Structural dependence of oxygen reduction reaction on palladium nanocrystals

Article abstract of DOI:10.1039/c1cc11004g

We have synthesized sub-10 nm Pd cubic and octahedral nanocrystals and then evaluated their activities towards oxygen reduction reaction (ORR). The ORR activity of Pd nanocubes was one order of magnitude higher than that of Pd octahedra, and comparable to

Full text of DOI:10.1039/c1cc11004g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 6566–6568
www.rsc.org/chemcomm
COMMUNICATION
Structural dependence of oxygen reduction reaction on palladium
nanocrystalsw
b
Minhua Shao,z*a Taekyung Yu,z Jonathan H. Odell,a Mingshang Jinb and Younan Xia*b
Received 21st February 2011, Accepted 22nd April 2011
DOI: 10.1039/c1cc11004g
We have synthesized sub-10 nm Pd cubic and octahedral nano-
crystals and then evaluated their activities towards oxygen
reduction reaction (ORR). The ORR activity of Pd nanocubes
was one order of magnitude higher than that of Pd octahedra,
and comparable to that of the state-of-the-art Pt catalysts.
{111} facets, and comparable to that of the state-of-the-art Pt
nanocatalysts.
The Pd nanocubes were synthesized in an aqueous solution
by reducing Na2PdCl4 with L-ascorbic acid in the presence of
Clꢀ and Brꢀ ions as the capping agents to promote the
formation of {100} facets, as reported previously.11 Fig. 1A
shows a transmission electron microscopy (TEM) image,
revealing that the sample contained 80% cubes with an edge
length of approximately 6 nm and 20% nanobars with an
aspect ratio of 1:2. The powder X-ray diffraction (XRD)
pattern (Fig. S1A in ESIw) of the Pd cubes confirms that they
had a face-centered cubic (fcc) structure (Fm3m, a = 3.958 A,
JCPDS Card No. 87-0641). The dimensions of the Pd cubes
calculated using the Scherrer formula was 6.4 nm, which was
in agreement with the TEM data. The high-resolution TEM
(HRTEM) image of a single Pd cube (Fig. 1B) clearly shows
continuous fringes with a period of 1.97 A, which is consistent
with the {200} lattice spacing of fcc Pd. The corresponding FT
pattern (Fig. 1B, inset) shows a square symmetry for the spots,
indicating that the Pd cube was a piece of single crystal bound
by {100} facets.
The slow kinetics of oxygen reduction reaction (ORR) is one
of the main obstacles hindering the commercialization of
low-temperature fuel cells.1,2 Platinum is by far the most
effective electrocatalyst for ORR, but it is an extremely rare
and expensive element. One way to address this problem is to
increase the ORR activity of a Pt catalyst by optimizing the
shape of constituent nanoparticles and thus maximizing the
expression of facets most reactive towards ORR reaction.3–6
In a non- or weakly-adsorbed electrolyte (such as HClO4)
solution, it has been established that the ORR activity on low-
index Pt surfaces increases in the order of Pt(100) o Pt(111) E
Pt(110).5 Another approach is to develop new catalysts based
on other noble metals. Palladium is a possible candidate but
the ORR activity of conventional Pd nanoparticles is at least
five times lower than that of their Pt counterparts, preventing
their direct use in fuel cells.9,10 It is not clear if facet engineer-
ing will be able to drastically enhance the activity of Pd
nanocrystals and thus lead to the development of Pd-based
catalysts for ORR. A recent study by Kondo et al. demon-
strated that the ORR activity on Pd single crystals followed a
trend opposite to that of Pt, i.e., Pd(110) o Pd(111) {
Pd(100), suggesting that the Pd(100) surface offers the most
active sites for ORR.7 However, Xiao et al. proposed that the
Pd(110) is more active than other sites based on their theoretical
calculations.8 Here we report a systematic study on the ORR
activities of shape-controlled Pd nanocrystals, including cubes
and octahedra. Our results show that the ORR activity of Pd
nanocubes enclosed by {100} facets were one order of
magnitude higher than that of Pd octahedra enriched with
a UTC Power, South Windsor, CT 06074, USA.
