Role and advantages of H2S in catalytic steam reforming over nanoscale CeO2-based catalysts

Article abstract of DOI:10.1016/j.jcat.2010.08.015

The activity of nanoscale CeO₂ and doped CeO₂ (with Gd, Y, Nb, La, and Sm) toward the steam reforming of CH₄ in the presence of H₂S was investigated for later application as an in-stack reforming catalyst in a solid oxide fuel cell. Although H₂S is commonly known as a poisonous gas for metallic-based catalysts, it was found that the presence of appropriate H₂S content increases the reforming activity of these CeO₂-based catalysts. According to postreaction catalyst characterizations by X-ray diffraction, X-ray photoelectron spectroscopy, temperature-programmed reduction, temperature-programmed desorption, H₂/H₂O + H₂S titration, and ¹⁸O/¹⁶O isotope exchange, it was revealed that this behavior is related to the formation of various Ce-O-S phases (Ce(SO ₄)₂, Ce₂(SO₄)₃, and Ce₂O₂S) during the reaction. Our studies indicated that the formation of Ce(SO₄)₂ promotes the oxygen storage capacity, the lattice oxygen mobility, and eventually the reforming activity, whereas the formation of Ce₂O₂S oppositely reduces both properties and lowers the reforming rate.

Full text of DOI:10.1016/j.jcat.2010.08.015

Journal of Catalysis 276 (2010) 6–15
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Role and advantages of H2S in catalytic steam reforming over nanoscale
CeO2-based catalysts
b
c
d
N. Laosiripojana a, , S. Charojrochkul , P. Kim-Lohsoontorn , S. Assabumrungrat
⇑
a The Joint Graduate School of Energy and Environment, CHE Center for Energy Technology and Environment, King Mongkut’s University of Technology Thonburi, Thailand
b National Metal and Materials Technology Center (MTEC), Pathumthani, Thailand
c Department of Chemical Engineering, Faculty of Engineering, Mahidol University, Thailand
d Department of Chemical Engineering, Faculty of Engineering, Chulalongkorn University, Thailand
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 June 2010
Revised 8 August 2010
Accepted 14 August 2010
Available online 29 September 2010
The activity of nanoscale CeO2 and doped CeO2 (with Gd, Y, Nb, La, and Sm) toward the steam reforming
of CH4 in the presence of H2S was investigated for later application as an in-stack reforming catalyst in a
solid oxide fuel cell. Although H2S is commonly known as a poisonous gas for metallic-based catalysts, it
was found that the presence of appropriate H2S content increases the reforming activity of these CeO2-
based catalysts. According to postreaction catalyst characterizations by X-ray diffraction, X-ray photo-
electron spectroscopy, temperature-programmed reduction, temperature-programmed desorption, H2/
H2O + H2S titration, and 18O/16O isotope exchange, it was revealed that this behavior is related to the for-
mation of various Ce–O–S phases (Ce(SO4)2, Ce2(SO4)3, and Ce2O2S) during the reaction. Our studies indi-
cated that the formation of Ce(SO4)2 promotes the oxygen storage capacity, the lattice oxygen mobility,
and eventually the reforming activity, whereas the formation of Ce2O2S oppositely reduces both proper-
ties and lowers the reforming rate.
Keywords:
CeO2
Steam reforming
H2S
Hydrogen
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
of catalytic applications [10–13]. One of the great potential appli-
cations of CeO2-based material is in solid oxide fuel cells (SOFC)
Hydrogen is widely regarded as a promising energy carrier for
fuel cells to generate energy with great improvements in air qual-
ity, human health, and climate [1]. It can be produced readily from
the reforming of hydrocarbons with oxygen-containing co-reac-
tants [2–5]. Metallic catalysts such as Ni, Rh, and Pd are known
to be active for these reactions, but catalyst deactivation due to
carbon formation and sulfur poisoning is a major concern when
heavy hydrocarbons and/or sulfur-containing feeds such as natural
gas, biogas, and liquefied petroleum gas are used [6,7]. Typically,
prereforming and/or desulfurization units are needed to reform
these feedstocks; however, both installations reduce the flexibility
and the potential for applying hydrogen/fuel cell technologies. Re-
search on developing catalysts with high resistance to carbon for-
mation and sulfur interaction is therefore continuing.
as cell materials and in-stack reforming catalysts (IIR-SOFC) [14–
19]. Furthermore, doping CeO2 with Gd, Nb, La, and Sm has also
been reported to improve the redox properties of CeO2, and these
doped forms are now widely used as catalysts in a wide variety
of reactions involving oxidation or partial oxidation of hydrocar-
bons (e.g., automotive catalysis). Previously, we successfully syn-
thesized nano-scale CeO2 with high specific surface area and
thermal stability by a cationic surfactant-assisted method [20].
We then studied H2O and CO2 reforming of CH4, C2H4, C2H6,
C3H8, C4H10, CH3OH, and C2H5OH over this ultrafine CeO2 and
found that this material efficiently converts these hydrocarbons
to H2-rich gas with high resistance toward carbon formation under
given conditions [20,21]. We proposed that the turnover rates are
strongly influenced by the amount of lattice oxygen (Ox ) in CeO2,
O
Cerium oxide (CeO2) is extensively used as a catalyst and sup-
port for a variety of reactions involving oxidation of hydrocarbons
[4,8,9]. This material contains a high concentration of highly
mobile oxygen vacancies, which act as local sources or sinks for
oxygen involved in reactions taking place on its surface; this
behavior renders CeO2-based materials of interest for a wide range
whereas the kinetic dependencies of hydrocarbon conversions
and the activation energies were unaffected by the material’s spe-
cific surface area, the doping element, the degree of oxygen storage
capacity (OSC), and the reactions (i.e., H2O reforming and CO2
reforming). Furthermore, identical turnover rates for H2O and
CO2 reforming (at similar CnHm partial pressure) with linear depen-
dence on CnHm partial pressure and independence of CO2 and H2O
partial pressures were observed. These results provide strong
evidence that the sole kinetically relevant elementary step is the
reaction of intermediate surface hydrocarbon species with OOx
⇑
Corresponding author. Fax: +66 662 872 6736.
E-mail address: navadol_l@jgsee.kmutt.ac.th (N. Laosiripojana).
0021-9517/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2010.08.015

N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
7
and that oxygen is replenished by a rapid surface reaction of re-
duced-state CeO2 with oxygen sources (i.e., CO2 or H2O) in the sys-
tem [21].
