and pH-Sensitive Structural Changes of the Poly(acrylic acid)

Department of Chemical Engineering, National Taiwan UniVersity of Science and Technology, ... ReceiVed January 24, 2006. In Final Form: May 18, 2006...
0 downloads 0 Views 286KB Size
Download PDF
6764

Langmuir 2006, 22, 6764-6770

Articles Effects of SDS on the Thermo- and pH-Sensitive Structural Changes of the Poly(acrylic acid)-Based Copolymer Containing Both Poly(N-isopropylacrylamide) and Monomethoxy Poly(ethylene glycol) Grafts in Water Yuan-Hung Hsu,† Wen-Hsuan Chiang,† Mu-Chin Chen,† Chorng-Shyan Chern,‡ and Hsin-Cheng Chiu*,† Department of Chemical Engineering, National Chung Hsing UniVersity, Taichung 402, Taiwan, and Department of Chemical Engineering, National Taiwan UniVersity of Science and Technology, Taipei 106, Taiwan ReceiVed January 24, 2006. In Final Form: May 18, 2006 The effects of SDS on the structural changes of the thermally induced polymeric micelles from a graft copolymer comprising poly(acrylic acid) (PAAc) as the backbone and poly(N-isopropylacrylamide) (PNIPAAm) and monomethoxy poly(ethylene glycol) (mPEG) as the grafts in aqueous solution are studied. At low temperature, SDS micelles form via the hydrophobic association of SDS molecules with the PNIPAAm grafts at a critical aggregation concentration of SDS (cacSDS) much lower than its critical micelle concentration. Consequently, the critical aggregation temperature of the graft copolymer is elevated. The corresponding structure of the thermally induced polymeric micelles is characterized by an abrupt reduction in the particle size and an increased tendency toward formation of the monocore structure with a more compact and hydrophobic PNIPAAm microdomain being developed. On the other hand, upon the polymeric micelle formation at high temperature, the copolymer-bound SDS micelle structure is disrupted and the dissociated SDS molecules migrate to the core-shell interface with their alkyl chains residing in the liquidlike region of the hydrophobic PNIPAAm microdomain. The correlation between the polymeric particles and copolymer-bound micelles is further substantiated by showing the change of the colloidal particle size in response to changes in cacSDS via adjusting the pH of the aqueous copolymer/SDS solutions.

Introduction Cooperative hydrophobic association of water-soluble polymer chains with ionic surfactant molecules and subsequent formation of micelle-like aggregates occurring at a critical aggregation concentration (cac) much lower than the critical micelle concentration (cmc) of the regular surfactant solution have been extensively studied.1-15 Attention has also been paid to the †

National Chung Hsing University. National Taiwan University of Science and Technology. * To whom correspondence should be addressed. Fax: 886-422854734. Tel: 886-422852636. E-mail: [email protected] (H.-C. Chiu). ‡

(1) Jonsson, B.; Lindman, B.; Holmberg, K.; Kronberg, B. Surfactants and Polymers in Aqueous Solution; John Wiley & Sons: New York, 1998. (2) Vangeyte, P.; Leyh, B.; Auvray, L.; Grandjean, J.; Misselyn-Bauduin, A.-M.; Jerome, R. Langmuir 2004, 20, 9019. (3) Castro, E.; Tobaoda, P.; Barbosa, S.; Mosquera, V. Biomacromolecules 2005, 6, 1438. (4) Brostein, L. M.; Chernyshov, D. M.; Vorontsov, E.; Timofeeva, G. I.; Dubrovina, L. V.; Veletsky, P. M.; Kazakov, S.; Khokhlov, A. R. J. Phys. Chem. B 2001, 105, 9077. (5) Garret-Flaudy, F.; Freitag, R. Langmuir 2001, 17, 4711. (6) Wesley, R. D.; Cosgrove, T.; Thompson, L.; Armes, S. P.; Baines, F. L. Langmuir 2002, 18, 5704. (7) Bastiat, G.; Grassl, B.; Khoukh, A.; Francois, J. Langmuir 2004, 20, 5759. (8) Dai, S.; Tam, K. C. Langmuir 2004, 20, 2177. (9) Qiu, Q.; Somasundaran, P.; Pethica, B. A. Langmuir 2002, 18, 3482. (10) Makhaeva, E. E.; Tenhu, H.; Khokhlov, A. R. Macromolecules 1998, 31, 1, 6112. (11) Gargallo, L.; Dadic, D.; Martinez-Pina, F. Eur. Polym. J. 1997, 33, 1767. (12) Ricka, J.; Meewes, M.; Nyffenegger, R.; Binkert, T. Phys. ReView letter 1990, 65, 657. (13) Meewes, M.; Ricka, J.; de Silva, M.; Nyffenegger, R.; Binkert, T. Macromolecules 1991, 24, 5811.

