Catalytic performance and characterization of Rh-CeO2/MgO catalysts for the catalytic partial oxidation of methane at short contact time

Article abstract of DOI:10.1016/j.jcat.2009.08.008

The addition of an optimum amount of CeO₂ (Ce/Rh = 4) to 1.0 wt% Rh/MgO promoted the catalytic partial oxidation (CPO) of methane with N₂ dilution. At the same time, it can suppress the temperature increase at the catalyst bed inlet

Full text of DOI:10.1016/j.jcat.2009.08.008

Journal of Catalysis 268 (2009) 1–8
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Catalytic performance and characterization of Rh–CeO2/MgO catalysts
for the catalytic partial oxidation of methane at short contact time
Hisanori Tanaka a, Rie Kaino a, Kazu Okumura b, Tokushi Kizuka a, Keiichi Tomishige a,c,
*
a Graduate School of Pure and Applied Sciences, University of Tsukuba, 1-1-1, Tennodai, Tsukuba, Ibaraki 305-8573, Japan
b Department of Chemistry and Biotechnology, Graduate School of Engineering, Tottori University, Koyama-cho Minami, Tottori 680-8552, Japan
c Japan Science and Technology Agency, CREST, 1-1-1, Tennodai, Tsukuba, Ibaraki 305-8573, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The addition of an optimum amount of CeO2 (Ce/Rh = 4) to 1.0 wt% Rh/MgO promoted the catalytic par-
tial oxidation (CPO) of methane with N2 dilution. At the same time, it can suppress the temperature
increase at the catalyst bed inlet during the CPO of methane without N2 dilution. The Rh–CeO2/MgO
(Ce/Rh = 4) has the ability to maintain a more reduced state in the CPO of methane, and this may be
related to the high ability to activate methane by the synergy of Rh metal surface and partially covering
Ce species. The catalyst bed of the CPO of methane consists of the initial oxidation zone and the subse-
quent steam reforming zone. The catalyst with higher resistance to the oxidation of Rh gives larger zone
for the steam reforming and it also enables the overlap of the exothermic oxidation zone with the endo-
thermic steam reforming zone.
Received 20 June 2009
Revised 18 August 2009
Accepted 19 August 2009
Available online 27 September 2009
Keywords:
Catalytic partial oxidation
Methane
Thermography
Hot spot
Ó 2009 Elsevier Inc. All rights reserved.
Rhodium
Ceria
1. Introduction
using the spatial analysis of gas composition and temperature pro-
files in the catalyst bed [19–21]. In the present study, the effect of
Conversion of natural gas to synthesis gas is important for the
production of methanol, dimethyl ether, and Fischer–Tropsch liq-
uids, and a more compact and energy-efficient process for syngas
production than conventional steam reforming is needed [1,2].
Much attention has recently been paid to the catalytic partial oxi-
dation (CPO) of methane, where high methane conversion and syn-
gas yield have been obtained at millisecond contact times under
autothermal conditions [3]. Therefore, CPO makes the reactors
simple and compact and it also enables fast responses and low heat
capacity. It has been confirmed that noble-metal-based catalysts
are highly active for the CPO of methane, and Rh is one of the suit-
able components [3–7]. It has been known that rhodium is applied
to various catalytic reactions other than partial oxidation of meth-
ane, and modification of Rh catalysts with CeO2 promoted many
reactions such as deNOx [8,9], CO oxidation [10], reforming of
hydrocarbons [11,12], gasification of biomass [13–15], catalytic
partial oxidation and autothermal reforming of ethanol [16,17],
and catalytic wet air oxidation of gasoline oxygenates [18]. In the
CPO of methane, Schmidt and co-workers have reported that the
effect of the washcoat and Ce addition was investigated over
CeO2 addition to Rh/MgO is investigated in the CPO of methane at
short contact time. In the case of the practical application of the
CPO of methane, the reaction pressure will be higher such as
1 MPa, although the CPO performance has been evaluated under
atmospheric pressure or below. Under pressurized conditions,
the suppression of carbon deposition will be more important.
MgO is known to be an effective support to the suppression of car-
bon deposition in the production of synthesis gas by the reforming
reactions [22,23]. The addition of CeO2 to MgO decreased the
amount of carbon deposition in the CPO of methane under pressur-
ized condition [24]. In addition, it has been reported that the MgO
support suppressed the sintering of Rh metal particles [25–27].
In this article, we investigated the dependence of the added
amount of CeO2 over 1 wt% Rh/MgO on the CPO of methane with
or without dilution. In particular, the effect of CeO2 addition on the
catalyst structure and the reaction zones of CPO are discussed on
the basis of the catalyst bed temperature measured by the infrared
thermographical observation and catalyst characterization results.
2. Experimental
Rh + CeO2-based catalysts supported on
a-Al2O3 foam monoliths
2.1. Catalyst preparation
* Corresponding author. Address: Graduate School of Pure and Applied Sciences,
University of Tsukuba, 1-1-1, Tennodai, Tsukuba, Ibaraki 305-8573, Japan. Fax: +81
29 853 5030.
A Rh/MgO catalyst was prepared by impregnating MgO with an
E-mail address: tomi@tulip.sannet.ne.jp (K. Tomishige).
aqueous solution of RhCl3ꢀ3H2O (Soekawa Chemicals). The MgO
0021-9517/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.08.008

2
H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
support was prepared by calcining MgO (UBE Material Industries
Ltd., Japan) at 1423 K for 3 h. After the impregnation, the solvent
was evaporated at 353 K and the sample was dried at 383 K for
12 h and then calcined at 773 K in air for 3 h. Rh–CeO2/MgO cata-
lysts were prepared by impregnating MgO with the mixed aqueous
solution of RhCl3ꢀ3H2O and Ce(NH4)2(NO3)6ꢀH2O (Wako Pure
Chemical Industries) in the co-impregnation method. After the
co-impregnation, the solvent was evaporated at 353 K. The sample
was dried at 383 K for 12 h and then calcined at 773 K for 3 h. The
loading amount of Rh on Rh/MgO and Rh–CeO2/MgO was 1.0 wt%.
The loading amounts of CeO2 on Rh–CeO2/MgO were represented
as the molar ratio of Ce to Rh (Ce/Rh) in parentheses as Rh–CeO2/
MgO (Ce/Rh = 4). The weight content of the CeO2 is listed out in Ta-
ble 1. As a reference, CeO2/MgO was also prepared by the impreg-
nation method. The loading amount of CeO2 was adjusted to the
corresponding Rh–CeO2/MgO. Catalysts in powder form were
pressed, then crushed and sieved into granules of 0.13–0.18 mm.
The catalysts were observed using a transmission electron
microscope (TEM) (JEOL JEM-2010) at an accelerating voltage of
200 kV. Catalysts after the H2 reduction pretreatment at 1123 K
for 0.5 h and after the activity test were dispersed in ethanol using
supersonic waves. Then they were put on Cu grids for TEM obser-
vation under air atmosphere. Average particle size in the form of a
P
P
surface mean diameter (ds) is calculated by ds ¼ n1d3= n1d2
1
1
(ni, number of pieces; di, particle size) [29–31].
