Synthesis of Cu-Pd alloy thin films by co-electrodeposition

Article abstract of DOI:10.1016/j.electacta.2011.05.052

This paper presents results on the synthesis of Cu-Pd alloy thin films on Ti substrates by co-electrodeposition of Pd and Cu from nitrate-base electrolytic baths. The deposition rates of Cu and Pd were determined by Electrochemical Quartz Crystal Microbalance as a function of the electrode potential and Cu⁺² and Pd⁺² concentrations. It is shown that electrodeposition of copper and palladium occurs simultaneously at -0.50 V vs. SCE and that Cu-Pd thin films over the entire composition range were obtained by changing the composition of the solution. X-ray diffraction analyses indicated that these films have a nanocrystalline single-phase face-centered cubic structure and scanning electron microscopy analyses showed that potentiostatically deposited films are rough and porous, which is appropriate for electrocatalysis applications. In an attempt to get denser deposits as required for H₂ purification applications, pulsed potential co-electrodeposition was performed and the effect of the deposition conditions on the roughness of the films was assessed by double layer capacitance measurements. It was shown that smooth Cu-Pd films (with R_f value as low as 8, as opposed to more than 120 for films prepared in the potentiostatic mode) could be obtained with a proper choice on the deposition conditions.

Full text of DOI:10.1016/j.electacta.2011.05.052

Electrochimica Acta 56 (2011) 7397–7403
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Synthesis of Cu–Pd alloy thin films by co-electrodeposition
Morgan Allemanda, Manuel. H. Martina, David Reytera, Lionel Rouéa, Daniel Guaya,∗
,
Carmen Andreib, Gianluigi A. Bottonb
a INRS Énergie, Matériaux et Télécommunications, 1650 Blvd. Lionel Boulet, Varennes (Qc), J3X 1S2, Canada
b Department of Materials Science and Engineering, Brockhouse Institute for Material Research, Canadian Centre for Electron Microscopy,
McMaster University, Hamilton, Ontario L8S 4L7, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
This paper presents results on the synthesis of Cu–Pd alloy thin films on Ti substrates by co-
electrodeposition of Pd and Cu from nitrate-base electrolytic baths. The deposition rates of Cu and Pd were
determined by Electrochemical Quartz Crystal Microbalance as a function of the electrode potential and
Cu+2 and Pd+2 concentrations. It is shown that electrodeposition of copper and palladium occurs simul-
taneously at −0.50 V vs. SCE and that Cu–Pd thin films over the entire composition range were obtained
by changing the composition of the solution. X-ray diffraction analyses indicated that these films have
a nanocrystalline single-phase face-centered cubic structure and scanning electron microscopy analyses
showed that potentiostatically deposited films are rough and porous, which is appropriate for electro-
catalysis applications. In an attempt to get denser deposits as required for H2 purification applications,
pulsed potential co-electrodeposition was performed and the effect of the deposition conditions on the
roughness of the films was assessed by double layer capacitance measurements. It was shown that smooth
Cu–Pd films (with Rf value as low as 8, as opposed to more than 120 for films prepared in the potentiostatic
mode) could be obtained with a proper choice on the deposition conditions.
Received 31 January 2011
Received in revised form 13 May 2011
Accepted 13 May 2011
Available online 21 June 2011
Keywords:
Palladium–copper alloy
Potentiostatic electrodeposition
Potential pulse electrodeposition
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
less plating through two main approaches. The first reports in this
domain [8,10–14,25,26] have dealt with sequential deposition of
Over the last few years, copper–palladium alloys have received
much attention because of their widespread potential use in
many electrocatalytic applications such as the reduction of oxy-
gen [1,2] and nitrate [3–6] and the oxidation of hydrazine [7].
Cu–Pd alloys have also demonstrated promising performance in
the field of hydrogen purification, since they exhibit high permit-
tivity coupled with high resistance to hydrogen embrittlement and
sulfur-poisoning [8–14].
In addition to studies using commercially available thick Cu–Pd
foil prepared by melting copper and palladium [15–18], numer-
ous papers have been published dealing with the deposition of
bimetallic/alloy copper–palladium coatings by a variety of tech-
niques. For instance, sputtering, chemical vapor deposition and
pulsed laser techniques [2,19–24] were used to prepare thin films
of Cu–Pd. Although these techniques are quite useful for laboratory
investigations, they can only be brought to large scale and vol-
ume production with difficulty. In contrast, electroless plating and
electrodeposition are two preparation techniques that offer several
interesting options for small (laboratory investigations) and large
scale productions. Thus, Cu–Pd thin films were prepared by electro-
copper and palladium, followed by annealing at high temperature
in reductive atmosphere to form the alloy. More recently, methods
based on colloidal co-deposition have been developed [1,27,28],
allowing the synthesis of Cu–Pd alloys with a large range of com-
position in a single deposition step. The plating conditions can be
varied and thin (micrometer range) Cu–Pd films can be formed on
various substrates. Nevertheless, in most cases, the plating of Cu–Pd
must be followed by a high temperature annealing treatment step
to obtain homogeneous alloys.
