Steam reforming of acetic acid over MgAl2O4-supported Co and Ni catalysts: Effect of the composition of Ni/Co and reactants on reaction pathways

Article abstract of DOI:10.1016/j.cattod.2017.04.023

The steam reforming of acetic acid (AcH), a model compound of pyrolysis-oil, was studied over MgAl₂O₄-supported Co-Ni catalysts with different Co/Ni ratios, prepared by impregnation using the incipient wetness method. The catalysts were characterized by X-ray powder diffraction, temperature programmed reduction, temperature programmed desorption, and thermogravimetry. In the steam reforming reaction of AcH, both Ni and Co catalysts suffered partial oxidation due to contact with the reaction mixture. The ketonization reaction occurred on the MgAl₂O₄ support and the presence of Co or Ni changed the reaction pathway of species adsorbed on the support, which suppressed the formation of acetone. The cleavage of C[sbnd]C bonds was favored on the Ni surface, resulting in CH_x species, which in the presence of H₂ were preferentially hydrogenated to CH₄ at low temperatures. On the other hand, on the Co surface the CH_x species were decomposed to C* and H* as the temperature increased. The addition of Co to Ni catalysts inhibited CH₄ formation and carbon accumulation.

Full text of DOI:10.1016/j.cattod.2017.04.023

Catalysis Today xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Steam reforming of acetic acid over MgAl O -supported Co and Ni catalysts:
2
4
Effect of the composition of Ni/Co and reactants on reaction pathways
a
a
b
a
Stefanie C.M. Mizuno , Adriano H. Braga , Carla E. Hori , João Batista O. Santos ,
José Maria C. Bueno
a
a,⁎
Department of Chemical Engineering, Federal University of São Carlos (UFSCar), P.O. Box 676, 13565-905 São Carlos, São Paulo, Brazil
Faculty of Chemical Engineering, Federal University of Uberlandia (UFU), 38408-144 Uberlandia, Minas Gerais, Brazil
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
2 4
The steam reforming of acetic acid (AcH), a model compound of pyrolysis-oil, was studied over MgAl O -
Steam reforming
Acetic acid
Catalyst
Hydrogen
Bimetallic catalyst
Nickel
supported Co-Ni catalysts with different Co/Ni ratios, prepared by impregnation using the incipient wetness
method. The catalysts were characterized by X-ray powder diffraction, temperature programmed reduction,
temperature programmed desorption, and thermogravimetry. In the steam reforming reaction of AcH, both Ni
and Co catalysts suffered partial oxidation due to contact with the reaction mixture. The ketonization reaction
occurred on the MgAl
on the support, which suppressed the formation of acetone. The cleavage of CeC bonds was favored on the Ni
surface, resulting in CH species, which in the presence of H were preferentially hydrogenated to CH at low
temperatures. On the other hand, on the Co surface the CH species were decomposed to C* and H* as the
temperature increased. The addition of Co to Ni catalysts inhibited CH formation and carbon accumulation.
2 4
O support and the presence of Co or Ni changed the reaction pathway of species adsorbed
Cobalt
x
2
4
x
4
1
. Introduction
candidates for steam reforming due to their high ability to break CeC
bonds [4,5]. However, the high cost of noble metals makes them
impractical for use on an industrial scale. On the other hand, metals
such as Ni and Co have the advantages of low cost and effectiveness in
breaking CeC bonds. Nonetheless, their performance suffers from
strong deactivation due to extensive carbon deposition and (especially
in the case of Co catalysts) deactivation due to oxidation of the active
metal [6–11]. Approaches such as the use of bimetallic catalysts,
modifying the support, have been employed to minimize deactivation
by carbon deposition and by oxidation. As an example, bimetallic Ni-Co
catalysts have shown better catalytic performance for steam reforming,
compared to monometallic Ni or Co catalysts [6–9,12].
Most of the energy used in the world comes from fossil fuels, which
are non-renewable and unsustainable. Economic growth strongly
influences world energy consumption, and energy demand is growing
rapidly. In order to sustain this economic growth, we need to develop
2
new renewable energy sources. Hydrogen (H ) produced from renew-
able sources is a viable alternative energy carrier because it has high
energy potential and can be employed to produce clean energy in fuel
cells [1].
Fast pyrolysis technology makes it possible to convert biomass to
bio-oil, which is a mixture of oxygenated compounds including acids,
alcohols, ketones, esters, ethers, aldehydes, and aromatics. Bio-oil is a
convenient feedstock in terms of its storage, transport, and processing
characteristics [1]. Catalytic steam reforming of bio-oil is an attractive
Many studies have shown that the support affects catalyst activity
and stability during steam reforming reactions [13]. For instance,
Al
2
O
3
-supported metal catalysts exhibit high selectivity towards H
2
in
way to produce sustainable H
2
[2,3]. However, it is very difficult to
steam reforming of acetic acid (SRAcH), but Al O
2 3
catalyzes dehydra-
design catalysts based on kinetic and spectroscopic studies of steam
reforming of bio-oil, due to its complexity. Therefore, representative
model compounds are often used and in this work, acetic acid was
chosen as a model compound to be used in steam reforming, since it
could represent the properties of acid bio-oil components that are
highly reactive with oxide supports.
tion reactions that lead to deactivation of the catalyst. Basic supports,
such as ZnO and MgO, have been used to avoid dehydration reactions.
However, other reactions that occur on the basic supports include
ketonization and oligomerization, which deactivate the catalyst due to
coke formation. Acetone is known to be readily produced from acetic
acid, and the reaction is catalyzed by a variety of basic oxides [14]. In
addition, the support used can affect the metal-support interface,
Catalysts based on noble metals such as Ir, Ru, and Rh are good
⁎
Corresponding author.
E-mail address: jmcb@ufscar.com.br (J.M.C. Bueno).
http://dx.doi.org/10.1016/j.cattod.2017.04.023
Received 13 January 2017; Received in revised form 29 March 2017; Accepted 10 April 2017
0
920-5861/ © 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Mizuno, S.C.M., Catalysis Today (2017), http://dx.doi.org/10.1016/j.cattod.2017.04.023

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
altering metal-support interactions as well as the reaction pathways
presence of metal in the supported catalyst may dominate the reaction
kinetics. One important parameter is the metal loading, which in the
case of noble metal catalysts is typically small (1%), hence favoring the
direct participation of the support in the reaction mechanism.
Considering that the effect of the support in the reaction pathways
for SRAcH is not clear, and that the acetone is formed by condensation
[
13].
Takanabe et al. [15,16] showed that on Pt/ZrO
occurred according to a bifunctional mechanism. On Pt black, pure
acetic acid decomposed to form H , CO , CO, carbon, and CH . The
formation of CH was attributed to hydrogenation of CO . Acetone and
ketene were observed during the reaction, with their formation
potentially occurring on coked Pt catalysts. Ketene and CH were
observed in small amounts during experiments using pure acetic acid on
ZrO , which could be explained by the dehydration of acetic acid or the
decomposition of acetone formed. Typical steam reforming products
were observed during SRAcH on Pt/ZrO , suggesting that acetic acid
activation occurred on Pt, while steam activation occurred on ZrO
Takanabe et al. [15,16] concluded that the reaction took place at the Pt
periphery, in close proximity to ZrO , and that the role of ZrO was in
the activation of water to form reactive hydroxyl groups.
