Plasmon-Enhanced Photocurrent Generation from Self-Assembled

Feb 12, 2009 - Department of Materials Physics and Chemistry, Graduate School of Engineering, Kyushu UniVersity,. 744 Moto-oka, Nishi-ku, Fukuoka ...
0 downloads 0 Views 589KB Size
Download PDF
Langmuir 2009, 25, 3887-3893

3887

Plasmon-Enhanced Photocurrent Generation from Self-Assembled Monolayers of Phthalocyanine by Using Gold Nanoparticle Films Kosuke Sugawa,† Tsuyoshi Akiyama,†,‡ Hirofumi Kawazumi,§ and Sunao Yamada*,†,‡,| Department of Materials Physics and Chemistry, Graduate School of Engineering, Kyushu UniVersity, 744 Moto-oka, Nishi-ku, Fukuoka 819-0395, Japan, Department of Applied Chemistry, Faculty of Engineering, Kyushu UniVersity, 744 Moto-oka, Nishi-ku, Fukuoka 819-0395, Japan, Department of Biological and EnVironmental Chemistry, Kinki UniVersity Kyushu, 11-6 Kayanomori, Iizuka, 820-8555, Japan, and Center for Future Chemistry, Kyushu UniVersity, 744 Moto-oka, Nishi-ku, Fukuoka 819-0395, Japan ReceiVed NoVember 19, 2008. ReVised Manuscript ReceiVed January 12, 2009 The effect of localized electric fields on the photocurrent responses of phthalocyanine that was self-assembled on a gold nanoparticle film was investigated by comparing the conventional and the total internal reflection (TIR) experimental systems. In the case of photocurrent measurements, self-assembled monolayers (SAMs) of a thiol derivative of palladium phthalocyanine (PdPc) were prepared on the surface of gold-nanoparticle film that was fixed on the surface of indium-tin-oxide (ITO) substrate via a polyion (PdPc/AuP/polyion/ITO) or on the ITO surface (PdPc/ITO). Photocurrent action spectra from the two samples were compared by using the conventional spectrometer, and were found that PdPc/AuP/polyion/ITO gave considerably larger photocurrent signals than PdPc/ITO under the identical concentration of PdPc. In the case of the TIR experiments for the PdPc/AuP/polyion/ITO and the PdPc/AuP/Glass systems, incident-angle profiles of photocurrent and emission signals were correlated with each other, and they were different from that of the PdPc/ITO system. Accordingly, it was demonstrated that the photocurrent signals were certainly enhanced by the localized electric fields of the gold-nanoparticle film.

1. Introduction Surface plasmons are collective oscillations of free electrons populated at the surfaces of metal films and nanoparticles, and lead to a generation of the local electric field at the metal/medium interface region by resonating with the light field, called surface plasmon resonance: SPR. It has been reported that the electric field on planar metal surfaces, associated with generating SPR at a resonant angle, leads to the enhancement of molecular excitation as has been verified with inducing the enhancement of fluorescence1-3 and photocurrent4-7 from the surface-anchored dye molecules under the attenuated total reflection (ATR) geometry. Especially, in the case of biosensing using an SPR sensor, the fluorescence enhancement has been well studied for high-sensitivity measurements of small molecules as surface-plasmon field-enhanced fluorescence spectroscopy (SPFS).8-20 * Author for correspondence. E-mail: [email protected]. † Department of Materials Physics and Chemistry, Graduate School of Engineering, Kyushu University. ‡ Department of Applied Chemistry, Faculty of Engineering, Kyushu University. § Department of Biological and Environmental Chemistry, Kinki University Kyushu. | Center for Future Chemistry, Kyushu University. (1) Ishida, A.; Sakata, Y.; Majima, T. Chem. Commun. 1998, 57–58. (2) Ishida, A.; Majima, T. Chem. Commun. 1999, 1299–1300. (3) Ishida, A.; Majima, T. Analyst 2000, 125, 535–540. (4) Ishida, A.; Sakata, Y.; Majima, T. Chem. Lett. 1998, 27, 267–268. (5) Ishida, A.; Majima, T. Chem. Phys. Lett. 2000, 242–246. (6) Fukuda, N.; Mitsuishi, M.; Aoki, A.; Miyashita, T. Chem. Lett. 2001, 30, 378–379. (7) Fukuda, N.; Mitsuishi, M.; Aoki, A.; Miyashita, T. J. Phys. Chem. B 2002, 106, 7048–7052. (8) Liebermann, T.; Knoll, W.; Sluka, P.; Herrmann, R. Colloid Surf., A 2000, 169, 337–350. (9) Liebermann, T.; Knoll, W. Colloid Surf., A 2000, 171, 115–130. (10) Neumann, T.; Johansson, M-L.; Kambhampati, D.; Knoll, W. AdV. Funct. Mater. 2002, 12, 575–586. (11) Yu, F.; Yao, D.; Knoll, W. Anal. Chem. 2003, 75, 2610–2617.

