Langmuir 2008, 24, 7559-7564
7559
Surface-Confined Photopolymerization of pH-Responsive Acrylamide/ Acrylate Brushes on Polymer Thin Films Gunnar Dune´r,†,‡ Henrik Anderson,‡,§ Annica Myrskog,‡,| Maria Hedlund,⊥ Teodor Aastrup,‡ and Olof Ramstro¨m*,† Royal Institute of Technology, Department of Chemistry, Teknikringen 30, S-10044 Stockholm, Sweden, Attana AB, Bjo¨rnna¨sVa¨gen 21, S-11347 Stockholm, Sweden, Uppsala UniVersity, Ångstro¨m Laboratory, Solid State Electronics, P.O. Box 534, S-75121 Uppsala, Sweden, Linko¨ping UniVersity, Department of Physics, Chemistry and Biology, S-58183 Linko¨ping, Sweden, and Uppsala UniVersity, Ångstro¨m Laboratory, Physics and Materials Science, P.O. Box 530, S-751 21 Uppsala, Sweden ReceiVed March 5, 2008. ReVised Manuscript ReceiVed April 17, 2008 Dynamic acrylamide/acrylate polymeric brushes were synthesized at gold-plated quartz crystal surfaces. The crystals were initially coated with polystyrene-type thin films, derivatized with photolabile iniferter groups, and subsequently subjected to photoinitiated polymerization in acrylamide/acrylate monomer feeds. This surface-confined polymerization method enabled direct photocontrol over the polymerization, as followed by increased frequency responses of the crystal oscillations in a quartz crystal microbalance (QCM). The produced polymer layers were also found to be highly sensitive to external acid/base stimuli. Large oscillation frequency shifts were detected when the brushes were exposed to buffer solutions of different pH. The dynamic behavior of the resulting polymeric brushes was evaluated, and the extent of expansion and contraction of the films was monitored by the QCM setup in situ in real time. The resulting responses were rapid, and the effects were fully reversible. Low pH resulted in full contractions of the films, whereas higher pH yielded maximal expansion in order to minimize repulsion around the charged acrylate centers. The surfaces also proved to be very robust because the responsiveness was reproducible over many cycles of repeated expansion and contraction. Using ellipsometry, copolymer layers were estimated to be ∼220 nm in a collapsed state and ∼340 nm in the expanded state, effectively increasing the thickness of the film by 55%.
Introduction Many polymeric systems show controlled dynamic behavior in response to external stimuli, such as temperature, pressure, pH, and chemical agents.1–18 The dynamic behavior may rely on different formats such as conformational, configurational and * Corresponding author. Fax: +46 8 7912333. E-mail:
[email protected]. † Royal Institute of Technology. ‡ Attana AB. § Uppsala University, Ångstro¨m Laboratory, Solid State Electronics. | Linko¨ping University. ⊥ Uppsala University, Ångstro¨m Laboratory, Physics and Materials Science. (1) Xu, F. J.; Li, J.; Yuan, S. J.; Zhang, Z. X.; Kang, E. T.; Neoh, K. G. Biomacromolecules 2008, 9, 331–339. (2) LeMieux, M. C.; Peleshanko, S.; Anderson, K. D.; Tsukruk, V. V. Langmuir 2007, 23, 265–273. (3) Minko, S. Polym. ReV. 2006, 46, 397–420. (4) Ionov, L.; Sapra, S.; Synytska, A.; Rogach, A. L.; Stamm, M.; Diez, S. AdV. Mater. 