Morphology-Dependent Electrochemistry and Electrocatalytical

Aug 1, 2007 - ... of electrode reactions, kinetic parameters, the formal potentials (E0'), and charge-transport ... Chemical Reviews 2017 117 (21), 13...
0 downloads 0 Views 284KB Size
Download PDF
Langmuir 2007, 23, 9487-9494

9487

Morphology-Dependent Electrochemistry and Electrocatalytical Activity of Cytochrome c Haiqing Liu, Yang Tian,* and Zifeng Deng Department of Chemistry, Tongji UniVersity, Siping Road 1239, Shanghai 200092, P. R. China ReceiVed March 21, 2007. In Final Form: June 7, 2007 The morphology-dependent electrochemistry and electrocatalytical activity of cytochrome c (cyt. c) were investigated at pyramidal, rodlike, and spherical gold nanostructures directly electrodeposited onto sputtered gold surfaces. Direct, reversible electron transfer of cyt. c, for the first time, was realized at nanorod-like and nanopyramidal gold surfaces without any mediators or promoters, while no redox reaction was observed at the nanospherical gold electrode. The electrochemical properties of cyt. c vary with the shape of gold nanostructures with respect to the reversibility of electrode reactions, kinetic parameters, the formal potentials (E0′), and charge-transport resistance (Rct), suggesting shape-dependent mechanisms for the electrode reactions of cyt. c. The experimental results manifest that cyt. c was stably immobilized on the nanostructured gold electrodes with different conformational changes of the heme microenvironment. Consequently, not only the electroactivity, but also the inherent biological activity of the immobilized cyt. c strongly depended on the shape of the electrode surfaces. The facilitated electron transfer combined with the intrinsic catalytical activity of cyt. c substantially constructed a third-generation H2O2 biosensor with high selectivity, quick response time, large linear range, and good sensitivity. The electrocatalytical activity of the immobilized cyt. c toward H2O2 was also found to be morphology dependent, and the linear range of H2O2 detection could be tuned by means of employing the nanostructured gold surfaces with different shapes.

Introduction Redox reactions play an important role, not only in physiological reactions and biotechnology, but also in the detection of biochemical species by transduction into an electrical signal in a biosensor or into electrical current in a biofuel cell.1-4 So, the routing of electrons from redox proteins (particularly redox enzymes, which are nonconductive) to electron conductors (i.e., electrodes) is the subject of extensive research. The approach to direct electron transfer requires interfaces that exhibit fast electrontransfer kinetics with biocompatibility, namely, without denaturation. Since direct contact between redox enzymes and metal electrode surfaces leads to significant structural and/or functional changes of enzymes,5,6 in the past couple of decades, a burst of research activity has been directed toward the modification of electrodes or proteins for creating accessible electron transfer interfaces. Numerous applications of chemically or nonchemically modified methods to the direct electron transfer of cytochrome c (cyt. c)7,8 and other redox proteins have been exploited.9,10 * To whom correspondence should be addressed. E-mail: yangtian@ mail.tongji.edu.cn. Tel: +86-21-65987075; Fax: +86-21-65982287. (1) Bard, A. J., Stratmann, M., Eds. Encyclopedia of Electrochemistry; WileyVCH: Weinheim, Germany, 2002; Vol. 9. (2) Barton, S. C.; Gallaway, J.; Atanassov, P. Chem. ReV. 2004, 104, 48674886. (3) Willner, I.; Katz, E. Angew. Chem. 2000, 112, 1230-1269; Angew. Chem. Int. Ed. 2000, 39, 1181-1218. (4) Katz, E.; Shipway, A. N.; Willner, I. Biofuel cells: Function, design and operation. In Handbook of Fuel Cell Technology; Vielstich, W., Gasteiger, H., Lamm, A., Eds.; Wiley: New York, 2002. (5) Holt, R. E.; Cotton, T. M. J. Am. Chem. Soc. 1989, 111, 2815-2821. (6) Yang, M.; Chung, F. L.; Thompson, M. Anal. Chem. 1993, 65, 37133716. (7) (a) Eddows, M. J.; Hill, H. A. O. J. Chem. Soc., Chem. Commun. 1977, 21, 771-772. (b) Eddows, M. J.; Hill, H. A. O. J. Am. Chem. Soc. 1979, 101, 4461-4464. (c) Gleria, K. D.; Hill, H. A. O.; Lowe, V. J.; Page, D. J. J. Electroanal. Chem. 1986, 213, 333-338. (d) Oliver, B. N.; Egekeze, J. O.; Murray, R. W. J. Am. Chem. Soc. 1988, 110, 2321-2322. (e) Taniguchi, I.; Iseki, M.; Toyosawa, K.; Yamaguchi, I.; Yasukouchi, K. J. Electroanal. Chem. 1984, 164, 385-391. (f) Taniguchi, I.; Iseki, M.; Yamaguchi, I.; Yasukouchi, K. J. Electroanal. Chem. 1984, 175, 341-348. (g) Taniguchi, I.; Higo, N.; Umetika, K.; Yasukouchi, K. J. Electroanal. Chem. 1986, 206, 341-348.

