durability study of pt-pd/c as pemfc cathode...
TRANSCRIPT
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 6
Avai lab le a t www.sc iencedi rec t .com
j ourna l homepage : www.e lsev ier . com/ loca te /he
Durability study of Pt–Pd/C as PEMFC cathode catalyst
Zhi-Min Zhou a,b, Zhi-Gang Shao a,*, Xiao-Ping Qin a,b, Xu-Guang Chen a,Zi-Dong Wei c, Bao-Lian Yi a
a Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Fuel Cell R&D Center, Dalian Liaoning 116023, Chinab Graduate School of Chinese Academy of Sciences, Beijing 100049, Chinac Chongqing University, Chongqing 400045, China
a r t i c l e i n f o
Article history:
Received 9 October 2009
Received in revised form
10 December 2009
Accepted 10 December 2009
Available online 31 December 2009
Keywords:
PEMFC
Pt–Pd/C
Catalyst
Durability
Potential cycling
Dynamic loading
* Corresponding author. Tel.: þ86 411 843791E-mail address: [email protected] (Z.-G
0360-3199/$ – see front matter ª 2009 Profesdoi:10.1016/j.ijhydene.2009.12.056
a b s t r a c t
In this paper, Pt–Pd/C and Pt/C catalysts were evaluated and compared. The catalysts were
evaluated as oxygen reduction reaction (ORR) catalysts in half cell test under potential
cycling, and cathode catalysts in single cell test under dynamic loading simulating the
vehicle operation. Physical and electrochemical techniques were applied to investigate the
structure, performance and durability of those catalysts. The electrochemical active
surface area (ECA) loss, particle size distribution, polarization behavior and electrochem-
istry impedance spectroscopy (EIS) suggested that the Pt–Pd/C showed a better durability
than Pt/C.
ª 2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.
1. Introduction water-heat management, and materials deterioration, etc [2].
Proton exchange membrane fuel cells (PEMFCs) are consid-
ered as promising alternative power sources for trans-
portation due to the high energy efficiency and clean
utilization. Within the last decades, much progress has been
made for PEMFCs in materials, system design, manufacturing
and application. Up to now, durability is thought to be one of
the major problems hindering the commercialization of
PEMFCs besides cost, hydrogen storage, etc. While the lifetime
for stationary used PEMFCs has already exceeded 3 years, the
lifetime for transportation used PEMFCs is still below 3000 h,
which is nearly half of the 2010/2015 target lifetime 5000 h set
by U.S. DOE [1]. It is known that PEMFCs will gradually degrade
in performance due to many factors including operation,
53.. Shao).sor T. Nejat Veziroglu. Pu
Membrane electrode assemble (MEA) is the most important
component of PEMFCs. The deterioration of MEA directly leads
to the degradation of PEMFCs. One of the most important
causes for MEA deterioration is the degradation of catalyst in
the cathode side due to the high voltage, low pH and oxidation
environment [3]. For the most popular Pt/C catalyst, it is
reported that the nano-sized platinum supported on carbon
black will gradually grow larger during long-term operation,
thus reduce the electrochemical active area (ECA) and resulted
in an irreversible performance loss [4].
This degradation will be even severer when PEMFCs are
operated under dynamic loadings. Borup investigated the
degradation behavior of single cell under cycling dynamic
loading and found that the growth extent of cathode catalyst
blished by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 61720
particles was greater than that observed during steady-state
testing [5]. The half cell electrochemical potential cycling test
also revealed that higher upper potential and scan frequency
lead to a higher degradation rate [5]. Darling and Meyer
calculated the Pt dissolution under potential cycling and found
platinum will dissolve rapidly when transitioning from low to
high potentials [6]. The proposed degradation mechanisms of
Pt/C catalysts include coarsening of Pt particles, Pt dissolution,
carbon corrosion [7,8] and loss of contact of ionomer and Pt [9].
Much work has been tried to improve the performance and
durability of the catalyst. One approach is using alloy or
bimetallic catalysts [10]. However, if cheap transition metals
are used, they may dissolve and move into the polymer phase.
