Ultradeep hydrodesulfurization of fuel over superior NiMoS phases constructed by a novel Ni(MoS4)2(C13H30N)2precursor

Article abstract of DOI:10.1039/d0cy01177k

This article presents novel decyltrimethylammonium bromide-dispersed Ni-Mo sulfide (DTMA-NiMo) as a precursor for preparing an efficient NiMoS/γ-Al₂O₃hydrodesulfurization (HDS) catalyst. The as-synthesized DTMA-NiMo is a sulfide containing both long-chain quaternary ammonium and Ni-Mo-S elements. The proposed method not only significantly improves the dispersion of Mo species but also greatly promotes the incorporation of Ni into MoS₂slabs, leading to an increase in the number of NiMoS phases. As a result, the DTMA-NiMo-based NiMoS/γ-Al₂O₃catalyst exhibits much higher activity for the HDS of 4,6-dimethyldibenzothiophene (4,6-DMDBT) and fluid catalytic cracking (FCC) diesel than NiMoS/γ-Al₂O₃catalysts prepared by the co-impregnation and tetrapropylammonium bromide (TPAB)-assisted methods. This novel strategy sheds a light on the facile and low-cost preparation of superior NiMoS phases without sulfidation treatment.

Full text of DOI:10.1039/d0cy01177k

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Ultradeep hydrodesulfurization of fuel over
superior NiMoS phases constructed by a novel
Cite this: DOI: 10.1039/d0cy01177k
Ni(MoS4)2(C13H30N)2 precursor†
*
Jundong Xu, Chenglong Wen, Shuisen He and Yu Fan
This article presents novel decyltrimethylammonium bromide-dispersed Ni–Mo sulfide (DTMA-NiMo) as a
precursor for preparing an efficient NiMoS/γ-Al2O3 hydrodesulfurization (HDS) catalyst. The as-synthesized
DTMA-NiMo is a sulfide containing both long-chain quaternary ammonium and Ni–Mo–S elements. The
proposed method not only significantly improves the dispersion of Mo species but also greatly promotes
the incorporation of Ni into MoS2 slabs, leading to an increase in the number of NiMoS phases. As a result,
the DTMA-NiMo-based NiMoS/γ-Al2O3 catalyst exhibits much higher activity for the HDS of
4,6-dimethyldibenzothiophene (4,6-DMDBT) and fluid catalytic cracking (FCC) diesel than NiMoS/γ-Al2O3
catalysts prepared by the co-impregnation and tetrapropylammonium bromide (TPAB)-assisted methods.
This novel strategy sheds a light on the facile and low-cost preparation of superior NiMoS phases without
sulfidation treatment.
Received 10th June 2020,
Accepted 27th July 2020
DOI: 10.1039/d0cy01177k
rsc.li/catalysis
To enhance the amount of Ni(Co)MoS phases, various
methods have been used to prepare Ni(Co)Mo catalysts, such
1. Introduction
Hydrodesulfurization (HDS) is one of the most widely used
processes to improve fuel quality in the petroleum refining
industry.1 Currently, Ni(Co)–Mo-based sulfides supported on
alumina are still the main catalysts used in diesel HDS.2
However, with the decline in crude-oil quality, the main
sulfur-containing compounds in diesel are those with steric
hindrance, such as 4,6-DMDBT, which are refractory to
removal; thus, the activity of the catalysts needs to be
improved.3–5 In general, bimetallic catalysts are prepared by
impregnating alumina with oxidic precursors such as
ammonium heptamolybdate and nickel or cobalt salt
solutions (simultaneously or successively), followed by drying,
calcination and activation via sulfidation.6 In these processes,
nickel or cobalt partially reacts with alumina to form Ni(Co)
Al2O4 spinels, which are inactive in the HDS reaction;
moreover, Mo species can also partially interact with alumina
to form Mo–O–Al bonds, which makes sulfidation of the Mo
species difficult and decreases the stacking number of MoS2
slabs.7–9 As a consequence, nickel or cobalt could not
decorate MoS2 appropriately to form Ni(Co)MoS phases with a
high stacking number, which are known as the highly active
HDS phases for refractory sulfur-containing compounds.
as chemical vapor deposition (CVD),10–12 the addition of
chelating agents (such as nitrilotriacetic acid (NTA), citric
acid and ethylenediaminetetraacetic acid (EDTA)),13,14 and
the use of different Ni(Co)Mo precursors (such as oxidic
Ni(Co)Mo-based polyoxometalates and Ni(Co)Mo-based
bimetallic
sulfides).15–20
Compared
to
ammonium
heptamolybdate and nickel or cobalt salts, Ni(Co)Mo-based
polyoxometalates and Ni(Co)Mo-based bimetallic sulfides
contain both Ni(Co) and Mo simultaneously, and the close
interaction between Ni(Co) and Mo is beneficial for
decorating MoS2 slabs with Ni(Co).21,22 Griboval et al.23–25
and Thomas et al.26,27 used Keggin-type polyoxometalates
(such as Co3/2PMo12O40, Co7/2PMo12O40 and Co2SiMo12O40)
and Anderson-type polyoxometalates (such as (NH4)3(CoMo6-
O
24H6)·7H2O and (NH4)6(Co2Mo10O38H4·7H2O)), respectively,
to prepare CoMo/γ-Al2O3 catalysts, and the as-prepared
catalysts possessed higher amounts of CoMoS phases than
conventional CoMo/Al2O3 HDS catalysts prepared by
impregnation. However, the low stability, the low Co/(Co +
Mo) atomic ratio (<0.3) and the difficulty of sulfidation of
both types of polyoxometalates limit their application.17
Taniguchi et al.28 used (Mo3NiS4Cl(H2O)9)3+ to prepare NiMo/
zeolite catalysts, and the as-prepared catalysts possessed
higher HDS activity than conventional NiMo/Al2O3 HDS
catalysts prepared by impregnation, but the low Ni/(Ni + Mo)
atomic ratio (<0.3), the poor stability and the complex
preparation process of (Mo3NiS4Cl(H2O)9)3+ limit its
application for the efficient HDS of refractory 4,6-DMDBT.29
State Key Laboratory of Heavy Oil Processing, China University of Petroleum,
Beijing, 102249, China. E-mail: fanyu@cup.edu.cn; Tel: +86 10 89734981
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0cy01177k
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Herein, we prepared a novel NiMoS/γ-Al2O3 catalyst with a
Catalysis Science & Technology
Al2O3-D was obtained. The levels of Ni and Mo loaded in s-
NiMo/γ-Al2O3-D were 3.3 and 10.5 wt%, respectively, as
detected using an X-ray fluorescence (XRF) spectrometer on
an X'Pert Axios apparatus.