On the other hand, localized surface plasmon resonance (LSPR), occurring in metal nanoparticles or clusters, has opened up the application in the highly sensitive spectroscopy and imaging based on surface-enhanced Raman scattering (SERS)21-24 and fluorescence enhancement,25-29 because LSPR generates highly confined electric fields.30 In addition, a noteworthy advantage of LSPR is that it is not necessary to use the complicated prism coupler for the ATR geometry and to adjust the incident angle of irradiation light for every excitation wavelength, because LSPR occurs simply by direct light irradiation of metal nanoparticles (12) Gryczynski, I.; Malicka, J.; Gryczynski, Z.; Lakowicz, J. R. Anal. Biochem. 2004, 324, 170–182. (13) Gryczynski, I.; Malicka, J.; Gryczynski, Z.; Lakowicz, J. R. J. Phys. Chem. 2004, 108, 12568–12574. (14) Ekgasit, S.; Thammacharoen, C.; Yu, F.; Knoll, W. Anal. Chem. 2004, 76, 2210–2219. (15) Ekgasit, S.; Yu, F.; Knoll, W. Langmuir 2005, 21, 4077–4082. (16) Ekgasit, S.; Yu, F.; Knoll, W. Sens. Actuators, B 2005, 104, 294–301. (17) Lo¨ssner, D.; Kessler, H.; Thumshirn, G.; Dahmen, C.; Wiltschi, B.; Tanaka, M.; Knoll, W.; Sinner, E-K.; Reuning, U. Anal. Chem. 2006, 78, 4524–4533. (18) Morigaki, K.; Tawa, K. Biophys. J. 2006, 91, 1380–1387. (19) Kasry, A.; Knoll, W. Appl. Phys. Lett. 2006, 89, 101106. (20) Kwon, S. H.; Hong, B. J.; Park, H. Y.; Knoll, W.; Park, J. W. J. Colloid Interface Sci. 2007, 308, 325–331. (21) Wang, H.; Levin, C. S.; Halas, N. J. J. Am. Chem. Soc. 2005, 127, 14992– 14993. (22) Suzuki, M.; Niidome, Y.; Kuwahara, Y.; Terasaki, N.; Inoue, K.; Yamada, S. J. Phys. Chem. B 2004, 108, 11660–11665. (23) Bozzini, B.; D’Urzo, L.; Mele, C.; Romanello, V. J. Phys. Chem. C 2008, 112, 6352–6358. (24) Yu, Q.; Guan, P.; Qin, D.; Golden, G.; Wallace, P. M. Nano Lett. 2008, 8, 1923–1928. (25) Kulakovich, O.; Strekal, N.; Yaroshevich, A.; Maskevich, S.; Gaponenko, S.; Nabiev, I.; Woggon, U.; Artemyev, M. Nano Lett. 2002, 2, 1449–1452. (26) Aslan, K.; Huang, J.; Geddes, D. J. Am. Chem. Soc. 2006, 128, 4206– 4207. (27) Ray, K.; Badugu, R.; Lakowics, J. R. J. Am. Chem. Soc. 2006, 128, 8998–8999. (28) Aslan, K.; Wu, M.; Lakowics, J. R.; Geddes, C. D. J. Am. Chem. Soc. 2007, 129, 1524–1525. (29) Tam, F.; Goodrich, G. P.; Johnson, B. R.; Halas, N. J. Nano Lett. 2007, 7, 496–501. (30) Lakowicz, J. R. Anal. Biochem. 2001, 298, 1–24.

10.1021/la803831c CCC: $40.75  2009 American Chemical Society Published on Web 02/12/2009

