Electrochimica Acta 59 (2012) 264–269
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Stability characteristics of Pt1 Ni1 /C as cathode catalysts in membrane electrode
assembly of polymer electrolyte membrane fuel cell
Yong-Hun Cho a , Tae-Yeol Jeon b , Sung Jong Yoo c , Kug-Seung Lee c , Minjeh Ahn b , Ok-Hee Kim b ,
Yoon-Hwan Cho b , Ju Wan Lim b , Namgee Jung b , Won-Sub Yoon d , Heeman Choe a , Yung-Eun Sung b,∗,1
a
School of Advanced Materials Engineering, Kookmin University, Seoul 136-702, South Korea
School of Chemical and Biological Engineering, College of Engineering, Seoul National University (SNU), Seoul, South Korea
c
Center for Fuel Cell Research, Korea Institute of Science and Technology, Seoul 136-791, South Korea
d
Department of Energy Science, Sungkyunkwan University, Suwon 440-746, South Korea
b
a r t i c l e
i n f o
Article history:
Received 6 July 2011
Received in revised form 19 October 2011
Accepted 21 October 2011
Available online 29 October 2011
Keywords:
Polymer electrolyte membrane fuel cell
(PEMFC)
Platinum–nickel alloy
X-ray photoelectron spectroscopy (XPS)
element mapping
Constant current operation
a b s t r a c t
To understand the difference in degradation characteristics between carbon-supported platinum (Pt/C)
and platinum–nickel alloy (Pt1 Ni1 /C) cathode catalysts in membrane electrode assemblies (MEAs) of a
polymer electrolyte membrane fuel cell (PEMFC), constant current operation of MEA in a single cell was
conducted for 1100 h. A significant change in cell potential for the Pt1 Ni1 /C MEA was observed throughout
the test. High-resolution transmission electron microscopy showed that sintering and detachment of
metal particles in the Pt1 Ni1 /C catalyst occurred more sparingly than in the Pt/C catalyst. Instead, X-ray
photoelectron spectroscopy element mapping revealed dissolution of Ni atoms in the Pt1 Ni1 catalysts
even when the Pt1 Ni1 /C catalyst used in the MEA was well synthesized.
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Studies on the lifespan of membrane electrode assemblies
(MEAs) have been conducted extensively over the past decade for
the purpose of polymer electrolyte membrane fuel cell (PEMFC)
commercialization. From prior investigations, it is obvious that
decay of MEA cathode catalysts is one of the most crucial factors in
the life expectancy of the PEMFC [1–3]. Use of carbon-supported Pt
alloy (Pt alloy/C) in conjunction with transition metals such as Co,
Ni, and Fe as cathode catalysts in PEMFCs reduces Pt consumption
and increases the oxygen reduction reaction (ORR) activity. Therefore, minimizing the degradation of MEA performance caused by
deactivation of the Pt alloy is necessary for commercializing PEMFCs [4,5]. The half-cell test and accelerated durability test (ADT) are
important for evaluating the electrochemical catalyst performance.
Previous studies on catalyst degradation have been based primarily on the half-cell test and on ADTs such as load cycling, start/stop
∗ Corresponding author. Tel.: +82 2 880 1889; fax: +82 2 888 1604.
E-mail address: ysung@snu.ac.kr (Y.-E. Sung).
1
International Society of Electrochemistry member.
0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.10.060
cycling, and RH cycling [6,7]. Those studies examined the effect
of Pt alloy/C on the degradation of PEMFC performance, but they
provided inconsistent results [8–12]. Those inconsistencies may
result from the different operating conditions of each durability
test. Nevertheless, it is widely accepted that Pt alloy/C catalysts
are electrochemically unstable compared to carbon-supported Pt
(Pt/C) catalysts under half-cell and ADT conditions. From a practical standpoint, the MEA test is the ultimate evaluation criterion for
characterization of catalysts [13–15]. Investigations of the durability of catalysts in MEAs are more complex and difficult than those
in half-cells because the temperature, pressure, and gas flow conditions play pivotal roles in determining accurate MEA performance.