2006, 18, 1453–1457. (5) Lupitskyy, R.; Roiter, Y.; Tsitsilianis, C.; Minko, S. Langmuir 2005, 21, 8591–8593. (6) Harnish, B.; Robinson, J. T.; Pei, Z.; Ramstro¨m, O.; Yan, M. Chem. Mater. 2005, 17, 4092–4096. (7) Gillies, E. R.; Fre´chet, J. M. J. Bioconjugate Chem. 2005, 16, 361–368. (8) Crowe, J. A.; Genzer, J. J. Am. Chem. Soc. 2005, 127, 17610–17611. (9) Brady, S.; Diamond, D.; Lau, K.-T. Sens. Actuators, A 2005, 119, 398– 404. (10) Xu, F. J.; Zhong, S. P.; Yung, L. Y. L.; Kang, E. T.; Neoh, K. G. Biomacromolecules 2004, 5, 2392–2403. (11) Toomey, R.; Freidank, D.; Ru¨he, J. Macromolecules 2004, 37, 882–887. (12) You, L.-C.; Lu, F.-Z.; Li, Z.-C.; Zhang, W.; Li, F.-M. Macromolecules 2003, 36, 1–4. (13) Ionov, L.; Minko, S.; Stamm, M.; Gohy, J.-F.; Jerome, R.; Scholl, A. J. Am. Chem. Soc. 2003, 125, 8302–8306. (14) Khongtong, S.; Ferguson, G. S. J. Am. Chem. Soc. 2002, 124, 7254–7255. (15) Jeong, B.; Anna, G. Trends Biotechnol. 2002, 20, 360. (16) De, S. K.; Aluru, N. R.; Johnson, B.; Crone, W. C.; Beebe, D. J.; Moore, J. J-MEMS 2002, 11, 544–555. (17) Park, Y. S.; Ito, Y.; Imanishi, Y. Chem. Mater. 1997, 9, 2755–2758. (18) Chun, S.-W.; Kim, J.-D. J. Controlled Release 1996, 38, 39–47.
constitutional changes,19–23 providing extensive control of a wide variety of systems. Such behavior possesses high potential for the development of new responsive materials, considering the many possibilities that the manipulation of chemistry, conformation, and dynamics of long polymer chains can offer.24,25 Of these, temperature- and pH-dependent polymeric systems have been most extensively studied because of their efficient control. Thermally dependent systems may, for example, include changes in adhesion properties, membrane permeability, and drug/gene delivery.14,18,26–29 Polymers such as poly(N-isopropylacrylamide) (PNIPAAm), poly(ethylene oxide) (PEO), and poly(propylene oxide) (PPO) are associated with a lower critical solution temperature (LCST). Upon LCST transition, the polymers go from a dissolved phase to an aggregated form and vice versa.30 Some systems have the potential of being used as sensing elements or active response elements.31 Materials dependent on pH have, for example, been designed for permeation control, controlled release formulations, new medical devices, and biosensors.7,17,28,29,32 (19) Lehn, J.-M. Chem. Soc. ReV. 2007, 36, 151–160. (20) Yuan, W.; Jiang, G.; Wang, J.; Wang, G.; Song, Y.; Jiang, L. Macromolecules 2006, 39, 1300–1303. (21) Xia, F.; Feng, L.; Wang, S.; Sun, T.; Song, W.; Jiang, W.; Jiang, L. AdV. Mater. 2006, 18, 432–436. (22) Yusa, S.-i.; Sakakibara, A.; Yamamoto, T.; Morishima, Y. Macromolecules 2002, 35, 5243–5249. (23) Hugel, T.; Holland Nolan, B.; Cattani, A.; Moroder, L.; Seitz, M.; Gaub Hermann, E. Science 2002, 296, 1103–1106. (24) Mather, P. T. Nat. Mater. 2007, 6, 93–94. (25) Russell, T. P. Science 2002, 297, 964–967. (26) Pennadam, S. S.; Ellis, J. S.; Lavigne, M. D.; Gorecki, D. C.; Davies, M. C.; Alexander, C. Langmuir 2007, 23, 41–49. (27) Howard, K. A.; Dong, M.; Oupicky, D.; Bisht, H. S.; Buss, C.; Besenbacher, F.; Kjems, J. Small 2007, 3, 54–57. (28) Schmaljohann, D. AdV. Drug DeliVery ReV. 2006, 58, 1655–1670. (29) Alexander, C.; Shakesheff, K. M. AdV. Mater. 2006, 18, 3321–3328. (30) Nath, N.; Chilkoti, A. AdV. Mater. 2002, 14, 1243–1247.
