Incorporation of the precursors of Mo and Ni oxides directly into the reaction mixture of sol-gel prepared γ-Al2O3-ZrO2 supports - Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization

Article abstract of DOI:10.1016/j.cattod.2014.10.035

Mo and NiMo oxides supported on γ-Al2O3-ZrO2 were successful one-pot synthesized and characterized by XRF, XRD, N2 sorption, UVVISDRS and Zeta Potential. The sulfide catalysts were evaluated between 240 and 300 °C in the thiophene HDS, used as a model reaction. XRD data evidenced the presence of γ-Al₂O₃ in the obtained solids, with its crystallinity being reduced with the incorporation of Zr and metallic species. XRD data also indicated that Zr, Ni and Mo were highly dispersed on the support. N₂ sorption and UVVISDRS analyses showed that the one-pot synthesis procedure led to Mo and NiMo supported solids with high SBET areas and to Mo species with a low polymerization degree. Zeta Potential and H2-TPR analyses showed the influence of Ni and Zr in the Mo redox properties that contributed to the well HDS performance of the studied catalysts. The NiMo-γ-Al2O3-ZrO2 catalysts prepared in this work by a one-pot synthesis procedure, especially the NiMo/80A20Z, presented specific activities (QTOF) in the same order that those obtained for a commercial and for a traditional catalyst prepared by wet impregnation. The used one-pot method reduced the synthesis steps, minimizing the preparation time and the energy consumption.

Full text of DOI:10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Incorporation of the precursors of Mo and Ni oxides directly into the
reaction mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports –
Evaluation of the sulfided catalysts in the thiophene
hydrodesulfurization
Eduardo Prado Bastona,∗, Alexandre Boscaro Franc¸ aa, Alano Vieira da Silva Netob,
Ernesto Antonio Urquieta-Gonzálezb
a São João Del Rei Federal University, Department of Chemical Engineering and Statistics, C. Postal 131, MG 443 Km 7, Ouro Branco, MG, Brazil
b São Carlos Federal University, Department of Chemical Engineering, C. Postal 676, São Carlos, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 May 2014
Received in revised form 6 October 2014
Accepted 26 October 2014
Available online xxx
Mo and NiMo oxides supported on ␥-Al2O3-ZrO2 were successful one-pot synthesized and characterized
by XRF, XRD, N2 sorption, UVVISDRS and Zeta Potential. The sulfide catalysts were evaluated between
240 and 300 ◦C in the thiophene HDS, used as a model reaction. XRD data evidenced the presence of ␥-
Al2O3 in the obtained solids, with its crystallinity being reduced with the incorporation of Zr and metallic
species. XRD data also indicated that Zr, Ni and Mo were highly dispersed on the support. N2 sorption and
UVVISDRS analyses showed that the one-pot synthesis procedure led to Mo and NiMo supported solids
with high SBET areas and to Mo species with a low polymerization degree. Zeta Potential and H2-TPR
analyses showed the influence of Ni and Zr in the Mo redox properties that contributed to the well HDS
performance of the studied catalysts. The NiMo-␥-Al2O3-ZrO2 catalysts prepared in this work by a one-
pot synthesis procedure, especially the NiMo/80A20Z, presented specific activities (QTOF) in the same
order that those obtained for a commercial and for a traditional catalyst prepared by wet impregnation.
The used one-pot method reduced the synthesis steps, minimizing the preparation time and the energy
consumption.
Keywords:
Synthesis one-pot, Alumina, Zirconia, NiMo
supported HDS catalysts
Thiophene
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
supported on ␥-Al2O3 [1–3], which have high catalytic perfor-
mance; however, many efforts are done to improve the catalyst
As well known, SOx emissions produce a series of contamination
problems and, consequently, the environmental regulations world-
wide have drastically reduced the limits of sulfur content in fuels. To
satisfy those limits, the petroleum refining industry has introduced
important improvements in the catalytic hydrodesulfurization
(HDS) process of hydrocarbons, which has been the traditional
route to produce cleaner and more environmental friendly fuels. In
the HDS process, organosulfur compounds, as mercaptans, sulfides,
disulfides, thiophenes, benzothiophenes (BT), dibenzothiophenes
(DBT) and others [1], are desulfurized generating sulfur free hydro-
carbons and H2S. The most used HDS commercial catalysts are
constituted by Mo or W sulfides promoted by Co or Ni sulfides
properties related with the active phases or supports aiming to
enhance the HDS levels.
The physico-chemical and textural properties of the support,
such as specific surface area, pore size and acidity, can deeply
influence the dispersion of the active metallic species and the
diffusion of reactants and products under the HDS operating con-
ditions. In order to improve the performance of ␥-Al2O3, supports
like ZrO2 [4–6], TiO2 [4,6], mesoporous molecular sieves [7] or
ZrO2-Al2O3 mixed oxides have been proposed [8–10]. On the
other hand, the method to incorporate the metallic active species
on the support can strongly influence the activity of the cata-
lysts. In comparison with the traditional support impregnation
procedure, the in situ incorporation of Mo, Co and Ni precursor
salts during the ‘sol’ preparation of the support (one-pot synthe-
sis) [11], has the advantage to minimize the time and to reduce
the energy consumption, resulting in a lower catalyst production
cost.
