3022
Langmuir 2005, 21, 3022-3027
Indium Tin Oxide Electrodes Modified with Tris(2,2′-bipyridine-4,4′-dicarboxylic acid) Iron(II) and the Catalytic Oxidation of Tris(4,4′-di-tert-butyl-2,2′-bipyridine) Cobalt(II) C. Michael Elliott* Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523
Stefano Caramori and Carlo A. Bignozzi* Dipartimento di Chimica, Universita` di Ferrara, Via L. Borsari, 46, 44100 Ferrara, Italy Received October 26, 2004. In Final Form: January 13, 2005 Indium tin oxide (ITO) electrodes modified by attachment of tris(2,2′-bipyridine-4,4′-dicarboxylic acid) iron(II) are examined. The mode of attachment is believed to be via the COOH functions in a manner similar to attachment of similar carboxylate-containing compounds to TiO2 surfaces. On the surface the complex resides as a stable electrochemically active monolayer. These modified electrodes can efficiently catalyze the oxidation of certain cobalt complexes, specifically, tris(4,4′-di-tert-butyl-2,2′-bipyridine) cobalt(II). On the unmodified ITO surfaces this cobalt complex is essentially electrochemically inert. The catalytic process approaches diffusional control at very slow scan speeds. Also, the electro-catalysis is sufficiently efficient that the peak oxidation current for Co2+, under certain conditions, exceeds the ip for the surface oxidation of the adsorbed Fe2+ by >×100 and the current for the uncatalyzed oxidation of Co2+ by considerably more than that.
Introduction Over the past decade, interest in dye-sensitized wide band gap semiconductors and their application in photovoltaic cells has risen markedly.1,2 A product of this intense effort has been the observation that carboxylic acid functions can serve as relatively robust “anchoring functions” for the dye. Moreover, when the dye is a metal bipyridine complex having COOH functions attached directly to the bipyridine(s) in the 4 and 4′ positions, not only is the surface attachment of the complex robust, but also the electronic coupling into the semiconductor band structure appears to be strong.3,4 As a result of this apparent strong coupling, injection of photoexcited electrons of the dye into the conduction band of the semiconductor is exceedingly fast (e.g., femtoseconds).5 Herein we report electrochemical and X-ray photoelectron spectroscopy (XPS) studies on the iron(II) complex, Fe(DCB)32+, bound to indium tin oxide (ITO) semiconductor electrode surfaces. DCB is 2,2′-bipyridine-4,4′-dicarbo(1) Hagfeldt, A.; Didriksson, B.; Palmqvist, T.; Lindstroem, H.; Soedergren, S.; Rensmo, H. Verification of high efficiencies for the Graetzel-cell. A 7% efficient solar cell based on dye-sensitized colloidal TiO2 films. Sol. Energy Mater. Sol. Cells 1994, 31, 481-488. (2) Nazeeruddin, M. K.; Pechy, P.; Renouard, T.; Zakeeruddin, S. M.; Humphry-Baker, R.; Comte, P.; Liska, P.; Cevey, L.; Costa, E.; Shklover, V.; Spiccia, L.; Deacon, G. B.; Bignozzi, C. A.; Graetzel, M. Engineering of Efficient Panchromatic Sensitizers for Nanocrystalline TiO2-Based Solar Cells. J. Am. Chem. Soc. 2001, 123, 1613-1624. (3) Kils, K.; Mayo, E. I.; Brunschwig, B. S.; Gray, H. B.; Lewis, N. S.; Winkler, J. R. Anchoring Group and Auxiliary Ligand Effects on the Binding of Ruthenium Complexes to Nanocrystalline TiO2 Photoelectrodes. J. Phys. Chem. B 2004, 108, 15640-15651. (4) Kuciauskas, D.; Freund, M. S.; Gray, H. B.; Winkler, J. R.; Lewis, N. S. Electron Transfer Dynamics in Nanocrystalline Titanium Dioxide Solar Cells Sensitized with Ruthenium or Osmium Polypyridyl Complexes. J. Phys. Chem. B 2001, 105, 392-403. (5) Tachibana, Y.; Moser, J. E.; Graetzel, M.; Klug, D. R.; Durrant, J. R. Subpicosecond interfacial charge separation in dye-sensitized nanocrystalline titanium dioxide films. J. Phys. Chem. 1996, 100, 20056-20062.
