Oxygen reduction reaction on electrodeposited Pt100-x-yNi xPdy thin films

Article abstract of DOI:10.1016/j.electacta.2010.08.013

The kinetics of the oxygen reduction reaction (ORR) were examined on a series of Pt_100-x-yNi_xPd_y ternary alloys. Films were produced by electrodeposition that involved a combination of underpotential and overpotential reactions. For Pt-rich Pt _100-x-yNi_xPd_y alloy films (x < 0.65) Ni co-deposition occurred at underpotentials while for Ni-rich films (x > 0.65) deposition proceeded at overpotentials. Rotating disk electrode (RDE) measurements of the ORR kinetics on Ni-rich Pt_100-x-yNi _xPd_y thin films revealed up to ～6.5-fold enhancement of the catalytic activity relative to Pt films with the same Pt mass loading. More than half of the electrocatalytic gain may be attributed to surface area expansion due to Ni dealloying. Surface area normalization based on the H _upd charge reduced the enhancement factor to a value less than 2. The most active ternary alloy film for ORR was Pt₂₅Ni ₇₃Pd₂. Comparison of the ORR on Pt, Pt₂₀Ni ₈₀, Pt₂₅Ni₇₃Pd₂ thin films indicate that the binary alloy is the most active with a H_upd normalized ORR enhancement factor of up to 3.0 compared to 1.6 for the ternary alloy.

Full text of DOI:10.1016/j.electacta.2010.08.013

Electrochimica Acta 55 (2010) 8938–8946
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Oxygen reduction reaction on electrodeposited Pt1
Ni Pd thin films
00−x−y
x
y
Sun-Mi Hwanga,b, Chang Hwa Lee , Jae Jeong Kim , Thomas P. Moffat
a
b
a,∗
a
Materials Science and Engineering Laboratory, National Institute of Standards and Technology, 100 Bureau Drive, Mail Stop 8551, Gaithersburg, MD 20899, United States
School of Chemical and Biological Engineering, Seoul National University, San 56-1, Sillim-dong, Gwanak-gu, Seoul 151-744, Republic of Korea
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 June 2010
Received in revised form 30 July 2010
Accepted 3 August 2010
Available online 10 August 2010
The kinetics of the oxygen reduction reaction (ORR) were examined on a series of Pt1
00−x−y x y
Ni Pd ternary
alloys. Films were produced by electrodeposition that involved a combination of underpotential and
overpotential reactions. For Pt-rich Pt100−x−yNixPdy alloy films (x < 0.65) Ni co-deposition occurred at
underpotentials while for Ni-rich films (x > 0.65) deposition proceeded at overpotentials. Rotating disk
electrode (RDE) measurements of the ORR kinetics on Ni-rich Pt100−x−yNixPdy thin films revealed up to
∼
6.5-fold enhancement of the catalytic activity relative to Pt films with the same Pt mass loading. More
Keywords:
than half of the electrocatalytic gain may be attributed to surface area expansion due to Ni dealloy-
ing. Surface area normalization based on the Hupd charge reduced the enhancement factor to a value
less than 2. The most active ternary alloy film for ORR was Pt25Ni73Pd2. Comparison of the ORR on Pt,
Pt20Ni80, Pt25Ni73Pd2 thin films indicate that the binary alloy is the most active with a Hupd normalized
ORR enhancement factor of up to 3.0 compared to 1.6 for the ternary alloy.
Electrocatalysis
Electrodeposition
Pt alloys
PtNiPd
Oxygen reduction reaction (ORR)
Published by Elsevier Ltd.
1
. Introduction
Platinum is the conventional catalyst for hydrogen oxidation
an effective strategy to substantially improve upon these limita-
tions [1,2,16–29]. Although the reason for the superior behavior
exhibited by various Pd-M (M = Ni, Fe, Co, Pt) or Pt-M (M = Co,
Ni, Cr, Mn, V) alloys is still debated, it is thought to be associ-
ated with perturbation of the electronic state and geometry of a
Pt enriched surface layer by the underlying alloy [1,6–8,19]. This
is often discussed in terms of electronic effects (d-band electron
vacancy), geometric factors (Pt–Pt bond distance), surface compo-
sition, and/or atomic ordering. Models continue to be developed
and evaluated with ever more sophistication [7,8,20]. Building on
these ideas new approaches to engineering Pt-rich skin/skeleton
and core–shell geometries are being actively explored [1,2,8]. In
addition to the prospect of enhanced catalytic performance these
structures also promise considerable cost savings by maximizing
Pt utilization [1,3]. In the case of Pt shell/Pt-free core particles,
the usage of Pt per unit power can be reduced by a factor of
5–10 compared to Pt alloy particles [3,31]. At the same time new
developments in electrode architecture remain an important area
for further improvement in catalyst utilization and performance.
One of the more interesting developments in the last decade has
been the fabrication of nanostructured thin film (NSTF) Pt and/or
Pt/TM (TM = Ni, Co, Mn, Fe) electrodes by vacuum sputter depo-
sition [32,33]. These materials have shown 5–10 times enhanced
activity toward ORR with reduced amount of Pt compared to con-
ventional membrane electrode assembly (MEA) [32]. Furthermore
the NSTF electrodes have been shown to be far more durable than
carbon supported Pt particles. The combination of improvements
speaks to new avenues for the design, development and utilization
of electrocatalyst. Nevertheless, a cost effective means for produc-
reaction (HOR) and/or oxygen reduction reaction (ORR) in polymer
electrolyte membrane fuel cells (PEMFCs). However, the sluggish
oxygen reduction reaction (ORR) associated with Pt cathodes in
PEMFC is one of the critical limitations hindering commercializa-
tion. The high cost combined with limited performance, durability,
and susceptibility to poisoning, remain major barriers to the uti-
lization of pure Pt. Consequently, significant research is underway
exploring the activity and utility of various platinum-based binary
(
PtM, M = Co, Ni, Fe, Cu, Cr, etc.) or ternary alloys (PtM M , M = Co,
1 2 1,2
Ni, Fe, Cu, Cr, Zr, Mn, etc.) [1–30]. Generally speaking many of
these alloys have been shown to improve the ORR activity com-
pared to unalloyed Pt by a factor that typically ranges between
2
and 4. Even more encouraging, Pt Ni(1 1 1) single crystals have
3
been reported to be 10-fold more active than Pt(1 1 1) and 90-
fold more active than state-of-the-art Pt/C catalyst [6]. Looking
beyond Pt-based alloys, Pd has been examined as a Pt substitute
based on its similar physical properties that arise from the peri-
odicity of elements [1,2,17–28]. However, detailed investigations
have revealed that Pd has lower activity and stability towards
the ORR. As with Pt, alloying Pd with other metals, M, Pd-M
(
M = Ni, Pt, Fe, Cu, Co, W, Sn, V, Au etc.) has been shown to be
∗ Corresponding author. Tel.: +1 301 975 2143; fax: +1 301 926 7679.
