Rheology of Protic Ionic Liquids and Their Mixtures - The Journal of

Oct 8, 2013 - At 20 °C, EtAN has the highest zero shear viscosity of 156.1 mPa·s, followed by PAN (89.3 mPa·s), EAN (35.9 mPa·s), EAF (23.1 mPa·s...
0 downloads 3 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCB

Rheology of Protic Ionic Liquids and Their Mixtures J. A. Smith,† Grant B. Webber,† Gregory G. Warr,‡ and Rob Atkin*,† †

Priority Research Centre for Advanced Particle Processing and Transport, The University of Newcastle, Callaghan, New South Wales 2308, Australia ‡ School of Chemistry, The University of Sydney, Sydney, New South Wales 2006, Australia S Supporting Information *

ABSTRACT: The rheological properties of five pure protic ionic liquids (ILs), ethylammonium nitrate (EAN), propylammonium nitrate (PAN), ethanolammonium nitrate (EtAN), ethylammonium formate (EAF), and dimethylethylammonium formate (DMEAF), are characterized and interpreted by considering the effects of both the H-bond network and the solvophobic nanostructure of the liquids. The results demonstrate that these effects are not, however, independent or simply additive. At 20 °C, EtAN has the highest zero shear viscosity of 156.1 mPa·s, followed by PAN (89.3 mPa·s), EAN (35.9 mPa· s), EAF (23.1 mPa·s), and DMEAF (9.8 mPa·s). The primary ammonium ILs behave as Newtonian fluids at low shear rates but shear thin at high shear. Fits to the Vogel−Fulcher−Tammann model reveal that nanostructure is not affected appreciably by temperature and that all the ILs studied are of intermediate fragility. The rheology of binary mixtures of these ILs was analyzed and used to demonstrate fundamental differences in the way IL cations and anions interact. IL mixtures containing both nitrate and formate anions resist flow more strongly than the pure liquids, which is a consequence of the difference in hydrogen bonding capacity of the anions. Mixing cations can give rise to complex behavior due to the offsetting effects of hydrogen bonding and solvophobic nanostructure formation.

1. INTRODUCTION Ionic liquids (ILs) are pure salts which melt below 373 K. Low melting points are generally achieved by employing large, sterically hindered ions, such that the charge is distributed over a large molecular volume, which weakens electrostatic attractions and prevents the ions packing neatly into a crystal lattice. ILs have attracted interest as lubricants,1,2 as solvents for synthesis and catalysis,3−7 as continuous phases for particle dispersions,8−12 and for many other applications.13−17 Apart from the omnipresent Coulombic effects, a wide array of other forces operate within ILs that can affect their structure and properties. Hydrogen bonds can have a significant effect, especially when present as a dense three-dimensional network, and are known to be important in protic ionic liquids exemplified by ethylammonium nitrate (EAN).18,19 In some ILs H-bond networks can give rise to solvophobic effects20 that augment van der Waals attractions between uncharged groups and generate amphiphilically nanostructured liquids, depending on the cation and anion chemistry.21−24 In this paper we explore how the rheological properties of pure protic ILs are affected by such nanostructure through systematic variation in the structure and H-bonding capacity of the cation and the anion. Control over viscosity is further tuned by examining mixtures of ILs with common cations and anions. Currently, about 200 papers have described the rheology of ILs,25 of which the vast majority focus on aprotic ILs. Tokuda et al.26 systematically varied the ionic structure to determine how this affects the viscosity of aprotic ILs. First, for the same © 2013 American Chemical Society

