Oil-Induced Anomalous Thermoresponsive Viscoelasticity in

12146

J. Phys. Chem. B 2007, 111, 12146-12153

Oil-Induced Anomalous Thermoresponsive Viscoelasticity in Fluorinated Surfactant Systems Suraj Chandra Sharma,†,‡ Carlos Rodrı´guez-Abreu,§ Lok Kumar Shrestha,† and Kenji Aramaki*,† Graduate School of EnVironment and Information Sciences, Yokohama National UniVersity, Tokiwadai 79-7, Hodogaya-ku, Yokohama 240-8501, Japan, Department of Pure and Applied Chemistry, Faculty of Science and Technology, Tokyo UniVersity of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan, and Institut d’InVestigacions Quı´miques i Ambientals de Barcelona, Consejo Superior de InVestigaciones Cientı´ficas (IIQAB/CSIC), Jordi Girona, 18-26, 08034 Barcelona, Spain ReceiVed: May 8, 2007; In Final Form: August 13, 2007

We have studied the rheology and structure of a mixed nonionic fluorinated surfactant, perfluoroalkyl sulfonamide ethoxylate, C8F17SO2N(C3H7)(CH2CH2O)nH abbreviated as C8F17EO10, and perfluorodecalin (C10F18) or perfluoropolyether oil, (C3F6O)nCOOH, in an aqueous system using rheometry and small-angle X-ray scattering (SAXS) techniques. In the absence of oil, the viscosity of surfactant solutions (10 and 15 wt %) first decreases slightly and then more strongly with temperature. Addition of a small amount of fluorinated oil to the wormlike micellar solution disrupts the network structure and decreases the viscosity sharply at lower temperature indicating a rod-sphere transition. The trend of the viscosity curve changes gradually and an anomalous viscosity maximum as a function of temperature appears. It is found that perfluoropolyether oil decreases the viscosity more effectively than perfluorodecalin. The generalized indirect Fourier transformation (GIFT) analysis of the SAXS data confirmed the formation of long rod-like particles in an oil-free, surfactant/water system at 20 °C. Addition of a trace amount of fluorinated oils induces modulation in the structure of the micelles and eventually short rods or spherical particles are formed. The decreasing trend in the viscosity with oil concentration is thus attributed to the microstructure changes induced by the added oils.

Introduction Fluorocarbon surfactants have particular features that make them irreplaceable in several applications. They reduce the surface tension much more than hydrocarbon surfactants, which is important for wetting agents.1 Additionally, they are chemically and thermally stable, therefore they can be used at high temperatures as in fire-fighting foams.2 On the other hand, their solubilization capacity for fluorinated oils is naturally higher than that of hydrocarbon surfactants, whose lipophilic parts are not compatible with fluorocarbon chains and tend to segregate.3 Moreover, it is known that fluorinated surfactants can form viscoelastic solutions of elongated micelles due to the high packing density of fluorocarbon chains that produce low interfacial areas and therefore induce micellar growth.4 Besides the above-mentioned applications, there is an increasing interest in the use of fluorocarbon surfactants for the synthesis of nanostructured materials, due to the very low aggregation concentration and the possibility of new morphologies and smaller dimensions. For the preparation of nanomaterials through cooperative self-assembly, it is of great importance to know the structure of the aggregates in the precursor solution and the ways to control it.5 For example, it has been reported that the pores in mesoporous materials can be tuned by addition of lipophilic additives that are incorporated inside the aggregates acting eventually as structure-directing agents.6 * To whom correspondence should be addressed. E-mail: aramakik@ ynu.ac.jp. Phone/fax: +81-45-339-4300. † Yokohama National University. ‡ Tokyo University of Science. § Consejo Superior de Investigaciones Cientı´ficas (IIQAB/CSIC).

These additives, however, can also drive structural changes that affect the final materials. When hydrocarbon oil is solubilized in elongated micelles, the viscosity is largely changed. Hence, it is considered that the surfactant layer curvature would be changed to less or more positive upon addition of oils. Due to the change in micellar shape and size, the viscosity either increases or decreases depending upon the type of solubilized oil. The oil-induced rod-sphere transition has been welldocumented in the literature for hydrocarbon surfactant systems7-10 but fewer data are available for fluorinated compounds.11,12 There are recent studies on the formation of mesoporous silica from mixtures of fluorinated surfactant and oil,13,14 but still there are some open questions on the effect of oil on fluorinated micelles. Temperature is another parameter that affects the micellar growth and rheological behavior of wormlike micellar solution, and the effect of temperature should be more pronounced in aqueous systems of nonionic surfactants, especially those having polyoxyethylene head groups because the hydration of oxyethylene units is sensitive to the temperature. In this context, we present a report on the rheology and structure of mixed fluorinated surfactant/oil systems, using rheometry and Small-Angle X-ray Scattering (SAXS). Experimental Section Materials. The fluorinated surfactant perfluoroalkyl sulfonamide ethoxylate, C8F17SO2N(C3H7)(C2H4O)10H, designated as C8F17EO10 was obtained from Mitsubishi Materials (Japan). The schematic molecular structure of the surfactant is shown in Scheme 1. The surfactant was purified by placing it under vacuum for several days to remove volatile components until