Recently, the regenerative H2S adsorption capability of CeO2 at
high temperature was reported [22]. In the present work, the activ-
ity of CeO2 and doped CeO2 (with several rare earths, Gd, Y, Nb, La,
and Sm), synthesized by a cationic surfactant-assisted method, to-
ward the steam reforming of CH4 in the presence of H2S was inves-
tigated under several operating conditions (i.e., various inlet H2S
contents, inlet steam/carbon (S/C) molar ratios, and operating tem-
peratures). Several characterizations, including X-ray diffraction
(XRD), X-ray photoelectron spectroscopy (XPS), temperature-pro-
grammed reduction (TPR), temperature-programmed desorption
(TPD) under reducing conditions, H2/H2O + H2S titration, and
18O/16O isotope exchange methods were also performed over the
fresh and spent catalysts from the reactions in order to observe
the changes in catalyst phase formation and the redox properties
associated with the OSC and the mobility of lattice oxygen. Based
on the experimental results and the above analyses, the mecha-
nism of methane steam reforming in the presence of H2S over
nano-scale CeO2-based catalysts was established. In addition, a
practical application with respect to the study using this ultrafine
CeO2 as a primary reforming catalyst to reform natural gas, biogas,
and liquefied petroleum gas (LPG) without prior desulfurization
was proposed.
2. Experimental
2.1. Catalyst preparations
Nano-scale CeO2-based materials were synthesized by a cat-
ionic surfactant-assisted method. We previously reported that
the preparation of ceria-based materials by this method can pro-
vide materials with ultrafine particle size, high surface area, and
good stability after thermal treatment [20,21]. The achievement
of high-surface-area material with good thermal stability by this
preparation technique is related to the interaction of hydrous oxide
with cationic surfactants under basic conditions and the incorpora-
tion of surfactants during preparation, which reduces the interfa-
cial energy and eventually decreases the surface tension of water
contained in the pores. It has been reported that this incorporation
reduces the shrinkage and collapse of the catalyst during heating,
which consequently helps the catalyst maintain high surface area
after calcination [23]. In the present work, the undoped CeO2
was prepared by mixing 0.1 M of cerium nitrate (Ce(NO3)3ꢀH2O
from Aldrich) solution with 0.1 M of cetyltrimethylammonium
bromide (from Aldrich) and keeping the [Ce]/[cetyltrimethylam-
monium bromide] molar ratio constant at 0.8. This solution was
stirred by magnetic stirring (100 rpm) for 3 h and then aqueous
ammonia was slowly added until the pH was 11.5. The mixture
was continually stirred, sealed, and placed in a thermostatic bath
maintained at 263 K. The precipitate was then filtered and washed
with deionized water and acetone to remove the free surfactant. It
was dried overnight in ambient air at 110 °C and then calcined in
flowing dry air by increasing the temperature to 900 °C at a rate
of 0.167 °C sꢁ1 and holding at 900 °C for 6 h. After calcination, fluo-
rite-structured CeO2 with good homogeneity was obtained.
According to the SEM image (Fig. 1), ultrafine particles of CeO2
can be achieved from this method (compared to the microscale
CeO2 obtained from the conventional precipitation method).
Doped CeO2 with Gd, Y, Nb, La, and Sm was prepared by mixing
Ce(NO3)3 with RE(NO3)x (Re = Gd, Nb, La, Y, and Sm) to achieve a
RE ratio in the material of 0.1; RE0.1–CeO2 (in the presence of
0.1 M cetyltrimethylammonium bromide solution). After treat-
ments, the specific surface areas of all doped and undoped CeO2
Fig. 1. SEM micrograph of (a) CeO2 prepared by precipitation method and (b) CeO2
prepared by cationic surfactant-assisted method (after calcination at 900 °C).
Table 1
Specific surface area of ceria-based materials at several calcination temperatures.
Catalysts
Surface area (m2 gꢁ1) after calcination at
700 °C
800 °C
900 °C
CeO2
52
68
61
62
40
59
35
51
47
45
23
41
18
26
23
23
11
20
La-doped CeO2
Gd-doped CeO2
Sm-doped CeO2
Nb-doped CeO2
Y-doped CeO2
were determined by BET measurement (Table 1), and as expected,
the surface area for all ceria decreased at high calcination
temperatures.
It is noted that, for comparison, Ni/Al2O3 and Rh/Al2O3 (5 wt.%
Ni and Rh) were also prepared and tested toward the steam
reforming reaction. In detail, these catalysts were prepared by
the wet impregnation of
a-Al2O3 with aqueous solutions of
Ni(NO3)2 and Rh(NO3)2 (from Aldrich). According to the fresh cat-
alyst (after reduction) characterizations by X-ray fluorescence
(XRF) analysis, temperature-programmed reduction (TPR) with
5% H2 in helium, and temperature-programmed desorption (TPD)
studies, the Ni and Rh weight content was 4.9% and 5.1%, the metal
reducibility was 92.1% and 94.8%, and the metal reducibility per-
centage was 4.87% and 5.04%, respectively.

8
N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
stoichiometry. The following equations present the calculations of
these selectivities:
2.2. Catalytic steam reforming and relevant reactions
ꢀ
ꢁ
To undertake the reaction testing, an experimental reactor sys-
tem was constructed. The feed gases, including the components of
interest, i.e. CH4, natural gas, biogas, LPG, and H2S, were controlled
and introduced to the system by the mass flow controllers, while
deionized H2O was fed by a syringe pump passing through an
evaporator. For methane steam reforming testing, the inlet concen-
tration of CH4 was 20%, whereas the steam concentration was var-
ied to achieve H2O/CH4 ratios between 0.5 and 3.0 and the inlet H2S
concentration was between 10 and 1000 ppm to capture a range of
H2S content in various fuels. As for the steam reforming of sulfur-
containing fuels (i.e., natural gas, biogas, LPG) testing, the inlet con-
centration of total hydrocarbons in these fuels was kept constant at
20%, whereas various steam concentrations were added to achieve
S/C ratios between 0.5 and 3.0. It is noted that the reforming tests
with and without predesulfurization of these hydrocarbons were
compared. Regarding the desulfurization unit, ZnO was applied
to adsorb H2S from the feed (the outlet gases were rechecked by
gas chromatography with a flame photometric detector (FPD) to
ensure that all H2S were removed before purging to the reforming
testing unit).
