Electrochemical preparation of porous copper surfaces in zinc chloride-1-ethyl-3-methyl imidazolium chloride ionic liquid

Article abstract of DOI:10.1149/1.2724154

The preparation of porous copper or copper-zinc surfaces by electrochemical formation of binary Cu-Zn alloys on Cu substrate and subsequent electrochemical etching of the zinc was investigated in a zinc chloride-1-ethyl-3- methylimidazolium chloride ionic liquid at 120°C. Cyclic voltammetry and X-ray diffraction measurements suggested that phase transformation from γ - to Β′ - Cu-Zn alloy occurred during constant potential dealloying. Essentially all the Zn content in the Cu-Zn could be removed from the alloy with dealloying at a sufficiently positive potential. Dealloyed materials exhibited well-developed bicontinuous porous structure. The dependence of the surface morphology of the porous Cu film on several experimental parameters, including deposition current and charge, and anodizing potential and temperature, were examined.

Full text of DOI:10.1149/1.2724154

D316
Journal of The Electrochemical Society, 154 ͑6͒ D316-D321 ͑2007͒
0013-4651/2007/154͑6͒/D316/6/$20.00 © The Electrochemical Society
Electrochemical Preparation of Porous Copper Surfaces in Zinc
Chloride-1-ethyl-3-methyl Imidazolium Chloride Ionic
Liquid
,z
*
Yi-Wen Lin, Chia-Cheng Tai, and I-Wen Sun
Department of Chemistry, National Cheng Kung University, Tainan, Taiwan 70101
The preparation of porous copper or copper-zinc surfaces by electrochemical formation of binary Cu–Zn alloys on Cu substrate
and subsequent electrochemical etching of the zinc was investigated in a zinc chloride-1-ethyl-3-methylimidazolium chloride ionic
liquid at 120°C. Cyclic voltammetry and X-ray diffraction measurements suggested that phase transformation from ␥- to ␤ -
Ј
Cu–Zn alloy occurred during constant potential dealloying. Essentially all the Zn content in the Cu–Zn could be removed from the
alloy with dealloying at a sufficiently positive potential. Dealloyed materials exhibited well-developed bicontinuous porous
structure. The dependence of the surface morphology of the porous Cu film on several experimental parameters, including
deposition current and charge, and anodizing potential and temperature, were examined.
© 2007 The Electrochemical Society. ͓DOI: 10.1149/1.2724154͔ All rights reserved.
Manuscript submitted November 29, 2006; revised manuscript received February 5, 2007. Available electronically April 18, 2007.
Porous copper is an interesting electrode material for the elec-
glove box filled with dry nitrogen by mixing proper amounts of
ZnCl2 ͑98% Aldrich͒ and EMIC at 90°C for 6 h. The resulting ionic
liquid was colorless.
trolysis process such as hydrogenation of CO2 into methanol.1-3 Po-
rous copper electrodes have been prepared by selective dissolution
of aluminum from CuAl2 alloy in NaOH solutions. This process is
known as dealloying and has also been applied for the preparation of
other porous metals, and the mechanism for formation of porous
metal surfaces by the dealloying process has been well discussed.4-12
In brief, the formation of porous structure is a result of a competi-
tion between the selective dissolution roughening process of the less
noble metal and the surface diffusion smoothing process of the more
noble metal. While thermal casting or sputter deposition methods
are often employed for the preparation of the precursor alloy CuAl2,
the electrodeposition method was not used because it is difficult to
find a suitable electrolyte for the electrodeposition of aluminum. To
overcome this, Cu–Zn, which can be formed electrochemically, may
be used in place of CuAl2.
Fabrication of porous copper surface.— The fabrication of po-
rous copper was conducted in a dry nitrogen-filled glove box.
