Kinetic study of the conversion of 2-methoxyphenol over supported Re catalysts: Sulfide and oxide state

Article abstract of DOI:10.1016/j.apcata.2015.08.010

The kinetics of guaiacol conversion over ReS₂/SiO₂ and ReOx/SiO₂ catalysts in a batch reactor at 300°C and 5 MPa hydrogen pressure was studied. The ReOx/SiO₂ catalyst was more active and selective towards deoxygenated products (mainly cyclohexane), while the ReS₂/SiO₂ catalyst was more selective towards the formation of phenol. Langmuir-Hinshelwood kinetic models were developed in order to identify changes in the reaction mechanism with the ReOx/SiO₂ and ReS₂/SiO₂ catalysts. The results for the guaiacol conversion over ReS₂/SiO₂ catalyst were fitted to a model that assumes the presence of two different kinds of active sites and that hydrogen adsorption is generated through heterolytic dissociation. The conversion of guaiacol over the ReOx/SiO₂ catalyst was fitted to a model that assumes hydrogen adsorption via homolytic dissociation and a single catalytic site.

Full text of DOI:10.1016/j.apcata.2015.08.010

Accepted Manuscript
Title: Kinetic study of the conversion of 2-methoxyphenol
over supported Re catalysts: sulfide and oxide state
´
Author: K. Leiva C. Sepulveda R. Garcıa D. Laurenti M.
Vrinat C. Geantet N. Escalona
PII:
S0926-860X(15)30105-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2015.08.010
APCATA 15504
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
28-5-2015
4-8-2015
5-8-2015
´
Please cite this article as: K.Leiva, C.Sepulveda, R.Garcia, D.Laurenti, M.Vrinat,
C.Geantet, N.Escalona, Kinetic study of the conversion of 2-methoxyphenol over
supported Re catalysts: sulfide and oxide state, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2015.08.010
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Kinetic study of the conversion of 2-methoxyphenol over supported Re catalysts:
sulfide and oxide state
K. Leivaa, C. Sepulvedaa, R. Garcíaa, D. Laurentib, M. Vrinatb, C. Geantetb, N. Escalonaa*
nescalona@udec.cl escalona.nestor@gmail.com
aUniversidad de Concepción, Facultad de Ciencias Químicas, Casilla160C, Concepción,
Chile.
bIRCELYON, UMR 5256 CNRS, Université Lyon 1, 2 avenue A. Einstein, 69626
Villeurbanne cedex, France.
*Corresponding author. Dr. Néstor Escalona B., Tel./Fax: (56-41) 2207236/ 2245374.
1

Graphical Abstract
2

Highlights
 Kinetics of guaiacol conversion over ReS2/SiO2 and ReOx/SiO2 catalysts were
studied
 Langmuir–Hinshelwood kinetic models were developed for these catalysts.
 Kinetic results show two kinds of active sites on the ReS2/SiO2 catalyst.
 ReOx/SiO2 catalyst was fitted to a model with one kind of active site.
3

Abstract
The kinetics of guaiacol conversion over ReS2/SiO2 and ReOx/SiO2 catalysts in a batch
reactor at 300 °C and 5 MPa hydrogen pressure was studied. The ReOx/SiO2 catalyst was
more active and selective towards deoxygenated products (mainly cyclohexane), while the
ReS2/SiO2 catalyst was more selective towards the formation of phenol. Langmuir–
Hinshelwood kinetic models were developed in order to identify changes in the reaction
mechanism with the ReOx/SiO2 and ReS2/SiO2 catalysts. The results for the guaiacol
conversion over ReS2/SiO2 catalyst were fitted to a model that assumes the presence of two
different kinds of active sites and that hydrogen adsorption is generated through heterolytic
dissociation. The conversion of guaiacol over the ReOx/SiO2 catalyst was fitted to a model
that assumes hydrogen adsorption via homolytic dissociation and a single catalytic site.
Keywords: Kinetic; Hydrodeoxygenation; 2-methoxyphenol; ReS2; ReOx; Langmuir–
Hinshelwood.
4

