- Email: [email protected]
Fluorescence Anisotropy of Probes Solubilized in Micelles of Tetradecyltrymethylammonium Bromide: Effect of Ethylene Glycol Added
Journal of Colloid and Interface Science 221, 262–267 (2000) doi:10.1006/1999.6555, available online at http://www.idealibrary.com on
Fluorescence Anisotropy of Probes Solubilized in Micelles of Tetradecyltrymethylammonium Bromide: Effect of Ethylene Glycol Added Crist´obal Carnero Ruiz1 Departamento de F´ısica Aplicada II, Escuela Universitaria Polit´ecnica, Universidad de M´alaga, Campus de El Ejido, 29013 M´alaga, Spain Received May 19, 1999; accepted September 23, 1999
This paper deals with the effect of ethylene glycol on the micelle formation of tetradecyltrimethylammonium bromide. The effect of ethylene glycol addition on the fluorescence anisotropy of several probe molecules residing in different regions of the micelle was investigated to address the solvent penetration in the micelle structure. Fluorescence depolarization measurements were carried out on micellar systems containing two different hydrophobic dyes, namely, perylene and diphenylbutadiene, and a hydrophilic one, fluorescein. The steady-state anisotropy values obtained in these experiments were used to estimate the microviscosity of the corresponding micellar regions. It is observed that the microviscosity in the hydrophobic regions of micelles were roughly constant with EG addition, indicating that the micellar interior does not undergo significant structural changes by the presence of cosolvent in the solution. However, the microviscosity at the micellar surface, as determined by using fluorescein as a probe, is found to increase with EG addition. This perturbation of the micellar surface is ascribed to the solvent penetration in this region of the micelle, where there is probably participation in the solvation layer of the micelle headgroups. °C 2000 Academic Press Key Words: fluorescence anisotropy; micelles; microviscosity; tetradecyltrimethylammonium bromide; ethylene glycol–water mixtures.
INTRODUCTION
The aggregation phenomenom of amphiphiles in nonaqueous media has received considerable attention in the past few years due to the increasing use of these materials in applications that require water-free or water-poor media (1, 2). The solvents used in these studies are strongly polar with water resembling properties, such as ethylene glycol, formamide, or glycerol (3–14). Most investigations carried out focused on determining aspects such as the critical micelle concentration, the shape and size of the aggregates, or the phase behavior of these systems. On the other hand, a large number of studies were concerned with the surfactant aggregation when water is gradually replaced with another polar solvent (15–23), as this allows one to explore a 1
E-mail: [email protected].
0021-9797/00 $35.00
C 2000 by Academic Press Copyright ° All rights of reproduction in any form reserved.
range of polarities. Moreover, information on the strengh and nature of the interactions involved in the aggregation process of surfactants in these media has been gained from the analysis of the thermodynamic properties of micellization, as obtained from the determination of the critical micelle concentration as a function of temperature (18, 20, 21, 23). Another interesting aspect refers to the solvent penetration into micelles formed in these media. However, only a few authors have addressed this problem. Belmajdoub et al. (24) have studied the heteronuclear Overhauser effect between protons of formamide and carbon-13 of cetyltrimethylammonium bromide (CTAB) at experimental conditions where this surfactant forms micelles in that solvent. These authors found experimental evidence of solvent penetration into micelles close to the micellar surface. More recently, the effect of glycerol on the molecular dymanics of lipid bilayers, by using different spectroscopic techniques, has been also investigated (25). In this investigation, significant changes in the lipid packing of the studied systems were found, which were ascribed to the replacement of water with glycerol in the lipid region. As ethylene glycol (EG) is a polar solvent with properties similar to those of formamide and glycerol, and with the objective of completing the recent study made in this laboratory (23), we have decided to investigate the problem of solvent penetration in micelles of tetradecyltrimethylammonium bromide (TTAB) formed in EG–water mixtures. With this purpose we have carried out studies of steady-state fluorescence anisotropy of several probes solubilized in the micellar microphase. The degree of depolarization of the fluorescence emission of a fluorophore is an index of the rotational diffusion during the lifetime of its excited state. The steady-state fluorescence anisotropy (r ) is related to the viscosity (η) around the probe (the so-called microviscosity) through the Perrin equation kT τ r0 =1+ , r Vη
[1]
where r0 is the limiting anisotropy obtained in the absence of rotational motion, k is the Boltzmann constant, T is the absolute temperature, and V and τ are the molecular volume and fluorescence lifetime of the probe, respectively. Since alterations in microstruture can be reflected by changes in the viscosity
262
263
FLUORESCENCE ANISOTROPY OF PROBES
around the probe and hence in r values, this technique has been widely used to explore modifications in structural properties of micelles, microemulsions, and other microheterogeneous systems (26–28). In this paper, we report the results of fluorescence anisotropy studies of perylene, diphenylbutadiene, and fluorescein solubilized in micellar solutions of TTAB formed in EG–water binary mixtures. Experimental data are used to estimate the microviscosity around the probe. Due to the difficulties inherent to the experimental technique used in this work (27–29), all the values of microviscosity presented here are interpreted and should be considered only as qualitative observations revealing the microenvironmental modifications induced by the presence of EG in the solvent mixture. EXPERIMENTAL
Materials The surfactant tetradecyltrimethylammonium bromide (99%+) and heavy paraffin oil were purchased from Sigma, and ethylene glycol of spectrophotometric grade (99%+) was from Aldrich, all these materials were used as received. Fluorescence probes pyrene (Sigma), perylene (Sigma), diphenylbutadiene (Aldrich), and fluorescein (Sigma) were also used without further purification. Stock solutions of pyrene and diphenylbutadiene (DPB) were prepared in absolute ethanol, whereas those of perylene and fluorescein were prepared in chloroform and water, respectively. All chemicals used were analytical grade. Water used to prepare all solutions was doubly destilled (Millipore). Instrumentation All fluorescence measurements were recorded on a Spex FluoroMax-2 spectrofluorometer, which uses a 150-W xenon lamp as the excitation source and is equipped with a thermostated cell housing. The temperature of the sample chamber was controlled between ±0.1◦ C with a Julabo F20 circulating water bath. Fluorescence anisotropy measurements were collected in the same apparatus equipped with a polarization accessory which uses the L-format instrumental configuration (30) and an automatic interchangeable wheel with Glan–Thompson polarizers. The fluorescence anisotropy value was determined as r=
IV − G IH , IV + 2G IH
[2]
where the subscripts of the fluorescence intensity values (I ) refer to vertical (V) and horizontal (H) polarizer orientations. The instrumental correction factor G, required for the L-format configuration, was automatically determined by the software supplied by the manufacturer. The anisotropy values were averaged over an integration time of 20 s and a maximum number of five measurements were made for each sample. All anisotropy values of the probes in the micellar solutions presented in this work are the mean values of three individual determinations.
