Size-Exclusion Properties of Nanoporous Films Derived from

Dec 10, 2008 - Single-Molecule Tracking Studies of Millimeter-Scale Cylindrical Domain Alignment in Polystyrene–Poly(ethylene oxide) Diblock Copolym...
2 downloads 0 Views 629KB Size
Download PDF
Anal. Chem. 2009, 81, 851–855

Size-Exclusion Properties of Nanoporous Films Derived from Polystyrene-Poly(methylmethacrylate) Diblock Copolymers Assessed Using Direct Electrochemistry of Ferritin Yongxin Li and Takashi Ito* Department of Chemistry, Kansas State University, 111 Willard Hall, Manhattan, Kansas 66506 This paper reports the size-exclusion properties of nanoporous films derived from polystyrene-poly(methylmethacrylate) diblock copolymers (PS-b-PMMA) for biomacromolecules. These properties were assessed by measuring cyclic voltammetry of ferritin (12 nm in diameter) adsorbed onto recessed nanodisk-array gold electrodes (RNEs) fabricated from the nanoporous films having different effective pore diameters and surface functionalities. RNEs having 20-nm-diameter nanopores modified with a poly(ethylene glycol) (PEG) layer showed the redox currents of ferritin after their immersion in a ferritin solution (5 mg/mL) for longer than 2 h. The currents originated from the direct electron transfer reaction of ferritin molecules immobilized on the underlying gold surface as a result of their penetration through the 20nm-diameter nanopores. The PEG modification of the nanopore surface was required for the penetration of ferritin, probably because it reduced the nonspecific adsorption of ferritin to the nanopore surface. In contrast, no redox current of ferritin was observed for RNEs having PEG-modified 15-nm-diameter nanopores after their immersion in the ferritin solution for 12 h, indicating the size-exclusion of ferritin from the 15-nm nanopores. The distinct size-exclusion properties of the PS-b-PMMAderived nanoporous films reflect their uniform diameters and shapes and will provide a means for fabricating separation membranes for biomolecules with high sizebased selectivity. This paper reports the electrochemical assessment of the size-exclusion properties of nanoporous films derived from polystyrene-poly(methylmethacrylate) diblock copolymers (PSb-PMMA) for biomacromolecules. These properties were investigated by measuring cyclic voltammetry (CV) of ferritin molecules (∼12 nm in diameter)1,2 adsorbed onto an underlying gold surface as a result of their penetration through the

nanopores (15∼25 nm in diameter; 30∼40 nm long).3,4 Understanding the size-exclusion properties of the nanopores is essential for developing size-selective separation membranes based on these materials. Nanoporous membranes capable of selectively separating molecules are interesting targets both in fundamental research and real applications.5 An ideal separation membrane contains nanoscale pores whose diameter and surface chemistry can be easily tailored for controlling the separation selectivity based on steric and chemical interactions. In particular, membranes containing nanoscale pores with well-defined structures and monodisperse sizes have made it possible to quantitatively correlate the pore size with the separation performance.6,7 Such membranes that have been employed for separating biomacromolecules or nanoparticles are fabricated from nanoporous anodic alumina membranes,8 track-etched polymer membranes,9-11 carbon nanotube membranes,12,13 and nanoporous silicon membranes.14 Separation of large protein molecules and nanoparticles using these membranes has shown distinct size-based separation selectivity as compared to conventional nanoporous media (e.g., gels) containing heterogeneous pore diameters and structures. Cylinder-forming block copolymers, including PS-b-PMMA, have been employed to fabricate membranes with cylindrical nanopores having narrow diameter distributions.15,16 This approach for fabricating nanoporous membranes has the following unique features. First, the nanoporous structure can be fabricated (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)

* To whom correspondence should be addressed. E-mail: [email protected]. Phone: 785-532-1451. Fax: 785-532-6666. (1) Thiel, E. C. Annu. Rev. Biochem. 1987, 56, 289–315. (2) St. Pierre, T. G.; Chan, P.; Bauchspiess, K. R.; Webb, J.; Betteridge, S.; Walton, S.; Dickson, D. P. E. Coord. Chem. Rev. 1996, 151, 125–143. 10.1021/ac802201w CCC: $40.75  2009 American Chemical Society Published on Web 12/10/2008

