In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to Steam Reforming of Ethanol reaction

Article abstract of DOI:10.1016/j.cattod.2016.02.029

The effect of ceria in cobalt-ceria-alumina catalyst was studied using in situ X-ray near edge spectroscopy (XANES) at Co K-edge and Ce L_III-edge. The introduction of ceria in this catalyst resulted in a significant removal of Co from CoAl₂O₄spinel phase to Co and CoO phases. The Co K-edge revealed the symmetry changes in Co according to the temperature and atmosphere in reduction process. We also showed the stability of Co sites as well the oxidation state of ceria in operando steam reforming of ethanol (SRE) reaction. After reduction of Ce⁴⁺to Ce³⁺, no significant changes were observed by XANES. Gas Chromatography (GC) analysis showed a high ethanol conversion at 500?°C, high hydrogen yield and low formation of undesired products as methane and ethylene. The results showed a Co/Ce/Al₂O₃catalyst as promising material to be applied in hydrogen production in SRE reaction.

Full text of DOI:10.1016/j.cattod.2016.02.029

G Model
CATTOD-10059; No. of Pages6
ARTICLE IN PRESS
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to Steam
Reforming of Ethanol reaction
Arthur E.Pastore de Limaa,b, Daniela C. de Oliveiraa,∗
a Brazilian Synchrotron Light Laboratory, LNLS, Caixa Postal 6192, CEP 13083-970 Campinas, SP, Brazil
b Universidade Estadual de Campinas, UNICAMP, Campinas, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
The effect of ceria in cobalt-ceria-alumina catalyst was studied using in situ X-ray near edge spectroscopy
(XANES) at Co K-edge and Ce LIII-edge. The introduction of ceria in this catalyst resulted in a significant
removal of Co from CoAl2O4 spinel phase to Co and CoO phases. The Co K-edge revealed the symme-
try changes in Co according to the temperature and atmosphere in reduction process. We also showed
the stability of Co sites as well the oxidation state of ceria in operando steam reforming of ethanol (SRE)
reaction. After reduction of Ce4+ to Ce3+, no significant changes were observed by XANES. Gas Chromatog-
raphy (GC) analysis showed a high ethanol conversion at 500 ◦C, high hydrogen yield and low formation
of undesired products as methane and ethylene. The results showed a Co/Ce/Al2O3 catalyst as promising
material to be applied in hydrogen production in SRE reaction.
Received 24 September 2015
Received in revised form 20 January 2016
Accepted 25 February 2016
Available online xxx
Keywords:
Hydrogen
Ethanol
Ceria
© 2016 Elsevier B.V. All rights reserved.
Catalyst
Cobalt
XANES
1. Introduction
The reaction pathway follows through dehydrogenation of
ethanol, acetaldehyde steam reforming reaction and water-gas
The growing need of a more rational use of natural resources
has drawn high interest in hydrogen (H2) production, especially
from renewable sources [1]. The use of H2 on fuel cells in automo-
biles involves the production of this gas at the time of use, to avoid
problems related to its storage and transportation.
shift reaction (WGSR). Parallel reactions may occur in Steam
Reforming of Ethanol reaction. Decomposition of ethanol produces
methane, carbon monoxide and hydrogen; and the dehydration of
ethanol produces ethylene, which can lead to coke formation on
catalyst surface, causing its deactivation [11].
The hydrogen production from methanol and hydrocarbons is
widely studied in reforming reactions; however, the reforming of
these compounds releases CO2, which is harmful to the environ-
ment [2,3]. But hydrogen can also be obtained from bio-ethanol,
for example, in a sustainable and clean way [4], through the Steam
Reforming of Ethanol (SRE), a very simple catalytic process. This
reaction also releases CO2, but the crops that produce bio-ethanol
(for SRE), such as sugarcane and corn, require CO2 to grow and
develop, balancing the emission of this gas [2,3,5].
Steam Reforming of Ethanol consists in reacting ethanol and
water, at moderate temperatures, to produce CO2 and H2. Liter-
ature reports different mechanisms for SRE, under the influence
of different catalysts supported in different materials [2–10]. The
general reaction is represented in Eq. (1) [5].
SRE has been widely studied and a large variety of materials
can be used under soft reaction conditions. The great interest on
the development of new active catalysts has contributed to under-
stand the mechanisms, to enhance the hydrogen selectivity while
the coke formation over the material can be minimized [4].
