© Copyright 2004 American Chemical Society
JULY 6, 2004 VOLUME 20, NUMBER 14
Letters Photoinduced Electron Storage and Surface Plasmon Modulation in Ag@TiO2 Clusters Tsutomu Hirakawa† and Prashant V. Kamat*,†,‡ Notre Dame Radiation Laboratory and Department of Chemical & Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556-0579 Received May 6, 2004 The reversible charging and discharging effects associated with photoexcitation of a TiO2 shell in a Ag@TiO2 composite are described. The photoinduced charge separation in the TiO2 shell is followed by electron injection into the silver core. Interestingly, the charging of the silver core is associated with the shift in the surface plasmon band from 460 to 430 nm. The stored electrons are discharged upon exposure of the charged Ag@TiO2 composite to an electron acceptor. As the electrons from the silver core are discharged, the original surface plasmon absorption of the Ag core is restored.
Introduction Photoinduced charge separation in semiconductor nanoparticles and nanostructured films provides the basis for photocatalytic degradation of organic contaminants,1,2 the operation of dye-sensitized photochemical solar cells,3 and chemical sensors.4-6 Charge recombination within the particle or at the grain boundary often limits the efficiency of light energy conversion. By employing two different semiconductors7 or semiconductor-metal composites,8 one * To whom correspondence should be addressed. E-mail:
[email protected]. Web: http://www/nd.edu/∼pkamat. † Notre Dame Radiation Laboratory, University of Notre Dame. ‡ Department of Chemical & Biomolecular Engineering, University of Notre Dame. (1) Kamat, P. V. Chem. Rev. 1993, 93, 267. (2) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. Rev. 1995, 95, 69. (3) Cahen, D.; Graetzel, M.; Guillemoles, J. F.; Hodes, G. Dye Sensitized Solar Cells: Principles of Operation. In Electrochemistry of Nanomaterials; Hodes, G., Ed.; Wiley-VCH Verlag GmbH: Weinheim, 2001; p 201. (4) Shipway, A. N.; Katz, E.; Willner, I. Chem. Phys. Chem. 2000, 1, 18. (5) Kamat, P. V.; Huehn, R.; Nicolaescu, R. J. Phys. Chem. B 2002, 106, 788. (6) Seker, F.; Meeker, K.; Kuech, T. F.; Ellis, A. B. Chem. Rev. 2000, 100, 2505. (7) Kamat, P. V. Pure Appl. Chem. 2002, 74, 1693.
can suppress the charge recombination process to some extent. Previous studies have shown that, when a gold nanoparticle comes in contact with a photoexcited semiconductor, the two particles undergo charge equilibration.9,10 The shift in the Fermi level toward the conduction band energy enhances the catalytic efficiency of the composite system.11 Monolayer-protected gold clusters have also been shown to display quantized double-layer capacitance charging in differential voltammetry experiments.12 In recent years, several research groups have synthesized metal core@oxide shell structures for the purpose of designing drug delivery systems to biological tags.13-17 Yet, little is known about the charge-transfer processes (8) Kamat, P. V. J. Phys. Chem. B 2002, 106, 7729. (9) Wood, A.; Giersig, M.; Mulvaney, P. J. Phys. Chem. B 2001, 105, 8810. (10) Jakob, M.; Levanon, H.; Kamat, P. V. Nano Lett. 2003, 3, 353. (11) Subramanian, V.; Wolf, E. E.; Kamat, P. V. J. Am. Chem. Soc. 2004, 126, 4943-4950. (12) Chen, S.; Murray, R. W. J. Phys. Chem. B 1999, 103, 9996. (13) Ung, T.; Liz-Marzan, L. M.; Mulvaney, P. Langmuir 1998, 14, 3740. (14) Caruso, F.; Spasova, M.; Saigueirino-Maceira, V.; Liz-Marzan, L. M. Adv. Mater. 2001, 13, 1090. (15) Oldfield, G.; Ung, T.; Mulvaney, P. Adv. Mater. 2000, 12, 1519. (16) Pastoriza-Santos, I.; Koktysh, D. S.; Mamedov, A. A.; Giersig, M.; Kotov, N. A.; Liz-Marzan, L. M. Langmuir 2000, 16, 2731.
