Ionic Liquid Tunes Microemulsion Curvature - Langmuir (ACS

Liping Liu, Pierre Bauduin, Thomas Zemb, Julian Eastoe and Jingcheng Hao* .... mm thicknesses with Kapton windows on a Huxley-Holmes high-flux camera...
0 downloads 0 Views 328KB Size
Langmuir 2009, 25, 2055-2059

2055

Ionic Liquid Tunes Microemulsion Curvature Liping Liu,† Pierre Bauduin,‡ Thomas Zemb,‡ Julian Eastoe,§ and Jingcheng Hao*,† Key Laboratory of Colloid and Interface Chemistry, Shandong UniVersity, Ministry of Education, Jinan 250100, PR China, Institut de Chimie Se´paratiVe de Marcoule, UMR 5257 CEA/CNRS/UM2/ ENSCM, BP 17171 CEA Marcoule, 30207 Bagnols-sur-Ce`ze, France, and School of Chemistry, UniVersity of Bristol, Bristol BS8 1TS, U.K. ReceiVed NoVember 1, 2008. ReVised Manuscript ReceiVed December 19, 2008 Middle-phase microemulsions formed from cationic dioctadecyldimethylammonium chloride (DODMAC), anionic sodium dodecylsulfate (SDS), n-butanol, and n-heptane were studied. An ionic liquid (IL), 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim][BF4]), was employed as the electrolyte in the aqueous media instead of inorganic salts usually used in microemulsion formulation. Studies have been carried out as a function of the concentrations of [bmim][BF4], n-butanol, total surfactant (cDODMAC+SDS), and temperature on the phase behavior and the ultralow interfacial tensions in which the anionic component is present in excess in the catanionic film. Ultralow interfacial tension measurements confirmed the formation of middle-phase microemulsions and the necessary conditions for stabilizing middle-phase microemulsions. Electrical conductivity, small-angle X-ray scattering (SAXS), and smallangle neutron scattering (SANS) experiments were also performed, indicating that the typical heptane domain size has an average radius of 360 Å and the ionic liquid induces softening of the charged catanionic film. Most interestingly, the IL concentration (cIL) is shown to act as an effective interfacial curvature-control parameter, representing a new approach to tuning the formulation of microemulsions and emulsions. The results expand the potential uses of ILs but also point to the design of new ILs that may achieve superefficient control over interfacial and self-assembly systems.

Introduction 1

Microemulsions (MEs), comprising amphiphiles and two immiscible fluids (usually oil and water), are important and interesting self-assembly systems with wide-ranging applications in research and industry. It is recognized that the interfacial rigidity and spontaneous curvature 1/Ro of the stabilizing amphiphile film are the fundamental control parameters for dictating the stability, properties, and structures of ME systems. When Ro is much larger than the typical surfactant molecular length lS, threephase Winsor III MEs may form. Winsor III MEs exhibit ultralow interfacial tensions (,10-2 mN · m-1), making them very efficient media for decontamination, solubilization, and enhanced oil recovery. Two main classes of MEs are known:2 (1) With singlechain nonionic amphiphiles, the interfacial film is very flexible, and thermal fluctuations dominate, giving rise to disordered domain structures. (2) With ionic surfactants, the films are generally stiffer, meaning that the limiting spontaneous curvature stabilizes well-defined dispersed domain shapes such as spheres, cylinders, or lamellae.3 For type-1 systems, interfacial curvature, phase stability, and structure can be readily tuned by temperature. However, with type-2 systems temperature effects are typically much weaker than for type-1 systems,4 so curvature control can be achieved only by adding electrolytes or cosurfactants.4 * Corresponding author. Tel/Fax: +86 531 88366074. E-mail: jhao@ sdu.edu.cn. † Shandong University, ‡ Institut de Chimie Se´parative de Marcoule, § University of Bristol. (1) Zana, R. Dynamics of Surfactant Self-Assemblies: Micelles, Microemulsions, Vesicles, and Lyotropic Phases; Surfactant Science Series; Marcel Dekker: New York, 2005; Vol. 125. (2) Chevalier, Y.; Zemb, Th. Rep. Prog. Phys. 1990, 53, 279. (3) Zemb, Th. Colloids Surf., A 1997, 129-130, 435. (4) Salager, J. L.; Anton, R. E. Ionic Microemulsions; In Handbook of Microemulsions Science and Technology; Kumar, P., Mittal, K., Eds.; . Marcel Dekker: New York, 1999; Chapter 8, pp 247-280.