With the development of nanomaterials, gold nanoparticles (AuNPs), which have been known for 2500 years, are the subject of an exponentially increasing number of publications, because of their fascinating research and promises for optical, electronic, magnetic, catalytic, and biomedical applications.11,12 More importantly, additional reasons for the present excitement in AuNPs research are the complete biocompatibility and stability of AuNPs. Actually, direct contact between naked AuNPs and proteins is common in histochemistry, where it has been demonstrated that electrostatically bound AuNPs-protein conjugates typically retain biological activity.13 Indeed, Crumbliss’ (8) (a) Wang, J. X.; Li, M. X.; Shi, Z, J.; Li, N. Q.; Gu, Z. N. Anal. Chem. 2002, 74, 1993-1997. (b) Jiang, X.; Shang, L.; Wang, Y.; Dong, S. Biomacromolecules 2005, 6, 3030-3036. (c) Yu, J.; Ju, H. Anal. Chem. 2002, 74, 35793583. (d) Wang, L.; Wang, E. Electrochem. Commun. 2004, 6, 49-54. (9) (a) Tian, Y.; Shioda, M.; Kasahara, S.; Okajima, T.; Mao, L.; Hisabori, T.; Ohsaka, T. Biochim. Biophys. Acta 2002, 1569, 151-158. (b) Tian, Y.; Mao, L.; Okajima, T.; Ohsaka, T. Anal. Chem. 2002, 74, 2428-2432. (c) Ohsaka, T.; Tian, Y.; Shioda, M.; Kasahara, S.; Okajima, T. Chem. Commun. 2002, 990-991. (d) Tian, Y.; Mao, L.; Okajima, T.; Ohsaka, T. Anal. Chem. 2004, 76, 41624168. (e) Tian, Y.; Ariga, T.; Takashima, N.; Okajima, T.; Mao, L.; Ohsaka, T. Electrochem. Commun. 2004, 6, 609-614. (10) (a) Heller, A. Acc. Chem. Res. 1990, 23, 128-134. (b) Hamachi, I.; Noda, S.; Kunitake, T. J. Am. Chem. Soc. 1991, 113, 9625-9630. (c) Rusling, J. F.; Nassar, A. E. F. J. Am. Chem. Soc. 1993, 115, 11891-11897. (d) Chen, H.; Ju, H.; Xun, Y. Anal. Chem. 1994, 66, 4538-4542. (e) Xiao, Y.; Patolsky, F.; Katz, E.; Hainfeld, J. F.; Willner, I. Science 2003, 299, 1877-1881. (11) (a) Kim, F.; Song, J. H.; Yang, P. J. Am. Chem. Soc. 2002, 124, 1431614317. (b) Hutchinson, Y. O.; Liu, Y.; Kiely, C.; Kiely, C. J.; Brust, M. AdV. Mater. 2001, 13, 1800-1803. (c) Sun, Y.; Xia, Y. AdV. Mater. 2003, 15, 695699. (d) Zhang, J.; Du, J.; Han, B.; Liu, Z.; Jiang, T.; Zhang, Z. Angew. Chem., Int. Ed. 2006, 45, 1116-1119. (e) Kim, F.; Connor, S.; Song, H.; Kuykendall, T.; Yang, P. Angew. Chem., Int. Ed. 2004, 43, 3673-3677. (f) Chen, Y.; Gu, X.; Nie, C.; Jiang, Z.; Xie, Z.; Lin, C. Chem. Commun. 2005, 4181-4183. (g) Sun, Y.; Xia, Y. Science 2002, 298, 2176-2179. (h) Shankar, S. S.; Rai, A.; Ankamwar, B.; Singh, A.; Ahmad, A.; Sastry, M. Nat. Mater. 2004, 3, 482-488. (i) Millstone, J. E.; Park, S.; Shuford, K. L.; Qin, L.; Schatz, G. C.; Mirkin, C. A. J. Am. Chem. Soc. 2005, 127, 5312-5313. (12) (a) Pendry, J. B. Science 1999, 285, 1687-1688. (b) Daniel, M.; Astruc, D. Chem. ReV. 2004, 104, 293-346. (c) Tian, Y.; Tatsuma, T. Chem. Commun. 2004, 1810-1811. (d) Tian, Y.; Tatsuma, T. J. Am. Chem. Soc. 2005, 127, 7632-7637. (13) (a) Geoghegan, W. D.; Ackerman, G. A. J. Histochem. Cytochem. 1977, 25, 1187-1200. (b) Bendayan, M. In Colloidal Gold: Principles, Methods and Applications; Hayat, M. A., Ed.; Academic Press: San Diego, CA, 1989; Vol. 2, Chapter 2.

10.1021/la700817y CCC: $37.00 © 2007 American Chemical Society Published on Web 08/01/2007