On the other hand, very few papers have been published on
the co-electrodeposition of Cu–Pd alloys [4,5,29–32]. Recently,
Reyter et al. [4] and Milhano and Pletcher [5] were able to co-
electrodeposit Cu–Pd alloys over a wide range of composition from
chloride-base and sulfate-base baths, respectively. However, up to
now, the influence of the deposition conditions on the composition
of the film and the film morphology has not been studied in details.
This paper presents a study of Cu–Pd alloys prepared by co-
electrodeposition from nitrate baths. Potentiostatic and potential
pulsed deposition were carried out, resulting in modification
of the film morphology. Scanning electron microscopy (SEM),
energy dispersive X-ray spectroscopy (EDXS), X-ray diffraction
(XRD), transmission electron microscopy (TEM) and electrochemi-
cal roughness measurements were used to characterize the Cu–Pd
deposits. As we will see, controlling the composition of the
∗
Corresponding author.
E-mail address: guay@emt.inrs.ca (D. Guay).
0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.05.052

7398
M. Allemand et al. / Electrochimica Acta 56 (2011) 7397–7403
deposition bath and the electrochemical deposition parameters
open interesting avenues to easily tune the composition and mor-
phology of Cu–Pd alloy thin films.
0
-1
-2
-3
-4
-5
2. Experimental
2.1. Electrodeposition
0.5 M HNO3
All electrodeposition experiments were carried out at room
temperature (23 1 ◦C) in a three-electrode, one-compartment
cell with a potentiostat/galvanostat Voltalab 40 type PGZ301
(Radiometer Analytical). Titanium disks were used as substrates
for the electrodeposition (diameter = 1.1 cm). Electrodes were first
polished on SiC abrasive paper (400 mesh), then cleaned in hot (ca.
80 ◦C) oxalic acid solution at the concentration of 100 g/L for 1 h. The
electrode was then dropped in water and agitated in an ultrasonic
bath for 15 min before being thoroughly rinsed in distilled water. A
saturated calomel electrode (SCE) (Accumet SCE electrode, +0.280 V
vs. RHE) was chosen as the reference electrode. All potentials in
the present paper are referred to this reference electrode. Plat-
inum gauze was used as counter-electrode. The deposition baths
were constantly stirred with a magnetic stirrer set at 500 rpm. The
solution was also deaerated before and during the deposition by
bubbling Ar in the solution. The thicknesses of the films varied from
500 to 800 nm.
All solutions were prepared with deionized water (Milli-
pore, Milli-Q gradient). The electrodeposition baths contained
(10 − x) mM Cu(NO3)2 (Alfa Aesar Puratronic 99.999%) +x mM
Pd(NO3)2 (Alfa Aesar 99.9%) with x = 0, 2, 4, 5, 6, 8, or 10, in 0.5 M
HNO3 (Fisherbrand, ACS grade).
Electrochemical quartz crystal microbalance measurements
(EQCM) (Princeton Applied Research QCM922) were carried out
with gold plated crystals (9 MHz AT-cut, geometric surface:
0.2 cm2). The calibration constant was 1.15 ng Hz−1. In order to
determine the Pd (Cu) deposition rate, a pre-layer of Pd (Cu) was
first electrodeposited on the bare gold-coated crystals at −0.6 V
during 300 s. Then, deposition rate measurements were performed
at the desired potential. The deposition rates were calculated from
the linear section of the f = f(t) curve, where f is the frequency of
the EQCM i.e. after the initial transient regime and by ignoring
the frequency variation during the pre-deposition step described
above.
0.5 M HNO3+ 10 mM Cu(NO)
0.5 M HNO3+ 10 mM Pd(NO3)2
3 2
-0.4
-0.2
0.0
0.2
0.4
Electrode potential vs SCE / V
Fig. 1. Linear potential sweep voltammetry (at 50 mV s−1) of bare Ti disk in 0.5 M
HNO3 (dotted line), in 0.5 M HNO3 + 10 mM Cu(NO3)2 (dashed line) and in 0.5 M
HNO3 + 10 mM Pd(NO3)2 (solid line).
80-300 operated at 300 kV and equipped with a spherical aber-
ration corrector of the imaging lens. All the spectral acquisition
and data analysis were carried out using the Inca EDXS software
provided by Oxford Instruments.
