Appearance of Pure Fluorocarbon Micelles Surveyed by Fluorescence

Tsuyoshi Asakawa,*,† Shouhei Ishino,† Per Hansson,‡ Mats Almgren,§. Akio Ohta,† and Shigeyoshi Miyagishi†. Department of Chemistry and Chem...
0 downloads 0 Views 122KB Size
Download PDF
6998

Langmuir 2004, 20, 6998-7003

Articles Appearance of Pure Fluorocarbon Micelles Surveyed by Fluorescence Quenching of Amphiphilic Quinoline Derivatives in Fluorocarbon and Hydrocarbon Surfactant Mixtures Tsuyoshi Asakawa,*,† Shouhei Ishino,† Per Hansson,‡ Mats Almgren,§ Akio Ohta,† and Shigeyoshi Miyagishi† Department of Chemistry and Chemical Engineering, Faculty of Engineering, Kanazawa University, Kanazawa 920-8667, Japan, Department of Pharmacy, Uppsala Biomedical Centre, P.O. Box 580, SE-75123 Uppsala, Sweden, and Department of Physical Chemistry, Uppsala University, P.O. Box 579, S 751 23, Uppsala, Sweden Received February 27, 2004. In Final Form: May 28, 2004

A halide-sensitive fluorescence probe was utilized to evaluate the miscibility of fluorocarbon and hydrocarbon surfactants in aqueous micellar systems. The fluorescence of 6-methoxy-N-1,1,2,2-tetrahydroheptadecafluorodecylquinolinium chloride, FC10MQ, was quenched by halide ions dissociated from the surfactant. The fluorescence in micellar solutions showed an initially rapid decay. This suggests that halide ions effectively quench FC10MQ fluorescence at the micellar surface. The subsequent slow decay corresponds to the quenching of FC10MQ fluorescence in the aqueous bulk phase by the free counterions. The Stern-Volmer plots for fluorescence quenching gave a distinct break at the critical micelle concentration of the cationic surfactants. The abrupt increase in fluorescence quenching is attributed to the solubilization of the probe in the micelles. The fluorescence quenching behavior provides direct information about the immiscibility of fluorocarbon and hydrocarbon species in micelles, and the results indicate that almost pure fluorocarbon micelles appear in surfactants mixtures.

Introduction Mixtures of fluorocarbon and hydrocarbon surfactants show unusual characteristics of micellization in aqueous solutions.1-3 Fluorescence probe methods have been extensively utilized to evaluate the physicochemical properties of such nonideal mixed micelles.4-7 Critical micelle concentrations (cmc’s) of surfactant mixtures are conveniently determined from changes in fluorescence spectra or intensities upon micelle formation. Micellar characteristics such as polarity and “microviscosity” can be investigated by pyrene and 1,3-pyrenylpropane, etc.8,9 Quenching of pyrene fluorescence by hexadecylpyridinium * To whom correspondence should be addressed. E-mail: [email protected]. † Kanazawa University. ‡ Uppsala Biomedical Centre. § Uppsala University. (1) Mukerjee, P.; Yang, A. Y. S. J. Phys. Chem. 1976, 80, 1388. (2) Shinoda, K.; Nomura, T. J. Phys. Chem. 1980, 84, 365. (3) Kissa, E. Fluorinated Surfactants; Marcel Dekker, Inc.: New York, 1994. (4) Muto, Y.; Esumi, K.; Meguro, K.; Zana, R. J. Colloid Interface Sci. 1987, 120, 162. (5) Asakawa, T.; Mouri, M.; Miyagishi, S.; Nishida, M. Langmuir 1989, 5, 343. (6) Tamori, K.; Ishikawa, A.; Kihara, K.; Ishii, Y.; Esumi, K. Colloids Surf. 1992, 67, 1. (7) Almgren, M.; Wang, K.; Asakawa, T. Langmuir 1997, 13, 4535. (8) Kalyanasundaram, K. Langmuir 1988, 4, 942. (9) Zana, R. J. Phys. Chem. B 1999, 103, 9117.