E-mail: minhua.shao@utcpower.com
b Department of Biomedical Engineering, Washington University,
Saint Louis, MO 63130, USA. E-mail: xia@biomed.wustl.edu
w Electronic supplementary information (ESI) available: Experimental
procedures, TEM images and XRD patterns of Pd cubes and octahedra.
TEM images of carbon-supported Pd nanocrystals, and additional
voltammetry curves. See DOI: 10.1039/c1cc11004g
Fig. 1 (A) TEM and (B) HRTEM images of Pd cubes synthesized by
heating an aqueous solution containing Na2PdCl4, L-ascorbic acid,
KCl, KBr, and PVP at 80 1C for 3 h. (C) TEM and (D) HRTEM
images of Pd octahedra synthesized by heating a water–ethanol
mixture containing Na2PdCl4, citric acid, and PVP at 80 1C for 3 h.
z Both authors contributed equally to this work.
c
6566 Chem. Commun., 2011, 47, 6566–6568
This journal is The Royal Society of Chemistry 2011

View Article Online
The Pd octahedra were prepared by heating a water–ethanol
mixture containing PVP, citric acid, and Na2PdCl4 at 80 1C for
3 h. As shown in Fig. 1C, the product mainly contained Pd
nanocrystals in an octahedral shape (490%) with an edge
length of 5–6 nm, which matched well with the size (5.8 nm)
obtained from the XRD data (Fig. S1B in ESIw). Both the
HRTEM image and corresponding FT pattern of a single Pd
octahedron recorded along the [011] zone axis reveal that the
Pd octahedron was also a single crystal enclosed by {111}
facets (Fig. 1D). The fringes with lattice spacing of 1.97 and
2.28 A can be indexed as {200} and {111} of Pd, respectively.
In our synthesis, the water–ethanol mixture was critical to the
formation of sub-10 nm Pd octahedra. When the synthesis was
conducted in an ethanol solution instead of a water–ethanol
mixture with other experimental parameters being kept the
same as in Fig. 1C, the product mainly contained polyhedral
nanocrystals with size of nearly 15 nm (Fig. S2A in ESIw). On
the other hand, small particles with sizes of less than 5 nm were
observed when the synthesis was conducted in an aqueous
solution without ethanol (Fig. S2B in ESIw). We believe that
ethanol could accelerate the reduction rate of Pd ions, thus
increase the sizes of Pd nanoparticles. However, a complete
understanding of the role of ethanol in this synthesis may
require further studies through not only experiments but also
simulations.
were also compared with those from the conventional
(cubo-octahedral) Pd nanoparticles (Pd/C, BASF). To minimize
the particle size effect, the as-received Pd/C powder was heat
treated at 500 1C in Ar for 2 h to achieve an average particle
size of 7.1 nm (denoted as Pd/C-HT). The Pd/C-HT sample
only showed a broad hydrogen adsorption peak at 0.22 V,
mainly from the contribution of (111) sites (Fig. S4 in ESIw).
The dominance of (111) signal can be explained by the fact
that B70% of the surface atoms on a 7.1 nm cubo-octahedral
nanoparticle are at (111) sites (Fig. S5 in ESIw).
