Journal of the Korean Physical Society, Vol. 61, No. 9, November 2012, pp. 1555∼1559
Fluorescence Correlation Spectroscopy Study on Room-temperature Ionic Liquids Seoncheol Cha and Doseok Kim∗ Department of Physics, Sogang University, Seoul 121-742, Korea (Received 2 December 2011, in final form 23 May 2012) Unique fluorescence behavior was reported in room-temperature ionic liquids having imidazolium cations. Due to the molecular electronic energy level of the imidazolium cation, fluorescence in the visible range from pure ionic liquids is not expected, but was readily observed. Fluorescence correlation spectroscopy (FCS) was used to find the nature of this unique fluorescence by investigating the fluorescence fluctuation and the number density of fluorophores. FCS signal was observed in pure room-temperature ionic liquids having different cations and anions, which is considered to come from the molecular aggregate of room-temperature ionic liquid. The size and the number density of aggregates in pure ionic liquids were measured by using FCS. PACS numbers: 87.64.Ni Keywords: Fluorescence spectroscopy, Correlation spectroscopy, Viscosity, Diffusion coefficient DOI: 10.3938/jkps.61.1555
I. INTRODUCTION Room-temperature ionic liquids (ILs) are a class of materials consisting only of cations and anions but existing in liquid phase at room temperature and atmospheric pressure. For example, the melting temperature of a prototypical ionic liquid consisting of 1-butyl-3methylimidazolium cation and hexafluorophosphate anion ([BMIM][PF6 ], Fig. 1) is 283 K [1]. Most ionic liquids have almost negligibly small vapor pressure and remain in liquid state without evaporation even at high vacuum conditions and elevated temperatures [1-9]. As they are chemically very stable and can dissolve many inorganic, organic, and polymeric materials, they have been known as environment-friendly ‘green solvents’ to replace conventional organic solvents [2-9]. Inorganic molten salts, such as sodium chloride at liquid phase where the temperature is higher than Tm = 801 K, are known to have homogeneous bulk structures in terms of the radial distribution functions of the consisting ions (e.g., Na+ and Cl− ) [10,11]. Figure 1 shows the chemical structures of the ionic liquid molecules investigated in this report: [BMIM][BF4 ], [OMIM][BF4 ], and [BMIM][PF6 ]. In contrast to the inorganic molten salts, ions consisting ILs are bulkier and asymmetric in shape. This molecular structure has influence on the physical properties of ILs to lower their melting points, as it increases the interionic distance to reduce the Coulomb interaction and its asymmetric shape hinders crystal∗ E-mail:
[email protected]; Fax: +82-2-711-4518
Fig. 1. Anions and cations of 1-alkyl-3-methylimidazolium based ionic liquids. (a) 1-butyl-3-methylimidazolium and (b) 1-butyl-3-octylimidazolium anions and (c) tetrafluoroborate and (d) hexafluorophosphate cations.
