Conversion of guaiacol over supported ReOx catalysts: Support and metal loading effect

Article abstract of DOI:10.1016/j.cattod.2017.04.002

In the present work, the catalytic activity and selectivity for the conversion of 2-methoxyphenol over ReOx catalysts supported on SiO₂, Al₂O₃ and SiO₂-Al₂O₃ in a batch reactor at 300 °C and 5 MPa of hydrogen pressure were studied. Additionally, several SiO₂-supported ReOx catalysts were synthesized with different metal loadings. These catalysts were characterized by N₂ adsorption, X-ray photoelectron spectroscopy (XPS), temperature programmed reduction (TPR), temperature-programmed desorption of ammonia (TPD-NH₃), Raman spectroscopy and ultraviolet visible (UV–vis-DR) spectroscopy. ReOx catalyst supported over SiO₂ revealed the highest catalytic activity and selectivity towards deoxygenated products. This was attributed to a higher formation of oxygen vacancies (active sites) on SiO₂, due to a weaker metal-support interaction. The increase of Re content on SiO₂ resulted in an increase of guaiacol conversion, reaching a maximum at about 1.8 atoms of Re per nm² of support. The decrease of activity over 1.8 atoms of Re per nm²of support was correlated to loss of active site due to the formation of aggregates of ReOx.

Full text of DOI:10.1016/j.cattod.2017.04.002

Catalysis Today xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Conversion of guaiacol over supported ReOx catalysts: Support and metal
loading effect
K. Leivaa, R. Garciaa, C. Sepulvedaa, D. Laurentib, C. Geantetb, M. Vrinatb, J.L. Garcia-Fierroc,
⁎
N. Escalonad,e,f,
a
Universidad de Concepción, Facultad de Ciencias Químicas, Casilla 160C, Concepción, Chile
IRCELYON, UMR 5256 CNRS, Université Lyon 1, 2 Avenue A. Einstein, 69626 Villeurbanne Cedex, France
Instituto de Catálisis y Petroquímica, CSIC, Cantoblanco, 28049 Madrid, Spain
Departamento de Ingeniería Química y Bioprocesos, Escuela de Ingeniería, Pontificia Universidad Católica de Chile, Avenida Vicuña Mackenna 4860, Macul, Santiago,
b
c
d
Chile
e
Departamento de Química Física, Facultad de Ciencias Químicas, Pontificia Universidad Católica de Chile, Chile
Centro de Investigación en Nanotecnología y Materiales Avanzados (CIEN-UC), Pontificia Universidad Católica de Chile, Santiago, Chile
f
A R T I C L E I N F O
A B S T R A C T
Keywords:
Guaiacol
Supports
ReOx
Biomass
HDO
In the present work, the catalytic activity and selectivity for the conversion of 2-methoxyphenol over ReOx
catalysts supported on SiO2, Al2O3 and SiO2-Al2O3 in a batch reactor at 300 °C and 5 MPa of hydrogen pressure
were studied. Additionally, several SiO2-supported ReOx catalysts were synthesized with different metal
loadings. These catalysts were characterized by N2 adsorption, X-ray photoelectron spectroscopy (XPS),
temperature programmed reduction (TPR), temperature-programmed desorption of ammonia (TPD-NH3),
Raman spectroscopy and ultraviolet visible (UV–vis-DR) spectroscopy. ReOx catalyst supported over SiO2
revealed the highest catalytic activity and selectivity towards deoxygenated products. This was attributed to a
higher formation of oxygen vacancies (active sites) on SiO2, due to a weaker metal-support interaction. The
increase of Re content on SiO2 resulted in an increase of guaiacol conversion, reaching a maximum at about 1.8
atoms of Re per nm2 of support. The decrease of activity over 1.8 atoms of Re per nm2of support was correlated
to loss of active site due to the formation of aggregates of ReOx.
1. Introduction
aliphatic alcohols in bio-oil [4–7]. To upgrade bio-oil, a catalytic
hydrotreating method known as hydrodeoxygenation (HDO) can be
The increase in worldwide energy demand, the decrease of fossil
fuel reserves and the negative environmental impact of fossil fuel have
led to focus on renewable and sustainable carbon sources. In this
context, lignocellulosic biomass is one of the main attractive alter-
natives to fossil fuels as this kind of biomass is available in many
countries and does not compete with food production.
Different thermochemical conversion processes can be employed to
convert biomass to chemicals and fuels [1]. For example, through fast
pyrolysis (400–500 °C under inert atmosphere), bio-oil is produced at
relatively high yields (60%), and this liquid is a potential source of
biofuels and chemicals [2,3]. However, the pyrolytic bio-oil cannot be
directly used as fuel due to their high viscosity, corrosiveness and
thermal instability making it impossible for storage and transport. This
is due to complex mixture of oxygen-containing compounds in the form
of phenol derivatives, aldehydes, ketones, carboxylic acids, esters and
utilized to remove oxygen in the form of H2O to improve the fuel
properties of bio-oil [8,9]. Because of the complexity of the composition
of bio-oil, the high amount of water and the presence of lignin
fragments, investigations dealing with bio-oil upgrading are focused
mainly on highly representative compounds. In this context, guaiacol
(2-methoxyphenol) which is representative of the lignin units in bio-oil
has become an attractive molecule for HDO studies [10–14].
To perform catalytic hydrodeoxygenation reactions, sulfides
[13,15–17], nitrides [18–20], and noble and non-noble metal based
catalysts [21–23] are the most commonly studied active phases. The
principal disadvantage of sulfide catalysts is the necessity of external
addition of a sulfiding agent to keep the sulfide phase. Thus, there has
been a recent shift on exploring non-sulfide catalysts for HDO. In this
context, it has been reported that reduced metallic catalysts displayed
higher activity and selectivity to hydrodeoxygenated compounds than
⁎
Corresponding author at: Departamento de Ingeniería Química y Bioprocesos, Escuela de Ingeniería, Pontificia Universidad Católica de Chile, Avenida Vicuña Mackenna 4860, Macul,
Santiago, Chile.