cation, larger anions generally produce less viscous materials. Cations with longer alkyl chains produce more viscous ionic liquids due to stronger solvophobic and van der Waals interactions. Similarly, cations with more localized charges produce more viscous liquids.26 Fewer papers have examined the viscoelastic properties of protic ILs,17,27−31 and none have examined the rheology of protic IL mixtures. Bouzón Capelo et al.29 recently reported temperature dependent viscosities and other liquid properties of ethylammonium nitrate (EAN), propylammonium nitrate (PAN), and butylammonium nitrate (BAN). In this paper the temperature dependent viscosities of EAN and PAN are reexamined and analyzed in light of what is now known about the bulk liquid nanostructure and hydrogen bonding in these liquids, which leads to somewhat different conclusions about the origins of viscoelastic behavior. Rheological studies of protic ILs frequently include mixtures with molecular solvents. Jacquemin et al.17 investigated the rheology of diisopropylethylammonium alkanoate ILs and their mixtures with both acetonitrile and water. Although this cation is unable to form a H-bond network, the rather long (C6 and C7) alkyl chains used are likely to favor amphiphilic aggregation. These low viscosity ILs were found to be Newtonian at low shear rates, but shear thickened above a shear rate of 2000 s−1. When they were mixed with water the Received: August 1, 2013 Revised: September 14, 2013 Published: October 8, 2013 13930

dx.doi.org/10.1021/jp407715e | J. Phys. Chem. B 2013, 117, 13930−13935

The Journal of Physical Chemistry B

Article

Neutron and X-ray scattering experiments19,41−45 have shown that even small protic ILs can be nanostructured in bulk. This results from electrostatic attractions and hydrogen bonding between charged groups leading to the formation of polar domains. The cation alkyl chains are expelled from these charged regions via the solvophobic effect46 and cluster together to form apolar domains within the liquid. To examine the influence of the various contributing interionic forces on the rheology of protic ILs, we compare three ILs known to form strongly amphiphilic nanostructures with long-range orderpropylammonium nitrate (PAN), ethylammonium nitrate (EAN), and ethylammonium formate (EAF)with two different but amphiphilically “unstructured” or reference ILs: ethanolammonium nitrate (EtAN) and dimethylethylammonium formate (DMEAF). EAN, EAF, and PAN all have spongelike bulk nanostructures.43,44 PAN has the most pronounced nanostructure43,45 with the clearest segregation between the charged and uncharged domains. EtAN can form a dense, three-dimensional H-bond network through multiple cation and anion donor and acceptor sites. The H-bonding capacity of the hydroxyl moiety renders the extent of amphiphilic association in EtAN negligible compared to primary alkylammonium ILs. As secondary and tertiary ammonium nitrates are explosive, the effect of alkyl substitution on the cation is probed using DMEAF, which cannot form a H-bond network; each DMEA+ cation has only a single H-bonding site.20 The bulkiness of the cation results in weak bulk structure.41

viscosity increased at all compositions, consistent with amphiphilic association being further promoted by water. The behavior of IL−acetonitrile mixtures was more complicated. At low IL concentrations the viscosity lay between that of the IL and acetonitrile, but at high IL concentrations the viscosity increased markedly, above that of both pure solvents. Burrell et al.28 investigated the viscoelastic properties of diethanolammonium acetate and its mixtures with water, and found significant shear thinning at modest shear rates (∼1000 s−1) in the pure IL. This IL is expected to form a dense H-bond network, but the extent of any solvophobic association is unknown. When water was added the viscosity decreased, as did the magnitude of the shear thinning. At very high water concentrations, shear thickening was observed. Both results were attributed to changes in the liquid structure. Rheology is also commonly used to assess the completeness of dissociation, or ionicity of ILs, using a method based on the Walden rule.27,32,33 The Walden rule was developed based on observations of dilute aqueous salt solutions, and it has since been shown to be valid for molten salts and ILs.34 The Walden plot (ln(conductivity) vs ln(viscosity−1) over a range of temperatures) for a completely ionized material should be linear and pass through the origin. The degree of ion dissociation in an IL is assessed by comparison to a reference aqueous electrolyte solution; 0.01 M aqueous KCl is used as the reference line, as it is completely ionized and the low concentration ensures that interactions between ions have negligible effect on ion conductivity.27 Because ILs are pure salts, ion−ion interactions do affect transport properties, and Walden plots for most ILs fall below the KCl reference line.35 When ion−ion interactions are accounted for,27 ILs that fall within ∼20% of the reference line are classed as “good” ILs, meaning they are highly dissociated. ILs with Walden plots more than ∼20% less than the KCl reference line are not regarded as fully dissociated, and they are classified as “poor” ILs.27 By this criterion, EAN, EtAN,36 and PAN29 are fully dissociated, “good” ILs, in agreement with previous work that also classified both EAN and ethylammonium formate (EAF) as good ILs.27 Many ILs are easily supercooled and are excellent glassformers. The variation of viscosity with temperature may also be used to characterize an IL’s fragility.37 A strong liquid more closely follows Arrhenius-like behavior consistent with a single activation energy for flow, whereas fragile liquids exhibit nonArrhenius temperature dependence. This is interpreted in terms of long-range structural correlations within the liquid that are often weak or poorly defined, and disintegrate when heated, giving a more complex flow mechanism.37 The general dependence of viscosity (or any appropriate transport property) on temperature can be modeled using the Vogel− Fulcher−Tammann37,38 (VFT) equation: η = η0 exp[D*(T0/(T − T0))]

