Article pubs.acs.org/JPCB
Effects of Dissolved Water on Li+ Solvation in 1‑Ethyl-3methylimidazolium Bis(trifluoromethanesulfonyl)amide Ionic Liquid Studied by NMR Tatsuya Umecky,*,† Toshiyuki Takamuku,† Tomoya Matsumoto,‡ Eiji Kawai,‡ Masaya Takagi,‡ and Toshitaka Funazukuri‡ †
Department of Chemistry and Applied Chemistry, Graduate School of Science and Engineering, Saga University, Honjo-machi, Saga 840-8502, Japan ‡ Department of Applied Chemistry, Faculty of Science and Engineering, Chuo University, 1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan S Supporting Information *
ABSTRACT: 1H and 7Li NMR chemical shifts of 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide−water solutions in the presence and absence of lithium bis(trifluoromethanesulfonyl)amide were determined at 293.2 K over a wide range of water concentrations from 0.0156 to 1.16 mol kg−1. These results revealed the attractive interaction between water molecule and Li+ as well as the hydrogen bonding among water molecules. Moreover, self-diffusion coefficients of water, 1-ethyl3-methylimidazolium cation, Li+, and bis(trifluoromethanesulfonyl)amide anion in the ionic liquid solutions at various water contents were determined by 1H, 7Li, and 19F NMR techniques. It was found that Li+ is averagely hydrated by eight water molecules in the ionic liquid solutions. Furthermore, 7Li longitudinal relaxation times of Li+ in the ionic liquid solutions at 293.2 K were measured with two different magnetic fields and various water contents. The mean one-jump distances of Li+ in the ionic liquid solutions were estimated from the correlation times and the self-diffusion coefficients. A comparison between the hydrodynamic radius and the mean one-jump distance of Li+ suggested the formation of water channels in the ionic liquid solutions.
■
depends on the Li+ concentration. A Li+ ion is solvated by a few mono- and/or bidentated TFSA− at low Li+ concentrations, whereas Li+ ions may aggregate by TFSA− bridging at high Li+ concentrations. On the other hand, water is a key substance as one of the impurities in ILs and as a counter-solvent in liquid− liquid extractions with hydrophobic ILs. It has given a way to the understanding that a trace of moisture as an impurity leads to significant changes in physico and electrochemical properties of hydrophobic ILs.10−14 A water molecule more strongly interacts with Li+ than other inorganic ions, such as Na+ and K+, resulting in lower hydration enthalpies.15 From not only a
INTRODUCTION
Ionic liquids (ILs) incorporating organic cations and organic or inorganic anions exhibit several excellent properties, for example, low volatility, low flammability, and high thermal and chemical stabilities, over traditional organic solvents. Hence, ILs are drawing attention as alternatives for volatile organic solvents in wide fields of chemistry and industry.1 Such microscopic properties as solution structure and ion dynamics of the mixtures of ILs and inorganic electrolytes like lithium salts are essential in designing and improving ILs as electrolytic media for secondary batteries and as solvents for liquid−liquid extractions. There are so far many experimental and simulation studies on solvation structures surrounding Li + in ILs composed of bis(trifluoromethanesulfonyl)amide anion (TFSA−).2−9 The solvation structure of Li+ in TFSA−-ILs © 2013 American Chemical Society
Received: September 18, 2013 Revised: November 17, 2013 Published: November 28, 2013 16219
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
of residual Br− in the IL was lower than the minimum limit of the detections. Sample Preparation. IL solutions involving LiTFSA at a constant concentration and water at some different concentrations were prepared as follows. First, two ILs at very different moisture contents were prepared. One was the relatively dried IL in which water is hardly dissolved (dry IL). The dry IL was obtained by drying under a reduced pressure at ∼333 K for more than 72 h. The other IL (wet IL) contained sufficient water under an ambient atmosphere. The wet IL was prepared by agitating for 24 h and then leaving to stand for 24 h at ∼333 K. The moisture contents in the two ILs determined by a Karl Fischer titration were 384 and 2.09 × 104 ppm, respectively. LiTFSA was dissolved into both dry and wet ILs. The LiTFSA concentrations were 0.0132 mol kg−1 for the dry IL and 0.0134 mol kg−1 for the wet IL. To prepare LiTFSA-doped IL-water solutions at different water contents, the dry and wet ILs where LiTFSA was dissolved were mixed with each other at various ratios. The LiTFSA concentrations of the solutions were in the range from 0.0132 to 0.0134 mol kg−1, which can be regarded as a constant concentration. The water concentrations of the LiTFSA-doped IL solutions were in a wide range from 0.0156 to 1.15 mol kg−1. A series of LiTFSA-free IL solutions at different water contents were also prepared for comparison. The water concentrations of the LiTFSA-free IL solutions ranged from 0.0213 to 1.16 mol kg−1. To avoid the change in water concentration, each IL solution was expeditiously sealed into an NMR tube with 4.1 mmϕ outer diameter (Shigemi, SC002). NMR Measurements. A double tube constructed by an inner tube and an outer tube filled with reference-doped D2O solution was used for NMR measurements. The sample solutions were sealed into the inner tube. The sample tube was kept for 30 min at 293.2 ± 0.1 K in the NMR probe to reach the constant temperature before measurements. 1 H and 7Li chemical shifts were observed with the JEOL ECA-500 spectrometer, where the resonance frequencies for 1H and 7Li were 500.2 and 194.4 MHz, respectively. Sodium 3(trimethylsilyl)-1-propanesulfonate at ∼0.020 mol dm−3 and lithium chloride at ∼0.020 mol dm−3 were used as external references for 1H and 7Li chemical shift measurements, respectively. 1 H, 7Li, and 19F self-diffusion coefficients (D) were determined with the JEOL ECP-300 spectrometer, where the resonance frequencies for 1H, 7Li, and 19F were 300.5, 116.8, and 282.7 MHz, respectively. Self-diffusion coefficients of C2mim+ (Dcation), TFSA− (Danion), Li+ (DLi), and water (Dwater) were estimated from 1H (for the H2 atom), 19F, 7Li, and 1H spectra, respectively. Pulsed-field gradient spin−echo sequence was used for the D measurements. The gradient magnitude and the diffusion time were fixed at 5.0 T m−1 and 50 ms, respectively. The applied gradient magnitudes were calibrated by using the self-diffusion coefficient of water at 293.2 K previously reported.18 A set of more than 20 different gradient pulse widths up to 5.0 ms was adopted for each measurement. 7 Li longitudinal relaxation times (T1,Li) were measured with both NMR instruments above. T1,Li values were determined by a standard inversion recovery sequence. In the T1,Li measurements, a set of more than 20 different recovery times was employed.
scientific but also an engineering standpoint, hence, it is important and imperative to elucidate how water molecules effect on solvation structure and dynamics of Li+ in the hydrophobic IL solutions. The solvation structure and dynamics of Li+ in hydrophobic IL−water binary mixtures are critical issues. However, such an experimental result is only available in the more recent literature.16 The present work was aimed at experimentally clarifying the effect of water dissolved into IL on solvation structure and dynamics of Li+. We focus on 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ([C2mim][TFSA]) as a hydrophobic IL. Figure 1 illustrated the structure and hydrogen
Figure 1. Structure of [C2mim]+ cation with the notation of the hydrogen atoms.
atom numbering for C2mim+ cation. [C2mim][TFSA] may contain moisture of 1.98 × 104 ppm at a maximum.14 The stability of TFSA− of this IL against water is relatively high among hydrophobic ILs involving other anions. The solvation structure of Li+ has been clarified in TFSA−-based IL as stated above. NMR chemical shifts, self-diffusion coefficients, and longitudinal relaxation times of the IL solutions were measured as a function of water content. The macroscopic properties of the binary [C2mim][TFSA]-water system, such as viscosity, are available in the previous literature.13
■
EXPERIMENTAL SECTION Chemicals. Chemicals were used as purchased from SigmaAldrich Chemicals: 1-methylimidazole (99%), 1-bromoethane (>99%), ethyl acetate (>99.7%, grade for HPLC), lithium bis(trifluoromethanesulfonyl)amide (LiTFSA) (99.95%), sodium 3-(trimethylsilyl)-1-propanesulfonate (97%), lithium chloride (>99.99%), and heavy water (99.9%D). Ultrapure water produced with Millipore Direct-Q system was used as light water solvent. Ionic Liquid Synthesis. [C2mim][TFSA] was synthesized according to the literature.17 Briefly, 1-methylimidazole was mixed with the equimolar amount of 1-bromoethane and then vigorously stirred at ∼343 K for more than 72 h. After the resulting liquid was washed with ethyl acetate, the precipitation of [C2mim]Br was obtained. Ethyl acetate phase was decanted, and ethyl acetate adsorbed to the precipitate was removed by heating at ∼343 K for 3 h under vacuum. To exchange Br− by TFSA−, an aqueous solution of LiTFSA was added to an aqueous solution of [C2mim]Br. After the biphasic mixture was stirred for 24 h, the upper aqueous phase was decanted, and the lower nonaqueous phase was washed more than five times with water. The nonaqueous phase was heated under vacuum at ∼343 K for 50 h or more to remove any excess water. The colorless viscous liquid thus obtained was identified as an intended [C2mim][TFSA] by 1H and 13C NMR spectroscopic measurements. There was no major impurity. In addition to this result, ion chromatography analysis and electrospray ionization mass spectrometry showed that the concentration 16220
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
■
Article
RESULTS AND DISCUSSION Chemical Shifts. The chemical shifts observed (δobs) were corrected for the susceptibility of the solutions into those δcor by using the following equation:19 4π δcor = δobs − (χ − χvol,ref ) × 106 (1) 3 vol,sam Here, χvol,sam and χvol,ref are volume susceptibilities of sample and reference solutions, respectively. The correction for the volume susceptibility of sample solution was explained in the Supporting Information. Additionally, the numerical data of δcation, δwater, and δLi corrected for the susceptibility are summarized in the Supporting Information (Table S1). In Figure 2, the 7Li chemical shifts (δLi) corrected for the
Figure 2. 7Li chemical shifts (δLi) of the IL solutions with LiTFSA as a function of the molality of water (mwater). The upper axis shows the molar ratio of [H2O]/[Li].