E-mail address: neescalona@ing.puc.cl (N. Escalona).
http://dx.doi.org/10.1016/j.cattod.2017.04.002
Received 5 February 2017; Received in revised form 21 March 2017; Accepted 4 April 2017
0920-5861/ © 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Leiva, K., Catalysis Today (2017), http://dx.doi.org/10.1016/j.cattod.2017.04.002

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
conventional sulfide catalysts [24].
heated in 5% H2/Ar mixture under 50 cm3 min−1flow.The samples
Transition metal carbides, nitrides and phosphides have also shown
promise in HDO. Boullosa-Eiras et al. [25] compared the catalytic
activity of TiO2-supported Mo carbide, nitride, phosphide and oxide
catalysts in phenol HDO reaction, and showed that MoO3 catalyst
displayed the highest activity. However, all the catalysts presented
similar selectivity to benzene and cyclohexene.
Shetty et al. [26] studied the HDO of m-cresol to toluene on
supported molybdenum oxide catalysts at very low H2 pressure
(≤1 bar). The authors found that catalysts supported on TiO2and
ZrO2 were highly active and stable due to stabilization of
MoO3oxidation states. Moreover, the authors proposed that the activity
of Mo oxide is related to coordinatively unsaturated Mo sites (oxygen
vacancies).
Rhenium is one of the non-conventional catalysts that have been
extensively used in hydrotreating reactions, in both sulfide and metallic
phases [13,14,17,27–32]. Initial efforts were focused on ReS2 as an
active phase in hydrodesulfurization [17,29,30,33], and HDO reactions
[13,14,17,28], in which ReS2/SiO2 catalyst presented a higher activity
than conventional NiMoS/Al2O3 catalyst. This behavior was attributed
to metal-like character of ReS2which favored hydrogenation [31].
Previously, we studied the hydrodeoxygenation of guaiacol over
silica-supported Re oxide, Re metal, and Re sulfide catalysts [34]. The
highest activity was obtained on ReOx/SiO2 catalyst; furthermore, this
catalyst was highly selective towards benzene and cyclohexane. In
contrast, ReS2/SiO2 and Re/SiO2 catalysts principally produced phenol.
The unique behavior of ReOx/SiO2 was attributed to the presence of
oxygen vacancies.
In order to further improve the performance of ReOx catalysts, it is
important to understand some fundamental factors that control reac-
tivity, such as the influence of support and metal loading. These
properties are yet to be studied for ReOx catalysts for HDO reactions.
Therefore, the objective of this work isto understand the effect of
support and metal content on the conversion of guaiacol as a lignin
model molecule. This was achieved by comparison of the reactivity of
well-characterized ReOx supported on SiO2, Al2O3 and SiO2-Al2O3
catalysts in the HDO of guaiacol at 300 °C and 5 MPa of hydrogen
pressure in a batch reactor. The effect of metal content was carried out
using the most active catalytic system.
were heated from 25 °C to 1050 °C with a heating rate of 10 °C min−1
.
XPS measurements were performed using a VG Escalab 200R
electron spectrometer equipped with a hemispherical electron analyzer
and Mg Kα (1253.6 eV) excitation source. Energy corrections were
performed using the line of each support as internal reference. The
intensity of the peaks was estimated by calculating the integral of each
peak after subtracting an S-shaped background and fitting the experi-
mental curve to a combination of Gaussian/Lorentzian lines.
The UV–vis-DR spectra of dried samples were measured using a
UV–vis-Lambda 35 PerkinElmer Spectrophotometer, equipped with a
quartz cell, provided with a diffuse reflectance sphere for powder
analysis (Labsphere). Before analysis, all samples were milled using an
agate mortar and diluted with BaSO4 to maximize reflectance. The
UV–vis-DR spectra were recorded in the wavelength range
200–1100 nm at ambient conditions. The spectra obtained were
decomposed with the PeakFit v4.12 software.
Raman spectra of the supported metal oxide catalysts were collected
from 0 to 3000 cm−1with a LabRam HM Raman spectrometer (Horiba-
JobinYvon) equipped with a BXFM confocal microscope, employing the
514.5 nm line of an Ar+ laser as the excitation source. All the recorded
data were treated using the LABSPEC software.
Temperature-programmed desorption of ammonia (TPD-NH3) ana-
lyses were performed in
a TPR/TPD Micromeritics 2900 system
equipped with athermal conductivity detector (TCD). Prior to analysis
the samples were cleaned under He flow of 50 mL min−1 at 383 K for
30 min, and then subjected to ammonia pulses until saturation of the
catalyst surface at 120 °C. Then, the samples were cooled to room
temperature under He, followed by flushing under He to remove weakly
adsorbed NH3. Once the baseline was stabilized, TPD-NH3 measure-
ments were performed while the temperature was increased linearly to
700 °C with a heating rate of 10 °C min−1
.
