Growth Orientation Control of Co Nanowires Fabricated by

Article Cite This: Cryst. Growth Des. XXXX, XXX, XXX−XXX

pubs.acs.org/crystal

Growth Orientation Control of Co Nanowires Fabricated by Electrochemical Deposition Using Porous Alumina Templates Mingliang Wang,*,†,‡ Zhigang Wu,†,§ Hong Yang,† and Yinong Liu*,† †

School of Mechanical and Chemical Engineering, The University of Western Australia, Perth, Western Australia 6009, Australia School of Materials Science & Engineering, Shanghai Jiao Tong University, No. 800 Dongchuan Road, Shanghai 200240, China § School of Civil Engineering, Guangzhou University, 230 Daxuecheng Outer Ring W Road, Guangzhou, Guangdong 510006, China ‡

S Supporting Information *

ABSTRACT: This paper reports an experimental and theoretical analysis of preferential orientation growth of metallic nanowires during electrochemical deposition using nanochanneled templates. In this work pure Co nanowire arrays were synthesized by electrochemical deposition using porous anodized aluminum oxide templates. The nanowire arrays are found to exhibit near complete preferential single axial orientation. The preferential orientation changed with increasing the applied voltage from [0002]hcp, [1010̅ ]hcp, [12̅10]hcp to [110]fcc. The observation is explained in terms of nucleation thermodynamics and crystal growth kinetics. The analysis demonstrates that at low applied voltages, when the wire growth is slow, the wire orientation is dictated by the criterion of minimum total surface energy, with the close-packed surfaces forming the external facets of the crystals. At high applied voltages, when the wire growth is fast, the crystal axial orientation is dictated by the growth kinetics, i.e., directions of the fastest growth velocity. These criteria also apply well to the preferential growth of fcc metal nanowires during electrochemical deposition, e.g., Ag, Au, Cu, and Ni.

1. INTRODUCTION Metallic nanowire arrays created by electrochemical deposition using regular nanochannel templates have stimulated keen research interests in recent years.1−5 These materials are characteristic of high length-to-diameter aspect ratios (typically greater than 1000),1 small lateral dimensions (typically 50−90 nm), and high population density (typically 108 wires/mm2) or large numbers (typically in the order of 4 × 109 over an area of 5 × 5 mm) can be fabricated in rare or complex substructures, like single crystals,6,7 bamboo structures,8 or core/shell structures.9 In addition to the above, there is also the possibility to create a large quantity of metallic nanowires of near perfect, defect-free crystal structures, which have the potential to approach the theoretical limit of metallic bonds for materials strength and elasticity. Owing to these unique characteristics and potentials, metallic nanowire arrays hold the promise to offer a wide range of rare properties for innovative applications. Examples of the interests include magnetic storage media,10−12 nanosensors,13−15 and electrocatalysts16−18 and as model materials for the understanding of fundamental magnetism theories at the nanoscale.19−27 Much of the performance of the anticipated applications of metallic nanowire arrays depends on, or can be optimized by, the crystallographic orientation of the wires. For example, magnetic properties of a nanowire array are strongly dependent on crystal orientation due to magnetic anisotropy of the crystal.28−30 The ultimate strength and elasticity of nanowires © XXXX American Chemical Society

are also dependent on nanowire orientation, due to the elastic constant anisotropy of the crystal structure.31 Owing to these reasons, it is of interest to establish the knowledge and to develop techniques to manipulate and control the orientations of nanowire arrays. Electrochemical deposition using porous anodized aluminum oxide (AAO) templates is a popular technique used for nanowire synthesis. It has been reported in the literature that the electrochemically deposited nanowires often grow with certain preferential axial crystallographic orientations, which appear to be influenced by electrochemical deposition potentials.7,10,32,33 Narayanan et al. has suggested that metal electrodeposition using a porous template is associated with a model involving mobility assisted growth.34,35 The growth of metal crystallites during electrodeposition in porous materials originates from the cathode surface at the bottom edge of the pore. The high surface area and presence of sites with low coordination number in the porous part of the alumina afford energetically favorable sites for initiating metal adsorption during electrodeposition. The observed axial orientations include both the normal directions of close-packed planes (e.g., [0001]hcp and [111]fcc) and non-close-packed planes (e.g., [1010̅ ]hcp and [110]fcc).28,34,35 Received: October 19, 2017 Revised: November 25, 2017 Published: December 4, 2017 A

DOI: 10.1021/acs.cgd.7b01464 Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

nanowire growth frontier, leading to different growth orientations and possibly different structures (e.g., fcc Co). Pan et al.7,32 suggested that at higher deposition potentials, H+ ions in the electrolyte stabilize the crystal planes of high surface energies (e.g., (110)fcc) by surface adsorption, promoting their growth.7,36,48 Sun et al.48,49 favored this model in their study of electrodeposition of Ni nanowires. Nevertheless, these explanations are considered unconvincing due to inconsistency with the experimental results and lacking of a quantified generic criterion responsible for the orientation transition. In this work, we studied the growth behavior of oriented pure Co nanowires synthesized by electrochemical deposition using AAO templates, in particular, the applied voltage dependence of the axial orientations of the nanowires. An analysis based on thermodynamics and crystallography is presented to determine the comparative criteria for the preferential orientation selection of metallic nanowires during electrochemical deposition using nanochanneled templates.

