Wet-Chemical Synthesis of Palladium Nanosprings - American

Aug 5, 2011 - Nanomaterial synthesis is one of the most critical topics in the nanoscience community, because it is the cornerstone of whole material ...
0 downloads 0 Views 851KB Size
LETTER pubs.acs.org/NanoLett

Wet-Chemical Synthesis of Palladium Nanosprings Lichun Liu,† Sang-Hoon Yoo,† Sang A. Lee,† and Sungho Park*,†,‡ †

Department of Chemistry and ‡Department of Energy Science, Sungkyunkwan University, Suwon, Gyeonggi-Do, 440-746, Korea

bS Supporting Information ABSTRACT: We report a methodology for synthesis of palladium (Pd) nanospring structures using an anodic aluminum oxide (AAO) membrane template and facile electrochemical deposition. The hydroxyl-terminated surfaces of alumina nanochannels and localized hydrogen evolution contribute to the growth of Pd atoms at peripheral positions of the alumina nanochannels in the presence of an effectual electric potential and a plating solution consisting of PdCl2, CuCl2, and HCl. Structural characterization including EDS line analysis and element mapping revealed Pd nanodomains curling up on the Cu nanorods. A clear Pd nanospring shape was observed after selectively removing Cu. The lengths of the nanosprings were dictated by the charges transported through electrodeposition, and the diameters of the nanosprings were tunable by altering the diameter of the alumina nanochannels. Screw dislocation is the most probable crystallographic defect responsible for the formation of coiled Pd nanostructures. Pd nanosprings have potential applications in nanomachines, nanosensors, nanoinductors, and metamaterials. We anticipate that our synthesis method will motivate and inform the synthesis of more advanced nanomaterials. KEYWORDS: Anodic aluminum oxide (AAO), electrodeposition, palladium, nanospring, double layer, nanorod

N

anomaterial synthesis is one of the most critical topics in the nanoscience community, because it is the cornerstone of whole material research and applications. A wide variety of wellknown and extensively studied nanomaterials with simple shapes, such as nanoparticles, nanorods, nanocubes, and nanotubes have been synthesized using two general approaches: bottom-up (growth) and top-down (decomposition) with template-assisted and template-free methods.1 A template (e.g., anodic aluminum oxide membrane, AAO, with highly ordered diameter-tunable nanochannels, Supporting Information Figure S1) can confine the chemical or physical reaction to a specific position in nanoscale space, thereby allowing the efficient development of many types of nanostructures (nanoparticles, nanodisks, nanorods, nanowires, and nanotubes).2 At present, there is great demand for more structurally complex nanomaterials because the shapes of nanomaterials affect their chemical and physical properties.3 5 A nanospring (coiled spring on the nanometer scale) is a typical example of a nanostructure with a complex shape; nanosprings could potentially serve as functional parts of nanomachines, nanosensors, nanoinductors, and photonic metamaterials.3 Until now, the majority of nanospring structures have been synthesized by chemical vapor deposition (CVD) on certain substrates, such as silicon carbide (SiC),6 boron carbide (BC),7 silicon dioxide (SiO2),8 and zinc oxide (ZnO),9 11 without the assistance of templates. This method usually requires high temperatures, high-purity chemicals, and expensive apparatus. Thus, it remains a challenge to synthesize complex nanostructures at low temperatures and in the aqueous phase. Herein, we report a novel methodology for synthesizing Pd nanosprings r 2011 American Chemical Society

with tunable lengths and diameters by using a simple template method and complete wet-chemical processing. Results and Discussion. Hydroxyl ( OH) group-terminated surfaces of an AAO template can selectively adsorb ions and molecules present in the adjacent liquid solution. The pH of the solution usually dictates the composition of the alumina surface; OH2+ functional groups are formed under acid conditions and O groups under alkaline conditions. A nanometer-thick interfacial double layer can be formed due to electrostatic interactions between the alumina surface and the ionic or molecular species in the solution.12 The distribution of ions in the electroneutral double layer clearly differs from that in the normal bulk solution. In an acid solution with an appropriate pH value, H+ ions can adsorb onto the alumina surface resulting in a compact layer (Gouy Chapman Stern model, Scheme 1A), thereby enriching the surface in H+ ions. When an effectual electric potential is externally applied through the nanochannels of the AAO, more hydrogen molecules or atoms will be generated at the walls of the alumina nanochannels than elsewhere. Because of the pivotal fact that hydrogen (either atoms or molecules) can effectively reduce PdCl42 under acidic conditions, the electrochemically induced disparity in the hydrogen concentration between peripheral positions and other positions in an alumina nanochannel results in the efficient formation of thin-walled palladium (Pd) nanotubes13 (Figure 1A) when PdCl42 and HCl are included in the plating solution. When Received: July 8, 2011 Revised: August 1, 2011 Published: August 05, 2011 3979

