Picosecond Time-Resolved Fluorescence Study on Solute− Solvent

4914

J. Phys. Chem. B 2007, 111, 4914-4919

Picosecond Time-Resolved Fluorescence Study on Solute-Solvent Interaction of 2-Aminoquinoline in Room-Temperature Ionic Liquids: Aromaticity of Imidazolium-Based Ionic Liquids† Koichi Iwata,*,‡ Minoru Kakita,§ and Hiro-o Hamaguchi*,§ Research Centre for Spectrochemistry and Department of Chemistry, School of Science, The UniVersity of Tokyo Hongo 7-3-1, Bunkyo-ku, Tokyo 113-0033, Japan ReceiVed: NoVember 1, 2006; In Final Form: February 26, 2007

Time-resolved fluorescence spectra and fluorescence anisotropy decay of 2-aminoquinoline (2AQ) have been measured in eight room-temperature ionic liquids, including five imidazolium-based aromatic ionic liquids and three nonaromatic ionic liquids. The same experiments have also been carried out in several ordinary molecular liquids for comparison. The observed time-resolved fluorescence spectra indicate the formation of π-π aromatic complexes of 2AQ in some of the aromatic ionic liquids but not in the nonaromatic ionic liquids. The fluorescence anisotropy decay data show unusually slow rotational diffusion of 2AQ in the aromatic ionic liquids, suggesting the formation of solute-solvent complexes. The probe 2AQ molecule is likely to be incorporated in the possible local structure of ionic liquids, and hence the anisotropy decays only through the rotation of the whole local structure, making the apparent rotational diffusion of 2AQ slow. The rotational diffusion time decreases rapidly by adding a small amount of acetonitrile to the solution. This observation is interpreted in terms of the local structure formation in the aromatic ionic liquids and its destruction by acetonitrile. No unusual behavior upon addition of acetonitrile has been found for the nonaromatic ionic liquids. It is argued that the aromaticity of the imidazolium cation plays a key role in the local structure formation in imidazolium-based ionic liquids.

1. Introduction Room-temperature ionic liquids are formed solely by ions. The positive and negative charges carried by these ions produce a variety of electrostatic environments that should affect the behaviors of electrons in ionic liquid systems. Because chemical reactions are, in a sense, relocation processes of valence electrons, the charge distribution in the ionic liquids should affect profoundly the chemical reactions proceeding in ionic liquids. It is therefore of considerable interest to examine the solute-solvent interaction and hence the molecular environments in ionic liquids with an electron-rich probe molecule, such as an aromatic molecule. One of the most widely used cations used for generating ionic liquids is the imidazolium cation. Because the imidazolium cation with six π-electrons is aromatic, an ionic liquid with the imidazolium cation will show the characters of an aromatic solvent in addition to the characters of an ionic liquid. It is thus important to examine the aromatic characters of imidazoliumbased ionic liquids and separate them from the characters of ionic liquids in general. We can establish the more general concept of ionic liquids by distinguishing the two characters observed for imidazolium-based ionic liquids. A specific behavior of an aromatic molecule in an ionic liquid has already been reported by us.1 We found that the rate of photoisomerization of S1 trans-stilbene in 1-butyl-3-methylimidazolium hexafluorophosphate (bmim[PF6]) was much larger than that expected from its high polarity and viscosity. From †

Part of the special issue “Physical Chemistry of Ionic Liquids”. * Corresponding authors. E-mail: [email protected], [email protected]. ‡ Research Centre for Spectrochemistry. § Department of Chemistry.