xylic acidsthe same ligand used to attach the so-called N3 dye6,7 to TiO2 surfaces in dye-sensitized solar cell studies. In our own studies on dye-sensitized solar cells, we have observed and reported that certain cobalt bipyridine complexes can serve as reasonable replacements for the commonly employed I-/I3- redox mediator couple.8 Two reasons why these cobalt complexes are successful at mediating electron transfers in TiO2 cells are that (1) their heterogeneous electron transfer both on bare TiO2 and on the fluoride-doped tin oxide electron collector is extremely slow and (2) the reaction with the photooxidized N3 dye on the TiO2 surface is fast. As part of the present study, we examine the electrochemistry of one of these Co-based mediators on unmodified and Fe(DCB)32+-modified ITO surfaces. The modified ITO surface exhibits an impressive catalytic oxidation of the Co(II) complex. Experimental Section I. Materials. Hydrochloric and nitric acids of analytical grade were obtained from Mallinckrodt. Photoresist AZ 1512 was obtained from Clariant Corp. Reagents used for complex synthesis, electrode preparation, and functionalization [acetone, acetonitrile (ACN), dimethylformamide (DMF), ethanol, 2-propanol, and sodium hydroxide] were all ACS grade purchased from Fisher. Optima grade ACN from Fisher was used for the electrochemical studies. TBAPF6, 4,4′-di-tert-butyl-2,2′-bipyridine (6) Nazeeruddin, M. K.; Kay, A.; Rodicio, I.; Humphry-Baker, R.; Mueller, E.; Liska, P.; Vlachopoulos, N.; Graetzel, M. Conversion of light to electricity by cis-X2bis(2,2′-bipyridyl-4,4′-dicarboxylate) ruthenium(II) charge-transfer sensitizers (X ) Cl-, Br-, I-, CN-, and SCN-) on nanocrystalline titanium dioxide electrodes. J. Am. Chem. Soc. 1993, 115, 6382-6390. (7) Nazeeruddin, M. K.; Graetzel, M.; Rillema, D. P. Compounds of general interest. An improved synthesis of cis-dithiocyanato-bis(4,4′dicarboxy-2,2′-BPY)Ru(II) sensitizer. Inorg. Synth. 2002, 33, 185-189. (8) Sapp, S. A.; Elliott, C. M.; Contado, C.; Caramori, S.; Bignozzi, C. A. Substituted Polypyridine Complexes of Cobalt(II/III) as Efficient Electron-Transfer Mediators in Dye-Sensitized Solar Cells. J. Am. Chem. Soc. 2002, 124, 11215-11222.
10.1021/la047364f CCC: $30.25 © 2005 American Chemical Society Published on Web 02/24/2005
Modified Indium Tin Oxide Electrodes (DTB), and iron(II) perchlorate hydrate were purchased from Aldrich. 4,4′-Dimethyl-2,2′-bipyridine (DMB) was obtained from Reilley Industries and recrystalized from ethyl acetate. Cobalt(II) perchlorate hydrate was obtained from GFS Chemical Co. DCB was prepared according to the literature procedure.9 Unless otherwise indicated all reagents were used without further purification. Low resistance (4-8 Ω) ITO glass slides were purchased from Delta Technologies, Ltd., Stillwater, MN. II. Synthesis. [Co(DTB)3](ClO4)2 (I). Complex I was prepared according to previously reported literature procedures.8 DTB (3 equiv) was dissolved under vigorous magnetic stirring in methanol at room temperature; when dissolution was complete, 1 equiv of [Co(H2O)6](ClO4)2 was added. Immediately the solution turned brown, and the stirring was continued for 4 h. The solvent was removed by rotary evaporation resulting in a brown solid that was redissolved in the minimum amount of ACN and precipitated by addition of a large excess of diethyl ether. The solid was collected by suction filtration, washed twice with diethyl ether, and allowed to dry at room temperature. To characterize the complex by NMR spectroscopy, the Co(DTB)32+ was oxidized to the diamagnetic Co(DTB)33+ form by adding a slight excess of the strong oxidizer NOBF4 to an ACN solution of the reduced form. The oxidation was complete almost instantaneously. The mixed salt, [Co(DTB)3](ClO4)x(BF4)3-x, was collected after complete evaporation of the solvent at 50 °C under a reduced pressure. The slight excess of NOBF4 present readily decomposes to gaseous products during this procedure by reaction with trace water in the solvent. The 1H NMR is consistent with the D3 geometry of the complex in which all the bipyridine ligands are magnetically equivalent. 