Effect of Lipophilic Tail Architecture and Solvent Engineering on the

Aug 31, 2010 - National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba Ibaraki ... International Young Researchers Empowerment Center, Sh...
0 downloads 0 Views 4MB Size
Download PDF
12008

J. Phys. Chem. B 2010, 114, 12008–12017

Effect of Lipophilic Tail Architecture and Solvent Engineering on the Structure of Trehalose-Based Nonionic Surfactant Reverse Micelles Lok Kumar Shrestha,*,†,‡ Takaaki Sato,§ Martin Dulle,⊥ Otto Glatter,⊥ and Kenji Aramaki‡ International Center for Young Scientists (ICYS), WPI Center for Materials Nanoarchitectonics (MANA), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba Ibaraki, 305-0044, Japan, Graduate School of EnVironment and Information Sciences, Yokohama National UniVersity, Tokiwadai 79-7, Hodogaya-ku, Yokohama 240-8501, Japan, International Young Researchers Empowerment Center, Shinshu UniVersity, 3-15-1 Tokida, Ueda 386-8567, Japan, and Department of Chemistry, UniVersity of Graz, Heinrichstrasse 28, 8010 Graz, Austria ReceiVed: April 6, 2010; ReVised Manuscript ReceiVed: July 30, 2010

We use small-angle X-ray scattering and dynamic light scattering to investigate the structural and dynamical properties of trehalose polyisostearate, abbreviated as TQ-n (n ) 3, 5, and 7), in different organic solvents, where n represents the number of isosterate chains per surfactant molecule. TQ-n spontaneously assembles into reverse micelles without addition of water at 25 °C. We found that for TQ-5 and TQ-7, steric hindrance of the lipophilic surfactant tail causes significant reduction of the aggregation number, whose scheme is clearly distinguished from the modification of the critical packing parameter. Increasing the hydrocarbon chain length of oils from octane to hexadecane favors one-dimensional micellar growth, leading to the formation of rodlike micelles due to different penetration tendencies of oils into the lipophilic shell of the surfactant. Subtle differences in solvent polarity also plays a crucial role in the micellar size, which is decreased when liquid paraffin is replaced with squalene. A further decrease is attained in more polar mixed triglyceride oils. A rising temperature also results in the same direction. The extrapolated structure factor to the zero scattering vector, S(q f 0), for the TQ-3/decane systems almost exactly follows that predicted for hard spheres, demonstrating that osmotic compressibility of the system is well explained if accounting for the excluded volume. However, we found that the effective diffusion coefficient decreases with surfactant concentration, which is an opposite trend to what is expected for hard spheres. This apparent contradiction is likely to be due to the occurrence of transient interdigitation between the lipophilic tails of neighboring reverse micelles at higher concentration. Our data highlight the relevance of the concept of “tunable reverse micellar geometry” in the novel trehalose-based nonionic surfactant binary mixtures, in which lipophilic tail architecture, solvent engineering, concentration, and temperature act as intrinsic parameters for the structure control of the reverse micelles. 1. Introduction Reverse micelles, comprising a hydrophilic polar core and a lipophilic nonpolar shell, have a structure opposite to those of conventional micelles in aqueous media and are often called inverse or inverted micelles. Studies on reverse micelles have attracted significant interest over the years because of their wide range of applications. They not only stabilize species that are essentially insoluble in nonpolar solvents but also can be utilized as a size-controlling microreactor for different aqueous chemical reactions1,2 as well as a template for nanomaterial synthesis.3–8 It has been found that the structure of the produced nanomaterials is closely linked to that of the template micelles.9 Previous studies have provided evidence for the formation of reverse micelles in ternary mixtures of surfactant/water/oil mostly in oil-rich regions, in which water was considered as an essential component, and also in aqueous systems of lipophilic surfactant in the surfactant-rich regions.10–19 For instance, water/ aerosol OT (AOT)/oil and water/lecithin/oil, are well-known * Corresponding author. Phone: +81-29-851-3354 (ext. 8903). Fax: +8129-860-4703. E-mail: [email protected]. † NIMS. ‡ Yokohama National University. § Shinshu University. ⊥ University of Graz.

and frequently studied systems.20–25 However, the formation of reverse micelles without addition of water is still a matter of discussion. Indeed, there exist only few reports in the literature so far that describe the formation of reverse micelles in surfactant/oil binary systems.26–30 A hydrophilic moiety of the conventional poly(oxyethylene) type nonionic surfactants tends to dissolve in many organic solvents without being able to form reverse micelles unless a certain amount of water is incorporated into the system. It has been found that in contrast, glycerolbased nonionic surfactants, generally more solvophobic than the conventional ethylene oxide (EO)-based nonionic surfactants, form self-assembled structures in nonpolar organic solvents without water addition.31,32 Recent phase behavior studies of mono- and diglycerol fatty acid esters in a variety of organic solvents have shown the formation of reverse micelles at elevated temperatures above melting points of solid and liquid crystal phases.29,33,34,35 Monoglycerol fatty acid esters do not form any self-assembled structures at room temperature, whereas diglycerol fatty acid esters efficiently do. They form an R-solid phase and a lamellar liquid crystal (LR) phase in the surfactant axis at higher surfactant concentration regions, depending on temperature, and also a dispersion of reverse vesicles in a dilute region at room

10.1021/jp103080b  2010 American Chemical Society Published on Web 08/31/2010

Structural, Dynamical Properties of TQ-n SCHEME 1: Schematic Molecular Structures of Trehalose Tri-Isostearate (TQ-3) and Selected Oils, Squalane, Squalene, and TIO

temperature. These surfactants also form reverse micelles with a wide range of morphologies, such as spheres, rods, and prolate ellipsoids, but only at high temperatures, which limits their potential for practical applications. Surfactants having potential for forming reverse micelles at room temperature have long been desired. Recently, we have formulated reverse micelles with branched-chain diglycerol polyisostearate nonionic surfactant/ oil systems and observed different types of reverse micellar structures controlled by temperature, surfactant concentration, the nature of the surfactant, solvent, and other outer conditions.36 In this contribution, we have studied the self-assembly of the novel trehalose polyisostearate (designated as TQ-n, n ) 3, 5, and 7) nonionic surfactants in a variety of organic solvents, concentrating on the effect of the lipophilic tail architecture of the surfactant and solvent engineering on the structure of the reverse micelles. The molecular structure of such trehalose-based nonionic surfactants would meet an increasing demand for environmentally friendly, biocompatible nonionic surfactants; in particular, from food or cosmetics oriented societies. We used small-angle X-ray scattering (SAXS) and dynamic light scattering (DLS) for structural characterization. Linear-chain hydrocarbons having different chain lengths, mixed-chain hydrocarbons, branched saturated and unsaturated hydrocarbon oil, mixed triglycerides, and ester oils were taken as solvents for systematic investigations, in which the SAXS data were evaluated by the generalized indirect Fourier transformation (GIFT) method and complemented by geometrical model fitting. 2. Experimental Section 2.1. Materials. The novel nonionic surfactants, trehalose polyisostearates with purity >99%, hereafter designated as TQ-n (n ) 3, 5, and 7), where n represents the number of isostearate chains per surfactant molecule, were a generous gift from the Nisshin OilliO Group, Ltd., Yokohama, Japan. The surfactants were used without further purification. The nonpolar organic solvents, n-octane, n-decane, n-dodecane, n-tetradecane, and n-hexadecane, were purchased from Tokyo Chemical Industry, Tokyo, Japan. These oils were 99.5% pure. Liquid paraffin, a blend of hydrocarbons with an average carbon number of 24; squalane; and squalene were also the product of Tokyo Chemical Industry. Olive oil was purchased from Wako Chemical Industry, Tokyo, Japan, and glycerol tris(2-ethylhexanoic) ester (TIO) was obtained from Nisshin OilliO Group, Ltd., Yokohama, Japan. All oils were used as they were supplied. The schematic molecular structures of TQ-3, squalane, squalene, and TIO are given in Scheme 1. 2.2. Methods. 2.2.1. Identification of Equilibrium Phases at 25 °C. Equilibrium phases of binary mixtures of trehalose polyisostearates, TQ-3, TQ-5, and TQ-7, with a series of

