Unique Half Embedded/Exposed PdFeCu/C Interfacial Nanoalloy as High-Performance Electrocatalyst for Oxygen Reduction Reaction

Article abstract of DOI:10.1002/cctc.201900469

Design of high-performance non-Pt electrocatalyst for fuel cell applications is greatly anticipated. Herein, we have developed a unique half-embedded and half-exposed interfacial PdFeCu nanoalloy anchored onto carbon matrix. The stable electronic coupling between the carbon matrix and PdFeCu nanoalloy possess very fast interfacial electron transfer which in turn enhances the electron conductivity. This makes the trimetallic nanoalloy high performing oxygen reduction reaction (ORR) electrocatalyst in both basic and acidic media. The PdFeCu/C nanoalloy exhibits enhanced electrochemically active surface area than various PdFe/C bimetallics as well as benchmark 20 wt% Pt/C and Pd/C. As a result, it offers larger active sites for ORR and eased the electron transport during the electrocatalysis. It exhibits 1.5- and 2.4-fold higher mass activity in comparison to the Pt/C and Pd/C. Furthermore, it exhibits long-term stability and low onset potential compared to those of the other catalysts. Thus, the present investigation shows potential strategy for the design and synthesis of Pt-free electrocatalyst with remarkable catalytic activity and stability.

Full text of DOI:10.1002/cctc.201900469

Accepted Article
Title: Unique Half Embedded/Exposed PdFeCu/C Interfacial Nanoalloy
as High-Performance Electrocatalyst for Oxygen Reduction
Reaction
Authors: Biraj Jyoti Borah, Himadri Saikia, Chiranjita Goswami, Kumar
Kashyap Hazarika, Yusuke Yamada, and Pankaj Bharali
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201900469
Link to VoR: http://dx.doi.org/10.1002/cctc.201900469
A Journal of
www.chemcatchem.org

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
Unique Half Embedded/Exposed PdFeCu/C Interfacial Nanoalloy
as High-Performance Electrocatalyst for Oxygen Reduction
Reaction
Biraj Jyoti Borah,a Himadri Saikia,a Chiranjita Goswami,a Kumar Kashyap Hazarika,a Yusuke Yamada,b
and Pankaj Bharalia*
Abstract: Design of high-performance non-Pt electrocatalyst for fuel
cell applications is greatly anticipated. Herein, we have developed a
unique half-embedded and half exposed interfacial PdFeCu
nanoalloy anchored onto carbon matrix. The stable electronic
coupling between the carbon matrix and PdFeCu nanoalloy possess
very fast interfacial electron transfer which in turn enhances the
electron conductivity. This makes the trimetallic nanoalloy high
performing oxygen reduction reaction (ORR) electrocatalyst in both
basic and acidic media. The PdFeCu/C nanoalloy exhibits enhanced
electrochemically active surface area than various PdFe/C
bimetallics as well as benchmark 20 wt% Pt/C and Pd/C. As a result,
it offers larger active sites for ORR and eased the electron transport
during the electrocatalysis. It exhibits 1.5 and 2.4 fold higher mass
activity in comparison to the Pt/C and Pd/C. Furthermore it exhibits
long term stability and low onset potential compared to those of the
other catalysts. Thus, the present investigation shows potential
strategy for the design and synthesis of Pt-free electrocatalyst with
remarkable catalytic activity and stability.
carbon on carbon nanotube (Pt@CNx/CNT), where the carbon
layer effectively prevents the Pt NPs from migration, dissolution,
detachment and aggregation during durability test and heat
treatment at 900 C. In another report, PtFe NPs with a very thin
„„dual purpose‟‟ N-doped carbon shell has been effectively
synthesized.[6] The carbon layer advantageously prevents PtFe
NPs from agglomeration during the thermal annealing at ~700
C. Even at this condition, the layer was still permeable for the
reactant molecule and the catalytic activity of the NPs was not
affected. Although, the processes exhibit good electrocatalytic
activity, they have serious operating limitations. Time-consuming
synthetic procedure, use of surfactants and very high pyrolysis
temperature (700-900C) effectively limits the electrocatalysts
for practical application. Thus, the engineering of highly stable
and efficient strategy for the realization of active fuel cell EC is
still desirable. In particular, fuel cell ECs for cathodic oxygen
reduction reaction (ORR) need to be very effective for overall
performance of fuel cells.[7b-e] This is because of the slow kinetics
associated with the ORR process which limits active fuel cell
materialization. In addition to this, the ORR activity of noble
metals and their compounds are greatly influenced by lattice
strain, electronic effect and surface defects.[8a,b,] The lattice
strain plays an important role in the ORR of noble metal
electrocatalysts, whereas the compressive lattice strain changes
the d-band center of the metal catalysts which weakens the
surface oxygen adsorption property of the catalyst surface.[9]
From the electronic effect, the alteration of the d band center for
Pt or Pd through the formation of alloys with second metal could
increase the kinetics of the ORR.[10a,b] On the other hand
abundant surface defect on NPs also alter the electrocatalytic
performance by proving the accessible active sites for the
reactant molecules.[8]
Inspired by various strategies to improve ORR performance,
we herein report the design of unique half embedded and half
exposed PdFeCu nanoalloy on carbon matrix with multiple
surface defects at relatively low temperature as presented in
Scheme 1. The half embedded portion in carbon matrix
effectively prevents the PdFeCu NPs from detachment,
migration, dissolution, and aggregation during ORR process.