Electrochemical roughness measurements were carried out by
cycling the Pd–Cu films in 0.1 M HClO4 (Aldrich, 99.999% trace metal
basis) in the double layer potential range (0–0.2 V) at various scan
rates (5, 10, 20 and 50 mV s−1). The slope of the linear regression
calculated from a plot of the current measured at 0.1 V vs. the scan
rate gave the value of the double layer capacitance, Cdl, with corre-
lation factor R2 better than 0.998. The real surface area was deduced
from Cdl by dividing by 25 ␮F cm−2, which is the value of the spe-
cific double layer capacity of Pd (and many others metals) in acidic
solutions [33–35]. Roughness factors, Rf, were finally calculated by
dividing the real surface area by the geometric surface area.
3. Results and discussion
3.1. Determination of the potential for electrodeposition of Cu–Pd
alloys by potentiostatic method
In order to determine the potential range where electrodeposi-
tion of copper and palladium on titanium can occur, linear sweep
voltammetry at 50 mV s−1 was carried out at a bare Ti electrode in
0.5 M HNO3 without any metallic cation and in presence of 10 mM
Pd(NO3)2 or 10 mM Cu(NO3)2. These voltammograms are shown in
Fig. 1. In 0.5 M H−NO3, the Ti electrode exhibits poor electrocatalytic
activity for NO3 reduction and hydrogen evolution. In the pres-
ence of copper salt, the negative sweep voltammogram shows a first
broad reduction peak, starting around −0.10 V and centered at ca.
−0.23 V, which is immediately followed by an important increase of
the reduction current at potential lower than −0.30 V. The first peak
is attributed to the reduction of Cu2+ to metallic copper deposited
on Ti (the Cu2+/Cu redox couple is 0.34 V vs. RHE), while the reduc-
tion wave is related to both bulk Cu electrodeposition and nitrate
reduction [4,36]. In the presence of palladium, Pd2+ reduction to
metallic Pd starts from the very beginning of the negative linear
sweep since the standard potential of the Pd2+/Pd redox couple is
0.635 V vs. SCE, well above the potential at the beginning of the neg-
ative linear sweep. Therefore, at the very beginning of the cathodic
sweep, metallic Pd starts to deposit on the Ti substrate (this will
be confirmed later on by EQCM measurements). Also, it is known
from the literature that H adsorption on Pd in acidic solution starts
at potential as low as +0.28 V vs. RHE [37,38]. The integrated charge
related to the peak at −0.05 V is ∼4 mC cm−2. Considering that the
2.2. Characterization of the deposits
The surface morphology of the deposits was examined by SEM
on a JEOL JSM-6300 microscope at an accelerating voltage of 15 kV.
Copper and palladium contents were estimated by EDX analyses.
The crystallographic structures were analyzed by X-ray diffraction
measurements on a Bruker AXSD8 Siemens X-ray diffractometer
˚
with Cu K␣ radiation (1.54184 A) operating at 40 kV and 40 mA.
XRD measurements were carried out in grazing incidence mode
(2◦) with an angular step size of 0.02◦ and a data acquisition time
of 6 s by step. The instrumental contribution to peak broadening
was estimated using a LaB6 microcrystalline sample. The broad-
ening of all diffraction peaks was corrected for the instrumental
contribution.
Cross-sectional samples for TEM observations were prepared by
focused ion beam using a Zeiss NVision 40 FIB/SEM. The specimens
were prepared using the lift-out method using milling with a Ga
ion beam at an accelerating voltage of 30 kV. SEM observations of
the samples for selections of sites of interest and general overview
were obtained with the same instrument.
High angle annular dark field (HAADF) images and energy dis-
persive X-ray spectroscopy maps were acquired with a FEI Titan

M. Allemand et al. / Electrochimica Acta 56 (2011) 7397–7403
7399
60
40
20
0
100
1 mM Cu(NO)
3 2
90
5 mM Cu(NO )
1 mM Pd(NO3)2
80
3 2
5 mM Pd(NO)
70
60
50
40
30
20
10
0
3 2
0
10 20 30 40 50 60 70 80 90 100
[Pd]/([Pd]+[Cu]) bath composition (at%)
-0.7
-0.6
-0.5
-0.4
-0.3
Electrode potential vs SCE
/
V
Fig. 3. Variation of the film composition with the bath composition. All films were
prepared by electrodeposition at −0.5 V during 30 min. The total metal cation con-
centration was 10 mM in all cases.
Fig. 2. Variation of the electrodeposition rate with the applied potential. The elec-
trodeposition rate was determined from EQCM measurements.
determined by EDXS, vs. the bath composition. Each experimental
point is the average of three different deposits. The deviation of the
film composition for the same bath composition was 5%. As shown
in Fig. 3, there is a one-to-one correspondence between the com-
position of the deposited film and the composition of the plating
bath. At Edep = −0.50 V, these results demonstrate that it is possible
to electrodeposit Cu–Pd films on Ti substrate over the entire com-
hydrogen adsorption charge on metallic Pd is 210 ␮C cm−2, this
would mean a roughness factor of 20, which is not un-realistic.
