J. Phys. Chem. B 2007, 111, 12829-12833
12829
Anion Conformation of Low-Viscosity Room-Temperature Ionic Liquid 1-Ethyl-3-methylimidazolium Bis(fluorosulfonyl) Imide Kenta Fujii,† Shiro Seki,‡ Shuhei Fukuda,§ Ryo Kanzaki,§ Toshiyuki Takamuku,† Yasuhiro Umebayashi,§ and Shin-ichi Ishiguro*,§ Department of Chemistry and Applied Chemistry, Faculty of Science and Engineering, Saga UniVersity, Honjo-machi, Saga 840-8502, Japan, Department of Chemistry, Faculty of Science, Kyushu UniVersity, Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan, and Materials Science Research Laboratory, Central Research Institute of Electric Power Industry, Komae, Tokyo 201-8511, Japan ReceiVed: June 4, 2007; In Final Form: August 23, 2007
Anion conformation of a low-viscosity room-temperature ionic liquid 1-ethyl-3-methylimidazolium bis(fluorosulfonyl) imide (EMI+FSI-) has been studied by Raman spectra and theoretical DFT calculations. Three strong Raman bands were found at 293, 328, and 360 cm-1, which are ascribed to the FSI- ion. These Raman bands show significant temperature dependence, implying that two FSI- conformers coexist in equilibrium. This is supported by theoretical calculations that the FSI- ion is present as either C2 (trans) or C1 (cis) conformer; the former gives the global minimum, and the latter has a higher SCF energy of about 4 kJ mol-1. Full geometry optimizations followed by normal frequency analyses show that the observed bands at 293, 328, and 360 cm-1 are ascribed to the C2 conformer. The corresponding vibrations at 305, 320, and 353 cm-1 were extracted according to deconvolution of the observed Raman bands in the range280-400 cm-1 and are ascribed to the C1 conformer. The enthalpy ∆H° of conformational change from C2 to C1 was experimentally evaluated to be ca. 4.5 kJ mol-1, which is in good agreement with the predicted value by theoretical calculations. The bis(trifluoromethanesulfonyl) imide anion (TFSI-) shows a conformational equilibrium between C1 and C2 analogues (∆H° ) 3.5 kJ mol-1). However, the profile of the potential energy surface of the conformational change for FSI- (the F-S-N-S dihedral angle) is significantly different from that for TFSI- (the C-S-N-S dihedral angle).
Introduction Room-temperature ionic liquids (RTILs) have been attracting attention as green solvents, as well as electrolytes for batteries and capacitors, that can be alternatives to conventional molecular solvents. However, among various RTILs, those involving the TFSI- ions are rather viscous as the electrolyte must be dealt with, and those of low viscosity have been explored. Recently, it is found that RTILs involving bis(fluorosulfonyl) imide (FSI-) show a significantly lower viscosity than the TFSI- analogues (e.g., EMI+FSI-, 18 mPa s, and EMI+TFSI-, 33 mPa s).1,2 The FSI- ion, {FS(O2)}2N-, has a molecular structure similar to that of the TFSI- ion, {CF3S(O2)}2N-. We reported that the 1-ethyl-3-methylimidazolium (EMI+) TFSI- ionic liquid involves two TFSI- conformers of C1 (cis) and C2 (trans) symmetries.3 It is suggested that the conformational change of TFSI- leads to a significant decrease in the melting point of its ionic liquids. The low viscosity of the RTILs involving the FSIion might also be related to conformation and its change. From a structural viewpoint, the FSI- ion, like TFSI-, is plausible to give the C1 and C2 conformers with respect to the F-S-NS-F skeleton. Indeed, the C1 conformer is found in crystals of M+FSI- (M+ ) Li+, K+, C6H6‚Ag+),4-6 whereas the C2 conformer is found in crystals of M+FSI- (M+ ) Cs+, CHCl3‚ Ph3C+, Ph3PH+, and (CH3)3Pb+ and a series of Xe(II) * To whom correspondence should be addressed. E-mail: analsscc@ mbox.nc.kyushu-u.ac.jp. † Saga University, Honjo-machi, Saga 840-8502, Japan, ‡ Central Research Institute of Electric Power Industry. § Kyushu University.