Rh K-edge EXAFS were measured at BL01B1 station in SPring-8
with support from the Japan Synchrotron Radiation Research Insti-
tute (JASRI) (Proposal Nos. 2006A1058, 2007A1156), and the mea-
surement method and the beam conditions were the same as those
reported previously [28]. The catalyst powder (750 mg) was trea-
ted using H2 at 1123 K for 0.5 h in a fixed-bed reactor and the sam-
ple was pressed into a self-supporting 7-mm-diameter wafer
under atmosphere, followed by treatment with H2 at 773 K for
0.5 h in a cell. After this pretreatment, the sample wafer was trans-
ferred to the measurement cell using a glove box filled with nitro-
gen to prevent exposure of the sample disk to air. The analysis
method for the EXAFS spectra was carried out in the same way
as that in the previous report [28].
2.2. Characterization of catalysts
Temperature-programed reduction (TPR) profiles were mea-
sured in a fixed-bed quartz reactor, and the procedures were the
same as those mentioned in the previous report [28]. Before TPR
measurement, the catalysts were treated in O2 at 773 K for 0.5 h
and then in Ar at 773 K for 0.5 h to remove adsorbed species such
as CO2. The sample weight was 50 mg, and the heating rate was
10 K minꢁ1 from room temperature to 1123 K, and 5% H2 diluted
in Ar (30 cm3 minꢁ1) was used. Consumption of hydrogen was esti-
mated from the integrated peak area of the profiles.
Measurement of H2 chemisorption was carried out in a high-
vacuum system using a volumetric method, and the procedures
were the same as those mentioned in the previous report [28]. Be-
fore the adsorption of H2, the catalysts were treated in H2 at 1123 K
for 0.5 h in a fixed-bed reactor. After this pretreatment, the sample
was transferred to a cell for adsorption measurements under air
atmosphere. Before each measurement, H2 pretreatment at 773 K
was carried out for 0.5 h in the cell. After evacuation at 773 K,
the sample was cooled to room temperature. The total amount of
H2 adsorption was measured at room temperature with a H2 pres-
sure at adsorption equilibrium of about 1.0 kPa. The dead volume
of the apparatus was 63.5 cm3 and the sample weight was 150 mg.
2.3. Partial oxidation of methane with or without N2 dilution
Activity tests for the catalytic partial oxidation (CPO) of meth-
ane with N2 dilution were carried out using a tubular fixed bed
flow reactor made of quartz. The reactor setup was the same as
that in the previous report [28]. The dilution of the reactant gases
with N2 (CH4/O2/N2 = 4/2/94) was applied to suppress the temper-
ature increase of the catalyst bed due to exothermic reactions. The
catalyst weight was 10 mg, and the height of the catalyst bed was
about 1.4 mm. Catalysts were reduced with hydrogen at 1123 K for
0.5 h before each activity test. The temperature on the reactor was
monitored with a thermocouple at the bottom of the catalyst bed,
and it was denoted as TTC. The total flow rate of the gases was
1500 cm3 minꢁ1, and the contact time was as short as 0.4 ms.
The temperature profile of the catalyst bed during the CPO of
methane with the N2 dilution was measured by using the infrared
thermography, and the profile is shown in Fig. S1. The bed temper-
ature was only 20 K higher than the TTC, indicating that the effect of
the temperature increase in the CPO of methane with N2 dilution is
Table 1
Properties of catalysts and results of activity test in partial oxidation of methane with N2 dilution.
Catalyst
Loading of
CeO2 (wt%)
BET surface
area (m2 gꢁ1
H2 adsorption
Rh metal particle Dispersion
Partial oxidation of methanec
size (nm)b
a
ꢁ1
)
ðl
mol g
Þ
H2
TEMd CH4
H2
CO
TOF
(sꢁ1
cat
e
adsorption
(H/Rh)
conversion
(%)
selectivity selectivity
)
(%)
(%)
Rh/MgO
–
6.4
6.9
9.8
11.2
6.0
5.6 0.3
4.1 0.3
0.24
0.13
0.20
0.27
69
81
83
93
81
90
125
274
Rh–CeO2/MgO
(Ce/Rh = 4)
Rh–CeO2/MgO
(Ce/Rh = 8)
Rh–CeO2/MgO
(Ce/Rh = 12)
Rh–CeO2/MgO
(Ce/Rh = 18.7)
6.4 wt% CeO2/
MgOf
12.8
19.2
30.0
6.4
11.5
12.8
15.0
10.2
14.6
4.8
4.0
4.0
–
–
0.10
0.09
0.09
–
–
72
70
64
0.3
0.5
88
87
87
28
36
84
84
80
12
9
303
352
323
–
–
–
3.2 0.3
0.34
–
–
–
30 wt% CeO2/
30.0
–
–
–
–
MgOf
a
b
c
d
e
f
H2 adsorption at 298 K.
Rh particle size (surface-averaged) is calculated from TEM images.
Reaction conditions: CH4/O2/N2 = 4/2/94, total flow rate 1500 cm3 minꢁ1; TTC = 973 K; total pressure 0.1 MPa; catalyst weight 10 mg; contact time 0.4 ms.
Dispersion from TEM is calculated from (1.098/(Rh particle size [nm])) [34].
Turnover frequency (TOF) in partial oxidation of methane is calculated on the basis of methane conversion rate and the amount of H2 adsorption at 298 K.
Loading amounts of CeO2 on 6.4 and 30 wt% CeO2/MgO were equal to those on Rh–CeO2/MgO (Ce/Rh = 4 and 18.7), respectively.

H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
3
much smaller than that in without N2 dilution, and this can be re-
garded as isothermal conditions.
spectrometer (QMS, QMA200; Pfeiffer Vacuum Technology AG).
Stable catalytic performance in CH4 conversion, H2 selectivity,
and CO selectivity during 10 pulses of the CPO of methane suggests
that the result of the pulse experiment can be reflected by that of
the steady-state condition.
