Au bimetallic nanostructures and their application in surface-enhanced Raman scattering

Au bimetallic nanostructures and their application in surface-enhanced Raman scattering

Thin Solid Films 520 (2012) 2701–2707 Contents lists available at SciVerse ScienceDirect Thin Solid Films journal homepage: www.elsevier.com/locate/...

2MB Sizes 2 Downloads 52 Views

Thin Solid Films 520 (2012) 2701–2707

Contents lists available at SciVerse ScienceDirect

Thin Solid Films journal homepage: www.elsevier.com/locate/tsf

Preparation of dendritic Ag/Au bimetallic nanostructures and their application in surface-enhanced Raman scattering Zao Yi a, b, Shanjun Chen b, Yan Chen a, b, Jiangshan Luo b, Weidong Wu b, Yougen Yi a,⁎, Yongjian Tang b,⁎⁎ a b

College of Physical Science and Technology, Central South University, Changsha 410083, China Research Center of Laser Fusion, China Academy of Engineering Physics, Mianyang 621900, China

a r t i c l e

i n f o

Article history: Received 26 February 2011 Received in revised form 10 November 2011 Accepted 17 November 2011 Available online 1 December 2011 Keywords: Ag/Au bimetallic dendrites Surface enhanced Raman scattering Copper foil

a b s t r a c t Dendritic Ag/Au bimetallic nanostructures have been synthesized via a multi-stage galvanic replacement reaction of Ag dendrites in a chlorauric acid (HAuCl4) solution at room temperature. After five stages of replacement reaction, one obtains structures with protruding nanocubes; these will mature into many porous structures with a few Ag atoms that are left over dendrites. The morphological and compositional changes which evolved with reaction stages were analyzed by using scanning electron microscopy, transmission electron microscopy, UV–visible spectroscopy, selected area electron diffraction and energy-dispersive X-ray spectrometry. The replacement of Ag with Au was confirmed. A formation mechanism involving the original development of Ag dendrites into porous structures with the growth of Au nanocubes on this underlying structure as the number of reaction stages is proposed. This was confirmed by surface-enhanced Raman scattering (SERS). The dendritic Ag/Au bimetallic nanostructures could be used as efficient SERS active substrates. It was found that the SERS enhancement ability was dependent on the stage of galvanic replacement reaction. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Recently, growing interest in bimetallic nanostructures comprised of noble metals such as gold (Au), silver (Ag), platinum (Pt), and palladium (Pd) has been justified by their fascinating optical [1], electronic [2], and catalytic [3] properties, leading to a wide range of applications, including surface-enhanced Raman scattering(SERS) [4], biosensors [5], and catalysis [6]. A variety of approaches to preparing bimetallic nanomaterials has been investigated, including simultaneous chemical reduction of mixed metalions [7], photochemical method [8], and successive reduction of metal ions on the surface of sacrificial nanoparticles, also known as galvanic replacement reaction (GRR) [4]. Among them, as an effective and simple yet still versatile tool, GRR has been extensively employed to synthesize bimetallic nanostructures like hollow/Porous [9] and core/shell particles [10] in aqueous [11] or organic [12] media. However, their applicability is still limited because of the assembly of individual particles from solution being a major challenge [13]. Therefore, growing efforts have been aimed toward creation of already assembled larger bimetallic structures, like nanowires [14] or thin films [15]. Thus, there are few reports about the creation of bimetallic dendrite structures [16]. In recent years, metallic dendrites have aroused intensive interest because of their attractive shapes and large surface area, resulting in

⁎ Corresponding author. Tel.: + 86 0816 2480827; fax: + 86 0816 2480830. E-mail address: [email protected] (Y. Yi). 0040-6090/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2011.11.042