Lemonidou et al. [17] showed that on 0.5 wt.% Rh supported on
/CeO -ZrO , reforming and water gas shift reactions occurred
even below 573 K, yielding H and CO . Acetic acid adsorbs dissocia-
tively on Rh, forming acetates that then decarboxylate to form surface
methyl groups, which are transformed to CO, CO , and H . According to
Lemonidou et al. [17], ketonization and decarboxylation of acetic acid
occurred on CeO -ZrO up to 873 K. In addition, they proposed that
2
catalysts, SRAcH
2
2
4
4
x
2 4 2 4
on MgAl O during SRAcH, we present a study of the effects of MgAl O
4
and metal composition (Ni, Co, and Ni-Co) on the reaction pathways for
SRAcH. The results reported in this work strongly indicate that the
presence of metal changes the reaction pathway of acetic acid activated
on the support.
2
2
2
.
2. Experimental section
2
2
2.1. Catalyst preparation
The catalysts were prepared by impregnation of Ni(NO ) ·6H O and
3
2
2
La
2
O
3
2
2
3 2 2 2 4
Co(NO ) ·6H O onto a MgAl O support synthesized according to the
2
2
sol-gel method, as previously described by Ávila-Neto et al. [11]. The Ni
and Co salts were dissolved in ethanol in order to give total metal
loadings of 8 wt.% of Co and Ni, and 20 wt.% only of Ni. In the case of
the bimetallic samples, the Co and Ni salts were diluted simultaneously
in ethanol to obtain samples with 4 wt.% Co–4 wt.% Ni, 2.5 wt.% Co-
2
2
2
2
acetate was formed on Rh crystallites and that acetyl was formed on the
periphery of the Rh particles. The ketonization reaction could only
occur on the periphery of the Rh crystallites.
DFT calculations have been used to propose the mechanisms for
SRAcH employing Co [18,19], Pt [20], and Ni [21] catalysts. The
results indicated that CeH and CeO scission are favored, relative to
CeC scission. For Co(111) and Ni(111) surfaces, acetic acid decom-
5
.5 wt.% Ni, and 5.5 wt.% Co-2.5 wt.% Ni. The support was then added
to the solution, under stirring, and the final material was dried at room
temperature under vacuum. The samples were then calcined at 550 °C
for 6 h, with heating from room temperature at a rate of 3 °C min
under a flow of synthetic air at 150 mL min . The catalysts were
labeled as 8Ni, 4Co4Ni, 2.5Co5.5Ni, 5.5Co2.5Ni, 8Co, and 20Ni,
representing 8Ni/MgAl
5·5Co2·5Ni/MgAl O 8Co/MgAl O , and 20Ni/MgAl O , where the
−
1
,
−1
2 4 2 4 2 4
O , 4Co4Ni/MgAl O , 2·5Co5·5Ni/MgAl O ,
3
position started with formation of the acetate (CH COO) intermediate,
2
4,
2
4
2 4
which adsorbed on the surface to form a bridge configuration involving
two O atoms bonded to the catalyst surface on the top site [19].
However, another path was also suggested, involving the formation of
numbers indicate the loadings of the metals.
2.2. Characterization
acyl species (CH
The SRAcH reaction was studied using a 3.2 wt.% Ni/CeO
catalyst, with S/C molar ratio of 5 [22]. Acetone was the main product
during SRAcH on CeO -ZrO , with the catalyst being strongly deacti-
vated, while on the CeO -ZrO -supported Ni catalyst, the conversion of
AcH exceeded 75% after 45 h of reaction, with stable selectivity
towards CO . Rapid initial deactivation of the catalyst was attributed
3
CO).
2
-ZrO
2
XRD patterns were obtained with a Rigaku Multiflex instrument,
using Cu Kα radiation, in the 2θ range from 10 to 90°, with step size of
.1°. The peaks were identified by comparison with the diffraction
patterns of cubic MgAl , cubic Co , fcc NiO, and cubic NiCo
obtained from the ICSD database.
2
2
0
2
2
2
O
4
3
O
4
2 4
O ,
x
The in situ reduction profiles of the calcined samples were evaluated
by temperature-resolved X-ray diffraction. The diffraction peaks were
measured at the XPD beamline of the Brazilian National Synchrotron
Light Laboratory, using wavelengths of 1.632 Å for the Co-containing
catalysts and 1.512 Å for the 8Ni-containing samples, which correspond
to energies immediately below the absorption K-edges of Co (7.6 eV)
and Ni (8.2 eV). This was performed to avoid absorption of incoming X-
ray beams and a fluorescence background. The detection was per-
formed with a linear detector (Mythen), and each diffraction data set in
to coking and the presence of strongly adsorbed species on the support
surface. During SRAcH, a very low yield of acetone was observed for the
supported Ni catalyst, compared to the bare support. The authors [22]
also observed the total yields of CO
x
and CH
4
during steam reforming of
acetone on 3.2 wt.% Ni/CeO -ZrO
2
2
at 973 K. It was concluded that if
acetone was formed on the Ni catalyst during SRAcH, it would be
expected to be converted to syngas.
We recently reported the effects of the support on the reaction
pathways for steam reforming of ethanol and acetone [12,13]. For
steam reforming of acetone on Ni, Co, and Ni-Co supported on
−
1
the Q range of 1.8–3.9 Å
took around 1 min to be acquired. The
reduction was performed from room temperature up to 750 °C, main-
−1
MgAl
of reduction of the Co-Ni nanoparticle surface. During steam reforming
with H in the feed, the Ni catalyst was reduced at low temperatures
200–300 °C) and was active for cleavage of the CeC bond of acetone,
forming CH . At high temperatures, the CH fragments formed in the
acetone activation were dehydrogenated to produce *C and H . The Co-
2
O
4
, the reaction pathway was strongly influenced by the degree
taining the final temperature for 1 h, under a flow of 100 mL min of
% H in He. Fittings of the (311) reflection of spinel Co , at around
2.57 Å , the (200) reflection of rock-salt oxides (CoO or NiO), at
5
2
3 4
O
−1
2
−1
(
around 2.96 Å , and the (200) reflection of metallic Ni and Co species,
−
1
4
x
at 3.50 Å , were performed using a pseudo-Voight function. The
molar fractions were calculated based on the integrated intensities of
2
containing catalysts remained oxidized and inactive at low tempera-
tures ( < 300 °C). At high temperatures, the catalysts became active
these peaks. The fitting of NiCo
low intensity of the (311) reflection and close proximity to the same
reflection from MgAl . Furthermore, differentiation of NiO and CoO
2 4
O mixed spinel was not possible due to
for dehydrogenation of CH
3
* species, while the hydrogenation of CH
x
*
2 4
O
to CH was not favored at high temperatures. We suggested that the
4
segregated phases, or mixed oxides, was not possible due to similarities
in the crystalline structure. Finally, the distribution of species was
expressed in terms of the molar fractions of NiO and Ni for the 8Ni
catalyst, Co O , CoO, and Co for the 8Co catalyst, and CoOeNiO and
3 4
CoeNi alloy for the bimetallic catalyst.
strong decrease in C accumulation for the Co-containing catalysts was
due to higher reaction rates for oxidation of *C by O and OH, as well as
the presence of CoO on the metal cores of the Co and Co-Ni catalysts.
