10700
Langmuir 2004, 20, 10700-10705
Electroactive Core-Shell Nanocluster Films of Heme Proteins, Polyelectrolytes, and Silica Nanoparticles Hongyun Liu,† James F. Rusling,‡,§ and Naifei Hu*,† Department of Chemistry, Beijing Normal University, Beijing 100875, China, Department of Chemistry, University of Connecticut, U-60, Storrs, Connecticut 06269-3060, and Department of Pharmacology, University of Connecticut Health Center, Farmington, Connecticut 06032 Received May 8, 2004. In Final Form: September 6, 2004 Novel protein core-shell nanocluster films were assembled layer by layer on solid surfaces. In the first step, positively charged heme protein hemoglobin (Hb) or myoglobin (Mb) and negatively charged poly(styrenesulfonate) (PSS) were alternately adsorbed on the surface of SiO2 nanoparticles, forming coreshell SiO2-(protein/PSS)m nanoclusters. In the second step, the SiO2-(protein/PSS)m nanoclusters and polycationic poly(ethylenimine) (PEI) were assembled layer by layer on various solid substrates, forming {[SiO2-(protein/PSS)m]/PEI}n films. Various techniques were used to characterize the nanoclusters and monitor the film growth. {[SiO2-(protein/PSS)m]/PEI}n films at pyrolytic graphite (PG) electrodes exhibited well-defined, chemically reversible cyclic voltammetric reduction-oxidation peaks characteristic of the heme FeIII/FeII redox couples. The proteins in the films retained near native conformations in the medium pH range, and the films catalyzed electrochemical reduction of oxygen and hydrogen peroxide. Advantages of the nanocluster films over the simple {SiO2/protein}n layer-by-layer films include a larger fraction of electroactive protein and higher specific biocatalytic activity. Using this approach, biocatalytic activity can be tailored and controlled by varying the number of bilayers deposited on the nanoparticle cores and the number of nanocluster layers on electrodes.
Introduction Thin films can provide favorable microenvironments for proteins to exchange electrons with underlying electrodes, and can be used as the basis of mediator-less biosensors.1,2 Electrostatic layer-by-layer self-assembly of oppositely charged polyions is a relatively simple and versatile technique adaptable to this purpose.3,4 Stable protein films assembled by this method were constructed and used for direct electrochemistry of proteins,5-11 biosensors,12 and enzyme-catalyzed synthesis.13 Layer-by-layer assembly has been extended to fabricate ultrathin films of proteins and nanoparticles. Hemoglobin * Corresponding author. Fax: +86-10-5880-2075. E-mail:
[email protected]. † Beijing Normal University. ‡ University of Connecticut. § University of Connecticut Health Center. (1) (a) Armstrong, F. A.; Hill, H. A. O.; Walton, N, J. Acc. Chem. Res. 1988, 21, 407. (b) Armstrong, F. A. Bioinorganic Chemistry: Structure and Bonding; Springer-Verlag: Berlin, 1990; Vol. 72, pp 137-236. (c) Chapin, M. F.; Bucke, C. Enzyme Technology; Cambridge University Press: Cambridge, U.K., 1990. (d) Armstrong, F. A.; Wilson, G. S. Electrochim. Acta 2000, 45, 2623. (2) (a) Rusling, J. F. Acc. Chem. Res. 1998, 31, 363. (b) Rusling, J. F. In Protein Architecture: Interfacing Molecular Assemblies and Immobilization Biotechnology; Lvov, Y., Mohwald, H., Eds.; Marcel Dekker: New York, 2000; pp 337-354. (c) Rusling, J. F.; Zhang, Z. In Handbook of Surfaces and Interfaces of Materials, Vol. 5. Biomolecules, Biointerfaces, and Applications; Nalwa, R. W., Ed.; Academic Press: San Diego, CA, 2001; pp 33-71. (d) Hu, N. Pure Appl. Chem. 2001, 73, 1979. (3) (a) Decher, G.; Hong, J. D. Macromol. Symp. 1991, 46, 4232. (b) Decher, G. Science 1997, 277, 1232. (c) Hammond, P. T. Curr. Opin. Colloid Interface Sci. 2000, 4, 430. (4) (a) Lvov, Y. M.; Decher, G. Crystallog. Rep. 1994, 39, 628. (b) Lvov, Y. In Protein Architecture: Interfacing Molecular Assemblies and Immobilization Biotechnology; Lvov, Y., Mohwald, H., Eds.; Marcel Dekker: New York, 2000; pp 125-166. (c) Lvov, Y. In Handbook of Surfaces and Interfaces of Materials, Vol. 3. Nanostructured Materials, Micelles and Colloids; Nalwa, R. W., Ed.; Academic Press: San Diego, CA, 2001; pp 170-189. (5) Lvov, Y.; Lu, Z.; Schenkman, J. B.; Zu, X.; Rusling, J. F. J. Am. Chem. Soc. 1998, 120, 4073.
