Effect of H2S and CO on the transformation of 2-ethylphenol as a model compound of bio-crude over sulfided Mo-based catalysts: Propositions of promoted active sites for deoxygenation pathways based on an experimental study

Article abstract of DOI:10.1039/c1gc15181a

Promoted (CoMo/Al₂O₃ and NiMo/Al₂O ₃) and unpromoted (Mo/Al₂O₃) catalysts were tested in the hydrodeoxygenation of 2-ethylphenol as a model compound of bio-crude under various partial press

Full text of DOI:10.1039/c1gc15181a

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 2441
www.rsc.org/greenchem
PAPER
Effect of H2S and CO on the transformation of 2-ethylphenol as a model
compound of bio-crude over sulfided Mo-based catalysts: propositions of
promoted active sites for deoxygenation pathways based on an experimental
study
Christophe Bouvier, Yilda Romero, Fre´de´ric Richard* and Sylvette Brunet
Received 18th February 2011, Accepted 17th May 2011
DOI: 10.1039/c1gc15181a
Promoted (CoMo/Al2O3 and NiMo/Al2O3) and unpromoted (Mo/Al2O3) catalysts were tested in
the hydrodeoxygenation of 2-ethylphenol as a model compound of bio-crude under various partial
pressures of H2S and CO. The catalytic tests were carried out at 340 ◦C under 7 MPa of total
pressure in a fixed-bed microreactor. H2S, needed to maintain the sulfidation level of the catalysts,
showed promoting or inhibiting effects depending on the catalyst tested and the deoxygenation
pathway considered. Over the three catalysts, H2S was found to slightly promote the HYD
pathway, which first involves the hydrogenation of the aromatic ring, whereas it strongly inhibited
the DDO pathway, which consists of a direct C–O bond scission, particularly on the CoMo/Al2O3
catalyst. In addition, a strong though reversible inhibition by CO was found over CoMo/Al2O3
for both deoxygenation pathways, while a limited effect was observed over NiMo/Al2O3.
Differences in inhibition induced by H2S and CO observed over both promoted catalysts
suggested that different active sites are involved in the hydrodeoxygenation of 2-ethylphenol,
depending on the nature and the localisation of promoters and on the transformation pathway.
rived from biomass, however, present a high oxygen content
Introduction
(between 20–55 wt%) depending on the biomass source and the
During the last century, economic development for a better
transformation process used6–11 due to numerous oxygenated
standard of living has been instrumental in damaging the
compounds, such as acid derivatives, aldehydes, alcohols and
environment. We are now in a position whereby we are
phenolic compounds.12–14 All these oxygenated compounds are
consuming more resources than ever, particularly fossil ones
responsible for some detrimental properties of bio-oils such as
for transportation. However, many major improvements have
high viscosity, low volatility, corrosiveness, immiscibility with
already been achieved in the last decades regarding emission
fossil fuels, thermal instability and a tendency to polymerize
abatement, energy consumption and development of alternative
under exposure to air. Therefore, bio-oil upgrading to liquid
technologies. The goals must be to give a positive answer in
fuel requires oxygen removal which can be achieved by a
topics such as to use renewable feedstocks, to reduce the total
hydrodeoxygenation (HDO) process. Such a process is carried
mass consumed and the amount of energy needed, to improve
out under hydrogen pressure (4–10 MPa) and at moderate
recyclability, to lessen waste production.
◦
temperature (300–500 C) where the catalysts usually used are
Most of the world’s energy demand is currently met by
fossil fuels and they are being depleted fast.1 This makes the
development of new liquid fuels coming from biomass an
attractive alternative.2,3 Following the fourth recommendation
of Anastas and Warner,4 the European Union is committed
to use non-food cellulosic and lignocellulosic biomass as a
sustainable alternative for bio-fuel production.5 Bio-oils de-
sulfided Mo/Al2O3 catalysts, promoted by metals of group VIII
(Co, Ni).3,15–17 Nevertheless, the sulfur content of bio-oils being
very low, the addition of sulfur-containing compounds to the
processed feed is required to maintain the sulfided state of such
catalysts. As organosulfur compounds present in petroleum
feedstocks lead to the formation of H2S by desulfurization, a
co-hydrotreatment of bio- and fossil oils could be a solution, as
already proposed recently in the literature.18–21 This would allow
operating the same production units and therefore obtaining an
upgraded bio-fuel at affordable prices. The study of the effect of
H2S during the HDO process using sulfided catalysts is therefore
relevant.
UMR CNRS 6503, Laboratoire de Catalyse en Chimie Organique,
Universite´ de Poitiers, 40 avenue du Recteur Pineau, 86022, Poitiers
Cedex, France. E-mail: frederic.richard@univ-poitiers.fr; Fax: +33 5
4945 3899; Tel: +33 5 4945 3519
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2441–2451 | 2441
©

View Online
Table 1 BET area, pore volume and chemical composition of the
In order to ease the understanding of complex catalytic
reactions, model molecules are often used. For the evalu-
ation of sulfided catalysts in HDO, benzofuran,22–25 pheno-
lic compounds,21,22,26–32 and acid derivatives such as methyl
heptanoate28,30,33–36 were recently chosen as probe oxygenated
molecules. The effect of H2S seems to be very dependent on
the oxygenated molecule used, the experimental conditions and
the type of catalyst (Ni(Co)Mo based catalysts). For example,
in a similar temperature range (300–340 ◦C), it was reported
that H2S inhibited the transformation of benzofuran over
NiMo/Al2O3 under 3.45 MPa of total pressure24 but enhanced
its conversion over NiMoP/Al2O3 under 7 MPa.22 An inhibiting
effect of H2S during the HDO of phenolic compounds using
sulfided catalysts is often described,26,30,35,37,38 even if a promoting
effect of H2S on the transformation of 2-ethylphenol over a
NiMoP/Al2O3 catalyst was recently reported.22 It is also known
that H2S strongly affects the selectivity of the catalysts. Indeed,
phenolic compounds mainly lead to aromatics and cycloalkanes
as deoxygenated products and their proportion depends on
the partial pressure of H2S used. In general, the formation of
aromatic compounds is more affected by H2S than the formation
of cycloalkanes, whatever the catalyst used (NiMo/Al2O3 or
CoMo/Al2O3).
catalysts in their oxide form29
Al2O3 Mo/Al2O3 CoMo/Al2O3 NiMo/Al2O3
BET area (m2 g-1)
252
251
255
0.64
9.2
4.2
—
257
0.66
9.3
—
Pore volume (cm3 g-1) 0.85 0.70
Mo (wt%)
Co (wt%)
Ni (wt%)
—
—
—
9.9
—
—
3.9
Table 2 Partial pressures in the gas phase in sulfidation and reaction
conditions29
Ptotal
P
Ptoluene P2-EtPh
P
P
PCO
H2
H2
S
CH4
(MPa) (MPa) (MPa) (kPa) (kPa) (kPa) (kPa)
Sulfidation
conditions
Reaction
4
7
2.674 1.068
—
129
129
—
5.750– 1.103– 49
5.813 1.181
10–49 10–49 0–30
conditions
HDO, with respect to the deoxygenation pathway and/or the
promoter.
