Microemulsion-like aggregation behaviour of an LCST-type ionic liquid in water

Abstract

The ionic liquid (IL), tetrabutylphosphonium trifluoroacetate ([P₄₄₄₄][CF₃COO]) showed a low critical solution temperature (LCST)-type phase transition in water. It was found that [P₄₄₄₄][CF₃COO] molecules can form some kind of long-living aggregates in aqueous solution under certain conditions before the phase separation. These aggregates displayed the characteristic properties of microemulsions, although no surfactants are used. For instance, the aggregate droplet size can be adjusted by temperature and concentration, which behaves like the swelling phenomenon of microemulsions; the formed aggregates showed beneficial solubilization capacity for apolar substances; in addition, a tunable micropolarity in the aggregates i̲s̲ o̲b̲s̲e̲r̲v̲e̲d̲ b̲y̲ t̲h̲e̲ d̲e̲t̲e̲c̲t̲i̲o̲n̲ o̲f̲ U̲V̲-Vis measurements. These are by far the simplest aggregates having the microemulsion characteristics. In nature, these microemulsion-like aggregates have mesoscopic phase separation, which is the intermediate state for macroscopic phase separation. This special system can be regarded as surfactant-free microemulsion-like aggregates and should be an effective platform to provide novel extraction or separation media.

---

Introduction

It is well known that microemulsions are clear, thermodynamically stable, isotropic liquid mixtures of oil, water and surfactant, sometimes with the aid of a co-surfactant. In these systems, dispersions of two immiscible liquids are stabilized by a surfactant. These dispersion systems have been extensively applied in many fields, such as in chemical reactions, separation science, oil recovery, material synthesis, and in the pharmaceutical industry.¹ In many cases, the use of a surfactant is only helpful during a special stage of a process, after which these surface active substances become a liability and block the separation of the components. The problem of how to destroy these microemulsion systems has not been resolved satisfactorily so far. Switchable surfactants could be used to solve this problem, but they still have several drawbacks, such as highly economic and environmental costs, possible contamination by the addition of these reagents and so on.²

Ionic liquids (ILs) are a class of tunable, designer solvents with essentially zero volatility and have been used in synthesis and catalytic reactions. We,³ and other researchers,⁴ have recently focused on using ILs to create microemulsions with various surfactants. These novel dispersions would provide hydrophobic or hydrophilic nanodomains that could expand the potential uses of ILs as reaction or separation media. Han and his co-workers were the first to find that 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim][BF₄]) assembles into small polar droplets when [bmim][BF₄] is dispersed in cyclohexane with the aid of Triton X-100.⁵ Later, they discovered that 1,1,3,3-tetramethylguanidinium type ILs can form reverse micelles in supercritical CO₂.⁶ They also reported that 1-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF₆]) can form microemulsion in another type of IL, propylammonium formate or ethylene glycol.^7,8 Based on Han's seminal work, Eastoe et al. used small-angle neutron scattering (SANS) to investigate the same microemulsion and observed a swelling behavior similar to traditional microemulsions.⁹ In addition, Sarkar and co-workers systematically investigated the solvent and rotational relaxation of coumarin fluorescence probes confined in various IL-containing aggregates.¹⁰ An environmentally friendly microemulsion was also created with water in which [bmim][PF₆] was used to substitute for organic solvents.^3c,11 These IL-containing assemblies have shown advantages over the traditional ones in different fields. For example, a recent study has found that IL-based microemulsions are good media in which to produce polymer nanoparticles, gels, open-cell porous materials and silica microrods.^12,13 Moreover, the IL microemulsions were also applied to organic and enzymatic reactions,^14,15 as well as bio-extraction.¹⁶

Despite these great advantages, an unusual feature of most of the IL-containing microemulsions is the necessity to include a large amount of surfactant,^4b,9 which may limit their applications in many practical processes. Therefore, it is of interest to create a surfactant-free microemulsion with ILs as substitutes for oil or water. In respect of surfactant-free microemulsions, Smith et al. demonstrated the first surfactant-free microemulsions consisting of n-hexane, isopropyl alcohol and water, in which the water droplets were stabilized by alcohol molecules absorbed on their surfaces.¹⁷ Moreover, effects of KOH and NaCl on the phase behavior were further investigated by Holt and his co-workers.¹⁸ These microemulsions represented the thermodynamically stable and optically transparent dispersions of aqueous microdroplets in a hydrocarbon. Recently, Hou et al. reported that surfactant-free microemulsions can be formed in the presence of benzene,¹⁹ or oleic acid²⁰ with the aid of alcohols. However, a common problem for these systems is that a hydrocarbon is inevitably used.