2. EXPERIMENTAL SECTION ILs were prepared by titrating concentrated acid (nitric or formic, Sigma-Aldrich) into a 5% excess of amine (ethanolamine, ethylamine, propylamine, or dimethylethylamine, SigmaAldrich), at a temperature maintained less than 10 °C. The resultant IL−water mixture was dried by rotary evaporation (for all ILs), followed by nitrogen purge and heating at 110 °C under an inert atmosphere (for nitrates only). The final pure ionic liquids were clear or faintly yellow. Nitrate-based ILs had water contents undetectable by Karl Fischer titration, while the formate ILs had water contents less than 0.1% by weight. Conductivity measurements were made using a Eutech Instruments con510 conductivity meter between 20 and 50 °C at 5 °C intervals. Rheological measurements were made with a TA Instruments AR-G2 rheometer using the cone and plate arrangement with a cone of 40 mm radius and angle of 1°59′36″. After each rheology experiment, the sample was analyzed by Karl Fischer titration to determine that water ingression was not appreciable. These measurements revealed that the water content did not increase by more than 0.1% w/w for any sample. This small increase in water content will have negligible effect on the rheological measurements. Temperature was controlled using a Peltier plate and jacket with chilled water. Rheological data were measured 15 times on samples from three separate IL batches (for each IL) and averaged. Temperature dependent viscosity was measured at a constant shear rate chosen to fall within the Newtonian response range for each liquid. Temperature was ramped from 20 to 50 °C during a 30 min period, with viscosity measured at each degree increment. Shear dependent viscosities were measured by sweeping the shear rate from 1 to 4000 s−1, accepting an average of five consecutive data points within 5% of one another at each shear rate in the sweep.

(1)

where η is the solution viscosity at a given temperature, T, and the fit parameters are the fragility (D*), the Vogel temperature (T0 < Tg), and the limiting high-temperature viscosity (η0). T0 for ILs is usually between 100 and 200 K, D* is typically between 1 and 10, and η0 is between 0.1 and 1 Pa s.27,29,33,39 Bouzón Capelo et al.29 reported Vogel temperatures of 169 K for EAN, 182 K for PAN, and 181 K for BAN and intermediate fragilities based on conductivity data, in accordance with results reported for other ILs.40 13931

dx.doi.org/10.1021/jp407715e | J. Phys. Chem. B 2013, 117, 13930−13935

The Journal of Physical Chemistry B

Article

Figure 1. (A) Zero shear viscosities of protic ILs EAN, PAN, EtAN, EAF, and DMEAF as a function of temperature, and (B) Arrhenius plots of the zero shear viscosities of these ILs. Symbols are experimental data, while solid lines show fits to the Vogel−Fulcher−Tammann equation.