J. Phys. Chem. B, Vol. 114, No. 37, 2010 12009 hydrocarbon oils were identified in a dilute region by visual inspection through a crossed-polarizer at 25 °C. The TQ-n surfactant/oil binary mixtures (n ) 3, 5, and 7) with surfactant weight fraction Ws between 5% and 35% were prepared in n-octane, n-decane, n-dodecane, n-tetradecane, and n-hexadecane in a 5 mL clean and dry glass ampule with a screw cap. The samples were mixed using a dry thermobath and a vortex mixer with repeated centrifugation to achieve homogeneity. The samples were kept in a temperature-controlled water bath at 25 °C for 2 h to observe the equilibrium phases. The accuracy of the temperature control was better than (0.5 °C. Isotropic solutions were observed for all the systems. 2.2.2. Small-Angle X-ray Scattering. SAXS experiments were performed on a series of the TQ-n surfactant/oil binary mixtures (n ) 3, 5, and 7) to study the influence of the lipophilic tail architecture, solvent properties, surfactant concentration, and temperature on the reverse micellar structure. The samples were placed in the water bath at 25 °C for 2 h before SAXS measurements. A SAXSess camera (Anton Paar, Austria) attached to a PW3830 sealed-tube anode X-ray generator (PANalytical, Netherlands) was operated at 40 kV and 50 mA. A Go¨bel mirror and a block collimator provided a focused monochromatic X-ray beam of Cu KR radiation (λ ) 0.1542 nm) with a well-defined line shape. A thermostatted sample holder unit (TCS 120, Anton Paar) controlled the sample temperature with an accuracy of 0.1 °C. The 2-D scattering pattern was recorded by an image plate (IP) detector (Cyclone, Perkin-Elmer, USA) and integrated into one-dimensional scattering intensities I(q) as a function of the magnitude of the scattering vector q ) (4π/λ) sin(θ/2) using SAXSQuant software (Anton Paar), where θ is the total scattering angle. All the measured intensities were semiautomatically calibrated for transmission by normalizing a zero-q attenuated primary intensity to unity. All I(q) data were corrected for the background scattering from the capillary and the solvents, and the absolute scale calibration was made using water as a secondary standard.37 In this study, we used the generalized indirect Fourier transformation (GIFT) method38–41 to obtain a real-space structure function of the reverse micellar aggregates, the socalled pair-distance distribution function, p(r). This procedure relies on a basic equation of one-component globular particle systems, I(q) ) n P(q) S(q), and its extension to polydisperse systems, where P(q) is the averaged form factor, S(q) is the averaged static structure factor, and n is the number of particles in unit volume. Since theoretically, P(q) is the Fourier transformation of p(r) as

P(q) ) 4π

∫0∞ p(r) sinqrqr dr

(1)

one needs to calculate the inverse Fourier transformation of an experimental P(q) to deduce p(r) from the scattering experiments. To suppress the influence of interparticle interference scattering on the evaluation of p(r) that generally leads to oscillations and to a highly underestimated maximum size of the scattering object, an interaction potential model for S(q) was to be involved in the analysis, where we chose the averaged structure factor model42,43 of hard sphere and Percus-Yevick closure relation to solve the Ornstein-Zernike equation. The detailed theoretical description of the method has been given elsewhere.44–46 When the axial length of an elongated scattering particle is sufficiently long, say, at least three times longer than the cross-

12010

J. Phys. Chem. B, Vol. 114, No. 37, 2010

Shrestha et al.

sectional diameter, we can apply a model-free cross section analysis. Theoretically, the radial electron density profile, ∆Fc(r), is related to the cross section pair-distance distribution function, pc(r), as47

pc(r) ) r∆F˜ c2(r)

(2)

Using a counterpart technique of indirect Fourier transformation (IFT), pc(r) can be calculated directly from the experimental scattering intensity, I(q), based on

I(q)q ) πL Ic(q) ) 2π2L

∫0∞ pc(r) J0(qr) dr

(3)

where J0(qr) is the zeroth-order Bessel function. The yielded pc(r) can then be used to calculate ∆Fc(r) by the deconvolution procedure.48,49 2.2.3. Dynamic Light Scattering. DLS experiments were performed on a laboratory-built goniometer equipped with single-mode fiber optics (OZ from GMP, Zu¨rich, Switzerland) and an ALV single-photon detector (ALV-Laser Vertriebsgesellschaft, Langen, Germany) for detection of a time-dependent scattered intensity. The light source was a Verdi V5 diode pumped solid-state laser (Coherent) with a wavelength of 532 nm and a maximum output of 5 W. The data acquisition was performed with an ALV 5000 multiple τ-digital correlator. The ALV-5000/E software package was used to calculate the intensity autocorrelation functions g2(τ). A number of 30 s DLS measurements were carried out at a scattering angle of 90° for individual samples. Tens of the resulting g2(τ) functions were averaged and converted into the mass-weighted distribution of the hydrodynamic radii D(RH) by an inverse Laplace transformation utilizing the optimized regularization technique (ORT).50 2.2.4. Densimetry. A high-precision DSA5000 densimeter (Anton Paar, Austria) was used to measure the densities of the oil and reverse micellar solution. The DSA5000 instrument is based on the conventional mechanical oscillator method, which measures the natural resonant frequency of a U-shaped glass tube, filled with a 1 mL sample. The highly tuned temperature control of the apparatus enables an accuracy of 10 mK in an absolute value. 3. Results and Discussions 3.1. Isothermal Phases in the Dilute Regions. Equilibrium phases of TQ-3, TQ-5, and TQ-7 in a variety of organic solvents, such as n-octane to n-hexadecane, liquid paraffin, squalane, squalene, olive oil, and TIO, were identified by visual inspection at 25 °C in 5% e Ws e 35%. We confirmed that the TQ-3, TQ-5, and TQ-7 surfactants form isotropic solutions in the aforementioned oils under ambient conditions. The detailed phase behavior (temperature-composition diagrams) study is to be done and will be reported elsewhere. 3.2. Effect of Lipophilic Tail Architecture on the Reverse Micellar Structure. Figure 1 shows the scattering intensities, I(q), and the pair-distance distribution functions, p(r), calculated with the GIFT method for the 5 wt % TQ-3, TQ-5, and TQ-7 in decane at 25 °C. The data confirm the presence of micellar aggregates in the systems. We point out that in the nonaqueous systems, the contrast (the electron density difference between solvent oil and the lipophilic part of the surfactant) is negligibly small so that SAXS can detect only the hydrophilic core of the reverse micelles for the form factor, P(q). Thereby, p(r) presented in Figure 1 should be considered as a measure of the

