Dielectric relaxation and ultrafast transient absorption spectroscopy of [ C 6 mim ] + [ Tf 2 N ] / acetonitrile mixtures

Mixtures of the ionic liquid (IL) [C6mim] [Tf2N] and acetonitrile have been investigated by a combination of dielectric relaxation spectroscopy (DRS) and ultrafast transient absorption techniques using the molecular probe 120-apo-b-carotenoic-120-acid (120CA). Steady-state absorption spectra of the 120CA molecule have also been recorded. The position of the probe’s S0 S2 absorption maximum correlates linearly with the polarizability of the mixture, suggesting that the bulk composition is a good approximation to the local composition. The lifetime t1 of the S1/ICT state of 12 0CA varies rather smoothly with composition between the value for pure acetonitrile (42 ps) and neat [C6mim] [Tf2N] (94 ps). At low IL contents there appears to be an influence of discrete ion pairs. Employing static dielectric constants from the DRS experiments, one finds that the lifetime of the probe in the IL mixtures is shorter than that in pure organic solvents with the same polarity parameter. This suggests an increased stabilization of the S1/ICT state in IL-containing mixtures, most likely due to IL-specific Coulombic interactions between the cation and the negative end of the probe’s dipole. An ultrafast solvation component is observed which is ca. 0.5 ps in pure acetonitrile, and approaches the value for the pure IL (2.0 ps) already around x(IL) = 0.3. This is interpreted in terms of an efficient perturbation of the cooperative solvation response of acetonitrile by the presence of small amounts of IL and possibly also the viscosity increase when adding IL. This view is also supported by the increase of the average longitudinal relaxation time of acetonitrile upon addition of small IL amounts extracted from the DRS experiments.


Introduction
Currently, experimental information regarding the polarity and solvation dynamics of ILs and their mixtures with other solvents is still sparse and rather heterogeneous.2][3] DRS applies an external ''macroscopic'' alternating field to determine the frequency-dependent complex dielectric function of a sample.Accurate data can be usually obtained in the frequency range from ca. 100 GHz down to 100 MHz. 4,5[10] Complementary approaches use local, ''microscopic'' molecular probes.This concept is more closely linked to chemical reactivity, because it is the local surrounding or the ''cybotactic'' region of a molecule which crucially influences the outcome of reactions.Probing of the cybotactic region can e.g.2][13][14] Care has to be taken in the interpretation of such steady-state spectroscopic studies when using solvatochromic probes, such as Reichardt's dye (E T N polarity scale), or the Kamlet-Taft parameter for dipolarity/polarizability, p*, where the influence of other solvent effects such as hydrogen-bonding may render unambiguous conclusions difficult.
6][17][18][19] Interestingly, this correlation was not followed by a large range of ILs. 20The lifetime of the S 1 /ICT state in these ILs was comparable to that in solvents such as ethanol or methanol, and thus the probe's local environment appears to be considerably more polar than the dielectric constants from DRS suggest.This is in line with earlier results based on solvatochromic probes. 11,12The underlying physical reason for the difference between macroscopic and ''local'' polarities is still unknown and therefore is an interesting question to us.
2][23] As outlined by Ernsting and co-workers, such an ultrafast probe can be regarded as a local THz emitter/detector, which essentially performs dielectric relaxation spectroscopy in the cybotactic region. 24Therefore, it is interesting to compare the results of such studies with timescales for solvent movements obtained from ''macroscopic'' DRS experiments.
The level of complexity is increased when the transition from neat ILs to IL mixtures is made.6][27] Mixtures of room temperature ILs and organic solvents allow a systematic variation of thermodynamic, polarity and solvation properties.Another interesting question is then, whether there is an ''ideal'' mixing behavior in binary mixtures of ILs or IL + organic solvents in the macroscopic and microscopic regimes.''Ideal'' behavior has been experimentally observed for some important thermodynamic quantities, such as viscosity and molar volume. 26 related issue is the concept of ''preferential solvation'' in spectroscopic studies of solvatochromic probes. 28,29n this study, we use a combined experimental approach using DRS and ultrafast transient absorption spectroscopy to study solvent polarity and solvation in binary mixtures of 1-nhexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)amide ([C 6 mim] + [Tf 2 N] À ) with the polar organic solvent acetonitrile.We use the carbonyl carotenoid 12 0 -apo-b-carotenoic-12 0 -acid (12 0 CA, see Fig. 1) as a micropolarity and local THz probe.The time-resolved experiments are complemented by steadystate absorption measurements.

