Sie sind auf Seite 1von 12

www.afm-journal.

de

FEATURE ARTICLE

Shape-Controlled Synthesis of Pd Nanocrystals in Aqueous Solutions


By Byungkwon Lim, Majiong Jiang, Jing Tao, Pedro H. C. Camargo, Yimei Zhu, and Younan Xia*

This article provides an overview of recent developments regarding synthesis of Pd nanocrystals with well-controlled shapes in aqueous solutions. In a solution-phase synthesis, the nal shape taken by a nanocrystal is determined by the twin structures of seeds and the growth rates of different crystallographic facets. Here, the maneuvering of these factors in an aqueous system to achieve shape control for Pd nanocrystals is discussed. L-ascorbic acid, citric acid, and poly(vinyl pyrrolidone) are tested for manipulating the reduction kinetics, with citric acid and Br ions used as capping agents to selectively promote the formation of {111} and {100} facets, respectively. The distribution of single-crystal versus multiple-twinned seeds can be further manipulated by employing or blocking oxidative etching. The shapes obtained for the Pd nanocrystals include truncated octahedron, icosahedron, octahedron, decahedron, hexagonal and triangular plates, rectangular bar, and cube. The ability to control the shape of Pd nanocrystals provides a great opportunity to systematically investigate their catalytic, electrical, and plasmonic properties.

1. Introduction
Palladium is a key catalyst invaluable to many industrial processes; notable examples include hydrogenation/dehydrogenation reactions, low-temperature reduction of automobile pollutants, and petroleum cracking.[13] It has also demonstrated remarkable performance in hydrogen storage at room temperature and atmospheric pressure.[4] In organic chemistry, a large number of carbon-carbon bond forming reactions such as Suzuki, Heck, and Stille coupling all depend on catalysts based upon Pd(0) or its compounds.[57] It has been shown that the activity and selectivity
[*] Prof. Y. Xia, Dr. B. Lim, P. H. C. Camargo Department of Biomedical Engineering, Washington University St. Louis, Missouri 63130 (USA) E-mail: xia@biomed.wustl.edu M. Jiang Department of Chemistry, Washington University St. Louis, Missouri 63130 (USA) Dr. J. Tao, Dr. Y. Zhu Condensed Matter Physics & Materials Science Department Brookhaven National Laboratory Upton, New York 11973 (USA)

DOI: 10.1002/adfm.200801439

of a catalyst can be greatly enhanced by the use of nanocrystals enclosed by specic crystal facets that are intrinsically more active for a particular reaction.[811] Since the facets exposed on a nanocrystal are determined by its shape, an exquisite shape control of Pd nanocrystals is therefore highly desired for tailoring their catalytic properties and also a prerequisite for high performance in various catalytic applications. Over the last few years, polyol synthesis has been a preferred method of preparing noble metal nanocrystals with well-dened shapes because of the ability of polyols such as ethylene glycol (EG) to dissolve many metal salts (precursors to the noble metals), and also due to the temperaturedependent reducing power of such polyols.[1228] The primary step of this process involves the reduction of a metal salt by a polyol at an elevated temperature in the presence of a polymeric stabilizer such as poly(vinyl pyrrolidone) (PVP). Despite its success in controlling the shape of many noble metal nanocrystals, however, the major products are often restricted to cuboctahedrons or truncated cubes due to the fast reduction and growth rates associated with the strong reducing power of a polyol.[13,17,18,22] In addition, polyol synthesis is often troubled by the irreproducible results associated with the shape of metal nanocrystals due to the presence of trace amounts of impurities (known or unknown) that are usually contained in commercial chemical reagents such as EG. For example, we have shown that for the polyol synthesis of Ag nanocrystals based upon EG, even the presence of a ppm level of Cl impurity could drastically alter the morphology of the nal products.[13] Furthermore, the mechanism by which metal ions are reduced in a polyol synthesis is still poorly understood. Our recent results indicate that in the temperature range of 140160 8C, the primary reducing agent is glycolaldehyde, being produced via thermal oxidation of EG by the oxygen in air, rather than acetaldehyde derived from the dehydration of EG, which has been assumed as the reducing agent in a typical polyol synthesis for several decades.[29] Of course, knowledge of the exact mechanism underlying the reaction pathway is essential to both reproducibility and scale-up production of metal nanocrystals with well-controlled shapes. Compared to polyol synthesis, a water-based system should provide a more environmentally sound route to the production of

Adv. Funct. Mater. 2009, 19, 189200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

189

www.afm-journal.de

noble metal nanocrystals because it does not involve toxic organic solvents. In a water-based synthesis, the reduction of a metal precursor can be readily achieved by introducing various reducing agents that are safe and easy to handle, with typical examples including L-ascorbic acid, citric acid, and alcohol. Importantly, by using chemicals with different reducing powers, one can easily manipulate the reduction kinetics and conduct a systematic study on the formation mechanism of differently shaped Pd nanocrystals. In addition, high-purity water is more readily accessible than an organic solvent, so we do not worry about unexpected results that might be caused by trace amounts of impurities. A water-based system also provide a number of other merits such as simplicity, convenience, and the potential for large-scale production.[3037] For these reasons, we and other groups have recently started to pay more attention to the waterbased system as a more attractive route to the shape-controlled synthesis of noble metal nanocrystals. This Feature Article provides a brief account of these efforts, with a primary focus on our own work. More specically, we want to demonstrate the facile synthesis of Pd nanocrystals with a rich variety of shapes including truncated octahedron, icosahedron, octahedron, decahedron, hexagonal and triangular thin plates, rectangular bar, and cube. It is worth noting that some of these shapes (e.g., octahedron and thin plates) could not be achieved using the polyol method.

tals.[32,35,39,40] For most chemical syntheses of metal nanocrystals, PVP has been widely used as a steric stabilizer to protect the product from agglomeration. As we recently demonstrated using 13 C NMR spectroscopy, however, the ends of commercially available PVP (if it is synthesized in an aqueous medium) are terminated in the hydroxyl (OH) group due to the involvement of water and hydrogen peroxide in polymerization.[30] Therefore, it can act like a long-chain alcohol and serve as a class of weak reducing agents. Note that the reducing power of an alcohol decreases as its alkyl chain becomes longer. In a water-based system, the difference in reducing power for these reagents enables one to control the reduction kinetics, and thus the shape of Pd nanocrystals. Table 1 summarizes the reaction conditions employed in this work and the shapes of Pd nanocrystals obtained under each condition. To appreciate the difference in reduction rate associated with the reducing agents, the concentration of [PdCl4]2 was monitored by UV-vis spectroscopy during the early stage of each reaction. As illustrated in Figure 2, the absorption peak at 425 nm that corresponds to [PdCl4]2 completely disappeared at t 10 min in the case of L-ascorbic acid, demonstrating the fast reduction of [PdCl4]2 by L-ascorbic acid under this reaction condition. In contrast, the reduction rate was much slower in the case of PVP, indicating the weak reducing power of PVP. In the case of citric acid, it exhibited a moderate reducing power between PVP and L-ascorbic acid. These results also demonstrate that a wide range of reduction rates could be accessed by using different reducing agents.