Zhang studied the Pt–Au/C catalyst and demonstrated that
this bimetallic catalyst exhibited fairly high durability under
half cell potential cycling test [11]. However, gold is still a very
expensive metal. In the past a few years, much attention has
been paid to Palladium and its alloys. Adzic and coauthors
made a series of study on the ORR activity of Pt-monolayer on
model Pd crystal surface substrate as well as Pd-monolayer on
model Pt crystal surface substrate through experimental and
computational methods [12–17]. They reported that when Pd
is added to Pt, a change in the electronic structure may lead to
a higher ORR activity in half cell evaluation. Li prepared Pt–Pd/
C catalyst using ethylene glycol (EG) reflux reduction method
and found that this kind of catalyst poses higher ORR activity
than Pt/C catalyst in half cell test [18]. Lim synthesized Pd–Pt
bimetallic nanodendrites consisting of a dense array of Pt
branches on a Pd core and found that the Pd–Pt nanodendrites
were two and half times more active on Pt mass for the ORR
than the state-of-the-art Pt/C catalyst [19]. Peng and Yang
using a sequential synthetic method prepared carbon sup-
ported Platinum-on-palladium bimetallic heterogeneous
nanostructure, in which 3-nm Pt particles grew on the
surfaces of 5-nm Pd nanoparticles. The electrochemical study
showed not only enhancement in electrocatalytic activity for
ORR but also improved stability in comparison to a commer-
cial platinum catalyst [20]. Based on the summaries above,
one can notice that Pt–Pd/C is a promising catalyst for PEMFCs.
In order to meet the practical applications, more investiga-
tions should be taken on its durability, especially under full
fuel cell test conditions under dynamic loading.
In this paper, the Pt–Pd/C and Pt/C catalyst was prepared by
an ethylene glycol reduction process. The designed metal
weight ratio is 50% which was preferred in practical PEMFC
application. In order to get a highly dispersion of fine metal
particles with such metal weight ratio, a high surface area
carbon black was used as the support. The Pt–Pd/C catalyst was
then compared with home-made Pt/C and commercial Pt/C
catalysts in terms of durability in both half cell and single cell
under dynamic loading simulating the vehicle operation.
2. Experimental
2.1. Catalyst preparation
The synthesis of the Pt–Pd/C catalyst was following the
ethylene glycol (EG) reflux reduction method described in Ref.
[18]. A high surface area carbon black was used as the support.
The carbon black bought from Dalian Sunrise Power Co., Ltd.
(P.R. China) has a high specific surface area of 770 m2 g�1.
Briefly, calculated amount of the carbon black was dispersed
in an EG solution by adding H2PtCl6 and PdCl2 under vigorous
stirring. The pH of the solution was then adjusted to 12 with
2 M NaOH. The mixture was kept at 130 �C for 3 h under the
protection of high purified nitrogen. After washing and drying
the Pt–Pd/C catalyst was obtained. The designed atom ratio
was Pt:Pd¼ 3:1. A Pt/C catalyst was prepared following the
same way. A commercial Pt/C catalyst from Tanaka Kikinzoku
Co., Ltd (Japan) was also used for comparison. For conve-
nience, the catalysts were denoted as Pt3Pd/S770, Pt/S770 and
TKK 46.6%, respectively.
2.2. Half cell test
The half cell test was conducted on a thin film rotating disk
electrode (TFRDE) in a three electrode electrochemical system.
It is a simple and fast way widely used to investigate the
activity and stability of fuel cell catalysts [21]. The TFRDE was
prepared as following: 5 mg catalyst was dispersed in
a mixture of 50 mL 5% Nafion solution and 4 ml ethanol. The
mixture was agitated by ultrasonic for 30 min to form an ink.
10 mL of this ink was then dropped on a glassy carbon rotating
disk electrode and dried to yield a thin film electrode. The
geometric area of the TFRDE is 0.1256 cm2 and the metal
loading is ca. 45 mg cm�2. During the durability test, the TFRDE
was subjected to a potential cycling test. Specifically, the
catalyst was repeatedly scanned from 0.6 V to 1.2 V versus
SHE at a rate of 50 mV s�1 in 0.5 M H2SO4 deaerated with high
purified N2. Cyclic voltammetry (CV) was recorded at the
beginning and every 100 cycles of the potential sweep by a CHI
600C potentistat from 0 to 1.2 V versus SHE at 50 mV s�1. The
ECA can be obtained by integrating the CV profile after sub-
tracting the double layer from the hydrogen adsorption peak
of each CV curve. The charge related to a monolayer H
adsorption to the catalyst crystal was supposed to be
210 mC cm�2.
