14188
Langmuir 2008, 24, 14188-14194
Poly(2-(dimethylamino)ethyl methacrylate)-Modified Nanoporous Colloidal Films with pH and Ion Response Olga Schepelina and Ilya Zharov* Department of Chemistry, UniVersity of Utah, Salt Lake City, Utah 84112 ReceiVed July 30, 2008. ReVised Manuscript ReceiVed September 23, 2008 Poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) brushes were grown using surface-initiated atom transfer radical polymerization on the nanopore surface inside the colloidal films assembled from 255 nm silica spheres. The molecular transport through PDMAEMA-modified colloidal nanopores was studied as a function of pH and ionic strength by measuring the flux of neutral and positively charged redox-active species across the colloidal films using cyclic voltammetry. Nanopores modified with PDMAEMA brushes exhibited pH- and ion-dependent behavior as follows. The diffusion rates decreased with decreasing pH as a result of electrostatic interactions and steric hindrance. At low pH (in the protonated state) the diffusion rates increased with increasing salt concentration because of the charge screening. We also quaternized the surface-grafted PDMAEMA, which led to the formation of strong polyelectrolyte brushes that hindered the diffusion of neutral molecules through the nanopores due to steric effects and the diffusion of positively charged species due to both electrostatic and steric effects.
Introduction “Smart” synthetic nanopores that can mimic the gating function of biological nanopores1 are attracting attention in chemical and biochemical sensors,2 novel medical devices,3 and in the separation of biomacromolecules and pharmaceuticals.4 A variety of techniques has been used to prepare such abiotic nanopores, including soft lithographic techniques,5 embedded carbon nanotubes,6 ion beam etching of silicon nitride and oxide,7 and tracketching of polymeric films.8 Transport in such nanopores has been controlled using grafted macromolecules that respond to environmental stimuli,9 including pH-responsive polypeptides,10,11 light-responsive spirobenzopyran-containing copolymers,12 and ion-responsive polymers.13 Silica colloidal crystals form via self-assembly of nanoscalesized silica spheres into a close-packed face-centered cubic (fcc) lattice with ordered arrays of three-dimensional interconnected nanopores.14 They offer a new approach for the preparation of ion-channel-mimetic nanoporous membranes with facile surface modification and characterization,15 and with highly selective transport. Recently, we showed that transport through the colloidal nanopores can be controlled by their surface modification with * Corresponding author. E-mail address:
[email protected]. (1) Yellen, G. Nature 2002, 419, 35–42. (2) Choi, Y.; Baker, L. A.; Hillebrenner, H.; Martin, C. R. Phys. Chem. Chem. Phys. 2006, 8, 4976–4988. (3) Leoni, L.; Desai, T. A. AdV. Drug DeliVery ReV. 2004, 56, 211–229. (4) Yu, S.; Lee, S. B.; Kang, M.; Martin, C. Nano Lett. 2001, 1, 495–498. (5) Saleh, O. A.; Sohn, L. L. Nano Lett. 2003, 3, 37–38. (6) Sun, L.; Crooks, R. M. J. Am. Chem. Soc. 2000, 122, 12340–12345. (7) Li, J.; Gershow, M.; Stein, D.; Brandin, E.; Golovchenko, J. A. Nat. Mater. 2003, 2, 611–615. (8) Harrell, C. C.; Kohli, P.; Siwy, Z. S.; Martin, C. R. J. Am. Chem. Soc. 2004, 126, 15646–15647. (9) Ito, Y.; Park, Y. S. Polym. AdV. Technol. 2000, 11, 136–144. (10) Smuleac, V.; Butterfield, D. A.; Bhattacharyya, D. Chem. Mater. 2004, 16, 2762–2771. (11) Ito, Y.; Park, Y. S.; Imahishi, Y. Langmuir 2000, 16, 5376–5381. (12) Ito, Y.; Park, Y. S.; Imahishi, Y. Macromolecules 1998, 31, 2606–2610. (13) Yamaguchi, T.; Ito, T. J. Am. Chem. Soc. 2004, 126, 6202–6203. (14) Wong, S.; Kitaev, V.; Ozin, G. A. J. Am. Chem. Soc. 2003, 125, 15589. (15) Ravoo, B. J.; Reinhoudt, D. N.; Onclin, S. Angew. Chem., Int. Ed. 2005, 44, 6282–6304.
