370
Y. Teng et al. / Journal of Catalysis 266 (2009) 369–379
where CDMDBT0 represents the 4,6-DMDBT concentration in the feed,
and CDMDBT, CTHDBT, and CHHDBT are the concentrations of 4,6-
DMDBT, TH-DMDBT, and HH-DMDBT in the HDS product.
To check whether or not MoS2 exists as a separate phase in
MoP/MCM-41(SR), a mixture of MoS2/MCM-41 and MoP/MCM-
41(R) was tested in the HDS of DBT. MoS2/MCM-41 was prepared
by sulfidation of MoO3/MCM-41 at 400 °C in 10% H2S/H2 for 3 h.
The MoP/MCM-41(R) sample was passivated with 0.5 mol% O2/Ar
at room temperature.
as the template, according to a procedure reported before [17].
MoP/MCM-41 oxidic precursors were prepared by the pore volume
impregnation method. In a typical preparation procedure, 3.6 g
(NH4)6Mo7O24ꢀ4H2O and 2.7 g (NH4)2HPO4 were dissolved in
15 mL of deionized water to form a transparent solution. A sample
of 3.0 g MCM-41 was wet impregnated with the above-mentioned
solution. The resulting slurry was kept for 12 h at room tempera-
ture, and then dried at 120 °C overnight, followed by calcination
in air at 500 °C for 3 h. In the preparation of the unsupported
MoP precursor, the above-mentioned transparent solution was
evaporated, and the resulting solid was dried at 120 °C overnight,
followed by calcination in air at 500 °C for 3 h. Our preliminary
investigation indicated that a maximum HDS activity was observed
at a MoO3 loading of 40 wt% with a P/Mo molar ratio of 1.0 in MoP/
MCM-41. Therefore, a MoO3 loading of 40 wt% and a Mo/P atomic
ratio of 1.0 were chosen for all supported catalysts in the present
study.
2.2. Catalyst characterization
X-ray diffraction (XRD) patterns of the catalysts were measured
on a Rigaku D/Max 2400 diffractometer using nickel-filtered Cu K
a
radiation at 40 kV and 100 mA. High-resolution transmission elec-
tron microscopy (HRTEM) was performed on an FEI Tecnai G2 F30
S-Twin electron microscope equipped with a field emission gun,
working at an acceleration voltage of 300 kV. Elemental composi-
tions were determined by means of an X-ray Fluorescence (XRF)
analyzer (RSR 3400X). X-ray photoelectron spectroscopy (XPS)
was performed on a Shimadzu ESCA 750 spectrometer with mono-
chromatic Mg K exciting radiation (8 kV, 30 mA) at 5 ꢂ 10ꢁ4 Pa.
The oxidic precursor of MoP, supported or unsupported, was pel-
leted, crushed, and sieved to 20–40 mesh. A sample of 0.2 g oxidic
precursor was charged into a trickle-bed reactor. The precursor
was heated according to the following program: from room temper-
ature to 400 °C at 10 °C minꢁ1 in 10% H2S/H2 and kept for 3 h, then
heated to 550 °C at 5 °C minꢁ1 in H2, to 650 °C at 1 °C minꢁ1, and
was finally kept at 650 °C for 2 h. The flow rate of H2S/H2 was
30 mL minꢁ1, and that of H2 was 150 mL minꢁ1. The supported cat-
alyst is denoted as MoP/MCM-41(SR). For comparison, supported
MoP was also prepared in a conventional in situ temperature-pro-
gramed reduction method [18]. The oxidic precursor was heated
from room temperature to 550 °C at 5 °C minꢁ1 in H2 at
150 mL minꢁ1, then to 650 °C at 1 °C minꢁ1, and finally kept at
650 °C for 2 h. The obtained catalyst is denoted as MoP/MCM-41(R).
The HDS reaction was performed in the same fixed-bed reactor
after the reduction. A solution of 0.8 wt% DBT or 4,6-DMDBT in
decalin was used as the feed. DBT was synthesized from biphenyl
and sulfur [19], and 4,6-DMDBT and decalin were of A.R. grade.
The HDS reactions of DBT and 4,6-DMDBT were conducted at
4.0 MPa. The catalytic performance was investigated either by
varying reaction temperature (300–360 °C) at constant weight
Binding energies were corrected using the
284.5 eV of adventitious carbon as a reference.
