Pd3Fe and Pt monolayer-modified Pd3Fe electrocatalysts for oxygen reduction

Article abstract of DOI:10.1524/zpch.2007.221.9-10.1175

In a search for electrocatalysts based on other metals with comparable attributes to very scarce and expensive Pt, we studied the kinetics of the oxygen reduction reaction (ORR) on carbon-supported Pd₃Fe alloy nanoparticles in HClO₄ and NaOH solutions. In acid solution, the electrocatalyst's activity for the ORR is slightly higher than that of commercial Pt/C. The reaction kinetics involves predominantly a four-electron reduction with the first charge-transfer step being the rate-determining one. The synthesis of a Pd overlayer on the Pd₃Fe alloy's surface at elevated temperatures is due to strong segregation effects. The major cause of this electrocatalyst's high activity appears to be the downshifting of the d-band center of Pd overlayer, resulting in a weaker interaction between the oxygen-containing species and the Pd surface. We demonstrated a further enhancement of the ORR kinetics with a Pt monolayer-covered Pd₃Fe/C electrocatalyst. In HClO₄ solution, the mass-specific activity of Pt/Pd₃Fe/C was about five times higher than that of commercial Pt/C, surpassing the Pt/Pd/C. In situ XANES data indicate that this elevated ORR activity may be due to the decreased formation of PtOH, and weaker oxygen adsorption on Pt/Pd₃Fe/C. by Oldenbourg Wissenschaftsverlag.

Full text of DOI:10.1524/zpch.2007.221.9-10.1175

Z. Phys. Chem. 221 (2007) 1175–1190 / DOI 10.1524/zpch.2007.221.9–10.1175
© by Oldenbourg Wissenschaftsverlag, München
Pd3Fe and Pt Monolayer-Modified Pd3Fe
Electrocatalysts for Oxygen Reduction
∗
By M. H. Shao, K. Sasaki, P. Liu, and R. R. Adzic
Chemistry Department, Brookhaven National Laboratory, Upton, NY 11973
Dedicated to Prof. Dr. Dieter M. Kolb on the occasion
of his 65th birthday
(Received June 4, 2007; accepted July 23, 2007)
Palladium Alloy / Platinum Monolayer / Oxygen Reduction Reaction /
Electrocatalyst / DFT
In a search for electrocatalysts based on other metals with comparable attributes to
very scarce and expensive Pt, we studied the kinetics of the oxygen reduction reaction
(ORR) on carbon-supported Pd3Fe alloy nanoparticles in HClO4 and NaOH solutions. In
acid solution, the electrocatalyst’s activity for the ORR is slightly higher than that of
commercial Pt C. The reaction kinetics involves predominantly a four-electron reduction
with the first charge-transfer step being the rate-determining one. The synthesis of
a Pd overlayer on the Pd3Fe alloy’s surface at elevated temperatures is due to strong
segregation effects. The major cause of this electrocatalyst’s high activity appears to be
the downshifting of the d-band center of Pd overlayer, resulting in a weaker interaction
between the oxygen-containing species and the Pd surface. We demonstrated a further
enhancement of the ORR kinetics with a Pt monolayer-covered Pd3Fe C electrocatalyst.
In HClO4 solution, the mass-specific activity of Pt Pd3Fe C was about five times higher
than that of commercial Pt C, surpassing the Pt Pd C. In situ XANES data indicate that
this elevated ORR activity may be due to the decreased formation of PtOH, and weaker
oxygen adsorption on Pt Pd3Fe C.
1. Introduction
Electrocatalytic oxygen reduction reaction (ORR) plays one of the key roles in
proton exchange membrane fuel cells (PEMFCs) [1–4]. Considerable on-going
research is focused on finding better ORR catalysts for fuel cells, which is con-
sidered to be a clean energy source in the near future. In all the ORR catalysts,
Pt and its alloys have the highest ORR activity [1, 3]. However, Pt is also very
expensive and of rare natural resources on earth. Some efforts have been made
*
Corresponding author. E-mail: adzic@bnl.gov
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1176
M. H. Shao et al.
to modify the Pt metal with other less expensive transition metals (Cr, Ti, Fe,
Co, Ni etc.), and observed some enhancement [5–10]. However, the loading
of Pt is still too high even being modified by other transition metals. Recently,
our group developed Pt monolayer electrocatalysts on suitable metal substrates
(Pd, Au) with high mass-specific ORR activity [11, 12]. The amount of Pt on
these catalysts is ultimately low since only one monolayer of Pt is used.
