HOME | SEARCH | PACS & MSC | JOURNALS | ABOUT | CONTACT US

Controlled synthesis of dendritic Au@Pt core-shell nanomaterials for use as an effective fuel cell electrocatalyst

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2009 Nanotechnology 20 025605 (http://iopscience.iop.org/0957-4484/20/2/025605) The Table of Contents and more related content is available

Download details: IP Address: 155.69.4.4 The article was downloaded on 30/03/2009 at 08:50

Please note that terms and conditions apply.

IOP PUBLISHING

NANOTECHNOLOGY

Nanotechnology 20 (2009) 025605 (9pp)

doi:10.1088/0957-4484/20/2/025605

Controlled synthesis of dendritic Au@Pt core–shell nanomaterials for use as an effective fuel cell electrocatalyst Shuangyin Wang1 , Noel Kristian1 , Sanping Jiang2,3 and Xin Wang1,3 1

School of Chemical and Biomedical Engineering, Nanyang Technological University, 50 Nanyang Drive, 639798, Singapore 2 School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Drive, 639798, Singapore E-mail: [email protected] and [email protected]

Received 22 September 2008, in final form 5 November 2008 Published 9 December 2008 Online at stacks.iop.org/Nano/20/025605 Abstract We report the controlled synthesis of dendritic Au@Pt core–shell nanomaterials. The size and morphology of the Au cores and the Pt shell thickness of the Au@Pt core–shell nanostructures could be easily tuned. It was found that the directing agent and the reducing agent play critical roles in the synthesis of dendritic Au@Pt core–shell nanomaterials. For comparison purposes, conventional Au@Pt core–shell nanoparticles and monometallic Pt nanoparticles were also synthesized by the successive reduction method. Transmission electron microscopy (TEM) observations demonstrated the dendritic surface of the products obtained. The UV–visible (UV–vis) spectroscopy results and a comparison of the average diameter between the dendritic Au@Pt and conventional Au@Pt confirmed the relatively loose Pt shells around Au cores for the dendritic Au@Pt. The as-prepared dendritic Au@Pt showed enhanced electrocatalytic activity for methanol oxidation in acid medium, compared to the conventional Au@Pt and monometallic Pt.

attention has been paid to branched Pt nanomaterials such as porous nanoparticles, multipods, nanowire networks, and dendritic nanoparticles [12–19]. Generally, the synthesis of branched Pt nanomaterials relies on the use of either hard templates, such as a porous silica template, or soft templates, such as various surfactants, polymers, and dendrimers [19, 20]. The hard template method generally produces low metallic interconnectivity due to the poor continuity of precursors in mesoporous silica templates. In order to remove the silica template, hydrofluoric acid needs to be used, which is a serious concern for the environment and safety. Therefore, using a soft template to prepare these branched nanostructures has been increasingly preferred by more and more research groups. Zhang and co-workers [14] reported the synthesis of threedimensional (3D) branched Pt nanoparticles by modulating the growth kinetics in oleylamine. Yang et al [12] have synthesized porous Pt nanoparticles with dendritic surfaces in diphenyl ether using hexadecylamine as the surfactant and reaction solvent, and the porous Pt obtained showed enhanced

1. Introduction Morphology-controlled design of metal nanostructures is of great significance because their physical and chemical properties strongly depend on the size, shape, and dimensions. Metal nanostructures have extensive applications in various fields, including optics, electronics, photochemistry, sensing, and catalysis [1]. To date, various forms of metal nanostructure have been synthesized by physical or chemical methods, including highly monodispersed nanospheres, cubic nanoparticles, and various anisotropic nanostructures such as nanorods, nanowires, nanotubes, and nanosheets [2–9]. Platinum, as a typical noble metal, is of particular importance for fundamental research and industrial applications [10]. In most cases, Pt is used in the form of small nanoparticles, since small size offers high surface area which is especially essential for catalysis applications [11]. Recently, a lot of 3 Authors to whom any correspondence should be addressed.