derivatives).7-12 This evidently indicates that the conformation strongly depends on an ionic environment around the FSI- ion. Furthermore, the sole C1 conformer of TFSI- is involved in EMI+TFSI- crystals,13 whereas both C1 and C2 conformers are present in the liquid and glassy states.14 This fact implies that the conformation of the FSI- ion in the liquid state at room temperature is different from that in crystals at low temperature. However, no information on the conformation of the FSI- ion in the liquid state has been obtained. In this work, we therefore investigated the conformation of the FSI- ion in the liquid state for EMI+FSI- using Raman and IR spectra and DFT calculations. Experimental Section Materials. 1-Ethyl-3-methylimidazolium bis(fluorosulfonyl) imide (EMI+FSI-) of spectroscopic grade (Dai-ichi Kogyo Seiyaku Co. Ltd.) was used without further purification. The water content was checked by a Karl Fischer test to be less than 100 ppm. The density of EMI+FSI-, measured using a densimeter (Kyoto Electronics, DA-310), is 1.5196 g cm-1 at 298.15 K. Raman and ATR-IR Spectroscopy. Raman spectra were measured using an FT-Raman spectrometer (Perkin-Elmer GXR) equipped with a Nd:YAG laser at 1064 cm-1. The optical resolution was 2.0 cm-1 and spectral data were accumulated 256 times to achieve a sufficiently high signal-to-noise ratio. The sample liquid was filled in a quartz cell thermostated at a given temperature. IR spectra were measured at 298 K using
10.1021/jp074325e CCC: $37.00 © 2007 American Chemical Society Published on Web 10/17/2007
12830 J. Phys. Chem. B, Vol. 111, No. 44, 2007
Fujii et al.
Figure 1. Torsion energy potential surface as a function of the S-NS-F dihedral angle of the FSI- ion obtained on the basis of B3LYP/ 6-31G(d) theory. One of two dihedral angles is fixed at 70°.
an FT-IR spectrometer (Perkin-Elmer GX-R) equipped with a KBr beam splitter and a single reflection ATR cell (Specac, GoldenGate) with KRS-5 lenses. The sample was placed on a diamond reflection element embedded in a plate made of tungsten carbide. The optical resolution was 2.0 cm-1, and spectral data were accumulated 64 times. The sample room was filled with dry nitrogen gas during the measurement. Raman spectra obtained were deconvoluted to extract single bands. A single band is assumed to be represented as a pseudo Voigt function, fv(ν) ) γfL(ν) + (1 -γ)fG(ν), where fL(ν) and fG(ν) stand for Lorentzian and Gaussian components, respectively, and the parameter γ (0 < γ < 1) is the fraction of the Lorentzian component. The intensity I of a single band was evaluated according to I ) γIL + (1 - γ)IG, where IL and IG denote integrated intensities of the Lorentzian and Gaussian components, respectively. A nonlinear least-squares curve-fitting program based on the Marquardt-Levenberg algorism15,16 was used throughout the analyses. Total intensity is thus obtained as the sum of intensities of single bands, which are calculated using the optimized band parameters. Total intensity Icalcd(ν) thus calculated was compared with the observed one Iobsd(ν), and the goodness of fitting was estimated in terms of the Hamilton R factor, R ) Σ{Iobsd(ν) - Icalcd(ν)}2/ΣIobsd(ν)2}1/2. Observed intensities were indeed reproduced with R < 0.01. The uncertainties were estimated by taking into account the standard error of the refined band parameters on the basis of the error propagation analysis. Theoretical Calculations. Evaluation of the torsion potential energy surface (PES) with respect to the F-S-N-S dihedral angle and full geometry optimization followed by normal frequency analyses for the single FSI- ion were carried out on the basis of Hartree-Fock (HF) theory, as well as density functional theory, taking into account an electron correlation effect with Becke’s three parameter and the Lee-Yang-Parr correlation function (B3LYP), coupled with various basis sets.17,18 All calculations were carried out using the Gaussian 03 program package.19 Results and Discussion Theoretical Calculations. As noted in the Introduction, the FSI- ion, like TFSI-, gives either C1 (cis) or C2 (trans) conformer; i.e., terminal F atoms locate either at the cis or trans position with respect to the F-S-N-S-F skeleton in crystals. Two S-N-S-F dihedral angles of the C1 conformer are 70.7
Figure 2. Typical Raman and IR spectrum of EMI+FSI- observed in the range 200-1700 cm-1.
