Temperature Invariance of NaCl Solubility in Water: Inferences fromSalt−Water Cluster Behavior of NaCl, KCl, and NH4ClPankaj Bharmoria, Hariom Gupta, V. P. Mohandas, Pushpito K. Ghosh,* and Arvind Kumar*

Central Salt and Marine Chemicals Research Institute, Council of Scientific & Industrial Research (CSIR), G. B. Marg, Bhavnagar 364002, Gujarat, India

*S Supporting Information

ABSTRACT: The growth and stability of salt−water clustershave been experimentally studied in aqueous solutions ofNaCl, KCl, and NH4Cl from dilute to near-saturationconditions employing dynamic light scattering and zetapotential measurements. In order to examine cluster stability,the changes in the cluster sizes were monitored as a function oftemperature. Compared to the other cases, the average size ofNaCl−water clusters remained almost constant over the studied temperature range of 20−70 °C. Information obtained from thetemperature-dependent solution compressibility (determined from speed of sound and density measurements), multinuclearNMR (1H, 17O, 35Cl NMR), and FTIR were utilized to explain the cluster behavior. Comparison of NMR chemical shifts ofsaturated salt solutions with solid-state NMR data of pure salts, and evaluation of spectral modifications in the OH stretch regionof saturated salt solutions as compared to that of pure water, provided important clues on ion pair−water interactions and waterstructure in the clusters. The high stability and temperature independence of the cluster sizes in aqueous NaCl shed light on thetemperature invariance of its solubility.

■ INTRODUCTION

The characteristics of aqueous salt solutions and their structures(existence of free ions, solvated ions, ion pair, and higher orderclusters) have been investigated employing different exper-imental methods which include thermodynamic,1 spectroscopic(IR absorption, Raman, or NMR),2 neutron and X-raydiffraction,3 dynamic light scattering,4 and zeta potentialmeasurements.5

Ultrafast infrared spectroscopy2f,6a and ultrafast optical Kerreffect study,6b,c which shed light on low-frequency modes ofaqueous alkali halide solutions, and molecular dynamicscalculations7 have revealed the dynamical character of aqueousion solvation. Along with the experimental studies, a largenumber of theories have been developed to explain the ionicassociations and higher order cluster formation in theelectrolyte solutions.8 The temperature invariance of NaClsolubility in water is an intriguing phenomenon of considerablebasic and applied research interest.9 This temperatureinvariance is exploited, for example, for the separation of KClfrom a solid mixture of KCl/NaCl by the “hot leaching”process.10 Surprisingly, there is scant information on thesubject, and we report here our experimental studies of salt−water clusters to shed light on the phenomenon.The presence of large clusters in unsaturated aqueous salt

solutions under ambient conditions was detected by Georgaliset al. from dynamic light scattering experiments (DLS).4

However, important issues such as the growth of clusters withincrease in concentration of salt, the nature of interactionsinvolved in cluster formation, the temperature stability ofclusters, and how the solubility of salts may be correlated with

cluster size and stability have not been addressed. We haveperformed concentration- and temperature-dependent DLSand zeta potential measurements in aqueous solutions of NaCl,KCl, and NH4Cl, and have examined the growth andtemperature stability of clusters. It has been shown throughseparate studies that the adiabatic compressibility determinedfrom complementary measurements of density and speed ofsound provides important information on structural changes ofwater as a consequence of addition of electrolytes.11 We havetherefore conducted accurate measurements of speed of soundand density for the above saturated salt solutions as a functionof temperature and derived information on their adiabaticcompressibility.Water is a structured liquid with distinct local structures that

vary with temperature.12 Such variations in water structure canbe explicitly involved in solubility limits of different salts.Besides temperature, dissolved electrolytes too affect the waterhydrogen bonding. In the aqueous electrolyte solutions thereexist (i) hydrating water (water in the interior of clustersurrounding ions/ion pairs and water on the surface of cluster)and (ii) bulk water.2 Although the nature of chemicalenvironments of the former do not alter significantly with thechanges in electrolyte concentration or other physicalparameters such as solution temperature and pressure, theproportions of the different forms of water and detectablechanges in the equilibrium structure of the “bulk” water are