high Ni/Mo ratio, high amount of NiMoS phases and high
dispersion of NiMoS phases using decyltrimethylammonium
bromide-dispersed Ni–Mo sulfide. In this method, the Ni and
Mo species were incorporated into the same sulfide with a
high Ni/Mo ratio, which is beneficial for Ni decorating on
MoS2 slabs to form additional NiMoS phases. The as-
prepared NiMoS/γ-Al2O3 catalyst exhibited higher activity for
HDS of 4,6-DMDBT and FCC diesel compared to NiMoS/γ-
Al2O3 catalysts prepared by co-impregnation and
tetrapropylammonium bromide (TPAB)-assisted methods.
For comparison, the other two NiMoS/γ-Al2O3 catalysts
were prepared, designated s-NiMo/γ-Al2O3-T and s-NiMo/γ-
Al2O3-CI. The difference between s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D was that DTMAB was replaced by TPAB
during the preparation process, and the as-prepared
precursor was denoted as TPA-NiMo. The s-NiMo/γ-Al2O3-CI
catalyst was prepared by co-impregnation of γ-Al2O3 with an
ethanolamine–deionized water solution of ATTM and nickel
nitrate. After co-impregnation, the sample was dried in a
vacuum oven at 80 °C for 24 h and calcined at 450 °C for 4 h
in a N2 atmosphere. The collected product was s-Mo/γ-Al2O3-
CI. The Ni loads of the s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-
CI catalysts detected by XRF were 3.2 and 3.5 wt%,
respectively; the Mo loads of the s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D catalysts were 10.6 and 10.8 wt%,
respectively.
2. Experimental
2.1. Materials
The γ-Al2O3 support contained 98.2 wt% Al2O3, with a pore
volume of 0.97 mL g−1, surface area of 322 m2 g−1 and
average diameter of 12.0 nm. 4,6-DMDBT (C14H12S, 99%) was
obtained from Beijing J & K Scientific Ltd (P. R. China).
Formamide (CHONH2, 99%), TPAB ((CH3CH2CH2)4NBr, 99%),
decyltrimethylammonium bromide (DTMAB, C10H21N(CH3)3-
Br, 99%) and ammonium tetrathiomolybdate ((NH4)2MoS4,
99%, denoted as ATTM) were obtained from Shanghai
Aladdin Biochemical Polytron Technologies Company, Ltd.
(P. R. China). Nickel nitrate hexahydrate (Ni(NO3)2·6H2O,
99%), ethanol (C2H5OH, 99%), naphthane (C10H18, 99%),
n-heptane (C7H16, 99%), cyclohexane (C6H12, 99%) and
carbon disulfide (CS2, 99%) were purchased from Beijing
Chemical Reagent Company (P. R. China). The above reagents
were used directly after purchase without purification.
2.3. Characterization
Elemental analysis of the precursor was performed through
inductively coupled plasma-optical emission spectrometry
(ICP-OES) on a Perkin-Elmer DV 4300 combined with a 2400
II CHN elemental analyzer. Elemental analyses of the sulfided
catalysts were performed on an X'Pert Axios XRF
spectrometer.
Raman experiments were performed on a Bruker Raman
spectrometer, and the spectra were recorded in 600–100 cm−1,
using a Nd:YAG laser excited at 532 nm as the laser source.
Fourier transform infrared (FT-IR) analyses were
performed on a Bruker 80v (Germany) FT-IR spectrometer
with a resolution of 4 cm−1. The recorded wavenumber range
of the samples was 400–4000 cm−1.
X-ray diffraction (XRD) characterization was performed on
a PANalytical Advance powder instrument that used Cu Kα
radiation (λ = 1.54 Å) and was operated at 40 kV and 40 mA.
The scanning speed was 5° min−1, and the diffraction lines of
2θ were between 10 and 80°.
The pore structure properties of the samples were
analyzed by N2 adsorption–desorption characterization and
these experiments were performed on a Micromeritics ASAP
2420 apparatus. Before the experiments, the samples were
degassed at 200 °C for 8 h. Then, the samples were subjected
to adsorption–desorption at −196 °C.
2.2. Synthesis of the sulfided NiMo precursor and
preparation of the catalysts
The sulfided NiMo-based precursor was synthetized as
follows: first, 0.64 g of (NH4)2MoS4 was dissolved in 15 mL of
formamide at 10 °C, and 0.36 g of Ni(NO3)2·6H2O was
dissolved in 35 mL of deionized water at 10 °C. Then, the
nickel nitrate aqueous solution was added slowly to the
ATTM solution at 10 °C. Then, 0.89 g of DTMAB was added
to the above mixed solution and stirred for 72 h at 10 °C
under a N2 atmosphere, and a resultant black suspension
was obtained. Finally, the suspension was suction filtered,
washed with ethanol, and dried at 120 °C for 12 h. The
acquired product was decyltrimethylammonium-linked NiMo
sulfide (denoted as DTMA-NiMo).