The viscosity measurements, required for the calibration curves of perylene and fluorescein, were made with a Haake Rotovisco VT550 concentric-cylinder rotational viscometer. The coaxial cylinder sensor system NV used was equipped with the outer cylinder, which was temperature-controlled between ±0.1◦ C. Methods In order to obtain the calibration curve for the microviscosity determination in the hydrophobic micellar region by using perylene as a probe, a solution of perylene (1 µM) in heavy paraffin oil was prepared as follows: a suitable volume of perylene stock solution was placed in a volumetric flask, and the solvent was evaporated off by slow passing of nitrogen. The heavy paraffin oil was then added, and the resultant solution was sonicated for 4 h. Steady-state anisotropy values of this solution in the temperature range of 6–65◦ C were recorded by setting the slit widths on the monochromators to bandwidth corresponding to 1.05 nm and excitation and emission wavelengths of 414.7 and 474.5 nm, respectively. The viscosity of the oil was measured in the same temperature range by using a shear rate of 162.3 s−1 until 50◦ C and of 324.0 s−1 for higher temperatures. In the case of fluorescein, with the aim of examining the behavior of the fluorescence anisotropy of this probe as a function of the viscosity, a solution of fluorescein (6.4 µM) in a 88.5 wt% EG solution was prepared. The anisotropy values of fluorescein in this solution were collected in a temperature range of 5.2–50◦ C by setting the slit widths on the monochromators to bandwidth corresponding to 1.05 nm and excitation and emission wavelengths of 504 and 521 nm, respectively. In the same way, the viscosity of the solvent mixture was determined in the same temperature range by using a shear rate of 2096.0 s−1 . For the microviscosity evaluation, the fluorescence lifetime (τ ) of the probes was estimated from the relation I τ = , τ0 I0
[3]
where τ0 is the fluorescence lifetime taken from the literature, I0 is the fluorescence intensity measured under the condition common with the τ0 measurement, and I is the measured fluorescence intensity of the probe in the micellar solution or in the reference solvent. Equation [3] can be applied when: (i) the emission decay is monoexponential, (ii) only collisional quenching occurs, and (iii) the behavior of the probe in the micellar medium is the same as in the reference solvent (31). This approach has been recently used to estimate the fluorescence lifetime of several fluorescence probes, including perylene and fluorescein, in different micellar media (31, 32). In the case of perylene, we took the literature values of 4.60 ns at 25◦ C in chloroform for τ0 (33), and 0.37 for r0 (34). Jobe and Verral (35) have studied the photophysical behavior of DPB in several ionic surfactants finding that this probe undergoes a monoexponential decay in TTAB micelles. Therefore, in the present case, we took 385 ps (35) for
´ CRISTOBAL CARNERO RUIZ
264
DPB in micellar solutions of TTAB (50 mM) in the absence of EG as the τ0 value. For this probe, to manipulate the Perrin ˚ 3 were employed equation, literature values of 0.346 and 205 A for r0 and V , respectively, (35). In this case, the slit widths on the monochromators were set to bandwidth corresponding to 1.05 nm, and excitation and emission wavelengths to 337 and 379 nm, respectively. The steady-state fluorescence anisotropy measurements in micellar media with the different fluorescence probes were carried out under the same optical conditions as described. In these experiments TTAB concentration was kept at 50 mM. RESULTS AND DISCUSSION
Microviscosity in the Micellar Hydrophobic Interior In order to obtain information about the structural changes in the micellar interior induced by the presence of EG in the solution, we have carried out fluorescence anisotropy studies with two different probes of hydrophobic character: perylene and DPB. Perylene is a fluorescence probe essentially insoluble in water which has been frequently used to estimate the microviscosity of the hydrophobic region of ionic (26, 34) and nonionic (31) micelles. In fact, this probe has a number of photophysical properties that are suitable for this purpose: first, it presents a high limiting anisotropy, r0 = 0.37, (34), a high fluorescence quantum yield (0.78 in cyclohexane), and a low fluorescence lifetime (4.60 ns in chloroform) (33). Therefore, perylene is a suitable probe to prevent significant contributions from micellar rotation to the overall anistropy (31). A calibration curve for perylene by using heavy paraffin oil as a reference solvent was prepared as described by Shinitzky et al. (34). It is assumed that this solvent provides an environment similar to that of the micellar hydrocarbon core, where perylene is expected to be solubilized. The calibration curve that we have obtained, plotted according the Perrin equation (Eq. [1]), is shown in Fig. 1. Contrary to what the Perrin equation predicts, the calibration curve does not present a linear behavior. This fact,
FIG. 2. Fluorescence anisotropy (d), r , and intensity (s) of perylene in TTAB micellar solutions (50 mM) as a function of EG solvent content, at 25◦ C.
which has been previously observed (31, 36), reflects the nonspherical nature of the probe, as perylene is a planar molecule with nonisotropic rotational diffusion. Figure 2 shows the results obtained for the fluorescence anisotropy and intensity of perylene in micellar medium as a function of the concentration of EG in the solution. As can be seen, while the fluorescence anisotropy remains roughly constant (around 0.02), the fluorescence intensity slightly diminishes. This reduction in the fluorescence intensity (≈16% for 60 wt% EG) is probably due to the quenching caused by the bromide ions. A priori, as the degree of counterion dissociation increases with increasing the EG concentration (23), this decrease is unexpected. However, the behavior of the fluorescence intensity for perylene observed in Fig. 2 may result from a reduction of the micellar size caused by the addition of EG (15), which should increase the quenching efficiency. Fluorescence anisotropy data together with the microviscosity values obtained by using the calibration curve in Fig. 1 are listed in Table 1. It must be noted that the microviscosity value obtained in the absence of EG compares well with literature values (26, 34). From data in Table 1, it can be seen that, within the experimental error, the microviscosity of the hydrophobic region of TTAB does not experience significant changes with increasing the EG content in the medium, TABLE 1 Fluorescence Anisotropy (r ) for Perylene and DPB and the Corresponding Microviscosity Values (η), Together with the Pyrene 1 : 3 Ratio (I 1 /I 3 ) in TTAB Micellar Solutions with Different EG Contents, at 25◦ C
FIG. 1. Calibration curve for perylene in heavy paraffin oil obtained by measuring the steady-state fluorescence anisotropy, r , in a temperature range. Data are plotted according to the Perrin equation (Eq. [1]).