In this report, we describe a surfactant-free microemulsion-like system that comprises only an ionic liquid tetrabutylphosphonium trifluoroacetate ([P₄₄₄₄][CF₃COO]) and water. Certain compositions of the [P₄₄₄₄][CF₃COO]–water binary system display the physical characteristics of microemulsions, such as swelling behavior, solubilization capacity and tunable micropolarity. The use of surfactants is avoided that would bring significant economic and environmental benefits. Moreover, no volatile organic compounds (VOCs) are used in the system – a substantial environmental advantage. To the best of our knowledge, this is by far the simplest aggregate system that has the characteristic properties of microemulsions. The microemulsion-like aggregates were characterized by means of dynamic light scattering (DLS), freeze-fracture transmission electron microscopy (FF-TEM), UV-Vis absorption spectroscopy and ¹H NMR spectroscopy. The present [P₄₄₄₄][CF₃COO]–water binary system should be an effective platform to provide the novel extraction or separation material. An improvement in the basic understanding of the system can contribute to the establishment of IL-based microemulsions as potential media.

Experimental section

Materials and supplies

Tetrabutylphosphonium hydroxide ([P₄₄₄₄][OH]) was obtained from Hokko Chem. Co. All other solvents and reagents are commercially available, analytical grade. Water was doubly deionized and distilled. The ionic liquid [P₄₄₄₄][CF₃COO] was prepared according to the literature.²¹ Typically, an aqueous solution of tetrabutylphosphonium hydroxide ([P₄₄₄₄][OH]) was first neutralized by trifluoroacetic acid. Water was then removed by using a rotary evaporator. The crude product was extracted with dichloromethane, and the dichloromethane was collected and washed with water several times. After removal of the solvent, the product was dried under reduced pressure at 343 K for 24 h. The purity of the product was confirmed by ¹H NMR and ¹³C NMR spectrometry. The spectra accord with those reported in the literature.²¹ The size and size distributions of the investigated microemulsions were determined by dynamic light scattering (DLS) using Zetasizer Nano S90 (ZEN1690) with a He–Ne laser operating at 633 nm. All measurements were made at a scattering angle of 90°. ¹H NMR measurements were carried out with a Bruker DXR-400 NMR spectrometer. The instrument was operated at a frequency of 400.23 MHz using tetramethylsilane (TMS) as an internal reference to confirm the purity of the IL. A co-axial NMR system was used for the variable-temperature NMR experiment, where the [P₄₄₄₄][CF₃COO]–D₂O mixture is contained in the inner compartment, while a very dilute solution of sodium 2,2-dimethylsilapentane-5-sulphonate (DSS) in D₂O is stored in the outer tube. DSS was used as an internal reference. The spectrometer was fitted with a 5 mm CPTXI-1H-13C/15N/2H probe head with z-gradients. UV-Vis absorption spectra were taken at room temperature on a UV-2500 (SHIMADZU) spectrometer. Freeze-fracture transmission electron microscopy (FF-TEM) observations on the replicated samples were carried out with a JEOL TEM 200CX electron microscope. The replicas were first prepared as follows: a small amount of sample was placed in a gold cup. The temperature of the sample was controlled to a desired temperature before the preparation of sample replicas. The gold cup was then swiftly plunged into a liquid Freon that has been cooled with liquid nitrogen in advance. The frozen samples were fractured and replicated in a freeze-fracture apparatus BAF 400 (Bal-Tec, Balzer, Liechtenstein) at 133 K. Pt/C was deposited at an angle of 45°.

Results and discussion

Phase behavior of the [P₄₄₄₄][CF₃COO]–water binary system

The ionic liquid, [P₄₄₄₄][CF₃COO], has been reported to exhibit a temperature-sensitive phase behavior in water, in contrast to most substances whose solubility increases with increasing temperature, [P₄₄₄₄]CF₃COO is miscible with water at low temperature but becomes immiscible when the temperature is increased.²¹ This exceptional phase behavior is classified as a lower critical solution temperature (LCST)-type phase transition.^21,22 Fig. 1 shows an example of the LCST-type phase transition of [P₄₄₄₄][CF₃COO]–H₂O binary mixture. A 40 wt% aqueous solution showed a homogeneous status at 298 K, and the solution was obviously phase separated into an upper water-rich phase and a lower IL-rich phase at 308 K or above. The two phases became homogeneous again when temperature was decreased down to 298 K. This means that such phase change was reversibly induced only by a small temperature change. Fig. 2 shows the phase transition temperature (T_c) of the [P₄₄₄₄][CF₃COO]–H₂O mixture in the full mixing range. T_c is found to decrease and then increase with increasing [P₄₄₄₄][CF₃COO] content. A similar phenomenon was also observed for other LCST-type ILs,²¹ but not all ILs have such LCST behaviour. Ohno et al. deduced that a LCST-type phase transition is only possible for those ILs whose hydrophilicity falls in a particular range.²³

Fig. 1 Temperature-sensitive LCST-type phase transition of a [P₄₄₄₄][CF₃COO]–water mixture consisting of 40 wt% [P₄₄₄₄][CF₃COO]. The bottom [P₄₄₄₄][CF₃COO]-rich phase was colored by red oil.

Fig. 2 Phase separation temperature (LCST) of [P₄₄₄₄][CF₃COO] after mixing with different amounts of water.

