ARTICLE pubs.acs.org/
Computational Study of Ion Distributions at the Air/Liquid Methanol Interface Xiuquan Sun,† Collin D. Wick,‡ and Liem X. Dang*,† † ‡
Pacific Northwest National Laboratory, Richland, Washington 99352, United States Louisiana Tech University, Ruston, Louisiana 71270, United States ABSTRACT: Molecular dynamic simulations with polarizable potentials were performed to systematically investigate the distribution of NaCl, NaBr, NaI, and SrCl2 at the air/liquid methanol interface. The density profiles indicated that there is no substantial enhancement of anions at the interface for the NaX systems, in contrast to what was observed at the air/ aqueous interface. The surfactant-like shape of the larger more polarizable halide anions, which is part of the reason they are driven to air/aqueous interfaces, was compensated by the surfactant nature of methanol itself. These halide anions had on average an induced dipole of moderate magnitude in bulk methanol. As a consequence, methanol hydroxy groups donated hydrogen bonds to anions where the negatively charged side of the anion induced dipole pointed, and methyl groups interacted with anions where the positively charged side of the anion-induced dipole pointed. Furthermore, salts were found to disrupt the surface structure of methanol. For the neat air/liquid methanol interface, there is relative enhancement of methyl groups at the outer edge of the air/ liquid methanol interface in comparison with hydroxy groups, but with the addition of NaX this enhancement was reduced somewhat. Finally, with the additional of salts to methanol, the computed surface potentials decreased, which is in contrast to what is observed in corresponding aqueous systems, where the surface potential increases with the addition of salts. Both of these trends have been indirectly observed with experiments. The surface potential trends were found to be due to the greater propensity of anions for the air/water interface that is not present at the air/liquid methanol interface.
I. INTRODUCTION Understanding the adsorption and distribution of ions at liquid interfaces is important to elucidate the mechanisms for many chemical and physical processes encountered in biological and industrial systems. Some examples include interfacial chemical reactions such as phase transfer catalysis, protein folding and unfolding, and the uptake and remediation of pollutant molecules.1-4 The distribution of ions at air/aqueous interfaces has been of significant interest for quite some time.5-22 In previous understandings, a depletion of ions at the air/aqueous interface was presented, but this has been revised recently with the development of polarizable molecular models and sophisticated spectroscopic techniques, observing that larger, more polarizable anions have a propensity for the air/aqueous interface.5-22 The nonuniform distribution of ions at the air/aqueous interface is considered to have major consequences for atmospheric, environmental, biological, and industrial applications.5,22-24 Recently, there has been an interest in understanding interfacial ion adsorption and the distributions of ions at the interface of other polar solvents with air. Experiments have been developed to be sensitive to ion interfacial activity and structure, including ion scattering spectroscopy, nonlinear spectroscopy, and surface potential measurements.13,25-28 Additionally, several molecular dynamics (MD) simulations were performed to study these types of systems.24,29-33 In these MD investigations, iodide was found r 2010 American Chemical Society
to possess an interfacial enhancement with respect to the bulk solution for solvation in formamide, ammonia, and ethylene glycol.29,30 These results are very similar to what is found at the air/aqueous interface, despite the fact that ion solvation in these nonaqueous systems varies significantly from that of water. Methanol is unique in that it has different ion solvation behavior in comparison with other polar solvents.30,31 For instance, no significant segregation was observed for NaI at the air/liquid methanol interface. Methanol represents one of the simplest organic solvents, and gaining insights into its ability to solvate ions, and the interfacial structure of methanol after ion solvation are of both fundamental and practical importance. One of the special features of methanol not present in water and many other polar solvents is that it contains a surfactant-like molecular structure with both a hydrophilic hydroxy group and a fairly hydrophobic methyl group. Because of this molecular structure, methanol was found to exhibit segregation from water in water/methanol systems.34 At the pure air/liquid methanol interface, the methanol methyl group was found to preferentially point toward the air, showing a small Special Issue: Victoria Buch Memorial Received: August 10, 2010 Revised: September 23, 2010 Published: October 12, 2010 5767
dx.doi.org/10.1021/jp107563e | J. Phys. Chem. A 2011, 115, 5767–5773
The Journal of Physical Chemistry A
ARTICLE
methyl propensity for the air/liquid methanol interface.35,36 This, along with other factors, such as weaker methanol hydrogen bond strength in comparison with water, will likely affect ion interfacial behavior. The effects of adding ions to a solvent on electrostatic surface potentials, which can be measured indirectly, have been found to show different qualitative behavior for methanol than water. For instance, the addition of NaX salts was found to increase the surface potential of water, while adding these salts decreased the surface potential of methanol.28 Linking these results to molecular structures, orientations, and ion distributions, though, can be a great challenge, in which molecular simulation can bring significant insights. This understanding is far from complete, and this work represents a MD study utilizing polarizable potentials to investigate the behavior of NaX salts at the air/liquid methanol interface in comparison with the air/aqueous interface.