∗
Corresponding author. Tel.: +55 31 8727 3329.
E-mail addresses: baston@ufsj.edu.br, eduardobaston@hotmail.com
(E.P. Baston).
http://dx.doi.org/10.1016/j.cattod.2014.10.035
0920-5861/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
2
E.P. Baston et al. / Catalysis Today xxx (2014) xxx–xxx
Considering the above discussed subject, the focus of the present
work was to study the physico-chemical and textural properties
of supported Mo and NiMo oxides, which were prepared directly
introducing under controlled pH conditions the respective precur-
sor salts into the reaction mixture during the sol–gel preparation of
␥-Al2O3-ZrO2 supports (one-pot synthesis). The supported Mo and
NiMo oxides were subsequently sulfided and the obtained catalysts
evaluated in the HDS of thiophene between 240 and 300 ◦C, used
as a model reaction. As a reference, a NiMo(i)/␥-Al2O3 catalyst was
also prepared by a traditional wet impregnation procedure.
spectrometer. The studied samples were also Zeta Potential ana-
lyzed using a Zeta Sizer (Malvern Instruments), equipped with
capillary cells (DTS1060 model) and a MPT-2 automatic titrator.
2.3. Catalytic evaluation
The catalytic evaluation using the HDS of thiophene as a model
reaction was performed at atmospheric pressure between 240 and
300 ◦C in a fixed-bed tubular micro-reactor using 0.1 g of catalyst
and 20 mL min−1 of a gas mixture of thiophene (8.2% mol/mol in
H2). The reaction conditions were selected to avoid external diffu-
sive effects and to obtain thiophene conversions lower than 15%.
The thiophene HDS was left during 50 min at the same tempera-
ture and for each 10 min the effluent products were CG analyzed in
a Varian CP 4900 equipped with a TCD and a 5CB Module. Before the
thiophene HDS, the supported Mo and NiMo oxides were sulfided
“in situ” using 30 ml.min−1 of a diluted H2S mixture (10% mol/mol
in H2).
In order to compare, a commercial NiMoP/Al2O3 catalyst was
also evaluated.
The thiophene conversion (XT) was calculated using Eq. (1),
where xiDS are the molar fractions of the hydrodesulfurized prod-
ucts and xiTOT the molar fractions of the hydrodesulfurized products
and the non-reacted thiophene.
2. Experimental
2.1. Mo and NiMo oxides supported on ꢀ-Al2O3-ZrO2
␥-Al2O3-ZrO2 supports were prepared using a gel with a molar
composition of: 1 (aluminum tri-sec-butoxide + zirconium propox-
ide):8.5 2-butanol:5 1,3-butanediol:10 H2O [11] and the nominal
composition of the Mo or NiMo supported oxides was 12 wt.% of
MoO3 or 12 wt.% of MoO3 and 3 wt.% of NiO, respectively. The sam-
ples were named Me-xAyZ, where Me represents Mo or NiMo; A:
␥-alumina; Z: zirconia; x is the wt.% of ␥-alumina in the support
being 100, 80 or 60 and y is the wt.% of zirconia = 100 − x. In a typ-
ical preparation, firstly aluminum tri-sec-butoxide (ATSB, Aldrich
97%) and/or zirconium propoxide (PZr, Aldrich 70%) and 2-butanol
(Aldrich, ≥ 99%) were mixed at 85 ◦C under stirring for 30 min. Then,
1,3-butanediol (Aldrich, 99+%) was added to the solution and left
under stirring for 1 h. Subsequently, the hydrolysis step was done
by adding deionized water and maintaining the stirring for one hour
more. The formed gel was aged at room temperature during 1 h and
then dried at 45 ◦C in a rotary evaporator under reduced pressure
during 48 h.
ꢀ
ꢀ
XT = 100x(
xiDS
/
xiTOT
)
(1)
3. Results and discussion
3.1. X-Ray diffraction
The XRD patterns of the pure ␥–Al2O3 (Fig. 1a) exhibits the typ-
ical peaks of the crystalline ␥–Al2O3 structure (JCPDS-48-0367).
From Fig. 1a it can also be seen that the diffractograms of the
80A20Z and 60A40Z show a decrease in the intensity of the ␥-Al2O3
diffraction peaks with the increase of the amount of Zr incorporated
in the support and the resulting solid becoming more amorphous
[8]. Moreover, the diffractograms of the xAyZ do not show any peak
related with any zirconia crystalline phase, which indicates that it
is either completely amorphous or constituted by small crystallites
(< 4 nm) [8,12].
The Mo and NiMo oxides supported on the xAyZ solids, precur-
sors of the HDS catalysts (Fig. 1b), exhibit similar patterns than
those of the respective supports (Fig. 1a), with the loss of crys-
tallinity being consequence of the presence of the incorporated Zr
in the support and of the partial X-ray absorption by the introduced
metallic species [13].