Table 1. Physical Properties and Fit Parameters for the Tested ILs at 20°C: Zero Shear Viscosity, η, Density, ρ, and Refractive Index, RI, with the Vogel−Fulcher−Tammann Fitting Parameters T0, η0, D*, and the Derived Fragility Index, B IL

η (mPa s)

ρ (g cm−3)

RI

T0 (K)

η0 (mPa s)

D*

B (×10−21 J)

EAN PAN EtAN EAF DMEAF

35.9 89.3 156 23.1 9.8

1.21 1.15 1.39 1.04 1.03

1.4535 1.4565 1.4850 1.4340 1.4220

168 169 187 166 146

0.35 0.39 0.35 0.30 0.19

3.5 4.0 3.5 3.4 4.0

8.13 9.35 9.03 7.80 8.07

3. RESULTS AND DISCUSSION Walden plots showing EtAN, EAN, PAN, and EAF to be fully dissociated or “good” ILs have been reported previously.27,29 The Walden plot for DMEAF also reveals it to be a strong IL. This means that the concentrations of neutral species in all of the ILs investigated in this work must be extremely low, and will not affect the viscoelastic properties of the liquids. Figure 1A presents the low shear viscosity of these protic ILs between 20 and 50 °C. For all ILs, the viscosity decreases with temperature. The viscosity at any temperature decreases in the order EtAN > PAN > EAN > EAF > DMEAF. The data for EAN and PAN are in agreement with that reported previously.29 Viscosity as a function of temperature is presented in Arrhenius form in Figure 1B, together with VFT fits (eq 1). Best fit parameters are listed in Table 1, together with the derived fragility index, B = D*kbT0, which characterizes a liquid’s viscous response to temperature.37 The fitted Vogel temperature for EAN agrees extremely well with that obtained by Bouzón Capelo et al.29 from conductivity, whereas the PAN value is a little lower (169 K versus 182 K).29 Despite their very different nanostructures, all of these ILs have Arrhenius-like, almost linear profiles over the temperature range examined, and also have similar fragility indices. The fragility indices obtained from viscosity are somewhat greater than those reported previously based on conductivity: 4.9 vs 3.8 for EAN and 5.6 vs 4.0 for PAN.29 This may be a result of the well-known differences in activation energies for conductivity and viscosity. The fact that the temperature range used in these measurements was narrower than that used by Bouzón Capelo et al.29 may also be a factor. The temperature range used in our experiments is a consequence of EAF undergoing an elimination reaction at elevated temperatures to produce the amide; this reaction also occurs at low temperatures but is sufficiently slow as to not affect the rheology measurements

appreciably. For ease of comparison, the same temperature range was used for all the ILs. Intermediate fragility means each IL’s response to heating is similar: the thermal motion of the ions increases but the liquid nanostructure does not change appreciably. This is consistent with negligible change in structure being observed by small angle neutron diffraction over a comparable temperature range for EAN and EtAN.47 On the basis of quite small differences in VFT fit parameters, Bouzón Capelo et al. asserted that the viscoelastic behavior of PAN was different from that of EAN and BAN because of “the fact that alkane chains with an odd number of carbon atoms are known to pack worse than their even number counterparts”.29 While our results do not exclude odd/even effects as a possible factor for liquid flow, they also do not indicate unusual rheological properties for PAN relative to EAN, and our recent neutron diffraction results suggest that the H-bond network is affected by increasing the cation alkyl chain length from ethyl to propyl to butyl.19 Elucidating the links between nanostructure, hydrogen bonding, and rheological properties for the EAN, PAN, and BAN series requires further investigation. The trend in viscosity for the ILs can be understood in terms of the relative strength of the IL bulk structures.41,43,44,48 Hydrogen bonding is the most significant intermolecular force in these systems (after electrostatics common to all ILs). Among these ILs, only DMEAF cannot form a dense H-bond network, as each DMEA+ can only form one hydrogen bond. As a result, DMEAF has the lowest viscosity of the protic ILs examined. This agrees well with the results reported for aprotic ILs where sterically hindered cations have low viscosities. EtAN has the densest H-bond network of the ILs examined and the highest viscosity.19,44 In PAN the hydroxyl group of EtAN is replaced by a −CH3 and the level of hydrogen bonding in the liquid decreases. PAN, however, has the most segregated nanostructure in this series, as H-bonding drives the cation’s propyl chains to solvophobically associate, yielding a bicontinuous, spongelike structure with long-range perio13932