10.1021/la048874c CCC: $27.50 © 2004 American Chemical Society Published on Web 06/02/2004
5646
Langmuir, Vol. 20, No. 14, 2004
associated with such systems. Many interesting issues remain to be addressed. Can such core-shell structures perform as superior catalysts? How does the presence of the metal core affect the photocatalytic properties? Can they serve as light energy storage systems? Applying an electrochemical bias to a nanostructured semiconductor film is another effective strategy to modulate the interfacial charge-transfer processes.18 However, there are no known methods to modulate the charge transfer in semiconductor particles that are suspended in a liquid medium using photoexcitation. By incorporating a metal core within a TiO2 shell, it should be possible to modulate the charge transfer properties under band gap excitation. In this preliminary communication, we demonstrate our initial success of photoinduced charging and dark discharging of a silver core as a means to modulate the surface plasmon band of Ag@TiO2 clusters. The reversible charging and discharging of electrons is useful in light energy storage and nanoelectronics applications. Basic understanding of the charge transfer process of metal core@semiconductor shell systems is likely to benefit the development of next generation catalysts.
Letters
Figure 1. (A) Absorption spectra of Ag@TiO2 colloids in ethanol. (a) Before UV irradiation, (b-d) after exposure to UV light (λ > 300 nm) for 10, 30, and 60 s, and (e) after exposure of the UV-irradiated suspension to air. Transmission electron micrograph is shown in the inset. (B) The shift in the plasmon absorption peak during UV excitation followed by its exposure to air.
Experimental Section Ag-core-TiO2-shell (Ag@TiO2) clusters were prepared in N,Ndimethylformamide (DMF) by modifying the procedure reported earlier.16 A total of 2 mL of an aqueous AgNO3 (15 mM) solution was mixed with 18 mL i-PrOH solution of titanium(triethanolaminato)-isopropoxide [N((CH2)2O)3TiOCH(CH3)2] (7.5 mM) and 10 mL of DMF. The concentrations of Ag+ and TiO2 were 1 and 5 mM, respectively, in the reaction mixture. The mixture was heated at reflux temperature with vigorous stirring. After 90 min, the sample solution was cooled to room temperature. Unless otherwise mentioned, all experiments were performed using Ag@TiO2 with a composition of 0.33 mM for Ag and 1.7 mM for TiO2. The stability test of dispersing Ag@TiO2 in HNO3 solution confirmed the capping of Ag by TiO2. Only the TiO2-capped Ag colloids remain stable in acid solution. Ag colloids with no capping or mixed Ag and TiO2 colloids readily dissolved in acid solution. Absorption spectra were recorded with a Shimadzu UV-visNIR 3101 PC spectrophotometer. Transmission electron microscopy (TEM) was carried out with a Hitachi H600 microscope. Specimens were prepared by applying a drop of the colloidal suspension to a carbon-coated copper grid. Particle sizes were determined from the micrographs recorded at a magnification of 150 000× using a Hitachi H600 transmission electron microscope. The photoexcitation of TiO2 and the Ag@TiO2 suspension was carried out using collimated light from a 150-W xenon lamp filtered through a CuSO4 solution (λ > 300 nm).
Results and Discussion The TEM image of Ag-core-TiO2-shell particles employed in the present study is shown in Figure 1A. The overall diameter of these particles is in the range of 2-4 nm. The dark spots of the metal core show that these colloids are well separated by the capping layer of TiO2. Colloids prepared without the TiO2 shell show aggregation of particles on the TEM grid. Furthermore, larger crystallites showed uniform capping of the TiO2 layer (not shown). Although the shell structures were not clearly evident in this TEM image, we were able to test the presence of TiO2 capping by suspending them in a HNO3 solution. Only the Ag colloids capped with TiO2 survived in the acid solution. The absorption spectrum of the Ag@TiO2 colloids in ethanol exhibits a plasmon absorption band at 470 nm (17) Makarova, O. V.; Ostafin, A. E.; Miyoshi, H.; Norris, J. R.; Meisel, D. J. Phys. Chem. B 1999, 103, 9080. (18) Vinodgopal, K.; Hotchandani, S.; Kamat, P. V. J. Phys. Chem. 1993, 97, 9040.
Figure 2. Photoinduced charge separation and charging of the metal core in Ag@TiO2.