(Hb), myoglobin (Mb), and horseradish peroxidase (HRP) exhibited well-defined and reversible electrochemical responses at underlying electrodes in films constructed with nanoparticles such as clay,7,8 SiO2,9-11 and MnO2.11 Films were also assembled layer by layer on curved surfaces of particles with diameters from tens of nanometers to a few micrometers, forming a core-shell structure.14 “Core” materials were nanoparticles,15 submicrometer-sized latex beads,16 biocrystals,17 or other colloids. (6) (a) Ma, H.; Hu, N.; Rusling, J. F. Langmuir 2000, 16, 4969. (b) He, P.; Hu, N.; Zhou, G. Biomacromolecules 2002, 3, 139. (c) Wang, L.; Hu, N. Bioelectrochemistry 2001, 53, 205. (d) Li, Z.; Hu, N. J. Colloid Interface Sci. 2002, 254, 257. (7) Zhou, Y.; Li, Z.; Hu, N.; Zeng, Y.; Rusling, J. F. Langmuir 2002, 18, 8573. (8) Li. Z.; Hu, N. J. Electroanal. Chem. 2003, 558, 155. (9) He, P.; Hu, N.; Rusling, J. F. Langmuir 2004, 20, 722. (10) He, P.; Hu, N. Electroanalysis 2004, 16, 1122. (11) Lvov, Y.; Munge, B.; Giraldo, O.; Ichinose, I.; Suib, S.; Rusling, J. F. Langmuir 2000, 16, 8850. (12) (a) Hodak, J.; Etchenique, R.; Calvo, E. J.; Singhal, K.; Bartlett, P. N. Langmuir 1997, 13, 2708. (b) Zhou, L.; Rusling, J. F. Anal. Chem. 2001, 73, 4780. (c) Mugweru, A.; Rusling, J. F. Anal. Chem. 2002, 74, 4044. (d) Zhou, L.; Yang, J.; Estavillo, C.; Stuart, J. D.; Schenkman, J. B.; Rusling, J. F. J. Am. Chem. Soc. 2003, 125, 1431. (e) Dennany, L.; Forster, R. J.; Rusling, J. F. J. Am. Chem. Soc. 2003, 125, 5213. (f) Zhang, X.; Sun, Y.; Shen, J. In Protein Architecture: Interfacing Molecular Assemblies and Immobilization Biotechnology; Lvov, Y., Mohwald, H., Eds.; Marcel Dekker: New York, 2000; pp 229-250. (13) Ariga, K.; Kunitake, T. In Protein Architecture: Interfacing Molecular Assemblies and Immobilization Biotechnology; Lvov, Y., Mohwald, H., Eds.; Marcel Dekker: New York, 2000; pp 169-192. (14) (a) Caruso, F.; Caruso, R. A.; Mohwald, H. Science 1998, 282, 1111. (b) Caruso, F. Chem.sEur. J. 2000, 6, 413. (c) Caruso, F. Adv. Mater. 2001, 13, 11. (15) (a) Uosaki, K.; Kondo, T.; Okamura, M.; Song, W. Faraday Discuss. 2002, 121, 373. (b) Kondo, T.; Okamura, M.; Uosaki, K. Chem. Lett. 2001, 930. (c) Song, W.; Okamura, M.; Kondo, T.; Uosaki, K. J. Electroanal. Chem. 2003, 554/555, 385. (d) Gittins, D. I.; Caruso, F. J. Phys. Chem. B 2001, 105, 6846. (e) Davies, R.; Schurr, G. A.; Meenan, P.; Nelson, R. D.; Bergna, H. E.; Brevett, C. A. S.; Goldbaum, R. H. Adv. Mater. 1998, 10, 1264. (f) Velikov, K. P.; van Blaaderen, A. Langmuir 2001, 17, 4779. (g) Oldenburg, S. J.; Averitt, R. D.; Westcott, S. L.; Halas, N. J. Chem. Phys. Lett. 1998, 288, 243. (h) Pham, T.; Jackson, J.; Halas, N. J.; Lee, T. R. Langmuir 2002, 18, 4915.