Choosing the catalysts is essential to SRE, because they may
improve the selectivity to desired product and minimize the par-
allel reactions. Ideally, catalyst must maximize ethanol conversion
and hydrogen selectivity, convert the CO produced to CO2 and avoid
coke formation over catalyst surface. Many transition metals have
been studied as catalyst active phase as Ni, Co, Cu, Fe [12]. For this
purpose, the non-noble metals Ni and Co have showed efficient
catalytic activity for SRE reaction [5]. The interest in cobalt catalyst
has been reinforced thanks to their activity to break C C bonds [13],
and their low cost, high activity and selectivity to hydrogen [12]. It
is then noteworthy that many authors [12–14] have demonstrated
that cobalt is an alternative to nickel due to the effectiveness of this
catalyst in the SRE.
C2H5OH + 3H2O → 2CO2 + 6H2
(1)
∗
Corresponding author.
E-mail address: doliveira@lnls.br (D.C. de Oliveira).
http://dx.doi.org/10.1016/j.cattod.2016.02.029
0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: A.E.Pastore de Lima, D.C. de Oliveira, In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to
Steam Reforming of Ethanol reaction, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.029

G Model
CATTOD-10059; No. of Pages6
ARTICLE IN PRESS
2
A.E.Pastore de Lima, D.C. de Oliveira / Catalysis Today xxx (2016) xxx–xxx
nH
Many different supports can be used in SRE, which can play
an active hole in the reaction depending on composition, mor-
phology and particle sizes. In particular, ceria has been widely
studied as active support in SRE with reports showing that dif-
ferent particles sizes have significant activity, especially at high
temperatures. However, the C C bond cleavage activity was seen
to be low [15–17]. Also, Song and Ozkan [18] reported on the pos-
itive effect of ceria to avoid deactivation by coke, due to the high
oxygen mobility in ceria oxide, which improves the catalytic stabil-
ity. All these interesting properties has highlighted the outstanding
catalytic properties of Co/ceria system in SRE [19].
Several techniques have been used to characterize inorganic
solids in order to obtain morphological, structural and/or textu-
ral information of catalysts [20]. Among other techniques, XAS
(X-Ray Absorption Spectroscopy) in the near edge region XANES (X-
Ray Absorption Near Edge Structure) provides information of the
oxidation state of absorber atom, the spatial arrangement of neigh-
borhood and the density of unoccupied states [21,22]. Moreover,
the high penetration of hard X-rays allows to investigate catalysts
under operando conditions [23].
2
RH
=
(3)
(4)
(5)
2
6 × (nin,EtOH − nout,EtOH
)
nj ×
j
ꢀ
S(j) =
i[ni × i]
nH × ꢁH
2
2
ꢀ
S(H2) =
i[ni × ꢁi]
where ␹EtOH is the ethanol conversion; nj is the amount of substance
of component j; RH2 is the hydrogen yield; S(j) is the selectivity for
component j; j is the number of carbon atoms present in molecule
j; and ꢁj is the number of hydrogen atoms present in molecule j. The
index i represent only the reaction products.
2.3. Catalyst characterization by XANES
XANES in situ experiments were done at the Brazilian Syn-
chrotron Light Laboratory (LNLS) on D04B (XAFS1) beamline. XAS
spectra were recorded at the cobalt K-edge (E = 7709 eV) and cerium
L
III-edge (E = 5723 eV) using a channel-cut Si(111) monochromator.
The harmonic contamination at the XAFS1 beamline is less than
0.1% above ∼4300 eV.
With all that in mind, the goal of this work was to study the
structure of Co-supported on ceria-alumina catalysts by XANES
during activation under H2 and in SRE reaction. We also aimed at
studing the catalytic activity for Steam Reforming of Ethanol reac-
tion, seeking for the optimal conditions to achieve high ethanol
conversion, hydrogen selectivity and catalyst stability. We show
that the presence of ceria affects the catalyst structure allowing
hydrogen production in our catalytic tests under SRE conditions.