Perylene
DPB
EG solvent content (wt%)
r
η (cP)
r
η (cP)
I1 /I3
0 10 20 32 40 52 60
0.019 0.020 0.019 0.020 0.019 0.019 0.020
20.4 22.3 19.7 21.5 19.0 18.4 19.5
0.228 0.234 0.241 0.246 0.251 0.253 0.257
15.0 15.7 17.6 17.4 19.3 16.7 17.5
1.36 1.36 1.36 1.36 1.37 1.39 1.41
FLUORESCENCE ANISOTROPY OF PROBES
FIG. 3. Fluorescence anisotropy (d), r , and intensity (s) of DPB in TTAB micellar solutions (50 mM) as a function of EG solvent content, at 25◦ C.
indicating that the interior structure of TTAB micelles is not affected by the presence of cosolvent in the solution. Alternatively, we have carried out a similar study by using DPB as a fluorescence probe. DPB is hydrophobic in nature and has been used to explore the structure of hydrophobic regions of micelles (35, 37). However, contrary to perylene, the fluorescence lifetime and quantum yield of DPB are very dependent upon the medium polarity (35). Figure 3 shows the results from the experiment performed in micellar media by using DPB as a probe. It is observed that while the fluorescence anisotropy slightly increases, the fluorescence intensity decreases (≈22% for 60 wt% EG). In this case, the reduction in the fluorescence intensity can be due to several factors: (i) the quenching produced by bromide ions, (ii) the penetration of the solvent into the micelle, (iii) the outer location of the probe in the micellar structure, where the polarity sensed by the probe would increase, (iv) or a combined effect of all or some of these factors. With the aim of gaining additional information on the changes in polarity of the hydrophobic region of TTAB micelles, we have carried out measurements of the pyrene 1 : 3 ratio, i.e., the relationship between the first and third vibronic peak, I1 /I3 , in a monomeric pyrene fluorescence spectrum. This parameter is well known to be an excellent index of the polarity in the probe microenvironment (26–28). The I1 /I3 values for the micellar media formed in the different solvent mixtures are shown in Table 1. The behavior of I1 /I3 apparently suggests that the polarity of the pyrene microenvironment remains constant until the solvent composition becomes 32 wt%, decreasing slightly when increases the EG content in the solution. Nevertheless, in this case there are also two possible explanations for this fact. It is known that pyrene is solubilized in the palisade layer of micelles, near the polar groups, where it senses a relatively high polar enviroment (27, 38). The reduction in the micellar size (aggregation number) with EG addition must be accompanied by an increase of the surface area per headgroups (39), inducing the incorporation of a greater number of solvent molecules to the palisade layer. Consequently, the polarity sensed by the probe is expected to increase. On the other hand, it is also possible that this increase
265
of surface area per headgroups forces the probe to locate in the outer regions in the micelle, which would contribute to a more polar environment. Although no definitive conclusions can be drawn about these effects, the small changes observed in the ratio I1 /I3 (see Table 1) indicate that these effects are little relevant. In view of these observations, we conclude that the decrease in the fluorescence intensity observed in the case of DPB is caused by the increased quenching efficiency of bromide ions, due to the reduction in the micellar size and, consequently, to the greater proximity between quencher and probe. In Table 1 we have also listed the experimental values of fluorescence anisotropy of DPB in TTAB micelles together with the corresponding microviscosity values as determined from the Perrin equation. From the data of microviscostiy obtained by using DPB as a probe we can conclude that, within the experimental error, the microviscosity of the solubilization site of DPB does not undergo significant changes with increasing concentration of EG in the solvent mixture. Also, the microviscosity values are, in this case, smaller than those obtained by using perylene as a probe, indicating probably that DPB is solubilized in an outer region than perylene, where the microviscosity is smaller. Microviscosity at the Micellar Interface To obtain a quantitative estimation on microviscosity changes at the micellar surface induced by EG addition, we have carried out measurements of fluorescence anisotropy of fluorescein solubilized in micellar solutions of TTAB. The use of fluorescein as a superficial probe in cationic micelles is justified by the anionic character of this probe, which ensures that the probe will be micellized, through electrostatic interactions, at the positively charged TTAB micellar surface. In fact, several spectroscopic studies previously reported (33, 40) have shown that fluorescein is solubilized in the inner part of the Stern region of CTAB micelles. As TTAB is a surfactant with surface properties similar to CTAB, it seems reasonable to assume that fluorescein will be micellized in the same site in TTAB micelles. Therefore, changes in the fluorescence anisotropy of fluorescein solubilized in TTAB micellar solutions will reflect changes in microviscosity at the micellar surface. A calibration curve was prepared by measuring the fluorescence anisotropy of fluorescein in a 88.5 wt% EG solution in a temperature range. This calibration curve, plotted according to the Perrin equation, is shown in Fig. 4. The excellent linear correlation found (r = 0.9996) indicates that the term kτ/V r0 remains constant across the temperature range studied. For this reason it was not neccessary to estimate the lifetime values of the probe in this case. The fluorescence anisotropy values of fluorescein in TTAB micellar media at different EG contents in the solvent system are shown in Fig. 5, where the behavior of fluorescein in homogeneous media of the same composition is also presented for comparison. Data in Fig. 5 show an emission much more polarized in TTAB micelles than in homogeneous media, revealing a strong interaction between probe and micelle. On the other hand, data in micellar media indicate that the micellar
266
´ CRISTOBAL CARNERO RUIZ
TABLE 2 Fluorescence Anisotropy (r ) of Fluorescein and the Corresponding Microviscosity Values (η) in TTAB Micellar Solutions with Different EG Contents, at 25◦ C
FIG. 4. Plot of the reciprocal of the fluorescence anisotropy, r −1 , vs T η−1 of fluorescein in an EG–water (88.5 wt%) binary solution.