(14) (15) (16)

Li, Y.; Maire, H. C.; Ito, T. Langmuir 2007, 23, 12771–12776. Li, Y.; Ito, T. Langmuir 2008, 24, 8959–8963. Davis, M. E. Nature 2002, 417, 813–821. Martin, C. R.; Nishizawa, M.; Jirage, K.; Kang, M.; Lee, S. B. Adv. Mater. 2001, 13, 1351–1362. Baker, L. A.; Jin, P.; Martin, C. R. Crit. Rev. Solid State Mater. Sci. 2005, 30, 183–205. Sano, T.; Iguchi, N.; Iida, K.; Sakamoto, T.; Baba, M.; Kawaura, H. Appl. Phys. Lett. 2003, 83, 4438–4440. Yu, S.; Lee, S. B.; Kang, M.; Martin, C. R. Nano Lett. 2001, 1, 495–498. Chun, K.-Y.; Stroeve, P. Langmuir 2002, 18, 4653–4658. Yu, S.; Lee, S. B.; Martin, C. R. Anal. Chem. 2003, 75, 1239–1244. Hinds, B. J.; Chopra, N.; Rantell, T.; Andrews, R.; Gavalas, V.; Bachas, L. G. Science 2004, 303, 62–65. Holt, J. K.; Park, H. G.; Wang, Y.; Stadermann, M.; Artyukhin, A. B.; Grigoropoulos, C. P.; Noy, A.; Bakajin, O. Science 2006, 312, 1034–1037. Striemer, C. C.; Gaborski, T. R.; McGrath, J. L.; Fauchet, P. M. Nature 2007, 445, 749–753. Hillmyer, M. A. Adv. Polym. Sci. 2005, 190, 137–181. Olson, D. A.; Chen, L.; Hillmyer, M. A. Chem. Mater. 2008, 20, 869–890.

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

851

Scheme 1

via simple processes, including annealing at high temperature to induce cylindrical domain formation via self-assembly and subsequent chemical etching of the domains, without expensive and sophisticated instruments.17,18 Second, the resulting membranes contain arrays of cylindrical nanopores having uniform diameters at high density. Their pore diameter can be tuned by using a diblock copolymer of appropriate molecular weight,15,16 and their surface properties can be controlled via chemical modification of the nanopore surface.4,19-22 These features will allow for designing novel separation membranes. Third, the geometry of the nanoporous monoliths can be controlled flexibly. Nanoporous membranes derived from diblock copolymers were previously employed as templates for nanoparticle deposition23,24 and filters for viruses,25,26 in addition to masks for lithography27 and templates for metal or silica nanowires synthesis.18,28 However, the size-based permeability of these membranes for biomacromolecules, which determine the separation selectivity and efficiency of the membranes, has not been reported so far. This paper reports the electrochemical assessment of the sizeexclusion properties of PS-b-PMMA-derived nanoporous films for a biomacromolecule, ferritin. Ferritin was chosen as a probe molecule, because its size (12 nm in diameter)1,2 is close to the (17) Thurn-Albrecht, T.; Steiner, R.; DeRouchey, J.; Stafford, C. M.; Huang, E.; Bal, M.; Tuominen, M.; Hawker, C. J.; Russell, T. P. Adv. Mater. 2000, 12, 787–791. (18) Thurn-Albrecht, T.; Schotter, J.; Kastle, G. A.; Emley, N.; Shibauchi, T.; Krusin-Elbaum, L.; Guarini, K.; Black, C. T.; Tuominen, M. T.; Russell, T. P. Science 2000, 290, 2126–2129. (19) Zalusky, A. S.; Olayo-Valles, R.; Wolf, J. H.; Hillmyer, M. A. J. Am. Chem. Soc. 2002, 124, 12761–12773. (20) Rzayev, J.; Hillmyer, M. A. J. Am. Chem. Soc. 2005, 127, 13373–13379. (21) Bailey, T. S.; Rzayev, J.; Hillmyer, M. A. Macromolecules 2006, 39, 8772– 8781. (22) Klaikherd, A.; Ghosh, S.; Thayumanavan, S. Macromolecules 2007, 40, 8518–8520. (23) Zhang, Q.; Xu, T.; Butterfield, D.; Misner, M. J.; Ryu, D. Y.; Emrick, T.; Russell, T. P. Nano Lett. 2005, 5, 357–361. (24) Bandyopadhyay, K.; Tan, E.; Ho, L.; Bundick, S.; Baker, S. M.; Niemz, A. Langmuir 2006, 22, 4978–4984. (25) Yang, S. Y.; Ryu, I.; Kim, H. Y.; Kim, J. K.; Jang, S. K.; Russell, T. P. Adv. Mater. 2006, 18, 709–712. (26) Yang, S. Y.; Park, J.; Yoon, J.; Ree, M.; Jang, S. K.; Kim, J. K. Adv. Funct. Mater. 2008, 18, 1371–1377. (27) Park, M.; Harrison, C.; Chaikin, P. M.; Register, R. A.; Adamson, D. H. Science 1997, 276, 1401–1404. (28) Kim, H.-C.; Jia, X.; Stafford, C. M.; Kim, D. H.; McCarthy, T. J.; Tuominen, M.; Hawker, C. J.; Russell, T. P. Adv. Mater. 2001, 13, 795–797. (29) Martin, T. D.; Monheit, S. A.; Niichel, R. J.; Peterson, S. C.; Campbell, C. H.; Zapien, D. C. J. Electroanal. Chem. 1997, 420, 279–290. (30) Cherry, R. J.; Bjornsen, A. J.; Zapien, D. C. Langmuir 1998, 14, 1971– 1973. (31) Pyon, M.-S.; Cherry, R. J.; Bjornsen, A. J.; Zapien, D. C. Langmuir 1999, 15, 7040–7046.

