Newly developed stepwise electroless deposition enables a remarkably facile synthesis of highly active and stable amorphous Pd nanoparticle electrocatalysts for oxygen reduction reaction

Article abstract of DOI:10.1021/ja500275r

This paper reports on highly active and stable amorphous Pd nanoparticle electrocatalysts for the oxygen reduction reaction. The amorphous catalysts were synthesized by a remarkably facile and quick electroless deposition process newly developed in this study (process time <5 min). An electrode substrate (glassy carbon, carbon cloth) was sequentially dipped for 10 s into two separate solutions of a reducing agent (sodium hypophosphite (NaH₂PO ₂)) and Pd ions to deposit amorphous Pd nanoparticles containing phosphorus (Pd-P). By repeating the deposition cycles, the specific and mass activities of the Pd nanoparticles can be actively tuned. Owing to the nanoscale amorphous nature, the obtained Pd-P nanoparticle electrocatalysts exhibited superior specific and mass activities compared with crystalline Pd nanoparticles synthesized by another reducing agent (N₂H₄) and commercial Pt-loaded carbon (Pt/C) and Pd-loaded carbon (Pd/C). The specific and mass activities of the amorphous Pd-P nanoparticles were over 4.5 times and 2.6 times higher than previously reported values of Pd and Pt catalysts.

Full text of DOI:10.1021/ja500275r

Communication
pubs.acs.org/JACS
Newly Developed Stepwise Electroless Deposition Enables a
Remarkably Facile Synthesis of Highly Active and Stable Amorphous
Pd Nanoparticle Electrocatalysts for Oxygen Reduction Reaction
Kee Chun Poon,†,‡ Desmond C. L. Tan, ,†,‡Thang D.T. Vo, Bahareh Khezri, Haibin Su,
†,‡
†,‡
§
∥
§
Richard D. Webster, and Hirotaka Sato*
†
∥
School of Mechanical & Aerospace Engineering and School of Materials Science & Engineering, Nanyang Technological University,
5
0 Nanyang Avenue, Singapore 639798
§
Division of Chemistry & Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University, 21
Nanyang Link, Singapore 637371
*
S Supporting Information
although various other amorphous particles and films exhibit
electrocatalytic activity.
23−27
ABSTRACT: This paper reports on highly active and
stable amorphous Pd nanoparticle electrocatalysts for the
oxygen reduction reaction. The amorphous catalysts were
synthesized by a remarkably facile and quick electroless
deposition process newly developed in this study (process
time <5 min). An electrode substrate (glassy carbon,
carbon cloth) was sequentially dipped for 10 s into two
separate solutions of a reducing agent (sodium hypo-
In this study, stable amorphous Pd nanoparticles were
synthesized by stepwise electroless deposition directly onto an
electrode. Unlike other processes that require harsh conditions
such as high temperature, pressure and long synthesis
durations, this process only requires ambient conditions and
1
6,17,19−21
a remarkably short time for the synthesis process.
Such a direct deposition method also does not require additives
2
8−30
phosphite (NaH PO )) and Pd ions to deposit amorphous
2
2
affecting the catalytic activity,
thereby significantly
Pd nanoparticles containing phosphorus (Pd−P). By
repeating the deposition cycles, the specific and mass
activities of the Pd nanoparticles can be actively tuned.
Owing to the nanoscale amorphous nature, the obtained
Pd−P nanoparticle electrocatalysts exhibited superior
specific and mass activities compared with crystalline Pd
nanoparticles synthesized by another reducing agent
improving the ORR. Amorphous Pd nanoparticles synthesized
by this process have higher catalytic and mass activities than
crystalline Pd nanoparticles and commercial Pt/C and Pd/C for
the ORR in KOH (0.1 M) solution. This paper also suggests
that the catalytic activity improvement is due to the amorphous
phase of the Pd nanoparticles. Moreover, such a high catalytic
activity is maintained for long-term, confirmed by a durability
test compared with the crystalline Pd nanoparticles, Pt/C, and
Pd/C.
(
N H ) and commercial Pt-loaded carbon (Pt/C) and
2 4
Pd-loaded carbon (Pd/C). The specific and mass activities
of the amorphous Pd−P nanoparticles were over 4.5 times
and 2.6 times higher than previously reported values of Pd
and Pt catalysts.