in micelles is one of the most valuable methods to determine the micelle aggregation number.10 Dynamic properties of micelles have been investigated by the behavior of probes around the solubilization site during the lifetime of fluorescence. Many investigators have studied the miscibility of fluorocarbon and hydrocarbon surfactants in micelles from the macroscopic solution properties, such as the change of the cmc with composition.1-3 However, it is necessary to clarify the microscopic aspects of the demixing. We have attempted to design a fluorescence probe method demonstrating the immiscibility of fluorocarbon and hydrocarbon surfactants in micelles.7 Recently, we found that water-soluble probes such as quinoline derivatives, the fluorescence of which are quenched by halide ions, could be utilized for estimating the concentration of counterion dissociated from cationic surfactants.11 The fluorescence quenching behavior of N-ethoxycarbonylmethyl-6-methoxyquinolinium bromide (MQAE) was used to determine simultaneously both the cmc of surfactants and the degree of micellar counterion dissociation (R). The value of R was quantified by the twosite model considering “free” ions in the aqueous bulk phase and “bound” ions in the micelle phase.12 On the (10) Grieser, F.; Drummond, C. J. J. Phys. Chem. 1988, 92, 5580 and references therein. (11) Asakawa, T.; Kitano, H.; Ohta, A.; Miyagishi, S.J. Colloid Interface Sci. 2001, 242, 284. (12) Stilbs, P.; Lindman, B. J. Phys. Chem. 1981, 85, 5, 2587.

10.1021/la049488y CCC: $27.50 © 2004 American Chemical Society Published on Web 07/21/2004

Micellar Immiscibility by Fluorescence Quenching

basis of Stern-Volmer plots, we estimated the concentration of “free” ions in micellar systems. The “free” halide ions give rise to fluorescence quenching in the aqueous pseudophase during the fluorescence lifetime of the probe. The “bound” halide electroneutralized with micelles gave no fluorescence quenching in the aqueous phase. No solvent polarity effect on the quenching was observed. Moreover, no quenching by oxygen was disturbing the experimental analysis. The quenching mechanism was reported in detail for lucigenin (bis-N-methylacridinium nitrate) fluorescence probe.13 Collisional quenching of the steady-state fluorescence of the probe is typically described by a Stern-Volmer relation. A diffusion-controlled quenching was found for lucigenin as probe and chloride ion as quencher. The Stern-Volmer constants were 390, 200, 145, and 118 M-1 for lucigenin, MQAE, 6-methoxy-Nethylquinolinium, and 6-methoxy-N-(3-sulfopropyl)quinolinium, respectively.14-16 That is, the Stern-Volmer constants increased with increasing effective positive charge because the positively charged fluorescence probe attracts the negatively charged chloride ion. This would result in increased quenching efficiency for the doubly charged cation of lucigenin. Amphiphilic probes having a quinolinium group would sense the local chloride concentration at the micellar surface. Measurements of fluorescence decay rates may allow a distinction between different local environments of probes. In this paper, the halide-sensitive fluorescence probe having a fluorocarbon chain was prepared to investigate the nonideal behavior for mixed micelles of fluorocarbon and hydrocarbon surfactants. The quenching behavior of amphiphilic fluorescence probes will enable us to estimate the immiscibility of fluorocarbon and hydrocarbon chains in the micellar phase. Experimental Procedures Materials. Alkyltrimethylammonium halides were obtained from Tokyo Kasei Kogyo Co., Ltd., and recrystallized twice from acetone-ethanol mixtures. 1,1,2,2-Tetrahydroheptadecafluorodecylpyridinium chloride, HFDePC, was prepared as reported previously.17 6-Methoxy-N-1,1,2,2-tetrahydroheptadecafluorodecylquinolinium chloride, FC10MQ, was prepared in a similar procedure as HFDePC. 6-Methoxy-N-dodecylquinolinium bromide, C12MQ, was synthesized by refluxing bromododecane with 6-methoxyquinoline. The amphiphilic probes were purified by the recrystallization from acetone and water. 6-Methoxy-Nethylquinolinium iodide (MEQ) was purchased from Molecular Probes, Inc., and used without purification. All the other reagents were of guaranteed grade. The solutions were made up in doubly distilled water. Measurements. The time-resolved fluorescence quenching measurements were performed as reported previously.18 Absorption spectra of the 1.3 × 10-5 M probe were recorded on a Hitachi U-3210 spectrometer. The fluorescence intensities of 1.0 × 10-6 M probe were measured at 449 nm by excitation at 348 nm using a Hitachi F-3010 spectrometer. All experiments were performed at 25 °C. No shift of the emission maximum was observed by the additions of surfactants and/or salts. The fluorescence intensity without quencher (I0) was used as a standard.