CHART 1: Geometries of C1 (cis) and C2 (trans) Conformers of the FSI- Ion Obtained by Optimization Using B3LYP/6-311+G(3df) Theory
and -73.6° in the K+FSI- crystal,5 and the corresponding angles of the C2 conformer are 69.9 and 75.8° in the Cs+FSI- crystal.9 As both C1 and C2 conformers involve a similar local structure of an S-N-S-F dihedral angle close to 70°, theoretical calculations were carried out by fixing one dihedral angle θ ) 70° and by varying another over the range -180° < θ < 180° to obtain the torsion potential energy surface (PES). A typical PES profile calculated on the basis of the B3LYP/6-31G(d) theory is shown in Figure 1. As seen, the global and local energy minima appear at θ ) 75 and -75°, respectively, indicating that the C2 conformer (θ ) 75°) is preferred to the C1 conformer (-75°). Using sets of θ’s as initial parameters, geometry optimizations for the C1 (70, -75°) and C2 (70, 75°) conformers were then carried out. The selected structural parameters (bond length, bond angle, and dihedral angle), dipole moments, and relative SCF energies thus optimized on the basis of the HF and B3LYP levels of theory coupled with various basis sets are summarized in Table 1. The same calculations have been reported for TFSI- in our previous paper.3 It has been found that the HF level of theory coupled with the 6-31G(d) basis set well reproduces the observed structural parameters but not the observed band frequencies. On the other hand, all calculations using the B3LYP level of theory well reproduce the observed band frequencies. Among others, the calculation coupled with the 6-311+G(3df)
A Low-Viscosity Room-Temperature Ionic Liquid
J. Phys. Chem. B, Vol. 111, No. 44, 2007 12831
TABLE 1: Structural Parameters (Bond Lengths/Å, Bond Angles/deg, Dihedral Angles/deg) of the C1 and C2 Conformers of FSI-, Dipole Moment/D, SCF Energy ∆ESCF/kJ mol-1, Enthalpy ∆H°/kJ mol-1, Entropy ∆S°/J K-1 mol-1, and Gibbs Energy ∆G°/kJ mol-1 for the Conformational Change from C2 to C1, and Scaling Factor for C2 C1
C2 B3LYP
param N-S S-O S-F
HF for 6-31G(d)
6-31G(d)
1.564 1.420 1.580
1.607 1.457 1.644
6-311G(d) 1.600 1.448 1.660
B3LYP
6-311+ G(d) 1.601 1.450 1.671
6-311+G(3df) Bond Lengths 1.574 1.430 1.620
HF for 6-31G(d)
6-31G(d)
1.568 1.423 1.576
1.609 1.456 1.642
6-311G(d) 1.603 1.448 1.659
6-311+G(d) 1.603 1.450 1.671
6-311+ G(3df) 1.577 1.432 1.616
S-N-S
125.50
122.45
124.53
124.85
Bond Angles 126.12
123.67
121.23
122.83
123.48
124.38
S2-N1-S6-F9 S6-N1-S2-F5 F5-S2-S6-F9
-87.44 58.16 -26.36
-84.67 61.63 -20.51
-78.72 70.74 -7.21
-88.99 66.12 -20.77
Dihedral Angles -85.41 66.56 -17.17
74.25 74.25 140.18
73.76 73.74 139.02
74.56 74.61 140.93
76.38 76.37 144.30
74.42 74.39 140.56
dipole moment ∆ESCF ∆H° ∆S° ∆G° scaling factor
1.916 6.4 6.3 6.5 4.5
1.496 1.5 3.6 3.2 2.7
0.911 2.8 3.8 15.1 -0.4
Other Parameters 0.750 1.374 3.5 2.5 3.5 3.6 4.2 7.6 2.4 1.5
basis set gave bond lengths similar to those calculated on the basis of the HF level of theory. Thus, we concluded that the B3LYP level of theory coupled with the 6-311+G(3df) basis set satisfactorily reproduces both structural parameters and band frequencies for the TFSI- conformers. The same applies also for the FSI- conformers, as will be discussed in a later section. If one looks at S-N-S-F dihedral angles in Table 1, a set of θ ) 74.4 and 74.4° for the C2 conformer and a set of θ ) 66.6 and -85.4° for the C1 conformer are obtained by the B3LYP/ 6-311+G(3df) calculation. Optimized geometries of the conformers are shown in Chart 1. Here, note that the corresponding C-S-N-S dihedral angles of TFSI- are 93.7 and 93.7° for the C2 conformer and 130.7 and -88.2° for the C1 conformer.3 With regard to the C2 conformer, the F atom within FSI- is less deviated from the S-N-S plane than the corresponding CF3 group within TFSI- to give a significantly shorter F5‚‚‚O7 (or F9‚‚‚O4) distance (3.28 Å) than the corresponding C‚‚‚O distance (3.85 Å) of