High Density Catalytic Hot Spots in Ultrafine Wavy Nanowires

May 29, 2014 - design of ultrafine wavy nanowires (WNWs) with a high density of accessible structural defects/grain boundaries as highly active cataly...
0 downloads 0 Views 8MB Size
Download PDF
Letter pubs.acs.org/NanoLett

High Density Catalytic Hot Spots in Ultrafine Wavy Nanowires Xiaoqing Huang,† Zipeng Zhao,† Yu Chen,† Chin-Yi Chiu,† Lingyan Ruan,† Yuan Liu,† Mufan Li,‡ Xiangfeng Duan,‡,§ and Yu Huang*,†,§ †

Department of Materials Science and Engineering, University of California, Los Angeles, California 90095, United States Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095, United States § California NanoSystems Institute, University of California, Los Angeles, California 90095, United States ‡

S Supporting Information *

ABSTRACT: Structural defects/grain boundaries in metallic materials can exhibit unusual chemical reactivity and play important roles in catalysis. Bulk polycrystalline materials possess many structural defects, which is, however, usually inaccessible to solution reactants and hardly useful for practical catalytic reactions. Typical metallic nanocrystals usually exhibit well-defined crystalline structure with few defects/grain boundaries. Here, we report the design of ultrafine wavy nanowires (WNWs) with a high density of accessible structural defects/grain boundaries as highly active catalytic hot spots. We show that rhodium WNWs can be readily synthesized with controllable number of structural defects and demonstrate the number of structural defects can fundamentally determine their catalytic activity in selective oxidation of benzyl alcohol by O2, with the catalytic activity increasing with the number of structural defects. X-ray photoelectron spectroscopy (XPS) and cyclic voltammograms (CVs) studies demonstrate that the structural defects can significantly alter the chemical state of the Rh WNWs to modulate their catalytic activity. Lastly, our systematic studies further demonstrate that the concept of defect engineering in WNWs for improved catalytic performance is general and can be readily extended to other similar systems, including palladium and iridium WNWs. KEYWORDS: noble metal, rhodium, nanowires, structural defect, catalytic oxidation

N

materials with tunable structural defects/grain boundaries that can be accessed by catalytic reactions. Figure 1 schematically illustrates the most widely investigated structures for catalysis to date. The typical metal nanostructures have very large surface area, but usually have well-defined crystalline structure with few grain boundaries/structural defects (Figure 1a). The traditional bulk polycrystalline materials have a large number of grain boundaries/structural defects, which are, however, mostly embedded in the bulk materials and not accessible by reactant molecules, and therefore less useful for practical catalysis. In general, the engineered nanostructures with a high population of solution accessible structural defects/grain boundaries could allow for the creation of highly active catalysts. To this end, we believe ultrafine one-dimensional nanowires represent an ideal structure because of their large high surface to bulk ratio and more importantly the ability to engineer the structural defects/ grain boundaries along the longitudinal direction of the nanowires (Figure 1b). To test this hypothesis, here we explore ultra fine wavy nanowires (WNWs) with variable and tunable structural defects

anostructured noble metals play a critical role in a wide range of heterogeneous catalytic reactions.1−4 Understanding the factors governing the catalytic performance of noble metal nanostructures is of both fundamental and practical importance for heterogeneous catalysis. In general, it is believed that the catalytic performance of metal nanostructures is highly sensitive to their size that controls surface area, to their shape that determines surface atomic arrangement/coordination, and to their structural defects.5−8 During the past decades, robust synthetic techniques have been developed for the synthesis of diverse noble metal nanostructures with well-controlled sizes and shapes,9−14 which has allowed systematic investigation of the size- and shape-effects on catalytic properties.15−25 The role of structural defects on the catalytic properties of noble metal nanoparticles, however, is much less explored. It has been well recognized that structural defects/grain boundaries can exhibit unusual activities in catalysis.26−28 However, traditional research in this area is typically limited to the structural defects/grain boundaries on the surface of bulk materials, which is, although fundamentally interesting, not useful for practical catalytic reactions in bulk solutions because of relatively small number of accessible active sites. A key challenge in exploiting the structural defects/grain boundaries for efficient catalysis is that the extreme difficulties in creating © 2014 American Chemical Society

