Effect of support modifications for CoMo/γ-Al2O 3 and CoMo/ASA catalysts in the hydrodeoxygenation of guaiacol

Article abstract of DOI:10.1016/j.apcata.2013.11.011

CoMo catalysts supported on a commercial alumina (γ-Al ₂O₃) modified with Na (1, 2 and 3 wt.%) and supported on amorphous aluminosilicates (ASA) with different Si/(Si + Al) ratios (0.15, 0.25 and 0.75) were prepared. Additionally, a CoMo catalyst supported on ASA modified with 3 wt.% Na was synthetized. These catalysts were evaluated in the hydrodeoxygenation (HDO) of guaiacol (2-methoxyphenol) as bio-oil model compound. The results show a decrease of catalytic activity with increasing amounts of Na added compared to an unmodified catalyst. This modification in the acid-base properties of the catalyst generates changes in selectivity such as the inhibition of methylated compounds formation, e.g., cresol and the increase of phenol production. For ASA-supported catalysts, a decrease in catalytic activity with the increase in the Si content took place, whereas the HDO degree was not significantly affected. Moreover, the ASA-supported catalysts led to a lower activity of the HDO of guaiacol compared to γ-Al₂O ₃-supported catalysts.

Full text of DOI:10.1016/j.apcata.2013.11.011

Applied Catalysis A: General 474 (2014) 59–68
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Effect of support modifications for CoMo/␥-Al2O3 and CoMo/ASA
catalysts in the hydrodeoxygenation of guaiacol
Iván D. Mora, Edna Méndez, Liseth J. Duarte, Sonia A. Giraldo∗
Centro de Investigaciones en Catálisis (CICAT), Escuela de Ing. Química, Universidad Industrial de Santander (UIS), Cra. 27 Calle 9, Bucaramanga, Colombia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 April 2013
Received in revised form 4 November 2013
Accepted 6 November 2013
Available online 15 November 2013
CoMo catalysts supported on a commercial alumina (␥-Al2O3) modified with Na (1, 2 and 3 wt.%) and
supported on amorphous aluminosilicates (ASA) with different Si/(Si + Al) ratios (0.15, 0.25 and 0.75)
were prepared. Additionally, a CoMo catalyst supported on ASA modified with 3 wt.% Na was synthetized.
These catalysts were evaluated in the hydrodeoxygenation (HDO) of guaiacol (2-methoxyphenol) as bio-
oil model compound. The results show a decrease of catalytic activity with increasing amounts of Na
added compared to an unmodified catalyst. This modification in the acid–base properties of the catalyst
generates changes in selectivity such as the inhibition of methylated compounds formation, e.g., cresol
and the increase of phenol production. For ASA-supported catalysts, a decrease in catalytic activity with
the increase in the Si content took place, whereas the HDO degree was not significantly affected. Moreover,
the ASA-supported catalysts led to a lower activity of the HDO of guaiacol compared to ␥-Al2O3-supported
catalysts.
Keywords:
Hydrodeoxygenation
Guaiacol
CoMo catalysts
Sodium
Alumina
Amorphous aluminosilicates
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
processed during the stabilization step due to the presence of
two O-containing functional groups, phenolic and methoxy. They
In response to the growing interest in finding alternative energy
sources to oil-derived products, extensive research has focused on
the study and use of biomass. One of the technologies used to
harness biomass energy is called fast pyrolysis. In this process a
liquid product called bio-oil is obtained. Their low concentrations
of sulfur, nitrogen, high availability and similarity to conventional
fossil fuels make this bio-oil attractive for its use. However, the
high oxygen content in bio-oil (30–40 wt.%) produces undesirable
properties, such as high thermal instability, high viscosity and low
heating value [1,2]. One alternative to improve the characteristics
of bio-oil is the catalytic HDO process. Elliot and Baker set up this
process in two steps due to the variety in reactivity of molecules in
bio-oil [3,4]. The first one is a stabilization step. This process will be
done at low hydrotreating temperatures to convert more reactive
groups, such as methoxyphenols, biphenols and others, into less
reactive molecules. This process is done to improve significantly
the quality of bio-oil for transportation and storage. The second
process will obtain a complete HDO as a result of a more severe
hydrotreating step [3,4].
tend to form high molecular weight compounds, and thus, grant
a low stability to the bio-oil [5]. The traditional catalysts in the
HDS processes such as CoMo/␥-Al2O3 have shown they are also
active in HDO reactions. However, the acidic properties of the ␥-
Al2O3 induce to the formation of heavy products, and therefore,
contribute to the coke formation [6,7].
Some authors have compared CoMo catalysts supported on ␥-
Al2O3 and active carbon and have found that the latter one limits
the formation of condensation products and allows a higher phe-
nol/catechol ratio; however, less catalytic activity compared with
catalysts supported on ␥-Al2O3 is evidenced [6,8]. Silica supported
catalysts have also been studied in the HDO of guaiacol using active
phase containing transition metals. Sepúlveda et al. [9] have stud-
ied the ReS2 catalysts supported on ␥-Al2O3 and SiO2. They have
reported that the use of ReS2/␥-Al2O3 catalyst led to the formation
of catechol, heavier compounds, and methylphenols principally.
Moreover, the ReS2/SiO2 catalyst was more selective toward phe-
nol and HDO compounds. These reports are evidence that the use
of acidic supports leads to higher conversion but lower selectiv-
ity to HDO compounds compared to neutral supports. Therefore,
a balance between activity and selectivity would be obtained by
modifying the acid properties of the supports. A good attempt to
reach this balance is by the addition of alkali metals to the sup-
ports. Pérez et al. [10] modified CoMo/␥-Al2O3 catalysts by adding
Na to the support and found out that increasing the Na content the
Model compounds have been used for HDO studies. Guaia-
col (2-methoxyphenol) is one of the molecules of interest to be
∗
Corresponding author. Tel.: +57 7 6344746; fax: +57 7 6344684.
E-mail addresses: sgiraldo@uis.edu.co, giraldosonia@yahoo.fr (S.A. Giraldo).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.11.011

60
I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
population of intermediate and strong acid sites diminished and
the population of weak acid sites increased; this led to changes in
the selectivity in FCC naphtha hydrotreating reactions.
determined based on the theory of Brunauer, Emmett, and Teller
(BET). The volume and the pore size distribution were calculated
based on the method of Barrett, Joyner, and Halenda (BJH) through
the programs incorporated in the equipment.