surface structure is affected by the EG addition. The microviscosity values at the surface of TTAB micelles, as determined from the calibration curve in Fig. 4, are listed in Table 2. Data in Table 2 indicate clearly that the increase of EG in the solvent system induces an increase in the microviscosity at the micellar surface. It is important to point out that this increase in microviscosity sensed by fluorescein is consistent with a certain interaction of solvent molecules with the headgroups of the surfactant, probably participating in the solvation of these groups. In effect, as an EG molecule is roughly 3.1 times larger than a water molecule, if EG replaces some water molecules in the solvation layers of the micelle headgroups, an increase of EG content will also cause an increase in the local viscosity sensed by the probe. It must be noted that this mechanism could also explain the reduction in the micellar size with EG addition. Since the direct participation of EG in the solvation of headgroups also causes an increase of the surface area per headgroup, then the micellar aggregation number is reduced, as it has been recently pointed out in the case of sodium dodecyl sulfate micelles formed in EG–water mixtures (20). Finally, it is important to mention that there is a certain possibility of that
EG solvent content (wt%)
r
η (cP)
0 10 20 32 40 52 60
0.078 0.082 0.089 0.096 0.102 0.111 0.117
8.1 8.5 9.6 10.6 11.6 13.0 14.1
the participation of EG in the solvation layer of TTAB micelles causes the transfer of some fluorescein molecules from the micellar surface to the bulk solution. In fact, a similar effect has been previously observed for different surface probe bound to micelles (41). Given the electrostatic nature of the interactions involved in the micellization process of fluorescein in TTAB micelles, this effect is expected to be of very little significance. However, it must be recognized that if the participation of EG in the solvation layer of TTAB micelles resulted in the displacement of several fluorescein molecules from the micellar surface to the bulk water, the microviscosity values in Table 2 at high EG content would be underestimated and, as a consequence, the effect caused by the EG addition would be still larger than that reflected from data in Table 2. CONCLUSIONS
Through the analysis of the photophysical response of different fluorescence probes solubilized in TTAB micelles, we have obtained reliable information on the microstructural changes in micelles induced by the presence of EG in the solvent system. From the steady-state fluorescence probe technique the microviscosity in different regions of TTAB micelles has been estimated as a function of the EG concentration in the solution. The most important conclusion that we can extract from the microviscosity values obtained in this study is that while the micellar interior does not present structural alterations when the EG content increases in the solution, the structure of the micellar surface is clearly affected by the presence of EG in the solvent system. These results seem to support the hypothesis of the solvent penetration into the micelle, close to the micellar surface, in good agreement with recent investigations previously carried out (24, 25) by using different amphiphiles and experimental techniques. REFERENCES
FIG. 5. Steady-state fluorescence anisotropy, r , of fluorescein in 50 mM TTAB micellar solutions (d) and in homogeneous media (s) as a function of EG solvent content, at 25◦ C.
1. Holmberg, K., and Laughlin, R. G., Curr. Opin. Colloid Interface Sci. 2, 453 (1997). 2. W¨arnheim, T., Curr. Opin. Colloid Interface Sci. 2, 472 (1997). 3. Ray, A., Nature 231, 313 (1971).
FLUORESCENCE ANISOTROPY OF PROBES 4. Ionescu, L. G., and Fung, D. S., J. Chem. Soc., Faraday Trans. 1 77, 2907 (1981). 5. W¨arnheim, T., and J¨onsson, A., J. Colloid Interface Sci. 125, 627 (1988). 6. Binana-Limbele, W., and Zana, R., Colloid Polym. Sci. 267, 440 (1989). 7. Jonstr¨oner, M., Sj¨oberg, M., and W¨arnheim, T., J. Phys. Chem. 94, 7549 (1990). 8. Evans, D. F., and Miller, D. D., in “Organized Solutions. Surfactants in Science and Technology” (S. E. Friberg and B. Lindman, Eds.), p. 33. Dekker, New York, 1992. 9. Gharibi, H., Palepu, R., Bloor, D. M., Hall, D. G., and Wyn-Jones, E., Langmuir 8, 782 (1992). 10. Sj¨oberg, M., Silveston, R., and Kronberg, B., Langmuir 9, 973 (1993). 11. Ceglie, A., Colafemmina, G., Della Monica, M., Olsson, U., and J¨onsson, B., Langmuir 9, 1449 (1993). 12. Perche, T., Auvray, X., Petipas, C., Anthore, R., Meliani, S., Perez, E., Rico, I., and Lattes, A., J. Phys. IV 3, 133 (1993). 13. Nagarajan, R., and Wang, C-C., J. Colloid Interface Sci. 178, 471 (1996). 14. Ylihautala, M. Vaara, J., Ingman, P., Johisaari, J., and Diehl, P., J. Phys. Chem. B 101, 32 (1997). 15. Backlund, S. Bergenstahl, B., Molander, O., and W¨arnheim, T., J. Colloid Interface Sci. 131, 393 (1989). 16. Sj¨oberg, M., Henriksson, U., and W¨arnheim, T., Langmuir 6, 1205 (1990). 17. Palepu, R., Gharibi, H., Bloor, D. M., and Wyn-Jones, E., Langmuir 9, 110 (1993). 18. Callagham, A., Doyle, R., Alexander, E., and Palepu, R., Langmuir 9, 3422 (1993). 19. Takisawa, N., Thomason, M., Bloor, D. M., and Wyn-Jones, E., J. Colloid Interface Sci. 157, 77 (1993). 20. Gracie, K., Turner, D., and Palepu, R., Can. J. Chem. 74, 1616 (1996). 21. Lee, D. J., and Huang, W. H., Colloid Polym. Sci. 274, 160 (1996). 22. Penfold, J., Staples, E., Tucker, I., and Cummins, P., J. Colloid Interface Sci. 185, 424 (1997).
267
23. Carnero Ruiz, C., Colloid Polym. Sci. 277, 701 (1999). 24. Belmajdoub, A., Mahieu, N., Tekely, P., and Canet, D., J. Phys. Chem. 96, 1011 (1992). 25. Oradd, G., Wikander, G., Lindblom, G., and Johansson, L. B-A., J. Chem. Soc. Faraday Trans. 90, 305 (1994). 26. Kalyanasundaran, K., “Photochemistry in Microheterogeneous Systems,” Academic Press, New York, 1987. 27. Zana, R., in “Surfactant Solutions: New Methods of Investigation” (R. Zana, Ed.), p. 241. Dekker, New York, 1987. 28. Kalyanasundaram, K., and Thomas, J. K., J. Am. Chem. Soc. 99, 2039 (1977). 29. Evertsson, H., and Nilsson, S., Macromolecules 30, 2377 (1997). 30. Lakowicz, J. R., “Principles of Fluorescence Spectroscopy.” Plenum Press, New York, 1987. 31. Komaromy-Hiller, G., and von Wandruszka, R., J. Colloid Interface Sci. 177, 156 (1996). 32. Hasegawa, M., Sugimura, T., Suzaki, Y., Shindo, Y., and Kitahara, A., J. Phys. Chem. 98, 2120 (1994). 33. Garcia Sanchez, F., and Carnero Ruiz, C., J. Lumin. 69, 179 (1996). 34. Shinitzky, M., Dianoux, A-C., Gitler, C., and Weber, G., Biochem. 10, 2106 (1971). 35. Jobe, D. J., and Verrall, R. E., Langmuir 6, 1750 (1990). 36. Cogan, U., Shinitzky, M., Weber, G., and Nishida, T., Biochem. 12, 521 (1973). 37. Carnero Ruiz, C., Colloids Surf. A 147, 349 (1999). 38. Thomas, J. K., Chem. Rev. 80, 283 (1980). 39. Israelachvili, J. N., in “Physics of Amphiphiles: Micelles, Vesicles and Microemulsions” (V. Degiorgio and M. Corti, Eds.), p. 24. North Holland, Amsterdam, 1985. 40. Hadjianestis, J., and Nikokavovouras, J., J. Photochem. Photobiol. A:Chem. 69, 337 (1993). 41. Carnero Ruiz, C., Mol. Phys. 86, 535 (1995) and references cited herein.
