Rh/Ce0.25Zr0.75O2 Catalyst for Steam Reforming of Propane at Low Temperature

Article abstract of DOI:10.1002/cctc.201801824

Solid oxide fuel cells (SOFCs) show high energy-conversion efficiency and thus emit less CO₂ than conventional combustion engines. Although SOFCs can directly convert hydrocarbons such as liquefied petroleum gas, these fuels readily induce coking on the electrodes of fuel cell stacks. To avoid coking, hydrocarbons can be subjected to a preliminary endothermic steam-reforming step at a relatively low temperature using waste heat from the stack. Herein, we report that a Rh/Ce_0.25Zr_0.75O₂ catalyst exhibited higher propane-steam-reforming activity than other Rh/Ce_1?xZr_xO₂ catalysts and Rh/γ-Al₂O₃. Catalyst characterization revealed that Rh/Ce_0.25Zr_0.75O₂ had excellent redox property and high H₂O-adsorption activity, which contributed to the activation of steam and thus enhanced the propane-steam-reforming activity of this catalyst.

Full text of DOI:10.1002/cctc.201801824

Accepted Article
Title: Rh/Ce0.25Zr0.75O2 Catalyst for Steam Reforming of Propane at
Low Temperature
Authors: Katsutoshi Nagaoka, Katsutoshi Sato, and Lin Yu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201801824
Link to VoR: http://dx.doi.org/10.1002/cctc.201801824
A Journal of
www.chemcatchem.org

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
Rh/Ce0.25Zr0.75O2 Catalyst for Steam Reforming of Propane at Low
Temperature
Lin Yu,[a] Katsutoshi Sato,[a,b] Katsutoshi Nagaoka*[a]
Consequently, catalysts based on noble metals, especially Rh,
Abstract: Solid oxide fuel cells (SOFCs) show high energy-
conversion efficiency and thus emit less CO2 than conventional
combustion engines. Although SOFCs can directly convert
hydrocarbons such as liquefied petroleum gas, these fuels readily
induce coking on the electrodes of fuel cell stacks. To avoid coking,
hydrocarbons can be subjected to a preliminary endothermic steam-
reforming step at a relatively low temperature using waste heat from
the stack. Herein, we report that a Rh/Ce0.25Zr0.75O2 catalyst exhibited
higher propane-steam-reforming activity than other Rh/Ce1-xZrxO2
catalysts and Rh/γ-Al2O3. Catalyst characterization revealed that
Rh/Ce0.25Zr0.75O2 had excellent redox property and high H2O-
adsorption activity, which contributed to the activation of steam and
thus enhanced the propane-steam-reforming activity of this catalyst.
are attracting increasing attention because of their high activity for
hydrocarbon reforming at low temperature.[7,34–36] Moreover,
Alphonse et al. have found that Ce0.5Zr0.5O2 and TiO2, which
interact strongly with Rh, are effective catalyst supports for low-
temperature steam reforming of propane. [37] However, as far as
we know, these investigators did not optimize the Ce/Zr ratio or
elucidate the function of the Ce0.5Zr0.5O2 support.
In this study, we prepared Rh catalysts on various Ce1-xZrxO2
supports, and we investigated their activity for steam reforming of
propane at 200–400 °C. We used a conventional Rh/γ-Al2O3
catalyst for comparison. Rh/Ce0.25Zr0.75O2 showed the highest
activity among the catalysts studied. To investigate the role of the
supports, we used characterization techniques such as
temperature-programmed reduction (TPR), H2O–temperature-
programmed desorption (TPD), and a transient response test
using H218O.
Introduction
Results and Discussion
Energy-saving technological innovations such as fuel cells are in
high demand around the world. Fuel cells can directly convert
chemical energy into electrical energy more efficiently than
internal-combustion engines and have therefore attracted
considerable attention as environmentally friendly power-
generation devices.[1–9] Among the various types of fuel cells, solid
oxide fuel cells (SOFCs) show the highest energy-conversion
efficiency and therefore emit less CO2 than conventional
combustion engines.[10–16] However, the operation temperature of
the currently available SOFC systems is so high (700–1000 °C)
that starting and shutting them down is time-consuming, and the
component parts deteriorate readily. Therefore, a considerable
amount of research on lowering the operation temperature of
SOFCs has been conducted.[12,17–23]
One potential fuel for SOFCs is liquefied petroleum gas (LPG),
which consists mainly of propane (C3H8). LPG has a higher
energy density than natural gas, is easy to store, and has a well-
established distribution infrastructure.[8,10,24–27] However, although
SOFCs can convert LPG directly, heavier hydrocarbons can
induce coking on the electrodes of fuel cell stacks and in the
system pipes.[27–31] Therefore, a catalytic low-temperature (300–
400 °C) pre-reforming step that eliminates heavier hydrocarbons
by converting them to CH4, H2, and COx using waste heat from
the stack is important. This pre-reforming step requires a noble-
metal catalyst that shows high activity at low temperature.
The temperature dependences of the catalytic activities of
Rh/Ce1-xZrxO2 and Rh/γ-Al2O3 are shown in Figure 1a. At 200 °C,
no propane consumption was observed over Rh/γ-Al2O3, and
conversion was low over all the Rh/Ce1-xZrxO2 catalysts. At 250 °C,
some of the propane was consumed, and the propane-conversion
activities of the catalysts decreased in the order Rh/Ce0.25Zr0.75O2
> Rh/Ce0.5Zr0.5O2 > Rh/Ce0.75Zr0.25O2 > Rh/CeO2 > Rh/ZrO2
>
Rh/γ-Al2O3. At the target temperature for this study (300 °C),
Rh/Ce0.25Zr0.75O2 showed the best activity: the propane
conversion reached 55%, which was 6 times that over Rh/γ-Al2O3.