3888 Langmuir, Vol. 25, No. 6, 2009

independent of the incident angle. Furthermore, the metal nanoparticles exhibit interesting properties which act as electron mediators31 and charge recombination centers.32 Therefore, LSPR-responsive metal nanoparticles are attractive for the development of photoelectrochemical devices. Some researchers have reported that the metal nanoparticles placed on the surface of the inorganic Si photodiode increased optical absorption33 and photocurrent response.33-37 It has also been reported that SPR, derived from the metal nanocluster and grating, increased the absorption and the photocurrent in the organic photovoltaic cell.38-42 However, detailed studies on the mechanisms of the enhancement phenomena have not be done. Recently, attention has been particularly focused on the localized electric fields derived from the coupling of the singleparticle LSPR, as can be seen in the particle dimers and aggregates. It has also been demonstrated that those fields are usually larger as compared with that of an isolated particle from both theoretical43-47 and experimental investigations in SERS44 and fluorescence enhancements.46 Therefore, the fabrication of the photoelectrochemical devices consisting of aggregated metal nanoparticles is promising for the development of highperformance devices. Furthermore, regarding photoelectric conversion in organic solar cells, it has been well recognized that the increase in the photoelectric conversion efficiency in the longer-wavelength region is one of the important goals to improve the performance of the cells.48-52 Since the film of aggregated gold nanoparticles (AuPs) exhibits the plasmon band in the longer-wavelength region (from 500 nm reaching as long as more than 1000 nm)45 the use of the film of aggregated AuPs as the electrode material is quite attractive, intended for enhancing the photocurrent generation in the far-red to near-infrared region. Previously, we fabricated a photofunctional electrode by selfassembling the porphyrin monolayer on the surface of an LSPRresponsive film of AuP aggregates, which was confined on an (31) Jena, B. K.; Raj, C. R. Anal. Chem. 2006, 78, 6332–6339. (32) Peumans, P.; Yakimov, A.; Forrest, S. R. J. Appl. Phys. 2003, 93, 3693– 3723. (33) Pillai, S.; Cathchpole, K. R.; Trupke, T.; Green, M. A. J. Appl. Phys. 2007, 101, 093105. (34) Derkacs, D.; Lim, S. H.; Matheu, W. Mar; Yu, E. T. Appl. Phys. Lett. 2006, 89, 093103. (35) Sundararajan, S. P.; Grady, N. K.; Mirin, N.; Halas, N. J. Nano. Lett. 2008, 8, 624–630. (36) Schaadt, D. M.; Feng, B.; Yu, E. T. Appl. Phys. Lett. 2005, 86, 063106. (37) Lim, S. H.; Mar, W.; Matheau, P.; Derkacs, D.; Yu, E. T. J. Appl. Phys. 2007, 101, 104309. (38) Tvingstedt, K.; Persson, N-K.; Ingana¨s, O. Appl. Phys. Lett. 2007, 91, 113514. (39) Yakimov, A.; Forrest, S. R. Appl. Phys. Lett. 2002, 80, 1667. (40) Rand, B. P.; Peumans, P.; Forrest, S. R. J. Appl. Phys. 2004, 96, 7519. (41) Stenzel, O.; Stendal, A.; Voigtsberger, K.; Vonborczyskowski, C. Sol. Energy Mater. Sol. Cells 1995, 37, 337–348. (42) Westphalen, M.; Kreibig, U.; Rostalski, J.; Luth, H.; Meissner, D. Sol. Energy Mater. Sol. Cells 2000, 61, 97–105. (43) Gersten, J. I.; Nitzan, A. Surf. Sci. 1985, 158, 165–189. (44) Nie, S.; Emory, R. Science 1997, 275, 1102–1106. (45) Xu, H.; Aizpurua, J.; Ka¨ll, M.; Apell, P. Phys. ReV. E 2000, 62, 4318– 4324. (46) Bek, A.; Jansen, R.; Ringler, M.; Mayilo, S.; Klar, T. A.; Feldmann, J. Nano Lett. 2008, 8, 485–490. (47) Futamata, M.; Maruyama, Y.; Ishikawa, M. J. Phys. Chem. B 2003, 107, 7607–7617. (48) Hagfeldt, A.; Gra¨tzel, M. Chem. ReV. 1995, 95, 49–68. (49) Kim, J. Y.; Lee, K.; Coates, N. E.; Moses, D.; Nguyen, T-Q.; Dante, M.; Heeger, A. J. Science 2007, 317, 222–225. (50) Muhlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Bradec, C. AdV. Mater. 2006, 18, 2884–2889. (51) Soci, C.; Hwang, I-W.; Moses, D.; Zhu, Z.; Waller, D.; Gaudiana, R.; Bradec, C. J.; Heeger, A. J. AdV. Funct. Mater. 2007, 17, 632–636. (52) Bidault, S.; Abajo, F. J. G.; Polman, A. J. Am. Chem. Soc. 2008, 130, 2750–2751.