It is important to remark that although the nature and structure of
the support may have a clear influence on the reaction pathways, the
Temperature-programmed reduction (TPR) was performed with a
2

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 1
a
Abundance of fragments for acetic acid reaction products used in the analysis of the TPD data .
Species
Mass fragments
2
12
15
15
16
27
8
28
2
29
5
39
6
41
4
42
10
43
100
44
45
58
25
60
76
Acetone
H
2
100
CO
CH
CO
5
5
10
2
100
12
100
2
4
2
80
18
12
4
1
6
100
2
92
HAc
3
100
a
Obtained based on NIST libraries, using electron ionization.
Micromeritics AutoChem II 2920 system. Prior to reduction, the
samples were treated at 200 °C under a flow of N . A 50 mg portion
of catalyst and support was placed in a quartz U-reactor and reduced by
550 °C. The spectra of the used samples were acquired using a Thermo
Scientific IS50 FT-IR spectrometer equipped with a mercury cadmium
telluride (MCT) detector and employing a diffuse reflectance Harrick
cell with ZnSe windows. The IR spectra were collected at room
temperature, with resolution of 4 cm and averaging 64 scans, which
took around 1 min.
2
−1
heating from room temperature up to 1000 °C under a 30 mL min
flow of 10% H in N . The H consumption was measured with a
−1
2
2
2
previously calibrated thermal conductivity detector.
Temperature-programmed desorption of acetic acid was performed
using the same TPR system. The samples (150 mg) were reduced at
2.4. Stability tests
−
1
2 2
750 °C for 1 h under a flow of 30 mL min of 10% H in N , followed
by cooling to 50 °C under a flow of He. The adsorption of HAc was
performed at this temperature, with pulses of acetic acid being
introduced to the sample by means of a calibrated loop (0.5 mL), until
Stability tests were performed at 550 °C using a fixed bed reactor.
−1
The catalysts were previously reduced under a 50 mL min flow of H
with heating from room temperature to 750 °C at a rate of 10 °C min
2
,
1
.
−
−1
saturation, using a 30 mL min
flow of He. The catalysts were then
The reactant mixture with composition H
2
O/acetic acid/He = 2.07/
acid molar ratio = 4;
aceticacid = 1.55 kPa) was fed to the reactor to obtain a W/Faceticacid
flushed with He for 30 min, after which the temperature was ramped
0
P
.51/97.42
vol.%
2
(H O/acetic
−1
from 50 °C to 900 °C at a rate of 10 °C min . The products of acetic
acid decomposition were analyzed using a Pfeiffer PrismaPlus mass
spectrometer.
−1
ratio of 6.4 gcat min gacetic acid . The reactor was maintained at 550 °C
for 6 h. Analysis of the reactor outflow was performed as described
above. The spent catalysts were submitted to thermogravimetric
analysis and the carbon accumulated was calculated from the first
derivative of the weight loss curve. Thermogravimetric analysis was
performed with a Shimadzu SDT 2960 DTA-TGA instrument. The
temperature was ramped from room temperature to 1000 °C, at a rate
The mass abundances obtained using electron ionization were
employed to interpret the TPD data, as several fragments showed
equivalent mass/charge ratios (Table 1). Based on the intensities, m/z
4
3 was selected for acetone, instead of m/z 58, due to its greater
intensity, while m/z 15 was used for methane, instead of m/z 16,
because the ionization of water, CO, and CO yielded fragments with
m/z 16. The mass fragment related to CO could not be separated from
the influence of a highly intense mass fragment due to CO , so m/z 28
considered the sum of the two carbon oxides. The following mass/
charge ratios and corresponding fragments were determined: 2 (H ); 4
He); 28 (CO + CO ); 15 and 16 (CH ); 18 (H O); 43 (acetic acid
acetone); 58 (acetone); 44 (CO ); and 60 (acetic acid).
−
1
−1
2
of 10 °C min , under a 30 mL min
. Results and discussion
2 4
.1. XRD analysis of the MgAl O support and catalysts
flow of synthetic air.
2
3
2
3
(
2
4
2
+
2
Fig. 1 shows the XRD patterns obtained for the previously calcined
MgAl support and the supported xCo-yNi catalysts containing x
+ y = 8 at various x/y ratios and 20 wt.% of Ni. As reported previously
11,12], the diffraction peaks of the support at 2θ equal to 31.3, 36.9,
44.9, 59.5, and 65.4° correspond to the spinel MgAl structure (ICSD
2 4
O
2.3. Catalytic tests
[
The catalytic tests were carried out in a continuous-flow tubular
fixed-bed quartz reactor at atmospheric pressure. Initially 100 mg of
2 4
O
collection code 40030). The calcined 8Co, 4Co4Ni, 2.5Co5.5Ni, and
5.5Co2.5Ni samples showed diffraction peaks characteristic of spinel
−
1
catalyst was previously reduced, under a 50 mL min flow of H
2
, with
−1
heating from room temperature to 750 °C at 10 °C min . The samples
were kept at the maximum temperature for 1 h, after which the reactor
structure MgAl
spinel Co and/or NiCo
2 4
lapping those of the MgAl O
2
O
4
, but the results could not exclude the formation of
structures with diffraction peaks over-
support. The 20Ni sample showed peaks
3
O
4
2 4
O
was cooled down to room temperature and the H
the SRAcH reaction mixture. The experiments were performed using
two reactant molar compositions: i) H O/acetic acid/He = 2.07/0.51/
7.42 vol.%; and ii) H O/acetic acid/He/H = 2.07/0.51/89.0/8.33
2
flow was replaced by
characteristic of the support and the NiO face-centered cubic structure
(JCPDS 75–1197), at 2θ equal to 43.2, 62.9, 75.3, and 79.3°. The 8Ni
sample showed peaks characteristic of the support, but did not show the
NiO face-centered cubic structure, which could be due to the low metal
content and it could indicate a high degree of metal dispersion.
Co-Ni bimetallic catalysts supported on different oxides have been
investigated by XRD. Szijjarto et al. [23] reported that the addition of
2
9
2
2
vol.%. The hydrogen in the feed was used for the purpose of hydro-
genation of possible intermediaries and decreased of their reactivity. In
both series of experiments, the following conditions were used: H
2
O/
acetic acid molar ratio = 4, Paceticacid = 1.55 kPa, and W/Faceticacid
−
1
ratio = 6.4 gcat min gacetic acid
(W = weight of catalyst; F = acetic
Co to Ni/MgAl
reflections, which was attributed to the formation of Ni-Co mixed
oxides. Althenayan et al. [24] described the formation of NiCo
Co , NiO, CoAl and NiAl phases on Co-Ni/Al oxides.
2 4
O oxides resulted in a slight shift in the NiO fcc
acid mass flow). Steam and acetic acid were introduced into the reactor
using two temperature-controlled saturators, with He as a carrier gas.
The two gas streams were saturated with water and acetic acid at 37
and 20 °C, respectively. The temperature of the catalytic bed was
increased from room temperature to 650 °C in steps of 50 °C.