A NiMoS/γ-Al2O3 catalyst, designated s-NiMo/γ-Al2O3-D,
was prepared using DTMA-NiMo as a precursor. The typical
procedure for preparing s-NiMo/γ-Al2O3-D was as follows:
1.60 g of the γ-Al2O3 powder was added into a 150 mL round
flask containing 50 mL of DTMA-NiMo suspension and
stirred for 48 h at 10 °C under a N2 atmosphere. The
obtained product was suction filtered and washed with
ethanol. Then, the product was dried under a N2 atmosphere
at 100 °C for 12 h, and the DTMA-NiMo/γ-Al2O3 composite
was obtained. Finally, the DTMA-NiMo/γ-Al2O3 composite was
calcined at 450 °C for 4 h in a N2 atmosphere, and s-NiMo/γ-
Transmission electron microscopy (TEM) images of the
samples were acquired on an FEI Tecnai G2 F20 instrument
equipped with an energy dispersive X-ray spectrometer (EDX).
The accelerating voltage was 200 kV, and the point-to-point
resolution was 1.9 A. Before the experiments, the catalysts
were kept under a N2 atmosphere to avoid oxidation. The
sample powder was first suspended in ethanol and then
dropped onto a carbon polymer-coated copper grid and dried
under an infrared lamp.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
The sulfided catalysts were analyzed by X-ray
photoelectron spectroscopy (XPS). The spectra were acquired
on a PHI Quantera SXMTM spectrometer equipped with Al
Kα radiation operating at 40 eV pass energy. Before the
experiments, the catalysts were kept under a N2 atmosphere
to avoid oxidation and then mounted on a holder using
double-sided adhesive tape and transferred into the
spectrophotometer immediately. The binding energies of the
determined elements were corrected by using C1s (284.6 eV)
as a reference. The spectra were decomposed by using
XPSPEAK version 4.1 software with the mixed Gaussian–
Lorentzian function to determine the amounts of Mo4+, Mo5+
and Mo6+ species and different kinds of Ni species.30
of FCC diesel are presented in Table S1.† The total sulfur
content of FCC diesel was measured using an Analytik Jena
Multi EA 5000 (Germany) instrument via the pyrolysis
method, in accordance with the international standard
(ASTM D5453).
The rate constant of HDS (kHDS
,
mol g−1 s−1) was
calculated by the following equation:32
ꢀ
ꢁ
F
m
1
kHDS
¼
ln
(5)
1 − x
where F is the 4,6-DMDBT rate (mol s−1), m is the catalyst
mass (g), and x is the 4,6-DMDBT conversion. The average
stacking number (N) and slab length (L) were expressed by
the following equations.33
The absolute quantification of species i (C(i)), the effective
Ni content in the NiMoS phase (CNiMoS), the effective Mo
P
P
P
P
content in the MoS2 phase (CMoS ), and the promoter ratio in
N = xiNi/ xi and L = xiLi/ xi
(6)
2
the slab edge of the active phase ((Ni/Mo)edge) were calculated
as follows:31
where Ni and Li are the layer number and length of MoS2
slabs in each stack, respectively, and xi is the number of
MoS2 slabs with Ni layers of length Li.
Ai=Si
CðiÞ ¼
× 100
(1)
n
X
Ai=Si
The Mo dispersion degree ( fMo) and the turnover frequency
(TOF) were calculated according to eqn (7) and (8).33
i¼1
t
X
ANiMoS
CNiMoS
¼
× CðiÞ
(2)
(3)
ð6ni − 6Þ
ANiMoS þ A
þ A
NiSx
NiOx
i¼1
f Mo
¼
(7)
t
X
ꢂ
ꢃ
AMoS
3n2i − 3ni þ 1
2
CMoS
¼
× CðMoÞ
MoO3
2
i¼1
A
MoS2
þ A
þ A
MoOxSy
F × x
C
NiMoS=C
MoS2
TOF ¼
nMo × f Mo
(8)
ðNi=MoÞedge
¼
(4)
f Mo
where t is the number of total slabs, ni is the number of Mo
atoms along one edge of a MoS2 slab, x is the conversion rate of
4,6-DMDBT, F is the rate of 4,6-DMDBT turnover (mol s−1) and
nMo is the amount of Mo atoms in the catalyst (mol).
where Ai is the calculated area of species i and Si is the
sensitivity factor of atom i.
The H2 temperature-programmed reduction (H2-TPR)
measurements of the sulfided catalysts were performed on an
Auto Chem II 2920 instrument. First, 100 mg of each catalyst
was pretreated at 400 °C for 1 h under an Ar atmosphere and
cooled to 100 °C. Subsequently, the catalyst was heated from
100 to 800 °C under a 10% H2/N2 flow at a rate of 10 °C
3. Results and discussion
3.1. Properties of DTMA-NiMo and determination of DTMA-
NiMo loading onto γ-Al2O3
min−1. H2 consumption was determined using
detector.
a TCD
To investigate the properties of DTMA-NiMo, elemental
analysis and Raman and FT-IR characterization were
performed. The elemental compositions of DTMA-NiMo
determined by ICP-OES combined with CHN analysis were
2.4. Catalyst assessment
The HDS performance of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T
and s-NiMo/γ-Al2O3-D was assessed in fixed-bed
a
microreactor. The catalyst was loaded into the microreactor
after it was sieved through a 20–40 mesh and mixed with
quartz granules of the same size and volume. The HDS
performance of the catalyst was estimated at a pressure of 4.0
MPa, a H2/oil volumetric ratio of 400, a temperature of 320
°C and a liquid hourly space velocity (LHSV) of 20 h−1 with a
feedstock of naphthane solution containing 0.8 wt%
4,6-DMDBT. The product was detected using a Finnigan trace
GC–MS. In addition, the HDS performance of the NiMo/γ-
Al2O3 catalysts was assessed using FCC diesel. The properties
Fig. 1 Raman spectra of ATTM and DTMA-NiMo (a); FT-IR spectra of
DTMAB, ATTM and DTMA-NiMo, the inset shows the 650–400 cm−1
range for ATTM and DTMA-NiMo (b).
