Nanocomposite Ion Gels Based on Silica Nanoparticles and an Ionic

Jul 9, 2008 - ... 79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan ... Seiji Tsuzuki , Wataru Shinoda , Kazuhide Ueno , Toshihiko Mandai , Hisash...
0 downloads 0 Views 254KB Size
Download PDF
J. Phys. Chem. B 2008, 112, 9013–9019

9013

Nanocomposite Ion Gels Based on Silica Nanoparticles and an Ionic Liquid: Ionic Transport, Viscoelastic Properties, and Microstructure Kazuhide Ueno,† Kenji Hata,† Toru Katakabe,† Masashi Kondoh,‡ and Masayoshi Watanabe*,† Department of Chemistry and Biotechnology and Instrumental Analysis Center, Yokohama National UniVersity, 79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan ReceiVed: April 3, 2008; ReVised Manuscript ReceiVed: May 17, 2008

The dispersion of silica nanoparticles made an ionic liquid, 1-ethyl-3-methyl imidazolium bis(trifluoromethanesulfonyl)amide ([C2mim][NTf2]), gelled even by the addition of 2-3 wt %, due to the formation of interconnected particulate silica networks in [C2mim][NTf2]. The ionic transport and viscoelastic properties of these nanocomposite ion gels were investigated in relation to the microstructure. Despite their solid-like behavior, the nanocomposite ion gels exhibited a high ionic conductivity of approximately 10-2 S cm-1 at 30 °C, which is comparable to that of neat [C2mim][NTf2]. Intriguing viscoelastic responses, such as shearthinning and shear-induced sol-gel transitions, were found in all of the nanocomposite ion gels. By adjusting the silica concentration, the elastic modulus (G′) could be precisely controlled in a range of more than 3 orders of magnitude and reached approximately 106 Pa without a considerable decrease in the ionic conductivity; the characteristic viscoelastic response was also maintained. For the aggregation mechanism in [C2mim][NTf2], the reaction-limited cluster aggregation (RLCA) model was proposed by rheology and light scattering measurements. Introduction Ionic liquids (ILs) are comprised entirely of ions and are fluid at ambient conditions. ILs have attracted considerable interest in many fields of chemistry and in the chemical industry because of their potential to replace the conventional organic solvents used in chemical and catalytic reactions, separations, and purifications.1 Since ILs possess high ionic conductivity and wide electrochemical window, the use of an IL as an electrolyte is also an attractive research area in the quest for future power sources and electrochemical devices.2 Furthermore, the wide variety of available cations and anions afford a high flexibility in material design. Because task-specific properties, such as proton conduction, lithium-ion conduction, and electronic charge transport, can be molecularly designed into the ILs, the scope and utility of ILs as effective electrolytes have expanded to fuel cells,3 lithium batteries,4 and solar cells.5 Among the different attempts to realize electrochemical applications of ILs, the solidification of ILs has received an upsurge of interest to fabricate a new solid electrolyte. Although the polymerization of an IL6 and the doping of ILs into polymers7 have been proposed as methods to prepare polymer electrolytes based on ILs, the gelation of ILs offers a simple route to the fabrication of IL-based solid electrolytes; it is technically easy to control their properties by changing the amount of gelling agent. These types of gel material, which include a large amount of IL, are referred to as “ion gels″. A number of procedures to create ion gels have been successfully performed by employing different gelling agents, such as carbon nanotubes,8 liquid crystals,9 low molecular weight gelators,10 crystalline polymers,11 silica networks,12 glycolipid bilayers,13 and block copolymers.14 * To whom correspondence should be addressed. Telephone/Fax: +8145-339-3955. E-mail: [email protected]. † Department of Chemistry and Biotechnology. ‡ Instrumental Analysis Center.

Our group also reported ion gels composed of a chemically cross-linked polymer network swollen by an IL and investigated the ionic transport, thermal, and mechanical properties of these gels.15 Such ion gels could be easily prepared by the in situ polymerization of suitable monomers and cross-linkers in ILs. Through our previous works,15 the intrinsic properties of ion gels (e.g., film-forming ability, transparency, flexibility, high thermal and mechanical stability, high ionic conductivity) were attributed to the compatibility between a small amount of a polymer network and a large amount of an IL, where the IL played an important role as conductive species as well as plasticizer. Recent research aiming at the development of an applicable solid electrolyte includes the improvement of ionic conductivity and the design of task-specific properties into the ILs of ion gels.16 The introduction of new functionalities into the “network matrix″ of ion gels has been another subject of more recent researches. In our previous works, we have presented two types of covalently cross-linked, thermo-sensitive ion gels showing reversible volume changes in response to temperature: upper critical solution temperature (UCST)17 and lower critical solution temperature (LCST)18 behavior in ILs. He and Lodge prepared a physical ion gel consisting of an IL and a thermo-sensitive block copolymer, which preserved a thermally reversible sol-gel transition.19 The addition of colloidal materials to fluids occasionally leads to the gelation of the fluids. Such gels frequently exhibit intriguing viscoelastic properties, such as shear-thinning and thixotropy. For electrolyte materials, the addition of inorganic colloidal particles to polyethylene glycol-based electrolytes improves their performance as lithium battery electrolytes.20 The solidification of IL-based electrolytes by the use of inorganic nanoparticles has been reported for high performance dyesensitized solar cells (DSSCs).21 However, these works have mainly dealt with the electrochemical reaction and the performance of DSSC. For nanocomposite ion gels in which colloidal particulate networks percolate throughout the whole volume of