Figure 3. 1H chemical shifts of the H2, H4, H5, H6, H7, and H8 atoms of C2mim+ (δcation) in the IL solutions with (closed symbols) and without (open symbols) LiTFSA as a function of the molality of water (mwater).
susceptibility are plotted as a function of the molality of water (mwater). The δLi value prominently shifts to the low field of ∼1.2 ppm with the increase in the mwater. This is attributed to the interaction between Li+ and water molecules in the IL solutions. Hence, the interaction of Li+ with water molecules is stronger than that with TFSA−. TFSA− anions in the first coordination shell of Li+ are replaced by water molecules. The increase in δLi is pronounced in the lower water content range of [H2O]/[Li] ≈ 20. Thus, the replacement of TFSA− by water molecules immediately occurs below ∼0.3 mol kg−1 ([H2O]/ [Li] ≈ 20), suggesting the preferential solvation of Li+ by water molecules in the IL solutions. Figure 3 shows the 1H chemical shifts of C2mim+ (δcation) corrected for the susceptibility in the IL solutions with and without LiTFSA as a function of the mwater. The presence of Li+ in the IL scarcely influences the δcation values, indicating no strong interaction of C2mim+ with Li+. The δcation of all of the H atoms of C2mim+ shifts to the low field with the increase in the water content. The effect of water is remarkably observed for the H2 atom among all of the H atoms of C2mim+. According to the previous report,20 the shift of the H2 atom of C2mim+ caused by the hydrogen bonding with methanol is ∼0.3 ppm. The δcation shifts to ∼0.04 ppm for the Li+-doped IL and the Lifree solutions with increasing mwater up to 1.15 and 1.16 mol kg−1, which correspond to the mole fractions of water of 0.32 and 0.41, respectively. The present results are comparable with the variation in the 1H chemical shifts of C2mim+ below the mole fraction of methanol of ∼0.4 for the [C2mim][TFSA]− methanol systems due to the hydrogen bonding. The small
increase in the δcation values of C2mim+ with the increase in the water concentration declares that C2mim+ more strongly interacts with water than does TFSA−. Figure 4a indicates the 1H chemical shifts of water (δwater) corrected for the susceptibility in the IL solutions with and without LiTFSA as a function of the mwater. The δwater shifts to ∼0.4 ppm toward the low field with increasing mwater. The variation in the δwater with mwater is larger as compared to those of C2mim+ in the solutions as shown in Figure 3. This is attributable to the interaction of water molecule with Li+ and other water molecules by the hydrogen bonding. This means the hydration of Li+ and the self-assembly of water molecules. Thus, a water-rich domain may form in the ILs, and Li+ mainly exists in the domain. Interestingly, the 1H chemical shifts of water in the IL solutions with and without LiTFSA are significantly different from each other. The δwater for the IL solutions without lithium salt monotonically increases with increasing mwater as shown by the quadratic function (a dotted line) in Figure 4a. In contrast, the mwater dependence of δwater for the IL solutions with Li salt is strange; the δwater is almost unchanged in the mwater range from 0.1 to 0.3 mol kg−1. The effect of LiTFSA on the δwater arises from the Li+−water interaction. To make more clear the effect of Li+ on the δwater, the differences (δwater,doped − δwater,free) of the 1H chemical shifts of water between the IL solutions with and without LiTFSA are plotted against the mwater (Figure 4b). Here, the δwater,doped 16221
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
Figure 5. (a) Self-diffusion coefficients (D) of cation (blue triangles), anion (purple diamonds), Li+ (red circles), and water (black squares) in the IL solutions with (closed symbols) and without (open symbols) LiTFSA as a function of the molality of water (mwater). (b) Plots of the D of C2mim+ (blue triangles), TFSA− (purple diamonds), Li+ (red circles), and water (black squares) of the IL solutions with (closed symbols) and without (open symbols) LiTFSA against the inverse of the solvent viscosity (η−1).
1
Figure 4. (a) H chemical shifts of water in the IL solutions with (δwater,doped, closed symbols) and without (δwater,free, open symbols) LiTFSA as a function of the molality of water (mwater). The dotted line presents the quadric curve obtained by a least-squares fit on δwater for the IL solutions without LiTFSA. (b) Difference (δwater,doped − δwater,free) of 1H chemical shifts of water between the IL solutions with and without LiTFSA as a function of mwater. δwater,doped values were experimentally estimated, whereas δwater,free values were obtained by the quadratic regression. The upper axis shows the molar ratio of [H2O]/ [Li].
of LiTFSA. C2mim+ cation may not remarkably interact with Li+. On the other hand, TFSA− anions should interact with Li+ to form the solvation structure of Li+ in the solutions at the low water contents as discussed above. However, the effect of Li+ on the Danion is not clearly observed because of the higher concentration of free TFSA− than the solvating ones. At the low mwater, the DLi values are somewhat smaller than the Dcation and Danion values. The order (DLi < Danion < Dcation) of the selfdiffusion coefficients previously reported21−30 can be interpreted by the slower migration of Li+ than the IL cation and anion under the conditions. The increase in the water content leads to the more remarkable increase in DLi than Dcation and Danion. At the highest mwater of 1.15 mol kg−1, the DLi is comparable with Dcation and larger by a factor of 1.5 than Danion, resulting in the order of Danion < Dcation ≈ DLi. The translational motion of Li+ becomes more active on the addition of water. In Figure 5a, the Dwater values are larger by a factor of ∼10 than the self-diffusion coefficients of the three ions whether Li+ is present or not. In addition, the change in Dwater with mwater is quite different from those of the ions. Without Li+ ions, the Dwater of the IL solution quickly declines and reaches a plateau as the mwater increases. The decrease in the Dwater for the [C2mim][TFSA] solutions in the present study is different from the increase in the Dwater for [C4mim][TFSA].31 This presumably means that water molecules more easily assemble in [C2mim][TFSA] than in [C4mim][TFSA]. It is conceivable that [C2mim][TFSA] has larger voids, where water molecules may accommodate, due to the shorter alkyl chain than
values for the IL solutions with LiTFSA were experimentally determined, and the δwater,free values were obtained by the interpolation of the IL solutions without LiTFSA using a quadratic function. When the mwater increases, the steep increase in the δwater,doped − δwater,free values at mwater < ∼0.1 mol kg−1 is followed by the gradual decrease in the difference. Below mwater = ∼0.1 mol kg−1, which corresponds to the molar ratio [H2O]/ [Li] = ∼8, water molecules gradually replace TFSA− on Li+. The Li+−water interaction is saturated at mwater = ∼0.1 mol kg−1. This suggests that Li+ ion is hydrated with eight water molecules averaged. In the range of mwater > ∼0.1 mol kg−1, water molecules do not further interact with Li+ but do with other water molecules. The increase in the number of the water−water interaction leads to the weakening of the effect of LiTFSA on the δwater. This is the reason for the diminishing of the difference of δwater,doped − δwater,free above mwater = ∼0.1 mol kg−1. Self-Diffusion Coefficients. The numerical data of Dcation, Danion, DLi, and Dwater of the IL solutions with and without LiTFSA were listed in the Supporting Information (Table S2). The plots of Dcation, Danion, DLi, and Dwater in the IL solutions with and without LiTFSA against the mwater are presented in Figure 5a. Dwater at the lowest mwater (0.0156 mol kg−1) could not be measured because of the very weak signal of water. The Dcation and Danion are not significantly influenced by the addition 16222
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
by LiTFSA. Importantly, the rcation and ranion values are almost kept until the molar ratios (∼0.46 and ∼0.69) of water to IL for the IL solutions with and without LiTFSA, respectively. These results are the evidence that the IL cation and anion do not strongly interact with Li+. The rLi shrinks from 0.26 to 0.14 nm with the increase in the amount of water. The rLi values are comparable with that of aqueous solution (0.24 nm) and thus larger than the radius of Li+ found in crystals, which ranges from 0.059 to 0.092 nm depending on the coordination number.15 The size variation arises from the change in the solvation structure of Li+. The reason for the decrease in the rLi is that the TFSA−, which solvates to Li+, is replaced by water molecules as water is added into the IL solutions. As the result, the Li+ diffusion activated on the addition of water (Figure 5) arises not only from the fluidization of the entire systems but also from the reduction of the size of Li+ solvation structure. This finding shows that the transport of Li+ in the IL solution is affected by its solvation structure. In Figure 6, with the increase in the amount of water, the rwater expands from 0.011 to 0.023 nm in the IL solution with LiTFSA and from 0.0091 to 0.025 nm in the solutions without LiTFSA. The rwater values are much smaller than those of the aqueous LiTFSA solution and pure water (0.11 nm for both). The diffusion mechanism of water in the IL solutions may