1H NMR, 300 MHz, CD OD: 9.0 ppm (d, 6H), 7.85 ppm 3 (m, 6H), 7.3 ppm (d, 6H), 1.5 ppm (s, 54 H). [Fe(DCB)3] (ClO4)2 (II). DCB (50 mg, 2.05 × 10-4 mol) was dissolved in hot DMF under magnetic stirring; after dissolution of the ligand, 17.5 mg (6.83 × 10-5 mol) of [Fe(H2O)6](ClO4)2 dissolved in 1 mL of DMF was slowly added. The solution immediately turned from colorless to dark red-purple. Heating was stopped, and the mixture was allowed to stir overnight at room temperature. Precipitation of the desired product was accomplished by addition of a large excess of diethyl ether. The purple solid was collected by suction filtration, washed twice with ether, and dried in an oven at 60 °C for about 4 h. The complex was characterized by cyclic voltammetry and ESI-MS (m/z ) 394.04 [M2+]) and used without further purification. III. Electrode Fabrication. To define a specific electrode area, patterned ITO electrodes were prepared according to the following procedure. Conductive ITO glass pieces (ca. 2.5 × 1 cm) were cleaned by rinsing with acetone, ethanol, and 2-propanol, to remove any oil and grease on the surface. After drying the electrodes under a nitrogen stream, Photoresist AZ 1512 was spin-coated onto the ITO surface at 2500 rpm for 40 s and baked in an oven at 90 °C for 20 min. An opaque mask of the desired shape and size was placed onto the electrodes which were then exposed to UV light generated by a Blak Ray 100 AP Long Wavelength UV lamp for about 1.5-2 min. The exposed photoresist is soluble in basic aqueous solution and was removed by immersion of the electrodes in a 0.5% NaOH solution for 15-20 s. The UV-unexposed photoresist remains intact protecting the underlying ITO from the next step. ITO was removed from the unprotected surface by immersion of the electrode in an “Aqua Regia solution” made of 45% concentrated HCl, 5% concentrated HNO3, and 50% water for 15-20 min. To ensure that the ITO layer was totally removed from the underlying insulating glass surface, the conductivity of the etched regions was checked. When the resistance was g20 MΩ the etch was deemed complete. After the etch, the protecting photoresist layer was removed with acetone, leaving a conductive ITO pattern of well-defined geometry on the glass surface. The active area of the resulting electrodes corresponds to a square of 0.49 cm2. Each electrode was carefully cleaned by sonication in an Alconox/deionized (DI) water solution (approximately 1/2 tbsp in 100 mL) for 15 min, rinsed with DI water and ethanol, sonicated (9) Nazeeruddin, M. K.; Kalyanasundaram, K.; Graetzel, M. Inorg. Synth. 1997, 32, 181.
Langmuir, Vol. 21, No. 7, 2005 3023 in ACN for 15 min, rinsed again with ACN, and blown dry with nitrogen. Finally, the surfaces were cleaned using a Harrick PDC3XG air plasma cleaner for 15 min prior to adsorption of the iron complex. IV. Functionalization of ITO Electrodes. Complex II (7 mg) was added to 10 mL of ACN. The dissolution of the complex was not complete, and the undissolved solid was removed by centrifugation. Freshly plasma-cleaned electrodes were immersed in the resulting clear red-purple solution and allowed to sit for 12 h under gentle heating (ca. 60 °C). The electrodes were removed from the solution, rinsed with ACN and ethanol, and dried under nitrogen prior to spectroscopic and electrochemical studies. V. Electrochemical Apparatus. Electrochemical studies were performed using a Bioanalytical System BAS 100 B Potentiostat-Galvanostat controlled by BAS 100 W software resident on an IBM-compatible personal computer. A standard three-electrode configuration was used for all the electrochemical experiments. The counter electrode was a large-area platinum flag, and a sodium saturated calomel electrode (SSCE) was used as the reference electrode. The working electrode was either an ITO electrode (modified or unmodified, vide supra) or a conventional electrode as specified. In all cases 0.1 M TBAPF6 in ACN was employed as the supporting electrolyte. VI. XPS Apparatus. XPS studies of electrode surfaces were carried out on a Physical Electronics PHI 5800 spectrometer equipped with a concentric hemispherical analyzer. The average pressure in the high vacuum chamber during the analysis was of the order of 5 × 10-9 Torr. Aluminum monochromatic KR radiation (1486.6 eV) was used as the excitation source, while photoelectrons were collected at a 15° take-off angle. Physical Electronics PC ACCESS and MULTIPAK resident softwares controlled data acquisition and processing. Full survey scans were performed setting a pass energy of 187.85 eV with an eV/ step of 0.800, while an acquisition time of 4 min was generally used. Utility scans were carried out with a pass energy of 117.4 eV and an eV/step of 0.5, acquiring C(1s) and N(1s) signals for 10 min. Finally, high-resolution acquisitions of the N(1s) and C(1s) peaks were performed employing 23.5 eV of pass energy and 0.1 eV/step for 15 min. Voigt profile fitting of the XPS spectra was achieved by means of the XPS PEAK 4.1 program.