At 300 °C, the catalyst activities for propane conversion
decreased in the order Rh/Ce0.25Zr0.75O2 > Rh/Ce0.5Zr0.5O2
>
Rh/ZrO2 > Rh/Ce0.75Zr0.25O2 > Rh/CeO2 (Figure 1b). At 350 °C,
propane conversion over Rh/Ce0.25Zr0.75O2 and Rh/ZrO2 reached
100%, and the activities of the other catalysts decreased in the
order Rh/Ce0.5Zr0.5O2 > Rh/Ce0.75Zr0.25O2 > Rh/CeO2. At 400 °C,
conversion reached 100% over all the catalysts except Rh/CeO2
and Rh/Ce0.75Zr0.25O2. Notably, Rh/γ-Al2O3 exhibited much lower
activity than the other catalysts at 200–350 °C. To determine why
Rh/Ce0.25Zr0.75O2 showed the highest activity over the entire
temperature range, we used various techniques to characterize
the catalysts.
Figure 2 shows the X-ray diffraction patterns of the Rh/Ce1-
xZrxO2 and Rh/γ-Al2O3 catalysts. In the pattern for Rh/γ-Al2O3, the
major peaks assigned to γ-Al2O3 were observed. The pattern for
Rh/CeO2 showed peaks that were assigned to a typical cubic
fluorite structure. With the addition of ZrO2 to CeO2, the diffraction
peaks assignable to the cubic fluorite structure shifted to higher
theta values. These results indicate that Zr4+, which has a smaller
cation radius than Ce4+ (0.84 vs. 0.97 Å, respectively), was
incorporated into the CeO2 lattice by cation exchange and that a
solid solution of CeO2 and ZrO2 was formed. Note also that in the
patterns for the Zr4+-doped catalysts we observed no extra peaks
due to a ZrO2 phase. For the Rh/ZrO2 catalyst, the major peaks
assigned to monoclinic zirconia were identified. Phases related to
the presence of Rh species could not be identified in any of the
Although Ni catalysts are widely used for steam reforming of
hydrocarbons in industry owing to their high activity and low
cost,[30–33] these catalysts tend to induce coke formation.
[a] L. Yu, Dr. K. Sato, Prof. Dr. K. Nagaoka
Department of Integrated Science and Technology, Faculty of
Science and Engineering, Oita University, Oita 870-1192 (Japan)
E-mail: nagaoka@oita-u.ac.jp
[b] Dr. K. Sato
Elements Strategy Initiative for Catalysis and Batteries (ESICB),
Kyoto University, Kyoto 615-8245 (Japan)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
●
● Cubic phaseCeO2 ■ ZrO2 ▲ γ-Al2O3
●
●
(a)
100
●
Rh/CeO2
●
●
Rh/Ce0.75Zr0.25O2
Rh/CeO2
Rh/Ce0.5Zr0.5O2
Rh/Ce0.25Zr0.75O2
Rh/ZrO2
80
60
40
20
0
Rh/Ce0.75Zr0.25O2
Rh/Ce0.5Zr0.5O2
Rh/γ-Al2O3
Rh/Ce0.25Zr0.75O2
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
Rh/ZrO2
▲
▲
▲
▲
▲
▲
Rh/γ-Al2O3
20
30
40
50
60
70
2θ (degree)
Figure 2. X-ray diffraction patterns of the supported Rh catalysts.
catalysts, owing to the low Rh loading.
200
250
300
350
400
The physical properties of the catalysts are shown in Table 1.
The Brunauer–Emmett–Teller surface areas of the Rh/Ce1-xZrxO2
catalysts were in the 10–33 m2 g−1 range, and the value for Rh/γ-
Al2O3 was 147 m2 g−1. It seems likely that the higher specific
surface area of the γ-Al2O3 support contributed to an increase in
Rh dispersion: the dispersion over this support was 85%,
compared with 41–50% for the Ce1-xZrxO2 supports. However, at
≤350 °C, Rh/γ-Al2O3 showed much lower propane conversion
than Rh/Ce1-xZrxO2. Therefore, the better propane-conversion
activity of Rh/Ce1-xZrxO2 could not be explained in terms of an
increase in the number of Rh active sites. These results indicate
that the properties of the support, rather than those of the Rh,
were crucial determinants of catalyst activity.
Temperature (ºC)
(b)
60
50
40
30
20
10
0
Rh/CeO2
CeO2
0.25
0.50
0.75
ZrO2
×3
Zr/(Zr+Ce) (molratio)
Figure 1. (a) Temperature dependence of propane conversion
over the supported Rh catalysts. (b) Propane conversion over the
supported Rh catalysts at 300 °C.
Rh/Ce0.75Zr0.25O2
Rh/Ce0.5Zr0.5O2
Rh/Ce0.25Zr0.75O2
×3
×3
Rh/ZrO2
Table 1. Physical properties of the supported Rh catalysts.
Rh/γ-Al2O3
0
200
400
600
800
1000
Catalyst
Specific surface area
(m2 g−1
Dispersion of Rh[a]
(%)
Temperature (℃)
)
Figure 3. H2-TPR profiles of the supported Rh catalysts.
