In situ surface chemistries and catalytic performances of ceria doped with palladium, platinum, and rhodium in methane partial oxidation for the production of syngas

Article abstract of DOI:10.1021/cs400070y

Methane partial oxidation (MPO) chemically transforms natural gas into syngas for the production of gasoline. CeO₂ doped with transition-metal ions is one type of catalyst active for MPO. A fundamental understanding of MPO on this type of catal

Full text of DOI:10.1021/cs400070y

Research Article
pubs.acs.org/acscatalysis
In Situ Surface Chemistries and Catalytic Performances of Ceria
Doped with Palladium, Platinum, and Rhodium in Methane Partial
Oxidation for the Production of Syngas
Yuan Zhu,† Shiran Zhang,† Jun-jun Shan,† Luan Nguyen, Sihui Zhan, Xiaoli Gu,
and Franklin (Feng) Tao*
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana 46556, United States
S
* Supporting Information
ABSTRACT: Methane partial oxidation (MPO) chemically trans-
forms natural gas into syngas for the production of gasoline. CeO2
doped with transition-metal ions is one type of catalyst active for
MPO. A fundamental understanding of MPO on this type of
catalyst is important for the development of catalysts with high
activity and selectivity for this process. CeO2-based catalysts,
including Pd-CeO2-air, Pd-CeO2-H2, Pt-CeO2-air, Pt-CeO2-H2, Rh-
CeO2-air, and Rh-CeO2-H2, were synthesized through doping
noble-metal ions in the synthesis of CeO2 nanoparticles. The
catalytic activity and selectivity in the production of H2 and CO
through MPO on these ceria-based catalysts as well as their surface chemistries during catalysis were investigated. They exhibit
quite different catalytic performances in MPO under identical catalytic conditions. In situ studies of their surface chemistries
during catalysis, using ambient-pressure X-ray photoelectron spectroscopy (AP−XPS), revealed quite different surface
chemistries during catalysis. Correlations between the catalytic performances of these catalysts and their corresponding surface
chemistries during catalysis were developed. Differing from the other four catalysts, Rh doped in the surface lattice of a CeO2
catalyst, including Rh-CeO2-air and Rh-CeO2-H2, is in a complete ionic state during catalysis. Correlations between the in situ
surface chemistry and the corresponding catalytic performance show that Rh ions and Pt ions doped in the lattice of CeO2 as well
as metallic Pd nanoparticles supported on CeO2 are active components for MPO. Among these catalysts, Rh-doped CeO2
exhibited the highest catalytic activity and selectivity in MPO.
KEYWORDS: in situ, methane partial oxidation, ceria, ambient-pressure XPS, synthesis
decreases the activation energy of C−H dissociation on oxygen
atoms of the surface lattice.
1. INTRODUCTION
In contrast to crude oil, natural gas is relatively abundant even if
the methane generated from biomass and the methane hydrate
from seashells are not taken into account. One utilization of
natural gas is the production of gasoline from synthesis gas
(syngas) through methane partial oxidation (MPO).1−4 MPO
is an economical strategy to generate syngas from natural
gas.4−33 The production of syngas through MPO using
different catalysts has been studied in the past decades.34
Several groups35−39,42,47 have studied methane activation over
CeO2 either doped or impregnated with precious metals. For
example, Pt-doped CeO2 catalysts were prepared with several
methods including coprecipitation,40 impregnation,41−46 depo-
sition−precipitation,47 sol−gel,47 and flame combustion.40,42,47
It was proposed that doping a reducible oxide with noble-metal
atoms could tune the catalytic activity of the oxide by shifting
the Fermi level through integrating doped ions into the lattice
of the oxide surface.48 In addition, this doping could change the
formation energy of oxygen vacancies and thus influence the
reactivity of the oxygen atoms of the surface lattice of a
reducible oxide toward the activation of C−H bonds.47 The
doping of Pt ions likely activates neighboring oxygen atoms and
Compared to Pt or Pd, Rh exhibits different binding energies
to oxygen atoms.49,50 A stronger binding of Rh ions to surface-
lattice oxygen atoms upon the doping of Rh ions could make
Rh-doped CeO2 exhibit a catalytic performance different from
that of CeO2 doped with Pd or Pt ions. Rh ions could activate
oxygen atoms of the surface lattice and thus promote C−H
dissociation significantly. Compared to Rh ions, Pd ions could
be reduced to metallic Pd readily. Thus, it is expected that Pd-
CeO2 exhibits a different surface chemistry and catalytic
performance for MPO. To understand how the different
noble-metal ions/atoms influence the catalytic performance of
MPO, it is necessary to perform parallel studies of CeO2
catalysts doped with different cations, including Pd, Pt, and Rh.
Although the catalytic performances of MPO on Pt-doped
CeO2 have been studied47 and the surface chemistries of these
catalysts have been characterized through ex situ approaches,
Received: January 28, 2013
Revised: September 11, 2013
Published: September 18, 2013
© 2013 American Chemical Society
2627
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
there is a lack of a direct correlation between the in situ surface
chemistry of catalysts during catalysis and their corresponding
catalytic performance. Here, we used an ambient-pressure X-ray
photoelectron spectrometer (AP−XPS) available in our group
to track the surface chemistries of CeO2 doped with Pd, Pt, and
Rh ions during MPO at different temperatures because AP−
XPS is an in situ technique with a high surface sensitivity.51−54
Pd-CeO2, Pt-CeO2, and Rh-CeO2 were synthesized with the
same method. A pretreatment in air or H2 resulted in six
catalysts (Pd-CeO2-air, Pd-CeO2-H2, Pt-CeO2-air, Pt-CeO2-H2,
Rh-CeO2-air, and Rh-CeO2-H2) that were evaluated for their
catalytic performance and their in situ surface chemistry during
catalysis (using AP−XPS).
Here, the catalytic performances of the six catalysts were
examined under the same catalytic conditions. Surface
chemistries of these catalysts during catalysis were tracked
with AP−XPS. Correlations between surface chemistry and the
corresponding catalytic performance were developed. These
correlations provide insight into the differences in catalytic
performance. Insights achieved from these correlations are
important for developing catalysts for MPO that have high
activity and selectivity.
Micrograph (Gatan) software. TEM samples were prepared by
grinding the catalyst with an agate mortar, dispersing 10 mg of
the sample in 1 mL of anhydrous ethanol, sonicating it for 10
min, and finally dropping the solution to a lacey-carbon-film-
coated 200 mesh copper grid (Electron Microscopy Sciences).
2.3. Measurements of Catalytic Performance. Measure-
ments of the catalytic activity and selectivity of these catalysts
for MPO were carried out at atmospheric pressure in a fixed-
bed tubular quartz microreactor (i.d. = 6 mm, length = 300
mm). One hundred milligrams of a sieved catalyst sample (40−
60 mesh) diluted with quartz sand (the bed height was ca. 5
mm) was packed in between two quartz wool plugs. The
catalyst was heated in a furnace equipped with a proportional-
integral-derivative (PID) temperature controller (Across
International). Pretreatment of the as-synthesized catalyst M-
CeO2 in air was performed at 500 °C for 3 h in a muffle furnace
(Across International). Before catalytic experiments using the
M-CeO2-air catalyst, it was flushed with pure Ar (99.99%,
Praxair, Inc.) for 30 min and then exposed to the reactant gases
flowing at a rate of 100 mL min−1 (controlled by a Dakota
Instruments, Inc. mass flow meter/controller). The gas
composition of the mixture of reactant gases was 5% CH4,
2.5% O2, and 92.5% Ar (Praxair, Inc.). For M-CeO2-H2, the
catalyst precursor was pretreated in 5% H2 at 500 °C (50 mL/
min) for 3 h and then cooled to near room temperature
naturally in 5% H2. Upon switching to the mixture of reactant
gases, the catalytic performance of M-CeO2-H2 for MPO was
measured. For comparison, the catalytic performance of MPO
on M-CeO2-H2 was measured under the same conditions as
those used for M-CeO2-air.
2. EXPERIMENTAL SECTION
2.1. Preparation of CeO2-Based Catalysts Doped with
5% Pd, Pt, or Rh. A solution-based hydrothermal procedure55
was used to prepare the Pd-, Pt-, and Rh-doped CeO2 catalysts.