1. - Introduction
Biomass has been considerably investigated as a new promising source for fuels and
chemicals due to environmental concerns and the need to decrease the overdependence on
fossil fuels. Pyrolysis of lignocellulosic biomass leads to compounds which have the
potential to constitute an alternative source of fuel (bio-oil) and value-added chemicals [1-
4]. Valorization of biomass is based on the production of chemicals like phenols or
aromatics (BTX: benzene, toluene, xylene) from the lignin polyphenolic structure [5].
However, the pyrolytic bio-oils need to be upgraded before direct application as fuel due to
the high content of oxygenated compound. Besides, the use of lignocellulosic compounds
as a precursor of carbonaceous materials has gained interest due to its low cost and high
availability [5-8]. Catalytic hydrodeoxygenation (HDO) has been demonstrated as an
interesting and effective route to improve the properties of biomass-derived oil. The main
oxygenated compounds in bio-oil originate from the lignin component of biomass. For that
reason, guaiacol (2-methoxyphenol) has been studied as a lignin-derived model compound
in order to elucidate the reaction pathway and the efficiency of the HDO process [9-12].
Conventional hydrotreating catalysts such as CoMoS, NiMoS and MoS2 have been found to
be highly active for the conversion of guaiacol. However, sulfide catalysts generally suffer
from continuous deactivation whose origin has been linked to the formation of coke or
partial re-oxidation of the sulfide phase [13-15]. Other promising catalysts are the Ni-based
catalysts studied by Bykova et al. [9]. These researchers found that the most active catalysts
among their library of catalysts in the HDO of guaiacol were Ni-based catalysts prepared
by a sol-gel method and stabilized with SiO2 and ZrO2 supports. According to TPR, XRD
and HRTEM analyses, the high activity of these catalysts were due to the high nickel
loading and the high specific area of active component as a result of the formation of nickel
oxide-silicate species. These studies demonstrated the importance of exploring other active
phases beside metals, oxides and conventional sulfides to improve the catalytic upgrading
process. Among unconventional supported metal sulfide catalysts, ReS2 catalysts have been
demonstrated to display a higher catalytic activity than a commercial sulfided NiMo/Al2O3
catalyst for the conversion of guaiacol [16, 17]. Recently, Leiva et al. [18] found that
ReOx/SiO2 catalyst displayed a higher activity for guaiacol conversion and a higher
selectivity towards deoxygenated compounds in comparison with ReS2/SiO2 and
commercial NiMo/Al2O3 catalysts.
5

On the other hand, reaction networks for guaiacol conversion have been widely
reported for different catalysts [13, 15, 19-21]; however, a fundamental understanding of
the reaction mechanisms is still lacking. Studying the kinetics and reaction mechanism of
furfural hydrogenation over Cu/SiO2 catalysts, Sitthisa et al. [22] found that the
transformation of furfural over these solids followed a Langmuir–Hinshelwood model
assuming: (1) dissociative adsorption of hydrogen, (2) that all adsorption sites are
equivalent and independent of coverage, and (3) that the surface reaction is the rate-
determining step. On the other hand, when Badawi et al. [23] studied the hydrogenation of
olefins over sulfide catalysts and used the Langmuir–Hinshelwood theory, they obtained a
pseudo Eley–Rideal model instead, after some approximations in agreement with the
proposal by Sitthisa et al. [22]. In addition, Badawi et al. [23] proposed, by theoretical
studies, that the adsorption of olefin is only possible on metal anionic vacancy, while the
reactant H2 is heterolytically dissociated on the sulfide catalyst. The authors then concluded
that the hydrogenation reaction occurred only on the sulfide phase, in agreement with the
mechanism previously proposed by Olguin et al. [24].
In our previous study, we compared ReOx/SiO2 and ReS2/SiO2 catalysts and
correlated the difference in activity to results obtained from XPS. The pre- and post-
reaction XPS results showed an increase in the amount of surface ReOx sites, consequently
leading to comparatively higher activity. In contrast, the lower activity of ReS2 catalysts
was attributed to decrease in dispersion caused by sintering. However, a greater
understanding of the kinetics involved in the conversion of guaiacol is still lacking.
Therefore, the objective of this study is to compare the behaviors of ReS2/SiO2 and
ReOx/SiO2 catalysts during guaiacol conversion. In contrasts to sulfide catalysts which
have been extensively studied in this kind of reactions [13-17, 19, 23 24-30], oxides
catalysts and especially Re oxides have rarely been investigated. The results obtained for
ReS2/SiO2 and ReOx/SiO2 catalysts in this study indicated large differences in activity and
selectivity. To explain those results, the reaction was modeled using a Langmuir-
Hinshelwood-type theory in order to identify changes in the reaction mechanism for
ReS2/SiO2 and ReOx/SiO2 catalysts.
6

2. - Experimental
2.1 Catalyst preparation
The catalysts were prepared by incipient wetness impregnation of an aqueous
solution of NH4ReO4 (Aldrich, 99%) on SiO2 support (Grace Davison 432). A Re nominal
loading of 2.7 atoms of Re per nm2 of support was the target. The impregnated catalysts
were left for maturation at room temperature for 24 h, dried at 393 K for 12 h and then
calcined at 573 K for 0.5 h. The calcined sample was denoted as ReOx/SiO2 catalyst; in
addition, a small portion of the calcined sample was sulfided at 673 K for 4 h under 15 %
H2S/N2 mixture and denoted as ReS2/SiO2 catalyst.
These catalysts were characterized by N2 adsorption, X-ray photoelectron
spectroscopy (XPS), Temperature programmed reduction (TPR) and Transmission electron
microscopy (TEM) in our previous studies [18].
2.2. Catalytic tests
HDO of guaiacol was carried out in a 0.25L batch reactor equipped with a hollow-
shaft 6-bladed mechanically driven turbine with four baffles on the wall of the reactor to
prevent vortex formation (model Parr 4841). 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, was introduced into the reactor. Approximately 0.200 g of catalyst with a
particle size of 100 mesh was added. The system was closed and, to avoid any air
contamination, N2 was bubbled through the solution for 10 min. Still under N2 the reactor
was heated to the reaction temperature of 573 K under stirring. The pressure was adjusted
to 5 MPa by H2 introduction into the reactor and kept constant during the course of the
experiment with a stirring rate of 645 rpm. Under these experimental conditions, the
reaction was not controlled by diffusional effects and the reaction was under kinetic control
[25]. Liquid samples were taken periodically during the reaction. The reaction products
were identified by gaseous chromatography (Perkin Elmer - Clarus 680) coupled to GCMS-
SQ8T equipped and quantified by gas chromatograph (Perkin-Elmer - Clarus 400) using a
Flame Ionization Detector (FID) and a CP-Sil 5 column (Agilent, 30 m x 0.53 mm x 1.0 μm
film thickness). For the kinetic study, the initial concentration of guaiacol was varied from
0.09 to 0.45 mol L-1. Catalytic activity on guaiacol conversion was expressed in terms of
7

