Effects of Monolayer Structures on Long-Range Electron Transfer in

Sep 13, 2008 - Effects of Monolayer Structures on Long-Range Electron Transfer in Helical Peptide Monolayer ... Telephone: +81-75-383-2400. ... Citati...
0 downloads 0 Views 434KB Size
12840

J. Phys. Chem. B 2008, 112, 12840–12850

Effects of Monolayer Structures on Long-Range Electron Transfer in Helical Peptide Monolayer Kazuki Takeda, Tomoyuki Morita, and Shunsaku Kimura* Department of Material Chemistry, Graduate School of Engineering, Kyoto UniVersity, Kyoto-Daigaku-Katsura, Nishikyo-ku, Kyoto 615-8510, Japan ReceiVed: June 29, 2008; ReVised Manuscript ReceiVed: August 14, 2008

Self-assembled monolayers of R-helical peptides were prepared on gold, and the effects of the monolayer structures (kind of constituent amino acid, molecular orientation, and molecular packing) on long-range electron transfer through the helical peptides were studied. The helical peptides were 16mer peptides having a thiophenyl linker at the N-terminal for immobilization on gold and a redox active ferrocene moiety at the C-terminal as an electron-transfer probe. The peptides were immobilized on gold by a gold-sulfur linkage and the electron transfer from the ferrocene moiety to gold was studied by electrochemical methods. When two types of the peptides, one with the repeating unit of Leu-Aib (Aib represents 2-aminoisobutyric acid) and the other with that of Ala-Aib, were compared, the electron transfer was found one order slower in the Leu-Aib peptide monolayer than that in the Ala-Aib peptide monolayer. The self-assembled monolayers of the Ala-Aib peptide with mixing of three different lengths of the peptides, 8mer, 12mer, and 16mer without a ferrocene moiety, were also prepared. The monolayer regularity in terms of molecular orientation and packing was higher roughly in the order of the monolayers mixed with 16mer > 12mer > no additive > 8mer, but the electron transfer became faster in the opposite order. The logarithms of the standard rate constants showed a nearly linear relationship with the direct distances between the ferrocene moiety and gold (β ) 0.32 Å-1). Some data deviated from this linear relationship, but the deviations could be explained from the difference in the molecular packing, which was evaluated from the monolayer capacitance. It is thus concluded that an electron is transferred along a few molecules along the surface normal so that the vertical orientation or the increase of the interchain backbone separation slows down the electron transfer. Further, it is demonstrated that a tightly packed monolayer, where vibrational mode is restricted, suppresses the electron transfer. Three models are proposed to account for the observed molecular dynamics effects on the basis of either electron-transfer mechanism of electron tunneling or sequential hopping. Introduction Electron transfer through peptide secondary structures has been a central issue to clarify efficient electron transport phenomena in biological systems in nature.1-6 Especially, electron transfer through helices has been attracting much attention because helices are considered to play an essential role in mediating an electron and even controlling electron transfer direction in protein assemblies.7 There have been a great number of studies on the electron transfer through a helical peptide by radiolysis8,9 and quenching experiments in solution,10 electrochemistry on self-assembled monolayer systems,11-17 and even scanning probe microscopy on a single peptide molecule immobilized on surface.18-20 There is a general consensus that a helical peptide is a good electron mediator for electron tunneling as indicated by a smaller decay constant over the electron-tunneling distance (β ) 0.5-0.7 Å-1)10,18,20-22 than that of a hydrocarbon chain (β ) 0.8-1.1 Å-1 assuming 1.25 Å along the chain per carbon atom).23-26 However, when the electron-transfer distance exceeds a certain point, its mechanism changes from simple electron tunneling to some other mechanism. Isied and co-workers observed a dramatic transition in β values from 1.4 Å-1 to 0.18 Å-1 when the donor-acceptor separation was over 20 Å in helical oligoprolines (0-9mer).27 * Corresponding author. Telephone: +81-75-383-2400. Fax: +81-75383-2401. E-mail: [email protected].

We previously showed that the decay constant of the electron transfer through long R-helical peptides (16mer and 24mer) could be as small as 0.02-0.04 Å-1 with adopting the electron tunneling mechanism if it was applied.17 Kraatz and co-workers working on collagen-like triple helices (12-27mer) reported a similar small decay constant (0.05 Å-1).28 There are two ideas to explain this extraordinary small β values for the long helical peptides. One is that a hopping mechanism is operative in the long helical peptides. In the hopping mechanism, a charge (electron or hole) is first injected from the donor or acceptor to the amide group in a peptide chain and then it hops along the chain to reach the distant acceptor or donor. The electron transfer rate decays exponentially with the distance increase in the tunneling mechanism,29 while it is inversely proportional to the distance in the hopping mechanism.29-39 Thus, the hopping mechanism will be dominant over the electron tunneling mechanism in a longer distance, giving apparently a small decay constant. The Isied group and our group have attributed the observed weak distance dependence to the hopping mechanism.11,12,17,27 May, Petrov, and co-workers theoretically reproduced the Isied’s observation using a hopping model.32,33,40,41 Schlag and co-workers have reported efficient hole hopping among the amide groups in a peptide chain in a gas phase.42-46 The other idea proposes the mechanism of molecular motion-assisted electron transfer or conformationally gated electron transfer. Kraatz and co-workers have adopted this

10.1021/jp805711v CCC: $40.75  2008 American Chemical Society Published on Web 09/13/2008

Effects of Monolayer Structures

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12841

Figure 1. Chemical structures of L16Fc, A16Fc, A8, A12, and A16.

Figure 2. Schematic illustration of the helical peptide SAMs studied in this work.

mechanism to explain the slow electron transfer with a weak distance dependence observed in their systems.6,14,16,28,47 In this mechanism, a peptide bridge thermally fluctuates to generate at a certain point an active conformer which readily transfer electrons through. However, note that the electron transfer mechanism is not limited to electron tunneling. A hopping process can also be assisted by molecular motion when some motion is related to efficient charge transfer among the amide groups. To gain a deeper insight on the mechanism, here we discuss effects of peptide backbone separation due to different constituent amino acids, molecular orientation, and molecular packing on long-range electron transfer through a helical peptide monolayer. Helical peptides having a thiophenyl group at the N-terminal and a redox-active ferrocene moiety at the C-terminal were immobilized on gold by a gold-sulfur linkage to form self-assembled monolayers (SAMs), and the electron transfer from the ferrocene moiety to gold was studied. Ferrocene is the most widely used to study electron transfer reactions through SAMs.25,48-52 First, we compared two different peptide monolayers with different backbone separation and molecular packing. One is a 16mer R-helical peptide having repeats of an L-leucine (Leu) and R-aminoisobutyric acid (Aib) sequence, and the other is a 16mer peptide having repeats of an L-alanine (Ala) and Aib sequence (L16Fc and A16Fc, Figure 1). The Leu-Aib peptide has bulky isobutyl side chains which make the peptide backbone separation large in the SAM compared with the SAM of the Ala-Aib peptide (Figure 2). Further, the Leu-Aib peptide

forms a tightly packed monolayer through interdigitation among the side chains.11,53,54 Second, we compared the peptide SAMs with different structural regularities. For this purpose, the A16Fc SAM was diluted with an equal amount of 8mer, 12mer, or 16mer Ala-Aib peptide without a ferrocene moiety (A8, A12, and A16, Figure 1) and with opposite dipole moment when it is immobilized on gold. When two kinds of helical peptides, one has an immobilizing group at the N-terminal and the other at the C-terminal, are immobilized together on gold, the monolayer generally becomes tightly packed with a small tilt angle of the helix axis from the surface normal compared with the SAM of single component.55,56 In such a mixed monolayer, antiparallel arrangement of helix dipole moments promotes tight molecular packing due to the favorable electrostatic interaction between the neighboring dipoles in the SAM. The monolayer regularity of the A16Fc SAM can be thus systematically variable just by mixing with A8, A12, or A16 with different molecular lengths (Figure 2). We assume that the A16Fc/A8 monolayer is loosely packed because of discrepancy in their lengths while the A16Fc/A16 monolayer is tightly packed by the length match. Infrared reflection-absorption spectroscopy (IRRAS), ellipsometry, and electrochemical blocking experiment were used for structural analyses of the peptide SAMs. The electron transfer from the ferrocene moiety to gold was studied by electrochemical methods including cyclic voltammetry (CV), chronoamperometry (CA), and electrochemical impedance spectroscopy (EIS). The results revealed that all the factors of backbone separation, molecular orientation, and molecular