Recently, ionic liquids (ILs) have received much attention because of special physical and chemical properties such as low volatility, a wide electrochemical window, nonflammability, high thermal stability, and a wide liquid range.5,6 It has been realized that ILs can be used as solvents for self-assembled structures such as micelles, reverse micelles, and liquid crystals.7 Electron microscopy (EM), laser light scattering (LLS), and small-angle neutron scattering (SANS) results have provided evidence for nanodomains in interfacially flexible type-1 IL-in-oil MEs composed of nonionic surfactant, cyclohexane, and IL 1-butyl3-methylimidazolium tetrafluoroborate ([bmim][BF4]).8,9 However, little attention has been focused on type-2 MEs with ILs, and further studies are still necessary to provide more details. In this article, it is shown that interfacially stiff type-2 IL MEs can also be formulated with ILs (i.e., ILs afford significant new possibilities for tailing the properties of microemulsions). These results pave the way for new potential applications of ILs in emulsion and microemulsion systems.

Experimental Section Chemicals. A typical ionic liquid, 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim][BF4], chemical structure Figure 1a insert), was synthesized as described in the literature.5 n-Heptane and n-butanol were obtained from Fluka Chemie GmbH. The singlechained anionic sodium dodecylsulfate (SDS) obtained from the West Centre Factory (Beijing) was recrystallized 3 times from ethanol, and the double-chained cationic dioctadecyldimethylammonium chloride (DODMAC, purity >99%) was supplied by Kao Soap Co. (5) Welton, T. Chem. ReV. 1999, 99, 2071. (6) Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc. 2002, 124, 14247. (7) Hao, J.; Zemb, Th. Curr. Opin. Colloid Interface Sci. 2007, 12, 129. (8) Gao, H.; Li, J.; Han, B.; Chen, W.; Zhang, J.; Zhang, R.; Yan, D. Phys. Chem. Chem. Phys. 2004, 6, 2914. (9) Eastoe, J.; Gold, S.; Rogers, S. E.; Paul, A.; Welton, T.; Heenan, R. K.; Grillo, I. J. Am. Chem. Soc. 2005, 127, 7302.

10.1021/la8036378 CCC: $40.75  2009 American Chemical Society Published on Web 01/22/2009

2056 Langmuir, Vol. 25, No. 4, 2009

Liu et al.

Figure 1. Phase behavior of middle-phase microemulsions as a function of (a) added IL ([bmim][BF4], chemical structure Figure 1a insert) at cDODMAC+SDS ) 2.0 wt % and cn-butanol ) 11.0 wt %; (b) added n-butanol at cDODMAC+SDS ) 2.0 wt % and cIL ) 2.31 wt %; (c) added total surfactants at cIL ) 2.38 wt % and cn-butanol ) 11.0 wt %; (d) effect of temperature at cIL ) 2.46 wt %, cDODMAC+SDS ) 2.0 wt % and cn-butanol ) 11.0 wt %. For all systems, the weight ratio cDODMAC/cSDS ) 1:4 was fixed, the oil is heptane, and the water-to-oil volume ratio is Vw/Vo ) 1:1. For the a, b, and c systems, T ) 25.0 ( 0.1 °C.