FEATURE ARTICLE

2. Reducing Agents
Reduction kinetics plays a key role in controlling the nucleation and growth of nanocrystals. In this work, we employed L-ascorbic acid, citric acid, and PVP as reducing agents (Fig. 1). Among them, L-ascorbic acid, which is commonly known as vitamin C, can serve as a strong reducing agent for fast reduction of a Pd precursor, as well as other noble metal salts.[38] Citric acid works as a reducing agent in a manner similar to the mechanism of a conventional citrate-based synthesis of noble metal nanocrys-

3. L-Ascorbic Acid for Fast Reduction and Formation of Truncated Octahedrons


In a solution-phase synthesis of metal nanocrystals, the number of twin planes in seeds plays the most important role in determining the shapes taken by the nal products.[25,41] When the reduction is relatively fast, there are sufcient Pd atoms that can be added to the surface of seeds for continuous growth, leading to a rapid size increase. In this case, the seeds tend to take a single-crystal or multiple-twinned structure in an attempt to minimize the total surface energy of the system under a given Younan Xia was born in Jiangsu, China, in 1965. He received a B.S. degree in chemical physics from the University of Science and Technology of China (USTC) in 1987, an M.S. degree in inorganic chemistry from the University of Pennsylvania in 1993 and a Ph.D. in physical chemistry from Harvard University in 1996. He is currently at Washington University in the Department of Biomedical Engineering, where his research centers on the design and synthesis of nanostructured materials with controlled properties.

Figure 1. Structural illustration of L-ascorbic acid, citric acid, and OHterminated PVP and their oxidized forms due to the redox reactions with Pd2 ions.

190

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2009, 19, 189200

www.afm-journal.de

Table 1. Summary of shapes that have been obtained under different experimental conditions.
Precursor (mM) Reducing agent (mM)
L-ascorbic

FEATURE ARTICLE

Capping agent (mM)

Stabilizer (mM) PVP (87)[a]

Solution temp. (8C) 100

Reaction time (h) 3

Shape

Schematic drawing[b]

Na2PdCl4 (17.4)

acid (31)

truncated octahedron

Na2PdCl4 (5.8)

citric acid (28)

citric acid (28)

PVP (29)[a]

90

26

icosahedron

Na2PdCl4 (7.4)

citric acid (28)

citric acid (28)

PVP (37)[a]

90

26

octahedron

Na2PdCl4 (17.4)

citric acid (84)

citric acid (84)

PVP (87)[a]

90

26

decahedron

Na2PdCl4 (17.4)

PVP (87)[a]

PVP (87)[a]

100

hexagonal/ triangular plate bar

Na2PdCl4 (17.4)

L-ascorbic

acid (31) Na2PdCl4 (17.4)


L-ascorbic

KBr (230) KBr (460) KBr (230) KBr (460)

PVP (87)[a] PVP (87)[a] PVP (87)[a] PVP (87)[a]

100

100

acid (31) Na2PdCl4 (17.4)


L-ascorbic

cube/ pentagonal rod cube

80

acid (31) Na2PdCl4 (17.4) PVP (87)[a]

100

cube

[a]The concentration of PVP was calculated in terms of the repeating unit. All of these syntheses were carried out with a xed molar ratio of PVP to Na2PdCl4 at 5. [b]The blue and gray colors represent the {100} and {111} facets, respectively. Twin planes are delineated with red lines.

Figure 2. UV-vis spectra from tests on 17.4 mM Na2PdCl4 solutions containing 31 mM L-ascorbic acid, 84 mM citric acid, and 87 mM PVP as a reducing agent after heating for 10 min; heating was performed at 100 8C for the solutions containing L-ascorbic acid and PVP and at 90 8C for the solution containing citric acid. These conditions correspond to those for the syntheses of Pd truncated octahedrons, decahedrons, and hexagonal and triangular plates, respectively. The absorption peak at 425 nm is directly proportional to the concentration of the [PdCl4]2 species.

volume. For a face-centered cubic (fcc) structure, the surface energies of the low-index crystallographic facets that typically encase a nanocrystal are in the order of g{111} < g{100} < g{110}. This sequence implies that a single-crystal seed should take an octahedral or tetrahedral shape in order to maximize the expression of {111} facets and minimize the total surface energy. Both shapes, however, have larger surface areas than a cube of the same volume. As a result, it is not unexpected for a single-crystal seed to evolve into a truncated octahedron (or the so-called Wulff polyhedron) enclosed by eight {111} and six {100} faces. This shape has a nearly spherical prole and thus the smallest surface area to minimize the total interfacial free energy. Multiple-twinned seeds with a vefold symmetry (e.g., icosahedral and decahedral seeds) are another class of favorable seeds typically observed in a solution-phase synthesis. In addition to single-crystal seeds, these vefold twinned seeds can be produced under the same reaction condition and grow into the socalled multiple-twinned particles (MTPs) including icosahedrons and decahedrons. These twinned species achieve the lowest total free energy by maximizing the surface coverage with the {111} facets. As MTPs grow rapidly into large sizes, however, the total free energy of the system will drastically go up because of the increase in strain energy as caused by the twin defects. For example, in an ideal decahedron, which can be described as an

Adv. Funct. Mater. 2009, 19, 189200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

191

www.afm-journal.de

pentagonal rod without any signicant increase in strain energy, if the {100} facets on the side surface can be stabilized.[16,42] In essence, the population of seeds with different twin structures is mainly determined by the statistical thermodynamics related to the free energies of different species in combination with the reduction kinetics regarding the generation and addition of metal atoms to the nuclei. In practice, the distribution of single-crystal versus twinned seeds can be further manipulated via the introduction of other processes such as oxidative etching, in which zero-valent metal atoms are oxidized back to ions. When the synthesis is conducted in air, a combination of a ligand from the metal ion such as Cl and O2 from air can result in a powerful etchant such as the O2/Cl pair for both the nuclei and seeds. Compared to twinned seeds, single-crystal seeds are more resistant to oxidative etching due to the lack of defect zones on the surface. By taking advantage of this selectivity, the population of differently structured seeds in a solutionphase synthesis can be manipulated in controllable fashion. For example, as we have demonstrated for Ag system, all twinned seeds can be removed from the solution by adding a trace amount of Cl to the reaction, leaving behind only single-crystal truncated octahedrons or cubes in the products.[13] Here we demonstrate that truncated octahedrons of Pd can be produced in high yields by coupling the fast reduction of a Pd precursor with oxidative etching for the selective removal of twinned structures. Experimentally, Na2PdCl4 is the most commonly used precursor for Pd because of its stability in air and good solubility in a variety of solvents. When the reaction is conducted in air with Na2PdCl4 as a precursor, no additional Cl is needed to initiate oxidative etching as this ligand will be released from Na2PdCl4 during the reaction. In this synthesis, L-ascorbic acid is used as a reducing agent to ensure the fast reduction of a Pd precursor, which is critical to the formation of thermodynamically favorable species such as truncated octahedrons and MTPs. In a typical protocol, we synthesized truncated octahedrons of Pd by heating 11 mL of an aqueous solution containing 17.4 mM Na2PdCl4, 31 mM L-ascorbic acid, and 87 mM PVP at 100 8C. Figure 3AC shows typical transmission electron microscopy (TEM) images of Pd samples taken at different stages of the reaction. At t 30 min, the sample contained both single-crystal truncated Figure 3. AC) TEM images of Pd samples obtained by heating 11 mL of an aqueous solution octahedrons and MTPs (Fig. 3A). As the containing 17.4 mM Na2PdCl4, 31 mM L-ascorbic acid, and 87 mM PVP at 100 8C for A) 30 min, reaction proceeded to t 1 h 30 min, however, B) 1 h 30 min, and C) 3 h. In (A), twinned particles are indicated by tw. D) HRTEM image of a all the twinned particles disappeared from the single truncated octahedron shown in (C) recorded along the [011] zone axis and the corre can be indexed as {200} and solution (Fig. 3B). During the next 1 h 30 min, sponding FT pattern (inset). The lattice spacings of 1.94 and 2.24 A there was no signicant change in shape, but {111} of fcc Pd, respectively. In the FT pattern, the spots circled and squared can be indexed as remaining truncated octahedrons the {200} and {111} reections, respectively. E) Illustration of the proposed mechanism by which the increased in size until they reached an average single-crystal truncated octahedrons were obtained in high yields. ensemble of ve tetrahedrons with twin-related adjoining faces, a gap of 7.35 8 is generated as the theoretical angle between two {111} planes of a tetrahedron is 70.53 8. As a result, the space must be compensated for by increasing the separation between adjacent atoms, giving rise to internal lattice strain. Similar to a decahedron, internal strains are also involved in closing the gaps formed in an icosahedron, which consists of twenty tetrahedrons. If the MTPs expand in a lateral dimension, the lattice strain will keep increasing and the low surface energy of the {111} facets can no longer remedy the excessive strain energy required to sustain the twinned structures. As a result, MTPs are thermodynamically favored primarily at relatively small sizes. Alternatively, a decahedron can grow along the vefold axis to generate a