dx.doi.org/10.1021/nl202332x | Nano Lett. 2011, 11, 3979–3982

Nano Letters Scheme 1a

a

(A) Schematic drawing indicating cross-sectional distribution of ions in an alumina nanochannel that obeys a Gouy Chapman Stern interfacial double layer model. The central circle in light gray color implies the bulk solution region. (M stands for Pd or Cu for the sake of simplicity). (B) Schematic illustration of the synthetic procedure for Pd nanosprings.

another type of metal ion such as Cu2+ is introduced into the solution, the system became more interesting, because Cu2+ ions are relatively less reactive with hydrogen than PdCl42 ions according to the activity series. The synthesis procedure that we used to synthesize Pd nanosprings is presented in Scheme 1B. Typically, we performed experiments using a solution containing 25 mM PdCl2, 20 mM CuCl2 3 2H2O, and 0.1 M HCl with an electric potential of 0.1 V (vs Ag/AgCl). After casting the solution into an electrochemical cell and applying an external electric potential, Cu and Pd ions are simultaneously reduced on the predeposited bottom Au cathode (Supporting Information Figure S2). It is noteworthy that thin-walled Pd nanotubes formed when Cu ions were absent from the solution (Figure 1A).13 The simultaneous reduction of two metal ions leads to the growth of homogeneous Pd Cu nanorods (Supporting Information Figure S3). These nanorods exhibited a periodical, indented, helical surface curvature in contrast to the smooth surfaces usually obtained. We observed this surface morphology in perpendicular nanorods (Figure 1B) and more clearly in dispersed nanorods (Figure 1C). TEM images revealed that individual nanorods were composed of a dark, spring-shaped wire swirling along the long axis of the nanorods and another bright component residing in the nanorod cores (Figure 1D). The composition of the nanorods was determined as Pd and Cu by energy dispersive spectroscopy (EDS) spectral analysis (Figure 1E H). To determine the atomic identities of the spring-shaped wire and filler, we conducted EDS line analysis to investigate the atomic spatial distributions in cylindrical nanorods. By setting the scanning trajectory line at the near edge of a Pd Cu nanorod, we clearly observed the periodical compositional change in the spectrum of EDS line analysis for Pd (Figure 1E, red traces), as expected. The periodical distribution of Pd was also consistent with the contrast difference visually observed for the nanorods on TEM images. The contrast difference is due not only to the indented surface curvature, but also the different electron transmission abilities of Pd and Cu metal atoms. The EDS element map (projected profile) of the Pd portion of a Pd Cu nanorod revealed that it had a spring shape, visualized by the density of colored pixels on the plane (Figure 1G). In contrast, the element map for the Cu portion (Figure 1H) showed less pixels at the positions occupied

