J. Phys. Chem. B 2006, 110, 24231-24237
24231
Probing the Behavior of Confined Water by Proton-Transfer Reactions G. Angulo, J. A. Organero, M. A. Carranza, and A. Douhal* Departamento de Quı´mica Fı´sica, Seccio´ n de Quı´micas, Facultad de Ciencias del Medio Ambiente, UniVersidad de CastillasLa Mancha, AVenida Carlos III, S.N., 45071 Toledo, Spain ReceiVed: July 6, 2006; In Final Form: August 30, 2006
The picosecond dynamics of a bifunctional and H-bonding molecule, 7-hydroxyquinoline (7HQ), has been studied in a reverse micelle with increasing water content. The fluorescence kinetics has a complex behavior as the water content is changed. All reactions are irreversible, and a two-step mechanism is invoked to explain the observations. H2O/D2O exchange and excitation energy effects show that the second step has a higher barrier and that the corresponding reaction occurs through tunneling. The results clearly indicate two regimes of water nanopool behavior switching at W0 ≈ 5 (W0 ) [water]/[surfactant]). Water collective dynamics explains these observations. The lower fluidity of confined water within the reverse micelle with respect to normal bulk water alters the related H-bond network dynamics and therefore is responsible for the slower proton-transfer processes.
1. Introduction It is of current interest to elucidate the role of water in biological functions with molecular and ultrafast resolution. Small and fast structural changes in the water network are decisive events in proton or H-atom transfer.1 However, biological water is very different from the bulk liquid. Furthermore, biological samples are most of the time too complex for research of the underlying fundamental processes. Therefore, vesicles and micelles have been proposed for mimicking nanostructured biological environments, such as protein pockets and cell membranes.2-4 Fluorescence probes sensitive to Hbonding and proton-transfer reactions are very often used as tools for exploring these media.5-7 Available time-resolved technology allows gaining insight into the very fast processes suffered by the probe, up to a femtosecond (fs) time scale, after initiation of the photoevents. The study of organic dyes having functional groups and structures comparable to DNA bases or amino acids in biological-like water is obviously relevant to the efforts for a better understanding of the dynamics of biological systems. In this work, we present studies of 7-hydroxyquinoline (7HQ)8-11 embedded in a nanopool of water inside reverse micelles formed by sodium bis(2-ethylhexyl) sulfosuccinate (AOT) (Scheme 1). 7HQ is a bifunctional probe with groups comparable to those in many biological molecules, including DNA base units. Adding water produces monodisperse microemulsions of water nanopools inside the reverse AOT micelles, and its content can be varied up to high relative water concentrations without breaking or reshaping them. 7HQ has been the subject of many studies, and its physical properties are well established.12-15 Photophysical studies of many compounds within micelles have been reported, revealing a complex water structure and pool-size-dependent fluorescence characteristics.16,17 Two water regions can be distinguished: a first layer solvating the negatively charged surfactant headgroups, and in which the so-called “bound” water is found, and a core * Corresponding author.
[email protected].
Fax:
+34-925-268840.
E-mail:
of “free” water. The former is much like crystallization water, while the latter does not behave like bulk water but has a higher viscosity, lower dielectric constant, and smaller activity, among other measured properties.12-15 This behavior is rather similar to that of water around proteins, as already stated.1 Therefore, the aim of this work is to show how a bifunctional compound like 7HQ probes the behavior of such an environment. While preparing this paper, we became aware of the recently published work on the same system.18 We report and discuss our results, and show that differences in the experimental conditions, such as excitation at selected wavelengths of the 7HQ-water absorption band, give another picture of the picosecond dynamics of this probe in the nanopool. We observed another prototropic species of the probe, and that the consequence of monitoring its photophysics in the nanopool is relevant for the discernment of water behavior in confined media. This allows us to distinguish two water regimes inside the micelle below W0 ≈ 5 (W0 ) [H2O]/[AOT]), where kinetics depend strongly on the amount of water, and above W0 ≈ 5, where the dependence almost disappears. 2. Materials and Methods 2.1. Materials. 7-Hydroxyquinoline (7HQ) (Acros, 99%), heptane (Acros, Spectrograde), deuterium oxide (Aldrich, 99.9%), and sodium bis(2-ethylhexyl) sulfosuccinate (AOT) (Fluka, Ultra >99.0%) were used as received. The Karl Fischer test shows that the percentage of water in the heptane micelle solution is about 0.15% w/v. Therefore, without adding water the reverse micelle already has a W0 ≈ 0.6. Milli-Q ultrapure water was used in all experiments. Reverse micelles were prepared by dissolving 0.2 M AOT in heptane and sonicating for about 30 min. This AOT concentration is well above the operational critical micelle concentration (cmc),15 and below the maximum for the monodispersivity criterion up to W0 ) 20.19 The concentration of 7HQ was less than 5 × 10-4 M in order to guarantee less than one probe molecule per micelle. Water content was changed by adding it to the stock solution of 7HQ/AOT in appropriate amounts.