II. MODELS AND METHODS Polarizable molecular models were used for NaX, with X being chloride, bromide, or iodide, and for SrCl2, methanol, and water, which were all developed previously by us.37-40 Each system contained 2000 methanol molecules and 80 ion pairs, corresponding to an ion pair concentration of 1M. As for the 1 M SrCl2 sample, we used 80 strontium and 160 chloride ions and 2000 methanol molecules. The aqueous interfacial systems contained 4000 water molecules and 67 ion pairs (i.e., NaCl, NaBr, and NaI), corresponding to an ion pair concentration of 1M. All systems were placed in a rectangular periodic box with a dimension of 41 41 180 Å and 52 52 150 Å, respectively, for the air-methanol and air-water systems, respectively. Due to the fact that the z-dimension was much longer than the xy dimensions, the liquid occupied approximately 40% of the box. As a result, two air/liquid interfaces were formed for each system bisecting the z-axis. A potential truncation of 11 Å was enforced with analytical tail corrections employed, and the 3D particle mesh Ewald summation technique was used to handle longranged electrostatic interactions.41 The simulations were carried out in the NVT ensemble, and the temperature was controlled by coupling it to a Langevin dynamics piston.42 The systems were equilibrated for 10 ns, and 10 ns trajectories were used to extract all equilibrium properties. Surface potentials were calculated using the atomic approach.11,43,44 For our polarizable potential model, the total electric potential difference across the interface is calculated from the partial charges and induced dipoles, Zz Δφq ðzÞ ¼ φq ðzÞ - φq ðz0 Þ ¼ Eðz0 Þdz0 ð1Þ z0
Δφind μ ðzÞ ¼ Ez ðzÞ ¼
1 ε0
1 ε0
Z
z
z0
Z
z
z0
0 0 ÆFind μ ðz Þædz
ÆFq ðz0 Þædz0
ð2Þ ð3Þ
Here, E(z) is the electric field along surface normal direction, ɛ0 is the dielectric permittivity in vacuum, and z0 is the reference point that is selected as a point far from the interface in the bulk liquid. ÆFq(z)æ is the ensemble averaged charge density profile, which was evaluated in slabs of 0.25 Å thickness along the z direction, and Fμind(z) is the z component of the averaged induced dipole moment.
Figure 1. Normalized number density profiles for air/liquid interface of salt methanol and (a) NaCl, (b) NaBr, and (c) NaI.
III. RESULTS AND DISCUSSION III.A. NaX Density Profiles. Figure 1 shows the normalized density profiles for the 1 M NaX in methanol solutions, with zero representing the Gibbs dividing surface (GDS) of methanol. For all three systems, sodium is depleted from the air/liquid methanol interface. The anion shows some concentration at the interface, but to a much lesser degree than has been observed for our models at the air/aqueous interface.17 For instance, chloride has a similar interfacial concentration as in the bulk, which is different than was observed for the air/aqueous interfaces, in which chloride has an enhanced interfacial concentration.17 Bromide, surprisingly, has a very similar relative interfacial concentration at the air/liquid methanol interface as in the bulk, which is similar to what was found for chloride. This, again, is not observed at the air/aqueous interface, in which bromide has a significantly greater propensity than chloride.17 This is unexpected, as while our bromide model has only a slightly larger diameter in its Lennard-Jones potential than chloride (4.5436 vs 4.3387 Å, respectively),37 which is consistent with their 5768
dx.doi.org/10.1021/jp107563e |J. Phys. Chem. A 2011, 115, 5767–5773
