Synergy effect of boron and cobalt in B2O3-SBA-15-(Co)Mo catalyst for efficient hydrodesulfurization of liquid fuels

Article abstract of DOI:10.1007/s11164-021-04475-x

The contributory effect of surface acidity is significant to desulfurize heavy recalcitrant organosulfur compounds effectively. This research explores the synergistic effect of boron and cobalt in CoMoS supported on B₂O₃-SBA-15 for ultradeep hydrodesulfurization (HDS) of DBT and MDBT in the model and diesel fuel, respectively. The catalysts prepared consisting of SBA-15-(Co)Mo, representing (SM and SMC) and B₂O₃-SBA-15-(Co)Mo for BSM and BSMC, were fully characterized to gain insight into the structural activity concerning the nature of the fuel and the organosulfur compound employed. The incorporation of boron into the mesoporous framework of SBA-15 improves the surface characteristics, viz. surface acidity, support metal interaction and textural properties, necessary for efficient catalytic HDS reaction. The catalysts BSM and BSMC outperformed their analogous catalysts SM and SMC. The catalytic efficiency of BSMC is outstanding and capable of desulfurizing diesel fuel (1000 ppmw-S) containing a more complex matrix with a 90% conversion of methyldibenzothiophene (MDBT). BSMC possessed favorable and lower activation energy (E_a) of 77.11 kJ mol^?1 than other catalysts and can be used as a commercial catalyst for ultradeep desulfurization of real fuel. Graphic abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s11164-021-04475-x

Research on Chemical Intermediates
https://doi.org/10.1007/s11164-021-04475-x
Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo
catalyst for efcient hydrodesulfurization of liquid fuels
Saheed A. Ganiyu1
Received: 14 August 2020 / Accepted: 19 April 2021
© The Author(s), under exclusive licence to Springer Nature B.V. 2021
Abstract
The contributory efect of surface acidity is signifcant to desulfurize heavy recal-
citrant organosulfur compounds efectively. This research explores the synergis-
tic efect of boron and cobalt in CoMoS supported on B2O3-SBA-15 for ultradeep
hydrodesulfurization (HDS) of DBT and MDBT in the model and diesel fuel,
respectively. The catalysts prepared consisting of SBA-15-(Co)Mo, representing
(SM and SMC) and B2O3-SBA-15-(Co)Mo for BSM and BSMC, were fully char-
acterized to gain insight into the structural activity concerning the nature of the
fuel and the organosulfur compound employed. The incorporation of boron into the
mesoporous framework of SBA-15 improves the surface characteristics, viz. surface
acidity, support metal interaction and textural properties, necessary for efcient cat-
alytic HDS reaction. The catalysts BSM and BSMC outperformed their analogous
catalysts SM and SMC. The catalytic efciency of BSMC is outstanding and capable
of desulfurizing diesel fuel (1000 ppmw-S) containing a more complex matrix with
a 90% conversion of methyldibenzothiophene (MDBT). BSMC possessed favorable
and lower activation energy (Ea) of 77.11 kJ mol−1 than other catalysts and can be
used as a commercial catalyst for ultradeep desulfurization of real fuel.
*
Saheed A. Ganiyu
gsadewale@kfupm.edu.sa
1
Department of Chemistry, King Fahd University of Petroleum and Minerals, Dhahran 31261,
Saudi Arabia
Vol.:(0123456789)
1 3

S. A. Ganiyu
Graphic abstract
Keywords Boron-SBA-15 · CoMo catalysts · Hydrodesulfurization · Diesel ·
Dibenzothiophene · Methyldibenzothiophene
Introduction
The stringent environmental regulations imposed on transportation fuels by US and
EU agencies to lower the amount of sulfur to as low as 10 ppm and the demand for
sulfur-free fuel for automobile catalytic converter have been the motivation for the
continued search for robust, highly dispersed and stable hydrodesulfurization (HDS)
catalysts [1–6].
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
HDS of heavier organosulfur compounds such as DBT and alkylated deriva-
tives requires highly porous support with sufcient dispersion of active nanospe-
cies (Ni(Co)Mo(W) for efcient difusion and catalytic conversion of the reactants
within a short time [7]. The HDS catalytic efciency of the catalysts is attributed
to the physicochemical properties of the support and the catalysts (Ni(Co)Mo(W)).
The properties include but not limited to the nature of the support, acidity, surface
area, pore size and volume of both support and catalyst, dispersion, reducibility and
stability of the catalyst active species [8, 9]. The role and infuence of support in
catalyst development for heterogeneous reactions are crucial. The properties of sup-
ports are mostly proportional to the catalytic efciency. Alumina-supported catalysts
are characterized by high surface acid sites and amorphous pore structures, with
thermal and mechanical stability [10]. Lately, research in HDS catalysts’ develop-
ment has been directed towards replacing gamma-Al2O3-Ni(Co)Mo(W) due to its
strong metal–support interaction (SMSI) from the Type-1 active phase that conse-
quently resulted in low catalytic activity and quick deactivation [11–14].
Furthermore, the alumina metal oxide support generally sufers from the low sur-
face area, which led to low catalytic activity [15]. Molecular sieves (mesoporous)
are a group of silica supports used as an alternative to alumina due to their large
surface area, ordered porous structure, thermal stability and ease of preparation with
tunable physicochemical properties [16–19]. These supports have boosted the high
dispersion of active species and better catalytic performance compared to alumina.
