Highly active graphene-supported NixPd100-x binary alloyed catalysts for electro-oxidation of ethanol in an alkaline media

Article abstract of DOI:10.1021/cs500103a

A series of graphene (G)-supported Ni_xPd_100-x binary alloyed catalysts (BACs) were prepared with variation of Ni and Pd metal loading through a facile chemical reduction method and were used as an anode catalyst for ethanol oxidation reaction (EOR), which is superior fuels for direct ethanol fuel cells (DEFCs). The X-ray diffraction (XRD) data reveals that the Ni _xPd_100-x/G catalysts were homogeneously alloyed, and Ni was present with the oxidized form. The transmission electron microscopy (TEM) images also suggest the alloyed formation with different shapes of metal nanoparticles (NPs). The electrochemical properties of the catalysts were evaluated using cyclic voltammetry (CV) and chronoamperometry (CA) in 1 M KOH electrolyte. The higher catalytic activity for EOR was increased in the order Ni₇₅Pd₂₅/G > Ni₀Pd₁₀₀/G > Ni₂₅Pd₇₅/G > Ni₅₀Pd₅₀/G. Among Ni_xPd_100-x/G BACs, the Ni₅₀Pd₅₀/G catalyst showed the highest onset potential (-0.8 V) on CV and long-term stability on amperometric measurements. The overall parameters of EOR study were determined that the Ni₅₀Pd₅₀/G was more favorable in various ethanol (EtOH) concentrations and scan rate.

Full text of DOI:10.1021/cs500103a

Research Article
pubs.acs.org/acscatalysis
Highly Active Graphene-Supported Ni Pd100−x Binary Alloyed
x
Catalysts for Electro-Oxidation of Ethanol in an Alkaline Media
Mohammad Shamsuddin Ahmed and Seungwon Jeon*
Department of Chemistry and Institute of Basic Science, Chonnam National University, 300 Yongbong-dong, Buk-gu, Gwangju
500-757, Republic of Korea
*
S Supporting Information
ABSTRACT: A series of graphene (G)-supported Ni Pd
binary alloyed catalysts
100−x
x
(
BACs) were prepared with variation of Ni and Pd metal loading through a facile
chemical reduction method and were used as an anode catalyst for ethanol oxidation
reaction (EOR), which is superior fuels for direct ethanol fuel cells (DEFCs). The X-ray
diffraction (XRD) data reveals that the Ni Pd100−x/G catalysts were homogeneously
x
alloyed, and Ni was present with the oxidized form. The transmission electron
microscopy (TEM) images also suggest the alloyed formation with different shapes of
metal nanoparticles (NPs). The electrochemical properties of the catalysts were evaluated
using cyclic voltammetry (CV) and chronoamperometry (CA) in 1 M KOH electrolyte.
The higher catalytic activity for EOR was increased in the order Ni Pd /G > Ni Pd100/
75
25
0
G > Ni Pd /G > Ni Pd /G. Among Ni Pd
/G BACs, the Ni Pd /G catalyst
25
75
50
50
x
100−x
50 50
showed the highest onset potential (−0.8 V) on CV and long-term stability on
amperometric measurements. The overall parameters of EOR study were determined that
the Ni Pd /G was more favorable in various ethanol (EtOH) concentrations and scan rate.
50
50
KEYWORDS: Ni−Pd, binary alloyed catalysts, ethanol oxidation reaction, electrocatalyst, direct ethanol fuel cell
INTRODUCTION
attracted widespread attention due to their low cost, stability,
and excellent catalytic activities.
Various carbon materials, which are used as supports for Pd
■
8
,15,19,20,22,23
Platinum-containing catalysts have been used as the best
electroactive catalysts for both anode and cathode in fuel cells
1
metal-based catalysts, such as carbon blacks, carbon nanotubes,
(
FCs). However, they still suffer from multiple problems such
2
,22−26
carbon nitride, and carbon nanofibers
have been studied
as slow reaction kinetics, CO poisoning, high cost, limited
reserve in nature, and poor durability.
2
−7
intensively as well in order to improve the catalytic perform-
ance. Also, two-dimensional thin material, graphene (G)-
For reducing the
usage of the precious Pt metal and hereafter the cost of FCs,
the progress of non-Pt electrocatalysts has thus generated a
1
0,27
11,28
supported Pd catalysts,
including alloy
and core−
29
8
,9
shell have been investigated. It is well-known that G has
attracted much attention in the last several years owing to its
huge interest.
Alternatively, palladium is an auspicious
catalyst because it is cheaper than Pt and has higher catalytic
activity for small alcoholic molecules (i.e., ethanol, EtOH)
oxidation for DEFCs in basic media,
considered as potential future power sources.
known that adsorbed hydroxide species could promote
dehydrogenation of adsorbed alcohol and desorption of
poisoning residues and thus enhance the oxidation of alcohol
on Pd catalysts in alkaline medium. Binary alloyed catalysts
BACs), which have been proved to be an effective way to
further improve the electrocatalytic activity and tolerance to
poisoning intermediates of Pd-based electrocatalysts. How-
ever, the addition of a transition metal element, such as nickel
to Pd can significantly improve the overall catalytic activities of
Pd due to the bimetallic promotional effect. In addition, the
surface morphologies and sizes of electrocatalysts are playing a
3
0,31
unique properties,
including thermal and electrical
8
,10−14
conductivity, incredible strength, higher explicit surface area,
which are
It is well-
32−35
15,16
and cost effectiveness.