Received: July 23, 2012Revised: August 31, 2012Published: August 31, 2012

Article

pubs.acs.org/JPCB

© 2012 American Chemical Society 11712 dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−11719

observed as a consequence of changes in hydrogen bondstrength. Ionic charge densities govern water/ion interactionsand define the relative importance of both water structureelectrostatics, i.e., ion−water dipole interactions, and the extentof hydrogen bonding between water molecules. Due toenhanced electrostatic ordering, ions with higher charge densityperturb the bulk hydrogen bonding to a greater extent thanlarger ions with lower charge density. The water that is on thesurface of clusters mediates between the “bulk” water and thehydrated ion pairs, and communicates these small shifts to theion nuclei. Through FTIR2,13 and NMR,2,14 which are sensitiveand noninvasive techniques, it is possible to detect the changesin the hydrogen-bonding environment of water and also smallshifts in ion nuclei. We therefore recorded temperature-dependent FTIR and NMR (1H, 17O, and 35Cl) spectraprecisely to complement the DLS and compressibilityinformation in elucidating the water structure, cluster stability,and ion pair−water interactions in the highly concentratedaqueous salt solutions. Since chloride was the common ionamong the three salts studied, 35Cl chemical shift providedcrucial information on the role of associated cations and helpedquantify the changes in strength of ion pairs and ion−dipoleinteractions in salt solutions arising from increase in temper-ature.The combination of DLS and compressibility data, together

with temperature-dependent FT-IR and NMR results, haveenabled us to account for the negligible influence oftemperature on the solubility of sodium chloride, unlike inthe case of KCl and NH4Cl.

■ EXPERIMENTAL SECTIONMaterials and Methods. NaCl, KCl, and NH4Cl (>99.5%

by mol) were obtained from SD Fine Chemicals (Mumbai).Salts were used after drying in an oven at 70 °C without furtherpurification. Aqueous salt solutions were prepared by weight,using an analytical balance with a precision of ±0.0001 g(Denver Instrument APX-200) in Millipore grade water.Solubility of different salts at 20 °C was determinedgravimetrically by taking excess salt and dissolving to thesaturation limit, taking care to ensure that some solid saltremains undissolved. Stirring was carried out for at least 3 h atworking temperature. Saturated solutions were filtered, waterwas evaporated, and the left solids were oven-dried andweighed. Solubility of salts was also confirmed by withdrawingthe solutions periodically and analyzing the chloride content byvolumetric method using standard AgNO3 solution.Dynamic Light Scattering. Cluster size and solution zeta

potential measurements were performed at varying temperaturefrom 20 to 70 °C on a Zetasizer Nano ZS light scatteringapparatus (Malvern Instruments, U.K.) with a He−Ne laser(633 nm, 4 Mw). Salt solutions were filtered directly into quartzcell using a membrane filter of 0.45 μm pore size. Prior tomeasurements the cell was rinsed several times with filteredwater and then filled with filtered sample solutions. For zetapotential, the system employs electrophoretic light scatteringmethod to measure electrophoretic velocities of colloidalclusters in an applied electric field. Temperature of themeasurements was controlled with an accuracy of ±0.1 K.Densimetry. The density (ρ) of the salt solutions at

different temperatures was measured with an Anton Paar(Model DMA 4500) vibrating-tube densimeter with aresolution of 5 × 10−2 kg·m−3. The temperature of theapparatus was controlled to within ±0.03 K by a built-in Peltier

device. Reproducibility of the results was confirmed byperforming at least three measurements for each sample.

Ultrasonics. The speed of sound (u) in the salt solutions atvarying temperatures was measured at 51 600 Hz using aconcentration analyzer (Model 87, SCM Laboratory SonicComposition Monitor) based on the sing-around techniquewith a single transducer cell, immersed in a water bath withtemperature controlled to ±0.01 K. The analyzer was calibratedby measurements of speeds of sound in water as a reference,and the error was estimated to be less than ±0.1 m·s−1.Measurements were carried out in a specially designed samplejar of low volume capacity. Sample jars were provided with anairtight Teflon covering to keep the samples moisture freeduring measurements. Not less than three experiments wereperformed at each temperature to check the reproducibility ofthe results.