pronounced surface reaction activity [17,18]. In particular, noble metal dendrites are widely utilized as catalysts [19]. Bimetals often exhibit better catalytic activities than the corresponding monometallic counterparts [20]. Compared to one-stage reaction approach, multi-stage reaction is more favorable for compositional and modality control. In this report, dendritic Ag/Au bimetallic nanostructures have been synthesized via a multi-stage galvanic replacement reaction (MGRR) of Ag dendrites in a chlorauric acid (HAuCl4) solution at room temperature. This results in bimetallic Ag/Au nanostructures with an interesting morphology. After five stages of replacement reaction, one obtains structures with protruding nanocubes; these will mature into very porous structures with little Ag left over. A series of characterizing techniques, including scanning electron microscopy (SEM), transmission electron microscopy (TEM), UV–visible spectroscopy (UV–vis), selected area electron diffraction (SAED) and energydispersive X-ray spectrometry (EDX) spectrometry were applied to systematically explore the changes in morphology, and composition. The EDX, TEM, UV–vis and SAED confirmed the replacement of Ag with Au. It proposed formation mechanism that the original Ag dendrites developing pores while growing Au nanocubes covering this underlying structure at more reaction stages confirmed by exploiting SERS. By Rhodamine 6G (R6G) as the probe molecule, we were able to confirm the suggested formation mechanisms via the structures' SERS effects. The dendritic Ag/Au bimetallic nanostructures could be used as highly efficient SERS active substrates. It was found that the SERS enhancement ability was dependent on the reaction stage of dendritic Ag/Au bimetallic nanostructures. When the reaction was one stage, the materials revealed the supreme SERS signal.

2702

Z. Yi et al. / Thin Solid Films 520 (2012) 2701–2707

2. Experimental procedure 2.1. Materials and preparation of Ag/Au bimetallic nanostructures on Cu substrate Rhodamine 6G was purchased from Sigma-Aldrich, AgNO3 was purchased from Beijing Chemical Plant, and they were both used without any purification. Copper foils were purchased from Tianjing Guangfu Fine Chemical Research Institute. The thin copper foil (99.9%) was first treated by hydrochloric acid to remove surface contamination and was rinsed with distilled water. Silver dendrite structures were produced by immersing the copper foil into the 50 mL of 0.05 M aqueous AgNO3 solution for 5 min at room temperature (about 298 K), followed by rinsing with a large amount of distilled water (50 mL) for three times. Au nanostructures were produced by immersing the copper foil into the 50 mL of 0.05 M aqueous HAuCl4 solution for 5 min at room temperature (about 298 K), followed by rinsing with a large amount of distilled water (50 mL) for three times. The products were finally air dried and stored in air for further usage. In the replacement strategy, the as-prepared Ag dendrites on the Cu substrate were immersed in 50 mL of 1 mM aqueous HAuCl4 solution for 2 min, and followed by rinsing with a large amount of distilled water (50 mL) for three times. Each replacement reaction was named one stage, and as the stage went on, the color of the Ag dendrites gradually changed from black to light gray in the replacement case. All experiments were conducted at room temperature (23 °C). 2.2. Characterization SEM analysis was performed on a FESEM microscope (Sirion XL, FEI, Hillsboro, OR) operated at 5 kV. TEM analysis was conducted with a JEOL microscope (JEM-200CX) operated at 200 kV. All UV– vis-NIR spectra were recorded within a 1-cm optical length quartz cell on a Perkin-Elmer Lambda 12 spectrophotometer. The compositional analysis was carried out in a SEM equipped with EDX (Inca Energy model, Oxford) operated at an accelerating voltage of 10 kV. The accumulation time for EDX spectra was 100 s. The EDX was calibrated

with pure elements Cu and Al. Raman spectra were obtained with a Renishaw 2000 model confocal microscopy Raman spectrometer with a CCD detector and a holographic notch filter (Renishaw Ltd, Gloucestershire, U.K.). The microscope attachment was based on a Leica DMLM system, and a 100× objective was used to focus the laser beam onto a spot size around 1 μm 2. Radiation of 514.5 nm from an air cooled argon ion laser (Spectra-Physics Model 163C4260) was used for excitation. All of the spectra reported were the results of a single 20-s accumulation. 3. Results and discussion Fig. 1a gives a typical SEM image of Ag dendrites before they reacted with HAuCl4. The image reveals the self-similar hierarchical structure expected for a fractal dendrite formed by diffusion-limited growth. SEM image taken under a higher magnification (Fig. 1b) clearly shows that these “dendrites” are composed of a large number of polygonal nanocrystallites, which self-assembled into a 3D hierarchical structure. The branching angles are all 60°. The EDX on the Ag dendrites structures (Fig. 1f) clearly shows a major peak of silver with a much weaker peak for Cu, arising from the copper substrate. When the above GRR is performed on the aqueous HAuCl4 solution under the same conditions, however, large scale irregular Au nanostructures are formed (Fig. 1d, e). Galvanic replacement reaction has been demonstrated as a general and effective means for preparing metallic nanostructures (e.g., thin films) by consuming the more reactive component [4]. Since the standard reduction potential of AuCl4−/Au pair (0.99 V vs standard hydrogen electrode, SHE) is higher than that of the Ag +/Ag pair (0.80 V vs SHE), silver would be oxidized into Ag + when silver nanostructures and HAuCl4 are mixed in an aqueous medium: −