0957-4484/09/025605+09$30.00

1

© 2009 IOP Publishing Ltd Printed in the UK

Nanotechnology 20 (2009) 025605

S Wang et al

adjustment of the Pt/Au molar ratios. The roles of CTAB as directing agents and L-ascorbic acid as weak reducing agent for the synthesis of dendritic Au@Pt nanomaterials were investigated. It was found that the presence of CTAB and the use of L-ascorbic acid are critical in obtaining the dendritic morphology. Finally, the dendritic Au@Pt nanomaterials obtained were used as the electrocatalyst for methanol oxidation in acid medium with a comparison to conventional spherical Au@Pt core–shell nanoparticles and monometallic Pt nanoparticles. The improved electrochemically active surface area and enhanced catalytic activity for methanol electrooxidation for this novel structure were demonstrated.

electrocatalytic activity for methanol oxidation. Using soft templates to form hexagonal liquid crystals as nanoreactors, Remita et al [21] have synthesized porous Pt nanoballs. Ha et al [16] have reported the pH-selective synthesis of 3D dendritic Pt nanoparticles using cetyltrimethylammonium bromide (CTAB) as a soft template. The as-prepared Pt nanoparticles show a decrease in the onset potential for oxygen reduction reaction. Similarly Yang et al [17] have reported the controllable synthesis of dendritic and porous Pt nanoparticles through the careful selection of reducing agent and the adjustment of the pH value of the reaction system in the presence of CTAB. Shelnutt et al [18] reported the synthesis of 2D and 3D dendritic Pt nanostructures by the seeding and autocatalytic reduction of Pt salts in aqueous surfactant solution. Although much work has been done on branched Pt nanostructures, branched bimetallic core–shell nanostructures with dendritic morphology have rarely been reported [22, 23]. Bimetallic core–shell nanoparticles have been attracting more and more attention due to their unusual optical, electronic, magnetic, and catalytic properties. A typical example is bimetallic Au core and Pt shell nanoparticles that exhibit improved catalytic activity as compared to simple mixtures of monometallic Pt and Au nanoparticles [24, 25]. Compared with conventional spherical Au@Pt core–shell nanoparticles, the branched structure of the Pt shell of dendritic Au@Pt core–shell nanoparticles would significantly increase the surface area of Pt, which is an essential factor in improving the catalytic activity of Pt. Recently, a few research groups have reported the synthesis of branched Au@Pt nanoparticles. Eichhorn et al [22] reported the synthesis of dendritic Au– Pt heteroaggregate nanostructures using Au nanoparticles of 10 nm as the cores. Dong et al [23] have reported the synthesis of spongelike Au@Pt core–shell nanomaterials with a hollow cavity, which were shown to have enhanced electrocatalytic activity. However, the size of the Au cores and the as-obtained spongelike Au@Pt core–shell nanoparticles is too big (30 nm and 70 nm for Au cores and spongelike Au@Pt, respectively), and does not meet the requirement of fine particle size for metal electrocatalysts in fuel cell applications. Direct methanol fuel cells (DMFCs) are promising power sources due to their high power density, relatively quick startup, rapid response to varying loading, and low operating temperatures [26, 27]. Platinum-based electrocatalysts are the most important catalysts in DMFCs [28, 29]. However, one of the most significant barriers for the widespread commercialization of DMFCs is the high cost of the precious Pt-based electrocatalysts. Correspondingly, improving the catalytic performance and utilization efficiency of Pt-based electrocatalysts can reduce the cost of DMFCs and thus speed up the commercialization of fuel cells. Much effort has been devoted to increasing the electrocatalytic activity and reducing the cost of metal electrocatalysts [28–31]. In this paper, we report the controlled synthesis of dendritic Au@Pt nanoelectrocatalysts in the presence of CTAB based on the seed-mediated growth method. The size and morphology of the Au cores and the Pt shell thickness of the dendritic Au@Pt could be easily controlled by choosing various Au cores with different size and morphology, and by

