Fuel 351 (2023) 128890

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Hydro-co-processing of a jatropha oil and gas oil blend with a sulfided

Ni–W catalyst supported on mesostructured materials Al(x)-SBA-15 type for
cleaner hybrid diesel production: Effect of the Al/Si molar ratio
Jonatan R. Restrepo-Garcia a, Diana Gomora-Herrera b, Pablo Torres-Mancera b, Ignacio Elizalde-
Martínez a, *
a
Centro Mexicano para la Producción más Limpia (CMP+L), Instituto Politécnico Nacional, Av. Acueducto s/n, Barrio La Laguna, Col. Ticomán, Gustavo A. Madero, C.
P. 07340, Mexico City, Mexico
b
Instituto Mexicano del Petróleo (IMP), Eje Central Norte Lázaro Cárdenas 152, Col. San Bartolo Atepehuacan, C.P.07730, Mexico City, Mexico

A R T I C L E I N F O A B S T R A C T

Keywords: Obtaining cleaner hybrid diesel fractions by hydro-co-processing vegetable oil and gas oil blends requires
Hydrotreating tailoring intrinsic properties (metal content and acidity) of a solid catalyst. Such catalyst must promote the
Al-SBA-15 hydrodesulfurization (HDS), hydrodeoxygenation (HDO), and hydrocracking (HCK) reactions. In that sense, the
Ni–W sulfides
present research work reports the synthesis (sol–gel and hydrothermal), characterization (N2 physisorption, XRD,
Vegetable oil blends
Green diesel
HR-TEM, 27Al-MAS-NMR, Pyridine-FTIR, and RAMAN spectroscopy), and catalytic evaluation (6 MPa of H2
initial pressure, 360 ◦ C, and 4 h) for Ni–W/Al(x)-SBA-15 sulfided catalysts during the hydroprocessing of a 20 vol
% of Jatropha Curcas L. oil blend with a mixed gas oil (50 vol% LGO and 50 vol% HGO). Metal content of the
catalysts was fixed at 2.5 wt% of NiO and 15 wt% of WO3, meanwhile, the Al/Si (x) molar ratio was adjusted in
0.1, 0.05, 0.033, 0.025, and 0.0 to study the effect of aluminum (Al) content during hydroprocessing. Liquid
products were analyzed by FTIR, 1H NMR, ESI-MS, Simulated Distillation (ASTM D2887) and ASTM D4294 for S
content. The effect of Al incorporation was mainly observed in the hydrotreating reactions (HDO and HDS) rather
than HCK reactions. Intermediate Al/Si molar ratios (0.05, and 0.03) promoted the highest sulfur removal of
about 13.1 % compared with the 10.7 % and 5.6 % for 0.1 and 0.025 Al/Si molar ratios, respectively. In addition,
the highest content of Brønsted (0.05 mmol g− 1) and Lewis (1.45 mmol g− 1) acid sites was observed for the
Ni–W/Al(0.05)-SBA-15 catalyst. Regarding HDO activity, it was the highest (57.3 %) for Ni–W/Al(0.1)-SBA-15
catalyst, and the lowest (30.9 %) for Ni–W/Al(0.025)-SBA-15, and Ni–W/Al(0.0)-SBA.15 sulfided catalysts.
Finally, by comparing HDS and HCK activities of Ni–W/Al(0.05)-SBA-15 and Ni–W/Al(0.0)-SBA-15 sulfided
catalysts, it was observed that HDS decreased 7.5 % and HCK increased 2.4 % when no Al was incorporated.

1. Introduction in heavier sources (e.g., vacuum gas oil). Additionally, it has been also
demonstrated that hydroprocessing mixtures of vegetable oil and
Hydroprocessing is a technology widely used to attain middle dis­ petroleum-derived feedstocks allows getting greener liquid fuels. But,
tillates (MD) with less content of Sulphur (S) and other contaminants (e. introducing the necessity of selectively removing oxygen (O) of carboxyl
g., N, V, etc.) by treating petroleum-derived feedstocks with solid cat­ and carbonyl groups or via hydrodeoxygenation (HDO) in the vegetable
alysts at conditions of high pressure and high temperature (2.5–8.8 MPa, oil (a mixture of oligomeric compounds) [2,3].
350–420 ◦ C). Bifunctional catalysts (hydrogenation/dehydrogenation Researchers have made several efforts in studying the reactivity of
and acidity) aim at selective removing heteroatoms and simultaneously commercial or up to date catalysts for hydro-co-processing mixtures of
converting heavier fractions in lighter ones [1]. Nonetheless, during the vegetable or animal oil and gas oil. Accordingly, De Paz Carmona et al.
last decades, a significant decrease in light crude sources has been studied the effect of using a Ni-Mo/ɤ-Al2O3 catalyst during hydro­
observed. This trend makes it more difficult to process larger molecules processing a 20 wt% mixture of waste cooking oil and animal fat with

* Corresponding author.
E-mail address: [email protected] (I. Elizalde-Martínez).

https://doi.org/10.1016/j.fuel.2023.128890
Received 14 December 2022; Received in revised form 14 April 2023; Accepted 1 June 2023
Available online 7 June 2023
0016-2361/© 2023 Elsevier Ltd. All rights reserved.
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

