Formation of a core/shell microstructure in Cu-Ni thin films

Article abstract of DOI:10.1149/1.2946709

Electrodeposition of Cu-Ni thin films can result in phase separation characterized by the formation of nodular features that exhibit a uniform columnar core/shell structure with a copper-rich core and nickel-rich shell. Here, we show that the core/shell microstructure is the result of differences in the nucleation and growth rates of the two components. In the potential range where the core/shell structure is observed, copper deposition is fast, resulting in the formation of a relatively low density of large hemispherical islands. Nickel deposition is characterized by slower kinetics, resulting in the formation of a high density of small islands surrounding the copper islands. These results provide a basis for understanding the formation of this core/shell microstructure in binary alloy systems.

Full text of DOI:10.1149/1.2946709

Formation of a Core/Shell Microstructure in Cu−Ni Thin Films
Zhu Liu, Lian Guo, Chia-Ling Chien and Peter C. Searson
J. Electrochem. Soc. 2008, Volume 155, Issue 9, Pages D569-D574.
doi: 10.1149/1.2946709
Email alerting
service
Receive free email alerts when new articles cite this article - sign up in the box at the
top right corner of the article or click here
To subscribe to Journal of The Electrochemical Society go to:
http://jes.ecsdl.org/subscriptions
©
2008 ECS - The Electrochemical Society
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org

Journal of The Electrochemical Society, 155 ͑9͒ D569-D574 ͑2008͒
D569
0
013-4651/2008/155͑9͒/D569/6/$23.00 © The Electrochemical Society
Formation of a Core/Shell Microstructure in Cu–Ni Thin Films
a,
a,
a,b
a,b,
,z
Zhu Liu, * Lian Guo, * Chia-Ling Chien, and Peter C. Searson **
a
b
Department of Materials Science & Engineering and Department of Physics & Astronomy, Johns
Hopkins University, Baltimore, Maryland 21218, USA
Electrodeposition of Cu–Ni thin films can result in phase separation characterized by the formation of nodular features that exhibit
a uniform columnar core/shell structure with a copper-rich core and nickel-rich shell. Here, we show that the core/shell micro-
structure is the result of differences in the nucleation and growth rates of the two components. In the potential range where the
core/shell structure is observed, copper deposition is fast, resulting in the formation of a relatively low density of large hemi-
spherical islands. Nickel deposition is characterized by slower kinetics, resulting in the formation of a high density of small islands
surrounding the copper islands. These results provide a basis for understanding the formation of this core/shell microstructure in
binary alloy systems.
©
2008 The Electrochemical Society. ͓DOI: 10.1149/1.2946709͔ All rights reserved.
Manuscript submitted March 24, 2008; revised manuscript received May 20, 2008. Published July 10, 2008.
The microstructure of thin films can be influenced by both ther-
modynamic and kinetic parameters associated with the deposition
process. For binary alloy systems this can lead to a rich array of
microstructures. Due to the similarity in lattice parameters ͑aNi
Ni͑H NSO ͒ ·4H O ͑99.99% Aldrich͒, 50 mM CuSO ·5H O ͑98%,
2 3 2 2 4 2
Aldrich͒, and 0.65 M HBO ͑Alfa Aesar͒ at pH 3.8. In all cases the
3
−
2
deposition charge was 3 C cm , corresponding to a film thickness
of about 1 ␮m with a deposition efficiency of Ն0.95. The electro-
chemical etching of copper was performed at +0.5 V for 20 min in
the same solution.
=
0.3524 nm, aCu=0.3615 nm͒ Cu–Ni is predicted to exhibit a
1-5
broad range of solid solubility. 6,H7 owever, calculations show a mis-
cibility gap below about 600 K.