The electrochemical experiments were accomplished with an
Autolab potentiostat/galvanostat controlled with GPES software. A
three-electrode electrochemical cell was used for the electrochemi-
cal experiments. The reference electrode was a Zn wire ͑Aldrich,
99.99%͒ placed in separated fritted glass tube containing pure
50–50 mol % ZnCl2-EMIC ionic liquid. The counter electrode was
a Zn spiral immersed in pure 50–50 mol % ZnCl2-EMIC contained
in a fritted glass tube. Preliminary studies showed that similar re-
sults were obtained at both Cu plates and Cu wires. The Cu ͑Ald-
rich, 99.99%͒ working electrode was cleaned by immersion in 2 M
HNO3 and rinsed with deionized water ͑specific resistivity
18.2 M cm͒ and dried before use. To fabricate the porous Cu elec-
trode, Cu–Zn surface alloys were first formed by electrodeposition
of Zn at the Cu wire in a 50.0–50.0 mol % ZnCl2-EMIC ionic liq-
uid. It was found that the electrodeposition of Zn could be promoted
by increasing the temperature. However, if the temperature was too
high, the electrodeposits became poorly adhered to the substrate. As
a result, the electrodeposition was performed at 120°C in this study.
After the electrodeposition, the Zn in the Cu–Zn surface alloys was
anodically stripped to create the porous surface. The microstructure
of the electrode was examined with a Philips XL40 field emission
scanning electron microscope ͑FESEM͒. The crystalline phases of
the metal samples were examined with a Shimadzu XD-D1 X-ray
diffractometer ͑XRD͒.
Zinc chloride reacts with 1-ethyl-3-methylimidazolium chloride
͑ZnCl2-EMIC͒, forming ionic liquids that are liquids near or below
ambient temperature.13 This ionic liquid system has been investi-
gated for electrodeposition applications.14-16 ZnCl2-EMIC is an
aprotic medium and provides a wide working temperature covering
from room temperature to above 150°C. The high working tempera-
ture may be advantageous for electrodeposition of alloys. Recently,
the fabrication of nanostructured platinum, gold, and silver films on
the surface of corresponding electrodes by electrochemical alloying/
dealloying in the ZnCl2-EMIC without using hazardous acids or
bases has been demonstrated.17-19 This method provides an easy way
to form porous metal films, and because the zinc͑II͒ species that was
consumed during the electrodeposition step was redissolved into the
ZnCl2-EMIC ionic liquid during the electrochemical dealloying
step, the ZnCl2-EMIC ionic liquid is reusable. In previous studies,
the effects of the experimental factors such as deposition current,
potential, and temperature on the structure of the porous metal films
have been examined. To verify the previous findings and further
explore this electrochemical alloying/dealloying approach, this
present paper examines the fabrication of porous copper films by
electrochemical formation and electrochemical dealloying of Cu–Zn
alloys in a 50–50 mol % ZnCl2-EMIC ionic liquid. TheCu–Zn alloy
phase change during the anodic dealloying step is noticed.
Results and Discussion
Cyclic voltammetry of Zn on Cu substrate.— In order to under-
stand the electrochemical behavior of the Zn͑II͒/Zn couple and the
formation of Cu–Zn alloy in the 50.0–50.0 mol % ZnCl2-EMIC
ionic liquid, cyclic voltammetry was conducted at a polycrystalline
Cu wire electrode at 120°C. A typical cyclic voltammogram is
shown in Fig. 1. For comparison, a voltammogram for the pure ionic
liquid recorded at a glassy carbon electrode ͑GCE͒ is also included
in this figure. As shown in Fig. 1A, the electrodeposition of Zn at
GCEs that occurs at potentials more negative than 0.0 V requires a
nucleation overpotential. Figure 1B shows that the electrodeposition
of Zn at Cu electrode occurs at a potential more positive than 0.0 V
without the nucleation current loop, indicating that the electrodepo-
sition of Zn at Cu may be facilitated by the formation of Cu–Zn
alloys. While Fig. 1A shows a single anodic wave for the stripping
of the Zn electrodeposits on the GCE, Fig. 1B shows that multiple
stripping waves are observed on the Cu electrode, indicating the
formation of different surface Cu–Zn alloys by the electrodeposited
Experimental
Synthesis of ZnCl2-EMIC ionic liquid.— EMIC was prepared
and purified according to the method described in literature.20 The
ZnCl2-EMIC ionic liquid was prepared in a Vacuum Atmospheres
*
Electrochemical Society Active Member.
z E-mail: iwsun@mail.ncku.edu.tw
Downloaded on 2014-07-16 to IP 132.203.227.63 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, 154 ͑6͒ D316-D321 ͑2007͒
D317
Figure 2. Cross-sectional SEM images and the distribution of Cu and Zn
contents in the samples that were obtained by electrodeposition of
2.54 C cm−2 Zn on Cu at 120°C. ͑A͒ The sample was kept at room tempera-
ture after the deposition. ͑B͒ The sample had been kept in the ionic liquid at
120°C for 12 h after the deposition.