Generally, constant current test has been recognized as traditional
durability test mode. In this paper, to overcome prior limitations,
catalyst-coated membranes (CCM) used as MEAs were fabricated
using Pt/C or Pt1 Ni1 /C as cathode catalysts, and they were operated
for 1100 h in constant current mode using a single cell. A comparison is reported between the performance degradation behaviors
of MEAs containing Pt/C or Pt1 Ni1 /C as cathode catalysts. In addition, X-ray photoelectron spectroscopy element mapping was used
to characterize degradation of the catalyst layer over a wide area
(above 0.78 mm2 ); this is a novel approach for analyzing degradation of the catalyst layer in a PEMFC. The unique transition of cell
Y.-H. Cho et al. / Electrochimica Acta 59 (2012) 264–269
265
Fig. 1. (a) TEM image of the carbon-supported Pt–Ni alloy nanoparticles, (b) powder X-ray patterns and (c) area ratios of various states fitted with XPS spectra of Pt 4f and
Ni 2p of Pt1 Ni1 /C. (d) Cyclic voltammograms obtained in 0.5 M H2 SO4 solution at a scan rate of 50 mV s−1 .
potential as a function of time on the Pt1 Ni1 /C MEA will be discussed
in comparison with the Pt/C MEA.
2. Experimental
Pt/C (40 wt.%, Johnson Matthey) and Pt1 Ni1 /C (40 wt.%, synthesized) were used as cathode catalysts in order to compare their
performance degradation behaviors. Pt/C was also used as an anode
catalyst in both MEAs. The 40 wt.% Pt1 Ni1 /C electrocatalyst was
synthesized by borohydride reduction in anhydrous ethanol containing dissolved sodium acetate (C2 H3 NaO2 ), as described in our
previous report [16]. The prepared Pt1 Ni1 /C and as-received Pt/C
were analyzed using X-ray diffraction (XRD, Rigaku D/Max2500)
with Cu K␣ radiation (40 kV, 200 mA). High-resolution transmission electron microscopy (HR-TEM) using a JEOL EM-2000 EXII
microscope at a 200 kV accelerating potential was conducted to
measure changes in the size and distribution of Pt/C and Pt1 Ni1 /C
particles. XPS was used using an Al K␣ source (ESCALAB 250 XPS
spectrometer, VG Scientifics) to determine chemical states of the
synthesized Pt1 Ni1 alloy. Binding energies were calibrated with
respect to C (1s) at 285 eV. Experimental data were curve-fitted
using XPSPEAK4.1 software. For the electrochemical characterization of synthesized Pt1 Ni1 /C and as-received Pt/C powders, the
catalyst slurry for the half-cell test was ultrasonically prepared
using 200 ␮l of deionized water, 572 ␮l of 5 wt.% Nafion solution (Aldrich), and 8 ml of isopropyl alcohol per 0.1 g of catalyst.
Cyclic voltammogram for half-cell tests was performed in a threeelectrode electrochemical cell that consisted of a glassy carbon
electrode coated with catalyst slurry for the working electrode,
saturated calomel electrode (SCE) for a reference electrode, and
Pt wire for a counter electrode. A cyclic voltammogram was examined in argon-saturated 0.5 M H2 SO4 with a 50 mV s−1 scan rate
at 20 ◦ C in order to determine the electrochemical active surface
area (ESA). The catalyst inks for MEAs were blended ultrasonically
with Nafion solution (5 wt.%, Aldrich), deionized water, and isopropyl alcohol. The CCMs for MEAs were fabricated as a polymer
electrolyte through a spraying method using Nafion 212 (Dupont),
as reported previously [17]. The active area of the electrode was
5 cm2 , and the metal catalyst loading was 0.2 mg cm−2 . The MEAs
were assembled from gas diffusion layers (GDLs), including a microporous layer and gaskets, using a single cell (CNL-PEM005-01, CNL
Energy). The long-term test as a function of time was performed for
1100 h using a fuel cell test station (CNL Energy) at a constant current density of 400 mA cm−2 , and the load was not disconnected
during the experiment duration without replacing the reactant
gases (H2 or air). The single-cell was operated using fully humidified
H2 gas and air on the anode and cathode with a stoichiometric ratio
of 2:2.5, respectively, and the cell temperature was maintained at
70 ◦ C under ambient pressure. The MEAs used with the Pt/C and
Pt1 Ni1 /C cathode catalysts are hereafter referred to as Pt/C MEA
and Pt1 Ni1 /C MEA, respectively. XPS element mapping of Pt1 Ni1 /C
MEA was performed to determine atomic ratios of Pt and Ni on
the catalyst layer surface before and after the long-term operation
in an ultra high vacuum (UHV) multipurpose surface analysis system (SIGMA PROBE, Thermo, UK) at base pressures <10−10 mbar.