Experimental
Dielectric spectra, e*(n) = e 0 (n) À ie 0 0 (n) (Fig. 2), where e 0 (n) is the relative permittivity and e 0 0 (n) the dielectric loss, were recorded at 25 AE 0.05 1C in the frequency range of 0.2 r n/GHz r 89.For 0.2 to 20 GHz a Hewlett-Packard model HP 85070M dielectric probe system consisting of a HP 8720D vector network analyzer (VNA) and a HP 85070M dielectric probe kit was used. 30Two waveguide interferometers were used at 27 r n/GHz r 89. 31 Additionally, a new reflectometer consisting of an Agilent E8364B VNA, connected to an electronic calibration module (ECal, Agilent N4693A) and a high-frequency dielectric probe kit (85070E), was used to cover the region 5 r n/GHz r 50 for selected samples.Whilst the interferometers yielded absolute data, the reflection measurements with the VNA probes were calibrated as described in detail previously. 2 To correct the experimentally accessible total loss, Z(n) = e 0 0 (n) + k/(2pne 0 ) (e 0 is the electric field constant), of the sample for the contribution from dc conductivity, k, 2 the latter was separately determined with the setup described in ref. 32.Samples for the DRS studies were prepared from [C 6 mim] + [Tf 2 N] À (IoLiTec) dried at a high-vacuum line (p o 10 À8 bar) for 7 days at B40 1C prior to use (water mass fraction o50 ppm) and from acetonitrile (Merck, >99.9%) distilled from CaH 2 and stored over activated 4 A ˚molecular sieves.Solutions were prepared individually on an analytical balance without buoyancy corrections.Solutions were transferred to the DRS apparatus using syringe techniques.Sample preparation, handling and measurement were performed under dry N 2 (g).
4][35][36] 12 0 CA was dissolved in the solvent or solvent mixture of interest and pumped through a thermostatted (298.15K) quartz flow-through  This journal is c the Owner Societies 2012 cuvette (path length 1.0 mm, window thickness 1.2 mm).The reporter molecule was excited at 480 nm (0.8À3 mJ pulse À1 ) and the subsequent ultrafast dynamics probed by a supercontinuum (340À770 nm) at magic-angle polarization.In the current experiments, the cross-correlation time was ca.80 fs and the time accuracy 10 fs.VIS pump (430 nm, o0.1 nJ pulse À1 )-near IR probe (860 nm, o1 nJ pulse À1 ) laser experiments were performed with another setup described previously using the same flow-cell. 17,37The time resolution of this setup was ca.130 fs.The 12 0 CA probe (>97% all-trans form) was generously provided by Dr Hansgeorg Ernst (BASF SE).[C 6 mim] + [Tf 2 N] À was purchased from IoLiTec and kept over molecular sieves (water mass fraction o100 ppm).Acetonitrile (Merck) had a specified purity of 99.5%.All samples were handled under an argon atmosphere.Steady-state absorption spectra were recorded using a Cary 5E spectrometer.