Therefore, the peak at −0.05 V can be attributed, at least partially,
to the underpotential deposition (UPD) of hydrogen on Pd. As sug-
gested in the literature [37,38], the second peak at −0.37 V and the
reduction wave past this potential are associated to H absorption
and the hydrogen evolution reaction (HER) at the Pd surface.
These voltammograms suggest that co-electrodeposition of Pd
and Cu from nitrate baths in potentiostatic mode should be possible
at potentials lower than ca. −0.30 V. In order to evaluate the most
efficient potential for co-electrodeposition for Cu and Pd, EQCM
experiments were carried out in nitrate-based baths with different
concentrations of Cu2+ or Pd2+ and at potentials ranging between
−0.3 and −0.7 V. The variation of the deposition rates with the
deposition potential, Edep, for baths containing 1 mM Cu2+, 5 mM
Cu2+, 1 mM Pd2+ and 5 mM Pd2+ in 0.5 M HNO3 are displayed in
Fig. 2. In the case of copper deposition, the deposition rate increases
with the Cu2+ concentration and a diminution of the electrode
potential, as expected for a charge transfer controlled reaction. In
the case of Pd-containing solutions, the deposition rate does not
vary for Edep = −0.30 and −0.40 V, indicating that deposition is lim-
ited by the diffusion of Pd2+ species to the electrode surface in this
potential range. For Edep lower than −0.40 V, the deposition rate
increases and goes through a maximum whose position depends
on the Pd+2 concentration. In this potential range, several reactions
involving hydrogen are expected, including hydrogen underpoten-
tial deposition, hydrogen absorption and hydrogen evolution. In
that later case, the formation of H2 bubbles induces solution stir-
ring and could also block part of the electrode. The occurrence of
a maximum in the Pd deposition rate must arise as a result of a
delicate balance between all these phenomena.
A
B
C
D
E
F
For the following, Edep = −0.50 V will be considered as an ade-
quate potential for the co-electrodeposition of Cu and Pd. Indeed,
this potential is close to the maximum deposition rate of Pd. More-
over, the deposition rates of both metals at this potential are quite
similar (less than 25% difference) in comparison with the values at
−0.60 V (more than 50% difference for [Cu+2] = [Pd+2] = 5 mM).
Ti
Pd
Cu
3.2. Potentiostatic electrodeposition at −0.50 V of Cu–Pd films
30
40
50
60
70
80
90
2θ / degree
In the light of the previous results, a potential of −0.50 V was
applied during 30 min for the co-electrodeposition of copper and
palladium on Ti substrates. Different baths containing (10 − x) mM
Cu(NO3)2 + x mM Pd(NO3)2 in 0.5 M HNO3 (x = 0, 2, 4, 5, 6, 8, or 10)
were tested. Fig. 3 presents the composition of the deposited films,
Fig. 4. XRD diffractograms of Cu–Pd films deposited from electrolytic baths with
different compositions. The composition of the films is (A) Pd, (B) Pd84Cu16, (C)
Pd65Cu35, (D) Pd48Cu52, (E) Pd15Cu85, and (F) Cu. The characteristic diffraction peaks
of Ti, Pd and Cu are also indicated at the bottom.

7400
M. Allemand et al. / Electrochimica Acta 56 (2011) 7397–7403
0.390
0.385
0.380
0.375
0.370
0.365
0.360
trodeposition allows the formation of mono-phased Cu–Pd alloys.
There is a progressive shift of the Bragg peaks to higher diffraction
angles as the Cu content of the Cu–Pd film is increased, indicating a
decrease of the lattice parameter as a result of the alloying process.
As far as we can tell from these XRD measurements, there is
no phase present other than the Fm-3m phase that is stable over
the whole composition range at temperature higher than ∼600 ◦C
according to the Cu–Pd binary phase diagram [39]. More specif-
ically, the ␤-CuPd phase (Pm-3m), which is the stable structure
at the Cu–Pd equimolar concentration and at temperature below
598 ◦C and whose main (1 1 0) diffraction peak occurs at 2ꢀ = 42.92◦
with a (1 0 0) characteristic peak at 2ꢀ = 30.09◦, is not observed.
Thus, Cu–Pd films prepared by electrodeposition spontaneously
adopted the high temperature phase structure. As expected, the
Fm-3m structure of Cu–Pd reverts to the stable Pm-3m structure
upon heat-treatment in an inert atmosphere at temperature below
598 ◦C (data not shown).
The variation of the lattice parameter with the composition of
the films is reported in Fig. 5. The dashed straight line shows the
variation of the lattice parameter expected from Vegard’s law. The
lattice parameter of the Cu–Pd thin films exhibits a positive devi-
ation from Vegard’s law. The deviation is maximum at equimolar
concentration. The same behavior was reported by Subramanian
and Laughlin [40] and Barrett and Massalski [41]. According to
Vegard’s law, the unit cell parameters should vary linearly with
composition for an ideal and continuous substitutional solid solu-
tion in which atoms that substitute for each other are randomly
distributed [42]. The deviation from ideal behavior observed pre-
viously may result from non-ideality of the solid solution enthalpy
of mixing [43] and weaker Cu–Pd interactions compared to pure
Cu–Cu and Pd–Pd [44].