FEATURE ARTICLE
192

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2009, 19, 189200

www.afm-journal.de

diameter of about 8 nm. Figure 3C shows a typical TEM image of the sample obtained at t 3 h, which revealed that truncated octahedrons of Pd had become the major product (>95%). The structure of these truncated octahedrons was supported by highresolution TEM (HRTEM) analysis. Figure 3D shows a HRTEM image of a single truncated octahedron recorded along the [011] zone axis and the corresponding Fourier transform (FT) pattern (inset); both data indicate that the Pd truncated octahedron is a piece of single crystal with the exposed facets being {111} and {100} planes. The fringes with lattice spacings of 1.94 and 2.24 A can be indexed as {200} and {111} of fcc Pd, respectively. Both TEM and HRTEM analyses conrmed the absence of Pd nanocrystals with twinned structures in the nal product. The higher reactivity of twinned structures towards oxidative etching can be attributed to the higher density of twin defects on their surfaces, which are much higher in energy relative to the single-crystal regions and thus are more susceptible to an oxidative environment.[13,17] As a result, MTPs are preferentially attacked by the enchant, oxidized, and dissolved into the solution in the early stage of a synthesis (Fig. 3E). During the initial dissolution stage, etching of MTPs could increase the concentration of [PdCl4]2. As the reaction continues, Pd atoms formed by reducing [PdCl4]2 with L-ascorbic acid are added to the existing truncated octahedrons, by which they grow into larger sizes. Our results demonstrate that the oxidative etching can also serve as a powerful means for purifying and thus controlling the shape of Pd nanocrystals for a water-based system.

octahedrons, and decahedrons whose surface is covered by the {111} facets. We synthesized Pd icosahedrons by heating 11 mL of an aqueous solution containing 5.8 mM Na2PdCl4, 28 mM citric acid, and 29 mM PVP at 90 8C for 26 h. Both scanning electron microscopy (SEM) and TEM studies revealed that Pd icosahedrons with edge lengths of approximately 25 nm were obtained as the major product ($80%) in addition to a small amount of octahedrons ($20%), as shown in Figure 4A and B. Interestingly, we found that the shape of Pd nanocrystals was highly sensitive to the concentration of Na2PdCl4. For instance, increasing the concentration of Na2PdCl4 by $25% while carefully keeping the concentration of citric acid and the molar ratio of PVP to Na2PdCl4 the same as in the synthesis of icosahedrons led to the formation of octahedrons as the major species ($90%) in the

FEATURE ARTICLE

4. Citric Acid for Moderate Reduction and Surface Capping in the Formation of Icosahedrons, Octahedrons, and Decahedrons
The multiple-twinned nanocrystals of Pd such as icosahedrons and decahedrons produced during the reaction are often difcult to retain in a typical solution-phase synthesis conducted in air due to the intrinsically corrosive environment, as discussed in Section 3. To remedy this issue, citric acid can be introduced into the reaction. Interestingly, citric acid (or citrate ions) can serve as not only a reducing agent but also a capping agent to stabilize these structures thanks to its strong binding to the {111} facets of Pd.[32,35] In addition, these species can effectively block oxidative etching by competing with oxygen adsorption onto the Pd surface or exhausting the adsorbed oxygen atoms.[32] In this way, it becomes possible to produce Pd icosahedrons and decahedrons. More specically, we demonstrate that the shape of the nal Pd nanocrystals can be controlled by varying the concentrations of Na2PdCl4 and citric acid to selectively produce Pd icosahedrons,

Figure 4. A) SEM and B) TEM images of Pd icosahedrons synthesized by heating 11 mL of an aqueous solution containing 5.8 mM Na2PdCl4, 28 mM citric acid, and 29 mM PVP at 90 8C for 26 h; C) SEM and D) TEM images of Pd octahedrons prepared under the same condition as in (A) except that the concentrations of Na2PdCl4 and PVP were increased to 7.4 and 37 mM, respectively; E) SEM and F) TEM images of Pd decahedrons prepared under the same condition as in (A) except that the concentrations of Na2PdCl4, PVP, and citric acid were increased to 17.4, 87, and 84 mM, respectively (modied with permission from [35], copyright 2007 Wiley-VCH).

Adv. Funct. Mater. 2009, 19, 189200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

193

www.afm-journal.de

product, with $10% other shapes including triangular plates and decahedrons (Fig. 4C and D). These results suggest that formation of octahedrons was more favorable than that of icosahedrons at relatively high Na2PdCl4 precursor concentrations. The reaction condition could be further manipulated to obtain Pd decahedrons by increasing both the Na2PdCl4 and citric acid concentrations beyond those employed for the formation of icosahedrons and octahedrons. Figure 4E and F shows typical SEM and TEM images of Pd decahedrons obtained by increasing the concentrations of Na2PdCl4 and citric acid to 17.4 and 84 mM, respectively. Both images revealed that Pd decahedrons with sizes of 2540 nm were produced as a major product (> 80%). As demonstrated in this work, citric acid favors the formation of Pd nanocrystals enclosed by the {111} facets, such as icosahedrons, octahedrons, and decahedrons. It has been shown by simulation that icosahedral, decahedral, and truncated octahedral clusters are favored for Pd at small (with the number of atoms N < 100), medium (100 < N < 6500), and large sizes (N > 6500), respectively.[4345] When the Pd precursor concentration is low, the seeds are most likely to adopt an icosahedral structure due to the slow addition of Pd atoms and remain small for a long period of time. As a result, Pd icosahedrons can be produced in high yields. When the concentration of Pd precursor is increased, the generation and addition of Pd atoms becomes faster and therefore most seeds will take a decahedral or truncated octahedral shape because of their rapidly increased size. At a relatively low concentration of citric acid, however, formation of decahedrons is less favorable than that of octahedrons because of lattice strain caused by twin defects and their preferential dissolution via oxidative etching in the presence of an O2/Cl pair. As a result, single-crystal octahedrons are more likely to remain in the nal product. To improve the yield of decahedrons, a higher concentration for the capping agent (citric acid or citrate ions in this case) is required to ensure sufcient capping of the {111} facets, which not only reduces the surface energy and thus compensates for the extra strain energy caused by twinning but also efciently protects them from oxidative environment. For these reasons, the decahedrons can be preserved and thus accumulate throughout the reaction.