It is expected that the addition of G
can not only maximize the specific surface area of nanosized
electrocatalyst for electron transfer but also provide enhanced
mass transport of reactants to the electrocatalyst. However, very
1
0,27
11,28
1
7
few G-supported Pd
and Pd-based alloys
have been
reported for the ethanol oxidation reaction (EOR), including
(
29
the core−shell from Zhang et al. So far, for the first time we
18
are applying on G-supported Ni Pd
(hereafter, denoted as
100−x
x
Ni Pd100−x/G) BACs for EOR.
x
On the basis of the above considerations, we synthesized
19
Ni Pd100−x/G BACs via simple chemical process, which has the
x
following merits: First, the Ni is easily alloyed with Pd to form
BAC in low temperature, because it is hard to alloy Ni and Pd
2
0
vital role in their catalytic activities. Moreover, when two
metals are alloyed, the d-band shift and the changes of
segregation energy of added metals should also be responsible
for bimetallic high activity of Pd catalysts, based on density
Received: January 23, 2014
Revised: March 31, 2014
Published: April 30, 2014
2
1,22
functional theory.
Recently, the synthesis of Pd with Ni has
©
2014 American Chemical Society
1830
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
Figure 1. TEM images of Ni Pd100/G (a), Ni Pd /G (b), Ni Pd /G (c), and Ni Pd /G (d) BACs. Insets: close-up view of selective particles
0
25
75
50
50
75
25
and the HRTEM of respective BACs (a′−d′).
2
2
37
using low reduction temperature. Second, the various
ascorbic acid. Figure1b shows that the cubic Pd nanocrystals
in Ni Pd /G exist with spherical-like structures upon addition
Ni Pd100−x/G BACs are easily synthesized, and Ni can be
x
25
75
found as an amorphous oxide form. Third, the as-prepared
of Ni , and the cubic crystals have relatively blunt angles.
Ni Pd100−x/G BACs exhibited higher electrocatalysis toward
Figure1c shows a different morphology in Ni Pd /G. Almost
x
50
50
EOR than that of Pd/G and Pt/C. Particularly, the results
all NPs are in spherical structures, although some cubic Pd
nanocrystals are still visible. Finally, in Figure1d, an aggregated
spherical structure can be observed due to excess Ni loading in
Ni Pd /G. The magnified TEM images in all insets are
prove that the Ni Pd /G exhibit outstanding electrocatalysis
5
0
50
for EOR. This study will be of great interest, and we expect our
work to pioneer the development of cost-effective G-supported
Pd-based alloy catalysts.
75
25
subjected to the close observation of size and shape. The
average size of the NPs were 35, 39, and 15 nm for Ni Pd100/G,
0
RESULTS AND DISCUSSION
Ni Pd /G, and Ni Pd /G, respectively. High-resolution
■
25
75
50
50
TEM (HRTEM) images are indicating that the prepared
Ni Pd
The lattice line spacing of NPs are 0.222, 0.234, and 0.229 nm,
which represented Pd(111), (101), and (111) planes of NiO/
Characterization of Various Ni Pd100−x/G. The trans-
mission electron microscopy (TEM) images of as-prepared
x
/G BACs are entirely crystalline (Figure 1a′−d′).
100−x
x
Ni Pd100−x/G BACs are shown in Figure 1, and the Pd
x
nanocrystals were obtained with varying Ni concentrations. As
shown in Figure1a, the Ni Pd100/G consists of a clear crystal
with variable size and shape on G, whereas the cubic shapes are
leading with a fairly sharp angle. The cubic Pd crystal formation
can be observed previous probably due to the addition of
Ni(OH)
. The similar shape variations of Pd nanocrystals were
2
0
observed previously upon various amount of KI addition as
38
additive. The probable shape formation mechanism of
Ni Pd100−x/G BACs was dependent on the ratio of AA and
x
1
831
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
Pd precursor upon Ni precursor addition. It is, however, more
or less similar to the reference. The Ni−Pd binary alloyed
formation was observed by the elemental mapping analysis
Ni Pd100/G (Figure 2b). This shift indicates that Ni atoms
0
38
around Pd sites had entered into the Pd lattice, forming a
2
2,39
bimetallic transition zone.
The numeric values are listed in
(
Figure S1). The energy-dispersive X-ray spectroscopy (EDS)
Table S1. No diffraction peak of Ni is observed for all BACs,
indicating the Ni in BACs was amorphous in nature, although
Ni can be easily oxidized in air.
pattern can be observed for the numerical analysis of elements
in all Ni Pd100−x/G BACs (Figure S2).
x
The X-ray diffraction (XRD) patterns of all Ni Pd100−x/G
BACs are shown in Figure 2, which characterizes their crystal
The average crystallite size of the catalysts was calculated
from Pd (220) peak using Scherrer eq 1
x
23
kλ
d(Å) =
βcos θ
(1)
where k is a coefficient (0.9), β the full-width at half-maximum
fwhm), λ the wavelength of X-ray used (1.5406 Å), and θ is
(
the angle of peak maximum. The calculated crystallite sizes
were 33, 35, 17, and 18 nm for Ni Pd /G, Ni Pd /G,
0
100
25
75
Ni Pd /G, and Ni Pd /G BACs, respectively, which are
50
50
75
25
similar to the HRTEM analysis. Also, on the basis of the same
peak value, the alloying percentages for all Ni Pd BACs
x
100−x
8
were calculated from Vegard’s law and are presented in Table
1
.