dx.doi.org/10.1021/jp407715e | J. Phys. Chem. B 2013, 117, 13930−13935

The Journal of Physical Chemistry B

Article

dicity.43 This association is weak and nonspecific, and results in less resistance to flow than EtAN. EAN also exhibits solvophobic nanostructure, albeit less pronounced than PAN’s, so viscosity is further reduced. The nature of these offsetting H-bond and nanostructure effects will be explored further below, where we consider IL mixtures. Replacing nitrate with formate in EAF reduces the hydrogen bonding capacity of the anion26 and weakens the hydrogen bond network in the polar domain.19 Hence the liquid flows more easily. The solvophobic nanostructures of EAN and EAF are strikingly similar,19 underscoring the dominant effect of Hbonding on viscosity. Figure 2 shows the viscosities of these pure ILs as a function of shear rate at 23 °C. EtAN, the most viscous IL examined, was

in a spongelike structure.53 The viscosity of the polar regions for primary ammonium ILs is likely to be comparable with that of EtAN, while the apolar region will have reduced viscosity. In contrast, aqueous bicontinuous microemulsions only shear thin above 103−104 s−1.54 The viscosities of the pure ionic liquids examined here vary over more than an order of magnitude (between 9.8 and 156 for DMEAF and EtAN, respectively). A significant challenge for the incorporation of ILs into real-world systems is to determine ways to retain the useful properties of the ILs while minimizing unfavorable characteristics; often high viscosity is cited as unfavorable. For example, in electrochemical systems high IL viscosity leads to relatively low conductivities given that ILs are pure salts. The use of IL mixtures is one approach to addressing these issues, and we have examined viscosity as a function of the composition for selected IL mixtures in order to tease out fundamental cause/effect relationships. Figure 3 shows the zero shear viscosities of binary mixtures of EAN, PAN, EtAN, and EAF at 23 °C determined using steady shear between 10 and 100 s−1 (see Figure A in the Supporting Information). Differences in viscosity upon mixing pure liquids are highlighted by the viscosity deviation, Δη (Figure 4).55,56 This is determined by linearly interpolating the viscosities of the two pure liquids, η1 and η2, and subtracting this from a mixture’s measured viscosity, ηm, according to Δη = ηm − (x1η1 + (1 − x1)η2)

(2)

where x1 is the mass fraction of liquid 1. Negative Δη values typically indicate that intermolecular (or ionic) forces between in the neat liquids are disrupted upon mixing, such as in water/ ethylene glycol mixtures where ethylene glycol disrupts water’s H-bond network.56 Positive Δη values indicate enhanced intermolecular forces, such as in mixtures of acetonitrile with methanol or 1-propanol.55 Mixing cations of nitrate ILs (Figures 3A and 4A) all lead to negative (or near-zero) viscosity deviations, whereas positive deviations arise whenever formate and nitrate anions are mixed (Figures 3B and 4B). Remarkably, Figure 3B shows that addition of small amounts of EAF increases the viscosity of all three nitrate ILs. In order to interpret the observed mixture viscosities, we consider the effects of both the H-bond network and the solvophobic nanostructure. These are not, however, independent or simply additive effects. Our recent structural studies of protic ILs have shown that increasing the alkyl chain length from ethylammonium through propylammonium to

Figure 2. Viscosity as a function of shear rate for EAN, PAN, EtAN, EAF, and DMEAF measured at 23 °C.

Newtonian over the range examined. DMEAF is also Newtonian up to quite high shear rates, but displays fluid dynamic instabilities above 1000 s−1 as a result of its low viscosity.49 In contrast, all three solvophobically nanostructured ILsEAN, EAF, and PANshear thin slightly at shear rates above ∼100 s−1. This is reminiscent of the shear thinning observed in surfactant sponge phases, and it suggests the onset of layering of polar and apolar regions in the shear gradient direction.50−52 This effect occurs at surprisingly low shear rates for systems comprised of such small and mobile ions. However, this probably results from the high viscosity of the H-bonded network present in the polar regions of these ILs as the onset of shear thinning should vary inversely with the “solvent” viscosity