Results and Discussion The fluorescence of a quinoline-based molecule is quenched by halide ions.14-16 Substitution of quinoline (13) Legg, K. D.; Hercules, D. M. J. Phys. Chem. 1970, 74, 4, 2114. (14) Verkman, A. S.; Seller, M. C.; Chao, A. C.; Leung, T.; Ketcham, R. Anal. Biochem. 1989, 178, 355. (15) Biwersi, J.; Verkman, A. S. Biochemistry 1991, 30, 7879. (16) Biwersi, J.; Tulk, B.; Verkman, A. S. Anal. Biochem. 1994, 219, 139. (17) Asakawa, T.; Hisamatsu, H.; Miyagishi, S. Langmuir 1995, 11, 478. (18) Hansson, P.; Jonsson, B.; Strom, C.; Soderman, O. J. Phys. Chem. B 2000, 104, 3496.

Langmuir, Vol. 20, No. 17, 2004 6999

Figure 1. Absorption and fluorescence spectra of FC10MQ in aqueous solutions. Quenching occurred in the presence of 1, 3, 5, and 10 mM HFDePC.

with an electron-donating group increased the halide ion sensitivity. The substitution of a methoxyl group at the 6-position on the quinoline ring gave a particulary high sensitivity toward fluorescence quenching. The nature of the N-substituent also influenced the sensitivity. An uncharged group as the N-substituent gave higher sensitivity of fluorescence quenching; e.g., the Stern-Volmer constant (KSV) of N-ethoxycarbonylmethyl-6-methoxyquinolinium bromide (MQAE) is larger than that of 6-methoxy-N-(3-sulfopropyl)quinolinium (SPQ). From the facts described above, we prepared amphiphilic fluorescence probes consisting of 6-methoxyquinoline N-substituted with a hydrophobic group. The amphiphilic fluorescence probe would sense the local concentration of halide ions at the micellar surface. The positive quaternary nitrogen is necessary for the halide ion sensitivity and water solubility. The absorption and fluorescence spectra of 6-methoxyN-1,1,2,2-tetrahydroheptadecafluorodecylquinolinium chloride (FC10MQ) are shown in Figure 1. FC10MQ had a molar extinction coefficient of 6100 M-1 cm-1 at 318 nm and 3700 M-1 cm-1 at 348 nm. The fluorescence had a single broad emission peak centered at 449 nm by the excitation at 348 nm. The FC10MQ fluorescence was quenched by chloride ions without change in the shape of the emission spectrum. No spectral shift was observed on addition of surfactants, not even in the presence of micelles. However, a remarkable decrease in fluorescence intensity was observed upon the formation of HFDePC micelles, showing that FC10MQ is solubilized in HFDePC micelles and senses the local chloride concentration at the micelle surface. To investigate the effect of solubilization, 6-methoxy-Nethylquinolinium iodide (MEQ) and 6-methoxy-N-dodecylquinolinium bromide (C12MQ) probes were also used with HFDePC micelles. The MEQ fluorescence is quenched by chloride ion with a linear Stern-Volmer relation. Figure 2 shows the Stern-Volmer plots for quenching of fluorescence probes in HFDePC systems. The distinct break points correspond to the cmc of HFDePC. The abrupt increase in quenching for FC10MQ upon micelle formation shows that FC10MQ is very strongly quenched in the micelles. In contrast, the quenching behavior of C12MQ was very similar to that of MEQ despite HFDePC micelle formation. This suggests that the amphiphilic C12MQ probe is scarcely solubilized by the fluorocarbon micelles owing to the immiscibility of fluorocarbon and hydrocarbon chains. The cmc value using MEQ was in fair agreement

