View Article Online / Journal Homepage / Table of Contents for this issue

Dalton Dynamic Article Links

Transactions
Cite this: Dalton Trans., 2011, 40, 11480

www.rsc.org/dalton PAPER
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.

Direct evidence of redox interaction between metal ion and support oxide in
Ce0.98 Pd0.02 O2-d by a combined electrochemical and XPS study
Sudhanshu Sharma,* Bhaskar Devu Mukri and M. S. Hegde*
Received 1st July 2011, Accepted 19th August 2011
DOI: 10.1039/c1dt11262g

A combined electrochemical method and X-ray photo electron spectroscopy (XPS) has been utilized to
understand the Pd2+ /CeO2 interaction in Ce1-x Pdx O2-d (x = 0.02). A constant positive potential
(chronoamperometry) is applied to Ce0.98 Pd0.02 O2-d working electrode which causes Ce4+ to reduce to
Ce3+ to the extent of ~35%, while Pd remains in the +2 oxidation state. Electrochemically cycling this
electrode between 0.01.2 V reverts back to the original state of the catalyst. This reversibility is
attributed to the reversible reduction of Ce4+ to Ce3+ state. CeO2 electrode with no metal component
reduces to CeO2-y (y ~ 0.4) after applying 1.2 V which is not reversible and the original composition of
CeO2 cannot be brought back in any electrochemical condition. During the electro-catalytic oxygen
evolution reaction at a constant 1.2 V for 1000 s, Ce0.98 Pd0.02 O2-d reaches a steady state composition with
Pd in the +2 states and Ce4+ : Ce3+ in the ratio of 0.65 : 0.35. This composition can be denoted as
Ce4+ 0.63 Ce3+ 0.35 Pd0.02 O2-d-y (y~0.17). When pure CeO2 is put under similar electrochemical condition, it
never reaches the steady state composition and reduces almost to 85%. Thus, Ce0.98 Pd0.02 O2-d forms a
stable electrode for the electro-oxidation of H2 O to O2 unlike CeO2 due to the metal support interaction.

Introduction real nature of interaction between metal and support during a

The promoting effect of CeO2 or CeO2 ZrO2 mixed oxide and
metal dispersed over these oxides is a very well known phe-
nomenon and is generally attributed to metal support interaction.
A large amount of work has been devoted to understanding the
metal/ceria systems and metal support interaction owing to their
use as the active catalyst in various industries. It has been reported
that the capacity of the metal to adsorb H2 or CO is suppressed
upon reduction when catalysts containing Group 8 metals are
supported on reducible metal oxides (TiO2 , CeO2 Nb2 O5 etc.).
This has been ascribed to a strong metalsupport interaction
(SMSI).1,2 Recently, to understand the metal support interaction
in Pt (1 wt %)/Ce0.6 Zr0.4 O2 catalysts, partial reduction of Ce4+ to
Ce3+ in the reducing environment has been shown3 which is said
to be due to the metal support interaction, however, they did not
comment on the exact Pt oxidation states. Pd/CeO2 is another
system which has been studied widely412 to understand the exact
nature of the interaction of metal and support. A recent review
by Bernel can be looked into for the advancement in the eld of
metal support interaction.13 It is found that most of the studies
which exist in the literature so far try to prove the negative effect
of the metal support interaction without giving much insight into
the actual interaction. In fact, there is no report dealing with the
Solid State and Structural Chemistry Unit, Indian Institute of Sci-
ence, Bangalore, 560012. E-mail: sudhanshu.iisc@gmail.com, mshegde@
sscu.iisc.ernet.in; Fax: +91-80-2360 1310; Tel: +91-80-22932614
Electronic supplementary information (ESI) available: See DOI:
10.1039/c1dt11262g
reaction. Metalsupport interaction has been studied by a great
array of experiments namely ESR,14 XPS,1517 UV-Vis,18 FT-IR,19
magnetic balance,18,2024 however the exact interaction between the
noble metal and the support oxide and the promoting action of
CeO2 is still debated.2527
We have been working on the synthesis and reactivity of Pd
ion substituted CeO2 represented by Ce1-x Pdx O2-d (x = 0.01, 0.02,
d~x) and found that Pd2+ ion can be stabilized in CeO2 as a solid
solution with oxide ion vacancies.28 Taking a model reaction of CO
oxidation, a mechanism on this catalyst was proposed where as
the rst step, adsorption of CO on Pd2+ ionic sites and dissociative
chemisorption of O2 on the oxide ion vacancies was shown. In the
second step, electron transfer from CO molecule to O2 via Pd2+ to
Ce4+ and to the oxide ion vacancy was shown to be responsible for
higher catalytic activity. Activity was found higher by manifold
compared to Pd metal particles dispersed on Al2 O3 or CeO2 .29,30
The backbone of this mechanism involves interaction between
Pd2+ /Pd0 and Ce4+ /Ce3+ redox couples as follows:
Ce4+ 1-x Pd2+ x O2-x + x CO [Ce4+ 1-x Pd0 x O2-2x
(1)
Ce4+ 1-3x Ce3+ 2x Pd2+ x O2-2x ] + xCO2
[Ce4+ 1-x Pd0 x O2 - 2x Ce4+ 1 - 3x Ce3+ 2x Pd2+ x O2 - 2x ] + 1/2x O2
(2)
Ce4+ 1-x Pd2+ x O2-x
While electronic structure of Ce1-x Pdx O2-d can be determined
by XPS before and after catalytic reaction, electron exchange
between Pd and Ce ion as stated above is difcult to ascertain from
gas/solid reaction. Therefore, in an attempt to understand electron