It is well accepted that the close-packed planes normally have the lowest surface energies and are thus expected to form the front exposed surface of a growing nanowire crystal.36,37 However, the explanation for nanowires grown along the nonclose-packed planes has yet to reach a consensus. A popular early explanation of the preferred orientation in electrodeposited metals was given by Finch et al.38,39 It was claimed that when deposition is carried out at low current densities and high temperatures, the crystallites are oriented in such a way that the most densely populated atom plane is parallel to the substrate, and the crystals grow by lateral expansion. For hcp metals, the axis of the preferred orientation (normal to the substrate) is [0001] for lateral growth. If the deposition is carried out at low temperatures and high current densities, the crystallites of the deposits are often oriented with the most densely populated atom plane being perpendicular to the substrate, i.e., the outward type of growth. For the hcp metals the axis of the preferred orientation (normal to the substrate) is [112̅0] for outward growth. It is clear that this is more a description of the experimental observations of orientation selection under different deposition conditions than an explanation of why such selection occurs. In addition, it has been observed experimentally that preferred axial orientations other than these two directions also occur, e.g., [101̅0]hcp. In this regard, the [0001] (lateral type) and the [112̅0] (outward type) growths are only special cases of crystal growth. Pangarov later proposed a two-dimensional (2D) nuclei theory in 1965, which was claimed as a satisfactory explanation of the development of preferred orientation axes with regard to deposition conditions at different overvoltage levels.40 The main idea of this theory is that the type of 2D nucleus is decisive in determining the preferred orientation of the crystallites and the subsequent crystal growth, and that the growth rate is dictated by the work required to form a 2D nucleus of a given lattice plane (hkl), i.e., W(hkl). That said, the rate of the 2D nucleus formation process will be fastest for the lattice plane (hkl) for which the work is the lowest. The preferred orientation axes calculated for hcp-Co using this theory was in the sequence of [0001] (low overvoltage) → [112̅0] (intermediate overvoltage) → [101̅0] (high overvoltage). However, this is not consistent with many experimental observations,41−44 in which [101̅0] orientation occurs at lower voltages than [1120̅ ] orientation. Tan and Chen had also argued, in their model, that the preferential growth should depend on the number of sites for dehydration of hydrated metal ions on a metal surface.45 Co is a less noble metal relative to hydrogen, and thus coreduction of hydrogen (2H+ + 2e− → H2) is expected to occur concurrently with the Co deposition at the cathode. Hydrogen reduction reaction will certainly affect the deposition rate of Co as estimated from deposition current measurement. It has also been suggested that hydrogen coreduction on the cathode surface affects crystal growth during electrochemical deposition. The growth and deposition mechanism of electrodeposited cobalt nanowires at different bath temperatures (25−60 °C) has been studied by Kaur et al.46,47 They found that a mixture of hcp and fcc phases grows at 25 °C, (200) textured fcc-Co phase grows at 50 °C, and (101̅0) and (112̅0) textured hcp-Co phase at 60 °C. They proposed that hydrogen adsorption and desorption have different kinetics on different metal crystalline surface planes; thus under a high codeposition rate (or high deposition overpotential) hydrogen production on the cathode may demand different crystallographic planes on the metal

2. EXPERIMENTAL PROCEDURES Porous anodized aluminum oxide (AAO) templates were prepared following a two-step anodization procedure.50 A high purity (99.99 at %) Al thin plate of 400 μm thick was used. Sample coupons were ultrasonically cleaned in acetone, ethanol, and deionized water and then electrochemically polished in an aqueous solution of sulfuric acid (25 wt %), phosphoric acid (25 wt %), and deionized water (50 wt %). The first anodization was performed at 40 V in a 3 wt % oxalic acid aqueous solution at room temperature for 6 h. The porous alumina formed was removed using an etchant solution containing 18 g/L chromic acid and 60 g/L phosphoric acid at 80 °C for 1 h. The second anodization was performed under the same conditions. The remaining Al was removed in a 10 wt % CuCl2 aqueous solution. The barrier layer was removed by etching in a 5 wt % H3PO4 aqueous solution at 30 °C for 60 min. The etching also widened the pore diameter to ∼60 nm. Electrochemical deposition of nanowires was performed in a twoelectrode cell. To facilitate electrodeposition, a 200 nm gold layer was coated by means of thermal evaporation onto the back of the AAO templates as the working electrode. The exposed AAO template surface was a circular area of 13 mm in diameter. Considering that the planar porosity of the AAO template was 34%, the exposed Au cathode area at the end of the deep AAO channels may be estimated to be ∼45 mm2. A Ti plate of 30 × 50 × 1 mm (with a surface area of ∼3000 mm2 from both sides) was used as the anode. The distance between the anode and the cathode was ∼60 mm. The electrolyte (pH = 4.6) used for Co nanowires deposition was an aqueous solution containing 0.2 M CoSO4 and 0.1 M H3BO3 at room temperature. A series of cell voltages (U) from 2.2 to 5.4 V were applied using a DC regulated power supply. The structure and morphology of the deposited nanowires were characterized by means of X-ray powder diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). XRD was conducted using a Siemens D5000 diffractometer with Cu Kα radiation (λ = 0.1542 nm). XRD was conducted on the surface perpendicular to the lengths of the nanowires. SEM examination was conducted using a Zeiss 1555 field emission scanning electron microscope. TEM was conducted using a JEOL 2100 analytical microscope. TEM samples were prepared as follows. A small piece of AAO/Co nanowires sample was immersed in 1 M NaOH solution at 30 °C for 1 h to dissolve the alumina template. The remaining Co nanowires, with some of them still being attached to the gold film and some having detached from the gold substrate and become loose wires, were washed using distilled water and ethanol for several times. The loose Co nanowires were dispersed into ethanol ultrasonically, and a drop of this solution was loaded on a Cu grid for TEM characterization. B



Article

3. RESULTS Figure 1 shows SEM micrographs of (a) an anodized porous alumina template (top view) and (b) Co nanowire arrays

deposition. The deposition current increased with increasing deposition voltage, as shown in Figure 2b, indicating an increase of the rate of deposition, as also confirmed by the observation that shorter times were required under higher deposition cell voltages for the deposition of nanowires of similar lengths. The integral of the current curves over time is a practical estimate of the total material deposited and the current at any time is a measure of the growth rate of the Co nanowires at that moment, as per eq 1. m=

Figure 1. SEM micrographs (a) a porous alumina template and (b) Co nanowires deposited at 3.0 V using the template (the alumina template has been removed by chemical etching in 1 M NaOH solution).

