Excited-State Proton Transfer in Complexes of Poly(methacrylic acid

Jun 16, 2004 - Department of Chemistry, Wichita State University, 1845 Fairmount,. Wichita, Kansas 67260-0051. Received February 17, 2004. In Final Fo...
0 downloads 0 Views 139KB Size
Download PDF
6158

Langmuir 2004, 20, 6158-6164

Excited-State Proton Transfer in Complexes of Poly(methacrylic acid) with Dodecyltrimethylammonium Chloride Svetlana V. Kombarova and Yuri V. Il′ichev* Department of Chemistry, Wichita State University, 1845 Fairmount, Wichita, Kansas 67260-0051 Received February 17, 2004. In Final Form: May 5, 2004 Proton-transfer reactions in aqueous solutions of poly(methacrylic acid) (PMA) were studied using a fluorescent probe and Fourier transform infrared (FTIR) spectroscopy. Protolytic photodissociation of 1-hydroxypyrene (HP) in water was found to be very slow. The PMA polyanion appeared to be very inefficient as a proton acceptor in the excited-state reaction with HP. However, a drastic increase in the deprotonation efficiency was observed in PMA solutions with the same pH values close to neutral when dodecyltrimethylammonium chloride (DTAC) was added. The protonated form of HP, as well as its anion, was shown to be solubilized in polyion-covered micelles. Time-resolved fluorescence data suggested at least two localization sites with different reactivities toward PMA. FTIR spectroscopy was used to quantify the degree of ionization of PMA in PMA-DTAC mixtures. The IR data indicated that protolytic dissociation of PMA could be well described by the Henderson-Hasselbach equation with an apparent pK of 6.6. In contrast, the fluorescent data revealed cooperative protonation of the PMA groups interacting with HP localized within surfactant assemblies. This selective protonation at a pH close to neutral may be associated with a conformational transition in the polymer-surfactant complex.

Introduction Interactions of polymers with surfactants have received a great deal of attention because of their fundamental importance and their numerous technological applications.1-4 Complexation of polyions and oppositely charged surfactants is cooperative and thermodynamically favorable, as can be judged by the critical aggregation concentration (CAC), which is typically much smaller than the critical micelle concentration (CMC), the concentration characterizing aggregation in the absence of polymers.1-3 The importance of electrostatic interactions in mixtures of surfactants and polyelectrolytes of opposite charge is evident from the strong dependence of the CAC on the concentration of polymer and other electrolytes, surfactant charge, and polyion charge density.5,6 The effects of polymer charge appear to be particularly complex. Although a concept of more favorable interactions at higher linear charge densities is theoretically sound and is supported by experimental data,7,8 some systems clearly show stronger interactions at lower charge densities.9-13 * To whom correspondence should be addressed. Current address: Department of Chemistry, Wichita State University, 317 McKinley Hall, 1845 Fairmount, Wichita, Kansas 67260-0051. E-mail: [email protected]. (1) Interactions of Surfactants with Polymers and Proteins; Goddard, E. D., Ananthapadmanabhan, K. P., Eds.; CRC Press: Boca Raton, FL, 1993; p 427. (2) Polymer-Surfactant Systems; Kwak, J. C. T., Ed.; Marcel Dekker: New York, 1998; p 482. (3) Goddard, E. D. J. Colloid Interface Sci. 2002, 256, 228-235 and references therein. (4) Goddard, E. D. Cosmet. Sci. Technol. Ser. 1999, 22, 181-215. (5) Konop, A. J.; Colby, R. H. Langmuir 1999, 15, 58-65. (6) Hansson, P. Langmuir 2001, 17, 4167-4180. (7) Hansson, P.; Almgren, M. J. Phys. Chem. 1996, 100, 9038-9046. (8) Satake, I.; Takahashi, T.; Hayakawa, K.; Maeda, T.; Aoyagi, M. Bull. Chem. Soc. Jpn. 1990, 63, 926-928. (9) (a) Hayakawa, K.; Santerre, J. P.; Kwak, J. C. T. Macromolecules 1983, 16, 1642-1645. (b) Shimizu, T.; Seki, M.; Kwak, J. C. T. Colloids Surf. 1986, 20, 289-301. (10) Chandar, P.; Somasundaran, P.; Turro, N. J. Macromolecules 1988, 21, 950-953.

