Electrochemical, Microscopic, and Spectroscopic Characterization of

Ionic liquid induced alterations in the physicochemical properties of aqueous solutions of sodium dodecylsulfate (SDS). Amalendu Pal , Sheena Chaudhar...
0 downloads 0 Views 738KB Size
Download PDF
Langmuir 2006, 22, 4027-4036

4027

Electrochemical, Microscopic, and Spectroscopic Characterization of Prevesicle Nanostructures and Vesicles on Mixed Cationic Surfactant Systems Emilio Aicart,† Patricia del Burgo,† Oscar Llorca,‡ and Elena Junquera*,† Departamento de Quı´mica Fı´sica I, Facultad de Ciencias Quı´micas, UniVersidad Complutense de Madrid, 28040-Madrid, Spain, and Centro de InVestigaciones Biolo´ gicas, CSIC, Ramiro de Maeztu 9, 28040 Madrid, Spain ReceiVed December 23, 2005. In Final Form: February 23, 2006 Several experimental techniques (conductivity, zeta potential, transmission electronic microscopy, and steady-state fluorescence spectroscopy) have been used to study the formation of mixed colloidal aggregates consisting of a cationic double-chain surfactant, di-dodecyldimethylammonium bromide (di-C12DMAB), and a single-chain alkyltrimethylammonium bromide with 10 and/or 14 carbon atoms (decyltrimethylammonium bromide, C10TAB, and/or tetradecyltrimethylammonium bromide, C14TAB). Special interest has been devoted to the prevesicle domain, within which the formation of aggregated nanostructures was first reported in our laboratory. For that purpose, studies have been carried out on the very dilute region by means of conductivity experiments, confirming the existence of two critical aggregation concentrations in that concentration domain: the so-called mixed critical aggregate concentration, CAC*, and the mixed critical vesicle concentration, CVC*. By carrying out TEM experiments on negatively stained samples, we were surprised to find a number of aggregates without a clear aggregation pattern and with a variety of sizes and shapes at concentrations below CAC*, where only monomers were expected. However, the nanoaggregates found at concentrations between CAC* and CVC*, also by TEM microscopy, show a clear and ordered “fingerprint”like aggregation pattern similar to the liquid-crystalline phases reported for DNA-liposome complexes and/or DNA packed with viral capsids. Finally, at total surfactant concentrations above CVC*, the aggregates were confirmed, by means of cryo-TEM micrographs and zeta potential measurements, to be essentially unilamellar spherical vesicles with a medium polydispersity and a net-averaged surface density charge of around 12 × 10-3 C m-2. The fluorescence emission of two probes, TNS (anionic) and PRODAN (nonionic), allows for the analysis of the micropolarity and microviscosity of the different microenvironments present in aqueous surfactant solutions where the above-mentioned vesicle and prevesicle aggregates are present.

I. Introduction Mixed colloidal systems have revealed in past decades to be powerful and interesting tools in industry and research.1-3 It is well known that the mixture of two or more single-chain surfactants drives the formation of mixed micelles, whereas mixed vesicles are the favored aggregates when the surfactants have two hydrophobic tails.1-5 However, the addition of a singlechain surfactant to a double-chain surfactant/water system drives the formation of a variety of mixed aggregates whose selforganization pattern and/or stability depends very much, among other factors, on the composition of the mixed system and/or the total surfactant concentration.6-8 Most studies reported in the literature deal with the medium-concentration domain where three aggregation phenomena occur: (i) the formation of mixed * To whom correspondence should be addressed. E-mail: [email protected]. http://www.ucm.es/info/coloidal/index.html. Fax: 34-913944135. † Universidad Complutense de Madrid. ‡ CSIC. (1) Holland, P. M.; Rubingh, D. N. Mixed Surfactant Systems; American Chemical Society: Washington, DC, 1992. (2) Christian, S. D.; Scamehorn, J. F. Solubilization in Surfactant Aggregates; Marcel Dekker: New York, 1995; Vol. 55. (3) Somasundaran, P.; Hubbard, A. Encyclopedia of Surface and Colloid Science.; Marcel Dekker: Santa Barbara, CA, 2002. (4) Tanford, C. J. Phys. Chem. 1974, 78, 2469. (5) Rosoff, M. Vesicles; Marcel Dekker: New York, 1996. (6) Junquera, E.; Arranz, R.; Aicart, E. Langmuir 2004, 20, 6619. (7) Junquera, E.; del Burgo, P.; Arranz, R.; Llorca, O.; Aicart, E. Langmuir 2005, 21, 1795. (8) Junquera, E.; del Burgo, P.; Boskovic, J.; Aicart, E. Langmuir 2005, 21, 7143.

vesicles at a total surfactant concentration above the critical mixed vesicle concentration, CVC*; (ii) the mixed micelles start to form and coexist with mixed vesicles at the critical mixed micelle concentration, CMC*; and (iii) the mixed vesicles are disrupted and solubilized by the mixed micelles at the total critical mixed micelle concentration, CMC/tot. This gradual solubilization of vesicle structures, due to micelle formation, points to these mixed aggregate systems as suitable models to mimic the solubilization and reconstitution processes of lipidic bilayers.9-15 Nevertheless, very few data have been reported on what happens at concentrations below the CVC* (in the extremely dilute concentration range in what we called the prevesicle domain), with the exception of our previous works,6-8 where another critical concentration, the so-called critical mixed aggregation concentration or CAC*, was presented. Those works showed that a different kind of colloidal aggregate, intermediate between the monomeric and the vesicle stages, appears.6,8 Nothing was said, however, about the aggregation pattern and/or structure of these new aggregates. With this background, we face in the present work a physical chemistry study of aqueous solutions of (9) Lichtenberg, D. Biochim. Biophys. Acta 1985, 821, 470. (10) Engberts, J. B. F. N.; Kevelam, J. Curr. Opin. Colloid Interface Sci. 1996, 1, 779. (11) Ollivon, M.; Lesieur, S.; Grabielle-Madelmont, C.; Paternostre, M. Biochim. Biophys. Acta 2000, 1508, 34. (12) Deo, N.; Somasundaran, P. Langmuir 2003, 19, 2007. (13) Majhi, P. R.; Blume, A. J. Phys. Chem. B 2002, 106, 10753. (14) Kawasaki, H.; Imahayashi, R.; Tanaka, S.; Almgren, M.; Karlsson, G.; Maeda, H. J. Phys. Chem. B 2003, 107, 8661. (15) Tan, A. M.; Ziegler, A.; Steinbauer, B.; Seelig, J. Biophys. J. 2002, 83, 1547.