Formulation for Optimal Conditions of Middle-Phase Microemulsions Obtained by an Orthogonal Design (OD). 10 Sample solutions were prepared by weight in graduated tubes, which were sealed with paraffin, as a function of IL concentration (wt %). The optimal formation conditions for middle-phase microemulsions in the system of a typical IL, [bmim][BF4], cationic (DODMAC)-anionic (SDS) surfactants, n-butanol, n-heptane, and water were thus obtained. The orthogonal design (OD) is an approach to formulating complex multicomponent, multiparameter systems such as microemulsions. The aim of OD is to understand how variations in one component or parameter influence the overall formulation and also isolate the effects of all of the individual components. The three-level L9 (34) table for OD was used to explore the conditions necessary for the formation of middle-phase microemulsions. The related data are tabulated in Supporting Information (Tables 1-3). From the results for the ratio of VM (mL · g-1, the volume of middle-phase microemulsions, Table 3 in Supporting Information) and cDODMAC +SDS (the total amount of surfactants), it can be concluded that the effective factors from the primary to the secondary are sequentially the ratios of SDS to DODMAC and the concentrations of n-butanol and [bmim][BF4]. The formation conditions for the middle-phase microemulsions are cDODMAC/cSDS ) 1:5, cn-butanol ) 11.0 wt %, and cIL ) 3.25 wt %. However, from the absolute values (Table 3 in Supporting Information) of ∆V ) |Vo - Vw| (Vo represents the volume of oil phase and Vw represents that of the water phase, Table 3 in Supporting Information), the effective factors from the primary to the secondary are sequentially the concentration of n-butanol, the ratios of SDS to DODMAC, and the concentration of [bmim][BF4]. The conditions are cDODMAC/cSDS ) 1:4, cn-butanol ) 12.0 wt %, and cIL ) 3.25 wt %. From the ∆V data, it can also be concluded that the three effective factors are almost the same because of the small change in ∆V values. With regard to the larger change in VM/cDODMAC+SDS values, this effect is the primary factor. Therefore, according to the experimental results, the optimal formulation conditions of the middle-phase microemulsions in the mixed DODMAC and SDS systems with [bmim][BF4] are listed in Table 1. (10) Geramita, A. V.; Seberry, J. Orthogonal Designs: Quadratic Forms and Hadamard Matrices; Lecture Notes in Pure and Applied Mathematics; Marcel Dekker: New York, 1979; Vol. 43.

Table 1. Optimal Formulation Conditions of the Middle-Phase Microemulsions in the Mixed DODMAC and SDS Systems with [bmim][BF4] compositions

optimal formulation conditions

cIL (wt %) cDODMAC/cSDS cn-butanol (wt %)

3.25 1:4 -1:5 11.0-12.0

Interfacial Tension Measurements. To measure the interfacial tensions between the middle-phase microemulsions and excess oil (γME-O) or water (γME-W), a spinning-drop interfacial tensiometer (Model-500, University of Texas) was used. A droplet (4 µL) of the low-density phase is injected into a matrix of the high-density phase in a horizontally mounted capillary. The capillary is rotated around its axis, and as the speed of rotation is increased, the drop deforms along the capillary axis until equilibrium is achieved. The interfacial tensions are accurate to within (10-4 mN · m-1. When equilibrium is reached and the length (x) of droplet exceeds 4 times the diameter (y) of the droplet, the approaches of Princen et al.11 and also Cayias et al.12 may be used to calculate the interfacial tension.

y ( n) ∆F γ ) 1.234 × 10 3

6

P2

(1)

where ∆F (g/cm3) is the density difference between the microemulsion and water or oil, x and y (cm) are the length and diameter of the droplet, respectively, n is the refractive index of the high-density phase, and P is the reciprocal of the rotational rate (also called the angular velocity). If x/y < 4, then the correction factor, f(x/y), should be different. For each measurement, the microemulsion systems were allowed to reach equilibrium at constant temperature (25.0 ( 0.1 °C), and the radius of the droplet was determined only after the (11) Princen, H. M.; Zia, I. Y. Z.; Mason, S. G. J. Colloid Interface Sci. 1967, 23, 99. (12) Cayias, J. L.; Schechter, R. S.; Wade, W. H. In Adsorption at Interfaces; ACS Symposium Series; Mittal, K. L., Ed.; American Chemical Society: Washington, DC, 1975; Vol. 8, pp 234-247.