interactions of surfactants with thermoresponsive polymers, particularly in studying the structure of the polymer-induced surfactant micelles and/or the coil-to-globule transition of the polymers in aqueous solutions upon heating.9-20 For example, formation of sodium n-dodecyl sulfate (SDS) micelles can be induced by poly(N-isopropylacrylamide) (PNIPAAm) in aqueous phase via hydrophobic association at an SDS concentration much lower than its cmc, leading to the increased water solubility and the elevated phase transition temperature of the polymer.14-20 Two different structures of PNIPAAm-induced SDS micelles have been proposed. First, extensive incorporation of polymer chain segments into the hydrophobic cores of SDS micelles leads to formation of pearl-necklace-like polymer-SDS complexes.16-18 Probably because of a sufficiently high molecular weight of PNIPAAm, the second type of SDS aggregates can be induced by being wrapped in the polymer loopy configuration with a similar interior structure to the regular SDS micelles.16 However, only few studies were concerned with the interaction between SDS and the PNIPAAm-based block or graft co(14) Schild, H.; Tirrell, D. Langmuir 1991, 7, 665. (15) Jean, B.; Lee, L. T.; Cabane, B. Langmuir 1999, 15, 7585. (16) Walter, R.; Ricka, J.; Quellet, C.; Nyffenegger, R.; Binkert, T. Macromolecules 1996, 29, 4019. (17) Kokufuta, E.; Zhang, Y.-Q.; Tanaka, T.; Mamada, A. Macromolecules 1993, 26, 1053. (18) Zhang, Y.-Q.; Tanaka, T.; Shibayama, M. Nature 1992, 360, 142. (19) Wu, C.; Zhou, S. J. Polym. Sci. Part B: Polym. Phy. 1996, 34, 1597. (20) Zhu, P. W.; Napper, D. H. Langmuir 1996, 12, 5992.

10.1021/la060229d CCC: $33.50 © 2006 American Chemical Society Published on Web 06/30/2006

Effects of SDS on the Structure of Copolymers

Langmuir, Vol. 22, No. 16, 2006 6765

Figure 1. Chemical structures of the PAAc/PNIPAAm/mPEG graft copolymer and SDS and their 1H NMR spectrum in D2O at 25 °C (the copolymer concentration was 10.0 mg/mL and that of SDS was 0.50 mg/mL). The assignments of the feature signal protons of the graft copolymer, SDS, and DMF as an external reference are also included.

polymers.21-24 Durand et al.24 studied the interaction between SDS and the PNIPAAm grafts attached to the poly(acrylic acid) (PAAc) backbone with the emphasis on the response of the thermally induced colloidal system comprising the hydrophobic PNIPAAm core structure to the addition of SDS. The hydrophobic glassy (solidlike) cores are gradually dissociated when the SDS concentration increases. However, the formation of the copolymerinduced SDS micelles was not described in this pioneering work. In our previous study,25 the PNIPAAm and PEG grafted copolymers at a concentration of 10.0 mg/mL with the backbone AAc residues being significantly ionized were shown to undergo self-assembly into large aggregates (> 600 nm) upon phase separation by slowly heating to the temperature range 32-35 °C. This has been attributed to the formation of the multicore structure and the intercore connections via the intramolecular bridging between the PNIPAAm and PEG grafts. It was shown that the solidlike hydrophobic inner core of PNIPAAm is surrounded by a significantly hydrated and liquidlike interfacial layer of the exceedingly interactive PNIPAAm and PEG grafts. The objective of this study is to investigate the effect of SDS on the morphological structure and hydrodynamic size of the thermally induced polymeric micelles. First, the formation of the graft copolymer-bound SDS micelles below the critical aggregation temperature of the copolymer (catpolymer) was characterized by the fluorescence study using pyrene as a polarity probe. The structural relation between the thermally induced polymeric aggregates and the copolymer-bound SDS micelles was then established by the variable temperature 1H NMR and (21) Huang, J.; Wu, X.-Y. J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 2667. (22) Mylonas, Y.; Staikos, G. Langmuir 2001, 17, 3586. (23) Durand, A.; Hourdet, D. Polymer 2000, 41, 545. (24) Durand, A.; Hourdet, D.; Lafuma, F. J. Phys. Chem. B 2000, 104, 9371. (25) Hsu, Y.-H.; Chiang, W.-H.; Chen, C.-H.; Chern, C.-S.; Chiu, H.-C. Macromolecules 2005, 38, 9757.