E-mail addresses: thomas.moffat@nist.gov, thomas.moffat@nist.gov
(
T.P. Moffat).
0
013-4686/$ – see front matter. Published by Elsevier Ltd.
doi:10.1016/j.electacta.2010.08.013

S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
8939
ing compositionally tailored nanoscale electrocatalysts remains to
be demonstrated.
beaker connected to the cell through a Luggin capillary to pre-
vent contamination of the main cell by chloride crossover. Cyclic
voltammetry (CV) was performed as a function of RDE rotation rate,
from 400 rpm to 4900 rpm, using a sweep rate of 50 mV/s. The ORR
data presented herein correspond to the positive-going potential
sweep collected at room temperature. For benchmarking purposes
Among the various synthetic methods being explored, elec-
trodeposition has advantages in terms of Pt utilization and
controlled mass loading since the material is deposited only on sites
where both electronic and ionic contact are insured [11,14,34,35].
Recently a simple electrodeposition process for co-deposition of
the activity of Pt1
with Pt and Pt
NixPdy ternary alloy films was compared
Nix grown under the same conditions. The stated
00−x−y
Pt-transition metal alloy thin films (Pt1 Mx, M = Ni, Co, Fe, Cu) has
−x
100−x
been studied [11,36–39]. In the case of Pt-rich alloys, co-deposition
of iron-group metals proceeds by underpotential deposition (upd)
concurrent with overpotential Pt deposition.
potentials are referred to the reversible hydrogen electrode (RHE)
using the offset RHE = SCE + 0.304 V that was determined previously
[11].
In this paper the electrodeposition of Pt1
ternary catalysts is detailed. Potentiostatic deposition is shown
to be an effective means to control the composition of the films.
NixPdy thin film
To estimate the electrochemical surface area of the electrode-
posited films, CV experiments were performed from 0 V RHE to
1.0 V RHE in Ar saturated 0.1 mol/L HClO4 at room temperature.
The integrated Hupd charge between 0.040 V RHE and 0.408 V RHE
00−x−y
The electrocatalytic behavior of Pt1
toward ORR is examined and compared to Pt1
NixPdy ternary alloy films
00−x−y
2
Nix binary and
was normalized to the conventional value of 0.210 mC/cm . The
00−x
elemental Pt films produced by the same deposition method.
importance of area correction to conclusions regarding the intrin-
sic ORR kinetics cannot be overstated, and the issue remains a topic
of central importance to the field. The reader is referred to the lit-
erature for a more detailed discussion of the matter [31,41,42]. In
the present study the estimated area may be viewed as a conserva-
tive measure since the Hupd charge includes double layer capacitive
effects as well as possible H absorption into the Pt–Ni–Pd thin
films. The latter is expected to be less significant for dilute Pd
alloys [43,44]. In addition to the magnitude of the electrode area,
morphology is also expected to exert an influence on the quan-
tification of the intrinsic ORR kinetics. When the voltammetric
measurements are such that the reaction occurs under mixed con-
trol conditions it is clear that significant segments of the recessed
surface area, quantified by “Hupd”, remain largely inaccessible to
O2 thereby leading to a conservative estimate for the measured
kinetics [41].
2
. Experimental
A series of Pt1
NixPdy ternary alloy films were electrode-
00−x−y
posited from an electrolyte containing 0.5 mol/L NaCl, 0.1 mol/L
NiCl ·6H O, 0.003 mol/L K PtCl , and 0.001 mol/L PdCl at a pH
2
2
2
4
2
of 2.5. Deposition was performed using a typical three-electrode
cell with a Pt anode and saturated calomel reference electrode
(
SCE). Working electrodes were prepared by using electron beam
evaporation to coat SiO /Si (1 0 0) wafers with a 3 nm Ti adhesion
2
layer followed by a 25 nm-thick Au seed layer. The Au seeded sub-
strates with an exposed area of 1.33 cm2 defined by 3M plater’s
tape were immersed into the electrolyte for alloy deposition at
room temperature. The effect of growth potential on film compo-
sition and microstructure was investigated over the range −0.1 V
SCE to −0.8 V SCE. The films were grown for a fixed time of 1800 s
yielding thickness values ranging from 140 nm to 600 nm. These
The effect of Pt1
NixPdy dealloying during ORR mea-
00−x−y
surements was quantified by examining the catalytic behavior
of films as a function of as-deposited film thickness. Likewise,
the evolution of the surface morphology and microstructure of
“
thick films” were used to characterize the potential dependence of
the structure and composition through the use of symmetric X-ray
diffraction (XRD) and energy dispersive X-ray spectroscopy (EDS).
Robust compositional determination by EDS requires careful anal-
ysis because of its low analytical sensitivity [40]. In this report a ZAF
Pt1
NixPdy films was investigated by field emission scanning
00−x−y
electron microscopy (FE-SEM) and XRD. For these experiments the
“thick” Pt1 NixPdy films grown on Au were used to ensure
sufficient scattering signal.