10.1021/la053474q CCC: $33.50 © 2006 American Chemical Society Published on Web 03/29/2006

4028 Langmuir, Vol. 22, No. 9, 2006 Chart 1. (a) Di-C12DMAB, (b) C10TAB, (c) C14TAB, (d) TNS, and (e) PRODAN Molecules

cationic mixed systems consisting of a double-chain surfactant, di-dodecyldimethylammonium bromide (di-C12DMAB), commercially used in a wide variety of applications,16 and an alkyltrimethylammonium bromide with 10 and/or 14 carbon atoms on the hydrophobic chain (Chart 1). Although part of this work is focused on the characterization of the mixed vesicles, special attention has been paid to the study of the aggregation phenomena that take place prior to vesicle formation. We try not only to detect these prevesicle aggregates but also to shed light on their aggregation pattern, structure, and so on. The critical aggregation concentrations, CVC*and CAC*, have been determined from conductivity experiments. The structure and morphology of the aggregates have been analyzed by using cryogenic transmission electron microscopy (cryo-TEM) and/or transmission electron microscopy on negatively stained samples (TEM). The vesicle/solution interface has also been studied by means of the zeta potential and the charge density of the aggregate surface, obtained from experimental electrophoretic mobility data. Additionally, the interaction of the anionic probe 2-(p-toluidino)naphthalene-6-sulfonic acid (TNS) and/or the nonionic probe 6-propionyl-2-dimethylaminonaphthalene (PRODAN) (Chart 1) with autoaggregated surfactant systems is well documented in the literature.17-24 The steady-state fluorescence emission of the above-mentioned probes has been analyzed in the presence of the mixed aggregates herein studied with the aim of further characterizing their interior as well as their surfaces. II. Experimental Section A. Materials. Di-C12DMAB and C14TAB, with purities of 99% or greater, were purchased from Aldrich Co, and C10TAB, with a (16) Jungermann, E. Cationic Surfactants; Marcel Dekker: New York, 1970. (17) Karukstis, K. K. Encapsulation of Fluorophores in Multiple MicroenVironments in Surfactant-Based Supramolecular Assemblies; Academic Press: London, 2001; Vol. 3. (18) Niu, S.; Gopidas, K. R.; Turro, N. J.; Gabor, G. Langmuir 1992, 8, 1271. (19) Parusel, A. B.; Nowak, W.; Grimme, S.; Ko¨hler, G. J. Phys. Chem. 1998, 102, 7149. (20) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic/Plenum: New York, 1999. (21) Krishnamoothy, G.; Dogra, S. K. Phys. Chem. Chem. Phys. 2000, 2, 2521. (22) Zhong, D.; Kumar-Pal, S.; Zewail, A. H. Chem. Phys. Chem. 2001, 2, 219. (23) Lobo, B. C.; Abelt, C. J. J. Phys. Chem. A 2003, 107, 10938. (24) Karukstis, K. K.; McCormack, S. A.; McQueen, T. M.; Goto, K. F. Langmuir 2004, 20, 64.

Aicart et al. purity of 98% or greater, was purchased from Fluka. The potassium salt of 2-(p-toluidino)naphthalene-6-sulfonic acid (TNS) is from Sigma, and 6-propionyl-2-dimethylaminonaphthalene (PRODAN) is from Molecular Probes. All were used without further purification. Distilled water was deionized using a Super Q Millipore system (with a conductivity lower than 18 µS cm-1) and finally was also degassed with a vacuum pump prior to the preparation of the solutions. B. Preparation of the Samples. Appropriate amounts of both components of the ternary systems, di-C12DMAB/C10TAB and diC12DMAB/C14TAB, were weighted and dissolved simultaneously in water to prepare a solution of a given total surfactant concentration at the desired mole fraction. These initial solutions, after being sonicated for 2 h in an ultrasonic bath, were homogeneous and totally clear on the visible scale. Their long-term thermodynamic stability was checked by confirming that conductivity, κ, and zeta potential, ζ, data remained unchanged 4 months later, within the experimental uncertainty. C. Conductometric Measurements. Conductivity data were collected at 298.15 K ((1 mK) with a Hewlett-Packard 4263A LCR meter. The equipment, preparation of mixtures, and fully computerized procedure were described previously.25 The accuracy of the specific conductivity, κ, obtained as an average of 2400 measurements for each concentration, is better than 0.03%. The conductivity measurements were made as a function of total surfactant concentration, [S]tot () [di-C12DMAB] + [CnTAB], n ) 10 or 14), at several constant values of the mole fraction R1. D. ζ-Potential Measurements. A laser Doppler electrophoresis (LDE) technique (Zetamaster 2000, Malvern Instruments Ltd.), which was previously described,6 was used to measure electrophoretic mobilities. The cell used is a Zetasizer 2000 standard quartz rectangular capillary electrophoresis cell of 5 × 2 × 50 mm3, which is calibrated with a zeta potential transfer standard of ζ ) (-50 ( 5) mV. Temperature was controlled at (298.15 ( 0.01 K). Each electrophoretic mobility data are taken as an average over 10 independent measurements. ζ potentials were calculated from the electrophoretic mobility measurements made at the mole fractions studied for conductivity experiments, most of them at two different [S]tot values per mole fraction. E. cryo-TEM Measurements. Samples for cryogenic transmission electron microscopy were initially prepared as explained in section B and vitrified according to the method devised by Dubochet et al.26 and also described in Llorca et al.27 In these experiments, 20 min of additional sonication was necessary. Then, 5 µL aliquots of the different samples were applied to glow-discharge carbon-coated grids for 1 min, blotted for 5 s, and frozen rapidly in liquid ethane at -180 °C and kept at this temperature throughout the whole procedure. Observations were conducted at a JEOL 1230 microscope operated at 100 kV. Micrographs were recorded on Kodak 4489 film at 0° tilt and a nominal magnification of 30 000×. Selected micrographs were digitalized in a Dimage Scan Multi Pro scanner (Minolta) with a sampling window of 7 Å/pixel at the specimen. Sample concentrations, [S]tot, were chosen, with the help of conductivity data, above CVC* to guarantee the presence of vesicles. F. TEM Measurements. TEM experiments were done for negatively stained samples, which were prepared following the method proposed by Kunitake and Okahata.28 One milliliter of sample solution (section B) was mixed with 1 mL of 2% (w/v) uranyl acetate aqueous solutions. The resulting mixtures were then sonicated in a water bath for 20 s, followed by incubation in ice for another 30 min. They were subsequently applied to glow-discharged copper-rhodium electron microscopy grids supporting a thin carbon film, blotted after 2 min, and air dried. Observations were conducted at a JEOL 1230 microscope operated at 100 kV. Micrographs were recorded on Kodak 4489 film at 0° tilt and a nominal magnification of 60 000× (25) Junquera, E.; Aicart, E. ReV. Sci. Instrum. 1994, 65, 2672. (26) Dubochet, J.; Adrian, M.; Chang, J. J.; Homo, J. C.; Lepault, J.; McDowall, A. W.; Schultz, P. Q. ReV. Biophys. 1988, 21, 129. (27) Llorca, O.; McCormack, E.; Hynes, G.; Grantham, J.; Cordell, J.; Carrascosa, J. L.; Willison, K. R.; Ferna´ndez, J. J.; Valpuesta, J. M. Nature 1999, 402, 693. (28) Kunitake, T.; Okahata, Y. J. Am. Chem. Soc. 1980, 102, 549.

