Electron Transfer through Clay Monolayer Films Fabricated by the

Japan Science and Technology Agency. ∇ The University of New Brunswick. † Current address: Department of Basic Science, Graduate School of Arts...
0 downloads 0 Views 592KB Size
Download PDF
Langmuir 2006, 22, 9591-9597

9591

Electron Transfer through Clay Monolayer Films Fabricated by the Langmuir-Blodgett Technique Jun Yoshida,§,† Kazuko Saruwatari,§ Jun Kameda,§ Hisako Sato,§,£ Akihiko Yamagishi,*,§,£,‡ Laisheng Sun,∇,# Maria Corriea,∇ and Gilles Villemure∇ Department of Earth and Planetary Science, Graduate School of Science, The UniVersity of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, CREST, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama, 332-0012, Japan, and Department of Chemistry, The UniVersity of New Brunswick, P.O. Box 45222, Fredericton, New Brunswick, Canada ReceiVed June 11, 2006. In Final Form: August 16, 2006 Hybrid films composed of amphiphilic molecules and clay particles were constructed by the modified LangmuirBlodgett (LB) method. Clays used were sodium montmorillonite (denoted as mont) and synthetic smectite containing Co(II) ions in the octahedral sites (denoted as Co). Two kinds of amphiphilic molecules were useds[Ru(dC18bpy)(phen)2](ClO4)2 (dC18bpy ) 4,4′-dioctadecyl-2,2′-bipyridyl and phen ) 1,10-phenanthroline) (denoted as Ru) and octadecylammonium choloride (ODAH+Cl- or denoted as ODAH). Three kinds of hybrid films (denoted as Rumont, Ru-Co, and ODAH-Co films) were prepared by spreading an amphiphilic molecule onto an aqueous suspension of a clay. Atomic force microscopy (AFM) analyses of the films deposited on silicon wafers indicated that closely packed films were obtained at 20 ppm for all the above three cases. Cyclic voltammetry (CV) was measured on an ITO electrode modified with a hybrid film or a monolayer film of pure Ru(II) complex salt (denoted as Ru film). The Ru(II) complexes incorporated in the Ru-mont film lost their redox activity, indicating that montmorillonite layers acted as a barrier against electron transfer. In contrast, the same complexes in the Ru-Co film were electrochemically active with the simultaneous appearance of the redox peaks due to the Co(II)/Co(III) (or Co(II)/ Co(IV)) couple. The results implied that electron transfer through cobalt clay layers was possible via mediation by Co(II) ions in a clay sheet. For an aqueous solution containing nitrite ions (NO2-) at pH 3.0, a large catalytic oxidation current was observed for both the electrodes modified with the Ru-mont and Ru-Co films. The results were interpreted in terms of the mechanisms that the charge separation of an incorporated Ru(II) complex took place to produce a pair of a Ru(III) complex and an electron and that the generated Ru(III) complex was reduced by a nitrite ion before it recombined with the electron.

Introduction Smectite-type clay minerals are characterized by a large cation exchange capacity (CEC) and the ability to stabilize a variety of cationic or polar molecules in their interlayer spaces. Those properties of clay minerals have opened application as catalysts, electrodes modifiers, and supports for photochemical reactions.1-5 Clay modified electrodes (denoted as CMEs) have been studied since they were first introduced in 1983.6 CMEs are usually constructed by a number of methods such as casting, spin-coating, or electrophoresis onto conducting substrates such as ITO or glassy carbon.2,3,6,7 Use of CMEs for biomolecular sensing, in particular, has attracted a lot of interest.8-11 * To whom correspondence should be addressed. E-mail: [email protected]. Phone: +81-3-5978-5575. Fax: +81-3-59785575. § University of Tokyo. £ Japan Science and Technology Agency. ∇ The University of New Brunswick. † Current address: Department of Basic Science, Graduate School of Arts and Sciences, The University of Tokyo, 3-8-1 Komaba, Meguro-ku, Tokyo 153-8902, Japan. ‡ Current address: Department of Chemistry, Faculty of Science, Ochanomizu University, 2-1-1, Otsuka, Bunkyo-ku, Tokyo 112-8610. # Current address: Department of Chemistry and Chemical Biology, Northeastern University, Boston, MA 0211-5000. (1) Pinnavaia, T. J.; Raythatha, R.; Lee, J. G. S.; Halloran, L. J.; Hoffman, J. F. J. Am. Chem. Soc. 1979, 101, 6891. (2) Macha, S. M.; Fitch, A. Mikrochim. Acta 1998, 128, 1. (3) Navra´tilova´, Z.; Kula, P. Electroanalysis 2003, 15, 837. (4) Ogawa, M.; Kuroda, K. Chem. ReV. 1995, 95, 399. (5) Shichi, T.; Takagi, K. J. Photochem. Photobiol., C 2000, 1, 113. (6) Ghosh, P. K.; Bard, A. J. J. Am. Chem. Soc. 1983, 105, 5691. (7) Song, C.; Villemure, G. J. Electroanal. Chem. 1999, 462, 143.