Rh/CeO2
22
27
41
50
50
51
45
85
To investigate the redox properties of the supported Rh catalysts,
we performed H2-TPR analysis (Figure 3). Owing to the
nonreducibility of Al3+, we attributed the reduction peaks observed
for Rh/γ-Al2O3 at temperatures lower than 300 °C to the reduction
of Rh2O3. In contrast, the H2-TPR profile of Rh/CeO2 showed
peaks in two regions: lower than 300 °C and higher than 600 °C.
At temperatures lower than 300 °C, the amount of hydrogen
consumed over Rh/CeO2 was larger than the amount consumed
over Rh/γ-Al2O3. This result indicates that not only Rh2O3 but also
surface CeO2 was reduced at low temperature. The reduction
peak at higher temperature corresponds to the reduction of bulk
CeO2.[38–40] As the Zr content was increased, the high-
temperature reduction peak shifted to lower temperature and
simultaneously decreased in intensity, owing to the decrease in
lattice tension resulting from incorporation of Zr4+ into the CeO2
Rh/Ce0.75Zr0.25O2
Rh/Ce0.5Zr0.5O2
Rh/Ce0.25Zr0.75O2
Rh/ZrO2
33
17
10
Rh/γ-Al2O3
165
[a] Determined from hydrogen-adsorption capacity by assuming H/Rh = 1.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
structure and the corresponding decrease in the amount of
reducible Ce4+.[41–43] When CeO2 is reduced, Ce4+ is converted to
Ce3+, which has a larger cation radius. So introduction of the
smaller Zr4+ cation into the CeO2 lattice decreased the lattice
tension during the reduction and promoted reduction of the
support. As a result, reduction of the support at and below 300 °C
was promoted by the increase in the amount of Zr4+. For Rh/ZrO2,
the reduction peaks at temperatures lower than 300 °C were
attributed to the reduction of Rh2O3, owing to the irreducibility of
Zr4+. These results reveal that addition of Zr4+ to the CeO2 support
markedly enhanced the redox properties of the support.
neither of which contains a redox-active element. The H2O-TPD
results indicate that adding Zr4+ to CeO2 improved the H2O-
activation ability of the reduced support and that the improvement
was greatest over Rh/Ce0.25Zr0.75O2. This result was due to
enhancement of the redox properties of the support, as evidenced
by the H2-TPR analysis.
At 300 °C, the Rh/ZrO2 catalysts exhibited higher propane-
conversion activity than the Rh/γ-Al2O3 catalyst, which indicates
that the ability of ZrO2 to adsorb steam was important for
catalyzing steam reforming of the propane. Furthermore, the fact
that Rh/Ce0.25Zr0.75O2 exhibited the highest propane-conversion
activity among the Rh/Ce1-xZrxO2 catalysts at 300 °C indicates that
the excellent H2O-activation and H2O-adsorption activities of this
catalyst efficiently promoted propane steam reforming at low
temperature.
(a)
6
Rh/CeO2
5
4
3
2
1
Rh/Ce0.75Zr0.25O2
Rh/Ce0.5Zr0.5O2
(a)
3.5
C18O2
Rh/γ-Al2O3
C18OO
3.0
Rh/Ce0.25Zr0.75O2
C18O or C2H6
H218
O
2.5
2.0
1.5
1.0
0.5
0
Rh/ZrO2
Rh/γ-Al2O3
0
100 200 300 400 500 600 700 800
Temperature (℃)
(b)
3
Rh/CeO2
2
1
0
Rh/Ce0.75Zr0.25O2
Rh/Ce0.5Zr0.5O2
0
5
10
15
20
25
30
Time (min)
(b)
Rh/Ce0.25Zr0.75O2
Rh/ZrO2
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
C18O2
Rh/Ce0.25Zr0.75O2
Rh/γ-Al2O3
C18OO
100 200 300 400 500 600 700 800
C18O or C2H6
Temperature (℃)
H218
O
Figure. 4. H2O-TPD profiles of the supported Rh catalysts, with
monitoring by mass spectrometry at (a) m/z 18 (H2O) and (b) m/z
2 (H2).
We carried out H2O-TPD analysis of the catalysts to evaluate
their ability to adsorb and activate H2O, which is one of the
reactants (Figure 4; see Table S1 in the supporting information for
the amounts of H2O desorbed). H2O desorption on Rh/CeO2 (3.0
μmol m−2) was observed at 130–450 °C. However, upon addition
of Zr4+ to the support, H2O desorption was observed at 200–
400 °C and the amount of H2O desorbed was increased
compared with that desorbed on Rh/CeO2; the amount of H2O
desorbed was the highest for Rh/ZrO2 (9.8 μmol m−2). These
results indicate that H2O-adsorption ability was enhanced by
increasing the amount of Zr4+ in the support. The amount of H2O
desorbed on Rh/γ-Al2O3 was 3.5 μmol m−2.
0
5
10
15
20
25
30
Time (min)
Figure 5. Amounts of 18O-containing products formed per specific
surface area over (a) Rh/γ-Al2O3 and (b) Rh/Ce0.25Zr0.75O2 at
300 °C.