differ from those in the aqueous LiTFSA solution and pure water. The increase in the rwater is attributed to the interactions of water molecule with both Li+ and other water molecules. The deceleration of the diffusion of water molecules in the IL solution without LiTFSA shown in Figure 5 is caused by the formation of the self-assembly of water molecules. For the IL solution with LiTFSA, the sharp rise in the rwater with increasing mwater up to ∼0.1 mol kg−1 may be ascribed to the interaction with Li+, and the moderate change in the rwater above mwater = ∼0.1 mol kg−1 may arise from the hydrogen bonding among water molecules. The Li+−water interaction conducts to the significant difference of the rwater between the IL solutions with and without LiTFSA. The inflection point of the rwater at mwater = ∼0.1 mol kg−1 corresponds to the molar ratio [H2O]/[Li] of ∼8, showing the hydration of Li+ with approximately eight water molecules. This is consistent with the results of the δwater (Figure 4a). In the range of [H2O]/[Li] > ∼8, the effect of the hydrogen bonding among water molecules on the motion of water molecules is larger than that of the addition of LiTFSA. The difference of the rwater between the IL solutions with and without LiTFSA becomes small in this range. The change in the rwater with increasing mwater shows both hydration of Li+ and hydrogen-bonded assembly among water molecules as well as δwater. Thus, water-rich domain may be formed in the IL solutions. Referring to the mwater dependence of Dwater of the IL solution with LiTFSA (Figure 5a), the deceleration of the motion of water molecules in the IL solutions below mwater = ∼0.1 mol kg−1 is attributed to the strong Li+−water interaction. In contrast, the retardation of the motion of water molecules above mwater = ∼0.1 mol kg−1 mainly arises from the hydrogen bonding among water molecules. Because the present [C2mim][TFSA] has the highly ordered liquid structure,34 therefore, the smaller rwater in the IL solutions than the aqueous solution and pure water may result from the occurrence of diffusion channel of water in the IL solution due to the partially segregated structure of IL. Longitudinal Relaxation Times. The numerical data of the 7Li longitudinal relaxation times (T1,Li) of Li+ in the IL
[C4mim][TFSA]. In the case of the IL solutions with LiTFSA, the Dwater rapidly decreases with increasing water content up to mwater = ∼0.1 mol kg−1, and then barely increases in the middle range from ∼0.1 to ∼0.3 mol kg−1, and the Dwater moderately decreases in the higher mwater range above ∼0.3 mol kg−1. The mechanism of the mwater dependence of Dwater is discussed in the following paragraph in terms of the hydration of Li+ and the self-assembly with water molecules. The D value is correlated with solvent viscosity (η) and temperature (T) as shown by the Stoke−Einstein relation, which can be obtained by combining classical hydrodynamics of Stokes law and kinetic theory of Einstein relation:
D=
kBT cπηr
(2)
where r is the hydrodynamic radius of the diffusing species and c is a constant. When a molecule diffuses with solvent molecules due to the friction between the diffusing and surrounding bodies (stick boundary limit), c should be 6.32 When a molecule migrates without any friction (slip boundary limit), c is set to be 4.33 On the basis of the Stoke−Einstein principle, the D values for all of the chemical species are plotted against the inverse of η (η−1) in Figure 5b. The η values were obtained from the interpolation of η for the binary [C2mim][TFSA]−water system as a function of the mole fraction of water.13 From eq 2, generally, the reduction of η contributes to the increase in the D value. However, the Dwater values do not come under this case. The increase in the D for C2mim+, TFSA−, and Li+ with increasing η−1 suggests that the acceleration of the ions is ascribed to the fluidization of the IL solutions on adding of water. As the η−1 value increases, the Dcation and Danion linearity grow, whereas the DLi shows a nonlinear relationship with η−1 (Figure 5b). This fact suggests that other reasons should contribute to the increase in DLi with increasing water content. On the assumption that each diffusing species is in sphere shape, their r values in the IL solutions can be determined from eq 2 with c = 6. Figure 6 gives the r values of C2mim+ (rcation), TFSA− (ranion), Li+ (rLi), and water (rwater) as a function of the mwater. The rcation (0.12 nm) and the ranion (0.20 nm) do not significantly change on adding LiTFSA. Thus, the diffusion mechanism of C2mim+ and TFSA− is not remarkably influenced
Figure 6. Hydrodynamic radii (r) of C2mim+ (blue triangles), TFSA− (purple diamonds), Li+ (red circles), and water (black squares) in the IL solutions with (closed symbols) and without (open symbols) LiTFSA as a function of the molality of water (mwater). The upper axis shows the molar ratio of [H2O]/[Li]. 16223
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
τc,Li values decrease with increasing mwater; notably, the τc,Li values significantly drop in the low mwater range. The mwater dependence of τc,Li is exponential-like and seems to be analogous to that on δLi (Figure 2), despite the opposite direction. Relationship between Self-Diffusion Coefficients and Correlation Times of Li+. 7Li nucleus relaxes by diffusing motion but not by tumbling one. The τc,Li value obtained from the T1,Li can be related to the DLi through the mean one-jump distance (RLi) of Li+:
solutions determined by using two different magnetic field strengths of 7.0 and 11.7 T are listed in the Supporting Information (Table S3). Figure 7 presents the T1,Li of Li+ in the
RLi =
6DLi τc,Li
(4)
Figure 9 shows the mwater dependence of RLi. The RLi at the lowest mwater is 0.33 nm and agrees with the value (∼0.3 nm)
Figure 7. 7Li longitudinal relaxation times (T1,Li) of the IL solutions with LiTFSA obtained by using two different magnetic field strengths of 7.0 T (closed symbols) and 11.7 T (open symbols) as a function of the molality of water (mwater).
IL solutions at 7.0 and 11.7 T as a function of the mwater. The T1,Li values of the IL solutions are ranged between ∼0.2 and ∼1.2 s with a minimum at mwater = ∼0.1 mol kg−1. All of the T1,Li at 11.7 T are larger than those at 7.0 T over the entire water content range studied. The remarkable differences between the T1,Li values at the two magnetic fields mean that Li+ ions in the IL solutions are not under the extreme narrowing condition of 2πvτ ≪ 1. Because 7Li with the spin quantum number of I = 3/2 relaxes through only the quadrupolar interaction, the observed T1,Li under no extreme narrowing condition is related to the correlation time (τc,Li) as follows:35 ⎛ ⎞ τc,Li 4τc,Li 1 ⎟⎟ = A⎜⎜ + 2 2 2 T1,Li 1 + 16π 2v 2τc,Li 2 ⎠ ⎝ 1 + 4π v τc,Li
Figure 9. Mean one-jump distances (RLi) of Li+ in the IL solutions as a function of the molality of water (mwater). The upper axis shows the molar ratio of [H2O]/[Li].
previously reported.30 When a trace of water is dissolved into the IL solution, the jump distance of Li+ suddenly shortens to 0.25 nm. The shortening of the RLi suggests that the distance between Li+ and the solvating species reduces due to the change in the solvation structure of Li+. The TFSA− anions that solvate to Li+ are replaced by water molecules. Above the molar ratio [H2O]/[Li] of ∼6, the change in the RLi values is very small. These jump distances are equal numerically to the hydrodynamic radius rLi (Figure 6). This shows that the jumping motion of Li+ occurs inside its solvation shell. The inflection point of the RLi at the molar ratio [H2O]/[Li] of ∼6 intends that six water molecules contribute to the jump motion of Li+ at the maximum. This finding is reasonable with the hydration structure of Li+ with approximately eight water molecules.
(3)
where A is a constant depending on sample system and v is a resonance frequency. τc,Li was estimated from a pair of the T1,Li measured and the v values on the assumption that the relaxation of 7Li occurs through the same mechanisms under two different magnetic fields. The τc,Li values are plotted against the mwater in Figure 8. All of the τc,Li values are in subnanosecond scale over the water content range studied. The present data for the dry IL solutions well agree with τc,Li in [C2mim][TFSA] solutions at 0.32 mol kg−1 of LiTFSA.30 The
■
CONCLUSIONS We carried out NMR measurements on the hydrophobic IL solutions with and without LiTFSA at 293.2 K in the wide water content range from 0.0156 to 1.16 mol kg−1. The δLi shifted to the low field as water was dissolved; peculiarly, the change in the δLi was remarkable in the low water content range. Li+ ions more strongly interact with water molecules than TFSA− in the IL solutions. The δwater remarkably increased with the water concentration and showed the difference between the values in the presence and absence of LiTFSA. These results were interpreted by that water molecules interact with Li+ as well as the hydrogen bonding among water molecules. As the water content increases, the Dcation, Danion, and
Figure 8. Correlation times (τc,Li) of Li+ in the IL solutions as a function of the molality of water (mwater). 16224
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