10.1021/jp073508y CCC: $37.00 © 2007 American Chemical Society Published on Web 10/04/2007

Oil-Induced Anomalous Thermoresponsive Viscoelasticity

J. Phys. Chem. B, Vol. 111, No. 42, 2007 12147

SCHEME 1 : Molecular Structure of C8F17EO10

weight became constant. The surfactant was used without further purification. Perfluoropolyether oil having the structure F-(CF2CF2CF2O)n-CF2CF2COOH designated as (C3F6O)nCOOH was kindly provided by Daikin Industries Ltd., Japan. The average molecular weight of (C3F6O)nCOOH is 3600, which gives n ≈ 21. The polydispersity is 1.14. Octadecafluorodecahydronaphthalene (perfluorodecalin) of purity 95% (mixture of cis and trans) was the product of Aldrich. Millipore water was used. Rheological Measurements. Samples for rheological measurements were homogenized and kept in a water bath at 25 °C for at least 24 h to ensure equilibration before performing measurements. The rheological measurements were performed in an ARES rheometer (Rheometric Scientific), using couette geometry (cup diameter 17 mm; bob diameter 16.5 mm; and bob length 13.3 mm) for low viscosity solutions and cone-plate geometry (diameter 50 mm; cone angle 0.04 rad) for highly viscous solutions. Temperature was controlled by using a circulating fluid from a temperature-controlled water bath. A sample cover provided with the instrument was used to minimize the change in sample composition by evaporation during the measurement. While studying the effect of temperature on the rheological behavior, samples were maintained at each temperature from 30 min (for samples of low viscosity) to 1 h (for samples of high viscosity) before the measurements, which was sufficient to attain stability in rheological parameters (η and Go), as suggested by separate rheological measurements as a function of time. Frequency sweep measurements were performed in the linear viscoelastic regime of the samples, as determined previously by dynamic strain sweep measurements. The zero-shear viscosity (ηo) of the samples was determined from a steady-shear measurement by extrapolating the viscosity-shear-rate curve to zero shear-rate for less viscous samples or from the values of Go and τR as obtained from oscillatory measurements for viscous samples by using the following relation:

η0 ) GoτR

(1)

where Go is the plateau modulus, which is related to the number density of entanglements in the transient network, and τR is the relaxation time, which is related to the average length of the wormlike micelles. Small-Angle X-ray Scattering (SAXS). SAXS measurements were carried out on surfactant/water and surfactant/water/ oil systems. The measurement was performed with a SAXSess camera (Anton Paar, PANalytical) attached to a PW3830 laboratory X-ray generator with a long fine focus sealed glass X-ray tube (KR wavelength of 0.1542 nm) (PANalytical). The apparatus was operated at 40 kV voltage and 50 mA current, respectively. The SAXSess camera is equipped with a focusing multiplayer optics and a block collimator for an intense and monochromatic primary beam with low background, and a translucent beam stop for the measurement of an attenuated primary beam at q ) 0. Samples were enclosed into a vacuumtight thin quartz capillary with an outer diameter of 1 mm and thickness of 10 µm and the same capillary was used for all measurements to attain exactly the same scattering volume and background contribution. The sample temperature was controlled with a thermostated sample holder unit (TCS 120, Anton Paar). The scattered intensity was first measured on an image plate (IP) detection system Cyclone (Perkin-Elmer, USA), and the

two-dimensional intensity data were finally transformed into one-dimensional scattering curves as a function of the magnitude of the scattering vector by using SAXSQuant software (Anton Paar). The scattering angle is related to the scattering vector by the following equation

q)

4π sin(θ/2) λ

(2)

where θ is the angle between the incident beam and the scattered radiation. All data were transmission-calibrated by normalizing an attenuated primary intensity at q ) 0 to unity, and were corrected for the background scattering from the capillary and the solvents. The absolute scale calibration was done with water as a secondary standard. For monodisperse spherical systems the total scattered intensity I(q) involving n particles in unit volume is generally given by

I(q) ) nP(q)S(q)