2. Experimental section 2.1. Materials The materials used in the present work included deionized (DI) water, sulfuric acid (99.5%, Fluka), methanol (Fluka), hexachloroplatinic (IV) acid (Sigma-Aldrich), sodium borohydride (Sigma-Aldrich), L-ascorbic acid (Alfa Aesar), cetyltrimethylammonium bromide (Acros), HAuCl4 (SigmaAldrich), AgNO3 (Sigma-Aldrich), carbon black (XC-72, Gashub), and trisodium citrate (Fluka). 2.2. Synthesis of ∼4 nm Au cores Au nanoparticles with the average diameter of ∼4 nm were prepared using the following procedure. 3.75 ml of 20 mM HAuCl4 was diluted by another 50 ml deionized water followed by mixing with 16.25 ml of 20 mM trisodium citrate. 10 ml of 10 mM NaBH4 was then added into the above solution dropwise under vigorous stirring. Finally, the total volume of the solution was adjusted to 80 ml by adding 10 ml DI water. After the addition of NaBH4 , the color of the solution changed from light yellow to light red very quickly, which demonstrates the successful formation of Au nanoparticles. 10 ml of the as-obtained suspension was collected for UV–vis and TEM characterization. The remaining 70 ml of the Au cores was heated to 40 ◦ C and kept for 30 min to decompose the excess NaBH4 . Finally, the as-obtained solution was stored at the room temperature. 2.3. Synthesis of Au nanorods Au nanorods were synthesized following the method reported by Wang et al [5]. First, the Au seeds were prepared by the following procedure. 0.125 ml of 20 mM HAuCl4 was mixed with 5 ml of 200 mM aqueous CTAB solution, followed by reduction by ice-cold NaBH4 . The Au seed solution obtained was kept at room temperature overnight to decompose the excess NaBH4 . Second, after the synthesis of Au seeds, the growth solution was prepared by mixing 3.75 ml of 20 mM HAuCl4 , 75 ml of 200 mM CTAB and 1.5 ml of 10 mM AgNO3 under stirring, followed by the addition of 1.05 ml of 200 mM ascorbic acid. The original yellow solution quickly became colorless. After gentle stirring, 180 µl of the above seed solution was added into the growth solution to initiate the growth of Au nanorods. Here AgNO3 was added to control 2

Nanotechnology 20 (2009) 025605

S Wang et al

Figure 1. TEM images of Au nanoparticles of ∼4 nm at low magnification (a) and high magnification (b).

spectrometer equipped with quartz cells. The morphology was observed using a JEOL 3010 transmission electron microscope at 160 kV. For the electrochemical characterization, the electrochemical signals were recorded with an Autolab PGSTAT302 potentiostat (Eco Chemie, Netherlands) in 0.5 M CH3 OH + 0.5 M H2 SO4 . The electrode potentials were measured and reported against a saturated calomel electrode (SCE) placed close to the working electrode. The counter electrode was Pt wire. The working electrode was prepared by depositing the catalyst onto a glassy carbon electrode based on a previously reported protocol [29]. The corresponding Pt loading on the working electrode could be calculated according to the specified molar ratios of Au@Pt in the catalysts. For the purpose of comparison, 20 wt% conventional carbon supported monometallic Pt catalysts were also prepared following the polyol method [29].