However, the main challenge with mesoporous silica is the weak metal–support
interaction and low support acidity, which resulted from the amorphous nature of the
silica wall [20]. The interaction of active metals with silica-based supports is usually
enhanced by integrating heteroatoms, such as Ti, Zr, Nb and Al, and is becoming
a more pragmatic approach to improve the catalytic activity [21–23]. The addition
of additives, such as fuorine, phosphates and chelating agents, has been reported
to improve the HDS catalytic activity by improving the dispersion of active species
through moderate metal–support interaction [24–28]. The moderate metal–support
interaction favors the formation of polyanions (polymolybdates or polytungstates),
which are readily sulfded to form molybdenum sulfde (MoS2) or tungstic sulfde
(WS2) crystallites. The heteroatoms’ incorporation to mesoporous silica enhanced
the support acidity (both Lewis and Brønsted). It has been reported that the addi-
tion of Zr and Ti to the framework of SBA-15 support in the SBA-15-NiMo catalyst
increased the available active sites of the catalyst by increasing the acid sites (Lewis)
in the modifed supported catalyst [29, 30].
Boron, a heteroatom belonging to the metalloid group of elements and with
chemical properties closer to aluminum, has been used for numerous catalytic appli-
cations [31–34], including HDS [35–37]. Mostly, boron is employed as a stabilizer
for sulfded Ni(Co)Mo(W)-supported alumina. Usman et al. optimized the weight
percent loading of boron on MoS2/B/Al2O3 and Co–MoS2/B/Al2O3 for thiophene
HDS and discovered that 0.6 wt.% of boron gives the best HDS experience [36]. The
fndings of Rashidi et al. described the combined efect of phosphorus and boron on
nano-alumina support that was impregnated with Co and Mo. They discovered that
the combined heteroatoms resulted in the catalysts’ suitable physicochemical prop-
erties that led to high HDS activity [37].
1 3

S. A. Ganiyu
Inspired by boron’s role on the catalytic property of Al2O3-CoMo catalysts, boron
was incorporated into the mesoporous silica framework and studied the support
efect on the HDS reaction of dibenzothiophene (DBT) and methyldibenzothiophene
(MDBT). After a thorough literature search and to the best of the author’s knowl-
edge, no known study has been reported on the role of boron on the structural prop-
erties and HDS activity of mesoporous silica (SBA-15)-supported CoMo catalysts.
The catalyst performance was compared with the analogous CoMo/Al-SBA-15 cata-
lyst. The structural reactivity of the catalysts was investigated by N2-physisorption,
DRS-UV–Vis spectroscopy, NH3-TPD, H2-TPR, XRD, Raman, Pyridine FTIR,
FESEM and XPS.
Methodology
Synthesis of boron‑modifed SBA‑15‑Mo with and without Co (boron‑SBA‑15‑Mo
and/or boron‑SBA‑15‑Mo/Co)
The boron-modifed mesoporous SBA-15-Mo catalyst was synthesized by incorpo-
rating boron and molybdenum into a synthesis mixture containing silica described
by Zhao et al. [38]. Firstly, an equivalent of 13 wt.% Mo from ammonium molybdate
(VI) tetrahydrate precursor solution was added to a continuously stirred sol solution
containing 60 g of 2 M HCl, 4.16 g of TEOS, 2 g of P123 and 15 g of deionized
water. The solution mixture was further stirred for 12 h before hydrothermal treat-
ment at 90 °C for 1 day. The resultant white powder, after centrifugation and dry-
ing, was calcined to form SBA-15-Mo. Boron and/or cobalt was introduced into the
SBA-15-Mo catalyst by adding an equivalent of Si/B=10 of boron from boric acid
precursor solution and/or 3 wt% equivalence of Co from CoCl2 solution, respec-
tively, to the solution of SBA-15-Mo before the hydrothermal treatment. The details
of the catalysts’ description can be found in Table SI-1. The as-prepared catalysts
described in Table 1 were pelletized to within 300–500 microns before activation
and HDS evaluation.
Characterization of catalysts
The catalysts surface area (SA), pore size and pore volume (PSPV) measurements
were conducted on a Micromeritics ASAP 2020. The analysis involves degassing
of catalysts at 250 °C for 3 h; afterwards, the N2 isotherms (adsorption–desorption)
were measured at 77 K. The XRD pattern of the catalysts was measured on a Rigaku
Ultima IV X-ray difractometer in the range 2θ=10°–70° and with a scanning speed
of 10°/min. The mesoporous difraction peak of SBA-15 was used to determine pore
wall thickness of catalysts after incorporation metals, using SAXS. The functional
groups in the catalysts were elucidated through the HORIBA, iHR320 Raman spec-
troscope with a CCD detector and a laser wavelength of 532 nm. UV–Vis difuse
refectance spectroscopy (DRS) studies were done on difuse refectance-attached
Hitachi U-4100 UV–Vis spectrophotometer. The elemental analysis was determined
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
1 3

S. A. Ganiyu
by ICP-OES PlasmaQuant Analytic Jena, after the sample (about 100 mg) was well
digested in an acidic mixture containing concentrated HNO3 and HCl (9:1, respec-
tively). The analysis was done in the region of 200–800 nm at ambient temperature.
