Effective Luminescence Quenching of Tris(2,2-bipyridine)ruthenium(II

The quenching rate was high when the clay with larger particle size was ...... Theng, B. K. G. The chemistry of clay organic reactions; Adam Hilger: L...
0 downloads 0 Views 118KB Size
Download PDF
7004

Langmuir 2004, 20, 7004-7009

Effective Luminescence Quenching of Tris(2,2-bipyridine)ruthenium(II) by Methylviologen on Clay by the Aid of Poly(vinylpyrrolidone) Norishige Kakegawa† and Makoto Ogawa*,‡ Graduate School of Science and Engineering, Waseda University, Nishiwaseda 1-6-1, Shinjuku-ku, Tokyo 169-8050, Japan, and Department of Earth Sciences, Waseda University, Nishiwaseda 1-6-1, Shinjuku-ku, Tokyo 169-8050, Japan Received November 25, 2003. In Final Form: June 7, 2004 Electron-transfer quenching of tris(2,2-bipyridine)ruthenium(II) by methylviologen in an aqueous suspension of clay in the presence of poly(vinylpyrrolidone) was investigated. The quenching behavior of the excited tris(2,2-bipyridine)ruthenium(II) on clay by the coadsorbed methylviologen indicated the homogeneous distribution of the adsorbed dyes. The quenching rate was high when the clay with larger particle size was used as the host. The adsorption of poly(vinylpyrrolidone) on clay resulted in the coadsorption of the tris(2,2-bipyridine)ruthenium(II) and methylviologen without segregation.

Introduction Studies on the intercalation of dyes on smectites and the photoprocesses of the adsorbed dyes have been extensively conducted, partly for the purpose of constructing functional supramolecular systems.1 For organizing photoactive species, smectite possesses various attractive features such as large surface area, swelling behavior, and ion exchange properties.2 Efforts have been made to accomplish controlled intermolecular reactions by organizing reactants on smectites. However, there are few examples to realize intermolecular reactions on smectites.3-6 Photoprocesses of tris(2,2-bipyridine)ruthenium(II) (abbreviated as [Ru(bpy)3]2+) intercalated in smectites have been studied.7-13 Electron-transfer quenching of adsorbed [Ru(bpy)3]2+ by transition metal ions, nitrobenzene, and dimethylaniline on smectites has been reported.7,10-12 On the other hand, the adsorbed methylviologen (abbreviated as MV2+) did not quench the excited [Ru(bpy)3]2+ on the clay, since the intercalation of the two cationic species results in segregation, where they are intercalated in different gallery spaces.9 * To whom correspondence should be addressed. E-mail: [email protected]. Tel: +81-3-5286-1511. Fax: +81-3-3207-4950. † Graduate School of Science and Engineering. ‡ Department of Earth Sciences. (1) Ogawa, M.; Kuroda, K. Chem. Rev. 1995, 95, 399-438. (2) Theng, B. K. G. The chemistry of clay organic reactions; Adam Hilger: London, 1974. (3) Margulies, L.; Rozen, H.; Cohen, E. Nature 1985, 315, 658-659. (4) Miyata, H.; Sugahara, Y.; Kuroda, K.; Kato, C. J. Chem. Soc., Faraday Trans. 1 1987, 83, 1851-1858. (5) Itoh, T.; Ishi, A.; Kodera, Y.; Matsushima, A.; Misao, H.; Nishimura, H.; Tsuzuki, T.; Kamachi, T.; Okura, I.; Inada, Y. Bioconjugate Chem. 1998, 9, 409-412. (6) Takagi, S.; Tryk, D. A.; Inoue, H. J. Phys. Chem. B 2002, 106, 5455-5460. (7) Dellaguardia, R. A.; Thomas, J. K. J. Phys. Chem. 1983, 87, 990998. (8) Schoonheydt, R. A.; Depauw, P.; Vliers, D.; Deschrijver, F. C. J. Phys. Chem. 1984, 88, 5113-5118. (9) Ghosh, P. K.; Bard, A. J. J. Phys. Chem. 1984, 88, 5519-5526. (10) Nakamura, T.; Thomas, J. K. Langmuir 1985, 1, 568-573. (11) Thomas, J. K. Acc. Chem. Res. 1988, 21, 275-280. (12) Kuykendall, V. G.; Thomas, J. K. J. Phys. Chem. 1990, 94, 42244230. (13) Awaluddin, A.; Deguzman, R. N.; Kumar, C. V.; Suib, S. L.; Burkett, S. L.; Davis, M. E. J. Phys. Chem. 1995, 99, 9886-9892.