The ORR curves were measured in an O2 saturated 0.5 M
H2SO4 from 1.0 V to 0.241 V versus SHE at 5 mV s�1 with
a rotating speed of 1600 rpm.
2.3. Single cell test
Based on the half cell test results, single cell test was con-
ducted to the Pt3Pd/S770 and TKK 46.6% catalysts on an Arbin
FCTS BT2000 model. The metal loading of all the Gas Diffusion
Electrode (GDE) was 0.5 mg cm�2. The MEA was manufactured
by hot pressing the GDE onto each side of a Nafion�-115
membrane. The single cell with an active area of 5 cm2 was
operated for 400 h under a cyclic dynamic loading. Within one
single cycle the current was varying in the following
sequence: 0 mA cm�2 for 60 s, 200 mA cm�2 for 60 s,
400 mA cm�2 for 180 s, 300 mA cm�2 for 60 s, 600 mA cm�2 for
60 s and 200 mA cm�2 for 60 s. The time of one cycle is 480 s, so
each single cell experienced 3000 cycles during the test. The
dynamic loading test was carried on an Arbin BT 2000 electric
load. A schematic presenting the controlled currents and
voltage reactions is given in Fig. 1. The cell temperature was
85 �C, which is between the typical 60–90 �C operation range
Fig. 1 – Schematic of the dynamic loading in single cell
dynamic loading test.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 6 1721
and high to accelerate the degradation rate. The reactant gas
was controlled as 35 ml min�1/800 ml min�1 for H2/air at
0.2 MPa. Polarization curves were recorded periodically under
the same operation parameters. After the I–V plot was recor-
ded, the in-situ electrochemical impedance spectroscopy (EIS)
was measured. The perturbation amplitude for the sine signal
was 20 mV over a frequency range from 10 kHz to 100 mHz.
CV was also periodically recorded to monitor the degradation
of ECA after purging the cathode with N2. The scan was from
0 to 1.0 V vs. SHE at 50 mV s�1. All the electrochemical char-
acterizations on single cell were carried on a potentiostat
Model 2273 A from EG&G Princeton Applied Research.
2.4. Physical characterization
X-ray diffraction (XRD) measurements were performed on
a PANalytical X’Pert Pro diffractometer with nickel-filtered
Cuka radiation at 40 kV and 40 mA. The lattice parameter A0
and the mean diameter of catalysts were calculated by Bragg
and Scherrer formula [22].
Transmission electron microscopy investigations of all
catalysts before and after the half cell test were carried on
a JEOL JEM-2011 electron microscope operating at 120 kV.
Fig. 2 – XRD patterns of Pt–Pd/C and Pt/C catalysts.
Particle size distribution diagrams were obtained by counting
all particles on each TEM picture.
Inductively coupled plasma-atomic emission spectroscopy
(ICP-AES) analyses were conducted for the element analysis
on Leeman Plasma–Spec–I equipment.
3. Results and discussion
3.1. Physical characterization
The powder XRD patterns and TEM images are shown in
Figs. 2 and 3, respectively. The lattice parameter (A0) and the
Fig. 3 – TEM images of fresh Pt–Pd/C and Pt/C catalysts: (a)
Pt3Pd/S770; (b) Pt/S770; (c) TKK 46.6.
Table 1 – Physical characterization of catalysts.
Catalyst Peak pos. 220/(2q) Lattice parameterA0 (nm)
Mean particlesize d (nm)
Metal weightpercent (ICP-AES)
Atomic ratioPt/Pd (ICP-AES)
Pt3Pd/S770 67.88 0.390 2.97 48 3.3
Pt/S770 67.26 0.393 1.60 48 –
TKK 46.6% 67.23 0.394 2.51 45 –
Fig. 4 – Particle size histograms of fresh Pt–Pd/C and Pt/C
catalysts (a) Pt3Pd/S770; (b) Pt/S770; (c) TKK 46.6%
measured from TEM images in Fig. 3.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 61722
mean particle size derived from (220) peak are summarized in
Table 1. All catalysts showed a face-centered cubic (FCC)
structure. The (220) peak of Pt3Pd/S770 is positively shifted
with a shortening of crystal parameter, indicating that Pd has
an obvious effect on the Pt crystal size. The particle size of
Pt3Pd/S770 increased by nearly 1.37 nm compared with
Pt/S770 due to the high precipitation tendency of Pd when
pH> 7. The atomic ratio of Pt/Pd is 3.3 close to the designed
value. The particle morphology of fresh catalysts is shown in
Fig. 3. Compared with Pt/S770, the metal particles for Pt3Pd/
S770 are larger and more agglomerates can be found, thus
resulted in a wider distribution of Pt3Pd/S770 than Pt/S770
presented in Fig. 4. The mean diameter of metal particles
measured by TEM is little larger than the value measured by
XRD, because particles smaller than 1 nm are hard to count in
TEM [23].