00−x−y
(
Z, atomic number; A, absorption; F, fluorescence) thin film matrix
correction was used for Pt-L␣, Pd-L␣, and Ni-K␣ spectra peaks col-
lected at the accelerating voltage of 15 keV in order to avoid peak
overlaps between Pt-M␣ and Au-M␣ lines.
3. Results and discussion
The ORR kinetics for Pt1
NixPdy ternary alloy films were
00−x−y
examined using a Pt rotating disk electrode (RDE) with a projected
3.1. Characteristics of Pt1
NixPdy
00−x−y
area of 0.196 cm2 as the substrate. Pt1
NixPdy films were
00−x−y
deposited on the Pt disk under potentiostatic conditions detailed
above, although deposition was limited to 30 s giving rise to much
thinner films than described in the previous paragraph. Prior to
each experiment, the Pt RDE was polished using 50 nm alumina
abrasive powder to remove the previously plated material. After
film deposition, the RDE was rapidly removed from the chloride
based plating bath and rinsed extensively with 18 Mꢀ cm deion-
ized (DI) water for tens of second in an attempt to remove any
X-ray diffraction patterns for Pt1
NixPdy films electrode-
00−x−y
posited on Au seeded Si for 1800 s at various potentials are shown
in Fig. 1. The invariant Au(1 1 1) and Au(2 2 2) peaks of the gold
seeded substrates are helpful as internal references for the analysis.
Likewise, Pt and Pt20Ni80 films deposited at −0.6 V SCE exhibit dis-
◦
◦
tinct (1 1 1) peaks at about 40.11 and 42.96 , respectively. For the
Pt1
NixPdy films, the fcc (1 1 1), (2 0 0) and (2 2 2) diffraction
peaks shift to higher angles as the deposition potential decreased
00−x−y
−
residual adsorbed Cl from the electrode surface. This is impor-
from −0.1 V SCE to −0.8 V SCE. The (1 1 1), (2 0 0) and (2 2 2) peaks
tant since halides are known to poison the ORR kinetics on Pt. The
RDE was then placed in a beaker containing 100 mL DI water fol-
lowed by storage in a second 100 mL beaker of DI water for transfer
to the RDE set up. During mounting of the RDE the electrode-
posited surface was covered with a water droplet. The electroplated
surface was then immersed into an oxygen saturated 0.1 mol/L
HClO4 solution (Mallinckrodt Chemicals 70%)* for measuring the
ORR kinetics. *(Certain trade names are mentioned for experimen-
tal information only; in no case does it imply a recommendation
or endorsement by NIST). The cell contained a Pt counter elec-
trode while the SCE reference electrode was held in an external
shifts for Pt1
NixPdy films correspond to a monotonic decrease
00−x−y
in the calculated lattice parameter from 0.3882 nm to 0.3607 nm
for the (1 1 1) (and from 0.3878 nm to 0.3622 nm for the (2 0 0)) as
shown in Fig. 2. The trends is congruent with increasing Ni content
in the films that eventually approaches the lattice parameter of bulk
fcc Ni (ao = 0.3520 nm). The corresponding potential dependence of
the ternary film compositions determined by energy dispersive X-
ray spectroscopy (EDS) is shown in Fig. 3a. Five different spots were
sampled on each film and the results averaged with the range indi-
cated by the length of the data bar. The atomic concentration of
Pt and Pd gradually decreased from 67% and 29% to 10% and 0.1%,

8
940
S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
Fig. 1. X-ray diffraction patterns for a series of Pt100−x−yNixPdy alloy films electrode-
posited for 1800 s as a function of growth potential.
Fig. 2. Lattice constants (a ) for electrodeposited Pt, Pt
Nix and Pt
100−x−y x
Ni Pdy
0
100−x
alloy films at various growth potentials.
respectively, while the Ni content increased from 4% to 90% as the
growth potential decreased. EDS data are in good qualitative agree-
ment with the lattice parameter change of Ni measured by XRD.
Quantitative comparison would require consideration of the stress
effects and a full measurement error analysis beyond the objectives
of the present study. The Nernst potential for the Ni/Ni2+ reaction is
close to −0.58 V SCE thus its co-deposition at more positive poten-
tials occurs by an underpotential process. In contrast both Pd and
Pt deposition occur at overpotentials. If Pt and Pd deposition occur
under mass transport limited conditions the ratio in the deposit
should reflect that of the electrolyte, namely 3:1 Pt:Pd. This is the
case for film deposited at −0.2 V SCE, −0.3 V SCE and −0.4 V SCE.
At −0.1 V SCE the ratio is slightly less and may reflect some kinetic
restraint on the Pt half reaction relative to Pd. The ratio increases
significantly to a factor in excess of 10 at potentials below −0.5 V
SCE. Importantly, the reversible potential for proton reduction is
2+
2+
−0.389 V SCE and both Pt and Pd species are subject to reduc-
tion by the parasitic hydrogen produced at more negative potential.
+
For pH 2.5, diffusion limited proton reduction (3.2 mmol/L H O )
3
would yield an upper bound for the interface H2 concentration
of 1.6 mmol/L. The relative concentration of the reactive species,
3 mmol/L Pt2+, 1 mmol/L Pd and 1.6 mmol/L H2 and prior reports
that the kinetics of Pd reduction tends to be faster than that of Pt
2+
Fig. 3. (a) Compositional analyses by EDS and (b) FE-SEM images for Pt100−x−yNixPdy alloy films electrodeposited for 1800 s as a function of growth potential.

S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
8941
Table 1
Grain size and film thickness of the Pt100−x−yNixPdy films grown on Au seeded sub-
strate at various potential for 1800 s.