Figure 1. (a) X-ray scattered intensities, I(q), of the 5 wt % TQ-3, TQ-5, and TQ-7 in decane at 25 °C in absolute scale; (b) the pair-distance distribution functions, p(r); (c) model fitting (solid lines) to the experimental intensities (symbols) with GIFT fit (dashed lines); and (d) p(r) curves obtained from the GIFT method (symbols) and model fittings (solid lines). The solid and broken lines in panel a represent the GIFT fit and the calculated form factor for n particles existing in unit volume, nP(q), respectively. The arrows in panels b and d indicate the maximum diameter of the micellar core, Dmax.

micellar core structure. On one hand, the structure factor S(q) should include all intermicellar interactions, without being affected by the contrast problems. The scattering behavior strongly differs, depending on the lipophilic tail architecture of the surfactant. As can be seen in Figure 1a, with an increase in the number of isostearate chains per surfactant molecule n from TQ-3 to TQ-5 or TQ-7, the forward-scattering intensities are greatly suppressed, which is accompanied by a shift of the scattering curves to higher q values, seen best at q > 2 nm-1. This observation is an indication of micellar shrinkage in the two directions of long and short axes of the micelle. Correspondingly, p(r) shown in Figure 1b gives a clear picture on the simultaneous decrease in the maximum diameter and the cross section radius; the maximum size of the micelles, Dmax, indicated by arrows, is significantly decreased upon changing surfactants from TQ-3 to TQ-5 or TQ-7, while the position of maximum in p(r) shifts toward lower-r side. The extended length of trehalose group is estimated to be ∼1.3 nm. The micellar core cross section diameter for TQ-3, which is approximately twice this value when judged from the feature of p(r), is naturally understood as the formation of short-rod reverse micelles. We found, however, that the micellar core cross section diameters for TQ-5 and TQ-7 are apparently smaller than twice the molecular size of trehalose. We point out that this trend cannot simply be explained, for instance, as due to short-rod to sphere transition, which is, indeed, beyond an increase in the critical packing parameter, cpp. Significant steric hindrance effect caused by voluminous lipophilic tail of TQ-5 and TQ-7 drastically decreases the aggregation number, producing a smaller micellar core. We postulate that the aggregates found in the TQ-5 and TQ-7 systems may be classified into “incomplete” reverse micelles. The GIFT analyses of the SAXS data were complemented by model fittings. Optimum fit curves to the experimental I(q) obtained by assuming homogeneous prolate ellipsoid and the resulting p(r) are presented, together with the experimental ones,

Structural, Dynamical Properties of TQ-n

J. Phys. Chem. B, Vol. 114, No. 37, 2010 12011

Figure 2. Comparison of the p(r) functions for different particle geometries: a homogeneous sphere (black), a prolate (red), and a cylinder (blue). Spherical particles give a symmetric, bell-shaped p(r), whereas that of a cylinder is characterized by the extended, linear-like high-r tail. A prolate particle is distinguished from a cylinder in terms of the downward convex curve in the high-r regime. We found that in most cases, the experimental p(r) for the TQ-3 based reverse micelles resembles that of a prolate ellipsoid (see Figures 1, 4, 7, 9, and 11). Alternatively, a distribution of polydisperse spheres could give a similar p(r). But micelles cannot be polydisperse enough due to packing constraints (only when swollen with water).

TABLE 1: Effect of Lipophilic Tail Architecture of Surfactant on the Geometrical Parameters, Short Axis (a), and Long Axis (b) of Prolate Ellipsoid Obtained from the Results of Model Fittings at 25 °C system

short axis (a)/nm

long axis (b)/nm

5 wt % TQ-3/decane 5 wt % TQ-5/decane 5 wt % TQ-7/decane

1.26 0.72 0.80

2.92 2.20 1.95

in Figure 1c and d. We found that among tested geometric models, only a prolate is able to simultaneously explain the shape of the experimental I(q) and the resulting p(r) as obtained by the GIFT/IFT technique. In Figure 2, we present the theoretically calculated p(r) functions for different particle geometries,51 such as sphere, cylinder, and prolate ellipsoid. The comparison of the shapes of the experimental and theoretical functions especially in real space, intuitively demonstrates the appropriate choice of the prolate ellipsoid model for the reverse micellar core of the investigated systems (see Figures 1d and 2). The geometrical parameters obtained from the model fitting, such as the short and long axes of the prolate ellipsoid, a and b, of the micelles, are given in Table 1. The structure factor curves and structure factor parameters effective volume fraction (φ), effective interaction radius (R), and the polydispersity (µ), obtained from the GIFT method, are supplied in the Supporting Information. The absence of the theoretically predicted minima in the high-q part of the experimental I(q) probably arises from polydispersity in size or small electron density fluctuations inside the core. It is well established that if polydisperse spheres are considered, such high-q minima can easily be smeared out with increasing polydispersity. This is because slightly different radii, R, give minima at slightly shifted q positions, and their superposition no longer represents distinct minima (see Figure S1 in the Supporting Information). We have tested polydispersity analysis based on IFT, assuming polydisperse spheres for the 5% TQ-3/decane system and found that, as shown in Figure 3, those having distributed radii of ∼1.0 - 3.0 nm in appearance fit the data well. However, a spherical micellar core comprising the hydrophilic moiety of the trehalose group could not attain a radius far greater than the extended length of ∼1.3 nm, which rules out the polydisperse spheres for the TQ-3/decane system.