Figure 1. Cyclic voltammograms recorded in
a 50.0–50.0 mol %
ZnCl2-EMIC ionic liquid at: ͑A͒ a glassy carbon disk electrode and ͑B͒ a
copper electrode at 120°C. ͑C͒ Stripping voltammograms recorded at a cop-
per electrode after different zinc deposition periods. The scan rate was
for 12 h after the conclusion of the electrodeposition. It can be seen
that although the initial deposited Zn layer was about 1 ␮m, the
Cu–Zn alloy layer has thickened to about 3 ␮m because of the in-
terdiffusion of Zn and Cu atoms. The fact that Cu also appears in the
top 1 ␮m deposited Zn layer indicates that the alloy layer was
formed not only by diffusion of the deposited Zn atoms into the Cu
substrate but also by diffusion of Cu atoms to the deposited Zn layer.
However, no significant alloy thickening was observed if the Zn-
deposited Cu sample was kept at room temperature after the elec-
trodeposition experiment, indicating that the interdiffusion of Zn and
Cu atoms was substantially retarded at lower temperature. This re-
sult is similar to that observed in the previous study on Ag–Zn
systems19 and confirms that temperature is an important factor on
the alloying process. To minimize the variation in the alloy thickness
due to the immersion period, in the subsequent experiments, the
dealloying step was performed immediately after the deposition
step.
50 mV s−1
.
Zn and the Cu substrate. According to the phase diagram of the
Cu–Zn system,21 ␣-Cu–Zn is stable at 0–30 atom % Zn. ␤ -CuZn is
Ј
approximately
a
1:1 compound, and ␥-CuZn exists at
60–67 atom % Zn. At even higher Zn content, ␧- and ␩-CuZn may
appear. Figure 1C shows the linear scan stripping voltammograms
recorded on the Cu electrode after the constant potential deposition
of Zn for different periods of time. This figure shows that when the
amount of deposited Zn was increased by increasing the deposition
period, waves a1 and a2 increased in magnitude whereas wave a3
approached a limiting value. From these results and the Cu–Zn
phase diagram, the broad anodic wave a1 may be attributed to the
dissolution of Zn from alloys such as ␥-CuZn that have high Zn
content, whereas wave a2 may due to the anodic dissolution of Zn
from ␤’-CuZn alloy. Finally, the dissolution of ␣-Cu–Zn occurs at
wave a3. Because the anodic stripping of the Zn in the Cu–Zn alloys
occurred at potentials far less positive than that for anodizing the Cu
substrate, it is possible to fabricate porous Cu through electrochemi-
cal etching ͑dealloying͒ of the Zn from the Cu–Zn alloy.
XRD measurements.— XRD measurements were conducted to
investigate the crystallographic properties of the Zn-deposited Cu
samples. Figure 3A shows the results of the as-deposited Cu–Zn
sample; this sample exhibits primarily the diffraction patterns of ␥-
CuZn ͑or CuZn2͒ alloy in addition to those of pure zinc and copper.