0
10 20 30 40 50 60 70 80 90 100
[Pd]/([Pd]+[Cu]) film concentration (at%)
Fig. 5. Variation of the lattice parameter of the fcc phase with respect to the Pd
content of the film. The short dashed straight line shows the variation expected on
the basis of the Vegard’s law.
position range simply by changing the composition of the plating
bath.
XRD measurements were carried out at the Cu–Pd films
deposited from different baths. Fig. 4 displays the X-ray diffrac-
tograms of all samples. Also, in that figure, the positions of the
diffraction peaks of pure Pd, Cu and Ti are indicated. All samples
display the characteristic diffraction peaks of pure Ti (hexagonal
close-packed, hcp) associated with the substrate. Considering these
diffractogramms were obtained in grazing incidence configuration
and the small intensity of the Ti peaks, the thickness of the films is
estimated to be in the micrometer range. The other peaks seen in
Fig. 4 can be assigned to Pd, Cu or Cu–Pd face centered cubic (fcc)
structure. Neither pure palladium nor pure copper diffraction peaks
are present in the bimetallic Cu–Pd coatings, indicating that elec-
The crystallite sizes of the various CuPd films were evaluated
from the full width at half maximum of the (1 1 1) peak of the XRD
patterns. The crystallite size decreases steadily from 21 nm for pure
Pd to 8 nm for Cu60Pd40 and then increases to 75 nm for pure Cu.
Fig. 6. SEM micrographs of Cu–Pd thin films prepared by co-electrodeposition of Cu and Pd. The composition of the film is (A) Cu, (B) Cu84Pd16, (C) Cu52Pd48, (D) Cu35Pd65
(E) Cu16Pd84, and (F) Pd. All films were prepared at −0.50 V vs. SCE.
,

M. Allemand et al. / Electrochimica Acta 56 (2011) 7397–7403
7401
Fig. 7. (A) High-angle annular dark-field scanning transmission electron microscopy image of cross-sectional Cu40Pd60 alloy thin film deposited on a Ti substrate. The Pd L␣1
(B) and Cu K␣1 (C) EDXS elemental maps of the same regions shown in (A) as highlighted with a frame. The carbon coating shown in (A) corresponds to the thin protective
layer deposited as part of the FIB sample preparation process.
Both EDXS and XRD analyses demonstrate that Cu–Pd alloys can be
deposited potentiostatically on Ti substrates over the entire com-
position range. Moreover, alloying of Cu and Pd reduces the mean
crystallite size and thus promotes the formation of nanoscale struc-
tures, which can be very interesting for electrocatalysis and surface
applications.
ing electrodeposition, the film morphology is strongly dependent
on the competition between nucleation and growth processes [45].
More precisely, the higher the nucleation rate and mass transport
current are during the electrodeposition process, the denser and
smoother the surface of the coating will be. An efficient way to
increase both the nucleation rate and the mass transport current is
to pulse the electrode potential during the electrodeposition pro-
cess [46–52].
The SEM micrographs of a few representative samples are
shown in Fig. 6. All Cu–Pd alloy films present a porous and
cauliflower-like structure, which should induce a large electroac-
tive area. It will be shown later that, in some cases, the roughness
factor of these films can exceed 100. The morphologies of pure
Pd (dendritic deposit) and pure Cu (large crystals) films are very
different from each other and from that of the Cu–Pd alloy films.
Imaging with high angle annular-dark field scanning transmis-
sion electron microscopy was carried out on one of the deposit
(Pd0.4Cu0.6). As shown in Fig. 7A, the Ti substrate is totally covered
by the electrodeposited film. Also, this micrograph confirms the
porous nature of the co-electrodeposited film in this range of com-
position. The element-specific maps obtained with EDXS shown in
Fig. 7B (Pd) and Fig. 7C (Cu) confirm that both elements are evenly
distributed in the thickness of the films.
3.3. Pulse electrodeposition
Since most of the Cu–Pd alloys used in hydrogen purification
applications have a composition around Pd40Cu60, all the Cu–Pd
films made by pulse electrodeposition were deposited from a 6 mM
Cu(NO3)2 + 4 mM Pd(NO3)2 bath in 0.5 M HNO3, which according to
the previous results, lead to a film composition in the appropriate
range for that application. In the following, all experiments were
carried out at room temperature under constant stirring.