10.1021/jp8029117 CCC: $40.75  2008 American Chemical Society Published on Web 07/09/2008

9014 J. Phys. Chem. B, Vol. 112, No. 30, 2008 the IL, the characterization of their intrinsic properties, such as ionic transport, mechanical properties, and the microstructure of network matrix in the IL, remains in a preliminary stage, even as does the basic colloidal science of ILs. In a previous work,22 we investigated the colloidal stability of silica particles in various ILs by using the Derjaguin-LandauVerwey-Overbeek23 (DLVO) theory, which readily explained the qualitative feature of colloidal stability in terms of the interplay between the London-van der Waals attraction and electrostatic repulsion. From the DLVO estimation, together with practical experiments, it was found that the interparticle electrostatic repulsion appeared to be inefficient, due to the extensively high ionic strength of the ILs and the resulting surface-charge screening. Consequently, the silica colloidal particles formed aggregates in the ILs. In this study, we have explored nanocomposite ion gels formed by the flocculation of silica nanoparticles in a widely used, hydrophobic ionic liquid, 1-ethyl-3-methyl imidazolium bis(trifluoromethanesulfonyl)amide ([C2mim][NTf2]), to present a facile methodology affording highly conductive solid electrolytes and also functionalized soft materials. The ionic transport properties were evaluated by means of conductometry and diffusivity measurements and compared with those of the IL itself24 and the poly(methyl methacrylate) (PMMA)-based ion gels reported previously.15c To understand the viscoelastic behavior of the nanocomposite ion gel, as well as the dilute dispersions in the IL, we studied the rheological properties with different silica concentrations. Furthermore, this paper focuses on the aggregate microstructure of the silica particles formed in the IL. The aggregation mechanism for the silica particles in the IL is discussed by using a cluster-cluster aggregation model. Experimental Section Materials. The silica nanoparticles used in this study, Aerosil 200, were kindly supplied by Nippon Aerosil Co., Ltd. The primary particles of the Aerosil 200 are 12 nm in diameter and have hydrophilic silanol (Si-OH) groups on their surface. The silica particles were dried for 24 h in a vacuum oven at 120 °C before use. [C2mim][NTf2] was prepared on the basis of a metathesis reaction of freshly prepared 1-ethyl-3-methyl imidazolium bromide, [C2mim]Br, and lithium bis(trifluoromethanesulfonyl)amide, Li[NTf2], as described in our previous work.24 The IL was dried for 24 h at 70 °C under a vacuum condition and stored in a dry box. The water content of the IL was determined to be 29 ppm by Karl Fischer titration using a Mitsubishi Chemical CA-07 moisture meter. Preparation of IL Dispersion of Silica Particles. All of the dispersions of the silica particles in the IL were prepared by mechanical mixing with a conditioning mixer (AR-250, THINKY, Tokyo) for 10 min to ensure homogeneous mixing, followed by a 3 min degassing to remove air bubbles in the samples. The obtained samples were again dried for 24 h under vacuum with heating at 70 °C prior to use for each measurement. Measurements. The ionic conductivity of the samples was measured by complex impedance measurements, using a computer-controlled multipotentiostat (VMP2, Princeton Applied Research) over the frequency range of 1 Hz to 500 kHz at an amplitude of 10 mV. These measurements were carried out in a cell with mirror finished stainless-steel electrodes at controlled equilibrium temperatures. The cell constant was determined by using a 0.1 M KCl aqueous solution as a reference at 25 °C. The pulsed-gradient spin-echo NMR (PGSE-NMR) measurements were conducted by using a JEOL AL400 spectrometer

Ueno et al.

