pubs.acs.org/NanoLett
Single-Molecule Mapping of Long-range Electron Transport for a Cytochrome b562 Variant Eduardo Antonio Della Pia,†,§ Qijin Chi,*,‡ D. Dafydd Jones,*,§ J. Emyr Macdonald,*,† Jens Ulstrup,‡ and Martin Elliott† †
School of Physics and Astronomy, Cardiff University, Queens’s Building, The Parade, Cardiff CF24 3AA, U.K., Department of Chemistry and NanoDTU, Technical University of Denmark, DK-2800 Lyngby, Denmark, and § School of Biosciences, Main Building, Cardiff University, Cardiff CF10 3AT, U.K. ‡
ABSTRACT Cytochrome b562 was engineered to introduce a cysteine residue at a surface-exposed position to facilitate direct selfassembly on a Au(111) surface. The confined protein exhibited reversible and fast electron exchange with a gold substrate over a distance of 20 Å between the heme redox center and the gold surface, a clear indication that a long-range electron-transfer pathway is established. Electrochemical scanning tunneling microscopy was used to map electron transport features of the protein at the single-molecule level. Tunneling resonance was directly imaged and apparent molecular conductance was measured, which both show strong redox-gated effects. This study has addressed the first case of heme proteins and offered new perspectives in singlemolecule bioelectronics. KEYWORDS Single-molecule electronics, nanobioelectronics, cytochrome b562, protein engineering, scanning tunneling microscopy, redox-gated tunneling resonance
M
any processes central to chemistry and biology such as photosynthesis, respiration, and enzyme electrocatalysis are driven by long-range protein electron transfer (LrET); the mechanism of electron transfer and the organization of the electron transfer elements in these protein systems are still not fully understood and remain a major area of research.1-4 From a new perspective, molecular electronics has recently emerged as a fascinating scientific frontier in the context of nanoscale science and technology. The studies of molecular electronics have provided significant advances in our understanding and measurement of charge transport events in various nanostructures.5 The development of ultrasensitive tools such as fluorescence spectroscopy, scanning tunneling microscopy (STM), and STM based techniques has led to a dramatic improvement over the past decade in the spatial resolution for mapping molecular charge transport. For small model organic molecules and transition metal complexes, measurements of their charge transport and molecular conductance have been achieved at the single-molecule level.6-19 However, the strategies used for small molecules has only limited applicability for biomacromolecules such as metalloproteins and enzymes. In terms of metalloproteins, the few reported cases have been largely limited to a blue copper protein azurin.20-25 The challenges for proteins arise mainly from their structural complexity and inherent dynamics
making measurements difficult. This is due to difficulties in the assembly of protein molecules to achieve sufficient stability, retention of biological function under the required conditions and control of molecular orientation to enable direct electron transfer between protein molecule and a solid electrode. These challenges can be partially overcome either by the molecular wiring assembly or by engineering the target protein through designed changes to the protein’s amino acid sequence. In the former approach, an external linking molecule is used to wire the protein molecule onto an electrode surface and to simultaneously control its orientation to promote direct electron transfer.26 Site-directed mutagenesis is employed in the latter approach to introduce linking residues such as cysteine and histidine on protein surfaces at defined positions based on analysis of the threedimensional (3D) atomic-level structure of the protein. This allows direct protein assembly on a solid electrode surface for single-molecule study. Electrochemical STM (ECSTM) is a technique that combines electrochemical control and STM high-resolution signatures such as molecular imaging and scanning tunneling spectroscopy (STS).19 The possibility of using ECSTM to observe single-molecule charge transport was speculated upon in the early 1990s.27 The first experimental demonstration was pioneered by Tao using a system of iron porphyrin molecules adsorbed on a highly ordered pyrolytic graphite (HOPG) surface, and the redox-tuned resonant tunneling effect was directly visualized by STM imaging.28 Since then, ECSTM has provided a powerful approach to study interfacial electron transfer and molecular conduc-
* To whom correspondence should be addressed. E-mail: (Q.C.)
[email protected]; (D.J.)
[email protected]; (J.E.M.)
