1632
J. Phys. Chem. B 2009, 113, 1632–1639
Reactive Dynamics in Confined Liquids: Interfacial Charge Effects on Ultrafast Torsional Dynamics in Water Nanodroplets Minako Kondo, Ismael A. Heisler, Jamie Conyard, Jasmine P. H. Rivett, and Stephen R. Meech* School of Chemical Sciences and Pharmacy, UniVersity of East Anglia, Norwich NR4 7TJ, U.K. ReceiVed: October 10, 2008; ReVised Manuscript ReceiVed: NoVember 20, 2008
The excited-state dynamics of a reactive dye molecule, auramine O, have been studied in nanoscale water droplets stabilized by a nonionic surfactant. Spectral dynamics were measured as a function of the radius of the water nanodroplet with 50 fs time resolution using time-resolved fluorescence up-conversion method. Qualitatively, the effect of confinement is to dramatically slow the rate of the reaction compared to that of bulk water. Data were quantitatively analyzed using the one-dimensional generalized Smoluchowski equation assuming a time-dependent diffusion coefficient. The results were contrasted with our earlier analysis of auramine O in aqueous nanodroplets stabilized by the ionic surfactant AOT. The excited-state reaction is slower in the nonionic surfactant, showing that interfacial charge is not required to suppress reactions in nanoscale water droplets. The location of the dye in the heterogeneous micelle is investigated by comparing the absorption spectra of AO in the micelle with those of a water- polyethyleneglycol mixture (to mimic the surfactant head group). The results suggest that the charged dye is located in the water phase. 1. Introduction Confined water appears widely in both man-made materials and nature. In biology the crowded environment of the intact cell presents numerous interfaces,1 channels, and pores in which water is spatially confined,2 while for proteins, a hydration layer a few molecules thick may be critical to structure and function.3-8 In materials science, porous solids and polymers can have a very high volume fraction of water dispersed in nanoscale droplets, profoundly influencing material properties.9,10 There is ample evidence that the structure and dynamics of confined water are very different from those of the bulk phase.11,12 This result is not unexpected as the properties of water are dominated by a hydrogen-bonding network which will be significantly perturbed by the presence of an interface with different (or no) H-bonding sites. It is because of the unusual properties, widespread appearance, and evident importance of confined water that an investigation of the dynamics of chemical reactions in confined aqueous media is necessary. In the previous paper in this series,13 we described the dramatic slowing down of the excited-state torsional-related dynamics of the dye auramine O (AO) in water droplets stabilized by an Aerosol OT (AOT) ionic surfactant. In this work, we extend those measurements to nonionic surfactants to investigate the role of interfacial charge and structure. One method of studying dynamics in confined water is the measurement of the solvation time correlation function, C(t). This is most often extracted from measurements of the timedependent fluorescence Stokes shift of a dissolved dipolar solute molecule.14 In bulk water, C(t) is dominated by a sub 50 fs component plus components of a few hundred femtoseconds.15 In contrast, measurements of C(t) in confined water in media as diverse as inverse micelles,16-20 lipid vesicles,21 cyclodextrins,22 proteins,23-26 and hydrogels10 reveal solvation dynamics on a very wide range of time scales, with components from * To whom correspondence should be addressed. E-mail: s.meech@ uea.ac.uk.
subpicosecond to tens or even hundreds of picoseconds being reported. Thus, C(t) measurements provide evidence of both inhomogeneity and a dramatic slowing down of molecular dynamics in nanoscale confined water. How this is reflected in the dynamics of chemical reactions in confined media is the topic addressed in this and related13 work. Perhaps the most widely studied example of nanoscale confined water is provided by water-in-oil or inverse micelles. Through control of the water to surfactant molar ratio (w0 ) [H2O]/[surfactant]) it is possible to control the radius of the dispersed water nanodroplet, while the choice of surfactant permits some control over the interface-water interaction. C(t)’s were measured in inverse micelles by time-resolved fluorescence on a range of time scales down to the subpicosecond by Levinger and co-workers19,20,27 and Bhattacharyya and co-workers.12,16,28,29 A characteristic slow multicomponent C(t) was reported and observed to be faster in larger water nanodroplets.20 Importantly, the C(t)’s were also found to be a strong function of the nature of the counterion in the head group of the ionic surfactant, suggesting a crucial role for the interface.27 Water nanodroplets in inverse micelles were also studied in time-resolved infrared experiments, which permitted a direct probe of the water vibrational and orientational dynamics in the OH stretch region.30-37 Many of those experiments could be analyzed on the basis of a core-shell model, in which slow dynamics were associated with the charged interfacial shell region, while the core of the dispersed nanodroplet was characterized by essentially bulk-like water dynamics.32,35,36 This model also points to a critical role for the interface in determining dynamics in the nanodroplet. Importantly, Piletic et al. found that orientational diffusion in an AOT stabilized nanodroplet was not well represented by the core-shell model, suggesting that coupling between core and shell regions influences the molecular dynamics.32 To further probe the role of interfacial charge, the same group compared orientational dynamics in an AOT stabilized inverse micelle with those in a nonionic surfactant with the same water nanodroplet radius.