The inlet gas mixtures were introduced to the reaction section,
in which a 10-mm-diameter quartz reactor was mounted vertically
inside a tubular furnace. The catalysts (500 mg) were diluted with
SiC (to obtain a total weight of 3.0 g) in order to avoid temperature
gradients and loaded into the quartz reactor, which was packed
with quartz wool to prevent the catalyst from moving. In the sys-
tem, a type-K thermocouple was placed in the annular space be-
tween the reactor and furnace. This thermocouple was mounted
in close contact with the catalyst bed to minimize the temperature
difference. It is noted that another type-K thermocouple, covered
by a closed-end quartz tube, was inserted into the middle of the
quartz reactor in order to recheck the possible temperature devia-
tion due to the heat transfer limitation. The record showed that the
maximum temperature fluctuation during the reaction was always
0.75 °C or less from the temperature specified for the reaction.
After the reactions, the exit gas mixture was transferred via
trace-heated lines (100 °C) to the analysis section, which consisted
of a Porapak Q column Shimadzu 14B gas chromatograph (GC) and
a quadrupole mass spectrometer (MS). The GC was applied in the
steady state studies, whereas the MS was used for the transient
experiments. In the present work, the outlet of the GC column
was directly connected to a thermal conductivity detector (TCD),
frame ionization detector, and FPD. In order to satisfactorily sepa-
rate all elements, the temperature setting inside the GC column
was programmed to vary with time. In the first 3 min, the column
temperature was constant at 60 °C; it was then increased steadily
at a rate of 15 °C minꢁ1 to 120 °C and last decreased to 60 °C.
In this study, the catalyst activity was identified in terms of the
turnover frequencies, H2 yields, and other outlet gas selectivities.
The turnover frequencies can be calculated from the equation [18]
ð%COÞ
SCO ¼ 100 ꢂ
ð%COÞ þ ð%CO2Þ þ ð%CH4Þ þ 2ð%C2H6Þ þ 2ð%C2H4Þ
ð2Þ
ꢀ
ꢁ
ð%CO2Þ
SCO ¼ 100 ꢂ
2
ð%COÞ þ ð%CO2Þ þ ð%CH4Þ þ 2ð%C2H6Þ þ 2ð%C2H4Þ
ð3Þ
ꢀ
ꢁ
ð%CH4Þ
ð%COÞ þ ð%CO2Þ þ ð%CH4Þ þ 2ð%C2H6Þ þ 2ð%C2H4Þ
SCH ¼ 100 ꢂ
4
ð4Þ
ꢀ
ꢁ
2ð%C2H4Þ
SC ¼ 100 ꢂ
2H4
ð%COÞ þ ð%CO2Þ þ ð%CH4Þ þ 2ð%C2H6Þ þ 2ð%C2H4Þ
ð5Þ
ꢀ
ꢁ
2ð%C2H6Þ
SC ¼ 100 ꢂ
2H6
ð%COÞ þ ð%CO2Þ þ ð%CH4Þ þ 2ð%C2H6Þ þ 2ð%C2H4Þ
ð6Þ
It is noted that, for the studies on methane and biogas steam
reforming, the terms for C2H6 and C2H4 are eliminated.
2.3. Measurement of carbon formation
After reaction, temperature-programmed oxidation (TPO) was
used to investigate the amount of carbon formed on the spent cat-
alyst surface by introducing 10% O2 in helium, after the system was
purged with helium. The operating temperature increased from
room temperature to 1000 °C at a rate of 10 °C minꢁ1. The amount
of carbon formation on the surface of catalysts was determined by
measuring the CO and CO2 yields from the TPO results (using Mic-
rocal Origin Software). The calibrations of CO and CO2 were per-
formed by injecting a known amount of these calibration gases
from the sampling loop. In addition to the TPO method, the amount
of carbon deposition was confirmed by the carbon balance calcula-
tion, in which the amount of carbon deposition theoretically equals
the difference between the inlet hydrocarbon fuel and the outlet
carbon components (e.g., CO, CO2, CH4, and C2+).
2.4. The study of CeO2 as prereforming catalyst
In the present work, a practical application using CeO2 as a pri-
mary reforming catalyst to reform natural gas, biogas, and LPG
(without prior desulfurization) was investigated. In detail, CeO2
was applied to adsorb H2S from the feed and primarily reform hea-
vy hydrocarbons (C2+) to light hydrocarbon (i.e., CH4). The product
from this section was continuously passed though a secondary
reforming bed, where Ni/Al2O3 was packed, to complete the con-
version and maximize H2 yield. The design of this system consists
of two tubular-containing CeO2 columns and one Ni/Al2O3 column
(with diameters of 25 mm and lengths of 50 cm). In each column,
25 g of either CeO2 or Ni/Al2O3 (mixed with SiC) was packed, and
these three columns were placed in the same burner, in which
the temperature was controlled isothermally at SOFC temperature
(900 °C) for later application as IIR-SOFC. Details of system opera-
tion are presented in Section 3.5.
rNA AN
mcS
2
turnover frequencies ¼
ð1Þ
where r is the moles of CH4 (or hydrocarbons) changing per unit
time (molCH minꢁ1), NA is Avogadro’s number, AN is the area occu-
4
2
pied by an adsorbed nitrogen molecule (16.2 ꢂ 10ꢁ20 m2); it is as-
sumed that all surface sites accessible by nitrogen adsorption. mc
is the weight of catalyst used, and S is the specific surface area of
the catalyst (m2 gꢁ1). The yield of H2 production (YH ) was defined
3. Results and discussion
2
as the molar fraction of H2 produced out of the total hydrogen-
based compounds in the products. Other by-product selectivities
(i.e., SCO, SCO ; SCH ; SC ; and SC ) were defined as the mole ra-
3.1. Preliminary experiments
2H6
2H4
2
4
tios of the specified component in the outlet gas to the total car-
bon-based compounds in the product, accounting for
Preliminary experiments were carried out to find a suitable con-
dition under which internal and external mass transfer effects are

N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
9
not predominant. Considering the effect of external mass transfer,
the total gas flow rate was varied between 10 and 200 cm3 minꢁ1
for a constant residence time of 5 ꢂ 10ꢁ4 gcat min cmꢁ3. It was found
that the turnover frequencies are independent of the gas velocity
when the gas flow rate is higher than 60 cm3 minꢁ1, indicating the
absence of external mass transfer effects at this high velocity
(Fig. 2). Furthermore, the reaction on different average sizes of cat-
alyst was studied in order to ensure that the experiments were car-
ried out within the region of intrinsic kinetics. It was observed that
H2S contents of 10, 100, 500, and 1000 ppm (capturing a range of
H2S content in various fuels) were added during the reaction at
900 °C with an H2O/CH4 ratio of 3.0. As shown in Fig. 3, we found
that the presence of H2S under appropriate conditions increases
the steam reforming rate. Without H2S, the steady state turnover
frequencies were 0.054 sꢁ1. When 10 ppm H2S was added, the
turnover frequencies increased steadily with time and reached a
higher steady state value at 0.074 sꢁ1. When more H2S was added
in the feed (100 and 500 ppm), the turnover frequencies increased
more rapidly to 0.098 and 0.095 sꢁ1. Nevertheless, when as much
as 1000 ppm H2S was introduced, although the turnover frequen-
cies initially increased to 0.115 sꢁ1, they later gradually dropped
to 0.071 sꢁ1. After exposure to H2S for 3 h, the experiment was con-
tinued by removing H2S from the feed. As also seen in this figure,
the turnover frequencies (after 10 ppm H2S was removed) remain
constant, whereas after the experiments with 100 and 500 ppm
H2S the turnover frequencies were slightly reduced to 0.085 and
0.086 sꢁ1, respectively. After 1000 ppm H2S was removed, the
turnover frequencies increased considerably before decreasing to
0.088 sꢁ1. It is noted according to the error analysis that the
the catalysts with particle size less than 200 lm showed no
intraparticle diffusion limitation under the range of conditions
studied. Therefore, in the following studies, the total flow rate was
kept constant at 100 cm3 minꢁ1, whereas the catalyst diameters
were kept within the above-mentioned range in all experiments.