(3)

where P(q) is the averaged form factor and S(q) is the static structure factor. For a dilute system S(q) is equal to one and the I(q) is simply given by P(q). However, in the semidilute or dense systems, the structure factor is no longer unity and the interparticle interaction comes into play. Therefore, proper choice of the structure factor model in the SAXS data analysis is very essential. In general, for concentrated systems, the contribution of the intraparticle scattering is related to the form factor P(q) and the interparticle correlations are related to the structure factor, S(q). An expression similar to eq 3 can be applied to polydisperse spherical and nonspherical systems by replacing S(q) with Seff(q).15 Seff(q) is no longer only a function of the particle distribution in space but depends also on the form amplitudes of the particles. The average form factor, P(q), is given by the Fourier transformation of the pair-distance distribution function (PDDF), p(r), as

P(q) ) 4π

∫0∞ p(r) sinqrqr dr

(4)

The PDDF of a particle of an arbitrary shape with a scattering density difference of ∆G(r) is given by

p(r) ) r2∆G˜ 2(r)

(5)

where ∆F˜ 2(r) is the convolution square of ∆F(r) averaged over all directions in space. For a spherical particle, averaging is not necessary because the scattering density difference, or electron density difference, ∆Gs(r), is only a function of the radial position and deconvolution of PDDF gives the radius contrast profile. For cylindrical scattering particles such as rodlike micelles the situation is similar to a sphere within a cross section, i.e., the contrast profile ∆Gc(r) is a function of the radial position within the cross section and overall PDDF can be obtained from the Fourier transformation of the I(q) curve. If the cylinder is at least 3 times longer than the cross-sectional diameter, it is possible to obtain the radial profile ∆Gc(r), which is related to the PDDF for the cross section, pc(r), by

pc(r) ) r∆G˜ c2(r)

(6)

The cross-sectional PDDF, pc(r), is related to the scattered intensity by a particle of length L by the following relationship:

12148 J. Phys. Chem. B, Vol. 111, No. 42, 2007

I(q) )

πLIc(q) 2π2L ) q q

∫0∞ pc(r)Jo(qr) dr

Sharma et al.

(7)

where Jo(qr) is the zero-order Bessel function and Ic(q) is the cross-sectional scattering function. The indirect Fourier transformation of the above equation gives pc(r), which in turn can be used to obtain ∆Gc(r). In the present study, the SAXS measurements were carried out on 3 wt % surfactant systems at 20 °C. Since it is a semidilute system, the SAXS data were analyzed by the generalized indirect Fourier transformation (GIFT) technique,16,17 to consider the interparticle interactions. Nevertheless, the model free indirect Fourier transformation (IFT) analysis of the SAXS data also yields almost identical results. The GIFT determines P(q) and S(q) to be simultaneously with minimal assumptions, letting P(q) be model-free in combination with an appropriate choice of the interparticle interaction potential model and the closure relation for the calculation of S(q). In the GIFT calculation, we used the averaged structure factor for a hardsphere (HS) interaction model, S(q)av,18,19 which considers the Gaussian distribution of the interaction radius σ to account for polydispersity µ, together with the choice of a PercusYevick (PY) closure relationship to solve the Ornstein-Zernike (OZ) equation. The detailed theoretical description of the method has been reported elsewhere.17,20-22 The S(q) analysis become no longer exact for the systems with long rod-like particles. However, the GIFT approach tries to reduce the influence of the interparticle interference scattering on the calculation of p(r), and deduces a realistic p(r).23-26 Results and Discussions Rheometry. Perfluorodecalin Systems. In a water/C8F17EO10 binary system, surfactant forms an aqueous micellar solution (Wm), a lamellar phase (LR), and a reverse micellar solution (Om) at 25 °C. Below 25 °C, hexagonal (H1) and bicontinuous cubic (V1) phases are also observed. Micellar solutions at low temperature (∼5 °C) become increasingly viscous with increasing surfactant concentration, and at compositions near the H1 phase, a highly viscous or gel-like solution is formed that consists of wormlike micelles. With increasing temperature the viscosity of the wormlike micellar solution gradually decreases and finally a less viscous easily flowing isotropic solution is formed. At higher temperature a phase separation occurs at the cloud point, which is typical of a nonionic surfactant system. A detailed phase behavior of this system in water is available elsewhere.27 Figure 1 shows data on viscosity as a function of temperature in C8F17EO10/water/perfluorodecalin systems at different concentrations of perfluorodecalin. In the absence of oil, the viscosity first decreases slightly and then more strongly with temperature; such a decrease does not follow an Arrheniustype behavior, as in the case of ionic surfactants.28,29 Note that the viscosity is quite high at low temperatures, due to the presence of elongated, worm micelles, as reported previously.30 At low surfactant concentration (Figure 1a), the trend of the viscosity curve changes gradually as oil (perfluorodecalin) is added, and a viscosity maximum as a function of temperature appears. The existence of a maximum has been attributed to temperature-induced micellar growth followed by micellar branching and/or micellar disruption.30 However, all curves coincide at high temperatures; namely, the systems are more oil sensitive at low temperatures. Such sensitivity with temperature almost disappears with an increase in surfactant concentration, as shown in Figure 1b. The results can be explained in terms of changes in the curvature of the surfactant layer and