An increase in the CAC with the degree of ionization has been observed for poly(acrylic acid) (PAA) and poly(methacrylic acid) (PMA).10,11 Molecular mechanisms implicated in these phenomena are far from being completely understood. Little is known about the dynamic properties of surfactant aggregates stabilized by weak polyelectrolytes with different degrees of ionization. Spectroscopic probes have found wide application in research focused on structural and dynamic aspects of microheterogeneous systems.14-16 Fluorescent probes undergoing excited-state proton transfer have been successfully used in studies of various surfactant assemblies17-25 and polymers.26-30 In this study, we demon(11) (a) Kiefer, J. J.; Somasundaran, P.; Ananthapadmanabhan, K. P. Stud. Polymer Sci. 1992, 11, 423-444. (b) Kiefer, J. J.; Somasundaran, P.; Ananthapadmanabhan, K. P. Langmuir 1993, 9, 1187-1192. (12) Katsuura, H.; Kawamura, H.; Manabe, M.; Kawasaki, H.; Maeda, H. Colloid Polym. Sci. 2002, 280, 30-37. (13) Proietti, N.; Amato, M. E.; Masci, G.; Segre, A. L. Macromolecules 2002, 35, 4365-4372. (14) Kalyanasundaram, K. Photochemistry in Microheterogeneous Systems; Academic: Orlando, FL, 1987; p 388. (15) (a) Winnik, F. M. Chem. Rev. 1993, 93, 587-614. (b) Winnik, F. M.; Regismond, S. T. A. Surfactant Sci. Ser. 1998, 77, 267-315. (16) Zana, R. In Amphiphilic Block Copolymers; Alexandridis, P., Lindman, B., Eds.; Elsevier: Oxford, U.K., 2000; pp 221-252. (17) Kuzmin, M. G. NATO ASI Ser., Ser. C 1992, 371, 279-294. (18) (a) Zaitsev, A. K.; Il’ichev, Yu. V.; Zaitsev, N. K.; Kuzmin, M. G. Proc. Acad. Sci. USSR, Phys. Chem. Sect. 1985, 283, 745-748. (b) Il’ichev, Yu. V.; Zaitsev, A. K.; Kuzmin, M. G. High Energy Chem. 1990, 24, 117-121. (19) Zaitsev, A. K.; Il’ichev, Yu. V.; Gorelik, O. F.; Zaitsev, N. K.; Kuzmin, M. G. Sov. J. Chem. Phys. 1989, 4, 2281-2291. (20) Il’ichev, Yu. V.; Demyashkevich, A. B.; Kuzmin, M. G. J. Phys. Chem. 1991, 95, 3438-3444. (b) Il’ichev, Yu. V.; Demyashkevich, A. B.; Kuzmin, M. G.; Lemmetyinen, H. J. Photochem. Photobiol., A 1993, 74, 51-63. (21) Il’ichev, Yu. V.; Shapovalov, V. L. Bull. Acad. Sci. USSR, Div. Chem. Sci. 1992, 41, 1762-1767. (22) Escabi-Perez, J. R.; Fendler, J. H. J. Am. Chem. Soc. 1978, 100, 2234-2236. (b) Politi, M. J.; Fendler, J. H. J. Am. Chem. Soc. 1984, 106, 265-273. (c) Politi, M. J.; Brandt, O.; Fendler, J. H. J. Phys. Chem. 1985, 89, 2345-2354. (23) Cohen, B.; Huppert, D.; Solntsev, K. M.; Tsfadia, Y.; Nachliel, E.; Gutman, M. J. Am. Chem. Soc. 2002, 124, 7539-7547.