Straight Run Gas Oil (SRGO) at 350 ◦ C, 5.5 MPa, and an LHSV of 2 h− 1 %)/Al-SBA-15 catalysts for solvent free catalytic deoxygenation of PFAD
[4]. Catalytic activity evidenced a paraffin (C15-C18) yield between 79 % at 350 ◦ C [15]. Si/Al molar ratio was varied as follows: 5, 25, 50, and 75
and 85 % depending on the type of oil used, waste cooking oil or animal for direct synthesis of Al-SBA-15. Depending on the Si/Al molar ratio a
fat, respectively. El-Sawy et al. evaluated a binary catalytic bed formed different product distribution was obtained. Gasoline selectivity was
by Ni-Mo/ɤ-Al2O3 y Ni–W/SiO2/Al2O3 catalysts, respectively [5]. For higher for the Si/Al of 5 (higher occurrence of tetrahedral Al (Brønsted
such a purpose, it was evaluated the hydro-co-processing (380–440 ◦ C, acid sites). However, diesel selectivity was the highest for Si/Al molar
7 MPa, LHSV of 1.5 h− 1) of a mixture of vacuum gas oil (VGO) and waste ratio of 50. In general, a hydrocarbon yield of 86 % was attained.
cooking oil (WCO), or waste lubricating oil (WLO). Results indicated Finally, Zhang et al. tested Ni reduced catalysts supported on acidic
that a higher amount of WLO increased the yield to diesel-like fraction, MCM-41 silica functionalized with phosphotungstic acid (HPW) during
while WCO content favored kerosene and gasoline-like fractions yield, 2 h hydrocracking of methyl palmitate at 390 ◦ C, 2 MPa [16]. Metal
varying vegetable oil content from 10 wt% to 20 wt%. The highest diesel loading was fixed at 10 wt% and HPW content varied in 0, 10, 20, and
yield was attained at 400 ◦ C, demonstrating the effect of temperature in 40 wt%. The highest selectivity to Jet Fuel fraction was achieved with a
hydrocracking during hydroprocessing. Similarly, Vlasova et al. studied 20 wt% of HPW due to the highest amount of acid sites it possessed
the effect of using Mo/Al2O3, and Ni-Mo/Al2O3-SAPO-11 during co- (0.730 mmol/g).
hydroprocessing a mixture of 30 wt% of Rapeseed Oil (RSO) and Despite recent advances in hydro-co-processing of vegetable oil and
SRGO at reaction conditions of 350–380 ◦ C, 4–7 MPa, and 1–1.5 h− 1 gas oil blends, there is a current challenge in developing an enhanced
(LHSV) in a Trickle Bed Reactor (TBR) [6]. In addition, experiments catalyst to treat heavier components in gas oil part and removing oxygen
were carried out in the presence of hydrogen sulfide (H2S) and ammonia in its vegetable oil part. Up to date, it is uncertain the effect of incor­
(NH3), evidencing that NH3 content inhibited hydrocracking reactions porating Al in SBA-15 type materials when hydro-co-processing vege­
while hydroprocessing. In general, Mo/Al2O3 catalysts favored the table oil and gas oil, and there is not a comparative study between
direct HDO of RSO at a lower temperature (330 ◦ C) [7]. By their side, reduced Ni-based metallic phases with its sulfided counterpart, which is
Wijaya et al. synthesized sulfated zirconia, SO4/ZrO2, as catalytic sup­ common in hydroprocessing of fossil fuel derivatives. Therefore, the
port for hydrocracking of WCO into bio-gasoline and employing reduced present study focuses on synthesizing and evaluating Ni–W sulfided
Ni as active phase. Ni (1, 2, and 3 wt%) incorporation on SO4/ZrO2 catalysts supported on Al-SBA-15 with variable Al/Si molar ratio (0.1,
increased acidity in the material. However, selectivity to gasoline was 0.05, 0.033, 0.025, and 0.0), aiming at enhancing yield to a cleaner (less
the highest (70.3 %) with the 1 % Ni reduced catalyst at 450 ◦ C of hy­ S) diesel-like fraction during hydro-co-processing of a mixture of vege­
drocracking reaction [8]. table oil (Jatropha Curcas L. oil) and gas oil. For this purpose, metallic
Alternative catalysts have been also developed and tested. For oxides content was fixed in 2.5 wt% of NiO and 15 wt% of WO3 to
instance, Burimsitthigul et al. evaluated an unsupported Ni-Mo sulfided elucidate the effect of the Al/Si molar at fixed operation conditions in a
catalyst during the hydrotreating of palm oil with the purpose of pro­ batch reactor during the hydro-co-processing of the mixture.
ducing green fuels (0.5–3 h, 280–320 ◦ C) using decane as solvent
(85–95 wt%) [9]. Results made it clear that decarbonylation and 2. Experimental
decarboxylation mechanisms were favored at higher reaction time and
temperature (2 h, 30 bar, and 300 ◦ C), and a 75.3 % yield for a C14-C18 2.1. Synthesis of Al (x)-SBA-15 catalytic supports
fraction was reached. However, in the last decade, it has been evidenced
a growing interest in synthesizing mesoporous materials such as MCM- Five(5) silica-based Al functionalized materials were directly syn­
41 (Mobil Composition of Matter N◦ 41), SBA-15(Santa Barbara thesized adapting previously reported procedures [12,17–19]. In a
Amorphous-15), and MCF (Mesostructured Cellular Foam) silica due to typical synthesis, about 9.8 g of Pluronic P123 (EO20PO70EO20, Mav =
the enhanced textural properties (~700 m2g− 1, 1–2 cm3g− 1, and 7–12 5800, PEG 30 wt%, Aldrich) were dissolved in 235 mL of a 1.2 M
nm) in comparison to traditional ɤ-Al2O3 (1). The objective is synthe­ aqueous solution of HCl (37.3 vol%, Merck) under continuous agitation
sizing catalysts capable of overcoming diffusion limitations when (~380 rpm). A calculated amount (depending on Al/Si molar ratio) of
treating heavier feedstocks. Nevertheless, such silica-based materials do aluminum isopropoxide (C9H21AlO3, Aldrich 98 %) was dissolved in
not possess acid sites (neutral silicon framework). For that reason, ~100 mL of a 1.2 M aqueous solution of HCl (37.3 vol%, Merck) under
introducing lower valence heteroatoms (e.g., Al, Ti, Zr, etc.) promotes continuous agitation (~500 rpm). Then, both HCl aqueous solutions
the occurrence of both Brønsted and Lewis acid sites. Acid sites are were mixed and kept under agitation (~380 rpm) for 1 h. Subsequently,
responsible for hydrocracking and isomerization reactions as it has been ~22.8 mL of tetraethyl orthosilicate (SiC8H20O4, TEOS, Aldrich 98 %)
reported for several authors [10,11]. were added dropwise. Temperature was increased from room tempera­
As a result, materials such as Al-SBA-15 have been also tested during ture (~20 ◦ C) to 38 ◦ C and kept at these conditions for 24 h. The milky
hydro-co-processing of vegetable oil–gas oil blends [12]. For instance, suspension thus obtained is transferred into a PTFE closed bottle for
Kamaruzaman et al. tested Co/SBA-15, Ni/SBA-15, and Ni-Co/SBA-15 hydrothermal treatment at 110 ◦ C for 24 h, using a heating rate of 2 ◦ C/
reduced catalysts during the catalytic cracking reaction of Palm Fatty min in an oven. The solid product is filtrated, flushed with deionized
Acid Distillate (PFAD) to get a green diesel fraction [13]. 10 wt% of the water, and dried at ambient conditions for 72 h. Finally, the solid
catalyst was used in a 3 h batch reaction at 350 ◦ C. According to reported product is placed on a furnace for drying at 110 ◦ C/12 h and calcination
results, the bimetallic Ni-Co catalysts showed a higher yield to a diesel- at 550 ◦ C/6h to remove the template (Pluronic P123) using a heating
like fraction (C8-C17) in comparison to only Ni counterpart (88.1 % and rate of 2 ◦ C/min. The resulting material was labeled Al(x)-SBA-15,
85.8 % respectively). Likewise, Oh et al. tested NiCu, NiMn, and NiZn where x is the Al/Si molar ratio. The nominal molar ratio of reactants
reduced bimetallic phases supported on pristine SBA-15 during hydro­ during the synthesis was the following: 1.0 TEOS:0.0169 P123:4.42
deoxygenation (HDO) of bio-oil at 3 MPa and temperature range of HCl:186H2O according to previously reported works (1, 10). Pristine
250–350 ◦ C during a reaction time of 3 h [14]. HDO of bio-oil resulted in SBA-15 catalytic support was labeled Al(0.0)-SBA-15.
four main products: gas, char, and a liquid product formed by two
immiscible liquids (heavy oil, and light oil). Principal remarks showed 2.2. Catalyst preparation
that heavy oil fraction yield in the liquid product increased from 40.8 %
to 60.6 % by using NiMn reduced at 350 ◦ C. Such fraction could be used Five (5) Ni–W/Al(x)-SBA-15 oxide phase catalysts were prepared by
as feedstock for hydro-co-processing reactions. the successive incipient wetness impregnation method. A calculated
There are also some reports regarding the use of functionalized SBA- amount (to get ~15 wt% of WO3) of ammonium metatungstate,
15 or MCM-41. First, Baharudin et al. synthesized and tested NiO (5 wt (NH4)6H2W12O40⋅H2O (Aldrich, 85 wt% WO3) was dissolved in ~25 mL

2
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

of deionized water. The previous solution was added to ~ 5 g of Al(x)- 2.3.7. FTIR spectra of liquid products during the hydro-co-processing of the
SBA-15 and mixed until homogenization and kept static for 24 h. The vegetable oil and gas oil
resulting material dried at 110 ◦ C/12 h and calcined at 550 ◦ C/6h with a IR spectra were recorded within the range of 400 to 4000 cm− 1 in an
heating rate of 2 ◦ C/min into a furnace. Then, the recovered solid was IR 2 IlluminatIR II JY Smiths Detection apparatus coupled to an
impregnated with ~20 mL of an aqueous solution of nickel nitrate, OLYMPUS BX41 optic microscope.
NiNO3⋅6H2O (Aldrich, 97 %) to get ~2.5 wt% of NiO, kept static for 24
h, and subsequently dried and calcined into a furnace at the same con­ 2.3.8. Electrospray ionization (ESI) mass spectrometry
ditions of the previous impregnation. All the synthesized solids were ESI spectra were acquired in a Bruker Daltonics micrOTOF-Q appa­
grounded and sieved down to a particle size between 25 and 75 µm for ratus scanning within the range of 50 to 2000 m/z and employing a
reaction experiments. positive ion polarity. Spectrum data was adjusted by subtracting sodium
(Na) molecular weight (~22.99 g mol− 1) from m/z units to qualitatively
determine the corresponding hydrocarbon fraction for the representa­
2.3. Characterization tive ions obtained at the highest intensities.