12842 J. Phys. Chem. B, Vol. 112, No. 40, 2008 packing affected the electron transfer. It is concluded that an electron should be transferred across a few peptide molecules in the SAM, and facile molecular motion should facilitate the electron transfer. Three mechanisms are proposed for the molecular dynamics effects: electron tunneling gated by global helix motion, electron tunneling coupled to helix conversion, and hole hopping assisted by local motion of the peptide chain. Materials and Methods Synthesis of Helical Peptides. L16Fc, A16Fc, A8, A12, and A16 were synthesized by the conventional liquid-phase method according to the literature.11,12,17,54 All the intermediates were identified by 1H NMR spectroscopy and the final products were further confirmed by FAB mass spectrometry. The purity of the compounds was checked by thin-layer chromatography. The identification data of the final products are shown below. L16Fc. TLC: (chloroform/methanol ) 10/1 v/v) Rf ) 0.39; (chloroform/methanol ) 5/1 v/v) Rf ) 0.72; (chloroform/ methanol/ammonia-water ) 13/5/1 v/v/v) Rf ) 0.87; (chloroform/methanol/acetic acid ) 95/5/3 v/v/v) Rf ) 0.16. 1H NMR (CDCl3, 400 MHz): δ (ppm) 0.8-1.1 (48H, LeuCδH3), 1.3-2.0 (72H, AibCH3, LeuCβH2, LeuCγH), 2.45 (3H, CH3CO), 3.3-3.8 (4H, NHCH2CH2NH), 3.8-4.1 (8H, LeuCRH), 4.1-4.4 (5H, ferrocene-H), 4.7-5.1 (2H, ferrocene-H), 6.9-9.0 (22H, amideH, C6H4). MS (FAB, matrix; nitrobenzylalcohol): m/z 2036.2 (calcd for C102H166FeN18O19S [(M + H)+] m/z 2036.2). A16Fc. TLC: (chloroform/methanol ) 10/1 v/v) Rf ) 0.36; (chloroform/methanol ) 5/1 v/v) Rf ) 0.66; (chloroform/ methanol/ammonia-water ) 13/5/1 v/v/v) Rf ) 0.77; (chloroform/methanol/acetic acid ) 95/5/3 v/v/v) Rf ) 0.15. 1H NMR(CDCl3, 400 MHz): δ (ppm) 1.4-1.7 (72H, AlaCβH3, AibCβH3), 2.45 (3H, CH3CO), 3.3-3.8 (4H, NHCH2CH2NH), 3.8-4.1 (8H, AlaCRH), 4.1-4.4 (5H, ferrocene-H), 4.7-5.1 (2H, ferrocene-H), 7.3-9.0 (22H, amide-H, C6H4). MS (FAB, matrix; nitrobenzylalcohol): m/z 1699.8 (calcd for C78H118FeN18O19S [(M + H)+] m/z 1699.8). A8. TLC: (chloroform/methanol ) 10/1 v/v) Rf ) 0.28; (chloroform/methanol ) 5/1 v/v) Rf ) 0.66; (chloroform/ methanol/ammonia-water ) 13/5/1 v/v/v) Rf ) 0.88; (chloroform/methanol/acetic acid ) 95/5/3 v/v/v) Rf ) 0.13. 1H NMR (CDCl3, 400 MHz) δ (ppm): 1.3-1.75 (36H, AlaCβH3, AibCβH3), 1.87 (3H, CH3CONH), 2.37 (3H, C6H4SCOCH3), 3.8-4.2 (4H, AlaCRH), 7.1-9.3 (13H, amide-H, C6H4). MS (FAB, matrix; nitrobenzylalcohol): m/z 834.4 (calcd for C38H60N9O10S [(M + H)+] m/z 834.4), m/z 856.4 (calcd for C38H59N9NaO10S [(M + Na)+] m/z 856.4). A12. TLC (chloroform/methanol ) 10/1 v/v) Rf ) 0.33; (chloroform/methanol ) 5/1 v/v) Rf ) 0.65; (chloroform/ methanol/ammonia-water ) 13/5/1 v/v/v) Rf ) 0.94; (chloroform/methanol/acetic acid ) 95/5/3 v/v/v) Rf ) 0.22. 1H NMR (CDCl3, 400 MHz) δ (ppm): 1.4-1.75 (54H, AlaCβH3, AibCβH3), 1.85 (3H, CH3CONH), 2.38 (3H, C6H4SCOCH3), 3.8-4.2 (6H, AlaCRH), 7.1-8.3 (17H, amide-H, C6H4). MS (FAB, matrix; nitrobenzylalcohol): m/z 1168.6 (calcd for C52H83N13NaO14S [(M + Na)+] m/z 1168.6). A16. TLC (chloroform/methanol ) 10/1 v/v) Rf ) 0.24; (chloroform/methanol ) 5/1 v/v) Rf ) 0.76; (chloroform/ methanol/ammonia-water ) 13/5/1 v/v/v) Rf ) 0.88; (chloroform/methanol/acetic acid ) 95/5/3 v/v/v) Rf ) 0.18. 1H NMR (CDCl3, 400 MHz) δ (ppm): 1.2-1.8 (72H, AlaCβH3, AibCβH3), 1.88 (3H, CH3CONH), 2.37 (3H, C6H4SCOCH3), 3.8-4.2 (8H, AlaCRH), 7.1-8.3 (21H, amide-H, C6H4). MS (FAB, matrix; nitrobenzylalcohol): m/z 1458.8 (calcd for C66H107N17O18S [(M + H)+] m/z 1458.8).