Fluorescence Anisotropy of Probes Solubilized in Micelles of Tetradecyltrymethylammonium Bromide: Effect of Ethylene Glycol Added Crist´obal Carnero Ruiz1 Departamento de F´ısica Aplicada II, Escuela Universitaria Polit´ecnica, Universidad de M´alaga, Campus de El Ejido, 29013 M´alaga, Spain Received May 19, 1999; accepted September 23, 1999
This paper deals with the effect of ethylene glycol on the micelle formation of tetradecyltrimethylammonium bromide. The effect of ethylene glycol addition on the fluorescence anisotropy of several probe molecules residing in different regions of the micelle was investigated to address the solvent penetration in the micelle structure. Fluorescence depolarization measurements were carried out on micellar systems containing two different hydrophobic dyes, namely, perylene and diphenylbutadiene, and a hydrophilic one, fluorescein. The steady-state anisotropy values obtained in these experiments were used to estimate the microviscosity of the corresponding micellar regions. It is observed that the microviscosity in the hydrophobic regions of micelles were roughly constant with EG addition, indicating that the micellar interior does not undergo significant structural changes by the presence of cosolvent in the solution. However, the microviscosity at the micellar surface, as determined by using fluorescein as a probe, is found to increase with EG addition. This perturbation of the micellar surface is ascribed to the solvent penetration in this region of the micelle, where there is probably participation in the solvation layer of the micelle headgroups. °C 2000 Academic Press Key Words: fluorescence anisotropy; micelles; microviscosity; tetradecyltrimethylammonium bromide; ethylene glycol–water mixtures.
INTRODUCTION
The aggregation phenomenom of amphiphiles in nonaqueous media has received considerable attention in the past few years due to the increasing use of these materials in applications that require water-free or water-poor media (1, 2). The solvents used in these studies are strongly polar with water resembling properties, such as ethylene glycol, formamide, or glycerol (3–14). Most investigations carried out focused on determining aspects such as the critical micelle concentration, the shape and size of the aggregates, or the phase behavior of these systems. On the other hand, a large number of studies were concerned with the surfactant aggregation when water is gradually replaced with another polar solvent (15–23), as this allows one to explore a 1
E-mail: [email protected].
0021-9797/00 $35.00
C 2000 by Academic Press Copyright ° All rights of reproduction in any form reserved.
range of polarities. Moreover, information on the strengh and nature of the interactions involved in the aggregation process of surfactants in these media has been gained from the analysis of the thermodynamic properties of micellization, as obtained from the determination of the critical micelle concentration as a function of temperature (18, 20, 21, 23). Another interesting aspect refers to the solvent penetration into micelles formed in these media. However, only a few authors have addressed this problem. Belmajdoub et al. (24) have studied the heteronuclear Overhauser effect between protons of formamide and carbon-13 of cetyltrimethylammonium bromide (CTAB) at experimental conditions where this surfactant forms micelles in that solvent. These authors found experimental evidence of solvent penetration into micelles close to the micellar surface. More recently, the effect of glycerol on the molecular dymanics of lipid bilayers, by using different spectroscopic techniques, has been also investigated (25). In this investigation, significant changes in the lipid packing of the studied systems were found, which were ascribed to the replacement of water with glycerol in the lipid region. As ethylene glycol (EG) is a polar solvent with properties similar to those of formamide and glycerol, and with the objective of completing the recent study made in this laboratory (23), we have decided to investigate the problem of solvent penetration in micelles of tetradecyltrimethylammonium bromide (TTAB) formed in EG–water mixtures. With this purpose we have carried out studies of steady-state fluorescence anisotropy of several probes solubilized in the micellar microphase. The degree of depolarization of the fluorescence emission of a fluorophore is an index of the rotational diffusion during the lifetime of its excited state. The steady-state fluorescence anisotropy (r ) is related to the viscosity (η) around the probe (the so-called microviscosity) through the Perrin equation kT τ r0 =1+ , r Vη
[1]
where r0 is the limiting anisotropy obtained in the absence of rotational motion, k is the Boltzmann constant, T is the absolute temperature, and V and τ are the molecular volume and fluorescence lifetime of the probe, respectively. Since alterations in microstruture can be reflected by changes in the viscosity
262
263
FLUORESCENCE ANISOTROPY OF PROBES
around the probe and hence in r values, this technique has been widely used to explore modifications in structural properties of micelles, microemulsions, and other microheterogeneous systems (26–28). In this paper, we report the results of fluorescence anisotropy studies of perylene, diphenylbutadiene, and fluorescein solubilized in micellar solutions of TTAB formed in EG–water binary mixtures. Experimental data are used to estimate the microviscosity around the probe. Due to the difficulties inherent to the experimental technique used in this work (27–29), all the values of microviscosity presented here are interpreted and should be considered only as qualitative observations revealing the microenvironmental modifications induced by the presence of EG in the solvent mixture. EXPERIMENTAL
Materials The surfactant tetradecyltrimethylammonium bromide (99%+) and heavy paraffin oil were purchased from Sigma, and ethylene glycol of spectrophotometric grade (99%+) was from Aldrich, all these materials were used as received. Fluorescence probes pyrene (Sigma), perylene (Sigma), diphenylbutadiene (Aldrich), and fluorescein (Sigma) were also used without further purification. Stock solutions of pyrene and diphenylbutadiene (DPB) were prepared in absolute ethanol, whereas those of perylene and fluorescein were prepared in chloroform and water, respectively. All chemicals used were analytical grade. Water used to prepare all solutions was doubly destilled (Millipore). Instrumentation All fluorescence measurements were recorded on a Spex FluoroMax-2 spectrofluorometer, which uses a 150-W xenon lamp as the excitation source and is equipped with a thermostated cell housing. The temperature of the sample chamber was controlled between ±0.1◦ C with a Julabo F20 circulating water bath. Fluorescence anisotropy measurements were collected in the same apparatus equipped with a polarization accessory which uses the L-format instrumental configuration (30) and an automatic interchangeable wheel with Glan–Thompson polarizers. The fluorescence anisotropy value was determined as r=
IV − G IH , IV + 2G IH
[2]
where the subscripts of the fluorescence intensity values (I ) refer to vertical (V) and horizontal (H) polarizer orientations. The instrumental correction factor G, required for the L-format configuration, was automatically determined by the software supplied by the manufacturer. The anisotropy values were averaged over an integration time of 20 s and a maximum number of five measurements were made for each sample. All anisotropy values of the probes in the micellar solutions presented in this work are the mean values of three individual determinations.