Ionic Liquid Tunes Microemulsion CurVature

Langmuir, Vol. 25, No. 4, 2009 2057

shape of the droplet remained constant for 20 min. For the middlephase microemulsion systems, x/y > 4 was confirmed (γME-O and γME-W represent the interfacial tensions between the middle-phase microemulsion and excess oil and excess water). To obtain valid interfacial tensions, attention has to be paid to the following: (1) washing the capillary three times with the high-densityphase solution before measurements, (2) using a uniform 4-mmdiameter capillary, and (3) aging some measurements for at least 0.5 h. Small-Angle X-ray Scattering (SAXS). SAXS experiments were performed in flat cells of 0.1 and 0.2 mm thicknesses with Kapton windows on a Huxley-Holmes high-flux camera.13 The X-ray source is a copper rotating anode operating at 15 kW, providing KR1 radiation (wavelength λ ) 1.54 Å). Spectra were recorded with a 2D gas detector of 0.3 m diameter, giving an effective q range of 0.02-0.4 Å-1 (q ) (4π/λ)sin θ, where θ is the scattering angle). Data correction, radial averaging, and absolute scaling were performed using routine procedures.14 To access a wider q range, additional spectra were obtained with a Guinier-Mering setup using a 2D image plate detector.15 The X-ray source was a molybdenum rotating anode operating at 3 keV. The q range covered by this instrument is 0.06 to 2 Å-1. Data correction, radial averaging, and absolute scaling were performed using routine procedures. The overlap of data obtained from both setups is good. Small-Angle Neutron Scattering (SANS). Experiments were performed at the Laboratoire Le´on Brillouin (LLB), Saclay, France, using the PAXE spectrometer on the Orphe´e reactor. The overall q range (0.014 < q (Å-1) < 0.554) was accessed by varying the sample-to-detector distance from 1 to 5 m. The acquisition time was 1 h for each sample, and the typical statistical error was on the order of 1%. Neutron detection was achieved with a built-in 2D sensitive detector composed of 64 cells × 64 cells. Absolute intensities were obtained using standard data treatment. SAXS and SANS Modeling. For SANS, ∆F is taken with good confidence to be the contrast between deuterated water and hydrogenated heptane, ∆F ) 6.94 × 1010 cm-2. For SAXS, the exact composition of the different phases (i.e., scattering particles and solvent-containing part) would be needed to determine ∆F precisely.

P(q) )

[

3[sin(qR) - qR cos(qR)] (qR)3

]

2

(2)

P(q) is the form factor, depending on the geometrical form of the scattering object. The SAXS spectrum of the middle-phase microemulsions was fitted by the Teubner-Strey expression,16,17 which can be used to describe small-angle scattering from flexible microemulsions (i.e., when the entropy of mixing overcomes topological and curvature constraints). The Teubner-Strey parametric expression for the scattering of microemulsions comprises three fitting parameters to describe the broad peak as well as the q-4 decay observed at the wave vector.18 The scattering function can be approximated using three parameters as

I(q) )

1 + Iin a2 + c1q2 + c2q4

(3)

with a2 > 0, c1 < 0, and c2 > 0 and the stability condition 4a2c2 - c12 > 0. From the fit, two length scales characterizing microemulsions can be obtained: ξ and d, where d represents the domain size and ξ represents a correlation length. The two length scales can be calculated according to eqs 4 and 5. (13) Zemb, Th.; Tache´, O.; Ne´, F.; Spalla, O. J. Appl. Crystallogr. 2003, 36, 800. (14) Ne´, F.; Gabriel, A.; Kocsis, M.; Zemb, Th. J. Appl. Crystallogr. 1997, 30, 306. (15) Ne´, F.; Gazeau, D.; Taboury, J.; Zemb, Th. J. Appl. Crystallogr. 1993, 26, 763. (16) Teubner, M.; Strey, R. J. Chem. Phys. 1987, 87, 3195. (17) Chen, S. H.; Chang, S. L.; Strey, R.; Samseth, J.; Mortensen, K. J. Phys. Chem. 1991, 95, 7427. (18) Zemb, Th.; Barnes, I. S.; Derian, P. J.; Ninham, B. W. Prog. Colloid Polym. Sci. 1990, 81, 20.