Ce(NO3)3·6H2O (1.7369 g, Sigma-Aldrich, 99% trace metals
basis) and a metal precursor, 0.0561 g of Pd(NO3)2·2H2O,
0.1090 g of H2PtCl6·6H2O, or 0.0541 g of RhCl3 (Sigma-
Aldrich), were dissolved in 5 mL of deionized water
simultaneously and added to 60 mL of a sodium hydroxide
solution in a Teflon bottle. The mixture was filled to 80 mL,
giving a final NaOH concentration of 6.0 M. This solution was
stirred for 15 min to form a milky slurry. The slurry was
subsequently transferred to a 100 mL Teflon-lined stainless
steel autoclave and heated in an oven at 100 °C for 24 h. The
autoclave was then cooled to room temperature. The product
was collected in a beaker by self-sedimentation and was washed
with deionized water. After decanting the supernatant until the
pH value reached ∼7, the sample was dried at 100 °C.
Calcination in air or H2 resulted in the six different catalysts.
Calcination at 500 °C in air offered catalysts Pd-CeO2-air, Pt-
CeO2-air, and Rh-CeO2-air. Calcination at 500 °C in 5% H2/Ar
provided catalysts Pd-CeO2-H2, Pt-CeO2-H2, and Rh-CeO2-H2.
The concentration of Pd in Pd-CeO2-air and Pd-CeO2-H2, of
Pt in Pt-CeO2-air and Pd-CeO2-H2, and of Rh in Rh-CeO2-air
and Pd-CeO2-H2 was measured with inductively coupled
plasma atomic emission spectroscopy (ICP-AES, PerkinElmer
Optima 3000XL) after the catalysis measurements. On the basis
of the ICP-AES measurements, the doping levels of these
reacted catalysts were 5.0 0.3%, very close to the atomic ratio
of Pd, Pt, or Rh to CeO2 of 5.0% used in syntheses.
A cold trap downstream of the reactor outlet was used to
condense the water formed during the reaction. The effluent
gas was connected to a gas chromatograph (8610C, SRI
1
Instruments) equipped with a HayeSep D (6′ × /8″) packed
column, a molecular sieve 13× (6′ × 1/8″) packed column, and
a thermal conductivity detector (TCD) for the analysis of both
the reactants and products. In this work, the calculations for the
conversion of methane (XCH ) and the selectivity of hydrogen
4
(SH ) and carbon monoxide (SCO) were calculated using the
2
following equations
XCH (%) = [(CCH ,inlet − CCH ,outlet)/CCH ,inlet] × 100
4
4
4
4
SH (%) = [moles of H2 produced
2
/(2 × moles of CH4 consumed)] × 100
SCO (%) = [moles of CO produced
/moles of CH4 consumed] × 100
where CCH ,inlet and CCH ,outlet are defined as the molar fractions
4
4
of CH4 at the reactor inlet and outlet, respectively.
2.2. Powder X-ray Diffraction and Transmission
Electron Microscopy. Powder X-ray diffraction (XRD)
patterns of the six catalysts were acquired on a Bruker D8
Advance XRD using nickel-filtered Cu Kα radiation (λ = 1.5418
Å) operating at 40 kV and 40 mA in a continuous mode with a
scanning rate of 4.2° min−1 in the 2θ range from 20° to 90°.
The size, shape, and lattice fringe of the CeO2-doped catalysts
were identified using high-resolution transmission electron
microscope (FEI, Titan 80-300) operated at an accelerating
voltage of 300 kV. Image analyses were performed with Digital
2.4. In Situ Studies using Ambient-Pressure X-ray
Photoelectron Spectroscopy. The surface chemistry of the
six catalysts was studied using an in-house ambient-pressure X-
ray photoelectron spectrometer (AP−XPS) with an Al Kα
monochromator (Microfocus 600, Specs) and energy analyzer
(Phoibos 150, Specs) in the Tao group.56 In terms of the
resolution of this spectrometer, the measured full-width at half-
maximum (fwhm) of the Ag 3d 5/2 spectrum in UHV is 0.3
eV. XPS peaks were calibrated to Au 4f 7/2 at 84.0 eV and Ag
3d 5/2 at 368.3 eV.57 Photoemission features of Ce 3d, Pt 4f,
2628
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
Pd 3d, Rh 3d, and O 1s were tracked during catalysis at
different temperatures. The interpretation of the photoemission
features and the assignment of the binding energies of the
photoelectrons excited from these subshells are based on
references of Ce 3d,57−59 Pt 4f,57,60 Pd 3d,57,61 Rh 3d,57,62 and
O 1s.57 All Ce 3d spectra were deconvoluted by building four
peaks of Ce3+ (u′, uo, v′, and vo) and six peaks of Ce4+ (u‴, u″,
the reaction cell is less than 10−15%. The existence of the gas
phase above the sample surface was confirmed by the
observation of the N 1s photoemission peak from nitrogen
atoms of N2 at 405 eV duing a test of an Ag thin film in the
nitrogen gas phase.56 The pressure in the reaction cell was
tuned through the mass flow meters installed on the tube for
gas introduction and/or through the pumping rate at the exit of
the reaction cell. Ce 3d, Pt 4f, Pd 3d, Rh 3d, and O 1s were
tracked when the catalyst was in the gas mixture of pure CH4 (2
Torr) and pure O2 (1 Torr) at different temperatures.
The pressure of the reactant gas, a mixture of pure CH4 and
O2 in a molar ratio of 2:1 around the catalyst in the reaction cell
of our AP−XPS system, was 3 Torr during the in situ studies of
the surface chemistry of these catalysts. In the experimental
evaluation of the catalytic performance, 5% CH4 was used. The
partial pressure was about 38 Torr. Because we assume that the
difference in the pressure of CH4 in the reaction cell of the
AP−XPS (3 Torr) and that in the reactor of catalysis test (38
Torr) does not result in a significant difference in the surface
chemistry of a catalyst, the surface chemistry revealed with AP−
XPS can represent that of the catalysts during catalysis in a flow
reactor.
59
u, v‴, v″, and v) based on the peak positions of pure Ce2O3
and pure CeO2 reported by Mullins et al.59 and ref 63. Here,
each Ce 3d spectrum was deconvoluted into 10 peaks,
including 880.5 with a fwhm of 2.5 eV, 882.7 with a fwhm of
2.0 eV, 885.1 with a fwhm of 2.4 eV, 888.9 with a fwhm of 4.6
eV, 898.5 with a fwhm of 1.9 eV, 899.2 with a fwhm of 2.5 eV,
901.2 with a fwhm of 2.0 eV, 903.5 with a fwhm of 2.4 eV,
907.5 with a fwhm of 4.6 eV, and 916.3 with a fwhm of 1.9 eV.
The details of the deconvolution of Ce 3d of the CeO2 particles
can be found in the Experimental Section and in the Supporting
Information of a recent publication.63 In the deconvolution of
Pt 4f, the peak positions of Pt0, Pt2+, and Pt4+ were assigned
with the peak positions reported in literature.60 In addition, the
fwhm of all components of the Pt 4f spectrum were constrained
to be the same. In the fitting of Pd 3d, peak positions of Pd0,
Pd2+, and Pd4+ were referred to the literature,61 and all
components maintained the same fwhm. Rh 3d spectra were
deconvoluted by building components of Rh3+ and Rh0 at the
corresponding peak positions reported in the literature62 and
maintaining the same fwhm of each component.
3. RESULTS AND DISCUSSION
3.1. Pd-Doped CeO2. 3.1.1. Pd-CeO2 at Room Temper-
ature. The as-synthesized catalyst (Pd-CeO2) was pretreated in
air or 5% H2/Ar. Figure 1a presents TEM images of Pd-CeO2
All XPS data presented in this article were collected from
catalysts at the listed temperature (25−600 °C) in a gas
environment of reactants with a listed pressure (2 Torr CH4
and 1 Torr O2). A catalyst sample was installed into a sample
stage located in the reaction cell for tracking the surface
chemistry during catalysis. The sample was heated via
conduction from an external wall of the reaction cell. The
external wall of the sample stage was heated via the
bombardment of electrons from an e-beam heater installed
on the vacuum side of the reaction cell. The heating of the
sample in the reaction cell is presented schematically in Figure
S1.
To exclude any potential side reactions of the nickel in the
stainless steel of the sample stage, we deliberately oxidized the
sample stage. The oxidation was confirmed with XPS. The
oxidation of the wall of the reaction cell excluded the
participation of the nickel in the stainless steel in catalysis. In
addition, the aperture was coated with a thick layer of graphite
(about 0.1 mm or thicker) when we received it. Because of a
thermal decomposition of the reactant CH4 or the byproducts
of the MPO at high temperature, the carbon dissociated from
CH4 or the byproducts of the MPO at high temperature could
have accumulated on the aperture. Because Au and graphite
were subsequently deposited on the aperture by the
manufacturer, the buried nickel of the stainless steel of the
aperture located below the graphite and Au thin films was not
exposed to any gases.