substantially slowing down the reduction rate. For this purpose, PVP can be used as an ideal reducing agent thanks to its weak reducing power, as described in Section 2, thereby enabling a kinetic control over both nucleation and growth. Among various shapes, 2D anisotropic nanostructures of noble metals such as hexagonal and triangular nanoplates have drawn increasing attention because of their unique optical properties and potential use in chemical and biological sensing. The nanoplates exhibit unique localized surface plasmon resonance (LSPR) features, such as quadrupole resonance peaks that are absent in small nanospheres, and are supposed to be particularly active for surface-enhanced Raman scattering (SERS) thanks to their sharp corners and edges.[4649] To date, a number of different synthetic routes have been developed to generate nanometer- or micrometer-sized thin plates of various noble metals, including light-induced conversion of nanospheres to nanoplates,[50,51] reduction of a metal precursor in the presence of a specic capping agent or surfactant,[52] and mild annealing of self-organized nanocrystals on carbon substrates.[53] Here we demonstrate that Pd nanoplates with hexagonal and triangular shapes can be simply produced by reducing Na2PdCl4 with PVP, without the involvement of additional capping agents or specic substrates. Figure 5A shows a TEM image of Pd nanoplates synthesized by heating 11 mL of an aqueous solution containing 17.4 mM Na2PdCl4 and 87 mM PVP at 100 8C for 3 h. Note that this reaction

FEATURE ARTICLE

5. PVP for Slow Reduction in the Formation of Hexagonal and Triangular Nanoplates
If the reduction becomes considerably slow, both nucleation and growth may deviate from a thermodynamically controlled pathway. This type of synthesis is known as a kinetically controlled process, and the nal nanocrystal shape typically deviates from those favored by thermodynamics (i.e., structures with higher free energies). In one case, thin plates with hexagonal and triangular shapes can be formed, with both top and bottom faces covered by the {111} facets. In practice, kinetically controlled synthesis can be achieved by
Figure 5. A) TEM image of hexagonal and triangular Pd nanoplates synthesized by heating 11 mL of an aqueous solution containing 17.4 mM Na2PdCl4 and 87 mM PVP at 100 8C for 3 h. Note that this reaction condition is the same as in Figure 3C except for the exclusion of Lascorbic acid from the reaction. In this case, PVP serves as a reducing agent. B) HRTEM image taken from the at top face of a single nanoplate and corresponding FT pattern (inset). In the FT pattern, the spots circled and squared can be indexed to the {220} and forbidden 1/ 3{422} reections, respectively, in which the latter indicates the presence of planar defects such as stacking faults in the {111} planes. C) Proposed mechanism for the formation of Pd nanoplates. Pd atoms nucleate to form nuclei with a metastable rhcp structure with the inclusion of stacking faults. At a slow reduction rate, these nuclei can evolve into plate-like seeds, which further grow into hexagonal and triangular nanoplates with their top and bottom faces enclosed by the {111} facets.

194

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2009, 19, 189200

www.afm-journal.de

condition is the same as in Figure 3C except for the exclusion of Lascorbic acid from the reaction. As shown in Figure 5A, the product mainly consists of nanoplates (> 90%) with both hexagonal and triangular shapes and sizes in the range of 50 80 nm. Most of the triangular nanoplates have truncated corners. Figure 5B shows a HRTEM image taken from the at top face of a single nanoplate and the corresponding FT pattern (inset). In the HRTEM image, the fringes with a lattice spacing of 1.4 A can be seen and indexed as {220} of fcc Pd. The FT pattern is composed of spots with a six-fold rotational symmetry, indicating that the top and bottom faces of Pd nanoplates are enclosed by the {111} planes. The spots circled and squared can be indexed to the {220} and forbidden 1/3{422} reections, respectively. Observation of the forbidden spots associated with 1/3{422} diffractions suggests that planar defects such as stacking faults are present in the {111} plane perpendicular to the electron beam.[54] The prevalence of a plate-like morphology in this synthesis can be attributed to the slow reduction rate, which is associated with the weak reducing power of PVP. As an fcc metal, the crystal structure of Pd provides no intrinsic driving force to grow into 2D nanostructures such as thin plates. Compared to a polyhedral nanocrystal of the same volume, a nanoplate with top and bottom faces covered by the {111} facets exhibits a large surface area, and thus its total free energy is relatively high regardless of the surface coverage by the {111} facets. As a result, formation of plate-like morphology is not favored in terms of thermodynamics. To obtain this class of highly anisotropic shapes, one needs to break the cubic symmetry of the lattice. One way to accomplish this is to incorporate planar defects such as stacking faults into the nanocrystals. For an fcc lattice, the stacking sequence of {111} layers should be ABCABCABC. When stacking faults are introduced, however, they disrupt the stacking sequence for one or two layers (e.g., ABCABABC). In the classical nucleation theory, it is assumed that the nucleus takes a spherical shape due to surface tension and the same crystal structure as the bulk solid. However, it has recently been shown by simulation that for fcc crystals, nuclei formed in the early stages of nucleation tend to take a random hexagonal close-packed (rhcp) structure a random mixture of both hexagonal close packing (hcp) and cubic close packing with the inclusion of stacking faults rather than a pure fcc phase because the strain energy caused by stacking faults is low and the rhcp structure is slightly more stable than fcc at this stage.[5557] When the reduction is relatively fast, these nuclei can evolve into polyhedral seeds such as single crystal, truncated octahedrons and vefold twinned decahedrons in an effort to lower the total surface energy, as discussed in Section 3. If the reduction is considerably slowed, however, these nuclei with a metastable rhcp structure could remain small for a long period of time, due to the slow addition of atoms, and gradually evolve into plate-like seeds while retaining their structure characterized by the presence of similar stacking faults in a vertical direction (Fig. 5C). These seeds can further grow into hexagonal and triangular nanoplates via preferential addition of Pd atoms onto their edges because they are bound by a mix of {110} and {100} facets with a higher surface energy than the top and bottom {111} faces. Therefore, the slow reduction rate derived from the weak reducing power of PVP seems to be most important factor in achieving the kinetically controlled synthesis of Pd nanoplates.