The viscosity measurements, required for the calibration curves of perylene and fluorescein, were made with a Haake Rotovisco VT550 concentric-cylinder rotational viscometer. The coaxial cylinder sensor system NV used was equipped with the outer cylinder, which was temperature-controlled between ±0.1◦ C. Methods In order to obtain the calibration curve for the microviscosity determination in the hydrophobic micellar region by using perylene as a probe, a solution of perylene (1 µM) in heavy paraffin oil was prepared as follows: a suitable volume of perylene stock solution was placed in a volumetric flask, and the solvent was evaporated off by slow passing of nitrogen. The heavy paraffin oil was then added, and the resultant solution was sonicated for 4 h. Steady-state anisotropy values of this solution in the temperature range of 6–65◦ C were recorded by setting the slit widths on the monochromators to bandwidth corresponding to 1.05 nm and excitation and emission wavelengths of 414.7 and 474.5 nm, respectively. The viscosity of the oil was measured in the same temperature range by using a shear rate of 162.3 s−1 until 50◦ C and of 324.0 s−1 for higher temperatures. In the case of fluorescein, with the aim of examining the behavior of the fluorescence anisotropy of this probe as a function of the viscosity, a solution of fluorescein (6.4 µM) in a 88.5 wt% EG solution was prepared. The anisotropy values of fluorescein in this solution were collected in a temperature range of 5.2–50◦ C by setting the slit widths on the monochromators to bandwidth corresponding to 1.05 nm and excitation and emission wavelengths of 504 and 521 nm, respectively. In the same way, the viscosity of the solvent mixture was determined in the same temperature range by using a shear rate of 2096.0 s−1 . For the microviscosity evaluation, the fluorescence lifetime (τ ) of the probes was estimated from the relation I τ = , τ0 I0
[3]
where τ0 is the fluorescence lifetime taken from the literature, I0 is the fluorescence intensity measured under the condition common with the τ0 measurement, and I is the measured fluorescence intensity of the probe in the micellar solution or in the reference solvent. Equation [3] can be applied when: (i) the emission decay is monoexponential, (ii) only collisional quenching occurs, and (iii) the behavior of the probe in the micellar medium is the same as in the reference solvent (31). This approach has been recently used to estimate the fluorescence lifetime of several fluorescence probes, including perylene and fluorescein, in different micellar media (31, 32). In the case of perylene, we took the literature values of 4.60 ns at 25◦ C in chloroform for τ0 (33), and 0.37 for r0 (34). Jobe and Verral (35) have studied the photophysical behavior of DPB in several ionic surfactants finding that this probe undergoes a monoexponential decay in TTAB micelles. Therefore, in the present case, we took 385 ps (35) for
´ CRISTOBAL CARNERO RUIZ
264
DPB in micellar solutions of TTAB (50 mM) in the absence of EG as the τ0 value. For this probe, to manipulate the Perrin ˚ 3 were employed equation, literature values of 0.346 and 205 A for r0 and V , respectively, (35). In this case, the slit widths on the monochromators were set to bandwidth corresponding to 1.05 nm, and excitation and emission wavelengths to 337 and 379 nm, respectively. The steady-state fluorescence anisotropy measurements in micellar media with the different fluorescence probes were carried out under the same optical conditions as described. In these experiments TTAB concentration was kept at 50 mM. RESULTS AND DISCUSSION
Microviscosity in the Micellar Hydrophobic Interior In order to obtain information about the structural changes in the micellar interior induced by the presence of EG in the solution, we have carried out fluorescence anisotropy studies with two different probes of hydrophobic character: perylene and DPB. Perylene is a fluorescence probe essentially insoluble in water which has been frequently used to estimate the microviscosity of the hydrophobic region of ionic (26, 34) and nonionic (31) micelles. In fact, this probe has a number of photophysical properties that are suitable for this purpose: first, it presents a high limiting anisotropy, r0 = 0.37, (34), a high fluorescence quantum yield (0.78 in cyclohexane), and a low fluorescence lifetime (4.60 ns in chloroform) (33). Therefore, perylene is a suitable probe to prevent significant contributions from micellar rotation to the overall anistropy (31). A calibration curve for perylene by using heavy paraffin oil as a reference solvent was prepared as described by Shinitzky et al. (34). It is assumed that this solvent provides an environment similar to that of the micellar hydrocarbon core, where perylene is expected to be solubilized. The calibration curve that we have obtained, plotted according the Perrin equation (Eq. [1]), is shown in Fig. 1. Contrary to what the Perrin equation predicts, the calibration curve does not present a linear behavior. This fact,
FIG. 2. Fluorescence anisotropy (d), r , and intensity (s) of perylene in TTAB micellar solutions (50 mM) as a function of EG solvent content, at 25◦ C.
which has been previously observed (31, 36), reflects the nonspherical nature of the probe, as perylene is a planar molecule with nonisotropic rotational diffusion. Figure 2 shows the results obtained for the fluorescence anisotropy and intensity of perylene in micellar medium as a function of the concentration of EG in the solution. As can be seen, while the fluorescence anisotropy remains roughly constant (around 0.02), the fluorescence intensity slightly diminishes. This reduction in the fluorescence intensity (≈16% for 60 wt% EG) is probably due to the quenching caused by the bromide ions. A priori, as the degree of counterion dissociation increases with increasing the EG concentration (23), this decrease is unexpected. However, the behavior of the fluorescence intensity for perylene observed in Fig. 2 may result from a reduction of the micellar size caused by the addition of EG (15), which should increase the quenching efficiency. Fluorescence anisotropy data together with the microviscosity values obtained by using the calibration curve in Fig. 1 are listed in Table 1. It must be noted that the microviscosity value obtained in the absence of EG compares well with literature values (26, 34). From data in Table 1, it can be seen that, within the experimental error, the microviscosity of the hydrophobic region of TTAB does not experience significant changes with increasing the EG content in the medium, TABLE 1 Fluorescence Anisotropy (r ) for Perylene and DPB and the Corresponding Microviscosity Values (η), Together with the Pyrene 1 : 3 Ratio (I 1 /I 3 ) in TTAB Micellar Solutions with Different EG Contents, at 25◦ C
FIG. 1. Calibration curve for perylene in heavy paraffin oil obtained by measuring the steady-state fluorescence anisotropy, r , in a temperature range. Data are plotted according to the Perrin equation (Eq. [1]).