spin-lattice relaxation time (T1) measurements of the feature protons of SDS molecules and the PNIPAAm and PEG grafts of the copolymer. Experimental Section Materials. Synthesis of the copolymer comprising PAAc as the backbone and PNIPAAm and mPEG as the grafts and its composition characterization were described in our previous work.25 SDS (Aldrich) with a purity greater than 99% was used as received. The chemical structures of the graft copolymer and SDS are illustrated in Figure 1. The composition of the graft copolymer is 1.0/18.1/13.3 in mole ratio of PAAc (Mw 20 000 g/mol)/PNIPAAm (3800)/mPEG (2000) and the corresponding weight-average molecular weight is 115 000 g/mol. Deionized water was produced from the Milli-Q Synthesis system (Millipore). Deuterium solvents for the liquid 1H NMR measurements were obtained from Cambridge Isotope. N,NDimethylformamide (DMF) as an external standard in 1H NMR measurements was obtained from Tedia and distilled under vacuum before use. Sample Preparation and Heating Approach. The graft copolymer concentration of 10.0 mg/mL in water was used throughout this study. The aqueous copolymer solutions in the presence of SDS of varying concentrations were prepared by dissolving the copolymer in aqueous SDS solutions at 4 °C with stirring for 12 h. The final pH of the aqueous polymer/SDS solutions was found to be ca. 4.3, irrespective of the SDS concentration studied. The target pH was adjusted by 0.01 N HCl or NaOH if necessary. The catpolymer in the absence and presence of SDS was determined by optical transmittance measurements at 550 nm. The results are shown as Supporting Information. Pyrene was selected as a polarity probe in the fluorescence measurements. Aliqouts (40.0 µL) of the pyrene stock solution (3.0 × 10-5 M) in acetone were evaporated in vials, and the aqueous copolymer (10 mg/mL)/SDS (at the preset concentrations in the range 0.00-1.00 mg/mL) solutions (2.0 mL) were then added, yielding the aqueous solutions with a constant pyrene concentration of ca. 6.0 × 10-7 M.26 Samples for 1H NMR measurements were

6766 Langmuir, Vol. 22, No. 16, 2006 prepared in D2O. The temperature of the polymer/SDS solutions was increased from 20 to 60 °C at a speed of 5 °C/min, but with equilibration at each preset temperature for 30 min. This heating profile has been referred to as fast heating in the literature.27-29 DLS Measurements. The hydrodynamic particle diameters (Dh) of the aqueous polymer/SDS solutions were determined by a Brookhaven 90 plus particle size analyzer (He-Ne laser 15 mW, λ ) 678 nm) at a fixed angle of 90°. The number of accumulation times was set at 50 and the data reported hereinafter represent an average of at least three measurements. Fluorescence Measurements. Fluorescence characterization was performed by measuring the fluorescence intensity ratios (I3/I1) of the third vibronic band at ca. 385.5 nm to the first at 373.5 nm of the emission spectra of pyrene in the aqueous polymer/SDS solutions at each preset temperature.30 The excitation was performed at 336 nm and the emission was recorded in the range from 350 to 500 nm on a Hitachi F-2500 fluorescence spectrometer equipped with a thermostat cell unit. The fluorescence emission spectra are illustrated in part as Supporting Information. Variable-Temperature NMR and Spin-Lattice Relaxation Time Measurements. 1H NMR spectra of the graft copolymer/ SDS/D2O system at different temperatures were obtained by a Varian Unity Inova-600 at 600 MHz in the absence of sample spinning. The pulse width of 4.9 µs with a relaxation delay of 2.0 s was utilized. DMF in a sealed capillary, an external reference in the determination of the copolymer composition in CDCl3, was repeatedly used by placing it coaxially in the sample tube. The proton signal from the formyl group of DMF at δ 8.14 ppm was selected as the resonance reference for assigning the feature signal positions and evaluating the signal integrals in comparison with those obtained in CDCl3 at ambient temperature. The spin-lattice relaxation times (T1) of the feature protons of the graft copolymer and SDS were evaluated at 600 MHz, using the standard inversion-recovery pulse sequence (180°-τ-90°). In each measurement, 13-15 variable delays were employed, and the waiting period was adjusted to be at least 5 times larger than the expected T1 value. T1 was thus calculated on the basis of a nonlinear least-squares fitting routine of the spectrometer.

Results and Discussion I. Polymer-Induced SDS Micelles and Their Effects on Polymeric Particle Size. The interaction between the graft copolymer and various levels of SDS in the aqueous solutions was first explored by the fluorescence intensity ratio of pyrene in response to the polarity of its residing microenvironment from 20 to 60 °C (Figure 2). In the absence of SDS, a significant increase in the I3/I1 values is observed in the temperature range consistent with its catpolymer (32-35 °C) determined by the optical transmittance measurements (Supporting Information). This is attributed to the thermally induced hydrophobic association of the PNIPAAm grafts in the copolymer and formation of polymeric aggregates, leading to the increased hydrophobicity of the microenvironments, in which most pyrene molecules reside, and thus the increased I3/I1 values. With respect to the experiments in the absence of SDS, the I3/I1 value is little influenced by SDS when the SDS concentration is lower than 0.30 mg/mL (Figure 2). On the other hand, at low temperatures and above 0.30 mg/ mL of SDS, the I3/I1 value increases considerably with increasing SDS concentration. As a consequence, with the temperature being increased from 25 to 60 °C, the abrupt increase of I3/I1, which signifies the phase transition of the graft copolymer at catpolymer, disappears. At 25 °C and above ca. 0.3 mg/mL of SDS, a rapid increase in I3/I1 with the SDS concentration occurs (Supporting (26) Schild, H. G.; Tirrell, D. A. Langmuir 1991, 7, 1319. (27) Virtanen, J.; Lemmetyinen, H.; Tenhu, H. Polymer 2001, 42, 9487. (28) Neradovic, D.; Soga, O.; Van Nostrum, C. F.; Hennink, W. E. Biomaterials 2004, 25, 2409. (29) Qiu, X.; Wu, C. Macromolecules 1997, 30, 7921. (30) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039.