amines,16,17 chiral selector moieties,18 and polymer brushes.19,20 Because pH- and ion-responsive polymers allow one to mimic the functions of biological nanopores, we decided to modify the surface of colloidal nanopores with poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA), a well-studied environmentally responsive polymer.21 This polymer presents a challenging system since its behavior is governed by both electrostatic and hydrophobic interactions, as described below. PDMAEMA is a weak cationic polyelectrolyte in aqueous solutions. It has been found that, at low pH, electrostatic repulsions between the protonated tertiary amine groups lead to PDMAEMA swelling.22 In contrast, at high pH, most of the amine groups are deprotonated and neutral. As a result, interactions between nonpolar groups in a polar solvent (hydrophobic interactions) are increased, which leads to a more compact PDMAEMA conformation. In addition, PDMAEMA exhibits ion-responsive behavior. Increasing ionic strength of the solution leads to a collapsed conformation of PDMAEMA at acidic pH as a result of charge screening in the protonated polymer.23 Recently, PDMAEMA brushes were prepared on the surface of silica,24 latex particles,25,26 and on flat silicon substrates27-30 (16) Newton, M. R.; Bohaty, A. K.; White, H. S.; Zharov, I. J. Am. Chem. Soc. 2005, 127, 7268–7269. (17) Newton, M. R.; Bohaty, A. K.; Zhang, Y.; White, H. S.; Zharov, I. Langmuir 2006, 22, 4429–4432. (18) Cichelli, J.; Zharov, I. J. Am. Chem. Soc. 2006, 128, 8130–8131. (19) Schepelina, O.; Zharov, I. Langmuir 2006, 22, 10523–10527. (20) Schepelina, O.; Zharov, I. Langmuir 2007, 23, 12704–12709. (21) Webber, G. B.; Wanless, E. J.; Butun, V.; Armes, S. P.; Biggs, S. Nano Lett. 2002, 2, 1307–1313. (22) Amalvy, J. I.; Wanless, E. J.; Li, Y.; Michailidou, V.; Armes, S. P. Langmuir 2004, 20, 8992–8999. (23) Gao, J.; Zhai, G.; Song, Y.; Jiang, B. J. Appl. Polym. Sci. 2008, 107, 3548–3556. (24) Zhou, L.; Yuan, W.; Yuan, J.; Hong, X. Mater. Lett. 2008, 68, 1372– 1375. (25) Zhang, M.; Liu, L.; Zhao, H.; Yang, Y.; Fu, G.; He, B. J. Colloid Interface Chem. 2006, 301, 85–91. (26) Zhang, M.; Liu, L.; Wu, C.; Fu, G.; Zhao, H.; He, B. Polymer 2007, 48, 1989–1997. (27) Sanjuan, S.; Perrin, P.; Pantoustier, N.; Tran, Y. Langmuir 2007, 23, 5769–5778. (28) Zhang, Q.; Xia, F.; Sun, T.; Song, W.; Zhao, T.; Liu, M.; Jiang, L. Chem. Commun. 2008, 1199–1201. (29) Kusumo, A.; Bombalski, L.; Lin, Q.; Matyjaszewski, K.; Schneider, J. W.; Tilton, R. D. Langmuir 2007, 23, 4448–4454.
10.1021/la802453z CCC: $40.75 2008 American Chemical Society Published on Web 11/17/2008
pH-/Ion-ResponsiVe PDMAEMA-Modified Colloidal Films
by surface-initiated living radical polymerization, and pHresponsive swelling,24-27 protein uptake,29 and wettability28 of PDMAEMA-modified surfaces have been studied. Liu et al.26 used dynamic light scattering (DLS) to show that PDMAEMA brushes are pH- and salt-responsive. The thickness of PDMAEMA brushes decreased with the increase of solution pH or salt concentration. Tran et al.27 used ellipsometry and neutron reflectivity to demonstrate the changes in the PDMAEMA brush thickness in response to changes in pH and salt concentration. They showed that PDMAEMA brush thickness increases significantly with pH changing from neutral to acidic. They also studied ion-responsive behavior of the quaternized PDMAEMA brushes and observed their salt-induced contraction at high salt concentration. Quaternized PDMAEMA brushes were also produced on flat surfaces by Huck et al., who found ion-responsive behavior31 and counterion-dependent wettability32 for those PDMAEMA-modified surfaces. PDMAEMA-modified membranes have been prepared from pure PDMAEMA or PDMAEMA-containing copolymers by phase inversion in a wet process.33-35 PDMAEMA has been also used to produce antibacterial surfaces36,37 and membranes38 as well as for the modification of chromatographic stationary phases.39 Importantly, Bruening and co-workers reported40 the use of layer-by-layer adsorption of macroinitiators on polymeric porous membranes followed by surface-initiated polymerization to produce PDMAEMA brushes inside the pores. Surface-initiated atom transfer radical polymerization (ATRP)41,42 is the most commonly used method for the preparation of well-defined polymer brushes.43-47 Recently, we demonstrated19,20 that surface-initiated ATRP allows one to perform polymerization inside a colloidal crystal and to control its nanopore size by varying the polymer brush thickness and by using a temperature-responsive polymer. In the present work we describe surface-initiated ATRP of 2-(dimethylamino)ethyl methacrylate (DMAEMA) inside the nanopores in the colloidal films assembled from 255 nm silica spheres. Using cyclic voltammetry to measure diffusion through the resulting polymermodified nanoporous films, we demonstrate the ability to control the molecular transport through the nanopores by changing the environmental pH and ionic strength. Cyclic voltammetry has been extensively used to study polyelectrolyte films and brushes immobilized onto gold and platinum electrodes.48 For example, (30) Ayres, N.; Boyes, S. G.; Brittain, W. J. Langmuir 2007, 23, 182–189. (31) Moya, S.; Azzaroni, O.; Huck, W. T. S. Angew. Chem., Int. Ed. 2005, 44, 4578–4581. (32) Azzaroni, O.; Brown, A. A.; Huck, W. T. S. AdV. Mater. 