Takeda et al. Preparation of SAMs. A slide glass was cleaned by sulfuric acid and rinsed with water and methanol. A gold substrate was prepared by vapor deposition of chromium and then gold (99.99%) onto the slide glass by a vacuum deposition system (N-KS350, Osaka Vacuum, Osaka). The thicknesses of the chromium and gold layers, monitored by a quartz oscillator, were approximately 300 and 2000 Å, respectively. The prepared gold substrate was immediately used for self-assembling. The SAMs were prepared by the following six different conditions (A-F). A: A 0.1 mM ethanol solution of the peptide was prepared, 28 wt % ammonia-water (10 µL per 1 mg of peptide) was added to remove the acetyl group, and a gold substrate was immersed in the solution at room temperature for 24 h. B: A 2 M dimethylamine methanol solution (50 equiv of the peptide) was added to a 0.1 mM peptide ethanol solution for deprotection of the acetyl group, the solution was stirred for a few minutes, and a gold substrate was immersed in the solution at room temperature for 24 h. C: A similar condition to B but at 50 °C for 24 h. D: A similar condition to C but for 48 h. The L16Fc monolayers were prepared in the conditions A and B, and the A16Fc monolayers were prepared in the conditions A, B, C, and D. E: A 0.1 mM A16Fc ethanol solution was treated with ammonia-water (200 µL per 1 mg of peptide) at room temperature for 3 h. On the other hand, a 0.1 mM A8 or A12 ethanol solution was treated with ammonia-water (total 400 µL per 1 mg of peptide, 200 µL at the beginning and another 200 µL after 3 h) at 40° for 6 h to remove the acetyl group at the C-terminal. Then, the two solutions were equally combined and a gold substrate was immersed in the solution for 50 °C for 48 h. The A16Fc/A8 and A16Fc/A12 mixed monolayers were prepared in this condition. F: 2 M dimethylamine methanol solution (50 equiv of the peptide) was added to a mixed methanol solution of A16Fc (0.05 mM) and A16 (0.05 mM), the solution was stirred for a few minutes, and a gold substrate was immersed in the solution at 50 °C for 48 h. The A16Fc/ A16 mixed monolayers were prepared in this condition. After each procedure, the substrate was rinsed with chloroform once and chloroform/methanol (1/1 v/v) twice, and dried in a N2 stream and in vacuum for 15 min. Monolayer Characterizations. IRRAS spectra were recorded on a Fourier transform infrared spectrometer (Nicolet 6700 FTIR, Thermo Fisher Scientific, MA) at room temperature with a reflection attachment (RMA-1DG/VRA, Harrick, NY). The incident angle was set at 80-85° from the surface normal. The number of interferogram accumulations was 200. The tilt angles of the helix axis from the surface normal was determined from the amide I/II absorbance ratio by using an equation in the literature.11,12,17,57-59 Ellipsometry was carried out by an autoellipsometer (DHA-OLX/S, Mizojiri Optical, Tokyo) at room temperature to determine the thicknesses of the monolayers. A helium-neon laser at a wavelength of 632.8 nm was used as the incident light, and its incident angle was set at 65°. The complex optical constant of the monolayer was assumed to be 1.50 + 0.00i. The thickness of the monolayer was calculated automatically by an equipped program. The thicknesses were measured on more than 5 different spots on the substrate and the data were averaged. The typical standard deviation was ca. 1 Å. Electrochemical Measurements. Electrochemical experiments were performed by a voltammetric analyzer (model 604, BAS, Tokyo) at room temperature, on a three-electrode system with the monolayer-modified gold substrate as the working electrode, Ag/AgCl in a 3 M NaCl aqueous solution as the reference electrode, and a platinum wire as the auxiliary

Effects of Monolayer Structures

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12843

Figure 3. Representative IRRAS spectra of the SAMs.

electrode. Milli-Q water was used to prepare the solutions. The solutions were flushed with N2 for 15 min prior to the experiments. All the applied potentials on the working electrode reported here are with respect to the reference electrode. The area of the working electrode exposed to the electrolyte solution was 0.9-1.1 cm2. The uncompensated resistance of the cell, estimated by EIS, was ca. 3-4 ohm. The blocking experiment by CV to assess the monolayer defects was carried out in a 1 mM K4[Fe(CN)6] and 1 M KCl aqueous solution. The electron transfer experiments were carried out in a 1 M HClO4 aqueous solution. CV was performed at various scan rates of 0.05-0.5 V/s. The capacitances of the monolayers were determined by dividing the capacitic current at 0.1 V by the scan rate in the cyclic voltammograms taken at a 0.1 V/s scan rate. The capacitic current is dominant at the potential. In CA, the time constant in the current follower of the potentiostat was set at 10-4 s. The potential was stepped from formal oxidation potentialoverpotential (V) to formal oxidation potential+overpotential (V) at the time zero. EIS was performed with a DC voltage of 50 mV at frequencies from 105 to 10-2 Hz. The Bode plot obtained at the formal oxidation potential was fitted by an equivalent circuit shown in Figure 7 consisting of the solution resistance (Rs), monolayer constant phase element (CPE, ZCPE ) (1/(Q(iω)n))), and electron-transfer resistance (Ret) and capacitance (Cet).60-62 In this study, a CPE was used instead of a simple capacitance to get better agreement between the experimental and simulated curves by accounting for the inhomogeneity of the electrode surface.63,64 The standard electron transfer rate constant (ket0) was calculated by ket0 ) 1/(2RetCet). Calculations of molecular length, monolayer thickness, and monolayer surface density. The molecular length of the 16mer helical peptide was estimated as the sum of the helix part and the linking parts which are between gold and the peptide and between the peptide and the ferrocene moiety. The helix length was estimated to be 24 Å with 1.5 Å per residue for an R-helix.65 Since the conformations of the linking parts are unknown, the half-value of the length of the all-trans extended chain was taken as approximation. The extended geometry of a virtual compound, Au-S-C6H5-CONH-(CH2)2-NHCO-ferrocene was generated by a CAChe WorkSystem software (ver. 6.1.1, Fujitsu, Tokyo) and optimized by the Molecular Mechanics program 2 (MM2), and the length between the Au atom and the ferrocene edge was estimated to be 15 Å. The length of the linking parts was thus assumed to be 7.5 Å, and the distance between the ferrocene moiety and gold along the

molecule was assumed to be 31.5 Å. The direct distance between them along the surface normal was calculated by 31.5 × cos (IRRAS tilt angle) (Å). The total molecular length was set to be 37.5 Å assuming ferrocene as a sphere with a 6 Å diameter. The theoretical thicknesses of the monolayers were calculated by 37.5 × cos (IRRAS tilt angle) (Å). On the other hand, the theoretical monolayer surface densities were estimated from the cross-sectional areas of the helices and the tilt angles. The crosssectional areas were taken to be 1.54 nm2 (14.0 Å diameter) for the Leu-Aib peptide and 0.68 nm2 (9.3 Å diameter) for the Ala-Aib peptide, respectively.55,66,67 Assuming hexagonal packing with a 0° tilt angle, the limiting surface densities were calculated to be 1.0 × 10-10 mol/cm2 for the Leu-Aib peptide and 2.2 × 10-10 mol/cm2 for the Ala-Aib peptides, respectively. Limiting surface density × cos (IRRAS tilt angle) gave the theoretical surface density of the monolayer. Results and Discussions SAM Preparation. The helical peptides (L16Fc, A16Fc, A8, A12, and A16; Figure 1) were synthesized by the liquid-phase synthesis. The L16Fc and A16Fc pure SAMs and A16Fc/A8, A16Fc/A12, and A16Fc/A16 mixed SAMs were prepared by immersion of a gold substrate into the respective peptide solution after removing the acetyl protecting group of the thiophenyl group by basic treatments. Preparation conditions (deprotecting base, period, temperature, and solvent system, see the experimental section for the details) were optimized for the respective systems to produce a monolayer with upright orientation (the tilt angle of the helix from the surface normal is less than 50°). For the pure monolayers, more than one conditions were used, producing monolayers with slightly different molecular orientation and packing degree. Monolayer Characterization. Molecular orientation was studied by IRRAS spectroscopy. Representative spectra are shown in Figure 3. Amide I and II bands were observed at around 1670 cm-1 and 1540 cm-1, respectively. These wave numbers are characteristic for helical conformation.59 The tilt angles of the helices from the surface normal were determined on the basis of the ratio of the amide I and II absorbances.11,12,17,57-59 The amide I/II absorbance ratios and derived tilt angles are summarized in Table 1. The tilt angles were 29-49°, indicating that the helical peptides did not lie down on the surface but took upright orientation. The L16Fc SAM showed 34 ( 1° tilt angles while the A16Fc SAM showed 38 ( 2° tilt angles. A smaller tilt angle in the L16Fc SAM can be explained by tight packing of neighboring helices by interdigi-

12844 J. Phys. Chem. B, Vol. 112, No. 40, 2008

Takeda et al.