Hsu et al.

Figure 2. Effect of temperature on the fluorescence intensity ratio (I3/I1) of pyrene in the aqueous copolymer/SDS solutions with the SDS concentrations of (O) 0.00, (9) 0.05, (2) 0.25, (3) 0.30, ([) 0.35, (]) 0.40, (f) 0.50, and (g) 1.00 mg/mL. SDS was dissolved in water at 4 °C prior to the addition of the graft copolymer (10.0 mg/mL). The pyrene concentration in the solutions was 6.0 × 10-7 M. The error bar at each point represents the standard deviation of triplicate measurements.

Information). The rapid reduction in the environmental polarity of pyrene at ambient temperature involves significant hydrophobic association of SDS molecules induced by binding themselves with the hydrophobic part of polymer chains at an onset concentration far below its cmc (ca. 2.4 mg/mL)1,14 and manifests the formation of the SDS-polymer complex in a micelle-like structure. The onset concentration at which the copolymer-bound SDS micelles start to form is referred to as the critical aggregation concentration of SDS (cacSDS) in this study. The formation of the copolymer-bound SDS micelles at ambient temperature with an SDS concentration ca. 8-fold lower than its cmc is ascribed to the screening effect of the surfactant-water interaction by PNIPAAm chain units.14 It was reported that formation of the polymer-bound SDS micelles can be induced, respectively, from the aqueous PNIPAAm,14-18 mPEG,31-33 and PAAc34,35 (at low pH, for example, pH 3.0) solutions at the SDS concentration lower than its cmc. It is therefore necessary to study which part of the graft copolymer strongly interacts with SDS species and then promotes the formation of the copolymer-bound SDS micelles in this work. The I3/I1 values of pyrene obtained from the aqueous PNIPAAm (Mw 3800 g/mol; 3.4 and 10.0 mg/mL at 25 °C) and mPEG (Mw 2000 g/mol; 10.0 mg/mL at 25 and 60 °C) solutions, respectively, as a function of the SDS concentration shown in Figure 3 indicate that SDS molecules interact readily with PNIPAAm as compared to the inactive mPEG. Obviously, formation of the copolymerinduced SDS micelles is primarily related to the association of SDS molecules with the PNIPAAm grafts, provided that the AAc residues in the backbone are significantly ionized at pH 4.3. With respect to the results in Figure 3 and the cacSDS values reported by Schild and Tirrell14 (0.23 mg/mL) and by Walter et al.16 (0.37 mg/mL), it appears that the cacSDS varies upon changing (31) Dai, S.; Tam, K. C. J. Phys. Chem. B 2001, 105, 10759. (32) Bernazzani, L.; Borsacchi, S.; Catalano, D.; Gianni, P.; Mollica, V.; Vitelli, M.; Asaro, F.; Feruglio, L. J. Phys. Chem. B 2004, 108, 8960. (33) Dhara, D.; Shah, D. O. Langmuir 2001, 17, 7233. (34) Anghel, D. F.; Toca-Herrera, J. L., Winnik, F. M.; Retting, W.; Klitzing, R. V. Langmuir 2002, 18, 5600. (35) Wang, C.; Tam, K. C. J. Phys. Chem. B 2005, 109, 5156.

Effects of SDS on the Structure of Copolymers

Langmuir, Vol. 22, No. 16, 2006 6767 Scheme 1. Local Illustration of the Polymeric Structure with the SDS Concentrations above and below cacSDS at Different Temperatures

Figure 3. Fluorescence intensity ratio (I3/I1) of pyrene as a function of the SDS concentration in the aqueous solutions of (b) 10.0 mg/ mL and (2) 3.4 mg/mL homoPNIPAAm (Mw 3800 g/mol) (equivalent to the concentration of the PNIPAAm grafts in the aqueous copolymer solution of 10 mg/mL) at 25 °C and of mPEG (Mw 2000 g/mol; 10 mg/mL) at (9) 25 and (0) 60 °C, respectively. The pyrene concentration was 6.0 × 10-7 M.