2007, 19, 151– 154. (33) Su, Y.; Li, C. J. Colloid Interface Sci. 2007, 316, 344–349. (34) Su, Y.; Li, C. J. Membr. Sci. 2007, 305, 271–278. (35) Du, R.; Zhao, J. J. Appl. Polym. Sci. 2004, 91, 2721–2728. (36) Lee, S. B.; Koepsel, R. R.; Morley, S. W.; Matyjaszewski, K.; Sun, Y.; Russel, A. J. Biomacromolecules 2004, 5, 887–882. (37) Huang, J.; Murata, H.; Koepsel, R. R.; Russel, A. J.; Matyjaszewski, K. Biomacromolecules 2007, 8, 1396–1399. (38) Zhai, G.; Shi, Z. L.; Kang, En. T.; Neoh, K. G. Macromol. Biosci. 2005, 5, 974–982. (39) Coad, B. R.; Kizhakkedathu, J. N.; Haynes, C. A.; Brooks, D. E. Langmuir 2007, 23, 11791–11803. (40) Jain, P.; Dai, J.; Grajales, S.; Saha, S.; Baker, G. L.; Bruening, M. L. Langmuir 2007, 23, 11360–11365. (41) Wang, J.; Matyjaszewski, K. Macromolecules 1995, 28, 7901–7910. (42) Pyun, J.; Kowalewski, T.; Matyjaszewski, K. Macromol. Rapid Commun. 2003, 24, 1043–1059. (43) Edmondson, S.; Osborne, V. L.; Huck, W. T. S. Chem. Soc. ReV. 2004, 33, 14–22. (44) Advincula, R. A. Polymer Brushes: Synthesis, Characterization, Applications; Wiley-VCH: Verlag, 2004. (45) Ejaz, M.; Tsujii, Y.; Fukuda, T. Polymer 2001, 42, 6811–6815. (46) Huang, X.; Wirth, M. J. Anal. Chem. 1997, 69, 4577–4580. (47) Xiau, D.; Wirth, M. J. Macromolecules 2002, 35, 2919–2925. (48) Choi, E.-Y.; Azzaroni, O.; Cheng, N.; Zhou, F.; Kelby, T.; Huck, W. T. S. Langmuir 2007, 23, 10389–10394.
Langmuir, Vol. 24, No. 24, 2008 14189
Crooks and Bergbreiter49 described electrochemical sensors based on Au electrodes coated with hyperbranched poly(acrylic acid) films modified with β-cyclodextrin moieties and capped with an ultrathin polyamine layer. Cyclic voltammetry of ionic redoxactive probe molecules at different pH values revealed a pHdependent selectivity resulting from the polyamine film grafting.
Experimental Section Materials and Reagents. Copper(I) chloride (anhydrous, 99.99%), copper(II) chloride (99.99%), 1,1,4,7,10,10-hexamethyltriethylenetetramine (HMTETA, 97%), tetraethyl orthosilicate (99.99%), ethyl 2-bromoisobutyrate (98%), 3-amino-propyl triethoxysilane (98%), 2-bromoisobutyryl bromide (98%), and ethyl bromide (98%) were obtained from Aldrich and used as received. The monomer DMAEMA (99%) was obtained from Aldrich and passed through a column of basic alumina before use. Hexaamineruthenium (III) chloride (Ru(NH3)63+; 99%, Strem Chemicals), 1,1′-ferrocenedimethanol (Fc(CH2OH)2; 98%, Aldrich), and potassium chloride (99%, Mallinckrodt) were used as received. Acetonitrile (Mallinckrodt, HPLC grade), triethylamine (98%, J. T. Baker), and dichloromethane (Mallinckrodt, HPLC grade) were distilled from CaH2 before use. Water (18 MΩ · cm) was obtained from a Barnsted “Epure” water purification system. Synthesis of 2-Bromo-2-methyl-N-(3-triethoxysilyl-propyl)propionamide (1). The initiator was synthesized using the previously reported procedure.50 Briefly, 1.5 mL (2.79 g, 12.1 mmol) of 2-bromoisobutyryl bromide were added dropwise to a solution of 2 mL (1.89 g, 8.55 mmol) of 3-aminopropyltriethoxysilane and 2.25 mL (1.64 g, 16.13 mmol) of triethylamine in 50 mL of dry dichloromethane. The solution was stirred for 16 h and the precipitate was filtered off. The solution was washed with dilute HCl and with deionized water, dried with magnesium sulfate, and evaporated to afford 1 in ca. 96% yield. Preparation and Modification of Silica Spheres in Solution. Solution of tetraethoxysilane (TEOS) in absolute ethanol was rapidly poured into a stirred mixture of ammonia and water in absolute ethanol at room temperature. The final concentrations of the reagents were 0.2 M TEOS, 0.4 M ammonia, and 17 M water. After the reaction mixture was stirred for 18 h, the spheres were isolated by repeated centrifugation and resuspention in absolute ethanol. The diameter of the spheres was found to be 255 ( 20 nm using scanning electron microscopy (SEM). Functionalization of silica spheres suspended in acetonitrile was achieved by the treatment with a 1.5-fold excess of 2-bromo-2methyl-N-(3-triethoxysilylpropyl)-propionamide for 18 h under N2 atmosphere at room temperature. The particles were isolated via centrifugation and washed by four cycles of centrifugation and resuspension in acetonitrile in order to remove any adsorbed initiator. The polymerization was performed using the previously reported procedure.37 Initiator-modified silica spheres (400 mg) were placed into a three-necked flask containing 5.05 mL (30 mmol) of the monomer, 11 µL (0.075 mmol) of the sacrificial initiator ethyl 2-bromoisobutyrate, and 5 mL of acetone. The mixture was degassed using three