C (1s) peak at
Temperature-programed reduction (TPR) profiles of MCM-41-
supported oxidic precursors were measured on a Chembet-3000
analyzer. Before the measurement, the sample was pretreated in
He at 200 °C for 2 h. A gas mixture of 10 mol% H2/Ar was used as
the reacting agent. The TPR profiles were measured from 100 to
950 °C at 10 °C minꢁ1
.
The CO uptake was measured using the pulsed chemisorption
on a Chembet-3000 analyzer according to the literature [20]. A
passivated sample (0.1 g), which had been exposed to air during
the transfer, was re-reduced in a H2 flow to remove the passivation
layer prior to the measurement. The reactor was then cooled to
30 °C in a flow of H2. An Ar flow (40 mL minꢁ1) was used to flush
the catalyst for 30 min to achieve a clean catalyst surface. After
pretreatment, 1.25 mL pulses of 1.0 mol% CO/Ar were injected into
a flow of Ar (80 mL minꢁ1). CO pulses were repeatedly injected un-
til no further CO adsorption was observed. The CO uptake of the
sample was calculated from the accumulated differences in the
peak areas of input and output signals.
time (50 g min molꢁ1
)
or by varying weight time (12–
62 g min molꢁ1) at 300 °C. Sampling of liquid products was started
6 h after the reaction conditions had been reached. For each run,
three to five liquid samples were collected at an interval of
20 min. Both feed and products were analyzed on an Agilent-
6890+ gas chromatograph equipped with an FID using an HP-5 cap-
3. Results
illary column (5% phenyl methyl polysiloxane, 30.0 m ꢂ 320
l
m ꢂ
3.1. Preparation and characterization
0.25 m).
l
In the HDS of DBT, two sulfur-containing intermediates, tetra-
hydrodibenzothiophene (TH-DBT) and hexahydrodibenzothioph-
ene (HH-DBT), were detected in the products. HDS conversion is
defined as follows:
Fig. 1 shows the XRD patterns of MoP/MCM-41 prepared by
temperature-programed reduction (MoP/MCM-41(R)) and sulfida-
tion–reduction (MoP/MCM-41(SR)). Diffraction peaks around
27.9°, 32.2°, 43.2°, and 57.3° were observed for both MoP/MCM-
41(R) and MoP/MCM-41(SR). They are characteristic of the MoP
crystal phase (PDF 24-771) and are similar to those reported by
others [4]. The half widths of the peaks at 32.2° and 43.2° in the
patterns of MoP/MCM-41(SR) (0.58° and 0.6°) were larger than
those of MoP/MCM-41(R) (0.52° and 0.54°). According to Scherrer
equation, MoP particle sizes were estimated to be 13.9 and
15.6 nm for MoP/MCM-41(SR) and MoP/MCM-41(R), respectively.
It is therefore suggested that higher dispersion was obtained in
MoP/MCM-41(SR).
Although the supported catalyst only showed the presence of
MoP, the possibility that another sulfur-containing phase is pres-
ent in low concentrations or well dispersed on the MCM-41 cannot
be excluded. Thus, the transformation of oxidic precursors in the
course was investigated in the preparation of bulk MoP. The XRD
patterns of the oxidic precursor and after different treatments
HDS conversion ¼ ðCDBT0 ꢁ CDBT ꢁ CTHꢁDBT ꢁ CHH-DBTÞ=CDBT0 ꢂ 100%
ð1Þ
where CDBT0 represents the DBT concentration in the feed; CDBT, CTH-
DBT, and CHH-DBT are the concentrations of DBT, TH-DBT, and TH-DBT
in the product, respectively.
Similarly, two sulfur-containing intermediates, tetrahydro-
dimethyldibenzothiophene (TH-DMDBT) and hexahydrodimethyl-
dibenzothiophene (HH-DMDBT), were observed in the HDS
products of 4,6-DMDBT. Decahydrodimethyldibenzothiophene
(DH-DMDBT) was not detected in the products. Therefore, the
HDS conversion of 4,6-DMDBT is defined as follows:
HDS conversion ¼ ðCDMDBT0 ꢁ CDMDBT ꢁ CTH-DMDBT
ꢁ CHH-DMDBTÞ=CDBT0 ꢂ 100%
ð2Þ