2 4
O ,
3
O
4
2
O
4
2
O
4
2 3
O
Nevertheless, the Co and Ni aluminates show the same spinel structure
and detailed XRD quantification is not possible. Hence, it is complex to
attribute spinel species in a nanostructured material based on XRD
The MgAl
2 4 2
O support was analyzed by FT-IR after SRAcH (H O/
acetic acid/He = 2.07/0.51/97.42 vol.%) for 1 h at 350, 400, 450, and
2 4
measurements. The formation of NiO phase on MgAl O with 8 wt.% of
3

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
2 4
Fig. 1. XRD patterns of the synthesized MgAl O support and the supported Ni, Co-Ni,
and Co samples. Star symbol: XRD pattern of NiO obtained from the Inorganic Crystal
Structure Database (collection code: 9866).
NiO was reported by Braga et al. [12]. NiO reflections were also
observed in this work (Fig. 1), but with lower intensity, indicating
greater dispersion than that observed by Braga et al. [12]. Quantifica-
tion of the individual NiO and CoO oxides was achieved by tempera-
ture-resolved X-ray diffraction as described below.
3.2. Temperature-programmed reduction
Fig. 2 presents the TPR profiles of the support and the xCo-yNi
catalysts. The MgAl O support (Fig. 2a) did not show any H uptake.
2 4 2
The 8Ni sample (Fig. 2b) showed a low temperature peak at 265 °C and
a broad reduction peak at a high temperature (about 710 °C). The peak
at low temperature could be attributed to the reduction of bulk NiO
clusters weakly bound to the support, while the reduction peak at
Fig. 2. TPR profiles of the catalysts: (a) MgAl
Co4Ni, (f) 2.5Co5.5Ni, and (g) 20Ni.
2
4
O , (b) 8Ni, (c) 8Co, (d) 5.5Co2.5Ni, (e)
4
710 °C could be assigned to the reduction of smaller NiO particles in
strong interaction with the support. Parravano [25] observed that the
reduction of pure NiO occurred at temperatures as low as 155 °C, while
reduction peaks between 500 and 800 °C were observed for the
the metal oxide weakly bonded to the support, while the high
temperature peak could be assigned to the reduction of metal oxide
in strong interaction with the support. Similar behavior was observed
for the other bimetallic samples. It can be seen in Fig. 2 that the
reduction of metal oxide at high temperature started at lower tempera-
tures for the bimetallic Ni-Co catalysts, compared to the monometallic
Ni catalyst. In addition, the reduction peak at high temperature shifted
to lower temperatures with increasing Ni content. Previous results
reduction of NiO supported on MgAl
2 4
O catalysts [26,27]. For 20Ni
(
Fig. 2g), the TPR showed a typical profile of large NiO particles with
different sizes, which were reduced at low temperature, and small NiO
particles that were reduced at about 650 °C. A similar effect was
observed by Hadian et al. [28]. The results presented here are in
agreement with literature data reported for TPR of NiO/γ-Al
2
O
3
and
[
33,34] showed that the lower reduction temperatures of bimetallic
NiO/MgAl
2 4
O [26,27,29,30].
catalysts, relative to those of monometallic catalysts, could be attrib-
uted to the formation of Ni and Co oxide species on the surfaces of the
bimetallic catalysts, which were more accessible than those in the bulk
structures.
The reductions of the 8Ni, 8Co, and bimetallic 4Co4Ni samples were
also analyzed by temperature-resolved X-ray diffraction. Fig. 3 shows
the contour maps obtained and the corresponding reduction profiles.
The reduction of NiO to metallic Ni in the 8Ni catalyst (Fig. 3a and b)
started at around 600 °C and increased with increasing temperature,
reaching complete reduction at 750 °C. The reduction of CoO to
metallic Co in the 8Co catalyst (Fig. 3e and f) started at about 750 °C
with rock-salt structure oxides such as CoO remaining present for a long
The 8Co sample (Fig. 2c) presented two low temperature H
2
uptake
to
→ CoO → Co ). According to
Jacobs et al. [31], the reduction of CoO species is markedly slower than
the reduction of Co . A peak at 825 °C was also observed for the 8Co
3 4
peaks, at 316 and 371 °C, characteristic of the reduction of Co O
0
0
3 4
Co , which occurs in two steps (Co O
3 4
O
sample, which could be explained by the reduction of small CoO
0
particles to Co , and probably the reduction of cobalt aluminate-like
compounds [32].
The TPR profiles for the bimetallic samples are shown in Fig. 2d–f.
The 5.5Co2.5Ni sample presented a shoulder at about 181 °C, two peaks
at low temperatures of 311 and 410 °C, and a high temperature peak at
780 °C. The low temperature peaks could be ascribed to the reduction of
4

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 3. Temperature-resolved X-ray diffraction analyses during heating of calcined precursor samples under a flow of H
2
. Contour maps of the diffractograms (left) for (a) 8Ni, (c) 4Co4Ni,
and (e) 8Co. Molar fractions of species identified in the diffractograms, as a function of temperature (right), for (b) 8Ni, (d) 4Co4Ni, and (f) 8Co.
time without undergoing reduction to metallic species. In the case of
the bimetallic mixed NiOeCoO oxide (in 4Co4Ni; Fig. 3c and d), the
reduction started at about 600 °C, reaching complete reduction at
structural defects in the Co-Ni mixed oxide with spinel structure
(NiCo ) could facilitate the reduction of this spinel oxide to rock-
salt structures, because the presence of defects can lead to oxygen
vacancy, easily activating hydrogen [35]. The monometallic 8Co
2 4
O
750 °C, and was faster than 8Ni catalyst. The greater presence of
5

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
2 4
Fig. 5. DRIFTS spectra collected at room temperature for the MgAl O support after 1 h
under SRAcH conditions at 350, 400, 450, and 550 °C.
3.3. Temperature-programmed desorption of acetic acid
Fig. 4 shows desorption profiles of the chemical compounds over the
Ni-Co catalysts and the MgAl
2
O
4
support. The mass fragment with m/
z = 15 was used to represent CH
4
. For the other products desorbed, the
most abundant fragments are provided in Table 1.
The TPD profiles obtained for the 8Ni, 8Co, and 4Co4Ni catalysts
2 2
showed similar desorption profiles, with desorption of H , CO, CO ,
acetone, and CH
occurred at around 400 °C, while CO
4
. Over the MgAl
2
O
4
support, acetone desorption
desorption was observed at
2
around 250 °C. The higher desorption temperature for acetone, com-
pared to CO , indicated that intermediates in acetone formation were
produced at low temperatures, with desorption of CO . Desorption of
acetone was not expected at a high temperature of around 400 °C, and it
is reasonable to suppose that the intermediates formed at low tempera-
tures on the support only recombined to form acetone at higher
temperatures.
2
Fig. 4. TPD spectra for the decomposition products of acetic acid on the Co, Co-Ni, and Co
catalysts, indicating the mass fragments used for each product.