Figure 1. Photographs of the dispersions of silica nanoparticles in [C2mim][NTf2] with different silica particle concentrations: (a) neat [C2mim][NTf2], (b) 1 wt %, (c) 3 wt %, (d) 5 wt %, and (e) 15 wt %.

with a 9.4 T narrow bore superconducting magnet equipped with a JEOL pulse field gradient probe and a current amplifier. The self-diffusion coefficients were measured using a simple Hahn spin-echo sequence, incorporating a sine gradient pulse in each τ period. The interval between two gradient pulses, ∆, was set at 50 ms, and the duration of the field gradient, δ, was varied. The quasi-solidified samples were inserted into a 5-mm (o.d.) NMR microtube (BMS-005J, Shigemi, Tokyo) by mild centrifugal treatment at approximately 3000 rpm for several minutes to avoid the inclusion of air bubbles into the samples. The measurements for the cationic and anionic self-diffusion coefficients were made by using 1H (399.7 MHz) and 19F (376.1 MHz) nuclei, respectively. Rheological measurements were performed with a rheometer (Physica MCR301, Anton Paar) under dry air conditions at 25 °C. Three different geometries were employed for the measurements: two types of cone-and-plate systems, one with a diameter of 25 mm and a cone angle of 2° and the other with a diameter of 50 mm and a cone angle of 1°, as well as a parallel plate system with a diameter of 25 mm. To erase any previous shear histories and to make the samples establish their equilibrium structures, a steady preshear was applied at a shear rate of 1 s-1 for 60 s followed by a 120 s rest period before each dynamic rheological measurement. To determine the linear viscoelastic limit, a dynamic measurement as a function of strain amplitude was conducted before each dynamic frequency sweep measurement. The frequency sweep was performed in this regime. Static light scattering (SLS) measurements were carried out with a DLS-8000 optical system (Otsuka Electronics) equipped with a 10 mW He-Ne laser at a wavelength of 633 nm. The dispersion of silica nanoparticles in the IL was placed on a collodion-coated TEM copper grid. TEM observation was performed with a JEOL JEM2000FXII operated at 100 kV. Results and Discussion Gelation of the IL. Figure 1 shows the appearance of the dispersions of silica nanoparticles in [C2mim][NTf2]. Although no gelation occurs with the addition of 1 wt % silica nanoparticles to the IL (Figure 1b), gelation can be observed even with the addition of 3 wt % nanoparticles (Figure 1c). A further addition increases the hardness of the nanocomposite ion gels. At 15 wt % silica content (Figure 1e), the composite looks like wet powder. To study the gelation of the nanocomposites in detail, rheological measurements were performed on the IL dispersions

Ion Gels Based on Silica Nanoparticles and an IL

J. Phys. Chem. B, Vol. 112, No. 30, 2008 9015

Figure 2. Shear rate dependencies of viscosity for IL dispersions with different silica particle concentrations.

Figure 3. Elastic (G′) and viscous (G″) moduli as functions of frequency for IL dispersions with different silica particle concentrations.

of silica nanoparticles with different silica concentrations. Figure 2 shows the steady shear response of the IL dispersions. For the neat IL (0 wt %), the viscosity is independent of the shear rate, indicating that the IL behaves as a Newtonian fluid in the measured range of shear rate. The shear-independent viscosity is fairly consistent with the reported viscosity of the IL (32.6 Pa s) at 25 °C.24 On the other hand, the viscosity decreases with an increase in the shear rate in all of the dispersions used in this study. Such shear-thinning behavior derived from the shear-induced disruption of interparticle physical bonds that can be observed in many of the typical dispersions of colloidal particles.25 Recently, the shear-thinning response has also been reported in a dispersion of hematite in an ionic liquid, 1-ethlyl3-methyl imidazolium ethylsulfate.26 At concentrations above 4 wt % of silica particles, the dispersions show a power-law decease with a slope of -1 at low shear rates. Such distinct rheological responses at higher silica concentrations imply the existence of a network structure having a yield stress.27 To study the viscoelastic behavior of the dispersions in the IL in more detail, we conducted oscillatory shear measurements. The results of the dynamic rheological properties measured under a constant strain amplitude within a linear viscoelastic regime are shown in Figure 3. The 1 wt % dispersion is a liquid; the viscous modulus (G″) exceeds the elastic modulus (G′) over the entire frequency range, and both moduli exhibit dependence on frequency. At 2 wt %, the dispersion shows a complicated response. The moduli exhibit a crossover at a high frequency. However, G′ is dominant and the moduli become relatively independent of frequency toward the low end of the frequency spectrum. Above 3 wt %, G′ is significantly larger than G″ and is independent of frequency over the entire measured range. Therefore, we can conclude that the dispersions containing above 2 wt % of the silica particles behave as a soft solid-like material (gel). In the literature, as reported by Huang et al.,21b the solidification of an IL, 1-butyl-3-methyl imidazolium tetrafluoroborate, by the addition of 2 wt % of the same silica particles were confirmed by a sample tube inversion method. The previously reported ion gel, which consisted of a PMMA network and [C2mim][NTf2], required that the gelling agent of the PMMA network be at least in the range of 6 wt % to 10 wt %, corresponding to the molar ratio of [MMA]/[[C2mim][NTf2]]