the packing parameter p ) V/(l a), where a is the interfacial area per amphiphile and V and l are the effective volume and effective length of the lipophilic part, respectively. If the added oil is solubilized in the core of aggregates, l would increase and, consequently, p will decrease, namely, aggregates with more positive curvature would tend to form. Hence, in the present system, there will be an oil-induced rod-sphere transition, which is responsible of the viscosity decrease at low temperatures. When surfactant concentration is increased a tends to decrease, which compensates the increase in l, and the curvature of aggregates would change less upon oil addition; therefore, the viscosity values would be similar. In nonionic surfactants systems, an increase in temperature will also lead to a decrease in a, but it also causes an increase in the diffusion coefficients and micellar breaking that surpasses the effect of a, therefore the viscosities will tend to be similar at high temperatures. The dynamics of wormlike micelles under oscillatory shear is described by considering two different processes. When a small strain is applied, the stress relaxation occurs by reptation, that is, a reptile-like motion of the micelle along its own contour. Besides this, micelles may undergo reversible scission.31 When the time scale of the reptation for an average micellar contour length is too short in comparison to the time scale of the scission, the viscoelastic micellar solutions behave as a Maxwell fluid32,33 with a single relaxation time,28 and variation of the elastic or storage modulus G′(ω) and the viscous or loss modulus G′′(ω) as a function of oscillatory-shear frequency, ω, is described by the following relationships:

G′(ω) )

G′′(ω) )

ω2τR2 1 + ω2τR2 ωτR 1 + ω2τR2

Go

(8)

Go

(9)

where Go and τR are the plateau modulus and the relaxation time, respectively. At high frequencies, G′ tends to attain a constant value equal to Go. The relaxation time τR may be estimated from the G′-G′′crossover frequency, that is τR ) 1/ω, when G′ ) G′′. Figure 2 shows a plot of the elastic modulus (G′) and loss modulus (G′′) as a function of oscillatory-shear frequency (ω) for a sample of 10 wt % surfactant solution containing perfluorodecalin at 20, 25, and 30 °C. At 20 °C, there is a clear viscoelastic behavior: at low frequencies, the viscous modulus G′′ is larger than the elastic modulus (G′) and the system behaves like a liquid; at high frequencies, G′ > G′′ and the system behaves like a solid. The system follows the Maxwell model of relaxation in the low-frequency range, but departs from it at high frequencies, which might be attributed to the existence of multiple relaxation times. This deviation is thought to have arisen from a transition of the relaxation mode from “slower” reptation to “faster” relaxation modes, namely, the Rouse modes.31,34 Maxwellian-type oscillatory rheological behavior of viscous micellar solutions, such as that shown in Figure 2, can be related to the transient network formed by the entanglement of wormlike micelles. It can be seen from the figure that with increasing temperature the plateau value of G′ increases, which corresponds to an increase in network density. The G′-G′′ crossover frequency shifts to the high-frequency region that suggests faster relaxation processes. The rheological parameters derived from fittings to the Maxwell model are presented in Figures 3 and 4. The plateau



Figure 1. Variation of zero-shear viscosity (ηo) with temperature for (a) 10 and (b) 15 wt % of C8F17EO10 in water/C8F17EO10/perfluorodecalin systems. The perfluorodecalin weight fractions are Wo ) 0 (circles), 0.001 (squares), 0.002 (triangles), and 0.003 (diamonds).

Figure 2. Variation of storage modulus (G′, filled symbols) and loss modulus (G′′ open symbols) with oscillatory shear frequency (ω) in water/C8F17EO10/perfluorodecalin systems (Wo ) 0.002, 10 wt % surfactant concentration) at different temperatures. The lines are best fits to the Maxwell model.

modulus (Go) decreases with oil content and increases with temperature and concentration. Go is proportional to the number of entanglements N in the system,35 then it seems that the increase in aggregate curvature induced by addition of oil results in a reduction in N. On the other hand, Go and, therefore, N increase with temperature or concentration, as for semidilute polymer solutions in good solvents.36 The Maxwellian relaxation time τR follows the same tendency of viscosity (Figure 1), namely, in the absence of oil τR decreases almost continuously with temperature but it shows a maximum when oil is added. The effect of oil is very small when surfactant concentration is increased. The results support the explanation based on the packing parameter. The increase in relaxation time before the maximum can be attributed to micellar growth induced by temperature (decrease of interfacial area a) whereas the decrease of τR after the maximum is probably a consequence of either micellar branching, because the micellar joints can slide along the micellar body allowing fast relaxation, or micellar disruption and formation of small globular micelles, especially in the presence of oil. Higher temperatures also result in higher diffusion coefficients according to the Stokes-Einstein equation, which will be reflected in shorter relaxation times. On the other hand, an increase of temperature accelerates the exchange of surfactant unimers between micelles, therefore unimers spend less time in micellar end-caps at a higher temperature.29 As a result, more end-caps can be formed, and this implies shorter micelles, and shorter relaxation times. (C3F6O)nCOOH Systems. Figure 5 shows the effect of (C3F6O)nCOOH on the viscosity of C8F17EO10/water systems at different temperatures and surfactant concentrations. The trends are similar to those of perfluorodecalin systems, but the effect of (C3F6O)nCOOH is stronger: at a given oil, surfactant concentration, and temperature, the viscosities are lower for (C3F6O)nCOOH, and the maximum as a function of temperature is more pronounced, i.e., the anomalous thermoresponsive