freeze-pump-thaw cycles, stirred and sonicated to obtain a uniform suspension. Next, 0.1 mL (0.36 mmol) of HMTETA, CuCl (29.4 mg, 0.3 mmol), and CuCl2 (8 mg, 0.06 mmol) were added under N2. The mixture was degassed again. The reaction mixture was stirred at room temperature. After 5 h of polymerization, a sample was taken using a syringe. The reaction was stopped by opening the flask to air after 20 h of polymerization. PDMAEMAmodified silica nanoparticles were precipitated into methanol, washed with a large amount of MeOH, distilled water, and acetone by repeated suspension and centrifugation, and dried. After the spheres were collected, the supernatant solutions were combined, and free PDMAEMA formed in solution was isolated by passing the mixture through a silica column and removing the solvent. (49) Dermody, D. L.; Peez, R. F.; Bergbreiter, D. E.; Crooks, R. M. Langmuir 1999, 15, 885–890. (50) Mulvihill, M. J.; Rupert, B. L.; He, R.; Hochbaum, A.; Arnold, J.; Yang, P. J. Am. Chem. Soc. 2005, 127, 16040–16041.
14190 Langmuir, Vol. 24, No. 24, 2008
Schepelina and ZharoV Scheme 1
Quaternization of the surface-grafted PDMAEMA brushes was performed using ethyl bromide according to a previously reported procedure.37 Pt Microdisk Electrodes. Pt microdisk electrodes (25 µm in diameter) shrouded in glass were prepared by first attaching a 1.0mm-diameter Cu wire (Alfa Aesar) to a 25-µm-diameter Pt wire using Ag paint (DuPont). The Pt wire was then flame sealed in a glass capillary; the capillary was bent into a U-shape, and the middle was cut orthogonal to the length of the capillary with a diamond saw to expose the Pt disk. The resulting electrodes were polished with Microcut Paper disks (Buehler), from 240 to 1200 grit in succession, until the surface was free from visible defects. Preparation of Initiator-Modified Colloidal Films. Colloidal films were deposited on the electrode surfaces by placing the electrodes vertically in a 1.5 wt % colloidal solution of 255 nm silica spheres in ethanol and letting the solvent evaporate for 2-3 days in a vibration-free environment. The surface of the silica spheres assembled into colloidal films on the Pt electrodes was modified by immersing the electrodes under nitrogen in dry acetonitrile containing 0.06 M of 2-bromo-2-methylN-(3-triethoxysilylpropyl)propion-amide. The reaction proceeded at room temperature for 18 h. After the modification, the electrodes were soaked and rinsed in dry acetonitrile. Surface-Initiated ATRP of DMAEMA in Colloidal Films. To grow PDMAEMA brushes inside the colloidal films, a DMAEMA polymerization solution was prepared (as described above). The electrodes with initiator-modified colloidal films were immersed into the polymerization mixture by hanging them vertically into the flask under N2 atmosphere. The reaction mixture was stirred at room temperature. The polymerization time was varied from 1 to 20 h. After the reaction, the electrodes were rinsed with MeOH and water. The polymer grown in solution was isolated as described above. Characterization. The molecular flux across the colloidal film was measured voltammetrically using a Par model 175 Universal Programmer and Dagan Cornerstone Chem-Clamp potentiostat. The voltammetric response of the bare, colloidal- and polymer-modified colloidal electrodes was measured in aqueous solutions of either 5.1 mM Ru(NH3)63+ or 1.6 mM Fc(CH2OH)2. Aqueous solutions were prepared using 18 MΩ · cm water and purged with nitrogen to remove dissolved oxygen. The pH of the aqueous solutions was adjusted by the addition of acetic acid and sodium hydroxide. Addition of KCl was used to change the ionic strength of the solutions. DLS, zetapotential analysis (NICOMP 380 ZLS), and SEM (Hitachi S3000N) were employed to perform size and surface charge characterization of polymer-modified silica spheres. Thermogravimetric analysis of polymer-modified silica particles was conducted using TGA Q500 (TA Instruments). Fourier transform infrared (FTIR) spectra were recorded using a Galaxy Series 3000 spectrophotometer. The molecular weight (MW) of PDMAEMA was determined using a gel permeation chromatograph (GPC; Waters, 717 plus) equipped with an HPLC pump (Waters, 515) at flow rate of 1 mL/min tetrahy-
Table 1. Zeta-Potential (ζ), Silica Sphere Diameter (d), and Polymer Brush Thickness (∆r) As a Function of Polymerization Time DLS
SEM
polym. time, h
ζ-potential, mV
d, nm
∆r, nm
d, nm
∆r, nm
0 5 20
-48 +36 +45
287 ( 40 316 ( 42 346 ( 47
0 15 30
255 ( 20 263 ( 25 294 ( 29
0 4 20
drofuran (THF) at 35 °C and four columns (guard, 105 Å, 103 Å, and 100 Å; Polymer Standards Services) in series. A calibration curve based on linear PMMA standards was used in conjunction with a differential refractive index (RI) detector (Waters, 2410).