TFSI-. With the C1 conformer, the F-S‚‚‚S-F dihedral angle of FSI- is -17.2°, which is also significantly different from the corresponding C-S‚‚‚S-C dihedral angle (42.5°) of TFSI-, also leading to a shorter F‚‚‚F distance of FSI- than the corresponding C‚‚‚C distance of TFSI-. These imply that the intramolecular repulsion within FSI- is weaker than that within TFSI-. It is also noted that, as seen in Figure 1, the PES profile of the FSI- ion shows two local energy maxima of about 14 kJ mol-1 at θ ) 0 and 150°. In contrast, the corresponding PES of TFSI- shows local energy maxima of 28 and 5 kJ mol-1 at θ ) -20 and 150°, respectively. The PES profiles imply that the activation enthalpy of conformational change is much less for TFSI- than FSI-. Energy barriers along the N-S bond imply that the FSI- ion equally rotates in both directions, whereas the TFSI- ion cannot and is swinging like a pendulum to change its conformation. Different mechanisms of the conformational change of FSI- and TFSI- might lead to different dynamics of their ionic liquids, as will be discussed in a following paper.20 Raman and IR Spectra. Typical Raman and IR spectra of EMI+FSI- are depicted in Figure 2. Here, Raman bands arising from intramolecular vibrations of the EMI+ ion have been established,21 and relatively strong Raman bands observed at
0.599
0.767
0.038
0.449
0.918
1.001
1.015
1.029
0.006
0.990
293, 328, 360, 456, 484, 526, 572, 726, 832, 1217, and 1367 cm-1 are ascribed to the vibrations of the FSI- ion. Also, strong IR bands observed at 564, 727, 826, 1099, 1165, 1174, 1360, and 1376 cm-1 are ascribed to the vibrations of the FSI- ion. Theoretical Raman and IR bands of a given conformer are predicted by full geometry optimizations followed by normal frequency analyses. Predicted frequencies of the C2 conformer of FSI- were then compared with the observed ones to give the scaling factor. The scaling factor is 0.918 on the basis of the HF level of theory with the 6-31G(d) basis set. The corresponding scaling factors on the basis of the B3LYP theory with 6-31G(d), 6-311G(d), 6-311+G(d), and 6-311+G(3df) basis sets are 1.001, 1.015, 1.029, and 0.990, respectively. The scaling factors are all close to unity, indicating that the B3LYP calculations well reproduce the observed frequencies, although the calculations are carried out for a single molecule. Observed and calculated frequencies are listed in Table S1 as Supporting Information. As the 6-311+G(3df) basis set better reproduce the bond lengths, we thus concluded that the B3LYP level of theory gives a satisfactory prediction among others. The EMI+FSI- ionic liquid gives three strong bands at 293, 328, and 360 cm-1 ascribable to the FSI- ion. As seen in Figure 3, these observed bands are evidently asymmetric, suggesting that molecular vibrations of two species are overlapped. The observed spectrum was then deconvoluted according to the procedure as described in the Experimental section. The spectrum can be satisfactorily deconvoluted into three strong bands at 293, 328, and 360 cm-1 and relatively weak bands at 305, 320, and 353 cm-1. Theoretical bands of the C1 and C2 conformers of FSI- obtained on the basis of the B3LYP/6311+G(3df) calculation are also shown in Figure 3, together with those of EMI+. The observed bands at 293, 328, and 360 cm-1 may correspond to the predicted bands at 275, 316, and 350 cm-1, respectively, of the C2 conformer, and the bands at 305, 320, and 353 cm-1 to the predicted bands at 278, 316, and 344 cm-1, respectively, of the C1 conformer. The same applies also for the IR spectra. Theoretical IR spectra of the C1 and C2 conformers, together with an observed one, in the range 10001500 cm-1 are given in Figure 4. The observed spectrum shows bands at around 1170 and 1370 cm-1, and the band-splitting is
12832 J. Phys. Chem. B, Vol. 111, No. 44, 2007
Fujii et al.
Figure 5. Raman spectra of EMI+FSI- obtained at various temperatures.
