Dealloying behavior of amorphous binary Ti-Cu alloys in hydrofluoric acid solutions at various temperatures

Article abstract of DOI:10.1016/j.jallcom.2013.07.144

A nanoporous copper structure was fabricated from amorphous binary TixCu_100-x (x = 40, 50 and 60 at.%) alloys in hydrofluoric acid solutions under a free corrosion condition. A bi-continuous nanoporous structure has a pore size of 25-75 nm and a ligament size of 46-79 nm in 0.03 M HF solution. In 0.13 M HF solution, a pore size of 85-380 nm and a ligament size of 80-338 nm were obtained. The dealloying process is thermal-activated and obeys the Arrhenius law. The surface diffusivity decreases with temperature and concentrations of dealloying solutions. The values of the activation energy of copper adatoms are estimated to be 40.3-78.1 kJ mol^-1, which depend not only on the alloy composition but also on the dealloying electrolyte composition.

Full text of DOI:10.1016/j.jallcom.2013.07.144

Journal of Alloys and Compounds 581 (2013) 567–572
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Dealloying behavior of amorphous binary Ti–Cu alloys in hydrofluoric
acid solutions at various temperatures
Zhenhua Dan a,b, , Fengxiang Qin , Shin-ichi Yamaura , Yu Sugawara , Izumi Muto , Nobuyoshi Hara
⇑
a
a
b
b
b
a
Institute for Materials Research, Tohoku University, Sendai 9808577, Japan
Department of Materials Science, Tohoku University, Sendai 9808579, Japan
b
a r t i c l e i n f o
a b s t r a c t
Article history:
A nanoporous copper structure was fabricated from amorphous binary Ti Cu
100ꢀx
(x = 40, 50 and 60 at.%)
x
Received 1 February 2013
Received in revised form 27 May 2013
Accepted 21 July 2013
alloys in hydrofluoric acid solutions under a free corrosion condition. A bi-continuous nanoporous struc-
ture has a pore size of 25–75 nm and a ligament size of 46–79 nm in 0.03 M HF solution. In 0.13 M HF
solution, a pore size of 85–380 nm and a ligament size of 80–338 nm were obtained. The dealloying pro-
cess is thermal-activated and obeys the Arrhenius law. The surface diffusivity decreases with tempera-
ture and concentrations of dealloying solutions. The values of the activation energy of copper adatoms
Available online 30 July 2013
Keywords:
ꢀ1
are estimated to be 40.3–78.1 kJ mol , which depend not only on the alloy composition but also on
Amorphous materials
Nanostructured materials
Corrosion
the dealloying electrolyte composition.
Ó 2013 Elsevier B.V. All rights reserved.
Diffusion
Transmission electron microscopy
1
. Introduction
Nanoporous metals have attracted much attention due to their
at the solid/electrolyte interface. Moreover, the dealloying, as a
thermally-activated process, obeys the Arrhenius relationship
[7,8,15–17]. The evolution of nanoporosity might be affected by
material factors (the composition in the alloy systems; the crystal-
line properties), the dealloying temperatures and dealloying elec-
trolytes. The activation energy for the coarsening of NPC during
potential in various applications, but especially as catalyst, sensors,
and actuators [1–4]. Nanoporous silver, gold, platinum were fabri-
cated from a dealloying process by many research groups [5–8].
Monolithic nanoporous copper (NPC) was fabricated from abun-
dant crystalline alloy systems, for instance, Al–Cu [5] and Mn–Cu
ꢀ1
dealloying process has been reported to be 47.7 and 145.3 kJ mol
in HCl solutions, which corresponds to NPCs with different pore
sizes formed in various regions where the phase separation occurs
during dealloying in the crystalline Al–Cu alloy systems [15]. The
surface diffusivity exhibits a dependency on the reciprocal of tem-
perature, and well obeys the Arrhenius law. Qian et al. [7] investi-
gated the dealloying behavior of binary Ag65Au35 alloy in
[
9,10]. Recently, the NPC was fabricated by dealloying amorphous
Ti–Cu alloys in hydrofluoric acids [11]. From an electrochemical
perspective, the standard electrode potentials of Ti and Cu
2
+
(
ꢀ1.630 V vs. the standard hydrogen electrode (SHE) for Ti/Ti
2+
and 0.342 V vs. SHE for Cu/Cu ) differ as much as 1.972 V [12],
which promise selective leaching of Ti from binary Ti–Cu alloys
in an appropriate HF solution, where Ti dissolves as stable ions
concentrated HNO
3
solution at temperatures of ꢀ25, 0, and 25 °C,
and they found that the coarsening rates and surface relaxing de-
pend on temperature and etching time. However, the dependency
of the NPC formation on amorphous Ti–Cu precursors in HF solu-
tions is still remaining unclear.