Dielectric data
The dielectric spectra of [C 6 mim] + [Tf 2 N] À + acetonitrile mixtures (Fig. 2) change smoothly from the single Debye-type (D) relaxation of pure acetonitrile 38 centred at B50 GHz to the broad distribution of relaxation times for the IL.In the investigated frequency range ((0.2 to 89) GHz) the IL spectrum can be modelled by the superposition of a lowfrequency Cole-Cole (CC) mode at B0.6 GHz and a small Debye relaxation at B200 GHz. 2 For imidazolium ILs, including [C 6 mim] + [Tf 2 N] À , the low-frequency mode can be essentially assigned to large-angle jump reorientation of the cation. 5However, the high-frequency mode is just a formal description of the significant contributions from librations and intermolecular vibrations dominating the far-infrared region. 5,8,25onsistent with the bimodal character of the spectra at intermediate concentrations (Fig. 2) and similar to IL + dichloromethane mixtures 25 all mixture spectra could be well fitted with the CC+D model (Fig. 2).We abstain here from giving the parameters (amplitudes, relaxation times and width parameter) of the CC+D model as their interpretation for the present IL + acetonitrile mixtures is not straightforward due to strong band overlap. 39However, as can be judged from Fig. 2a, the extrapolation e = lim n-0 e 0 (n) yielding the static permittivity, e, of the [C 6 mim] + [Tf 2 N] À + acetonitrile mixtures, is always unproblematic.This key quantity of the present dielectric measurements is given in Table 1, together with the average relaxation time of the samples, t av = (2pn peak ) À1 , associated with the frequency n peak of their loss peak, Max[e 0 0 (n)].The significant increase of t av from 5.8 ps at x = 0.08403 to 42 ps at x = 0.2595 reflects the switch-over from the acetonitrile-dominated spectrum at a low [C 6 mim] + [Tf 2 N] À content to IL-dominated spectra at x > 0.25.
Fig. 3 shows the dependence of the static permittivity on the IL mole fraction, x.As expected from the significant difference in the static permittivities of both compounds (e = 35.96for pure acetonitrile 40 and 12.7 for [C 6 mim] + [Tf 2 N] À2 ), e increases strongly with decreasing x.Interestingly, however, for vanishing IL content e does not smoothly extrapolate to the value of pure acetonitrile (B35.9), 38,40,41but reaches e = 37.0 at the lowest investigated IL concentration, x = 0.00235 (inset of Fig. 3).A similar behaviour was observed for mixtures of ILs with dichloromethane. 3,25Additionally, for x o 0.6 the static permittivity of the mixtures estimated from the analytical concentrations of IL and acetonitrile (dash-dotted line in Fig. 3) with the help of the effective dipole moments of the pure components 3 is significantly smaller than the experimental data.These observations are strong indication for the presence of an additional species with large dipole moment at low IL content.Most likely, this is a [C 6 mim] + [Tf 2 N] À ion pair.Note that ion pairs of imidazolium ILs were not only found in dichloromethane 3,25 but also in acetonitrile and even water. 422 Solvatochromy of 12 0 CA in the IL/organic solvent mixture Fig. 4a shows the dependence of the S 0 -S 2 absorption band of 12 0 CA on mixture composition.For the sake of clarity, only the maximum absorption region of the normalized spectra in the range 0.9-1.0 is presented.Fig. 4b contains a plot of the  2). Tere is a clear nonlinear red-shift of the steady-state absorption spectrum with increasing x(IL).For the 12 0 CA probe molecule and a range of other carotenoids, we previously demonstrated in organic solvents and supercritical fluids that the absorption shift depends linearly on the polarizability function R(n) = (n 2 À 1)/(n 2 + 2), a behavior which is typical for many carotenoids.35,43 In the current study, we were able to confirm this behavior for the IL/organic solvent mixture: Fig. 5 contains a plot of the absorption maximum against R(n), with our measured refractive indices n and R(n) values summarized in Table 2.The linear dependence clearly suggests that the bulk composition of the mixture is a good approximation to the local mixture composition in the vicinity of the probe, i.e., there is no indication for ''preferential'' solvation of the probe molecule by either of the mixture components. We note that including the absorption maxima of 12 0 CA for the previously studied organic solvents and ILs 19,20 (not shown in Fig. 5) still leads to the same qualitative dependence on polarizability, however with a stronger scatter of the points.This is understandable considering the strong variation in the solvent type, e.g.regarding H-bonding, shape, size, etc.We also note that results, such as those shown in Fig. 4b, have been frequently interpreted as an indication of specific solvation effects.29 However, it is clear that the nonlinearities in Fig. 4b can be simply traced back to the much larger molar polarizability of the IL compared to acetonitrile.

