Communication: Unusual structure and transport in ionic liquid-hexane mixtures Min Liang, Sufia Khatun, and Edward W. Castner Jr. Citation: The Journal of Chemical Physics 142, 121101 (2015); doi: 10.1063/1.4916388 View online: http://dx.doi.org/10.1063/1.4916388 View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/142/12?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Composition dependent structural organization in trihexyl(tetradecyl)phosphonium chloride ionic liquidmethanol mixtures J. Chem. Phys. 142, 134503 (2015); 10.1063/1.4916308 A joint theoretical/experimental study of the structure, dynamics, and Li+ transport in bis([tri]fluoro[methane]sulfonyl)imide [T]FSI-based ionic liquids J. Chem. Phys. 139, 034502 (2013); 10.1063/1.4813413 Studies on the translational and rotational motions of ionic liquids composed of N -methyl- N -propylpyrrolidinium ( P 13 ) cation and bis(trifluoromethanesulfonyl)amide and bis(fluorosulfonyl)amide anions and their binary systems including lithium salts J. Chem. Phys. 133, 194505 (2010); 10.1063/1.3505307 Molecular-dynamics study of anomalous volumetric behavior of water-benzene mixtures in the vicinity of the critical region J. Chem. Phys. 123, 244507 (2005); 10.1063/1.2145749 1 H , 7 Li , and 19 F nuclear magnetic resonance and ionic conductivity studies for liquid electrolytes composed of glymes and polyetheneglycol dimethyl ethers of CH 3 O ( CH 2 CH 2 O ) n CH 3 (n=3–50) doped with LiN ( SO 2 CF 3 ) 2 J. Chem. Phys. 117, 5929 (2002); 10.1063/1.1501279

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 142.109.1.196 On: Thu, 30 Apr 2015 18:09:19

THE JOURNAL OF CHEMICAL PHYSICS 142, 121101 (2015)

Communication: Unusual structure and transport in ionic liquid-hexane mixtures Min Liang, Sufia Khatun, and Edward W. Castner, Jr.a) Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, Piscataway, New Jersey 08854, USA

(Received 11 February 2015; accepted 17 March 2015; published online 26 March 2015) Ionic liquids having a sufficiently amphiphilic cation can dissolve large volume fractions of alkanes, leading to mixtures with intriguing properties on molecular length scales. The trihexyl(tetradecyl)phosphonium cation paired with the bis(trifluoromethylsulfonyl)amide anion provides an ionic liquid that can dissolve large mole fractions of hexane. We present experimental results on mixtures of n-C6D14 with this ionic liquid. High-energy X-ray scattering studies reveal a persistence of the characteristic features of ionic liquid structure even for 80% dilution with n-C6D14. Nuclear magnetic resonance self-diffusion results reveal decidedly non-hydrodynamic behavior where the self-diffusion of the neutral, non-polar n-C6D14 is on average a factor of 21 times faster than for the cation. Exploitation of the unique structural and transport properties of these mixtures may lead to new opportunities for designer solvents for enhanced chemical reactivity and interface science. C 2015 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4916388]

INTRODUCTION

Ionic liquids are becoming ubiquitous in several fields of chemistry, materials science, and energy research. Some of the many applications of these fascinating solvents are separations,1 pre-treatment of lignocellulosic biomass for applications to liquid fuels,2 catalysis using nanoparticles,3,4 and electrochemical applications such as batteries and supercapacitors,5 in addition to a variety of energy applications.6 Modern ionic liquids (ILs) often have amphiphilic anions, cations, or both. Examples of such amphiphilic ions include alkyl sulfate anions and tetraalkyl- ammonium or phosphonium cations. When the alkyl tails on such anions or cations are long enough, a characteristic peak in the observed liquid structure function, the so-called first sharp diffraction peak (FSDP) is routinely observed both in X-ray and neutron scattering7,8 as well as in molecular simulations.9 For a cationic hydrocarbon chain of length n = 5 or longer, the FSDP has been shown to lead to nanoscale phase segregation between polar head groups and non-polar alkyl tails.7,10,11 Less hydrophobic substituents, such as oligoether chains, are not sufficiently amphiphilic to lead to a FSDP.12 One of the most hydrophobic IL cations is trihexyl (tetradecyl)phosphonium (P14,6,6,6+), which is commercially available in bulk quantities from Cytec. The P14,6,6,6+ cation is frequently paired with a variety of anions, but it is most commonly used with the chloride, dicyanamide, and bis(trifluoromethylsulfonyl)amide (NTf −2 ) anions. The bulk liquid structures of both the P14,6,6,6+/Cl− and the P14,6,6,6+/NTf −2 ILs have been studied in detail using a combination of Xray scattering and molecular simulations.9,13 For both the P14,6,6,6+/Cl− and P14,6,6,6+/NTf −2 ILs, three peaks are observed in the inter-molecular region of the liquid structure funca)E-mail: [email protected]