from 2/8 to 3/7, to solidify the IL.15c In sharp contrast, in the case of the nanocomposite ion gel, the gelation was achieved by the addition of 2 wt % of silica. The self-standing ion gel can be obtained with a much lower concentration of the silica particles used as a gelling agent. For the nanoconposite gel formed by space-spanning particle aggregates, it is also known that particle size strongly affects the gelation behavior. The critical volume fraction for the gelation has been theoretically derived in the reported literature, taking the gravity effect and particle radius into account.28 While we mainly used silica nanoparticles having 12 nm in diameter in this study, the use of larger particles resulted in a higher critical concentration for the gelation of the IL. The details are shown in the Supporting Information. Ionic Transport. To confirm their potential use as solid electrolytes, the ionic transport properties of the nanocomposite ion gels were investigated and compared to those of a PMMAbased ion gel. The temperature dependence of the ionic conductivity for the nanocomposite ion gels is shown in Figure 4, along with that for neat [C2mim][NTf2]24 and for the PMMAbased ion gel.15c The ionic conductivity of the nanocomposite ion gels is higher than that of the PMMA-based ion gel over the entire measured range of silica concentrations, although for comparison we chose the PMMA-based ion gel (10 wt % of PMMA network) that had the highest ionic conductivity. Despite its quasi-solidified appearance, the ionic conductivity at low silica concentrations reaches a value of approximately 10-2 S cm-1 at 30 °C, which is comparable to that of [C2mim][NTf2]. Figure 5 presents the temperature dependency of the selfdiffusion coefficients of the cationic and anionic species in a nanocomposite ion gel, along with those in neat [C2mim][NTf2]24 and in the PMMA-based ion gel.15c As seen in [C2mim][NTf2] and in the PMMA-based ion gel, the [C2mim] cation diffuses faster than the [NTf2] anion in the nanocomposite ion gels. The dependence of the ionic transport on the concentration of silica particles is shown in Figure 6. The addition of silica particles has little influence on the ionic conductivity (Figure 6a). As shown in Figure 6b, the relative diffusivity, Dg/D0, of the ions decreases slightly with an increase in the silica concentration, due to the obstruction of the silica

9016 J. Phys. Chem. B, Vol. 112, No. 30, 2008

Ueno et al.

Figure 6. Concentration dependencies of ionic transport properties for nanocomposite ion gels at 30 °C: (a) ionic conductivity and (b) ionic self-diffusion coefficient. Figure 4. Temperature dependencies of ionic conductivity for [C2mim][NTf2], nanocomposite ion gels, and PMMA-based ion gel. Data for [C2mim][NTf2] and PMMA-based ion gel are cited from refs 24 and 15c, respectively.

Figure 5. Temperature dependencies of ionic self-diffusion coefficient for [C2mim][NTf2], 5 wt % nanocomposite ion gel, and PMMA-based ion gel. Data for [C2mim][NTf2] and PMMA-based ion gel are cited from refs 24 and 15c, respectively.

particles by the gelling agent; where D0 and Dg are the selfdiffusion coefficients of the diffusive species, measured by PGSE-NMR, in the neat IL and in the nanocomposite ion gels, respectively. Here we consider the differences in the ionic transport properties between the nanocomposite ion gel and PMMA-based ion gel. The ionic conductivity is affected by the ionic mobility and ionic carrier concentration. The higher conductivity of the nanocomposite ion gels is due to both a higher mobility and higher carrier concentration in the system. In the PMMA-based ion gels,15b the PMMA network polymer and the IL are compatible due to the interaction between them. In contrast, the silica-based nanocomposite ion gels can be characterized as phase-separated binary systems where the ionic transport is