2.3. Catalytic tests
The conversion of guaiacol was carried out in an autoclave reactor
operating in batch mode. The liquid reactant feed, consisting of
guaiacol (0.232 mol L−1) in n-dodecane (80 mL) with hexadecane
(0.0341 mol L−1) as an internal standard, were introduced into the
reactor together with0.200 g of solid catalyst. The system was closed,
2. Experimental
and N2 was bubbled through the solution for 10 min with
a
100 mL min−1 flow to purge the system. The reactor was heated under
stirring to the reaction temperature of 300 °C under N2 atmosphere. The
initial reaction time (t0) was defined when the reaction temperature
was reached and the pressure was adjusted to 5 MPa of H2. The pressure
was kept constant during the course of the experiment. Liquid samples
were taken periodically during the reaction(0, 10, 15, 30, 60, 120, 180,
240 and 300 min), and were analyzed by gas chromatography (Perkin-
Elmer Clarus 680) GCMS-SQ8T, and quantified by gas chromatography
(Perkin-Elmer Clarus 400) equipped with a Flame Ionization Detector
(FID) and a CP-Sil 5 column (Agilent, 30 m × 0.53 mm × 1.0 μm film
thickness).The specific rate for the total conversion of guaiacol was
calculated from the initial slope of conversion as a function of time plot
according to the following equation:
2.1. Catalyst preparation
The catalysts were prepared by incipient wetness impregnation with
aqueous solution of NH4ReO4 (Aldrich, 99%) on different supports
(Silica Grace 432, Al2O3 SPH 501, and SiO2-Al2O3Ketjen HA 100 SP
(26.2%Al2O3)). For the study of the support effect, the catalysts were
prepared with a Re nominal loading of 1.8 atoms of Re per nm2 of
support. The impregnated catalysts were left for maturation at room
temperature for 24 h, dried at 120 °C for 12 h and then calcined at
300 °C for 0.5 h. For the study of the effect of metal loading, the
catalysts were prepared with a loading of 0.6-2.7 atoms of Re per nm2
of SiO2 support following the previously described procedure. The Re
content was determined by ICP using the line of emission 221.426 nm
of Re. The values are summarized in Table 1.
[b × n]
rs =
(1)
m
2.2. Characterization of the supports and catalysts
Where rs is the specific rate (moles of guaiacol transformed per gram of
catalyst per second and expressed in mol. gcat−1s−1), b represents the
initial slope of the conversion vs. time plot (s−1), n is the initial moles of
guaiacol (mol), and m is the mass of catalyst (g). The intrinsic rate was
calculated from the specific rate according to the following equation:
The BET surface area (SBET) and the total pore volume (Vp) of
catalysts and supports were determined from nitrogen adsorption-
desorption isotherms at −196 °C using a Micromeritics-TriStar II
3020instrument.
Temperature programmed reduction (TPR) studies were carried out
in a quartz cell on a conventional system equipped with a thermal
conductivity detector. In each experiment, 100 mg of the sample was
rs
nRe
ri =
× Nav
(2)
Where ri is the intrinsic rate (i.e. molecules of guaiacol transformed per
2

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 1
Textural properties of supports and catalysts.
Samples
Re2O7 content
(%)
Re surface density
SBET
Vp
Vo
Vm
(atoms nm−2
)
(m2g−1
)
(cm3g−1
)
(cm3g−1
)
(cm3g−1
)
SiO2
SiO2-Al2O3
γ-Al2O3
ReOx(1.8)/SiO2-Al2O3
ReOx(1.8)/γ-Al2O3
ReOx(0.6)/SiO2
ReOx(1.3)/SiO2
ReOx(1.8)/SiO2
ReOx(2.2)/SiO2
ReOx(2.7)/SiO2
–
–
–
13
15
6
11
14
17
21
–
–
–
1.8
1.8
0.6
1.3
1.8
2.2
2.7
305
352
240
214
171
306
281
265
236
224
1.1
0.20
0.20
0.14
0.15
0.10
0.14
0.13
0.12
0.11
0.10
0.90
0.40
0.36
0.25
0.20
0.89
0.75
0.67
0.61
0.58
0.60
0.50
0.40
0.30
1.03
0.88
0.88
0.72
0.68
Re atoms per second and expressed in molec. Re at−1s−1), nRe is the
number of Re atoms per gram of catalyst, and NAV is the Avogadro’s
number.
The selectivities (%) were determined at 20% and 50% of guaiacol
conversion, according to the following equation:
Xi
XT
S% =
× 100
(3)
Where Xi is the percentage of product I formed, and XT is the guaiacol
conversion.
3. Results and discussion
3.1. Support effect
The BET surface areas (SBET), micropore (Vo), mesopore (Vm) and
total (Vp) pore volumes of the supports and catalysts are summarized in
Table 1. It can be noted that the surface area of the supports varied in
the order SiO2-Al2O3 > SiO2 > γ-Al2O3. Also, it shows that the SBET
and pore volumes of the three catalysts with the same Re content (1.8
atoms of Re per nm2 of support) decreased in comparison with the
corresponding supports; more specifically, a decrease in the SBET of 13%
for SiO2, 39% of SiO2-Al2O3 and 29% for Al2O3.These results suggest
−
that ReO4 species were best dispersed on silica where it is observed
that the micropore and mesopore volume decrease only slightly after
impregnation, indicating that blockage of the mouth of the pores is
virtually absent. Meanwhile, the highest decrease of SBET observed for
ReOx(1.8)/SiO2-Al2O3in comparison with SiO2-Al2O3support suggests a
partial blocking at the mouth of the pores.
Fig. 1a shows the TPR profiles of ReOx(1.8)/support catalysts. It can
be seen that the ReOx(1.8)/SiO2catalystpresented a single reduction
peak at 341 °C. This suggests a single reduction step of ReOx species to
metallic Re on silica, in agreement with the results of Arnoldy et al.
[35]. However, the TPR profile of ReOx(1.8)/γ-Al2O3displayed two
reduction peaks centered at 361 °C and 383 °C. Similarly, the ReOx
(1.8)/SiO2-Al2O3catalystsdisplayed two reduction peaks, but they were
centered at 363 °C and 378 °C. The inset of Fig. 1a shows that reduction
of NH4ReO4 precursor occurs at 387 °C, while that of bulk ReOx
(obtained after calcinations of NH4ReO4 at 300 °C for 0.5 h) occurs at
a lower temperature of 358 °C. Therefore, the low-temperature peak
observed for ReOx(1.8)/γ-Al2O3and ReOx(1.8)/SiO2-Al2O3can be at-
tributed to ammonium perrhenate species, which were not completely
decomposed during calcination, in accordance with previous observa-
tions made by Arnoldy et al. [35]. On the other hand, the higher-
temperature peak could be assigned to rhenium oxide species present
on the surface of the catalyst. The difference in behavior between the
SiO2-supported catalyst and the other supported catalysts can be
attributed to different acid-base properties of each support, affecting
the strength of the interaction between ReOx or ReO4− species and the
surface of the supports. Another notable observation from Fig. 1a is the
difference in intensity of the reduction peaks (particularly the higher-
Fig. 1. TPR profiles of a) ReOx(1.8)/support and b) ReOx(x)/SiO2 catalysts. Inset Fig. 1a:
TPR of NH4ReO4 and ReOx (after calcination of NH4ReO4 for 0.5 h at 300 °C).