Figure 3B shows that the diffraction patterns of ␥-CuZn have trans-
Cross-sectional SEM images of the Zn-deposited Cu
samples.— Figure 2 shows the typical cross-sectional SEM images
and the distribution of Cu and Zn contents in the samples that were
obtained by electrodeposition of zinc on copper at 120°C. Figure 2A
shows the image of a Cu sample that had been galvanostatically
deposited with Zn and was immediately removed from the ionic
liquid after the electrodeposition process. On this sample, the depos-
ited Zn layer has a thickness slightly larger than 1 ␮m. The Zn
content decreased from the surface toward the inner substrate form-
ing alloy with Cu. Figure 2B shows the image of a sample that had
been prepared with the same manner as Fig. 2A, but this sample had
been immersed in the ionic liquid at 120°C at open-circuit potential
formed to those of the ␤ -CuZn ͑Cu Zn and CuZn͒ alloy after
Ј
5
8
anodic treatment of the as-deposited Cu–Zn sample at 0.10 V. How-
ever, patterns of ␣-Cu–Zn were not observed when the Cu–Zn alloy
was anodized at 0.3 V, probably due to the very small amount of
this alloy compared to other constituents. When the Cu–Zn alloy
samples were anodized at 0.4 V, XRD diffraction patterns like that
shown in Fig. 3C were obtained, which indicates that only Cu re-
mained and essentially all the Zn in the surface alloy had been
removed. Figure 3D shows the patterns of a Cu–Zn sample that has
been immersed in the ionic liquid at 120°C for 12 h after the elec-
trodeposition ͑the sample similar to that shown in Fig. 2B͒. This
Downloaded on 2014-07-16 to IP 132.203.227.63 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D318
Journal of The Electrochemical Society, 154 ͑6͒ D316-D321 ͑2007͒
Figure 5. Plan-view SEM images of copper wire samples that have been
electrodeposited at 120°C with 2.54 C cm−2 of zinc followed by dealloying
at 0.4 V. The amount ͑in C cm−2͒ of the zinc that was dealloyed is ͑A͒
1.27 C cm−2, ͑B͒ 1.59 C cm−2, ͑C͒ 1.91 C cm−2, and ͑D͒ 2.54 C cm−2. The
dealloying temperature was 120°C.
Figure 3. XRD patterns ͑Cu K␣͒ of Zn-deposited Cu samples from
ZnCl2-EMIC ionic liquid at 120°C: ͑A͒ as deposited, ͑B͒ after being deal-
loyed at 0.10 V, ͑C͒ after being dealloyed at 0.40 V, and ͑D͒ Cu–Zn sample
immersed in the ionic liquid for 12 h after the electrodeposition step ͑i.e., the
sample shown in Fig. 2B͒.
difference is not known, but may be the result from the relative
differences in the rates of the alloy phase transformation and diffu-
sion of the metal atoms.
figure indicates that the ␥-CuZn has transformed to ␤ -CuZn. The
Ј
The evolution of the porous copper surface during dealloying is
illustrated by the SEM microstructural images shown in Fig. 5.
Shown in Fig. 5A is the as-deposited surface which contains the
Zn–Cu alloy grains. As the dealloying process started, zinc atoms
were selectively removed from the outermost alloy surface, leading
to the formation of tiny pits. It appeared that the removal of Zn from
the alloy occurs preferentially at the grain boundaries than from
other areas, and as a result, during the early stage of the dealloying,
the grains turned into islands having tiny pits on them and separated
by crevices between them ͑Fig. 5B͒. As the dealloying proceeded,
the selective dissolution of Zn atoms released more copper atoms
onto the surface. The surface diffusion of these copper atoms results
in larger ligaments and increased pore size ͑Fig. 5C͒. Finally, the
grains turned into separated porous Cu islands of aggregated liga-
ments ͑Fig. 5D͒.
