New Insights on the Photophysical Behavior of PRODAN in Anionic

748

J. Phys. Chem. B 2007, 111, 748-759

New Insights on the Photophysical Behavior of PRODAN in Anionic and Cationic Reverse Micelles: From Which State or States Does It Emit? Mercedes Novaira, M. Alicia Biasutti, Juana J. Silber, and N. Mariano Correa* Departamento de Quı´mica, UniVersidad Nacional de Rıó Cuarto, Agencia Postal #3 (X5804ALH), Rıó Cuarto, Argentina ReceiVed: August 25, 2006; In Final Form: NoVember 1, 2006

6-Propionyl-2-(N,N-dimethyl)aminonaphtahalene, PRODAN, is widely used as a fluorescent molecular probe because of its significant Stokes shift in polar solvents. It is an aromatic compound with intramolecular chargetransfer states (ICT) that can be particularly useful as a sensor. The nature of the emissive states has not yet been established despite the detailed experimental and theoretical investigations done on this fluorophore. In this work, we performed absorption, steady-state, time-resolved fluorescence (TRES) and time-resolved area normalized emission (TRANES) spectroscopies on the molecular probe PRODAN in the anionic water/sodium 1,4-bis-2-ethylhexylsulfosuccinate (AOT)/n-heptane and the cationic water/benzyl-n-hexadecyl dimethylammonium chloride (BHDC)/benzene reverse micelles (RMs). The experiments were done by varying the surfactant concentrations at a fixed molar ratio (W ) [H2O]/[Surfactant]) and changing the water content at a constant surfactant concentration. The results obtained varying the surfactant concentration at W ) 0 show a bathochromic shift and an increase in the intensity of the PRODAN emission band due to the PRODAN partition process between the external solvent and the RMs interface. The partition constants, Kp, are quantified from the changes in the PRODAN emission spectra and the steady-state anisotropy () with the surfactant concentration in both RMs. The Kp value is larger in the BHDC than the AOT RMs, probably due to the interaction between the cationic polar head of the surfactant and the aromatic ring of PRODAN. The partition process is confirmed with the TRES experiments, where the data fit to a continuous model, and with the time-resolved area normalized emission spectroscopy (TRANES) spectra, where only one isoemissive point is detected. On the other hand, the emission spectra at W ) 10 and 20 show a dual fluorescence with a new band that emerges in the low-energy region of the spectra, a band that was previously assigned to the PRODAN emission from the water pool of RMs. Our studies demonstrate that this band is due to the emission from an ICT state of the molecular probe PRODAN located at the interface of the RMs. These results are also confirmed by the lifetime measurements, the TRES experiments where the results fit to a two-state model, and the time-resolved area normalized emission spectroscopy (TRANES) spectra where three or two isoemissive points are detected in the AOT and BHDC RMs, respectively. In the AOT RMs, Kp values obtained at W ) 10 and 20 are almost independent of the water content; the values are higher for the BHDC RMs due to the higher micropolarity of this interface.

Introduction Several surfactants are able to aggregate in nonaqueous solvents to yield reverse micellar systems (RMs). Small solute particles can be located in three different compartments: (a) the external organic solvent, (b) the micellar interface formed by a surfactant monolayer, and (c) the internal water pool. Subsequently, these systems contain aqueous microdroplets entrapped in a film of surfactant and dispersed in a low-polarity bulk solvent.1-12 Among the anionic surfactants that form RMs, the best known are the systems derived from the AOT (sodium 1,4-bis-2ethylhexylsulfosuccinate) in different nonpolar media. AOT has a well-known V-shaped molecular geometry, giving rise to stable RMs without cosurfactant. In addition, it has the remarkable ability to solubilize a large amount of water with values of W (W ) [H2O]/[AOT]) as large as 40 to 60 depending on the surrounding nonpolar medium, the solute, and the * Author to whom correspondence should be addressed: mcorrea@ exa.unrc.edu.ar.

temperature. However, the droplets size depends only on W.2,3,5 To compare between different interfaces, we choose the cationic surfactant benzyl-n-hexadecyl dimethylammonium chloride (BHDC), which forms RMs in benzene and water can be solubilized up to W ∼ 25 without addition of a cosurfactant.13,14 The BHDC RMs seems to have properties that are characteristic of other RMs. This is brought out by the nature of the water pool in the BHDC RMs, which show properties similar to that of the bulk water only after there is enough water for the surfactant solvation.13,14 The photoinduced intramolecular charge transfer (ICT) of various organic molecules containing electron donor (generally a dialkylamino group) and acceptor groups in its moiety has been the growing interest of recent investigations because it is a possible mechanism for biological and chemical energy conversion.15-18 Among them, those who can experience an ICT process are the most interesting. The ICT state is formed from the initially excited planar state. The formation of the ICT state could give a dual fluorescence phenomenon where two emission bands will be observed: the normal emission from the local

10.1021/jp065528q CCC: $37.00 © 2007 American Chemical Society Published on Web 01/06/2007

Photophysical Behavior of PRODAN in Reverse Micelles

J. Phys. Chem. B, Vol. 111, No. 4, 2007 749

SCHEME 1: Local Excited (LE) and Intramolecular Charge Transfer (ICT) Structures of PRODAN

excited state (LE) and a new low-energy band that corresponds to the ICT state. Sometimes, this state may involve a twisting from the planar structure to the so-called twisted intramolecular charge-transfer state (TICT).17,19 6-Propionyl-2-(N,N-dimethyl)aminonaphtahalene, PRODAN, has been the subject of many studies in the last two decades due to its high sensitivity to the environment, which makes it useful as a fluorescent probe for different kinds of media such as RMs and other membranes’ mimickers.17,20-29 It is a fluorescent probe that exhibits strong shifts in the absorption and emission spectra, varying the environment. It emits an intense, single broad fluorescence band that is strongly redshifted with increasing polarity-polarizability (π*) and the hydrogen donor ability (R) of the media.26,30-32 Recently, we demonstrate that PRODAN can be used as solvent polarity parameter because the transition energy (expressed in kcal/mol) of the absorption and emission bands correlates quite well with the well-known polarity parameter ET(30).26 Theoretical calculations33,34 show that PRODAN is planar in the ground state, and its twisted excited conformer emits from an ICT fluorescence state in polar media because of strong solute-solvent interactions. In nonpolar solvents, the fluorescence occurs from the LE state; in n-butyl alcohol, the initially excited state undergoes relaxation to an energetically lower ICT state with both bands considerably overlapping.17,35-40 Despite the fact that numerous papers report applications of PRODAN,17 its photophysics seems to be far from clarity, probably due to the lack of sufficient experimental evidence. Semiempirical36 and ab initio calculations36,37 arrive at an explanation of the emission properties of PRODAN in terms of TICT model. The low-lying excited state in the planar (ground-state) geometry, has significant CT character, but the formation of an ICT state by twisting of either the -NMe2 (N-TICT) or the -(CdO)CH2CH3 group (O-TICT) (Scheme 1) leads to a large increase of the dipole moment. Samantha et al.41 have determined the excited-state dipole moment of PRODAN by the transient dielectric loss method in benzene and dioxane. The authors interpret the value obtained (about 10 D) as evidence against the pure ICT character of the emitting state because the value is too low for an N-TICT or O-TICT states. They conclude that the excited-state structure is not very different from that in the ground state, and the large solvent shifts are rather caused by specific solvation. It must be clarified that those studies were accomplished in very weakly polar solvents where, as it was previously suggested,17,42 the PRODAN emission comes from the LE state. Very recently, Abelt et al.43,44 claimed that the energy calculated for the TICT states is greatly dependent on the choice of the Onsager’s radius. The solvent stabilization by a radius of 4.6 Å favors the TICT state, whereas a slightly higher value, around 5.6 Å, gives the planar ICT (PICT) state as the lowest.36 They have proposed, using some related compounds, that PRODAN emits from a PICT state instead of the TICT state proposed before. Regarding the use of PRODAN as molecular probe to monitor RMs, some authors have investigated its photophysics in water/

