pubs.acs.org/Langmuir © 2010 American Chemical Society
Swelling and Collapse of an Adsorbed pH-Responsive Film-Forming Microgel Measured by Optical Reflectometry and QCM Shaun C. Howard,† V. S. J. Craig,*,†,^ Paul A. FitzGerald,‡ and Erica J. Wanless§ †
Department of Applied Mathematics, Research School of Physics and Engineering, Australian National University, Canberra, ACT 0200, Australia, ‡School of Chemistry, F11, The University of Sydney, NSW 2006, Australia, §School of Environmental and Life Sciences, The University of Newcastle, Callaghan, New South Wales, 2308, Australia, and ^Department of Chemistry, The Royal University of Phnom Penh, Russian Federation Boulevard, Toul Kork, Phnom Penh, Cambodia Received June 8, 2010. Revised Manuscript Received August 14, 2010 The swelling and deswelling of a pH-responsive electrosterically stabilized poly[2-(diethylamino)ethyl methacrylate] microgel adsorbed to silica surfaces have been quantified using the techniques of optical reflectometry (OR) and quartz crystal microbalance (QCM). It is shown that by utilizing and comparing OR measurements performed on wafers with differing oxide layer thicknesses the adsorbed amount and film thickness of the adsorbed microgel in both the swollen and deswollen forms can be determined. Also, the kinetics of the transition can be followed, revealing that collapse is a slower process than swelling, and direct support is provided for the formation of a dense outer layer or skin during collapse that slows the deswelling process. It is shown that the adsorption of this low glass transition temperature filmforming microgel latex is robust to changes in pH after an initial swelling event which is responsible for desorption of a large and variable fraction of the initially adsorbed polymer. Subsequent deswelling and swelling of the adsorbed film indicates that adsorption to a surface greatly hinders the volumetric swelling capacity of the microgel film. In its swollen state the film is only 3-4 times thicker than the collapsed film, whereas for particles in bulk the volume increases by a factor of 20 upon protonation of the tertiary amine residues. QCM results show that even in the collapsed form the film contains a considerable amount of water. Further, the viscoelasticity of the deswollen film is similar to that of the swollen film, suggesting that the degree of cross-linking is the primary determinant of viscoelasticity.
Introduction Aqueous microgel dispersions consist of insoluble lightly crosslinked polymer latex particles of colloidal dimensions dispersed in a poor solvent.1 The particles swell substantially when exposed to a good solvent, typically through a change in solution pH, temperature, or electrolyte concentration. The low level of cross-linking within the particles ensures swelling rather than dissolution of the polymer with the improvement in solvent quality. The swelling is highly reversible when the stimulus is temperature, harnessing for example the polymer LCST in the case of a poly(N-isopropylacrylamide) (PNIPAM) microgel.2 For pH-responsive microgels, the critical swelling pH is defined by the polymer average pKa,3 as reported for the protonation of basic 2-vinylpyridine residues below pH 4.1, resulting in increased solubility and solvent uptake by the P2VP microgel particles.4 The swelling of pH-responsive microgel particles is also generally reversible, although multiple swelling cycles lead to electrolyte buildup and may ultimately override the swelling capacity of the particles. Colloidal stability can be ensured throughout multiple swelling and collapse cycles through the incorporation of a steric stabilizer on the surface of the latex during synthesis. A number of applications for stimulusresponsive microgel particles have been proposed, such as viscosity *Corresponding author. E-mail:
[email protected]. (1) (a) Ballauff, M.; Lu, Y. Polymer 2007, 48(7), 1815–1823. (b) Hoare, T.; Pelton, R. Curr. Opin. Colloid Interface Sci. 2008, 13(6), 413–428. (2) Pelton, R. Adv. Colloid Interface Sci. 2000, 85(1), 1–33. (3) Tan, B. H.; Tam, K. C. Adv. Colloid Interface Sci. 2008, 136(1-2), 25–44. (4) Dupin, D.; Fujii, S.; Armes, S. P.; Reeve, P.; Baxter, S. M. Langmuir 2006, 22(7), 3381–3387. (5) Kiminta, D. M. O.; Luckham, P. F.; Lenon, S. Polymer 1995, 36(25), 4827– 4831.