10.1021/jp808991g CCC: $40.75 2009 American Chemical Society Published on Web 01/15/2009
Reactive Dynamics in Confined Liquids
Figure 1. Molecular structures of AO and IG.
They found that a charged interface was not critical to the observation of slow dynamics in water nanodroplets.31 A number of molecular dynamics simulations of water droplets stabilized by inverse micelles have been reported, usually based on models of the better characterized ionic surfactants.38-47 These all point to the importance of an effectively bound interfacial water layer exhibiting greatly reduced mobility. Simulations suggest that orientational dynamics in this region may be slower than that in bulk water by factors of hundreds. Senapati and Berkowitz characterized the water nanodroplet confined by an ionic surfactant into three regions, 0-5, 5-9, and beyond 9 Å from the charged interface.44 Each region was characterized by progressively faster dynamics for locations closer to the center of the water nanopool, and in the region beyond 9 Å, translational and rotational dynamics approached to within 20% of the bulk water value. Significantly, Senapati and Berkowitz also conducted similar simulations using an uncharged head group for the surfactant.47 They found that the orientational dynamics of water molecules in the first layer became faster in the absence of charges but remained very slow compared to that of bulk water. Dynamics in the second layer were similar for charged and uncharged systems and remained distinctly slower than that for bulk water. Faeder and Ladanyi simulated C(t) for a solvation probe dispersed in a water nanodroplet stabilized by an ionic negatively charged head group.48 Very different dynamics were found depending on the charge assigned to the solvation probe. A positively charged probe was bound in the interfacial region and showed very slow and w0-dependent solvation dynamics, while a negatively charged probe was located in the center of the water nanodroplet and had faster solvation dynamics which were independent of w0. Thus, both time-resolved IR and MD simulations point to the importance of a number of competing factors in determining dynamics in water nanodroplets: interactions between water molecules and the interface; interactions between water molecules and the charged head group; the size of the micelle; the charge and location of the probe; interaction of the probe with the water molecules. In this work, we investigate the role of interfacial charge on the dynamics of chemical reactivity in inverse micelles as a function of w0. As a reactive probe molecule, we have chosen AO (Figure 1) because its excited-state dynamics are relatively well characterized.49-53 In the excited electronic state, the phenyl rings undergo barrierless torsional motion about the bond to the central carbon atom. Movement along this coordinate leads to a nonfluorescent intermediate, which subsequently relaxes back to the ground state. Because of the large structural reorganization involved, the reaction is very sensitive to medium friction53 and has been studied in detail in bulk solvents.51,52 In a previous paper, we investigated the excited-state dynamics of AO in AOT stabilized water nanodroplets, compared them with bulk solvent data, and modeled them with a generalized
J. Phys. Chem. B, Vol. 113, No. 6, 2009 1633 Smoluchoski equation.13 It was found that the excited-state dynamics were dramatically slowed in the dispersed aqueous medium. The dynamics were a function of w0 but, even in the largest micelle, did not approach the very fast reaction found in bulk aqueous solution. In this paper, the same AO probe molecule is studied in water nanodroplets, in this case stabilized by a polar nonionic surfactant, Igepal (IG, Figure 1). By contrasting these data with those obtained earlier for the AOT stabilized micelle,13 we interrogate the influence of the surfactant-water nanodroplet interface on reactions in the aqueous nanodroplets. The most striking difference between the two surfactants is the absence of a head group charge in IG, but in addition, the IG head group is more extended than that in AOT, comprising a 5 unit oxyethylene chain (Figure 1). The extended IG head group may be expected to influence the structure of the dispersed nanodroplet, just as the acyl chains influence the dynamics of the oil phase in oil-in-water micelles.54 However, a number of experiments suggest that the major portion of the dispersed water exists as a nanodroplet, rather than being sequestered by the oxyethylene chain.55,56 2. Experimental Section To accurately follow the time course of the AO reaction requires subpicosecond time resolution. The procedure used here is based around an up-conversion spectrometer which records time-resolved fluorescence with sub 50 fs time resolution; the method was described in our preceding paper.13 The timedependent fluorescence spectra were constructed from fluorescence profiles measured at a range of wavelengths across the emission spectrum. The resulting spectra were fit to a log-normal function from which the experimental parameters were extracted for comparison with theoretical calculations. AO (dye content > 80%), IG (Igepal CO 520, with an average molecular weight of Mn ∼ 441), cyclohexane (chromasolv, plus, for HPLC g99.9%), and n-hexane (chromasolv) were purchased from Sigma-Aldrich. Polyethylene glycol 200 (PEG5, mol wt 190-220) was purchased from Fluka. These chemicals were used without further purification. An IG stock solution with a concentration of 0.8 mol dm-3 was prepared by dissolving IG into a 50:50 wt % mixture of cycolhexane/n-hexane. The micelle samples were prepared using a procedure similar to that for AOT micelle samples.13 The IG inverse micelles have been characterized by Lipgens et al.55 The radius of the central water droplet can be calculated through the following expression55
rw ) 0.19w0 + 0.70 where
w0 ) [H2O]/[IG] In this expression, rw represents the water core radius which forms once the shell solvation has been completed by filling the oxyethylene chain region with water. The AO concentration was calculated to be less than one molecule per micelle, and the OD was in the range of 0.1-0.2 at an excitation wavelength around 415 nm in a 1 mm cuvette. Viscosities were measured in a shortened form BS/IP/SL(S) suspended level viscometer (size No. 3-6). The sample viscosity was calculated through the kinematic viscosity expression, cS ) Ct, where C is a calibration constant and t represents the measured flow time in seconds. The calibration constants were C ) 0.01019cS/s for size no. 3; 0.03327cS/s for size no. 4; 0.1037sC/s for size no. 5; and 03256cS/s for size no. 6.