TABLE 1: Summary of the Monolayer Characterizations. SAM L16Fc A16Fc

A16Fc/A8 A16Fc/A12

A16Fc/A16

a

preparation conditiona

tilt angle (deg)

theoretical thickness from IRRAS (Å´)

A B B A B B C C D E E E E E E E F F F F

33 34 34 34 38 38 39 39 39 49 45 45 30 30 29 29 31 31 34 34

31.5 31.1 31.1 31.1 29.6 29.6 29.1 29.1 29.1 24.6 26.5 26.5 32.5 32.5 32.8 32.8 32.1 32.1 31.1 31.1

experimental thickness from ellipsometry 18.0 25.0 25.0 16.0 29.0 29.0 25.0 25.0 26.0 18.8 20.2 20.2 29.0 29.0 25.2 25.2 27.4 27.4 25.6 25.6

See the Materials and Methods section for the details.

tation of the side chains to form a more highly ordered monolayer. On the other hand, the A16Fc/A8 mixed SAM gave larger tilt angles (46 ( 2°) than that of the pure A16Fc SAM (38 ( 2°), while the A16Fc/A12 (30 ( 1°) and A16Fc/A16 (33 ( 2°) mixed monolayers showed more vertical orientation than the pure monolayer. This result suggests that the antiparallel arrangement of A16Fc with A12 or A16 which has the opposite dipole moment on the surface relieves dipole-dipole electrostatic repulsion to allow more vertical orientation,56 whereas A8 is not long enough to form a well-packed antiparallel monolayer with A16Fc (Figure 2). The theoretical thicknesses of the monolayers were estimated from the molecular length (37.5 Å) and the tilt angles determined by IRRAS (Table 1). The theoretical values fairly agree with the experimental monolayer thicknesses measured by ellipsometry (Table 1), showing that the peptides form a well-defined monolayer with

homogeneous orientation. Using the tilt angles, the theoretical values of the surface density of the peptide (Table 2) and the direct distance between the ferrocene moiety and gold along the surface normal were also calculated. Electrochemical Measurements. To study the electron transfer between the ferrocene moiety and gold, cyclic voltammetry was performed in an HClO4 solution. Figure 4 shows the representative cyclic voltammograms of the respective SAMs. In all the monolayers, reversible redox peaks of the ferrocene oxidation were clearly observed. The formal potentials for ferrocene oxidation were ca. 0.47 V for the Leu-Aib SAM and 0.44 V for the Ala-Aib pure and mixed SAMs, respectively. The peak separations were ca. 40 mV for the Leu-Aib SAM and 25-35 mV for the Ala-Aib SAMs, suggesting slower electron transfer in the Leu-Aib SAM. The peak full widths of the half-maximum were ca. 130 mV in all the cases, which is slightly larger than the ideal value (90 mV). This suggests that there is a certain interaction among the ferrocene moiety because they are arranged at a relatively high density on the densely packed peptide monolayer.28,68,69 The surface densities of the ferrocene moiety were calculated by integration of the anodic peaks (Table 2). The ferrocene densities of the pure SAMs (L16Fc and A16Fc) agree well with the theoretical peptide densities, and the ferrocene densities in the mixed SAMs (A16Fc/A8, A16Fc/A12, and A16Fc/A16) are nearly half of the peptide densities, which are agreeable for the mixed SAMs of 1:1. This agreement confirms that the peptides form a regular monolayer with uniform orientation, and indicated that the comparable amount of diluent peptides (A8, A12, and A16) to A16Fc are incorporated in the mixed SAMs. A large background current was seen in the voltammogram of the A16Fc/A8 mixed SAM, showing that the monolayer is loosely packed and thus has a larger capacitance and allows faradaic currents through the defects. The approximate monolayer capacitances were determined from the capacitic current in the voltammograms as a quantitative value of monolayer packing (Table 2). The average capacitances are 28 ( 9 µF/cm2 for the L16Fc SAM, 37 ( 10 µF/cm2 for the A16Fc SAM, 50 ( 27 µF/cm2 for the A16Fc/A8 SAM, 39 ( 4 µF/cm2 for the A16Fc/A12 SAM, and 26 ( 8 µF/cm2 for the A16Fc/A16 SAM, respectively. The order of capacitances is generally in line with the order of tilt angles,

TABLE 2: Summary of the Electrochemical Studies SAM L16Fc A16Fc

A16Fc/A8 A16Fc/A12

A16Fc/A16

experimental ferrocene surface density from CV (× 10-11 mol/cm2)

theoretical peptide suface density from IRRAS (× 10-11 mol/cm2)

8.7 8.3 8.3 14.1 13.8 16.6 12.6 12.2 16.0 5.6 8.3 7.9 10.3 9.8 11.8 10.6 10.0 11.6 11.5 11.3

8.4 8.3 8.3 18.2 18.2 17.3 17.3 17.1 17.1 17.1 14.4 15.6 15.6 19.1 19.1 19.2 19.2 18.9 18.9 18.2

ket0 from CA

57 44 964 557 1233 857 833 1474 1444 393 521 317 325 844 532

ket0 from EIS 51 45 41 375 689 112 1263 466 714 1198 1478 1816 211 243 322 413 111 217 509 124

k′et0 from EIS

69 85 14 140 52 79 40 82 101 56 65 93 120 27 53 94 23

monolayer capacitance from CV (µF/cm2) 33.3 32.2 17.9 30.7 35.5 38.5 34.3 27.0 54.3 20.3 74.1 56.2 41.1 37.9 42.1 34.1 32.0 25.7 30.9 13.9

Effects of Monolayer Structures

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12845

Figure 4. Representative cyclic voltammograms of the SAMs in a 1 M HClO4 aqueous solution at a scan rate of 0.5 V/s.

Figure 5. Cyclic voltammograms of a bare gold substrate (dashed lines) and the SAM-modified substrates (solid lines) in a 1 mM K4[Fe(CN)6] and 1 M KCl aqueous solution at a scan rate of 0.1 V/s.