2.3.1. Textural properties 2.3.9. Analysis of liquid products by 1H NMR

Samples between 0.05 and 0.1 g were outgassed at 180 ◦ C/12 h H+ Liquid state Nuclear Magnetic Resonance (NMR) analyses were
before analysis. Outgassed samples were analyzed in a Quantachrome performed in Bruker Ascend TM apparatus for all liquids products of the
ASQIA equipment for nitrogen [email protected] ◦ C. Isotherms hydro-co-processing of the vegetable oil and gas oil blend and pure
were recorded with twenty acquisition points during adsorption and Jatropha Curcas L. oil. Samples were previously dissolved in deuterated
seventeen during desorption. Specific surface area (SBET,m2g− 1) was chloroform, CDCl3. Spectra were acquired at 750 MHz employing a
determined by the BET (Brunauer, Emmet and Teller) method at relative relaxation delay of 1 s, receiver gain of 4, pulse width of 8 s, and a total
pressures (P/P0) within 0.04 and 0.25 according to IUPAC recommen­ number of scans of 32.
dation [20]. Pore volumes (cm3g− 1) was determined at 0.99 of relative
pressure, while pore diameter (nm) was determined by the BJH (Bayer 2.4. Catalytic evaluation
Joyner and Halenda) method.
Hydro-co-processing experiments were carried out in a 100 mL batch
2.3.2. Materials crystallinity by XRD reactor Parr 4598 under the following operational conditions: 360 ◦ C, 6
Previously grounded samples between 0.05 and 0.1 g were analyzed MPa of initial hydrogen (H2) partial pressure, ~2 wt% of sulfided
in a Rigaku Ultima IV X-ray diffractometer coupled with an ultra-high- catalyst, 700 rpm, and a reaction time of 4 h. Prior to the experiments, a
speed detector for low angle (2θ from 0.4◦ to 4◦ ) measurements. Wide mixture 50 % vol. of LGO(Light Gas Oil) and HGO (Heavy Gas Oil) was
angle (2θ from 10◦ to 80◦ ) measurements were performed to identify prepared and labeled as GOM (Gas Oil Mix) GOM was further mixed
changes in metallic phases and crystallinity of the synthesized catalysts. with Jatropha Curcas L. Oil (JO) to make a mixture JO/GOM volume
Copper Kα1 radiation (1.54 A◦ ) was employed with a speed of 0.5◦ min− 1 ratio of 20/80. About 50 mL of the final mixture was loaded into the
each 0.02 s. reactor with the required amount of previously sulfided Ni–W/Al(x)-
SBA-15 catalyst.
2.3.3. Porous structure by HR-TEM Sulfidation process (transforming NiO and WO3 in their respective
Micrographs with a magnification of ~50 nm were taken for Al(x)- sulfides) was conducted ex-situ in the same batch reactor. But, using a
SBA-15, and Ni–W/Al(0.1)-SBA-15 materials using a 200 kV JEOL feedstock formed by LGO (~1.4 wt% S content) mixed with the required
(Japan) JEM-ARM200F microscope with a cold field emission gun amount of dimethyl disulfide (DMDS, C2H6S2, Aldrich 99 %) to attain
(ColdFEG). Images acquisitions were carried out with a Orius model ~2.3 wt% of S content. Reactor operating conditions are the following:
Gatan camera employing Gatan Microscopy Suite: Digital micrograph 25 mL of LGO, 3 MPa of H2(g) initial pressure, 750 rpm, 2.5 g of oxidized
2.31.734 version. These analyses were performed to elucidate changes Ni–W precursor of the catalyst. Temperature was successively increased
in porous structure by Al, Ni and W incorporation during synthesis from ~25 ◦ C to 150 ◦ C, and kept for 30 min. Subsequently, the tem­
procedure. perature was raised from 150 ◦ C to 350 ◦ C kept for 3 h at a heating rate
of ~2.5 ◦ C/min. The sulfided catalyst were recovered by filtration.
2.3.4. Materials acid sites distribution (Brønsted/Lewis) by Pyridine-FTIR
About 0.03 g of each Al(x)-SBA-15 and Ni–W/Al(x)-SBA-15 materials 2.5. Catalytic activity
were analyzed in an FTIR spectrometer Nicolet 8700 within a range of
400 to 4000 cm− 1, and an average number of scans of 50. Prior to the 2.5.1. Hydrodesulfurization (HDS) percentage (sulfur removal)
analysis, samples were outgassed at 400 ◦ C. About 50 µL of Pyridine was Sulfur content was determined according to the ASTM D4294 stan­
absorbed at room temperature inside a quartz cell, and desorbed pro­ dard in a SINDIE 2622-10 equipment. Then, sulfur removal percentage
gressively under vacuum at 50, 100, 200, 300, and 400 ◦ C. was determined by computing the relative deviation percentage be­
tween the feedstock (as reference) and the liquid product for each Ni–W/
2.3.5. Molecular structure of aluminum species by 27Al-MAS-NMR Al(x)-SBA-15 catalyst.
measurements
27
Al Nuclear magnetic resonance (NMR) spectra were recorded in a 2.5.2. Hydrodeoxygenation (HDO) percentage (oxygen removal)
Bruker Advance III spectrometer attached to a 4 mm magic-angle spin­ To quantify the oxygen removal by hydro-co-processing the feed­
ning (MAS) probe, using direct excitation, about 30◦ of pulse angle, a stock with the Ni–W/Al(x)-SBA-15 catalysts a determination curve,
recycle delay of 0.1 s, and a sample spin-rate of about 7 kHz. Spectra vegetable oil vs 1750 cm− 1 FTIR absorbance, (R2 = 0.986), was carried
were acquired at 104.26 MHz. out by preparing samples with 0, 5.7, 10.8, 19.7, and 25.1 wt% of
Jatropha Curcas L. oil and the GOM. Changes in intensity for the bands
2.3.6. Chemical functional groups by RAMAN spectroscopy about 1750 cm− 1 and 1710 cm− 1 were used to determine the HDO
RAMAN spectra were acquired in the interval of 100 to 1300 cm− 1 in performance by computing the respective vegetable oil percentage.
a LabRam HR8000 equipment employing a laser of 633 nm, nominal Then, the relative deviation percentage between the liquid product for
power of 13 mW, exposure time of 6 s, and an accumulation time of 8 each catalyst and the initial vegetable oil content in the feedstock as
repetitions. references was computed.