FT-IR Measurements. FT-IR measurements of saturatedaqueous salt solutions were carried out at room temperatureusing Nicolet 6700 FT-IR spectrometer. For recording spectra,a cell with BaF2 windows and a Teflon spacer was used; theoptical path length was 0.02 mm. For each spectrum, 132 scanswere made with a selected resolution of 4 cm−1. Temperature-dependent FTIR spectra were recorded on a IRPrestige-21(Shimadzu) machine.

Multi-Nuclear Magnetic Resonance (NMR) Measure-ments. The 1H, 17O, and 35Cl NMR spectra of saturatedaqueous salt solutions were recorded using a Bruker 500spectrometer. The proton and oxygen chemical shifts werereferenced with respect to D2O.

17O spectra were obtained at67.79 MHz. 35Cl spectra of various saturated aqueous saltsolutions were recorded corresponding to a salt solution of 0.1m concentration at 49 MHz. At least 32, 2000−3000, and 256scans were made for proton, oxygen and chlorine NMRrespectively. Temperature-dependent measurements werecarried out with an accuracy of ±0.1 K using an inbuiltvariable-temperature control unit.

■ RESULTS

The aqueous solubility (molal) of the studied salts as a functionof temperature is shown in Figure 1. The investigated salts hadvarying solubility with temperature, with the highest increasefor NH4Cl followed by KCl and nearly constant solubility inthe case of NaCl. Temperature invariance of solubility of NaClwas probed by examining the solution structures of NaCl, KCl,

Figure 1. Solubility of the studied salts in water as a function oftemperature.

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911713

and NH4Cl prepared at 20 °C. DLS measurements of saltsolutions (from very dilute to saturated) showed a bimodal sizedistribution, i.e., a small component corresponding to solvatedions (∼5 Å) and a large component corresponding to biggersize clusters. A typical plot of the apparent hydrodynamicdiameter (Dh) (calculated from the CONTIN method) versusconcentration (Figure 2) showed distinct growth pattern of

clusters in different salt solutions. The smallest cluster sizesobserved at very low concentrations (∼1 × 10−6 m) were 240,222, and 168 nm for NaCl, KCl, and NH4Cl solution,

respectively, at 20 °C. In the case of KCl solutions the sizeof the cluster grew linearly until saturation and acquired amaximum average size of ∼450 nm. In NaCl solutions thecluster size grew linearly up to a concentration of ∼4 m andthen became almost constant (∼560 nm) until saturation. Apeculiar behavior of cluster growth was observed in aqueousNH4Cl solutions wherein clusters grew very slowly until aconcentration of ∼4 m, and an exponential growth wasobserved thereafter (∼380 nm at saturation). Temperature-dependent DLS measurements were performed to investigatethe temperature stability of different salt clusters at saturationconcentrations. Plots of apparent hydrodynamic diameter (Dh)versus intensity of scattered light in the saturated salt solutionsas a function of temperature showing the large component areprovided in Figure 3a−c. With the increase in temperature from20 to 70 °C, the size of the cluster reduced by 20, 140, and 200nm for NaCl, KCl, and NH4Cl, respectively (Figure 3d),indicating that clusters formed in NaCl solutions weretemperature stable whereas those in aqueous KCl and NH4Clshrank to a sizable extent. The stability of different salt clusterswas probed through the measurements of zeta potential (theeffective surface potential at the hydrodynamic “shear surface”close to the cluster−bulk water interface), a key parametergoverning the electrokinetic behavior of particles in solution.This parameter determines the electrophoretic mobility of theparticles within an external electric field, as well as theelectrostatic repulsion between particles (or between a particleand a bounding surface) that acts to prevent or promoteparticle attraction and adhesion.5 Very high negative zeta

Figure 2. Cluster growth pattern in the studied salt solutions plottedagainst concentration at 20 °C.

Figure 3. Hydrodynamic diameter Dh distributions plotted against % intensity normalized to unity at different temperatures of (a) NaCl, (b) KCl,and (c) NH4Cl. Panels d and e show the plots of size distribution and zeta potential, respectively, for various salt solutions at different temperatures.

