1776
J. Phys. Chem. A 2010, 114, 1776–1782
Melting Behavior and Ionic Conductivity in Hydrophobic Ionic Liquids Miriam Kunze,† Maria Montanino,‡ Giovanni B. Appetecchi,‡ Sangsik Jeong,† Monika Scho¨nhoff,† Martin Winter,† and Stefano Passerini*,† Department of Physical Chemistry, Westfälische Wilhelms-UniVersita¨t Mu¨nster, Corrensstraβe 28/30, 48149 Mu¨nster, Germany, and Agency for the New Technologies, Energy and the EnVironment (ENEA), Via Anguillarese 301, 00123 Rome, Italy ReceiVed: October 16, 2009; ReVised Manuscript ReceiVed: December 9, 2009
Four room-temperature ionic liquids (RTILs) based on the N-butyl-N-methyl pyrrolidinium (Pyr14+) and N-methyl-N-propyl pyrrolidinium cations (Pyr13+) and bis(trifluoromethanesulfonyl)imide (TFSI-) and bis(fluorosulfonyl)imide (FSI-) anions were intensively investigated during their melting. The diffusion coefficients of 1H and 19F were determined using pulsed field gradient (PFG) NMR to study the dynamics of the cations, anions, and ion pairs. The AC conductivities were measured to detect only the motion of the charged particles. The melting points of these ionic liquids were measured by DSC and verified by the temperature-dependent full width at half-maximum (FWHM) of the 1H and 19F NMR peaks. The diffusion and conductivity data at low temperatures gave information about the dynamics at the melting point and allowed specifying the way of melting. In addition, the diffusion coefficients of 1H (DH) and 19F (DF) and conductivity were correlated using the Nernst-Einstein equation with respect to the existence of ion pairs. Our results show that in dependence on the cation different melting behaviors were identified. In the Pyr14based ILs, ion pairs exist, which collapse above the melting point of the sample. This is in contrast to the Pyr13-based ILs where the present ion pairs in the crystal dissociate during the melting. Furthermore, the anions do not influence the melting behavior of the investigated Pyr14 systems but affect the Pyr13 ILs. This becomes apparent in species with a higher mobility during the breakup of the crystalline IL. Introduction Ionic liquids (ILs) are molten salts with a melting point below 100 °C. Not only this fact, but also that these liquids have a low vapor pressure and a high chemical and electrochemical stability make them labeled as green solvents.1-5 Therefore, ILs are being investigated for a wide range of applications. One field is the utilization as electrolyte component in electrochemical devices. This includes lithium batteries,6-14 fuel cells,15 electrochemical (super or ultra) capacitors,16 electrochemical actuators,17 light-emitting electrochemical cells,18 etc. For these electrochemical applications, one has to have exact requirements including a high chemical, thermal, and electrochemical stability, a wide liquid range for operation at low and high temperatures, and a high ionic conductivity. Especially for lithium batteries the ionic conductivity plays an important role. In case a lithium salt is added to the IL, it is generally found that the Li+ conductivity scales with the conductivity of the IL. Thus, batteries can be discharged at faster rates with high conductivity ILs. Furthermore, the use (charge and discharge) of the lithium batteries shall be possible at very low temperatures. Because of that fact, it is important to gain knowledge of the molecular motion at low temperatures and throughout the melting transition of the material. Until now, in the literature it is often reported that above the melting temperature TM ion pair formation is taking place in the IL.19-22 Here, we report the physical and electrochemical properties of four different ILs based on the N-butyl-N-methyl pyrrolidinium cation (Pyr14+), N-methyl-N-propyl pyrrolidinium cation * Corresponding author. E-mail:
[email protected]. † Westfaelische Wilhelms-Universita¨t Mu¨nster. ‡ ENEA.