AOT/n-heptane20,21,23 and in water/1-hexanol/dodecyl trimethyl ammonium bromide (DTAB) or cetyl trimethyl ammonium (CTAB)/n-heptane22 reverse micelles. Karukkstis et al.21,22 concluded that PRODAN is a powerful probe to the features of RMs, and they interpret their results considering that PRODAN exists in multiple locations within the aggregates: the free water pool, the bound water layer, the AOT interfacial region, and the bulk heptane continuum. A conclusion arrived after the deconvolution of the overall PRODAN fluorescence emission spectrum into a sum of overlapping Gaussian functions. This procedure was also applied for PRODAN in different microenvironments.24,45 On the other hand, Lissi et al.23 quantified the distribution of PRODAN between the external nonpolar solvent and the aggregate interface using fluorescence techniques. The results show a low value of the distribution constant, which is independent of the water content. For this reason, the authors claimed that there is an association of the probe to the RMs interfaces but the majority of PRODAN remains in the n-heptane pseudophase. Sengupta et al.20 have interpreted the fluorescence spectrum of PRODAN at a fixed value of AOT concentration and at different water content values. They consider the molecules to be distributed in three regions of the AOT RMs: one part of the population exists in the interfacial region with emission occurring in the blue region, some of the molecules are in the tail surfactant region of the reverse micelles where they face a nonpolar environment, and the remaining molecules exist in the polar head group region. At higher values of W, PRODAN moves toward the bound water phase at the interface with a new emission band around λmax ) 500 nm. The authors are not in agreement with the suggestion of Karukkstis et al.21,22 that there is a fraction of PRODAN molecules in the bulk water phase, because PRODAN’s solubility in water is low. They have also suggested that further studies are necessary in order to gain insights in the behavior of this probe in AOT RMs. However, all the studies consider the PRODAN emission to be from a unique excited-state geometry: the LE state. In light of all this controversy about the emitting state(s) of PRODAN, here we perform systematic studies of PRODAN in the anionic water/AOT/n-heptane and the cationic water/BHDC/ benzene RMs using absorption and emission (stationary, timeresolved emission spectra (TRES) and time-resolved area normalized emission (TRANES)46-48) spectroscopies. TRES is frequently used to study the excited-state dynamics and kinetics of fluorescent molecules in solution.38 The standard interpretation of TRES assumes a prior knowledge of the number of fluorescent species in the ground state, usually a single species. This assumption may fail if the fluorophore is present in a complex environment such as a microheterogeneous media. Time-resolved area normalized emission spectroscopy (TRANES)46-48 is a step forward, and with this analysis, it is possible to determine the number of species in the sample that contributes to the observed fluorescence emission. In this way, TRANES, which is a modified version of TRES, is obtained without assumption of ground- or excited-state kinetics. A useful

750 J. Phys. Chem. B, Vol. 111, No. 4, 2007 feature of TRANES is that an isoemissive point in the spectra supports any model that involves two emitting species in the sample. On the other hand, multiple isoemissive points in TRES spectra at different time intervals are observed if there are more than two emitting species. For example, two isoemissive points are found if there are three emitting species with emission spectra and fluorescence lifetimes distinctly separate.46-48 We present our results showing first the data at W ) 0 where the PRODAN photophysics are consistent with the emission from a LE state. Also, the results show a partition process of the molecular probe between the organic pseudophase and the RM interface in both RMs. The partition constant value (Kp) in BHDC is larger than the one obtained in the AOT RMs because of the interaction between the cationic polar head of the surfactant and the PRODAN aromatic ring. After that, we present the data at W ) 10 and 20 where the PRODAN photophysics is more complex because of the emission from two different excited states: the LE and the ICT. The results show that the Kp values in the AOT RMs are independent of the water content but they depend on W for the BHDC RMs until a value of W ) 10. This is characteristic of a molecular probe that exists at the RM interface.5,13,49 The emission spectra of PRODAN inside the RMs varying the surfactant concentration at these W values show a new emission band that emerges around λmax ) 500 nm. We demonstrate that this new band is due to the emission of the probe located at the RMs interfaces from an ICT state and not due to the emission of PRODAN molecules that exist in the water pool of the RMs in the ground state as it was previously suggested. The PRODAN ICT state is formed from its LE state during its excited lifetime. These results are confirmed with the lifetime measurements where a negative pre-exponential factor appeared, the TRES experiments where the two-state model is employed, and the TRANES spectra where three and two isoemissive points are detected in the AOT and BHDC RMs, respectively. Experimental Section Sodium 1,4-bis-(2-ethylhexyl) sulfosuccinate (AOT) (Sigma, >99% purity) was used as received. Benzyl-n-hexadecyldimethylammonium chloride (BHDC) from Sigma (>99%) was recrystallized twice from ethyl acetate.13 Both surfactants were kept under vacuum over P2O5 to minimize H2O absorption. The absence of acidic impurities was confirmed through the 1-methyl-8-oxyquinolinium betaine (QB) absorption bands.13 n-Heptane and benzene (Sintorgan HPLC quality) were used as received, and Ultrapure water was obtained from Labonco equipment model 90901-01. The absorption spectra were measured by using Shimadzu 2401 equipment at 25 ( 0.1 °C unless otherwise indicated. A Spex fluoromax apparatus was employed for the fluorescent measurements. Corrected fluorescence spectra were obtained using the correction file provided by the manufacturer. The path length used in the absorption and emission experiments was 1 cm. The λmax was measured by taking the midpoint between the two positions of the spectrum where the absorbance is equal to 0.9Amax. The uncertainties in λmax are about 0.1 nm. Fluorescence decay data were measured with the time correlated single photon counting technique (Edinburgh Instrument FL-900) with a PicoQuant subnanosecond Pulsed LED PLS 370 (emitting at 370 nm) < 600 ps fwhm. Fluctuations in the pulse and intensity were corrected by making an alternate collection of scattering and sample emission. The quality of the fits was determined by the reduced χ2. For the best fit χ2, it must be around 1.0.50

Novaira et al. TRES were generated from a set of emission decay times taken at 5 nm intervals spanning the fluorescent spectrum (typically 40 decays). To resolve the convolution with the instrument response on the time-resolved decay data at each emission wavelength, a multiexponential fit was used. The purpose of these fits is simply to represent the decay curves, and no physical meaning is ascribed to the derived exponential parameters.38 Two components were generally required to obtain a satisfactory fit to the data. TRES have been constructed following the procedure described.51 TRANES spectra, which are modified versions of TRES, were constructed by normalizing the area of each spectrum in TRES such that the area of the spectrum at time t is equal to the area of the spectrum at the shorter t.46-48 Steady-state anisotropy measurements were performed in a Hitachi 2500 spectrofluorometer with a Glan-Thomson polarizer for the excitation and emission analyzer. In the micellar media, background fluorescence, as well as light scattering, was removed by subtraction of a spectrum recorded on a blank solution. Fluorescence anisotropy values were obtained at 25 °C using the expression ) (IVV - GIVH)/(IVV + 2GIVH) where IVV and IVH are the vertically and horizontally polarized components of PRODAN after excitation by vertically polarized light and G is the sensitivity factor of the detection system.38 The stock solutions of AOT and BHDC in the hydrocarbon solvent were prepared by mass and volumetric dilution. To obtain optically clear solutions, they were shaken in a sonicating bath. To introduce the probe, a concentrated solution of PRODAN was prepared in acetonitrile (Sintorgan HPLC quality). The appropriate amount of this solution to obtain the desired final concentration of PRODAN in the micellar system, i.e., 5 × 10-6 M, was transferred into a volumetric flask, and the acetonitrile was removed by bubbling dry N2. n-Heptane or benzene was added to the residue, and the resulting solution was used to prepare the surfactant-containing samples. The appropriate amount of stock surfactant solution to obtain a given concentration of surfactant in the micellar media was transferred into the cell, and the water was added using a calibrated microsyringe. The amount of water present in the system is expressed as the molar ratio between water and the surfactant (W ) [H2O]/[surfactant]). Results and Discussion In a recent work26 with Kamlet and Taft’s solvatochromic comparison method,52 we showed that PRODAN’s absorption and emission bands are sensitive to the polarity-polarizability (π*) and the hydrogen donor ability (R) of the media. The bands are more sensitive to these parameters for PRODAN in the excited state. In this work, we show that the PRODAN emission spectrum in water shows only one band and no dual fluorescence is present. The fluorescence lifetime values obtained are consistent with the known superposition of two π f π* transitions and are not due to other excited-state process. Also, we have demonstrated that PRODAN in water aggregates because of its very low solubility. On the other hand, although the PRODAN emission feature has been used to characterize AOT RMs,20-24 the complexity in its photophysics means that the nature of the PRODAN emitting species is not yet completely understood. We also believe that the lack of sufficient experimental evidence in organized media is the reason for the controversy about the nature of the PRODAN emitting state or states.



