J. Phys. Chem. B 2007, 111, 3935-3942
3935
Influence of Electrolytes on the Microenvironment of F127 Triblock Copolymer Micelles: A Solvation and Rotational Dynamics Study of Coumarin Dyes Manoj Kumbhakar*,† and Rajib Ganguly‡ Radiation & Photochemistry and Chemistry DiVisions, Bhabha Atomic Research Center, Mumbai 400 085, India ReceiVed: NoVember 23, 2006; In Final Form: February 16, 2007
Dynamic Stokes’ shift and fluorescence anisotropy measurements of coumarin 153 (C153) and coumarin 151 (C151) as fluorescence probes have been carried out to understand the influence of electrolytes (NaCl and LiCl) on the hydration behavior of aqueous (ethylene oxide)100-(propylene oxide)70-(ethylene oxide)100 (EO100PO70-EO100, F127) block copolymer micelles. A small blue shift in the fluorescence spectra of C153 has been observed in presence of electrolytes due to the dehydration of the oxyethylene chains in the PEO-PPO region, although fluorescence spectra of C151 remain unaltered. The close vicinity of bulk water for C151 probably negates the effect of dehydration in the PEO region. Fluorescence anisotropy measurements indicate a gradual increase in microviscosity with electrolyte concentrations. The partial collapse of copolymer blocks in the presence of electrolytes has been suggested as a reason for the increase in microviscosity along with the strong hydration of ions in the corona region. The interplay between the ion hydration and the mechanically trapped water content, and specific interaction of ions, such as complexation of Li+ ions with the copolymer block, is found to control solvation dynamics in the corona region. In addition to that, it has been established that Na+ ions reside deep into the corona region whereas Li+ ions prefer to reside closer to the surface. Owing to its higher lyotropicity, LiCl influences the corona hydration to a greater extent than NaCl and sets in micelle-micelle interaction above the 2 M LiCl concentration, as reflected in the saturation of solvation time constants. The formation of larger clusters of F127 micelles above 2 M LiCl has been confirmed by dynamic light scattering measurements; however, such cluster formation is not evident with NaCl.
1. Introduction Amphiphilic copolymer micelles consisting of both reasonably hydrophilic (polyethylene oxide, PEO) and reasonably hydrophobic (polypropylene oxide, PPO) blocks drew immense attention as new models for confined reaction media in numerous applications because of their diverse structural behaviors.1-15 Extensive studies have been performed to understand the influence of cosolutes and cosolvents on critical micellar temperature (CMT), critical micellar concentration (CMC), cloud point behavior, shape and phase transitions, etc., employing techniques such as neutron scattering, NMR, static and dynamic light scattering, fluorescence spectroscopy, calorimetry, etc.16-29 Unlike fluorescence spectroscopy, most of these techniques do not involve external probes. The interaction of probe molecules with the micellar internal environments and their dynamics is essential for better understanding and utilization of amphiphilic copolymer micelles for a particular purpose. There are a few reports that aimed to comprehend the micellar microenvironment with proper selection of fluorescent probes.30-38 It has been demonstrated that due to micellization above the CMT probes experience higher microviscosity.31-33 Dynamics of water molecules in the corona region are also reported in literature.34-36 Temperature-dependent solvation dynamic studies * Author to whom correspondence should be addressed. Fax: 91-2225505151 or 91-22-25519613. E-mail:
[email protected]. † Radiation & Photochemistry Division ‡ Chemistry Division.
show formation of water clusters in the corona region due to dehydration of PEO blocks.36 In this paper, we report the influence of electrolytes (NaCl and LiCl) on the hydration behavior of F127 (EO100-PO70-EO100) micelles above the CMT, using coumarin 153 (C153) and coumarin 151 (C151) as fluorescence probes. The CMT value of 5% w/v aqueous solution of F127 is 282.5 K.3 The aim of the present study is to comprehend the effect of electrolyte on the dynamics of water molecules in the micellar corona region and the microviscosity experienced by the probes. It is wellknown that the hydrophobic character of PEO and PPO blocks increases in the presence of inorganic salts due to dehydration.23-29 However, the influence of dehydration on the interaction dynamics of probes dissolved in the micellar phase is still not clear. Moreover, unlike other neutral micelles, such as TritonX-100 ((CH3)3CCH2C(CH3)2-(C6H4)-(OCH2CH2)nOH where n ) 10, TX-100), Brij-35 (CH3(CH2)10CH2-(OCH2 CH2)23OH, BJ35), etc., the degree of hydration is quite less in block copolymer micelles.36 In addition to that, the micellar core (PPO region) of block copolymers can have few water molecules as well.2 Therefore, it is very exciting to know the extent of penetration of inorganic ions inside the micelle. In F127 micelles, C153 resides reasonably deeper in the corona region while C151 resides reasonably closer to the micellar surface.36 So, we will be able to explore two regions of F127 micelles. PEO-PPOPEO block copolymer micelles can be broadly compared with the Triton-X series ((CH3)3CCH2C(CH3)2-(C6H4)-(OCH2CH2)nOH) of micelles, at least for the behavior
10.1021/jp067803e CCC: $37.00 © 2007 American Chemical Society Published on Web 03/24/2007
3936 J. Phys. Chem. B, Vol. 111, No. 15, 2007 CHART 1
Kumbhakar and Ganguly I|(t) and I⊥(t) are the decays for parallel and perpendicular polarizations with respect to the vertically polarized excitation beam. From I|(t) and I⊥(t) decays, the anisotropy decay function r(t) was constructed using the following relation48
