Room temperature hydrogen production from ethanol over CeNiXHZOY nano-oxyhydride catalysts

Article abstract of DOI:10.1002/cctc.201300087

CeNi_XH_ZO_Y nano-oxyhydride catalysts were developed for the highly efficient sustainable hydrogen production from ethanol and water in the oxidative steam reforming reaction. After an insitu treatment in hydrogen in the temperature range of 200-300°C, the cerium-nickel binary mixed oxides became hydrogen reservoirs, which were called oxyhydrides, in the presence of hydrogen species of the hydride nature in the anionic vacancies of mixed oxides. A novel technology was developed for the room temperature hydrogen production by using the chemical energy released from the reaction between CeNi_XH_ZO_Y nano-oxyhydride catalysts and oxygen, which completely converted ethanol specifically at 60°C (oven temperature) and simultaneously produced hydrogen, carbon dioxide, and carbon monoxide along with small amounts of methane and ethanal. CeNi_XH_ZO_Y nano-oxyhydride catalysts demonstrated excellent catalytic stability, which was attributed to the graphitic filamentous carbon formed during the reaction. The unique activation phenomenon of the reaction (a huge variation in the temperature between the catalyst bed and the oven) was demonstrated in detail. Finally, the correlations among the catalyst properties, the catalytic performances, and the characterizations were thoroughly discussed.

Full text of DOI:10.1002/cctc.201300087

CHEMCATCHEM
FULL PAPERS
FULL PAPERS
W. Fang, C. Pirez, S. Paul, M. Capron,
H. Jobic, F. Dumeignil,
L. Jalowiecki-Duhamel*
&& – &&
Room Temperature Hydrogen
Production from Ethanol over
CeNiXHZOY Nano-Oxyhydride Catalysts
Taking alcohol from the nanobar:
CeNiXHZOY nano-oxyhydride catalysts
enable us to completely convert ethanol
at 608C and produce hydrogen
(45 mol% in the gas phase products)
through the oxidative steam reforming
reaction. The oxyhydrides are obtained
through the in situ activation of CeNiXOY
nanocompounds in hydrogen at 2508C.
This technology can save energy by
using the strong exothermic chemical
reaction between the hydride species in
the catalyst and O2.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&1
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
DOI: 10.1002/cctc.201300087
Room Temperature Hydrogen Production from Ethanol
over CeNiXHZOY Nano-Oxyhydride Catalysts
Wenhao Fang,[a] Cyril Pirez,[a] Sꢁbastien Paul,[a, b] Mickaꢂl Capron,[a] Hervꢁ Jobic,[c]
Franck Dumeignil,[a, d] and Louise Jalowiecki-Duhamel*[a]
CeNiXHZOY nano-oxyhydride catalysts were developed for the
highly efficient sustainable hydrogen production from ethanol
and water in the oxidative steam reforming reaction. After an
in situ treatment in hydrogen in the temperature range of
200–3008C, the cerium–nickel binary mixed oxides became hy-
drogen reservoirs, which were called oxyhydrides, in the pres-
ence of hydrogen species of the hydride nature in the anionic
vacancies of mixed oxides. A novel technology was developed
for the room temperature hydrogen production by using the
chemical energy released from the reaction between CeN-
iXHZOY nano-oxyhydride catalysts and oxygen, which complete-
ly converted ethanol specifically at 608C (oven temperature)
and simultaneously produced hydrogen, carbon dioxide, and
carbon monoxide along with small amounts of methane and
ethanal. CeNiXHZOY nano-oxyhydride catalysts demonstrated
excellent catalytic stability, which was attributed to the graph-
itic filamentous carbon formed during the reaction. The
unique activation phenomenon of the reaction (a huge varia-
tion in the temperature between the catalyst bed and the
oven) was demonstrated in detail. Finally, the correlations
among the catalyst properties, the catalytic performances, and
the characterizations were thoroughly discussed.
Introduction
Hydrogen is widely regarded as a promising future energy
source to address a series of crisis of the oil shortage, climate
change, and environmental pollution, which derives from the
exploitation and utilization of fossil fuels.[1–5] Hydrogen energy
is critical to any clean energy policy based on its high-energy
capacity and environmentally friendliness. Globally, the steam
reforming of natural gas, especially methane, is still the most
commonly used and the most economically competitive
method for H2 production. To support a clean and sustainable
H2 economy, it is urgently desirable to produce H2 from renew-
able energy sources, such as biomass or biomass-derived ma-
terials.[6–10] The current mature biomass-fermentation technolo-
gy makes bioethanol a highly available and low-cost feedstock
for H2 production.[11–13] Of the platform molecules derived from
biomass, bioethanol has the advantages of high H2 content,
low toxicity, and ease of delivering. In addition, as bioethanol
is a mixture of ethanol and water, it involves water, which is
the most renewable, abundant, and cleanest H2 carrier, in the
H2 generation process, for instance, the steam reforming of
ethanol [SRE; Eq. (1)].
C2H5OH þ 3 H2O ! 6 H2 þ 2 CO2
C2H5OH þ 1:5 O2 ! 3 H2 þ 2 CO2
DH0
¼ 174 kJ molꢀ1
298K
ð1Þ
DH0
¼ ꢀ545 kJ molꢀ1
298K
ð2Þ
In theory, the SRE reaction provides an almost closed carbon
recycle because all CO2 produced during the reforming process
can be fed back to biomass through the photosynthesis reac-
tion. In reality, the strongly endothermic SRE reaction is not fa-
vorable thermodynamically at low temperatures [Eq. (1)] and
essentially requires an extra energy input.[14,15] Such high-tem-
perature operations (ꢁ6508C) not only consume plenty of
energy but also cause inconvenience to the application of H2
to fuel cells.[16-18] On the contrary, the energy needed can be
supplied by adding some O2 or air to the feed. Thus, the exo-
thermic partial oxidation of ethanol [POE; Eq. (2)] can occur
thermodynamically at low temperatures,[11,19,20] during which
a portion of ethanol is burned to produce the required energy
so as to simultaneously achieve the SRE reaction [Eq. (1)].[1,21-24]
In such a context, a technology combining the SRE and POE
processes, generally called the oxidative steam reforming of
ethanol (OSRE), would be promising for the onboard H2 pro-
duction, that is, for the mobile applications that require high
H2 selectivity and rapid response. Moreover, the OSRE reaction
could reduce the carbon deposition and thus the addition of
[a] W. Fang, Dr. C. Pirez, Prof. Dr. S. Paul, Dr. M. Capron, Prof. Dr. F. Dumeignil,
Dr. L. Jalowiecki-Duhamel
Universitꢀ Lille Nord de France
CNRS UMR8181, Unitꢀ de Catalyse et Chimie du Solide
59655 Villeneuve d’Ascq Cedex (France)
E-mail: louise.duhamel@univ-lille1.fr
[b] Prof. Dr. S. Paul
Ecole Centrale de Lille
59655 Villeneuve d’Ascq (France)
[c] Dr. H. Jobic
IRCELyon Institut de Recherches sur la Catalyse et I’Environnement de Lyon
2 Avenue Albert Einstein, 69626 Villeurbanne Cedex (France)
[d] Prof. Dr. F. Dumeignil
Institut Universitaire de France
Maison des Universitꢀs
103 Boulevard Saint-Michel, 75005 Paris (France)
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
2
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
O2 to the feed enables the elimination of carbon on catalyst
particles.[25]
catalytic bed and the oven) is demonstrated in detail in the
present study. The correlations among the catalyst properties,
the catalytic performances, and the characterizations are dis-
cussed to better understand this new technology.