In this study, electron-transfer quenching of [Ru(bpy)3]2+ by methylviologen in aqueous suspensions of clays in the presence of poly(vinylpyrrolidone) (abbreviated as PVP) was investigated. We have already reported the photoprocesses of the intercalated [Ru(bpy)3]2+ in the interlayer space of a swelling mica with PVP.14,15 The cointercalated PVP was forced to surround [Ru(bpy)3]2+ in close contact in the statically limited interlayer spaces to cause luminescence shifts and suppression of self-quenching. This result suggested that two cationic species ([Ru(bpy)3]2+ and exchangeable sodium ion) were intercalated in the same interlayer space of the swelling mica without segregation. In the present study, coadsorption of the [Ru(bpy)3]2+ and MV2+ on clays in aqueous suspensions in the presence of PVP was conducted. The added PVP in a clay aqueous suspension depressed the flocculation of [Ru(bpy)3]2+/MV2+-clay. The quenching behavior of the excited [Ru(bpy)3]2+ and MV2+ in suspension was examined to reveal the spatial distribution of the adsorbed species on clays. Experimental Section Materials. Synthetic sodium-saponite (Sumecton SA, Kunimine Industries Co.; reference clay sample of the Clay Science Society of Japan), synthetic sodium-hectorite (Laponite XLG, Rockwood Additives Ltd.), and sodium swelling mica (abbreviated as ME-100, Co-op Chemical Co., Ltd.) were used as received. Sodium fluorotetrasilicic mica (abbreviated as TSM, Topy Ind. Co.) was used after removing nonexpandable impurities by a dispersion-sedimentation method.16 TSM is a synthetic 2:1 type layered clay17 and exhibits ion exchange and adsorptive properties similar to those of smectites.14,15,18,19 ME-100 is made from talc by a solid-state reaction.20,21 The chemical composition of ME100 is almost the same as that of TSM (NaMg2.5SiO10F2). The (14) Ogawa, M.; Inagaki, M.; Kodama, N.; Kuroda, K.; Kato, C. J. Phys. Chem. 1993, 97, 3819-3823. (15) Ogawa, M.; Tsujimura, M.; Kuroda, K. Langmuir 2000, 16, 42024206. (16) TSM was used after removing the nonexpandable impurities and the size fraction. The TSM (100 g) was dispersed in water (100 mL). The TSM aqueous suspension was centrifuged (3000 rpm, 10 min) to remove nonexpandable impurities. The supernatant was centrifuged again (20 000 rpm, 10 min) to remove finite size fractions, and the precipitate was used as the host material. (17) Kitajima, K.; Daimon, N. Nihon Kagakukaisi 1975, 4, 11681174. (18) Soma, M.; Tanaka, A.; Seyama, H.; Hayashi, S.; Hayamizu, K. Clay Sci. 1990, 8, 1.