Received: March 27, 2014 Revised: May 20, 2014 Published: May 29, 2014 3887

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

Letter

hydroformylation, and hydrocarbonylation reactions.29−34 Through a nanoparticle attachment mechanism,35−37 we show that ultrafine Rh WNWs can be prepared with systematically variable density of grain boundaries/structural defects by controlling the growth duration. We found that the ultrathin Rh WNWs exhibit high activity in the selective oxidation of benzyl alcohol under mild condition by O2 gas, and their performances are highly dependent on the population of structural defects, with catalytic activity increasing with the number of structural defects. X-ray photoelectron spectroscopy (XPS) and cyclic voltammograms (CVs) studies demonstrate that the chemical state of surface atoms of the ultrathin Rh WNWs can be significantly altered by the high degree of structural defects, which is believed to be the main contributor to the enhanced catalytic activity. The generality of the defect engineering in ultrafine WNWs for improve catalytic activity was further demonstrated with other metals, such as palladium and iridium WNWs. The ultrathin Rh WNWs were prepared by reduction of sodium hexachlororhodate(III) (Na3RhCl6) using ethylene glycol (EG) as solvent and sodium ascorbate (NaAA) as reducing agent, polyvinylpyrrolidone (PVP) as surfactant, and sodium iodide (NaI) as a structure-directing agent. In a typical preparation, Na3RhCl6, PVP, NaI, NaAA, H2O, and EG were mixed in a vial and ultrasonicated for around 5 min to result in a homogeneous mixture (see Supporting Information for

Figure 1. Schematic illustration of different structures. (a) A schematic illustration of the typical and well-defined nanocrystals. (b) A schematic illustration of a wavy nanowire for creating a high population of accessible structural defects.

that can be accessed by catalytic reaction for a systematic investigation of the correlation among structural, chemical, and catalytic properties. We focus on Rh WNWs as an initial example because Rh nanocatalysts have been reported as efficient catalysts in a wide range of oxidation, hydrogenation,

Figure 2. Morphology and structure analyses for the ultrathin rhodium wavy nanowires prepared in a 120 min reaction (Rh WNWs-120 min). Representative low-magnification TEM (a), high-magnification TEM (b), HRTEM (d, e, f) images, and the PXRD pattern (c) of the ultrathin Rh WNWs-120 min. SAED pattern of ultrathin Rh WNWs-120 min is shown in the inset of (a). Red arrows in (d) highlight the presence of distinct defects/grain boundaries. Dashed lines in Figure (e) and (f) highlight the presence of twin defects cross-sectioning the nanowire. 3888

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

Letter

Figure 3. Analyses of the intermediates to the ultrathin Rh WNWs. (a) TEM images of the Rh WNWs intermediates collected from the reactions with various reaction times. (b) Changes of the size (particles) or diameter (nanowires) distribution as a function of reaction time. (c) Statistics of the length of the products as a function of reaction time. (d) Reduction of Na3RhCl6 during the reaction calculated from the ICP-AES measurements. (e) HRTEM images (Please see Supporting Information Figure S10 for enlarged images), (f) statistical analysis of the number of defects per nanostructure, and (g) the calculations of the relative total number of defects of the Rh WNWs intermediates collected at various reaction times. The data in (g) were generated by dividing data in (f) by those in (c).

details). The mixture was then heated to 170 °C and kept at 170 °C for 2 h in an oil bath before cooling down to room temperature. The resulting colloidal products were collected by centrifugation and washed five times with an ethanol/acetone mixture. To achieve fine control over the synthetic process and produce high yield of Rh WNWs, an extensive set of control experiments were conducted to optimize the synthetic conditions (Supporting Information Figure S1−5). The morphologies of the prepared Rh nanostructures were initially determined by transmission electron microscopy (TEM) (Figure 2a−b, Supporting Information Figure S6). It is evident that the nanowires were the dominant products with a typical yield approaching 100%. Those nanowires were frequently bent on TEM grid, indicating their highly flexible nature. The Rh nanowires exhibited high aspect ratios with the length in the range of tens of nanometers and the diameter as thin as ∼2 nm. It is noticed that all nanowires surveyed exhibit

an uneven ultrathin diameter along its entire length with a highly wavy morphology or rough surface. We therefore refer them as Rh wavy nanowires (Rh WNWs) herein. Despite their ultrathin nature, the as-prepared Rh WNWs are highly crystalline. The selected-area electron diffraction (SAED) pattern (Figure 2a, right inset) and powder X-ray diffraction pattern (PXRD, Figure 2c) shows that the Rh WNWs have a face-center-cubic (fcc) structure that is typical for noble metals. To further evaluate their structural characteristics, the detailed structure of the Rh WNWs was investigated by highresolution TEM (HRTEM). As illustrated in the HRTEM image (Figure 2d), the majority of displayed facets show lattice fringes with interplanar spacing of 0.22, consistent with {111} plane of fcc Rh. The lattice fringes are not continuous and their orientations vary mainly due to highly wavy morphology of the nanowire. A survey of tens of Rh WNWs by HRTEM revealed that, multiple convex/concave, bent and relatively straight 3889