An electrocatalyst composed of relatively less expensive metals but with
comparable electrocatalytic properties to Pt catalysts would significantly
broaden the applications of fuel cells. Recent studies by our group and
others demonstrated that Pd-alloys may be promising candidates for cathode
materials [13–18], since they have comparable ORR activity to Pt but are
cheaper. Another distinct advantage is that Pd-alloys are less active in oxidizing
methanol [13, 14, 16, 18, 19]. This property is very helpful in designing high
methanol tolerance electrocatalysts for direct methanol fuel cells (DMFCs).
A similar ORR activity to that of Pt C was reported for carbon-supported
binary [13, 15, 18, 19] and ternary Pd-alloys [14, 16]. In this paper, we de-
scribe the results of our measurements of the kinetics of oxygen reduction on
Pd3Fe C using a rotating disk electrode method. Furthermore, we discuss the
very high ORR activity in HClO4 solution that we found for a Pt monolayer
deposited on Pd3Fe C.
2. Experimental
Pd-Fe alloy nanoparticles were prepared with a molar ratio of Pd : Fe = 3 : 1.
Carbon-supported Pd nanoparticles (Pd C, 19 wt.%) obtained from ETEK, and
an FeCl3 (Aldrich) solution were mixed and sonicated for 30 min, and then
dried at 80 ◦C in air in an oven. The dried mixture next was heated at 500 ◦C
in H2 for 2 5 h and allowed to cool down in an Ar atmosphere. For compari-
son, we correspondingly heat-treated one sample of Pd C lacking Fe to obtain
particles of a similar size.
Thin film electrodes with nano-electrocatalysts were prepared by placing
a certain amount of the nanoparticle suspension (1 0 mg ml−1) onto a flat glassy
carbon electrode (5-mm diameter, Pine Instrument). The noble metal (Pd or
Pt) loading of each sample was controlled at 10 µg cm−2, except where other-
wise stated. After drying in air, the electrode was covered with 5 µl of a dilute
Nafion solution (2 µg 5 µl).
The Pt monolayer was prepared by galvanic replacement of underpoten-
tially deposited (upd) Cu monolayer on a thin film Pd C or Pd3Fe C electrode
with Pt atoms. The detailed procedure is described elsewhere [20]. In brief,
after the upd of Cu from a 50 mM H2SO4 + 50 mM CuSO4 solution, the thin
film electrode covered with Cu monolayer was rinsed and kept in a 1 0 mM
K2PtCl4 + 50 mM H2SO4 solution for about 2 min to complete the replacement
of Cu with Pt.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1177
The Pt monolayer electrode then was characterized using a cyclic voltam-
metry before measuring ORR activity in rotating disk electrode (RDE, Pine
Instrument) experiments. All these operations were carried out in a multi-
compartment cell under a nitrogen atmosphere. For depositing the Pt mono-
layer, the Pd content of the substrates (Pd C and Pd3Fe C) was controlled at
10 nmol cm−2.
Electrolyte solutions were prepared from Optima sulfuric- or perchloric-
acid obtained from Fisher, and UV-plus water (Millipore). An Ag AgCl, Cl−
electrode and a Pt flag were used as the reference and counter electrode, respec-
tively. All potentials are given with respect to a reversible hydrogen electrode
(RHE). All the electrochemical measurements were taken at the room tempera-
ture.
Powder X-ray diffraction (XRD) patterns were obtained by a commercial
diffractometer (Phillips 3100) using Cu K radiation (1.54056 Å). We used the
diffraction line at (220) to calculate Pd’s inter-atomic distances and particle
sizes after Lorentzian fitting.
We characterized the size and distribution of particles using a JEOL-3000F
transmission electron microscopy (TEM), equipped with a Schottky field-
emission source operated at 300 keV. Carbon-supported nanoparticles were
dispersed on a Cu mesh grid coated with a lacey amorphous carbon-film.