the very high viscosity of bmim[PF6] (312 mPa s at 303 K) with a polarity comparable with alcohols, it was expected that the rate of photoisomerization was as small as 9.7 × 108 s-1. However, the observed value was much larger (6.6 × 109 s-1). We argued that the solvent properties of ionic liquids may not be well accounted for by conventional macroscopic parameters like polarity and viscosity, which have been extensively used for molecular liquids. This finding led us to study the liquid structure of ionic liquids, imidazolium-based ionic liquids in particular, from many different physicochemical approaches.2-6 A working hypothesis has thus been proposed; ionic liquids have specific local structures that cause many of their intriguing properties.7,8 Ionic liquids appear to be homogeneous macroscopically, but they are likely to be heterogeneous microscopically in the nanometer scale. Recent molecular dynamics simulation results have supported this view.9 Study of solutesolvent interaction will shed more light on the possible local structures in ionic liquids. Time-resolved fluorescence spectroscopy is sensitive to the solute-solvent interaction. This method has been widely used for elucidating the solvation environments in ionic liquids.10 By measuring the time-dependent changes of fluorescence signals in ionic liquids, solvation,11-18 rotation,17-19 photoisomerization,1,19 or excimer formation20 process of a probe molecule has been examined. It should be noted that in ordinary molecular solutions many solute-solvent complexes, including excimers and exciplexes, show characteristic fluorescence bands indicating the formation of these complexes, in addition to the monomer fluorescence bands.21 In this study, we observe time-resolved fluorescence spectra of 2-aminoquinoline (2AQ, Figure 1), an aromatic molecule, in five aromatic ionic liquids, three nonaromatic ionic liquids,

10.1021/jp067196v CCC: $37.00 © 2007 American Chemical Society Published on Web 04/12/2007

Solute-Solvent Interaction of 2-Aminoquinoline

J. Phys. Chem. B, Vol. 111, No. 18, 2007 4915

Figure 1. Structure of 2-aminoquinoline (2AQ).

and aromatic and nonaromatic ordinary molecular solvents. We chose 2AQ because we had studied the double proton-transfer reaction of 2-aminopyridine, an analogue of 2AQ, with picosecond time-resolved fluorescence spectroscopy.22 The 2AQ molecule also serves as a relatively small fluorescence probe comparable to the ring portion of the imidazolium cation. 2. Experimental Section The details of our picosecond time-resolved fluorescence spectrometer have been already reported.22 In short, the third harmonic (266 nm, 1 kHz, 2mW) of the output from a Ti:sapphire regenerative amplifier (Clark Instruments, 800 nm, 1 kHz, 400 mW) seeded by a Ti:sapphire oscillator (Spectra Physics, Tsunami) pumped by a Nd:YLF laser (Spectra Physics, Millennia, 523 nm, 5 W) was used as the pump light for the fluorescence excitation. The sample solution was stirred in a quartz cell with the path length of 1 cm during the fluorescence measurement. A portion of the emitted light whose polarization was 54.7° (magic angle) to the excitation polarization was collected selectively by a polarizer for normal measurements while the parallel and perpendicular components were recorded separately for the fluorescence anisotropy measurements. The collected fluorescence signals were dispersed with an astigmatismcorrected spectrograph (Chromex 500). With the use of a streak camera (Hamamatsu C2909), the time and spectral profiles of the fluorescence signals were recorded simultaneously as a twodimensional image. The instrumental response function of the spectrometer, estimated by the rising edge of the excimer fluorescence of benzene, was 9 ps. Steady-state absorption spectra were measured with a commercial spectrophotometer (Hitachi U-3500 UV-vis). Fluorescence and fluorescence excited spectra were measured with a commercial spectrofluorometer (JASCO FP-6500). The sample of 2AQ was purchased from Tokyo Kasei Kogyo Co., Ltd. It was recrystallized from hexane and was subsequently dried in vacuum before the spectroscopic measurements. We use 1-butyl-3-methylimidazolium hexafluorophosphate (bmim[PF6]), 1-butyl-3-methylimidazolium tetrafluoroborate (bmim[BF4]), 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (bmimTf2N), 1-octyl-3-methylimidazolium hexafluorophosphate (omim[PF6]), and 1-ethyl-3-methylimidazolium ethylsulfate (emimSO4Et) as the aromatic ionic liquids, methyltrioctylammonium bis(trifluoromethylsulfonyl)imide (N1888Tf2N), trihexyl(tetradecyl)phosphonium bis(trifluoromethylsulfonyl)imide (P666 14 Tf2N), and trihexyl(tetradecyl)phosphonium chloride (P666 14 Cl) as the nonaromatic ionic liquids, and benzene, toluene, chlorobenzene, acetonitrile, 2,6,10,15,19,23hexamethyltetracosane (squalane), and liquid paraffin as the ordinary organic solvents. Bmim[PF6] and omim[PF6] were synthesized in our laboratory as described in the literature.23 Bmim[BF4], N1888Tf2N, P666 14 Tf2N, and P666 14 Cl were purchased from Aldrich (Fluka). BmimTf2N (lot number 609041) was purchased from Kanto Chemical Co., Ltd. Benzene (HPLC grade), toluene (HPLC grade), chlorobenzene (special grade), acetonitrile (spectroscopy grade), squalane (special grade), and liquid paraffin (d ) 0.86 to 0.89) were purchased from Wako Chemicals Co., Ltd. EmimSO4Et (>99%) was given