During the HDO process, the oxygen atom can be eliminated
through water and/or carbon oxides production. In the case
of benzofuran and phenolic compounds, water is the only
oxygenated by-product while for acid derivatives, in addition
to water, CO and/or CO2 are expected. It has been shown that
water has a limited effect on deoxygenation of either phenolic
compounds37,39,40 or acid derivatives.34 However, carbon oxides
may have a negative impact on the deoxygenation rate. Indeed,
carbon monoxide is frequently used as an infra-red probe
molecule to characterize sulfided phases, since it is known to
strongly interact with the edge sites of sulfided catalysts.41–44 The
high adsorption observed in these conditions suggests that some
catalytic drawbacks can be expected if CO is formed during the
HDO process or through the reverse water-gas shift reaction
from CO2. Indeed, an important inhibiting effect of CO was
recently reported over sulfided catalysts for hydrodesulfurization
(HDS) reactions.45,46 However, to the best of our knowledge, its
impact on hydrodeoxygenation has never been reported.
The aim of the present paper is first to confirm the influ-
ence of the promoter type (Co and Ni) on the activity and
selectivity of three sulfided catalysts (Mo/Al2O3, NiMo/Al2O3
and CoMo/Al2O3) in the transformation of 2-ethylphenol
used as a model oxygenated molecule. The effect of different
partial pressures of H2S, produced in the reactional mixture
by decompostion of dimethyldisulfide (DMDS), as well as the
effect of CO are also investigated over the three catalysts.
Despite many proposals from different studies, a complete
understanding of the influence of H2S on the activity and
selectivity of sulfided catalysts is still missing. In addition, as
infra-red measurements of CO adsorption allow to distinguish
promoted from unpromoted sites of sulfided catalysts, it can
be expected that active sites could have a specific interaction
with CO under reaction conditions. Potential differences in
activity and selectivity in the course of a catalytic test might
therefore lead to valuable information and help determine
the nature and/or location of the active sites involved in
Experimental
Chemicals and catalysts
Dimethyldisulfide (DMDS, 98%), 2-ethylphenol (2-EtPh, 99%)
and toluene (99%) were purchased from Aldrich. Carbon
monoxide (1% in H2) and hydrogen (H2O < 5 ppm, O2 < 5
ppm) were provided by Air Liquide. 2-Ethylphenol was distilled
prior to use.
The three catalysts (Mo/Al2O3, NiMo/Al2O3 and
CoMo/Al2O3) were provided by Total. The same alumina
(g-Al2O3) was used to support all catalysts. Textural properties
of the catalysts were determined by nitrogen physisorption on a
ASAP 2000 Micromeritics instrument. The specific surface area
was calculated from the linear portion of BET plots (P/P0 =
0.05–0.30) and the pore volume at P/P0 = 0.99. The Co or Ni
composition was measured by XRF and the Mo content by
ICP. Catalysts composition and support textural properties are
presented in Table 1 for non-sulfided catalysts. All catalysts
have comparable textural properties and Mo and/or promoter
content.
The extrudates were crushed and sieved to a 250–315 mm size
range. The catalysts (50 mg of NiMo/Al2O3 and CoMo/Al2O3
or 100 mg of Mo/Al2O3) were diluted in carborundum to keep
the volume of the catalytic bed constant. The catalyst weight
was chosen in order to obtain a total molar conversion of 2-
EtPh near 20%.
Sulfidation and reaction conditions
The catalysts were sulfided in situ in a dynamic flow reactor
(length: 40 cm, inner diameter: 1.25 cm) using a sulfiding feed
(Table 2) made up of hydrogen, DMDS and toluene. The
sulfiding solution was injected at a starting temperature of
150 ◦C. After 1 h, the temperature was raised to 350 ◦C at a
5 ◦C min-1 rate and then kept at 350 ◦C for 14 h. It was finally
lowered to the reaction temperature (340 ◦C).
2442 | Green Chem., 2011, 13, 2441–2451
This journal is
The Royal Society of Chemistry 2011
©

View Online
The catalytic tests were performed right after sulfidation in
the same fixed-bed microreactor at 340 ◦C under 7 MPa of
total pressure. DMDS was added to the feed to generate a
constant partial pressure of H2S ranging from 10 to 49 kPa.
The partial pressure of 2-EtPh was fixed at 49 kPa. The effect
of H2S was measured with independent experiments. However,
the effect of carbon monoxide was evaluated during a single
experiment for each catalyst. CO was introduced after catalyst
stabilization (after about 20 h on stream) at partial pressures
set in the range of 2.5–30 kPa. After each exposition to CO
(about 4 h), the catalytic activity was measured without CO
until stabilization was observed. The total pressure was kept
constant at 7 MPa by adjusting the partial pressure of hydrogen
(Table 2). For a typical run, the contact time, defined as the ratio
between the volume of the catalyst and the gas flow of 2-EtPh
was close to 1.1 min for promoted catalysts (NiMo/Al2O3 and
CoMo/Al2O3) and near 2.3 min for the unpromoted catalyst
(Mo/Al2O3). No transformation of toluene was observed under
these experimental conditions.
Product analysis
The reactor effluents were condensed and liquid samples
were periodically analyzed with a Varian 430 chromatograph
equipped with a DB1 capillary column (length: 30 m; inside
diameter: 0.25 mm; film thickness: 0.25 mm) and a flame
ionization detector. The oven temperature was first set during
5 min at 55 ◦C, then raised to 175 ◦C at a rate of 4 ◦C min-1 and
kept at 175 ◦C for 5 min.
Gaseous effluents (CO, CO2, H2S and CH4) were analyzed
online with a Varian 450 chromatograph equipped with an
automatic sampling valve, two HayeSep Q columns (length:
0.5 m; inside diameter: 1/8 in; film thickness: 2.0 mm), a 13X
molecular sieve (length: 1.5 m; inside diameter: 1/8 in; film
thickness: 2.0 mm) and a thermal conductivity detector. A
backflush procedure allowed the removal of organic compounds.
The oven, block and filament temperatures were set at 40, 160
and 200 ◦C, respectively.
For example, the total catalyst activity (global reaction rate)
was calculated according to the following equation:
Fig. 1 Effect of time on stream on the transformation of 2-EtPh over
sulfided Mo/Al2O3 (a), NiMo/Al2O3 (b) and CoMo/Al2O3 (c) at 340 ◦C
under 7 MPa of total pressure in presence of 30 kPa of H2S in the feed.
Total activity (ꢀ); HYD activity (᭜); DDO activity (ꢀ); ACI activity
(ꢁ).
(1)
Atot = X.F/W
where Atot is the catalyst total activity (in mmol g-1 h-1), X the
conversion of 2-EtPh, F the molar flow rate of 2-EtPh (in mmol
h-1) and W the weight of catalyst (in g).
in the reactional mixture owing to its high reactivity. By
dehydration, it yielded a mixture of cycloalkenes (1- and 3-
ethylcyclohexenes) and ethylcyclohexane as the main product
(HYD pathway). As ethylbenzene was not hydrogenated into
ethylcyclohexane under these conditions, these two pathways
were parallel and independent. The third route (ACI pathway)
mainly involving the acid properties of the catalyst support led to
oxygenated products (phenol, 3-ethylphenol and diethylphenol
isomers) by disproportionation and isomerization reactions.29
They could be further deoxygenated following either the DDO
or the HYD pathway, yielding mainly benzene, cyclohexane,
diethylcyclohexanes and diethylbenzenes.