temperature peak) of ReOx(1.8)/SiO2-Al2O3 and ReOx(1.8)/γ-Al2O3
despite the nearly identical reduction temperatures. This is because the
catalysts were prepared with identical Re surface density but contain
different metal loadings (in wt%) due to the different SBET of the
supports. Furthermore, Arnoldy et al. [35] studied the reduction
temperature of Re2O7catalysts supported on Al2O3, SiO2 and carbon,
and found that the difference in reducibility was due to a variation in
the strength and heterogeneity of the Re7+-support interaction, which
depended on the support used, and decreased in the order,
Al2O3 > SiO2 > carbon. Mitra et al. [36] found that the reduction
temperature of ReOx species correlated with the binding energy
3

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 3
XPS binding energies (eV) and surface atomic ratios of Si2p, Al2p, O1 s and Re4f species
for the oxides catalysts.
Catalysts
Si2p
(eV.)
Al2p O1s
(eV.) (eV.)
Re4f
(eV.)
Re/M (Si, Al, Si + Al)
at
ReOx(1.8)/Al2O3
ReOx(1.8)/SiO2-
Al2O3
–
74.5
532.7 46.3
532.8 46.2
0.073
0.153
103.4 75.2
ReOx(0.6)/SiO2
103.4
103.4
103.4
103.4
103.4
–
–
–
–
–
532.8 46.1 (65) 0.022
47.3 (35)
532.8 45.9 (55) 0.040
47.5 (45)
532.7 45.9 (53) 0.054
47.5 (47)
532.7 45.7 (63) 0.041
47.4 (37)
ReOx(1.3)/SiO2
ReOx(1.8)/SiO2
ReOx(2.2)/SiO2
ReOx(2.7)/SiO2
532.7 45.6 (59) 0.042
47.3 (41)
Fig. 2. Temperature-programmed desorption of ammonia (TPD-NH3) for ReOx(1.8)/
SiO2, ReOx(1.8)/Al2O3 and ReOx(1.8)/SiO2-Al2O3 catalysts.
species, shifting their reduction peak to higher temperature.
Fig. 2 presents the temperature-programmed desorption of ammo-
nia (TPD-NH3) of ReOx(1.8)/SiO2, ReOx(1.8)/Al2O3 and ReOx(1.8)/
SiO2-Al2O3 catalysts. The strength of the acid sites of the catalysts are
related to the temperature of NH3 desorption: weak acid sites
(< 300 °C), medium acid sites (between 300 °C and 500 °C), and strong
acid sites (> 500 °C)[37]. According to this classification, Fig. 2 shows
that all the three catalysts possess weak and medium acid sites, but with
different peak intensities, indicating an influence of the support.
Table 2 presents the acid site distribution and the total acid sites
expressed as quantity of ammonia desorbed per gram of catalyst. The
ReOx(1.8)/A2O3 and ReOx(1.8)/SiO2 catalysts presented a predomi-
nance of weak acid sites, while ReOx(1.8)/SiO2-Al2O3 displayed the
highest amount of medium acid sites. The total acid sites decreased in
order: ReOx(1.8)/SiO2-Al2O3 > ReOx(1.8)/Al2O3 > ReOx(1.8)/SiO2
catalysts. However, it must be pointed out that ReOx(1.8)/A2O3 had
the highest number of strong acid sites.
Table 2
Surface acid sites calculated from NH3 desorption (TPD-NH3).
Catalysts
Weak acid
sites
Medium acid
sites
Strong acid
sites
Total acid
sites
(10−4 mol
(10−4 mol
(10−4 mol
(10−4 mol
NH3 g−1
)
NH3 g−1
)
NH3 g−1
0.29
)
NH3 g−1
)
ReOx(1.8)/
SiO2-
Al2O3
ReOx(1.8)/
Al2O3
1.14
1.41
3.17
1.59
0.95
0.31
0.17
0.78
0.14
2.68
1.26
ReOx(1.8)/
SiO2
between the metal and the support, which decreased in the following
order: Re-O-Al > Re-O-Si-Al > Re-O-Si. Therefore, these results suggest
that less acidic supports like SiO2 promote a weak interaction with the
ReOx or NH4ReO4species, favoring only one low-temperature reduction
peak. Meanwhile, the acid sites present on the γ-Al2O3 and SiO2-Al2O3
XPS spectra of Re 4f region of ReOx(1.8)/SiO2, ReOx(1.8)/Al2O3
and ReOx/SiO2-Al2O3 catalysts are shown in Fig. 3. Curve fitting of the
spectra for ReOx(1.8)/SiO2 revealed two partially overlapped doublet,
each one containing the Re 4f7/2 and 4f5/2 peaks. Meanwhile, curve
fitting of the spectra of ReOx(1.8)/Al2O3 and ReOx/SiO2-Al2O3 cata-
−
supports could favor a stronger interaction with ReO4 and ReOx
Fig. 3. X-ray photoelectron spectra of the Re 4f region on: a) ReOx(1.8)/SiO2, b) ReOx(1.8)/Al2O3 and ReOx(1.8)/SiO2-Al2O3 catalysts.
4

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 4. Raman spectroscopy of calcined Re(x)/SiO2 catalysts.
Fig. 6. Relationship between the XPS Re/Si atomic ratio and the Re content for ReOx(x)/
Fig. 5. UV–vis-DR spectra of oxide ReOx(x)/SiO2 catalysts.