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Ni: 6.10 wt%, Mo: 21.22 wt%, C: 35.04 wt%, H: 6.40 wt%, N:
3.08 wt%, and S: 28.16 wt%. Therefore, the DTMA-NiMo
composition was C2.87H6.35N0.22Ni0.11Mo0.22S0.88, almost the
same as Ni(MoS4)2(C13H30N)2. To determine the inorganic
core of DTMA-NiMo, the Raman spectra of ATTM and DTMA-
NiMo are shown in Fig. 1a. ATTM shows two bands at 458
and 480 cm−1, which are ascribed to the symmetric and
asymmetric Mo–S vibrations of MoS42−, respectively.34 DTMA-
NiMo displays the vibrational bands at 433, 445, 490 and 510
cm−1, in which the bands at 433 and 445 cm−1 belong to
bridging Mo–S stretches and the bands at 490 and 510 cm−1
belong to the symmetric and asymmetric Mo–S vibration
bands, respectively.35,36 The symmetric and asymmetric
stretching vibration bands of the Mo–S bond in DTMA-NiMo
Fig. 2 N2 adsorption–desorption isotherms (a) and BJH pore size
distribution curves (b) of γ-Al2O3, DTMA-NiMo/γ-Al2O3 and s-NiMo/γ-
Al2O3-D.
γ-Al2O3 and s-NiMo/γ-Al2O3-D are type IV, indicating that they
have the characteristics of mesoporous materials; the hysteresis
ring is type H3, suggesting the presence of slit-type pores.43
Fig. 2b shows that the introduction of DTMA-NiMo into γ-Al2O3
reduces the size of the latter most probable pore size. The
results in Table 1 show that the average pore diameter, pore
volume and specific surface area of DTMA-NiMo/γ-Al2O3 are
smaller than those of γ-Al2O3, suggesting the loading of DTMA-
NiMo onto γ-Al2O3. The specific surface area, pore volume and
average pore diameter of s-NiMo/γ-Al2O3-D with calcination at
450 °C are higher than those of DTMA-NiMo/γ-Al2O3 without
calcination due to the decomposition of the organic carbon
chains in DTMA-NiMo, which proves that DTMA-NiMo is
loaded onto γ-Al2O3. To further prove the loading of DTMA-
NiMo onto γ-Al2O3, an s-NiMo/γ-Al2O3-DE catalyst was prepared
using alumina extrudates as a support. s-NiMo/γ-Al2O3-DE has
almost the same TEM results (Fig. S1†) and the same
4,6-DMDBT HDS activity (Table S2†) as s-NiMo/γ-Al2O3-D. These
results prove the successful deposition of DTMA-NiMo into the
support pores.
migrates to
a
higher frequency than those in ATTM,
2−
indicating that MoS4 is coordinated with the central metal
atom Ni to form Ni(MoS4)2 in DTMA-NiMo.35,36 The above
2−
analyses demonstrate that the inorganic core of DTMA-NiMo
2−
is Ni(MoS4)22−. Thus, Ni(MoS4)2 anions interact with DTMA+
cations to form
determine the
a
DTMA-NiMo compound. To further
structure of DTMA-NiMo, FT-IR
characterization of DTMAB, ATTM and DTMA-NiMo was
carried out, and the results are shown in Fig. 1b. In DTMAB,
the peaks at 2856 and 2928 cm−1 are assigned to the
stretching vibrations of –(CH2)n– (n > 4) and –CH2–,37
respectively, the peak at 1630 cm−1 is attributed to the
asymmetric bending vibration of (CH3)3N+–(NR4 ) in DTMAB,
+
and the peak at 1468 cm−1 belongs to the shear vibration of
–CH2–.38 In the FT-IR spectrum of DTMA-NiMo, there are also
characteristic peaks of DTMAB in DTMA-NiMo, and the
characteristic peak attributed to the –CH3 bending vibration
appears at 1384 cm−1, indicating that DTMA+ is contained in
DTMA-NiMo. In the range of 400–600 cm−1, DTMA-NiMo has
vibration peaks at 513, 485, 446 and 435 cm−1. According to
the literature,35,39–41 the peaks at 513 and 485 cm−1 belong to
the symmetric and asymmetric stretching vibrations of the
MoS bond, respectively, and the peaks at 446 and 435 cm−1
are attributed to bridging Mo–S stretches. The asymmetric
stretching vibration peak of the MoS bond in DTMA-NiMo
migrates to a higher frequency than that in ATTM (the inset
of Fig. 1b). This is because the vibration of the MoS bond
when MoS42− is coordinated with the central metal atom Ni is
stronger than that of the MoS bond in the uncoordinated
3.2. Characterization of the sulfided NiMoS/γ-Al2O3 catalysts
3.2.1. XRD and N2 adsorption–desorption. To explore the
dispersion of active metals on γ-Al2O3, the as-prepared
NiMoS/γ-Al2O3 catalysts were characterized by XRD, and the
results are shown in Fig. 3. All the catalysts presented
characteristic peaks at 2θ = 37.4°, 45.7° and 67° (JCPDS card
no. 29-1486), which are attributed to γ-Al2O3. In addition, s-
NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T also have peaks at 2θ =
33.5° and 58.3° (JCPDS card no. 37-1492) belonging to the
MoS2 (101) and (110) planes. However, there is no
2−
MoS42−, which shows that Ni(MoS4)2 is formed by the
2−
interaction of Ni and MoS4 in DTMA-NiMo.35,42 In addition,
in the DTMA-NiMo FT-IR spectrum, except for the peaks
2−
attributed to DTMA+ and Ni(MoS4)2
, no other peaks
Table 1 Textural properties of γ-Al2O3, s-NiMo/γ-Al2O3-CI, s-NiMo/γ-
Al2O3-T and s-NiMo/γ-Al2O3-D
appeared. The above analysis shows that DTMA-NiMo is a
2−
composite of Ni(MoS4)2
impurities.