lization. The alkyl chain attached to the cation is nonpolar and hydrophobic, and the unique heterogeneous bulk structures resulting from these dissimilar moieties, like nano-structural organizations or molecular aggregates consisting of polar and nonpolar domains of ILs, were suggested by molecular dynamic simulations and X-ray or neutron scattering experiments [12-14]. Ionic liquids having 1-alkyl-3-methylimidazolium ([Cn MIM]) cations also have shown unique fluorescence behaviors [15-19]. As it consists of single imidazolium core with alkyl chain attached, electronic absorption of [Cn MIM] falls naturally in the UV (t , < F (t) >2t
(2)
where F (t) is the fluorescence intensity in the detection volume. An analytic form of Eq. (2) could be obtained under the assumptions that (1) the origin of the fluorescence fluctuation is the free-diffusion dye in the liquid media, and (2) the beam profile is Gaussian. 1 G(τ ) = 1 + ¯ N
τ 1+ τd
1
1+
r02 z02
2
,
(3)
τ τd
where τd = r02 /4D, and r0 and z0 are the lateral and the longitudinal distances (the distance from the focal
Fluorescence Correlation Spectroscopy Study on Room-temperature · · · – Seoncheol Cha and Doseok Kim
-1557-
Table 1. Obtained physical parameter from measured FCS data. Ionic Liquid [BMIM][BF4 ] [BMIM][PF6 ] [OMIM][BF4 ]
Diffusion Coefficient (µm2 /s) 3.4 ± 0.1 1.2 ± 0.1 0.75 ± 0.2
Concentration (nM) 16.0 ± 10 27.5 ± 11 86 ± 34
Viscosity (cP) [28-34] 110 – 219 207 – 371 325 – 439
Hydrodynamic Radius of Particle (˚ A) 3.0 – 5.9 4.9 – 8.7 6.6 – 8.9
Fig. 3. (Color online) (a) Absorption spectra and (b) emission spectra of three different room-temperature ionic liquids excited at 520 nm. The excitation line is not shown.
point at which the intensity drops down to 1/e of its maximum value), respectively. N is the average number of dye molecules in the confocal volume, and D is the diffusion coefficient. Figure 2 shows the measured correlation function of 1 nM TAMRA dye molecules dissolved in water solution. The size of TAMRA dye molecules in water is well known, and r0 and z0 were determined as described previously [24]. (G(0) − 1) is inversely proportional to the number of dye molecules in the detection volume as expected by Eq. (2). With r0 = 330 nm and z0 = 2.3 µm, the detection volume is about 1 fL, and with 1-nM concentration of TAMRA, there is only ∼1 molecule in the above detection volume. G(τd ) ∼ 1.9 in Fig. 2 agrees with the above dye concentration, and τd ∼ 0.1 ms also agrees with the diffusion time of the TAMRA dye molecules in the detection volume.
IV. RESULTS & DISCUSSIONS The absorption spectra of three different ILs are shown in Fig. 3(a). These ILs have strong absorption in the UV region from the electronic transition in the imidazole ring [15-17,20]. As reported previously, unexpected weak absorption was observed at visible wavelengths longer than 450 nm [15-17]. The fluorescence spectra in Fig. 3(b) excited at 520 nm are similar between these three ILs, and show a Stokes shift of ∼200 nm. As a rough estimate, the fluorescence from [BMIM][BF4 ] excited at 520 nm is ∼30 times stronger than the Raman scattering signal from the OH-stretch mode of water detected with
Fig. 4. (Color online) Selected FCS results for pure ionic liquids. Dots are experimental data and solid lines are fitted results.
the same spectrometer in the cuvette filled with pure water. (Raman signal from water: ∼0.1 count, RTIL: 3 counts). The fluorescence intensity from [OMIM][BF4 ] is appreciably stronger than those from [BMIM][BF4 ] and [BMIM][PF6 ]. The visible fluorescence emission region was then investigated by FCS by using 532-nm laser excitation. The measured correlation functions of pure [BMIM][BF4 ], [BMIM][PF6 ], [OMIM][BF4 ] are shown in Fig. 4, and the obtained fit parameters are listed in Table 1. As already mentioned, only fluorescence fluctuations from very low concentrations of dye (∼few nM) could normally be observed by FCS technique because of signal-to-noise ratio. Pure IL samples in this study correspond to a few M in concentration. As this concentration suggests (G(0) − 1) ∼ 10−9 from Eq. (3), they shouldn’t show any measurable correlation function above the background noise.