3Ag ðsÞ þ AuCl4 ðaqÞ → AuðsÞ þ 3AgClðsÞ þ Cl ðaqÞ:

ð1Þ

The oxidation of Ag 0 into Ag + leads to the gradual consumption of Ag and, at the same time, the production of Au 0, which is deposited on the Ag dendrites.

Fig. 1. SEM images of (a, b) Ag dendrites on the Cu surface, (d, e) Au nanostructures on the Cu surface; (c, f) EDX spectrum of Ag dendrites and Au nanostructures.

Z. Yi et al. / Thin Solid Films 520 (2012) 2701–2707

The typical SEM images are used to show the morphological and structural changes involved in various stages of the replacement reaction between silver dendrites and HAuCl4 in Fig. 2. After the silver dendrites have reacted with the first stage, the dimensions of each silver dendrites exhibit no apparent change in the initial stage of this reaction (Fig. 2a), only some small holes are formed at a specific site of the surface, indicating that the replacement reaction is initiated locally rather than over the entire surface. As shown in Fig. 2b, the holes can be clearly observed as black spots on the surfaces of some dendrites. The newly formed surfaces containing holes should represent

2703

the most active sites for further replacement reaction. If the reaction stages are for the second (Fig. 2d), the dimensions of each silver dendrites exhibit no apparent change also. The gold atoms are deposited on the surface of each dendrite as a very thin (most likely, incomplete) shell. As a result, the replacement reaction continues to generate larger holes (Fig. 2e). As further stages go on (in Fig. 2h), the dendrite is covered with polycrystalline bimetallic Ag/Au nanocubes that stand on their edges on top of it. It is clear that these nanocubes are smooth surface crystals with sharp corners and with an average edge length of 125 nm. These nanocubes also often mutually overlap.

Fig. 2. SEM images of Ag/Au structures after different stage of MGRR: (a, b) 1, (d, e) 3, (g, h) 5, (j, k) 7; (c, f, i, l) EDX spectrum of Ag/Au structures after different stages of GRR.

2704

Z. Yi et al. / Thin Solid Films 520 (2012) 2701–2707

Moreover, from this figure, one can already discern how the nanocubes form the walls of the cavities that are the most recognizable feature of the structures determines at even longer reaction times (we will henceforth always call Au-wall-build pores “cavities” to clearly distinguish them from the pores due to removal of Ag from the original Ag dendrite). If the reaction proceeds for seven stages, almost all the nanocubes seem to disappear and numerous cavities are visible instead (Fig. 2k). To examine the composition of the structures, EDX measurements were also performed. The gradual depression of Ag peaks and the development of Au ones (Fig. 2e, f, i, l) are direct evidence for the ongoing replacement of Ag with Au during the whole reaction stages. Semiquantitative EDX analysis estimates the Au atoms content of the four samples to be 4, 9, 13, and 20%for 1, 3, 5, and 7 stages, respectively. Fig. 3a shows a TEM image of two typical Ag dendrites whose shape is discussed above with the help of SEM observations (Fig. 1a, b). From the SAED pattern of a region in a tip (see the inset), we identify the (111), (200), and (311) planes of FCC Ag. The dendrite is highly crystalline, yet the two sets of diffraction spots also indicate the existence of a twinned structure. From the HRTEM data (Fig. 3b), we can also detect a lattice spacing of 0.235 nm which is the interplanar spacing of Ag (111). After reacting with AuCl4− ions for 5 stages, the Ag dendrites have become porous (Fig. 3c) because of the consumption of Ag. The rectangle region's SAED pattern (top inset of Fig. 3c) displays discontinuous concentric rings, which implies a polycrystalline bimetallic Ag/Au structure due to the deposition of Au on Ag dendrites. The generation of Au is reanalyzed by the EDX recorded from the same samples (Fig. 2i). The Ag/Au bimetal has an Au atomic