In the preparation of Mo or NiMo supported oxides, their
respective precursor salts were directly introduced in the reaction
mixture of the supports during the hydrolysis step (in situ incorpo-
ration or one-pot synthesis as the procedure is commonly named in
the literature). It was firstly added the solution of ammonium hep-
tamolybdate tetrahydrate (pH = 9) and after 10 min under stirring
it was added the solution of nickel (II) nitrate hexahydrate, with the
stirring remaining for one hour more. The resulting gel was aged
during 1 h and then dried under the same conditions used in the
preparation of the pure supports.
The pure supports and the precursor catalysts were calcined
under air at 500 ◦C during 4 h.
A reference sample was prepared by a traditional wet impreg-
nation procedure of Mo and Ni salts on a pure ␥-Al2O3 support
(NiMo(i)/100A0Z), subsequently being dried at 100 ◦C during 24 h
and then calcined under air at 500 ◦C during 4 h.
From Fig. 1b it is clear that no diffraction peaks attributed to
Mo or Ni species are possible to distinguish, which suggests that
Mo species are highly dispersed on the supports and constituted
by small crystallites amorphous to X-ray radiation [6,7]. According
with Bergwerff et al. [14], the Mo oxides are normally distributed
in a monolayer due to their tendency to spread on the support sur-
face at high temperatures. This behavior shows that the used Mo
and NiMo direct incorporation procedure, can in fact result in a
well oxide dispersion, however, it is necessary to emphasize that
it is possible that part of the Mo or NiMo active species could be
occluded into the support or even making part of a ternary metal-
lic phase containing Mo, Al and Zr that could explain the lower
intensity of the diffraction peaks of the Mo-xAyZ or NiMo-xAyZ
samples when compared with the pure support (Fig. 1a). The inten-
sity decrease of the diffraction peaks of the Mo-xAyZ or NiMo-xAyZ
samples could also be consequence of the presence of defects or dis-
tortions in the Al-O bonds due to Zr incorporation in the support
[8].
2.2. Characterization
The Mo, Ni, Al and Zr contents were determined by X-ray
fluorescence spectroscopy (XRF) on a Rigaku Rix 3100 X-ray flu-
orescence spectrometer equipped with an end window 4 kW Rh.
X-ray diffraction (XRD) patterns were obtained on a Rigaku Multi-
˚
flex Diffractometer operating with CuK␣ (ꢁ = 1.5406 A) radiation.
The specific surface area and mean pore diameter were deter-
mined from N2 adsorption/desorption isotherms obtained at 77 K
on a Quantachrome Corporation Nova-1200 apparatus. H2 tem-
perature programmed reduction (H2-TPR) analyses were carried
out on a SAMP3 chemisorption analyzer equipped with a thermal
conductivity detector (TCD). In each TPR analysis were used 60 mg
of the sample, a flow of 25 mL min−1of H2 (5% V/V in N2) and a
heating rate of 10 ◦C min−1. UV–visible Diffuse Reflectance Spec-
troscopy (UVvisDRS) analyses were performed in a Varian Cary 5G
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
E.P. Baston et al. / Catalysis Today xxx (2014) xxx–xxx
3
Fig. 1. X-ray diffractograms of: (a) ␥-Al2O3-ZrO2 and ZrO2 pure supports; (b) Mo and NiMo oxides supported on ␥-Al2O3-ZrO2.
3.2. N2 adsorption/desorption
Zepeda et al. [21] demonstrate that there is a direct relationship
between the reducibility of the precursor NiMo oxides and the HDS
activity of the sulfided catalysts. According to those authors, the
lower the reduction temperature of the HDS active metallic species
the higher will be the amount of the generated sulfided ones, thus
resulting in highly active catalysts.
All the analyzed solids showed isotherms of type IV (not shown).
The mean pore diameter (dBJH) [15], specific surface area (SBET
)
[16] and total pore volume (Vt) of the xAyZ supports and of the
Mo-xAyZ and NiMo-xAyZ precursor catalysts are shown in Table 1.
As it can be seen, all the sol–gel prepared samples present high
and very similar values of specific surface area depending on the
zirconium content and, coherently, the higher the Zr content the
lower was the SBET. Contrarily to supported species prepared by
conventional impregnation methods [9], the one-pot synthesis of Ni
or NiMo oxides supported on xAyZ resulted in SBET values slightly
higher than the pure supports. This promising result is attributed
to the used one-pot procedure and to the pH control of the used
ammonium heptamolybdate tetrahydrate solution [14,17,18]. The
significant decrease of the mean pore diameter and pore volume
of the Mo and NiMo precursor catalysts when compared with the
pure supports evidences that Mo and NiMo oxides were formed
inside of the ␥–Al2O3-ZrO2 mesopores.
3.4. UVvisDRS
The UVvisDRS spectra of the studied Mo and NiMo supported
oxides, determined between 200 and 800 nm, are shown in Fig. 3.