[email protected]. Received for review: 09/21/2010 Published on Web: 11/24/2010 © 2011 American Chemical Society
176
DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176–182
FIGURE 1. (a) Three-dimensional structure and sizes of cyt b562 D50C, (b) a schematic illustration of cyt b562 D50C self-assembled, along with small molecules 1,4-dithiothreitol (DTT) on a Au(111) surface, and (c,d) three-dimensional STM images obtained in a phosphate buffer (10 mM, pH 6.2) under electrochemical control. The ECSTM imaging parameters: the tunneling current (It) ) 35 pA, bias voltage (Vb) ) -0.4 V, working electrode potential (Ew) ) -0.10 V vs SCE; image sizes (c) 150 nm × 150 nm and (d) 90 nm × 90 nm.
tance of electroactive species at the single-molecule level in an electrochemical environment.13,19,29 Most studies have, however, focused on the STS measurements rather than STM imaging for direct visualization of single-molecular charge transport.8-19 In this study, we have taken the advantages of singlecrystal electrochemistry and ECSTM as powerful tools to disclose the electronic structures and properties of a bacterial electron transfer protein, cytochrome b562 (cyt b562),30,31 selfassembled on a Au(111) surface in a natural “quasi-physiological” environment. Cyt b562 is a four-helix bundle protein comprised of 106 residues that binds a single heme group, a biologically active redox cofactor via axial amino acid residues, a biologically important redox-active cofactor. The protein binds heme close to the termini through two ligands that coordinate to the heme iron moiety; the Sδ atom of a methionine at position 7 and Nε2 atom of a histidine at position 102. Cyt b562 has a reduction potential of +167 mV versus NHE,32 which is lower than those for c-type cytochromes (+230 to +270 mV vs NHE) in which the heme group is also bound via thio- either links to the porphyrin ring to the protein. Cyt b562 has proved a very useful model for understanding molecular properties of proteins and has been the subject of extensive structure, folding, and proteinengineering studies.33 To facilitate direct self-assembly of cyt b562 on a metal surface such as gold or platinum, the protein has been engineered to introduce a thiol-linking group via the amino acid cysteine on the protein surface. Site-directed mutagen© 2011 American Chemical Society
esis was used to generate the cyt b562 D50C variant (Figure 1a), in which aspartic acid (D) at position 50 was substituted with cysteine (C). Spectrophotometry and hemin titration indicate that the D50C mutation had no detectable effect on the structure or heme binding properties of cyt b562 (Figures S1 and S2 in the Supporting Information). The mutation had little effect on the redox potential of cyt b562 (vide infra). Cyt b562 D50C was observed to self-assemble on a Au(111) surface via thiol-gold chemistry to form a monolayer or submonolayer, depending on the protein concentration and adsorption time. To facilitate single-molecule measurements, 1,4-dithiothreitol (DTT) was coadsorbed with the protein to yield a mixed monolayer (Figure 1b). Here DTT played two crucial roles: (1) to act as a reducing agent to prevent the formation of protein dimers in solution via disulfide bonds, and (2) to serve as a surface diluent to generate isolated protein molecules. With such a refined molecular assembly, high-resolution STM imaging of individual holo-cyt b562 D50C (heme bound to protein) molecules was consistently achieved in a biological-friendly buffer and under electrochemical control. Figure 1c,d shows two threedimensional (3D) STM images with different scanned areas as examples. The image in Figure 1c was obtained from the samples prepared with relatively high protein population, while the low surface coverage (as shown in Figure 1d) is favored for single-molecule measurements. Further examples of STM images of proteins over a range of scanned areas are provided in the Supporting Information (Figure S4). 177
DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176-–182