The X-ray photoelectron spectroscopic (XPS) spectra of GO
and all Ni Pd100−x/G BACs are shown in Figure 3. The XPS
x
recorded signals from C, O, Pd, and N for all Ni Pd100−x/G
x
BACs, with that attributable to Pd being the most significant.
However, tiny Ni 2p peaks were observed upon Ni addition, as
can be seen in Figure 3a. The atomic ratio of C/O significantly
increased (from 1.6 to 4.1 for GO and Ni Pd100/G,
0
respectively) upon the addition of reduction (Figure S3). The
core level of C 1s XPS spectra of GO and all Ni Pd100−x/G
x
BACs are shown in Figure 3b. The peak at ∼284.9 eV is
2
attributable to the graphitic sp bonds. The peaks located at
2
87.6 and 289.2 eV are due to the oxygenated functional groups
3,10,40
such as CO and O−CO, respectively.
The intensity
of each class of carbon can be seen in Figure S4. The spectra of
all Ni Pd100−x/G BACs show a significant reduction of
x
oxygenated functional groups compared to the GO, which is
also confirmed by FTIR (Figure S5). For instance, in FTIR, the
Figure 2. (a) XRD patterns of graphene supported Ni Pd100−x BACs.
Inset: XRD pattern of GO. (b) Enlarged pattern of Pd(111), Pd(200),
and Pd(220) peaks.
−1
x
CO (1730 cm ) peak of GO was slightly shifted to ∼1747
−1
cm for Ni Pd
all BACs, with a small appearance upon
100−x
x
reduction. Indicating O species were reduced, and residual O
atoms of G interacted with Ni Pd particles by chem-
x
100−x
4
structure. In each pattern of Figure 2a, there are three
characteristic typical peaks. The peaks at 2θ = 40.05°, 46.58°,
and 68.06° are allocated to the Pd (111), (200), and (220),
respectively. The peak at 10.8° is raised from the graphite (002)
plane of GO (inset) that shifted to the higher angle at around
isorption on the surfaces of the G (C−O/Ni Pd
). The Pd
100−x
/G BACs (Figure 3c) show two
x
3d spectra of all Ni Pd
x
100−x
peaks at 355.6 and 340.8 eV for the 3d5/2 and 3d3/2
,
respectively, whereas those peaks are located at 355.3 and
340.5 eV for Ni Pd100/G (Figure3c). Such positively shifted
0
2
4.7° upon reduction. It is, however, those Pd peaks mentioned
values indicates that the Pd in Ni Pd100−x/G BAC was present
x
above that slightly shift to higher angles in the patterns of
Ni Pd100−x/G BACs compared with those in the pattern of
in a different electronic environment due to the alloy
2
4,41
formation.
If we consider the more magnified spectra of
x
Table 1. Physical Characterization and Electrochemical Performance Comparison between of All Ni Pd100−x/G BACs
x
Ni Pd100/G
0
Ni Pd /G
Ni Pd /G
Ni Pd /G
75 25
25
75
50
50
Pd wt % from XPS analysis
Pd loading (μg cm−2)
alloy formation (%)
50.64
35.82
46.51
32.89
48.6
61
39.34
27.82
73.4
77
24.85
17.58
36.3
46
ECSA (m2g −P d1
)
68
−0.74
0.57
113.7
9.4
EOR onset (V vs Ag/AgCl)
−0.77
0.93
307.8
8.0
−0.8
0.82
614.6
21.7
142
−0.59
1.13
155.7
6.2
jf/jb
−
1
mass activity (j , mA mg
)
Pd
f
stability at 10 k s (%)
Tafel slope (mV dec−1)
165
160
591
1
832
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
Figure 3. (a) XPS spectra, (b) core level spectra of C 1s region, (c) Pd 3d region, and (d) Ni 2p region (with trendline) for Ni Pd100−x/G BACs.
x
Figure 4. CVs of graphene supported Ni Pd
BACs recorded in 1 M KOH electrolyte in the absence (dotted lines) and presence (solid lines) of
x
100−x
−
1
0.1 M EtOH at a scan rate of 50 mV s .
Pd 3d3/2, the maximum positive shifts cab be seen at Ni Pd /
ethanol are displayed in Figure 4. The dominating reduction
50
50
G (Figure S6), probably due to the higher degree of binary
alloyed formation. Figure 3d shows the core level spectra of Ni
peak of PdO at around −0.45 V in the reverse scan can be seen
in all CV curves of all Ni Pd
/G BACs in Figure 4, and
x
100−x
8
,27
2
p peaks, which shows a strong peak at ∼855 eV, which is an
similar CVs were also observed in previous report.
The
indication that the Ni was in an oxide form. Also, another three
salient peaks can be observed at 862, 874, and 880 eV that are
electrochemical surface area (ECSA) for all catalysts were
calculated using Coulombic charge (Q) for the reduction of
PdO, which is located at −0.45 V potential region (colored
attributed to the several oxo-Ni species (i.e., Ni(OH) , and
2
11,8,27
NiOOH). This is probably the reason to increase the oxygen at
peaks in Figure 4).