the reaction kinetics of the reduction of HAuCl4 . It has been demonstrated that only spherical Au nanoparticles could be formed without the addition of AgNO3 in the growth solution during the synthesis of Au nanorods. 2.4. Synthesis of ∼10 nm Au cores The synthesis of ∼10 nm Au cores was performed by the modified seed-mediated growth method. Briefly, 10 ml of the above ∼4 nm Au seeds was used as the seeds and was mixed with 75 ml of 200 mM CTAB, followed by heating to 50 ◦ C. After gentle stirring, 0.5 ml of 20 mM HAuCl4 was introduced into the above solution under vigorous stirring. 0.25 ml of 200 mM ascorbic acid was then added to start the reduction. 2.5. Synthesis of dendritic Au@Pt When using ∼4 nm Au nanoparticles as the cores, 20 ml of the as-obtained Au cores was mixed with 20 ml of 200 mM CTAB under vigorous stirring. Then certain amount of 200 mM ascorbic acid was added into the above solution, followed by heating until boiling. To form the Pt nanoshells on Au cores, a certain amount of H2 PtCl6 was dropped into the boiling solution and the solution was kept boiling for 1 h to complete the reduction of Pt(IV) to Pt(0). The specific amounts of ascorbic acid and H2 PtCl6 are based on the desired Pt/Au molar ratio. The molar ratio of ascorbic acid to H2 PtCl6 was kept at 5:1 to ensure the complete reduction of Pt salts. In the case of using Au nanorods and 10 nm Au nanoparticles as the cores, since the CTAB already existed in the system, extra addition of CTAB was not necessary. Thus ascorbic acid and Pt salts could be introduced following the same procedure as in the case of Au nanoparticles as the cores. For the synthesis of conventional Au@Pt core–shell nanoparticles, the same approach was adopted except that 20 ml DI water was added into the Au core solution instead of the above CTAB solution. All the nanostructures obtained were deposited on carbon black (XC-72) to form the catalyst, while keeping the Au loading at 10 wt%.

3. Results and discussion Au nanoparticles with different sizes (4 and 10 nm) and Au nanorods were prepared and used as cores for dendritic Pt shell growth. Figure 1 shows the typical TEM images of Au nanoparticles with the diameter of about 4 nm. Figure 2 shows TEM images of the nanostructures obtained after the growth of the dendritic Pt shells. Apparently, the surfaces of the dendritic Au@Pt are quite rough, indicating the dendritic nature of the Pt shell. The thickness of the Pt shell and the overall diameter of the as-obtained core–shell nanostructure were tuned via control of the Pt/Au molar ratio. At the molar ratio of 1:2, it was clearly observed that parts of the Au cores are still not covered by Pt shells, as shown in figure 2(a). As the molar ratios reached 1:1, complete dendritic Pt shells can be regarded to form. With the further increase of the Pt/Au ratio to 2:1 and 3:1, the thickness of the Pt shells and the diameter of the core–shell structures both increase. The average diameters of dendritic Pt/Au core–shell nanoparticles are about 6.5 nm, 10.2 nm, 12.5 nm, and 13.4 nm for Pt/Au molar ratios 1:2, 1:1, 2:1 and 3:1, respectively. It was also observed that the dendritic Pt shells became denser with the increase of Pt/Au, and at high Pt/Au molar ratio (Pt/Au = 3:1), we can clearly observe the shell structure as shown in the inset of figure 2(d). To confirm the existence of Pt and Au elements, an energy dispersive x-ray spectrometer (EDS) attached to the TEM equipment was used. A typical EDS spectrum for the sample in figure 2(d) is shown

2.6. Characterization methods The UV–vis spectra of the Au and Au@Pt core–shell nanomaterials were measured with a Shimadzu UV2450 3

Nanotechnology 20 (2009) 025605

S Wang et al

Figure 2. TEM images of dendritic Au@Pt at different Pt/Au molar ratios: 1:2 (a), 1:1 (b), 2:1 (c), and 3:1 (d).

Figure 3. The EDS spectrum of the samples in figure 2(d).

in figure 3. The quantitative analysis indicates that the Pt/Au molar ratio is 2.6, slightly smaller than the theoretical feeding ratio of 3. It is believed that the formation of a dendritic Pt shell, instead of a compact Pt shell, is due to the presence of CTAB. Herein, CTAB functions as the directing agent. In order to obtain shape-controlled metal nanocrystals, directing agents are always used. However, when the interaction between the directing agents and the metal is too strong, the catalytic activity of the metal would be reduced. For example, polyvinylpyrrolidone (PVP), as the most widely used directing polymer for the shape-controlled synthesis of metal nanocrystals, strongly interacts with the platinum surface and

thus blocks many active sites of the Pt surface. In this work, CTAB has been carefully chosen as the directing agent, as it has been demonstrated that the interaction between CTAB and the Pt surface is significantly weaker and it offers the possibility to control the morphology of metal nanostructured materials while still preserving catalytically active sites. Yang et al [17] have reported the synthesis of porous Pt nanoparticles using a similar method, except for the absence of Au cores or seeds. In their case, the use of ascorbic acid as the reducing agent would lead to the formation of porous Pt nanoparticles, instead of dendritic structures. However, in our present study, the introduction of Au seeds provides the substrates for the further deposition of newly formed Pt atoms 4