grams (CVs) recorded for this system. A pair of well-defined redox peaks were observed, whereas no Faradaic signal was detected for either wild-type cyt b562 or apo-cyt b562 D50C. The formal redox potential (Eo′), estimated from CVs, was -80 ( 6 mV versus SCE (equivalent to +165 mV vs NHE). The Eo′ value agrees well with that for wild-type cyt b562 determined by redox titration32 and measured by electrochemistry with the protein in homogeneous solution,34 although a slightly more positive redox potential (ca. -10 mV vs SCE) was recently reported for the wild-type cyt b562 immobilized electrostatically on the amine-terminal SAMs on a silver electrode.35 In addition, it is noted that the capacitive background current of holo-protein monolayers is significantly larger than that for apoprotein monolayers (Figure 2b). This is most likely attributed to structural differences between apo- and holocyt b562 (Figure S3 in the Supporting Information). The apocyt b562 D50C is partially folded and is much less ordered. The hydrophobic patches in the polypeptide could thus be exposed partially to the electrolyte solution to increase surface hydrophobicity of apoprotein monolayers. In addition to the absorbance spectra (Figure S1 in the Supporting Information), the electrochemical results further confirm that the D50C mutation does not disrupt the protein. The observed linear relationship between the redox peak currents and scan rates (Figure S7 in the Supporting Information) is a clear indication of a diffusionless electrochemical electron transfer reaction and supports the fact that the redox protein is confined on the electrode surface. Furthermore, this system is robust enough to acquire CVs with high scan rates (Figure S8 in the Supporting Information). The electron transfer rate constant can thus be estimated using theLavironmethod(FigureS9intheSupportingInformation),36,37 which gave a value of 44 s-1 that is quite fast in view of a 20 Å electron transfer distance. This demonstrates that introducing a cysteine in place of a non-gold binding aspartate at residue 50 establishes highly efficient electron transfer pathways between the heme center and the gold electrode. Direct visualization of single-molecule protein electron transfer by STM was conducted in the constant-current mode at a small tunneling current of 35 pA. STM imaging was first performed over a larger scan area such as 150 × 150 nm2 to obtain images of the protein submonolayer at molecular resolution and then focused on a few individual molecules by gradually reducing the scan area. By keeping a constant bias voltage between the substrate and the tip, STM imaging started with the substrate potential set around the equilibrium redox potential of cyt b562 D50C (i.e., zero overpotential, η ) Ew - Eo′). Imaging was continued toward either positive or negative overpotentials by adjusting the substrate and tip potentials in parallel (i.e., with fixed bias voltage) and finally returned to the equilibrium potential. As a result, a series of STM images at various overpotentials were acquired. Figure 3 shows some representative images focused on seven protein molecules. The single-molecule
FIGURE 2. Cyclic voltammograms of (a) reductive desorption for the holocyt b562 D50C and DTT mixture monolayers and of (b) Faradic responses for holocyt b562 D50C (red line) and apocyt b562 D50C (blue line) molecules self-assembled on Au(111) surfaces in a phosphate buffer (10 mM, pH 6.2). Scan rates: (a) 50 and (b) 500 mV s-1.
Electrochemical measurements were carried out using a three-electrode system. The experimental details are described in the Supporting Information. The wild-type protein (no cysteine mutation) and apo-cyt b562 D50C (no heme bound to protein) were used as control references (or controls). The 3D STM images (Figure 1c,d) suggest that cyt b562 D50C was bound in the desired molecular orientation (Figure 1b). The direct surface self-assembly of protein molecules was facilitated via Au-S bonding. This is supported clearly by the feature of reductive desorption observed for the mixed monolayers, as shown in Figure 2a. Two cathodic peaks at -0.70 and -0.79 V (vs SCE) correspond to the reductive desorption of DTT and protein molecules from the gold surface, respectively. With a molecular assembly illustrated in Figure 1b the distance between the heme redox center and the Au(111) surface measured would then be at least 20 Å (Figure 1a), according with the notion of long-range electron transfer. However, direct electron transfer is consistently detected by normal cyclic voltammetry, indicating that electronic coupling between the heme center and the electrode is quite effective. Figure 2b shows an example of cyclic voltammo© 2011 American Chemical Society
178
DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176-–182
FIGURE 3. ECSTM images for direct visualization of single-molecule protein electron transfer arising from individual cytochrome b562 D50C molecules with imaging focused on seven protein molecules. The images obtained at different working overpotentials: (a) -250, (b) -10, (c) +50, (d) +150, and (e) +200 mV. The bias voltage (Vb) ) -0.4 V, and tunneling current (It) ) 35 pA. The size of images: 32 nm × 32 nm. Other experimental conditions are the same as those in Figure 1c,d.