Figure 4. Temperature dependence of the hydrodynamic diameter (Dh) of the graft copolymer in water in the presence of SDS at the concentrations of (9) 0.00, (O) 0.05, (2) 0.25, (3) 0.35, ([) 0.50, and (g) 1.00 mg/mL. SDS was dissolved in water at 4 °C prior to the addition of the graft copolymer (10.0 mg/mL).

the local hydrophobic environments, depending on the concentration, molecular weight, and, as described by Walter et al.,16 the tacticity of PNIPAAm in the aqueous phase. Figure 4 shows that, in the absence of SDS, a rapid increase in the hydrodynamic diameter (Dh) of polymeric aggregates is observed around the catpolymer (ca. 32-35 °C), consistent with the formation of polymeric aggregates by the thermally induced hydrophobic association of the PNIPAAm grafts shown in Figure 2. Upon further increasing the temperature, Dh remains relatively unchanged, although the 1H NMR data demonstrate the continuous dehydration of the interior structure of polymeric aggregates (shown below). Little change in Dh is observed when the SDS concentration is below the cacSDS (ca. 0.3 mg/mL). In agreement with the pyrene fluorescence study, the structure and size of polymeric micelles are hardly influenced in the lack of sufficient binding of SDS molecules in micelle form with the PNIPAAm

grafts occurring prior to the polymer micellization. On the other hand, when the SDS concentration is above the cacSDS, the polymer particle size becomes abruptly reduced due to the preexistence of the PNIPAAm-bound SDS micelles. Figure 4 also shows that Dh decreases and catpolymer increases with increasing SDS concentration (i.e., the amount of PNIPAAmbound SDS micelles). Association of SDS micelles with the PNIPAAm grafts enables the latter to carry negative charges and, therefore, enhances their water solubility and increases the catpolymer (Figure 4).13,14 In our previous study,25 we observed the extensive interactions between the PNIPAAm and mPEG grafts of this copolymer around the micellization temperature (ca. 3235 °C) and, as a consequence, the significant intercore connections generated by both the mPEG and PNIPAAm bridging within the multicore structured polymeric aggregates. Accompanied by the decreased polymer chain association upon increasing the PNIPAAm-bound SDS micelles, the probability for the thermally induced hydrophobic PNIPAAm cores to be connected to (or physically cross-linked with) each other is greatly reduced. Therefore, Dh of ca. 200 nm at 60 °C in the SDS-free colloidal system decreases rapidly down to 61 nm, with the SDS concentration being increased to 1.00 mg/mL. The copolymerinduced SDS micelles are concomitantly destroyed while the graft copolymer molecules undergo thermally induced hydrophobic association and polymeric micelle formation. The effect of the copolymer-bound SDS micelles on the thermally induced micellization of the graft copolymer is illustrated in Scheme 1. In support of this schematic model, Dh was found to be maintained essentially unchanged when SDS (up to a concentration of 1.20 mg/mL > cacSDS) was added into the polymeric micelle solutions at 60 °C (data not shown here). This is simply due to the incapability of forming SDS micelles along the PNIPAAm grafts under these experimental conditions and of reducing the intercore connections within polymeric micelles. In addition, the variation of cacSDS was achieved by changing the pH of the aqueous copolymer/SDS solution to further examine the relation between the colloidal particle size and copolymerbound SDS micelles at a copolymer concentration of 10.0 mg/ mL. Figure 5 shows the profile of cacSDS as a function of pH obtained from the pyrene fluorescence experiments. In the pH

6768 Langmuir, Vol. 22, No. 16, 2006

Hsu et al.

Table 1. Hydrodynamic Diameters (Dh) of Polymeric Particles at the Polymer Concentration of 10.0 mg/mLa Dh (nm) pH 3.0

pH 4.3

pH 5.0

SDS (mg/mL)

25 °C

60 °C

25 °C

60 °C

25 °C

60 °C

0.00 0.05 0.35 0.50

63.2 ( 5 (11.3)b 30.4 ( 3 (3.9) 17.2 ( 2 (6.0) 16.5 ( 3 (5.8)

288.1 ( 4 218.0 ( 3 110.5 ( 5 100.1 ( 4

8.0 ( 5 (1.3) 6.1 ( 2 (1.7) 8.1 ( 2 (1.8) 7.8 ( 3 (1.6)

206.6 ( 4 203.0 ( 5 118.1 ( 9 90.7 ( 5

6.3 ( 4 (1.6) 7.8 ( 3 (1.2) 8.9 ( 4 (1.4) 6.6 ( 3 (1.0)

335.4 ( 5 340.3 ( 7 231.6 ( 7 189.6 ( 3

a

The heating approach is described in the Experimental Section. SDS was dissolved in the aqueous media prior to the addition of the graft copolymer at 4 °C with stirring. b In parentheses are the scattering intensity of the DLS measurements in kcps.