Size and morphology of the aggregates

In the course of our study for the necessary factors for LCST-type phase transition of IL–water mixtures, we unexpectedly found that [P₄₄₄₄][CF₃COO] molecules can form some kinds of aggregates in an aqueous solution of certain concentrations when the temperature was below the LCST. This was identified by the dynamic light scattering (DLS) measurements. DLS is known to be a powerful technique to detect inhomogeneities at the mesoscopic scale. Importantly, the aggregates are stable as no obvious phase separation takes place even after 2 weeks' standing. Because [P₄₄₄₄][CF₃COO] is not a surfactive active substance, we can exclude the possibility of amphiphilic assembly formation. This unusual phenomenon compels us to probe into the nature of these long-living aggregates. To this end, the aggregation behavior was further investigated by DLS. A binary system consisting of [P₄₄₄₄][CF₃COO]–water (4 : 6, w/w) was first chosen as an example. The critical temperature of IL–water (4 : 6, w/w) mixture was determined to be 303 K, which compares closely with the value observed by Ohno et al. of 302 K.²¹ Fig. 3 shows the average size and size distributions of the [P₄₄₄₄][CF₃COO] aggregates ([P₄₄₄₄][CF₃COO]–water = 4 : 6, w/w) at different temperatures. The size of the aggregates increases with the increase of temperature, concomitantly with an increase in the breadth of the size distribution. Since the phase separation can be controlled by temperature, it is necessary to clarify the relationship between the structure of the aggregates and temperature. The insert of Fig. 3 shows the variation of the average size of [P₄₄₄₄][CF₃COO] aggregates in relation to temperature. It is easy to see that the size increases more quickly when the temperature approaches the critical point, i.e. LCST of [P₄₄₄₄][CF₃COO] in water (4 : 6, w/w). After that, a further increase in temperature beyond the LCST will lead to the phase separation. The whole process is reversible, which means that the aggregate size can be tuned by temperature. From this result, we primarily suppose that the phase separation of the [P₄₄₄₄][CF₃COO]–water mixture may undergo a transition from molecular solution or mesoscopic phase separation to a macroscopically heterogeneous mixture, i.e. macroscopic phase separation, passing through a stable intermediate process in which there exits a microscopically heterogeneous structure in the solution.²⁴ In other words, [P₄₄₄₄][CF₃COO] molecules may form some long-living aggregates in water before the phase separation.

Fig. 3 Size and size distribution of [P₄₄₄₄][CF₃COO] aggregates in water (4 : 6, w/w) at different temperatures; ■: 283 K, ●: 288 K, ▲: 293 K, ▼: 298 K, and ◀: 300 K.

Since the phase transition behaviour of the [P₄₄₄₄][CF₃COO]–water mixture is also concentration-dependent, as revealed in Fig. 2, the temperature dependence of the average size of [P₄₄₄₄][CF₃COO] aggregates at different [P₄₄₄₄][CF₃COO] concentrations was also investigated. The variation of average aggregate size with temperature at various IL contents is shown in Fig. 4. All systems demonstrated a similar variation tendency, that is to say, whatever the concentration is, the aggregate size will increase gradually with increasing temperature, followed by a sharp increase as the temperature approaches the phase transition temperature, and the phase separation will eventually happen once the temperature reaches the LCST. This result further suggests that the aggregates are an intermediate state for the macroscopic phase separation. Besides, it is clear that at the same temperature, the concentration of [P₄₄₄₄][CF₃COO] had a remarkable effect on the aggregate size. For example, when the temperature was fixed at 298 K, upon increasing the concentration from 20 to 25 and then 40 wt%, the aggregate size was increased from 5.3 to 10.7 and further to 30.5 nm. As the concentration was increased up to 50 wt%, the size of aggregates was 31.0 nm, close to that of the 40 wt% solution. However, with further increasing the concentration up to 60 wt%, the size decreased to 19.7 nm. We ascribe this phenomenon to the role diversion of [P₄₄₄₄][CF₃COO] from solute to solvent with water. In the case where the size increases with the concentration of [P₄₄₄₄][CF₃COO], the water content is more than that of the [P₄₄₄₄][CF₃COO], the [P₄₄₄₄][CF₃COO] molecules form self-assembled aggregates in water. We take it for granted that big aggregates are prone to appear in a higher-concentration solutions. In the case where the size decreases with further increases of the concentration, water is dissolved in [P₄₄₄₄][CF₃COO] and forms liquid-like associated aggregates, which have been observed in hygroscopic ILs with anions of stronger basicity including [CF₃COO].^25,26 At this stage, as the [P₄₄₄₄][CF₃COO] concentration is increased, a fraction of water is dissolved in the surrounding [P₄₄₄₄][CF₃COO] phase as trapped water,²⁷ which will lead to the decrease in aggregate size. Therefore, these aggregates are both temperature and concentration-dependent.

Fig. 4 Variation of the size of [P₄₄₄₄][CF₃COO] aggregates with temperature in solutions containing different amounts of [P₄₄₄₄][CF₃COO].