Since the Mo6+ ion has a d0 electronic configuration, the only
absorption band that occurred between 200 and 400 nm in the
electronic spectra of the Mo-xAyZ samples (Fig. 3a) is due to
the ligand–metal charge transfer transition (LMCT): O2− → Mo6+
[17,20–23]. Higashimoto et al. [22] attribute the absorption bands
at 200 nm and between 250 to 280 nm to Mo+6 species in tetra-
hedral coordination and the bands between 300 and 400 nm
attributed to the absorption of Mo
O Mo with octahedral coor-
dination. Fournier et al. [17] observed that with the increase of the
size of Mo clusters in the supported catalysts occurs an enlargement
and displacement of that band to higher wavelength. On the other
hand, they concluded that with the increasing of Mo dispersion
occurs a thinning and displacement of that same band to smaller
wavelengths. The NiMo-xAyZ samples (Fig. 3b) show more similar
absorption bands related with Mo6+ that were above discussed for
the Mo-xAyZ samples (Fig. 3a). However, the NiMo-xAyZ solids also
show absorption bands at 390 nm and between 600 and 800 nm
that are related with octahedrally coordinated Ni+2 species [24].
Besides to evidence the Mo oxidation state, the UVvisDRS data
allow analyzing the dispersion of the Mo species, determining
their aggregation degree (number of Mo neighbors species in a
cluster). From this analysis, it is possible to obtain the absorp-
tion edge energy of the Mo species according as was proposed by
Webber [25], who correlated the absorption edge energy bands of
the UVvisDRS spectrum with the average number of Mo neighbors
species. The method consists in applying the UVvisDRS results to
build a graphic of [F(R ) × hꢂ]2 × hꢂ, where F(R ) is the Kubelka-
3.3. H2-TPR
The hydrogen temperature programmed reduction (H2-TPR)
profiles of the Mo-xAyZ samples (Fig. 2a) showed hydrogen con-
sumption related to the reduction of Mo species [5,7] in a first step
of Mo6+ to Mo4+ (between 400 and 600 ◦C) and then Mo4+ to Mo0
(between 800 and 950 ◦C). From Fig. 2a it is also verified a signif-
icant decrease in the reduction temperature of Mo6+ species with
the increase of the amount of Zr incorporated into the support.
According to Zhang et al. [9], the incorporation of ZrO2 into the
support decreased the interaction between the active metal and the
composite support, leading to form Mo species which are easier to
be reduced.
From the H2-TPR profiles of NiMo-xAyZ samples (Fig. 2b) it is
also clearly verified a substantial decrease in the reduction tem-
perature of Mo6+ that is consequence of two effects, the first one,
as discussed above, is related to Zr incorporation into the sup-
port and the second one to the well-known synergetic redox effect
of Ni in the reduction of Mo6+ species [5–7,20]. That effect is
related to the lower reduction temperature of Ni2+ to Nio (typically
between 300 and 500 ◦C) [19], which promotes the H2 spillover
phenomenon [20] that generates the reduction of Mo6+ at lower
temperatures. The presence of Ni in the supported NiMo oxides
was confirmed by X-ray fluorescence (Table 1) and DRS-UVvis (dis-
cussed below).
∞
∞
Munk function and hv the energy of the incident photon. Then, one
must determine the intersection of the tangent line at the inflec-
tion point of the curves in the lower energy region with the x-axis,
being the value found the energy of the absorption edge band.
Table 2 shows that the studied Mo-xAyZ and NiMo-xAyZ sam-
ples, present values of that absorption edge energy between 3.67
and 4.16 eV. As proposed by Webber [25], those values indicate
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
4
E.P. Baston et al. / Catalysis Today xxx (2014) xxx–xxx
Table 1
Chemical composition and textural properties of pure supports and Mo and NiMo oxides supported on ␥-Al2O3-ZrO2.
Chemical composition (wt.%)
Sample
dBJHa[nm]
SBETb[m2 g−1
]
Vtc[cc g−1
]
NiO
MoO3
Al2O3
ZrO2
60A40Z
80A20Z
100A0Z
Mo-60A40Z
Mo–80A20Z
Mo–100A0Z
NiMo-60A40Z
NiMo–80A20Z
NiMo–100A0Z
NiMo(i)/100A0Z
8.7
9.0
7.0
3.7
4.8
4.0
4.1
4.5
3.7
4.2
310
368
451
342
397
488
367
399
459
305
0.72
1.18
1.20
0.44
0.70
0.67
0.50
0.69
0.55
0,47
–
–
–
–
–
–
2.9
3.0
3.1
3.3
–
–
–
10.3
9.8
9.0
11.7
11.3
8.7
13.5
54
79
100
59
78
100
56
77
100
100
46
21
–
41
22
–
44
23
–
–
a
dBJH: mean pore diameter.
SBET: specific surface area.
Vt: total pore volume.
b
c
that the formation of Mo species with high polymerization degree
was minimized and that the studied samples are mainly consti-
tuted by dimers (Mo2O7)2−, where molybdenum is in tetrahedral
3.5. Zeta Potential
Changes in the surface characteristics of the supports (xAyZ)
were analyzed trough the values of the isoelectric point (IEP). The
obtained Zeta Potential data (not shown) can be explained by Eqs.