Therefore, the role of Na in the reduction of acid proper-
ties of the ␥-Al2O3 is evident. Amorphous aluminosilicates (ASA)
are of interest as supports due to among other properties, their
acidic character, which can be controlled with the variation of the
Si/(Si + Al) ratio. For FCC naphtha reactions, it has been reported
that these changes affect the behavior of the reactions, allowing
a better control over the selectivity toward hydrogenation reac-
tions, isomerization and cracking of olefins, during the HDS of
2-methylthiophene [11,12]. This knowledge about the modifica-
tion of the supports and its effect in HDS reactions could be used as
a guide to obtain catalyst to be tested in HDO reactions.
This work aims to study the effect of the acid-base properties
of CoMo supported on either ␥-Al2O3 catalysts by varying the Na
amount impregnated to the support or ASA catalysts with different
Si/(Si + Al) ratios on the catalytic activity and selectivity for the HDO
of guaiacol in the stabilization step conditions.
2.2.2. Proton affinity distributions (PAD)
The acid–basic properties of catalysts in the oxide state were
determined by potentiometric titration using a titrator TitroLine
alpha (SCHOTT, 0.001 pH units). The catalyst (0.5 g; Dp < 0.74 mm)
was suspended in 50 ml of aqueous solution of NaNO3 0.1 M with
constant magnetic stirring. HNO3 0.1 N was used as acid titrant for
␥-Al2O3-supported catalysts and HCl 0.1 M for ASA-supported cat-
alysts; NaOH 0.1 N was used as basic titrant agent in both cases.
These solutions were added (0.05 ml) every 90 s until reaching a
final pH of 3 or 11. The pH in function of the accumulated volumes
of acid or base added allowed calculation of the consumption of
protons using proton balance and thus determine a function of pro-
ton consumption (F(Log K)) in function of pH, which is equivalent
to the local adsorption isotherm. The proton distribution function
was determined using the method proposed by Contescu et al.
[13,14]. The quantitative analysis of the OH groups presents in the
surface of the supports was carried out by mean of the decon-
volution of the curves using Gauss functions as was reported by
Contescu et al. [15]. The area under each peak represents the acid
sites (mmol H+ g−1 cat) presents in the supports. The software used
was Origin Pro 8.5.1.
2. Experimental
2.1. Catalyst preparation
The CoMo catalysts supported on gamma alumina (␥-Al2O3,
Procatalyse) and ASA, containing 10 wt.% MoO3 and 2.5 wt.% CoO,
were prepared by successive incipient wetness impregnation with
an aqueous solution of (NH4)6Mo7O24·4H2O (Merck, 99%) and
Co(NO3)2·6H2O (Merck, 99%) respectively. ASA supports with
different Si/(Si + Al) molar ratios (0.15, 0.25, 0.75) were pre-
pared by the sol–gel procedure described by La Parola et al.
[12] using tetraethyl orthosilicate (TEOS, Sigma, 98%) and alu-
minum tri-sec-butoxide (ASTB, Sigma, 97%) as precursors. For
ASA-supported catalysts after each impregnation step, the solids
were dried under air flow at 343 K for 2 h and calcined at 773 K
for 12 h. The ASA supports and catalysts were named ASAxx and
CoMo/ASAxx respectively, where xx represents the molar percent-
age of Si/(Si + Al). The ASA and ␥-Al2O3-supported catalysts used
in this study were prepared in the same way that the materials
tested in previous studies performed in our laboratory by Pérez
et al. [10,11].
For the ASA25 modified with 3 wt.% of Na and for the ␥-Al2O3
modified with 1, 2 and 3 wt.% nominal content of Na, an initial
impregnation step was performed with a solution of NaNO3 (Merck,
99%). Subsequently, the Co and Mo were impregnated following
the procedure described above. After each impregnation step, for
ASA25 the solids were dried under air flow at 343 K for 2 h and cal-
cined at 773 K for 12 h. In this way, the ASA25-Na(3) support and
CoMo/ASA25-Na(3) catalyst were obtained. For ␥-Al2O3-supported
catalysts the solids were dried at 393 K for 12 h and calcined at 773 K
for 4 h both under air flow. A CoMo catalyst supported on pure
␥-Al2O3 was also prepared in the same way as the reference cata-
lyst (named CoMo/A). The CoMo catalysts modified with Na were
named CoMo/A-Na(x), where x represents the Na nominal content.
The Na, Mo, and Co content of these catalysts was verified using
atomic absorption spectroscopy (results not shown).
2.2.3. UV–vis diffuse reflectance (UV–vis DRS)
To determine the chemical coordination of cobalt and molyb-
denum ions of the catalysts in oxide state, UV–vis DRS
spectra of CoMo/A, CoMo/A-Na(1), CoMo/A-Na(2), CoMo/A-Na(3),
CoMo/ASA15, CoMo/ASA25, CoMo/ASA75 catalysts were recorded
using a UV-2401PC Shimadzu spectrophotometer equipped with
an ISR 240A integrating sphere accessory. The spectra were
recorded in the range of 200–800 nm and using reference BaSO4.
Kubelka–Munk function was calculated from reflectance by (F(R)),
where R is the diffuse reflectance depending on the wavelength.
2.2.4. X-ray diffraction (XRD)
In order to determine the phases presents in the catalyst
CoMo/A, CoMo/A-Na(3) and CoMo/ASAxx, XRD measurements of
powdered samples were carried out in a Bruker Advance diffrac-
tometer with DaVinci geometry using Ni-filtered CuK␣1 radiation
(40 kV, 30 mA). The 2ꢀ range was scanned between 10◦ and 80◦
with a step size of 0.01526◦ 2ꢀ and a counting time of 0.4 s per step.
The qualitative analysis of the phases present in the samples
was performed by comparing the profile observed with diffraction
profiles reported in the database PDF-2 ICDD.
2.2.5. FTIR
Aiming to determine the effect of Na over the OH groups of ␥-
Al2O3, FTIR spectroscopy of ␥-Al2O3 and ␥-Al2O3–Na(3) was carried
out in a Bruker Tensor 27 FTIR model supplemented with Bruker
ATR Platinum. Before tests, the samples were dried ex situ for 12 h.