Figure 2. UV absorption spectrum (left axis, dotted curve) and fluorescence spectrum (right axis, solid curve) of 2AQ in acetonitrile. The excitation wavelength for the fluorescence measurement was 266 nm. The concentration of 2AQ for the measurements was 5.25 × 10-5 mol dm-3.

by Solvent Innovation. Viscosity of the sample was measured with a viscometer (Brookfield DV-I+ Digital Viscometer). 3. Results and Discussion 3.1. Formation of the Solute-Solvent Complex in Aromatic Solvents. We first examined the steady-state absorption and fluorescence spectra of 2AQ in a nonaromatic solvent. The spectra of 2AQ in a 5.3 × 10-5 mol dm-3 acetonitrile solution are shown in Figure 2. The excitation wavelength for the fluorescence measurement was 266 nm. These spectra agree well with the reported results.24 2AQ has two absorption bands at 330 and below 250 nm. There are two fluorescence bands at 310 and 380 nm, apparently corresponding to the two absorption bands. When we excite 2AQ for the fluorescence measurement, the excited-state first created by the photoexcitation at 266 nm is different from the fluorescent state whose absorption maximum is located at 330 nm. The fluorescence spectrum of the concentration of 5.3 × 10-5 mol dm-3 agreed with the spectrum measured at 8.9 × 10-3 mol dm-3, which is more concentrated than the solutions used for the time-resolved fluorescence measurement (6.9 × 10-3 mol dm-3, see below). Time dependence of the 2AQ fluorescence band was observed with the picosecond time-resolved fluorescence spectrometer described in the Experimental section. The results for acetonitrile and benzene solutions are shown in panels a and b, respectively, of Figure 3. Fluorescence spectra averaged for the time period of 0-0.5, 0.5-1, 1-1.5, 1.5-2, and 2-2.5 ns after the photoexcitation are shown in the figures. The excitation wavelength was 266 nm. The 2AQ concentration for the two solutions was 1.18 × 10-2 mol dm-3. In acetonitrile, the fluorescence band appears at 380 nm immediately after the photoexcitation (Figure 2a). It decays in approximately 2 ns, changing neither its position nor its shape. The experimental results obtained in benzene were different from those in acetonitrile. As in acetonitrile, there is a fluorescence band observed at 380 nm (Figure 3b). In addition to the 380 nm band, however, there is another broad fluorescence feature between 450 and 550 nm. This broad component is observed only for the time period of 0.5-1 ns and is decayed completely after 1.5 ns. It has much shorter lifetime compared with the 380 nm component.

4916 J. Phys. Chem. B, Vol. 111, No. 18, 2007

Iwata et al.

