J. Phys. Chem. B 2005, 109, 799-803
799
Aggregation of Perfluoropolyether Carboxylic Salts in Aqueous Solutions. Fluorescence Probe Study Konrad Sulak,† Marian Wolszczak,† Alba Chittofrati,‡ and Ewa Szajdzinska-Pietek*,† Institute of Applied Radiation Chemistry, Technical UniVersity of Lodz, Wroblewskiego 15, 93-590 Lodz, Poland, and Physical Chemistry of Interphases, SolVay Solexis - Research & Technology, Viale Lombardia 20, 20021 Bollate (Milano), Italy ReceiVed: September 16, 2004; In Final Form: October 25, 2004
Aqueous solutions of anionic surfactants Cl(C3F6O)nCF2COOX, consisting of n ) 2 and 3 perfluoroisopropoxy units and the counterion X ) Na+ or NH4+, were studied by the method of fluorescence quenching with the use of (1-pyrenylbutyl)trimethylammonium bromide as a luminophore, and 1,1′-dimethyl-4,4′bipyridinium dichloride (methyl viologen) as a quencher. From the kinetics of fluorescence decay (time-resolved experiments) micellar aggregation numbers, N, and rate constants of the intramicellar quenching were determined for a wide range of surfactant concentrations, on the basis of the model developed by Infelta and Tachiya. The results are discussed in terms of the shape of the aggregates and the degree of counterion binding. The most important conclusions include: (i) a significant increase of N with increasing surfactant concentration suggests that spherical micelles formed at critical micellar concentration (CMC) transform into ellipsoidal aggregates, (ii) the degree of counterion binding to micelles is higher for NH4+ than for Na+, leading to higher N values in the case of the ammonium salt (n ) 2), and (iii) at concentrations close to CMC the longer chain surfactant (n ) 3) forms loose aggregates suggesting significant permeation with water molecules. An additional finding of this study is that the micelle-bound luminophore and quencher can form a ground-state complex, and for this reason the N values cannot be evaluated properly from the steady-state fluorescence intensity data using the equation proposed by Turro and Yekta.
Introduction Perfluorinated amphiphiles, in comparison to their protiated analogues, have received much attention in the literature for their unique hydrophobic and lipophobic character that, in combination with other specific properties, is of interest for actual and potential practical applications,1,2 while supporting an increasing demand for fundamental studies on the selfassembly of fluoro-amphiphilic systems. The aim is to provide valuable knowledge in the design of materials and devices with specialized structures at the nano- or mesoscale for specialized functions where dynamic aspects can play a major role. In such a framework, recently reviewed by Krafft,3 the simple micellar solutions4 and their details on molecular assemblies can be regarded as a first step toward more complex systems. Among fluoroamphiphiles, the functional perfluoropolyethers (PFPE) of the type Cl(C3F6O)nCF2COOX are particularly interesting due to the impact of lateral perfluoromethyl groups on their self-assembly features. Their synthesis and chemical analysis has been described by Tonelli at al.5 In the past, narrow MW distributions of monocarboxylic PFPEs in water had been examined, either with perfluoromethyl or chlorine end groups, combining several techniques to identify micellar and liquid crystalline structures.6-10 Recently, the Tonelli’s synthesis route enabled the preparation of laboratory scale samples of increased purity with respect to the general formula above with the number of isopropoxy units n ) 2 or 3. * Corresponding author. E-mail:
[email protected]; phone: +48.42.6313189; fax: +48.42.6840043. † Technical University of Lodz. ‡ Solvay Solexis - Research & Technology.
Several samples of this type have then been examined in aqueous solution to provide information on the concentration threshold for liquid crystal formation,11 and CMC by surface tension,11 conductivity,12 and NMR.13 The micellar structure of ammonium or potassium n ) 2 salt solutions has been investigated in detail by SANS as a function of concentration and temperature, providing first evidence of transition from spherical to ellipsoidal micellar aggregates (in isotropic solutions well below the onset of detectable liquid crystal appearance) as well as information on the aggregation number and ionization degree.14,15 However, no study has been reported by the fluorescence quenching technique, which enables a dynamic approach to the evaluation of aggregation numbers,16,17 and comparison of the data with those obtained from scattering studies (at least within the constraints posed by the purity of the specific samples). An important issue in the application of fluorescence probes to study fluorinated systems is appropriate choice of the luminophore/quencher pair. The protiated probes commonly used in the studies of hydrocarbon surfactant systems appear incompatible with fluorinated hosts, but this incompatibility may be counterbalanced by positive electrostatic interactions. We have recently proposed to use a cationic luminophore [(1pyrenebutyl)trimethylammonium bromide (PBTMA)] and a cationic quencher [1,1′-dimethyl-4,4′-bipyridinium dichloride (methyl viologen, MV)] to determine aggregation numbers of anionic fluoro-surfactants in aqueous solutions, based on timeresolved fluorescence quenching (TRFQ) measurements.18 In the present work we extend this method for the studies of aqueous micelles of PFPE surfactants of the general formula
10.1021/jp0457949 CCC: $30.25 © 2005 American Chemical Society Published on Web 12/07/2004
800 J. Phys. Chem. B, Vol. 109, No. 2, 2005
Sulak et al.
