Nickel alloying effect on formation of cobalt nanoparticles and nanowires via electroless deposition under a magnetic field

Article abstract of DOI:10.1149/2.062202jes

Co-Ni nanowires 40-100 nm in diameter with several dozen m of length were prepared via electroless deposition under a magnetic field. The formation of Co-Ni nanowires was investigated by an in-situ mixed potential measurement and a cyclic voltammetry combined with a quartz crystal microbalance electrode. These electrochemical measurements revealed that the deposition rate of nickel on nickel is lower than that of cobalt on cobalt. Nickel works as an inhibitor in a growth process of Co-Ni alloys. Thus, the thickness of Co-Ni alloy nanowires is drastically reduced by the addition of Ni(II) in solution. The morphology of Co-Ni nanowires is also strongly affected by the magnetization of Co-Ni alloy nanoparticles which are precursors in the formation of nanowires. Both the magnetization and the deposition rate in Co-Ni alloys are the key parameters to control the aspect ratio and surface morphology of nanowires.

Full text of DOI:10.1149/2.062202jes

Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
E37
0013-4651/2012/159(2)/E37/8/$28.00 © The Electrochemical Society
Nickel Alloying Effect on Formation of Cobalt Nanoparticles and
Nanowires via Electroless Deposition under a Magnetic Field
Makoto Kawamori,a,∗,z Shunsuke Yagi,b,∗∗ and Eiichiro Matsubaraa
aDepartment of Materials Science and Engineering, Kyoto University, Kyoto 606-8501, Japan
bNanoscience and Nanotechnology Research Center, Osaka Prefecture University, Osaka 599-8570, Japan
Co-Ni nanowires 40–100 nm in diameter with several dozen μm of length were prepared via electroless deposition under a magnetic
field. The formation of Co-Ni nanowires was investigated by an in-situ mixed potential measurement and a cyclic voltammetry
combined with a quartz crystal microbalance electrode. These electrochemical measurements revealed that the deposition rate of
nickel on nickel is lower than that of cobalt on cobalt. Nickel works as an inhibitor in a growth process of Co-Ni alloys. Thus, the
thickness of Co-Ni alloy nanowires is drastically reduced by the addition of Ni(II) in solution. The morphology of Co-Ni nanowires
is also strongly affected by the magnetization of Co-Ni alloy nanoparticles which are precursors in the formation of nanowires. Both
the magnetization and the deposition rate in Co-Ni alloys are the key parameters to control the aspect ratio and surface morphology
of nanowires.
© 2011 The Electrochemical Society. [DOI: 10.1149/2.062202jes] All rights reserved.
Manuscript submitted August 29, 2011; revised manuscript received October 26, 2011. Published December 20, 2011.
Nanowires of iron group metals (Fe, Co, Ni) are attractive for ap-
plications to catalytic materials as well as magnetic materials with
the magnetic shape anisotropy.1 Numerous methods for synthesis of
iron group nanowires have been reported using the template-assisted
electrodeposition,2–4 organometallic routes,5,6 polyol reduction,7,8
and electroless deposition under an external magnetic field.9,10 Es-
pecially, the electroless deposition has a great advantage because the
nanowires with a wide variety of compositions and sizes can be fab-
ricated in a large-scale with a relatively simple method.
We have extensively studied the fabrication of metallic nanopar-
ticles such as copper,11,12 cobalt,13,14 nickel,15 and Co-Ni alloy16 by
the electroless deposition method and also developed the monitoring
method of the reaction process by adopting a concept of a mixed
potential which is a good indicator of a reduction ability in solution.
Recently, we have succeeded in a fabrication of nickel nanowires by
applying this method of nanoparticles and also proposed the formation
mechanism of nanowires.17 In the synthesis of nanowires under a mag-
netic field, the magnetic properties of nanoparticles definitely affect
the morphology of nanowires because the nanoparticles are firstly
formed prior to the formation of nanowires.17 Thus, in the present
work, the effect of a magnetic field on the formation of nanowires
was investigated in the Co-Ni system. First, the electrochemical be-
havior in the formation of Co-Ni nanoparticles was carefully studied
using an in-situ mixed potential measurement, an in-situ quartz crystal
microbalance (QCM) measurement, and a cyclic voltammetry with a
QCM electrode. Then, the formation of Co-Ni nanowires was dis-
cussed.
solutions was kept at 353 K with nitrogen gas bubbling to remove
the dissolved oxygen. The metallic salt solution and the hydrazine
solution were mixed at 353 K to start the reaction. The reaction so-
lution was agitated at a rate of 500 rpm with a magnetic stirring unit
at 353 K during the reaction for the syntheses of Co-Ni particles. For
the syntheses of Co-Ni wires, the reaction solution was kept at 353 K
in a water bath located inside the two parallel neodymium magnets
(100 × 100 mm) separated 100 mm apart. The magnetic field inside
the magnets was about 1.0–2.0 kOe, which was measured by a Tesla
meter (KANETEC Co., Ltd., TM-601). After the reaction, particles
and wires were washed several times with ethanol.
