Complex Dynamical Aspects of Organic Electrolyte ... - ACS Publications

Article pubs.acs.org/JPCB

Complex Dynamical Aspects of Organic Electrolyte Solutions Francesca Palombo,† Paola Sassi,‡ Marco Paolantoni,*,‡ Chiara Barontini,‡ Assunta Morresi,‡ and Maria Grazia Giorgini§ †

School of Physics, University of Exeter, EX4 4QF Exeter, United Kingdom Department of Chemistry, University of Perugia, Via Elce di Sotto 8, I-06123 Perugia, Italy § Department of Industrial Chemistry - ‘Toso Montanari’, University of Bologna, Viale del Risorgimento 4, I-40136 Bologna, Italy ‡

ABSTRACT: Molecular dynamics of acetone−alkali metal halide (LiBr, LiI) solutions were investigated using depolarized Rayleigh scattering (DRS) and low-frequency Raman spectroscopy in the frequency range from ∼0.5 to 200 cm−1 (∼20 GHz to 6 THz). These experiments probe fast dynamical fluctuations of the polarizability anisotropy at picosecond and sub-picosecond time scales that are mainly driven by acetone orientational dynamics. Two distinct contributions were revealed: a fast process (units of picosecond, ps) related to the essentially unperturbed bulk solvent and a slow one (tens of ps) assigned to acetone molecules forming Li+ solvation shells, decelerated by the motional constraint imposed by the cation. The increase of LiBr and LiI concentration significantly slows down the overall solvent relaxation as a consequence of the increased fraction of acetone molecules involved in the ion solvation shells. The global retardation is larger in LiI than LiBr solutions consistently with viscosity trends. This is explained in terms of ion association (at least ion pairing) more favorably promoted by Br− than I−, with reduced Li+−acetone interactions in LiBr than LiI solutions. Anion-induced modulation of the Li+···OC contacts, largely responsible for electrostriction phenomena, also affects the reduced THz-Raman spectral density, ascribed to ultrafast librational motions of acetone molecules. Overall, these findings enlighten the interplay between ion−dipole and ion−ion interactions on the fast solvation dynamics in electrolyte solutions of a typical polar aprotic solvent.

1. INTRODUCTION Ion solvation plays a key role in many chemical, technological, and biological processes, since it influences electric conductivity, diffusion, and other transport phenomena in condensed phases.1 While hydration processes modulating the microscopic structure and dynamics of aqueous electrolyte solutions have been extensively studied, less attention has generally been paid to ion solvation within organic solvents. Acetone (AC) is a dipolar (μ = 3.1 D)2 aprotic liquid with intermediate solvating power and donor strength (relative permittivity εr = 20.6, donor number DN = 17.0 kcal/mol),3 which makes it a suitable candidate for studying ion solvation and association phenomena. With its carbonyl group relevant to charge migration mechanisms including ion transport across cell membranes,4 AC can be considered to some extent a model for complex amide-bond-containing molecules, namely, proteins or ionophores. In the context of ion-binding proteins such as those forming ion channels, Br− and I− anions are strong denaturants, since they “salt in” the peptide group, thus showing a preferential interaction with the unfolded rather than native form of a protein.5 The Li+ cation is also relevant for its applications in the technological (rechargeable batteries6) and pharmacological fields (psychotherapeutic drugs7). The interaction of Li+ with the carbonyl group of proteins has been examined with a focus on its configuration related to protein structure and folding.8 Lithium salts, mainly halides and perchlorates, dissolved in dipolar aprotic solvents have been used in various studies aimed at investigating how the ions © XXXX American Chemical Society

affect the structure of the liquid at a microscopic level. Ion solvation and association in electrolyte solutions have been extensively analyzed in the past by IR, Raman, and NMR spectroscopy.9−12 The interest in this subject has since expanded to comprise the investigation of fine spectroscopic details of these complex systems. Among them, the noncoincidence effect (NCE) of the CO stretching Raman band in AC solutions of alkali metal halides and perchlorates was explored.13−16 With the support of ab initio calculations, the NCE could be interpreted along two points of view: as the effect of intermolecular vibrational coupling of CO oscillators pointing to the same central ion (Mm+) and as the separation of the in-phase and out-of-phase CO stretching bands in tetrahedral (or octahedral) ACn Mm+ clusters.15 Buchner and colleagues17 used dielectric relaxation spectroscopy to study the dynamics of a large variety of aqueous and organic electrolyte solutions. In particular, dielectric spectra of acetonitrile electrolyte (including LiBr) solutions revealed the existence of ion pairs even in low salt concentration solutions.18 Time-dependent fluorescence Stokes shift (TDFSS) spectroscopy has been widely employed to study the solvation dynamics taking place after rapid light-induced charge redistribution on a probe chromophore, thus giving relevant information on the relaxation features of polar solvents.19−21 In Received: July 5, 2013 Revised: November 18, 2013

A

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

2. MATERIALS AND METHODS 2.1. Materials. Spectroscopic grade AC (≥99.9% pure) and anhydrous LiBr (≥99.0% pure) and LiI (≥98.0% pure) were Sigma-Aldrich products. Solutions of different salt concentration, up to near saturation, were prepared in a drybox under nitrogen atmosphere and used for the experiments reported here. 2.2. Measurements. The density (ρ) and shear viscosity (η) of the solutions were measured at different concentrations and temperatures applying previously described procedures32 (Table 1).