As mentioned above, excellent redox property was the most
important determinant of the high catalytic activity of the
Rh/Ce0.25Zr0.75O2 catalyst. To elucidate how the redox properties
of this catalyst contributed to activation of H2O and propane
during the propane steam reforming, we performed transient
response experiments using H218O. For comparison, we used
Rh/γ-Al2O3, which does not contain a redox-active element. After
H2 pretreatment, C3H8 and H218O + H2O were supplied to the
These experiments also revealed that H2 formed on Rh/CeO2 at
the same time as H2O desorption (Figure 4b), by means of the
following reaction:
CeO2-y + yH2O → CeO2 + yH2
(1)
The H2-formation temperature decreased with increasing ZrO2
content. However, H2 did not form on Rh/ZrO2 or Rh/γ-Al2O3,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
catalyst for 30 min at 300 °C. Subsequently, the reaction vessel
was purged with Ar, C3H8 and H2O were fed to the catalyst, and
the products were analyzed by mass spectrometry (MS). Only
small amounts of 18O-containing species formed over Rh/γ-Al2O3
(Figure 5a), whereas large amounts of 18O-containing species
formed over Rh/Ce0.25Zr0.75O2 (Figure 5b). On the basis of these
results, along with the H2-TPR and H2O-TPD results, we attributed
the formation of products containing 18O (CO18O and C18O2) to the
participation of H218O-derived lattice oxygen atoms in the
Ce0.25Zr0.75O2 support. Note that distinguishing the mass spectra
of C18O and C2H6 was difficult because the C18O fragment and the
molecular ion of C2H6 have the same m/z value (30). Because CO
was not detected over Rh/Ce0.25Zr0.75O2 (see Figure 6b for product
distribution) during the catalytic activity test, the mass spectra was
not ascribed to the mass spectra of C18O. On the other hand,
formation of a large amount of C18O2 indicates that C18O produced
by the reaction between propane and lattice oxygen atoms was
further reacted with lattice oxygen atoms [Eq. (2)]. Therefore,
these results demonstrate that the redox property of Ce0.25Zr0.75O2
strongly activated both steam and propane.
(Figure S2 in the supporting information).
(a)
100
80
60
40
20
0
H2O conv.
C3H8 conv.
CH4 selectivity
CO2 selectivity
CO selectivity
C2H6 selectivity
200
250
300
350
400
Temperature (ºC)
CO + H2O → CO2 + H2
(2)
(b)
Next, to clarify the reaction mechanism over Rh/Ce0.25Zr0.75O2,
we investigated the product distribution and the propane and H2O
conversions (Figure 6; see Figure S1 in the supporting information
for the equilibrium yields). Over Rh/γ-Al2O3 (Figure 6a), a high
concentration of CO2 and low concentrations of CH4, CO, and
C2H6 were observed at 250 °C. When the reaction temperature
was increased, the CO selectivity initially increased slightly (at
300 °C) and then decreased. In contrast, the CO2 and C2H6
selectivities decreased with increasing temperature, whereas the
CH4 selectivity increased. These results indicate that the water–
gas shift reaction [Eq. (2)] predominated at 250 °C and that CO
methanation [Eq. (3)] became faster than the shift reaction as the
temperature was increased. Note that the decrease in C2H6
concentration with increasing temperature indicates that steam
cracking of propane [Eq. (4)] was retarded and steam reforming
was enhanced.
100
80
60
40
20
0
H2O conv.
C3H8 conv.
CH4 selectivity
CO2 selectivity
CO selectivity
C2H6 selectivity
200
250
300
Temperature (ºC)
350
400
CO + 3H2 → CH4 + H2O
(3)
(4)
Figure 6. Effect of reaction temperature on product distributions
of propane steam reforming over (a) Rh/γ-Al2O3 and (b)
Rh/Ce0.25Zr0.75O2.
C3H8 + H2O → C2H6 + 2H2 + CO
The results obtained over Rh/Ce0.25Zr0.75O2 are presented in
Figure 6b. At 200 °C, high concentrations of CO2 and C2H6 were
produced, whereas the CH4 concentration was low and CO
generation was not observed. When the reaction temperature
was increased, the CH4 selectivity increased and the C2H6
selectivity decreased, owing to the predominance of propane
steam reforming over cracking at the higher temperatures. The
CO2 selectivity also decreased slowly with increasing temperature
over the entire range. CO methanation was accelerated at higher
temperature until 300 °C, as in the reaction over Rh/γ-Al2O3.
Notably, CO was not detected over Rh/Ce0.25Zr0.75O2 at any
temperature, which contrasts with the results obtained over Rh/γ-
Al2O3. This difference indicates that the rates of CO-consuming
reactions such as the water–gas shift reaction and CO
methanation were much faster over Rh/Ce0.25Zr0.75O2 than over
Rh/γ-Al2O3. Regarding this, we observed that at 200–350 °C, H2O
conversion over Rh/Ce0.25Zr0.75O2 was higher than that over Rh/γ-
Al2O3, a result that is due to the high H2O-activation activity of
Rh/Ce0.25Zr0.75O2. In addition, at low temperatures (200–300 °C),
higher H2 concentrations were observed over Rh/Ce0.25Zr0.75O2
than over Rh/γ-Al2O3, whereas at high temperatures (350–
400 °C), H2 concentrations were comparable for both catalysts
C3H8 + 3H2O → 3CO + 7H2
(5)
Next, we compared the rates of the CO-consuming reactions that
occurred after propane steam reforming over Rh/γ-Al2O3 and
Rh/Ce0.25Zr0.75O2. For these experiments, a reactant mixture
consisting of CO/H2/H2O in a ratio of 3/7/3 was fed to the catalyst
on the assumption that the propane was completely consumed by
steam reforming to produce CO and H2 [Eq. (5)]. When the
reaction was carried out over Rh/γ-Al2O3 (Figure 7a), reaction of
CO started at 250 °C, and conversion reached 9% at 300 °C, 58%
at 350 °C, and 100% at 400 °C. CH4 was the only product at
250 °C. As the reaction temperature was increased, the CH4
selectivity decreased and the CO2 selectivity gradually increased.