The half exposed part, i.e. without protective carbon layer,
become highly active as the reactant molecules (O2 diffusion
layer) could directly interact with these sites.[8a] To the best of
our knowledge, there is no report on the synthesis of half
embedded and half exposed PdFeCu nanoalloys with lattice
strain, and abundant surface defects for ORR in both acidic and
alkaline electrolytes. It exhibits 1.5 and 2.4 fold higher mass
activity in comparison to the Pt/C and Pd/C. Furthermore it
shows long term stability and low onset potential compared to
those of the Pd/C, Pt/C and other compositions of PdFe/C. Thus,
Introduction
The low physical and chemical stability of nanoparticles (NPs)
mainly restrict the practical application of electrocatalysts (ECs)
for fuel cell. Throughout the fuel cell operating conditions, NPs
are often easily oxidized, dissolved in electrolyte, or separate
from the support and agglomerated into larger particles, and
thus losing their electrochemical catalytic performance during
cycling.[1] Therefore, numerous efforts have attempted to
address the hurdles. These include modification of NPs with
protective layer by carbon and/or modified carbon shells,[2]
inorganic oxide[3] and graphitic material[4,5] etc. Eventually the
coated NPs exhibit improved electrocatalytic activity and long
term durability compared to the uncoated one. Unfortunately, it
was assessed that the protective layer can also decrease the
activity by blocking the active sites of the EC.[6] For instance,
Guo et al.[7a] reported embedded Pt nanocrystals into N-doped
[a]
[b]
B.J. Borah, Dr. H. Saikia, C. Goswami, K.K. Hazarika, Dr. P. Bharali
Department of Chemical Sciences, Tezpur University, Tezpur,
Assam-784 028, India
E-mail: pankajb@tezu.ernet.in
Prof. Y. Yamada
Department of Applied Chemistry & Bioengineering, Graduate
School of Engineering, Osaka City University, 3-3-138 Sugimoto,
Sumiyoshi-ku, Osaka 558-8585, Japan
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
this present strategy provides an efficient route for the synthesis
of unique structured PdFeCu/C nanoalloy by altering d-band
centre with rich surface defects and lattice strain.
Pd (JCPDS no. 87-0641, a = 0.3961 nm, d = 0.228 nm) and
larger than bulk Fe (JCPDS no. 88-2324, a = 0.343 nm, d =
0.198 nm), and bulk Cu (JCPDS no. 85-1326, a = 0.361 nm, d =
0.208 nm) indicating the formation of PdFe bimetallic and
PdFeCu trimetallic alloys. The relatively smaller values of „a‟
and„d‟ for Pd3Fe0.5Cu0.5/C, as compared to the other PdFe alloys
also implies effective lattice contraction. The Figure S3B
(Supporting information) shows the N2 adsorption-desorption
isotherm of Pd3Fe0.5Cu0.5/C which exhibits a high BET specific
surface area of 69.8 m2/g and a large pore volume of 0.104 cc/g.
Results and Discussion
The compositions of as prepared nanocatalysts are determined
by EDS and ICP-OES analyses. The EDS analysis of
Pd3Fe0.5Cu0.5/C shows that only Pd, Fe, Cu and C are present in
the EDS pattern with 22 wt% loading of metals on carbon
matrix. The corresponding EDS elemental maps show that
homogeneous dispersion of metals on the carbon (Figure S1,
Supporting information).
Figure 1. Typical (A) low, (B,C) medium, and (D-F,H) high resolution-TEM
images of Pd3Fe0.5Cu0.5/C, (G) corresponding SAED pattern, (E) is magnified
HR-TEM image taken from (D), marked by the rectangle.
Scheme 1. Synthesis of unique half embedded and half exposed PdFeCu
nanoalloy on carbon matrix.
The shape and microstructure of Pd3Fe0.5Cu0.5/C NPs are
further studied by TEM and HR-TEM and presented in Figure 1.
From the TEM images (Figure 1A-C) it could be observed that
many nearly-spherical Pd3CuFe NPs of dark contrast are highly
dispersed on the larger particles of Vulcan XC72R carbon of
lighter contrast. The average particle-size of Pd3Fe0.5Cu0.5 NPs
is estimated to be about ~ 6-7 nm (inset in Figure 1A). The NPs
exhibit various crystal defects on their surfaces, for instance
vacancy, dislocation and low-coordinate atomic steps, as
manifested in HR-TEM images (Figure.1D,E). Because of these
coordinative unsaturation the NPs can serve as catalytic sites,
successfully enhancing the electrocatalytic activity.[14a]
Furthermore, as shown in Figure 1F, Pd3CuFe NPs are half
embedded in the carbon matrix which generate rich interface
that favors high electrocatalytic activity. The concentric circles
with well-defined crystalline facets in the SAED pattern (Figure
1G) infer high crystallinity of the Pd3CuFe/C NPs. The
comprehensible d-spacings ∼0.22 nm (Figure 1H) can be
observed in the lattice fringes of a Pd3CuFe/C NPs, close to the
fcc Pd (111) facets, signifying high crystallization. The detailed
morphological characterization of Pd3Fe/C is also performed by
TEM and HR-TEM analyses and presented in Figure S4
(Supporting information). As estimated, the Pd3Fe/C alloy shows
an analogous morphology and surface/interface structure with
that of Pd3Fe0.5Cu0.5/C NPs. Based on the above discussion, we
could deduce that spherical and nearly homogeneous Pd-based
NPs can be effectively synthesized by simple solvothermal
The Pd, Fe and Cu atomic ratio are ca. 74.81%, 14.56% and
10.61%, which corroborate well with ICP-OES analyses. The
loading of the metal obtained by TGA analysis is ∼22% in the
case of the Pd3Fe0.5Cu0.5/C (Figure S2, Supporting information).