In parallel with their practical applications, extensive work has been done to clarify the basic mechanism of charge transport in CMEs.7,12-19 For example, it has been shown that the electrochemically active fraction of redox molecules incorporated in clays is very low, or less than a few percent of the cationexchanged ions.13,15 Several hypotheses have been proposed to explain this. One is that molecules in contact with a conducting substrate are electrochemically active. Another is that molecules adsorbed in excess of the clay CEC are electroactive. A major difficulty in understanding the mechanism of charge transport in CMEs is that there are many electron-transfer routes to consider. With the purpose of establishing the fundamental aspects of charge transport in CMEs, we have developed a method of constructing the hybrid films composed of amphiphilic molecules and a single clay layer by the modified Langmuir-Blodgett (LB) technique.20,21 This method involves the spreading of cationic (8) Zen, J.-M.; Lo, C.-W.; Chen, P.-J. Anal. Chem. 1997, 69, 1669. (9) Zhou, Y.; Hu, N.; Zeng, Y.; Rusling, J. F. Langmuir 2002, 18, 211. (10) Mousty, C. Appl. Clay Sci. 2004, 27, 159. (11) Zen, J.-M.; Senthil Kumar, A. Anal. Chem. 2004, 76, 205A. (12) White, J. R.; Bard, A. J. J. Electroanal. Chem. 1986, 197, 233. (13) Rudzinski, W. E.; Bard, A. J. J. Electroanal. Chem. 1986, 199, 323. (14) King, R. D.; Nocera, D. G.; Pinnavaia, T. J. J. Electroanal. Chem. 1987, 236, 43. (15) Villemure, G.; Bard, A. J. J. Electroanal. Chem. 1990, 282, 107. (16) Villemure, G.; Bard, A. J. J. Electroanal. Chem. 1990, 283, 403. (17) Kaviratna, P. D.; Pinnavaia, T. J. J. Electroanal. Chem. 1995, 385, 163. (18) Yao, K.; Shimazu, K.; Nakata, M.; Yamagishi, A. J. Electroanal. Chem. 1998, 442, 235. (19) Sun, L.; Villemure, G. J. Electroanal. Chem. 2005, 583, 84. (20) Inukai, K.; Hotta, Y.; Taniguchi, M. Tomura, S.; Yamagishi, A. J. Chem. Soc., Chem. Commun. 1994, 959. (21) Tamura, K.; Setsuda, H.; Taniguchi, M.; Yamagishi, A. Langmuir 1999, 15, 6915.

10.1021/la061668f CCC: $33.50 © 2006 American Chemical Society Published on Web 09/23/2006

9592 Langmuir, Vol. 22, No. 23, 2006

Yoshida et al.

of clay minerals; a sodium montmorillonite (denoted as mont) and a synthetic Co(II)-smectite in which Co(II) ions are located in the octahedral sites (denoted as Co). By changing the concentration of a clay in the range of 0-50 ppm, the surface structure of a deposited film was studied by atomic force microscopy (AFM). Cyclic voltammetry (CV) measurements on electrodes modified with these hybrid films were performed in contact with either a 0.1 M NaClO4 solution or a sodium acetate buffer solution at pH 3.0. The effect of the presence of nitrite ions was also studied at pH 3.0. It was expected that nitrite ions would change the electrochemical responses of hybrid films due to their electron donating ability. Experimental

Figure 1. Schematic diagram of possible electron-transfer routes in an electrode modified with a clay monolayer.