S 1 /ICT lifetime as a function of mixture composition
For a first overview, Fig. 6 compares selected PSCP broadband absorption spectra for [C 6 mim] + [Tf 2 N] À , acetonitrile and four selected mixture compositions at 10 ps.All spectra consist of a negative-going S 0 -S 2 ground state bleach (GSB, 350-450 nm) and the characteristic double-peak of the S 1 /ICT -S n excited state absorption (ESA, peaks at ca. 500 and 600 nm).We note that an additional smaller ESA band peaking at ca. 720 nm arises from absorption of the 12 0 CA radical cation (presumably the D 0 -D 3 transition).The radical cation is produced by two-photon ionization,    This journal is c the Owner Societies 2012 when a sufficiently high pump pulse energy is used. 22At long times (not shown), after the S 1 /ICT -S n ESA band of neutral 12 0 CA has completely decayed, only the characteristic longlived absorption of the 12 0 CA radical cation absorption remains, with a peak at 720 nm and an extended tail toward shorter wavelengths (see e.g.Fig. 2 in ref. 22, which also shows the remaining GSB due to the missing S 0 population of neutral 12 0 CA).Under the selected excitation conditions of the current PSCP experiments, this radical cation absorption overwhelms the well-known S 1 /ICT -S 0 stimulated emission of neutral 12 0 CA. 20The trends observed for the transient absorption spectra confirm the larger red-shift with increasing mole fraction of IL (Fig. 4) also for the S 1 /ICT -S n ESA band and the cation absorption band.
Next, we recorded transient pump-probe absorption signals of 12 0 CA for different mole fractions of IL at extremely low pump power, which suppresses the formation of the radical cation completely.Two representative examples are shown in Fig. 7 for x(IL) = 0.1 and 0.9, respectively (l pump = 430 nm, l probe = 860 nm).The typical signal shape consists of a transient S 2 -S n absorption peak at early times which due to the internal conversion (IC) process S 2 -S 1 /ICT quickly develops into pronounced S 1 /ICT -S 0 stimulated emission (SE), with t 2 = 120 fs.The SE finally decays because of the IC process S 1 /ICT -S 0 .This decay directly provides the S 1 /ICT lifetime t 1 , which is summarized in Table 2 for all mixtures.
Table 2 clearly demonstrates that there is a pronounced increase of the lifetime upon addition of IL.The fast decay of pure acetonitrile (42 ps) is expected for such a polar solvent and is slightly faster than for methanol (49 ps).The value for pure [C 6 mim] + [Tf 2 N] À (94 ps) is consistent with the value measured recently by us. 20For comparison of the transient absorption and dielectric data the following strategy was employed: the decrease of t 1 from [C 6 mim] + [Tf 2 N] À to acetonitrile can be understood in terms of a polarity-induced reduction of the S 1 /ICT-S 0 energy gap DE, which accelerates the nonradiative IC process. 16This type of behavior is also well-known for other carbonyl-substituted carotenoids, e.g., peridinin or fucoxanthin. 44Such a dependence can be described in terms of an energy-gap-law approach: 16,45 ln(k 1 /s À1 ) = ln({t 1 /s} À1 ) = A À BÁDE/cm À1 (1) Here, k 1 (= t 1

À1
) is the IC rate constant and A and B are fit constants.DE can be approximately related to the Stokes shift Dv˜S tokes , because both quantities are measures of the solventdependent energetic separation between the S 1 /ICT and S 0 electronic states.Simple continuum theories of solvation predict that Dv˜S tokes is proportional to the polarity parameter Df, 46 defined as: Of course, the shortcomings of continuum-type approaches for the description of IL properties are well documented, 23,[47][48][49] and so far no physically sound model for a reliable description of ILs exists.Still, we believe that one obtains reasonable qualitative trends for the mixtures of the current study by using correlations based on quantities such as Df.Therefore, we compare the different experiments by plotting the transient absorption results in terms of ln(k 1 /s À1 ) and the dielectric data in terms of Df vs. the mole fraction x(IL).This is shown in Fig. 8a, where the y axes are scaled in a way that the ln(k 1 ) and Df values for pure acetonitrile and the pure IL are on top of each other to highlight the dependence on mixture