Results and discussion
Effect of time on stream on the transformation of 2-EtPh over
sulfided catalysts-probable causes of deactivation
The transformation of 2-EtPh over three sulfided catalysts
(Mo/Al2O3, NiMo/Al2O3 and CoMo/Al2O3) was evaluated
in the presence of 30 kPa of H2S in the feed (Fig. 1). Over
these catalysts, 2-EtPh reacted according to three pathways,
as already reported29 and are presented in Scheme 1. The
first pathway involved a direct C–O bond scission (direct
deoxygenation route – DDO pathway) yielding ethylbenzene
only. The second one led, after hydrogenation of the aromatic
ring, to 2-ethylcyclohexanol. This alcohol was never observed
Fig. 1 shows the effect of time on stream on the activity
of an unpromoted catalyst (Mo/Al2O3) and two promoted
catalysts (CoMo/Al2O3 and NiMo/Al2O3). NiMo/Al2O3 was
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2441–2451 | 2443
©

View Online
Scheme 1 Transformation pathways of 2-EtPh over sulfided Mo-based catalysts.29
Table 3 Chemical composition in carbon and sulfur of sulfided and
the least stable of all catalysts used in the present study. Indeed,
its residual total activity, calculated as the ratio between the
total activity obtained after 30 h and after 1 h on stream,
was 0.7 whereas it was close to 0.9 for both Mo/Al2O3 and
CoMo/Al2O3.
Contrary to the HYD route, the activity of the DDO pathway
was quite stable for the three catalysts, since the residual
activities were always close to 0.9. The difference in behavior
for deactivation between these two pathways seems to indicate
that the nature or the location of active sites for each pathway
could be different.
The activity of the ACI pathway was also stable for all
catalysts, as indicated in Fig. 1. Products formed through
this pathway were mainly oxygenated and represented about
70 mol% of the products obtained by this route over both
promoted catalysts and 84 mol% over the unpromoted catalyst.
They were produced by disproportionation and isomerization
reactions involving the acid properties of the support, as already
mentioned.
The main reason for the deactivation of sulfided catalysts
involves a loss of active sites that could be due to several
phenomena such as coke formation and/or modification of
the sulfided phase.47 As shown in Table 3, all spent catalysts
presented an increase in carbon content that could be con-
tributed to coking. This amount was higher over NiMo/Al2O3,
which could explain the important deactivation observed for this
catalyst (Fig. 1). These results seem in contradiction with those
reported by Guichard et al.48 which stated that carbon content
was more important over CoMo/Al2O3 than over NiMo/Al2O3
after desulfurization of gas oil feedstocks. Consequently, the
authors suggested that active sites on the CoMoS phase are more
likely to be covered by coke than those on the NiMoS phase.48,49
The different behaviors can be explained in part by the different
feedstocks used. Nevertheless, Murti et al.50 observed that the
carbon content was 1.3 times higher over NiMo/Al2O3 than over
CoMo/Al2O3 after the hydrodesulfurization of straight run gas
spent catalysts
% C
% S
Sulfided catalystsa
Mo/Al2O3
CoMo/Al2O3
NiMo/Al2O3
Spent catalystsb
Mo/Al2O3
1.3
1.2
0.4
5.3
7.4
7.9
3.9
3.3
4.7
5.1
6.4
6.1
CoMo/Al2O3
NiMo/Al2O3
a Sulfided catalysts were obtained after the sulfidation procedure de-
scribed in the experimental part. b Spent catalysts were obtained after
30 h on stream in the transformation of 2-EtPh at 340 ◦C under 7 MPa
of total pressure, in presence of 30 kPa of H2S.
oil, as observed in the present study. Even if carbon deposition
should mainly take place on alumina,48,51 it seems that the carbon
content had no deactivating effect on the ACI pathway.
A second possible source of deactivation is the modification of
the sulfided phases. It is now well accepted that they are made of
crystallites which expose two types of edges: metallic and sulfur
edges, and that active sites are coordinatively unsaturated sites
(CUS) present on both edges. It has recently been proposed that
coke precursors adsorb preferentially on metallic edges.52 As the
HYD pathway was the one showing the highest deactivation for
the three catalysts, this might indicate that it probably requires
active sites mainly located on the metallic edge, more sensitive
to coke formation.
The sulfur content of the spent unpromoted catalyst seemed
stable, as indicated in Table 3, even if it was lower than those
measured for promoted catalysts. However, it is known that the
presence of promoter (Ni or Co) allows a higher sulfidation
level.53 The loss of sulfur content for NiMo/Al2O3 (about 23%)
was the most significant and could also explain in part the
important deactivation observed for this catalyst. The sulfur
2444 | Green Chem., 2011, 13, 2441–2451
This journal is
The Royal Society of Chemistry 2011
©

View Online
loss over NiMo/Al2O3 has actually been proposed as a main
cause of deactivation during the desulfurization of thiophene.54
Another possibility is the loss of the promoted phase through
the segregation of promoters.55 For example, Guichard et al.48,49
have recently reported that promoter segregation over CoMoS
and NiMoS phases could be linked with sulfided catalysts
deactivation. The authors suggested that Ni segregated from
both the sulfur and metallic edges of crystallites, while Co
might only segregate from the metallic edge. Their study49 over
NiMoP/Al2O3 and CoMoP/Al2O3 led them to conclude that
with Co, deactivation was mainly due to coking, while with Ni
as promoter, segregation and the MoS2 phase modifications were
responsible for the catalyst deactivation. In our case, as time on
stream was relatively short and H2S was present in the reaction
mixture, the conditions were probably insufficient to allow the
segregation of the promoters (Ni or Co).
The difference in behavior between the two promoted catalysts
could be explained in part by a different localisation of the
promoters, Ni and Co. Based on density functional theory
(DFT) calculations, it has been established that the most
stable position for promoter atoms is in the substitution of
molybdenum on the edges of the sulfided phase55,59–62 leading to
the Ni(Co)–Mo–S mixed phase model developed by Topsøe and
coworkers.15,63,64 For a CoMoS crystallite, close to a hexagon,
the sulfur edge is supposed to be fully promoted while the
metallic edge could be only partially substitued, depending on
the sulfo-reductive conditions.55 As far as the NiMoS crystallite
is concerned, it is close to a deformed hexagon exposing more
metallic than sulfur edges. Moreover, it is proposed that Ni
is preferentially located on metallic edges. As promoters are
supposed to increase the number of vacancies, it is expected that
the amount of active sites is more important on the sulfur edges
of a CoMoS phase and on the metallic edges of a NiMoS phase.
As the DDO pathway was promoted more over CoMo/Al2O3, it
seems that DDO active sites could be preferentially found on the
sulfur edges. On the opposite, the HYD active sites could likely
be located on the metallic edges, as already proposed above to
explain the deactivation of the HYD route.
It is worth mentioning that the adsorption mode of 2-EtPh
is reasonably different for both pathways. Indeed, the carbon-
oxygen bond scission involved in the DDO pathway requires a
h1 adsorption through the oxygen atom.65 On the other hand,
hydrogenation of the aromatic ring, needed in the HYD route,
probably proceeds through a flat adsorption (h5 type).