The morphology of precipitates was observed using a field-
emission-scanning electron microscope (JEOL Ltd., JSM-6500F).
The mean diameter and size distribution of the Co-Ni particles were
determined by image analysis for randomly selected 600 particles in
each sample. The compositions of Co-Ni particles and wires were an-
alyzed with energy-dispersive X-ray (EDX) spectrometry. The crys-
talline structure of precipitates was investigated by X-ray diffrac-
tion (XRD, Rigaku Co., Ltd., RINT-2200) using Cr Kα radiation.
A superconducting quantum interference device (Quantum Design,
Inc., MPMS SQUID XL) was used to measure the magnetic hystere-
sis curves of the particles at room temperature. During the synthe-
ses of Co-Ni particles, the gold-sputtered QCM electrode (SEIKO
EG&G QA-A9M-AU) was immersed in the reaction solution and the
mixed potential of the QCM electrode was measured by a potentio-
stat/galvanostat (Hokuto Denko Co., Ltd., HA-151) using a Ag/AgCl
(3.33 M KCl) electrode (Horiba 2565A-10T) as a reference electrode.
The weight of deposits on the QCM electrode, ꢀm, was calculated
from the change in resonance frequency of the QCM electrode, ꢀf,
Experimental
by Sauerbrey’s equation;18
The reaction solutions were prepared using cobalt chloride hexahy-
drate (CoCl2 · 6H2O) and nickel chloride hexahydrate (NiCl2 · 6H2O)
as a source of cobalt and nickel ions, ethylene glycol (EG) as a sol-
vent, and hydrazine monohydrate (N2H4 · H2O) as a reducing agent.
Sodium hydroxide (NaOH) was used as a source of OH− ions. Chloro-
platinic acid hexahydrate (H2PtCl6 · 6H2O) was used as a nucleating
agent. These reagents are all reagent-grade (Nacalai Tesque, Inc.) and
used without further purification.
√
2 f02
A
ρq μq
ꢀm = −
ꢀf
[1]
where f0 is the frequency of the QCM electrode before the deposition,
A is the active area of the QCM electrode (0.196 cm2), ρq is the
density of quartz (2.648 g cm−3), and μq is the shear modulus of
quartz (2.947 × 1011 g cm−1
s
−2). The cyclic voltammograms were
also measured by the potentiostat/galvanostat using a gold or nickel-
sputtered QCM electrode (SEIKO EG&G QA-A9M-NI) as a working
electrode, a platinum electrode (20 × 20 mm) as a counter electrode,
and a Ag/AgCl electrode as a reference electrode.
First, 27 cm3 EG solution containing 0.10 M metallic salts ([CoCl2]
+ [NiCl2] = 0.10 M) and 0.3 M NaOH was prepared, where M is mol
dm−3. For the experiments with a nucleating agent, 0.20 mM H2PtCl6
was added. The same amount of EG solution (27 cm3) containing 1.00
M N2H4 and 0.1 M NaOH was also prepared. The temperature of the
Results
Figure 1 shows scanning electron microscopy (SEM) images and
size distributions of particles synthesized in the different concentration
ratios of Co(II) to Ni(II) in the reaction solutions. The mean diameter
∗
Electrochemical Society Student Member.
Electrochemical Society Active Member.
∗∗
z E-mail: kawamori@makoto.mbox.media.kyoto-u.ac.jp
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E38
Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
Figure 1. SEM images and size distribution of the particles synthesized from the solutions containing (a) 0.050 M Co(II), (b) 0.035 M Co(II) - 0.015 M Ni(II),
(c) 0.025 M Co(II) - 0.025 M Ni(II), (d) 0.015 M Co(II) - 0.035 M Ni(II), and (e) 0.050 M Ni(II) in the absence H2PtCl6, and (f) 0.050 M Co(II), (g) 0.035 M
Co(II) - 0.015 M Ni(II), (h) 0.025 M Co(II) - 0.025 M Ni(II), (i) 0.015 M Co(II) - 0.035 M Ni(II), and (j) 0.050 M Ni(II) in the presence of 0.10 mM H2PtCl6.