0021-9606/2015/142(12)/121101/4/$30.00

tion S(q). For both of these ILs, the FSDP is observed at a value of q = 0.42 Å−1, corresponding to a domain size of 15 Å. In addition to the structural information, other physical properties of ILs based on the P14,6,6,6+ cation have been reported by Ito et al.14 In addition to collective properties such as density, refractive index, and viscosity, Ito et al. reported on the dynamical properties of the solvent reorganization about a solvatochromic probe molecule, coumarin 153 (C153). The solvent reorganization dynamics and orientational friction experienced by the probe molecule were shown to depend strongly on the choice of the anion paired with the P14,6,6,6+ cation. The average lifetime for the strongly nonexponential solvation dynamics for C153 was 10.6 ns at room temperature.14 Mixtures of ILs with water and organic solvents have been studied in detail.15–17 Blesic et al. studied the solubility of various neutral organic species in a series of ILs based on the P14,6,6,6+ cation and anions included acetate, dicyanamide, triflate, and NTf −2 .16 Unlike most other ILs, this series of P14,6,6,6+-based ILs showed very high solubilities of alkanes. Alkane solubility in P14,6,6,6+ ILs correlates with the volume fraction of the alkane, so that a larger mole fraction of hexane could be dissolved than for tetradecane.16 Blesic et al. also compared the alkane solubilities in both P14,6,6,6+/NTf −2 and 1-methyl-3-tetradecylimidazolium/ NTf −2 ILs, finding that the high degree of solubility depends strongly on the nature of the cationic group, not solely on the length of the alkyl chain.16 For the case of hexane, the solubility limit at ambient temperature in P14,6,6,6+/NTf −2 corresponds to a mole fraction of x(C6 H14) = 0.9, while the corresponding volume fraction is about two-thirds. The explanation given for the large solubility was that the alkane finds a favorable solvation environment in a region comprising the alkyl tails of the P14,6,6,6+ cation that is locally segregated from the polar parts of the IL anions and cations.

142, 121101-1

© 2015 AIP Publishing LLC

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 142.109.1.196 On: Thu, 30 Apr 2015 18:09:19

121101-2

Liang, Khatun, and Castner, Jr.

J. Chem. Phys. 142, 121101 (2015)

Motivated by the report from Blesic et al., we have investigated the structural and transport properties of a series of mixtures of varying concentrations of n-C6D14 in the strongly hydrophobic IL P14,6,6,6+/NTf −2 . The mole fractions studied were x C6 D14 = 0.0, 0.02, 0.05, 0.2, 0.5, 0.8, and 1.0. Our hypothesis was that if the n-C6D14 were phase segregated, at high concentrations it should ultimately form nanoscale puddles in the midst of a larger IL domain, where the IL organization resulting from strong charge-charge interactions could also be locally preserved. To investigate these ideas, the structure of the mixtures was measured using high-energy Xray scattering, while the self-diffusion coefficients of the IL anions, cations, and the neutral n-C6D14 were measured using pulse-gradient spin echo (PG-SE) nuclear magnetic resonance (NMR) methods. EXPERIMENTAL METHODS

High-energy X-ray scattering experiments were performed using 115 keV X-ray photons at the Advanced Photon Source, beamline 11-ID-C, using experimental protocols and data analysis procedures described extensively in recent publications.9,11,12,18 All measurements were carried out at 295 K. Self-diffusion coefficients were obtained from PG-SE NMR data, measured using a Varian DirectDrive NMR at a 1H frequency of 400 MHz, using procedures described previously.19 PG-SE intensities were measured as a function of gradient strength using the bipolar pulse pair stimulated echo sequence (dbppste).20 A Doty Scientific model 16-38 gradient probe was used with gradient field strengths between 5 and 300 G/cm. Self-diffusion coefficients for the NTf −2 anion, the P14,6,6,6+ cation, and the n-C6D14 solute were measured using 19 F, 1H and 2H signals, respectively. Temperature-dependent viscosity measurements of the n-C6D14 - P14,6,6,6+/NTf −2 mixtures were made using a Cambridge Viscosity ViscoLab 4100 instrument, as described previously.21 Further experimental details on X-ray, viscosity and NMR experiments are available in the supplementary material.22 RESULTS AND DISCUSSION