The kinetics of copper nucleation and growth was studied by
Electrodeposition in the Cu–Ni system has focused primarily on
the deposition of multilayer thin films or nanowires from solution
containing both Cu͑II͒ and Ni͑II͒. In this case pure copper is depos-
ited at more positive potentials and a Cu–Ni alloy at more negative
potentials. To minimize the amount of copper in the nickel layer, the
depositing copper from solution containing 50 mM CuSO ·5H O
4
2
and 0.65 M HBO . The nickel deposition kinetics was studied by
3
depositing nickel from solution containing 400 mM
Ni͑H NSO ͒ ·4H O and 0.65 M HBO . In both cases, current–time
2
3 2
2
3
transients were recorded at different potentials and the island size
and density were determined using scanning electron microscopy
͑SEM͒.
8-12
Ni͑II͒/Cu͑II͒ ratio is usually Ն100.
Under these conditions, the
nickel layer is a single-phase Ni–Cu solid solution with the mole
fraction of copper usually less than about 5%. Intermediate Cu–Ni
alloy compositions can be accessed by deposition from solution with
The morphology of Cu–Ni thin films before and after etching the
copper was characterized by SEM ͑JEOL 6700F͒. After etching, the
back side of the thin films was imaged by using conductive tape to
peel the films from the substrate. The near-neighbor pore and island
distributions were obtained from analysis of SEM images using NIH
Image. The microstructure of the films was characterized by X-ray
diffraction ͑Phillip’s X Pert 3040͒ with a Cu K␣ source.
13
lower Ni͑II͒/Cu͑II͒ ratios. While solid solutions have been re-
5
ported at intermediate compositions using pulse plating, phase
1
3
separation has been reported for deposition at constant potential,
indicating that the equilibrium state can be achieved in electrodepo-
sition.
In previous work, we have shown evidence for the miscibility
gap in electrodeposited Cu–Ni thin films that exhibit a remarkable
core/shell columnar microstructure characterized by nodular features
Results and Discussion
14
Core/shell microstructure of Cu–Ni thin films.— In previous
work, we have shown that phase separation in Cu–Ni thin films is
dependent on the deposition potential and the Cu͑II͒/Ni͑II͒ concen-
with a copper-rich core and a nickel-rich shell. By confining depo-
sition in a patterned structure with dimensions corresponding to the
size of the nodular features, the core/shell microstructure can be
exploited for the formation of individual ferromagnetic rings and
14
tration ratio in solution. Figure 1 shows X-ray diffraction ͑XRD͒
1
4
patterns for Cu–Ni thin films deposited at potentials from −0.7 to
−1.1 V. In this potential range the copper deposition rate is diffu-
tubes. The core/shell structure is of particular interest for the syn-
1
5,16
17-20
thesis of ferromagnetic rings
and tubes,
and for applications
21
in bioseparations and magnetoresistance random access memory
2
2
devices.
While thermodynamics provides a driving force for phase sepa-
ration, it does not explain the columnar core/shell structure. Here,
we report on the kinetics of copper and nickel deposition and show
that differences in the nucleation and growth rates are essential for
the formation of the core/shell structure.
Experimental
Electrochemical deposition was performed on Au films sputter
deposited onto Si͑111͒ wafers in a three-electrode cell with a plati-
num mesh ͑99%, Sigma Aldrich͒ counter electrode and a Ag/AgCl
͑
3 M NaCl͒ reference electrode ͑Bioanalytical Systems͒. All poten-
tials are reported vs the Ag/AgCl reference ͑Ueq = 0.200 V vs stan-
dard hydrogen electrode͒. Thin films were deposited in the potential
range from −0.7 to −1.1 V from solution containing 400 mM
Figure 1. XRD patterns for Cu–Ni thin films deposited from solution con-
taining 0.4 M Ni͑N2HSO3͒2, 0.05 M CuSO4, and 0.65 M HBO3. The dotted
lines show the positions of the Cu͑111͒ and Ni͑111͒ peaks.
*
Electrochemical Society Student Member.