Fig. 5 – Electrochemical characterization of fresh Pt–Pd/C
and Pt/C catalysts: (a) CV curves; (b) ORR curves.
Fig. 6 – Comparison of normalized ECA for various catalysts
during potential cycling.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 6 1723
3.2. Electrochemistry test in half cell
The CV and ORR curves of as-prepared Pt3Pd/S770 and Pt/S770
catalysts are shown in Fig. 5. The hydrogen adsorption/
desorption peak area of the Pt3Pd/S770 declines as its particle
is larger. Another difference in Fig. 5(a) is the Pt–O reduction
peak of Pt3Pd/S770 shifts by 0.024 V compared to Pt/S770. It is
believed that a shift of the Pt–O reduction peak to higher
potential reflects a decrease of the adsorption strength for the
oxygen containing species on the surface, e.g. OHads [24,25].
Although Pt weight and ECA of Pt3Pd/S770 is less than Pt/S770,
its ORR curve is very close to the latter, which means Pt3Pd/
S770 poses a high activity comparable to pure Pt/S770.
However, the improvement of the ORR as Li[18] described was
not found, which may result from different carbon supports
and metal/carbon ratios.
Potential cycling is a widely used way to test the stability
and durability of fuel cell catalysts [24,26,27]. The potential
window chosen in this paper is 0.6–1.2 V vs. SHE, which is
a typical operation region of H2/O2 PEMFC. Fig. 6 shows the
normalized ECA variation of different catalyst during the half
cell potential cycling test. The ECA was normalized to the
initial ECA before potential cycling. As the results indicated,
all catalysts gradually decrease in ECA as the potential scan
proceeds. After the half cell test, nearly 55% of the initial ECA
of Pt3Pd/S770 has been lost, compared to 68% of Pt/S770 and
65% of TKK 46.6%. It can be clearly seen that Pt3Pd/S770 has
the lowest decreasing rate, while Pt/S770 has the highest one.
The ECA degradation is listed in Table 2.
Table 2 – Catalyst degradation in half cell.
Catalyst ECA loss Mean particle size by TEM (nm)
Before test After test
Pt3Pd/S770 55% 3.18 4.52
Pt/S770 68% 2.24 5.36
TKK 46.6% 65% 2.75 5.27
Fig. 7(a)–(c) presents the TEM images of all the catalysts after
the potential cycling test. The size distribution derived from
Fig. 7 is presented in Fig. 8. Particle sizes before and after the
potential cycling are also listed in Table 2. After the potential
cycling all particles grow larger and many agglomerates can be
clearly observed. However, the metal particle of Pt3Pd/S770
becomes the smallest one and the Pt/S770 becomes the largest
one. After the potential cycling, the size distribution of Pt/S770
becomes much wider with a tail toward to large particle size. It
is reported this shape of the particle size distribution may
ascribe to the micrometer-scale platinum dissolution–diffu-
sion–precipitation mechanism [23]. However, the same distri-
bution behavior is not found on Pt3Pd/S770, which suggests that
Fig. 7 – TEM images of Pt–Pd/C and Pt/C catalysts after
potential cycling: (a) Pt3Pd/S770; (b) Pt/S770; (c) TKK 46.6%.
Fig. 8 – Particle size histograms of Pt–Pd/C and Pt/C
catalysts after potential cycling (a) Pt3Pd/S770; (b) Pt/S770;
(c) TKK 46.6% measured from TEM images in Fig. 7.
Fig. 9 – Comparison of polarization behavior after different
operation time for various catalysts: (a) Pt3Pd/S770; (b) TKK
46.6%.
Table 3 – Catalyst degradation in single cell.