r(t) ) of dissolved water in the corona region.36,39 Water molecules in the corona region can be either thermodynamically bound to the ether oxygen’s or nonspecifically associated in the corona region (mechanically trapped water).40 Due to dehydration with the addition of electrolyte, bound water content decreases, and on the contrary mechanically trapped water content increases.41-44 From the resemblance of the corona region of TX-100 to block copolymer micelles one can expect water cluster formation in block copolymer micelles as well, although the extent will depend on the degree of hydration of the micellar system. Hence, comparison of present results with those reported in TX-10045-47 is expected to further elaborate the effect of electrolyte on the microenvironment of block copolymer micelles. The chemical structures of C153 and C151 probes are shown in Chart 1. 2. Experimental Section The F127 copolymer was obtained from Sigma and used without further purification. Laser grade C153 and C151 were obtained from Exciton, USA, and used as received. NaCl and LiCl were obtained from Fluka, Switzerland, and used without further purification. Nanopure water, having a conductivity of ∼0.1 µS cm-1, was obtained by passing distilled water through a Barnstead Nanopure Water System and used for the preparation of the micellar solutions. In the experimental solutions, the concentration of F127 was kept at 5% w/v. The required amount of block copolymer was weighed and kept overnight under refrigeration after adding the requisite amount of water in a sealed container. Samples were prepared by adding C153 and C151 in the aqueous solutions of block copolymers at room temperature. The concentration of C153 or C151 was kept much lower in comparison to the micelle concentration, ∼50 µM. Steady-state absorption spectra were recorded using a JASCO (Tokyo, Japan) model V530 spectrophotometer. Fluorescence spectra were recorded using a Hitachi (Tokyo, Japan) model F-4010 spectrofluorometer. Time-resolved fluorescence measurements were carried out using a diode-laser-based timecorrelated single-photon-counting (TCSPC)48 spectrometer from IBH, U. K. In the present work a 408 nm diode laser (1 MHz) was used as the excitation source, and a TBX4 detection module (IBH, U. K.) coupled with a special Hamamatsu photomultiplier tube (PMT) was used for the detection of the fluorescence decays. For the present setup the instrument response function was ∼230 ps at the full width at half-maximum. The time resolution achievable with the present setup following deconvolution analysis of the fluorescence decays is ∼50 ps. For solvation dynamics studies, fluorescence decays were recorded with a vertically polarized excitation beam, and fluorescence was collected at the magic angle (54.7°). Measurements were repeated three times, both to check the reproducibility and to obtain the average values of the relaxation times. All of the measurements were carried out at ambient room temperature (298 ( 1 K) using a microprocessor-based temperature controller (model DS from IBH). Fluorescence anisotropy measurements were carried out by measuring polarized fluorescence decays, I|(t) and I⊥(t), where
I|(t) - GI⊥(t) I|(t) + 2GI⊥(t)
(1)
where G is a correction factor for the polarization bias of the detection setup. The G factor was independently obtained by using the horizontally polarized excitation beam and measuring the two perpendicularly polarized fluorescence decays with respect to the excitation polarization. Time-resolved emission spectra (TRES) were generated from a set of fluorescence decays collected at different wavelengths, covering the entire emission band of the probe. All of the decays were fitted as three exponential functions. The emission spectrum at a given time t, S(λ,t), was obtained from the series of the fitted decays, D(λ,t), after their normalization with respect to the steady-state fluorescence spectrum, S0(λ), and using the following equation49
S(λ,t) ) D(λ,t)
S0(λ)
∫0
∞
(2)
D(λ,t) dt
The method used to obtain the spectral shift correlation function C(t) from the TRES is well-described by Maroncelli and Fleming.49 In the present context it is to be mentioned that the steady-state fluorescence spectrum has been used as the infinite time spectrum, S0(λ), while constructing the C(t) curve. Dynamic light scattering (DLS) measurements of the solutions were performed using a Malvern 4800 Autosizer employing a 7132 digital correlator. The light source was an argon ion laser operated at 514.5 nm with a maximum output power of 2 W. The electric field autocorrelation function [g1(τ)] versus time data collected by the correlator were analyzed, and the average decay rates of the correlation functions were obtained. Measurements were made at five different angles ranging from 50° to 130°. The average decay rates (Γ) of the scattering species have been found to vary linearly with q2, (Γ ) Dq2), indicating translational diffusion of the scatterers (q being the magnitude of the scattering vector given by q ) [4πn sin(θ/2)]/λ, where n is the refractive index of the solvent, λ is the wavelength of the laser light, and θ is the scattering angle). From the slopes of the Γ vs q2 plots, the apparent translational diffusion coefficient (D) values were estimated, and from these values the apparent equivalent hydrodynamic radii (RH) of the micelles were calculated using the Stokes-Einstein relationship
RH ) kT/6πηD
(3)
where η is viscosity of the solvent, k is the Boltzmann constant, and T is the temperature. When k, T, η, and RH are expressed in SI units (kg m2 s-2 K-1, K, kg m-1 s-1, and m, respectively) the unit of D is m2 s-1. 3. Results and Discussions Steady-state fluorescence spectra of C153 and C151 were recorded in the absence and the presence of electrolytes, and the emission maxima, λmax flu , are listed in Table 1. Figure 1 shows the fluorescence spectra of both the dyes in the absence and the presence of 3 M NaCl. A close look at Figure 1 reveals a blue shift of 5 nm for C153 in the presence of electrolytes while the spectrum remains almost unaltered for C151, indicat-
F127 Triblock Copolymer Micelles
J. Phys. Chem. B, Vol. 111, No. 15, 2007 3937
Figure 1. Normalized fluorescence spectra of C153 (9) and C151 (b) in F127 micellar solution. In the presence of 3 M NaCl, C151 spectra (O) do not change, but C153 spectra (0) show a slight blue shift.