9488 Langmuir, Vol. 23, No. 18, 2007

group found that several enzymes could maintain their biocatalytical and electrochemical activities when immobilized on AuNPs.14 Natan and his co-workers realized the reversible electrochemistry of cyt. c with 12 nm-diameter AuNPs and suggested that the size of AuNPs plays an important role in protein electrochemistry.15 Cyt. c, which has the special biological function of transferring electrons between membrane-bound enzyme complexes of cyt. c reductase and cyt. c oxidase, is an excellent model for studying the electron transfer of typical metalloproteins from a structural point of view.16 In the present work, it is the first time that the shape-dependent electrochemistry of cyt. c and its electrocatalytical activity toward hydrogen peroxide (H2O2) were investigated at pyramidal, rodlike, and spherical gold nanostructures directly electrodeposited onto sputtered gold substrates. In our previous work, nanopyramidal, nanorod-like, and nanospherical gold surfaces were fabricated on polycrystalline gold substrates through electrochemical overpotential deposition by easily manipulating the deposited potentials and concentrations of HAuCl4.17 Here, direct, reversible electron transfer of cyt. c is clearly observed at pyramidal and rodlike nanostructured surfaces, while it is not obtained at nanospherical surfaces. Besides the size of AuNPs,15 the shape of gold nanostructures plays a key role in protein electrochemistry, and experimental results demonstrate that cyt. c strongly confines onto nanostructured gold surfaces with different conformational changes. In addition, cyt. c processes the inherent enzymatic activity after being immobilized on nanorod-like and nanopyramidal gold surfaces, which also enables the direct electron transfer of cyt. c itself. Consequently, these cyt. c-pyramidal and -rodlike gold nanocomposites are successfully constructed for third-generation H2O2 biosensors. The electrocatalytical activity of cyt. c toward H2O2 also depends on the shape of the gold nanostructures. Experimental Section Chemicals and Materials. Hydrogen tetrachloroaurate(III) (HAuCl4) trihydrate was purchased from Aldrich (St. Louis, MO) and used as supplied. Horse heart cyt. c (MW 13 000) was obtained from Sigma and used without further purification. Phosphate-buffered solution (PBS, 25 mM) was prepared by mixed stock standard K2HPO4 and KH2PO4 solutions, and the pH of the PBS was adjusted by a pH meter. Other reagents were of analytical grade and used as received. Indium tin oxide (ITO)-coated glass plates with a square resistance of 10-20 Ω cm-2 were obtained from Asahi Glass (Japan). All the solutions were prepared with Milli-Q water and were deaerated with high purity nitrogen before experiments. All electrochemical experiments were carried out at room temperature. Preparation of Gold Nanostructures and Modification. ITOcoated glass plates were thoroughly cleaned by sonication for 30 min in the following solvents successively: soapy water, neat ethanol, 1 M NaOH, and water. Then, a gold film with a thickness of about 50 nm estimated by atomic force microscopy (AFM) was sputtered on the clean ITO glass plate. Pyramidal, rodlike, and spherical gold nanostructures were electrodeposited from aqueous solutions of 0.1 M HClO4 containing 40, 4, and 40 mM HAuCl4, respectively, at -0.08, -0.08, and -0.2 V vs Ag|AgCl, respectively, for 2 min. (14) (a) Crumbliss, A. L.; Stonehuerner, J. G.; Henkens, R. W.; Zhao, J.; O’Daly, J. P. Biosens. Bioelectron. 1993, 8, 331-337. (b) Zhao, J.; Henkens, R. W.; Stonehuerner, J. G.; O’Daly, J. P.; Crumbliss, A. L. J. Electroanal. Chem. 1992, 327, 109-119. (15) Brown, K. R.; Fox, A. P.; Natan, M. J. J. Am. Chem. Soc. 1996, 118, 1154-1157. (16) (a) Fedurco, M. Coord. Chem. ReV. 2000, 209, 263-331. (b) Oellerich, S.; Wackerbarth, H.; Hildebrandt, P. J. Phys. Chem. B 2002, 106, 6566-6580. (c) Gong, J.; Yao, P.; Duan, H.; Jiang, M.; Gu, S.; Li, C. Biomacromolecules 2003, 4, 1293-1300. (17) (a) Tian, Y.; Liu, H.; Zhao, G.; Tatsuma, T. J. Phys. Chem. B 2006, 110, 23478-23481. (b) Tian, Y.; Liu, H.; Deng, Z. Chem. Mater. 2006, 18, 58205822.

Liu et al.

Figure 1. AFM images of (a) a sputtered gold surface and (b) electrodeposited spherical, (c) rodlike, and (d) pyramidal gold surfaces. Similar images were observed when different spots were scanned. Pyramidal, rodlike, and spherical gold electrodes were modified with cyt. c by immersing the nanostructured gold electrodes into a PBS (pH 7.2) of cyt. c (0.1 mM) for about 18-72 h at 3 °C in a refrigerator. (The enzymatic activity of the cyt. c solution stored in the refrigerator was monitored by UV-vis absorption spectroscopy. No obvious UV-vis spectrum change of cyt. c solution was observed up to 30 days.) Hereafter, the cyt. c-modified gold substrate and the cyt. c-modified nanospherical, nanorod-like, and nanopyramidal gold electrodes will be referred to as Au/cyt. c, Au-NS/cyt. c, Au-NR/cyt. c, and Au-NP/cyt. c, respectively. The real surface area of gold nanostructures was determined by calculating the charge consumed during the formation of the surface oxide monolayer in H2SO4 solution.18 Instruments and Measurements. A CHI 660 electrochemical work station (CH Instruments,) was employed in all electrochemical measurements, which were carried out with a conventional twocompartment three-electrode electrochemical cell. The reference electrode was a KCl-saturated Ag|AgCl electrode, while the auxiliary electrode was a platinum wire. An ∼50-nm-thick Au film was sputtered onto the clean ITO-coated glass with an EMS-550X highvacuum metal sputter coater (EMS, Inc.). An SPA-300HV atomic force microscope (Seiko Instruments, Inc., Japan) was employed to record the images of the nanostructured gold surfaces. An X-ray diffraction (XRD) pattern was obtained by a D/max2550VB3+/PC X-ray diffractometer using Cu (40 kV, 100 mA). The UV-vis reflectance spectrum was collected by a BWS003 UV-vis transmittance/reflectance spectrophotometer (BWTEK Instruments). The UV-vis absorption spectrum of the solution was recorded by an Agilent 8453 UV-vis-near-infrared spectrophotometer (Agilent Instruments). An infrared spectrum was recorded by a Nexus 912 AO446 FT-IR spectrophotometer (Thermo).

