Effect of pre-activation treatment temperature on hydrodesulfurization catalytic activity of CoMoS/KIT-6

Article abstract of DOI:10.1016/j.cattod.2019.10.019

Mesoporous silica-supported cobalt–molybdenum hydrodesulfurization catalysts have been prepared by wetness impregnation, drying and a thermal pre-treatment of impregnated support before activation. Characterization of materials was made by nitrogen adsorption, thermogravimetric analysis, Raman spectroscopy, TEM and XPS. Catalytic activity was determined by hydrodesulfurization of dibenzothiophene (DBT). The results indicate that the thermal pre-treatment carried out at the lower temperature allows the formation of smaller and less stacked active phase structures, with a higher amount of CoMoS phase.

Full text of DOI:10.1016/j.cattod.2019.10.019

Journal Pre-proof
Effect of pre-activation treatment temperature on hydrodesulfurization
catalytic activity of CoMoS/KIT-6
Karen A. Beltr a´ n, Lorena Alvarez-Contreras, Anabel D. Delgado,
C e´ sar C. Leyva-Porras, Alfredo Aguilar-Elguezabal
PII:
S0920-5861(19)30553-X
DOI:
https://doi.org/10.1016/j.cattod.2019.10.019
CATTOD 12524
Reference:
To appear in:
Catalysis Today
Received Date:
Revised Date:
Accepted Date:
16 February 2019
28 September 2019
1 October 2019
Please cite this article as: Beltr a´ n KA, Alvarez-Contreras L, Delgado AD, Leyva-Porras CC,
Aguilar-Elguezabal A, Effect of pre-activation treatment temperature on hydrodesulfurization
catalytic activity of CoMoS/KIT-6, Catalysis Today (2019),
doi: https://doi.org/10.1016/j.cattod.2019.10.019
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. 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.
©
2019 Published by Elsevier.

Effect of pre-activation treatment temperature on hydrodesulfurization catalytic
activity of CoMoS/KIT-6
*
Karen A. Beltrán, Lorena Alvarez-Contreras , Anabel D. Delgado, César C. Leyva-Porras,
Alfredo Aguilar-Elguezabal.
Centro de Investigación en Materiales Avanzados, S. C. Miguel de Cervantes No. 120,
Complejo Industrial Chihuahua, Chih., México, CP 31136.
*
Corresponding author: lorena.alvarez@cimav.edu.mx
Graphical abstract
Highlights.

CoMoS/KIT-6 catalysts with different pre-activation thermal treatments were
synthesized.

Metal loading of 16% wt. CoMo was used.
1



Lower pre-activation temperatures lead to smaller and less stacked CoMoS crystals.
Pre-activation thermal treatment is a key factor to improve catalytic activity.
Abstract
Mesoporous silica-supported cobalt–molybdenum hydrodesulfurization catalysts have been
prepared by wetness impregnation, drying and a thermal pre-treatment of impregnated
support before activation. Characterization of materials was made by nitrogen adsorption,
thermogravimetric analysis, Raman spectroscopy, TEM and XPS. Catalytic activity was
determined by hydrodesulfurization of dibenzothiophene (DBT). The results indicate that the
thermal pre-treatment carried out at the lower temperature allows the formation of smaller
and less stacked active phase structures, with a higher amount of CoMoS phase.
Keywords
Ordered mesoporous silica, supported HDS catalysts, CoMoS catalysts, KIT-6
2

1
. Introduction.
Due to the environmental damages associated to the release of sulfur compounds as a
consequence of fuels burning and the increasing demand for fuels, environmental laws have
become stricter in sulfur content in these substances, nowadays in many countries the
maximum content of sulfur must be lower than 10 ppm[1–5]. This requirement on fuels
promotes the adjustment of hydrotreatment processes on crude oil refineries, some of these
improvements include the development of more active catalysts, i.e., with modifications of
morphology or composition of supports, active phases or even the use of additives, among
other changes[6–12]. For hydrodesulfurization (HDS) processes it is known that the key in
obtaining a highly active catalyst is related to the crystalline structure, and the most accepted
model is the formation of the so-called CoMoS phase[13,14], where electronic effects
induced by the insertion of Co promoter facilitate the removal of the sulfur atom from
molecules that contain this atom in its structure, moreover Ramos et al proposed that Co also
modifies textural and structural properties of MoS
on metals sulfidation degree has been scarcely studied. Some of the important steps to
modulate the formation of sulfides of active phases are the gas flow, H S/H ratio, time of
2
slabs [15].Effect of activation treatments
2
2
treatment and temperature, although another important characteristic is the chemical structure
of precursors of active phase before activation step.
In order to obtain ultra-low sulfur content diesel, is essential to remove sulfur from refractory
molecules such as dibenzothiophenes and alkyl dibenzothiophenes, like 4,6-
dimethyldibenzothiophene, making this task a challenge [3,5,16–20]. Thus, to reach
environmental regulations, researchers have focused on the development of more effective
catalysts for deep removal of sulfur, considering also the reduction of processing costs by
lowering the severity of the reaction conditions such as hydrogen pressure and temperature.
3