Figure 3. Viscosity as a function of composition for (A) nitrate−nitrate and (B) nitrate−formate mixtures. Lines are to guide the eye. 13933

dx.doi.org/10.1021/jp407715e | J. Phys. Chem. B 2013, 117, 13930−13935

The Journal of Physical Chemistry B

Article

Figure 4. Viscosity deviations for (A) nitrate−nitrate and (B) nitrate−formate mixtures at 20 °C. Solid lines are drawn to guide the eye.

cation tail groups in the apolar phase or weakening hydrogen bonding in the polar phase, reduce viscosity; both effects contribute to DMEAF having the lowest viscosity. Increasing temperature reveals that all the ILs examined have similar, intermediate, fragilities, consistent with little or no change in nanostructure. The effect of mixing different anions and cations has also been examined. Mixtures containing both nitrate and formate anions have increased the resistance to flow, due to the differences in hydrogen bonding capacity of the anions. Mixing cations can give rise to complex behavior due to the offsetting effects of hydrogen bonding and solvophobic nanostructure formation.

butylammonium not only leads to a more pronounced spongelike nanostructure, but also changes the H-bond network between the cationic ammonium groups and anions.19 The almost linear behavior of PAN−EtAN mixtures seems at first to be most straightforward. As previously remarked, substitution of PA+ for EtA+ equates to removing H-bond capacity from the liquid, but also to the offsetting effect of the formation of a pronounced, solvophobically induced liquid nanostructure. The slight deviations from linearity suggest two regions. PAN-rich compositions are characterized by a negative Δη. Here the solvophobic nanostructure is progressively disrupted by EtAN addition. EtAN-rich compositions exhibit positive Δη, as PAN’s propensity to self-assemble changes the H-bonding network. Both EAN−PAN and EAN−EtAN mixtures show strong negative Δη values at all compositions. EAN’s weaker spongelike nanostructure is more easily disrupted, and it evolves smoothly into the EtAN H-bond network as the EtA+ content increases. From another perspective, EAN has insufficiently segregated structure to affect the EtAN H-bond network at low concentration. This behavior is qualitatively similar to that of ethylene glycol/water mixtures, which are also characterized by the gradual disruption of a strong water Hbond network.47 Mixtures of PAN and EAN similarly show a smooth transition from a less-pronounced to more-pronounced spongelike nanostructure (Figure 4A) with similar H-bond networks in their polar domains. The viscosity deviations of mixtures of EAF with all three nitrate-based ILs are strikingly alike, exhibiting similar positive deviations over a wide composition range. Given the low viscosity of EAF, its ability to increase the viscosity of even EtAN is all the more remarkable. As the nanostructures of the three nitrate ILs differ greatly, we attribute this effect to changes in the H-bond network due to the incorporation of formate. Further work is required to elucidate this, but we speculate that the presence of two anions allows a more optimal H-bond network to be formed in these mixtures.

■

ASSOCIATED CONTENT

* Supporting Information S

Shear rate dependent viscosities for the ionic liquid mixtures. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +61 2 49217107. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by Australian Research Council (ARC) Discovery Projects. R.A. wishes to acknowledge the ARC for a Future Fellowship. J.S. wishes to thank the University of Newcastle for provision of a Ph.D. scholarship.

■

REFERENCES

(1) Werzer, O.; Cranston, E. D.; Warr, G. G.; Atkin, R.; Rutland, M. W. Phys. Chem. Chem. Phys. 2012, 14, 5147−5152. (2) Sweeney, J.; Hausen, F.; Hayes, R.; Webber, G. B.; Endres, F.; Rutland, M. W.; Bennewitz, R.; Atkin, R. Phys. Rev. Lett. 2012, 109, 155502. (3) Clavel, G.; Larionova, J.; Guari, Y.; Guerin, C. Chem.Eur. J. 2006, 12, 3798−3804. (4) Liu, N.; Luo, F.; Wu, H.; Liu, Y.; Zhang, C.; Chen, J. Adv. Funct. Mater. 2008, 18, 1518−1525. (5) Seddon, K. R. J. Chem. Technol. Biotechnol. 1997, 68, 351−356. (6) Sheldon, R. Chem. Commun. 2001, 2399−2407. (7) Welton, T. Chem. Rev. 1999, 99, 2071−2083. (8) Prechtl, M. H. G.; Scariot, M.; Scholten, J. D.; Machado, G.; Teixeira, S. R.; Dupont, J. Inorg. Chem. 2008, 47, 8995−9001.