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911714

potentials (<−120 mV at 20 °C) were observed for all theinvestigated salt solutions (Figure 3e).Precise measurements of density and speed of sound (Table

S1 in the Supporting Information) allow determination ofsolution compressibility (κS) by using the Newton−Laplaceequation (κS = 1/u2ρ), where u is the speed of sound and ρ isthe solution density.11 Figure 4a shows the bar diagram

indicating change in compressibility with temperature for waterand the three saturated salt solutions, and plots of κS versus Tfor the salt solutions are shown in Figure 4b. Saturated solutionof NaCl was found to be the least compressible with a marginalrise in κS with temperature, whereas in the case of pure waterand saturated KCl and NH4Cl solutions, there was a drop in κSwith T, the trend being H2O > NH4Cl > KCl.FT-IR spectroscopy has been used to study the structure of

water under the influence of salts.13 Curve fitting of watervibrational spectrum into four deconvoluted bands is often usedto describe the structure of liquid water by the continuummodel proposed by Walrafen13a and subsequently developed byothers.13d,e The most prominent bands are around 3250 and3420 cm−1 corresponding to the strongly H-bonded patches ofmolecules with strongly and weakly H-bonded molecules,respectively. The component bands at 3550 and 3620 cm−1 are

attributed to the coupling of OH asymmetric stretchingvibrational modes,13c and the dangling OH stretching modesfrom 3- and 2-coordinate water molecules, respectively.13f Thedeconvoluted FT-IR spectra of water and the three saltsolutions are shown in Figure 5a−d. In the case of NH4Cl−water solution, the vibrational bands of NH4

+ and OHstretching of water were deconvoluted according to literatureprocedure13g and only the bands of OH were considered(Figure 5d). The band area ratio (BAR) of strongly bound tothe band area of the weakly H-bonded water molecules [i.e.,A∼3250/A∼3420] provided important information about thechanges in hydrogen-bonding network of water with theaddition of salts. Temperature dependence of FTIR spectra ofsaturated NaCl solution and BAR of strongly bound to all theweakly H-bonded water molecules are shown in Figure 6, a andb, respectively. Only a small reduction in spectral intensity wasobserved with the rise in solution temperature from 25 to 70°C, and the value of BAR hardly changed with temperature.The extent of interactions of ion pairs of various salts with

water molecules in the vicinity can be well understood from theNMR of various nuclei.14 Figure 7 shows the NMR spectra of1H, 17O, and 35Cl for the three saturated salt solutions as afunction of temperature. 1H shifts were found to decrease onlyslightly for both NaCl and KCl, while an increase in 17O shiftswere observed for both the cases. The 35Cl resonance appearedmarkedly downfield in saturated NH4Cl, whereas it was leastdownfield in the case of saturated NaCl. To obtain betterinsight on the Cl---H−O−H bond strengths in different salt−water clusters, the solid-state 35Cl spectra of the various studiedsalts were compared with those of the corresponding saturatedaqueous solutions (Figure 8). The chemical shifts in solid statefor NH4Cl, KCl, and NaCl were 75.2, 4.41, and −44.8 ppm,respectively (very close to literature values14h) and thecorresponding values for the solutions were 17.99, 6.5, and4.04 ppm, respectively. For the solid-state spectra, the shifts inthe chemical shift values were 49 and 2 ppm downfield forNaCl and KCl, respectively, while it was 57 ppm upfield forNH4Cl.

■ DISCUSSIONTemperature-dependent DLS (Figure 3d) and zeta potential(Figure 3e) measurements indicated that the clusters in thesaturated salt solutions were highly stable. A combination offactors such as electrostatic interactions among oppositelycharged ions, hydrophobic interactions among ions caused byexclusion of ions by water, and bridging of water molecules byanions and/or cations is reported to be responsible for thecluster stability.15 In the saturated salt solutions (particularly forthe salts having high solubility), there are only a few watermolecules available per ion which will necessarily be involved inthe first solvation shell of the ions, and most of the saltmolecules will be existing in the form of contact ion pairs (CIP)or solvent-separated ion pairs (SSIP) with the overlappinghydration shells. SSIP species are reported to be thermody-namically predominant in the larger clusters.16 These ion pairswill influence the water hydrogen bonding strongly and alsoenhance the water dipole moment. Under such conditions, theforce constants for the OH---Cl− bonds are even larger than forthe OH---O (water−water) H-bond.6c,17 It is further reportedthat, when the hydration shell of an anion is shared by a cationforming a Y+OH---X−structure (in the present case Y = Na, K,NH4; X = Cl), the cation modifies the properties of theassociated anion−water H-bond.6c As the charge on the cation

Figure 4. (a) Bar diagram showing comparison of isentropiccompressibility (κS) of water and aqueous salt saturated solutions(saturated at 20 °C) as function of temperature. (b) Plots ofcompressibility vs temperature for different salt solutions: (●) NaCl;(■) KCl; and (▲) NH4Cl.