852

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

size of the PS-b-PMMA-derived nanopores employed here (15-25 nm in diameter) and thus would be suitable to examine the sizeexclusion properties of the films. In addition, its redox-active properties29-33 made it possible to electrochemically assess its penetration through the nanoporous film. A thin PS-b-PMMAderived nanoporous film was prepared on a planar gold substrate to give recessed nanodisk-array electrodes (RNEs; Scheme 1).3,4 The penetration of ferritin through the nanoporous films with different pore diameters and surface functionalities was assessed by measuring CVs on the RNEs to demonstrate the size-exclusion properties of the films. If ferritin can penetrate through the nanoporous film and adsorb onto the underlying gold surface, redox peaks will be observed due to its direct electron transfer reaction on the underlying electrode, EXPERIMENTAL SECTION Materials. All solutions were prepared with water having a resistivity of 18 MΩ cm or higher (Barnstead Nanopure Systems). Two types of PS-b-PMMA (57K PS-b-PMMA: Mn ) 39 800 g/mol for PS and 17 000 g/mol for PMMA, Mw/Mn ) 1.06; 71K PS-bPMMA: Mn ) 50 000 g/mol for PS and 21 000 g/mol for PMMA, Mw/Mn ) 1.08) were purchased from Polymer Source and used as received. Potassium dihydrogen phosphate (Fisher Chemical), potassium hydroxide (Fisher Chemical), potassium hydrogen phosphate (Fisher Chemical), ferritin (from equine spleen, Sigma), cysteamine hydrochloride (Fluka), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC; Chem-impex international), N-hydroxysuccinimide (N-HSS; Acros Organics), and tetraethyleneglycol monoamine ((PEO)4NH2; Molecular Biosciences Inc.) were used as received. Gold-coated Si wafers, which were prepared by sputtering 10 nm of Ti followed by 20 nm of Au onto Si(100) wafers, were purchased from LGA Thin Films (Foster City, CA). Electrode Preparation. RNEs based on PS-b-PMMA-derived nanoporous films on gold substrates were prepared according to procedures reported previously.3,4 In short, a PS-b-PMMA film (30-40 nm thick; measured using a J. A. Woollam Alpha-SE spectroscopic ellipsometer) was prepared by spin-coating its toluene solution onto a clean gold-coated Si wafer and then annealed at 170 °C in vacuum (∼0.3 Torr) for g60 h. The cylindrical PMMA domains in the film were removed by UV irradiation (254 nm, 20 mW/cm2, 80 min) under an Ar atmosphere and subsequent rinsing with glacial acetic acid. Chemi(32) Zapien, D. C.; Johnson, M. A. J. Electroanal. Chem. 2000, 494, 114–120. (33) Tominaga, M.; Ohira, A.; Yamaguchi, Y.; Kunitake, M. J. Electroanal. Chem. 2004, 566, 323–329.