2
6
ꢀ
such as TiF . The uniform nanoporous structure was fabricated
more readily from the amorphous alloys because amorphous alloys
can be characterized by their disordered atomic-scale structure
and an absence of defects and heterogeneous microstructures, i.e.
grain boundaries, second phases and element segregation [13,14].
The evolution of the nanoporosity is regarded as a spinodal
decomposition process, and the dealloying is dynamically evoked
The present research focused on the dealloying behavior of
amorphous Ti–Cu precursors in aspects of their electrochemical
reactions and evolution of nanoporous structures. The starting
x
materials, amorphous binary Ti Cu100ꢀx (x = 40, 50 and 60 at.%) al-
loys, were immersed in HF solutions under a free corrosion condi-
tion at different dealloying temperatures. The characteristic scale
length of the nanopores and ligaments of NPCs was investigated
at different temperatures. The surface diffusivity and activation en-
ergy during the NPC formation was also investigated.
⇑
Corresponding author. Address: Institute for Materials Research, Tohoku
University 2-1-1, Katahira, Aoba-ku, Sendai 9808579, Japan. Tel.: +81 22 215
2
718; fax: +81 22 215 2714.
E-mail address: zhenhuadan@imr.tohoku.ac.jp (Z. Dan).
0
925-8388/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jallcom.2013.07.144

5
68
Z. Dan et al. / Journal of Alloys and Compounds 581 (2013) 567–572
2
. Experimental procedure
2 2
Cu O overlapped with those from Cu. The formation of Cu O
phases was confirmed on the residual porous Cu on dealloyed Ti60-
Cu40 ribbons. On the contrary, no visible Cu O peaks appeared on
dealloyed Ti50Cu50 and Ti40Cu60 ribbons. Therefore, the residue of
Binary Ti–Cu alloys with nominal compositions of 40, 50, 60 at.% Cu were pre-
pared by arc melting of the mixture of pure Ti (99.99 wt%) and pure Cu (99.9 wt%) in
an Ar atmosphere. The melt spinning method was used to prepare Ti–Cu amor-
2
phous ribbons with 20
formed in 0.03 M and 0.13 M HF solutions under a free corrosion condition open to
air for 10.8 ks at temperatures of 298 K, 323 K and 348 K. The pH of 0.03 M and
l
m in thickness and 2 mm in width. The dealloying was per-
dealloyed alloys had the structure of face-centered cubic (fcc) Cu
(JCPDS card No.: 02-1225) and trace Cu O (JCPDS card No.: 74-
2
1
230) in some cases. The fcc Cu residue was confirmed for all rib-
0
.13 M HF solution was 3.3 and 2.90, respectively. The amorphicity of the as-spun
bons dealloyed at three different temperatures. Sherrer‘s equation
[18] was used to estimate the grain size for Ti60Cu40, Ti50Cu50 and
Ti40Cu60 ribbons after dealloying in 0.13 M HF solution at different
temperatures. The values of the calculated grain size are listed in
Table 1. The Cu grain sizes decreased with an increase in Cu con-
tents in the precursors.
Ti–Cu samples and the dealloyed alloys was confirmed by XRD using an X-ray dif-
fractometer (Rigaku, RINT 4200). The microstructures of dealloyed Ti–Cu ribbons
were investigated by a transmission electron microscope (JEOL, JEM-HC2100) and
a high-resolution transmission electron microscope (JEOL, JEM-ARM210F). The
TEM sampling was conducted by an ion milling method. The morphology of the
dealloyed samples was observed by
a scanning electron microscope (JEOL,
JIB-4610F) equipped with an energy dispersive X-ray spectroscope (EDX).