LETTER

by the Pd wire. The obscure contrasts of spatial pixel densities for Cu at different locations were due to stronger or weaker alloy interactions between Pd and Cu. To further characterize the Pd nanosprings, we selectively etched away Cu from the Pd Cu nanorods using 30% (volume ratio) nitric acid. Figure 2A shows a high yield of Pd nanospring structures with a shape resembling that of a real spring as drawn in Figure 2B. Both TEM images (Figure 2D) and high-resolution SEM images (Figure 2C) of Pd nanosprings revealed their obvious spring shapes. A little difference of Pd nanosprings diameters shown in Figure 2A was caused by inhomogeneous nanochannels diameters. The outer diameter of Pd nanosprings was equal to the diameter of AAO nanochannels. The inner diameter and wire diameter of Pd nanosprings were estimated to be about 130 and 60 nm for large ones (Figure 2D) and 30 and 25 nm for small ones (Figure 3B), respectively. It is worthy to note that as-synthesized nanosprings were not chiral with respect to the growth direction. The availability of these polycrystalline (Supporting Information Figure S4) Pd nanosprings was highly dependent on the concentration of PdCl2 and CuCl2 used for electrodeposition. The key requirement for Pd nanospring synthesis is the appropriate composition of Pd and Cu ions in the nanochannels. A high concentration of PdCl2 in the plating solution adversely affected the production of Pd nanospring structures; only slight slits were left on the surface of the nanorods after selectively dissolving Cu (Supporting Information Figure S5). In addition, a high concentration of CuCl2 in the plating solution led to the production of porous Pd Cu nanorods even before etching away the Cu (Supporting Information Figure S6). The as-synthesized nanorods structure was reproducible when the concentration of HCl was from 100 to 300 mM (PdCl2 at 25 mM). The measured pH of the solution before and after deposition did not show obvious change because the amount of reacted H+ ions under a moderate reduction potential ( 0.1 V) was much less than that in the solution. The effective applied potentials for electrochemical codeposition were limited in a narrow range ( 0.05∼ 0.3 V vs Ag/AgCl). Hydrogen evolution rate was much enhanced in more negative potential (< 0.4 V), releasing large amount of hydrogen gas bubbles that apparently disturbed the regular arrangement of atoms. Other applied potentials and concentrations of HCl gave Pd nanosprings with a low yield or a failed structure. The synthesis strategy described above can be used to synthesize Pd nanosprings with variable lengths and diameters. By monitoring the charge transported during electrodeposition, we were able to easily control the lengths of the Pd nanosprings. For 1 C/cm2 (C: Coulomb) deposition, the length of spring was about 1.5 μm (Supporting Information Figure S7A). Longer Pd springs (∼6 μm, Supporting Information Figure S7B) were producible with a 4 C/cm2 deposition. Furthermore, by using AAO templates with the desired diameters of nanochannels, we were able to flexibly tune the diameters of the Pd nanosprings; we were able to synthesize Pd nanosprings as small as ∼70 nm in diameter (Figure 3A,B) when an in-house made AAO template with smaller nanochannels was used. The successful synthesis of thin Pd nanosprings using these narrow nanochannels (70 nm) was benefiting from still effective double layer model (presented in Scheme 1A) because the thickness of double layer is usually in several nanometers or in subnanometers. To understand the mechanism underlying the growth of Pd nanosprings, it is necessary to have basic knowledge of crystal growth and electrochemical deposition. In a previous study,13 we 3980

dx.doi.org/10.1021/nl202332x |Nano Lett. 2011, 11, 3979–3982

Nano Letters

LETTER

Figure 1. (A) A typical SEM image of Pd nanotubes formed from a solution containing 25 mM PdCl2 and 100 mM HCl under 0.1 V (vs Ag/AgCl). (B) SEM image of perpendicular nanorods showing periodical helical indented curvature on the surface. These nanorods were synthesized by only adding 20 mM CuCl2 3 2H2O into plating solution used in (A). (C) SEM image of disperse nanorods synthesized in (B) showing clear rotating spring shape in nanorods. (D) SEM image of nanorods revealing helical nanospring structure in nanorods. (E) EDS line analysis at the edge of a nanorod, peaks in upper panel show the alternating existence of Pd (red trace) and Cu (blue trace). (F) EDS of nanorods showing the composition of nanorods. (G) Element map for Pd portion in nanorods revealing the spring shape of Pd. (H) Element map for Cu portion in nanorods.

characterized Pd nanotubes in detail. The generation of nonalloyed nanotubes when CuCl2 was present in the plating solution was attributed to failure in constructing an effective depletion region. The formation of a depletion region on an electrochemical electrode can result in deposition of reduced metal atoms at elevated positions, such as the walls of nanotubes. In this system, Pd atoms were still able to grow along the walls of nanochannels, but Cu atoms were not. When the growth of nanorods started, Pd atoms nucleated to clusters and further grew to grains (Supporting Information Figure S8) nearby the wall of the AAO, simultaneously repelling reduced Cu atoms because of considerable atomic misfits. This is probably responsible for the slightly indented Cu on the surfaces of the nanorods. According to this spatially different growth process, thus other nanotube formation systems (e.g., employing anchoring agents, diffusion limits, and high current density) are potentially applicable to fabricate nanospring structure or other complicated nanostructures by virtue of bottom-up codeposition and topdown chemical etching. Pd nanospring growth was invalid when the concentration of CuCl2 in the solution was high because of the strong influence of a large amount of Cu atoms involved in

nanostructure (Supporting Information Figure S5). The observation of spring-shaped nanostructures provides insights into the probable formation mechanism of screw dislocation.14 Screw dislocation is generally considered a nuisance because it adversely affects the mechanical, electrical, and electromagnetic properties of materials; however, it provides a way to synthesize a new class of nanomaterials.15 17 Screw dislocation is one of the chief defects in crystal growth and is notoriously difficult to observe. Screw dislocation in our system can take place during the growth of Pd grains because of the actions of stored internal strain and stress, and may have a major influence on the characteristics of deposition and trigger the propagation of atom growth preferentially on the steps18 of a crystal surface in a spiral fashion. The barrier energy required for screw dislocationinduced crystal growth is thousands of times smaller than that required to create a new surface for crystal growth. To support these theory-based hypotheses, more comprehensive and indepth studies are required to collect more information about the unusual growth of Pd nanosprings in alumina nanochannels. In conclusion, we reported the first successful synthesis of Pd nanosprings with tunable lengths and diameters. The synthesis 3981

dx.doi.org/10.1021/nl202332x |Nano Lett. 2011, 11, 3979–3982

Nano Letters

LETTER

growth of coiled Pd nanosprings. Spring-shaped Pd nanomaterials could potentially be used as components of nanomachines, nanosensors, nanoinductors, and metamaterials. In addition, the synthesis method reported in this study can be potentially applied to the advanced synthesis of other nanomaterials.