The material most used as support of HDS catalyst is gamma-alumina [21–24], nevertheless
researchers have also explored other materials like nanoestructured silica (SBA-15, MCM-41
and KIT-6) [2,4,25–32]. Particularly, KIT-6 is a mesoporous silica with tridimensional
interconnected channels that makes this material a promising support for HDS catalysts. KIT-
6
has high surface area and pore diameter over 6 nm, that allows better metal dispersion and
the diffusion of large molecules through pores, moreover KIT-6 has thermal and mechanical
stability [2,4]. This work demonstrates that there is an effect of thermal treatment of
precursors once impregnated on support and previously to activation treatment, on the
activity for hydrodesulfurization, which is due to the dispersion and structure of active phases
obtained.
2
. Materials and methods
Synthesis of catalyst consists on preparation of support, impregnation of cobalt precursor on
support, after that impregnation of molybdenum precursor, then thermal treatment, and
finally, the activation step. Catalysts were synthesized with 16% wt. total metal loading (11.2
%
wt. of Mo and 4.8% wt. of Co).
2
.1. Supports Impregnation
The synthesis of KIT- 6 (mesoporous silica) was made according to the modified
methodology previously reported by Freddy Kleitz et al. [33]. Triblock copolymer (pluronic P
1
23 from Sigma Aldrich) was used as structure-directing agent. Synthesis is as follows: for a
typical synthesis, pluronic P123 (15.9 g) is dissolved in a mixture of distilled water (575 mL)
and hydrochloric acid (26.3 mL, J. T. Baker 36.5-38 %), later, butanol (19.6 mL, Sigma
Aldrich 99.8 %) is added under constant stirring at 308 K. After 1 h, tetraethyl orthosilicate
(36.6 m/L TEOS, Sigma Aldrich 98%) was added dropwise under continuous stirring for 2 h
at 308 K. The solution was then transferred to a Teflon autoclave heated at 393 K for 24 h.
4

Finally, the solid was filtered, washed with a mixture of ethanol (HYCEL 96 %) and
hydrochloric acid and dried, followed by calcination at 823 K to remove the template. Then
wetness impregnation of precursor salts was made; firstly, an appropriate amount of cobalt
nitrate 6-hydrate (from J. T. Baker 99.1% of purity) in aqueous solution was used for
o
impregnation. Subsequently, the impregnated support was dried during 2h at 60 C. In a
second step, a solution of ammonium heptamolybdate tetrahydrate ((NH
4
)
6
Mo O24, HMA,
7
from Sigma Aldrich 99.9% of purity) was used as molybdenum precursor. The obtained
sample were then calcined at 135, 225, 300, 450, or 500 °C for 3 h with a heating rate of 4
°
C/min, and the CoMo catalysts precursors were named as is shown in the Table 1.
.2. Catalysts Activation.
2
The activation of catalysts was made “ex-situ”. For this purpose, the impregnated and
thermally treated supports were placed in a quartz tube inside a tubular furnace under gas
flow of 45 mL/min of H
2
S/H
2
mixture (15 % H S). Temperature was raised from ambient to
2
o
o
4
00 C with a heat rate of 4 C/min, and kept under these conditions during 4 h. Afterwards,
the solid was transferred to a sealed bottle under a nitrogen atmosphere for later
characterization and HDS activity study.
2
2
.3. Oxides and sulfided catalysts characterization.
.3.1. X-ray diffraction (XRD).
X-ray diffractograms were obtained using a Panalytical Xpert’PRO Power Diffractometer
CuKα radiation, 0.154056 nm) operating at 35 mA, 40 kV, step size of 0.05°, and step time
of 500 s. Diffraction patterns were analyzed using standard JCPDS files.
.3.2. Nitrogen physical adsorption characterization
(
2
5

Nitrogen adsorption and desorption measurements were performed using a Quantachrome
o
o
apparatus (Autosorb-1) at –196 C after degasification of at 120 C for 12 h for oxides, while
o
support and sulfided catalyst were outgassed at 250 C for 8 h. The specific surface area
(SBET) was calculated from the BET (Brunauer-Emmett-Teller) method, and the pore-size
distribution and pore volume was obtained using the BJH (Barret-Joyner-Halenda) method.
The normalized SBET (NSBET) was calculated according to the following equation (
Equation 1) [25].
ꢀꢁꢂꢃ
표푓푐푎푡푎푙푦푠푡
표푓푠푢푝푝표푟푡]
Equation 1
푁푆퐵퐸푇 = [
(1−푦)∗ꢀꢁꢂꢃ
Where y, is the content fraction of precursors, metal oxides or metal sulfides, depending on
the step during catalyst synthesis.
2
.3.3. Transmission electron microscopy (TEM)
The morphology and microstructure of materials were determined by transmission electron
microscopy using a Hitachi ™ 7700 microscope with a tungsten filament at an accelerating
voltage of 100kV. Samples were suspended in 2-propanol and sonicated; then a drop was
added on a copper grid coated with lacey carbon. The length and stacking number of the
MoS slabs were calculated from at least 500 crystals from each catalyst. All measurements
2
were made with the software Image Pro Plus version 6.0 and the obtained average value was
used to calculate the Mo dispersion (fMo) using equations 2 and 3.
푡
∑
(6푛 ꢅ 6)
2
Equation 2
Equation 3
푖ꢆ1
ꢄ
푀표
=
푡
∑
(3푛 ꢅ 3푛 + ꢇ)
푖ꢆ1
퐿
푛 = + 0.5
ꢈ.4
6