Reaction tests for partial oxidation of methane (CH4/O2 = 2/1)
without N2 dilution were also carried out using a fixed-bed quartz
reactor. The reactor setup was the same as that reported previously
[28]. In the case of the CPO of methane without the dilution, the
catalyst temperature during the reaction was measured by the
infrared thermography (TH31; NEC San-ei Instruments Ltd.), as re-
ported in the previous reports [28,32,33]. The apparatus detected
2.5. Ability of the catalysts to activate methane with or without
preadsorbed oxygen
the infrared with 3.2 lm wavelength and the temperature resolu-
To evaluate the ability of the catalysts to activate methane, the
reactivity of CH4 was measured by the CH4 pulse reaction. In addi-
tion, the reactivity was also compared between reduced and oxy-
gen-pretreated catalysts because the catalyst bed in the CPO of
methane has two zones: oxidation zone and the reforming zone
[19]. Catalysts without preadsorbed oxygen were reduced catalysts
with hydrogen at 1123 K for 0.5 h. On the other hand, the O2 pulse
tion was 5 K, and the spatial resolution was 0.8 mm. The calibra-
tion was done on the basis of the catalyst bed temperature with the
flow of an inert gas (N2) and heating only by the furnace, whose
temperature was controlled by the thermocouple. The calibration
includes the emission coefficient of the catalysts. The temperature
profiles of catalyst bed were measured from
a window
(15 mm ꢂ 15 mm) on furnace. The furnace temperature was con-
trolled at 723 K. The total flow rate was 300 cm3 minꢁ1, the cata-
lyst weight was 10 mg, the height of the catalyst bed was 3 mm,
and the contact time was 2.0 ms.
(0.45 lmol) was supplied at 373 K to the reduced catalysts repeat-
edly and the amount of adsorbed and absorbed oxygen was ad-
justed on the basis of the QMS analysis during the O2 pulse
feeding. The O2 treatment temperature was set to as low as
373 K because O2 molecule can be trapped by the absorption into
Rh near the catalyst bed inlet at high temperature. After purging
In both reaction tests of the CPO with and without N2 dilution,
the catalysts were reduced with hydrogen at 1123 K for 0.5 h be-
fore each activity test. The contact time in the experimental result
is represented on the basis of the density of Rh–CeO2/MgO (Ce/
Rh = 4), since a density of catalysts increased 20% monotonously
from Rh/MgO to Rh–CeO2/MgO (Ce/Rh = 18.7), and the contact
time is also changed when the catalyst weight is fixed. The effluent
gas was analyzed using an FID-GC equipped with a methanator for
CO, CH4, and CO2, a TCD-GC (carrier gas: Ar) for H2 and a TCD-GC
(carrier gas: He) for O2. The on-line sampling of the six-way valves
was used. Methane conversion and CO and H2 selectivities are cal-
culated as
the reactor with He, the CH4 pulse (1.2
at 100 cm3 minꢁ1) at 773 K was supplied to 0.1 g catalyst contain-
ing 9.3 mol Rh. The temperature of the methane pulse reaction
lmol with He carrier gas
l
was lower than the CPO experiments to keep the amount of the
converted methane smaller than that of the surface Rh. At this
temperature, the methane conversion can reflect the probability
of methane activation of surface Rh. At high temperature, catalytic
methane decomposition to solid carbon and H2 can proceed and
methane conversion is too high to evaluate the probability of
methane activity. In this experiment, only hydrogen was observed
as a product, and the formation of CO and CO2 was below the
detection limit of the QMS analysis.
CH4 conversion ð%Þ ¼ ðCCO þ CCO Þ=ðCCH þ CCO þ CCO Þ ꢂ 100;
2
4
2
CO selectivity ð%Þ ¼ CCO=ðCCO þ CCO Þ ꢂ 100;
2
H2 selectivity ð%Þ ¼ CH =ððCCO þ CCO Þ ꢂ 2Þ ꢂ 100;
2
2
3. Results and discussion
where C is the concentration of each component in the effluent
gases. Throughout the experiments, no carbon-containing product
other than CO and CO2 was formed. Amount of deposited carbon
was negligible based on the thermogravimetric analysis (TGA) of
the catalysts after the reaction. An example of the TGA profile is
shown in Fig. S2, and no weight decrease with an exothermic
behavior due to the carbon combustion was observed at all.
3.1. Characterization of Rh–CeO2/MgO catalysts
Fig. 1 shows TEM images of Rh/MgO, Rh–CeO2/MgO (Ce/Rh = 4
and 18.7), and CeO2/MgO catalysts after the H2 reduction pretreat-
ment. Small spherical dark spots were observed on three Rh-con-
taining catalysts (Fig. 1a–c), and they are not observed on CeO2/
MgO (Fig. 1d), therefore, these spots are assigned to Rh metal par-
ticles. The average Rh particle sizes (surface-averaged) were deter-
mined from the TEM observation and the dispersions calculated
from them are summarized in Table 1 [34]. The Rh dispersions esti-
mated by H2 chemisorption are also shown in Table 1. In the case
of Rh–CeO2/MgO catalysts, it is found that the dispersion (H/Rh)
obtained from H2 adsorption is much smaller than that estimated
by TEM result, and it is suggested that the surface of Rh metal par-
ticles can be partially covered with Ce species. This kind of struc-
ture agreed well with the previous reports on Rh + CeO2-based
catalysts [35–37]. It should be noted that Rh metal particle size de-
creased gradually with increasing CeO2 amount.
The structure of Rh was characterized by Rh K-edge EXAFS. The
EXAFS spectra are shown in Fig. S3. Curve fitting results of Rh K-
edge EXAFS after H2 reduction are presented in Table 2, where a
good fitting was attained using only the Rh–Rh bond. Intensities
of the peak due to the Rh–Rh bond in the Fourier transform spectra
(Fig. S3b) decreased gradually with increasing CeO2 amount. At the
same time, the coordination number (CN) of the Rh–Rh bond de-
creased with increasing CeO2 amount. Even in the case of Rh–
CeO2/MgO (Ce/Rh = 18.7) with smaller Rh particles, other bonds
like Rh–O than Rh–Rh cannot be detected. The dependence of the
2.4. Titration of adsorbed oxygen during the partial oxidation of
methane
To examine the oxidation degree of catalysts during partial oxi-
dation of methane, titration of adsorbed/absorbed oxygen with H2
pulses was carried out after the pulse reaction of partial oxidation
of methane. The reactor setup was the same as that for the CPO
without the dilution. Pulse gases were supplied to the reactor using
six-way valves and He carrier gas. The catalyst amount was 10 mg,
which contained 0.93
0.5 h, 10 pulses of CH4 + O2 were introduced at 873 K. The single
pulse for the partial oxidation of methane contained 4.0 mol
CH4 + 2.0
mol O2 with He carrier gas at 300 cm3 minꢁ1, and the
lmol Rh. After H2 pretreatment at 1123 K for
l
l
amount was much larger than total Rh amount and the number
of surface Rh atoms, indicating that the large turnover number
can be obtained. After the pulse reaction, the reactor was purged
with He rapidly, and then the H2 pulse (0.46 lmol) was supplied
repeatedly at 1123 K, which was the same temperature as that in
the reduction pretreatment. The total consumption of H2 pulses
corresponds to the oxidation during the pulse CPO of methane.
Reactants and products were analyzed using a quadrupole mass

4
H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
Fig. 1. TEM images of Rh/MgO (a), Rh–CeO2/MgO (Ce/Rh = 4 (b) and 18.7 (c)), and 6.4 wt% CeO2/MgO (d) catalysts after H2 reduction at 1123 K.