The electrochemical cell employed for using X-ray absorption near edge
fine structure spectroscopy (XANES) data comprised two machined acrylic
plastic parts with windows for passing the X-ray beam. A Pt Pd3Fe C catalyst
film, a proton-exchange membrane (Nafion 117, DuPont Chemical Co.), and
a carbon-counter electrode (Grafoil, Union Carbide Corp.) were sandwiched
and held in PTFE gaskets, and all the cell components were clamped tightly to-
gether by the two acrylic plastic bodies with an O-ring. Electrical contact was
made through the Pt and Au leads that were pressed against the anode and cath-
ode electrodes, respectively. A reference electrode (Ag|AgCl, 3 M Cl− was
inserted in a compartment machined in one of the acrylic blocks. 1 M HClO4
was added to the cell as the electrolyte. A detailed description of the electro-
chemical cell is given elsewhere [21, 22]. XANES measurements were carried
out at the National Synchrotron Light Source (NSLS), Brookhaven National
Laboratory (BNL) using Beam Line X11B. The monochromator (Si(111) crys-
tals) was used to reject higher harmonics. The XANES data were processed
using the ATHENA data analysis software.
3. Results
3.1 Structure characterization of Pd3Fe C nanoparticles
Figure 1 displays the XRD patterns of Pd C (ETEK) and Pd3Fe C nanoparti-
cles exposed at 500 ◦C; both show a typical face-centered cubic (fcc) pattern.
However, all the diffraction lines have shifted to higher angles reflecting the in-
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1178
M. H. Shao et al.
Fig. 1. Comparison of XRD patterns of Pd3Fe C and Pd C nanoparticles. The dashed line
represents the standard (220) reflection line of Pd.
corporation of Fe atoms into the Pd lattice; there are no reflection lines from
Fe phase itself. The broad line at about 25◦ (2 ) represents the reflection from
carbon particles. The particles’ sizes and lattice constants calculated from the
reflection line (220) are 9 7 nm and 0 385 nm for Pd3Fe C, and 10 2 nm and
0 389 nm for Pd C, respectively. A superlattice, which is expected to exist in
the composition range of 65–85 at % Pd [23], was not observed in Fig. 1. Its
absence may indicate that the phase ordering in Pd-Fe system does not readily
occur [24].
Figure 2a shows the histogram of the distribution of particle sizes of
Pd3Fe C, derived from TEM images (inset). The particles are uniformly dis-
tributed, and their size is around 9 nm on average, a little smaller than the
value calculated from XRD data. Figure 2b is a high-resolution TEM image of
a single Pd3Fe particle. The atomic rows have a spacing of 0 216 nm corres-
ponding to the (111) interplanar distance of Pd3Fe. For an intrinsic Pd particle,
the measured Pd(111) interplanar distance is 0 226 nm (HRTEM image is not
shown), close to the theoretical value of 0 225 nm calculated from the Pd lat-
tice constant [25]. Even though our measurements of interplanar distance are
not precise, the results demonstrate that the lattice constant of Pd3Fe alloy is
smaller than that of intrinsic Pd, consistent with the XRD data.
3.2 Structure characterization of Pd3Fe C nanoparticles
The curve for UPD of Cu on Pd3Fe C nanoparticles, exhibiting several
pairs of current peaks shown in Fig. 3 (blue line), is similar to the one for
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1179
Fig. 2. Size distribution of Pd3Fe C nanoparticles derived from TEM images (inset) (a)
and a high resolution image of a Pd3Fe particle (b).
Pd C [12]. The integrated charge density corresponding to the Cu upd peaks
is 1 55 mC cm−2 that is about double the value obtained from the upd deposi-
tion of hydrogen on Pd3Fe (black line in Fig. 3). The voltammetry curve of Pt
monolayer on Pd3Fe C shows that the PtOH formation has shifted to a higher
potential compared with that of Pd3Fe C, and also, that the current peak due to
oxide reduction is much smaller.
3.3 Oxygen reduction reaction in 0.1 M HClO4 solution
Figure 4 shows the polarization curves recorded at different rotation rates on
a Pd3Fe C thin film electrode in a 0 1 M HClO4 solution. The Koutecky–
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1180
M. H. Shao et al.
Fig. 3. Voltammetry curves for the underpotential deposition of Cu on a Pd3Fe C thin film
electrode in 50 mM H2SO4 + 50 mM CuSO4 (blue line), and the voltammogramms of
the same electrode in 50 mM H2SO4 solution before (black line) and after (red line) Pt
monolayer deposition; sweep rate 50 mV s−1
.
Fig. 4. Polarization curves for the ORR on a Pd3Fe C thin film electrode in an oxygen-
saturated 0 1 M HClO4 solution at different rotation rates; sweep rate 10 mV s−1. Inset
shows the Koutecky–Levich plots at different potentials.
Levich (K–L) plots, derived from the data in Fig. 4, are presented as an inset.
The K–L plots are linear at all potentials. Both their linearity and the paral-
lelism indicate a first order kinetics with respect to molecular oxygen [26].