TABLE 1: Fluorescence Maxima (λmax flu ) and Lifetime (τ0) Values of C153 and C151 in the Presence of Different Concentrations of NaCl and LiCl in a 5% Aqueous Solution of F127a C153 NaCl
C151 LiCl
NaCl
LiCl
electrolyte concentration (M)
λmax flu (nm)
τ0 (ns)
λmax flu (nm)
τ0 (ns)
λmax flu (nm)
τ0 (ns)
λmax flu (nm)
τ0 (ns)
0 0.2 0.5 1.0 1.5 2.0 2.5 3.0
530 529 528 527 527 526 525 525
5.2 5.3 5.3 5.3 5.3 5.3 5.3 5.2
529 529 528 527 526 526 526
5.2 5.3 5.3 5.3 5.3 5.3 5.3
489 489 489 489 489 489 490 490
5.9 5.8 5.8 5.8 5.7 5.7 5.6 5.6
489 489 489 489 489 490 490
5.8 5.8 5.8 5.7 5.6 5.6 5.6
The error limits in λmax flu and τ0 are (0.5 nm and (0.1 ns, respectively. a
ing a change in polarity for the C153 microenvironment. Though the blue shift for C153 is small, it is significant in comparison to unaltered spectra of C151 in the presence of electrolytes. Being more hydrophilic in nature, C151 resides relatively more toward the micellar surface in the PEO region than C153, which is reported to reside in the PPO-PEO region.36 Moreover, these unlike probe solubilization sites for C153 and C151 in F127 micelles possibly result in different spectral shifts in the presence of electrolytes. Dehydration of polyethyleneoxide chains in the presence of electrolytes has been reported earlier in case of TX100 micelles.41-45 Assuming similar dehydration in F127 micelles in the presence of electrolytes, a marginal blue shift is expected as coumarin dyes are very sensitive to the changes in micropolarity. However, such a shift in emission spectra has not been observed in TX-100 micelles,45-47 probably due to a higher degree of hydration in TX-100 micelles compared to that in F127 micelles. Comparison of Stokes’ shifts in these two micelles indicates greater hydration in TX-100 than in F127.36 Though the exact reason is not yet clear to us, it is probably related to the compact structure of F127 micelles in comparison to TX-100 micelles.36 Moreover, comparison of electrolyte influence on the fluorescence spectra of C153 and C151 suggests that the influence of dehydration is comparatively larger in the PPO-PEO interface region than that in the PEO region. Perhaps the close vicinity of bulk water downplays the effect of dehydration in the PEO region.
Figure 2. (A) Fluorescence anisotropy decays of C153 in F127 micellar solutions in the absence (b) and the presence (O) of 3 M NaCl. The solid line represents the biexponential fit curves. (B) 〈τr〉 as a function of electrolyte concentrations for C153-NaCl (9), C153-LiCl (0), C151-NaCl (b), and C151-LiCl (O) systems in F127 micellar solution. Solid lines show the overall trend in 〈τr〉 values. 〈τr〉 values of C153 (open squares with a cross) and C151 (open circles with a cross) in the absence of electrolytes in F127 micellar solutions are also shown in the figure.
Fluorescence decays were measured at the emission maxima of the dyes in F127 micellar solution as a function of electrolyte concentration and are seen to be fit well with a singleexponential function. It is observed that the fluorescence lifetimes (τ0, Table 1) of the dyes do not change appreciably in the presence of electrolytes. In the present work, fluorescence anisotropy decays of both of the probes in F127 micelles were measured as a function of electrolyte concentration. Anisotropy decays in the present systems are seen to be biexponential in nature, as also reported in the literature for other micellar systems.31,35,36,45-47 Fluorescence anisotropy decays of C153 in the absence and the presence of 3 M NaCl in F127 micellar solutions are shown in Figure 2A. Rotational relaxation times for the present systems along with their relative contributions at different salt concentrations are listed in Table 2. Average rotational relaxation times, 〈τr〉, also listed in Table 2, were calculated using the following relation
〈τr〉 ) ar1τr1 + ar2τr2
(4)
where τr1 and τr2 are the fast and the slow time constants and ar1 and ar2 are their relative contributions, respectively. In a micellar solution with the probe in the hydrated layer, three different kinds of motions can contribute to determine the fluorescence anisotropy decay dynamics, r(t). They are: wobbling motion of the probe in the micellar corona region, lateral diffusion of the probe along the micellar surface, and rotational motion of the whole micelle.31,36,39,45-47 Due to the interplay of these motions, as rigorously treated in the well-known twostep model,50,51 the anisotropy decays in micellar solutions
3938 J. Phys. Chem. B, Vol. 111, No. 15, 2007
Kumbhakar and Ganguly
TABLE 2: Time Constants of the Anisotropy Decays as Obtained from the Biexponential Fit along with the Average Anisotropy Decay Times of C153 and C151 in 5% w/v Aqueous Solution of F127 as a Function of Electrolyte Concentrationsa electrolyte
probe
NaCl
C153
C151
LiCl
C153
C151
concentration (M)
τr1 (ns)
ar1 (%)
τr2 (ns)
ar2 (%)
〈τr〉 (ns)
0 0.2 0.5 1.0 1.5 2.0 2.5 3.0 0 0.2 0.5 1.0 1.5 2.0 2.5 3.0
0.9 0.5 0.6 0.5 0.6 0.8 0.8 0.8 0.6 0.6 0.7 0.7 0.7 0.7 0.7 0.9
21.9 4.0 3.4 8.8 8.9 8.5 10.8 12.1 13.6 10.2 9.8 8.1 8.3 8.8 9.7 16.0
2.9 2.6 2.7 3.1 3.1 3.3 3.4 3.5 3.6 3.6 3.7 3.7 3.7 3.7 3.8 4.0
78.1 96.0 96.6 91.2 91.7 91.5 89.2 87.9 86.4 91.2 90.2 90.9 91.7 91.2 90.3 84.0
2.5 2.5 2.6 2.9 2.9 3.1 3.1 3.2 3.2 3.3 3.4 3.4 3.5 3.4 3.5 3.5
0.2 0.5 1.0 1.5 2.0 2.5 3.0 0.2 0.5 1.0 1.5 2.0 2.5 3.0
0.6 0.5 0.6 0.5 0.6 0.6 0.7 0.6 0.6 0.7 0.6 0.7 0.7 0.8
7.9 6.7 8.9 9.6 9.2 9.6 9.4 8.2 7.4 9.2 8.2 8.4 8.9 11.4
2.8 2.8 3.0 3.0 3.2 3.3 3.4 3.7 3.6 3.8 3.7 3.8 3.8 4.0
92.1 93.3 91.1 90.4 90.8 90.4 90.6 91.8 92.6 90.8 91.8 91.6 91.1 88.6
2.6 2.7 2.8 2.8 3.0 3.0 3.2 3.5 3.4 3.5 3.5 3.5 3.5 3.6
a The error limits in τ , τ , and 〈τ 〉 are (0.1, (0.2, and (0.2 ns, r1 r2 r2 respectively.