Where t is the number of layers, n is the number of Mo atoms in one edge, and L is the length
of MoS crystals. The molybdenum dispersion (fMo) was calculated assuming that crystals of
MoS are perfect hexagons [27].
.3.4. X-ray photoelectron spectroscopy (XPS)
X-Ray Photoelectron Spectroscopy (XPS) measurements were performed with a Thermo
2
2
2
−9
Scientific Escalab 250Xi spectrometer, samples were analyzed under vacuum (<10 mbar)
with a monochromatic Al Kα (1486.6 eV) X-ray source and a spot size of 650 µm, and the
binding energies (BE) were recorded taking C1s (284.8 eV) as reference. The Mo3d, Co2p
and S2p spectra were analyzed applying a Shirley background subtraction and mixed
Gaussian-Lorentzian functions. Surface atomic ratios were calculated from the peak area
ratios normalized by their atomic sensitivity factors. For S2s the binding energies and peak
areas were constrained by the Equation 4, as previously reported Chen et al. [34].
ꢉꢊ(푆ꢋꢌ) = ꢉꢊ ꢍ푆ꢋꢎꢏ ꢐ + 6ꢑ.3, 퐴ꢒ푒ꢓ(푆ꢋꢌ) = 퐴ꢒ푒ꢓ(푆 ꢋꢎ) ∗ 0.66
⁄
2
Equation 4
The molybdenum, cobalt and sulfur surface atomic concentration [C(Mo), C(Co) and C(S)]
for each species was calculated through the Equation 5.
ꢕꢖ
⁄
ꢀ
퐹ꢖ
퐶(ꢔ) =
∗ ꢇ00
푠
ꢕꢖ
ꢗ
∑
⁄
ꢖꢘꢙ
ꢀ퐹ꢖ
Equation 5.
Where C(i)
s
is the species atomic concentration, A
i
is the measured spectra area from the
different species, SF
i
is the sensitivity factor of each element.
On the other hand, the percentages of Mo species (Mo +4, +5 or +6) were calculated using the
Equation 6.
퐴푀표ꢝꢞ
4
ꢜ
ꢚꢛ푆 ꢛꢚꢛ (%) =
∗ ꢇ00
2
퐴푀표ꢝꢞ + 퐴푀표ꢟꢞ + 퐴푀표ꢠꢞ
2

Equation 6.
2
+
Similarly, the Co species (CoMoS, Co
9
S
8
and Co ) were obtained by Equation 7.
퐴ꢡ표푀표ꢀ
퐶ꢛꢚꢛ푆(%) =
∗ ꢇ00
퐴ꢡ표푀표ꢀ + 퐴ꢡ표9ꢀ8 + 퐴ꢡ표ꢢꢞ
Equation 7.
2
.3.5. Raman spectroscopy
Raman spectra were obtained using a Renishaw inVia Raman Microscope, the single-line
laser at 244 nm from an Ar ion laser was used as the Raman excitation source for recording
the spectra, the samples were placed in an aluminum capsule to perform the analysis and data
acquisition was collected after 30 min of laser exposure time. The spectrum was normalized,
and the background was removed for further comparison of intensities.
2
.4. Catalytic performance tests
HDS reaction was performed in a high-pressure batch reactor (Parr model 4842). Activity
tests were made adding catalyst (0.5 g) into the reactor vessel with a solution of DBT (Sigma
Aldrich 98 %, 119.3 mmol/L) in decahydronaphthalene (Sigma-Aldrich, 99%, 4250 ppm of
o
sulfur). The reaction was carried out at 350 C and under 3.4 MPa of H
2
pressure for 5 h and
maintaining constant stirring of 600 rpm. Liquid samples were taken every 30 minutes during
the 5 h. Samples were analyzed in a gas chromatograph coupled with mass spectrometer from
Shimadzu, GCMS-QP2010 SE equipped with a capillary column RX-1-5Sil MS, length of 30
m and 0.25 mm of diameter. The catalytic performance of the catalysts towards the HDS of
DBT was measured by calculating the kinetic rate constants, assuming a pseudo-first order
kinetics. The calculations were based on the changes in the concentration of DBT present in
the aliquots taken from the reaction system at intervals of 30 min starting when the
3

temperature of the reactor reached 350 °C. The DBT conversion was calculated as follows
35]:
[
ꢡꢣꢁꢃꢤ−ꢡꢣꢁꢃꢖ
ꢡꢣꢁꢃꢤ
푋퐷퐵푇
=
Equation 8
-1
where 퐶퐷퐵푇ꢤ is initial molar concentration of DBT (molL ) and 퐶
is DBT molar
퐷퐵푇ꢖ
concentration at a determined time i.
Reaction rate constant was calculated using a pseudo-first order kinetic as follows [35]:
ꢒ푎 = 푘퐶퐷퐵푇ꢖꢥ푘퐶퐷퐵푇ꢖ = 퐶퐷퐵푇ꢤ(ꢇ ꢅ 푋퐷퐵푇
)
Equation 9
The pseudo-first order rate constant is obtained from [35]:
푘ꢦ = ꢅꢧ푛(ꢇ ꢅ 푋퐷퐵푇
)
Equation 10
-1
-1
-1
where r is the rate constant (mols g ), k is rate constant (s ), t is time, and XDBT is DBT
a
conversion.
The HYD/HDS selectivity is calculated using the following equation:
ꢡꢨꢩꢁꢜꢡꢣꢨꢩꢜꢡꢃꢩꢣꢁꢃ
푆퐻푌퐷⁄
퐻퐷ꢀ
=
Equation 11
ꢡꢁ푃
Where CCHB, CDCH, CTHDBT and CBP are the CHB, DCH, THDBT and BP concentrations in
reaction product, respectively
3
3
3
. Results and discussion
.1 Characterization of KIT-6 support and catalysts.
.1.1. X-ray diffraction analysis.
To confirm the KIT-6 structure, low angle X-ray diffraction was taken from synthesized
supports. Figure 1 shows the KIT-6 low angle X-ray diffraction pattern. As can be seen, XRD
pattern shows a sharp intense peak at 2= 0.95 corresponding to (211) plane and 2= 1.1 a
hump attributed to (220) plane. The XRD pattern clearly indicates that material presents an
ordered mesoporous structure and belongs to bicontinuous cubic space group Ia3d [4, 33, 36].
4