From the Koutecky–Levich equation:
1
1
1
1 2
=
+
B
(1)
j
jk
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1181
Fig. 5. Polarization curves for the ORR on a Pt monolayer deposited on a Pd3Fe C thin
film electrode in an oxygen-saturated 0 1 M HClO4 solution at different rotation rates;
sweep rate 10 mV s−1. Inset shows the Koutecky–Levich plots at different potentials.
where j is the measured current density, jk is the kinetic current density, B is
a constant, and is the rotation rate. The calculated value of B from pub-
lished data is 0 092 mA s−1 2 for a four-electron exchange reaction [19]. From
the slopes of the K–L plots, i.e., the constant B, we obtained the number of
electrons exchanged in the reduction of an oxygen molecule. Our experimental
value of B = 0 105 mA s−1 2, evaluated from Fig. 4, agrees with the calculated
value for a four-electron reduction of O2.
The data recorded at different rotation rates for a Pt Pd3Fe C thin film
electrode are depicted in Fig. 5, with the K–L plots shown as an inset. The
plots also are parallel lines similar to those of Pd3Fe C, indicating that the ki-
netic mechanisms for these two are the same. The constant B was evaluated as
0 098 mA s−1 2, again pointing to a four-electron reduction reaction.
Figure 6 compares the polarization curves for the ORR on different elec-
trocatalysts in 0 1 M HClO4 solution at 1600 rpm. The half-wave potentials in-
crease in the order Pt C (835 mV) Pd3Fe C (849 mV) Pt Pd C (860 mV)
Pt Pd3Fe C (875 mV). The ORR activity of the non-Pt catalyst, Pd3Fe C
is slightly better than that of the state-of-the-art electrocatalysts Pt C. Inter-
estingly, the half-wave potential of the Pt monolayer catalyst on Pd3Fe C is
15 mV higher than that of Pt Pd C, which we found in a previous study, to be
a better catalyst than Pt C [12]. Figure 7 affords a clearer comparison of activ-
ity by showing histograms of the total noble metal (Pt + Pd) mass activity of
different catalysts. At 0 8 V, the activities of Pd3Fe C and Pt Pd3Fe C are, re-
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1182
M. H. Shao et al.
Fig. 6. Comparison of polarization curves for the ORR in an oxygen-saturated 0 1 M
HClO4 solution for the following electrocatalysts: Pt C (10 µgPt cm−2), Pd3Fe C
(10 µgPd cm−2), Pt Pd C (10 nmolPd cm−2), and Pt Pd3Fe C (10 nmolPd cm−2). Sweep rate
10 mV s−1; room temperature. The inset shows the Tafel plots for Pt C, Pd3Fe C, and
Pt Pd3Fe C at 1600 rpm.
spectively, about 2 and 5 times larger than that of Pt C, while Pt Pd C has an
enhancement factor of 3.
The mass-transport corrected Tafel plots (kinetic currents, jk of Pt C,
Pd3Fe C and Pt Pd3Fe C are shown as the inset of Fig. 6. The jk was calcu-
lated based on the Eq. (1) and K–L plots. For Pd3Fe C, in the low overpotential
range of 0 94 V
the high overpotential range of 0 87 V
E
0 87 V, the estimated slope is −52 mV dec−1, while in
E
0 75 V, a slope of −80 mV dec−1
is observed. For Pt Pd3Fe C, the Tafel slopes are −61 and −93 mV dec−1 in
the low and high overpotential range, respectively. The Tafel slopes of Pd3Fe C
are lower than those of Pt Pd3Fe C in all the potential ranges. The lower Tafel
slopes than that of Pt were also observed for other Pd-transition metal bulk
alloys [17] as well as for nanoparticles [19] in acid solutions.
3.4 Oxygen reduction reaction in 0.1 M NaOH solution
Figure 8 gives the polarization curves of Pd C, Pd3Fe C and Pt Pd3Fe C
recorded at 1600 rpm in an oxygen-saturated 0 1 M NaOH solution. The
Pd3Fe C has higher ORR activity than the intrinsic Pd C, with a half-wave po-
tential shifted positively by 20 mV. The activity of Pt Pd3Fe C is similar to
that of Pd C. The Tafel plots of these three catalysts are presented as an inset in
Fig. 8. At low overpotentials, the Tafel slope of Pt Pd3Fe C (−70 mV dec−1) is
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1183
Fig. 7. Comparison of total noble metal (Pd + Pt) mass-specific activity for the ORR on
Pt C, Pd3Fe C, Pt Pd C, and Pt Pd3Fe C electrocatalysts in 0 1 M HClO4 solution at
0 8 V and 0 85 V.