The amorphous Pd nanoparticles were synthesized by a
stepwise dipping process. A glassy carbon rotating disc
electrode (GC-RDE) (3 mm diameter, Bioanalytical Systems,
Inc.) was first dipped into 0.2 M NaH PO (reducing agent) for
2
2
1
0 s. The electrode was then air-blow dried to remove excess
he oxygen reduction reaction (ORR) has attracted much
solution adhered to it. The electrode was then dipped into a 2
mM palladium chloride (PdCl ) solution for 10 s, resulting in
T
interest because of its importance in the cathodic reaction
2
1
in fuel cells. This interest, particularly with respect to alkaline
the reduction of the Pd ions to Pd metal nanoparticles by
utilizing only the reducing agent adsorbed onto the electrode
surface (i.e., Pd electroless deposition). These steps can be
repeated to tune and optimize the catalytic activity of Pd on the
electrode surface as in Figure 2. For comparison, the sequence
was replicated by varying the reducing agents used (e.g.,
fuel cells, has resurfaced because of recent advances in alkaline
2
anion-exchange membranes. Currently, platinum (Pt) electro-
catalysts are the best and most widely used catalyst for the
ORR. However, due to the high cost of Pt, cheaper alternatives
such as Pt-based alloys and non-Pt metals have been
3
−15
researched.
Pd has emerged as a strong candidate for
hydrazine (N
2 4 2 2
H ) instead of NaH PO ).
The Pd nanoparticles synthesized with NaH
replacing Pt owing to its significantly lower cost and high
2
PO (denoted
2
1
6−19
by Pd−P) were small and sparsely dispersed (Figure 1A).
Moreover, the halo electron diffraction pattern indicates that
the Pd nanoparticles were amorphous. The EDS spectrum of
these Pd nanoparticles showed that they contain 11 at% of
phosphorus. The X-ray diffraction profile (Figure S1) of the Pd
catalytic activity.
Recent studies on Pd as an electrocatalyst have shown many
interesting advancements such as the ability to manipulate the
1
6
20
size and shape of Pd nanoparticles, the combination of Pd
2
1,22
with other metals to give bimetallic catalysts,
and the use of
various support materials to improve the catalytic activity of
3
,5,6,9,10,16−19
Pd.
However, very few have studied Pd nano-
Received: January 10, 2014
Published: March 24, 2014
particle morphology tuning to improve the catalytic activity,
©
2014 American Chemical Society
5217
dx.doi.org/10.1021/ja500275r | J. Am. Chem. Soc. 2014, 136, 5217−5220

Journal of the American Chemical Society
Communication
Figure 1. Transmission electron microscopy images, electron
diffraction patterns, and energy-dispersive X-ray spectroscopy results
from Pd NPs by the stepwise electroless deposition process using
reducing agents (A) NaH PO and (B) N H . From the EDS spectra,
2
2
2
4
Pd:P ratio for (A) is 89:11, whereas that for (B) indicates that the
particles are pure Pd.
Figure 2. ORR polarization curves for Pd nanoparticles synthesized
with (A) NaH PO and (B) N H . The numbers correspond to the
2
2
2
4
cycle numbers of the stepwise electroless deposition process. The
measurements were performed in 0.1 M O -saturated KOH solution.
Scan rate: 10 mV s . Rotational speed: 1600 rpm.
2
nanoparticles further supports that the Pd nanoparticles were
amorphous. On the other hand, Pd nanoparticles synthesized
with N H (Figure 1B) were also small, but the electron
−1
2
4
Table 1. Half-Wave Potential and Pd (and Pt) Specific and
diffraction pattern shows clear Debye−Scherrer rings, proving
a
Mass Activity Values
that the Pd nanoparticles synthesized with N H were
2
4
crystalline.
half-wave
specific activity
mass activity
−1
catalyst
Pt/C
potential (V)
(mA cm−
2
)
(mA μg )
Linear sweep voltammetry using rotating disk electrodes
RDE) were performed with varying number of deposition
(
0.74
0.83
0.82
0.78
0.85
0.83
0.88
0.82
0.47
1.58
2.31
2.16
5.16
2.87
6.85
3.53
0.068
0.17
1.79
1.11
2.21
1.33
1.34
1.44
cycles to analyze the effect of amorphous Pd−P nanoparticles
on the ORR in O -saturated KOH (0.1 M) solution compared
with that of crystalline Pd nanoparticles, Pt/C, and Pd/C.