11480 | Dalton Trans., 2011, 40, 1148011489 This journal is The Royal Society of Chemistry 2011

exchange between the redox ions in the catalyst we have now
devised a novel electrochemical method where Ce1-x Pdx O2-d (x~d)
is subjected to precise electrochemical oxidation or reduction
potential. The catalyst can be oxidized or reduced in static or
dynamic mode. At each stage, electronic states of Ce and Pd are
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.
examined ex situ by X-ray photo electron spectroscopy (XPS).
This method had been used by Kim et al. to investigate the oxide
lm over Pt and Pd.31,32 Our group used this probe to understand
the metal support interaction in Pt substituted CeO2 .33 Pd doped
CeO2 in the form of Ce1-x Pdx O2-d has been fully characterized in
our earlier studies34 hence in this paper our aim is to show that Pd2+
and Ce4+ have reversible electronic interaction in Ce1-x Pdx O2-d .
This phenomenon is attributed to the metal ion and support oxide
interaction in the doped metal oxide systems.
Experimental
The aqueous solution containing ceric ammonium nitrate, oxa-
lyldihydrazide (ODH) and PdCl2 is rapidly heated at 500 C to
form nano crystalline Ce1-x Pdx O2-d . In a typical preparation of
Ce0.98 Pd0.02 O2-d , ceric ammonium nitrate (0.06 M), ODH (0.14 M)
and PdCl2 (1.2 10-3 M) solution were combusted at 500 C to
give Ce0.98 Pd0.02 O2-d .
10(1-x)(NH4 )2 Ce(NO3 )6 + 24(1-x) (CONHNH2 )2 + 10PdCl2
10Ce1-x Pdx O2-d + 88(1-x)N2 + 48(1-x)CO2 + (102112)H2 O
+ 20HCl
CeO2 alone was prepared by the same solution combustion
method. In a typical preparation, 2.0 g of ceric ammonium nitrate,
1.03 g of ODH were dissolved in 20 ml of distilled water to make
a clear solution. Resulting solution was heated and combusted in
the same way as mentioned above to make CeO2-d .
10(NH4 )2 Ce(NO3 )6 + 24(CONHNH2 )2 10CeO2 + 88N2 +
48CO2 + 112H2 O
Structural properties of combustion synthesized CeO2 has
been fully characterized by X-ray diffraction (XRD) and X-
ray photoelectron spectroscopy (XPS) in a recent publication.35
Structural and morphological properties of Ce0.98 Pd0.02 O2-d have
been fully characterized by XRD, XPS, EXAFS, TEM where Pd
is present in +2 oxidation state and Pd ions are substituted for Ce4+
ions forming Ce0.98 Pd0.02 O2-d (d~0.02) retaining the parent uorite
structure.34,36
Morphological properties of CeO2 and Ce0.98 Pd0.02 O2-d were
performed using Transmission Electron Microscopy (FEI-Technai
G2 Sphera Microscope with a eld emission gun 200 kV, Point
resolution 0.24 nm; Line res. 0.14 nm).
Electrochemical studies were performed using a conventional
three electrode system. A working electrode was made by mixing
a known amount of CeO2 or Ce0.98 Pd0.02 O2-d with 30 wt% of
graphitic carbon and 5 wt% of PVDF (poly-vinyledene diuoride)
binder followed by depositing it on graphite paper over a known
area. Platinum foil was used as a counter electrode. Saturated
calomel electrode (SCE) was used as a reference electrode. H2 SO4
(0.5 M) was used as an electrolyte. Potentio static experiments
(chronoamperometry) were performed at a particular potential
for 1000 s. Potentio dynamic experiments (cyclic-voltametry (CV))
experiments were carried out at different scan rates within the
potential range of 0.0 to 1.2 V/1.6 V. Electronic states of Ce and
Pd ions of electrochemically oxidized, reduced and cycled samples
This journal is The Royal Society of Chemistry 2011
View Article Online
were recorded from an X-ray photoelectron spectrometer (Thermo
Fisher Scientic Multilab 2000). Sampling depth of XPS was 80
100 A. The electrode was taken out from the electrochemical cell,
dried and quickly mounted directly inside the XPS chamber. Since
the catalyst was mixed with graphite powder and deposited on
graphite paper there was no charging effect. Binding energies are
with respect to C(1 s) at 284.5 eV and they are accurate within
0.1 eV. XRD patterns of the electrodes were recorded before and
after applying the electrochemical potential.
To calculate the surface composition of Ce or Pd, before and
after the electrochemical measurements, the area under the main
peak was considered. The linear background was subtracted
before calculating the area. The relative sensitivity factor was
also considered for each element. Detail of the formula is given
elsewhere.28 For example, in the case of mixed valent Ce4+ and Ce3+
spectra, one can easily identify the main peaks (Ce3d7/2 ) related
to Ce4+ and Ce3+ . Having done that, the area under the main peak
for both Ce4+ and Ce3+ is calculated and the ratio is determined.
In this way one can approximately assess the composition of Ce4+
to Ce3+ in the compound. To further conrm the ratio, a second
method was also consulted. In this method, XPS spectra of Ce(3d)
core level region in pure Ce3+ (in Ce2 O3 ) and pure Ce4+ (in CeO2 )
were added in different proportions and the shape of the resultant
spectra is matched with the experimental data and the ratio is
veried. This method has been adopted in our previous work and
it is quite accurate.37 An example of this approach has been given in
Fig. S0 in the supporting information where XPS spectra of Ce(3d)
core level region in CeO2 and Ce2 O3 are added in 60 : 40 ratio
and the resulting spectra is shown. Catalysis/electrocatalysis is a
surface phenomenon hence XPS which is also a surface analysis
technique gives a good correlation.
Catalytic experiments were performed using a microreacter (6
mm quartz tube) equipped with a gas chromatograph (GC) and
temperature controller. Catalyst in the form of 40/60 mesh, is
lled in the quartz tube and both ends of the catalyst were
plugged using quartz wool. Temperature programmed reactions
(TPR) were performed using CO (0.5 vol%) and O2 (0.5 vol%)
purchased from Chemix Gases. N2 is further added to make
the total ow of 100 cc min-1 . This gives the space veloc-
ity (catalyst volume/volumetric ow rate) of around 43 000
h-1 . Output gases were analyzed using GC having Heysep-R
column.
Results and discussion
TEM of CeO2 and Ce0.98 Pd0.02 O2-d shows the cubic crystallites of
CeO2 in both of the compounds. TEM of CeO2 shows crystallite
size in the range 2030 nm (Fig. S1). The crystallite sizes of
Ce0.98 Pd0.02 O2-d are in the range of 3035 nm (Fig. S2). Some
of the crystallites also show lattice fringes. The big particles are
CeO2 particles (conrmed by EDAX analysis). Separated Pd metal
particles are not observed. It is expected as Ce0.98 Pd0.02 O2-d solid
solution should only show the impression of host CeO2 because
metal is substituted and no longer separated.
Detailed XRD results suggest that the lattice parameter for
CeO2 is 5.4113(2) A.34,35 Upon Pd ion substitution there is only a
small decrease in the lattice parameter, with a value of 5.4107(1)
A.34 XRD patterns also suggest that crystallite size of CeO2 and
Ce0.98 Pd0.02 O2-d calculated using the Debye-Sherrer equation are in
Dalton Trans., 2011, 40, 1148011489 | 11481