percentage of 12.4%. The polycrystalline nature of the bimetallic nanostructures is also seen in the HRTEM image (Fig. 3d). Analyzing the lattice spacings of the various domains reveals ones that can be assigned to (111) and (200) planes. Since the lattice structures of Au and Ag are too similar, it cannot be established whether these domains are Au, Ag, or mixed regions. The mutual solubility of Ag and Au and high diffusion rates at high reaction temperatures (e.g., boiling) result in the formation of single crystalline Au/Ag alloys [21]. The polycrystalline nature of the obtained nanostructures is therefore to be expected from our experiments. One can conclude that the MGRR turned almost single crystalline Ag dendrites (up to twinning) into polycrystalline bimetallic Ag/Au nanostructures. Compared to one-stage replacement approach, MGRR is more favorable for compositional and modality control. The MGRR is found to play a critical role in forming uniform Ag/Au Bimetallic dendrites nanostructures. As shown in Fig. 4(a), the Ag dendrites are transformed into structures with asymmetrical and irregular modality when the reaction is performed under the similar of conditions used in Fig. 2, except that the replacement reaction is one one-stage and the reaction time is 14 min. Here appears bigger ball. Fig. 4(b) and (c) is Ag/Au structures corresponding magnified views. As shown in Fig. 4(c), the ball is covered with flakes that stand on their edges on top of it. The flakes have a smooth surface and uniform thickness. The inset is a magnified image, from which one finds that the flakes also often mutually intersect and overlap. The striking difference in morphology observed here can be attributed to such major factors: AgCl solid has a smaller solubility product constant (ksp) at room temperature, which might lead to the simultaneous formation of AgCl precipitation during the replacement reaction. The byproduct

Fig. 3. (a) TEM image of a typical Ag dendrite. The inset is the SAED pattern from the rectangle region. (b) HRTEM image of the rectangle area in panel a. (c) High-magnification TEM image of an Ag/Au (5 stage) sample together with the SAED pattern from the rectangle region. (d) HRTEM image of the rectangle area in panel c.

Z. Yi et al. / Thin Solid Films 520 (2012) 2701–2707

2705

Fig. 5. UV–vis spectra of the as-prepared: (a) typical Ag dendrites, (b–e) Ag/Au structures after different stages of MGRR: (b) 1, (c) 3, (d) 5, (e) 7.

Fig. 4. SEM images of Ag/Au structures after one stage of MGRR and corresponding magnified views.

accompanied with the production of particles will be accumulated after one stage and a long reaction time, leading to spoil the optimal growing conditions. As is known, UV–vis spectroscopy can investigate the surface plasmon resonance (SPR) property that is commonly used to monitor the growth of the dendritic Ag/Au bimetallic nanostructures. The absorption spectra of products after different stages of replacement are shown in Fig. 5. The initial dendritic Ag nanostructures exhibited three peaks: a small peak at 335 nm, a shoulder peak at 450 nm, and a weak peak at 670 nm, which can be ascribed to the out-ofplane quadrupolar, inplane quadrupolar and in-plane dipolar SPR absorption band of dendritic Ag nanostructures respectively [22]. The in-plane dipolar band is always paid more attention, as it holds the strongest peak and is most sensitive to change of the anisotropy and component of the nanoparticles. Fig. 5(b–e) reveals that during the multi-stage replacement, the out of-plane quadrupolar band hardly shifts, the in-plane quadrupolar band redshifts slightly, and the in-plane dipolar band redshifts from around 670 nm to near 1000 nm step by step, entering the near infrared region. The redshift of the SPR peak may be the UV/vis absorption spectra have a contribution from both Ag as well as Au plasmon absorptions. Another fact may also be responsible for the hollow cavity of the