The photoelectron spectra were excited by an Al K␣ (1486.6 eV)
anode operating at a constant power of 100 W (15 kV and 10 mA).
The mapping area of catalyst layer surfaces in the Pt1 Ni1 /C MEA
was 0.78 mm2 (975 ␮m × 800 ␮m), and the number of pixels was
1248 (39 × 32). Beam size pixel sizes were 15 ␮m and 625 ␮m2
(25 ␮m × 25 ␮m), respectively.
3. Results and discussion
As-prepared Pt1 Ni1 /C powder was characterized by HR-TEM, as
shown in Fig. 1(a). Pt1 Ni1 nanoparticles were well dispersed on
the carbon support with an average particle size of approximately
2.5 nm. As seen in Fig. 1(b), the (2 2 0) peak of the as-prepared
266
Y.-H. Cho et al. / Electrochimica Acta 59 (2012) 264–269
Fig. 2. Cell potentials of the single cell with (a) Pt/C and (b) Pt1 Ni1 /C MEAs as a
function of time on load operated at constant current of 400 mA cm−2 for 1100 h.
Pt1 Ni1 /C (70.6◦ ) shifted to a higher angle compared to that of
the commercial Pt/C (67.7◦ ), which suggests that Pt1 Ni1 was well
alloyed. The degree of alloying for Pt1 Ni1 was calculated using the
(2 2 0) peak position according to Vegard’s law, yielding 69% [18].
This degree of alloying is relatively high compared with previously
reported results [19]. Fig. 1(c) shows the ratio of the Pt and Ni
metallic phases to oxides. The ratio of Pt and Ni metallic phases
is higher than that in a previous study [20]. As the alloying level
increases, the resulting alloy approaches toward homogeneous
alloying of Pt and Ni without Ni-loss or unalloyed Ni phases such
as NiO and Ni(OH)2 , as seen in Fig. 1(c). The ESA of the as-prepared
Pt1 Ni1 /C (28.2 m2 /gPt–Ni ) was larger than that of commercial Pt/C
(24.4 m2 /gPt ) due to an increase in the surface/volume ratio and
the relatively small particle size for the Pt–Ni alloy nanoparticles,
as shown in Fig. 1(d). Consequently, all data shown in Fig. 1 led to
the conclusion that the as-prepared Pt1 Ni1 /C was well synthesized
with a high degree of alloying.
Fig. 2 shows performance degradation trends of (a) Pt/C and (b)
Pt1 Ni1 /C as cathode catalysts in MEAs during long-term operation
(1100 h). For the Pt/C MEA (Fig. 2(a)), cell potential was initially
approximately 0.69 V and decreased to approximately 0.45 V in the
final stage of the long-term test (1100 h). For Pt1 Ni1 /C (Fig. 2(b)),
the initial cell potential of 0.67 V decreased to a final potential of
0.35 V. In other words, the Pt/C MEA showed the performance decay
of approximately 35%, whereas the performance of the Pt1 Ni1 /C
MEA declined by roughly 48% at a constant operating current of
400 mA cm−2 , which indicates that Pt1 Ni1 /C is less electrochemically stable than the Pt/C catalyst. As shown in Fig. 2, the initial
cell potential of the Pt1 Ni1 /C MEA was slightly lower than that
of the Pt/C MEA. We think that comparing the durability of Pt/C
and Pt1Ni1/C MEA would be worth although the composition of
ionomer and solvent in catalyst ink and the annealing condition
of synthesized catalysts were not optimized. Fig. 2 shows that
the performances of both Pt/C MEA and Pt1 Ni1 /C MEA drastically
decreased during the initial 300 h. After 300 h of operation, the cell
potential peak for the Pt/C MEA repeated itself at rest times during the replacement of gases until the end of the long-term test,
whereas the performance of the Pt1 Ni1 /C MEA remained steady
from 300 h to 700 h and then continuously declined until 1100 h.