IM nF

(1)

where m is the growth rate (g/m), I is the current, M is the molar mass of the element (for cobalt M = 58.93 g/mol), n is the valence number (for cobalt n = 2), and F is the Faraday constant (F = 96485.3 C/mol). It should be noted that such growth rate calculated is only valid with the assumption of 100% current efficiency for deposition, which is unlikely the case in a real situation due to concurrent hydrogen coreduction on the cathode. Figure 3 shows XRD analysis of the deposited Co nanowires. Figure 3a are the XRD spectra of the Co nanowires deposited at different applied voltages. The XRD spectra were taken of the plane perpendicular to the wire axial directions, as

deposited at 3.0 V using the template. It is seen that the pores are in hexagonally ordered arrangement with diameters of ∼70 nm. In Figure 1b, the alumina template has been dissolved in 1 M NaOH solution, exposing the wires. The Co nanowires had a length of ∼40 μm. Figure 2 shows the current−time curves of the deposition of Co nanowires at different deposition voltages (Figure 2a). It is seen that after an initial drop the current at each deposition voltage stabilized at a constant level during the period of

Figure 3. Deposition voltage dependence of preferential axial orientation of Co nanowires. (a) XRD spectra of electrochemically deposited Co nanowires; (b) schematic of the spatial relationship between the nanowires and the X-ray for XRD experiments.

Figure 2. Deposition of Co nanowires. (a) Current−time curves under different applied voltages; (b) effect of applied voltage on steady state deposition current of Co nanowires. C



Article

applied voltage. This is in total agreement with the XRD observations shown in Figure 3. It is seen that there is a polymorphism transition induced by higher applied voltages, from hcp-Co to fcc-Co at U ≥ 3.6 V.

schematically shown in Figure 3b. At the bottom of the spectra standard diffraction peak positions and intensities of hcp-Co and fcc-Co are shown for reference. The XRD spectra of the samples deposited at U = 2.2 and 2.4 V are fully indexed to hcpCo with clear preferential orientation of (0002)hcp perpendicular to the wire axes. A small diffraction peak for (101̅0)hcp is also visible. When deposited at U = 2.7 V, the Co nanowires displayed complete orientation selection of (101̅0)hcp. At U = 3.0 V, the Co nanowires showed a clear peak at 2θ = 75.9°. At the meantime, the (101̅0)hcp peak (at 2θ = 41.7°) has become much weaker, as shown in the inset. At U ≥ 3.6 V, the peak at 2θ = 75.9° became the sole peak, indicating complete orientation selection. This peak can be indexed to either (112̅0)hcp or (220)fcc, which have practically the same 2θ position for Co. Figure 4 shows TEM micrographs with the corresponding selected area electron diffraction (SAED) patterns of Co

4. ANALYSIS AND DISCUSSION 4.1. Governing Criteria for Preferential Axial Orientation of Co Nanowire Growth. The nucleation of metal nanowires by electrochemical deposition has been discussed by many researchers6,7,32,48 based on the classic thermodynamics nucleation theory. It is considered that in electrochemical deposition, the free energy of formation of a nucleus of N atoms (ΔG(N)) has two components:36 ΔG(N ) = −ΔGv (N ) + φ(N )

(2)

where ΔGv(N) is the volume free energy change related to the transfer of N ions from solution to the crystal phase nucleus, and φ(N) is the surface energy of the nucleus created. ΔGv(N) is a function of the applied voltage (more precisely the reduction potential of the ions on the crystal electrode) and provides the driving force for the nucleation, whereas φ(N) is a resistive force. Therefore, ΔG(Nc) = 0, or ΔGv(Nc) = φ(Nc), defines the critical size of the nucleus. On the basis of this theory, Plieth et al.37 proposed the relation between the critical dimension Nc (minimum number of atoms in the nucleus) and the surface energy of the formed crystal plane (hkl). Considering a metallic nanowire nucleus as a hemisphere-shaped crystal with radius of r, the critical dimension Nc can be determined as NC =

16πσhkl 3 3z 3F 3 |η|3

(3)

where z is the ionic valence, η is the overpotential of the reduction reaction of the ions, and σhkl is the surface energy of the crystal face (hkl) that forms the exterior of the nucleus, and F is Faraday constant. It is clear that crystal planes of low surface energies are more favored as the exterior surface of nuclei for nucleation. The surface energy of an atomic plane is related to the planar atomic packing density of the plane. Surface energies of hcp-Co structure increase in the order of {0001}, {101̅0}, and {12̅10}.51 Table 1 shows structural parameters of Co (hcp)

Figure 4. TEM micrographs and SAED patterns of Co nanowires deposited at different applied voltages: (a) U = 2.2 V, (b) U = 2.7 V, (c) U = 3.0 V, and (d) U = 3.6 V; (e) shows a schematic of the dependence of the preferential axial orientation of the wires on applied voltage.