Growth potential (V) SCE
Grain size from
1 1 peak (nm)
Film thickness (nm)
1
Pt100−x−yNixPdy at −0.1 V
Pt100−x−yNixPdy at −0.2 V
Pt100−x−yNixPdy at −0.3 V
Pt100−x−yNixPdy at −0.4 V
Pt100−x−yNixPdy at −0.5 V
Pt100−x−yNixPdy at −0.6 V
Pt100−x−yNixPdy at −0.7 V
Pt100−x−yNixPdy at −0.8 V
10.99
10.54
9.80
9.62
8.25
7.77
7.29
6.54
260 ± 1.0
320 ± 11.0
410 ± 13.4
590 ± 77.6
140 ± 7.1
210 ± 8.4
500 ± 5.5
Peeled off
[
45,46] may account for the selective fall off in Pd incorporated in
the alloy deposit grown at potentials below −0.5 V SCE. This expla-
nation is congruent with the surface morphologies observed by
SEM where, as shown in Fig. 3b, a limited volume fraction of large
spherical particle precipitates are evident in films grown between
−
0.5 V SCE and −0.7 V SCE. Qualitative EDS indicates a high con-
centration of Pd in these particles compared to the neighboring
surface.
Furthermore, the corresponding film thickness measured by
FE-SEM examination of cleaved cross-sections reveals a distinct
transition in the same potential regime. As shown in Table 1 for
a fixed deposition time, the film thickness increased from 260 nm
to 590 nm as the growth potential decreased from −0.1 V SCE to
−
0.4 V SCE. At −0.5 VSCEthe thickness suddenlydropped to140 nm
at −0.5 V SCE. The thickness increased again to 500 nm at −0.7 V
SCE. The film deposited at −0.8 V SCE was found to exhibit a mud
crack surface pattern, with typical crack separation distance of
5
␮m. During cross-sectioning the film spalled from the substrate
preventing an accurate determination of its thickness.
The width of the (1 1 1) diffraction peak was used to estimate
grain size of the films using the Scherrer equation. As shown in
Table 1 the grain size decreased monotonically from 11 nm to 6 nm
as the growth potential decreased from −0.1 V SCE to −0.8 V SCE.
3
.2. ORR studies on Pt1
NixPdy
Fig. 4. (a) Cyclic voltammetries in Ar saturated 0.1 mol/L HClO4 and (b) RDE mea-
surements for ORR in O2 saturated 0.1 mol/L HClO4 with 1600 rpm on a series of
Pt100−x−yNixPdy films grown for 30 s.
00−x−y
The kinetics of the ORR on Pt1
NixPdy films were exam-
00−x−y
ined using a RDE. In contrast to the “thick” films used for structural
and compositional evaluation, “thin” films were grown on the Pt
RDE for 30 s at the specified potentials. Prior work with Pt alloy
co-deposition indicates that there is little change in film com-
position with thickness [35,36] so that the relationship between
structure and composition established with the “thick films” can
be applied to the “thin films”. Prior to measuring the ORR kinet-
ics, the electroactive surface area of the deposited films was
assayed by voltammetrically measuring the Hupd charge in Ar sat-
Following saturation of the electrolyte with O2 the positive-
going voltammetric curves for ORR on a series of Pt1
x y
Ni Pd
00−x−y
thin films were collected using a RDE and normalized by the geo-
2
metrically projected area (0.196 cm ). The ORR kinetics on the
Pt1
x y
Ni Pd alloy films exhibit half-wave potentials (E1/2) that
00−x−y
are 26–65 mV greater than that of pure Pt grown at −0.6 V SCE as
2
shown in Fig. 4b. A diffusion limited current density of ∼6 mA/cm
is observed below 0.7 V RHE followed by slight decrease below
urated 0.1 mol/L HClO . Quasi steady-state voltammograms for the
0.3 V RHE due to contributions from the anodic oxidation of Hupd.
4
Pt1
NixPdy alloy films are shown in Fig. 4a. The average charge
The mass transfer corrected kinetics can be obtained using the
Levich–Koutecky equation where ikinetic is the kinetic current at
a given potential and ilimited is the limiting current.
00−x−y
measured for the anodic and cathodic waves between 0.039 V RHE
and 0.408 V RHE is given in Table 2. Interestingly the Hupd charge
tends to increase with the as-desposited Ni content in the films
although the trend is not monotonic. For example, the Hupd charge
for the film grown at −0.6 V SCE is less than that for the more
Pt-Pd rich films grown −0.3 V SCE. Although surface roughness is
expected to increase with as-deposited Ni content due to subse-
quent dealloying, the as-deposited film thickness might be less
due to the parasitic effects of proton reduction alluded to early.
Pd additions may also lead to contributions arising from hydrogen
absorption and thereby an overestimate of the electroactive area.
1
1
1
= i
+ i
(1)
imeasured
kinetic
limited
Graphically, the slope of the inverse of the measured current ver-
sus the square root of the reciprocal of the RDE rotation rate (not
−
shown) is congruent with diffusion limited 4e reduction of O2 to
H O consistent with other work on Pt and Pt alloys [48,49]. The
2
mass transport corrected ORR kinetics for each alloy at 0.9 V RHE is
summarized in Table 2. Compared to electrodeposited Pt the appar-
ent ORR kinetics for the alloys are substantially greater, in some
cases as much as a factor of 6–8 times. However consideration of the
Hupd charge suggests that much of this increase is due to expansion
of the electroactive area that accompanies Ni dealloying. As shown
In the first approximation the Hupd charge for the Pt
NixPdy
100−x−y
2
films was normalized by 210 ␮C/cm ; the commonly accepted
value for a monolayer charge of adsorbed hydrogen on polycrys-
talline Pt [47].

8
942
S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
Table 2
ORR kinetics at 0.9 V RHE as a function of growth potentials (the films were grown for 30 s at each growth potential).