SiO2 catalysts.
lysts shows a doublet containing Re 4f7/2 and 4f5/2 peaks. Table 3
summarizes the binding energies (BE) of the most intense Re 4f7/2
component of each doublet, their relative proportion and surface Re/Si
dispersion was highest over SiO2-Al2O3 support, while Re was similarly
dispersed over Al2O3 and SiO2supports.
(Al, Si + Al) atomic ratios. Table
3 shows that the ReOx(1.8)/
3.2. Metal loading effect
SiO2catalyst presents two Re4f7/2 contributions with BE at
45.7
0.2 eV and 47.4
0.1 eV. The BE of 45.7
0.2 eV is in
The textural properties of the catalysts are shown in Table 1, where
it is possible to notice a gradual decrease in the surface area, micropores
and mesopores volume with increase in the Re content over SiO2
support. This behavior is consistent with the homogeneous deposition
of Re species in the pores of the support.
Fig. 1b depicts the reduction profiles for the ReOx/SiO2catalysts as a
function of Re metallic loading. This figure shows that the ReOx(x)/
SiO2 catalysts with 0.6 and 1.3 atoms of Re per nm2 of support
displayed two reduction peaks. These results suggest that at low Re
metal content, the ReOx and NH4ReO4 species interacted strongly with
the support. However, this interaction decreases at higher metal
contents and it can be attributed to an increase in crystallite size by
increasing Re loading.
the lower range of Re7+ (45.5–47.4) [38–43], which could be assigned
to oxides of Re species with minor oxygen atoms in its environment
(ReOx). Meanwhile, the BE of 47.4
0.1 eV was assigned to Re+7
−
species corresponding to Re2O7or ReO4 species [38–43]. However,
these species were not differentiated by TPR, suggesting that these
species have the same reducibility on SiO2 support. On the other hand,
the ReOx(1.8)/Al2O3 and ReOx(1.8)/SiO2-Al2O3 catalysts displayed a
contribution of Re 4f7/2 with BE of 46.3 eV and 46.2 eV, respectively.
The BEs of 46.3 eV and 46.2 eV are in the range of Re+7 corresponding
to ReO4 species [38–43]. Table 3 also shows that the Re/Si (Al or Si
+ Al) surface atomic ratio decreases in the following order: Re/(Si
+ Al) > Re/Al ≈ Re/Si. These results suggest that the apparent Re
−
5

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 7. Yield of product in function of guaiacol conversion of a) ReOx(1.8)/SiO2, b) ReOx(1.8)/Al2O3 and c) ReOx(1.8)/SiO2-Al2O3 catalyst.
In order to identify the different ReOx species on SiO2 support, the
acteristic bands of solid perrhenate (used as precursor). However for
ReOx(1.3)/SiO2, the peaks are not intense and we had to use a lower
range to get a spectrum. This figure also shows that from 1.8 atoms of
ReOx(x)/SiO2 catalysts were characterized by Raman spectroscopy and
their spectra are shown in Fig. 4. The spectrum of NH4ReO4, included in
this figure as a reference, shows 4 characteristic bands centered at 342,
891, 914 and 965 cm−1. The bands centered at 342 and 965 cm−1are
attributed to symmetric stretching of the terminal Re = O bonds of
Re per nm2 of support,
a
new Raman band centered at
990 cm−1appears. This band has also been assigned to tetrahedral
ReO4− surface species [36,44,45], and suggests the formation of larger
ReO4−crystals.
−
ReO4 surface species [36,44–47]. Meanwhile, the band centered at
891 cm−1 corresponds to ReeOeRe bond vibration [36,44–47].
Fig. 5 depicts the UV–vis-DR spectra of the ReOx(x)/SiO2 catalysts.
The four ReOx/SiO2catalysts (Fig. 4) present the same four char-
The catalysts containing 0.6 atoms of Re per nm2 exhibited a very
6

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 8. Reaction network for the conversion of guaiacol [35].
intense single adsorption band at 246 nm. Meanwhile, at higher Re
contents, the catalysts exhibited bands at 246, 322 and 456 nm, in
which the intensities increased with Re content. Wang and Hall [48]
relative proportion and surface Re/Si atomic ratios. The ReOx(x)/SiO2
catalysts presented two contributions of Re 4f7/2with BE of
45.9
0.3 eV and 47.5
0.2 eV. The BE of 45.7
0.2 eV is in
the lower range of Re7+ (45.5–47.4)[38–43], which could be assigned
−
reported that only tetrahedral ReO4 species with bands near 205 and
230 nm were found on catalysts with up to 2.9% of Re supported on
alumina. However, degeneration increases when ReO4 ion is dis-
to oxides of Re species with minor oxygen atoms in its environment
0.1 eV was assigned to Re+7
−
(ReOx). Meanwhile, the BE of 47.4
−
torted, and three bands appear near 205, 230 and 295 nm. Stoyanaba
et al. [49] studied the Re content of the oxide catalysts supported on
alumina for olefin metathesis. They observed by UV–vis two adsorption
bands for the catalysts which were centered at 330 nm and
235–240 nm, and increased in intensity with increasing Re content.
The band at 235 nm was assigned to Re7+species with tetrahedral
coordination (Td), while the band at 350 nm was attributed to mixtures
of Re oxide species. According to the results obtained for ReOx(x)/SiO2
species corresponding to Re2O7 or ReO4 species [38–43], in agree-
ment with Raman spectroscopy and UV–vis-DR results. Also, Table 3
shows that Re/Si atomic ratio gradually increases with Re content up to
1.8 atoms of Re nm−2 of SiO2 support, then decreases and remains
substantially constant. This trend is better observed in Fig. 6. The
gradual increase observed in the Re/Si surface atomic ratio with Re
content up to 1.8 atoms of Re per nm2 support suggests a high
dispersion of Re oxide species (probably monolayer-like) on the SiO2
support. The observed deviation from linearity above 1.8 Re atoms per
nm2 (decrease in surface atomic ratio) suggests the formation of the
agglomerates of Re oxide species, in agreement with Raman spectro-
scopy results.