and DTMA+, without other
Sample
SBETa (m2 g−1
)
Vpb (cm3 g−1
)
Dpc (nm)
γ-Al2O3
322
224
258
263
297
0.97
0.52
0.64
0.66
0.82
12.0
9.3
9.9
10.0
11.0
To detect the deposition of DTMA-NiMo on γ-Al2O3, an N2
adsorption–desorption experiment was performed. The N2
adsorption–desorption isotherms and BJH pore size
distribution diagrams of γ-Al2O3, DTMA-NiMo/γ-Al2O3 and s-
NiMo/γ-Al2O3-D are shown in Fig. 2. As seen from Fig. 2a, the
N2 adsorption–desorption isotherms of γ-Al2O3, DTMA-NiMo/
DTMA-NiMo/γ-Al2O3
s-NiMo/γ-Al2O3-CI
s-NiMo/γ-Al2O3-T
s-NiMo/γ-Al2O3-D
a
b
c
BET specific surface area. Pore volume. Average pore diameter
(4 Vp/Sg).
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
shows that s-NiMo/γ-Al2O3-D has the best pore structure
among the three catalysts, which is beneficial for promoting
the diffusion of 4,6-DMDBT in the catalyst and thereby
improving the HDS activity of s-NiMo/γ-Al2O3-D.
3.2.2. HRTEM. The typical HRTEM images of s-NiMo/γ-
Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D are shown
in Fig. 4. According to the EDX analysis results, the black
thread-like fringes in the images are MoS2 slabs. The
interlayer distance between the black thread-like fringes is
0.62 nm, which is consistent with the distance between the
(002) characteristic plane of the MoS2 crystal.44 Most of the
MoS2 slabs on s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T are
long while most of the MoS2 slabs on s-NiMo/γ-Al2O3-D are
short, indicating that s-NiMo/γ-Al2O3-D has superior active-
metal dispersion.
To further understand the lengths and stacking numbers
of the supported active metals, statistical analyses were
conducted on the basis of approximately 20 images,
containing 400–600 black thread-like fringes obtained from
different regions of each catalyst, and the statistical results
are shown in Fig. 5. The lengths of MoS2 slabs in s-NiMo/γ-
Al2O3-CI and s-NiMo/γ-Al2O3-T are mainly concentrated at
3.0–5.0 nm, while those in s-NiMo/γ-Al2O3-D are mainly
concentrated at 2.0–4.0 nm. For s-NiMo/γ-Al2O3-CI and s-
NiMo/γ-Al2O3-T, the percentages of MoS2 slabs with lengths
longer than 4.0 nm are 44.7% and 55.2%, respectively; for s-
NiMo/γ-Al2O3-D, only 19.5% of MoS2 slabs has lengths longer
than 4.0 nm, indicating that the use of DTMA-NiMo as a
precursor significantly decreases the lengths of MoS2 slabs.
From Fig. 5b, the distributions of the stacking numbers of
MoS2 slabs in s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D are all concentrated in 3–4 layers, indicating
that the three catalysts have almost identical stacking
numbers of MoS2 slabs.
Fig. 3 XRD patterns of γ-Al2O3 (a), s-NiMo/γ-Al2O3-CI (b), s-NiMo/γ-
Al2O3-T (c) and s-NiMo/γ-Al2O3-D (d).
characteristic peak attributed to MoS2 in s-NiMo/γ-Al2O3-D,
indicating that compared with s-NiMo/γ-Al2O3-CI and s-NiMo/
γ-Al2O3-T, s-NiMo/γ-Al2O3-D presents a smaller particle size of
Mo species on the surface of γ-Al2O3. Because of the low
loading of nickel in the NiMoS/γ-Al2O3 catalysts, no
characteristic peak of nickel species appears in the XRD
pattern of NiMoS/γ-Al2O3.
The pore structure parameters of γ-Al2O3, s-NiMo/γ-Al2O3-
CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D are listed in
Table 1. Compared with the γ-Al2O3 support, the specific
surface areas of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D are reduced by 19.9%, 18.3% and 7.8%,
respectively; the pore volumes are reduced by 34.0%, 32.0%,
and 15.5%, respectively; and the average pore diameters are
decreased by 17.5%, 16.7%, and 8.3%, respectively. These
decreases are attributed to the deposition of active metals. s-
NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T have similar average
pore diameters, pore volumes and specific surface areas;
however, compared with s-NiMo/γ-Al2O3-CI and s-NiMo/γ-
Al2O3-T, s-NiMo/γ-Al2O3-D has a larger average pore diameter,
pore volume and specific surface area. According to the above
XRD characterization results, the particle size of Mo species
on s-NiMo/γ-Al2O3-D is smaller than those on s-NiMo/γ-Al2O3-
CI and s-NiMo/γ-Al2O3-T, so it has minimal influence on the
pore structure of the γ-Al2O3 support. The above analysis
According to eqn (6), the average stacking numbers and
the average lengths of MoS2 slabs in s-NiMo/γ-Al2O3-CI, s-
NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D were calculated as
shown in Table 2. Combined with eqn (7), the dispersion
degree (fMo) of MoS2 slabs in the three bimetallic catalysts
was calculated as shown in Table 2. The dispersion degree,
the average stacking number and the average length of MoS2
slabs in s-NiMo/γ-Al2O3-CI are 0.25, 3.3 layers and 4.2 nm,
respectively. The average stacking number of MoS2 slabs in s-
NiMo/γ-Al2O3-T is almost the same as that of s-NiMo/γ-Al2O3-
Fig. 5 Statistical distributions of the lengths (a) and stacking numbers
(b) of MoS2 slabs on s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D.