The sample temperature was controlled through tuning the
emission current of the filament of the e-beam heater. Reactant
gases were introduced to the reaction cell through the entrance
pore (Figure S1), flowed through the sample, and exited
through both the exit pore (Figure S1) and the aperture
adapted to a prelens. Gas pressure in the reaction cell was
measured through averaging the readings of two gauges
installed at the entrance and exit tubes of the reaction cell.
The difference between the pressures at the entrance and exit of
Figure 1. High-resolution TEM images of Pd-CeO2-air before (a) and
after (b) catalysis. The average size of CeO2 doped with Pd is 13.0 nm
based on a statistical accounting of the nanoparticles in the TEM
images from different parts of the catalyst sample.
pretreated in air (termed Pd-CeO2-air). There are not any
identifiable nanoclusters on the surface of the Pd-CeO2-air,
showing that Pd ions were doped in the lattice of CeO2 instead
of the formation of PdOx or Pd nanoparticles on CeO2. The
interplanar distances of CeO2 doped with Pd ions (Pd-CeO2-
air) were measured in TEM studies. They are very close to pure
CeO2. The almost identical interplanar distance could partially
result from the small difference between the ionic radius of
Ce4+ (101 pm) and the doped ions (e.g., Pt2+ (94 pm), Pd2+
(100 pm), Pd4+ (75.5 pm), and Rh3+ (80.5 pm)).64
Figures 2a1 and 2b1 are the photoemission features of Pd 3d
of Pd-CeO2-air and Pd-CeO2-H2 in the mixture of reactants at
room temperature. They represent the surface chemistry of Pd-
CeO2-air and Pd-CeO2-H2 before catalysis because Pd-CeO2-air
and Pd-CeO2-H2 are not active for MPO at room temperature.
In the following section, we will use the surface chemistry of Pt-
CeO2-air, Pt-CeO2-H2, Rh-CeO2-air, and Rh-CeO2-H2 at room
temperature in the mixture of CH4 and O2 to represent their
2629
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
Figure 3. High-resolution TEM images of Pd-CeO2-H2 before (a) and
after (b) catalysis. The arrows mark the nanoparticles in Pd-CeO2-H2
after the catalysis reaction. The average size of CeO2 doped with Pd is
34.0 nm based on a statistical accounting of the nanoparticles in TEM
images from different parts of the catalyst sample.
nanoparticles with a size of ∼1−3 nm. Compared to the lack
of metal nanoparticles in Pt-CeO2-H2 (Figure 8) and Rh-CeO2-
H2 (Figure 13) upon pretreatment in H2 at 500 °C, Pd in Pd-
CeO2-H2 (Figure 3) forms Pd metal nanoparticles upon the
same treatment. One potential reason for the formation of Pd
nanoparticles in Pd-CeO2 upon H2 pretreatment could relate to
the surface energy of metallic Pd nanoparticles, which is lower
than that of Pt and Rh. It is reasonable to use the surface
energies of the (111) faces of Pd, Pt, and Rh to represent the
surface energies of their metal nanoparticles in general because
of the lack of surface energy of the nanoparticles of these
metals. Pd(111) has a surface energy of 1.88 J m−2 in contrast
to 2.35 J m−2 of Pt (111) and 2.78 J m−2 of Rh(111).65 The
lower surface energy of Pd makes the formation of Pd metal
nanoparticles relatively thermodynamically favorable in contrast
to Pt and Rh. In addition, Pd can catalyze the dissociation of H2
even at room temperature. Hydrogen atoms could further
reduce other Pd ions to metal Pd atoms under a mild condition
and even accelerate the nucleation to form Pd nanoparticles.
3.1.2. Catalytic Performance of Pd-CeO2-air and Pd-CeO2-
H2. Figure 4 presents the catalytic conversion and selectivity of
Pd-CeO2-air for MPO. The conversion of CH4 increases as a
function of temperature and reaches 84.0% at 700 °C (Table
1). Notably, the selectivity for the production of H2 increases
Figure 2. XPS spectra of the in situ studies of Pd 3d of Pd-CeO2-air
(a), Pd 3d of Pd-CeO2-H2 (b), Ce 3d of Pd-CeO2-air (c), and Ce 3d
of Pd-CeO2-H2 (d) during catalysis. The fwhm of the fitted peaks of
Pd 3d in panels a and b were fixed at 1.4 and 1.2 eV, respectively. The
ratio of the partial pressures of CH4 to O2 in the mixture reactant is
2:1, with a total pressure of 3 Torr. The flow rate is 6 mL min−1. All
AP−XPS data were collected at the same pressure and flow rate of
CH4 and O2. The only variable is the reaction temperature. The
spectra collected at 25 °C is used to represent the surface chemistry of
the catalysts before catalysis because none of these catalysts is active
for MPO at 25 °C.
surface chemistries before catalysis because none of them are
active for MPO at room temperature.
It is noted that there is no charging for Pd 3d and Ce 3d of
the sample at room temperature. The lack of charging is
evidenced by the fact that the peak position u‴ of Ce 3d from
Ce4+ at ∼916.3 eV observed in Figure 2c1,d1 is identical to
those reported in refs 59 and 63. Because a Ce 3d spectrum of
CeO2 is typically deconvoluted into six peaks of Ce4+ and four
peaks of Ce3+,59,63 the fraction of Ce3+ could largely influence
the shape of the Ce 3d spectrum. For a Ce 3d spectrum of
CeO2 with a very low fraction of Ce3+, there is one obvious
valley at ∼883 eV (Figure 2c2,d2). If the fraction of Ce3+,
∼30%, is relatively high, then the valley is filled by one peak of
Ce3+ at 883 eV (Figure 2d1). Thus, the shape of Ce 3d
spectrum cannot be used to judge whether there is charging or
not.
Before the catalysis of MPO, Pd of Pd-CeO2-air exists in an
oxidized state (Figure 2a1) and exists mainly in a metallic state
for Pd-CeO2-H2 at room temperature (Figure 2b1). The
photoemission features of Pd 3d and Ce 3d at a temperature
higher than 25 °C during catalysis will be discussed in the
Results and Discussion, section 3.5 even though they were
presented in Figure 2 along with those of Pd-CeO2-air and Pd-
CeO2-H2 before catalysis.
Different from Pd-CeO2-air, the as-synthesized CeO2 doped
with Pd ions was pretreated in 5% H2/Ar. It is termed Pd-
CeO2-H2. Figure 3a is a TEM image of Pd-CeO2-H2 before
catalysis. Unfortunately, Pd aggregated and formed some
Figure 4. Catalytic performance of Pd-CeO2-air. The left axis (red)
and the red line show the conversion of methane. The right axis (blue)
and the two blue lines show the selectivities for the production of H2
(solid dot) and CO (solid triangle). SH2 (%) is defined as [moles of H2
produced/(2 × moles of CH4 consumed)] × 100. SCO (%) is defined
as [moles of CO produced/(moles of CH4 consumed)] × 100.
2630
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
a
Table 1. Catalytic Performance of Methane Partial Oxidation on Pd-CeO2-air, Pt-CeO2-air, and Rh-CeO2-air catalysts
temperature (oC)
300
catalyst
XCH4 (%)
SH2 (%)
SCO (%)
H2/CO ratio
Pd-CeO2-air
Pt-CeO2-air
Rh-CeO2-air
Pd-CeO2-air
Pt-CeO2-air
Rh-CeO2-air
Pd-CeO2-air
Pt-CeO2-air
Rh-CeO2-air
Pd-CeO2-air
Pt-CeO2-air
Rh-CeO2-air
Pd-CeO2-air
Pt-CeO2-air
Rh-CeO2-air
13.2
1.8
0.0
0.0
0.0
0.0
5.3
0.0
0.0
400
500
600
700
28.7
28.8
37.4
45.9
43.9
61.4
61.5
53.1
83.6
84.0
74.6
95.2
48.4
44.8
52.8
78.8
72.2
83.3
87.6
74.2
89.9
92.5
88.5
92.9
11.8
5.8
7.2
14.7
10.4
73.7
4.2
11.2
43.0
36.6
53.3
71.2
56.6
77.5
91.9
85.8
85.7
3.1
2.6
2.6
2.3
2.2
2.3
2.2
a
The oxygen conversion was always 100%.
quickly in the temperature range of 300−500 °C. At 500 °C, it
reaches 83% with a following increase to 92.5% at 700 °C.
Compared to H2, the selectivity of producing CO is less than
that for H2 by ∼25−35% at a temperature lower than 500 °C.
At temperatures higher than 500 °C, the difference in selectivity
between CO and H2 decreases. At 700 °C, it exhibits the same
selectivity for the production of H2 and CO. As shown in
Figure 5, the ratio of H2 to CO decreases with the temperature
and reaches the stoichiometric ratio of 2 at a temperature
higher than 500 °C.