Fig. 2 shows some representations of the promoted active sites
proposed in several DFT studies44,55,57,62,66 under typical hydro-
treatment conditions and likely to be involved in deoxygenation
Effect of the promoter type-propositions of promoted active sites
Table 4 shows the activity for the three catalysts measured after
30 h on stream with 30 kPa of H2S in the feed. The total activity
was largely increased over promoted catalysts compared to the
unpromoted one. The promoting effect was measured as the
ratio between the activity of promoted (by Ni or Co) catalysts
and the activity of the unpromoted catalyst. It ranged from 2.0 to
2.3, depending on the promoter. These results are in agreement
with the fact that cobalt and nickel are known for weakening
the metal-sulfur bond and therefore for increasing the number
of sulfur vacancies (CUS) proposed as active sites for such a
reaction.56–58
The relative importance of each pathway was very dependent
on the catalyst. Over Mo/Al2O3, the contribution of both main
pathways (ACI and HYD) was about the same after 30 h on
stream in the presence of 30 kPa of H2S (Table 4). When Ni was
present, the HYD pathway was enhanced and therefore became
predominant. Indeed, the promoter effect of nickel on the HYD
route was close to 3, while no promoting effect was measured
on the DDO route. Over CoMo/Al2O3, both deoxygenation
pathways were promoted, the DDO route in a larger extend
than the HYD one. Therefore, the DDO/HYD selectivity was
very dependent on the catalyst used and was always the highest
over CoMo/Al2O3. These results are in line with previous studies
reported with phenolic compounds.28,37
Table 4 Activity and DDO/HYD selectivity measured after 30 h on
stream in the transformation of 2-EtPh over sulfided catalysts at 340 ◦C
under 7 MPa of total pressure in presence of 30 kPa of H2S in the
feed. Promoting effects (defined as the ratio between the activity of a
promoted and the unpromoted catalyst) are given in brackets for Co and
Ni
Mo/Al2O3 CoMo/Al2O3 NiMo/Al2O3
Activity (mmol g-1 h-1) Total 10.5
HDOa 6.8
24.6 (2.3)
20.7 (3.0)
8.7 (1.9)
10.4 (6.9)
5.5 (1.3)
1.18
21.4 (2.0)
17.3 (2.5)
14.5 (3.2)
1.3 (0.9)
5.6 (1.3)
0.09
HYD 4.6
DDO 1.5
ACI 4.4
Fig. 2 Representations of probable promoted active sites from DFT
studies described in the literature. NiMoS promoted metallic edge (a);
CoMoS promoted metallic edge (b); CoMoS fully promoted sulfur edge
(c); CoMoS fully promoted sulfur edge after S-shift (d).
DDO/HYD selectivity
0.33
a Activity in total deoxygenation.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2441–2451 | 2445
©

View Online
reactions. Fig. 2a presents a NiMoS metallic edge where Ni has a
square-planar sulfur coordination. Due to the higher density of
Ni atoms on this edge and the fact that they are not covered by
S atoms, the probability of the occurrence of two vacancies on
two paired Ni atoms is increased, favoring a flat adsorption of
2-EtPh. Therefore, we propose that such a site could be active in
the HYD pathway of 2-EtPh. In the case of the CoMoS phase,
the metallic edge presents a lower promotion rate with five-fold
Mo atoms and square-planar four-fold Co atoms, as depicted in
Fig. 2b. Over these type of sites, the flat adsorption of 2-EtPh
is probably slightly more hindered than over Ni-promoted sites
presented in Fig. 2a, resulting in a higher promoting effect of
Ni compared to Co for the HYD pathway, as observed in Table
4. Moreover, such sites could also allow an adsorption through
the oxygen atom, contributing to the DDO activity.
On the other hand, as discussed earlier, the promotion level
achieved on the sulfur edge is proposed to be more important
when Co is used as promoter. Furthermore, the substitution of
Mo by Co does not induce an important structural modification
of the sulfur edge. As a consequence, Co is four-fold coordinated
with S atoms in bridging positions as shown in Fig. 2c. On this
edge, the creation of a vacancy is required to adsorb oxygenated
compounds. It could be obtained by a shift of a S atom towards
a neighboring Co atom, as reported recently by Badawi et al.67
and presented in Fig. 2d. In addition, the authors proposed that
the creation of such a vacancy is energy demanding contrary to
the one presented in Fig. 2b. Active sites shown in Fig. 2b and 2d
could be involved in the DDO pathway and therefore both the
metallic and the sulfur edges could be active in the direct scission
of the C–O bond. As a consequence, an important promoting
effect of Co on the DDO pathway was observed, as can be
seen in Table 4. This is in accordance with results obtained by
DFT calculations in the case of thiophene, which indicate that
the adsorption energy through the sulfur atom is similar on
both edges.45 With respect to NiMo/Al2O3, due to the limited
presence of Ni on the sulfur edges and the probable preference
for a flat adsorption of the aromatic ring instead of an adsorption
through the oxygen atom over metallic promoted edges (Fig. 2a),
no promoting effect on the DDO pathway was noticed.
Effect of H2S on the activity and selectivity of sulfided promoted
and unpromoted catalysts
The effect of H2S on the activity of the three sulfided catalysts
measured after 30 h on stream was studied in the range of
10–49 kPa. The three catalysts showed a different tolerance
to H2S: the activity of both Mo/Al2O3 and NiMo/Al2O3
increased with the partial pressure of H2S whereas its effect
on the activity of CoMo/Al2O3 was more complex (Fig. 3a).
Indeed, CoMo/Al2O3 was more active in the presence of a low
H2S pressure (10 kPa), whereas NiMo/Al2O3 was more active
when the pressure of H2S was increased. This can be explained
by the fact that the main pathway over CoMo/Al2O3 is the
DDO route, which is strongly inhibited by H2S (Fig. 3c), while
the HYD pathway, which is the main pathway observed over
NiMo/Al2O3, appears to be slightly promoted by H2S (Fig.
3b). This is confirmed by negative apparent kinetic orders with
respect to H2S for the DDO activity and the slightly positive ones
observed for the HYD activity (Table 5). In all cases, Mo/Al2O3
was roughly twice less active than the promoted catalysts. The
Fig. 3 Influence of the partial pressure of H2S on activities measured after 30 h on stream in the transformation of 2-EtPh over sulfided Mo/Al2O3
(᭜), NiMo/Al2O3 (ꢁ) and CoMo/Al2O3 (ꢀ) at 340 ◦C under 7 MPa of total pressure. Total activity (a); HYD activity (b); DDO activity (c); ACI
activity (d).
2446 | Green Chem., 2011, 13, 2441–2451
This journal is
The Royal Society of Chemistry 2011
©

View Online
Table 5 Apparent kinetic orders with respect to H2S for the transfor-
mation of 2-EtPh
Total
HYD
DDO
ACI
Mo/Al2O3
CoMo/Al2O3
NiMo/Al2O3
0.77
-0.17
0.58
0.15
0.17
0.18
-0.36
-1.40
-0.54
1.48
1.20
1.70
DDO/HYD selectivity was therefore higher for a lower partial
pressure of H2S, especially in the case of CoMo/Al2O3 where it
was above 1 in presence of less than 30 kPa of H2S (Fig. 4).