Scheme 2

cal modification of the nanopores with poly(ethylene glycol) (PEG) was performed via EDC-mediated amidation of the surface -COOH groups on the nanopores with (PEO)4NH2.4 The gold surface of an RNE was modified with cysteamine by immersion in an aqueous solution of cysteamine hydrochloride (10 mM) for 1.5 h, followed by rinsing with water.34 The cysteamine monolayer will electrostatically capture ferritin molecules (pI ) 4∼5)1 and enhance the direct electron transfer reaction of ferritin.33 Planar gold-coated Si wafers modified with cysteamine were prepared in the same manner for control experiments. Electrochemical Measurements. A cysteamine-modified RNE or planar gold-coated Si wafer was immobilized at the bottom of a cell to serve as the working electrode (Scheme 1).35 The PSb-PMMA-derived nanoporous films after UV irradiation for 80 min were robust enough to be immobilized using an O-ring due to the cross-linking of the PS matrix during the UV irradiation.36 The diameter of the film area exposed to the solution was defined by the O-ring (0.66 cm; A ) 0.34 cm2). Then, a 5 mg/mL ferritin solution (0.1 M KH2PO4-K2HPO4 buffer, pH 7.0, bubbled by Ar) was added into the chamber of the cell. The electrode was exposed to the solution for a certain period of time (20 min, 40 min, 1 h, 2 h, 4 h, or 8 h) under Ar atmosphere and then washed with phosphate buffer (pH 7.0) several times. CVs on the resulting electrode in the buffer were measured in a threeelectrode cell containing a Ag/AgCl (3 M KCl) reference electrode and a Pt counter electrode using a CH Instruments model 618B electrochemical analyzer under Ar. Since the redox reaction of ferritin involves the loading and unloading of iron in the apoferritin shell,30,31,33 we measured CVs at a fixed scan rate (0.1 V/s) and did not measure their scan-rate dependence. RESULTS AND DISCUSSION In this study, CVs were measured on three types of RNEs having different effective pore diameters and surface functionalities in addition to planar gold substrates (Scheme 2). These RNEs showed redox currents of Fe(CN)63-, 1,1′-ferrocenedimethanol, and Ru(NH3)63+, indicating that the nanoporous films are permeable to the small molecules.4 The effective diameters of nanopores in a RNE (deff), which were measured using CVs of (34) Raj, C. R.; Ohsaka, T. Electroanalysis 2002, 14, 679–684. (35) Ito, T.; Audi, A. A.; Dible, G. P. Anal. Chem. 2006, 78, 7048–7053. (36) Jeong, U.; Ryu, D. Y.; Kim, J. K.; Kim, D. H.; Russell, T. P.; Hawker, C. J. Adv. Mater. 2003, 15, 1247–1250.

Figure 1. Multiple potential cycles for ferritin immobilized (a) on a PEG-modified RNE derived from 71K PS-b-PMMA (deff ∼ 20 nm) and (b) on a planar gold electrode (with no nanoporous film). Prior to the measurements, the electrodes were immersed in a 5 mg/mL ferritin solution (0.1 M KH2PO4-K2HPO4 buffer, pH 7.0) for 2 h and then rinsed with the buffer. CVs (scan rate, 0.1 V/s) were measured in 0.1 M KH2PO4-K2HPO4 buffer at pH 7.0. The geometrical electrode area was 0.34 cm2, and the gold surface was modified with cysteamine.