It displays a sudden weight loss of the sample at about ~350 °C
which is due to the oxidation of carbon to carbon dioxide in the
sufficient oxidative environment. The residue obtained after
burning of all the carbonaceous materials gives information
about the loading of the total metal content. The powder XRD
patterns of Pd3Fe0.5Cu0.5/C, Pd3Fe/C, PdFe/C, and PdFe3/C
(Figure S3A, Supporting information) show mainly characteristic
peaks of face centred cubic (fcc) crystalline Pd signifying
multiphase disordered structure. The diffraction peaks centred at
~40o, ~47o, and ~68o correspond to the (111), (200), and (220)
reflections, respectively. These peaks are slightly shifted to
higher angles relative to the pure Pd (JCPDS no. 87-0641)
indicating the formation of an alloyed nanostructure.[11] A broad
peak observed at around ∼25 is ascribed to the (002) reflection
of a hexagonal structure in Vulcan XC-72R carbon. From
Bragg‟s law[12] the cell parameter (a) values of 0.3910 (PdFe3/C),
0.3906 (PdFe/C), 0.39161 (Pd3Fe/C) and 0.38918 nm
(Pd3Fe0.5Cu0.5/C) are calculated. The calculated „d‟ values of
0.2261 (Pd3Fe/C) and 0.2246 nm (Pd3Fe0.5Cu0.5/C) agrees well
with the spacing between planes observed using HR-TEM (See
Figure 1H). These values are rather smaller than those for bulk
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
method without using any harmful structure directing solvents
and surfactants. XPS is extensively performed to elucidate the
chemical composition of NPs. Figure 2A-D illustrates the XP
spectra of C 1s, Pd 3d, Fe 2p and Cu 2p regions of
Pd3Fe0.5Cu0.5/C. The C 1s core-level XP spectrum (Figure 2A) of
Pd3Fe0.5Cu0.5/C is slightly asymmetric and hence, can be
deconvoluted into three different peaks viz., a prominent peak at
284.6 eV along with two small peaks at 285.9 and 289.8 eV,
consistent to form –C=C−, −C−O−, and O=C−O− bonds,
respectively.[ 14b,c]
efficient PdFe nanoalloy formation (See Figure S5 in the
Supporting information). The electrocatalytic behavior of
Pd3Fe0.5Cu0.5/C and binary PdFe/C NPs modified glassy carbon
electrode (GCE) are measured in N2 and O2 saturated 0.1 M
KOH, as shown in Figure 3 and Figure S6-S8 in the Supporting
information. As presented in the cyclic voltammogram (CV) in
Figure 3A, a strong and well-defined reduction peak is observed
in O2-saturated 0.1 M KOH solution whereas no such feature is
observed in N2-saturated electrolyte. This signifies that
Pd3Fe0.5Cu0.5/C NP shows good ORR activity. To determine the
kinetics of ORR, rotating disk electrode (RDE) measurement is
performed in O2 saturated 0.1 M KOH solution at a rotating rate
of 400, 900, 1600, 2500 and 3600 rpm as shown in Figure 3B.
Based on the ORR polarization curves, the number of electrons
(n) transferred during ORR can be evaluated using the following
Koutecky−Levich (K−L) equation.[18b]
1/j=1/jk+ 1/B ω0.5
B= 0.62 nF(Do2)2/3v1/6 Co2
Where, j = current density, jk = kinetic current density, ω=
rotating rate of the electrode, B = value of slope obtained from
the K−L plots, F = Faraday constant (96485 C mol−1), Do2
=
diffusion coefficient of O2 in 0.1 M KOH solution, v = kinetic
viscosity (0.01 cm2 s−1), Co2 = bulk concentration of O2 (1.2 ×
10−6 mol cm−3), and n = the number of electrons transferred in
the ORR process. The K−L (j−1 vs. ω−1/2) plots obtained on the
basis of the corresponding ORR polarization curves for
Pd3Fe0.5Cu0.5/C NP at a potential of −0.40, −0.50, −0.60, and
−0.70 V show good linearity (Figure 3C) thereby implying a first-
order kinetics for ORR. Figure 3D gives the corresponding plot
of „n‟ vs. potential (V). It can be seen that the „n‟ values from
−0.40 to −0.70 V are ~4.0, suggesting that the ORR mechanism
on Pd3Fe0.5Cu0.5/C NP follows the direct “4e−” pathway (O2
+
Figure 2. (A) C 1s, (B) Pd 3d, (C) Fe 2p, and (D) Cu 2p XP spectra of
2H2O + 4e− → 4OH−). The electrocatalytic ORR mechanism of
binary PdFe/C NPs also follow the direct “4e−” pathway in the
case of Pd3Fe/C and PdFe/C NPs whereas “2e−” pathway
possibly in the case of PdFe3/C NPs (See Figure S6-S8 in the
Supporting information).
Pd3Fe0.5Cu0.5/C.