(5) Shirai, A.; Fujii, K.; Seki, S.; Umebayashi, Y.; Ishiguro, S.; Ikeda, Y. Solvation of Lithium Ion in N,N-Diethyl-N-methyl-N-(2methoxyethyl)ammonium Bis(trifluoromethanesulfonyl)amide Using Raman and Multinuclear NMR Spectroscopy. Anal. Sci. 2008, 24, 1291−1296. (6) Lassegues, J.-C.; Grondin, J.; Aupetit, C.; Johansson, P. Spectroscopic Identification of the Lithium Ion Transporting Species in LiTFSI-Doped Ionic Liquids. J. Phys. Chem. A 2009, 113, 305−314. (7) Umebayashi, Y.; Mori, S.; Fujii, K.; Tsuzuki, S.; Seki, S.; Hayamizu, K.; Ishiguro, S. Raman Spectroscopic Studies and Ab Initio Calculations on Conformational Isomerism of 1-Buthyl-3-methylimidazolium Bis-(trifluoromethanesulfonyl)amide Solvated to a Lithium Ion in Ionic Liquids: Effects of the Second Solvation Sphere of the Lithium Ion. J. Phys. Chem. B 2010, 114, 6513−6521. (8) Umebayashi, Y.; Hamano, H.; Seki, S.; Minofar, B.; Fujii, K.; Hayamizu, K.; Tsuzuki, S.; Kameda, Y.; Kohara, S.; Watanabe, M. Liquid Structure of and Li+ Ion Solvation in Bis(trifluoromethanesulfonyl)amide Based Ionic Liquids Composed of 1-Ethyl-3-methylimidazolium and N-Methyl-N-propylpyrrolidinium Cations. J. Phys. Chem. B 2011, 115, 12179−12191. (9) Li, Z.; Smith, G. D.; Bedrov, D. Li+ Solvation and Transport Properties in Ionic Liquid/Lithium Salt Mixtures: A Molecular Dynamics Simulation Study. J. Phys. Chem. B 2012, 116, 12801− 12809. (10) Schröder, U.; Wadhawan, J. D.; Compton, R. G.; Marken, F.; Suarez, P. A. Z.; Consorti, C. S.; de Souza, R. F.; Dupont, J. WaterInduced Accelerated Ion Diffusion: Voltammetric Studies in 1-Methyl3-[2,6-(S)-dimethylocten-2-yl]imidazolium Tetrafluoroborate, 1-Butyl3-methylimidazolium Tetrafluoroborate and Hexafluorophosphate Ionic Liquids. New J. Chem. 2000, 24, 1009−1015. (11) Huddleston, J. G.; Visser, A. E.; Reichert, W. M.; Willauer, H. D.; Broker, G. A.; Rogers, R. D. Characterization and Comparison of Hydrophilic and Hydrophobic Room Temperature Ionic Liquids Incorporating the Imidazolium Cation. Green Chem. 2001, 3, 156− 164. (12) Bakers, N.; Baker, G. A.; Bright, F. V. Temperature-Dependent Microscopic Solvent Properties of ‘dry’ and ‘wet’ 1-Butyl-3methylimidazolium Hexafluorophosphate: Correlation with ET(30) and Kamlet−Taft Polarity Scales. Green Chem. 2002, 4, 165−169. (13) Widegren, J. A.; Laesecke, A.; Magee, J. W. The Effect of Dissolved Water on the Viscosities of Hydrophobic Room-Temperature Ionic Liquids. Chem. Commun. 2005, 41, 1610−1612. (14) Jacquemin, J.; Husson, P.; Padua, A. A. H.; Majer, V. Density and Viscosity of Several Pure and Water-Saturated Ionic Liquids. Green Chem. 2006, 8, 172−180. (15) CRC Handbook of Chemistry and Physics, 86th ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 2005. (16) Martins, V. L.; Nicolau, B. G.; Urahata, S. M.; Ribeiro, C. C.; Torresi, R. M. Influence of the Water Content on the Structure and Physicochemical Properties of an Ionic Liquid and Its Li+ Mixture. J. Phys. Chem. B 2013, 117, 8782−8792. (17) Bonhôte, P.; Dias, A.; Papageorgiou, N.; Kalyanasundaram, K.; Grätzel, M. Hydrophobic, Highly Conductive Ambient-Temperature Molten Salts. Inorg. Chem. 1996, 35, 1168−1178. (18) Holz, M.; Heil, S. R.; Sacco, A. Temperature-Dependent SelfDiffusion Coefficients of Water and Six Selected Molecular Liquids for Calibration in Accurate 1H NMR PFG Measurements. Phys. Chem. Chem. Phys. 2000, 2, 4740−4742. (19) Becker, E. D. High-Resolution NMR: Theory and Chemical Applications, 3rd ed.; Academic Press: New York, 2000; pp 83−117. (20) Takamuku, T.; Honda, Y.; Fujii, K.; Kittaka, S. Aggregation of Imidazolium Ionic Liquids in Molecular Liquids Studied by SmallAngle Neutron Scattering and NMR. Anal. Sci. 2008, 24, 1285−1290. (21) Hayamizu, K.; Aihara, Y.; Nakagawa, H.; Nukuda, T.; Price, W. S. Ionic Conduction and Ion Diffusion in Binary Room-Temperature Ionic Liquids Composed of [emim][BF4] and LiBF4. J. Phys. Chem. B 2004, 108, 19527−19532. (22) Nicotera, I.; Oliviero, C.; Henderson, W. A.; Appetecchi, G. B.; Passerini, S. NMR Investigation of Ionic Liquid-LiX Mixtures:
DLi increased in one direction, but the Dwater strangely behaved. The r values of all of the species in the IL solutions were evaluated from their D values. The accelerations of the three ion species are due mostly to the fluidization of the entire systems. In addition, the acceleration of the diffusion of Li+ arises from the replacement of TFSA− by water molecules. The water content dependence of the rwater also supported the hydration of Li+ as well as self-assembly of water molecules by hydrogen bonds. The τc,Li values were determined from the T1,Li values, and the jump distance RLi in the IL solution can be estimated from the DLi and τc,Li values. The RLi decreased from 0.33 to 0.25 nm with increasing the molar ratio [H2O]/[Li] up to ∼6. This indicated the reducing of the distance between Li+ and the solvating species, suggesting the change in the solvation structure of Li+. In conclusion, most of the present results revealed that the solvating species of TFSA− for Li+ in the IL solutions were replaced by water molecules in the solution at around mwater = 0.1 mol kg−1. The concentration corresponds to the molar ratio of [H2O]/[Li] = ∼8. Consequently, a Li+ ion is averagely hydrated by about eight water molecules on the present NMR scales. It becomes clear that Li+ diffuses in the water-rich domain (water channel) and at the same time jumps inside the hydration shell in subnanosecond scale.
■
ASSOCIATED CONTENT
S Supporting Information *
Details of the magnetic susceptibility correction, the numerical data of δcation, δwater, and δLi of the IL solutions with and without LiTFSA (Table S1), the Dcation, Danion, DLi, and Dwater of both IL solutions (Table S2), and the T1,Li of the IL solutions with LiTFSA (Table S3). This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported partly by a Grant-in-Aid for Young Scientists (B) (No. 24750019) from the Japan Society for the Promotion of Science and Dean’s Grant for Progressive Research Projects from Saga University.
■
REFERENCES
(1) Ionic Liquids as Green Solvents; Rogers, R. D., Seddon, K. R., Eds.; ACS Symposium Series 856; American Chemical Society: Washington, DC, 2003. (2) Lassegues, J.-C.; Grondin, J.; Talaga, D. Lithium Solvation in Bis(trifluoromethanesulfonyl)imide-Based Ionic Liquids. Phys. Chem. Chem. Phys. 2006, 8, 5629−5632. (3) Umebayashi, Y.; Mitsugi, T.; Fukuda, S.; Fujimoto, T.; Fujii, K.; Kanzaki, R.; Takeuchi, M.; Ishiguro, S. Lithium Ion Solvation in Room-Temperature Ionic Liquids Involving Bis(trifluoromethanesulfonyl) Imide Anion Studied by Raman Spectroscopy and DFT Calculations. J. Phys. Chem. B 2007, 111, 13028− 13032. (4) Duluard, S.; Grondin, J.; Bruneel, J.-L.; Pianet, I.; Grelard, A.; Campet, G.; Delville, M.-H.; Lassegues, J.-C. Lithium Solvation and Diffusion in the 1-Buthyl-3-methylimidazolium Bis(trifluoromethanesulfonyl)imide Ionic Liquid. J. Raman Spectrosc. 2008, 39, 627−632. 16225
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
Pyrrolidinium Cations and TFSI− Anions. J. Phys. Chem. B 2005, 109, 22814−22819. (23) Saito, Y.; Umecky, T.; Niwa, J.; Sakai, T.; Maeda, S. Existing Condition and Migration Property of Ions in Lithium Electrolytes with Ionic Liquid Solvent. J. Phys. Chem. B 2007, 111, 11794−11802. (24) Hayamizu, K.; Tsuzuki, S.; Seki, S.; Ohno, Y.; Miyashiro, H.; Kobayashi, Y. Quaternary Ammonium Room-temperature Ionic Liquid Including an Oxygen Atom in Side Chain/Lithium Salt Binary Electrolytes: Ionic Conductivity and 1H, 7Li, and 19F NMR Studies on Diffusion Coefficients and Local Motions. J. Phys. Chem. B 2008, 112, 1189−1197. (25) Umecky, T.; Saito, Y.; Okumura, Y.; Maeda, S.; Sakai, T. Ionization Condition of Lithium Ionic Liquid Electrolytes under the Solvation Effect of Liquid and Solid Solvents. J. Phys. Chem. B 2008, 112, 3357−3364. (26) Frömling, T.; Kunze, M.; Schönhoff, M.; Sundermeyer, J.; Roling, B. Enhanced Lithium Transference Numbers in Ionic Liquid Electrolytes. J. Phys. Chem. B 2008, 112, 12985−12990. (27) Le, M. L. P.; Alloin, F.; Strobel, P.; Leprêtre, J.-C.; del Valle, C. P.; Judeinstein, P. Structure-Properties Relationships of Lithium Electrolytes Based on Ionic Liquid. J. Phys. Chem. B 2010, 114, 894−903. (28) Bayley, P. M.; Lane, G. H.; Lyons, L. J.; MacFarlane, D. R.; Forsyth, M. Undoing Lithium Ion Association in Ionic Liquids through the Complexation by Oligoethers. J. Phys. Chem. C 2010, 114, 20569− 20576. (29) Castiglione, F.; Raggh, E.; Mele, A.; Appetecchi, G. B.; Montanino, M.; Passerini, S. Molecular Environment and Enhanced Diffusivity of Li+ Ions in Lithium-Salt-Doped Ionic Liquid Electrolytes. J. Phys. Chem. Lett. 2011, 2, 153−157. (30) Hayamizu, K.; Tsuzuki, S.; Seki, S.; Umebayashi, Y. Nuclear Magnetic Resonance Studies on the Rotational and Translational Motions of Ionic Liquids Composed of 1-Ethyl-3-methylimidazolium Cation and Bis(trifluoromethanesulfonyl)amide and Bis(fluorosulfonyl)amide Anions Their Binary Systems Including Lithium Salts. J. Chem. Phys. 2011, 135, 084505. (31) Rollet, A.-L.; Porion, P.; Vaultier, M.; Billard, I.; Deschamps, M.; Bessada, C.; Jouvensal, L. Anomalous Diffusion of Water in [BMIM][TFSI] Room-Temperature Ionic Liquid. J. Phys. Chem. B 2007, 111, 11888−11891. (32) Einstein, A. Investigations on the Theory of the Brownian Movement; Dover: New York, 1956; pp 68−85. (33) McLaughlin, E. Viscosity and Self-Diffusion in Liquids. Trans. Faraday Soc. 1959, 55, 28−38. (34) Lopes, J. N. A. C.; Pádua, A. A. H. Nanostructural Organization in Ionic Liquids. J. Phys. Chem. B 2006, 110, 3330−3335. (35) Abragam, A. The Principles of Nuclear Magnetism; Oxford University Press: New York, 1989; pp 313−315.