The H2-TPR and NH3-TPD analyses of the catalysts’ oxides were measured using
an AutoChem II-2920 Micromeritics Chemisorption analyzer with hydrogen and
ammonia, a probe molecule, respectively. The analyses were performed by degas-
sing 100 mg of the catalysts at 500 °C for one hour and cooling to room tempera-
ture, both under the 99.99% pure helium fow. Later, 10% H2 in helium (for H2-TPR)
was fowed through the catalysts at a 20 ml/min rate, and the amount of H2 con-
sumed at the reducible temperature (elevated to 1000 °C at a 10 °C/min rate) was
recorded. Similarly, the NH3-TPD was carried out by fowing 10 wt.% NH3 to the
degassed catalysts at 100 °C for 0.5 h, and the surplus NH3 was ejected by purging
99.99% pure helium for 1 h. The catalysts were subsequently heated to 800 °C at
10 °C/min, and the desorbed ammonia was recorded with a thermal conductivity
detector (TCD). In situ pyridine FTIR analysis was performed by fowing pyridine
vapor for 30 min to the catalysts, and the excess pyridine was expelled by degas-
sing at 150 °C. Before that, the catalysts were pretreated under vacuum at 500 °C
for 1 h, and after that, the temperature was brought down to 150 °C. The structure
of catalysts (in the sulfded form) was visualized with the aid of a feld emission
scanning electron microscope (FESEM) (TESCAN, LYRA 3). Chemical (oxidation)
states of the sulfded catalysts elements were evaluated by X-ray photoelectron spec-
troscopy (XPS) using a PHI 5000 Versa Probe II ULVAC-PHI Inc. spectroscope,
and the analysis was conducted with mini Al Kα radiation maintained under ultra-
high vacuum. During analysis, the chamber’s pressure was 9×10−8 mBar, while
the sample was mounted on the carbon tape. The Avantage software using Gaussian-
type baseline was employed for peak deconvolution, and the analysis was corrected
for adventitious carbon at 285 eV.
Catalyst sulfdation and activity evaluation
The catalysts (in the pelletized form) were sulfded by frst reducing the Co and Mo
oxides under the fow of 5% H2/He at 400 °C for 3 h and presulfding the reduced
metals under the fow of 2 wt.% CS2 in cyclohexane at 350 °C for 5 h. The HDS
activity studies of the activated catalysts were done on a Micro Bench Top Reactor
(Parr 4590) ran at a pressure of 5 MPa, temperature of 350 °C and stirring rate of
100 rpm. About 15 ml of fuel in either form of 1000 ppm DBT in dodecane (fuel I)
or 1000 ppm MDBT in commercial diesel (fuel II) was added to the reactor vessel
containing 25 mg of the activated catalysts. The HDS reaction was run for 3 h after
attaining a stabilized reaction condition, and sampling of reaction products was done
at one and a half-hour intervals. The amount of sulfur and the reaction products
were analyzed using GC–SCD and GC–MS, respectively. The products quantifca-
tion from the instruments was triplicated to ensure that the uncertainty and errors
associated with the instrument are within the acceptable standard, having previously
standardized and stabilized the reactor used for the performance of the catalysts.
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
Results and discussion
Textural properties studies of catalysts
The surface area and pore size distribution (PSD) of all the sulfded catalysts
are presented in Table 1. It was observed that the SM and SMC catalysts have
nearly the same BET surface area; however, a lower percentage of the BET sur-
face area is available as mesoporous surface area in SMC than in SM. This is
ascribed to the loading of Co on the SM catalyst’s surface, which blocked some
of the mesopores in the support and ultimately led to a decrease in pore size
(adsorption branch is used) and pore volume of catalyst and increase in the cat-
alyst microporosity as depicted in Fig. 1. However, incorporating boron to the
SM catalyst boosts the BET surface area to 580 m2/g, with almost 90% available
as mesoporous surface area. Generally, the introduction of heteroatoms such as
Ce and Ti is known to expand the silica structure due to the heteroatoms’ rela-
tively large atomic sizes [39, 40]. It was similarly observed that Co to BSM’s
addition to forming BSMC result in the decrease in BET surface area and poros-
ity of the catalyst, as shown in Fig. 1. The low/small-angle XRD (SAXS) was
used to investigate the mesoporosity, pore wall thickness of the pure SBA-15 and
the catalysts, and the comparison was made with respect to textural properties at
mesoporous level. The pore wall thickness of SBA-15 was found to be 4.68 nm,
while the incorporation of Mo and Co without B reduces the thickness. However,
upon incorporation of boron in BSM and BSMC, it redisperses the Mo and Co
on SBA-15 with better wall thickness closer to reference SBA-15 material syn-
thesized in this study. The SAXS for all the catalysts and SBA-15 is presented
as inset in Fig. 1 PSD. The boron efect on the catalyst characteristic outweighs
that of cobalt, as evident by surface area, pore volume and pore wall thickness.
The elemental analysis of catalyst’s components was determined and reported in
Table SI-1, the total mole ratio of Co/Co + Mo in BSMC catalyst was found to be
0.29, and a good agreement with the theoretical value at 0.30.
Fig. 1 N2 adsorption–desorption isotherm of the catalysts and the pore size distribution plot
1 3

S. A. Ganiyu
Difraction pattern of catalysts
The XRD pattern (Fig. 2 left) of sulfded catalysts shows a broad difraction peak
at nearly 25 2θ (°) in all the catalysts, and this is ascribed to the amorphous nature
of silica-based SBA-15 support [30, 41]. There was no difraction pattern of boron
oxide in all the catalysts, implying the incorporation of boron oxide into the silica
framework [42]. Perhaps the amount of boron oxide is below the detectable level by
the XRD. Similarly, no difraction pattern of the active metals oxides was observed
except in SM and BSM, where weak peaks at 24, 26 and 28 2θ (°) were noticed, and
these peaks were assigned to the difraction pattern of crystalline MoO3 [43]. This
implies that the presence of Co in SMC and BSMC catalysts has enhanced both the
dispersion and sulfdation of the MoO3 in the SBA-15 support, through surface acid-
ity and other characteristics as explained vide infra.
