Effect of Waveform of ac Voltage on the Morphology and Crystallinity

Apr 23, 2014 - The electrodes were contacted with a tip-probing station (Karl Suss). ...... Talukdar, Ishan; Ozturk, Birol; Flanders, Bret N.; Mishima...
0 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/Langmuir

Effect of Waveform of ac Voltage on the Morphology and Crystallinity of Electrochemically Assembled Platinum Nanowires Alexander Nerowski,† Markus Pötschke,† Ulrich Wiesenhütter,§ Jürgen Nicolai,∥ Ulana Cikalova,∥ Arezoo Dianat,† Artur Erbe,§ Jörg Opitz,∥ Manfred Bobeth,† Larysa Baraban,*,† and Gianaurelio Cuniberti†,‡ †

Institute for Materials Science and Max Bergmann Center of Biomaterials and ‡Center for Advancing Electronics Dresden, Dresden University of Technology, 01062 Dresden, Germany § Helmholtz-Zentrum Dresden-Rossendorf, Institute of Ion Beam Physics and Materials Research, 01314 Dresden, Germany ∥ Fraunhofer-Institute for Non-Destructive Testing, 01109 Dresden, Germany S Supporting Information *

ABSTRACT: Here we present electrochemically grown ultrathin platinum nanowires and demonstrate that their morphology and crystalline structure can be tuned by the waveform of the alternating voltage applied to the microelectrodes. The structure of the nanowires was analyzed by scanning and transmission electron microscopy. The voltage signal, applied to grow the nanowires, consisted of several Fourier components of a squareshaped wave. We observed that, depending on the number of Fourier components, the morphology of the nanowires changed from branched dendritic-like patterns to straight wires and the wire crystallinity changed from polycrystalline to highly oriented growth with the [111] direction of platinum crystallites along the nanowire axis. We propose a simple model to explain this intriguing observation.

1. INTRODUCTION High-aspect-ratio one-dimensional nanostructures, e.g., nanowires or nanotubes, fabricated via bottom-up self-assembly of nanoscaled objects in solution represent sound and low-cost alternatives to realizing novel, ultrasmall circuitry with rich functionalities, following the so-called “More than Moore” trend.1−4 Nowadays, one-dimensional nanostructures have gained considerable attention due to their unique properties, such as thermal and mechanical stability or electron and phonon transport.5,6 Significant efforts have been made to explore bottom-up fabrication techniques and applications of both semiconducting and metal nanowires.7,8 A particularly promising method to produce metal nanowires relies on the electrochemically assisted growth from a solution containing metal compounds.9 The undoubted advantage of this technique is that the nanowires can be grown directly between microfabricated electrodes of predefined design. If the metalbearing objects in the solution are neutral clusters or molecules, then the attraction of these objects to the nanowire tip is due to dielectrophoresis (DEP), the movement of a neutral particle within a spatially inhomogeneous electric field.10 In the case of charged species present in solution, the nanowire growth is termed directed electrochemical nanowire assembly (DENA),11,12 cf. the schematics in Figure 1. Whereas most DENA experiments rely on the application of a square-shaped alternating (ac) voltage, DEP-grown wires are often obtained with a voltage possessing a sine waveform.13,14 A lot of effort has been dedicated to controlling the nanowire growth process, allowing the fabrication of straight nanowires © 2014 American Chemical Society

Figure 1. Schematics of DENA setup (not to scale). ac potential is applied to the gold electrodes immersed in aqueous chloroplatinic acid solution. Wires grow at the side of the electrodes as well as at the tip.

with controlled morphology and diameter.15,16 For instance, Kawasaki et al.17 investigated the branching of growing platinum nanowires. The angle of branching was related to the crystallographic axes of a face-centered cubic (fcc) crystal, assuming the wires to be monocrystalline. In this work, the formation of branched wire structures is explained on the basis of the theory of dendritic solidification (e.g., ref 12). To our knowledge, there are few works concerning the crystalline structure of the nanowires, although it is an important feature regarding the application of the 1D nanostructures. It Received: January 22, 2014 Revised: March 16, 2014 Published: April 23, 2014 5655