6. Bromide as a Capping Agent in Promoting the {100} Facets


The preparation of single-crystal Pd nanocrystals encased with only one type of facet is desirable, particularly for systematic studies of, and applications in, catalysis. Once the seed is xed in terms of twin structure, the nal shape taken by the nanocrystal will be determined by the relative rates at which different crystallographic facets grow, in which the facets with a slower growth rate will be exposed more on the nanocrystals surface. For example, if the fast growing facets correspond to the {111} of a truncated octahedron, the nal crystal shape will be a cube enclosed by the slow growing {100} facets. In contrast, if the fast growing facets correspond to the {100} faces, the nal crystal shape will be an octahedron enclosed by slow growing {111} facets. Such a dynamic evolution can selectively enlarge one set of facets at the expense of others on a nanocrystal. In a solutionphase synthesis, the seeds can grow into nanocrystals with drastically different shapes by controlling the relative growth rates of different facets. In particular, impurities or capping agents can change the order of free energies of different facets through their chemical interaction with a metal surface. This alternation may signicantly affect the relative growth rates of different facets and thus lead to different morphologies for the nal products.[22,24,26,58,59] For example, as we have demonstrated for a Ag system, PVP can serve as a capping agent whose oxygen atoms bind most strongly to the {100} facets of Ag. For single-crystal Ag seeds terminated with only {111} and {100} facets, this preferential capping drives the addition of Ag atoms primarily onto the poorly passivated {111} facets, resulting in the formation of nanocubes enclosed by the {100} facets.[12,60,61] However, PVP is too big to have a capping effect on small Pd nanocrystals. As a result, single-crystalline Pd nanocrystals (typically < 10 nm) tend to take the truncated octahedral shape when prepared in the presence of PVP as shown in Figure 3C.[17] Interestingly, Br ions can provide this function as a small ionic capping agent, capable of preferentially chemisorbing onto the {100} facets of Pd nanocrystals.[26] In this way, Br ions stabilize the {100} facets and thus favor the formation of Pd nanocrystals enclosed by the {100} facets, as for nanobars and nanocubes.

FEATURE ARTICLE

6.1. Synthesis of Nanobars 1D nanostructures of Pd are of particular interest because they are promising building blocks for fabricating nanoscale electronic devices. For instance, it has been shown that Pd can be used for resistance-based detection of hydrogen gas due to its exceptional sensitivity towards hydrogen.[62] In addition, Pd can form reliable and reproducible ohmic contacts with carbon nanotubes (CNTs) as it has a relatively high work function and can easily wet the carbon surface, which makes it useful for CNT-based devices such as eld-effect transistors.[63,64] Here we demonstrate that Pd nanobars with 1D anisotropic structure can be produced in high yields under the fast reduction of a Pd precursor by L-ascorbic acid in the presence of Br ions. During the course of this work, it was found that the selective activation process could initiate anisotropic growth of Pd nanocrystals if the concentration of

Adv. Funct. Mater. 2009, 19, 189200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

195

www.afm-journal.de

Br ions was relatively low, resulting in the formation of nanobars. Unlike nanorods commonly observed in Ag and Au systems,[16,42,65] Pd nanobars are single crystals with a rectangular cross section and bound by the {100} facets. We synthesized Pd nanobars by heating 11 mL of an aqueous solution containing 17.4 mM Na2PdCl4, 31 mM L-ascorbic acid, 87 mM PVP, and 230 mM KBr at 100 8C for 3 h. These reaction conditions are the same as in the synthesis of truncated octahedrons shown in Figure 3C except for the addition of KBr. Figure 6A and B shows TEM images of the resulting product at different magnications. It can be clearly seen that the product mainly consists of nanobars (>95%) with a larger dimension along one direction than those along the other two directions. The width and aspect ratio of the nanobars were 68 nm and 24 nm, respectively. The structure of nanobars was characterized by HRTEM study, an image from which is Figure 6C. The HRTEM image of a single nanobar recorded along the [001] zone axis displayed well-resolved, continuous fringes with a lattice spacing of 1.94 A, which can be indexed as Figure 6. A and B) TEM images of Pd nanobars prepared under the same condition as in {200} of fcc Pd. The corresponding FT pattern is Figure 3C except that the reaction was conducted in the presence of 230 mM KBr. C) HRTEM composed of spots with a square symmetry. These image of a single nanobar and the corresponding FT pattern (inset). The lattice spacing of results conrm that the Pd nanobar is a piece of 1.94 A can be indexed as {200} of Pd. In the FT pattern, the spots circled and squared can be single crystal enclosed by the {100} facets. indexed to the {200} and {220} reections, respectively. D) Schematic illustration of the As discussed in Section 3, fast reduction mechanism responsible for the formation of nanobars. coupled with oxidative etching yields truncated octahedrons, which are expected to further evolve into nanocubes in the presence of Br ions due to their preferential chemisorption on the {100} facets of truncated Figure 7A and B shows TEM images of the Pd sample prepared octahedrons. At a relatively low concentration of Br ions, the under the same condition as in Figure 6A except that the formation of nanobars can be attributed to anisotropic growth of concentration of KBr was increased to 460 mM. It can be seen that cubic Pd nanocrystals driven by selective activation of one of their Pd nanocubes with an average size of 10 nm were formed as the six {100} faces (Fig. 6D).[26] When the concentration of Br ions is major product ($95%) in addition to a small amount of vefold low, the surface of a nanocube is most likely coated by a relatively twinned pentagonal nanorods ($5%) with a diameter of about thin layer of Br ions. In this case, oxidative etching could remove 8 nm and lengths up to 200 nm. The HRTEM image of a single some of the Br ions from the Pd surface. As in the case of nanocube (Fig. 7C) clearly shows continuous fringes with a corrosion of Pd nanocubes by pitting process or galvanic period of 1.94 A, which is consistent with the {200} lattice spacing replacement between Ag nanocubes and HAuCl4, the oxidative of fcc Pd. The corresponding FT pattern (Fig. 7C, inset) reveals a etching can selectively take place on only one of the six {100} square symmetry for the spots. These results indicate that the Pd faces, making this particular face more active than others. As a nanocube is a piece of single crystal bound by the {100} facets. result, this localized oxidative etching creates a favorable site for Figure 7D represents a HRTEM image taken from the end of a the subsequent addition of Pd atoms (i.e., growth) and thus single pentagonal nanorod, which reveals that the {111} twin facilitates the preferential growth on this face. This selective boundary is straight and continuous along the entire longitudinal activation process could break the symmetry of a nanocube and axis of the nanorod. The corresponding FT pattern (Fig. 7D, inset) eventually lead to its anisotropic growth into a nanobar. can be interpreted as the overlapping of the [100] and [112] zone axes of fcc Pd. The presence of spots with a square symmetry indicates that the side faces of the nanorod are bound by the {100} facets. 6.2. Synthesis of Nanocubes In the formation of a cubic morphology, the presence of Br ions at a relatively high concentration is signicant. In this case, all the faces of a Pd nanocube could be stabilized by a monolayer Noble metal nanocrystals with a cubic shape have recently been of Br ions, which could effectively block the localized oxidative synthesized by a polyol method in the presence of a polymeric or [12,21,22] etching and thus prohibit its anisotropic growth into a nanobar. At ionic capping agent such as PVP or Ag ions. Here we the same time, some of the decahedral seeds may evolve into show that Pd nanocubes can be produced simply by modifying pentagonal nanorods under these reaction conditions. It has been the reaction conditions used for the synthesis of Pd nanobars.