Results and Discussion Characterization of the Nanostructured Gold Surfaces. The morphology of the nanostructured gold surfaces was characterized by AFM as shown in Figure 1. The background (a sputtered gold surface) shows a spherical image with a diameter of around 50 nm (Figure 1a). Nanostructured gold surfaces with different shapes were electrochemically deposited under different experimental conditions, such as the applied potentials and the (18) El-Deab, M. S.; Arihara, K.; Ohsaka, T. J. Electrochem. Soc. 2004, 151, E213-E218.

Electrochemistry/Electrocatalytic ActiVity of Cyt. c

concentrations of the HAuCl4 solution. As demonstrated in Figure 1b, nanospheres with diameters ranging from 70 to 100 nm formed at a sufficiently negative potential (-0.2 V vs Ag|AgCl in a 40 mM HAuCl4 solution). On the other hand, at a lower HAuCl4 concentration (4 mM), rodlike nanostructures (Figure 1c) grew out to about 100 nm wide and up to 200-300 nm long, or more. As Figure 1d depicts, more featured nanopyramidal structures with 50-200 nm edge length along the bottom and several hundreds of nanometers in height were obtained at a more positive potential (-0.08 V vs Ag|AgCl in a 40 mM HAuCl4 solution). The crystalline characteristic of various gold films was investigated by recording the XRD pattern, as reported in our previous works.17 Diffraction peaks of sputtered Au substrate on ITO (111), (200), (220), and (311) were observed, and the intensity of the (111) diffraction peak was a little stronger than that of the other peaks, whereas a polished polycrystalline gold disk exhibits a variety of diffraction peaks, namely, (111), (200), (220), and (311), and other peaks are relatively stronger than (111). So, the present sputtered gold substrate may hold a more excellent quality of crystal orientation than a polycrystalline gold disk. Incidentally, (222), (441), and (622) peaks of the ITO substrate are also observed. The diffraction peaks of the electrodeposited spherical, rodlike, and pyramidal gold nanostructures are investigated. The observed peaks corresponding to (111), (200), (220), and (311) facets indicate that the electrodeposited gold is composed of pure crystalline gold with the face-centered cubic structure. The intensity ratios of the (200) peak to the (111) peak obtained for the pyramidal (0.059), rodlike (0.078), and spherical (0.13) structures were much lower than that reported in the standard file (JCPDS, 0.33),19 indicating that the gold nanostructures, nanopyramids in particular, were preferentially dominated by (111) facets, and the dominating (111) orientation of the gold nanostructures decreased in the order nanopyramids > nanorods > nanospheres. To characterize the electrochemical properties of the three kinds of gold nanostructured surfaces, the cyclic voltammetry of electron transfer indicators such as Fe(CN)63-/4- was first studied at the gold nanostructured electrodes and compared with that at gold substrates. As shown in Figure 2a, quasi-reversible electron redox behavior of ferricyanide was obtained at a bare gold substrate with a peak-to-peak separation (∆Ep) of about 125 mV at a potential scan rate of 100 mV s-1. However, at the nanostructured gold surfaces (Figure 2b-d), a pair of well-defined redox peaks were observed with a ∆Ep of ∼60 mV and a near unity of the anodic-to-cathodic peak current ratio (Iap/Icp), which is typical of the reversible process of the redox species in solution phase. The standard electron-transfer rate constant (ks) was calculated by the Nicholson theory20 to be 0.14, 0.18, 0.23 cm s-1 at nanospherical (Figure 2b), nanorod-like (Figure 2c), and nanopyramidal (Figure 2d) gold surfaces, respectively, all of which are much greater than that obtained at a gold substrate (about 0.06 cm s-1). It indicates that electron transfer is greatly facilitated at the nanostructured gold surfaces, which was further confirmed by electrochemical impedance spectroscopy (EIS), as demonstrated in Figure 3. The charge-transfer resistance (Rct) of the Fe(CN)63-/4- redox couple is near 20 Ω at the gold substrate (Figure 3a), but it decreases to 10, ∼1, and ∼1 Ω at nanospherical (Figure 3b), nanorod-like (Figure 3c), and nanopyramidal (Figure 3d) gold surfaces, respectively. The decreases in the chargetransfer resistance at the nanostructured gold surfaces, particularly at nanopyramidal and nanorod-like surfaces, intrinsically enhance (19) Sun, X.; Dong, S.; Wang. E. Angew. Chem., Int. Ed. 2004, 43, 63606363. (20) Nicholson, R. S. Anal. Chem. 1965, 37, 1351-1355.

Langmuir, Vol. 23, No. 18, 2007 9489

Figure 2. CVs obtained at (a) a sputtered gold substrate and (b) nanospherical, (c) nanorod-like, and (d) nanopyramidal gold nanostructured electrodes in 0.1 M KCl solution containing 1 mM [Fe(CN)6]3-/4-. Potential scan rate: 100 mV s-1.