1,1′-ferrocenedimethanol,4 were varied by using PS-b-PMMA of different molecular weights and by PEG-modification of the nanopore surface. The deff was ∼20 nm (derived from 71K PSb-PMMA, with a PEG layer; Scheme 2a), ∼24 nm (derived from 71K PS-b-PMMA, without a PEG layer; Scheme 2b), and ∼15 nm (derived from 57K PS-b-PMMA, with a PEG layer; Scheme 2c).4 The film thickness, measured using spectroscopic ellipsometry prior to the UV irradiation, was ∼30 and ∼40 nm for 57K and 71K PS-b-PMMA-derived films, respectively. These electrodes were immersed in 5 mg/mL ferritin solution for a certain period of time (tim), washed with phosphate buffer, and then employed for CV measurements in the buffer containing no ferritin (Scheme 1). Thus, the redox peaks in the CVs are ascribed to the electrode reactions of ferritin molecules immobilized on the underlying gold surface33 as a result of their permeation through the nanoporous film during immersion. CVs of Ferritin Adsorbed on RNEs Based on PEGModified 20-nm Nanopores. Figure 1a shows CVs on an RNE based on a PEG-modified 20-nm nanoporous film derived from 71K PS-b-PMMA (Scheme 2a) after immersion in a 5 mg/mL ferritin solution for 2 h. A cathodic peak around -0.41 V and an anodic peak around -0.16 V were observed in the first cycle. These redox peaks were due to the direct electron transfer reaction of ferritin immobilized on the underlying electrode surface,29-33 indicating that ferritin could penetrate through the nanopores of deff ∼ 20 nm during the immersion of the RNEs in a ferritin solution (Scheme 2a). In the subsequent cycles, the cathodic peak shifted to a more positive potential (-0.30 V), probably reflecting the reconstruction of the ferritin layer as previously reported on tin-doped indium oxide electrodes.30 In addition, the cathodic current gradually decreased probably due to the release of Fe(II) from the ferritin molecules.30,31,33 A similar Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

853

Figure 3. Cyclic voltammograms on 71K PS-b-PMMA-derived RNEs with and without a PEG layer (deff ∼ 20 and 24 nm, respectively) after immersion in a 5 mg/mL ferritin solution (0.1 M KH2PO4-K2HPO4 buffer, pH 7.0) for 2 h and then rinsed with the buffer. CVs (scan rate, 0.1 V/s) were measured in 0.1 M KH2PO4-K2HPO4 buffer at pH 7.0. The geometrical electrode area was 0.34 cm2, and the gold surface was modified with cysteamine.

Figure 2. Cyclic voltammograms of ferritin immobilized (a) on PEGmodified RNEs derived from 71K PS-b-PMMA (20 nm in effective pore diameter) and (b) on planar gold electrodes at different electrode immersion times. All CVs (scan rate, 0.1 V/s) were measured in 0.1 M KH2PO4-K2HPO4 buffer at pH 7.0. The geometrical electrode area was 0.34 cm2, and the gold surface was modified with cysteamine. (c) Relationship between charge for ferritin reduction and electrode immersion time in a 5 mg/mL ferritin solution on planar gold electrodes (O) and on PEG-modified RNEs derived from 71K PS-b-PMMA (deff ∼ 20 nm; b). The plots and error bars indicate the average and standard deviation, respectively, measured using three separate electrodes.

shift in the cathodic potential and a gradual decrease in the charge were observed on cysteamine-modified planar gold substrates with no nanoporous film (Figure 1b). The slight difference in cathodic potential (e.g., -0.41 V vs -0.40 V in the first cycle for parts a and b of Figure 1, respectively) may reflect a difference in the conformational change of ferritin that occurred during its penetration and/or immobilization processes. In the subsequent sections, CVs in the second potential cycle were discussed rather than those in the first potential cycle, because the latter peak was sometimes observed unclearly due to the large background current at the more negative potential. CVs of Ferritin at Different Immersion Time. The CVs of ferrtin adsorbed on the underlying electrode were measured at different immersion times (tim). Figure 2 shows the CVs on (a) RNEs with 20-nm nanopores and (b) planar electrodes after immersion for different periods of time (10∼ 480 min). Figure 2c summarizes the charge of ferritin reduction at different tim. The redox currents of ferritin were observed at tim g 120 min for the RNEs, whereas it was observed at much shorter tim (g10 min) for the planar electrodes. On the planar electrodes, the cathodic charge gradually increased from 10 to 60 min and then did not change at longer tim (Figure 2c). At tim g 60 min, the electrode surface was fully covered with ferritin, as 854