The Journal of Physical Chemistry A experimental ionic radii,45 it has a larger polarizability. This suggests that polarizability plays little effect on anion interfacial behavior at the air/liquid methanol interface. There is an increase in sodium concentration near the air-water interface for both the NaCl and NaBr systems, which is somewhat larger for the NaBr system. Upon careful examination of Figures 1a and 1b, we can make the following observation. The probability of being at the methanol surface is slightly enhanced for the bromide anion when compared to the chloride anion. However, the difference is less significant than that observed for the case of these ions at an aqueous interface. Iodide, on the other hand, shows a clear double layer with enhancement at the air/liquid methanol interface, followed by a depletion layer. This, though, is still fairly minor, and is only a little different than for the NaBr and NaCl solutions, in contrast to the air/aqueous interface, where iodide shows a large concentration enhancement at the interface with respect to bulk. III.B. Analysis of Interfacial Driving Forces. The general picture of ion solvation is a competition between favorable electrostatic interactions of ions with the solvents and the penalty to create a cavity in the solvent. Both water and methanol have strong hydrogen bonding interactions, even though methanol has somewhat weaker hydrogen bond interaction than water. As a result, they both have strong interactions with ions. These include water and methanol hydroxy hydrogens strongly interacting with anions, and their oxygens strongly interacting with cations. Cations appear to behave in a similar fashion in both water and methanol, being repelled from both air/liquid interfaces. Anions, though, show a clear difference at the air/liquid methanol interface than at the air/aqueous interface. In bulk water, it has been found that more polarizable halide anions have an anisotropic solvation structure, which is due to the fact that halide anions in bulk water have an average induced dipole magnitude greater than zero.46 A consequence of this is that these anions have a surfactant-like structure in bulk water with a hydrophilic side and a hydrophobic side.46 Figure 2a shows a schematic of what we believe is a typical arrangement of methanol and water molecules near a polarizable anion. There are clearly two sides to the anion shown in Figure 2a, and the different sides are defined as follows. The side in which the induced dipole negative charge points will be called the hydrophilic side (the top in Figure 2a); and the other side, in which the induced dipole positive charge points, will be called the hydrophobic side (bottom of Figure 2a). For water, the hydrophilic side of a polarizable anion accepts hydrogen bonds from water molecules, due to the fact that water molecules on this side of the anion will interact both with the anion negative charge, and the negative charge of the anion induced dipole. On the hydrophobic side of a polarizable anion, there is a small cavity present in water devoid of any water hydrogens or oxygens.46 Farther away from the anion, but still on the hydrophobic side, there is a region in which it was found that water showed no strong orientational preference with respect to the anion position,46 which is consistent with hydrophobic interactions. One of the driving forces for larger, more polarizable anions to have higher concentrations at the air/aqueous interface is the fact that at the interface these surfactant-like anions have their hydrophobic side pointing toward the air. Methanol, on the other hand, is already surfactant-like in its shape, so that the anion’s hydrophobic side can have favorable interactions with the methanol methyl group, which reduces the penalty for forming this hydrophobic cavity in methanol. To better illustrate this, we calculated the
ARTICLE
Figure 2. (a) Schematic of a typical arrangement of water or methanol molecules surrounding a polarizable anion. (b) RADF between iodide and the hydrogens of the methanol methyl group (left) and between iodide and the hydrogens of the methanol hydroxy group (right). Schematics of how we define the angles and distances for each are given below their respective RADF diagrams.