3
. Results
3.2. Morphology of dealloyed Ti–Cu alloys
3
.1. Characterization of microstructure of as-spun Ti–Cu alloys
Fig. 2 shows the surface morphology of dealloyed Ti60Cu40 and
Ti50Cu50 ribbon alloys in 0.03 M and 0.13 M HF solutions at
298 K. The dealloyed ribbons exhibited an open, bi-continuous
interpenetrating pore-to-ligament structure with length scales of
few tenth nanometers to several hundredth nanometers. The
cross-sectional morphology show that the NPCs had three-dimen-
sional structures, and the NPC pore sizes decreased while Cu atom-
ic ratios in the Ti–Cu ribbons increased. The figures which are
shown below the SEM graphs indicate the distribution of pore
and ligament sizes of NPCs. The pore sizes and ligament sizes were
estimated by the single chord length method over 150 sites of SEM
morphology. The mean values of nanopores of Ti60Cu40 and Ti50-
Cu50 ribbon alloys were found to be 37 nm and 31 nm in 0.03 M
HF solution, and 86 nm and 185 nm in 0.13 M HF solution, respec-
tively. The mean values of ligament sizes were confirmed to be
The XRD patterns of as-spun and dealloyed Ti–Cu ribbons are
showed in Fig. 1. A broad diffraction peak appeared in the XRD pat-
terns of as-spun Ti40Cu60, Ti50Cu50 and Ti60Cu40 ribbons at an angle
around 2h of 41°, which is typical of the amorphous alloys. The
amorphicity of Ti–Cu alloys remained at a wide range of Cu con-
tents. The diffraction peaks of dealloyed ribbons were assigned to
Cu (111), (200) and (220). The small peak at 2h = 36.6° in
2
Fig. 1a was risen from Cu O (111), and other diffraction peaks from
4
6 nm and 57 nm in 0.03 M HF solution, and 94 nm and 185 nm
in 0.13 M HF solution, respectively. The nanoporous structures
formed in dilute HF solutions had smaller pores and narrow liga-
ments, which indicated their higher nanoporosity. With an in-
crease in Cu contents, the nanopores became smaller for Ti60Cu40
and Ti40Cu60 ribbon alloys.
Fig. 3 shows the nanoporous structures of dealloyed Ti60Cu40
and Ti40Cu60 ribbons after immersion of 10.8 ks at 323 K and
3
3
48 K. The nanopores formed on Ti60Cu40 ribbons at 323 K and
48 K had a characteristic pore size of 170 nm and 227 nm, and
those of Ti40Cu60 ribbon samples were 121 nm and 246 nm, respec-
tively. In general, it could be observed that the pore sizes increased
with the dealloying temperature. The EDX spectra of dealloyed
Ti40Cu60 ribbon alloy indicated that residue was Cu, and other ele-
ments, such as C and Re, were from the adhesive tapes and SEM
electron guns. The residue consisted of trace Ti element, and Ti
contents seemed slightly higher on the samples dealloyed at higher
temperatures.
The changes in pore sizes and ligament sizes measured from
SEM morphology are summarized in Fig. 4 as a function of dealloy-
ing temperature. The nanopores increase with increasing tempera-
ture as well as HF concentration except some cases for Ti50Cu50 and
Ti60Cu40 samples. The nanopores formed on amorphous Ti50Cu50
Table 1
Cu grain sizes of NPCs obtained from amorphous ribbons after dealloying in 0.13 M
HF solution for 10.8 ks at different temperatures. Grain sizes are calculated according
to Scherer equation [18] (unit: nm).
Alloys
Temperature
98 K
2
323 K
348 K
Ti40Cu60
Ti50Cu50
Ti60Cu40
24
21
16
19
17
14
25
24
19
Fig. 1. XRD patterns of Ti60Cu40 (a), Ti50Cu50 (b) and Ti40Cu60 (c) amorphous ribbons
before (I) and after dealloying in 0.13 M HF solution for 10.8 ks at different
temperatures of 298 K (II), 323 K (III) and 348 K (IV).