indicating that vertical orientation makes the monolayer tightly packed or vice versa. Qualitative assessment of monolayer defects were performed by cyclic voltammetry in an aqueous solution containing ferrocyanide redox ions. The cyclic voltammograms of a control bare gold substrate and the SAMmodified substrates are shown in Figure 5. Clear redox peaks of [Fe(CN)6]3-/[Fe(CN)6]4- were seen in the bare substrate. On the other hand, the redox peaks were significantly reduced in the SAMs, showing that these monolayers do not have large defects which allow the ferrocyanide ions to reach the gold surface. In the A16Fc, A16Fc/A8, and A16Fc/A12 SAMs, an oxidation peak of the ferrocene moiety without following reduction was observed at 0.40-0.46 V. This is a typical behavior of a SAM with a ferrocene moiety in a redox solution in which the oxidized ferrocene moiety is readily reduced by the redox species in the aqueous phase giving no following reduction.70 The oxidation peak is most remarkable in the A16Fc/A8 SAM followed by the A16Fc and A16Fc/A12 SAMs, and the peak was not observed in the L16Fc and A16Fc/A16 SAMs. Since the peak intensity shows how readily the electron transfer occurs, it is expected that the electron transfer may be faster in the order of A16Fc/A8 > A16Fc > A16Fc/A12 > A16Fc/A16 and L16Fc. CA was performed in an HClO4 solution to determine the standard electron transfer rate constant (ket0). Time courses of the anodic currents after applying a +0.01 V overpotential are shown in Figures 6a in the form of a semilog plot. Each

current-time curve has a linear part, the slope of which is the electron transfer rate constant (ket).69,71,72 ket values were collected at various overpotentials (0.01-0.05 V), they were plotted versus the overpotentials (Figure 6b), and ket0 was determined by extrapolation to the zero overpotential. The ket0 values from the positive and negative overpotentials are comparable for the A16Fc/A12, A16Fc/A16, and the L16Fc SAMs, while they are different for the A16Fc/A8 and A16Fc SAMs. In the latter two monolayers, the cathodic currents are accompanied by background currents allowed by the defects of the monolayer as seen in the distorted cathodic peaks. The noticeable overpotential dependences of the rate constants also support this interpretation. Accordingly, the ket0 value from the negative overpotentials should deviate from the true value at least for these monolayers. Therefore, the ket0 values from the positive overpotentials are adopted in this study. Interestingly, the ket values are almost independent of the overpotentials. This behavior is quite different from other reports on SAMs of redoxterminated thiols, in which a 0.05 V overpotential accelerates the electron tunneling by 2-3 times.69,71,73,74 This negligible overpotential dependence suggests that an electron is not transferred directly from the ferrocene moiety to gold because the driving force of the direct electron tunneling is equal to the overpotential. On the other hand, if a hopping mechanism is assumed in which the rate determining hopping step is not influenced by the Fermi level of gold, the overpotential dependence would be zero. This needs to be clarified by a

12846 J. Phys. Chem. B, Vol. 112, No. 40, 2008

Takeda et al.

Figure 6. (a) Representative the current-time curves of the SAMs with a 0.01 V overpotential applied at the time zero in a semilog plot, and (b) the Tafel plots to determine the standard electron transfer rate constants (ket0).

theoretical calculation taking into account the potential profile in the monolayer. The determined ket0 values are shown in Table 2. Setting aside the differences in the tilt angles and monolayer capacitances here, as a general trend, the electron transfer is faster in the order of A16Fc/A8 > A16Fc > A16Fc/A12 > A16Fc/A16 > L16Fc, which agrees with the prediction from the blocking experiment results. Unfortunately, in some cases (blank entires in Table 2), the CA measurements were unsuccessful because of a background current due to the monolayer defects making it hard to find a linear part in the current-time curve. To supplement the CA results, EIS was carried out to determine ket0. The representative results obtained at the formal oxidation potential are shown in Figure 7 in the form of a Bode plot (absolute impedance (|Z|) and phase vs frequency). The Bode plots were fitted with an equivalent circuit composed of the solution resistance (Rs), monolayer constant phase element (CPE), the electron-transfer resistance (Ret), and capacitance (Cet) shown in the top panel of Figure 7. The ket0 values were determined by ket0 ) 1/(2RetCet), and the results are summarized in Table 2. The ket0 values determined by EIS generally agree with the values determined by CA, confirming the reliability of these measurements. Since the EIS measurements were successful on all the SAMs, the following discussion on the electron transfer is made on the basis of the EIS results. The simply averaged ket0 values are 46 ( 5 s-1 for the L16Fc, 603 ( 289 s-1 for the A16Fc, 1497 ( 309 s-1 for the A16Fc/A8, 297 ( 57 s-1 for the A16Fc/A12, and 240 ( 206 s-1 for the A16Fc/A16 monolayers, respectively. There are average values showing large deviations by simple average, and thus we discuss what causes the data dispersion in terms of molecular orientation and molecular packing in the following sections. Discussion on the Electron Transfer. First, it is clear that the electron transfer is slower in the Leu-Aib peptide SAM (L16Fc SAM) compared to any Ala-Aib peptide SAMs. The features of the Leu-Aib peptide SAM are vertical orientation, large backbone separation and tight monolayer packing. If the backbone separation is responsible, the slower electron transfer can be interpreted as the result of lowered mediation of electron tunneling by the leucine isobutyl side chains or suppression of interchain hole hopping among the amide groups. In both cases,

the electron transfer should be an intermolecular process (Figure 8b). The suppression of electron transfer by vertical orientation can also be understood by considering the intermolecular electron transfer process. On the other hand, if the tight monolayer packing is responsible, an effect of molecular dynamics should be considered. Faster electron transfer is often observed in a loosely packed peptide monolayer.14,16,17 The slower electron transfer may be then attributed to restricted molecular motion in a tight monolayer. Next, the Ala-Aib peptide monolayers are compared. It is evident that the electron transfer is much faster in the A16Fc/ A8 mixed SAM compared to the other SAMs. The most striking point of the A16Fc/A8 SAM is the large tilt angle. If the electron transfer occurs intramolecularly (Figure 8a), there would be no dependence on the tilt angles. It is thus suggested that an electron is transferred along the surface normal (Figure 8b). Figure 9a shows the plot of ln(ket0) versus the direct distance between the ferrocene moiety and gold. The data of the Ala-Aib peptide SAMs can be roughly fitted by a linear function with a slope of -0.32 Å-1. This finding shows that the electron transfer is an intermolecular process via a few peptide molecules (Figure 8b) which is facilitated by tilted orientation shortening the direct distance of electron transfer. However, this decay factor is smaller than the values reported by Sisido and co-workers (β ) 0.66 Å-1)10 and by Sek and co-workers (β ) 0.50 Å-1).20 Moreover, if the mechanism is electron tunneling along the surface normal, the decay factor would be larger than those values for intrachain electron tunneling because a hydrocarbon chain is not a better electron mediator (β ) 0.8-1.1 Å-1)23-26 and noncovalent gaps are also involved. Therefore, the observed small decay factor cannot be explained by simple electron tunneling. These SAMs differ from each other in terms of the direct distance from the ferrocene to gold and molecular packing. In order to extract the effect of molecular packing, the ket0 values of the Ala-Aib peptide SAMs were normalized to those of the same direct distance of 31.5 Å under the assumption of β ) 0.32 Å-1 to afford the normalized rate constants (k′et0) by an equation, k′et0 ) ket0 × exp[0.32(direct distance - 31.5)]) (Table 2). The ln(k′et0) values are plotted over the direct distance in Figure 9b. The scattering of k′et0 represents the deviation from the linear relationship. When the ln(k′et0) values are plotted

Effects of Monolayer Structures

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12847

Figure 7. Equivalent circuit used to analyze the EIS data, and the Bode plots of the SAMs at the ferrocene oxidation potential. The open circles and triangles show the experimental absolute impedance (|Z|) and phase, and the solid lines are the fitting curves.