The bicontinuous porous Cu structure obtained in this study is
similar to those obtained for Au18 and Ag19 but different from that
obtained for Pt17 in the previous studies using the same alloying/
dealloying approach. For Pt, nano and microsized cracks rather than
pores were produced after dealloying. Such differences may be a
result of the very high melting point and thus, very low surface
diffusivity of Pt compared to that of the other three metals.23
phase transformation behavior has also been reported during the
dealloying of Ni–Zn alloy in a ZnCl2–NaCl molten salt system.22
Formation of porous copper.— Porous copper samples were
prepared by selective anodic stripping of Zn off the electrochemi-
cally formed Cu–Zn alloys. The total dissolution rate of the Zn is
affected by the charge-transfer rate ͑which is determined by the
applied anodic potential͒, the changing in the active surface area for
the dissolution, and the mass transport of the metal atoms. Figure 4
shows the typical current vs time curves during constant potential
dealloying of the Cu–Zn alloys at 0.40 V at 120°C. This figure
shows that the Zn dissolution rate ͑current͒ quickly reached a
steady-state value until most of the Zn was dealloyed, and then
decayed afterwards. The steady-state dealloying current suggests
that the rate of the creation of new active sites equals to that of the
removal of active dissolution sites. It is worth to mention that the
steady-state i-t behavior observed in this figure is different from that
was observed during the dealloying of Zn from Ag–Zn alloy. For the
dealloying of Ag–Zn, the dealloying current increased with time
until a maximum value was reached, indicating that the creation of
active sites was slow in the early part of the dealloying but increased
with time during the dealloying process. The exact reason of such a
Effect of dealloying potential.— The cyclic voltammogram
shown in Fig. 1 shows multiple anodic stripping waves due to the
various Cu–Zn alloys. To determine the proper potential for the
dealloying of Zn from the Cu–Zn surface alloy, constant potential
electrolysis experiments on electrochemically formed Cu–Zn
samples were conducted at different dealloying potentials covering
the range between 0.05 and 0.40 V. In each experiment, the deal-
loying electrolysis was continued until the dealloying current de-
creased to the background value. It was found that consistent with
the cyclic voltammogram shown in Fig. 1, when the dealloying po-
tential was less positive than 0.1 V, the dealloying current was fairly
low. When the dealloying potential was increased to values more
positive than 0.2 V, the dealloying rate ͑current͒ increased and the
time for completing the dealloying decreased significantly. The deal-
loying potential not only determines the dealloying rate but also
determines the amount of Zn that could be dealloyed from the
Cu–Zn alloy. The amount of the Zn dealloyed from the samples at
each dealloying potential was estimated from the number of charge
passed during the dealloying. By comparing the total amount of the
deposited Zn and the amount of the dealloyed Zn, the percentage of
Figure 4. Current-time curves during the anodic dealloying of copper
samples that have been electrodeposited with 6.35 C cm−2 of zinc and an-
odically dealloyed at 0.4 V. The temperature was 120°C.
Downloaded on 2014-07-16 to IP 132.203.227.63 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, 154 ͑6͒ D316-D321 ͑2007͒
D319
Table I. The percentage of zinc remaining in CuZn samples after
dealloying at different potentials.
Dealloying potential ͑V͒
Percentage remaining zinc ͑%͒
0.05
0.10
0.15
0.20
0.25
0.30
0.35
52
41
14
11
4
3
3
the Zn remaining in the Cu–Zn alloy was calculated and presented
in Table I. Table I indicates, as can be expected from the XRD
patterns shown in Fig. 3, that at less positive dealloying potential,
the percentage of the Zn remaining in the alloy is high because only
the Cu–Zn alloys with Zn content no less than ␥-CuZn could be
dealloyed, leading to the ␤ -CuZn and ␣-Cu–Zn that require more
Ј
positive potentials to dealloy. The amount of Zn that remained in the
alloy decreased with increasing dealloying potential, and a potential
more positive than 0.35 V is required to bring the Zn content to a
substantial low level. Shown in Fig. 6 are the plan-view SEM im-
ages of these samples. This figure shows that not only a porous Cu
Figure 7. Plan-view SEM images of the porous copper samples prepared by
electrodeposition of ͑A͒ 6.36, ͑B͒ 3.17, and ͑C͒ 1.59 C cm−2 of zinc and
anodically dealloyed at 0.40 V. The temperature was 120°C.
surface could be prepared but also a porous Cu–Zn surface could be
prepared. As the dealloying potential became more positive, more of
the Zn was stripped from the Cu–Zn film, and the coral-like struc-
ture of the sample evolved more evidently. It should be mentioned
that if the dealloying potential was so positive ͑Ͼ0.50 V͒ that dis-
solution of Cu may occur, the porous structure may be destroyed.
For the subsequent experiments, a dealloying potential of 0.40 V
was employed.
Effect of deposited zinc quantity.— The amount of zinc that is
electrodeposited at the copper surface may affect the composition of
the alloy and thus the microstructure of the porous Cu produced
after dealloying. A series of porous copper wires was prepared at
120°C by constant current ͑current density 0.64 mA/cm2͒ elec-
trodeposition of different amounts of Zn onto the copper substrate
and followed by selective dealloying at a constant potential of
0.40 V. It was found that because increasing the amounts of the
deposited Zn resulted in thicker alloy layers, the time period for
complete stripping of the Zn from the alloy increased accordingly.