However, even at 400 °C, CO methanation predominated.
Furthermore, small amounts of C2H6 and C3H8 were generated at
300–350 °C, indicating that some C–C bond formation occurred
at that temperature range in this condition. In contrast, when the
reaction was carried out over Rh/Ce0.25Zr0.75O2 (Figure 7b),
reaction of CO started at 250 °C, as was the case over Rh/γ-Al2O3,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
Rh/Ce0.25Zr0.75O2, the conventional reaction pathway was
accompanied by a reaction between propane and lattice oxygen
atoms derived from H2O. Although we suspected that differences
in the electronic state of Rh might be responsible for the different
catalytic activities of the catalysts, Rh K-edge XANES spectra for
Rh/Ce0.25Zr0.75O2, Rh/γ-Al2O3, and Rh/ZrO2 after H2 treatment
were comparable with the spectrum of Rh foil (Figure S3 in the
supporting information), indicating that Rh was present in the
catalysts in the metallic state.
(a)
100
CO conv.
CH4 selectivity
CO2 selectivity
C2H6 selectivity
C3H8 selectivity
80
60
40
20
0
CH4, H2, CO2
H2O 1/2H
C3H8
2
200
250
300
350
400
H2
Temperature (ºC)
Rh0
H
(b)
O2-
O-
O2-
100
CO conv.
CH4 selectivity
CO2 selectivity
C2H6 selectivity
C3H8 selectivity
-O2—Ce4+-O2--Zr4+-O2--
80
60
40
20
0
Rh/Ce0.25Zr0.75O2
Figure 8. Mechanism of propane steam reforming over
Rh/Ce0.25Zr0.75O2 at low temperature.
100
80
60
40
20
0
200
250
300
350
400
Temperature (ºC)
Figure 7. Effect of temperature on reactions of CO/H2/H2O over
(a) Rh/γ-Al2O3 and (b) Rh/Ce0.25Zr0.75O2.
but CO conversion increased drastically with temperature,
reaching 100% at 350 °C, which was lower by 50 °C than the
corresponding temperature over Rh/γ-Al2O3. These results reveal
the much higher activity of Rh/Ce0.25Zr0.75O2 for catalyzing CO
consumption. Furthermore, at and below 300 °C, the product
distribution over Rh/Ce0.25Zr0.75O2 was completely different from
that over Rh/γ-Al2O3; specifically, the CO2 concentration over
Rh/Ce0.25Zr0.75O2 was higher than the CH4 concentration. These
results indicate that the water–gas shift reaction was much faster
over Rh/Ce0.25Zr0.75O2 than over Rh/γ-Al2O3 and that CO was
effectively consumed owing to the excellent H2O-activation
activity of Rh/Ce0.25Zr0.75O2, which was due to its excellent redox
properties. When the reaction temperature was increased to
350 °C, the CO2 selectivity decreased further. At temperatures of
350°C and higher, the CH4 concentration was higher than the CO2
concentration, indicating that CO methanation was faster than the
water–gas shift reaction.
It has been reported that during steam reforming reactions, the
support is the principal site for water activation and that metal
sites are responsible for hydrocarbon activation.[44–46] In the cases
of Rh/γ-Al2O3 and Rh/ZrO2, propane was adsorbed on active Rh
surface sites, and steam was adsorbed on support surface sites.
The higher H2O-adsorption activity of Rh/ZrO2 resulted in its
higher activity for propane steam reforming. In the case of
0
10
20
30
40
50
Reaction time (h)
Figure 9. C3H8 conversion as a function of reaction time during
propane steam reforming over Rh/Ce0.25Zr0.75O2. Conditions: (●)
350 °C, space volume = 3.9 L h−1 g−1 and (●) 380 °C, space
volume = 77.1 L h−1 g−1.
On the basis of the characterization results, we propose the
reaction mechanism outlined in Figure 8 for steam reforming of
propane over the Rh/Ce0.25Zr0.75O2 catalyst at low temperature.
The H2-TPR profile revealed that the redox property of the support
is high and that reduction of the support occurs readily at low
temperature owing to the presence of the smaller Zr4+ cation in
the CeO2 lattice. During H2 reduction, Ce0.25Zr0.75O2 is reduced,
and an oxygen defect is thus generated. Activation of steam by
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
the oxygen defect is accompanied by H2 formation, and propane
reacts with the activated-oxygen species to form CO. In addition,
the water–gas shift reaction [Eq. (2)] is promoted by the
Rh/Ce0.25Zr0.75O2 catalyst, owing to the presence of the above-
mentioned activated-oxygen species, and this reaction efficiently
oxidizes CO to CO2. Therefore, CO generation is not observed.