Langmuir 2010, 26(18), 14615–14623
modifiers5 and switchable particulate emulsifiers.4,6 Capture and release applications which harness the tunable swelling have also been suggested including delivery capsules for drug molecules7 and biosensors.8 An alternative family of pH-responsive microgels is that based on basic tertiary amine methacrylates such as 2-(diethylamino)ethyl methacrylate, DEA.9 These acid-swellable microgels can be readily synthesized by emulsion polymerization to include either a nonionic steric stabilizer or indeed a cationic electrosteric stabilizer to ensure colloidal stability across a broad pH range. Since the average pKa of DEA homopolymer (PDEA) is around pH 7.0-7.3, a PDEA microgel swells at physiological pH unlike other pH-responsive microgels previously reported. For example, a 1% cross-linked PDEA microgel sterically stabilized by a poly[2-(dimethylamino)ethyl methacrylate] macromonomer was measured to swell by a factor of ∼20 in volume from a 455 nm diameter latex above pH 6.5-7.0 to the swollen microgel at pH values below this range. A critical swelling pH that is nearly neutral suggests potential biomedical applications for these microgels. The latex-to-microgel swelling transition occurred rapidly, in less than 10 s as detected via crude turbidity measurements.9 A subsequent detailed stopped flow analysis of P2VP microgel swelling (6) Dupin, D.; Rosselgong, J.; Armes, S. P.; Routh, A. F. Langmuir 2007, 23(7), 4035–4041. (7) Oh, J. K.; Drumright, R.; Siegwart, D. J.; Matyjaszewski, K. Prog. Polym. Sci. 2008, 33(4), 448–477. (8) (a) Kim, J.; Singh, N.; Lyon, L. A. Biomacromolecules 2007, 8(4), 1157–1161. (b) Su, S. X.; Ali, M.; Filipe, C. D. M.; Li, Y. F.; Pelton, R. Biomacromolecules 2008, 9(3), 935–941. (9) Amalvy, J. I.; Wanless, E. J.; Li, Y.; Michailidou, V.; Armes, S. P.; Duccini, Y. Langmuir 2004, 20(21), 8992–8999.
Published on Web 08/26/2010
DOI: 10.1021/la1023218
14615
Article
kinetics showed that the characteristic swelling time correlates linearly with the mean particle diameter, as predicted by the Tanaka equation.6 The swelling of isolated PDEA microgel particles is expected to follow this same relationship. Interestingly, microgel deswelling required many hours for complete recovery to the collapsed state, indicating a significant hysteresis. The hindered collapse was attributed to the excretion of electrolyte from the microgel interior being retarded by the ingress of water due to substantial differences in osmotic pressure between the microgel interior and the bulk solution.9 There is a growing interest in microgel activity at interfaces for application as functional coatings in controlled uptake and release applications,7 micro-optics,10 water purification,11 biofouling,12 and biosensors.8 In the case of pH-responsive microgels, adsorption to a surface allows the solvent to be easily exchanged, overcoming the negative effects of electrolyte buildup on microgel swelling, as a result of multiple pH cycling. The facile synthesis of microgel particles via emulsion polymerization provides a cost-effective means of stimulus-responsive particle formation. Films can subsequently be formed via spin-coating, dip-coating, or adsorption, and if microgel character is retained, then swelling and collapse of the film opens up the potential for small target molecule capture and release on the substrate. Recent reports have focused on characterization of the dried coatings, rehydration, and retention of microgel character. For example, multilayers of anionic PNIPAM-co-AA microgel and cationic poly(allylamine hydrochloride) (PAH) spin-coated films have been used for the thermally controlled uptake and release of doxorubicin.13 Kazakov et al. prepared lipid bilayer-coated poly(NIPAM-co-N-vinylimidazole) microgels and then adsorbed these onto mica to study their deformation using tapping mode atomic force microscopy (TMAFM) in air.14 After incubation for 2 h, the partially dehydrated particles became flattened and formed a uniform, homogeneous structure, described as