SCHEME 1: Cations and Anions Used for the Ionic Liquids
(Pyr13+), bis(trifluoromethanesulfonyl)imide (TFSI-), and bis(fluorosulfonyl)imide (FSI-) anion (Scheme 1) over a wide temperature range including the dynamics below TM. Mixtures of these ionic liquids with lithium TFSI have been proved successfully for lithium insertion in graphite.23-25 Hence, the melting temperatures TM were determined by DSC and NMR, and for the dynamic processes the measured diffusion coef-
10.1021/jp9099418 2010 American Chemical Society Published on Web 01/08/2010
Melting Behavior and Ionic Conductivity in Hydrophobic ILs
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1777
ficients of 1H and 19F in the solid and liquid state with the corresponding AC conductivities were correlated and discussed.
chamber (Binder GmbH MK53) located in the dry room. The entire setup was controlled by software developed at ENEA. NMR Measurements. A Bruker Avance 400 MHz spectrometer with an Ultra Shield 89 mm cryomagnet with a magnetic flux density of 9.4 T was employed for all NMR measurements. A gradient probe head (Bruker, Diff 30) provided a maximum gradient strength of 12 T/m for diffusion measurements. The gradient coils were cooled by a water circulation unit. For both 1H and 19F experiments, a tunable RF insert (376-400 MHz) was used, which, by appropriate isolation, was especially configured for low and high temperature measurements. Self-diffusion coefficients of 1H and 19F were measured using pulsed field gradient NMR (PFG-NMR).27,28 The stimulated echo pulse sequence was used, which contains three 90° pulses: {π/2-τ-π/2-T-π/2-τ-ECHO}. This sequence was combined with two gradient pulses of strength g and duration δ after the first and the third 90° pulse within the delay time τ, respectively. τ was kept constant for all measurements, τ ) 6.2 ms. The waiting time between the two gradient pulses is the diffusion time ∆. For the 1H and 19F diffusion measurements, ∆ ranged from 100 ms at low temperatures to 70 ms at high temperatures, respectively. The gradient strength g was varied at 1180 G cm-1 for all experiments, and δ was set to 4 ms below the melting point and 2.5 ms above the melting point of the ionic liquid. The analysis of the echo decays was based on the area of the resonances of 1H and 19F for the cation and anion, respectively. The decay of the signal as a function of g resulted in the diffusion coefficient D by fitting the exponential decay function:
Experimental Section Sample Preparation. The four ILs (Pyr13FSI, Pyr13TFSI, Pyr14FSI, and Pyr14TFSI) (cf., Scheme 1) were synthesized through a procedure developed at ENEA and described in detail elsewhere.26 The chemicals N-methylpyrrolidine (97 wt %), 1-propylbutane (99%), 1-bromobutane (99 wt %), and ethyl acetate (ACS grade, >99.5 wt %) were purchased from Aldrich and previously purified (with the exception of ethyl acetate) using activated carbon (Aldrich, Darco-G60) and alumina (acidic, Aldrich Brockmann I). Lithium bis(trifluoromethanesulfonyl)imide, LiTFSI (99.9 wt %, battery grade), and lithium (or potassium) bis(fluorosulfonyl)imide, Li(K)FSI (99.9 wt %, battery grade), were purchased from 3 M and Dai-Ichi Kogyo Seiyaku Co. Ltd., respectively, and used as received. Deionized H2O was obtained using a Millipore ion-exchange resin deionizer. The N-alkyl-N-methylpyrrolidinium bromide precursors (Pyr13Br and Pyr14Br) were synthesized by reacting N-methylpyrrolidine with the appropriate amount of bromoalkyl in the presence of ethyl acetate. The precursors were repeatedly rinsed with ethyl acetate to remove the reagents excess and the soluble impurities. The four ionic liquids were obtained by reacting aqueous solutions of the precursors (Pyr13Br and Pyr14Br) with