−
catalysts, the band at 246 nm can be attributed to tetrahedral ReO4
species and the band observed at 322 nm can be assigned to mixtures of
Re oxide species, in agreement with Raman spectroscopy results.
Meanwhile, the band centered at 456 nm can be assigned to ReOx
species that could be interacting with the support, according to
previous findings [33].
XPS spectra of the Re 4f region of ReOx(x)/SiO2catalysts revealed
two partially overlapped doublets, each one containing the Re 4f7/2 and
4f5/2 peaks (not shown here).Table 3 summarizes the binding energies
(BE) of the most intense Re 4f7/2 component of each doublet, their
3.3. Catalytic activity
Fig. 7 shows the yield of products as a function of guaiacol
conversion over Re(1.8)/support catalysts. The main reaction products
obtained over all the catalysts were phenol, BTX (mainly benzene and
7

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 9. Products distribution on ReOx(1.8)/support catalysts calculated at a) 20% and b) 50% of guaiacol conversion.
Fig. 10. a) Initial specific rates and b) intrinsic rates over ReOx(1.8)/support catalysts.
toluene, traces of xylene) and cyclohexane. As side products, the
formation of cyclohexanol and anisole were observed. The formation
of methanol, cyclohexene, hexane, methylcyclohexane, cyclohexanone
and methyl phenol were also observed but the total yield of these
products were very small (< 2%). Phenol and anisole production
increased with time and then decreased, indicating that both com-
pounds are intermediate products. Thus, these results are consistent
with the guaiacol conversion scheme previously reported by Leiva et al.
[34] for ReOx/SiO2 catalyst, which was adapted from Laurent and
Delmon [50], Bui et al. [15,51] and Nimmanwudipong et al. [52], and
shown in Fig. 8. On the basis of this reaction scheme, guaiacol can be
transformed by direct deoxygenation route (DDO) to form anisole, by
direct demethoxylation (DMO) to form phenol, by demethylation
(DME) to form catechol (it was not detected) and by hydrogenation
(HYD) to form methoxyclyclohexanone. Then, benzene can be formed
by DDO of phenol or by DMO of anisole. Cyclohexene can be formed by
HYD of benzene or DMO of methoxyclyclohexanone to form cyclohex-
anone, followed by hydrogenation and dehydration. On the other hand,
methylation (ME) of phenol to methyphenol can occur, followed by
DDO to toluene. Xylene (detected in trace amount) can be formed either
by successive methylation of phenols and DDO or by ME of toluene.
Meanwhile, methylcyclohexane can be formed by HYD of toluene or
methylcyclohexene. Finally, cyclohexane (traces) could form hexane
through the cracking route (CR). These results show that ReOx/support
catalysts have a high potential for the hydrogenolysis of CeOH bond.
The products distribution calculated at 20% and 50% of guaiacol
conversion over the catalysts dispersed on different supports (ReOx
(1.8)/supports) is shown in Fig. 9. The catalysts displayed similar
products of reaction with phenol as the main product. However,
methylphenol was only observed with the ReOx(1.8)/SiO2-Al2O3 and
ReOx(1.8)/Al2O3 catalysts. Since this product is formed via a methyla-
tion step which is promoted in the presence of acid sites, this result is
consistent with TPD-NH3 results which shows these two catalysts to be
the most acidic catalysts, and agrees with the findings of Bui et. al
[15,51]. It should be pointed out that unlike studies over other active
phases, such as CoMo sulfide [51], metallic Ni [53], metal nitrides [19],
among others, which have shown a strong dependence of products
distribution on the surface properties of the support, ReOx species
appear to have a stronger influence on the reaction pathway than the
support. This result further highlights the unique abilities of ReOx
8

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 11. Yield of products as a function of guaiacol conversion over a) ReOx(0.6)/SiO2, b) ReOx(1.3)/SiO2, c) ReOx(1.8)/SiO2, d) ReOx(2.2)/SiO2and e) ReOx(2.7)/SiO2 catalysts.
catalysts in HDO reactions.
(Al, Si + Al) atomic ratio. On the other hand, TPR results showed that
Re oxides were reduced at a lower temperature on the SiO2 support
than over Al2O3 and SiO2-Al2O3 supports. If it is considered that the
active sites on ReOx species are oxygen vacancy sites created by
interaction of hydrogen with ReOx-support, according to kinetic model
proposed recently by Leiva et al. [34], then ReOx have a higher
possibility of vacancy formation on SiO2support during the initial stage
of the reaction. This interpretation is supported by XPS results which
The specific (per mass of catalyst) and intrinsic (per Re atoms)
catalytic activities for the ReOx(1.8)/support catalysts are depicted in
Fig. 10. Fig. 10a and 10b show ReOx(1.8)/SiO2to be the most active
catalyst, followed by ReOx(1.8)/SiO2-Al2O3 and ReOx(1.8)/Al2O3
catalysts. This trend cannot entirely be related to the relative Re
dispersion obtained from XPS results. In fact, XPS results show that
the most active catalyst, ReOx(1.8)/SiO2, displayed the lowest Re/Si
9

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
The results of initial rate over ReOx(x)/SiO2 catalysts are depicted
in Fig. 13: the rate increases with increasing Re content up to 1.8 atoms
of Re nm−2, and then decreases. The initial lineal increase of the rate of
guaiacol conversion reflects an increase in the amount well-dispersed
Re oxide species, resulting in an increase in the formation of oxygen
vacancies (active sites). This is consistent with XPS results which
showed an increase in Re dispersion up to a loading of 1.8 atoms of
Re per nm2of support. Along the same line, the formation of Re
aggregates as evidence by XPS, UV–vis-DR and Raman results led to
decrease in the initial rate at higher loadings.
4. Conclusion
Re oxide catalysts supported on SiO2, Al2O3and SiO2-Al2O3 were
synthesized to study the support effect on catalytic activity and
selectivity for the conversion of guaiacol (as a lignin model compound).