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661

Langmuir

Article

Figure 2. SEM images of the evolution of the nanowire morphology depending on the waveform. The morphology varies from (a) dendrites grown with the superposition of three sine waves to (b, c) rectified wirelike structures grown with the superposition of four and six sine waves, respectively, to (d) completely straight wires for a square-shaped voltage signal. Scale bars are 500 nm. conducted within 3 h after dilution.24 A drop of 1.5 μL of the prepared solution was put onto the electrodes, and an ac voltage signal with a frequency of f = 500 kHz and a root-mean-square value of VRMS = 7 V was applied. After a growth time of 5 min, the function generator was switched off. In all wire growth experiments, the H 2 PtCl 6 concentration and the RMS value and frequency of the voltage were fixed. The morphology and crystal structure of the nanowires were characterized by SEM (Philips XL 30 ESEM-FEG) and TEM (FEI Titan 80-300).

determines, for instance, their electrical transport properties and their suitability for plasmonics.20,21 Electron diffraction of indium wires with diameters of about 370 nm revealed that the unbranched wire segments were single-crystalline.18 Also, diffraction studies on 75 nm gold nanowires led to the conclusion of single-crystal wire growth.19 Here, we address the control of the crystalline structure of electrochemically assembled platinum nanowires, realized via application of an ac voltage between gold electrodes in a solution. We observe drastic changes in the morphology and crystalline structure of the nanowires by varying the waveform of the applied ac voltage. Starting with a sine function, we increase the number of higher Fourier components in the Fourier series of a square wave until an almost square-shaped waveform is achieved. The resulting changes in the nanowires’ structure are investigated by scanning and transmission electron microscopy (SEM and TEM, respectively). Remarkably, the crystalline structure within the wires changes from random polycrystalline to [111] oriented growth, corresponding to the number of Fourier components. Finally, based on numerical results, we propose a mechanism for explaining the observed structural changes as a function of the applied waveform.

3. RESULTS AND DISCUSSION 3.1. Nanowire Growth. By varying the applied ac voltage signal, nanowires with diameters from 25 to 100 nm and with various degrees of branching and straightness can be fabricated. We observed the nucleation of the nanowires at both the tips and sides of the microelectrodes, when an ac voltage was applied to the electrodes. In order to investigate the effect of the waveform of the applied voltage signal on the resulting nanostructures, in our growth experiments we gradually increased the number N of Fourier components in the truncated Fourier series expansion of a square-shaped signal sin((2n − 1)ωt ) 2n − 1 N 1 ∑n = 1 (2n − 1)2

N

V (t ) =

2. EXPERIMENTAL SECTION According to the setup shown in Figure 1, platinum nanowires were grown from aqueous chloroplatinic acid (H2PtCl6) solution via applying an ac voltage between gold microelectrodes. The distance between electrodes was 8 μm. The electrodes were fabricated on various substrates, depending on the method of nanowire investigation. For SEM observations, a piece of silicon wafer with a 300 nm oxide layer was prepared as described previously,22 except for the exposure, which was made by laser lithography (Heidelberg Instruments DWL66 fs) with a 405 nm laser wavelength and a 4 mm writehead. For the TEM investigations, electron beam lithography was performed on a 50 nm thin Si3Ni4 membrane (Tedpella). A positive resist (PMMA 495 K A4, Micro Chem) was spin-coated onto the substrate at a speed of 3500 rpm to reach a thickness of 100 nm. The samples were exposed using e-beam lithography (Raith 150TWO) and developed with methylisobutylketon/isopropanol (ratio 1:3, Micro Chem) and isopropanol as a stopper. The developed structures were then coated by thermal evaporation with 3 nm Cr (adhesion layer) and 15 nm Au (electrode material). The lift-off was done with acetone. The width of the electrodes was for both SEM and TEM investigations 2 μm, and the length was at least 20 μm. The electrodes were contacted with a tip-probing station (Karl Suss). The electrical signal was provided by a function generator (Tektronix AFG320, maximum slew rate 3.5 × 108 V/s) and observed with an oscilloscope (Tektronix TDS3014) with an input resistance of 1 MΩ. The solution for the nanowire growth was prepared by dilution of a commercial H2PtCl6 stock solution (Sigma Aldrich, 8 wt %) with deionized water up to a concentration of 200 μM. At this concentration, we were able to grow comparatively thin and straight nanowires with diameters of about 20−100 nm.23 Since with increasing pH the solution becomes unstable, the experiments were