Ce/Rh ratio on the CN of the Rh–Rh bond can be explained by the
decrease of Rh metal particle size caused by CeO2 addition, and this
behavior is supported by TEM results.
3.2. Catalytic performance in partial oxidation of methane with N2
dilution
Fig. 2 shows TPR profiles of the catalysts with H2, and Table 3
lists out H2 consumption data. Reduction of Rh species on MgO
proceeds mainly at 400–800 K [25–27]. According to the previous
reports, H2 consumption below 673 K was assigned to the reduc-
tion of Rh2O3, and H2 consumption above 673 K was assigned to
the reduction of MgRh2O4 [25–27]. As listed out in Table 3, the
total amount of H2 consumption on Rh/MgO was explained by
the stoichiometry of the reduction of Rh3+ to Rh0 in Rh2O3 and
MgRh2O4 (Rh3+ + 3/2H2 ? 2Rh0 + 3H+), indicating that almost all
the Rh species are present in a metallic state after the reduction.
On CeO2/MgO, the CeO2 species on MgO were reduced at around
860 K. For Rh–CeO2/MgO, a TPR peak at 460 K grew with increas-
ing CeO2 amount, where the H2 was consumed by the reduction
of both Rh2O3 and CeO2 because the H2 consumption in the tem-
perature range of r.t. to 600 K was larger than 1.5 times of total
Rh mol. In addition, the amount of the reduced CeO2 on Rh–
CeO2/MgO (Ce/Rh = 8) was larger than that on CeO2/MgO with
the same CeO2 loading amount. It should be noted that the
amount of the reduced Ce species increased with increasing
CeO2 amount, but the ratio of reduced Ce species to the total
Ce species decreased with CeO2 loading. From these results it
can be observed that the presence of Rh can promote the reduc-
tion of CeO2 within the region where spilt-over hydrogen can
reach [38,39]. In particular, the increase of the peak intensity be-
low 600 K with increasing CeO2 amount indicates the strong
interaction between Rh and CeO2, and this interaction is con-
nected to the formation of Rh metal particles covered with
CeO2 suggested by the results given in Table 1. Decoration of me-
tal particles with reduced oxide species has been known as SMSI
(strong metal–support interaction) and SMOI (strong metal–oxide
interaction) phenomena [40,41].
Table 1 shows how the amount of CeO2 affects methane conver-
sion and selectivity over Rh–CeO2/MgO and CeO2/MgO. The con-
version and selectivity increased by the addition of CeO2 at Ce/
Rh = 4, but they decreased with increasing CeO2 amount in the
range of Ce/Rh = 4–18.7. As a result, the optimum amount of
CeO2 over Rh/MgO is determined to be Ce/Rh = 4. As a reference,
we prepared 6.4 and 30 wt% CeO2 supported on MgO, whose load-
ing amounts corresponded to that supported on Rh–CeO2/MgO
(Ce/Rh = 4 and 18.7), and it is found that the activity of CeO2/
MgO was very low, and the increase of conversion and selectivity
on Rh–CeO2/MgO (Ce/Rh = 4) cannot be explained by the contribu-
tion of CeO2 itself. In addition, it should be noted that the adsorp-
tion amount of H2 decreased gradually with increasing CeO2
amount, and this tendency does not agree with the behavior of
methane conversion. Reasons for high methane conversion on
Rh–CeO2/MgO (Ce/Rh = 4) are discussed later.
3.3. Catalytic performance in partial oxidation of methane without N2
dilution
Without the dilution, the catalyst bed temperature is strongly
influenced by the catalytic performance, regarding the CPO of
methane [28]. Fig. 3 shows the results of thermographical observa-
tion during the CPO of methane without N2 dilution, and the re-
sults of the test are summarized in Table 4. The highest bed
temperature on Rh–CeO2/MgO (Ce/Rh = 4) was lower than those
on the other catalysts tested, and the bed temperature increased
with increasing CeO2 amounts on the Rh–CeO2/MgO in the range
of Ce/Rh > 4. This tendency in the bed temperature was opposite
to the catalytic activity in the CPO of methane with N2 dilution
(Table 1). Generally speaking, it is not easy to distinguish the

H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
5
Table 2
Curve fitting results of Rh K-edge EXAFS of the reduced catalysts.
CNa
Rb (10ꢁ1 nm)
rc (10ꢁ1 nm)
DE0 (eV)
Rfe (%)
d
Catalyst
Shells
Rh/MgO
Rh–Rh
Rh–Rh
Rh–Rh
Rh–Rh
Rh–Rh
10.8 0.8
9.6 1.0
9.0 0.8
8.3 1.0
7.6 0.9
2.68 0.01
2.67 0.01
2.67 0.01
2.67 0.01
2.66 0.01
0.068 0.002
0.069 0.003
0.069 0.003
0.069 0.003
0.070 0.004
0.6 1.9
3.1 2.2
1.9 2.0
ꢁ1.8 1.6
0.5 2.6
0.26
0.29
0.27
0.20
0.47
Rh–CeO2/MgO (Ce/Rh = 4)
Rh–CeO2/MgO (Ce/Rh = 8)
Rh–CeO2/MgO (Ce/Rh = 12)
Rh–CeO2/MgO (Ce/Rh = 18.7)
Rh foil
Rh–Rh
12.0
2.68
0.060
0
a
b
c
Coordination number.
Bond distance.
Debye–Waller factor.
d
e
Difference in the origin of photoelectron energy between the reference and the sample.
Residual factor. Fourier transform range: 30–160 nm, Fourier filtering range: 0.153–0.273 nm.
0.2
1323
1223
1123
1023
923
12.8 wt%CeO2/MgO
Rh-CeO2/MgO
(Ce/Rh=18.7)
Rh-CeO2/MgO
(Ce/Rh=8)
Rh-CeO2/MgO
Ce/Rh = 18.7
Gas feeding
Rh/MgO
12
8
4
0
Rh-CeO2/MgO
(Ce/Rh=4)
823
1123 K for 0.5 h
1100
300
500
700
900
Temperature / K
723
Catalyst bed
(length 3.0 mm)
Fig. 2. TPR profiles of Rh–CeO2/MgO and CeO2/MgO.
623
Fig. 3. Result of thermographical observation during the partial oxidation of
methane. Reaction conditions: CH4/O2 = 2/1, total flow rate 300 cm3 minꢁ1; total
pressure 0.1 MPa; catalyst weight 10 mg; and contact time 2.0 ms. Furnace
temperature 723 K.
different reaction zones, such as reforming, partial oxidation, and
combustion, only from the thermography as quasi-spatially re-
solved technique, and it is necessary to analyze the gas composi-
tion at each position in the catalyst bed [5,42,43]. Here, the effect
of CeO2 is attempted to be discussed on the basis of the tempera-
ture profiles of the catalyst bed. As reported previously, it is
thought that the CPO of methane at millisecond contact times con-
sists of an initial oxidation zone in which predominantly CO and
methane is highly endothermic reaction, the catalyst bed temper-
ature can be decreased in the steam reforming zone. The initial oxi-
dation zone is located near the catalyst bed inlet and this can
explain the temperature increase in Fig. 3. The highest temperature
is influenced by the selectivity of oxidation to CO or CO2 and cata-
lytic activity of the oxidation reaction.