Mean diameter dm. is also indicated.
of the particles is also indicated. In the absence of H2PtCl6, the mean
particle diameters are 1300 nm and 330 nm at [Co(II)] : [Ni(II)]
= 100 : 0 and 0 : 100, respectively (Fig. 1a and 1e). The particle size
decreases and the size distribution becomes sharper with an increase
in a concentration of Ni(II) (Fig. 1b–1d). By the addition of H2PtCl6,
the particle size largely decreased in a whole ratio of Co(II) to Ni(II)
(Fig. 1f–1j). The nucleating agent such as H2PtCl6, whose redox po-
tential is much higher than that of the objective metal, provides many
heterogeneous nucleation sites, resulting in the formation of abundant
and fine particles.19,20 The provided platinum nuclei divides the nu-
cleation and growth stages during the formation of Co-Ni particles,
which gives us a sharper size distribution. On the contrary, a reduction
of the Co-Ni particle size by the addition of Ni(II) is observed with
and without H2PtCl6. The standard electrode potential of Ni(II)/Ni
redox pair (ENi(II)/Ni = −0.250 V vs. SHE) is close to that of Co(II)/Co
pair (ECo(II)/Co = −0.277 V vs. SHE).21 Thus, the formation of smaller
particles by the addition of Ni(II) is not explained like the case of the
nucleating agent of H2PtCl6.
The atomic ratios of cobalt to nickel in the particles are almost
equal to the ratios of Co(II) to Ni(II) ions in solution as it is seen
in Table I. Figure 2 shows the atomic ratios of cobalt and nickel at
the points indicated in the SEM images of Co-Ni alloy particles. A
small composition gradient is observed in the particles at all concen-
tration ratios. The ratio of cobalt is slightly higher at the core of the
particles than at the edge. In contrast, that of nickel is lower at the
core of the particles than at the edge. Consequently, although atomic
concentrations of the particles are equal to the nominal concentra-
tion in a macro-scopic scale, the concentration fluctuations exist in a
nano-meter scale.
peaks due to fcc structure becomes dominant. By the addition of
H2PtCl6, the peak width slightly becomes broader in each concentra-
tion ratio (Fig. 3f–3j), which is due to a decrease in the crystallite size.
In pure cobalt, a hcp structure is stable at the room temperature and
transforms to the fcc phase at 695 K according to the Co-Ni phase
diagram.22 This allotropic transformation temperature decreases by
alloying with nickel. At present reaction temperature of 353 K, the
hcp and fcc phases are stable below about 25 atom%Ni and above
35 atom%Ni, respectively. Both phases coexist between these nickel
concentrations. Thus, both the fcc and hcp phases are the equilibrium
phase at Co : Ni = 70 : 30 (atom%) and the fcc phase at Co : Ni
= 50 : 50 and 30 : 70 (atom%) at 353 K. Small diffuse peaks due
to hcp cobalt disappear by annealing the as-prepared Co-Ni alloy
particles at 623 K in Ar atmosphere for 16 hours. Figure 4 shows
the SEM images of the annealed particles. Neither sintering nor
Table I. Atomic ratio of metal in the particles synthesized by
changing the concentration ratio of Co(II) to Ni(II) in the
reaction solution.
Concentration ratio
of metallic species in
the reaction solution
Atomic ratio of
metal (atom%)
With or without
H2PtCl6
[Co(II)] : [Ni(II)]
Co
Ni
(a) 100 : 0
(b) 70 : 30
(c) 50 : 50
(d) 30 : 70
(e) 0 : 100
(f) 100 : 0
(g) 70 : 30
(h) 50 : 50
(i) 30 : 70
(j) 0 : 100
−
−
−
−
−
x
x
x
x
x
100.0
70.3
50.4
29.9
0.0
100.0
69.8
49.4
30.3
0.0
0.0
29.7
49.6
70.1
100.0
0.0
30.2
50.6
69.7
100.0
Figure 3 shows XRD patterns of the particles. The hexagonal close-
packed (hcp) peaks of cobalt and the face-centered cubic (fcc) peaks
of nickel are observed at [Co(II)] : [Ni(II)] = 100 : 0 and 0 : 100,
respectively. In the XRD patterns of Co-Ni alloy particles, both fcc
and hcp peaks are observed. The intensity of (100) hcp cobalt peak
around 63.8◦ decreases with increasing a concentration of Ni(II). Ad-
ditionally, a symmetry of the largest peak around 68.5◦ improves at
a higher concentration of Ni(II), indicating that the contribution of
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
E39
Figure 2. SEM images and atomic ratios of cobalt and nickel of particles synthesized from the solutions containing (a) 0.035 M Co(II) - 0.015 M Ni(II), (b) 0.025
M Co(II) - 0.025 M Ni(II), and (c) 0.015 M Co(II) - 0.035 M Ni(II).