2 VRMS

∑n = 1

(1)

where V(t) is the time-dependent voltage, VRMS is its root mean square, ω = 2πf is the angular frequency, and t is the time. Figure 2 shows a series of SEM images illustrating the evolution of the nanowire morphology as a function of the signal shape of the applied ac voltage (see Figure S1 in the Supporting Information (SI)). The corresponding insets in Figure 2 display the waveforms of the applied electrical signal and their number N of Fourier components. No growth was observed for N = 1 (only the first harmonic), in accordance with results reported by Thapa et al.14 Observable growth of the wires occurs upon application of an ac signal consisting of three Fourier components (N = 3), as demonstrated in Figure 2a. The case N = 3 corresponds to an initial voltage change of 8.3 × 107 V/s, defined as the rate of the voltage increase at the rising edge of the signal. Nanostructures with predominantly highly branched morphology were obtained in this case. Increasing the number of Fourier components yielded straighter wires, as shown in Figure 2b,c for four (initial voltage change 1.1 × 108 V/s) and six Fourier components (1.6 × 108 V/s), respectively. Finally, we observed that the application of a square-shaped signal (3.5 × ×108 V/s, device limitation) results in the growth of straight and unbranched wires (panel d). These observations were additionally supported by analyzing the fractal dimension of the wire structures grown at different voltage signals. The fractal 5656

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661

Langmuir

Article

steps which might cause structural damage to the wires. Figure 3 displays the morphology of the nanowires in more detail. A wire grown with an applied ac signal with three Fourier components (N = 3, see panel a) reveals an overall directed growth with short side branches and a comparatively large diameter of 100 nm. When a square-shaped ac signal was applied, straight and unbranched wires with a diameter of around 25 nm were formed (Figure 3b). The enlarged image of Figure 3b in Figure 3c reveals a rough wire surface. The wire represents an agglomerate of 2−5 nm clusters. The observation of such a wire structure was unexpected since only negatively charged platinum complexes [PtCl6]2−, [PtCl5(H2O)]−, and [PtCl5(OH)]2− should exist in the solution.24 To understand the formation of the grainlike structure, we analyzed the orientation of the platinum clusters within the wire and conducted the Fourier transform of the obtained TEM images. Figure 4 displays a close-up image of the wire in Figure 3a, which was grown by applying a signal consisting of three Fourier components (N = 3). The crystal orientations of the platinum nanocrystallites were analyzed at different locations of the nanowires. Figure 4c−e displays the (111) lattice planes of the nanocrystallites. These images clearly reveal a crystalline atomic order. However, the orientation of the crystallites within the nanowire differs. The polycrystalline wire structure is also confirmed by the Fourier transform (FT) of the image in Figure 4a, which exhibits several diffuse ringlike spots. The reflections can be assigned to the (111) lattice planes of a platinum face-centered cubic crystal (fcc), which have a lattice spacing of d111 = 0.226 nm. Presence of chloride ions in the H2PtCl6 solution could potentially lead to its incorporation into the growing nanowire. To our knowledge, platinum crystals with incorporated chlorine atoms exhibit typical crystalline structures that have to be reflected in a corresponding Fourier analysis.25 Since FT images display only peaks peculiar to fcc order, we therefore conclude that Pt is the only constituting element of the nanowire. This statement is strengthened by conducting an elemental analysis of a wire using electron energy loss spectroscopy, proving the platinum composition of the wire. More details are provided in the SI (Figures S4 and S5). The ring sections indicate that the wire is polycrystalline, confirming the different orientations of the (111)-lattice planes found in the high-resolution images.26

Figure 3. TEM images of Pt nanowires grown on the sides of the gold microelectrodes. (a) Dendritic-like wire grown with a voltage signal with N = 3 Fourier components. (b) Straight wire grown with a square-shaped voltage signal. (c) Details of the wire in panel b.

dimension decreases from 1.65 for N = 3 to 1.24 for N → ∞ (see Figures S2 and S3 in the SI). The electric field around the opposing electrodes strongly affects the straightness of the wires.23 In particular, wires grown not directly between the electrodes but into the free space on the other electrode sides were found to be straighter and less branched. Therefore, our TEM investigations were performed on wires growing on those sides of the microelectrodes. We achieved very thin DENA-grown wires with diameters of around 25 nm. The low diameters are correlated with the low concentration of the H2PtCl6 solution of 200 μM.23 The observed dependence of the wire morphology on the ac voltage signal raises the question of the underlying growth mechanism. The applied root-mean-square voltage VRMS was kept constant for all experiments in Figure 2 and thus cannot cause the morphology change. To gain further insight into the growth mechanism, we focused on the elucidation of the crystalline structure of the nanowires by means of TEM. 3.2. TEM Investigations of Platinum Nanowires. For the TEM investigations, nanowires were grown at electrodes on a 50 nm thin Si3Ni4 membrane. On this substrate, nanowires were directly investigated without intermediate preparation