tunneling contrast is clearly tuned by the redox state of the protein, with a maximum contrast around the overpotential of -10 to -20 mV (Figure 3b), very close to the equilibrium redox potential. The contrast decreases upon applying either negative (Figure 3a) or positive (Figure 3c-e) overpotentials with the effects being quite symmetric. The dependence of the apparent contrast height on the overpotentials applied is given in the Supporting Information (Figure S5). The apparent contrast enhancement can be up to 2 Å, equivalent to an increase in the tunneling current by ca. 1 order of magnitude assuming that the decay factor of the tunneling current is 1.0 Å-1.28 This is further demonstrated by the correlation between the normalized contrast and the overpotential (Figure 4a) that shows a maximum increase by a factor of 2-3. As a comparison, similar STM imaging was performed but at a fixed tip potential, which means that the bias voltage changed with changing the working electrode potentials. The correlation between the apparent tunneling contrast and the overpotential (Figure 4b) is largely unchanged across the overpotentials applied under such experimental conditions where two external parameters (i.e., the bias voltage and substrate potential) are changed simultaneously. This is further discussed below. The experimental observations represented in Figure 4a,b can be explained by a two-step electron transfer mechanism in the STM redox process. The energy levels of the STM substrate (or working electrode), the STM tip, and a redox molecule located in the substrate-tip gap may all be modified by changing the substrate potential, but the difference between the substrate and tip energy levels will remain constant if the bias voltage is fixed as the experimental conditions applied for Figures 3 and 4a. As a consequence, the redox level is shifted relative to the substrate and tip Fermi levels by the overpotential. Furthermore, the redox level is strongly coupled to the environment with an initially oxidized (vacant) level above the tip Fermi level and an initially reduced (occupied) level below the substrate Fermi level, when the bias voltage (Vb ) ET - EW, where ET and EW are the potentials applied at the substrate and tip, respectively) is negative as for the present case. Nuclear fluctuations initially bring the redox levels into the energy © 2011 American Chemical Society
region close to the Fermi level of the electrodes, which in turn induces a two-step electron transfer process.38 At a negative bias voltage, electron transfer proceeds first from the tip to the vacant redox level and then from the temporarily occupied redox level to the substrate. The tunneling current (It) is thus gated by the molecular redox level, which is displayed directly by the changes in STM contrast. In other words, the contrast changes observed are due to the redoxgated tunneling resonance. In contrast, the bias voltage will change with changing the substrate potential when the tip potential is fixed. The molecular energy levels can be mostly located either above or below the tip Femi level, resulting in no significant resonant tunneling tuned by the substrate potential. In the two-step STM redox model, the tunneling current (It) can be approximately quantified by the interfacial electron transfer rate constants, leading to a transparent view on the ECSTM electron transfer mechanisms. In principle, long-range interfacial electron transfer (>20 Å) as in the present case suggests that the electronic coupling between the heme group and the enclosing substrate and tip electrodes is relatively weak and the diabatic weak-coupling notion should apply. In addition, weak coupling also means that the tunneling current is small and only a single or a few electrons are transmitted in a single ECSTM event. However, significant tunneling currents are observed throughout the overpotential range traversing the equilibrium redox potential. This implies that the electronic coupling through the Au-S contact, while still weak enough to maintain the functional integrity of the immobilized protein, can bring the overall electron transfer process much closer to the adiabatic limit of electrode-protein interactions. The following simple two-step tunneling current form illuminates the present situation. The tunneling current (It) dependence on the effective overpotential (eξη) and the bias voltage (eVbias) can be recast in the steady-state electron transfer form in the following combination.19,38
It ) 2en
179
b k r/o k o/rb k +b k r/o
bo/r
(1)
DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176-–182
kr/o can be given in the following The rate constants b ko/rand b forms at relatively small bias voltage.20
b k o/r ) κtFt
b k r/o ) κsFs
(
)
(λ - eVbias + eξη + eγVbias)2 ωeff 2kBT exp 2π αs 4λkBT
(2)
)
(3)
where κt and κs are electronic transmission coefficients (toward unity in the adiabatic limit) for electron transfer between the tip and the protein, and between the protein and the substrate, respectively; Ft and Fs are the electronic level densities of the tip and the substrate; ωeff is the effective nuclear vibrational frequency, Rt and Rs are the transfer coefficients for electron transfer between the tip and the protein and between the protein and the substrate, respectively; λ is the reorganization free energy; ξ is the fraction of the substrate-solution potential drop, η is the overpotential; γ is the fraction of the bias voltage drop at the site of the molecular redox center. Other symbols have their usual meaning. These equations support quantitatively the experimental observations above. As a note of observation, the parameters ξ and γ represent the electric potential distribution in the ECSTM tunneling gap. These two parameters are mutually dependent and determined by the ionic strength.39 However, further discussion of these crucial parameters is beyond the scope of this communication and will instead be addressed in a future report. To measure apparent molecular conductance, the tunneling current (It)-bias voltage (Vb) curves for single protein molecules were acquired at given substrate potentials. Examples of the It-Vb correlations are provided in the Supporting Information (Figure S11). Tunneling current-bias voltage relations of interfacial condensed matter singlemolecule redox processes offer new challenges as the molecular conduction mechanisms are quite different for “small” and “large” bias voltages.19,39 We shall offer such analysis elsewhere but presently address the data by a much simpler pragmatic frame developed recently by Vilan.40 This method is based on the Simmons tunneling model by an approximation approach leading to a linear presentation. A typical linearized presentation of experimental It-Vb data is given by eq 4.