Z. Dan et al. / Journal of Alloys and Compounds 581 (2013) 567–572
569
Fig. 2. SEM morphology of dealloyed Ti60Cu40 and Ti50Cu50 amorphous ribbons after dealloying in 0.03 M (a and c) and 0.13 M (b and d) HF solution for 10.8 ks at 298 K. The
size distribution ratios of nanopores and ligaments are shown below the corresponding SEM morphology.
ribbon alloys in both 0.03 and 0.13 M HF solutions are relatively
larger than those obtained on Ti60Cu40 and Ti40Cu60 ribbons.
pattern) of dealloyed Ti40Cu60 ribbons immersed in 0.03 M HF solu-
tion at 348 K for 10.8 ks. The three-dimensional bi-continuous por-
ous structure is clearly visible, and the characteristic pore size was
confirmed to be 38 nm for the internal nanoporous structures. On
the other hand, the characteristic pore size confirmed from SEM
images was 68 nm, which indicated that the nanoporous structure
formed in the internal regions had a finer structure. As shown in
3.3. Characteristics of dealloyed Ti–Cu alloys
Fig. 5a–c shows the bright-field TEM image, the high-resolution
TEM (HR-TEM) image and selective area diffraction patterns (SAD

5
70
Z. Dan et al. / Journal of Alloys and Compounds 581 (2013) 567–572
Fig. 3. SEM morphology of dealloyed Ti60Cu40 and Ti40Cu60 ribbons after dealloying in 0.13 M HF solution at 323 K (a and c) and 348 K (b and d) for 10.8 ks and the typical EDX
spectra (e) of dealloyed Ti40Cu60 ribbons after dealloying in 0.13 M HF solution for 10.8 ks at different temperatures.
Fig. 4. The mean pore size, d, and mean ligament size, l, of dealloyed Ti40Cu60, Ti50Cu50 and Ti60Cu40 ribbon alloys as a function of temperature, T, in 0.03 M HF solution (a) and
0
.13 M HF solution (b).
Fig. 5b, the interplanar distance from the adjacent lattice fringes
marked in the HR-TEM images was 0.209 nm, corresponding to
the {111} planes of fcc Cu. The lattice constants of residue Cu esti-
mated from XRD analysis were found to have values of 0.3609–
T is the temperature, and a is the lattice constant. The surface en-
ꢀ2
ergy is considered to be 1.79 J m for Cu [7,19]. The lattice con-
stants of NPCs obtained from the XRD data were used to calculate
the surface diffusivity.
0
0
.3619 nm, which are slightly larger than that of pure Cu of
.3607 nm. The HR-TEM image shows that the NPCs have a poly-
Since the dealloying process is considered to be a thermally-
activated process, the measurement of the activation energy will
be helpful for understanding the dealloying mechanisms. The
Arrhenius equation used for describing for surface diffusion pro-
cess is as follows [20]:
crystalline structure, indicating the original Cu atoms are rear-
ranged to form a crystalline structure as they were dealloyed
[
16,17]. As shown in Fig. 5c, the diffraction rings are assigned to
ꢀ
ꢁ
Cu (111), (200), (220), (311) and (222). The Cu O specie was
2
E
RT
a
not detected in the TEM observation, which is consistent with
XRD patterns (Fig. 1c). The SAD pattern indicated that the deal-
loyed Ti40Cu60 ribbons have the structure of fcc Cu.
D
s
¼ D
0
exp
ꢀ
ð2Þ
where D is the diffusion constant, R is the gas constant and E is the
0
a
activation energy for the surface diffusion process.
The surface diffusivity of Ti60Cu40 ribbon alloy dealloyed in
3
.4. Estimation of surface diffusivity
ꢀ19
2 ꢀ1
0
2
.03 M HF solution was estimated to be 7.08 ꢂ 10
m s
at
ꢀ18
2 ꢀ1
98 K and 6.66 ꢂ 10
m s at 348 K. On the other hand, the sur-
On the basis of the surface diffusion controlled coarsening
face diffusivity of Ti60Cu40 ribbon alloy dealloyed in 0.13 M HF
s
mechanism, the surface diffusivity, D , at various dealloying tem-
perature was estimated by the equation [19]:
ꢀ
17
2 ꢀ1
solution was estimated to be 2.16 ꢂ 10
m s
at 298 K and
ꢀ15
2 ꢀ1
1
.21 ꢂ 10
m s at 348 K. The surface diffusivity increased al-
4
most one order of magnitude due to the increase in dealloying
temperature, and the drastic increase in the surface diffusivity in
the concentrated HF solutions was more than 2 orders of magni-
tude, which indicates that the effect of dealloying temperature
on evolution of nanoporosity in concentrated HF solutions was
½
dðtÞꢁ kT
D
s
¼
ð1Þ
4
32cta
where k is Boltzmann constant (1.3806 ꢂ 10ꢀ23 J K ),
ꢀ1
c
is surface
energy,t is the dealloying time (10 800 s), d(t) is the pore size at t,

Z. Dan et al. / Journal of Alloys and Compounds 581 (2013) 567–572
571
nanopore size and surface diffusivity presented here is consistent
with the recent kinetic Monte Carlo simulation [17]. The value of
activation energy and diffusion constants extracted from the slopes
(
ꢀE
a 0
/R) and the intercepts (lnD ) of Fig. 6 are listed in Table 2. The
activation energy was found to have values in the range from 40.3
ꢀ1
to 78.1 kJ mol for Ti–Cu alloy systems. The activation energy is
strongly depended on the alloy composites of the amorphous
Ti–Cu precursors.