should be effectively biexponential in nature. In relation to the present results, the contribution of the whole micelle rotation to the observed anisotropy decays seems to be quite negligible, as the correlation time constants for the rotation of the whole micelles (τM) are roughly more than one order of magnitude higher than even the longer anisotropy component, τr2, observed for the present systems.36 Thus, we can consider that the observed r(t) decays in the present systems are mainly determined by the wobbling motion and lateral diffusion of the probe in the micelle and the rotation of the whole micelle does not contribute significantly to the observed decays.36,39 Accordingly, the changes in the rotational relaxation times are not directly related to the changes in the micellar size with the added electrolyte concentrations. We thus infer that the changes in the rotational relaxation times reflect the changes in the microviscosity inside the micellar corona region. In the present context we will simply consider the average rotational relaxation times directly to obtain a qualitative understanding of the changes in microviscosity in the micellar corona region with added electrolytes. It is indicated from Table 2 that both τr2 and τr1 increase with the addition of electrolytes for C151. Though τr values for C153 also increase marginally, they show a slight decrease at very low electrolyte concentrations. The fluctuations in τr values are probably related to the limited time resolution of the present TCSPC setup. Average rotational relaxation times increase linearly with electrolyte concentration (Figure 2B). From the similarities of the corona regions of the TX-100 and F127 micelles, it is suggested that the ions reside in the corona region, and due to the hydration of the ions, the water molecules
in the corona region undergo a kind of clustering.41-47 Hence, microviscosity in fact increases rather than decreasing as expected due to increased mechanically trapped water content.45-47 A partial collapse of the surfactant chains due to their dehydration also adds to the increase in microviscosity and the consequent retardation in relaxation dynamics in the presence of electrolytes.42,45-47 The collapse of PEO blocks due to inorganic ions also has been reported by Florin et al.27 and Bahadur et al.28 Higher values of rotational relaxation times for C151 compared to those for C153 are probably due to the differences in probe locations in the micellar phase. Although, the possibility of hydrogen bonding with the surrounding water molecules cannot be neglected for C151. To substantiate the possibility of hydrogen bonding with the surrounding water molecules for C151, we have carried out anisotropy measurements of both of the dyes in F127 micelles with D2O as the solvent instead of H2O. Because D2O forms stronger hydrogen bonds than H2O, rotational hindrance is expected to increase more for C151 than for C153. Additionally, due to stronger hydrogen bonds, viscosity in D2O is a bit higher than that in H2O. In fact, anisotropy decays were observed to be slower in D2O than in H2O, but the extent is higher for C151 than for C153. Particularly, the fast rotational time constant (τr1) for C151 increases from 0.6 to 0.9 ns (∼50%) in D2O whereas this increase is lesser for C153, from 0.9 to 1.1 ns (∼22%). Slow rotational time constants in D2O were observed to increase with respect to H2O by 17% for C151 (3.6 to 4.2 ns) and by 14% for C153 (2.9 to 3.3 ns). More pronounced hindrance for the wobbling motion (τr1) of C151 in D2O than for C153 is expected only when the former probe is involved in a hydrogen-bond interaction with the surrounding water molecules. Therefore, a larger extent of increase in rotational time constants in D2O for C151 than for C153 permit us to infer that the hydrogen-bonding ability of C151 also contributes to the higher rotational relaxation times in the present case. Moreover, Figure 2B also indicates that the rise in rotational hindrance in the presence of electrolyte is somewhat higher for C153 than for C151. Perhaps collapse of copolymer blocks as a result of dehydration starts form inside the micelles, i.e., the PPO-PEO interface region, and its extent is less in the PEO region. This is in accordance with the steady-state fluorescence measurements. Wavelength-dependent fluorescence decays obtained for both of the dyes in F127 micellar solutions indicate the presence of time-dependent Stokes’ shifts for the emission spectra at all electrolyte concentrations studied. The TRES of the probes in F127 micellar solutions at various electrolyte concentrations were generated using the spectral reconstruction method described in the literature.49 The dynamic Stokes’ shifts of the TRES were used to construct the normalized correlation function C(t) as
C(t) )
ν(t) - ν(∞) ν(0) - ν(∞)
(5)
where ν(0), ν(t), and ν(∞) correspond to the frequencies of the emission maxima at times 0, t, and ∞, respectively. The C(t) curves as obtained for C151 in the presence and the absence of LiCl are shown in Figure 3. It is clearly indicated from these C(t) curves that solvation dynamics in F127 micelles follow a non-single-exponential behavior. Solvation time constants for the present systems were obtained following biexponential analysis of the C(t) curves and are listed in Table 3 along with their percentage contributions. In micellar media we suppose that the longer solvation component (τs2) is mainly determined by the exchange rates between the mechanically trapped and
F127 Triblock Copolymer Micelles
J. Phys. Chem. B, Vol. 111, No. 15, 2007 3939
Figure 3. Normalized spectral shift correlation functions C(t) of C151 in the presence of 0 M LiCl (dash line), 0.2 M LiCl (solid line), and 3 M LiCl (dotted line). Inset: For better comparison of slow solvation time constants, C(t)’s have been plotted on a log scale.
TABLE 3: Solvation Times as Obtained from the Biexponential Analysis of the Normalized Spectral Shift Response Function C(t) for C153 and C151 in F127 Micellar Solution at Different Electrolyte Concentrations along with Their Relative Contributionsa ∆νjsol (103 cm-1) electrolyte probe NaCl
C153
C151
LiCl
C153
C151
concentration τs1 (M) (ns)
as1 (%)
τs2 (ns)
as2 (%)
∆νjsol jsol obs ∆ν total
0 0.2 0.5 1.0 1.5 2.0 2.5 3.0 0 0.2 0.5 1.0 1.5 2.0 2.5 3.0
0.3 0.4 0.3 0.2 0.4 0.5 0.3 0.4 0.8 0.7 0.3 0.3 0.4 0.3 0.3 0.4
53.24 48.81 34.97 27.05 43.89 42.64 47.28 46.38 51.18 54.73 53.21 53.34 57.63 54.79 57.26 59.54
2.6 2.4 1.8 1.6 2.0 2.2 2.1 2.3 4.1 3.4 3.0 2.8 3.2 3.3 3.2 3.5
46.76 51.19 65.03 72.95 56.11 57.36 52.72 53.62 48.82 45.22 46.79 46.66 42.37 45.21 42.74 40.46
1.3 1.2 1.1 1.0 1.0 1.0 1.1 1.1 0.5 0.5 0.6 0.5 0.6 0.6 0.6 0.6
0.2 0.5 1.0 1.5 2.0 2.5 3.0 0.2 0.5 1.0 1.5 2.0 2.5 3.0
0.3 0.4 0.4 0.4 0.4 0.4 0.3 0.3 0.3 0.3 0.2 0.2 0.3 0.3
48.11 41.64 44.96 41.27 44.46 48.43 43.13 55.25 56.69 59.41 62.74 63.75 53.37 56.12
2.3 2.3 2.3 2.5 2.4 2.3 2.3 2.8 3.0 3.1 3.5 3.2 3.2 3.1
51.89 58.36 55.04 58.73 55.54 51.57 56.87 44.75 43.31 40.59 37.26 36.25 46.63 43.88
1.2 1.1 1.1 1.1 1.1 1.2 1.1 0.6 0.6 0.7 0.8 0.8 0.7 0.7
1.3
1.5
a The Stokes’ shift (∆ν jsol) due to solvation is also listed in the table. The error limit in τs1 and τs2 are (0.1 and (0.2 ns, respectively. ∆νjsol values were reproduced within (100 cm-1.