10.1021/la036213u CCC: $27.50 © 2004 American Chemical Society Published on Web 07/23/2004

Luminescence Quenching of [Ru(bpy)3]2+ swelling and cation exchange properties of ME-100 are similar to those of smectites. The cation exchange capacity of the clays was 71, 63, 86, and 97 mequiv/100 g of host for Sumecton, Laponite, ME-100, and TSM, respectively. The clays (100 mg) were dispersed in 200 mL of water in order to measure the particle size of the clays. From a light scattering measurement, it was confirmed that the particle sizes of the Sumecton and Laponite were smaller than 0.02 µm and the average particle sizes of TSM and ME-100 were 1 and 5 µm, respectively. [Ru(bpy)3]2+ chloride hexahydrate, MV2+ chloride, and PVP (Mw ) 10 000) were obtained from Tokyo Kasei Ind. Co. and were used as received. Syntheses of [Ru(bpy)3]2+-Clay Intercalation Compounds. [Ru(bpy)3]2+-clay intercalation compounds were prepared by cation exchange reaction of Na-type clays with an ethanol solution of [Ru(bpy)3]2+ chloride at room temperature for 1 day. The loading amount of [Ru(bpy)3]2+ was 1.0 mequiv/ 100 g of host. Preparation of [Ru(bpy)3]2+/MV2+-Clay-PVP Suspensions. Preparation of [Ru(bpy)3]2+-clay-PVP suspensions has been reported by Ogawa et al.15 The [Ru(bpy)3]2+-clay intercalation compounds (5.0 mg) were added to an aqueous solution of PVP (10, 9.95, 9.75, 9.5, 9.75, and 9.0 mL), and the mixtures were allowed to react at room temperature for 4 h. The added PVP amount was 2.5 mg. The molar ratio of a monomeric unit of PVP to [Ru(bpy)3]2+ on clay was calculated to be 900:1. An aliquot of MV2+ aqueous solution (0.5 mmol/L) was added to the [Ru(bpy)3]2+-clay-PVP aqueous suspensions, and the suspensions were magnetically stirred for 1 day in the dark at room temperature. The added MV2+ aqueous solution amounts were 0, 0.05, 0.25, 0.50, 0.75, and 1.0 mL. The total volume of the suspensions was adjusted to 10.0 mL. Consequently, the concentration of [Ru(bpy)3]2+ in the aqueous suspension was 0.0025 mmol/L and the molar ratios of the added MV2+/[Ru(bpy)3]2+ were 0, 1, 5, 10, 15, and 20. Characterization. The particle size of the clay dispersed in water was measured by a HORIBA LA-920 particle size distribution analyzer. UV-vis absorption spectra of the products were recorded on a Shimadzu UV-3100PC spectrophotometer. Steady-state luminescence spectra were recorded on a Hitachi F-4500 fluorospectrophotometer with the excitation wavelength of 468 nm. Luminescence lifetimes were measured by a singlephoton-counting technique on a HORIBA NAES-700 timecorrelated spectrophotometer equipped with a hydrogen lamp. These measurements were conducted under ambient atmosphere at room temperature. Kinetic data were extracted from the bestfit simulate transients. The best fits were determined by the method of nonlinear least-squares. Data fittings were performed with Origin7.0 softwear (Originlab Co.).

Results Sample Preparation. By the reactions between Natype clays and [Ru(bpy)3]2+, orange solids were obtained. The quantitative adsorption of the added [Ru(bpy)3]2+ was confirmed by the visible absorption spectra of the supernatants where the metal-to-ligand charge transfer (MLCT) absorption band (at around 452 nm) of [Ru(bpy)3]2+ was not detected. The [Ru(bpy)3]2+-TSM-PVP and [Ru(bpy)3]2+-ME-100-PVP aqueous suspensions were stable for several days without significant flocculation. No flocculation was observed for the [Ru(bpy)3]2+-LaponitePVP and [Ru(bpy)3]2+-Sumecton-PVP suspensions for several months. The quantitative adsorption of the added MV2+ was also confirmed by the visible absorption spectra of the supernatants where the π-π* absorption band (at around 257 nm) of MV2+ was not detected. The suspension was stable even after the addition of MV2+. However, when the loading amount of MV2+ on (19) Ogawa, M.; Hama, M.; Kuroda, K. Clay Miner. 1999, 34, 213220. (20) Tateyama, H.; Nishimura, S.; Tsunematsu, K.; Jinnai, K.; Adachi, Y.; Kimura, K.; Furusawa, T. Clays Clay Miner. 1992, 40, 180-185. (21) Tateyama, H.; Tsunematsu, K.; Jinnai, K.; Noma, H.; Adachi, Y. J. Am. Ceram. Soc. 1996, 179, 3321-3324.