2.2.6. SEM-EDS
Scanning Electron Microscopy with X-ray microanalysis was
used to obtain morphological observations of the surface of
CoMo/ASA15, CoMo/ASA75 and CoMo/A-Na(3) samples in the oxi-
dized state; mapping of the elements was carried out to determine
the distribution on the surface of the catalyst. The technique was
performed in a scanning electron microscopy FEI Quanta 650 FEG
operating with an electron voltage of 4.00 kV and coupled with ana-
lyzer of EDS X-ray. The study was carried out with a magnification
of 10,000×.
2.2. Characterization of catalysts
2.2.1. Textural properties
The specific surface area (ABET), pore volume (PV) and average
pore diameter (PD), of catalysts were estimated by mean of the
adsorption-desorption isotherms using nitrogen UAP grade as an
adsorbate with a Quantachrome NOVA 1200 instrument. The solids
were previously outgassed in vacuum by 12 h at 70 ◦C. The ABET was

I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
61
Table 1
2.3. Catalytic evaluation
Textural properties of the different catalysts.
Catalytic tests were made in a batch stainless-steel reactor (Parr
instruments) under H2 atmosphere at 523 K, 5.5 MPa pressure and
stirring rate of 750 rpm. The solution of reactants that occupied a
volume of 250 ml was prepared using 3.7 wt.% of guaiacol (Sigma,
98%), xylenes containing ethylbenzene (Baker, 99.9%) as solvent,
and 2 wt.% dodecane (Sigma, 99%) as an internal standard for the
chromatographic analysis. The catalyst (1.7 g, 0.18–0.6 mm frac-
tion) was sulfided ex situ under H2S/H2 (15/85% v/v) flow at 673 K
for 4 h. The catalyst-to-guaiacol ratio used in the experiments was
0.21. The reactor was sealed immediately after presulfurized cata-
lyst was added to the reaction mixture. The air was evacuated by
pressurization–depressurization cycles with nitrogen. The mixture
was then heated to 523 K under constant stirring of 750 rpm. H2 was
subsequently added to reach a pressure of 5.5 MPa. After reaching
these conditions the first liquid sample was taken. It was consid-
ered to correspond to time zero of the reaction. Liquid samples
were taken every 10 min during the first hour of reaction, every
20 min during the second hour, and 30 min during the third and
fourth hours. The volume of liquid taken was always measured.
Liquid samples were analyzed by gas chromatography with a HP
6890 equipped with a HP-1 column (100 m × 0.25 mm × 0.5 ␮m)
and FID detector. The compounds in the liquid sample were iden-
tified by GC–MS analysis using a column HP-5 MS. The closure in
mass balances was verified for all cases.
Catalyst
ABET (m2 g−1
)
PV (cm3 g−1
)
PD (Å)
A
238
194
168
164
160
312
384
115
205
239
68
0.67
0.52
0.47
0.45
0.45
0.52
0.71
–
0.37
0.45
0.24
0.68
0.48
112
109
114
113
106
67
74
–
72
82
CoMo/A
CoMo/A-Na(1)
CoMo/A-Na(2)
CoMo/A-Na(3)
ASA15
ASA25
ASA75
CoMo/ASA15
CoMo/ASA25
CoMo/ASA75
ASA25-Na(3)
CoMo/ASA25-Na(3)
144
76
78
158
133
ABET, specific surface area BET; PV, pore volume; PD, average pore diameter; A, alu-
mina; ASA, amorphous silicates aluminum. (x) = Na nominal mass content. ASAxx,
xx = Si/(Si + Al) molar ratio.
observed that the ABET decreases when the Na content is increased.
The deterioration of the textural properties after active metal incor-
poration for CoMo/A and CoMo/ASA catalysts when comparing
them with their respective supports is also evidenced. From the
analysis of the adsorption–desorption curve hysteresis, it can be
noticed that all materials are mesoporous, but there is a presence
of a small proportion of micropores for catalysts supported on ASA
which is explained by Pérez et al. [11] as due to the basic hydrolysis
used in the sol–gel synthesis of these materials. In reference to the
textural properties of ASA support there is not a defined trend of
the ABET with the Si/(Si + Al) ratio.
2.4. Expression of results
The activity of the catalysts as characterized by the disappear-
ance of the reactant is reported as pseudo first-order reaction with
rate constants k in Eq. (1), as reported by other authors [6,16],
introducing a volume correction according to Eq. (2) proposed by
Gevert et al. [17].
3.1.2. PAD
Fig. 1 shows the PAD curves for ␥-Al2O3, ␥-Al2O3 modified with
Na and ␥-Al2O3-supported catalysts. The hydroxyl groups present
on each catalyst before they are sulfurized can be identified. The ␥-
Al2O3, ␥-Al2O3 doped with Na and CoMo/A catalysts present three
well-defined peaks in pH regions corresponding to the OH groups
type III, IA, and IB as classified by Knözinger et al. [18].
Table 2 shows the quantification of the types of OH groups in
(mmol H+ g−1 cat) for some of the materials studied. In these results,
the decrease in the OH groups type III and IIA compared with ␥-
Al2O3 can be observed when 2 and 3% of Na is added to the support.
This decrease is especially significant for the sites type III, showing
higher selectivity of the alkaline metal to be deposited in the more
acidic sites of the ␥-Al2O3. On the other hand, the other groups are
not significantly affected. Concerning to the type IA and IB sites for
␥-Al2O3 doped with 2 and 3% Na, a tendency due to Na addition is
not observed.
ꢀ
ꢁ
ꢀ ꢁ
Ci
t
−ln
= kWf
(1)
(2)
C0
V
n
ꢀ ꢁ
ꢂ
t
ti − ti−1
Vi
f
=
V
i=1
where Ci and C0 [g/ml] are the reactant concentrations of sample i
at time t and at time zero, respectively; k [min−1 g−1 ml] is the rate
constant; W [g] is the catalyst weight; t [min] is the time and V [ml]
is the solution volume.
Product yield was expressed according to Eq. (3). Where molj,
are the moles of product j and molgua are the initial moles of gua-
0
iacol.
molj
Yield =
(3)
molgua
For the ␥-Al2O3-supported catalysts modified with Na (Fig. 1b),
the elimination of the peak in pH 3 is evidenced. A new peak in pH 5,
which increases with increasing amounts of Na appears, and a dis-
placement of all peaks to the right is also observed; this is evidenced
0
The HDO degree is calculated at 30% of guaiacol conversion and
including deoxygenated and partially deoxygenated compounds
such as phenol and cresol as shown in Eq. (4), where mk is the mass
of phenol, cresol, toluene, benzene and cyclohexane, and mgua is
0
the initial mass of guaiacol.