Figure 3. Nyquist plots obtained at (a) a sputtered gold substrate and (b) nanospherical, (c) nanorod-like, and (d) nanopyramidal gold electrodes in 1 mM KCl solution containing 0.1 M [Fe(CN)6]3-/4-. EIS conditions: potential, 0.25 V; alternative voltage, 5 mV; frequency range, 0.1-105 Hz.

the electron transfer, which is in good agreement with the results obtained in cyclic voltammetry. Morphology-Dependent Electrochemistry of Cyt. c. Figure 4A shows typical cyclic voltammograms (CVs) obtained at Au/ cyt. c, Au-NS/cyt. c, Au-NR/cyt. c, and Au-NP/cyt. c electrodes in 25 mM PBS (pH 7.2) containing no cyt. c at the potential scan rate of 100 mV s-1. Well-defined reversible redox waves were observed at both Au-NR/cyt. c (c) and Au-NP/cyt. c (d), while no obvious voltammetric peak was observed at either Au/cyt. c (a) or Au-NS/cyt. c (b). Additionally, only the charge currents were obtained at the gold substrate (a), and nanospherical (b), nanorod-like (c), and nanopyramidal (d) gold surfaces in 25 mM PBS (pH 7.2) containing no cyt.c (Figure 4B). It is clear that the electron transfer of cyt. c is strikingly facilitated at nanorod-like and nanopyramidal gold surfaces, but not at the nanospherical one. The shape of the gold nanostructures plays a key role in the direct electron transfer of cyt. c. The formal potentials (E0′ )

9490 Langmuir, Vol. 23, No. 18, 2007

Liu et al.

Figure 4. (A) CVs obtained at (a) Au/cyt. c, (b) Au-NS/cyt. c, (c) Au-NR/cyt. c, and (d) Au-NP/cyt. c electrodes in 25 mM PBS (pH 7.2). (B) CVs obtained at (a) a sputtered gold substrate and (b) nanospherical, (c) nanorod-like, and (d) nanopyramidal gold electrodes in 25 mM PBS (pH 7.2). Potential scan rate: 100 mV s-1.

(Ep,a + Ep,c)/2) of Au-NR/cyt. c and Au-NP/cyt. c are estimated to be 58 and 61 mV vs Ag|AgCl, respectively, which are almost consistent with the formal potential of native cyt. c reported as 60 mV vs Ag|AgCl in a neutral solution of pH 7.0.21 It suggests that gold nanorods and nanopyramids may hold cyt. c in an orientation in which the heme edge is close to the electrode surface, while the heme ligation and the protein envelope may be preserved well, which will be further discussed later. CVs of Au-NR/cyt. c and Au-NP/cyt. c in PBS (pH 7.2) at various scan rates are demonstrated in Figure 5, panels A and B, respectively. We found that both the anodic and cathodic peak currents (Iap and Icp) vary linearly with the potential scan rate (V) in the range of 10-500 mV s-1 (the inset in Figure 5), and the ratio of Icp/Iap at a given V is nearly unity. In addition, the CVs remained essentially unchanged upon consecutive potential scanning up to 100 cycles at a sweep rate of 20 mV s-1, indicating that cyt. c is stably confined on nanorod-like and nanopyramidal gold surfaces. This should be compared with the usual cases, for example, the redox reaction of cyt. c at a SnO2 electrode modified with 12-nm-diameter colloidal Au particles and a cysteinemodified gold electrode, where the peak currents increased linearly with V1/2 as expected for the diffusion-controlled electrode process of solution-phase species.7c,15 In our case, cyt. c could be regarded as “permanently” immobilized on gold nanorodlike and nanopyramidal surfaces. According to Laviron’s procedure,22 from the potential scan rate dependence of the anodic (21) Jiang, X.; Zhang, L.; Jiang, J.; Qu, X.; Wang, E.; Dong, S. Chem. Phys. Chem. 2005, 6, 1613-1621.

Figure 5. CVs obtained at (A) Au-NR/cyt. c and (B) Au-NP/cyt. c electrodes in 25 mM PBS (pH 7.2) at different scan rates: 10, 20, 40, 60, 80, 100, 200, 300, 400, and 500 mV s-1 (from the inner to the outer). The inset shows the relationship between peak current and scan rate.

and cathodic peak potentials, the relevant kinetic parameters of the electrode reaction, that is, the rate constant of the electrochemical process (ks) and the anodic and cathodic transfer coefficients (Ra and Rc), were estimated: ks ) 1.06, Ra ) 0.63, Rc ) 0.37 and ks ) 1.13, Ra ) 0.57, Rc ) 0.43 at nanorod-like and nanopyramidal gold surfaces, respectively. The direct immobilization of cyt. c molecules at the nanopyramidal gold electrode is further verified by the measurement of the open-circuit potential (Eoc) of a Au-NP/cyt. c electrode in N2-saturated 25 mM PBS (pH 7.2). A 180 mV negative shift of Eoc (from +250 mV to +70 mV vs Ag|AgCl) was observed when the electrode was applied at a potential of -300 mV for 180 s. This negative shift of Eoc indicates that an electron is transferred to the redox center of the cyt. c (i.e., the Fe3+ moiety is converted into Fe2+), whereas the Eoc was returned back to +250 mV when it was measured after holding the electrode at +400 mV for 180 s, indicating the oxidation of the Fe2+ to Fe3+ again. A similar result was also obtained at the nanorod-like gold (22) Laviron, E. J. Electroanal. Chem. 1979, 101, 19-28.