FIGURE 4. Dependence of electronic tunneling contrast (a and b) and apparent molecular conductance (G0) (c) on working electrode overpotential (η ) E - Eo′). The data in (a) and (b) were obtained at a fixed bias voltage and a fixed tip potential, respectively. The solid line in (a) shows a theoretical fit and the solid line in (c) is a Gauss fit.
where b ko/r and b kr/o are the rate constants, respectively, for electron transfer between the tip and cyt b562 D50C and between the protein and the substrate. The arrows indicate the electron flow direction from the tip to the protein molecule and from the protein molecule to the substrate when a negative bias voltage is applied. n is the number of electrons transmitted in a single electron transfer event (n is large in the adiabatic limit) and e the electronic charge. © 2011 American Chemical Society
(
(λ - eξη - eγVbias)2 ωeff 2kBT exp 2π αt 4λkBT
It = G0(1 + CVb2) Vb
(4)
where G0 is the equilibrium conductance, that is, the first derivative of the current with respective to voltage near 0 V 180
DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176-–182
(i.e., G0 ) dI/dV|f0). Thus, a plot of It/Vb against the square of the bias voltage is expected to yield the two functional parameters, G0 and C. The experimental data were fitted by eq 4 to extract the G0 values at different overpotentials (Figure S12 in the Supporting Information). The dependence of apparent molecular conductance on the substrate potential is shown in Figure 4c. The apparent conductance of single protein molecules is modified (or tuned) significantly by their redox state, with a maximum conductance around the equilibrium redox potential. In addition, the protein molecules appear to be slightly more conductive in their reduced forms than in their oxidized state. The overall pattern is largely similar to that for the molecular electronic contrast (Figure 4a), but a much larger enhancement by a factor of 6-8 (cf. a factor of 2-3 for the contrast increase) is observed. However, the physical relation between redox-gated tunneling resonance and molecular conductance is complicated. This will constitute interesting and challenging issues for further theoretical studies to address electron transfer properties of electroactive species in the ECSTM configuration. In summary, engineering the electron transfer protein cyt b562 to incorporate a thiol-linking group in place of Asp50 allows not only the protein molecules to selfassemble robustly on a metal surface but also opens a feasible electron transfer pathway. Electronic communication between the protein and gold electrode can be directly detected by normal cyclic voltammetry with fast electron transfer rates even over a distance of 20 Å. Highresolution ECSTM enables us to visualize redox-gated tunneling resonance at the single-molecule level. The maximum enhancement (i.e., the on/off ratio) is up to a factor of 2-3, which is slightly smaller than that observed for the protein azurin (ca. 9)20 and significantly smaller than that for small molecules such as iron porphyrin (ca. 20)28 or Os-polypyridine complexes.17 The following factors may account for the relatively small apparent enhancement: (a) a relatively large bias (e.g., -0.4 V) is needed to obtain molecular resolution STM images for cyt b562 D50C. (b) The tunneling current needed to be set at a small value (35 pA) in comparison with the azurin case where the tunneling current was set at 100-200 pA. (c) Because of a larger distance (20 