Figure 4. (a) Detailed TEM image of the wire in Figure 3a. The rectangular sections represent the detailed images in panels c−e. (b) Fourier transform of panel a indicating the polycrystallinity of the wire with different nanocrystal orientations. (c−e) (111) lattice planes within the nanowire sections from panel a. White stripes are a guide to the eye. The orientation of the lattice planes in the panels differs, e.g., the angle between (111) planes in panels c and e is 52°. 5657

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661

Langmuir

Article

Figure 5. (a) Detailed TEM image of the wire in Figure 3b,c. The rectangular sections represent the detailed images in panels c−e. (b) Fourier transform of panel a shows two spots, indicating a single-crystalline orientation among all nanocrystallites incorporated into the nanowire. (c−e) (111) lattice planes of the nanocrystallites from the sections in panel a. White stripes are a guide to the eye. All planes exhibit the same orientation (angle between (111)-crystal planes in panels c and d: 5°).

Detailed images of the wire in Figure 3b,c are depicted in Figure 5. In Figure 5c−e, sections of the wire in Figure 5a are presented, where (111) lattice planes are shown in more detail. In contrast to the wire in Figure 4, we find equal orientations of lattice planes all over the wire, except small deviations. For example, the angle between the lattice planes in Figure 5c,e is 5°. The Fourier transform of Figure 5a, shown in Figure 5b, reveals two spots in the axial wire direction, which correspond to (111) lattice planes. Due to small orientation variations of these planes, the spots are slightly smeared. In summary, the Fourier transform in Figure 5b suggests a common orientation of all nanocrystallites with (111) lattice planes perpendicular to the axial wire direction. Concerning the growth mechanism of the nanowires, it is clear that a highly oriented growth as shown in Figure 5 is not due to the close packing of the platinum clusters, which preexisted in solution (e.g., because of spontaneous reduction). Instead, an epitaxial growth of platinum from complex anions and possibly a small number of platinum oligomers with no distinct crystallographic structure is assumed. For voltage signals with three Fourier components (N = 3), the rare deposition of larger clusters with random orientation obviously creates a seed for a new orientation of nanocrystal growth in the wire. The TEM image in Figure 6 shows the straight wire in Figure 3b at the interface to the gold electrode. The Fourier transform in the inset displays the platinum (111) reflections in accordance with Figure 5b and three reflections which originate from the gold electrode. The three gold reflections of the same lattice spacing show that the electrode consists of grains with different crystal orientations. The reflections prove that the wire did not grow epitaxially onto the electrode. 3.3. Influence of Voltage Signal on Wire Crystal Structure and Morphology. To understand the effect of the waveform of the applied ac voltage on the wire morphology and crystallinity (cf. Figures 2 to 5), we analyzed the deposition process of platinum at the wire tip in more detail. The chemical reaction occurring during platinum deposition is poorly understood. We assume here that platinum is provided by anionic platinum complexes in the solution as, for example, in the form of [PtCl5(H2O)]−. The concentration of these anionic

Figure 6. TEM image of the nanowire in Figure 3b at the interface to the gold electrode. The Fourier transform in the inset shows one Pt reflection from the monocrystalline wire and three Au reflections from the polycrystalline electrode.

complexes near the nanowire tip strongly changes with time due to the voltage oscillations. For a quantitative analysis, we performed numerical simulations of the spatiotemporal concentration evolution of ions around the nanowire tip. The ion motion was described by the modified Poisson−Nernst− Planck equations

∂ci + ∇·ji ⃗ = 0 ∂t

(2)

⎛ ci ze ⎞ + i φ⎟ ji ⃗ = −Dci∇⎜log cmax − (c1 + c 2) kBT ⎠ ⎝

(3)

e ∇2 φ = − (z1c1 + z 2c 2) (4) ε (For more details, see refs 27 and 28.) The index i refers to cations (i = 1) and anions (i = 2). Equations 2 and 3 determine the evolution of the cation and anion concentrations ci. The ion fluxes ji⃗ include ion diffusion and electric-field-driven ion migration. The electric potential φ is governed by the Poisson equation (eq 4). The ions were supposed to be monovalent (z1 5658