Figure 5. Effect of the pH of the aqueous copolymer (10.0 mg/mL) solution on cacSDS. Data were obtained from the fluorescence measurements of the aqueous copolymer/SDS solutions at 25 °C, using pyrene (6.0 × 10-7 M) as a polarity probe.

range 3.0-3.6, where a significant fraction of the backbone AAc residues remains un-ionized, cacSDS is significantly reduced (ca. 0.02-0.03 mg/mL) due to the enhanced hydrophobic association of the graft copolymer. The hydrophobic associations initiated most likely by forming hydrogen bonds among the PAAc, PNIPAAm, and mPEG chains, leading to formation of the polymeric complex at low temperature.36-39 The polymeric particles with an average Dh of 63.2 nm at 25 °C in the absence of SDS was observed by DLS measurements (Table 1). With respect to the reported cacSDS values for homoPNIPAAm (Figure 3) and PAAc at pH 3.0 (ca. 0.50 mg/mL),34,35 the cacSDS at ca. 0.02 mg/mL indicates that significant hydrophobic association of the polymeric complex seems rather essential in cooperatively facilitating the SDS micelle formation. An abrupt increase in cacSDS from 0.02 to 0.30 mg/mL occurs with raising the pH from 3.6 to 4.3, due to the significantly increased ionization extent of the backbone AAc residues. As a result, the hydrophobic interactions of the AAc residues with the PNIPAAm and mPEG grafts become mostly disrupted, and the amount of SDS required for the extensive association with the graft copolymer is increased accordingly. With the pH being raised above 4.3, cacSDS remains relatively unchanged. This behavior strongly suggests that PAAc backbones, upon significant ionization, interact with neither SDS molecules nor the PNIPAAm and mPEG grafts. The copolymerbound SDS micelles are thus generated by the extensive association with the PNIPAAm grafts. (36) Chen, H.; Zhang, Q.; Li, J.; Ding, Y.; Zhang, G.; Wu, C. Macromolecules 2005, 38, 8045. (37) Khutoryanskiy, V. V.; Dubolazov, A. V.; Nurkeeva, Z. S.; Mun, G. A. Langmuir 2004, 20, 3785. (38) Vasile, C.; Bumbu, G. G.; Mylonas, Y.; Cojocaru, I.; Staikos, G. Polym. Int. 2003, 52, 1887. (39) Poe, G. D.; Jarrett, W. L.; Scales, C. W.; McCormick, C. L. Macromolecules 2004, 37, 2603.

The response of the polymeric particle size (Dh) to changes in pH and the SDS concentration at 25 and 60 °C, respectively, is summarized in Table 1. At pH 3.0, the particle sizes at both 25 and 60 °C are significantly reduced with the SDS concentration greater than its cacSDS (ca. 0.02 mg/mL). Apparently, the copolymer-bound SDS micelles even formed at 0.05 mg/mL still exhibit a pronounced effect in reducing the polymer chain association by the electrostatic repulsion force. As expected, the polymer chain association is further restricted and the particle size concomitantly reduced when the amount of the copolymerbound SDS micelles increases. At pH 4.3 and 5.0, the graft copolymer is well dissolved in water at 25 °C, irrespective of the SDS concentration. Despite the differences in Dh between the samples at pH 4.3 and 5.0 at 60 °C, a significant reduction in Dh is only observed above the cacSDS (ca. 0.30 mg/mL). II. Structural Characterization of Copolymer/SDS Complexes with Temperature. In 1H NMR measurements, the formyl proton signal of the external standard DMF at δ 8.140 ppm was selected as the reference resonance throughout this work, due to the temperature invariance of its chemical shift in nature (Figure 1). Figure 6 shows the locally magnified spectra of the feature signals of methyl protons from both the PNIPAAm grafts and SDS molecules in D2O. At low temperatures, the methyl proton signal of SDS is split into a triplet, due to the spin-spin coupling of the adjacent methylene protons. A similar triplet signal from the methyl protons of SDS alone in D2O in either monomeric or micelle form was observed (data not shown here). Although the signal integral remains unchanged, the triplet is gradually transformed into a broad singlet, while the polymeric micelles are thermally induced, indicating the relocation of SDS molecules from the more homogeneous to heterogeneous microenvironment. Concomitantly, the signal of SDS (0.50 mg/mL) becomes upfield shifted from δ 0.875 to 0.855 ppm, whereas a slight downfield shift from 0.852 to 0.855 ppm occurs with the SDS concentration (0.10 mg/mL) below the cacSDS. The chemical shift (δ 0.875 ppm) of the methyl proton signal of SDS (0.50 mg/mL) in the copolymer-bound micelles is essentially identical to that of SDS (26.0 mg/mL) in regular micelles but deviates from that in the monomeric form in the presence (δ 0.852 ppm) or absence (0.836 ppm) of the graft copolymer. The deviation in chemical shifts is caused primarily by the deshielding of the methyl protons in the (regular and copolymer-bound) SDS micelles due to the partial translocation of the surfactant alkyl protons from the gauche to trans conformation in order to release the hydrophobic tail contraction.40,41 Therefore, it becomes evident that, upon the polymer micelle formation, the PNIPAAm-bound SDS micelles are destroyed and the dissociated SDS molecules migrate to the core/shell interfaces within the polymeric micelles, thereby inducing the change of the methyl proton signal from a triplet (40) Roscigno, P.; Asaro, F.; Pellizer, G.; Ortona, O.; Paduano, L. Langmuir 2003, 19, 9638. (41) Hannak, R. B.; Farber, G.; Konrat, R.; Krautler, B. J. Am. Chem. Soc. 1997, 119, 2313.