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

Letter

complete within the first 20 min. Figure 3e and Supporting Information Figure S10 show HRTEM images depicting the structure transformation of the Rh WNWs. Collectively, three distinct stages of growth can be identified: (a) In the initial stage of the growth, only small particles were found when the Rh precursor began to reduce. After the reaction had proceeded for 20 min, larger nanoparticles with average size of 2.2 nm were found in the solution, which were likely formed through continuous growth of the small nanoparticles. (b) With the depletion of Rh precursor after 20 min, the growth of nanoparticles stops and the particle attachment occurs, which is dominated by nanoparticles interacting with each other to form wavy-like nanowires. During the attachment of the particles, their wavy feature became more and more prominent with increasing length of the nanowires. Highly wavy nanowires formed with a highest population of structural defect was obtained by the end of 120 min-reaction, as also confirmed by the statistics of number of structural defects per structure (Figure 3f). (c) Because of the relatively low energy barrier, the Rh atoms around the joints can go through an Oswald ripening process and diffuse along the nanowire to result in relatively smooth surfaces under the high synthesis temperature, transforming highly wavy nanowires to relatively smooth nanowires with further increasing reaction time (240 and 300 min).36 It is evident that the local surface roughness decreased substantially at this stage, even though the nanowires remain to exhibit a high degree of curvature (Figure 3e). Statistical analysis indicates that number of grain boundaries is significantly reduced (Figure 3f−g and Supporting Information Figure S10). In this way, the Rh WNWs can be synthesized with tunable degree of particle attachment and thereby the variable structural defects by manipulating the reaction duration. This well-understood growth mechanism of WNWs presented here can be easily extended to the synthesis of other similar noble metal WNWs (Figure 6 and Supporting Information Figure S11−12). The successful preparation of ultrathin Rh WNWs with controlled structural defects can readily enable us to investigate the structural defect-dependent catalytic activity. The Rh-based nanomaterials are highly active catalysts for a wide range of organic reactions. As an example, we chose the benzyl alcohol oxidation by O2 gas to examine catalyst efficiency of the Rh WNWs. Benzyl alcohol oxidation product benzaldehyde is important intermediate in pharmaceuticals, dyes, food preservative and so on.38,39 In a typical catalytic reaction, benzyl alcohol is transformed into benzaldehyde with O2 gas over the Rh catalysts in an aqueous solution at 90 °C and 1 atm O2 gas. To investigate the structural defect enhanced catalytic performance in Rh WNWs, the same amount of intermediate of Rh WNWs grown from 20 min, 30 min, 60 min, 90 min, 120 min, 180 min, 240 min, and 300 min reactions were collected (designated as Rh WNWs-20 min, Rh WNWs-30 min, Rh WNWs-60 min, Rh WNWs-90 min, Rh WNWs-120 min, Rh WNWs-180 min, Rh WNWs-240 min, and Rh WNWs-300 min, respectively) and used as the catalysts for the benzyl alcohol oxidation reaction under the same conditions. Similar plots of reaction products against time (Supporting Information Figure S14) were analyzed for all the catalysts. Table 1 summarizes the catalytic properties of the various Rh WNWs. All the Rh WNWs show effective catalytic behavior for the oxidation reaction but with highly distinct conversion efficiencies. For example, nearly 100% conversion was achieved in 3 h for the Rh WNWs-120 min, but 57.6% conversion was