nanoparticles, which is in accordance with the literature reported previously [23]. Rhodamine 6G is chosen as an analyte in the present study because it has been well characterized by SERS and by resonance Raman spectroscopy. The SERS spectra of single R6G molecules have been obtained and most of the prominent Raman bands have been assigned [24]. The 514.5 nm line from an argon ion laser is used in the present study. This wavelength corresponds to a resonant excitation of R6G, so the SERS is acquired. The SERS is accepted that the increase in signal is due to an increase in the apparent cross-section of the molecules. Fig. 6 shows the SERS spectra that the R6G molecules are adsorbed onto (a) the cleaned thin copper plate; (b) Ag dendrites on the Cu surface. Fig. 6(a) shows the SERS of the R6G deposited film that is prepared by dropping 5 × 10 − 2 mL R6G methanol solution (1 × 10 − 6 M) onto the cleaned thin copper plate. As expected, we cannot distinguish any signal. Fig. 6(b) shows the SERS of the 5 × 10 − 2 mL R6G methanol solution (1 × 10 − 6 M) is deposited on Cu substrate covered by Ag dendrites. Several strong bands at 1650, 1598, 1574, 1507, 1361, 1310, 1187, 774 and 611 cm − 1 are observed on the substrate. The bands at 1650, 1574, 1507 and 1361 cm − 1 are assigned to aromatic C\C stretching; 1598 and 1130 cm − 1 are assigned to C_C stretching; 1187 cm − 1 are assigned to aromatic C\H bending; 611 cm − 1 are assigned to aromatic bending, respectively [25]. As is known, the shape and the aggregate of nanoparticles are important parameters for the surface enhancement in terms of the electromagnetic theory of SERS. Fig. 7 shows Raman spectra of substrates

Fig. 6. Raman spectra of 10− 6 M R6G on (a) the cleaned thin copper plate and on substrate covered by Ag dendrites (b).

2706

Z. Yi et al. / Thin Solid Films 520 (2012) 2701–2707

inelastic scattering intensity per molecule between the presence and absence of the SERS structure. The Raman enhancement factor can be written as [31,32]:



Fig. 7. Raman spectra of 10− 6 M R6G on (a) Au nanostructures on the Cu surface (b) Ag dendrites or (c–f) Ag/Au structures after different stages of MGRR: (c) 1, (d) 3, (e) 5, (f) 7.

prepared with one drop of 5 × 10 − 2 mL methanol solution (1 × 10 − 6 M). From Fig. 7, we find that the Raman enhancements by the Ag/Au samples in 1 and 3 stages are both greater than that of pure Ag dendrites, although the SERS effectiveness of Ag is usually better than that of Au or as investigated alloys [26]. This SERS enhancement may be related to several factors. First, surfaces of bimetals provide more possibilities for molecules to deposit on the boundaries between Ag and Au domains [27]. Second, adequate amounts of Au in a homogeneous alloy (solid solution) may intrinsically enhance the SERS activity. However, since the intrinsic activity of Ag is much higher [28], the most important reason may not be related to the added Au directly, but rather the corresponding morphological change of the underlying Ag dendrite. From the SEM of Fig. 2 a and d, we know that the MGRR leaves pores where Ag is depleted. High-curvature features can cause very large enhancement (lightening rod effect) for molecules adsorbed on the tips of needles or on edges [29]. Two close metallic surfaces can enhance the electromagnetic (EM) field around molecules absorbed between them, which lead to SERS enhancement [30]. However, as the MGRR stages increases from 1 to more, SERS enhancement again decreases. This confirms the increase in SERS-inactive Au covering active Ag. Especially the 5 and 7 stage samples also possess rough surfaces and even more nanocubes. However, as shown in curves e and f, their SERS enhancement is much below that of even the initial unmodified Ag dendrites. This is because compared with that of the 3 stage sample, the content of the SERS inert Au increases further. Again, this may be due to complex alloy surface chemistry or in the case of MGRR simply due to the fact that deposited Au increasingly starts to change and even more simply hide the rough SERS active Ag structure below (see Fig. 2). In conclusion, the roughness of Ag increases strongly at the beginning (e.g., stage = 1). However, as the stage further increases (e.g., stage = 5), a larger amount of Au nanocubes covers the Ag, which, although it may still increase in roughness, is also hidden beneath the Au surfaces. Finally, we calculate the Raman enhancement factor (G) from the spectra in Fig. 7. The enhancement factor is defined as the ratio of