10.1021/la0488598 CCC: $27.50 © 2004 American Chemical Society Published on Web 10/19/2004
Electroactive Core-Shell Nanocluster Films
Core-shell nanoclusters containing enzymes were used as bioreactors. The {enzyme/polyion}n multilayers deposited on solid core particles formed a bioactive “shell”, and the total catalytic activity was controlled by varying the thickness of the shell.16e,18-21 We felt that core-shell nanoclusters with redox protein shells assembled on nanoparticle cores could be further assembled layer by layer into porous films on electrode surfaces. The proteins in these novel films should exhibit direct electrochemistry and electrocatalysis. In the present paper, heme proteins (Hb and Mb) were assembled with poly(styrenesulfonate) (PSS) on the surface of SiO2 nanoparticles, forming core-shell SiO2-(protein/PSS)m nanoclusters. These core-shell nanoclusters were then assembled with oppositely charged poly(ethylenimine) (PEI) on electrodes, forming {[SiO2-(protein/PSS)m]/PEI}n films. The films exhibited reversible electrochemistry for protein heme FeIII/FeII. The stable core-shell nanocluster protein films catalyzed the electrochemical reduction of oxygen and hydrogen peroxide. Catalytic activity was controlled by the number of protein layers on the nanoparticle shells and the number of core-shell layers on the electrodes. Experimental Section Chemicals. Bovine hemoglobin (Hb, Mw ) 66 000) and horse heart myoglobin (Mb, Mw ) 17 800) were from Sigma and used as received. Poly(ethylenimine) (PEI, 90%, Mw ∼ 15 000), poly(sodium 4-styrenesulfonate) (PSS, Mw ∼ 70 000), and 3-mercapto1-propanesulfonate (MPS, 90%) were from Aldrich. SiO2 nanoparticles (15 ( 5 nm) were from Zhoushan Nanoparticle Technology. Hydrogen peroxide (H2O2, 30%) was from Beijing Chemical Engineering Plant. Other chemicals were reagent grade. The water used was purified twice successively by ion exchange and distillation. Preparation of Core-Shell SiO2-(Protein/PSS)m Nanoclusters. The layer-by-layer assembly of SiO2-(protein/PSS)m nanoclusters was similar to that described previously.18 Assembly of SiO2-(Hb/PSS)m is described briefly as an example. Hb has net positive charge at pH 5.0 with its isoelectric point at pH 7.422 (the isoelectric point of Mb is pH 6.823), while SiO2 nanoparticles have negative surface charge in pH 5.0 aqueous suspensions with the zero charge point at pH 224 or 3.25 In the first step, 2 mL of 3 mg mL-1 Hb at pH 5.0 and 0.1 M NaAc and 0.1 M KBr as electrolytes were added into a centrifuge tube containing 2 mL of the SiO2 suspension (3 mg mL-1 in pH 5.0 buffer), and Hb was adsorbed onto SiO2 for 30 min with occasional stirring. The mixture was then centrifuged at 6000 rpm for 4 min to separate the Hb-coated SiO2 nanoparticles from the supernatant. The SiO2-Hb particles were redispersed in pH 5.0 buffer and centrifuged, and the supernatant was discarded. This redispersion/centrifugation cycle was repeated two additional times to ensure removal of all unadsorbed Hb from the SiO2-Hb particles. In the second step, 2 mL of 3 mg mL-1 PSS in pH 5.0 buffers was (16) (a) Caruso, F.; Lichtenfeld, H.; Mohwald, H.; Giersig, M. J. Am. Chem. Soc. 1998, 120, 8523. (b) Salgueirino-Maceira, V.; Caruso, F.; Liz-Marzan, L. M. J. Phys. Chem. B 2003, 107, 10990. (c) Khopade, A. J.; Caruso, F. Nano Lett. 2002, 2, 415. (d) Rhodes, K.; Davis, S. A.; Caruso, F.; Zhang, B.; Mann, S. Chem. Mater. 2000, 12, 2832. (e) Caruso, F.; Mohwald, H. J. Am. Chem. Soc. 1999, 121, 6039. (17) (a) Caruso, F.; Trau, D.; Mohwald, H.; Renneberg, R. Langmuir 2000, 16, 1485. (b) Jin, W.; Shi, X.; Caruso, F. J. Am. Chem. Soc. 2001, 123, 8121. (c) Yu, A.; Caruso, F. Anal. Chem. 2003, 75, 3031. (18) Lvov, Y.; Caruso, F. Anal. Chem. 2001, 73, 4212. (19) Caruso, F.; Fiedler, H.; Haage, K. Colloids Surf., A 2000, 169, 287. (20) Caruso, F.; Schuler, C. Langmuir 2000, 16, 9595. (21) Schuler, C.; Caruso, F. Macromol. Rapid Commun. 2000, 21, 750. (22) Matthew, J. B.; Hanania, G. I. H.; Gurd, F. R. N. Biochemistry 1979, 18, 1919. (23) Bellelli, A.; Antonini, G.; Brunori, M.; Springer, B. A.; Sligar, S. J. J. Biol. Chem. 1990, 265, 18898. (24) Franks, G. V. J. Colloid Interface Sci. 2001, 249, 44. (25) Fisher, M. L.; Colic, M.; Rao, M. P.; Lange, F. F. J. Am. Ceram. Soc. 2001, 84, 713.