4. CONCLUSION The rheological behavior of pure protic ionic liquids and their mixtures depends on the molecular structures of both anion and cation components. The H-bond network strength and that of the solvophobic, spongelike nanostructure both have a significant impact on viscosity, whereas shear thinning is primarily related to nanostructure. Factors that reduce liquid structure, such as the frustrated packing of sterically hindered 13934

dx.doi.org/10.1021/jp407715e | J. Phys. Chem. B 2013, 117, 13930−13935

The Journal of Physical Chemistry B

Article

(9) Smith, J. A.; Werzer, O.; Webber, G. B.; Warr, G. G.; Atkin, R. J. Phys. Chem. Lett. 2010, 1, 64−68. (10) Ueno, K.; Imaizumi, S.; Hata, K.; Watanabe, M. Langmuir 2009, 25, 825−831. (11) Ueno, K.; Inaba, A.; Kondoh, M.; Watanabe, M. Langmuir 2008, 24, 5253−5259. (12) Wang, B.; Wang, X.; Lou, W.; Hao, J. Nanoscale Res. Lett. 2011, 6, 259. (13) Dupont, J.; de Souza, R. F.; Suarez, P. A. Z. Chem. Rev. 2002, 102, 3667−3691. (14) Niga, P.; Wakeham, D.; Nelson, A.; Warr, G. G.; Rutland, M.; Atkin, R. Langmuir 2010, 26, 8282−8288. (15) Earle, M. J.; Seddon, K. R. Pure Appl. Chem. 2000, 72, 1391− 1398. (16) Endres, F.; Hoefft, O.; Borisenko, N.; Gasparotto, L. H.; Prowald, A.; Al-Salman, R.; Carstens, T.; Atkin, R.; Bund, A.; Zein, E. A. S. Phys. Chem. Chem. Phys. 2010, 12, 1724−1732. (17) Jacquemin, J.; Anouti, M.; Lemordant, D. J. Chem. Eng. Data 2011, 56, 556−564. (18) Evans, D. F.; Chen, S.-H.; Schriver, G. W.; Arnett, E. M. J. Am. Chem. Soc. 1981, 103, 481−482. (19) Hayes, R.; Imberti, S.; Warr, G. G.; Atkin, R. Angew. Chem., Int. Ed. 2013, 52, 4623−4627. (20) Ray, A. Nature 1971, 231, 313−315. (21) Greaves, T. L.; Drummond, C. J. Chem. Rev. 2008, 108, 206− 237. (22) Greaves, T. L.; Kennedy, D. F.; Mudie, S. T.; Drummond, C. J. J. Phys. Chem. B 2010, 114, 10022−10031. (23) Greaves, T. L.; Weerawardena, A.; Fong, C.; Drummond, C. J. J. Phys. Chem. B 2007, 111, 4082−4088. (24) Greaves, T. L.; Weerawardena, A.; Krodkiewska, I.; Drummond, C. J. J. Phys. Chem. B 2008, 112, 896−905. (25) According to the SciFinder journal search service available from Chemical Abstracts Service. https://www.cas.org/products/scifinder (accessed July 21, 2013). (26) Tokuda, H.; Tsuzuki, S.; Susan, M. A. B. H.; Hayamizu, K.; Watanabe, M. J. Phys. Chem. B 2006, 110, 19593−19600. (27) Angell, C. A.; Byrne, N.; Belieres, J.-P. Acc. Chem. Res. 2007, 40, 1228−1236. (28) Burrell, G. L.; Dunlop, N. F.; Separovic, F. Soft Matter 2010, 6, 2080−2086. (29) Bouzón Capelo, S.; Mendez-Morales, T.; Carrete, J.; Lopez, L. E.; Vila, J.; Cabeza, O.; Rodriguez, J. R.; Turmine, M.; Varela, L. M. J. Phys. Chem. B 2012, 116, 11302−11312. (30) Lopez-Barron, C. R.; Wagner, N. J. Langmuir 2012, 28, 12722− 12730. (31) Lopez-Barron, C. R.; Basavaraj, M. G.; DeRita, L.; Wagner, N. J. J. Phys. Chem. B 2012, 116, 813−822. (32) Xu, W.; Cooper, E. I.; Angell, C. A. J. Phys. Chem. B 2003, 107, 6170−6178. (33) Yoshizawa, M.; Xu, W.; Angell, C. A. J. Am. Chem. Soc. 2003, 125, 15411−15419. (34) Campbell, A. N.; Kartzmark, E. M.; Anad, S. C.; Cheng, Y.; Dzikowski, H. P.; Skrynyk, S. M. Can. J. Chem. 1968, 46, 2399−2407. (35) Schreiner, C.; Zugmann, S.; Hartl, R.; Gores, H. J. J. Chem. Eng. Data 2009, 55, 1784−1788. (36) Angell, C. A.; Byrne, N.; Belieres, J.-P. Acc. Chem. Res. 2007, 40, 1228−1236. (37) Angell, C. A. J. Phys. Chem. Solids 1988, 49, 863−871. (38) Ediger, M. D.; Angell, C. A.; Nagel, S. R. J. Phys. Chem. 1996, 100, 13200−13212. (39) Belieres, J.-P.; Angell, C. A. J. Phys. Chem. B 2007, 111, 4926− 4937. (40) MacFarlane, D. R.; Forsyth, M.; Izgorodina, E. I.; Abbott, A. P.; Annat, G.; Fraser, K. Phys. Chem. Chem. Phys. 2009, 11, 4962−4967. (41) Greaves, T. L.; Kennedy, D. F.; Mudie, S. T.; Drummond, C. J. J. Phys. Chem. B 2010, 114, 10022−10031.

