Bi and Te thin films synthesized by galvanic displacement from acidic nitric baths

Article abstract of DOI:10.1016/j.electacta.2009.09.038

Bismuth (Bi) and tellurium (Te) thin films were formed by galvanic displacement of different sacrificial iron group thin films [i.e. nickel (Ni), cobalt (Co) and iron (Fe)] where the formation was systematically investigated by monitoring the change of op

Full text of DOI:10.1016/j.electacta.2009.09.038

Electrochimica Acta 55 (2010) 743–752
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Bi and Te thin films synthesized by galvanic displacement from acidic
nitric baths
a
a
b
b
Chong Hyun Chang , Youngwoo Rheem , Yong-Ho Choa , Dong Hyuk Shin ,
c,∗
a,∗
Deok-Yong Park , Nosang V. Myung
Department of Chemical and Environmental Engineering, University of California-Riverside, Riverside, CA 92521, USA
Division of Materials and Chemical Engineering, Hanyang University, Ansan 426-791, Republic of Korea
Department of Applied Materials Engineering, Hanbat National University, Daejeon 305-719, Republic of Korea
a
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 July 2009
Received in revised form 9 September 2009
Accepted 11 September 2009
Available online 16 September 2009
Bismuth (Bi) and tellurium (Te) thin films were formed by galvanic displacement of different sacrificial
iron group thin films [i.e. nickel (Ni), cobalt (Co) and iron (Fe)] where the formation was systemati-
cally investigated by monitoring the change of open circuit potential (OCP), surface morphology and
microstructure. The surface morphologies and crystal structures of galvanically displaced Bi or Te thin
films strongly depended on the type and thickness of the sacrificial materials. Continuous Bi thin films
were successfully deposited with the sacrificial Co. However, dendrites and nanoplatelets were formed
from the Ni and Fe thin films. Te thin films were synthesized with all the three sacrificial thin films.
Chemical dissolution rate of the sacrificial thin films and mixed potential strongly influenced formation
of Bi or Te thin films.
Keywords:
Galvanic displacement
Electroless deposition
Bismuth
Tellurium
© 2009 Elsevier Ltd. All rights reserved.
Thermoelectric materials
1
. Introduction
Bismuth (Bi) and tellurium (Te) are one of well-known materials
injection [8], evaporation or sputtering [9], electron-beam lithog-
raphy [2], low-temperature hydrothermal reduction [10], room
temperature aqueous chemical route [11], solvothermal method
[12], and vapor-phase deposition technique [13] were used to syn-
thesize these nanostructures.
used in thermoelectric devices, which convert thermal energy from
temperature gradient into electrical energy (the Seebeck effect),
or vice versa (the Peltier effect), without any moving parts [1].
Te with a p-type narrow band gap (direct band gap energy:
0.35 eV) and a highly anisotropic crystal structure has been studied
because of several potential applications in high-efficiency pho-
toconductors, thermoelectric and piezoelectric devices, electronic
and optoelectronic devices [14,15]. Te exhibits several interest-
ing chemical and physical properties such as catalytic activity,
piezoelectricity, thermoelectricity, non-linear optical response,
and photoconductivity [15,16]. Specially, trigonal tellurium (t-Te)
shows a highly anisotropic crystal structure which consists of heli-
cal chains of covalently bonded Te atoms and is bound together
through van der Waals interactions in a hexagonal lattice [14,16].
This anisotropic structure results in one-dimensional growth (e.g.
nanowire, nanorod, nanobelt or nanotube) [16].
◦
Bi with a low melting point (271 C) and a rhombohedral crystal
structure has been extensively studied because of its unusual elec-
tronic properties; small effective electron mass (0.001me), large
carrier mean free path (∼100 nm at 300 K and ∼400 ␮m at 4 K),
large Fermi wavelength ꢀF (40 nm at room temperature), highly
anisotropic Fermi surface, small band overlap (∼38 meV at 77 K)
and low charge-carrier density [2,3]. Most of research works for
Bi have been focused on fabrication methods, electronic prop-
erties, thermoelectric properties, and its applications. One- or
two-dimensional Bi nanostructures (i.e. nanowires, nanotubes or
thin films) have been recently used to investigate quantum con-
finement [2], finite-size effect [4], magnetoresistance (MR) effect
[
5,6], and thermoelectric effect [7] due to the potential applications
In addition, various functional materials can be synthesized
using the reaction between tellurium and other elements (Bi [1],
Cd [17], Pb [18–20] and Sb [21]). Most of research works for Te
have been focused on synthesis of nanostructures and thin films by
using the various techniques including refluxing [16], microwave-
assisted synthesis [22], thermal decomposition [23], solvothermal
or hydrothermal methods [24], biomolecule-assisted routes [25]
and chemical vapor deposition [14,26].
in thermoelectric devices, photonics and optoelectronics. Several
different methods such as electrodeposition [3,5], high-pressure
∗ Corresponding author.
E-mail addresses: dypark@hanbat.ac.kr (D.-Y. Park),
myung@engr.ucr.edu (N.V. Myung).
0
013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2009.09.038

7
44
C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
Table 1
been utilized to synthesize less common metals and semiconduct-
ing materials, although extensive studies for Bi, Te and their alloys
were investigated [1–7,14–16]. In our prior work, we demonstrated
the feasibility of synthesizing of BixTey nanotubes and thin films
by galvanic displacement [45,46]. However, there is a lack of sys-
tematic studies on the formation of pure Bi and Te from sacrificial
metals.