these experiments, short-range chromophore−solvent interactions may influence the observed relaxation especially at subpicosecond time scales.20 It is worth noting that the ultrafast solvation component is believed to be significant in determining the ionic mobility.22 TDFSS experiments performed in ionic non-aqueous solutions23 evidenced well separated relaxation domains: a slow process on a time scale of up to nanoseconds related to ion mobility and a faster picosecond process due to the ion-perturbed solvent dynamics. Solvation dynamics in electrolyte solutions was further addressed by detailed theoretical and computational studies;24,25 while the former24 recognizes the inertial character of the very fast component, consistently with the pure solvent case, the latter25 confirms the onset, at long time scales, of ion− solvent dynamical correlations associated with the ion exchange between the ion solvation shell and the solution bulk. Recently, dynamical aspects of organic electrolyte solutions have been investigated using advanced ultrafast pump−probe methods such as the optical Kerr effect (OKE)26 and 2D-IR spectroscopy.27,28 While earlier OKE experiments have revealed29 no change in the fast relaxation dynamics of electrolyte (NaI and LiClO4) acetonitrile and ethyl acetate solutions compared to the pure solvents, noticeable effects were observed in a very recent investigation on formamide ionic solutions.30 Depolarized Rayleigh scattering (DRS) and lowfrequency Raman spectroscopy are especially relevant for the purpose of these studies. The measured response, corrected for the thermal occupation factor, represents the spectral counterpart of the time-domain response of OKE and relates to the relaxation dynamics of the anisotropic components of the total polarizability of the system.31,32 In our previous work,32 DRS and low-frequency Raman scattering have been used to study the molecular dynamics of pure liquid AC at different temperatures. Insight into the diffusive orientational relaxation and intermolecular, librational dynamics of this dipolar aprotic liquid has been obtained, hence forming the basis for the present study. Here, we have investigated the molecular mobility in AC alkali halide (LiBr and LiI) solutions at various concentrations, using DRS−lowfrequency Raman spectra and their susceptibility χ″(ν̃) representation. The spectral response in these experiments is expected to be dominated by the orientational dynamics of anisotropic AC molecules, considering that spherical monatomic ions contribute to the DRS signal only through interaction-induced effects,33,34 which are generally less important.33−35 In principle, in the case of a presence of long living ion−ion or ion−solvent aggregates, direct contributions originating from reorientations of ionic clusters must be taken into account. However, it will be shown that, even if ion−ion and AC−ion aggregation play a relevant role in these electrolyte solutions, DRS−low-frequency Raman experiments in the 0−200 cm−1 frequency range essentially probe AC mobility and its modification under the influence of the ionic species, providing selective information on the effect of ions on the solvent dynamics. This latter is found to critically depend on the competitive balance between ion−ion and ion−solvent interactions. Dynamical information extracted from the spectra is relevant to elucidate the effects of ion solvation and association on transport and other macroscopic properties in electrolyte solutions.

Table 1. Density (ρ) and Viscosity (η) of Lithium Bromide and Lithium Iodide Solutions in AC, at 25 °C acetone LiBr

LiI

x

c (M)

ρ (g mL−1)

η (cP)

Eaa (kcal mol−1)

0.01 0.03 0.05 0.08 0.005 0.01 0.05 0.08

0.14 0.42 0.72 1.17 0.069 0.14 0.74 1.24

0.7839(1) 0.7966(1) 0.8259(1) 0.8512(1) 0.8833(1) 0.7953(1) 0.8034(1) 0.9012(1) 0.9775(1)

0.297(1) 0.346(1) 0.401(1) 0.512(2) 0.654(2) 0.324(1) 0.385(1) 0.754(2) 1.450(5)

1.68(1) 1.79(2) 1.84(2) 1.88(2) 2.26(3) 1.69(1) 1.83(1) 2.19(1) 2.60(1)

The activation energy for the viscous flow, Ea, was derived from an Arrhenius fit of the viscosity data in the temperature range 15−45 °C (Figure 1c and d). Ea for liquid AC is close to previous data: 1.67(3)74 and 1.7075 kcal mol−1. Numbers in brackets indicate the error on the last digit of each value. a

DRS and low-frequency Raman spectroscopic measurements were performed exciting the sample with an argon ion laser (λ = 514.5 nm, 500 mW) and analyzing the scattered radiation, collected at 90° and in the HV polarization configuration, with a U1000 ISA Jobin-Yvon double monochromator. Details of the experimental setup and procedures applied have been reported in a previous work.32 DRS spectra were acquired over the range −30 to 30 cm−1 with a resolution of 0.4 cm−1, while anisotropic Raman spectra were measured between 1 and 200 cm−1 with a 0.8 cm−1 resolution. Each spectrum was obtained by averaging over four successive scans. Measurements were performed at constant temperature, 25.0 (±0.1) °C, using freshly prepared solutions of different mole fraction, x. The solutions were filtered through a 0.20 μm Sartorius MinisartSRP 15 PTFE filter and transferred into an optical quality fused silica cell. For an intermediate composition of the LiBr solution, x = 0.05, spectra were also measured at variable temperature, in the range 5.5−44.6 °C (with a 10 °C interval).

3. RESULTS AND DISCUSSION 3.1. Density and Viscosity. Figure 1 shows the evolution of density (a, b) and viscosity (c, d) of LiBr−AC and LiI−AC solutions as a function of temperature; corresponding values at 25 °C are listed in Table 1. A remarkable increase of the solution shear viscosity (at every temperature) is observed with increasing salt concentration. The variation is stronger with LiI than LiBr. The apparent activation energy for the viscous flow (Table 1) increases with increasing concentration, with values systematically higher for LiI than LiBr solutions. From the measured densities, the total molar concentration n, expressed as moles of solute plus solvent per liter of solution, has been calculated. Figure 2a shows the evolution of n for the B

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

Figure 1. (a, b) Plot of the density of (a) LiBr−AC and (b) LiI−AC solutions versus temperature, in the range from 5 to 45 °C. The error bar is within the symbol size. (c, d) Semilogarithmic Arrhenius plot of the shear viscosity of the same solutions in the temperature range 15−45 °C. For each Arrhenius fit (R2 ≥ 0.999; dashed line), the resulting activation energy for the viscous flow is reported in Table 1.