considered to be only affected by the obstruction of the interconnecting silica network. A large open network structure derived from the lower concentrations of the gelling agent minimizes the obstruction of ionic migration in the nanocomposite ion gels. Due to the gelation even by the addition of a small amount of silica particles, the reduction in the total ionic carrier concentration is also kept to a minimum. Consequently, the higher ionic conductivity of the nanocomposite ion gels can be achieved. Interestingly, the reduction in Dg/D0 for the [C2mim] cation is found to be larger than that of the [NTf2] anion in all of the nanocomposite ion gels (Figure 6b). This suggests that [C2mim] cations preferentially interact with the silica surface. Recently, the structure of the IL/SiO2 interface has been studied by using sum-frequency vibrational spectroscopy; the imidazolium cation was detected at the IL/SiO2 interface.29 In contrast to a nanocomposite ion gel, a larger decrease in Dg/D0 for the [NTf2] anion has been observed in a PMMA network-based ion gel, indicating that the [NTf2] anion, rather than the [C2mim] cation, strongly interacts with the PMMA network matrix.15c Viscoelastic Properties. The effects of particle concentration on the mechanical properties of a nanocomposite ion gel were studied by dynamic shear measurements. Figure 7 shows the elastic G′ and viscous G″ moduli as functions of the strain amplitude for the nanocomposite ion gels containing different amounts of silica nanoparticles. While G′ is larger than G″ within the region of linear viscoelasticity, the transition from a solid-like (G′ > G″) to a liquid-like (G′ < G″) behavior is detected for the high strain amplitudes in all of the nanocomposite ion gels. These results indicate that the nanocomposite ion gels behave as pseudoplastic fluids showing a shear-induced sol-gel transition. The same data for G′ are plotted in Figure 8 as a function of stress amplitude. Distinct two step decreases in G′ are observed, suggesting the existence of two types of yield stresses. This characteristic feature has also been reported in the literature.20a,30 The first and second yield stresses are recognized as the stress required to barely disrupt the network and the minimum stress needed to cause the sample to flow, respectively. In this study, we define the first yield stress as the appropriate value of yield stress (τy) for the network structures in the nanocomposite ion gels. Moreover, G′ and τy are considerably enhanced as the concentration of silica particles is increased, and the value of G′ reaches a value as high as 106 Pa at a concentration of 15 wt %.

Ion Gels Based on Silica Nanoparticles and an IL

Figure 7. Dynamic strain sweep for nanocomposite ion gels with different silica concentrations. The frequency was held constant at 1 s-1.

Figure 8. Elastic modulus (G′) as a function of dynamic stress amplitude for nanocomposite ion gels with different silica concentrations.

It is known that the G′ of a physical gel formed from colloidal aggregates exhibits power-law dependence on the volume fraction (φ) of particles; G′ ∝ φn.31 As shown in Figure 9, relatively good correlations between G′, τy, and φ, according to the power-law, are found for our nanocomposite ion gels; for G′, n ) 4.65, and for τy, n ) 4.53. Thus, the mechanical properties, G′ and τy, can be precisely controlled over more than 3 orders of magnitude by adjusting the amount of silica particles, while all of the nanocomposite ion gels show fluidic behavior under sufficient shear application, which can be an advantage in processing. Microstructure. Even under a high vacuum condition and electron beam irradiation, ILs can be observed by a scanning electron microscope and a transmission electron microscope (TEM) without evaporation and without the accumulation of

J. Phys. Chem. B, Vol. 112, No. 30, 2008 9017

Figure 9. Dependence of elastic modulus (G′) and yield stress (τy) on volume fraction of silica particles (φ). The solid lines represent the results of power-law fitting.

Figure 10. In situ TEM images for 5 wt % nanocomposite ion gel. The focus cannot be sharp due to the presence of IL.

electronic charges.32 Therefore, by in situ TEM observations using ILs as dispersion media, we can evaluate visually the dispersibility and microstructure of colloidal materials in ILs. The particulate network structure of silica was confirmed by in situ TEM observation. As shown in Figure 10, due to poor colloidal stability in the IL, silica particles form interconnected aggregates having a loosely outspread fractal structure, and thereby the aggregated clusters percolate throughout the whole volume of the IL, leading to the gelation of the IL even with the addition of a small amount of silica particles. It is known that the fractal structure of the aggregated clusters is strongly influenced by the aggregation mechanism in a dispersion medium. Two distinct limiting regimes for colloidal aggregation have been identified: diffusion-limited cluster aggregation (DLCA) and reaction-limited cluster aggregation (RLCA).33 Through the numerous experiments and computer simulations made to date, the fractal dimension (df) has been found to have

9018 J. Phys. Chem. B, Vol. 112, No. 30, 2008

Ueno et al. Conclusion

Figure 11. Normalized scattering intensity as a function of wave vector (q) for IL dispersions of silica nanoparticles, measured at 25 °C. The broken line represents the result from power-law fitting. For the calculation of q, a refractive index of 1.42 for [C2mim][NTf2] was used from ref 22.