viscoelasticity is stronger. There is also a clear shift of the maximum to higher temperatures as (C3F6O)nCOOH concentration is increased. This effect is compensated by an increase in surfactant concentration, as shown in Figure 5b. Nevertheless, the viscosity values for (C3F6O)nCOOH and perfluorodecalin systems are very similar and independent of oil concentration for T > 25 °C, suggesting that other mechanisms different from curvature change, such as increased diffusion and micellar breaking, are more relevant. Similar to perfluorodecalin systems, systems in the presence of (C3F6O)nCOOH are highly viscoelastic at low temperatures. The Maxwell model again fails to describe the systems at high frequencies, as multiple modes of relaxation appear. The maxwellian parameters for (C3F6O)nCOOH systems show tendencies similar to those for perfluorodecalin systems. Go and therefore N increase with temperature and surfactant concentration and decreases with oil content. τR values decrease with temperature and (C3F6O)nCOOH concentration, the effect of oil being larger at low surfactant concentrations. However, Go and τR values are lower than those corresponding to perfluorodecalin systems, therefore, the effect of oil depends on its structure, as will be discussed below. In nonionic hydrocarbon surfactant systems, the average aggregation number in the limit of strong micellar growth can be expressed as37

N ) No + (KX)1/2

(10)

where No is the aggregation number of the globular micelle before the uniaxial growth, X is the surfactant mole fraction, and K is given approximately by

K = exp(-Ec/kBT)

(11)

where kB is the Boltzmann constant and Ec is the free energy required to form two end caps in the cylindrical body of the micelle and controls the extent of micellar growth. According to eqs 10 and 11, the micellar growth will be favored by temperature. On the other hand, the average worm length Lh (proportional to N) is predicted to decrease exponentially with temperature according to the equation28

Lh = φ1/2 exp(Es/2kBT)

(12)

Here, φ is the volume fraction of worms and Es is the micellar scission energy. Then, temperature exerts two competitive effects. It can be argued that in the low-temperature range, Ec > Es and therefore micellar length and viscosity would increase with temperature, whereas in the high-temperature range Es > Ec, hence micellar disruption would prevail, causing the viscosity and relaxation time to decrease, as seen in Figures 4 and 8. The shift of the viscosity maximum with (C3F6O)nCOOH


Sharma et al.

Figure 3. Variation of plateau modulus (Go) with temperature for (a) 10 and (b) 15 wt % of C8F17EO10 in water/C8F17EO10/perfluorodecalin systems. The perfluorodecalin weight fractions are Wo ) 0 (circles), 0.001 (squares), 0.002 (triangles), and 0.003 (diamonds).

Figure 4. Variation of relaxation time (τR) with temperature for (a) 10 and (b) 15 wt % of C8F17EO10 in water/C8F17EO10/perfluorodecalin systems. The perfluorodecalin weight fractions are Wo ) 0 (circles), 0.001 (squares), 0.002 (triangles), and 0.003 (diamonds).

Figure 5. Variation of zero-shear viscosity (ηo) with temperature for (a) 10 and (b) 15 wt % of C8F17EO10 in water/C8F17EO10/(C3F6O)nCOOH systems. The (C3F6O)nCOOH weight fractions are Wo ) 0 (circles), 0.001 (squares), 0.002 (triangles), and 0.003 (diamonds).

Figure 6. Variation of storage modulus (G′, filled symbols) and loss modulus (G′′ open symbols) with oscillatory shear frequency (ω) in water/C8F17EO10/(C3F6O)nCOOH systems (Wo ) 0.002, 10 wt % surfactant concentration) at different temperatures. The lines are best fits to the Maxwell model.

concentration observed in Figure 5 could be explained if one takes into account that upon oil addition the curvature becomes more positive and Ec will be lower, so a higher temperature will be needed to induce an increase in viscosity. Although anomalous thermoresponsive behavior has been found in some ionic surfactant systems38-40 to our knowledge this is the first report of such a behavior in nonionic surfactants in the presence of oil.