radial-angular distribution function (RADF) between iodide and methanol hydrogens. This calculation was carried out in a cubic box with 1000 methanol molecules and 1 iodide anion. Two RADFs were calculated, one between iodide and hydroxy hydrogens, and the other between iodide and methyl hydrogens. The RADF is defined by the iodide-hydrogen distance, and the angle between the iodide induced dipole vector and the anionhydrogen vector. The RADF diagram is consistent with Figure 2a, in that the top is where the iodide induced dipole’s negative charge is pointing, and the bottom is where the iodide induced dipole’s positive charge is pointing. These results are given in Figure 2b for iodide in methanol with the right panel showing the RADF between iodide and hydroxy hydrogens and the left panel showing the RADF between iodide and methyl hydrogens. On the bottom of Figure 2b is a schematic showing how the distances and angles are calculated for each RADF. The methanol hydroxy hydrogens have a strong preference to be found in the hydrophilic side of iodide in the RADF, which is consistent with what is shown in Figure 2a. This preference is even present in the anion’s second solvation shell due to additional methanol molecules hydrogen bonding with the first-solvation shell methanol molecules. Toward the hydrophobic side of iodide, there are not any hydroxy hydrogens present in 5769
dx.doi.org/10.1021/jp107563e |J. Phys. Chem. A 2011, 115, 5767–5773
The Journal of Physical Chemistry A the first solvation shell, but some are present at a distance of around 4 Å. This shows that, as is the case for water, there is a small cavity on the hydrophobic side of iodide. The RADF of the methyl hydrogens show a strong preference for methyl hydrogens to be located at an angle of around 45 on the hydrophilic side of iodide, but these are clearly the methyl groups attached to hydrogen-bonding hydroxy groups. There is also a preference for methyl hydrogens to be located on the hydrophobic side of iodide 3.5 Å and greater from iodide. This shows that the surfactant shape of methanol accommodates the solvation of polarizable anions, since the hydrophobic side of the anions can interact with the methanol methyl groups. In water, there is no surfactant-like shape to accommodate the solvation of polarizable anions, so polarizable anions get driven to their surface so that the anion’s hydrophobic side can point toward air. There is another contribution that has been cited for driving anions to the air/aqueous interface.37,47 Anion-induced dipoles are known to be enhanced at the air/aqueous interface due to the symmetry-breaking of a polar air/liquid interface. The air/liquid methanol interface also breaks the symmetry of the system, so it would be expected that iodide would have a greater induced dipole at the air/aqueous interface than in bulk methanol. Figure 3 gives the average induced dipole of iodide as a function of z-position in methanol and water with zero representing the GDS. It can be observed that at both the air/aqueous and air/ liquid methanol interfaces, the iodide induced dipole increases with respect to the bulk, and by a similar degree of around 1 D. This is despite the fact that iodide has a lower average induced dipole in bulk methanol than in bulk water. This similar increase as iodide approaches the air for both solutions shows that the energetic benefit with regard to stronger interactions with water or hydroxy hydrogens due to their greater induced dipoles should be similar also. As a result, this effect is likely not a major reason ion interfacial behavior differs so much between water and methanol. There may be one more thing that contributes to anion interfacial behavior. An anion at the interface will disrupt the interfacial structure of the solvent molecules. Interfacial methanol molecules have been found to orient with their methyl groups pointing toward the air at the neat air/liquid methanol interface.33,35,36 It is established by our previous discussion that hydroxy hydrogens strongly bind with iodide aligned with its induced dipoles. For an iodide ion at the air/liquid methanol
Figure 3. The average iodide induced dipole as a function of z-position for the air/aqueous and air/methanol interface.
ARTICLE
Figure 4. The difference in the normalized number density profiles between atoms of the methanol methyl group and the hydroxy group for pure methanol and NaI in methanol as a function of z-position. Positive values represent situations in which the methyl groups have a higher density than hydroxy groups.
interface, it will most likely have to displace some methanol methyl groups. There is likely an additional penalty associated with this. To illustrate this, we calculated the difference between the relative methyl atomic density and the hydroxy atomic density, and the profile for this difference with respect to z-position is given in Figure 4 with zero representing the GDS. For the pure air/liquid methanol interface, there is a strong preference for the methyl group to be located toward the outer edge of the interface and for the hydroxy group to be located on the inner edge of the interface. With the addition of NaI, the preference for the hydroxy group on the inner edge of the air/ liquid methanol interface is not affected significantly. On the other hand, the methyl group shows a significant depletion in preference for the outer edge of the air/liquid methanol interface. As stated earlier, there should be a penalty associated with this, as the methanol interfacial structure is clearly disturbed significantly. The picture we are developing is one in which there are three factors that need to be considered when investigating the driving forces for anions to the air/liquid methanol interface. There is an energetic advantage for polarizable anions to be located at the interface due to a larger induced dipole, but this is similar to the air/aqueous interface. Also, in methanol, since methanol itself is a surfactant, the formation of a hydrophobic cavity toward the positive pole of the anion-induced dipole is compensated by interactions with methanol methyl groups. Finally, there is also some penalty associated with interfacial anions having to replace methanol methyl groups, which prefer to orient toward the air. Figure 5 gives a snapshot of the air/liquid methanol interface with NaI present. It can be observed that the methyl groups prefer to orient toward the air. III.C. SrCl2 Density Profiles. The normalized number density profiles for the ions of SrCl2 dissolved in methanol are shown in figure 6. Chloride is found to be closer to the interface than strontium and an ion double layer can be observed. The number density of chloride is similar to its bulk value at the air/liquid methanol interface, but there is a region of significant strontium enhancement near the interface. This region of enhanced strontium density is followed by a region of depletion, in which chloride density is depleted also. These observations are significantly different than for NaCl solutions and show a 5770
dx.doi.org/10.1021/jp107563e |J. Phys. Chem. A 2011, 115, 5767–5773
The Journal of Physical Chemistry A
ARTICLE
Figure 5. A snapshot taken from the simulation of air/liquid NaI methanol interface. Sodium is represented as blue and iodide as yellow, with methanol as an ice blue stick.