surfactants at the air-water interface over a dilute clay suspension. The negatively charged clay particles are adsorbed electrostatically onto the cationic parts of floating amphiphilic molecules.22-24 The hybrid films produced are transferred onto a solid surface by either the vertical or horizontal dipping methods. This modified LB method is also applicable to other inorganic layered materials, leading to the bottom-up construction of inorganic-organic hybrid films.25,26 Employing this method, functional inorganic-organic hybrid films have been constructed with novel nonlinear optical and photosensitive magnetic properties.27-29 The hybrid films could also be used for sensing chiral molecules by using chiral Os(II) complexes as templates.30-32 In the present work, electrodes were modified with the hybrid films composed of a clay monolayer and a monolayer of a metal complex. It was attempted to determine the electron-transfer routes in CMEs on a monolayer scale. On such electrodes, only three electron transfer routes are considered to be plausible (Figure 1): directly through the clay monolayer (route 1), through vacant spaces between clay particles (route 2), and through metal complexes in direct contact with a conducting electrode (route 3). If a closely packed clay monolayer is deposited on a conducting substrate, route 1 is considered to be the only possible pathway. Three kinds of hybrid films have been prepared by employing two kinds of amphiphilic molecules; [Ru(dC18bpy)(phen)2](ClO4)2 (dC18bpy ) 4,4′-dioctadecyl-2,2′-bipyridyl and phen ) 1,10-phenanthroline) (abbreviated by Ru) and octadecylammonium chloride (ODAH+Cl- or denoted as ODAH) and two types (22) Umemura, Y.; Yamagishi, A.; Schoonheydt, R.; Persoons, A.; Schryver, F. D. Langmuir 2001, 17, 449. (23) Ras, R. H. A.; Ne´meth, J.; Johnston, C. T.; DiMasi, E.; De´ka´ny, I.; Schoonheydt, R. A. Phys. Chem. Chem. Phys. 2004, 6, 4174. (24) Umemura, Y.; Shinohara, E. Langmuiir 2005, 21, 4520. (25) Taguchi, Y.; Kimura, R.; Azumi, R.; Tachibana, H.; Koshizaki, N.; shimomura, M.; Momozawa, N.; Sakai, H.; Abe, M.; Matsumoto, M. Langmuir 1998, 14, 6550. (26) Saruwatari, K.; Sato, H.; Kameda, J.; Yamagishi, A.; Domen, K. Chem. Commun. 2005, 1999. (27) Kawamata, J.; Ogata, Y.; Taniguchi, M.; Yamagishi, A.; Inoue, K. Mol. Cryst. Liq. Cryst. 2000, 343, 53. (28) Umemura, Y.; Yamagishi, A.; Schoonheydt, R.; Persoons, A.; Schryver, F. D. J. Am. Chem. Soc. 2002, 124, 992. (29) Yamamoto, T.; Umemura, Y.; Sato, O.; Einaga, Y. Chem. Mater. 2004, 16, 1195. (30) Okamoto, K.; Tamura, K.; Takahashi, M.; Yamagishi, A. Colloids Surf. A 2000, 169, 241. (31) He. J. X.; Sato, H.; Yang, P.; Yamagishi, A. Electrochem. Commun. 2003, 5, 388. (32) He. J. X.; Sato, H.; Umemura, Y.; Yamagishi, A. J. Phys. Chem. B 2005, 109, 4679.

Materials. NaNO2, NaClO4, and ODAH+Cl- were used as received. ODAH+Cl- was dissolved in a 1:19 (v/v) methanol/ chloroform mixture at a concentration of 2.1 × 10-4 mol dm-3. All aqueous solutions were prepared using HPLC-grade water (Wako, Japan). The pH of a solution was controlled by adding a 0.1 M CH3COOH/CH3COONa buffer to a 0.1 M NaClO4 aqueous solution. An amphiphilic Ru(II) complex salt with two long alkyl chains, [Ru(dC18bpy)(phen)2](ClO4)2, was prepared according to a previously reported method.21 Clay samples used in the present studies were sodium montmorillonite (Kunipia P, Kunimine Ind., Japan) (denoted as mont) with an elemental composition of (Na0.38Ca0.04K0.01)(Al1.57Mg0.34Fe3+0.10)(Si3.86Al0.14)O10(OH)2 and a cation exchange capacity (CEC) of 115 mequiv per 100 g and a synthetic smectite (Co(II)-smectite) (denoted as Co) with Na0.71(Co5.86Si7.89)O20(OH)4 and a CEC of 67 mequiv per 100 g. The cobalt clay was prepared by hydrothermal treatment of a mixture of cobalt chloride, silicic acid, and NaOH.33 These clays were dispersed in deionized water at the concentration of 1000 ppm as a stock solution. They were diluted to 1-50 ppm as a subphase in a Langmuir trough. The clay suspension was placed at room temperature for 2 days before being used to confirm the complete exfoliation of clay layers. Film Preparation. A solution of either the Ru(II) complex salt or ODAH+Cl- was spread onto a clay suspension as a subphase in a 10.0 cm × 13.0 cm LB trough (USI System, Japan). Temperature was maintained by circulating water at 20 °C. After 40 min, the surface was compressed at a rate of 10 cm2 min-1 until the surface pressure reached 15 mN m-1. The floating film was left for 10 min at 15 mN m-1 to complete the hybridization. The film was thereafter transferred onto ITO or silicon substrates by the vertical dipping method at a dipping rate of 10 mm min-1. AFM and SEM Observations. AFM images were recorded at room temperature in air with a Nanoscope III scanning probe microscope (DI Instruments). Nanoprobe integral cantilevers (Si3N4) were used, having tips with a spring constant of 33 N m-1 (Park Scientific). AFM images were obtained in the tapping mode with the filters off. SEM studies on clay LB films were made using a Hitachi S-4500 SEM equipped with a cold field-emission gun and a Thermo Noran Phase ID system.34 Electrochemical Measurements. ITO glass electrodes were cut into 1.0 cm × 1.5 cm pieces. Cyclic voltammograms were recorded with a potentiostat (Toho Technical, Japan) in a three-electrode cell using a Ag|AgCl|KCl (sat) electrode as reference and a platinum mesh as an auxiliary electrode.