Growth potential (V) SCE
ikinetic, geo (mA/cm2)
Hupd charge (mC/cm2)
Hupd area (cm2)
ikinetic, Hupd (mA/cm2)
Pt at −0.3 V
3.23
1.58
2.98
3.2
6.84
5.39
5.31
6.22
9.81
7.79
10.14
5.02
13.31
1.50
0.48
1.20
1.09
2.11
1.45
1.58
1.42
2.10
1.68
1.81
2.11
3.0
7.13
2.26
5.73
5.18
10.07
6.89
7.50
6.78
10.01
8.01
8.63
0.45
0.7
Pt at −0.6 V
Pt100−x−yNixPdy at −0.10 V
Pt100−x−yNixPdy at −0.20 V
Pt100−x−yNixPdy at −0.30 V
Pt100−x−yNixPdy at −0.35 V
Pt100−x−yNixPdy at −0.40 V
Pt100−x−yNixPdy at −0.45 V
Pt100−x−yNixPdy at −0.50 V
Pt100−x−yNixPdy at −0.55 V
Pt100−x−yNixPdy at −0.60 V
Pt100−x−yNixPdy at −0.65 V
Pt100−x−yNixPdy at −0.70 V
0.52
0.62
0.68
0.78
0.71
0.92
0.98
0.97
1.18
0.50
0.91
10.03
14.68
in Table 2, Hupd normalization results in the ORR kinetics increas-
ing as the growth potential is made more negative. The maximum
area corrected ORR enhancement factor is reduced to a factor of 1.7
times Pt for the Pt25Ni73Pd2 film grown at −0.6 V SCE.
0.9 V RHE is on the order of 1.7 as shown in Fig. 5c and d. This is
slightly less than the factor of 2.5 observed for binary Pt72Ni28 films
produced by the same means [11].
The influence of dealloying-derived area expansion on the ORR
kinetics was examined for a series of alloy films grown to dif-
A more detailed comparison of the ORR behavior of Pt57Ni27Pd16
film grown at −0.35 V SCE (a Pt Ni analog) to that of Pt grown
ferent thicknesses at −0.6 V SCE. The H
electroactive area of
3
upd
at a similar potential, −0.3 V SCE, is given in Fig. 5. Since the Pt
deposition reaction proceeds at the diffusion limit the total Pt mass
loading for both films is identical, the alloy film only being thicker
due to co-deposition of Ni and Pd. Voltammetry in the absence of O2
reveals practically identical Hupd charge and, by inference, a sim-
ilar electroactive area. Subsequent ORR measurements reveal the
enhanced activity of the alloy. The kinetic enhancement factor at
each was determined prior to ORR measurements. The half-wave
potential of the ORR response exhibits a monotonic shift with
the as-deposited film thickness as shown in Fig. 6a. A compli-
cation is clearly evident for the thickest films where the shape
of the curve deviates sharply at 0.9 V RHE presumably due to
some complex hydrodynamic interactions between the thick deal-
loyed material and the electrolyte. The Levich equation was used
Fig. 5. (a) Cyclic voltammetries in Ar saturated 0.1 mol/L HClO4 on Pt and Pt57Ni27Pd16 films grown at −0.3 V SCE and −0.35 V SCE for 30 s, (b) RDE measurements, (c) mass
transport corrected Tafel plot, and (d) Hupd area corrected Tafel plot for ORR in O2 saturated 0.1 mol/L HClO4 with 1600 rpm on Pt and Pt57Ni27Pd16 films grown at −0.3 V SCE
and −0.35 V SCE for 30 s.

S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
8943
The apparent loss for the thicker films may be attributed to the
inability of O2 to access the recessed dealloyed regions that are
assayed by the Hupd measurements. The effect is particularly accen-
tuated at potentials at or below the half-wave potential where a
substantial O2 depletion gradient exists in the interfacial region. A
more suitable measurement of the intrinsic activity of the thicker
dealloyed layer is obtained at smaller overpotentials where con-
volution of the O2 depletion with the porous electrode is less
severe.
3.3. Comparison of ORR on Pt1
NixPdy and Pt
Nix
00−x−y
100−x
The ORR electrocatalytic activity of Pt25Ni73Pd2 was evaluated
by comparing it with Pt20Ni80 and Pt films electrodeposited under
similar conditions. Electrodeposited Ni-rich Pt1 Nix alloy films
00−x
(
x = ∼80%) have been reported to exhibit 4-fold higher, H
area
upd
corrected ORR kinetics compared to pure Pt films. For this study
Pt, Pt20Ni80, and Pt25Ni73Pd2 films were deposited on a Pt RDE at
−
0.6 V SCE for 30 s. For all three samples Pt deposition occurs under
transport limited conditions thereby resulting in the same Pt mass
loading for a given deposition time. Subsequent Ni dealloying dur-
ing voltammetric cycling in 0.1 mol/L HClO leads to area expansion
4
as reflected by changes in the Hupd with cycling. The steady-state
voltammogram for the three materials in Ar saturated 0.1 mol/L
HClO4 are shown in Fig. 7a. The shape of the Hupd region is notably
different for the Pd alloy and may reflect a contribution from H
absorption. Comparison of the expansion in charge associated with
the oxide formation and reduction suggests that area change in
the Hupd region is still dominated by hydrogen adsorption versus
absorption as has been reported for Pt-rich Pt–Pd alloys [43,44].
Measurement of the ORR, shown in Fig. 7b, reveals a positive shift
in the half-wave potential for Pt25Ni73Pd2 and Pt20Ni80, relative to
Pt. The respective shifts of 61 mV and 66 mV translate to an increase
in the mass transport corrected rate constant (ikinetic, geo) by a factor
of 5 and 6, respectively, when evaluated at 0.9 V RHE as summarized
in Table 3. Thus, for a fixed Pt loading, formation and subsequent
dealloying of the thin alloy films yields an increase in the ORR Pt
mass activity that is more than a factor of 5. The area normalized
ORR reactivity, based on the Hupd charge, reduces this to a factor
of 1.4 and 2.3, respectively. Mass transport corrected Tafel plots
for the three specimens with fixed Pt loading are given in Fig. 7c
while the Hupd area corrected versions are shown in Fig. 7d. Clearly,
the enhanced ORR kinetics for the alloys arise from a combination
of enhanced electrocatalysis and area expansion due to dealloying.