10.1021/la049585r CCC: $27.50 © 2004 American Chemical Society Published on Web 06/16/2004

Excited-State Proton Transfer in Complexes of PMA

Langmuir, Vol. 20, No. 15, 2004 6159

strated how a fluorescent photoacid can be utilized to elucidate mechanisms of protolytic reactions and conformational changes in mixtures of weak polyelectrolytes and surfactants. Experimental Section An aqueous solution of sodium poly(methacrylate) (Mw ∼ 9500 g/mol, Mw/Mn ) 1.76) was purchased from Aldrich. Dodecyltrimethylammonium chloride (DTAC) (Fluka, >99%) and 1-hydroxypyrene (HP) (Aldrich, 98%) were used as received. An aqueous solution of HP with a concentration of ∼5 µM was prepared as follows: an exact quantity of HP was dissolved in methanol (Fisher), solvent was evaporated under a mild flow of argon, a required amount of HPLC water (Fisher) was added, and the solution was filtered through a Millipore Millex filter (2 µm pore size). PMA-DTAC solutions with different pHs were prepared by diluting sodium poly(methacrylate) stock solution (11 g/L, pH ∼ 8) with HP solution and adding a calculated volume of 0.3 M DTAC solution. pH was varied by adding 1 N NaOH (Fisher), 1 N HCl (Fisher), or water to a total volume of 3.5 mL. A Φ-340 pH meter (Beckman Coulter) equipped with a Futura microcombination electrode and an automatic temperature compensation (ATC) probe was used to measure the solution pH. All the measurements were performed with aerated solutions at 23 ( 1 °C. IR absorption spectra were recorded with a Nexus 870 Fourier transform infrared (FTIR) spectrometer (Thermo Nicolet) equipped with a deuterated triglycine sulfate (DTGS) detector and a KBr beam splitter. The instrument was continuously purged with dry air. Samples were placed into a demountable cell with CaF2 windows and a 25 µm Teflon spacer. Typically, 128 interferograms were accumulated at a spectral resolution of 4 cm-1 and Happ-Genzel apodization was applied. Water absorption was subtracted from all spectra. Ultraviolet-visible (UV-vis) absorption and fluorescence spectra were recorded with a Hitachi U-2010 spectrophotometer and a Jobin-Yvon-Spex Fluorolog Tau-3 lifetime system, respectively. Emission spectra were corrected to account for the wavelength-dependent sensitivity of the photodetection components. A correction file was generated from the spectra of a set of secondary emission standards.31 Fluorescence quantum yields were measured relative to quinine sulfate (Fluka) in 0.1 N H2SO4 (φ ) 0.546).32 Fluorescence response functions were collected by using the timecorrelated single-photon counting technique. A femtosecond Tisapphire oscillator (home-built from a Concept Design Production (CDP) laser kit) was pumped by a Millennia 6-J solid-state NdYVO4 laser (Spectra Physics). The oscillator output (∼500 mW, 83 MHz, ∼50 fs) was directed to a second harmonic generator (CDP) which generated light pulses (∼375 nm) used for sample excitation. The laser power on the sample ( 7.5. Fluorescence excitation spectra recorded at 480 nm gave a pKa estimate of 9.3. In the pH range from 4.1 to 5.9, phase separation was observed and no fluorescent data were collected. The white precipitate completely dissolved at pH e 4, and the intensity of UV light scattered by acidic solutions was found to be similar to that at pH > 8. The plot of the (φ0φ′)/(φφ′0) ratio versus pH was analyzed within a simple model described by Scheme 2. Effective concentrations in eq 2 can be evaluated by using an empirical equation for the degree of ionization (R):

R ) [RCOO-]/C0 ) 1/(1 + 10n(pK-pH))

(3)

Here, C0 ) [RCOO-] + [RCOOH] is the total concentration of the PMA repeating units, pK is a parameter analogous to the pKa of a monoprotic acid, and n can be interpreted as an apparent number of protons released upon deprotonation. Plots of R versus pH in polyacid solutions are often described by eq 3 with n < 1.52 After substituting eq 3 into eq 2, we obtain

φ0φ′ 1 ) φφ′0 A + B‚10n(pK-pH)