Catalyst Voltage at 200 mA cm�2 (V) ECA loss
Before test After test
Pt3Pd/S770 0.758 0.749 37%
TKK 46.6% 0.748 0.703 58%
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 61724
the addition of Pd may change the degradation mechanism of
the catalyst. Mayrhofer [28] observed that Pt particles detached
from the carbon support under potential cycling. So ECA should
depend on two terms: one is the number of particles survived
on carbon support and the other is the size of those particles.
Actually, one could see that the particles on carbon are much
less in Fig. 8b than Fig. 8a. A further discussion will be con-
ducted later.
3.3. Durability test in single cell
Based on the results of half cell test, Pt3Pd/S770 and TKK 46.6%
were chosen to be evaluated under single cell condition,
which is more similar to the real operation conditions of
vehicles. Both cells experienced open circuit voltage, mediate
current density and high current density in each cycle which
was repeated 3000 times during the whole 400 h test. The
polarization curves of cells with Pt3Pd/S770 and TKK 46.6% are
shown in Fig. 9. As listed in Table 3, the initial performance of
Fig. 11 – Comparison of normalized ECA for various
catalysts cathode under single cell operation.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 6 1725
the two cells are close to each other, though the cell using
Pt3Pd/S770 are slightly better. The voltage at 200 mA cm�2 is
0.758 V for cell with Pt3Pd/S770 and 0.748 V for cell with TKK
46.6%. After 400 h operation, the performances of both cells
are degraded. The voltage at 200 mA cm�2 decreased to 0.749 V
for cell with Pt3Pd/S770 and 0.703 V for cell with TKK 46.6%.
The Nyquist plots at 200 mA cm�2 are shown in Fig. 10. All
plots present a single arc indicating the electrode process is
dominated by the interfacial kinetics of the ORR process [29]. It
is believed that the high-frequency intercept of the impedance
arc on the real axis Ro represents the total ohmic resistance of
the cell, while the diameter of the arc, Rct, is a measure of the
charge transfer resistance of the ORR. From Fig. 10, it can
be seen that the Ro of both cells did not change much during the
operation suggesting the100%RH humidification waskept well.
However, the Rct of both cells did increase obviously after the
test. The Rct of cell using TKK 46.6% increased much larger than
the Rct of cell using Pt3Pd/S770. This may be an explanation of
higher performance in the latter cell after the long time test.
The ECA change of single cell with running time is also
shown in Fig. 11 and Table 3. It is observed that the degrada-
tion phenomenon is similar to what found in half cell test. The
ECA declines faster in the cell using TKK 46.6% than in the cell
of Pt3Pd/S770. And the ECA loss may make a major contribu-
tion to the drop in ORR kinetics.
3.4. Proposed degradation mechanisms
The single cell test results along with the half cell test results
demonstrated that addition of Pd may improve the durability
of the Pt/C catalysts for PEMFC cathode. The direct evidences
are the growth of particle size and ECA loss. As pointed out in
the Introduction section, there are several mechanisms sug-
gested to explain this degradation. Shao-Horn and coauthors
discussed a few different mechanisms through experimental
and theoretical computation, and conclude that the sub-
monolayer dissolution of Pt nanoparticles may be the gov-
erning factor for the surface area loss [7]. In details, the Pt can
be oxidized to Pt–O between ca. 0.9 and 1.2 V. The oxidized Pt
then can be chemically dissolved into the solution, thus
causing a Pt loss or precipitation by reduction. A mathematic
Fig. 10 – Comparison of EIS for single cell using various
catalysts.
model of this mechanism built by Darling [6] also demon-
strated that the oxidation played an important role in the
dissolution of Pt. Although this mechanism needs more
investigations to be identified, it is acceptable to use it to
explain the durability enhancement of Pt–Pd/C catalyst in this
paper. Gasteiger studied the ORR activity at different Pt crystal
surface and supposed that the specific ORR activity can be
reflected by the position of Pt–O reduction peak [25]. When the
peak position is more positive, it means the catalyst binds
oxygen species more weakly and thus the specific ORR activity
is higher. As observed in Section 3.2, the position of Pt–O
reduction peak is shifted to more positive, which means the
Pt–Pd binds oxygen species more weakly than pure Pt. It is
reported that the addition of Pd may change the electronic
structure and d-band center of the catalyst surface, thus
weaken the adsorption of oxygen species [16]. So, here we
proposed a possible mechanism as following: the introduction
of Pd can alleviate the oxidation of catalyst, reduce its disso-
lution and finally improve the durability.