Sugawa et al.

indium-tin oxide (ITO) electrode.53 Preliminary investigations indicated that this electrode showed an appreciable enhancement of photocurrent ascribed to porphyrin excitation in the wavelength region of the plasmon band of the film. The results suggested that the enhancement was derived from the increase of porphyrin molecule excitation assisted by the localized electric fields of the AuP film as well as from the increment of the immobilization of porphyrin. In the previous study, however, we measured the photocurrent generation from the porphyrin immobilized on the surface of AuP film by the direct irradiation geometry. Therefore, the resultant photocurrents should be the result from the direct photoexcitation as well as the enhancement of electronic excitation by the localized electric fields. From these viewpoints, the purpose of the present study is to clarify the effects of localized electric fields on the photocurrent responses. Thus, we have studied the photocurrent generation through the LSPR excitation of the AuP film under the direct irradiation geometry and the total internal reflection (TIR) geometry to generate the evanescent field, and by using phthalocyanine (Pc) having strong absorption bands in the farred region where they well overlapped with the wavelength region of LSPR. In order to evaluate the effects of localized electric fields, emission and photocurrent signals from the immobilized Pc on the surface of the AuP film were measured. Furthermore, the resultant photocurrent signals were compared with the case of excitation by the evanescent field alone as a reference.

2. Experimental Section Materials. The colloidal aqueous solution of AuPs capped with citric ion was prepared by the reduction of HAuCl4 with sodium citrate acid as described previously,54 and the mean diameter evaluated from transmission electron micrograph (TEM) image was 18 ( 1 nm. A thiol derivative of palladium phthalocyanine (PdPc, the chemical structure as shown in Figure 1) as a photoresponsive dye molecule was supplied from Mitsui Chemicals, Inc.55 Poly(styrenesulfonate) (PSS, Mw ) 70000), poly(ethyleneimine) (PEI, Mw ) 50000-100000), and other chemicals were used as received. An ITO electrode (thickness of ITO: ∼30 nm) was obtained from Sanyo Vacuum Industries Co., Ltd. Preparation of Au Planar Film on Glass Plate. The glass plate (18 mm × 32 mm × 0.12-0.17 mm) was sonicated with acetone for 20 min and then treated in an NH3 aq (28%)/H2O2 aq (30%) (1/1 v/v) mixed solution at 100 °C, followed by washing with enough water and drying in air. Then a planar gold film on the glass (denoted as Au/Glass) was prepared by vacuum deposition of titanium (3 nm) followed by gold (40 nm). The thickness of the deposited film was measured with the quartz crystal microbalance method. Preparation of PdPc-Modified AuP Film on Glass Plate. The preparation procedure for the PdPc-modified AuP film, denoted as PdPc/AuP/Glass, is shown in Figure 1a. The glass plate (18 mm × 32 mm × 0.12-0.17 mm) was sonicated with acetone for 20 min and then treated in an NH3 aq (28%)/H2O2 aq (30%) (1/1 v/v) mixed solution at 100 °C, followed by washing with enough water. The preparation procedure for the AuP film at the liquid/liquid interface has been reported previously,56,57 namely the liquid/liquid interface of the aqueous colloidal gold solution (30 mL) and hexane (10 mL) was formed in a beaker (50 mL), where the glass plate was placed at its bottom. Then, methanol (15 mL) was vigorously injected into the colloidal solution, resulting in instantaneous color change of the (53) Akiyama, T.; Nakada, M.; Terasaki, N.; Yamada, S. Chem. Commun. 2006, 4, 395–397. (54) Turkevich, J.; Stevenson, P. C.; Hiller, J. Discuss. Faraday Soc. 1951, 11, 55–75. (55) Yamada, S.; Akiyama, T.; Inoue, S.; Misawa, T. Jpn. Pat. 2003/123863, 2003. (56) Suzuki, M.; Niidome, N.; Terasaki, N.; Kuwahara, Y.; Yamada, S. Jpn. J.Appl. Phys 2004, 43, L554–L556. (57) Akiyama, T.; Kawahara, T.; Arakawa, T.; Yamada, S. Jpn. J. Appl. Phys 2008, 47, 3063–3066.

Photocurrent Generation through the LSPR Excitation

Langmuir, Vol. 25, No. 6, 2009 3889

Figure 1. Preparation procedures of (a) PdPc/AuP/Glass, (b) PdPc/AuP/polyion/ITO, and (c) PdPc/ITO.