Figure 3. Picosecond time-resolved fluorescence spectra of 2AQ in acetonitrile (a), toluene (b), and bmim[PF6] (c). The concentration was 1.18 × 10-2 mol dm-3 for acetonitrile and toluene, and 6.9 × 10-3 mol dm-3 for bmim[PF6]. The time delay for each spectrum is shown in the figure.

The broad fluorescence band was observed also in ionic liquids. The time-resolved fluorescence spectra of a bmim[PF6] solution of 2AQ is shown in Figure 3c. The concentration was 6.9 × 10-3 mol dm-3. Both of the fluorescence band at 400 nm and the broad component at the longer wavelength side are observed in the set of time-resolved spectra. However, the relative intensities of the two fluorescence components and their time profiles are different from the results obtained for ordinary molecular solutions. The difference between the ionic liquids and the molecular solvents is discussed in details below. The spectral shape of 2AQ fluorescence is different depending on the solvent, in particular at early time delays. We average the fluorescence spectra of 2AQ in various solvents for the time period of 0-0.5 ns. The averaged transient spectra are shown in Figure 4. The upper seven traces in the figure are the transient spectra of 2AQ measured in aromatic organic solvents, benzene (a), toluene (b), and chlorobenzene (c), and in aromatic ionic liquid, omim[PF6] (d), bmim[PF6] (e), emimSO4ET (f), and bmim[BF4] (g). The lower four traces are the spectra measured in nonaromatic organic solvent, acetonitrile (h), and in nonaromatic ionic liquids, N1888Tf2N (i), P666 14Tf2N (j), and P666 14Cl (k). In the figure, there is a difference between the transient spectra observed in most of the aromatic solvents (Figure 3af) and those observed in the nonaromatic solvents (Figure 3hk). The spectra measured in the aromatic solvents show the broad fluorescence component at the longer wavelength side, in addition to the 380 nm band. On the contrary, none of the spectra measured in the nonaromatic solvents shows the broad component. It is not crucial on determining the shape of the observed transient fluorescence spectrum whether the solvent is molecular or ionic. The broad component appears only in the aromatic solvents. In bmim[BF4], however, the broad component was not detected. We assign the fluorescence band at 380 nm, commonly observed in the transient spectra (Figure 4) as well as in the steady-state spectrum (Figure 2), to the monomer fluorescence of 2AQ. We consider that the broad fluorescence component at the longer wavelength side represents the fluorescence from

Figure 4. Transient fluorescence spectra of 2AQ solutions of benzene (a), toluene (b), chlorobenzene (c), omim[PF6] (d), bmim[PF6] (e), bmimSO4Et (f), bmim[BF4] (g), acetonitrile (h), N1888Tf2N (i), P666 14Tf2N (j), and P666 14Cl (k). Time-resolved fluorescence spectra for the time period of 0-0.5 ns are averaged for obtaining each transient spectrum.

a solute-solvent complex, judging from its broad structureless shape and from the fact that it is observed only in the aromatic solvents and not in the nonaromatic solvents. It has been reported that a fluorescence band from the exciplex between 2-amionpyridine and p-nitroaniline is observed at 410 nm in EPA (etherpentane-ethanol) while a monomer fluorescence band of 2-aminopyridine is observed at 340 nm.25 Because the solute

Solute-Solvent Interaction of 2-Aminoquinoline

Figure 5. Time-dependence of fluorescence signals from 2AQ in bmim[PF6] (a, 6.9 × 10-3 mol dm-3), omim[PF6] (b, 6.9 × 10-3 mol dm-3), emimSO4Et (c, 6.9 × 10-3 mol dm-3), chlorobenzene (d, 1.2 × 10-2 mol dm-3), benzene (e, 1.2 × 10-2 mol dm-3), and toluene (f, 1.2 × 10-2 mol dm-3). The fluorescence signals from 450 to 500 nm have been averaged.