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911715

increases, the effect becomes more pronounced, with increasingpolarization leading to stronger H-bonds. Therefore, thearrangement of water molecules around the ion pairs/contaction pairs of Na+ with Cl− would have stronger order in thehydration shells with respect to the ion pairs/contact ion pairsK+ or NH4

+ with Cl−. Consequently, the water moleculeswould be most strongly anchored in NaCl−water clusters. Thesolubility of the three salts in water as a function of temperature(Figure 1) could be correlated to the temperature stability ofthe corresponding salt−water clusters (Figure 3a−d). In NaClclusters, the dipoles resulting from this local alignment of

solvated water may grow large enough, slowing down theequilibrium of bulk and solvated water, thus minimizing thehydration of additional amounts of NaCl even with the rise intemperature. On the other hand, solvated water moleculesavailable in the outer hydration shell were released from KCl−water and NH4Cl−water clusters with the rise in temperatureas evident from the observed reduction in the cluster sizes(Figure 3d). The released water was utilized for the dissolutionof more salt.There was also good correlation between the solubility

differences among the three salts and the κS differences of their

Figure 5. Gaussian decomposition of the vibration FTIR spectra of (a) water, (b) saturated NaCl−water, (c) saturated KCl−water, and (d) saturatedNH4Cl−water.

Figure 6. (a) Temperature dependence of FTIR absorption spectra of saturated NaCl−water solution in the range 2500−4000 cm−1 and (b) bandarea ratio of strongly bound (icelike) to the band area of the fraction of all weakly H-bonded water molecules (liquidlike).

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911716

saturated solutions, the latter being influenced strongly by the

perturbation of the water hydrogen-bonding network (Figure

4). Recently, it has been shown that the oxygen−oxygenstructure is strongly dependent on the ionic concentration

while it is almost independent of the cation. The hydrogen

bonding is preserved at all concentrations and temperatures.

The main effect of increasing the ionic concentration is the

tendency of the water structure to assume the high-density

Figure 7. 35Cl, 17O, and 1H NMR spectra at variable temperature of aqueous (a) NaCl, (b) KCl, and (c) NH4Cl solutions.

Figure 8. Comparison of 35Cl NMR spectra of aqueous salt saturated solutions with their corresponding solid salts.

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911717

liquid.18 Our κS results though favor these results partially thathigh-density (less compressible) solutions are formed withincrease in salt concentrations, but the influence of cation sizeand charge on water hydrogen bonding cannot be certainlyignored as is evident from different κS values of salt solutions ofsimilar concentrations (Figure 4b). The κS differences could beexplained on the basis of the partial molar volumes and chargedensities of the respective salts. In the case of NH4Cl, theammonium and chloride ions both have partial molar volumesclose to that of water (∼18 cm3 mol−1), and these do notchange appreciably with concentration.19 Hence the perturba-tion of the water structure is small. From the very low values ofκS for NaCl−water solution (Figure 4), we inferred that thesodium ion in had pronounced electrostrictive effect. Due tothe high charge density and comparatively smaller size, the Na+

perturbed the water hydrogen bonding to a greater extent.NaCl−water solutions were thus very rigid with greatly reducedfree volume and, consequently, having little scope for furthersolubilization of salt.The extent of hydrogen-bonding distortion in different

saturated salt solutions is further evident from FTIR results(Figure 5). The electric field of the cation polarizes the O−H---O and O−H---Cl− hydrogen bonds of water molecules adjacentto the cation. The strength of the hydrogen bonds increaseswith increasing polarization and alters O−H stretch vibration ofan H2O molecule solvating the anion.2f The small BAR value(0.38) for NaCl−water compared to that of pure water (1.07)obtained from Figure 5 indicated perturbation of waterhydrogen bonding network by sodium ions to a marked extentin line with the compressibility data. The temperature-dependent studies of BAR indicated that the ratio was virtuallyunaffected by temperature (Figure 6b). The BAR value of 0.71in KCl−water revealed moderate perturbation of waterhydrogen bonding network and, therefore, had some freevolume to accommodate more KCl. In case of NH4Cl−water,the value of BAR was even higher (1.33) than that of purewater. The high hydrophobic hydration and good compatibilityof ammonium ions with water molecules meant that there islittle reduction in free volume of solution. Thus, besides thefactors such as high hydration energy and low lattice energy, thenegligible perturbation of free volume allowed maximumsolubilization of NH4Cl with the rise in temperature.The changes in 35Cl resonance also provided valuable clues