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661

Langmuir

Article

= −z2 = 1 (ion charge numbers), ε = 80ε0 (electrolyte permittivity), e (elementary charge), kB (Boltzmann constant), and T = 293 K (temperature)). For simplicity, the ion diffusion coefficient D = 4.3 × 10−10 m2/s29 was assumed to be the same for all ion species. The ion concentration at the surface of the nanowire tip can become very large. To limit this ion crowding in our model to a reasonable maximum value, the term cmax − (c1 + c2) was introduced into the denominator in eq 3 (e.g., ref 27). In this way, steric effects due to finite ion sizes were approximately considered. As the maximum concentration, we chose cmax = 10 M, which is on the order of the solubility of salts such as sodium or potassium chloride at room temperature. In our model calculations, the nanowire tip was approximated by a sphere with radius r0 = 10 nm and the counter electrode, by a concentric hollow sphere with 1 μm radius. The potential was subject to the boundary conditions φ = V(t) at the nanowire electrode and φ = 0 at the counter electrode. Since the wire growth velocity is small compared to the ion velocities,28 the movement of the wire tip relative to the liquid was neglected. The deposition of platinum at the electrode surface could occur, for example, by the following cathodic and anodic partial reactions with electron transfer between the electrode surface and the anions

Figure 7. Calculated anion concentration profiles near the nanowire tip at different times for sine and square waveforms of the voltage signal. The plots reveal a higher anion concentration near the wire tip for the square waveform at equal field strength on the electrode surface. The field changes sign between the first and second times. The times are referred to the zero crossing of the voltage. The electric field exhibits a slight phase shift with respect to the voltage. The times t in ns and field strength E in MV/m corresponding to the curves are given as pairs (t, E). Blue (sine wave): (45, +2.4), dashed line; (54, −58), dotted line; (65, −107), solid line. Red (square wave): (2.3, +2.2), dashed line; (3.1, −53), dotted line; (3.8, −108), solid line. Parameters: VRMS = 7 V, f = 500 kHz, r0 = 10 nm, and mean anion concentration cb = 400 μM.

[PtCl5(H 2O)]− (aq) + 4e− → Pt(s) + 5Cl−(aq) + H 2O (5)

4Cl−(aq) → 2Cl 2(g) + 4e−

includes platinum complexes as well as chlorine ions. Assuming a predominant dissociation into monovalent complexes as [PtCl 5 (H 2 O)] − and correspondingly Cl − (instead of [PtCl6]2−), the anion concentration was chosen to be 400 μM. For each waveform of the voltage signal, concentration profiles at three points in time are plotted in Figure 7. The first time is chosen shortly before polarity reversal of the electric field at the wire tip, and the other times, afterward. Because of the very different slew rates of square and sine waves, the corresponding times at equal electric fields for these cases differ strongly. Before the polarity change, the anions are attracted. For the square-wave signal, a higher concentration at the electrode surface is reached, but the concentration decays more strongly with distance from the surface than for the sine-wave signal. After reversal, anions are repelled and the surface concentrations strongly diminish. Note that for similar field strengths the concentrations in the tip vicinity (r − r0 < 3 nm) are much smaller for the sine-wave signal. We suggest that platinum reduction (cf. eq 5) as a precondition for platinum deposition is strongly affected by the electric field at the nanowire tip surface. For a square-wave signal, due to the steep signal slope at polarity reversal, many anions are still in the vicinity of the nanowire tip so that a field-activated reduction of platinum could occur. The reduced platinum atoms presumably deposit on the nanowire surface. In contrast, for a sine-wave signal with a comparatively slow increase in the electric field, platinum complexes have already been repelled a certain distance from the electrode surface before the electric field is sufficiently high to activate platinum reduction. If this distance is too large, then charge transfer from the electrode to the complex cannot take place and nanowire growth is inhibited. For voltage signals with two higher Fourier components (N = 2), nanowire growth was observed. However, the wires are branched and polycrystalline with a wide orientation variation of nanocrystallites (cf. Figures 2a, 3a, and 4). Obviously, the reduction of platinum is still possible in this case, although anionic platinum complexes are repelled further from the

(6)

Accordingly, the overall reaction reads [PtCl5(H 2O)](aq)− → 2Cl 2(g) + H 2O + Pt(s) + Cl(aq)− (7)