The OSRE reaction is generally studied in the catalytic fixed-
bed reactors[26–28] or the microchannel membrane reactors.[29–31]
The critical factor is the development of highly efficient, stable,
and low-cost catalysts, that is, the design of the appropriate
active sites that fulfill the above-mentioned features of the
OSRE reaction, especially the design of the non-noble metal
catalysts.[32–34] Till date, some metal catalytic fixed-bed systems
have been reported for the autothermal reforming of ethanol,
generally the OSRE reaction, such as the oxide-supported
active metal catalysts[11,24,26,27,35] and the Ni-, Co-, and Cu-based
mixed oxides.[36,37] There is no doubt that the Ni-based catalyst
is the most attractive because of its good activity in the cleav-
age of the CꢀC and CꢀH bonds.[38] However, the drawback of
these systems is still reflected in the thermal management. An
ideal OSRE process would develop a catalyst working at low
temperatures to build long life and safe reformers, which
could be used for portable applications.[39,40]
Results and Discussion
Catalytic behavior of the CeNiXHZOY nano-oxyhydride cata-
lysts for the OSRE reaction
After an in situ treatment in H2 at 2508C, the active CeNiXHZOY
nano-oxyhydride catalysts obtained are studied for H2 produc-
tion from ethanol in the presence of water and O2. The reac-
tants with
a
molar composition of H2O/EtOH/O2/N2 =
3:1:1.6:1.3 are introduced into the reactor in a specific order
(given in the Experimental Section). The reaction can be acti-
vated at a certain temperature, from which the temperature of
the catalytic bed (TCat.) increases quickly until it can no longer
increases within a few minutes. The initial and final tempera-
tures in such an activation phenomenon are recorded as TIni.
and TFin. At the same moment, the heating supply is lowered
to room temperature and the variation in the reaction temper-
ature between the catalytic bed and the oven (DTR =
TCat.ꢀTOven) is measured precisely when the catalytic system is
stable.
In our previous communication, we reported a novel route
for the room temperature H2 production on the CeNiHZOY
nano-oxyhydride catalyst.[28] This promising technology can
save energy by means of a combination of the exothermic re-
action between the hydride species stored in the catalyst and
O2 (chemical energy) and the exothermic reaction between
ethanol and O2 (POE reaction). The CeNiHZOY nano-oxyhydride
catalyst can simultaneously activate ethanol, produce H2 at
room temperature, and provide hydride species to sustain the
chemical reaction with O2, which is a remarkable advancement
in the field. In our laboratory, CeNiXOY nanocompounds have
been studied largely in the SRE reaction because of the strong
interactions that exist between the Ni and Ce species in the
solid.[15,41–43] As a fluorite-type oxide, CeO2 has been used in
various reactions so as to take advantage of its O2 storage ca-
pacity and/or O2 diffusion property. The release and uptake of
O2 by CeO2 enable CeO2 to participate in the redox reactions,
and the increase in the formation of oxygen vacancies attribut-
ed to the reduction of Ce cations has been shown to be of im-
The gas phase product distribution in the OSRE reaction on
the CeNi1HZOY (Ni loading 24 wt%) catalyst are shown in
Figure 1 as an example. Complete ethanol conversion is ach-
ieved, which is accompanied by the H2 formation of approxi-
portance to catalytic activity.[44,45] A redox process among Ce4+
,
Ce3+, Ni0, and Ni2+ species has been proposed.[46–49] Thus, the
active Ni species in CeNiXOY nanocompounds are characteristi-
cally and easily reduced and reoxidized owing to their close in-
teractions with the Ce species.
Figure 1. Ethanol conversion and product distribution in the oxidative steam
reforming of ethanol reaction on the CeNi1HZOY catalyst. Ethanol conversion
~
&
^
^ *
*
(
), H2 ( ), CO2 ( ), CO ( ), CH4 ( ), and CH3CHO ( ). Reaction conditions:
Not only the CeNiHZOY (Ni loading 20 wt%) nano-oxyhydride
catalyst but also the CeNiXHZOY oxyhydride family could dem-
onstrate catalytic activity in the OSRE reaction, which depends
on the Ni content. Furthermore, the CeNiXHZOY oxyhydrides
with various Ni contents have different hydrogen storage ca-
pacities in terms of hydride species,[46,47] which are strongly as-
sociated with the chemical energy (O2) consumed to sustain
the OSRE reaction at room temperature. For this reason, the
effect of the Ni content is likely to bring about interesting cata-
lytic behaviors for the OSRE reaction. Herein, we report the
CeNiXHZOY oxyhydrides as efficient and stable catalysts for H2
production from ethanol at room temperature. Complementa-
ry to our previous communication, the unique reaction phe-
nomenon (a huge variation in the temperature between the
catalyst 30 mg, H2O/EtOH/O2/N2 =3:1:1.6:1.3, TOven =608C, and TCat. =3458C.
mately 45 mol% relative to all the gas phase products (dry
basis). The other products analyzed are mainly CO2 (41 mol%)
and CO (12 mol%); small amounts of CH4 and CH3CHO are also
detected. This is still the best result that has ever been report-
ed for a low-cost catalyst. The reaction temperature (TCat.) is
measured at 3458C, whereas the oven temperature remains at
only 608C, which is much lower than that required. The energy
released from the reaction between the hydride species stored
in the oxyhydride catalyst and O2 (chemical energy) and from
the exothermic POE reaction is combined to provide the
power necessary to convert ethanol and produce H2. Thus,
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
3
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
only a little extra energy is needed to maintain the reaction.
The CeNi1HZOY catalyst demonstrates excellent catalytic stabili-
a higher Ni content. The CO2 molar fraction presents a mini-
mum at Ni/MT ꢂ0.4.
ty after 75 h of the reaction even if a small amount of carbon
Complementary to Equations (1) and (2) leading to the for-
mation of H2 and CO2, different byproducts can be obtained,
owing to the co-occurrence of different reactions that are also
involved. For example, ethanol can be decomposed into a mix-
ture of CO, CH4, and H2 [Eq. (3)] at a low temperature on the
catalysts with a strong capability of the CꢀC bond rupture,
such as the Ni catalyst.[42,43,50] The water-gas shift reaction that
is slightly exothermic is favored at low temperature and ena-
bles the transformation of CO [Eq. (4)]. The dehydrogenation
of ethanol [Eq. (5)] and the decomposition of CH3CHO [Eq. (6)]
can occur at the intermediate temperatures (300–3508C).
Moreover, in the presence of O2, the partial oxidation of meth-
ane [Eq. (7), which is also exothermic] can lead to the forma-
tion of H2 and CO (or CO2 at a higher O2 concentration).