-1558-
Journal of the Korean Physical Society, Vol. 61, No. 9, November 2012
But the correlation functions values in Fig. 4 were about (G(0) − 1) ∼ 0.01, suggesting that the concentration of fluorescent species in these pure ILs is only a few hundred nM. The concentrations of fluorescent species in ILs were compared between these three samples. The number of fluorescent species in [BMIM][BF4 ] and [BMIM][PF6 ] are not very different, and it is more than four times larger in [OMIM][BF4 ]. We propose that the chain length attached to the imidazole ring is related to the number of fluorophores. The ILs having longer carbon chains are known to form the aggregates more easily [13], with the sizes of the aggregates also related to the chain length from the results obtained using X-ray scattering and molecular dynamics simulations [13,14]. The diffusion dynamics of fluorescent species could also be studied by FCS technique. The time scale of the correlation decay of ∼10−3 s indicates that that the correlation originates solely from the translational diffusion motion of the fluorescent species in the ILs, not from the photophysics like triplet-state transition or photobleaching [24,27]. To proceed further, these diffusion times can be used to determine the size of the fluorescent species by Stokes-Einstein relation provided the viscosity of the medium is known. The exact size determination is limited by the error in the reported values of the viscosity in Table 1 [28-34]. The reported viscosity values of ILs differ widely in Table 1, due to the nature of IL material of which the viscosity is very sensitive to water impurity [35]. With this error in mind, the hydrodynamic radius of the fluorescent species in [BMIM][BF4 ] is smallest, and is larger for [BMIM][PF6 ] and [OMIM][BF4 ], as shown in Table 1. These hydrodynamic radii are comparable to, or a bit larger than the values of the corresponding cations in the ILs. These measured hydrodynamic radii can also be compared with the size of nanostructures suggested by other studies. For example, the length scale of the aggregate structures from molecular dynamic simulation is about 11 – 20 ˚ A for [Cn MIM][PF6 ] (n = 2 to 12) ionic liquids [13]. The size of the nanostructure from X-ray scattering A study of [Cn MIM][Cl] (n = 4 to 10) is about 13 – 27 ˚ [14]. These values are all a bit larger than the radius of the fluorescence species deduced from the FCS, presumably fluorescent species identified by FCS is not exactly the same as the nanostructure in the simulation of X-ray scattering. The reason for the difference originates from the difference between the hydrodynamic radius and the van der Walls radius. Furthermore, Stokes-Einstein relation assumes a perfectly spherical shape of diffusion material, but the aggregate of molecules may have an asymmetric structure. A more quantitative comparison of the sizes between theses ILs in the future would require taking into account the subtle change of the focal volume following change of experimental condition, such as the refractive index of the samples [36-38].
V. CONCLUSION Fluorescence correlation spectroscopy was performed to study imidazolium-based pure ILs having different cations and anions. From the measured correlation functions, the diffusion coefficient and the concentration of the fluorescent species of these ILs were estimated. The fluctuation could be originated from the diffusional motion of the fluorescent species, which were presumed to consist of the cations of the IL molecules forming molecular aggregates.
ACKNOWLEDGMENTS D. Kim acknowledges support by the National Research Foundation (NRF) grant funded by the Korea government (MEST, Ministry of Education, Science and Technology) No. 2011-0017435.