Figure 2. Schematic diagram for SPR and fluorescence measurements for TIR configuration.

aqueous phase from wine red to light pink and the formation of the liquidlike film of AuPs at the interface. The resultant liquidlike film was transferred onto the surface of the glass plate by lifting it up normal to the interface, followed by drying in air, to give the AuP film on the plate, denoted as AuP/Glass. The self-assembled monolayer of PdPc on AuP/Glass was prepared by immersing the substrate into a CH2Cl2 solution containing 0.1 mM PdPc for 36 h, followed by rinsing with CH2Cl2 to obtain the sample of PdPc/ AuP/Glass. Preparation of PdPc-Modified AuP Film on ITO Electrode. The AuP film was also transferred onto the surface of an ITO electrode. In this case, however, the ITO surface was modified with some precursor polyion films in order to prevent the AuP film from peeling off during the photoelectrochemical measurements; the preparation procedure is shown in Figure 1b. The ITO electrode (20 mm × 25 mm × 0.7 mm) was first sonicated in acetone for 20 min and then washed with enough acetone. This treatment was repeated twice. Then the ITO electrode was cleaned by ozone etching for 15 min, followed by washing with enough water and then drying with N2 gas. Next, the ITO electrode was immersed into an aqueous solution of PEI (1.5 mM) for 20 min, an aqueous solution of PSS (3 mM) for 20 min, and again the above-described PEI solution for 20 min with intermediate water rinsing to generate cationic charges on the ITO surface. Then, the liquidlike AuP film formed at the liquid/liquid interface was transferred onto the surface of the positively

charged ITO electrode. The PEI/PSS/PEI precursor film satisfactorily improved the physical stability of the AuP film. Finally the selfassembled monolayer of PdPc was prepared in a manner similar to that for PdPc/AuP/Glass, denoted as: PdPc/AuP/polyion/ITO (see Figure 1b). As a reference, the self-assembled monolayer of PdPc was also prepared on the bare ITO electrode, without polyion precursor films.58 The ITO electrode was immersed into the CH2Cl2 solution of PdPc (0.1 mM) for nine days, followed by washing with CH2Cl2, giving the monolayer of PdPc on ITO, denoted as PdPc/ ITO(see Figure 1c). Measurements. Absorption spectra of the samples were measured with a JASCO V-670 UV-vis near IR (NIR) spectrophotometer attached with an integrating sphere. Scanning electron microscope (SEM) observations were carried out using a HITACHI S-5000 microscope. The schematic diagram of emission (front side from the prism) and reflectivity (rear side) measurements under TIR geometry is shown in Figure 2. The rear surface of the glass plate of PdPc/ AuP/Glass was joined on the bottom face of a BK-7 semi-cylindrical prism via an index-matching oil (see Figure 2). A linearly p-polarized tungsten-halogen light (for reflectivity measurement, power: 10 mW, LS-1, Ocean Optics Inc.) or the He-Ne laser (for photoluminescence measurement, λ ) 632.8 nm, power: 4 mW) was used as an incident light through a polarizer. The reflected light from the (58) Obeng, Y. S.; Bard, A. J. Langmuir 1991, 7, 195–201.

3890 Langmuir, Vol. 25, No. 6, 2009

Figure 3. Schematic diagram for photocurrent measurement system under TIR configuration.

sample was collected with a lens and an optical fiber, and then the resultant signal was detected by a multichannel spectroscope. On the other hand, the photoluminescence signal from the sample was collected with a couple of lenses and an optical fiber and was detected with a photomultiplier through a monochromator (f ) 10 cm). Then the signal was amplified with an amplifier (NF P-61), and the output signal was recorded with a recorder. The schematic diagram for photocurrent measurements is shown in Figure 3. The p-polarized He-Ne laser light was used as an excitation light source. The sample electrode, PdPc/AuP/polyion/ ITO or PdPc/ITO, was joined on the bottom face of the prism in the same manner as shown in Figure 2. Photocurent measurements were carried out in an aqueous solution containing NaClO4 (0.1 mM), using a homemade three-electrode photoelectrochemical cell with PdPc/AuP/polyion/ITO or PdPc/ITO as a working electrode, Ag/ AgCl (sat. KCl) as a reference electrode, and a platinum wire as a counterelectrode respectively. The volume of the cell was approximately 10 mL. In all measurements, methyl viologen (5.0 mM) was added to the electrolyte solution as an electron acceptor. The steady-state photocurrents were detected using a Fuso HECS-318C potentiostat.

3. Results and Discussion Morphology Analysis and Optical Properties of AuP Film. SEM photographs of AuP/Glass and AuP/polyion/ITO were measured to investigate the morphologies of the AuP films of the substrates, as shown in panels a and b of Figure 4. Both AuP films had nearly two-dimensional and densely packed structures59 (Supporting Information, Figure S1) with the coverages of 70% for AuP/Glass and 69% for AuP/polyion/ITO, respectively. As shown in panels c and d of Figure 4, absorption spectra of both samples exhibited clear plasmon bands around 700 nm, though AuP/Glass were broader than AuP/polyion/ITO. In both cases, the plasmon bands showed a considerable red-shift as compared with the case in solution (∼520 nm). The peak position of AuP/ polyion/ITO showed slight blue-shift as compared with AuP/ Glass. This may due to a lower refractive index (1.45) of the polyion layer as compared with that of the glass substrate (1.53). In addition, the interaction of AuPs with polyion generally reduced broadening of the interparticle plasmon band of the AuP film, although the detailed reason is not clear at this stage. Optical Characterization of PdPc Modified on AuP Film and ITO. Figure 5 shows the absorption spectra of PdPc/AuP/ polyion/ITO and AuP/polyion/ITO, together with those of the aqueous colloidal gold solution and the dichloromethane solution of PdPc. As to AuP/polyion/ITO (Figure 5a), a broadband ascribed to the plasmon oscillation of interparticle plasmon coupling is (59) Sugawa, K.; Kawahara, T.; Akiyama, T.; Yamada, S. Jpn. J. Appl. Phys 2009, xxx. In press.