It was suspected that oxidation–dissolution of copper during the
pulse at Eoff could be a critical issue in our attempt to get homo-
geneous and smooth films with a controlled composition. Hence
Eoff was set at a potential sufficiently negative to avoid this phe-
nomenon, i.e. −0.10 V. From the voltammograms shown in Fig. 1,
it is evident that Pd2+ can be reduced at this potential, while the
current for the reduction of Cu+2 to Cu0 is negligible. Hence, for
Although a porous and high surface area structure is very
convenient for electrocatalytic applications, denser and more
mechanically resistant films are required for the preparation of
metallic membranes used in the field of hydrogen purification. Dur-

7402
M. Allemand et al. / Electrochimica Acta 56 (2011) 7397–7403
Table 1
not investigated in details since we are concerned with decreas-
ing the Rf values. Note that in the perspective of H2 purification,
dense Cu–Pd films could be topped with a rougher surface of the
same composition in a single experiment, just by switching from
potential pulse co-electrodeposition to potentiostatic deposition.
This could be a very attractive approach to increase the roughness
factor of the utmost surface layer of an otherwise dense mem-
brane and offers an interesting avenue if surface reactions are the
rate-determining step in the purification process.
Variation of the average roughness factor with toff, ton and Eon. All films were pre-
pared with Eoff = −0.1 V and under solution stirring (500 rpm).
a
b
c
Effect of toff on Rf
Effect of ton on Rf
Effect of Eon on Rf
toff (ms)
Rf
ton (ms)
Rf
Eon (V)
Rf
0
5
50
122 14
12.3 1.5
11.8 1.9
1
5
10
15 11
11.8 1.9
10.4 0.3
−0.3
−0.4
−0.5
−0.6
9
4
7.7 1.5
11.8 1.9
45 35
a
Finally, the last set of results presented in Table 1 concerns
the effect of Eon. Although there is some dispersion in the values
of films prepared at Eon = −0.30 and −0.60 V, the data of Table 1
indicate that the film roughness increases as the applied poten-
tial is decreased (more negative). This effect is most particularly
noticeable for Eon = −0.60 V. At that potential, electrodeposition
of Pd is accompanied by strong hydrogen evolution. Accordingly,
Eon = −0.50 V offers the best compromise between a smooth mor-
phology and a high deposition rate.
Eon = −0.5 V; ton = 5 ms.
Eon = −0.5 V; toff = 50 ms.
ton = 5 ms; toff = 50 ms.
b
c
films prepared with Eoff = −0.10 V, the Pd content was expected to
vary with toff. However, in sulfuric acid, it is known that Cu can be
underpotentially deposited (UPD) on palladium at potentials lower
than 0.5 V vs. RHE, which corresponds to 0.22 V vs. SCE [53–55].
Thus, UPD of Cu on Pd will occur at Eoff = −0.10 V vs. SCE and Pd
enrichment during pulsed potential deposition is not expected to
be as important as it could be expected from the previous con-
siderations. Indeed, using a 6 mM Cu(NO3)2 + 4 mM Pd(NO3)2 bath,
the composition of the films prepared by pulsed electrodeposi-
tion ([Cu] = 60.7 at.% and [Pd] = 39.3 at.%) was not found to vary
significantly from that of films prepared by potentiostatic elec-
trodeposition in the same bath electrolyte ([Cu] = 57.8 at.% and
[Pd] = 42.2 at.%).
Three deposition parameters were varied to evaluate their effect
on the morphology of the film: Eon, the most negative potential at
which the deposition was performed; ton, the duration of the pulse
at Eon; toff, the duration of the pulse at Eoff. In order to investigate
the influence of these three parameters, the following deposition
conditions were chosen as a reference state: Eon = −0.5 V, ton = 5 ms,
toff = 50 ms. Although SEM is an excellent method to qualitatively
observe the effect of each parameters on the morphology of films, it
is not well suited for quantification purposes. Thus, an evaluation of
the roughness factor by an electrochemical method was performed
to quantitatively evaluate the effect of the different deposition
parameters on the morphology of the film. This method was chosen
since it offers a quick and in situ evaluation of the roughness factor.
Details on the procedure used to determine the roughness factor
are given in the experimental part. Table 1 summarizes the values
of the average roughness factor, Rf, for all tested conditions. Aver-
age values were calculated from at least six different films made in
each deposition conditions.
The first parameter that was examined was toff, as setting it to
toff = 0 is equivalent to potentiostatic deposition. As seen in Table 1,
the Rf values of films made in the pulsed electrodeposition mode
(toff =/ 0 ms) are a factor of about ten lower than films made in the
potentiostatic mode (toff = 0 ms). This confirms the high efficiency
of the pulse electrodeposition mode for producing smoother Cu–Pd
deposits. Taking into account the deviation of the values, no sig-
nificant difference is observed between films prepared at toff = 5 or
50 ms. From these results, toff = 5 ms could be considered as an opti-
mum value since it insures that smooth films can be prepared in
the quickest manner.