universal values of df ) 1.8 for the DLCA and df ) 2.1 for the RLCA.34 Furthermore, the exponent of n in the dependence of G′ on φ (G′ ∝ φn), as shown in Figure 8, also provides information on the colloidal aggregation mechanism and the microstructure of clusters; n ) 3.5 ( 0.2 for the DLCA, and n ) 4.5 ( 0.2 for the RLCA.31 From the results of the viscoelastic measurements (Figure 9), the experimental value of n for our nanocomposite ion gel (n ) 4.65) agrees very well with the reported value of n for the RLCA model. SLS measurements were also conducted to determine the actual value of df for the clusters formed in the IL, where the scattering intensity I(q) shows power-law decay against the wave vector q; I(q) ∝ q-df.35 Figure 11 shows a normalized I(q) as a function of q for the IL dispersions. The obtained value of df from the slope in Figure 11 (df ) 2.14) is again fairly consistent with the predicted df for the RLCA. These experimental evidence clearly indicate that the aggregation in the IL is limited by the RLCA model, while the DLCA-dominated aggregation has been reported in a similar system based on the same silica particles in a mineral oil.36 In the DLCA model, where colloidal particles have no or much weaker repulsion compared to the particle energy of kT, every collision causes the formation of a new cluster. In contrast, in the RLCA model, only a fraction of the collisions leads to the formation of a new cluster. In this case, the particles possess a moderate repulsive energy barrier comparable to or larger than kT, and the aggregation is limited by the probability of overcoming such a repulsive barrier according to the Bolzmann distribution.37 Hence, the RLCA in the IL suggests the existence of a moderate repulsive-interaction between the particles. However, the electrostatic repulsion was not effective enough to stabilize the colloidal particles in the IL due to the extremely high ionic strength and the resulting surface-charge screening.22 In the present case, such moderate repulsion is not likely to come from an electrostatic force but rather from steric hindrance and/or the solvation force38 of the IL itself. At the silica surfaces, possibly the [C2mim] cation rather than the [NTf2] anion preferentially interacts with the surface. In fact, we observed a larger decrease in D/D0 for the [C2mim] cation with increasing silica concentrations (Figure 6b).

In this paper, nanocomposite ion gels consisting of interconnected networks of silica nanoparticles and an IL, [C2mim][NTf2], were studied. We demonstrated a facile route to the preparation of ion gels that exhibit high ionic conductivity comparable to that of the IL at room temperature and a shearresponsive sol-gel transition. In the nanocomposite ion gels, the high ionic conductivity can be attributed to the high ionic mobility and high ionic concentration arising from the fact that the nanocomposite ion gels are a phase-separated system containing a small amount of silica nanoparticles forming interconnected network structures. The intriguing viscoelastic properties (e.g., shear thinning, pseudoplastic fluidity, and dependence of G′ and τy on φ) offer advantages for the fabrication of high performance solid electrolytes and new functional soft materials based on ILs in terms of easy processability and precise controllability of mechanical strength. From the investigation of the fractal microstructure formed in the IL, we found that colloidal aggregation is limited by the RLCA model, implying the presence of moderate interparticle repulsion in the IL. The assumed repulsive interaction may be due to steric hindrance and/or the solvation force of the IL itself, weakly associated with the silica surfaces. It appears possible that the [C2mim] cations are associated with the surface in our case, as detected by diffusivity measurements. A RLCA model for ILs was found in the literature by Zhou et al. The reactionlimited aggregation of TiO2 nanoparticles resulted in mesoporous spherical aggregates having a sponge-like nanostructure in 1-butyl-3-methyl imidazolium tetrafluoroborate.39 Among a number of studies on ILs with colloids, a striking stabilization of colloidal materials in ILs has been reported, even in the absence of any stabilizers, such as surfactants and polymers.8,40 Nevertheless, only a few studies have been performed on the characterization of colloidal behavior in ILs;41 the details for such specific stabilization in ILs have not been understood completely. Although our silica particles were unstable in the IL, leading to the formation of a network, the evidence for the RLCA in the IL may provide information on such striking stabilization by IL-based steric and/or solvation forces in ILs, where electrostatic stabilization is strongly suppressed. Investigations of the interaction between the silica surface and ILs using spectroscopic methods and the dependence of the microstructure and viscoelastic properties on the ionic species of ILs need to be explored to gain a greater understanding of the fundamental colloidal science of ILs. Acknowledgment. This work was supported by a Grant-inAid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (No. 452-17073009). The authors thank Nippon Aerosil Co., Ltd., for providing the silica nanoparticles used in this work. K.U. acknowledges financial support provided by JSPS. Supporting Information Available: Dependence of particle size on gelation of the IL. This material is available free of charge via the Internet at http://pubs.asc.org. References and Notes (1) (a) Welton, T. Chem. ReV. 1999, 99, 2071. (b) Holbrey, J. D.; Seddon, K. R. Clean Products Process 1999, 1, 223. (c) Wasserscheid, P.; Keim, W. Angew.Chem., Int. Ed. 2000, 39, 3772. (2) Galonski, M.; Lewandowski, A.; Stepniak, I. Electrochim. Acta 2006, 51, 5567. (3) Noda, A.; Susan, M. A. B. H.; Kubo, K.; Mitsushima, S.; Hayamizu, K.; Watanabe, M. J. Phys. Chem. B 2003, 107, 4024.