agglomeration among the particles is observed. Additionally, the sur-
face of the annealed particles is relatively smoother than that of the
as-prepared particles. The XRD patterns of annealed particles and the
lattice parameters calculated by the position of the fcc 200 peak at
around 80◦ are shown in Figure 5. The peaks in the XRD patterns
of annealed particles become sharper because of the grain growth by
annealing. The 100 hcp peak at 63.8◦ observed in the as-prepared
particles at [Co(II)] : [Ni(II)] = 50 : 50 and 30 : 70 (Fig. 3c and 3d)
completely disappeared after annealing. This is consistent with the
phase expected in the phase diagram. As shown in Fig. 5B, the lattice
parameters change approximately with the linear function of the com-
position of Ni(II) between the two lattice parameters of metastable
fcc cobalt and fcc nickel, obeying the Vegard’s law.22,23 This indicates
that Co-Ni solid solutions were formed in the particles after annealing.
Figure 6 shows the weight change of metals deposited on the QCM
substrate during the nanoparticles formation with different concentra-
tion ratios of Co(II) to Ni(II). At [Co(II)] : [Ni(II)] = 100 : 0 in
the absence of H2PtCl6 (Fig. 6a), the cobalt deposition is observed
at about 4 minutes after mixing the solutions containing Co(II) and
hydrazine. The weight of cobalt reaches 105 μg in 17 minutes, which
Figure 3. XRD patterns of the particles synthesized
from the solutions containing (a) 0.050 M Co(II),
(b) 0.035 M Co(II) - 0.015 M Ni(II), (c) 0.025 M
Co(II) - 0.025 M Ni(II), (d) 0.015 M Co(II) - 0.035 M
Ni(II), (e) 0.050 M Ni(II) in the absence of H2PtCl6,
and (f) 0.050 M Co(II), (g) 0.035 M Co(II) - 0.015 M
Ni(II), (h) 0.025 M Co(II) - 0.025 M Ni(II), (i) 0.015
M Co(II) - 0.035 M Ni(II), (j) 0.050 M Ni(II) in the
presence of 0.10 mM H2PtCl6.
Figure 4. SEM images of the particles of (a) to
(e) in Figure 1 annealed at 623 K for 16 hours.
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E40
Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
Figure 5. (A) XRD patterns of the particles of (a) to (e) in Figure 1 annealed at 623 K for 16 hours and (B) change in lattice parameters of annealed particles with
composition of Ni(II).
is the maximum value observable in the present system. The slope
of the curve, which is the deposition rate, becomes smaller and the
final weight of metal deposited on a QCM substrate decreases with an
increase in a concentration of Ni(II). Also, the amounts of cobalt in
the presence of H2PtCl6 (Fig. 6f) is about 4 μg which is much smaller
than that in the absence of H2PtCl6 (Fig. 6a). Considering that the
same amount of Co(II) dissolves in the reaction solution in the both
cases and the formation of platinum particles provides nucleation sites
of cobalt which suppresses the heterogeneous nucleation on the QCM
substrate, we conclude that the decrease in the amounts deposited on
a QCM substrate by the addition of Ni(II) is due to the enhancement
of the homogeneous nucleation in the solution.
(Fig. 7a), the surface of Co-Ni wires becomes rougher at a higher
concentration of Ni(II) (Fig. 7b–7d) and beads-like wires were ob-
tained at [Co(II)]: [Ni(II)] = 0:100 (Fig. 7e). The diameter of wires
decreases with an increase in a concentration of Ni(II). Cobalt and Co-
Ni nanowires whose diameter is smaller than 100 nm were obtained
by the addition of 0.10 mM H2PtCl6. Beads-like nickel nanowires of
40 nm in diameter were formed from the solution containing 0.050 M
Ni(II). It is likely that the formation of wire becomes more difficult at
a higher Ni(II) concentration. It is also noted that the diameter of the
nanoparticles without a magnetic field is close to that of the nanowires
with a magnetic field at the same concentration ratio of Co(II) to Ni(II)
(see Fig. 1 and Fig. 7, respectively).
Table II shows the atomic ratios of metals in the nanowires. The
alloy compositions are close to the ratios of Co(II) to Ni(II) ions in
solution as it is seen in the particles of Table I. Figure 8 shows XRD
patterns of the wires. The peak positions and peak widths are almost
the same as those of the particles in Fig. 3. Thus, the basic formation
processes are similar in both nanoparticles and nanowires.
Figure 7 shows SEM images of wires synthesized from the solu-
tions under a magnetic field at the different concentration ratios of
Co(II) to Ni(II). While the surface of cobalt wires is relatively smooth
Table II. Atomic ratio of metal in the wires synthesized by
changing the concentration ratio of Co(II) to Ni(II) in the
reaction solution.