Langmuir, Vol. 20, No. 17, 2004 7005

Figure 1. Absorption spectra of the [Ru(bpy)3]2+/MV2+-TSMPVP aqueous suspension (a), the [Ru(bpy)3]2+/MV2+-TSM aqueous suspension (b), the MV2+ aqueous solution (c), and the MV2+-PVP aqueous solution (d). The concentration of MV2+ was 50 mmol/L, and the molar ratio of MV2+ to [Ru(bpy)3]2+ was 20:1.

clays was over 20 mequiv/100 g clay, the [Ru(bpy)3]2+/ MV2+-clay-PVP flocculated in water. On the other hand, the flocculation of [Ru(bpy)3]2+/MV2+-clay without PVP occurred even when the added amount of the photoactive species was smaller than 20 mequiv/100 g clay. Thus, the added PVP depressed the flocculation of the [Ru(bpy)3]2+/ MV2+-clay. Spectroscopic Studies of the Adsorbed [Ru(bpy)3]2+ and MV2+ on Clay-PVP Compounds. As typical examples, visible absorption spectra of a [Ru(bpy)3]2+/MV2+-TSM-PVP aqueous suspension, a [Ru(bpy)3]2+/MV2+-TSM aqueous suspension, a MV2+ aqueous solution, and a MV2+-PVP aqueous solution are shown in Figure 1a-d, respectively. The concentration of MV2+ was 50 mmol/L, and the molar ratio of MV2+ to [Ru(bpy)3]2+ was 20:1. In the visible absorption spectra of the [Ru(bpy)3]2+/MV2+-TSM-PVP and [Ru(bpy)3]2+/MV2+-TSM aqueous suspensions (Figure 1a,b, respectively), the MLCT absorption band of [Ru(bpy)3]2+ appeared at a longer wavelength region (at 468 nm) if compared with that (452 nm) observed for the [Ru(bpy)3]2+ aqueous solution. The π-π* absorption band of the adsorbed MV2+ also appeared at a longer wavelength region (at 269 and 273 nm for the absorption band shown in Figure 1a,b, respectively) if compared with those (257 nm) observed for the MV2+ (Figure 1c) and MV2+-PVP aqueous solutions (Figure 1d). The steady-state luminescence spectra of the [Ru(bpy)3]2+/MV2+-TSM-PVP aqueous suspension are shown in Figure 2. The MLCT luminescence of the [Ru(bpy)3]2+ was observed at 600 nm. The luminescence maximum was almost constant irrespective of the hosts and the MV2+ loading. On the other hand, the emission intensity became weak as the concentration of the added MV2+ increased. Figure 3 shows emission decays of the [Ru(bpy3)]2+/ MV2+-TSM-PVP suspensions. The decay for the suspension without MV2+ followed a single-exponential kinetics. On the other hand, the decays for the [Ru(bpy3)]2+/ MV2+-TSM-PVP suspensions followed a double-exponential kinetics. The emission of the adsorbed [Ru(bpy3)]2+ decayed fast with increasing amounts of the added MV2+.

7006

Langmuir, Vol. 20, No. 17, 2004

Figure 2. Luminescence spectra of [Ru(bpy)3]2+/MV2+-TSMPVP suspensions.