Table 2
ꢃ
Quantification of types of OH groups of some catalysts.
mk
HDO degree =
(4)
Catalyst
Type sites (mmol H+ g−1 cat)
mgua
0
III
IIA
IA
IB
3. Results
A
2.596
2.022
1.878
1.535
1.012
1.250
0.996
0.920
0.155
0.218
0.602
0.593
0.791
0.472
0.785
0.727
0.613
1.016
0.321
0.465
A–Na(2)
A–Na(3)
CoMo/A
CoMo/ASA75
3.1. Characterization of catalysts
3.1.1. Textural properties
Table 1 shows the BET specific surface (ABET), pore volume, and
average pore diameter of different catalysts and supports. It is
ASA, amorphous silicates aluminum. (x) = Na nominal mass content. ASAxx,
xx = Si/(Si + Al) molar ratio.

62
I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
Fig. 1. PAD curves for not activated: (a) ␥-Al2O3 and ␥-Al2O3–Na(x) supports, (b)
CoMo/A-Na(x) catalysts. A = ␥-Al2O3; (x) = Na nominal mass content.
Fig. 2. PAD curves for not activated CoMo/ASAxx catalysts: (a) with differ-
ent Si/(Si + Al) molar ratios, (b) CoMo/ASA25 and CoMo/ASA25-Na(3) catalysts.
ASA = amorphous ailicates aluminum. (x) = Na nominal mass content. ASAxx,
xx = Si/(Si + Al) molar ratio.
speciallywith the peak correspondingto the IIAgroup. With the dis-
placement, the IIA group is transformed into a peak with weaker
acid character as reported by Contescu et al. [13]. Additionally, for
the catalysts modified with sodium, a decrease in the peaks size is
observed when compared with CoMo/A catalyst.
sol–gel method [19]. As shown in Fig. 2a, the CoMo/ASA75 catalyst
presents stronger acids sites than the other ASA-supported cata-
lysts as shown by the peaks at lower pH ranges. This catalyst has
the higher Si/(Si + Al) ratio. According to other authors an increase
of Si leads to an increase in the acidic strength of these materials
[11].
The effect of Na on the acidic properties of the catalyst sup-
ported on ASA25 is shown in Fig. 2b. The Na addition leads to a
decrease of the peak size corresponding to the pH 4.3 (type IIA
sites for ␥-Al2O3). This behavior is similar to the one presented for
CoMo catalysts supported on ␥-Al2O3.
Even though OH type groups for ASA support are not defined,
they can be analyzed as in the case of the sites defined for ␥-Al2O3
as reported by other authors [19]. PAD curves for the ASA15 support
and CoMo/ASA catalysts are shown in Fig. 2a, and PAD curves for
the CoMo/ASA25 catalyst modified and unmodified with Na are
shown in Fig. 2b. A peak in pH 3 for the CoMo/ASA75 catalyst is
observed; it can be related to the type III sites, which correspond
to the acid sites of the ␥-Al2O3 according to the ranges reported
in literature [14,20]. Moreover, a peak at pH 6 appears due to the
presence of silanol groups of the silica surface as reported in other
studies [11,13,19]. A third peak appears in pH 9.3 and can be due
to the presence of type IA sites of the ␥-Al2O3 or siloxane groups of
the silica [14,19,20].
Referring to the ASA15 support, CoMo/ASA15 and CoMo/ASA25
catalysts, the first peak at pH 4.3 and the final peak at pH 10 indi-
cate the presence of type IIA and IB sites of the ␥-Al2O3 respectively
[14,20]. The intermediate peak observed around pH 8 could repre-
sent siloxane groups present in these materials. However, this peak
was also attributed to the appearance of new OH centers in the
surface of the ␥-Al2O3 due to the material was synthesized by the
3.1.3. UV–vis DRS
Fig. 3 shows the UV-DRS spectrum in the range 200–800 nm
for the CoMo catalysts supported on ␥-Al2O3 and ASA respec-
tively. According to the results reported in literature [21–25], in
the region between 200 and 400 nm, it can be identified the bands
corresponding to Mo tetrahedral (240–270 nm) and Mo octahedral
(290–320 nm) species. In the visible region of the spectrum the
bands attributed to the Co octahedral species and Co tetrahedral
species can be identified.

I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
63
Fig. 3. UV-DRS spectra of some not activated catalysts. (a) and (b) CoMo/A-Na(x) catalysts (c) and (d) CoMo/ASAxx catalysts. A = ␥-Al2O3; ASA = amorphous silicates aluminum.
(x) = Na nominal mass content. ASAxx, xx = Si/(Si + Al) molar ratio.
For CoMo/A catalysts, a decrease in the intensity of the peaks
located in the region of Mo octahedral and tetrahedral can be
observed when the support is modified with Na. This behavior was
also observed by Lycourghiotis et al. [26]; they concluded that for
atomic relations of Na/Co < 2, the perturbations caused by the pres-
ence of the alkaline metal leads to the formation of MoO4 at the
expense of the two dimensional polymeric species.
3.1.4. FT-IR
Fig. 4 shows FTIR-ATR spectra for ␥-Al2O3 and ␥-Al2O3-Na(3).
In general the bands shown in IR spectra are attributed to hydroxyl
groups presents on ␥-Al2O3. In accordance with Digne et al. [28],
In the region of 400–800 nm occurs an increase on the inten-
sity of the triplet band centered around 600 nm assigned to the
supported cobalt at tetrahedral symmetry (CoAl2O4) [27]. Addi-
tionally, for the Na-modified catalysts, an increase in the peaks
around 450 nm and 750 nm can be observed, which is assigned to
Co3O4 [27].
The bands corresponding to the ASA-supported catalysts (Fig. 3)
show a tendency to the decrease in the peaks corresponding to the
octahedral and tetrahedral Mo with increasing amounts of Si. With
the catalysts CoMo/ASA15 and CoMo/ASA25, these triplet peaks
around 600 nm not show a significant change; a triplet band associ-
ated with the Co species in tetrahedral coordination (CoAl2O4), as
in the case for ␥-Al2O3-supported catalysts is detected. However
with the CoMo/ASA75 catalyst there is a drop in the intensity of
the peaks located in the region around 600 nm. A shoulder above
700 nm with a small peak around 730 nm can be detected, which
indicates the presence of Co3O4 when comparing with the other
catalysts [27].