Cl(C3F6O)nCF2COOX, PFPE-n-X, with the counterion X ) Na+ or NH4+ for n ) 2, and X ) Na+ for n ) 3. The main objective of our study is to determine the aggregation numbers and rate constants of intramicellar fluorescence quenching as a function of the surfactant concentration, and to evaluate the effects of counterion and PFPE chain length. The results will be compared with those obtained recently by SANS.14,15 In addition to TRFQ experiments, we present the steady-state fluorescence quenching results for the PBTMA/ MV probes, which, however, do not allow a proper estimation of aggregation numbers of the examined surfactants. Preliminary data for the PFPE-2-X systems have been presented earlier.19 Materials and Methods The PFPE-n-X surfactants were supplied by Solvay Solexis - Research & Technology, Bollate (Milano), Italy; their purity was comparable to similar samples previously described,11,14,15 typically around 99% and 98% for PFPE-2-X and for PFPE3-X, respectively. Although impurities in traces cannot be fully excluded and may deserve further attention soon, it is worth stressing that the present PFPE-2-Na displayed by surface tension the same cmc value of analogues previously reported within 8% experimental deviation. In view of the large effect of impurities on CMC and surface tension,20 this result was taken as a sufficient indicator of quality for this series of n ) 2 samples. The PFPE-3-Na material might contain traces of n ) 4 analogue, silica/silicates from glassware were detected in amounts of 187 µg/g referred to Si, and residual alkalinity of 55 µg/g was determined by titration. PBTMA (Molecular Probes) and MV (Aldrich) were used as received. Aqueous stock solutions of surfactant/PBTMA and MV were prepared in Millipore deionized water; MV concentration was determined spectrophotometrically using the molar absorption coefficient ) 20 500 dm3 mol-1 cm-1 at λmax ) 258 nm.21 The examined samples were obtained by mixing proper volumes of these stocks and water. All solutions remained clear and isotropic upon long-term equilibration, though it is worth recalling that the concentration limit for solution free of any detectable appearance of liquid crystalline phase at 25 °C is 25 and 45 wt % for PFPE-2-NH4 and PFPE-2-Na, respectively, against 2 wt % for PFPE-3-Na. Concentration of the luminophore was e8 µmol dm-3, ensuring the average probe/micelle ratio e0.01, and that of the quencher was not higher than the micelle concentration. All measurements were done at room temperature (24 ( 1 °C) for samples carefully deaerated by bubbling with argon gas (40 min at least). TRFQ measurements were carried out using the photolysis setup based on a nitrogen laser (pulse duration 300 ps). Steady-state fluorescence spectra were recorded on Aminco-Bowman Series 2 spectrofluorometer, and absorption spectra were recorded on a Varian Cary 5E spectrophotometer. Additional details have been described earlier.22 Results TRFQ Measurements. The decay of excited PBTMA in surfactant solutions in the absence of the quencher was exponential with rate constants k0 ) (6.4 ( 0.2) × 106 s-1 for sodium salts and k0 ) (6.7 ( 0.1) × 106 s-1 for the ammonium salt, independent of surfactant concentration; these values are lower than that determined for homogeneous PBTMA/H2O solution, (7.4 ( 0.1) × 106 s-1,18 indicating that the luminophore is bound to micelles. In the presence of the quencher the kinetics remains firstorder in water but nonexponential decay curves have been
Figure 1. Fluorescence decay monitored at 405 nm for the system PFPE-3-Na/PBMTA/MV (surfactant concentration 10.7 mmol dm-3, quencher concentration 0.22 mmol dm-3). Lower panel: noisy curve, experimental run; smooth curve, the best fit with the use of eq 1, where A2 ) 6.20 × 106 s-1, A3 ) 0.508, and A4 ) 4.67 × 107 s-1.
recorded for surfactant solutions, and the data were analyzed in the frames of the model developed by Infelta and Tachiya.23,24 The model assumes that a micelle contains a maximum of one probe, which does not exit the micelle during the lifetime of excited state, and the occupation of micelles by the solutes (probe and quencher) follows Poisson statistics. Using a nonlinear least-squares procedure the fluorescence decay curves I(t) have been fitted by
I(t) ) A1 exp{-A2t - A3[1 - exp(-A4t)]}
(1)
where A1 ) I(0) is the emission intensity at zero time (the end of laser pulse) and A2, A3, and A4 are parameters; cf. Figure 1 for the example of data fitting. For all examined systems the A2 parameter was independent of quencher concentration and equal to k0, indicating that there was no intermicellar migration of the MV during the lifetime of the excited PBTMA. In this case the model predicts A3 ) [Q]/[M] and A4 ) kqm, where [Q] and [M] are concentrations of the quencher and micelles, respectively, and kqm is the rate constant for the first-order intramicellar quenching process. Consistently, for a given surfactant system the best fitting A4 parameters were independent of the quencher concentration, within uncertainty e10%, and the A3 parameter linearly increased with [Q], as illustrated in Figure 2. Concentration of micelles was determined from the slope of the plot A3 vs [MV] (standard deviation