Another important point is that the temperature gradients can
decrease due to the overlap of the oxidation and the reforming
zones [19,46–48]. In order to discuss the contributions of the oxi-
dation, the reforming of methane, and the overlap, we have to get
the information of the degree of oxidation of catalyst surface in the
H2O are formed (CH4 + 1.5O2 ? CO + 2H2O,
(1)), followed by steam reforming zone (CH4 + H2O ? CO + 3H2,
H = 210 kJ/mol, (2)) [44,45]. In the initial oxidation zone, the total
oxidation (combustion) of methane (CH4 + 2O2 ? CO2 + 2H2O,
D
H = ꢁ519 kJ/mol,
D
D
H = ꢁ802.3 kJ/mol, (3)) can also proceed as a side reaction. When
the oxidation of methane with O2 such as (1) and (3) proceeds, the
catalyst bed temperature was increased by these highly exother-
mic reactions. On the other hand, since the steam reforming of
Table 3
Results of temperature-programed reduction.
a
ꢁ1
ꢁ1
ꢁ1
Catalyst
Ce/Rh
Loading amount ð
l
mol g
CeO2
–
Þ
H2 consumption ð
r.t. to 600 K
37
l
mol g
Þ
Amount of reduced CeO2
ð
l
mol g
Þ
cat
cat
cat
Rh
600–1123 K
r.t. to 1123 K
–
Rh/MgO
0
93
98
Rh–CeO2/MgO
4
8
12
18.7
93
93
93
93
373
746
1119
1744
150
201
218
265
150
158
165
243
320
438
488
735
12.8 wt% CeO2/MgOb
–
–
746
11
163
347
a
Assuming that all the Rh species are reduced (Rh3+ ? Rh0) and stoichiometry of the CeO2 reduction (2CeO2 + H2 ? Ce2O3 + H2O).
Loading amount of CeO2 on 12.8 wt% CeO2/MgO was equal to that on Rh–CeO2/MgO (Ce/Rh = 8).
b

6
H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
Table 4
Catalyst performance and bed temperature in partial oxidation of methane without N2 dilution.
Catalyst
Partial oxidation of methanea
CH4 conversion (%)
Temperature (K)
Highest bed
H2 selectivity (%)
CO selectivity (%)
Outlet
Rh/MgO
78
77
76
76
71
93
93
87
87
87
91
90
89
88
85
1109
1052
1122
1135
1201
5
5
5
5
5
941
928
922
923
994
5
Rh–CeO2/MgO (Ce/Rh = 4)
Rh–CeO2/MgO (Ce/Rh = 8)
Rh–CeO2/MgO (Ce/Rh = 12)
Rh–CeO2/MgO (Ce/Rh = 18.7)
5
5
5
5
a
Reaction conditions: CH4/O2 = 2/1, total flow rate 300 cm3 minꢁ1; total pressure 0.1 MPa; catalyst weight 10 mg; contact time 2.0 ms.
catalyst bed during the CPO. In addition, in the CPO of methane
without N2 dilution, methane conversion was in the range of 76–
78% for all catalysts tested except for Rh–CeO2/MgO (Ce/
Rh = 18.7) (71%), and it seems that the conversion was not influ-
enced by the catalysts. This suggests that the effluent gas compo-
sition of the CPO without N2 dilution almost reaches the
equilibrium of steam reforming and water gas shift reactions with-
in the experimental errors [19]. Fig. S4 shows the TEM image of
Rh–CeO2/MgO (Ce/Rh = 4) after the CPO of methane for 3 h. The
catalytic performance was stable for 3 h and the TEM image indi-
cates that the size of Rh metal particles was not changed
(ds = 3.8 0.2 nm), and no deposited carbon such as whisker car-
bon was observed.
to H2 molecule and the desorption of H2 and CO are very favorable.
Adsorbed oxygen atom can give considerable coverage since the
dissociative adsorption of O2 molecule is a fast elementary step.
Therefore, the coverage of the reaction intermediates except for
oxygen atom is so small that it may be possible to quench the sur-
face state during the steady-state and pulse reactions. The order of
the H2 consumption amount was as follows: Rh/MgO > Rh–CeO2/
MgO (Ce/Rh = 18.7) > Rh–CeO2/MgO (Ce/Rh = 4), and this also
means the order of oxidation degree during the CPO of methane
(Table 5). An important point is that the CeO2 addition at the opti-
mum amount can maintain the reduced state of catalysts during
the reaction. The surface state of the catalysts can be strongly influ-
enced by methane activation ability in the CPO of methane [28].
In addition, the H2 adsorption on the Rh–CeO2/MgO (Ce/Rh = 4)
after the reaction was measured by using the present pulse tech-
nique, where the sample temperature was decreased to room tem-
3.4. Oxidation degree during the CPO of methane
perature. The amount of H2 consumption at the room temperature
Table 5 lists out the results of the CH4 + O2 pulse reaction and
subsequent titration of surface oxygen species. In the CH4 + O2
pulse reaction, the amount of CH4 and O2 was much larger than
the adsorption amount of H2 (number of surface atoms) as listed
out in Table 1. The pulse of CH4 + O2 was supplied 10 times and
the activity was not changed from the initial, and Table 1 lists
out the stable activity. The order of methane conversion was as fol-
lows: Rh–CeO2/MgO (Ce/Rh = 4) > Rh/MgO > Rh–CeO2/MgO (Ce/
Rh = 18.7). This activity order is similar to that of the CPO of meth-
ane with N2 dilution as shown in Table 1. After the pulse reactions,
we measured the amount of adsorbed and absorbed oxygen by the
reaction of H2 (O(a) + H2 ? H2O) at 1123 K. Generally, it is difficult
to quench the catalyst under the steady-state conditions at high
temperature without change in the surface state of the catalyst
during ignition/extinction in the CPO of methane. Therefore,
although the present results on the estimation of the degree of oxi-
dation are qualitative, they are thought to be sufficiently sugges-
tive. Methane-derived carbon-containing surface species such as
CHx and surface carbide have very low coverage due to much high-
er reactivity than that of methane. At the same time, the coverage
of adsorbed hydrogen atom and CO can be very low due to high
reaction temperature where the recombination of hydrogen atoms
ꢁ1
was 4:6
l
mol g and this was close to the H2 adsorption of the
cat
fresh Rh–CeO2/MgO (Ce/Rh = 4) (Table 1). This suggests no drastic
structural and morphological change during the reaction.