Electrochemistry/Electrocatalytic ActiVity of Cyt. c

Figure 6. Nyquist plots obtained at (a) Au-NS/cyt. c, (b) Au-NR/ cyt. c, and (c) Au-NP/cyt. c electrodes in 0.1 M KCl solution containing 1 mM [Fe(CN)6]3-/4-. EIS conditions: potential, 0.25 V; alternative voltage, 5 mV; frequency range, 0.1-105 Hz.

electrode. A controlled experiment was performed at the bare nanostructured gold electrodes by holding the electrode potential at -300 mV for 180 s, and no obvious negative shift of the Eoc was observed. This behavior provides another evidence for the successful immobilization of the cyt. c at the nanopyramidal or nanorod-like gold surface. The adsorbed behavior of cyt. c and its surface coverage immobilized on the nanostructured gold surfaces could also be followed by EIS. Figure 6 shows Nyquist plots obtained at Au-NS/cyt. c, Au-NR/cyt. c, and Au-NP/cyt. c electrodes in 0.1 M KCl solution containing 0.1 M [Fe(CN)6]3-/4-. The charge-transfer resistance of cyt. c immobilized at nanospherical (Figure 6a), nanorod-like (Figure 6b), and nanopyramidal (Figure 6c) surfaces is estimated to be 2, 2.5, and 10 KΩ, respectively. The charge-transfer resistances of cyt. c-immobilized gold electrodes increased by a factor of ∼103, compared with that obtained at bare nanostructured gold surfaces as shown in Figure 3. It confirmed again that cyt. c adsorbed onto electrode surfaces and inhibited the electrochemical communication between the electron transfer indicator (Fe(CN)63-/4-) and the nanostructured gold surfaces. The surface coverage (θ) of cyt. c confined at the nanostructured gold surfaces can be derived from the equation θ ) 1 - Rct/Rcyt.c ct , where Rct is the chargetransfer resistance of the nanostructured gold surfaces, and Rcyt.c ct is the corresponding resistance of cyt. c-immobilized surfaces. The surface coverage of cyt. c confined at nanospherical (Figure 6a), nanorod-like (Figure 6b), and nanopyramidal (Figure 6c) surfaces is 99.90%, 99.95%, and 99.96%, respectively. It should be noted that the charge-transfer resistance obtained at the AuNS/cyt. c surface is much larger than that at either Au-NP/cyt. c or Au-NR/cyt. c surfaces, suggesting that cyt. c was actually adsorbed on the nanospherical gold surface, and the surface coverage of cyt. c adsorbed on gold nanospheres is comparable with that on gold nanopyramids or nanorods, although no obvious electroactive characteristic of cyt. c was observed at the nanospherical gold surface. Ideal Model of Electron Transfer between the Nanostructured Gold Surfaces and Cyt. c. It is well-known that direct electron transfer of cyt. c is hard to be realized at a bare gold or platinum electrode because irreversible adsorption of cyt. c easily blocks the electrode surface, and unfolding of adsorbed cyt. c also causes poisoning and deactivation of the bare

Langmuir, Vol. 23, No. 18, 2007 9491

Figure 7. (a) Infrared spectrum of cyt. c. (b-d) ATR-FTIR spectra of cyt. c adsorbed on (b) nanospherical, (c) nanorod-like, and (d) nanopyramidal gold films.

electrode.23 However, the direct electron transfer of cyt. c was greatly enhanced at nanorod-like and nanopyramidal gold surfaces and experimental results demonstrated that cyt. c was stably confined on gold nanorods and nanopyramids. So, it is desirable to clarify whether cyt. c retains its inherent biological activity after immobilization on the nanorods or nanopyramids, which enables the direct electron transfer of cyt. c itself. To understand the effect of the shape of gold nanostructures on the biological activity and conformational change of cyt. c, we studied attenuated total reflection-Fourier transform infrared spectroscopy (ATRFTIR) and UV-vis diffuse reflectance spectroscopy (DRS) of cyt. c immobilized on nanospherical, nanorod-like, and nanopyramidal gold surfaces. ATR-FTIR is a highly sensitive means for probing the secondary structure of proteins. It is well-known that, among the nine characteristic vibrational bands or group frequencies that arise from the amide groups of protein, the amide I and II bands provide the most detailed information on the secondary structure of proteins. The amide I band (1700-1600 cm-1) is almost entirely due to the CdO stretch vibrations of the peptide linkages, while the amide II band (1600-1500 cm-1) results from the combination of N-H bending and C-N stretching. Figure 7a demonstrates the infrared spectrum of cyt. c itself. Two distinctive peaks located at 1655 and 1546 cm-1 were observed, corresponding to the absorption bands arising from amide I and amide II. From the ATR-FTIR spectra of cyt. c adsorbed on nanospherical (Figure 7b), nanorod-like (Figure 7c), and nanopyramidal (Figure 7d) gold surfaces, we can see that the amide I band (1655 cm-1) has almost no changes after cyt. c adsorbed on the all three kinds of nanostructured gold films. On the other hand, the amide II band (1546 cm-1) shows no variation after cyt. c immobilized on the nanopyramidal gold surface, whereas it has a little shift (1534 cm-1) after cyt. c adsorbed on the nanorodlike gold surface, indicating the difference in the conformational changes of the heme microenvironment for the cyt. c immobilized on nanopyramids and nanorods. However, the amide II band becomes broad and almost disappears in the region of 16001500 cm-1 after cyt. c adsorbed on the nanospherical gold film, implying cyt. c may hold little biological activity, and may even (23) (a) Allen, H.; Hill, O.; Hunt, N. I.; Bond, A. M. J. Electroanal. Chem. 1997, 436, 17-25. (b) Szucs, A.; Hitchens, G. D.; Bockris, J. O. M. Electrochim. Acta 1992, 37, 403-412.

9492 Langmuir, Vol. 23, No. 18, 2007

Liu et al.