Figure 1. Normalized absorption and emission spectra of PRODAN in n-heptane. [PRODAN] ) 5 × 10-6 M.

Figure 1 shows the absorption and emission spectra of PRODAN in n-heptane, the nonpolar solvent used to create AOT RMs. The emission spectrum consists of a single band that is independent of excitation wavelength. This indicates that only one species is present in this nonpolar solvent. PRODAN in n-heptane was also explored using TRES. The experiments were carried out at λexc ) 342 nm and at two emission wavelengths: λem ) 390 and 420 nm. The fluorescence decay of the molecule in n-heptane fits well to a single-exponential function with a fluorescence lifetime of τ ) 0.14 ( 0.05 ns, χ2 ) 1.01, which is close to the one reported in the literature.20 This result demonstrates that only one PRODAN species emits in n-heptane. In benzene, the organic solvent used to prepare the BHDC RMs, the fluorescence decay of PRODAN, fits nicely to a singleexponential function with a fluorescence lifetime of 2.42 ( 0.05 ns, χ2 ) 1.04 at λem ) 390 and 418 nm. In both organic solvents, the dual fluorescence is not observed and the PRODAN emission comes from an LE state because of the low polarity of these solvents.17,42 Studies in the Reverse Micelle Systems at W ) 0. Figure 2A,B shows the absorbance and steady-state emission spectra of PRODAN in water/AOT/n-heptane at W ) 0, respectively. As the AOT concentration increases, there is practically no variation in the maxima position of the absorption band (λmax ) 341 nm) and there is a bathochromic shift and an increase in the intensity of the emission band. The red shifts show that the polarity of the microenvironment sensed by PRODAN increases26 in comparison to that in n-heptane because of the gradual incorporation of the molecule into the AOT RMs interface. Moreover, different emission maxima are observed at different excitation wavelengths (results not shown) because of the PRODAN heterogeneity in the ground state of the molecule.53 Indeed, PRODAN emits from two different microenvironments: the organic pseudophase and the RMs interface because of the partition process that PRODAN undergoes, as was suggested in the literature.20,23 Table 1 shows the fluorescence lifetime of PRODAN at an AOT concentration of 7 × 10-2 and 0.2 M and at three emission wavelengths: λem ) 390, 420, and 500 nm. At λem ) 390 and 420 nm, the emission decay fits well to a double-exponential function with the shorter component (τ1) that corresponds to the fluorescence lifetime of PRODAN in n-heptane; the other (τ2) corresponds to the probe in the AOT RM interface. Note that the contribution of the lifetime of PRODAN in the RM interface (τ2) increases with the surfactant concentration. At λem ) 500 nm, the contribution of PRODAN from n-heptane

Figure 2. (A) Absorbance and (B) steady-state emission (λexc ) 330 nm) spectra of PRODAN varying [AOT] at W ) 0 in water/AOT/nheptane reverse micelles. [PRODAN] ) 5 × 10-6 M.

disappears because the molecule has no emission at this wavelength (Figure 1). The emission decay is monoexponential with only the component (τ2) detected at the AOT RMs interface. PRODAN in the cationic BHDC/benzene RMs shows no variation in the absorption spectra with the increase in the surfactant concentration (results not shown) as is found in the AOT RM (Figure 2A). Figure 3 shows the steady-state emission spectra of PRODAN in BHDC/benzene RM at λexc ) 350 nm. The figure shows a red shift and a decrease in the PRODAN emission as the surfactant concentration increases. Also, there is an isoemissive point at λ ∼ 430 nm. The results also prove that PRODAN undergoes a partition process between the two pseudophases. Interestingly, the intensity of its emission band decreases as the BHDC concentration increases, as was also observed for this probe dissolved in another RMs made with different cationic surfactants.24 It is known that PRODAN in cationic RMs of didodecyldimethyl ammonium bromide and cetyltrimethylammonium bromide22,24 interacts with the cationic polar head through its aromatic π electron cloud. Similar interaction is found with another aromatic probe, QB, in the cationic BHDC RMs.13 Indeed, the decrease in the PRODAN emission intensity as the BHDC concentration increases may be due to the specific interaction with the cationic surfactant. Table 1 also shows the fluorescence lifetime for [BHDC] ) 0.2 M and λem ) 420 nm. The fluorescence decay exhibits a biexponential decay with τ1 that corresponds to PRODAN in benzene and the shorter component, τ2, which we assign to

752 J. Phys. Chem. B, Vol. 111, No. 4, 2007

Novaira et al.

TABLE 1: Fluorescence Lifetimes (τ) of PRODAN in Water/AOT/n-Heptane and Water/BHDC/Benzenea [Surf]/M [AOT] ) 7 × 10

-2

[AOT] ) 0.2

λem

τ1/ns (%)b

τ2/ns (%)b

0

390 420 500 390 420 500 390 420 500 390 420 500 420 420 495

0.19 ( 0.02 (81) 0.18 ( 0.02 (31)

2.18 ( 0.07 (19) 2.49 ( 0.05 (69) 3.02 ( 0.02 2.12 ( 0.05 (37) 2.54 ( 0.04 (83) 2.85 ( 0.02 1.30 ( 0.06 (15) 2.60 ( 0.02 (44) 2.70 ( 0.02 (0.057)c 1.35 ( 0.04 (30) 2.50 ( 0.08 (33) 2.82 ( 0.05 (0.059)c 1.07 ( 0.02 (60) 1.07 ( 0.02 (24) 3.12 ( 0.04 (0.036)c

0

[AOT] ) 7 × 10-2

10

[AOT] ) 0.2

10

[BHDC] ) 0.2 [BHDC] ) 0.2

0 10

0.22 ( 0.03 (63) 0.20 ( 0.07 (17) 0.16 ( 0.02 (85) 0.17 ( 0.02 (25) 0.14 ( 0.02 (70) 2.38 ( 0.03 (40) 2.59 ( 0.02 (76)

τ3/ns (%)b

χ2

0.81 ( 0.03 (31) 0.82 ( 0.06 (-0.036)c

1.02 1.23 1.30 1.15 1.04 1.14 0.98 1.02 1.04 1.16 1.28 1.04 1.16 1.15 1.32

0.70 ( 0.03 (67) 0.71 ( 0.03 (-0.054)c 0.90 ( 0.02 (-0.035)c

[PRODAN] ) 5 × 10 M. Values in parentheses are the contribution of the species obtained from the biexponential or triexponential fitting. Values in parentheses are the pre-exponential factor of the species obtained from the biexponential fitting. a

c

W

-6

b

PRODAN in the RM cationic interface. The interaction invoked above may explain why the fluorescence lifetime of PRODAN inside the cationic RMs is lower than the value in benzene. The partition of PRODAN between the AOT or BHDC RMs and the external solvent will be treated within the framework of the pseudophase model.5,23,53-60 This model considers the RMs as a distinct pseudophase whose properties are independent of the AOT concentration and are only determined by the value of the characteristic parameter W. In this model, only two solubilization sites are considered, that is, the external solvent and the RM interface (i.e., all the surfactant molecules). In this way, the distribution of PRODAN between the micelles and the external solvent pseudophase defined in eq 1 can be expressed in terms of the partition constant Kp shown in eq 2.