Langmuir, Vol. 20, No. 24, 2004 10701 Chart 1. Schematic Diagram of the Idealized Structure of Layer-by-Layer {[SiO2-(Hb/PSS)2]/PEI}n Films on a PG/PEI/PSS/PEI Surface
added into the tube and mixed with the SiO2-Hb nanoparticles for 30 min. The negatively charged PSS was adsorbed onto SiO2Hb. A redispersion/centrifugation cycle was repeated for three times to remove the unadsorbed PSS, and the SiO2-(Hb/PSS) nanoclusters were collected. The assembly of a Hb/PSS bilayer shell on the surface of the SiO2 nanoparticle core was repeated until the desired number of Hb/PSS bilayers was achieved. The assembly of SiO2-(Mb/PSS)m nanoclusters was the same as that for Hb except that the original concentration of Mb solution was 2 mg mL-1. Assembly of Layer-by-Layer {[SiO2-(Protein/PSS)m]/ PEI}n Films on Solid Substrates. For the electrochemical studies, basal plane pyrolytic graphite (PG, Advanced Ceramics, geometric area 0.16 cm2) disk electrodes were used as the underlying substrate, and films of core-shell SiO2-(protein/ PSS)m nanoclusters with PEI were then fabricated on the surface. Before assembly, a precursor film of PEI/PSS/PEI was deposited on the PG surface by alternate adsorption of PEI and PSS from their 3 mg mL-1 aqueous solutions containing 0.5 M NaCl. {[SiO2-(protein/PSS)m]/PEI}n films were then fabricated on the surface. Taking Hb as an example, the PG/PEI/PSS/PEI electrode was alternately immersed for 20 min in an aqueous dispersion of negatively charged core-shell SiO2-(Hb/PSS)m nanoclusters and polycationic PEI solutions (containing 0.5 M NaCl) with intermediate water washing. This cycle was repeated until the {[SiO2-(Hb/PSS)m]/PEI}n film was assembled as in Chart 1. Apparatus and Procedures. A CHI 660A electrochemical workstation (CH Instruments) was used for cyclic voltammetry (CV).26 A three-electrode cell was used with a saturated calomel electrode (SCE) as the reference electrode, a platinum wire as the counter electrode, and a PG disk with films as the working electrode. Buffers were purged with high purity nitrogen for at least 20 min prior to a series of experiments, and then, a nitrogen environment was maintained. All experiments were done at ambient temperature (18 ( 2 °C). The electrophoretic mobility of bare SiO2 nanoparticles and the core-shell SiO2-(Hb/PSS)m nanoclusters was measured in pH 5.0 buffers with a Nicomp 380/ZLS zeta-potential analyzer (Particle Sizing Systems). The electrophoretic mobility (u) of the charged particles was determined by measuring the Doppler shift in frequency of light scattered from the moving particles in an applied electric field. The zeta-potential (ξ) was calculated from electrophoretic mobility by using the Smoluchowski relation ξ ) uη/, where η and are the viscosity and permittivity of the solution, respectively. Quartz crystal microbalance (QCM) measurements were done with a CHI 420 electrochemical analyzer. AT-cut quartz resonators (8 MHz) coated by thin gold films (geometric area 0.196 cm2) were used. The gold electrode was pretreated as described previously.9 The Sauerbrey equation27 was used to relate frequency decreases to mass increases. UV-vis absorption spectroscopy was performed with a Cintra 10e spectrophotometer (GBC). (26) Chen, X.; Hu, N.; Zeng, Y.; Rusling. J. F.; Yang, J. Langmuir 1999, 15, 7022. (27) Sauerbrey, G. Z. Phys. 1959, 155, 206.