We also evaluated the long-term stability of the Rh/Ce0.25Zr0.75O2
catalyst (Figure 9). When steam reforming was carried out for 50
h, propane conversion decreased from 24% to 20% at 380 °C and
from 99% to 98% at 350 °C. In addition, the product selectivities
remained largely unchanged with time (Figures S4 and S5 in the
supporting information). After the reaction, carbon deposition on
the used catalyst was quantified by means of temperature-
programmed oxidation, which revealed that a trace amount of
carbon had been deposited on the catalysts (0.11 and 0.10 wt%
at 350 and 380 °C, respectively), revealing the high stability of the
catalyst.
determined by assuming total conversion of CO to CO2 in the
entire reaction [Eq. (6)]. The steam-to-carbon ratio was kept at 2/1.
Each catalyst (500 or 100 mg) was reduced in pure H2 (total flow
rate, 20 mL min−1) at 600 °C for 1 h. The reaction was carried out
either at 200–400 °C or at 380 °C. The rates of product formation
were determined by means of gas chromatography (GC) with
thermal conductivity detection. The amount of C3H8 converted
and selectivity for carbon-containing products were calculated
from the carbon balance, because the amount of coke deposited
was very small to negligible. The equations used are shown in the
supporting information.
C3H8 + 6H2O → 3CO2 + 10H2
(6)
For measurement of the activity of the catalysts for CO-
consuming reactions, the ratio in the feed gas was 3/7/3/7 vol %
CO/H2/H2O/N2 (total flow rate, 45.8 mL min−1; space velocity, 5.5
L h−1 g−1). Each catalyst (500 mg) was reduced in pure H2 at a
total flow rate of 20 mL min−1 at 600 °C for 1 h. The reaction
temperatures were in the 200–400 °C range. The rates of product
formation were analyzed by means of GC with thermal
conductivity detection. The amount of CO converted and
selectivity for carbon-containing products were calculated from
the carbon balance. The equations used are shown in the
supporting information.
Conclusions
We found that at 300 °C, a temperature suitable for SOFC
systems, Rh/Ce0.25Zr0.75O2 exhibited higher propane-steam-
reforming activity than Rh/γ-Al2O3 and other Rh/CexZr1-xO2
catalysts. In addition to the conventional mechanism involving
water adsorption and subsequent activation on the oxide support,
H2O was also activated by oxygen defect sites that formed on the
Rh/Ce0.25Zr0.75O2 catalyst, owing to the excellent redox property
of the support. This activation mechanism promoted propane
steam reforming and subsequent water–gas shift reaction.
Stability tests revealed that Rh/Ce0.25Zr0.75O2 resisted coking and
showed stable activity for at least 50 h, making the catalyst a
promising candidate for pre-reforming of hydrocarbons for SOFC
applications.
Transient response tests using labeled H218O were conducted to
ascertain the contribution of lattice oxygen in the support. Each
catalyst (500 mg) was reduced in flowing H2 at 600 °C, cooled to
300 °C in Ar, and then analyzed as follows:
(1) Reactant gas containing C3H8/H216O (17.5% H218O)/Ar at a
ratio of 1/6/7 vol% and a total flow rate of 32.1 mL min−1 was
supplied to the catalyst for 30 min.
(2) The reaction vessel was purged with Ar for 3 h.
(3) Reaction gas containing C3H8/H216O/Ar at a ratio of 1/6/7 and
a total flow rate of 32.1 mL min−1 was supplied to the catalyst, and
simultaneously the exit gas was analyzed by means of
quadrupole MS. Ar was added as an internal standard. The
following 18O species were analyzed: m/z 20 (H218O), m/z 30
(C18O), m/z 46 (CO18O), and m/z 48 (C18O2).
Experimental Section
Catalyst preparation
Characterization
For preparation of the Ce1-xZrxO2 supports, an aqueous solution
containing ZrO(NO3)2·2H2O (Wako Pure Chemical Industries) and
Ce(NO3)3·6H2O (Wako) was added to a 25% aqueous NH3
solution. Co-precipitation occurred, and the co-precipitates were
kept in suspension by stirring at room temperature overnight and
were then filtered, washed with distilled water, dried overnight at
70 °C, and precalcined at 700 °C to afford Ce1-xZrxO2, where x =
0, 0.25, 0.5, 0.75, and 1.0. A γ-Al2O3 sample (JRC-ALO-8) was
obtained from the Catalysis Society of Japan, precalcined at
700 °C, and used for wet impregnation with aqueous
Rh(NO3)3·nH2O (Mitsuwa Pure Chemicals). The final
concentration of Rh was 0.5 wt%. All the samples were dried at
room temperature, and subsequently at 70 °C overnight, and the
dried samples were calcined at 450 °C in flowing air. The powdery
catalysts were pressed into pellets, crushed, and sieved to obtain
grains with diameters between 250 and 500 μm.
X-ray diffraction patterns were recorded on a SmartLab X-ray
diffractometer with a Cu Kα radiation (Rigaku, Japan) source
operating at 40 kV and 30 mA.
The specific surface areas of the catalysts were estimated by the
Brunauer–Emmett–Teller method with
a
BELSORP-mini
instrument (BEL Japan Inc., Japan) at 196 °C with N2 as the
analysis gas. Before each measurement, the sample was
outgassed at 300 °C for 2 h.
H2-TPR measurements were performed by heating the catalysts
(50 mg) from room temperature to 1000 °C at a rate of 10 °C min−1
in flowing H2/Ar gas (5 vol%; flow rate, 30 mL min−1). H2
consumption was monitored by means of GC with thermal
conductivity detection.