There are also other possible mechanisms. Holby and
coauthors studied the impact of particle size distribution on
related degradation rate by experiment and theoretic simu-
lation [30]. They found that larger particles are more stable
than smaller ones due to lower Gibbs–Thomson energy.
Adding Pd made the catalyst size larger than pure Pt but
without obvious deterioration of ORR activity. This may at the
same time enhance the stability of the catalyst. Carbon
corrosion also can cause coalescence of particles and
detachment of Pt particles from the support, just like what the
Fig. 8b of Pt/S770 reveals. Whether the addition of Pd could
alleviate the carbon corrosion is not clear yet. And the
degradation mechanisms still needs more work to be clarified.
4. Conclusions
The Pt–Pd/C catalyst was prepared and compared with home-
made and commercial Pt/C catalysts. Both half cell test and
single cell test demonstrated that this bimetal catalyst
showed a comparable performance and better durability than
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 5 ( 2 0 1 0 ) 1 7 1 9 – 1 7 2 61726
conventional Pt/C catalyst. The possible reasons for the
durability enhancement are discussed. Firstly, Pd may modify
the electronic structure of catalyst surface and thus alleviate
the dissolution of it; secondly, increased particle size may also
improve the resistance to degradation; lastly, less detachment
from carbon support has been observed for Pt–Pd catalyst.
Further work is still needed to clarify the degradation
mechanisms.
Acknowledgements
This work was financially supported by the National High
Technology Research and Development Program of China (863
Program, No. 2007AA05Z137, 2008AA11A105) and the National
Natural Science Foundations of China (No. 20636060, 20936008).
r e f e r e n c e s
[1] Cleghorn SJC, Mayfield DK, Moore DA, Moore JC, Rusch G,Sherman TW, et al. A polymer electrolyte fuel cell life test: 3years of continuous operation. J Power Sources 2006;158:446–54.
[2] Schmittinger W, Vahidi A. A review of the main parametersinfluencing long-term performance and durability of PEMfuel cells. J Power Sources 2008;180:1–14.
[3] Borup R, Meyers J, Pivovar B, Kim YS, Mukundan R,Garland N, et al. Scientific aspects of polymer electrolyte fuelcell durability and degradation. Chem Rev 2007.
[4] Xie J, Wood DL, More KL, Atanassov P, Borup RL.Microstructural changes of membrane electrode assembliesduring PEFC durability testing at high humidity conditions.J Electrochem Soc 2005;152:A1011–20.
[5] Borup RL, Davey JR, Garzon FH, Wood DL, Inbody MA. PEMfuel cell electrocatalyst durability measurements. J PowerSources 2006;163:76–81.
[6] Darling RM, Meyers JP. Mathematical model of platinummovement in PEM fuel cells. J Electrochem Soc 2005;152:A242–7.
[7] Shao-Horn Y, Sheng WC, Chen S, Ferreira PJ, Holby EF,Morgan D. Instability of supported platinum nanoparticles inlow-temperature fuel cells. Top Catal 2007;46:285–305.
[8] Tang H, Qi Z, Ramani M, Elter JF. PEM fuel cell cathode carboncorrosion due to the formation of air/fuel boundary at theanode. J Power Sources 2006;158:1306–12.
[9] Zhang F-Y, Advani SG, Prasad AK, Boggs ME, Sullivan SP,Beebe Jr TP. Quantitative characterization of catalyst layerdegradation in PEM fuel cells by X-ray photoelectronspectroscopy. Electrochim Acta 2009;54:4025–30.
[10] Antolini E, Salgado JRC, Gonzalez ER. The stability of Pt–M(M¼ first row transition metal) alloy catalysts and its effecton the activity in low temperature fuel cells – a literaturereview and tests on a Pt–Co catalyst. J Power Sources 2006;160:957–68.
[11] Zhang J, Sasaki K, Sutter E, Adzic RR. Stabilization ofplatinum oxygen-reduction electrocatalysts using goldclusters. Science 2007;315:220–2.