10702
Langmuir, Vol. 20, No. 24, 2004
Liu et al.
Figure 1. Zeta-potential and electrophoretic mobility of (Hb/ PSS)m multilayers assembled layer by layer on the surface of SiO2 nanoparticles with the adsorption step: (+) pure SiO2 nanoparticles; (b) the Hb adsorption step; (O) the PSS adsorption step.
Results Assembly and Characterization of Core-Shell SiO2-(Protein/PSS)m Nanoclusters. Taking SiO2-(Hb/ PSS)m as an example, the core-shell assembly (see the Experimental Section) was monitored by zeta-potential or electrophoretic mobility (Figure 1). The negatively charged SiO2 nanoparticles showed a zeta-potential at ∼ -10 mV in pH 5.0 buffer. When positively charged Hb was adsorbed on SiO2, the zeta-potential increased to ∼ -5 mV. In the next step, the adsorption of polyanionic PSS resulted in a decrease of zeta-potential to ∼ -13 mV. The zeta-potential was then switched between around -4 and -13 mV when Hb and PSS were alternately adsorbed onto the nanoparticle surface as the outermost layer. When positively charged Hb was the outermost layer, the zeta-potentials were not positive as expected but always negative. A similar situation was observed when (HRP/PSS)m shells were assembled on a PS latex bead surface.20 This could involve incomplete coverage of the protein layer and exposure of some of the underlying PSS. Nevertheless, regular alternating zeta-potentials were observed depending on whether Hb or PSS was the outermost layer, providing qualitative evidence for the stepwise alternate adsorption of Hb and PSS and suggesting multilayer growth of charged macromolecule shells on SiO2 nanoparticle cores. The multilayer growth of the (Hb/PSS)m shell on the SiO2 core was also observed by transmission electron microscopy (TEM) (Supporting Information Figure S1). The uncoated SiO2 nanoparticles showed diameters of 2560 nm, larger than the original 15 nm, indicating a certain degree of aggregation of SiO2 nanoparticles in the dispersions.9 When the (Hb/PSS)2 shell was assembled on the SiO2 surface, the size of particles increased to 50-85 nm in diameter. Compared with bare SiO2 particles, the diameter of the SiO2-(Hb/PSS)2 nanoclusters increased by ∼25 nm, corresponding to a (Hb/PSS)2 shell thickness of ∼12.5 nm. The dimensions of Hb are 5.0 × 5.5 × 6.5 nm3,28 and the PSS layer is likely to be 1-2 nm.29,30 Thus, the predicted thickness of the (Hb/PSS)2 shell is 12-17 nm, consistent with the TEM results. Evidence of the growth of Hb/PSS multilayer shells on the SiO2 nanoparticle surface was also provided by UVvis spectroscopy. In pH 5.0 buffers, Hb displayed an iron (28) Perutz, M.; Muirhead, H.; Cox, J.; Goaman, L.; Mathews, L.; McGandy, E.; Webb, L. Nature 1968, 219, 29. (29) Caruso, F.; Furlong, D. N.; Ariga, K.; Ichinose, I.; Kunitake, T. Langmuir 1998, 14, 4559. (30) Caruso, F.; Lichtenfeld, H.; Donath, E.; Mohwald, H. Macromolecules 1999, 32, 2317.