the initial reaction rate calculated from the initial slope of the conversion vs. time plot,
according to the following equation:
[bn]
r 
Eq. (1)
s
m
where rs is the specific rate expressed in mol g−1s−1, b is the initial slope of the conversion
of guaiacol vs. time (τ/t), n is the initial number moles of guaiacol, and m is the mass of
catalyst (g). The deoxygenation (HDO) rate was calculated from guaiacol conversion to O-
free compounds (benzene, cyclohexane, cyclohexene and analogs). On the other hand, the
selectivities (%) were determined at 20% and 60% of guaiacol conversion, according to Eq.
(2).
Xi
S% 
100
Eq. (2)
XT
where Xi is the percentage of product i formed and XT is the guaiacol conversion.
3. - Result and discussion
3.1 Catalytic activity
Figure 1 shows the total and hydrodeoxygenation initial rates of the conversion of
guaiacol over the ReS2/SiO2 and ReOx/SiO2 catalysts. As can be observed, the ReOx/SiO2
catalyst was the more active catalyst, in agreement with our previous results [18].
Moreover, in our previous work, it was found that the ReOx/SiO2 catalyst exhibited higher
total and hydrodeoxygenation initial rates of 2.9 and 4.8 times, respectively, than a
conventional NiMoS/Al2O3 catalyst, demonstrating the potential of ReOx/SiO2 catalyst for
the valorization of lignin-derived compounds.
Figure 2 shows the conversion and yield of products obtained for the transformation
of guaiacol over the ReS2/SiO2 and ReOx/SiO2 catalysts as a function of time. These results
show a clear difference between the catalysts. The main products with the ReS2/SiO2
catalyst were phenol, catechol and cyclohexane (Fig. 2(a)). Catechol, detected in small
quantities, was an intermediate product. Other compounds such as anisole, methylphenol,
cyclohexanol, cyclohexane and methanol were also detected in minor quantities, while-
methoxycyclohexanone, hexane, cyclohexene, methylcyclohexane and BTX (benzene,
toluene and xylene) were detected in trace amounts. With the ReOx/SiO2 catalyst (Fig.
8

2(b)), phenol, anisole and cyclohexanol were the main intermediate products which led to
the production of cyclohexane and BTX at higher conversions. Methylcyclohexane,
methoxycyclohexanone and methanol were the minor products obtained, while
methylphenol, cyclohexene, methylcyclohexanone and hexane were only detected as traces.
In addition to these observations, the ReOx/SiO2 catalyst achieved a maximum conversion
after 2 hours (~80%) and then remained relatively constant over the course of the reaction.
However, after less than 2 hours of reaction the concentrations of phenol, anisole and
cyclohexanol began to decrease while guaiacol conversion was about 60%. This is due to
the competitive adsorption occurring between guaiacol and these products on the catalytic
site.
Figure 3 shows the network of guaiacol conversion in agreement with Laurent and
Delmon [13, 26], Bui et al. [15] and Leiva et al. [18]. On the basis of this network, guaiacol
is mainly converted to phenol via two pathways: a demethoxylation (DMO) pathway
without catechol as an intermediate and a demethylation (DME) route to form catechol,
which subsequently undergoes hydrogenolysis to form phenol. However, over the
ReOx/SiO2 catalyst phenol was formed at low conversion (without going through the DME
route) and rapidly transformed to products, including BTX (mainly benzene) and
cyclohexane. The absence of catechol in the product stream could be due to its rapid
transformation to phenol.
Table 1 shows the products distribution calculated at 20 and 60% of guaiacol
conversion: both catalysts were highly effective in converting guaiacol into phenol (60-
75% selectivity at 20% of guaiacol conversion). Also, at 20% of guaiacol conversion, the
ReOx/SiO2 catalyst displayed higher amounts of deoxygenated products such as benzene
and cyclohexane than the ReS2/SiO2 catalyst. Furthermore, catechol was detected in higher
amount with the ReS2/SiO2 catalyst. These results suggest that ReS2/SiO2 and ReOx/SiO2
catalysts have different active sites and acidic properties. In order to clarify these
differences, the products distribution was calculated at 60% of guaiacol conversion, where
the conversion was still linear with the time. There were clear differences observed at this
higher conversion as can be seen in Table 1. The ReS2/SiO2 catalyst was highly selective
towards phenol formation (75%), while the ReOx/SiO2 catalyst presented a higher
distribution of products such as BTX (17.5%), cyclohexane (28%) and phenol (34%). This
9

