2
N. Escalona et al. / Applied Catalysis A: General 481 (2014) 1–10
hydrocarbons [1,2]. The fast pyrolysis of lignocellulosic biomass in
order to produce bio-oil is another attractive route for the process
[15] reported no dependence of the reaction temperature on the
product distribution for guaiacol conversion with Pt/Al O catalyst.
2
3
[
3,4]. Bio-oil is a complex mixture of water and about 300 chem-
However, the authors demonstrated that an increase in the hydro-
gen pressure (from atmospheric pressure to 127 MPa) favored the
formation of benzene, toluene, anisole, phenol and o-cresol com-
pounds. Other non-classical catalysts, such as Ni P/SiO , Co P/SiO
icals. Among them, the following prominent compounds can be
mentioned: guaiacol, catechol, syringol, vanillin, furancarboxalde-
hydes, isoeugenol, pyrones, acetic acid, formic acid, carboxylic acids
derivates, hydroxyaldehydes, hydroxyketones, sugars, carboxylic
acids and phenolic compounds [3,4]. This makes bio-oil a poten-
tial feedstock for chemicals within the biorefinery concept. Lignin
derivatives in bio-oils can be as much as 30 wt% [3] and offer partic-
ular promise as a major source of aromatic compounds due to their
unique structure. However, they have received limited attention
because they make bio-oil thermally unstable, corrosive and also
they have the proclivity to polymerize under air exposure [4,5]. Dif-
ferent methods have been studied to extract chemicals from bio-oil,
with the use of polar and non-polar solvents [3,5]. Hydrotreating
process, intended to reduce the oxygen content of lignin deriva-
tives to indirectly address many of the limitations of bio-oil, has
also been widely investigated [6–8]. Guaiacol (2-methoxyphenol)
has been extensively used as a lignin model compound because
guaiacyl-type compounds with their two functional groups (–OH
and –OCH ) are significant constituents of bio-oil [3,5]. These
studies have been conducted using different catalysts and reactor
configurations [9–21].
Previously, Laurent and Delmon [9] carried out catalytic
hydrodeoxygenation (HDO) reaction of guaiacol on conventional
hydrodesulfurization catalysts (CoMoS/␥-Al O and NiMoS/␥-
Al O ), primarily producing catechol. Bui et al. [10] stud-
ied the effect of support on the catalytic conversion of
guaiacol with MoS2 supported on ␥-Al O , TiO2 and ZrO2
catalysts. They found that the support modifies the activ-
ity and product distribution. The observed trends in prod-
uct distribution were as follows: for MoS /␥-Al O catalyst,
2
2
2
2
and Fe P/SiO2 have also been studied for guaiacol conversion:
2
Ni P/SiO2 and Co P/SiO2 catalysts displayed high activity, prin-
2
2
cipally producing benzene, whereas Fe P/SiO2 catalyst was less
2
active and favored the production of phenol [16]. Bykova et al. [24]
reported it to be active in the guaiacol conversion into cyclohexane,
1-methylcyclohexane-1,2-diol and cyclohexanone compounds as
the main product over Ni-based catalysts. In general, the metal cata-
lysts have shown a high activity and selectivity to different products
in the conversion of guaiacol, though the stability of the active
phase under reaction condition is low and has had little attention.
An attractive option to stabilize metal particles on mixed oxides is
by using perovskite-type oxides as precursors. In fact, Escalona et al.
[25] stabilized Co nanoparticles by reducing LaFe1 CoxO3 perov-
−x
skites to high temperature, displaying a high activity in the Fischer
Tropsch reaction. In general, these oxides can be represented by
the general formula ABO3 structures, where A is an alkaline earth
or lanthanide and B is a redox d-transition series metal, generally
responsible for the catalytic activity. There is a considerable num-
ber of potentially interesting perovskites in oxidation reactions,
owing to the number of A and B cations that can be inserted into
this structure. When perovskite-type oxides are reduced, the low
thermal behavior of these materials leads to highly dispersed metal
nanoparticles [26]. Additionally, the oxygen storage capacity of the
substituted perovskites may also catalyze the oxidation of carbona-
ceous materials deposited on the active sites of the catalyst, leading
to enhanced stability. Therefore, the objective of the current work
was to study the guaiacol conversion in high added value chemi-
cals over highly dispersed Ni particles supported on a mixed oxide.