Types of acidity by pyridine‑FTIR
The presence of Lewis and Brønsted acidity of the catalysts was evaluated by
adsorption of pyridine under vacuum using FTIR vibrational spectroscopy. The
acidic centers of the catalysts were measured through pyridinium ion peak assign-
ments resulted from acid–base interaction between the pyridine (base) and the acid
sites (Lewis and Brønsted). The coordinatively unsaturated ionic metal oxide cati-
ons at the surface are related to the Lewis acid sites, while the associated surface
OH groups through adsorbed water are assigned to Brønsted sites [44]. As shown
in Fig. 2 (right), the strong Lewis site is noticeable at 1450 and 1610 cm−1, while
the weak Lewis site peak is extremely low. The contribution from both Lewis and
Brønsted acid sites is noticed at 1492 cm−1, with Brønsted acid site peaks observed
at 1542 and 1640 cm−1, accordingly. The relative amount of both acidity types gave
insights into each catalyst’s distribution concerning their composition (see Table 1).
The BSM and BSMC possessed higher acidity (Lewis and Brønsted) than SM and
SMC, incorporating boron into the catalysts. This observation is expected to bring
about an improved performance to both catalysts [45].
Fig. 2 XRD pattern (left) and pyridine-FTIR (right) of the catalysts
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
Fig. 3 Raman spectra and DRS-UV–Vis of the catalysts
Table 2 NH3-TPD and H2-TPR of the catalysts. (Error= 3.0%)
Catalysts
H2-TPR
NH3-TPD
Peak Temp. (s)
Quantity (mmol/g STP)
Peak Temp. (s)
Quantity (mmol /g STP)
SM
160, 428, 671
133, 429, 670
132, 442, 626
137, 438, 620
111.1, 13.7, 6.9
119.0, 42.1, 12.4
122.7, 26.7, 11.0
121.4, 49.0, 14.5
239, 654
228, 721
216, 726
206, 668
0.223, 0.007
0.314, 0.006
0.317, 0.007
0.318, 0.014
SMC
BSM
BSMC
H2‑TPR of the catalysts
The H2-TPR of the catalysts oxides informs the ease of reducing the catalysts, and the
amount of H2 consumed during the reduction process. The catalysts’ TPR profles and
the amount of H2 consumed by the catalysts are presented in Fig. 3 and Table 2, respec-
tively. The main reduction temperatures for MoO3–MoO2 and MoO2–Mo(0) were
observed at 428–442 °C and 620–671 °C, respectively. It was noted that the boron-
incorporated catalysts present lesser reduction temperature for Mo2+–Mo(0) and also
have more quantity of consumed H2 than the corresponding non-boron-containing cata-
lysts. This further confrms that boron aids the dispersion and ease of reducibility of
the MoO3 in the boron-containing catalysts. The ease of reducibility of boron-modifed
catalysts (BSM and BSMC) pointed to the active phase’s availability to a high degree
of sulfdation and dispersion, achieved via moderate support interaction for efcient
HDS activities.
1 3

S. A. Ganiyu
NH3‑TPD of catalysts
The NH3-TPD analysis is a useful tool employed to explicate the relative amount
of acid sites present in a catalyst oxide [46]. The amount of ammonia desorbed is
shown in Table 2, and the TPD profle of the catalysts is shown in Fig. 3b. The TPD
profles showed that desorption occurs in two temperature ranges: 206–239 °C and
654–726 °C, which correspond to weak/moderate acid sites and strong acid sites,
respectively [47]. The amount of ammonia desorbed by the catalysts increased in
the sequence SM<SMC<BSM~BSMC in the 206 – 239 °C temperature range.
Precisely, B to SM’s addition increases the catalyst’s moderate acid sites by 42%,
and almost the same acidity efect was noticed upon the incorporation of Co to SM.
However, the amount of moderate acid sites in BSMC did not signifcantly difer
from SMC and BSM except in the strong acid sites, which is 100% more than in
SMC and BSM. It is largely reported that the number of active sites is related to the
amount of surface acidity obtained through the number NH3 desorbed from the tem-
perature-programmed analysis [7]. Evidently, the boron addition to BSM and BSMC
catalysts enhances the surface acidity for efcient HDS reaction.
Raman spectra of catalysts
Raman spectroscopy is a great tool that is often used to understand the Mo com-
pound’s chemical structure [48]. The Raman spectra of the sulfded catalysts, pre-
sented in Fig. 4a, showed peaks at 605 and 850 cm−1, characteristics of the hydroxyl
group’s vibrational stretching mode Si–OH and the symmetrical stretching of the
Si–O–Si groups due to the SBA-15 support. These peaks are more intense in the SM
and BSM catalysts and weaker in the SMC and BSMC catalysts, which implies that
the loading of Co has resulted in some Co–O–Si interactions that probably afect
the stretching of the Si–OH and Si–O–Si bonding in the SMC and BSMC catalysts.
Notably, the Raman peaks characteristics of boron oxide were not observed in both
BSM and BSMC catalysts, confrming the XRD results that boron oxide has been
incorporated into the SBA-15 framework. Other Raman peaks were observed at 284,
Fig. 4 H2-TPR and NH3-TPD of the catalysts in their metal oxide form
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
650, 810 and 980 cm−1 in SM and BSM catalysts, and these peaks correspond to the
wagging, symmetrical and asymmetrical stretching respectively of the Mo–O–Mo in
MoO3. These peaks were mostly in low intensity in the BSMC catalyst, thus afrm-
ing that most of the MoO3 have been reduced and sulfded in the BSMC than in the
other catalysts.