Ion Gels Based on Silica Nanoparticles and an IL (4) MacFarlane, D. R.; Huang, J.; Forsyth, M. Nature 1999, 402, 792. (5) Papageorgiou, N.; Athanassov, Y.; Armand, M.; Bonhoˆte, P.; Pettersson, H.; Azam, A.; Gra¨tzel, M. J. Electrochem. Soc. 1996, 143, 3099. (6) (a) Yoshizawa, M.; Ohno, H. Chem. Lett. 1999, 889. (b) Hirao, M.; Ito, K.; Ohno, H. Electrochim. Acta 2000, 45, 1291. (7) (a) Watanabe, M.; Yamada, S.; Sanui, K.; Ogata, N. J. Chem. Soc., Chem. Commun. 1993, 929. (b) Lewandowski, A.; Swiderska, A. Solid.State Ionics 2004, 169, 21. (c) Singh, B.; Sekhon, S. S. J. Phys. Chem. B 2005, 109, 16539. (8) Fukushima, T.; Kosaka, A.; Ishimura, Y.; Yamamoto, T.; Takigawa, T.; Ishii, N.; Aida, T. Science 2003, 300, 2072. (9) Firestone, M. A.; Dzielawa, J. A.; Zapol, P.; Curtiss, L. A.; Seifert, S.; Dietz, M. L. Langmuir 2002, 18, 7258. (10) Hanabusa, K.; Fukui, H.; Suzuki, M.; Shirai, H. Langmuir 2005, 21, 10383. (11) (a) Carlin, R. T.; Delong, H. C.; Fuller, J.; Trulove, P. C. J. Electrochem. Soc. 1994, 141, L73. (b) Kawaushi, T.; Kumaki, J.; Okoshi, K.; Yashima, E. Macromolecules 2005, 38, 9155. (12) Ne´ouze, M.-A.; Bideau, J. L.; Gaveau, P.; Bellayer, S.; Vioux, A. Chem. Mater. 2006, 18, 3931. (13) Kimizuka, N.; Nakashima, T. Langmuir 2001, 17, 6759. (14) He, Y.; Boswell, P. G.; Bu¨hlmann, P.; Lodge, T. P. J. Phys. Chem. B 2007, 111, 4645. (15) (a) Noda, A.; Watanabe, M. Electrochim. Acta 2000, 45, 1265. (b) Seki, S.; Susan, M. A. B. H.; Kaneko, T.; Tokuda, H.; Noda, A.; Watanabe, M. J. Phys. Chem. B 2005, 109, 3886. (c) Susan, M. A. B. H.; Kaneko, T.; Noda, A.; Watanabe, M. J. Am. Chem. Soc. 2005, 127, 4976. (16) (a) Shobukawa, H.; Tokuda, H.; Susan, M. A. B. H.; Watanabe, M. Electrochim. Acta 2005, 50, 3872. (b) Li, Z.; Liu, Y.; Liu, H.; He, P.; Zhang, Q.; Li, J. Solid State Ionics 2006, 177, 1281. (17) Ueki, T.; Watanabe, M. Chem. Lett. 2006, 35, 964. (18) Ueki, T.; Watanabe, M. Langmuir 2007, 23, 988. (19) (a) He, Y.; Lodge, T. P. Chem. Commun. 2007, 2732. (b) He, Y.; Lodge, T. P. Macromolecules 2008, 41, 167. (20) (a) Raghavan, S. R.; Riley, M. W.; Fedkiw, P. S.; Khan, S. A. Chem. Mater. 1998, 10, 244. (b) Zhang, X.; Li, Y.; Kahn, S. A.; Fedkiw, P. S. J. Electrochem. Soc. 2004, 151, A1257. (21) (a) Wang, P.; Zakeeruddin, S. M.; Comte, P.; Exnar, I.; Gra¨tzel, M. J. Am. Chem. Soc. 2003, 125, 1166. (b) Yang, H.; Yu, C.; Song, Q.; Xia, Y.; Li, F.; Chen, Z.; Li, X.; Yi, T.; Huang, C. Chem. Mater. 2006, 18, 5173. (c) Kato, T.; Kado, T.; Tanaka, S.; Okazaki, A.; Hayase, S. J. Electrochem. Soc. 2006, 153, A626. (d) Katakabe, T.; Kawano, R.; Watanabe, M. Electrochem. Solid-State Lett. 2007, 10, F23. (22) Ueno, K.; Inaba, A.; Kondoh, M.; Watanabe, M. Langmuir 2008, 24, 5253.