The Pd 3d XP spectrum as presented in Figure 2B displays
a doublet signal with binding energies of 335.7 and 341.0 eV for
Pd 3d5/2 and Pd 3d3/2, respectively, corresponding to the Pd
signal. Additionally, small doublets observed around 336.9 and
342.5 eV could be indexed to the Pd 3d5/2 and Pd 3d3/2 peaks of
PdO.[15] Figure 2C shows the Fe 2p core-level spectrum for the
NP. Two prominent peaks at 711.1 and 724.9 eV corresponds to
the Fe 2p3/2 and Fe 2p1/2 core-level energies of iron oxide. The
binding energy positions clearly indicates that the iron oxide in
the NP is in the form of magnetite (Fe3O4).[16] The emerging
photoelectron peaks of Fe3+ may be attributed to the easy
oxidation of zero valent iron (ZVI) in air, which could result in a
layer of iron oxide around the ZVI NPs. The peaks observed at
932.0 and 951.8 eV from the Cu 2p3/2 and Cu 2p1/2 spin−orbit
doublets, respectively, correspond to a metallic copper (Figure
2D), which is consistent with Cu NPs.[17] The binding energy
peaks of the three constituent elements Pd 3d5/2, Fe 2p3/2 and
Cu 2p3/2 are shifted as compared to that of the standard
literature values. For Pd, a positive shift of ~0.50 eV, while, for
Fe and Cu negative shifts of ~0.67 eV and ~0.50 eV,
respectively, are observed in the binding energy values. The
strong charge transfer between Pd, Fe and Cu is responsible for
the alloy formation.[18a] For the bimetallic Pd3Fe/C, the clear shift
of binding energy for both Pd and Fe peaks in comparison to the
standard monometallic counterparts is a strong evidence of
Figure 3. CV plots of Pd3Fe0.5Cu0.5/C in N2- and O2-saturated 0.1 M KOH
solution with a scan rate of 50 mV s−1, (B) Rotating rate-dependent ORR
polarization curves for respective NP with the scan rate of 10 mV s−1, (C) K-L
plots of j−1 vs. ω−1/2 at different potential obtained from (B), and (D) The plot of
the number of transferred electrons vs. potential for Pd3Fe0.5Cu0.5/C NPs.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
To elucidate the electrocatalytic behavior of the NPs, the
electrochemically active surface areas (ECSA) are calculated
using CV curves in N2 saturated 0.1 M KOH (Figure 4A). The
peaks in the potential range −0.5 0 V originate from the
palladium oxide reduction.[18c] The calculated ECSA of
Pd3Fe0.5Cu0.5/C (63.22 m2g1) is higher than Pd3Fe/C (32.99
m2g1), PdFe/C (17.87 m2g1), PdFe3/C (8.3 m2g1), standard
Pt/C (8.7 m2g1), and standard Pd/C (1.42 m2g1). The higher
ECSA values of the binary and ternary NPs can be attributed to
the synergistic effect between the elements. Figure 4B shows
the ORR polarization curves of Pd3Fe0.5Cu0.5/C, PdFe3/C,
PdFe/C, Pd3Fe/C, Pd/C, and Pt/C in O2-saturated 0.1 M KOH
393.74 mV dec1) < Pt/C (84 and 443.4 mV dec1) > Pd/C
(92.38 and 603.2 mV dec1), in the high and low potential range,
respectively. This indicates that Pd3Fe0.5Cu0.5/C has higher
efficiency towards ORR as compared to that of the other catalyst
systems studied and several reported Pd-based catalysts as
detailed in Table S1 (Supporting information).
The mass activity (MA) and specific activity (SA) are the two
important parameters for better assessment of the catalytic
activity. The MA of various NPs under different potential at 1600
rpm are presented in Figure 4D. The MA and SA of
Pd3Fe0.5Cu0.5/C at 0.4 V (595.8 mA mg1 and 84 mA cm2) are
much higher than Pt/C (382.2 mA mg1 and 53.89 mA cm2) and
Pd/C (241.6 mA mg1 and 34.06 mA cm2) under the identical
experimental conditions. The MA and SA values of
Pd3Fe0.5Cu0.5/C are also larger than those of the PdFe/C NPs
which follows the order: Pd3Fe/C (434.9 mA mg1 and 61.32 mA
cm2) > PdFe/C (349 mA mg1 and 49.20 mA cm2) > PdFe3/C
(288.5 mA mg1 and 40.67 mA cm2). These experimental
findings are evident of the enriched kinetic ability of the
Pd3Fe0.5Cu0.5/C, making it an auspicious ORR electrocatalyst.
electrolyte at a rotation of 1600 rpm. The onset potential (Eonset
0.044 V) and half-wave potential (E1/2 0.145 V) of
Pd3Fe0.5Cu0.5/C NP are found to be more positive than those of
standard Pd/C (Eonset = 0.103 V, E1/2 = 0.212 V), Pt/C (Eonset
0.069 V, E1/2 = 0.154 V), PdFe3/C (Eonset = 0.047 V, E1/2
0.168 V), PdFe/C (Eonset = 0.036 V, E1/2 = 0.194 V), and
Pd3Fe/C (Eonset = 0.030 V, E1/2 = 0.147 V) under identical
conditions, suggesting superior ORR activity of the ternary
Pd3Fe0.5Cu0.5/C NP. The overall ORR behavior follows the order:
Pd3Fe0.5Cu0.5/C > Pd3Fe/C > PdFe/C > PdFe3/C > Pt/C > Pd/C.
Moreover, the apparent current density values follow the same
order.
=
=
=
=
Figure 5. (A) CA curves of different electrocatalysts, recorded at −0.3 V in O2-
saturated 0.1
M KOH solution with a rotation rate of 1600 rpm, ORR
polarization curves of (B) Pt/C and (C) Pd3Fe0.5Cu0.5/C at 1600 rpm before and
after the stability test at different potential scans, (D) Comparison of mass
activity after the stability test at different potential scans at 1600 rpm.
Figure 4 (A) CV plots of all the catalysts at N2-saturated 0.1 M KOH solution at
a scan rate of 50 mV s−1, (B) Comparison of rotating rate-dependent ORR
polarization curves for different catalyst at 1600 rpm, (C) Tafel plots, and (D)
Mass activities at 1600 rpm. (These values are calculated as per the loading
mass of the NPs).