3
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

2.5.3. Hydrocracking (HCK) and fractions compositions catalysts surface, and the lowest Al/Si molar ratio catalyst depicted the
Simulated Distillation was carried out according to the ASTM D2887 more reduced hysteresis loop. Therefore, the lowest adsorption capacity
procedure in a GC-2010 Shimadzu equipment, which was set up for such as previously commented.
purposes. Distillation curve data was analyzed in ASPEN HYSYS V9 to
determine the composition of each fraction as follows: Naphtha, Kero­ 3.1.2. Mesoporous structure of Al(x)-SBA-15 catalytic supports and Ni–W/
sene, Light Diesel, Heavy Diesel, Gas Oil, and Residue. Diesel-like frac­ Al(x)-SBA-15 catalysts crystallinity
tion composition was computed by adding Light Diesel and Heavy Diesel Fig. 2a) depicts low angle X-ray diffraction (LA-XRD) profiles for Al
compositions, respectively. (x)-SBA-15, and SBA-15 reference. All materials exhibited three clear
peaks at 2θ: 0.8◦ 1.4◦ and 1.7◦ ascribed to indexed planes (1 0 0), (1 1 0),
3. Results and discussion and (2 0 0) respectively. The occurrence of such peaks indicates that all
materials possess a hexagonal symmetry with p6mm space group of SBA-
3.1. Characterization of catalytic supports and catalysts 15 [21]. The direct incorporation of Al during the synthesis of Al(x)-
SBA-15 materials did not alter the ordering pattern of SBA-15. There­
3.1.1. Textural properties of catalytic supports and catalysts fore, Al might be incorporated as extra-framework Al or at lower Al/Si
According to data reported in Table 1, all Al(x)-SBA-15 supports molar ratio it might be completely incorporated into silica framework.
exhibited a pore volume (PV) of 1.2 cm3g− 1 and a pore diameter (PD) of Regarding wide angle X-ray diffraction (WA-XRD) in Fig. 2. b), all
~6.6 nm with a variation among samples of ~6.0 % and ~2.6 % catalysts samples depicted peaks at 2θ: 23.6◦ , 33.6◦ , 41.6◦ , 54.7◦ , and
respectively. Regarding the specific surface area, SBET, the average de­ 60.2◦ . Such peaks could be ascribed to NiWO4 (PDF 00-0072-0480
viation in such property is ~5.6 %. No linear tendence could be observed ICDD) [22], indicating the formation of a mixed compound of W and
for textural properties variations with Al/Si molar ratio. Thus, the dif­ Ni components during catalysts preparation. On the other hand, WO3
ferences among such properties could be possibly attributed to inherent species signals are ascribed at 2θ: 23.1◦ , 23.6◦ , and 24.3◦ . Hence, the
errors during measurements, indicating all catalytic supports shown acute signal at ~23.6◦ could be associated to WO3. No occurrence of
similar properties disregarding the Al content. In addition, Table 1 also signals for NiO at 2θ: 37.4◦ , 43.4◦ , and 62.7◦ were observed [15,23,24].
reports the textural properties for Ni–W oxide phase catalysts (2.5 wt% Hence, the latter ratifies the mixture among Ni and W precursors pre­
NiO, and 15 wt% WO3) supported on Al(x)-SBA-15 materials. Average viously commented. Regarding to Al species, no signal could be
deviations in PV and PD were ~3.6 % and 2.1 % respectively, while SBET observed, suggesting that there were not Ni, W, and Al mixed phases or
average deviation was ~1.5%. However, it could be observed that SBET either those were below the detection limit of XRD. Additionally, WA-
decreased at expenses of the decrease of Al/Si molar ratio, indicating XRD profile for Ni–W/Al(x)-SBA-15 and Ni–W/Al(0.0)-SBA-15 (no Al
that Al influenced changes in the mesoporosity when adding metals (Ni content) were the same. Hence, the formation of any phase related to Al
and W) on the Al(x)-SBA-15 catalytic supports. Such changes were was not clearly evidenced. Finally, NiAl2O4 signals ascribed at 2θ: 19.3◦ ,
ratified by comparing textural properties among Al(x)-SBA-15 materials 31.5◦ , 37.2◦ , 45.2◦ , 59.9◦ , and 65.7◦ were not observed [25].
with Ni–W/Al(x)-SBA-15 oxide phase catalysts. It was evidenced the According to the information given in Table 2, all catalysts exhibited
reduction of both SBET and PV when adding metals on the catalysts an average particle size ~4.4 nm. Particle size was computed by Debye-
surface or adsorption capacity, and PD remains the same. Accordingly, Scherrer equation for the strongest signal at 2θ ~ 23.6◦ for WO3
Ni and W species did not block pores during the synthesis of the oxide diffraction signal, and b is the Full Width at Half Maximum (FWHM) of
phase catalysts. such peak. It is noteworthy that the minimum particle size that can be
Fig. 1a) shows nitrogen physisorption isotherms for all Al(x)-SBA-15 observed by XRD is ~4 nm. Therefore, catalysts average particle size
catalytic supports and SBA-15 reference. All materials exhibited a type was 0.3 nm larger that such minimum.
IV isotherm with an H1 hysteresis loop. Type IV isotherm type indicates
a mesoporous material, and the kind of hysteresis loop indicates that all 3.1.3. Porous structure of catalytic supports and catalysts
materials possess tubular and uniform pores possibly opened both sides. As far as can be observed in Fig. 3, SBA-15 reference, and Al(x)-SBA-
As a result, it can be inferred that Al incorporation did not affect the 15 samples clearly exhibited the hexagonal 2D-array of cylindrical pores
structure of pores due to the no significant changes in the shape of such as LA-XRD diffractograms indicated (Section 3.1.2). For all materials
loop at relative pressures (P/P0) between 0.6 and 0.8 were observed. tubular uniform channels were observed. Hence, it was confirmed that
For Ni–W/Al(x)-SBA-15 catalysts, Fig. 1b) shown that all catalysts direct Al incorporation did not influence significant changes in the
depicted Type IV isotherm and an H1 hysteresis loop as the pristine Al mesoporous pattern of SBA-15 by the proposed method of synthesis.
(x)-SBA-15 catalytic supports. Nevertheless, in comparison with iso­ Accordingly, those results suggests that Al species might be uniformly
therms presented in Fig. 1a), these catalysts shown a reduced adsorbed distributed on silica surface for Al(x)-SBA-15 catalytic supports. In
volume due to the reduction and partial deformation of the hysteresis addition, in Fig. 3f) it can be observed that Ni and W nanoparticles were
loop at relative pressures (P/P0) between 0.6 and 0.8. Such deformation grouped or agglomerated over Al(0.1)-SBA-15 as it was shown in Table 2
indicates a change in pores shape by Ni and W incorporation on the for the size of WO3 nanoparticles. The latter indicates that the formation
of NiWO4 was not homogeneous during the synthesis process.

Table 1 3.1.4. Materials acid sites distribution (Brønsted/Lewis) by Py-FTIR

Textural Properties for Al(x)-SBA-15 supports and Ni–W/Al(x)-SBA-15 catalysts; Acid sites (Brønsted and Lewis) distribution can be qualitatively
x: Al/Si molar ratio. analyzed by Py-FTIR spectra. In such spectra, three main bands could be
Sample SBET [m2/g] PV [cm3/g] PD [nm] observed. First, about 1444 cm− 1 ascribed to Lewis (L) acid sites, second,
Al(0.1)-SBA-15 766.3 1.3 6.7 about 1489 cm− 1 ascribed to Brønsted and Lewis (B + L) acid sites, and
Al(0.05)-SBA-15 750.0 1.3 6.8 about 1539 cm− 1 ascribed to Brønsted (B) acid sites. The band about
Al(0.033)-SBA-15 748.0 1.2 6.6 1595 cm− 1 corresponds to the vibrational modes of pyridine adsorbed
Al(0.025)-SBA-15 803.0 1.3 6.5 on the materials surface. Accordingly, in Fig. 4a), for catalytic supports,
Al(0.0)-SBA-15 728.4 1.2 6.8
Ni–W/Al(0.1)-SBA-15 524.3 0.9 7.0
there were not observed intense signals for Brønsted acid sites. Instead, it
Ni–W/Al(0.05)-SBA-15 514.2 0.9 6.9 was observed that the predominant signal was for Lewis acid sites. In
Ni–W/Al(0.033)-SBA-15 513.6 0.9 6.7 general, Lewis and Brønsted acid sites bands increased with Al content in
Ni–W/Al(0.025)-SBA-15 501.6 0.8 6.6 comparison with Al(0.0)-SBA-15 reference. On the other hand, Fig. 4b)
Ni–W/Al(0.0)-SBA-15 516.1 0.9 6.8
for Ni–W/Al(x)-SBA-15 oxide phase catalysts, it was observed that there

4
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

Fig. 1. [email protected] ◦ C adsorption–desorption isotherms for a) Al(x)-SBA-15 materials b) Ni–W/Al(x)-SBA-15 oxide phase catalysts; x: Al/Si molar ratio.

Fig. 2. XRD patterns for a) Al(x)-SBA-15 materials and SBA-15 reference (LA, low angle), b) Ni–W/Al(x)-SBA-15 oxide phase catalysts (WA, wide angle); x: Al/Si
molar ratio.

et al. reported a similar result, 0.057 mmol g−, 1 for an Pt/Al(0.14)-SBA-

Table 2
15 catalyst at 150 ◦ C [12]. Similarly, as Li et al. reported for Al(0.2)-SBA-
Crystallite size [nm] for WO3 in Ni–W/Al(x)-SBA-15 catalysts, x: Al/Si molar
15, and Ni–W/Al(0.2)-SBA-15 a total Brønsted acid sites concentration
ratio.
of 0.043 mmol g− 1, and 0.023 mmol g− 1 respectively at 200 ◦ C [24].
Catalyst θm (2θ/2) [◦ ] b [◦ ] b[rad] d[nm] On the other hand, Table 4 depicts the amount of pyridine adsorbed
Ni–W/Al(0.0)-SBA-15 11.8454 1.8371 0.0321 4.5 (mmol g− 1) on Lewis acid sites for Ni–W/Al(x)-SBA-15 oxide phase
Ni–W/Al(0.1)-SBA-15 11.7489 1.8904 0.0330 4.3 catalysts. At temperatures higher than 50 ◦ C the amount of pyridine
Ni–W/Al(0.05)-SBA-15 11.7020 1.8992 0.0331 4.3
adsorbed on Lewis acid sites (mmol g− 1) decreased dramatically (about
Ni–W/Al(0.033)-SBA-15 11.6999 1.9031 0.0332 4.3
Ni–W/Al(0.025)-SBA-15 11.7155 1.8666 0.0326 4.4 97.9 %) for all catalysts. For instance, for Ni–W/Al(0.05)-SBA-15, Lewis
sites concentration decreased from 1.450 mmol g− 1 (50 ◦ C) to 0.030
mmol g− 1 (400 ◦ C). In comparison, Jaroszewska et al., reported a similar
was an increase in intensity of both bands (Brønsted and Lewis), indi­ result, 0.089 mmol g− 1 for an Pt/Al(0.14)-SBA-15 catalyst at 150 ◦ C
cating that Ni and W incorporation enhanced the formation of acid sites [12]. Similarly, Li et al., reported for Al(0.2)-SBA-15, and Ni–W/Al(0.2)-
on the catalysts surface. SBA-15 [24] a total Lewis acid sites concentration of 0.209 mmol g− 1,
For Ni–W/Al(x)-SBA-15 oxide phase catalysts, presented in this and 0.332 mmol g− 1, respectively at 200 ◦ C. Hence the Ni–W/Al(x)-SBA-
work, the amount of pyridine adsorbed (mmol g− 1) on Brønsted acid 15 catalysts presented in this work showed a higher concentration of
sites is depicted in Table 3. At temperatures higher than 50 ◦ C the Lewis acid sites than Brønsted acid sites and compared to previously
amount of pyridine adsorbed on Brønsted acid sites (mmol g− 1) reported measurements for similar materials.
decreased dramatically (about 98 %) for all catalysts. For example, for Finally, the dramatical decrease in concentration of both kind of acid
Ni–W/Al(0.05)-SBA-15, Brønsted sites concentration decreased from sites (Brønsted and Lewis) at temperatures higher than 100 ◦ C indicated
0.050 mmol g− 1 (50 ◦ C) to 0.001 mmol g− 1 (400 ◦ C). However, Ni–W/Al that the oxide phase catalysts presented in this study possess weak
(0.05)-SBA-15 catalyst exhibited the highest concentration of Brønsted acidity, and for that reason pyridine is readily desorbed.
acid sites of 0.050 mmol g− 1 at 50 ◦ C. Regarding the latter, Jaroszewska

5
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

Fig. 3. HR-TEM micrographs for Al(x)-SBA-15 supports and catalyst, a) Al(0.0)-SBA-15, b) Al(0.1)-SBA-15, c)Al(0.05)-SBA-15, d) Al(0.033)-SBA-15, e) Al(0.025)-
SBA-15, f) Ni–W/Al(0.1)-SBA-15. x: Al/Si molar ratio.