Fig. 5 Saturated CoMoS fully promoted sulfur edge carrying SH
groups.
and dissociation of H2S and similarly the adsorption and the
C–O bond scission involved in the DDO pathway.
Over the three catalysts, H2S had no inhibiting effect on
the HYD route indicating no competitive adsorption between
H2S and 2-EtPh over HYD sites. The same observation was
made by Viljava et al.26 for the deoxygenation of phenol over a
CoMo/Al2O3 catalyst. This could be explained by the fact that
hydrogenation sites are proposed as less electrophilic than DDO
sites32,37 and could be less sensitive to the presence of H2S.
A slight promoting effect of H2S was even noted since
positive kinetic orders (close to 0.2) were calculated, as shown
in Table 5. A similar promoting effect was also reported in the
Fig. 4 Influence of the partial pressure of H2S on the DDO/HYD
selectivity measured after 30 h on stream in the transformation of 2-
EtPh over sulfided Mo/Al2O3 (᭜), NiMo/Al2O3 (ꢁ) and CoMo/Al2O3
(ꢀ) at 340 ◦C under 7 MPa of total pressure.
The difference in behavior between the DDO and HYD
pathways as a function of the partial pressure of H2S seems
to confirm that the active sites involved in both deoxygenation
routes are, at least in part, different. Similarly, the two parallel
routes involved in the desulfurization of dibenzothiophenic
compounds (namely HYD and DDS) were affected in a different
way by H2S. The DDS pathway was always more inhibited than
the HYD route.68,69 This is in good agreement with the proposal
that sites involved in HDO and HDS reactions could be similar
in nature.15
The inhibiting effect of H2S on the DDO pathway of
CoMo/Al2O3, as observed in Table 5, is in line with results
presented in the literature.22,26,28,37 It can be explained by a
competitive adsorption between H2S and 2-EtPh on the DDO
active sites. After adsorption, H2S can dissociate heterolytically
on a DDO site described in Fig. 2d leading to two SH groups,
as shown in Fig. 5. Such a modified active site cannot allow an
adsorption through the O atom anymore, leading to a loss of
the DDO activity when the partial pressure of H2S increases.
The effect of H2S on the DDO pathway for both Mo/Al2O3
and NiMo/Al2O3 was more limited since this route was less
favored than over the CoMo/Al2O3 catalyst. The absence of
promoting effect of Ni and the comparable effect of H2S on the
DDO activity of both catalysts indicated that the DDO sites were
similar in nature for Mo/Al2O3 and NiMo/Al2O3. There could
be a vacancy on a Mo atom on an unpromoted fragment. As H2S
had a limited effect over Mo/Al2O3 and NiMo/Al2O3 compared
to CoMo/Al2O3, the vacancies present on an unpromoted phase
seem to be less electrophilic than the vacancies involved over a
CoMoS phase. Therefore the latters could favor the adsorption
22
transformation of phenolic compounds over a NiMoP/Al2O3
and a CoMo/Al2O3.37 As a consequence, if the HYD active
sites of the NiMoS phase are, for example, those described in
Fig. 2a, they seem to be insensitive to the presence of H2S,
but cannot justify the increase of the HYD activity when the
partial pressure of H2S increases. The contribution of other sites
is needed to explain the positive role of H2S in hydrogenation
reactions. Accordingly, fully sulfided sites showing no vacancies
could act as hydrogenation sites, as already proposed by several
authors in HDS.54,69–73 Therefore, an increase of the partial
pressure of H2S would result in a more important HYD activity,
as observed experimentally. This saturation of vacancies could
explain both the inhibition observed on the DDO pathway
by loss of electrophilic active sites and the promotion of the
HYD activity by an increase of hydrogenating sites. It could
therefore be seen as a conversion of hydrogenolysis sites into
hydrogenation sites in presence of H2S, as proposed by several
groups.37,74,75
As shown in Fig. 3d, the increase of the partial pressure of H2S
also had an influence on the ACI pathway, which mainly involves
the acidic properties of the support. Indeed, the apparent kinetic
orders with respect to H2S were between 1.20 and 1.70 for this
route (Table 5). It was reported that adsorption of H2S on
alumina can increase the Brønsted acidity of the support54,76 and
therefore ease isomerization and disproportionation reactions
involved in this pathway. The presence of Co or Ni did not have
any major influence on the activity of this pathway. This seems
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2441–2451 | 2447
©

View Online
Table 6 Partial pressures of H2S and CH4 measured in the presence
and absence of CO during the transformation of 2-EtPh over sulfided
catalysts
to indicate that only the acidity of alumina is responsible for the
presence of isomerization and disproportionation reactions.
a
b
a
b
Decomposition of DMDS in presence of carbon monoxide
PH (kPa) PH (kPa) PCH (kPa) PCH (kPa)
2 S
2 S
4
4
Decomposition of DMDS towards H2S and CH4 (eqn (2)) occurs
over naked Al2O3 but is twice more important over supported
sulfided catalysts.
Mo/Al2O3
CoMo/Al2O3 32.8
NiMo/Al2O3 30.7
29.9
29.4
29.1
30.9
29.8
31.0
31.6
29.0
29.5
28.8
a In absence of CO. b In presence of 30 kPa of CO.
(2)
DMDS + 3H2 → 2CH4 + 2H2S
Table 6 summarizes the average partial pressures observed
during the transformation of 2-EtPh over the three sulfided
catalysts, in the presence or absence of CO. Considering that
deoxygenation of 2-EtPh leads to the formation of water, two
reactions can take place in the presence of CO and could modify
the partial pressures in the gas phase:
known that metallic nickel catalyzes the methanation reaction.
In addition, over all catalysts, the introduction of CO slightly
decreased the decomposition of DMDS towards both H2S and
CH4 in similar proportions (less than 5%) as derived from Table
6.
CO + 3H2 ꢀ CH4 + H2O
(3)
(4)
Impact of CO on the transformation of 2-EtPh over sulfided
catalysts in presence of 30 kPa of H2S
CO + H2O ꢀ CO2 + H2
The influence of carbon monoxide on the transformation of 2-
EtPh over the three sulfided catalysts was evaluated in presence
of 30 kPa of H2S in the feed. Fig. 6 presents the activity obtained
over the three sulfided catalysts as a function of the CO partial
pressure. For all catalysts and whatever the partial pressure of
CO applied, it was possible to fully recover the activity level
obtained before exposition to CO, indicating, when observed,
the totally reversible adsorption of CO on the sulfided phase.
No influence of CO was observed over Mo/Al2O3 for
both partial pressures of CO investigated (15 and 30 kPa).
Consequently, the DDO/HYD selectivity was unmodified in
Eqn (3) and (4) being the methanation and the water-gas shift
equilibria, respectively.
Nevertheless, as CO2 was never detected whatever the catalyst
used, it seems that the water-gas shift reaction did not occur.