SAXS Measurements. To investigate the microstructure changes in terms of shape and size of the micellar aggregates induced by the fluorinated oils SAXS measurements were carried out on 3 wt % surfactant solutions at different concentrations of added oils. We have performed SAXS measurements in the semidilute system, although the rheological measurements were carried out at higher surfactant concentrations, in order to avoid the possible interparticle interference scattering effects. The temperature-viscosity plot for the 3 wt % surfactant solution possesses maximum viscosity at 20 °C.30 Therefore, all the SAXS measurements were carried out at this temperature in order to observe the clear oil effect on the micellar structure and hence on the rheology. Figure 9 shows the results on SAXS measurements in perfluorodecalin samples. The slope of the intensity curves (Figure 9a) at low q seems to decrease as perfluorodecalin is added, suggesting a rod-sphere transition. This is more clearly seen in the pair-distance distribution functions p(r) derived from the GIFT analysis, which give information on the shape of aggregates. In the absence of oil, the p(r) profiles are highly asymmetric, which is typical of elongated aggregates. The value of r at which p(r) becomes zero (Dmax) gives in principle an estimation of the maximum length of the aggregates. However, in the present system the values of Dmax should be taken as



Figure 7. Variation of plateau modulus (Go) with temperature for (a) 10 and (b) 15 wt % of C8F17EO10 in water/C8F17EO10/(C3F6O)nCOOH systems. The (C3F6O)nCOOH weight fractions are Wo ) 0 (circles) and 0.001 (squares).

Figure 8. Variation of relaxation time (τR) with temperature for (a) 10 and (b) 15 wt % of C8F17EO10 in water/C8F17EO10/(C3F6O)nCOOH systems. The (C3F6O)nCOOH weight fractions are Wo ) 0 (circles) and 0.001 (squares).

Figure 9. SAXS analysis results for water/C8F17EO10/perfluorodecalin systems at 20 °C. The C8F17EO10 concentration is 3 wt %. (a) Scattering curves (lines are best fittings) and (b) the corresponding PDDF functions. The perfluorodecalin weight fractions are Wo ) 0 (circles), 0.0015 (squares), 0.0034 (triangles), and 0.0051 (diamonds).

qualitative since the electron density contrast is not high enough. From Figure 9b it is clear that Dmax decreases upon addition of perfluorodecalin and that the curves become symmetric, both indications that the micelles are getting shorter and spherical (globular). As can be seen in Figure 10, the effect of (C3F6O)nCOOH on the size and shape of aggregates is qualitatively similar to that of perfluorodecalin but stronger. The p(r) profiles again indicate an oil induced rod-sphere transition, in agreement with rheometry results. The effect of oils on the morphology of aggregates is more clearly observed in Figure 11. Dmax is plotted together with Dratio, defined as Dratio ) Dmax/(2rm); rm is the value of r at the maximum of the p(r) curves in Figures 9 and 10, which would give the radius of spherical aggregates.41 For spherical micelles, Dratio ) 1, whereas for elongated micelles, Dratio >1. Both Dmax and Dratio decrease with oil content, suggesting that aggregates become smaller. Dratio approaches unity as oil is added, indicating a decreasing axial ratio and a rod-sphere transition.

The decrease in Dratio is steeper for (C3F6O)nCOOH. It can be argued that the long chain (C3F6O)nCOOH tends to migrate more into the micellar cores, therefore having a larger effect on the interfacial curvature than perfluorodecalin. After the geometry of the micelles is determined, it is possible to calculate the cross-sectional PDDF, pc(r), of the cylindrical geometry. Nevertheless, the procedure is strictly valid only when the axial length of the cylindrical micelles is at least three times longer than its cross-sectional diameter. The pc(r) gives very important information about the internal structure of the particles, as for example, the value of r at which pc(r) ) 0 gives an estimate of maximum cross-sectional dimension of the cylindrical micelles. Figure 12a shows the pc(r) of the surfactant/ water and the surfactant/water/oil systems. The maximum crosssectional dimension of the surfactant/water system (∼5.95 nm) increases upon addition of a trace amount of perfluorodecalin oil (as is indicated by an arrow in Figure 12a) reflecting the formation of an oil pool at the micellar core. The increasing tendency of the cross-section with added oil may semiquanti-


Sharma et al.