3.5. Ability to activate CH4 of the catalysts
High methane activation ability is needed in order to maintain
the reduced state even in the initial oxidation zone. Therefore,
the ability to activate methane of the catalyst with the reduced
and partially oxidized surface was evaluated using the pulse tech-
nique. Table 6 lists out the result of CH4 pulse reaction on reduced
and O2-preadsorbed catalysts. It should be noted that the amount
of methane feeding was less than or comparable to the number
of Rh surface atom determined by the H2 adsorption. On the re-
duced catalysts, the order of the methane reactivity was as follows:
Rh–MgO > Rh–CeO2/MgO (Ce/Rh = 4) > Rh–CeO2/MgO (Ce/Rh =
18.7). On the other hand, since CeO2/MgO has no reactivity of
CH4, the reactivity of methane normalized by surface Rh atoms is
considered. The probability of activation, which is calculated by
(the amount of reacted CH4)/(the amount of surface Rh atoms)
was estimated to be 0.36, 0.63, and 0.73 over Rh/MgO and Rh–
Table 5
H2 titration after CH4 + O2 reaction over Rh/MgO and Rh–CeO2/MgO (Ce/Rh = 4 and 18.7).
Catalyst
Number of surface Rh atoms
mol)a
Pulse CH4 + O2 reactionb (4.0/2.0
l
mol)
CO selectivity
H2 titrationc
(
l
CH4 conversion
(%)
H2 selectivity
(%)
H2 consumption
(lmol)
Rh-based oxidation degree
(%)d
(%)
Rh/MgO
Rh–CeO2/MgO (Ce/
Rh = 4)
0.22
0.12
78
85
74
77
69
71
0.48
0.05
34.4
3.6
Rh–CeO2/MgO (Ce/
Rh = 18.7)
0.08
54
35
63
0.25
17.9
a
b
c
H2 adsorption.
CH4/O2 reaction conditions: CH4/O2 = 4.0/2.0
H2 = 0.46
mol (100 cm3 minꢁ1 He carrier at 1123 K).
(H2 consumption [ mol])) ꢂ 100.
mol])/(1.5 ꢂ (Rh amount [
l lmol-Rh).
mol (300 cm3 minꢁ1 He carrier; 2.0 ms); TTC = 873 K; total pressure 0.1 MPa; catalyst weight 10 mg (0.93
l
d
l
l

H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
7
Table 6
Results of the pulse reaction of methane on reduced and O2-preadsorbed catalysts.
Catalyst
Amount (
l
mol)
CeO2
0
Amount of preadsorbed oxygen atom (
l
mol)
CH4 pulse reaction (
Reacted CH4
l
mol)a
Rh
Surface Rhb
2.2
Formed H2
Rh/MgO
9.3
0
4.6
0
4.8
0
4.8
0
0.79
0.81
0.76
0.84
0.58
0.67
0
0.71
0.20
0.61
0.65
0.51
0.48
0
Rh–CeO2/MgO (Ce/Rh = 4)
Rh–CeO2/MgO (Ce/Rh = 18.7)
6.4 wt% CeO2/MgOc
9.3
9.3
–
37
1.2
0.8
–
174
37
4.8
0
0
a
b
c
Reaction conditions: CH4 = 1.2
H2 adsorption.
Loading amount of CeO2 on 6.4 wt% CeO2/MgO was equal to that on Rh–CeO2/MgO (Ce/Rh = 4).
l
mol (100 cm3 minꢁ1 He carrier); TTC = 773 K; total pressure 0.1 MPa; catalyst weight 100 mg (9.3
lmol-Rh).
CeO2/MgO (Ce/Rh = 4 and 18.7). The results indicate that the addi-
tion of CeO2 enhanced the probability of methane activation. In the
reaction, hydrogen was formed as a product and the selectivity to
hydrogen was similar among the catalysts. Carbon-containing
products such as CO and CO2 were not observed at all, and this rep-
resents that carbon-containing surface species such as CHx and sur-
face carbide can be present on the catalyst surface. In this case, the
amount of carbon-containing species was smaller than the number
of the surface Rh atoms, and the formation of solid carbon such as
whisker carbon is not expected. When the catalysts were treated
with oxygen pulses up to O/Rh = 0.5, the reactivity of CH4 on Rh/
MgO was almost unchanged, but the H2 formation was strongly
suppressed. The Rh/MgO with preadsorbed oxygen can also acti-
vate methane to give H2O, which cannot be detected by our exper-
imental system. In contrast, on Rh–CeO2/MgO (Ce/Rh = 4 and 18.7),
the reactivity of methane increased slightly by preadsorption of
oxygen and the amount of H2 formation was almost maintained.
This behavior suggests that the preadsorbed oxygen is adsorbed
on reduced Ce species, and metallic Rh surface is maintained. At
the same time, partial dewetting of Ce species by the oxidation
[35] can increase the number of surface Rh metal atom; this can
enhance the amount of reacted methane and formed H2. The role
of added CeO2 on Rh–CeO2/MgO has two aspects. One is to prevent
the oxidation of Rh with O2 as mentioned above. The other is the
enhancement of methane activation rate per the Rh surface site.
Based on the characterization results, the surface of Rh metal par-
ticles is partially covered with the reduced CeO2 species, indicating
that Rh–CeO2/MgO catalysts have large interface between Rh and
Ce species, and it is suggested that the role of the interface on
the methane activation can be due to oxide ions present in the
interface. It has been reported that the activation of methane on
metal surfaces can be promoted by the adsorbed oxygen atom to
give methoxy and hydroformyl species according to the previous
reports [49–51].
most unchanged [19,52]. Much different degree of oxidation
suggests that the length of oxidation zone may be changed. Further
investigation is necessary regarding the mass transfer limitation of
O2 in the present catalytic systems.
The steam reforming of methane is a main reaction at down-
stream, and the metallic surface has the catalytic activity of steam
reforming of methane. The activity of steam reforming of methane
was evaluated on the reduced Rh/MgO and Rh–CeO2/MgO (Ce/
Rh = 4), and the results are listed out in Table S1. The CH4 conver-
sion in methane steam reforming on Rh–CeO2/MgO (Ce/Rh = 4)
was a little lower than that on Rh/MgO. This behavior almost
agreed with the ability to activate methane of the catalysts without
preadsorbed oxygen (Table 6). On the other hand, high steam
reforming activity on the reduced Rh/MgO than that on the re-
duced Rh–CeO2/MgO (Ce/Rh = 4) cannot explain the results of the
CPO with or without N2 dilution.
In the case of the CPO with N2 dilution (Table 1), methane con-
version on Rh/MgO was lower than that on Rh–CeO2/MgO (Ce/
Rh = 4). This tendency is interpreted to be due to Rh/MgO with dee-
per oxidation, which can decrease the amount of reduced and ac-
tive surface for steam reforming. In contrast, the oxidation of Rh–
CeO2/MgO (Ce/Rh = 4) is suppressed and the amount of reduced
and active surface is maintained, although Rh–CeO2/MgO (Ce/
Rh = 4) was not so active for steam reforming as Rh/MgO in the
fully reduced state.