Results Fabrication of the Amphiphilic Molecule-Clay Hybrid Films. Four types of films were prepared. Three were hybrid films consisting of amphiphilic molecules and clay particles. The fourth was a LB film containing pure [Ru(dC18bpy)(phen)2](ClO4)2 (denoted as Ru film) as a reference. Three types of hybrid films were as follows. [Ru(dC18bpy)(phen)2]2+ with (33) Sun, L. Ph.D. Thesis, University of New Brunswick, 2005. (34) Ono, A. Bull. Ceram. Soc. Jpn. 2004, 39, 911.

Electron Transfer through Clay Monolayer Films

montmorillonite (Ru-mont film), [Ru(dC18bpy)(phen)2]2+ with Co(II)-smectite (Ru-Co film) and ODAH+ with Co(II)smectite (ODAH-Co film). In fabricating these hybrid films, the concentration of a clay in a subphase was varied from 0 to 50 ppm. The π-A isotherms were obtained when [Ru(dC18bpy)(phen)2](ClO4)2 was spread onto an aqueous suspension of sodium montmorillonite at the concentration of 0, 1, 5, 10, 20, and 50 ppm, respectively. Although a lift-off area was not determined precisely from these isotherms, the surface pressure rose from zero around 1.2-1.4 nm2 molecule-1. As the clay concentration increased, the lift-off areas increased, suggesting the occurrence of hybridization of the Ru(II) complexes with clay particles at an air-water interface. These isotherms were nearly identical with the previous results on [Os(dC18bpy)(phen)2]2+-clay hybrid films.30 The π-A isotherms were obtained in a similar way when ODAH+Cl- was spread onto an aqueous suspension of the Co(II)-smectite at the concentration of 0, 5, 10, 20, and 50 ppm, respectively. The lift-off area increased when the clay concentration increased from 0.15 to 0.30 nm2 molecule-1, again suggesting that ODAH+ was effectively hybridized with Co(II)-smectite particles. The π-A isotherms were obtained when [Ru(dC18bpy)(phen)2](ClO4)2 was spread onto aqueous suspensions of the Co(II)smectite at 0, 5, 10, 20, and 50 ppm, respectively. These π-A isotherms were similar to those shown for the Ru-mont films. The lift-off area shifted toward higher values around 1.2 nm2 molecule-1. with the increase of a clay concentration, indicating the hybridization of the Co(II)-smectite particles with [Ru(dC18bpy)(phen)2]2+. Analyses of Deposited Hybrid Films by AFM and SEM. The amphiphilic molecule-clay hybrid films (Ru-mont, ODAH-Co, and Ru-Co films) obtained at various clay concentrations (0-50 ppm) were transferred onto either silicon wafers (denoted as Si) or ITO substrates by the vertical dipping method. The transfer ratio (Rt) was nearly equal to 1.0 for the clay concentration of 0, 5, 10, and 20 ppm for both the Rumont and Ru-Co films, while Rt was about 0.8 for the ODAHCo films. Rt was less than 0.6 in all cases when the clay concentration was 50 ppm. One possible reason for the low Rt at 50 ppm was that clay particles were aggregated at this higher clay concentration to form a heterogeneous film partially composed of multiple clay sheets. The hybrid films deposited onto Si substrates were analyzed by AFM. In the case of the Ru-mont films, the surface density of clay particles increased with the increase of the clay concentration in the subphase (Figure 2). For the concentrations of 5, 10, and 20 ppm, there was little overlap between neighboring clay particles. The most closely packed film was prepared at 20 ppm. Similarly, the ODAH-Co and Ru-Co films became more closely packed as the clay concentration increased, although some aggregation of clay particles was seen at the highest clay concentrations (Figures 3 and 4). For both the ODAH-Co and Ru-Co films, the most closely packed films were prepared at the clay concentration of 20 ppm. The AFM observation could not be performed successfully on hybrid films deposited onto ITO electrodes because the surface of an ITO substrate was too rough.32 Instead, the surface structures of the films deposited on ITO were investigated by SEM. According to the results on the Ru films deposited at the clay concentration of 20 ppm, bright spots were seen uniformly over the whole surface. It indicated that electrons were emitted from an ITO substrate, passing through the Ru film. In contrast, there

Langmuir, Vol. 22, No. 23, 2006 9593

Figure 2. AFM images of Ru-mont films transferred onto Si substrates at the montmorillonite concentration of (a) 1, (b) 5, (C) 10, and (d) 20 ppm, respectively. The hybrid films were deposited at 15 mN m-1.

Figure 3. AFM images of ODAH-Co films transferred onto Si substrates at the Co(II)-smectite concentration of (a) 5, (b) 10, and (c) 20 ppm. The hybrid films were deposited at 15 mN m-1.

were distinctively dark and bright regions seen on the surface of the Ru-mont film. The dark regions were due to the inhibition of electron emission by the clay particles, while the bright regions were covered by the monolayer layer of pure Ru(II) complexes. On the basis of this view, the SEM images were regarded as an indication that clay particles were transferred onto an ITO substrate accompanying the amphiphilic monolayers. Electrochemical Studies in Contact with an Aqueous NaClO4 Solution. CV curves were measured on ITO electrodes modified with various films when the electrodes were in contact with an aqueous 0.1 M NaClO4 solution. Figure 5 shows the CV curves on the electrodes modified with either Ru or Ru-mont films. Reversible peaks corresponding

9594 Langmuir, Vol. 22, No. 23, 2006

Yoshida et al.