H2 were prepared with the same method as Pd-doped CeO2-air
and Pd-doped CeO2-H2, respectively. Figure 7a is one
representative TEM image of the Pt-CeO2-air before catalysis.
TEM studies showed there are no nanoclusters identified,
suggesting a homogeneous distribution of Pt ions in Pt-CeO2-
air before catalysis. A representative TEM image of Pt-CeO2-air
after catalysis is shown in Figure 7b. The lack of identifiable
nanoparticles in Pt-CeO2-air after catalysis suggests that there is
not an aggregation of Pt atoms during catalysis. Figure 8a is a
TEM image of Pt-CeO2-H2 before catalysis. Compared to Pd-
CeO2-H2 before catalysis (Figure 3a), there are not any Pt
nanoparticle identifiable in Pt-CeO2-H2 before catalysis (Figure
8a). In addition, the TEM image of Pt-CeO2-H2 after catalysis
(Figure 8b) shows that there is no aggregation of Pt atoms
during catalysis.
The deconvolution of the Pt 4f photoemission feature of Pt-
CeO2-air before catalysis (Figure 9a1) suggests that there could
be three different types of Pt atoms, which each exhibit different
bonding to oxygen atoms. There are also three different
oxidation states of Pt atoms (Pt0, Pt2+, and Pt4+) in Pt-CeO2-H2
before catalysis (Figure 9b1). It is noted that there are no Pt
metal nanoparticles observed in the high-resolution TEM
images even though there is a component with a binding energy
of 71.2 eV in Pt-CeO2-H2 before catalysis. The lack of Pt metal
nanoparticles is supported by detailed analysis of TEM and
XPS. There were no Pt nanoparticles found in extensive studies
of Pt-CeO2-air and Pt-CeO2-H2. Alternatively, a Pt atom could
bond to Ce atoms upon pretreatment in H2. The removal of
oxygen atoms bonding to Ce atoms could have created oxygen
vacancies that a Pt atom nearby could fill in to form Pt−Ce
bonds; in this case, Pt atoms do not form any Pt nanoparticles
even though Pt exhibits a binding energy close to metallic Pt.
We call this type of bonding of Pt the “pseudo-metallic state”.
The pseudo-metallic state of Pt will be further discussed in the
Results and Discussion, section 3.5. In addition, a clear
difference in the Ce 3d spectrum of Pt-CeO2-air before
catalysis (Figure 9c1) and Pt-CeO2-H2 before catalysis (Figure
9d1) is that the fraction of Ce3+ in Pt-CeO2-H2 before catalysis,
26%, is obviously larger than that in Pt-CeO2-air before
catalysis. The large density of oxygen vacancies could mean a
large fraction of Pt atoms bonding to Ce atoms.
Figure 5. Molar ratio of H2 to CO produced from MPO on Pd-CeO2-
air, Pt-CeO2-air, and Rh-CeO2-air in the temperature range of 400−
700 °C.
In contrast to Pd-CeO2-air, Pd-CeO2-H2 exhibits quite a
different catalytic performance (Figure 6). There is no catalytic
activity of MPO on Pd-CeO2-H2 at a temperature ≤500 °C.
However, the conversion of CH4 on Pd-CeO2-H2 rapidly
increases in the temperature range of 550−700 °C (Figure 6a).
This catalyst exhibits an evolution of the selectivity of H2 that is
very similar to that of CO (Figure 6b,c). In contrast to the
sharp increase in both conversion and selectivity on Pd-CeO2-
H2 at a temperature higher than 550 °C, Pd-CeO2-air exhibits a
progressive increase of activity and selectivity over the
temperature range of 350−700 °C.
3.2.2. Catalytic Performance of Pt-CeO2-air and Pt-CeO2-
H2. The catalytic performances of Pt-CeO2-air and Pt-CeO2-H2
3.2. Pt-Doped CeO2. 3.2.1. Pt-CeO2-air and Pt-CeO2-H2 at
Room Temperature. Pt-doped CeO2-air and Pt-doped CeO2-
2631
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
Figure 7. High-resolution TEM images of Pt-CeO2-air before (a) and
after (b) catalysis. The average size of CeO2 doped with Pt is 9.5 nm
based on a statistical accounting of the nanoparticles in the TEM
images from different parts of the catalyst sample.
Figure 8. High-resolution TEM images of Pt-CeO2-H2 before (a) and
after (b) catalysis. The average size of CeO2 doped with Pt is 28.5 nm
based on a statistical accounting of the nanoparticles in the TEM
images from different parts of the catalyst sample.
temperature range of 400−500 °C (Figure 5) largely decreases.
At a temperature higher than 500 °C, this ratio is further
decreased to 2.
Compared to Pt-CeO2-air, Pt-CeO2-H2 exhibits different
catalytic performance. The conversion rate of CH4 on Pt-CeO2-
H2 gradually increases in the temperature range of 400−700 °C
(Figure 11a). In addition, the conversion of CH4 on Pt-CeO2-
H2 is always lower than Pt-CeO2-air by ∼15% up to 700 °C. In
terms of selectivity for the production of H2 and CO, Pt-CeO2-
H2 is inferior to Pt-CeO2-air throughout the whole temperature
range (350−700 °C) (Figure 11b,c).
3.3. Rh-Doped CeO2. 3.3.1. Rh-CeO2-air and Rh-CeO2-H2
at Room Temperature. Rh-CeO2-air and Rh-CeO2-H2 were
synthesized with the same method as those of Pt-CeO2-air and
Pt-CeO2-H2, respectively. The high-resolution TEM images in
Figures 12 and 13 show that there are no nanoparticles in Rh-
CeO2-air before catalysis (Figure 12a), in Rh-CeO2-air after
catalysis (Figure 12b), in Rh-CeO2-H2 before catalysis (Figure
13a), and in Rh-CeO2-H2 after catalysis (Figure 13b). They
further suggest that Rh ions were homogeneously distributed in
CeO2. Similar to Pd-CeO2-H2 before catalysis and Pt-CeO2-H2
before catalysis, the fraction of Ce3+ in the total Ce3+ and Ce4+
of Rh-CeO2-H2 before catalysis, 25%, is relatively high. In
addition, the homogeneous doping of Rh ions in Rh-CeO2-air
before catalysis is supported by the lack of a Rh 3d
photoemission feature of metallic Rh (Figure 14a1). Compared
to Rh-CeO2-air before catalysis, Rh in Rh-CeO2-H2 before
catalysis is in a metallic state. However, there are no Rh
nanoparticles observed in the TEM studies of Rh-CeO2-H2
before catalysis (Figure 13a). This pseudo-metallic state of Rh
Figure 6. Catalytic performance of Pd-CeO2-H2 for MPO. Conversion
of methane (a), selectivity to H2 (b), and selectivity to CO (c). For the
convenience of comparison, the catalytic performance of Pd-CeO2-air
(red line) under the same catalytic condition as Pd-CeO2-H2 (blue
line) is listed in a parallel manner.
were measured. Figure 10 presents the catalytic conversion and
selectivity for MPO on Pt-CeO2-air. The conversion of CH4 on
Pt-CeO2-air increases as a function of the reaction temperature
and reaches 74.6% at 700 °C (Table 1). Notably, the selectivity
for the production of H2 increases rapidly in the temperature
range of 300−500 °C. At 500 °C, it reaches 72.2%, with a
following increase to 88.5% at 700 °C. Compared to the
selectivity for H2 production, the selectivity for the production
of CO is lower by 10−20% at a temperature lower than 500 °C.
At 700 °C, the selectivity for CO is the same as that for H2
(Figure 10). Because of the higher H2 selectivity of MPO on Pt-
CeO2-air at a low temperature such as 400 °C, the molar ratio
of H2 to CO at low temperatures (Figure 5) is much higher
than the stoichiometric ratio of 2. The ratio of H2 to CO in the
2632
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
Figure 9. XPS spectra of the in situ studies of Pt 4f of Pt-CeO2-air (a),
Pt 4f of Pt-CeO2-H2 (b), Ce 3d of Pt-CeO2-air (c), and Ce 3d of Pt-
CeO2-H2 (d) during catalysis. The fwhm of the fitted peaks of Pt 4f in
panels a and b were fixed at 1.2 eV, respectively. The ratio of the
partial pressures of CH4 to O2 in the mixture reactant is 2:1. The flow
rate is about 6 mL min−1. All data were collected at the same pressure
and flow rate of CH4 and O2. The only variable is the reaction
temperature.