the range of 3040 nm (Fig. S3 and Fig. S4). Where the mean size
Kl
of the particle (t) is calculated as follows: t = , where K is
b cos q
shape factor, l is the wavelength of X-ray used, b is line broadening
(FWHM) in radians and q is the Bragg angle.
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.
This is in close agreement of TEM results. Detailed EXAFS
studies on CeO2 and Pd doped CeO2 have shown the change the
coordination environment around Pd which conrms that Pd is
doped for Ce in CeO2 host lattice.34
Detailed XPS study of Ce0.98 Pd0.02 O2-d has been carried out by
Priolkar et al. where it was shown that Pd is in +2 state and Ce
is in +4 state.34 Cl(2p) peaks are not detectable in the XPS of
Ce0.98 Pd0.02 O2-d . This means that solution combustion synthesis
gives the compounds without any halogen impurity from the
precursor which is known to affect the catalytic activity.38
Cyclic voltamogram (CV) of CeO2 in 0.5 M H2 SO4 in the
potential range of 0.01.6 V is shown in Fig. 1a. Scan rate is
100 mV s-1 . The direction of the voltamogram is shown by arrows.
Only the 1st and 20th cycles are plotted to show the prominent
changes occurring due to cycling. Prominent peaks at 1.3 V and
1.1 V in the forward and backward sweep are due to the Ce3+ /Ce4+
oxidation and reduction peaks. It is clear from the CV that from
the 1st to the 20th cycle there is an increase in the anodic current at
1.3 V as well as the cathodic current at 1.1 V. Increase in this redox
peak implies that a change in the Ce3+ /Ce4+ ratio is occurring.
Further, a new second peak at around 0.8 V in the backward scan
is observed in the 20th cycle. Notice the absence of this peak in the
Fig. 1 CV at a scan rate of 100 mV s-1 of (a) CeO2 ; (b) Ce0.98 Pd0.02 O2-d ; (c) Chronoamperometry of CeO2 and Ce0.98 Pd0.02 O2-d at 1.2 V in 0.5 M H2 SO4
solution.
11482 | Dalton Trans., 2011, 40, 1148011489
View Article Online
1st scan. This peak starts forming from the 2nd scan. This peak is
a characteristic feature of the presence of Ce3+ in the system. So
increase in this peak suggests the increase in the Ce3+ amount.
The origin of this peak is not clear but one can conrm this
behavior by running the CV of an aqueous solution of Ce(NO3 )3
in acidic medium using Pt both as working and the counter
electrode.
CV of Ce0.98 Pd0.02 O2-d in the same experimental conditions is
shown in Fig. 1b. Unlike the CeO2 electrode, at 1.3 V and 1.1 V
there is almost no difference in the 1st and the 20th cycle indicative
of the very less change in the Ce4+ /Ce3+ ratio. Further, there is no
indication of the formation of a new broad peak at around 0.8 V
like in CeO2 . Recall that this peak is a characteristic peak of the
presence of Ce3+ in the system. All of these observations indicate
that redox processes of Ce in Ce0.98 Pd0.02 O2-d are more reversible
than CeO2 and there is negligible reduction of Ce4+ to Ce3+ . In
simple words one can say that from the 1st to the 20th cycle over
all changes in the CV of Ce0.98 Pd0.02 O2-d are not as extensive as
in the case of CeO2 . Therefore, Ce0.98 Pd0.02 O2-d electrode appears
more stable than CeO2 during the multiple cycling.
The chronoamperometry experiment for oxygen evolution at
1.2 V over CeO2 working electrode is given in Fig. 1c. Continuous
decrease in the oxygen evolution current is apparent which falls
down almost to zero. The chronoamperometry experiment at 1.2 V
for Ce0.98 Pd0.02 O2-d is also shown in Fig. 1c. At 1.2 V, the oxygen
evolution current initially decreases and reaches a steady state
current value.
This journal is The Royal Society of Chemistry 2011
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23. View Article Online

Fig. 2 XPS of (a) CeO2 (as prepared); (b) CeO2 after 20 cycles between 0.01.6 V at a scan rate of 100 mV s-1 ; (c) CeO2 after putting an electrochemical
potential of 1.2 V for 1000 s; (d) after 10 cycles at a scan rate of 100 mV s-1 in the potential range of 0.01.2 V and (e) after applying an electrochemical
potential of -0.2 V for 1000 s.