*
* Electrochemical Society Active Member.
z
E-mail: searson@jhu.edu
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org

D570
Journal of The Electrochemical Society, 155 ͑9͒ D569-D574 ͑2008͒
Figure 2. Cu–Ni phase diagram ͑adapted from Ref. 7͒. ͑᭿͒: The mole frac-
tion of Cu in Cu–Ni thin films where phase separation is observed.
sion limited and hence effectively independent of potential. How-
ever, nickel deposition is in the kinetic regime and the deposition
rate increases exponentially as the deposition potential becomes
more negative. At −0.7 V, where the nickel deposition rate is low, a
2
3
single broad peak close to the Cu͑111͒ peak ͑2␪ = 43.41°͒ is seen,
1
4
indicating a single-phase Cu Ni
−
alloy with x Ϸ 0.97. At
x
1−x
Figure 4. ͑Color online͒ Distributions of near-neighbor distances for films
deposited at −0.9 V: ͑a͒ nodular features in as-deposited films, ͑b͒ pores on
the top of the film after etching, and ͑c͒ pores on the back side of the film
after etching. ͑d͒ Summary of near-neighbor distances vs deposition poten-
tial: ͑᭺͒ before etching, ͑ᮀ͒ after etching ͑top͒, and ͑᭝͒ after etching ͑back
side͒.
0.8 V, two peaks are observed, indicating phase separation into a
Cu-rich phase and a Ni-rich phase. As the potential becomes more
negative, the nickel deposition rate increases and the intensity of the
Ni-rich peak increases. The mole fraction of copper in the films
where phase separation is observed is in the range 0.3–0.8, consis-
7
,24
tent with the miscibility gap in the Cu–Ni phase diagram,
shown in Fig. 2.
as
Figure 3 shows plan-view SEM images of Cu–Ni thin films be-
fore and after etching the copper-rich phase. Thin films were depos-
ited from 50 mM Cu͑II͒ and 400 mM Ni͑II͒ at −0.8, −0.9, and
8
by the formation of a thin oxide layer. As shown in Fig. 3, after
etching, the thin films have uniform pores. Cross-section images
show that the pores are cylindrical and uniform throughout the film.
At the bottom of the film there is a thin, approximately 50 nm thick,
transition region with a higher density of pores.
−
1.0 V where phase separation is observed. Before etching, the
as-deposited thin films exhibit nodular features with an average size
that decreases slightly as the potential becomes more negative.
The distribution of the copper- and nickel-rich phases cannot be
determined from the SEM images of the as-deposited films; how-
ever, this can be conveniently accomplished by selective etching of
Comparison of the thin films before and after etching the copper-
rich phase reveals that each individual nodular feature is correlated
with each individual pore. This can be demonstrated quantitatively
from the near-neighbor distributions, as shown in Fig. 4 for the film
deposited at −0.9 V. From the distribution of the nodular features
we obtain an average near-neighbor distance of 230 Ϯ 59 nm,
which is very close to the near-neighbor distance ͑242 Ϯ 50 nm͒ of
the cylindrical pores on the top side of etched thin films. On the
back side of the films the near-neighbor distance is 133 Ϯ 29 nm,
somewhat smaller than the average spacing on the top of the films.
This is consistent with the higher density of pores on the bottom side
in the transition region ͑see Fig. 5͒. Because the pores are derived
from the copper-rich phase, it is clear that each individual nodule is
characterized by a columnar structure with a copper-rich core and a
nickel-rich shell. The near-neighbor distances are only weakly de-
pendent on the deposition potential, as shown in Fig. 4d.
8,13,14
the copper-rich phase.
In sulfamate solution, copper can be
selectively etched at positive potentials where nickel is passivated
The SEM images of the Cu–Ni thin films reveal a uniform core/
shell microstructure. The miscibility gap provides a thermodynamic
driving force for phase separation, but does not explain the separa-
tion into nodular features with a copper-rich core and a nickel-rich
shell. We show that the growth kinetics of the individual compo-
nents is key to the evolution of this unique core/shell structure.
Kinetics of copper and nickel deposition.— The influence of
nucleation and growth on the evolution of film microstructure in
alloy deposition is poorly understood. To provide insight into the
evolution of the core–shell microstructure observed here, we ana-
lyze the nucleation and growth of each component individually. As
we show below, the nucleation and growth kinetics provides impor-
tant clues into the formation of the core–shell microstructure in this
system.