Bath compositions and operation conditions for galvanic displacement reactions to
synthesize Bi and Te thin films.
◦
Composition
Electrolyte
Temperature ( C)
Agitation
10 mM Bi3 + 1 M HNO3
+
Bi
Te
RT
RT
No
No
10 mM HTeO2+ + 1 M HNO3
In this paper, we conducted systematic investigation to monitor
the formation of Bi and Te by galvanic displacement of the sacrificial
iron group metallic films (Fe, Co and Ni). Electrochemical technique
such as open circuit potential (OCP) measurement was utilized to
monitor the process. SEM and XRD were utilized to observe the
surface morphology and microstructure.
Among different techniques to synthesize Bi, Te and its alloys,
electrochemical processes such as electrodeposition, electroless
(
autocatalytic) deposition and galvanic displacement (also called
as immersion plating or cementation) are considered as one of the
promising processes due to several advantages over vacuum pro-
cesses [27]. Galvanic displacement among other electrochemical
methods can be considered as a simple and fast process to deposit
Bi and Te thin films on the substrate and to tailor the morphol-
ogy, crystal structure and composition by varying experimental
conditions (e.g. sacrificial material, temperature, composition of
electrolyte). Also, galvanic displacement can offer the advantage of
selective deposition because reduction of metal ions in electrolyte
is coupled with oxidation and dissolution of the sacrificial material
on substrate. For example, galvanic displacement has been used to
selectively deposit Cu and other metals onto Si for applications such
as integrated circuits, microelectromechanical systems (MEMS),
surface-enhanced Raman spectroscopy, catalysis and microchan-
nel chemical reactors [28–30]. Recently, galvanic displacement has
been utilized to deposit pure metals and binary alloys by several
research groups [31–46] including noble (i.e. Au, Ag, Pt and Pd)
2. Experimental
In order to synthesize Bi and Te thin films by galvanic displace-
ment, three different sacrificial iron group (i.e. Ni, Co and Fe) thin
films with approximately 3 ␮m thickness were electrodeposited
on Pt-coated Ti/SiO /Si substrates from chloride baths where
2
the substrates consist of Pt(200 nm)/Ti(20 nm)/SiO (100 nm)/Si. Pt
2
(platinum) and Ti (titanium) were formed on SiO /Si substrate by
2
electron-beam evaporation process. Pt was used as a seed layer and
Ti was used as an adhesion layer between Pt and SiO2 layer. Elec-
trodeposition was conducted using a Princeton Applied Research
−
2
Potentiostat (VMP2) at current density of 5 mA cm and room tem-
perature without stirring. Saturated calomel electrode (SCE) and
Pt-coated Ti anode were used as a reference and a counter electrode,
respectively. Electrolyte compositions and operation conditions for
electrodeposition of the sacrificial thin films can be found else-
where [46]. Calcium chloride was used as a supporting electrolyte
[
32–37] and less-noble (i.e. Ni, Cu and Pb) [38–42] metals. Xia and
co-workers have also synthesized binary alloys including Au–Ag,
Pd–Ag and Pt–Ag [43,44]. However, galvanic displacement has not
Fig. 1. Open circuit potential (OCP) as a function of galvanic reaction time; the sacrificial materials (Ni, Co and Fe) are immersed into nitric acid baths containing (a) 10 mM
Bi , (b) initial stages of OCP curves in (a) within 3 min, (c) 10 mM HTeO2+, and (d) initial stages of OCP curves in (c) within 5 min. Solid, dash, and dotted lines represent Fe,
3+
Ni, and Co, respectively.

C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
745
and l-ascorbic acid was used to prevent oxidation of Fe2+ in the
electrolyte for electrodeposition of Fe thin film. Solution pH was
controlled to be 3 with HCl or NaOH.
(step II) as soon as the sacrificial thin film is completely covered
by Bi thin film. After finishing galvanic displacement reaction, OCP
curve reached to be another steady-state value at near 0 V vs SCE
(step III). Reaction time for galvanic displacement with the sacri-
ficial Co thin film was measured to be approximately 10 min and
less than 1 min for the sacrificial Fe and Ni thin films. Absence of
clear three steps in OCP curve for the sacrificial Ni thin films may be
attributed to much slower dissolution rate of the sacrificial Ni layer
than reduction rate of Bi3+ ions in the electrolyte. However, much
faster dissolution rate of the sacrificial Fe layer than reduction rate
of Bi3+ ions in the electrolyte is responsible for very short step I.