Table 1. Structural Parameters of Co (hcp) Crystal Planes

nanowire samples deposited under different cell voltage levels. Micrograph (a) shows a wire deposited at U = 2.2 V. The single strand Co wire is ∼60 nm in diameter. The wire is single crystalline and the growth direction is [0002]hcp, as derived from its SAED pattern. SAED patterns taken at different locations along the length of the wire confirm the single crystallinity and the common axial orientation. Similarly, the wires deposited at U = 2.7 V (Figure 4b), 3.0 V (Figure 4c), and 3.6 V (Figure 4d) are also found to be single crystals. The axial orientations of the nanowires are determined to be [101̅0]hcp for U = 2.7 V, [12̅10]hcp for U = 3.0 V and [110]fcc for U = 3.6 V. In addition, the SAED allowed the differentiation between (12̅10)hcp and (220)fcc. Figure 4e presents a schematic of the dependence of the preferential orientation on applied voltage, as observed experimentally. The Co nanowire orientation evolves from [0002]hcp, [101̅0]hcp, [12̅10]hcp to [110]fcc with increasing

low index plane

planar atomic packing density

dspace (R: atomic radius)

surface energy (for Co51) (J/m2)

(0001) (101̅0) (12̅10)

0.907 0.802 0.555

1.64R 1.16R R

2.994 3.421 4.841

crystal planes, including the planar atomic packing density, interplane spacing (dspace) and surface energy of the three low index planes. Crystal planes of higher atomic packing densities generally have smaller surface energies. It is seen that the surface energies of the planes follow the increasing order of σ(0001) < σ(10 1̅ 0) < σ(1210) (4) ̅ Referring to the above, it is easy to see that the critical nucleus sizes when the respective planes are exposed for hcp Co follow the relationship of NC(0001) < NC(10 1̅ 0) < NC(1210) ̅ D

(5)



Article

planar coordination number of the three low index planes has the relationship of n(0001) < n(10 1̅ 0) < n(1210) (6) ̅

According to this criterion, nuclei with (0001) as the exposed surface are the easiest to form; i.e., the (0001) plane will be the face plane of the disk shaped nuclei. This is consistent with the preferred orientation of the Co nanowires deposited at low voltages (U ≤ 2.4 V), when the deposition rate is low. 4.2. Crystal Growth Kinetics. In this work, Co nanowires in [101̅0] and [12̅10] preferential axial orientations are formed at higher applied voltages. This obviously deviates from the minimum surface energy criterion expressed above. This growth habit is attributed to the influence of growth kinetics, or the difference in growth rate of the crystal in different directions. The kinetics of growth of a crystal in the direction normal to a plane is determined by the probability of an arriving atom to adhere to a site on the plane. The probability of adherence is determined by the strength of the interatomic bonds within the solid and the number of bonds of the site. For a given crystal, the bond strength between contacting atoms is the same regardless of the plane surface in question, and thus the number of such bonds is the determining factor for the growth probability. The number of bonds for a receiving site for an arriving atom is the surface coordination number. Figure 5

This implies that an atom in the solution will find it easiest to adhere to the (12̅10) plane and least ease to adhere to the (0001) plane. Reciprocally, it is also correct to say that an atom adhered to the (0001) plane is much easier to break away from the solid plane into the solution than an atom adhered to (12̅10). This discussion is schematically illustrated in Figure 6. Figure 6a illustrates a nanowire electrochemically deposited inside an

Figure 6. Schematic drawings of (a) nanowire electrochemical deposition along AAO channels, (b) free energy analysis of electrochemical deposition process at a given applied voltage, and (c) summary of the influence of applied voltage for axial preferential of electrochemically synthesized metallic nanowires.

AAO channel, and Figure 6b expresses the free energy state of an atom during electrochemical deposition at a given applied voltage. An ion in the solution has the same energy state regardless of onto which plane it may be deposited. For an ion in the electrolyte to convert to one atom in the solid, it needs to overcome an energy barrier, i.e., the activation energy of the conversion, of Q1. At the same time, an exposed atom on the surface of the solid may also ionize back into the electrolyte. The activation energy required for this conversion will be different for different planes, because of the different coordination numbers discussed above. The (121̅ 0) plane has the highest coordination number, and thus for an atom to break away from this plane it requires the highest activation energy Q2. In comparison, an atom adhered to the (0001) plane has the highest energy state (least stable) due to its fewest coordination number among the three planes. Therefore, it requires the least activation energy to break away, as illustrated in Figure 6b. Thus, following relationship (6), relationship (7) can be established:

Figure 5. Structure analysis of hcp crystal planes.

presents a structure analysis of the main low-index crystal planes of hcp structure. The first column shows the top view of the growing plane (growing in the normal direction to the plane). The atom in red is an arriving atom (ion) from the solution. The second column shows the side view parallel to the growing plane. The third column shows the stacking order of the growing plane. The fourth column shows the coordination number for the arriving atom on the growing plane. The first row shows the case of the (0001) plane. This is the closed-packed plane of hcp structure and has a stacking order of [AB] along the [0001] direction. The arriving atom has three nearest neighbors, i.e., a coordination number of 3. For the (101̅0) plane, the stacking order is [ABCD] along its normal direction. The arriving atom has four nearest coordinators, including two in the top layer and two in the layer below. The situation of (12̅10) plane is shown in the last row. This plane has a stacking order of [AB] and a coordination number of 5 (four in the top layer and 1 in the layer below). Therefore, the

Q 2(0001) < Q 2(10 1̅ 0) < Q 2(1210) ̅

(7)

The probability (P1) of one ion from the electrolyte to convert to an atom and adhere to a site on the solid is dictated by the activation energy of this process, as per an Arrhenius type equation: P1 = E

⎛ Q ⎞ kBT exp⎜ − 1 ⎟ ⎝ RT ⎠ h

(8) DOI: 10.1021/acs.cgd.7b01464 Cryst. Growth Des. XXXX, XXX, XXX−XXX


Article

Table 2. Summary of Electrodeposited Element Nanowires of Typical hcp and fcc Metals metal

structure

growth direction

deposition condition

nanowires in template or Films on substrate

Co

NW

28, 58

fcc

NW

59

Au

fcc

NW

6

Cu

fcc

NW

Cu

fcc

NW

60 61 62

Ni

fcc

NW

49

Ni

fcc

−1.45 V (applied V) −3.0 V (applied V) 5 mA/cm2 50 mA/cm2 −0.05 V (Ag/AgCl) −0.25 V (Ag/AgCl) −0.75 V (SCE) −1.2 V (SCE) −0.15 V (SCE) −0.45 V (SCE) 0.5 V (SCE) 2 V (SCE) 0.5 mA/cm2 2.5 mA/cm2 0.4 V (applied V) 4 V (applied V)