The values of the ECSAs are
%
and decrease the C/O ratio in all BACs. The intensity of all
summarized in Table 1. The highest value was obtained for
Ni Pd /G among all catalysts and which is 1.13 times higher
peaks were reflected to the at% of all elements. The numeric
results of all BACs have been listed in Table S2.
50
50
compared to Ni Pd100/G. It is known that the ECSA could
0
Electrochemical Ethanol Oxidation. The CVs of the
Ni Pd /G, Ni Pd /G, Ni Pd /G, and Ni Pd /G BACs
first investigated in Ar pursed 1 M KOH solution in the
reveal to what degree of Pd is utilized, this result implies a
better utilization of the comparatively small amount of Pd on
the surface of the Ni Pd /G BAC with the assistance of Ni
0
100
25
75
50
50
75
25
50
50
absence (dotted lines) and presence (solid lines) of 0.1 M
alloy formation.
1
833
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
The EOR on the Ni Pd /G, Ni Pd /G, Ni Pd /G, and
at Ni Pd /G and Ni Pd /G and less alloy formation than
0
100
25
75
50
50
25
75
50
50
Ni Pd /G BACs in alkaline medium was performed by CVs
amorphous Ni at Ni Pd /G. The ratio of the forward peak
75
25
75 25
(Figure 4). The CV curves are similar in appearance: two well-
current density (j ) and the backward peak current density (j ),
f b
defined peaks are observed in each curve. All two peaks are
located at ∼ −0.3 V (f) in the forward sweep and located at
j /j , is generally used to determine the poisoning tolerance to
f b
11,45
the carbonaceous species for Pd catalysts.
Also, a higher
∼
−0.4 V (b) in the backward sweep, strongly suggesting that
ratio of j /j is attributed to more efficient oxidation of EtOH
f
b
EOR can occur on all Ni Pd
/G BACs in a alkaline
and less accumulation of carbonaceous species to the catalysts.
x
100−x
2
2,24,43
medium, which is similar to other reports.
The
The onset potential, current density (j ), and j /j ratio were
f
f
b
magnitude of the forward scan peak is due to the oxidation
of freshly chemisorbed species that come from the adsorption
of EtOH; however, the backward scan peak is mostly associated
with the removal of carbonaceous species that not completely
determined by CV and are summarized in Table 1. The j /j of
f b
all BACs was higher than that of Ni Pd100/G, which indicates
0
that the Pd surface was less poisoned on the BACs than
Ni Pd100/G (inset).
0
42
oxidized during forward scan. However, among four CVs for
EOR, the current density was increased with the increasing of
Ni adding up to 50 at%, and the highest current density was
found with Ni Pd /G BAC, which is approximately 8, 4, and
The similar CV patterns indicate that the same EOR pathway
had occurred on BACs. The EOR goes through a dual-pathway
46
mechanism in alkaline media. In the C pathway, the strongly
1
adsorbed COads and CHx,ads species can be generated by C−C
50
50
1
.7 times superior than that of Ni Pd , Ni Pd100/G, and
bond breaking. In the C pathway, acetaldehyde and acetate
7
5
25
0
2
Ni Pd /G BACs, respectively. Very importantly, negatively
(partially oxidized) are produced as a result of the difficulty in
breaking the C−C bond. A previous report showed that the
carbonate formation toward EOR in alkaline media on Pd
25
75
47
shifted onset potential (−0.8 V) was observed on Ni Pd /G
50
50
BAC at room temperature (RT), which is also better than that
8
,10,15,19,20,22−24,43
of other previous reports.
The EOR mechanism on Pd
ceous intermediates including (CH CO) can be strongly
electrode was only 2.5%. Therefore, the EOR at Ni Pd100−x/G
x
1
9,28
47
suggests that the carbona-
probably proceed through the C pathway.
2
The Tafel plots measured for kinetic analysis are shown in
Figure 5 ( see also Figure S8). Tafel slopes derived from the
3
ads
adsorbed on the Pd surface and can block the activity (in the
forward scan). The rate-determining step will take place in the
reaction-starting region (∼−0.8 to −0.7 V), where Pd begins to
−
−
adsorb OH . Due to the OH adsorption, the strongly
adsorbed carbonaceous species swiftly exposed away and led
to the increase in current density (peak f). It is, however, at a
higher potential in the scan, and the PdO formation may block
the further adsorption of reactive species and lead to a decrease
44
in current density. In the backward sweep, the earlier formed
PdO would be reduced to active Pd, thus, leading to the
regaining of EOR current. The newly produced Pd surface
should be directly exposed to fresh ethanol. At high potential,
−
the quick adsorption of OH will help with the generation of
high current density and lead to a sharp backward peak. The
peak b will assign to the elimination of carbonaceous species
44
that are not completely oxidized in the forward scan, and
rationally, those intermediates may diffuse into the bulk
solution.