further illustrates that the transformation of guaiacol over ReS2/SiO2 and ReOx/SiO2
catalysts clearly follows different kind of mechanisms. Besides, the differences observed
for the selectivities confirm that both ReS2/SiO2 and ReOx/SiO2 systems have different
active sites which were not modified during the course of the reaction under our operating
conditions.
3.2 Effect of guaiacol concentration
With the objective to clarify and understand the previous reactivity results, the
effect of guaiacol concentration over ReS2/SiO2 and ReOx/SiO2 catalysts was studied. The
results obtained are shown in Fig. 4 which illustrates the variation of the initial rate as a
function of guaiacol concentration over ReS2/SiO2 and ReOx/SiO2 catalysts. There was a
fairly linear increase in the initial rate with the concentration of guaiacol over the
ReS2/SiO2 catalysts. On the contrary, the initial rates over the ReOx/SiO2 catalyst increased
up to guaiacol concentration of 0.183 mol L-1 and then decreased. These results confirm
that the reaction mechanisms involved over the ReS2/SiO2 and ReOx/SiO2 catalysts are
different.
3.3 Mechanism and rate equation
In order to get more information on the reaction mechanism involved in the
conversion of guaiacol over ReS2/SiO2 and ReOx/SiO2 catalysts, two kinetic models based
on Langmuir-Hinshelwood concept were proposed.
3.3.1 Mechanism and rate equation over ReS2/SiO2
Various types of adsorption of dihydrogen are considered for hydrogenation
reactions are in the literature, such as non-dissociative or homolytic dissociative adsorption
which are expected to occur on metal catalysts [23]. In general, kinetic models for
hydrotreating reactions differ by: (1) the use of one or two types of sites, and (2) the
molecular adsorption or the homolytic dissociation of hydrogen [30]. However, over sulfide
catalysts, adsorption mechanisms are thought to be different [23, 30]. The mechanism over
10

sulfide catalysts has been widely studied for hydrogenation and hydrodesulfurization
reactions [24, 27-29]. The kinetic models for these kinds of reactions as previously
mentioned are often based on the molecular adsorption of H2 and H2S and sometimes on
their homolytic dissociation [23, 24, 27]. However, it has been demonstrated in several
studies that the reaction mechanism over sulfide catalysts involved the existence of two
different catalytic sites and a heterolytic dissociative adsorption of hydrogen [22-24, 28 and
30]. These active sites correspond to a coordinatively unsaturated metal ion and a strongly
bonded stable sulfur ion under the experimental conditions studied [23, 24, 30].
On the basis of these studies, we propose a mechanism for guaiacol conversion over
ReS2/SiO2 in accordance with the reaction network given in Fig. 3. This mechanism
designated as Model 1 involves two active sites denoted as M= metal ion vacancy and X2--
M= Sulfur stable ion. The set of elementary steps considered in Model 1 was identical to
the one previously developed by Kasztelan et al. [30] and Olguin et al. [24]. In addition,
this model was expressed in terms of the initial reaction rate, where the principal reaction
products over ReS2/SiO2 catalysts were phenol and methanol.
The set of elementary steps considered in Model 1 was:
Adsorption of GUA: GUA + M  GUA-M
Equilibrium KGUA (3)
H2 dissociative adsorption: H2 + M + X2--M  H--M + H+-X2—M Equilibrium KH2 (4)
Addition of H-: GUA-M + H--M  GUAH--M + M
Equilibrium K3 (5)
Rate constant K4 (6)
Addition of H+: GUAH--M + H+-X2--M  Ph + MeOH + 2M
The rate determining step was considered to be the second catalyst’s H+ attack on
the oxygen of the methoxy group according to numerous researches [24, 30-32]. The
adsorption and desorption of reactants (H2 and guaiacol) are much faster than the surface
reaction step and are assumed to be in equilibrium. The number of active sites is assumed
to be constant, while desorption of the products and the adsorption-desorption of the
solvent were not considered [23]. The development of the kinetic law assumed a weak H2
11

adsorption on metal active sites [24], and hence the coverage of proton and hydride species
on the surface of the catalyst is lower than the coverage of guaiacol.
Thus, the reaction rate for Model 1 is given by the expression:
rGUA  k4CGUAH MCH X2 M (7)
-
+
2-
-M
From equation (5) and (4), we can deduce the concentration of CGUAH -M and CH -X
,
respectively.
K3 CH  M CGUAM
CGUAH M