Discussion The π-π* adsorption band maximum (273 nm) for MV2+ in the [Ru(bpy)3]2+/MV2+-TSM aqueous suspension (Figure 1b) appeared at a longer wavelength region than that (257 nm) in the [Ru(bpy)3]2+/MV2+-PVP aqueous suspension (Figure 1c). Similar red shifts have been previously reported when MV2+ was absorbed onto MV2+ smectite.22,23 The π-π* absorption maximum observed at 275 nm24 and 280 nm22,25 was attributed to MV2+ intercalated into smectite layers. The red-shift reflected that the clay surface provided the environment with different polarity compared with the water. The other reason for the red shift is the difference in the planarity of the pyridine.25 When the molecule is in a planar conformation such as the intercalated porphyrin in the interlayer space of smectite,26 the number of functional π electrons is maximized, causing a red shift in the π-π* transitions.27 The π-π* adsorption band maximum (273 nm) of MV2+ in the spectrum of the [Ru(bpy)3]2+/MV2+-TSM aqueous suspension appeared at a longer wavelength region than that (269 nm) of the [Ru(bpy)3]2+/MV2+-TSM-PVP aqueous suspension. It was thought that coadsorbed PVP also affected the electronic state of the adsorbed MV2+ on TSM.14,15 We have reported the relationship between the wavelength of the luminescence maxima of the adsorbed [Ru(bpy)3]2+ on TMS-PVP and the average distance of the [Ru(bpy)3]2+.15 The luminescence maxima did not change further when the average [Ru(bpy)3]2+-[Ru(bpy)3]2+ distance was longer than 6 nm. In the present system, the average [Ru(bpy)3]2+-[Ru(bpy)3]2+ distance on clay is longer than 8 nm, which is determined by the ideal surface area (supposed to be 700 m2/g of clay, which is calculated from the surface area of each cell)28 and the (22) Haque, R.; Lilley, S.; Coshow, W. R. J. Colloid Interface Sci. 1970, 33, 185-189. (23) Villemure, G.; Detellier, C.; Szabo, A. G. Langmuir 1991, 7, 1215-1221. (24) Kakegawa, N.; Kondo, T.; Ogawa, M. Langmuir 2003, 19, 35783582. (25) Rytwo, G.; Nir, S.; Margulies, L. Soil Sci. Soc. Am. J. 1996, 60, 601-610. (26) Cady, S. S.; Pinnavaia, T. J. Inorg. Chem. 1978, 17, 1501-1507. (27) Berlmann, B. I. J. Phys. Chem. 1970, 74, 3085-3093.

Kakegawa and Ogawa

composition, indicating that the intermolecular interactions are thought to be negligible. In addition, the wavelength of the luminescence maxima observed in the present study was almost constant (600 nm) irrespective of the MV2+ loadings and the kind of hosts. This result suggested that the distribution of the adsorbed [Ru(bpy)3]2+ on TSM did not change by the adsorbed MV2+. When the adsorbed amounts of the [Ru(bpy)3]2+ were lower than 1.0 mequiv/100 g clay, the position of the luminescence maxima (600 nm) and the luminescence intensity of the adsorbed [Ru(bpy)3]2+ on clay were almost constant irrespective of the presence of PVP. Therefore, [Ru(bpy)3]2+-[Ru(bpy)3]2+ interactions are thought to be negligible at the present low loadings of [Ru(bpy)3]2+. The emission decay of the [Ru(bpy)3]2+-TSM-PVP suspensions was fitted by a double-exponential model (Figure 3) indicating that the [Ru(bpy)3]2+ adsorbed on two sites of the clay surfaces even when the coadsorbed MV2+ was absent.12 The normalized emission decays of the [Ru(bpy)3]2+/ MV2+-TSM-PVP suspensions are shown in Figure 4. The vertical axis expressed the relative emission intensity of I(t)/I(0), where I(t) and I(0) are emission intensity at time t and 0, respectively. The emission decays of the suspension are almost linear. The decay curves of the suspensions were fitted by the Albery model.29 This model assumes that the dispersion in the kinetics is the result of a change in energy of activation for the reaction at different sites in the system. The Albery model assumes that the observed kinetics can be described by a Gaussian distribution of the energies of activation or the natural logarithm of the decay rate constant. In this model, the emission intensity is given by