Results and Discussion I. XPS Studies. It is known from several studies that carboxylic functions are useful attachment groups to bind a wide variety of molecules and metal complexes to certain metal oxide semiconductor surfaces such as In2O3, SnO2, TiO2, and NiO.10,11 A number of different possibilities and hypotheses about the nature of the acid derivative binding are also reported including physical adsorption, hydrogen bonding, coordination of the carboxylate group to the hard Lewis acids such as In3+, Sn4+, and so forth, or formation of a covalent bond via esterification of the carboxylic acid with the hydroxylic functions present at the surface of the metal oxide.12 It is also probable that, to some extent, multiple modes of interaction occur leading to the formation of a relatively stable monolayer of the carboxylic acid derivatized species on the semiconductor surface. The XPS survey spectra of a bare ITO electrode, after air plasma cleaning, is shown in Figure 1 and is similar to what has been reported previously.13,14 Oxygen, tin, and indium photoelectron emissions are responsible for (10) Rotzinger, F. P.; Kesselman-Truttmann, J. M.; Hug, S. J.; Shklover, V.; Graetzel, M. Structure and Vibrational Spectrum of Formate and Acetate Adsorbed from Aqueous Solution onto the TiO2 Rutile (110) Surface. J. Phys. Chem. B 2004, 108, 5004-5017. (11) Nazeeruddin, M. K.; Humphry-Baker, R.; Liska, P.; Graetzel, M. Investigation of Sensitizer Adsorption and the Influence of Protons on Current and Voltage of a Dye-Sensitized Nanocrystalline TiO2 Solar Cell. J. Phys. Chem. B 2003, 107, 8981-8987. (12) Meyer, T. J.; Meyer, G. J.; Pfennig, B. W.; Schoonover, J. R.; Timpson, C. J.; Wall, J. F.; Kobusch, C.; Chen, X.; Peek, B. M.; et al. Molecular-Level Electron Transfer and Excited State Assemblies on Surfaces of Metal Oxides and Glass. Inorg. Chem. 1994, 33, 39523964.
3024
Langmuir, Vol. 21, No. 7, 2005
Figure 1. XPS spectrum of an ITO electrode after plasma cleaning, survey scan.
Figure 2. XPS spectrum of ITO after functionalization with Fe(DCB)32+, survey scan. Inset: expansion of the N(1s) and C(1s) regions.