ξ)

[( ) ] [( ) ] 1 a2 2 c2

d ) 2π

1⁄2

1 a2 2 c2

+

1⁄2

1 c1 4 c2

-

-1⁄2

1 c1 4 c2

(4) -1⁄2

(5)

Results and Discussion Phase Behavior Investigations. Previous work studied a model water-in-heptane ionic microemulsion system stabilized by a negatively charged stiff catanionic surfactant mixture19 comprising an 8:1 mole ratio of single-chained anionic SDS and doublechained cationic DODMAC. To generate stable systems, it was necessary to add a cosurfactant, n-butanol (11 to 12 wt %). Close to room temperature, this formulation was shown to generate large three-phase Winsor III regions, with Ro essentially being zero.19 The interfacial curvature of this model microemulsion can be adjusted by adding NaCl, and over the salinity range of 0.25-0.77 mol · L-1 a Winsor III phase is observed.19 If now instead of an inorganic electrolyte the water-soluble IL [bmim][BF4] is added, this can also be used to tune spontaneous curvature Ro. The optimal formulations for obtaining the extensive stabilization of Winsor III middle-phase microemulsions with these mixtures were obtained from three-level L9 (34) OD tables for orthogonal experiments (Experimental Section).19 This resulted in compositions of cDODMAC/cSDS ) 1:4-1:5 and cn-butanol(%) ) 11.0-12.0. The phase behavior observed on addition of IL is shown in Figure 1a. In the absence of added IL, the system is biphasic, consistent with a Winsor I system of an oil-in-water ME in equilibrium with excess oil. With added IL, the Winsor I system persists up to cIL ) 1.776 wt %. Above this critical concentration, a three-phase Winsor III system appears, which consists of a middle phase microemulsion in equilibrium with lower aqueous and upper oily phases. The Winsor III system is stable up to a [bmim][BF4] concentration of cIL ) 2.38 wt %, and then for higher IL concentrations a Winsor II system appears, comprising an upper-phase water-in-oil ME equilibrated with a denser aqueous phase. Owing to the chemical nature of the IL, it might be expected that [bmim][BF4] partitions into the interfacial film,20 effectively acting as a cosurfactant as well as a solute in the aqueous phase but being largely absent from the organic phases. This kind of system can be contrasted with the commonly observed Winsor I f III f II transition driven by the variation of electrical double layer around microemulsion drops compressed by the increasing concentration of NaCl.21 To investigate the phase behavior of ME systems in more detail, the effect varying alcohol content, total surfactant concentration, and temperature on phase behavior were followed (Figure 1b-d). It can be seen that the phase transition follows the sequence Winsor I f III f II. In Figure 1b, when the concentration of n-butanol is lower than 12.96 wt %, the ME system is Winsor I. From 12.96 to 13.77 wt %, the ME systems are Winsor III with middle-phase microemulsions coexisting with excess water and oil phases. When the butanol concentration is higher than 13.77 wt %, a Winsor II system forms. On changing the total surfactant concentration at fixed weight ratio cDODMAC/ cSDS ) 1:4, the phase sequence was the same as seen with n-butanol, but when the total surfactant was between 0.8 and 2.8 wt %, the volume of the middle-phase microemulsions grew. (19) Hao, J.; Wang, H.; Shi, S.; Lu, R.; Wang, T.; Li, G.; Sun, H. Sci. China, Ser. B 1997, 40, 225. (20) Petrache, H. I.; Zemb, Th.; Belloni, L.; Parsegian, V. A. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 7982. (21) Chan, K. S.; Shah, D. O. J. Dispersion Sci. Technol. 1980, 1, 55.

2058 Langmuir, Vol. 25, No. 4, 2009

Liu et al.

Figure 3. SAXS and SANS (with D2O to provide contrast) from middlephase Winsor III systems for an oil volume fraction of φO ) 0.64. Solid and open circles represent the SANS spectra before and after background subtraction, respectively. On the left side, the open squares represent the SAXS spectra, and the overlapping dark line is the corresponding Teubner--Strey fit, giving parameters of ξ ) 60 Å and d ) 360 Å.

Figure 2. Ultralow interfacial tensions (γME-O and γME-W) between middlephase microemulsions and excess oil and water phases induced by adding ionic liquid: (a) [bmim][BF4], (b) n-butanol, and (c) total surfactants. T ) 25.0 ( 0.1 °C.