Fig. 4. FTIR spectrum of both A-Na(3) and A supports. A = ␥-Al2O3; (x) = Na nominal
mass content.

64
I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
Table 3
Pseudo first-order rate constants for the guaiacol conversion of all catalysts.
Catalyst
Activity
Catalyst
Activity
(g−1 min−1 ml)
(g−1 min−1 ml)
A
0.5
0.23
0.3
0.65
0.6
0.32
CoMo/A-Na(3)
CoMo/ASA15
CoMo/ASA25
CoMo/ASA75
ASA25-Na(3)
0.25
0.55
0.34
0.18
0.08
0.26
ASA15
ASA25
CoMo/A
CoMo/A-Na(1)
CoMo/A-Na(2)
CoMo/ASA25-Na(3)
A, alumina; ASA, amorphous silicates aluminum. (x) = Na nominal mass content.
ASAxx, xx = Si/(Si + Al) molar ratio.
catalysts have the broad bands typical of the ␥-Al2O3 with poor
crystallinity. The formation of the compound Na2MoO4 was not
detected in the spectra of ␥-Al2O3-supported catalysts modified
with Na.
For ASA-supported catalysts, the crystallographic profile is typ-
ical of the amorphous nature of the silica–alumina. As in the case
of ␥-Al2O3-supported catalysts, there were not observed the peaks
corresponding to the Co or Mo phase, excepting the CoMo/ASA75
catalyst. A peak corresponding to Co or Mo phases appears [30,31].
Fig. 5. XRD of CoMo/A-Na(x) and CoMo/ASAxx catalysts. A = ␥-Al2O3;
ASA = amorphous silicates aluminum. (x) = Na nominal mass content. ASAxx,
xx = Si/(Si + Al) molar ratio.
the main OH bands observed in the high-frequency region of the
IR spectra correspond to OH species exhibiting distinct local envi-
ronments. The peaks near at 3785 cm−1 correspond to hydroxyl
bound to a single octahedrally coordinated aluminum atom (type
IB)[18]. The peak near at 3690 cm−1 was assignedto hydroxylgroup
bound to three octahedrally coordinated aluminums (type III) [18].
The appearance of the reverse band at 3746 cm−1 indicates both
tetrahedral Al–OH and octahedral Al–OH [29]. The band between
3600 and 2900 cm−1 was assigned to hydroxyl group from sample
humidity.
3.1.6. SEM
Fig. 6 shows the SEM micrograph for Mo species obtained
from the CoMo/ASA15, CoMo/ASA75 and CoMo/A-Na(3) catalyst
in oxide state. It can observed the differences of CoMo/ASA15 and
CoMo/A-Na(3) respect to CoMo/ASA75. This may indicate that the
Mo species distribution is uniform on the surface of both catalysts
CoMo/ASA15 and CoMo/A-Na(3). Meanwhile, the CoMo/ASA75 can
have some agglomerates as shown in Fig. 6b.
According to Fig. 4, the peaks at 3785 and 3690 cm−1 and the
reverse band at 3746 cm−1 are perturbed by the Na addition. This
shows that the addition of sodium can suppress and neutralize
the surface acidic hydroxyl groups on ␥-Al2O3. Dignea et al. [28]
suggested that the active site contained a special configuration in
which a tetrahedrally coordinated Al3+ ion was adjacent to IA type
hydroxyl group. Knozinger and Ratnasamy in their review, also pro-
posed an active site on alumina, which contained a tetrahedrally
coordinated AI3+ and a tetrahedrally coordinated aluminum atom
bound to a hydroxyl (type IA) connected to the former by means of
an oxygen atom in the lower plane [18].
3.2. Catalytic activity and selectivity
The activities of various catalysts are reported in Table 3. All
supports without any modification showed capacity for guaiacol
transformation. The ␥-Al2O3 has the highest constant rate of all
supports and similar to the CoMo/ASA15 catalyst; however, the
␥-Al2O3 only reached a conversion of 50% within 4 h of reaction,
whereas CoMo/A and CoMo/ASA15 reached a conversion approxi-
mately 98 and 82% respectively.
CoMo/A is the most active catalyst, demonstrating the influence
of metal sulfide phase in guaiacol transformation when compared
with the activity of ␥-Al2O3 support. For ASA-supported catalysts,
CoMo/ASAS15 showed the higher activity with a constant slightly
smaller than CoMo/A. The observations lead to infer that with the
increase in the Si/(Si + Al) ratio, a decrease in catalytic activity takes
place.
3.1.5. XRD
The XRD results for both the ␥-Al2O3-supported catalysts and
the ASA-supported catalysts are shown in Fig. 5. It was not observed
the formation of the peaks corresponding to the Co or Mo which is
associated with a good dispersion of these on the different supports
[11]. Additionally, is observed that the spectra of ␥-Al2O3 supported
Fig. 6. SEM microscopy of (a) CoMo/ASA15; (b) CoMo/ASA75 and (c) CoMo/A-Na(3). ASA = amorphous silicates aluminum. ASAxx, xx = Si/(Si + Al) molar ratio. A = ␥-Al2O3.
(x) = Na nominal mass content.

I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
65
Table 4
The deoxygenation degree at 30% of guaiacol conversion for catalysts evaluated.
Catalyst
HDO degree
CoMo/A
0.109
0.109
0.204
0.214
0.079
0.054
0.127
0.071
CoMo/A-Na(1)
CoMo/A-Na(2)
CoMo/A-Na(3)
CoMo/ASA15
CoMo/ASA25
CoMo/ASA25-Na(3)
CoMo/ASA75
A, alumina; ASA, amorphous silicates aluminum. (x) = Na nominal mass content.
ASAxx, xx = Si/(Si + Al) molar ratio.
Concerning to the support modification with Na, activity is nega-
tively affected by the presence of this alkaline metal in all catalysts.
For CoMo/A the modification with 1 wt.% Na generates no signif-
icant change in the activity; however, the addition of 2 to 3 wt.%
of Na leads to a drop up to 50% in activity. For the ASA catalyst
modified with 3 wt.% Na, the variation in the constant rate is not as
marked as the one with ␥-Al2O3-supported catalyst. Comparison
of the rate constants between the supports modified with Na and
unmodified clearly reveals the effect of Na on the catalytic activity.