J. Phys. Chem. B, Vol. 112, No. 30, 2008 9019 (23) (a) Derjaguin, B. V.; Landau, L. Acta. Physicochim. U.S.S.R. 1941, 14, 633. (b) Verway, E. J. W.; Overbeek, J. T. G. Theory of the Stability of Liophobic Colloids; Elsevier: Amsterdam, 1948. (24) Tokuda, H.; Ishii, K.; Susan, M. A. B. H.; Tsuzuki, S.; Hayamizu, K.; Watanabe, M. J. Phys. Chem. B 2006, 110, 2833. (25) Buscall, R.; Corner, T.; Stageman, J. F. Polymer Colloids; Elsevier Applied Science Publishers: New York, 1985; Chapter 6. (26) Altin, E.; Gradl, J.; Peukert, W. Chem. Eng. Technol. 2006, 29, 1347. (27) Macosko, C. W. Rheology: Principles, Measurements and Applications; VCH Publishers: New York, 1994. (28) Allain, C.; Cloitre, M.; Wafra, M. Phys. ReV. Lett. 1995, 74, 1478. (29) Fitchett, B. D.; Conboy, J. D. J. Phys. Chem. B 2004, 108, 20255. (30) (a) Bonnecaze, R. T.; Brady, J. F. J. Rheol. 1992, 36, 73. (b) Husband, D. M.; Aksel, N.; Gleissle, W. J. Rheol. 1993, 37, 215. (31) Buscall, R.; Mills, P. D. A.; Goodwin, J. W.; Lawson, D. W. J. Chem. Soc., Faraday Trans. 1 1988, 84, 4249. (32) (a) Scheeren, C. W.; Machado, G.; Dupont, J.; Fichtner, P. F. P.; Texeira, S. R. Inorg. Chem. 2003, 42, 4738. (b) Kuwabata, S.; Kongkanand, S.; Oyamatsu, D.; Torimoto, T. Chem. Lett. 2006, 35, 600. (33) Lin, M. Y.; Lindsay, H. M.; Weitz, D. A.; Ball, R. C.; Klein, R.; Meakin, P. Nature 1989, 339, 360. (34) For example, (a) Weitz, D. A.; Oliveria, M. Phys. ReV. Lett. 1984, 52, 1433. (b) Schaefer, D. W.; Martin, J. E.; Wiltzius, P.; Cannell, D. S. Phys. ReV. Lett. 1984, 52, 2371. (c) Meakin, P. Phys. ReV. Lett. 1983, 51, 1119. (35) Lin, M. Y.; Lindsay, H. M.; Weitz, D. A.; Ball, R. C.; Klein, R.; Meakin, P. J. Phys.: Condens. Matter 1990, 2, 3093. (36) Khan, S. A.; Zoeller, N. J. J. Rheol. 1993, 37, 1225. (37) Lin, M. Y.; Lindsay, H. M.; Weitz, D. A.; Ball, R. C.; Klein, R.; Meakin, P. Phys. ReV. A 1990, 41, 2005. (38) Israelachvili, J.; Wennerstro¨m, H. Nature 1996, 379, 218. (39) Zhou, Y.; Antonietti, M. J. Am. Chem. Soc. 2003, 125, 14960. (40) (a) Wojtko´w, W.; Trzeciak, A. M.; Choukroun, R.; Pellegatta, J. L. J. Mol. Catal. A: Chem. 2004, 224, 81. (b) Gelesky, M. A.; Umpierre, A. P.; Machado, G.; Correia, R. R. B.; Magno, W. C.; Morais, J.; Ebeling, G.; Dupont, J. J. Am. Chem. Soc. 2005, 127, 4588. (c) Torimoto, T.; Okazaki, K.; Kiyama, T.; Hirahara, K.; Tanaka, N.; Kuwabata, S. Appl. Phys. Lett. 2006, 89, 243117. (41) (a) Ott, L. S.; Cline, M. L.; Deetlefs, M.; Seddon, K. R.; Finke, R. G. J. Am. Chem. Soc. 2005, 127, 5758. (b) Fonseca, G. S.; Machado, G.; Teixeira, S. R.; Fecher, G. H.; Morais, J.; Alves, M. C. M.; Dupont, J. J. Colloid Interface Sci. 2006, 301, 193.

JP8029117