PRODANf h PRODAN#b Kp )

[PRODAN]#b [PRODAN]f

(1)

(2)

The terms in brackets represent free (f) and bound (b) PRODAN in terms of local concentrations. If [PRODAN]b is the analytical (bulk) concentration of micelle bound substrate, eq 3 holds.

[PRODAN]#b )

[PRODAN]b [Surf]

[PRODAN]b [PRODAN]f[Surf]

sample at the working excitation wavelength is low,62,63 eq 5 can be deduced:53

I)

I0(φf + φbKp[Surf]) (1 + Kp[Surf])

(5)

(3)

and hence Kp can be expressed as in eq 4

Kp )

Figure 3. Steady-state emission spectra (λexc ) 350 nm) of PRODAN varying [BHDC] at W ) 0 in water/BHDC/benzene reverse micelles. [PRODAN] ) 5 × 10-6 M.

(4)

where [PRODAN]b is the analytical concentration of the substrate incorporated in the RMs, [PRODAN]f is the concentration of the substrate in the organic solvent, and [Surf] is the micellized surfactant (total AOT concentration minus the “operational CMC” = 10-4 M obtained using the absorption or emission bands’ shift with the AOT concentration for different molecular probes at different water content).5,61 This equation applies at a fixed value of W and when [PRODAN]T , [Surf] where [PRODAN]T is the probe analytical concentration. The value of Kp can be determined from the change in the fluorescent intensity of the probe with the surfactant concentration measured at a given wavelength. If the analytical concentration of the probe is kept constant and the absorbance of the

where I0 is the incident light, If and Ib are the fluorescent intensities when the probe is present in the external solvent and the disperse pseudophase, respectively, I is the fluorescence intensity measured at the surfactant concentration considered, and φf and φb are the fluorescent quantum yields of PRODAN in the organic solvent and bound to the RM interface, respectively. Figure 4 shows the variation of the steady-state anisotropy for PRODAN with the BHDC concentration at W ) 0. A similar trend is found for PRODAN in the AOT RMs. The increase in the values shows that the molecular probe senses a microviscosity higher inside the RMs pseudophase than the one in the bulk solvent. The trend observed in Figure 3 allows us to also determine the value of Kp. Equation 6 is deduced from the additivity law for anisotropy38,53,64 and shows the relationship between and the surfactant concentration.

)

+ Kp[Surf] (1 + Kp[Surf])

(6)


Figure 4. Steady state anisotropy of PRODAN in water/BHDC/ benzene reverse micelles varying [BHDC] at W ) 0. Dotted line values fitted with eq 6.

TABLE 2: Equilibrium Constants (Kp) for the Partition of PRODAN in Water/AOT/n-Heptane in Water/BHDC/ Benzene Reverse Micellesa reverse micelles AOT BHDC

W

KpEm/M-1 b,c

KpAn/M-1 d

0 10 20 0 10 20

2.5 ( 0.4 2.1 ( 0.4 2.7 ( 0.5 8.1 ( 0.5 16.4 ( 1.1 17.4 ( 1.1

1.9 ( 0.5 9.7 ( 0.9

a [PRODAN] ) 5 × 10-6 M. T ) 25 °C. b Equation 5. c K value p for AOT obtained from ref 21 is Kp ) 1 ( 0.2. d Equation 6.

where is the anisotropy of the mixture, and are the anisotropies of the free and bound PRODAN species, respectively, and [PRODAN]T is the analytical probe concentration. Table 2 shows the value of Kp in both RMs calculated using a least-squares fit of eqs 5 and 6. For AOT/n-heptane RMs, λem ) 412 nm, and for BHDC/benzene RMs λem ) 417 nm. There is a very good agreement in the Kp values obtained from the emission and anisotropy data. The value of Kp for PRODAN in AOT/n-heptane RMs are in agreement with previous results found in the literature (see footnote in Table 2) but were obtained using a different approach.23 On the other hand, it is of interest to highlight that the Kp value is almost four times larger in the BHDC/benzene than in the AOT/n-heptane RMs, reflecting the specific interaction between the cationic polar head of the surfactant and the PRODAN aromatic ring invoked before.22 It seems that this specific interaction is a powerful driving force for the molecular probe to reach the cationic RM interface. We have shown previously13 that a similar interaction is the principal reason for the aromatic probe QB to exist exclusively at the BHDC/ benzene interface while this molecular probe partitions between the organic and the micellar pseudophase in the AOT/benzene reverse micelles system.61 Studies at W ) 10 and 20. The steady-state emission behavior of PRODAN in water/AOT/n-heptane RMs varying AOT concentration is conspicuously different upon the water addition. The absorption spectra are quite similar to the ones in Figure 2A, with the maxima at the same wavelength as at W ) 0 (λmax ) 341 nm). It shows no dependence on the surfactant concentration. Moreover, even at W ) 20, the λmax does not


Figure 5. Steady-state emission spectra (λexc ) 330 nm) of PRODAN varying [AOT] at W ) 10 in water/AOT/n-heptane reverse micelles. [PRODAN] ) 5 × 10-6 M.

match the value of the maxima in bulk water, λmax ) 360 nm.26 Figure 5 shows the steady-state emission spectra at W ) 10. The same trend is observed at W ) 20. It is clear that dramatic changes occur in the emission profile varying the surfactant concentration. Contrary to what is found at W = 0 (Figure 2B) where the emission band shifts bathochromically and its intensity increases as the AOT concentration increases, at W = 10 and 20, the spectra show that, as the AOT concentration increases, the band that peaked at ∼423 nm increases at the expense of the band that peaked at ∼390 nm. Also, a new band emerges near 500 nm, which peaks 25 nm at higher energy in comparison with the emission maxima in water (λem ) 520 nm),26 and there is a clear isoemissive point at λ ) 398 nm. In previous works,20-23 these spectral changes were attributed to the existence of PRODAN in a variety of sites where the partition may be influenced by electrostatic, hydrophobic, dipolar, and cation-π interactions. Two different assignments were given for the longer wavelength component of the fluorescence spectra. Some authors21,22 assigned it to the emission of probe molecules deep within the free water pool, whereas other authors20,23 assigned this band to the emission of the molecules associated to the fully hydrated microaggregate interface, with the hydrophobic tails projected toward the organic solvent and not to the RMs water pool because of the low water solubility of PRODAN. We will show a different interpretation of the origin of this emission band in the next section. Figure 6 shows the steady-state emission spectra of PRODAN in water/BHDC/benzene RMs at W ) 10. A similar trend is also observed at W ) 20. The results show that, as BHDC concentration increases, there is a decrease in the intensity and a bathochromic shift of the band at λem ∼ 420 nm (PRODAN in benzene). Also, a new band emerges around 495 nm, and there is a clear isoemissive point at λ ) 458 nm. Indeed, these results also show that the distribution between the two pseudophases that the molecular probes undergoes at W ) 0 is still present at W ) 10 and 20 in both RMs. Kp values were evaluated using eq 5 at λem ) 425 nm for AOT RMs and at λem ) 417 nm for BHDC RMs. The results are shown in Table 2. Practically, there is no variation of Kp values with the water content in the AOT RMs probably because PRODAN resides mostly at the oil side of the RMs interface as other authors23 claimed. Moreover, the fact that Kp is independent of the water content rules out the possibility that PRODAN can exists in the AOT RMs water pool as other authors have suggested.21,22

754 J. Phys. Chem. B, Vol. 111, No. 4, 2007

Figure 6. Steady-state emission spectra (λexc ) 380 nm) of PRODAN varying [BHDC] at W ) 10 in water/BHDC/benzene reverse micelles. [PRODAN] ) 5 × 10-6 M.