10.1021/jp064257g CCC: $33.50 © 2006 American Chemical Society Published on Web 11/04/2006
24232 J. Phys. Chem. B, Vol. 110, No. 47, 2006
Angulo et al.
SCHEME 1: Molecular Structures of 7HQ and AOT and Schematic Representation of 7HQ/H2O in AOT/n-heptane Micellea
a For simplicity we located the probe in the center of the nanopool. The arrows indicate the proton-transfer reactions between the probe and confined water molecules.
Figure 1. Absorption spectra of 7HQ in AOT/n-heptane reverse micelles with increasing W0. Inset: change of absorbance at 410 nm against W0, fitted by 1:2 complex formation model.
2.2. Methods. UV-visible absorption and emission spectra were recorded on Varian (Cary E1) and Perkin-Elmer (LS-50B) spectrophotometers, respectively. Picosecond (ps) time-resolved emissions at magic-angle measurements were done by using a time-correlated single-photon-counting spectrophotometer (FluoTime 200) exciting at 371 or 433 nm (instrument response function (IRF) 65 ps operated at 10 MHz).20 The decays were fitted by multiexponential functions convoluted with the IRF using the Fluofit package. Decay components down to 10 ps can be resolved after convolution, and this has been checked using short-lived (less than 10 ps decay time component) excited species and confirmed using femtosecond observation from this laboratory. The quality of the fits was characterized in terms of residual distribution, its autocorrelation, and the reduced χ2 value. Time-resolved emission spectra (TRES) have been recorded from the decays measured at the different wavelengths of emission (32 decays) with a spectral interval of typically 6 nm. The spectra have been then constructed using the Fluofit package. The zero time has been defined at the channel corresponding to half of the maximum of the IRF at its rising part. The time-resolved area-normalized emission spectra (TRANES) have been constructed by area normalization of the TRES. All experiments were done at 293 ( 1 K. 3. Results and Discussion 3.1. Steady-State Observation. Figure 1 shows the absorption spectra of 7HQ in the reverse micelles upon increasing
Figure 2. Fluorescence emission spectra of 7HQ in AOT/n-heptane reverse micelles with increasing W0. Excitation wavelength: 371 nm. The spectra are corrected for the wavelength dependence of the detection and normalized for the absorption at the excitation wavelength. Inset: intensities at 430 and 600 nm divided by the respective intensities for W0 ) 0.6, as a function of W0. The arrows indicate the trend of both bands as W0 increases from 0.6 to 20.6.
water concentration. In the absence of water, the spectrum is of the enol form (E).21 The increase of water content (W0 ) 0.6 f 20.6) makes evident the appearance of the keto (K) or zwitterionic (Z) form (hereafter called tautomer, T, form) absorption at about 410 nm. W0 ) 0.6 corresponds to the sample without addition of water. At W0 ) 20.6, the absorption spectrum is still far from showing as much T absorption as in bulk water.21,22 The new absorption band (410 nm) increases following a 1:2 stoichiometry complex formation, with an apparent equilibrium constant of 2 M-2 (inset in Figure 1). This process should involve the complexation of E with two water molecules to give an anion (A) which is converted to T. The emission spectra of these samples upon excitation at 371 nm are shown in Figure 2. In the absence of added water the emission band (maximum intensity at 430 nm) does not correspond to any of the bands seen in pure water.21 It is wellknown that the E form emits at 380 nm, while the cation (C) and A emit at 450 and 490 nm, respectively.21 For W0 < 5 or in the absence of water within the AOT micelle, the dielectric constant of the pool is significantly lower than that of water,5 and C cannot be produced with an anionic pool. This emission corresponds to an E form of 7HQ strongly bound (E‚‚‚AOT) to the headgroups of AOT, with a strong anionic character.