10.1021/la800700h CCC: $40.75 2008 American Chemical Society Published on Web 06/19/2008
7560 Langmuir, Vol. 24, No. 14, 2008
Dune´r et al.
Scheme 1. Quartz Crystal Coating and Surface Polymerization
Responsive polymers may be deposited at surfaces, for example, by chemical grafting, in situ polymerized by, for example, surface-catalyzed polymerization, atom-transfer radical polymerization (ATRP), reversible addition-fragmentation chain transfer polymerization (RAFT), or nitroxide-mediated polymerization (NMP).8,33–40 ATRP has in these cases been more preferred because of the possibility of controlled microarchitectural design.41 However, another photobased polymerization technique makes use of the iniferter system, pioneered by Otsu and co-workers.42,43 An iniferter acts as initiator, transfer agent, and terminator and allows for the control of polymer thickness and block copolymerization.44,45 The photoinitiated iniferter technique is thus a useful tool for creating responsive polymer surfaces of the desired functionality, giving rise to a variety of applications. Moreover, the quartz crystal microbalance (QCM) technique offers a convenient method for following mass-induced surfaces phenomena and may be used to record pH responses,6 thermal responses,1 and polymer growth.46 Herein, we report an application of the iniferter technique showing the fast, responsive, reversible behavior of a copolymer of acrylamide (AAm) and acrylic acid (AAc) to changes in pH at the surface of a goldplated quartz crystal (Scheme 1).
Experimental Section General. Acrylamide (AAm) from Bio-Rad, acrylic acid (AAc) from Fluka, sodium diethyldithiocarbamate trihydrate (NaDEC) from Aldrich, and poly(vinylbenzyl chloride) (PVBC) from Aldrich (60/ 40 mixture of 3- and 4- isomers, average Mn≈ 55 000 and average Mw ≈ 100 000 as determined by GPC) were used as received. 1H NMR spectra were recorded with a Bruker DMX 500 instrument at (31) Luzinov, I.; Minko, S.; Tsukruk, V. V. Prog. Polym. Sci. 2004, 29, 635– 698. (32) Cai, Q.; Zeng, K.; Ruan, C.; Desai, T. A.; Grimes, C. A. Anal. Chem. 2004, 76, 4038–4043. (33) Lowe, A. B.; McCormick, C. L. Prog. Polym. Sci. 2007, 32, 283–351. (34) Ayres, N.; Boyes, S. G.; Brittain, W. J. Langmuir 2007, 23, 182–189. (35) Favier, A.; Charreyre, M.-T. Macromol. Rapid Commun. 2006, 27, 653– 692. (36) Bai, D.; Elliott, S. M.; Jennings, G. K. Chem. Mater. 2006, 18, 5167– 5169. (37) Kaholek, M.; Lee, W.-K.; LaMattina, B.; Caster, K. C.; Zauscher, S. Nano Lett. 2004, 4, 373–376. (38) Edmondson, S.; Osborne, V. L.; Huck, W. T. S. Chem. Soc. ReV. 2004, 33, 14–22. (39) Pyun, J.; Kowalewski, T.; Matyjaszewski, K. Macromol. Rapid Commun. 2003, 24, 1043–1059. (40) Yakushiji, T.; Sakai, K.; Kikuchi, A.; Aoyagi, T.; Sakurai, Y.; Okano, T. Langmuir 1998, 14, 4657–4662. (41) Gil, E. S.; Hudson, S. M. Prog. Polym. Sci. 2004, 29, 1173–1222. (42) Otsu, T.; Matsunaga, T.; Doi, T.; Matsumoto, A. Eur. Polym. J. 1995, 31, 67–78. (43) Otsu, T. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2121–2136. (44) Lambrinos, P.; Tardi, M.; Polton, A.; Sigwalt, P. Eur. Polym. J. 1990, 26, 1125–1135. (45) Turner, S. R.; Blevins, R. W. Macromolecules 1990, 23, 1856–1859. (46) Moya, S. E.; Brown, A. A.; Azzaroni, O.; Huck, W. T. S. Macromol. Rapid Commun. 2005, 26, 1117–1121.