Fig. 4 HRTEM images of s-NiMo/γ-Al2O3-CI (a), s-NiMo/γ-Al2O3-T (b)
and s-NiMo/γ-Al2O3-D (c).
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
and Mo3d3/2 binding energies in the MoO3 (Mo6+) phase are
232.2 0.1 eV and 235.3 0.1 eV, respectively.30 The binding
Table
2 HRTEM statistical results for s-NiMo/γ-Al2O3-CI, s-NiMo/γ-
Al2O3-T and s-NiMo/γ-Al2O3-D
energy at 226.0
0.1 eV is ascribed to S2s.47 In s-NiMo/γ-
Catalyst
L (nm)
N
fMo
Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D, there are no
peaks belonging to the oxidation state MoOxSy (Mo5+) phase.
This is because the above three catalysts were all prepared
by using a sulfur-containing molybdenum compound as a
precursor, which is directly decomposed into MoS2 during
calcination without going through the oxidic MoOxSy
intermediates. The Mo sulfidation degree (Mosulfidation) is
defined as Mosulfidation = Mo4+/(Mo4+ + Mo5+ + Mo6+),32,48 and
Mo4+, Mo5+ and Mo6+ represent the levels of the MoS2 (Mo4+),
MoOxSy (Mo5+) and MoO3 (Mo6+) phases, respectively. The
XPS fitting results in Table 3 show that the Mo sulfidation
degrees of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-
Al2O3-D are 93.2%, 93.5% and 93.6%, indicating almost full
sulfidation of Mo species in the three catalysts. The much
higher sulfidation degrees of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-
Al2O3-T and s-NiMo/γ-Al2O3-D compared to that of
commercial NiMo/Al2O3 result from the use of ammonium
tetrathiomolybdate ((NH4)2MoS4) as the molybdenum
precursor rather than ammonium heptamolybdate.49
The experimental and fitting results of the Ni2p spectra
for s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D
are shown in Fig. 7. In general, on the sulfided NiMo
bimetallic catalysts, nickel species exist in the form of NiOx,
NiSx (Ni2S3, Ni9S8 and NiS) and NiMoS phases. According to
the literature,30,50,51 the peaks of Ni2p spectra were
decomposed. The Ni 2p spectra contain NiOx peaks at 855.8,
858.4, and 861.7 eV, NiSx peaks at 852.9 and 855.0 eV, and
NiMoS peaks at 853.3 and 856.0 eV, and the relative levels of
the NiOx, NiSx and NiMoS phases calculated are listed in
Table 3. The percentage of the NiMoS phase in s-NiMo/γ-
Al2O3-D is 79.1%, which is higher than those in s-NiMo/γ-
Al2O3-CI (52.7%) and s-NiMo/γ-Al2O3-T (67.2%). The higher
percentage of the NiMoS phase in s-NiMo/γ-Al2O3-D results
from the promoting effect of its high MoS2 dispersion on the
generation of NiMoS phases because the MoS2 slabs can
serve as a secondary support for nickel species.52,53
s-NiMo/γ-Al2O3-CI
s-NiMo/γ-Al2O3-T
s-NiMo/γ-Al2O3-D
4.2
4.0
2.9
3.3
3.2
3.1
0.25
0.27
0.36
CI; however, the average length of MoS2 slabs in s-NiMo/γ-
Al2O3-T is slightly shorter than that in s-NiMo/γ-Al2O3-CI, and
the dispersion degree of MoS2 slabs in s-NiMo/γ-Al2O3-T is
slightly higher than that in s-NiMo/γ-Al2O3-CI. The average
stacking number of MoS2 slabs in s-NiMo/γ-Al2O3-D is almost
the same as that in s-NiMo/γ-Al2O3-CI, but the average length
of MoS2 slabs in s-NiMo/γ-Al2O3-D is much lower than that in
s-NiMo/γ-Al2O3-CI, and the dispersion degree of MoS2 slabs
in s-NiMo/γ-Al2O3-D is much higher than that in s-NiMo/γ-
Al2O3-CI. These results indicate that the catalyst prepared
with DTMA-NiMo as a precursor has good Mo dispersion.
3.2.3. XPS. To further understand the chemical state,
dispersion degree, sulfurization degree and surface
composition of the active metals in the prepared catalysts,
XPS characterization of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T
and s-NiMo/γ-Al2O3-D was performed. According to the
literature,45,46 the background of the Mo3d secondary
electron spectrum was subtracted, and then the peak
decomposition was performed (Shirley method). The ratio of
the half width of 3d3/2 to the half width of 3d5/2 is
approximately 1.2, and the relative area of 3d5/2 to the relative
area of 3d3/2 is approximately 1.5.
The experimental and fitting results of the Mo3d XPS
spectra of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-
Al2O3-D are shown in Fig. 6. The Mo3d5/2 and Mo3d3/2
binding energies in the MoS2 (Mo4+) phase are 229.0 0.1 eV
and 232.1
0.1 eV, respectively; the Mo3d5/2 and Mo3d3/2
binding energies in the intermediate MoOxSy (Mo5+) phase
are 230.2 0.1 eV and 233.4 0.1 eV, respectively; the Mo3d5/2
In Table 3, the (Ni/Mo)edge ratio of s-NiMo/γ-Al2O3-D (1.22)
is higher than those of s-NiMo/γ-Al2O3-CI (0.81) and s-NiMo/
γ-Al2O3-T (0.98), which indicates that the decoration degree of
Ni on the edges of MoS2 in s-NiMo/γ-Al2O3-D is better than
those in s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T.47 The much
higher relative content of Ni in NiMoS and the higher (Ni/
Mo)edge ratio over s-NiMo/γ-Al2O3-D than those over s-NiMo/γ-
Al2O3-CI and s-NiMo/γ-Al2O3-T indicate that, compared with
the co-impregnation and tetrapropylammonium bromide
(TPAB)-assisted methods, the novel precursor method
facilitates Ni decoration on MoS2 slabs to form a high
number of NiMoS phases.