The data collection was carried out in transmission mode using
ion chamber detectors to measure the incident (I0) and transmit-
ted (I1) x-rays intensities. The reference samples used in this study
were obtained from commercial sources with 99,99% purity. To
acquire the spectrum we used 0.5 and 0.3 eV energy step at XANES
region of Ce LIII-edge and Co K-edge around the edge, respectively,
counting 1s/point. The total time to measure each spectrum was
about 20 min at Ce LIII-edge and 10 min at Co K-edge. The Co5Ce10Al
catalyst was studied by XANES along the in situ activation pro-
cess (reduction of Co oxide) and in SRE operando conditions. The
required amount of samples were mixed with boron nitride to
achieve the optimal sample thickness with a jump of ꢂ␮ = 0.58 and
ꢂ␮ = 0.14 for Ce LIII-edge and Co K-edges measurements, respec-
tively. Cobalt and chromiun metal foils were used as a reference for
energy calibration. An in situ cell was used in order to carry out the
x-ray absorption measurement at realistic SRE reaction conditions.
The cell consists of a quartz tube in a ceramic furnace heated by a
Cr–Al thread. The quartz tube is sealed by a Kapton window sup-
ported in a connector with standard Swagelok 1/4” where unions
fixed at both extremity of the tube permit a gas flux through the
sample pellet. Gas flow was supplied by a mobile gas distribution
set controlled by commercial flow meters. The temperature pro-
grammed reduction was carried out by heating the pellet up to
800 ◦C under 100 mL/min flow of 5% H2/He, keeping the temper-
ature constant by 1 h before stopping heating and then cooling
the sample to the ambient temperature under He flow. Following,
the sample was heated to 400 ◦C and we introduced the ethanol
and water to start the SRE; then the reaction was also performed
at 500 ◦C. All this process was followed by XANES spectra every
∼10 min and mass spectrometry (MS) to check the SRE reactants
and products.
2. Experimental
2.1. Catalyst preparation
The catalyst was prepared by the wet impregnation method. An
aqueous solution of CoCl2·6H2O and Ce(NO3)3·6H2O was added to
Al2O3 (all Sigma-Aldrich reactants). The mixture was stirred over
1 h and then dried overnight at 100 ◦C. The solid obtained was cal-
cined at 800 ◦C over 8 h in air to remove residual components. We
defined the initial concentrations to obtain a final solid with molar
composition of 5% of CoO, 10% of Ce2O3 and 85% of Al2O3. The
sample was named Co5Ce10Al.
2.2. Catalytic reaction
Preliminary catalytic tests were performed to find the exper-
imental conditions of activation and reaction using
a mass
spectrometer (MS). To assess the catalytic activity of the Co5Ce10Al
sample, a gas chromatograph (GC-Agilent) equipped with a FID
(flame ionization detector) and a TCD (thermal conductivity detec-
tor) was used. The reaction was carried out in a quartz tube reactor
with 12 mm of diameter. First we activated the sample at 800 ◦C
for 1 h under continuous flow of diluted hydrogen (100 mL/min, 5%
H2/He). Then we cooled down the furnace temperature to 200 ◦C
and the reaction mixture was fed into the reactor. Catalytic tests
were carried out from 200 ◦C to 700 ◦C. We used a ratio of 3:1 of
H2O:ethanol (steam); a total flux of 100 mL/min of He was used
to fed the reactants into the reactor. We analyzed the products of
reaction each 100 ◦C over 1 h (5 injections per temperature). The
quantification (Table 1) was obtained by comparing the area under
the corresponding GC peak with the calibration performed prelim-
inary with each component of the reaction products and reactants.
Ethanol conversion, hydrogen yield and products selectivity where
calculated with the following equations [24,25].
3. Results and discussion
3.1. Catalyst characterization
3.1.1. XANES study at Co K-edge
From XANES data at Co K-edge the information about geometry
and oxidation state can be attributed by qualitative analysis. Fig. 1
shows the XANES of Co K-edge of some cobalt compounds. The Co
from Co5Ce10Al sample is in CoAl2O4 structure, which has spinel
geometry where the cobalt ions (Co2+) occupy only tetrahedral (Td)
sites while the aluminum ions are in the octahedral (Oh) sites. The
other common compound with spinel structure is Co3O4 with two
nin,EtOH − nout,EtOH
ꢀEtOH
=
(2)
nin,EtOH
Please cite this article in press as: A.E.Pastore de Lima, D.C. de Oliveira, In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to
Steam Reforming of Ethanol reaction, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.029

G Model
CATTOD-10059; No. of Pages6
ARTICLE IN PRESS
A.E.Pastore de Lima, D.C. de Oliveira / Catalysis Today xxx (2016) xxx–xxx
3
Table 1
Ethanol conversion, hydrogen yield and selectivity to the SRE products to the Co5Ce10Al catalyst.