I(t) 1 ) I(0) xπ

∫-∞∞ exp(-x2) exp[-(kavt)eγx] dx

(1)

where kav is the average rate constant and γ is the distribution width. The calculated values are summarized in Table 1. Average emission lifetimes τ ()1/kav) became short as the concentration of the added MV2+ increased. The γ values (ca. 0.37) were constant, irrespective of the added amounts of the MV2+. Luminescence lifetimes decreased with the increase of the loading amount of MV2+, suggesting that the excited [Ru(bpy)3]2+ was quenched by the adsorbed MV2+. The rate of electron transfer from [Ru(bpy)3]2+ to MV2+ on clay should vary depending on the distance r between the [Ru(bpy)3]2+ and MV2+ on the clay. In such a case, there should be a dispersion of r according to a normal Gaussian distribution, exp(-x2), about some mean value (rav): r ) rav - γx. When γ ) 0, there is no distribution, and the system will behave in a classical homogeneous fashion. In the present system, the γ values were low (ca 0.37) and constant, irrespective of the added amounts of MV2+ (Table 1). This result indicated that the adsorbed [Ru(bpy)3]2+ and MV2+ dispersed on clays almost homogeneously. The γ values obtained from the fitting data of the emission decay of the adsorbed [Ru(bpy)3]2+ on Laponite-PVP were low and constant (ca. 0.3), irrespective of the loading amount of MV2+. Accordingly, the distribution of the adsorbed [Ru(bpy)3]2+ and MV2+ on Laponite was also homogeneous. (28) Van Olphen, H. An introduction to clay colloid chemistry, 2nd ed.; Wiley-Interscience: New York, 1977. (29) Albery, W. J.; Bartlett, P. N.; Wilde, C. P.; Darwent, J. R. J. Am. Chem. Soc. 1985, 107, 1854-1858.

Luminescence Quenching of [Ru(bpy)3]2+

Langmuir, Vol. 20, No. 17, 2004 7007

Figure 3. Logarithmic plots of the emission decay of the adsorbed [Ru(bpy)3]2+ on TSM at different concentrations of MV2+. [MV2+] in the suspension is 0 (a), 0.0025 (b), 0.0050 (c), 0.0075 (d), and 0.0100 (e) mmol/L. Solid lines are fits to the double-exponential model.

Figure 4. The decays of the relative emission intensity of the [Ru(bpy3)]2+/MV2+-TSM-PVP suspensions. To avoid the effect of the excitation light, I(50 ns) was used as I(0). The molar ratios of MV2+/[Ru(bpy)3]2+ were 0 (a), 5 (b), 10 (c), 15 (d), and 20 (e). Table 1. Emission Lifetimes and Fitting Parameters for the Added [Ru(bpy3)]2+ Quenching by MV2+ on TSM, Calculated from Nonlinear Least-Squares Fitting of Experimental Data to the Albery Kinetic Model molar ratio of MV2+/[Ru(bpy3)]2+ kav × 106, s-1 τ × 10-7, s 0 5 10 15 20

2.7 5.9 7.0 10.1 8.8

3.70 1.69 1.42 0.99 1.13

g

irreducibleχ2

0.39 0.37 0.37 0.32 0.36

0.00009 0.00009 0.00007 0.00009 0.00008

Stern-Volmer plots obtained from the steady-state emission data of the [Ru(bpy)3]2+/MV2+-TSM-PVP aqueous suspensions (Figure 2) and the average emission lifetimes derived from the curve fitting data by the Albery model (Figure 4 and Table 1) and by the double-exponential model (τ1 and τ2, summarized in Table 1) are shown in Figure 5. These plots were obtained from Stern-Volmer equation (eq 2),

I/I0 or τ/τ0 ) 1 + Ksv[MV2+]

(2)

where I0(τ0) and I(τ) are the luminescence intensity

Figure 5. Stern-Volmer plots of [Ru(bpy3)]2+/MV2+-TSMPVP suspensions derived from the steady-state emission data (b) and lifetimes obtained from the Albery model (2) and the double-exponential model (0).