Figure 2. UV-vis absorption spectra in pH 5.0 buffer for (a) 2 × 10-6 M Hb, (b) 0.5 mg mL-1 PSS, (c) 0.5 mg mL-1 SiO2 nanoparticles, (d) SiO2-(Hb/PSS) nanoclusters, and (e) SiO2(Hb/PSS)2 nanoclusters.
heme Soret absorption band31 at 405 nm (Figure 2a), while PSS and uncoated SiO2 particles showed no peak in this region (Figure 2b and c). The SiO2-(Hb/PSS)m nanoclusters in pH 5.0 dispersions also demonstrated a small Soret band at 405 nm, and the absorbance increased with the number of bilayers (m) (Figure 2d and e), suggesting successful layer-by-layer assembly. The SiO2-(Hb/PSS)m nanoclusters showed Soret bands at the same position as the native Hb in pH 5.0 buffers. Reflectance absorption infrared (RAIR) spectroscopy was used to characterize the protein conformation in the core-shell nanoclusters (Supporting Information Figure S2). SiO2-(Hb/PSS)m nanocluster films with different m values showed the second derivative amide I and II bands in almost the same positions as pure Hb films at 1655 and 1540 cm-1, respectively. Adsorption protein in each bilayer on SiO2 nanoparticles was estimated indirectly by UV-vis spectroscopy. For example, the absorbance of the Soret band of Hb at 405 nm was measured for the initial Hb solution at pH 5.0 before it was mixed with SiO2 nanoparticle dispersions. After adsorption, the Soret band was measured again for the supernatant after centrifugation. In the following washing steps, UV-vis spectra were also measured for supernatants. After three cycles of redispersion/centrifugation steps, the amount of free or unadsorbed Hb in the supernatant could not be detected. The amount of adsorbed Hb on the SiO2 nanoparticle surface was then estimated by subtracting the absorbance of the supernatants at 405 nm from that of the initial Hb solution with volume adjustment. From these data, it was estimated that over 15% of the total Hb in solution was adsorbed onto the SiO2 particles, and the ratio of adsorbed Hb to SiO2 was 0.46 ( 0.11 (mg Hb/mg SiO2). Ignoring the loss of SiO2 nanoparticles during the assembly, the number of Hb molecules adsorbed on the surface of each SiO2 particle was estimated at 21 ( 5, taking into account the Hb molecular weight (66 000), the average diameter of SiO2 nanoparticles (35 nm), and the density of SiO2 (0.223 g cm-3). The results for the second and third bilayer in SiO2(Hb/PSS)m nanoclusters are listed in Supporting Information Table S1, as are the results for the Mb nanoclusters. With the increase of the number of bilayers (m), the diameter of the core-shell nanocluster increased accordingly. (31) Theorell, H.; Ehrenberg, A. Acta Chem. Scand. 1951, 5, 823.
Electroactive Core-Shell Nanocluster Films
Figure 3. QCM frequency shift (∆F) with the adsorption step for (a) {[SiO2-(Hb/PSS)]/PEI}n and (b) {[SiO2-(Hb/PSS)2]/PEI}n films layer-by-layer assembled on a Au/MPS/PEI surface: (b) the PEI adsorption step; (9) the SiO2-(Hb/PSS) adsorption step; (0) the SiO2-(Hb/PSS)2 adsorption step.