If this would be the case, Kp should change over the whole W range studied and increase notably upon the water pool formation (W > 10). Note that the value of Kp is very small; thus, even at [AOT] ) 0.2 M, the emission comes mainly from PRODAN in the n-heptane pseudophase (Table 1, λem ) 390 nm). For BHDC RMs, the Kp values are greater at W ) 10 and 20 than at W ) 0, probably reflecting the greater micropolarity of the cationic interface upon the water addition.13 Indeed, Kp value increases up to W ) 10 and then remain unchanged. It seems to us that, contrary to that observed in the AOT RMs, PRODAN in BHDC RMs resides mainly at the waterside of the interface. The fact that Kp changes with W until the complete RM interface hydration (W ) 10) and remains unchanged after that is consistent with a probe that exists exclusively at the water side of the RM interface.5,13,49 It was not possible to fit the experimental data of vs [Surf] at W ) 10 and 20 using eq 6 as it was done at W ) 0 because of the data dispersion, which makes the fitting not statistically correct. In conclusion, the results in both RMs media are consistent with the molecular probe molecules that exist at the RMs interface instead of the water pool. Sengupta et al.20 carried out a preliminary study of PRODAN lifetimes in different RMs, and they explain their results considering the heterogeneity in microenvironments where PRODAN can emits. Nevertheless, these studies lack information on the pre-exponential factors of their biexponential fluorescence decay. Here, we perform a study of the PRODAN fluorescence lifetimes at different surfactant concentrations. For AOT RMs, the decay is monitored around the following emission wavelengths: 390, 420, and 500 nm (Figure 5) and for BHCD RMs around λem ) 420 and 495 nm. The results are given in Table 1 with the pre-exponential factor included for the band that peaked at lower energy. A good deal of information can be extracted from Table 1. Hence, we discuss first the results obtained at λem ) 390 and 420 nm for AOT. At [AOT] ) 7 × 10-2 M and λem ) 390 nm, the emission decay fits nicely to a double-exponential function with a major contribution coming from the fluorescence lifetimes of PRODAN in n-heptane (τ1). We assign the other component (τ2) to the contribution of PRODAN at the interface of the AOT RMs. At λ ) 420 nm, the emission decay fits well to a three-exponential function with a new component of τ3 of 0.81 ns for which significance and interpretation will be discussed later in detail. Moreover, at [AOT] ) 0.2 M and λem ) 390 nm, the contribution of PRODAN at the interface (τ2) increases in

Novaira et al. comparison with the contribution shown at [AOT] ) 7 × 10-2 M. On the other hand, at [AOT] ) 0.2 M and λem ) 420 nm, the contribution of PRODAN from n-heptane (τ1) is no longer present, and although the emission decay fits to a double exponential function, the two lifetimes (τ2 and τ3) correspond to PRODAN at the interface of the RMs (Vide infra). The results for PRODAN at [BHDC] ) 0.2 M, W ) 10, and λem ) 420 nm show that the emission decay fits nicely to a double-exponential function with fluorescence lifetimes that correspond to PRODAN in benzene (τ1) and PRODAN in the RMs interface (τ2). A completely different and novel situation comes out when the longer wavelength emission band is analyzed in the different RMs: λem ) 500 nm for AOT and λem ) 495 nm for BHDC. Surprisingly, with both surfactants, the fluorescence decays fit well to a double-exponential function, but the shorter component (τ3) has a negative pre-exponential factor (Table 1). A priori, the observation of negative amplitudes indicates that, before the radiative deexcitation, a fast process compared to the emission lifetime exists, leading to an increase in the emitting population observed by fluorescence. In other words, there is an excited-state process leading to a new emitting state that is different from the initially excited state, which can explain the PRODAN dual fluorescence observed in the RMs media at W ) 10 and 20.38,65,66 This possibility is considered and discussed in the next section. TRES of PRODAN in Reverse Micelles Systems at W ) 0, 10, and 20. In order to gain more information on the photophysics of PRODAN inside the different RMs, we study the PRODAN behavior using time-resolved fluorescence measurements in both RMs media at W ) 0 and 10 and at [Surf] ) 0.2 M. Figure 7A,B shows the fluorescence decays measured at several wavelengths across the emission spectrum of PRODAN in AOT/n-heptane RMs with [AOT] ) 0.2 M and W ) 0 and 10, respectively. A similar trend is observed in BHDC/benzene RMs at both W values. Figure 7A shows that the decays are faster at shorter emission wavelengths and they are also wavelength-dependent. However, no rise in intensity is observed, which is typical for a fluorophore that only experiences groundstate heterogeneity, as was shown in the previous section. On the contrary, for the decays at W ) 10 shown in Figure 7B, even when they are also wavelength-dependent, there is evidence of a rise time at longer wavelengths (e.g., 515 nm). This is characteristic of an excited-state process where a rise time can only be observed if the emission is not directly excited but rather forms from a previously excited state.38,65,66 The time-dependent decays were used to construct the TRES, and the results are shown in Figure 8A-C for AOT/n-heptane RMs at W ) 0 and 10 and water/BHDC/benzene RMs at W ) 10, respectively. For all the RMs, the surfactant concentration is 0.2 M. Figure 8 shows that the TRES is different depending on the water content. In both RMs at W ) 0 (results in BHDC/ benzene not shown), the emission spectra shift progressively to longer wavelengths at longer times and the band shape shows little or no variation with time. Also, at W ) 0, the fluorescence lifetimes obtained change over the whole emission spectra. In other words, the lifetimes are emission wavelength-dependent. These facts are in agreement with the continuous model for a spectral relaxation, which can explain TRES that came from a heterogeneity in the ground state and/or a multitude of solvent fluorophore interactions.38,65 On the other hand, TRES obtained at W ) 10 in anionic and cationic RMs (Figure 8 B,C) show a type of relaxation that can be represented by the two-state model.38,65 In this model, the



Figure 7. Normalized fluorescence decays of PRODAN in water/AOT/ n-heptane reverse micelles at (A) W ) 0 and (B) W ) 10. [PRODAN] ) 5 × 10-6 M.

initially excited short-wavelength state is assumed to change to the longer-wavelength relaxed state, each one displaying a discrete emission spectrum. This is the model used to describe excited-state processes such as chemical reactions or formation of the ICT state. Moreover, the two lifetimes shown in Table 1 at [Surf] ) 0.2 M are constant over the emission spectrum (410-600 nm) with only the relative amplitudes varying with the emission wavelength and with a negative amplitude for the short-lifetime component in the lower energy region corresponding to 475-550 nm. Identity of PRODAN Excited-State Processes in Reverse Micelles at W ) 10 and 20. As was shown in the Introduction, PRODAN is a kind of molecule that is capable of simultaneously creating LE and ICT excited states, which makes its photophysics very complex and hard to understand.17,30,33,35-40 Here, we show that PRODAN at the interface of the RMs with [AOT] > 5 × 10-2 M and [BHDC] > 0.1 M and at W ) 10 and 20 undergoes an excited-state process. We attribute this process as being due to a dual fluorophore emission from two different states: the LE with the emission band at λmax ) 420 nm (water/AOT/n-heptane) or λmax ) 424 nm (water/BHDC/ benzene) and the ICT state with the emission band at λmax ) 500 nm (water/AOT/n-heptane) or λmax ) 490 nm (water/ BHDC/benzene). Interestingly enough, these emission bands have similarities in their excitation spectra (results not shown). Moreover, the excitation spectra of the PRODAN’s LE and ICT