the appropriate amounts of LiTFSI or Li(K)FSI. The lithium to potassium content in the ionic liquids was tested to be below 2 ppm by atomic absorption spectroscopy (AAS). The reactions led to the formation of the hydrophobic ionic liquids and hydrophilic LiBr or KBr. After removal of the aqueous phase, the ionic liquids were rinsed several times with deionized water to remove water-soluble LiBr or KBr and excess of LiTFSI or Li(K)FSI. Next, the ionic liquids were purified with activated carbon and acidic alumina. The liquid fractions were separated from the solid phases by vacuum filtering and then placed in a rotary evaporator at 80 °C under vacuum to remove the solvent (ethyl acetate). Finally, the ionic liquids were dried using an oil-free vacuum pump at 60 °C for at least 2 h and then at 120 °C for at least 18 h with yields ranging from 85 to 90 mol %. The materials were stored in sealed glass tubes in a controlled environment (dry-room, R.H. DF in the TFSI--based ILs could be the existence of nano domains.31 The Pyr14-ion could create via the alkyl chain some kind of domains or agglomerates and can be surrounded by the anion. Between the Pyr14-ion agglomerates and free Pyr14-ions, a fast exchange is maybe present, which could result a fast diffusing cation. Besides this, the surrounding anions could build some pathway. Along this path of negative charges, the cation can easily diffuse. The existence and the development of ions and ion pairs can be proven by the comparison of the diffusion coefficients D with the AC conductivity σAC. This can be done in two ways: one way is plotting the diffusion coefficients and the AC conductivity as a function of temperature to compare their formation at the melting points (cf., Figure 6); the other is using the Nernst-Einstein equation (cf., Figure 7).32,33
σDC )
e2NV(DH + DF) kBT
(2)
where e is the elementary charge, kB is the Boltzmann constant, and NV is the number density of the cations or the anions, which can be calculated with the density F given in Table 1. The diffusion coefficients DH and DF contain the motion of all cations and anions, respectively, independent of whether they carry a charge or are part of a neutral ion pair. The comparison via the temperature dependence of D and σAC is displayed for all ionic liquids in Figure 6a-d. The increase of the AC conductivity is roughly correlated to the separation of DH and DF. At the temperature where DH and DF start to differ, σAC approximately begins to increase by about 4 orders of magnitude. This is valid for all investigated ionic liquids. A closer look at the Pyr14-based ionic liquids (Figure 6a and c) shows that there are almost solely ion pairs and neutral
Figure 6. AC conductivity σAC and diffusion coefficients D of the cation (blue O) and the anion (blue 9) as a function of temperature. The dotted lines represent the melting temperatures. (a) Pyr14FSI; (b) Pyr13FSI; (c) Pyr14TFSI; and (d) Pyr13TFSI.
clusters in the small liquid fraction of the ionic liquid (cf., Figure 5) existing before and within the melting takes place.
Melting Behavior and Ionic Conductivity in Hydrophobic ILs
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1781 TABLE 1: Densities of the Investigated Ionic Liquids T/K
F(Pyr14FSI) g/mL
F(Pyr13FSI) g/mL
F(Pyr14TFSI) g/mL
F(Pyr13TFSI) g/mL
293.15 313.15
1.310 1.302
1.343 1.327
1.399 1.381
1.433 1.423
TABLE 2: σAC/σNMR at Temperatures 5 K above TM TM + 5/K σAC/σNMR
Figure 7. AC conductivity σAC (O) and conductivity σNMR (0) calculated from the self-diffusion coefficients with the Nernst-Einstein equation. The green “9” represent the ratio of σAC and σNMR. The dotted lines represent the melting temperatures. (a) Pyr14FSI; (b) Pyr13FSI; (c) Pyr14TFSI; and (d) Pyr13TFSI.