regions can exist in each WNW (Figure 2d−f and Supporting Information Figure S7−8). Characteristically, structural defects were frequently observed around the vicinity of the crystal domains of the nanowire, as indicated by red arrows. Particularly, twin defects cross-sectioning the nanowire were frequently observed at the grain boundaries, as indicated by the dashed lines (Figure 2e) and also at joints where the nanowire changes growth direction (Figure 2f). The presence of highdensity twin defects and grain boundaries cross-sectioning the nanowire can serve as possible channels for O2 incorporation into the surface region therefore can facilitate the oxidation reactions. Those unusual features of the ultrathin Rh WNWs reported here suggested that a unique mechanism might dominate their growth. To probe the formation mechanism of ultrathin Rh WNWs in our system, we have systematically monitored their growth process and examined the growth intermediates at different reaction times. As illustrated by TEM analyses, only Rh nanoparticles in the sub-2-nm regime were seen at 10 min (Supporting Information Figure S9a). Those nanoparticles continued to grow with the average size increased to 2.2 nm at 20 min (Figure 3a). The formation of short Rh nanowires was found after a 30 min reaction. The nanowires exhibit an average diameter of about 2.2 nm, the same as those of coexisting nanoparticles (Figure 3b). The Rh nanoparticles continually evolved into one-dimensional nanostructures with increasing reaction time. As a result, the population of nanoparticles decreases and the population of longer nanowires increases with increasing reaction time (Figure 3c). The WNWs with high aspect ratios were obtained at 120 min and onward. At this point, the Rh nanowires display a relatively rough surface with a large number of convex/concave regions and distinct domains (Figures 2f and 3e). This intermediate apparently has the largest total number of defects (Figure 3f; as Rh WNWs are not single crystal, we counted the number of structural defects where the crystal orientations are not continuous and where grain boundaries are clearly present) and the highest degree of wavy-like features. The relative total number of defects in each sample is shown in Figure 3g, which was generated by dividing data in Figure 3f by those in Figure 3c. When the relative total number of defects for Rh WNWs-120 min was fixed at 1, the relative total number of defects was calculated at 0.24, 0.62, 0.90, 1.0, 0.57, 0.34, and 0.34 for the Rh WNWs-30 min, Rh WNWs-60 min, Rh WNWs-90 min, Rh WNWs-120 min, Rh WNWs-180 min, Rh WNWs-240 min, and Rh WNWs-300 min, respectively. The relative total number of defects in each sample shows a volcano type behavior and the Rh WNWs-120 min have the highest relative total number of defects. By further increasing the reaction time to 180 and 240 min, the product becomes less wavy as compared with the Rh WNWs collected at 120 min. The average diameter of the Rh WNWs collected at 180 and 240 min is around 2.1 nm. No significant change in the overall shape is observed in products beyond 240 min (Supporting Information Figure S9b). The growth kinetics and structure transformation of the Rh WNWs was further investigated by using inductively coupled plasma atomic emission spectrometry (ICP-AES) and high resolution TEM (HRTEM). Figure 3d shows the percentage of Na3RhCl6 reduced as a function of time. We can see that, the reduction of Na3RhCl6 was very fast and the percentage of Na3RhCl6 reduced was 35.2% at t = 10 min and then 89.2, 90.1, 92.3, 91.6, 92.8, 93.1 and 93.5% at 20, 30, 60, 90, 120, 180, and 240 min, respectively. The reduction of Na3RhCl6 was almost 3890

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

Letter

duration. The Rh WNWs-120 min has smallest corners/edges/ surface area due to the particle attachment and largest number of defects (Figure 3g). Considering that the Rh WNWs-120 min still have the best catalytic activity with the highest TOF number, the observation of the exceptional catalytic activity in Rh WNWs-120 min is particularly significant, which strongly demonstrates that defects other than the corners/edges/surface area plays more important role in the catalytic activity of Rh WNWs. It is also noted the Rh WNWs-120 min exhibit excellent durability during the repeated cycles of the oxidation reaction (Table 1, Entry 12−13). There was only a slight decrease in the conversion after five cycles. Considering that the morphology and structure of the Rh WNWs is largely retained after five catalytic reaction cycles (Supporting Information Figure S15), we believe the slight decrease of the conversion is largely due to the inevitable catalyst loss during the centrifugation for collection after each recycle. Figure 4a shows the plots of reaction products against time. Significantly, with the Rh WNWs 120 min as the catalyst, nearly 100% conversion of benzyl alcohol was achieved in 3 h with nearly 100% selectivity toward benzaldehyde. There are negligible byproducts (e.g., benzyl acid) in the reaction products. The control experiment without the catalyst shows that essentially no transformation of benzyl alcohol into benzaldehyde can be observed with O2 alone (Figure 4b). To further evaluate the catalytic activity of the RhWNWs, we have synthesized Rh nanocrystals with other shapes (∼6.5 nm Rh nanocubes (Supporting Information Figure S13) and ∼3.1 nm Rh nanoparticles (Supporting Information Figure S1a,b)) as controls to compare with the Rh WNWs in catalytic oxidation studies. Under the same reaction condition and with the same amount of Rh nanocubes, only 9.9% of benzyl alcohol was converted in the 3-h oxidation reaction. Elongating the reaction time to 10 h did not improve much the conversion of benzyl alcohol to benzaldehyde (22.9% conversion of benzyl alcohol). Similarly, replacing the Rh WNWs-120 min with Rh nanoparticles did not result in high-yield conversion of benzyl alcohol either (44.7% conversion of benzyl alcohol in the 3-h reaction). Compared to the conversion efficiency at the first 30 min reaction, Rh WNWs-120 min (29.33% conversion) is much more active than that of the 6 nm Rh nanocubes (1.55% conversion) and 3 nm Rh nanoparticles (4.56%). Considering the fact that the 2 nm WNWs roughly have the same surface to