Figure 2. Evolution of the (a) total molar concentration and (b) relative shear viscosity of LiBr and LiI−AC solutions as a function of salt mole fraction, in the range up to x = 0.08 (near saturation), at 25 °C.

two electrolytes at 25 °C as a function of the mole fraction x: n increases with increasing concentration, more for LiI than LiBr. The observed trend might be largely influenced by electrostriction effects regulating the changes of free volume in the solution: smaller volume for the solvent in the ion solvation shell than in the bulk.36 Figure 2b illustrates the mole fraction dependence of the relative viscosity, ηr (=η/η0, with η indicating the viscosity of the solution at a given x and η0 the viscosity of the pure solvent), at 25 °C. Also, in this case, the changes due to LiI are remarkably larger than those due to LiBr; a similar situation holds at the other temperatures. Viscosity data are often interpreted in the framework of the Jones−Dole equation.37−40 For moderate electrolyte concentrations (c
LiI.42 More recent C

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

a convolution of the “true” spectrum with the instrumental function:

investigations of electrolytes in low-permittivity solvents, including AC, have confirmed the presence of ion pairs even at low concentrations,18,27,41,43,44 as well as a stronger aggregation tendency for LiBr than LiI.41,43,45 3.2. Relaxation Dynamics. Figure 3 shows the evolution of the DRS spectrum of LiBr−AC solutions as a function of salt

IHV(ν)̃ = G(ν)̃ ⊗ [K pδ(ν)̃ + a1L1(ν)̃ + a 2L 2(ν)] ̃

(1)

where δ(ν̃) represents the spurious polarized contribution and Kp, a1, and a2 are the intensities of the components.46 Since the effect of G(ν̃) (full width at half height, ΓG = 0.40 cm−1) on both L1(ν̃) and L2(ν̃) is negligible (given their width, Γ1 = 9.3 and Γ2 = 45 cm−1 for liquid AC at 25 °C,32 more than 10 times larger than ΓG), eq 1 can be rewritten as IHV(ν)̃ ≈ K pG(ν)̃ + a1L1(ν)̃ + a 2L 2(ν)̃

(2)

In this model, each Lorentzian accounts for an exponential relaxation process. Hence, two solvent relaxation times are derived: τ1 = 1.1 ps and τ2 = 0.2 ps for pure AC at 25 °C.32 The datum for τ1 is close to the corresponding OKE result at 24 °C, τ1 = 1.21 ps.2 The slow time τ1 is assigned to the molecular diffusive reorientation of liquid AC,32 while the fast subpicosecond time τ2 is tentatively ascribed to a local (“cage”) structural relaxation, in analogy with the case of other nonbonded liquids.47−49 As already pointed out,32 the molecular origin of this almost universally detected sub-picosecond process is still unclear and its exponential character can also be questionable. Nevertheless, beyond any specific assignment, the DRS spectrum of liquid AC can be mainly ascribed to anisotropic polarizability fluctuations of reorientational origin,32,35 even if translational contributions cannot be totally neglected. TDFSS experiments,19,21 which mainly probe the orientational relaxation of a polar solvent, gave acetone solvation relaxation times of 0.187 and 1.09 ps, with a resulting average solvation time of 0.58 ps in quantitative agreement with DRS results. Note that, in principle, different observables are probed by the two techniques.50 Fit analysis of the DRS spectrum of the electrolyte solutions required an additional narrower Lorentzian, L0′(ν̃), with eq 2 thus becoming

Figure 3. Semilog plot of the DRS spectrum of LiBr−AC solution as a function of composition, in the whole solubility range, at 25 °C. Spectra were normalized to the high-frequency wings.

mole fraction at 25.0 °C. After normalization to the higher frequency tails of the distribution, a progressive increase of intensity for the central part with increasing salt concentration is observed. A similar trend is observed for the LiI−AC solution. To obtain quantitative information from these spectra, a fit analysis was applied (Figure 4) based on previous work on the pure solvent (see inset of Figure 4).32 In the range −30 to 30 cm−1, the DRS spectrum of liquid AC is well reproduced using a sum of two Lorentzians, L1(ν̃) and L2(ν̃), with L2(ν̃) broader than L1(ν̃), and a narrow Gaussian G(ν̃) for the instrument response.32 The measured spectrum IHV(ν̃) can be expressed as

IHV(ν)̃ ≈ K pG(ν)̃ + a0′L0′(ν)̃ + a1L1(ν)̃ + a 2L 2(ν)̃

(3)

For the new component L0′(ν̃), which is affected by experimental resolution, the “true” width (Γ0) was derived from a numerical deconvolution51 based on the assumption L0′(ν̃) ≈ [G(ν̃) ⊗ L0(ν̃)]. For all investigated solutions, eq 3 was a suitable fit function to the DRS profile over the range −30 to 30 cm−1. An example of fit results is shown in Figure 4, while all parameters are listed in Table 2. To evaluate the effect induced by electrolytes on the overall relaxation of the system, an average relaxation time ⟨τ⟩ weighted on the intensity of each component was calculated (Table 2). The relative average relaxation time, ⟨τ⟩r = ⟨τ⟩/⟨τ⟩0 (average time of the solution scaled to the value of the pure solvent), was also obtained and plotted versus composition in Figure 5. For both systems, ⟨τ⟩r sensibly increases with increasing concentration, but more for LiI than LiBr, consistently with the ηr data in Figure 2b. It is possible to rationalize the results obtained from the analysis of DRS spectra of AC electrolytic solutions as follows. The measured profiles essentially reveal the effect of electrolytes on the dynamics of liquid AC, specifically on its orientational rearrangements. Solvent relaxations are more effectively retarded by LiI than LiBr salt, thus suggesting stronger ion−AC interactions in LiI solution. IR and Raman spectra of AC solutions of Li+ salts reveal a dominating cation effect on liquid AC reorganization.12,15 This finding can be

Figure 4. Fit results for the DRS spectrum of a LiBr−AC solution with x = 0.08, at 25 °C, using a sum of three Lorentzians and an instrumental Gaussian. Inset: fit analysis of the pure AC spectrum including two Lorentzians and an instrumental Gaussian. Related parameters are listed in Table 2. D

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

Table 2. Fit Parameters of the DRS Spectrum of LiBr and LiI−AC Solutions, at 25 °Ca x acetone LiBr