Thus, Ce0.98 Pd0.02 O2-d is stable and CeO2 is unstable under
steady state electro-oxidizing condition as conrmed by CV and
chronoamperometry.
Chronoamperometric behavior prompts us to know the reason
behind the stability of Pd ion substituted CeO2 compared to
pure CeO2 . Electronic as well as crystal structures of CeO2 and
Ce0.98 Pd0.02 O2-d at different static and dynamic electrochemical
potential conditions are analyzed to gauge this phenomenon.
XPS of Ce(3d) shows the characteristic satellites in CeO2 and
Ce2 O3 and they have been discussed in detail by Kotani et al.39
Accordingly, a deconvoluted X-ray photoelectron spectrum (XPS)
of Ce(3d5/2, 3/2 ) core level region of as prepared CeO2 is shown in
Fig. 2a. Binding energies of Ce4+ (3d5/2 ) at 882.6 eV (A) along
with its satellites at 889.6 (A) and 898.7 eV (A) and Ce4+ (3d3/2 )
seen at 901.1 eV (B) along with its satellites at 908.1 (B) and
917.1 eV (B) which are characteristics of CeO2 . A small amount
of Ce3+ is also present as described in the spectrum by C, C and
D, D. Combustion synthesized CeO2 in known to have a small
percentage of Ce3+ and oxide ion vacancies due to this.40
XPS of Ce(3d5/2, 3/2 ) core level in CeO2 after 20 cycles of CV (from
Fig. 1a) is shown in Fig. 2b where a substantial reduction in Ce4+
to Ce3+ is noticed (notice the shape change in the spectra). These
two observations reveal that during cycling there is continuous
reduction in CeO2 electrode which is irreversible and it conrms

This journal is The Royal Society of Chemistry 2011 Dalton Trans., 2011, 40, 1148011489 | 11483
 View Article Online

the explanation of CV behavior in Fig. 1. Considering the area

under the main peak related to Ce4+ and Ce3+ and their relative
sensitive factor, Fig. 2b gives approximately 80% of Ce4+ reduction
to Ce3+ . This means CeO2 is electrochemically reduced to CeO2-y
(y ~ 0.4).
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.

Ce(3d5/2,3/2 ) core level spectrum after keeping CeO2 electrode at

an electrochemical potential of 1.2 V for 1000 s (chronoamperom-
etry) is given in Fig. 2c. The shape of the spectrum is completely
changed and the spectrum now looks similar to Ce2 O3 39 (also
compare it with the Ce2 O3 spectrum shown in Fig. S0). It is very
important to notice that applying a positive oxidizing potential
is causing reduction rather than the oxidation in the system. As
positive potential always induces the oxidation, reduction in CeO2
should be accompanying with a parallel oxidation reaction causing
reduction in the CeO2 . The accompanying parallel oxidation
process can only be the oxidation of the negatively charged oxygen Fig. 3 XRD of (a) CeO2 initially; (b) CeO2 after applying an electro-
from the CeO2 lattice; O2- O2 +2e- . An important conclusion chemical potential of 1.2 V for 1000 s; (c) Ce0.98 Pd0.02 O2-d initially; (b)
Ce0.98 Pd0.02 O2-d after putting an electrochemical potential of 1.2 V for
which can be drawn from this observation is that lattice oxygen can
1000 s.
take part during the electrochemical processes. This in turn tells
that oxygen evolution over CeO2 surface in acidic medium must
proceed via CeO2 reduction. Thus, oxygen evolution reaction over
CeO2 surface in acidic medium can now be written as follows:
2 CeO2 + H2 O Ce4+ 1-x Ce3+ x O2-x/2 + 2H+ + 2e + (6 + x)/4
(3)
O2 (x ~ 0.8)