a “nanofilm”. Monolayer particle films have been reported where swollen PNIPAM-based microgel particles were adsorbed onto poly(ethyleneimine)-coated silicon wafers, followed by drying to obtain films.15 The rehydrated films could be reversibly swollen and deswollen by cycling either the solution pH or temperature. Absorption of a cationic surfactant within these films was explored as a prelude to controlled uptake/release applications. The swelling and collapse of rehydrated PDEA-PMAA copolymer microgel coatings has recently been followed by spectroscopic ellipsometry with the assumption that the microgel is a slab of uniform refractive index in order to solve for the thickness and the refractive index.16 The in situ characterization of microgel coatings is pertinent to understanding the coating formation and subsequent stimulusresponsive character. FitzGerald et al. were the first to monitor in situ the swelling and collapse of P2VP microgel particles adsorbed on mica and silica using AFM.17 With a glass transition temperature, Tg above 100 C, these latexes retained their integrity in the adsorbed coating, facilitating ready calculation of the (10) Serpe, M. J.; Kim, J.; Lyon, L. A. Adv. Mater. 2004, 16(2), 184. (11) Morris, G. E.; Vincent, B.; Snowden, M. J. J. Colloid Interface Sci. 1997, 190(1), 198–205. (12) Bridges, A. W.; Singh, N.; Burns, K. L.; Babensee, J. E.; Lyon, L. A.; Garcia, A. J. Biomaterials 2008, 29(35), 4605–4615. (13) Serpe, M. J.; Yarmey, K. A.; Nolan, C. M.; Lyon, L. A. Biomacromolecules 2005, 6(1), 408–413. (14) Kazakov, S.; Kaholek, M.; Kudasheva, D.; Teraoka, I.; Cowman, M. K.; Levon, K. Langmuir 2003, 19(19), 8086–8093. (15) Nerapusri, V.; Keddie, J. L.; Vincent, B.; Bushnak, I. A. Langmuir 2006, 22(11), 5036–5041. (16) Bradley, M.; Liu, D.; Keddie, J. L.; Vincent, B.; Burnett, G. Langmuir 2009, 25(17), 9677–9683. (17) FitzGerald, P. A.; Dupin, D.; Armes, S. P.; Wanless, E. J. Soft Matter 2007, 3(5), 580–586.
14616 DOI: 10.1021/la1023218
Howard et al.
adsorbed amount per area of substrate from the AFM images. The initial adsorbed amount was less than that predicted by the standard random sequential adsorption model (RSA) for hard particle adsorption and was explained by the unexpected deformation of these high-Tg particles due to their strong electrostatic attraction to the solid-liquid interface. The rearrangement and desorption of nearly 75% of the adsorbed microgel particles upon exposure to acid and swelling were attributed to the inability of the nonionic steric stabilizer to anchor the particles sufficiently strongly so as to resist the lateral repulsive forces induced by swelling neighboring particles. The incorporation of a cationic steric stabilizer on PDEA microgel particles was indeed found to more firmly anchor the adsorbed microgel latex particles on mica and silica substrates.18 The far lower Tg of these particles (-5 C) ensured extensive particle flattening and particle coalescence into a genuine featureless film upon adsorption as measured by AFM. This loss of particle integrity, however, prevented ready determination from AFM images of the adsorbed amount in the latex form at pH 8.5 together with the degree of desorption, if any, when the coating swelled upon exposure to acid. In situ AFM force measurement supported the swelling and protonation of the coating at pH 4 and the deswelling upon return to pH 8.5. This prior work has therefore qualitatively demonstrated film formation with loss of particle integrity and retention of pH-responsive microgel character when adsorbed on a solid substrate of opposite charge. Quantitative information on the surface excess, interfacial critical swelling volume, and robust nature of the film are required prior to deployment of this type of coating for application as an active coating. Here we combine the surface sensitive techniques of optical reflectometry (OR) and quartz crystal microbalance (QCM)19 to follow the adsorbed amount and swelling response of this filmforming electrosterically stabilized PDEA microgel.