Figure 3. Observed Raman bands of EMI+FSI- and theoretical Raman bands of the FSI- and EMI+ conformers.
energy than the C1 conformer. The Gibbs energy, enthalpy, and entropy of conformational change were then evaluated as follows. Integrated intensity I(ν) of a given single band centered at ν cm-1 is represented as I(ν) ) (J(ν))m, where J and m denote the Raman scattering coefficient and molality of a given conformer; i.e., the relationships IC1(ν) ) (JC1(ν))mC1 and IC2(ν) ) (JC2(ν))mC2 apply for the C1 and C2 conformers, respectively. The equilibrium constant K ()mC1/mC2) for the reaction C2 ) C1 is thus given according to ln K ) ln{IC1(ν)/ IC2(ν)} - ln{JC1(ν)/JC2(ν)}. Combining the equation with -R ln K ) ∆H°/T - ∆S°, we obtain the following equation:
-R ln{IC1(ν)/IC2(ν)} ) ∆H°/T - ∆S° - R{JC1(ν)/JC2(ν)} (1) Here ∆H° and ∆S° denote the enthalpy and entropy of conformational change from C2 to C1 and R is the gas constant. The -R ln{IC1(ν)/IC2(ν)} vs T-1 plots give the ∆H° and -∆S° - R{JC1(ν)/JC2(ν)} values as the slope and intercept, respectively. The procedure was carried out using the 293 and 360 cm-1 bands of the C2 conformer, coupled with the corresponding 305 and 353 cm-1 bands of the C1 conformer, to give the ∆H° values of 4.3(1) and 4.8(2) kJ mol-1, respectively, as shown in Figure 6. On the other hand, as the total molality mT of FSI- is given as mT ) mC1 + mC2, we obtain the relationship as Figure 4. Observed IR bands of EMI+FSI- and theoretical IR bands of the FSI- and EMI+ conformers.
also consistent with the presence of two conformers in equilibrium. Theoretical prediction indeed indicates that the C2 conformer gives bands at a lower frequency side than the corresponding bands of the C1 conformer. Thermodynamics of Conformational Change. Figure 5 shows Raman spectra of EMI+FSI- measured at various temperatures in the range 250-400 cm-1. As seen, intensities of the 293, 328, and 362 cm-1 bands weaken with increasing temperature and isosbestic points are seen at around 301, 324, and 353 cm-1. This evidently indicates that two conformers coexist in equilibrium. The variation profile is consistent with the theoretical prediction that the C2 conformer has a lower SCF
IC1(ν) ) mTJC1(ν) - {JC1(ν)/JC2(ν)}IC2(ν)
(2)
The IC1(ν) vs IC2(ν) plots at various temperature give a straight line, if we assume that the J value is independent of temperature. The JC1(ν)/JC2(ν) value is indeed obtained as a slope using the 360 and 353 bands, and the ∆S° value is obtained by subtracting the JC1(353)/JC2(360) value from the intercept of the straight line in Figure 6. The ∆S° value thus obtained is 19(5) J K-1 mol-1, and the Gibbs energy is then calculated to be -1(2) kJ mol-1 at 298 K, according to ∆G° ) ∆H° - T∆S°. The ∆G° value indeed close to zero, also indicating that the C1 and C2 conformers coexist in equilibrium in the liquid state. The dipole moment for the FSI- ion is 1.37 and 0.0 D for the C1 and C2 conformers, respectively, on the basis of the B3LYP/6-311+ G(3df) calculation. As described, the C1 and C2 conformers of FSI- are present almost to the same extent in the liquid at 298 K. The EMI+ ion involves also two conformers,
A Low-Viscosity Room-Temperature Ionic Liquid
J. Phys. Chem. B, Vol. 111, No. 44, 2007 12833 Acknowledgment. This work has been financially supported by Grant-in-Aids for Scientific Research Nos. 17350037, 18850017, 19350033, 19550022, and 19750062 from the Ministry of Education, Culture, Sports, Science, and Technology. Supporting Information Available: Observed Raman and theoretical IR/Raman spectra (Table S1). This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes
Figure 6. -R ln{IC1(ν)/IC2(ν)} vs T-1 plots using bands of the C2 conformer at around 293 (O) and 360 (b) cm-1, together with the corresponding bands of the C1 conformer.