LiI

0.01 0.03 0.05 0.08 0.005 0.01 0.05 0.08

A0 0.05 0.08 0.13 0.15 0.03 0.11 0.21 0.26

τ0 (ps)

A1

τ1 (ps)

A2

τ2 (ps)

⟨τ⟩ (ps)

3.5 4.4 6.6 18 6.8 7 15 19

0.49 0.49 0.45 0.39 0.39 0.47 0.43 0.35 0.31

1.1 1.0 1.1 1.2 1.5 1.3 1.1 1.7 1.6

0.51 0.46 0.47 0.48 0.46 0.50 0.46 0.44 0.43

0.24 0.19 0.21 0.21 0.24 0.23 0.20 0.25 0.23

0.68 0.77 0.94 1.4 2.6 0.87 1.3 3.2 4.1

a

Values are reported as follows: A, normalized intensity of the components (Ai = ai/Σan, with ai being the integral area of the i-component over the range −30 to 30 cm−1); τ, relaxation time derived from the Lorentzian bandwidth (τi = 1/2πcΓi); ⟨τ⟩, average relaxation time (⟨τ⟩ = ΣAnτn). Spectra were reproduced to R2 ≥ 0.999. A deconvolution of the instrumental slit function was applied to derive τ0, according to the method in ref 51.

aggregates (e.g., ion pairs) will be the one containing the smaller number of dynamically retarded solvation molecules. The data shown in Figure 5 are then indicative of a lower degree of ion association and correspondingly higher number of Li+−AC contacts in LiI than LiBr solution. A similar picture has been derived for the LiX−methanol system using mass spectrometry, 41 consistently with an increased Li + −X − association propensity going from I− to Br−.42,43,45 Therefore, the retardation of solvent dynamics selectively probed by DRS is largely determined by Li+−AC interactions and by their interplay with counterion effects through ion association. The viscosity trends in Figure 2b can thus be justified on the basis of the same molecular picture. 3.3. Spectral Assignment. We attempted a more specific assignment of the observed relaxation processes. Data in Table 2 denote an increase in L0(ν̃) intensity, at the expenses of the other contributionsmainly L1(ν̃), with increasing LiBr/I concentration. A marked slowdown of the corresponding τ0 process is also observed. On the other hand, the time constants of the faster contributions, also detected for the neat solvent,32 are only marginally affected. This enables the assignment of L0(ν̃) to the relaxation of solvating AC molecules, while the other components relate to dynamically unperturbed (bulk) AC. Distinct bulk and solvation relaxation processes have already been revealed by extended light scattering spectra of biologically relevant aqueous solutions (with solutes including sugars,56 peptides,57 and protein58); results were also supported by MD simulations.59 This evidence of distinct “shell” and “bulk” contributions implies their dynamics being faster than the mutual solvent exchange between these two environments (i.e., slow exchange limit). The same must hold for the AC alkali metal halide solutions studied here. The slower contribution L0(ν̃) is systematically more intense for LiI than LiBr, consistently with a larger fraction of AC molecules affected by the solute in the former solution. It is worth remarking that the determination of L0(ν̃) is limited by instrumental resolution such that inaccuracies may be beyond the statistical errors of the fit analysis considered in Table 2, especially for the higher concentrations (slower dynamics, i.e., smaller bandwidth). However, sugar relaxation processes on a time scale of tens of picoseconds determined with this setup were found to be accurate,46 based on a comparison with higher resolution interferometric measurements.60,61 Thus, the presence of a slow process related to solvent mobility up to 1 order of magnitude slower than the bulk can be reasonably considered. The τ0 relaxation becomes slower with increasing salt concentration, with a more pronounced change in LiI−AC

Figure 5. Plot of the relative average relaxation time of LiBr and LiI− AC solutions versus salt mole fraction, in the range up to x = 0.08 (near saturation), at 25 °C.

explained on the basis of the remarkable charge density of Li+ and on the chemical nature of the AC carbonyl group, whose negatively charged terminus is prone to interactions with the cation (anion−AC interactions instead involve the partially hinderedby two methyl groupspositive side of the CO). The absence of an electron acceptor group in AC has been inferred as a possible cause for weak anion−AC interactions.52 Even for the more polar ethylene carbonate (μ = 4.8 D), the calculated entropy and enthalpy solvation changes are significantly more negative for the relatively large K+ than for Br− and I−.53 Hence, Li+ is expected to play a major role in determining structural and dynamical variations on AC.41,43,44 Moreover, changes induced by the anions, if present, should decrease going from Br− to I− due to the concomitant reduction of ion−solvent (i.e., ion−dipole) interactions,53,54 while an opposite trend is observed here (Figures 2b and 5). This can be explained on the basis of the interplay of Li+−AC interactions and Li+−X− pairing, along with a different association tendency for Br− and I−.41,42 With increasing salt concentration, some AC molecules in the solvation shell of Li+ (hence with restricted mobility) would be replaced by the counterion X− and then released to the bulk (with larger mobility). This interpretation is supported by previous neutron scattering investigations on LiBr−acetonitrile and LiBr−AC solutions at similar concentrations, indicating that on average Li+ is surrounded by one Br− and three solvent molecules.44,55 Thus, at any given mole fraction, the solution with more ion E