(4)

where A ) (k2τ0C0)-1 and B ) (1 + k-2τ′0C0)/(k2τ0C0). Fitting eq 4 to the data shown in Figure 3b yielded pK ) 6.52 and n ) 3.9. The n value suggests a cooperative transition takes place in the PMA-DTAC solution at pH ∼ 6.5. The apparent rate constants obtained from the fitting (k2 ) 7.1 × 109 M-1 s-1 and k-2 ) 4.9 × 109 M-1 s-1) include activity coefficients and reflect HP reactivity averaged over all possible localization sites. The k2 value practically coincides with that obtained for the HP reaction with (CH3)3CCOO- in CTAB micelles (7.0 × 109 M-1 s-1).18b In contrast, the k-2 value for the monomeric acid was by a factor of 56 smaller than that obtained for PMA. It is worth mentioning that the experimental data show a systematic, albeit small, deviation from the theoretical curve at pH < 6.3. The (φ0φ′)/(φφ′0) ratio remains much larger than the value expected from the fit and also than that observed in the absence of DTAC (see Figures 2b and 3b). FTIR spectroscopy was utilized to characterize the protolytic equilibria of PMA in the PMA-DTAC mixture. Figure 4a presents difference IR absorption spectra of 100 mM PMA solutions with different pHs. All solutions contain 10 mM DTAC, which does not contribute significantly to IR absorption in the spectral region shown. We used two distinct bands to quantify the degree of ionization. The strong band at 1543 cm-1 originates from asymmetric stretching vibration of COO-. The 1188 cm-1 band is (51) The fluorescence quantum yield ratio was calculated as described for bulk water. We used the φ0/φ′0 ratio of 0.71 obtained from the HP spectra measured at pH 1.7 and 12. (52) Oosawa, F. Polyelectrolytes; Marcel Decker: New York, 1971; p 75.

Figure 4. (a) Stacked difference IR absorption spectra of a PMA-DTAC mixture (100 mM PMA and 10 mM DTAC) in H2O at different pHs. The vertical dotted lines show two wavenumbers that were used to calculate the degree of ionization of (R) PMA. (b) Plot of R vs solution pH. R was calculated from the absorbance at 1188 cm-1 (open symbols) and 1543 cm-1 (filled symbols). The solid line corresponds to the best fit of the 1188 cm-1 data to eq 3. The dashed line shows R calculated from the parameters obtained by fitting the fluorescence data (see the text).

clearly associated with the COOH group of the protonated form. The pH profiles calculated from absorbance changes at these two wavenumbers coincide within the experimental accuracy (Figure 4b). Surprisingly, these profiles are well described by the Henderson-Hasselbach equation (eq 3 with n)1) with an apparent pK of 6.58. When n was used as an adjustable parameter, we obtained n values of 0.99 and 0.83 by fitting the data at 1188 cm-1 and 1543 cm-1, respectively. The dashed line in Figure 3b shows the pH profile which was calculated by using pK and n values from the fluorescence data. The IR data demonstrate that strong suppression of the excited-state proton transfer observed at pH < 6.5 is not caused by a dramatic change in the overall concentration of the proton acceptor. The fluorescence response functions of HP collected in PMA-DTAC solution at pH 7.13 are depicted in Figure 5. The results of global analysis of the kinetic curves measured at four different wavelengths are also presented. Three exponential terms are required to fit the data. The longest time found (3.5 ns) was slightly larger than the lifetime of the HP anion (τ′0 ) 3.0 ns); two other times (0.2 and 1.5 ns) were found that were significantly shorter. The relative amplitude of the shortest component remains approximately constant within the anion emission band. In contrast, the relative contribution of the intermediate component increases with the wavelength. The amplitude sum is close to zero at the red side of the emission spectrum. The steady-state fluorescent data (see Figure 3b) suggest an essentially irreversible reaction at pH ∼ 7 and, therefore, monoexponential decay for *ArOH and biexponential kinetics for *ArO-. We believe that the large