The area normalized result indicates that the intrinsic ORR reac-
tivity spans a range of 0.7–1.7 depending on the alloy composition
and potential.
3.4. Stability of Pt20Ni80 and Pt25Ni73Pd2
The dealloying of electrodeposited Pt25Ni73Pd2 and Pt20Ni80
was evaluated by monitoring changes in the cyclic voltammetric
behavior in Ar saturated 0.1 mol/L HClO . At the same time X-ray
4
diffraction was used to characterize the resulting structural evo-
lution of the films. The “thick” electrodeposited films (>100 nm)
grown for 1800 s on Au substrates were used to ensure strong X-
ray scattering power for the conventional laboratory diffraction
experiment.
As shown in Fig. 8a, the initial cyclic voltammogram for
Pt25Ni73Pd2 reveals a potential independent current associated
with the growth, dissolution and capacitance of the passive nickel
oxide film. Previous work indicated that cycling the potential into
the Hupd region enables activation of Ni dissolution in 0.1 mol/L
Fig. 6. (a) RDE measurements, (b) mass transport corrected Tafel plot, and (c) Hupd
area corrected Tafel plot for the ORR in O2 saturated 0.1 mol/L HClO4 with 1600 rpm
on Pt25Ni73Pd2 films grown at −0.6 V SCE as a function of growth time.
to isolate the kinetic contribution and, as shown in Fig. 6b, the
mass transport corrected ORR kinetics (i
at 0.9 V RHE)
kinetic, geo
increased monotonically from 7.5 mA/cm2 to 15 mA/cm2 as the
film growth time increased from 10 s to 120 s. Thereafter the ORR
kinetics scatter between 15 mA/cm2 and 20 mA/cm . The corre-
sponding Hupd normalized ORR kinetics (ikinetic, Hupd ) are shown
2
HClO with significant dealloying occurring with continued cycling
4
in Fig. 6c with a maximum activity observed for the film grown
for 120 s followed by decreasing performance for the thicker films.
[32,50]. The voltammetric characteristics of Hupd and oxide for-
mation and reduction on Pt become increasingly evident. The

8
944
S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
Fig. 7. (a) Cyclic voltammetries in Ar saturated 0.1 mol/L HClO4 on electrodeposited Pt, Pt20Ni80, and Pt25Ni73Pd2 films, (b) positive-going polarization curves, (c) mass
transport corrected Tafel plot, and (d) Hupd area corrected Tafel plot for the ORR in O2 saturated 0.1 mol/L HClO4 with 1600 rpm on electrodeposited Pt, Pt20Ni80, and
Pt25Ni73Pd2 films.
magnitude of the Hupd wave for Pt25Ni73Pd2 increases, passing
through a maximum around 20 cycles before settling towards a
quasi steady-state result that reflects the behavior of Pt. FE-SEM
images of the film surface before and after cycling are shown in
Fig. 8b and c, and Ni dealloying is clearly evident by the formation
of cracks and very fine scale porosity. Cross-sections of the film
reveal dealloying leads to decrease in film thickness from 210 nm
to 100 nm. X-ray diffraction shown in Fig. 8d reveals a shift in the
Consistent with a previous study [11] similar trends were
observed for a Pt20Ni80 film although interestingly the area expan-
sion accompanying dealloying was more substantial (4×) than for
the Pd containing alloy as revealed by comparing the H
and
upd
Pt oxidation and reduction waves in Fig. 9a compared to Fig. 8a.
SEM examination of the cross-section reveals a decrease in the film
thickness from 446 nm to 358 nm upon dealloying. The more sub-
stantial area increase might be associated with the thicker film
observed for the Pt–Ni system 446 nm relative to the Pt–Ni–Pd
films (210 nm) grown at this potential. The more extensive deal-
loying of Pt20Ni80 compared to Pt25Ni73Pd2 may also be ascribed
to the lower combined Pt–Pd content, 20% versus 27%, in the
as-deposited alloy. For the magnifications examined, the FE-SEM
images of the as-deposited (Fig. 9b) and dealloyed (Fig. 9c) mate-
rial were effectively indistinguishable from those for Pt25Ni73Pd2.
X-ray diffraction shown in Fig. 9d reveals a shift in the position
◦
position of the 1 1 1 and 2 0 0 peaks to lower angles, from 43.59
◦
◦
◦
to 41.05 for the 1 1 1 and from 50.88 to 47.65 for the 2 0 0. Com-
parison to the 1 1 1 peak position for pure Pt grown under similar
conditions indicates that significant Ni remains within the deal-
loyed Pt25Ni73Pd2 film. This was qualitatively confirmed by EDS
measurements of the porous dealloyed layer. The lattice parame-
ter shift from 0.3593 nm to 0.3805 nm for the (1 1 1) suggests that
the dealloyed reminant has a composition close to Pt70Ni30 stoi-
chiometry.
◦
◦
of the 1 1 1 peak from 42.89 to 41.20 while the 2 0 0 shifts from
Table 3
Comparison of ORR kinetics at 0.9 V RHE and 0.95 V RHE on Pt25Ni73Pd2, Pt20Ni80, and Pt films electrodeposited at −0.6 V SCE.
Materials
ikinetic, geo at 0.9 V RHE (mA/cm2)
Hupd charge (mC/cm2)
Hupd area (cm2)
ikinetic, Hupd at 0.9 V RHE (mA/cm2)
Pt
2.43
15.04
12.33
0.48
1.29
1.70
2.26
6.14
8.08
1.07
2.45
1.53
Pt20Ni80
Pt25Ni73Pd2
Materials
ikinetic, geo at 0.95 V RHE (mA/cm2)
Hupd charge (mC/cm2)
Hupd area (cm2)
ikinetic, Hupd at 0.95 V RHE (mA/cm2)
Pt
0.48
2.76
2.18
0.48
1.29
1.70
2.26
6.14
8.08
0.21
0.45
0.27
Pt20Ni80
Pt25Ni73Pd2

S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
8945
Fig. 9. (a) Multicyclic voltammogram for Pt20Ni80 films grown for 1800 s at −0.6 V
SCE, FE-SEM images of Pt20Ni80 film (b) as-deposited, (c) after CVs, and (d) X-ray
diffraction patterns of Pt20Ni80 films before and after CVs in Ar saturated 0.1 mol/L
HClO4.