16226
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
Effects of Dissolved Water on Li+ Solvation in 1‑Ethyl-3methylimidazolium Bis(trifluoromethanesulfonyl)amide Ionic Liquid Studied by NMR Tatsuya Umecky,*,† Toshiyuki Takamuku,† Tomoya Matsumoto,‡ Eiji Kawai,‡ Masaya Takagi,‡ and Toshitaka Funazukuri‡ †
Department of Chemistry and Applied Chemistry, Graduate School of Science and Engineering, Saga University, Honjo-machi, Saga 840-8502, Japan ‡ Department of Applied Chemistry, Faculty of Science and Engineering, Chuo University, 1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan S Supporting Information *
ABSTRACT: 1H and 7Li NMR chemical shifts of 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide−water solutions in the presence and absence of lithium bis(trifluoromethanesulfonyl)amide were determined at 293.2 K over a wide range of water concentrations from 0.0156 to 1.16 mol kg−1. These results revealed the attractive interaction between water molecule and Li+ as well as the hydrogen bonding among water molecules. Moreover, self-diffusion coefficients of water, 1-ethyl3-methylimidazolium cation, Li+, and bis(trifluoromethanesulfonyl)amide anion in the ionic liquid solutions at various water contents were determined by 1H, 7Li, and 19F NMR techniques. It was found that Li+ is averagely hydrated by eight water molecules in the ionic liquid solutions. Furthermore, 7Li longitudinal relaxation times of Li+ in the ionic liquid solutions at 293.2 K were measured with two different magnetic fields and various water contents. The mean one-jump distances of Li+ in the ionic liquid solutions were estimated from the correlation times and the self-diffusion coefficients. A comparison between the hydrodynamic radius and the mean one-jump distance of Li+ suggested the formation of water channels in the ionic liquid solutions.
■
depends on the Li+ concentration. A Li+ ion is solvated by a few mono- and/or bidentated TFSA− at low Li+ concentrations, whereas Li+ ions may aggregate by TFSA− bridging at high Li+ concentrations. On the other hand, water is a key substance as one of the impurities in ILs and as a counter-solvent in liquid− liquid extractions with hydrophobic ILs. It has given a way to the understanding that a trace of moisture as an impurity leads to significant changes in physico and electrochemical properties of hydrophobic ILs.10−14 A water molecule more strongly interacts with Li+ than other inorganic ions, such as Na+ and K+, resulting in lower hydration enthalpies.15 From not only a
INTRODUCTION
Ionic liquids (ILs) incorporating organic cations and organic or inorganic anions exhibit several excellent properties, for example, low volatility, low flammability, and high thermal and chemical stabilities, over traditional organic solvents. Hence, ILs are drawing attention as alternatives for volatile organic solvents in wide fields of chemistry and industry.1 Such microscopic properties as solution structure and ion dynamics of the mixtures of ILs and inorganic electrolytes like lithium salts are essential in designing and improving ILs as electrolytic media for secondary batteries and as solvents for liquid−liquid extractions. There are so far many experimental and simulation studies on solvation structures surrounding Li + in ILs composed of bis(trifluoromethanesulfonyl)amide anion (TFSA−).2−9 The solvation structure of Li+ in TFSA−-ILs © 2013 American Chemical Society
Received: September 18, 2013 Revised: November 17, 2013 Published: November 28, 2013 16219
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
of residual Br− in the IL was lower than the minimum limit of the detections. Sample Preparation. IL solutions involving LiTFSA at a constant concentration and water at some different concentrations were prepared as follows. First, two ILs at very different moisture contents were prepared. One was the relatively dried IL in which water is hardly dissolved (dry IL). The dry IL was obtained by drying under a reduced pressure at ∼333 K for more than 72 h. The other IL (wet IL) contained sufficient water under an ambient atmosphere. The wet IL was prepared by agitating for 24 h and then leaving to stand for 24 h at ∼333 K. The moisture contents in the two ILs determined by a Karl Fischer titration were 384 and 2.09 × 104 ppm, respectively. LiTFSA was dissolved into both dry and wet ILs. The LiTFSA concentrations were 0.0132 mol kg−1 for the dry IL and 0.0134 mol kg−1 for the wet IL. To prepare LiTFSA-doped IL-water solutions at different water contents, the dry and wet ILs where LiTFSA was dissolved were mixed with each other at various ratios. The LiTFSA concentrations of the solutions were in the range from 0.0132 to 0.0134 mol kg−1, which can be regarded as a constant concentration. The water concentrations of the LiTFSA-doped IL solutions were in a wide range from 0.0156 to 1.15 mol kg−1. A series of LiTFSA-free IL solutions at different water contents were also prepared for comparison. The water concentrations of the LiTFSA-free IL solutions ranged from 0.0213 to 1.16 mol kg−1. To avoid the change in water concentration, each IL solution was expeditiously sealed into an NMR tube with 4.1 mmϕ outer diameter (Shigemi, SC002). NMR Measurements. A double tube constructed by an inner tube and an outer tube filled with reference-doped D2O solution was used for NMR measurements. The sample solutions were sealed into the inner tube. The sample tube was kept for 30 min at 293.2 ± 0.1 K in the NMR probe to reach the constant temperature before measurements. 1 H and 7Li chemical shifts were observed with the JEOL ECA-500 spectrometer, where the resonance frequencies for 1H and 7Li were 500.2 and 194.4 MHz, respectively. Sodium 3(trimethylsilyl)-1-propanesulfonate at ∼0.020 mol dm−3 and lithium chloride at ∼0.020 mol dm−3 were used as external references for 1H and 7Li chemical shift measurements, respectively. 1 H, 7Li, and 19F self-diffusion coefficients (D) were determined with the JEOL ECP-300 spectrometer, where the resonance frequencies for 1H, 7Li, and 19F were 300.5, 116.8, and 282.7 MHz, respectively. Self-diffusion coefficients of C2mim+ (Dcation), TFSA− (Danion), Li+ (DLi), and water (Dwater) were estimated from 1H (for the H2 atom), 19F, 7Li, and 1H spectra, respectively. Pulsed-field gradient spin−echo sequence was used for the D measurements. The gradient magnitude and the diffusion time were fixed at 5.0 T m−1 and 50 ms, respectively. The applied gradient magnitudes were calibrated by using the self-diffusion coefficient of water at 293.2 K previously reported.18 A set of more than 20 different gradient pulse widths up to 5.0 ms was adopted for each measurement. 7 Li longitudinal relaxation times (T1,Li) were measured with both NMR instruments above. T1,Li values were determined by a standard inversion recovery sequence. In the T1,Li measurements, a set of more than 20 different recovery times was employed.
scientific but also an engineering standpoint, hence, it is important and imperative to elucidate how water molecules effect on solvation structure and dynamics of Li+ in the hydrophobic IL solutions. The solvation structure and dynamics of Li+ in hydrophobic IL−water binary mixtures are critical issues. However, such an experimental result is only available in the more recent literature.16 The present work was aimed at experimentally clarifying the effect of water dissolved into IL on solvation structure and dynamics of Li+. We focus on 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ([C2mim][TFSA]) as a hydrophobic IL. Figure 1 illustrated the structure and hydrogen
Figure 1. Structure of [C2mim]+ cation with the notation of the hydrogen atoms.
atom numbering for C2mim+ cation. [C2mim][TFSA] may contain moisture of 1.98 × 104 ppm at a maximum.14 The stability of TFSA− of this IL against water is relatively high among hydrophobic ILs involving other anions. The solvation structure of Li+ has been clarified in TFSA−-based IL as stated above. NMR chemical shifts, self-diffusion coefficients, and longitudinal relaxation times of the IL solutions were measured as a function of water content. The macroscopic properties of the binary [C2mim][TFSA]-water system, such as viscosity, are available in the previous literature.13
■
EXPERIMENTAL SECTION Chemicals. Chemicals were used as purchased from SigmaAldrich Chemicals: 1-methylimidazole (99%), 1-bromoethane (>99%), ethyl acetate (>99.7%, grade for HPLC), lithium bis(trifluoromethanesulfonyl)amide (LiTFSA) (99.95%), sodium 3-(trimethylsilyl)-1-propanesulfonate (97%), lithium chloride (>99.99%), and heavy water (99.9%D). Ultrapure water produced with Millipore Direct-Q system was used as light water solvent. Ionic Liquid Synthesis. [C2mim][TFSA] was synthesized according to the literature.17 Briefly, 1-methylimidazole was mixed with the equimolar amount of 1-bromoethane and then vigorously stirred at ∼343 K for more than 72 h. After the resulting liquid was washed with ethyl acetate, the precipitation of [C2mim]Br was obtained. Ethyl acetate phase was decanted, and ethyl acetate adsorbed to the precipitate was removed by heating at ∼343 K for 3 h under vacuum. To exchange Br− by TFSA−, an aqueous solution of LiTFSA was added to an aqueous solution of [C2mim]Br. After the biphasic mixture was stirred for 24 h, the upper aqueous phase was decanted, and the lower nonaqueous phase was washed more than five times with water. The nonaqueous phase was heated under vacuum at ∼343 K for 50 h or more to remove any excess water. The colorless viscous liquid thus obtained was identified as an intended [C2mim][TFSA] by 1H and 13C NMR spectroscopic measurements. There was no major impurity. In addition to this result, ion chromatography analysis and electrospray ionization mass spectrometry showed that the concentration 16220
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
■
Article
RESULTS AND DISCUSSION Chemical Shifts. The chemical shifts observed (δobs) were corrected for the susceptibility of the solutions into those δcor by using the following equation:19 4π δcor = δobs − (χ − χvol,ref ) × 106 (1) 3 vol,sam Here, χvol,sam and χvol,ref are volume susceptibilities of sample and reference solutions, respectively. The correction for the volume susceptibility of sample solution was explained in the Supporting Information. Additionally, the numerical data of δcation, δwater, and δLi corrected for the susceptibility are summarized in the Supporting Information (Table S1). In Figure 2, the 7Li chemical shifts (δLi) corrected for the
Figure 2. 7Li chemical shifts (δLi) of the IL solutions with LiTFSA as a function of the molality of water (mwater). The upper axis shows the molar ratio of [H2O]/[Li].