View Article Online
This journal is c the Owner Societies 2012 Phys.Chem.Chem.Phys., 2012, 14, 3596-3603 3601 composition.Error bars for the ln(k 1 ) values (and also for the t 3 values later on presented in Fig. 8b) were extracted from the fit results of ca.5-15 transients recorded for each x(IL).One observes a similar slightly curved dependence of ln(k 1 ) and Df on IL mole fraction.For the data representation chosen, deviations become larger below x(IL) E 0.4.Of particular interest is the maximum of ln(k 1 ) around x(IL) = 0.05 meaning that addition of a small amount of IL to pure acetonitrile leads to an acceleration of the IC process.The effect might be due to the formation of discrete ion pairs of the IL with large dipole moment. 2719,21 Specifically for 12 0 CA in organic solvents, we established a smooth correlation between t 1 and the polarity parameter Df.This is shown in Fig. 9. Lifetimes varied between 230 ps in n-hexane and 49 ps in methanol. 20The current experiment in the highly polar acetonitrile fits very well into this correlation.Interestingly, ILs featuring imidazolium, pyrrolidinium, tetraalkylammonium, and trialkylsulfonium cations did not follow the correlation (open circles in Fig. 9).1][52][53] Instead, for a series of chemically and structurally different cations featuring the same counter-anion [Tf 2 N] À , we succeeded in establishing an empirical correlation between t 1 and the IL cation radius, which suggests the importance of Coulombic interactions of cations with the negatively charged carbonyl end of the reporter molecule. 20This probably leads to a stronger stabilization of the S 1 /ICT state of the probe.
The current work allows us to investigate the applicability of this correlation to IL mixtures with the polar aprotic solvent acetonitrile.For that, we use Df values directly calculated from the static dielectric constants (Table 1) and the refractive indices (Table 2).The results are included in Fig. 9 as half-filled circles.The mixtures smoothly connect the highly polar part of the t 1 -Df correlation with the IL region.As a result, mixtures with increasing IL composition progressively deviate from the correlation.This behavior supports our previous interpretation that simple descriptions based on dielectric continuum theory, which work surprisingly well for dipolar organic solvents, fail for imidazolium-based ILs, because (cation) monopole-dipole contributions dominate the IL cation-probe interaction.In fact, this is also consistent with our previous finding that an addition of a large amount of Li + cations to organic solvents markedly reduces t 1 of 12 0 CA. 20

Solvation dynamics as observed by the local probe and dielectric spectroscopy
The transient absorption signals contain additional information related to the solvation response upon the change of the probe's dipole moment.Because the S 1 /ICT lifetime of 12 0 CA in the mixtures studied is fairly short (o100 ps), we restrict our discussion to the short time dynamics.23]54 The former one corresponds to high frequency solvent modes in the dielectric spectrum.In the near-IR transient absorption experiments, the ultrafast response is visible at early times as a curvature on the signals which is due to the transient Stokes shift of the SE band of 12 0 CA, see the inset in Fig. 7. Fitting of the transients provides an additional time constant t 3 , which is plotted as a function of IL mole fraction in Fig. 8b including error bars deduced from typically 5-15 experiments at a given mole fraction.Values are summarized in Table 2. Regarding possible variations of t 3 with probe wavelength, we note that for another closely related carbonyl carotenoid system shifts in the probe wavelength on the order of 20 nm (upper limit estimated from the spectral shift seen in Fig. 6) should not change t 3 by more than typically 10%. 17or pure acetonitrile, the response is extremely fast, which is in agreement with findings for other probe molecules, such as coumarin 153. 23,55For the pure IL, this response is considerably slower, around 2 ps.The interesting observation is that already at low IL mole fractions the solvation response is markedly slowed down.We tentatively explain this by a perturbation of the ultrafast cooperative response of the acetonitrile network by the constituents of the IL and the increasing viscosity of the mixture upon adding IL.Such an interpretation is consistent with a recent Raman spectroscopic study of Aleksa et al. 56 and earlier results of computer simulations for acetonitrile and CO 2 by Ladanyi and Maroncelli. 57Further support for this view comes from a comparison with the results from dielectric spectroscopy.The connection between solvation dynamics and dielectric relaxation of the solvent is through the longitudinal relaxation time, t L . 58For a solute dipole embedded in a solvent showing Debye-type relaxation the latter is approximately given by Such a simple relation does not exist for the present samples but may be tentatively used with t av = t for x o 0.2 because of the predominance of the acetonitrile contribution to the spectra (Fig. 2).Quantitative agreement of t L av with t 3 from the transient absorption signals cannot be expected because of the approximate nature of the average relaxation time, t av .However, at low IL content t L av nicely parallels the increase of t 3 (Fig. 8b) corroborating the interpretation of t 3 as an indicator for fast solvation dynamics.In fact, there might be even a slight systematic shift in t 3 values, because due to the first ultrafast S 2 -S 1 /ICT IC step (ca.120 fs) the fastest part of the solvation response might be somewhat ''smeared out''.This would mean that the t 3 values from transient absorption should be taken as upper limits.This effect would be more pronounced for the acetonitrile-rich mixtures showing fast relaxation.In fact, such a reduction, possibly by perhaps 0.1-0.2ps, would actually bring t 3 even closer to t L av from DRS (Fig. 8b).