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

solvation shell exchanges should be mentioned. For LiBr in acetonitrile (another dipolar aprotic liquid), the ion-pair rotational relaxation time was derived by dielectric relaxation spectroscopy, ∼80 ps (more than 20 times slower than the solvent relaxation, 3.4 ps), and found to be independent of solute concentration.18 Also, the orientational relaxation time of LiSCN ion pairs in N,N-dimethylformamide estimated by ultrafast IR spectroscopy is ∼60 ps,27 and both dielectric and ultrafast IR spectroscopies suggest that ion pair association/ dissociation is slower.18,27 A detailed 2D infrared study on a LiNCS ion pair and ion-pair dimer association in polar solvents provided times of chemical exchange between them (ion pair and ion-pair dimer) larger than 20 ps and an ion-pair dimer dissociation time larger than 40 ps.28 Concerning the dynamical exchange of solvent molecules between bulk and solvation shells, typical residence times of aprotic solvents around Li+ are of the order of 0.1−2 ns,54 while the reorientation of stable Li+−AC4 clusters15 is likely DRS inactive for symmetry reasons. On the basis of that, the assignment of τ0 to a dynamical rearrangement of AC molecules temporarily localized in the cation solvation shell seems reasonable. Note that, for LiNCS ion pairs, CN-stretching spectral diffusion components at picosecond time scales measured by 2D infrared spectroscopy were mainly assigned to structural reorganization of Li+coordinated solvent molecules.28 3.4. Ultrafast Dynamics. While the central part of the DRS spectrum of simple liquids mainly originates from relaxation processes at picoseconds, the high frequency portion extending up to ∼200 cm−1 derives from ultrafast intermolecular dynamics generally associated to librations (hindered rotations) within solvent cages.32,34,63−65 Translations activated by an interaction-induced (II) mechanism may also contribute to the fluctuations of anisotropic polarizability at these frequencies.33,34,59 Possible coupling between molecular (orientational) and II (orientational and translational) terms complicates the interpretation.31,65−67 This spectral portion is conveniently analyzed using the formalism of the imaginary part of the dynamical susceptibility, χ″(ν̃), as already applied for the neat solvent.32 Here, on the basis of experimental trends, the broad distribution centered at around 50 cm−1 was mainly attributed to molecular librations, in line with previous experimental and computational works suggesting that, for anisotropic molecules, orientational dynamics is the prevailing contribution even in this spectral region.35,66,67 We note that the II term may also depend on molecular orientations.34,67 In the following, we assume that, also for the studied mixtures, the AC molecular response dominates the measured spectrum, neglecting possible ion−AC II contributions. Figure 7 shows the evolution of the normalized χ″(ν̃) of LiBr−AC solution as a function of composition, up to near saturation. Progressive changes are revealed over a broad region above 20 cm−1 (Figure 7a), namely, an intensity increase at 80−150 cm−1, and toward very low frequencies ( n(LiBr) toward higher x (Figure 2a) can be mainly ascribed to the larger electrostriction effects in the former system owing to minor ion association degree. The rescaling observed in Figure 11 is thus consistent with the assignment of the rRSDs to AC librations probing cation−AC interactions. These forces are expected to play a major role in determining both volumetric and librational trends. On the other hand, halide anions contribute only indirectly by modulating Li+−AC interactions through ion association.

4. CONCLUSIONS Ion solvation through ion−carbonyl interaction is crucial to a number of processes in the technological, biological, and pharmaceutical fields. DRS and low-frequency Raman spectra of LiBr−AC and LiI−AC solutions up to near saturation were measured and deeply analyzed. Spectra revealed changes in the anisotropic polarizability response corresponding to the picosecond orientational relaxation and ultrafast librations of acetone. Contributions from the solute mobility are not significant here, such that ion solvation effects can be suitably investigated. Both LiBr and LiI added to liquid AC caused a slowdown in the molecular reorientation and a parallel increase in viscosity owing to ion−dipole interactions. Li+, having a strong charge density compared to the bigger Br− and I−, plays a major role in the perturbation of solvent dynamics. However, ion−dipole interactions are modulated by ion pairing (and/or higher association) to an anion-dependent extent. The picosecond orientational relaxation is slower in LiI than LiBr solution, indicating stronger cation solvation in the former system and larger ion association in the latter. The interplay between Li+− AC and Li+−X− interactions is also evident through the trend in relative viscosity of the solutions, suggesting a major effect of ion−AC contacts.

Figure 10. Comparison between normalized rRSDs of LiBr and LiI− AC solutions with x = 0.08 derived from fit analysis of corresponding χ″(ν̃) at 25 °C.

profile is slightly blue-shifted consistently with a higher number of Li+···OC contacts than in LiBr solution. An average librational frequency, expressed as first moment M1 of the asymmetric multi-Gaussian spectrum (G1 + G2 + G3), was also calculated.32 Its value for pure liquid AC at different temperatures has been found to correlate with the macroscopic H