Figure 3. Plan-view SEM images of Cu–Ni thin films. Film deposited at
−
0.8 V ͑a͒ as deposited, ͑b͒ after etching, and ͑c͒ the back side of the film
after etching. Film deposited at −0.9 V ͑d͒ as deposited, ͑e͒ after etching,
and ͑f͒ the back side of the film after etching. Film deposited at −1.0 V ͑g͒
as deposited, ͑h͒ after etching, and ͑i͒ the back side of the film after etching.
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org

Journal of The Electrochemical Society, 155 ͑9͒ D569-D574 ͑2008͒
D571
Figure 5. SEM images of the Cu–Ni thin film cross section deposited at
1.0 V after etching the Cu-rich phase.
−
Figure 6a shows a cyclic voltammogram for gold in 400 mM
Ni͑II͒ solution. The equilibrium potential for the Ni͑II͒/Ni couple in
this solution is −0.48 V. The onset for nickel deposition is about
−
0.9 V, corresponding to a nucleation barrier of 0.22 V. In the
potential range from −0.8 to −1.1 V where phase separation is ob-
served in the Cu–Ni films, nickel deposition is kinetically limited, as
evidenced by the exponential increase in current with potential. On
the reverse scan, there is no stripping peak due to the formation of
8,13
the nickel passivation layer.
For copper deposition, the onset potential is close to the equilib-
rium potential ͑Ueq = 0.04 V͒, as shown in Fig. 6b. In the potential
range from −0.8 to −1.1 V where phase separation is observed,
copper deposition is diffusion limited and hence independent of po-
tential. In the positive scan, a large striping peak is observed.
Figure 6c shows cyclic voltammetry ͑CV͒ for gold in a solution
containing 400 mM Ni͑II͒ and 50 mM Cu͑II͒. The onset potential
around 0.04 V is followed by a constant current due to the diffusion-
limited deposition of copper, as can be seen by comparison with Fig.
6
b. At −0.7 V, the onset of nickel deposition results in an exponen-
tial increase in current, as seen by comparison with Fig. 6a. The CV
of Cu–Ni solution reveals similar features of individual Cu͑II͒ and
Ni͑II͒ solution, and hence the independent deposition of Cu and Ni
can be used to understand the formation mechanism of Cu–Ni
codeposition.
Figure 6. Current–voltage curve for gold in solution containing ͑a͒ 0.4 M
Ni͑N2HSO3͒2 and 0.65 M HBO3, ͑b͒ 0.05M CuSO4 and 0.65 M HBO3, and
͑c͒ 0.4 M Ni͑N2HSO3͒2, 0.05M CuSO4, and 0.65 M HBO3. The equilibrium
potential for the Ni͑II͒/Ni couple is −0.48 V and for the Cu͑II͒/Cu couple is
Figure 7a shows current–time transients for copper deposition
from 50 mM Cu͑II͒ solution. Up to about 50 ms, the current asso-
ciated with copper nucleation and growth is very high whereas at
longer times the current decreases. As shown in the inset of Fig. 7a,
the initial current is approximately constant and independent of po-
tential with an average value of 130 mA cm . Due to the low bulk
Cu͑II͒ concentration ͑50 mM͒, the concentration at the surface is
quickly depleted and copper growth becomes diffusion limited after
about 50 ms. For a diffusion-limited process the current is given by
the Cottrell equation
−
1
+
0.04 V. The scan rate was 20 mV s .
−
2
for copper deposition, even though the Ni͑II͒ concentration is almost
an order of magnitude larger ͓Ni͑II͒/Cu͑II͒ = 8͔. This indicates that
the kinetics for copper deposition is initially much faster than for
nickel deposition. After about 5 s, when the gold surface is almost
completely covered with nickel, the current is exponentially depen-
dent on the applied potential, as shown in Fig. 8b. The inverse slope
gives a Tafel coefficient of 0.09 V per decade, close to published
nFD1/2c
i =
͓1͔
␲1
/2 1/2
t
2
8
where F is Faraday’s constant, D is the diffusion coefficient, c is the
bulk ion concentration, and n is the number of electrons evolved.