The high durability of the electrocatalysts towards ORR is
also an essential factor for energy conversion systems. Thus to
check the durability, chronoamperometry (CA) test was
performed in O2-saturated 0.1 M KOH at 0.3V (vs. Ag/AgCl)
with a rotation speed of 1600 rpm. It can be observed from
Figure 5A that the initial activity decay for Pt/C and Pd3Fe/C NP
are ~25.5% and ~3.8%, respectively after 7000 sec of testing,
while the Pd3Fe0.5Cu0.5/C NP unveils only a loss of ~1.4% of its
initial activity after 7000 sec, under identical experimental
conditions. This demonstrates the excellent stability of the
Pd3Fe0.5Cu0.5/C NPs. The structure and morphology of the
Pd3Fe0.5Cu0.5/C NPs as measured by TEM analyses after the CA
test are shown in Figure S10 (Supporting information). There is
meager change in the morphology of the Pd3Fe0.5Cu0.5/C NPs
after the CA test. To support the structural stability, an
The Tafel plots derived from the ORR polarization curves for
all NPs are presented in Figure 4C and Figure S9 (Supporting
information). The Tafel slope calculated for Pd3Fe0.5Cu0.5/C in
the high potential range (i.e., 3.9 V to 2.5 V) is 70.35 mV dec1,
while the other in the low potential range is 311.93 mV dec1.
The two Tafel slopes indicates different ORR behaviors in
different potential ranges.[19] These two slopes are comparatively
smaller than those of Pt/C (84 and 443.4 mV dec1), Pd/C (92.38
and 603.2 mV dec1) as well as for other catalysts compositions
and follows the order: Pd3Fe/C (72 and 348.66 mV dec1) <
PdFe/C (74.59 and 373.17 mV dec1) < PdFe3/C (78.92 and
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
accelerated durability test (ADT) is also performed for 1000
cycles for Pt/C and Pd3Fe0.5Cu0.5/C NPs at 1600 rpm (Figure
5B,C). The MA of the Pd3Fe0.5Cu0.5/C NPs has been deteriorated
to a lower extent (5.17%) in comparison to Pt/C (11%) after
1000 potential CV cycles. There is a clear drop of onset potential
in the case of Pt/C, whereas, the onset potential remains almost
same after 500 and 1000 cycles for Pd3Fe0.5Cu0.5/C. The E1/2 of
Pt/C (26 mV) also shows a larger change compared to that of
Pd3Fe0.5Cu0.5/C (10 mV). Furthermore, methanol tolerance ability
of Pd3Fe0.5Cu0.5/C was performed by adding methanol (0.1 M)
into the O2-saturated 0.1 M KOH electrolyte in the CA test
(Figure S11, Supporting information). After the addition of
methanol (t = 150 s), a remarkable loss of current is observed in
case of standard Pt/C signifying fast degradation of its
electrocatalytic activity whereas there is no loss of activity in
case of Pd3Fe0.5Cu0.5/C up to 700 s. The above discussion
indicates that Pd3Fe0.5Cu0.5/C NP is the most efficient, stable and
methanol tolerant electrocatalyst. To generalize the ORR
behavior of the Pd3Fe0.5Cu0.5/C NPs in the acidic electrolyte, we
conducted electrocatalytic studies in 0.5 M H2SO4 solution. As
presented in Figure S12 (Supporting information), it is apparent
that Pd3Fe0.5Cu0.5/C is very active towards ORR in 0.5 M H2SO4
solution as well. It can be seen that the „n‟ values from 0.1 to
0.05 V are ~4.0, suggesting that the ORR mechanism on
Pd3Fe0.5Cu0.5/C NPs follows the direct “4e−” pathway (O2 + 4H +
4e− → 2 H2O). To evaluate the durability, CA test was performed
in O2-saturated 0.5 M H2SO4 at 0.3 V (vs. Ag/AgCl) at 1600 rpm
for Pd3Fe0.5Cu0.5/C and Pt/C NPs. It is noticed from the Figure
S13 (Supporting information) that the activity decay for
Pd3Fe0.5Cu0.5/C and Pt/C NPs are ~40% and ~70.5%,
respectively after 7000 sec of testing. This shows that
Pd3Fe0.5Cu0.5/C is more stable even in acidic electrolyte over
Pt/C NPs. However, the stability of NPs in acidic electrolyte is
inferior in comparison to the basic electrolyte.
we have calculated lattice strain from XRD data that
Pd3Fe0.5Cu0.5/C has the smallest lattice parameter among all
compositions of PdFe/C, Pd/C and Pt/C. Moreover, the
formation of alloy with the second metals can cause the
alteration of d band center of Pd and Pt metal. For oxygen
adsorption, an upshift of d-band centre can increase the
interaction of the 2p states oxygen with the metal d states, i.e.,
forms a stronger metal-oxygen bond. On the other hand, a
downshift of d-band causes a weak interaction with oxygen. The
positive shifts of binding energy represent a downshift of d-band
center with respect to Fermi level. The downshift of Pd d-band
center can further be closely correlated with the weaker Pd-O
bond energy, by means of which the intermediate OHad species
adsorption on the Pd surface can be decreased,[10,21] thus
sustaining the more active sites for ORR and at the end resulting
in the enrichment in ORR activity.[10,22] The alteration of the d
band center for Pt or Pd through the formation of alloys with
second metals could increase the kinetics of the ORR. There are
seminal theoretical reports of ORR where DFT has been used
over various catalyst alloys with compositions of Pt3X or Pd3X (X
= Cu, Ni, Co, Fe, Ag, Au, Ir, etc.) by using d-band center
model.[21,23] The upshift of ~0.50 and ~0.40 eV in the binding
energies of Pd 3d5/2 for Pd3Fe0.5Cu0.5/C and Pd3Fe/C NPs in
comparison to Pd/C as shown is Figure S14 (Supporting
information) implies the aforesaid down shift trends of Pd d-band
center which leads to the enhancement of the catalytic
performance and acceleration in terms of mass activity of the
ORR activity.