Fig. 8. (a) Multicyclic voltammogram for Pt25Ni73Pd2 films grown for 1800 s at
0.6 V SCE, FE-SEM images of Pt25Ni73Pd2 film (b) as-deposited, (c) after CVs, and
d) X-ray diffraction patterns of Pt25Ni73Pd2 film before and after CVs in Ar saturated
.1 mol/L HClO4.
−
(
0

8
946
S.-M. Hwang et al. / Electrochimica Acta 55 (2010) 8938–8946
◦
◦
5
0.20 to 47.54 . This corresponds to a change in the lattice param-
[13] U.A. Paulus, A. Wokaun, G.G. Scherer, T.J. Schmidt, V. Stamenkovic, N.M.
Markovic, P.N. Ross, Electrochim. Acta 47 (2002) 3787.
eter from 0.3649 nm to 0.3792 nm for the 1 1 1 plane spacing and
from 0.3632 nm to 0.3822 nm for the 2 0 0. Comparison with lattice
parameter data for electrodeposited Pt–Ni films [37] indicates the
[
14] K. Jayasayee, V.A. Dam, T. Verhoeven, S. Celebi, F.A. de Bruijn, J. Phys. Chem. C
1
13 (2009) 20371.
[15] F.H.B. Lima, J.R.C. Salgado, E.R. Gonzalez, E.A. Ticianelli, J. Electrochem. Soc. 154
2007) A369.
(
dealloyed electrode has a composition close to Pt65Ni35
.
[
[
16] A. Seo, J. Lee, K. Han, H. Kim, Electrochim. Acta 52 (2006) 1603.
17] J.F. Whitacre, T.I. Valdez, S.R. Narayanan, Electrochim. Acta 53 (2008) 3680.
Ni-rich Pt–(Pd)–Ni alloys show significant Ni dealloying during
extended voltammetric cycling at room temperature. However the
dealloying appears to be limited by a threshold that likely repre-
sents a convolution of kinetic (surface diffusion) and geometric
[18] A. Bonakdarpour, K. Stevens, G.D. Vernstrom, R. Atanasoski, A.K. Schmoeckel,
M.K. Debe, J.R. Dahn, Electrochim. Acta 53 (2007) 688.
[
19] J.H. Cho, W.J. Roh, D.K. Kim, J.B. Yoon, J.H. Choy, H.S. Kim, J. Chem. Soc., Faraday
Trans. 94 (1998) 2835.
(
i.e. percolation) factors [51,52] associated with an average alloy
[20] G.E. Ramirez-Caballero, Y. Ma, R. Callejas-Tovar, P.B. Balbuena, Phys. Chem.
Chem. Phys. 12 (2010) 2209.
composition in the range Pt70Ni30 as determined by a Vegard’s
law approximation. The actual composition of the dealloyed mate-
rials is probably slightly different due to capillarity and related
stress effects associated with dealloying and the remnant mate-
rial.
[
[
[
21] E. Antolini, Energy Environ. Sci. 2 (2009) 915.
22] M.H. Shao, K. Sasaki, P. Liu, R.R. Adzic, Z. Phys. Chem. 221 (2007) 1175.
23] B. Lim, M.J. Jiang, P.H.C. Camargo, E.C. Cho, J. Tao, X.M. Lu, Y.M. Zhu, Y.A. Xia,
Science 324 (2009) 1302.
[
[
[
[
24] G. Ramos-Sanchez, H. Yee-Madeira, O. Solorza-Feria, Int. J. Hydrogen Energy 33
(
2008) 3596.
25] X.P. Wang, N. Kariuki, J.T. Vaughey, J. Goodpaster, R. Kumar, D.J. Myers, J. Elec-
trochem. Soc. 155 (2008) B602.
26] W.M. Wang, D. Zheng, C. Du, Z.Q. Zou, X.G. Zhang, B.J. Xia, H. Yang, D.L. Akins,
J. Power Sources 167 (2007) 243.
4
. Conclusions
The kinetics of the ORR were measured on a series of elec-
27] J. Kim, T. Momma, T. Osaka, J. Power Sources 189 (2009) 909.
trodeposited Pt1
NixPdy films. The alloys were produced by
[28] S. Guerin, B.E. Hayden, C.E. Lee, C. Mormiche, A.E. Russell, J. Phys. Chem. B 110
2006) 14355.
29] R. Wang, H. Li, S. Ji, H. Wang, Z. Lei, Electrochim. Acta 55 (2010) 1519.
[30] M.T. Paffett, J.G. Beery, S. Gottesfeld, J. Electrochem. Soc. 135 (1988) 1431.
[31] S. Gottesfeld, in: M.T.M. Koper (Ed.), Fuel Cell Catalysis, John Wiley & Sons, 2009
00−x−y
(
an electrodeposition process that involved a combination of under-
NixPdy
[
potential and overpotential reactions. For Pt-rich Pt1
00−x−y
alloy films (x < 0.65) Ni co-deposition occurred at underpotentials
while for Ni-rich films (x > 0.65) the deposition proceeded at over-
potentials. Potentiostatic deposition enables control of the alloy
composition that is a monotonic function of the growth potential.
For thin film with fixed Pt loading the Pt25Ni73Pd2 and Pt20Ni80
alloys exhibit a 5–6-fold enhancement of the mass transport cor-
rected ORR kinetics. However, based on Hupd measurements a
significant portion of this is associated with surface area expansion
that accompanies partial dealloying of Ni. Using Hupd to correct for
(
Chapter 1).
32] M.K. Debe, A.K. Schmoeckel, G.D. Vernstrorn, R. Atanasoski, J. Power Sources
61 (2006) 1002.