The unit cell parameter a0 of calcined sample was also calculated, getting values of 227.7 Å
211) and 195.7 Å (220), which are in good agreement with the reported in the literature and
(
suggests that mesoporous silica is KIT-6 with body centered cubic symmetry [4].
Powder X-ray diffractograms of the pre-activated catalysts O-135, O-225, O-300, O-450 and
O-500 are presented in Figure 1 (b). All samples showed the broad signal between 16° and
3
0° 2θ typical for amorphous silica [36, 37]. The materials calcined at 450 and 500 °C show
major broad diffraction peaks at 2θ values of 23.3, 26.4 and 27.2°. The presence of these
peaks could be attributed to the Mo species and the peaks intensity is related to degree of
crystallinity in each material, which increases at a higher temperature. The diffraction peaks
at 2θ value of 23.4, 25.7 and 27.3° correspond to (110), (040) and (021) planes of the
orthorhombic crystalline phase of MoO
3
(ICDD PDF No. 00-076-1003) [38,39]. The
or even MoO . Therefore, the
additional peaks could be associated to Mo forms like MoO
3
2
presence of these peaks depicts an additional evidence of the successful impregnation of the
KIT-6 with the metal species. There are no evident peaks that indicate the Co presence; this
could be explained with the low content of Co with respect to Mo [40]. X-ray diffraction
patterns of activated catalysts C-135, C-225, C-300, C-450 and C-500 are showed in Figure 1
(
c). In all catalysts, there are four major broad diffraction peaks at 2θ values of 14.5, 33.4,
8.2 and 58.5° which correspond to (002), (100), (103) and (110) plane of the MoS
hexagonal-phase (ICDD PDF No. 00-037-1492) [36,41]. As expected, the intensity of the
MoS peaks increased with the increase of calcination temperature, indicating the formation
of more agglomerated MoS nano-particles at higher calcination temperature [41, 42].
.1.2. Textural characterization.
3
2
2
2
3
2
A surface area of 647 m /g was determined for KIT-6 by nitrogen adsorption, and as expected
(Table 2), this support presents type IV adsorption-desorption isotherm corresponding to
5

mesoporous materials with H1 hysteresis loops for cylindrical pores. KIT-6 has a pore
volume of 1.3 cc/g and a pore diameter of 69 Å, which was calculated by the BJH method.
The pre-activated catalysts, i.e., supports impregnated and thermally treated, shown the same
kind of adsorption-desorption isotherms and hysteresis, indicating that the mesoporous
structure remains unaltered (Figure 2), however, the specific surface area presents a
2
significant reduction of around 200 m /g (Table 2). Nevertheless, among the pre-activated
catalysts an increase in specific surface area was observed as temperature of thermal
treatment increases, this could be explained as result of precursor degradation at higher
2
temperatures, thus samples O-450 and O-500 has better textural properties, 430 m /g and 0.9
cc/g for SBET and pore volume, respectively, due to the release of ammonia and water from
cobalt and molybdenum precursors molecules, reducing obstructions into available pores of
support.
The average pore diameter from pre-activated and sulfided samples remains close to the pure
support value, around 66 Å as shown Table 2, which allows assuming that pores were not
internally covered by deposition of multi-layers of precursors, and surface area reduction is
mainly due to complete obstruction of some pores. Even more, the same behaviour was
detected in specific surface area and pore volume. To determine the specific surface area
reduction from the support to pre-activated materials and sulfided catalysts, the NSBET was
calculated from
Equation 1. For each NSBET, (1-y) value must
be calculated considering on whether dispersed cobalt and molybdenum are precursors oxides
or sulfided species, the NSBET values are summarized in
Table 2. The obtained values are between 0.77 to 0.82 for pre-activated materials, while after
activation NSBET values are from 0.67 to 0.75. Accordingly, to López-Mendoza et al. [32],
values near to 0.9 are characteristic of high-dispersed catalysts, while values below 0.6
6

indicates that a fraction of pores was obstructed by species dispersed on support. The
reduction of NSBET values after activation, despite the higher contribution of surface area
from higher fraction of support (1-y), indicates that sulfided species obstruct a higher amount
of surface originally available on pores of KIT-6.
3
.1.3 Raman Spectroscopy.
The collected Raman spectra of pre-activated samples are shown in Figure 3, which were
normalized in order to compare the intensity of signals among the studied materials. All
-1
samples exhibit a broad peak at 970 cm attributed to large polymolybdate clusters such as
4
-
Mo
at 875.7 cm associated to Mo-O-Co stretching vibrations in CoMoO
cobalt ions are placed in an octahedral environment). These latter species are often related to
a decrease in HDS catalytic activity, since is widely known that CoMoO cannot be easy
8
O26 (Mo=O Raman band of octahedral coordination) [43,44] and also present the signal
-1
4
species (in which
4
-1
transformed into the CoMoS phase [44]. Additionally, the signal located at 690 cm related
to Co-O stretching vibrations [45] in the oxide state, have been detected in all samples. At
-
1
1
,065cm a small signal due to nitrate residue from the cobalt precursor has been observed,
this signal is not detected at highest temperature treatments, since nitrates decompose around
o
3
3
3
00 C.
.2. Characterization of sulfided catalysts
.2.1. Active Phase Morphology.
The TEM study is focused on the study of the active phase obtained after activation treatment
of catalyst supported on KIT-6. Figure 4 shows TEM micrographs of materials, showing the
typical morphology of MoS
2
fringes, the insets present the length distribution of MoS
2
crystals, with Gaussian curve fit. The length and stacking average are summarized in Table 3,
both length and stacking tend to increase as the drying temperature increases. For catalyst C-
7