The dispersion of Rh metal was determined by a H2 pulse
chemisorption method. Each catalyst (200 mg) was reduced in
flowing H2 at 600 °C for 1 h (heating rate, 5 °C min−1). After
reduction, the catalyst was heated at 600 °C for 30 min under
flowing Ar to desorb any H2 that spilled over the support and was
then cooled to room temperature under flowing Ar. The H2 uptake
measurements were performed at 76 °C.
H2O-TPD measurements were performed with a BELCAT II- VP-
C chemisorption analyzer (Microtrac-BEL). Catalyst (200 mg) was
loaded into a quartz reactor and reduced in a stream of H2 at
Catalytic activity tests
Tests of catalyst activity for steam reforming of propane were
conducted in a fixed-bed flow reactor system at atmospheric
pressure. The C3H8/H2O/N2 ratio in the feed gas was 1/6/7 vol%
(total flow rate, 32.1 mL min−1; space velocity, 3.9 L h−1 g−1; or
total flow rate, 128.5 mL min−1; space velocity, 77.1 L h−1 g−1), as
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
600 °C for 1 h; the reactor was then purged with a stream of He
for 1 h and cooled to 100 °C. After 5% H2O in He gas (flow rate,
30 mL min−1) was fed to the catalyst for 30 min at 100 °C, the oven
temperature was increased at 10 °C min−1 to 800 °C. The H2O
and H2 desorption profiles were monitored by quadrupole MS at
m/z 18 and m/z 2, respectively.
[11] A. Yan, S. Huang, S. Li, R. Chen, P. Ohodnicki, M. Buric, S. Lee, M. J.
Li, K. P. Chen, Sci. Rep. 2017, 7, 9360.
[12] P. Arunkumar, R. Ramaseshan, S. Dash, K. S. Babu, Sci. Rep. 2017, 7,
3450.
[13] Q. Zhang, X. Du, S. Tan, D. Tang, K. Chen, T. Zhang, Sci. Rep. 2017, 7,
5355.
The formation of the carbonaceous deposits was quantified by
temperature-programmed oxidation of the used catalyst. Used
catalyst (50 mg) was placed in a quartz reactor, and the reactor
was heated from room temperature to 1000 °C at a rate of 10 °C
min−1 under a flow of O2/Ar gas (5 vol%; flow rate, 28.5 mL min−1).
The effluents were analyzed by GC with flame ionization detection.
X-ray absorption fine structure measurements for the Rh K-
edges were performed on the BL01B1 beamline at SPring-8. A Si
(311) two-crystal monochromator was used for the
measurements of absorption edges. After being reduced under a
flow of pure H2 at 600 °C for 1 h, the samples were cooled in Ar,
recovered from the tubular reactor, and encapsulated in plastic
bags filled with Ar. All the procedures for recovery and transfer of
the samples to the plastic bags were carried out under Ar to avoid
altering the samples through air exposure. The plastic bags were
transferred to the beamline and spectra of the catalyst disks were
measured. Measurements of the Rh K-edge were carried out in
transmittance mode with two ion chambers. The X-ray energy was
calibrated against the spectra of standard Rh foils. Data analysis
was performed with the Athena and Artemis programs (ver.
0.9.25) included in the Demeter package. [47]
[14] N. Minh, Solid State Ionics 2004, 174, 271-277.
[15] V. Eveloy, W. Karunkeyoon, P. Rodgers, A. Al Alili, Int. J. Hydrogen
Energy 2016, 41, 13843-13858.
[16] M. Cimenti, J. Hill, Energies 2009, 2, 377-410.
[17] T. H. Santos, J. P. F. Grilo, F. J. A. Loureiro, D. P. Fagg, F. C. Fonseca,
D. A. Macedo, Ceram. Int. 2018, 44, 2745-2751.
[18] C. Duan, D. Hook, Y. Chen, J. Tong, R. O'Hayre, Energy Environ. Sci.
2017, 10, 176-182.
[19] T. Fukui, Nanoparticle Technology Handbook (Third Edition) 2018, 499-
503.
[20] E. D. Wachsman, K. T. Lee, Science 2011, 334, 935-939.
[21] Z. Gao, L. V. Mogni, E. C. Miller, J. G. Railsback, S. A. Barnett, Energy
Environ. Sci. 2016, 9, 1602-1644.
[22] C. Xia, Solid State Ionics 2002, 149, 11-19.
[23] X. Zhang, M. Robertson, C. Deĉes-Petit, W. Qu, O. Kesler, R. Maric, D.
Ghosh, J. Power Sources 2007, 164, 668-677.
[24] L. Li, X. Wang, K. Shen, X. Zou, X. Lu, W. Ding, Chin. J. Catal. 2010, 31,
525-527.
[25] H. Jeong, M. Kang, Appl. Catal., B 2010, 95, 446-455.
[26] Y. Lu, J. Chen, Y. Liu, Q. Xue, M. He, J. Catal. 2008, 254, 39-48.
[27] L. Yu, K. Sato, T. Toriyama, T. Yamamoto, S. Matsumura, K. Nagaoka,
Chemistry — Eur. J. 2018, 24, 8742-8746.
[28] K. Shen, X. Wang, X. Zou, X. Wang, X. Lu, W. Ding, Int. J. Hydrogen
Energy 2011, 36, 4908-4916.
Acknowledgements
[29] M. Krumpelt, Catal. Today 2002, 77, 3-16.
[30] K. Nagaoka, K. Sato, H. Nishiguchi, Y. Takita, Catal. Commun. 2007, 8,
1807-1810.