Moreover, the fact that the partial pressures of H2S and CH4,
in the absence and presence of CO, were comparable (Table
6) clearly outlined that no conversion of CO towards CH4
was achieved. In conclusion, under our reaction conditions,
CO was unreactive. The absence of Ni segregation under our
experimental conditions can be highlighted since it is well
Fig. 6 Effect of the CO partial pressure on activities measured after catalyst stabilization on the transformation of 2-EtPh over sulfided Mo/Al2O3
(᭜), CoMo/Al2O3 (ꢀ) and NiMo/Al2O3 (ꢁ) at 340 ◦C under 7 MPa of total pressure in presence of 30 kPa of H2S in the feed. Total activity (a);
HYD activity (b); DDO activity (c); ACI activity (d).
2448 | Green Chem., 2011, 13, 2441–2451
This journal is
The Royal Society of Chemistry 2011
©

View Online
the presence of CO (Fig. 7). This result clearly indicates that
unpromoted sites are not affected by CO.
both the HYD and DDO activities of CoMo/Al2O3 would have
been quite the same as the activities of the unpromoted catalyst.
As shown in Fig. 7, the DDO/HYD selectivity of CoMo/Al2O3
decreased with the increase of the quantity of CO introduced.
The DDO route was the main transformation pathway of 2-
EtPh in the absence of CO (DDO/HYD selectivity close to
1.2) while the HYD pathway became predominant for any CO
partial pressure investigated (DDO/HYD selectivity near 0.5 in
presence of 33.5 kPa of CO).
It is interesting to note that similar trends in terms of
inhibition were obtained over the three catalysts for both the
transformation of 2-EtPh and the decomposition of DMDS.
Indeed, even if the latter was only scarcely inhibited, the effect
was more pronounced over CoMo/Al2O3.
For all catalysts, as expected, the apparent inhibition of the
ACI pathway was negligible (Fig. 6d), since CO does not modify
the acidity of alumina, which is responsible for isomerisation
and disproportionation reactions involved in this pathway.
Nevertheless, for both promoted catalysts, the ACI product
distribution revealed that oxygenated compounds accounted for
75% over NiMo/Al2O3 and 90% over CoMo/Al2O3, when they
represented only 70% in absence of CO for these two catalysts,
confirming both the inhibition of the deoxygenation level and
their different tolerance to CO.
Our results clearly show a competitive adsorption between
CO and an oxygenated compound on the active sites promoted
by Co, leading to a strong inhibition of the deoxygenation of 2-
EtPh. Indeed, CO can be adsorbed through its carbon atom on
cobalt over both edges, as depicted in Fig. 8, in accordance with
DFT studies.44,45,67 Such an inhibiting effect was not observed
for the unpromoted catalyst. These results are in agreement with
those reported by Travert et al.,44 which indicate a considerably
lower adsorption of CO over an unpromoted MoS2 phase
compared to any edge or promotion degree of a CoMoS phase.
Similarly, the absence of inhibition by CO over Mo/Al2O3 is
in good agreement with the results obtained when the effect
of H2S was investigated. Indeed, we propose that unpromoted
vacancies involved in the DDO pathway are less electrophilic
than the vacancies associated to Co and therefore the former
should also be less sensitive to CO. With respect to active sites
involved in the HYD pathway of the unpromoted catalyst, they
Fig. 7 DDO/HYD selectivity as a function of the CO partial
pressure in the transformation of 2-EtPh over sulfided Mo/Al2O3 (᭜),
CoMo/Al2O3 (ꢀ) and NiMo/Al2O3 (ꢁ) at 340 ◦C under 7 MPa of
total pressure in presence of 30 kPa of H2S in the feed.
On the other hand, both promoted catalysts suffered from an
inhibition due to CO, but in completely different proportions, as
shown in Fig. 6. Over NiMo/Al2O3, the loss of total activity was
practically linear and limited to less than 10% in the presence
of the highest CO content studied (30 kPa). This decrease was
mainly due to a slight inhibition of the HYD pathway (Fig. 6b)
whereas both the DDO (Fig. 6c) and ACI (Fig. 6d) pathways
were practically unaffected by CO addition. Nevertheless, the
DDO/HYD selectivity was lower than 0.1 (Fig. 7), showing
that the HYD route remained largely predominant over Ni-
promoted sulfided catalyst.
With respect to CoMo/Al2O3, the increase of CO content
resulted in an increase of the inhibition level of its total activity
(Fig. 6a). Indeed, it was already reduced from one third for an
exposition to a low partial pressure of CO (2.5 kPa), and was
halved at partial pressures above 20 kPa. As expected, a negative
kinetic order with respect to CO was calculated between 2.5
and 33.5 kPa and found equal to -0.11 (Table 7). Interestingly,
both deoxygenation pathways were affected at different levels.
Whatever the partial pressure of CO applied, the DDO activity
was the most inhibited with up to 75% of inhibition in the
presence of 33.5 kPa of CO, but a low partial pressure of CO
(2.5 kPa) already caused a loss of activity close to 50% (Fig. 6c).
The same phenomenon was observed to a lesser extend for the
HYD activity, for which the inhibition reached up to 55% in
the presence of 33.5 kPa of CO (Fig. 6b). The apparent kinetic
orders with respect to CO were -0.11 and -0.25 for the HYD
and DDO pathways, respectively (Table 7), confirming a more
significant inhibiting effect on the DDO route than on the HYD
route. In fact, it seems that, in presence of a larger amount of CO,
Table 7 Apparent kinetic orders with respect to CO for the transfor-
mation of 2-EtPh
Total
HYD
DDO
ACI
a
CoMo/Al2O3
NiMo/Al2O3
-0.11
-0.02
-0.11
-0.05
-0.25
-0.03
0.01
0.00
b
a Calculated between 2.5 and 33.5 kPa of CO. b Calculated between 6.5
and 29.7 kPa of CO.
Fig. 8 Adsorption of CO on a CoMoS fully promoted sulfur edge after
S-shift (a), and on a CoMoS promoted metallic edge (b).
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2441–2451 | 2449
©

View Online
are likely to be fully sulfided sites and consequently unaffected
by both H2S and CO, as observed experimentally.
to the presence of H2S and CO due to a competitive adsorption
with the oxygenated compound. In the case of NiMo/Al2O3,
the DDO sites are similar to those present on the unpromoted
catalyst since both DDO activities are comparable whatever the
partial pressures of H2S and CO studied. In addition as such
sites are hardly affected by both H2S and CO, they could be less
electrophilic than those present over a CoMoS phase.
The differences in the inhibition of the DDO and HYD
pathways observed over CoMo/Al2O3 clearly indicate that both
routes require different active sites. The levelling of the inhibition
of both pathways at higher CO partial pressures towards the
activity of an unpromoted catalyst might suggest that each route
involves at least two distinct active sites: one involving cobalt,
strongly disabled by CO and one due to molybdenum only, not
affected by CO in the range of partial pressures investigated. The
latter could reasonably be similar in nature to the ones present
over the unpromoted catalyst and would explain why the activity
of CoMo/Al2O3 and Mo/Al2O3 are similar in the presence of a
large amount of CO, as observed in Fig. 6.
The differences in CO inhibition between CoMo/Al2O3 and
NiMo/Al2O3 confirm that different active sites are involved for
both promoted catalysts in the deoxygenation of 2-EtPh. The
weaker inhibiting effect of CO on the activity of NiMo/Al2O3
could be explained by the lower adsorption energy of CO on Ni
compared to Co, as already reported.44 In addition, since the sites
involved in the DDO route of NiMo/Al2O3 are not affected by
the presence of CO, it is possible to confirm the hypothesis made
to explain the effect of H2S, i.e. that they are similar in nature to
those present over Mo/Al2O3. This is in good agreement with
the fact that Ni has no promoting effect on this pathway, as
indicated in Table 4. On the other hand, CO slightly inhibited
the HYD pathway of NiMo/Al2O3, the only Ni promoted route.