Self-diffusion in ILs is often well described by the slip boundary condition Stokes-Einstein (S-E) or fractional StokesEinstein (FSE) equations; these are given by D = k BT/(4πηr) and D ∝ (T/η)m , respectively, where T is the absolute temperature, r is the effective spherical radius, and η is the shear viscosity. Harris showed for a wide variety of liquids that a FSE relation given by D ∝ (T/η)m more precisely describes liquid self-diffusion, with an average value of the exponent m = 0.90 for several common ILs, and for n-C6H14.23 Our observations for neat P14,6,6,6+/NTf −2 and n-C6D14 are fully consistent with this picture, but the mixtures show diffusion coefficients for n-C6D14 in solutions of P14,6,6,6+/NTf −2 that deviate strongly from FSE behavior. Temperature-dependent viscosities were measured for n-C6D14, P14,6,6,6+/NTf −2 , and mixtures of the two; the data are best fit to the Vogel-Fulcher-Tammann (VFT) equation, ln(η(T)) = A + B/(T − T0). Acceptable fits are obtained to the Arrhenius equation. The values of A, B, T0, and Ea are

FIG. 1. Plots of self-diffusion coefficients for mixtures of n-C6D14 with P14,6,6,6+/NTf −2 vs. the ratio of temperature to viscosity. Top: n-C6D14; middle: P14,6,6,6+; bottom: NTf −2 . Dashed lines denote the predictions from the Stokes-Einstein equation. Linear plots in T /η are in the supplementary material.22

given in the supplementary material.22 On going from neat P14,6,6,6+/NTf −2 to mixtures diluted with n-C6D14, the viscosity consistently decreases as does the value of Ea . The results from the PG-SE NMR self-diffusion measurements are analyzed in Fig. 1, where the self-diffusion coefficients are plotted versus the ratio of absolute temperature to shear viscosity on log-log axes. These data represent selfdiffusion measurements of the neat IL and of solutions with mole fractions x C6 D14 = 0.02, 0.05, 0.2, 0.5, and 0.8, for a set of three temperatures — 298, 308, and 318 K. From top to bottom, the graphs show self-diffusion coefficients for n-C6D14, P14,6,6,6+, and NTf −2 , respectively. The predicted diffusion coefficient from the Stokes-Einstein hydrodynamic equation for slip boundary conditions is plotted in the dashed line using measured temperatures and viscosities and an effective spherical radius calculated from the van der Waals volume. The effective spherical radii used were: 5.18, 3.36, and 3.00 Å for P14,6,6,6+, NTf −2 , and n-C6D14, respectively. The P14,6,6,6+ cation self-diffusion is well described by slip hydrodynamics for the neat IL and for n-C6D14 mixtures with the IL up to x C6 D14 = 0.50, above which point the diffusion coefficient becomes faster than the S-E prediction. Self-

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 142.109.1.196 On: Thu, 30 Apr 2015 18:09:19

121101-3

Liang, Khatun, and Castner, Jr.

