Photochemical Reactivity of Gold Clusters: Dependence on Size and

Jul 2, 2009 - Noble metal clusters of sizes comparable to the Fermi wavelength are known to exhibit molecule-like transitions owing to the discretion ...
1 downloads 14 Views 2MB Size
pubs.acs.org/Langmuir © 2009 American Chemical Society

Photochemical Reactivity of Gold Clusters: Dependence on Size and Spin Multiplicity† Masanori Sakamoto, Takashi Tachikawa, Mamoru Fujitsuka, and Tetsuro Majima* The Institute of Scientific and Industrial Research (SANKEN), Osaka University, Mihogaoka 8-1, Ibaraki, Osaka 567-0047, Japan Received April 30, 2009. Revised Manuscript Received June 10, 2009 Noble metal clusters of sizes comparable to the Fermi wavelength are known to exhibit molecule-like transitions owing to the discretion of the density of states. In the present article, the important factors influencing the reactivity of excited gold (Au) clusters are examined from the viewpoint of molecular photochemistry. The investigation of the differently sized Au clusters embedded in a polymer film using single-molecule fluorescence spectroscopy facilitates the further understanding of their size-dependent photoreactivity. In addition, it was discovered that the spin multiplicity of the excited state (i.e., singlet or triplet excited state) governs the photoreactivity.

Introduction Noble metal clusters of sizes comparable to the Fermi wavelength have recently attracted much attention because this unexplored size domain may yield interesting size-dependent optical, electronic, and chemical properties applicable to optoelectronic devices, catalysts, and fluorescent labels.1-8 In contrast to bulk metals, metal clusters in this size regime (∼1 nm in diameter) exhibit molecule-like transitions owing to the discretion of the density of states. Photochemical and electrical studies have demonstrated that specific noble metal clusters exhibit bright fluorescence following their transition from the excited state.9-11 However, most studies regarding excited noble metal clusters have focused on the fluorescence properties; other features have received much less attention. The photochemical reactivity of a metal cluster determines its fluorescence properties and, from what little in known, could hold the key to new insights concerning the characteristics of clusters comprising a countable number of atoms.12 However, thus far, all attempts to elucidate the photochemical reactivity of such metal † Part of the “Langmuir 25th Year: Nanoparticles synthesis, properties, and assemblies” special issue. *Corresponding author. E-mail: [email protected].

(1) Chen, S.; Ingram, R. S.; Hostetler, M. J.; Pietron, J. J.; Murray, R. W.; Schaaff, T. G.; Khoury, J. T.; Alvarez, M. M.; Whetten, R. L. Science 1998, 280, 2098. (2) Peyser, L. A.; Vinson, A. E.; Bartko, A. P.; Dickson, R. M. Science 2001, 291, 103. (3) Haruta, M. Chem. Rec. 2003, 3, 75. (4) Lee, T.-H.; Gonzalez, J. I.; Zheng, J.; Dickson, R. M. Acc. Chem. Res. 2005, 38, 534. (5) Branham, M. R.; Douglas, A. D.; Mills, A. J.; Tracy, J. B.; White, P. S.; Murray, R. W. Langmuir 2006, 22, 11376. (6) Zheng, J.; Nicovich, P. R.; Dickson, R. M. Annu. Rev. Phys. Chem. 2007, 58, 409. (7) Tsunoyama, H.; Ichikuni, N.; Tsukuda, T. Langmuir 2008, 24, 11327. (8) Rodrı´ guez-Vazquez, M. J.; Blanco, M. C.; Lourido, R.; Vazquez-Vazquez, C.; Pastor, E.; Planes, G. A.; Rivas, J.; Lopez-Quintela, M. A. Langmuir 2008, 24, 12690. (9) Vosch, T.; Antoku, Y.; Hsiang, J.-C.; Richards, C. I.; Gonzalez, J. I.; Dickson, R. M. Proc. Nat. Acad. Sci. U.S.A. 2007, 104, 12616. (10) Lin, C.-A. J.; Yang, T.-Y.; Lee, C.-H.; Huang, S. H.; Sperling, R. A.; Zanella, M.; Li, J. K.; Shen, J.-L.; Wang, H.-H.; Yeh, H.-I.; Parak, W. J.; Chang, W. H. ACS Nano 2009, 3, 395. (11) Dı´ ez, I.; Pusa, M.; Kulmala, S.; Jiang, H.; Walther, A.; Goldmann, A. S.; M€uller, A. H. E.; Ikkala, O.; Ras, R. H. A. Angew. Chem. 2009, 48, 2122. (12) Sakamoto, M.; Tachikawa, T.; Fujitsuka, M.; Majima, T. J. Am. Chem. Soc. 2009, 131, 6.