PreVesicle Nanostructures and Vesicles

Langmuir, Vol. 22, No. 9, 2006 4029 Table 1. Values of [S]tot, CAC*, CVC*, KDa, f(KDa), ζ, and σζ at Several Mole Fractions r1 of the Mixed System di-C12DMAB (1) + C10TAB (2) R1

[S]tot (mM)

CAC* (mM)

CVC* (mM)

κDa

f(κDa)

ζ (mV)

103σζ (C/m2)

0.200 0.401 0.500 0.647 0.847 1.000

0.797 1.097 1.458 1.904 2.145 0.305

0.348 0.405 0.441 0.496 0.598

0.441 0.524 0.615 0.823 1.740 0.165

1.7 2.0 2.3 2.6 2.8 1.0

1.06 1.07 1.08 1.09 1.0 1.04

86 ( 2 83 ( 2 84 ( 3 79 ( 3 76 ( 3 88 ( 9

11 ( 2 12 ( 2 14 ( 2 14 ( 2 14 ( 2 9(2

Table 2. Values of [S]tot, CAC*, CVC*, KDa, f(KDa), ζ, and σζ at Several Mole Fractions r1 of the Mixed System di-C12DMAB (1) + C14TAB (2) R1 0.196 0.397 0.496 0.651

Figure 1. Plot of specific conductivity, κ, as a function of total surfactant concentration, [S]tot, in the very dilute range at 298.15 K at constant mole fraction, R1, for the mixed system di-C12DMAB (1) + C10TAB (2). under low-dose conditions. Selected micrographs were digitized in a Dimage Scan Multi Pro scanner (Minolta) at 800 dpi corresponding to a sampling window of 5.3 Å/pixel at the specimen. Sample concentrations, [S]tot, were chosen with the help of conductivity data, considering that it must be within the range for which the prevesicle aggregates are present in solution, either below CAC* or at CAC* < [S]tot < CVC*. A blank sample, consisting of a 1:1 mixture of an aqueous solution and uranyl acetate and sonicated and treated with the same protocol as for the actual sample, was also used as a control test to be observed under the electron microscope. G. Fluorescence Measurements. Steady-state fluorescence experiments were carried out with a Perkin-Elmer LS-50B luminescence spectrometer29 connected to a PC. A 10 mm stoppered rectangular silica cell was placed in a stirred cuvette holder whose temperature was kept constant at 298.15 ( 0.01 K. Following a previously described procedure,8 probe concentrations are kept constant at [TNS] ) 2.07 µM and/or [PRODAN] ) 5.02 µM, and the surfactant concentrations range from pre- to postvesicle values at each mole fraction of the ternary system. All surfactant/probe solutions thus prepared were first stirred for 2 to 3 h and were subsequently allowed to equilibrate for more than 60 h prior to running the spectra. Excitation wavelengths of 330 and 340 nm were chosen, and emission wavelengths were varied from 340 to 600 nm and from 350 to 650 nm to record the fluorescence emission of TNS and PRODAN, respectively. In all cases, excitation and emission band slits were fixed at 2.5 nm, and the scan rate was 240 nm/min.

III. Results and Discussion Electrochemical Study. Conductivity experiments were run to determine the concentrations at which possible structural changes, such as those accompanying aggregation phenomena, occur in aqueous solutions. These measurements have been carried out in the very dilute concentration range as a function of the total surfactant concentration for a series of runs where the mole fraction was kept constant. Figure 1 shows a plot of specific conductivity versus total surfactant concentration for three compositions of the mixed system di-C12DMAB (1) + C10TAB (29) Junquera, E.; Pen˜a, L.; Aicart, E. Langmuir 1997, 13, 219.

0.751 0.849 1

[S]tot (mM)

CAC* (mM)

CVC* (mM)

1.001 0.626 1.283 0.800 1.591 0.996 2.338 1.461 1.951 2.905 0.305

0.334

0.452

0.415

0.560

0.454

0.624

0.513

0.884

0.578 0.611

1.197 1.992 0.165

κDa

f(κDa)

ζ (mV)

103σζ (C/m2)

2.6 2.1 2.9 2.3 3.3 2.6 4.0 3.1 3.6 4.4 1.0

1.09 1.07 1.09 1.08 1.10 1.08 1.12 1.10 1.11 1.13 1.04

90 ( 2 83 ( 1 88 ( 3 79 ( 2 82 ( 4 79 ( 2 85 ( 3 77 ( 2 71 ( 3 76 ( 2 89 ( 9

12 ( 2 9(2 13 ( 2 9(2 13 ( 2 10 ( 2 16 ( 3 11 ( 2 11 ( 2 15 ( 3 9(2

(2) as an example: (i) with a deficit in the double-chain surfactant (R1 ) 0.200); (ii) the equimolecular composition (R1 ) 0.500); and (iii) with a deficit in the single-chain surfactant (R1 ) 0.847). In all cases, the critical concentrations, calculated by means of the third derivative method,31 are again used for the systems studied in this work in Tables 1 and 2. The effect of composition is clear; only one critical concentration is detected for the pure components (R1 ) 0, CMC is 66.5 mM for C10TAB32 and 3.63 mM C14TAB;33 R1 ) 1, CVC is 0.165 mM for di-C12DMAB6), whereas two changes are found in the very dilute concentration range for the mixed system, CAC* and CVC*. Furthermore, the conductivity plot shows a zigzag pattern, which is more evident as long as the mole fraction increases, as can be clearly observed in Figure 1. If attention is paid to the concentration range between CAC* and CVC*, then a slight decrease in conductivity can be observed at mole fractions greater than 0.6 or 0.7. We do think that it may be due to either the degree of dissociation readjustments and/or structural changes with increasing total surfactant concentrations, as will be discussed later. This zigzag pattern was first found by our group for another cationic mixed system, di-C12EDMAB + C12EDMAB,6 but it was not present in the cationic-nonionic mixed systems, di-C12DMAB + OBG and di-C10DMAB + OBG, or for the cationic di-C10DMAB + C12EDMAB system.7,8,34 We do believe that the occurrence of the zigzag on conductivity plots, which means that there is a detectable concentration range within which the prevesicle aggregates exist, has nothing to do with the cationic or nonionic character of the surfactants; the studies support the fact that it does depend on whether the critical concentrations CVC* and CMC* are close to each other. When these critical concentrations are separated enough (in concentration units), the zigzag is detected, whereas the contrary is true when they are close. Also observable in Figure 1 is that the concentration region within which prevesicle (30) Lichtenberg, D.; Robson; Dennis, E. A. Biochim. Biophys. Acta 1983, 737, 285. (31) Phillips, J. N. Trans. Faraday Soc. 1955, 51, 561. (32) Junquera, E.; Aicart, E.; Tardajos, G. J. Phys. Chem. 1992, 96, 4533. (33) Lainez, A.; del Burgo, P.; Junquera, E.; Aicart, E. Langmuir 2004, 20, 5745. (34) del Burgo, P.; Aicart, E.; Junquera, E. Colloids Surf., A, 2005, submitted for publication.