Figure 2 shows that the Pd nanoparticles prepared with just
three deposition cycles (total synthesis time = 60 s, including
Pd/C
Pd−P (3)
Pd−N H (3)
2
2
4
Pd−P (9)
Pd−N H (9)
2
4
1
0 s each in the reducing agent and Pd ion solution per cycle)
Pd−P (18)
exhibit superior catalytic activity than Pt/C, as indicated by the
positive shift in the ORR polarization curve. This trend was
Pd−N H (18)
2
4
a
Mass activity and specific activity calculations were according to the
1
9
observed for Pd nanoparticles synthesized with NaH PO2
method described elsewhere. The values for Pt/C are consistent with
2
3
1−35
(Figure 2A) and N H (Figure 2B). The positive shift in the
that in other papers.
The values are the average of several
2
4
curves increased with the number of deposition cycles.
Notably, a comparison of Pd nanoparticles synthesized with
NaH PO and N H (denoted by Pd−N H ) at the same
numbers of repeated experiments. The current was measured at 0.85 V
vs RHE at 1600 rpm. The numbers in the bracket correspond to the
cycle numbers of the stepwise electroless deposition process.
2
2
2
4
2
4
number of cycles reveals that Pd−P shows superior catalytic
activity than Pd−N H . As summarized in Table 1 from Figure
2
4
2
, after three cycles, Pd−P exhibited a more positive half-wave
From Table 1, Pd−P nanoparticles have the highest active
surface among Pt/C, Pd/C, and Pd−N H as seen from it
potential (0.82 V) than Pd−N H (0.78 V) and also an almost
2
4
2
4
similar half-wave potential to that of Pd/C (0.83 V). This
increase in catalytic activity of Pd−P nanoparticles was more
pronounced at higher number of cycles. In fact, Pd−P at 9
cycles exhibited a more positive half-wave potential (0.85 V)
than Pd−N H (0.83 V) at 9 cycles or even 18 cycles (0.82 V).
having the highest specific activity, which is kinetic current
normalized by electrochemically active surface area (ECSA).
−2
The maximum specific activity (6.85 mA cm ) obtained at 18
cycles of Pd−P is over 4.5 times higher than previously
19
reported values of Pd and Pt catalysts. The specific activity is
higher at greater number of cycles (9, 18 cycles) than at lower
(3 cycles), which suggests initially synthesized Pd−P particles
that deposited directly on the glassy carbon surface at lower
number of cycles is less active than subsequently synthesized
2
4
The most positive half-wave potential was 0.88 V obtained at
8 cycles of Pd−P. Overall, the results prove that Pd−P
nanoparticles exhibit superior catalytic activity than Pd−N H ,
1
2
4
Pd/C, and Pt/C.
5
218
dx.doi.org/10.1021/ja500275r | J. Am. Chem. Soc. 2014, 136, 5217−5220

Journal of the American Chemical Society
Communication
3
7−39
Pd−P particles that deposited on previously deposited Pd−P
particles at greater number of cycles.
structures.
The presence of low-coordination sites has
38−43
been shown to improve catalytic activity in the ORR.
The stepwise electroless deposition reasonably saved the Pd
loading, resulting in very high mass activity, the kinetic current
normalized by mass. Table 1 also shows the mass activity for
Pd−P and Pd−N H at different numbers of cycles. The
As in Figure 3, the Pd nanoparticles were deposited directly
onto carbon cloth. This experiment demonstrates the
developed stepwise electroless deposition is suitable for
practical preparation of the catalyst sheet for fuel cell.
Compared to traditional casting of Pt loaded carbon paste
onto carbon cloth, our stepwise electroless deposition is more
facile, simpler, and less costly because it deposits catalyst
nanoparticles directly onto carbon cloth under ambient
condition, and we can actively tune the catalyst load by simply
repeating the dipping cycle.