Fig. 4. Py-FTIR spectra at 50 ◦ C a) Al(x)-SBA-15 catalytic supports, b) Ni–W/Al(x)-SBA-15 oxide phase catalysts. x: Al/Si molar ratio.

Table 3 Table 4
Brønsted acid sites concentration for Ni–W/Al(x)-SBA-15 catalysts, x: Al/Si Lewis’ acid sites concentration for Ni–W/Al(x)-SBA-15 catalysts, x: Al/Si molar
molar ratio. ratio.
Material\ T[◦ C] Brønsted sites [mmol g− 1] Material\ T[◦ C] Lewis sites [mmol g-1]

50 100 200 300 400 50 100 200 300 400

Ni–W/Al(0.1)-SBA-15 0.040 0.007 0.005 0.006 0.004 Ni–W/Al(0.1)-SBA-15 0.790 0.190 0.060 0.040 0.020
Ni–W/Al(0.05)-SBA-15 0.050 0.021 0.010 0.002 0.001 Ni–W/Al(0.05)-SBA-15 1.450 0.180 0.050 0.030 0.030
Ni–W/Al(0.033)-SBA-15 0.028 0.013 0.002 0.000 0.000 Ni–W/Al(0.033)-SBA-15 0.600 0.140 0.050 0.020 0.010
Ni–W/Al(0.025)-SBA-15 0.023 0.015 0.007 0.001 0.000 Ni–W/Al(0.025)-SBA-15 0.660 0.170 0.040 0.010 0.000

3.1.5. Molecular structure of aluminum species by 27Al-MAS-NMR framework of SBA-15. The latter, due to the occurrence and intensity of
measurements the signal at about 53 ppm ascribed to tetrahedral aluminium [26–28]. It
Nuclear magnetic resonance (NMR) in the solid state provides in­ is noteworthy that, the occurrence of octahedral aluminum species
formation regarding chemical structure of specific elements in a mate­ (extra framework Al) at a chemical shift about 0 ppm is evidenced for
rial. In this case, 27Al-MAS-NMR spectra for Al(x)-SBA-15 and Ni–W/Al the highest Al/Si molar ratios (0.1 and 0.05). For Ni–W/Al(x)-SBA-15 in
(x)-SBA-15 are shown in Fig. 5 for Al species. Specifically, in Fig. 5a) it Fig. 5b), the occurrence of both kind of Al species was also evident. In
could be observed that most of the Al added during the synthesis of Al comparison with Al(x)-SBA-15 results in Fig. 5a), the intensity of octa­
(x)-SBA-15 catalytic supports was directly incorporated into the Si hedral Al was increased by the addition of Ni and W, suggesting that Ni

6
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

27
Fig. 5. Al-MAS-NMR spectra a) Al(x)-SBA-15 catalytic supports, b) Ni–W/Al(x)-SBA-15 oxide phase catalysts. x: Al/Si molar ratio.

and W interacted with Al. However, for the catalyst Ni–W/Al(0.025)- compound for NiWO4 [29]. Regarding the latter, catalysts with Al/Si
SBA-15 the corresponding signals to tetra and octahedral aluminum molar ratios of 0.05 and 0.033 exhibited a sharper signal at ~892 cm− 1,
species were not clearly observed. corresponding to NiWO4, specifically to vibrational modes for distorted
and octahedral WO6 clusters as reported by Lima et al. [30]. Finally, for
3.1.6. Chemical functional groups by RAMAN spectroscopy Ni–W/Al(0.025)-SBA-15 it was observed a weak signal at 606 cm− 1
According to the spectra presented in Fig. 6, the occurrence of ascribed to NiAl2O4 [31]. Nevertheless, the presence of Ni and Al
different functional groups on the surface for Ni–W/Al(x)-SBA-15 cata­ compounds was not detected by WA-XRD measurements as reported in
lysts varies in function of Al/Si molar ratio. In common, there were Section 3.1.2. On the other hand, the formation of Aluminum Tungstate
bands at 715 cm− 1 and 805 cm− 1, which are ascribed to WO3 species in was discarded since there was no signal about 1054 cm− 1 for any of the
agreement with the WA-XRD profile shown in Fig. 2b) and data reported catalysts [29].
in Table 2. An extra common band was observed at about 975 cm− 1 for
all Ni–W/Al(x)-SBA-15 oxide phase catalysts. For Ni–W/Al(0.025)-SBA- 3.1.7. FTIR spectra of liquid products during the hydro-co-processing of the
15 oxide phase catalyst such signal was the highest. That signal could be vegetable oil and gas oil
associated with the formation to an octahedrally coordinated tungsten According to Fig. 7, major changes in FTIR spectra were qualitatively

Fig. 6. RAMAN spectra for Ni–W/Al(x)-SBA-15 oxide phase catalysts. x: Al/Si Fig. 7. FTIR spectra for liquid products using Ni–W/Al(x)-SBA-15 oxide phase
molar ratio. catalysts. x: Al/Si molar ratio.