Figure 6. Cyclic voltammogram on an ITO electrode modified with an ODAH-Co film fabricated using a clay concentration of 20 ppm. (a) 3rd, (b) 5th, (c) 10th, (d) 15th, and (e) 20th scans. The scan speed was 50 mV s-1.

Figure 4. AFM images of Ru-Co films transferred onto Si substrates at the Co(II)-smectite concentration of (a) 5, (b) 10, and (c) 20 ppm, respectively. The hybrid films were deposited at 15 mN m-1.

Figure 7. Cyclic voltammograms on an ITO electrode modified with a Ru-Co film fabricated using the clay concentration of 20 ppm: The curves corresponded to (a) 3rd, (b) 5th, (c) 10th, (d) 15th (solid curve), and (e) 20th dotted curve) scans, respectively. Noted that curves (d) and (e) were nearly identical. The scan speed was 50 mV s-1.

Figure 5. Cyclic voltammograms on ITO electrodes modified with (a) a Ru film and with Ru-mont films fabricated at the concentration of (b) 5, (c) 10, and (d) 20 ppm, respectively. Curve (e) indicates the CV on a bare ITO as a reference (3rd scan, 50 mV s-1).

to the metal-based Ru2+/Ru3+ redox couple were observed at 1190 (oxidation) and 1110 mV (reduction).35,36 When the CV curves were compared among the Ru-mont films deposited at 5, 10, and 20 ppm, the peak currents were found to decrease with an increase of the clay concentration. In other words, as clay particles covered an electrode surface more efficiently, the electrode showed lower electrochemical activities. It seemed that the films lost such activities when the surface was fully covered with clay films. Figure 6 shows the CV curves on the ITO electrode modified with the ODAH-Co film deposited at 20 ppm. No redox reaction was seen in the first several scans. However, a redox peak gradually appeared around 970 mV on repeating the scans. It was attributed to the oxidation of Co(II) in the clay layers. The same behavior was observed previously on ITO electrodes covered with the thicker films of the same Co(II)-smectite by a spincoating method.37,38 (35) Staniewicz, R. J.; Sympson, R. F.; Hendricker, D. G. Inorg. Chem. 1977, 16, 2166. (36) Kim, B. H.; Lee, D. N.; Park, H. J.; Min, J. H.; Jun, Y. M.; Park, S. J.; Lee, W.-Y. Talanta, 2004, 62, 595. (37) Xiang, Y.; Villemure, G. Clays Clay Miner. 1996, 44, 515.

Figure 7 shows the CV curves on the ITO electrode modified with the Ru-Co film deposited at 20 ppm. The electrode showed a large redox activity from the very first scan in contrast with that of the electrode modified with the ODAH-Co film. On repeating the scans, the oxidation peak shifted to the lower potential until a stable peak appeared around 1100 mV was obtained after the 15th scan. In correspondence to this change, the reduction peak, which was observed initially around 1100 mV, showed a small shoulder at lower potential around 970 mV. When these were compared with the results for an electrode modified with the ODAH-Co film, it was concluded that both of the Co(II) and Ru(II) ions in the Ru-Co films were oxidized. A small shoulder in the reduction peak near 970 mV was attributed to the electrochemical reduction of the cobalt sites. The results suggested that the Co(III) (or Co(IV)) in the clay was catalytically reduced by Ru(II). Effects of Nitrite Ions. The effect of sodium nitrite in an electrolyte on the CV curves was investigated. The nitrite ion was chosen because of its electron-donating ability. This anion is known to be oxidized at lower pH to nitrate by way of a number of intermediates.39,40 The CV curves were obtained at the nitrite concentration from 1.5 × 10-5 to 5.8 × 10-4 M. The ODAH-Co film showed no electrochemical oxidation activity toward nitrite even at a high concentration of nitrite. On the other (38) Xiang, Y.; Villemure, G. J. Phys. Chem. 1996, 100, 7143. (39) Chen, S.-M. J. Electroanal. Chem. 1998, 457, 23. (40) Pan, K.-C.; Chuang, C.-S.; Cheng, S.-H.; Su, Y. O. J. Electroanal. Chem. 2001, 501, 160.

Electron Transfer through Clay Monolayer Films

Langmuir, Vol. 22, No. 23, 2006 9595

possibilities exist

[Ru(dC18bpy)(phen)2](ClO4)2 + Na-montmorillonite T [Ru(dC18bpy)(phen)2](ClO4)-montmorillonite + NaClO4 (1) [Ru(dC18bpy)(phen)2](ClO4)2 + 2Na-montmorillonite T [Ru(dC18bpy)(phen)2]-(montmorillonite)2 + 2NaClO4 (2)

Figure 8. Cyclic voltammograms on ITO electrodes modified with (a) a Ru film, (b) a Ru-Co film, and (c) a Ru-mont film in 0.1 M sodium acetate buffer pH 3 containing 1.5 × 10-4 M sodium nitrite. (d) Bare ITO electrode shown for reference (3rd scans, 50 mV s-1).