Let us emphasize that the detailed reaction path from anionic platinum complexes to metallic platinum is presently unknown. However, we hypothesize that the overall reaction is supported by the electric field and is accompanied by a charge transfer from the tip to the complexes. Note that the number of anions does not change in the overall reaction (eq 7), where [PtCl5(H2O)](aq)− is replaced by Cl(aq)−, and therefore the mean anion concentration in the solution is constant. Our calculations showed that in the vicinity of the nanowire tip the fraction of platinum complexes incorporated into the wire is small. Consequently, the ion fluxes at the electrode surfaces are negligible. This represents the boundary condition for eq 2. After applying an ac voltage to the electrodes, an asymptotically periodic state between the platinum consumption at the growing nanowire tip and the transport of new anionic complexes to the tip is established within some transient time. In this asymptotic state, the concentration of platinum complexes is proportional to the total concentration of all anions (Figure 7). In our theoretical analysis, we compared two waveforms of the voltage signal, namely, the sine wave (N = 1) and squarewave signals (N → ∞). In particular, we focused on the time when the electric field polarity changes from positive (anion attraction) to negative (anion repulsion and Pt reduction). Figure 7 shows results of our simulations of the spatiotemporal evolution of the anion concentration. Concentration profiles near the nanowire tip are plotted at different times, corresponding to different electric field strengths. The amplitude and frequency of the voltage were chosen as in the experiments. The total anion concentration in our calculations 5659

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661

Langmuir

Article

clusters seem to develop in the solution in front of the nanowire tip. Their random deposition could be responsible for the wire branching and polycrystalline structure. Further chemical analysis and more accurate modeling of the ionic motion are necessary to substantiate our present hypothesis on the platinum deposition mechanism. The controlled growth of nanowires with desired crystalline structure might be useful for certain applications as in nanoelectronics (electrical transport, and interconnects) or photonics (waveguides). Furthermore, intriguing applications might also appear in view of recently discovered ferromagnetism and the high magnetic anisotropy of nanosized platinum clusters and ultrathin platinum wires.29,30

electrode surface than for square-wave signals (cf. Figure 7). However, due to the larger distance from the wire surface, platinum deposition on the surface becomes less likely. Instead, agglomeration of platinum complexes to oligomers and nanoclusters takes place in the solution shortly before the nanowire tip. Depending on their charge state, these agglomerates can be attracted to or repelled from the wire tip by the Coulomb force. On the other hand, they can be attracted by the dielectrophoretic force. We assume that finally the mechanism of attraction is strongly dependent on the size of the attracted clusters.25 The deposition of single platinum atoms and tiny oligomers is expected to lead to crystallographically coherent nanowire growth. In contrast, the deposition of larger, randomly oriented nanoclusters at the tip could cause the observed variation in the orientation of nanocrystallites in the nanowires (Figure 3). Also, wire branching is presumably attributed to the random deposition of nanoclusters on the wire tip, which serve as a seed for the growth of branches (Figure 2). An analysis of the fractal dimension of the observed nanowire morphologies showed a decreasing dimension with a greater number of Fourier components in the waveform of the voltage signal (see Supporting Information). For nanowire growth by means of square-wave voltage signals, the deposition of platinum atoms or tiny oligomers presumably prevails compared to the deposition of platinum nanoclusters. This seems to be a prerequisite for the growth of straight and highly textured nanocrystalline or even singlecrystalline wires exhibiting [111] crystal orientation along the wire axis. In summary, we suggest that a high rate of the electric field change on the electrode surface at polarity reversal is crucial for growing straight nanowires from anionic platinum complexes. A high slew rate of the electric field can also be achieved for a sine-wave voltage with sufficiently high frequency. Then, due to shorter pulse duration, the increase in the concentration of attracted anionic complexes near the wire tip is smaller compared to that at low frequencies. Consequently, fewer complexes are reduced per pulse, and as a tendency, the formation of smaller platinum clusters is expected, which should be favorable for growing single-crystal nanowires.

■

ASSOCIATED CONTENT

S Supporting Information *

Analysis of the fractal dimension of the nanowires. Elemental analysis of the nanowire using EELS. This material is available free of charge via the Internet at http://pubs.acs.org.