3.2. Activity of CeO2 toward steam reforming of CH4 in the presence of
H2S
First, the steam reforming of CH4 over nano-scale CeO2 in the
presence of various H2S contents was investigated. In our studies,
0.12
0.1
0.08
0.06
0.04
0.02
0
900C
950C
925C
975C
0
20
40
60
80
100
120
140
160
180
200
-1
Gas flow rate (cm3min )
Fig. 2. Effect of the total gas flow rate on the turnover frequencies from the steam reforming at 900–975 °C with constant residence time 5 ꢂ 10ꢁ4 g min cmꢁ3
.
0.18
50 ppm H2S
100 ppm H2S
500 ppm H2S (CeO2)
500 ppm H2S (Rh/Al2O3)
1000 ppm H2S
CeS4 = Ce(SO4)2
CeS3 = Ce2(SO4)3
CeS = Ce2O2S
CeO2 = CeO2
0.16
0.14
0.12
0.1
Rh/Al2O3
92%CeS4
8%CeO2
79%CeS4
4%CeS3
8%CeS
87%CeS4
13%CeO2
74%CeS4
26%CeO2
9%CeO2
88%CeS4
12%CeO2
0.08
0.06
0.04
0.02
0
31%CeS4
69%CeO2
68%CeS4
32%CeO2
71%CeS4
7%CeS3
14%CeS
8%CeO2
66%CeS4
34%CeO2
CeO2
59%CeS4
41%CeO2
Removal of H2S from the feed
Feeding H S along with CH and H2O
2
4
0
25
50
75
100
125
150
175
200
225
250
275
300
Time (min)
Fig. 3. Effect of H2S on the turnover frequencies from methane steam reforming and Ce–O–S phase compositions at various H2S concentrations.

10
N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
deviations of these turnover frequencies are in the range of 4.5%.
Furthermore, the turnover frequencies observed in the present
work are in good agreement with the values previously reported
in the literature [18,20].
CeO2
Ce(SO4)2
Ce2O2S
For comparison, the effect of H2S on the steam reforming activ-
ity of highly active Rh/Al2O3 was also studied by adding 500 ppm
H2S along with CH4 and H2O. It was found that, in the presence
of H2S, the turnover frequencies from the steam reforming over
Rh/Al2O3 dramatically dropped from 0.172 to 0.013 sꢁ1 in a short
time and could not be recovered even when H2S was removed from
the system (Fig. 3). It is noted that the XRD study on the spent Rh/
Al2O3 catalyst indicated the formation of rhodium sulfide, which is
rarely regenerated.
CeO2 after removal of H2S from the feed
CeO2 after reaction at 900ºC
(with 1000 ppm H2S) for 2 h
3.3. Effects of temperature and inlet H2O/CH4 ratio
As the next step, the effects of operating temperature and inlet
H2O/CH4 molar ratio on the steam reforming of CH4 over CeO2 in
the presence of H2S were determined by varying the temperature
(from 900 °C to 925, 950, and 975 °C) and inlet H2O/CH4 molar ra-
tio (from 3.0 to 2.0, 1.5, and 1.0). It was found that the influence of
H2S on the turnover frequencies over CeO2 strongly depends on the
operating temperature and inlet H2O/CH4 ratio (Fig. 4). When
100 ppm H2S was added at 900 °C, the turnover frequencies in-
creased by 81% (from 0.054 to 0.098 sꢁ1); this difference increased
to 83% (from 0.063 to 0.117 sꢁ1), 86% (from 0.075 to 0.139 sꢁ1), and
88% (from 0.089 to 0.168 sꢁ1) when the temperature increased to
925, 950, and 975 °C, respectively. In contrast, at 900 °C, when
the inlet H2O/CH4 ratio was reduced to 2.0, 1.5, and 1.0, the posi-
tive deviation of the turnover frequencies with H2S was reduced
to +53%, +25%, and +7.9%; and when an H2O/CH4 ratio below 1.0
was applied, H2S then reduced the turnover frequencies (from
0.054 sꢁ1 to 0.041, 0.038, and 0.033 sꢁ1 at inlet H2O/CH4 ratios of
0.75, 0.5, and 0.25). It is suggested from the XRD studies on spent
CeO2 after several reaction periods and conditions that this contra-
dictory effect of H2S is related to the formation of various Ce–O–S
phases: Ce(SO4)2, Ce2(SO4)3, and Ce2O2S, during the reaction (as
illustrated in Fig. 5). The relevant reactions between CeO2 and
H2S, which result in these phases, may be.
CeO2 after reaction at 900ºC
(with 1000 ppm H2S) for 30 min
Fresh CeO2
20
30
40
50
60
70
80
Degrees (2-Theta)
Fig. 5. XRD patterns of CeO2 at various reaction times.