4030 Langmuir, Vol. 22, No. 9, 2006

Aicart et al.

Figure 2. Plot of CAC* (open symbols) and CVC* (solid symbols) as a function of mole fraction R1 for the mixed systems consisting of (1) di-C12DMAB and (2) C10TAB (squares), C14TAB (circles), or C12EDMAB (triangles, data taken from ref 6). The inset show a plot of the prevesicle aggregate domain as a function of R1 for di-C12DMAB + C10TAB (solid diamonds) and di-C12DMAB + C14TAB (open diamonds).

aggregates exist (i.e., between CAC* and CVC*) increases with R1, as can be seen in Figure 2, where both critical concentrations and/or the prevesicle concentration domain (inset) are plotted as a function of the mole fraction for different cationic systems. This feature confirms that either prevesicle aggregates or vesicles tend to form at higher concentrations, as long as the total content in the double-chain surfactant increases. It is also clearly inferred from the Figure that neither the length of the hydrophobic tail nor the polar head of the single-chain surfactant affects the concentration limits of the monomer, prevesicles aggregates, and/or vesicle domains. The analysis of the effect of the length of the single-chain surfactant tail on aggregation parameters can be also followed in Figure 3, which shows a plot of the specific conductivity versus total surfactant concentration for the equimolecular solution in the very dilute region (from 0 to 1.5 mM). For that analysis, this experiment was also run for the system di-C12DMAB + C16TAB under the same conditions of concentration and mole fraction. Also included in this Figure are the data previously reported for the mixed system di-C12DMAB + C12EDMAB.6 This Figure deserves some remarks: (i) the zigzag pattern observed for di-C12DMAB + CnTAB with n ) 10, 12, or 14 is clearly lost for n ) 16, which confirms that this is the difference among the critical concentrations CVC* and CMC* that conditions the appearance of the zigzag; and (ii) the substitution of a methyl group by an ethyl group on the polar head of the single-chain surfactant does not affect the aggregation behavior. The electrochemical study was completed with the analysis of the charge on the aggregates’ outer surfaces by means of their ζ potentials. ζ potentials were calculated from the experimental electrophoretic mobilities by using the Henry equation35,36

ζ)

3η µ 20r f(κDa) E

(1)

where the permittivity of the vacuum, 0, the relative permittivity

Figure 3. Plot of specific conductivity, κ, as a function of total surfactant concentration, [S]tot, in the very dilute range at 298.15 K at constant mole fraction R1 around 0.5 for the mixed systems consisting of di-C12DMAB and C10TAB, C14TAB, or C16TAB. Also included are the data for di-C12DMAB + C12EDMAB, taken from ref 6.

of the medium, r, and the viscosity of water, η, were taken to be 8.854 × 10-12 J-1 C2 m-1 and 78.36 and 8.904 × 10-4 N m-2 s, respectively. The factor f(κDa), which is called the Henry function, depends on the reciprocal Debye length, κD, and the particle radius, a, and has been estimated by the Ohshima equation,37 with relative errors of less than 1%:

f(κa) )

[

2 1+ 3

(

2 1+

1 2.5 κa(1 + 2e-κa)

)] 3

(2)

A value of a ) 17.5 nm has been used for pure di-C12DMAB vesicles,6 whereas in the case of mixed di-C12DMAB/C10TAB and/or di-C12DMAB/C14TAB vesicles averaged aggregate radii were taken from the cryo-TEM data reported in this work. Tables 1 and 2 list κDa, f(κDa) and ζ-potential values calculated at several mole fractions R1 for the mixed systems studied herein. Notice that the ζ potential is almost constant with R1, around (83 ( 6) and (82 ( 5) mV, for di-C12DMAB/C10TAB and di-C12DMAB/ C14TAB systems, respectively, within the experimental error. This value is comparable to that obtained for other cationic and/ or cationic-nonionic mixed vesicles previously studied in our laboratory,6-8 revealing that the ionic polar head of the dialkylsurfactant governs the zeta potential of the interface. The surface charge density enclosed by the shear plane σζ has been calculated, assuming a Gouy-Chapman double layer using38

σζ )

[ ( )

( )]

20rκDkBT zeζ zeζ 2 tanh sinh + ze 2kBT κDa 4kBT

(3)

where e is the elemental charge, z is the valence of the ion, kB (35) Hunter, R. J. Zeta Potential in Colloids Science: Principles and Applications; Academic Presss: London, 1981. (36) Delgado, A. V. Interfacial Electrokinetics and Electrophoresis; Marcel Dekker: New York, 2002; Vol. 106. (37) Ohshima, H.; Furusawa, K. Electrical Phenomena at Interfaces: Fundamentals, Measurements, and Applications; Marcel Dekker: New York, 1998.

PreVesicle Nanostructures and Vesicles

Langmuir, Vol. 22, No. 9, 2006 4031

Figure 4. (a and b) TEM fields of a di-C12DMAB + C10TAB solution at R1 ) 0.500 and [S]tot ) 0.272 mM after negative staining with uranyl acetate. Scale bar: 50 nm. (c) The image shown in panel b is magnified to highlight the structural details within the aggregates. Scale bar: 25 nm.

is the Boltzmann constant, and T is the absolute temperature (Tables 1 and 2). As can be seen, σζ values remain almost constant with R1, with an average value of (12 ( 3) × 10-3 C m-2 for both di-C12DMAB/C10TAB and di-C12DMAB/C14TAB vesicles. This σζ value is in agreement with those reported for other mixed vesicles studied in our laboratory.6-8 However, it seems that σζ is slightly higher for cationic-nonionic mixed vesicles.8 This feature may point to a balance between two possible effects: the presence of nonionic molecules on the aggregate surface may shield the electrostatic repulsion between the cationic heads, thus increasing the counterion degree of dissociation and, consequently, the effective charge, but at the same time, they contribute to the increment of surface area making the charge density decrease. This is a complex process that results in a net increase of the surface charge density. Microscopic Study. Transmission electron microscopy techniques have been used to analyze the shape, size, and morphology of the mixed aggregates studied in this work. For that purpose, TEM experiments were done for three negatively stained samples for each mole fraction: (i) at [S]tot ) 0 (blank or control test); (ii) at [S]tot below CAC*, which is supposed to be the monomers’ domain (test 1); and (iii) at [S]tot between CAC* and CVC* (i.e., the prevesicle aggregates domain, test 2). Additionally, cryoTEM experiments were run on one sample for each mole fraction (38) Loeb, A. L.; Overbeek, J. T. G.; Wiersema, P. H. The Electrical Double Layer around a Spherical Colloid Particle; MIT Press: Cambridge, MA, 1961.