The above equation is constructed with 80% reduction of Ce4+

to Ce3+ .
Further, CeO2 electrode is now cycled between 0.01.2 V under
potentio-dynamic condition (CV) at a scan rate of 100 mV s-1 (20
cycles). XPS of Ce(3d) core level spectra of the electrode after the
experiment is shown in Fig. 2d. A comparison of Ce(3d) spectra
in CeO2 (Fig. 2a) and after cycling (Fig. 2d) shows Ce4+ reduction
to Ce3+ . Ce4+ : Ce3+ ratio is 0.25 : 0.75. Thus the electrochemical
cycling cannot give back the original CeO2 .
CeO2 is now put at the reduction potential of -0.2 V. XPS of
Ce(3d) spectra after applying -0.2 V for 1000 s is shown in Fig. 2e.
One can see that ~90% of Ce4+ reduces to Ce3+ and the resulting
XPS after the reduction looks almost similar to Ce2 O3 39 (also see
Fig. S0 in the supporting information).
XRD patterns of CeO2 electrode before and after applying
1.2 V potential for 1000 s are shown in Fig. 3a, b. Both CeO2 (as
prepared) and CeO2-x (after chronoamperometry) are in uorite
structure. The lattice parameter increased from 5.411(2) A in
CeO2 to 5.440(2) A in CeO2-x . Increase in the lattice parameter
is expected due to the formation of Ce3+ ion and also the oxide ion
vacancies.
Thus, CV, XRD and XPS studies demonstrate that CeO2 con-
tinuously gets reduced even though uorite structure is retained.
The reduced CeO2 cannot get oxidized back at any electrochemical
condition.
XPS of Pd(3d) region in Pd metal foil, PdO, PdCl2 and
Ce0.98 Pd0.02 O2-d are given in Fig. 4. The binding energy of Pd(3d5/2 )
in the metal foil and as-prepared Ce0.98 Pd0.02 O2-d appears at 335.1
and 338.0 eV respectively. Which suggests Pd is in higher oxidation
state in Ce0.98 Pd0.02 O2-d . The binding energy of Pd(3d5/2 ) is 336.7 eV
in PdO, 338.0 eV in PdCl2 , 338.2 eV in Pd(NO3 )2 and 338.7 in
PdSO4 .41 The binding energy of Pd0 in Pd metal, Pd2+ in PdO,
PdCl2 , Pd(NO3 )2 and PdSO4 agrees well with the literature cited
value.41 Pd is in the +2 oxidation state in PdO, PdCl2 , Pd(NO3 )2 ,
Fig. 4 XPS of Pd(3d) core level region in Pd metal, PdO2 and PdCl2 and
Ce0.98 Pd0.02 O2-d .
and PdSO4 , but the binding energy of Pd(3d5/2 ) varies from 336.7
to 338.7 eV. Such a variation is due to the increase of effective
charge on the Pd2+ ion. Therefore, the oxidation state of palladium
is +2 state in Ce0.98 Pd0.02 O2-d . XPS of Pd(3d) and Ce(3d) of freshly
prepared Ce0.98 Pd0.02 O2-d are shown in Fig. 5(a, b). Pd (3d5/2 , 3d3/2 )
peaks at 338.1 and 343.4 eV represent Pd in +2 state (as explained
earlier). Ce is mostly in +4 state as shown in Fig. 5b. Each time
Ce0.99 Pd0.02 O2-d is synthesized, electronic states of Pd and Ce as
shown above are reproducible.
XPS of analysis of Ce0.98 Pd0.02 O2-d working electrode is carried
out after 20 cycles of CV (after the experiment shown in Fig. 1b).
Pd(3d) core level spectrum looks unchanged before (Fig. 5a) and
after the experiment (Fig. 6a) and Pd is in +2 oxidation state only.
Hence, Pd in Ce0.98 Pd0.02 O2-d is stable to redox cycling. Ce(3d) core
level spectra (Fig. 6b) after 20 cycles of CV (after the experiment
shown in Fig. 1b) shows partially reduced CeO2 . From the areas
under the peaks of Ce4+ and Ce3+ and the method of adding the
CeO2 and Ce2 O3 spectra (Fig. S0), approximately 35% of Ce4+
is reduced to Ce3+ . This reduction is much smaller compared to
CeO2 after similar cycling.

11484 | Dalton Trans., 2011, 40, 1148011489 This journal is The Royal Society of Chemistry 2011
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23. View Article Online

Fig. 5 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d (as prepared); (b) Ce(3d) core level region in Ce0.98 Pd0.02 O2-d (as prepared).

Fig. 6 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d after 20 cycles between 0.01.6 V at a scan rate of 100 mV s-1 in 0.5 M H2 SO4 , (b) Ce(3d)
core level region in Ce0.98 Pd0.02 O2-d after 20 cycles between 0.01.6 V at a scan rate of 100 mV s-1 in 0.5 M H2 SO4 .

Fig. 7 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d after applying an electrochemical potential of 1.2 V for 1000 s; (b) Ce(3d) core level region
in Ce0.98 Pd0.02 O2-d after putting an electrochemical potential of 1.2 V for 1000 s.

XPS of Pd(3d) region in Ce0.98 Pd0.02 O2-d working electrode kept
at a positive potential of 1.2 V for 1000 s is shown in Fig. 7a, b.
Pd(3d) does not show any change and remains in Pd2+ at this
oxidizing potential (Fig. 7a). Since there is no change in the
Pd binding energy on applying the oxidizing potential of 1.2 V,
PdO2 formation is unlikely. In other words, the highest attainable
oxidation state for Pd in CeO2 matrix is +2. Ce(3d) region on the
other hand shows Ce4+ reduction to Ce3+ after the experiment (Fig.
7b). From the areas of Ce4+ and Ce3+ peaks approximately 40%
of Ce4+ reduction is noticed (also compare Fig. S0). Importantly,

This journal is The Royal Society of Chemistry 2011 Dalton Trans., 2011, 40, 1148011489 | 11485