Figure 8. Schematic illustrations of (a) intramolecular electron transfer and (b) intermolecular electron transfer through the helical peptide monolayer.

against the monolayer capacitances determined by CV (Figure 9c), it shows a saturation behavior indicated by a solid line. This finding shows that the electron transfer is accelerated as the monolayer becomes loose up to a certain point but the effect saturates from there. In other words, some molecular motion associated with the electron transfer is restricted in a tightly packed monolayer, but once the monolayer is loosened to a certain extent, the molecular motion is free to occur and no longer affects the electron transfer. Therefore, the dynamics (motion) in question should not be a big motion such as helix bending or tilting, but some motions which are allowed even in such vertically oriented peptide monolayers with relatively few defects (voids). Getting back to the slow electron transfer in the Leu-Aib peptide SAM, it can be ascribed to the combination of vertical orientation or large backbone separation hindering intermolecular electron transfer, and the tight molecular packing suppressing this type of molecular motion. Discussions on the Molecular Dynamics Effect. It has been proposed that molecular dynamics can have a large influence on peptide electron transfer. Kraatz and co-workers have proposed a conformationally gated electron transfer mechanism.6,14,16,28,47 Although they do not specify the electron transfer mechanism, let us here consider electron tunneling first. Collective vibrations of the C-C and C-N bonds of the peptide backbone generate global motions such as stretching, contraction, bending, and other deformations of the helix.75,76 During these global motions, a specific conformation enabling a strong electronic coupling through the peptide bridge is formed, when an electron instantly tunnels through the bridge. Substantial contraction or bending of the helix which brings the ferrocene

moiety close to gold, can be one of those specific conformations (Figure 10a). Since those motions are in the time scale of picosecond,77-80 another frequency factor is needed to account for the rate constants in the order of 100-1000 s-1 observed in the present systems. However, a global bending of the helix to reduce the distance considerably between ferrocene moiety and gold (Figure 10a, right) is not plausible even in a loosely packed monolayer. The observed saturation behavior of the rate constants with increase of the monolayer capacitance also suggests that such a global lateral motion of the helix is excluded. For the same reason, tilting motion of the helix, which is not a vibration-based motion though, is also unlikely to occur. Therefore, only the vertical helix contraction (Figure 10a, middle) appears reasonable if global motion and electron tunneling are responsible. It is well-known that vibronic coupling plays an important role in the electron transfer between donor and acceptor molecules.81-86 In the classical theory, only outer-sphere reorganization due to the rotations and vibrations of solvent molecules surrounding the reactant and the product is considered.87,88 The energies of those degrees of freedom are much lower than the thermal energy. Thus the potential curves of the initial and final states are depicted as two continuous parabolas, and the activation energy at their intersection is provided by the thermal energy. On the other hand, when the electron transfer involves changes of the bond lengths of the donor, acceptor, and bridge if exists (inner-sphere reorganization) and the vibration energies are higher than the thermal energy, the classical description with parabolas is no longer valid and a quantum treatment is needed. In the semiclassical theory, the transition probability is determined by the overlap of nuclear wave functions in the initial and final states.81-86 It has been reported that electron tunneling between a ferrocene moiety and gold in a self-assembled monolayer is governed by solvent reorganization.47,73,89,90 However, it is sure that the electron transfer is not controlled simply by electron tunneling in the present system. If a certain bond-length change is involved during the electron transfer, molecular packing can affect the vibration and accordingly influence the electron transfer. If electron tunneling is assumed, there should be some difference in conformations before and after the electron is tunneled. One possibility is conversion from

12848 J. Phys. Chem. B, Vol. 112, No. 40, 2008

Takeda et al.

Figure 9. (a) Dependence of ln(ket0) values determined by EIS on the direct distances between the ferrocene moiety and gold for the Leu-Aib (square) and Ala-Aib peptide monolayers (circle) with a linear fitting curve for the Ala-Aib peptide monolayer (solid line), (b) plot of the ln(k′et0) versus the direct distances, where k′et0 is the normalized ket0 to take the same distance (31.5 Å) in all the Ala-Aib peptide monolayers with a decay constant of 0.32 Å-1, and (c) plot of ln(k′et0) versus the monolayer capacitance determined by CV with a fitting curve (solid line).

Figure 10. Proposed mechanisms for the molecular dynamics effect on the electron transfer, (a) global motion-gated electron tunneling, (b) electron tunneling coupled with helix conversion from R-helix to 310-helix, and (c) hole hopping among the amide groups assisted by a local backbone motion.

R-helix to 310-helix which causes vertical elongation of the helix (Figure 10b). In this case, the vibrations of the peptide backbone whose transition is parallel to the helix axis (amide I for example) should be involved in the electron transfer process and is thus affected by the molecular packing. We have demonstrated previously that a helical peptide could switch between the two helix states depending on the polarity of the external electric field in the scanning tunneling microscopy system.91 Electrostatic repulsion between the oxidized ferrocene cation and the positively biased metal surface may therefore stretch the helix from R-helix to 310-helix. Such a molecular dynamics driven by electrostatic force between a molecular terminal and a metal surface has also been reported in a selfassembled monolayer.92 However, the extremely small β of 0.02-0.04 Å-1 in our previous report may not be well explained by this mechanism. It is no more than a speculation without any quantitative discussion at this stage. Another possible mechanism is electron hopping among the amide groups (Figure 10c).11,12,17 Our proposed mechanism is that the nearest amide group to gold first transfers an electron to gold (possibly via the sulfur atom) to form a cation radical of the amide group (hole), and the hole hops among the amide groups until it reaches to the ferrocene moiety to complete the process. Schlag and co-workers have studied a hole hopping process along a peptide chain initiated by photoionization of a residue having a lower ionization potential in a gas phase.42-46,93 They theoretically demonstrated that the dihedral angles of the peptide backbone significantly change to form an active

conformation when a hole is readily transferred from one amide group to another. In the active conformation, a strong electronic coupling is attained between the neutral form and cation radical of amide groups and also the activation barrier is negligible. If the hole injection is possible from gold to the amide group and is the rate-determining step, the local vibrational motion at the interface may be hindered by tight molecular packing of a monolayer to suppress the hole hopping process resulting in decrease of the overall electron transfer rate. This explanation is based on the idea that the electron hopping process among the amides should be fast because an electronic coupling between neighboring amide groups in a helix should be strong as reported in the literature.5 However, the hole hopping among the amides may happen to be slowed down to the same order as the hole injection from gold under the suppressed local motion due to the hydrogen bonding network in a helix. The tight molecular packing therefore may suppress not only the hole injection but also the hole hopping in this case. A question about the hole hopping in the SAM is a plausibility of interchain hole jump, since the dependence of the electron transfer on the direct distance indicates intermolecular electron transfer from one peptide chain to another to take a shortcut. Generally, such an interchain hole jump is not certain because each peptide backbone is separated by at least the size of the side chain. However, it may be possible in the Ala-Aib peptide SAM with the closest molecular packing. Then the slow electron transfer