When 10 and 100 ppm H2S was introduced during the reaction
(with an H2O/CH4 ratio of 3.0 at 900 °C), Ce(SO4)2 was formed (via
Eqs. (7) and (8)) and its proportion increased with increasing oper-
ating time (from 31% after 20 min to 68% after 2 h for 10 ppm H2S;
and from 59% after 20 min to 87% after 2 h for 100 ppm H2S). In the
experiments with 500 and 1000 ppm H2S, as well as Ce(SO4)2 for-
mation (79% and 71% for 500 and 1000 ppm H2S), Ce2(SO4)3 (4%
and 7% for 500 and 1000 ppm H2S) and Ce2O2S (8% and 14% for
500 and 1000 ppm H2S) were also observed. At higher operating
temperatures, the portion of Ce(SO4)2 phase increased consider-
ably. On the other hand, when inlet H2O/CH4 ratios less than 1.0
were applied, Ce2O2S (occurs via Eq. (9)) became the dominant
Ce–O–S phase. It has been suggested that with decreasing oxygen
fugacity, the sulfate form of ceria decomposes to CeO2 and then to
2CeO2 þ 8H2O þ 3H2S $ Ce2ðSO4Þ3 þ 11H2
Ce2ðSO4Þ þ 4H2O þ H2S $ 2CeðSO4Þ2 þ 5H2
2CeO2 þ 3H2 þ H2S $ Ce2O2S þ 2H2O
ð7Þ
ð8Þ
ð9Þ
0.18
0.16
0.14
0.12
0.1
Feeding H2S (100 ppm) along with CH4 and H2O
Removal of H2S from the feed
85%CeS4
15%CeO2
92%CeS4
8%CeO2
95%CeS4
5%CeO2
81%CeS4
19%CeO2
89%CeS4
11%CeO2
77%CeS4 23%CeO2
● 900°C (H2O/CH4=3.0) (Based case)
0.08
0.06
0.04
0.02
0
71%CeS4 7%CeS3
45%CeS4 55%CeO2
47%CeS4 12%CeS3
37%CeS 4%CeO2
15%CeS 7%CeO2
Effect of Temp.
Effect of H2O/CH4
18%CeS4 21%CeS3
59%CeS 2 %CeO2
16%CeS4
22%CeS
62%CeO2
63%CeS4
37%CeO2
925°C (H2O/CH4=3.0)
900°C (H2O/CH4=2.0)
■ 950°C (H2O/CH4=3.0) ◊ 900°C (H2O/CH4=1.5)
975°C (H2O/CH4=3.0)
25 50
900°C (H2O/CH4=0.75)
0
75
100 125
150
175
200
225
250
275
300
Time (min)
Fig. 4. Effect of H2S on the turnover frequencies from methane steam reforming and Ce–O–S phase compositions at various temperatures and H2O/CH4 ratios.

N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
11
Ce2O2S [24]. After H2S removal, XRD patterns revealed that the
Ce2O2S phase disappeared (due to the reversal of Eq. (9)), while
the portion of Ce(SO4)2 phase slightly decreased but that of the
CeO2 phase increased.
gate the lattice oxygen mobility of the samples (CeO2 and
Ce(SO4)2). The sample (200 mg) was placed in the quartz reactor
and thermally treated under
a flow of high-purity helium
(99.995%) at the desired temperatures for 1 h. Then 18O2 (in he-
lium) was multiply pulsed into the system and the outlet gases
were monitored by the MS.
3.4. Determination of the OSC and lattice oxygen mobility
We found in the studies that the TPR of Ce(SO4)2 indicated a
sharp reduction band at 470 °C and a broader band at 810 °C,
whereas for CeO2 smaller peaks were detected at slightly higher
temperatures (600 and 850 °C). The TPD under reducing conditions
showed an amount of H2S desorption approximately correspond-
ing to half of the initial coverage of sulfate relevant to the forma-
tion of Ce2O2S. After TPD, all sulfates were reduced to Ce2O2S.
The H2/H2O + H2S titration was conducted to ensure that Ce2O2S
can be reoxidized to sulfate forms. In five consecutive cycles, the
amounts of H2 uptake were nearly identical (Fig. 6), suggesting that
the redox behavior of Ce(SO4)2 and CeO2 is reversible. The amounts
of H2 uptake and H2S produced were applied to indicate the
amount and percentage of reducible oxygen in the catalysts. From
the calculation, the amount of reducible oxygen for Ce(SO4)2 was
estimated to be 1.27 mmol gꢁ1 (17.3% of total oxygen in catalyst)
compared to 0.71 mmol gꢁ1 (10.1% of total oxygen in catalyst) for
CeO2; this clearly indicates the higher OSC of Ce(SO4)2. Further-
more, regarding the 18O/16O isotope exchange study, it is known
that the exchange of 18O/16O isotopes over CeO2 surface theoreti-
cally consists of (i) homoexchange in the gas phase (18O2 (g) +
16O2 (g) ? 18O16O (g)) and (ii) heteroexchange with the participa-
Based on the results from Sections 3.1 and 3.2, we suggest that
the formation of Ce(SO4)2 during the reaction leads to a high
reforming activity, whereas the presence of Ce2O2S reduces the
activity. To test this idea, the redox properties associated with
the OSC and the mobility of lattice oxygen for Ce(SO4)2 were exam-
ined and compared to those for CeO2 by applying TPR, TPD under
reducing condition, H2/H2O + H2S titration, and 18O/16O isotope ex-
change methods. Furthermore, the ratio of Ce3+/Ce4+ under reduc-
ing and oxidizing conditions was characterized by XPS. In detail,
the TPR experiment was carried out in a 10-mm-diameter quartz
reactor, which was mounted vertically inside tubular furnace. A
type-K thermocouple was placed into the annular space between
the reactor and furnace, while another thermocouple, covering
by a closed-end quartz tube, was inserted into the middle of the
quartz reactor in order to recheck the possible temperature gradi-
ent. The sample (100 mg) was heated from 25 to 1000 °C under 5%
H2 in nitrogen with a flow rate of 50 cm3 minꢁ1, and the amount of
H2 consumed during the TPR process at different temperatures was
monitored online by the TCD and quantified by calibrating the
peak areas against the TPR of a known amount of CuO. The TPD
over Ce(SO4)2 under the reducing-condition experiment was per-
formed in the same scale of reactor with the same weight of sam-
ple as the TPR study (under 5%H2 in nitrogen), but the effluent
gases from the TPD experiment were monitored by the MS. It is
noted that the MS signals were calibrated with a known amount
of outlet gases (e.g., H2S) to determine the absolute coverages cor-
responding to these TPD signals. As for H2/H2O + H2S titration, this
experiment was conducted over Ce(SO4)2 to confirm the consecu-
tive cycles of Ce(SO4)2 M CeO2 M Ce2O2S. Similarly to the TPR and
TPD experiments, the sample (100 mg of Ce(SO4)2) was placed in
the middle of a quartz reactor packed with two layers of quartz
wool to prevent the sample from moving. After the system was
purged with helium for 1 h, known amounts of H2 and H2O + H2S
were sequentially pulsed to the reactor for five consecutive cycles
and the effluent gases were monitored with the MS. Last, an
18O/16O isotope exchange experiment was carried out to investi-
tion of oxygen atoms from CeO2 ( O
18O2 (g) + 16O2 (S) ? 18O16
(g) + 18O (S) and 18O16O (g) + 16O2 (S) ? 16O2 (g) + 18O (S)). Accord-
ing to our results at 600 °C, the production of 16O2 and 18O16O for
Ce(SO4)2 is 24 and 17%, whereas the production of 16O2 and
18O16O for CeO2 is 12 and 2%. Therefore, the homoexchange in
the gas phase is negligible under these operating conditions, since
18O16O concentration from both materials should be identical if the
exchange in the gas phase is dominant for the overall reaction [25].