this reduction is much less compared to the reduction in CeO2
after the same chronoamperometry experiment (compare Fig. 7b
and Fig. 2c).
Thus, as explained earlier, at 1.2 V in Ce0.98 Pd0.02 O2-d , oxygen
evolution should result from lattice oxygen as well as due to water
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.
splitting. Therefore, the initial reaction at 1.2 V can be written as
follows:
Ce0.98 Pd0.02 O2-d + H2 O Ce0.98-y Pd0.02 O2-y/2-d + 2H+ + 2e +
(4)
y/2 O2 , (where y = 0.4)
Reaction (4) explains that O2 evolution reaction at the initial
stage has lattice oxygen contribution. Once the lattice oxy-
gen is partially removed giving a steady state composition as
Ce0.98-y Pd0.02 O2-y/2- d , O2 evolves due to water splitting only giving
a steady state current (see Fig. 1c). In this way this stable inter-
mediate redox couple Ce0.98-y Pd0.02 O2-y/2-d acts as the conventional
electrocatalyst and O2 evolution reaction on this electrode can be
written as:
2H2 O 4H+ + 4 e + O2 (5)
XRD patterns of Ce0.98 Pd0.02 O1.95 (or Ce0.98 Pd0.02 O2-d ) and
Ce0.98-y Pd0.02 O2-y/2-d , (or Ce0.98 Pd0.02 O1.77 ) are shown in Fig. 3c, d.
Ce0.98 Pd0.02 O2-d and Ce0.98 Pd0.02 O1.77 are both in uorite structure.
The lattice parameter of the electrochemically reduced sample
(Ce0.98-y Pd0.02 O2-y/2-d ) increases to 5.430(1) A compared to 5.407(2)
A of the unreduced compound, Ce0.98 Pd0.02 O2-d .
Further, a higher potential (1.4 V) is applied to Ce0.98 Pd0.02 O1.95
electrode for 1000 s to check the response in more vigorous
electrochemical conditions. After the experiment, Pd(3d) core level
spectra (Fig. 8a) shows a small percentage of Pd metal (335.0 eV)
due to the reduction of Pd2+ . Ce(3d) core level spectra (Fig. 8b)
in Ce0.98 Pd0.02 O2-d , shows Ce4+ reduction to the extent of 50%.
This reduction is higher than the reduction that occurred at
1.2 V (compare Fig. 8b and 7b). A slight reduction in Pd2+ at
a positive oxidizing potential is intriguing. It has already been
established that reduction in CeO2 in Ce0.98 Pd0.02 O1.95 is due to
the oxidation of negatively charged lattice oxygen (O2- O2 +2e- )
which eventually leaves the surface to cause the reduction. But
slight reduction in Pd2+ to Pd metal is a clear indication of the
Fig. 8 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d after applying an electrochemical potential of 1.4 V for 1000 s; (b) Ce(3d) core level region
in Ce0.98 Pd0.02 O2-d after applying an electrochemical potential of 1.4 V for 1000 s.
11486 | Dalton Trans., 2011, 40, 1148011489
View Article Online
electron or the charge distribution between metal ions (Pd2+ ) and
the oxide support (CeO2 ) in Ce0.98 Pd0.02 O1.95 . Apparently, there is
an electronic interaction between the metal ion and the support
oxide. But how exactly this electron distribution is taking place
is still unclear. Even at 1.4 V, Ce is not completely reduced while
at the same potential CeO2 showed complete reduction to Ce2 O3
(not shown).
It will be imperative to check the changes in the electronic states
of Ce0.99 Pd0.02 O2-d surface after the redox cycling in the potentio-
dynamic condition (CV) in the potential range of 0.01.2 V. XPS
of Ce0.98 Pd0.02 O2-d working electrode after the cycling at a scan
rate of 20 mV s-1 is shown in Fig. 9a, b. XPS of Pd(3d) after
20 cycles shows Pd in +2 oxidation state conrming that Pd in
Ce0.98 Pd0.02 O2-d is reversible (Fig. 9a) at this scan rate. However,
after CV, Ce4+ gets reduced to the extent of 40% after cycling (Fig.
9b). Compare it with Fig. S0 where the addition of 60% of Ce3d
spectra in CeO2 and 40% in Ce2 O3 gives similar spectra. This in
turns tells that Ce4+ at this scan rate in not completely reversible.
Ce0.98 Pd0.02 O2-d is now cycled at a higher scan rate of 40 mV s-1 in the
potential range of 0.01.2 as shown in Fig. 10a,b. Pd as expected
remains unchanged as Pd2+ (Fig. 10a). Cerium at this scan rate
shows almost 85% Ce4+ (Fig. 10b) suggesting that Ce4+ states are
reversible at this high scan rate. Note that as prepared compound
has 90% of cerium in +4 oxidation state. Thus, Ce0.98 Pd0.02 O2-d is
stable to redox cycle at 40 mV s-1 of scan rate, in the potential
range of 0.01.2 V and it can be reduced and oxidized. Increasing
the scan rate further to 80 mV s-1 does not make much difference
in the Pd2+ and Ce4+ : Ce3+ ratios (not shown).
It is also important to check the redox chemistry of
Ce0.98 Pd0.02 O2-d working electrode in the negative potential condi-
tions. XPS of the Ce0.98 Pd0.02 O2-d working electrode after applying
-0.2 V for 1000 s is given in Fig. 11a, b. The XPS spectrum now
shows peaks at 335 eV and Pd2+ at 338.1 eV related to Pd0 and
Pd2+ respectively. It is important to notice that even at this negative
potential, Pd has not reduced completely. Further, Ce(3d) core
level XPS shows that 50% of Ce4+ is reduced to Ce3+ . This clearly
demonstrates that due to the interaction between Pd and CeO2 ,
CeO2 gets reduced to the extent of ~50% (Fig. 11b) retaining Pd
in the oxidized state.
This journal is The Royal Society of Chemistry 2011
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23. View Article Online

Fig. 9 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d after 10 cycles (CV) at a scan rate of 20 mV s-1 ; (b) Ce(3d) core level region in Ce0.98 Pd0.02 O2-d
after 10 cycles (CV) at a scan rate of 20 mV s-1 .

Fig. 10 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d after 10 cycles (CV) at a scan rate of 40 mV s-1 ; (b) Ce(3d) core level region in Ce0.98 Pd0.02 O2-d
after 10 cycles (CV) at a scan rate of 40 mV s-1 .

Fig. 11 XPS of (a) Pd(3d) core level region in Ce0.98 Pd0.02 O2-d after applying an electrochemical potential of -0.2 V for 1000 s; (b) Ce(3d) core level region
in Ce0.98 Pd0.02 O2-d after applying an electrochemical potential of -0.2 V for 1000 s.