ꢀ1
is formed (with a formation rate of 28 mgh
g
ꢀ1) after the
cat
reaction. A stable activity in the SRE reaction on such a type of
catalyst was attributed to the relatively homogeneous filamen-
tous carbon species formed.[43]
Under the same OSRE conditions, a high and stable conver-
sion of approximately 100% has already been reported, with
the main gas phase products of H2 (44 mol%), CO2 (31 mol%),
and CO (21 mol%) accompanied by approximately 1 mol% of
CH4 and 3 mol% of CH3CHO.[28] Herein, the product distribution
is improved, which can be probably due to 1) a higher Ni con-
tent of 24 wt% and 2) a larger variation in the reaction temper-
ature (DTR). Instead of a TCat. of 2808C obtained previously,
a higher TCat. of 3458C is observed here. The carbon formation
rate also decreases compared to the former rate of
ꢀ1. CO formation rate is also found to decrease,
ꢀ1
C2H5OH ! CO þ CH4 þ H2
CO þ H2O ! H2 þ CO2
C2H5OH ! CH3CHO þ H2
CH3CHO ! CO þ CH4
ð3Þ
ð4Þ
ð5Þ
ð6Þ
ð7Þ
63 mgh
g
cat
whereas CO2 formation rate increases.
The ethanol conversion and the gas phase product composi-
tion as a function of the Ni content in the OSRE reaction after
20 h are shown in Figure 2. The results obtained in our previ-
ous communication are also reported for comparison.[28] Etha-
nol is completely converted, with an activity constantly higher
CH4 þ 0:5 O2 ! CO þ 2 H2
The amount of carbon formed is measured after the reac-
tion, and it is strongly associated with the Ni content
(Figure 2). The CeNi1HZOY catalyst forms the smallest amount
ꢀ1
ꢀ1
of carbon with a formation rate of 79 mgh
g
after 20 h of
cat
the reaction. With exactly the same catalyst, a much smaller
amount of carbon is formed with a formation rate of
ꢀ1
ꢀ1
28 mgh
g
after 75 h. The result reveals that carbon is not
cat
linearly generated with the time on stream; in contrast, the
formation rate is relatively in inverse proportion to the reaction
time. It indicates that most of the carbon is deposited in the
initial stage and may be before introducing O2. In reality,
carbon can be formed through the total decomposition of eth-
anol. Moreover, solid carbon can also be generated through
the disproportionation of CO and/or the decomposition of
CH4.[14,51]
Figure 2. Effect of the Ni content on the oxidative steam reforming of etha-
~
^
*
*
nol reaction on CeNiXHZOY catalysts. H2 ( ), CO2 ( ), CO ( ), CH4 ( ),
~
&
CH3CHO ( ), and carbon formation ( ). Reaction conditions: catalyst 30 mg,
H2O/EtOH/O2/N2 =3:1:1.6:1.3, TOven =608C, and time 20 h. Ethanol conversion
is approximately 100% independent of the Ni content. For comparison, the
result reported in our previous paper is represented by blue symbols.[28]
To understand the product distribution, it is necessary to in-
vestigate the effect of the Ni content on the activation phe-
nomenon, which plays a dominant role in the OSRE reaction.
As shown in Figure 3, when the oven temperature is set at
608C, the highest reaction temperature of 3458C is measured
on the CeNi1HZOY catalyst whereas the lowest reaction temper-
ature of 3018C is measured on the CeNi0.3HZOY catalyst. Nota-
bly, no variation in the reaction temperature could be ob-
served under exactly the same reaction conditions without the
pretreatment of the catalyst in H2. Moreover, the total conver-
sion of ethanol is obtained regardless of the Ni content, where-
as the variation in the temperature depends on the Ni content.
Therefore, the exothermic POE cannot justify the reaction at
room temperature. The activation phenomenon is mostly at-
tributed to the exothermic reaction between the hydride spe-
cies of the solid and O2. Thus, the variation in the reaction tem-
perature depends on the concentration of the hydrogen
than 97% on CeNiXHZOY catalysts with various Ni contents.
Only small amounts of CH4 and CH3CHO are detected in outlet
gases. The total molar fraction of these two products decreases
slightly with an increase in the Ni content. The other products
in the gas phase do not show monotonic change with
a change in the Ni content. Approximately 32% H2 is formed
on the CeNi0.3HZOY catalyst (Ni/MT =0.23). Afterward, the H2
molar fraction increases with the Ni content and reaches an
optimum value of approximately 45% on the CeNi1HZOY cata-
lyst. If the Ni/MT ratio increases to 0.85 (CeNi5HZOY), H2 decreas-
es slightly to 40%. CO formed is approximately 19% at Ni/
MT =0.2 and reaches the maximum value of 24% at Ni/MT
ꢂ0.3, followed by a rapid decrease to approximately 11% at
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
4
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Table 1. Ni loading, specific surface area, and crystallite size of CeNiXOY
nanocompounds.
Sample
Ni loading
[wt%]
Ni/MT
SBET
[m2 gꢀ1
dNiO
[nm]
dCeO
2
]
[nm]
CeNi0.02OY
CeNi0.3OY
CeNi0.5OY
CeNi1OY
CeNi5OY
0.3
7.9
12.2
24.0
52.8
0.02
0.23
0.31
0.50
0.85
76
79
109
119
84
–
–
8
11
10
6
5
5
4
4
Figure 3. Variation in temperatures versus Ni content in the oxidative steam
~
the mixed oxides have a relatively large surface area, which is
no less than 75 m2 gꢀ1; CeNi0.5OY and CeNi1OY particularly pres-
ent the highest specific areas in the range of approximately
110–120 m2 gꢀ1.
^
), TIni.
reforming of ethanol reaction on CeNiXHZOY catalysts. TCat.
(
(
), and
&
TFin. ( ). Reaction conditions: catalyst 30 mg, H2O/EtOH/O2/N2 =3:1:1.6:1.3,
TOven =608C, and time 20 h.
stored in the solid as hydride species. In reality, it has been dis-
closed by a chemical titration technique that the CeNi1OY
nanocompound possesses the highest hydrogen storage ca-
pacity among the CeNiXOY series of catalysts.[47,48] The evolution
of DTR is found to be quite similar with the curve of H2 forma-
tion versus the Ni content. A part of hydride species are con-
sumed by O2 to provide enough chemical energy to drive the
catalytic reaction at room temperature; meanwhile, the hy-
dride species are generated continuously from ethanol to sus-
tain the reaction.
Analysis by using XRD
The XRD patterns of the CeNiXOY nanocompounds with various
Ni contents are shown in Figure 4. A CeO2-like phase (JCPDS
file 34-0394) with similar intensity is clearly detected in each
sample, whereas the NiO phase (JCPDS file 4-0835) starts to
emerge at Xꢁ0.5. Furthermore, the diffraction peaks of crystal-
To explain how the Ni content affects the carbon formation,
the initial and final temperatures are investigated during the
activation phenomenon (Figure 3). Notably, the reaction can
be activated at 155–1718C on the catalysts with Ni/MT ꢁ0.5
after O2 is added to the feed. In such a case, the final tempera-
tures are recorded entirely at approximately 3858C. In the case
of the lower Ni molar ratio of 0.3, the reaction has to be start-
ed at a much higher temperature of 2768C, although the final
temperature is also found to be higher at approximately
4638C. Of the complete series of catalysts, the CeNi1HZOY cata-
lyst still demonstrates the optimum behavior, which gives the
largest variation between the initial and final temperatures
during the activation phenomenon of the reaction. The varia-
tion between the initial and final temperatures is found to
obey the similar regulation of DTR. Thus, the CeNi1HZOY and
CeNi0.5HZOY catalysts can inhibit the carbon formation, because
these catalysts can activate the reaction at lower temperatures
(coke formation stage) and drive the reaction at higher tem-
peratures (coke removal and balance stage). On the contrary,
the CeNi5HZOY and CeNi0.3HZOY catalysts produce more solid
carbon, which can be ascribed to the high Ni content and the
much higher initial temperature, respectively. Regardless of the
carbon formation rate, the CeNiXHZOY series of catalysts all pos-
sess excellent catalytic stability.