Conclusions
The combination of dielectric relaxation spectroscopy (DRS) and ultrafast transient absorption techniques has provided a comprehensive view of the polarity and dynamics of [C 6 mim] + [Tf 2 N] À + acetonitrile mixtures.DRS finds a strong decrease of the static dielectric constant from acetonitrile (ca.35.9) to the pure IL (12.7).For vanishing IL content, however, there is no smooth extrapolation of e to the value of pure acetonitrile, and instead the value e = 37.0 is found at the lowest IL content investigated (x = 0.00235).This effect can be most likely traced back to the influence of [C 6 mim] + [Tf 2 N] À ion pairs.The picture is confirmed by ultrafast transient absorption measurements employing the local polarity and THz probe 12 0apo-b-carotenoic-12 0 -acid (12 0 CA).The lifetime t 1 of the probe's S 1 /ICT state, which is a convenient indicator of polarity, shows a strong increase from pure acetonitrile to the pure IL, corresponding to a substantial decrease of polarity with increasing IL content in the mixture.Interestingly, as in DRS, at small IL concentrations the local environment of the probe appears to be slightly more polar than that in pure acetonitrile which again points toward an influence of IL ion pairs at low IL content.In any case, the linear shift of the S 0 -S 2 absorption maximum with mixture polarizability strongly indicates that ''preferential solvation'' of the probe is negligible.
Using the static dielectric constants from DRS, one finds that the probe's lifetime does not follow the correlation with the dipolarity parameter Df, which was previously established in organic solvents. 20The behavior, which becomes more and more evident with increasing IL content, points toward a substantial stabilization of the 12 0 CA S 1 /ICT state by charge-dipole interactions of IL cation(s) with the terminal carbonyl group.
The 12 0 CA probe also identified a fast solvent relaxation component which is around 0.5 ps in acetonitrile and about 2 ps in the pure IL.A small addition of IL to acetonitrile is already sufficient to slow down the reorientation dynamics of the acetonitrile network and/or increase the viscosity of the mixture.The picture is in agreement with the results from DRS which finds an increase of the average longitudinal relaxation time in the same range of mixture composition.
For future studies, it will be worthwhile to extend this combined approach to other IL mixtures, containing, e.g., more polar ILs or less polar organic solvents, in order to obtain a more complete overview of the dynamics in these interesting binary solvent systems.

Fig. 4
Fig. 4 (a) Maximum region of the steady-state absorption spectra of 12 0 CA in [C 6 mim] + [Tf 2 N] À /acetonitrile mixtures as a function of wavelength; (b) absorption maxima (in cm À1 ) as a function of x(IL).

Fig. 5
Fig. 5 Steady-state absorption maxima of 12 0 CA as a function of the solvent polarizability function R(n) for different x(IL) (numbers).

Fig. 7
Fig. 7 Time-resolved transient absorption/stimulated emission signals of 12 0 CA in the near IR region for mixtures of [C 6 mim] + [Tf 2 N] À and acetonitrile with x(IL) = 0.1 and 0.9.The inset shows a magnification at early times.l pump = 430 nm, l probe = 860 nm.

Fig. 8
Fig. 8 (a) Open circles with error bars: logarithm of the S 1 /ICT -S 0 internal conversion rate constant ln(k 1 /s À1 ) = ln((t 1 /s) À1 ) as a function of IL mole fraction for 12 0 CA in [C 6 mim] + [Tf 2 N] À /acetonitrile mixtures, calculated from the values inTable 2; crosses: Df values calculated using eqn (2) from e values in Table 1 and interpolated n values from Table 2. (b) Fast solvation component t 3 of 12 0 CA (open squares with error bars) and average longitudinal relaxation time t L
a Ref. 41. b Ref. 38. c Ref. 40. d Ref. 2.Fig.3 Static relative permittivity, e (K, solid line as a guide to the eye; .ref. 40), as a function of IL mole fraction, x, of mixtures of [C 6 mim] + [Tf 2 N] À + acetonitrile at 25 1C.The dash-dotted line indicates the static permittivity expected from the molar concentrations of acetonitrile and IL present in the mixtures (see the text).View Article Online This journal is c the Owner Societies 2012 Phys.Chem.Chem.Phys., 2012, 14, 3596-3603 3599 corresponding maximum position v˜m ax vs. IL mole fraction (for values see Table
; crosses: Df values calculated using eqn (2) from e values in Table 1 and interpolated n values from Table 2. (b) Fast solvation component t 3 of 12 0 CA (open squares with error bars) and average longitudinal relaxation time t L av from dielectric spectroscopy (open triangles) as a function of x(IL).