1634 J. Phys. Chem. B, Vol. 113, No. 6, 2009
Figure 2. Absorption spectra of (a) 5 × 10-5 mol dm-3 AO in IG stabilized aqueous nanodroplets as a function of radius and (b) 3 × 10-6 mol dm-3 AO in pentanol for different degrees of acidification.
Kondo et al.
3. Results and Discussion
Figure 3. (a) Fluorescence decay data for AO in IG stabilized aqueous nanodroplets, compared with the data for AOT surfactant.13 (b) Plot of the mean lifetime determined from three exponential fits to the data, compared with the steady-state fluorescence yield.
Steady-State Spectroscopy. The absorption spectra for AO in IG reversed micelles are shown in Figure 2 as a function of w0 and compared with the spectrum in bulk water. For water nanodroplets of calculated radius greater than rw ) 2.5 nm (w0 > 9.5), the AO spectrum is independent of radius and very similar to that measured in bulk water. This contrasts with measurements on AOT stabilized nanodroplets,13 where the AO spectrum approaches that in bulk water only in the largest micelles (rw > 10). The most significant size effect on the absorption spectrum is seen for the smallest micelle studied (rw ) 2 nm) where the intensity in the UV region increases at the expense of the main 430 nm peak. This can be ascribed to an equilibrium between the charged protonated and the deprotonated forms of AO. This is demonstrated in Figure 2b, where the AO spectrum is shown in pentanol and its acidic and basic forms. The deprotonated form of AO exists as a minor component in pentanol but can be completely populated by addition of base or driven completely to the charged form upon acidification. A w0 dependence of the position of an acid/base equilibrium was reported previously for the dye acridine orange in aqueous nanodroplets stabilized by TritonX-100 surfactant (which has a nine unit oxyethylene head group).57 The result suggests that for the smallest micelles, the oxyethylene head group is incompletely hydrated and can thus act as a proton acceptor to generate the neutral AO. As seen in Figure 2, the charged form of AO dominates for all larger micelles, and the neutral from will not interfere with time-resolved measurements, as it is not significantly excited at the 415 nm excitation wavelength used. Ultrafast Time-Resolved Fluorescence. Time-resolved AO fluorescence decay data were measured as a function of rw. The fluorescence decay is a function of wavelength, but in Figure
3, the data measured at the peak of the fluorescence intensity are compared. Along with the data in the IG surfactant, we also show data for AOT13 and for bulk water. It is evident that the AO decay is dramatically slowed down in the IG micelle. This is certainly true with respect to bulk water, where the mean decay time for the rw ) 2 micelle has increased around 35 times. Significantly, the AO decay times in IG stabilized inverse micelles are also slower than those in AOT for all nanodroplet radii studied. Thus, the slowing down of the excited-state reaction reported in AOT is not simply a function of location of the probe molecule in the highly constrained ionic head group region of AOT. The observation is consistent with the OH vibrational and orientational relaxation data of Moilanen et al., who found that the slow dynamics observed in AOT stabilized nanodroplets persisted in IG micelles.31 Also shown in Figure 3 are the average decay times (at the peak fluorescence intensity) and the steady-state fluorescence yield as a function of rw. The AO fluorescence is a weak function of rw and does not change significantly between rw ) 3 and 4 nm and increases by 40% between rw ) 3 and 2 nm, the smallest micelle studied. This contrasts with the behavior seen in AOT, where the AO fluorescence yield was a function of rw between 1 and 10 nm.13 This suggests that the AO environment does not change significantly as a function of rw in the small IG stabilized micelles studied here To check that the AO is indeed probing the interior of the IG reverse micelle, we made a measurement on a sample with an identical w0 but a higher mole fraction of water. In this case, the macroscopic solution becomes viscous. The fluorescence decay was measured to be the same for both samples, showing that the AO indeed probes the dispersed water phase and is not
Reactive Dynamics in Confined Liquids
J. Phys. Chem. B, Vol. 113, No. 6, 2009 1635
Figure 4. Time-dependent emission spectra for AO in IG stabilized nanodroplets (rw ) 2). The data (points) are fit to the log-normal function (lines).