7000

Langmuir, Vol. 20, No. 17, 2004

Asakawa et al.

Figure 2. Effect of alkyl chain of quinoline derivatives on fluorescence quenching in HFDePC systems: (O) MEQ; (4) FC10MQ; (0) C12MQ. The solid lines above the cmc indicate calculated curves using eq 5.

Figure 3. Stern-Volmer plots for quenching of FC10MQ fluorescence by bromide ions: (O) NaBr; (b) DTAB; (2) TTAB; (9) CTAB. The solid lines above the cmc indicate calculated curves using eq 5.

Scheme 1. Fluorescence Quenching of MEQ and FC10MQ in Aqueous Micellar Solutions

which is given with respect to the concentration of micellar surfactant, [S]m ) [S]tot - cmc. The observed fluorescence intensity should be described as, I ) Im + Iaq

Im ) A(τm/τ0)[MQ]m

(2)

Iaq ) A(τaq/τ0)[MQ]aq ) A(1 + KSV[X-])-1[MQ]aq ) A{1 + KSV(cmc + R[S]m)}-1[MQ]aq (3) where A is a proportionality constant, τ0 is the lifetime of fluorescence probe in the absence of quencher, τm and τaq are the lifetimes in the micelle and aqueous bulk phases, respectively, [X-] is the concentration of free halide ions in the aqueous solution, and the Stern-Volmer expression has been utilized. R is the degree of micellar counterion dissociation. [X-] is given by the sum of cmc and the concentration of dissociated counterion from micelles. The Stern-Volmer intensity ratio becomes

I0/I ) A[MQ]tot/{A(τm/τ0)[MQ]m + A(τaq/τ0)[MQ]aq} (4)

with that using the conductivity method. The decrease in slope of the Stern-Volmer plot above the cmc can be ascribed to the counterion binding of the cationic micelles. This behavior is in accord with that using MQAE probe as reported previously.11 The MEQ fluorescence is quenched by the free halide ions dissociated from the surfactant in the aqueous bulk phase as shown in Scheme 1. In contrast, the amphiphilic FC10MQ fluorescence is effectively quenched by the halide ions at the micellar surface. In the pseudophase approximation the fluorescence probe is regarded as partitioned between a micelle phase and an aqueous bulk phase according to eq 1.

K)

[MQ]m [MQ]aq[S]m

(1)

MQ and S are the fluorescence probe and surfactant, subscripts m and aq mean in micelle phase and aqueous bulk phase, respectively. K is the distribution constant,

If we assume that the intensity from the probes in the micelles is negligible, the equation is simplified. As long as the concentration of probe in the micelles is not much larger than that in the aqueous phase, the validity of the assumption can be ascertained from the lifetimes.