2
catalyst showed strong interaction of CoO with the support, since 20%
of this species still remained at 750 °C. The TPR results showed that
greater Ni loading in the bimetallic samples decreased the temperature
of reduction of species at high temperature (Fig. 2), suggesting the
formation of NiOeCoO mixed oxides in the bimetallic catalysts. Kuboon
Fig. 5 shows the FT-IR spectra of the MgAl
exposure under a flow of acetic acid/H O for 1 h at various tempera-
tures. Bands at around 3016, 2983, and 2936 cm
2 4
O support after
2
−1
were observed at
50 °C. The intensities of these bands increased with temperature from
50 to 400 °C, and the bands were suppressed at 450 °C. Well-defined
et al. [36] pointed out that mixed spinel phases such as NiCo
2 4
O or
3
3
Ni eCo are easily formed in systems heated to around 500 °C in
3
O
4
3 4
O
air. Continuous heating at higher temperature (800 °C) can easily form
rock-salt structures of CoOeNiO [28–30]. Since the lattice parameters
of NiO and CoO are 4.178 and 4.263 Å, respectively, the mixed oxides
could be formed during the reduction of Ni-Co spinels. In previous
works [9,12], investigation of the reduction profiles of Co-Ni catalysts
using X-ray absorption spectroscopy (XAS) demonstrated the reduction
−1
bands at 3016 and 2983 cm
stretching of adsorbed acetate. A band at 2936 cm
attributed to symmetric CeH stretching [37]. At 350 °C, other bands
were observed at around 1639 (sh), 1600, 1560 (sh), 1466, and 1440
could be assigned to antisymmetric CH
−1
could be
3
−1
−1
(
sh) cm . The band at 1639 cm
could be assigned to C]O
stretching of acetyl species [23]. When the temperature was increased
of mixed spinel Ni
x
Co(1-x)
O
4
to CoOeNiO mixed oxides, with further
−1
to 550 °C, the bands at 1560 and 1440 cm remained stable, while the
simultaneous reduction of NiO and CoO upon heating under H
2
.
x
CH bands disappeared, indicating that the stable bands could be
characteristic of carbonate species [38,39]. Increasing the temperature
revealed correlation between the intensities of the bands at 1600 and
6

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 6. Acetic acid conversion and distribution of products formed during the steam reforming of acetic acid, as a function of temperature. Paceticacid = 1.55 kPa; H
molar ratio = 2.07/0.51/97.42 (H O/acetic acid molar ratio = 4). Catalysts reduced at 750 °C.
2
O/acetic acid/He
2
1
466 cm−1, which could be assigned to two CeO stretching vibrations
CH
CH
2
COO* → CH
2
* + CO
2
*
(3)
(4)
of the COO group of adsorbed acetate [23,40]. These bands were
present for the material treated at 450 °C, while the bands correspond-
ing to CH species were absent, which was characteristic of acetic acid
x
adsorbed as a monolayer at low coverage [40]. The FT-IR results
corroborated the TPSR data, indicating that at about 400 °C, the acetate
adsorbed on the support condensed to form acetone.
2
* + 2H* → CH
4
Although the reactions of CH
hydrogenation of CH * species (Eq. (4)) are thermodynamically favor-
able [42,43], the decomposition of CH COO* species to CO (Eq. (3))
was not found to be favorable on a Co stepped surface [18]. Therefore,
these results could not explain our finding of the concomitant formation
of H
effects, with activation of CH
sidered in the DFT calculations. The TPD results, showing that
CH COOH was activated on the support, with formation of acetone,
while acetone was not formed on the metal-containing catalysts,
indicated that the reaction pathway was modified by the presence of
metal. This could be one apparent disagreement between the experi-
mental and calculation results. It is reasonable to suppose that at low
3
COOH activation (Eq. (1)) and the
2
2
2
Interestingly, acetone desorption was not observed for the Co- and
Ni-containing catalysts, with only desorption of H
being observed. These results strongly indicated that the metal presence
changed the reaction pathway for activation of acetic acid on the
support. The similar temperatures for desorption of CO
suggested that the formation of these products was related. The
different temperatures for adsorption of CO and CO, and lower
temperature for CO desorption, were not indicative of strong adsorp-
tion of CO on a support with basic properties [41] and suggested that
CO and CO were formed in distinct reaction routes. The extended
temperature ranges and the desorption profiles of H
2
, CO, CO
2
, and CH
4
2
, CH
4
, and CO
2
at low temperatures. Nevertheless, the interface
3
COOH on the support, were not con-
2 4
and CH
3
2
2
2
2
indicated that H
2
3
temperatures, the CH COO* activated on the support decomposed in
the metal-support interface, according to the reaction sequence:
2
was formed by different reactions related to the formation of CO
2
and
CO.
*
CH
3
COOH → CH
3
COO* + H*
(5)
The DFT calculation [18] for decomposition of acetic acid on a Co
stepped surface showed that the most kinetically favorable decomposi-
tion pathway involved the formation of CO according to the reaction
CH
3
COO* → CH
3
* + CO
2
*
(6)
sequence: CH
111) surface, the most favorable pathway found was:
CH COOH → CH COO → CH CO → CH C → CH →CH → CH [21];
while for Co(111) it was: CH COOH → CH CO → CH CO → CH → CH
18]. These results suggest that formation of the CH COO species
depends on the surface structure and the nature of the metal. Although
the route via CH COO was not feasible on the Co stepped surface, which
represents a metal surface with defects more similar to our real catalyst,
the presence of CH COO and CH COO could be considered as pre-
cursors of CO formation. The concomitant formation of CO , H , and
, observed in the TPD experiments, could involve reactions such as:
3
COOH → CH
3
CO → CH
2
CO → CH
2
→ CH. On a flat Ni
*
Considering that the hydrogenation of CH
x
species on a metal is
(
favorable at low temperatures [42,43], the TPD results suggest that the
reaction involving metal oxide could have been responsible for the
decomposition of acetic acid to CH , H , and CO at low temperatures.
4 2 2
3
3
3
3
3
2
3
3
2
2
[
3
3
.4. Performance of the Ni, Co-Ni, and Co catalysts in the steam reforming
3
of acetic acid
3
2
Figs. 6a and 7a present the TPSR results for acetic acid conversion
and the composition of products formed on the support, performed at
2
2
2
CH
CH
4
3
an H
in the reactor feed, respectively. The results show that independent
of the presence of H , there was formation of CO , CH , and acetone at
2
O/acetic acid molar ratio of 4/1, with and without the addition of
H
2
COOH* → CH
2
COOH* + H*
(1)
2
2
4
temperatures in the region of 350–400 °C. These products were in
agreement with those observed in TPD profiles at temperatures higher
CH
2
COOH* → CH
2
COO* +H*
(2)
7

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 7. Acetic acid conversion and distribution of products formed during the steam reforming of acetic acid, as a function of temperature, with H
aceticacid = 1.55 kPa; H O/acetic acid/He/H molar ratio = 2.07/0.51/89.0/8.33 (H O/acetic acid molar ratio = 4). Catalysts reduced at 750 °C.
2
in the reactor feed.
P
2
2
2
than about 350 °C, while the CO
experiments at low temperatures were not observed in TPSR. The
hydration of the support surface due to the presence of H O in the TPSR
experiments could provide an explanation for the differences in the
decomposition of acetic acid at low temperatures.