UV–Vis (DRS) spectra of catalysts
The DSR is explored to have some insights into the structure and electronic states of
catalysts oxides [49]. The DRS spectra of all the catalysts are presented in Fig. 4b,
and in all the spectra, there was no absorption band due to SBA-15 since it is trans-
parent in the UV–Vis region of the spectra. The broad absorption band recorded at
200–350 nm in all the catalysts are characteristics of octahedral molybdate, owing
to ligand-to-metal charge transfer from oxygen to Mo6+. It is of particular interest
to note that incorporating boron to SM and SMC has led to a signifcant decline in
the absorption band’s intensity, which implies that the incorporation of boron has
enhanced the metal–support interaction result in better dispersion of the active met-
als. Furthermore, the presence of single band in BSM and BSMC pointing to the
fact that Mo species are mainly ascribed to the polymolybdate Oh Mo types. The
absorption band at 520 nm in SMC and BSMC is ascribed to the Co2+ in the tetrahe-
dral coordination [49].
FESEM of catalysts
The FESEM analysis of the sulfded catalysts (Fig. SI-1 and Fig. 5) shows dispersed
micro-sized hexagonal rods that are consistent with the structure of SBA-15 [50],
and the dimension of the rods was noticed to have increased in the boron-modifed
Fig. 5 FESEM of BSM and BSMC catalysts
1 3

S. A. Ganiyu
catalysts, revealing characteristics structure of modifed mesoporous SBA-15 [51]. It
is essential to mention that the morphology of the SBA-15 was preserved even after
incorporating other catalyst’s components (Co, Mo, B and S) through their respec-
tive precursors. This pointed out that the strategy adopted for preparation is success-
ful, and the ordered mesoporosity properties of the SBA-15 were transferred to the
sulfded HDS catalysts. The elemental analysis of sulfded catalysts was observed by
EDX as shown in Fig. SI-2-5. It was observed that all catalysts exhibited better dis-
persion of each component, with emphasis on the boron-incorporated catalyst which
showed high dispersion of sulfur.
XPS analysis of the catalysts
The oxidation states of boron, molybdenum and sulfur elements in the catalysts were
identifed and quantifed by analyzing the catalysts XPS spectra. The XPS spec-
tra deconvolution of the catalysts shows that the catalysts (BSM and BSMC) have
boron in the chemical state of boric oxide, B2O3 (Fig.SI-6) implies that the boron
heteroatom is incorporated into the silica structure in the catalysts. The XPS spec-
tra of Mo (Fig. 6) in all the catalysts show that Mo exists in three chemical states:
Fig. 6 Deconvoluted XPS spectra of the catalysts showing the chemical states of Mo
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
Table 3 Atomic percent of chemical states of Mo and S observed in the catalysts
Mo3d 228.5 eV
Mo3d 231.5 eV
Mo3d 234.8 eV
S2p 162.3 eV
S2p 170.0 eV
SM
–
–
50.05
48.25
39.95
38.82
23.36
39.41
19.85
18.69
13.93
12.34
19.99
26.47
12.66
–
10.53
SMC
BSM
BSMC
9.68
16.02
–
Fig. 7 HDS performance plot of boron-modifed catalysts. Fuel=1000 ppm dibenzothiophene in dode-
cane
Mo3d5/2 (228.5 eV), Mo3d5/2 (230.4 eV) and Mo3d5/2 (232.5 eV), and these cor-
respond to the (Mo4+(3d5/2)), (Mo5+(3d5/2) and Mo6+(3d5/2)), respectively [12]. The
relative amount (%) of the molybdenum species in all the catalysts is summarized in
Table 3. It was observed that BSM and BSMC catalysts have Mo4+(3d5/2) species,
which is an indication of the presence of surface MoS2 in these catalysts. Also, the
percentage of MoS2 in BSMC was higher than in BSM, which may imply the BSMC
catalyst’s better reactivity than the others in the series. The sulfur chemical states
in the catalysts (Fig. SI-8) also show two sulfur species: the sulfde and the oxy-
sulfde observed at 162.3 eV and 170 eV, respectively. Correspondingly, the sulfde
in the catalysts is highest in BSMC, which corroborates the claim that BSMC has
more MoS2 phases that the other catalysts. The incorporation of boron resulted into
improved surface acidity, textural properties, better reduction potential and presence
of polymolybdate (Oh), which can produce highly reducible and dispersed sulfded
phase necessary for efcient HD operation. These observations are supported by
TPD, pyridine-FTIR, surface area and porosity, TPR and DRS analysis, respectively.
Catalysts activity studies using fuel I
The catalysts’ activity was frst studied using fuel I, consisting of 1000 ppm DBT in
dodecane. The catalysts’ HDS performance plot is presented in Fig. 7, and the sum-
marized percentage conversion is shown in Table SI-2. The percent DBT removal
after one and a half hour of HDS reaction for SM catalyst is 68%, and the SMC
1 3

S. A. Ganiyu
has an increased conversion, of 73%, due to the Co-promotion efect. Interestingly,
the boron-modifed SM catalyst (BSM) showed a much higher conversion of 90%.