Perylene
DPB
EG solvent content (wt%)
r
η (cP)
r
η (cP)
I1 /I3
0 10 20 32 40 52 60
0.019 0.020 0.019 0.020 0.019 0.019 0.020
20.4 22.3 19.7 21.5 19.0 18.4 19.5
0.228 0.234 0.241 0.246 0.251 0.253 0.257
15.0 15.7 17.6 17.4 19.3 16.7 17.5
1.36 1.36 1.36 1.36 1.37 1.39 1.41
FLUORESCENCE ANISOTROPY OF PROBES
FIG. 3. Fluorescence anisotropy (d), r , and intensity (s) of DPB in TTAB micellar solutions (50 mM) as a function of EG solvent content, at 25◦ C.
indicating that the interior structure of TTAB micelles is not affected by the presence of cosolvent in the solution. Alternatively, we have carried out a similar study by using DPB as a fluorescence probe. DPB is hydrophobic in nature and has been used to explore the structure of hydrophobic regions of micelles (35, 37). However, contrary to perylene, the fluorescence lifetime and quantum yield of DPB are very dependent upon the medium polarity (35). Figure 3 shows the results from the experiment performed in micellar media by using DPB as a probe. It is observed that while the fluorescence anisotropy slightly increases, the fluorescence intensity decreases (≈22% for 60 wt% EG). In this case, the reduction in the fluorescence intensity can be due to several factors: (i) the quenching produced by bromide ions, (ii) the penetration of the solvent into the micelle, (iii) the outer location of the probe in the micellar structure, where the polarity sensed by the probe would increase, (iv) or a combined effect of all or some of these factors. With the aim of gaining additional information on the changes in polarity of the hydrophobic region of TTAB micelles, we have carried out measurements of the pyrene 1 : 3 ratio, i.e., the relationship between the first and third vibronic peak, I1 /I3 , in a monomeric pyrene fluorescence spectrum. This parameter is well known to be an excellent index of the polarity in the probe microenvironment (26–28). The I1 /I3 values for the micellar media formed in the different solvent mixtures are shown in Table 1. The behavior of I1 /I3 apparently suggests that the polarity of the pyrene microenvironment remains constant until the solvent composition becomes 32 wt%, decreasing slightly when increases the EG content in the solution. Nevertheless, in this case there are also two possible explanations for this fact. It is known that pyrene is solubilized in the palisade layer of micelles, near the polar groups, where it senses a relatively high polar enviroment (27, 38). The reduction in the micellar size (aggregation number) with EG addition must be accompanied by an increase of the surface area per headgroups (39), inducing the incorporation of a greater number of solvent molecules to the palisade layer. Consequently, the polarity sensed by the probe is expected to increase. On the other hand, it is also possible that this increase
265
of surface area per headgroups forces the probe to locate in the outer regions in the micelle, which would contribute to a more polar environment. Although no definitive conclusions can be drawn about these effects, the small changes observed in the ratio I1 /I3 (see Table 1) indicate that these effects are little relevant. In view of these observations, we conclude that the decrease in the fluorescence intensity observed in the case of DPB is caused by the increased quenching efficiency of bromide ions, due to the reduction in the micellar size and, consequently, to the greater proximity between quencher and probe. In Table 1 we have also listed the experimental values of fluorescence anisotropy of DPB in TTAB micelles together with the corresponding microviscosity values as determined from the Perrin equation. From the data of microviscostiy obtained by using DPB as a probe we can conclude that, within the experimental error, the microviscosity of the solubilization site of DPB does not undergo significant changes with increasing concentration of EG in the solvent mixture. Also, the microviscosity values are, in this case, smaller than those obtained by using perylene as a probe, indicating probably that DPB is solubilized in an outer region than perylene, where the microviscosity is smaller. Microviscosity at the Micellar Interface To obtain a quantitative estimation on microviscosity changes at the micellar surface induced by EG addition, we have carried out measurements of fluorescence anisotropy of fluorescein solubilized in micellar solutions of TTAB. The use of fluorescein as a superficial probe in cationic micelles is justified by the anionic character of this probe, which ensures that the probe will be micellized, through electrostatic interactions, at the positively charged TTAB micellar surface. In fact, several spectroscopic studies previously reported (33, 40) have shown that fluorescein is solubilized in the inner part of the Stern region of CTAB micelles. As TTAB is a surfactant with surface properties similar to CTAB, it seems reasonable to assume that fluorescein will be micellized in the same site in TTAB micelles. Therefore, changes in the fluorescence anisotropy of fluorescein solubilized in TTAB micellar solutions will reflect changes in microviscosity at the micellar surface. A calibration curve was prepared by measuring the fluorescence anisotropy of fluorescein in a 88.5 wt% EG solution in a temperature range. This calibration curve, plotted according to the Perrin equation, is shown in Fig. 4. The excellent linear correlation found (r = 0.9996) indicates that the term kτ/V r0 remains constant across the temperature range studied. For this reason it was not neccessary to estimate the lifetime values of the probe in this case. The fluorescence anisotropy values of fluorescein in TTAB micellar media at different EG contents in the solvent system are shown in Fig. 5, where the behavior of fluorescein in homogeneous media of the same composition is also presented for comparison. Data in Fig. 5 show an emission much more polarized in TTAB micelles than in homogeneous media, revealing a strong interaction between probe and micelle. On the other hand, data in micellar media indicate that the micellar
266
´ CRISTOBAL CARNERO RUIZ
TABLE 2 Fluorescence Anisotropy (r ) of Fluorescein and the Corresponding Microviscosity Values (η) in TTAB Micellar Solutions with Different EG Contents, at 25◦ C
FIG. 4. Plot of the reciprocal of the fluorescence anisotropy, r −1 , vs T η−1 of fluorescein in an EG–water (88.5 wt%) binary solution.