2
and CH
4
products observed in the TPD
The temperatures for initiation of acetic acid conversion, using the
reactants acetic acid/H O, were similar for the catalysts with 8 wt.% of
metal and different Co:Ni ratios (Figs. 6d and 6e). This suggested that
the metal oxides on a core of Co-Ni were reduced by acetic acid at
similar temperatures for different Co-Ni compositions [12]. However, in
2
2
In the case of the 8Ni catalyst (Fig. 6b), the activity started at about
2
the presence of H , the temperature for AcH activation was lower for
3
50 °C, with formation of H
2
, CO, and CO
2
as the main products,
. The TPD results
Fig. 4) for the 8Ni catalyst indicated that products such as CO , H
, and CO would be expected to be formed at temperatures below
in the
reactor feed (Fig. 7b) showed that the activity started at a low
temperature of around 250 °C, with formation of CH (in higher
amount), CO, and CO , in agreement with the TPD results. These
results suggest that the metal at surface available to reaction is partially
oxidized by the H O present in the reactor feed. In fact, previous in situ
the 8Ni catalyst (Fig. 7b) and increased with Co content (Figs. 7c and
7d). For the 8Co catalyst, the temperature for initiation of acetic acid
together with small amounts of acetone and CH
(
CH
3
4
2
2
,
conversion was similar with (Fig. 6f) and without (Fig. 7c) H
feed. These findings suggested that at low temperatures, with increas-
ing Co content, the oxidation of the metal by H O at the surface of the
nanoparticles was stronger, relative to the reduction by H . Interest-
ingly, the TPSR results obtained with H in the reactor feed (Fig. 7)
showed that the selectivity to CH decreased with increasing Co
content. The hydrogenation of CO , CO, and CH species formed by
scission of the CeC bond of the acetic acid molecule is favored by the
presence of Ni [42,44]. The selectivity to CH increased with tempera-
ture, with a maximum reached at about 450–500 °C, while a decrease at
higher temperatures indicated that the decomposition of CH species,
with successive abstraction of H, was favored over hydrogenation to
CH . A significant increase in CO formation at high temperature (Fig. 7)
2
in the
4
50 °C. However, the TPSR results for the 8Ni catalyst with H
2
2
2
4
2
2
4
2
x
2
XAS experiments [12] demonstrated that a fraction of the Ni was
oxidized at low temperatures in the presence of the reactants acetone
and H
4
2
O, and that the Ni catalyst was reduced by acetone at
x
temperatures above 350 °C.
Interestingly, for the 20Ni catalysts (Fig. 6c) in the presence of
4
acetic acid/H
2
O, the activity started at about 300 °C, which was lower
could indicate the occurrence of the reverse shift reaction [45,46].
The selectivity to acetone was very low. According to the TPD
results (Fig. 4), acetone formation would be expected to be suppressed
in the presence of metal, indicating a change in the reaction pathway,
with the species formed by activation of acetic acid on the support,
which are precursors of acetone, migrating to the metal surface to be
than the temperature of 350 °C observed for the 8Ni catalyst (Fig. 6b). It
has been shown that the oxidation-reduction equilibrium of metal
nanoparticles depends on particle size, with the oxidation being favored
with smaller nanoparticles [12]. The results indicated that the surface
of the 8Ni catalyst was oxidized by H
acetic acid/H O at temperatures lower that 350 °C. In the presence of
in the feed, the Ni particle surfaces became reduced and active at
lower temperatures than in the absence of H . In the absence of H , the
2
O in the presence of the reactants
2
reformed. The TPD results suggested the formation of CO
lower temperatures than CO. As discussed above, the DFT calculation
results [18] showed that the formation of CO was not favorable on a
Co stepped surface. Nevertheless, the experimental TPSR results for the
20Ni catalyst, in the absence or presence of H in the reactor feed
(Figs. 6c and 7e, respectively), clearly showed the formation of CO at
low temperature, where very low selectivity to CO was observed. In this
case, it is clear that the presence of CO could not be explained by the
presence of the shift reaction. The higher selectivity to CO , relative to
2
at slightly
H
2
2
2
2
increase of particle size from 4 nm (8Ni catalyst) to 8 nm (20Ni
catalyst) led to a change in the free energy of the surface, and it is
expected that reduction of the surface occurs at lower temperatures for
the 20Ni sample, due to the presence of large Ni particles (20Ni). In the
2
2
presence of H
high CH /(CO + CO
by cleavage of the CeC bond of the acetic acid molecule were
hydrogenated to CH at low temperatures.
2
, the Ni catalysts were highly selective to CH
4
, with the
2
4
2
) molar ratio indicating that the fragments formed
2
CO, was less evident at decreased metal loading and increased Co
content. These results suggested that metal-support interface effects
4
8

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 2
References
Carbon accumulation rates of the catalysts after 6 h under SRAcH conditions.
[
1] R.R. Davda, J.W. Shabaker, G.W. Huber, R.D. Cortright, J.A. Dumesic, review of
catalytic issues and process conditions for renewable hydrogen and alkanes by
aqueous-phase reforming of oxygenated hydrocarbons over supported metal
catalysts, Appl. Catal. B: Environ. 56 (2005) 171–186.
−1
h−1
)
Sample
Carbon accumulation rate (mgcarbon
g
cat
8
2
4
5
8
Ni
22.7
8.9
8.9
8.2
5.4
.5Co5.5Ni
Co4Ni
.5Co2.5Ni
Co
[
2] J. Remón, F. Broust, G. Volle, L. García, J. Arauzo, Hydrogen production from pine
and poplar bio-oils by catalytic steam reforming. Influence of the bio-oil composi-
tion on the process, Int. J. Hydrogen Energy 40 (2015) 5593–5608, http://dx.doi.
org/10.1016/j.ijhydene.2015.02.117.
[
[
3] D. Li, X. Li, J. Gong, Catalytic reforming of oxygenates: state of the art and future
prospects, Chem. Rev. 116 (2016) 11529–11653, http://dx.doi.org/10.1021/acs.
chemrev.6b00099.
4] J.H. Sinfelt, D.J.C. Yates, Catalytic hydrogenolysis of ethane over the noble metals
of group VIII, J. Catal. 8 (1967) 82–90, http://dx.doi.org/10.1016/0021-9517(67)
90284-9.
5] G.A. Somorjai, Y. Li, Introduction to Surface Chemistry and Catalysis, second ed.,
Wiley-VCH Verlag GmbH & Co KGaA, 2010.
6] X. Hu, G. Lu, Comparative study of alumina-supported transition metal catalysts for
hydrogen generation by steam reforming of acetic acid, Appl. Catal. B: Environ. 99
altered the reaction pathway, with formation of CO
tures. The TPSR results for the 20Ni catalyst showed that in the
presence of H , CH was formed with high selectivity, indicating the
presence of CH species on the metal surface. These results suggest that
the metal-support interface and the presence of Ni favored decomposi-
tion of the activated acetic acid molecules to CO and CH species,
which were hydrogenated to CH in the presence of H . However, CO
2
at low tempera-
2
4
[
x
[
2
x
(2010) 289–297, http://dx.doi.org/10.1016/j.apcatb.2010.06.035.
4
2
[
[
7] K.K. Pant, P. Mohanty, S. Agarwal, A.K. Dalai, Steam reforming of acetic acid for
hydrogen production over bifunctional Ni-Co catalysts, Catal. Today 207 (2013)
36–43, http://dx.doi.org/10.1016/j.cattod.2012.06.021.
8] P.G.M. Assaf, F.G.E. Nogueira, E.M. Assaf, Ni and Co catalysts supported on alumina
applied to steam reforming of acetic acid: representative compound for the aqueous
phase of bio-oil derived from biomass, Catal. Today 213 (2013) 2–8, http://dx.doi.
org/10.1016/j.cattod.2013.02.012.
9] S. Andonova, C.N. de Ávila, K. Arishtirova, J.M.C. Bueno, S. Damyanova, Structure
and redox properties of Co promoted Ni/Al2O3 catalysts for oxidative steam
reforming of ethanol, Appl. Catal. B: Environ. 105 (2011) 346–360, http://dx.doi.
org/10.1016/j.apcatb.2011.04.029.
could be formed by the decomposition of acetic acid on the metal
surface, as described by Li et al. [18].