We also considered polydisperse prolates and cylinders. A fixed short axis of a ∼ 1.26 nm accounting for the length of the trehalose group and varied long axes of 2.0 e b/nm e 3.0 for prolate ellipsoid result in the appearance of the minima at nearly identical q (∼3.3 and 5.8 nm-1). A fixed cross section radius, r, and varied length, L, for cylinders creates a similar situation. Thus, polydispersity alone could not explain the absence of the high-q minima in the experimental I(q). Thus, it is likely to be due to the small electron density fluctuation inside the micellar core. We will come back to this point in the following section. 3.3. Effect of Solvent Engineering on the Reverse Micellar Structure. 3.3.1. Effect of Hydrocarbon Chain Length in Linear Chain Oils. Figure 4 presents I(q) and p(r) for the 5 wt % TQ-3 binary mixtures with various alkane oils, CmH2m+2, with different chain lengths at 25 °C. For better visibility, the scattering functions are multiplied by factors of 10, 50, 200, and 500, respectively, for decane (m ) 10), dodecane (m ) 12), tetradecane (m ) 14), and hexadecane (m ) 16). The low-q slope increases with increasing hydrocarbon chain length of the solvents. This can be taken as evidence for solventinduced micellar growth. Similar trends were readily observed in glycerol-based nonionic surfactant/oil systems.52 This feature can be seen clearly in the p(r) functions presented in Figure 4b. A slightly asymmetric, bell shaped p(r) of the TQ-3/octane system confirms globular type reverse micelles having a maximum core diameter, Dmax, of ∼4.5 nm. Asymmetry in the shape of the p(r) function becomes more pronounced with an increasing chain length of the hydrocarbon oils, m, while the position of the maximum in p(r) remains essentially unchanged, demonstrating one-dimensional micellar growth, or an increasing population of highly elongated micelles in the polydisperse systems for the higher m. This way, Dmax for the TQ-3/ hexadecane system reaches ∼10 nm. To highlight this structural evolution, the normalized p(r) functions, p(r)/p(rmax), are presented in Figure 4c. The cross section diameters of TQ-3 in various hydrocarbon oils from octane to hexadecane are estimated to be roughly 2.5-2.6 nm from the inflection point of p(r) seen on the higher-r side of the maximum, which is virtually constant for all the systems. Figure 4d compares the experimental intensities (symbols), GIFT fit (dashed lines), and model fitting (solid lines) on the basis of a geometrical model of homogeneous prolate ellipsoids, which confirms the consistency of the results obtained from these two analyses, except a visible difference in the depth of the minimum in the experimental and theoretical I(q) functions, implying again polydispersity or moderate electron density fluctuation inside the core. Figure 5 shows the effects of different electron density fluctuations in the micellar core for the 5 wt % TQ-3/hexadecane system, in which we compare homogeneous and inhomogeneous electron density distribution. An optimum inhomogeneous electron density profile for the TQ-3/hexadecane was obtained as the output of cross-section IFT analysis. Such different fluctuations result in different shapes of the cross section pair distance distribution functions, pc(r), as shown in panel b, and as a consequence, lead to the presence and absence of distinct minima in the high-q regime of I(q), as clearly demonstrated in panel a. This implies that the absence of distinct minima in I(q) observed for a series of the TQ-3/oil systems arise mainly from the small electron density fluctuation inside the core, rather than polidispersity in the length. The geometrical parameters of the micelles obtained from the results of model fitting are supplied in Table 2.

12012

J. Phys. Chem. B, Vol. 114, No. 37, 2010

Shrestha et al.

Figure 3. Geometrical model fitting for the SAXS data of the TQ-3/decane system based on (A) polydisperse spheres (green), (B) prolate ellipsoid (blue), and (C) cylinder (cyan). Polydisperse spheres having distributed radii of ∼1.0-3.0 nm in appearance fit the experimental form factor, P(q)exp without creating any minima. The radii significantly greater than 1.3 nm, however, are unrealistic for spherical micellar cores consisting of the hydrophilic moiety of the trehalose group.

Figure 5. Effects of electron density distribution in the TQ-3 reverse micellar core: (a) the scattering functions, I(q); (b) the cross section pair distance distribution functions, pc(r); and (c) the electron density profiles, ∆F(r), of the 5 wt % TQ-3/hexadecane system. Homogeneous (red) and inhomogeneous (black) electron density fluctuations are compared; all black curves are the output of cross section IFT analysis. The results clearly indicate that a homogeneous distribution causes distinct minima in I(q), whereas an inhomogeneous distribution easily smears them out. Figure 4. (a) The X-ray scattered intensities, I(q), of 5 wt % TQ-3/oil binary systems at 25 °C in absolute scale; (b) the corresponding p(r) curves; (c) normalized p(r) functions, p(r)/p(rmax); and (d) comparison of the experimental intensities (symbols), the model fitting (solid lines), and GIFT fit (dashed lines). The solid and broken lines in panel a represent the GIFT fit and the calculated form factor for n particles existing in unit volume, nP(q), respectively. The arrows in panels b and c indicate the maximum diameter of the micellar core, Dmax.

For more quantitative evaluation of the cross section structure of the micellar core, we applied an extended technique of indirect Fourier transformation (IFT) to the TQ-3 systems with long hydrocarbon oils (m > 12), in which elongated reverse micelles are produced. The cross-sectional pair-distance distribution function, pc(r), can be directly calculated from the experimental scattered intensity, I(q), based on eq 3. The further deconvolution procedure of the pc(r) yields the radial difference electron density distribution profile, ∆Fc(r).48,49 Figure 6 shows the resulting pc(r) and ∆Fc(r) for the TQ-3/dodecane, TQ-3/ tetradecane, and TQ-3/hexadecane systems. The positive values of ∆Fc(r) in the entire r range as shown in Figure 6b confirm the electron-rich hydrophilic reverse micellar core. The calculated cross section diameters from the position of Dc max in pc(r) for the TQ-3/dodecane, TQ-3/tetradecane, and TQ-3/hexadecane systems are almost identical, giving diameters of ∼2.6 nm and close to those roughly estimated from the inflection point of p(r). As naturally expected from pc(r), the maximum core radius, Rcore, ∼ 1.3 nm, is read out from ∆Fc(r). Oil penetration into the palisade layer of surfactant molecules at the water/oil interface is well-known.53,54 The penetration phenomenon modifies the control parameter and generally makes

TABLE 2: Effect of Hydrocarbon Chain Length of Linear Chain Oils on the Geometrical Parameters, Short Axis (a), and Long Axis (b) of Ellipsoidal Prolate Obtained from the Results of Model Fittings at 25 °C system 5 5 5 5 5

wt wt wt wt wt

% % % % %

TQ-3/octane TQ-3/decane TQ-3/dodecane TQ-3/tetradecane TQ-3/hexadecane

short axis (a)/nm

long axis (b)/nm

1.24 1.26 1.28 1.22 1.20

2.91 2.92 3.40 3.98 5.40

spontaneous curvature negative. The observed difference in the aggregate structure in the TQ-3/oil systems depending on the nature of solvents can be explained in terms of their different penetration tendency into the hydrophilic/lipophilic interface of the surfactant molecule so as to avoid energetically unfavorable direct contact between solvent oil and the hydrophilic group of the surfactant. In studies of nonaqueous phase behavior of diglycerol monomyristate, it has been found that the solubility of the surfactant decreases with an increasing molecular weight of oils and with an increasing alkyl chain length of the hydrocarbon oils. Thus, we can anticipate that a short-chain oil, such as octane, has a more pronounced penetration tendency to the hydrophilic/lipophilic interface of the surfactant as compared with long-chain oils. This could be the reason for the observed globular type reverse micelles in the 5 wt % TQ-3/octane system. This is because octane may go close to the interface and makes spontaneous curvature more negative. As the chain