Table 1. Catalytic Results of the Oxidation of Benzyl Alcohol Using Various Rh Catalystsa conversion /%

TOF/h−1

entry

catalyst

0.5 h

3h

0.5 h

3h

1 2 3 4 5 6 7 8 9 10 11 12 13

Blank Rh nanocubes Rh nanoparticles Rh WNWs-20 min Rh WNWs-30 min Rh WNWs-60 min Rh WNWs-90 min Rh WNWs-120 min Rh WNWs-180 min Rh WNWs-240 min Rh WNWs-300 min Rh WNWs-120 min-second Rh WNWs-120 min-fifth

0.01 1.55 4.56 7.57 10.63 15.21 21.18 29.33 18.29 13.25 13.21 27.64 26.83

0.09 9.88 44.78 57.58 69.89 76.10 88.38 99.39 81.39 71.42 71.30 93.43 90.11

6 930 2736 4542 6378 9126 12 708 17 598 10 974 7950 7924 16 584 16 098

9 988 4478 5758 6989 7610 8838 9939 8139 7142 7130 9343 9011

a

Reaction conditions: benzyl alcohol, 3.0 mmol; catalysts, 1.0 mL, 0.1 mM Rh nanocrystals in H2O; solvent, 9.0 mL H2O; 1 atm O2 gas; reaction temperature, 90 °C.

obtained for the Rh WNWs-20 min in the same reaction time. The catalytic activities were further normalized to the intrinsic turnover frequency (TOF, defined as the benzyl alcohol conversion per mole metal per hour). Such normalization allows to directly compare the effects of structural defects on the catalysis. The TOFs (at 0.5 h catalytic reaction) of Rh WNWs-20 min, Rh WNWs-30 min, Rh WNWs-60 min, Rh WNWs-90 min, Rh WNWs-120 min, Rh WNWs-180 min, Rh WNWs-240 min, and Rh WNWs-300 min are 4542, 6378, 9126, 12708, 17598, 10974, 7950, and 7924 h−1 (Table 1, Entry 4−11), respectively, which are all much higher than those of cubic Rh nanocrystals and Rh nanoparticle (the TOFs of Rh nanocubes and Rh nanoparticles are 930 and 2736 h−1, respectively), demonstrating superior catalytic performance of the Rh WNMs catalysts. Remarkably, the Rh WNWs-120 min shows the best catalytic activity with the highest TOF number, which is even higher than the state-of-art catalyst.39 On the other hand, based on the growth kinetics and the structure transformation study for the Rh WNWs, the Rh WNWs can be synthesized with tunable degree of particle attachment and thereby the variable defects by manipulating the reaction

Figure 4. Catalytic activities of various Rh catalysts. (a) The time-dependent conversion of benzyl alcohol and the product selectivity in the oxidation reaction catalyzed by Rh WNWs-120 min at 90 °C and 1 atm O2 gas. (b) Benzyl alcohol conversion with the reaction time catalyzed by using different Rh catalysts. 3891

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

Letter

Figure 5. Fundamental properties of various Rh catalysts. (a) Rh 3d XPS spectra of the Rh WNWs-120 min. (b) CV curves of several typical Rh WNWs on glassy carbon surface, in 0.5 M H2SO4. (c) (Top) Variation of the fraction of the Rh+ in various Rh WNWs. All Rh+ fractions in this figure were generated from Supporting Information Figure S16. (Middle) Variation of the current density (J) (at 0.7 V) of various Rh WNWs. All current density in this figure were obtained from Supporting Information Figure S17. (Bottom) Variation of TOF (at 3 h catalytic reaction) of various Rh WNWs in the selective oxidation of benzyl alcohol.