Co5Ce10Al
400 ◦
C
500 ◦
C
600 ◦
C
700 ◦
C
Ethanol conversion
H2 yield
S(H2)
S(CO2)
S(CH4)
S(Acetaldehyde)
S(CO)
S(Acetic acid)
S(Ethylene)
S(Acetone)
(22.1 0.7)%
(51.4 0.8)%
(67.4 0.8)%
(15.6 0.5)%
(3.0 1)%
(61.7 0.7)%
(13.0 2)%
–
(84 2)%
(100 0)%
(64.3 0.2)%
(97.6 0.03)%
(59.0 2)%
(2.4 0.03)%
–
(38.0 2)%
–
–
–
(100 0)%
(62.0 0.3)%
(95.4 0.1)%
(59.0 1)%
(4.51 0.09)%
–
(36.2 0.9)%
–
–
–
(51.8 0.4)%
(77.7 0.5)%
(41.0 1)%
(2.2 0.3)%
(20.0 2)%
(16.1 0.8)%
–
–
(4.7 0.2)%
(15.0 2)%
(6.1 0.4)%
Fig. 2. (top) Relative Co compounds fractions in the Co5Ce10Al sample as a function
of reduction temperature extracted from the XANES at Co K-edge reported in the
bottom part; (bottom) evolution of the Co K-edge XANES of Co5Ce10Al sample along
the reduction process.
Fig. 1. (top) XANES at Co K-edge from reference compounds; from top to bottom:
CoAl2O4, CoO, Co3O4 and Co metal; (bottom) XANES at Co K-edge from Co5Ce10Al
sample under different conditions along the activation and SRE reaction.
thirds of the cobalt ions occupying Oh sites with oxidation state
3+ and a third of the cobalt ions occupying Td sites with oxidation
state 2+. But to the later compound the intensity of the pre-edge
peak, which is attributed to the 1s–3d transition, is lower. The pre-
edge intensity is related to the occupancy as well the symmetry of
the 3d shell. In the case of Td coordination, the dipole-forbidden
1s–3d transition becomes partially dipole-allowed by the asym-
metry and it results in a pre-edge peak. With Oh coordination, the
pre-edge peak is less intense or absent, only appearing if the center
of symmetry is broken, allowing the 1s–3d transition. Another ref-
erence sample such as CoO has only Oh sites for Co, and therefore
the pre-edge feature is very weak. In Co5Ce10Al initial sample, a
well defined pre-edge peak can be observed around 7709 eV. Other
important feature observed in the initial sample is the strong edge
shoulder, characteristic of the higher occupancy of Co2+ in Td sites
as CoAl2O4 [26].
XANES measurements in Temperature Programmed Reduction
(TPR) were used to obtain information about phase changes of Co as
a function of the temperature under hydrogen atmosphere showed
in Fig. 2. The first changes in the spectra were observed at low tem-
peratures up to ∼490 ◦C, where the white line starts to decrease (the
main absorption peak owing to 1s–4p transition, with area under
the peak directly related to the number of unoccupied states in the
4p), which can be attributed to the water removal from the sample.
There is no energy shift along the reduction up to ∼700 ◦C and the
pre-edge as well as the white line remain unchanged. From ∼700 ◦C
to 800 ◦C the decrease in the intensity of white line is evident. After
that temperature, the increase in the pre-edge is concomitant to
Please cite this article in press as: A.E.Pastore de Lima, D.C. de Oliveira, In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to
Steam Reforming of Ethanol reaction, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.029

G Model
CATTOD-10059; No. of Pages6
ARTICLE IN PRESS
A.E.Pastore de Lima, D.C. de Oliveira / Catalysis Today xxx (2016) xxx–xxx
4
the white line decrease with the reduction process carrying on at
constant temperature of 800 ◦C.
XANES Co K-edge measurements were also performed during
SRE tests, also showed in Fig. 1. At 400 ◦C, just before start the
SRE, the sample spectrum has basically the same features of the
final reduced sample, with a very small increase in the white line
intensity attributed to some oxidation and/or temperature effect.