1
35 the average stacking degree is the smallest (3.15 layers) of the catalysts series with an
average length of 4.23 nm, whereas the longest crystal corresponds to C-500 with 5.19 nm
and 3.9 MoS sheets. The average length and stacking are associated to dispersion of Mo
atoms available to carry out HDS reaction. The average fraction of Mo atoms on the edge
2
surface (fMo
)
is
a
Mo dispersion indicator, which is calculated from
Equation 3. The Table 3 shows the obtained
Equation 2 and
f
Mo values, and as can be seen, higher dispersion is obtained at lower temperature of thermal
treatment, except for C-450.
.2.2.XPS analysis of activated catalysts.
3
The surface chemical states of metal species on the sulfided catalysts were investigated by
XPS spectroscopy. The Mo3d spectra of catalysts are presented in Figure 5, showing the
doublets corresponding to the contributions of Mo 3d5/2 and Mo 3d3/2 of different Mo species
+4
+5
+6
such as sulfided (Mo ), partially sulfided (Mo ) and oxidized (Mo ) [43,46,47],
additionally, a single signal is also presented from S2s at 226.0 eV. The spectra
deconvolution shows the presence of Mo species at 228.4 and 232.2 eV, which correspond to
4
+
4+
Mo 3d5/2 and Mo 3d3/2, respectively. Doublets at 230.9 and 234.1 eV correspond to MoS
species; while at 234.3 and 237.2 eV the signals are associated with MoO polymorphs. The
Co2p spectra were also collected, and the corresponding deconvolution are presented in
Figure 6. The contribution from different Co species was observed, Co at 777.8 eV,
CoMoS at 780.2 eV, being the latter commonly considered responsible of most of HDS
x
O
y
x
S
9 8
+2
activity, therefore it is the most expected species, and finally, Co from CoO at 782.9 eV
x
+2
and the satellite of Co . From XPS spectra, differences on signals intensity were observed.
Table 4 shows molybdenum and cobalt percentage present in the catalysts. The main
+4
difference is related whit molybdenum composition (specifically Mo ), obtaining 36 y 25 %
for C-500 and C-135, respectively. Similar behaviour is observed from CoMoS content,
8

reaching 41 % for C-500, and 33 % for C-135 catalyst. The catalyst with highest sulfidation
degree is the calcined at 500 °C, this temperature could improve the formation of MoS
hence the mayor formation of CoMoS phase.
2
and
3
.2.3. Hydrodesulfuration of Dibenzothiophene.
The effect of calcination temperature over the hydrodesulfurization activity of the catalysts
C-135, C-225, C-300, C-450 and, C-500 was studied through the HDS reaction, using DBT
as a model molecule. DBT conversion, reaction rate constant and selectivity were calculated
using equations 8, 10 and 11, respectively; the results are presented in Table 5. A reduction in
the DBT conversion was observed as follows 71% (C-135) > 65% (C-225) > 62% (C-300) >
6
1% (C-450) > 58% (C-500). According to the literature [1–4], the HDS of DBT occurs
through two desulfurization parallel routes: direct desulfurization (DDS) and hydrogenation
HYD). On DDS path, sulfur is removed by hydrogenolysis of C-S bond without aromatic
ring hydrogenation producing biphenyl (BP). On the other hand in the hydrogenation route
HYD), aromatic ring saturation is required before the desulfuration step, producing
(
(
intermediate compounds such as tetra hydrodibenzothiophene (THDBT). Once sulfur is
removed, then cyclohexylbenzene (CHB) and dicyclohexyl (DCH) are generated. The DDS
was the main route observed for studied catalysts, since the principal product detected was
biphenyl (BP). In Table 5 the correlation between the drying temperature and selectivity is
summarized, no significant effect was detected over selectivity on the catalyst HDS, since the
selectivity values were comparable (0.35-0.36). According to the HDS catalytic results, C-
-7
-7
1
35 has the highest reaction rate constant 21x10 mol/s•g, and the lowest is 15x10 mol/s•g
for C-500 catalyst. Considering these results, the increment in the temperature of pre-
activation thermal treatment leads to a reduction of the catalytic activity of the catalyst.
3
.3. Discussion
9

According to catalytic activity results, the best HDS performance was obtained for the
catalyst that was thermally treated at 135 °C, this enhanced performance was due to the
combination of several factors such as a good Mo dispersion (fMo=0.28), smaller MoS
2
crystals (4 nm) and low stacking degree (3 slabs), as well as to have 95 % of MoS promoted
2
-7
with cobalt, and those characteristics allowed to obtain a rate constant of 21x10 mol/g•s. On
the other hand, the catalyst C-500 showed the lowest HDS activity, the poor performance
could be attributed to a poor Mo dispersion (fMo=0.23) derived from the average length of
crystals (5 nm), despite all the other characteristics are very close to the values of C-135.
Raman spectra for all catalyst presented a small signal of CoMoO
4
, which is not desirable in
264+. These results
HDS catalyst synthesis; the strongest signal shown in the spectra was Mo O
8
along with Mo dispersion and the ratios of CoMoS/MoS
active catalyst.
2
are the best prediction to a highly
In order to explain more profoundly the differences observed in the obtained materials, a
thermogravimetric analysis of the Mo precursor (HMA) used for the impregnation of the
catalysts was carried out. The thermogram obtained is shown in Figure 7. Total weight loss of
o
1
8.6% can be observed as a result of calcination until 500 C. The HMA precursor
degradation is carried out in three steps, the first step occurs with a 7% weight loss in which
ammonia is released; according to the work of Said et al. [47,48], second and third steps
consist on the elimination of chemically bound water, with weight losses of 4 and 7.6% for
o
each case. The degradation reactions are carried out at 110, 210 and 290 C. This allows
considering that the present species in materials treated at low temperatures are related to
high coordination molybdenum-hydrated molecules, while the MoO species start to appear
3
o
o
after 210 C until becoming the only phase at 500 C. The decrease in the magnitude of the
rate constant as the calcination temperature increases is a clear indication that calcination at
higher temperatures reduces the activity of the catalysts. The observed phenomenon could be
1
0