Figure 9. Plots of peak current vs the concentration of sodium nitrite (from 1.5 × 10-5 to 5.8 × 10-4 M) in the cyclic voltammograms of electrodes modified with the three kinds of hybrid films: (a) Ru film, (b) Ru-mont film, and (c) Ru-Co film. The values were corresponded to the third scans at 50 mV s-1.

hand, three other films (Ru, Ru-mont and Ru-Co) showed large oxidation currents in the presence of nitrite (Figure 8). The CV curves were characteristic of catalytic reactions with the absence of reduction peaks in the backward scans. The oxidation peak currents increased linearly when they were plotted against the concentration of sodium nitrite for all three films (Figure 9). There was little difference in the slope of the plots among the three electrodes (Ru, Ru-Co and Ru-mont).

Discussion π-A Isotherms. The π-A isotherm for [Ru(dC18bpy)(phen)2] (ClO4)2 exhibited a clear change in the presence of clay particles in a subphase. The slope of the isotherm was steeper when the subphase contained montmorillonite, while the lift-off area was not significantly affected. The cross section of the headgroup of [Ru(dC18bpy)(phen)2]2+ (denoted as Sc) is estimated to be about 1.4 nm2 per molecule. On the basis of this, the π-A isotherm for [Ru(dC18bpy)(phen)2] (ClO4)2 on pure water is expected to show the lift-off area at about 1.4 nm2. This coincided with the experiments that the surface pressure increased from zero to around 1.2-1.4 nm2. The results indicated that [Ru(dC18bpy)(phen)2]2+ formed a monolayer at an air-water interface with their alkyl chains upright from the surface. During the hybridization process with clay minerals, the Ru(II) complexes were thought to be stabilized on clay particles by cation exchange. For the hybridization of amphiphilic Ru(II) complex salts with montmorillonite particles, the following two

The area per charge of montmorillonite clay is calculated to be about 1.02 nm2 (denoted by Sm) on the basis of its elemental composition.41 Therefore, one Ru(II) complex would occupy 1.02 nm2 according to reaction 1. Since this Sm is less than the Sc of [Ru(dC18bpy)(phen)2]2+, the Ru(II) complexes would form at least partially a bimolecular layer on a clay surface at the full ion exchange. In contrast, according to reaction 2, one Ru(II) complex would occupy 2Sm or 2.04 nm2. This is more than the Sc of a clay, and thus, the Ru(II) complex could form a monolayer on a clay surface even after the occurrence of the full ion exchange. The observed lift-off area in the π-A isotherms (1.2-1.4 nm2) was intermediate between Sm and 2Sm. Thus, both reactions 1 and 2 likely took place simultaneously when the Ru(II) complex was hybridized with montmorillonite particles at an air-water interface. The π-A isotherms of ODAH+Cl- on pure water and on Co(II)-smectite suspensions also showed a clear effect by the presence of clay particles in a subphase. The lift-off area of ODAH+Cl- on pure water was a little smaller than the previous reports for ODAH+Cl- on NaClO4 solutions.22 This might be because some of the ODAH+Cl- dissolved into a water subphase. The lift-off area increased drastically as the clay concentration increased. A detailed structural analysis of the Co(II)-smectite has not yet been done. However, its area per charge (Sm) can be estimated to be about 1.35 nm2 on the basis of its chemical composition. It is assumed here that the unit cell of Co(II)-smectite has the dimension similar to that of known trioctahedral smectite-type clays (0.51-0.54 nm × 0.90-0.93 nm).23 This is much larger than the Sm of montmorillonite clay and very close to the Sc of [Ru(dC18bpy)(phen)2](ClO4)2. This explains why the lift-off areas were much less affected by the presence of a clay in a subphase of Co(II)-smectite than in the case of a montmorillonite suspension. The higher Sm or the lower charge density means that, in the ODAH-Co films, a clay surface was not totally covered by ODAH+ ions and/or also that the alkyl chains of ODAH+ were inclined toward the clay surface. Surface Structures of Deposited Hybrid Films. The AFM observation on hybrid films transferred onto Si substrates showed that the clay concentration in the subphase influenced greatly the packing of clay particles. Clay particles were more closely packed with the increase of the concentration of a clay (Figures 2-4). The most closely packed films were prepared at the clay concentration of 20 ppm. Below this concentration, clay particles were less densely packed with vacant spaces between them. The transfer ratios were always close to unity even at the low clay concentrations, indicating that these vacant parts were likely covered with the monolayer of pure amphiphilic molecules. In other words, the films fabricated at the low clay concentrations were composed of two different regions: one was pure metal complexes and the other clay/metal complex hybrids.23 In accord with the above results, the SEM micrographs showed that clay particles were present on an ITO electrode coated with a Rumont film, but that they did not fully cover the substrate. (41) Okada, T.; Morita, T.; Ogawa, M. Appl. Clay Sci. 2005, 29, 45.