Effects of extremely slow dissolution rate of the sacrificial Ni layer
and very fast dissolution rate of the sacrificial Fe layer on forma-
tion of Bi thin film will be discussed further with SEM microscopic
images in details. As the surface coverage of Bi or Te increased with
the reduction of sacrificial film surface, it shifted the open circuit
potential value toward more positive value during step II. After Bi
thin films completely covered the surface, the potential reached a
steady state mixed potential between Bi/Bi+3 and dissolved oxygen
oxidation. Fig. 1(c) shows OCP curves of HTeO2+ (with a concentra-
tion of 10 mM) ion with the sacrificial Ni, Co and Fe thin films for
galvanic displacement during 60 min. Clear step I in OCP curve for
all the sacrificial thin films (Co, Ni and Fe) were observed. Reaction
time for galvanic displacement with the sacrificial Fe thin film was
measured to be approximately 15 min. However, reaction times for
the sacrificial Ni and Co thin films were measured to be greater than
60 min. From comparison of open circuit potential values at step III
in OCP curves of Bi3+ [∼−0.03 V for the sacrificial Ni, Co and Fe thin
For galvanic displacement, electrodeposited Ni, Co and Fe thin
films were immersed into the bismuth or tellurium containing elec-
trolytes for 1 h (or 3 h). Solution pH for galvanic displacement was
fixed at 1. OCP during galvanic displacement reaction was mea-
sured in a three-electrode cell to monitor the effect of the sacrificial
materials. Electrolyte compositions for galvanic displacement are
listed in Table 1. Bismuth nitrate [Bi(NO ) ·5H O], tellurium oxide
3
3
2
(
TeO , 99.99%; Alfa Aesar) and nitric acid (HNO ) were used to make
2 3
the electrolytes for galvanic displacement. Concentrations of metal
ion sources (Bi for Bi thin film or HTeO2+ for Te thin film) were
3+
fixed at 10 mM and HNO3 at 1 M for comparison.
Surface morphology and film composition of the thin films
obtained from electrodeposition and galvanic displacement were
examined using a scanning electron microscope (SEM) (model:
XL30-FEG, Phillips) and an energy-dispersive spectroscopy (EDS)
(
model: EDAX, Phoenix). An X-ray diffractometer (XRD) (model:
D8 advance diffractometer, Bruker) with Cu K␣ radiation (operat-
ing at 40 kV) was used for the identification of the phase in the thin
◦
films. The conditions of XRD were a scanning range of 20–80 with
◦
0.02 increments and a 0.5 s collection time per increment.
3
. Results and discussion
Open circuit potentials (OCP) as a function of reaction time dur-
ing galvanic displacement were measured to monitor the reactions
between metal ions (Bi3 and HTeO2+) in electrolytes and the sac-
rificial metal films (Ni, Co and Fe) as shown in Fig. 1. The sacrificial
thin films are galvanically displaced to form Bi or Te thin films
because there exists the difference of redox potentials between
+
+
films in Fig. 1(a)] and HTeO2 [∼0.3 V for the sacrificial Fe, ∼0 V
for the sacrificial Ni, and ∼−0.1 V for the sacrificial Co thin films
in Fig. 1(c)] ions, it can be suggested that the sacrificial Ni and Co
thin films were not completely displaced by Te during galvanic dis-
placement reaction. In order to completely displace the sacrificial
Ni, Co and Fe thin films with Bi or Te metals by galvanic displace-
ment, surface morphologies during galvanic displacement reaction
should be porous for metal ions (Bi or HTeO2+) in the electrolyte
metal ions (Bi or HTeO2+) in electrolyte and the sacrificial thin
3+
films (Ni, Co and Fe) as the following equations [1,20,45,47]:
Bi3 + 3e− → Bi(s)
+
E = 0.317 V vs SHE
0
(1)
3+
3
+
+
HTeO2 + 3H+ + 4e− → Te(s) + 2H O
+
0
to reach to the sacrificial layers. If the metal ions (Bi or HTeO2 )
in the electrolyte cannot reach and contact the sacrificial layers,
galvanic displacement cannot be continued. This will be discussed
further in Fig. 6. It was clearly observed that shape of OCP curve
or potential value after finishing galvanic displacement reaction
strongly depend on both type of the metal ion in the electrolyte
and that of the sacrificial materials.
E = 0.551 V vs NHE (2)
2
Ni2 + 2e− → Ni(s)
+
E = −0.257 V vs SHE
0
(3)
(4)
(5)
Co2 + 2e− → Co(s)
+
E = −0.28 V vs SHE
0
Fe2+ + 2e− → Fe(s)
0
E = −0.44 V vs SHE
Galvanic displacement occurs spontaneously because Bi or Te met-
als are more noble (easier to reduce) than the sacrificial materials
After finishing galvanic displacement reaction with the sacrifi-
cial metal thin films (Ni, Co and Fe) for 60 min, surface morphologies
of galvanically synthesized Bi were observed using SEM, as shown
in Fig. 2. Fig. 2(a) shows the surface morphology of electrode-
posited sacrificial Ni thin film from chloride bath before galvanic
displacement reaction [46]. Surface morphology of Bi galvanically
synthesized with the electrodeposited Ni thin film exhibited sev-
eral isolated particles [marked with red arrows in Fig. 2(b)], in the
absence of uniform thin film. Needle-like shape surface in Fig. 2(a)
and (b) comes from the electrodeposited Ni thin film. Those parti-
cles in Fig. 2(a) were verified to be Bi using EDS analysis. Therefore,
it can be concluded that dissolution (oxidation) rate of the elec-
trodeposited Ni thin film during galvanic displacement in nitric
acid solution is extremely slow. On the other hand, surface mor-
phology of the galvanically synthesized Bi thin film (Fig. 2(d)) with
the electrodeposited Co thin film as a sacrificial thin film exhibits
a granular structure with a polygon shape arrangement and some-
what similar to surface morphology of the electrodeposited Co thin
film (Fig. 2(c)). EDS analysis confirm the formation of Bi thin film.