on the nature of the interactions. The high downfield shiftobserved for the saturated aqueous NH4Cl solution withrespect to very dilute standard solution (Figure7c) was theconsequence of larger deshielding of chlorine nucleus under theinfluence of more electropositive NH4

+ cation.14h On the otherhand, the large upfield 35Cl shift with respect to solid state(Figure 8) and comparatively high 17O shift (Figure 7c) wereindicative of weak Cl---H and O---NH4 bridging therebyforming weaker clusters, as was also indicated by DLS resultswhich showed a significant reduction in the cluster size with therise in temperature. The high 35Cl downfield shift observed inthe case of saturated aqueous NaCl solution with respect tosolid state (Figure 8) suggested very strong Cl---H−O−Hbonds whereas, in the case of KCl, the much smaller extent ofdownfield shift was indicative of moderate strength of Cl---H−O−H bonds. Therefore, it was concluded that strong Cl---Hbridges as compared to the other cases were possiblyresponsible for the thermally stable large NaCl−water clustersdetected. The strong connections between water oxygen andNa+ ions and weakened water intermolecular hydrogen

bonding (as inferred from very low 17O and 1H NMR chemicalshifts in saturated NaCl solutions, Figure 7) also contributed tostrengthening of the NaCl−water clusters. When we looked atchanges in 35Cl chemical field in saturated aqueous solution as afunction of temperature (Figure 7), it went marginally upfield(∼3 ppm), indicating some decrease in hydrogen-bondingstrength between Cl− ions and water protons, which is alsoperhaps the cause of marginal rise of solution compressibility ofsaturated NaCl−water solution with the rise in temperature(Figure 4). However, as can be seen from the large downfieldshift (∼49 ppm) with respect to solid state (Figure 8), thedecrease in hydrogen-bonding strength as a function oftemperature was not enough to break the Cl---H−O−Hcontacts involved in the NaCl−water cluster and hence theseremained quite stable in the investigated temperature range.

■ CONCLUSIONSThe marked variations in the aqueous solubilities of NaCl, KCl,and NH4Cl as a function of temperature were correlated withtheir cluster stabilities. The nature of the clusters and theirstabilities were probed by a combination of measurementsincluding dynamic light scattering, compressibility, and IR/NMR spectroscopy. Slow exchange of solvated water moleculeswith bulk water as a consequence of formation of (i) stableNaCl−water clusters, (ii) highly perturbed water hydrogenbonding network with reduced capacity to solvate ions/ionpairs, and (iii) high rigidity of aqueous NaCl saturatedsolutionpreventing penetration of sodium chloride furtherinto the solutionis the probable reason behind the nearlyconstant solubility of sodium chloride as a function oftemperature. As mentioned in the text, this phenomenon isof importance in practical processes such as the separation ofKCl from KCl/NaCl solid mixture. Unravelling the phenom-enon at a more fundamental level would be a logical extensionof the above work. Additionally, the study of the temperaturedependence of salt−water clusters on colligative properties maybe of considerable importance.

■ ASSOCIATED CONTENT*S Supporting InformationHydrodynamic diameter (Dh) distributions plotted against %intensity normalized to unity of NaCl−water clusters (FiguresS1−S3), and density (ρ) and speed of sound (u) data as afunction of temperature (Table S1). This material is availablefree of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: arvind@csmcri.org (A.K.); pkghosh@csmcri.org(P.K.G.). Tel.: +91-278-2567039. Fax: +91-278-2567562.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank the reviewers for their critical review of the work andimportant suggestions which were incorporated in the revisedmanuscript. The authors are thankful to Prof. Nobuo Kimizukaand Dr. Tejwant Singh for temperature-dependent FTIRmeasurements. We also thank the Analytical Discipline andCentralized Instrumental Facility of our Institute for helpfulassistance. CSIR India is gratefully acknowledged forsupporting the research as part of an in-house project.