sensitive to the macroscopic sample viscosity. Thus, the main experimental results are that the excited-state reaction leading to decay of AO is dramatically slowed in aqueous nanodroplets stabilized by IG compared with that in bulk water and is even somewhat slower than that in AOT stabilized micelles of similar radius. However, the fluorescence in IG micelles is only a weak function of nanodroplet size and certainly does not tend to the bulk aqueous lifetime in the rw range investigated here. The absence of a large rw dependence argues for a dominant interfacial location of the dye rather than a preference for the aqueous phase, even though the AO itself is charged. However, the absorption spectra described above suggest that the AO is located on the aqueous side of the interface. Analysis of Time-Dependent Emission Spectra. The fluorescence decay data for AO in IG reverse micelles are nonsingle exponential and a strong function of the emission wavelength. Thus, the steady-state intensity and mean decay time at the peak emission wavelength are only suitable for a qualitative analysis. For a more quantitative treatment, the time dependence of the entire emission spectrum must be measured and analyzed. Fluorescence decay profiles were collected at a number of wavelengths across the AO emission spectrum and normalized to the steady-state emission spectrum to create a three-dimensional time-resolved fluorescence spectral surface. The resulting time-dependent emission spectra are shown in Figure 4 fit to a log-normal function. In Figure 5, the parameters extracted from the log-normal fit are displayed and compared with those obtained for an AOT stabilized inverse micelle.13 The qualitative results discussed above are reflected in the quantitative analysis. The integrated fluorescence intensity for IG with rw ) 2 nm decays only slightly more slowly than for rw ) 4 nm, and for both, the decay is slower than that for AO in the smallest (rw ) 1 nm) AOT stabilized nanodroplet (Figure 5a). The rate of spectral shift is also slower for the IG micelles compared to that for AOT (Figure 5b). Significantly, the total magnitude of the shift is larger for AOT. Finally, the behavior of the time-dependent spectral bandwidth (Figure 5c) is different for IG and AOT micelles, with the latter broadening during the spectral shift, while for IG, the width remains approximately constant. The excited-state decay of AO can be modeled in terms of diffusive phenyl rotational motion on a barrierless excited-state potential energy surface leading to a curve crossing with a nonemissive state of charge-transfer character.49-52,58 The diffusive excited-state evolution can be calculated from the generalized Smoluchowski equation. This approach has been applied successfully to AO in alcohol solution51 and, in our earlier paper,13 to AO in AOT stabilized micelles. The initial population on
Figure 5. Parameters of AO in IG stabilized nanodroplets with rw ) 2 (square) and 4 nm (triangle) and AO in AOT stabilized nanodroplets with rw ) 1 nm (diamond), extracted from log-normal fits to the timeresolved emission spectra. (a) Integrated fluorescence intensity, (b) mean frequency, and (c) full width at half-maximum height.
the ground-state surface is transferred in a Franck-Condon transition to an unstable position on the excited-state surface. The subsequent evolution of the excited-state population, F(z,t), along the reaction coordinate (z, assumed to be the phenyl torsional coordinate) is given by the generalized Smoluchowski equation59,60
(
)
∂ ∂ ∂ 1 ∂ F(z, t) ) D(t) + S (z) F(z, t) - κΓ(z)F(z, t) ∂t ∂z ∂z kBT ∂z r (1) from which F(z,t) can be determined for any potential surface, Sr(z), and diffusion coefficient, D(t), which may, in general, be time-dependent.61-63 In addition to the excited-state evolution, the final term in eq 1 allows for decay from the excited state to the ground state with a rate coefficient κ, which may itself be a function of z (we employ the “Gaussian sink model”,59 and therefore, Γ(z) is a Gaussian function).
1636 J. Phys. Chem. B, Vol. 113, No. 6, 2009
Kondo et al.
Figure 6. (a) Fit of the time-resolved spectra to the Smoluchowski model for the rw ) 2 AO in IG micelle. (b) The fit to the mean frequency (first moment) and (c) spectral width as a function of time.