The effect of temperature on phase behavior is shown in Figure 1d. In this case, the influence of temperature on the phase behavior was not as great as observed with nonionic surfactants: Winsor II systems were observed within the temperature range studied, indicating that the Winsor III system might possibly be stable at higher temperatures. Ultralow Interfacial Tensions. Ultralow interfacial tensions between the middle-phase microemulsions and excess oil (γME-O) or water (γME-W) were determined by the spinning-drop technique at 25.0 ( 0.1 °C. Figure 2a indicates that γME-O decreases whereas γME-W increases with increasing cIL. The cross-over point in Figure 2a, where γME-O ) γME-W defines the optimal IL concentration cIL* for the stabilization of middle-phase microemulsions. (For inorganic salt MEs, this corresponds to the optimal salinity). At cIL*, the ratio of partition coefficients of surfactant in oil and water phases is very close to 1.16 For these two systems, the following optimal conditions are achieved: cDODMAC/cSDS ) 1:4, cDODMAC+SDS ) 2.0 wt %, and cn-butanol ) 11.0 wt %. From Figure 2a, cIL* is found to be ∼2.09 wt %, for which the optimal interfacial tension γ* ) γME-O ) γME-W ) 5.2 × 10-3 mN · m-1. We took the results from Figures 1a and 2a together with the volume of the middle-phase microemulsion (VIL* ) 5.80 - 4.10 ) 1.70 mL) at cIL*. These parameters compare favorably with the celebrated AOT-water-heptane system for which cNaCl* ≈ 2.8 wt % and γ* ≈ 5.2 × 10-4 mN · m-1.22 Interfacial tensions were also determined by changing the concentrations of both n-butanol and total surfactant. Figures 2b and c indicate that

γME-O decreases and γME-W increases with increasing cn-butanol or cDODMAC+SDS. The control parameters cn-butanol* and cDODMAC+SDS* can be calculated from the cross point of the interfacial tension lines. The optimal interfacial tensions γ* are 9.2 × 10-3 and 3.6 × 10-3 mN · m-1, respectively. Structural Investigations by SAXS and SANS on the Middle-Phase MEs. On the basis of the mass density of the optimal Winsor III ME (0.881 g · mL-1), it can be estimated that the oil volume fraction is φO ≈ 0.64, which might indicate a water-in-oil structure. The electrical conductivity of the same system was 6 µS · cm-1, representing ∼60% of the conductivity of the polar phase. Because these results give no immediate indication of the microstructrue, X-ray and neutron scattering experiments were performed to shed light on the nanodomain structure, as shown in Figure 3. The SANS spectrum is given in Figure 3. The specific area Σ (cm2/cm3) is obtained from a Porod treatment. The background noise (BG) due to incoherent scattering that has to be subtracted from the experimental intensity is 0.665 (Figure 3). Σ is found to be around 130 m2 · g-1. Combined with a typical domain size of 80 Å and a value for a typical size of 360 Å, a dimensionless product of Σd of about 24 is nearly twice the highest value reported so far. Domains in the form of connected cylinders can give values of the dimensionless product peak multiplied by a specific area of up to 12, whereas standard flexible microemulsions without topology or curvature constraints give the values of up to only 6.23 The scattering spectrum does not show a broad peak as usually observed with bicontinuous microemulsions. No oscillation can be seen at high q, demonstrating that the sample is not in a globular droplet regime. The common model in the case of concentrated (oil volume fraction of φO ) 0.64) nonspherical microemulsions is the Talmon-Prager model based on space tessellation. This model of random domains that initially tessellate space was proposed by Talmon and Prager and then modified by neglecting the bending energy at edges. It is hence adapted for describing situations with low bending energy and was then (22) Aveyard, R.; Binks, B. P.; Clark, S.; Mead, J. J. Chem. Soc., Faraday Trans. 1 1986, 82, 125. (23) Zemb, Th. C. R. Chim. 2008, doi:10.1016/j.crci.2008.10.008.