Figure 4. XRD patterns of CeNiXOY nanocompounds: X=0.02 (a), 0.3 (b), 0.5
^
*
(c), 1 (d), and 5 (e). CeO2 ( ), NiO ( ).
lized NiO become more intense for the samples with higher Ni
content. As already reported, the addition of Ni has an effect
not only on the broadness of CeO2 peaks but also on their po-
sitions, which was attributed to the substitution of Ce4+ cat-
ions with Ni2+ cations inside the CeO2 lattice and was inter-
preted through the formation of the Ce–Ni solid solution.[46]
The ionic radius of the Ni cation (0.07 nm) is smaller than that
of the Ce cation (0.09 nm). The CeNiXOY compounds can be as-
cribed to a mixture of nanoparticles consisting of CeO2 and
NiO and to a solid solution consisting of both Ce and Ni spe-
cies when X<1.[41,42,46] The highest proportion of the solid so-
lution has been found for the CeNi0.5OY compound in this
series of catalysts.[46] Thus, our current results are in agreement
with the literature.
Characterizations of CeNiXOY catalysts
Analysis by using the Brauner-Emmett-Teller method and in-
ductively coupled plasma-mass spectrometry
The crystallite sizes of NiO and CeO2 particles are estimated
from the diffraction lines by taking into account the (111),
(200), and (220) planes. The average crystallite size of CeO2
particles is approximately 5 nm (Table 1), which decreases
The precise Ni loadings and the surface areas of the CeNiXOY
nanocompounds were measured and are listed in Table 1. All
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
5
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
slightly with the Ni content. In the case of NiO particles, the
average crystallite size increases slightly from 8 to 11 nm with
an increase in the Ni content. It is necessary to recall that the
crystallites smaller than 2 nm cannot be examined by using
the XRD technique, owing to the detection limit. However,
such small-sized NiO nanoparticles have been proved to be
present in the CeNiXOY nanocompounds.[43,52]
Analysis by using Raman spectroscopy
The CeNiXOY nanocompounds were also characterized by using
Raman spectroscopy. The results show a strong frequency shift
and broadening for the first-order (F2g) CeO2 peak located near
460 cmꢀ1 (Figure 5) ascribed to fluorite nanocrystalline
Figure 6. Temperature-programmed reduction profiles of CeNiXOY nanocom-
pounds: X=0.02 (a), 0.3 (b), 0.5 (c), 1 (d), and 5 (e).
Ni species is observed at approximately 2708C. When the Ni/
Ce molar ratio increases from 0.02 to 1, a main reduction peak
at approximately 3708C becomes more intense and shifts
slightly toward a higher temperature of 3908C. The TPR analy-
sis of CeO2 reveals two peaks at approximately 500 and 8208C,
and thus the reduction peak obtained at approximately 8208C
is assigned to the reduction of bulk Ce4+ to Ce3+
.
The in situ treatment of CeNiXOY nanocompounds in H2
leads to the formation of CeNiXHZOY nano-oxyhydride catalysts.
The treatment conditions have been optimized in our former
papers. To activate the Ni species in the small NiO nanoparti-
cles and/or in the Ce–Ni solid solution, the appropriate treat-
ment temperature in H2 corresponds to the first TPR peak at
approximately 2708C. Further investigation revealed that
Figure 5. Raman spectra of CeNiXOY nanocompounds: X=0 (a, CeO2),
0.02 (b), 0.3 (c), 0.5 (d), 1 (e), and 5 (f).
a
higher reduction temperature of approximately 3708C
CeO2.[53,54] It shifts to lower frequencies (438 cmꢀ1) compared
to those reported previously on Ce–Ni mixed oxides
(443 cmꢀ1).[55] This behavior can be ascribed to the solubility of
Ni into CeO2, which is in agreement with the presence of
a solid solution.[56] In the present case, the distortion of the lat-
tice and the creation of oxygen vacancies are higher. Secon-
dary peaks at approximately 230 and 330 cmꢀ1 ascribed to
CeO2 nanostructures are also observed.[55] The phonon mode
at 570 cmꢀ1 is characteristic of oxygen vacancies in the CeO2
lattice.[53,57] The weak shoulders between 500 and 650 cmꢀ1
were observed on doped Ce oxides and attributed to oxygen
vacancies created due to the incorporation of the dopant.[58]
Two modes positioned at 540 and 600 cmꢀ1 were ascribed to
the local vibrations of different oxygen vacancy (VO) com-
plexes. The mode at 600 cmꢀ1 originates from the existence of
Ce3+–VO complexes in the CeO2 lattice.[56] Thus, the broad NiO
signal at approximately 520 cmꢀ1 cannot be well detected, be-
cause of a shadowing effect by the long tail of the first-order
CeO2 peak.[59] However, for the low Ni content, two peaks at
540 and 620 cmꢀ1 are observed.
cannot improve the catalytic activity; on the contrary, it result-
ed in a decrease in the ethanol conversion at low temperatures
under the SRE reaction conditions.[41,42] In addition to this,
chemical titration of the hydrogen species stored in the solids
has shown that CeNiXOY nanocompounds can store a large
amount of hydrogen and the Ni content affects the hydrogen
storage capacity of CeNiXOY compounds.[47,48] The CeNi1OY
sample is found to store the largest amount of reactive and ex-
tractable hydrogen of all the CeNiXOY series of catalysts. This
could well explain the optimum catalytic behavior of the CeN-
i1HZOY catalyst for the product distribution and the highest DTR
obtained in the OSRE reaction. To understand the process of
hydrogen storage in the catalyst, it has been proposed that H2
endures the heterolytical dissociation at an anionic vacancy
and an O2ꢀ species of the mixed oxide [Eq. (8)].