Table 1 also demonstrates that varying ton from 1 ms to 10 ms
does not have any dramatic effect on the film roughness, although
the reproducibility of films prepared with ton = 1 ms is less than that
prepared at longer ton. In the former case, this is thought to arise
as a consequence of electrodeposition occurring mainly during the
early transient state that follows the application of Eon. Obviously, if
ton is too long, films prepared in the pulse deposition mode will not
differ significantly from those prepared in the potentiostatic mode
(which is equivalent to the pulse deposition mode with ton = ∞). The
effect of increasingly longer ton period on the film morphology was
4. Conclusion
From this study, the following conclusions can be drawn:
1. Mixed Cu–Pd thin film can be obtained by potentiostatic co-
electrodeposition on Ti substrate over the entire composition
range simply by changing the composition of the plating bath.
2. In all cases, X-ray diffraction measurements revealed that all
coatings are made of a single phased CuxPd100−x fcc structure.
3. Films prepared in the potentiostatic mode were rough and
porous, with roughness factor exceeding 120. These deposits
could be suitable for electrocatalysis applications.
4. However, it is possible to get denser and more mechanically
resistant deposits, more appropriate for hydrogen purification,
by using pulsed potential co-electrodeposition. In that case, films
smooth and dense films are obtained with roughness factor as
low as ca. 8.
Acknowledgments
The authors would like to thank the Natural Sciences and Engi-
neering Research Council of Canada (NSERC), the NSERC Hydrogen
Canada (H2CAN) Strategic Research Network and the Fonds Québé-
cois de la Recherche sur la Nature et les Technologies (FQRNT) for
supporting this work. Transmission electron microscopy was car-
ried out at the Canadian Centre for Electron Microscopy, a facility
supported by NSERC and McMaster University.
References
[1] X. Wang, N. Kariuki, J.T. Vaughey, J. Goodpaster, R. Kumar, D.J. Myers, J. Elec-
trochem. Soc. 155 (2008) B602.
[2] F. Fouda-Onana, S. Bah, O. Savadogo, J. Electroanal. Chem. 636 (2009) 1.
[3] O. Ghodbane, M. Sarrazin, L. Roué, D. Bélanger, J. Electrochem. Soc. 155 (2008)
F116.
[4] D. Reyter, D. Bélanger, L. Roué, J. Phys. Chem. C 113 (2009) 290.
[5] C. Milhano, D. Pletcher, J. Electroanal. Chem. 614 (2008) 24.
[6] Y. Wang, J. Qu, Water Environ. Res. 78 (2006) 724.
[7] C.C. Yang, A.S. Kumar, M.C. Kuo, S.H. Chien, J.M. Zen, Anal. Chim. Acta 554 (2005)
66.
[8] S.K. Gade, P.M. Thoen, J.D. Way, J. Membr. Sci. 316 (2008) 112.
[9] P. Kamakoti, B.D. Morreale, M.V. Ciocco, B.H. Howard, R.P. Killmeyer, A.V. Cugini,
D.S. Sholl, Science 307 (2005) 569.
[10] A. Kulprathipanja, G.O. Alptekin, J.L. Falconer, J.D. Way, J. Membr. Sci. 254 (2005)
49.
[11] A. Qiao, K. Zhang, Y. Tian, L. Xie, H. Luo, Y.S. Lin, Y. Li, Fuel 89 (2010) 1274.
[12] F. Roa, M.J. Block, J.D. Way, Desalination 147 (2002) 411.
[13] F. Roa, J.D. Way, R.L. McCormick, S.N. Paglieri, Chem. Eng. J. 93 (2003) 11.
[14] P.M. Thoen, F. Roa, J.D. Way, Desalination 193 (2006) 224.
[15] B.D. Morreale, M.V. Ciocco, B.H. Howard, R.P. Killmeyer, A.V. Cugini, R.M. Enick,
J. Membr. Sci. 241 (2004) 219.

M. Allemand et al. / Electrochimica Acta 56 (2011) 7397–7403
7403
[16] B.H. Howard, R.P. Killmeyer, K.S. Rothenberger, A.V. Cugini, B.D. Morreale, R.M.
Enick, F. Bustamante, J. Membr. Sci. 241 (2004) 207.
[17] C. Decaux, R. Ngameni, D. Solas, S. Grigoriev, P. Millet, Int. J. Hydrogen Energy
35 (2010) 4883.
[18] J.B. Miller, C. Matranga, A.J. Gellman, Surf. Sci. 602 (2008) 375.
[19] J.Y. Yang, C. Nishimura, M. Komaki, J. Membr. Sci. 282 (2006) 337.
[20] J. Chakraborty, U. Welzel, E.J. Mittemeijer, J. Appl. Phys. 103 (2008) 113512.
[21] J. Chakraborty, U. Welzel, E.J. Mittemeijer, Thin Solid Films 518 (2010) 2010.