Å), electron transfer between the protein and the substrate is much slower (44 s-1) than that either for the iron porphyrin case where iron porphyrin molecules were in direct contact with the HOPG surface and electron transfer is almost completely reversible or for the azurin case where the distance between the copper redox center and the gold surface was only 11 Å leading to an electron transfer rate larger than 500 s-1. The apparent molecular conductance is also strongly dependent on the redox state of cyt b562 D50C, and the reduced form appears to be more conductive. The conductance of the cyt b562 molecules is enhanced by 6-8 folds close to its equilibrium redox state. We have thus © 2011 American Chemical Society
demonstrated that large and structurally complicated molecules of significant biological and bionanotechnological significance can be studied in a similar fashion to small redox molecules by combining smart protein engineering, surface self-assembly chemistry, and highresolution tools such as ECSTM. The ability to utilize, measure, and modulate the single-molecule electron transfer characteristics of a protein directly at a metal electrode without linker molecules and in a biologically friendly environment will further benefit the development of bioelectronics. Acknowledgment. We thank Dr. Jingdong Zhang for assistance in the STM experiments. E.D.P. was supported by a Cardiff University Richard Whipp Interdisciplinary Research studentship and the Charles Coles Traveling Scholarship. This work was supported by EPSRC (Grants EP/D076072/1 to J.E.M. and M.E.) and BBSRC (Grant BB/E001084 to D.D.J.) and by the Danish Research Council for Technology and Production Sciences (to Q.C. and J.U., Contract No. 274-07-0272). Supporting Information Available. Experimental materials and methods, DNA and protein sequences, UV-Vis spectra, hemin titration data, additional STM images and electrochemical data, and STM tunneling current-bias voltage relations. This material is available free of charge via the Internet at http://pubs.acs.org. REFERENCES AND NOTES (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)
181
Berg, J. M.; Tymoczko, J. L.; Stryer, L. Biochemistry: International Edition; W. H. Freeman: New York, 2006. (a) Gray, H. B.; Winger, J. R. Q. Rev. Biophys. 2003, 36, 341–372. (b) Gray, H. B.; Winger, J. R. Annu. Rev. Biochem. 1996, 65, 537– 561. Electron Transfer from Isolated Molecules to Biomolecules, Advances in Chemical Physics; Jortner, J., Bixon, M. , Eds.; Wiley: New York, 1998; pp 106-107. Kuznetrov, A. M.; Ulstrup, J. Electron Transfer in Chemistry and Biology: An Introduction to the Theory; Wiley: Chichester, U.K., 1999. (a) Nitzan, A.; Ratner, M. A. Science 2003, 300, 1384–1389. (b) Hush, N. S. Ann. N.Y. Acad. Sci. 2003, 1006, 1–20. Xu, B. Q.; Tao, N. J. Science 2003, 301, 1221–1223. Li, X.; He, J.; Hihath, J.; Xu, B. Q.; Lindsay, S. M.; Tao, N. J. J. Am. Chem. Soc. 2006, 128, 2135–2141. Tao, N. J. Nat. Nanotechnol. 2006, 1, 172–181. Chen, F.; Hihath, J.; Huang, Z.; Li, X.; Tao, N. J. Annu. Rev. Phys. Chem. 2007, 58, 535–564. Haiss, W.; Nichols, R. J.; Van Zalinge, H.; Higgins, S. J.; Bethell, D.; Schiffrin, D. J. Phys. Chem. Chem. Phys. 2004, 6, 4330–4337. Haiss, W.; Wang, C. S.; Grace, I.; Batsanov, A. S.; Schiffrin, D. J.; Higgins, S. J.; Bryce, M. R.; Lambert, C. J.; Nichols, R. J. Nat. Mater. 2006, 5, 995–1002. Martin, S.; Haiss, W.; Higgins, S. J.; Nichols, R. J. Nano Lett. 2010, 10, 2019–2023. Nichols, R. J.; Haiss, W.; Higgins, S. J.; Leary, E.; Martin, S.; Bethell, D. Phys. Chem. Chem. Phys. 2010, 12, 2801–2815. Li, Z.; Han, B.; Meszaros, G.; Pobelov, I.; Wandlowski, T.; Blaszczyk, A.; Mayor, M. Faraday Discuss. 2006, 131, 121–143. Pobelov, I. V.; Li, Z.; Wandlowski, T. J. Am. Chem. Soc. 2008, 130, 16045–1054. Li, Z.; Liu, Y.; Mertens, F. L.; Pobelov, I. V.; Wandlowski, T. J. Am. Chem. Soc. 2010, 132, 8187–8193. DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176-–182