O2ꢀMnþ þ H2 ! OH MnþHꢀ
&
Mnþ represents a cation
ꢀ
&
ð8Þ
represents an anionic vacancy
It has already been pointed out that the metallic Ni species
can adsorb H2.[60] Thus, for CeNiXOY nanocompounds, in addi-
tion to the heterolytical route described above, the homolytic
dissociation of H2 on Ni0 species should be considered and dif-
ferent hydrogen species could coexist. However, the in situ
Analysis by using H2 temperature-programmed reduction
The profiles of the temperature-programmed reduction (TPR)
in H2 of CeNiXOY nanocompounds are shown in Figure 6. For
a low Ni content (X=0.3), one reduction peak ascribed to the
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
6
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
XRD analysis in H2 does not confirm the presence of Ni0 spe-
cies in the CeNi1OY solid at 2508C after 10 h until the treatment
temperature increases to 3008C.[28]
For the reduction temperatures lower than 6008C, a linear
relationship can be established between the total H2 consump-
tion during TPR and the Ni content of CeNiXOY nanocom-
pounds (Figure 7), which indicates that H2 is completely con-
Figure 8. Inelastic neutron scattering spectra of CeNiXHZOY catalysts obtained
after a treatment in H2 at 2508C: a) X=0.5 and b) X=1 (with the subtraction
of the background).
pound has been subtracted from that of the CeNiXHZOY com-
pound (treated in H2 at 2508C). The hydrogen species with the
hydride nature ascribed to the peak at approximately 460 cmꢀ1
appear clearly after a treatment in H2 at 2508C. The large band
at approximately 870 cmꢀ1 observed for the CeNi1HZOY com-
pound was assigned to the hydrogen species in the interaction
with metallic Ni0 species.[61,62] Even if no Ni0 species is observed
by using XRD, some small particles could still exist, because
such small particles are beyond the detection limit of the XRD
technique. However, the concentration of this last hydrogen
species disappear at Ni/Ce=0.5 (Figure 8), which is in agree-
ment with the existence of a high proportion of the solid solu-
tion of Ce and Ni in this compound.[46]
Figure 7. H2 consumption in temperature-programmed reduction as a func-
tion of the Ni content of CeNiXOY nanocompounds.
sumed and reduces the Ni species in this range of tempera-
tures. Moreover, the reduction peak at low temperature was
attributed to the Ni species present in 1) the solid solution
and/or 2) small NiO particles, which are easily reducible; how-
ever, with the simultaneous reoxidation of a part of these spe-
cies through the reduction of the Ce4+ cations in the proximity
to the Ce3+ cations as the existence of a redox system was es-
tablished [Eqs. (9) and (10)],[46,52] larger NiO crystallites are re-
duced when the temperature is increased.
Characterization of carbon formed
0
Ni2þ þ O2ꢀ þ H2 ! Ni þ H2O þ
Analysis by using Raman spectroscopy
&
ð9Þ
2 Ce4þ þ Ni0 $ 2 Ce3þ þ Ni2þ
In the SRE reaction, the homogeneous filamentous carbon can
contribute to the catalytic stability.[43] Because of the small
amount of carbon deposited during the reaction, Raman spec-
troscopy is used, which is the most available and powerful
technique to characterize the carbonaceous materials.[63–65] As
shown in Figure 9, two main peaks, known as the D band and
G band, are visible regardless of the Ni content of the catalyst.
ð10Þ
The reduction peak at approximately 2708C confirms that
this peak ascribes to the Ni species in small NiO nanoparticles
and/or in the solid solution. This phenomenon has been veri-
fied for different Ni contents. Thus, the CeNiXOY nanocom-
pounds with a large amount of small-sized NiO and CeO2
nanoparticles can improve the strong interactions between the
Ni and Ce species, which enables the mixed oxides to be easily
and reversibly reduced and reoxidized.
Characterization of the CeNiXHZOY nano-oxyhydride catalysts
Analysis by using inelastic neutron scattering
The introduction of hydride species into the solid is shown
here by using inelastic neutron scattering (INS).[61] The H2 treat-
ment generates an increase in the INS spectrum level as well
as the emergence of new peaks associated with hydrogen vi-
bration bands. To observe more precisely the hydrogen species
created during the activation treatment in H2, in Figure 8 the
INS spectrum of the corresponding CeNiXOY (calcined) com-
Figure 9. Raman spectra of the carbon species formed on CeNiXHZOY nano-
oxyhydride catalysts during the oxidative steam reforming of ethanol reac-
tion: X=0.3 (a), 0.5 (b), 1 (c), and 5 (d).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
7
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Notably, the carbon deposit without acid treatment in the
present result provides information obtained from the spent
catalyst. The D band observed between 1343 and 1350 cmꢀ1 is
ascribed to the vibration of the sp3 hybridized carbon atom
with the dangling bonds in the disordered carbonaceous spe-
cies, whereas the G band observed between 1586 and
1595 cmꢀ1 is ascribed to the stretching mode of the sp2 hybrid-
ized carbon atom in the ordered graphite. Comparisons of the
frequency shift between carbon nanotubes (CNTs) and carbon
nanofibers (CNFs) were reported, and only small differences
could be observed. The D band was obtained at 1350 cmꢀ1 for
CNFs and 1355 cmꢀ1 for CNTs, whereas the G band was ob-
served at 1592 cmꢀ1 for CNFs and 1582 cmꢀ1 for CNTs.[66] The
excellent stability of the CeNiXHZOY catalysts is certainly due to
the formation of the filamentous carbon species, which proba-
bly relate to the CNFs and/or CNTs, although this hypothesis
requires more proofs from other characterizations, such as TEM
and temperature-programmed oxidation (TPO).
vant peak ascribed to amorphous carbon is observed at tem-
peratures lower than 4008C; this is highly reactive toward O2
and easily oxidized from the Ni surface. The carbon species
formed on the CeNi1HZOY catalyst is graphitic in nature, which
clearly indicates an intense oxidation peak at 6038C. In TPO
analysis, temperatures at approximately 5108C were ascribed
to the single-walled carbon nanotubes whereas temperatures
at approximately 6108C were ascribed to the multiwalled
carbon nanotubes. CNFs were affirmed to burn off in lower
temperature range than did single-walled carbon nano-
tubes.[64,67,68] Compared to the TPO profile of the carbon spe-
cies generated on the CeNi1OY catalyst under the SRE reaction
conditions,[43] a narrower peak is found that shifts toward
a higher oxidation temperature, which suggests that the
carbon formed on the CeNi1HZOY catalyst under the OSRE reac-
tion conditions is more graphitic in structure. Therefore, the
carbon formed during the present process can be observed as
an added value, as CNFs and multiwalled carbon nanotubes
are strategic materials in various fields and are interesting cata-
lyst support materials for diverse catalyst applications.[68]
The intensity ratio ID/IG can be defined as a measure of the
degree of graphitization. Hence, ID/IG is characteristic of the
degree of order of the carbon materials.[67] The smaller the ID/IG
ratio, the higher the degree of order of the substances. The ID/
IG ratios do not depend monotonously on the Ni content, but
show the minimum value for the CeNi1HZOY catalyst in a de-
creasing order of 1.64 (X=5)>1.33 (X=0.3)>1.23 (X=0.5)>
0.98 (X=1). Thus, the optimal behavior of the CeNi1HZOY cata-
lyst ascribed to both the catalytic performance and the physi-
cochemical property could be due to the formation of fibrous
carbon (CNTs and/or CNFs) with the highest degree of
graphitization.
Analysis by using TEM
TEM is used to study the morphology of the carbon species
formed (Figure 11). The carbon filaments are clearly observed
on the CeNi1HZOY catalyst after the OSRE reaction, which is in
accordance with the results of TPO analysis. Moreover, Ni parti-
Analysis by using O2-TPO
To gain insight into the nature of the carbon species formed
on the CeNi1HZOY catalyst, TPO is used to evaluate the nature
of carbon structure, that is, whether it is amorphous and/or
graphitic in nature (Figure 10). It has already been well estab-
lished in the literature that the types of carbonaceous species
can be identified by the TPO peaks related to different reactivi-
ties toward oxidation. It is immediately apparent that no rele-
Figure 11. TEM images of the spent CeNi1HZOY nano-oxyhydride catalyst
after the oxidative steam reforming of ethanol reaction. The 5 nm scale
image focuses on carbon species.
cles are observed at the tip of the carbon filaments. Such
types of filamentous carbon species have already been report-
ed as the benefit to the catalytic stability for the SRE reaction
on the CeNi1OY nanocompound.[43] The carbon filaments are
composed of graphitic sheets, and the graphitic structure is
shown in Figure 11. This result is in good agreement with the
Raman spectrum of the carbon species formed on the CeN-
i1HZOY catalyst, in which the G band observed at 1595 cmꢀ1 is
ascribed to the stretching mode of the sp2 hybridized carbon
atom in the ordered graphite. Moreover, the CeNi1HZOY catalyst
has the minimum ID/IG ratio, which corresponds to a higher
degree of order of substances.