Excited-State Proton Transfer in Complexes of PMA

Figure 5. Fluorescence response functions of HP in a PMADTAC mixture (100 mM PMA and 10 mM DTAC, pH 7.13) collected at 409 nm (1), 440 nm (2), 480 nm (3), and 500 nm (4). The results of global fitting to a three-exponential function are shown as solid lines. The parameters obtained, the reduced χ2, and the standard deviations are also presented.

amplitude of the 3.5 ns component in the fluorescence decay measured at 409 nm is mainly due to a strong overlap between ArOH and ArO- fluorescence spectra (the anion emission contributes ∼20% to the total fluorescence intensity at 409 nm; see Figure 3a). The decay times 0.2 and 1.5 ns can be attributed to the presence of two subpopulations of HP bound to PMA-DTA aggregates. When we used an average of these two times and τ0 ) 15.3 ns, we obtained k2 ) (1/τ - 1/τ0)/C0 ) 6.4 × 109 M-1s-1, which is in good agreement with the steady-state data. Surfactant aggregates stabilized by polyions are generally believed to be similar to micelles formed in the presence of small counterions.1-3 However, the results of this study indicate that significant differences may be anticipated when the kinetics of the reactions involving counterions is analyzed. For complexes of oppositely charges polymers and surfactants, one should expect a radial distribution of polyion charges and the existence of micellar regions that are not covered by the polymer.1,7 Therefore, different distances between the OH group of HP solubilized in polymer-coated micelles and COOgroups are anticipated. A bimodal distribution of the distances between two reactants may result in kinetic nonequivalence, as observed for HP, if exchange between different sites is slow in comparison to the proton-transfer reactions. Considering that the HP lifetime is not very sensitive to microenvironment, the two subpopulations have to be associated with large differences in the local concentration of COO- groups, in their reactivity toward *ArOH, or in both. Interestingly, deviations from the twostate behavior and kinetic nonequivalence have never been observed for reactions of HP and other hydroxyaromatics with monomeric carboxylate anions in cationic micelles.18 An aggregation number (N) in the range 40-50 has been reported for DTAC micelles.47 An N value of 100 has been obtained for DTAB in the presence of PMA.45 If we

Langmuir, Vol. 20, No. 15, 2004 6163

assume that aggregates of similar size are formed in the PMA-DTAC mixture studied, we conclude that ∼90% of fully ionized PMA is present as free polyion. A possible explanation for our fluorescent and IR data is based on pH-dependent localization of HP either in polyelectrolytecoated micelles or in hydrophobic pockets of the contracted form of PMA, where no excited-state deprotonation occurs. Previous studies40,41 utilizing pyrene derivatives as fluorescent probes demonstrated that there is no change in the probe microenvironment at pH g 6 and only a small variation is seen at lower pH in a PMA-DTAB mixture of similar composition.45 This suggests that HP transfer from DTA aggregates to the PMA coil can be driven mainly by a concentration factor (ratio of volume fractions of the compact-coil form and micelles). However, this factor can hardly be significant at pH ∼ 6.5. Recall that PMA itself shows a conformational change to the compact-coil form at pH ∼ 5.5. Another explanation assuming that HP fluorescence reflects conformational changes in polymer-surfactant aggregates seems to be more feasible. Excited-state proton transfer in PMA-DTA complexes likely involves only the groups that are in close proximity to the proton donor, that is, those of the polymer chain wrapping the micelle. Selective protonation of such carboxylates seems to be highly unfavorable under conditions when almost half of the acidic sites are ionized (pH ∼ 6.5). However, the polymer chain may form protruding loops to compensate for differences in the distance between adjacent charges on the PMA chain and those on the micellar surface.6 Despite their distant location, the COO- groups in these loops may be more reactive because they are easily protonated; that is, they have a higher apparent pK than those forming ion pairs with the surfactant. Our data suggest selective protonation of the COO- groups in the polymer loops triggers a conformational transition which results in a substantial increase in the local concentration of the COOH groups “seen” by HP. Taken together, these changes in the concentration of both the proton acceptor and donor cause the excited-state proton transfer to be strongly suppressed within a very narrow pH range. Summary Protolytic photodissociation of 1-hydroxypyrene (HP) in bulk water was found to be very slow. The small rate constant for the excited-state proton transfer is consistent with the reaction energetics (pKa* ∼ 4) and the nature of the lowest singlet excited state of HP. Simple photophysics and the very slow reaction with water make HP an ideal fluorescent probe for studying bimolecular proton-transfer reactions in complex environments. The polyanion of poly(methacrylic acid) (PMA) was found to be very inefficient as the proton acceptor for the excited HP. The excitedstate proton-transfer reaction between HP and PMA was significantly accelerated in the presence of a cationic surfactant but only when polymer-surfactant aggregates were formed ([DTAC] > CAC) and HP was solubilized in these aggregates. Time-resolved fluorescence data indicated that the kinetics of proton-transfer reactions in cationic surfactant assemblies stabilized by a reactive polyanion differs markedly from that observed in the presence of small counterions. FTIR data demonstrated that protolytic dissociation of PMA in the presence of DTAC can be well described by the Henderson-Hasselbach equation with an effective pK of 6.6, and therefore, no dramatic change in the overall concentration of the COO- groups should be observed in a narrow pH range. In contrast, the fluorescent data for