the intense peaks in the binding energy region from 1010 to 450 eV. The two peaks at the lowest binding energies, under 50 eV, correspond to electrons expelled from the 4d orbitals of tin and indium, respectively. Only a minor residual contamination of carbon is evident as a small, poorly resolved peak around 285 eV. After chemisorption of II the survey spectrum reported in Figure 2 shows clearly the appearance of the N(1s) peak at around 400 eV and a net increase of the C(1s) peak. The intensity of the new peaks is, however, always much smaller than that of the In or Sn signals due to the small cross sections for photoelectron emissions of these light atoms (15-20 times smaller than the indium or tin). It was almost impossible to directly observe Fe2+ photoelectron peaks due to the close proximity and overlap of the much stronger indium and tin signals. The integrated peak areas, normalized for the cross sections of carbon and nitrogen15 obtained from the utility (13) Armstrong, N. R.; Carter, C.; Donley, C.; Simmonds, A.; Lee, P.; Brumbach, M.; Kippelen, B.; Domercq, B.; Yoo, S. Interface modification of ITO thin films: organic photovoltaic cells. Thin Solid Films 2003, 445, 342-352. (14) Donley, C.; Dunphy, D.; Paine, D.; Carter, C.; Nebesny, K.; Lee, P.; Alloway, D.; Armstrong, N. R. Characterization of Indium-Tin Oxide Interfaces Using X-ray Photoelectron Spectroscopy and Redox Processes of a Chemisorbed Probe Molecule: Effect of Surface Pretreatment Conditions. Langmuir 2002, 18, 450-457. (15) Chastain, J.; King, R. G., Jr. Handbook of X-ray Photoelectron Spectroscopy; Physical electronics, Inc.: Eden Prairie, MN, 1995.
Elliott et al.
Figure 3. Fitting of N(1s) and C(1s) photoelectron peaks, highresolution acquisition mode.
scan acquisition give a C/N ratio of 7.9, close to the expected ratio of 6, for the DCB ligands of II. An excess of carbon is not surprising given the potential for environmental contamination by carbonaceous material when the surface is exposed to the atmosphere and during the adsorption procedure. Some contamination was evident immediately after plasma cleaning. The C(1s) peak was, in the high resolution mode, fitted best with three Gaussian-Lorentian functions centered respectively at 285.5, 286.6, and 289.5 eV (Figure 3). The peak at the highest binding energy has been assigned to the carboxylic acid carbon. The peak at 286.6 eV probably originates from carbons in the ortho and para positions of the pyridine ring where the electronwithdrawing effect of the carboxylic group is strongest. Finally, the 285.5 eV peak is most likely due to the meta carbons of the aromatic ring. The relative areas of the peaks, roughly in the ratio 2:3:1, support this interpretation, when consideration is made for the presence of carbon contaminants. The nitrogen peak was fitted with a single Voigt profile centered at 400.7 eV (Figure 3). The substantially higher binding energy with respect to that commonly reported for ammonia-like nitrogen (398 eV) is most probably the result of Fe(II) coordination.15,16 II. Electrochemistry. Electrochemical oxidation of [Fe(DCB)3](ClO4)2 in solution at an ITO working electrode results in a reversible voltammogram as shown in Figure 4. The half-wave potential is 1350 mV versus SSCE, and the peak separation is 85 mV. The oxidation process involves the removal of one electron from a predominantly iron-centered d orbital. The oxidation potential is substantially higher than iron complexes such as [Fe(bpy)3]2+, which has a half wave potential close to 1 V versus SSCE. This more positive potential is consistent with the strong electron-withdrawing effect of the carboxylic acid functions which greatly increases the back-bonding between the fully occupied dπ orbitals of the metal and the π* orbitals of the ligands, ultimately decreasing the charge density on the metal center. The cyclic voltammetry of Fe(DCB)32+ adsorbed on ITO (Figure 5) exhibits a negative shift of 200 mV in the E1/2 relative to the complex in solution (from 1.35 V to about 1.15 V). This shift is ostensibly caused by the direct interaction of the carboxylate functions with the surface. The peak separation of 50 mV is consistent (16) Briggs, D.; Seah, M. P. Practical Surface Analysis, 2nd ed.; John Wiley and Sons, Inc.: New York, 1983; Vol. 1.
Modified Indium Tin Oxide Electrodes
Figure 4. Cyclic voltammetry of [Fe(II)(DCB)3](ClO4)2 recorded using an ITO electrode in 0.1 M ACN/TBAPF6, scan speed 100 mV/s.