57

Ag

[101̅0] [110] [0001] [101̅0] [111] [110] [111] [111]&[100]&[110] [111] [110] [111]&[100]&[110] [110] [111]&[100]&[110] [110] [111]&[100]&[110] [110]

NW

Co

hcp fcc hcp

NW

7

where kB is Boltzmann Constant, T is the absolute temperature, h is Plank constant, R is the ideal gas constant, and Q1 is the activation energy for the conversion of one ion in the electrolyte to one atom. At the same time, the probability (P2) of one atom adhered on an atomic plane migrating into the electrolyte is expressed as P2 =

⎛ Q 2(hkl) ⎞ kBT ⎟ exp⎜ − h RT ⎠ ⎝

growth when the kinetic growth characters of different crystallographic directions and planes are intrinsically altered, for example by ion adsorption in solution7,32,53−55 or changing the chemical bond of a certain crystallographic plane of the crystal.52 Sun et al.52 studied the formation of ZnO nanocrystals and reported that the primary growth directions can be altered by modifying surface bonding lengths between of Zn−O. Chen et al.53 synthesized Cu2O crystals using hydrothermal method, and the growth orientations and shapes of Cu2O crystals can be tuned using surfactants. In the case of electrochemical deposition of nanowires, a kinetic model for growth orientation of nanowires has been established in terms of surface adsorption of H+ ions.7,32 However, this model is phenomenological and lacks of a clear criterion for the orientation selection. In another study, Li et al.56 reported the kinetic crystal growth by oriented attachment of iron oxyhydroxide nanoparticles. However, this is more a case of nanocrystal assembling than crystal growth.” 4.3. Crystal Growth of fcc Metals. The above analysis can also be applied to fcc metals (e.g., Ag, Au, Cu, and Ni). Table 2 shows a summary of electrodeposited elemental nanowires of typical hcp and fcc metals. It is seen that the axial orientations of fcc metal nanowires evolve with increasing deposition voltage in the order of

(9)

where Q2(hkl) is the activation energy for the conversion of an atom in the solid (hkl) plane to an ion in the electrolyte. It is obvious that P2 is (hkl) dependent, as discussed above. The difference between P1 and P2 is the net flux of atoms (ions) between the electrolyte and the solid, i.e., the growth rate of the crystal. Therefore, the probability of net deposition along the normal direction of a (hkl) plane at a given applied voltage is given as P[hkl] = P1 − P2 =

⎛ Q 2(hkl) ⎞⎤ kBT ⎡ ⎛ Q 1 ⎞ ⎢exp⎜ − ⎟⎥ ⎟ − exp⎜ − h ⎢⎣ ⎝ RT ⎠ RT ⎠⎥⎦ ⎝ (10)

Considering eqs 7 and relationship (10), the net atomic adhesion probabilities, i.e., the growth rates along the normal directions of the three planes during electrochemical deposition, can be expressed as P[0001] < P[10 1̅ 0] < P[1210] ̅

reference

[111] → [100] → [110]

(12)

Summarizing all the information available in the literature as well as the findings of this work, the orientation dependence of electrochemically deposited metallic nanowires on deposition voltage may be expressed as in Figure 6c. Table 3 shows crystal plane parameters for the three lowindex crystal planes of fcc structure, as in the case of Ni, including the crystal planar atomic packing density, interplane spacing (dspace) and surface energy. It is seen that the surface energy (e.g., Ni63) is in the order of σ(111) < σ(100) < σ(110),

(11)

Summarizing the discussions above, it can be concluded that at low applied voltages, when the growth rate is low, the formation orientation of the nanowire crystals is dictated by minimum surface energy criterion, i.e., thermodynamically controlled. At high applied voltages, when the growth rate is high, the nanowire axial orientation is dictated by the growth kinetics criterion. These predictions are obviously consistent with the experimental observations summarized in Figure 4e. Kinetics-controlled crystal growth has also been reported in other studies.2,11,52−56 The intrinsic growth rate of a specific crystal plane can be predicted by chemical bond theory calculation.52−54 This study reveals the selection of preferential axial direction of single crystal nanowires when the kinetic condition is changed externally (applied voltage in its case). Some early studies discussed kinetically controlled crystal

Table 3. Structure Parameters of Ni (fcc) Crystal Planes

F

low index plane

planar atomic packing density

dspace (R: atomic radius)

surface energy (for Ni63) (J/m2)

(111) (100) (110)

0.907 0.802 0.555

1.64R 1.414R R

1.171 1.305 1.417



Article

implying that the critical dimension for nanowire nucleation is in the order of NC(111) < NC(100) < NC(110)

At low applied voltages the axial orientation is dictated by nucleation thermodynamics, which obeys the criterion of minimum surface energy. This criterion dictates that the round face of a thin disk shaped nucleus is (0001) for hcp metals and (111) for fcc metals. At higher applied voltages, the axial orientation is dictated by the criterion of maximum crystal growth kinetics. Crystal planes of higher planar coordination numbers have higher growth rates, and thus their normal directions are the preferred growth directions of the nanowires. This criterion dictates that the axial orientation is the normal direction of (121̅ 0) for hcp metals and (110) for fcc metals. (3) The criteria for nanowire axial preferential orientation in electrochemical deposition proposed above appears to be applicable to both hcp and fcc metals.

(13)

Figure 7 shows the structure analysis of the three low-index crystal planes of fcc structure (calculation of the crystal planar

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.cgd.7b01464. For Co (hcp) structure, the crystal planar atomic packing density of three low index planes, including (0001), (101̅0), and (12̅10), are calculated in Figure S1. For fcc structure, the crystal planar atomic packing density of three low index planes, including (100), (110), and (111), are calculated in Figure S2 (PDF)

Figure 7. Structure analysis of fcc crystal planes.