The plan-view SEM images of the porous copper wires obtained
from the above experiments are shown in Fig. 7. As can be seen
from these images, all these samples exhibit coral-like structures
formed by aggregated ligament islands. While the ligaments in these
samples are of similar sizes, the ligaments’ aggregated islands are of
Figure 6. Plan-view SEM images of copper samples that have been elec-
trodeposited with 2.54 C cm−2 of zinc with a constant current density of
0.64 mA cm−2 at 120°C, and dealloyed at ͑A͒ 0.1, ͑B͒ 0.2, and ͑C͒ 0.35 V.
Downloaded on 2014-07-16 to IP 132.203.227.63 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D320
Journal of The Electrochemical Society, 154 ͑6͒ D316-D321 ͑2007͒
Figure 8. Plain-view SEM images of the copper samples that have been
electrodeposited with 2.54 C cm−2 of Zn at a constant density ͑mA cm−2͒ of
͑A͒ 0.4 mA and ͑B͒ 1.6 mA, and completely dealloyed at 0.40 V. The tem-
perature was 120°C.
different sizes depending on the quantity of the deposited Zn; alloys
having higher Zn content resulted in larger ligament aggregated is-
lands. This phenomenon may be related to the fact that the different
charges passed during the deposition step produced deposits of dif-
ferent grain sizes and each grain turned into a ligament aggregate
after dealloying. Because increasing the charge for deposition of Zn
increases the deposited grain size, larger ligament aggregates were
produced after dealloying. It may be worthy to note that the pres-
ence of wide crevices between the ligament aggregates could be
advantageous for applications in which diffusion of reactants to the
deeper layers of the porous material is important.
Figure 9. Plan-view SEM images of the copper samples that have been
electrodeposited with 2.54 C cm−2 of zinc at 120°C and dealloyed at 0.40 V.
The dealloying temperature was: ͑A͒ 90, ͑B͒ 120, and ͑C͒ 150°C.
due to the increased surface diffusion rate of the Zn and Cu atoms at
higher temperature, which increased the rearrangement and coarsen-
ing rate of the Cu porous structure
Effect of zinc deposition current.— The Zn deposition rate
͑deposition current͒ may affect the electrodeposited morphology, the
time period required for completing the electrodeposition, and the
extent of the Cu–Zn alloy formation which in turn may affect the
structure of the porous Cu that was produced after dealloying.
Figure 8 shows the plan-view SEM images of the porous Cu
after dealloying the Cu–Zn surface alloys that were prepared by
electrodeposition of the same amounts of Zn ͑2.54 C/cm2͒ with
different deposition currents. It can be seen that a more uniform
porous structure was obtained when the Cu–Zn surface alloy was
prepared with a lower deposition current ͑Fig. 8A͒. This is because
at the lower deposition current the deposited Zn is more compact.
When the deposition current was high, the deposits became less
compact and dealloying of such Cu–Zn film resulted in coarser po-
rous Cu ͑Fig. 8B͒.
Conclusion
The preparation of porous copper by electrochemical alloying
and dealloying of Cu–Zn alloy from the ZnCl
2-EMIC ionic liquid
was studied. The formation of Cu–Zn alloy by electrodeposition was
indicated by cyclic voltammetry, cross-sectional SEM, and XRD
measurements. Phase transformation during different stages of deal-
loying was revealed by XRD measurements. The current-time be-
havior during the dealloying of the Cu–Zn system was different
from that of the Ag–Zn system. While the porous structure produced
for the Cu in this study is similar to those obtained for Au and Ag in
previous studies, it is different from that for Pt using the same
alloying/dealloying approach. The morphologies of the dealloyed
material also depend on the dealloying temperature; higher dealloy-
ing temperature results in larger pore size due to the increased sur-
face diffusion rate of the metal atoms. The applications of the porous
Cu will be studied further.