FEATURE ARTICLE
196

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2009, 19, 189200

www.afm-journal.de

concentration, 1D anisotropic growth of nanocubes into nanobars could be greatly suppressed by lowering the reaction temperature. When the reaction was performed at 80 8C while keeping other reaction parameters the same as in Figure 6A, nanobars with an aspect ratio of 24 were rarely found in the product and most of the Pd nanocrystals were of a cubic shape with sizes of 1012 nm, although some of them were slightly elongated along one direction (Fig. 8A and B). As demonstrated in our previous studies, the etching power of the O2/Cl pair is reduced at low reaction temperatures.[33] In this case, the localized oxidative etching could be signicantly eliminated and, as a result, the initially formed nanocubes could maintain their cubic morphology and accumulate throughout the reaction without growing into nanobars via selective activation processes. In another demonstration, we found that introduction of Br ions into a kinetically controlled process (with PVP as a reducing agent) could enable the production of Pd nanocubes at high yields. Figure 8C and D shows TEM images of Pd nanocubes with an average size of $10 nm prepared under the same condition as in the synthesis of the hexagonal and triangular nanoplates shown in Figure 5A except for the presence of 460 mM KBr in the reaction solution. It is clear that the Br ions induced a cubic morphology for the resulting Pd nanocrystals. The formation of nanocubes in this synthesis implies that singleFigure 7. A and B) TEM images of a Pd sample prepared under the same condition as in crystal seeds are involved in the nucleation step, in Figure 6A except that the concentration of KBr was increased to 460 mM. C and D) HRTEM which Br ions could break the rhcp structure of images and the corresponding FT patterns (insets) of a single nanocube and pentagonal initially formed nuclei through their chemical nanorod, respectively. The lattice spacings of 1.94 and 1.4 A can be indexed as {200} and interaction with Pd surfaces and lead to its {220} of Pd, respectively. In the FT patterns, the spots circled, squared, and triangled can be transformation into an fcc structure. During the indexed to the {200}, {220} and {112} reections, respectively. E) Schematic illustration of growth step, these single-crystal seeds could the mechanism responsible for the formation of pentagonal nanorods. further evolve into nanocubes via preferential chemisorption of Br ions on the {100} facets. It should be pointed out that pentagonal nanorods are absent in the product because the multiple-twinned seeds are suggested that the twin defects in a decahedron, which bring not favored in this kinetically controlled synthesis. about internal strain in the lattice, are largely responsible for the anisotropic growth of pentagonal nanorods.[16,42] In principle, when atoms in a decahedron are located far from the central axis, the strain in the lattice will be extremely high. Thus, the strain will 7. Seeded Growth be greatly increased if a decahedron grows in lateral dimensions. In contrast, elongation of a decahedron in a direction parallel to With regards to crystal growth, pre-formed nanocrystals with the twin planes does not increase the lattice strain. Consequently, well-dened shapes can serve as primary sites, i.e., seeds, for the decahedrons can preferentially grow along the vefold axis heterogeneous nucleation of added metal atoms. When the added into pentagonal nanorods to retain a low strain energy. This kind atoms have the same crystal structure and lattice constant as the of anisotropic growth requires the presence of a capping agent for seed, the crystal structure of the seed is transferred to the entire the stabilization of newly formed {100} side faces. Accordingly, nanocrystal via epitaxial growth, although the nal shape of a when a sufcient amount of Br ions are introduced into the nanocrystal may deviate from that of the initial seed due to the reaction at the fast reduction rate, some of the decahedral seeds crystal habit governed by the growth rates of different crystalcan quickly evolve into pentagonal nanorods with their side {100} lographic facets. This so-called seeded growth approach offers an faces being stabilized by chemisorbed Br ions. alternative way for synthesizing uniform nanocrystals by taking By tuning the experiment condition, we could further improve advantage of the absence of homogenous nucleation during the the yield of Pd nanocubes. It was found that at a low Br ion growth step. Although seeded growth has been proven to be

FEATURE ARTICLE
197

Adv. Funct. Mater. 2009, 19, 189200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.afm-journal.de

of 1.94 and 2.24 A can be indexed as {200} and {111} of Pd, respectively. When the concentrations of Na2PdCl4 and PVP were increased to 21 and 105 mM, respectively, with other reaction parameters being kept the same, Pd nanocrystals with a regular octahedral shape were obtained in a high yield (>95%), as shown in Figure 9D and E. Recently, it was reported that Ag and Au nanocrystals with a cubic morphology could evolve into truncated and then regular octahedrons as more atoms were added to the {100} faces of nanocubes.[69,71] The shape conversion of Pd nanocrystals observed in the present work might also occur through a similar process. In this case, the initial addition of Pd atoms on a nanocube results in the generation of the {111} faces at each corner of the nanocube. In the absence of specic capping agents, the continued addition of Pd atoms prefers to the {100} faces of the Pd nanocrystal as the relative surface energies of the low-index crystallographic facets for an fcc metal are in the order of g{111} < g{100} < g{110}. As the crystal growth continues, the surface fraction of the slow growing {111} faces Figure 8. A and B) TEM images of Pd nanocubes prepared under the same condition as in Figure 6A except that the reaction temperature was lowered to 80 8C. C and D) TEM images increases at the expense of the faster growing of Pd nanocubes prepared under the same condition as in Figure 5A except that the reaction {100} faces. At a relatively low concentration of Na2PdCl4, this process yields truncated octahewas conducted in the presence of 460 mM KBr. drons bound by both the {111} and {100} facets. When the concentration of Na2PdCl4 is further increased, the {100} facets completely disappear via continued extremely powerful in Au, Ag, and Pt systems,[6670] little addition of Pd atoms, leaving behind regular octahedrons attention has been paid to the seeded growth of Pd nanocrystals. enclosed by the {111} facets. We expect that this approach based Here we demonstrate that Pd nanocrystals can be completely on seeded growth could be utilized to further control the shape converted from cubes to truncated and regular octahedrons via and size of Pd nanocrystals by taking advantage of a rich variety of seeded growth. The Pd nanocubes with an average size of shapes that have been already achieved. $10 nm shown in Figure 8C were rst prepared as described in Section 6.2 and then used as the seeds to initiate further nanocrystal growth when additional Pd precursor was added and then slowly reduced by PVP. As more Pd atoms were added to the Pd nanocubes, they grew into truncated or regular octahedrons depending on the amount of Na2PdCl4 added to the reaction (Fig. 9A). In a typical synthesis, we obtained Pd truncated octahedrons by adding 1 mL of the as-prepared Pd nanocube suspension to 9 mL of an aqueous solution containing 10.5 mM Na2PdCl4 and 52.5 mM PVP and heating at 90 8C for 3 h. As shown in Figure 9B, the resulting product mainly contained Pd nanocrystals with a truncated octahedral shape (>95%) and edge lengths of 1215 nm. HRTEM imaging of a single truncated octahedron recorded along the Figure 9. A) Schematic illustration of seeded growth of Pd octahedrons with and without truncation [011] zone axis and the corresponding FT at corners from cubic Pd seeds. B) TEM and C) HRTEM images of Pd truncated octahedrons pattern indicated that the Pd truncated obtained by adding 1 mL of the as-prepared Pd nanocube solution (shown in Fig. 8C) to 9 mL of an octahedron was a piece of single crystal with aqueous solution containing 10.5 mM Na2PdCl4 and 52.5 mM PVP and heating at 90 8C for 3 h. the exposed facets of both {111} and {100} D) TEM and E) HRTEM images of Pd octahedrons prepared under the same condition as in (B) (Fig. 9C). The fringes with lattice spacings except that the concentrations of Na2PdCl4 and PVP were increased to 21 and 105 mM, respectively.