Effects of Monolayer Structures in the Leu-Aib peptide SAM can be explained by suppression of the interchain hole jump due to the increase of the interchain distance. Conclusions In this study, various types of self-assembled monolayers were prepared on gold from helical peptides carrying a ferrocene moiety at the terminal, and the effects by backbone separation, molecular orientation, and molecular packing on the electron transfer from the ferrocene moiety to gold were examined by electrochemical measurements. A leucine-based 16mer peptide monolayer, an alanine-based 16mer peptide monolayer, and mixed monolayers of the alanine-based 16mer and a 8mer, 12mer, or 16mer peptide without a ferrocene moiety, were compared. Infrared spectroscopy and ellipsometry confirmed that the peptides form a well-defined monolayer with vertical orientation with helix tilt angles of 29-49°. The standard electron transfer rate constants ranged from 50 to 1300 s-1 depending on the molecular orientation and molecular packing. A general conclusion is that more vertical orientation and tighter packing of the monolayer suppress the electron transfer, indicating that the electron transfer occurs intermolecularly and the electron transfer is associated with some molecular motion. To explain the molecular dynamics effect, three different mechanisms are proposed: (1) electron tunneling gated by a global helix contraction, (2) electron tunneling coupled with helix conversion from R-helix to 310-helix, and (3) electron hopping among the amide groups assisted by a local backbone motion. To determine the mechanism, further experiments on temperature dependence, distance dependence extended to much longer regions, and effects of underlying metals, are underway. Acknowledgment. This work is partly supported by Grantin-Aids for Young Scientists B (16750098), for Young Scientists A (20685009), for Exploratory Research (17655098), and for Scientific Research B (15350068), and Global COE program, International Center for Integrated Research and Advanced Education in Materials Science, from the Ministry of Education, Culture, Sports, Science, and Technology, Japan. References and Notes (1) Mayo, S. L.; Ellis, W. R.; Crutchley, R. J.; Gray, H. B. Science 1986, 233, 948–952. (2) Beratan, D. N.; Onuchic, J. N.; Winkler, J. R.; Gray, H. B. Science 1992, 258, 1740–1741. (3) Langen, R.; Chang, I. J.; Germanas, J. P.; Richards, J. H.; Winkler, J. R.; Gray, H. B. Science 1995, 268, 1733–1735. (4) Isied, S. S. Prog. Inorg. Chem. 1984, 32, 443–517. (5) Shin, Y. G. K.; Newton, M. D.; Isied, S. S. J. Am. Chem. Soc. 2003, 125, 3722–3732. (6) Long, Y. T.; Abu-Rhayem, E.; Kraatz, H. B. Chem.sEur. J. 2005, 11, 5186–5194. (7) Hol, W. G. J. Prog. Biophys. Mol. Biol. 1985, 45, 149–195. (8) Vassilian, A.; Wishart, J. F.; Vanhemelryck, B.; Schwarz, H.; Isied, S. S. J. Am. Chem. Soc. 1990, 112, 7278–7286. (9) Ogawa, M. Y.; Wishart, J. F.; Young, Z. Y.; Miller, J. R.; Isied, S. S. J. Phys. Chem. 1993, 97, 11456–11463. (10) Sisido, M.; Hoshino, S.; Kusano, H.; Kuragaki, M.; Makino, M.; Sasaki, H.; Smith, T. A.; Ghiggino, K. P. J. Phys. Chem. B 2001, 105, 10407–10415. (11) Morita, T.; Kimura, S. J. Am. Chem. Soc. 2003, 125, 8732–8733. (12) Watanabe, J.; Morita, T.; Kimura, S. J. Phys. Chem. B 2005, 109, 14416–14425. (13) Galka, M. M.; Kraatz, H. B. ChemPhysChem 2002, 3, 356–359. (14) Kraatz, H. B.; Bediako-Amoa, I.; Gyepi-Garbrah, S. H.; Sutherland, T. C. J. Phys. Chem. B 2004, 108, 20164–20172. (15) Sek, S.; Tolak, A.; Misicka, A.; Palys, B.; Bilewicz, R. J. Phys. Chem. B 2005, 109, 18433–18438. (16) Mandal, H. S.; Kraatz, H. B. Chem. Phys. 2006, 326, 246–251.

J. Phys. Chem. B, Vol. 112, No. 40, 2008 12849 (17) Kai, M.; Takeda, K.; Morita, T.; Kimura, S. J. Pept. Sci. 2008, 14, 192–202. (18) Xiao, X. Y.; Xu, B. Q.; Tao, N. J. J. Am. Chem. Soc. 2004, 126, 5370–5371. (19) Sek, S.; Swiatek, K.; Misicka, A. J. Phys. Chem. B 2005, 109, 23121–23124. (20) Sek, S.; Misicka, A.; Swiatek, K.; Maicka, E. J. Phys. Chem. B 2006, 110, 19671–19677. (21) Antonello, S.; Formaggio, F.; Moretto, A.; Toniolo, C.; Maran, F. J. Am. Chem. Soc. 2003, 125, 2874–2875. (22) Polo, F.; Antonello, S.; Formaggio, F.; Toniolo, C.; Maran, F. J. Am. Chem. Soc. 2005, 127, 492–493. (23) Hsu, C. P. J. Electroanal. Chem. 1997, 438, 27–35. (24) Xu, B. Q.; Tao, N. J. J. Science 2003, 301, 1221–1223. (25) Liu, B.; Bard, A. J.; Mirkin, M. V.; Creager, S. E. J. Am. Chem. Soc. 2004, 126, 1485–1492. (26) Morita, T.; Lindsay, S. J. Am. Chem. Soc. 2007, 129, 7262–7263. (27) Malak, R. A.; Gao, Z. N.; Wishart, J. F.; Isied, S. S. J. Am. Chem. Soc. 2004, 126, 13888–13889. (28) Dey, S. K.; Long, Y. T.; Chowdhury, S.; Sutherland, T. C.; Mandal, H. S.; Kraatz, H. B. Langmuir 2007, 23, 6475–6477. (29) Adams, D. M.; Brus, L.; Chidsey, C. E. D.; Creager, S.; Creutz, C.; Kagan, C. R.; Kamat, P. V.; Lieberman, M.; Lindsay, S.; Marcus, R. A.; Metzger, R. M.; Michel-Beyerle, M. E.; Miller, J. R.; Newton, M. D.; Rolison, D. R.; Sankey, O.; Schanze, K. S.; Yardley, J.; Zhu, X. Y. J. Phys. Chem. B 2003, 107, 6668–6697. (30) Segal, D.; Nitzan, A.; Davis, W. B.; Wasielewski, M. R.; Ratner, M. A. J. Phys. Chem. B 2000, 104, 3817–3829. (31) Berlin, Y. A.; Burin, A. L.; Ratner, M. A. J. Am. Chem. Soc. 2001, 123, 260–268. (32) Petrov, E. G.; May, V. J. Phys. Chem. A 2001, 105, 10176–10186. (33) Petrov, E. G.; Shevchenko, Y. V.; Teslenko, V. I.; May, V. J. Chem. Phys. 2001, 115, 7107–7122. (34) Berlin, Y. A.; Burin, A. L.; Ratner, M. A. Chem. Phys. 2002, 275, 61–74. (35) Berlin, Y. A.; Hutchison, G. R.; Rempala, P.; Ratner, M. A.; Michl, J. J. Phys. Chem. A 2003, 107, 3970–3980. (36) Berlin, Y. A.; Ratner, M. A. Radiat. Phys. Chem. 2005, 74, 124– 131. (37) Davis, W. B.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2001, 123, 7877–7886. (38) Weiss, E. A.; Ahrens, M. J.; Sinks, L. E.; Gusev, A. V.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2004, 126, 5577–5584. (39) Weiss, E. A.; Tauber, M. J.; Kelley, R. F.; Ahrens, M. J.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2005, 127, 11842–11850. (40) Petrov, E. G.; Hanggi, P. Phys. ReV. Lett. 2001, 86, 2862–2865. (41) Petrov, E. G.; May, V.; Hanggi, P. Chem. Phys. 2002, 281, 211– 224. (42) Weinkauf, R.; Schanen, P.; Yang, D.; Sonkara, S.; Schlag, E. W. J. Phys. Chem. 1995, 99, 11255–11265. (43) Weinkauf, R.; Schanen, P.; Metsala, A.; Schlag, E. W.; Burgle, M.; Kessler, H. J. Phys. Chem. 1996, 100, 18567–18585. (44) Schlag, E. W.; Sheu, S. Y.; Yang, D. Y.; Selzle, H. L.; Lin, S. H. J. Phys. Chem. B 2000, 104, 7790–7794. (45) Sheu, S. Y.; Schlag, E. W.; Yang, D. Y.; Selzle, H. L. J. Phys. Chem. A 2001, 105, 6353–6361. (46) Schlag, E. W.; Sheu, S. Y.; Yang, D. Y.; Selzle, H. L.; Lin, S. H. Angew. Chem., Int. Ed. 2007, 46, 3196–3210. (47) Orlowski, G. A.; Chowdhury, S.; Kraatz, H. B. Langmuir 2007, 23, 12765–12770. (48) Chidsey, C. E. D. Science 1991, 251, 919–922. (49) Sikes, H. D.; Smalley, J. F.; Dudek, S. P.; Cook, A. R.; Newton, M. D.; Chidsey, C. E. D.; Feldberg, S. W. Science 2001, 291, 1519–1523. (50) Smalley, J. F.; Finklea, H. O.; Chidsey, C. E. D.; Linford, M. R.; Creager, S. E.; Ferraris, J. P.; Chalfant, K.; Zawodzinsk, T.; Feldberg, S. W.; Newton, M. D. J. Am. Chem. Soc. 2003, 125, 2004–2013. (51) Smalley, J. F.; Sachs, S. B.; Chidsey, C. E. D.; Dudek, S. P.; Sikes, H. D.; Creager, S. E.; Yu, C. J.; Feldberg, S. W.; Newton, M. D. J. Am. Chem. Soc. 2004, 126, 14620–14630. (52) Creager, S.; Yu, C. J.; Bamdad, C.; O’Connor, S.; MacLean, T.; Lam, E.; Chong, Y.; Olsen, G. T.; Luo, J. Y.; Gozin, M.; Kayyem, J. F. J. Am. Chem. Soc. 1999, 121, 1059–1064. (53) Kitagawa, K.; Morita, T.; Kawasaki, M.; Kimura, S. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 3493–3500. (54) Kitagawa, K.; Morita, T.; Kimura, S. J. Phys. Chem. B 2004, 108, 15090–15095. (55) Fujita, K.; Bunjes, N.; Nakajima, K.; Hara, M.; Sasabe, H.; Knoll, W. Langmuir 1998, 14, 6167–6172. (56) Yasutomi, S.; Morita, T.; Imanishi, Y.; Kimura, S. Science 2004, 304, 1944–1947. (57) Tsuboi, M. J. Polym. Sci. 1962, 59, 139. (58) Gremlich, H. U.; Fringeli, U. P.; Schwyzer, R. Biochemistry 1983, 22, 4257–4264.