attributed to the higher crystallinity degree of MoO
evident in the XRD patterns (Figure 1) and therefore, sulfidation becomes more difficult.
. Conclusion.
3
species at higher temperatures, which is
4
In this work, CoMo catalysts, supported over mesoporous silica (KIT-6) were prepared
through a thermal treatment before activation of impregnated cobalt and molybdenum
precursors on KIT-6 support, and the resulting efficiency was measured in DBT
desulfurization. According to HDS catalytic results, the best catalyst was obtained at the
o
lower temperature used for the thermal treatment, before activation (135 C); higher
temperatures used for other catalysts do not improve the activity results. Thus, thermal
treatment after drying step and before activation become an easy way to improve HDS
activity of a catalyst. Raman study of species formed during pre-activation thermal treatment
and XPS and TEM analyses of active phases formed after activation treatment allows
understanding catalytic properties of studied materials from the information of species
formed in each step and the dispersion and structures of CoMoS active phase dispersed on
KIT-6 support.
Declaration of interests
☒
The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
The authors declare the following financial interests/personal relationships which may be
☐
considered as potential competing interests:
Acknowledgement.
The authors gratefully acknowledge the technical support of Raúl Ochoa, Luis Silva, Luis de
la Torre and Dr. Victor Ramos. The authors appreciatively acknowledge the Mexican Council
of Science and Technology (CONACYT) for support through the projects “Laboratorio
Nacional de Micro y Nanofluídica“ no. 293442 and CONACYT-SENER-Hidrocarburos No.
1
20135.
1
1

References
[1] I. Babich, Science and technology of novel processes for deep desulfurization of oil
refinery streams: a review⋆ , Fuel. 82 (2003) 607–631. doi:10.1016/S0016-
2
361(02)00324-1.
[
2] H. Wu, A. Duan, Z. Zhao, D. Qi, J. Li, B. Liu, et al., Preparation of NiMo/KIT-6
hydrodesulfurization catalysts with tunable sulfidation and dispersion degrees of active
phase by addition of citric acid as chelating agent, Fuel. 130 (2014) 203–210.
doi:10.1016/j.fuel.2014.04.038.
[
[
[
3] T. Fujikawa, H. Kimura, K. Kiriyama, K. Hagiwara, Development of ultra-deep HDS
catalyst for production of clean diesel fuels, Catal. Today. 111 (2006) 188–193.
doi:10.1016/j.cattod.2005.10.024.
4] K. Soni, B.S. Rana, A.K. Sinha, A. Bhaumik, M. Nandi, M. Kumar, et al., 3-D ordered
mesoporous KIT-6 support for effective hydrodesulfurization catalysts, Appl. Catal. B
Environ. 90 (2009) 55–63. doi:10.1016/j.apcatb.2009.02.010.
5] C. Song, An overview of new approaches to deep desulfurization for ultra-clean
gasoline, diesel fuel and jet fuel, Catal. Today. 86 (2003) 211–263.
doi:10.1016/S0920-5861(03)00412-7.
[
6] J. Ramirez, F. Sanchez-Minero, Support effects in the hydrotreatment of model
molecules, Catal. Today. 130 (2008) 267–271. doi:10.1016/j.cattod.2007.10.103.
7] A. Rochet, B. Baubet, V. Moizan, E. Devers, A. Hugon, C. Pichon, et al., Influence of
the Preparation Conditions of Oxidic NiMo/Al2O3 Catalysts on the Sulfidation
Ability: A Quick-XAS and Raman Spectroscopic Study, J. Phys. Chem. C. 119 (2015)
[
2
3928–23942. doi:10.1021/acs.jpcc.5b06219.
[8] M. Breysse, P. Afanasiev, C. Geantet, M. Vrinat, Overview of support effects in
hydrotreating catalysts, Catal. Today. 86 (2003) 5–16. doi:10.1016/S0920-
1
2

5
861(03)00400-0.
[
9] H. YIN, T. ZHOU, Y. LIU, Y. CHAI, C. LIU, Study on the structure of active phase in
NiMoP impregnation solution using Laser Raman spectroscopy II. Effect of organic
additives, J. Fuel Chem. Technol. 39 (2011) 109–114. doi:10.1016/S1872-
5
813(11)60012-1.
[
10] Y. Espinoza-Armenta, J. Cruz-Reyes, F. Paraguay-Delgado, M. Del Valle, G. Alonso,
S. Fuentes, et al., CoMoW sulfide nanocatalysts for the HDS of DBT from novel
ammonium and alkyltrimethylammonium-thiomolybdate-thiotungstate-cobaltate (II)
precursors, Appl. Catal. A Gen. 486 (2014) 62–68. doi:10.1016/j.apcata.2014.08.017.
11] F. Trejo, M. Rana, J. Ancheyta, CoMo/MgO–Al2O3 supported catalysts: An
alternative approach to prepare HDS catalysts, Catal. Today. 130 (2008) 327–336.
doi:10.1016/j.cattod.2007.10.105.
[
[
[
[
[
[
12] S. a. Hanafi, M.S. Mohamed, Recent Trends in the Cleaning of Diesel Fuels via
Desulfurization Processes, Energy Sources, Part A Recover. Util. Environ. Eff. 33
(2011) 495–511. doi:10.1080/15567030903058360.
13] R.R. Chianelli, A.F. Ruppert, S.K. Behal, B.H. Kear, A. Wold, R. Kershaw, The
reactivity of MoS2 single crystal edge planes, J. Catal. 92 (1985) 56–63.
doi:10.1016/0021-9517(85)90236-2.
14] R.R. Chianelli, G. Berhault, B. Torres, Unsupported transition metal sulfide catalysts:
1
00 years of science and application, Catal. Today. 147 (2009) 275–286.
doi:10.1016/j.cattod.2008.09.041.
15] M. Ramos, F. Galindo-Hernández, I. Arslan, T. Sanders, J.M. Domínguez, Electron
tomography and fractal aspects of MoS2 and MoS2/Co spheres, Sci. Rep. 7 (2017)
1
2322. doi:10.1038/s41598-017-12029-8.
16] C. Kwak, J.J. Lee, J.S. Bae, K. Choi, S.H. Moon, Hydrodesulfurization of DBT, 4-
1
3