Pivovar and co-workers [21] demonstrated that the performance
decay could be classified as recoverable or unrecoverable. The
unrecoverable values could be attributed to irreversible changes
in MEA compositions such as the loss of electrochemical surface
area and decrease of ionic conductivity in the membrane and catalyst layer. Conversely, the recoverable values could be related
to reversible changes in the MEA such as partial flooding, platinum surface oxidation, and catalyst poison by impurities. The cell
potential recovery for the Pt/C MEA at approximately 300 h suggests that severe unrecoverable degradation of Pt/C did not occur
after 300 h and that subsequent performance was subsequently
maintained. This also suggests that the potential of the Pt/C MEA
declined owing to recoverable properties after 300 h. In contrast,
continuously declining cell potential with unrecoverable performance decay was observed for the Pt1 Ni1 /C MEA after 700 h, which
indicates that irreversible destruction of Pt1 Ni1 /C continued over
time. In addition, the Pt1 Ni1 /C MEA experienced larger fluctuations
in cell potential than the Pt/C MEA over the entire test period,
as shown in Fig. 2(b). This result indicates that Pt1 Ni1 /C used as
the cathode catalyst in MEA is less electrochemically stable than
the Pt/C catalyst and causes continuous degradation of the MEA
induced by deactivation of the alloy catalyst, which manifests as
unrecoverable catalyst degradation. On the other hand, recovered
performance of Pt/C MEA and Pt1 Ni1 /C MEA during the durability test period might be caused by removing accumulated water
and impurities in the catalyst layer owing to shut-down and restart procedure. Fig. 3(a and c) shows HR-TEM images of Pt/C and
Pt1 Ni1 /C from the MEAs before the long-term operation. The mean
particle sizes of Pt and Pt1 Ni1 were 4.0 and 2.7 nm, respectively,
as measured from HR-TEM images. Further, the metal particles in
both the MEAs were highly dispersed on the carbon supports. However, some Pt and PtNi nanoparticles agglomerated near junctions
among the primary carbon particles, whereas certain areas on the
surfaces of carbon supports had no metal particles. These defects
in catalysts might occur due to ultrasonication during the fabrication of catalyst inks. Fig. 3(b) shows a HR-TEM image of Pt/C
from the MEA after the long-term test. The image shows scattered
Pt cluster particles (particle agglomeration) on carbon supports as
well as detached Pt particles; the average particle size increased to
5.9 nm. Remarkably, as shown in Fig. 3(d), the mean particle size
of Pt1 Ni1 /C in MEA after the test was 3.7 nm, with apparent structure of the catalyst remaining essentially the same; there was no
particle aggregation owing to Pt1 Ni1 particles falling from carbon
supports. These results clearly indicate that Pt1 Ni1 particles have
more resistance to the sintering of metal particles on the carbon
supports than pure Pt particles. A similar result prompted authors
to refer to “the anchor effects of Ni to Pt on carbon supports” by
Popov and co-workers [22]. Fig. 4 shows the particle size distributions of Pt/C and Pt1 Ni1 /C before and after the long-term test.
For Pt/C, as seen in Fig. 4(a), the width of the curve for Pt particle
size distribution before the test was relatively narrow, whereas the
width of the curve for the Pt/C MEA broadened and shifted to a
Y.-H. Cho et al. / Electrochimica Acta 59 (2012) 264–269
267
Fig. 3. HR-TEM micrographs of (a–b) Pt/C and (c–d) Pt1 Ni1 /C before and after long-term operation, respectively.
higher value after the test. This result is originated from the typical
particle growth mechanisms of Ostwald ripening and particle coalescence. Conversely, the particle size distribution of Pt1 Ni1 /C was
narrower than that of Pt/C before the test, as shown in Fig. 4(c).
In addition, the width of the curve for Pt1 Ni1 /C after the test also
was smaller than that of Pt/C after the test. Fig. 5 compares the
atomic ratios of surface Pt and Ni in the Pt1 Ni1 /C catalyst layer MEA
using XPS element mapping (a) before and (b) after the long-term
operation. XPS element mapping visualized the atomic ratios of Pt
and Ni and also showed that the screen turns red when the catalyst layer surface has a higher Ni ratio. Fig. 5(a) shows the XPS
mapping image of the Pt1 Ni1 /C MEA catalyst layer surface before
testing; the atomic ratio of Pt to Ni was 41.1:58.9. XPS showed that
the low fraction of Pt on the catalyst surface was due to a difference in the reduction rates of Pt and Ni during the synthesis of the
Pt1 Ni1 alloy phase. Since Pt has a faster reduction time than Ni,
Pt1 Ni1 nanoparticles have a higher surface composition of Ni. On
the other hand, Fig. 5(b) shows an XPS element mapping image of
the Pt1 Ni1 /C MEA catalyst layer surface after testing; the atomic
ratio of Pt to Ni was 17.4:82.6. Changes in the ratio of Pt to Ni
were caused by dissolution of Ni from the Pt1 Ni1 alloy phase. The
cause for the increased Ni ratio might be that dissolved Ni was redeposited on the catalyst layer surface. It was previously reported
that dissolution of transition metals such as Ni and Co from a Pt
alloy catalyst is a common phenomenon in PEMFC operation [23].