Materials and Methods The preparation and bulk aqueous solution behavior of the electrosterically stabilized poly[2-(diethylamino)ethyl methacrylate] microgel particles used in this study have been reported previously.9 The lightly cross-linked (1%) PDEA latex was synthesized by aqueous emulsion polymerization. The cationic steric stabilizer, PDMASt, is a styrene-capped poly[2-(dimethylamino)ethyl methacrylate] macromonomer (mean degree of polymerization= 50). These particles will be referred to here as PDMASt-stabilized PDEA microgel or more simply as PDEA microgel. Above pH 6.5-7.0, the microgel is in its nonswollen latex form with a hydrodynamic diameter of 455 ( 10 nm as measured by dynamic light scattering. The mobility (and therefore zeta potential) is positive below the isoelectric point at pH 9.2 and increases with decreasing pH. The microgel has a critical swelling range of pH 7.0 down to 6.5, over which the latex rapidly becomes protonated and swells appreciably. Below this pH, the microgel reaches a diameter of 1230 ( 100 nm and a maximum zeta potential of þ46 ( 2 mV. The size changes that occur during pH cycling are reversible. For these experiments, a 0.01 wt % (100 ppm) microgel particle dispersion was used. At this concentration the dispersions show little turbidity to the naked eye and are therefore suited to in situ optical measurements. Our investigation focused on quantifying the adsorption of the PDEA microgel on silica. The particles were adsorbed at pH 8.5 where they are in the nonswollen latex form, but below the isoelectric point so as to have a positive zeta potential of þ12 mV,9 and thus electrostatically adsorb onto anionic silica substrates. (18) FitzGerald, P. A.; Amalvy, J. I.; Armes, S. P.; Wanless, E. J. Langmuir 2008, 24(18), 10228–10234. (19) Enarsson, L. E.; Wagberg, L. Biomacromolecules 2009, 10(1), 134–141.
Langmuir 2010, 26(18), 14615–14623
Howard et al. The pH of the solution adjacent to the adsorbed microgel layer was then cycled between pH 8.5 ( 0.1 and 4.3 ( 0.3, corresponding to bulk solution conditions of nonswollen and swollen particles, respectively. These pH values were chosen to enable correlation with the previous AFM imaging study which demonstrated formation of genuine uniform films of coalesced particles for this low glass transition temperature latex when adsorbed on silica.18 Millipore water was used for rinsing glassware and preparing solutions. All experiments employed 10-2 M potassium nitrate solutions (Sigma-Aldrich) as background electrolyte. Potassium hydroxide (Merck and Univar) solution was used to adjust the solution to pH 8.5 ( 0.1, Nitric acid (AnalaR and Univar) was used to adjust the solution to pH 4.3 ( 0.3. Optical Reflectometry. In optical reflectometry (OR) the system is arranged such that the polarization of the reflected light changes in proportion to the adsorbed mass of material and is insensitive to changes in conformation of the adsorbed material.20 Therefore, one might expect that OR cannot be employed as a tool to follow conformational changes of an adsorbed microgel. However, the refractive index of the adsorbed microgel is also changed in response to a pH change, and this leads to a signal change that can be followed. A possible complication of using OR to investigate these systems is that when the film is highly swollen and consists largely of aqueous solution, the optical system is no longer in the regime where the change in polarization is linearly related to the adsorbed mass. As we show later, this has to be accounted for. We have also employed QCM to explore the swelling and collapse of the adsorbed microgel as QCM is sensitive to the conformation of the adsorbed film both through the amount of solvent that is coupled to the surface, which influences the resonant frequency, and through changes in the viscosity of the film, which can be followed through the dissipation signal where the dissipation is defined as D = Edissipated/2πEstored, where Edissipated is the energy dissipated during one oscillation and Estored is the energy stored in the oscillating system.21 Additionally, the conformation of the adsorbed polymer layer can be probed through the relationship between dissipation change (ΔD) and frequency change (Δf ).22 Pieces of silicon wafer (50 mm 10 mm, rms roughness 0.6 nm over 10 μm 10 μm) with a 122 or 323 nm oxide layer were used. The 323 nm oxide layer surfaces were made to order by Silicon Valley Microelectronics, Santa Clara, CA. The 122 nm oxide layer surfaces were grown in-house from a silicon wafer in an oxygen atmosphere at 1000 C. As the oxide layer is grown at high temperature it produces surfaces with low hydroxyl group density due to condensation reactions at the silica surface that result in the formation of siloxane bonds.23 The remaining hydroxyl groups are isolated and therefore less likely to participate in hydrogenbonded stabilization of hydronium ions at the surface.23 In solution, these isolated hydroxyl groups are therefore more acidic, and the silica surface will be more highly charged than a silica surface which has geminal hydroxyl groups which can hydrogen bond and thereby prevent ionization. Silica of this type is known as pyrogenic. Pyrogenic silica will slowly rehydroxylate when immersed in water resulting in hydroxylated silica. In this work, the initially pyrogenic silica was cleaned with piranha solution (H2O2 80% w/w; 20% w/w HNO3) for