Concentration ratio
of metallic species in
the reaction solution
Atomic ratio of
metal (atom%)
With or without
H2PtCl6
[Co(II)] : [Ni(II)]
Co
Ni
(a) 100 : 0
(b) 70 : 30
(c) 50 : 50
(d) 30 : 70
(e) 0 : 100
(f) 100 : 0
(g) 70 : 30
(h) 50 : 50
(i) 30 : 70
(j) 0 : 100
−
−
−
−
−
x
x
x
x
x
100.0
70.5
50.3
30.4
0.0
100.0
71.2
51.4
29.1
0.0
0.0
29.5
49.7
69.6
100.0
0.0
28.8
48.6
70.9
100.0
Figure 6. Weight of metal deposited on the gold-sputtered QCM substrate
from the solutions containing (a) 0.050 M Co(II), (b) 0.035 M Co(II) - 0.015
M Ni(II), (c) 0.025 M Co(II) - 0.025 M Ni(II), (d) 0.015 M Co(II) - 0.035 M
Ni(II), (e) 0.050 M Ni(II), and (f) 0.050 M Co(II). H2PtCl6 is added only in
the solution (f).
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
E41
Figure 7. SEM images of the wires synthesized from the solutions containing (a) 0.050 M Co(II), (b) 0.035 M Co(II) - 0.015 M Ni(II), (c) 0.025 M Co(II) - 0.025
M Ni(II), (d) 0.015 M Co(II) - 0.035 M Ni(II), and (e) 0.050 M Ni(II) in the absence H2PtCl6, and (f) 0.050 M Co(II), (g) 0.035 M Co(II) - 0.015 M Ni(II), (h)
0.025 M Co(II) - 0.025 M Ni(II), (i) 0.015 M Co(II) - 0.035 M Ni(II), and (j) 0.050 M Ni(II) in the presence of 0.10 mM H2PtCl6.
Discussion
substrates as a working electrode. The anodic current due to the
oxidation of hydrazine is observed above −0.94 and −0.90 V vs.
Ag/AgCl on the cobalt and nickel substrates, respectively. This in-
dicates that excess potential for the hydrazine oxidation is higher
on the nickel substrate than on the cobalt substrate. The exchange
current density of the hydrazine oxidation reaction is much smaller
on the nickel substrate than on the cobalt substrate. Namely, nickel
has a lower catalytic activity than cobalt in the oxidation reaction of
hydrazine.
Formation of Co-Ni nanoparticles via electroless deposition.— In
the formation process of nanoparticles by electroless deposition, the
particle growth proceeds in the simultaneous oxidation reactions of
the reducing agent and reduction reactions of metallic ions. Therefore,
the polarization curves enable us to evaluate the particle growth rate.
In the present process, the following hydrazine oxidation reaction
mainly occurs as an anodic reaction:
N2H4 + 4OH− → N2 + 4H2O + 4e
Figure 9 shows anodic polarization curves measured in EG con-
taining 0.2 M NaOH and 0.50 M N2H4 using cobalt and nickel
[2]
Both a weight of deposited metal and a total current density are
simultaneously measured by voltammetry combined with QCM.
The following reduction reaction of Co(II) and Ni(II) species was
Figure 8. XRD patterns of the wires synthesized from
the solutions containing (a) 0.050 M Co(II), (b) 0.035
M Co(II) - 0.015 M Ni(II), (c) 0.025 M Co(II) - 0.025
M Ni(II), (d) 0.015 M Co(II) - 0.035 M Ni(II), and
(e) 0.050 M Ni(II) in the absence H2PtCl6, and (f) 0.050
M Co(II), (g) 0.035 M Co(II) - 0.015 M Ni(II), (h) 0.025
M Co(II) - 0.025 M Ni(II), (i) 0.015 M Co(II) - 0.035 M
Ni(II), and (j) 0.050 M Ni(II) in the presence of 0.10 mM
H2PtCl6.
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E42
Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
−0.77 V during the anodic sweep. The dissolution of cobalt was
seen above −0.77 V. Thus, the redox potential of Co(II)/Co can be
determined about −0.77 V. Similarly, the redox potential of Ni(II)/Ni
is about −0.70 V, which is slightly more positive value than that of
Co(II)/Co. The reduction current density of Ni(II) species is much
smaller than that of Co(II) species.
Consequently, both the anodic current density of hydrazine oxida-
tion and the cathodic current density of metal deposition are smaller
on nickel than on cobalt. Thus, it would appear that the growth rate
of nickel particles is slower than that of cobalt particles. Once nickel
deposits on the surface of particles in the formation process of Co-Ni
alloy particles, both the oxidation reaction of hydrazine and the de-
position of metal are suppressed. Therefore, nickel has an inhibiting
effect on the Co-Ni particle growth and the increase in the concentra-
tion of Ni(II) results in the decrease of growth rate of Co-Ni particles.