(2) and (3), which according Guevara-Lara, Bacaud and Vrinat [26],
represent the reaction between OH− and H+ ions with the oxide
surface. These authors suggest that if the pH decreases, the H+
coordination, by poly-anions of (Mo7O24)
2− where molybdenum is
in octahedral coordination or even by a mixture of those two Mo
species. This result confirms that the one-pot synthesis procedure
led to highly dispersed Mo species on the ␥-Al2O3-ZrO2 supports,
as it was suggested by the DRX data.
Fig. 2. H2-TPR profiles of: a) Mo; (b) NiMo oxides supported on ␥-Al2O3-ZrO2.
Fig. 3. UVvisDRS spectra of: (a) Mo; (b) NiMo oxides supported on ␥-Al2O3-ZrO2.
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
E.P. Baston et al. / Catalysis Today xxx (2014) xxx–xxx
5
Table 2
Absorption edge energy of the Mo and NiMo oxides supported on ␥-Al2O3-ZrO2.
Catalyst
Absorption edge Eenergy [eV]
Mo-60A40Z
Mo-80A20Z
Mo-100A0Z
NiMo-60A40Z
NiMo-80A20Z
NiMo-100A0Z
3.67
3.76
4.16
3.81
3.86
3.71
concentration increases and the negative charges of the surface
are neutralized, so Zeta Potential becomes more positive (Eq. (2)).
On the other hand, if the pH increases, the OH− concentration
increases, then the surface charge will be saturated and the Zeta
Potential is negative (Eq. (3)).
−MOH + H+→ − MOH2+↔ − M+ + H2O
−MOH + OH−→ − M(OH)2−↔ − MO− + H2O
(2)
(3)
Fig. 4. Specific activity (QTOF) of Mo-␥-Al2O3-ZrO2, NiMo-␥-Al2O3-ZrO2, reference
NiMo(i)/␥-Al2O3-ZrO2 and commercial NiMoP/alumina catalysts as a function of the
reaction temperature during the thiophene HDS.
The point that Zeta Potential is zero (ꢃ = 0) is defined as the iso-
electric point or point of zero charge (PZC). The obtained results
showed that an increase in the Zr content in the ␥-Al2O3-ZrO2 sup-
port led to a decrease in the IEP. These result shows that the surface
of the supports containing Zr exhibit intermediate characteristics of
the pure alumina and zirconia. According to Prada Silvy et al. [27],
above the PZC the surface is negatively charged, therefore, nega-
tively charged ions will not be able to adsorb on the surface due to
electrostatic repulsion. On the other hand, negatively charged ions
which are in a solution with pH below than that corresponding to
the PZC, can easily adsorb on the solid surface. Thus, the ZrO2 sur-
face has a tendency to adsorb preferably hydroxyl groups (OH−), or
present a behavior of a weak acid, on the contrary, alumina surface
has a behavior of a weak base.
The Zeta Potential data confirmed that the increase of Zr content
in the support generates modifications in their physico-chemical
properties, which influenced their acid–base properties that led to
a decrease of the reduction temperatures of Mo6+ species, as dis-
cussed in the H2-TPR results. Those acid-base properties must also
influence the adsorption mechanism of the organosulfur molecules
on the surface of the studied HDS catalysts.
the active catalytic site [31,32]. This phenomenon occurs due to the
higher reducibility of the Ni species in relation to the molybdenum
species, as it was discussed above in the H2-TPR analyses of the
NiMo-xAyZ samples.
From Table 3 it can also be verified that with the increase of
zirconium content occurs a tendency to a slightly increase in the
thiophene conversion. This result is attributed to an increase in the
Mo6+ reducibility in the samples containing Zr, as it was evidenced
in the H2-TPR analyses of Mo-xAyZ and NiMo-xAyZ samples (Fig. 2).
In Table 3 are displayed the reaction rate calculated at 300 ◦C
(molthiophene gcat−1 s−1) and the apparent activation energy (Eaap
)
obtained from an Arrhenius plot (ln (−rA) vs. 1/T), built from thio-
phene conversion on the ␥-Al2O3-ZrO2 supported Mo and NiMo
sulfided catalysts. As observed, the Eaap values of the NiMo-xAyZ
catalysts are, coherently, slightly lower than the Mo-xAyZ ones
and the reaction rate increases with the increase of the reac-
tion temperature and also with the increase of the Zr content.
The studied NiMo catalysts, when compared with a reference
NiMo(i)/100A0Z catalyst prepared with the same content of Mo
and Ni oxides via a traditional wet impregnation procedure and
with a commercial NiMoP/Alumina catalyst (19 wt.% MoO3, 5 wt.%
NiO and 6 wt.% P2O5), show lower conversion levels (Table 3),
however, their reaction rates are in the same order (close to
100 × 10−5 molthiophene gcat−1 s−1).
