saturated water content of hydrophobic ionic liquids

Article abstract of DOI:10.1039/c2cc36119a

The saturated water content in a hydrophobic ionic liquid, 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)-imide, was improved from 0.4 wt% to 17.8 wt% by adding a 3-(1-butyl-3-imidazolio)propanesulfonate- type zwitterion in appropriate amounts. The mixture containing 17.8 wt% water successfully dissolved horse heart cytochrome c without significant change of the higher ordered structure.

Full text of DOI:10.1039/c2cc36119a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 11220–11222
www.rsc.org/chemcomm
COMMUNICATION
Addition of suitably-designed zwitterions improves the saturated water
content of hydrophobic ionic liquidsw
Yoritsugu Ito, Yuki Kohno, Nobuhumi Nakamura and Hiroyuki Ohno*
Received 23rd August 2012, Accepted 30th September 2012
DOI: 10.1039/c2cc36119a
The saturated water content in a hydrophobic ionic liquid,
1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)-imide,
was improved from 0.4 wt% to 17.8 wt% by adding a 3-(1-butyl-
3-imidazolio)propanesulfonate-type zwitterion in appropriate
amounts. The mixture containing 17.8 wt% water successfully
dissolved horse heart cytochrome c without significant change of
the higher ordered structure.
control of the water content of hydrophobic ILs has not been
reported. This is because hydrophobic ILs are characterised by
their poor miscibility with water. To control the water content
of hydrophobic ILs, component ions could be designed with
suitable hydrophilicity.14,15 It is difficult to tune the desired
amount of saturated water in the ILs accurately by this method,
however. Other methods include simple mixing with some
hydrophilic ILs. Mixing two or more ILs is a much easier way
to fine-tune the saturated water content of ILs than structural
design of the ILs. Unfortunately the possibility of ion exchange
among the mixed ions could give rise to unexpected changes in
the physico-chemical properties of these ILs.
Ionic liquids (ILs) are organic molten salts with very low melting
temperature.1–3 They are capable of exhibiting a wide variety of
liquid properties including polarity,4 viscosity,5 and hydrophobicity6
by appropriate combination of cation and anion species. Some
hydrophobic ILs readily undergo liquid–liquid phase separation
after mixing with water. These hydrophobic IL–water biphasic
systems have been used in novel energy-saving processes
for the extraction, separation, and condensation of various
compounds. As an example, the dissolving and stabilising of
proteins in hydrophobic ILs would enable these ILs to act as media
for enzymatic reactions and/or product separation. It would be
convenient to separate the product after transfer to an aqueous
phase following an enzymatic reaction in the hydrophobic IL phase.
Most enzymes were partitioned to an aqueous phase, however,
because enzymes are scarcely soluble in hydrophobic ILs. There are
ways to dissolve proteins in hydrophobic ILs, such as modification
of the proteins with amphiphilic polymers,7,8 addition of crown
ethers,9 and formation of aqueous microemulsion droplets in a
hydrophobic IL.10
Here, we propose zwitterions (ZIs) as additives suitable for
controlling the water content of hydrophobic ILs. ZIs should
prevent ion exchange in the IL mixtures, because their component
cations and anions are both covalently tethered. In the present
study, we investigated the effect of the added ZI on the saturated
water content of a hydrophobic IL. The solubility and stability of
proteins in the hydrophobic IL–ZI–water mixture were also
analysed. Cytochrome c (cyt.c) from horse heart was chosen as
a model protein, since it is a typical heme protein which has been
intensively studied.
We synthesised four ZIs comprising alkylimidazolium
cations and sulfonate anions, as shown in Scheme 1. The
synthesis of these ZIs is summarised in ESI.w The resulting ZIs
were all obtained as white solids. All ZIs used in this study
were soluble in water. Solubility of these ZIs in water depends
on their structure.
Hydrophobic ILs would be more useful if more compounds
could be dissolved in them. The solubility of water-soluble
compounds (including proteins) in ILs is influenced by the
water content. In the case of hydrophilic ILs it is easy to control
the water content of hydrophilic ILs, simply by adding water.