Results and Discussion Surface Modification of Silica Spheres in Solution. First, we investigated the surface modification of the silica spheres with PDMAEMA in their colloidal solution. We assume that similar processes take place on silica sphere surfaces after their assembly into the colloidal crystal. We synthesized the surface modifying initiator 1 according to the previously reported procedure48 and modified 255 nm diameter silica spheres with 1 in solution (Scheme 1). The resulting modified silica spheres were treated with a DMAEMA solution to form the polymer brushes (Scheme 1). The formation of the polymer brushes was initially confirmed by IR spectroscopy, which after the polymerization showed the characteristic band for the CdO group at 1730 cm-1. Zeta-potential measurements showed a dramatic change in the surface charge of the silica spheres after their modification (Table 1). It increased from -48 mV for the unmodified silica spheres to +36 mV after 5 h of polymerization and to +45 mV after 20 h of polymerization. The charge of the PDMAEMA brushes is expected to be pH-dependent and based on protonation-deprotonation equilibria.21,27 Thus, we performed zeta-potential measurements for the PDMAEMA-modified silica spheres (20 h of polymerization) at varied pH and observed surface charge decreasing to +28 mV as pH was increased from 7 to 10, indicating partial deprotonation of the polymer chains. As pH was brought below 7, the zeta-potential remained approximately the same, indicating that at neutral pH the polymer chains have reached an equilibrium with respect to their protonation. To estimate the degree of protonation of the PDMAEMA chains at neutral pH, we performed the quaternization of the tertiary amine groups of the polymer according to the previously reported procedure.37 We assume that, after the
pH-/Ion-ResponsiVe PDMAEMA-Modified Colloidal Films
Figure 1. TGA of the initiator-modified silica (blue), PDMAEMAmodified silica 5 h (red), PDMAEMA-modified silica 20 h (green), and quaternized PDMAEMA-modified silica (purple).
treatment with ethyl bromide for 24 h, the number of charges on the polymer chains is fixed and is associated with the number of the polymer amine groups. The zeta-potential of the PDMAEMA-modified silica spheres increased from +45 mV to +107 mV after the quaternization, indicating that only half of the amine groups of the original surface-immobilized PDMAEMA chains were protonated at neutral pH. Next, we measured the hydrodynamic diameters of the polymer-modified silica particles (20 h of polymerization) at varied pH. When pH was lowered from neutral to pH ) 3.8, the hydrodynamic diameter increased from 346 ( 47 nm to 370 ( 32 nm. This indicates that PDMAEMA chains became more protonated and attained a more extended conformation. At more acidic pH (2.4) the hydrodynamic diameter increased slightly to 374 ( 33nm. At pH 10 the hydrodynamic diameter decreased to 335 ( 28 nm, indicating a collapsed conformation of the polymer chains due to their deprotonation. Thermogravimetric analysis (Figure 1) revealed that the weight loss for the polymer-modified silica particles is higher by 13% than that for the initiator-modified silica particles, indicating the presence of the polymer. Weight loss for the quaternized PDMAEMA-modified nanoparticles is higher by another 7% compared to the unquaternized polymer-modified particles, indicating an additional increase in the polymer weight after the quaternization. The main weight loss for the polymer-modified silica particles was observed between 250 and 450 °C. To determine the thickness of the polymer brush on the silica spheres as a function of polymerization time, we measured the sphere diameters using DLS and SEM. The results of these measurements are shown in Table 1. The diameter of the unmodified silica spheres determined by DLS is ca. 30 nm larger compared to that determined by SEM. This result is likely due to the fact that DLS measures the hydrodynamic diameter in aqueous solution in contrast to SEM measurements, which are performed in the dry state. For the polymer-modified spheres, the difference between DLS and SEM results is even higher (ca. 50 nm), which is likely the result of the polymer brushes swelling in aqueous solution (as measured by DLS) in contrast to their dry state (as measured by SEM). Overall, the polymer brush thickness determined by DLS is 3 times higher than that measured by SEM for silica particles after 5 h of polymerization, and 1.5 times higher for 20 h of polymerization. These results suggest that the shorter polymer chains can collapse onto the silica surface more efficiently than the longer chains. The polymer thickness measured by DLS of the polymermodified silica spheres in solution allows estimating the molecular weight of the PDMAEMA chains formed on the silica surface, assuming that the chains are fully extended42 in water and taking the length of the monomeric unit to be ca. 0.25 nm. For polymer brushes obtained after 5 h of polymerization, the MW can be
Langmuir, Vol. 24, No. 24, 2008 14191
Figure 2. A representative GPC trace for PDMAEMA grown in solution for 20 h. Mn ) 11 800; Mw ) 13 600; PDI ) 1.15.