The current–time transients are replotted as i−2 vs t in Fig. 7b show-
ing a linear region from about 50 ms to 1 s, consistent with Eq. 1.
The diffusion coefficient, obtained from the slopes, is 7.7 Ϯ 0.3
values. The results presented in Fig. 7 and 8 confirm that, in the
potential range of interest, copper deposition is diffusion limited
while nickel deposition is kinetically limited. The current density for
copper during the early stages of nucleation and growth is much
higher than for nickel, even though the Ni͑II͒ concentration is much
larger ͓Ni͑II͒/Cu͑II͒ = 8͔.
−
6
2
−1
ϫ 10 cm s , in excellent agreement with values of 6-8
−
6
2
−1
25-27
ϫ 10 cm
s
reported.
Figure 9 shows plan-view SEM images of copper and nickel
islands deposited at different potentials. After 100 ms, copper forms
relatively large hemispherical islands with a relative low island den-
sity. These results are typical for the deposition of copper from
Figure 8a shows current–time transients for nickel deposition
from 400 mM Ni͑II͒ solution at −0.8, −0.9, and −1.0 V. The initial
current for nickel deposition is much lower than the initial current
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org

D572
Journal of The Electrochemical Society, 155 ͑9͒ D569-D574 ͑2008͒
Figure 7. ͑a͒ Current–time curves for deposition of copper from 50 mM
Cu͑II͒ at −0.8, −1.9, and −1.0 V. The inset shows the initial current vs
−
2
potential. ͑b͒ Current–time transients replotted as i vs t: ͑solid line͒
0.8 V, ͑dotted line͒ −0.9 V, and ͑dashed line͒ −1.0 V. The inset shows
Figure 8. ͑a͒ Current–time curves for deposition of nickel from 400 mM
Ni͑N HSO ͒ and 0.65 M HBO . ͑b͒ Current at −0.8, −0.9, and −1.0 V
−
2
3
2
3
the diffusion coefficient, calculated from the slopes of the linear regions vs
deposition potential.
after 5 s vs potential.
islands. Nickel has a much higher island density that, coupled with
the phase separation, results in relatively small nickel islands sur-
rounding the large copper islands. As described above, during the
first 50 ms ͑35 equivalent ML͒, the copper growth rate is much
2
5,26,29-33
different solutions.
As shown in Fig. 10, the island density
is independent of deposition time. The first measurement at 50 ms
corresponds to the shortest time where the copper islands can be
easily imaged. Analysis of the distribution of copper islands results
in a near-neighbor distance of 447 nm, within a factor of 2 of the
near-neighbor distance of the nodular features in the Cu–Ni thin
films, and is in good agreement with the near-neighbor distance
value of nodular features.
The initial rate of deposition of nickel is much slower than for
copper, as described above, and hence the nickel islands can only be
imaged after longer deposition times. After 5 s, nickel forms very
11
−2
small islands with a high island density ͑Ͼ5 ϫ 10 cm ͒ which is
3
orders of magnitude than the island density of copper.
The results from the analysis of copper and nickel deposition
reveal two important features. First, the initial copper current is very
high with a low island density, resulting in the fast formation of
large copper islands. Second, the initial nickel current is relatively
low and the island density is very high, resulting in the formation of
small islands.
Figure 11 shows the difference in current from the independent
experiments of copper and nickel deposition. At short times ͑50 ms͒,
the copper growth rate is much larger than the nickel growth rate.
The difference in growth rates then decreases as the copper ions at
the surface are depleted. At longer times, after about 10 s, steady-
state growth is achieved, which is consistent with the 50 nm transi-
tion region described above.
From these independent measurements of the nucleation and
growth of pure copper and pure nickel, we can provide new insight
into Cu–Ni codeposition, as shown schematically in Fig. 12. At
short times, copper growth is very fast and the island density is
relatively low, resulting in the formation of relatively large copper
Figure 9. ͑Left͒ Plan-view SEM images of Cu islands deposited from 50
mM Cu͑II͒ solution after 100 ms at −0.8, −0.9, and −1.0 V. ͑Right͒ Plan-
view SEM images of Ni islands deposited from 400 mM Ni͑II͒ solution after
5 s at −0.8, −0.9, and −1.0 V.