(42) Greaves, T. L.; Kennedy, D. F.; Weerawardena, A.; Tse, N. M. K.; Kirby, N.; Drummond, C. J. J. Phys. Chem. B 2011, 115, 2055− 2066. (43) Hayes, R.; Imberti, S.; Warr, G. G.; Atkin, R. Phys. Chem. Chem. Phys. 2011, 13, 13544−13551. (44) Hayes, R.; Imberti, S.; Warr, G. G.; Atkin, R. Phys. Chem. Chem. Phys. 2011, 13, 3237−3247. (45) Atkin, R.; Warr, G. G. J. Phys. Chem. B 2008, 112, 4164−4166. (46) Ronis, D.; Martina, E.; Deutch, J. M. Chem. Phys. Lett. 1977, 46, 53−55. (47) Hayes, R.; Imberti, S.; Warr, G. G.; Atkin, R. Phys. Chem. Chem. Phys. 2011, 13, 3237−3247. (48) Hayes, R.; El, A. S. Z.; Atkin, R. J. Phys. Chem. B 2009, 113, 7049−7052. (49) Whitcomb, P. J.; Macosko, C. W. J. Rheol. 1978, 22, 493−505. (50) Porcar, L.; Hamilton, W. A.; Butler, P. D.; Warr, G. G. Phys. Rev. Lett. 2002, 89, 168301. (51) Porcar, L.; Hamilton, W. A.; Butler, P. D.; Warr, G. G. Langmuir 2003, 19, 10779−10794. (52) Porcar, L.; Hamilton, W. A.; Butler, P. D.; Warr, G. G. Phys. Rev. Lett. 2004, 93, 198301. (53) Cates, M. E.; Milner, S. T. Phys. Rev. Lett. 1989, 62, 1856−1859. (54) Warr, G. G. Colloids Surf., A 1995, 103, 273−279. (55) Nikam, P. S.; Shirsat, L. N.; Hasan, M. J. Chem. Eng. Data 1998, 43, 732−737. (56) Tsierkezos, N. G.; Molinou, I. E. J. Chem. Eng. Data 1998, 43, 989−993.

13935

dx.doi.org/10.1021/jp407715e | J. Phys. Chem. B 2013, 117, 13930−13935