Fig. 7 shows the Arrhenius plots from 18O/16O isotope exchange
studies over Ce(SO4)2 compared to CeO2. It was found that the con-
version of 18O2 increases with increasing temperature to form 16O2
and 18O16O for both materials and the production of 16O2 and
18O16O from Ce(SO4)2 is greater than that from CeO2, indicating
its higher oxygen mobility. Furthermore, the observed activation
energy from 18O/16O isotope exchange over Ce(SO4)2 is 85 kJ molꢁ1
whereas that over CeO2 was 110 kJ molꢁ1
,
.
Ce(SO4)2
2nd cycle
1st cycle
3th cycle
5th cycle
4
th cycle
CeO2
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18
Time
Fig. 6. H2 uptake from H2/H2O + H2S titration over Ce(SO4)2 and CeO2 for five cycles.

12
N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
48
46
44
42
40
38
36
34
32
30
CeO2, which lead to the higher reforming activity, can be
confirmed.
3.5. Long-term stability testing, effect of doped CeO2 (with Gd, Y, Nb,
La, and Sm), and activity toward the steam reforming of several
hydrocarbons
Ea = 85 kJ mol-1
Long-term stability testing with consecutive cycles at various
scenarios (i.e., CH4 + H2O/CH4 + H2O + H2S/CH4 + H2O and CH4 +
H2O/CH4 + H2O + H2S/O2 with different inlet H2S concentrations)
was carried out (Fig. 8). Over 50 h with 10 cycles, good stability
and reversibility were observed for all conditions. Furthermore,
the effect of doping CeO2 with Gd, Y, Nb, La, and Sm on the contra-
dictory behavior of H2S was also studied, since doping with these
rare earths is known to affect the OSC of CeO2 [26,27]. It was first
observed that doping with La, Sm, Gd, and Y increased the turnover
frequencies of CeO2 (which relates to their OSC improvement),
whereas doping with Nb shows an inhibitory effect due to the
strong segregation of Nb from the CeO2 surface. It was then found
that the effect of H2S on doped and undoped CeO2 followed the
same trend but with different magnitude depending on the OSC
of materials. Based on TPR and H2–O2 titration testing, the amounts
of reducible oxygen over doped CeO2 with Gd, Y, Nb, La, and Sm
were 0.93, 0.82, 0.45, 1.09, and 0.95 mmol gꢁ1 compared to
0.71 mmol gꢁ1 for undoped CeO2. These values are closely related
to the turnover frequencies of the materials, in that the order of
turnover frequencies without and with 100 ppm H2S is La–CeO2
(0.083 sꢁ1/0.121 sꢁ1) > Sm–CeO2 (0.071 sꢁ1/0.113 sꢁ1) > Gd–CeO2
Ea = 110 kJ mol-1
1.5
1.7
1.9
2.1
2.3
2.5
2.7
1000/T
Fig. 7. Arrhenius plots from 18O/16O isotope exchange studies over (s) Ce(SO4)2
and (d) CeO2.
Last, the XPS studies were also carried out to quantify the Ce4+
and Ce3+ levels of reduced and oxidized states of sulfate-form sam-
ples (Ce2O2S and Ce(SO4)2) compared to reduced and oxidized
states of CeO2. The XPS spectra implied that, under reducing condi-
tions, the contents of Ce3+ for the sulfate-form sample and CeO2
were 34.6% and 24.9%, respectively, while under oxidizing condi-
tions, the contents of Ce3+ for the sulfate-form sample and CeO2
were 17.3% and 19.8%, respectively. These results suggest that
CeO2 in sulfate form promotes the reduction of Ce4+ to Ce3+ under
reducing conditions and the oxidation of Ce3+ to Ce4+ under oxidiz-
ing conditions. It is noted that these results should be further con-
firmed by an in situ XPS study, which can determine the Ce4+/Ce3+
values during the reaction. Nevertheless, based on all characteriza-
tion results, the greater redox properties of Ce(SO4)2 compared to
(0.069 sꢁ1/0.109 sꢁ1) > Y–CeO2
(0.061 sꢁ1/0.104 sꢁ1) > undoped
CeO2 (0.054 sꢁ1/0.098 sꢁ1) > Nb–CeO2 (0.041 sꢁ1/0.066 sꢁ1).
With regard to more practical applications, the steam reforming
of sulfur-containing hydrocarbons, i.e., natural gas (67%CH4,
8.6%C2H6, 4.5%C3H8, 2.5%C4H10, and 14.5%CO2 with 50 ppm H2S sup-
0.12
0.1
(a)
0.08
0.06
10 ppm H2S
0.04
10 Cycles of:
100 ppm H2S
1000 ppm H2S
CH4+H2O / CH4+H2O+H2S / CH4+H2O
0.02
0
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
0.12
0.1
(b)
0.08
0.06
0.04
0.02
0
10 Cycles of:
CH4+H2O / CH4+H2O+H2S / O2
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Time (h)
Fig. 8. Prolonged testing with 10 consecutive cycles of CH4 + H2O/CH4 + H2O + H2S/CH4 + H2O (a) and CH4 + H2O/CH4 + H2O + H2S/O2 (b) at three different H2S concentrations.

N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
13
Table 2
Turnover frequencies, H2 yield, and other outlet gas selectivities from the steam reforming of natural gas, LPG, and biogas over CeO2 at several temperatures and S/C molar ratios
(with and without prior desulfurization).