Fig. 8. Comparison of polarization curves for the ORR for the following electrocatalysts in
an oxygen-saturated 0 1 M NaOH solution: Pd C (10 µgPt cm−2), Pd3Fe C (10 µgPt cm−2),
and Pt Pd3Fe C (10 nmolPd cm−2). Sweep rate 10 mV s−1; room temperature. Inset shows
the Tafel plots for Pd C, Pd3Fe C and Pt Pd3Fe C at 1600 rpm.
larger than those of Pd C and Pd3Fe C (both are −55 mV dec−1), while at high
overpotentials, all three catalysts show similar slopes, around −115 mV dec−1.
Figure 9 compares mass activities indicating the activity of Pd3Fe C about
twice those of Pd C and Pt Pd3Fe C.
Figure 10 displays the polarization curves recorded at different rotation
rates for a Pd3Fe C thin film electrode. The K–L plots are given as an inset.
Again, the near four-electron transfer reaction is observed. However, the K–L
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1184
M. H. Shao et al.
Fig. 9. Comparison of total noble metal (Pd + Pt) mass-specific activity for the ORR on
Pd C, Pd3Fe C, and Pt Pd3Fe C electrocatalysts in 0 1 M NaOH solution at 0 9 V and
0 95 V.
Fig. 10. Polarization curves for the ORR on a Pd3Fe C thin film electrode in an oxygen-
saturated 0 1 M NaOH solution at different rotation rates. Inset shows Koutecky–Levich
plots at different potentials at 1600 rpm.
plots are not quite parallel, which may be due to the influence of the carbon
supports on the ORR in the NaOH solution.
4. Discussion
4.1 ORR on Pd3Fe C
The Pd3Fe C electrocatalyst exhibits high ORR activity both in acid- and
alkaline-solutions. In a 0 1 M HClO4 solution, its noble metal mass-activity is
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1185
Fig. 11. Turnover frequencies (TOFs) for the ORR on Pt C, Pd3Fe C, Pt Pd C, and
Pt Pd3Fe C in a 0 1 M HClO4 solution at 1600 rpm; room temperature.
slightly higher than that of Pt C. However, if we take into account the size
of the particles (3 nm for Pt C versus 9 nm for Pd3Fe C from TEM meas-
urements), the difference of ORR activity between Pd3Fe C and Pt C will be
even larger since Pt C has much higher dispersion. The turnover frequency
(TOF), defined as the number of reduced oxygen molecules per time unit and
per electrochemically active atom, is an useful parameter for evaluating the
surface-specific activity of catalysts [5]. The Pd and Pt surface atomic densities
(normalized to the geometric area of the electrode) on Pd3Fe C and Pt C sur-
face calculated from the Hupd charge are 16 and 18 5 nmol cm−2, respectively.
The plot in Fig. 11 comparing the TOFs reveal that the enhancement factor
ranges from 1.2 (low overpotential region) to 2.5 (high over potential region)
per noble metal surface atom.
Bard et al. [15] and Balbuena et al. [27] proposed thermodynamic guide-
lines for designing non-Pt electrocatalysts for the ORR. For Pd-Co alloys,
Bard et al. [15] suggested that Co provides sites for breaking the O–O bonds
of oxygen. The O atoms then can migrate to Pd sites, to be reduced to water
at a relatively high potential. For Pd, with fully occupied valence d-orbitals,
Balbuena et al. [27] proposed that alloying with transition metals such as
Co, having unoccupied valence d-orbitals, reduces significantly the Gibbs
free energy both for the first charge-transfer step and those steps involv-
ing the reduction of intermediates (O OH). However, these thermodynamic
models did not take into account the strong surface segregation in the Pd-Co
system.
Recent theoretical calculations conducted by Norskov [28], Marvrikakis et
al. [29] and by Kolb’s group [30] suggested that the d-band center of the cata-
lysts might play a crucial role on catalytic activity. In Norskov’s model, when
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1186
M. H. Shao et al.
the d band of the metal surface shifts to a position of high energy, the reac-
tivity of the catalyst increases, and thus the adsorption of adsorbates becomes
stronger. A downshift of the d-band center has the reverse effect on reactiv-
ity and adsorption strength. Thus, the d band center of a good catalyst should
have the proper energy. Recent work [19] on a Pd monolayer on different single
crystals demonstrated that the d-band center of Pd monolayer on Pt (111) shifts
to a lower energy levels compared with the bulk Pd metal, yielding a better
electrocatalyst for oxygen reduction.