REFERENCES [1] J. D. Holbrey and K. R. Seddon, J. Chem. Soc., Dalton Trans., 2133 (1999). [2] P. Wasserscheid and T. Welton, Ionic Liquids in Synthesis (Wiley-VCH, West Sussex, 2003). [3] N. V. Plechkova, R. D. Rogers and K. R. Seddon, Ionic Liquids: From Knowledge to Application (Oxford University Press, New York, 2009). [4] R. Shelden, Chem. Commun. 37, 2399 (2011). [5] J. G. Huddleston and R. D. Rogers, Chem. Commun. 34, 1765 (1998). [6] S. Dai, Y. H. Ju and C. E. Barnes, J. Chem. Soc., Dalton Trans., 1201 (2001). [7] M. C. Buzzeo, R. G. Evans and R. G. Compton, ChemPhysChem 5, 1106 (2004). [8] P. Wang, S. M. Zakeeruddin, J. Moser and M. Gratzel, J. Phys. Chem. B 107, 13280 (2003). [9] H. Itoh, K. Naka and Y. Chujo, J. Am. Chem. Soc. 126, 3026 (2004). [10] F. G. Edwards, J. E. Enderby, R. A. Howe and D. I. Page, J. Phys. C: Solid State Phys. 8, 3483 (1975). [11] S. Biggin, M. Gay and J. E. Enderby, J. Phys. C: Solid State Phys. 17, 977 (1984). [12] Y. Wang and G. A. Voth, J. Am. Chem. Soc. 127, 12192 (2005). [13] J. N. A. C. Lopes and A. A. H. Padua, J. Phys. Chem. B 110, 3330 (2006). [14] A. Triolo, O. Russina, H.-J. Bleif and E. D. Cola, J. Phys. Chem. B 111, 4641 (2007). [15] A. Paul, P. K. Mandal and A. Samanta, Chem. Phys. Lett. 402, 375 (2005). [16] A. Paul, P. K. Mandal and A. Samanta, J. Phys. Chem. B 109, 9148 (2005). [17] T. Singh and A. Kumar, J. Phys. Chem. B 112, 4079 (2008). [18] X. Chen, J. Liu and J. Wang, J. Phys. Chem. B 115, 1524 (2011).
Fluorescence Correlation Spectroscopy Study on Room-temperature · · · – Seoncheol Cha and Doseok Kim [19] H. Izawa, S. Wakizono and J. Kadokawa, Chem. Commun. 46, 6359 (2010). [20] R. Katoh, Chem. Lett. 36, 1256 (2007). [21] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd ed. (Springer Science+Bussiness Media, New York, 2006). [22] J. H. Werner, S. N. Baker and G. A. Baker, Analyst 128, 786 (2003). [23] J. Gao, G. A. Baker, P. C. Hillesheim, S. Dai, R. W. Shaw and S. M. Mahurin, Phys. Chem. Chem. Phys. 13, 12395 (2011). [24] S. Cha, S. H. Kim and D. Kim, J. Korean Phys. Soc. 56, 1315 (2010). [25] S. H. Kim, T. Shim, D. Kim and B. Z. Tang. J. Korean Phys. Soc. 56, 1264 (2010). [26] M. Magde, E. Elson and W. W. Webb, Phys. Rev. Lett. 29, 705 (1972). [27] O. Krichevsky and G. Bonnet, Rep. Prog. Phys. 65, 251 (2002). [28] K. R. Seddon, A. Stark and M. Torres, Clean Solvents: Alternative Media for Chemical Reaction and Processing (Oxford University Press, New York, 2002).
-1559-
[29] J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D. Willauer, G. A. Broker and R. D. Rogers, Green Chem. 3, 156 (2001). [30] J. Wang, Y. Tian, Y. Zhao and K. Zhuo, Green Chem. 5, 618 (2003). [31] Z. Jiqin, C. Jian, L. Chengyue and F. Weiyang, J. Chem. Eng. Data 52, 812 (2007). [32] S. N. Baker, G. A. Baker, M. A. Kane and F. V. Bright, J. Phys. Chem. B 105, 9663 (2001). [33] K. R. Harris, M. Kanakubo and L. A. Woolf, J. Chem. Eng. Data 52, 1080 (2007). [34] J. Restolho, A. P. Serro, J. L. Mata and B. Saramago, J. Chem. Eng. Data 54, 950 (2009). [35] J. A. Widegren, A. Laesecke and J. W. Magee, Chem. Commun. 1610 (2005). [36] I. Gregor, D. Patra and J. Enderlein, ChemPhysChem 6, 164 (2005). [37] J. Enderlein, I. Gregor, D. Patra, T. Dertinger and U. B. Kaupp, ChemPhysChem 6, 2324 (2005). [38] S. T. Hess and W. W. Webb, Biophys. J. 83, 2300 (2002).