■ ■

AUTHOR INFORMATION

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported by the European Union (European Social Fund and European Regional Development Fund) and the Free State of Saxony (Saechsische Aufbaubank) in the young researcher group InnovaSens (SAB no. 080942409) and via TP A2 (MolFunc/MolDiagnostik) of the Cluster of Excellence “European Center for Emerging Materials and Processes Dresden” (ECEMP). We gratefully acknowledge support from the German Excellence Initiative via the Cluster of Excellence EXC 1056 “Center for Advancing Electronics Dresden” (CfAED). We thank Hagen Eckert, Prof. Sybille Gemming, Dr. Denys Makarov, and Dr. Mark Rümmeli for fruitful discussions. The sample preparation for the elemental analysis in the Supporting Information was performed by Y. Ritz, Fraunhofer-Institute for Non-Destructive Testing in Dresden, Germany. The corresponding TEM images and electron energy loss spectroscopy were recorded by U. Mühle, Fraunhofer-Institute for Non-Destructive Testing in Dresden, Germany.

4. CONCLUSIONS We have presented the electrochemical growth of ultrathin polycrystalline platinum nanowires between electrodes and demonstrated that the variation of the waveform of the applied alternating voltage has a strong influence on the morphology as well as on the crystallinity of the nanowires. The waveform was varied by including different numbers of Fourier components of a Fourier series of a square-wave signal. For a pure sine wave, no growth occurred. With an increasing number of Fourier components, growth changed from branched dendritic-like to completely straight wire growth. Also, the crystallite orientations within the nanocrystalline wires changed from a wide distribution to a very narrow one with [111] crystal orientation parallel to the wire axis. A numerical estimation of the spatiotemporal evolution of the anion concentration near the nanowire tip suggested that fast polarity reversal of the electric field leads to comparatively high anion concentrations in the immediate tip vicinity at the beginning of anion repulsion. This should enable platinum reduction by electron transfer from the tip to the anionic platinum complexes. For slower field change, larger platinum

■

ABBREVIATIONS SEM, scanning electron microscopy; TEM, transmission electron microscopy; DENA, directed electrochemical nanowire growth; DEP, dielectrophoresis

■

REFERENCES

(1) He, B.; Morrow, T. J.; Keating, C. D. Nanowire Sensors for Multiplexed Detection of Biomolecules. Curr. Opin. Chem. Biol. 2008, 12, 522−528. (2) Nie, Z.; Petukhova, A.; Kumacheva, E. Properties and emerging applications of self-assembled structures made from inorganic nanoparticles. Nat. Nanotechnol. 2009, 5, 15−25. (3) Arico, A. S.; Bruce, P.; Scrosati, B.; Tarascon, J.-M.; van Schalkwijk, W. Nanostructured Materials for Advanced Energy Conversion and Storage Devices. Nat. Mater. 2005, 4, 366−377. (4) Wu, Z.; Grossman, J. C. Prediction of Ultra-High Aspect Ratio Nanowires from Self-Assembly. Nano Lett. 2008, 8, 2697−2705. (5) Wang, Z. L., Ed.; Nanowires and Nanobelts: Materials, Properties, and Devices; Kluwer Academic Publishers: Boston, 2003; Vol. I.