In analogy with CoMo/Al2O3, we can consider that active sites
of this catalyst are both unpromoted and promoted sites. As
discussed above, the former are insensitive to CO. Considering
that the promoted sites are mainly located on the metallic edge
and are possibly described in Fig. 2a, they are scarcely affected
by CO, confirming the low effect observed with H2S. This type
of sites could be less electrophilic than those promoted by cobalt
and described in Fig. 2b. This correlates with the difference of
CO adsorption energy between Co and Ni.
Concerning the HYD route, where a flat adsorption of
the phenolic compound is required, the hydrogenation of the
aromatic ring can take place over at least two different active
sites. The first one could be two neighboring vacancies formed
by two paired promoter atoms located on metallic edges. Such a
site should be more present in the case of a NiMoS phase since
Ni promoters are known to be more abundant on metallic edges
than Co atoms. This could explain the higher promoting effect
of Ni on the HYD pathway. The second type of HYD site could
be a totally sulfided surface which is insensitive to H2S and CO
and could explain the HYD activity of an unpromoted catalyst.
If one aims for a lower hydrogen consumption during a
sustainable HDO process, our results show that CoMo/Al2O3
is the catalyst of choice as it is favoring the DDO pathway.
Nevertheless, the probable presence of CO from deoxygenation
of acid derivatives present in a large amount in bio-oils will
considerably decrease its activity, whereas the one observed over
a NiMo/Al2O3 catalyst should not be affected in a large extend.
Similarly, H2S generated by desulfurization of fossil fuels will
bring a limitation to the use of a CoMo/Al2O3 catalyst compared
to NiMo/Al2O3, in the case of a co-hydro-treatment of fossil
fuels and bio-oils. Consequently, despite its lower selectivity for
the DDO route, a sulfided Ni-promoted catalyst seems to be
the catalyst of choice for HDO reactions as it is less sensitive to
inhibition by inherent by-products (CO and H2S).
Acknowledgements
This work has been performed within the ANR funded Pro-
gramme National de Recherche sur les bioe´nergies-ECOHDOC,
a joint project of the Centre National de la Recherche Scien-
tifique (CNRS), the Universities of Caen, Lille and Poitiers, and
Total.
Conclusions
The present study evaluated both the influence of the hydrogen
sulfide content and the impact of carbon monoxide on the
hydrodeoxygenation of 2-EtPh over an unpromoted catalyst
(Mo/Al2O3) and two promoted catalysts (NiMo/Al2O3 and
CoMo/Al2O3). Over this type of sulfided catalysts, 2-EtPh was
deoxygenated via two parallel routes: one involving first the
hydrogenation of the aromatic ring (HYD pathway) and the
second consisting in a direct C–O bond scission (DDO pathway).
The global different tolerance to H2S of CoMo/Al2O3 com-
pared to NiMo/Al2O3 can be explained by the fact that H2S
strongly inhibits the DDO route, which is only promoted by
Co and slightly increased the HYD route, which is mainly
promoted by Ni. This clearly indicates that the sites involved
in both pathway are, at least in part, different. Moreover, the
strong affinity of CO for active sites present on CoMo/Al2O3
compared to the weak inhibition observed over NiMo/Al2O3
confirm differences between the active sites of both catalysts.
With respect to the DDO sites, they are vacancies that could
be located on both edges (metallic and sulfur) of sulfided slabs.
In the case of a CoMoS phase, this type of site is very sensitive
Notes and references
1 C. Di Blasi, Prog. Energy Combust. Sci., 2008, 34, 47.
2 D. O. Hall, Forest Ecol. Manage., 1997, 91, 17.
3 A. G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044.
4 P. T. Anastas, J. C. Warner, Green Chemistry: Theory and Practice,
Oxford University Press, Oxford, 1998.
5 Off. J. Eur,. Union, 2009, L140/88.
6 L. Wei, S. Xu, L. Zhang, H. Zhang, C. Liu, H. Zhu and S. Liu, Fuel
Process. Technol., 2006, 87, 863.
7 D. Mohan, C. U. Pittman and P. H. Steele, Energy Fuels, 2006, 20,
848.
8 A. Demirbas, Prog. Energy Combust. Sci., 2007, 33, 1.
9 D. C. Elliot, Energy Fuels, 2007, 21, 1792.
10 O. Senneca, Fuel Process. Technol., 2007, 88, 87.
11 D. M. Alonso, J. Q. Bond and J. A. Dumesis, Green Chem., 2010, 12,
1493.
12 L. Ingram, D. Mohan, M. Bricka, P. H. Steele, D. Strobel, D. Crocker,
B. Mitchell, J. Mohammad, K. Cantrell and C. U. Pittman, Energy
Fuels, 2008, 22, 614.
2450 | Green Chem., 2011, 13, 2441–2451
This journal is
The Royal Society of Chemistry 2011
©

View Online
13 J. H. Marsman, J. Wildschut, F. Mahfud and H. J. Heeres, J.
Chromatogr., A, 2007, 1150, 21.
14 J. H. Marsman, J. Wildschut, P. Evers, S. de Koning and H. J. Heeres,
J. Chromatogr. A, 2008, 1188, 17.
15 E. Furimsky, Appl. Catal., A, 2000, 199, 147.
16 G. W. Huber and A. Corma, Angew. Chem., Int. Ed., 2007, 46,
7184.
17 H. Topsøe, B. S. Clausen, F. E. Massoth, in Hydrotreating Catalysis,
Science and Technology, ed. J. R. Anderson and M. Boudart, Springer
Verlag, Berlin, 1996, vol. 11.
45 F. Pelardy, C. Dupont, C. Fontaine, E. Devers, A. Daudin, F.
Bertoncini, P. Raybaud and S. Brunet, Appl. Catal., B, 2010, 97,
323.
46 P. Ghosh, A. T. Andrews, R. J. Quann and T. R. Halbert, Energy
Fuels, 2009, 23, 5743.
47 E. Furimsky and F. E. Massoth, Catal. Today, 1999, 52, 381.
48 B. Guichard, M. Roy-Auberger, E. Devers, B. Rebours, A. A.
Quoineaud and M. Digne, Appl. Catal., A, 2009, 367, 1.
49 B. Guichard, M. Roy-Auberger, E. Devers, B. Rebours, A. A.
Quoineaud and M. Digne, Appl. Catal., A, 2009, 367, 9.
50 S. D. S. Murti, K-H Choi, Y. Korai and I. Mochida, Appl. Catal., A,
2005, 280, 133.
18 A. Pinheiro, D. Hudebine, N. Dupassieux and C. Geantet, Energy
Fuels, 2009, 23, 1007.
19 V. N. Bui, G. Toussaint, D. Laurenti, C. Mirodantos and C. Geantet,
Catal. Today, 2009, 143, 72.
20 F. de Miguel Mercader, M. J. Groeneveld, S. R. A. Kersyen, N. W. J.
Way, C. J. Schaverien and J. A. Hogendoorn, Appl. Catal. B, 2010,
86, 57.