Figure 8. (a) UV-vis absorption spectrum of 10 µM cyt. c in PBS (pH 7.2). (b-d) UV-vis DRS of cyt. c adsorbed on (b) nanospherical, (c) nanorod-like, and (d) nanopyramidal gold films. The relative reflectance of a film, F(R), is given by the Kubelka-Munk equation [F(R) ) (1 - R)2/2R, where R is the ratio of the reflected intensities of the sample and the corresponding nanostructured gold films.

probably denature after being immobilized on the nanospherical gold surface.24 Figure 8a depicts the UV-vis absorption spectrum of 10 µM cyt. c in PBS. Two obvious peaks located at 409 and 522 nm were observed, corresponding to the Soret band and Q-band of cyt. c. The intense Soret band and Q-band result from the π-π* transitions of the porphyrin ring in cyt. c. To qualitatively investigate the biological activity of cyt. c immobilized on nanostructured gold surfaces with various shapes, we focused on the changes in the Soret band because the interaction of cyt. c with gold nanostructures will affect the Q-band. The optical absorption properties of cyt. c immobilized on nanospherical (Figure 8b), nanorod-like (Figure 8c), and nanopyramidal (Figure 8d) gold surfaces were conducted by means of UV-vis DRS. From the DRS spectra, we can see that all of the optical intensity of cyt. c in the Soret band decreases, accompanied by a red shift after cyt. c immobilized on the nanostructured gold surfaces. Also, clear optical intensity in the Soret band and in the Q-band was observed at either nanopyramidal or nanorod-like gold surfaces, whereas no obvious optical peak and a weak peak of cyt. c in the Soret band and in the Q-band were obtained at nanospherical gold surfaces, suggesting that cyt. c adsorbed on the gold nanospheres without electrochemical activity, probably holding little inherent activity. However, cyt. c processes the biological activity after being immobilized on nanopyramidal or nanorod-like gold electrodes, which also enables the direct (24) (a) Schlereth, D. D.; Mantele, W. Biochemistry 1993, 32, 1118-1126. (b) Nassar, A. F.; Willis, W. S.; Rusling, J. F. Anal. Chem. 1995, 67, 2386-2392. (c) Zhou, J.; Lu, X.; Hu, J.; Li, J. Chem.sEur. J. 2007, 13, 2847-2853. (d) Liu, S.; Dai, Z.; Chen, H.; Ju, H. Biosens. Bioelectron. 2004, 19, 963-969.

Figure 9. CVs obtained at (A) Au-NP/cyt. c and (B) Au-NR/cyt. c electrodes in the absence (a) and presence (b) of 5 × 10-4 M H2O2 in 25 mM PBS (pH 7.2). Potential scan rate: 100 mV s-1.

electron transfer of cyt. c. This result is in a good agreement with that obtained by ATR-FTIR. It is known that the additive effects of the transition dipole moments between two orbital excitations (a2u-b1u and a2u-eg) will affect the intense Soret band and Q-band, respectively.25 So, the intensity of the Soret band and Q-band will be affected by changes in the symmetry of the porphyrin ring. As shown in Figure 8c,d, the Q-band shows a red-shift (21-543 nm) with an increase in optical intensity and an appearance of a shoulder peak (501 nm) when cyt. c is immobilized on a nanorod-like gold surface, whereas a red-shift (39-561 nm) with an increase in intensity is obtained when cyt. c is confined on a nanopyramidal gold surface. It suggests the difference in the conformational changes of the heme microenvironment for the cyt. c immobilized on nanopyramids and nanorods. The UV-vis spectrum of cyt. c adsorbed on a nanorod-like gold surface in the Q-band is similar to that reported by Rosell et al.26 Taking into account their result, it is proposed that the thioether bonds formed by Cys14 and Cys17 with a heme prosthetic group are affected heavily due to the immobilization of cyt. c on gold nanorod-like structures. So we may deduce that the electrostatic reaction sites of cyt. c are located at the right groove, which is a positively charged surface enriched by lysine. On the contrary, the red shifting of the Q-band (25) Wu, L. L.; Huang, H. G.; Li, J. X.; Luo, J.; Lin, Z. H. Electrochim. Acta 2000, 45, 2877-2881. (26) Rosell, F. I.; Mauk, A. G. Biochemistry 2002, 41, 7811-7818.

Electrochemistry/Electrocatalytic ActiVity of Cyt. c

Langmuir, Vol. 23, No. 18, 2007 9493

Figure 10. Schematic mechanism of the catalytical reduction of H2O2 by cyt. c immobilized on nanorod-like or nanopyramidal gold electrodes.

Figure 11. Typical steady-state amperometric responses obtained at (a) Au-NP/cyt. c, (b) Au-NR/cyt. c, and (c) Au-NS/cyt. c electrodes at an applied potential of -100 mV in 25 mM PBS solution (pH 7.2) upon the successive addition of 5 × 10-5 M H2O2.