The catalysts were prepared by reduction of high temperature of
3
2
3
2
3
2
3
2
2
3
catechol > methylcatechol > methylphenol > heavy products; for
MoS /TiO and MoS /ZrO catalysts, phenol > catechol » light prod-
2
2
2
2
ucts (mainly methanol) > HDO compounds. The large extent of
methylation reaction displayed by MoS /␥-Al O catalyst was
La1 Ce NiO3 (x = 0.0, 0.1, 0.3, 0.5, 0.7, 0.9) perovskites precursor
−x 1−x
3+
with different substitution of lanthanum cation La by an oxygen
storage cation such as cerium (which arises from the redox behav-
ior of Ce3+/Ce ). The Ni nanoparticles displayed a high activity and
selectivity, however segregation of La O and CeO oxides depend-
2
2
3
attributed to the acid strength of the ␥-Al O3 support which pro-
2
4+
motes the formation of heavy products, acts as a precursor to coke
and hence decreases the overall catalytic activity. Along these lines,
Zhu et al. [11] studying the influence of acid sites of Pt/zeolite cat-
alysts for the transformation of methoxybenzene found that acid
sites favor the formation of methyl phenolic compounds (methyl-
catechols, methylphenols, etc.). Despite the high activity of the
sulfided catalysts and the possibility to get different products, these
catalysts are quickly deactivated by coke deposition reaction con-
ditions [22,23].
2
3
2
ing on grade of substitution were observed. The Ni-reduced catalyst
that displayed the most activity in batch reactor was compared
with sulfides NiMoS/SiO2 catalyst and its stability was studied in a
continuous flow reactor.
2. Experimental
On the other hand, several studies have demonstrated the via-
bility of non-sulfided materials as a new class of catalysts that
are active and selective for guaiacol conversion. Gutierrez et al.
2.1. Preparation
[
13] reported high conversion and selectivity in the hydrogenation
Substituted La1 CexNiO3 (x = 0.0, 0.1, 0.3, 0.5, 0.7, 0.9) perov-
−x
of guaiacol at lower reaction temperatures on zirconia supported
monometallic and bimetallic–noble metallic (Rh, Pd, Pt) catalysts.
The monometallic catalyst produces cyclohexanol as its princi-
pal product, whereas 1-methyl-1,2-cyclohenanediol was obtained
from bimetallic catalysts. However, it is well known that the prod-
uct distribution is strongly dependent on the respective support.
Lee et al. [14] reported that noble metal (Rh, Pd, Pt, Ru) catalysts
skites were prepared by the self-combustion method [27]. Glycine
(H NCH CO H), used as ignition promoter, was added to an aque-
2
2
2
ous solution of lanthanum, ceria and nickel nitrates with the
−
appropriate stoichiometry in order to get a NO3 /NH2 (molar
ratio) = 1. The resulting solution was slowly evaporated until a vitre-
ous green gel was obtained. The gel was heated up to around
◦
250 C, temperature at which the ignition reaction occurs produc-
supported on carbon black, Al O3 and SiO –Al O were able to
ing a powdered precursor which still contains carbon residues.
The solids were crushed and sieved to obtain the required parti-
cle size (<200 m) prior to calcination at 700 C in air for 10 h in
order to eliminate the remaining carbon and leads to the formation
of the perovskite structure. Even though in some of the calcined
solids the perovskite structure was not obtained, the nomenclature
2
2
2
3
modify the product distribution. Noble metals supported on carbon
black favored the formation of 2-methoxycyclohexanol, whereas
those supported on SiO –Al O favored cyclohexane as the main
◦
2
2
3
product. On the other hand, noble metals supported on more acidic
materials were reported to favor deoxygenation reactions. Reac-
tion conditions have also been studied in order to determine their
potentialinfluenceon guaiacolconversion. Nimmanwudiponget al.
La1 CexNiO3 was used throughout this study for the sake of uni-
−x
formity. Since hydrogen was used to activate the perovskites prior