In the Pd3Fe C system, a Pd-rich overlayer is expected to form during the
high temperature treatment due to the large difference in surface segregation
energy between Pd and Fe [31]. The results of the DFT calculations reveal that
the d-band center of the Pd overlayer on Pd3Fe shifts to a lower position com-
pared with bulk Pd and even lower than that of Pd Pt(111) [32]. This change
weakens the bonds between the Pd overlayer and oxygen species that enhances
the ORR kinetics, even though the reactivity of Pd overlayer declines. Such
an interpretation is consistent with Xu et al’s DFT calculations on the Pt over-
layer on Pt3Fe alloy [33]. Our DFT results predict that the ORR activity of Pd
overlayer on Pd3Fe surpasses that of Pt (111) [32].
The argument for formation of the Pd overlayer on Pd3Fe is further
supported by the fact that the value for surface distribution of Pd atoms
(16 nmol cm−2) obtained electrochemically is much larger than the one calcu-
lated from the general dispersion equation [34, 35]:
D = 1 13 d
(2)
where D and d are the ratio of the number of surface atoms to the total
number of atoms, and the particle size, respectively. The Pd atom density
is about 9 5 nmol cm−2 considering that only about 75% of the total surface
atoms are Pd. Without Fe atoms on the surface, the Pd atom density would
be 12 5 nmol cm−2, i.e., close to the above value of 16 nmol cm−2. It would be
desirable to verify the formation of a Pd overlayer and the ORR kinetics on
well-ordered Pd3Fe single crystal surfaces.
4.2 ORR on Pt monolayer electrocatalysts
The ORR activities of Pt monolayer on Pd C and Pd3Fe C both surpass that of
Pt C in acidic solution. The enhanced ORR activity of Pt Pd C was reported
recently by our group. The increase in mass activity for Pt Pd3Fe C is larger
than that of Pt Pd C (5 times vs. 3 times) in comparison with Pt C. The TOFs
of these two catalysts are also shown in Fig. 11. The enhancement factor of
Pt Pd3Fe C with respect to Pt C varies from 8 to 10.
The possible origin of the observed high ORR activity may also be the
shift of the d-band center of the Pt monolayer. The Pt monolayer is actually
deposited on a Pd overlayer, not on the surface that reflects the composition
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1187
Fig. 12. In situ XANES spectra of the Pt L3 edge of Pt Pd3Fe C in 1 M HClO4 solution
at different potentials. The inset compares the change in white-line intensity as a function
of potential for pure Pt C and Pt monolayer electrocatalysts.
of the Pd3Fe alloy. DFT calculations were undertaken to clarify the changes
in the d-band center and the oxygen adsorption energy on a Pt monolayer
on Pd Pd3Fe. The findings show that oxygen adsorption energy is weaker
by 0 2 eV on the Pt monolayer (−1 83 eV) than on a intrinsic Pt surface
(−2 04 eV) due to the down-shift of the d-band center. It is becoming apparent
that oxygen adsorption on Pt is slightly stronger than the optimal value [29].
Thus, a decrease in oxygen bonding strength by about 0 2 eV, should give
a higher ORR activity than for bulk Pt [36]. We also note that the oxygen ad-
sorption energy of Pt monolayer on Pd Pd3Fe is weaker by 0 1 eV than that of
a Pt monolayer on Pd (−1 94 eV).
Zhang et al. [12] demonstrated that the positive shift of onset potential of
OH adsorption on Pt monolayer played a main role on enhancing ORR ac-
tivity, findings that are in agreement with the inhibitory effect of OH [37].
Stamenkovic et al. [8] reached a similar conclusion for the Pt “skin” on Pt
alloy cathodes. The OH group adsorbed on the Pt surface not only blocks the
active sites, but has a negative electronic effect on ORR kinetics [38]. We
observed the delayed formation of PtOH on Pt Pd3Fe C by in situ XANES
measurements. Figure 12 shows the Pt L3 edge spectra of Pt Pd3Fe C at dif-
ferent potentials. At potentials in the double layer region (0.42 and 0 6 V), the
white-line intensities of the Pt L3 edge show no significant change. As po-
tential increases to 1 0 V, where OH adsorbs on Pt surface, an increase in the
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1188
M. H. Shao et al.
white-line intensity is observed due to the depletion of Pt’s d-band. At 1 12 V,
the white line has continued to shift to a higher level intensity as more PtOH
and PtO accumulate on the surface. However, the rate of increase of the white
line intensity of the Pt monolayer as a function of potential is smaller than
that of Pt C (Fig. 12 inset), suggesting that the formation of PtOH on a Pt
monolayer requires higher potentials than is needed on Pt nanoparticles. Con-
sequently, the coverage of OH on a Pt monolayer is considerably smaller than
on Pt C.