Figure 5. CVs of graphene supported Ni Pd
in 0.1 M EtOH containing 1 M KOH electrolyte at 50 mV s scan
BACs were recorded
x
100−x
−1
It is, however, when Ni is involved that the formed Ni(OH)2
rate. Inset: peak potentials and j /j variation with respect to Ni
f
b
of Ni Pd
BACs will enhance the EOR by increasing OH
x
100−x
loading.
species that lead to the increase of onset potential and current
density, as observed in Figure 5 (the forward scan’s current is
regularized to the mass of the Pd metal in Figure S7). In the
positive and negative scans, the hydrogen adsorption/
desorption region (−1.0 V to −0.8 V) was significantly
depressed because of the surface blocking by EtOH
decomposition. However, as the Ni amount increases, the
current density at both peak f and b increases accordingly on
linear region (from −0.7 to −0.4 V) are listed in Table 1. The
slope values are 160 and 142 mV dec− for Ni Pd /G and
1
2
5
75
Ni Pd /G BACs, respectively, which are lower than that of
5
0
50
−1
Ni Pd /G (165 mV dec ). Upon lower Tafel slopes, the
0
100
EOR charge-transfer kinetics on the catalysts were faster than
1
4,22
Ni Pd100/G. The lowest Tafel slope
0
was found on
Ni Pd /G that contained the highest degree of Ni and Pd
50
50
Ni Pd /G > Ni Pd /G> Ni Pd /G and suddenly decrease
0
100
25
75
50
50
alloy. Because the Ni has a strong interaction with electrolyte to
form oxygenated species, it helps the EOR process through
oxidation of intermediates and results in faster charge-transfer
on Ni Pd /G (inset). This behavior may occur for a couple of
75
25
probable reasons. First, the formation of Ni(OH) during the
2
−
−1
test increases OH and/or (OH)ads concentration into the
system and facilitates to the EOR, as depicted in eq 2
kinetics. The highest Tafel slope was found (591 mV dec ) on
Ni Pd /G, indicating slow kinetics due to Pd surface blocking
1
9,22
75
25
by excess Ni.
Ni(OH)2
‐
‐
Figure 6 illustrates the potentiodynamic measurements of
Pd − (CH CO) + OH ⎯⎯⎯⎯⎯⎯⎯ ⎯→ Pd − CH COOH + e
3
ads
3
EOR at Ni Pd /G catalyst in Ar saturated 1 M KOH solutions
(
2)
50
50
with various EtOH concentrations. As shown in Figure 6, with
increasing EtOH concentrations, the EOR starts at −0.8 V and
gives rise to a peak with the same onset potential. The current
Second, excess Ni may block the Pd sites at Ni Pd /G (as
observed in TEM image), leading to poor EOR due to the
active surface blocking. Third, there is excellent alloy formation
7
5
25
1
4,48
density is much better than that of other reports.
There are
1
834
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
Figure 7. CVs on Ni Pd /G catalyst in 1 M KOH at various scan
rates. Insets: plots of current density vs v (a) and f-peak shift vs ln of
scan rates (b).
Figure 6. (a) CVs of Ni Pd /G recorded in 1 M KOH electrolyte
50
50
50
50
1/2
−1
with the addition of various concentrations of EtOH at 50 mV s scan
rate. Inset: enlarged CVs at −1.0 to −0.4 V. (b) Peak potential and (c)
j /j vs concentrations of EtOH plots.
f
b
potential of −0.4 V. As can be seen in Figure 8a, initial rapid
current decreases could be attributed to the accumulation of
strongly adsorbed intermediates onto the surface of reactive
sites. Consequently, the relative current slowly decreased and
reached a pseudo-steady state. The time-dependent comparison
some significant observations involved in the experiment: First,
the onset potential of EOR on the Ni Pd /G is fixed (inset a).
The concentration independence of the onset potential for
5
0
50
EOR on the Ni Pd /G indicates that both EtOH dissociation
50
50
and the oxidation of intermediate behavior at Ni Pd /G are
can be seen in the inset. The stability of Ni Pd100−x/G BACs
5
0
50
x
the same. Also, low poisoning species may be formed at
Ni Pd /G. Second, the CV of Ni Pd /G shows a couple of
were also investigated by the long-term CV sweep for 500
cycles (Figure S10), and the first CV cycle of all Ni Pd100−x/G
50
50
50
50
x
peaks at −0.85 and −0.38 V in the forward sweep for low
concentration of EtOH (0.05 M) (inset a). The first peak is
from the partial reaction current for carbonate formation (in
alkaline condition), and the next peak appears due to the
BACs were compared with 20% Pt/C in Figure S11. The
current density of forward peaks from CVs is shown in Figure
8b. From Figure 8, the stability of Ni Pd /G catalyst in terms
50
50
of relative current is about 2.3, 2.7, and 3.5 times (from CA)
4
9
acetate formation (in alkaline condition). For higher
concentration, however, only one peak is observed in the
forward scan. Third, at the −1.0 to −0.4 V potential regions,
the lesser hysteresis was observed at the Ni Pd /G among the
and 2.4, 5.0, and 1.2 times (from CV) higher than Ni Pd100/G,
0
Ni Pd /G and Ni Pd /G, respectively. The indicates highly
7
5
25
25
75
1,20,22
efficiency and poisoning tolerance on Ni Pd /G catalyst.
50
50
Both CV and chronoamperometric measurements conclude
that Ni Pd /G has optimal EOR activity among Ni Pd100−x/G
5
0
50
forward and backward sweep for EOR. Although the hysteresis
is often caused by the strongly adsorbed species poisoning on
50
50
x
BACs.