at [S]tot above CVC*, where mixed vesicles are supposed to be present (test 3). As expected, the control test, performed with the uranyl acetate aqueous solution instead of the sample, leads to micrographs where only a continuous background is seen, whereas TEM micrographs for samples with [S]tot < CAC* (test 1), for which no aggregates were expected, surprisingly lead to micrographs where a series of aggregates of various shapes and sizes and without a defined, clear aggregation pattern and/or structure can be seen. Figure 4a and b shows, as an example, images of such prevesicle aggregates for di-C12DMAB/C10TAB mixed system at R1 ) 0.500. However, these “disordered” structures turn to ordered patterns for samples for which CAC* < [S]tot < CVC* (test 2), as can be observed in Figure 5. This Figure, which shows, as an example, a gallery of TEM fields for the above-mentioned mixed system at the same mole fraction, displays a collection of structures clearly arranged as circles of various diameters that are made of an internal collection of slices (Figure 5a). To determine if these circles are made of a regular arrangement of rings, boxes of 80 pixels × 80 pixels were extracted after selecting the center of each of these circles using the “boxer” command from EMAN39 and were processed without CTF correction. Up to ∼20 of these images were boxed and analyzed as single-aggregate images using reference-free classification and averaging methods as implemented in EMAN.39 A circular mask was applied beyond a certain radius, and the (39) Ludtke, S. J.; Baldwin, P. R.; Chiu, W. J. Struct. Biol. 1999, 128, 82.

4032 Langmuir, Vol. 22, No. 9, 2006

Aicart et al.

Figure 5. (a) Gallery of images extracted from electron microscopy micrographs obtained after negative staining of a di-C12DMAB + C10TAB solution at R1 ) 0.500 and [S]tot ) 0.551 mM. Scale bar: 32 nm. (b) Single images from the fields shown in panel a were used to extract boxes of 80 pixels × 80 pixels after selecting the center of the aggregates. These images were analyzed by image processing methods as single particles and as implemented in the Eman software package.39 A circular mask was applied beyond a certain radius, and the resulting images were also normalized to homogenize the data. A gallery of the resulting images is shown. (c) Two-dimensional average obtained for the data set shown in panel b after classification and 2D alignment of these images using reference-free methods.39 After alignment into a congruent orientation, pixels in similar positions for each image were averaged to generate a final 2D average (shown) of the input data. This average has a higher signal-to-noise ratio than the individual images, therefore revealing and highlighting that the aggregates are made of a repetitive unit of around 3.2 nm in section. (d) Plot of the gray levels along a straight line across the 2D average shown in Figure 5c. The maximum densities are revealed as peaks within a distance of five to six pixels, with six pixels being the most frequently observed value and corresponding to 3.2 nm (after assuming 0.53 nm/pixel after the scanning of the micrographs).

resulting images were also normalized to homogenize the data. A gallery of 10 of the resulting images is shown in Figure 5b. Most of the extracted aggregates were capable of averaging into a 2D image where the internal structure is more clearly defined, as can be observed in Figure 5c. The spacing within repetitive elements of the 2D average was measured by drawing a line across the image and plotting the intensities of the gray levels. The number of pixels between several contiguous peaks was then calculated and found to be 6 pixels on average (Figure 5d).

This value was multiplied by the 0.53 nm per pixel at the specimen after digitization of the micrographs, which yielded a spacing of ∼32 Å in section for the example shown in Figure 5. Finally, the same image processing protocol was also applied to several micrographs of the aggregates found below CAC*; an averaged 2D image with a clear pattern was not found in this case, confirming the disordered character of these aggregated structures. Figure 6 shows an example of the micrographs taken in cryoTEM experiments (test 3) of di-C12DMAB/C10TAB at R1 )

PreVesicle Nanostructures and Vesicles

Langmuir, Vol. 22, No. 9, 2006 4033

Figure 6. (a) Representative cryo-TEM field of a di-C12DMAB + C10TAB solution at R1 ) 0.500 and [S]tot ) 1.458 mM. (b) Gallery of selected vesicles that have been extracted and magnified.

0.500 and [S]tot ) 1.458 mM. These photographs revealed a large number of polydisperse spherical vesicles, essentially unilamellar, with an average diameter of about (37 ( 18) nm. Similar micrographs have been visualized for di-C12DMAB/ C14TAB mixed vesicles at R1 ) 0.496 and [S]tot ) 0.994 mM, which are also polydisperse, unilamellar, and spherical, with an average diameter of about (50 ( 30) nm. It is worth mentioning that, to our knowledge, this work is the first evidence of (i) the existence of disordered aggregates below CAC*, within a concentration region where only monomers were supposed to exist and (ii) such a clear, reproducible, ordered fingerprint pattern shown by the prevesicle aggregates at CAC* < [S]tot < CVC*, which remind us of those reported by some research groups for DNA complexed by cationic liposomes (or genosomes)40-43 and/or DNA molecules packed within viral capsids.44 Furthermore, the fingerprint pattern reported for some genosomes has been resolved by Fourier transform techniques into diffraction images that point to inverted hexagonal HCII (40) Huebner, S.; Battersby, B. J.; Grimm, R.; Cevc, G. Biophys. J. 1999, 76, 3158. (41) Mel’nikova, Y. S.; Mel’nikov, S. M.; Lofroth, J. E. Biophys. Chem. 1999, 81, 125. (42) Chesnoy, S.; Huang, L. Annu. ReV. Biophys. Biomol. Struct. 2000, 29, 27. (43) Tarahovsky, Y. S.; Rakhmanova, V. A.; Epand, R. M.; MacDonald, R. C. Biophys. J. 2002, 82, 264. (44) Agirrezabala, X.; Martin-Benito, J.; Caston, J. R.; Miranda, R.; Valpuesta, J. M.; Carrascosa, J. L. EMBO J. 2005, 24, 3820.