9596 Langmuir, Vol. 22, No. 23, 2006

Redox Activity of Ru and Ru-mont Films in NaClO4 Solutions. A reversible redox peak due to the Ru(II)/Ru(III) couple was seen in an ITO electrode modified with the pure Ru(II) complex (Ru film) (Figure 5a). If it is assumed that the whole surface of an ITO substrate (1.0 cm × 1.5 cm) was covered by [Ru(dC18bpy)(phen)2]2+ ions whose surface density was one molecule per 1.4 nm2, the oxidation of the Ru(II) centers in the monolayer would produce about 17.2 µC of charge. This is close to the value of the charge (or 21 µC) as obtained by integrating the oxidation peak in Figure 5a. In the Ru(II) complex-clay hybrid films (Ru-mont films), the Ru(II) complexes were hybridized with clay particles. As discussed above, there are two possible cation-exchange mechanisms (eqs 1 and 2), and the π-A isotherms suggested that both reactions occurred in parallel during the hybridization process. Depending on the relative contributions of reactions 1 and 2, the amount of electrons that would be produced by oxidation of all the Ru(II) complexes in the Ru-mont hybrid films varies from 17.2 (1.4 nm2 per molecule for reaction 1) to 11.8 µC (2.04 nm2 per molecule for reaction 2). The experimental values determined from the area of the redox peaks in cyclic voltammograms were estimated to be 12.5, 4.3, and 1.8 µC for the films fabricated at the clay concentrations of 5, 10, and 20 ppm, respectively. In other words, [Ru(dC18bpy)(phen)2]2+ lost much of its electrochemical activity upon hybridization with montmorillonite. One explanation for this is that even a single clay layer prohibited electron transfer between the ruthenium headgroups and the ITO substrates. In other words, electron transfer by route 1 (Figure 1) was less efficient than by routes 2 or 3. This is consistent with the decrease in the peak currents as clay particles became more closely packed with the increase of the clay concentration in a subphase. Redox Activity of ODAH-Co Films in NaClO4 Solutions. The voltammetric redox peaks on ITO electrodes modified with ODAH-Co films as seen in Figure 6 are attributed to the redox activity of Co(II) sites in a Co(II)-smectite. When all the Co(II) sites in the film were oxidized electrochemically, the produced charge was estimated to be about 290 µC, based on the chemical composition of the clay. The estimation was based on the assumption that the unit cell of the Co(II) smectite was close to those of known trioctahedral smectites (ca. 0.48 nm2)23 and that clay particles covered the whole electrode surface uniformly. Integration of the oxidation peak in the 15th scan in Figure 6 gave a charge of 15.6 µC. This value was only 5% of the charge when the whole Co(II) sites were assumed to be active electrochemically. This fraction (or 5%) was, however, larger than what had been reported for the electrodes modified with the spin-coated films of the same Co(II)-smectites. The fraction of electroactive Co(II) ions on the spin-coated electrode was found to be les than 0.3%.37 Two reasons were invoked to account for the observed low electroactive fraction in the case of spin-coated films. The first was that activity was restricted to Co(II) in defects in a clay where Co(II) ions were exposed. The second was that activity was restricted to Co(II) sites in the first few clay layers in direct contact with the ITO substrates. The observation of a significantly larger electroactive fraction on the present electrodes argues for the second possibility. In other words, all or nearly all the clay layers were in direct contact with the substrate in case of the films as deposited by the LB method,. Figure 6 shows that the voltammetric peaks in the ODAHCo films increased with the number of scans, while the peak positions shifted slightly toward higher potentials. This likely reflects changes in the clay structure during the redox reactions. Oxidation of octahedral metal ions in clay layers requires

Yoshida et al.