(8)
CM
K
CH C 2 M CM
H2
2
X
CH X
(9)
2 M
CH
 M
Substituting equations (8) and (9) into (7):
rGUA  k4KH K3CH CX 2 M CGUAM (10)
2
2
From equation (3) the concentration of CGUA-M can be deduced and substituted into equation
(10):
rGUA  k4K3KGUAKH CGUACH CM CX (11)
2 M
2
2
For the two kinds of sites, two equations for the conservation of sites can be considered:
(CM )o  CM  CH M  CGUAM  CGUAH M (12)
(CX 2 M )o  CX
 CH X
(13)
2 M
2 M
In absence of H2S the surface coverage by proton and hydride species is very low [23].
Then,
CH
 CGUAM  CGUAH
 CM and CH 
 CX
 M
 M
X
2 M
2 M
Therefore, equation (13) becomes,
12

(CX 2 M )0  CX
(14)
2 M
Finally, and considering the initial rate and the weak adsorption of H2 on the metal active
site, the coverage of proton and hydride species on the surface of the catalysts are thus
-
lower than the coverage of guaiacol (CGUA-M + CM > CGUAH -M); therefore, equation (12)
becomes:
(CM )0  CM  CGUAM (15)
From equation (1):
CGUAM  KGUACGUACM (16)
Substituting equation (16) in (15):
(CM )0  CM  KGUACGUACM (17)
Finally:
(CM )0
CM 
(18)
(1  KGUACGUA
)
Substituting equation (14) and (18) into (11):
k4K3KGUAKH CGUACH (CX 2 M )0 (CM )0
2
2
rGUA

(19)
(1  KGUACGUA
)
Since k4K3KH (CX 2 M )(CM )0 are constant, equation (19) can be expressed as:
2
kGUAKGUACGUACH
2
rGUA

(20)
(1  KGUACGUA
)
where: kGUA = k4K3KH (CX 2 M )(CM )0
2
Linearizing equation (20):
1
1
1


(21)
rGUA kGUA
C
kGUAKGUAC CGUA
H2
H2
13

3.3.1.1 Corroboration of the kinetic model over ReS2/SiO2
In order to corroborate the kinetic law for the conversion of guaiacol over the
ReS2/SiO2 catalyst, the initial rate values from linearized equation, as a function of the
initial guaiacol concentration, were plotted in Fig. 5, according to Model 1 (equation 21)).
According to Model 1, the results must be adjusted to a straight line with positive a
slope. Figure 5 clearly shows that the conversion of guaiacol over the ReS2/SiO2 catalyst
can fit the equation for Model 1 which follows a linear behavior and a correlation of
0.99645, similar to the classical Langmuir–Hinshelwood kinetic model. This result
confirms the presence of two different active sites on the sulfide catalysts as has been
previously proposed by Kasztelan et al. [30], Badawi et al. [23] and Olguin et al. [24]. For
the ReOx/SiO2 catalyst, the catalytic activity results did not fit the equation for Model 1
(inset of Fig. 5), suggesting that the ReOx/SiO2 catalyst followed another kind of
mechanism for the conversion of guaiacol.
3.3.2 Mechanism and rate equation over ReOx/SiO2
As previously stated, kinetic models for hydrotreating reactions involves either one
or two catalytic sites as well as either molecular adsorption of hydrogen or homolytic
dissociation of hydrogen [30]. Dissociative adsorption of hydrogen occurs relatively easily
on many metals. On metal oxides, however, dissociation of the strong H-H bond seems
only to occur on the most reactive surfaces and is promoted by the presence of surface
defects. Homolytic dissociative adsorption results in two hydroxyls or, less commonly, two
metal-hydrogen species. Heterolytic adsorption is also possible; however, the metal-
hydrides formed are not very stable species, and so the more general route is the formation
of hydroxyl groups [33, 34]. Unlike sulfide catalysts, the oxide catalysts have been less
studied for these kinds of reaction. As discussed earlier, the Re oxide catalyst follows a
kinetic model which differs from the Re sulfide catalyst for the reaction under study.
Therefore, the mechanism proposed for this catalyst is designated as Model 2 and assumes
the presence of one kind of active sites designated as M and hydrogen adsorption by
homolytic dissociation. This active site can be associated with the metal oxide defects
created by hydrogen attack, generating H2O removal on the ReOx/SiO2 catalyst [18].
14