phases.41,42,45,46 These techniques have also been applied to the images obtained in this work, but the inverted hexagonal phase was not found. Further studies in this sense will be welcomed. Spectroscopic Study. TNS and PRODAN belong to this type of compound with an electron donor-acceptor moiety linked by a single bond to an aromatic ring (Chart 1). This arrangement confers the molecules’ characteristic properties, which have been explained in the literature.17-24 Figures 7 and 8 show, as an example, the emission spectra of TNS and PRODAN, respectively, for a series of di-C12DMAB/C10TAB/probe aqueous solutions at different mole fractions R1 and total surfactant concentration [S]tot. These concentrations were chosen to have either prevesicle aggregates or vesicles, according to CAC* and CVC* values at each mole fraction, previously obtained from conductivity experiments. It is known20 that the ground states of the fluorescent probes used in this work, which are characterized as possessing a naphthalene ring and a phenyl ring (TNS) or a dimethylamino moiety (PRODAN) in their molecules, yield, upon excitation, a locally excited state (LE) (via a π f π* electronic transition). Among the different approaches reported in the literature17,19,20,23,24,47-50 for interpreting the mechanism (45) Tarahovsky, Y. S.; Arsenault, A. L.; MacDonald, R. C.; McIntosh, T. J.; Epand, R. M. Biophys. J. 2000, 79, 3193. (46) Safinya, C. R. Curr. Opin. Struct. Biol. 2001, 11, 440. (47) Zachariasse, K. A.; Grobys, M.; van der Haar, T.; Hebecker, A.; Il’ichev, Y. V.; Morawski, O.; Riicker, I.; Kiihnle, W. J. Photochem. Photobiol., A 1997, 105, 373.

4034 Langmuir, Vol. 22, No. 9, 2006

Figure 7. Emission fluorescence spectra of TNS ([TNS] ) 2.07 µM) immersed in aqueous solutions of di-C12DMAB (1) + C10TAB (2) at (i) disordered prevesicle concentrations (dashed lines, in increasing intensity order); R1 ) 1, [S]tot ) 0.067 mM and R1) 0.647, [S]tot ) 0.069 mM; (ii) ordered prevesicle concentrations (dotted lines, in increasing intensity order); R1) 0.647, [S]tot ) 0.656 mM and R1 ) 0.847, [S]tot ) 1.498 mM; and (iii) mixed vesicle concentrations (solid lines, in increasing intensity order); R1 ) 1, [S]tot ) 0.306 mM; R1) 0.500, [S]tot ) 1.454 mM; R1 ) 0.200, [S]tot ) 0.795 mM; R1 ) 0.647, [S]tot ) 1.662 mM; R1 ) 0.847, [S]tot ) 1.498 mM; R1 ) 0.401, [S]tot ) 1.094 mM; and R1 ) 0.647, [S]tot ) 1.899 mM. The inset shows the deconvolution of the emission fluorescence band of TNS immersed in a vesicle solution of diC10DMAB + C10TAB (R1) 0.647, [S]tot ) 1.899 mM) for three Gaussian components.

by which this excited state is formed and deactivated, we have followed in this work the one50 that considers the π f π* emission band of the fluorescent probe as consisting of different bands, each of which is attributable to the π f π* emission of the probe immersed within different aggregate microenvironments where the probe may be housed. Because each microenvironment is characterized by its hydrophobicity, microviscosity, rigidity, and/ or solvation features, the corresponding π f π* emission is shifted to different wavelengths. Accordingly, experimental emission spectra were analyzed, after the conversion of wavelength to frequency, by deconvolution into overlapping Gaussian curves with a commercial nonlinear least-squares multipeak fitting procedure that uses an iterative Marquardt-Levenberg fitting algorithm. Both the procedure and the control of the goodness of the fits have been explained elsewhere.8 The insets at the tops of Figures 7 and 8 show, as examples, the deconvolutions of TNS and PRODAN emission for the mixed systems studied herein, whereas Tables 3-6 resume the fitting parameters. In all cases, the mathematical treatment points to the presence of three microenvironments, which have been assigned as follows. TNS Emission. Figure 7 shows two groups of peakssthose corresponding to the disordered prevesicle aggregates solutions (dashed lines) and those corresponding to both ordered prevesicle aggregates (dot lines) and/or vesicle solutions (solid lines). The Gaussian components of the bands corresponding to the disordered aggregates, as can be seen in the first two lines of the tables (48) Kawski, A. Z. Naturforsch. 1999, 54, 379. (49) Rettig, W.; Paeplow, B.; Herbst, H.; Mullen, K.; Desvergne, J. P.; BouasLaurent, H. New J. Chem. 1999, 23, 453. (50) Karukstis, K. K.; Zieleniuk, C. A.; Fox, M. J. Langmuir 2003, 19, 10054.

Aicart et al.

Figure 8. Emission fluorescence spectra of PRODAN ([PRODAN] ) 5.02 µM) immersed in aqueous solutions of di-C12DMAB (1) + C10TAB (2) at (i) disordered prevesicle concentrations (dashed lines, in increasing intensity order); R1) 1, [S]tot ) 0.067 mM and R1) 0.647, [S]tot ) 0.069 mM; (ii) ordered prevesicle concentrations (dotted lines, in increasing intensity order); R1 ) 0.647, [S]tot ) 0.655 mM; and (iii) mixed vesicle concentrations (solid lines, in increasing intensity order); R1 ) 1, [S]tot ) 0.326 mM; R1 ) 0.401, [S]tot ) 1.097 mM; R1 ) 0.647, [S]tot ) 1.665 mM; R1 ) 0.500, [S]tot ) 1.458 mM; and R1 ) 0.647, [S]tot ) 1.904 mM. The inset shows the deconvolution of the emission fluorescence band of PRODAN immersed in a vesicle solution of di-C12DMAB + C10TAB (R1 ) 0.647, [S]tot ) 1.665 mM) for three Gaussian components.

Tables 3 and 4, are centered at around 393, 410, and 436 nm for both di-C12DMAB/C10TAB/TNS and di-C12DMAB/C14TAB/ TNS, whereas those for the ordered prevesicle aggregates and/or vesicle bands are centered at around 422, 445, and 476 nm in both cases. We notice that λi remains almost constant within each group, irrespective of (i) R1; (ii) the prevesicle aggregate concentration, [S]agg,tot, and/or vesicle concentration, [S]v; (iii) the aggregation pattern; or (iv) the length of the single-chain surfactant tail. These bands have been assigned to different microenvironments as follows. In the presence of vesicles (last six lines in Table 3 and last four lines in Table 4), (i) the most blue-shifted emission at around 422 nm has been attributed to TNS inside the hydrophobic bilayer, which is the more hydrophobic region; (ii) the emission at around 476 nm must come from TNS in the vesicle surface; and (iii) the 445 nm emission has been assigned to TNS in the bulk solvent. Notice that TNS experiences a lowerenergy transition when it is housed in the surface than that one found in the bulk solvent, indicating that the charged di-C12DMA+ heads yield a more polar microenvironment than the bulk. It is also remarkable that the bands (and their Gaussian components) corresponding to the ordered prevesicle aggregates (lines three and four in Table 3) are centered at the same wavelengths. This feature means that the different microenvironments where TNS may be housed are similar for both the vesicles and the fingerprint-like aggregates. In a previous paper,8 the three bands found for the emission of TNS in solutions with [S]tot < CAC* were assigned to TNS housed within hydrophobic clusters formed by monomers of the double-chain and the single-chain surfactants (λ1), which is due to either to the formation of ion-pair complexes, as suggested