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org

Journal of The Electrochemical Society, 155 ͑9͒ D569-D574 ͑2008͒
D573
Figure 10. Cu island density vs time at −0.8, −0.9, and −1.0 V.
faster than the nickel growth rate, even though the Ni͑II͒ concentra-
tion in solution is almost an order of magnitude larger. At longer
times, the Cu deposition rate decreases, and steady-state growth is
achieved after about 10 s, corresponding to the transition region
described above ͑see Fig. 5͒. The miscibility gap provides the driv-
ing force for phase separation and leads to the formation of a co-
lumnar microstructure.
On the basis of our results, we can summarize the conditions
required to form a core/shell microstructure in a binary alloy system.
First, the system must exhibit a miscibility gap in the composition
range of interest, providing a thermodynamic driving force for phase
separation. Second, to form the core requires a material with a high
initial current and low island density. The fast initial kinetics results
in the formation of a low density of relatively large islands at short
times. Third, the shell requires a material with a relatively low initial
current and high island density, resulting in the formation of rela-
tively small islands. Based on these considerations, it should be
possible to extend the core/shell structure to other systems such as
Co–Cu.
Figure 12. ͑Color online͒ Schematic illustration of the evolution of the
core/shell microstructure. Initially, the growth rate of copper is very fast
compared to nickel and the relatively low island density results in the for-
mation of large copper islands surrounded by smaller nickel islands. As the
copper growth rate slows down, the core/shell microstructure evolves as
steady state is reached.
island density is relatively large, resulting in the formation of a high
density of nickel islands surrounding the larger copper islands. At
longer times, the copper growth and nickel growth reaches steady
state due to the decreased rate of copper deposition. The differences
in nucleation and growth kinetics, coupled with the miscibility gap,
result in this unique core/shell microstructure.
Acknowledgment
This work was supported by the JHU MRSEC ͑NSF grant no.
DMR05-20491͒.
Conclusion
Phase separation in electrodeposited Cu–Ni thin films results in
nodular features with a copper-rich core and nickel-rich shell. We
show that the evolution of this microstructure can be explained by
differences in nucleation and growth between copper and nickel. At
short times, copper deposition is initially fast and exhibits a rela-
tively low island density, resulting in the formation of large hemi-
spherical islands that become the cores in the nodular features. In
contrast, the kinetics of nickel deposition is initially slow but the
Johns Hopkins University assisted in meeting the publication costs of this
article.
References
1
. M. Hansen, Constitution of Binary Alloys, 2nd ed., p. 1305, McGraw-Hill, New
York ͑1958͒.
2. V. P. Glibin, B. V. Kuznetsov, and T. N. Vorobyova, J. Alloys Compd., 386, 139
2005͒.
͑
3
4
. T. A. Green, A. E. Russell, and S. Roy, J. Electrochem. Soc., 145, 875 ͑1998͒.
. E. Toth-Kadar, L. Peter, T. Becsei, J. Toth, L. Pogany, T. Tarnoczi, P. Kamasa, I.
Bakonyi, G. Lang, A. Cziraki, et al., J. Electrochem. Soc., 147, 3311 ͑2000͒.
. I. Mizushima, M. Chikazawa, and T. Watanabe, J. Electrochem. Soc., 143, 1978
5
6
7
͑
1996͒.
. Q. Wang, G. Wang, X. Han, X. Wang, and J. G. Hou, J. Phys. Chem. B, 109,
3326 ͑2005͒.
. ASM, Binary Alloy Phase Diagrams, ASM International, Materials Park, OH
1996͒.
2
͑
8
9
. S. Menezes and D. P. Anderson, J. Electrochem. Soc., 137, 440 ͑1990͒.
. M. Chen, C. L. Chien, and P. C. Searson, Chem. Mater., 18, 1595 ͑2006͒.