In comparison, the boron-modifed SMC (BSMC) presented conversion of nearly
82% at the one and a half hour of HDS reaction. It was further observed that the
percent conversion increases after the three-hour reaction time, and the SMC and
BSMC catalysts presented the highest increase, by 19.8% and 17.9% of the initial
conversion, respectively, compared to the other catalysts (SM and SMC). The obser-
vations show that incorporating boron to SM improved the catalyst activity better
than incorporating Co to SM, at least at the initial reaction stage. This is related to
both the textural and structural properties of the catalysts. As discussed in Sect. 3.1,
the low mesoporous surface area of SMC due to Co incorporation is why the cata-
lyst’s HDS activity is lower than that of BSM. More so, the TPR and TPD results
have shown that boron heteroatom increased both moderate acid sites and metals
dispersion, which are required for better HDS activity. However, the catalysts’ activ-
ity is subject to its strong resistance to sulfur poisoning due to H2S gas generation at
a longer reaction time. Comparatively, the BSM activity after 3-h reaction time was
slightly lower than expected, whereas the BSMC activity is already at 99.9%. The
increase in performance of BSMC, which is by almost 18% of the initial conversion,
indicated that the HDS reaction is more favored in the presence of both boron and
cobalt. It is well reported that the incorporation of promoter (Ni or Co) favors Mo
species’ dispersion to achieve more Type-II active phase for better conversion of
organosulfur compounds [52]. An analogous catalyst containing alumina-modifed
SBA-15-CoMo (ASMC) was tested and evaluated in the same HDS reaction with
DBT to establish the favorable acidity with BSMC. BSMC conversion rates at 90
and 180 min were 4% and 12% higher than the alumina-based catalyst, respectively.
The observed results indicated that the boron being less acidic introduced moder-
ate acidity with excellent active phase dispersion than alumina. The rate constants
for the HDS of DBT reaction at both 1.5-h (kHDS (90 min)) and 3-h (kHDS (180 min))
reaction time and their ratio (kHDS (180 min)/kHDS (90 min)) for all the catalysts
(Table SI-2) further support the claim that the BSMC catalyst has higher activity
than the other catalysts in the series after 180 min of reaction, and this is due to
better structural characteristics such as improved surface area, pore wall thickness,
better acidity and presence of highly reducible polymolybdate (Oh) possessed by the
catalyst over others. The product analysis of DBT in model fuel was evaluated at
a half-hour reaction interval when the HDS catalytic conversion of most catalysts
was below 50% except in the BSM catalyst. Generally, the HDS reaction of DBT
proceeds via two most noticeable routes; direct sulfur cleavage from organosulfur
aromatic ring (known as direct desulfurization, DDS) to form biphenyl (BP). The
hydrogenation of the aromatic ring undergoes intermediate products tetrahydro-
dibenzothiophene (THDBT) and hexahydro-dibenzothiophene (HHDBT) before
cleavage of C-S to form cyclohexylbenzene (CHB) as an alternative reaction route
[53]. As shown in Table 4, each catalyst’s products’ distribution varied according to
the surface characteristics. BSM and BSMC catalysts’ reactivity was higher than SM
and SMC at the 30 min reaction interval, as observed by their higher rate constants.
This is due to the high amount of Lewis and Brønsted acidity possessed by these
catalysts (see Table 1). Furthermore, the predominant reaction route for all catalysts
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
Table 4 Catalyst performance test results: product distribution and frst-order rate constants for HDS of
DBT at 350 °C at 30 min
Catalysts %-S conversion Product distribution
(wt.%)
kHDS ×103 kDDS ×103 kHYD ×103 kDDS/kHYD
(min−1
)
(min−1
)
(min−1
)
CHB BP
THDBT
SM
32
39
52
47
41
16.3 82.6 1.1
14.8 85.1 0.1
20.1 76.5 3.4
18.6 79.5 1.9
22.2 75.4 2.4
12.9
16.5
24.5
21.2
17.6
10.6
14.0
18.7
16.8
13.3
2.3
2.5
5.8
4.4
4.3
4.7
5.7
3.3
3.9
3.1
SMC
BSM
BSMC
ASMC
is through direct scission of C–S bond (DDS), while the ratio of reactivity of direct
desulfurization to hydrogenation (kDDS/kHYD) varies. It should be mentioned that the
hydrogenation rate constants (kHYD) of BSM and BSMC are higher than SM and
SMC, indicating the tendency of the reactant to react through the additional route
with sufcient surface acidity as observed by pyridine FTIR [54].
Catalysts activity studies using fuel II
The observed results in Sect. 3.9 motivated the further study of the catalysts’ per-
formance in a more complex fuel system that contains 1000 ppm of MDBT dis-
solved in commercial diesel fuel. The commercial diesel consists of saturated and
aromatic hydrocarbons, which can react competitively during the HDS reaction. The
refractory sulfur compound, MDBT, can undergo an isomerization reaction before
the HDS reaction, and the strength of the catalyst’s acid sites signifcantly afects
the isomerization process. The activity of the catalysts in the HDS of MDBT in die-
sel fuel is plotted in Fig. 8. As expected, the BSMC catalyst performed better than
HDS Acꢀvity @ 350 oC
100
89.2
87
80.9
80
60
40
20
0
71.6
66.9
60.9
90 min
180 min
SMC
BSM
Catalysts
BSMC
Fig. 8 HDS performance plot of boron-modifed catalysts. Fuel=1000 ppm methyldibenzothiophene in
diesel
1 3

S. A. Ganiyu
the other catalysts in the series, and it is observed that the SMC catalyst performs
slightly better than the BSM catalyst. This further confrmed the BSM catalyst’s
high activity is only at ab initio in the dodecane model fuel. In a more complex
system like commercial diesel, the activity is low at the onset of the reaction. This
low activity is due to the BSM catalyst’s moderate acidity and the absence of a Co-
promoter. Unlike the BSM, the BSMC has both strong acid sites and Co-promoting
efect. There is an electronic interaction due to the B and Co species, typical of a
synergetic efect that could withstand the complexity’s reaction matrix.
Furthermore, signifcant surface acidity promotes additional HDS reaction con-
version routes through hydrogenation and isomerization, especially for bulky sul-
fur molecules [55]. The efect of operating temperature on the HDS activity of the
BSMC catalyst was further studied using the diesel fuel containing 1000 ppm of
MDBT. The observed result (Fig. 9) showed that the catalyst activity increases with
temperature. The activation energy (Ea) of the MDBT HDS reaction was subse-
quently determined using the rate constants obtained at the diferent temperatures.