Nanotechnology 20 (2009) 025605

S Wang et al

Figure 4. The TEM images of conventional Au@Pt at different Pt/Au molar ratios: 1:1 (a), 3:1 (b), and EDS spectrum of conventional Au@Pt at Pt/Au = 1:1 (c).

by the weak reducing agent, ascorbic acid. After mixing Au seeds with CTAB, microscopically, the molecular chains of CTAB would be adsorbed on the surface of Au seeds due to the interaction of Au and bromide ions of the CTAB molecules. The existence of CTAB around Au surfaces would direct the formation of the branched Pt nanostructures during their reduction and deposition. As a result, dendritic Au@Pt core– shell nanostructures were obtained. To confirm the function of CTAB in the synthesis of dendritic Au@Pt core–shell nanostructures, a comparative experiment was performed in the absence of CTAB. Without the CTAB template during the synthesis, the deposition of Pt atoms on Au cores would not be directed and conventional Au@Pt core–shell nanostructures with a compact Pt shell would be expected to form on the Au core, which is confirmed by the TEM analysis (figures 4(a) and (b)) for the two samples with Pt/Au molar ratios 1:1 and 3:1. Clearly, solid Au@Pt core–shell nanoparticles with relatively smooth surfaces were obtained, instead of the rough and dendritic shell in the case of the presence of CTAB. Figure 4(c) shows the EDS spectrum of Pt/Au = 1:1, which confirms the successful deposition of Pt on Au cores. The average diameter of the samples with Pt/Au = 3:1 is bigger than that of Pt/Au = 1:1 (8.5 nm versus 6.1 nm), suggesting the further deposition of Pt onto the Au core. The comparison of the average diameters of dendritic and conventional Au@Pt at different Pt/Au molar ratios is presented in figure 5. Obviously, at the same Pt/Au molar ratio, the dendritic Au@Pt nanostructures always show significantly bigger diameter than conventional Au@Pt. This result demonstrates that the dendritic Pt shells have looser aggregation around Au cores compared with the conventional

Figure 5. Plots of the average diameter of dendritic Au@Pt (– –) and conventional Au@Pt (––) against the Pt/Au molar ratio.

Au@Pt core–shell nanostructures, due to the presence of CTAB during the deposition of Pt atoms on Au cores. The UV–vis spectra of Au@Pt core–shell nanomaterials were collected to trace the effect of Pt shell thickness on the absorption peak of Au cores. The UV–vis spectra of the dendritic Au@Pt are presented in figure 6(a). The adsorption peak at 530 nm due to the surface plasmon resonance of Au could be clearly observed for the Au cores without the Pt shells. With the increase of the Pt/Au molar ratio, the absorption 5

Nanotechnology 20 (2009) 025605

S Wang et al

peaks were gradually suppressed, as a result of the increase of Pt shell thickness. When the ratio reaches 3:1, the Au plasmon absorption peak could be considered to vanish. A similar phenomenon was observed for conventional Au@Pt core–shell nanoparticles, as shown in figure 6(b). However, the complete suppression of the absorption peaks occurs at much lower Pt/Au ratio (1:1). The delayed suppression for dendritic Au@Pt relative to conventional Au@Pt also demonstrates the loose shell structure of dendritic Au@Pt core–shell nanostructures, which is consistent with the conclusion based on the comparison of the average diameters between the dendritic and conventional core–shell nanostructures at fixed Pt/Au molar ratios. The dendritic structure can be formed on Au cores with different size or different morphology using the same seedmediated growth method, as stated in section 2. As an example of the core size change, figure 7 shows the TEM images of the Au cores obtained with average diameter of 10 nm and the corresponding dendritic Au@Pt structure with Pt/Au ratios of 1:1, 3:1 and 8:1. The dendritic Pt shell on bigger cores seems denser compared with that on 4 nm cores with the same Pt/Au ratio. We also synthesized one-dimensional dendritic Pt/Au core–shell nanorods using Au nanorods as the cores. Au nanorods with smooth surface were first synthesized with an aspect ratio of ∼2.5. Once we coated the Au nanorods with Pt shells in the presence of CTAB, a flocky surface was also observed, as shown in figure 8. It can be seen that both the dendritic Au@Pt nanostructures still retain their core’s shape, be it sphere or nanorod. During the synthesis of dendritic Au@Pt nanoparticles, ascorbic acid, which is a relatively weak reducing agent,