It is known that conductivity measurements can give useful information for microemulsions and the microstructure can be determined by this method.¹⁵ Fig. 5 shows the conductivity of the [P₄₄₄₄][CF₃COO]–water system as a function of temperature. It can be seen that the conductivity of the [P₄₄₄₄][CF₃COO]–water (1 : 4, w/w) increased with increasing temperature but a turning point was observed when the temperature was 299 K, which is similar to the transition point in the DLS data (a turning point can be clearly observed at 298 K if the DLS curve of the system is shown separately). A similar result was also obtained for the [P₄₄₄₄][CF₃COO]–water (1 : 2, w/w) system. After the turning point, it is clear that the slope of the increasing trend decreases with temperature. We relate this phenomenon to the transition of dissociated [P₄₄₄₄] and [CF₃COO] in water into the [P₄₄₄₄][CF₃COO] ionic liquid as it is well known that the conductivity of pure ionic liquids is generally lower than their aqueous solution.

Fig. 5 Variation of the logarithm of the conductivity value of the [P₄₄₄₄][CF₃COO]–water (1 : 4, w/w) system as a function of temperature.

Freeze-fracture transmission electron microscopy (FF-TEM) can provide a direct image of the aggregates and their morphology in a liquid sample. The morphology of the [P₄₄₄₄][CF₃COO] aggregates in water was thus examined with FF-TEM. No self-assembled structure is observable in the FF-TEM images when the [P₄₄₄₄][CF₃COO] content is less than 5 wt% (therefore these pictures are not shown). This is almost in accordance with our DLS measurements, thus confirming that a molecular solution does exist at that stage. Aggregates can form only when the [P₄₄₄₄][CF₃COO] concentration reaches a certain extent. Fig. 6a provides a representative image for the [P₄₄₄₄][CF₃COO]–water (40 wt%) system at 283 K, which clearly shows a spherical structure. The droplet size distribution is quite narrow and the mean droplet diameter is about 10.0 nm. However, when the temperature is increased to 298 K, the morphology of the aggregates remains spherical, but the size is increased to ca. 30.0 nm (Fig. 6b), in accordance with the DLS measurements. As the temperature is further increased to 302 K, a tendency toward aggregation of the small droplets is observed (Fig. 6c). Such aggregation of small droplets into a cluster structure was reported to be a universal phenomenon in ionic liquid-containing microemulsions.^3b Thus, the aggregates not only show the “swelling” behavior of traditional microemulsions but also have a unique behavior of ionic liquid microemulsions.

Fig. 6 FF-TEM images of [P₄₄₄₄][CF₃COO] aggregates in water (4 : 6, w/w) at 283 K (top-left), 298 K (top-right), and 302 K (bottom).

Solubilization capacity of the aggregates

The LCST behaviour of the IL–water binary mixture has been utilised to extract protein cytochrome C (cyto C), the use of the phase separation considerably accelerated reaching the equilibrium.²⁸ The reason may be due to the appearance of small aggregate droplets that have a large interface area and thus accelerate the rate of extraction. We then investigated if these aggregates possess the solubilization capacity, like microemulsions. Toluene is soluble in [P₄₄₄₄][CF₃COO] but immiscible with water. Thus, the solubilization of toluene in the aggregates was studied in this section. As expected, no phase separation was observed when a small amount of toluene was added to the [P₄₄₄₄][CF₃COO]–water (4 : 6, w/w) system, suggesting that toluene can be solubilized into the aggregates. Therefore, it can be suggested that the [P₄₄₄₄][CF₃COO] aggregates demonstrated a similar solubilization behavior to microemulsions.

It has been reported that a dramatic increase in the number of water molecules per ion pair took place in the LCST-type phase separation.²¹ Keeping this in mind, we assume that the microenvironment of the [P₄₄₄₄][CF₃COO] aggregate droplets will be affected greatly when organic solvents are added to the system. For this reason, the effect of toluene on the average size of the [P₄₄₄₄][CF₃COO] aggregates at different temperatures was studied. Fig. 7 shows the temperature dependence of the average size of the [P₄₄₄₄][CF₃COO] aggregates (P₄₄₄₄][CF₃COO]–water = 4 : 6, w/w) with various toluene contents. First, we found that the aggregate size increased slowly with temperature, followed by a dramatic increase when the temperature approaches the critical temperature, which is similar to the system without toluene. This means that the aggregate structure may not be destroyed by the addition of toluene. In addition, under the same temperature conditions, the addition of toluene was found to increase the aggregate size. The possible reason is that toluene provided an apolar environment for [P₄₄₄₄][CF₃COO] molecules and promoted the aggregation of these molecules. To confirm this, another apolar substance, chloroform, was used to detect the solubilization behavior and structural change of the aggregates. The same result was obtained when chloroform was used instead of toluene (not shown here), further indicating that the aggregates have the solubilization capability for apolar substances.

Fig. 7 Temperature dependence of the average size of [P₄₄₄₄][CF₃COO] aggregates in water (4 : 6, w/w) without and with different toluene contents (the weight percent shown in the figure is relative to the weight of the microemulsion).