7
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

observed by comparing the feedstock spectrum with the liquid products oil into the feedstock. Therefore, depending on the Al content (Al/Si
spectra for Ni–W/Al(x)-SBA-15 catalysts and the Ni–W/Al(0.0)-SBA-15 molar ratio) hydroprocessing of vegetable oil occurs via formation of
catalyst reference. The band at about 1160 cm− 1, ascribed to C-O bond free fatty acids, such as oleic acid, which were later hydroconverted in
disappeared completely, indicating that HDO of vegetable oil occurs via lighter hydrocarbons (C14 to C29). The less Al on the catalysts the
separation of linear hydrocarbons from carboxylic acid chains during higher concentration of lighter fractions within m/z range between 200
HDO from carbon and oxygen simple bond. On the other hand, the band and 400. The more Al on the catalysts the highest content of free fatty
about 1750 cm− 1, ascribed to carbonyl group (acyl groups) moved acids in the product mixture. On the other hand, FTIR results (Section
leftwards to about 1710 cm− 1. The latter indicates that there was a 3.1.7) indicated that the highest reduction of carbonyl group at ~1710
change in product distribution and molecules containing the carbonyl cm− 1 was observed at Al/Si molar ratio of 0.1. Therefore, hydro­
group due to hydroprocessing from triglyceride to carboxylic acid [32]. conversion of acyl groups in triglycerides into fatty acids could be
Such change was the highest for Ni–W/Al(0.1)-SBA-15, which showed observed. In addition, Ni–W/Al(0.1)-SBA.15 sulfided catalyst evidenced
the lowest intensity for the ~1710 cm− 1 FTIR band. However, it is a lower occurrence of lighter hydrocarbons fractions.
noteworthy that by FTIR analysis it cannot be determined if the reduc­
tion of 1710 cm− 1 band was derived from the decomposition of only one 3.1.9. H NMR spectra for liquid products and vegetable oil
carboxylic acid or a group of compounds with similar carbon and oxygen Fig. 9 depicts H-NMR spectra for the liquid products during hydro-
double bond groups. The latter due to that Jatropha Curcas L. oil pos­ co-processing the mixture of vegetable oil and gas oil. There, it is also
sesses three kinds of carboxylic acids: oleic, linoleic, and linolenic acids, presented the spectra for Jatropha Curcas L. oil as a reference for
which exhibits the same characteristic band. comparative purposes. According to the figure, about 9 signals could be
clearly observed for pure vegetable oil at 0.89 ppm for fatty acids
3.1.8. Qualitative hydrocarbon distribution by ESI mass spectrometry excluding linolenic acid, 1.31 ppm and 1.61 ppm for acyl chains, 2.04
In Fig. 8 are depicted the ESI-MS spectra for all the liquid products ppm for mono- and polyunsaturated fatty acids, 2.31 ppm for acyl chains
during hydroprocessing the mixture of vegetable oil and gas oil. Ac­ in unsaturated fatty acids, 2.77 ppm for linoleic and linolenic acid, 4.27
cording to the results, all samples exhibited the occurrence of most of the ppm and 5.28 ppm for triacylglycerols, and at 5.35 ppm for mono- and
fractions within the m/z range between 200 and 400, such range cor­ polyunsaturated fatty acid [33,34]. The latter signals could be observed
responds to C14 to C29 hydrocarbons. Nevertheless, the mixture for the feedstock at lower intensity due to the concentration of vegetable
(feedstock) exhibited the most intense signal at m/z ~ 619.16 corre­ oil in such mixture (~20 vol%). However, by comparing with the
sponding to the C44 alkane chain, followed by the signal at ~591.108 spectra of liquid products for Ni–W/Al(x)-SBA-15 and Ni–W/Al(0.0)-
for C42, and the occurrence of signals between C59 and C63 at 829.550 SBA-15 sulfided catalysts, it could be observed only two signals at 1.31
and 885.654 respectively. Hence, the feedstock possessed a major con­ ppm and 0.89 ppm. As a result, hydro-co-processing the mixture
centration of heavy fractions. On the other hand, when using Ni–W/Al occurred via degradation of fatty acids via HDO of the vegetable oil
(x)-SBA-15 sulfided catalysts, the product distribution evidenced a component. The occurrence of the remaining signals indicated that there
reduction in heavy components at expenses of decreasing the Al/Si were fractions of acyl group components unconverted from carboxylic
molar ratio. Specifically, for Ni–W/Al(0.1)-SBA-15, it was observed the acid. The latter in agreement with FTIR spectra in Section 3.1.7 for the
strongest signal at 282.25 corresponding to C18-C20 hydrocarbons, analysis of the band at about 1710 cm− 1, which remained disregarding
followed by the signals at 304.23 and 390.28 ascribed to C21 and C28 Al content. Signals at higher chemical shift of 1.31 ppm were not
respectively. Even so, the signal ascribed to C18-C20 might be possible detectable when using Ni–W/Al(x)-SBA-15 sulfided catalysts. Therefore,
attributed to some type of free fatty acids presented in Jatropha Curcas L. mono- and polyunsaturated fatty acid in the vegetable oil were

Fig. 8. ESI-MS spectra for the liquid product of Ni–W/Al(x)-SBA-15 sulfided phase catalysts and the feedstock. x: Al/Si molar ratio.

8
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

3.2.2. Effect of the Al content in the hydrodeoxygenation (HDO) properties

of Ni–W/Al(x)-SBA-15 sulfided catalysts
Fig. 11 depicts changes in reduction (%HDO) of the FTIR absorption
band ascribed to carbonyl group for carboxylic acid (about 1710 cm− 1)
as function of Al/Si molar ratio. %HDO decreased at expenses of Al/Si
molar ratio, indicating that a major incorporation of Al promotes the
carboxylic acid formation from triglycerides during the hydro-co-
processing the mixture of vegetable oil and gas oil. The highest HDO
yield was for Ni–W/Al(0.1)-SBA-15 (57.3 %), and the lowest for Ni–W/
Al(0.025)-SBA-15 (30.9 %). The absence of Al in the catalyst, Ni–W/Al
(0.0)-SBA-15, showed a 30.8 % of HDO as observed for Ni–W/Al(0.025)-
SBA-15. However, as previously mentioned. the best yield was attained
at the highest Al/Si molar ratio. Hence. Al effect in HDO was clearly
demonstrated. The previous results agreed to the qualitative FTIR
analysis shown in Section 3.1.7.
By comparing HDO and HDS yields, the HDO prevailed over the HDS
at the highest Al/Si molar ratio as shown in the previous section (Sec­
tion 3.2.2) where Ni–W/Al(0.1)-SBA 0.15 sulfided catalyst did not show
the highest HDS activity (Section 3.2.1). In this respect, Sépulveda et al.
[35] demonstrated that during the co-processing of guaiacol and 4,6-
DMDBT with sulfided ReS2/SiO2 and ReS2/Al2O3 catalysts, guaiacol
Fig. 9. H-NMR spectra for the liquid product of Ni–W/Al(x)-SBA-15 sulfided
inhibited HDS rate and catalysts acidity influenced HDO. They also
phase catalysts and the feedstock. x: Al/Si molar ratio.
highlighted that water formation as byproduct during the HDO also
might decrease HDS activity. Water formation could promote the
completely hydroconverted. occurrence of oxy-sulfides and oxides over the catalysts surface. The
latter was stated by Infantes-Molina et al. [36], demonstrating that
3.2. Catalytic evaluation of Ni–W/Al(x)-SBA-15 sulfided catalysts during HDO of phenol the oxidation/hydroxylation of ruthenium sulfide
(RuS2) phase occurred forming a ruthenium oxy-sulfided and ruthenium
3.2.1. Effect of the Al content in the hydrodesulfurization properties of oxide. The latter according to XPS measurements for the spent catalysts
Ni–W/Al(x)-SBA-15 sulfided catalysts supported on SBA-15 and modified with Ir or Pd, RuS2/SBA-15. On the
Fig. 10 depicts changes in sulfur removal expressed as %HDS as other hand, Şenol et al. [37] studied H2S effect during the HDO of ar­
function of Al/Si molar ratio. According to the results, there was not a omatic and aliphatic oxygenates. HDO conversion for phenol decreasing
linear tendency between the %HDS and the Al content. It could only be at expenses of increasing H2S content. On the contrary, increasing
inferred that the sulfided catalyst Ni–W/Al(0.05)-SBA-15 demonstrated hydrogen sulfide content during HDO of methyl heptanoate enhanced
the highest sulfur removal yield (13.1 %). The catalysts Ni–W/Al HDO conversion. As a result, one can infer that hydrogen sulfide for­
(0.025)-SBA-15 and Ni–W/Al(0.0)-SBA-15 evidenced similar results, mation during HDS improves HDO reactions when hydro-co-processing
5.6, and 6.0 % respectively. A plausible explanation for such trend could sulfur and oxygenated compounds.
be related with that Al was undetectable by 27Al-MAS-NMR for the For our Ni–W/Al(0.1)-SBA-15 sulfided catalysts, a higher HDO could
catalyst with Al/Si molar ratio of 0.025 as shown in Fig. 5b). The less Al lead to a higher production of water, reducing the occurrence of sulfided
on the catalysts the less HDS activity. Hence, Al influences sulfur active sites via formation of Ni and W oxy-sulfides and oxides. In this
removal reactions via HDS at the reaction conditions evaluated. Addi­ respect, a gas phase H2/H2S blend is required to reactivate the catalysts,
tionally, it could also be inferred that catalysts acidity influenced sulfur restoring the fully sulfided coordinated metal sulfide. Hence, hydrogen
removal. The latter due to that the Ni–W/Al(0.05)-SBA-15 catalyst dis­ sulfide formation during HDS enhanced HDO by regenerating the active
played the highest occurrence of Brønsted and Lewis acid sites and the sites used during HDO. Additionally, there might be a competition
highest HDS activity as shown in Table 3, and Fig. 10, respectively. among oxygenated and sulfur compounds on reacting over the surface of
the catalysts as the above mentioned authors have reported before with

Fig. 10. HDS expressed as sulfur removal percentage for Ni–W/Al(x)-SBA-15 Fig. 11. HDO expressed reduction of absorbance at FTIR band of about 1710
oxide phase catalysts. x: Al/Si molar ratio. cm− 1 for Ni–W/Al(x)-SBA-15 oxide phase catalysts. X: Al/Si molar ratio.