[22] S.K. Ryi, J.S. Park, S.H. Kim, D.W. Kim, K.I. Cho, J. Membr. Sci. 318 (2008) 346.
[23] V. Bhaskaran, P. Atanasova, M.J. Hampden-Smith, T.T. Kodas, Chem. Mater. 9
(1997) 2822.
[24] V.M. Ievlev, E.K. Belonogov, A.A. Maksimenko, Phys. Met. Metallogr. 105 (2008)
574.
[25] S.N. Paglieri, K.Y. Foo, J.D. Way, J.P. Collins, D.L. Harper-Nixon, Ind. Eng. Chem.
Res. 38 (1999) 1925.
[26] R. Javaid, H. Kawanami, M. Chatterjee, T. Ishizaka, A. Suzuki, T.M. Suzuki, Catal.
Commun. 11 (2010) 1160.
[27] N.N. Kariuki, X. Wang, J.R. Mawdsley, M.S. Ferrandon, S.G. Niyogi, J.T. Vaughey,
D.J. Myers, Chem. Mater. 22 (2010) 4144.
[28] M. Shao, K. Shoemaker, A. Peles, K. Kaneko, L. Protsailo, J. Am. Chem. Soc. 132
(2010) 9253.
[29] F. Gobal, R. Arab, J. Electroanal. Chem. 647 (2010) 66.
[30] S.N. Vinogradov, G.N. Mal’tseva, N.A. Gulyaeva, Russ. J. Electrochem. 37 (2001)
715.
[31] X. Gao, J. Tang, Y. Zuo, Y. Tang, J. Xiong, Corros. Sci. 51 (2009) 1822.
[32] M. Łukaszewski, K. Klimek, A. Zurowski, T. Kedra, A. Czerwin´ ski, Solid State
Ionics (2010), doi:10.1016/j.ssi.2010.03.00.
[35] P. Waszczuk, P. Zelenay, J. Sobkowski, Electrochim. Acta 40 (1995) 1717.
[36] A. Ramos, M. Miranda-Hernandez, I. Gonzalez, J. Electrochem. Soc. 148 (2001)
C315.
[37] L. Birry, A. Lasia, Electrochim. Acta 51 (2006) 3356.
[38] H. Duncan, A. Lasia, Electrochim. Acta 53 (2008) 6845.
[39] H. Baker, H. Okamoto (Eds.), ASM Handbook, Alloy Phase Diagrams, Volume 03,
ASM International, 1992.
[40] P.R. Subramanian, D.E. Laughlin, J. Phase Equilib. 12 (1991) 231.
[41] C.S. Barrett, T.B. Massalski, Structure of Metals: Crystallographic Methods, Prin-
ciples and Data, Pergamon, New York, 1993.
[42] K.T. Jacob, S. Raj, L. Rannesh, Int. J. Mater. Res. 98 (2007) 776.
[43] V.S. Urusov, J. Struct. Chem. 33 (1992) 68.
[44] The CRC, Handbook of Solid State Electrochemistry, CRC Press, New York, 1997.
[45] E.B. Budevski, G.T. Staikov, W.J. Lorenz, Electrochemical Phase Formation and
Growth: An Introduction to the Initial Stages of Metal Deposition, VCH, Wein-
heim, 1996.
[46] A. Rodchanarowan, R. Bhide, M.L. Free, Hydrometallurgy 2008: Proceedings of
the 6th International Symposium, 2008, p. 632.
[47] H.O. Beckmann, H. Gerischer, D.M. Kolb, G. Lehmpfuhl, Faraday Symp. Chem.
Soc. 12 (1977) 51.
[48] C.H. Chen, A.A. Gewirth, J. Am. Chem. Soc. 114 (1992) 5439.
[49] J.P. Ganon, J. Clavilier, Surf. Sci. 145 (1984) 487.
[50] B.W. Gregory, D.W. Suggs, J.L. Stickney, J. Electrochem. Soc. 138 (1991) 1279.
[51] Y. Nakai, D.M. Kolb, G. Lehmpfuhl, M.S. Zei, Ber. Bunsen. Phys. Chem. 88 (1984)
340.
[52] M.S. Zei, G. Qiao, G. Lehmpfuhl, D.M. Kolb, Ber. Bunsen. Phys. Chem. 91 (1986)
349.
[33] H. Duncan, A. Lasia, Electrochim. Acta 52 (2007) 6195.
[34] M.L. Tremblay, M.H. Martin, C. Lebouin, A. Lasia, D. Guay, Electrochim. Acta 55
(2010) 6283.
[53] F. Hernandez, H. Baltruschat, J. Solid State Electron. 11 (2007) 877.
[54] M. Nakamura, Y. Sakurai, M. Ito, J. Electroanal. Chem. 563 (2004) 63.
[55] J. Okada, J. Inukai, K. Itaya, Phys. Chem. Chem. Phys. 3 (2001) 3297.