Assembly of {[SiO2-(Protein/PSS)m]/PEI}n Films. The overall surface charge of core-shell SiO2-(protein/ PSS)m nanoclusters is negative, since the negatively charged PSS is the outermost layer. Thus, films of the negative nanoclusters and polycationic PEI were assembled (see the Experimental Section). Figure 3 shows the QCM frequency shift (∆F) for the adsorption of {[SiO2-(Hb/PSS)m]/PEI}n films. A roughly linear decrease of frequency with the number of adsorption steps showed that the build-up of {[SiO2-(Hb/PSS)m]/PEI}n films occurred in a reproducible manner. The amounts of SiO2(Hb/PSS)m nanoclusters and PEI adsorbed at each cycle were nearly the same (Supporting Information Table S2). The amount of adsorbed SiO2-(Hb/PSS)2 (m ) 2) was more than twice that of SiO2-(Hb/PSS) with m ) 1. However, the adsorbed amounts of SiO2-(Mb/PSS)2 and SiO2-(Mb/PSS) nanoclusters were similar (Supporting Information Table S2). Regular and reproducible layerby-layer growth of {[SiO2-(Hb/PSS)2]/PEI}n films on the surface of PEI-coated quartz slides was confirmed by UVvis Soret absorption (Supporting Information Figure S3). The assembly of {[SiO2-(protein/PSS)m]/PEI}n multibilayer films on PG/PEI/PSS/PEI surfaces was monitored by cyclic voltammetry (CV). Taking {[SiO2-(Hb/PSS)2]/ PEI}n films as an example, after each adsorption cycle creating a new [SiO2-(Hb/PSS)2]/PEI bilayer, the electrode was washed in water and transferred to pH 7.0 buffer containing no Hb, and CV was performed. Well-defined, nearly reversible CV peaks were observed at midpoint -0.35 V vs SCE (Figure 4A), characteristic of the Hb heme FeIII/FeII redox couples.32 Reduction and oxidation peak currents grew with the number of nanocluster/PEI bilayers (n) until n ) 8. At n > 8, no further increase in the peak currents was observed. Similar results were observed for {[SiO2-(Mb/PSS)2]/PEI}n films (Figure 4B). Cyclic voltammograms of {[SiO2-(protein/PSS)2]/PEI}n films showed roughly symmetric redox peak shapes and nearly equal heights of reduction and oxidation peaks. For a given film, reduction peak current increased linearly with scan rate from 0.05 to 2.0 V s-1. These results suggest diffusionless, surface-confined electrochemical behavior.33 In this case, the integration of the CV reduction peak gave the total charge (Q) passing through the electrode in the reduction of heme FeIII and was used to estimate the surface concentration of electroactive protein (Γ*, mol cm-2).33 Γ* for {[SiO2-(Hb/PSS)2]/PEI}n films increased nonlinearly with n up to 8 (Supporting Information Figure S4A) and then reached a plateau. From the data of the (32) Huang, Q.; Lu, Z.; Rusling, J. F. Langmuir 1996, 12, 5472. (33) Murray, R. W. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1984; Vol. 13, pp 191-368.
Langmuir, Vol. 20, No. 24, 2004 10703
average adsorbed amount of SiO2-(protein/PSS)2 nanoclusters in each bilayer (Supporting Information Table S2) and the weight ratio of adsorbed protein over SiO2 nanoparticles in the nanocluster (Supporting Information Table S1), the surface coverage of adsorbed protein (Γ) in each bilayer was estimated. Thus, the surface concentration in each bilayer for {[SiO2-(protein/PSS)2]/PEI}n films was around 3.0 × 10-11 mol cm-2 for Hb and 2.2 × 10-11 mol cm-2 for Mb. The value of Γ* determined by CV using PG electrodes was compared to the Γ value mainly estimated from QCM. For n < 8, the fraction of electroactive proteins (Γ*/Γ) showed a general decrease with n. At n > 8, additional layers did not add electroactivity and the ratio approached zero (Supporting Information Figure S4A). Mb film electrodes showed similar trends (Supporting Information Figure S4B). The total amount of adsorbed proteins (Γ), surface concentration of electroactive protein (Γ*), and ratio of Γ*/Γ for films are listed in Table 1. In previous work,9,10 heme proteins were assembled layer by layer with SiO2 nanoparticles directly on PG electrodes, and the direct electrochemistry of the proteins in the films was observed. The peak potentials of {[SiO2-(protein/PSS)2]/PEI}n films were very similar to those of corresponding {SiO2/protein}n films at the same assembling pH (Table 1). The CV peak heights are also similar for both films at the steady state. The amount of total adsorbed protein (Γ) and electroactive protein (Γ*) in the {[SiO2-(protein/PSS)m]/PEI}8 films is much less than that in the corresponding {SiO2/protein}6 films. However, the fraction of electroactive protein (Γ*/ Γ) in the core-shell films is larger than that in the {SiO2/ protein}6 films, especially with a large increase found for Mb (Table 1). Stability. Taking {[SiO2-(Hb/PSS)2]/PEI}8 films as an example, after 3 weeks of storage either in buffers or in air, the CV peak potentials of the films on PG showed no change, and the peak heights decreased