the thermodynamically bound (hydrogen-bonded to oxyethylene chains) water molecules that are adjacent to the probe, and the shorter solvation component (τs1) is mainly determined by the collective response of a relatively large number of water molecules that are somewhat away from the probe.36,39 The response of mechanically trapped water molecules is expected to be comparatively faster than that of the bound water molecules and should lead to rapid solvation.
Figure 4. Solvation time constants for C153 and C151 as a function of (A) NaCl and (B) LiCl concentrations in F127 micellar solution. The overall trend in the values of τs2 for C153 (O) and C151 (0) is similar for an electrolyte. Variations in the τs1 values for C153 (b) and C151 (9) are within experimental fluctuations.
It is to be noted that due to faster solvation a significant fraction of the initial dynamic Stokes’ shift is beyond our time resolution and only the slower part is observed especially with C151. (In Table 3, compare ∆νjsol jsol obs and ∆ν total, the observed and total expected Stokes’ shifts due to solvation. ∆νjsol total was estimated from the Stokes’ shifts for the probes in a nonpolar solvent cyclohexane and in block copolymer micelles, as obtained from absorption and steady-state fluorescence spectra.) In the present study, observed dynamics Stokes’ shifts (∆νjsol obs) are approximately 1100 and 600 cm-1 for C153 and C151, respectively. A comparison of ∆νjsol obs values indicates that the contribution of the inertial component is much greater in the PEO region (C151) than in the PEO-PPO region (C153). Therefore, the microenvironments for C153 and C151 are not alike in the F127 micelle. In the construction of the C(t) curves, in the present work we always used the initial spectra obtained at the shortest possible time window (∼50 ps, following the deconvolution analysis of the fluorescence decays) with the present TCSPC setup. Individual time constants for both of the probes have been plotted against electrolyte concentration in Figure 4 for better visual comparison. The interesting point to be noted from Figure 4 and Table 3 is that the change in the C(t) function with electrolyte is not a gradual one. It has been observed that slow solvation time (τs2) initially decreases with the addition of NaCl in the F127 micellar solution up to a concentration of 1 M and above that it increases gradually. On the contrary, τs2 decreases suddenly on the addition of LiCl (0.2 M) and then increases up to a concentration of 2 M and almost saturates on further addition of LiCl. The overall trend in the variation of τs2 with NaCl or LiCl is quite similar for both the probes although the extent of variation is different (Figure 4). Moreover, the fast solvation time constant (τs1) for C151 decreases in the presence of electrolytes, while τs1 remains almost unaltered for C153. Fast solvation time constants observed in the present cases are close to the instrument response function of our TCSPC setup, and the fluctuations in τs1 values are within the experimental error limit. It is now interesting to compare the trend of τs2 values in the presence of NaCl and LiCl. Increased mechanically trapped water in the micellar corona region with added electrolytes should increase solvation rates.39 On the contrary, the motion
3940 J. Phys. Chem. B, Vol. 111, No. 15, 2007 of these water molecules in the micellar phase effectively becomes retarded due to strong hydration of the ions, causing slow solvation.45 Therefore, there is interplay of two opposing factors in the presence of electrolytes in micellar systems. At low electrolyte concentrations of NaCl, increased mechanically trapped water content rules over the strong hydration of the ions, resulting in a decrease in τs2 values. However, above 1 M NaCl, it is the strong hydration of the ions that dominates over mechanically trapped water and controls solvation dynamics. This behavior with LiCl is little different, as unlike Na+ ions, Li+ ions are reported to undergo complexation with the ether groups of the surfactant molecules, where the latter act as the polydendate ligand.46,47,52 Thus, at low LiCl concentrations, Li+ ions in the corona region are engaged in complex formation with the ether groups. These Li+ ions do not perturb the mobility of the mechanically trapped water molecules in the micellar corona region, as otherwise happens with Na+. As a result, at low LiCl concentrations solvation dynamics are expected to be faster due to enhanced mechanically trapped water. Hence, the τs2 value decreases sharply at low LiCl concentrations (0.2 M). At higher LiCl concentrations, the binding of Li+ ions saturates, and they start residing in the corona region. Once free Li+ ions are available in the PEO region, they retard the motion of mechanically trapped water molecules in the corona region. The extent of influence on the mobility of mechanically trapped water molecules will be greater with Li+ ions than that with Na+ ions, due to its higher lyotropicity.43,44 As a consequence, we see a moderately steep increase in τs2 values above 0.2 M LiCl (Figure 4). On the whole, we see a rather steep decrease and then an increase in τs2 values around 0.2 M LiCl unlike the slow decrease and then gradual increase with NaCl around 1 M. In the presence of 1 M NaCl, the τs2 value for C151 and C153 decreases by almost a similar extent from 4.1 to 2.8 ns (∼32%) and 2.6 to 1.6 ns (∼38%), respectively. However, in the presence of LiCl (0.2 M), the τs2 value decreases to 2.8 ns (∼32%) and 2.3 ns (∼12%) for C151 and C153, respectively. These values indicate that the extent of influence of NaCl and LiCl on the microenvironment of C151 in F127 micelle is almost equal. However, the influence on the C153 microenvironment is more pronounced in the presence of NaCl than that with LiCl. It has been previously established that C151 resides in the PEO region, closer to the surface, while C153 resides quite deep in the PEO-PPO interface region. Therefore, present results indicate that though Na+ ions penetrate deep into the PEO-PPO region Li+ preferably resides in the PEO region. This is also expected from the consideration of the lower degree of hydration of triblock copolymer micelles along with the higher lyotropicity of Li+ ions compared to that of Na+ ions. However, the influence of hydration on solvation times for a probe residing closer to the surface is expected to be less compared to the one that resides deep in the micellar corona phase. The contribution from bulk water will be higher for the probe residing closer to the surface. The increase in the τs2 value due to the hydration of free Li+ ions above 0.2 M LiCl is distinctly higher for C151 than that for C153 (Figure 4). This observation further corroborates our above conclusion that Li+ ions reside close to the micellar surface only. Therefore, in the present systems, observed variations in solvation dynamics are due to different probe solubilization sites in the Corona region along with the nature of the cations. However, Aswal et al. suggest that the nature of the anions is more important than that of the cations and the magnitude of the changes depends on the size of the anions.25 Also, changes in the microenvironment for C153 and