along with the increase in intensity may result from the increase in the cyt. c hydrophobic nature due to the increase of the R-helix and the decrease of the random coil when cyt. c is immobilized on gold nanopyramidal structures. Accordingly, the ideal model of electron transfer between cyt. c and gold nanostructures is proposed. Not only the morphological changes, but also the variation in the crystallographic orientation ratios of the single crystalline domains could be taken into account. The isolated and sharp pyramidal or rodlike gold nanostructures directly electrodeposited onto gold substrates may penetrate into the shells (insulators) wrapped around the metal center of cyt. c and reduce the effective electron-transfer distance. On the other hand, the crystallographic orientation may also play an important role in the electron transfer, as reported for a silver surface.27 The nanopyramidal or nanorod-like gold surface, which is enriched in Au (111), may provide the favorable adsorption orientation of cyt. c, and, subsequently, it is probably beneficial for the direct and fast electron transfer of cyt. c. The electrostatic interactions between the intrinsic negative charge on gold nanorods and the positive charge of the right groove of cyt. c enriched by lysine were probably the main forces causing the adsorption of cyt. c on gold nanorod-like surfaces, whereas the hydrophobic interactions may be the main force for the adsorption of cyt. c on gold nanopyramidal electrodes. The electrostatic and/or hydrophobic interactions may favor the protein orientation and provide a close approach of the redox metal center to the electrode surface, and subsequently remarkably facilitate the electron transfer between cyt. c and gold nanorods or nanopyramids. Also, the tips of pyramids and rods probably act as electron antennae that effectively funnel electrons between the electrode and cyt. c. However, the roundish and consecutive nanospherical gold structures, which are a little enriched in Au(110) or Au(100), are considered to be difficult to approach the metal redox center of cyt. c entrapped in an insulated shell, resulting in the

hard electron transfer between nanospheres and cyt. c. So, the electron transfer of cyt. c could be remarkably facilitated at nanopyramidal and nanorod-like gold surfaces, rather than at nanospherical surfaces. Electrocatalytical Activity toward H2O2. As demonstrated above, cyt. c processes the biological activity after being immobilized on nanopyramidal and nanorod-like surfaces, which also facilitate the electron transfer of cyt. c itself. This formed a strong basis for the development of a third-generation biosensor for H2O2 because the heme has been well documented as the active site for the catalysis to the reduction of H2O2.28 Figure 9 compares the CVs obtained at Au-NP/cyt. c- (A) and Au-NR/ cyt.c-based (B) electrodes in the absence (curves a) and presence (curves b) of H2O2. The addition of H2O2 to 25 mM PBS (pH 7.2) obviously increases the cathodic currents and decreases the anodic currents of the cyt. c confined on nanopyramidal and nanorod-like gold electrodes, indicating the good electrocatalytical activity of the immobilized cyt. c for the reduction of H2O2. It should be mentioned that the same response was obtained neither at bare nanopyramidal surfaces nor at nanorod-like electrodes. As illustrated in Figure 10, such an electromediation of the cyt.

(27) (a) Foresti, M. L.; Guideli, R. J. Electroanal. Chem. 1993, 346, 251-259. (b) Valette, G. J. Electroanal. Chem. 1989, 260, 425-431. (c) Valette, G. J. Electroanal. Chem. 1987, 224, 285-294.

(28) (a) Zhao, G.; Yin, Z.; Zhang, L.; Wei, X. Electrochem. Commun. 2005, 7, 256-260. (b) Dai, Z.; Liu, S.; Ju, H. Electrochim. Acta 2004, 49, 2139-2144. (c) Feng, J.; Zhao, G.; Xu, J.; Chen, H. Anal. Biochem. 2005, 342, 280-286.

Figure 12. Calibration curves of steady-state currents obtained at (A) Au-NP/cyt. c and (B) Au-NR/cyt. c electrodes against the concentration of H2O2. The data in panels A and B were taken from Figure 11, panels a and b, respectively.

9494 Langmuir, Vol. 23, No. 18, 2007

c-based biosensors is essentially constructed on the inherent biological activity of cyt. c for H2O2, namely, cyt. c catalyzes the reduction of H2O2 via a redox cycle of active heme, and on the direct electron transfer of cyt. c confined on the nanopyramidal and nanorod-like gold surfaces. Interestingly, the increased cathodic current obtained at Au-NP/cyt. c is much greater than that at Au-NR/cyt. c, and no obvious response was observed at a Au-NS/cyt. c electrode upon the same addition of H2O2, suggesting the shape-dependent electrocatalytical activity of the immobilized cyt. c toward H2O2. Amperometric responses of Au-NP/cyt. c (a) and Au-NR/cyt. c (b), to successive concentration changes of H2O2 were examined at the applied potential of -100 mV, and the corresponding current-time responses are shown in Figure 11. Well-defined steady-state current responses were obtained at both nanopyramidal and nanorod-like gold electrodes, and the currents increased stepwise with successive additions of H2O2. No obvious amperometric response was obtained at either the Au-NS/cyt. c electrode, as shown in Figure 11c, or the Au/cyt. c surface (data not shown). The calibration plots obtained from Figure 11a,b are shown in Figure 12. The steady-state currents at nanopyramidal and nanorod-like gold electrodes were proportional to H2O2 in the examined range of 10 µM-1 mM and 50 µM-1.5 mM, respectively, implying that the linear range of H2O2 detection could be tuned by means of employing the nanostructured gold surfaces with different shapes. The sensitivity of Au-NP/cyt. c was estimated to be 20 µA/(cm-2 mM), nearly 3 times as large as that of Au-NR/cyt. c (7 µA/(cm-2 mM)). The detection limit was evaluated on the basis of a signal-to-noise ratio of 3:1 and calculated to be 1.56 µM at nanopyramidal gold surfaces and 3.70 µM at nanorod-like gold electrodes. The response time of the sensor was measured as the time to reach 95% of the maximum change in response to a step injection of H2O2 and was found to be less than 5 s. The interferences from the coexistence of a variety of biological compounds, such as ascorbic acid, dopamine, and uric acid, with the concentrations of approximate extracellular fluid levels,29 were investigated at the applied potential of -100 mV. The experimental results demonstrated that all these interferences were negligible (