The ORR activities of the Pd nanocrystals were measured
with a thin film-rotating disk electrode (TF-RDE). The results
are compared to Pd/C-HT in Fig. 3A. The oxygen polarization
curve of Pd/C cubes shifts to much more positive potentials
compared to those of octahedra and Pd/C-HT. This indicates
that the Pd/C cubes were much more active for oxygen
reduction. The intrinsic/specific ORR activities of Pd atoms
Fig. S3 shows typical TEM images of the carbon-supported
Pd cubes and octahedra, respectively, suggesting that all the
Pd nanocrystals were uniformly dispersed on the carbon
particles. Fig. 2 shows cyclic voltammetry curves for the Pd
nanocrystals in 0.1 M HClO4 solution. Since the surfaces of
the as-synthesized Pd nanocrystals were covered by capping
agents such as Brꢀ, Clꢀ, and PVP, no clear hydrogen
adsorption features were observed (dashed lines). The capping
agents could be completely removed by a non-destructive
proprietary method at room temperature. For Pd cubes, a
well-defined hydrogen adsorption peak was observed at 0.16 V
after the removal of capping agents. On the other hand, a
broader hydrogen adsorption peak appeared at 0.21 V for Pd
octahedra. The higher hydrogen adsorption potential implies
that the adsorption energy of hydrogen on Pd octahedra is
stronger than that on Pd cubes. The CVs of Pd nanocrystals
Fig. 3 (A) Anodic polarization curves for the ORR on Pd cubes and
octahedra supported on carbon black in 0.1 M HClO4. The data of
regular Pd/C (BASF) annealed at 500 1C for 2 h (Pd/C-HT) is also
included for comparison. Sweep rate = 10 mV sꢀ1; rotating speed =
1600 rpm; and room temperature. The currents were normalized to the
geometric area of the rotating disk electrode. The electrochemically
active areas of Pd/C cubes, Pd/C octahedra, and Pd/C-HT were 1.27,
1.4, and 0.98 cm2, respectively. Inset: Comparison of specific activities
for Pd/C cubes, Pd/C octahedra, Pd/C-HT, Pt/C, and Pt/C-HT at
0.9 V. (B) Anodic polarization curves in 0.1 M HClO4 and coverage of
OH of Pd/C cubes, Pd/C octahedra, and Pd/C-HT after subtracting
the double layer current density. The currents were normalized to the
electrochemical active area. The coverage of OH was calculated by
integrating the charge associate with OHad formation and normalized
to half of the charge of Cu UPD.
Fig. 2 Voltammetry curves of carbon-supported Pd nanocrystals
before and after removal of capping agents in a nitrogen-saturated
0.1 M HClO4 solution. Scanning rate = 50 mV sꢀ1. The currents were
normalized to the geometric area of the rotating disk electrode
(0.196 cm2).
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6566–6568 6567

View Article Online
in different samples were obtained by normalizing the kinetic
currents at 0.9 V to the electrochemically active areas (ECAs)
and compared in the inset of Fig. 3A. To avoid the interference
of hydrogen absorption in calculating the hydrogen adsorption
charge, the ECAs of Pd were obtained by integrating the
stripping charge of a Cu monolayer underpotentially deposited
on Pd, assuming 420, 490, and 460 mC cmꢀ2 for full Cu
monolayer coverage on cubes, octahedra, and cubo-octahedra,
respectively (Fig. S6 in ESIw).12,13 The specific activities of
Pd/C cubes, Pd/C octahedra, and Pd/C-HT were 0.31, 0.033
and 0.055 mA cmꢀ2, respectively. The activity enhancement of
Pd cubes was about 10 and 6 times over Pd/C octahedra and
Pd/C-HT, respectively. The Pd/C cubes were even more active
than the state-of-the-art Pt/C catalysts (TKK, 46.6 wt%) with
an average particle size of 2.8 nm, and comparable to Pt/C-HT
with a similar size range for the particles (TKK, 50 wt%,
7 nm). Our results demonstrate that Pd(100) sites are much
more active than Pd(111) at the nanoscale, consistent with the
extended surface study.7
{100} facets being much more active than the {111} facets. The
activity of Pd cubes was even comparable to that of the
state-of-the-art Pt electrocatalysts. With a much lower price
for Pd ($654.1/Oz.) as compared to Pt ($1796.9/Oz.), the cost
of fuel cells could be considerably lowered by switching from
Pt- to Pd-based catalysts.9 We believe that our results are also
important to guide the design of more active catalysts for fuel
cells and other applications. Further work on the catalytic
activity of Pd cubes will address the question of their fuel cell
performance and long-term stability.