(17) Albrecht, T.; Guckian, A.; Ulstrup, J.; Vos, J. G. Nano Lett. 2005, 5, 1451–1455. (18) Albrecht, T.; Moth-Poulsen, K.; Christensen, J. B.; Hjelm, J.; Bjørnholm, T.; Ulstrup, J. J. Am. Chem. Soc. 2006, 128, 6574–6575. (19) Zhang, J.; Kuznetsov, A. M.; Medvedev, I. G.; Chi, Q.; Albrecht, T.; Jensen, P. S.; Ulstrup, J. Chem. Rev. 2008, 108, 2737–2791. (20) Chi, Q.; Farver, O.; Ulstrup, J. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 16203–16208. (21) Alessandrini, A.; Salerno, M.; Frabboni, S.; Facci, P. Appl. Phys. Lett. 2005, 86, 133902. (22) Chi, Q.; Zhang, J.; Jensen, P. S.; Christensen, H. E. M.; Ulstrup, J. Faraday Discuss. 2006, 131, 181–195. (23) Alessandrini, A.; Corni, S.; Facci, P. Phys. Chem. Chem. Phys. 2006, 8, 4381–4397. (24) Davis, J. J.; Wang, N.; Morgan, A.; Zhang, T.; Zhao, J. Faraday Discuss. 2006, 131, 167–179. (25) Salverda, J. M.; Patil, A. V.; Mizzon, G.; Kuznetsova, S.; Zauner, G.; Akkilic, N.; Canters, G. W.; Davis, J. J.; Heering, H. A.; Aartsma, T. J. Angew. Chem., Int. Ed. 2010, 49, 5776–5779. (26) (a) Song, S.; Clark, R. A.; Bowden, E. F.; Tarlov, M. J. J. Phys. Chem. 1993, 97, 6564–6572. (b) Chi, Q.; Zhang, J.; Andersen, J. E. T.; Ulstrup, J. J. Phys. Chem. B 2001, 105, 4669–4679. (27) (a) Schmickler, W.; Widrig, C. J. Electroanal. Chem. 1992, 336, 213–217. (b) Kuznetsov, A. M.; Sommer-Larsen, P.; Ulstrup, J. Surf. Sci. 1992, 275, 52–60. (28) (a) Tao, N. J. Phys. Rev. Lett. 1996, 76, 4066–4069. (b) Schmickler, W.; Tao, N. J. Electrochim. Acta 1997, 42, 2809–2815.
© 2011 American Chemical Society
(29) Li, C.; Mishchenko, A.; Pobelov, I.; Wandlowski, T. Chimia 2010, 64, 383–390. (30) Feng, Y.; Sligar, S. G. Biochemistry 1991, 30, 10150–10155. (31) Hamada, K.; Bethge, P. H.; Mathews, F. S. J. Mol. Biol. 1995, 247, 947–962. (32) Springs, S. L.; Bass, S. E.; Bowman, G.; Nodelman, I.; Schutt, C. E.; McLendon, G. L. Biochemistry 2002, 41, 4321–4328. (33) See for examples: (a) Faraone-Mennella, J.; Gray, H. B.; Winkler, J. R. Proc. Natl. Acad.Sci. U.S.A. 2005, 102, 6315–6319. (b) Jones, D. D.; Barker, P. D. Angew. Chem., Int. Ed. 2005, 44, 6337–6341. (c) Baldwin, A. J.; Bader, R.; Christodoulou, J.; MacPhee, C. E.; Dobson, C. M.; Barker, P. D. J. Am. Chem. Soc. 2006, 128, 2162– 2163. (34) Barker, P. D.; Butler, J. L.; de Oliverira, P.; Hill, H. A. O.; Hunt, N. I. Inorg. Chim. Acta 1996, 252, 71–77. (35) Zuo, P.; Albrecht, T.; Barker, P. D.; Murgida, D. H.; Hildebrandt, P. Phys. Chem. Chem. Phys. 2009, 11, 7430–7436. (36) Laviron, E. J. Electroanal. Chem. 1979, 101, 19–28. (37) Jensen, P. S.; Chi, Q.; Grumsen, F. B.; Abad, J. M.; Horsewell, A.; Schiffrin, D. J.; Ulstrup, J. J. Phys. Chem. C 2007, 111, 6124–6132. (38) Zhang, J.; Chi, Q.; Kuznetsov, A. M.; Hansen, A. G.; Wackerbarth, H.; Christensen, H. E. M.; Andersen, J. E. T.; Ulstrup, J. J. Phys. Chem. B 2002, 106, 1131–1152. (39) Zhang, J.; Kuznetsov, A. M.; Ulstrup, J. J. Electroanal. Chem. 2003, 541, 133–146. (40) Vilan, A. J. Phys. Chem. C 2007, 111, 4431–4444.
182
DOI: 10.1021/nl103334q | Nano Lett. 2011, 11, 176-–182