298 K in CDCl3 using the residual signal from CHCl3 (1H: δ ) 7.28 ppm) as an internal standard. A 450 W medium-pressure mercury lamp (ACE-Hanovia) was used for polymerizations. QCM experiments were conducted with Attana 100 or Attana 80 biosensor instrumentation (Attana). The Attester software (Attana) was used to monitor frequency changes. An ESCA 300 (Scienta) was used for the elemental analysis of coated crystals. Spin-coated crystals were prepared using a spin coater (model P6700 series, Specialty Coating Systems). Fourier-transform infrared reflection absorption spectroscopy (IRAS) was performed using IFS66 instrumentation (Bruker). The buffer solutions used were phosphate-buffered saline (PBS, 10 mM) or phosphate buffer (PB, 10 mM) from Sigma-Aldrich, adjusted with HCl or NaOH using an EcoScan pH 5/6 and ion 6 pH meter. Preparation of Iniferter-Derivatized Poly(vinylbenzyl chloride) (PVBDij). In a typical example, PVBC (1.0 g, 6.6 mmol) and NaDEC (0.30 g, 1.3 mmol) were dissolved in THF (50 mL), and the mixture was stirred at ambient temperature overnight. After the evaporation of the solvent, the resulting white powder was rinsed twice with deionized water and then with methanol. The product was dissolved in chloroform (75 mL), and water-soluble substances were extracted twice with deionized water. Evaporation and drying in vacuo at 40 °C yielded 1.15 g (quant) of PVBD51 as a white powder. 1H NMR (CDCl3): δ 7.07-6.49 (m, 20H, C6H4), 4.45 (s, 10H, C6H4-CH2), 4.01 (m, 2H, NCH2), 3.72 (m, 2H, NCH2), 2.2-1.2 (m, 21H, CHCH2, CHCH2, CH2CH3). Coating of Crystals. Gold-plated, AT-cut 10 MHz quartz crystals (Attana) were initially ultrasonicated for 5 min in ethanol and dried under nitrogen. PVBDij (15 mg) was dissolved in toluene, and the solution was either carefully drop coated (solvent evaporated under a stream of nitrogen) or spin coated (2000 rpm, 60 s) onto the crystals. Preparation of Poly(AAm), Poly(AAc), and Poly(AAm-co-AAc). In a typical preparation, AAm (1.44 g, 20.3 mmol) and AAc (160 µL, 2.3 mmol) were dissolved in a 1:1 (v/v) mixture (60 mL) of deionized water and ethanol and deaerated prior to polymerization. The QCM crystals were placed in Pyrex glass tubes with the monomer feed solution, and a stream of nitrogen was blown over the solution throughout the process. Polymerization was initiated by UV irradiation at a measured intensity of 13.3-13.5 mW/cm2 at 360 nm unless otherwise stated, and irradiation was allowed to proceed for 120 min. QCM Measurements and Determination of Dissociation Constants. QCM crystals subjected to polymerizations were dried under nitrogen and inserted into the QCM instrument for evaluation of the frequency differences in their dry state. The increase in mass was estimated according to the Sauerbrey equation (eq 1)
∆m )
-∆fANFq f02
(1)
where ∆m is the change in mass, ∆f is the frequency change or frequency shift, A is the area of the electrode (0.16 cm2), N is the frequency constant (1670 × 102 Hz cm), Fq is the density of quartz (2.65 g/mL), and f0 is the fundamental frequency of the crystal (10
pH-ResponsiVe Acrylamide/Acrylate Brushes
Langmuir, Vol. 24, No. 14, 2008 7561
Scheme 2. Preparation of Iniferter-Derivatized Polymer PVBDij
MHz). With the given quantities, a shift of 1 Hz corresponds to a mass change of 700 pg.47 Estimations of apparent dissociation constant values (pKapp values) were conducted with a liquid flow of 25 µL/min unless otherwise stated. Values were estimated from frequency shifts, recorded in triplicate, for samples ranging from pH 3.0 to 8.0. The results were subjected to nonlinear regression analysis according to eq 2 using the software package GraphPad Prism (GraphPad Software). ∆fmin is the minimum frequency difference, and ∆fmax is the maximum frequency difference.