Figure 6. UV–vis spectra of dendritic Au@Pt (a) and conventional Au@Pt (b). (This figure is in colour only in the electronic version)

Figure 7. TEM images of Au cores of ∼10 nm (a) and of dendritic Au@Pt using 10 nm Au nanoparticles as the cores with different Pt/Au molar ratios: 1:1, (b) 3:1 (c), and 8:1 (d).

6

Nanotechnology 20 (2009) 025605

S Wang et al

Figure 8. (a) TEM images of as-prepared Au nanorods. (b) Dendritic Au@Pt using Au nanorods as the cores. The Pt/Au molar ratio is 3:1.

Figure 9. TEM image of Au–Pt sample synthesized in the presence of CTAB using NaBH4 as the reducing agent.

was used to reduce Pt(IV) to Pt(0). To study the effect of reducing agents, a strong reducing agent, NaBH4 , was used instead of ascorbic acid during the synthesis of Au@Pt core– shell nanoparticles in the presence of CTAB. As shown in figure 9, no dendritic core–shell Au@Pt nanoparticles were observed. Au nanoparticles with the average particle size of 4 nm and Pt nanoparticles with average particle size of 12 nm were separately distributed. Due to the strong reducing ability of NaBH4 , the addition of NaBH4 leads to the fast nucleation of a large amount of Pt atoms protected by CTAB, and the as-formed Pt atoms would quickly aggregate to form big Pt nanoparticles, instead of gradual attachment on Au cores. In contrast, the reduction rate is relatively slow for the case of ascorbic acid. The initially formed Pt atoms are of low concentration; they prefer to grow on the existing Au cores/substrate, instead of by self-nucleation to form individual Pt nanoparticles. This observation confirms the importance of ascorbic acid as the reducing agent for the formation of dendritic Au@Pt core–shell nanoparticles. As a demonstration of the potential application of this novel structure, the electrocatalytic activity was tested for methanol oxidation in acid medium on dendritic Au@Pt, conventional Au@Pt and monometallic Pt nanoparticles supported on carbon black. Figure 10 shows the cyclic voltammetry (CV) results from the three catalysts measured in nitrogen-saturated 0.5 M H2 SO4 (figure 10(a)) and 0.5 M

Figure 10. The mass normalized cyclic voltammetry (CV) on Pt/C (solid line), conventional Au@Pt (dashed line) and dendritic Au@Pt (dotted line) in nitrogen-purged 0.5 M H2 SO4 (a) and 0.5 M CH3 OH + 0.5 M H2 SO4 (b) at room temperature. The current densities were normalized based on the Pt loading in the catalysts. The scan rate is 50 mV s−1 .

H2 SO4 + 0.5 M CH3 OH (figure 10(b)) at room temperature, where the current density was normalized by the Pt loading. The electrochemical surface area (ECSA) of the three catalysts can be obtained from the area of the hydrogen desorption peak after correcting for the double layer charging current from the CV results in 0.5 M H2 SO4 (figure 10(a)). It can be seen that the dendritic Au@Pt nanoparticles supported on carbon 7