(lifetime) in the absence and the presence of MV2+, respectively. Ksv is the Stern-Volmer constant. The Stern-Volmer plot obtained from the steady-state emission data (Figure 5, filled circles) shows good linearity with the positive inclination to the increase of the loading amount of the MV2+. The Stern-Volmer plots obtained from the emission lifetimes (Figure 5, filled triangles and squares) also are almost linear, and the inclination of the plots is almost the same as the inclination of the SternVolmer plot obtained from the steady-state emission data. The good linearity of the Stern-Volmer plots (Figure 5) indicated that the adsorbed MV2+ quenched the excited [Ru(bpy)3]2+ effectively. Note that the Ksv value of the Stern-Volmer plot became significantly smaller when the concentration of the MV2+ in the suspension was over 20 mequiv/100 g clay. This result reflected the flocculation of the [Ru(bpy)3]2+-MV2+clay-PVP in water. Figure 6 shows the Stern-Volmer plots obtained from the steady-state emission data of the [Ru(bpy)3]2+/MV2+TSM-PVP aqueous suspensions with different concen-

7008

Langmuir, Vol. 20, No. 17, 2004

Kakegawa and Ogawa

Figure 6. Quasi-Stern-Volmer plots of [Ru(bpy3)]2+/MV2+TSM-PVP suspensions stirred for 1 day (b) and for 1 week (0). (2, ×) Data obtained after diluting suspensions stirred for 1 week (0) by 2 and 10 times, respectively. Table 2. Stern-Volmer Constants, Luminescence Lifetime of [Ru(bpy)3]2+, and Bimolecular Quenching Rate Constants of [Ru(bpy)3]2+ and MV2+ in the Suspensions and Aqueous Solution Ksv, (mol/L)-1

silicates used as host TSM

ME-100 Laponite Sumecton

stirred for 1 day stirred for 1 week diluted 2 timesa diluted 10 timesa stirred for 1 day stirred for 1 day stirred for 1 day

42.6 × 103 43.3 × 103 96.7 × 103 419 × 103 23.4 × 103 20.1 × 103 15.9 × 103

a The suspension, which was magnetically stirred for a week, was diluted by 2 or 10 times.

trations and stirring times. The Ksv values obtained from the inclination of the Stern-Volmer plots shown in Figure 6 are summarized in Table 2. The Ksv value obtained from the emission data of the [Ru(bpy)3]2+/MV2+-TSM-PVP suspensions was hardly changed after the suspensions were magnetically stirred for 1 week (Figure 6, open squares and filled circles, before and after stirring for 1 week, respectively). This fact suggested that the segregation of the adsorbed [Ru(bpy)3]2+ and MV2+ and flocculation of the clay-PVP did not occur by stirring the suspension for such period. Distribution of the adsorbed photoactive species on clay also was homogeneous. The Ksv values derived from the Stern-Volmer plots obtained from the emission data of the suspensions, which were diluted 2 and 10 times by water (Figure 6, filled triangles and cross marks, respectively), increased almost proportionally to the degree of the dilution. The concentration of the MV2+ in suspension was decreased by the dilution. However, assuming a TSM particle as a unit, the concentration of the adsorbed MV2+ on TSM did not change before and after the dilution. If the concentration of MV2+ did not change before and after the dilution, the Ksv values (Figure 6, filled circles, open squares, triangles, and cross marks) were almost the same. This result indicated that the electron-transfer quenching of the [Ru(bpy)3]2+ by the MV2+ mainly occurred in an intrasheet of the silicate. If the electron transfer occurred between