3.5. Discussion
The formation of NPCs during dealloying includes several steps:
(
2
1) The breakdown of native TiO oxide layer. (2) The selective dis-
solution of Ti atoms. (3) The diffusion of Cu adatoms. (4) The rear-
rangements of Cu adatoms and the formation of ligaments.
2
From the perspectives of electrochemistry, the native TiO layer
with a thickness about 2–5 nm was broken down since the fluoride
ions triggered the formation reactions of high soluble fluoride com-
2
6
ꢀ
2þ
plex (i.e. TiF , TiF2 ) [21]. Meanwhile, the bare Ti substrates were
ꢀ
attacked by F ions continuously, which is a selective dissolution
process of Ti atoms from Ti–Cu matrix. Our former research results
indicated that the dissolution of Ti elements was so fast in 0.13 M
HF solution that the Cu contents in the surface regions became
over 90 at.% for Ti40Cu60 alloy, over 97 at.% for Ti50Cu50 alloy and
over 99 at.% for Ti60Cu40 alloy after the first immersion of 600 s.
On the other hand, the change of Cu contents in 0.03 M HF solution
ꢀ
was relatively slow [11]. The F ions had an ability to produce
more active corrosion potential and to decrease the anodic and
cathodic (hydrogen evolution) Tafel slopes, which resulted in the
increase of dissolution rates of Ti atoms in the concentrated HF
Fig. 5. Bright-field TEM image (a), HR-TEM image (b) and SAD pattern (c) of
Ti40Cu60 ribbons dealloying in 0.03 M HF solution after dealloying for 10.8 ks at
2 4
solutions. The critical corrosion currents of Ti electrodes in H SO
3
48 K.
and NaF increased more than 3 orders of magnitude when the
NaF concentration increased from 0 to 0.12 M [21]. The behavior
described above is also enhanced by temperature. With increasing
temperature from 297 K to 353 K, the critical corrosion currents of
much stronger than that in diluted HF solutions. This fact revealed
that the larger nanopores formed preferentially in the concen-
trated HF solution at the higher dealloying temperatures. The sur-
2 4
Ti electrodes in the mixture solution of 0.5 M H SO and 0.59 mM
ꢀ
16
2
ꢀ1
NaF increased almost one order in magnitude. The buffer ability of
the dealloying solutions also played an important role for the for-
mation of nanoporous structures [22]. Since the HF solution had no
face diffusivity at 348 K was estimated to be 2.30 ꢂ 10
m s
ꢀ15
2 ꢀ1
for Ti40Cu60 ribbon alloy and 9.45 ꢂ 10
m s for Ti50Cu50 rib-
bon alloy in 0.13 M HF solution. The surface diffusivity of Ti50Cu50
ribbon alloys dealloyed in 0.13 M HF solution was larger than the
other binary alloy systems, which in fact caused the severe surface
relaxation and the formation of rough nanopores. Fig. 6 shows the
buffer ability, the dissolution of Ti
hydrogen ions to generate hydrogen gas bubbles during dealloying
11]. As a result, the pH of the solution was increased. The forma-
tion of Cu O on Ti60Cu40 ribbon alloys was considered to be a direct
evidence for the increase in pH. According to E–pH diagram of Cu
in H O at 298 K, the critical pH for the formation of Cu O is higher
2
O and Ti atoms consumed
[
2
Arrhenius plots of lnD
.13 M HF solutions. Good linear relationships were obtained for
the experimental data. The strong temperature dependence of the
s
vs. 1000/T of each Ti–Cu alloy in 0.03 M and
0
2
2
than 6 [23]. When the pH of the solution was increased from 3 to 5,
the critical corrosion current decreased more than one order of
magnitude [24]. Therefore, the nanopores formed in Ti60Cu40 rib-
bons were relatively smaller than those of Ti50Cu50 ribbons.