A small composition gradient in the Co-Ni alloy particles as shown
in Fig. 2 can be explained by these different electrode behaviors
between cobalt and nickel in the reduction reactions of metallic ions
and the oxidation reactions of hydrazine. The atomic ratio of cobalt is
higher at the core of Co-Ni alloy particles than at the edge because the
deposition of cobalt is preferentially occurred in the growth process.
According to the Co-Ni alloy phase diagrams, the hcp phase is stable
at a higher concentration of cobalt. It is likely that cobalt-rich core of
particles mainly has the hcp phase and the edge of particles has the fcc
phase. In the result, the mixture of hcp and fcc phases are observed in
the XRD patterns of the as-prepared Co-Ni alloy particles and wires
in Fig. 3b–3d and 3g–3i. During the annealing at 623 K for 16 hours, a
composition gradient in the Co-Ni alloy particles is reduced by atomic
diffusion, which results in the disappearance of the hcp phase and the
grain growth of the fcc phase in the particles containing more than
25 atom% Ni.
Figure 11 shows the time-dependence of mixed potential in the
solution containing 0.50 N2H4, 0.2 M NaOH, and (a) 0.050 M CoCl2
or (b) 0.050 M NiCl2. The redox potentials of Co(II)/Co and Ni(II)/Ni
are also indicated. At 0.050 M CoCl2 (Fig. 11a), the mixed potential
gradually dropped below the redox potential of Co(II)/Co after about
4 minutes of mixing the metallic salt solution and the hydrazine so-
lution. This indicates that the cobalt deposition is thermodynamically
possible at 4 minutes after the start of the reaction. Actually, the cobalt
deposition is observed after 4 minutes of reaction by the in-situ QCM
measurement as shown in Fig. 6a. At 0.050 M NiCl2 (Fig. 11b), the
mixed potential quickly dropped to around −1.04 V vs. Ag/AgCl just
after mixing the reaction solutions.
Figure 9. Anodic polarization curves measured on cobalt and nickel substrates
in EG containing 0.2 M NaOH and 0.50 M N2H4. Sweep rate is 1 mV s−1
.
separately evaluated from the weight change of metal deposited on
the QCM substrate.
Co(II) + 2e → Co
[3]
Ni(II) + 2e → Ni
[4]
Figure 10 shows cyclic voltammograms measured in 120 cm3 EG
containing 0.050 M CoCl2 and 0.2 M NaOH (Fig. 10a), and 0.050
M NiCl2 and 0.2 M NaOH (Fig. 10b) using the gold-sputtered and
nickel-sputtered QCM substrates as a working electrode, respectively.
As shown in Fig. 10a (dotted line), the cathodic current due to
the decomposition of the solvent is observed below about −0.90 V
vs. Ag/AgCl during the cathodic sweep. The contribution of this
reductive decomposition of the solvent to the total current density is
dominant in this process, and the current density due to the cobalt
deposition is hardly observed by a normal voltammetry. In a cyclic
voltammogram evaluated from the weight change observed by QCM
(Fig. 10a, solid line), the cathodic current due to the cobalt deposition
is observed below about −0.80 V during the cathodic sweep. After
the gold-sputtered QCM substrate was completely covered with
deposited cobalt, the deposition current of cobalt is observed below
The difference between the redox potential and the mixed potential
corresponds to the driving force of the metal deposition. Thus, the
driving force of nickel metal deposition is higher than that of cobalt and
the homogeneous nucleation is more favorable during the formation of
Figure 10. Cyclic voltammograms measured by potentiostat (dotted line) and QCM (solid line) at 1 mV s−1 in EG containing (a) 0.050 M CoCl2, 0.2 M NaOH,
and (b) 0.050 M NiCl2, 0.2 M NaOH.
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
E43
Table III. Magnetic properties of particles synthesized by
changing the concentration ratio of Co(II) and Ni(II) in the
reaction solution.
Concentration ratio of metallic
species in the reaction solution
[Co(II)] : [Ni(II)]
Saturation
magnetization, Ms
Coercivity, Hc
(emu g−1
)
(Oe)
(a) 100 : 0
(b) 70 : 30
(c) 50 : 50
(d) 30 : 70
(e) 0 : 100
145.5
124.1
110.3
80.1
186
92
41
61
102
47.5
to this mechanism, the magnetic interaction and the deposition rate of
particles strongly affect the formation of wires. Namely, the formation
of wires becomes easier at a stronger magnetic interaction and a slower
deposition rate of particles.