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

Lithium, and Magnesium Perchlorates in Acetone at 25°C. Russ. J. Phys. Chem. 1975, 49, 76−79. (11) Cahen, Y. M.; Handy, P. R.; Roach, E. T.; Popov, A. I. Spectroscopic Studies of Ionic Solvation. XVI. Lithium-7 and Chlorine-35 Nuclear Magnetic Resonance Studies in Various Solvents. J. Phys. Chem. 1975, 79, 80−85. (12) Popov, A. I.; Wong, M. K.; McKinney, W. J. Spectroscopic Studies of Ionic Solvation. VIII. Alkali Metal Salts in Acetone Solutions. J. Phys. Chem. 1971, 75, 56−61. (13) Bukowska, J. Intermolecular Coupling of the CO Stretching Vibrations in Electrolyte Solutions of Carbonyl Compounds. J. Mol. Struct. 1986, 143, 309−312. (14) Kecki, Z.; Sokolowska, A. Crossing of Anisotropic and Isotropic Raman Components in the Intermolecular Resonance Coupling of Vibrations. IILiClO4 and LiI Solutions in Acetone. J. Raman Spectrosc. 1994, 25, 723−726. (15) Giorgini, M. G.; Torii, H.; Musso, M.; Venditti, G. Influence of Ions on the Structural Organization of Dipolar Liquids Probed by the Noncoincidence Effect: Experimental and Quantum Chemical Results. J. Phys. Chem. B 2008, 112, 7506−7514. (16) Giorgini, M. G.; Torii, H.; Musso, M. The Influence of Alkaline Earth Ions on the Structural Organization of Acetone Probed by the Noncoincidence Effect of the ν(CO) Band: Experimental and Quantum Chemical Results. Phys. Chem. Chem. Phys. 2010, 12, 183− 192. (17) Buchner, R.; Hefter, G. Interactions and Dynamics in Electrolyte Solutions by Dielectric Spectroscopy. Phys. Chem. Chem. Phys. 2009, 11, 8984−8999. (18) Barthel, J.; Kleebauer, M.; Buchner, R. Dielectric Relaxation of Electrolyte Solutions in Acetonitrile. J. Solution Chem. 1995, 24, 1−17. (19) Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. Subpicosecond Measurements of Polar Solvation Dynamics: Coumarin 153 Revisited. J. Phys. Chem. 1995, 99, 17311−17337. (20) Biswas, R.; Bagchi, B. Solvation Dynamics in Nonassociated Polar Solvents. J. Phys. Chem. A 1999, 103, 2495−2500. (21) Bagchi, B.; Jana, B. Solvation Dynamics in Dipolar Liquids. Chem. Soc. Rev. 2010, 39, 1936−1954. (22) Bagchi, B.; Biswas, R. Ionic Mobility and Ultrafast Solvation: Control of a Slow Phenomenon by Fast Dynamics. Acc. Chem. Res. 1998, 31, 181−187. (23) Chapman, C. F.; Maroncelli, M. Fluorescence Studies of Solvation and Solvation Dynamics in Ionic Solutions. J. Phys. Chem. 1991, 95, 9095−9114. (24) Chandra, A.; Patey, G. N. Solvation Dynamics in Electrolyte Solutions. J. Chem. Phys. 1994, 100, 1552−1558. (25) Neria, E.; Nitzan, A. Numerical Simulations of Solvation Dynamics in Electrolyte Solutions. J. Chem. Phys. 1994, 100, 3855− 3868. (26) Fujisawa, T.; Nishikawa, K.; Shirota, H. Comparison of Interionic/Intermolecular Vibrational Dynamics between Ionic Liquids and Concentrated Electrolyte Solutions. J. Chem. Phys. 2009, 131, 244519. (27) Lee, K.-K.; Park, K.-H.; Kwon, D.; Choi, J.-H.; Son, H.; Park, S.; Cho, M. Ion-Pairing Dynamics of Li+ and SCN− in Dimethylformamide Solution: Chemical Exchange Two-Dimensional Infrared Spectroscopy. J. Chem. Phys. 2011, 134, 064506. (28) Ji, M.; Hartsock, R. W.; Sun, Z.; Gaffney, K. J. Interdependence of Conformational and Chemical Reaction Dynamics during Ion Assembly in Polar Solvents. J. Phys. Chem. B 2011, 115, 11399−11408. (29) Deuel, H. P.; Cong, P.; Simon, J. D. Probing the Effects of Electrolytes on Intermolecular Interactions in Solution by OpticalHeterodyne Detected Raman-Induced Kerr Effect Spectroscopy (OHD-RIKESH). J. Raman Spectrosc. 1995, 26, 523−526. (30) Palombo, F.; Meech, S. R. Dynamics of Formamide Ionic Solutions Investigated by Ultrafast Optical Kerr Effect. J. Phys. Chem. B 2012, 116, 13481−13489. (31) Isegawa, M.; Kato, S. Electronic Polarization Effect on LowFrequency Infrared and Raman Spectra of Aprotic Solvent: Molecular Dynamics Simulation Study with Charge Response Kernel by Second

Two distinct solvent relaxation terms were identified. The slower (tens of ps) was assigned to Li+-solvating AC, with partly diminished mobility, and the fast one (units of ps) to the bulk solvent. The relative extent of dynamic slowdown (5−12) appears to depend on both ion concentration and counterion size. The reduced Raman spectral density (rRSD) derived from the spectra was analyzed by a multicomponent fit. An increase in local interaction forces (torques) around AC molecules upon the formation of Li+···OC contacts, with increasing ion concentration, was reflected through the progressive blue-shift of the rRSD dominated by AC librations. The correlation of the first moment of the librational spectrum with the total molar concentration of the solutions validated the assignment and suggested that (anion-modulated) Li+···AC interactions are the main contribution to electrostriction phenomena. The ability to identify distinct solvent contributions in these complex solutions is an invaluable strength point of the method applied here. Salt-induced perturbations on the slow diffusive relaxation and ultrafast librational dynamics of liquid AC are related to the reorientation of Li+-solvating molecules, in turn modulated by ion−ion interactions. The correlation between solvation dynamics and ion association was demonstrated by these experiments.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +390755855580. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This project was funded by PRIN 2010-2011 of the Italian Ministry of University and Research (MIUR).

■

REFERENCES

(1) Marcus, Y. Ion Solvation; John Wiley & Sons: Chichester, U.K., 1985. (2) Shirota, H.; Fujisawa, T.; Fukazawa, H.; Nishikawa, K. Ultrafast Dynamics in Aprotic Molecular Liquids: A Femtosecond RamanInduced Kerr Effect Spectroscopic Study. Bull. Chem. Soc. Jpn. 2009, 82, 1347−1366. (3) Stevens, J. R.; Jacobsson, P. A Comparison of Acetone and Poly(Propylene Glycol) as Solvents for Lithium Triflate and Lithium Perchlorate. Can. J. Chem. 1991, 69, 1980−1984. (4) Gouaux, E.; MacKinnon, R. Principles of Selective Ion Transport in Channels and Pumps. Science 2005, 310, 1461−1465. (5) Baldwin, R. L. How Hofmeister Ion Interactions Affect Protein Stability. Biophys. J. 1996, 71, 2056−2063. (6) Olsher, U.; Izatt, R. M.; Bradshaw, J. S.; Dalley, N. K. Coordination Chemistry of Lithium Ion: a Crystal and Molecular Structure Review. Chem. Rev. 1991, 91, 137−164. (7) Kovacsics, C. E.; Gottesman, I. I.; Gould, T. D. Lithium’s Antisuicidal Efficacy: Elucidation of Neurobiological Targets Using Endophenotype Strategies. Annu. Rev. Pharmacol. 2009, 49, 175−198. (8) Chakrabarti, P. Systematics in the Interaction of Metal Ions with the Main-Chain Carbonyl Group in Protein Structures. Biochemistry 1990, 29, 651−658. (9) Perelygin, I. S.; Klimchuk, M. A. Infrared Spectra and Structure of Non-Aqueous Electrolyte Solutions. V. Sodium, Lithium, and Magnesium Perchlorate Solutions in Acetone at Temperatures between −90° and 45°C. Russ. J. Phys. Chem. 1974, 48, 1466−1469. (10) Perelygin, I. S.; Klimchuk, M. A. Infrared Spectra and Structure of Non-Aqueous Electrolyte Solutions. IV. Solutions of Sodium, I