Figure 11. Catalytic performance of Pt-CeO2-H2. Conversion of
methane (a), selectivity to H2 (b), and selectivity to CO (c). For the
convenience of comparison, the catalytic performance of Pt-CeO2-air
(red line) under the same catalytic condition as Pt-CeO2-H2 (blue
line) is listed in a parallel manner.
Figure 10. Catalytic performance of Pt-CeO2-air. The left axis (red)
and the red line show the conversion of methane. The right axis (blue)
and the blue lines show the selectivities for the production of H2 (solid
dot) and CO (solid triangle).
for CO is less than that for H2 by 20−30% at a temperature
lower than 500 °C. As shown in Figure 5, the H2-to-CO ratio in
the temperature range of 400−500 °C is far greater than 2; at
temperatures higher than 500 °C, the ratio is decreased to 2.
As shown in Figure 16, Rh-CeO2-H2 exhibits a catalytic
performance almost identical to Rh-CeO2-air in the temper-
ature range of 350−700 °C. This is different from the obvious
variation between Pt-CeO2-air and Pt-CeO2-H2 (Figure 11)
and that between Pd-CeO2-air and Pd-CeO2-H2 (Figure 6). In
addition, among Pd-CeO2-air, Pd-CeO2-H2, Pt-CeO2-air, Pt-
CeO2-H2, Rh-CeO2-air, and Rh-CeO2-H2, the difference in
catalytic conversion and selectivity to CO and H2 between Pd-
CeO2-air and Pd-CeO2-H2 is much larger than those between
Pt-CeO2-air and Pt-CeO2-H2 and between Rh-CeO2-air and
Rh-CeO2-H2. In the case of Pd-CeO2-H2, this could result from
in Rh-CeO2-H2 before catalysis can be rationalized with the two
arguments made for Pt-CeO2-air and Pt-CeO2-H2 in the
Results and Discussion, section 3.2.1.
3.3.2. Catalytic Performance of Rh-CeO2-air and Rh-CeO2-
H2. The catalytic performance of Rh-CeO2-air and Rh-CeO2-H2
was measured under the same conditions as the other catalysts.
Figure 15 presents the catalytic conversion and selectivity of
MPO on Rh-CeO2-air. The conversion of CH4 increases rapidly
as a function of the reaction temperature and reaches 95.2% at
700 °C (Table 1). The selectivity for the production of H2
increases fast in the temperature range of 300−500 °C (Figure
15). At 500 °C, it reaches 83.3% followed by a further increase
to 92.9% at 700 °C. Compared to H2 selectivity, the selectivity
2633
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
Figure 12. High-resolution TEM images of Rh-CeO2-air before (a)
and after (b) catalysis. The average size of CeO2 doped with Rh is 22.5
nm based on a statistical accounting of the nanoparticles in the TEM
images from different parts of the catalyst sample.
Figure 14. XPS spectra of the in situ studies of Rh 3d of Rh-CeO2-air
(a), Rh 3d of Rh-CeO2-H2 (b), Ce 3d of Rh-CeO2-air (c), and Ce 3d
of Rh-CeO2-H2 (d) during catalysis. The fwhm of the fitted peaks of
Rh 3d in panels a and b were constrained between 1.2 and 1.4 eV,
respectively. The ratio of the partial pressures of CH4 to O2 in the
mixture reactant is 2:1, with a total pressure of 3 Torr. The flow rate is
6 mL/min. All AP−XPS data were collected at the same pressure and
flow rate of CH4 and O2. The only variable is the reaction temperature.
Figure 13. High-resolution TEM images of Rh-CeO2-H2 before (a)
and after (b) catalysis. The average size of CeO2 doped with Rh is 30.5
nm based on a statistical accounting of the nanoparticles in the TEM
images from different parts of the catalyst sample.
some side reactions on the Pd metal nanoparticles. At 700 °C,
Rh-CeO2-air and Rh-CeO2-H2 have a similar selectivity for H2
and CO of 88−94%, respectively (Figure 16). At this
temperature, the ratio of H2 to CO is the stoichiometric
ratio, 2.
3.4. Turnover Frequency and Apparent Activation
Energy. For a parallel comparison of the catalytic perform-
ances in a kinetics-controlled regime, turnover frequencies
(TOF) and apparent activation energy (Ea) for MPO on the six
catalysts were measured. In all of these measurements, the
highest conversion of CH4 remained less than 15%. Because Pd
metal nanoparticles were clearly observed in Pd-CeO2-H2 after
catalysis (Figure 3b), we did not calculate the TOF of Pd-
CeO2-H2. For the other five catalysts (Pd-CeO2-air, Pt-CeO2-
air, Pt-CeO2-H2, Rh-CeO2-air, and Rh-CeO2-H2), we exper-
imentally evaluated their TOFs under the assumption that all
atoms of Pd, Pt, or Rh on the surface layers of the five catalysts
are active sites for MPO regardless of whether they appear in a
metallic or ionic state. In the Results and Discussion, section
3.5, we rationalize that the pseudo-metallic state of Pd in Pd-
CeO2-air and of Pt in Pt-CeO2-H2 and Pt-CeO2-air observed
during catalysis are not metal atoms of metal nanoparticles. The
ratios of the noble metal (Pd, Pt, or Rh) of the topmost surface
layer to Ce of the topmost surface layer during catalysis at
different temperatures were calculated using the peak areas of
Pd 3d, Pt 4f, Rh 3d, and Ce 3d collected during catalysis
through Beer’s law. The calculated atomic ratios of Pd, Pt, and
Rh to Ce during catalysis are listed in Figure S4. They are
approximately close to the bulk ratios of doped ions in CeO2.
The procedure of calculating TOFs is also described in the
Figure 15. Catalytic performance of Rh-CeO2-air. The left axis (red)
and the red line show the conversion of methane. The right axis (blue)
and the blue lines show the selectivities to the production of H2 (solid
dot) and CO (solid triangle).
Supporting Information. The calculated TOF at different
temperatures is plotted in Figure 17. Obviously, Rh-doped
catalysts (Rh-CeO2-H2 and Rh-CeO2-air) exhibit catalytic
activity higher than Pt- and Pd-doped catalysts (Pd-CeO2-air,
Pt-CeO2-H2, and Pt-CeO2-air). In addition, they also exhibit
higher catalytic selectivity for the production of CO and H2
than Pd- and Pt-doped catalysts based on the catalytic
performances presented in Figures 4−6, 10, 11, 15, and 16.
The measurements of the apparent activation energy, Ea, for
the six catalysts were performed by controlling the catalytic
conversion of CH4 to be lower than 15%. Figure 18 is the
Arrhenius plots of MPO on the six catalysts at different
2634
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
Figure 17. Turnover frequency of MPO on Rh-CeO2-H2, Rh-CeO2-
air, Pd-CeO2-H2, Pd-CeO2-air, Pt-CeO2-H2, and Pt-CeO2-air meas-
ured in a kinetics-controlled regime.
Figure 18. Arrhenius plots of MPO on catalysts Rh-CeO2-H2, Rh-
CeO2-air, Pd-CeO2-H2, Pd-CeO2-air, Pt-CeO2-H2, and Pt-CeO2-air. In
LnX, X is the conversion rate.
the identity of the surface chemistry of the two catalysts during
catalysis (Figure 14).
3.5. In Situ AP−XPS Studies and Correlation to
Catalytic Performance. To understand the different catalytic
behaviors of these catalysts, in situ studies of the surface
chemistries of these catalysts during catalysis were performed
with our in-house ambient-pressure XPS. Photoelectrons were
excited with a monochromatic Al Kα source for a better
resolution. In situ AP−XPS studies of the six catalysts were
performed at a temperature up to 600 °C for M-CeO2-air or
500 °C for M-CeO2-H2. The collected spectra have a good
signal-to-noise ratio. A continuous accumulation of atomic
carbon that dissociated from CH4 at high temperatures blocked
the nozzle between the reaction cell and the energy analyzer
and thus prevented it from collecting photoelectrons during in
situ studies of Pd-CeO2-H2, Pt-CeO2-H2, and Rh-CeO2-H2 at
600 °C and higher. The temperature of the aperture must have
been very high because the distance between the sample surface
and the aperture is only 1 mm or less. Thus, we have to limit
the highest temperature for taking spectra during catalysis of
Pd-CeO2-H2, Pt-CeO2-H2, and Rh-CeO2-H2 to 500 °C
(Figures 2b, 9b, and 14b).