Conclusions
In summary, an efficient and long-term stable ORR
The enhanced ORR performance of Pd3Fe0.5Cu0.5/C NPs is
due to their unique structural and compositional superiorities.
From the HR-TEM images, the unique half-embedded and half
expose interface structure was seen. i.e., one-half of NPs are
embedded into the carbon matrix and another part exposed from
the carbon matrix. Therefore, there is a stable electronic
coupling between the carbon matrix and Pd3Fe0.5Cu0.5 NPs. This
means that the half embedded part possess very faster
interfacial electron transfer ultimately enhancing the conductivity
of the Pd3Fe0.5Cu0.5 NPs. Moreover the other exposed part is
also highly active as the reactant molecule (O2 diffusion layer)
directly interacts with the catalyst layer. The carbon matrix also
acts as superior support as they effectively prevent the NPs
aggregation, improving the stability of the Pd3Fe0.5Cu0.5 in the
electrocatalyst,
Pd3Fe0.5Cu0.5/C
has
been
synthesized
successfully. The modified lattice strain and d-band shift in
Pd3Fe0.5Cu0.5/C improves the catalytic efficiency relative to
various compositions of PdFe/C. The unique half-embedded and
half exposed interface structure of the nanoalloy offers stable
electronic coupling between the carbon matrix and Pd3Fe0.5Cu0.5
NPs. As a result it impressively displays superior activity and
stability towards ORR in comparison to commercial Pd/C and
Pt/C electrocatalysts.
Experimental Section
Chemicals. The chemicals used are palladium(II) chloride (PdCl2),
copper(II) chloride dihydrate (CuCl22H2O), iron(II) chloride tetrahydrate
(FeCl24H2O), sodium hydroxide (NaOH), hydrazine hydrate
(NH2NH26H2O), potassium hydroxide (KOH) etc. were purchased from
Merck, India. Nafion, 20wt% Pt/C and 20wt% Pd/C were purchased from
Alfa Aesar. Vulcan XC-72 R was obtained from Cabot Corporation in the
form of research gift. All the chemicals were used as received and
distilled water was used throughout the experiment.
electrochemical
process.
Furthermore,
Pd3Fe0.5Cu0.5/C
nanoalloy exhibits many crystal defects on their surfaces,
including vacancy, atomic steps and dislocation defects, that can
work as abundant catalytically active sites accessible to the
reactant, ultimately enhancing ORR activity. One of the salient
parameters to understand the enhancement of ORR activity is
the lattice strain effect in catalysts. The lattice strain changes the
d-band center of the metal catalyst, which resolve the surface
oxygen adsorption property on the catalyst surface.[9,20] A down
shift in the d-band center of Pd weakens the binding energy of
Synthesis of Pd3Fe0.5Cu0.5/C NPs. The 20wt% of Pd3Fe0.5Cu0.5/C NPs
was synthesized by the typical solvothermal method. To prepare the NPs,
to a mixture solutions of 25 mL distilled water and 25 mL C2H5OH was
mix, 0.197 mmol PdCl2 (0.0350 g), 0.065 mmol FeCl24H2O (0.0132 g),
0.065 mmol CuCl22H2O (0.0112 g), and 0.1 g Vulcan XC-72 R were
absorption of intermediates such as OH− and COads thereby
ads
conserves more active sites on the catalyst surface. In this case,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
added sequentially at room temperature with constant stirring. The
mixture solution was ultra-sonicated for 5 min and 30 mL 0.05 M (0.2 g)
NaOH solution was added dropwise at room temperature under constant
stirring. After ca. 20 min, 2 mL 85% hydrazine hydrate was added
dropwise to the mixture and stirred vigorously for another 25 min. The
solution was then transferred into a Teflon cup (250 mL capacity) in a
stainless steel autoclave. The autoclave was sealed and kept in an oven
at 120 C for 6 h and allowed to cool naturally (~2.5 h); the final product
was washed several times with distilled water and then 50% C2H5OH to
remove the unreacted NaOH and Cl, and finally collected by
centrifugation. The product was dried in a vacuum oven at 55 C.
(0.1 M KOH and 0.5 M H2SO4) solution with scan rate 50 mV s-1. RDE
measurements were performed in O2-saturated 0.1 M KOH. Oxygen was
purged for at least 30 min before measurement and continuously bubbled
through the electrolyte, in order to ensure the saturation of the electrolyte
with O2 and then blanketing the solution with an O2 atmosphere during
the entire experiment and all experiments were performed at room
temperature. The electrochemical active surface area (ECSA) of the
catalysts was evaluated by CVs in N2-saturated 0.1 M KOH solution. For
Pd-based catalysts, to evaluate the ECSA of the samples the columbic
charge for the reduction of Pd−O monolayer, formed on Pd catalysts at
the forward scan, was applied. The ECSA were calculated using the
following equation: ECSA = Q/SL, where L is the Pd loading, Q is the
collected charge that calculated from the Pd−O stripping, and S is a
constant of 210 μC cm−2 that assumes a monolayer of Pd−O on the
surface. The stability of the ORR electrode was tested using
chronoamperometry. The chronoamperometric response for the ORR
was obtained either at −0.3 V (vs. Ag/AgCl) in O2-saturated 0.1 M KOH
or 0.3 V (vs. Ag/AgCl) in O2-saturated 0.5 M H2SO4 solution at 1600 rpm.
CA response also performed to see the methanol tolerance ability at −0.3
V (vs. Ag/AgCl) in O2-saturated 0.1 M KOH by using 0.1 M methanol.