Figure 10. Temperature-programmed oxidation profile of the carbon species
formed on the CeNi1HZOY nano-oxyhydride catalyst during the oxidative
steam reforming of ethanol reaction.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
8
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
by combining the exothermic chemical reaction between the
hydride species stored in the CeNiXHZOY catalysts and O2 with
the exothermic ethanol partial oxidation. Hence, a huge varia-
tion in the temperature between the catalyst bed and the
oven can be observed. Such a unique advantage can com-
pletely convert ethanol and produce H2, with the need of only
a small amount of extra energy, and it lowers the oven temper-
ature down to 608C. CeNiXHZOY catalysts show excellent cata-
lytic stability owing to the graphitic filamentous carbon species
formed during the reaction. The Ni content has a significant
effect on the product distribution, whereas the activity is as-
cribed to the hydrogen storage capacity of the solid depend-
ing on the Ni content. The CeNi1HZOY nano-oxyhydride is
proved to be the best catalyst in the series analyzed. Complete
ethanol conversion and 45% H2 molar fraction in all the gas
phase products are achieved at 3458C when the oven temper-
Active site modeling
The ethanol dehydrogenation step requires the abstraction of
hydrogen species from alcohol. As it has often been remarked
that the abstraction of hydrogen from alcohol is the rate-deter-
mining step, the ability of the solid to accept hydrogen is an
important factor. The variation in the temperature measured is
mainly attributed to the reaction between the hydride species
of the solid and O2 (exothermic), and it depends on the con-
centration of hydrogen stored in the solid in the hydride
form.[47] This phenomenon enables the decrease in the oven
temperature. A part of hydride species formed from ethanol
react with O2, which provides chemical energy and maintains
the catalytic reaction. The reaction is sustainable because hy-
dride species are provided by ethanol. An active site involving
an anionic vacancy and an O2ꢀ species of the solid can be en-
visaged for the heterolytic dissociation of ethanol [Eq. (11)], as
it has been proposed for the heterolytic dissociation of H2
[Eq. (8)].
ature is at only 608C. In addition, a small amount of carbon is
ꢀ1
formed (with a formation rate of 28 mgh
g
ꢀ1) after 75 h of
cat
the reaction. This is the best result that has ever been reported
for a low-cost catalytic system.
ꢀ
O2ꢀMnþ þ C2H5OH ! OH MnþHꢀ þ CH3CHO
&
ð11Þ
Such a site has already been modeled by an ensemble of
two cations in the close interaction.[28] It has been shown in
the laboratory that three coordinatively unsaturated sites (3M)
are a prerequisite condition for alkadiene hydrogenation activi-
Experimental Section
Catalysts preparation
The CeNiXOY nanocompounds were prepared by using the copreci-
pitation method. The Ni and Ce nitrate solutions (0.5 molLꢀ1) were
coprecipitated with triethylamine. After filtration, the solids were
dried at 1008C and finally calcined in air at 5008C for 4 h.
X
Y
ty and different ensembles Mꢀ M’ (in which X and Y are the
number of unsaturations, i.e., anionic vacancies on each
cation) have been proposed to be the active sites; each ele-
mentary ensemble is associated with a particular reaction. De-
pending on the unsaturation degree of this active site, ethanol
transformation can lead to different products. With a low unsa-
turation degree (one anionic vacancy), CH3CHO and H2 can be
obtained through the heterolytic abstraction of hydrogen. On
a site with a higher unsaturation degree obtained after a treat-
ment in H2 at 2508C, ethanol can be transformed into H2 and
other products.[28] This is in agreement with the observed in-
crease in O vacancies induced by doping CeO2 with Ni[15,55]
even if the existence of a hydride species has not been
envisaged.[15]
Catalytic reaction
The catalytic reaction was performed under the atmospheric pres-
sure in a fixed-bed quartz reactor (inner diameter 8 mm) fitted in
a programmable oven. The catalyst (30 mg) was previously treated
in situ in H2 at 2508C for 10 h. The water–ethanol mixture with
a stoichiometric molar ratio of 3 was pumped and vaporized in the
heating chamber. The gas stream of H2O/EtOH/O2/N2 =3:1:1.6:1.3
was then fed to the reactor, with a total flow rate of 60 mLminꢀ1
The outlet gases were analyzed online with a TRACE GC ULTRA gas
chromatograph equipped with a thermal conductivity detector
and a flame ionization detector.
.
Conclusions
The reactants were introduced in a specific order, and O2 was fed
in the last to avoid the immediate reoxidation of the oxyhydride.
After all the reactants were fed, the reaction could be activated at
a certain temperature, depending on the ability of the oxyhydride
catalyst to react with O2. The heat of the oven was reduced as
soon as the measured reaction temperature started to increase
drastically, and the variation in the reaction temperature was re-
corded.
We have developed a novel catalytic technology to support
a sustainable H2 production from ethanol and water at room
temperature. We report highly efficient and stable CeNiXHZOY
nano-oxyhydride catalysts for the oxidative steam reforming of
ethanol. A facile coprecipitation method is used to prepare the
CeNiXOY nanocompounds, which possess a high proportion of
active Ni species in small-sized NiO nanoparticles and/or in the
solid solution and can be reduced and reoxidized easily and re-
versibly because of the existence of strong interactions be-
tween the Ni and Ce cations. The inelastic neutron scattering
characterization confirms that the hydride species are intro-
duced into the CeNiXOY nanocompounds to fabricate CeN-
iXHZOY nano-oxyhydride catalysts after an in situ treatment in
H2 at 2508C. This new technology enables us to save energy
Catalytic performances were measured by calculating ethanol con-
version (XEtOH) and product molar composition (Ci) (dry basis)
[Eqs. (12) and (13)].
nEtOH;in ꢀ nEtOH;out
XEtOH
¼
ꢃ 100 %
ð12Þ
ð13Þ
nEtOH;in
ni
P
Ci ¼
ꢃ 100 %
products ni
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&
9
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Characterizations
technique de la Fꢀdꢀration Chevreul CNRS FR 2638”, for their val-
uable technical help. The Fonds Europꢀen de Dꢀveloppement Re-
gional, CNRS, Rꢀgion Nord Pas-de Calais, and Ministꢀre de l’Edu-
cation Nationale de l’Enseignement Supꢀrieur et de la Recherche
are also acknowledged for the funding of XRD analysis.
The Ni and Ce contents (in wt%) were analyzed by using inductive-
ly coupled plasma-mass spectrometry (CNRS, Service Central d’Ana-
lyse, Vernaison, France). For CeNiXHZOY catalysts, the Ni/MT molar
ratio is the Ni molar proportion in all the metals (MT =X+1; Ni/
MT =X/(X+1)).