50
the surface of electrodes, Ni Pd /G may involve less
5
0
50
poisoning as well as stable intermediate formation or the
oxidation of these species may be fast. Fourth, small variation of
peak potential ratio can be observed (inset b) with the
increasing of ethanol concentration, indicating a small
CONCLUSION
In summary, a series of electrocatalyst, G-supported Ni Pd
■
x
100−x
BACs has been successfully synthesized via a simple chemical
method and introduced to the EOR with superior catalytic
activity. We were able to prepare the Pd and Ni alloy at low
temperature with Ni being oxidized in the BACs through
combined characterization techniques. In particular, the
Ni Pd /G catalyst demonstrated higher peak current density,
47
uncompensated resistance (IR ) in the test system. Fifth,
u
the j /j is almost constant at around 0.8 but in the 0.05 M
f
b
ethanol (inset c). The nearly constant j /j is also indicating that
f
b
the stable intermediates are formed at the Pd site. The j /j for
f
b
50
50
0
.05 M ethanol is higher than others, which is probably due to
mass activity, and much negatively shifted onset potential (−0.8
V vs Ag/AgCl) in CV in EtOH 1 M KOH containing solution.
Also, it showed long-term stability compared to the other BACs
in both CA and CV. The Ni Pd /G catalyst had the lowest
minimal carbonate formation earlier, as mentioned above. Also,
the reproducibility has been checked in an 0.5 M EtOH-
containing 1 M KOH solution (Figure S9).
50
50
Figure 7 demonstrates the CVs on Ni Pd /G at different
5
0
50
Tafel slope and highest level of alloy content that lead to a
superior charge-transfer rate toward EOR. The overall
dependence of both ethanol concentration and scan rate
indicated that the Ni Pd /G is the more promising EOR
scan rates in an 0.4 M ethanol-containing 1 M KOH solution.
The relationship between both f and b-peak current density and
1/2
v
is displayed (inset a). The current densities for EOR
50
50
become bigger with an increasing scan rate. The linear
relationship can be observed between current densities and
catalyst. By comparing CVs and Tafel slopes, we concluded that
the EOR on Ni Pd /G catalyst might mostly proceed by
50
50
1
/2
v , which suggests the EOR on Ni Pd /G is a diffusion-
5
0
50
forming acetic acid. Also, we hope that this report will
encourage other groups to prepare G-supported Ni−Pd
materials for Pt-free catalytic applications including DEFCs.
5
1
controlled process. Furthermore, the f-peak potential
increases with increasing scan rate, and a straight line
relationship can be observed between f-peak potentials and ln
of scan rates (inset b), suggesting the EOR on Ni Pd /G is an
EXPERIMENTAL SECTION
Synthesis of Graphene-Supported Ni Pd100−x. Graphene oxide
5
0
50
■
51
irreversible electrode process.
x
CA has used to evaluate the stability of the Ni Pd100−x/G
(GO) was obtained from graphite powder via a modified Hummer’s
x
1
0,11,29,36
BACs in 0.5 M EtOH-containing 1 M KOH with an applied
method.
Briefly, a mixture of 1 g of graphite powder and 6 g
1
835
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
Figure 8. CA measurements and the plots of forward peak current density and cycle number (from Figure S10) of Ni Pd100−x/G BACs in 1 M KOH
x
with 0.5 M EtOH solution. Inset: time-dependent current density comparison.
of KMnO was added into a mixture of concentrated H SO (120 mL)
AUTHOR INFORMATION
4
2
4
■
and H PO (13 mL) in a round-bottom flux and was then stirred at 60
3
4
Corresponding Author
E-mail: swjeon3380@naver.com. Fax: +82 62 530 3389. Tel.:
82 62 530 0064.
°C for 16 h. The reaction was poured onto 200 mL of ice followed by
*
adding 2 mL of 30% H O after cooling at RT. The yellow product
2
2
+
was centrifuged at 4000 rpm, and the precipitate was then washed with
−
3
0% HCl, water (up to Cl was not removed), and EtOH. The
Notes
remaining solid material was dispersed in ether and filtered. The final
product, GO, was then vacuum-dried at 60 °C for 24 h.
The authors declare no competing financial interest.
The 0.01 M of K PdCl and 0.01 M of Ni(NO ) ·6H O were mixed
in a 20 mL vial (the atomic ratios of Ni and Pd were 0:100, 25:75,
0:50, and 75:25), followed by the addition of 5 mg of GO into the
2
4
3
2
2
ACKNOWLEDGMENTS
■
This research has supported by the National Research
Foundation of Korea (NRF) funded by the Ministry of
Education,Science and Technology (2010-0007864).
5
vial and prepare a yellow-brown homogeneous solution by ultrasonic
agitation with no visible aggregated particles. This solution was kept
under stirring at RT. Subsequently, 1 mL of 0.01 M ascorbic acid was
added under continuous stirring for 30 min to prepare metal
nanoparticles (NPs), and the color of the solution was changed into
ash. Thereafter, 5 μL of 65% hydrazine (N H ) was added under
REFERENCES
■
(1) Xu, H.; Ding, L.-X.; Liang, C.-L.; Tong, Y.-X.; Li, G.-R. NPG Asia
Mater. 2013, 5, e69−e79.
(2) Lee, E.; Park, I.-S.; Manthiram, A. J. Phys. Chem. C 2010, 114,
10634−10640.
2
4
stirring for another 30 min, followed by continuous heating in an oven
at 65 °C for overnight to produce a homogeneous black suspension.