2AQ is also an aromatic molecule, it is highly likely that the solute-solvent complex is stabilized by the π-π interaction between the aromatic molecule and the aromatic imidazolium cation. The formation of the solute-solvent complex in most of the imidazolium-based ionic liquids may be explained by the aromaticity of the imidazolium cation. However, the lifetime of the complex seems quite different between the molecular aromatic solvents and the ionic liquids in which the complex is formed. In bmim[PF6], for example, the fluorescence from the complex is observed with a large intensity even after 2 ns (Figure 3c). In benzene, however, this fluorescence component disappears after 1 ns (Figure 3b). There is a marked difference in the 2AQ-solvent complex lifetime in the excited-state between the molecular liquids and the ionic liquids. For examining the dynamic behavior of the 2AQ-solvent complex in more details, we extracted the fluorescence decay curves from the recorded two-dimensional (wavelength-time) fluorescence images. The fluorescence intensities from 450 to 500 nm were averaged and plotted against the time delay. The results are shown in Figure 5 for bmim[PF6] (a), omim[PF6] (b), emimSO4Et (c), benzene (d), toluene (e), and chlorobenzene (f). It is clear from the figure that the lifetime of the fluorescence from the solute-solvent complex in bmim[PF6], emimSO4Et, or in omim[PF6] is much longer than the lifetimes in benzene, toluene, or chlorobenzene. The least-squares fitting analysis showed that the fluorescence decay curves obtained in the three aromatic molecular solvents were well represented by singleexponential decay functions. The obtained lifetimes were 13 ( 2 ps for benzene, 17 ( 1 ps for toluene, and 15 ( 2 ps for chlorobenzene. In the ionic liquids, however, the singleexponential decay function was not a good model. A double exponential decay function gave a successful fit when the lifetimes were 280 ( 20 ps and 3.1 ( 0.2 ns for bmim[PF6], 300 ( 10 ps and 3.5 ( 0.1 ns for omim[PF6], and 340 ( 30 ps and 3.3 ( 0.1 ns for emimSO4Et. The lifetime of the fluorescence from the 2AQ-solvent complex is longer in the

J. Phys. Chem. B, Vol. 111, No. 18, 2007 4917

Figure 6. Typical example from the fluorescence polarization measurement. The results for the acetonitrile molar ratio of 0.02 in bmim[PF6] are shown. Parallel (I|(t)) and perpendicular (I⊥(t)) decay components (upper panel) and calculated anisotropy (lower panel). Single-exponential decay function best fitted to the anisotropy decay curve is shown with a solid curve.

three ionic liquids than in the ordinary solvents by more than 2 orders of magnitude. The unsuccessful fit by a single-exponential function suggests that the solvation environments for 2AQ are not uniform in the ionic liquids, unlike in the molecular liquids. We use the double exponential decay as a model function for characterizing a decay kinetics that is not represented well with a single-exponential decay function. We do not discuss the physical meaning of the two time constants determined from the analysis. However, the variety of the environments in the ionic liquid may be large enough to change the lifetime of the 2AQ complex, which represents the structure and stability of the complex, by 10 times. 3.2. Rotational Diffusion of 2AQ in Ionic Liquid and in Ordinary Solvents. The pump light pulse for the fluorescence excitation was linearly polarized. By measuring the parallel and perpendicular components of the emitted fluorescence signals, we can calculate the fluorescence anisotropy at a given time delay t, r(t), by

r(t) )

I|(t) - I⊥(t) I|(t) + 2I⊥(t)

(1)

where I|(t) and I⊥(t) are the observed parallel and perpendicular fluorescence intensities at the time delay t. A typical set of observed parallel and perpendicular fluorescence decay curves is shown in Figure 6. In the upper panel, the parallel and perpendicular intensities, I|(t) and I⊥(t), are indicated by solid and dotted curves, respectively. The anisotropy values r(t) calculated from I|(t) and I⊥(t) by using eq 1 are shown in the lower panel. The anisotropy value is about 0.15 at time 0 and decreases with time, indicating the formation of aligned excited molecules by linearly polarized excitation light and subsequent decay of the alignment by the rotational diffusion of the excited molecules. By fitting the calculated anisotropy decay curve, we can obtain the rotational diffusion time of the probe molecule 2AQ. After the photoexcitation at 266 nm, as mentioned earlier, the singlet state immediately formed by the excitation is

4918 J. Phys. Chem. B, Vol. 111, No. 18, 2007

Iwata et al.