adjustments in the structure to maintain electric neutrality to accommodate the decrease in the clay negative surface charge density. Analogies can be made with the extensive literature on the reduction and oxidation of iron in natural clays that is accompanied by extensive changes in the structure and properties of the clays that have still not been completely elucidated. The literature on the redox activity of iron in natural smectite clays has been recently reviewed.42 Redox Activity of Ru-Co Films in NaClO4 Solutions. Three differences were noted when the CV of an ITO electrode modified with a Ru-Co film deposited at 20 ppm (Figure 7) was compared with those of the corresponding Ru-mont (Figure 5) and ODAH-Co films (Figure 6). First, in the Ru-Co film, large voltammetric peaks were seen from the very first scan. The growth of the peaks with the number of scans was much less dramatic than for the ODAH-Co film, and a stable voltammogram was obtained after 15th scans. Second, the peak potential was higher than for the ODAH-Co film, much closer to that of the peaks seen in the Ru or Ru-mont films. It was indicated that they were mostly due to the redox activity of Ru(II) in the headgroups of the amphiphillic molecules. Redox activity of the Co(II) sites was responsible only for small shoulders on the main peak at lower potentials. More importantly, the peak near 1100 mV was much larger than in the corresponding Ru-mont film. Integration of the oxidation peak in the 15th scan of Figure 6 gave 19 µC as compared to only 1.8 µC for the Ru-mont film (Figure 5d). This suggested that electron transfer though the clay (route 1 in Figure 1) was more efficient for the redox-active Co(II) smectite than for the natural montmorillonite. In other words, the Co(II) sites in the Co(II) smectites acted as electron relays between the Ru(II) in the headgroups of the amphiphilic molecules and the ITO substrates. This is consistent with a previous report of large enhancement of the voltammetric response of [Ru(bpy)3]2+ cations adsorbed in the spin-coated films of synthetic Co(II) smectites compared to that obtained with natural montmorilonites that was attributed to electron transfer between Co(II) sites in the synthetic clays and electrochemically oxidized ruthenium cations.38 Effects of Nitrite Ions. When sodium nitirite was added to the solution at pH 3.0, all the films containing the Ru(II) species (Ru, Ru-mont, and Ru-Co) gave large catalytic oxidation currents in proportion to the concentration of sodium nitrite. Further, the three types of films had nearly the same activity. The observed effect of nitrite ions is rationalized in terms of catalytic oxidation mediated by Ru(III) according to the following mechanism (eqs 3 and 4): k-3

RuII {\ } RuIII + ek

(3)

3

k

4 RuIII + 21HNO2 + 21H2O98RuII + 21NO3- + 23H+

(4)

in which k3, k-3, and k4 denote the rate constants of the indicated steps, respectively. The following equations can be derived from the above mechanism:

d[RuII] ) -k3[RuII] + k-3[RuIII][e-] + dt [H2O] k4 [RuIII][HNO2] (5) 2 2 (42) Stucki, J. W. In Handbook of Clay Science; Bergaya, F., Theng, B. K. G., Lagaly, G., Eds.; Elsevier: Amsterdam, 2005; Chapter 8.

Electron Transfer through Clay Monolayer Films

d[e-] ) k3[RuII] - k-3[RuIII][e-] - k5[e-] dt

Langmuir, Vol. 22, No. 23, 2006 9597

(6)

Here, k5 is the rate constant for an electron passing through a film. Assuming the stationary state conditions for reactions 3 and 4, both the amounts of RuIII complexes and electrons are constant:

d[RuIII] d[e-] ) 0 and )0 dt dt

(7)

The oxidation current can thus be derived as:

i ) k5[e-] ) 4k4[H2O][RuIII][NO2-] ) k4′[NO2-] (8) where

(k4′ ) 4k4[H2O][RuIII] ) constant)

(9)

Equations 8 and 9 show that the current should be proportional to the amount of nitrite ions in a solution, as was found experimentally in Figure 9. Since k4 is the rate constant of the reduction of Ru(III) by nitrite, the slope of the plots of the current versus [NO2-] should not be dependent on the presence or absence of clay or on the type of clay used. This again coincides with the experimental results. In other words, the presence of clay particles in the films of Ru(II) complexes modifying an ITO substrate, whether it might be either montmorillonite or Co(II) smectite, was found to have little effect on the catalytic oxidation of nitrite ions in an electrolyte solution. In the previous reports, we observed a similar catalytic oxidation of binaphthol on ITO electrodes modified with Os(II) complex-clay hybrid films.32 We can therefore postulate that electron transfer through a clay monolayer is possible when an sufficiently electron-donating

species is present in an electrolyte. This is thought to reduce the backward reaction of reaction 3.

Conclusions Three types of amphiphilic molecule-clay hybrid films (Rumont, ODAH-Co, and Ru-Co) and one LB film of pure Ru(II) complex salt (Ru) were fabricated by the modified LB method. The effect of clay concentration on the packing of clay particles in these hybrid films was analyzed by AFM. The most closely packed films were obtained at the clay concentration of 20 ppm for all hybrid films. CV measurements on the hybrid films were done in 0.1 M NaClO4 and in a pH 3.0 buffer solution. In the following two cases, the clear evidence of electron transfer through a clay monolayer was found. (1) For the films prepared with a redox-active synthetic cobalt Smectite, the voltammetric peak current peaks were nearly 10 times larger than those in a equivalent film prepared with a natural aluminum clay, and (2) in the presence of nitrite ions, Ru(III) was found to mediate catalytic oxidation either in the presence or the absence of a clay monolayer between the ITO substrates and the ruthenium headgroups. Acknowledgment. This work was supported by CREST of JST (Japan Science and Technology Agency). This work was financially supported by a Grant-in-Aid for Scientific Research on Priority Areas (417) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government. Financial support was also provided by the Natural Science and Engineering Research Council of Canada (NSERC). Thanks are due to Prof. T. Kogure (The University of Tokyo) for his comments on the SEM results and Prof. A. Kobayashi (The University of Tokyo) for the use of equipment. Supporting Information Available: Experimental details; π-A isotherms; and AFM, SEM, and CV data. This material is available free of charge via the Internet at http://pubs.acs.org. LA061668F