It is well known that dissolved metal ion from the PEMFC catalyst
migrates to the polymer electrolyte or ionomer on the catalyst layer
and is subsequently re-deposited [24,25]. However, our XPS mapping results clearly show the re-deposition of Ni on the catalyst
layer surface with no observation of Ni migration to the polymer
electrolyte, suggesting that dissolved Ni moves to the gas diffusion
layer, but not to the polymer electrolyte. Nevertheless, it is difficult to determine exactly which phase (PtNi, Ni, Pt) of the catalysts
will be placed on the catalyst layer surface, though the mapping
image after testing obviously shows that Ni was dissolved from the
Pt1 Ni1 alloy. XPS was performed to determine the de-alloying of
Pt1 Ni1 catalyst surface before and after durability test. Fig. 6(a–c)
shows the Pt 4f core-level peaks of Pt/C MEA before durability test,
Pt1 Ni1 /C MEA before durability test, and Pt1 Ni1 /C MEA after dura-
Table 1
The binding energy from XPS results with Pt chemical states in each MEA.
Pt/C MEA before test
Pt1 Ni1 /C MEA before test
Pt1 Ni1 /C MEA after test
Oxidation state
Binding energy(eV)
Pt
PtO
PtO2
Pt
PtO
PtO2
Pt
PtO
PtO2
70.98
72.00
72.73
71.14
72.13
72.97
71.03
72.12
72.89
268
Y.-H. Cho et al. / Electrochimica Acta 59 (2012) 264–269
Fig. 4. Histograms of particle size distributions of (a–b) Pt/C and (c–d) Pt1 Ni1 /C before and after long-term operation, respectively.
bility test. From curve-fitting, it was noticed that binding energy of
metallic Pt in Pt1 Ni1 /C MEA after durability test was significantly
shifted from 71.14 eV to 71.03 eV. This shift in binding energy to
a lower energy means that Ni dissolution from Pt1 Ni1 nanoparticles was generated during the durability test. The binding energies
from XPS results with Pt chemical states in each MEA are listed in
Table 1. Based on these results, the Pt/C MEA had showed the initial
severe performance decay induced by the general particle growth
mechanism for approximately 300 h, after which it maintained performance with the recoverable performance degradation until the
conclusion of the test. The continuous decrease in performance
of the Pt1 Ni1 /C MEA is due to dissolution of Ni from Pt1 Ni1 alloy
nanoparticles, indicating that XPS element mapping is a powerful
tool for studying degradation of Pt alloy catalysts.
Fig. 5. Surface images of the Pt1 Ni1 /C catalyst layer using XPS element mapping (a) before and (b) after long-term operation. (c) Photograph of the fabricated CCM as a MEA.
Y.-H. Cho et al. / Electrochimica Acta 59 (2012) 264–269
269
detachment of metal particles. In addition, surfaces of the MEA electrodes can be characterized using XPS element mapping for the
study of Pt1 Ni1 /C electrocatalyst degradation.
Acknowledgments
This work was supported by the Technology Innovation
Program (10029897, Development of MEA fabrication process
using new catalysts and the application technology for direct
methanol fuel cell) and New & Renewable Energy R&D Program
(2008NFC08P030000) funded by the MKE. This research was also
supported by the Human Resources Development of the Korea
Institute of Energy Technology Evaluation and Planning (KETEP)
grant funded by the Korea government MKE (2008-N-BL-HM-E-010000). The work at Kookmin University was supported by Priority
Research Centers Program through NRF funded by the MEST (20090093814).
References
Fig. 6. Pt4f XPS spectra of (a) Pt/C MEA before test, (b) Pt1 Ni1 /C MEA before test, (c)
Pt1 Ni1 /C MEA after test.