■

atomic packing density are shown in Figure S2). It is seen that the planar coordination number of the planes, hence the growth rate of the nanowires in the normal directions, is in the order of n(111) < n(100) < n(110) (14)

AUTHOR INFORMATION

Corresponding Authors

*(Y.L.) E-mail: [email protected]. Tel: 61 8-6488-3132. *(M.W.) E-mail: [email protected]. ORCID

Mingliang Wang: 0000-0003-4866-9371

Relationships (13) and (14) imply that for electrochemical deposition of fcc metal nanowires, [111] is the preferential axial orientation at low applied voltages, due to low surface energy of the (111) plane, and [110] is the preferential axial orientation at high applied voltages, due to its high growth kinetics. This conclusion is obviously consistent with the experimental observations summarized in Table 2. As a final note, we wish to point out that we did not attempt to analyze and explain the polymorphic change of electrochemically deposited Co nanowires from hcp to fcc with increasing the deposition voltage to above 3.6 V in the given system. For Co, (12̅10)hcp and (110)fcc planes have the same structure parameters, including planar atomic packing density, planar coordination number, and interplane spacing. The theory presented above does not provide a means to explain this transition.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the facilities and the technical assistance of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy, Characterisation & Analysis, The University of Western Australia, a facility funded by the University, State and Commonwealth Governments, and the support of the Western Australian node of the Australian National Fabrication Facility, The University of Western Australia.

■

REFERENCES

(1) Kuchibhatla, S. V. N. T.; Karakoti, A. S.; Bera, D.; Seal, S. One dimensional nanostructured materials. Prog. Mater. Sci. 2007, 52, 699− 913. (2) Shingubara, S. Fabrication of nanomaterials using porous alumina templates. J. Nanopart. Res. 2003, 5, 17−30. (3) Li, W.; Zhang, J.; Shen, T.; Jones, G. A.; Grundy, G. J. Magnetic nanowires fabricated by anodic aluminum oxide template-a brief review. Sci. China: Phys., Mech. Astron. 2011, 54, 1181−1189. (4) Chan, T. C.; Lin, Y. M.; Tsai, H. W.; Wang, Z. M.; Liao, C. N.; Chueh, Y. L. Growth of large-scale nanotwinned Cu nanowire arrays from anodic aluminum oxide membrane by electrochemical deposition process: controllable nanotwin density and growth orientation with enhanced electrical endurance performance. Nanoscale 2014, 6, 7332− 7338.

5. CONCLUSIONS (1) In this work, pure Co nanowires are synthesized by electrochemical deposition into the nanochannels of porous anodized aluminum oxide templates. The nanowires formed are predominantly single crystals with uniform preferential axial orientation. The preferential orientation is found to change with increasing deposition voltage, in the order of [0002]hcp, [1010̅ ]hcp, [121̅ 0]hcp, and [110]fcc. (2) The variation of the preferential axial orientation of the nanowires is explained in terms of two governing criteria. G



Article

anodic alumina membranes. J. Solid State Electrochem. 2014, 18, 2963− 2972. (25) Ivanov, Y. P.; Trabada, D. G.; Chuvilin, A.; Kosel, J.; ChubykaloFesenko, O.; Vázquez, M. Crystallographically driven magnetic behaviour of arrays of monocrystalline Co nanowires. Nanotechnology 2014, 25, 475702. (26) Holanda, J.; Silva, D. B. O.; Padrón-Hernández, E. Angular dependence of the coercivity in arrays of ferromagnetic nanowires. J. Magn. Magn. Mater. 2015, 378, 228−231. (27) Esmaeili, A.; AlmasiKashi, M.; Ramazani, A.; Montazer, A. H. Tailoring magnetic properties in arrays of pulse-electrodeposited Co nanowires: The role of Cu additive. J. Magn. Magn. Mater. 2016, 397, 64−72. (28) Darques, M.; Encinas, A.; Vila, L.; Piraux, L. Controlled changes in the microstructure and magnetic anisotropy in arrays of electrodeposited Co nanowires induced by the solution pH. J. Phys. D: Appl. Phys. 2004, 37, 1411−1416. (29) Cullity, B. D.; Graham, C. D. Introduction to Magnetic Materials; JohnWiley & Sons, Inc.: Hoboken, NJ, 2008. (30) Zhang, J.; Jin, Y.; Wang, H.; Ye, C.; Tong, W.; Wang, H. Growth and magnetic properties of single crystalline Ni nanowire arrays prepared by pulse DC electrodeposition. Sci. China: Phys., Mech. Astron. 2011, 54, 1244−1248. (31) Dubois, S.; Colin, J.; Duvail, J. L.; Piraux, L. Evidence for strong magnetoelastic effects in Ni nanowires embedded in polycarbonate membranes. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, 14315−14318. (32) Pan, H.; Sun, H.; Poh, C.; Feng, Y.; Lin, J. Single-crystal growth of metallic nanowires with preferred orientation. Nanotechnology 2005, 16, 1559−1564. (33) Zhu, Y.; Dou, X.; Huang, X.; Li, L.; Li, G. Thermal Properties of Bi Nanowire Arrays with Different Orientations and Diameters. J. Phys. Chem. B 2006, 110, 26189−26193. (34) Narayanan, T. N.; Shaijumon, M. M.; Ajayan, P. M.; Anantharaman, M. R. Synthesis of High Coercivity Cobalt Nanotubes with Acetate Precursors and Elucidation of the Mechanism of Growth. J. Phys. Chem. C 2008, 112, 14281−14285. (35) Narayanan, T. N.; Shaijumon, M. M.; Ci, L.; Ajayan, P. M.; Anantharaman, M. R. On the growth mechanism of nickel and cobalt nanowires and comparison of their magnetic properties. Nano Res. 2008, 1, 465−473. (36) Paunovic, M.; Schlesinger, M. Fundamentals of Electrochemical Deposition; Wiley: New York, 1998. (37) Plieth, W. Electrochemistry for Materials Science; Elsevier: Oxford, 2008. (38) Finch, G. I.; Wilman, H.; Yang, L. Crystal growth at the cathode. Discuss. Faraday Soc. 1947, 1, 144−158. (39) Burgers, W. G.; Dippel, C. J. On the crystalline state of thin calciumfluoride films. Physica 1934, 1, 549−560. (40) Pangarov, N. A. Preferred orientations in electro-deposited metals. J. Electroanal. Chem. 1965, 9, 70−85. (41) Huang, X.; Li, L.; Luo, X.; Zhu, X.; Li, G. OrientationControlled Synthesis and Ferromagnetism of Single Crystalline Co Nanowire Arrays. J. Phys. Chem. C 2008, 112, 1468−1472. (42) Wang, X. W.; Yuan, Z. H.; Sun, S. Q.; Duan, Y. Q.; Bie, L. J. Thermal expansion behaviors of hcp and fcc Co nanowire arrays. Phys. Lett. A 2009, 373, 2887−2889. (43) Mehmood, T.; Shahzad Khan, B.; Mukhtar, A.; Chen, X.; Yi, P.; Tan, M. Mechanism for formation of fcc-cobalt nanowires in electrodeposition at ambient temperature. Mater. Lett. 2014, 130, 256−258. (44) Cortés, A.; Lavín, R.; Denardin, J. C.; Marotti, R. E.; Dalchiele, E. A.; Valdivia, P.; Gómez, H. Template assisted electrochemical growth of cobalt nanowires: influence of deposition conditions on structural, optical and magnetic properties. J. Nanosci. Nanotechnol. 2011, 11, 3899−3910. (45) Tan, M.; Chen, X. Growth Mechanism of Single Crystal Nanowires of fcc Metals (Ag, Cu, Ni) and hcp Metal (Co) Electrodeposited. J. Electrochem. Soc. 2012, 159, K15−K20.