Figure 7. Dependence of the rotational diffusion time of 2AQ on the viscosity of the bulk solvent.

converted internally to a different singlet state, which serves as the initial state of the fluorescence emission. Because 2AQ is not a highly symmetrical molecule, it is likely that the direction of the transition moment for the absorption (Sn-S0) does not match the direction of the emission (S0-S1). This may explain why the calculated anisotropy value for time 0 is different from 0.4. The rotational diffusion time of a fluorescent molecule changes depending on the microscopic solvation environment. The rotational diffusion time can be a good index characterizing the strength of the solute-solvent and solvent-solvent interactions and their distribution. The Debye-Stokes-Einstein equation gives a simple relation between the rotational diffusion time τrot and the viscosity η

τrot )

ηV kT

(2)

where V is the hydrodynamic volume of the solute molecule, k is the Boltzmann constant, and T is the temperature. We measured the rotational diffusion time of 2AQ in various aromatic and nonaromatic solvents including ionic liquids and try to examine the solvation environment in each solvent. We used bmim[PF6], bmim[BF4], bmimTf2N, emimSO4Et, N1888Tf2N, P666 14Tf2N, P666 14Cl, acetonitrile, tetradecane, squalane, and liquid paraffin as the solvent. The rotational diffusion time of 2AQ measured in these solvents is plotted against the viscosity of the bulk solvent in Figure 7. The viscosity values for bmim[PF6],26 bmim[BF4],27 acetonitrile,28 and squalane29 were obtained from the literature. All the other viscosity values were measured with the viscometer mentioned in the Experimental section. In Figure 7, there is a positive correlation between the solvent viscosity and the rotational correlation times for the results from the molecular solvents, even if there is more than 400 times difference between the viscosity of acetonitrile (0.37 mPa s) and that of liquid paraffin (160 mPa s). If we include the rotational diffusion times observed in ionic liquids, however, the correlation seems to be lost. For ionic liquids, microscopic solvation environment represented by the rotational diffusion time is not well correlated with the macroscopic solvent properties such as viscosity. It should be noted that rotational diffusion time of 2AQ measured in aromatic ionic liquids

Figure 8. Dependence of the diffusion time of 2AQ on molar ratio of ionic liquid in the binary mixture of ionic liquid and acetonitrile. The results for four aromatic ionic liquids are compared.

(bmim[PF6], bmim[BF4], bmimTf2N, and emimSO4Et) are located in the upper area compared with other molecular solvents of similar viscosity in Figure 7, while there is no significant deviation observed for the nonaromatic ionic liquids N1888Tf2N and P666 14Cl. The rotational diffusion in P666 14Tf2N is as slow as in the aromatic ionic liquids. It is possible that aromaticity of the imidazolium cation contributes to the π-π interaction between 2AQ and the cation and stabilizes the 2AQ-cation complex, which decelerates the rotational motion of 2AQ. This result is consistent with the observation of the broad component in the transient fluorescence spectrum of 2AQ in bmim[PF6], omim[PF6], and emimSO4Et, which we think indicates the formation of the stable solute-solvent complex. It has been reported that some imidazolium-based ionic liquids form liquid clathrates when mixed with aromatic hydrocarbons.30 The 1:2 or 2:1 mixture of 1,3-dimethylimidazolium hexafluorophosphate (dmim[PF6]) and benzene forms a crystal in which benzenes are aligned between the two parallel π-planes of the imidazolium cations.30,31 Solute-solvent complexes with similar structures are probably formed in dilute solutions of 2AQ in imidazolium ionic liquids. If the 2AQ molecule is dissolved in binary mixtures of two solvents, the property of the solvent can be changed continuously when the mixing ratio of the two solvents is changed. We measured the rotational diffusion time of 2AQ in the mixed solvents of acetonitrile and aromatic ionic liquids. Four aromatic ionic liquids, bmim[PF6], bmim[BF4], bmimTf2N, and bmimSO4Et, were mixed with acetonitrile with various molar ratios and used as the solvent when the rotational diffusion time was measured. The results are shown in Figure 8. For the four aromatic ionic liquids, the observed rotational diffusion time of 2AQ has similar dependence on the molar ratio of the mixed acetonitrile. When the molar ratio of acetonitrile is increased, the rotational diffusion time starts decreasing steeply. For bmim[PF6], the rotational diffusion time changes from 2.0 to 0.86 ns when the molar ratio of acetonitrile is changed from 0 to 0.02. It drops again to 0.39 ns when the ratio is changed to 0.04. The presence of the 4% acetonitrile is enough to accelerate the rotation of the solute molecule by a factor of 5. The effect of the addition of acetonitrile, however, is not observed for the mixture of nonaromatic ionic liquids. The molar