For comparison, a commercial 20 wt % Pd/C and Pt/C (Alfa Aesar) were
measured under the identical experimental conditions.
Synthesis of Pd4xFex/C NPs. The synthesis procedures for Pd4xFex/C
NPs (x = 1, 2 and 3) were similar to that of Pd3Fe0.5Cu0.5/C NPs keeping
other conditions unaltered and only varying the initial mole ratio of Pd
and Fe salt precursors to 1:3, 1:1 and 3:1.
Synthesis of Pd/C NPs. The Pd/C NPs were also synthesized using the
method similar to that for Pd3Fe0.5Cu0.5/C. In this case, only PdCl2 was
employed as the metal precursors.
Materials characterization. The X-ray diffraction (XRD) patterns of the
samples were recorded on a Bruker AXS Model D8 focus instrument
using nickel-filtered CuKα (0.15418 nm) radiation source at the 2θ
between 10° and 80°, the scan rate was 0.05° s1
.
The Inductively
Acknowledgements
coupled plasma-optical emission spectrophotometry (ICP-OES) analysis
was accomplished by using Perkin Elmer, USA instrument (Model:
Optima 2100 DV) with the software Winlab-32. The loading of NPs on the
carbon support was determined by Thermogravimetric analysis (Model
TGA-50, Shimadzu). The flow rate of air atmosphere for the analysis was
maintained at 30 mL min1 and heating rate at 10 °C min1 up to 700 °C.
X-ray photoelectron spectroscopy (XPS) data were collected using a
Shimadzu ESCA-3400 equipped with a Mg-anode X-ray gun (10 kV, 20
mA). Samples were Ar-etched for 60 s before the measurements.
Accumulation times were 10-100 depending on signal intensity. The
binding energies of the samples were charge-corrected with respect to
the adventitious carbon (C 1s) peak at 284.6 e V. Transmission and high-
resolution transmission electron microscopy (TEM and HR-TEM)
The authors gratefully acknowledge the generous financial
support from Science and Engineering Research Board
(SERBDST No: SB/FT/CS-048/2014), New Delhi. SAIF, North-
Eastern Hill University Shillong is acknowledged for TEM
facilities.
Keywords: Nanoalloy, fuel cell, electrocatalysis, non-Pt
electrocatalyst, oxygen reduction reaction.
[1]
[2]
Y. Shao, G. Yin, Y. Gao, J. Power Sources 2007, 171, 558–566.
J. Deng, P. Ren, D. Deng, X. Bao, Angew. Chemie Int. Ed. 2015, 54,
2100–2104.
investigations were performed on
a FEI-Technai (G2 20 S-TWIN)
instrument equipped with slow-scan CCD camera and at an
a
[3]
[4]
N. Cheng, M. N. Banis, J. Liu, A. Riese, X. Li, R. Li, S. Ye, S. Knights, X.
Sun, Adv. Mater. 2015, 27, 277–281.
accelerating voltage of 200 kV. Energy-dispersive X-ray spectroscopy
(EDS) elemental mapping were performed on an Oxford instrument
attached to the “JEOL, JSM Model 6390 LV” scanning electron
microscope operated at 15 kV. The Brunauer-Emmett-Teller (BET)
C. Baldizzone, S. Mezzavilla, H. W. P. Carvalho, J. C. Meier, A. K.
Schuppert, M. Heggen, C. Galeano, J.-D. Grunwaldt, F. Schüth, K. J. J.
Mayrhofer, Angew. Chemie Int. Ed. 2014, 53, 14250–14254.
Y. Hu, J. O. Jensen, W. Zhang, L. N. Cleemann, W. Xing, N. J. Bjerrum,
Q. Li, Angew. Chemie Int. Ed. 2014, 53, 3675–3679.
surface areas were determined by N2 physisorption using
Quantachrome Instruments (Model: NOVA1000e).
a
[5]
[6]
[7]
D. Y. Chung, et al., J. Am. Chem. Soc. 2015, 137, 15478–15485. ("see
Supporting Information")
Electrocatalytic measurements. The catalytic activities of the freshly
prepared Pd4xFex/C NPs (x = 1, 2 and 3) and Pd3Fe0.5Cu0.5/C catalysts
for the electrocatalytic oxygen reduction were measured following a
typical procedure. In brief, glassy carbon (GC) electrode (3 mm diameter,
Rotating disk electrode, RDE) was polished with 0.005-μm Al2O3 powder.
A typical suspension of the carbon-supported catalysts was prepared by
adding 5 mg of fresh catalyst to 0.5 mL 2-propanol, deionized water and
0.5% Nafion solution, respectively. The resultant mixture was sonicated
for 30 min until it became a uniform dark ink. Then, 3 μL of the
suspension was quantitatively deposited onto the surface of the polished
GC electrode. The electrode coated with a thin film of the ink was dried at
35°C under N2 atmosphere for 6 h under vacuum. The modified GC
electrode was inserted into an electrochemical cell and used as the
working electrode. The cell includes two additional electrodes: a KCl
saturated Ag/AgCl (Metrohm) as the reference electrode and a platinum
wire (Metrohm) as the counter electrode. Cyclic voltammetry (CV) was
a) L. Guo, W.-J. Jiang, Y. Zhang, J.-S. Hu, Z.-D. Wei, L.-J. Wan, ACS
Catal. 2015, 5, 2903–2909; b) F. Liu, K. Sun, Z. Rui, J. Liu, T. Juan,
R. Liu, J. Luo, Z. Wang, Y. Yao, L. Huang, P. Wang, Z. Zou,
ChemCatChem, 2018,10, 3049 –3056; c) C. Goswami, K. K. Hazarika,
P. Bharali, Mater. Sci. Energy Technol. 2018, 1, 117–128; d) K. K.
Hazarika, C. Goswami, H. Saikia, B. J. Borah, P. Bharali, Mol. Catal.