Our group has found that some hydrophilic ILs containing a
small amount of water, classified as ‘‘hydrated ILs’’, were
capable of dissolving many kinds of proteins.11,12 Furthermore,
the dissolved proteins remained in their native state in the
hydrated ILs for at least one year.13 Unfortunately, systematic
We chose 1-butyl-3-methylimidazolium bis(trifluoromethane-
sulfonyl)imide ([C4mim][Tf2N]) as a model hydrophobic IL,
because it has very low saturated water content (0.4 wt%). To
analyse the miscibility between [C4mim][Tf2N] and the ZIs
prepared, an equimolar ZI was mixed with [C4mim][Tf2N], and
the resulting solutions were stirred vigorously. None of the added
Department of Biotechnology, Tokyo University of Agriculture and
Technology, 2-24-16, Naka-cho, Koganei, Tokyo, 184-8588, Japan.
E-mail: ohnoh@cc.tuat.ac.jp; Fax: +81 42-388-8588;
Tel: +81 42-388-7024
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c2cc36119a
Scheme 1 Structure of zwitterions used in this study.
c
11220 Chem. Commun., 2012, 48, 11220–11222
This journal is The Royal Society of Chemistry 2012

View Article Online
ZIs were fully soluble in the IL, and they underwent solid–liquid
type phase separation. Differing amounts of water were then
added to the solution, which were stirred. We analysed the
miscibility and phase behaviour of the resulting mixtures. When
the ZIs 3-(1-methyl-3-imidazolio)propanesulfonate (C1Im3S),
3-(1-ethyl-3-imidazolio)propanesulfonate (C2Im3S), 3-(1,2-
dimethyl-3-imidazolio)propanesulfonate (C1,1Im3S), and 3-(1-
butyl-3-imidazolio)propanesulfonate (C4Im3S) were individually
mixed with the [C4mim][Tf2N], no homogeneous phase was
found, and a solid–liquid type phase separation prevailed. No
homogeneous mixtures were obtained even after adding water
to these mixtures except for C4Im3S. In all cases, when excess
water was added to the mixture, they underwent liquid–liquid
phase separation. In the case of mixing with C4Im3S, a clear
homogeneous phase was obtained by mixing a suitable amount
of water (the molar ratio of C4Im3S to water molecules varied
from 1 : 3 to 1 : 8). These results suggested that suitable
hydrophilicity of ZIs is required to improve the saturated water
content of hydrophobic ILs. Both C1Im3S and C2Im3S were
rather hydrophilic and showed low solubility in the hydro-
phobic IL, resulting in the obvious phase separation even after
mixing with a small amount of water. Since C4Im3S proved to
be the ZI most compatible with [C4mim][Tf2N], we chose
C4Im3S as an additive to control the saturated water content
of the hydrophobic IL. C4Im3S containing 8-fold molar water
was further added to the [C4mim][Tf2N] in different mixing
ratios, and the water content of the mixture was analysed by
the Karl–Fischer titration method. The water content of
[C4mim][Tf2N] increased with the amount of the C4Im3S–water
mixture (Fig. 1). This result suggests that the saturated water
content of hydrophobic ILs can be controlled by adding the
ZI–water mixture. As seen in Fig. 1, addition of C4Im3S
improves the saturated water content of [C4mim][Tf2N] from
0.4 wt% to 17.8 wt%.
Fig. 2 Photograph of cyt.c mixed with [C4mim][Tf2N]–C4Im3S–water
mixture (left) and [C4mim][Tf2N] (right).