When we start SRE reaction at 400 ◦C, we observe that the water and
ethanol presence doesn’t affect significantly the Co local geometry
except for another small amount of oxidation that tends to disap-
pear along the time. When the temperature is increased to 500 ◦C,
the white line intensity decrease, which indicate a cobalt reduction
due to the temperature and/or products from reaction.
The relative composition of sample presented in Fig. 2 (top) was
obtained by linear combination fitting of standards spectra. The
initial sample was obtained ∼97% of CoAl2O4. At the end of reduc-
tion at 800 ◦C, the composition is 60% of Co metal, 20% of CoAl2O4
and 20% CoO while at 25 ◦C the composition changes to 45% of Co
metal, 45% of CoAl2O4 and 10% CoO. It is worth mentioning that
the Co3O4 standard did not improve the fitting procedure. After
1 h under SRE at 400 ◦C we have ∼33% of reduced Co (metal) and at
500 ◦C it changes to ∼46%, showing structural changes as a function
of the temperature of SRE. The final sample at 25 ◦C after removing
the ethanol and water feed (Fig. 1) has ∼45% of metallic Co; beside
this, the final sample and the sample under SRE do not present
an intense pre-edge which indicate the stability of Co symmetry
and no more Td sites are formed. The concentration of Co (Td) is
stabilized at about 30% under SRE and after the reaction.
We found many reports in the literature [27] showing that the
CoAl2O4 compound is not favorable to catalytic processes because
it removes the active phase (Co) from the surface to form this sta-
ble phase, spinel, which is inactive or very few active in SRE. But
in our sample, after heating at 800 ◦C in H2 atmosphere we esti-
mated by linear combination fit that the sample is formed by 60%
of reduced cobalt (Co◦) and no Co3O4 was detected; then decreas-
ing the temperature to 25 ◦C under He after reduction we achieve
∼45% of reduced Co. This variation of reduction state indicates that
the Co structure is strongly affected by temperature even under H2
or an inert atmosphere, and the CoAl2O4 phase is not a cumber to
reduce the Co in this case.
Fig. 3. (top) CeO2 and Ce(OH)CO3 references of Ce4+ and Ce3+ respectively; (bottom)
XANES at Ce LIII-edge from Co5Ce10Al sample under different conditions along the
activation and SRE reaction.
3.1.2. XANES study at Ce LIII-edge
The XANES data from Ce LIII-edge is frequently used to character-
ize the electronic properties of ceria-based materials. The spectrum
of Ce4+ presents two intense features in XANES region [28] while
one intense feature in the white line is the main characteristic of
Ce3+ spectrum (Fig. 3, up). Fig. 3 (down) shows the XANES of Ce
LIII-edge of fresh sample, reduced sample, under SRE reaction and
after the reaction. The characteristic spectrum of Ce4+ was obtained
from our fresh sample. After the reduction, the XANES Ce LIII-edge
presents a typical intense white line of Ce3+ oxidation state. As
the data were collected in transmission mode, we observed that
all ceria was reduced from CeO2 → Ce2O3 by this activation pro-
cess. Fig. 4 (up) shows the relative composition of Ce ions along the
reduction under H2, obtained by linear combination fit of spectra in
Fig. 4 (down) with standards of Ce3+ and Ce4+. Significant changes in
Ce spectra were observed at ∼580 ◦C; we achieve 50% of Ce reduc-
tion at ∼700 ◦C and the full reduction at 800 ◦C. After reduction, at
25 ◦C, we observed a very small decrease in the white line intensity,
a small increase in the feature at ∼5737 eV and a small energy shift
(∼0.5 eV), which can indicate that some small content of Ce4+ can
be present in low temperatures even under H2. When the sample
is under SRE at 400 ◦C, 500 ◦C or under He at 25 ◦C after reaction,
no more significant changes in the Ce LIII-edge were observed. The
mixed oxidation state of Ce (Ce3+ and Ce4+) was stable in this sample
under SRE reaction conditions.
Fig. 4. (top) Relative Ce(IV) and Ce(III) fractions in the Co5Ce10Al sample as a func-
tion of reduction temperature extracted from the XANES spectra reported in the
bottom part; (bottom) evolution of the Ce LIII-edge XANES of Co5Ce10Al along the
reduction process.