The surface atomic ratios of Mo/Al and Ni/Al in s-NiMo/γ-
Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D were
detected by XPS, and the results are shown in Table 3.
Though the three catalysts possess almost the same metal
contents, s-NiMo/γ-Al2O3-D has a higher surface atomic ratio
Fig. 6 Mo 3d XPS patterns of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T
and s-NiMo/γ-Al2O3-D.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
Table 3 Fitting results of XPS on s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D
Ni fraction (%)
Mosulfidation
(%)
Catalyst
Mo/Al
Ni/Al
NiOx
NiSx
NiMoS
(Ni/Mo)edge
s-NiMo/γ-Al2O3-CI
s-NiMo/γ-Al2O3-T
s-NiMo/γ-Al2O3-D
0.092
0.121
0.142
0.032
0.037
0.042
93.2
93.5
93.6
18.4
—
—
28.9
32.8
20.9
52.7
67.2
79.1
0.81
0.98
1.22
than the other two catalysts, indicating the better dispersion
of NiMoS phases supported on s-NiMo/γ-Al2O3-D as compared
to s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T.54
3.3. Catalytic activity
A naphthane solution containing 0.8 wt% 4,6-DMDBT was
used as a feedstock to evaluate the HDS activity of s-NiMo/γ-
Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D. The HDS
reaction rate constants (kHDS) and TOF values of s-NiMo/γ-
Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D at 4 MPa, a
H2/oil volumetric ratio of 400 and 320 °C are listed in
Table 4. The kHDS value of s-NiMo/γ-Al2O3-D is 6.35 × 10−7
mol g−1 s−1, which is 2.05 times that of s-NiMo/γ-Al2O3-CI and
1.69 times that of s-NiMo/γ-Al2O3-T. The TOF value of s-NiMo/
γ-Al2O3-D is 10.01 × 10−4 s−1, which is 1.66 times that of s-
NiMo/γ-Al2O3-CI and 1.40 times that of s-NiMo/γ-Al2O3-T. The
TOF and kHDS values of s-NiMo/γ-Al2O3-D are higher than
those of s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T, indicating
that s-NiMo/γ-Al2O3-D has better HDS activity than s-NiMo/γ-
Al2O3-CI and s-NiMo/γ-Al2O3-T.
The reaction pathways of 4,6-DMDBT on NiMoS/γ-Al2O3
are shown in Fig. 9.59 The HDS of 4,6-DMDBT on NiMoS/γ-
Al2O3 mainly includes the following pathways: hydrogenation
reaction (HYD) and direct desulfurization reaction (DDS).
The primary products detected during 4,6-DMDBT HDS on
NiMoS/γ-Al2O3 are: 3,3′-dimethylbicyclohexane (3,3′-DMBCH),
4,6-hexahydrodimethyldibenzothiophene
3,3′-dimethylbiphenyl
dimethylcyclohexyltoluene
3.2.4. H2-TPR. The H2-TPR characterization of the sulfided
s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D
catalysts provides the information about the strength of the
Mo–S bond and the relative number of surface active
sites.55,56 Fig. 8 shows the H2-TPR patterns of s-NiMo/γ-Al2O3-
CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D. The three
catalysts have similar characteristic peaks, corresponding to
the high-intensity resolution peak in the low-temperature
region (210 °C) and the low-intensity broad peak in the high-
temperature region. According to the literature,56–58 the
reduction peak in the low-temperature region is ascribed to
the reduction of sulfur atoms at the edge of MoS2 with weak
Mo–S bonding strength. The low-temperature reduction peak
temperatures of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D are 223, 219 and 213 °C, respectively. The
temperature of the low-temperature reduction peak on s-
NiMo/γ-Al2O3-D is lower than those on s-NiMo/γ-Al2O3-CI and
s-NiMo/γ-Al2O3-T because the former has a higher dispersion
of Mo sulfides as demonstrated by the above HRTEM
characterization. These results indicate that s-NiMo/γ-Al2O3-D
has a higher reducibility than s-NiMo/γ-Al2O3-CI and s-NiMo/
γ-Al2O3-T.56 In addition, s-NiMo/γ-Al2O3-D has a much higher
low-temperature peak area than s-NiMo/γ-Al2O3-CI and s-
NiMo/γ-Al2O3-T because the former has more NiMoS phases
as demonstrated by the above XPS characterization.
(4,6-HHDMDBT),
(3,3′-DMBP),
(3,3′-MCHT)
3,3′-
and
4,6-tetrahydrodimethyldibenzothiophene (4,6-THDMDBT).3,60
At the identical conversion (50%) of 4,6-DMDBT, the HDS
product distributions of the three bimetallic catalysts were
compared, and the results are listed in Table 4. The molar
ratios of 3,3′-MCHT to 3,3′-DMBP in the reaction products of
4,6-DMDBT HDS on all catalysts are higher than 1, which
shows that the HDS of 4,6-DMDBT on s-NiMo/γ-Al2O3-CI, s-
Fig. 7 Ni 2p XPS spectra of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and
s-NiMo/γ-Al2O3-D.