∆f ) ∆fmin + (∆fmax - ∆fmin) ⁄ (1 + 10pKapp - pH)
(2)
Ellipsometry Measurements and Thickness Estimation. Measurements were performed in solutions of pH 3.0-7.0 with a Rudolph Research/AutoEL instrument using a He-Ne laser at 632.8 nm and at an angle of incidence of 70°. For estimation of the optical thickness, a three-phase model was adopted with the iniferter layer as the substrate, the copolymer as the film, and the pH solutions as the medium. Refractive indices of the polystyrene-like iniferter substrate (1.575), the collapsed (1.440) and swollen (1.339) copolymers, and the medium (1.335) were set to reported values.48–51 Grafting Density. The grafting density was estimated using eq 3
σ)
mfilmNA MpA
(3)
where σ is the grafting density in chains nm-2, mfilm is the mass of the film in the collapsed state, NA is Avogadro’s number, Mp is the molecular mass of the average polymer, and A is the area of the QCM electrode.
Results and Discussion To form a stable, yet removable, layer on the quartz crystal surfaces, the adsorption of polystyrene-type polymers was adopted. This allows for simple coverage of the surface by drop coating or spin coating and enables the straightforward derivatization of the formed films. The resulting layers are sufficiently robust to withstand a wide variety of conditions and can in principle be desorbed from the quartz crystal surface by treatment with solvent blends. Poly(vinylbenzyl chloride) (PVBC) was thus chosen as the foundation layer precursor. Functionalization of the precursor polymer was performed by reacting PVBC with sodium diethyldithiocarbamate (NaDEC), in different ratios, to produce the active iniferter-derivatized polymer (PVBDij) (Scheme 2). Polymers resulting from a lower NaDEC/PVBC ratio generated hydrophilic surfaces in preliminary UV-irradiation assays; consequently, these polymers were mainly chosen for (47) Sauerbrey, G. Z. Phys. 1959, 155, 206–222. (48) Beines, P. W.; Klosterkamp, I.; Menges, B.; Jonas, U.; Knoll, W. Langmuir 2007, 23, 2231–2238. (49) Harmon, M. E.; Kuckling, D.; Frank, C. W. Macromolecules 2003, 36, 162–172. (50) Hu, X.; Shin, K.; Rafailovich, M.; Sokolov, J.; Stein, R.; Chan, Y.; Williams, K.; Wu, W. L.; Kolb, R. High Perform. Polym. 2000, 12, 621–629. (51) Karlsson, C. A. C.; Wahlgren, M. C.; Christian Tra¨gårdh, A. Colloids Surf., B 1996, 6, 317–328.
Figure 1. ESCA spectra of coated gold-plated quartz crystals. (Top) PVBC. (Bottom) Iniferter-derivatized polymer film (PVBD01) with complete dithiocarbamate substitution.