fractions of Rh+ in Rh WNWs-180 min and Rh WNWs-240 min are 26.30% and 20.18%, respectively. In addition to XPS results, the structural and chemical nature of the Rh WNWs can also be characterized by cyclic voltammograms (CVs). For this purpose, we relied on the characteristic features of oxygen adsorption/desorption which are very sensitive to the nature of the metal structure and oxygen atoms preferably adsorb on atoms near steps, kinks or structural defects.41 Importantly, the CV measurements shows different Rh WNWs exhibit remarkably different features in the oxygen region (0 to 0.80 V) (Figure 5b and Supporting Information Figure S17). It is evident that oxygen adsorption/ desorption current on Rh WNWs-120 min is significantly larger than that on the other Rh WNWs. The variation of oxygen adsorption/desorption in various Rh WNWs is also very similar to that of the XPS results (Figure 5c) and in agreement with TEM results and catalytic studies. The Rh WNWs-120 min shows the largest fraction of the Rh+ and biggest oxygen adsorption/desorption current, which can be attributed to the degree of structural defects in the Rh WNWs (Figure 3e,f and Supporting Information Figure S10). In general, most of the atoms in the vicinity of the grain boundaries or structural defects usually have a low coordination number to create more traps for the activation of O2. Together, these systematic structural, chemical, and catalytic studies clearly demonstrates that by engineering the structural defect in the Rh WNWs, it is possible to significantly alter the chemical states of the surface atoms, which can fundamentally determine their catalytic activities. Importantly, the concept of defect engineering in ultrafine WNWs for improved catalytic properties is general and can be readily extended to other metals, such as palladium WNWs (Pd WNWs) and iridium WNWs (Ir WNWs). Briefly, Pd WNWs

bulk ratio as the 3 nm Rh nanoparticles, the observation of the exceptional catalytic activity in WNWs is particularly significant, which strongly indicates that other factors (e.g., structural defects/grain boundaries) than the surface area plays an important role in the catalytic activity. It has been previously suggested that the surface defects can serve as possible channels for O2 incorporation into the surface region.40 Therefore, the presence of structural defects in WNWs can fundamentally affect the adsorption/dissociation of O2 and to modulate their catalytic activity. When the O2 is incorporated into the surface through the defect sites, the interaction between the oxygen and the electron cloud of the adjacent Rh atoms can positively shift the chemical state of Rh atoms and activate O2 for the subsequent oxidation reaction. To support this hypothesis, we have conducted X-ray photoelectron spectroscopy (XPS) analysis to determine the chemical states of the different Rh WNWs. In general, XPS spectra of all the Rh WNWs show two peaks that can be assigned to Rh 3d5/2 and Rh 3d3/2 states (Figure 5 and Supporting Information Figure S15). These two peaks can be further split into two doublets, associated with Rh0 and Rh+ chemical states. Importantly, it is noticed that the total Rh+ fraction in the Rh WNWs is quite different from each other. In particular, the total Rh+ fraction in the Rh WNWs shows a volcano type behavior, which is similar to their corresponding TOF changes (Figure 5c), and resembles that of the structural defects population (Figure 3f). Rh WNWs-120 min exhibits the highest fraction of the Rh+ (33.97%) and the highest TOF. The fraction of Rh+ is observed to increase from Rh WNWs-20 min to Rh WNWs-30 min, Rh WNWs-60 min, and Rh WNWs-90 min and finally peaks at Rh WNWs-120 min. Further prolonging the growth time beyond 120 min, however, lead to a decrease of the fraction of Rh+ in the Rh WNWs. The 3892

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

Letter

Figure 6. Structures and catalytic activities of various Ir WNWs and Pd WNWs. Representative HRTEM images of (a) the Ir WNWs intermediates collected from various reaction times and (c) the Pd WNWs intermediates collected from various reaction times. Please see Supporting Information Figure S12 for enlarged images. Red arrows in (a) and (c) indicate the presence of distinct defects. (b) The time-domain conversion of nitrobenzene in Ir WNWs-catalyzed hydrogenation reaction. (d) Electrocatalytic properties of Pd WNWs prepared with different reaction time. The arrows in the CV curves indicate the scan direction. The CV curves were recorded in 0.5 M H2SO4 + 0.25 M HCOOH solution at a scan rate of 100 mV/s.

such ultrafine WNWs represent an ideal platform for investigating structural defect dependent catalytic properties and demonstrate that their catalytic activity is highly dependent on their structural defects, with increasing catalytic activity in WNWs with more structural defects. This unusual catalytic behavior can be attributed to the unique chemical states altered by the degree of structural defects. Together, our systematic studies demonstrate ultrafine WNWs as a unique structure that can allow the creation of a high population of accessible structural defects as catalytic hot spots, which is not readily possible in other nanostructures or bulk materials. With the advancement in characterization techniques the roles of structural defects are expected to be further understood by correlating the experimental information with theoretical models.42,43 It can therefore open a new pathway to highly active catalysts through defect engineering and can significantly impact broad areas including energy conversion, fine chemical synthesis, and beyond.