Figure 6. Normalized number density profiles for SrCl2 solvated in methanol.
uniqueness for systems of divalent cations. Also, this is significantly different than what has been observed for SrCl2 the air/ aqueous interface, in which both strontium and chloride have very strong enhancement at the surface.39 III.D. Electrostatic Potentials. The surface potential is very sensitive to the distribution of species at the interface, and provides valuable information of the molecular orientation and ion distributions at interfaces. Figure 7a shows the electrostatic potentials for pure methanol, NaCl, NaBr, and NaI methanol solutions as a function of z-position with zero representing the GDS. The surface potential value for pure methanol was calculated to be -0.38 V and agrees with experimental and other theoretical values well.26-28,48 The nonzero value of the surface potential indicates the existence of orientational ordering of methanol molecule at the interface. The surface potential decreases with the addition of salt. The additional of NaCl and NaBr reduces the surface potential by 20%, and the reduction of
Figure 7. The electrostatic potentials across the interface for sodium halides in (a) air/liquid methanol and (b) air/aqueous interface systems.
surface potential by the addition of NaI is less, about 10%. The surface potentials of pure water and salt aqueous solutions are shown in Figure 7b. The pure water surface potential is ∼0.48 V. The addition of salt increases the surface potential; and the larger the anions, the greater the increase in surface potential. This is 5771
dx.doi.org/10.1021/jp107563e |J. Phys. Chem. A 2011, 115, 5767–5773
The Journal of Physical Chemistry A
ARTICLE
and will cause a decrease in surface potential. Since this is similar for both systems, it is not the reason for the different qualitative behavior observed for the air/liquid water and methanol interfaces with the addition of salt. As stated before, the differences observed for the air/liquid methanol and air/aqueous interfaces are primarily due to stronger ion double-layer that forms at the air/aqueous interface with the addition of NaCl.
Figure 8. Decomposition of the electrostatic potentials for (a) pure and NaCl in methanol and (b) pure and NaCl in water systems into contributions from static charges (q) and induced dipoles (μ).
consistent with both indirect experimental and theoretical studies.28,44 To gain insights into the different trends for the methanol and water systems, the surface potentials of pure solvent and NaCl in both methanol and water were decomposed into contributions from charge contributions and contributions from induced dipoles. The results are shown in Figure 8a for methanol and 8b for water. Both the charge and induced dipole contributions are significantly different for both systems. With the addition of NaCl into methanol, the charge contribution of the surface potential initially decreases by a similar degree as the pure system, until just inside the GDS. After that, there is a strong divergence for the NaCl and pure methanol systems likely due to the double-layer formed by the NaCl ions. Similar behavior can be observed for the charge contribution to the electrostatic potential at the air/ aqueous interface, except that the divergence is greater, leading to a total increase of 0.9 V with the addition of NaCl, whereas for methanol, the total increase is around 0.7 V. This difference is due to the NaCl double-layer that is formed at the air/aqueous interface that is not present at the air/liquid methanol surface. A double-layer with higher anion concentration toward the air causes an increase in potential drop, which is observed for the air/ aqueous interface. The lack of a significant ion double layer at the air/liquid methanol interface is the reason adding NaCl does not increase the surface potential, but decreases it. The induced dipole contribution to the electrostatic potential diverges rapidly from the pure and NaCl systems for the air/liquid interfaces of water and methanol. Also, there is a total decrease of around 0.8 V for methanol and 0.85 V for water with the addition of NaCl for both solvents. The divergence from the pure system is mostly due to induced dipoles from the anions, which point toward the bulk,