The Ea was obtained by plotting the natural logarithms of the rate constants versus
the reciprocal of the temperatures (in kelvin), and the plot’s slope, which is equal
to -Ea/R, was used to calculate the Ea. The obtained Ea (77.108 KJmol−1) is lower
than most of the previously reported activation energies of MDBT HDS reactions
[56–58] which further corroborate the efectiveness of the BSMC catalyst in the
HDS reaction of MDBT.
Conclusion
The role of boron heteroatom in the modifcation of (Co)Mo supported on SBA-15
catalyst has been studied by single-pot preparation of the catalysts. The catalysts:
SM, SMC, BSM and BSMC have been tested for the HDS of DBT in dodecane
model fuel and HDS of MDBT in commercial diesel. The BSM and BSMC catalysts
performed best in the HDS of DBT, whereas the BSMC catalyst possessed superior
Fig. 9 Efect of process temperature on the HDS performance of BSMC catalyst. Fuel=1000 ppm
methyldibenzothiophene in diesel
1 3

Synergy efect of boron and cobalt in B2O3‑SBA‑15‑(Co)Mo…
activity in the HDS of MDBT. The competitive performance of BSM catalyst in the
HDS of DBT relative to SM and SMC catalysts was ascribed to its large mesoporous
surface area, moderate acid sites, ease of reducibility and good dispersion of the
active Mo metal. The exceptional performance of BSMC in the HDS of both model
and diesel fuels is considered due to its strong acid sites, the Co-promoting efect,
high degree of sulfdation and dispersion and the electronic interaction between the
boron heteroatom and the active metals in the catalyst. The BSMC catalyst is suit-
able for potential ultradeep HDS of transportation fuels with alkylated derivatives of
DBT.
Supplementary Information The online version contains supplementary material available at https://doi.
org/10.1007/s11164-021-04475-x.
Acknowledgements The fnancial support provided by King Fahd University of Petroleum & Minerals
(KFUPM) for funding this work through Project No. DSR NUS15105 is greatly acknowledged.
References
1. C. Song, Catal. Today 77, 17 (2002)
2. W. Lai, L. Pang, J. Zheng, J. Li, Z. Wu, X. Yi, W. Fang, L. Jia, Fuel Process. Technol. 110, 8 (2013)
3. M. Zhang, J. Fan, K. Chi, A. Duan, Z. Zhao, X. Meng, H. Zhang, Fuel Process Technol 156, 446
(2017)
4. H. Song, F. Zhang, N. Jiang, M. Chen, F. Li, Z. Yan, Res. Chem. Intermed. 44, 5285 (2018)
5. S.T. Hong, D.R. Park, S.J. Yoo, J.D. Kim, H.S. Park, Res. Chem. Intermed. 32, 857 (2006)
6. H. Song, Q. Yu, N. Jiang, Z. Yan, T. Hao, Z. Wang, Res. Chem. Intermed. 44, 3629 (2018)
7. D. Zhang, A. Duan, Z. Zhao, C. Xu, J. Catal. 274, 273 (2010)
8. G. Pérot, Catal. Today 86, 111 (2003)
9. Y. Saih, K. Segawa, Appl. Catal. A Gen. 353, 258 (2009)
10. T. Li, A. Duan, Z. Zhao, B. Liu, G. Jiang, J. Liu, Y. Wei, H. Pan, Fuel 117, 974 (2014)
11. S. Badoga, A.K. Dalai, J. Adjaye, Y. Hu, Ind. Eng. Chem. Res. 53, 2137 (2014)
12. A. Tanimu, S.A. Ganiyu, S. Adamu, K. Alhooshani, React. Chem. Eng. 4, 724 (2019)
13. A.A. Asadi, S.J. Royaee, S.M. Alavi, M. Bazmi, Fuel Process. Technol. 187, 36 (2019)
14. A. Safari, M. Vesali-Naseh, J. Clean. Prod. 220, 1255 (2019)
15. Y. Dong, Y. Xu, Y. Zhang, X. Lian, X. Yi, Y. Zhou, W. Fang, Appl. Catal. A Gen. 559, 30 (2018)
16. Q. Gao, Y. Zhang, K. Zhou, H. Wu, J. Guo, L. Zhang, A. Duan, Z. Zhao, F. Zhang, Y. Zhou, RSC
Adv. 8, 28879 (2018)
17. X. Zhou, A. Duan, Z. Zhao, Y. Gong, H. Wu, J. Li, Y. Wei, G. Jiang, J. Liu, Y. Zhang, J. Mater.
Chem. A 2, 6823 (2014)
18. M.A. López-Mendoza, R. Nava, C. Peza-Ledesma, B. Millán-Malo, R. Huirache-Acuña, P. Skewes,
E.M. Rivera-Muñoz, Catal. Today 271, 114 (2016)
19. C. Suresh, L. Pérez-Cabrera, J.N. Díaz de León, T.A. Zepeda, G. Alonso-Núñez, S.F. Moyado,
Catal. Today 296, 214 (2017)
20. H. Li, Y. Sakamoto, Z. Liu, T. Ohsuna, O. Terasaki, M. Thommes, S. Che, Microporous Mesoporous
Mater. 106, 174 (2007)
21. K. Chandra Mouli, S. Mohanty, Y. Hu, A. Dalai, J. Adjaye, Catal. Today 207, 133 (2013)
22. W. Fu, L. Zhang, T. Tang, Q. Ke, S. Wang, J. Hu, G. Fang, J. Li, F.-S. Xiao, J. Am. Chem. Soc. 133,
15346 (2011)