Three-micron thick electrodeposited Fe film had dense and uni-
form deposit (Fig. 2(e)). However, Bi thin film was not formed by
galvanic displacement reaction (Fig. 2(f)). EDS also confirmed the
finding which might be attributed to complete chemical dissolution
(
Ni, Co and Fe) [45]. Fig. 1(a) shows three OCP curves of Bi3+ ions
for galvanic displacement with the sacrificial Ni, Co and Fe thin
films during 60 min. It was observed in Fig. 1(b) that OCP curves
started from negative potential values (−0.37 V for Fe < −0.15 V
for Co < −0.07 V for Ni) at early initial stage of galvanic displace-
ment reaction and the starting potential values in OCP curve
were matched to the sequence of standard redox potential values
(
−0.44 V vs. SHE for Fe < −0.28 V vs. SHE for Co < −0.257 V vs. SHE
for Ni). The negative potential at the early initial stage in OCP curve
may indicate the mixed potentials between reduction of Bi3 in
the electrolyte and oxidation of the sacrificial metal thin films. The
driving force for galvanic displacement reaction is the difference of
redox potentials between Bi3 in the electrolyte and the sacrificial
metal thin films [1,20,45]. After the early initial stage, OCP curve
for the sacrificial Ni thin film leveled off without showing step I.
However, OCP curve exhibited very short step I for the sacrificial
Fe thin film and clear step I for the sacrificial Co thin film, respec-
tively. Clear three steps were observed only for the sacrificial Co
thin film and not for the sacrificial Ni and Fe thin films. Existence
of step I during galvanic displacement reaction strongly depend on
type of the sacrificial thin film. OCP curve sharply increase again
+
+

7
46
C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
Fig. 2. Surface morphologies of sacrificial (a) Ni, (c) Co and (e) Fe thin films electrodeposited from chloride baths [46]. Surface morphologies of Bi thin films deposited by
galvanic displacement reaction for 60 min with the sacrificial (b) Ni, (d) Co and (f) Fe thin films.
of Fe thin film prior to galvanic displacement. To further investigate
the process, the effect of film thickness of Fe on the formation of Bi
was conducted which will be discussed in later section. Also, only
Bi particles were obtained using the sacrificial Ni thin film during
galvanic displacement reaction. The exact mechanism leading to
those phenomena in Fig. 2(b and f) has not yet been identified. It is
obvious that formation of the galvanically synthesized Bi thin films
with the sacrificial Ni, Co and Fe thin films strongly depend on type
of the sacrificial thin films. From SEM observation in Fig. 2, it is
clear that the galvanically synthesized Bi particles resulted from
extremely slow dissolution rate of the sacrificial Ni thin films. Also,
very fast dissolution rate of the sacrificial Fe thin films resulted in
no formation of Bi thin film by galvanic displacement of thin Fe
films (approx. 3 ␮m).
ment with the electrodeposited Ni (Fig. 3(a)), Co (Fig. 3(b)) and Fe
(Fig. 3(c)) thin films showed completely different surface morphol-
ogy compared to the surface morphologies of the electrodeposited
sacrificial Ni (Fig. 2(a)), Co (Fig. 2(c)), and Fe (Fig. 2(e)) thin films
[46]. Surface morphology of the galvanically synthesized Te thin
film with the electrodeposited Ni thin film exhibited a well-defined
polygon shape as shown in Fig. 3(a). On the other hand, surface
morphology of the galvanically synthesized Te thin film with the
electrodeposited Co thin film in Fig. 3(b) showed a flower-like gran-
ular structure with very fine polygon shapes. Also, galvanically
synthesized Te thin film with the electrodeposited Fe thin film had
very fine polygon shape as shown in Fig. 3(c). It is clear that surface
morphologies of galvanically synthesized Te thin films with the sac-
rificial Ni, Co and Fe thin films were not depended on those of the
sacrificial thin films (Fig. 2(a) for Ni, Fig. 2(c) for Co and Fig. 2(e) for
Fe).
Surface morphology of galvanically synthesized Te thin films
with the sacrificial metal thin films (Ni, Co and Fe) for 60 min was
observed using SEM, as shown in Fig. 3. Te thin films were suc-
cessfully deposited with the three sacrificial thin films by galvanic
displacement reaction. Te thin films deposited by galvanic displace-
Fig. 4 shows XRD patterns of Bi thin films deposited by galvanic
displacement with the different sacrificial Ni, Co, and Fe thin films.
Bi thin film was clearly formed by galvanic displacement only with

C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
747
analysis of Fig. 4(c) was observed by galvanic displacement with the
sacrificial Fe thin film. Further analysis was conducted using EDS
because it is not clear whether Bi thin film was formed by galvanic
displacement with the sacrificial Fe thin film, or not. Neither Bi nor
Fe was not detected at all in EDS analysis. This result indicates that
Bi thin film was not formed during galvanic displacement reaction
with 3 ␮m thick Fe thin film.