2
4
amount of Pd loading was measured by extracting the Pd
nanoparticles from the electrode surfaces using concentrated
HNO3 solution and analyzing the samples by inductively
coupled plasma-mass spectrometry (ICP-MS). Pt/C followed
by Pd/C had the lowest mass activity compared with both Pd−
P and Pd−N2H4. Comparing with Pd−N H , Pd−P is superior
30
2
4
at almost all deposition cycle numbers. For 3 cycles, the mass
Figure 4, showing the durability test for the ORR of Pd
nanoparticles synthesized with NaH PO or N H and the
−1
activity of Pd−P (1.79 mA μg ) was ∼1.6 times that of Pd−
2
2
2
4
−1
N H (1.11 mA μg ). This trend was further reflected in
2
4
samples prepared with 9 cycles (Pd−P 1.6 times greater). For
1
8 cycles, the mass activity for Pd−P was almost similar to that
of Pd−N2H4. To our best knowledge, the mass activity (2.21
−1
mA μg ) given by 9 cycles Pd−P is the highest value among
ever reported electrocatalysts. For instance, it is 2.6 times
1
9
higher than the one reported in a known paper, while the
current density of the Pd−P is higher to that in the paper; thus,
the Pd−P synthesized by the stepwise electroless deposition is
much more cost-effective as an ORR electrocatalyst.
Figure 3 suggests that the superior catalytic activity is due to
the amorphous nature of the Pd−P nanoparticles, as indicated
Figure 4. Chronoamperometry response of Pd nanoparticles
synthesized with NaH PO2 or N H . The measurements were
2
2
4
performed in 0.1 M O -saturated KOH solution, as per the procedure
2
reported elsewhere (potential: 0.666 V vs RHE. Rotational speed: 200
1
rpm).
commercial Pt/C and Pd/C, demonstrates that the durability of
the amorphous Pd−P nanoparticles is enough to keep its
catalytic activity being higher than that of Pt/C, Pd/C, and Pd−
N H nanoparticles prepared with the same number of
2
4
deposition cycles as well as to that of Pd−N H nanoparticles
2
4
prepared with a greater number of cycles.
In summary, a facile and quick method for synthesizing Pd
nanoparticles using NaH PO as the preferred reducing agent
Figure 3. ORR polarization curves of Pd nanoparticles synthesized on
carbon cloth. Annealing was performed at 700 °C for 1 h. The
2
2
measurements were performed in 0.1 M O -saturated KOH solution.
by stepwise electroless deposition was developed. The specific
and mass activities of the obtained Pd nanoparticles containing
phosphorus (Pd−P) were higher than any previously reported
values. The Pd nanoparticles obtained were amorphous, which
may account for their superior specific and mass activities
compared with those of crystalline Pd nanoparticles, Pt/C, and
Pd/C. Furthermore, the mass activities of these Pd nano-
particles are tunable by varying the number of deposition
cycles. The stepwise electroless deposition method was also
proven to be effective for practical use with carbon cloth,
indicating its suitability for fuel cell industrial applications. The
amorphous Pd nanoparticles were durable enough to keep the
high catalytic activity for long-term. Overall, these amorphous
Pd nanoparticles are a promising alternative to commercial Pd/
C and Pt/C for the ORR in alkaline solutions. The stepwise
electroless deposition process has the following benefits: (1)
Handling is remarkably facile (sequential dipping into two
solutions of simple chemical compositions), and reaction time
is short (20 s per deposition cycle); (2) catalytic activity is
tunable by simply repeating the deposition cycle; (3) no
2
−1
Scan rate: 10 mV s . The inserted figure is the EDS spectra of Pd−P
after the annealing, indicating that P remains even after the annealing.
by the polarization curve of the carbon cloth sample annealed
at 700 °C for 1 h. Annealing at this temperature and duration
36
changes the amorphous Pd to crystalline phase. After both
samples were annealed, the catalytic activity of the Pd−P
sample became similar to that of the Pd−N H sample
2
4
(
crystalline Pd). Also, phosphorus remained in the Pd
nanoparticles even after the annealing, which is confirmed by
EDS (Figure 3). Hence, the amorphous nature of Pd−P is the
key factor in its superior catalytic activity. The surfaces of Pd−P
are more structurally disordered and have more accessible
reactive sites with low coordination number in the amorphous
nanostructure which may account for the catalytic activity of
Pd−P being superior to that of the crystalline Pd−N H . Such
2
4
low coordination sites, represented by defects, kinks and edges,
are abundant in amorphous structures but not so in crystalline
5
219
dx.doi.org/10.1021/ja500275r | J. Am. Chem. Soc. 2014, 136, 5217−5220

Journal of the American Chemical Society
Communication
vacuum, high temperature, or an external power source is
needed; and (4) less chemical waste is produced.