MDBT, and 4,6-DMDBT on fluorinated CoMoS/Al2O3 catalysts, Appl. Catal. A Gen.
00 (2000) 233–242. doi:10.1016/S0926-860X(00)00635-9.
2
[17] M. EGOROVA, R. PRINS, The role of Ni and Co promoters in the simultaneous HDS
of dibenzothiophene and HDN of amines over Mo/γ-Al2O3 catalysts, J. Catal. 241
(
2006) 162–172. doi:10.1016/j.jcat.2006.04.011.
18] T.C. Ho, Deep HDS of diesel fuel: chemistry and catalysis, Catal. Today. 98 (2004) 3–
8. doi:10.1016/j.cattod.2004.07.048.
[
[
[
1
19] R. Shafi, G.J. Hutchings, Hydrodesulfurization of hindered dibenzothiophenes: an
overview, Catal. Today. 59 (2000) 423–442. doi:10.1016/S0920-5861(00)00308-4.
20] A. Stanislaus, A. Marafi, M.S. Rana, Recent advances in the science and technology of
ultra low sulfur diesel (ULSD) production, Catal. Today. 153 (2010) 1–68.
doi:10.1016/j.cattod.2010.05.011.
[
[
[
21] J.A. Mendoza-Nieto, F. Robles-Méndez, T.E. Klimova, Support effect on the catalytic
performance of trimetallic NiMoW catalysts prepared with citric acid in HDS of
dibenzothiophenes, Catal. Today. 250 (2015) 47–59. doi:10.1016/j.cattod.2014.05.002.
22] M.J. Girgis, B.C. Gates, Reactivities, reaction networks, and kinetics in high-pressure
catalytic hydroprocessing, Ind. Eng. Chem. Res. 30 (1991) 2021–2058.
doi:10.1021/ie00057a001.
23] P. a. Nikulshin, V. a. Salnikov, A.V. Mozhaev, P.P. Minaev, V.M. Kogan, A. a.
Pimerzin, Relationship between active phase morphology and catalytic properties of
the carbon–alumina-supported Co(Ni)Mo catalysts in HDS and HYD reactions, J.
Catal. 309 (2014) 386–396. doi:10.1016/j.jcat.2013.10.020.
[
24] D. Laurenti, B. Phung-Ngoc, C. Roukoss, E. Devers, K. Marchand, L. Massin, et al.,
Intrinsic potential of alumina-supported CoMo catalysts in HDS: Comparison between
γc, γT, and δ-alumina, J. Catal. 297 (2013) 165–175. doi:10.1016/j.jcat.2012.10.006.
1
4

[
25] R. Huirache-Acuña, B. Pawelec, C.V. Loricera, E.M. Rivera-Muñoz, R. Nava, B.
Torres, et al., Comparison of the morphology and HDS activity of ternary Ni(Co)-Mo-
W catalysts supported on Al-HMS and Al-SBA-16 substrates, Appl. Catal. B Environ.
1
25 (2012) 473–485. doi:10.1016/j.apcatb.2012.05.034.
[
26] M. a. Guzmán, R. Huirache-Acuña, C.V. Loricera, J.R. Hernández, J.N. Díaz de León,
J. a. de los Reyes, et al., Removal of refractory S-containing compounds from liquid
fuels over P-loaded NiMoW/SBA-16 sulfide catalysts, Fuel. 103 (2013) 321–333.
doi:10.1016/j.fuel.2012.07.005.
[27] O.Y. Gutiérrez, T. Klimova, Effect of the support on the high activity of the
(Ni)Mo/ZrO2–SBA-15 catalyst in the simultaneous hydrodesulfurization of DBT and
4
,6-DMDBT, J. Catal. 281 (2011) 50–62. doi:10.1016/j.jcat.2011.04.001.
[
28] D. Valencia, T. Klimova, Citric acid loading for MoS2-based catalysts supported on
SBA-15. New catalytic materials with high hydrogenolysis ability in
hydrodesulfurization, Appl. Catal.
doi:10.1016/j.apcatb.2012.09.006.
B
Environ. 129 (2013) 137–145.
[
[
[
29] T.E. Klimova, D. Valencia, J.A. Mendoza-Nieto, P. Hernández-Hipólito, Behavior of
NiMo/SBA-15 catalysts prepared with citric acid in simultaneous hydrodesulfurization
of dibenzothiophene and 4,6-dimethyldibenzothiophene, J. Catal. 304 (2013) 29–46.
doi:10.1016/j.jcat.2013.03.027.
30] S. Badoga, K.C. Mouli, K.K. Soni, A.K. Dalai, J. Adjaye, Beneficial influence of
EDTA on the structure and catalytic properties of sulfided NiMo/SBA-15 catalysts for
hydrotreating of light gas oil, Appl. Catal. B Environ. 125 (2012) 67–84.
doi:10.1016/j.apcatb.2012.05.015.
31] M.Á. Calderón-Magdaleno, J.A. Mendoza-Nieto, T.E. Klimova, Effect of the amount
of citric acid used in the preparation of NiMo/SBA-15 catalysts on their performance
1
5