The ␥-Al2O3 doped with 3 wt.% Na showed an initial rate constant
close to zero, and the activity for ASA25-Na(3) support is very low
compared to the same support without Na.
Regarding to the appearance of reaction products, it is observed
that ␥-Al2O3 and ASA used as catalysts led to the formation of
catechol as the primary product, whilst neither phenol nor deoxy-
genated products were detected. The 1,2-dimethoxybenzene
and 2-methoxy-4-methylphenol are observed in very small
quantities when the supports are used as catalysts. These last
two compounds are also produced for the CoMo/ASAxx cata-
lysts. With CoMo/ASA15, the benzene, cyclohexane, and toluene
were produced but only in low amounts. For ASA-supported
catalyst modified with 3 wt.% Na, approximately the same pro-
portion of catechol and phenol is produced, whereas, with the
unmodified catalyst the production of catechol is higher than
phenol.
Fig. 7. Product yield to 30% of guaiacol conversion for: (a) CoMo/ASAxx-Na(x) cat-
alysts and (b) CoMo/A-Na(x). A = alumina; ASA = amorphous silicates aluminum.
(x) = Na nominal mass content. ASAxx, xx = Si/(Si + Al) molar ratio.
With catalysts supported on ␥-Al2O3, the metal sulfide phase
leads to the formation of phenol and deoxygenated compounds
such as toluene, cresol, benzene, and cyclohexane; however,
toluene is not observed for Na modified catalysts. For CoMo/A-
Na(2) and CoMo/A-Na(3) catalysts the phenol quantity is always
maintained higher than catechol quantity. This effect did not occur
from the beginning of the reaction with CoMo/A and CoMo/A-Na(1).
The maximum deviation in the mass balance closure was of 6%,
which can indicate that coke was not formed under these reaction
conditions.
Fig. 7 presents the product yield of guaiacol HDO at 30% conver-
sion. In Fig. 7a, it can be observed that there is not a remarkable
change in selectivity as a function of Si/(Si + Al) ratio. For the
ASA-supported catalyst modified with 3 wt.% Na, the product 1,2-
dimethoxybenzene is not detected. Moreover, this catalyst does not
lead to the production of cyclohexane or deoxygenated compounds.
The addition of Na to ␥-Al2O3 and ASA-supported catalysts leads
to an increase in phenol/catechol ratio at 30% conversion. As the Na
content increases on ␥-Al2O3-supported catalysts, the production
of phenol increases, but this behavior is most evident with increas-
ing the Na content up to 2 wt.%. For catalysts containing 2 and 3 wt.%
of Na cresol is not produced, but the formation of cyclohexane and
benzene is evidenced.
tendency for deoxygenation degree is not clear with the Si vari-
ation. On the other hand, the CoMo/ASA25-Na(3) catalyst clearly
shows an increase on its HDO degree, compared with the same
catalyst unmodified.
4. Discussion
In the present work, the catalytic activities and selectivities of
␥-Al2O3 and ASA-supported CoMo catalysts have been investigated
under the same conditions of reaction.
In reference to the results shown in Table 1, the deteriora-
tion of the textural properties after active metal incorporation for
CoMo/ASA catalysts can be attributed to the blocking of the pores by
the active metal deposition [32]. A similar effect of deterioration of
the textural properties for ␥-Al2O3-supported catalysts after active
metal incorporation and sodium addition is also observed.
4.1. Reaction scheme
In accordance with the product distribution results outlined
before, the guaiacol transformation could occur according to two
principal routes as shown in Fig. 8. The first route is a demethylation
(DME) where the hydrogenolysis of the methyl–oxygen bond of the
methoxyl group takes place to form catechol and methane in the
first stage followed by the elimination of one of the hydroxyl groups
Table 4 shows the deoxygenation degree of the catalysts eval-
uated to 30% of total conversion. The increase of deoxygenation
degree for ␥-Al2O3-supported catalysts with 2 and 3% Na compared
with CoMo/␥-Al2O3 is observed. For ASA-supported catalysts the

66
I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
Fig. 8. Principal reaction network for guaiacol HDO.
Adapted from Bui et al. [33].
to produce phenol. Guaiacol can also be converted by demethoxyla-
tion (DMO) where the elimination of the methoxy group to directly
produce phenol and methanol occurs as proposed by Bui et al.
[33]. The conversion of phenol can occur by either a direct deox-
ygenation route (DDO) to form benzene or a consecutive route
including hydrogenation (HYD) with subsequent elimination of
oxygen to form deoxygenated products such as cyclohexane. The
methyl group from initial methoxy function in guaiacol can be
transferred in the ortho position to obtain 3-methylcatechol [33].
The elimination of one of the OH groups of this molecule leads to
the formation of cresol and toluene in a subsequent step.
For ␥-Al2O3-supported catalysts, trace amounts of methanol
were presented, and were confirmed in GC–MS; this indicates
that the transformation of guaiacol occurs by the DMO route. The
apparition time of phenol is always higher than catechol for ASA-
supported catalysts. In addition, the fact that methanol was never
detected in any reaction with ASA-supported catalysts is an indi-
cation that phenol is not produced by the DMO of guaiacol route.
The activity (Table 3) and products yield results (Fig. 7) indi-
cate that both supports led to the production of catechol and
2-methoxy-4-methylphenol. In the presence of the metallic sul-
fide phase, the formation of phenol and deoxygenated compounds
was observed; this is an evidence that the active sites of the sulfide
phases promotes the Carom O bond hydrogenolysis as indicated by
Ferrari et al. [34], i.e., the support promotes preferably the guaia-
col transformation by DME route rather than DMO. Moreover, it has
been reported that the acid sites of the support promote the methyl
substitution reactions of the benzene ring. Bui et al. [7,33] described
how the adsorption of methyl (result of the guaiacol DME) can be
carried out in the acid sites of the support surface, and subsequently
embedded in the benzene ring leading methylated compounds as
results. In some cases, these methylated compounds are highly
refractory to HDO [7].
leads to a decrease of the acidity in the catalysts. The decrease of
the acidity is confirmed by the PAD curves quantification for the
␥-Al2O3 and ␥-Al2O3 with Na (Table 2) and the PAD curves (Fig. 1)
indicate that a blockage as the mentioned above can exists for cat-
alysts supported on ␥-Al2O3 with the reduction principally in the
peak size corresponding to type III with Na addition. This result
could be related to a decrease in the quantity of the Al+3 sites of
␥-Al2O3, also evidenced in FTIR results where the reducing of the
peaks corresponding to the type III for the deposition of Na in this
sites are evident [35]. These results suggest that the type III sites
are the principally responsible for the guaiacol transformation over
␥-Al2O3 supported catalysts via DME. These sites correspond to
the hydroxyl group bounded to triple octahedral coordinated alu-
minums. A similar result is observed for the CoMo/ASA25 catalyst;
the comparison between the PAD catalyst modified with Na curve
and the unmodified catalyst curve (Fig. 2b) shows a decrease in the
peak size corresponding to the most acid pH, and the decrease of
activity also is evident.