The reactive potential energy surface, S(z), was constructed from coupling two harmonic potential surfaces associated with the directly excited emissive state and the dark intermediate state.50,64,65 The shape parameters and coupling strength were as used previously for AO in AOT micelles,13 but the absolute energies were adjusted to fit the steady-state and time-resolved spectra. The emission spectra were then calculated according to64-66
Ifl ∝
∫ dz g(ν0(z), ν(z) - ν0(z))M2(z)F(z, t)ν3
(2)
in which M(z) is a coordinate-dependent transition moment, decreasing as a function of z as the population evolves from the optically allowed transition state to the dark state. The g(ν0(z), ν(z) - ν0(z)), is a line shape function (a log-normal function was used here), and ν0(z) is the torsional-angledependent energy gap between the excited and ground states. Accurately fitting the time-dependent emission spectra (Figure 6) required a time-dependent diffusion coefficient, as previously discussed for AOT.13 A time dependence in the diffusion coefficient is not unexpected in the case of ultrafast reactions, where the high-frequency part of the friction may contribute most to motion along the potential surface.62 The time-dependent diffusion coefficient is defined by67,68
D(t) ) -〈(δz)2〉
˙ (t) ∆ ∆(t)
(4)
in which ∆(t) is the normalized reaction coordinate time correlation function ∆(t) ) 〈δz · δz(t)〉/〈(δz)2〉, where δz ) z zeq and 〈(δz)2〉 is the mean-square fluctuation of the reaction coordinate, which, for the potential surface used here, we obtain 〈(δz)2〉 ) 0.059. It has been proposed that the evolution from the bright to dark state in AO corresponds to a transition to an internal charge-transfer state. In such cases, to a first approximation, ∆(t) can be taken from the solvation time correlation function C(t).64,65 In the absence of a measured C(t) for IG
stabilized micelles, we use as initial values for ∆(t) those extracted for AO in AOT, which were themselves based on C(t) measurements.13 These values were adjusted to optimize the fit to the data. The final parameters required to achieve a best fit to all three experimental data sets (spectral width, spectral first moment and integrated intensity) are shown in Table 1, which also contains the value for the fitted decay coefficient, κ. The quality of the best fit is illustrated in Figure 6 for the rw ) 2 micelle. The one-dimensional Smoluchowski model provides an accurate description of the integrated intensity and timedependent spectral shift for both 2 and 4 nm radius IG micelles. However, the model always predicts a small time-dependent increase in the spectral width, which is not observed in the experimental data. The failure to observe the expected broadening may arise simply from the fact that the spectral width is the parameter least well defined in the analysis due to the range over which the data was collected (Figure 5a). Origin Of The Increased Friction In Micelles. The fit parameters (Table 1) show that the reaction coordinate experiences rather similar frequency-dependent friction in AOT and IG stabilized micelles, and in both cases, the friction is much greater than that in bulk aqueous solution. This is entirely consistent with the qualitative result that reactions in dispersed water are dramatically slowed compared with those in bulk water. However, suppression of the AO torsional dynamics in the excited-state is not the only effect operating. The analysis also suggests that the decay rate from the dark state back to the ground state (κ) is a strong function of the environment. The decay rate is strongly suppressed upon going from water to AOT stabilized nanodroplets.13 There is a further suppression upon going from ionic AOT to nonionic IG surfactants, which may suggest a role for the ionic medium in promoting the transition from the dark state to the ground state. First, we address the possible nature of the medium probed by the excited-state reaction of AO in the micelle and the origin of the friction experienced by the reaction coordinate. One explanation is simply that confinement of a spherical water droplet by a surfactant capable of forming strong H-bonds is itself sufficient to provide the high friction observed. However, the IG-water interface is complex; the polar head group comprises a five monomer oxyethylene chain, which may itself trap the ionic probe molecule. To investigate this point further, we contrast the observations for AO in IG stabilized aqueous nanodroplets with measurements for AO in mixtures of water and PEG5, characterized by the ratio w ) [H2O]/[PEG5]. The mean fluorescence decay time at the wavelength of maximum intensity is shown in Figure 7 as a function of w and w0. At all values where the ratios overlap, the mean AO lifetime in solution is much less than that in the IG stabilized nanodroplet. However, the AO lifetime is a strong function of w, particularly in PEG5-rich regions. At w < 3, the mean AO lifetime increases considerably. This reflects both increasing medium viscosity and the suppression of the quenching mechanism, which is specific to AO in an aqueous environment (see also Figure 3 and measurements in pure water discussed in more detail in the accompanying ref 13). On the basis of Figure 7 alone, a plausible explanation for the long relaxation time in IG micelles could be that AO is somehow preferentially located in the head group, in a region which is not fully solvated by water molecules. However, this seems improbable; there is ample evidence that the surfactant head groups are solvated at relatively low w0 (e.g., w0 ) 8 for the TritonX-100 surfactant, which has an oxyethylene head group twice as long as IG) and that additional water exists as
Reactive Dynamics in Confined Liquids
J. Phys. Chem. B, Vol. 113, No. 6, 2009 1637
TABLE 1: Parameters Used Fitting the Time Dependent Spectra of Micelles and Bulk Water to the Generalized Smoluchowski Equation D(t)a/ps-1
a
medium
Se /cm-1
Sd /cm-1
a1
a2
τD1 /ps
τD2 /ps
〈τ〉/ps
κb/ps-1
IG rw ) 2 IG rw ) 4 AOT rw ) 1 AOT rw ) 5 water
22280
21430
0.83
0.17
0.33
3.2
0.82
0.04
22230
21380
0.88
0.12
0.23
1.9
0.43
0.05
22640
21790
0.83
0.17
0.32
1.60
0.54
0.14
22230
21380
0.94
0.06
0.33
1.20
0.39
0.26
22120
21270
nm nm nm nm
D ) 0.75
1.80
Fitting parameters with bi-exponential D(t), D(t) ) a1 exp(-τD1 /t) + a2 exp(-τD2 /t). b The sink parameters were z 0 ) 0.8 and σ ) 0.2.