Figure 3. 1H chemical shifts of the H2, H4, H5, H6, H7, and H8 atoms of C2mim+ (δcation) in the IL solutions with (closed symbols) and without (open symbols) LiTFSA as a function of the molality of water (mwater).
susceptibility are plotted as a function of the molality of water (mwater). The δLi value prominently shifts to the low field of ∼1.2 ppm with the increase in the mwater. This is attributed to the interaction between Li+ and water molecules in the IL solutions. Hence, the interaction of Li+ with water molecules is stronger than that with TFSA−. TFSA− anions in the first coordination shell of Li+ are replaced by water molecules. The increase in δLi is pronounced in the lower water content range of [H2O]/[Li] ≈ 20. Thus, the replacement of TFSA− by water molecules immediately occurs below ∼0.3 mol kg−1 ([H2O]/ [Li] ≈ 20), suggesting the preferential solvation of Li+ by water molecules in the IL solutions. Figure 3 shows the 1H chemical shifts of C2mim+ (δcation) corrected for the susceptibility in the IL solutions with and without LiTFSA as a function of the mwater. The presence of Li+ in the IL scarcely influences the δcation values, indicating no strong interaction of C2mim+ with Li+. The δcation of all of the H atoms of C2mim+ shifts to the low field with the increase in the water content. The effect of water is remarkably observed for the H2 atom among all of the H atoms of C2mim+. According to the previous report,20 the shift of the H2 atom of C2mim+ caused by the hydrogen bonding with methanol is ∼0.3 ppm. The δcation shifts to ∼0.04 ppm for the Li+-doped IL and the Lifree solutions with increasing mwater up to 1.15 and 1.16 mol kg−1, which correspond to the mole fractions of water of 0.32 and 0.41, respectively. The present results are comparable with the variation in the 1H chemical shifts of C2mim+ below the mole fraction of methanol of ∼0.4 for the [C2mim][TFSA]− methanol systems due to the hydrogen bonding. The small
increase in the δcation values of C2mim+ with the increase in the water concentration declares that C2mim+ more strongly interacts with water than does TFSA−. Figure 4a indicates the 1H chemical shifts of water (δwater) corrected for the susceptibility in the IL solutions with and without LiTFSA as a function of the mwater. The δwater shifts to ∼0.4 ppm toward the low field with increasing mwater. The variation in the δwater with mwater is larger as compared to those of C2mim+ in the solutions as shown in Figure 3. This is attributable to the interaction of water molecule with Li+ and other water molecules by the hydrogen bonding. This means the hydration of Li+ and the self-assembly of water molecules. Thus, a water-rich domain may form in the ILs, and Li+ mainly exists in the domain. Interestingly, the 1H chemical shifts of water in the IL solutions with and without LiTFSA are significantly different from each other. The δwater for the IL solutions without lithium salt monotonically increases with increasing mwater as shown by the quadratic function (a dotted line) in Figure 4a. In contrast, the mwater dependence of δwater for the IL solutions with Li salt is strange; the δwater is almost unchanged in the mwater range from 0.1 to 0.3 mol kg−1. The effect of LiTFSA on the δwater arises from the Li+−water interaction. To make more clear the effect of Li+ on the δwater, the differences (δwater,doped − δwater,free) of the 1H chemical shifts of water between the IL solutions with and without LiTFSA are plotted against the mwater (Figure 4b). Here, the δwater,doped 16221
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
Figure 5. (a) Self-diffusion coefficients (D) of cation (blue triangles), anion (purple diamonds), Li+ (red circles), and water (black squares) in the IL solutions with (closed symbols) and without (open symbols) LiTFSA as a function of the molality of water (mwater). (b) Plots of the D of C2mim+ (blue triangles), TFSA− (purple diamonds), Li+ (red circles), and water (black squares) of the IL solutions with (closed symbols) and without (open symbols) LiTFSA against the inverse of the solvent viscosity (η−1).
1
Figure 4. (a) H chemical shifts of water in the IL solutions with (δwater,doped, closed symbols) and without (δwater,free, open symbols) LiTFSA as a function of the molality of water (mwater). The dotted line presents the quadric curve obtained by a least-squares fit on δwater for the IL solutions without LiTFSA. (b) Difference (δwater,doped − δwater,free) of 1H chemical shifts of water between the IL solutions with and without LiTFSA as a function of mwater. δwater,doped values were experimentally estimated, whereas δwater,free values were obtained by the quadratic regression. The upper axis shows the molar ratio of [H2O]/ [Li].
of LiTFSA. C2mim+ cation may not remarkably interact with Li+. On the other hand, TFSA− anions should interact with Li+ to form the solvation structure of Li+ in the solutions at the low water contents as discussed above. However, the effect of Li+ on the Danion is not clearly observed because of the higher concentration of free TFSA− than the solvating ones. At the low mwater, the DLi values are somewhat smaller than the Dcation and Danion values. The order (DLi < Danion < Dcation) of the selfdiffusion coefficients previously reported21−30 can be interpreted by the slower migration of Li+ than the IL cation and anion under the conditions. The increase in the water content leads to the more remarkable increase in DLi than Dcation and Danion. At the highest mwater of 1.15 mol kg−1, the DLi is comparable with Dcation and larger by a factor of 1.5 than Danion, resulting in the order of Danion < Dcation ≈ DLi. The translational motion of Li+ becomes more active on the addition of water. In Figure 5a, the Dwater values are larger by a factor of ∼10 than the self-diffusion coefficients of the three ions whether Li+ is present or not. In addition, the change in Dwater with mwater is quite different from those of the ions. Without Li+ ions, the Dwater of the IL solution quickly declines and reaches a plateau as the mwater increases. The decrease in the Dwater for the [C2mim][TFSA] solutions in the present study is different from the increase in the Dwater for [C4mim][TFSA].31 This presumably means that water molecules more easily assemble in [C2mim][TFSA] than in [C4mim][TFSA]. It is conceivable that [C2mim][TFSA] has larger voids, where water molecules may accommodate, due to the shorter alkyl chain than
values for the IL solutions with LiTFSA were experimentally determined, and the δwater,free values were obtained by the interpolation of the IL solutions without LiTFSA using a quadratic function. When the mwater increases, the steep increase in the δwater,doped − δwater,free values at mwater < ∼0.1 mol kg−1 is followed by the gradual decrease in the difference. Below mwater = ∼0.1 mol kg−1, which corresponds to the molar ratio [H2O]/ [Li] = ∼8, water molecules gradually replace TFSA− on Li+. The Li+−water interaction is saturated at mwater = ∼0.1 mol kg−1. This suggests that Li+ ion is hydrated with eight water molecules averaged. In the range of mwater > ∼0.1 mol kg−1, water molecules do not further interact with Li+ but do with other water molecules. The increase in the number of the water−water interaction leads to the weakening of the effect of LiTFSA on the δwater. This is the reason for the diminishing of the difference of δwater,doped − δwater,free above mwater = ∼0.1 mol kg−1. Self-Diffusion Coefficients. The numerical data of Dcation, Danion, DLi, and Dwater of the IL solutions with and without LiTFSA were listed in the Supporting Information (Table S2). The plots of Dcation, Danion, DLi, and Dwater in the IL solutions with and without LiTFSA against the mwater are presented in Figure 5a. Dwater at the lowest mwater (0.0156 mol kg−1) could not be measured because of the very weak signal of water. The Dcation and Danion are not significantly influenced by the addition 16222
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
by LiTFSA. Importantly, the rcation and ranion values are almost kept until the molar ratios (∼0.46 and ∼0.69) of water to IL for the IL solutions with and without LiTFSA, respectively. These results are the evidence that the IL cation and anion do not strongly interact with Li+. The rLi shrinks from 0.26 to 0.14 nm with the increase in the amount of water. The rLi values are comparable with that of aqueous solution (0.24 nm) and thus larger than the radius of Li+ found in crystals, which ranges from 0.059 to 0.092 nm depending on the coordination number.15 The size variation arises from the change in the solvation structure of Li+. The reason for the decrease in the rLi is that the TFSA−, which solvates to Li+, is replaced by water molecules as water is added into the IL solutions. As the result, the Li+ diffusion activated on the addition of water (Figure 5) arises not only from the fluidization of the entire systems but also from the reduction of the size of Li+ solvation structure. This finding shows that the transport of Li+ in the IL solution is affected by its solvation structure. In Figure 6, with the increase in the amount of water, the rwater expands from 0.011 to 0.023 nm in the IL solution with LiTFSA and from 0.0091 to 0.025 nm in the solutions without LiTFSA. The rwater values are much smaller than those of the aqueous LiTFSA solution and pure water (0.11 nm for both). The diffusion mechanism of water in the IL solutions may differ from those in the aqueous LiTFSA