ISERS =NSERS : IRef =NRef

ð2Þ

ISERS is the enhanced intensity of the adsorbed R6G molecules on the SERS substrate. The value of ISERS mainly arises from a single molecule layer covering a nanoparticle array, from which other additional molecules layers of analytes on the SERS substrate, as previously reported [33], do not contribute to Raman gain and can be neglected. IRef is the spontaneous Raman scattering intensity from the bulk R6G molecules under the laser spot on the blank Cu substrate. NSERS is the number of the single-layer molecules covering the SERS substrate under the laser spot. NRef is the number of the bulk molecules excited by laser on the surface of the regular substrate. In order to obtain the values of these four parameters, we follow the same procedures based on the published literatures [32–34]. Using the 100× objective lens, we determine the area of the laser spot size at around 1 μm 2. We also calculate the area of a single molecule of R6G to be approximately 1 × 10 4 nm 2. Thus, the value of NSERS under the laser excitation is approximately 1 × 10 2 molecules. The focusing scope of laser beam is approximately 5 μm, thus scattering from all R6G molecules underneath the laser spot is detected. Assuming a uniform distribution of R6G over the droplet area of 2 mm 2, the value of NRef is approximately 1.5 × 10 7 molecules. Therefore, from Eq. (2), the enhancement factor G is simply (1.5 × 10 5) · (ISERS/IRef). The enhancement factors of each assigned Raman peak as measured from different SERS substrates are also shown in Table 1. The results show that the highest enhancement factor can be achieved at the average value of 2.34 × 10 9 from the spectrum (c), which obtained from the optimized SERS substrate when the MGRR stage is at 1 stage. 4. Conclusion In conclusion, bimetallic Ag/Au nanostructures are synthesized via MGRR. Compared to one-stage replacement approach, MGRR is more favorable for compositional and modality control. Different reaction stages resulted in different morphologies, compositions, and crystal structures of the corresponding products, which were characterized by a series of techniques. On the basis of the results, the growth mechanism as the MGRR unfolds is discussed. The original Ag dendrites develop pores, while the growing Au nanocubes slowly cover this underlying structure. The dendritic Ag/Au bimetallic nanostructures could be used as highly efficient SERS active substrates. The exploitation of SERS effects in this work is an application of SERS based on its more traditional use in the detection of molecules. Acknowledgments The work was supported by the National Natural Science Foundation of China (No. 10804101), the State Key Development Program for Basic Research of China (Grant No. 2007CB815102), and the Science

Table 1 Enhancement factors (G) calculated from the SERS spectra (a)–(f) as shown in Fig. 7. Raman shift (cm− 1) G factor