The X-ray absorption experiments were performed with the
approval of the Japan Synchrotron Radiation Research Institute
(JASRI) (Proposal Nos. 2015A1534, 2018B1711).This work was
partly supported by the ACCEL, Japan Science and Technology
Agency (grant number JPMJAC1501). K.S. thanks the Program
for Elements Strategy Initiative for Catalysts & Batteries (ESICB)
commissioned by MEXT. This work was supported by JSPS
KAKENHI Grant Number 15H04188.
[31] K. Nagaoka, K. Sato, H. Nishiguchi, Y. Takita, Appl. Catal., A 2007, 327,
139-146.
[32] J. R. Rostrup-Nielsen, J. Sehested, J. K. Nørskov, Adv. Catal. 2002, 47,
65-139.
[33] P. Ferreira‐Aparicio, M. J. Benito, J. L. Sanz, Catal. Rev. 2005, 47, 491-
588.
[34] A. Qi, S. Wang, C. Ni, D. Wu, Int. J. Hydrogen Energy 2007, 32, 981-991.
[35] M. Ferrandon, T. Krause, Appl. Catal., A 2006, 311, 135-145.
[36] S. Ayabe, H. Omoto, T. Utaka, R. Kikuchi, K. Sasaki, Y. Teraoka, K.
Eguchi, Appl. Catal., A 2003, 241, 261-269.
Keywords: redox • steam reforming • heterogeneous catalysis •
reaction mechanism • low temperature
[37] P. Alphonse, F. Ansart, J. Colloid Interface Sci. 2009, 336, 658-666.
[38] K. Sato, K. Adachi, Y. Takita, K. Nagaoka, ChemCatChem 2014, 6, 784-
789.
[1]
[2]
[3]
S. Choi, C. J. Kucharczyk, Y. Liang, X. Zhang, I. Takeuchi, H.-I. Ji, S. M.
Haile, Nat. Energy 2018, 3, 202-210.
C. Li, Q. Yuan, B. Ni, T. He, S. Zhang, Y. Long, L. Gu, X. Wang, Nat.
Commun. 2018, 9, 3702.
[39] H. C. Yao, Y. F. Yu Yao, J. Catal. 1984, 86, 254-265.
[40] S. R. Gomes, N. Bion, D. Duprez, F. Epron, Appl. Catal., B 2016, 197,
138-145.
W. Zheng, L. Wang, F. Deng, S. A. Giles, A. K. Prasad, S. G. Advani, Y.
Yan, D. G. Vlachos, Nat. Commun. 2017, 8, 418.
M. Schrope, Nature 2001, 414, 682-684.
[41] P. Fornasiero, R. Dimonte, G. R. Rao, J. Kaspar, S. Meriani, A. Trovarelli,
M. Graziani, J. Catal. 1995, 151, 168-177.
[4]
[5]
[6]
D. L. Trimm, Z. I. Önsan, Catal. Rev. 2001, 43, 31-84.
K. Takehira, T. Shishido, P. Wang, T. Kosaka, K. Takaki, J. Catal. 2004,
221, 43-54.
[42] K. Otsuka, Y. Wang, M. Nakamura, Appl. Catal., A 1999, 183, 317-324.
[43] K. Sato, K. Kawano, A. Ito, Y. Takita, K. Nagaoka, ChemSusChem 2010,
3, 1364-1366.
[7]
[8]
Y. Li, X. Wang, C. Xie, C. Song, Appl. Catal., A 2009, 357, 213-222.
X. Zou, X. Wang, L. Li, K. Shen, X. Lu, W. Ding, Int. J. Hydrogen Energy
2010, 35, 12191-12200.
[44] T. R. Phillips, J. Mulhall, G. E. Turner, J. Catal. 1969, 15, 233-244.
[45] K. S. M. Bhatta, G. M. Dixon, Ind. Eng. Chem. Prod. Res. Dev. 1969, 8,
324-331.
[9]
M. C. Steil, S. D. Nobrega, S. Georges, P. Gelin, S. Uhlenbruck, F. C.
Fonseca, Appl. Energy 2017, 199, 180-186.
[46] J. R. Rostrup-Nielsen, J. Catal. 1973, 31, 173-199.
[47] B. Ravel, M. Newville, J. Synchrotron Radiat. 2005, 12, 537-541.
[10] C. Duan, R. J. Kee, H. Zhu, C. Karakaya, Y. Chen, S. Ricote, A. Jarry, E.
J. Crumlin, D. Hook, R. Braun, N. P. Sullivan, R. O'Hayre, Nature 2018,
557, 217-222.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201801824
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Redox property: A
L. Yu, K. Sato, K. Nagaoka*
CO2
Rh/Ce0.25Zr0.75O2 catalyst exhibited
higher propane-steam-reforming
CH4
Page No. – Page No.
H2O
H2
activity than other Rh/Ce1-xZrxO2
catalysts and Rh/γ-Al2O3. The
excellent redox property and high
H2O-adsorption activity of
Rh/Ce0.25Zr0.75O2 activated the
steam and thus enhanced the
activity of the catalyst for propane
steam reforming.
Rh/Ce0.25Zr0.75O2 Catalyst for
Steam Reforming of Propane at
Low Temperature
C3H8
Rh
O2-
Oxygen defect
O2-
Lattice oxygen
Rh/Ce0.25Zr0.75O2
This article is protected by copyright. All rights reserved.