FEATURE ARTICLE
198

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2009, 19, 189200

www.afm-journal.de

8. Concluding Remarks and Future Direction


We have demonstrated the capability and feasibility of using a water-based system for the high-yield production of Pd nanocrystals with a rich variety of shapes, including truncated octahedrons, icosahedrons, octahedrons, decahedrons, hexagonal and triangular thin plates, rectangular bars, and cubes. The success of this approach depends on a number of parameters such as the reduction kinetics, oxidative etching, and surface capping. These parameters could be combined to provide an effective route to maneuver both the twin structures of seeds and the surface facets, which are two key factors in determining the nal shape of a nanocrystal. Specically, the twin structures of seeds could be controlled by introducing different reducing agents and thus manipulating the reduction kinetics. When the reduction rate was relatively fast, the reaction was under thermodynamic control and, as a result, single-crystal and multiple-twinned seeds were dominant in the nucleation stage. The distribution of these seeds could be further altered by employing or blocking oxidative etching. In one case, the multiple-twinned seeds could be selectively removed from the solution by the O2/Cl etchant, leaving behind single-crystal truncated octahedrons as a major product. Meanwhile, citric acid (or citrate ions) facilitated the formation of icosahedrons and decahedrons by blocking the oxidative etching through its strong binding to the {111} surfaces. When the reduction rate was considerably slowed, the reaction became kinetically controlled. In this case, plate-like seeds with planar defects such as stacking faults formed at the initial nucleation stage and then grew into hexagonal and triangular nanoplates that deviated from the thermodynamically favored shapes. In most cases, the binding selectivity of a capping agent had a profound impact on the crystal habit of Pd nanocrystals. They could interact more strongly with specic facets and thus changed the order of free energies for different crystallographic facets. This kind of chemisorption or surface capping provided a means for controlling the relative growth rates of different facets and thus the nal shape of Pd nanocrystals. As demonstrated in this work, Br ions bind most strongly to the {100} facets of Pd, resulting in the formation of nanobars and nanocubes. Controlling the shape of noble metal nanocrystals provides a powerful tool for tailoring their electronic, plasmonic, and catalytic properties. For instance, we have recently demonstrated that Pd nanoplates of 45 nm in edge length exhibit LSPR peaks redshifted to the visible region (up to 610 nm)[31] compared to small Pd nanocrystals (typically < 10 nm in size) whose LSPR peaks are located in the UV region. Since the position of the LSPR peak determines the wavelength of excitation for the maximum electromagnetic eld enhancement, these Pd nanoplates with redshifted LSPR peaks are expected to interact more intensely with the laser of a commercial Raman spectrometer (514 and 785 nm) and, as a result, exhibit stronger SERS activity. Another important property of Pd nanocrystals is their catalytic activity and selectivity towards chemical or electrochemical reactions. For example, it was recently found that in the electrocatalytic oxidation of formic acid with Pd nanocrystals, the peak current measured for Pd cubes was ve times higher than that for Pd octahedrons.[9] This difference in activity was attributed to the different oxidation rate of formic acid on the {100} and {111}

facets of Pd and illustrates the dependence of the electrochemical activity of Pd nanocrystals on the shape. However, the shape effect of Pd nanocrystals on the activity and selectivity towards various chemical reactions such as hydrogenation and cross-coupling reactions still remains largely unexplored. We expect that the methodology for controlling the shape of Pd nanocrystals demonstrated in this work could provide a great opportunity to systematically evaluate their shape-dependent catalytic properties, as well as to fully explore their applications in elds of sensing, storage of hydrogen gas, SERS, and fabrication of nanoelectronic devices.

FEATURE ARTICLE

Acknowledgements
This work was supported in part by NSF (both DMR-0451788 and DMR0804088), ACS (PRF, 44353-AC10), and a 2006 Directors Pioneer Award from NIH (5DP1D000798). B. L. was also partially supported by the Korea Research Foundation Grant funded by the Korean Government (KRF-2006352-D00067). J. T. and Y. Z. were supported by the U.S. DOE/BES (DEAC02-98CH10886). Received: September 26, 2008 Published online: December 12, 2008

a, nez-Arias, L. N. Salamanca, J. M. Coronado, [1] M. Fernandez-Garc A. Mart J. A. Anderson, J. C. Conesa, J. Soria, J. Catal. 1999, 187, 474. [2] Y. Nishihata, J. Mizuki, T. Akao, H. Tanaka, M. Uenishi, M. Kimura, T. Okamoto, N. Hamada, Nature 2002, 418, 164. [3] J. M. Thomas, B. F. G. Johnson, R. Raja, G. Sankar, P. A. Midgley, Acc. Chem. Res. 2003, 36, 20. [4] L. Schlapbach, A. Zuttel, Nature 2001, 414, 353. [5] M. T. Reetz, E. Westermann, Angew. Chem, Int. Ed. 2000, 39, 165. [6] Y. Li, X. M. Hong, D. M. Collard, M. A. El-Sayed, Org. Lett. 2000, 2, 2385. [7] S.-W. Kim, M. Kim, W. Y. Lee, T. Hyeon, J. Am. Chem. Soc. 2002, 124, 7642. [8] R. Narayanan, M. A. El-Sayed, Nano Lett. 2004, 4, 1343. [9] S. E. Habas, H. Lee, V. Radmilovic, G. A. Somorjai, P. Yang, Nat. Mater. 2007, 6, 692. [10] K. M. Bratlie, H. Lee, K. Komvopoulos, P. Yang, G. A. Somorjai, Nano Lett. 2007, 7, 3097. [11] C. Wang, H. Daimon, T. Onodera, T. Koda, S. Sun, Angew. Chem, Int. Ed. 2008, 47, 3588. [12] Y. Sun, Y. Xia, Science 2002, 298, 2176. [13] B. Wiley, T. Herricks, Y. Sun, Y. Xia, Nano Lett. 2004, 4, 1733. [14] T. Herricks, J. Chen, Y. Xia, Nano Lett. 2004, 4, 2367. [15] F. Kim, S. Connor, H. Song, T. Kuykendall, P. Yang, Angew. Chem, Int. Ed. 2004, 43, 3673. [16] B. Wiley, Y. Sun, Y. Xia, Langmuir 2005, 21, 8077. [17] Y. Xiong, J. Chen, B. Wiley, Y. Xia, S. Aloni, Y. Yin, J. Am. Chem. Soc. 2005, 127, 7332. [18] Y. Xiong, J. Chen, B. Wiley, Y. Xia, Y. Yin, Z.-Y. Li, Nano Lett. 2005, 5, 1237. [19] Y. Xiong, J. M. McLellan, J. Chen, Y. Yin, Z.-Y. Li, Y. Xia, J. Am. Chem. Soc. 2005, 127, 17118. [20] J. Chen, T. Herricks, Y. Xia, Angew. Chem, Int. Ed. 2005, 44, 2589. [21] H. Song, F. Kim, S. Connor, G. A. Somorjai, P. Yang, J. Phys. Chem. B 2005, 109, 188. [22] D. Seo, J. C. Park, H. Song, J. Am. Chem. Soc. 2006, 128, 14863. [23] B. J. Wiley, Y. Xiong, Z.-Y. Li, Y. Yin, Y. Xia, Nano Lett. 2006, 6, 765. [24] B. J. Wiley, Y. Chen, J. M. McLellan, Y. Xiong, Z.-Y. Li, D. Ginger, Y. Xia, Nano Lett. 2007, 7, 1032.