Figure 10. SAXS analysis results for water/C8F17EO10/(C3F6O)nCOOH systems at 20 °C. The C8F17EO10 concentration is 3 wt %. (a) Scattering curves (lines are best fittings) and (b) PDDF curves extracted from the GIFT analysis of the SAXS data. The (C3F6O)nCOOH weight fractions are Wo ) 0 (circles), 0.0015 (squares), 0.0035 (triangles), and 0.0051 (diamonds)

nm (0% and 0.15% oil systems) is because of the -C3H7 moiety of low electron density that might have balanced the positive contribution of N, S, and O, bringing the overall contrast profile to a low value. A slight increase of the electron density value, which eventually leads to zero after the minimum, is attributed to the ethylene oxide unit of the surfactant. As can be seen from Figure 12b, the minimum position in the contrast profile, which gives an estimation of micellar core radius, shifts toward the higher r side in the oil-added system. This emphasizes that the cross-sectional length of the particle goes on increasing with oil content and eventually a spherical shaped micelle is formed. These data well support the results extracted from Figures 9 and 10. Figure 11. Variation of Dmax (circles) and Dratio (squares) as a function of weight fractions of oil (Wo), perfluorodecalin (filled symbols), and (C3F6O)nCOOH (open symbols). The C8F17EO10 concentration is fixed at 3 wt %.

tatively indicate the cylinder-to-sphere type of transition in the structure of the particles. In Figure 12b, we present the contrast profiles ∆Fc(r) and ∆Fs(r) obtained by the deconvolution of the pc(r) and p(r), respectively. The ∆Fc(r) and ∆Fs(r) represent the contrast profile of cylindrical and spherical particles, respectively. In the present systems, the micellar core has an electron-rich fluorocarbon chain. Therefore, it shows a high contrast in the electron density profile. The high contrast of the system is due to the fact that the overall electron density of the fluorocarbon chain is pretty much higher than that of the solvent water. A minima observed in the contrast profile around r ) 2

Conclusions The network structure of the wormlike micellar solution is disrupted due to addition of a trace amount of fluorinated oil and the viscosity decreases dramatically at lower temperatures (below 15 °C) indicating rod-sphere transition. A viscosity maximum as a function of temperature appears as oil is solubilized in the wormlike solution. This anomalous thermoresponsive behavior is attributed to the competitive effects between micellar elongation and disruption in non-ionic surfactant systems. It is found that perfluoropolyether oil decreases the viscosity more strongly than perfluorodecalin. The SAXS data well support the data obtained from the rheological measurements. Namely, the decrease in the viscosity with added oils is due to the microstructure change. The axial length of rod-like micelles formed in the surfactant/water system decreases

Figure 12. (a) The cross-sectional PDDF, pc(r), for the weight fraction of oil, Wo ) 0 (circles), and perfluorodecalin, Wo ) 0.0015 (squares), and (b) the corresponding contrast profile, ∆Fc(r), ∆Fs(r) for the weight fractions of oil Wo ) 0 (circles), perfluorodecalin, Wo ) 0.0015 (squares), perfluorodecalin, Wo ) 0.0051 (triangles), and (C3F6O)nCOOH, Wo ) 0.0051 (diamonds).

Oil-Induced Anomalous Thermoresponsive Viscoelasticity upon addition of perfluorodecalin and perfluoropolyether oil and eventually spherical particles are formed. The present SAXS data give a clear picture on the oil-induced rod-to-sphere transition in the fluorinated surfactant system. Besides, the data show the possible route to control over the structure of the aggregates. Acknowledgment. This work was partly supported by Core Research for Evolution Science and Technology (CREST) of JST Corporation. S.C. Sharma and L.K. Shrestha are thankful to the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan for being given a Monbukagakusho Scholarship. References and Notes (1) Fletcher, P. D. I. In Specialist Surfactants; Robb, I. D., Ed.; Blackie A&P: Cambridge, UK, 1997; p 135. (2) Kissa, E. Fluorinated Surfactants and Repellents, 2nd ed.; Marcel Dekker: New York, 2001; Chapter 8. (3) Johan, C.; Peter, S. J. Phys. Chem. 1984, 88, 4410. (4) Hoffmann, H.; Wu¨rtz, J. J. Mol. Liquids 1997, 72, 191. (5) Lazzari, M.; Rodrı´guez, C.; Rivas, J.; Lo´pez-Quintela, A. J. Nanosci. Nanotechnol. 2006, 6, 892. (6) Patrick, S.-W.; Charles, J. G.; Galen, D. S. Langmuir 2000, 16, 356. (7) Hoffmann, H.; Ulbricht, W. J. Colloid Interface Sci. 1989, 129, 388. (8) Rodrı´guez, C.; Aramaki, K.; Tanaka, Y.; Lo´pez-Quintela, M. A.; Ishitobi, M.; Kunieda, H. J. Colloid Interface Sci. 2005, 291, 560. (9) Sato, T.; Acharya, D. P.; Kaneko, M.; Aramaki, K.; Singh, Y.; Ishitobi, M.; Kunieda, H. J. Dispersion Sci. Technol. 2006, 27, 611. (10) Pokhriyal, N. K.; Joshi, J. V.; Goyal, P. S. Colloids Surf. A 2003, 218, 201. (11) Sharma, S. C.; Rodrı´guez, C.; Esquena, J.; Kunieda, H. J. Colloid Interface Sci. 2006, 299, 297. (12) Sharma, S. C.; Acharya, D. P.; Aramaki, K. Langmuir 2007, 23, 5324. (13) Blin, J. L.; Ste´be´, J. M. J. Phys. Chem. B 2004, 108, 11399. (14) Blin, J. L.; Ste´be´, J. M. Microporous Mesoporous Mater. 2005, 87, 67. (15) Weyerich, B.; Brunner-Popela, J.; Glatter, O. J. Appl. Crystallogr. 1999, 32, 197. (16) Brunner-Popela, J.; Glatter, O. J. Appl. Crystallogr. 1997, 30, 431.

