E.-M. Ryymin et al. / Applied Catalysis A: General 389 (2010) 114–121
115
over Ni- and Co-promoted Mo catalysts with and without sulphur
vacancies.
and as mixtures. In all tests, n-decane (Merck, >98%) was added as
an internal standard. All gases were obtained with 99.999% purity
from AGA. The effect of sulphur additive was tested in both types of
experiments. The partial pressure of H2S was adjusted with gaseous
H2S (up to 2000 ppm) or via the decomposition of a liquid sul-
phur component, dimethyl disulphide (0.2–0.8 wt%), DMDS (Fluka,
≥98%).
Density functional theory (DFT) studies have provided new
insights into the structure of sulphided catalysts and the active sites
for hydrogenation. Besenbacher et al. [11] summarise the latest DFT
studies as follows: unpromoted and promoted MoS2 catalysts have
specific metallic edge sites that give rise to bright brims in scanning
tunneling microscopy (STM). These edge sites could play a role in
hydrogenation reactions and also in hetero-atom removal. The DFT
studies indicate the weak inhibiting effect of H2S on hydrogenation
reactions, since it is reported that H2S is unable to bind to the fully
coordinated brim site.
Previously, we studied the HDO of phenol and methyl hep-
tanoate over sulphided NiMo/␥-Al2O3 and CoMo/␥-Al2O3 catalysts
[12]. Phenol was used as an aromatic model component and methyl
heptanoate as an aliphatic model component for biomass-based
fuels and the reactants were tested separately. Phenol was found
to be less reactive than methyl heptanoate in the experiments car-
ried out in a flow reactor at 250 ◦C under 1.5 MPa pressure. In some
tests, H2S was added to the gas feed to maintain the activity of
the sulphided catalyst. The effect of the sulphiding agent on the
HDO of phenol was opposite to its effect on methyl heptanoate:
the HDO conversion of phenol decreased as a function of increas-
methyl heptanoate increased. From this, the HDO of the aromatic
alcohol and the aliphatic ester was concluded to occur on dissimilar
active sites.
Compositions of the reaction mixtures (T = 250 ◦C and
P = 7.5 MPa) were as follows: mixture of phenol and methyl
heptanoate (both 3 wt%) was tested with H2S (650 and 2000 ppm),
with DMDS (0.2, 0.4 and 0.8 wt%) and without any sulphur addi-
tives. In addition, phenol alone (3 wt%) and methyl heptanoate
alone (3 wt%) were tested with DMDS (0.4 wt%) and without any
sulphur additives. To determine the effect of co-reactant amount
on the product distribution, lower (2 wt%) and higher (4 wt%)
amounts of phenol and methyl heptanoate were also tested
(keeping the amount of the other co-reactant at 3 wt% and in the
absence of any sulphur additive).
Supplementary experiments were carried out with benzene
(Fluka, >99.5%), cyclohexanol (Fluka, ≥99.0%) and cyclohexene
(Aldrich, 99%) to clarify the reaction pathways. These reactants
were studied both individually and as reactant mixtures together
with phenol.
The effect of H2 pressure was evaluated at 6.0, 7.5 and 9.0 MPa
at 200 ◦C and with 0.4 wt% DMDS at 7.5 and 9.0 MPa at 250 ◦C. Tem-
perature of 200 ◦C was used for three experiments to obtain a lower
reaction rate and allow clarification of the initial reaction steps. In
addition, the stirring rate was varied from 1000 to 1750 rpm to
assess the diffusion limitations at 250 ◦C and 7.5 MPa.
Earlier studies on the HDO of mixtures [10,13,14] have been
carried out with aromatic components. Clearly, however, there are
differences in the behaviour of aromatic and aliphatic components
over sulphided catalysts and in this study we focus on the HDO
of mixtures of aromatic and aliphatic oxygen-containing model
components, i.e., phenol and methyl heptanoate. The aim is to
understand the reactions of aromatic and aliphatic structures over
a sulphided NiMo catalyst and to derive mechanistic information
that applies both to the reactants separately and in their mixtures
with and without sulphur additive.
Catalytic character of commercial ␥-Al2O3 was explored with
methyl heptanoate as a reactant. The loading of the support was
0.4 g. Pretreatment of the catalyst was similar to the one described
above for NiMo/␥-Al2O3.
2.3. Analytical methods
Liquid samples were analysed off-line with an Agilent Technolo-
gies 7890A gas chromatograph equipped with a capillary column
(HP-1, 60 m × 0.25 mm × 1 m) and a flame ionisation detector.
The products were quantified by internal standard method. Sup-
plementary qualitative analyses were obtained with an Agilent
Technologies 5975C mass spectrometer connected to the gas chro-
matograph.
2. Experimental
2.1. Reactor
The HDO experiments were performed in a 50 ml batch reactor
(Autoclave Engineers) equipped with a fixed catalyst basket and a
magnetic stirrer. The stirring rate was 1000 rpm.
2.4. Definitions
Calculations were carried out on the basis of the analysed liq-
uid samples. Molar concentration of the product is the number of
moles of product divided by the total number of moles, including
unreacted reactant and products, and multiplied by 100%. The total
amount of hydrocarbon is the sum of the products containing no
hetero atoms. Hence, complete deoxygenation is achieved when
the total amount of hydrocarbon is 100%.
In this work, terms phenol and methyl heptanoate refer to
experiments where either of the reactants was used alone together
with the solvent. Mixture, in turn, describes experiments where
both phenol and methyl heptanoate, with the solvent, were used
together. Some of the experiments were performed in the presence
of a sulphur additive, either H2S or DMDS.
2.2. Experiments
Commercial NiMo/␥-Al2O3 catalyst was crushed and sieved to a
fraction of 0.59–0.75 mm, dried at 100 ◦C for 5 h and packed (0.5 g)
into the catalyst basket. The catalyst was presulphided in situ before
the activity tests. At the start of the pretreatment period, the reac-
tor was heated up to 350 ◦C under N2 flow (atmospheric pressure),
and the catalyst was dried at this temperature for 2 h. After drying,
the catalyst was sulphided at 350 ◦C under H2S/H2 (5 mol%, atmo-
spheric pressure) flow for 2 h. Activity test temperature (200 or
250 ◦C) was achieved under N2 flow during 30 min.
The liquid reactant solution was introduced to the preheated
reactor, and the total pressure was set with H2, typically to 7.5 MPa.
The liquid added to the reactor occupied 1/3 of the reactor volume.
Duration of the reaction tests varied between 1 and 5 h. The test
procedure has been described in detail previously [15].
3. Results
3.1. Reactivity in the HDO of phenol and methyl heptanoate
The main reactants were phenol (Fluka, ≥99%) and methyl hep-
tanoate (Merck, >98%) diluted with n-dodecane (Merck, ≥99%). The
reactions of the model components were studied both separately
HDO reactions were studied at 250 ◦C over the sulphided NiMo
catalyst. Using the reactant mixtures, we first tested two H2 pres-