Furthermore, in our previous work [18], we showed by XPS analysis of the post-reaction
ReOx/SiO2 catalyst that in-situ partial reduction of around 10% of the oxide phase occurred
o
after 4 hours of reaction under 300 C and 50 bar H2 pressure. However, the catalytic
models were determined at initial reaction rates where the catalyst is assumed to be in its
oxidic state.
The set of elementary steps considered in Model 2 were:
Adsorption of GUA: GUA + M  GUA-M
Equilibrium KGUA
Equilibrium KH2
Equilibrium K3
Rate constant K4
(22)
(23)
(24)
(25)
H2 Dissociative adsorption: H2 + 2M  2H*-M
Addition of H*: GUA-M + H*-M  GUAH*-M + M
Addition of H*: GUAH*-M + H*-M  Ph + MeOH + 2M
The rate determining step of the reaction is assumed to be the addition of the second H
atom, same as for Model 1. Thus, the reaction rate for Model 2 is given by the expression:
rGUA  k4CGUAH*MCH*M (26)
From equation (23) and (24), we can deduce the concentration for CGUAH*-M and CH*-M
,
respectively.
K3CH* M CGUAM
CGUAH* M

(27)
CM
CH*M  K1H/2C1H/2CM (28)
2
2
Substituting equations (27) and (28) into (26):
rGUA  k4K3KH CGUAM CH CM (29)
2
2
15

From equation (22) the concentration of CGUA-M can be deduced and substituted into
equation (29):
CGUA  KGUACGUACM (30)
rGUA  k4K3KH KGUACGUACH CM2 (31)
2
2
From the balance of active sites, the following equations could be obtained:
(CM )0  CM K1H/ 2C1H/ 2CM  KGUACGUACM  K3KGUAK1H/ 2CGUAC1H/2CM (32)
2
2
2
2
(CM )0
1 K1H/2C1H/2  KGUACGUA  K3KGUAK1H/ 2CGUAC1H/2
CM 
(33)
2
2
2
2
Considering that the concentration of H2 is low and a weak adsorption of H2 on the metal
active site is assumed, equation (33) becomes:
(CM )0
CM 
(34)
1 KGUACGUA
Substituting equation (34) into (29):
k4K3KH KGUACGUACH (CM )02
2
2
rGUA

(35)
2
(1  KGUACGUA
)
Since k4K3KH (CM )02 are constant, equation (35) can be expressed by:
2
kGUAKGUACGUACH
2
rGUA

(36)
2
(1  KGUACGUA
)
where: kGUA = k4K3KH (CM )02
2
Linearizing equation (36), we get:
16

CGUA
rGUA
KGUACGUA
1


(37)
kGUA
kGUAKGUAC
H2
3.3.2.1 Corroboration of the kinetic model over ReOx/SiO2
To corroborate the proposed kinetic law for guaiacol conversion over the
ReOx/SiO2 catalyst, the initial rate values from the linearized equation as a function of the
initial guaiacol concentration (equation (37)) were plotted and shown in Fig. 6.
With respect to Model 2, the variation of catalytic activity with guaiacol
concentration in the reaction mixture should be able to fit a straight line with a positive
slope, which should cut the Y axis without passing through the origin. For the ReOx/SiO2
catalyst, the initial rate vs. guaiacol conversion line fits Model 2 which assumes the
presence of one kind of active site, and follows a linear behavior and a correlation of
0.94614 (as can be seen in Fig. 6). Furthermore, this model allows us to explain the trend
found in the catalytic activity when the concentration of guaiacol is increased. Unlike for
the ReS2/SiO2 catalyst, where an increase in the conversion of guaiacol was observed even
at high concentrations, the activity of the ReOx/SiO2 catalyst increased until a guaiacol
concentration of 0.183 molL-1 (Fig. 4 (b)). This is obviously due to competitive adsorption
between hydrogen and guaiacol on the active site of the catalyst, which is more prominent
at high guaiacol concentration. Accordingly, the observed increase up to 0.183 mol L-1 and
the subsequent decrease at higher concentrations was attributed to guaiacol probably
adsorbing more strongly on the ReOx/SiO2 catalyst than the hydrogen, and thus limits the
adsorption of H2 on the surface. However, at higher guaiacol concentrations, the expression
2

1 KGUACGUA

in equation (36) from Model 2 plays an important role because KGuaCGUA
becomes very large, leading to a decrease in catalytic activity. Accordingly, these results
confirm that guaiacol is strongly adsorbed on the active site on the ReOx/SiO2 catalyst,
leading to a greater surface coverage of guaiacol and a decrease in the concentration of
hydrogen adsorbed. This consequently leads to a decrease of guaiacol conversion, which
confirms the competitive adsorption between guaiacol and H2, and explains the maximum
observed activity, in agreement with Fig.4 (b). Therefore, the results clearly show that the
conversion of guaiacol over the ReOx/SiO2 catalyst fits Model 2 while the ReS2/SiO2 did
not (the inset of Fig. 6), corroborating the proposed kinetic Model 1.
17