Figure 3. SEM image of the colloidal film assembled from 255 nm silica spheres, after surface modification with initiator moieties and ATRP polymerization of DMAEMA for 20 h. Size bar ) 2 µm.
estimated as ca. 9400 Da, while, for those obtained after 20 h, it is ca. 19 600 Da. The surface-initiated polymerization of DMAEMA was carried out in the presence of a small amount of the sacrificial initiator. This resulted in the formation of a free polymer, which was isolated and analyzed by GPC to estimate the molecular weight of the polymer chains formed on the surface. Indeed, it has been reported that the chains formed at the surface and those formed in solution have similar molecular weight characteristics.43,47,51 Figure 2 shows a representative GPC trace for PDMAEMA after 20 h polymerization with MW of 13,600 Da. Polymerization Inside the Colloidal Nanopores. In order to determine whether the colloidal crystal lattice would remain unperturbed by the surface-initiated polymerization of DMAEMA, we performed ATRP for colloidal films assembled on glass slides using 1.5 wt% solution 255 nm silica spheres. The surface of the silica spheres was modified with the initiator moieties (Scheme 1), and polymerization of DMAEMA was performed for 20 h. The SEM image of the resulting hybrid PDMAEMA/ colloidal film is shown in Figure 3. It is clear that the crystalline lattice remained intact. It is also apparent from Figure 3 that no thick polymer layer is formed over the colloidal film, and that the colloidal nanopores are still present. Unfortunately, SEM images do not allow directly observing the polymer brushes inside the colloidal nanopores. Assuming the polymer growth rate inside the nanopore to be approximately half of that on the silica spheres in the colloidal solution,19 the polymer brush thickness inside the nanopores after 20 of polymerization may be estimated as 10-15 nm, thus filling most of the void inside the colloidal crystal. Indeed, the size of the nanopores, i.e., the distance from their center to the nearby silica sphere surface for the colloidal films comprised of 255 nm spheres is 19 nm (15% of the sphere radius19). We found the PDMAEMA-modified colloidal films to be significantly more robust compared to the unmodified colloidal films, as they did not form cracks and adhered well to the glass surface. (51) Husseman, M.; Malmstroem, E. E.; McNamara, M.; Mate, M.; Mecerreyes, D.; Benoit, D. G.; Hedrickthat, J. L.; Mansky, P.; Huang, E.; Russel, T. P.; Hawker, C. J. Macromolecules 1999, 32, 1424–1431.
14192 Langmuir, Vol. 24, No. 24, 2008
Schepelina and ZharoV
Figure 4. Representative voltammetric responses for PDMAEMA-colloidal film Pt electrodes (20 h polymerization) at different pH for Ru(NH3)63+ (A) and for Fc(CH2OH)2 (B). Voltammograms recorded above pH 5 are shown in blue, and those below pH 4 are shown in red.
Figure 5. Limiting current (Ru(NH3)63+) as a function of increasing pH for PDMAEMA-colloidal film Pt electrodes for 5 h (A) and for 20 h (B) polymerizations.
Figure 6. Voltammetric responses of the protonated PDMAEMAmodified colloidal film Pt electrodes for Ru(NH3)63+ reduction in water as a function of KCl concentration. KCl concentrations are shown adjacent to each voltammogram.