PreVesicle Nanostructures and Vesicles

Langmuir, Vol. 22, No. 9, 2006 4035

Table 3. Parameters of the Deconvoluted Gaussian Components of the TNS Fluorescence Emission Spectra at Various Mole Fractions r1 and Total Surfactant Concentrations [S]tot of the Mixed System di-C12DMAB (1) + C10TAB (2)a R1

[S]tot (mM)

CAC* (mM)

CVC (mM)

0.647 1.0 0.647 0.847 0.200 0.401 0.500 0.647 0.647 1.0

0.069 0.067 0.656 1.498 0.795 1.094 1.454 1.662 1.899 0.306

0.496

0.822 0.160 0.822 1.740 0.441 0.524 0.615 0.822 0.822 0.160

0.496 0.598 0.348 0.405 0.441 0.496 0.496 -

[S]agg,tot (mM)

0.160 0.900 0.447 0.689 1.013 1.166 1.403

[S]v (mM)

λmax (nm)

Imax

λ1 (nm)

A1 %

W1

λ2 (nm)

A2 %

W2

λ3 (nm)

A3 %

W3

r2

0.354 0.570 0.839 0.840 1.076 0.146

400.0 398.4 428.6 432.3 432.1 432.4 431.9 432.0 430.5 434.0

499.8 235.9 827.4 957.7 687.0 973.7 578.2 938.0 999.9 203.5

392.4 392.9 421.6 421.3 421.8 421.9 421.4 421.6 421.5 424.2

20.2 21.6 33.0 32.2 33.0 34.9 31.9 32.8 32.3 37.2

16 17 30 29 30 30 30 29 29 40

409.4 411.0 445.6 445.2 445.3 446.4 444.3 445.4 445.0 450.1

48.3 48.0 46.3 47.0 46.0 45.5 45.2 46.7 46.8 43.4

28 30 40 40 40 40 40 40 40 51

434.9 438.1 476.1 475.4 475.5 476.8 473.8 475.8 475.0 486.9

31.4 30.4 20.7 20.7 21.0 19.6 22.9 20.4 21.0 19.4

49 51 61 60 62 60 61 60 60 77

0.9992 0.9989 0.9999 0.9999 0.9999 0.9999 0.9999 0.9999 0.9999 0.9995

a

Wavelength λι, width Wi, and area Ai in terms of the percent contribution to the overall fluorescence emission area. [S]agg,tot and [S]v are the total aggregated surfactant concentration and the surfactant concentration in vesicle form, respectively. Table 4. Parameters of the Deconvoluted Gaussian Components of the TNS Fluorescence Emission Spectra at Various Mole Fractions r1 and Total Surfactant Concentrations [S]tot of the Mixed System di-C12DMAB (1) + C14TAB (2)a R1

[S]tot (mM)

CAC* (mM)

CVC* (mM)

0.496 1 0.496 0.496 0.651 1

0.070 0.067 0.784 1.010 1.041 0.306

0.454

0.624 0.160 0.624 0.624 0.884 0.160

0.454 0.454 0.513 --

[S]agg,tot (mM)

0.330 0.556 0.528

[S]v (mM)

λmax (nm)

Imax

λ1 (nm)

A1%

W1

λ2 (nm)

A2%

W2

λ3 (nm)

A3%

W3

r2

0.159 0.386 0.157 0.146

399.0 398.4 429.0 431.0 431.0 434.0

147.8 235.9 832.4 802.2 883.2 203.5

394.0 392.8 421.2 421.8 421.6 424.2

21.4 21.6 30.9 33.3 33.1 37.2

16 17 29 30 30 40

409.9 411.0 444.7 445.4 445.3 450.1

44.0 48.0 46.7 45.6 45.7 43.4

26 30 40 40 40 51

433.6 438.1 474.6 475.8 475.3 486.9

34.6 30.4 22.4 21.1 21.2 19.4

47 51 61. 61 60 77

0.9991 0.9989 0.9999 0.9999 0.9999 0.9995

a Wavelength λι, width Wi, and area Ai in terms of the percent contribution to the overall fluorescence emission area. [S]agg,tot and [S]v are the total aggregated surfactant concentration and the surfactant concentration in vesicle form, respectively.

Table 5. Parameters of the Deconvoluted Gaussian Components of the PRODAN Fluorescence Emission Spectra at Various Mole Fractions r1 and Total Surfactant Concentrations [S]tot of the Mixed System di-C12DMAB (1) + C10TAB (2)a R1

[S]tot (mM)

0.647 1 0.647 0.401 0.500 0.647 0.647 1

0.069 0.067 0.655 1.097 1.458 1.665 1.904 0.326

CAC*(mM) 0.496 0.496 0.405 0.441 0.496 0.496

CVC (mM) 0.822 0.160 0.822 0.524 0.615 0.822 0.822 0.160

[S]agg,tot (mM)

0.160 0.689 1.017 1.169 1.408

[S]v (mM)

λmax (nm)

Imax

λ1 (nm)

A1%

W1

λ2 (nm)

A2%

W2

λ3 (nm)

A3%

W3

r2

0.571 0.843 0.842 1.082 0.166

519.5 519.0 489.0 483.0 483.5 483.0 481.8 493.5

253.0 250.0 789.2 632.2 842.2 701.9 899.2 590.4

459.8 456.3 474.2 470.7 466.9 463.3 461.9 479.9

12.1 11.7 76.1 69.7 64.9 55.9 52.7 63.9

67 66 62 62 58 56 55 67

515.6 515.2 515.0 511.4 508.1 502.3 501.4 509.6

82.8 83.4 22.0 28.7 33.8 43.1 46.2 34.4

48 49 47 51 51 56 56 52

575.8 576.3 574.1 576.6 575.8 576.9 577.5 577.8

5.1 4.8 1.8 1.6 1.3 1.0 1.0 1.6

35 34 37 37 35 33 34 31

0.9994 0.9993 0.9998 0.9997 0.9998 0.9998 0.9998 0.9994

a Wavelength λι, width Wi, and area Ai in terms of the percent contribution to the overall fluorescence emission area. [S]agg,tot and [S]v are the total aggregated surfactant concentration and the surfactant concentration in vesicle form, respectively.