Figure 5. Fe(DCB)32+ adsorbed on an ITO electrode recorded in 0.1 M ACN/TBAPF6, scan speed 200-1000 mV/s.
with an adsorbed species.17,18 The peak width at half height of 260 mV is substantially larger than ideal (90 mV) and is most probably due to inhomogeneities of the surface giving rise to a dispersion in redox potentials and to repulsive lateral interactions between the adsorbed species.19 Voltammogramms showed no significant change over hundreds of potential cycles and were reproducible from day to day over a period of at least 1 week. The absence of any relevant peak shift over a wide range of scan speeds, from 0.1 to 2 V/s, is a further indication of the fast electron transfer and minimal IR drop. This observation provides further evidence that the non-zero peak separation results from lateral interaction of the redox species and dispersion in the E1/2. Finally, as expected for reversible absorbed species, the faradic peak current exhibits an excellent linear dependence on the scan rate (Figure 6). Integration of the anodic wave, after correction for the capacitive current, provides the charge exchanged in the faradic process (3.1 × 10-6 C/cm2) and the surface concentration. The surface concentration calculated from voltammograms recorded at different scan speeds all yield, within experimental error, the same value giving a mean and standard deviation of 6.3 ( 0.2 × 10-11 mol/cm2. (17) Zaniron, E. J. Electroanal. Chem. 1974, 52, 395. (18) Zaniron, E. J. Electroanal. Chem 1974, 52, 355. (19) Brown, A. P.; Anson, F. C. Anal. Chem. 1977, 49, 1589.
Langmuir, Vol. 21, No. 7, 2005 3025
Figure 6. Fe(DCB)32+ adsorbed on an ITO electrode, anodic peak current versus scan speed.
Figure 7. Fe(DCB)32+ adsorbed on ITO after DMB treatment recorded in 0.1 M ACN/TBAPF6, scan speed 200-800 mV/s.
Approximating the footprint of an adsorbed Fe(DCB)32+ as a square of 150 pm on a side, the surface concentration of one monolayer would be 7.4 × 10-11 mol/cm2, which is in close agreement to the experimental value. Scanning electron microscopy images of the ITO glass used for fabricating the electrodes shows no discernible features at the highest magnification possible (×100 000); thus, if for the sake of comparison a roughness factor of 1.5 V versus SSCE. On these surfaces, the electron transfer, especially the oxidation, is extremely slowsso slow that in cyclic voltammogram (CV) mode it is impossible to distinguish the faradic process from the capacitive current at any reasonable scan rate (i.e., >10 mV/s). A comparison of CVs of a 1 × 10-3 M Co(DTB)32+ solution obtained on a 0.2-cm2 ITO electrode and a 0.07-cm2 glassy carbon (GC) electrode is shown in Figure 8. The reasons for the dramatic dependence of the Co(DTB)32+/3+ heterogeneous electron transfer rate on the chemical nature of the electrode surface are still unclear and are the subject of current investigation. Probably there are specific interactions between the surface and the electroactive species which modify the electronic coupling and/or the activation barrier for the electron-transfer reaction. The surface-dependent electrochemistry of Co(DTB)32+ and related cobalt complexes of substituted bipyridines and terpyridines is not unique to ITO. The electron transfer behavior of such complexes varies widely over the entire array of common electrode materials (e.g., Pt, Au, and graphite, as well as GC and ITO).8 The significant observation here, however, is that the oxidation of Co(DTB)32+ is efficiently catalyzed by the Fe(DCB)32+modified ITO surface via an EsurfaceC′ mechanism.20,21Figure (20) Saveant, J. M.; Vianello, E. Electrochim. Acta 1965, 10, 905. (21) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Fundamentals and Applications, 2nd ed.; John Wiley and Sons, Inc.: New York, 2001; pp 501 ff.
Elliott et al.
Figure 9. Catalytic oxidation of Co(II) on Fe(II)-modified ITO electrodes recorded in 0.1 M ACN/TBAPF6, scan speed 100 mV/s.
Figure 10. Variations of the catalytic current with the scan speed: [Co(DTB)32+] ) 3.27 × 10-3 M, recorded in 0.1 M ACN/ TBAPF6, scan speed 20-1000 mV/s.