Figure 16. Catalytic performance of Rh-CeO2-H2. Conversion of
methane (a), selectivity to H2 (b), and selectivity to CO (c). For the
convenience of comparison, the catalytic performance of Rh-CeO2-air
(red line) under the same catalytic condition as Rh-CeO2-H2 (blue
line) is listed in a parallel manner.
temperature ranges. In these plots, the variable is the
conversion rate of CH4. Thus, the calculated activation energies
from the slopes of these plots (Figure 18) are only the apparent
activation energies. For Pt-CeO2-air, Pd-CeO2-air, Rh-CeO2-air,
and Rh-CeO2-H2, the measurements of Ea were conducted in
the temperature range of 360−410 °C. Among them, Pd-CeO2-
air and Pt-CeO2-air have rather high apparent activation
barriers of 204.9 and 338.0 kJ mol−1, respectively (Figure 18) in
this temperature region. These quite high activation energies
could result from the coexistence of both oxidizing and metallic
Pt and Pd atoms in Pd-CeO2-air and Pt-CeO2-air, which offset
the measured Ea. Rh-CeO2-air has a much lower activation
energy of 30.2 kJ mol−1 in contrast to Pd-CeO2-air and Pt-
CeO2-air. The active phase of Rh-CeO2-air is CeO2 doped with
Rhn+ ions. It is worth noting that Rh-CeO2-H2 exhibits an
activation barrier very similar to Rh-CeO2-air, consistent with
Figure 14 presents the photoemission features of Rh 3d of
Rh-CeO2-air and Rh-CeO2-H2 at catalytic temperatures in a
mixture of 2 Torr CH4 and 1 Torr O2. For Rh-CeO2-air during
catalysis (Figure 14a2−a5), Rh is in an oxidizing state in the
2635
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
temperature range of 300−500 °C. TEM studies of Rh-CeO2-
air and Rh-CeO2-H2 after catalysis suggested that there are no
Rh or RhOx nanoparticles on or in CeO2 (Figures 12b and
13b). These experimental data from in situ studies showed the
preservation of an ionic state of Rh in Rh-CeO2-air before
catalysis, during catalysis, and after catalysis. The preservation
of an ionic state of Rh of Rh-CeO2-air during catalysis can be
rationalized by the high bond strength of Rh−O bonds.50
In contrast to Rh-CeO2-air, Rh in Rh-CeO2-H2 before
catalysis (Figure 14b1) is in a metallic state in the surface
region. However, the TEM studies of the catalyst, Rh-CeO2-H2
before catalysis, showed there are no Rh metal nanoparticles. In
fact, the high density of oxygen vacancies in catalyst Rh-CeO2-
H2 before catalysis (Figure 14d1) supports that Rh likely bonds
to Ce atoms upon the removal of some of the surface lattice
oxygen atoms of CeO2. The direct bonding of a Rh atom to Ce
atoms makes Rh appear to be in a metallic state although there
are no Rh nanoparticles formed. Here, we call the low binding
energy of Rh 3d of Rh bonding to Ce atoms a pseudo-metallic
state because the Rh atoms do not form any Rh metal
nanoparticles.
AP−XPS studies showed that Rh of Rh-CeO2-H2 was
oxidized to an ionic state during catalysis at 300 °C (Figure
14b2). TEM studies of Rh-CeO2-H2 after catalysis (Figure 13b)
showed there are no RhOx nanoparticles. The upshift in Rh 3d
was driven by the oxidization of Rh atoms bonding to Ce atoms
(in an pseudo-metallic state) in an ionic state in the reactant
mixture (CH4 and O2) during catalysis. The similarity in surface
chemistry (Figure 14a4,b4) of Rh-CeO2-air and Rh-CeO2-H2
during catalysis is consistent with the similarity in their catalytic
performance (Figure 16).
In situ AP−XPS showed that the surface chemistries of Pt-
CeO2-H2 and Pt-CeO2-air during catalysis are different even
though the catalytic conditions are exactly the same. In the case
of Pt-CeO2-H2, 76% Pt atoms remain in a pseudo-metallic state
in the temperature range of 300−500 °C during catalysis
(Figure 9b). However, about 52% ionic Pt of Pt-CeO2-air is
reduced to a pseudo-metallic state during catalysis (Figure 9a).
Compared to the preservation of the ionic state of Rh of Rh-
CeO2-air during catalysis, the reduction of the ionic state Pt of
Pt-CeO2-air to a pseudo-metallic state during catalysis results
from the low bond strength of Pt−O. Rh−O bonds have a
larger formation energy than Pt−O.66,67
Through correlating the surface chemistries of Pt-CeO2-air
and Pt-CeO2-H2 during catalysis to the corresponding catalytic
performances, we deduce that the high conversion of methane
on Pt-CeO2-air (Figure 11a) could result from the high fraction
of Pt ions in Pt-CeO2-air compared to Pt-CeO2-H2 during
catalysis (Figure 9). This is consistent with the understanding
of the catalytic performance on CeO2 doped with Pt for MPO
based on DFT calculation.47 The DFT calculation47 showed
that the ionic-state Pt doped in the lattice of CeO2 can decrease
the bond strength of surface-lattice oxygen to Ce ions, which
further decreases the activation barrier for breaking C−H on
the surface-lattice oxygen atoms.47
because Pt or Pd in a pseudo-metallic state are formed based on
the in situ AP−XPS studies. WGS consumes CO and H2O to
produce CO2 and H2. A high reaction temperature is not
favorable for the conversion because WGS is slightly
exothermic. However, at a relatively low temperature, some
CO and H2O are converted to CO2 and H2 through WGS
because H2O is formed during the combustion of CH4. The
WGS at a relatively low temperature increased the overall
selectivity for the production of H2, leading to a higher ratio of
H2 to CO at a relatively low temperature. At a high
temperature, the WGS is not thermodynamically favorable;
thus, the influence of WGS on the ratio of H2 to CO is
negligible, and the ratio of H2 to CO decreases to the
stoichiometric ratio of MPO, 2:1, at high temperatures, such as
700 °C.
The high ratio of H2 to CO of MPO on Rh-CeO2-air and Rh-
CeO2-H2 cannot be rationalized with the above argument. This
is because Rh is in an ionic state during MPO. There are no
sites for WGS. Alternatively, the high ratio of H2 to CO in
MPO on Rh-CeO2-air and Rh-CeO2-H2 could result from a
different side reaction, CO oxidation. By considering the
reported high activity of CO oxidation on a RhOx thin film
formed on Rh nanoparticles,69,70 the Rh-ions-doped oxide
surface may catalyze CO oxidation with a low activation energy.
CO oxidation on Rh-CeO2-air and Rh-CeO2-H2 can consume
CO and thus increase the H2/CO ratio. A computational study
was suggested to provide insights for the expected activity of
CO oxidation on this Rhn+-doped oxide surface.
Compared to Rh-CeO2-H2, Rh-CeO2-air, Pt-CeO2-H2, and
Pt-CeO2-air, the evolution of the surface chemistries of Pd-
CeO2-H2 and Pd-CeO2-air are more complicated. As shown in
Figure 2a1,b1, the surface chemistries of Pd-CeO2-air and Pd-
CeO2-H2 before catalysis are quite different. Pd exhibits an
oxidizing state in Pd-CeO2-air before catalysis in contrast to Pd-
CeO2-H2 before catalysis in which the majority of Pd is in the
metallic state (Figure 2d1). This is supported by the
observation of Pd metal nanoparticles in the TEM studies. In
situ studies using AP−XPS showed that ionic Pd of Pd-CeO2-
air is partially reduced to metallic Pd during catalysis and the
fraction of metallic Pd nanoparticles of Pd-CeO2-air increases
with the rising reaction temperature in the temperature range of
300−600 °C (Figure 2a). This is partially due to the reduction
of the doped Pd ions by the products of MPO, H2, and/or CO.
All Pd in the Pd-CeO2-air during catalysis at 600 °C exists in a
pseudo-metallic state (Figure 2a5) because there are no Pd
metal nanoparticles observed in the Pd-CeO2-air after catalysis
(Figure 1b). Correspondingly, Pd-CeO2-air at 600 °C exhibits a
high conversion and selectivity for the production of H2 and
CO (Figure 4). Correlations of the surface chemistries during
catalysis and the corresponding catalytic performances show
that the pseudo-metallic state of Pd is a component of an active
site.
Figure 2b1−b4 suggest that the Pd metal nanoparticles of
Pd-CeO2-H2 before catalysis were oxidized during catalysis.