DRS results can provide information about the reduction in
catalytic activity when Na is added. An effect of decrease in the
Mo octahedral and tetrahedral species which indicate the pres-
ence of phases that are once sulfide, are active in the HDS process
[11]. Moreover, an increase in the phases CoAl2O4 and Co3O4, is
observed. These last two compounds may have a negative effect in
HDO reactions similar to the one presented in the catalytic activity
for HDS reactions.
Additionally, several authors have reported the formation of
Na2MoO4 compounds in Na-doped catalyst [10,26]; these com-
pounds are difficult to sulfide and, therefore, leads to a less active
sulfide phase in HDS reactions [10] and they could also generate a
similar effect in HDO reactions. Although in the XRD results (Fig. 5)
the formation of this compound is not observed, it is suspected that
this compound can be present in these catalysts, but as reported
by Pérez et al. [11], the Na2MoO4 compound cannot be detected
because the size of the crystals is not high enough to be detected by
XRD analysis, which can occur when the atomic relation Na/Mo < 2
[26]. This compound was detected in previous work of our group
in catalysts modified with 5% Na [10], as well as by other authors
[26,31].
4.2. Effect of catalysts modified with sodium in guaiacol HDO
The catalytic activity results (Table 3) show a drop in the ini-
tial rate constant of guaiacol when the supports of the catalysts are
modified with Na compared with the unmodified catalysts. This is
more evidenced with increasing the Na amount. The reduction on
catalytic activity could be related to the blockage of the acid sites
of the support due to the deposition of Na on these sites, which
As seen in the result section (Fig. 7), the sodium addition gen-
erates a significant change in the selectivity of the catalysts. The
introduction of Na to the support generates an increase in the

I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
67
deoxygenated yield, which is more evident for catalysts with more
5. Conclusion
than 2% Na added, compared with undoped catalysts. Another
important effect when the Na is added is the decrease of methylated
compounds production, such as cresol for ␥-Al2O3-supported cat-
alysts and 1,2-dimethoxybenzene and 2-metoxy-4-methylphenol
for the CoMo/ASA25-Na catalyst, that may be partly caused by the
blockage of the strong acid sites of the catalyst by Na, as evidenced
in PAD and FTIR results with decrease of the peaks of type III
sites.
The increase in HDO can be attributed to a change in the inter-
action of the metal with the support phase caused by the decrease
in acidity by adding Na. This phenomenon can generate easier to
sulfide species, which are responsible for removal heteroatoms
breaking the bond Carom O.
Moreover the blocking of acid sites in the support leads the
guaiacol molecule to be adsorbed at the active sites of the sulfide
phases where the Carom O hydrogenolysis takes place instead of
the acid sites of the support, which promotes the DME routes and
the methyl substitution reactions. This may explain the increase
in the deoxygenation degree for both catalysts ␥-Al2O3-supported
and ASA-supported with Na content higher than 2%. (Table 4, Fig. 7).
These results show that it is possible to generate positive changes in
selectivity for the catalysts tested, modifying the acid-base charac-
teristics of the support with the Na addition. However, is necessary
to control this modification in order to not strongly affect other
aspects of the reaction, such as the catalytic activity, as evidenced
in this work.
The CoMo catalysts supported on amorphous aluminosilicates
and alumina modified with Na were tested in the hydrodeoxygena-
tion of guaiacol. The modification of the catalyst with Na caused a
decrease in the catalytic activity, which was higher by increasing
the amount of the Na impregnated. However, the blockage of the
acid sites of the support and the consequent decrease in the acid-
ity of the catalysts due to the presence of the alkali metal caused
a reduction of the reactions associated with the support as the
methyl substitution, and increased the reactions associated with
the active phase such as the Carom O bond hydrogenolysis. These
phenomena could explain the increase in the phenol/catechol
ratio and the reduction in the methylated compounds production.
For the ASA-supported catalysts, it was found that the catalytic
activity decreased with the increase in the Si content; however,
the variations in the Si/(Si + Al) ratio did not generate significant
changes in selectivity. Compared to the ␥-Al2O3-supported cata-
lysts, they showed a lower capacity for the Carom O hydrogenolysis,
evidenced by the low tendency to produce fully HDO compounds.
Acknowledgements
This work was possible due to financial support of the VIE-UIS in
the frame of the project: “Producción de energía a partir de residuos
provenientes de la transformación de la biomasa”, code 5451. Authors
especially thank Y.M. Eugenio and Y.S. Vásquez for its collaboration
performing some of the experiments.
4.3. Effect of the variation of the Si/(Si + Al) ratio in guaiacol HDO
References
[1] Q. Zhang, J. Chang, T. Wang, Y. Xu, Energy Conv. Manag. 48 (2007) 87–92.
[2] P.M. Mortensen, J.D. Grunwaldt, P.A. Jensen, K.G. Knudsen, A.D. Jensen, Appl.
Catal. A – Gen. 407 (2011) 1–19.
[3] B.D. Elliott, E. Bridgwater, A.V. Kuester, J.L. (Eds.), (1988) 883.
[4] R. Maggi, B. Delmon, A review of catalytic hydrotreating processes for the
upgrading of liquids produced by flash pyrolysis, in: B.D.G.F. Froment, P. Grange
(Eds.), Studies in Surface Science and Catalysis, vol. 106, Elsevier, 1997, pp.
99–113.
[5] E. Furimsky, Appl. Catal. A – Gen. 199 (2000) 147–190.
[6] E. Laurent, A. Centeno, B. Delmon, in: B. Delmon, G.F. Froment (Eds.), Stud. Surf.
Sci. Catal., Elsevier, 1994, pp. 573–578.