3.6. Catalytic evaluation
As it can be observed from Table 3, the HDS of thiophene on cata-
lysts containing only Mo resulted in thiophene conversions ≤2%. On
the other hand, the NiMo catalysts present significant higher thio-
phene conversions, confirming the well-known strong promoter
character of Ni species in the desulfurization mechanism involv-
ing Mo species [7,28,29]. Many theories have been presented in
the literature to elucidate the HDS reaction mechanism on sul-
fided NiMo and CoMo catalysts, as that proposed by Wang et al.
[20] for the HDS of thiophene, dibenzothiophene, and their deriva-
tives dimethyl-dibenzothiophenes. According to those authors, Ni
acts as a promoter exerting attractive forces on the sulfur atom of
the organosulfur molecule, extracting it. According to this mech-
anism the strength of the Mo S bond is higher than that of the
Ni S bond; however the Mo S bond attached to an edged Mo atom
is more difficult to cleave than a Ni S bond. However, more evi-
dences related with this promoting effect have been explained by
the model involving a structural association between the metal-
lic phases, known as “CoMoS” or “NiMoS” [30], where the Ni or
Co are atomically dispersed at the edge of the MoS2 structure.
These species are claimed to be the mainly responsible for the
HDS catalytic activity, where Ni or Co metals are able to gener-
ate a “spillover” of hydrogen, which migrate from the donor phase
(NiSx or CoSx) to the MoS2 recipient phase, creating or modifying
As the number of active sites in the studied Mo or NiMo sulfided
catalysts is not known, the specific activity of those active sites
in the used heterogeneous reactor working at steady-state condi-
tions can be close estimated in terms of a quasi-turnover frequency
(QTOF) calculated according to Eq. (4):
QTOF = [X × F]/m
(4)
Where QTOF represent a specific activity (mol × molMe−1 s−1),
X is the conversion of thiophene, F is its molar flow rate (mol s−1
)
and m refers to the mol number of metal atoms per gram of catalyst
(molMe gcat−1), with molme = (molMo + molNi).
As it can be seen from Fig. 4, the NiMo-xAyZ catalysts, and espe-
cially, the NiMo-80A20Z, present QTOF values closer than those of
the reference and commercial catalysts, confirming that the used
one-pot synthesis procedure led to highly active HDS catalysts with
the important advantages in reducing the preparation time, energy
consumption and, consequently, reducing the catalyst production
cost.
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
6
E.P. Baston et al. / Catalysis Today xxx (2014) xxx–xxx
Table 3
Conversion, reaction rate and apparent activation energy (Eaap) of the studied Mo and NiMo sulfided catalysts supported on ␥-Al2O3-ZrO2.
a
Catalyst
Conversion (%)
240 ◦C
Reaction Rate x10−5 (molthiophene gcat−1 s−1
)
Eaap × 104 (J mol−1
)
260 ◦
C
280 ◦
C
300 ◦
C
Mo-60A40Z
Mo-80A20Z
Mo-100A0Z
NiMo-60A40Z
NiMo-80A20Z
NiMo-100A0Z
NiMo(i)/100A0Z
NiMoP/alumina
0.1
0.2
0.1
1.0
1.6
1.1
2.4
2.9
0.3
0.4
0.3
2.5
3.1
2.7
4.6
6.0
0.6
0.6
0.5
4.2
5.1
4.5
7.3
10.4
1.0
1.0
0.6
6.6
9.5
7.2
11.4
17.4
11
12
8
70
100
86
7.0
7.0
7.6
6.1
6.3
5.9
5,7
6.7
97
113
a
Reaction rate at 300 ◦C.
[4] Y. Okamoto, K. Ochiai, M. Kawano, K. Kobayashi, T. Kubota, Effects of support
on the activity of Co–Mo sulfide model catalysts. Appl. Catal. A 226 (2002)
115–127.
[5] L. Kaluza, M. Zdrazil, Preparation of zirconia-supported hydrodesulphurisation
catalysts by water-assisted spreading. Appl. Catal. A 329 (2007) 58–67.
[6] J. Escobar, M.C. Barrera, J.A. Reyes, J.A. Toledo, V. Santes, J.A. Colín, Effect of
chelating ligands on Ni–Mo impregnation over wide-pore ZrO2–TiO2. J. Mol.
Catal. A: Chem. 287 (2008) 33–40.
The thiophene HDS resulted in sulfur-free n-butane, butene,
trans-butene and cis-butene compounds with typical selectivities
(not shown) that well matched with those describe in the literature
[33].
4. Conclusions
[7] S. Garg, K. Soni, G.M. Kumaran, M. Kumar, J.K. Gupta, L.D. Sharma, G.M. Dhar,
Effect of Zr-SBA-15 support on catalytic functionalities of Mo, CoMo, NiMo
hydrotreating catalysts. Catal. Today 130 (2008) 302–308.
[8] G. Li, W. Li, M. Zhang, K. Tao, Morphology and hydrodesulfurization activity
of CoMo sulfide supportedon amorphous ZrO2 nanoparticles combined with
Al2O3. Appl. Catal. A 273 (2004) 233–238.