Fig. 5 shows XRD patterns of Te thin films obtained by gal-
vanic displacement with the different sacrificial metals such as Ni
(Fig. 5(a)), Co (Fig. 5(b)), and Fe (Fig. 5(c)). From XRD results, it is
obvious that Te thin films were successfully synthesized by gal-
vanic displacement with the sacrificial Ni and Co films, even with
the sacrificial Fe thin film, compared to XRD results of Bi thin films
(Fig. 4(a and c)). The galvanically synthesized Te thin films with
the sacrificial Ni, Co and Fe thin films show a hexagonal structure
(JCPDS No. 36-1452).
It is difficult to clearly observe total film thickness of the gal-
vanically deposited Bi thin film using cross-sectional SEM image
because of the formation of Bi particles instead of Bi thin film as
shown in Fig. 2(b) and no formation of Bi thin film as shown in
Fig. 2(f). Therefore, it is meaningless to compare thickness of Bi thin
film obtained by galvanic displacement with the three different sac-
rificial thin films. However, Te thin films were successfully formed
during galvanic displacement with the three different sacrificial
thin films. Thicknesses of Te thin films galvanically synthesized
from nitric acid baths for 60 min and the sacrificial Ni, Co and Fe
thin films electrodeposited from chloride baths could be clearly
observed using cross-sectional SEM image as shown in Figs. 6–8.
Nickel thin film with a thickness of ∼3 ␮m electrodeposited from
chloride bath has a typical dimple structure (sponge-like structure)
of ductile metal as shown in Fig. 6(a). After galvanic displacement
for 60 min, Ni thin film was partially replaced by Te thin film with
a polygon shape and thickness of Te thin film was measured to be
∼
0.8 ␮m thick. Te thin film exhibited columnar growth (Fig. 6(b and
c)). Columnar growth is commonly observed in the thin/thick films
fabricated by other electrochemical processes such as electrodepo-
sition [27,48] or electroless deposition [49–52]. Fig. 6(b) exhibits
Te thin film with ∼0.8 ␮m thick after galvanic displacement reac-
tion for 60 min with the sacrificial Ni thin film. The sacrificial Ni
thin film with ∼2.8 ␮m thick still remained as shown in Fig. 6(b).
This result may come from much slower dissolution rate of the
sacrificial Ni thin film compared to reduction rate of HTe O ion in
+
2
nitric acid solution and is in good agreement with the result of OCP
curve for the sacrificial Ni thin film (Fig. 1(c)). That is, OCP value
for the galvanically deposited Te thin film with the sacrificial Ni
thin film in Fig. 1(c) was measured to be −0.04 V and indicated that
the sacrificial Ni thin film was partially displaced by galvanic dis-
placement. Te thin film with ∼0.8 ␮m thick was formed by galvanic
displacement the sacrificial Ni thin film for 1 h. After galvanic dis-
placement reaction for 3 h, thickness of Te thin film was measured
to be ∼2.5 ␮m as shown in Fig. 6(c). It was observed that growth
rate of Te thin film was linearly proportional to time of galvanic
displacement reaction.
Fig. 3. Surface morphologies of galvanically displaced Te thin films with sacrificial
a) Ni, (b) Co and (c) Fe thin films.
(
the sacrificial Co thin films as shown in Fig. 4(b). Bi thin film in
Fig. 4(b) exhibits a typical XRD pattern of Bi with a dominant (0 1 2)
planes and a rhombohedral structure (JCPDS No. 05-0519). How-
ever, only a very weak peak of Bi was observed during galvanic
displacement with the sacrificial Ni thin films as shown in Fig. 4(a).
The weak Bi peak resulted from the formation of Bi particles instead
of the formation of continuous Bi thin film. This XRD result is in
good agreement with SEM observation of Bi particles in Fig. 2(b).
Therefore, it is clear from SEM, EDS and XRD analyses that only
some amounts of Bi particles with the sacrificial Ni thin film were
deposited by galvanic displacement due to extremely slow dissolu-
tion rate of Ni thin films into the electrolyte and low mixed potential
to drive the reaction. On the other hand, no peak for Bi from XRD
Fig. 7(a) shows the electrodeposited sacrificial Co thin film with
∼3 ␮m thick. After galvanic displacement reaction with the sacri-
ficial Co thin film for 1 h, thickness of Te thin film was ∼1.8 ␮m
thick as shown in Fig. 7(b). Te thin film was deposited galvani-
cally in powdery form and showed very rough surface as shown
in Fig. 7(d). After galvanic displacement for 1 h, significant amount
of the sacrificial Co thin film in Fig. 7(b) still remained. This result
may come from much slower dissolution rate of the sacrificial Co
+
thin film compared to reduction rate of HTe O ion and is in good
2
agreement with result of OCP curve for the sacrificial Co thin film
(Fig. 1(c)).
On the other hand, cross-sectional SEM image for the electrode-
posited Fe thin film with a thickness of ∼3 ␮m is shown in Fig. 8(a).

7
48
C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
Fig. 4. XRD patterns of the galvanically displaced Bi thin films with sacrificial (a) Ni, (b) Co and (c) Fe thin films. (S: substrate – Pt and Si).
Fig. 5. XRD patterns of galvanically displaced Te thin films with sacrificial (a) Ni, (b) Co, and (c) Fe thin films. (S: substrate – Pt and Si).