13888 DOI: 10.1021/la901552f

clusters have been hampered by the necessary use of capping agents, which stabilize the cluster but also tend to block direct access to the cluster surface. In the present article, the aim is to further the understanding of the photochemical reactivity through the investigation of differently sized gold (Au) clusters embedded in a polymer film using single-molecule fluorescence spectroscopy (SMS). SMS is a powerful and effective method of reaching the individual emissive Au clusters underlying the heterogeneous characteristics of the materials.9,12,13 The observation of molecule-accessible clusters isolated in the polymer matrix enables us to reveal the photochemical reactivity. With regard to the molecular photochemistry, the photochemical reactivity of the molecule-like clusters is speculated to depend on the structure (i.e., the size in the present study) and the spin multiplicity of the excited state (i.e., singlet or triplet excited state). Because the fluorescence spectra of Au clusters are characteristic of the number of atoms, SMS provides an opportunity to investigate the size dependence of the photochemical reactivity. Furthermore, we discovered strong evidence that spin multiplicity plays an important role in the reactivity of excited Au clusters.

Experimental Section Preparation of Sample Film. The poly(vinyl acetate) (PVAc, Mw 113 000, Aldrich) film containing a radical precursor (2-hydroxy-40 -(2-hydroxy-ethoxy)-2-methyl-propiophenone) and HAuCl4 (Aldrich) (PVAc) was prepared as follows. First, the precursor (5 mM) and HAuCl4 (1 mM) were dissolved in an acetonitrile solution of PVAc (20 wt %) and then poured in a Petri dish. Second, the sample was placed in a glovebox with constant humidity (20%) and temperature (20 °C) and dried for 1 week. Under these conditions, a film (thickness ca. 400-500 μm) containing the entrapped precursor and AuCl4- was obtained. The color of the PVAc film was pale yellow as a result of the absorption of the precursor and AuCl4-. Aging of the PVAc Film after UV Light Irradiation. The PVAc film was irradiated by the 365 nm UV lamp (2.8 mW, LUV-16, (13) Zheng, J.; Dickson, R. M. J. Am. Chem. Soc. 2002, 124, 13982. (14) Sakamoto, M.; Fujitsuka, M.; Majima, T. J. Photochem. Photobiol., C 2009, 10, 33.

Published on Web 07/02/2009

Langmuir 2009, 25(24), 13888–13893

Sakamoto et al.

Article

Figure 1. (A) Absorption spectra of the PVAc film before (black line, a) and after (red line, b) UV irradiation. The absorption spectrum of the aged PVAc film for 14 days after UV irradiation is also shown (blue line, c). (B) Time-resolved fluorescence spectra of the PVAc film after UV irradiation. The film was excited by 380 nm light. (C) Excitation and emission spectra of Au clusters. Black lines a and b are excitation and emission spectra of Aun, respectively. Red lines c and d are excitation and emission spectra of Aum (m>n), respectively. (D) Photograph of the emission from Au clusters in the PVAc film under the excitation of UV light. The Chinese character “light” was patterned using the photomask. The color of fluorescence changed from blue to pink during aging. Scheme 1. Formation and Growth Process of Au Clusters (Aun and Aum (m > n)) in the PVAc Film

Az one) for 30 min and then stored in the dark for 2 weeks (aging). During aging, the color of the PVAc film changed to vivid pink.