Sugawa et al.

observed at 670 nm. This band shifts to the longer-wavelength region as compared with the isolated plasmon oscillation mode appearing in the colloidal solution (∼520 nm; Figure 5d).60,61 In the case of PdPc/AuP/polyion/ITO, broad bands around 750 and 670 nm are observed (Figure 5b). The band around 750 nm almost disappeared when PdPc/AuP/polyion/ITO was immersed in dichloromethane solution containing an excess of dodecanethiol (DT), due to the replacement of the immobilized PdPc with DT as shown in Figure 5c, where the substrate after replacement with DT is denoted as DT/AuP/polyion/ITO. A slight peak-shift of plasmon band in DT/AuP/polyion/ITO may be due to some environmental and/or morphological changes by the replacement treatment. Thus, this band is ascribed to the Q-band of PdPc, and the band at 670 nm is ascribed to the plasmon oscillation. Because the absorption band of PdPc well overlaps with the plasmon band of AuP film, the enhanced excitation of PdPc is expected. The absorption spectrum of PdPc/ITO revealed the immobilization of PdPc on ITO, because the Q-band of PdPc was clearly observed (Supporting Information, Figure S2). In addition, its peak position showed no appreciable shift as compared with that in dichloromethane (Supporting Information, Figure S2). These results indicate no appreciable aggregation of PdPc in PdPc/ITO. Photocurrent Action Spectra of PdPc Modified on AuP film and ITO. In the case of the TIR geometry, the photocurrent measurements were successful only by using the He-Ne laser (beam diameter, 1 mm: Figure 3). It was practically impossible in our experimental system to obtain the action spectrum in the TIR geometry, because the diameter of the optical spot of the white light was 4 mm at the normal position and the photon density changed dramatically with different incident angles. Thus, the photocurrent action spectrum was obtained by direct irradiation of PdPc/AuP/polyion/ITO or PdPc/ITO based on the front-side geometry (incident angle: 0 °) by using the standard threeelectrode photoelectrochemical cell (irradiation area: 0.28 cm2). The results are shown in Figure 6. For each point, the signal was obtained by averaging the values measured for three times under the identical condition. Both samples showed stable photocurrent responses in the cathodic direction and the action spectra were well correlated with the absorption spectrum of PdPc in CH2Cl2 (see Figure 5e). The number of PdPc immobilized on each electrode was estimated as follows. First, each electrode was dipped into 10 mL of the dichloromethane solution containing DT (300 mM) for 7 days and then was sonicated in the solution. After withdrawing the electrode from the solution, it was washed with dichloromethane; it was then collected into the above DTcontaining solution. After these treatments, the absorption peaks of PdPc on each electrode almost completely disappeared (see Figure 5c). While the combined dichloromethane solution containing dissolved PdPc was concentrated to 10 mL, and its absorption spectrum was measured to evaluate the ratio of immobilized PdPc between the two electrodes. As a result, the ratio of immobilized PdPc for PdPc/AuP/polyion/ITO to PdPc/ ITO was 7:1. In contrast, the average ratio of photocurrents for PdPc/AuP/polyion/ITO to PdPc/ITO in the 600-750 nm region, where the plasmon band was clearly seen, was 14. Previously, Imahori et al.62 reported photocurrent generation quenched from an organic dye that was self-assembled on the gold surface. Nevertheless, considerable enhancement of photocurrents was (60) Schmitt, J.; Decher, G.; Dressick, W. J.; Brandow, S. L.; Geer, R. E.; Shashidhar, R.; Calvert, J. M. AdV. Mater. 1997, 9, 61–65. (61) Su, K.-H.; Wei, Q.-H.; Zhang, X.; Mock, J. J.; Smith, D. R.; Schultz, S. Nano Lett. 2003, 3, 1087–1090. (62) Imahori, H.; Norieda, H.; Ozawa, S.; Ushida, K.; Yamada, H.; Azuma, T.; Tamaki, K.; Sakata, Y. Langmuir 1998, 14, 5335–5338.