Micropolarity of aggregates

In order to get more information about the aggregates in the [P₄₄₄₄][CF₃COO]–water (4 : 6, w/w) mixture, UV-Vis spectroscopy combined with methyl orange as an absorption probe was used to detect the micropolarity of aggregates. We found that methyl orange preferred to locate in the lower [P₄₄₄₄][CF₃COO]-rich phase rather than in the upper water-rich phase after the phase separation. It is thus deduced that methyl orange will stay in the [P₄₄₄₄][CF₃COO] aggregates but not in the surrounding water phase once the aggregates are formed by decreasing the temperature. So, the interior polarity of such a droplet can be reflected by the absorption maximum, λ_max, of the UV/Vis spectroscopy. Fig. 8 shows the variation of λ_max of methyl orange in the [P₄₄₄₄][CF₃COO]–water mixture with temperature. The λ_max decreases from 443.2 to 432.1 nm as the temperature is increased from 278 to 298 K, showing that the micropolarity of the [P₄₄₄₄][CF₃COO] droplets is obviously decreased. We also determined the polarity of pure water and [P₄₄₄₄][CF₃COO], respectively, at different temperatures. It was found that the λ_max of methyl orange in pure [P₄₄₄₄][CF₃COO] was 425.2 nm at 283 K and shifted to 424.5 and 423.9 nm when temperature was increased to 293 K and 303 K, respectively (a total shift of 1.3 nm). In the case of pure water, the λ_max of methyl orange is 464.2 nm at 283 K and moved to 462.7 nm at 303 K (1.5 nm shift). Obviously, the λ_max of methyl orange in both pure [P₄₄₄₄][CF₃COO] and water changed with temperature, but not as much as in the [P₄₄₄₄][CF₃COO]–water mixture, showing that the decrease in the micropolarity of [P₄₄₄₄][CF₃COO] aggregates can be attributed to the change of microenvironment. The micropolarity of traditional microemulsion droplets is known to be dependent on the loaded amount of the dispersed phase. For example, in a water-in-oil microemulsion the polarity of water droplets increases with increasing water content.²⁷ From this point of view, we believe that the [P₄₄₄₄][CF₃COO] aggregates have an adjustable micropolarity like traditional microemulsions. However, the difference is that the decrease in the micropolarity of [P₄₄₄₄][CF₃COO] aggregates is due to the decreased number of water molecules per ion pair, that is to say, water molecules were squeezed out of the aggregates on heating.²¹ It is noteworthy that the micropolarity of the [P₄₄₄₄][CF₃COO] aggregates can be adjusted reversibly by only changing the temperature. This is a great advantage over the traditional microemulsions where the micropolarity of the microemulsion droplets can only be changed through the irreversible addition of the dispersed phase.

Fig. 8 The λ_max of methyl orange in the [P₄₄₄₄][CF₃COO] aggregates in water (4 : 6, w/w) dependence on temperature.

Essence of the aggregates

Recently, the aggregation behaviour of ILs in water has been widely reported.²⁹ ILs with longer alkyl chains acted as surfactants and formed micellar aggregates at a certain concentration. However, in the currently studied system, this possibility can be excluded as [P₄₄₄₄][CF₃COO] is not surface-active. Over the years, the salt-induced phase separation of water-miscible organic solvents, such as 3-methylpyridine (MP), methanol, acetonitrile, and 1,4-dioxane with water, has been investigated by various technologies.^30–34 A microheterogeneous phase was found as large-sized clusters in these mixtures. These microheterogeneities are considered to be related to the nonequilibrium mesoscopic aggregates.^31,35,36 It has been proposed that the organic small molecules are shielded from the added ions by water molecules, and at the same time, ions form a charged layer, which helps to stabilize the clusters.^31a,d,35

ILs have been widely reported to possess a similar polarity to short-chain alcohols.³⁷ Therefore, they have similar properties of polar organic small molecules, including methanol, MP, and 1,4-dioxane, as mentioned above. Besides, ILs are essentially molten salts. So, we reason that these aggregates in the [P₄₄₄₄][CF₃COO]–water system could be similar to the salt-induced mesoscopic phase separation of the mixture of polar organic solvents and water. The mesoporous phase separation also occurred in the [bmim][BF₄]–water mixture.²⁴ The driving force for the formation of this microheterogeneous structure was believed to be the attraction among [bmim][BF₄] molecules in water. In addition, it is difficult to break down the 3D network of water molecules by adding low-polar small molecules.³⁸ These two factors were considered to drive the formation of a cluster structure in water.