were obtained by introducing Na2PdCl4 into water in the presence of PVP as surfactant and Ir WNWs were produced by using Na3IrCl6 as the precursor and PVP as the surfactant (see Supporting Information for details). In both the cases of Pd WNWs and Ir WNWs, similar nanoparticles or short nanowires can be obtained initially, which can be transformed into longer wavy rough nanowires with a high density of structural defects and then less rough nanowires with increasing growth time (Figure 6a,c and Supporting Information Figure S11−12). A similar evolution of catalytic properties with structural defects was also observed in these two WNWs. In the hydrogenation of nitrobenzene, it is evident that the activity of Ir WNWs-12h is higher than that on Ir WNWs-6h and Ir WNWs-18h. After the addition of the Ir WNWs-12h, we can see that the hydrogenation reaction proceeds very fast, and nearly 100% conversion of nitrobenzene to aniline was achieved within 50 min. In contrast, only 69% and 63% conversion were obtained in the same reaction time with the Ir WNWs-6h and Ir WNWs18h as the catalysts (Figure 6b). The Pd WNWs also showed enhanced electrocatalytic activity for the oxidation of formic acid. The maximum forward current densities of the Pd WNWs-4h were measured to be 444 mAmg−1, 1.46 times and 1.65 times of that of Pd WNWs-2h and Pd WNWs-6h, respectively (Figure 6d). In conclusion, we have shown that diverse noble metal WNWs with highly variable structural defects can be synthesized in a highly controllable manner. We show that

■

ASSOCIATED CONTENT

S Supporting Information *

Information on materials and methods, additional TEM and HRTEM images, and cyclic voltammograms. This material is available free of charge via the Internet at http://pubs.acs.org. 3893

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894

Nano Letters

■

Letter

(26) Bucur, R. V.; Indrea, E. Acta Metall. Mater. 1987, 35, 1325− 1332. (27) Maier, J. Angew. Chem., Int. Ed. 1993, 32, 313−335. (28) Maier, J. Angew. Chem., Int. Ed. 1993, 32, 528−542. (29) Leibsle, F. M.; Murray, P. W.; Francis, S. M.; Thornton, G.; Bowker, M. Nature 1993, 363, 706−709. (30) Ungvary, F. Coord. Chem. Rev. 2007, 251, 2087−2102. (31) Park, K. H.; Jang, K.; Kim, H. J.; Son, S. U. Angew. Chem., Int. Ed. 2007, 46, 1152−1155. (32) Grass, M. E.; Zhang, Y. W.; Butcher, D. R.; Park, J. Y.; Li, Y. M.; Bluhm, H.; Bratlie, K. M.; Zhang, T. F.; Somorjai, G. A. Angew. Chem., Int. Ed. 2008, 47, 8893−8896. (33) Zhang, H.; Xia, X. H.; Li, W. Y.; Zeng, J.; Dai, Y. Q.; Yang, D. R.; Xia, Y. N. Angew. Chem., Int. Ed. 2010, 49, 5296−5300. (34) Yuan, Y.; Yan, N.; Dyson, P. J. ACS Catal. 2012, 2, 1057−1069. (35) Liao, H. G.; Cui, L. K.; Whitelam, S.; Zheng, H. M. Science 2012, 336, 1011−1014. (36) Zhu, C.; Peng, H. C.; Zeng, J.; Liu, J. Y.; Gu, Z. Z.; Xia, Y. N. J. Am. Chem. .Soc. 2012, 134, 20234−20237. (37) Wang, Y.; Choi, S. I.; Zhao, X.; Xie, S. F.; Peng, H. C.; Chi, M. F.; Huang, C. Z.; Xia, Y. N. Adv. Funct. Mater. 2013, DOI: 10.1002/ adfm.201302339. (38) Mallat, T.; Baiker, A. Chem. Rev. 2004, 104, 3037−3058. (39) Enache, D. I.; Edwards, J. K.; Landon, P.; Solsona-Espriu, B.; Carley, A. F.; Herzing, A. A.; Watanabe, M.; Kiely, C. J.; Knight, D. W.; Hutchings, G. J. Science 2006, 311, 362−365. (40) Blume, R.; Niehus, H.; Conrad, H.; Bottcher, A. J. Phys. Chem. B 2004, 108, 14332−14339. (41) Furuya, N.; Koide, S. Surf. Sci. 1990, 226, 221−225. (42) Marks, L. D. Rep. Prog. Phys. 1994, 57, 603−649. (43) Chen, C.-C.; Zhu, C.; White, E. R.; Chiu, C.-Y.; Scott, M. C.; Regan, B. C.; Marks, L. D.; Huang, Y.; Miao, J. Nature 2013, 496, 74− 77.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the support from ARO through Award No. 54709-MS-PCS and the ONR under Award N00014-08-10985. We thank Electron Imaging Center of Nanomachines at CNSI for the TEM support. Y.H. acknowledges the support from Sloan Research Fellowship. X.D. acknowledges the support from the U. S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Science and Engineering through Award DE-SC0008055.