Ionic Liquid Tunes Microemulsion CurVature

refined by Milner and Welberry.24-26 In the case of microemulsions made of droplets, connected lamellae, or connected cylinders, the Teubner-Strey formula always converges to give a typical size. In the case of flexible films, the correlation length obtained from fitting is usually one-third of the typical domain size.27 The Teubner-Strey empirical formula allows us to derive a typical size of oil or water domains, even in the absence of a well-defined peak in the scattering. It enables the evaluation of the typical size of domains. The best fit to Teubner-Strey expressions was shown in Figure 3, which is obtained for parameters d ) 360 Å and ξ ) 60 Å. The correlation length is surprisingly low compared to d, and it is normally found that the ratio of d to ξ ranges from 1 to 3. This can be interpreted by considering that the ionic liquid pair acts as a strong cosurfactant, which reduces the rigidity of the film. Derivation of local microstructure in the form of globules and connected cylinders of locally lamellar structures would require a dilution plot (i.e., variation of the apparent size of domains obtained while varying the relative volume fractions of water and oil at constant interfacial area). The same data in Figure 3 could be fitted by a standard model of polydisperse spheres. An average droplet size of 80 Å would be obtained, indicating that there is first-order balance between attractive and repulsive terms (i.e., the structure factor S(q) is close to 1 for any q value). In any case, the average curvature of the microemulsion is directed toward the inner material and is high. If the microstructure consists of droplets, then the packing parameter would be 0.81 (i.e., an average curvature radius of 5 times the surfactant film thickness). If the microstructure consists of random Voroni cells (i.e., Talmon-Prager-like random microemulsions), then the average curvature would be half of that value. In any case, adding ionic liquid from a balanced microemulsion has produced a curved interface with the curvature directed toward water: ionic liquids screen the lateral interactions in the film very efficiently. As compared to simple salts for (24) Welberry, T. R.; Zemb, Th J. Colloid Interface Sci. 1988, 123, 413. (25) Milner, S. T.; Safran, S. A.; Andelman, D.; Cates, M. E.; Roux, D. J. Phys. (Paris) 1988, 49, 1065. (26) Zemb, Th.; Welberry, T. R. Colloid Polym. Sci. 1993, 271, 124. (27) Zemb, Th. Scattering of Micelles and Microemulsions; In X-ray, Neutron and Light Scattering; Lindner P., Zemb, Th., Eds.; Elsevier: Amsterdam, 2002.

Langmuir, Vol. 25, No. 4, 2009 2059

microemulsion formation,19,28 the curvature of microemulsions is of the same type; over the salinity range of 0.25 to 0.77 mol · L-1 a Winsor III phase is observed. Thus, ionic liquids are still considered to be salts and can be commonly used as electrolytes to tune the interfacial tension and phase behavior of microemulsions with ionic surfactant.

Conclusions The concentration of a model ionic liquid [bmim][BF4] can be used as a control parameter for film curvature tuning in interfacially stiff microemulsions stabilized by cationic-anionic surfactants. The IL acts in a similar fashion to electrolyte in standard systems, promoting ultralow interfacial tensions (,10-2 mN · m-1) and evolution of the phase behavior from Winsor I f Winsor III f Winsor II MEs as a function of concentration cIL. The lower amount of IL needed compared to normal NaCl makes ILs the most effective additives for curvature adjustment in ionic microemulsions, and ionic liquids including [bmim][BF4] used in the present study could be employed to tune the microemulsion curvature. This enhanced efficiency is most likely a direct consequence of the unique structure of ILs, combining two ions of large volume, which can perform multiple functions at the interface: structure breaking as well as binding and penetration into the stabilizing film to act as cosurfactants. These results point to new possibilities for ionic liquids as functional components in emulsions and microemulsions. Interestingly, ILs such as [bmim][BF4] can act in two ways, simultaneously fulfilling the roles of polar phase and electrolyte. As such, IL microemulsions or emulsions represent flexible systems, offering approaches for supplying structured fluids through a reduction in the number of components necessary to achieve the optimization of interfacial curvature. Acknowledgment. This work was supported by the NSFC (grant no. 20625307) and National Basic Research Program of China (973 program, 2009CB930103). Supporting Information Available: Formulation for the optimal conditions of middle-phase microemulsions obtained by an orthogonal design. This material is available free of charge via the Internet at http://pubs.acs.org. LA8036378 (28) Hellweg, Th. Curr. Opin. Colloid Interface Sci. 2002, 7, 50.