The BET surface area was measured by using N2 physisorption at
77 K with a Micromeritics Tristar II 3020 surface area and porosime-
try analyzer. Samples were previously outgassed under vacuum at
1508C for 3 h.
Keywords: ethanol
sustainable reactions
· hydrogen · nickel · oxyhydride ·
XRD analysis was performed with a Bruker D8 Advance X-ray dif-
fractometer equipped with a fast detector-type LynxEye with
a copper anticathode. The XRD patterns were registered at 2q=
20–908, with a measured step of 0.028, and the integration time
was set at 0.3 s. The crystallite size was calculated by using the
Scherrer equation.
[1] G. A. Deluga, J. R. Salge, L. D. Schmidt, X. E. Verykios, Science 2004, 303,
993–997.
[2] A. Midilli, M. Ay, I. Dincer, M. A. Rosen, Renewable Sustainable Energy
Rev. 2005, 9, 273–287.
[3] M. Momirlan, T. N. Veziroglu, Int. J. Hydrogen Energy 2005, 30, 795–802.
[4] V. A. Goltsov, T. N. Veziroglu, L. F. Goltsova, Int. J. Hydrogen Energy 2006,
31, 153–159.
[5] K. Christopher, R. Dimitrios, Energy Environ. Sci. 2012, 5, 6640–6651.
[6] J. A. Turner, Science 2004, 305, 972–974.
[7] M. Ni, D. Y. C. Leung, M. K. H. Leung, Int. J. Hydrogen Energy 2007, 32,
3238–3247.
[8] R. M. Navarro, M. C. Sanchez-Sanchez, M. C. Alvarez-Galvan, F. d. Valle,
J. L. G. Fierro, Energy Environ. Sci. 2009, 2, 35–54.
[9] D. Chen, L. He, ChemCatChem 2011, 3, 490–511.
[10] L. He, D. Chen, ChemSusChem 2012, 5, 587–595.
[11] S. Dutta, RSC Adv. 2012, 2, 12575–12593.
[12] N. Bion, D. Duprez, F. Epron, ChemSusChem 2012, 5, 76–84.
[13] L. V. Mattos, G. Jacobs, B. H. Davis, F. B. Noronha, Chem. Rev. 2012, 112,
4094–4123.
[14] E. Y. Garcꢃa, M. A. Laborde, Int. J. Hydrogen Energy 1991, 16, 307–312.
[15] G. Zhou, L. Barrio, S. Agnoli, S. D. Senanayake, J. Evans, A. Kubacka, M.
Estrella, J. C. Hanson, A. Martꢃnez-Arias, M. Fernꢄndez-Garcꢃa, J. A. Rodri-
guez, Angew. Chem. 2010, 122, 9874–9878; Angew. Chem. Int. Ed. 2010,
49, 9680–9684.
[16] M. Z. Jacobson, W. G. Colella, D. M. Golden, Science 2005, 308, 1901–
1905.
[17] J. C. Ruiz-Morales, D. Marrero-Lopez, J. Canales-Vazquez, J. T. S. Irvine,
RSC Adv. 2011, 1, 1403–1414.
H2-TPR was performed with a Micromeritics AutoChem II chemi-
sorption analyzer, and the H2 consumption was measured with
a thermal conductivity detector. The sample (50 mg) was treated in
a 5 vol% H2–95 vol% Ar mixture, with a flow rate of 30 mLminꢀ1
.
The temperature was increased to 10008C, with a heating rate of
108Cminꢀ1
.
O2-TPO was performed with a Micromeritics AutoChem 2920 chem-
isorption analyzer. The sample was treated in a 5 vol% O2–95 vol%
He mixture, with a flow rate of 50 mLminꢀ1. The temperature was
increased to 10008C, with a heating rate of 58Cminꢀ1. The desorp-
tion species from the sample were traced by using an OmniStar
GSD 300 O mass spectrometer.
INS experiments were conducted with the IN1 BeF spectrometer at
the Institut Laue–Langevin, Grenoble, France. The solid (36 g) was
placed in a stainless steel container and treated in high purity H2
(10 h). INS analysis was performed at 10 K with a Cu monochroma-
tor. The scattering cross section is much greater for H2 [80 barn
(1 barn=10ꢀ28 m2)] than for other elements (5 barn); therefore, INS
emphasizes the motions of hydrogen species.
[18] M. Contestabile, G. J. Offer, R. Slade, F. Jaeger, M. Thoennes, Energy Envi-
ron. Sci. 2011, 4, 3754–3772.
[19] J. R. Salge, G. A. Deluga, L. D. Schmidt, J. Catal. 2005, 235, 69–78.
[20] S. N. Hsu, J. L. Bi, W. F. Wang, C. T. Yeh, C. B. Wang, Int. J. Hydrogen
Energy 2008, 33, 693–699.
[21] K. Sato, K. Kawano, A. Ito, Y. Takita, K. Nagaoka, ChemSusChem 2010, 3,
1364–1366.
[22] W. Cai, F. Wang, E. Zhan, A. C. Van Veen, C. Mirodatos, W. Shen, J. Catal.
2008, 257, 96–107.
[23] S. M. de Lima, I. O. da Cruz, G. Jacobs, B. H. Davis, L. V. Mattos, F. B.
Noronha, J. Catal. 2008, 257, 356–368.
[24] H. Chen, H. Yu, F. Peng, H. Wang, J. Yang, M. Pan, J. Catal. 2010, 269,
281–290.
[25] F. Frusteri, S. Freni, V. Chiodo, S. Donato, G. Bonura, S. Cavallaro, Int. J.
Hydrogen Energy 2006, 31, 2193–2199.
The Raman spectra were recorded on a LabRam Infinity Raman
spectrometer (Horiba Jobin Yvon) using a visible laser with an
output laser power of l=532 nm at RT.
TEM images were obtained with a FEI Tecnai G2 20 transmission
electron microscope operating at an acceleration voltage of
200 kV. The sample was ultrasonically dispersed in acetone, and
then drops of the suspension were placed on a copper grid-sup-
ported transparent carbon film.
The amount of carbon formed was determined by measuring mass
variation.
[26] E. B. Pereira, P. Ramꢃrez De La Piscina, S. Martꢃ, N. Homs, Energy Environ.
Sci. 2010, 3, 487–493.
[27] S. Andonova, C. N. de ꢅvila, K. Arishtirova, J. M. C. Bueno, S. Damyanova,
Appl. Catal. B 2011, 105, 346–360.
Acknowledgements
W.F. gratefully acknowledges a grant from Erasmus Mundus
Tandem. C.P. gratefully acknowledges a grant from the French
Ministry. A sincere appreciation is dedicated to the Institut Laue–
Langevin, Grenoble, France, for the financial support of the INS
experiments and to the help of Dr. A. Ivanov (for IN1). The au-
thors thank Ms. L. Burylo (for XRD), Mr. O. Gardoll (for TPR), and
Mr. J.C. Morin (for Raman spectroscopy) from the Unitꢀ de Cata-
lyse et Chimie du Solide and Mr. A. Addad (for TEM) from the
Centre Commun de Microscopie of the University of Lille, “plateau
[28] C. Pirez, M. Capron, H. Jobic, F. Dumeignil, L. Jalowiecki-Duhamel,
Angew. Chem. 2011, 123, 10375–10379; Angew. Chem. Int. Ed. 2011, 50,
10193–10197.