solution and pure water. The increase in the rwater is attributed to the interactions of water molecule with both Li+ and other water molecules. The deceleration of the diffusion of water molecules in the IL solution without LiTFSA shown in Figure 5 is caused by the formation of the self-assembly of water molecules. For the IL solution with LiTFSA, the sharp rise in the rwater with increasing mwater up to ∼0.1 mol kg−1 may be ascribed to the interaction with Li+, and the moderate change in the rwater above mwater = ∼0.1 mol kg−1 may arise from the hydrogen bonding among water molecules. The Li+−water interaction conducts to the significant difference of the rwater between the IL solutions with and without LiTFSA. The inflection point of the rwater at mwater = ∼0.1 mol kg−1 corresponds to the molar ratio [H2O]/[Li] of ∼8, showing the hydration of Li+ with approximately eight water molecules. This is consistent with the results of the δwater (Figure 4a). In the range of [H2O]/[Li] > ∼8, the effect of the hydrogen bonding among water molecules on the motion of water molecules is larger than that of the addition of LiTFSA. The difference of the rwater between the IL solutions with and without LiTFSA becomes small in this range. The change in the rwater with increasing mwater shows both hydration of Li+ and hydrogen-bonded assembly among water molecules as well as δwater. Thus, water-rich domain may be formed in the IL solutions. Referring to the mwater dependence of Dwater of the IL solution with LiTFSA (Figure 5a), the deceleration of the motion of water molecules in the IL solutions below mwater = ∼0.1 mol kg−1 is attributed to the strong Li+−water interaction. In contrast, the retardation of the motion of water molecules above mwater = ∼0.1 mol kg−1 mainly arises from the hydrogen bonding among water molecules. Because the present [C2mim][TFSA] has the highly ordered liquid structure,34 therefore, the smaller rwater in the IL solutions than the aqueous solution and pure water may result from the occurrence of diffusion channel of water in the IL solution due to the partially segregated structure of IL. Longitudinal Relaxation Times. The numerical data of the 7Li longitudinal relaxation times (T1,Li) of Li+ in the IL
[C4mim][TFSA]. In the case of the IL solutions with LiTFSA, the Dwater rapidly decreases with increasing water content up to mwater = ∼0.1 mol kg−1, and then barely increases in the middle range from ∼0.1 to ∼0.3 mol kg−1, and the Dwater moderately decreases in the higher mwater range above ∼0.3 mol kg−1. The mechanism of the mwater dependence of Dwater is discussed in the following paragraph in terms of the hydration of Li+ and the self-assembly with water molecules. The D value is correlated with solvent viscosity (η) and temperature (T) as shown by the Stoke−Einstein relation, which can be obtained by combining classical hydrodynamics of Stokes law and kinetic theory of Einstein relation:
D=
kBT cπηr
(2)
where r is the hydrodynamic radius of the diffusing species and c is a constant. When a molecule diffuses with solvent molecules due to the friction between the diffusing and surrounding bodies (stick boundary limit), c should be 6.32 When a molecule migrates without any friction (slip boundary limit), c is set to be 4.33 On the basis of the Stoke−Einstein principle, the D values for all of the chemical species are plotted against the inverse of η (η−1) in Figure 5b. The η values were obtained from the interpolation of η for the binary [C2mim][TFSA]−water system as a function of the mole fraction of water.13 From eq 2, generally, the reduction of η contributes to the increase in the D value. However, the Dwater values do not come under this case. The increase in the D for C2mim+, TFSA−, and Li+ with increasing η−1 suggests that the acceleration of the ions is ascribed to the fluidization of the IL solutions on adding of water. As the η−1 value increases, the Dcation and Danion linearity grow, whereas the DLi shows a nonlinear relationship with η−1 (Figure 5b). This fact suggests that other reasons should contribute to the increase in DLi with increasing water content. On the assumption that each diffusing species is in sphere shape, their r values in the IL solutions can be determined from eq 2 with c = 6. Figure 6 gives the r values of C2mim+ (rcation), TFSA− (ranion), Li+ (rLi), and water (rwater) as a function of the mwater. The rcation (0.12 nm) and the ranion (0.20 nm) do not significantly change on adding LiTFSA. Thus, the diffusion mechanism of C2mim+ and TFSA− is not remarkably influenced
Figure 6. Hydrodynamic radii (r) of C2mim+ (blue triangles), TFSA− (purple diamonds), Li+ (red circles), and water (black squares) in the IL solutions with (closed symbols) and without (open symbols) LiTFSA as a function of the molality of water (mwater). The upper axis shows the molar ratio of [H2O]/[Li]. 16223
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
τc,Li values decrease with increasing mwater; notably, the τc,Li values significantly drop in the low mwater range. The mwater dependence of τc,Li is exponential-like and seems to be analogous to that on δLi (Figure 2), despite the opposite direction. Relationship between Self-Diffusion Coefficients and Correlation Times of Li+. 7Li nucleus relaxes by diffusing motion but not by tumbling one. The τc,Li value obtained from the T1,Li can be related to the DLi through the mean one-jump distance (RLi) of Li+:
solutions determined by using two different magnetic field strengths of 7.0 and 11.7 T are listed in the Supporting Information (Table S3). Figure 7 presents the T1,Li of Li+ in the
RLi =
6DLi τc,Li
(4)
Figure 9 shows the mwater dependence of RLi. The RLi at the lowest mwater is 0.33 nm and agrees with the value (∼0.3 nm)
Figure 7. 7Li longitudinal relaxation times (T1,Li) of the IL solutions with LiTFSA obtained by using two different magnetic field strengths of 7.0 T (closed symbols) and 11.7 T (open symbols) as a function of the molality of water (mwater).
IL solutions at 7.0 and 11.7 T as a function of the mwater. The T1,Li values of the IL solutions are ranged between ∼0.2 and ∼1.2 s with a minimum at mwater = ∼0.1 mol kg−1. All of the T1,Li at 11.7 T are larger than those at 7.0 T over the entire water content range studied. The remarkable differences between the T1,Li values at the two magnetic fields mean that Li+ ions in the IL solutions are not under the extreme narrowing condition of 2πvτ ≪ 1. Because 7Li with the spin quantum number of I = 3/2 relaxes through only the quadrupolar interaction, the observed T1,Li under no extreme narrowing condition is related to the correlation time (τc,Li) as follows:35 ⎛ ⎞ τc,Li 4τc,Li 1 ⎟⎟ = A⎜⎜ + 2 2 2 T1,Li 1 + 16π 2v 2τc,Li 2 ⎠ ⎝ 1 + 4π v τc,Li
Figure 9. Mean one-jump distances (RLi) of Li+ in the IL solutions as a function of the molality of water (mwater). The upper axis shows the molar ratio of [H2O]/[Li].
previously reported.30 When a trace of water is dissolved into the IL solution, the jump distance of Li+ suddenly shortens to 0.25 nm. The shortening of the RLi suggests that the distance between Li+ and the solvating species reduces due to the change in the solvation structure of Li+. The TFSA− anions that solvate to Li+ are replaced by water molecules. Above the molar ratio [H2O]/[Li] of ∼6, the change in the RLi values is very small. These jump distances are equal numerically to the hydrodynamic radius rLi (Figure 6). This shows that the jumping motion of Li+ occurs inside its solvation shell. The inflection point of the RLi at the molar ratio [H2O]/[Li] of ∼6 intends that six water molecules contribute to the jump motion of Li+ at the maximum. This finding is reasonable with the hydration structure of Li+ with approximately eight water molecules.
(3)
where A is a constant depending on sample system and v is a resonance frequency. τc,Li was estimated from a pair of the T1,Li measured and the v values on the assumption that the relaxation of 7Li occurs through the same mechanisms under two different magnetic fields. The τc,Li values are plotted against the mwater in Figure 8. All of the τc,Li values are in subnanosecond scale over the water content range studied. The present data for the dry IL solutions well agree with τc,Li in [C2mim][TFSA] solutions at 0.32 mol kg−1 of LiTFSA.30 The
■
CONCLUSIONS We carried out NMR measurements on the hydrophobic IL solutions with and without LiTFSA at 293.2 K in the wide water content range from 0.0156 to 1.16 mol kg−1. The δLi shifted to the low field as water was dissolved; peculiarly, the change in the δLi was remarkable in the low water content range. Li+ ions more strongly interact with water molecules than TFSA− in the IL solutions. The δwater remarkably increased with the water concentration and showed the difference between the values in the presence and absence of LiTFSA. These results were interpreted by that water molecules interact with Li+ as well as the hydrogen bonding among water molecules. As the water content increases, the Dcation, Danion, and
Figure 8. Correlation times (τc,Li) of Li+ in the IL solutions as a function of the molality of water (mwater). 16224
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