4. Conclusion
The conversion of guaiacol over ReS2/SiO2 and ReOx/SiO2 catalysts showed many
differences in activity and selectivity. The ReOx/SiO2 catalyst displayed a higher activity
and selectivity to deoxygenated products (mainly cyclohexane), while the ReS2/SiO2
catalyst was more selective towards the formation of phenol.
The kinetic models developed showed that the ReOx/SiO2 and ReS2/SiO2 catalysts
displayed different reaction mechanisms. The kinetic results obtained for the ReS2/SiO2
catalyst fitted a model assuming the presence of two distinct catalytic sites and hydrogen
heterolytic dissociative adsorption (Model 1). On the contrary, the kinetic result for the
ReOx/SiO2 catalyst was adjusted to a kinetic model involving a single kind of active site,
and a homolytic dissociative adsorption of hydrogen (Model 2). This suggests that guaiacol
and hydrogen adsorbed on the same active site on the ReOx/SiO2 catalyst. Therefore, a
clear competition was observed between the reactants for the active site on the ReOx/SiO2
catalyst at high concentrations of guaiacol. The observed decrease of the catalytic activity
was attributed to the higher adsorption of guaiacol which is probably more strongly
adsorbed on the ReOx/SiO2 than the hydrogen.
Acknowledgments
The authors thank CONICYT-Chile for FONDECYT Nº 1140528, PFB-27, ECOS-
CONICYT NºC13E02, 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.
18

References
[1] M.V. Bykova, D.Yu. Ermakov, S.A. Khromova, A.A. Smirnov, M.Yu. Lebedev, V.А.
Yakovlev, Catalysis Today, 220–222 (2014) 21-31.
[2] Yaseen Elkasabi, Charles A. Mullen, Anna L.M.T. Pighinelli, Akwasi A. Boateng Fuel
Processing Technology, 123 (2014) 11-18.
[3] S. Boullosa-Eiras, R. Lødeng, Håkon Bergem, M. Stöcker, L. Hannevold, E. A.
Blekkan, Catalysis Today, 223 (2014) 44-53.
[4] V.A. Yakovlev, S.A. Khromova, O.V. Sherstyuk, V.O. Dundich, D.Yu. Ermakov, V.M.
Novopashina, M.Yu. Lebedev, O. Bulavchenko, V.N. Parmon, Catalysis Today, 144 (2009)
362–366.
[5] B. Joffres, C. Lorentz, M. Vidalie, D. Laurenti, A.-A. Quoineau, N. Charon, A. Daudin,
A. Quignard, C. Geantet, Applied Catalysis B: Environmental, 145 (2014) 167-176.
[6] C.N. Hamelinck, A.P.C. Faaij, H. den Uil, and H. Boerrigter, Energy, 29 (2004) 1743-
1771.
[7] L. Devi, K.J. Ptasinski, and F. Janssen, Fuel Processing Technology, 86 (2005) 707-
730.
[8] S. Czernik, A. V. Bridgwater, Energy Fuels, 18 (2004) 590-598.
[9] M.V. Bykova, D.Yu. Ermakov, V.V. Kaichev, O.A. Bulavchenko, A.A. Saraev, M.Yu.
Lebedev, V.A. Yakovlev, Applied Catalysis B: Environmental, 113– 114 (2012) 296– 307
[10] T. Mochizuki, S.-Y. Chen, M. Toba, Y. Yoshimura, Applied Catalysis B:
Environmental, 146 (2014) 237– 243.
[11] Y-K. Hong, D-W. Lee, H-J. Eom, K-Y. Lee, Applied Catalysis B: Environmental,
150– 151 (2014) 438– 445.
[12] P. Grange, E. Laurent, R. Maggi, A. Centeno, B. Delmon, Catalysis Today, 29 (1996)
297-301.
[13] E. Laurent, B. Delmon, Applied Catalysis A: General, 109 (1994) 77-96.
[14] V. N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Applied Catalysis B: Environmental,
101 (2011) 239-245.
[15] V. N. Bui, D. Laurenti, P. Delichère, and C. Geantet, Applied Catalysis B:
Environmental, 101 (2011) 246-255.
19

[16] P. E. Ruiz, K. Leiva, R. Garcia, P. Reyes, J. L. G. Fierro, and N. Escalona, Applied
Catalysis A: General, 384 (2010) 78-83.
[17] C. Sepúlveda, N. Escalona, R. García, D. Laurenti, M. Vrinat, Catalyis Today, 195
(2012) 101-105.
[18] K. Leiva, N. Martinez, C. Sepulveda, R. García, C. Jiménez, D. Laurenti, M. Vrinat, J.
L. G. Fierro, I. T. Ghampson, N. Escalona, Applied Catalysis A: General, 490 (2015) 71-
79.
[19] M. Ferrari, S. Bosmans, R. Maggi, B. Delmon, P. Grange, Catalysis Today, 65 (2001)
257-264.
[20] T. Nimmanwudipong, R. C. Runnebaum, D. E. Block, and B.C. Gates, Energy &
Fuels, 25 (2011) 3417-3427.
[21] T. Nimmanwudipong, R. C. Runnebaum, D. E. Block, B. C. Gates, Catalysis letters,
141 (2011) 779-783.
[22] S. Sitthisa, T. Sooknoi, Y. Ma, P. B. Balbuena, D. E. Resasco, Journal of Catalysis,
277 (2011) 1-13.
[23] M. Badawi, L. Vivier, D. Duprez, Journal of Molecular Catalysis A: Chemical, 320
(2010) 34–39.
[24] E. Olguin Orozco, M. Vrinat, Applied Catalysis A: General, 170 (1998) 195-206.
[25] C. Calais, M. Lacroix, C. Geantet, M. Breysse, Journal of Catalysis, 144 (1993) 160-
174.
[26] E. Laurent, B. Delmon, Applied Catalysis A: General, 109 (1994) 97-115.
[27] M. Egorova, R. Prins, Journal of Catalysis, 224 (2004) 278-287.
[28] M. L. Vrinat, Applied Catalysis, 6 (1983) 137-158.
[29] M.J Girgis, B.C. Gates, Ind. Eng. Chem. Res. 30 (1991) 2021-2058.
[30] S. Kasztelan and D. Guillaume, Ind. Eng. Chem. Res. 33 (1994) 203-210.
[31] A. Daudin, A.F. Lamic, G. Pérot, S. Brunet, P. Raybaud, C. Bouchy, Catalysis Today,
130 (2008) 221-230.
[32] A. Daudin, S. Brunet, G. Pérot, P. Raybaud, C. Bouchy, Jurnal of Catalysis, 248
(2007) 111-119.
[33] V. E. Henrich and P.A. Cox “The Surface Science of metal oxides” Cambridge
University Press. Cambridge, 1994.
20