Next, we assembled the colloidal films on the surface of Pt microelectrodes shrouded in glass, modified film surfaces with initiator moieties, and performed the polymerization of DMAEMA on the electrodes, while removing them from solution after different periods of time. We then measured the limiting currents of Ru(NH3)63+ and Fc(CH2OH)2 at pH ) 7 for these electrodes and compared the limiting currents before and after the polymerization. The limiting currents measured for the electrodes after 20 h of the polymerization decreased by 30-40% for Ru(NH3)63+ and only by ∼10% for Fc(CH2OH)2. This suggests that a polymer brush has been formed inside the nanopores. The difference in the limiting current of Ru(NH3)63+ and Fc(CH2OH)2 at neutral pH may be explained by the fact that the polymer is partially positively charged, as shown above, so diffusion of positively charged species Ru(NH3)63+ is hindered as a result of both electrostatic and steric interactions, while the diffusion of the neutral Fc(CH2OH)2 is hindered only sterically. The chain conformation of polyelectrolyte PDMAEMA is governed by the electrostatic interactions between the charged monomer units,27 and, at neutral pH, the polymer chains may attain a partially
collapsed conformation. This would explain the relatively small decrease in diffusion rate for Fc(CH2OH)2. Thus, it is hard to estimate the PDMAEMA chain length inside the colloidal nanopores using cyclic voltammetry as reported by us earlier for neutral polymers.19,20 However, we have previously demonstrated that the change in the diffusion rates is associated with the growth of the polymer brushes inside the colloidal nanopores and that the chains tend to be shorter compared to those grown on silica spheres in colloidal solution,19,20 which likely results from a slower diffusion of the monomer to the silica sphere surface inside the nanopores. pH-Responsive Behavior of the PDMAEMA-Modified Colloidal Films. Next, we studied the pH response for PDMAEMA-modified nanoporous colloidal films using positively charged Ru(NH3)63+. As can be clearly seen in Figure 4A, the limiting current for PDMAEMA-modified colloidal film electrodes is highly dependent on pH. This effect is summarized in Figure 5. The limiting current increases with increasing pH by ∼80% with an abrupt change at pH ∼ 4-5. This result is consistent with the previous reports for surface-grafted PDMAEMA brushes,24-27 as discussed above. According to the previous reports,52,53 PDMAEMA is a weak polybase with a pKa of about 7.0 in aqueous solution, which is significantly higher than the pKa of 4-5 we found for the surface-grafted PDMAEMA. This difference can be explained by the fact that the surface-bound chains are located in close proximity to each other, and electrostatic repulsions between the neighboring groups result in their deprotonation at lower pH. The polymer brushes’ behavior inside the nanopores is also different compared to that on the surface of the silica particles in colloidal solution. According to (52) van de Wetering, P.; Zuidam, N. J.; van Steenbergen, M. J.; van der Houwen, O. A. G. J.; Underberg, W. J. M.; Hennink, W. E. Macromolecules 1998, 31, 8063–8068. (53) Liu, S.; Weaver, J. V. M.; Tang, Y.; Billingham, N. C.; Armes, S. P.; Tribe, K. Macromolecules 2002, 35, 6121–6131.
pH-/Ion-ResponsiVe PDMAEMA-Modified Colloidal Films
Langmuir, Vol. 24, No. 24, 2008 14193
Figure 7. (A) Limiting current (Ru(NH3)63+) at pH 2 (red) and at pH 7 (blue). (B) Limiting current (Ru(NH3)63+) at 0.05 M KCl (red) and at 0 M KCl at pH ) 2 for PDMAEMA-modified colloidal film Pt electrodes after polymerization for 5 h.
Figure 8. Representative voltammetric responses of Ru(NH3)63+ (A) and Fc(CH2OH)2 (B) of PDMAEMA-colloidal film Pt electrodes (5 h polymerization) before (blue) and after (red) quaternization and after quaternization in the presence of 0.5 M KCl (green).