Catalytic activity due to Pd and CeO2 interaction and comparison
to CeO2 and Pt doped CeO2
To understand the role of metal support interaction it is imperative
to see if this electronic interaction is involved in enhancing the real
catalytic activity. Gas phase CO oxidation activity using O2 is
carried out for CeO2 , Pd substituted CeO2 (Ce0.98 Pd0.02 O2-d ) and
Pt substituted CeO2 (Ce0.98 Pt0.02 O2-d ) which is shown in Fig. 12.
Comparison of Ce0.98 Pt0.02 O2-d with Ce0.98 Pd0.02 O2-d is imperative
because in our earlier study it has been shown that electronic
interaction does exist between Pt and CeO2 also which affects its
redox properties. From Fig. 12 it is clear that the temperatures at

This journal is The Royal Society of Chemistry 2011 Dalton Trans., 2011, 40, 1148011489 | 11487

Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.
Fig. 12 Temperature programmed reaction (TPR) for CO oxidation using
a stoichiometric amount of oxygen (0.5 vol% of both CO and O2 ) for
Ce0.98 Pd0.02 O2-d , Ce0.98 Pt0.02 O2-d and CeO2 .
which CO is completely converted to CO2 are 350, 150 and 125
C for CeO2 , Ce0.98 Pd0.02 O2-d and Ce0.98 Pt0.02 O2-d respectively. Due
to the interaction between Pd2+ /Pt2+ and CeO2 , there is ~300 C
decrease in the temperature for CO oxidation activity (temperature
at which CO is fully converted to CO2 ) in comparison to CeO2
alone. To understand it more clearly, one can say that in the
case of CeO2 , once the lattice oxygen is removed by CO at some
temperature, it is hard to replenish it using gas phase O2 as shown
by electrochemical cycling (refer to Fig. 2c,d). However, in the
case of Ce0.98 Pd0.02 O2-d or Ce0.98 Pt0.02 O2-d , lattice oxygen is more
mobile and easily replenished by gas phase oxygen as shown by
electrochemical cycling in Fig. 10, 11. Thus the catalytic process
(CO + 1/2O2 CO2 ) continues via lattice oxygen utilization which
is reected in high activity at low temperature compared to CeO2 .
Comparing the Ce0.98 Pd0.02 O2-d and Ce0.98 Pt0.02 O2-d , activity looks
similar which is not surprising because both Ce0.98 Pd0.02 O2-d and
Ce0.98 Pt0.02 O2-d can be cycled in the electrochemical conditions.
Light off temperature, at which 50% conversion takes place is the
same (50 C) however, the temperature at which 100% conversion
takes place is 150 and 125 C for Ce0.98 Pd0.02 O2-d and Ce0.98 Pt0.02 O2-d
respectively showing the slight difference in the activity.
Surface concentration of Pd relative to Ce in the as prepared
compound is found to be 5% from XPS study. No noticeable
change in the surface concentration is observed after the elec-
trochemical experiments. Increase in the lattice parameter due to
the reduction of Ce4+ suggests the bulk reduction of oxide in the
electrodes. The concentration of metal ions is estimated within
56% error yet the changes in the ionic concentration ratios are
much higher than 5%.
In summary, the main observations are: (a) Ce0.98 Pd0.02 O2-d
shows partial (40%) reduction of Ce4+ unlike CeO2 showing
almost complete (90% reduction) when kept for 1000 s at an
electrochemical potential of 1.2 V. Moreover, Ce0.98 Pd0.02 O2-d
reaches a steady state in this condition; (b) Ce0.98 Pd0.02 O2-d can be
electrochemically cycled between 0.01.2 V at and above 40 mV s-1
unlike CeO2 ; (c) At the negative potential (-0.2 V), Ce0.98 Pd0.02 O2-d
shows only partial reduction in Ce4+ and Pd unlike CeO2 which
is completely reduced to Ce2 O3 ; (d) compound retains its primary
11488 | Dalton Trans., 2011, 40, 1148011489
View Article Online
uorite structure in all the electrochemical conditions as conrmed
by the XRD data.
In conclusion, the redox catalytic property of Ce0.98 Pd0.02 O2-d
as given in eqn (1) and eqn (2) is associated with the interaction
between Pd2+ and Ce4+ /Ce3+ redox couple within the oxide catalyst.
This is a general method and redox properties of other noble
metal ions substituted CeO2 , TiO2 and other such supports can be
studied to evaluate metal support interaction.
Acknowledgements
The authors thank for nancial support from the Departments
of Science and Technology, Govt. of India. Mr. Preetam Singh is
thankfully acknowledged for helping with XRD experiments.
References
1 S. J. Tauster, S. C. Fung, R. T. K. Baker and J. A. Horsley, Science,
1981, 211, 1121.
2 S. J. Tauster, S. C. Fung and R. L. Garten, J. Am. Chem. Soc., 1978,
100, 170.
3 G. Deganello, F. Giannici, A. Martorana, G. Pantaleo, A. Prestianni,
A. Balerna, L. F. Liotta and A. Longo, J. Phys. Chem. B, 2006, 110,
8731.
4 A. Badri, C. Binet and J. C. Lavalley, J. Chem. Soc., Faraday Trans.,
1996, 92, 16031608.
5 A. Bensalem, F. Bozon-Verduraz and V. Perrichon, J. Chem. Soc.,
Faraday Trans., 1995, 91, 21852189.
6 A. Bensalem, J. C. Muller, D. Tessier and F. Bozon-Verduraz, J. Chem.
Soc., Faraday Trans., 1996, 92, 32333237.
7 C. Binet, A. Jadi, J. Lavalley and M. Boutonnet-Kizling, J. Chem. Soc.,
Faraday Trans., 1992, 88, 20792084.
8 D. Ciuparu, A. Bensalem and L. Pfefferle, Appl. Catal., B, 2000, 26,
241255.
9 L. Kepinski and M. Wolcyrz, Appl. Catal., A, 1997, 150, 197220.
10 W.-J. Shen and Y. Matsumura, J. Mol. Catal. A: Chem., 2000, 153,
165168.
11 L. Yang, O. Kresnawahjuesa and R. J. Gorte, Catal. Lett., 2001, 72,
3337.
12 Z. Yang, Z. Lu, G. Luo and K. Hermansson, Phys. Lett. A, 2007, 369,
132139.
13 S. Bernal, J. J. Calvino, M. A. Cauqui, J. M. Gatica, C. Larese, J. A.
Perez Omil and J. M. Pintado, Catal. Today, 1999, 50, 175206.
14 J. Cunningham, S. OBrian, J. Sanz, J. M. Rojo, J. A. Soria and J. G.
Fiero, J. Mol. Catal., 1990, 57, 379.
15 D. I. Kondarides and X. E. Verykios, J. Catal., 1998, 174, 52.
16 J. Stubenrauch and J. M. Vohs, Catal. Lett., 1997, 47, 21.
17 Y. Zhou, M. Nakashima and J. M. White, J. Phys. Chem., 1988, 92,
812.
18 A. Laachir, V. Perrichon, A. Badri, J. Lamotte, E. Catherine, J. C.
Lavalley, J. E. Fallah, L. Hilaire and F. L. Normand, J. Chem. Soc.,
Faraday Trans., 1991, 87, 1601.
19 C. Binet, A. Jadi, J. C. Lavalley and M. Boutonnet-Kizling, J. Chem.
Soc., Faraday Trans., 1992, 88, 2079.
20 A. Bensalem, F. Bozon-Verduraz and V. Perrichon, J. Chem. Soc.,
Faraday Trans., 1995, 91, 2185.
21 S. Bernal, G. Blanco, J. M. Gatica, C. Larese and H. Vidal, J. Catal.,
2001, 200, 411.
22 S. Bernal, J. J. Calvino, G. A. Cifredo and J. M. Rodriguez-Izquierdo,
J. Phys. Chem., 1995, 99, 11794.
23 V. Perrichon, A. Laachir, G. Bergeret, R. Frety, L. Tournayan and O.
Touret, J. Chem. Soc., Faraday Trans., 1994, 90, 773.
24 S. Bernal, J. J. Calvino, G. A. Cifredo, A. Laachir, V. Perrichon and J.
M. Herrmann, Langmuir, 1994, 10, 717.
25 W. Lin, A. A. Herzing, C. J. Kiely and I. E. Wachs, J. Phys. Chem. C,
2008, 112, 5942.
26 C. M. Y. Yeung, K. M. K. Yu, Q. J. Fu, D. Thompsett, M. I. Petch and
S. C. Tsang, J. Am. Chem. Soc., 2005, 127, 1801018011.
27 A. Y. Stakheev and L. M. Kustov, Appl. Catal., A, 1999, 188, 335.
This journal is The Royal Society of Chemistry 2011