The [C4mim][Tf2N]–C4Im3S–water mixture, in a ratio of
1 : 1 : 8, was used. We found that cyt.c was soluble (3 mg mlÀ1
)
in the mixture. Fig. 2 shows a photograph of the cyt.c in the
mixture, with and without the C4Im3S–water mixture. Since cyt.c
did not dissolve in the pure [C4mim][Tf2N], it is confirmed visually
that the C4Im3S–water mixture facilitated the dissolution of cyt.c
in the [C4mim][Tf2N]. Resonance Raman (RR) spectroscopy of
cyt.c in the mixture was then undertaken in order to determine the
environment of the heme vicinity. Several relevant modes observed
in the RR spectra are sensitive both to axial coordination and the
spin state of the iron ion of heme.16,17 In the RR spectrum of cyt.c
in buffer solution at pH 7.0, the n4 band at around 1372 cmÀ1
represents the valency of the heme iron. The n2 (1586 cm–1) and
n3 (1502 cmÀ1) bands are both helpful in determining the
coordination state and the spin state, and these bands indicate
a six-coordinated low spin state. Fig. 3 shows the RR spectra
of cyt.c in phosphate buffer (a, black line), cyt.c in the
[C4mim][Tf2N]–C4Im3S–water mixture (b, red line), and blank
solution without cyt.c (c, blue line). The RR spectrum of cyt.c in
the [C4mim][Tf2N]–C4Im3S–water mixture included intense
bands at 1372 (n4), 1502 (n3), 1586 (n2), and 1634 cmÀ1 (n10),
which indicates that the central iron ion of the heme complex is
in oxidized form and in a six-coordinated low spin state
(Fig. 3b). These characteristic bands are similar to those for
cyt.c in buffer solution, suggesting that the dissolution of cyt.c
in the [C4mim][Tf2N]–C4Im3S–water mixture did not give rise
to a drastic change in the heme vicinity. Both solubility and
The addition of water to ILs improves the solubility of
water-soluble compounds, including proteins such as cyt.c. In the
case of hydrophobic ILs we did not attempt to evaluate the effect
of water content on the solubility of cyt.c, because of the limited
water content in the hydrophobic ILs. We examined the
solubility of cyt.c in our [C4mim][Tf2N]–C4Im3S–water mixture.
Fig. 1 Relation between water content of [C4mim][Tf2N] and frac-
tion of C4Im3S in [C4mim][Tf2N]–C4Im3S mixtures. The molar ratio
of [C4mim][Tf2N] to C4Im3S in [C4mim][Tf2N]–C4Im3S–water is
shown in the upper axis. The C4Im3S–water ratio was 1 : 8 in all cases.
Fig. 3 Resonance Raman spectra of cyt.c in phosphate buffer (black
line, a), cyt.c in [C4mim][Tf2N]–C4Im3S–water mixture (red line, b),
and blank solution without cyt.c (blue line, c).
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 11220–11222 11221

View Article Online
C4Im3S dissolved homogeneously in [C4mim][Tf2N] without
disruption of the hydration state.
There was an obvious liquid–liquid phase separation
between the IL phase and aqueous ZI phase upon adding
further water to the mixture (>1 : 8). We determined the
water content of the separated IL phase of the mixture
(the molar ratio of C4Im3S, [C4mim][Tf2N], and water was
1 : 1 : 9) by the Karl–Fischer titration method. The water
content was 1.8 wt%, indicating that most C4Im3S had
transferred to the aqueous phase. After the phase separation,
we found that all cyt.c had transferred from the IL phase to
the aqueous ZI phase. This result clearly shows that the
recycling of [C4mim][Tf2N] and the re-extraction of cyt.c were
both facilitated by simple addition of excess water. This is the
basis for a novel reaction–extraction process for bio-related
polymers.
Fig. 4 DSC profiles of C4Im3S after mixing with differing amounts
of water.
In conclusion, we have demonstrated that the saturated
water content of hydrophobic [C4mim][Tf2N] is controlled
accurately up to 17.8 wt% by mixing with C4Im3S in suitable
proportions. The drastic change in the saturated water content
of the hydrophobic [C4mim][Tf2N] involves water molecules
bound strongly to the C4Im3S (8 water molecules per C4Im3S).
This study was supported by a Grant-in-Aid for Scientific
Research from the Japan Society for the Promotion of Science
(No. 21225007).
stabilisation of cyt.c in [C4mim][Tf2N] are accomplished by
adding the C4Im3S–water mixture.
Since hydrophobic [C4mim][Tf2N] has no capability to
dissolve cyt.c, it is plausible that the hydration state of
C4Im3S affects the stable dissolution of cyt.c. Nikawa et al.