our zeta-potential measurements, the PDMAEMA brush on silica particles is deprotonated above pH ) 7. However, the polymer brush inside the nanopores becomes deprotonated at lower pH (4-5). This may be due to the surface of the nanopores being more crowded with polymer chains than the surface of silica particles in solution, and, as a result of neighboring chains’ interpenetration and electrostatic repulsion, the deprotonation occurs at lower pH. For polyelectrolyte brushes, the chain conformation is governed by the electrostatic interactions between the charged monomer units.27 At high pH, the amine groups of the polymer chains are deprotonated, and the polymer is considered to be neutral. As a result, the polymer chains tend to attain a collapsed conformation due to the hydrophobic interactions, and the diffusion of the positively charged Ru(NH3)63+ is not sterically hindered by the polymer brush, nor it is repelled electrostatically. In contrast, at low pH, the polymer brush becomes protonated and stretches away from the surface as a result of electrostatic repulsions between the charged monomer units and between the polymer chains. The diffusion of Ru(NH3)63+ in this case is blocked as a result of both electrostatic repulsion from the positively charged polymer chains and as a result of the steric hindrance. To isolate the pH effect on the polymer conformation, we examined the diffusion of a neutral redox-active molecule, Fc(CH2OH)2, assuming that it does not electrostatically interact with the polymer chains. The limiting current of Fc(CH2OH)2 decreased only by ∼30% (Figure 4B) with no abrupt change when pH was lowed from neutral to acidic, which should result exclusively from the conformational changes in the PDMAEMA chains. Ion-Responsive Behavior of the PDMAEMA-Modified Colloidal Films. Next, we studied the influence of the solution ionic strength on the diffusion across the protonated PDMAEMAmodified colloidal films by measuring the limiting current of Ru(NH3)63+ and Fc(CH2OH)2 as a function of KCl concentration. Figure 6 shows the dependence of the limiting current of
Ru(NH3)63+ on the salt concentration at pH 2.4 where the PDMAEMA chains are protonated. The limiting current increases with increasing KCl concentration. The addition of KCl progressively screens the charge within the polymer brush. As a result, the diffusion of Ru(NH3)63+ becomes easier. Similar results were observed for the colloidal films obtained after both 5 and 20 h polymerization. These results are in good agreement with previously reported studies for PDMAEMA-modified surfaces.24-27 We did not observe any significant effect of the salt concentration on the diffusion of Fc(CH2OH)2 at low pH, which suggests that, under these conditions, the conformation of the PDMAEMA chains is not significantly affected by the salt concentration.25,26 Both pH- and ion-dependent changes in limiting current were reversible (Figure 7). The nanoporous PDMAEMA-modified colloidal films can be cycled between the low and high pH and between the low and high ionic strength regimes without apparent loss of responsiveness. In order to confirm that only the polymer formed inside the nanopores affects the limiting current, a bare Pt microelectrode shrouded in glass was treated with initiator-moieties, and DMAEMA was polymerized for 20 h. The resulting electrodes did not show a pH-responsive behavior. Quaternization of the PDMAEMA-Modified Colloidal Films. To demonstrate that the “weak” polyelectrolyte PDMAEMA can be converted into a “strong” polyelectrolyte brush with a fixed, pH-independent number of charges, we performed the quaternization of the polymer by treating the PDMAEMAcolloidal film electrodes with ethyl bromide.37 We then measured the limiting current of Ru(NH3)63+ and Fc(CH2OH)2 using these quaternized electrodes at neutral pH (Figure 8). The limiting current of both neutral and positively charged species decreased significantly after the quaternization. This suggests that the polymer brush carrying a large number of quaternary ammonium ions blocks the diffusion of ions and molecules almost completely, regardless of pH as a result of
14194 Langmuir, Vol. 24, No. 24, 2008
both strong electrostatic repulsion and steric hindrance. Next, we measured the limiting current for Ru(NH3)63+ and Fc(CH2OH)2 at 0.5 M concentration of KCl. We did not observe any significant change in the limiting current for Ru(NH3)63+ with increased ionic strength (Figure 8A), indicating that the positive charge of the quaternized polymer brush was not significantly screened under these conditions. At the same time, for Fc(CH2OH)2, an ∼10% increase in the limiting current was observed (Figure 8B), suggesting small conformational changes in the polymer chains. This result is in agreement with the previously reported data,27 which demonstrated that, at moderate salt concentration, the thickness of strong polyelectrolyte brushes does not strongly depend on the salt concentration due to the osmotic pressure of the counterions trapped inside the brush (osmotic brush regime).
Schepelina and ZharoV
performed the quaternization of the surface-grafted PDMAEMA, which led to the formation of strong polyelectrolyte brushes that hinder the diffusion of neutral molecules through the nanopores due to steric effects and the diffusion of the positively charged species due to both electrostatic and steric effects. Similar responsive behavior has been observed for PDMAEMA-containing ultrafiltration membranes prepared by phase inversion method,33,34 where the flux of water was controlled using pH and ionic strength. However, the water flux for those membranes was not reversible. To the best of our knowledge, our observations are the first example of the pH- and ionic strength-controlled diffusion of small molecules through a nanoporous membrane modified with PDMAEMA brushes. We believe that PDMAEMA-modified colloidal films can be potentially used as ion-channel-mimetic nanoporous membranes in separations and sensing.
Conclusions We performed surface-initiated ATRP of DMAEMA inside the nanopores of the silica colloidal films and studied the molecular transport across the resulting PDMAEMA-modified nanoporous films as a function of pH and salt concentration for positively charged and neutral redox-active species. We found that the transport through the nanopores modified with PDMAEMA brushes exhibits a complex pH- and ion-responsive behavior. The diffusion rates decrease with decreasing pH, and at low pH they increase with increasing salt concentration. We also
Acknowledgment. This work was supported by the NSF CAREER Award (CHE-0642615) and NSF Grant CHE-0616505. I.Z. is grateful to the Camille and Henry Dreyfus Foundation for a New Faculty Award. We thank Professor Chuck White and Ms. Jeramie Jergins (Chemistry Department, University of Utah) for their assistance with TGA measurements, and Dr. Andrew Bohaty (Department of Chemistry, Carnegie Mellon University) for his help with GPC measurements. LA802453Z