In the case of the CPO without N2 dilution, Rh–CeO2/MgO (Ce/
Rh = 4) showed lower catalyst bed temperature than Rh/MgO
(Fig. 3). The highest bed temperature can be determined by the
exothermic reaction in the oxidation zone and the effect of the
overlap with the endothermic steam reforming zone [19,46–48].
In the case of Rh–CeO2/MgO (Ce/Rh = 4), the formation of the oxi-
dized surface can be suppressed as expected from the results of the
CPO with N2 dilution (Tables 1 and 5). This enables larger overlap
of the oxidation reaction zone with the reforming zone since the
reforming reaction can proceed on the reduced metallic surface.
In addition, Rh metal particles partially covered with reduced Ce
species on the reduced Rh–CeO2/MgO are suggested by character-
ization results. In contrast, it has been reported that this decoration
of Rh particles with Ce species is reversible under O2-containing
atmosphere [35]. Therefore, the available Rh surface atoms on
Rh–CeO2/MgO under the CPO may be larger than those on the re-
duced Rh–CeO2/MgO. In fact, this phenomenon can explain the in-
crease of the reacted CH4 amount by O2 preadsorption on Rh–CeO2/
MgO (Ce/Rh = 4) in Table 6. In this experiment, the Rh-based oxida-
tion degree is calculated as 0.34. On the other hand, the oxidation
degree of Rh–CeO2/MgO (Ce/Rh = 4) determined in Table 5 is 0.036
and it was much smaller. Judging from the small oxidation degree
of Rh–CeO2/MgO (Ce/Rh = 4), the increase of the available Rh sur-
face atoms by dewetting of Ce species is not a main reason for
the effect of CeO2 addition in the CPO of methane.
3.6. Interpretation of temperature profiles in the catalytic partial
oxidation of methane
Based on the previous reports, the CPO of methane has two
reaction zones: the initial oxidation of methane to CO and CO2,
and the stream reforming of residual methane at downstream
[44,45]. In the initial oxidation zones, O2 is present in the gas
phase, and the catalysts tend to be oxidized. The obtained oxida-
tion degree of catalysts by H2 titration after the pulse CPO (Table
5) is associated with the initial oxidation zone. The results given
in Table 5 show that the degree of oxidation is strongly dependent
on the catalysts and that Rh/MgO was more deeply oxidized than
Rh–CeO2/MgO (Ce/Rh = 4). This behavior does not agree with the
previous results that the O2 consumption rate is typically con-
trolled by the mass transfer and the length of oxidation zone is al-

8
H. Tanaka et al. / Journal of Catalysis 268 (2009) 1–8
In the CPO of methane, water gas shift (WGS) reaction (CO +
H2O ? CO2 + H2,
H = ꢁ41 kJ/mol) proceeds as well as the oxida-
Appendix A. Supplementary data
D
tion of methane and the steam reforming of methane. Since the
heat due to the WGS is much smaller than those due to the meth-
ane oxidation and steam reforming, it is difficult to recognize the
reaction zone by our experimental setup. Based on the outlet tem-
perature and the composition of the effluent gas in the CPO of
methane without N2 dilution (Table 4), we calculate the equation
below [19]:
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2009.08.008.
References
[1] J.R. Rostrup-Nielsen, J. Sehested, J.K. Nørskov, Adv. Catal. 47 (2002) 65.
[2] J.R. Rostrup-Nielsen, Catal. Today 145 (2009) 72.
[3] D.A. Hickman, L.D. Schmidt, J. Catal. 138 (1992) 267.
[4] D.A. Hickman, L.D. Schmidt, Science 259 (1993) 343.
[5] R. Horn, K.A. Williams, N.J. Degenstein, A. Bitsch-Larsen, D. Dalle Nogare, S.A.
Tupy, L.D. Schmidt, J. Catal. 249 (2007) 380.
[6] C. Elmasides, T. Ioannides, X.E. Verykios, ALChE J. 46 (2000) 1260.
[7] B.C. Enger, R. Lødeng, A. Holmen, Appl. Catal. A 346 (2008) 1.
[8] H.C. Yao, Y.F. Yu Yao, J. Catal. 86 (1984) 254.
[9] N.R. Collins, M.V. Twigg, Top. Catal. 42–43 (2007) 323.
[10] A. Gayen, K.R. Priolkar, P.R. Sarode, V. Jayaram, M.S. Hedge, G.N. Subbanna, S.
Emura, Chem. Mater. 16 (2004) 2317.
[11] K. Nagaoka, K. Sato, S. Fukuda, S. Nakashiki, H. Nishiguchi, J.A. Lercher, Y.
Takita, Chem. Mater. 20 (2008) 4176.
[12] R. Wang, H. Xu, X. Liu, Q. Ge, W. Li, Appl. Catal. A 305 (2006) 204.
[13] M. Asadullah, S. Ito, K. Kunimori, M. Yamada, K. Tomishige, J. Catal. 208 (2002)
255.
[14] T. Miyazawa, K. Okumura, K. Kunimori, K. Tomishige, J. Phys. Chem. C 112
(2008) 2574.
[15] P.J. Dauenhauer, B.J. Dreyer, N.J. Degenstein, L.D. Schmidt, Angew. Chem., Int.
Ed. 46 (2007) 5864.
[16] J.R. Salge, G.A. Deluga, L.D. Schmidt, J. Catal. 235 (2005) 69.
[17] W. Cai, F. Wang, A.C. Van Veen, H. Provendir, C. Mirodatos, W. Shen, Catal.
Today 138 (2008) 152.
ꢀ
ꢁ
KEXP
KEQ
PH PCO
1
2
2
¼
ꢀ
PCO
P
KEQ;WGS
H2
O
WGS
Here, KEQ is the equilibrium constant calculated with the outlet
temperature, and Pi are the partial pressures of species i. The values
of Rh/MgO and Rh–CeO2/MgO (Ce/Rh = 4) are 2.4 and 2.8, respec-
tively. The difference is so small that it is difficult to discuss the ef-
fect of CeO2 on the equilibrium approach of the WGS reaction.
On the other hand, Rh–CeO2/MgO (Ce/Rh = 18.7) showed low
methane conversion in the CPO of methane with N2 dilution and
high bed temperature in the CPO without dilution. As shown in Ta-
ble 6, the ability to activate methane of Rh–CeO2/MgO (Ce/
Rh = 18.7) was inferior to Rh/MgO and Rh–CeO2/MgO (Ce/Rh = 4),
although the small increase of reacted methane amount that could
be interpreted by dewetting effect of Ce oxidation was observed.
Low ability to activate methane is related to high oxidation degree
in the CPO of methane as listed out in Table 5. Both low steam
reforming activity and high oxidation degree during the CPO bring
about the bed temperature increase in the CPO without N2 dilution
and the decrease of methane conversion in the CPO with N2 dilu-
tion on Rh–CeO2/MgO (Ce/Rh = 18.7).