Photocurrent Generation through the LSPR Excitation

Langmuir, Vol. 25, No. 6, 2009 3891

Figure 4. SEM photographs of (a) AuP/Glass and (b) AuP/polyion/ITO; absorption spectra of (c) AuP/Glass and (d) AuP/polyion/ITO.

Figure 5. Absorption spectra of AuP/polyion/ITO (a: ___), PdPc/AuP/ polyion/ITO (b: ---), DT/AuP/polyion/ITO (c: · · · ), colloidal aqueous solution of gold nanoparticles (d: - · - · ) and PdPc in dichloromethane (e: - · · - · · ).

Figure 7. Reflection spectra of (a) AuP/Glass and (b) Au/Glass under TIR geometry.

Figure 6. Photocurrent action spectra of PdPc/AuP/polyion/ITO (b) and PdPc/ITO (2).

achieved in PdPc/AuP/polyion/ITO in this study. These results suggest that the photocurrent signal of PdPc was certainly

enhanced in the presence of AuP film, probably because the excitation efficiency of immobilized PdPc is increased by the enhanced electric field caused by surface plasmon. Reflectance Spectra of AuP Film and Gold Film under TIR Geometry. Figure 7a shows incident angle dependences of the reflection spectra for the AuP/Glass under the TIR geometry. In each spectrum, when the incident angle was above the critical angle (42 °), the reflected signal showed a dip around 650 nm

3892 Langmuir, Vol. 25, No. 6, 2009

Sugawa et al.

Figure 9. Emission spectra obtained from a conventional fluorescence spectrophotometer (λex ) 633 nm) (AuP/Glass: - - -, PdPc/AuP/ Glass: ___) and from the present experimental system shown in Figure 2 (AuP/Glass: O, PdPc/AuP/Glass: 0).

Figure 8. Incident angle dependence of reflection intensities (b) and emission properties of (a) PdPc/AuP/Glass (0) and (b) AuP/Glass (O) excited at 632.8 nm.

and dipped lower with increases in the incident angle. This feature is characteristic of AuPs63 but not for the Au/Glass where a dip is shifted to shorter wavelength with increasing the incident angle (Figure 7b).64 These results strongly suggest that LSPR of AuPs occurs due to the interaction between the evanescent field around 650 nm and the surface plasmon of the AuP film. Fluorescence Measurements of PdPc Modified on AuP Film under TIR Geometry. Next, we tried to clarify that AuP/Glass exhibits enhancement of the localized electric field due to LSPR under the TIR geometry. Recently, a number of studies about the fluorescence enhancement derived from the increment of excitation and emission efficiencies of molecules by the localized electric field of LSPR have been reported.25-29,65-67 The fluorescence from a copper phthalocyanine film was also observed by using a tip-enhanced plasmon technique.68 In this study, the emission from PdPc of PdPc/AuP/Glass was investigated. Because the absolute emission signal was very small and the incident angles were changed in the 41-75° regions, we have used the glass plate instead of the ITO electrode in the fluorescence measurements to avoid possible dispersion of the light beam by the ITO layer. As shown in Figure 7, the reflection intensity dependence on the incident angle shows a SPR curve feature in the AuP film. Figure 8a shows the angular dependence of emission from the PdPc/AuP/Glass at 725 nm where the incident light of 632.8 nm is used under TIR geometry. Considerably larger emission signals were observed for PdPc/AuP/Glass as compared (63) Brown, K. R.; Lyon, L. A.; Fox, A. P.; Reiss, B. D.; Natan, M. J. Chem. Mater. 2000, 12, 314–323. (64) Frutos, A. G.; Weibel, S. C.; Corn, R. M. Anal. Chem. 1999, 71, 3935– 3940. (65) Tovmachenko, O. G.; Graf, C.; Heuvel, D. J.; Blaaderen, A.; Gerritsen, H. C. AdV. Mater. 2006, 18, 91–95. (66) Chen, Y.; Munechika, K.; Ginger, D. S. Nano Lett. 2007, 7, 690–696. (67) Nooney, R. I.; Stranik, O.; McDonagh, C.; MacCraith, B. Langmuir 2008, 24, 11261–11267. (68) Uemura, T.; Furumoto, M.; Nakano, T.; Akai-Kasaya, M.; Saito, A.; Aono, M.; Kuwahara, Y. Chem. Phys. Lett. 2007, 448, 232–236.