Figure 8. Normalized TRES of PRODAN in water/AOT/n-heptane reverse micelles at (A) W ) 0; (B) W ) 10, [AOT] ) 0.20 M; and in (C) water/BHDC/benzene. [BHDC] ) 0.20 M at W ) 10. [PRODAN] ) 5 × 10-6 M.

bands are identical with the absorption spectrum, which means that they have a common precursor (the ground-state molecules that exist at the RMs interface), a fact that it is necessary for an excited-state process that can lead to the dual fluorescence being observed.67 Therein, it is not likely that the emission band at λem ∼ 500 corresponds to the emission due to the excitation of the PRODAN ground-state molecules that exist in the water pool of the RMs systems, as was previously suggested for the AOT RMs.20-23 It seems to us that the origin of these two emission bands is the dual fluorescence that PRODAN at the RMs interface presents and not because of the different locations of the molecular probe (in its ground state) inside the RMs media invoked before.20-23 In the literature68 is shown that PRODAN dissolved in the gel phase of small unilaminar vesicles made of 1,2-dipalmitoyl-

756 J. Phys. Chem. B, Vol. 111, No. 4, 2007 1-sn-glycerol-3-phosphocholine (DPPC) presents dual fluorescence in the steady-state emission spectra at 25 °C. The authors suggested that this dual fluorescence could be due to the emission from the LE and the TICT state of PRODAN, the contribution depending on the bilayer phase: LE from the gel phase and TICT from the liquid phase. However, a TRES study was not performed in order to gain more insight about the nature of the PRODAN excited-state process. On the other hand, a study performed using a related molecule, LAURDAN, in homogeneous media, reveals that LAURDAN undergoes an excited-state reaction that can be attributed to the interconversion between the excited LE and ICT states, which could involve the rotation of the dimethylamino groups (TICT). In polar media, LAURDAN is subjected to an important excited-state stabilization due to the solvent molecules, and most of the fluorescence of the probe arises from the ICT or TICT states.66 At present, follow the discussion about our explanation of the PRODAN fluorescence lifetime shown in Table 1 for the different RMs at W ) 10. In the water/AOT/n-heptane RM, at [AOT] ) 7 × 10-2 Μ and λem ) 420 nm, the emission decay is triexponential. It is clear that τ1 ) 0.17 ns corresponds to PRODAN in the n-heptane pseudophase, but the origin of the other two lifetimes (τ2 and τ3) needs a detailed discussion. The lifetimes obtained at λem ) 500 nm, where the contribution from PRODAN in n-heptane is no longer present, show that the shortlifetime component (τ3) has the negative pre-exponential factor. Hence, this lifetime is associated with the initially excited species (LE) in the RM interface, and the long lifetime component (τ2) is assigned to the newly formed species (ICT).38,65 The same is valid at [AOT] ) 0.2 M. On the other hand, the PRODAN emission decay in the water/ BHDC/benzene RMs is always biexponential at both wavelengths. Analyzing the results at the λem ) 495 nm band, the short-lifetime (τ3) component also has the negative amplitude like the anionic RMs. Thus, this component can be assigned to the PRODAN LE state. The τ2 ) 3.12 ns is then attributed to the ICT species of the molecular probe. At λem ) 420 nm, the contribution of the τ1 ) 2.59 ns is 76% because this is an average value between the lifetimes of the PRODAN ICT state (around 3 ns) and the lifetime of PRODAN in benzene (τ ) 2.38 ns). Taking into account the Kp values shown in Table 1, PRODAN in benzene contributes with approximately 20% of the total emission at λem ) 420 nm. Because our equipment cannot resolve these two close lifetimes, it gives the average value of the fluorescence lifetime. At λem ) 495 nm, the contribution in the emission spectra from PRODAN in benzene is practically nothing (Figure 6); thus, the lifetime of τ2 ) 3.12 ns is correctly assigned to the ICT state at the RM interface. Another possible mechanism in the PRODAN excited state that can account for the results shown in this work is the solvent relaxation, where the initially excited state relaxes to a solvent relaxed state.38,69,70 It is known that a TICT process is generally controlled by the medium polarity rather than their viscosity; the opposite is admitted for a solvent relaxation process.38,65,71 In other words, the TICT process is favored in a high-polarity medium and the solvent relaxation process is favored in a high viscosity medium. In RMs without the addition of water, W ) 0, it is recognized that the microviscosity at the interface is higher than that in the presence of water and the micropolarity is lower. The micropolarity increases with the water addition up to a W value of around 10 and then remains constant.5,13,72 Figures 2B and 5 show that the appearance of the new lowenergy emission band occurs precisely when water is sequestrated inside the RMs. In this case, the microviscosity is lower

Novaira et al. and the micropolarity is higher than their value at W ) 0. Consequently, without the addition of water, only the partition process that PRODAN undergoes is responsible for the emission changes, as was previously demonstrated. Thus, it is not likely that the PRODAN emission feature in the interface of the RMs when water is sequestrated in the water pool could be due to the medium relaxation process, as will be demonstrated in the next section. TRANES in Reverse Micelles Systems at W ) 0, 10, and 20. As the standard interpretation of TRES assumes a prior knowledge of the number of fluorescent species in the ground state, usually a single species, and as this assumption may fail if the fluorophore is present in a complex environment such as a RMs, we decided to also use time-resolved area normalized emission spectroscopy (TRANES).46-48 This is a step forward, and with this analysis, it is possible to determine the number of species in the sample that contribute to the observed fluorescence emission without assumption of ground- or excitedstate kinetics.46-48 Here, we use this novel and powerful method to know and to confirm how many PRODAN species emit from the different AOT and BHDC RMs. Figure 9A-C shows the TRANES spectra of PRODAN in AOT/n-heptane at W ) 0 and 10 and in BHDC/benzene RMs at W ) 10, respectively. In both systems, the surfactant concentration is 0.2 M. Figure 9A shows the existence of one isoemissive point located at λ ) 417 nm, which indicates that there are only two species that emit, one in the organic pseudophase and the other at the RM interface, both from an LE state. The presence of this isoemissive point in the TRANES spectra rules out the possibility that the PRODAN emission feature inside the RMs is due to the medium relaxation process where only a shift of the band with time would be observed.46-48 A different situation is shown in Figure 9B, which shows two isoemissive points, one peaked at λ ) 427 nm and the other at λ ) 485 nm. Therefore, in water/AOT/ n-heptane RM at W ) 10, there are three PRODAN species that emit, each one having different fluorescence lifetimes (Table 1): (i) PRODAN in n-heptane; (ii) PRODAN LE state, and (iii) PRODAN ICT state species at the RMs interface. On the other hand, Figure 9 shows that, in the water/BHDC/ benzene RM at W ) 10, despite the existence of three PRODAN species that emit in this RM, only one isoemissive point is detected (λ ) 465 nm). We interpret this result because two of the species: PRODAN in benzene and the PRODAN ICT state at the RMs interface have similar fluorescence lifetimes (see Table 1). A necessary condition for multiple isoemissive points is that the lifetimes of the species be significantly different from each other.46-48 Thus, our results confirm that it is not likely that the origin of the PRODAN photophysic in RMs is due to multiple PRODAN species that emit from several sites within the aggregates, as is suggested in the literature.21-24 Undoubtedly, PRODAN exists in two different environments: the organic solvent, where the emission came from the PRODAN LE state, and the RMs interface, where they undergo an excitedstate process that yields the PRODAN ICT state. Thus, inside the RMs, the emission came from both the PRODAN LE and ICT states. It seems that the RMs interface is a unique microenvironment to create an ICT state. More on the Effect of the Water Addition. Figure 10 shows the emission spectra of PRODAN in AOT/n-heptane at [AOT] ) 0.2 M and λexc ) 330 nm as a function of W. The results show that, at W ∼ 4, the new low-energy emission band that we assign to the PRODAN ICT state emerges. Upon the water addition, it shifts slightly to the red, which reflects the increase



Figure 10. Steady-state emission spectra (λexc ) 330 nm) of PRODAN in water/AOT/n-heptane reverse micelles varying the water content, W. [AOT] ) 0.2 M. [PRODAN] = 5 × 10-6 M.