21 M. Philippe, F. Richard, D. Hudebine and S. Brunet, Appl. Catal.,
A, 2010, 383, 14.
22 Y. Romero, F. Richard, Y. Rene`me and S. Brunet, Appl. Catal., A,
2009, 353, 46.
51 A. Popov, E. Kondratieva, J.-M. Goupil, L. Mariey, P. Bazin, J.-P.
Gilson, E. Travert and F. Mauge´, J. Phys. Chem. C, 2010, 114, 15661.
52 B. Guichard, M. Roy-Auberger, E. Devers, C. Legens and P. Raybaud,
Catal. Today, 2008, 130, 97.
53 N. Frizi, P. Blanchard, E. Payen, P. Baranek, C. Lancelot, M.
Rebeilleau, C. Dupuy and J.-P. Dath, Catal. Today, 2008, 130, 32.
54 B. M. Vogelaar, P. Steiner, T. F. von der Zijden, A. D. van Langeveld,
S. Eijsbouts and J. A. Moulijn, Appl. Catal., A, 2007, 318, 28.
55 E. Krebs, B. Silvi and P. Raybaud, Catal. Today, 2008, 130, 160.
56 S. Harris and R. R. Chianelli, J. Catal., 1996, 98, 17.
57 J.-F. Paul, S. Cristol and E. Payen, Catal. Today, 2008, 130, 139.
58 L. P. Nielsen, S. V. Christensen, H. Topsøe and B. S. Clausen, Catal.
Lett., 2000, 67, 81.
59 P. Raybaud, J. Hafner, G. Kresse, S. Kasztelan and H. Toulhoat, J.
Catal., 2000, 190, 128.
60 J. V. Lauritsen, J. Kibsgaard, G. H. Olesen, P. G. Moses, B.
Hinnemann, S. Helveg, J. K. Nørskov, B. S. Clausen, H. Topsøe,
E. Laegsgaard and F. Besenbacher, J. Catal., 2007, 249, 220.
61 P. Raybaud, Appl. Catal., A, 2007, 322, 76.
23 A. Y. Bunch and U. S. Ozkan, J. Catal., 2002, 206, 177.
24 A. Y. Bunch, X. Wang and U. S. Ozkan, J. Mol. Catal. A: Chem.,
2007, 270, 264.
25 A. Y. Bunch, X. Wang and U. S. Ozkan, Appl. Catal., A, 2008, 346,
96.
26 T.-R. Viljava, R. S. Komulainem and A. O. I. Krause, Catal. Today,
2000, 60, 83.
27 F. E. Massoth, P. Politzer, M. C. Concha, J. S. Murray, J. Jakowski
and J. Simons, J. Phys. Chem. B, 2006, 110, 14283.
28 O. I Senol, E.-M. Ryymin, T.-R. Viljava and A. O. I. Krause, J. Mol.
Catal. A: Chem., 2007, 277, 107.
29 Y. Romero, F. Richard and S. Brunet, Appl. Catal., B, 2010, 98, 213.
30 E.-M. Ryymin, M. L. Honkela, T.-R. Viljava and A. O. I. Krause,
Appl. Catal., A, 2010, 389, 114.
31 V. N. Bui, D. Laurenti, P. Afanasiev and C. Geantet, Appl. Catal. B,
2010, 101, 239.
62 M. Sun, A. E. Nelson and J. Adjaye, J. Catal., 2004, 226, 32.
63 H. Topsøe, B. S. Clausen, R. Candia, C. Wivel and S. Morup, J.
Catal., 1981, 68, 433.
64 H. Topsøe and B. S. Clausen, Catal. Rev., 1984, 26, 395.
65 M. Badawi, S. Cristol, J.-F. Paul and E. Payen, C. R. Chim., 2009,
12, 754.
32 V. M. L. Whiffen and K. J. Smith, Energy Fuels, 2010, 24, 4728.
33 O. I. Senol, T.-R. Viljava and A. O. I. Krause, Catal. Today, 2005,
100, 331.
34 O. I. Senol, T.-R. Viljava and A. O. I. Krause, Catal. Today, 2005,
106, 186.
66 H. Schweiger, P. Raybaud and H. Toulhoat, J. Catal., 2002, 212, 33.
67 M. Badawi, J.-F. Paul, S. Cristol and E. Payen, Catal. Commun.,
2011, 12, 901.
68 V. Rabarihoela-Rakotovao, S. Brunet, G. Perot and F. Diehl, Appl.
Catal., A, 2006, 306, 34.
35 O. I. Senol, T.-R. Viljava and A. O. I. Krause, Appl. Catal., A, 2007,
326, 236.
36 E.-M. Ryymin, M. L. Honkela, T.-R. Viljava and A. O. I. Krause,
Appl. Catal., A, 2009, 358, 42.
69 M. Egorova and R. Prins, J. Catal., 2004, 225, 417.
70 F. Besenbacher, M. Brorson, B. S. Clausen, S. Helveg, B. Hinnemann,
J. Kibsgaard, J. V. Lauritsen, P. G. Moses, J. K. Nørskov and H.
Topsøe, Catal. Today, 2008, 130, 86.
37 E. Laurent and B. Delmon, Ind. Eng. Chem. Res., 1993, 32, 2516.
38 B. S. Gevert, M. Eriksson, P. Eriksson and F. E. Massoth, Appl.
Catal., A, 1994, 117, 151.
39 E. O. Odebunmi and D. F. Ollis, J. Catal., 1983, 80, 56.
40 E. Laurent and B. Delmon, Appl. Catal., A, 1994, 109, 97.
41 F. Mauge´ and J.-C. Lavalley, J. Catal., 1992, 137, 69.
42 B. Muller, A. D. van Langeveld, J. A. Moulijn and H. Knozinger, J.
Phys. Chem., 1993, 97, 9028.
71 P. G. Moses, B. Hinnemann, H. Topsøe and J. K. Nørskov, J. Catal.,
2007, 248, 188.
72 P. G. Moses, B. Hinnemann, H. Topsøe and J. K. Nørskov, J. Catal.,
2009, 268, 201.
73 A. Logadottir, P. G. Moses, B. Hinnemann, N. Y. Topsøe, K. G.
Knudsen, H. Topsøe and J. K. Nørskov, Catal. Today, 2006, 111, 44.
74 C. Moreau, C. Aubert, R. Durand, N. Zmimita and P. Geneste, Catal.
Today, 1988, 4, 117.
43 J. Bachelier, M. J. Tilliette, M. Cornac, J.-C. Duchet, J.-C. Lavalley
and D. Cornet, Bull. Soc. Chim. Belg., 1984, 93, 743.
44 A. Travert, C. Dujardin, F. Mauge´, E. Veilly, S. Cristol, J.-F. Paul and
E. Payen, J. Phys. Chem. B, 2006, 110, 1261.
75 J. van Gestel, L. Finot, J. Leglise and J.-C. Duchet, Bull. Soc. Chim.
Belg., 1995, 104, 189.
76 A. Travert, O. V. Manoilova, A. A. Tsyganenko, F. Mauge´ and J.-C.
Lavalley, J. Phys. Chem. B, 2002, 106, 1350.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2441–2451 | 2451
©