a nanoscale droplet. However, to investigate this possibility further, the absorption spectra of AO in the mixed solvent and the micelle are compared in Figure 8. The behavior as a function of w is somewhat unexpected, in that as the water content increased, the AO spectrum initially red shifted at w < 5 and then underwent a progressive blue shift. The blue shift with increasing w can reasonably be ascribed to increasing water-AO H-bonding interactions stabilizing the ground electronic state. The blue shift at low w (which correlates with the longer fluorescence decay time) is less readily explained but may reflect a solvation of the AO ion by the PEG5. Polyethylene glycols have an established role in polymer electrolyte solutions.69 As the water content increases, the polymer AO solvation will be increasingly replaced by aqueous solvation. The key feature of Figure 8 is that the spectral maxima of the IG samples all match closely that of the neat aqueous
Figure 7. (a) Mean decay time for AO in IG stabilized nanodroplets (blue) and as a function of composition in the water/PEG5 mixed solvent (magenta). (b) The relative fluorescence yield for AO water/ PEG5 (magenta) and an ethanol/PEG5 (blue) mixtures, plotted as a function of the measured viscosity.
solution. This argues strongly in favor of localization of AO in the aqueous phase rather than in a region of the head group unoccupied by water. This in turn confirms that the very high
Figure 8. (a) Absorption spectra of AO in the water/PEG5 mixed solvent as a function of w. (b) Peak wavelength as a function of w. The value for bulk water is shown as a dashed line. (c) Spectral width of the main absorption band.
1638 J. Phys. Chem. B, Vol. 113, No. 6, 2009 friction experienced by the reactive coordinate arises within nanodispersed water rather than through incorporation of AO within the IG head group region. This observation is in agreement with the results of transient IR spectroscopy, which showed that dispersed water nanodroplets of similar size stabilized by ionic (AOT) and nonionic (IG) surfactants have similar dynamical properties.31 The IR data show a slowing of the dynamics compared to that of bulk water by factors between 2 and 10, depending on the property being measured.31,35 The mean lifetime for AO in an IG stabilized micelle is ∼6 ps, about 30 times larger than that in bulk water. Such a large effect on the excitedstate reaction may arise from a combination of two causes. First, the friction experienced by the reaction coordinate (a phenyl twist) is likely to involve many more water molecules than are probed by the IR measurement; therefore, a cumulative effect may be expected to exaggerate the effect of confinement on the dispersed water dynamics. Second, the excited-state decay of AO appears to be accelerated in bulk water due to a specific solvent-solute interaction (Table 1). Possibly, this specific interaction does not take place in the dispersed phase. Some evidence for this can be seen in the width of the absorption spectra. Water has a broader spectrum than either the micelle or the PEG5/water mixtures. In this sense, the broad spectrum correlates with the short lifetime of AO in bulk water. 4. Conclusions The dynamics of the AO torsional excited-state reaction have been measured in nanoscale water droplets stabilized by a nonionic surfactant. The results have been quantitatively analyzed in terms of a one-dimensional modified Smoluchowski model. The results have been contrasted with the behavior in droplets stabilized by an ionic surfactant (AOT).13 The first conclusion is that the rate of reaction is decreased dramatically in the dispersed phase and equally in ionic and nonionic surfactants; the effect is not simply one of interfacial charge. The origin of the friction in the IG stabilized nanodroplets was investigated by comparing the data with measurements of AO in PEG5/water, where the PEG is a model for the IG head group. The data suggest that the AO is located on the aqueous side of the interface. A similar conclusion was reached by Dutt for charged solutes.70 Thus, it is concluded that reactive dynamics in the nanoscale dispersed water droplet are slowed due to confinement of the water by the surfactant. Acknowledgment. We are grateful to EPSRC for financial support of this project. M.K. thanks MEXT for a studentship. J.C. and J.R. were supported by summer studentship bursaries from the Nuffield Foundation. References and Notes (1) Persson, E.; Halle, B. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 6266. (2) Kim, J.; Lu, W. Y.; Qiu, W. H.; Wang, L. J.; Caffrey, M.; Zhong, D. P. J. Phys. Chem. B 2006, 110, 21994. (3) Zhang, L. Y.; Wang, L. J.; Kao, Y. T.; Qiu, W. H.; Yang, Y.; Okobiah, O.; Zhong, D. P. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 18461. (4) Daidone, I.; Ulmschneider, M. B.; Di Nola, A.; Amadei, A.; Smith, J. C. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 15230. (5) Schroder, C.; Rudas, T.; Boresch, S.; Steinhauser, O. J. Chem. Phys. 2006, 124. (6) Rudas, T.; Schroder, C.; Boresch, S.; Steinhauser, O. J. Chem. Phys. 2006, 124. (7) Russo, D.; Murarka, R. K.; Copley, J. R. D.; Head-Gordon, T. J. Phys. Chem. B 2005, 109, 12966.