(5) Zhang, L.; Zhang, J.; Yang, C.; Zhao, G.; Mu, J.; Wang, Y. Freestanding Cu nanowire arrays on Ti/Cr/Si substrate as tough nonenzymatic glucose sensors. RSC Adv. 2015, 5, 82998. (6) Tian, M.; Wang, J.; Kurtz, J.; Mallouk, T. E.; Chan, M. H. W. Electrochemical Growth of Single-Crystal Metal Nanowires via a TwoDimensiona lNucleation and Growth Mechanism. Nano Lett. 2003, 3, 919−923. (7) Pan, H.; Liu, B.; Yi, J.; Poh, C.; Lim, S.; Ding, J.; Feng, Y.; Huan, C. H. A.; Lin, J. Growth of Single-Crystalline Ni and Co Nanowires via Electrochemical Deposition and Their Magnetic Properties. J. Phys. Chem. B 2005, 109, 3094−3098. (8) Su, Y. K.; Qin, D. H.; Zhang, H. L.; Li, H.; Li, H. L. Microstructure and magnetic properties of bamboo-like CoPt/Pt multilayered nanowire arrays. Chem. Phys. Lett. 2004, 388, 406−410. (9) Zhang, D.; Liu, Z.; Han, S.; Li, C.; Lei, B.; Stewart, M. P.; Tour, J. M.; Zhou, C. Magnetite (Fe3O4) Core:Shell Nanowires: Synthesis and Magnetoresistance. Nano Lett. 2004, 4, 2151−2155. (10) Chen, S. F.; Wei, H.; Liu, C. P.; Hsu, C. Y.; Huang, J. C. A. Microstructural effects on the magnetic and magneto-transport properties of electrodeposited Ni nanowire arrays. Nanotechnology 2010, 21, 425602. (11) Nielsch, K.; Müller, F.; Li, A. P.; Gösele, U. Uniform Nickel Deposition into Ordered Alumina Pores by Pulsed Electrodeposition. Adv. Mater. 2000, 12, 582−586. (12) Zeng, H.; Zheng, M.; Skomski, R.; Sellmyer, D. J.; Liu, Y.; Menon, L.; Bandyopadhyay, S. Magnetic properties of self-assembled Co nanowires of varying length and diameter. J. Appl. Phys. 2000, 87, 4718−4720. (13) Whitney, T. M.; Jiang, J. S.; Searson, P. C.; Chien, C. L. Fabrication and Magnetic Properties of Arrays of Metallic NanoWires. Science 1993, 261, 1316−1319. (14) Guo, F.; Cheng, K.; Ye, K.; Wang, G.; Cao, D. Preparation of nickel-cobalt nanowire arrays anode electro-catalyst and its application in direct urea/hydrogen peroxide fuel cell. Electrochim. Acta 2016, 199, 290−296. (15) Kurowska, E.; Brzózka, A.; Jarosz, M.; Sulka, G. D.; Jaskuła, M. Silver nanowire array sensor for sensitive and rapid detection of H2O2. Electrochim. Acta 2013, 104, 439−447. (16) Leontiev, A. P.; Brylev, O. A.; Napolskii, K. S. Arrays of rhodium nanowires based on anodic alumina: Preparation and electrocatalytic activity for nitrate reduction. Electrochim. Acta 2015, 155, 466−473. (17) Mahshid, S. S.; Mahshid, S.; Dolati, A.; Ghorbani, M.; Yang, L.; Luo, S.; Cai, Q. Template-based electrodeposition of Pt/Ni nanowires and its catalytic activity towards glucose oxidation. Electrochim. Acta 2011, 58, 551−555. (18) Ruan, D.; Gao, F.; Gu, Z. Enhanced Electrochemical Properties of Surface Roughed Pt Nanowire Electrocatalyst for Methanol Oxidation. Electrochim. Acta 2014, 147, 225−231. (19) Kim, S. I.; Lee, J. H.; Chang, Y. W.; Hwang, S. S.; Yoo, K. H. Reversible resistive switching behaviors in NiO nanowires. Appl. Phys. Lett. 2008, 93, 033503. (20) Piraux, L.; George, J. M.; Despres, J. F.; Leroy, C.; Ferain, E.; Legras, R.; Ounadjela, K.; Fert, A. Giant magnetoresistance in magnetic multilayered nanowires. Appl. Phys. Lett. 1994, 65, 2484− 2486. (21) Liu, K.; Nagodawithana, K.; Searson, P. C.; Chien, C. L. Perpendicular giant magnetoresistance of multilayered Co/Cu nanowires. Phys. Rev. B: Condens. Matter Mater. Phys. 1995, 51, 7381−7384. (22) Cattaneo, L.; Franz, S.; Albertini, F.; Ranzieri, P.; Vicenzo, A.; Bestetti, M.; Cavallotti, P. L. Electrodeposition of hexagonal Co nanowires with large magnetocrystalline anisotropy. Electrochim. Acta 2012, 85, 57−65. (23) Srivastav, A. K.; Shekhar, R. Crystal anisotropy induced temperature dependent magnetization in cobalt nanowires electrodeposited within alumina template. J. Magn. Magn. Mater. 2014, 349, 21−26. (24) Dobosz, I.; Gumowska, W.; Czapkiewicz, M. Structure and magnetic properties of Conanowires electrodeposited into the pores of H