I0/I ) (1 + K[S]m) {1 + KSV(cmc + R[S]m)}

(5)

With R and the cmc determined from measurements on an aqueous probe (MEQ), and the Stern-Volmer constant from measurements below the cmc, the simulation of the data above the cmc should give the distribution constant K. The variations of I0/I for FC10MQ and C12MQ were simulated using K ) 1.2 × 103 and 22 mol-1 dm3, respectively, as shown in Figure 2. HFDePC is much favorable for solubilizing FC10MQ. FC10MQ probe was tried with hydrocarbon surfactant systems in order to further clarify the relation between the fluorescence quenching behavior and the immiscibility of fluorocarbon and hydrocarbon species. Figure 3 shows the quenching behavior of FC10MQ fluorescence in alkyltrimethylammonium bromide systems. The fluorescence is quenched by NaBr and follows a linear Stern-Volmer

Micellar Immiscibility by Fluorescence Quenching

Figure 4. Fluorescence decay curves from FC10MQ in the absence and presence of TTAB: (a) 0; (b) 2.5; (c) 7.5; (d) 10 mM TTAB. Table 1. The Critical Micelle Concentrations, the Degree of Micellar Counterion Dissociation, and the Distribution Constant between Micelle and Aqueous Bulk Phases MEQ

FC10MQ

surfactant

cmc (mM)

a

C8F17CH2CH2NC5H5Cl (HFDePC) C14H29N(CH3)3Cl (TTAC) C12H25N(CH3)3Br (DTAB) C14H29N(CH3)3Br (TTAB) C16H33N(CH3)3Br (CTAB)

2.6 5.5 16.0 3.7 1.0

0.23 0.18 0.23 0.16 0.13

cmc K (mM) (mol-1 dm3) 2.3 5.5 15.1 4.1 1.0

1.2 × 103 3.1 × 102 5.7 × 102 3.1 × 102 2.6 × 102

plot. The fluorescence behavior for surfactant systems below the cmc is similar to that in NaBr aqueous solution, whereas an abrupt increase in quenching was observed above the cmc. Interestingly, the slope for the SternVolmer plot increased with decreasing length of the alkyl chain. The behavior was simulated by the increase of distribution constant (K) between micelles and aqueous bulk phases. The cmc values, the degree of micellar counterion dissociation (R), and the distribution constant (K) are summarized in Table 1. The decreases of cmc and R with increasing length of the alkyl chain are generally in agreement with those reported previously.11 The cmc values determined by FC10MQ gave good agreement with those determined by MEQ. The FC10MQ probe was much more solubilized with HFDePC micelles than with TTAC ones by a factor of about 4. The K values decreased with increasing length of the alkyl chain. This suggests the immiscibility of fluorocarbon chain of probe and hydrocarbon chain of surfactants. Measurements of the fluorescence decay rates of FC10MQ in aqueous micellar solutions were attempted to distinguish between different environments of the probe. Figure 4 shows the FC10MQ fluorescence decay in the absence of and presence of TTAB. The lifetime of FC10MQ in aqueous solution was 26.2 ns, which is similar to 25.3 ns of SPQ.14 The fluorescence lifetimes of FC10MQ decreased with increasing concentration of TTAB. The fluorescence decay curves for FC10MQ are monoexponential below the cmc of TTAB as seen in (a) and (b), whereas they appear to be biexponential in the presence of TTAB micelles as seen in (c) and (d). The shorter lifetime corresponds to decay in the micellar environment, while the longer lifetime can be attributed to decay in the aqueous bulk phase. The observed fluorescence intensity,

Langmuir, Vol. 20, No. 17, 2004 7001

Figure 5. Stern-Volmer plots for the longer lifetime component of FC10MQ in comparison with plots of fluorescence intensities: (O) τ0/τ; (b) I0/I.

F(t), can be expressed by the following equation as a function of time

F(t) ) B1 exp(-t/τm) + B2 exp(-t/τaq)

(6)

where τm and τaq are the lifetimes of FC10MQ in the micelle and aqueous bulk phases, respectively, and B values are the amplitudes. Since free bromide ions quench FC10MQ fluorescence in the bulk, the longer lifetimes decreased with increasing TTAB concentration, i.e., 14.3, 11.0, and 10.4 ns in 2.5, 7.5, and 10 mM TTAB, respectively. The initially rapid decay of the quenched curve reflects the short average distance between the probe and bromide ions at the micellar surface. The fluorescence lifetimes in micelles were 0.2 ns or less. It was difficult to estimate accurately the extent of FC10MQ partitioning between the micellar and aqueous bulk phases. When quenching occurs by the collisional interaction of the fluorescence probe and halide ion quencher, the variation of fluorescence intensity is related to the concentration of quencher [Q] by the Stern-Volmer equation