Figure 12 shows the magnetization-magnetic field (M-H) curves
for the particles synthesized by changing the concentration ratios of
Co(II) to Ni(II) in solution without H2PtCl6. The hysteresis is ob-
served in each curve and it would appear that the particles behave
as magnetic multidomain particles.24 The magnetic properties of the
particles are summarized in Table III. The saturation magnetization
of cobalt and nickel particles is 145.5 and 47.5 emu g−1 which is
lower than that of the bulk metal (161 and 54 emu g−1 at 293 K,
respectively).25 The saturation magnetization of Co-Ni alloy particles
decreases with increasing a concentration ratio of Ni(II). The decrease
of the observed saturation magnetization of cobalt and nickel parti-
cles with respect to the bulk values is possibly due to the impurities
such as carbon, oxygen, and hydrogen.24,26 Actually, the mass losses
about 2% were observed in cobalt particles (Fig. 1a) after the ther-
mal treatment at 623 K in H2 atmosphere for 16 hours, indicating that
the particles contain the organic species, organometallic phases or
unreacted hydroxides which are thermally unstable.
At 1.0–2.0 kOe, which corresponds to the magnetic field in the syn-
theses of wires, the magnetization of nickel particles is much lower
than that of cobalt and Co-Ni alloy particles as shown in the inset
of Fig. 12. In the formation of cobalt wires, it takes less time for
the particles to align along a magnetic field due to the stronger mag-
netic interaction between cobalt particles. Thus, the particles align
prior to the end of the reduction reaction and the relatively-smooth
wires were formed as shown in Fig. 7a and 7f despite the higher
growth rate of particles. On the contrary, the weak magnetic inter-
action between nickel particles results in the formation of beads-like
wires as shown in Fig. 7 while the growth rate of nickel particles
is lower than that of cobalt. In order to fabricate the nickel wires
with a smooth surface, it is necessary to decrease the deposition rate
of nickel particles, which is achieved17 by adjusting a concentration
of NaOH and adding a complexing agent such as trisodium citrate
dihydrate (Na3C6H5O7 · 2H2O). Consequently, by carefully control-
ling both the deposition rate and the magnetization of the particles,
one can successfully prepare nanowires with various diameters and
surface morphology.
Figure 11. Time dependence of mixed potential in the solution containing
0.50 N2H4, 0.2 M NaOH, and (a) 0.050 M Co(II) or (b) 0.050 M Ni(II).
nickel particles than that of cobalt particles. Furthermore, the slower
particle growth rate of nickel divides the nucleation and growth stage
during the formation, which results in smaller particles with a sharper
size distribution as shown in Figs. 1e and 1j. An inhibiting effect on the
Co-Ni particle growth by the addition of Ni(II) becomes stronger with
increasing a concentration of Ni(II) and the nucleation is prompted in
comparison. This is the formation mechanism of smaller particles by
the addition of Ni(II).
Formation of Co-Ni nanowires under a magnetic field.— In the
previous report,17 we discussed the following formation mechanism
of nanowires via electroless deposition under a magnetic field. First,
particles are formed in the reaction solution, and are aligned along a
magnetic field due to the magnetic interaction among the particles.
Then, the necks between particles are preferentially covered with
deposited metal to decrease the interfacial energy. Finally, the particles
are tightly connected and form rigid wires. Therefore, the necessary
condition for the formation of wires is that the alignment of particles
occurs prior to the terminal point of the deposition reaction. According
Conclusions
In the present work, the formation of Co-Ni particles and wires via
electroless deposition was electrochemically investigated. The forma-
tion of smaller particles and thinner wires by the addition of Ni(II) is
explained by different electrode behaviors between cobalt and nickel
in the oxidation reaction of hydrazine and the reduction reactions of
metallic ions. The catalytic activity of both reactions is much lower
on nickel than that on cobalt, indicating that the growth rate of nickel
particles is lower than cobalt. Deposited nickel suppresses the growth
of Co-Ni particles. A small composition gradient observed in Co-Ni
alloy particles is also explained by the different deposition rate of
Figure 12. The magnetization-magnetic field (M-H) curves obtained at
room temperature for the particles synthesized from the solutions containing
(a) 0.050 M Co(II), (b) 0.035 M Co(II) - 0.015 M Ni(II), (c) 0.025 M Co(II) -
0.025 M Ni(II), (d) 0.015 M Co(II) - 0.035 M Ni(II), (e) 0.050 M Ni(II) in the
absence of H2PtCl6.
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E44
Journal of The Electrochemical Society, 159 (2) E37-E44 (2012)
cobalt and nickel. The magnetization of particles as well as the de-
position rate strongly affects the morphology of wires formed under
a magnetic field. For example, a lower deposition rate is required in
fabrication of smooth wires whose magnetization is relatively weak,
such as nickel.