The finishing products (Ni Pd /G, Ni Pd /G, Ni Pd /G, or
0
100
25
75
50
50
(3) Ahmed, M. S.; Jeon, S. J. Power Sources 2012, 218, 168−173.
(4) Lu, Y.; Jiang, Y.; Wu, H.; Chen, W. J. Phys. Chem. C 2013, 117,
2926−2938.
Ni Pd /G,) were collected through filtration and dried in a vacuum
75
25
oven at 40 °C for 12 h.
Characterizations of Physical Parameters. All TEM images and
EDS were obtained by TECNAI model FI-20 (FEI, Netherland). The
FTIR measurements were performed on an FTIR spectroscope
PerkinElmer). XPS were gained using a MultiLab 2000 with a 14.9
keV Al K X-ray source. XRD spectra were carried out on a Rigaku D/
(5) Xia, B. Y.; Wu, H. B.; Wang, X.; Lou, X. W. J. Am. Chem. Soc.
2012, 134, 13934−13937.
(6) Zhang, L.; Li, N.; Gao, F.; Hou, L.; Xu, Z. J. Am. Chem. Soc. 2012,
134, 11326−11329.
(
(7) Yamauchi, Y.; Tonegawa, A.; Komatsu, M.; Wang, H.; Wang, L.;
Nemoto, Y.; Suzuki, N.; Kuroda, K. J. Am. Chem. Soc. 2012, 134,
5100−5109.
max-2500, using filtered Cu Kα radiation.
Electrochemical Characterization. The suspension of Ni Pd
/
0
100
−
1
G, Ni Pd /G, Ni Pd /G, or Ni Pd /G in ethanol (1 mg mL )
(8) Dutta, A.; Datta, J. J. Mater. Chem. A 2014, 2, 3237−3250.
(9) Lim, B.; Kobayashi, H.; Yu, T.; Wang, J.; Kim, M. J.; Li, Z.-Y.;
Rycenga, M.; Xia, Y. J. Am. Chem. Soc. 2010, 132, 2506−2507.
25
75
50
50
75
25
was prepared separately by a 30 min sonication. A 5 μL portion of the
Ni Pd /G, Ni Pd /G, Ni Pd /G, or Ni Pd /G suspension was
0
100
25
75
50
50
75
25
(
(
10) Mondal, A.; Jana, N. R. ACS Catal. 2014, 4, 593−599.
11) Wang, Y.; Zhao, Y.; Yin, J.; Liu, M.; Dong, Q.; Su, Y. Int. J.
Hydrogen Energy 2014, 39, 1325−1335.
(12) Ma, L.; He, H.; Hsu, A.; Chena, R. J. Power Sources 2013, 241,
then dropped onto the prepolished glassy carbon electrode (GCE, 0.3
cm in diameter) trailed by dropping 5 μL of Nafion solution as a
binder (0.5 wt % in ethanol). The Pt/C suspension (20%) was
−1
prepared by dispersing 1 mg mL of 20% Pt/C in ethanol in the
presence of 5 μL of 5% Nafion solution in aliphatic alcohol and used 5
μL from the suspension onto GCE. All CVs and CA were taken using
a three-electrode potentiostat [CHI 700C electrochemical workstation
6
(
96−702.
13) Ding, L.-X.; Wang, A.-L.; Ou, Y.-N.; Li, Q.; Guo, R.; Zhao, W.-
X.; Tong, Y.-X.; Li, G.-R. Sci. Rep. 2013, 3, 1181−1188.
14) Du, W.; Mackenzie, K. E.; Milano, D. F.; Deskins, N. A.; Su, D.;
Teng, X. ACS Catal. 2012, 2, 287−297.
15) Shen, S. Y.; Zhao, T. S.; Xu, J. B.; Li, Y. S. J. Power Sources 2010,
95, 1001−1006.
16) Ahmed, M. S.; Kim, D.; Jeon, S. Electrochim. Acta 2013, 92,
68−175.
(17) Awasthi, R.; Singh, R. N. Carbon 2013, 51, 282−289.
18) Zhang, Z. H.; Ge, J. J.; Ma, L.; Liao, J. H.; Lu, T. H.; Xing, W.
Fuel Cells 2009, 2, 114−120.
(
(
U.S.A.)]. A Pt wire and Ag/AgCl electrodes were used as auxiliary
electrode and reference electrode, respectively. All electrochemical
experiments were performed in a high purity Ar purged (for at least 30
min) 1 M KOH solutions at RT.
(
1
(
1
ASSOCIATED CONTENT
■
(
*
S
Supporting Information
Supplemental TEM images, EDS spectra, XPS spectra, FTS
spectra, and other additional data. This material is available free
of charge via the Internet at http://pubs.acs.org.
(
19) Zhang, M.; Yan, Z.; Xie, J. Electrochim. Acta 2012, 77, 237−243.
(20) Lee, K.; Kang, S. W.; Lee, S.-U.; Park, K.-H.; Lee, Y. W.; Han, S.
W. O. ACS Appl. Mater. Interfaces 2012, 4, 4208−4214.
1
836
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837

ACS Catalysis
Research Article
(21) Demirci, U. B. J. Power Sources 2007, 173, 11−18.
(22) Zhang, Z.; Xin, L.; Sun, K.; Li, W. Int. J. Hydrogen Energy 2011,
3
(
6, 12686−12697.