In order to confirm that the [P₄₄₄₄][CF₃COO] aggregates are a salt-induced critical behavior of fluids, we attempted to add additional inorganic salt, NaCl, to the [P₄₄₄₄][CF₃COO]–water (4 : 6, w/w) mixture. Fig. 9 shows the temperature dependence of the average size of [P₄₄₄₄][CF₃COO] aggregates in the presence of different NaCl concentrations. The size increased with increasing NaCl concentration, indicating that NaCl promoted the [P₄₄₄₄][CF₃COO] molecules to aggregate. Therefore, we believe that the [P₄₄₄₄][CF₃COO] aggregates may be also salt-induced. Although NaCl played a similar role to apolar toluene and chloroform in leading to an increase in aggregate size, the mechanism of action is different. In the former case, we have mentioned above that organic substance provides an apolar environment for the [P₄₄₄₄][CF₃COO] molecules to accelerate aggregation. Toluene is a good solvent for [P₄₄₄₄][CF₃COO], the addition of toluene “pulls” the [P₄₄₄₄][CF₃COO] molecules to assemble together. However, in the latter case, NaCl prefers to be dissolved in water since NaCl is not sufficiently soluble in pure [P₄₄₄₄][CF₃COO]. Hydrophilic ILs can be salted out when in contact with concentrated solutions of the water-structuring NaCl.³⁹ Figuratively, if we consider toluene to “pull” the [P₄₄₄₄][CF₃COO] molecules together to aggregate, then NaCl would “push” them together. Moreover, we found that with the addition of NaCl, the LCST of the [P₄₄₄₄][CF₃COO]–water mixture decreased. The same phenomenon was also observed in the MP/water/NaBr system,^30,31a where the added NaCl was considered to reduce the strength of hydrogen-bond forces, and thus the LCST. This result further reveals that the [P₄₄₄₄][CF₃COO]–water binary mixture behaves like a MP/water/NaBr system and the aggregates are the salt-induced microheterogeneities. In this currently studied system, [P₄₄₄₄][CF₃COO] is not only the polar organic molecule, like MP, but also the electrolyte, like NaBr in traditional nonequilibrium microheterogeneous phase.

Fig. 9 Temperature dependence of the average size of [P₄₄₄₄][CF₃COO] aggregates in water (4 : 6, w/w) with different NaCl contents (the weight percent shown in the figure is relative to the weight of the microemulsion).

The mixture of MP/water/NaBr is the one of the most intensively investigated nonequilibrium systems. With an increasing amount of ions, MP molecules exsolve from the homogeneous solution and the mesoscopic MP-rich aggregates appear.³⁰ This means that the water content in the MP-rich aggregates is decreased with increasing ion concentration. To this end, the intermolecular interaction between [P₄₄₄₄][CF₃COO] and water was also investigated. ¹H NMR spectroscopic analysis can give detailed information about the interaction between the molecules and thus we used ¹H NMR to gain insight into the valuable information of the water molecules in the [P₄₄₄₄][CF₃COO] aggregates. Fig. 10 shows the ¹H NMR spectra of the [P₄₄₄₄][CF₃COO]–water (4 : 6, w/w) system at different temperatures. It can be seen that the proton signals on the tetrabutylphosphonium cation shifted downfield with increasing temperature. Ghandi and coworkers have used proton chemical shifts to investigate the intermolecular interactions in mixtures of trihexyl(tetradecyl)phosphonium chloride IL and water.⁴⁰ The chemical shifts of protons on alkyl chains were also found to move downfield as the temperature was increased, as well as when the water concentration in the IL was decreased. However, the temperature coefficient was found to be very much smaller at high water concentrations, which means that the effect of temperature on the chemical shift of can be ignored. The relationship between water concentration and the position of the CH₂ and CH₃ chemical shifts could be attributed to the partial transfer of electron density from the water's oxygen lone pairs to the cation, and therefore, to the protons of the alkyl chains on the IL.³⁹ This would result in an increase in the electron density of the CH₃ protons, and thus result in the chemical shifts of these protons moving further upfield when more water was added. In other words, a downfield shift reveals that the water content in the IL was decreased with temperature in the [P₄₄₄₄][CF₃COO]–water system, which is also in accordance with the measurements by Ohno et al.²¹ Obviously, whatever we increase the temperature for the [P₄₄₄₄][CF₃COO]–water system or add NaBr to the MP–water mixture, as described in the literature, it comes to the same result in the end, that is, a microheterogeneous phase or nonequilibrium mesoscopic aggregates will be induced to appear. Although the mesoscopic domains of [P₄₄₄₄][CF₃COO] have similar properties to microemulsions, the formation mechanism is very different. Therefore, we prefer to name the system microemulsion-like aggregates. A schematic representation of the phase behaviour transition process of the [P₄₄₄₄][CF₃COO]–water binary system with temperature is given in Fig. 11. The foregoing discussion then leads one to suspect that changing the temperature of other LCST-type IL–water binary mixtures should also cause a critical phenomenon. The relevant experiments have not yet been undertaken but clearly deserve serious consideration.

Fig. 10 ¹H NMR spectra of the [P₄₄₄₄][CF₃COO]–water mixture (4 : 6, w/w) at different temperatures. From the bottom up: 288 K, 293 K, and 298 K.

Fig. 11 Schematic models of the [P₄₄₄₄][CF₃COO] aggregates in water at different temperatures.

Conclusions

In summary, we describe a binary microemulsion-like system consisting of ionic liquid tetrabutylphosphonium trifluoroacetate ([P₄₄₄₄][CF₃COO]) and water, in which [P₄₄₄₄][CF₃COO] can form long-living aggregates without using surfactants under certain conditions. The droplets can be “swollen” by changing temperature, and apolar substances, like toluene or chloroform can be solubilized into the [P₄₄₄₄][CF₃COO] microdroplets. In addition, the aggregates demonstrate a similar tunable micropolarity to traditional microemulsions. These microemulsion-like aggregates are essentially the immediate state of a phase transition from a homogeneous phase to macroscopic phase separation.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (21273235 and 21073189). Also, the authors would like to thank the Natural Scientific Foundation of Shandong Province of China (ZR2011BM017) for the financial support.