Table 2 shows the carbon accumulation rates for the catalysts,
evidencing a decrease of carbon accumulation with increasing Co
content. Previous results have demonstrated that a fraction of Co is
oxidized in reforming reactions, with formation of a surface consisting
of reduced and oxidized Co (Co-CoO) [10]. The TPD results discussed
above, together with the DFT calculations [18], suggest the decom-
[
[
10] C.N. Ávila-neto, J.W.C. Liberatori, A.M. Silva, D. Zanchet, C.E. Hori, F.B. Noronha,
J.M.C. Bueno, Understanding the stability of Co-supported catalysts during ethanol
reforming as addressed by in situ temperature and spatial resolved XAFS analysis, J.
Catal. 287 (2012) 124–137, http://dx.doi.org/10.1016/j.jcat.2011.12.013.
position of acetic acid on the metal surface, with formation of CH
species that are hydrogenated to CH at low temperatures or are
decomposed to C* and H at high temperatures. The presence of Co-
x
4
2
CoO sites could promote the oxidation of C* species and decrease
carbon accumulation.
[11] C.N. Ávila-Neto, D. Zanchet, C.E. Hori, R.U. Ribeiro, J.M.C. Bueno, Interplay
between particle size, composition, and structure of MgAl 2 O 4 −supported Co-Cu
catalysts and their influence on carbon accumulation during steam reforming of
ethanol, J. Catal. 307 (2013) 222–237, http://dx.doi.org/10.1016/j.jcat.2013.07.
0
25.
4. Conclusions
[12] A.H. Braga, E.R. Sodré, J.B.O. Santos, C.M. de Paula Marques, J.M.C. Bueno, Steam
reforming of acetone over Ni- and Co-based catalysts: effect of the composition of
reactants and catalysts on reaction pathways, Appl. Catal. B: Environ. 195 (2016)
The pathway of the reaction studied depended on catalyst Ni-Co
1
6–28, http://dx.doi.org/10.1016/j.apcatb.2016.04.047.
composition, reactants composition, and temperature. In the presence
of acetic acid and H O, the catalyst surface was reduced at temperatures
higher than about 350 °C, with formation of CO, CO , and H . In the
presence of H , the Ni-containing catalyst surface was reduced at
temperatures of around 200 °C, while the temperature for reduction
of the catalyst surface increased with the presence of Co. In the
[13] D. Zanchet, J.B.O. Santos, S. Damyanova, J.M.R. Gallo, J.M.C. Bueno, Toward
understanding metal-catalyzed ethanol reforming, ACS Catal. 5 (2015) 3841–3863,
http://dx.doi.org/10.1021/cs5020755.
2
2
2
[
14] R. Pestmam, R.M. Koster, J.A.Z. Pieterse, V. Ponec, Reactions of carboxylic acids on
oxides 2. Bimolecular reaction of aliphatic acids to ketones, J. Catal. 168 (1997)
265–272, http://dx.doi.org/10.1006/jcat.1997.1624.
2
[
15] K. Takanabe, K. ichi Aika, K. Inazu, T. Baba, K. Seshan, L. Lefferts, Steam reforming
of acetic acid as a biomass derived oxygenate: bifunctional pathway for hydrogen
formation over Pt/ZrO2 catalysts, J. Catal. 243 (2006) 263–269, http://dx.doi.org/
10.1016/j.jcat.2006.07.020.
presence of H
the CeC bond, and the CH
hydrogenated to CH . At low temperature, the selectivity to CH
suppressed by the presence of Co in the Ni-based catalysts.
Over MgAl , the ketonization reaction occurred at temperatures
2
, the Ni-containing catalysts were active for scission of
, CO, and CO species formed were
was
x
2
[
16] K. Takanabe, K.I. Aika, K. Seshan, L. Lefferts, Sustainable hydrogen from bio-oil −
steam reforming of acetic acid as a model oxygenate, J. Catal. 227 (2004) 101–108,
http://dx.doi.org/10.1016/j.jcat.2004.07.002.
4
4
2
O
4
[17] A.A. Lemonidou, E.C. Vagia, J.A. Lercher, Acetic acid reforming over Rh supported
on La /CeO –ZrO : catalytic performance and reaction pathway analysis, ACS
2
O
3
2
2
of around 400 °C. The presence of metal on the catalysts suppressed the
ketonization reactions, due to changes of the reaction pathway. The
species adsorbed on the support, such as acetyl and acetate, were
decomposed in the metal-support interface. On the reduced catalyst
surface, the decomposition of acetic acid resulted in the formation of
Catal. (2013) 1919–1928, http://dx.doi.org/10.1021/cs4003063.
[
[
[
[
[
18] X. Li, S. Wang, Y. Zhu, G. Yang, P. Zheng, DFT study of bio-oil decomposition
mechanism on a Co stepped surface: acetic acid as a model compound, Int. J.
Hydrogen Energy 40 (2015) 330–339, http://dx.doi.org/10.1016/j.ijhydene.2014.
1
1.004.
19] S. Wang, X. Li, L. Guo, Z. Luo, Experimental research on acetic acid steam reforming
over Co-Fe catalysts and subsequent density functional theory studies, Int. J.
Hydrogen Energy 37 (2012) 11122–11131, http://dx.doi.org/10.1016/j.ijhydene.
CO and CO
2
according to different reaction pathways. The formation of
CO involved metal-support interfacial sites.
2
2
012.05.011.
20] P. Catalysts, R. Alcala, J.W. Shabaker, G.W. Huber, M.A. Sanchez-castillo,
J.A. Dumesic, Experimental and DFT studies of the conversion of ethanol and acetic
acid on, J. Phys. Chem. B 109 (2005) 2074–2085, http://dx.doi.org/10.1021/
jp049354t.
21] S. Wang, X. Li, F. Zhang, Q. Cai, Y. Wang, Z. Luo, Bio-oil catalytic reforming without
steam addition: application to hydrogen production and studies on its mechanism,
Int. J. Hydrogen Energy 38 (2013) 16038–16047, http://dx.doi.org/10.1016/j.
ijhydene.2013.10.032.
22] T.M.C. Hoang, B. Geerdink, J.M. Sturm, L. Lefferts, K. Seshan, Steam reforming of
acetic acid − a major component in the volatiles formed during gasification of
humin, Appl. Catal. B: Environ. 163 (2015) 74–82, http://dx.doi.org/10.1016/j.
apcatb.2014.07.046.
Conflicts of interest
The authors declare that there are no conflicts of interest regarding
the publication of this article.
Acknowledgments
This work was supported by CNPq (grant #407030/2013-1) and the
São Paulo State Research Foundation (FAPESP, grant #2013/10858-2).
We also thank LNNano for the provision of TEM facilities (proposal
[
[
23] G.P. Szijjártó, Z. Pászti, I. Sajó, A. Erdohelyi, G. Radnóczi, A. Tompos, Nature of the
active sites in Ni/MgAl2O4-based catalysts designed for steam reforming of ethanol,
J. Catal. 305 (2013) 290–306, http://dx.doi.org/10.1016/j.jcat.2013.05.036.
24] F.M. Althenayan, S. Yei Foo, E.M. Kennedy, B.Z. Dlugogorski, A.A. Adesina,
#
19921) and LNLS for use of the XPD beamline (proposal #19027).