C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
749
Fig. 6. Cross-sectional SEM images of Te thin films deposited with the sacrificial Ni thin film by galvanic displacement; (a) the sacrificial Ni thin films electrodeposited from
chloride bath, (b) Te thin film deposited by galvanic displacement with the sacrificial Ni thin film for 1 h and (c) Te thin film deposited by galvanic displacement with the
sacrificial Ni thin film for 3 h.
Fig. 7. Cross-sectional SEM images of Te thin films deposited with the sacrificial Co thin film by galvanic displacement; (a) the sacrificial Co thin films electrodeposited from
chloride bath, (b) Te thin film deposited by galvanic displacement with the sacrificial Co thin film for 1 h, and (c) high magnification of Te thin film marked as region “A” in
(
b), (d) tilting top view of Te thin film in (b).

7
50
C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
Fig. 8. Cross-sectional SEM images of Te thin films deposited with increasing thickness of the sacrificial Fe thin film by galvanic displacement; (a) the sacrificial Fe thin film
3 ␮m thick) electrodeposited from chloride bath, (b) Te thin film deposited by galvanic displacement with the sacrificial Fe thin film (3 ␮m thick) for 1 h, (c) tilted top view
(
of Te thin film of (b), (d) Te thin film deposited by galvanic displacement with the sacrificial Fe thin film (10 ␮m thick) for 1 h, (e) tilted top view of Te thin film of (d), (f) Te
thin film deposited by galvanic displacement with the sacrificial Fe thin film (20 ␮m thick) for 1 h, (g) top view of Te thin film in (f), and (h) XRD patterns of Te thin film for
(
d) and (f).
After galvanic displacement reaction for 1 h, Te thin films with tri-
angular and spine-shape was formed. Thickness of Te thin film in
Fig. 8(b) was measured to be ∼0.4 ␮m. Compared to the sacrificial
was observed. This result is in good agreement with the result of
OCP curve for the sacrificial Fe thin film in Fig. 1(c); time to reach
∼0.3 V, at which Te thin film was completely displaced by galvanic
displacement reaction with the sacrificial Fe thin film, was ∼15 min.
Dissolution rates of the sacrificial Ni, Co, and Fe thin films in nitric
Ni and Co layers, much faster dissolution rate of the sacrificial Fe
+
thin film than reduction rate of HTe O ion in nitric acid solution
2

C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
751
Fig. 9. Surface morphologies of galvanically displaced Bi from sacrificial Fe thin films with a thickness of (a) 10 ␮m, low magnification. (b) 10 ␮m, high magnification (c)
0 ␮m, low magnification and (d) 20 ␮m, high magnification for 1 h.
2
acid solution (pH 1 in this study) were measured to be in order of
Co ≈ Ni < Fe from galvanic displacement reaction. Dissolution rate
of the sacrificial layer may strongly depend on surface morpholo-
gies of Te deposited galvanically on the sacrificial layers at the initial
stage of galvanic displacement; after covering initially the sacrifi-
cial layer by galvanic displacement, more porous surface of Te thin
film on the sacrificial layer result in faster dissolution rate of the
sacrificial layer. It was also observed that surface morphologies of
Te thin films with both sacrificial Ni (Fig. 3(a) or Fig. 6(b)) and Co
From OCP measurements (Fig. 1) and SEM results (Figs. 2 and 9
for Bi thin film, and Figs. 6–8 for Te thin films), it was obvious that
the deposited thickness of galvanically synthesized Bi or Te thin
films strongly depends on type of the sacrificial thin films. Also,
it can be suggested that faster dissolution rate of the sacrificial Fe
thin film than reduction rate of Bi ions in galvanic displacement for
Bi thin film resulted in very short step I in OCP curve as shown in
Fig. 1(a). Furthermore, extremely slow dissolution rate of the sac-
rificial Ni thin film may caused absence of step I in OCP curve. On
the other hand, clear step I in OCP curve was observed in galvanic
displacement of Te thin film due to relatively slow dissolution rate
of the sacrificial Co and Ni than reduction rate of Te ions. There-
fore, galvanic displacement reactions to fabricate Bi or Te thin films
could be successfully monitored by OCP measurements. Also, it
is clear from the observation of cross section morphologies that
type of sacrificial materials has a strong influence on the structure
of Te thin film. In other words, microstructure of the galvanically
deposited thin film could be adjusted by the type of the sacrificial
materials.
(
Fig. 3(b) or Fig. 7(d)) layers were formed in more compact than
that of Te with the sacrificial Fe layer (Fig. 3(c) or Fig. 8(b)). In order
to investigate the influence of the sacrificial Fe layer thickness on
surface morphology and thickness of Te thin film, thickness of the
sacrificial Fe layer was changed from 3 to 10 and 20 ␮m. Thicker and
more compact Te thin films by galvanic displacement were formed
with increasing thickness of the sacrificial Fe layer as shown in
Fig. 8(b, d and f). From XRD analysis in Fig. 8(h), (1 0 1) peak (JCPDS
No. 36-1452) was more clearly observed compared to XRD result
in Fig. 5(c). Also, it can be suggest from XRD results that Te thin
film may be very close to amorphous phase rather than crystalline
phase.
4. Conclusions
To understand the effects of Fe layer thickness on the forma-
tion of Bi similar to Te, the thickness was varied from 3 to 10 and
Bi and Te thin films were synthesized by galvanic displacement
with three different types of the sacrificial thin films (Ni, Co and Fe).