5. Conclusions
Carbon-supported Pd nanoparticles modified by Fe atoms with a Pd : Fe ratio
of 3 : 1 show very high activities for the ORR, both in acid and alkaline solu-
tions. The activity in the HClO4 solution is higher than that of commercial Pt C
electrocatalysts. Our study of the ORR kinetics demonstrated that O2 reduction
involves predominantly a four-electron reaction with the first charge-transfer
being the rate-determining step, as is the case for a Pt surface. A Pd overlayer
forms on the Pd3Fe alloy’s surface due to strong segregation effects at ele-
vated temperatures. The downshift of the d-band center of the Pd overlayer,
which weakens the adsorption of oxygen-containing species on the Pd surface,
appears to be the main cause of the high ORR activity of Pd3Fe C. Further elu-
cidation of formation of Pd overlayer and its role in enhancing the activity for
the ORR requires studies on well-defined single crystals of the alloy.
We also studied a Pt monolayer on Pd3Fe C electrocatalysts in acid and al-
kaline solutions. In HClO4 the ORR activity of Pt Pd3Fe C is, respectively,
three and five times more than that of Pd3Fe C and Pt C. A 15 mV positive
shift of the half-wave potential of this catalyst was observed compared with
that of Pt Pd C. The in situ XANES measurements indicate that the observed
high ORR activity may be due to the delayed formation of PtOH on Pt mono-
layer, which acts as a catalytic poison. The oxygen adsorption energy of a Pt
monolayer on Pd3Fe C calculated by DFT is lower by 0 2 eV than that of an in-
trinsic Pt surface. Thus, the weaker binding between the reaction intermediates
and the Pt monolayer is reflected in faster ORR kinetics.
Acknowledgement
This work is supported by US Department of Energy, Divisions of Chem-
ical and Material Sciences, under the Contract No. DE-AC02-98CH10886.
Research was carried out in part at the Center for Functional Nanomaterials,
Brookhaven National Laboratory. M. H. Shao acknowledges partial support
from Department of Materials Science and Engineering, State University of
New York at Stony Brook. The authors also thank L. H. Zhang for assisting
with the TEM measurements.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

Pd Alloy and Pt Monolayer Electrocatalysts
1189
References
1. R. R. Adzic, In: Electrocatalysis. J. Lipkowski and P. N. Ross (Eds.). Wiley, New
York (1998), p. 197.
2. S. Gottesfeld and T. A. Zawodzinski, In: Advances in Electrochemical Science and
Engineering. Vol. 5, R. C. Alkire and D. M. Kolb (Eds.). Wiley, Weinheim (1997).
3. N. M. Markovic, T. J. Schmidt, V. Stamenkovic, and P. N. Ross, Fuel Cells 1 (2001)
105.
4. H. A. Gasteiger, S. S. Kocha, B. Sompalli, and F. T. Wagner, Appl. Catal. B 56
(2005) 9.
5. U. A. Paulus, A. Wokaun, G. G. Scherer, T. J. Schmidt, V. Stamenkovic, N. M. Mar-
kovic, and P. N. Ross, Electrochim. Acta 47 (2002) 3787.
6. A. Sode, W. Li, Y. G. Yang, P. C. Wong, E. Gyenge, K. A. R. Mitchell, and D. Biz-
zotto, J. Phys. Chem. B 110 (2006) 8715.
7. J. N. Soderberg, A. H. C. Sirk, S. A. Campbell, and V. I. Birss, J. Electrochem. Soc.
152 (2005) A2017.
8. V. Stamenkovic, T. J. Schmidt, P. N. Ross, and N. M. Markovic, J. Phys. Chem. B
106 (2002) 11970.
9. T. Toda, H. Igarashi, and M. Watanabe, J. Electroanal. Chem. 460 (1999) 258.
10. H. Yang, W. Vogel, C. Lamy, and N. Alonso-Vante, J. Phys. Chem. B 108 (2004)
11024.