IV. CONCLUSION Molecular dynamics simulations were carried out to investigate ion distributions at the air/liquid methanol interface with polarizable potentials. There was a small preference for anions to be present at the outer edge of the air/liquid methanol interface, but no significant interfacial enhancement was observed. This is in contrast to what was observed for the air/aqueous interface, in which iodide and bromide showed strong interfacial enhancement. We investigated three aspects in relation to anion solvation and interfacial behavior. (i) Anions at the air/liquid methanol interface showed a higher induced dipole than in the bulk, and the increase in induced dipole at the interface was similar to the increase observed at the air/aqueous interface versus bulk water. (ii) The methanol structure itself, due to its surfactant-like shape, accommodated the solvation of polarizable anions better, which also have a surfactant-like shape due to the fact that polarizable anions have an average induced dipole significantly greater than zero. The solvation structure surrounding iodide showed that where the negatively charged side of iodide’s induced dipole pointed, there was a hydrophilic region that accepted hydrogen bonds from both water and methanol molecules. Where the positively charged side of the induced dipole pointed, there was a hydrophobic side, in which methanol methyl groups interacted with the iodide. For water, this hydrophobic side cannot be accommodated as well as in methanol, so there is a preference for the hydrophobic side of iodide to be expelled to the air/aqueous interface pointing toward the air. (iii) Also, the methanol interfacial structure, which has methyl groups preferentially oriented toward the air, was disrupted to a degree with the presence of interfacial anions, in which the methyl group preference was reduced in comparison with pure methanol. Although we believe both factors ii and iii play a role in why NaX salts show different interfacial behavior at air/liquid methanol surfaces than at air/ aqueous surfaces, which one plays a more important role is not indicated in our results. We also investigated the behavior of SrCl2 at the air/liquid methanol interface, in which there was an enhancement of both strontium and chloride in a region near the air/liquid methanol interface, but not as strong as was observed ion water. With the addition of sodium halides in methanol, the surface potential was found to decrease, which is in contrast with what is observed in water, in which the surface potential increases. This observation was found to be in agreement with previous experimental and theoretical analyses. ’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT This work was performed at the Pacific Northwest National Laboratory (PNNL) and was supported by the Division of Chemical Sciences, Geosciences and Biosciences, Office of Basic 5772
dx.doi.org/10.1021/jp107563e |J. Phys. Chem. A 2011, 115, 5767–5773
The Journal of Physical Chemistry A Energy Sciences, U.S. Department of Energy (DOE). PNNL is operated by Battelle for the DOE.