[29] B. Neltner, B. Peddie, A. Xu, W. Doenlen, K. Durand, D. S. Yun, S. Speak-
man, A. Peterson, A. Belcher, ACS Nano 2010, 4, 3227–3235.
[30] A. Iulianelli, A. Basile, Catal. Sci. Technol. 2011, 1, 366–379.
[31] N. Zhu, X. Dong, Z. Liu, G. Zhang, W. Jin, N. Xu, Chem. Commun. 2012,
48, 7137–7139.
[32] M. Wang, L. Sun, ChemSusChem 2010, 3, 551–554.
[33] R. M. Navarro Yerga, M. C. ꢅlvarez-Galvꢄn, N. Mota, J. A. Villoria de La
Mano, S. M. Al-Zahrani, J. L. G. Fierro, ChemCatChem 2011, 3, 440–457.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&10
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[34] C. Descorme, P. Gallezot, C. Geantet, C. George, ChemCatChem 2012, 4,
1897–1906.
[53] G. Li, R. L. Smith, Jr., H. Inomata, J. Am. Chem. Soc. 2001, 123, 11091–
11092.
[35] W. Cai, F. Wang, C. Daniel, A. C. van Veen, Y. Schuurman, C. Descorme,
H. Provendier, W. Shen, C. Mirodatos, J. Catal. 2012, 286, 137–152.
[36] S. Velu, N. Satoh, C. S. Gopinath, K. Suzuki, Catal. Lett. 2002, 82, 145–
152.
[37] R. Guil-Lꢆpez, R. M. Navarro, M. A. PeÇa, J. L. G. Fierro, Int. J. Hydrogen
Energy 2011, 36, 1512–1523.
[54] S.-L. Zhong, L.-F. Zhang, L. Wang, W.-X. Huang, C.-M. Fan, A.-W. Xu, J.
Phys. Chem. C 2012, 116, 13127–13132.
[55] L. Barrio, A. Kubacka, G. Zhou, M. Estrella, A. Martꢃnez-Arias, J. C.
Hanson, M. Fernꢄndez-Garcꢃa, J. A. Rodriguez, J. Phys. Chem. C 2010,
114, 12689–12697.
´
ˇ
´
´
ˇ
´
[56] N. Paunovic, Z. Dohcevic-Mitrovic, R. Scurtu, S. Askrabic, M. Prekajski, B.
´
´
[38] C. Bianchini, P. Fornasiero, ChemCatChem 2012, 4, 45–46.
[39] G. Frenette, D. Forthoffer, Int. J. Hydrogen Energy 2009, 34, 3578–3588.
[40] P. Leung, A. Tsolakis, J. Rodriguez-Fernandez, S. Golunski, Energy Environ.
Sci. 2010, 3, 780–788.
[41] L. Jalowiecki-Duhamel, C. Pirez, M. Capron, F. Dumeignil, E. Payen, Int. J.
Hydrogen Energy 2010, 35, 12741–12750.
[42] L. Jalowiecki-Duhamel, C. Pirez, M. Capron, F. Dumeignil, E. Payen, Catal.
Today 2010, 157, 456–461.
[43] W. Fang, C. Pirez, M. Capron, S. Paul, T. Raja, P. L. Dhepe, F. Dumeignil, L.
Jalowiecki-Duhamel, RSC Adv. 2012, 2, 9626–9634.
[44] S. Yin, Y. Minamidate, S. Tonouchi, T. Goto, Q. Dong, H. Yamane, T. Sato,
RSC Adv. 2012, 2, 5976–5982.
[45] H. Li, G. Wang, F. Zhang, Y. Cai, Y. Wang, I. Djerdj, RSC Adv. 2012, 2,
12413–12423.
[46] C. Lamonier, A. Ponchel, A. D’Huysser, L. Jalowiecki-Duhamel, Catal.
Today 1999, 50, 247–259.
[47] L. Jalowiecki-Duhamel, Int. J. Hydrogen Energy 2006, 31, 191–195.
[48] L. Jalowiecki-Duhamel, J. Carpentier, A. Ponchel, Int. J. Hydrogen Energy
2007, 32, 2439–2444.
[49] L. Jalowiecki-Duhamel, H. Zarrou, A. D’Huysser, Int. J. Hydrogen Energy
2008, 33, 5527–5534.
Matovic, Z. V. Popovic, Nanoscale 2012, 4, 5469–5476.
[57] A. Mineshige, T. Taji, Y. Moroi, M. Kobune, S. Fujii, N. Nishi, M. Inaba, Z.
Ogumi, Solid State Ionics 2000, 135, 481–485.
[58] M. Palard, J. Balencie, A. Maguer, J.-F. Hochepied, Mater. Chem. Phys.
2010, 120, 79–88.
[59] N. Dharmaraj, P. Prabu, S. Nagarajan, C. H. Kim, J. H. Park, H. Y. Kim,
Mater. Sci. Eng. B 2006, 128, 111 – 114.
[60] C.-j. Liu, J. Ye, J. Jiang, Y. Pan, ChemCatChem 2011, 3, 529–541.
[61] H. Jobic, Handbook of Heterogeneous Catalysis, 2nd ed. (Eds.: G. Ertl, H.
Knçzinger, F. Schꢇth, J. Weitkamp), Wiley-VCH, Weinheim, 2008,
pp. 971.
[62] L. Jalowiecki-Duhamel, S. Debeusscher, H. Jobic, E. Payen, Int. J. Nuclear
Hydrogen Prod. Appl. 2009, 2, 148–158.
[63] P. Serp, E. Castillejos, ChemCatChem 2010, 2, 41–47.
[64] J.-P. Tessonnier, M. Becker, W. Xia, F. Girgsdies, R. Blume, L. Yao, D. S. Su,
M. Muhler, R. Schlçgl, ChemCatChem 2010, 2, 1559–1561.
[65] M. Inagaki, H. Orikasa, T. Morishita, RSC Adv. 2011, 1, 1620–1640.
[66] Y. Liu, C. Pan, J. Wang, J. Mater. Sci. 2004, 39, 1091–1094.
[67] Y. Kobayashi, M. Sano, ChemCatChem 2010, 2, 397–401.
[68] E. Sairanen, R. Karinen, M. Borghei, E. I. Kauppinen, J. Lehtonen, Chem-
CatChem 2012, 4, 2055–2061.
[50] M. A. Ebiad, D. R. Abd El-Hafiz, R. A. Elsalamony, L. S. Mohamed, RSC
Adv. 2012, 2, 8145–8156.
[51] J. Rass-Hansen, C. H. Christensen, J. Sehested, S. Helveg, J. R. Rostrup-
Nielsen, S. Dahl, Green Chem. 2007, 9, 1016–1021.
[52] A. Ponchel, A. Huysser, C. Lamonier, L. Jalowiecki-Duhamel, Phys. Chem.
Chem. Phys. 2000, 2, 303–312.
Received: January 31, 2013
Revised: February 28, 2013
Published online on && &&, 0000
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 11
&11
&
ÞÞ
These are not the final page numbers!