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

Order Møller-Plesset Perturbation Method. J. Chem. Phys. 2007, 127, 244502. (32) Palombo, F.; Paolantoni, M.; Sassi, P.; Morresi, A.; Giorgini, M. G. Molecular Dynamics of Liquid Acetone Determined by Depolarized Rayleigh and Low-Frequency Raman Scattering Spectroscopy. Phys. Chem. Chem. Phys. 2011, 13, 16197−16207. (33) Heisler, I. A.; Mazur, K.; Meech, S. R. Low-Frequency Modes of Aqueous Alkali Halide Solutions: An Ultrafast Optical Kerr Effect Study. J. Phys. Chem. B 2011, 115, 1863−1873. (34) Paolantoni, M.; Ladanyi, B. M. Polarizability Anisotropy Relaxation in Liquid Ethanol: A Molecular Dynamics Study. J. Chem. Phys. 2002, 117, 3856−3873. (35) Zhong, Q.; Fourkas, J. T. Optical Kerr Effect Spectroscopy of Simple Liquids. J. Phys. Chem. B 2008, 112, 15529−15539. (36) Marcus, Y. Electrostriction, Ion Solvation, and Solvent Release on Ion Pairing. J. Phys. Chem. B 2005, 109, 18541−18549. (37) Jones, G.; Dole, M. The Viscosity of Aqueous Solutions of Strong Electrolytes with Special Reference to Barium Chloride. J. Am. Chem. Soc. 1929, 51, 2950−2964. (38) Hefter, G.; May, P. M.; Sipos, P.; Stanley, A. Viscosities of Concentrated Electrolyte Solutions. J. Mol. Liq. 2003, 103−104, 261− 273. (39) Jenkins, H. D. B.; Marcus, Y. Viscosity B-Coefficients of Ions in Solution. Chem. Rev. 1995, 95, 2695−2724. (40) Corridoni, T.; Mancinelli, R.; Ricci, M. A.; Bruni, F. Viscosity of Aqueous Solutions and Local Microscopic Structure. J. Phys. Chem. B 2011, 115, 14008−14013. (41) Megyes, T.; Radnai, T.; Wakisaka, A. Complementary Relation between Ion−Counterion and Ion−Solvent Interaction in Lithium Halide−Methanol Solutions. J. Phys. Chem. A 2002, 106, 8059−8065. (42) Savedoff, L. G. Conductance of Electrolytes in Anhydrous Acetone. J. Am. Chem. Soc. 1966, 88, 664−667. (43) Das, D.; Das, B.; Hazra, D. K. Conductance of Some 1:1 Electrolytes in N,N-Dimethylacetamide at 25°C. J. Solution Chem. 2002, 31, 425−431. (44) Kameda, Y.; Kudoh, N.; Suzuki, S.; Usuki, T.; Uemura, O. Solvation Structure of Lithium Bromide in Concentrated Acetone Solutions. Bull. Chem. Soc. Jpn. 2001, 74, 1009−1014. (45) Richardi, J.; Fries, P. H.; Krienke, H. The Solvation of Ions in Acetonitrile and Acetone: A Molecular Ornstein-Zernike study. J. Chem. Phys. 1998, 108, 4079−4089. (46) Paolantoni, M.; Sassi, P.; Morresi, A.; Santini, S. Hydrogen Bond Dynamics and Water Structure in Glucose-Water Solutions by Depolarized Rayleigh Scattering and Low-Frequency Raman Spectroscopy. J. Chem. Phys. 2007, 127, 024504. (47) Ricci, M.; Bartolini, P.; Chelli, R.; Cardini, G.; Califano, S.; Righini, R. The Fast Dynamics of Benzene in the Liquid Phase - Part I. Optical Kerr Effect Experimental Investigation. Phys. Chem. Chem. Phys. 2001, 3, 2795−2802. (48) Chelli, R.; Cardini, G.; Ricci, M.; Bartolini, P.; Righini, R.; Califano, S. The Fast Dynamics of Benzene in the Liquid Phase. Part II. A Molecular Dynamics Simulation. Phys. Chem. Chem. Phys. 2001, 3, 2803−2810. (49) Smith, N. A.; Meech, S. R. Ultrafast Dynamics of Polar Monosubstituted Benzene Liquids Studied by the Femtosecond Optical Kerr Effect. J. Phys. Chem. A 2000, 104, 4223−4235. (50) Castner, E. W., Jr; Maroncelli, M. Solvent Dynamics Derived from Optical Kerr Effect, Dielectric Dispersion, and Time-Resolved Stokes Shift Measurements: an Empirical Comparison. J. Mol. Liq. 1998, 77, 1−36. (51) Asthana, B. P.; Kiefer, W. Deconvolution of the Lorentzian Linewidth and Determination of Fraction Lorentzian Character from the Observed Profile of a Raman Line by a Comparison Technique. Appl. Spectrosc. 1982, 36, 250−257. (52) Piekarski, H. Dissolution Enthalpies of NaI in Binary Mixtures of Water with Aliphatic Alcohols and Methylcellosolve at 298.15 K. Enthalpic Pair Interaction Coefficients of NaI-Nonelectrolyte in Water Solution. Can. J. Chem. 1983, 61, 2203−2208.