4. Conclusions
Pt1 Ni1 /C MEA exhibited greater fluctuations in potential than
the Pt/C MEA throughout the long term testing. Moreover, HR-TEM
images showed that the degree of sintering and detachment of
metal particles in the Pt1 Ni1 /C catalyst layer were less than those
in the Pt/C catalyst layer, and Ni dissolution in Pt1 Ni1 /C catalysts
was observed by XPS element mapping. These results indicate that
the Pt1 Ni1 catalyst is electrochemically and physically more unstable than the Pt/C under actual PEMFC operating conditions, even
though the Pt1 Ni1 /C catalyst is more resistant to the sintering and
[1] S. Mommura, K. Kawahara, T. Shimohira, Y. Teraoka, J. Electrochem. Soc. 155
(2008) A29.
[2] X.G. Yang, F.Y. Zhang, A.L. Lubawy, C.Y. Wang, Electrochem. Solid-State Lett. 7
(2004) A408.
[3] V.A. Sethuraman, J.W. Weinder, A.T. Haug, S. Motupally, L.V. Protsailo, J. Electrochem. Soc. 155 (2008) B50.
[4] E. Antolini, J.R.C. Salgado, E.R. Gonzalez, J. Power Sources 160 (2006) 957.
[5] M.S. Wilson, F.H. Garzon, K.E. Sickafus, S. Gottesfeld, J. Electrochem. Soc. 140
(1993) 2872.
[6] R.L. Borup, J.R. Davey, F.H. Garzon, D.L. Wood, M.A. Inbody, J. Power Sources
163 (2006) 76.
[7] J.P. Meyers, R.M. Darling, J. Electrochem. Soc. 153 (2006) A1432.
[8] D.A. Stevens, M.T. Hicks, G.M. Haugen, J.R. Dahn, J. Electrochem. Soc. 152 (2005)
A2309.
[9] H.A. Gasteiger, S.S. Kocha, B. Sompalli, F.T. Wagner, Appl. Catal. B 46 (2005) 9.
[10] G. Chen, S.R. Bare, T.E. Mallouk, J. Electrochem. Soc. 194 (2002) A1092.
[11] J. Luo, P.N. Njoki, Y. Lin, L. Wang, C.J. Zhong, Electrochem. Commun. 8 (2006)
581.
[12] A. Bonakdarpour, J. Wenzel, D.A. Stevens, S. Sheng, T.L. Monchesky, R. Lobel,
R.T. Atanasoski, A.K. Schmoeckel, G.D. Vernstorm, M.K. Debe, J.R. Dahn, J. Electrochem. Soc. 152 (2005) A61.
[13] H.R. Colon-Mercado, B.N. Popov, J. Power Sources 155 (2006) 253.
[14] Z. Wei, H. Guo, Z. Tang, J. Power Sources 62 (1996) 233.
[15] J.R.C. Salgado, E. Antolini, E.R. Gonzalez, J. Phys. Chem. B 108 (2004) 17767.
[16] T.-Y. Jeon, S.J. Yoo, Y.-H. Cho, S.H. Kang, Y.-E. Sung, J. Phys. Chem. C 113 (2009)
19732.
[17] Y.-H. Cho, Y.-H. Cho, H.-S. Park, D.-S. Jung, H.-Y. Park, Y.-E. Sung, J. Power Sources
172 (2007) 89.
[18] H. Yang, W. Vogel, C. Larmy, N. Alonso-Vante, J. Phys. Chem. B 108 (2004) 11024.
[19] L. Colmenarens, E. Guerrini, Z. Jusys, K.S. Nagabhushushana, E. Dinjus, S.
Behrens, W. Habicht, H. Bonnemann, R.J. Behm, J. Appl. Electrochem. 37 (2007)
1413.
[20] L.G.R.A. Santos, C.H.F. Oliveira, I.R. Moraes, E.A. Ticianelli, J. Electroanal. Chem.
596 (2006) 141.
[21] Y.S. Kim, M. Einsla, J.E. Mcgrath, B.S. Pivovar, J. Electrochem. Soc. 157 (2010)
B1602.
[22] H.R. Colon-Mercado, H. Kim, B.N. Popov, Electrochem. Commun. 6 (2004) 795.
[23] G. Liu, H. Zhang, Y. Zhai, Y. Zhang, D. Xu, Z. Shao, Electrochem. Commun. 9
(2007) 135.
[24] P. Yu, M. Pemberton, P. Plasse, J. Power Sources 144 (2005) 11.
[25] L. Kim, C.G. Chung, Y.W. Sung, J.S. Chung, J. Power Sources 183 (2008) 524.