Effects of SDS on the Structure of Copolymers

Langmuir, Vol. 22, No. 16, 2006 6769

Figure 7. Detectable fraction (%) of the PNIPAAm methyl protons (at δ 1.14 ppm) of the graft copolymer (10.0 mg/mL) with SDS at the concentrations of (9) 0.00, (b) 0.10, and (2) 0.50 mg/mL, respectively, determined by 1H NMR measurements in D2O as a function of temperature with respect to its signal integral in CDCl3 at 25 °C, using the DMF signal at 8.14 ppm as a reference.

Figure 6. Locally magnified 1H NMR spectra (in the range from δ 0.7 to 1.5 ppm) of the graft copolymer (10.0 mg/mL) and SDS [(a) 0.10 and (b) 0.50 mg/mL] in D2O at different temperatures. The inset in part a shows the variation with temperature of the local spectra of the methyl protons of SDS at 0.10 mg/mL. SDS was dissolved in D2O at 4 °C prior to the addition of the graft copolymer.

to a broad singlet and its upfield shift due to the slightly increased hydration of the alkyl chains. With respect to the unchanged signal position with temperature, the signal integral of the methyl protons of PNIPAAm decreases rapidly as the polymer micellization becomes thermally initiated (Figure 6). However, substantial fractions of the methyl proton

signal from the PNIPAAm grafts remain detectable at high temperatures, suggesting that the hydrophobic PNIPAAm microdomains of polymeric micelles comprise the solidlike inner core surrounded by the liquidlike interfacial layer.25 The structural difference between the liquid- and solidlike core structure is usually manifested by a dramatic decrease in the proton spinlattice relaxation time (T1) of the latter, rendering their proton signals undetectable in the liquid 1H NMR measurements.42,43 With respect to the signal integral measured in CDCl3 at 25 °C, the signal integral of the detectable fraction of the PNIPAAm methyl groups as a function of temperature is illustrated in Figure 7. In the absence of SDS, although polymeric micelles start to form at 32-35 °C and then exhibit little change in Dh when temperature is further increased (Figure 5), the detectable fraction of PNIPAAm continuously decreases beyond ca. 35 °C due to the progressive transformation of the liquidlike PNIPAAm interfacial layer into the solidlike structure. As reported previously,25 the presence of the liquidlike interfacial layer is ascribed to the extensive interactions between the PNIPAAm and mPEG grafts, rendering its structure looser and more hydrated. The feature proton signal of the mPEG grafts in the interfacial layer at δ 3.65 ppm thus remains fully detectable (data not shown here). As expected, the effect of SDS at 0.10 mg/mL on the phase transition of the graft copolymer is negligible and a similar temperature-dependent structural change of the hydrophobic PNIPAAm microdomains to that in the absence of SDS is validated (Figure 7). At 0.50 mg/mL of SDS, the formation of polymeric micelles is initially retarded by the preexistence of the PNIPAAm-bound SDS micelles due to the reduced hydrophobic association of PNIPAAm grafts. However, upon increasing the temperature to 50 °C, the undetectable solidlike fraction of the hydrophobic PNIPAAm microregions is appreciably enhanced because of the reduced interactions between the mPEG and PNIPAAm grafts. This is consistent with the corresponding increase in I3/I1 of pyrene within polymeric micelles (Supporting (42) Heald, C. R.; Stolnik, S.; Kujawinski, K. S.; Matteis, C. D.; Garnett, M. C.; Illum, L.; Davis, S S.; Purkiss, S. C.; Barlow, R. J.; Gellert, P. R. Langmuir 2002, 18, 3669. (43) Hrkash, J. S.; Peracchia, M. T.; Domb, A.; Lotan, N.; Langer, R. Biomaterials 1997, 18, 27.

6770 Langmuir, Vol. 22, No. 16, 2006

Hsu et al.