subsequent assays. However, iniferter-derivatized polymers with higher dithiocarbamate content were also produced. The photoactive polymers were then coated onto the gold electrodes of the QCM crystals (Scheme 1). ESCA was used to verify the iniferter synthesis and coating. Gold crystals were spin coated with the iniferter-derivatized polymer in which all benzyl groups were substituted (PVBD01). As a reference, a PVBC-coated electrode was used. Figure 1 shows the resulting spectra for the PVBC-coated gold electrode and an iniferter-derivatized polymer film (PVBD01), respectively. As can be seen, covering the gold electrode with iniferterderivatized polymer results in nearly complete suppression of the gold signals. Furthermore, the disappearance of chloride signals and the presence of nitrogen and sulfur signals in the PVBD-film spectrum indicate complete dithiocarbamate substitution in agreement with NMR data. Initial photopolymerization tests with PVBD01 films, where all Cl groups were substituted with dithiocarbamate groups, were not successful. Therefore, for the synthesis of polymer-brush surfaces, the molar proportions of PVBC and NaDEC were optimized to create dilute iniferter-derivatized polymers to prevent possible cross linking between adjacent initiation sites upon UV irradiation. The selected iniferter-derivatized polymer (PVBD51) used for subsequent grafting studies had a ratio of 1/5 of unsubstituted chlorides and active diethyldithiocarbamate groups. All resulting iniferter-derivatized polymers could be stored at room temperature for extended periods of time without degradation. The frequency responses of the crystals were measured under dry conditions before and after drop coating with the iniferter-derivatized polymers. The typical frequency shift resulting from the coating process was 3-7 kHz, corresponding to 2-5 µg according to the Sauerbrey equation.47 In a subsequent step, the samples were submerged in the monomer feed solutions. All monomer preparations were made in water/ethanol (1:1 v/v). This composition enabled the solubility of the monomers as well as wetting of the hydrophobic iniferter-derivatized polymer surface without disruption of the film. After UV irradiation in the monomer solutions, the frequency responses of the crystals
7562 Langmuir, Vol. 24, No. 14, 2008
Figure 2. Dependence of the frequency shift from UV irradiation time using AAc as the monomer at an intensity of 5 mW/cm2 at 360 nm.
Figure 3. IRAS analyses of photogenerated QCM surfaces: (a) PVBD51, (b) poly(AAc)-grafted PVBD51, c) poly(AAm)-grafted PVBD51, and (d) poly(AAm-co-AAc)-grafted (10% AAc) PVBD51. Mean values from 3000 scans (10 min); resolution 2 cm-1; reference HS(CD2)15CD3 on gold.
were recorded, resulting in frequency shifts of 200-2100 Hz (0.2-1.5 µg). The presence of hydrophilic groups at the surface of the QCM crystals was verified by qualitative contact angle measurements clearly indicating hydrophilic polymer formation. The amount of formed polymer as a result of UV irradiation varies with time as seen in Figure 2. This nonlinear effect is more pronounced than in previous studies;52 however, it is comparable within the 20 min time window used. In the present case, two domains can be discerned: an initial phase showing slow chain growth and a secondary phase showing an increased growth rate. The nonlinearity in the present study is a source of continuous investigation and may have several causes; one possible factor could be phase-related because the monomers are not solvated in the vicinity of the iniferter-derivatized polymer thin film. Other factors that may affect the extent of the polymerization are the monomer concentration, the radiation intensity, and the dithiocarbamate surface density.52 However, the polymerization can be stopped at any moment by turning off the UV light and likewise restarted by switching it on. The presence and composition of the polymers were further verified by infrared reflection-absorption spectroscopy (IRAS). Figure 3 shows the resulting spectra of spin-coated iniferterderivatized polymer (PVBD51) devoid of any grafting and PVBD51 grafted with poly(AAm), poly(AAc), and poly(AAm-co-AAc) (10% AAc). The peaks at 3341, 3201, 1693, and 1615 cm-1 in Figure 3 show the addition of amides to the iniferter-derivatized polymer surface. The peaks at 1736 cm-1 in Figure 3 indicate the presence of carboxylic groups. Thus, both AAm and AAc could be efficiently incorporated into the growing polymer brushes at the (52) Nakayama, Y.; Matsuda, T. Macromolecules 1999, 32, 5405–5410.
Dune´r et al.