7. D. Ung, Y. Soumare, N. Chakroune, G. Viau, M.-J. Vaulay, V. Richard, and F. Fie´vet,
Chem. Mater., 19, 2084 (2007).
8. Y. Soumare, A. Dakhlaoui-Omrani, F. Schoenstein, S. Mercone, G. Viau, and N.
Jouini, Solid State Communications, 151, 284 (2011).
9. V. V. Sviridov, G. P. Shevchenko, A. S. Susha, and N. A. Diab, J. Phys. Chem., 100,
19632 (1996).
10. M. D. L. Balela, S. Yagi, and E. Matsubara, J. Electrochem. Soc., 158, D210
(2011).
11. S. Yagi, H. Nakanishi, E. Matsubara, S. Matsubara, T. Ichitsubo, K. Hosoya, and Y.
Matsuba, J. Electrochem. Soc., 155, D474 (2008).
12. S. Yagi, H. Nakanishi, T. Ichitsubo, and E. Matsubara, J. Electrochem. Soc., 156,
D321 (2009).
13. S. Yagi, M. Kawamori, and E. Matsubara, Electrochem. Solid-State Lett., 13, E1
(2010).
14. M. D. L. Balela, S. Yagi, and E. Matsubara, Electrochem. Solid-State Lett., 13, D4
(2010).
Acknowledgments
This research was supported by a grant-in-Aid for JSPS (Japan So-
ciety for the Promotion of Science) fellows, grant-in-Aid for Knowl-
edge Cluster Initiative (Kyoto Nanotechnology Cluster), grant-in-Aid
for the Global COE Program (International Center for Integrated Re-
search and Advanced Education in Materials Science), and grant-in-
Aid for Challenging Exploratory Research (23656477), all from the
Japan Society for the Promotion of Science, Ministry of Education,
Culture, Sports, Science and Technology of Japan.
15. S. Yagi, T. Koyanagi, H. Nakanishi, T. Ichitsubo, and E. Matsubara, J. Electrochem.
Soc., 155, D583 (2008).
16. S. Yagi, M. Kawamori, and E. Matsubara, J. Electrochem. Soc., 157, E92 (2010).
17. M. Kawamori, S. Yagi, and E. Matsubara, J. Electrochem. Soc., 158, E79 (2011).
18. A. Janshoff, H. J. Galla, and C. Steinem, Angew. Chem. Int. Ed., 39, 4004 (2000).
19. G. Viau, P. Toneguzzo, A. Pierrard, O. Acher, F. Fie´vet-Vincent, and F. Fie´vet, Scr.
Mater., 44, 2263 (2001).
References
20. M. D. L. Balela, S. Yagi, Z. Lockman, A. Aziz, A. Jr. Amorsolo, and E. Matsubara,
1. Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F. Kim, and H. Yan, Adv.
Mater., 15, 353 (2003).
2. X. Y. Zhang, L. D. Zhang, W. Chen, G. W. Meng, M. J. Zheng, L. X. Zhao, and
J. Electrochem. Soc., 156, E139 (2009).
21. M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, p. 322,
Cebelcor, Bru¨ssel (1966).
F. Phillipp, Chem. Mater., 13, 2511 (2001).
22. T. Nishizawa and K. Ishida, Bull. Alloy Phase Diagram, 4, 390 (1983).
23. P. Villars and L. D. Calvert, Pearson’s Handbook of Crystallographic Data for
Intermetallic Phases, American Society of Metals, Materials Park, Ohio, (1985).
24. C. Luna, M. Morales, C. J. Serna, and M. Vazquez, Nanotechnology, 14, 268
(2003).
25. B. D. Cullity and C. D. Graham, Introduction to Magnetic Materials, Second Edition,
p. 531, The Institute of Electrical and Electronics Enginners, Inc. (2009).
26. G. Viau, F. Fie´vet-Vincent, and F. Fie´vet, J. Mater. Chem., 6, 1047 (1996).
3. K. Nielsch, F. Mu¨ller, A.-P. Li, and U. Go¨sele, Adv. Mater., 12, 582 (2000).
4. M. Tian, J. Wang, J. Kurtz, T. E. Mallouk, and M. H. W. Chan, Nano Lett., 3, 919
(2003).
5. F. Dumestre, B. Chaudret, C. Amiens, M. C. Fromen, M. J. Casanove, P. Renaud,
and P. Zurcher, Angew. Chem., 114, 4462 (2002).
6. Z. Zhang, D. A. Blom, Z. Gai, J. R. Thompson, J. Shen, and S. Dai, J. AM. CHEM.
SOC., 125, 7528 (2003).
Downloaded on 2014-12-10 to IP 169.230.243.252 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).