Table 2. Spin-Lattice Relaxation Times (T1 in s) of the Feature Protons of SDS, PNIPAAm, and PEG Grafts in SDS/D2O, Copolymer/ SDS/D2O, and Copolymer/D2O Systemsa copolymer/SDS/D2O CSDS ) 0.5 mg/mL

SDS/D2O (mg/mL)

copolymer/SDS/D2O CSDS ) 0.1 mg/mL

copolymer/D2O CSDS ) 0.0 mg/mL

temp (°C)

0.5

26

SDS

PNIPAAm

PEG

SDS

PNIPAAm

PEG

PNIPAAm

PEG

20 40

2.78 3.75

1.31 1.90

1.06 1.27

0.86 1.05

0.95 1.49

2.17 2.67

0.87 1.07

0.78 0.99

0.85 0.99

0.73 1.05

a

The concentration of the graft copolymer, if present, is 10.0 mg/mL.

Information). The polymeric micelles thus exhibit a reduced particle size and an increased tendency toward the formation of the monocore structure with a more compact and hydrophobic PNIPAAm microdomain being developed (Scheme 1). The reduced interaction between the mPEG and PNIPAAm grafts by the presence of the PNIPAAm-bound SDS micelles at ambient temperature leads to an enhanced local segmental mobility of the mPEG grafts. This is confirmed by the increase of their proton spin-lattice relaxation times (T1) at 20 and 40 °C, respectively, in comparison with those samples with the SDS concentration (0.10 mg/mL) below the cacSDS and without SDS (Table 2). Accompanied by the observed signal position of the methyl protons of SDS at δ 0.875 ppm by forming copolymerbound micelles, the corresponding proton T1 is significantly reduced because of a decrease in the molecular mobility. At the concentration (0.10 mg/mL) below the cacSDS, the T1 value of the methyl protons of SDS in the copolymer/SDS/D2O solutions at 20 °C is somewhat lower than that in the SDS/D2O solutions below the cmc. Obviously, in the presence of the graft copolymer, the segmental mobility of the SDS alkyl chains is slightly restricted, despite the inability to form the copolymer-bound SDS micelles. Being confirmed by the difference in the chemical shifts (at δ 0.852 and 0.836 ppm) between the methyl proton signals of monomeric SDS molecules in the presence and absence of the graft copolymer, the data reveal that individual molecules of SDS below cacSDS are associated with the PNIPAAm grafts at low temperatures. Nevertheless, due to the void of the copolymer-bound micelles, the binding of monomeric SDS molecules with the PNIPAAm grafts is rather weak and easily disrupted by the polymer chain association process. Therefore, the interactions between the PNIPAAm and mPEG grafts remain almost intact, as evidenced by the comparable T1 values of the ethylene protons of mPEG grafts at 20 and 40 °C to those of samples in the absence of SDS (Table 2). At 40 °C above the catpolymer, the migration of SDS molecules to the liquidlike coreshell interfaces renders the T1 (2.67 s) of their methyl protons lower than that (3.75 s) of monomeric SDS molecules in the polymer-free SDS solution but higher than the T1 (1.27 s) of the SDS molecules translocating from the PNIPAAm-bound micelles

to the more “mPEG-deficient” liquidlike interfacial layers of polymeric micelles (Table 2). In contrast to an increase in T1 of the protons of the mPEG grafts due to the formation of the copolymer-bound SDS micelles, the T1 of the methyl protons of the PNIPAAm grafts are rather insensitive not only to their varying interactions with the mPEG grafts but also to their thermally induced phase transition (data not shown here). This is primarily a consequence of the high segmental mobility of the methyl protons of the PNIPAAm grafts in the liquidlike interfacial layers.25 In summary, the results presented in this work confirm that the effects of SDS on the structure and size of the thermally induced polymeric particles occur primarily as a consequence of the copolymer-bound SDS micelles formed at low temperature. The structural changes of polymeric micelles are primarily caused by the reduced interactions between the PNIPAAm and mPEG grafts by binding SDS micelles along the PNIPAAm grafts. On the other hand, the copolymer-bound SDS micelles are destroyed and the dissociated SDS molecules migrate to the core-shell interface of polymeric micelles with their alkyl chains being buried in the liquidlike interfacial layers of the hydrophobic PNIPAAm microdomains when the temperature is increased. Such a unique role of the copolymer-bound SDS micelles in the formation of polymeric particles is further demonstrated by changes in the polymeric particle size in response to the change of cacSDS via adjusting the pH of the aqueous polymer/SDS solutions. Acknowledgment. This work was supported by the National Science Council of Taiwan (NSC93-2213-E-005-028). Supporting Information Available: The critical aggregation temperature of the graft copolymer (catpolymer) with varying levels of SDS determined by optical transmittance measurements, the emission spectra of pyrene in aqueous copolymer/SDS solutions with temperature, and the fluorescence intensity ratio (I3/I1) of pyrene in aqueous copolymer solution as a function of the SDS concentration at 25 and 55 °C. This material is available free of charge via the Internet at http://pubs.acs.org. LA060229D