surface structure is affected by the EG addition. The microviscosity values at the surface of TTAB micelles, as determined from the calibration curve in Fig. 4, are listed in Table 2. Data in Table 2 indicate clearly that the increase of EG in the solvent system induces an increase in the microviscosity at the micellar surface. It is important to point out that this increase in microviscosity sensed by fluorescein is consistent with a certain interaction of solvent molecules with the headgroups of the surfactant, probably participating in the solvation of these groups. In effect, as an EG molecule is roughly 3.1 times larger than a water molecule, if EG replaces some water molecules in the solvation layers of the micelle headgroups, an increase of EG content will also cause an increase in the local viscosity sensed by the probe. It must be noted that this mechanism could also explain the reduction in the micellar size with EG addition. Since the direct participation of EG in the solvation of headgroups also causes an increase of the surface area per headgroup, then the micellar aggregation number is reduced, as it has been recently pointed out in the case of sodium dodecyl sulfate micelles formed in EG–water mixtures (20). Finally, it is important to mention that there is a certain possibility of that
EG solvent content (wt%)
r
η (cP)
0 10 20 32 40 52 60
0.078 0.082 0.089 0.096 0.102 0.111 0.117
8.1 8.5 9.6 10.6 11.6 13.0 14.1
the participation of EG in the solvation layer of TTAB micelles causes the transfer of some fluorescein molecules from the micellar surface to the bulk solution. In fact, a similar effect has been previously observed for different surface probe bound to micelles (41). Given the electrostatic nature of the interactions involved in the micellization process of fluorescein in TTAB micelles, this effect is expected to be of very little significance. However, it must be recognized that if the participation of EG in the solvation layer of TTAB micelles resulted in the displacement of several fluorescein molecules from the micellar surface to the bulk water, the microviscosity values in Table 2 at high EG content would be underestimated and, as a consequence, the effect caused by the EG addition would be still larger than that reflected from data in Table 2. CONCLUSIONS
Through the analysis of the photophysical response of different fluorescence probes solubilized in TTAB micelles, we have obtained reliable information on the microstructural changes in micelles induced by the presence of EG in the solvent system. From the steady-state fluorescence probe technique the microviscosity in different regions of TTAB micelles has been estimated as a function of the EG concentration in the solution. The most important conclusion that we can extract from the microviscosity values obtained in this study is that while the micellar interior does not present structural alterations when the EG content increases in the solution, the structure of the micellar surface is clearly affected by the presence of EG in the solvent system. These results seem to support the hypothesis of the solvent penetration into the micelle, close to the micellar surface, in good agreement with recent investigations previously carried out (24, 25) by using different amphiphiles and experimental techniques. REFERENCES
FIG. 5. Steady-state fluorescence anisotropy, r , of fluorescein in 50 mM TTAB micellar solutions (d) and in homogeneous media (s) as a function of EG solvent content, at 25◦ C.
1. Holmberg, K., and Laughlin, R. G., Curr. Opin. Colloid Interface Sci. 2, 453 (1997). 2. W¨arnheim, T., Curr. Opin. Colloid Interface Sci. 2, 472 (1997). 3. Ray, A., Nature 231, 313 (1971).
FLUORESCENCE ANISOTROPY OF PROBES 4. Ionescu, L. G., and Fung, D. S., J. Chem. Soc., Faraday Trans. 1 77, 2907 (1981). 5. W¨arnheim, T., and J¨onsson, A., J. Colloid Interface Sci. 125, 627 (1988). 6. Binana-Limbele, W., and Zana, R., Colloid Polym. Sci. 267, 440 (1989). 7. Jonstr¨oner, M., Sj¨oberg, M., and W¨arnheim, T., J. Phys. Chem. 94, 7549 (1990). 8. Evans, D. F., and Miller, D. D., in “Organized Solutions. Surfactants in Science and Technology” (S. E. Friberg and B. Lindman, Eds.), p. 33. Dekker, New York, 1992. 9. Gharibi, H., Palepu, R., Bloor, D. M., Hall, D. G., and Wyn-Jones, E., Langmuir 8, 782 (1992). 10. Sj¨oberg, M., Silveston, R., and Kronberg, B., Langmuir 9, 973 (1993). 11. Ceglie, A., Colafemmina, G., Della Monica, M., Olsson, U., and J¨onsson, B., Langmuir 9, 1449 (1993). 12. Perche, T., Auvray, X., Petipas, C., Anthore, R., Meliani, S., Perez, E., Rico, I., and Lattes, A., J. Phys. IV 3, 133 (1993). 13. Nagarajan, R., and Wang, C-C., J. Colloid Interface Sci. 178, 471 (1996). 14. Ylihautala, M. Vaara, J., Ingman, P., Johisaari, J., and Diehl, P., J. Phys. Chem. B 101, 32 (1997). 15. Backlund, S. Bergenstahl, B., Molander, O., and W¨arnheim, T., J. Colloid Interface Sci. 131, 393 (1989). 16. Sj¨oberg, M., Henriksson, U., and W¨arnheim, T., Langmuir 6, 1205 (1990). 17. Palepu, R., Gharibi, H., Bloor, D. M., and Wyn-Jones, E., Langmuir 9, 110 (1993). 18. Callagham, A., Doyle, R., Alexander, E., and Palepu, R., Langmuir 9, 3422 (1993). 19. Takisawa, N., Thomason, M., Bloor, D. M., and Wyn-Jones, E., J. Colloid Interface Sci. 157, 77 (1993). 20. Gracie, K., Turner, D., and Palepu, R., Can. J. Chem. 74, 1616 (1996). 21. Lee, D. J., and Huang, W. H., Colloid Polym. Sci. 274, 160 (1996). 22. Penfold, J., Staples, E., Tucker, I., and Cummins, P., J. Colloid Interface Sci. 185, 424 (1997).
267
23. Carnero Ruiz, C., Colloid Polym. Sci. 277, 701 (1999). 24. Belmajdoub, A., Mahieu, N., Tekely, P., and Canet, D., J. Phys. Chem. 96, 1011 (1992). 25. Oradd, G., Wikander, G., Lindblom, G., and Johansson, L. B-A., J. Chem. Soc. Faraday Trans. 90, 305 (1994). 26. Kalyanasundaran, K., “Photochemistry in Microheterogeneous Systems,” Academic Press, New York, 1987. 27. Zana, R., in “Surfactant Solutions: New Methods of Investigation” (R. Zana, Ed.), p. 241. Dekker, New York, 1987. 28. Kalyanasundaram, K., and Thomas, J. K., J. Am. Chem. Soc. 99, 2039 (1977). 29. Evertsson, H., and Nilsson, S., Macromolecules 30, 2377 (1997). 30. Lakowicz, J. R., “Principles of Fluorescence Spectroscopy.” Plenum Press, New York, 1987. 31. Komaromy-Hiller, G., and von Wandruszka, R., J. Colloid Interface Sci. 177, 156 (1996). 32. Hasegawa, M., Sugimura, T., Suzaki, Y., Shindo, Y., and Kitahara, A., J. Phys. Chem. 98, 2120 (1994). 33. Garcia Sanchez, F., and Carnero Ruiz, C., J. Lumin. 69, 179 (1996). 34. Shinitzky, M., Dianoux, A-C., Gitler, C., and Weber, G., Biochem. 10, 2106 (1971). 35. Jobe, D. J., and Verrall, R. E., Langmuir 6, 1750 (1990). 36. Cogan, U., Shinitzky, M., Weber, G., and Nishida, T., Biochem. 12, 521 (1973). 37. Carnero Ruiz, C., Colloids Surf. A 147, 349 (1999). 38. Thomas, J. K., Chem. Rev. 80, 283 (1980). 39. Israelachvili, J. N., in “Physics of Amphiphiles: Micelles, Vesicles and Microemulsions” (V. Degiorgio and M. Corti, Eds.), p. 24. North Holland, Amsterdam, 1985. 40. Hadjianestis, J., and Nikokavovouras, J., J. Photochem. Photobiol. A:Chem. 69, 337 (1993). 41. Carnero Ruiz, C., Mol. Phys. 86, 535 (1995) and references cited herein.