Solute-Solvent Interaction of 2-Aminoquinoline

J. Phys. Chem. B, Vol. 111, No. 18, 2007 4919 They have also indicated that the aromaticity of the imidazolium cation plays a crucial role in the local structure formation, most probably through the π-π interaction. It seems that the imidazolium-based ionic liquids are, though studied extensively as prototype ionic liquids, better regarded as aromatic ionic liquids that have distinct properties characteristic of the aromaticity of the cation. Acknowledgment. This work is supported by Grant-in-Aid for Creative Scientific Research (No.11NP0101) from Japan Society for the Promotion of Science (JSPS), and Grant-in-Aid for Scientific Research on Priority Areas (Area 452, No. 17073003) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government. References and Notes

Figure 9. Dependence of the diffusion time of 2AQ on molar ratio of ionic liquid in the binary mixture of ionic liquid and acetonitrile. The results for nonaromatic ionic liquids are compared.

ratio dependence of the rotational diffusion time of 2AQ for the mixture of three nonaromatic ionic liquids, N1888Tf2N, P666 14Tf2N, and P666 14Cl and acetonitrile is shown in Figure 9. The results for the squalane-acetonitrile mixture are also shown in the figure. Unlike the aromatic ionic liquids, there is no steep decrease of the rotational diffusion time observed when small amounts of acetonitrile are added to the nonaromatic ionic liquids. Instead, a steady decrease, which may be represented by a linear dependence, is observed for all the mixing ratios from 0 to 1. The results in Figure 9 indicate that the nonaromatic ionic liquids and acetonitrile form uniform mixtures, when monitored in a molecular scale by the rotational diffusion time. The present experiment suggests that neither the ionic liquid nor acetonitrile forms a specific local structure at any molar ratio. For the aromatic ionic liquid, however, the results were different. The steep decrease of the rotational diffusion time, shown in Figure 8, suggests that the ionic liquid and acetonitrile do not mix freely at the molecular level. It is possible that rotational diffusion process is accelerated by the addition of acetonitrile because the acetonitrile molecules destroys the rigid local structure of the aromatic ionic liquid in which the 2AQ is solvated, probably between the parallel π planes. Or, though it is unlikely, the 2AQ molecule stabilized through the π-π interaction may be “extracted” to a microdroplet of acetonitrile in which the rotational diffusion is much faster. Either way, the structure of the mixtures of the aromatic ionic liquid and acetonitrile are not uniform. The present results have given further support for our working hypothesis that specific local structures are formed in imidazolium-based ionic liquids and that those ionic liquids are not liquids in the conventional sense but are “nanostructured fluids”.