Nanotechnology 20 (2009) 025605

S Wang et al

black exhibit a higher hydrogen desorption peak than that of conventional Au@Pt and monometallic Pt. This indicates that the dendritic Au@Pt nanoparticles have higher ECSAs, apparently due to the branched structure of their Pt shells. This also demonstrates that dendritic Au@Pt supported on carbon black is more electrochemically accessible, which is very important for electrocatalyst applications in fuel cells. As a result of the higher ECSA for dendritic Au@Pt, high electrocatalytic activity for the electro-oxidation reaction of methanol was observed (figure 10(b)). The Faradaic current for the reaction in de-aerated 0.5 M H2 SO4 + 0.5 M CH3 OH solution exhibits the well-known features of methanol oxidation on Pt-based electrocatalysts. The activity of the three electrocatalysts for methanol oxidation can be represented by the magnitude of the forward anodic current peak. The peak current density is 203.9, 136.1, and 81.4 mA mg−1 Pt for dendritic Au@Pt, conventional Au@Pt and monometallic Pt, respectively. The significantly higher anodic current for the reaction on dendritic Au@Pt and conventional Au@Pt indicates a much higher electrocatalytic activity of Au@Pt with the core–shell structure than traditional monometallic Pt nanoparticles, because of a better Pt utilization on these structures, which is consistent with our previous report [20]. The higher electrocatalytic activity for methanol oxidation on dendritic Au@Pt than that on conventional Au@Pt again shows the excellence of the branched Pt shells for dendritic Au@Pt, which is consistent with those reported by other authors related to branched Pt nanoparticles [10, 14]. This can also be explained by the higher ECSA of the dendritic structure. The results demonstrate the promising potential of branched Pt nanostructures as highly efficient catalysts for the development of Pt-based electrocatalysts in fuel cells. Finally, we investigated the dependence of the catalytic activity of dendritic Au@Pt core–shell nanoparticles on the morphology (shape and size) of the Au cores. As shown in figure 11, the linear sweep voltammetry curves for methanol oxidation on dendritic Au@Pt nanoparticles using different Au cores (4 nm Au nanoparticles, 10 nm nanoparticles, and Au nanorods) at the same Pt/Au molar ratio (Pt/Au = 1:1) were collected. It was found that dendritic Au@Pt using 4 nm Au nanoparticles as the cores shows the highest current density for methanol oxidation. This result could be attributed to the smaller Au core size. Both 10 nm Au nanoparticles and Au nanorods have bigger particle size, compared with 4 nm Au nanoparticles; and at the same Pt/Au molar ratio, the corresponding Pt shells would be thicker for both 10 nm Au nanoparticles and Au nanorods as the cores. The thicker Pt shell would not only suppress the promotional role of the Au cores, such as the electronic effect, but also lead to the lower utilization of Pt. On the other hand, dendritic Au@Pt nanorods show slightly higher current density than dendritic Au@Pt nanoparticles with 10 nm Au nanoparticles as the cores, probably due to the advantageous one-dimensional nature of dendritic Au@Pt nanorods [8].

Figure 11. The linear sweep voltammetry curves for methanol oxidation on dendritic Au@Pt nanoparticles using 4 nm Au nanoparticles (dotted line), 10 nm Au nanoparticles (solid line) and Au nanorods (dashed line) as the cores in nitrogen-purged 0.5 M CH3 OH + 0.5 M H2 SO4 at room temperature. All three catalysts have the same Pt/Au molar ratio of Pt/Au = 1:1. The current densities were normalized based on the Pt loading in the catalysts. The scan rate is 20 mV s−1 .

reducing agent. Both the size of the Au cores and the Pt shell thickness of the dendritic Au@Pt could be tuned. The critical role of CTAB and ascorbic acid for the synthesis of dendritic Au@Pt core–shell nanoparticles was demonstrated. The loose structure of the Pt shell provides a larger electrochemically active surface area, which is particularly important for the use of Pt electrocatalysts in DMFCs. The as-obtained dendritic Au@Pt nanoparticles show enhanced electrocatalytic activity for methanol electro-oxidation in acid medium, compared with conventional Au@Pt and monometallic Pt nanoparticles. The as-obtained nanoelectrocatalysts have potential applications in fuel cells due to the improved Pt utilization and catalytic activity.