[9] D. Zhang, A. Duan, Z. Zhao, G. Wan, Z. Gao, G. Jiang, K. Chi, K.H. Chuang,
Preparation, characterization and hydrotreating performances of ZrO2–Al2O3-
supported NiMo catalysts. Catal. Today 149 (2010) 62–68.
[10] M.A. Al-Daous, S.A. Ali, Deep desulfurization of gas oil over NiMo catalysts
supported on alumina–zirconia composites. Fuel 97 (2012) 662–669.
[11] E.P. Baston, E.A. Urquieta-Gonzalez, Direct addition of the precursor salts of Mo,
Co or Ni oxides during the sol formation of ␥-Al2O3 and ZrO2 - The effect on
metal dispersion. Stud. Surf. Sci. Catal. 175 (2010) 671–674.
[12] P. Jabbarnezhad, M. Haghighi, P. Taghavinezhad, Sonochemical synthesis of
NiMo/Al2O3–ZrO2 nanocatalyst: Effect of sonication and zirconia loading on
catalytic properties and performance in hydrodesulfurization reaction. Fuel
Process. Technol. 126 (2014) 392–401.
[13] F. Dumeignil, K. Sato, M. Imamura, N. Matsubayashi, E. Payen, H. Shimada, Char-
acterization and hydrodesulfurization activity of CoMo catalysts supported on
boron-doped sol–gel alumina. Appl. Catal. A 315 (2006) 18–28.
[14] J.A. Bergwerff, M. Jansen, B.G. Leliveld, T. Visser, K.P. de Jong, B.M. Weck-
huysen, Influence of the preparation method on the hydrotreating activity of
MoS2/Al2O3 extrudates: A Raman microspectroscopy study on the genesis of
the active phase. J. Catal. 243 (2006) 292–302.
[15] E.P. Barret, L.G. Joyner, P.P. Halenda, The Determination of Pore Volume
and Area Distributions in Porous Substances. I. Computations from Nitrogen
Isotherms. J. Am. Chem. Soc. 73 (1951) 373–380.
We synthesized Mo and NiMo oxides supported on ␥-Al2O3-
ZrO2 containing 0, 20 or 40% of ZrO2 via a one-pot synthesis
procedure and the resulting solids were characterized by XRF,
XRD, N2 sorption measurements, UVVISDRS and Zeta Potential.
These HDS catalysts precursors were subsequently sulfided and
evaluated between 240 and 300 ◦C in the thiophene HDS, used as
a model reaction. The XRD data evidenced the presence of ␥-Al2O3
phase in the obtained pure supports and in the Mo and NiMo
supported solids, with the ␥-Al2O3 crystallinity decreasing with
the incorporation of Zr and metallic species. No diffraction peaks
related with any crystalline phase of zirconia, Mo and Ni were
observed in the diffractograms of the NiMo-␥-Al2O3-ZrO2 solids,
indicating that they are formed by highly dispersed nanometric
crystallites. N2 sorption measurements and UVVISDRS analyses,
showed that the one-pot synthesis procedure led, respectively, to
pure support and ␥-Al2O3-ZrO2 supported Mo and NiMo oxides
with high specific surface areas and to Mo species with a low
polimerization degree. Zeta Potential and H2-TPR data showed the
influence of Ni and Zr in the Mo redox properties that resulted in a
decrease in the reduction temperatures of Mo6+, which contributed
to the well HDS catalytic performance.
[16] S. Brunauer, P.H. Emmett, E.J. Teller, Adsorption of Gases in Multimolecular
Layers. J. Am. Chem. Soc. 60 (1938) 309–319.
As expected, the thiophene HDS showed that the NiMo-
sulfided catalysts were highly more active than the Mo-sulfided
ones, confirming the well-known Ni promoting effect. The NiMo-
[17] M. Fournier, C. Louis, M. Che, P. Chaquin, D. Masure, Polyoxometallates as
models for oxide catalysts: Part I. An UV-visible reflectance study of polyoxo-
molybdates: Influence of polyhedra arrangement on the electronic transitions
and comparison with supported molybdenum catalysts. J. Catal. 119 (1989)
400–414.
[18] A. Duan, G. Wang, Z. Zhao, C. Xu, Y. Zheng, Y. Zhang, T. Dou, X. Bao, K.
Chung, Characterization and activity of Mo supported catalysts for diesel deep
hydrodesulphurization. Catal. Today 119 (2007) 13–18.
[19] J. Barrientos, M. Lualdi, M. Boutonnet, S. Järås, Deactivation of supported nickel
catalysts during CO methanation. Appl. Catal. A 486 (2014) 143–149.
[20] A. Wang, Y. Wang, T. Kabe, Y. Chen, A. Ishihara, W. Qian, P. Yao, Hydrodesul-
furization of Dibenzothiophene over Siliceous MCM-41-Supported Catalysts:
II. Sulfided Ni–Mo Catalysts. J. Catal. 210 (2002) 319–327.
[21] T.A. Zepeda, T. Halachev, B. Pawelec, R. Nava, T. Klimova, G.A. Fuentes,
J.L.G. Fierro, Hydrodesulfurization of dibenzothiophene over CoMo/HMS and
CoMo/Ti-HMS catalysts. Catal. Commun. 7 (2006) 33–41.