diffusion for the NTf −2 anion is somewhat slower than predicted by the S-E theory. In sharp contrast, the self-diffusion of n-C6D14 is not at all well described by the S-E hydrodynamic predictions, consistently remaining more than order of magnitude larger than the calculated value with an average ratio of 21, with ratios in the range from 16 to 28. Kaintz et al. reported self-diffusion of small concentrations of benzene in several ILs, including P14,6,6,6+/NTf −2 . The ratio of diffusivities for C6H6 to that for P14,6,6,6+ was 28.2 at 298 K.24 It is interesting to compare the values of the Arrhenius activation enthalpy for the shear viscosity Ea,visc with the values of the self-diffusion coefficients for n-C6D14, P14,6,6,6+, and NTf −2 obtained from PG-SE experiments, Ea, PG−S E . We have done this for x C6 D14 = 0.5 to find that the values of Ea,visc = 31.8 kJ/mol are quite similar to Ea, PG−S E (NTf −2 ) = 29.5 kJ/mol. Since the self-diffusion of n-C6D14 is more than an order of magnitude faster than predicted by the SE equation, it is not as surprising to find a lower value for Ea, PG−S E (n-C6D14) = 17.5 kJ/mol. The value of Ea, PG−S E (P14,6,6,6+) lies between the values for n-C6D14 and NTf −2 at 21.8 kJ/mol. It seems clear that self-diffusion of the anions is the primary determinant of the viscosity. Further details are presented in the supplementary material.22 The structure of P14,6,6,6+/NTf −2 has been discussed in detail by Kashyap et al.9 Molecular dynamics (MD) simulations provide a quantitative comparison between experiment and theory, and the agreement obtained is very good. Analysis of the simulations permits unambiguous assignment of the correlations that contribute to S(q).18 These show that there is a sharp separation between intermolecular correlations in S(q) located in the range from 0.2 to 2.0 Å−1, with correlations in S(q) above 2.0 Å−1 being assigned wholly to intramolecular correlations. The peak at q = 1.37 Å−1 results from strong near-neighbor correlations between the IL anions and cations.9 The next peak in the inter-molecular region of S(q) at q = 0.75 Å−1 is assigned to self-charge-charge correlation resulting from ionic alternation. The pre-peak or FSDP at q = 0.42 Å−1 is the characteristic signature of the intermediate range order in ILs that have an amphiphilic ion with a longer alkyl chain substituent.9 The charge-charge correlation component is observed as a weak shoulder in S(q). However, we are confident of the assignment because of the quantitative agreement between the measured S(q) and that calculated from simulations. Analysis of the partial structure factors shows that the strong constructive interference in both the anionanion and cation-cation correlations is offset by a negative amplitude arising from destructive interference in the anioncation correlations for q = 0.75 Å−1.9 The structure of neat deuterated n-hexane was investigated by Venturi et al. using both neutron diffraction and molecular dynamics simulations.25 The lowest-q peak in the distinct neutron structure function and overall scattering cross section both show a maximum near q = 1.4 Å−1, in excellent agreement with the primary peak in the high-energy X-ray scattering liquid structure function for n-C6D14. Fig. 2 presents the liquid structure functions S(q) for neat P14,6,6,6+/NTf −2 , neat n-C6D14 and for the mixtures, measured using high-energy X-ray scattering. The most significant change in the liquid structure that results from addition of

J. Chem. Phys. 142, 121101 (2015)

FIG. 2. Liquid structure functions for mixtures of n-C6D14 in P14,6,6,6+/ NTf −2 ; mole fractions range from x C 6 D 14 = 0.0 to 1.0.

n-C6D14 to the P14,6,6,6+/NTf −2 IL is shown in the inset to Fig. 2. The dashed line tilting to the left indicates the shift of the FSDP maximum from q = 0.417 Å−1 for the neat IL to q = 0.353 Å−1 for x C6 D14 = 0.8. If one interprets the FSDP peak as indicating the domain size of the liquid intermediate range order, this means that this domain grows from an effective size of 15 Å for the neat IL to 17.8 Å for the mixture with x C6 D14 = 0.8. Blesic et al. presented a snapshot from a MD simulation of P14,6,6,6+/Cl− illustrating the domain sizes of the dominant non-polar regions that arise from segregation of the tetradecyl and hexyl groups on the P14,6,6,6+ cation from the polar domains. In the work of Kashyap et al., the sizes of the polar domains are indicated by plots of the partial pair distribution functions that show that the cation-cation (phosphorus atom) and anionanion (nitrogen atom) correlations are both peaked at 10 Å.9 Additional analysis of the neat-IL simulations by Kashyap et al. shows that the effective distances from the polar centers of the NTf −2 anion and the P14,6,6,6+ cations are roughly half this size, or 5 Å, and that the NTf −2 anions are tightly clustered in a tetrahedral distribution about the central phosphorus atom of P14,6,6,6+, with a coordination number of 3.6.9 Additional analysis of the liquid structure functions shown in Fig. 2 shows a simple intermolecular structure for n-C6D14, with a single peak in the inter-molecular part of S(q) at q = 1.41 Å−1 that corresponds to an intermolecular spacing of 4.5 Å. Thus, the natural sizing of the neat n-C6D14 liquid is coincidentally equal to the anion-cation spacing of the neat IL. The width of the primary intermolecular peak for n-C6D14 centered at 1.41 Å−1 is 0.23 Å−1, which is significantly broader than the peak at the same position that is observed for both P14,6,6,6+/NTf −2 and all of the mixtures with n-C6D14. For neat P14,6,6,6+/NTf −2 , the maximum of the anion-cation adjacency peak is found at q = 1.37 Å−1, with a width of

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 142.109.1.196 On: Thu, 30 Apr 2015 18:09:19

121101-4

Liang, Khatun, and Castner, Jr.