11. K. Sasaki, Y. Mo, J. X. Wang, M. Balasubramanian, F. Uribe, J. McBreen, and
R. R. Adzic, Electrochim. Acta 48 (2003) 3841.
12. J. Zhang, Y. Mo, M. B. Vukmirovic, R. Klie, K. Sasaki, and R. R. Adzic, J. Phys.
Chem. B 108 (2004) 10955.
13. R. R. Adzic and J. X. Wang, Solid State Ionics 150 (2002) 105.
14. J. L. Fernandez, V. Raghuveer, A. Manthiram, and A. J. Bard, J. Am. Chem. Soc.
127 (2005) 13100.
15. J. L. Fernandez, D. A. Walsh, and A. J. Bard, J. Am. Chem. Soc. 127 (2005) 357.
16. V. Raghuveer, A. Manthiram, and A. J. Bard, J. Phys. Chem. B 109 (2005) 22909.
17. O. Savadogo, K. Lee, K. Oishi, S. Mitsushimas, N. Kamiya, and K.-I. Ota, Elec-
trochem. Commun. 6 (2004) 105.
18. M. H. Shao, K. Sasaki, and R. R. Adzic, J. Am. Chem. Soc. 128 (2006) 3526.
19. M. H. Shao, T. Huang, P. Liu, J. Zhang, K. Sasaki, M. B. Vukmirovic, and R. R.
Adzic, Langmuir 22 (2006) 10409.
20. S. R. Brankovic, J. X. Wang, and R. R. Adzic, Surf. Sci. 477 (2001) L173.
21. G. D. Adzic, J. R. Johnson, J. J. Reilly, J. McBreen, S. Mukerjee, M. P. S. Kumar,
W. Zhang, and S. Srinivasan, J. Electrochem. Soc. 142 (1995) 3429.
22. J. McBreen, W. E. Ogrady, K. I. Pandya, R. W. Hoffman, and D. E. Sayers, Lang-
muir 3 (1987) 428.
23. M. Hansen, Constitution of Binary Alloys. McGraw-Hill, New York (1958).
24. M. Chen and D. E. Nikles, J. Appl. Phys. 91 (2002) 8477.
25. JCPDS-ICDD PDF-2 Sets 1-45 database (Card 5-681).
26. N. A. Anastasijevic, V. B. Vesovic, and R. R. Adzic, J. Electroanal. Chem. 229
(1987) 305.
27. M. M. Waje, X. Wang, W. Z. Li, and Y. S. Yan, Nanotechnology 16 (2005) S395.
28. B. Hammer and J. K. Norskov, In: Advances in Catalysis. Vol. 45, B. Gates and
H. Knoezinger (Eds.), Academic Press, New York (2000), pp. 71–129.
29. J. L. Zhang, M. B. Vukmirovic, Y. Xu, M. Mavrikakis, and R. R. Adzic, Angew.
Chem.-Int. Edit. 44 (2005) 2132.
30. L. A. Kibler, A. M. El-Aziz, R. Hoyer, and D. M. Kolb, Angew. Chem. Int. Ed. 44
(2005) 2080.
31. A. V. Ruban, H. L. Skriver, and J. K. Norskov, Phys. Rev. B 59 (1999) 15990.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM

1190
M. H. Shao et al.
32. M. H. Shao, P. Liu, J. Zhang, and R. R. Adzic, J. Phys. Chem. B 111 (2007) 6772.
33. Y. Xu, A. V. Ruban, and M. Mavrikakis, J. Am. Chem. Soc. 126 (2004) 4717.
34. S. Y. Wang, S. H. Moon, and M. A. Vannice, J. Catal. 71 (1981) 167.
35. R. Pattabiraman, Appl. Catal. A 153 (1997) 9.
36. V. Stamenkovic, B. S. Mun, K. J. J. Mayrhofer, P. N. Ross, N. M. Markovic, J. Ross-
meisl, J. Greeley, and J. K. Norskov, Angew. Chem. Int. Ed. 45 (2006) 2897.
37. R. R. Adzic, In: Modern Aspects of Electrochemistry. Vol. 21 R. E. White,
J. O. M. Bockris, and B. E. Conway (Eds.), Plenum Press, New York (1990),
pp. 163–236.
38. J. X. Wang, N. M. Markovic, and R. R. Adzic, J. Phys. Chem. B 108 (2004) 4127.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/10/15 11:59 AM