774 a b c d e f

1187 7

5.71 × 10 6.85 × 108 2.12 × 109 8.86 × 108 1.86 × 108 7.44 × 107

1361 7

6.04 × 10 7.63 × 108 2.21 × 109 9.46 × 108 2.06 × 108 8.65 × 107

1507 7

7.05 × 10 8.76 × 108 2.33 × 109 1.01 × 109 2.16 × 108 9.74 × 107

1650 7

7.65 × 10 9.45 × 108 2.46 × 109 1.06 × 109 2.26 × 108 9.98 × 108

Average 7

8.03 × 10 1.08 × 109 2.58 × 109 1.21 × 109 2.38 × 108 1.01 × 108

6.90 × 107 8.70 × 108 2.34 × 109 1.02 × 109 2.14 × 108 9.18 × 107

Z. Yi et al. / Thin Solid Films 520 (2012) 2701–2707

and Technology Development Foundation of Chinese Academy of Engineering Physics (Grant No. 2007B08007). References [1] A. Steinbruck, A. Csaki, G. Festag, W. Fritzsche, Plasmonics 1 (2006) 79. [2] N. Nishida, Y. Shiraishi, S. Kobayashi, N. Toshima, J. Phys. Chem. C 112 (2008) 20284. [3] X.L. Li, B.Z. Li, M.H. Cheng, Y.K. Du, X.M. Wang, P.J. Yang, Mol. Catal. A: Chem. 284 (2008) 1. [4] W. Song, Y.C. Cheng, H.Y. Jia, W.Q. Xu, B. Zhao, J. Colloid Interface Sci. 298 (2006) 765. [5] X.L. Ren, X.W. Meng, F.Q. Tang, Sens. Actuators, B 110 (2005) 358. [6] L.H. Lu, G.Y. Sun, H.J. Zhang, H.S. Wang, S.Q. Xi, J.Q. Hu, Z.Q. Tian, R. Chen, J. Mater. Chem. 14 (2004) 1005. [7] M.J. Kim, K.Y. Lee, G.H. Jeong, J. Jang, S.W. Han, Chem. Lett. 36 (2007) 1350. [8] M. Treguer, C.D. Cointet, H. Reita, J. Khatouri, M. Mostafavi, J. Amblard, J. Phys. Chem. B 102 (1998) 4310. [9] Y.D. Jin, S.J. Dong, J. Phys. Chem. B 107 (2003) 12902. [10] B. David, S.L. Wang, E.J. Scott, H. Liang, J. Phys. Chem. C 111 (2007) 13665. [11] Y.G. Sun, J. Phys. Chem. C 114 (2010) 2127. [12] J. Yang, J.Y. Lee, H.P. Too, S. Valiyaveettil, J. Phys. Chem. B 110 (2006) 125. [13] V. Bansal, H. Jani, J.D. Plessis, P.J. Coloe, S.K. Bhargava, Adv. Mater. 20 (2008) 717. [14] X.W. Teng, Q. Wang, P. Liu, J. Am. Chem. Soc. 130 (2008) 1093. [15] L.T. Viyannalage, R. Vasilic, N. Dimitrov, J. Phys. Chem. C 111 (2007) 4036.

2707

[16] Y. Hong, Y. Rheem, M. Lai, D.M. Cwiertny, S.L. Walker, N.V. Myung, Chem. Eng. J. 151 (2009) 66. [17] S.F. Pang, T. Kondo, T. Kawai, Chem. Mater. 17 (2005) 3636. [18] Y.W. Hu, S. Liu, S.Y. Huang, W. Pan, Thin Solid Films 519 (2010) 1314. [19] M.H. Rashid, T.K. Mandal, J. Phys. Chem. C 111 (2007) 16750. [20] J.H. Liu, A.Q. Wang, Y.S. Chi, H.P. Lin, C.Y. Mou, J. Phys. Chem. B 109 (2005) 40. [21] S.E. Skrabalak, J.Y. Chen, Y.G. Sun, X.M. Lu, L. Au, C.M. Cobley, Y.N. Xia, Acc. Chem. Res. 41 (2008) 1587. [22] K.L. Kelly, E. Coronado, L.L. Zhao, J. Phys. Chem. B 3 (2003) 668. [23] A.M. Schwartzberg, T.Y. Olson, C.E. Talley, J.Z. Zhang, J. Phys. Chem. B 110 (2006) 19935. [24] W. Grochala, A. Kudelski, J.J. Bukowska, Raman Spectrosc. 29 (1998) 681. [25] X.S. Xu, H. Ming, Q.J. Zhang, Y.S. Zhang, J. Opt. A: Pure Appl. Opt. 4 (2002) 237. [26] L. Rivas, S. Sanchez-Cortes, J. Garcia-Ramos, G. Morcillo, Langmuir 16 (2000) 9722. [27] Y.C. Liu, S. Yang, J. Electrochim. Acta 52 (2007) 1925. [28] R.G. Freeman, M.B. Hommer, K.C. Grabar, M.A. Jackson, M.J. Natan, J. Phys. Chem. 100 (1996) 718. [29] X.Q. Zou, E.B. Ying, S.J. Dong, J. Colloid Interface Sci. 306 (2007) 307. [30] M.L. Xu, M.J. Dignam, J. Chem. Phys. 100 (1994) 197. [31] C.D. Keating, K.K. Kovaleski, M.J. Natan, J. Phys. Chem. B 102 (1998) 9414. [32] G. Laurent, N. Félidj, J. Aubard, G. Lévi, J.R. Krenn, A. Hohenau, G. Schider, A. Leitner, F.R. Aussenegg, Phys. Rev. B 71 (2005) 045430. [33] G.N. Xiao, S.Q. Man, Chem. Phys. Lett. 447 (2007) 305. [34] B. Guo, G.Y. Han, M.Y. Li, S.Z. Zhao, Thin Solid Films 518 (2010) 3228.