0.16 Å−1. For x C6 D14 = 0.2, significant broadening of this adjacency peak by about 7% is observed, despite the negligible volume or atom fraction for this concentration of hexane. While the explanation for this broadening of the X-ray peaks in the liquid mixtures on addition of small quantities of hexane is likely not a simple one, it does seem intriguing that this apparently indicates a weakening of the intrinsic structure of the IL. Further details of this structural analysis are available in the supplementary material.22

CONCLUSIONS

Addition of hexane (n-C6D14) to the P14,6,6,6+/NTf −2 IL leads to unexpected changes to several physical properties. Remarkably, the characteristic structure of the IL is preserved for concentrations through x C6 D14 = 0.8, near the saturation point for hexane in P14,6,6,6+/NTf −2 . Results from high-energy X-ray scattering measurements of the liquid structure function indicate that the major structural change is that the FSDP, characteristic of intermediate range order, shifts to lower values of q. This is consistent with the growth of the domain from a size of 15 to 17.8 Å, or a growth in volume by a factor of 1.67 on going from the neat IL to x C6 D14 = 0.8, corresponding to a volume fraction estimate of 40%. The measured viscosities for the n-C6D14 mixtures with P14,6,6,6+/NTf −2 are fit well by the empirical VFT equation, but an Arrhenius analysis, while statistically inferior, still provides reasonable fits in this temperature range. The self-diffusion coefficients show behavior that is in moderately good agreement with hydrodynamic predictions for the NTf −2 anion and the P14,6,6,6+ cation, but the hydrodynamic prediction is too small by more than an order of magnitude for the self-diffusion coefficient for n-C6D14 in the mixtures. Clearly, this indicates that the hexane experiences a very different environment than the polar regions, pointing to the likelihood of nanophase segregation of hexane in a domain together with the non-polar tetradecyl and hexyl tails on the P14,6,6,6+ cation. The unique structural and transport properties of these alkane-IL mixtures relate directly to their use as highly tunable solvent mixtures. A trivial adjustment of the concentration of alkane can lead to precise adjustment of the size of hydrophobic domains, density, viscosity, and degree of solvent amphiphilicity. We anticipate that these results will be useful for precise tailoring of amphiphilic solvent properties for challenging chemical reactions and separations. To this end, we have begun a number of further studies, including large-scale molecular simulations of these mixtures with over 1000 ion pairs. ACKNOWLEDGMENTS

We gratefully acknowledge support for this work from the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences under Contract No. DE-SC0001780 at Rutgers. We thank Dr. Nagarajan Murali for help with NMR experiments. This research used beamline 11-ID-C at the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Sci-

J. Chem. Phys. 142, 121101 (2015)

ence User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC0206CH11357. For expert assistance with experiments at the APS, we thank Dr. Yang Ren (APS), Professor N. Sanjeeva Murthy, Mr. Luiz F. O. Faria, and Ms. Jessalyn Devine (Rutgers). We thank Professors Mark Maroncelli, Claudio Margulis, and Dr. Juan Carlos Araque for helpful discussions. 1X.