5660

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661

Langmuir

Article

(6) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim, F.; Yan, H. One-Dimensional Nanostructures: Synthesis, Characterization, and Applications. J. Adv. Mater. 2003, 15, 353−389. (7) Weber, W. M.; Geelhaar, L.; Unger, E.; Chèze, C.; Kreupl, F.; Riechert, H.; Lugli, P. Silicon to Nickel-Silicide Axial Nanowire Heterostructures for High Performance Electronics. Phys. Status Solidi B 2007, 244, 4170−4175. (8) Chen, J.; Wiley, B. J.; Xia, Y. One-Dimensional Nanostructures of Metals: Large-Scale Synthesis and Some Potential Applications. Langmuir 2007, 23, 4120−4129. (9) Hermanson, K. D.; Lumsdon, S. O.; Williams, J. P.; Kaler, E. W.; Velev, O. D. Dielectrophoretic Assembly of Electrically Functional Microwires From Nanoparticle Suspensions. Science 2001, 294, 1082− 1086. (10) Pohl, H. A. Dielectrophoresis; Cambridge University Press: Cambridge, U.K., 1978. (11) Flanders, B. N. Directed Electrochemical Nanowire Assembly: Precise Nanostructrue Assembly via Dendritic Solidification. Mod. Phys. Lett. B 2012, 26, 1130001. (12) Saito̅, Y. Statistical Physics of Crystal Growth; World Scientific: Singapore, 1996. (13) Ranjan, N.; Mertig, M.; Cuniberti, G.; Pompe, W. Dielectrophoretic Growth of Metallic Nanowires and Microwires: Theory and Experiments. Langmuir 2009, 26, 552−559. (14) Thapa, P. S.; Ackerson, B. J.; Grischkowsky, D. R.; Flanders, B. N. Directional Growth of Metallic and Polymeric Nanowires. Nanotechnology 2009, 20, 235307. (15) Gierhart, B. C.; Howitt, D. G.; Chen, S. J.; Smith, R. L.; Collins, S. D. Frequency Dependence of Gold Nanoparticle Superassembly by Dielectrophoresis. Langmuir 2007, 23, 12450−12456. (16) Bhatt, K. H.; Velev, O. D. Control and Modeling of the Dielectrophoretic Assembly of On-Chip Nanoparticle Wires. Langmuir 2004, 20, 467−476. (17) Kawasaki, J. K.; Arnold, C. B. Synthesis of Platinum Dendrites and Nanowires Via Directed Electrochemical Nanowire Assembly. Nano Lett. 2011, 11, 781−785. (18) Talukdar, I.; Ozturk, B.; Flanders, B. N.; Mishima, T. D. Directed growth of single-crystal indium wires. Appl. Phys. Lett. 2006, 88, 221907. (19) Ozturk, I.; Flanders, B. N.; Grischkowsky, D.; Mishima, T. D. Single-step growth and low resistance interconnecting of gold nanowires. Nanotechnology 2007, 18, 175707. (20) Ranjan, N.; Vinzelberg, H.; Mertig, M. Growing OneDimensional Metallic Nanowires by Dielectrophoresis. Small 2006, 2, 1490−1496. (21) Ditlbacher, H.; Hohenau, A.; Wagner, D.; Kreibig, U.; Rogers, M.; Hofer, F.; Aussenegg, F.; Krenn, J. Silver Nanowires as Surface Plasmon Resonators. Phys. Rev. Lett. 2005, 95, 257403. (22) Nerowski, A.; Poetschke, M.; Bobeth, M.; Opitz, J.; Cuniberti, G. Dielectrophoretic Growth of Platinum Nanowires: Concentration and Temperature Dependence of the Growth Velocity. Langmuir 2012, 28, 7498−7504. (23) Nerowski, A.; Opitz, J.; Baraban, L.; Cuniberti, G. Bottom-Up Synthesis of Ultrathin Straight Platinum Nanowires: Electric Field Impact. Nano Res. 2013, 6, 303−311. (24) Shelimov, B.; Lambert, J. F.; Che, M.; Didillon, B. Application of NMR to Interfacial Coordination Chemistry: A Pt-195 NMR Study of the Interaction of Hexachloroplatinic Acid Aqueous Solutions with Alumina. J. Am. Chem. Soc. 1999, 121, 545−556. (25) von Schnering, H. G.; Chang, J.-H.; Peters, K.; Peters, E.-M.; Wagner, F. R.; Grin, Y.; Thiele, G. Structure and Bonding of the Hexameric Platinum(II) Dichloride, Pt_6Cl_12 (β-PtCl_2). Z. Anorg. Allg. Chem. 2003, 629, 516−522. (26) Kumar, V.; Kawazoe, Y. Evolution of Atomic and Electronic Structure of Pt Clusters: Planar, Layered, Pyramidal, Cage, Cubic, and Octahedral Growth. Phys. Rev. B 2008, 77, 205418. (27) Olesen, L. H.; Bazant, M. Z.; Bruus, H. Strongly Nonlinear Dynamics of Electrolytes in Large AC Voltages. Phys. Rev. E 2010, 82, 011501.

(28) Poetschke, M.; Bobeth, M.; Cuniberti, G. Ion fluxes and electroosmotic fluid flow in electrolytes around a metallic nanowire tip under large applied ac voltage. Langmuir 2013, 29, 11525−11534. (29) Sakamoto, Y.; Oba, Y.; Maki, H.; Suda, M.; Einaga, Y.; Sato, T.; Mizumaki, M.; Kawamura, N.; Suzuki, M. Ferromagnetism of Pt Nanoparticles Induced by Surface Chemisorption. Phys. Rev. B 2011, 83, 104420. (30) Smogunov, A.; Dal Corso, A.; Delin, A.; Weht, R.; Tosatti, E. Colossal Magnetic Anisotropy of Monatomic Free and Deposited Platinum Nanowires. Nat. Nanotechnol. 2008, 3, 22−25.

5661

dx.doi.org/10.1021/la5002946 | Langmuir 2014, 30, 5655−5661