in HDS of dibenzothiophene-type compounds, Catal. Today. 220–222 (2014) 78–88.
doi:10.1016/j.cattod.2013.06.002.
[
32] M.A. López-Mendoza, R. Nava, C. Peza-Ledesma, B. Millán-Malo, R. Huirache-
Acuña, P. Skewes, et al., Characterization and catalytic performance of Co-Mo-W
sulfide catalysts supported on SBA-15 and SBA-16 mechanically mixed, Catal. Today.
2
71 (2016) 114–126. doi:10.1016/j.cattod.2015.07.049.
[
33] F. Kleitz, S. Hei Choi, R. Ryoo, Cubic Ia3d large mesoporous silica: synthesis and
replication to platinum nanowires, carbon nanorods and carbon nanotubesElectronic
supplementary information (ESI) available: TEM images of mesoporous cubic silica
and Pt networks, XRD patterns during forma, Chem. Commun. (2003) 2136.
doi:10.1039/b306504a.
[34] W. Chen, X. Long, M. Li, H. Nie, D. Li, Influence of active phase structure of
CoMo/Al2O3catalyst
hydrodearomatization,
on
the
selectivity
Today.
of
hydrodesulfurization
and
Catal.
292 (2017) 97–109.
doi:10.1016/j.cattod.2016.09.029.
35] I. Vázquez, A. López, G. Berhault and A. Guevara, Effect of support on the acidity of
NiMo/Al -MgO and NiMo/TiO -Al catalysts and on the resulting competitive
[
2
O
3
2
2
O
3
hydrodesulfurization/hydrodenitrogenation reactions, Fuel. 236 (2019) 55-64.
doi:10.1016/j.fuel.2018.08.053.
[36] A.D. Delgado, L. Alvarez Contreras, K.A. Beltrán, A. Aguilar Elguezabal, Effect of the
SiO support morphology on the hydrodesulfurization performance of NiMo catalysts,
2
J. Mater. Res. 33 (2018) 3646–3655. doi:10.1557/jmr.2018.382
[
37] F.Kleitz, S. H. Choi and R. Ryoo, “Cubic Ia3d large mesoporous silica: synthesis and
replication to platinum nanowires, carbon nanorods and carbon nanotubes,” Chemical
communications (Cambridge, England), (2003) 2136–2137. doi:10.1039/b306504a
1
6

[
38] F. Méndez, G. Ascención, M. Mota, I. Lee, X. Bokhimi and T.E. Klimova, “NiMo
catalysts supported on Al, Nb, Ti or Zr-containing MCM-41 for dibenzothiophene
hydrodesulfurization” Catal. Today, 2010 (2018). doi:10.1016/j.cattod.2018.03.039
39] International Center for Diffraction Data, PCPDFWIN v.2.02. PDF-2 Data Base,
Newtown Philadelphia, (1995).
[
[
[
40] T. C. Ho, “Deep HDS of diesel fuel: chemistry and catalysis”, Catal. Today, 98 (2008)
3
-18. doi:10.1016/j.cattod.2004.07.048
41] C. Yin, L Zhao, Z. Bai, H. Liu, Y Liu and C. Liu, " A novel porous ammonium nickel
molybdate as the catalyst precursor towards deep hydrodesulfurization of gas oil.
"
Fuel, 107 (2013) 873−878. doi:10.1016/j.fuel.2013.02.001
[
[
[
[
[
42] H. Liu, C. Yin, B. Liu, X. Li, Y. Li, Y. Chai and C. Liu, “Effect of Calcination
Temperature of Unsupported NiMo Catalysts on the Hydrodesulfurization of
Dibenzothiophene”, Energy and Fuels, 28 (2014) 2429-2436. doi:10.1021/ef500097u
43] L. Peña, D. Valencia, T. Klimova, CoMo/SBA-15 catalysts prepared with EDTA and
citric acid and their performance in hydrodesulfurization of dibenzothiophene, Appl.
Catal. B Environ. 147 (2014) 879–887. doi:10.1016/j.apcatb.2013.10.019.
44] R. Nava, R.A. Ortega, G. Alonso, C. Ornelas, B. Pawelec, J.L.G. Fierro, CoMo/Ti-
SBA-15 catalysts for dibenzothiophene desulfurization, Catal. Today. 127 (2007) 70–
8
4. doi:10.1016/j.cattod.2007.02.034.
45] P. Mazoyer, C. Geantet, F. Diehl, S. Loridant, M. Lacroix, Role of chelating agent on
the oxidic state of hydrotreating catalysts, Catal. Today. 130 (2008) 75–79.
doi:10.1016/j.cattod.2007.07.013.
46] H. Yuan, Qiherima, G.-T. Xu, H.-F. Li, L.-J. Lu, Study of oxidic and sulfided selective
hydrodesulfurization catalysts for gasoline using Raman spectroscopy, Chinese Chem.
Lett. 24 (2013) 1041–1044. doi:10.1016/j.cclet.2013.07.009.
1
7

[
47] E. Rodríguez-Castellón, A. Jiménez-López, D. Eliche-Quesada, Nickel and cobalt
promoted tungsten and molybdenum sulfide mesoporous catalysts for
hydrodesulfurization, Fuel. 87 (2008) 1195–1206. doi:10.1016/j.fuel.2007.07.020.
48] a. F.. Sanders, A.. de Jong, V.H.. de Beer, J. a. . van Veen, J.. Niemantsverdriet,
Formation of cobalt–molybdenum sulfides in hydrotreating catalysts: a surface science
approach, Appl. Surf. Sci. 144–145 (1999) 380–384. doi:10.1016/S0169-
[
4
332(98)00831-9.
1
8

Figure Captions
Figure 1. X-Ray diffraction spectra of low angle support (a), pre-activated catalysts (b) and
sulfided catalysts (c).
1
9

Figure 2. Adsorption-desorption isotherms of support and catalysts in the pre-activated (a)
and sulfided state (c), pore size distribution of support and pre-activated catalysts (b) and
sulfided catalysts (d).
2
0

Figure 3. Deconvoluted Raman spectra of pre-activated samples, a) O-135, b) O-235, c) O-
00, d) O-450 and e) O-500.
3
2
1

2
2

Figure 4. TEM micrographics of sulfided catalysts, a) C-135, b) C-225, c) C-300, d) C-450
and e) C-500. Crystal length distribution (inset)
Figure 5. XPS spectra of Mo3d of sulfides catalysts, a) C-135, b) C-225, c) C-300, d) C-450,
e) C-500.
2
3

Figure 6.XPS of Co2p of sulfided catalysts a) C-135, b) C-225, c) C-300, d) C-450, e) C-500.
Figure 7. Thermogravimetric analysis from HMA precursor.
2
4