9
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

experiments with sulfur and oxygenated model compounds [36,37]. conditions employed as previously commented in sections 3.2.1 and
Therefore, the catalytic activity observed on simultaneous HDO-HDS 3.2.2. In addition, it has been also reported that for NiWO4 sulfided
with Ni–W/Al(x)-SBA-15 sulfided catalysts may be governed by the catalysts there is an enhancement in HDS activity due to the formation of
regeneration step in the catalytic cycle, specifically, after oxygen a single sulfided phase over the catalysts surface as detected by tem­
removal. perature programmed sulfidation [22]. Hence, Al incorporation influ­
According to the latter, it is not always possible to attain a catalyst enced the formation of a single NiWO4 sulfided phase due to that
with the best yielding for HDO and HDS simultaneously when hydro-co- RAMAN spectra for Ni–W/Al(x)-SBA-15 catalysts showed decreased
processing a mixture of vegetable oil and gas oil. Intrinsic properties of intensity for the signal at ~970 cm− 1 for higher Al/Si molar ratios, thus
the catalyst like acidity and metallic character influences the catalytic promoting hydrotreating reactions. On the other hand, the occurrence of
activity. As a result, tailoring such properties will depend on the final an acute signal at 2θ of 23.6◦ for WO3(Fig. 2 for WA-XRD) suggests the
purpose of the catalyst. In this research our purpose was obtaining a formation of the type II active phase (multi-layer) ascribed to the cor­
catalyst yielding a diesel-like fraction with the less sulfur content as responding sulfide, which can be ratified by the acute signal at ~805
possible. cm− 1 in the RAMAN spectra at higher Al/Si molar ratios. Acute RAMAN
signals suggest less dispersed active phases while wider signals more
3.2.3. Effect of the Al content in the hydrocracking (HCK) properties of dispersed active phases. However, type II active phase enhanced HDS
Ni–W/Al(x)-SBA-15 sulfided catalysts and hydrogenation (HG) activity of dibenzothiophene (DBT) in com­
ASTM D2887 simulated distillation curves were processed in ASPEN parison to type I active phase (single layer) using Ni-Mo and Co-Mo
HYSYS 9 to compute the corresponding mass fractions for five fractions: sulfided catalysts supported on γ-Al2O3 as reported by Kagami et al.
Naphtha, Kerosene, Diesel, Gasoil and Residue. According to the results [38]. For that reason, the catalytic activity is not necessarily governed
shown in Fig. 12, the highest content of Al evidenced the lowest diesel- by active phase dispersion on the catalyst surface, rather by the type of
like fraction percentage (39.5 %) for Ni–W/Al(0.1)-SBA-15 catalyst, and active phase site. In our case, as shown in Fig. 5 for 27Al-MAS-NMR
the catalyst with Al/Si molar ratio of 0.0 showed the highest yield (47.2 spectra, Al interacted with metals increasing octahedral Al at expenses
%). However, Ni–W/Al(0.05)-SBA-15 showed a 44.8 % diesel-like of decreasing tetrahedral Al. Consequently, the interaction among
fraction percentage, which is 2.4 % units lesser than Ni–W/Al(0.0)- metals and the Al2O3formed during the synthesis of Al(x)-SBA-15 sup­
SBA-15. Same tendency was observed for the other Ni–W/Al(x)-SBA-15 ports gave birth to the already mentioned type II and type I active phases
sulfided catalysts. Accordingly, it could not be concluded that Al content sites on the sulfided catalysts.
did not influence HCK reactions, instead it was clear that Al/Si molar Finally, as evidenced in Fig. 12, WO3 and NiWO4 oxide precursors for
ratio of 0.05 promoted the highest yield to a diesel-like fraction when a sulfided active phase improved hydrocracking reactions as demon­
hydro-co-processing the mixture of vegetable oil and gas oil. Similar strated by the formation of higher contents of diesel, naphtha, and
trends could be observed for the other fractions. Specifically, Residue kerosene fractions with the no Al sulfided catalyst, Ni–W/Al(0.0)-SBA-
mass fraction was reduced from 33 % to an average 20.3 % by hydro-co- 15.
processing the feedstock with Ni–W/Al(x)-SBA-15 sulfided catalysts
with Al/Si molar ratio higher than 0.1. One way to explain the latter 4. Conclusions
results relies on the fact that at higher Al content the catalysts exhibited
the occurrence of grouped metallic phases with ~4.4 nm of WO3 crys­ Hydro-co-processing a mixture of Jatropha Curcas L. oil and gas oil
tallite size over the catalyst surface as shown in TEM micrograph in blend was evaluated employing a series of unconventional Ni–W/Al(x)-
Fig. 3f) and Table 2. Respectively. Catalytic activity depends strongly on SBA-15 sulfided catalysts. Al effect in hydroprocessing relied on pro­
the way the active phases distribute on the catalysts surface, and it was moting heteroatoms (S and O) removal instead of hydrocracking re­
shown in Fig. 6 for RAMAN spectra, Al incorporation influenced actions as was demonstrated by testing Ni–W/Al(0.0)-SBA-15 sulfided
chemical structure of the catalysts. As a result, an improvement in catalyst (no Al). In addition, that sulfided catalyst enhanced the highest
hydrotreating reactions (HDS and HDO) when Al was added to the diesel-like fraction yield of about 47.2 % compared to 44.8 % for Ni–W/
synthesis of the catalytic support was observed. Accordingly, Al addition Al(0.05)-SBA-15 sulfided catalyst as a direct effect of NiWO4 as active
favored heteroatoms removal instead of hydrocracking at the reaction phase. However, Ni–W/Al(x)-SBA-15 sulfided catalysts could remove

Fig. 12. Lumps mass fraction percentage distribution for Ni–W/Al(x)-SBA-15 oxide phase catalysts. x: Al/Si molar ratio.