This work was supported by UTC Power. As a visiting
student from Xiamen University, M.J. was also partially
supported by the China Scholarship Council. Part of the work
was performed at the Nano Research Facility (NRF), a
member of the National Nanotechnology Infrastructure
Network (NNIN), which is supported by the NSF under
award no. ECS-0335765.
Notes and references
The ORR kinetics is controlled by the amount of available
active sites on the catalyst’s surface and the interaction
between the surface and oxygen-containing species (e.g., O2,
O, OH, and OOH).14,15 The chemisorbed OH (OHad) acts as a
poison species in the potential range where oxygen reduction is
under combined kinetic-diffusion control, since it blocks the
surface sites for O2 adsorption.1,14 The anodic branches of the
voltammetry curves (0.65–0.9 V) for Pd/C cubes, Pd/C
octahedra and Pd/C-HT after subtracting the double layer
currents are compared in Fig. 3B. The onset potential of OHad
formation for both Pd octahedra and Pd/C-HT was B0.7 V,
which is more than 50 mV lower than that on Pd cubes
(40.75 V). The coverage of OHad (yOH) on the Pd surface
was calculated and also compared in Fig. 3B. It’s clear that the
OHad coverage on Pd cubes was much lower than other
surfaces in the potential range 4 0.7 V. Thus, the higher
ORR activity on Pd cubes can be attributed to its lower OHad
coverage and consequently more available reaction sites.
In summary, we thoroughly studied the structure dependence
of ORR activity for shape-controlled Pd nanocrystals with a
particle size B6 nm. The Pd cubes enclosed by {100} facets
were one order of magnitude more active than Pd octahedra
enclosed by {111} facets towards ORR. Our results demon-
strated that the ORR activity was strongly dependent on the
atomic structure on the surface of a Pd nanocatalyst with the
1 H. A. Gasteiger, S. S. Kocha, B. Sompalli and F. T. Wagner, Appl.
Catal., B, 2005, 56, 9.
2 W. Vielstich, A. Lamm and H. A. Gasteiger, Handbook of Fuel
Cells-Fundamentals, Technology and Applications, John Wiley &
Sons, Chichester, 2003, vol. 1–4.
3 R. R. Adzic, in Electrocatalysis, ed. J. Lipkowski and P. N. Ross,
Wiley, New York, 1998, pp. 197.
4 F. El Kadiri, R. Faure and R. Durand, J. Electroanal. Chem., 1991,
301, 177.
5 N. M. Markovic, R. R. Adzic, B. D. Cahan and E. B. Yeager,
J. Electroanal. Chem., 1994, 377, 249.
6 V. R. Stamenkovic, B. Fowler, B. S. Mun, G. Wang, P. N.
Ross, C. A. Lucas and N. M. Markovic, Science, 2007, 315,
493.
7 S. Kondo, M. Nakamura, N. Maki and N. Hoshi, J. Phys. Chem.
C, 2009, 113, 12625.
8 L. Xiao, L. Zhuang, Y. Liu, J. Lu and H. D. Abruna, J. Am. Chem.
Soc., 2008, 131, 602.
9 M. Shao, J. Power Sources, 2011, 196, 2433.
10 M. Shao, K. Sasaki and R. R. Adzic, J. Am. Chem. Soc., 2006, 128,
3526.
11 M. Jin, H. Liu, H. Zhang, Z. Xie, J. Liu and Y. Xia, Nano Res.,
2011, 4, 83.
12 T. Chierchie and C. Mayer, Electrochim. Acta, 1988, 33, 341.
13 J. Okada, J. Inukai and K. Itaya, Phys. Chem. Chem. Phys., 2001,
3, 3297.
14 N. M. Markovic, T. J. Schmidt, V. R. Stamenkovic and
P. N. Ross, Fuel Cells, 2001, 1, 105.
15 J. K. Norskov, J. Rossmeisl, A. Logadottir, L. Lindqvist,
J. R. Kitchin, T. Bligaard and H. Jonsson, J. Phys. Chem. B,
2004, 108, 17886.
c
6568 Chem. Commun., 2011, 47, 6566–6568
This journal is The Royal Society of Chemistry 2011