[18] I. Cuautémoc, G.D. Angel, G. Torres, V. Bertin, Catal. Today 133–135 (2008) 588.
[19] A. Donazzi, B.C. Michael, L.D. Schmidt, J. Catal. 260 (2008) 270.
[20] B.C. Micael, A. Donazzi, L.D. Schmidt, J. Catal. 265 (2009) 117.
[21] A. Bitsch-Larsen, N.J. Degenestein, L.D. Schmidt, Appl. Catal. B 78 (2008) 364.
[22] K. Tomishige, Y.G. Chen, K. Fujimoto, J. Catal. 181 (1999) 91.
[23] Y.G. Chen, K. Tomishige, K. Yokoyama, K. Fujimoto, Appl. Catal. A 165 (1997)
335.
[24] K. Imagawa, A. Nagumo, R. Kanai, T. Minami, K. Tomishige, Y. Suehiro, Stud.
Surf. Sci. Catal. 167 (2007) 415.
4. Conclusions
[25] H.Y. Wang, E. Ruckenstein, J. Catal. 186 (1999) 181.
[26] E. Ruckenstein, H.Y. Wang, J. Catal. 190 (2000) 32.
[27] E. Ruckenstein, H.Y. Wang, Appl. Catal. A 198 (2000) 33.
[28] S. Naito, H. Tanaka, S. Kado, T. Miyao, S. Naito, K. Okumura, K. Kunimori, K.
Tomsihge, J. Catal. 259 (2008) 138.
[29] J.R. Anderson, Structure of Metallic Catalysts, Academic Press, New York, 1975.
p. 358.
[30] D.G. Mustard, C.H. Bartholomew, J. Catal. 67 (1981) 186.
[31] N. Seshu Babu, N. Lingaiah, R. Gopinath, P. Siva Sankar Reddy, P.S. Sai Prasad, J.
Phys. Chem. C 111 (2007) 6447.
[32] B. Li, S. Kado, Y. Mukainakano, T. Miyazawa, T. Miyao, S. Naito, K. Okumura, K.
Kunimori, K. Tomishige, J. Catal. 245 (2007) 144.
[33] Y. Mukainakano, B. Li, S. Kado, T. Miyazawa, K. Okumura, T. Miyao, S. Naito, K.
Kunimori, K. Tomishige, Appl. Catal. A 318 (2007) 252.
[34] J.R. Anderson, Structure of Metallic Catalysts, Academic Press, New York, 1975.
p. 289.
[35] S. Bernal, F.J. Bontana, J.J. Calvino, G.A. Cifredo, J.A. Pérez-Omil, J.M. Pintado,
Catal. Today 23 (1995) 219.
[36] C. Force, J.P. Belzunegui, J. Sanz, A. Martínez-Arias, J. Soria, J. Catal. 197 (2001)
192.
[37] J. Stubenrauch, J.M. Vohs, Catal. Lett. 47 (1997) 21.
[38] K. Tomishige, A. Okabe, K. Fujimoto, Appl. Catal. A 194–195 (2000) 383.
[39] R. Ueda, T. Kusakari, K. Tomishige, K. Fujimoto, J. Catal. 194 (2000) 14.
[40] G.L. Haller, D.E. Resasco, Adv. Catal., vol. 36, Academic Press, New York, 1989.
p. 173.
[41] S.J. Tauster, Acc. Chem. Res. 20 (1987) 389.
[42] R. Horn, N.J. Degenstein, K.A. Williams, L.D. Schmidt, Catal. Lett. 110 (2006)
169.
[43] R. Horn, K.A. Williams, N.J. Degenstein, L.D. Schmidt, J. Catal. 242 (2006) 92.
[44] T. Liu, C. Snyder, G. Veser, Ind. Eng. Chem. Res. 46 (2007) 9045.
[45] D.D. Nogare, N.J. Degenstein, R. Horn, P. Canu, L.D. Schmidt, J. Catal. 258 (2008)
131.
[46] K. Tomsihgie, S. Kanazawa, K. Suzuki, M. Asadullah, M. Sato, K. Ikushima, K.
Kunimori, Appl. Catal. A 233 (2002) 35.
[47] B. Li, K. Maruyama, M. Nurunnabi, K. Kunimori, K. Tomishgie, Appl. Catal. A
275 (2004) 157.
[48] B. Li, S. Kado, Y. Mukainakano, M. Nurunnabi, T. Miyao, S. Naito, K. Kunimori, K.
Tomishige, Appl. Catal. A 304 (2006) 62.
The additive amount of CeO2 over Rh–CeO2/MgO was opti-
mized at Ce/Rh = 4 in terms of highest methane conversion in
the catalytic partial oxidation (CPO) of methane with N2 dilution
and lowest catalyst bed temperature in the CPO of methane
without N2 dilution. Characterization results indicated that the
addition of CeO2 to Rh/MgO decreased Rh metal particle size
monotonously with increasing CeO2 amount, and that Rh metal
particles were partially covered with the Ce species on Rh–
CeO2/MgO catalysts. As a result of the pulse CPO of methane,
the optimized Rh–CeO2/MgO (Ce/Rh = 4) exhibited much lower
oxidation degree of the catalysts than Rh/MgO and Rh–CeO2/
MgO (Ce/Rh = 18.7). High resistance to oxidation of Rh–CeO2/
MgO (Ce/Rh = 4) can be caused by high ability to activate meth-
ane, on the reduced catalyst and even on the catalyst with
preadsorbed oxygen. The catalyst bed of the CPO of methane
mainly consists of the oxidation and the steam reforming zones.
High resistance to oxidation of Rh caused by the presence of
CeO2 is thought to decrease the ratio of the oxidized catalysts
and to increase simultaneously the ratio of the catalysts with
the reduced surface. High methane conversion in the CPO of
methane over Rh–CeO2/MgO (Ce/Rh = 4) with N2 dilution can
be due to higher ratio of the reduced catalyst. At the same time,
the decrease in the catalyst bed temperature in the CPO of meth-
ane without N2 dilution on the Rh–CeO2/MgO (Ce/Rh = 4) with
higher resistance to the oxidation of Rh can be due to the pro-
motion of the overlap between the exothermic oxidation reac-
tion zone and the endothermic reforming zone.
[49] T. Sasaki, K. Nakao, K. Tomishige, K. Kunimori, Appl. Catal. A 328 (2007)
140.
[50] T. Sasaki, K. Nakao, K. Tomishige, K. Kunimori, Chem. Commun. (2006) 3821.
[51] M.A. Quinlan, B.J. Wood, H. Wise, Chem. Phys. Lett. 118 (1985) 478.
[52] M. Maestri, A. Beretta, G. Groppi, E. Tronconi, P. Forzatti, Catal. Today 105
(2005) 709.
Acknowledgments
We sincerely thank Japan Oil, Gas and Metals National Corpora-
tion (JOGMEC) and Chiyoda Corporation for financial support.