τ0/τ ) I0/I ) 1 + KSV[Q]

(7)

where τ0 and I0 and τ and I are the fluorescence lifetimes and intensities in the absence and presence of quencher, respectively, and KSV is the Stern-Volmer constant. Figure 5 shows the Stern-Volmer plots for the longer lifetime component of FC10MQ in comparison with plots of fluorescence intensities as a function of TTAB concentration. The slope in τ0/τ plot above the cmc was smaller than that below the cmc. The behavior can be ascribed to the micellar counterion binding of bromide in similar to an aqueous MEQ because the determined τ0/τ correspond to the ratio of longer lifetimes for FC10MQ in the aqueous bulk phase. On the other hand, the slope in I0/I plot above the cmc was significantly larger than that below the cmc owing to the effective quenching of micelle solubilized FC10MQ. Thus the fluorescence quenching behavior demonstrates the miscibility of FC10MQ probe and TTAB micelles. On the basis of these results, the fluorescence quenching behavior enables us to detect the micelle formation in fluorocarbon and hydrocarbon surfactant mixtures. An abrupt increase in quenching of FC10MQ fluorescence will correspond to the formation of fluorocarbon-rich micelles. Stern-Volmer plots for quenching of FC10MQ in a series of TTAC-HFDePC mixtures are shown in Figure 6. The Stern-Volmer constant below the cmc was 247 M-1, which

7002

Langmuir, Vol. 20, No. 17, 2004

Asakawa et al.

Figure 6. FC10MQ fluorescence quenching in TTAC-HFDePC mixtures: (b) TTAC; (4) XF 0.20; (0) XF 0.33; (3) XF 0.43; (O) HFDePC.

Figure 8. C12MQ fluorescence quenching in TTAC-HFDePC mixtures: (a) (O) XF 0.10, (4) XF 0.30, (O) XF 0.60; (b) 15 mM TTAC-HFDePC. Figure 7. MEQ fluorescence quenching in TTAC-HFDePC mixtures: (b) TTAC; (4) XF 0.42; (O) HFDePC.

was constant over all compositions. This suggests almost no effect on fluorescence quenching by the surfactant monomers. The fact that there is no difference in quenching between TTAC and HFDePC monomers indicates that the fluorescence probe is selective and sensitive only to free chloride ions and not those bound by the surfactant micelles. The slope above the cmc increased with increasing mole fraction of HFDePC. Above a mole fraction of HFDePC ) 0.33 the slope became similar to that of pure HFDePC and was almost constant with increasing total surfactant concentrations. This behavior would correspond to the formation of fluorocarbon-rich micelles with a micelle composition that remained constant with increasing total surfactant concentrations. The quenching behavior seems to indicate formation of almost pure HFDePC micelles even in mixtures. Precise cmc measurements for the mixtures are also necessary to estimate the miscibility of TTAC-HFDePC in micelles. The MEQ fluorescence quenching behavior gave the cmc values of TTAC-HFDePC mixtures as shown in Figure 7. For a mole fraction of HFDePC ) 0.42, the first break point corresponds to cmc (4.0 mM), and a second break point was observed far above the cmc. Above the second break point, the slope was similar to that in a pure TTAC system. This behavior can be explained as follows. At this composition, the HFDePC-rich micelles first appear at the mixture cmc because of the low cmc of HFDePC. As the total concentration increases at fixed composition, the monomer concentration of TTAC increases and