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911718

■ REFERENCES(1) (a) Bernal, J. D.; Fowler, R. H. J. Chem. Phys. 1933, 1, 515−548.(b) Onsager, L.; Fuoss, R. M. J. Phys. Chem. 1932, 36, 2689−2778.(c) Markus, Y. Chem. Rev. 1988, 88, 1475−1498. (d) Jenkins, H. D.W.; Markus, Y. Chem. Rev. 1995, 95, 2695−2724. (e) Marcus, Y.;Hefter, G. Chem. Rev. 2006, 106, 4585−4621. (f) Ohtali, H.; Randnai,T. Chem. Rev. 1993, 93, 1157−1204. (g) Robinson, R. A.; Stokes, R. H.Electrolyte Solutions: The Measurement and Interpretation of Con-ductance, Chemical Potential and Diffusion in Solutions of SimpleElectrolytes; Academic Press: New York, 1959.(2) (a) Kropman, M. F.; Bakker, H. J. Science 2001, 291, 2118−2120.(b) Max, J.-J.; Chapados, C. J. Chem. Phys. 2001, 115, 2664−2675.and references therein. (c) Cerreta, M. K.; Berglund, K. A. J. Cryst.Growth 1987, 84, 577−587. (d) Dillon, S. R.; Dougherty, R. C. J. Phys.Chem. A 2003, 107, 10217−10224. (e) Li, R.; Jiang, Z.; Yang, H.;Guan, Y. J. Mol. Liq. 2006, 16, 14−20. (f) Bakker, H. J. Chem. Rev.2008, 108, 1456−1473.(3) (a) Neilson, G. W.; Mason, P. E.; Ramos, S.; Sullivan, D. Philos.Trans. R. Soc. London, Ser. A 2001, 359, 1575−1591. (b) Neilson, G.W.; Adya, A. K.; Ansell, S. Annu. Rep. Prog. Chem. C 2002, 98, 273−322.(4) Georgalis, Y.; Kierzek, A. M.; Saenger, W. J. Phys. Chem. B 2000,104, 3405−3406.(5) (a) Hunter, R. J. Zeta Potential in Colloid Science; Academic Press:London, 1981. (b) Miller, J. D.; Yalamanchili, M. R. Langmuir 1992, 8,1464−1469.(6) (a) A Fayer, M. D. Annu. Rev. Phys. Chem. 2009, 60, 21−38.(b) Heisler, I. A.; Meech, S. R. Science 2010, 327, 857−860.(c) Heisler, I. A.; Mazur, K.; Meech, S. R. J. Phys. Chem. B 2011,115, 1863−1873.(7) (a) Lin, Y. S.; Auer, B. M.; Skinner, J. L. J. Chem. Phys. 2009, 131,144511−144513. (b) Laage, D.; Hynes, J. T. Proc. Natl. Acad. Sci.U.S.A. 2007, 104, 11167−11172.(8) (a) Fuoss, R. M. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 34−38.(b) Bjerrum, N. Niels Bjerrum Selected Papers; Einar Munksgaard:Copenhagen, 1949; p 108. (c) Evans, R. Adv. Phys. 1979, 28, 143−200.(9) (a) Choen-Adad, R.; Lorimer, J. W. Solubility Data Series 47;Pergamon Press: Oxford, UK, 1991; (b) Franks, F., Ed. The Physicsand Physical Chemistry of Water; Plenum Press: New York, 1972; Vol.1.(10) (a) Ghosh, P. K.; Langalia, K. J.; Gandhi, M. R.; Dave, R. H.;Joshi, H. L.; Vohra, R. N.; Mohandas, V. P.; Daga, S. L.; Halder, K.;Deraiya, H. H. et al. Process for recovery of sulphate of potash. US Patent7041268, 2006; (b) Blasdale, W. C. J. Ind. Eng. Chem. 1918, 10, 344−347.(11) (a) Passynski, A. Acta Physicochim. 1938, 8, 385−418.(b) Mathieson, J. G.; Conway, B. E. J. Solution Chem. 1974, 3, 455−477. (c) Gucker, F. T.; Stubley, D.; Hill, D. J. J. Chem. Thermodyn.1975, 7, 865−873. (d) Onori, G. Acoustics Lett. 1990, 14, 7−15.(12) (a) Osborne, N. S.; Stimson, H. F.; Ginnings, D. C. J. Res. Natl.Bur. Stand. 1939, 23, 197−260. (b) Bett, K. E.; Cappi, J. B. Nature1965, 207, 620−621. (c) Bridgman, P. W. Proc. Am. Acad. Arts Sci.1911−12, 47, 441. (d) Dougherty, R. C.; Howard, L. N. J. Chem. Phys.1998, 109, 7379−7393. (e) Wernet, Ph.; Nordlund, D.; Bergmann, U.;Cavalleri, M.; Odelius, M.; Ogasawara, H.; Naslund, L. Å.; Hirsch, T.K.; Ojamae, L.; Glatzel, P.; et al. Science 2004, 304, 995−999.(f) Alphonse, N. K.; Dillon, S. R.; Dougherty, R. C.; Galligan, D. K.;Howard, L. N. J. Phys. Chem. A 2006, 110, 7577−7580.(13) (a) Walrafen, G. E. In Water, A Comprehensive Treatise; Franks,F., Ed.; Plenum: New York, 1972; Vol. I, Chapter 5. (b) Narten, A. H.;Levy, H. A. In Water, A Comprehensive Treatise; Franks, F., Ed.;Plenum: New York, 1972; Vol. I, Chapter 8. (c) Du, Q.; Superfine, R.;Freysz, E.; Shen, Y. R. Phys. Rev. Lett. 1993, 70, 2313−2316.(c) Kusalik, P. G.; Svishchev, I. M. Science 1994, 265, 1219−21.(d) Tanaka, H. Chem. Phys. Lett. 1998, 282, 133−138. (e) Green, J. L.;Lacey, A. R.; Sceats, M. G. J. Phys. Chem. 1986, 90, 3958−3964.(f) Hare, D. E.; Sorensen, C. M. J. Chem. Phys. 1992, 96, 13−23.(g) Rowland, B.; Fisher, M.; Delvin, J. P. J. Chem. Phys. 1991, 95,