the dipole moments of which are 3.78 and 5.46 D for the planar and nonplanar conformers, respectively, on the basis of the B3LYP/6-311+G(3df) calculation.21 If the homogeneous distribution of conformers is assumed, the stronger dipole-dipole interaction of the C1 conformer with EMI+ ions might lead to the favorable formation of the conformer in the EMI+FSI- ionic liquid at 298 K. The same applies also for the TFSI- ion. The corresponding dipole moments for the C1 and C2 conformers of TFSI- are 4.40 and 0.30 D, respectively. The value for the C1 conformer of TFSI- is even larger than that of FSI-, implying that the C1 conformer is even more favored for TFSI- than for FSI- in their EMI+ ionic liquids. The intermolecular interaction in the bulk is thus supposed to be weaker for EMI+FSI- than EMI+TFSI-. Conclusion It is established from both Raman and IR spectra that the EMI+FSI- ionic liquid involves two conformers of the FSIion in equilibrium at room temperature. Theoretical calculations for the FSI- ion indicate that the C1 (cis) and C2 (trans) conformers respectively give the local and global energy minima with the energy difference of ca. 4.5 kJ mol-1. The Gibbs energy, enthalpy, and entropy of conformational change from C2 to C1 were experimentally evaluated. The obtained enthalpy value is in agreement with the predicted value. The entropy value is positive, and the Gibbs energy is close to zero. It is thus concluded that the two conformers of the FSI- ion coexist in equilibrium in the liquid state. The dipole moments for the FSI- ion are 1.37 and 0.0 D for the C1 and C2 conformers, respectively, on the basis of the B3LYP/6-311+G(3df) calculation. A stronger dipole-dipole interaction is thus expected for the C1 conformer with EMI+ ions, which might lead to the favorable formation of the conformer in the EMI+FSI- ionic liquid at 298 K.
(1) Matsumoto, H.; Sakaebe, H.; Tatsumi, K.; Kikuta, M.; Ishiko, E.; Kono, M. J. Power Sources 2006, 160, 1308. (2) Ishikawa, M.; Sugimoto, T.; Kikuta, M.; Ishiko, E.; Kono, M. J. Power Sources 2006, 162, 658. (3) Fujii, K.; Fujimori, T.; Takamuku, T.; Kanzaki, R.; Umebayashi, Y.; Ishiguro, S. J. Phys. Chem. B 2006, 110, 8179. (4) Beran, M.; Prˇ´ıhoda, J.; Zˇ a´, K, Z.; C ˇ ernı´k, M. Polyhedron 2006, 25, 1292. (5) Zˇ a´, K, Z.; Ru˚zˇicˇka, A.; Michot, Ch. Z. Kristallogr. 1998, 213, 217. (6) Begley, M. J.; Sowerby, D. B.; Verma, R. D.; Vig, A. J. Organometallic Chem. 1994, 481, 243. (7) Hiemisch, O.; Henschel, D.; Blaschette, A.; Jones, P. G. Z. Anorg. Allg. Chem. 1997, 623, 324. (8) Farrugia, L. J.; Holfter, H.; Klapo¨tke, T. M. J. Fluorine Chem. 1996, 78, 51. (9) Hiemisch, O.; Henschel, D.; Jones, P. G.; Blaschette, A. Z. Anorg. Allg. Chem. 1996, 622, 829. (10) Schumacher, G. A.; Schrobilgen, G. J. Inorg. Chem. 1983, 22, 2178. (11) Sawyer, J. F.; Schrobilgen, G. J.; Sutherland, S. J. Inorg. Chem. 1982, 21, 4064. (12) Faggiani, R.; Kennepohl, K.; Lock, C. J. L.; Schrobilgen, G., J. Inorg. Chem. 1986, 25, 563. (13) Choudhury, A. R.; Winternton, N.; Steiner, A.; Cooper, A. I.; Johnson, K. A. CrystEngComm 2006, 8, 742. (14) Lasse`gues, J.; Grondin, J.; Holomb, R.; Johansson, P. J. Raman Spectrosc. 2007, 38, 551. (15) Marquardt, D. W. J. Soc. Ind. Appl. Math. 1963, 11, 431. (16) Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T. Numerical Recipes; Cambridge University Press: Cambridge, U.K., 1989. (17) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (18) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (b) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200. (19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H., Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C. and Pople, J. A. Gaussian 03, rev. B.04; Gaussian, Inc.: Pittsburgh, PA, 2003. (20) Canongia Lopes, J. N.; Shimizu, K.; Pa´dua, A. A. H.; Umebayashi, Y.; Fukuda, S.; Fujii, K.; Ishiguro, S. To be submitted for publication. (21) Umebayashi, Y.; Fujimori, T.; Sukizaki, T.; Asada, M.; Fujii, K.; Kanzaki, R.; Ishiguro, S. J. Phys. Chem. A 2005, 109, 8976.