J. Phys. Chem. B, Vol. 111, No. 42, 2007 12153 (17) Glatter, O.; Fritz, G.; Lindner, H.; Brunner-Popela, J.; Mittelbach, R.; Strey, R.; Egelhaaf, S. U. Langmuir 2000, 16, 8692. (18) Pusey, P. N.; Fijnaut, H. M.; Vrijm, A. J. Chem. Phys. 1982, 77, 4270. (19) Salgi, P.; Rajagopolan, R. AdV. Colloid Interface Sci. 1993, 43, 169. (20) Strey, R.; Glatter, O.; Schubert, K.-V.; Kaler, E. W. J. Chem. Phys. 1996, 105, 1175. (21) Glatter, O. Progr. Colloid Polym Sci. 1991, 84, 46. (22) Fritz, G.; Bergmann, A. J. Appl. Crystallogr. 2004, 37, 815. (23) Shrestha, L. K.; Masaya, K.; Sato, T.; Acharya, D. P.; Iwanaga, T.; Kunieda, H. Langmuir 2006, 22, 1449. (24) Shrestha, L. K.; Sato, T.; Acharya, D. P.; Iwanaga, T.; Aramaki, K.; Kunieda, H. J. Phys. Chem. B 2006, 110, 12266. (25) Shrestha, L. K.; Sato, T.; Aramaki, K. J. Phys. Chem. B 2007, 111, 1664. (26) Moitzi, C.; Freiberger, N.; Glatter, O. J. Phys. Chem. B 2005, 109, 16161. (27) Esquena, J.; Rodrı´guez, C.; Solans, C.; Kunieda, H. Microporous Mesoporous Mater. 2006, 92, 212. (28) Cates, M. E.; Candau, S. J. J. Phys.: Condens. Matter 1990, 2, 6869. (29) Tung, S. H.; Huang, Y. E.; Raghavan, S. R. Langmuir 2007, 23, 372. (30) Acharya, D. P.; Sharma, S. C.; Rodrı´guez, C.; Aramaki, K. J. Phys. Chem. B 2006, 110, 20224. (31) Granek, R.; Cates, M. E. J. Chem. Phys. 1992, 96, 4758. (32) Larson, R.G. The Structure and Rheology of Complex Fluids; Oxford University Press: New York, 1999. (33) Marin, G. In Rheological Measurement, 2nd ed.; Collyer, A. A., Clegg, D. W., Eds.; Chapman and Hall: London, UK, 1998. (34) Doi, M.; Edwards, S. F. The theory of polymer Dynamics; Oxford University Press: Clarendon, U.K., 1986. (35) Rehage, H.; Hoffmann, H. J. Phys. Chem. 1988, 92, 4712. (36) Oda, R.; Narayanan, J.; Hassan, P. A.; Manohar, C.; Salkar, R. A.; Kern, F.; Candau, S. J. Langmuir 1998, 14, 4364. (37) Missel, P. J.; Mazer, N. A.; Benedek, G. B.; Young, C. Y.; Carey, M. C. J. Phys. Chem. 1980, 84, 1044. (38) Tobita, K.; Sakai, H.; Kondo, Y.; Yoshino, N.; Kamogawa, K.; Momozawa, N.; Abe, M. Langmuir 1998, 14, 4753. (39) Abe, M.; Tobita, K.; Sakai, H.; Kamogawa, K.; Momozawa, N.; Kondo, Y.; Yoshino, N. Colloids Surf. A 2000, 167, 47. (40) Kalur, G. C.; Frounfelker, B. D.; Cipriano, B. H.; Norman, A. I.; Raghavan, S. R. Langmuir 2005, 21, 10998. (41) Sharma, S. C.; Rodrı´guez-Abreu, C.; Shrestha, L. K.; Aramaki, K. J. Colloid Interface Sci. 2007, 314, 223.

Oil-Induced Anomalous Thermoresponsive Viscoelasticity in

Recommend Documents