(1) Ozawa, R.; Hamaguchi, H. Chem. Lett. 2001, 736. (2) Hayashi, S.; Ozawa, R.; Hamaguchi, H. Chem. Lett. 2003, 32, 498. (3) Saha, S.; Hayashi, S.; Kobayashi, A.; Hamaguchi, H. Chem. Lett. 2003, 32, 740. (4) Ozawa, R.; Hayashi, S.; Saha, S.; Kobayashi, A.; Hamaguchi, H. Chem. Lett. 2003, 32, 948. (5) Katayanagi, H.; Hayashi, S.; Hamaguchi, H.; Nishikawa, K. Chem. Phys. Lett. 2004, 392, 460. (6) Shigeto, S.; Hamaguchi, H. Chem. Phys. Lett. 2006, 427, 329. (7) Hamaguchi, H.; Saha, S.; Ozawa, R.; Hayashi, S. ACS Symp. Ser. 2005, 901, 68. (8) Hamaguchi, H.; Ozawa, R. AdV. Chem. Phys. 2005, 131, 85. (9) Lopes, J. N. A C.; Pa´dua, A. A. H. J. Phys. Chem. B 2006, 110, 3330. (10) Samanta, A. J. Phys. Chem. B 2006, 110, 13704. (11) Karmakar, R.; Samanta, A. J. Phys. Chem. A 2002, 106, 6670. (12) Karmakar, R.; Samanta, A. J. Phys. Chem. A 2003, 107, 7340. (13) Arzhantsev, S.; Ito, N.; Heitz, M.; Maroncelli, M. Chem. Phys. Lett. 2003, 381, 278. (14) Saha, S.; Mandal, P. K.; Samanta, A. Phys. Chem. Chem. Phys. 2004, 6, 3106. (15) Mandal, P. K.; Samanta, A. J. Phys. Chem. B 2005, 109, 15172. (16) Arzhantsev, S.; Jin, H.; Ito, N.; Maroncelli, M. Chem. Phys. Lett. 2006, 417, 524. (17) Ito, N.; Arzhantsev, S.; Heitz, M.; Maroncelli, M. J. Phys. Chem. B 2004, 108, 5771. (18) Ito, N.; Arzhantsev, S.; Maroncelli, M. Chem. Phys. Lett. 2004, 396, 83. (19) Chakrabarty, D.; Chakraborty, A.; Hazra, P.; Seth, D.; Sarkar, N. Chem. Phys. Lett. 2004, 397, 216. (20) Karmakar, R.; Samanta, A. Chem. Phys. Lett. 2003, 376, 638.S (21) Mataga, N.; Chosrowjan, H.; Taniguchi, S. J. Photochem. Photobiol., C 2005, 6, 37. (22) Ishikawa, H.; Iwata, K.; Hamaguchi, H. J. Phys. Chem. A 2002, 106, 2305. (23) Huddleston, J. G.; Willauer, H. D.; Swatloski, R. P.; Visser, A. E.; Rogers, R. D. Chem. Comm. 1998, 1765. (24) Kovi, P. J.; Capomacchira, A. C.; Schulman, S. G. Anal. Chem. 1972, 44, 1161. (25) Wolleben, J.; Testa, A. C. J. Phys. Chem. 1975, 79, 1137. (26) Tomida, D.; Kumagai, A.; Qiao, K.; Yokoyama, C. Int. J. Thermophys. 2006, 27, 39. (27) Zhou, A;, Wang, L. S.; Chen, H. P. J. Chem. Eng. Data 2006, 51, 905. (28) Hickey, K.; Waghorne, W. E. J. Chem. Eng. Data 2001, 46, 851. (29) Tripathi, N. Int. J. Thermophys. 2005, 26, 693. (30) Holbrey, J. D.; Reichert, W. M.; Nieuwenhuyzen, M.; Sheppard, O.; Hardacre, C.; Rogers, R. D. Chem. Comm. 2003, 476. (31) Deetlefs, M.; Hardacre, C.; Nieuwenhuyzen, M.; Sheppard, O.; Soper, A. K. J. Phys. Chem. B 2005, 109, 1593.