Fuel
Temp. (°C)
S/C ratio
Turnover frequencies (sꢁ1
)
H2 yield (%)
By-product selectivity (%)
CH4
C2H6
C3H8
C4H10
CO
CO2
CH4
C2+
Natural gas
900
925
950
1000
900
900
900
900
3.0
3.0
3.0
3.0
2.0
1.5
1.0
0.5
0.070 (0.043)a
0.085 (0.052)
0.105 (0.059)
0.138 (0.078)
0.070 (0.043)
0.056 (0.043)
0.049 (0.043)
0.043 (0.038)
0.018 (0.017)
0.019 (0.017)
0.021 (0.018)
100 (0.019)
0.018 (0.017)
0.018 (0.017)
0.017 (0.017)
0.017 (77)
0.010 (0.010)
0.010 (0.010)
0.010 (0.010)
0.010 (0.010)
0.010 (0.010)
0.010 (0.010)
0.010 (0.010)
0.010 (0.010)
0.006 (0.006)
0.006 (0.006)
0.006 (0.006)
0.006 (0.006)
0.006 (0.006)
0.006 (0.006)
0.006 (0.006)
0.006 (0.006)
49 (32)
57 (37)
69 (41)
81 (50)
48 (30)
43 (28)
37 (26)
28 (24)
37 (20)
48 (26)
57 (32)
65 (40)
39 (23)
33 (25)
31 (26)
29 (28)
25 (18)
23 (16)
21 (15)
19 (11)
23 (15)
20 (12)
12 (10)
9 (3)
38 (59)
29 (58)
22 (53)
16 (49)
38 (59)
47 (60)
55 (60)
59 (63)
0 (3)
0 (0)
0 (0)
0 (0)
0 (3)
0 (3)
2 (4)
3 (6)
C3H8
C4H10
CO
CO2
CH4
C2+
LPG
900
925
950
1000
900
900
900
900
3.0
3.0
3.0
3.0
2.0
1.5
1.0
0.5
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
0.108 (0.108)
69 (61)
76 (65)
83 (68)
89 (71)
65 (58)
59 (55)
54 (52)
45 (49)
58 (50)
65 (57)
72 (64)
78 (71)
57 (52)
55 (54)
53 (55)
51 (56)
27 (21)
24 (19)
20 (17)
17 (14)
21 (18)
18 (15)
13 (10)
6 (7)
13 (21)
10 (19)
8 (18)
2 (8)
0.2 (5)
0 (1.4)
0 (0)
5 (8)
7 (9)
5 (15)
17 (22)
20 (22)
24 (25)
30 (26)
10 (10)
13 (11)
CO
CO2
CH4
Biogas
900
925
950
1000
900
900
900
900
3.0
3.0
3.0
3.0
2.0
1.5
1.0
0.5
0.071 (0.054)
43 (33)
50 (38)
58 (40)
75 (51)
37 (31)
31 (29)
24 (28)
21 (25)
27 (24)
34 (30)
42 (37)
54 (45)
26 (27)
25 (30)
23 (33)
21 (35)
33 (31)
31 (29)
29 (26)
26 (24)
32 (28)
31 (25)
30 (21)
30 (18)
40 (45)
35 (41)
29 (37)
20 (31)
42 (45)
44 (45)
47 (46)
49 (47)
0.088 (0.067)
0.108 (0.078)
0.147 (0.101)
0.062 (0.054)
0.056 (0.054)
0.047 (0.052)
0.041 (0.049)
a
Values in blanket are those observed from the steam reforming of feedstocks with prior desulfurization.
Fig. 9. Design and operation of CeO2 as prereforming catalyst in the present work.

14
N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
plied by PPT Plc., Thailand), LPG (60%C3H8, 40%C4H10 with 100 ppm
H2S), and biogas (60%CH4, 40%CO2 with 1000 ppm H2S) with and
without desulfurization over CeO2 was examined. Table 2 presents
the catalytic activity in terms of turnover frequencies, H2 yield,
and by-product selectivities for various operating conditions. At
suitable conditions, i.e., inlet S/C ratio 3.0, higher turnover frequen-
cies could be achieved from the steam reforming of these feedstocks
without desulfurization. Furthermore, after reaction for 18 h, the
amounts of carbon deposition under each condition were analyzed
by TPO and carbon balance calculation. It was found that, with an in-
let S/C molar ratio of 3.0, the observed carbon deposition from the
steam reforming of natural gas, LPG, and biogas (at 900 °C) was in
the range of 0.22–0.26, 0.62–0.65, and 0.04–0.07 mmol gꢁca1t, respec-
tively. These low amounts of carbon deposition indicate the excel-
lent resistance toward carbon deposition of CeO2.
bons (mainly CH4) were detected in the product, indicating incom-
plete conversion. Thus, we proposed the pairing of CeO2 with a
suitable metallic catalyst to achieve the benefits of self-desulfur-
ization, improved resistance to carbon deposition, and higher
reforming activity. For the approach in the present work, CeO2
was applied as a prereforming catalyst in order to adsorb H2S from
the feed and primarily reform heavy hydrocarbons (C2+) in the feed
to CH4. The product from this section was continuously passed
though a secondary reforming bed, where Ni/Al2O3 was packed,
to complete the conversion and to maximize H2 yield. Design
and operation of this prototype system are shown in Fig. 9. As
for the operation, the primary fuels (i.e., natural gas, LPG, and bio-
gas) were mixed with steam and flowed past a switching valve
through the first CeO2 column to remove all H2S and partially re-
form all heavy hydrocarbons to CH4. At the end of this tube, the
gas mixture flowed backward past another switching valve to the
Ni/Al2O3 column to complete the reforming reaction. At proper
Nevertheless, as seen from this table, CeO2 alone as a reforming
catalyst gives relatively low reforming activity and some hydrocar-
100
80
From 1st CeO2 column
▲
○ From 2rd CeO2 column
60
40
Primary fuel
: Natural Gas
Switching time: 15 h
20
0
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
100
80
60
40
20
0
Primary fuel
Switching time: 12 h
: LP G
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
100
80
60
40
20
0
Primary fuel
Switching time: 5 h
: biogas
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Time (h)
Fig. 10. H2 yield produced from the use of CeO2 as prereforming catalyst fed by different primary fuels at 900 °C with appropriate switching times (from first CeO2 column to
second CeO2 column).
Table 3
Activity in terms of turnover frequencies, H2 yield, other outlet gas selectivities, and the amount of carbon formation from the steam reforming of natural gas, LPG, and biogas
(using CeO2 as primary reforming catalyst and Ni/Al2O3 as secondary reforming catalyst) at 900 °C with S/C molar ratio 3/1.
Turnover frequencies (sꢁ1
)
H2 yield (%)
By-product selectivity (%)
ꢁ1
Fuel
Regeneration time (h)
Carbon formation (mmol g
)
cat
CH4
C2H6
C3H8
C4H10
CO
CO2
CH4
C2H4
Natural gas
LPG
Biogas
15
12
5
0.173
–
0.214
0.022
–
–
0.011
0.108
–
0.006
0.108
–
99.7
99.0
99.7
66.3
74.6
61.3
33.5
25.0
38.5
0.2
0.4
0.2
0
0
–
0.36a (0.33)b
0.91 (0.95)
0.09 (0.12)
a
Calculated using CO and CO2 yields from the TPO study.