28 P. Bera, K. C. Patil, V. Jayaram, G. N. Subbanna and M. S. Hegde, J.
Catalysis, 2000, 196.
29 P. Bera, K. R. Priolkar, A. Gayen, P. R. Sarode, M. S. Hegde, S. Emura,
R. Kumashiro, V. Jayaram and G. N. Subbanna, Chem. Mater., 2003,
15, 2049.
30 P. Bera, A. Gayen, M. S. Hegde, N. P. Lalla, L. Spadaro, F. Frusteri
Published on 26 September 2011. Downloaded by Indian Institute of Technology Hyderabad on 24/12/2014 17:50:23.
and F. Arena, J. Phys. Chem. B, 2003, 107, 6122.
31 K. S. Kim, N. Winograd and R. E. Davis, in J. Am. Chem. Soc., 1971,
p. 23.
32 K. S. Kim, A. F. Gossmann and N. Winograd, Anal. Chem., 1974, 46,
197.
33 S. Sharma and M. S. Hegde, J. Chem. Phys., 2009, 130, 114706114708.
34 K. R. Priolkar, P. Bera, P. R. Sarode, M. S. Hegde, S. Emura, R.
Kumashiro and N. P. Lalla, Chem. Mater., 2002, 14, 21202128.
This journal is The Royal Society of Chemistry 2011
View Article Online
35 A. Gupta, M. S. Hegde, K. R. Priolkar, U. V. Waghmare, P. R. Sarode
and S. Emura, Chem. Mater., 2009, 21, 58365847.
36 T. Baidya, A. Marimuthu, M. S. Hegde, N. Ravishankar and G. Madras,
J. Phys. Chem. C, 2007, 111, 830839.
37 S. Sharma, P. Singh and M. Hegde, J. Solid, State Electrochem., in
press.
38 D. O. Simone, T. Kennelly, N. N. Brungard and R. J. Farrauto, Appl.
Catal., 1991, 70, 87100.
39 A. Kotani and H. Ogasawarab, J. Electron Spectrosc. Relat. Phenom.,
1992, 60, 257.
40 B. Murugan and A. V. Ramaswamy, J. Am. Chem. Soc., 2007, 129,
30623063.
41 D. Briggs and M. P. Seah, Practical Surface Analysis, John Wiley &
Sons Ltd, 2004.
Dalton Trans., 2011, 40, 1148011489 | 11489