The sacrificial Ni, Co and Fe thin films were electrodeposited on
2
(
0 ␮m. Although Bi was not deposited from ∼3 ␮m thick Fe film
Fig. 2(f)), the dendrites and hexagonal nanoplatelets of Bi were
deposited via galvanic displacement with increasing film thickness
Fig. 9). The morphology of the deposits was strongly influenced by
the sacrificial film thickness where dendrites were observed with
10 ␮m thick Fe thin film [Fig. 9(a–b)]. Approximately 20 ␮m thick
Pt(nm)/Ti(nm)/SiO (nm)/Si substrate at room temperature with-
2
(
out stirring. Galvanic displacement reactions to fabricate Bi and Te
thin films could be successfully monitored by OCP measurements.
It was clearly observed that a good agreement between OCP mea-
surement, XRD analysis and SEM observation for the galvanically
deposited Bi and Te thin films is existed. Surface morphologies and
crystal structures of the as-synthesized thin films by galvanic dis-
placement strongly depend on the type of the sacrificial materials.
∼
Fe thin films led to the formation of hexagonal Bi nanoplatelets on
the dendrites [Fig. 9(c–d)]. The X-ray diffraction patterns confirmed
the formation of Bi with preferred orientation in (0 1 2) and (1 0 4)
planes (JCPDS No. 05-0519) (Data not shown).

7
52
C.H. Chang et al. / Electrochimica Acta 55 (2010) 743–752
Bi thin films were successfully deposited with the sacrificial
[14] P. Mohanty, T. Kang, B. Kim, J. Park, J. Phys. Chem. B 110 (2006) 791.
[
[
[
15] G. Xi, Y. Peng, W. Yu, Q. Tian, Cryst. Growth Des. 5 (2005) 325.
16] B. Mayers, Y. Xia, J. Mater. Chem. 12 (2002) 1875.
17] K. Arai, S. Hagiwara, S. Takayama, K. Murase, T. Hirato, Y. Awakura, Electrochem.
Commun. 8 (2006) 605.
Co thin films. However, only small amounts of Bi particles were
deposited by galvanic displacement with the 3 ␮m thick sacrifi-
cial Ni thin films due to slow dissolution rate. Similar to nickel,
Bi was not galvanically displaced on the 3 ␮m thick Fe thin films.
By increasing the Fe film thickness, dendrites with hexagonal
nanoplatelets of Bi were deposited.
Te thin films were successfully deposited using the sacrificial Ni,
Co, and Fe thin films during galvanic displacement. Thicker sacri-
ficial Fe films led to the formation of thicker and more compact of
Te. Moreover, Te thin films exhibited different surface morpholo-
gies compared to the surface morphologies of the sacrificial thin
films. The galvanically deposited Bi and Te thin films were suc-
cessfully investigated using open circuit potential (OCP), surface
morphology and microstructure.
[
18] M.R. Khairulaev, D.Kh. Dadaev, M.M. Rasulov, Russ. J. Inorg. Chem. 51 (2006)
1706.
[
19] G. Henshaw, I.P. Parkin, G.A. Shaw, J. Chem. Soc. Dalton Trans. (1997) 231.
[20] F. Xiao, B.Y. Yoo, M.A. Ryan, K.-H. Lee, N.V. Myung, Electrochim. Acta 52 (2006)
1101.
[
21] R. Venkatasubramanian, T. Colpitts, E. Watko, M. Lamvik, N. El-Masry, J. Cryst.
Growth 170 (1997) 817.
[22] Y.J. Zhu, W.W. Wang, R.J. Qi, X.L. Hu, Angew. Chem. Int. Ed. 43 (2004) 1410.
[23] Z. Tang, Y. Wang, K. Sun, N.A. Kotov, Adv. Mater. 17 (2005) 358.
[
[
[
24] X.L. Li, G.H. Cao, C.M. Feng, Y.D. Li, J. Mater. Chem. 14 (2004) 244.
25] Q. Lu, F. Gao, S. Komarneni, Adv. Mater. 16 (2004) 1629.
26] B. Geng, Y. Lin, X. Peng, G. Meng, L. Zhang, Nanotechnology 14 (2003) 983.
[27] D.-Y. Park, N.V. Myung, M. Schwartz, K. Nobe, Electrochim. Acta 47 (2002) 2893.
[28] C. Carraro, R. Maboudian, L. Magagnin, Surf. Sci. Rep. 62 (2007) 499.
[29] C. Carraro, L. Magagnin, R. Maboudian, Electrochim. Acta 47 (2002) 2583.
[30] C.-H. Wang, D.-C. Sun, X.-H. Xia, Nanotechnology 17 (2006) 651.
[31] V. Bansal, H. Jani, J. Du Plessis, P.J. Coloe, S.K. Bhargava, Adv. Mater. 20 (2008)
Acknowledgements
717.
[
32] Y.G. Sun, B.T. Mayers, Y.N. Xia, Nano Lett. 2 (2002) 481.
This research work was supported by the National Research
Foundation of Korea (NRF) funded by the Ministry of Education,
Science and Technology (Grant No. 2009-0082820), and Hanbat
National University, Republic of Korea.
[33] Y.G. Sun, B. Mayers, Y.N. Xia, Adv. Mater. 15 (2003) 641.