(18) Papandrew, A. B.; Chisholm, C. R. I.; Zecevic, S. K.; Veith, G.
M.; Zawodzinski, T. A. J. Electrochem. Soc. 2013, 160, F175.
(
19) Seo, M. H.; Choi, S. M.; Kim, H. J.; Kim, W. B. Electrochem.
Commun. 2011, 13, 182.
20) Erikson, H.; Sarapuu, A.; Alexeyeva, N.; Tammeveski, K.; Solla-
Gullon, J.; Feliu, J. M. Electrochim. Acta 2012, 59, 329.
21) Maheswari, S.; Sridhar, P.; Pitchumani, S. Electrochem. Commun.
013, 26, 97.
22) Neergat, M.; Gunasekar, V.; Rahul, R. J. Electroanal. Chem. 2011,
In addition, by introducing other metal ion solutions and
their mixtures into the process, other pure metals, alloys, or
composite nanoparticles can be synthesized. Owing to these
benefits, nanoparticles can be widely designed and tailored
from theoretical models, not only for ORR electrocatalysis but
also for various purposes, and the synthetic process can be
automated toward combinatorial chemistry and high-through-
put screening.
(
́
(
2
(
658, 25.
(23) Budniok, A.; Kupka, J. Electrochim. Acta 1989, 34, 871.
(
24) Kreysa, G.; Hak
Electrochem. 1986, 201, 61.
25) Podesta, J. J.; Piatti, R. C. V. Int. J. Hydrogen Energy 1997, 22,
53.
26) Xiao, L.; Zhuang, L.; Liu, Y.; Lu, J.; Abrun
Chem. Soc. 2008, 131, 602.
(27) Antolini, E. Energy Environ. Sci. 2009, 2, 915.
28) Yin, H.; Tang, H.; Wang, D.; Gao, Y.; Tang, Z. ACS Nano 2012,
, 8288.
̊
ansson, B. J. Electroanal. Chem. Interfacial
ASSOCIATED CONTENT
Supporting Information
■
(
7
(
́
*
S
Experimental procedures and XRD patterns of Pd deposits
obtained with NaH PO /N H . This material is available free of
̃
a, H. c. D. J. Am.
2
2
2
4
charge via the Internet at http://pubs.acs.org.
(
6
AUTHOR INFORMATION
Corresponding Author
hirosato@ntu.edu.sg
■
(29) Lee, Y.; Loew, A.; Sun, S. Chem. Mater. 2009, 22, 755.
(30) Li, D.; Wang, C.; Tripkovic, D.; Sun, S.; Markovic, N. M.;
Stamenkovic, V. R. ACS Catalysis 2012, 2, 1358.
(31) Niu, K.; Yang, B.; Cui, J.; Jin, J.; Fu, X.; Zhao, Q.; Zhang, J. J.
Power Sources 2013, 243, 65.
Author Contributions
‡
These authors contributed equally.
(32) Yang, W.; Fellinger, T.-P.; Antonietti, M. J. Am. Chem. Soc.
Notes
2
010, 133, 206.
The authors declare no competing financial interest.
(33) Cui, Z.; Burns, R. G.; DiSalvo, F. J. Chem. Mater. 2013, 25, 3782.
(34) Dong, H.-Q.; Chen, Y.-Y.; Han, M.; Li, S.-L.; Zhang, J.; Li, J.-S.;
ACKNOWLEDGMENTS
Lan, Y.-Q.; Dai, Z.-H.; Bao, J.-C. J. Mater. Chem. A 2014, 2, 1272.
35) Han, C.; Wang, J.; Gong, Y.; Xu, X.; Li, H.; Wang, Y. J. Mater.
Chem. A 2014, 2, 605.
■
(
This study was financially supported by the Nanyang Assistant
Professorship (NAP, M4080740). The authors appreciate Ms.