Calculated from the balance of carbon-based compounds in the system.
b

N. Laosiripojana et al. / Journal of Catalysis 276 (2010) 6–15
15
exposure times (5 h for biogas, 12 h for LPG, and 15 h for natural
Acknowledgments
gas, where H2S start came out in the outlet gas), two switching
valves as mentioned above automatically switched the port direc-
tion and the primary fuels were flowed through the second CeO2
column instead. Simultaneously, the air was purged through the
first CeO2 column to remove all sulfur elements in the column
and vent out from the system. In our study, the switching process
was repeated five times or for 50 h without detection of any activ-
ity deactivation, as shown in Fig. 10. The results in Table 3 also
indicate that almost 100% H2 yield could be achieved from all feed-
stocks; furthermore, low amounts of carbon deposition (less than
1.0 mmol gꢁca1t) were detected from all reactions after the prolonged
testing. We concluded that the novelties of this reforming unit are
the flexibility of inlet fuels and the nonrequirements for a desulfur-
ization unit, a separate prereforming unit, and/or the use of expen-
sive noble metal catalysts to reform sulfur-containing heavy
hydrocarbon feedstocks. Importantly, this reforming unit would
promote the practical application of IIR-SOFC, particularly with
the use of a sulfur-tolerant SOFC anode, by eliminating the require-
ments for costly desulfurization units.
Financial support from the Thailand Research Fund (TRF)
throughout this project is gratefully acknowledged. The authors
acknowledge Dr. John T.H. Pearce for the English correction of this
manuscript.
References
[1] M.Z. Jacobson, W.G. Colella, D.M. Golden, Science 308 (2005) 901.
[2] J.R. Rostrup-Nielsen, J. Sehested, J.K. Nørskov, Adv. Catal. 47 (2002) 65.
[3] J.R. Rostrup-Nielsen, T.S. Christensen, I. Dybkjaer, Stud. Surf. Sci. Catal. 113
(1998) 81.
[4] G.A. Deluga, J.R. Salge, L.D. Schmidt, X.E. Verykios, Science 303 (2004) 993.
[5] G.W. Huber, J.W. Shabaker, J.A. Dumesic, Science 300 (2003) 2075.
[6] J.R. Rostrup-Nielsen, Stud. Surf. Sci. Catal. 147 (2004) 121.
[7] G. Jones, J.G. Jakobsen, S.S. Shim, J. Kleis, M.P. Andersson, J. Rossmeisl, F. Abild-
Pedersen, T. Bligaard, S. Helveg, B. Hinnemann, J.R. Rostrup-Nielsen, I.
Chorkendorff, J. Sehested, J.K. Nørskov, J. Catal. 259 (2008) 147.
[8] A. Grirrane, A. Corma, H. García, Science 322 (2008) 1661.
[9] J.A. Rodriguez, S. Ma, P. Liu, J. Hrbek, J. Evans, M. Pérez, Science 318 (2007)
1757.
[10] A. Trovarelli, Catal. Rev. Sci. Eng. 38 (1996) 439.
[11] P. Fornasiero, G. Balducci, R.D. Monte, J. Kaspar, V. Sergo, G. Gubitosa, A.
Ferrero, M. Graziani, J. Catal. 164 (1996) 173.
[12] T. Miki, T. Ogawa, M. Haneda, N. Kakuta, A. Ueno, S. Tateishi, S. Matsuura, M.
Sato, J. Phys. Chem. 94 (1990) 339.
4. Conclusions
[13] C.T. Campbell, C.H.F. Peden, Science 309 (2005) 713.
[14] T. Hibino, A. Hashimoto, T. Inoue, J. Tokuno, S. Yoshida, M. Sano, Science 288
(2000) 2031.
The presence of appropriate H2S content makes it possible to
promote the steam reforming rate of nano-scale CeO2 and doped
CeO2 (with Gd, Y, Nb, La, and Sm). This contradictory effect was re-
lated to the formation of various Ce–O–S phases (i.e., Ce(SO4)2,
Ce2(SO4)3, and Ce2O2S) during the reaction. It was revealed that
the formation of Ce(SO4)2 promotes the OSC, the lattice oxygen
mobility, and eventually the reforming activity, whereas the for-
mation of Ce2O2S reduces the OSC, the lattice oxygen mobility,
and the reforming rate. The capability of these CeO2 catalysts to ad-
sorb and use poisoning H2S gas to enhance its catalytic activity of-
fers great benefits in terms of energy and environmental
management, and importantly, this behavior would help to pro-
mote the hydrogen/fuel cell economy by eliminating the require-
ments for costly desulfurization units.
[15] Z. Zhan, S.A. Barnett, Science 308 (2005) 844.
[16] R.J. Gorte, J.M. Vohs, S. McIntosh, Solid State Ion. 175 (2004) 1.
[17] O. Costa-Nunes, R.J. Gorte, J.M. Vohs, J. Power Sources 141 (2005) 241.
[18] E. Ramírez-Cabrera, A. Atkinson, D. Chadwick, Appl. Catal. B 47 (2004) 127.
[19] D.J.L. Brett, A. Atkinson, D. Cumming, E. Ramírez-Cabrera, R. Rudkin, N.P.
Brandon, Chem. Eng. Sci. 60 (2005) 5649.
[20] N. Laosiripojana, S. Assabumrungrat, Appl. Catal. B 60 (2005) 107.
[21] N. Laosiripojana, S. Assabumrungrat, Appl. Catal. B 82 (2008) 103.
[22] M. Flytzani-Stephanopoulos, M. Sakbodin, Z. Wang, Science 312 (2006) 1508.
[23] D. Terribile, A. Trovarelli, J. Llorca, C. Leitenburg, G. Dolcetti, J. Catal. 178 (1998)
299.
[24] D.A.R. Kay, W.G. Wilson, V. Jalan, J. Alloys Compd. 192 (1993) 11.
[25] Y. Zeng, S. Kaytakoglu, D.P. Harrison, Chem. Eng. Sci. 55 (2000) 4893.
[26] H. He, H.X. Dai, K.W. Wong, C.T. Au, Appl. Catal. A 251 (2003) 61.
[27] L. Cao, C. Ni, Z. Yuan, S. Wang, Catal. Commun. 10 (2009) 1192.