Figure 9. Normalized TRANES of PRODAN in water/AOT/n-heptane reverse micelles at (A) W ) 0; (B) W ) 10, [AOT] ) 0.20 M; and in (C) water/BHDC/benzene [BHDC] ) 0.20 M at W ) 10. [PRODAN] ) 5 × 10-6 M.

in the micropolarity of the RM interface.5,13,72 After W > 10, the band remains unchanged. Note that, even at the highest W ) 40, when there is a very well defined aqueous pool,5 the dual florescence is still present and the ICT band peaks at λmax ) 500 nm, far from the value in water.26 This fact also rules out the possibility that the solvent relaxation processes is the cause of the PRODAN emission feature and that the ICT species exists deep inside the water pool. Recently,16 the behavior of the ICT state of p-N,N-dimethylaminobenzoic acid and p-N,N-dimethylaminobenzonitrile in water, methanol, or acetonitrile/AOT/n-heptane, RMs systems was studied. The authors have shown that hydrogen bond interactions between the probes and the hydrogen bond donor solvents, water and methanol, may play a major role in the fluorescence quenching of the fluorophores’ ICT state observed by the deactivation via internal conversion (IC). Furthermore,

it was shown that the IC de-excitation rate of the hydrogenbonded ICT state is much larger than that of a non-hydrogenbonded ICT state.73 In our system, it is clear that this IC deexcitation pathway, which would cause an ICT emission quenching, is not likely to occur to PRODAN because its ICT emission band is still observed even at W ) 40. PRODAN has a hydrogen-bond acceptor group in its structure, the carbonyl group (Scheme 1), so we predict that the probe molecule exists at the interface of the AOT RM with its acceptor group toward the hydrocarbon tails, buried from the water pool and the negative charge of the AOT polar head. With this orientation, water cannot interact through hydrogen bonding with the probe molecules and the PRODAN ICT emission is not quenched. On the contrary, the results for PRODAN in the BHDC/benzene system at [BHDC] ) 0.3 M varying W (not shown) show that its ICT band is not as well defined as in the AOT system and its intensity decreases notably up to W ) 22, the maximum amount of water that this system accepts. It seems to us that PRODAN exists in the cationic interface with the carbonyl group toward the water pool, and the dimethylamino group remains buried at the interface and far from the positive charge of the BHDC polar head. With this orientation, the hydration water molecules at the RM interface can interact with the carbonyl group of PRODAN through hydrogen-bond interactions with the consequent quenching of the PRODAN ICT emission. It is worthwhile to emphasize that previous results26 for PRODAN in bulk water show that the molecule does not present the dual fluorescence in its emission spectra, its fluorescence decay does not show the negative amplitude in the shorter fluorescence lifetimes, and the emission maxima in the RMs are far from the emission maxima of the molecular probe in water, facts that discard the possibility of an excited-state process for PRODAN in the water pool of the RMs. Conclusions In this work, we investigate the photophysics of the molecular probe PRODAN in water/AOT/n-heptane and water/BHDC/ benzene RMs using absorption, steady-state, TRES, and TRANES spectroscopies. The results show the existence of a partition process for PRODAN between the organic pseudophase and the different RMs interfaces. The partition constants, Kp, are quantified from the changes in the emission spectra and the steady-state anisotropy () with the surfactant concentration in both RMs.

758 J. Phys. Chem. B, Vol. 111, No. 4, 2007 The Kp value is larger in the BHDC than in the AOT RMs, which reflects the specific interaction between the cationic polar head of the surfactant and the aromatic ring of PRODAN. At W ) 0, the results are consistent with the PRODAN emission from only a LE excited state of the molecular probe at the organic solvent and at the RMs interfaces. TRES data at W ) 0 fit a continuous model characteristic for a heterogeneous ground state. On the other hand, at W ) 10 and 20, with increasing surfactant concentration, a new emission band in the low-energy part of the spectra emerges and it is assigned to the emission of a PRODAN ICT state at the interface of the aggregates. Lifetime measurements and TRES experiments corroborate these results. The results show biexponential emission decay with a negative pre-exponential factor for the short component of the PRODAN fluorescence lifetime. This is characteristic of an excited-state process that yields a new emitting state, the ICT state, different from the initially excited state (LE). TRANES results show one or two isoemissive points in the different systems depending on if there are two PRODAN species that emit in the RMs as at W ) 0 (both from a LE excited state) or three PRODAN species as in water/AOT/nheptane at W ) 10 and 20, each one with different fluorescence lifetimes. The PRODAN species that emit at these W values are (i) one in the organic solvent, where the emission came from the LE state and (ii) two at the RMs interfaces, the LE and the ICT excited state. In this way, the LE species undergoes an excited-state process that yields its ICT state with the consequent PRODAN emission from both excited states (dual fluorescence). As the molecular probe does not show dual fluorescence either in n-heptane and benzene or in water, it seems to us that the RMs interfaces represent a unique environment for PRODAN to experience the ICT process. This is a very interesting result because it seems to us that RMs can be good models to study the electron-transfer processes that occur with biological molecules embedded in biological membranes. Acknowledgment. Financial support from the Consejo Nacional de Investigaciones Cientı´ficas y Tećnicas (CONICET), Universidad Nacional de Rıó Cuarto, Fundacioń Antorchas, and Agencia Nacional de Promocioń Cientı´fica y Tećnica is gratefully acknowledged. J.J.S., M.A.B., and N.M.C. hold a research position at CONICET. M.N. thanks CONICET for a research fellowship. References and Notes (1) Fendler, J. H. Membrane Mimetic Chemistry; Wiley: New York, 1982. (2) Pileni, M. P. Structure and ReactiVity of ReVerse Micelles; Elsevier: New York, 1989. (3) De, T. K.; Maitra, A. AdV. Colloid Interface Sci. 1995, 59, 95. (4) Moulik, S. P.; Paul, B. K. AdV. Colloid Interface Sci. 1998, 78, 99. (5) Silber, J. J.; Biasutti, M. A.; Abuin, E.; Lissi, E. AdV. Colloid Interface Sci. 1999, 82, 189. (6) Monduzzi, M.; Caboi, F.; Moriconi, C. Colloids Surf., A 1997, 129130, 327. (7) Hamada, K.; Ikeda, T.; Kawai, T.; Kon-No, K. J. Colloid Interface Sci. 2001, 233, 170. (8) Kene´z, P. H.; Carlstro¨m, G.; Furo´, I.; Halle, B. J. Phys. Chem. 1992, 96, 9524. (9) Pileni, M. P. J. Phys. Chem. 1993, 97, 6961. (10) Luisi, P. L.; Straub, B. ReVerse Micelles, 8th ed.; Plenum Press: London, 1984. (11) Ballesteros, A.; Bornscheuer, U.; Capewel, A.; Combes, D.; Condoret, J. S.; Koenig, K.; Kolisis, F. N.; Marty, A.; Mengue, U.; Scheper, T.; Stamatis, H.; Xenakism, A. Biocatal. Biotransform. 1995, 13, 1. (12) Falcone, R. D.; Biasutti, M. A.; Correa, N. M.; Silber, J. J.; Lissi, E.; Abuin, E. Langmuir 2004, 20, 5732. (13) Correa, N. M.; Biasutti, M. A.; Silber, J. J. J. Colloid Interface Sci. 1996, 184, 570.