Figure 4. ESCA analysis of the AAm/AAc monomer ratio in the copolymer brushes; the respective peaks are indicated. The inset shows a scan of O 1s: hydroxyl oxygen (a) and carbonyl oxygen (b).
surfaces. As further characterization of the brushes, the AAm/ AAc ratio was determined by ESCA analysis. Figure 4 shows a total survey of a poly(AAm-co-AAc) film. The area ratio of indicated peaks is 0.205 (O 1s), 0.142 (N 1s), and 0.653 (C 1s). The inlet shows an O 1s scan indicating two types of oxygen atoms present in the copolymer, hydroxyl (a) and carbonyl (b). The hydroxyl oxygen from the acrylate corresponds to 18% of the total oxygen measured. The AAm/AAc ratio is thus calculated to be 3.7. This value is lower than the monomer feed solution molar ratio of 8.7, which indicates that the incorporation of monomers in the growing brushes is not isotropic. This may be due to a higher intrinsic apparent polymerization rate constant for AAc than for AAm in water;53 in addition, monomers of AAm present at higher abundance are more likely to dimerize, reducing the concentration of free AAm monomers.53 In accordance, reactivity ratios tend to favor poly(AAc) over poly(AAm).54 The prepared crystals were inserted into the flow-through QCM instrument with a running buffer of pH e3. In a flow-cell system, the poly(AAm-co-AAc) film expands easily upon deprotonation of the carboxylic groups and is registered as frequency shifts. The left side of Figure 5 shows how a sample of poly(AAmco-AAc)-grafted PVBD51 responds to PBS injections of pH 3-9 in the QCM instrument at a measured flow rate of 50 µL/min. The right side of Figure 5 shows the corresponding case when only AAm was used as a monomer in the grafting process. In the first case, the frequency shifts increase dramatically at higher pH, showing a pH dependence, whereas for the latter surface no such dependence could be recorded. For the AAm reference sample, the small background shifts are at the same level and nonincreasing over the entire pH range. The different responses to the pH gradient indicate that acrylic acid was successfully incorporated into the copolymer matrix. Different ratios of AAc were further evaluated, and it was found that an 10 mol % AAc was sufficient to achieve a responsive polymer. To characterize the copolymer surfaces further, ellipsometry measurements were performed to estimate the film thicknesses in the collapsed and swollen states. The results of the calculations showed an average thickness of 220 nm in the contracted state and 340 nm in the expanded state, thus indicating an apparent film expansion of 55%. From the determined height of the fully extended polymers and a calculated value of the polymer molecular mass of 98 000 g mol-1, the grafting density can be estimated to be 0.57 chains nm-2. This can be compared to the high values of 0.65 and 0.67 chains nm-2 reported in the literature.55,56 (53) Seabrook, S. A.; Tonge, M. P.; Gilbert, R. G. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 1357–1368.
pH-ResponsiVe Acrylamide/Acrylate Brushes
Langmuir, Vol. 24, No. 14, 2008 7563
Figure 7. Frequency responses to injections of PB at pH 2.1 to 8.0 for poly(AAm-co-AAc)-grafted (10% AAc) PVBD51-coated QCM crystals. Apparent dissociation constant pKapp ) 5.24, and R2 ) 0.9943.
Figure 5. Frequency responses of polymer-coated QCM crystals to injections of PBS at pH ranging from 3.0 to 9.0. (Top) Poly(AAmco-AAc)-grafted (10% AAc) PVBD51-coated crystals. (Bottom) Poly(Aam)-grafted PVBD51-coated crystals.
Figure 6. Frequency responses of poly(AAm-co-AAc)-grafted (10% AAc) PVBD51-coated QCM crystals to repeated PBS injections at pH 6.5.