[34] A.L. Dent and R. J. Kokes, PhysChem. 73 (1969) 3781-3790.
21

Figure Captions
Figure 1. Initial rate and HDO rate of (0.232 mol L-1) guaiacol conversion over (a)
ReS2/SiO2 and (b) ReOx/SiO2 catalysts.
Figure 2. Yield of products and (0.232 molL-1) guaiacol conversion over: (a) ReS2/SiO2
and (b) ReOx/SiO2 catalysts.
Figure 3. Reaction network for the conversion of guaiacol over ReS2/SiO2 and ReOx/SiO2
catalysts. (Adapted from Leiva et al. [18]).
Figure 4. Initial rate as a function of the initial concentration of guaiacol over (a)
ReS2/SiO2 and (b) ReOx/SiO2 catalysts.
Figure 5. Linearization of the kinetic Model 1 for ReS2/SiO2 catalyst. Inset: linearization of
the kinetic model 1 for ReOx/SiO2 catalyst.
Figure 6. Linearization of the kinetic Model 2 for ReOx/SiO2 catalyst. Inset: linearization
of the kinetic model 2 for ReS2/SiO2 catalyst.
22

Figure 1
18
15
12
9
Total rate
HDO rate
6
3
0
ReOx/SiO2
ReS2/SiO2
Active Phase
23

Figure 2
100
80
100
80
100
80
60
40
20
0
100
80
60
40
20
0
(b)
(a)
60
60
40
40
20
20
8
6
4
2
0
8
6
4
2
0
0
3000
6000
9000
12000
15000
0
3000
6000
9000
12000
15000
Time (s)
Time (s)
24

Figure 3
25

Figure 4
10
8
(a)
6
4
2
0
0,15
0,20
0,25
0,30
0,35
0,40
0,45
0,50
Initial concentration of Guaiacol (molL-1)
15
(b)
12
9
6
3
0
0,05
0,10
0,15
0,20
0,25
0,30
Initial concentration of Guaiacol (molL-1)
26

Figure 5
0,60
0,45
0,30
0,15
0,00
0,16
0,12
0,08
0,04
ReOx/SiO
2
ReS /SiO
2
2
R=0,44916
2
4
6
8
10
12
1/CGUA
R=0,99645
2
4
6
8
10
1/CGUA
27

Figure 6
0,24
0,20
0,16
0,12
0,08
ReS /SiO
2
2
0,220
0,216
0,212
0,208
R= 0,5131
ReOx/SiO2
0,0
0,1
0,2
0,3
0,4
0,5
-1
C (molL )
GUA
R= 0,94614
0,0
0,1
0,2
0,4
0,5
0,-31
C (molL )
GUA
28

Tables
Table 1. Products distribution on ReS2/SiO2 and ReOx/SiO2 catalysts calculated at 20 and
60% of guaiacol conversion.
At 20% Conversion
ReS2/SiO2 ReOx/SiO2
0.5 10
At 60% Conversion
ReS2/SiO2 ReOx/SiO2
0.8 17.5
Products (%)
BTX
Methanol
2.5
1.5
3.0
0.0
0.0
75.5
2.0
1.0
1.5
12.5
0.0
1.5
0.0
12.5
2.0
1.0
57
7.0
1.0
8.0
0.0
0.0
2.2
3.3
8.4
0.3
0.0
75.0
2.5
1.7
2.2
3.3
0.3
0.5
0.5
27.8
1.8
1.2
34
Cyclohexene
Cyclohexane
Hexane
Methylcyclohexane
Phenol
Cyclohexanol
Methylphenol
Anisole
9
0.5
7.2
0.0
0.0
Catechol
Methoxycyclohexanone
29