␥-Al2O3-ZrO2 catalysts prepared in this work by
a one-pot
synthesis procedure, especially the NiMo/80A20Z, presented spe-
cific reaction rates in the same order that those obtained for a
NiMo(i)/␥-Al2O3 catalyst prepared via a traditional wet impregna-
tion procedure and for a commercial NiMoP/␥-Al2O3. This result
is highly attractive because the one-pot procedure reduces the
synthesis steps, therefore, reducing the preparation time, energy
consumption and consequently, the catalyst production cost.
Acknowledgements
[22] S. Higashimoto, Y. Hu, R. Tsumura, K. Iino, M. Matsuoka, H. Yamashita, Y.G. Shul,
M. Che, M. Anpo, Synthesis, characterization and photocatalytic reactivities of
Mo-MCM-41 mesoporous molecular sieves: Effect of the Mo content on the
local structures of Mo-oxides. J. Catal. 235 (2005) 272–278.
The authors gratefully acknowledge the financial support of
CNPq and PETROBRAS.
[23] P.T. Mancera, J. Ramírez, R. Cuervas, A.G. Alejandre, F. Murrieta, R. Luna, Hydro-
desulfurization of 4,6-DMDBT on NiMo and CoMo catalysts supported on
B2O3-Al2O3. Catal. Today. 107–108 (2005) 551–558.
References
[24] P. Salerno, S. Mendioroz, A.L. Agudo, Al-pillared montmorillonite-based NiMo
catalysts for HDS and HDN of gas oil: influence of the method and order of Mo
and Ni impregnation. Appl. Catal. A 259 (2004) 17–28.
[25] R.S. Weber, Effect of Local Structure on the UV-Visible Absorption Edges of
Molybdenum Oxide Clusters and Supported Molybdenum Oxides. J. Catal. 151
(1995) 470–474.
[1] C. Song, Catal. , An overview of new approaches to deep desulfurization for
ultra-clean gasoline, diesel fuel and jet fuel. Today 86 (2003) 211–263.
[2] S. Brunet, D. Mey, G. Pérot, C. Bouchy, F. Diehl, On the hydrodesulfurization of
FCC gasoline: a review. Appl. Catal. A 278 (2005) 143–172.
[3] I.V. Babich, J.A. Mouljin, Science and technology of novel processes for deep
desulfurization of oil refinery streams: a review. Fuel 82 (2003) 607–631.
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035

G Model
CATTOD-9324; No. of Pages7
ARTICLE IN PRESS
E.P. Baston et al. / Catalysis Today xxx (2014) xxx–xxx
7
[26] A. Guevara-Lara, R. Bacaud, M. Vrinat, Highly active NiMo/TiO2–Al2O3 cata-
lysts: Influence of the preparation and the activation conditions on the catalytic
activity. Appl. Catal. A 328 (2007) 99–108.
[27] R. Prada-Silvy, F. Lopez, Y. Romero, E. Reyes, V. León, R. Galiasso, Influence of
titania loading on tungsten adsorption capacity, dispersion, acidic and zero
point of charge properties of W/TiO2-Al2O3 catalysts. Stud. Surf. Sci. Catal. 91
(1995) 281–290.
[28] J.M. Herrera, J. Reyes, P. Roquero, T. Klimova, New hydrotreating NiMo catalysts
supported on MCM-41 modified with phosphorus. Microporous Mesoporous
Mater. 83 (2005) 283–291.
[29] M. Egorova, R. Prins, Hydrodesulfurization of dibenzothiophene and 4,6-
dimethyldibenzothiophene over sulfided NiMo/␥-Al2O3, CoMo/␥-Al2O3, and
Mo/␥-Al2O3 catalysts. J. Catal. 225 (2004) 417–427.
[30] H. Topsoe, B. Hinnemann, J.K. Norskov, J.V. Lauritsen, F. Besenbacher, P.L.
Hansen, G. Hytoft, R.G. Egenerg, K.G. Knudsen, The role of reaction pathways
and support interactions in the development of high activity hydrotreating
catalysts. Catal. Today 107-108 (2005) 12–22.
[31] S. Eijsbouts, On the flexibility of the active phase in hydrotreating catalysts.
Appl. Catal. A 158 (1997) 53–92.
[32] P. Grange, X. Vanhaeren, Hydrotreating catalysts, an old story with new chal-
lenges. Catal. Today 36 (1997) 375–391.
[33] M.J.B. Souza, B.A. Marinkovic, P.M. Jardim, A.S. Araujo, A.M.G. Pedrosa, R.R.
Souza, HDS of thiophene over CoMo/AlMCM-41 with different Si/Al ratios.
Appl. Catal. A 316 (2007) 212–218.
Please cite this article in press as: E.P. Baston, et al., Incorporation of the precursors of Mo and Ni oxides directly into the reaction
mixture of sol–gel prepared ␥-Al2O3-ZrO2 supports – Evaluation of the sulfided catalysts in the thiophene hydrodesulfurization, Catal.
Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.035