Effect of dealloying temperature.— The effect of the dealloying
temperature on the porous structure was investigated by elec-
trodeposition of zinc on copper wires at a current density of
0.64 mA cm−2 at 120°C, and followed by dealloying the formed
Ag–Zn film at 0.40 V at 90, 120, and 150°C, respectively. The
corresponding SEM images of these samples are shown in Fig. 9.
These SEM images show that the pore size and the ligament size
both increased with increasing dealloying temperature. This may be
Acknowledgment
This work was supported by the National Science Council of the
Republic of China, Taiwan.
Downloaded on 2014-07-16 to IP 132.203.227.63 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, 154 ͑6͒ D316-D321 ͑2007͒
D321
10. H. W. Pickering and C. J. Wang, J. Electroanal. Chem. Interfacial Electrochem.,
114, 698 ͑1967͒.
11. M. J. Pryor and J. C. Fister, J. Electrochem. Soc., 131, 1230 ͑1984͒.
12. D. V. Pugh, A. Dursun, and S. G. Corcoran, J. Mater. Res., 18, 216 ͑2003͒.
13. S. I. Hsiu, J. F. Huang, I. W. Sun, C. H. Yuan, and J. Shiea, Electrochim. Acta, 47,
4367 ͑2002͒.
14. P. Y. Chen and I. W. Sun, Electrochim. Acta, 46, 1169 ͑2001͒.
15. J. F. Huang and I. W. Sun, J. Electrochem. Soc., 149, E348 ͑2002͒.
16. M. C. Lin, P. Y. Chen, and I. W. Sun, J. Electrochem. Soc., 148, C653 ͑2001͒.
17. J. F. Huang and I. W. Sun, Chem. Mater., 16, 1829 ͑2004͒.
18. J. F. Huang and I. W. Sun, Adv. Funct. Mater., 15, 989 ͑2005͒.
19. F. H. Yeh, C. C. Tai, J. F. Huang, and I. W. Sun, J. Phys. Chem. B, 110, 5215
͑2006͒.
20. J. S. Wilkes, J. A. Levisky, and R. A. Wilson, Inorg. Chem., 21, 1263 ͑1982͒.
21. H. Baker, Alloy Phase Diagrams, ASM International, Materials Park, OH ͑2006͒.
22. A. Katagiri and M. Nakata, J. Electrochem. Soc., 150, C585 ͑2003͒; T. Fukumizu,
F. Kotani, A. Yoshida, and A. Katagiri, J. Electrochem. Soc., 153, C629 ͑2006͒.
23. E. G. Seebauer and C. E. Allen, Prog. Surf. Sci., 49, 265 ͑1995͒.
National Cheng-Kung University assisted in meeting the publication
costs of this article.
References
1. J. B. Friedrich, M. S. Wainwright, and D. J. Young, J. Catal., 80, 1 ͑1983͒.
2. A. J. Smith, T. Tran, and M. S. Wainwright, J. Appl. Electrochem., 29, 1085
͑1999͒.
3. J. Toyir, M. Saito, I. Yamauchi, S. C. Luo, J. G. Wu, I. Takahara, and M. Takeuchi,
Catal. Today, 45, 245 ͑1998͒.
4. A. J. Forty, Nature (London), 282, 597 ͑1979͒.
5. J. D. Fritz and H. W. Pickering, J. Electrochem. Soc., 138, 3209 ͑1991͒.
6. T. P. Moffat, F. R. F. Fan, and A. J. Bard, J. Electrochem. Soc., 138, 3224 ͑1991͒.
7. R. C. Newman, S. G. Corcoran, J. Erlebacher, M. J. Aziz, and K. Sieradzki, MRS
Bull., 24, 24 ͑1999͒.
8. J. Erlebacher, M. J. Aziz, A. Karma, N. Dimitrov, and K. Sieradzki, Nature (Lon-
don), 410, 450 ͑2001͒.
9. Y. Ding, Y. J. Kim, and J. Erlebacher, Adv. Mater. (Weinheim, Ger.), 16, 1897
͑2004͒; J. Erlebacher, J. Electrochem. Soc., 151, C614 ͑2005͒.
Downloaded on 2014-07-16 to IP 132.203.227.63 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).