Adv. Funct. Mater. 2009, 19, 189200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

199

www.afm-journal.de

[25] B. Wiley, Y. Sun, Y. Xia, Acc. Chem. Res. 2007, 40, 1067. [26] Y. Xiong, H. Cai, B. J. Wiley, J. Wang, M. J. Kim, Y. Xia, J. Am. Chem. Soc. 2007, 129, 3665. [27] Y. Xiong, Y. Xia, Adv. Mater. 2007, 19, 3385. [28] C. Li, K. L. Shuford, Q.-H. Park, W. Cai, Y. Li, E. J. Lee, S. O. Cho, Angew. Chem, Int. Ed. 2007, 46, 3264. [29] S. E. Skrabalak, B. J. Wiley, M. Kim, E. V. Formo, Y. Xia, Nano Lett. 2008, 8, 2077. [30] I. Washio, Y. Xiong, Y. Yin, Y. Xia, Adv. Mater. 2006, 18, 1745. [31] Y. Xiong, I. Washio, J. Chen, H. Cai, Z.-Y. Li, Y. Xia, Langmuir 2006, 22, 8563. [32] Y. Xiong, J. M. McLellan, Y. Yin, Y. Xia, Angew. Chem, Int. Ed. 2007, 46, 790. [33] Y. Xiong, H. Cai, Y. Yin, Y. Xia, Chem. Phys. Lett. 2007, 440, 273. [34] Y. Xiong, I. Washio, J. Chen, M. Sadilek, Y. Xia, Angew. Chem, Int. Ed. 2007, 46, 4917. [35] B. Lim, Y. Xiong, Y. Xia, Angew. Chem, Int. Ed. 2007, 46, 9279. [36] B. Lim, J. Wang, P. H. C. Camargo, M. Jiang, M. J. Kim, Y. Xia, Nano Lett. 2008, 8, 2535. [37] B. Lim, P. H. C. Camargo, Y. Xia, Langmuir 2008, 24, 10437. [38] H. Lee, S. E. Habas, S. Kweskin, D. Butcher, G. A. Somorjai, P. Yang, Angew. Chem, Int. Ed. 2006, 45, 7824. [39] A. Henglein, M. Giersig, J. Phys. Chem. B 2000, 104, 6767. [40] M. Maillard, P. Huang, L. Brus, Nano Lett. 2003, 3, 1611. [41] B. Wiley, Y. Sun, B. Mayers, Y. Xia, Chem. Eur. J. 2005, 11, 454. [42] Y. Sun, B. Mayers, T. Herricks, Y. Xia, Nano Lett. 2003, 3, 955. [43] F. Baletto, R. Ferrando, Phys. Rev. B 2001, 63, 155408. [44] F. Baletto, R. Ferrando, A. Fortunelli, F. Montalenti, C. Mottet, J. Chem. Phys. 2002, 116, 3856. [45] F. Baletto, R. Ferrando, Rev. Mod. Phys. 2005, 77, 371. [46] J. E. Millstone, S. Park, K. L. Shuford, L. Qin, G. C. Schatz, C. A. Mirkin, J. Am. Chem. Soc. 2005, 127, 5312. [47] S. S. Shankar, A. Rai, A. Ahmad, M. Sastry, Chem. Mater. 2005, 17, 566. [48] C. S. Ah, Y. J. Yun, H. J. Park, W.-J. Kim, D. H. Ha, W. S. Yun, Chem. Mater. 2005, 17, 5558. [49] J. E. Millstone, G. S. Metraux, C. A. Mirkin, Adv. Funct. Mater. 2006, 16, 1209.

[50] R. Jin, Y. Cao, C. A. Mirkin, K. L. Kelly, G. C. Schatz, J. G. Zheng, Science 2001, 294, 1901. [51] R. Jin, Y. C. Cao, E. Hao, G. S. Metraux, G. C. Schatz, C. A. Mirkin, Nature 2003, 425, 487. [52] Y. Sun, Y. Xia, Adv. Mater. 2003, 15, 695. [53] A. Courty, A.-I. Henry, N. Goubet, M.-P. Pileni, Nat. Mater. 2007, 6, 900. [54] C. Lofton, W. Sigmund, Adv. Funct. Mater. 2005, 15, 1197. [55] J. Zhu, M. Li, R. Rogers, W. Meyer, R. H. Ottewill, STS-73 Space Shuttle Crew. W. B. Russel, P. M. Chaikin, Nature 1997, 387, 883. [56] S. Auer, D. Frenkel, Nature 2001, 409, 1020. [57] U. Gasser, E. R. Weeks, A. Schoeld, P. N. Pusey, D. A. Weitz, Science 2001, 292, 258. [58] S. Maksimuk, X. Teng, H. Yang, Phys. Chem. Chem. Phys. 2006, 8, 4660. [59] S. Maksimuk, X. Teng, H. Yang, J. Phys. Chem. C 2007, 111, 14312. [60] S. H. Im, Y. T. Lee, B. Wiley, Y. Xia, Angew. Chem, Int. Ed. 2005, 44, 2154. [61] A. R. Siekkinen, J. M. McLellan, J. Chen, Y. Xia, Chem. Phys. Lett. 2006, 432, 491. [62] F. Favier, E. C. Walter, M. P. Zach, T. Benter, R. M. Penner, Science 2001, 293, 2227. [63] A. Javey, J. Guo, Q. Wang, M. Lundstrom, H. Dai, Nature 2003, 424, 654. [64] D. Mann, A. Javey, J. Kong, Q. Wang, H. Dai, Nano Lett. 2003, 3, 1541. [65] C. J. Murphy, T. K. Sau, A. M. Gole, C. J. Orendorff, J. Gao, L. Gou, S. E. Hunyadi, T. Li, J. Phys. Chem. B 2005, 109, 13857. guez[66] A. Sanchez-Iglesias, I. Pastoriza-Santos, J. Perez-Juste, B. Rodr Gonzalez, F. J. G. De Abajo, L. M. Liz-Marzan, Adv. Mater. 2006, 18, 2529. guez-Gonzalez, J. Pacico, I. Pastoriza-Santos, J. [67] E. Carbo-Argibay, B. Rodr Perez-Juste, L. M. Liz-Marzan, Angew. Chem, Int. Ed. 2007, 46, 8983. [68] C. Xue, J. E. Millstone, S. Li, C. A. Mirkin, Angew. Chem, Int. Ed. 2007, 46, 8436. [69] D. Seo, C. I. Yoo, J. C. Park, S. M. Park, S. Ryu, H. Song, Angew. Chem, Int. Ed. 2008, 47, 763. [70] M. A. Mahmoud, C. E. Tabor, Y. Ding, Z. L. Wang, M. A. El-Sayed, J. Am. Chem. Soc. 2008, 130, 4590. [71] A. Tao, P. Sinsermsuksakul, P. Yang, Angew. Chem, Int. Ed. 2006, 45, 4597.

FEATURE ARTICLE
200

2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2009, 19, 189200

Das könnte Ihnen auch gefallen