12850 J. Phys. Chem. B, Vol. 112, No. 40, 2008 (59) Kennedy, D. F.; Crisma, M.; Toniolo, C.; Chapman, D. Biochemistry 1991, 30, 6541–6548. (60) Steinem, C.; Janshoff, A.; Ulrich, W. P.; Sieber, M.; Galla, H. J. Biochim. Biophys. ActasBiomembr. 1996, 1279, 169–180. (61) Yamada, T.; Nango, M.; Ohtsuka, T. J. Electroanal. Chem. 2002, 528, 93–102. (62) Creager, S. E.; Wooster, T. T. Anal. Chem. 1998, 70, 4257–4263. (63) Wu, X. Z.; Zhang, W. Z.; Hou, Y. J. Electroanal. Chem. 1995, 398, 1–4. (64) Protsailo, L. V.; Fawcett, W. R. Electrochim. Acta 2000, 45, 3497– 3505. (65) Benedetti, E.; Di Blasio, B.; Pavone, V.; Pedone, C.; Santini, A.; Crisma, M.; Toniolo, C. In Molecular Conformation and Biological Interactions; Balaram, P., Ramaseshan, S., Eds.; Indian Academy of Science: Bangalore, India, 1991; pp 497-502. (66) Otoda, K.; Kitagawa, Y.; Kimura, S.; Imanishi, Y. Biopolymers 1993, 33, 1337–1345. (67) Fujita, K.; Kimura, S.; Imanishi, Y.; Okamura, E.; Umemura, J. Langmuir 1995, 11, 1675–1679. (68) Brown, A. P.; Anson, F. C. Anal. Chem. 1977, 49, 1589–1595. (69) Sek, S.; Palys, B.; Bilewicz, R. J. Phys. Chem. B 2002, 106, 5907– 5914. (70) Niwa, M.; Morikawa, M.; Higashi, N. Angew. Chem., Int. Ed. 2000, 39, 960–963. (71) Finklea, H. O.; Ravenscroft, M. S.; Snider, D. A. Langmuir 1993, 9, 223–227. (72) Finklea, H. O.; Hanshew, D. D. J. Am. Chem. Soc. 1992, 114, 3173– 3181. (73) Sek, S.; Misicka, A.; Bilewicz, R. J. Phys. Chem. B 2000, 104, 5399–5402. (74) Finklea, H. O.; Yoon, K.; Chamberlain, E.; Allen, J.; Haddox, R. J. Phys. Chem. B 2001, 105, 3088–3092. (75) Idiris, A.; Alam, M. T.; Ikai, A. Protein Eng. 2000, 13, 763–770.

Takeda et al. (76) Papamokos, G. V.; Demetropoulos, I. N. J. Phys. Chem. A 2004, 108, 8160–8169. (77) Dicapua, F. M.; Swaminathan, S.; Beveridge, D. L. J. Am. Chem. Soc. 1991, 113, 6145–6155. (78) Arteca, G. A. Macromolecules 1996, 29, 7594–7601. (79) Pak, Y.; Wang, S. M. J. Chem. Phys. 1999, 111, 4359–4361. (80) Zanuy, D.; Aleman, C.; Munoz-Guerra, S. Macromol. Theo. Simul. 2000, 9, 543–549. (81) Efrima, S.; Bixon, M. Chem. Phys. Lett. 1974, 25, 34–37. (82) Efrima, S.; Bixon, M. Chem. Phys. 1976, 13, 447–460. (83) Jortner, J. J. Chem. Phys. 1976, 64, 4860–4867. (84) Jortner, J.; Bixon, M. J. Chem. Phys. 1988, 88, 167–170. (85) Jortner, J.; Bixon, M.; Langenbacher, T.; Michel-Beyerle, M. E Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 12759–12765. (86) Redmore, N. P.; Rubtsov, I. V.; Therien, M. J. J. Am. Chem. Soc. 2003, 125, 8769–8778. (87) Marcus, R. A. J. Chem. Phys. 1965, 43, 679–701. (88) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265– 322. (89) Weber, K. S.; Creager, S. E. J. Electroanal. Chem. 1998, 458, 17– 22. (90) Orlowski, G. A.; Kraatz, H. B. Electrochim. Acta 2006, 51, 2934– 2937. (91) Kitagawa, K.; Morita, T.; Kimura, S. Angew. Chem., Int. Ed. 2005, 117, 6488–6491. (92) Lahann, J.; Mitragotri, S.; Tran, T. N.; Kaido, H.; Sundaram, J.; Choi, I. S.; Hoffer, S.; Somorjai, G. A.; Langer, R. Science 2003, 299, 371–374. (93) Schlag, E. W.; Sheu, S. Y.; Yang, D. Y.; Selzle, H. L.; Lin, S. H Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 1068–1072.

JP805711V