Structural, Dynamical Properties of TQ-n

J. Phys. Chem. B, Vol. 114, No. 37, 2010 12013

Figure 6. (a) The cross-sectional pair-distance distribution function, pc(r), of the 5 wt % TQ-3/dodecane, TQ-3/tetradecane, and TQ-3/ hexadecane systems; and (b) the resulting cross section radial electron density profile, ∆Fc(r), calculated with the deconvolution procedure of pc(r) using program DECON. Solid lines in panel (a) represent the optimum fit curve in the deconvolution analysis.

length of linear-chain hydrocarbon oils increases, the penetration tendency decreases and, hence, aggregates with less negative curvature are formed. The present results well support the results of Kunieda et al.,55 in which they studied the effect of oil on the surfactant molecular curvature in liquid crystals. They found that m-xylene penetrates into the surfactant palisade layer, and hexagonal to lamellar liquid crystal (H1-LR) transition occurs in a hydrophilic C12EO7 system. This implies that the penetration leads to the production of aggregates with negative or less positive curvature. Furthermore, they found that long-chain oils, such as squalane, have a lower degree of penetration as compared with a shortchain hydrocarbon oil decane. Decane induces LR-H2 phase transitions in a lipophilic C12EO3 surfactant system due to the penetration effect. The detailed information about the effect of oil solubilization in surfactant self-assemblies can be found elsewhere.55–57 3.3.2. Effect of Mixed- And Branched-Chain Oils with Higher Polarity. In this section, we describe the effect of solvent polarity on the structure of the trehalose-based reverse micelles. SAXS measurements were performed on 5 wt % TQ-3 in liquid paraffin, squalane, squalene, olive oil, and TIO. The results are presented in Figure 7. Liquid paraffin, which is a blend of hydrocarbons with an average carbon number of 24, has a nonpolar character. Squalane is a branched, saturated hydrocarbon, but the molecular weight of squalane is larger than that of liquid paraffin. Squalene has six double bonds per molecule and, hence, is more polar than the other two. As can be seen from Figure 7a, the low-q forward scattering intensity increases upon changing an oil from squalene to squalane via liquid paraffin, indicating the formation of bigger reverse micelles in squalane. However, since the scattering behavior at higher-q regions is essentially the same, we expect similar internal structures in these oils. In all three investigated oils, the p(r) functions depicted in Figure 7b show the characteristics of prolate ellipsoids. The maximum diameter of the micellar core, Dmax, decreases from ∼7.0 to 4.2 nm when squalane is replaced with squalene, but the micellar cross section structure remains essentially unchanged, judging from the virtually unchanged peak position in the p(r) and the similar low-r behavior for all solvent oils. As shown in Figure 7d, use of more polar oils, such as olive oil and TIO, leads to a further decrease of the micellar size, producing nearly spherical micelles. We assume that the direct contact between the trehalose moiety and oils with lower polarity is energetically less favorable. This leads to the formation of more elongated structures as a result of a balance between minimizing solvent/ micelle contact area. Figure 8 compares the results obtained from GIFT analysis and the model fittings. One can clearly sees the large

Figure 7. (a) SAXS intensities, I(q), of 5 wt % TQ-3/liquid paraffin, TQ-3/squalane, and TQ-3/squalene binary systems at 25 °C in absolute scales; (b) the corresponding p(r) curves; (c) the I(q) of 5 wt % TQ3/olive oil and TQ-3/TIO; and (d) the corresponding p(r) curves. The solid and broken lines in panels a and c represent the GIFT fit and the calculated form factor for n particles existing in unit volume, nP(q), respectively. The arrows in panels b and d indicate the maximum diameter of the micellar core, Dmax.

Figure 8. (a) Experimental X-ray scattered intensities (symbols) and the model fitting results of 5 wt % TQ-3/liquid paraffin, TQ-3/squalane, and TQ-3/squalene binary systems at 25 °C in absolute scale; (b) the corresponding p(r) obtained from GIFT (symbols) and based on the geometrical model (solid lines); (c) the experimental I(q) (symbols) and the model fitting results of 5 wt % TQ-3/olive oil and TQ-3/TIO; and (d) the corresponding p(r) from GIFT (symbols) and the geometrical model (solid lines). The solid and broken lines in panels a and c represent the model and GIFT fits. The arrows in panels b and d indicate the maximum diameter of the micellar core, Dmax.

consistency between the I(q) and p(r) functions obtained from GIFT and model fittings. The geometrical parameters of the micelles obtained from the results of model fitting are supplied in Table 3. 3.4. Effect of Surfactant Concentration on the Reverse Micellar Structure and Diffusion Dynamics. To study the concentration effect on the micellar structure, SAXS experiments were carried out on the TQ-3/decane system at 25 °C over a wider concentration range (5 to 20 wt %). The results are shown in Figure 9.

12014

J. Phys. Chem. B, Vol. 114, No. 37, 2010

Shrestha et al.

TABLE 3: Effect of Mixed and Branched with Higher Polarity Oils on the Geometrical Parameters, Short Axis (a), and Long Axis (b) of Prolate Ellipsoid Obtained from the Results of Model Fittings at 25 °C system 5 5 5 5 5

wt wt wt wt wt

% % % % %

TQ-3/liquid paraffin TQ-3/squalane TQ-3/squalene TQ-3/olive oil TQ-3/TIO

short axis (a)/nm

long axis (b)/nm

1.02 1.03 1.03 0.92 0.87

2.95 3.45 2.40 1.75 1.50

With increasing surfactant concentration, the forward intensity is suppressed, as can be seen in Figure 9a, and the micelle-micelle correlation peak at q ∼ 1.2 nm-1 develops, which indicates the growing (net) repulsive intermicellar interactions at higher surfactant concentrations.58,59 According to the semiempirical extension of the PercusYevick (PY) theory derived by Carnahan and Starling,60 the extrapolated structure factor to zero scattering vector, S(q f 0), is given as a function of volume fraction of hard sphere (HS), φHS, as

S(q f 0) )

(1 - φHS)4 1 + 4φHS + 4φHS2 + 4φHS3 + 4φHS4

(4) which provides an accurate analytical expression for the osmotic compressibility of a monodisperse, hard sphere dispersion. From densimetry, density of the solvent decane and specific density of TQ-3 in the binary mixtures with decane are, respectively, 0.7265 and 1.003 g/cm3 at 25 °C. Using these values, we converted the weight concentration of the surfactant, Ws, into the surfactant volume fraction, φs. As for aqueous micellar solutions, one should account for the volume of hydration water for the micellar volume fraction.61,62 For the present nonaqueous reverse micellar solutions, we assume that the surfactant volume