[
[
34] L.A. Porter, H.C. Choi, A.E. Ribbe, J.M. Buriak, Nano Lett. 2 (2002) 1067.
35] L.A. Porter, H.C. Choi, J.M. Schmeltzer, A.E. Ribbe, L.C.C. Elliott, J.M. Buriak, Nano
Lett. 2 (2002) 1369.
[36] M.R.H. Nezhad, M. Aizawa, L.A. Porter Jr., A.E. Ribbe, J.M. Buriak, Small 1 (2005)
076.
37] M. Aizawa, A.M. Cooper, M. Malac, J.M. Buriak, Nano Lett. 5 (2005) 815.
[38] O.M.R. Chyan, J.-J. Chen, H.Y. Chien, J. Sees, L. Hall, J. Electrochem. Soc. 143
1996) 92.
1
[
References
(
[
[
[
[
1] M.S. Mart ␫´ n-González, A.L. Prieto, R. Gronsky, T. Sands, A.M. Stacy, J. Elec-
trochem. Soc. 149 (2002) C546.
2] S.B. Cronin, Y.-M. Lin, O. Rabin, M.R. Black, J.Y. Ying, M.S. Dresselhaus, P.L. Gai,
J.-P. Minet, J.-P. Issi, Nanotechnology 13 (2002) 653.
3] T.W. Cornelius, J. Brötz, N. Chtanko, D. Dobrev, G. Miehe, R. Neumann, M.E.
Toimil Molares, Nanotechnology 16 (2005) S246.
4] M. Lu, R.J. Zieve, A. van Hulst, H.M. Jaeger, T.F. Rosenbaum, S. Radelaar, Phys.
Rev. B 53 (1996) 1609.
[39] J.S. Jeon, S. Raghavan, H.G. Parks, J.K. Lowell, I. Ali, J. Electrochem. Soc. 143 (1996)
2870.
[40] D. Niwa, T. Homma, T. Osaka, J. Phys. Chem. B 108 (2004) 9900.
[41] D. Niwa, T. Homma, T. Osaka, J. Electrochem. Soc 152 (2005) C54.
[42] C.Y. Wang, M.Y. Lu, H.C. Chen, L.J. Chen, J. Phys. Chem. C 111 (2007) 6215.
[43] Y. Sun, Y. Xia, Adv. Mater. 16 (2004) 264.
[44] J. Chen, B. Wiley, J. McLellan, Y. Xiong, Z.-Y. Li, Y. Xia, Nano Lett. 5 (2005) 2058.
[45] F. Xiao, B.Y. Yoo, K.-H. Lee, N.V. Myung, J. Am. Chem. Soc. 129 (2007) 10068.
[46] C.H. Chang, Y.W. Rheem, Y.-H. Choa, D.-Y. Park, N.V. Myung, Galvanic displace-
ment of BixTey thin films from sacrificial iron group thin films Electrochim.
Acta, in press.
[
[
[
[
5] F.Y. Yang, K. Liu, C.L. Chien, P.C. Searson, Phys. Rev. Lett. 82 (1999) 3328.
6] K. Liu, C.L. Chien, P.C. Searson, K. Yu-Zhang, Appl. Phys. Lett. 73 (1998) 1436.
7] A. Boukai, K. Xu, J.R. Heath, Adv. Mater. 18 (2006) 864.
8] Z. Zhang, X. Sun, M.S. Dresselhaus, J.Y. Ying, J.P. Heremans, Appl. Phys. Lett. 73
[47] http://www.webelements.com.
(
1998) 1589.
[48] M. Schwartz, in: R.F. Bunshah (Ed.), Handbook of Deposition Technologies for
Films and Coatings, Second ed., Noyes Publications, Park Ridge, NJ, 1994, p. 524.
[49] Electroless plating: fundamentals and applications, in: G.O. Mallory, J.B. Hajdu
(Eds.), American Electroplaters and Surface Finishers Society, Orlando, Florida,
1990, p. 111, 476.
[50] T. Homma, Y. Kita, T. Osaka, J. Electrochem. Soc. 147 (2000) 160.
[51] T. Homma, K. Inoue, H. Asai, K. Ohrui, T. Osaka, IEEE Trans. Magn. 27 (1991)
4909.
[
9] N. Garcia, Y.H. Kao, M. Strongin, Phys. Rev. B 5 (1972) 2029.
10] Y. Li, J. Wang, Z. Deng, Y. Wu, X. Sun, D. Yu, P. Yang, J. Am. Chem. Soc. 123 (2001)
[
9
904.
[
[
11] B. Yang, C. Li, H. Hu, X. Yang, Q. Li, Y. Qian, Eur. J. Inorg. Chem. 2003 (2003) 3699.
12] X.-Y. Liu, J.-H. Zeng, S.-Y. Zhang, R.-B. Zheng, X.-M. Liu, Y.-T. Qian, Chem. Phys.
Lett. 374 (2003) 348.
[
13] J. Heremans, C.M. Thrush, Y.-M. Lin, S. Cronin, Z. Zhang, M.S. Dresselhaus, J.F.
Mansfield, Phys. Rev. B 61 (2000) 2921.
[52] A. Brenner, D.W. Couch, E.K. Williams, J. Res. Nat. Bur. Stand. 44 (1950) 109.