Han and D. W. Armstrong, Acc. Chem. Res. 40, 1079–1086 (2007). Wang, G. Gurau, and R. D. Rogers, Chem. Soc. Rev. 41, 1519–1537 (2012). 3A. Riisager, R. Fehrmann, M. Haumann, and P. Wasserscheid, Eur. J. Inorg. Chem. 695–706 (2006). 4J. Dupont, Acc. Chem. Res. 44, 1223–1231 (2011). 5M. Armand, F. Endres, D. R. MacFarlane, H. Ohno, and B. Scrosati, Nat. Mater. 8, 621–629 (2009). 6J. F. Wishart, Energy Environ. Sci. 2, 956–961 (2009). 7A. Triolo, O. Russina, H.-J. Bleif, and E. Di Cola, J. Phys. Chem. B 111, 4641–4644 (2007). 8C. Hardacre, J. D. Holbrey, C. L. Mullan, T. G. A. Youngs, and D. T. Bowron, J. Chem. Phys. 133, 074510 (2010). 9S. M. Urahata and M. C. C. Ribeiro, J. Chem. Phys. 120, 1855–1863 (2004); H. K. Kashyap, C. S. Santos, H. V. R. Annapureddy, N. S. Murthy, C. J. Margulis, and E. W. Castner, Jr., Faraday Discuss. 154, 133–143 (2012). 10J. N. A. Canongia Lopes and A. A. H. Pádua, J. Phys. Chem. B 110, 3330–3335 (2006). 11O. Russina, A. Triolo, L. Gontrani, R. Caminiti, D. Xiao, J. Hines, G. Larry, R. A. Bartsch, E. L. Quitevis, N. Pleckhova, and K. R. Seddon, J. Phys.: Condens. Matter 21, 424121 (2009); H. K. Kashyap, C. S. Santos, N. S. Murthy, J. J. Hettige, K. Kerr, S. Ramati, J. Gwon, M. Gohdo, S. I. LallRamnarine, J. F. Wishart, C. J. Margulis, and E. W. Castner, Jr., J. Phys. Chem. B 117, 15328–15337 (2013). 12A. Triolo, O. Russina, R. Caminiti, H. Shirota, H. Y. Lee, C. S. Santos, N. S. Murthy, and E. W. Castner, Jr., Chem. Commun. 48, 4959–4961 (2012); H. K. Kashyap, C. S. Santos, R. P. Daly, J. J. Hettige, N. S. Murthy, H. Shirota, E. W. Castner, Jr., and C. J. Margulis, J. Phys. Chem. B 117, 1130–1135 (2013). 13L. Gontrani, O. Russina, F. L. Celso, R. Caminiti, G. Annat, and A. Triolo, J. Phys. Chem. B 113, 9235–9240 (2009). 14N. Ito, S. Arzhantsev, M. Heitz, and M. Maroncelli, J. Phys. Chem. B 108, 5771–5777 (2004). 15T. Gutel, C. C. Santini, A. A. H. Pádua, B. Fenet, Y. Chauvin, J. N. Canongia Lopes, F. Bayard, M. F. C. Gomes, and A. S. Pensado, J. Phys. Chem. B 113, 170–177 (2009). 16M. Blesic, J. N. Canongia Lopes, M. F. C. Gomes, and L. P. N. Rebelo, Phys. Chem. Chem. Phys. 12, 9685–9692 (2010). 17J. N. Canongia Lopes, M. F. C. Gomes, P. Husson, A. A. H. Pádua, L. P. N. Rebelo, S. Sarraute, and M. Tariq, J. Phys. Chem. B 115, 6088–6099 (2011); M. F. C. Gomes, L. Pison, A. S. Pensado, and A. A. H. Pádua, Faraday Discuss. 154, 41–52 (2012); M. Tariq, M. G. Freire, B. Saramago, J. A. P. Coutinho, J. N. Canongia Lopes, and L. P. N. Rebelo, Chem. Soc. Rev. 41, 829–868 (2012); N. Dias, K. Shimizu, P. Morgado, E. J. M. Filipe, J. N. C. Lopes, and F. V. Chavéz, J. Phys. Chem. B 118, 5772–5780 (2014). 18C. S. Santos, N. S. Murthy, G. A. Baker, and E. W. Castner, Jr., J. Chem. Phys. 134, 121101 (2011); C. S. Santos, H. V. R. Annapureddy, N. S. Murthy, H. K. Kashyap, E. W. Castner, Jr., and C. J. Margulis, ibid. 134, 064501 (2011). 19S. H. Chung, R. Lopato, S. G. Greenbaum, H. Shirota, E. W. Castner, Jr., and J. F. Wishart, J. Phys. Chem. B 111, 4885–4893 (2007); T. A. Fadeeva, Ph.D. thesis, Rutgers, The State University of New Jersey, Graduate School, New Brunswick, 2011. 20D. Wu, A. Chen, and C. S. J. Johnson, J. Magn. Reson. A115, 260–264 (1995). 21A. M. Funston, T. A. Fadeeva, J. F. Wishart, and E. W. Castner, Jr., J. Phys. Chem. B 111, 4963–4977 (2007). 22See supplementary material at http://dx.doi.org/10.1063/1.4916388 for more detailed information on high-energy X-ray scattering, viscosity, and PG-SE NMR experiments. 23K. R. Harris, J. Chem. Phys. 131, 054503 (2009). 24A. Kaintz, G. Baker, A. Benesi, and M. Maroncelli, J. Phys. Chem. B 117, 11697–11708 (2013). 25G. Venturi, F. Formisano, G. J. Cuello, M. R. Johnson, E. Pellegrini, U. Bafile, and E. Guarini, J. Chem. Phys. 131, 034508 (2009). 2H.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 142.109.1.196 On: Thu, 30 Apr 2015 18:09:19