becomes constant above the second cmc where TTAC-rich micelles are formed. Thus the slope between the first and second cmc was larger than that of a single surfactant system owing to the increase in monomer concentration. Above the second cmc, the monomer concentrations remained constant because the monomers are in equilibrium with two kinds of mixed micelles with fixed compositions. Thus the slope above the second cmc reflects the degree of micellar counterion dissociation for the two kinds of mixed micelles, which was similar to that of pure micelles. The amphiphilic fluorescence probe with hydrocarbon chain (C12MQ) was also used in the TTAC-HFDePC mixtures. Since the C12MQ probe was scarcely solubilized in HFDePC micelles as shown in Figure 2, the abrupt quenching of C12MQ fluorescence can be expected only if TTAC-rich micelles appear in the solution. The SternVolmer plots for quenching of C12MQ fluorescence are shown in Figure 8a. The break point for the mole fraction 0.1 of HFDePC corresponds to the mixture cmc, whereas the break point for the mole fraction 0.6 of HFDePC occurs far above the mixture cmc determined using the MEQ probe. The quenching behavior of C12MQ fluorescence was also measured as a function of the mole fraction of HFDePC at a fixed concentration of 15 mM TTAC-HFDePC as shown in Figure 8b. Effective quenching occurred in the TTAC-rich region. The plot of I0/I vs mole fraction of HFDePC gave a linear relation for mole fractions of HFDePC between 0.16 and 0.71. The linearity suggests a decrease in concentration of TTAC-rich micelles with increasing mole fraction of HFDePC. The TTAC-rich micelles solubilize the C12MQ probe as shown schematically.

Micellar Immiscibility by Fluorescence Quenching

Langmuir, Vol. 20, No. 17, 2004 7003

under the assumption of coexistence of two kinds of mixed micelles with fixed compositions, that is

cmc2 ) CAZ

XF - XAZ XF - x

(8)

XAZ - XH cmc2 ) CAZ x - XH

Figure 9. Micellar pseudophase diagram for TTAC-HFDePC mixtures: cmc (O) MEQ, (4) C12MQ, (0) FC10MQ; second cmc (b) MEQ, (2) C12MQ. The solid line is the cmc curve predicted by the group contribution method.17 The dashed line is the calculated second cmc if micellar demixing occurs between XH ) 0.11 and XF ) 1.

On the basis of fluorescence data using the different quinoline derivatives, a micellar pseudophase diagram of the TTAC-HFDePC system was constructed as shown in Figure 9. The HFDePC-rich micelles first appear at a mole fraction of HFDePC between 0.3 and 1 owing to the low cmc of HFDePC. As the total surfactant concentration increases, the concentration of monomeric TTAC increases until the second cmc, at which TTAC-rich micelles appear owing to the immiscibility of fluorocarbon and hydrocarbon surfactants. The position of the second cmc may be calculated from the material balance of the two surfactants

where CAZ and XAZ are the concentration and composition under azeotropic conditions, XH and XF are the compositions of HFDePC in hydrocarbon-rich and fluorocarbonrich micelles, respectively, and x is the composition of fluorocarbon surfactant in mixture. The larger value of cmc2 in eq 8 leads to the second cmc.5 From a comparison of measured and simulated second cmc data, the micellar compositions are XH ) 0.11 and XF ) 1. TTAC-rich micelles solubilize HFDePC to a certain extent, whereas the HFDePC-rich micelles hardly solubilize TTAC at all. This trend in mutual solubilities was in accord with the results reported previously.19 It should be noted that the HFDePC micelles almost without hydrocarbon species are revealed not only by the analysis of the micellar pseudophase diagram but also by the fluorescence quenching behavior of C12MQ in the HFDePC system. In contrast, the TTAC-rich micelles solubilized fluorocarbon species to a certain extent in aqueous surfactants mixtures. The fluorescence quenching method designed in this paper demonstrated directly the appearance of almost pure HFDePC micelles. LA049488Y (19) Kadi, M.; Hansson, P.; Almgren, M.; Furo, I. Langmuir 2002, 18, 9243.