References

1. (a) C. Solans and H. Kunieda, Industrial Application of Microemulsions, Surfactant Science Series, Dekker, New York, 1997, vol. 66, pp. 175–197; (b) J. Eastoe, S. Gold, S. Rogers, P. Wyatt, D. C. Steytler, A. Gurgel, R. K. Heenan, X. Fan, E. J. Beckman and R. M. Enick, Angew. Chem., Int. Ed., 2006, 45, 3675–3677.
2. Y. Liu, P. G. Jessop, M. Cunningham, C. A. Eckert and C. L. Liotta, Science, 2006, 313, 958–960.
3. (a) Y. A. Gao, J. Zhang, H. Y. Xu, X. Y. Zhao, L. Q. Zheng, X. W. Li and L. Yu, ChemPhysChem, 2006, 7, 1554–1561; (b) Y. A. Gao, A. Voigt, L. Hilfert and K. Sundmacher, ChemPhysChem, 2008, 9, 1603–1609; (c) Y. A. Gao, N. Li, L. Q. Zheng, X. Y. Zhao, S. H. Zhang, B. X. Han, W. G. Hou and G. Z. Li, Green Chem., 2006, 8, 43–49; (d) Y. A. Gao, N. Li, L. Q. Zheng, X. T. Bai, L. Yu, X. Y. Zhao, J. Zhang, M. W. Zhao and Z. Li, J. Phys. Chem. B, 2007, 111, 2506–2513; (e) Y. A. Gao, L. Hilfert, A. Voigt and K. Sundmacher, J. Phys. Chem. B, 2008, 112, 3711–3719.
4. (a) R. Pramanik, S. Sarkar, C. Ghatak, V. G. Rao and N. Sarkar, J. Phys. Chem. B, 2011, 115, 2322–2330; (b) R. Atkin and G. G. Warr, J. Phys. Chem. B, 2007, 111, 9309–9316; (c) J. H. Porada, M. Mansueto, S. Laschat and C. Stubenrauch, Soft Matter, 2011, 7, 6805–6810; (d) Z. Bai and T. P. Lodge, J. Am. Chem. Soc., 2010, 132, 16265–16270; (e) A. Chandran, K. Prakash and S. Senapati, J. Am. Chem. Soc., 2010, 132, 12511–12516; (f) Y. W. Lian and K. S. Zhao, Soft Matter, 2011, 7, 8828–8837; (g) R. Atkin, S. M. C. Bobillier and G. G. Warr, J. Phys. Chem. B, 2010, 114, 1350–1360.
5. H. X. Gao, J. C. Li, B. X. Han, W. N. Chen, J. L. Zhang, R. Zhang and D. D. Yan, Phys. Chem. Chem. Phys., 2004, 6, 2914–2916.
6. J. H. Liu, S. Q. Cheng, J. L. Zhang, X. Y. Feng, X. G. Fu and B. X. Han, Angew. Chem., Int. Ed., 2007, 46, 3313–3315.
7. S. Q. Cheng, J. L. Zhang, Z. F. Zhang and B. X. Han, Chem. Commun., 2007, 2497–2499.
8. S. Q. Cheng, X. G. Fu, J. H. Liu, J. L. Zhang, Z. F. Zhang, Y. L. Wei and B. X. Han, Colloids Surf., A, 2007, 302, 211–215.
9. J. Eastoe, S. Gold, S. E. Rogers, A. Paul, T. Welton, R. K. Heenan and I. Grillo, J. Am. Chem. Soc., 2005, 127, 7302–7303.
10. (a) D. Seth, A. Chakraborty, P. Setua and N. Sarkar, J. Chem. Phys., 2007, 126, 224512; (b) D. Seth, P. Setua, A. Chakraborty and N. Sarkar, J. Chem. Sci., 2007, 119, 105–111; (c) D. Seth, A. Chakraborty, P. Setua and N. Sarkar, J. Phys. Chem. B, 2007, 111, 4781–4787; (d) D. Seth, A. Chakraborty, P. Setua and N. Sarkar, Langmuir, 2006, 22, 7768–7775; (e) A. Chakraborty, D. Seth, P. Setua and N. Sarkar, J. Phys. Chem. B, 2006, 110, 5359–5366; (f) A. Chakraborty, D. Seth, D. Chakrabarty, P. Setua and N. Sarkar, J. Phys. Chem. A, 2005, 109, 11110–11116; (g) D. Chakrabarty, D. Seth, A. Chakraborty and N. Sarkar, J. Phys. Chem. B, 2005, 109, 5753–5758; (h) D. Chakrabarty, A. Chakraborty, D. Seth and N. Sarkar, J. Phys. Chem. A, 2005, 109, 1764–1769; (i) D. Chakrabarty, A. Chakraborty, P. Hazra, D. Seth and N. Sarkar, Chem. Phys. Lett., 2004, 397, 216–221.
11. Y. A. Gao, S. B. Han, B. X. Han, G. Z. Li, D. Shen, Z. H. Li, J. M. Du, W. G. Hou and G. Y. Zhang, Langmuir, 2005, 21, 5681–5684.
12. F. Yan and J. Texter, Chem. Commun., 2006, 2696–2698.