Novaira et al. (14) McNeil, R.; Thomas, J. K. J. Colloid Interface Sci. 1981, 83, 57. (15) Grabowski, Z. R. Pure Appl. Chem. 1993, 65, 1751. (16) Hazra, P.; Chakrabarty, D.; Chakraborty, A.; Sarkar, N. J. Photochem. Photobiol., A 2004, 167, 23. (17) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Chem. ReV. 2003, 103, 3899. (18) Zachariasse, K. A. Chem. Phys. Lett. 2000, 320, 8. (19) Rettig, W. Angew. Chem., Int. Ed. Engl. 1986, 25, 971. (20) Sengupta, B.; Guharay, J.; Sengupta, P. K. Spectrochim. Acta, Part A 2000, 56, 1433. (21) Karukstis, K. K.; Frazier, A. A.; Martula, D. S.; Whiles, J. A. J. Phys. Chem. 1996, 100, 11133. (22) Karukstis, K. K.; Frazier, A. A.; Loftus, C. T.; Tuan, A. S. J. Phys. Chem. B 1998, 102, 8163. (23) Lissi, E. A.; Abuin, E. B.; Rubio, M. A.; Ceroń, A. Langmuir 2000, 16, 178. (24) Karukstis, K. K.; Zieleniuk, C. A.; Fox, M. J. Langmuir 2003, 19, 10054. (25) Weber, G.; Farris, F. J. Biochemistry 1979, 18, 3075. (26) Moyano, F.; Biasutti, M. A.; Silber, J. J.; Correa, N. M. J. Phys. Chem. B 2006, 110, 11838. (27) Krasnowska, E. K.; Gratton, E.; Parasassi, T. Biophys. J. 1998, 74, 1984. (28) Parasassi, T.; Krasnowska, E. K.; Bagatolli, L. A.; Gratton, E. J. Fluoresc. 1998, 8, 365. (29) Krasnowska, E. K.; Bagatolli, L. A.; Gratton, E.; Parasassi, T. Biochim. Biophys. Acta 2001, 1511, 330. (30) Balter, A.; Nowak, W.; Pawelkiewicz, W.; Kowalczyk, A. Chem. Phys. Lett. 1988, 143, 565. (31) Catalan, J.; Perez, P.; Laynez, J.; Blanco, F. G. J. Fluoresc. 1991, 1, 215. (32) Moreno, F.; Corrales, S.; Sevilla, P.; Blanco, F. G.; Diaz, C.; Catalań, J. HelV. Chim. Acta 2001, 84, 3306. (33) Nowak, W.; Adamczak, P.; Balter, A.; Sygula, A. J. Mol. Struct. (THEOCHEM) 1986, 139, 13. (34) Ilich, P.; Prendergast, F. G. J. Phys. Chem. 1989, 93, 4441. (35) Heisel, F.; Miehe´, J. A.; Szemik, A. W. Chem. Phys. Lett. 1987, 138, 321. (36) Parusel, A. B. J.; Nowak, W.; Grimme, S.; Ko¨hler, G. J. Phys. Chem. A 1998, 102, 7149. (37) Parusel, A. B. J.; Schneider, F. W.; Ko¨hler, G. J. Mol. Struct. (THEOCHEM) 1997, 398/399, 341. (38) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic: New York, 1999. (39) Parusel, A. J. Chem. Soc., Faraday Trans. 1998, 94, 2923. (40) Balter, A.; Nowak, W.; Pawelkiewicz, W.; Kowalczyk, A. Chem. Phys. Lett. 1988, 143, 565. (41) Samanta, A.; Fessende, R. W. J. Phys. Chem. A 2000, 104, 8972. (42) Rollison, A. M.; Drickamer, H. G. J. Chem. Phys. 1980, 73, 5981. (43) Lobo, B. C.; Abelt, C. J. J. Phys. Chem. A 2003, 107, 10938. (44) Davis, B. N.; Abelt, C. J. J. Phys. Chem. A 2005, 109, 1295. (45) Karukstis, K. K.; Suljak, S. W.; Waller, P. J.; Whites, J. A.; Thompson, E. H. Z. J. Phys. Chem. 1996, 100, 11125. (46) Koti, A. S. R.; Krishna, M. M. G.; Periasamy, N. J. Phys. Chem. A 2001, 105, 1767. (47) Koti, A. S. R.; Periasamy, N. J. Chem. Phys. 2001, 115, 7094. (48) Periasamy, N.; Koti, A. S. R. Proc. Indian Natl. Sci. Acad. 2003, 69A, 41. (49) Baruah, B.; Roden, J.; Sedgwick, M.; Correa, N. M.; Crans, D. C.; Levinger, N. E. J. Am. Chem. Soc. 2006, 128, 12758. (50) O’Connor, D. V.; Phillips, D. Time-Correlated Single Photon Counting; Academic Press: New York, 1983; Chapter 6. (51) Maroncelli, M. P.; Fleming, J. R. J. Chem. Phys. 1987, 86, 6221. (52) Chalet, M. J.; Abboud, J. L. M.; Abraham, M. H.; Taft, R. W. J. J. Org. Chem. 1983, 48, 2877. (53) Falcone, R. D.; Correa, N. M.; Biasutti, M. A.; Silber, J. J. J. Colloid Interface Sci. 2006, 296, 356. (54) Silber, J. J.; Falcone, R. D.; Correa, N. M.; Biasutti, M. A.; Abuin, E.; Lissi, E.; Campodonico, P. Langmuir 2003, 19, 2067. (55) Falcone, R. D.; Correa, N. M.; Biasutti, M. A.; Silber, J. J. Langmuir 2002, 18, 2039. (56) Lissi, E.; Engel, T. Langmuir 1992, 8, 452. (57) Rubio, M. A.; Lissi, E. A. J. Colloid Interface. Sci. 1989, 128, 458. (58) Encinas, M. V.; Lissi, E. Chem. Phys. Lett. 1986, 132, 545. (59) Aguilar, L. F.; Abuin, E.; Lissi, E. Arch. Biochem. Biophys. 2001, 388, 231. (60) Falcone, R. D.; Biasutti, M. A.; Correa, N. M.; Silber, J. J.; Lissi, E.; Abuin, E. Langmuir 2004, 20, 5732. (61) Ueda, M.; Schelly, Z. A. Langmuir 1989, 5, 1005. (62) Lissi, E.; Abuin, E. In Solubilization in surfactant Aggregates; Christian, S. D., Scamehorn, J. F., Eds.; Marcel Dekker: New York, 1995.

Photophysical Behavior of PRODAN in Reverse Micelles (63) Barltrop, J. A.; Coyle, J. D. Principles of Photochemistry; Wiley & Sons: New York, 1978. (64) Xu, Y.; Mc Carroll, M. E. J. Phys. Chem. B 2005, 109, 8144. (65) Krishna, M. M. G. J. Phys. Chem. A 1999, 103, 3589. (66) Virad, M.; Gallay, J.; Vincent, M.; Meyer, O.; Robert, B.; Paternostre, M. Biophys. J. 1997, 73, 2221. (67) Rettig, W.; Chandross, A. J. Am. Chem. Soc. 1985, 107, 5617. (68) Kokzyra, K. A.; Heldt, J. R.; Gonder, G.; Kwiek, G. P.; Heldt, J. Z. Naturforsch. 2004, 59a, 809.

J. Phys. Chem. B, Vol. 111, No. 4, 2007 759 (69) Maroncelli, M. P.; Fleming, J. R. J. Chem. Phys. 1987, 86, 6221. (70) Bhattacharyya, K. Acc. Chem. Res. 2003, 36, 95. (71) Hicks, J.; Vandersall, M.; Babarogic, Z.; Eisenthal, K. B. Chem. Phys. Lett. 1985, 116, 18. (72) Correa, N. M.; Biasutti, M. A.; Silber, J. J. J. Colloid. Interface Sci. 1995, 172, 71. (73) Kwok, W. M.; George, M. W.; Grills, D. C.; Ma, C.; Matousek, P.; Parker, A. W.; Philips, D.; Toner, W. T.; Towrie, M. Angew. Chem., Int. Ed. Engl. 2003, 42, 1826.

New Insights on the Photophysical Behavior of PRODAN in Anionic

Recommend Documents