1
0. T. P. Moffat, J. Electrochem. Soc., 142, 3767 ͑1995͒.
11. D. M. Tench and J. T. White, J. Electrochem. Soc., 137, 3061 ͑1990͒.
12. J. Yahalom and O. Zadok, J. Mater. Sci., 22, 499 ͑1987͒.
13. L. Sun, C. L. Chien, and P. C. Searson, Chem. Mater., 16, 3125 ͑2004͒.
14. Z. Liu, G. Q. Xia, F. Zhu, S. Kim, N. Markovic, C. L. Chien, and P. C. Searson, J.
Appl. Phys., In press.
1
1
1
1
5. F. Q. Zhu, D. L. Fan, X. C. Zhu, J. G. Zhu, R. C. Cammarata, and C. L. Chien, Adv.
Mater. (Weinheim, Ger.), 16, 2155 ͑2004͒.
6. F. Q. Zhu, G. W. Chern, O. Tchernyshyov, X. C. Zhu, J. G. Zhu, and C. L. Chien,
Phys. Rev. Lett., 96, 027205 ͑2006͒.
7. J. C. Bao, C. Y. Tie, Z. Xu, Q. F. Zhou, D. Shen, and Q. Ma, Adv. Mater. (Wein-
heim, Ger.), 13, 1631 ͑2001͒.
8. K. Nielsch, F. J. Castano, C. A. Ross, and R. Krishnan, J. Appl. Phys., 98, 034318
͑2005͒.
Figure 11. The difference in current for the individual components ⌬i
=
icu − ini vs time at −0.8, −0.9, and −1.0 V.
19. K. Nielsch, F. J. Castano, S. Matthias, W. Lee, and C. A. Ross, Adv. Eng. Mater., 7,
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org

D574
Journal of The Electrochemical Society, 155 ͑9͒ D569-D574 ͑2008͒
2
17 ͑2005͒.
0. J. Escrig, P. Landeros, D. Altbir, E. E. Vogel, and P. Vargas, J. Magn. Magn. Mater.,
08, 233 ͑2007͒.
1. S. J. Son, J. Reichel, B. He, M. Schuchman, and S. B. Lee, J. Am. Chem. Soc., 127,
316 ͑2005͒.
148, C41 ͑2001͒.
2
2
27. T. I. Quickenden and Q. Z. Xu, J. Electrochem. Soc., 143, 1248 ͑1996͒.
28. A. J. Bard and H. Lund, Encyclopedia of Electrochemistry of the Elements, M.
Dekker, New York ͑1973͒.
29. A. Radisic, F. M. Ross, and P. C. Searson, J. Phys. Chem. B, 110, 7862 ͑2006͒.
30. A. Radisic, P. M. Vereecken, P. C. Searson, and F. M. Ross, Surf. Sci., 600, 1817
͑2006͒.
3
7
2
2
2. J. G. Zhu, J. Appl. Phys., 87, 6668 ͑2000͒.
3. NIST, Joint Committee on Powder Diffraction Standards–International Centre for
Diffraction Data ͑2001͒.
4. C. P. Wang, X. J. Liu, M. Jiang, I. Ohnuma, R. Kainuma, and K. Ishida, J. Phys.
Chem. Solids, 66, 256 ͑2005͒.
31. A. Radisic, A. C. West, and P. C. Searson, J. Electrochem. Soc., 149, C94 ͑2002͒.
32. G. Oskam, P. M. Vereecken, and P. C. Searson, J. Electrochem. Soc., 146, 1436
͑1999͒.
2
2
2
5. D. Grujicic and B. Pesic, Electrochim. Acta, 47, 2901 ͑2002͒.
6. A. Radisic, J. G. Long, P. M. Hoffmann, and P. C. Searson, J. Electrochem. Soc.,
33. M. Zheng, M. Willey, and A. C. West, Electrochem. Solid-State Lett., 8, C151
͑2005͒.
Downloaded on 2013-01-05 to IP 150.108.161.71 address. Redistribution subject to ECS license or copyright; see www.esltbd.org