[
1
[33] J.H. Wee, K.Y. Lee, S.H. Kim, J. Power Sources 165 (2007) 667.
[34] E.J. Taylor, E.B. Anderson, N.R.K. Vilambi, J. Electrochem. Soc. 139 (1992) L45.
[
35] S.M. Ayyadurai, Y.-S. Choi, P. Ganesan, S.P. Kumaraguru, B. Popov, J. Elec-
trochem. Soc. 154 (2007) B1063.
[36] J.J. Mallett, U. Bertocci, J.E. Bonevich, T.P. Moffat, J. Electrochem. Soc. 156 (2009)
D531.
[
37] J.J. Mallett, E.B. Svedberg, J.E. Bonevich, A.J. Shapiro, W.F. Egelhoff, T.P. Moffat,
J. Electrochem. Soc. 155 (2008) D1.
the electroactive area indicate that the Pt25Ni73Pd2 and Pt20Ni80
alloys have an intrinsic activity toward ORR at 0.9 V RHE that is 1.4
and 2.3-fold greater, respectively, than Pt.
[38] J.J. Mallett, E.B. Svedberg, S. Sayan, A.J. Shapiro, L. Wielunski, T.E. Madey, P.J.
Chen, W.F. Egelhoff, T.P. Moffat, Electrochem. Solid-State Lett. 8 (2005) C15.
[
39] J.J. Mallett, E.B. Svedberg, S. Sayan, A.J. Shapiro, L. Wielunski, T.E. Madey, W.F.
Egelhoff, T.P. Moffat, Electrochem. Solid-State Lett. 7 (2004) C121.
[
40] Z. Samard zˇ ija, Ro zˇ manF K. Zˇ ., S. Kobe, Mater. Charact. 60 (2009) 1241.
References
[41] K.J.J. Mayrhofer, D. Strmcnik, B.B. Blizanac, V. Stamenkovic, M. Arenz, N.M.
Markovic, Electrochim. Acta 53 (2008) 3181.
[
[
42] Y. Liu, S. Bliznakov, N. Dimitrov, J. Phys. Chem. C 113 (2009) 12362.
43] M. Grden, A. Czerwinski, J. Golimowski, E. Bulska, B. Krasnodebska-Ostrega, R.
Marassi, S. Zamponi, J. Electroanal. Chem. 460 (1999) 30.
[
[
[
1] R.R. Adzic, J. Zhnag, K. Sasaki, M.B. Vukmirovic, M. Shao, J.X. Wang, A.U. Nilekar,
M. Mavrikakis, J.A. Valerio, F. Uribe, Top. Catal. 46 (2007) 249.
2] A.U. Nilekar, Y. Xu, J.L. Zhang, M.B. Vukmirovic, K. Sasaki, R.R. Adzic, M.
Mavrikakis, Top. Catal. 46 (2007) 276.
3] J. Zhang, F.H.B. Lima, M. Shao, K. Sasaki, J.X. Wnag, J. Hanson, R.R. Adzic, J. Phys.
Chem. B 109 (2005) 22701.
[
44] M. Grden, A. Piascik, Z. Koczorowski, A. Czerwinski, J. Electroanal. Chem. 532
(
2002) 35.
[
[
45] R. Hoyer, L.A. Kibler, D.M. Kolb, Electrochim. Acta 49 (2003) 63.
46] H.F. Waibel, M. Kleinert, L.A. Kibler, D.M. Kolb, Electrochim. Acta 47 (2002)
[
[
4] L. Xiong, A. Manthiram, J. Electrochem. Soc. 152 (2005) A697.
5] T. Toda, H. Igarashi, H. Uchida, M. Watanabe, J. Electrochem. Soc. 146 (1999)
1461.
[
47] F.C. Nart, W. Vielstich, in: W. Vielstich, H.A. Gasteiger, A. Lamm (Eds.), Handbook
of Fuel Cells—Fundamentals, Technology and Applications, vol. 2, John Wiley &
Sons, 2003 (Chapter 21).
3
750.
[
[
[
[
6] V.R. Stamenkovic, B. Fowler, B.S. Mun, G.F. Wang, P.N. Ross, C.A. Lucas, N.M.
Markovic, Science 315 (2007) 493.
7] V.R. Stamenkovic, B.S. Mun, K.J.J. Mayrhofer, P.N. Ross, N.M. Markovic, J. Ross-
meisl, J. Greeley, J. Nørskov, Angew. Chem. Int. Ed. 45 (2006) 2897.
8] V.R. Stamenkovic, B.S. Mun, M. Arenz, K.J.J. Mayrhofer, C.A. Lucas, G.F. Wang,
P.N. Ross, N.M. Markovic, Nat. Mater. 6 (2007) 241.
[
[
[
48] V. Stamenkovic, T.J. Schmidt, P.N. Ross, N.M. Markovic, J. Electroanal. Chem. 554
(
2003) 191.
49] V. Stamenkovic, T.J. Schmidt, P.N. Ross, N.M. Markovic, J. Phys. Chem. B 106
2002) 11970.
50] M.Y. Rusanova, P. Polaskova, M. Muzikar, W.R. Fawcett, Electrochim. Acta 51
2006) 3097.
(
9] P. Mani, P. Srivastava, P. Strasser, J. Phys. Chem. C 112 (2008) 2770.
(
[
[
[
10] P. Strasser, S. Koh, J. Greeley, Phys. Chem. Chem. Phys. 10 (2008) 3670.
11] T.P. Moffat, J.J. Mallett, S.-M. Hwang, J. Electrochem. Soc. 156 (2009) B238.
12] S. Mukerjee, S. Srinivasan, J. Electroanal. Chem. 357 (1993) 201.
[
[
51] D.M. Artymowicz, J. Erlebacher, R.C. Newman, Philos. Mag. 89 (2009) 1663.
52] J. Snyder, J. Erlebacher, J. Electrochem. Soc. 157 (2010) C125.