6164

Langmuir, Vol. 20, No. 15, 2004

Kombarova and Il’ichev

HP localized within PMA-DTA complexes revealed cooperative protonation of PMA. This was attributed to selective protonation of the COO- groups interacting with the photoacid. The pH-dependent transition might also be associated with a conformational change in the polymer-surfactant complex.

1/τ1 + 1/τ2 ) µ′ + µ ) 1/τ0 + k1 + 1/τ′0 + k-1[H3O+] (A7)

Acknowledgment. This work was supported by Wichita State University.

At pH ∼ 5-8, all bimolecular processes involving hydronium ions are too slow to compete with any unimolecular processes (k1, 1/τ0, and 1/τ′0). Under these conditions, the ArOH fluorescence decay becomes single exponential (A f 0) with a lifetime of τN:

Appendix

1/τ1 ) µ ) 1/τN ) 1/τ0 + k1

According to Scheme 1, the fluorescence response functions for *ArOH (I) and *ArO- (I ′) are described by two-exponential functions:

I(t) ) I0[exp(-t/τ1) + A exp(-t/τ2)]

(A1)

I ′(t) ) I ′0[exp(-t/τ2) - exp(-t/τ1)]

(A2)

The decay times (τ1 and τ2) and relative amplitude (A) can be written as follows:

1/τ1,2 ) (µ + µ′)/2 ( [(µ - µ′)2/4 + k1k-1[H3O+]]1/2 (A3) (A4) A ) (1/τ1 - µ)/(µ - 1/τ2) where

µ ) 1/τ0 + k1 ) (1/τ1 + A/τ2)/(1 + A) +

µ′ ) 1/τ0 + k-1[H3O ] ) 1/τ1 + 1/τ2 - µ Combining eqs A5 and A6, we obtain

(A8)

The fluorescence kinetics of anionic species remains two-exponential with a rise time equal to the decay time of ArOH (τ1 ) τN) and a decay time equal to the lifetime of the anion (τ2 ) τ′0). This behavior is experimentally observed for many aromatic hydroxycompounds,20 which are typically characterized by τ0 < τ′0 and, therefore, 1/τ1 > 1/τ2. However, it has to be pointed out that I ′0 in eq A2 is proportional to 1/(1/τ1 - 1/τ2), as expected for the standard scheme of two sequential irreversible reactions. If τ0 > τ′0 and k1 is relatively small, such as in the case of HP, 1/τ1 will be smaller than 1/τ2. In this case, formation of the excited anion is slower than its decay and the rise time in the anion fluorescence kinetics is equal to the lifetime of the anion under conditions of direct excitation. Keeping I ′0 in eq A2 positive, we can write the following for the anion fluorescence at a pH close to neutral:

(A5) (A6)

I ′(t) ) I ′0[exp(-t/τN) - exp(-t/τ′0)] LA049585R

(A9)