Figure 10. (a) The effective diffusion coefficient, Deff, and (b) the apparent hydrodynamic radius, RHapp, as a function of the volume fraction, φmic, of the reverse micelle for the TQ-3/decane systems at 25 °C and (c) the comparison of the S(q f 0) values as obtained by SAXS experiments (solid circles) and those calculated from an extended PY theory (solid curve) for a dispersion of hard sphere. The dotted lines given in panels a and b represent the concentration dependence of Deff and RHapp theoretically predicted for the HS system.

faction, φs, is virtually identical to micellar volume fraction, φmic. We can then immediately calculate the expected concentration dependence of S(q f 0) by substituting φmic for φHS in eq 4. The comparison between the experimental S(q f 0) obtained by SAXS and the predicted values for the TQ-3/decane systems is shown in Figure 10c. S(q f 0) for the TQ-3/decane systems almost exactly follows that predicted for a hard sphere system in a wide concentration range, which demonstrates that osmotic compressibility of the system is well explained in terms of the excluded volume of the reverse micelles, although they are not exactly hard sphere due to a number of nonideal aspects, for example, to some extent elongated shape, anticipated polydispersity, possible mutual interpenetration, and so forth. We performed DLS measurements on the TQ-3/decane system at 25 °C as a function of concentration to examine the concentration effect on the collective translational diffusion dynamics of the reverse micelles, in which the intermicellar interacting schemes may be manifested. The intensity autocorrelation function, g2(t), measured in the homodyne mode is connected to the normalized field correlation function, g1(t), via the relation

g2(t) ) 1 + β|g1(t)| 2

(5)

where β is the coherence factor. The diffusion constant, D0, can be related to the hydrodynamic radius, RH, of the hard sphere via the Stokes-Einstein relation

D0 )

Figure 9. (a) The I(q) curves of the TQ-3/decane system at different surfactant concentrations: 5, 10, 15, and 20 wt % on absolute scale at 25 °C; (b) the corresponding real-space p(r) functions; (c) the normalized p(r) by surfactant weight fraction, p(r)/Ws; and (d) the static structure factor S(q). The solid and broken lines in panel a represent the GIFT fit and the calculated form factor for n particles existing in a unit volume, nP(q), respectively. The arrows in panels b and c highlight the maximum diameter of the micellar core.

k BT 6πηRH

(6)

where kB is the Boltzmann constant, T is the temperature, and η is the solvent viscosity. Extrapolating the concentrationdependent DLS data to the zero concentration, D0 is determined to be 8.53 × 10-11 m2/s, corresponding to RH ) 2.85 nm. In the hard sphere interaction approximation, the effective diffusion constant at volume fraction φHS, D(φHS)eff, is given by63

Structural, Dynamical Properties of TQ-n

D(φHS)eff ) D0(1 + 1.56φHS)

J. Phys. Chem. B, Vol. 114, No. 37, 2010 12015

(7)

which is equivalent to the expression of the apparent hydrodynamic radius, R(φHS)app

R(φHS)app )

RH 1 + 1.56φHS

(8)

The experimentally obtained effective diffusion coefficient, D(φHS)eff, and apparent hydrodynamic radius, R(φHS)app, are shown in Figure 10a-b. In the systems consisting of hard sphere colloidal particles, D(φHS)eff is an increasing function of concentration, as given by eq 7. This type of behavior was observed for various aqueous micellar solutions.61,62,64 However, we found a trend of translational diffusion dynamics for the TQ-3 reverse micelles opposite to what is generally expected for hard sphere; that is, a decreasing collective diffusion coefficient with increasing surfactant concentration, as observed for ternary sugar ester microemulsion systems.65 As we have already mentioned, the osmotic compressibility of the TQ-3/ decane systems manifested in the S(q f 0) can be well accounted for by the hard sphere approximation. This apparent contradiction for static and dynamic scattering results appears to indicate the occurrence of transient interdigitation between the lipophilic tails of neighboring reverse micelles, which causes weak transient clustering at higher concentration. We found that the S(q) peak position, q*∼ 1.2 nm-1, or the averaged interparticle distance, d*() 2π/q*) ∼ 5.2 nm, are almost independent of the concentration. If we define the 1/2 eff effective micellar core radius, Reff c , as Rc ) (ab) , where a and b are the short and long axes of the micellar core, both the results of GIFT analysis and the model fitting give Reff c ∼ 1.9 nm. The extended length of the isostearate chain, lis, is ∼2 nm. eff We observed the relation 2Reff c < d* < 2(Rc + lis) at all investigated φmic. The revealed d*, which is essentially independent of the concentration and always shorter than the expected value for the direct contact, 2(Reff c + lis), provides evidence for the interdigitation of the lipophilic surfactant tail, which supports the DLS results. We infer that such a transient clustering induces a notable slowdown of the translational diffusion dynamics, but is so weak that the osmotic compressibility of the system is still very close to that expected from the excluded volume of the micelles. 3.5. Temperature Induced Transition in the Reverse Micellar Structure. Figure 11 shows the temperature effects on the reverse micellar structure, in which I(q) and p(r) for the 5 wt % TQ-3/decane system at different temperatures of 25, 50, and 75 °C are presented. The decreasing low-q intensity with increasing temperature from 25 to 75 °C without affecting the scattering behavior in the high-q regions is reflected in the decrease in Dmax in p(r), as shown in Figure 11b. The core size is decreased from ∼6.0 to 4.5 nm upon the increasing temperature by 50 °C. Figure 11c and d compares the results from the model fitting, confirming broad consistency between the GIFT and the modeling results. Discussing the micellar size based on SAXS data, interpretation in particular for elongated structures at higher temperatures might suffer from an artifact caused by different contrast. In general, increasing temperature worsens the contrast of amphiphilic core-shell systems, which may lead to critical underestimation of the micellar size. However, in the nonaqueous systems, the reverse micelles generally possess positive electron density fluctuation in the hydrophilic core, whereas the

Figure 11. Effect of temperature on the reverse micellar structures of 5 wt % TQ-3/decane system as obtained by SAXS: (a) The X-ray scattering intensities, I(q), in absolute scale at different temperatures of 25, 50, and 75 °C; (b) the corresponding real-space p(r) functions calculated with the GIFT method; (c) the experimental intensities (symbols), model fitting (solid lines), and GIFT fit (dashed lines); and (d) comparison of the experimental p(r) derived with GIFT (symbols) and those obtained from model fitting (solid lines). The data were fitted using a model of a homogeneous prolate ellipsoid. The solid and broken lines in panel a represent the GIFT fit and the calculated form factor for n particles in unit volume nP(q), respectively. Arrows in panels b and d highlight the maximum diameter of the micellar cores.