’ REFERENCES (1) Hu, J. H.; Shi, Q.; Davidovits, P.; Worsnop, D. R.; Zahniser, M. S.; Kolb, C. E. J. Phys. Chem. 1995, 99, 8768. (2) Magi, L.; Schweitzer, F.; Pallares, C.; Cherif, S.; Mirabel, P.; George, C. J. Phys. Chem. A 1997, 101, 4943. (3) Thomas, J. L.; Jimenez-Aranda, A.; Finlayson-Pitts, B. J.; Dabdub, D. J Phys Chem A 2006, 110, 1859. (4) Zhang, Y. J.; Cremer, P. S. Curr. Opin. Chem. Biol. 2006, 10, 658. (5) Knipping, E. M.; Lakin, M. J.; Foster, K. L.; Jungwirth, P.; Tobias, D. J.; Gerber, R. B.; Dabdub, D.; Finlayson-Pitts, B. J. Science 2000, 288, 301. (6) Dang, L. X.; Chang, T. M. J. Phys. Chem. B 2002, 106, 235. (7) Garrett, B. C. Science 2004, 303, 1146. (8) Petersen, P. B.; Saykally, R. J. Chem. Phys. Lett. 2004, 397, 51. (9) Raymond, E. A.; Richmond, G. L. J. Phys. Chem. B 2004, 108, 5051. (10) Benjamin, I. Chem. Rev. 2006, 106, 1212. (11) Chang, T. M.; Dang, L. X. Chem. Rev. 2006, 106, 1305. (12) Gopalakrishnan, S.; Liu, D. F.; Allen, H. C.; Kuo, M.; Shultz, M. J. Chem. Rev. 2006, 106, 1155. (13) Hofft, O.; Borodin, A.; Kahnert, U.; Kempter, V.; Dang, L. X.; Jungwirth, P. J. Phys. Chem. B 2006, 110, 11971. (14) Jungwirth, P.; Tobias, D. J. Chem. Rev. 2006, 106, 1259. (15) Shen, Y. R.; Ostroverkhov, V. Chem. Rev. 2006, 106, 1140. (16) Cheng, J.; Hoffmann, M. R.; Colussi, A. J. J. Phys. Chem. B 2008, 112, 7157. (17) Wick, C. D.; Dang, L. X. Chem. Phys. Lett. 2008, 458, 1. (18) Bian, H. T.; Feng, R. R.; Guo, Y.; Wang, H. F. J. Chem. Phys. 2009, 130. (19) Feng, R. R.; Bian, H. T.; Guo, Y.; Wang, H. F. J. Chem. Phys. 2009, 130. (20) Horinek, D.; Herz, A.; Vrbka, L.; Sedlmeier, F.; Mamatkulov, S. I.; Netz, R. R. Chem. Phys. Lett. 2009, 479, 173. (21) Sun, X. Q.; Dang, L. X. J. Chem. Phys. 2009, 130, 124709. (22) dos Santos, A. P.; Diehl, A.; Levin, Y. Langmuir 2010, 26, 10778. (23) Bostrom, M.; Williams, D. R. M.; Stewart, P. R.; Ninham, B. W. Phys. Rev. E 2003, 68. (24) Jungwirth, P.; Winter, B. Annu. Rev. Phys. Chem. 2008, 59, 343. (25) Andersson, G.; Krebs, T.; Morgner, H. Phys. Chem. Chem. Phys. 2005, 7, 2948. (26) Barraclough, C. G.; Mctigue, P. T.; Ng, Y. L. J. Electroanal. Chem. 1992, 329, 9. (27) Case, B.; Parsons, R. Trans. Faraday Soc. 1967, 63, 1224. (28) Randles, J. E. B. Phys. Chem. Liq. 1977, 7, 107. (29) Andersson, G.; Morgner, H.; Cwiklik, L.; Jungwirth, P. J. Phys. Chem. C 2007, 111, 4379. (30) Cwiklik, L.; Andersson, G.; Dang, L. X.; Jungwirth, P. Chemphyschem 2007, 8, 1457. (31) Dang, L. X. J. Phys. Chem. A 2004, 108, 9014. (32) Faralli, C.; Pagliai, M.; Cardini, G.; Schettino, V. J. Phys. Chem. B 2006, 110, 14923. (33) Matsumoto, M.; Kataoka, Y. J. Chem. Phys. 1989, 90, 2398. (34) Dixit, S.; Crain, J.; Poon, W. C. K.; Finney, J. L.; Soper, A. K. Nature 2002, 416, 829. (35) Chang, T. M.; Dang, L. X. J. Phys. Chem. B 2005, 109, 5759. (36) Chen, H.; Gan, W.; Lu, R.; Guo, Y.; Wang, H. F. J. Phys. Chem. B 2005, 109, 8064. (37) Dang, L. X. J. Phys. Chem. B 2002, 106, 10388. (38) Dang, L. X.; Chang, T. M. J. Chem. Phys. 2003, 119, 9851. (39) Sun, X. Q.; Wick, C. D.; Dang, L. X. J. Phys. Chem. B 2009, 113, 13993. (40) Dang, L. X.; Chang, T. M. J. Chem. Phys. 1997, 106, 8149. (41) Ulrich, E.; Lalith, P.; Max, L. B.; Tom, D.; Hsing, L.; Lee, G. P. J. Chem. Phys. 1995, 103, 8577.
ARTICLE
(42) Pastor, R. W.; Brooks, B. R.; Szabo, A. Mol. Phys. 1988, 65, 1409. (43) Wilson, M. A.; Pohorille, A.; Pratt, L. R. J. Chem. Phys. 1988, 88, 3281. (44) Wick, C. D.; Dang, L. X.; Jungwirth, P. J. Chem. Phys. 2006, 125. (45) Shannon, R. Acta Crystallogr., Sect. A 1976, 32, 751. (46) Wick, C. A.; Xantheas, S. S. J. Phys. Chem. B 2009, 113, 4141. (47) Vrbka, L.; Mucha, M.; Minofar, B.; Jungwirth, P.; Brown, E. C.; Tobias, D. J. Curr. Opin. Colloid Interface Sci. 2004, 9, 67. (48) Yang, B.; Sullivan, D. E.; Gray, C. G. J. Chem. Phys. 1991, 95, 7777.
5773
dx.doi.org/10.1021/jp107563e |J. Phys. Chem. A 2011, 115, 5767–5773