[7] V.N. Bui, D. Laurenti, P. Delichère, C. Geantet, Appl. Catal. B – Environ. 101 (2011)
246–255.
The variation in Si/(Si + Al) ratio for CoMo/ASA modifies mainly
the activity as shown in the result section (Table 3). The reduction
in catalytic activity by increasing the Si amount may be analyzed
by mean of the DRS spectra showed in Fig. 3.
Concerning the decrease in catalytic activity related to the
increase in the Si amount in ASA supported catalysts, differ-
ent hypothesis can be proposed based on the results outlined
before. DRS spectra show a decrease in the bands corresponding
to Mo octahedral and tetrahedral with increasing amounts of Si.
These species, specially the Mo octahedral, have been reported
as precursors of the active phase for HDS reactions [30]. This
behavior observed in the DRS spectra of ASA-supported catalysts
can be correlated to the catalytic activity results for HDO reac-
tions; in these results it is confirmed that the catalyst CoMo/ASA75
presents higher reduction in this species Mo and Co mentioned
above.
Furthermore, the SEM results (Fig. 6b) show the formation
of agglomerate with CoMo/ASA75, which is not observed with
CoMo/ASA15. These agglomerates may correspond to Mo species.
Moreover, as observed in the XRD spectra (Fig. 5), a peak cor-
responding to Co or Mo phases appears. This may indicate that
the increase of Si amount leads to the dispersion being negatively
affected [30,31], and consequently generates a low activity of gua-
iacol conversion.
[8] A. Centeno, E. Laurent, B. Delmon, J. Catal. 154 (1995) 288–298.
[9] C. Sepúlveda, N. Escalona, R. García, D. Laurenti, M. Vrinat, Catal. Today 195
(2012) 101–105.
[10] D.J. Pérez-Martı´nez, P. Eloy, E.M. Gaigneaux, S.A. Giraldo, A. Centeno, Appl. Catal.
A – Gen. 390 (2010) 59–70.
[11] D.J. Pérez-Martínez, E.M. Gaigneaux, S.A. Giraldo, A. Centeno, J. Mol. Catal. A –
Chem. 335 (2011) 112–120.
[12] V. La Parola, G. Deganello, S. Scirè, A.M. Venezia, J. Solid State Chem. 174 (2003)
482–488.
[13] C. Contescu, J. Jagiello, J.A. Schwarz, in: J. Martens, B. Delmon, P.A. Jacobs, G.
Poncelet, P. Grange (Eds.), Stud. Surf. Sci. Catal., Elsevier, 1995, pp. 237–252.
[14] C. Contescu, J. Jagiello, J.A. Schwarz, Langmuir 9 (1993) 1754–1765.
[15] M. Adachi, C. Contescu, J.A. Schwarz, J. Catal. 158 (1996) 411–419.
[16] E. Laurent, B. Delmon, Appl. Catal. A – Gen. 109 (1994) 77–96.
[17] B.S. Gevert, et al., Kinetics of the HDO of methyl-substituted phenols, Appl.
Catal. 31 (1) (1987) 119–131.
[18] H. Knözinger, P. Ratnasamy, Catal. Rev. – Sci. Eng. 17 (1978) 31–70.
[19] T.J. Bandosz, C. Lin, J.A. Ritter, J. Colloid Interface Sci. 198 (1998) 347–353.
[20] C. Contescu, A. Contescu, J.A. Schwarz, J. Phys. Chem. 98 (1994) 4327–4335.
[21] H. Praliaud, J. Less – Common Met. 54 (1977) 387.
[22] J.H. Ashley, P.C.H. Mitchell, J. Chem. Soc. A (1968) 2821–2827.
[23] J. Vakros, K. Bourikas, Ch. Kordulis, A. Lycourghiotis, J. Phys. Chem. B 107 (2003)
1804.
[24] L.G.A Van De Water, J.A. Bergwerff, B.R.G. Leliveld, B.M. Weckhuysen, K.P. de
Jong, J. Phys. Chem. B 109 (2005) 14513.
[25] L.G. Van De Water, J.A. Bergwerff, T.A. Nijhuis, K.P. de Jong, B.M. Weckhuysen,
J. Am. Chem. Soc. 127 (2005) 5024.
[26] A. Lycourghiotis, C. Defosse, F. Delannay, B. Delmon, J. Chem. Soc. Faraday Trans.
I 76 (1980) 2052–2064.
[27] Ch. Papadopoulou, J. Vakros, H.K. Matralis, Ch Kordulis, A. Lycourghiotis, J.
Colloid Interface Sci. 261 (2003) 146.
Comparing the products distribution for ASA and ␥-Al2O3-
supported catalysts the low capacity to produce fully HDO
compounds and no production of phenol via DMO for ASA-
supported compared to ␥-Al2O3-supported catalysts is notable.
This indicates that the first one has a lower capacity for Carom
O
bond hydrogenolysis. Since both supports have the same metallic
phase, this behavior may be related to differences in the sulfide
phase and/or in the support-metal sulfide phase interaction. This
result is important considering that the main objective of catalytic
HDO of these molecules is to achieve the highest possible degree
of deoxygenation.
[28] M. Dignea, P. Sautet, P. Raybaudc, P. Euzend, H. Toulhoate, J. Catal. 211 (2002)
1–5.

68
I.D. Mora et al. / Applied Catalysis A: General 474 (2014) 59–68
[29] J. Liu, P. Ying, Q. Xin, C. Li, Appl. Spectrosc. 53 (1999) 40–42.
[30] D.J. Pe´rez-Martı´nez, E.M. Gaigneaux, S.A. Giraldo, Appl. Catal. A – Gen. 421
(2012) 48–57.
[33] V.N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Appl. Catal. B – Environ. 101 (2011)
239–245.
[34] M. Ferrari, S. Bosmans, R. Maggi, B. Delmon, P. Grange, Catal. Today 65 (2001)
[31] V. La Parola, G. Deganello, A.M. Venezia, Appl. Catal. A – Gen. 260 (2004)
237–247.
[32] V.L. Barrio, P.L. Arias, J.F. Cambra, M.B. Güemez, B. Pawelec, J.L.G. Fierro, Fuel 82
(2003) 501–509.
257–264.
[35] S. Srinivasan, C.R. Narayanan, A. Biaglow, R. Gorte, A.K. Datye, Appl. Catal. A –
Gen. (1995) 271–287.