TABLE 4: Effect of Temperature on the Structure Parameters of Reverse Micelles Obtained from the Results of Model Fittings system

short axis (a)/nm

long axis (b)/nm

5 wt % TQ-3/decane-25 °C 5 wt % TQ-3/decane-50 °C 5 wt % TQ-3/decane-75 °C

1.26 1.26 1.26

2.92 2.80 2.60

lipophilic shell has almost negligible contrast in organic solvents. This offers a better situation for the nonaqueous systems. As shown in Figure 8b, we observed decreased micellar size by ∼20% with increasing temperature from 25 to 75 °C. This notable difference cannot be attributed solely to the contrast difference. In nonaqueous systems, increasing temperature enhances the penetration of oil in the surfactant chain, resulting in more negative curvature. Increasing temperature has an effect similar to decreasing the hydrocarbon chain length of solvent alkanes (see section 3.3.1). The present result supports our previous studies on reverse micelles from glycerol-based surfactants, in which the temperature effect on the reverse micellar structure was found to be more efficient.34,66 The structure factor curves and at different temperatures is supplied in the Supporting Information. The geometrical parameters obtained from the results of model fittings are supplied in Table 4. 4. Conclusion Using SAXS and DLS, we have systematically investigated the effect of lipophilic tail architecture, solvent engineering, surfactant concentration, and temperature on the reverse micellar structure of the novel trehalose-based nonionic surfactant, TQ-n (n ) 3, 5, and 7), n being the number of isosterate chains per surfactant molecule. The SAXS data were evaluated with the

12016

J. Phys. Chem. B, Vol. 114, No. 37, 2010

GIFT method, which yielded the real space picture of the micellar structure, and the model fitting complements the results. Our SAXS data confirm the spontaneous formation of the reverse micelles in TQ-n/oil binary systems at ambient temperature without addition of water. We found that steric hindrance arising from the voluminous lipophilic surfactant tail of TQ-5 and TQ-7 significantly reduces the aggregation number in the micelles. This reduction mechanism is classified into a different category from rod-to-sphere transition due to the modulation of cpp. On one hand, one-dimensional micellar growth, that is, sphere-to-rod transition, takes place in the TQ3/oil systems when a longer-hydrocarbon-chain oil such as hexadecane is substituted for octane. This type of structural evolution is considered as due to a restricted penetration tendency for longer-chain oils into the lipophilic shell of the surfactant. We also observed a decrease in the micellar size when nonpolar squalane or liquid paraffin is replaced with only slightly polar oil squalene. A further notable decrease in size occurs in more polar mixed triglyceride oils, such as olive oil and glycerol tris (2-ethylhexanoic) ester (TIO). We suggest that as direct contact of trehalose moiety with solvent oils becomes energetically unfavorable for longer-chain hydrocarbon oils or lower polarity oils, elongated structures are more favored so as to optimize the exposure of the trehalose group to the surroundings. Smaller micelles are also formed at higher temperature. The comparison between the structure factor S(q) analysis of the SAXS data and the DLS results gives important insights into the interacting scheme of the reverse micelles: the experimental S(q f 0) for the TQ-3/decane systems and that predicted for a hard sphere system from an extended PY theory are found to be nearly identical, which demonstrates that osmotic compressibility of the system is well explained only in terms of the excluded volume of the reverse micelles. However, we found a totally opposite trend for the DLS data; that is, the decreasing collective diffusion coefficient with surfactant volume fraction. This apparent contradiction for static and dynamic scattering results is likely to be due to the occurrence of transient interdigitation between the lipophilic tails of neighboring reverse micelles at higher concentration. This view is supported by the revealed averaged interparticle distance, d* ∼ 5.2 nm, which is apparently smaller than the micellar diameter, and essentially independent of the concentration. We think that such a clustering that causes a notable slowdown of the translational diffusion dynamics is not permanent, but rather transient. Therefore, the excluded volume of the micelles still largely determines the osmotic compressibility of the system. We are able to successfully control the geometry of the reverse micelles in the novel trehalose-based nonionic surfactant binary mixtures without addition of polar additives such as water. We have found that lipophilic tail architecture, solvent engineering, concentration, and temperature can be treated as intrinsic parameters for the structure control, showing the relevance of the concept of “tunable reverse micellar geometry”, which is realized by the optimization of an energetically unfavorable exposure of the hydrophilic trehalose group to the solvents. Acknowledgment. L.K.S. thanks Japan Society for the Promotion of Science (JSPS) for the postdoctoral fellowship for foreign researcher. Authors are thankful to The Nisshin OilliO Group, Yokohama for the supply of surfactants. This work is supported in part by Grant-in-Aid for JSPS Fellow (No. 20 08044).