Koh Joo Luang, Ms. Heng Chee Hoon, Ms. Yong Mei Yoke
and Mr. Leong Kwok Phui at Material/Biological & Chemical
Laboratories at MAE, NTU, for their continuous support and
effort to set up and maintain an excellent experimental
environment.
(36) Gullman, L. O. J. Less-Common Met. 1966, 11, 157.
(37) Lee, J.-S.; Park, G. S.; Lee, H. I.; Kim, S. T.; Cao, R.; Liu, M.;
Cho, J. Nano Lett. 2011, 11, 5362.
38) Yang, J.; Xu, J. J. Electrochem. Commun. 2003, 5, 306.
(
(39) Lima, F. H. B.; Calegaro, M. L.; Ticianelli, E. A. J. Electroanal.
Chem. 2006, 590, 152.
(
40) Cheng, F.; Su, Y.; Liang, J.; Tao, Z.; Chen, J. Chem. Mater. 2009,
2, 898.
41) Macia,
Electroanal. Chem. 2004, 564, 141.
42) Kuzume, A.; Herrero, E.; Feliu, J. M. J. Electroanal. Chem. 2007,
99, 333.
43) Cheng, F.; Shen, J.; Peng, B.; Pan, Y.; Tao, Z.; Chen, J. Nat
Chem 2011, 3, 79.
2
REFERENCES
■
(
́
̃
M. D.; Campina, J. M.; Herrero, E.; Feliu, J. M. J.
(
1) Guo, S.; Zhang, S.; Wu, L.; Sun, S. Angew. Chem., Int. Ed. 2012,
1, 11770.
5
(
5
(
(
2) Park, J.-S.; Park, S.-H.; Yim, S.-D.; Yoon, Y.-G.; Lee, W.-Y.; Kim,
C.-S. J. Power Sources 2008, 178, 620.
3) Chen, Z.; Waje, M.; Li, W.; Yan, Y. Angew. Chem., Int. Ed. 2007,
6, 4060.
4) El- Deab, M. S.; Ohsaka, T. Electrochem. Commun. 2002, 4, 288.
5) Fernandez, J. L.; Walsh, D. A.; Bard, A. J. J. Am. Chem. Soc. 2004,
27, 357.
6) Kim, J.; Park, J.-E.; Momma, T.; Osaka, T. Electrochim. Acta 2009,
4, 3412.
(
4
(
(
1
(
5
(44) Qu, L.; Liu, Y.; Baek, J.-B.; Dai, L. ACS Nano 2010, 4, 1321.
́
(
(
(
7) Lin, Y.; Cui, X.; Ye, X. Electrochem. Commun. 2005, 7, 267.
8) Matter, P. H.; Zhang, L.; Ozkan, U. S. J. Catal. 2006, 239, 83.
9) Shao, M.; Yu, T.; Odell, J. H.; Jin, M.; Xia, Y. Chem. Commun.
2
011, 47, 6566.
(
10) Suo, Y.; Zhuang, L.; Lu, J. Angew. Chem., Int. Ed. 2007, 46, 2862.
(
(
11) Wang, B. J. Power Sources 2005, 152, 1.
12) Yan, X.-Y.; Tong, X.-L.; Zhang, Y.-F.; Han, X.-D.; Wang, Y.-Y.;
Jin, G.-Q.; Qin, Y.; Guo, X.-Y. Chem. Commun. 2012, 48, 1892.
13) Liang, Y.; Li, Y.; Wang, H.; Zhou, J.; Wang, J.; Regier, T.; Dai,
H. Nat. Mater. 2011, 10, 780.
14) Oezaslan, M.; Hasche, F.; Strasser, P. J. Electrochem. Soc. 2012,
59, B444.
15) Tang, Q.; Jiang, L.; Qi, J.; Jiang, Q.; Wang, S.; Sun, G. Appl.
Catal., B 2011, 104, 337.
16) Jiang, L.; Hsu, A.; Chu, D.; Chen, R. J. Electrochem. Soc. 2009,
56, B643.
17) Mentus, S.; Abu Rabi, A.; Jasi
74.
(
(
́
1
(
(
1
(
̌
n, D. Electrochim. Acta 2012, 69,
1
5
220
dx.doi.org/10.1021/ja500275r | J. Am. Chem. Soc. 2014, 136, 5217−5220