Please cite this article in press as: A.E.Pastore de Lima, D.C. de Oliveira, In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to
Steam Reforming of Ethanol reaction, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.029

G Model
CATTOD-10059; No. of Pages6
ARTICLE IN PRESS
A.E.Pastore de Lima, D.C. de Oliveira / Catalysis Today xxx (2016) xxx–xxx
5
with the increasing on CO2 concentration, evidencing the water-
gas shift reaction. From 600 ◦C to 700 ◦C we observed an increasing
on CO concentration, which is expected since the shift reaction is
not favoredat higher temperatures. The highestamountof methane
was only 5% at 700 ◦C and the ethylene was detected only in 500 ◦C
in concentrations lower than 5%.
This catalyst presented a low selectivity to ethylene, which is
one of the main routes for coke formation (the polymerization of
ethylene derived from ethanol dehydration) [29]. The coke can
cover the active sites decreasing the catalytic conversion. In this
case, the Ce can be the responsible to avoid carbon deposits because
one of the important properties of cerium oxide is its oxygen stor-
age capacity, which can provide oxygen to the gas mixture in
catalytic tests or oxygen mobility over the catalyst surface [30],
removing the carbon over the active phase.
During SRE, many authors reported acidic supports like Al2O3
favor dehydration and thereby increase the tendency for coke for-
mation due to the polymerization of ethylene [31]. However, on
considered being a basic support as ceria the dehydration is limited
and its redox properties inhibit coke formation [32].
Biswas et al. [33] found that a high value of Ce/Zr ratio in
Ni–ceria–zirconia catalyst for SRE has a lower tendency for coke
formation. Many authors also reported CeO2-rich formulations to
allow high hydrogen yield [34]. Indeed, the absence of ethylene for-
mation can be the main reason for the low coke formation of our
cobalt-ceria-alumina catalysts, even with low ceria content (∼10%)
as we showed. The oxygen exchange capacity of cerium oxide is
associated with its ability to reversibly change the cerium oxida-
tion states between Ce4+ and Ce3+ [35]. This mixed oxidation state
was detected in XANES Ce LIII-edge spectra after the reduction at
25 ◦C and also along all SRE catalytic test, where the ethylene was
detected only in very low concentration.
The GC results also showed a low formation of methane, which
indicate the undesired reactions that can decrease the hydrogen
production (as decomposition of ethanol and acetaldehyde and
methanation of CO and CO2) are not favored with this catalyst.
On the basis of the above in situ and operando XANES studies and
the catalytic test results, it is possible to propose a reliable struc-
tural model, as shown in Fig. 6. The initial state of cobalt species
in the sample before reduction exists as CoAl2O4 while the ceria is
in CeO2 form (Ce4+). The reduction transforms aluminate to cobalt
oxide (CoO) and metal clusters (Co); the ceria is fully reduced from
Ce4+ to Ce3+; the cobalt species attaches on the surface of ceria
(now Ce2O3 in high temperatures). It is, therefore, concluded that
the cobalt species are dispersed on the ceria surface, interacting
with ceria in the Co5Ce10Al catalyst. During SRE up to the end of
reaction, the Co/CoO ratio increase with the increase of tempera-
ture and the Co concentration in CoAl2O4 site is kept constant; a
small amount of Ce4+ is formed; the mixture of Ce4+/Ce3+ is stable
and no CoAl2O4 formation is detected.
Fig. 5. (top) Variation of selectivity to hydrogen, CO2, CO, CH4, ethylene, acetone,
acetaldehyde as a function of temperature; (bottom) Ethanol conversion and hydro-
gen yield during SRE reaction performed with Co5Ce10Al catalyst.
In Co-alumina supported catalysts, the Co in Td symmetry is a
signature of the CoAl2O4. where Co atoms are not available as active
site to catalysis. In Co5Ce10Al sample we aimed to remove all the Co
atoms from this structure by adding ceria to our catalyst. By XANES
analysis in the Co K-edge and Ce LIII-edge, the activation process and
the presence of Ce in the support seemed to be enough to remove
most of Co from this site; later, there was no migration of Co to Td
site again by changing to different atmospheres and temperatures.
The Ce can be the responsible to remove the Co from the stable
structure of CoAl2O4 and then to keep them in Oh symmetry, being
irreversible in the tested conditions.