1378−85. (h) Gopalakrishnan, S.; Jungwirth, P.; Tobias, D. J.; Allen,H. C. J. Phys. Chem. B 2005, 109, 8861−8872.(14) (a) Reuben, J. J. Am. Chem. Soc. 1969, 91, 5725−5729.(b) Svishchev, I. M.; Kusalik, P. G. J. Am. Chem. Soc. 1993, 115, 8270−8274. (c) Li, R.; Jiang, Z.; Shi, S.; Yang, H. J. Mol. Struct. 2003, 645,69−75. (d) Irish, D. E. In Ionic interactions; Petruchi, S., Ed.; AcademicPress: New York, 1971; Vol. II, p 187. (e) Bloor, E. G.; Kidd, R. G.Can. J. Chem. 1968, 46, 3425−3430. (f) Erlich, R. H.; Popov, A. I. J.Am. Chem. Soc. 1971, 93, 5620−5623. (g) Miller, A. G.; Franz, J. A.;Macklin, J. W. J. Phys. Chem. 1985, 89, 1190−1193. (h) Weeding, T.L.; Veeman, W. S. J. Chem. Soc., Chem. Commun. 1989, 945−946.(15) Degreve, L.; da Silva, F. L. J. Chem. Phys. 1999, 110, 3070−3079.(16) Peslherbe, G. H.; Ladanyi, B. M.; Hynes, J. T. J. Phys. Chem. A2000, 104, 4533−4548.(17) (a) Hiraoka, K.; Mizuse, S.; Yamabe, S. J. Phys. Chem. 1988, 92,3943−3952. (b) Heuft, J. M.; Meijer, E. J. J. Chem. Phys. 2003, 119,11788−11791.(18) Gallo, P.; Corradinib, D.; Roverea, M. Phys. Chem. Chem. Phys.2011, 13, 19814−19822.(19) Millero, F. J. Chem. Rev. 1971, 71, 147−176.

The Journal of Physical Chemistry B Article

dx.doi.org/10.1021/jp307261g | J. Phys. Chem. B 2012, 116, 11712−1171911719