Analytical Chemistry, Vol. 81, No. 2, January 15, 2009

Figure 4. Cyclic voltammograms on PEG-modified RNEs derived from 57K and 71K PS-b-PMMA (deff ∼ 15 and 20 nm, respectively) after immersion in a 5 mg/mL ferritin solution (0.1 M KH2PO4-K2HPO4 buffer, pH 7.0) for 12 h (57K PS-b-PMMA-derived RNE) and for 2 h (71K PS-b-PMMA-derived RNE) and then rinsed with the buffer. CVs (scan rate, 0.1 V/s) were measured in 0.1 M KH2PO4-K2HPO4 buffer at pH 7.0. The geometrical electrode area was 0.34 cm2, and the gold surface was modified with cysteamine.

supported by the AFM image of the substrate (Supporting Information, Figure S1). The gradual increase in the charge probably reflects the slow adsorption kinetics of ferritin onto the cysteamine-modified electrode surface.33 In contrast, the cathodic charge on the RNEs gradually increased from 1 to 4 h (Figure 2c). The delay of the increase in the charge may originate from several processes, such as the slow penetration of ferritin due to chemical interactions with the nanopore wall, hindered mass transport,37 and/or by the slow kinetics of ferritin adsorption within the restricted nanoporous space. More importantly, the maximum charges observed on these electrodes in Figure 2c permit us to determine the density of ferritin adsorbed onto the RNEs at full coverage. Assuming the planar cysteamine-modified gold surface was completely covered with ferritin upon immersion for 1 h (i.e., 8 × 103 molecules/ µm2;33 Figure S1 in the Supporting Information), the density of ferritin immobilized on the recessed gold surface was calculated from the maximum charge values in Figure 2c. The density was 1500 ± 500 molecules/µm2. Considering the density of the nanopores (∼ 700 pores/µm2),4 two ferritin molecules at the bottom of each nanopore are likely involved in the direct electron transfer reaction with the underlying gold substrate. (37) Deen, W. M. AIChE J. 1987, 33, 1409–1425.

Influence of PEG Modification of PS-b-PMMA-Derived Nanopores on Ferritin Permeation. Figure 3 shows CVs of RNEs based on 71K PS-b-PMMA-derived nanoporous films with and without the PEG modification after immersion in ferritin solution for 2 h. In contrast to the PEG-modified nanopores (Figure 3, gray), the redox peaks of ferritin were not observed on nanopores without a PEG layer (Figure 3, black), despite the larger effective pore diameter of the PEG-free nanopores (deff ∼ 24 nm).4 This result suggests that ferritin could not penetrate through the PEG-free nanopores due to the irreversible nonspecific adsorption onto the nanopore (Scheme 2b). Size-Exclusion of Ferritin by PEG-Modified 15-nm Nanopores. Figure 4 shows CVs on RNEs based on PEG-modified nanopores derived from 71K PS-b-PMMA (deff ∼ 20 nm) and 57K PS-b-PMMA (deff ∼ 15 nm) after immersion in a 5 mg/mL ferritin solution. In contrast to the RNE with 20-nm nanopores (Figure 4, gray), the redox peaks of ferritin were not observed on the RNE with 15-nm nanopores that were treated with the ferritin solution for 12 h (Figure 4, black). The absence of the redox peaks in the latter RNEs indicates that the ferritin molecules could not pass through the nanopores because their effective pore diameter (∼ 15 nm) was too close to the diameter of ferritin (∼12 nm) (Scheme 2c). The uniform diameter of the nanopores (relative standard deviation