’ ASSOCIATED CONTENT

bS

Supporting Information. Experimental details and additional SEM images of control experiments are included. This material is available free of charge via the Internet at http://pubs. acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

Figure 2. (A) SEM image of large quantity of Pd nanosprings synthesized using ∼250 nm nanochannel of commercial AAO template. (B) A drawing of real spring structure. (C) Magnified SEM image of Pd nanosprings showing their clear spring shapes. (D) TEM image of a typical Pd nanospring with ∼60 nm in wire diameter and ∼238 nm in spring diameter.

Figure 3. (A) TEM image of dispersed Pd nanosprings synthesized using smaller-sized nanochannel of homemade AAO template (∼70 nm). (B) SEM image of perpendicular Pd nanosprings synthesized using homemade AAO.

’ ACKNOWLEDGMENT This work was supported by the National Research Foundation of Korea (World Class University (WCU), R31-2008-10029; Nano R&D program, 2010-0019149, 2010-0015457; and Priority Research Centers Program, NRF-20100029699). ’ REFERENCES (1) Burda, C.; Chen, X.; Narayanan, R.; El-Sayed, M. A. Chem. Rev. 2005, 105 (4), 1025–1102. (2) Martin, C. R. Science 1994, 266 (5193), 1961–1966. (3) Korgel, B. A. Science 2005, 309 (5741), 1683–1684. (4) Gansel, J. K.; Thiel, M.; Rill, M. S.; Decker, M.; Bade, K.; Saile, V.; von Freymann, G.; Linden, S.; Wegener, M. Science 2009, 325 (5947), 1513–1515. (5) Tian, B.; Zheng, X.; Kempa, T. J.; Fang, Y.; Yu, N.; Yu, G.; Huang, J.; Lieber, C. M. Nature 2007, 449 (7164), 885–889. (6) Zhang, D. Q.; Alkhateeb, A.; Han, H. M.; Mahmood, H.; McIlroy, D. N.; Norton, M. G. Nano Lett. 2003, 3 (7), 983–987. (7) McIlroy, D. N.; Zhang, D.; Kranov, Y.; Norton, M. G. App. Phy. Lett. 2001, 79 (10), 1540–1542. (8) Zhang, H.-F.; Wang, C.-M.; Buck, E. C.; Wang, L.-S. Nano Lett. 2003, 3 (5), 577–580. (9) Gao, P. X.; Ding, Y.; Mai, W.; Hughes, W. L.; Lao, C.; Wang, Z. L. Science 2005, 309 (5741), 1700–1704. (10) Gao, P. X.; Wang, Z. L. Small 2005, 1 (10), 945–9. (11) Gao, P. X.; Mai, W.; Wang, Z. L. Nano Lett. 2006, 6 (11), 2536– 2543. (12) Kasprzyk-Hordern, B. Adv. Colloid Interf. Sci. 2004, 110 (1 2), 19–48. (13) Liu, L.; Park, S. Chem. Mater. 2011, 23 (6), 1456–1460. (14) Hull, D.; Bacon, D. J. Introduction to dislocations, 4th ed.; Butterworth-Heinemann: Woburn, MA, 2001. (15) Morin, S. A.; Bierman, M. J.; Tong, J.; Jin, S. Science 2010, 328 (5977), 476–480. (16) Morin, S. A.; Jin, S. Nano Lett. 2010, 10 (9), 3459–3463. (17) Bierman, M. J.; Lau, Y. K. A.; Kvit, A. V.; Schmitt, A. L.; Jin, S. Science 2008, 320 (5879), 1060–1063. (18) Lagally, M. G.; Zhang, Z. Nature 2002, 417 (6892), 907–910.

method that we used is robust and uncomplicated, as it only involves simple electrodeposition, chemical etching, and an AAO template. We proposed that an interfacial double layer, hydrogen evolution, and screw dislocation contributed to the successful 3982

dx.doi.org/10.1021/nl202332x |Nano Lett. 2011, 11, 3979–3982