Shrestha et al. Supporting Information Available: The geometrical parameters of the micelles obtained from the GIFT fits and model fittings and the S(q) curves depending on lipophilic tail architecture of the surfactant, solvent nature, and temperature is supplied in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Luisi, P. L. Strab, B. E., Eds. ReVerse Micelles: Biological and Technological releVance of Amphiphilc Structures in Apolar Media; Plenum Press: New York, 1987. (2) Pileni, M. P. Structure and ReactiVity in ReVerse Micelles; Pileni, M. P., Ed.; Elsevier: Amsterdam, 1989; Vol 65. (3) Boutonnet, M.; Kizling, J.; Stenius, P. Colloids Surf. 1982, 5, 209. (4) Lisiecki, I.; Pileni, M. P. J. Am. Chem. Soc. 1993, 115, 3887. (5) Pileni, M. P. Langmuir 1997, 13, 3266. (6) Lo´pez-Quintela, M. A. Curr. Opin. Colloid Interface Sci. 2003, 8, 137. (7) Lo´pez-Quintela, M. A.; Tojo, C.; Blanco, M. C.; Garcı´a Rio, L.; Leis, J. R. Curr. Opin. Colloid Interface Sci. 2004, 9, 264. (8) Cushing, B. L.; Kolesnichenko, V. L.; O’Connor, C. J. Chem. ReV. 2004, 104, 3893. (9) Pileni, M. P. AdV. Colloid Interface Sci. 1993, 46, 139. (10) Kunieda, H.; Shinoda, K. J. Colloid Interface Sci. 1980, 75, 601. (11) Kunieda, H.; Shinoda, K. J. Dispers. Sci. Technol. 1982, 3, 233. (12) Friberg, S. E.; Blute, I.; Kunieda, H.; Stenius, P. Langmuir 1986, 2, 659. (13) Kunieda, H.; Solans, C.; Parra, J. L. Colloids Surf. 1987, 24, 225. (14) Solans, C.; Pons, R.; Davis, H. T.; Evans, D. F.; Nakamura, K.; Kunieda, H. Langmuir 1993, 9, 1479. (15) Uddin, H. Md.; Rodriguez, C.; Watanabe, K.; Lo´pez-Quintela, M. A.; Kato, T.; Furukawa, H.; Harashima, A.; Kunieda, H. Langmuir 2001, 17, 5169. (16) Kunieda, H.; Tanimoto, M.; Shigeta, K.; Rodriguez, C. J. Oleo Sci. 2001, 50, 633. (17) Kaneko, M.; Matsuzawa, K.; Uddin, H. Md.; Lo´pez-Quintela, M. A.; Kunieda, H. J. Phys. Chem. B 2004, 108, 12736. (18) Kunieda, H.; Uddin, Md. H.; Horii, M.; Furukawa, H.; Harashima, A. J. Phys. Chem B 2001, 105, 5419. (19) Kunieda, H.; Shigeta, K.; Ozawa, K.; Suzuki, M. J. Phys. Chem. B 1997, 101, 7952. (20) De, T.; Maitra, A. AdV. Colloid Interface Sci. 1995, 59, 95. (21) Riter, R. E.; Kimmel, J. R.; Undiks, E. P.; Levinger, N. E. J. Phys. Chem. B. 1997, 101, 8292. (22) Cason, J. P.; Roberts, C. B. J. Phys. Chem. B. 2000, 104, 1217. (23) Li, Q.; Li, T.; Wu, J. J. Phys. Chem. B. 2000, 104, 9011. (24) Kanamaru, M.; Einaga, Y. Polymer 2002, 43, 3925. (25) Tung, S.-H.; Huang, Y.-E.; Raghavan, S. R. J. Am. Chem. Soc. 2006, 128, 5751. (26) Forster, S.; Zisenis, M.; Wenz, E.; Antonietti, M. J. Chem. Phys. 1996, 104, 9956. (27) Zhong, X. F.; Varsheny, S. K.; Eisenberg, A. Macromolucules 1992, 25, 7160. (28) Desjardins, A.; van de Ven, T. G. M.; Eisenberg, A. Macromolecules 1992, 25, 2412. (29) Shrestha, L. K.; Masaya, K.; Sato, T.; Acharya, D. P.; Iwanaga, T.; Kunieda, H. Langmuir 2006, 22, 1449. (30) Rodriguez, C.; Uddin, Md. H.; Watanabe, K.; Furukawa, H.; Harashima, A.; Kunieda, H. J. Phys. Chem. B. 2002, 106, 22. (31) Herrington, T. M.; Shali, S. S. J. Am. Oil Chem. Soc. 1988, 65, 1677. (32) Rodriguez-Aberu, C.; Acharya, D. P.; Hinata, S.; Ishitobi, M.; Kunieda, H. J. Colloid Interface Sci. 2003, 262, 500. (33) Shrestha, L. K.; Sato, T.; Acharya, D. P.; Iwanaga, T.; Aramaki, K.; Kunieda, H. J. Phys. Chem. B 2006, 110, 12266. (34) Shrestha, L. K.; Sato, T.; Aramaki, K. Langmuir 2007, 23, 6606. (35) Shrestha, L. K.; Sato, T.; Aramaki, K. J. Phys. Chem. B. 2007, 111, 1664. (36) Shrestha, L. K.; Shrestha, R. G.; Oyama, K.; Matsuzawa, M.; Aramaki, K. J. Phys. Chem. B 2009, 113, 12669. (37) Orthaber, D.; Bergmann, A.; Glatter, O. J. Appl. Crystallogr. 2000, 33, 218. (38) Fritz, G.; Bergmann, A.; Glatter, O. J. Chem. Phys. 2000, 113, 9733. (39) Glatter, O.; Fritz, G.; Lindner, H.; Brunner, P. J.; Mittelbach, R.; Strey, R.; Egelhaaf, S. U. Langmuir 2000, 16, 8692. (40) Brunner-Popela, J.; Mittelbach, R.; Strey, R.; Shubert, K. V.; Kaler, E. W.; Glatter, O. J. Chem. Phys. 1999, 110, 10623. (41) Weyerich, B.; Brunner-Popela, J.; Glatter, O. J. Appl. Crystallogr. 1999, 32, 197.

Structural, Dynamical Properties of TQ-n (42) Pusey, P. N.; Fijnaut, H. M.; Vrijm, A. J. Chem. Phys. 1982, 77, 4270. (43) Salgi, P.; Rajagopolan, R. AdV. Colloid Interface Sci. 1993, 43, 169. (44) Fritz, G.; Bergmann, A. J. Appl. Crystallogr. 2004, 37, 815. (45) Glatter, O. Prog. Colloid Polym. Sci. 1991, 84, 46. (46) Glatter, O. J. Appl. Crystallogr. 1980, 13, 577. (47) Glatter, O. J. Appl. Crystallogr. 1980, 13, 7. (48) Glatter, O. J. Appl. Crystallogr. 1981, 14, 101. (49) Glatter, O.; Hainisch, B. J. Appl. Crystallogr. 1984, 17, 435. (50) Schnablegger, H.; Glatter, O. Appl. Opt. 1991, 30 (33), 4889. (51) Glatter, O. J. Appl. Crystallogr. 1979, 12, 166. (52) Shrestha, L. K.; Shrestha, R. G.; Varade, D.; Aramaki, K. Langmuir 2009, 25, 4435. (53) Evans, D. F.; Wennerstrom, H. The Colloidal Domain; VCH: New York, 1994; Chapter 11. (54) Aveyard, R.; Binks, B. P.; Cooper, P.; Fletcher, P. D. I. AdV. Colloid Interface Sci. 1990, 33, 59. (55) Kunieda, H.; Ozawa, K.; Huang, K.-L. J. Phys. Chem. B 1998, 102, 831.

J. Phys. Chem. B, Vol. 114, No. 37, 2010 12017 (56) Kunieda, H.; Horii, M.; Koyama, M.; Sakamoto, K. J. Colloid Interface Sci. 2001, 236, 78. (57) Rodriguez, C.; Shigata, K.; Kunieda, H. J. Colloid Interface Sci. 2000, 223, 197. (58) Stradner, A.; Sedgwick, H.; Cardinaux, F.; Poon, W. C. K.; Egelhaaf, S. U.; Schurtenberger, P. Nature 2004, 432, 492. (59) Stradner, A.; Glatter, O.; Schurtenberger, P. Langmuir 2000, 16, 5354. (60) Carnahan, N. F.; Starling, K. E. J. Chem. Phys. 1969, 51, 635. (61) Sato, T.; Hossain, Md. K.; Acharya, D. P.; Glatter, O.; Chiba, A.; Kunieda, H. J. Phys. Chem. B 2004, 108, 12927. (62) Tomsˇicˇ, M.; Besˇter-Rogacˇ, M.; Jamnik, A.; Kunz, W.; Touraud, D.; Bergmann, A.; Glatter, O. J. Phys. Chem. B 2004, 108, 7021. (63) Fijnaut, H. M. J. Chem. Phys. 1981, 74, 6857. (64) Lindner, H.; Scherf, G.; Glatter, O. Phys. ReV. E 2003, 67, 0614021-9. (65) Glatter, O.; Orthaber, D.; Stradner, A.; Scherf, G.; Fanun, M.; Garti, N.; Cle´ment, V.; Leser, M. E. J. Colloid Interface Sci. 2001, 241, 215. (66) Shrestha, L. K.; Sato, T.; Aramaki, K. Phys. Chem. Chem. Phys. 2009, 11, 4251.

JP103080B