UV-Vis Absorption and Fluorescence Spectral Measurements. The UV-vis absorption and fluorescence spectra were measured by using a Shimadzu UV-3100 spectrometer and a Hitachi 850 spectrometer, respectively. Femtosecond Laser Flash Photolysis. The fluorescence lifetime of Au clusters was measured by the single-photon counting method using a streak scope (Hamamatsu Photonics, C4334-01) equipped with a polychromator (Acton Research, SpectraPro150). The second harmonic oscillation (420 nm) of the output of the femtosecond laser (Spectra-Physics, Tsunami 3941-M1BB; fwhm 80 fs; 840 nm) pumped by a diode-pumped solid-state laser (Spectra-Physics, Millennia VIIIs) was used to excite a sample. The instrument response function was also determined by measuring the scattered laser light in order to analyze a temporal profile. By using this method, we could obtain a time resolution of about 50 ps after the deconvolution procedure. The temporal emission profile was well fit with a single-exponential function. Nanosecond Laser Flash Photolysis System. Nanosecond laser flash photolysis was carried out using the second harmonic oscillation (532 nm) of a Nd3þ:YAG laser (Continuum, Surelite II-10; fwhm 5 ns). The emission from the sample was collected by a focusing lens and directed through a grating monochromator (Koken, SG-100) to a photomultiplier tube (PMT) (Hamamatsu Photonics, C9525) that was located in the chamber (Hamamatsu Photonics, C6544-20) cooled to -80 °C by utilizing liquid nitrogen Langmuir 2009, 25(24), 13888–13893

as the coolant medium. The signal from the PMT was amplified through the preamplifier (Stanford Research Systems, SR445A) and detected by a gated photon counter (Stanford Research Systems, SR400). Transmission Electron Microscopy (TEM). For the TEM studies, a PVAc film containing a Au cluster was dissolved in acetone and cast on a carbon-coated TEM grid. TEM images were obtained by using a Hitachi H-9000 TEM equipped with a tilting device and operated at 300 kV. The images were recorded under axial illumination by using a CCD camera (model XR-100 by AMT). Experimental Setup for Single-Cluster Microscopy. The experimental setup for single-cluster microscopy was based on an Olympus IX71 inverted fluorescence microscope. Continuous wave (CW) light, emitted from a 405 nm diode laser (Olympus, LD405, 30 mW) or a 532 nm laser (Photop Suwtech, DPGL2050F, 50 mW) that passed through an objective lens (Olympus, UPlanSApo, 1.40 NA, 100) after reflection by a dichroic mirror was used to excite the radical precursor and Aun or Aum, respectively. (For 405 nm CW laser excitation, we used an Olympus DM455 and an Omega Optical 445DRLP for imaging and spectral measurements, respectively. For 532 nm CW laser excitation, we used an Olympus DM570 for imaging and spectral measurements, respectively.) The emission from single clusters on the cover glass was collected by using the same objective and magnified by the built-in 1.6 magnification changer. (Therefore, the net magnification was 160.) Then, it was passed through an DOI: 10.1021/la901552f

13889

Article

Sakamoto et al.

emission filter to remove the undesired scattered light and imaged by an electron-multiplying charge-coupled device (EM-CCD) camera (Roper Scientific, Cascade II:512). For 405 nm CW laser excitation, we used an Olympus BA475 and an Omega Optical 435ALP for imaging and spectral measurements, respectively. For 532 nm CW laser excitation, we used an Olympus BA575IF for imaging and spectral measurements, respectively. The images were processed by using ImageJ software (http://rsb.info.nih.gov/ ij/). A background threshold is modified by the shot noise in each image above the background. Counts above the threshold were then considered to be the fluorescence signal. For spectroscopy, only the emission that passed through a slit entered the imaging spectrograph (Acton Research, SP-2356) that was equipped with an EM-CCD camera (Princeton Instruments, PhotonMAX:512B). The width of the slit was 100 μm, which corresponded to 0.6 μm at the specimen because the images at the slit were magnified by 160. The spectra were typically integrated for 1 s. The spectrum detected by the EM-CCD camera was stored and analyzed by using a personal computer. All of the experimental data were obtained at room temperature. The oxygen concentration was measured by an oxygen analyzer, MaxO2þ (Maxtec).