The activity decreased in the order: SiO2 > SiO2-Al2O3 > Al2O3. TPR
results showed that Re oxides are reduced at a lower temperature on
SiO2 support than over Al2O3 and SiO2-Al2O3 supports, suggesting that
the binding energy of Re-O-support on SiO2 is lower, which translates to
a more readily formation of oxygen vacancy sites (active sites asso-
ciated with metal oxide defects created during course of the reaction).
This interpretation is also supported by XPS results which showed the
presence of two ReOx species on the SiO2-supported catalyst, unlike the
more difficult-to-reduce Al2O3 and SiO2-Al2O3-supported catalysts in
which only one ReOx species was observed. Meanwhile, the higher
activity of the SiO2-Al2O3-supported catalyst over Al2O3-supported
catalyst is related to its higher Re dispersion as suggested by XPS
results. For the effect of metal loading of ReOx(x)/SiO2catalysts, the
activity trend correlates with Re dispersion obtained from XPS, where-
by the highest catalytic activity was obtained with 1.8 atoms of Re per
nm2 of support (14% of Re) which exhibited the highest dispersion
(formation of new active site). On the other hand, the loss of catalytic
activity at higher rhenium contents is due to Re metal beginning to
form aggregates, leading to loss of active sites in agreement with XPS,
UV-vis-DR and Raman results.
Fig. 12. Products distribution on ReOx(x)/SiO2 catalysts calculated at 20% of guaiacol
conversion.
Acknowledgments
The authors thank CONICYT-Chile for FONDECYT N° 1140528,
PFB-27, ECOS-CONICYT N°C13E02, FONDEQUID EQM 120096 grants
and Red Doctoral REDOC.CTA, MINEDUC project UCO1202 at the
University of Concepción. K. Leiva is indebted to CONICYT for doctoral
grant.
Fig. 13. Specific rates over ReOx(x)/SiO2 catalysts.
show the presence of two ReOx species on the surface of the ReOx
(1.8)/SiO2 catalyst, while the other two catalysts contain one ReOx
species. This suggests that the ReOx(1.8)/SiO2 catalyst is more easily
reducible, in agreement with TPR results, leading to an increase in the
number of vacancy sites on this catalyst, and consequently the highest
activity. In the case of ReOx(1.8)/SiO2-Al2O3 and ReOx(1.8)/Al2O3
catalysts, in which one ReOx species was present, the higher activity of
the former is attributed to its higher relative Re dispersion. Therefore,
the data reveals that while the presence of oxygen vacancy sites is the
dominant factor in the HDO of guaiacol over ReOx/supports catalysts,
Re dispersion cannot be ignored when the catalysts exhibit similar
reduction tendencies.
Fig. 11 shows the yield of products as a function of guaiacol
conversion over ReOx(x)/SiO2 catalysts. The main products on all the
ReOx(x)/SiO2 catalysts were phenol, BTX (mainly benzene and toluene
and xylene in trace) and cyclohexane. Thus, these results follow the
previously-described reaction network for guaiacol transformation in
Fig. 8. The products distribution calculated at 20% of guaiacol
conversion over the ReOx(x)/SiO2 catalysts are shown in Fig. 12 and
reveals nearly identical behavior over all the catalysts, indicating that
the active sites present on these catalysts are similar. Furthermore, the
formation of Re aggregates did not affect the products distribution.
References
[1] Y. Nakagawa, Y. Shinmi, S. Koso, K. Tomishige, J. Catal. 272 (2010) 191–194.
[2] S. Koso, Y. Nakagawa, K. Tomishige, J. Catal. 280 (2011) 221–229.
[3] S. Koso, H. Watanabe, K. Okumura, Y. Nakagawa, K. Tomishige, Appl. Catal B
111–112 (2012) 27–37.
[4] G. Busca, Mater. Chem. Phys. 19 (1988) 157–165.
[5] M. Khurshid, M.A. Al-Daous, H. Hattori, S.S. Al-Khattaf, Appl. Catal A 362 (2009)
75–81.
[6] M. Khurshid, S.S. Al-Khattaf, Appl. Catal A 368 (2009) 56–64.
[7] Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Appl. Catal B
105 (2011) 117–127.
[8] S. Koso, N. Ueda, Y. Shinmi, K. Okumura, T. Kizuka, K. Tomishige, J. Catal. 267
(2009) 89–92.
[9] D.E. Resasco, S. Crossley, AIChE J. 55 (2009) 1082–1089.
[10] E. Laurent, B. Delmon, Appl. Catal A 109 (1994) 97–115.
[11] P. Grange, E. Laurent, R. Maggi, A. Centeno, B. Delmon, Catal. Today 29 (1996)
297–301.
[12] S.D. Robertson, B.D. McNicol, J.H. De Baas, S.C. Kloet, J.W. Jenkins, J. Catal. 37
(1975) 424–431.
[13] C. Sepúlveda, R. García, P. Reyes, I.T. Ghampson, J.L.G. Fierro, D. Laurenti,
M. Vrinat, N. Escalona, Appl. Catal A 475 (2014) 427–437.
[14] K. Leiva, C. Sepulveda, R. Garcia, J.L.G. Fierro, N. Escalona, Catal. Commun. 53
(2014) 33–37.
[15] V.N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Appl. Catal B 101 (2011) 239–245.
[16] M. Ferrari, R. Maggi, B. Delmon, P. Grange, J. Catal. 198 (2001) 47–55.
10

K. Leiva et al.
Catalysis Today xxx (xxxx) xxx–xxx
[17] C. Sepulveda, N. Escalona, R. Garcia, D. Laurenti, M. Vrinat, Catal. Today 195
(2012) 101–105.
C. Geantet, J.L.G. Fierro, I.T. Ghampson, N. Escalona, Appl. Catal. A 490 (2015)
71–79.