different particles, the quenching rate should become lower with the decrease of encounter probability between the [Ru(bpy)3]2+/MV2+-TSM-PVP particles in the suspension. However, in the present system, the electron transfer was not affected by the concentration of the clay in the suspension but was affected by the population of the adsorbed MV2+ on clay. Therefore, it is thought that photoinduced electron transfer between coadsorbed [Ru(bpy)3]2+ and MV2+ on the same clay particle occurred. In addition, the bimolecular rate constant for the quenching kq ()Ksv/τ0) in the [Ru(bpy)3]2+/MV2+-TSM-PVP suspension (1.1 × 1011 mol/L-1 s-1) is much higher than that in the [Ru(bpy)3]2+ and MV2+ aqueous solution (5.0 × 108 mol/L-1 s-1),30 indicating that the quenching of excited [Ru(bpy)3]2+ by adsorbed MV2+ occurred at a close distance on clay. Accordingly, the electron transfer is thought to occur mainly between the dyes adsorbed on the same clay particles. Ghosh and Bard9 reported that the intercalation of [Ru(bpy)3]2+ and MV2+ into the interlayer space of smectite resulting in segregation and photoinduced electron transfer between [Ru(bpy)3]2+ and MV2+ on smectite did not occur. In the present system, the adsorption of PVP on clay resulted in the adsorption of [Ru(bpy)3]2+ and MV2+ on the clay without segregation. The electron-transfer quenching of the adsorbed [Ru(bpy)3]2+ by MV2+ occurred mainly on the intrasheet of the hosts. In addition, the distribution of the adsorbed photoactive species on clay is almost homogeneous. Note that the methylviologen radical cation in suspension was not detected by the UV-vis absorption spectroscopy. This result suggested that the lifetime of the methylviologen radical cation was so short that the formed radical cation could not be detected. However, the color of the precipitate changed to blue upon UV irradiation. Miyata et al. reported that the oxidation of the viologen radical cation in a viologen-montmorillonite-PVP film was affected by the film thickness.4 The gas permeability might depend on the nanostructure of the intercalation compounds. Therefore, in the present systems, the difference between the encounter probability of oxygen to the methylviologen radical cation in suspension and that in precipitate might affect the lifetime of the formed methylviologen radical cation. Thus, we concluded that the luminescence quenching of the adsorbed [Ru(bpy)3]2+ in the suspension was due to electron transfer. Cu(II) was used as a quencher instead of MV2+. The Ksv value derived from the Stern-Volmer plots obtained from the emission data of the Cu(II)/[Ru(bpy)3]2+-TSM-PVP suspensions is almost same as that of the Cu(II)/ [Ru(bpy)3]2+-TSM suspensions. This result agreed with the previous study, which confirmed the coadsorption of Cu(II) and [Ru(bpy)3]2+ on smectite.12 On the other hand, the Ksv value derived from the Stern-Volmer plots obtained from the emission data of the MV2+/[Ru(bpy)3]2+-TSM suspensions was ca. 0, indicating that the segregation of MV2+ and [Ru(bpy)3]2+ on TSM occurred. Therefore, distribution of the adsorbed dye on clay was affected strongly by the coadsorbed PVP. Figure 7 shows the Stern-Volmer plots obtained from the steady-state emission data of the [Ru(bpy)3]2+/MV2+TSM-PVP, [Ru(bpy)3]2+/MV2+-ME-100-PVP, [Ru(bpy)3]2+/MV2+-Laponite-PVP, and [Ru(bpy)3]2+/MV2+Sumecton-PVP suspensions. The Ksv values derived from the inclination of the Stern-Volmer plots are summarized in Table 2. Irrespective of the kinds of clays, the plots (30) Kalyanasundaram, K.; Kiwi, J.; Gra¨tzel, M. Helv. Chim. Acta 1978, 61, 2721-2730.

Luminescence Quenching of [Ru(bpy)3]2+

Figure 7. Quasi-Stern-Volmer plots of [Ru(bpy3)]2+/MV2+TSM-PVP (b), [Ru(bpy3)]2+/MV2+-ME-100-PVP (0), [Ru(bpy3)]2+/MV2+-Laponite XLG-PVP (2), and [Ru(bpy3)]2+/ MV2+-Sumecton SA-PVP (×) suspensions.

show linearity with the positive inclination to the increase of the loading MV2+ amount. The Ksv value obtained from the emission data of the [Ru(bpy)3]2+/MV2+-TSM-PVP suspensions is higher than the other Ksv values. The difference of the quenching rate reflected the impurity and particle size of the hosts. ME-100 contains iron as an impurity (