Kondo et al. (8) Bandyopadhyay, S.; Chakraborty, S.; Balasubramanian, S.; Bagchi, B. J. Am. Chem. Soc. 2005, 127, 4071. (9) Grant, C. D.; DeRitter, M. R.; Steege, K. E.; Fadeeva, T. A.; Castner, E. W. Langmuir 2005, 21, 1745. (10) Datta, A.; Das, S.; Mandal, D.; Pal, S. K.; Bhattacharyya, K. Langmuir 1997, 13, 6922. (11) Rasaiah, J. C.; Garde, S.; Hummer, G. Annu. ReV. Phys. Chem. 2008, 59, 713. (12) Bhattacharyya, K.; Bagchi, B. J. Phys. Chem. A 2000, 104, 10603. (13) Heisler, I. A.; Kondo, M.; Meech, S. R. Preceding article in this Journal. (14) Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. J. Phys. Chem. 1995, 99, 17311. (15) Jimenez, R.; Fleming, G. R.; Kumar, P. V.; Maroncelli, M. Nature 1994, 369, 471. (16) Satoh, T.; Okuno, H.; Tominaga, K.; Bhattacharyya, K. Chem. Lett. 2004, 33, 1090. (17) Corbeil, E. M.; Riter, R. E.; Levinger, N. E. J. Phys. Chem. B 2004, 108, 10777. (18) Elles, C. G.; Levinger, N. E. Chem. Phys. Lett. 2000, 317, 624. (19) Riter, R. E.; Undiks, E. P.; Levinger, N. E. J. Am. Chem. Soc. 1998, 120, 6062. (20) Riter, R. E.; Willard, D. M.; Levinger, N. E. J. Phys. Chem. B 1998, 102, 2705. (21) Sen, P.; Satoh, T.; Bhattacharyya, K.; Tominaga, K. Chem. Phys. Lett. 2005, 411, 339. (22) Sen, P.; Roy, D.; Mondal, S. K.; Sahu, K.; Ghosh, S.; Bhattacharyya, K. J. Phys. Chem. A 2005, 109, 9716. (23) Qiu, W. H.; Kao, Y. T.; Zhang, L. Y.; Yang, Y.; Wang, L. J.; Stites, W. E.; Zhong, D. P.; Zewail, A. H. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 13979. (24) Qiu, W. H.; Zhang, L. Y.; Okobiah, O.; Yang, Y.; Wang, L. J.; Zhong, D. P.; Zewail, A. H. J. Phys. Chem. B 2006, 110, 10540. (25) Pal, S. K.; Zewail, A. H. Chem. ReV. 2004, 104, 2099. (26) Bhattacharyya, S. M.; Wang, Z. G.; Zewail, A. H. J. Phys. Chem. B 2003, 107, 13218. (27) Pant, D.; Riter, R. E.; Levinger, N. E. J. Chem. Phys. 1998, 109, 9995. (28) Dutta, P.; Bhattacharyya, K. J. Chem. Sci. 2004, 116, 5. (29) Bhattacharyya, K. Acc. Chem. Res. 2003, 36, 95. (30) Park, S.; Moilanen, D. E.; Fayer, M. D. J. Phys. Chem. B 2008, 112, 5279. (31) Moilanen, D. E.; Levinger, N. E.; Spry, D. B.; Fayer, M. D. J. Am. Chem. Soc. 2007, 129, 14311. (32) Piletic, I. R.; Moilanen, D. E.; Spry, D. B.; Levinger, N. E.; Fayer, M. D. J. Phys. Chem. A 2006, 110, 4985. (33) Tan, H. S.; Piletic, I. R.; Fayer, M. D. J. Chem. Phys. 2005, 122. (34) Tan, H. S.; Piletic, I. R.; Riter, R. E.; Levinger, N. E.; Fayer, M. D. Phys. ReV. Lett. 2005, 94. (35) Dokter, A. M.; Woutersen, S.; Bakker, H. J. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15355. (36) Cringus, D.; Bakulin, A.; Lindner, J.; Vohringer, P.; Pshenichnikov, M. S.; Wiersma, D. A. J. Phys. Chem. B 2007, 111, 14193. (37) Cringus, D.; Lindner, J.; Milder, M. T. W.; Pshenichnikov, M. S.; Vohringer, P.; Wiersma, D. A. Chem. Phys. Lett. 2005, 408, 162. (38) Harpham, M. R.; Ladanyi, B. M.; Levinger, N. E.; Herwig, K. W. J. Chem. Phys. 2004, 121, 7855. (39) Harpham, M. R.; Ladanyi, B. M.; Levinger, N. E. J. Phys. Chem. B 2005, 109, 16891. (40) Ladanyi, B. M.; Faeder, J.; Martins, L. R.; Skaf, M. S. Abstr. Pap. Am. Chem. Soc. 2004, 228, U219. (41) Faeder, J.; Albert, M. V.; Ladanyi, B. M. Langmuir 2003, 19, 2514. (42) Faeder, J.; Ladanyi, B. M. J. Phys. Chem. B 2001, 105, 11148. (43) Faeder, J.; Ladanyi, B. M. J. Phys. Chem. B 2000, 104, 1033. (44) Senapati, S.; Berkowitz, M. L. J. Phys. Chem. A 2004, 108, 9768. (45) Lu, L. Y.; Berkowitz, M. L. J. Am. Chem. Soc. 2004, 126, 10254. (46) Senapati, S.; Berkowitz, M. L. J. Phys. Chem. B 2003, 107, 12906. (47) Senapati, S.; Berkowitz, M. L. J. Chem. Phys. 2003, 118, 1937. (48) Faeder, J.; Ladanyi, B. M. J. Phys. Chem. B 2005, 109, 6732. (49) Changenet, P.; Zhang, H.; van der Meer, M. J.; Glasbeek, M.; Plaza, P.; Martin, M. M. J. Fluoresc. 2000, 10, 155. (50) Martin, M. M.; Plaza, P.; Changenet, P.; Meyer, Y. H. J. Photochem. Photobiol., A 1997, 105, 197. (51) van der Meer, M. J.; Zhang, H.; Glasbeek, M. J. Chem. Phys. 2000, 112, 2878. (52) Changenet, P.; Zhang, H.; van der Meer, M. J.; Glasbeek, M.; Plaza, P.; Martin, M. M. J. Phys. Chem. A 1998, 102, 6716. (53) Oster, G.; Nishijima, Y. J. Am. Chem. Soc. 1956, 78, 1581. (54) Hunt, N. T.; Jaye, A. A.; Meech, S. R. J. Phys. Chem. B 2003, 107, 3405. (55) Lipgens, S.; Schubel, D.; Schlicht, L.; Spilgies, J. H.; Ilgenfritz, G.; Eastoe, J.; Heenan, R. K. Langmuir 1998, 14, 1041.
Reactive Dynamics in Confined Liquids (56) Caldararu, H.; Caragheorgheopol, A.; Vasilescu, M.; Dragutan, I.; Lemmetyinen, H. J. Phys. Chem. 1994, 98, 5320. (57) Andrade, S. M.; Costa, S. M. B. Photochem. Photobiol. Sci. 2002, 1, 500. (58) Glasbeek, M.; Zhang, H.; van der Meer, M. J. J. Mol. Liq. 2000, 86, 123. (59) Bagchi, B.; Fleming, G. R.; Oxtoby, D. W. J. Chem. Phys. 1983, 78, 7375. (60) Denny, R. A.; Bagchi, B. J. Phys. Chem. A 1999, 103, 9061. (61) Grote, R. F.; Vanderzwan, G.; Hynes, J. T. J. Phys. Chem. 1984, 88, 4676. (62) Hynes, J. T. J. Phys. Chem. 1986, 90, 3701. (63) Vanderzwan, G.; Hynes, J. T. J. Chem. Phys. 1983, 78, 4174.
J. Phys. Chem. B, Vol. 113, No. 6, 2009 1639 (64) Tominaga, K.; Walker, G. C.; Jarzeba, W.; Barbara, P. F. J. Phys. Chem. 1991, 95, 10475. (65) Tominaga, K.; Walker, G. C.; Kang, T. J.; Barbara, P. F.; Fonseca, T. J. Phys. Chem. 1991, 95, 10485. (66) Tominaga, K.; Kliner, D. A. V.; Johnson, A. E.; Levinger, N. E.; Barbara, P. F. J. Chem. Phys. 1993, 98, 1228. (67) Okuyama, S.; Oxtoby, D. W. J. Chem. Phys. 1986, 84, 5824. (68) Okuyama, S.; Oxtoby, D. W. J. Chem. Phys. 1986, 84, 5830. (69) Olender, R.; Nitzan, A. J. Chem. Phys. 1995, 102, 7180. (70) Dutt, G. B. J. Chem. Phys. 2008, 129 .
JP808991G