10
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

about 1267 ppm of the feedstock and about 57.3 % of acyl groups in [10] Restrepo-Garcia JR, Baldovino-Medrano VG, Giraldo SA. Improving the selectivity
in hydrocracking of phenanthrene over mesoporous Al-SBA-15 based Fe–W
vegetable oil. Therefore, these catalysts are promising candidates for
catalysts by enhancing mesoporosity and acidity. Appl Catal A 2016;510:98–109.
getting cleaner (less S), greener and lighter hydrocarbons blend via [11] Ramírez J, Gutiérrez-Alejandre A, Sánchez-Minero F, Macías-Alcántara V, Castillo-
hydroprocessing of vegetable oil and gas oil blends. Finally, Al direct Villalón P, Oliviero L, et al. HDS of 4,6-DMDBT over NiMoP/(x)Ti-SBA-15 catalysts
incorporation during the synthesis of SBA-15 did not collapse the mes­ prepared with H3PMo12O40. Energy Fuel 2012;26(2):773–82.
[12] Jaroszewska K, Fedyna M, Trawczyński J. Hydroisomerization of long-chain n-
ostructured ordering pattern of SBA-15 maintaining textural properties, alkanes over Pt/AlSBA-15+zeolite bimodal catalysts. Appl Catal B 2019;255:
generating acidity, and promoting hydroprocessing reactions on the 117756.
catalysts surface. [13] Kamaruzaman MF, Taufiq-Yap YH, Derawi D. Green diesel production from palm
fatty acid distillate over SBA-15-supported nickel, cobalt, and nickel/cobalt
catalysts. Biomass Bioenergy 2020;134:105476.
CRediT authorship contribution statement [14] Oh S, Lee JH, Choi I-G, Choi JW. Enhancement of bio-oil hydrodeoxygenation
activity over Ni-based bimetallic catalysts supported on SBA-15. Renew Energy
2020;149:1–10.
Jonatan R. Restrepo-Garcia: Conceptualization, Methodology, [15] Baharudin KB, Taufiq-Yap YH, Hunns J, Isaacs M, Wilson K, Derawi D. Mesoporous
Formal analysis, Writing – original draft, Writing – review & editing. NiO/Al-SBA-15 catalysts for solvent-free deoxygenation of palm fatty acid
distillate. Microporous Mesoporous Mater 2019;276:13–22.
Pablo Torres-Mancera: Conceptualization, Methodology, Formal
[16] Zhang Ze, Cheng J, Zhu Y, Guo H, Yang W. Jet fuel range hydrocarbons production
analysis, Resources, Supervision, Writing – original draft, Writing – re­ through competitive pathways of hydrocracking and isomerization over HPW-Ni/
view & editing. Ignacio Elizalde-Martínez: Conceptualization, Meth­ MCM-41 catalyst. Fuel 2020;269:117465.
odology, Formal analysis, Resources, Supervision, Funding acquisition, [17] Fedyna M, Żak A, Jaroszewska K, Mokrzycki J, Trawczyński J. Composite of Pt/
AlSBA-15+zeolite catalyst for the hydroisomerization of n-hexadecane: the effect
Writing - original draft, Writing - review & editing. of platinum precursor. Microporous Mesoporous Mater 2020;305:110366.
[18] Jaroszewska K, Masalska A, Grzechowiak JR, Grams J. Hydroconversion of 1-
methylnaphthalene over Pt/AlSBA-15–Al2O3 composite catalysts. Appl Catal A
Declaration of Competing Interest 2015;505:116–30.
[19] Ochoa-Hernández C, Yang Y, Pizarro P, de la Peña O’Shea VA, Coronado JM,
Serrano DP. Hydrocarbons production through hydrotreating of methyl esters over
The authors declare that they have no known competing financial Ni and Co supported on SBA-15 and Al-SBA-15. Catal Today 2013;210:81–8.
interests or personal relationships that could have appeared to influence [20] Rouquerol J, Rouquerol F. 3 - Methodology of Gas Adsorption. In: Rouquerol F,
Rouquerol J, Sing KSW, Llewellyn P, Maurin G, editors. Adsorption by Powders and
the work reported in this paper.
Porous Solids. Second Edition. Oxford: Academic Press; 2014. p. 57–104.
[21] Zhao D, Feng J, Huo Q, Melosh N, Fredrickson GH, Chmelka BF, et al. Triblock
Data availability copolymer syntheses of mesoporous silica with periodic 50 to 300 angstrom pores.
Science 1998;279(5350):548–52.
[22] Bi Y, Zeng S, Nie H, Li M, Yang Q, Xia G, et al. Self-complexation: the nature for
Data will be made available on request. high hydrodesulfurization activity of NiWO4 nanoparticles. Catal Commun 2014;
57:5–8.
[23] Lei Z, Gao L, Shui H, Chen W, Wang Z, Ren S. Hydrotreatment of heavy oil from a
Acknowledgments direct coal liquefaction process on sulfided Ni–W/SBA-15 catalysts. Fuel Process
Technol 2011;92(10):2055–60.
The authors acknowledge the support given by the Science and [24] Li Y, Pan D, Yu C, Fan Y, Bao X. Synthesis and hydrodesulfurization properties of
NiW catalyst supported on high-aluminum-content, highly ordered, and
Technology National Council (CONACyT) in Mexico for doctoral hydrothermally stable Al-SBA-15. J Catal 2012;286:124–36.
research studies grant in 2020, the PIFI program in the National Poly­ [25] Li D, Li Y, Liu X, Guo Y, Pao C-W, Chen J-L, et al. NiAl2O4 spinel supported Pt
technique Institute, IPN, for the research grant in 2022, both given to J. catalyst: high performance and origin in aqueous-phase reforming of methanol.
ACS Catal 2019;9(10):9671–82.
R. Restrepo-Garcia, and to the Centre of Nanosciences and Micro and
[26] Blanchard J, Breysse M, Fajerwerg K, Louis C, Hédoire CE, Sampieri A, et al. Acidic
Nanotechnologies, CNMN-IPN, for experimental support in data acqui­ zeolites and Al-SBA-15 as supports for sulfide phases: application to hydrotreating
sition for characterization of materials and catalysts during the devel­ reactions. In: J. Čejka NŽ, Nachtigall P, editors. Studies in Surface Science and
opment of this research. Catalysis. Volume 158, Part B: Elsevier; 2005. p. 1517-24.
[27] Dragoi B, Dumitriu E, Bennici S, Auroux A. Acidic and adsorptive properties of Al
modified SBA-15 samples. In: Antoine Gédéon PM, Florence B, editors. Studies in
References Surface Science and Catalysis, Volume 174. Part B: Elsevier; 2008. p. 953–6.
[28] Gómez-Cazalilla M, Infantes-Molina A, Moreno-Tost R, Maireles-Torres PJ, Mérida-
Robles J, Rodríguez-Castellón E, et al. Al-SBA-15 as a support of catalysts based on
[1] Restrepo-Garcia JR, Ramírez GE, Baldovino-Medrano VG. Hydroprocessing of
chromium sulfide for sulfur removal. Catal Today 2009;143(1-2):137–44.
Phenanthrene Over Sulfided Fe–W Supported on Modified SBA-15. Catal Lett 2018;
[29] Thomas R, Kerkhof FPJM, Moulijn JA, Medema J, de Beer VHJ. On the formation
148(2):621–41.
of aluminum tungstate and its presence in tungsten oxide on γ-alumina catalysts.
[2] Templis C, Vonortas A, Sebos I, Papayannakos N. Vegetable oil effect on gasoil HDS
J Catal 1980;61(2):559–61.
in their catalytic co-hydroprocessing. Appl Catal B 2011;104(3):324–9.
[30] Lima NA, Alencar LDS, Siu-Li M, Feitosa CAC, Mesquita A, M’peko J-C, et al.
[3] Varakin AN, Mozhaev AV, Pimerzin AA, Nikulshin PA. Toward HYD/DEC
NiWO4 powders prepared via polymeric precursor method for application as
selectivity control in hydrodeoxygenation over supported and unsupported Co(Ni)-
ceramic luminescent pigments. J Adv Ceram 2020;9(1):55–63.
MoS2 catalysts. A key to effective dual-bed catalyst reactor for co-hydroprocessing
[31] Cancino-Trejo F, Santes V, Cardenas JAA, Gallardo M, Maldonado YG, Miranda AL,
of diesel and vegetable oil. Catal Today 2020;357:556–64.
et al. Active Ni and Fe species on catalysts Ni/Al2O3 and NiFe/Al2O3 for the
[4] De Paz Carmona H, de la Torre Alfaro O, Brito Alayón A, Romero Vázquez MA,
oxidative dehydrogenation (ODH) of ethane to ethylene assisted by CO2. Chem Eng
Macías Hernández JJ. Co-processing of straight run gas oil with used cooking oil
J Adv 2022;12:100404.
and animal fats. Fuel 2019;254:115583.
[32] Scaldaferri CA, Pasa VMD. Production of jet fuel and green diesel range
[5] El-Sawy MS, Hanafi SA, Ashour F, Aboul-Fotouh TM. Co-hydroprocessing and
biohydrocarbons by hydroprocessing of soybean oil over niobium phosphate
hydrocracking of alternative feed mixture (vacuum gas oil/waste lubricating oil/
catalyst. Fuel 2019;245:458–66.
waste cooking oil) with the aim of producing high quality fuels. Fuel 2020;269:
[33] Barison A, da Silva CW, Campos FR, Simonelli F, Lenz CA, Ferreira AG. A simple
117437.
methodology for the determination of fatty acid composition in edible oils through
[6] Vlasova EN, Porsin AA, Aleksandrov PV, Nuzhdin AL, Bukhtiyarova GA. Co-
1H NMR spectroscopy. Magn Resonance Chem 2010;48(8):642–50.
hydroprocessing of straight-run gasoil – Rapeseed oil mixture over stacked bed
[34] Siudem P, Zielińska A, Paradowska K. Application of 1H NMR in the study of fatty
Mo/Al2O3 + NiMo/Al2O3-SAPO-11 catalysts. Fuel 2021;285:119504.
acids composition of vegetable oils. J Pharm Biomed Anal 2022;212:114658.
[7] Vlasova EN, Porsin AA, Aleksandrov PV, Nuzhdin AL, Bukhtiyarova GA. Co-
[35] Sepúlveda C, Escalona N, García R, Laurenti D, Vrinat M. Hydrodeoxygenation and
processing of rapeseed oil — straight run gas oil mixture: Comparative study of
hydrodesulfurization co-processing over ReS2 supported catalysts. Catal Today
sulfide CoMo/Al2O3-SAPO-11 and NiMo/Al2O3-SAPO-11 catalysts. Catal Today
2012;195(1):101–5.
2021;378:119–25.
[8] Wijaya K, Utami M, Damayanti AK, Tahir I, Tikoalu AD, Rajagopal R, et al. Nickel-
modified sulfated zirconia catalyst: Synthesis and application for transforming
waste cooking oil into biogasoline via a hydrocracking process. Fuel 2022;322:
124152.
[9] Burimsitthigul T, Yoosuk B, Ngamcharussrivichai C, Prasassarakich P.
Hydrocarbon biofuel from hydrotreating of palm oil over unsupported Ni–Mo
sulfide catalysts. Renew Energy 2021;163:1648–59.

11
J.R. Restrepo-Garcia et al. Fuel 351 (2023) 128890

[36] Infantes-Molina A, Pawelec B, Fierro JLG, Loricera CV, Jiménez-López A, [38] Kagami N, Vogelaar BM, Langeveld ADV, Moulijn JA. Reaction pathways on NiMo/
Rodríguez-Castellón E. Effect of Ir and Pt addition on the HDO performance of Al2O3 catalysts for hydrodesulfurization of diesel fuel. Appl Catal A 2005;293:
RuS2/SBA-15 sulfide catalysts. Top Catal 2015;58(4-6):247–57. 11–23.
[37] Şenol Oİ, Ryymin EM, Viljava TR, Krause AOI. Effect of hydrogen sulphide on the
hydrodeoxygenation of aromatic and aliphatic oxygenates on sulphided catalysts.
J Mol Catal A Chem 2007;277(1):107–12.

12