Behavior of Confined Water
J. Phys. Chem. B, Vol. 110, No. 47, 2006 24233
TABLE 1: Values of Fluorescence Time Constants (τi) and Normalized Preexponential Factors (% Ai) at Three Emission Wavelengths (λem) and Increasing Water Content (W0) of 7HQ/Water/AOT/n-Heptane W0
λem/nm
τ1/ns
% A1
0.6
430 580 430 505 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600 430 505 600
0.71 0.29 0.60 0.04 0.21 0.01
10 10 16 (-)12 14 (-)46
0.19 0.02 0.06 0.15 0.01 0.06 0.14 0.01 0.04 0.14 0.01 0.03 0.14 0.01 0.05 0.12 0.01 0.04 0.11 0.01 0.03 0.11 0.01 0.03 0.10 0.01 0.03 0.10 0.01 0.03 0.08 0.01 0.04 0.06 0.01 0.04
23 (-)51 (-)13 26 (-)52 (-)15 27 (-)50 (-)19 31 (-)53 (-)31 30 (-)53 (-)17 34 (-)56 (-)18 39 (-)54 (-)24 40 (-)51 (-)24 41 (-)54 (-)25 41 (-)51 (-)24 39 (-)58 (-)20 36 (-)52 (-)19
1.1 1.6 2.6 3.6 4.1 4.6 5.1 5.6 8.1 10.6 13.1 15.6 18.1 20.6
a
τ2/ns
% A2
1.68 0.79 0.62 1.33
27 4 (-)21 31
0.83 0.81 0.75 0.71 0.79 0.50 0.66 0.78 0.34 0.63 0.75 0.34 0.63 0.66 0.28 0.61 0.59 0.27 0.51 0.52 0.27 0.50 0.47 0.26 0.51 0.43 0.27 0.52 0.36 0.28 0.50 0.30 0.29 0.50
(-)28 27 5 (-)26 30 6 (-)25 29 7 (-)21 29 7 (-)25 28 10 (-)23 28 13 (-)20 27 14 (-)20 27 14 (-)18 25 14 (-)17 27 12 (-)18 29 13 (-)18
τ3/ns
% A3
χ2
7.45b 8.75b 7.01b 8.89b 6.20b 7.3b 7.15b 4.93b 5.45b 3.62b 3.46 3.48 3.6b 3.5b 3.00b 3.46b 2.93 2.47 2.90b 2.91 2.51 3.07 2.46 2.13 2.76 1.90 1.93 2.73 1.65 1.96 2.58 1.48 1.96 2.58 1.35 2.07 2.55 1.27 2.09 2.54 1.19 2.16 2.57
90 90 84 88 59 50 73 46 49 59 47 44 58 44 44 56 40 40 47 41 40 59 38 35 58 32 33 57 33 34 56 33 33 57 34 35 58 34 30 62 35 35 63
1.397 1.030 1.199 1.098 1.179 1.079 1.066 1.132 1.136 1.058 1.171 1.286 1.045 1.140 1.168 1.109 1.210 1.128 1.103 1.207 1.128 1.077 1.125 1.195 1.093 1.107 1.189 1.20 1.107 1.138 1.137 1.079 1.125 1.277 1.041 1.158 1.170 1.049 1.096 1.205 1.123 1.167 1.335
(-), negative amplitude, percentage calculated with absolute values. b The fixed parameter during fitting, obtained at longer time scales.