23. P. Jabbarnezhad, M. Haghighi, P. Taghavinezhad, Fuel Process. Technol. 126, 392 (2014)
24. C.E. Santolalla-Vargas, V.A.S. Toriello, J.A. de los Reyes, D.K. Cromwell, B. Pawelec, J.L.G.
Fierro, Mater. Chem. Phys. 166, 105 (2015)
25. J. Chen, F. Maugé, J. El Fallah, L. Oliviero, J. Catal. 320, 170 (2014)
26. R. Palcheva, L. Kaluža, L. Dimitrov, G. Tyuliev, G. Avdeev, K. Jirátová, A. Spojakina, Appl. Catal.
A Gen. 520, 24 (2016)
1 3

S. A. Ganiyu
27. X. Guo, M. Song, X. Zhao, L. Zhao, J. Fuel Chem. Technol. 44, 1326 (2016)
28. W. Cui, W. Li, R. Gao, H. Ma, D. Li, M. Niu, X. Lei, Energy Fuels 31, 3768 (2017)
29. O.Y. Gutiérrez, S. Singh, E. Schachtl, J. Kim, E. Kondratieva, J. Hein, J.A. Lercher, ACS Catal. 4,
1487 (2014)
30. S.A. Ganiyu, K. Alhooshani, Energy Fuels 33, 3047 (2019)
31. Y. Wang, H. Li, J. Yao, X. Wang, M. Antonietti, Chem. Sci. 2, 446 (2011)
32. Y. Zheng, Y. Jiao, L. Ge, M. Jaroniec, S.Z. Qiao, Angew. Chemie Int. Ed. 52, 3110 (2013)
33. A.-M. Alexander, J.S.J. Hargreaves, Chem. Soc. Rev. 39, 4388 (2010)
34. Y. Cao, H. Yu, J. Tan, F. Peng, H. Wang, J. Li, W. Zheng, N.-B. Wong, Carbon N. Y. 57, 433 (2013)
35. T.A. Saleh, S.A. AL-Hammadi, A.M. Al-Amer, Process Saf. Environ. Prot. 121, 165 (2019)
36. A. Usman, T. Kubota, Y. Araki, K. Ishida, Y. Okamoto, J. Catal. 227, 523 (2004)
37. F. Rashidi, T. Sasaki, A.M. Rashidi, A.N. Kharat, K.J. Jozani, J. Catal. 299, 321 (2013)
38. D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science 279,
548 (1998)
39. A. Tanimu, K. Alhooshani, Energy Fuels 33, 2810 (2019)
40. S.A. Ganiyu, S.A. Ali, K. Alhooshani, Ind. Eng. Chem. Res. 56, 5201 (2017)
41. P. Velmurugan, J. Shim, K.-J. Lee, M. Cho, S.-S. Lim, S.-K. Seo, K.-M. Cho, K.-S. Bang, B.-T. Oh,
J. Ind. Eng. Chem. 29, 298 (2015)
42. Q. Dai, X. Wang, G. Chen, Y. Zheng, G. Lu, Microporous Mesoporous Mater. 100, 268 (2007)
43. J. Gong, W. Zeng, H. Zhang, Mater. Lett. 154, 170 (2015)
44. G. Busca, Catal. Today 41, 191 (1998)
45. S.A. Ganiyu, S.A. Ali, K. Alhooshani, Chem. Sel. 4, 370 (2019)
46. S.A. Ganiyu, S.A. Ali, K. Alhooshani, RSC Adv. 7, 21943 (2017)
47. F. Lónyi, J. Valyon, Microporous Mesoporous Mater. 47, 293 (2001)
48. Y. Li, Z. Feng, Y. Lian, K. Sun, L. Zhang, G. Jia, Q. Yang, C. Li, Microporous Mesoporous Mater.
84, 41 (2005)
49. J.A. Mendoza-Nieto, O. Vera-Vallejo, L. Escobar-Alarcón, D. Solís-Casados, T. Klimova, Fuel 110,
268 (2013)
50. S. Che, K. Lund, T. Tatsumi, S. Iijima, S.H. Joo, R. Ryoo, O. Terasaki, Angew. Chemie Int. Ed. 42,
2182 (2003)
51. S.A. Ganiyu, K. Alhooshani, S.A. Ali, Appl. Catal. B Environ. 203, 428 (2017)
52. S.L. Amaya, G. Alonso-Núñez, T.A. Zepeda, S. Fuentes, A. Echavarría, Appl. Catal. B Environ.
148, 221 (2014)
53. S.A. Ali, S. Ahmed, K.W. Ahmed, M.A. Al-Saleh, Fuel Process. Technol. 98, 39 (2012)
54. Y. Li, K. Chi, H. Zhang, P. Du, D. Hu, C. Xiao, H. Li, Z. Zhao, A. Duan, C. Xu, Fuel Process. Tech-
nol. 180, 56 (2018)
55. Q. Yu, L. Zhang, R. Guo, J. Sun, W. Fu, T. Tang, T. Tang, Fuel Process. Technol. 159, 76 (2017)
56. J. Ancheyta, M.J. Angeles, M.J. Macías, G. Marroquín, R. Morales, Energy Fuels 16, 189 (2002)
57. A. Venezia, V. La Parola, G. Deganello, B. Pawelec, J.L. Fierro, J. Catal. 215, 317 (2003)
58. E. Reséndiz, A. Rosales-Quintero, G. Marroquín, Fuel 86, 1247 (2007)
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional afliations.
1 3