These ion pairs and the solid ionic liquid as well break directly after the melting into ions and cause a conductivity
Pyr14FSI
Pyr13FSI
Pyr14TFSI
Pyr13TFSI
259 0.66
269 0.63
272 0.68
289 0.66
in the range of (10-2-10-1) S m-1. This characteristic is given for Pyr13FSI as well. The main difference from Pyr13FSI to the Pyr14-based ionic liquids is the minor difference in the diffusion coefficients directly above melting. Anyhow, one reaches high conductivities in Pyr13FSI, which can be explained with large diffusion coefficients of the ions (Figure 6b). As compared to these three ionic liquids, Pyr13TFSI shows a different evolution of ion formation in the region of the melting point (Figure 6d). It is still valid that the conductivity increases with the separation of the diffusion coefficients, but this happens before the melting point TM is reached. A good conductivity in the range of 10-1 S m-1 is registered about 3 K below TM. This can be correlated to the large liquid fraction, which is present below TM. In conclusion, the solid phase of Pyr13TFSI breaks into ion pairs at temperatures much below the melting point. This mobile fraction does not cause a good conductivity. At a certain liquid fraction, which is roughly about 16%, the ion pairs dissociate into ions and result in good conductivities, although the melting point is not reached. For Pyr13TFSI, one has a smaller fraction of ion pairs at its melting point than for the other three investigated ionic liquids. The existence of ion pairs as an intermediate state in the melting process is proven by the calculation of a conductivity σNMR using the Nernst-Einstein equation (eq 2) and calculating the ratio σAC/σNMR. σNMR represents only the conductivity in the liquid fraction of the ionic liquids. Both σNMR and σAC as well as the ratio σAC/σNMR are displayed in Figure 7. Although there is only a small fraction of the ionic liquid mobile below TM, which yields a diffusion coefficient, this comparison illustrates very well the existence of the ion pairs. For Pyr14FSI, Pyr13FSI, and PYR14TFSI, one can see that the ratio σAC/σNMR is increasing after the melting, while the ratio for Pyr13TFSI increases before the melting point is achieved. Hence, there is the intermediate state “ion pair” during the melting present in Pyr14FSI, Pyr13FSI, and PYR14TFSI, while this state is run through at lower temperatures for Pyr13TFSI. In all cases, σAC/σNMR reaches a constant value below 1 after TM. In case that σAC/σNMR equals 1, there would be only single ions existing in the ionic liquid. Consequently, in every ionic liquid there are ion pairs existing in the liquid state, but their amount is lower than the amount of ions. This is not valid for the solid phase, where σAC/σNMR is almost zero. Remarkable is a ratio in the range of 0.65 for all four ionic liquids after the melting (Table 2). This value, which has been already reported for Pyr14TFSI,21,22 describes a quite good dissociation for the ionic liquids as compared to other ionic liquids especially made of cations with longer side chains.34 In our case, the variation of the cation by one CH2 group or the anion does not seem to have a strong influence on the dissociation of the ionic liquid. This ratio stays nearly constant for Pyr13FSI and Pyr14TFSI over a large temperature range.
1782
J. Phys. Chem. A, Vol. 114, No. 4, 2010
Conclusions NMR, DSC, and conductivity measurements have been used to investigate the low temperature behavior of various ionic liquids. The changes of the peak width in the NMR-spectra confirmed TM from the DSC measurements, and an additional phase transition was registered for Pyr13FSI with these methods as well. Furthermore, it has been shown that in dependence on the cation in the IL there are different melting behaviors related to the conductivity and the diffusion coefficients of the cation and the anion. The existence of ions pairs below and above TM was presented and proven by the ratio of σAC/σNMR. These ion pairs start to partially dissociate after the melting but do not fully disappear. However, this is different for Pyr13TFSI. Here, a liquid fraction of about 5% below TM is present, and the dissociation of the ion pairs is taking place even before the melting point is reached. Acknowledgment. We are thankful for the financial support of the European Commission within the FP6 STREP Project ILLIBAT (contact no. NMP3_CT_2006_033181). References and Notes (1) Dupont, J.; de Souza, R. F.; Suarez, P. A. Z. Chem. ReV. 2002, 102, 3667. (2) Earle, M. J.; Seddon, K. R.; McCormac, P. B. Green Chem. 2000, 2, 261. (3) Wilkes, J. S. Green Chem. 2002, 4, 73. (4) Tarascon, J. M.; Armand, M. Nature 2001, 414, 359. (5) Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc. 2002, 124, 14247. (6) Ishikawa, M.; Sugimoto, T.; Kikuta, M.; Ishiko, E.; Kono, M. J. Power Sources 2006, 162, 658. (7) Matsumoto, H.; Sakaebe, H.; Tatsumi, K.; Kikuta, M.; Ishiko, E.; Kono, M. J. Power Sources 2006, 160, 1308. (8) Nakagawa, H.; Izuchi, S.; Kuwana, K.; Nukuda, T.; Aihara, Y. J. Electrochem. Soc. 2003, 150, A695. (9) Saint, J.; Best, A. S.; Hollenkamp, A. F.; Kerr, J.; Shin, J. H.; Doeff, M. M. J. Electrochem. Soc. 2008, 155, A172. (10) Sakaebe, H.; Matsumoto, H.; Tatsumi, K. Electrochim. Acta 2007, 53, 1048.