However, we cannot exclude the very few molecules of water (0.15%) that are attached to these heads and that might be involved in the formation of this blue emitting species. Indeed these molecules are necessary to stabilize the transferred proton to the AOT heads. The emission (430 nm) resembles the A band in R2PI (resonant two photon ionization) supersonic-jet experiments with 7HQ‚(NH3)n clusters (444 nm).23 Note that the excitation at 330 nm gives emission bands at 380 and 530 nm according to a previous work.18 As the water content increases, the emission at 430 nm is quenched and slightly shifts up to 10 nm to the red. Clearly, at small water content, W0 ≈ 2, another emission band appears at 530 nm and its intensity continues to increase until saturation at W0 ≈ 10. Based on previous reports, we assign this band to the emission of the T form.21,23 It does not suffer any shift. The fact that the quenching efficiency of the bound E form emission grows monotonically (inset in Figure 2) and the emission of T saturates indicates that T photoformation comes from both excited-state reaction of E‚‚‚AOT and direct excitation of T populated at the ground state. 3.2. Time-Resolved Observation. Upon excitation at 371 nm, compared to the observation at larger W0 values, the fluorescence decay of 7HQ within the micelle at W0 ) 0.6 shows
a small variation with the emission wavelength. Though the emission is not monoexponential (0.71 and 7.45 ns at 430 nm, and 0.29 and 8.75 ns at 580 nm; Table 1), only one band can be recorded in the time-resolved spectrum, shown in Figure 3. The band is attributable to the E strongly bound to the heads of AOT as explained above. Within the picosecond regime, the constructed solvation dynamics correlation function, C(t), shows a biexponential decay with time constants of 20 ps (13%) and 1.81 ns (87%) (inset in Figure 3). Because in nonpolar and nonH-bonding solvents the emission is due to E (380 nm), it is not possible to use, as has been done in other systems,24 the related spectra to extract the zero-time emission spectrum of the E form bound to the AOT headgroups. The involved species of 7HQ in an apolar medium and in the AOT system are completely different. Therefore, zero time was defined as explained in section 2.2, Methods. The extracted times are different from those of the previously reported C(t) for coumarin 152 in the same environment,25 where a ∼15 ns component was explained in terms of ionic motion inside the pool. Although the mean emission lifetime observed for 7HQ (7.45 ns) in the reverse micelle is much longer than that observed for the coumarin (4 ns), we could not detect the 15 ns component in C(t) decay. This difference could reflect either a different degree of
24234 J. Phys. Chem. B, Vol. 110, No. 47, 2006
Angulo et al.
Figure 3. Time-resolved emission spectra of 7HQ in AOT/n-heptane reverse micelles at W0 ) 0.6. Excitation wavelength: 371 nm. The fluorescence maximum energies at t ≈ 0 and ∞ and total dynamic Stokes shift are, respectively, νj0 ) 23 534 cm-1, νj∞ ) 23 004 cm-1, and ∆νj0-∞ ) 530 cm-1. The inset shows the corresponding solvation correlation function, C(t), fitted (time constants in nanoseconds) with C(t) ) 0.13 exp(-t/0.02) + 0.87 exp(- t/1.81).
attachment of these probes to the micellar headgroups or a different localization inside the micelle (trapped between the alkyl chainssunlikely for 7HQsor inside the “pool”). When we increased the water content, several features of the decay noticeably changed: a rise was observable at long wavelengths as the A and the T forms appeared, and all processes accelerated. Figure 4 shows the emission signal for W0 ) 20.6 and at different wavelengths of observation. The figure also shows how the short components, decaying at 430 nm, and rising at 600 nm, evolve with W0. To begin with a simple and interesting observation, two main behaviors can be inferred following the value of W0. Below W0 ≈ 5, the time constants change dramatically (100% at 430 nm), and above this limit there is no or weak variation. The value of W0 ≈ 5 corresponds to the amount of water argued to be needed to start having “free” water molecules inside the nanopool.26 Terahertz spectroscopic measurements have found that the nanopool for W0 below 3 is substantially different from that above this value. The collective relaxation modes of water are suppressed by the confinement when the nanopool of free water has a radius smaller than 8 Å.13 Furthermore, at a W0 value of 5 the viscosity of the water nanopool reaches a plateau (