23) Ding, K.; Yang, H.; Cao, Y.; Zheng, C.; Rapole, S. B.; Guo, Z.
Mater. Chem. Phys. 2013, 142, 403−411.
24) Ramulifho, T.; Ozoemena, K. I.; Modibedi, R. M.; Jafta, C. J.;
Mathe, M. K. Electrochim. Acta 2012, 59, 310−320.
25) Chun, C.; Yu, F.; Meng, H.; Chunying, W.; Guoqiang, S.;
Lingyan, Z. Appl. Catal. B: Environ. 2013, 142−143, 553−560.
26) Barakat, N. A. M.; Motlak, M.; Elzatahry, A. A.; Khalil, K. A.;
(
(
(
Abdelghani, E. A. M. Int. J. Hydrogen Energy 2014, 39, 305−316.
(
(
27) Singh, R. N.; Awasthi, R. Catal. Sci. Technol. 2011, 1, 778−783.
28) Seo, M. H.; Choi, S. M.; Seo, J. K.; Noh, S. H.; Kim, W. B.; Han,
B. Appl. Catal. B: Environ. 2013, 129, 163−171.
29) Zhang, M.; Yan, Z.; Sun, Q.; Xie, J.; Jing, J. New J. Chem. 2012,
6, 2533−2540.
30) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.;
(
3
(
Katsnelson, M. I.; Grigorieva, I. V.; Dubonos, S. V.; Firsov, A. A.
Nature 2005, 438, 197−200.
(31) Geim, A. K.; MacDonald, A. H. Phys. Today. 2007, 60, 35−41.
(32) Balandin, A. A.; Ghosh, S.; Bao, W.; Calizo, I.; Teweldebrhan,
D.; Miao, F.; Lau, C. N. Nano Lett. 2008, 8, 902−907.
33) Lee, C.; Wei, X.; Kysar, J. W.; Hone, J. Science 2008, 321, 385−
88.
34) Stankovich, S.; Dikin, D. A.; Dommett, G. H. B.; Kohlhaas, K.
(
3
(
M.; Zimney, E. J.; Stach, E. A.; Piner, R. D.; Nguyen, S. B. T.; Ruoff, R.
S. Nature 2006, 442, 282−286.
(
(
35) Guo, S.; Dong, S.; Wang, E. ACS Nano 2010, 4, 547−555.
36) Marcano, D. C.; Kosynkin, D. V.; Berlin, J. M.; Sinitskii, A.; Sun,
Z.; Slesarev, A.; Alemany, L. B.; Lu, W.; Tour, J. M. ACS Nano 2010, 4,
806−4814.
37) Zhang, H.; Jin, M.; Wang, J.; Kim, M. J.; Yang, D.; Xia, Y. J. Am.
4
(
Chem. Soc. 2011, 133, 10422−10425.
(
(
38) Niu, W.; Zhang, L.; Xu, G. ACS Nano 2010, 4, 1987−1996.
39) Wang, Y.; Wang, X.; Li, C. M. Appl. Catal. B: Environ. 2010, 99,
2
(
(
29−234.
40) Ahmed, M. S.; Han, H. S.; Jeon, S. Carbon 2013, 61, 164−172.
41) Li, R.; Wei, Z.; Huang, T.; Yu, A. Electrochim. Acta 2011, 56,
6
860−6865.
42) Maiyalagan, T.; Scott, K. J. Power Sources 2010, 195, 5246−
251.
43) Wang, Y.; Shi, F.-F.; Yang, Y.-Y.; Cai, W.-B. J. Power Sources
013, 243, 369−373.
44) Qin, Y. H.; Yang, H. H.; Zhang, X. S.; Li, P.; Zhou, X. G.; Niu,
L.; Yuan, W. K. Carbon 2010, 48, 3323−3329.
45) Feng, Y.-Y.; Liu, Z.-H.; Kong, W.-Q.; Yin, Q.-Y.; Du, L.-X. Int. J.
Hydrogen Energy 2014, 39, 2497−2504.
46) Lai, S. C. S.; Kleijn, S. E. F.; Ozturk, F. T. Z.; Vellinga, V. C. V.
R.; Koning, J.; Rodriguez, P.; Koper, M. T. M. Catal. Today 2010, 154,
2−104.
47) Zhou, Z.-Y.; Wang, Q.; Lin, J.-L.; Tian, N.; Sun, S.-G.
Electrochim. Acta 2010, 55, 7995−7999.
48) Chen, L. Y.; Chen, N.; Hou, Y.; Wang, Z. C.; Lv, S. H.; Fujita,
T.; Jiang, J. H.; Hirata, A.; Chen, M. W. ACS Catal. 2013, 3, 1220−
230.
49) Jiang, L.; Hsu, A.; Chu, D.; Chen, R. Int. J. Hydrogen Energy
010, 35, 365−372.
50) Lai, S. C. S.; Koper, M. T. M. Phys. Chem. Chem. Phys. 2009, 11,
0446−10456.
51) Zhao, Y.; Yang, X.; Tian, J.; Wang, F.; Zhan, L. Int. J. Hydrogen
Energy 2010, 35, 3249−3257.
(
5
(
2
(
(
(
9
(
(
1
(
2
(
1
(
1
837
dx.doi.org/10.1021/cs500103a | ACS Catal. 2014, 4, 1830−1837