13. Z. H. Li, J. L. Zhang, J. M. Du, B. X. Han and J. Q. Wang, Colloids Surf., A, 2006, 286, 117–120.
14. G. P. Zhang, H. H. Zhou, J. Q. Hu, M. Liu and Y. F. Kuang, Green Chem., 2009, 11, 1428–1432.
15. M. Moniruzzaman, N. Kamiya, K. Nakashimaa and M. Goto, Green Chem., 2008, 10, 497–500.
16. Y. Shu, D. H. Cheng, X. W. Chen and J. H. Wang, Sep. Purif. Technol., 2008, 64, 154–159.
17. G. D. Smith, C. E. Donelan and R. E. Barden, J. Colloid Interface Sci., 1977, 60, 488–496.
18. (a) B. A. Kelser, D. Varie, R. E. Barden and L. H. Holt, J. Phys. Chem., 1979, 83, 1276–1280; (b) N. F. Borys, S. L. Holt and R. E. Barden, J. Colloid Interface Sci., 1979, 71, 526–532.
19. F. Song, J. Xu and W. G. Hou, Chin. Chem. Lett., 2010, 21, 880–883.
20. J. Xu, A. L. Yin, J. K. Zhao, D. X. Li and W. G. Hou, J. Phys. Chem. B, 2013, 117, 450–456.
21. Y. Kohno, H. Arai, S. Saita and H. Ohno, Aust. J. Chem., 2011, 64, 1560–1567.
22. K. Fukumoto and H. Ohno, Angew. Chem., Int. Ed., 2007, 46, 1852–1855.
23. Y. Kohno and H. Ohno, Phys. Chem. Chem. Phys., 2012, 14, 5063–5070.
24. Y. Chen, F. Y. Ke, H. P. Wang, Y. M. Zhang and D. H. Liang, ChemPhysChem, 2012, 13, 160–167.
25. L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton, Phys. Chem. Chem. Phys., 2001, 3, 5192–5200.
26. A. Maiti, A. Kumar and R. D. Rogers, Phys. Chem. Chem. Phys., 2012, 14, 5139–5146.
27. G. W. Zhou, G. Z. Li and W. J. Chen, Langmuir, 2002, 18, 4566–4571.
28. Y. Kohno, S. Saita, K. Murata, N. Nakamura and H. Ohno, Polym. Chem., 2011, 2, 862–867.
29. B. Dong, N. Li, L. Q. Zheng, L. Yu and T. Inoue, Langmuir, 2007, 23, 4178–4182.
30. V. Balevicius, Z. Gdaniec, J. Tamuliene and H. Fuess, Phase Transitions, 2008, 81, 293–301.
31. (a) M. A. Anisimov, J. Jacob and A. Kumar, Phys. Rev. Lett., 2000, 85, 2336–2339; (b) J. Jacob, A. Kumar, M. A. Anisimov, A. A. Povodyrev and J. V. Sengers, Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top., 1998, 58, 2188–2200; (c) A. F. Kostko, M. A. Anisimov and J. V. Sengers, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2004, 70, 026118; (d) J. Jacob, A. Kumar, S. Asokan, D. Sen, R. Chitra and S. Mazumder, Chem. Phys. Lett., 1999, 304, 180–186.
32. T. Takamuku, A. Yamaguchi, D. Matsuo, M. Tabata, T. Yamaguchi, T. Otomo and T. Adachi, J. Phys. Chem. B, 2001, 105, 10101–10110.
33. E. Hawlicka and D. Swiatla-Wojcik, Phys. Chem. Chem. Phys., 2000, 2, 3175–3180.
34. K. Sadakane, N. Iguchi, M. Nagao, H. Endo, Y. B. Melnichenkoe and H. Seto, Soft Matter, 2011, 7, 1334–1340.
35. J. Jacob, M. A. Anisimov, A. Kumar, V. A. Agayan and J. V. Sengers, Int. J. Thermophys., 2000, 21, 1321–1338.
36. M. Wagner, O. Stanga and W. Schroer, Phys. Chem. Chem. Phys., 2004, 6, 580–589.
37. (a) R. B. Mellein, S. N. V. K. Aki, R. L. Ladewski and J. F. Brennecke, J. Phys. Chem. B, 2007, 111, 131–138; (b) J. M. Lee, S. Ruckes and J. M. Prausnitz, J. Phys. Chem. B, 2008, 112, 1473–1476; (c) L. Crowhurst, P. R. Mawdsley, J. M. Perez-Arlandis, P. A. Saltera and T. Welton, Phys. Chem. Chem. Phys., 2003, 5, 2790–2794.
38. F. H. Stillinger, Science, 1980, 209, 451–457.
39. K. E. Gutowski, G. A. Broker, H. D. Willauer, J. G. Huddleston, R. P. Swatloski, J. D. Holbrey and R. D. Rogers, J. Am. Chem. Soc., 2003, 125, 6632–6633.
40. J. Dwan, D. Durant and K. Ghandi, Cent. Eur. J. Chem., 2008, 6, 347–358.

---