(5) Shirai, A.; Fujii, K.; Seki, S.; Umebayashi, Y.; Ishiguro, S.; Ikeda, Y. Solvation of Lithium Ion in N,N-Diethyl-N-methyl-N-(2methoxyethyl)ammonium Bis(trifluoromethanesulfonyl)amide Using Raman and Multinuclear NMR Spectroscopy. Anal. Sci. 2008, 24, 1291−1296. (6) Lassegues, J.-C.; Grondin, J.; Aupetit, C.; Johansson, P. Spectroscopic Identification of the Lithium Ion Transporting Species in LiTFSI-Doped Ionic Liquids. J. Phys. Chem. A 2009, 113, 305−314. (7) Umebayashi, Y.; Mori, S.; Fujii, K.; Tsuzuki, S.; Seki, S.; Hayamizu, K.; Ishiguro, S. Raman Spectroscopic Studies and Ab Initio Calculations on Conformational Isomerism of 1-Buthyl-3-methylimidazolium Bis-(trifluoromethanesulfonyl)amide Solvated to a Lithium Ion in Ionic Liquids: Effects of the Second Solvation Sphere of the Lithium Ion. J. Phys. Chem. B 2010, 114, 6513−6521. (8) Umebayashi, Y.; Hamano, H.; Seki, S.; Minofar, B.; Fujii, K.; Hayamizu, K.; Tsuzuki, S.; Kameda, Y.; Kohara, S.; Watanabe, M. Liquid Structure of and Li+ Ion Solvation in Bis(trifluoromethanesulfonyl)amide Based Ionic Liquids Composed of 1-Ethyl-3-methylimidazolium and N-Methyl-N-propylpyrrolidinium Cations. J. Phys. Chem. B 2011, 115, 12179−12191. (9) Li, Z.; Smith, G. D.; Bedrov, D. Li+ Solvation and Transport Properties in Ionic Liquid/Lithium Salt Mixtures: A Molecular Dynamics Simulation Study. J. Phys. Chem. B 2012, 116, 12801− 12809. (10) Schröder, U.; Wadhawan, J. D.; Compton, R. G.; Marken, F.; Suarez, P. A. Z.; Consorti, C. S.; de Souza, R. F.; Dupont, J. WaterInduced Accelerated Ion Diffusion: Voltammetric Studies in 1-Methyl3-[2,6-(S)-dimethylocten-2-yl]imidazolium Tetrafluoroborate, 1-Butyl3-methylimidazolium Tetrafluoroborate and Hexafluorophosphate Ionic Liquids. New J. Chem. 2000, 24, 1009−1015. (11) Huddleston, J. G.; Visser, A. E.; Reichert, W. M.; Willauer, H. D.; Broker, G. A.; Rogers, R. D. Characterization and Comparison of Hydrophilic and Hydrophobic Room Temperature Ionic Liquids Incorporating the Imidazolium Cation. Green Chem. 2001, 3, 156− 164. (12) Bakers, N.; Baker, G. A.; Bright, F. V. Temperature-Dependent Microscopic Solvent Properties of ‘dry’ and ‘wet’ 1-Butyl-3methylimidazolium Hexafluorophosphate: Correlation with ET(30) and Kamlet−Taft Polarity Scales. Green Chem. 2002, 4, 165−169. (13) Widegren, J. A.; Laesecke, A.; Magee, J. W. The Effect of Dissolved Water on the Viscosities of Hydrophobic Room-Temperature Ionic Liquids. Chem. Commun. 2005, 41, 1610−1612. (14) Jacquemin, J.; Husson, P.; Padua, A. A. H.; Majer, V. Density and Viscosity of Several Pure and Water-Saturated Ionic Liquids. Green Chem. 2006, 8, 172−180. (15) CRC Handbook of Chemistry and Physics, 86th ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 2005. (16) Martins, V. L.; Nicolau, B. G.; Urahata, S. M.; Ribeiro, C. C.; Torresi, R. M. Influence of the Water Content on the Structure and Physicochemical Properties of an Ionic Liquid and Its Li+ Mixture. J. Phys. Chem. B 2013, 117, 8782−8792. (17) Bonhôte, P.; Dias, A.; Papageorgiou, N.; Kalyanasundaram, K.; Grätzel, M. Hydrophobic, Highly Conductive Ambient-Temperature Molten Salts. Inorg. Chem. 1996, 35, 1168−1178. (18) Holz, M.; Heil, S. R.; Sacco, A. Temperature-Dependent SelfDiffusion Coefficients of Water and Six Selected Molecular Liquids for Calibration in Accurate 1H NMR PFG Measurements. Phys. Chem. Chem. Phys. 2000, 2, 4740−4742. (19) Becker, E. D. High-Resolution NMR: Theory and Chemical Applications, 3rd ed.; Academic Press: New York, 2000; pp 83−117. (20) Takamuku, T.; Honda, Y.; Fujii, K.; Kittaka, S. Aggregation of Imidazolium Ionic Liquids in Molecular Liquids Studied by SmallAngle Neutron Scattering and NMR. Anal. Sci. 2008, 24, 1285−1290. (21) Hayamizu, K.; Aihara, Y.; Nakagawa, H.; Nukuda, T.; Price, W. S. Ionic Conduction and Ion Diffusion in Binary Room-Temperature Ionic Liquids Composed of [emim][BF4] and LiBF4. J. Phys. Chem. B 2004, 108, 19527−19532. (22) Nicotera, I.; Oliviero, C.; Henderson, W. A.; Appetecchi, G. B.; Passerini, S. NMR Investigation of Ionic Liquid-LiX Mixtures:
DLi increased in one direction, but the Dwater strangely behaved. The r values of all of the species in the IL solutions were evaluated from their D values. The accelerations of the three ion species are due mostly to the fluidization of the entire systems. In addition, the acceleration of the diffusion of Li+ arises from the replacement of TFSA− by water molecules. The water content dependence of the rwater also supported the hydration of Li+ as well as self-assembly of water molecules by hydrogen bonds. The τc,Li values were determined from the T1,Li values, and the jump distance RLi in the IL solution can be estimated from the DLi and τc,Li values. The RLi decreased from 0.33 to 0.25 nm with increasing the molar ratio [H2O]/[Li] up to ∼6. This indicated the reducing of the distance between Li+ and the solvating species, suggesting the change in the solvation structure of Li+. In conclusion, most of the present results revealed that the solvating species of TFSA− for Li+ in the IL solutions were replaced by water molecules in the solution at around mwater = 0.1 mol kg−1. The concentration corresponds to the molar ratio of [H2O]/[Li] = ∼8. Consequently, a Li+ ion is averagely hydrated by about eight water molecules on the present NMR scales. It becomes clear that Li+ diffuses in the water-rich domain (water channel) and at the same time jumps inside the hydration shell in subnanosecond scale.
■
ASSOCIATED CONTENT
S Supporting Information *
Details of the magnetic susceptibility correction, the numerical data of δcation, δwater, and δLi of the IL solutions with and without LiTFSA (Table S1), the Dcation, Danion, DLi, and Dwater of both IL solutions (Table S2), and the T1,Li of the IL solutions with LiTFSA (Table S3). This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported partly by a Grant-in-Aid for Young Scientists (B) (No. 24750019) from the Japan Society for the Promotion of Science and Dean’s Grant for Progressive Research Projects from Saga University.
■
REFERENCES
(1) Ionic Liquids as Green Solvents; Rogers, R. D., Seddon, K. R., Eds.; ACS Symposium Series 856; American Chemical Society: Washington, DC, 2003. (2) Lassegues, J.-C.; Grondin, J.; Talaga, D. Lithium Solvation in Bis(trifluoromethanesulfonyl)imide-Based Ionic Liquids. Phys. Chem. Chem. Phys. 2006, 8, 5629−5632. (3) Umebayashi, Y.; Mitsugi, T.; Fukuda, S.; Fujimoto, T.; Fujii, K.; Kanzaki, R.; Takeuchi, M.; Ishiguro, S. Lithium Ion Solvation in Room-Temperature Ionic Liquids Involving Bis(trifluoromethanesulfonyl) Imide Anion Studied by Raman Spectroscopy and DFT Calculations. J. Phys. Chem. B 2007, 111, 13028− 13032. (4) Duluard, S.; Grondin, J.; Bruneel, J.-L.; Pianet, I.; Grelard, A.; Campet, G.; Delville, M.-H.; Lassegues, J.-C. Lithium Solvation and Diffusion in the 1-Buthyl-3-methylimidazolium Bis(trifluoromethanesulfonyl)imide Ionic Liquid. J. Raman Spectrosc. 2008, 39, 627−632. 16225
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
The Journal of Physical Chemistry B
Article
Pyrrolidinium Cations and TFSI− Anions. J. Phys. Chem. B 2005, 109, 22814−22819. (23) Saito, Y.; Umecky, T.; Niwa, J.; Sakai, T.; Maeda, S. Existing Condition and Migration Property of Ions in Lithium Electrolytes with Ionic Liquid Solvent. J. Phys. Chem. B 2007, 111, 11794−11802. (24) Hayamizu, K.; Tsuzuki, S.; Seki, S.; Ohno, Y.; Miyashiro, H.; Kobayashi, Y. Quaternary Ammonium Room-temperature Ionic Liquid Including an Oxygen Atom in Side Chain/Lithium Salt Binary Electrolytes: Ionic Conductivity and 1H, 7Li, and 19F NMR Studies on Diffusion Coefficients and Local Motions. J. Phys. Chem. B 2008, 112, 1189−1197. (25) Umecky, T.; Saito, Y.; Okumura, Y.; Maeda, S.; Sakai, T. Ionization Condition of Lithium Ionic Liquid Electrolytes under the Solvation Effect of Liquid and Solid Solvents. J. Phys. Chem. B 2008, 112, 3357−3364. (26) Frömling, T.; Kunze, M.; Schönhoff, M.; Sundermeyer, J.; Roling, B. Enhanced Lithium Transference Numbers in Ionic Liquid Electrolytes. J. Phys. Chem. B 2008, 112, 12985−12990. (27) Le, M. L. P.; Alloin, F.; Strobel, P.; Leprêtre, J.-C.; del Valle, C. P.; Judeinstein, P. Structure-Properties Relationships of Lithium Electrolytes Based on Ionic Liquid. J. Phys. Chem. B 2010, 114, 894−903. (28) Bayley, P. M.; Lane, G. H.; Lyons, L. J.; MacFarlane, D. R.; Forsyth, M. Undoing Lithium Ion Association in Ionic Liquids through the Complexation by Oligoethers. J. Phys. Chem. C 2010, 114, 20569− 20576. (29) Castiglione, F.; Raggh, E.; Mele, A.; Appetecchi, G. B.; Montanino, M.; Passerini, S. Molecular Environment and Enhanced Diffusivity of Li+ Ions in Lithium-Salt-Doped Ionic Liquid Electrolytes. J. Phys. Chem. Lett. 2011, 2, 153−157. (30) Hayamizu, K.; Tsuzuki, S.; Seki, S.; Umebayashi, Y. Nuclear Magnetic Resonance Studies on the Rotational and Translational Motions of Ionic Liquids Composed of 1-Ethyl-3-methylimidazolium Cation and Bis(trifluoromethanesulfonyl)amide and Bis(fluorosulfonyl)amide Anions Their Binary Systems Including Lithium Salts. J. Chem. Phys. 2011, 135, 084505. (31) Rollet, A.-L.; Porion, P.; Vaultier, M.; Billard, I.; Deschamps, M.; Bessada, C.; Jouvensal, L. Anomalous Diffusion of Water in [BMIM][TFSI] Room-Temperature Ionic Liquid. J. Phys. Chem. B 2007, 111, 11888−11891. (32) Einstein, A. Investigations on the Theory of the Brownian Movement; Dover: New York, 1956; pp 68−85. (33) McLaughlin, E. Viscosity and Self-Diffusion in Liquids. Trans. Faraday Soc. 1959, 55, 28−38. (34) Lopes, J. N. A. C.; Pádua, A. A. H. Nanostructural Organization in Ionic Liquids. J. Phys. Chem. B 2006, 110, 3330−3335. (35) Abragam, A. The Principles of Nuclear Magnetism; Oxford University Press: New York, 1989; pp 313−315.
16226
dx.doi.org/10.1021/jp409324k | J. Phys. Chem. B 2013, 117, 16219−16226