In light of the ellipsometry measurements and the frequency shifts, the pH dependence can be explained by the increased amount of water that resonates within the gel upon deprotonation and concomitant extension of the polymer brushes. We have previously reported similar hydrogel-like behavior for films of poly(4-vinylpyridine) (P4VP), in this case showing the opposite pH dependence.6 It was shown that NaCl/KCl was not specifically absorbed by the expanded layer. Furthermore, gold and PS surfaces were devoid of any mass change upon buffer injections, suggesting that the mass change depended on incorporated solvent molecules to minimize the electrostatic repulsion of deprotonated carboxylic groups. To evaluate the stability of the system and to verify the reversibility of the brush conformational states, a repetition study was performed. Figure 6 shows repeated injections of PBS at pH 6.5 over a poly(AAm-co-AAc)-grafted PVBD51-coated crystal. The average frequency shift for the signals in Figure 6 is 284 ( 1.0 Hz showing both reproducibility and reversibility in the expansion/contraction process. The results from Figure 6 also confirm the complete reversibility of the brush expansion/ contraction states. The regularity of the pH-dependent state changes could furthermore be used to evaluate the dissociation constant of the AAc groups.
Figure 7 shows average frequency shift responses to buffer injections from pH 2.1 to 8.0 for a poly(AAm-co-AAc)-grafted PVBD51-coated crystal with 10 mol % AAc in the monomer feed solution. The transition between contracted and expanded states followed the degree of deprotonation, and nonlinear regression resulted in an apparent dissociation constant (pKapp) of 5.24, with high accuracy. Surfaces with AAc feed percentages of 10, 30, 50, and 100 were evaluated, but no correlation between the pKapp of the polymer and the percentage of AAc in the monomer feed solution could be noted. Miller and colleagues found that the pKa in homopolymers of poly(acrylic acid) is independent of molecular weight and conclude that the dissociation of a given carboxyl is influenced by adjacent groups in the same chain and not on nearby loops.57 From these results, a carboxyl group on a neighboring chain, as is the presumed case on the iniferterderivatized polymer surface, will not affect the acidic strength if individual brushes behave like solution-phase homopolymers. However, in our case intermolecular hydrogen bonding between carboxyl groups on close enough chains may be expected as a result of the relatively ordered attachment at the surface, which is most pronounced for the acrylic high-percentage polymers. However, no correlation was found from the collected data that would suggest any inter- or intramolecular influence.
Conclusions The demonstrated application of the iniferter polymerization technique combined with QCM monitoring provides an efficient means of studying responsive surfaces leading to a direct or an indirect mass change. Frequency shifts of poly(AAm-co-AAc)grafted PVBD-coated crystals were registered upon administration of a pH gradient. Because of the deprotonation of the carboxylic hydrogens, the polymer brushes expand, resulting in increased water content co-resonating with the polymer films. The grafted copolymer shows both reproducibility and reversibility in the expansion/contraction process, indicating the unhindered expansion of the brushes. In accordance with previous results, neighboring chains do not affect carboxyl deprotonation. All steps include simple handling. The density of active iniferter sites can be easily governed in the one-step synthesis by adjusting the molar percentage of dithiocarbamate groups compared to that of benzyl chloride residues. The crystal preparation is straightforward, and the equipment is easily operated. The long, polymeric, styrene-based chains were deposited without difficulty (54) Paril, A.; Alb, A. M.; Giz, A. T.; Catalgil-Giz, H. J. Appl. Polym. Sci. 2007, 103, 968–974. (55) Ohno, K.; Morinaga, T.; Koh, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2005, 38, 2137–2142. (56) Stenzel, M. H.; Zhang, L.; Huck, W. T. S. Macromol. Rapid Commun. 2006, 27, 1121–1126. (57) Miller, M. L.; Rauhut, C. E. J. Colloid Sci. 1959, 14, 524–532.
7564 Langmuir, Vol. 24, No. 14, 2008
on the gold surface of the quartz crystal by drop coating or spin coating and were attached firmly by physisorption. The time of UV irradiation determines the thickness of the hydrogel, and the apparent dissociation constant of the produced hydrogel surfaces can be determined. Acknowledgment. We thank the Swedish Research Council, the Royal Institute of Technology, and Attana AB for financial
Dune´r et al.
support. Dr. Ulrik Gelius is gratefully thanked for assistance with ESCA analyses. Supporting Information Available: AFM images of a gold surface, an iniferter-coated surface, and a copolymer surface. This material is available free of charge via the Internet at http://pubs.acs.org. LA800700H