analysed the hydrated state of choline dihydrogenphosphate-
type ILs by means of differential scanning calorimetric (DSC)
measurement, and found that the superior hydrated state of the
IL facilitated the stable dissolution of proteins.18 Accordingly,
we also analysed the hydration state of C4Im3S after mixing
with different amounts of water. Fig. 4 shows typical DSC
heating curves of C4Im3S after mixing with differing amounts
of water (the heating rate was 5 1C minÀ1). C4Im3S formed a
homogeneous solution when more than 6 water molecules were
added per C4Im3S molecule. In DSC heat-flow profiles of the
mixture (the mole fraction of C4Im3S to water was 1 : 6), there
was no endothermic peak due to melting of water. This implies
that the added water in the solution exists as non-freezing water
which interacted strongly with C4Im3S. In the case of the
C4Im3S–water mixture with a molar ratio of 1 : 8, both
exothermic and endothermic peaks were observed, at around
À50 1C and À20 1C, respectively. These peaks are characteristic
of freezing bound water, which is often seen in biocompatible
polymers.19,20 In the case of C4Im3S–water mixtures with a
molar ratio exceeding 1 : 10, the exothermic peak disappeared
completely, and only an endothermic peak was observed. This
endothermic peak is known to be derived from ‘‘free’’ water
molecules that do not interact strongly with the ions. These
results indicate that C4Im3S forms a ‘‘hydrated ZI’’, in which
water molecules interact strongly with C4Im3S when 6 to
8 water molecules are added for each C4Im3S molecule. The
[C4mim][Tf2N]–C4Im3S–water mixture formed a clear homo-
geneous solution. This result suggests that the hydration state
of C4Im3S would be maintained in the hydrophobic
[C4mim][Tf2N]. To check this, we measured the DSC of the
hydrated C4Im3S–[C4mim][Tf2N] mixture (the molar ratio
of C4Im3S, [C4mim][Tf2N], and water was 1 : 1 : 8). The
endothermic peak derived from freezing bound water was also
observed in this mixture (see ESI,w Fig. S1). It is apparent that
Notes and references
1 J. S. Wilkes and M. J. Zaworotko, J. Chem. Soc., Chem. Commun.,
1992, 965.
2 T. Welton, Chem. Rev., 1999, 99, 2071.
3 P. Wasserscheid and T. Welton, Ionic Liquids in Synthesis, Wiley-
VCH, 2003.
4 L. Crowhurst, P. R. Mawdsley, J. M. Perez-Arlandis, P. A. Salter
and T. Welton, Phys. Chem. Chem. Phys., 2003, 5, 2790.
5 P. Bonhote, A. P. Dias, N. Papageorgiou, K. Kalyanasundaram
¨
¨
and M. Gratzel, Inorg. Chem., 1996, 35, 1168.
6 K. R. Seddon, A. Stark and M. J. Torres, Pure Appl. Chem., 2000,
72, 2275.
7 N. V. Kawahara and H. Ohno, Solid State Ionics, 1998, 114, 161.
8 K. Nakashima, T. Maruyama, N. Kamiya and M. Goto, Chem.
Commun., 2005, 4297.
9 K. Shimojo, H. Nakagawa, N. Kamiya and M. Goto, Biomacro-
molecules, 2006, 7, 2.
10 M. Moniruzzaman, N. Kamiya, K. Nakashima and M. Goto,
Green Chem., 2008, 10, 497.
11 K. Fujita, D. R. MacFarlane and M. Forsyth, Chem. Commun.,
2005, 4804.
12 K. Fujita and H. Ohno, Biopolymers, 2010, 93, 1093.
13 K. Fujita, D. R. MacFarlane, M. Forsyth, M. Yoshizawa-Fujita,
K. Murata, N. Nakamura and H. Ohno, Biomacromolecules, 2007,
8, 2080.
14 Y. Kohno and H. Ohno, Phys. Chem. Chem. Phys., 2012, 14, 5063.
15 Y. Kohno and H. Ohno, Chem. Commun., 2012, 48, 7119.
16 Cytochrome c: A Multidisciplinary Approach, ed. R. A. Scott and
A. G. Mauk, University Science Books, Sausalito, CA, 1996.
17 C. Y. Huang, G. Balakrishnan and T. G. Spiro, J. Raman Spectrosc.,
2006, 37, 277.
18 Y. Nikawa, K. Fujita, H. Ohno, 4th Internat. Congr. Ionic Liquids,
2011, 222, abstr.
19 T. Hatakeyama, M. Tanaka and H. Hatakeyama, Acta Biomater.,
2010, 2077.
20 M. Tanaka, T. Motomura, N. Ishii, K. Shimura, M. Onishi,
A. Mochizuki and T. Hatakeyama, Polym. Int., 2000, 49, 1709.
c
11222 Chem. Commun., 2012, 48, 11220–11222
This journal is The Royal Society of Chemistry 2012