9

S.C.M. Mizuno et al.
Catalysis Today xxx (xxxx) xxx–xxx
Bimetallic Co–Ni/Al2O3 catalyst for propane dry reforming: estimation of reaction
metrics from longevity runs, Chem. Eng. Sci. 65 (2010) 66–73, http://dx.doi.org/
0.1016/j.ces.2009.03.037.
25] G. Parravano, The reduction of nickel oxide by hydrogen, J. Am. Chem. Soc. 74
1952) 1194–1198, http://dx.doi.org/10.1021/ja01125a016.
26] A. Parmaliana, F. Arena, F. Frusteri, N. Giordano, Temperature-programmed
reduction study of NiO–MgO interactions in magnesia-supported Ni catalysts and
NiO–MgO physical mixture, J. Chem. Soc. Faraday Trans. 86 (1990) 2663–2669,
http://dx.doi.org/10.1039/FT9908602663.
[35] E.L. Rodrigues, A.J. Marchi, C.R. Apesteguia, J.M.C. Bueno, Promoting effect of zinc
on the vapor-phase hydrogenation of crotonaldehyde over copper-based catalysts,
Appl. Catal. A: Gen. 294 (2005) 197–207, http://dx.doi.org/10.1016/j.apcata.
2005.07.029.
[36] S. Kuboon, Y.H. Hu, Study of niO-CoO and Co3O4 −Ni3O4 solid solutions in
multiphase Ni-Co-O systems, Ind. Eng. Chem. Res. 50 (2011) 2015–2020, http://dx.
doi.org/10.1021/ie101249r.
[37] S.R. Tong, L.Y. Wu, M.F. Ge, W.G. Wang, Z.F. Pu, Heterogeneous chemistry of
monocarboxylic acids on??-Al 2O3 at different relative humidities, Atmos. Chem.
Phys. 10 (2010) 7561–7574, http://dx.doi.org/10.5194/acp-10-7561-2010.
[38] C. Binet, M. Daturi, J.-C. Lavalley, {IR} study of polycrystalline ceria properties in
oxidised and reduced states, Catal. Today 50 (1999) 207–225, http://dx.doi.org/10.
1016/S0920-5861(98)00504-5.
1
[
[
(
[
27] Z. Mosayebi, M. Rezaei, A.B. Ravandi, N. Hadian, Autothermal reforming of
methane over nickel catalysts supported on nanocrystalline MgAl2O4 with high
surface area, Int. J. Hydrogen Energy 37 (2012) 1236–1242, http://dx.doi.org/10.
1
016/j.ijhydene.2011.09.141.
[
28] N. Hadian, M. Rezaei, Z. Mosayebi, F. Meshkani, CO2 reforming of methane over
[39] R. Knapp, S.A. Wyrzgol, A. Jentys, J.A. Lercher, Water-gas shift catalysts based on
ionic liquid mediated supported Cu nanoparticles, J. Catal. 276 (2010) 280–291,
http://dx.doi.org/10.1016/j.jcat.2010.09.019.
nickel catalysts supported on nanocrystallineMgAl2O4, J. Nat. Gas Chem. 21 (2012)
2
00–206.
[
29] H. Özdemir, M.A.F. Öksüzömer, M.A. Gürkaynak, Effect of the calcination
temperature on Ni/MgAl2O4 catalyst structure and catalytic properties for partial
oxidation of methane, Fuel 116 (2014) 63–70, http://dx.doi.org/10.1016/j.fuel.
[40] C. Xu, B.E. Koel, Adsorption and reaction of CH3COOH and CD3COOD on the MgO
(100) surface: a Fourier transform infrared and temperature programmed deso-
rption study, J. Chem. Phys. 42 (1995) 1115, http://dx.doi.org/10.1063/1.469227.
2
013.07.095.
[41] J.I. Di Cosimo, V.K. Dıez, M. Xu, E. Iglesia, C.R. Apesteguıa, Structure and surface
́ ́
[
[
30] J. Guo, H. Lou, H. Zhao, D. Chai, X. Zheng, Dry reforming of methane over nickel
and catalytic properties of Mg-Al basic oxides, J. Catal. 178 (1998) 499–510,
http://dx.doi.org/10.1006/jcat.1998.2161.
catalysts supported on magnesium aluminate spinels, Appl. Catal. A: Gen. 273
(
2004) 75–82, http://dx.doi.org/10.1016/j.apcata.2004.06.014.
[42] R.M. Watwe, H.S. Bengaard, J.R. Rostrup-Nielsen, J.A. Dumesic, J.K. Nørskov,
Theoretical studies of stability and reactivity of CHx species on Ni(111), J. Catal.
189 (2000) 16–30, http://dx.doi.org/10.1006/jcat.1999.2699.
[43] P. Siegbahn, I. Panas, A theoretical study of CH, chemisorption and Ni(111) surfaces
on the Ni(100), Surf. Sci. 240 (1990) 37–49.
31] G. Jacobs, J.A. Chaney, P.M. Patterson, T.K. Das, B.H. Davis, Fischer-Tropsch
synthesis: study of the promotion of Re on the reduction property of Co/Al2O3
catalysts by in situ EXAFS/XANES of Co K and Re LIII edges and XPS, Appl. Catal. A:
Gen. 264 (2004) 203–212, http://dx.doi.org/10.1016/j.apcata.2003.12.049.
32] A. Sirijaruphan, A. Horváth, J.G. Goodwin, R. Oukaci, Cobalt aluminate formation
in alumina-supported cobalt catalysts: effects of cobalt reduction state and water
vapor, Catal. Lett. 91 (2003) 89–94, http://dx.doi.org/10.1023/B:CATL.
[
[44] H. Bengaard, Steam reforming and graphite formation on Ni catalysts, J. Catal. 209
(2002) 365–384, http://dx.doi.org/10.1006/jcat.2002.3579.
[45] E.C. Lovell, A. Fuller, J. Scott, R. Amal, Enhancing Ni-SiO2 catalysts for the carbon
dioxide reforming of methane: reduction-oxidation-reduction pre-treatment, Appl.
Catal. B: Environ. 199 (2016) 155–165, http://dx.doi.org/10.1016/j.apcatb.2016.
05.080.
0
000006322.80235.8f.
[33] J. Zhang, H. Wang, A.K. Dalai, Effects of metal content on activity and stability of
Ni-Co bimetallic catalysts for CO2 reforming of CH4, Appl. Catal. A: Gen. 339
(
2008) 121–129, http://dx.doi.org/10.1016/j.apcata.2008.01.027.
[46] T. Mondal, N. Kaul, R. Mittal, K.K. Pant, Catalytic steam reforming of model
oxygenates of bio-oil for hydrogen production over La modified Ni/CeO2–ZrO2
catalyst, Top. Catal. 59 (2016) 1343–1353, http://dx.doi.org/10.1007/s11244-
016-0662-3.
[
34] K. Nagaoka, K. Takanabe, K.I. Aika, Modification of Co/TiO2 for dry reforming of
methane at 2 MPa by Pt, Ru or Ni, Appl. Catal. A: Gen. 268 (2004) 151–158, http://
dx.doi.org/10.1016/j.apcata.2004.03.029.
1
0