Acknowledgments This work is supported by a start-up grant of Nanyang Technological University, supplementary equipment purchase fund (RG116/06), academic research fund AcRF tier 1 (RG40/05) and AcRF tier 2 (ARC11/06), Ministry of Education, Singapore.

References [1] Daniel M and Astruc D 2004 Chem. Rev. 104 293 [2] Habas S E, Lee H, Radmilovic V, Somorjai G A and Yang P 2007 Nat. Mater. 6 692 [3] Wang C, Daimon H, Lee Y, Kim J and Sun S 2007 J. Am. Chem. Soc. 129 6974 [4] Sakai G, Yoshimura T, Isohata S, Uota M, Kawasaki H, Kuwahara T, Fujikawa D and Kijima T 2007 Adv. Mater. 19 237 [5] Wang C, Ma Z and Su Z 2006 Nanotechnology 17 1819 [6] Lee E P, Peng Z, Cate D M, Yang H, Campbell C T and Xia Y 2007 J. Am. Chem. Soc. 129 10634

4. Conclusions In summary, we successfully synthesized dendritic Au@Pt using CTAB as the directing agent, and ascorbic acid as the 8

Nanotechnology 20 (2009) 025605

S Wang et al

[19] Song Y, Garcia R M, Dorin R M, Wang H, Qiu Y, Coker E N, Steen W A, Miller J E and Shelnutt J A 2007 Nano Lett. 7 3650 [20] Han Y, Kim J M and Stucky G D 2000 Chem. Mater. 12 2068 [21] Surendran G, Ramos L, Pansu B, Prouzet E, Beaunier P, Audonnet F and Remita H 2007 Chem. Mater. 19 5045 [22] Zhou S, Mcllwrath K, Jackson G and Eichhorn B 2006 J. Am. Chem. Soc. 128 1780 [23] Guo S, Fang Y, Dong S and Wang E 2007 J. Phys. Chem. C 111 17104 [24] Kristian N and Wang X 2008 Electrochem. Commun. 10 12 [25] Zhao D and Xu B 2006 Angew. Chem. Int. Edn 45 4955 [26] Shukla A K, Raman R K and Scott K 2005 Fuel Cells 5 436 [27] Arico A S, Srinivasan S and Antonucci V 2001 Fuel Cells 1 133 [28] Wang S, Wang X and Jiang S P 2008 Langmuir 24 10505 [29] Wang S, Jiang S P and Wang X 2008 Nanotechnology 19 265601 [30] Hu X, Wang T, Wang L, Guo S and Dong S 2007 Langmuir 23 6352 [31] Kim J, Jung C, Rhee C K and Lim T 2007 Langmuir 23 10831

[7] Wang S, Wang X and Jiang S P 2008 Nanotechnology 19 455602 [8] Wang S, Kristian N, Jiang S and Wang X 2008 Electrochem. Commun. 10 961 [9] Fan F, Liu D, Duan S, Xie Z, Jiang Z and Tian Z 2008 J. Am. Chem. Soc. 130 6949 [10] Reddington E, Sapienza A, Gurau B, Viswanathan R, Sarangapani S, Smotkin E S and Mallouk T E 1998 Science 280 1735 [11] Kim P, Joo J B, Kim W, Kim J, Song I K and Yi J 2006 J. Power Sources 160 987 [12] Teng X, Liang X, Maksimuk S and Yang H 2006 Small 2 249 [13] Wakayama H and Fukushima Y 1999 Chem. Commun. 391 [14] Zhang H, Ding J and Chow G 2008 Langmuir 24 375 [15] Teng X and Yang H 2005 Nano Lett. 5 885 [16] Ullah M H, Chuang W, Kim I and Ha C 2006 Small 2 870 [17] Lee H, Habas S E, Kweskin S, Butcher D, Somorjai G A and Yang P 2006 Angew. Chem. Int. Edn 45 7824 [18] Song Y, Yang Y, Medforth C J, Pereira E, Singh A K, Xu H, Jiang Y, Brinker C J, Swol F V and Shelnutt J A 2004 J. Am. Chem. Soc. 126 635

9