Kumbhakar and Ganguly C151 in the presence of electrolytes can also be studied by Raman spectroscopy. The trend in Raman shifts can follow alterations in the microenvironment, as they are very sensitive to the different chemical environments. In this respect it is worth mentioning the works of Aliotta et al.53 and Dillon et al.54 They have demonstrated the existence of bulk water clusters and hydrated electrolyte clusters using Raman spectra of electrolyte solutions. Extending such studies to micellar media will be complex due to the involvement of different kinds of water, e.g., mechanically trapped water, thermodynamically bound water, hydrated electrolyte water clusters, bulk water, and the tethered water molecules located near the PPO blocks of the copolymers. However, in pursuit of intricate details of water structures in these organized assemblies, Raman spectroscopy can be thought of as a complementary tool in future studies. At this juncture it is worth comparing the microviscosity changes in the corona region of the F127 micelles with those of the corona region of the TX-100 micelles.45-47 The increase in microviscosity for both of the probes in the presence of NaCl is fairly similar in both F127 and TX-100 micelles but very different in presence of LiCl. It has been observed that rotational relaxation times of C153 follow similar trends to the solvation times with the addition of LiCl in TX-100 micelles; i.e., initially microviscosity decreases and then increases at approximately 1.5 M LiCl.46 No such behavior has been observed in F127 micelles. Rotational relaxation times for both of the probes increase gradually. Possible reasons could be (i) that the larger microviscosity experienced by the probes in the F127 micelles compared to that in the TX-100 micelles suppresses the decrease in the microviscosity due to enhanced mechanically trapped water content as a result of Li+ ion binding to the PEO chains at low concentrations, (ii) that due to less hydration Li+ ions do not go into the PEO region for complexation, or (iii) that due to less hydration very few Li+ ions actually bind to the ether oxygen, leading to a marginal change in microviscosity that is hard to detect. Very few Li+ ions actually bind to the oxyethylene blocks of F127, which is also evident from the fact that free Li+ ions are available for coordination with the mechanically trapped water molecules above 0.2 M LiCl in F127 micelles whereas in TX-100 it occurs only above 1.5 M.46 Though it is hard to address specifically on the basis of the above causes responsible for the present observation, the last one seems to be more plausible. Above 2 M LiCl the τs2 value more or less saturates around 3.2 ns (Figure 4). However, no such behavior has been observed for NaCl; instead the τs2 value increases gradually up to 3 M. Saturation in slow solvation times at higher LiCl concentrations indicates the participation of something other than ion hydration and water cluster formation. It is well-known that with increases in electrolyte concentration the number of water molecules that are thermodynamically bound to PPO and PEO blocks decreases.23-29,41-44 These water molecules hinder micellar interaction.27,28,43 The decrease in thermodynamically bound water content consequently increases the chances of micellemicelle interactions and thus reduces the cloud point.27,28,43 As Li+ ions are more hydrated than Na+ ions, the cloud point of F127 is expected to be much lower in the presence of LiCl than in the presence of NaCl.43 Hence, the saturation above 2 M LiCl in F127 is probably an indication of micelle-micelle interaction, which is still absent in the case of NaCl up to a concentration of 3 M. To comprehend these aspects of hydration and micelle-micelle interaction, we carried out DLS studies of F127 micelles in the presence of various concentrations of LiCl and NaCl.
F127 Triblock Copolymer Micelles
J. Phys. Chem. B, Vol. 111, No. 15, 2007 3941 colloids with hard sphere interaction. Linear regressions to the data give values of D0 as 2.42 × 10-11 and 1.93 × 10-11 m2 s-1 and k0 values as 5.1 and 7.7 for 0 and 2 M LiCl, respectively. The observed values of k0 are much higher than what is expected for hard sphere colloids, which suggests that due to a large degree of hydration in the corona region the effective volume fraction of the solvated micelles is significantly more than the volume fraction of the polymer.23,58 The higher value of k0 in presence of 2 M LiCl suggests that an enhanced degree of micellar hydration leads to the observed increase in the size of the micelles upon the addition of LiCl. This enhanced hydration is somewhat unusual since the electrolyte enhances the hydrophobic character of the copolymer and reduces the CMC, CMT, and cloud point of the solutions.28,59 As discussed earlier, this enhanced hydration results from the enhanced amount of mechanically trapped water within the corona region of the micelles. The presence of these compensates more than the amount of water lost in the corona region due to the dehydration of PEO chains in presence of electrolytes. The behaviors of the copolymer solutions containing 3 and 4 M LiCl concentrations are also shown in Figure 5A. At these electrolyte concentrations, the correlation function diagrams are no longer representative of the presence of only micelles but also larger clusters along with them. The analysis of the correlation data was thus done based on the biexponential equation
Figure 5. (A) Correlation function, g1(τ), vs time plots for 5% F127 micellar solutions with 0 M LiCl (0), 2 M LiCl (O), 3 M LiCl (4), and 4 M LiCl ()) concentrations at 303 K. The solid lines are fits to the data points. Inset: Variation of the hydrodynamic radius, RH, of the micelles as a function of LiCl concentration. The symbols include error limits for the calculated RH values. (B) Diffusion coefficient (D) vs copolymer volume fraction plots for solutions containing 0 M (0) and 2 M LiCl (O). The symbols include error limits for the calculated D values.