9 shows the CVs obtained from an Fe(DCB)32+-modified ITO electrode in solutions containing various concentrations of Co(DTB)32+ ranging from 0.0 to 3.27 × 10-3 M. The peak current observed for the highest concentration of Co(DTB)32+ is on the order of ×100 larger than that for the absorbed Fe(DCB)32+ alone at this scan rate. The peak potential for the catalytic oxidation process is ca. 1 V more positive than the E1/2 for Co(DTB)32+ measured on GC and matches the potential of the oxidation of adsorbed Fe(DCB)32+. Figure 10 shows the scan rate dependence of oxidation of Co(DTB)32+ (3.27 mM) on the same Fe(DCB)32+-modified ITO surface. For the data shown in Figures 9 and 10, in neither case is the absolute value of the peak current nor its dependence on respectively concentration and scan rate consistent with a diffusion controlled process for Co(DTB)32+ oxidation. This fact is most evident in Figure 11 where the peak current from Figure 10 is plotted versus v1/2.22 Included also in this figure is a plot of the theoretical diffusion-controlled current calculated using the diffusion coefficient for Co(DTB)32+ obtained from chronoamperometry at a platinum electrode in ACN (9.86 × 10-6 cm2 s-1).23 At every scan rate considered the experimental value (22) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Fundamentals and Applications, 2nd ed.; John Wiley and Sons, Inc.: New York, 2001; p 231.
Modified Indium Tin Oxide Electrodes
Figure 11. Catalytic peak current versus v1/2. [Co(DTB)32+] ) 3.27 × 10-3 M. Scheme 1 . EsurfaceC′ Mechanism
of ip is considerably less than that theoretically predicted, and the difference increases markedly with scan rate. At the fastest scan rate considered (1.0 V s-1) this difference is about ×5. The rate-limiting step must, therefore, involve one of the two processes shown in Scheme 1. Even though there is no indication of slow electrontransfer kinetics for absorbed Fe(DCB)32+/3+ in the data of Figure 5, when Co(DTB)32+ is present in solution the absolute current is much larger than for the adsorbed Fe(DCB)32+ alone. Therefore, the possibility that the Fe(DCB)32+/3+ process could become rate limiting with Co(DTB)32+ present should not be dismissed out of hand. However, from the data in Figures 9 and 10 there is only a moderate shift in Ep over the entire collection of voltammograms presented in the two figures (ca. 150 mV). Also, the shift in Ep for both sets of data is linear with the increase in ip. Finally, the magnitude of the potential shift (23) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Fundamentals and Applications, 2nd ed.; John Wiley and Sons, Inc.: New York, 2001, pp 161-164.
Langmuir, Vol. 21, No. 7, 2005 3027
is commensurate with what is expected on the basis of the uncompensated resistance (ca. 100 Ω for the ITO electrode itself). If the rate-limiting step was Fe(DCB)32+ oxidation, a much stronger (and nonlinear) dependence of Ep on ip would be observed. Thus, it is reasonable to conclude from these data that the slow electron-transfer process is between the adsorbed Fe(DCB)33+ and Co(DTB)32+ in solution. Finally, it was not possible in this study to access either the pure diffusion-controlled or pure kinetic-controlled regimes. In the latter case, the solubility of [Co(DTB)32+](ClO4-)2 limits the maximum concentration attainable, but more significantly, the maximum potential scan rate is limited by the IR drop. The inherent resistance of ITO is relatively large and so is the electrode area. Unfortunately, the method used to pattern the ITO working electrode requires that the electrode be relatively large to ensure a well-defined geometric area. With alternate approaches to working electrode fabrication it should be possible to have smaller areas and, thus, access faster scan rates that would allow for a quantitative evaluation of the kinetics of catalysis. Conclusion In this study we have described the preparation and the characterization of ITO electrodes modified by attachment of iron(II) polypyridyl complexes. These complexes reside on the surface as stable, electrochemically active monolayers. We have also demonstrated that these modified electrodes can efficiently catalyze the oxidation of certain cobalt complexes that are otherwise essentially electrochemically inert on the unmodified ITO surfaces. The catalytic process approaches diffusional control at very slow scan speeds, but that condition was not actually achievedsnor was it possible to pass into the regime of pure kinetic control in these studies. The catalysis is sufficiently efficient that the peak oxidation current for Co(DTB)32+ exceeds the ip for the surface oxidation of Fe(DCB)32+ by >×100 and the current for the uncatalyzed oxidation of Co(DTB)32+ by considerably more than that. Acknowledgment. Support for this work (C.M.E.) from the National Science Foundation (CHE-0139637) and Department of Energy (DE-FG0204ER15591) is gratefully acknowledged. C.A.B. and S.C. acknowledge INSTM consortium and FIRB Contract No. RBNE019H9K for support. LA047364F