Kunze et al. (11) Shin, J. H.; Henderson, W. A.; Passerini, S. Electrochem. SolidState Lett. 2005, 8, A125. (12) Shin, J. H.; Henderson, W. A.; Scaccia, S.; Prosini, P. P.; Passerini, S. J. Power Sources 2006, 156, 560. (13) Shin, J. H.; Henderson, W. A.; Tizzani, C.; Passerini, S.; Jeong, S. S.; Kim, K. W. J. Electrochem. Soc. 2006, 153, A1649. (14) Sakaebe, H.; Matsumoto, H.; Tatsumi, K. J. Power Sources 2005, 146, 693. (15) Judeinstein, P.; Iojoiu, C.; Sanchez, J. Y.; Ancian, B. J. Phys. Chem. B 2008, 112, 3680. (16) Balducci, A.; Henderson, W. A.; Mastragostino, M.; Passerini, S.; Simon, P.; Soavi, F. Electrochim. Acta 2005, 50, 2233. (17) Ding, J.; Zhou, D. Z.; Spinks, G.; Wallace, G.; Forsyth, S.; Forsyth, M.; MacFarlane, D. Chem. Mater. 2003, 15, 2392. (18) Yang, C. H.; Sun, Q. J.; Qiao, J.; Li, Y. F. J. Phys. Chem. B 2003, 107, 12981. (19) Tokuda, H.; Hayamizu, K.; Ishii, K.; Abu Bin Hasan Susan, M.; Watanabe, M. J. Phys. Chem. B 2004, 108, 16593. (20) Tokuda, H.; Ishii, K.; Susan, M. A. B. H.; Tsuzuki, S.; Hayamizu, K.; Watanabe, M. J. Phys. Chem. B 2006, 110, 2833. (21) Tokuda, H.; Tsuzuki, S.; Susan, M. A. B. H.; Hayamizu, K.; Watanabe, M. J. Phys. Chem. B 2006, 110, 19593. (22) Castiglione, F.; Moreno, M.; Raos, G.; Famulari, A.; Mele, A.; Appetecchi, G. B.; Passerini, S. J. Phys. Chem. B 2009, 113, 10750. (23) Lux, S. F.; Schmuck, M.; Appetecchi, G. B.; Passerini, S.; Winter, M.; Balducci, A. J. Power Sources 2009, 192, 606. (24) Appetecchi, G. B.; Montanino, M.; Balducci, A.; Lux, S. F.; Winter, M.; Passerini, S. J. Power Sources 2009, 192, 599. (25) Schmuck, M.; Balducci, A.; Rupp, B.; Kern, W.; Passerini, S.; Winter, M. J. Solid State Electrochem. 2009, DOI:10.1007/s10008/008/ 0763/4. (26) Appetecchi, G. B.; Scaccia, S.; Tizzani, C.; Alessandrini, F.; Passerini, S. J. Electrochem. Soc. 2006, 153, A1685. (27) Price, W. S. Concepts Magn. Reson. 1997, 9, 299. (28) Stilbs, P. Prog. NMR Spectrosc. 1987, 19, 1. (29) Zhou, Q.; Henderson, W. A.; Appetecchi, G. B.; Montanino, M.; Passerini, S. J. Phys. Chem. B 2008, 112, 13577. (30) Henderson, W. A.; Passerini, S. Chem. Mater. 2004, 16, 2881. (31) Triolo, A.; Russina, O.; Fazio, B.; Appetecchi, G. B.; Carewska, M.; Passerini, S. J. Chem. Phys. 2009, 130, 164521. (32) Nicotera, I.; Oliviero, C.; Henderson, W. A.; Appetecchi, G. B.; Passerini, S. J. Phys. Chem. B 2005, 109, 22814. (33) Aihara, Y.; Arai, S.; Hayamizu, K. Electrochim. Acta 2000, 45, 1321. (34) Tokuda, H.; Hayamizu, K.; Ishii, K.; Susan, M. A. B. H.; Watanabe, M. J. Phys. Chem. B 2005, 109, 6103.
JP9099418