3.2. Catalytic reaction
During XANES measurements, the hydrogen production was
detected by mass spectrometry (MS) at the tested temperatures
(400 and 500 ◦C). In order to analyze quantitatively the ethanol
conversion and hydrogen production, we used Gas Chromatogra-
phy (GC) to perform catalytic tests with the Co5Ce10Al sample.
We attributed the catalytic activity and the structural changes after
activation to the Ce presence, which can decrease the thermal sta-
bility of CoAl2O4 and favored the atomic diffusion of Co in the
catalyst, under temperature and H2 flux.
All the products evaluated by GC are showed in Fig. 5. The quan-
titative analysis showed a significant ethanol conversion only at
400 ◦C and the dehydrogenation of ethanol to acetaldehyde can be
observed at this temperature. The higher hydrogen production is
also observed at 400 ◦C. At 500 ◦C the ethanol conversion drastically
increases, and we observe the decreasing on acetaldehyde yield
4. Conclusions
We studied the influence of Ce on Co/CeO2/Al2O3 catalyst. The
Ce presence contributed to activate the catalyst and even with the
presence of CoAl2O4 we obtained a sample with a mixture of com-
pounds which presented high hydrogen selectivity by SRE reaction.
The interaction between Co and Ce showed a significant improve-
ment in the reducibility of Co and an important hole to avoid the
formation of CoAl2O4 again.
The catalyst evaluation showed high ethanol conversion fol-
lowed by high hydrogen yield at 500 ◦C. The products analysis at
this temperature showed a reaction pathway through the acetalde-
hyde route. The CO2 selectivity showed a water-gas shift reaction
consuming CO to produce hydrogen. The methane formation was
Please cite this article in press as: A.E.Pastore de Lima, D.C. de Oliveira, In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to
Steam Reforming of Ethanol reaction, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.029

G Model
CATTOD-10059; No. of Pages6
ARTICLE IN PRESS
A.E.Pastore de Lima, D.C. de Oliveira / Catalysis Today xxx (2016) xxx–xxx
6
Fig. 6. Structural model proposed to the Co5Ce10Al catalyst in different steps of reaction. The initial state shows the presence of CoAl2O4 compound; the activation step
shows the reduction of Ce4+ to Ce3+ and the structural change in the Co site; the final structure shows a small oxidation of Ceria and Co predominant in Oh site.
below 4.5% in all tested temperatures. The methane presence can
be an issue to decrease the hydrogen production since the methane
reforming reaction requires higher temperatures. The ethylene can
be a source to carbon formation over the active sites, but this cata-
lyst also presented a low selectivity to ethylene.
[12] J.L. Contreras, J. Salmones, J.A. Colín-Luna, L. Nuno, B. Quintana, I. Cordova, B.
Zeifert, C. Tapia, G.A. Fuentes, Int. J. Hydrogen Energy 39 (2014) 18835–18853.
[13] J. Llorca, N. Holms, P.R. de la Piscina, J. Catal. 227 (2004) 556–560.
[14] J. Llorca, P.R. de la Piscina, J.A. Dalmon, J. Sales, N. Holms, Appl. Catal. B:
Environ. 43 (2003) 355–369.
[15] I.I. Soykal, H. Sohn, U.S. Ozkan, ACS Catal. 2 (2012) 2335–2348.
[16] A. Erdohelyi, J. Raskó, T. Kecskés, M. Tóth, M. Dömök, K. Baán, Catal. Today 116
(2006) 367–376.
Acknowledgements
[17] G. Busca, U. Costantino, T. Montanari, G. Ramis, C. Resini, M. Sisani, Int. J.
Hydrogen Energy 35 (2010) 5356–5366.
[18] H. Song, U.S. Ozkan, J. Catal. 261 (2009) 66–74.
[19] E. Varga, Z. Ferencz, A. Oszkó, A. Erdohelyi, J. Kiss, J. Mol. Catal. A: Chem. 397
(2015) 127–133.
[20] A.I. Frenkel, J.A. Rodriguez, J.G. Chen, ACS Catal. 2 (2012) 2269–2280.
[21] M. Fernandez-Garcia, Catal. Rev. Sci. Eng. 44 (2002) 59–121.
[22] S. Bordiga, E. Groppo, G. Agostini, J.A. van Bokhoven, C. Lamberti, Chem. Rev.