This is quite different from the partial reduction of the doped
Pd ions of Pd-CeO2-air during catalysis. In fact, this difference
can be rationalized by their catalytic activity. As shown in
Figure 6a, Pd-CeO2-H2 is active for MPO only at a temperature
higher than 500 °C. At a temperature ≤500 °C, the reactive
environment is a mixture of CH4 and O2 without any
component of a reducing gas such as CO or H2. Certainly,
O2 readily oxidized the Pd metal nanoparticles of Pd-CeO2-H2
during catalysis to ionic Pd (Figures 2b1−b4). As the oxidation
In the catalysis of MPO on Pt-CeO2-air, Pt-CeO2-H2, Pd-
CeO2-air, and Pd-CeO2-H2, the ratio of H2 to CO is always
higher than the stoichiometric ratio of MPO, 2, at a
temperature lower than 650−700 °C (Figure 5). This means
that the selectivity for the production of CO on these catalysts
is always lower than that of H2 at a temperature lower than
650−700 °C. This is probably related to a side reaction, a
water-gas shift (WGS) performed at the metal-oxide interface,68
2636
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
of Pd metal nanoparticles must be performed on the external
surface of Pd nanoparticles, a core−shell structured Pd@PdOx
was formed in the temperature range of 300−500 °C (Figures
2b2−b4). The lack of activity of Pd@PdOx at 300−500 °C
suggests that the PdOx shell is not active for MPO. It is noted
that the coordination environment of Pd ions of the PdOx shell
formed through oxidation of the Pd layer of Pd metal plates is
different from that of Pt ions or Rh ions doped in the lattice of
CeO2. The Pd ions doped in the lattice of CeO2 likely activate
neighboring oxygen atoms of CeO2 and thus decrease the
activation energy of C−H dissociation on oxygen atoms of the
surface lattice. However, Pd ions of PdOx formed on Pd-CeO2-
H2 in the mixture of CH4 and O2 (Figure 2b) cannot activate
oxygen atoms of CeO2 because Pd ions in this case are in a
PdOx shell of a Pd@PdOx nanoparticle instead of the surface
lattice of CeO2. Thus, the PdOx shell formed on Pd-CeO2-H2
in the temperature range of 300−500 °C in the mixture of CH4
and O2 is not active for MPO. Thus far, we have developed the
correlations of the in situ surface chemistries of the six catalysts
during catalysis and their corresponding catalytic performances.
Because Pd-CeO2-air, Pt-CeO2-air, and Rh-CeO2-air exhibit
better catalytic performance than Pd-CeO2-H2, Pt-CeO2-H2,
and Rh-CeO2-H2, respectively, we performed tests of stability
of MPO on Pd-CeO2-air, Pt-CeO2-air, and Rh-CeO2-air at 600
°C. As shown in Figure 19, the catalytic performances of the
three catalysts, including conversion, selectivity to H2 and CO,
and the ratio of H2 to CO, remain nearly constant in the first 24
h.
4. CONCLUSIONS
Pd-CeO2-air, Pd-CeO2-H2, Pt-CeO2-air, Pt-CeO2-H2, Rh-
CeO2-air, and Rh-CeO2-H2 catalysts were synthesized. Pd, Pt,
and Rh ions were successfully doped into CeO2 in Pd-CeO2-air,
Pt-CeO2-air, Pt-CeO2-H2, Rh-CeO2-air, and Rh-CeO2-H2,
although Pd metal nanoparticles were formed on CeO2 of
Pd-CeO2-H2 before catalysis. These well-controlled, parallel
studies of the six different catalysts (Pd-CeO2-air, Pd-CeO2-H2,
Pt-CeO2-air, Pt-CeO2-H2, Rh-CeO2-air, and Rh-CeO2-H2)
revealed quite different catalytic performances even though
they were synthesized with the same method. Pd-CeO2-air
exhibits higher catalytic activity than Pd-CeO2-H2 in the
temperature range of 300−550 °C. In situ studies of the surface
chemistry of the catalysts suggested that the pseudo-metallic
state of Pd is active for MPO. For Pt-CeO2-H2, Pt ions were
partially reduced. A comparative study of Pt-CeO2-air and Pt-
CeO2-H2 during catalysis suggests that Pt ions doped in the
surface lattice of CeO2 are a component of the active sites of
MPO. Compared to Pd-CeO2-air, Pd-CeO2-H2, Pt-CeO2-air,
and Pt-CeO2-H2, Rh in Rh-CeO2-air and Rh-CeO2-H2 always
remains in an ionic state. Clearly, Rh ions doped in the CeO2
lattice are a component of the active sites.
Figure 19. Stability tests of conversion, selectivity, and ratio of H2 to
CO of Pd-CeO2-air (a), Pt-CeO2-air (b), and Rh-CeO2-air (c) at 600
°C.
Overall, we developed correlations between the evolution of
catalytic performances of the six catalysts and their correspond-
ing surface chemistries at different temperatures through in situ
studies of the surface chemistry of the six CeO2-based catalysts.
These correlations between the experimental evaluations of
catalytic performances and the corresponding surface chem-
istries during catalysis suggest the necessity of in situ studies in
understanding catalysis. Without in situ studies of surface
chemistry of catalysts during catalysis, the different surface
chemistries of Rh-CeO2-air and Rh-CeO2-H2 before catalysis
(Figure 14a1,b1) identified with ex situ studies contradict the
same catalytic performance of MPO on Rh-CeO2-air and Rh-
CeO2-H2 during catalysis.
ASSOCIATED CONTENT
* Supporting Information
■
S
Schematic showing gas flow and heating in the catalysis reactor
integrated in the APXPS system; catalytic performance of pure
CeO2 for methane partial oxidation; XRD patterns of catalysts
after catalysis; atomic ratio of Pd, Pt, or Rh in the topmost
surface layer to Ce of the topmost surface layer of the catalysts
during catalysis; and calculation of turn-over frequency. This
2637
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
(27) Basile, F.; Benito, P.; Fornasari, G.; Monti, M.; Scavetta, E.;
Tonelli, D.; Vaccari, A. Catal. Today 2010, 157, 183−190.
(28) Karelovic, A.; García, X.; Wojcieszak, R.; Ruiz, P.; Gordon, A. L.
Appl. Catal., A 2010, 384, 220−229.
(29) Tanaka, H.; Kaino, R.; Nakagawa, Y.; Tomishige, K. Appl. Catal.,
A 2010, 378, 187−194.
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
(30) Tanaka, H.; Kaino, R.; Okumura, K.; Kizuka, T.; Nakagawa, Y.;
Tomishige, K. Appl. Catal., A 2010, 378, 175−186.
(31) Boullosa-Eiras, S.; Zhao, T.; Vanhaecke, E.; Chen, D.; Holmen,
A. Catal. Today 2011, 178, 12−24.
*E-mail: ftao@nd.edu. Tel.: (+1) 574 631 1394.
Author Contributions
†These authors equally contributed to this work.
́
(32) Mateos-Pedrero, C.; Duquesne, S.; Carrazan, S. R. G.; Soria, M.
Notes
A.; Ruíz, P. Appl. Catal., A 2011, 394, 245−256.
(33) Berger-Karin, C.; Wohlrab, S.; Rodemerck, U.; Kondratenko, E.
V. Catal. Commun. 2012, 18, 121−125.
(34) Liander, H. Trans. Faraday Soc. 1929, 25, 462.
(35) Ashcroft, A. T.; Cheetham, A. K.; Foord, J. S.; Green, M. L. H.;
Grey, C. P.; Murrell, A. J.; Vernon, P. D. F. Nature 1990, 344, 319−
321.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work is partially supported by the Chemical Sciences,
Geosciences and Biosciences Division, Office of Basic Energy
Sciences, Office of Science, U.S. Department of Energy, under
the Grant DE-FG02-12ER1635.
(36) Hargreaves, J. S. J.; Hutchings, G. J.; Joyner, R. W. Nature 1990,
348, 428−429.
(37) Choudhary, V. R.; Mamman, A. S.; Sansare, S. D. Angew. Chem.,
Int. Ed. Engl. 1992, 31, 1189−1190.
REFERENCES
■
(38) Hickman, D. A.; Schmidt, L. D. Science 1993, 259, 343−347.
(39) Ashcroft, A. T.; Cheetham, A. K.; Green, M. L. H.; Vernon, P.
D. F. Nature 1991, 352, 225−226.
(1) Goetsch, D. A.; Schmidt, L. D. Science 1996, 271, 1560−1562.
(2) Wilson, J. N.; Pedigo, R. A.; Zaera, F. J. Am. Chem. Soc. 2008, 130,
15796−15797.
(40) Pino, L.; Vita, A.; Cordaro, M.; Recupero, V.; Hegde, M. S. Appl.
Catal., A 2003, 243, 135−146.
(3) Donazzi, A.; Livio, D.; Maestri, M.; Beretta, A.; Groppi, G.;
Tronconi, E.; Forzatti, P. Angew. Chem., Int. Ed. 2011, 50, 3943−3946.