Table 6. Parameters of the Deconvoluted Gaussian Components of the PRODAN Fluorescence Emission Spectra at Various Mole Fractions r1 and Total Surfactant Concentrations [S]tot of the Mixed System di-C12DMAB (1) + C14TAB (2)a R1

[S]tot (mM(

CAC* (mM)

CVC* mM)

0.496 1 0.496 0.496 0.651 1

0.070 0.067 0.784 1.014 1.260 0.326

0.454

0.624 0.160 0.624 0.624 0.884 0.160

0.454 0.454 0.513

[S]agg,tot (mM)

0.330 0.560 0.747

[S]v (mM)

λmax (nm)

Imax

λ1 (nm)

A1%

W1

λ2 (nm)

A2 %

W2

λ3 (nm)

A3%

W3

r2

0.160 0.390 0.376 0.166

518.0 519.0 485.0 481.0 479.6 493.5

172.9 250.0 426.9 559.4 691.2 590.4

434.2 456.3 477.8 473.3 466.9 479.9

5.0 11.7 79.3 77.1 63.5 63.9

50 66 68 63 58 67

515.5 515.2 513.9 513.2 506.3 509.6

89.6 83.4 19.2 21.2 35.2 34.4

49 49 51 49 53 52

575.8 576.3 579.33 574.9 575.8 577.8

5.3 4.8 1.6 1.7 1.3 1.6

34 34 35 39 36 31

0.9994 0.9993 0.9996 0.9998 0.9998 0.9994

a Wavelength λι, width Wi, and area Ai in terms of the percent contribution to the overall fluorescence emission area. [S]agg,tot and [S]v are the total aggregated surfactant concentration and the surfactant concentration in vesicle form, respectively.

by some other authors,51 or to the presence of certain microdomains (λ2) and to TNS in the bulk solvent (λ3). In this work, the presence of aggregates with a disordered aggregation pattern and with different sizes and shapes has been confirmed at [S]tot concentrations below CAC*, as presented in a previous section (Figure 4). This new finding allows us to reinforce our assignment of (i) the emission at around λ1 ) 393 nm to TNS housed within the disordered aggregates such as those shown in Figure 4b and (51) Karukstis, K. K.; Savin, D. A.; Loftus, C. T.; D’Angelo, N. D. J. Colloid Interface Sci. 1998, 203, 157.

c; (ii) the peak at around 410 nm to TNS placed within the aqueous regions surrounded by regions of aggregates, obviously with a micropolarity higher than that inside the aggregates but lower than that of the bulk solvent (Figure 4a); and (iii) the 436 nm emission assigned to TNS in the bulk solvent. Other authors18,51 have suggested that the association of the fluorescent probe with, for example, monomers or clusters of surfactant in aqueous media could be the responsible for the emission observed at concentrations below the CVC or CMC, but this work is the first experimental evidence of the existence of these “clusters”

4036 Langmuir, Vol. 22, No. 9, 2006

or disordered aggregates. Another interesting feature is that the micropolarity found in TNS experiments within the microdomains below CAC* is lower than that found within the vesicles, with this fact producing a blue shift of the spectra with respect to the spectra obtained when vesicles are present, as can be observed in Figure 7. These features are in agreement with previous results obtained by other authors18 for TNS in surfactant solutions of C16TAB. PRODAN Emission. Emission spectra of PRODAN shown in Figure 8 reveal the presence of two groups of bands as well as those corresponding to disordered prevesicle aggregates (dashed lines) and those corresponding to ordered prevesicle aggregates (dotted lines) and/or vesicles (solid lines). In both di-C12DMAB/ C10TAB/PRODAN and di-C12DMAB/C14TAB/PRODAN spectra, the Gaussian components obtained for the disordered prevesicle aggregates (first two lines in Tables 5 and 6) are centered at around 457, 515, and 576 nm, whereas those for the ordered prevesicle aggregates and/or vesicle solutions are centered at around 470, 510, and 576 nm. Again, λi remains quite constant within each group of spectra, following the same behavior previously explained for TNS. Also in this case, the bands for ordered prevesicle aggregates are centered at the same wavelength as those for vesicles. The assignments of the above-mentioned PRODAN bands have been done as follows: (i) the emission at around 576 nm, the unique peak in common in all cases, has been assigned to PRODAN in the bulk solvent. In the presence of vesicle aggregates, the emissions at around 470 and 510 nm have been attributed to PRODAN inside the hydrophobic bilayer and in the vesicle surface, respectively. In disordered prevesicle solutions, the assignment of the other two peaks is similar to that proposed for TNS (i.e., the emissions at around 457 and 515 nm have also been attributed to PRODAN immersed within the disordered aggregates such as those shown in Figure 4b and c and within the aqueous regions surrounded by regions of aggregates (Figure 4a), respectively). Besides the wavelength assignment, it is also convenient to remark on other interesting features regarding the intensities (in

Aicart et al.

terms of areas, Ai) of the bands. As can be noticed in Tables 3-6, both probes are predominantly solubilized within the aggregates irrespectively of the length of the single-chain surfactant tail (i.e., (A1 + A3) > A2 for TNS and (A1 + A2) > A3 for PRODAN. However, the preferred solubilization site is the hydrophobic core because A1 > A3 for TNS and A1 > A2 for PRODAN, in contrast with the results previously found for a similar system (di-C10DMAB/OBG/Probe) where both probes were found to prefer mainly the vesicle surface as the solubilization site.8 Nevertheless, the anionic TNS remains within the bulk solvent up to a considerable extent (A2 is about 45% of the total fluorescence), whereas the nonionic PRODAN are almost totally housed within the vesicles (A3 is very low). Finally, using the reported λmax of TNS and PRODAN in a series of pure solvents of known dielectric constant,17,52 several conclusions can be drawn: (i) the results obtained in the case of TNS indicate that the hydrophobic core of di-C12DMAB/ C10TAB or di-C12DMAB/C14TAB vesicles (λ1 ) 422 nm) presents a polarity similar to that of a C3-C4 alcohol with  around 17-20, whereas that of the vesicle surface (λ3 ) 476 nm) resembles an ethylene glycol environment ( ) 41.4) and (ii) the results obtained from PRODAN spectra reveal a hydrophobic core (λ1 ) 470 nm) similar to that of a larger alcohol, such as a decanol environment ( ) 8.1), and a palisade (λ2 ) 510 nm) with a polarity similar to or slightly higher than that of methanol, with  ) 32.6. Considering that the vesicles are the same in both cases, these results point to the fact that a nonionic probe such as PRODAN is located more deeply in the hydrophobic core than its anionic counterpart (TNS), as previously found for other cationic-nonionic mixed vesicles.8 Acknowledgment. We thank the Spanish Ministry of Education, project no. BQU2002-586. O.L. thanks the MEC of Spain (project no. SAF2002-01715) for supporting TEM and cryo-TEM measurements. LA053474Q (52) Lide, D. R. CRC Handbook of Chemistry and Physics; CRC Press: Boca Raton, FL, 2004.