113 (2013) 1736–1850.
[23] M.A. Newton, A.J. Dent, J. Evans, Chem. Soc. Rev. 31 (2002) 83–95.
[24] F. Frusteri, S. Freni, V. Chiodo, S. Donato, G. Bonura, S. Cavallaro, Int. J.
Hydrogen Energy 31 (2006) 2193–2199.
We thank Dr. Daniela Zanchet and Felipe Moreira Pinto for the
GC analysis. We also thank the financial support from FAPESP (proc.
2013/18062-2 and 2014/2666-1) and the Brazilian Synchrotron
Light Laboratory (LNLS) at CNPEM for the use of XAFS1 beamline
installation (internal proc.). Narcizo Souza-Neto is acknowledged
for the manuscript revision.
References
[25] E.C. Wanat, K. Venkataraman, L.D. Schmidt, Appl Cat A: Gen. 276 (2004)
155–162.
[26] H.Y. Jeong, D.-C. Koh, K.-S. Lee, H.H. Lee, Appl. Catal. B: Environ. 127 (2012)
68–76.
[27] S.D. Davidson, H. Zhang, J. Sun, Y. Wang, Dalton Trans. 43 (2014)
11782–11802.
[28] X. Wang, J.A. Rodriguez, J.C. Hanson, D. Gamarra, A. Martínez-Arias, M.
Fernández-García, Top. Catal. 49 (2008) 81–88.
[29] D. Li, L. Zeng, X. Li, X. Wang, H. Ma, S. Assabumrungrat, J. Gong, Appl. Catal. B:
Environ. 176–177 (2015) 532–541.
[1] V.A. Goltsov, T.N. Veziroglu, L.F. Goltsova, Int. J. Hydrogen Energy 31 (2)
(2006) 153–159.
[2] E. Seker, Int. J. Hydrogen Energy 33 (2008) 2044–2052.
[3] C. Graschinsky, M. Laborde, N. Amadeo, A. Le Valant, N. Bion, F. Epron, D.
Duprez, Ind. Eng. Chem. Res. 49 (24) (2010) 12383–12389.
[4] J.P. Breen, R. Burch, H.M. Coleman, Appl. Catal. B: Environ. 39 (1) (2002) 65–74.
[5] M. Ni, D.Y.C. Leung, M.K.H. Leung, Int. J. Hydrogen Energy 32 (15) (2007)
3238–3247.
[6] T. Apichai, S. Tawee, T. Supaporn, Sci. Asia 27 (2001) 193–198.
[7] F. Haga, T. Nakajima, H.M.S. Mishima, Catal. Lett. 48 (1997) 223–227.
[8] J. Llorca, N. Homs, J. Sales, P.R. de la Piscina, J. Catal. 209 (2002) 306–317.
[9] M. Benito, J.L. Sanz, R. Isabel, R. Padilla, R. Arjona, L. Daza, J. Power Sources 151
(0) (2005) 11–17.
[30] D. Zanchet, J.B.O. Santos, S. Damyanova, J.M.R. Gallo, J.M.C. Bueno, ACS Catal. 5
(2015) 3841–3863.
[31] M. Badlani, I.E. Wachs, Catal. Lett. 75 (2001) 137–149.
[32] L.V. Mattos, B.H. Jacobs, F.B. Noronha, Chem. Rev. 112 (2012) 4094–4123.
[33] P. Biswas, D. Kunzru, Int. J. Hydrogen Energy 32 (2007) 969–980.
[34] M. Liao, W. Wang, R. Ran, Z. Shao, J. Power Sources 196 (2011) 6177–6185.
[35] J. Kaspar, P. Fornasiero, M. Graziani, Catal. Today 50 (1999) 285–298.
[10] H. Wang, Y. Liu, L. Wang, Y.N. Qin, Chem. Eng. J. 145 (2008) 25–31.
[11] C. Resini, T. Montanari, L. Barattini, G. Ramis, G. Busca, S. Presto, P. Riani, R.
Marazza, M. Sisani, F. Marmottini, U. Costantino, Appl. Catal. A: Gen. 355
(2009) 83–93.
Please cite this article in press as: A.E.Pastore de Lima, D.C. de Oliveira, In situ XANES study of Cobalt in Co-Ce-Al catalyst applied to
Steam Reforming of Ethanol reaction, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.029