(Figure 1B). The presence of the TiO2 shell causes the plasmon absorption peak to shift to the red region. When the aerated suspension of Ag@TiO2 in ethanol was irradiated with UV light, we observe a blue shift in the plasmon absorption band. After 1 min of UV irradiation (λ > 300 nm), the plasmon peak shifts to 430 nm. Upon exposure to air, the plasmon absorption reverts to a position close to the original absorption at 470 nm. The small difference between the spectrum a and the spectrum e is attributed to the interfacial changes associated with UV irradiation. The interesting aspect of the present study is our ability to charge the metal core with electrons under UV excitation and discharge them on demand. The charging and discharging effects in Ag@TiO2 colloids can be monitored from the shift in maximum plasmon absorption (Figure 1B). It should be emphasized that the shift in the plasmon absorption is quick and can be reproduced during repeated charge and discharge cycles. Under UV excitation, charge separation occurs within the TiO2 shell. While the holes are scavenged by the surrounding ethanol, the electrons are injected into the silver core (Figure 2). The electron transfer continues until the two systems achieve Fermi level equilibration. The lack of absorption in the red-infrared region is an indication that there is no major electron accumulation within the TiO2 shell. (It should be noted that if we excite neat TiO2 particles in deaerated ethanol with UV light, we observe blue coloration as a result of electron accumulation.) As shown earlier,10 the addition of metal colloids to the UV-irradiated TiO2 suspension results in charge equilibration. In the present experiments, the metal core is surrounded by the TiO2 shell. As we scavenge the photogenerated holes with ethanol, electron
Letters
Figure 3. Absorption spectra recorded following UV irradiation of Ag@TiO2 and 17 µM thionine in deaerated ethanol.
transfer from the TiO2 shell to the silver core is, thus, facilitated. Further confirmation for the observed storage of electrons and its effect on the plasmon absorption was obtained by carrying out UV excitation in the presence of an electron acceptor, thionine (Figure 3). The dye absorption at 600 nm disappears upon its reduction to the leuco form. As long as we have the electron acceptor, thionine present in the solution, we do not observe any shift in the plasmon absorption. Once all the dye molecules are reduced, the plasmon absorption shifts to blue as electrons get stored in the Ag core. Two major factors influence the surface plasmon absorption peak of metal clusters: (i) dielectric of the medium and (ii) the density of electrons of the metal cluster. Whereas earlier studies demonstrated the dependence on the medium dielectric, little is known about the dependence of plasmon absorption on the varying electron density of suspended metal particles. According to the Drude model, the bulk plasmon frequency (ωp ) 2πc/λp) is proportional to the square root of the electron density of the particle.19 Because it is difficult to obtain an accurate estimate of the electron density in a colloidal particle, we correlated the trend of the plasmon shift and excess stored electrons with UV irradiation time. We estimated the number of excess electrons in the silver core by titrating with thionine.11 By assuming a uniform particle distribution and an average diameter as 3.7 nm, we estimate a concentration of 2.1 × 1017 Ag@TiO2 particles/L. The plasmon absorption maximum and number of stored electrons per particle are plotted versus UV(19) Mulvaney, P. Langmuir 1996, 12, 788.
Langmuir, Vol. 20, No. 14, 2004 5647
Figure 4. Dependence of the plasmon absorption peak and number of stored electrons per particle on the UV-irradiation times. The deaerated samples were subjected to different times of UV irradiation to monitor the changes in the surface plasmon band, and the stored electrons were titrated with thionine.
irradiation time (Figure 4). The blue shift in the plasmon absorption peak parallels the trend observed with that of stored electrons. Earlier spectroelectro-chemical studies have also shown a similar dependence of plasmon absorption upon charging the particles electrochemically20 or chemically.21 Increased electron density in the core also drives the Fermi level close to the conduction band of TiO2. The maximum number of electrons that can be stored in each Ag core is dictated by the Fermi level shift that is necessary to equilibrate with the TiO2 Fermi level. The experiments described here demonstrate the ability to modulate surface plasmon absorption by means of controlled charging of a metal core with electrons. These core@shell structures with their ability to store photogenerated electrons should be explored further for light energy conversion and storage. The charging and discharging events presented in this study open up new ways to utilize semiconductor-metal composites as microcapacitors and to tailor the photocatalytic properties of the TiO2-based particulate system. Acknowledgment. The work described herein was supported by the Office of the Basic Energy Sciences of the U.S. Department of Energy. This is contribution no. 4518 from the Notre Dame Radiation Laboratory. LA048874C (20) Ung, T.; Dunstan, D.; Giersig, M.; Mulvaney, P. Langmuir 1997, 13, 1773. (21) Oldfield, G.; Ung, T.; Mulvaney, P. Adv. Mater. 2000, 12, 1519.