From the physical perspectives, the diffusion and rearrange-
ment of Cu adatoms are regarded to be the key steps for the forma-
tion of NPCs. These steps evoke mainly at the interfacial regions
and these surface diffusion controlled processes exhibit a strong
Table 2
Activation energy and diffusion constant of dealloyed Ti40Cu60, Ti50Cu50 and Ti60Cu40
amorphous alloys derived from Eqs. (1) and (2).
Alloy system
Solutions (M HF)
Activation energy
Diffusion constant
(D , m s )
0
ꢀ
1
2 ꢀ1
(E
a
, kJ mol
)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
6
Ti40Cu60
Ti40Cu60
Ti50Cu50
Ti50Cu50
Ti60Cu40
Ti60Cu40
0.03
0.13
0.03
0.13
0.03
0.13
78.1
77.9
55.3
67.9
40.3
74.1
8.45 ꢂ 10
7.14 ꢂ 10
3.00 ꢂ 10
2.03 ꢂ 10
7.60 ꢂ 10
2.42 ꢂ 10
4
7
6
12
4
Fig. 6. Arrhenius plot of natural logarithm of surface diffusivity, ln(D
reciprocal of temperature, 1000/T.
s
), vs. the

5
72
Z. Dan et al. / Journal of Alloys and Compounds 581 (2013) 567–572
dependency on temperature [7]. Evangelakis et al. investigated the
self-diffusion of Cu adatoms by the molecular dynamic method
are prevailed at higher temperatures, which resulted in the forma-
tion of rough nanoporous structure.
[
25]. They found that Cu adatom diffusion took place exclusively
by hopping from one adatom position to an adjacent one and that
multiple jumps are frequent at low temperatures. Montalenti et al.
found that the long jump occurred frequently at high temperatures
Acknowledgements
The authors gratefully acknowledge financial support by the
Ministry of Education, Culture, Sports, Science, and Technology
(MEXT) through Grant-In-Aid for Science Research in a Priority
Area on ‘‘Research and Development Project on Advanced Materi-
als Development and Integration of Novel Structured Metallic
and Inorganic Materials’’ and a Grant-in-Aid for Young Scientists
(B) under Grant No. 24760567. We would like to thank Mr. Kobay-
ashi for his help with TEM observations.
[
26]. The in-channel diffusion, such as single jump or double jump,
is preferred at low temperatures. On the contrary, cross-channel
jump diffusion, such as long jump or double exchange jump, may
appear at high temperatures. When the temperature increased
from 450 K to 600 K, the percentage of the long jump increased
from 6% to 15% for Cu adatoms, and the single jumps were domi-
nant at 450 K. Compared with Au, the long jumps are practically
absent in the above mentioned temperature ranges, which is con-
tributed to the formation of fine nanoporous structures for deal-
loyed Ag–Au alloys [7]. Boisvert et al. has reported that the
frequency of jump is higher than that of exchange at the lower
temperatures [27]. Therefore, in our experiments, the rough nano-
porous structures formed at higher temperatures were mainly
attributed to the increase in the cross-channel jump diffusion,
which prevails in a longer diffusion distance and rearranges in rel-
ative larger length scales. Because the diffusion and rearrangement
of Cu adatoms proceeded in a relative large scale length, the nano-
porous structures had a large nanopores and wider ligaments at
high temperatures. The formation of the narrow ligaments on Ti60-
Cu40 ribbon in 0.03 M HF solution at 348 K might be due to the
complex effects of temperature and local pH changes.
References
[
[
[
1] D. Kramer, R.N. Viswanath, J. Weissmüller, Nano Lett. 4 (2004) 793–796.
2] G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng. 41 (1999) 319–388.
3] T. You, O. Niwa, M. Tomita, S. Hirono, Anal. Chem. 75 (2003) 2080–2085.
[4] J.R. Weissmueller, N. Viswanath, D. Kramer, P. Zimmer, R. Wuerschum, H.