Article

(46) Kaur, D.; Pandya, D. K. Hydrogen Co-Deposition Induced Phase and Microstructure Evolution of Cobalt Nanowires Electrodeposited in Acidic Baths. J. Electrochem. Soc. 2016, 163, D221−D229. (47) Kaur, D.; Pandya, D. K.; Chaudhary, S. Texture Changes in Electrodeposited Cobalt Nanowires with Bath Temperature. J. Electrochem. Soc. 2012, 159, D713−D716. (48) Sun, H.; Li, X.; Chen, Y.; Li, W.; Li, F.; Liu, B.; Zhang, X. The control of the growth orientations of electrodeposited single-crystal nanowire arrays: a case study for hexagonal CdS. Nanotechnology 2008, 19, 225601. (49) Sun, H.; Yu, Y.; Li, X.; Li, W.; Li, F.; Liu, B.; Zhang, X. Controllable growth of electrodeposited single-crystal nanowire arrays: The examples of metal Ni and semiconductor ZnS. J. Cryst. Growth 2007, 307, 472−476. (50) Masuda, H.; Fukuda, K. Ordered metal nanohole arrays made by a two-step replication of honeycomb structures of anodic alumna. Science 1995, 268, 1466−1468. (51) Zhang, J. M.; Wang, D. D.; Xu, K. W. Calculation of the surface energy of hcp metals by using the modified embedded atom method. Appl. Surf. Sci. 2006, 253, 2018−2024. (52) Sun, C.; Xue, D. Tailoring Anisotropic Morphology at the Nanoregime: Surface Bonding Motif Determines the Morphology Transformation of ZnO Nanostructures. J. Phys. Chem. C 2013, 117, 5505−5511. (53) Chen, K.; Sun, C.; Song, S.; Xue, D. Polymorphic crystallization of Cu2O compound. CrystEngComm 2014, 16, 5257−5267. (54) Sun, C.; Xu, D.; Xue, D. Direct in situ ATR-IR spectroscopy of structural dynamics of NH4H2PO4 in aqueous solution. CrystEngComm 2013, 15, 7783−7791. (55) Burrows, N. D.; Hale, C. R. H.; Penn, R. L. Effect of pH on the Kinetics of Crystal Growth by Oriented Aggregation. Cryst. Growth Des. 2013, 13, 3396−3403. (56) Li, D.; Nielsen, M. H.; Lee, J. R. I.; Frandsen, C.; Banfield, J. F.; De Yoreo, J. J. Direction-Specific Interactions Control Crystal Growth by Oriented Attachment. Science 2012, 336, 1014−1018. (57) Wang, X. W.; Fei, G. T.; Tong, P.; Xu, X. J.; Zhang, L. D. Structural control and magnetic properties of electrodeposited Co nanowires. J. Cryst. Growth 2007, 300, 421−425. (58) Darques, M.; Piraux, L.; Encinas, A.; Bayle-Guillemaud, P.; Popa, A.; Ebels, U. Electrochemical control and selection of the structural and magnetic properties of cobalt nanowires. Appl. Phys. Lett. 2005, 86, 072508. (59) Riveros, G.; Green, S.; Cortes, A.; Gomez, H.; Marotti, R. E.; Dalchiele, E. A. Silver nanowire arrays electrochemically grown into nanoporous anodic alumina templates. Nanotechnology 2006, 17, 561− 570. (60) Gao, T.; Meng, G. W.; Zhang, J.; Wang, Y. W.; Liang, C. H.; Fan, J. C.; Zhang, L. D. Template synthesis of single-crystal Cu nanowire arrays by electrodeposition. Appl. Phys. A: Mater. Sci. Process. 2001, 73, 251−254. (61) Gao, T.; Meng, G.; Wang, Y.; Sun, S.; Zhang, L. Electrochemical synthesis of copper nanowires. J. Phys.: Condens. Matter 2002, 14, 355. (62) Thongmee, S.; Pang, H. L.; Ding, J.; Lin, J. Y. Fabrication and magnetic properties of metallic nanowires via AAO templates. J. Magn. Magn. Mater. 2009, 321, 2712−2716. (63) Wen, Y. N.; Zhang, J. M. Surface energy calculation of the fcc metals by using the MAEAM. Solid State Commun. 2007, 144, 163− 167.

I


Growth Orientation Control of Co Nanowires Fabricated by

Recommend Documents