■

REFERENCES

(1) Vougioukalakis, G. C.; Grubbs, R. H. Chem. Rev. 2010, 110, 1746−1787. (2) Magano, J.; Dunetz, J. R. Chem. Rev. 2011, 111, 2177−2250. (3) Climent, M. J.; Corma, A.; Iborra, S. Chem. Rev. 2011, 111, 1072−1133. (4) Zhang, Y.; Cui, X. J.; Shi, F.; Deng, Y. Q. Chem. Rev. 2012, 112, 2467−2505. (5) Wang, Z. L. J. Phys. Chem. B 2000, 104, 1153−1175. (6) Turner, M.; Golovko, V. B.; Vaughan, O. P. H.; Abdulkin, P.; Berenguer-Murcia, A.; Tikhov, M. S.; Johnson, B. F. G.; Lambert, R. M. Nature 2008, 454, 981−984. (7) Tian, N.; Zhou, Z. Y.; Sun, S. G. J. Phys. Chem. C 2008, 112, 19801−19817. (8) Oliver-Meseguer, J.; Cabrero-Antonino, J. R.; Dominguez, I.; Leyva-Perez, A.; Corma, A. Science 2012, 338, 1452−1455. (9) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.; El-Sayed, M. A. Science 1996, 272, 1924−1926. (10) Sun, Y. G.; Xia, Y. N. Science 2002, 298, 2176−2179. (11) Wang, X.; Zhuang, J.; Peng, Q.; Li, Y. D. Nature 2005, 437, 121−124. (12) Tian, N.; Zhou, Z. Y.; Sun, S. G.; Ding, Y.; Wang, Z. L. Science 2007, 316, 732−735. (13) Huang, X. Q.; Tang, S. H.; Mu, X. L.; Dai, Y.; Chen, G. X.; Zhou, Z. Y.; Ruan, F. X.; Yang, Z. L.; Zheng, N. F. Nat. Nanotechnol. 2011, 6, 28−32. (14) Chiu, C. Y.; Li, Y. J.; Ruan, L. Y.; Ye, X. C.; Murray, C. B.; Huang, Y. Nat. Chem. 2011, 3, 393−399. (15) Murphy, C. J.; San, T. K.; Gole, A. M.; Orendorff, C. J.; Gao, J. X.; Gou, L.; Hunyadi, S. E.; Li, T. J. Phys. Chem. B 2005, 109, 13857− 13870. (16) Wiley, B.; Sun, Y. G.; Xia, Y. N. Acc. Chem. Res. 2007, 40, 1067− 1076. (17) Xia, Y. N.; Xiong, Y. J.; Lim, B.; Skrabalak, S. E. Angew. Chem., Int. Ed. 2009, 48, 60−103. (18) Jones, M. R.; Osberg, K. D.; Macfarlane, R. J.; Langille, M. R.; Mirkin, C. A. Chem. Rev. 2011, 111, 3736−3827. (19) Chen, M.; Wu, B. H.; Yang, J.; Zheng, N. F. Adv. Mater. 2012, 24, 862−879. (20) Habas, S. E.; Lee, H.; Radmilovic, V.; Somorjai, G. A.; Yang, P. Nat. Mater. 2007, 6, 692−697. (21) Stamenkovic, V. R.; Fowler, B.; Mun, B. S.; Wang, G. F.; Ross, P. N.; Lucas, C. A.; Markovic, N. M. Science 2007, 315, 493−497. (22) Wang, C.; Daimon, H.; Onodera, T.; Koda, T.; Sun, S. H. Angew. Chem., Int. Ed. 2008, 47, 3588−3591. (23) Tsung, C. K.; Kuhn, J. N.; Huang, W. Y.; Aliaga, C.; Hung, L. I.; Somorjai, G. A.; Yang, P. D. J. Am. Chem. Soc. 2009, 131, 5816−5822. (24) Zeng, J.; Zhang, Q.; Chen, J. Y.; Xia, Y. N. Nano Lett. 2010, 10, 30−35. (25) Wu, J. B.; Gross, A.; Yang, H. Nano Lett. 2011, 11, 798−802. 3894

dx.doi.org/10.1021/nl501137a | Nano Lett. 2014, 14, 3887−3894