The influence of LiCl on the structure of the F127 micelles is shown in Figure 5. Figure 5A shows the autocorrelation function [g1(τ)] vs time plots as a function of LiCl concentrations. The nature of the plot remains the same up to 2 M LiCl concentrations and reflects the presence of only micellar species. The decay rates (Γ) of the correlation functions for solutions up to 2 M LiCl were obtained by analyzing the plots using the method of cumulants.55,56 The diffusion coefficient and the corresponding hydrodynamic radius values were obtained from the Γ vs q2 plots. As shown in the inset of Figure 5A, the hydrodynamic size of the micelles increases upon the addition of LiCl; this is manifested in a shift of the correlation function diagram to a higher time scale. The F127 micelles are coreshell particles with the core being composed of the hydrophobic PPO units and the shell being composed of hydrated PEO units.4,22,57 The core size, which determines the micellar aggregation number, and the shell size, which determines the degree of micellar hydration, can independently control the size of the F127 micelles. To understand the role of LiCl in increasing the micellar size, we have studied the variation of the diffusion coefficient with the concentration of the copolymer both in the presence and in the absence of 2 M LiCl (Figure 5B). The observed variation of the diffusion coefficient with the volume fraction of the copolymer, φ, can be expressed as
Deff ) D0(1+ k0φ)
(6)
where Deff is the effective diffusion coefficient at any volume fraction φ, D0 is diffusion coefficient at infinite dilution, and k0 is the diffusion virial coefficient that has a value of 1.56 for
g1(τ) ) Af exp(-τΓf) + As exp[-(τΓs)]
(7)
where Af and As are the amplitudes for the fast and slow relaxation modes corresponding to the decay rates of the correlation functions Γf and Γs of the micelles and the clusters, respectively. The hydrodynamic radii of the micelles obtained from this analysis still show an increasing trend with an increase in the LiCl concentration (inset of Figure 5A). Certain pluronics are known to form similar clusters along with the micelles at temperatures close to the CMT. Ho et al.60 observed similar cluster formation in the aqueous solutions of PEO and suggested that hydrogen bonding between PEO monomer groups mediated by water molecules is responsible for this. In the present case the enhanced amount of mechanically trapped water in the corona region of the micelles in the presence of LiCl could play a role in the formation of these clusters. A detailed study on the influence of LiCl on the CMC and CMT of F127 in the aqueous medium will probably shed some light on the formation of clusters in the present case. Though gradual increase in micellar size and hydration has also been observed with the addition of NaCl, larger clusters have not been detected at higher NaCl concentrations. Moreover, present DLS results substantiate our assumption that above 2 M LiCl there is micelle-micelle interaction leading to the formation of larger clusters, which is absent in the case of NaCl. The observation of unchanged τs2 values above 2 M LiCl is probably related to the changes in the hydration behavior of the corona region due to the formation of larger clusters in the copolymer solutions. Therefore, not only the hydration behavior of the micelles but also their interaction with other micelles can be addressed from solvation time constants. 4. Conclusions Effects of electrolyte on the hydration behavior of F127 triblock copolymer micelles have been studied. Partial collapse of the surfactant chains due to their dehydration in presence of electrolytes impedes rotational relaxation dynamics. In the
3942 J. Phys. Chem. B, Vol. 111, No. 15, 2007 presence of electrolytes in the micellar corona region, the interplay of ion hydration and the mechanically trapped water content governs solvation dynamics. Variation in the trends of solvation time constants in the presence of NaCl and LiCl is due to the complexation of Li+ ions with the copolymer block. Different locations for C151 and C153 in the corona region have also been argued. It has been established that Na+ ions reside comparatively deep in the corona region in comparison to Li+ ions. Above 2 M LiCl in F127 micellar solution, micellemicelle interaction initiates, which is absent in the case of NaCl. Acknowledgment. We are thankful to Dr. H. Pal, Dr. S. K. Sarkar, and Dr. T. Mukherjee for their encouragement and support. References and Notes (1) Alexandridis, P.; Lindman, B. Amphiphilic Block Copolymers: SelfAssembly and Applications; Elsevier: New York, 2000. (2) Chu, B.; Zhou, Z. Physical chemistry of polyoxyalkylene block copolymer surfactants. In Nonionic Surfactants; Nace, V. M., Ed.; Surfactant Science Series 60; Marcel Dekker: New York, 1996; p 67. (3) Alexandridis, P.; Holzwarth, J. F.; Hatton, T. A. Macromolecules 1994, 27, 2414. (4) Mao, G.; Sukumaran, S.; Beaucage, G.; Saboung, M.-L.; Thiyagarajan, P. Macromolecules 2001, 34, 552. (5) Wanka, G.; Hoffmann, H.; Ulbricht, W. Macromolecules 1994, 27, 4145. (6) Mortensen, K.; Pedersen, J. S. Macromolecules 1993, 26, 805. (7) Noda, T.; Hashidzume, A.; Morishima, Y. Langmuir 2000, 16, 5324. (8) Duxin, N.; Liu, F.; Vali, H.; Eisenberg, A. J. Am. Chem. Soc. 2005, 127, 10063. (9) Alexandridis, P.; Hatton, T. A. Colloids Surf., A 1995, 96, 1. (10) Zhang, K.; Khan, A. Macromolecules 1995, 28, 3807. (11) Scherlund, M.; Welin-Berger, K.; Brodin, A.; Malmsten, M. Eur. J. Pharm. Sci. 2001, 14, 53. (12) Allen, C.; Maysinger, D.; Eisenberg, A. Colloids Surf., B 1999, 16, 3. (13) Varshney, M.; Morey, T. E.; Shah, D. O.; Flint, J. A.; Moudgil, B. M.; Seubert, C. N.; Dennis, D. M. J. Am. Chem. Soc. 2004, 126, 5108. (14) Zhang, R.; Liu, J.; Han, B.; He, J.; Liu, Z.; Zhang, J. Langmuir 2003, 19, 8611. (15) Ma, Y.; Qi, L.; Ma, J.; Cheng, H.; Shen, W. Langmuir 2003, 19, 9079. (16) Ivanova, R.; Lindman, B.; Alexandridis, P. Langmuir 2000, 16, 3660. (17) Alexandridis, P.; Ivanova, R.; Lindman, B. Langmuir 2000, 16, 3676. (18) Jansson, J.; Schillen, K.; Nilsson, M.; Soderman, O.; Fritz, G.; Bergmann, A.; Glatter, O. J. Phys. Chem. B. 2005, 109, 7073. (19) Elisseeva, O. V.; Besseling, N. A.; Koopal, L. K.; Cohen Stuart, M. A. Langmuir 2005, 21, 4954. (20) Su, Y.; Wei, X.; Liu, H. J. Colloid Interface Sci. 2003, 264, 526. (21) Lisi, R. D.; Milioto, S. Langmuir 2000, 16, 5579. (22) Ganguly, R.; Aswal, V. K.; Hassan, P. A.; Gopalakrishnan, Kulshreshtha, S. K. J. Phys. Chem. B 2006, 110, 9843. (23) Ganguly, R.; Aswal, V. K.; Hassan, P. A.; Gopalkrishnan, I. K.; Yakhmi, J. V. J. Phys. Chem. B 2005, 109, 5653. (24) Malmsten, M.; Linse, P. Macromolecules 1992, 25, 5440.