2018, 451, 153–160; e) L. Zhang, X.-F. Zhang, X.-L. Chen, A.-J. Wang,
D.-M. Han, Z.-G. Wang, J.-J. Feng, J. Colloid Interface Sci. 2019, 536,
556–562.
[8]
[9]
a) C. Li, Q. Yuan, B. Ni, T. He, S. Zhang, Y. Long, L. Gu, X. Wang, Nat.
Commun. 2018, 9, 3702(1-9); b) A.-J. Wang, X.-F. Zhang, L.-Y. Jiang,
L. Zhang, and J.-J. Feng, ChemCatChem 2018, 10, 3319 – 3326.
W. Xiao, M. A. Liutheviciene Cordeiro, M. Gong, L. Han, J. Wang, C.
Bian, J. Zhu, H. L. Xin, D. Wang, J. Mater. Chem. A 2017, 5, 9867–
9872.
performed using
a
computer-controlled electrochemical analyzer
[10] a) J. Greeley, I. E. L. Stephens, A. S. Bondarenko, T. P. Johansson, H.
A. Hansen, T. F. Jaramillo, J. Rossmeisl, I. Chorkendorff, J. K. Nørskov,
Nat. Chem. 2009, 1, 552–556; b) H.-Y. Chen, M.-X. Jin, L. Zhang, A.-J.
(Metrohm Autolab PGSTAT204 workstation, The Netherlands) at room
temperature. The CV was measured in high purity N2 and O2 saturated
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
Wang, J. Yuan, Q.-L. Zhang, J.-J. Feng, J. Colloid Interface Sci. 2019,
543, 1–8.
[11] Y. Liu, S. Liu, Z. Che, S. Zhao, X. Sheng, M. Han, J. Bao, J. Mater.
Chem. A 2016, 4, 16690–16697.
[12] C. V. Boone, G. Maia, Electrochim. Acta 2017, 247, 19–29.
[13] F. Wang, C. Li, L.-D. Sun, C.-H. Xu, J. Wang, J. C. Yu, C.-H. Yan,
Angew. Chemie Int. Ed. 2012, 51, 4872–4876.
[14] a) A.-X. Yin, X.-Q. Min, W. Zhu, H.-S. Wu, Y.-W. Zhang, C.-H. Yan,
Chem. Commun. 2012, 48, 543–545; b) J. Du, C. Chen, F. Cheng J.
Chen, Inorg. Chem., 2015, 54, 5467–5474; c) Y. Lu, S. Zhao, R. Yang,
D. Xu, J. Yang, Y. Lin, N.-E. Shi, Z. Dai, J. Bao, M. Han, ACS Appl.
Mater. Interfaces, 2018, 10, 8155–8164.
[15] J. Chai, F. Li, Y. Hu, Q. Zhang, D. Han, L. Niu, J. Mater. Chem. 2011,
21, 17922-17929.
[16] T. Yamashita, P. Hayes, Appl. Surf. Sci. 2008, 254, 2441–2449.
[17] F. Li, J. Li, X. Lin, X. Li, Y. Fang, L. Jiao, X. An, Y. Fu, J. Jin, R. Li, J.
Power Sources 2015, 300, 301–308.
[18] a) A. Ghosh, P. Chandran, S. Ramaprabhu, Appl. Energy 2017, 208,
37–48; b) K. Parvez, S. Yang, Y. Hernandez, A. Winter, A. Turchanin, X.
Feng and K. Müllen, ACS Nano, 2012, 6, 9541–9550; c) F. Dehghani
Sanij, H. Gharibi, Colloids Surfaces A Physicochem. Eng. Asp., 2018,
538, 429–442.
[19] Q. Shi, C. Zhu, C. Bi, H. Xia, M. H. Engelhard, D. Du, Y. Lin, J. Mater.
Chem. A 2017, 5, 23952–23959.
[20] L. A. Kibler, A. M. El-Aziz, R. Hoyer, D. M. Kolb, Angew. Chemie Int. Ed.
2005, 44, 2080–2084.
[21] M. Shao, P. Liu, J. Zhang, R.R. Adzic, J. Phys. Chem. B 2007, 111,
6772–6775.
[22] C. Koenigsmann, A. C. Santulli, K. Gong, M. B. Vukmirovic, W. Zhou, E.
Sutter, S. S. Wong, R. R. Adzic, J. Am. Chem. Soc. 2011, 133, 9783–
9795.
[23] J. Greeley, I. E. L. Stephens, A. S. Bondarenko, T. P. Johansson, H. A.
Hansen, T. F. Jaramillo, J. Rossmeisl, I. Chorkendorff, J. K. Nørskov,
Nat. Chem. 2009, 1, 552–556.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900469
ChemCatChem
FULL PAPER
Entry for the Table of Contents
FULL PAPER
The rich interface between the
Biraj Jyoti Borah, Himadri Saikia,
embedded PdFeCu nanoparticles and
the carbon layer markedly affects the
electronic structure of PdFeCu
Chiranjita Goswami, Kumar Kashyap
Hazarika, Yusuke Yamada, and Pankaj
Bharali*
nanoalloy and contributed to the
improvement in the catalytic activity
and stability of PdFeCu/C. As a result,
the PdFeCu/C catalyst exhibited an
excellent stability and activity for
oxygen reduction reaction.
Page No. 1-6
Unique Half Embedded/Exposed
PdFeCu/C Interfacial Nanoalloy as
High-Performance Electrocatalyst for
Oxygen Reduction Reaction
This article is protected by copyright. All rights reserved.