[18] I.T. Ghampson, C. Sepulveda, R. Garcia, L.R. Radovic, J.L.G. Fierro, W.J. DeSisto,
N. Escalona, Appl. Catal A 439–440 (2012) 111–124.
[19] C. Ghampson, R. Sepulveda, J.L. Garcia, N. Garcia Fierro, W.J. Escalona, Appl.
Catal. A 435–436 (2012) 51–60.
[35] P. Arnoldy, E.M. van Oers, O.S.L. Bruinsma, V.H.J. de Beer, J.A. Moulijn, J. Catal.
93 (1985) 231–245.
[36] B. Mitra, X. Gao, I.E. Wachs, A.M. Hirt, G. Deo, Phys. Chem. Chem. Phys. 3 (2001)
1144–1152.
[20] I.T. Ghampson, C. Sepulveda, R. Garcia, B.G. Frederick, M.C. Wheeler, N. Escalona,
[37] R. Nava, B. Pawelec, P. Castaño, M.C. Álvarez-Galván, C.V. Loricera, J.L.G. Fierro,
W.J. DeSisto, Appl. Catal. A 413–414 (2012) 78–84.
Appl. Catal. B 92 (2009) 154–167.
[21] R.N. Olcese, M. Bettahar, D. Petitjean, B. Malaman, F. Giovanella, A. Dufour, Appl.
Catal. B 115–116 (2012) 63–73.
[38] A. Naor, N. Eliaz, L. Burstein, E. Gileadi, Electrochem. Solid State Lett. 13 (2010)
D91–D93.
[22] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige, Green
[39] W.T. Tysoe, F. Zaera, G.A. Somorjai, Surf. Sci. 200 (1988) 1–14.
[40] L.L. Murrell, N.C. Dispenziere, R.T.K. Baker, J.J. Chludzinski, Strong metal-support,
Am. Interact. Chem. Soc. (1986) 195–199.
Chem. 9 (2007) 582–588.
[23] Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Appl. Catal. B 94 (2010)
318–326.
[41] J.M. Herrmann, Strong metal-support interactions, Am. Interact. Chem. Soc. (1986)
[24] A. Gutierrez, R.K. Kaila, M.L. Honkela, R. Slioor, A.O.I. Krause, Catal. Today 147
(2009) 239–246.
[25] A. Shimao, S. Koso, N. Ueda, Y. Shinmi, I. Furikado, K. Tomishige, Chem. Lett. 38
(2009) 540–541.
200–211.
[42] M. Spichiger-Ulmann, A. Monnier, M. Koudelka, J. Augustynski, Strong metal-
support interactions, Am. Interact. Chem. Soc. (1986) 212–227.
[43] K. Fang, J. Ren, Y. Sun, J. Mol. Catal. A: Chem. 229 (2005) 51–58.
[44] B, C. Peng, S. Zhao, S. Kasakov, Foraita, J.A. Lercher, Chem. – A Eur. J. 19 (2013)
4732–4741.
[26] M. Shetty, K. Murugappan, T. Prasomsri, W.H. Green, Y. Román-Leshkov, J. Catal.
331 (2015) 86–97.
[27] N. Escalona, M. Vrinat, D. Laurenti, F.J. Gil Llambías, Appl. Catal. A 322 (2007)
113–120.
[45] A.B. Dongil, I.T. Ghampson, R. Garcia, J.L.G. Fierro, N. Escalona, RSC Adv. 6 (2016)
2611–2623.
[28] P.E. Ruiz, K. Leiva, R. Garcia, P. Reyes, J.L.G. Fierro, N. Escalona, Appl. Catal. A 384
[46] N. Koike, S. Hosokai, A. Takagaki, S. Nishimura, R. Kikuchi, K. Ebitani, Y. Suzuki,
S.T. Oyama, J . Catal. 333 (2016) 115–126.
(2010) 78–83.
[29] G. Lagos, R. García, A.L. Agudo, M. Yates, J.L.G. Fierro, F.J. Gil-Llambías,
N. Escalona, Appl. Catal. A 358 (2009) 26–31.
[30] C. Sepulveda, V. Bellière, D. Laurenti, N. Escalona, R. García, C. Geantet, M. Vrinat,
Appl. Catal. A 393 (2011) 288–293.
[47] A. Cho, H. Kim, A. Iino, A. Takagaki, S. Ted Oyama, J. Catal. 318 (2014) 151–161.
[48] S. Ted Oyama, H. Zhao, H.-J. Freund, K. Asakura, R. Włodarczyk, M. Sierka, J.
Catal. 285 (2012) 1–5.
[49] C. Cristiani, P. Forzatti, G. Busca, J. Catal. 116 (1989) 586–589.
[50] E. Laurent, B. Delmon, Appl. Catal. A 109 (1994) 77–96.
[51] V.N. Bui, D. Laurenti, P. Delichère, C. Geantet, Appl. Catal. B 101 (2011) 246–255.
[52] T. Nimmanwudipong, R.C. Runnebaum, D.E. Block, B.C. Gates, Energy Fuels 25
(2011) 3417–3427.
[31] C. Sepulveda, R. Garcia, N. Escalona, D. Laurenti, L. Massin, M. Vrinat, Catal. Lett.
141 (2011) 987–995.
[32] K. Leiva, C. Sepúlveda, R. García, J.L.G. Fierro, G. Aguila, P. Baeza, M. Villarroel,
N. Escalona, Appl. Catal. A 467 (2013) 568–574.
[33] N. Escalona, J. Ojeda, R. Cid, G. Alves, A. López Agudo, J.L.G. Fierro,
[53] Y. Yang, C. Ochoa-Hernández, V.A. de la Peña O'Shea, P. Pizarro, J.M. Coronado,
́
F.J.G. Llambıas, Appl. Catal. A 234 (2002) 45–54.
D.P. Serrano, Appl. Catal. B 145 (2014) 91–100.
[34] K. Leiva, N. Martinez, C. Sepulveda, R. García, C.A. Jiménez, D. Laurenti, M. Vrinat,
11