Kumbhakar and Ganguly (25) Aswal, V. K.; Kohlbrecher, P. S. G.; Bahadur, P. Chem. Phys. Lett. 2000, 349, 458. (26) Su, Y.; Liu, H.; Wang, J. Langmuir 2002, 18, 865. (27) Florin, E.; Ejellander, R.; Eriksson, J. C. J. Chem. Soc., Faraday Trans. 1984, 80, 2889. (28) Bahadur, P.; Li, P.; Almgren, M.; Brown, W. Langmuir 1992, 8, 1903. (29) Desai, P. R.; Jain, N. J.; Sharma, R. K.; Bahadur, P. Colloids Surf., A 2001, 178, 57. (30) Lee, J. C.-M.; Law, R. J.; Discher, D. E. Langmuir 2001, 17, 3592. (31) Dutt, G. B. J. Phys. Chem. B 2005, 109, 4923. (32) Grant, C. D.; Steege, K. E.; Bunagan, M. R.; Castner, E. W., Jr. J. Phys. Chem. B 2005, 109, 22273. (33) Grant, C. D.; DeRitter, M. R.; Steege, K. E.; Fadeeva, T. A.; Castner, E. W., Jr. Langmuir 2005, 21, 1745. (34) Humpolickova, J.; Stepanek, M.; Prochazka, K.; Hof, M. J. Phys. Chem. A 2005, 109, 10803. (35) Sen, P.; Ghosh, S.; Sahu, K.; Mondal, S. K.; Roy, D.; Bhattacharyya, K. J. Chem. Phys. 2006, 124, 204905. (36) Kumbhakar, M.; Goel, T.; Nath, S.; Mukherjee, T.; Pal, H. J. Phys. Chem. B 2006, 110, 25646. (37) Alexandridis, P.; Nivaggioli, T.; Hatton, T. A. Langmuir 1995, 11, 1468. (38) Kabanov, A. V.; Nazarova, I. R.; Astafieva, I. V.; Batrakov, E. V.; Alakhov, V. Y.; Yaroslavov, A. A.; Kabanov, V. A. Macromolecules 1995, 28, 2303. (39) Kumbhakar, M.; Goel, T.; Mukherjee, T.; Pal, H. J. Phys. Chem. B 2004, 108, 19246. (40) Streletzky, K.; Phillies, G. D. J. Langmuir 1995, 11, 42. (41) Phillies, G. J. D.; Yambert, J. E. Langnuir 1996, 12, 3431. (42) Charlton, I. D.; Doherty, A. P. J. Phys. Chem. B 2000, 104, 8327. (43) Molina-Bolivar, J. A.; Aguiar, J.; Ruiz, C. C. J. Phys. Chem. B 2002, 106, 870. (44) Molina-Bolivar, J. A.; Aguiar, J.; Ruiz, C. C. Mol. Phys. 2001, 99, 1729. (45) Kumbhakar, M.; Goel, T.; Mukherjee, T.; Pal, H. J. Phys. Chem. B 2005, 109, 14168. (46) Kumbhakar, M.; Goel, T.; Mukherjee, T.; Pal, H. J. Phys. Chem. B 2005, 109, 18528. (47) Kumbhakar, M.; Mukherjee, T.; Pal, H. Chem. Phys. Lett. 2005, 413, 142. (48) O’Connor, D. V.; Philips, D. Time Correlated Single Photon Counting; Academic Press: New York, 1984. (49) Maroncelli, M.; Fleming, G. R. J. Chem. Phys. 1987, 86, 6221. (50) Quitevis, E. L.; Marcus, A. H.; Fayer, M. D. J. Phys. Chem. 1993, 97, 5762. (51) Maiti, N. C.; Krishna, M. M. G.; Britto, P. J.; Periasamy, N. J. Phys. Chem. B 1997, 101, 11051. (52) Schott, H.; Royce, A. E.; Hans, S. K. J. Colloid Interface Sci. 1984, 98, 196. (53) Aliotta, F.; Fontana, M. P.; Maisano, G.; Migliardo, P.; Wanderlingh, F. Opt. Acta 1980, 27, 931. (54) Dillon, S. R.; Dougherty, R. C. J. Phys. Chem. A 2002, 106, 7647. (55) Brown, J. C.; Pusey, P. N.; Dietz, R. J. J. Chem. Phys. 1975, 62, 1136. (56) Hassan, P. A.; Kulshreshtha, S. K. J. Colloid Interface Sci. 2006, 300, 744. (57) Pedersen, J. S.; Gerstenberg, C. Macromolecules 1996, 29, 1363. (58) Lindner, H.; Scherf, G.; Glatter, O. Phys. ReV. E 2003, 67, 061402. (59) Jain, N. J.; Aswal, V. K.; Goyel, P. S.; Bhadur, P. J. Phys. Chem. B 1998, 102, 8452. (60) Ho, D. L.; Hammouda, B.; Kline, S. J. Polym. Sci., Part B: Polym. Phys. 2003, 41, 135.