(4) Nagaoka, K.; Sato, K.; Takita, Y. J. Catal. 2012, 287, 86−92.
(5) Tsang, S. C.; Claridge, J. B.; Green, M. L. H. Catal. Today 1995,
23, 3−15.
(6) Pena, M. A.; Gomez, J. P.; Fierro, J. L. G. Appl. Catal., A 1996,
144, 7−57.
(41) Pantu, P.; Gavalas, G. R. Appl. Catal., A 2002, 223, 253−260.
(42) Pino, L.; Recupero, V.; Beninati, S.; Shukla, A. K.; Hegde, M. S.;
Bera, P. Appl. Catal., A 2002, 225, 63−75.
(43) Souza, M. M. V. M; Schmal, M. Catal. Lett. 2003, 91, 11−17.
(44) Passos, F. B.; de Oliveira, E. R.; Mattos, L. V.; Noronha, F. B.
Catal. Today 2005, 101, 23−30.
(7) Sokolovskii, V. D.; Coville, N. J.; Parmaliana, A.; Eskendirov, I.;
Makoa, M. Catal. Today 1998, 1998, 191−195.
(8) Rostrup-Nielsen, J. R. Catal. Today 2002, 71, 243−247.
(9) York, A. P. E.; Xiao, T.; Green, M. L. H. Top. Catal. 2003, 22,
345−358.
(45) Passos, F. B.; Oliveira, E. R.; Mattos, L. V.; Noronhe, F. B. Catal.
Lett. 2006, 110, 261−267.
(46) Ahn, K.; Chung, Y.-C.; Oh, J. H.; Hari Prasad, D.; Kim, H.; Kim,
H.-R.; Son, J.-W.; Lee, H.-W.; Lee, J.-H. J. Nanosci. Nanotechnol. 2011,
11, 6414−6419.
(10) York, A. P. E.; Xiao, T.-C.; Green, M. L. H.; Claridge, J. B. Catal.
Rev. 2007, 49, 511−560.
(47) Tang, W.; Hu, Z.; Wang, M.; Stucky, G. D.; Metiu, H.;
McFarland, E. W. J. Catal. 2010, 273, 125−137.
(48) Krcha, M. D.; Mayernick, A. D.; Janik, M. J. J. Catal. 2012, 293,
103−115.
(11) Christian Enger, B.; Lødeng, R.; Holmen, A. Appl. Catal., A
2008, 346, 1−27.
(12) Choudhary, T. V.; Choudhary, V. R. Angew. Chem., Int. Ed.
2008, 47, 1828−1847.
(49) Jankowiak, J.; Barteau, M. J. Catal. 2005, 236, 366−378.
(50) Salmeron, M.; Brewer, L.; Somorjai, G. A. Surf. Sci. 1981, 112,
207−228.
(13) Otsuka, K.; Ushiyama, T.; Yamanaka, I. Chem. Lett. 1993,
1517−1520.
(14) Otsuka, K.; Wang, Y.; Sunada, E.; Yamanaka, I. J. Catal. 1998,
175, 152−160.
(51) Joyner, R. W.; Roberts, M. W.; Yates, K. Surf. Sci. 1979, 87,
501−509.
(15) Au, C.-T.; Ng, C.-F.; Liao, M.-S. J. Catal. 1999, 185, 12−22.
(16) Ruckenstein, E.; Wang, H. Y. J. Catal. 1999, 187, 151−159.
(17) Ruckenstein, E. J. Catal. 2000, 190, 32−38.
(18) Basile, F.; Fornasari, G.; Gazzano, M.; Vaccari, A. J. Mater. Chem.
2002, 12, 3296−3303.
(52) Joyner, R. W.; Roberts, M. W. Chem. Phys. Lett. 1979, 60, 459−
462.
(53) Bukhtiyarov, V. I.; Kaichev, V. V.; Prosvirin, I. P. Top. Catal.
2005, 32, 3−15.
(54) Knop-Gericke, A.; Kleimenov, E.; Haevecker, M.; Blume, R.;
Teschner, D.; Zafeiratos, S.; Schloegl, R.; Bukhtiyarov, V. I.; Kaichev,
V. V.; Prosvirin, I. P.; Nizovskii, A. I.; Bluhm, H.; Barinov, A.; Dudin,
P.; Kiskinova, M. In Advances in Catalysis; Gates, B. C., Knozinger, H.,
Eds., Elsevier: Amsterdam, The Netherlands, 2009; Vol. 52, pp 213−
272.
(19) Horn, R.; Williams, K.; Degenstein, N.; Schmidt, L. J. Catal.
2006, 242, 92−102.
(20) Eriksson, S.; Rojas, S.; Boutonnet, M.; Fierro, J. L. G. Appl.
Catal., A 2007, 326, 8−16.
̀
(21) Montini, T.; Condo, A. M.; Hickey, N.; Lovey, F. C.; De
Rogatis, L.; Fornasiero, P.; Graziani, M. Appl. Catal., B 2007, 73, 84−
(55) Mai, H. X.; Sun, L. D.; Zhang, Y. W.; Si, R.; Feng, W.; Zhang, H.
P.; Liu, H. C.; Yan, C. H. J. Phys. Chem. B 2005, 109, 24380−24385.
(56) Tao, F.; Zhang, S. R.; Nguyen, L.; Zhang, X. Q. Chem. Soc. Rev.
2012, 41, 7980−7993.
(57) Moulder, J. F.StickleW. F.; Sobol, P. E.; Bomben, K. D.
Handbook of X-ray Photoelectron Spectroscopy; Perkin-Elmer Corpo-
ration: Eden Prairie, MN, 1992.
97.
(22) Bitsch-Larsen, A.; Degenstein, N. J.; Schmidt, L. D. Appl. Catal.,
B 2008, 78, 364−370.
(23) Bitsch-Larsen, A.; Horn, R.; Schmidt, L. D. Appl. Catal., A 2008,
348, 165−172.
(24) Donazzi, A.; Michael, B. C.; Schmidt, L. D. J. Catal. 2008, 260,
270−275.
(58) Nolan, P. D.; Lutz, B. R.; Tanaka, P. L.; Mullins, C. B. Surf. Sci.
(25) Michael, B. C.; Donazzi, A.; Schmidt, L. D. J. Catal. 2009, 265,
117−129.
1998, 419, L107−L113.
(59) Mullins, D. R.; Overbury, S. H.; Huntley, D. R. Surf. Sci. 1998,
409, 307−319.
(26) Tanaka, H.; Kaino, R.; Okumura, K.; Kizuka, T.; Tomishige, K.
J. Catal. 2009, 268, 1−8.
2638
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639

ACS Catalysis
Research Article
(60) Bancroft, G. M.; Adams, I.; Coatsworth, L. L.; Bennewitz, C. D.;
Brown, J. D.; Westwood, W. D. Anal. Chem. 1975, 47, 586−588.
(61) Kim, K. S.; Gossmann, A. F.; Winograd, N. Anal. Chem. 1974,
46, 197−200.
(62) Contour, J. P.; Mouvier, G.; Hoogewys, M. J. Catal. 1977, 48,
217−228.
(63) Wen, C.; Zhu, Y.; Ye, Y. C.; Zhang, S. R.; Cheng, F.; Liu, Y.;
Wang, P.; Tao, F. ACS Nano 2012, 6, 9305−9313.
(64) Handbook of Inorganic Compounds; Perry, D. L., Phillips, S. L.,
Eds.; CRC Press: Boca Raton, FL, 1995.
(65) Skriver, H. L.; Rosengaard, N. M. Phys. Rev. B 1992, 46, 7157−
7168.
(66) Samsonov, G. V. The Oxide Handbook,. 2nd ed.; IFI/Plenum:
New York, 1982.
(67) Bichowsky, F. R.; Rossini, F. D. Thermochemistry of the Chemical
Substances; Reinhold Publishing: New York, 1936.
(68) Fu, Q.; Saltsburg, H.; Flytzani-Stephanopoulos, M. Science 2003,
301, 935−938.
(69) Blomberg, S.; Lundgren, E.; Westerstrom, R.; Erdogan, E.;
Martin, N. M.; Mikkelsen, A.; Andersen, J. N.; Mittendorfer, F.;
Gustafson, J. Surf. Sci. 2012, 606, 1416−1421.
(70) Grass, M. E.; Zhang, Y.; Butcher, D. R.; Park, J. Y.; Li, Y.;
Bluhm, H.; Bratlie, K. M.; Zhang, T.; Somorjai, G. A. Angew. Chem.,
Int. Ed. 2008, 47, 8893−8896.
2639
dx.doi.org/10.1021/cs400070y | ACS Catal. 2013, 3, 2627−2639