[12] Zhang J, Mo Y, Vukmirovic MB, Klie R, Sasaki K, Adzic RR.Platinum monolayer electrocatalysts for O-2 reduction: Pt
monolayer on Pd(111) and on carbon-supported Pdnanoparticles. J Phys Chem B 2004;108:10955–64.
[13] Zhang J, Vukmirovic MB, Sasaki K, Nilekar AU, Mavrikakis M,Adzic RR. Mixed-metal Pt monolayer electrocatalysts forenhanced oxygen reduction kinetics. J Am Chem Soc 2005;127:12480–1.
[14] Zhang JL, Vukmirovic MB, Xu Y, Mavrikakis M, Adzic RR.Controlling the catalytic activity of platinum-monolayerelectrocatalysts for oxygen reduction with differentsubstrates. Angew Chem Int Ed 2005;44:2132–5.
[15] Shao MH, Huang T, Liu P, Zhang J, Sasaki K, Vukmirovic MB,et al. Palladium monolayer and palladium alloyelectrocatalysts for oxygen reduction. Langmuir 2006;22:10409–15.
[16] Shao MH, Liu P, Zhang JL, Adzic R. Origin of enhancedactivity in palladium alloy electrocatalysts for oxygenreduction reaction. J Phys Chem B 2007;111:6772–5.
[17] Adzic RR, Lima FHB. In: Wolf Vielstich HY, Gasteiger HubertA, editors. Handbook of fuel cells – fundamentals,technology and applications, vol. 5. John Wiley & Sons. Ltd.;2009. p. 5–17.
[18] Li HQ, Sun GQ, Li N, Sun SG, Su DS, Xin Q. Design andpreparation of highly active Pt–Pd/C catalyst for the oxygenreduction reaction. J Phys Chem C 2007;111:5605–17.
[19] Lim B, Jiang M, Camargo PHC, Cho EC, Tao J, Lu X, et al. Pd–Ptbimetallic nanodendrites with high activity for oxygenreduction. Science 2009;324:1302–5.
[20] Peng Z, Yang H. Synthesis and oxygen reductionelectrocatalytic property of Pt-on-Pd bimetallicheteronanostructures. J Am Chem Soc 2009;131.
[21] Shao Y, Yin G, Gao Y. Understanding and approaches for thedurability issues of Pt-based catalysts for PEM fuel cell.J Power Sources 2007;171:558–66.
[22] Radmilovic V, Gasteiger HA, Ross PN. Structure andchemical-composition of a supported Pt–Ru electrocatalystfor methanol oxidation. J Catal 1995;154:98–106.
[23] Ferreira PJ, la O’G J, Shao-Horn Y, Morgan D, Makharia R,Kocha S, et al. Instability of Pt/C electrocatalysts in protonexchange membrane fuel cells – a mechanistic investigation.J Electrochem Soc 2005;152. A2256–A2a71.
[24] Arico AS, Stassi A, Modica E, Ornelas R, Gatto I,Passalacqua E, et al. Performance and degradation of hightemperature polymer electrolyte fuel cell catalysts. J PowerSources 2008;178:525–36.
[25] Gasteiger HA, Kocha SS, Sompalli B, Wagner FT. Activitybenchmarks and requirements for Pt, Pt-alloy, and non-Ptoxygen reduction catalysts for PEMFCs. Appl Catal B Environ2005;56:9–35.
[26] Yu P, Pemberton M, Plasse P. PtCo/C cathode catalyst forimproved durability in PEMFCs. J Power Sources 2005;144:11–20.
[27] Merzougui B, Swathirajan S. Rotating disk electrodeinvestigations of fuel cell catalyst degradation due topotential cycling in acid electrolyte. J Electrochem Soc 2006;153:A2220–6.
[28] Mayrhofer KJJ, Meier JC, Ashton SJ, Wiberg GKH, Kraus F,Hanzlik M, et al. Fuel cell catalyst degradation on thenanoscale. Electrochem Commun 2008;10:1144–7.
[29] Yuan X, Wang H, Colin Sun J, Zhang J. AC impedancetechnique in PEM fuel cell diagnosis – a review. Int JHydrogen Energy 2007;32:4365–80.
[30] Holby EF, Sheng W, Shao-Horn Y, Morgan D. Pt nanoparticlestability in PEM fuel cells: influence of particle sizedistribution and crossover hydrogen. Energy Environ Sci2009;2:865–71.