Result and Discussion Formation and Growth of Au Clusters in the Polymer Film. A poly(vinyl acetate) (PVAc) film containing a radical precursor (2-hydroxy-40 -(2-hydroxyethoxy)-2-methylpropiophenone) and HAuCl4 (denoted as PVAc) was prepared. Upon photoexcitation, the precursor yields radicals via a Norrish-type-I R-cleavage. The radicals work as reducing agents for the Au ions to generate Au clusters (Scheme 1).12,14 Figure 1A shows the absorption spectra of PVAc before and after UV light irradiation. After UV light irradiation (3 mW/cm2) for 30 min, a broad absorption band peak appeared at 420 nm. The film exhibited a fluorescence peak at 480 nm that originated from the formation of the emissive Au clusters mixture (Aun, n= atom number) (Figure 1B-D).12 When the UV-irradiated film was stored in the dark for a sufficient duration (aging), a new peak appeared at 542 nm, and the color turned to vivid pink.15 The time-resolved absorption and fluorescence spectra of the PVAc film are shown in Figure 1A,B, respectively. Following the absorption spectral change, the emission peak at 480 nm decreased, and a new emission appeared at the longer wavelength. The newly observed emission showed the excitation and emission spectra maxima at 550 and 583 nm, respectively (Figure 1C,D). The transmission electron microscopy (TEM) image revealed that the average diameter of the Au clusters increased from 1.0 to 1.5 nm during aging (Figure 2).12 This fact indicates that the changes in the absorption and fluorescence spectra of the PVAc film were caused by the formation of new Au clusters with larger numbers of atoms (Aum (m>n)). Irradiation with a UV lamp for a longer period of time or with a stronger power of light did not result in the formation of Aum. Only an adequate aging duration allowed the formation of Aum. SMS images of single Aun and Aum (m > n) are shown in Figure 3A,B. Aun and Aum can be excited selectively by employing (15) The shape of the excitation spectrum of Aum is similar to that of the absorption spectrum. In addition, the diameter of clusters is less than 3 nm, which does not show the surface plasmon band.16,17 Thus, it is concluded that the absorption band of aged PVAc film does not originated from the surface plasmon band of the larger gold nanoparticles (>3 nm). (16) Shimizu, T.; Teranishi, T.; Hasegawa, S.; Miyake, M. J. Phys. Chem. B 2003, 107, 2719. (17) Wyrwas, R. B.; Alvarez, M. M.; Khoury, J. T.; Price, R. C.; Schaaff, T. G.; Whetten, R. L. Euro. Phys. J. D 2007, 43, 91.

13890 DOI: 10.1021/la901552f

Figure 2. (A) TEM images of Aun and Aum (m > n). (B) Size distribution of Aun (blue bar) and Aum (red bar).

the appropriate combination of sample and light source. The sample of Aun was prepared by spin coating (3000 rpm, 40 s) the acetonitrile solution containing PVAc, precursor, and HAuCl4 onto the cleaned cover glass (to a thickness of approximately 50 μm). The 405 nm CW laser generated Aun via a photochemical reaction and concurrently excited the formed Aun.12 In contrast, the sample of Aum was prepared by spin coating (3000 rpm, 40 s) an acetone solution of the aged PVAc film containing Aum. Because the absorption and emission spectra of Aum did not change with the dissolution of the PVAc film in acetone, it is believed that the distribution of Aum does not significantly change during sample preparation (Supporting Information). The use of a 532 nm CW laser enabled the selective excitation of Aum in the sample. Although the bulk fluorescence spectrum is broad, single Aum exhibited a narrower spectrum (Figure 4A). The fluorescence spectra of single Au clusters revealed that the bulk spectrum of Aum (m>n) was mainly composed of the two spectral peaks at 1.97 and 2.04 eV, which may correspond to Au21 and Au19, respectively.18 In our previous paper, the fluorescence spectra of single Aun was assigned to Au