(53) Peruzzi, N.; Ninham, B. W.; Lo Nostro, P.; Baglioni, P. Hofmeister Phenomena in Nonaqueous Media: The Solubility of Electrolytes in Ethylene Carbonate. J. Phys. Chem. B 2012, 116, 14398−14405. (54) Gering, K. L. Prediction of Electrolyte Viscosity for Aqueous and Non-Aqueous Systems: Results from a Molecular Model Based on Ion Solvation and a Chemical Physics Framework. Electrochim. Acta 2006, 51, 3125−3138. (55) Cartailler, T.; Kunz, W.; Turq, P.; Bellisent-Funel, M. C. Lithium Bromide in Acetonitrile and Water: a Neutron Scattering Study. J. Phys.: Condens. Matter 1991, 3, 9511−9520. (56) Paolantoni, M.; Comez, L.; Gallina, M. E.; Sassi, P.; Scarponi, F.; Fioretto, D.; Morresi, A. Light Scattering Spectra of Water in Trehalose Aqueous Solutions: Evidence for Two Different Solvent Relaxation Processes. J. Phys. Chem. B 2009, 113, 7874−7878. (57) Perticaroli, S.; Comez, L.; Paolantoni, M.; Sassi, P.; Morresi, A.; Fioretto, D. Extended Frequency Range Depolarized Light Scattering Study of N-Acetyl-leucine-methylamide-Water Solutions. J. Am. Chem. Soc. 2011, 133, 12063−12068. (58) Perticaroli, S.; Comez, L.; Paolantoni, M.; Sassi, P.; Lupi, L.; Fioretto, D.; Paciaroni, A.; Morresi, A. Broadband Depolarized Light Scattering Study of Diluted Protein Aqueous Solutions. J. Phys. Chem. B 2010, 114, 8262−8269. (59) Lupi, L.; Comez, L.; Paolantoni, M.; Fioretto, D.; Ladanyi, B. M. Dynamics of Biological Water: Insights from Molecular Modeling of Light Scattering in Aqueous Trehalose Solutions. J. Phys. Chem. B 2012, 116, 7499−7508. (60) Fioretto, D.; Comez, L.; Gallina, M. E.; Morresi, A.; Palmieri, L.; Paolantoni, M.; Sassi, P.; Scarponi, F. Separate Dynamics of Solute and Solvent in Water−Glucose Solutions by Depolarized Light Scattering. Chem. Phys. Lett. 2007, 441, 232−236. (61) Gallina, M. E.; Comez, L.; Morresi, A.; Paolantoni, M.; Perticaroli, S.; Sassi, P.; Fioretto, D. Rotational Dynamics of Trehalose in Aqueous Solutions Studied by Depolarized Light Scattering. J. Chem. Phys. 2010, 132, 214508. (62) Jarek, R. L.; Miles, T. D.; Trester, M. L.; Denson, S. C.; Shin, S. K. Solvation of Li+ by Acetone, THF, and Diethyl Ether in the Gas Phase and the Ion−Molecule Association Mechanism. J. Phys. Chem. A 2000, 104, 2230−2237. (63) Lynden-Bell, R. M.; Steele, W. A. A Model for Strongly Hindered Molecular Reorientation in Liquids. J. Phys. Chem. 1984, 88, 6514−6518. (64) Sassi, P.; Paolantoni, M.; Perticaroli, S.; Palombo, F.; Morresi, A. A Study of Collective Motions in Liquid Tert-Butanol from LowWavenumber Raman Scattering. J. Raman Spectrosc. 2009, 40, 1279− 1283. (65) Polok, K.; Idrissi, A.; Gadomski, W. Low Frequency Response of Methanol/Acetone Mixtures: Optical Kerr Effect and Molecular Dynamics Simulations. J. Mol. Liq. 2012, 176, 29−32. (66) Ryu, S.; Stratt, R. M. A Case Study in the Molecular Interpretation of Optical Kerr Effect Spectra: Instantaneous-NormalMode Analysis of the OKE Spectrum of Liquid Benzene. J. Phys. Chem. B 2004, 108, 6782−6795. (67) Elola, M. D.; Ladanyi, B. M. Polarizability Response in Polar Solvents: Molecular-Dynamics Simulations of Acetonitrile and Chloroform. J. Chem. Phys. 2005, 122, 224506. (68) Hunt, N. T.; Jaye, A. A.; Meech, S. R. Ultrafast Dynamics in Complex Fluids Observed through the Ultrafast Optically-Heterodyne-Detected Optical-Kerr-Effect (OHD-OKE). Phys. Chem. Chem. Phys. 2007, 9, 2167−2180. (69) Smith, N. A.; Meech, S. R. Optically-Heterodyne-Detected Optical Kerr Effect (OHD-OKE): Applications in Condensed Phase Dynamics. Int. Rev. Phys. Chem. 2002, 21, 75−100. (70) McMorrow, D.; Lotshaw, W. T. Intermolecular Dynamics in Acetonitrile Probed with Femtosecond Fourier-Transform Raman Spectroscopy. J. Phys. Chem. 1991, 95, 10395−10406. (71) Bardak, F.; Xiao, D.; Hines, L. G.; Son, P.; Bartsch, R. A.; Quitevis, E. L.; Yang, P.; Voth, G. A. Nanostructural Organization in Acetonitrile/Ionic Liquid Mixtures: Molecular Dynamics Simulations J

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry B

Article

and Optical Kerr Effect Spectroscopy. ChemPhysChem 2012, 13, 1687−1700. (72) Guillot, B.; Marteau, P.; Obriot, J. Investigation of Very Fast Motions in Electrolyte Solutions by Far Infrared Spectroscopy. J. Chem. Phys. 1990, 93, 6148−6164. (73) Davis, S. L. Model Polarizabilities and Multipoles for Ionic Compounds. Alkali Halides. J. Chem. Phys. 1988, 88, 1080−1087. (74) CRC Handbook of Chemistry and Physics, 90th ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 2009−2010. (75) Adams, W. A.; Laidler, K. J. Electrical Conductivities of Quaternary Ammonium Salts in Acetone. Part II. The Mechanisms of Transport. Can. J. Chem. 1968, 46, 1989−2003.

K

dx.doi.org/10.1021/jp4066604 | J. Phys. Chem. B XXXX, XXX, XXX−XXX