Gleiter, Science 300 (2003) 312–315.
[
5] Z.H. Zhang, Y. Wang, Z. Qi, W.H. Zhang, J.Y. Qin, J. Frenzel, J. Phys. Chem., C 113
2009) 12629–12636.
[6] J.S. Yu, Y. Ding, C.X. Xu, A. Inoue, T. Sakurai, M.W. Chen, Chem. Mater. 20 (2008)
548–4550.
(
4
[
[
7] L.H. Qian, M.W. Chen, Appl. Phys. Lett. 91 (2007) 083105-1–083105-3.
8] Y. Ding, Y.J. Kim, J. Erlebacher, Adv. Mater. 16 (2004) 1897–1900.
[9] J.R. Hayes, A.M. Hodge, J. Biener, A.V. Hamza, J. Mater. Res. 21 (2006) 2611–
2616.
[
[
10] L.Y. Chen, T. Fujita, Y. Ding, M.W. Chen, Adv. Funct. Mater. 20 (2010) 2279–
2
285.
11] Z.H. Dan, F.X. Qin, Y. Sugawara, I. Muto, N. Hara, Intermetallics 29 (2012) 14–
0.
2
4
. Conclusion
[12] D.R. Lide, CRC Handbook of Chemistry and Physics, 84th ed., CRC Press LLC,
Boca Raton, London, New York, Washington, DC, 2003.
[
[
13] W. Klement, R.H. Jun, Willens, P. Duwez, Nature 187 (1960) 869–870.
14] A. Inoue, Acta Mater. 48 (2000) 263–279.
In conclusion, an evolution of nanoporous structure from binary
amorphous Ti
x
Cu100ꢀx (x = 40, 50 and 60 at.%) alloys was investi-
[15] W.B. Liu, S.C. Zhang, N. Li, J.W. Zheng, Y.L. Xing, Micropor. Mesopor. Mater. 138
2011) 1–7.
16] J. Erlebacher, M.J. Aziz, A. Karma, N. Dimitrov, K. Sieradzki, Nature 410 (2001)
50–453.
(
gated from an aspect of Arrhenius relationships. An interpenetrat-
ing three-dimensional bi-continuous nanoporous structure of fcc
Cu was formed at different temperatures with a pore size of
[
4
[17] J. Erlebacher, J. Electrochem. Soc. 151 (2004) C614–C626.
[18] A.L. Patterson, Phys. Rev. 56 (1939) 978–982.
[
[
[21] M.J. Mandry, G. Rosenblatt, J. Electrochem. Soc. 119 (1972) 29–33.
[22] Z.H. Dan, F.X. Qin, Y. Sugawara, I. Muto, N. Hara, Intermetallics 31 (2012) 157–
2
5–380 nm and a ligament size of 46–338 nm. The characteristic
19] J.M. Dona, J. Gonzalez-Velasco, J. Phys. Chem. 97 (1993) 4714–4719.
20] E.G. Seebauer, C.E. Allen, Prog. Surf. Sci. 49 (1995) 265–330.
scale length of nanopores and ligaments at lower temperatures
was almost one order of magnitude smaller than those formed at
higher temperatures. The finer nanopores formed on Ti40Cu60 rib-
bons, which had a pore size of 25 nm in 0.03 M HF solution and
164.
[
23] R.M. Garrels, C.L. Christ, Solutions, Minerals and Equilibria, Harper & Row, New
York, 1965. pp. 216–222.
8
5 nm in 0.13 M HF solution are due to the high Cu content. The
significant increases in the surface diffusivity from the dealloying
temperatures accordingly formed the rougher nanostructures.
[24] M. Nakasawa, S. Matsuya, T. Shiraishi, M. Ohta, J. Dent. Res. 78 (1999) 1568–
572.
25] G.E. Evangelakis, G.C. Kallinteris, N.I. Papanicolaou, Surf. Sci. 394 (1997) 185–
91.
[26] F. Montalenti, R. Ferrando, Phys. Rev. B 59 (1999) 5881–5891.
27] G. Boisvert, L.J. Lewis, Phys. Rev. B 56 (1997) 7643–7655.
1
[
ꢀ1
The activation energy was estimated to be 40.3–78.1 kJ mol
,
1
which depended on the alloy composites and concentrations of
dealloying solutions. The long jump and cross-channel diffusion
[