to those for AuP/Glass (Figure 8a) above the critical angle. Weak emission signals were observed from the AuP/Glass and were roughly constant above the critical angle, probably due to scattering of stray light (Figure 8b). These strongly suggest that the emission signal for PdPc/AuP/Glass mostly arises from the immobilized PdPc. In addition, the emission signal steeply increased above the critical angle (∼41 °) and was correlated with the reflectivity response. The excitation wavelength of 632.8 nm overlaps well with both the absorption wavelength of LSPR of AuPs and the absorption band of PdPc. Furthermore, the wavelength dependence of the emission signal from the PdPc/AuP/Glass was measured under the TIR geometry, as shown in Figure 9. It is clear that the emission spectrum is similar to the fluorescence spectrum of PdPc/AuP/Glass measured by the conventional fluorescence spectrophotometer (λex ) 633 nm). In addition, the emission signal from PdPc/AuP/Glass was considerably larger as compared with that from AuP/Glass in the 680-850 region. These results clearly show that the observed emission arose mainly from the fluorescence from the immobilized PdPc. The presence of AuPs certainly enhanced the fluorescence signal from the immobilized PdPc on AuP/Glass. Photocurrent Measurements of PdPc Modified on AuP Film under TIR Geometry. Finally, the effects of localized plasmon excitation on the photocurrent generation from PdPc/ AuP/polyion/ITO were investigated by using a three-electrode photoelectrochemical cell containing the electrolyte solution. The results are shown in Figure 10. The photocurrent was evidently observed from the monolayer-covered PdPc on AuP via the TIR excitation. In PdPc/ITO as the reference system, the photocurrent signal appeared near the critical angle and increased gradually with increasing the incident angle (61.8 °). It reached maximum at around the critical angle and then decreased slightly above the critical angle as shown in Figure 10b. This profile of the photocurrent signal can be explained as follows. First, we assumed the three phase of water/ITO(thickness: 30 nm)/prism and simulated the angular-dependence of reflectivity and electric field intensity at the water/ITO interface when the p-polarized light (λ ) 632.8 nm) was incident from the prism by using the Fresnel reflectivity calculation as shown in Figure 11.69 We considered the refractive index of ITO to be 1.9 + (69) The calculations closely follow those published in the article by Hansen, W. J. Opt. Soc. Am. 1968, 58, 380–390. Although the equations follow Hansen’s N-phase method, this particular calculation is written to perform only a fourphase calculation. http://unicorn.ps.uci.edu/calculations/fresnel/fcform.html.

Photocurrent Generation through the LSPR Excitation

Langmuir, Vol. 25, No. 6, 2009 3893

Figure 11. Calculated incident angle dependence of ITO/Prism: (a) reflectivity curve (___) and electric field; (b) Ex (---) and (c) Ez (- - -).

Figure 10. Incident angle dependence of reflection intensities (9) and following photocurrent properties (b) of (a) PdPc/AuP/polyion/ITO and (b) PdPc/ITO.

0.01i.70 The reflectivity showed the total reflection above the critical angle. On the other hand, the intensity of the electric field Ex was zero at every incident angle, and that of Ez increased gradually with increasing the incident angle below the critical angle and reached maximum at the critical angle. In addition, Ez decreased with increasing the incident angle above critical angle. These results show that the photocurrent profile for PdPc/ ITO (Figure 10b) was similar to the intensity profile of Ez. Therefore, we suggest that the photocurrent is induced by the evanescent field as the excitation light. On the contrary, the intensity dependence of the photocurrent on the incident angle of the excitation light from PdPc/AuP/ (70) Vicikauskas, V.; Bremer, J.; Hunderi, O.; Antanavicius, R.; Januskevicius, R. Thin Solid Films 2002, 411, 262–267.

polyion/ITO (Figure 10a) was roughly correlated with the corresponding emission response of PdPc (Figure 8a), although the critical angles were different between the photocurrent measurements (61° in water: Figure 10a) and the fluorescence measurements (42° in air: Figure 8a). The photocurrent is certainly ascribed from the electron-transfer from the electronically excited state of PdPc to MV2+. Thus, we successfully demonstrated the enhanced excitation of self-assembled organic dye molecules by LSPR of AuPs, leading enhancement of photocurrent signals.

Conclusion We have fabricated the photocurrent measurement system consisting of PdPc and the two-dimensional film of AuPs to realize the utilization of LSPR to generate photoelectric conversion. We have carefully studied the LSPR properties as functions of incident angle and wavelength and have demonstrated that the photocurrent signals of PdPc immobilized on the AuP film were certainly enhanced by localized electric fields even in the case of the TIR. As a result, the combination of the TIR geometry and the AuP film was useful for obtaining a larger photocurrent signal by making the most of incident light. Supporting Information Available: This material is available free of charge via the Internet at http://pubs.acs.org. LA803831C