Fig. 8 H2-TPR patterns of s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and
s-NiMo/γ-Al2O3-D.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 4 HDS results for 4,6-DMDBT on s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D
KHDS
TOFa × 104
(s−1
a
Product selectivity (%)
Product ratio
Catalyst
(10−7 mol g−1 s−1
)
)
TH + HH
MCHT
DMBCH
DMBP
(TH + HH)/MCHT
MCHT/DMBP
s-NiMo/γ-Al2O3-CI
s-NiMo/γ-Al2O3-T
s-NiMo/γ-Al2O3-D
3.10
3.76
6.35
6.03
7.14
10.01
10.3
7.5
5.3
54.7
59.4
63.7
4.0
4.9
5.5
31.0
28.2
24.5
0.19
0.13
0.08
1.76
2.11
2.60
a
Calculated with the 4,6-DMDBT conversion at about 10%.
NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D is mainly carried out by
the HYD route. This is mainly due to the steric hindrance
effect of 4,6-DMDBT at its positions 4 and 6, which inhibits
the interaction between the sulfur atoms and the active metal
centers, making it difficult for 4,6-DMDBT to generate σ
adsorption with the active metal centers through the vertical
adsorption of sulfur atoms.61 However, the methyl groups at
the 4 and 6 positions will not affect 4,6-DMDBT to generate σ
adsorption with the active metal centers through the
horizontal adsorption of the aromatic ring.62 Therefore, the
desulfurization reaction via the HYD pathway is dominant
during 4,6-DMDBT HDS.63 The molar ratio of 3,3′-MCHT to
3,3′-DMBP in the HDS product of 4,6-DMDBT on s-NiMo/γ-
Al2O3-D is higher than those on s-NiMo/γ-Al2O3-CI and s-
NiMo/γ-Al2O3-T, which shows that s-NiMo/γ-Al2O3-D has a
better HYD activity than s-NiMo/γ-Al2O3-CI and s-NiMo/γ-
Al2O3-T.
Conventional NiMo/γ-Al2O3 (3.8 wt% Ni and 11.0 wt% Mo,
designated NiMo/γ-Al2O3-C) prepared by impregnating
alumina with the solutions of ammonium heptamolybdate
and nickel nitrate was also selected as a reference catalyst to
evaluate the activities and stabilities of the as-prepared
catalysts. The time course of changes in total sulfur levels in
the 4,6-DMDBT model oil HDS product with reaction on
NiMo/γ-Al2O3-C, s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-
NiMo/γ-Al2O3-D is shown in Fig. 10. Under the reaction
conditions of 320 °C, 4.0 MPa, a LHSV of 20 h−1, and a H2/oil
volumetric ratio 400, NiMo/γ-Al2O3-C, s-NiMo/γ-Al2O3-CI, s-
NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D exhibit stable sulfur
levels in the HDS products of 4,6-DMDBT model oil within
142 h of operation time. This proves that the four catalysts
have good HDS stability. In addition, the HDS activities of
the four catalysts using FCC diesel as a feedstock were
further investigated. The total sulfur content in the feed was
3640 μg g−1, and the reaction conditions were 350 °C, 6.0
MPa, a LHSV of 2.5 h−1, and a H2/oil volumetric ratio 400.
From Table 5, s-NiMo/γ-Al2O3-D has the highest HDS ratio
among the four catalysts, indicating its superiority in the
HDS of FCC diesel. These results indicate that s-NiMo/γ-
Al2O3-D has application potential in the ultra-deep HDS of
FCC diesel.
3.4. Origin of the active phase structures of the catalysts
The above characterization results indicate that s-NiMo/γ-
Al2O3-D prepared with DTMA-NiMo as a precursor has high
Mo dispersion and a great number of NiMoS phases. To
explore the influence of TPA-NiMo and DTMA-NiMo as
precursors on the morphology of Mo species in the prepared
s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D, the TPA-NiMo and
DTMA-NiMo precursor suspensions were characterized by
TEM, as shown in Fig. 11. The statistical analysis result for
the nanoparticle size on DTMA-NiMo, obtained by measuring
at least 500 nanoparticles, is shown in the inset of Fig. 11a.
The nanoparticle size of DTMA-NiMo is mainly distributed at
2.4–3.2 nm. The DTMA-NiMo composites are highly
dispersed and uniform nanoparticles with an average particle
size of approximately 3.0 nm (Fig. 11a and b). Unlike the
uniformly monodispersed nanoparticles of DTMA-NiMo, TPA-
NiMo mainly exists in the form of large aggregates
(Fig. 11c and d). The different existing states of DTMA-NiMo
and TPA-NiMo are explained as follows: in DTMA-NiMo, each
DTMA-NiMo nanoparticle has two long carbon chains, and
the long carbon chains have strong hydrophobic–hydrophilic
Fig. 10 Total sulfur levels in the product of 4,6-DMDBT HDS over the
reaction time on NiMo/γ-Al2O3-C (a), s-NiMo/γ-Al2O3-CI (b), s-NiMo/γ-
Al2O3-T (c) and s-NiMo/γ-Al2O3-D (d).
Fig. 9 Reaction network of 4,6-DMDBT HDS on these catalysts.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
Table 5 FCC diesel HDS results for s-NiMo/γ-Al2O3-CI, s-NiMo/γ-Al2O3-T and s-NiMo/γ-Al2O3-D
Catalyst
Total sulfur content in feedstock (μg g−1
)
Total sulfur content in feedstock (μg g−1
)
HDS ratio (%)
NiMo/γ-Al2O3-C
s-NiMo/γ-Al2O3-CI
s-NiMo/γ-Al2O3-T
s-NiMo/γ-Al2O3-D
3640
3640
3640
3640
208.7
194.5
132.3
7.2
94.3
94.7
96.4
99.8
repulsion with the surrounding water molecules.64 This effect
makes DTMA-NiMo uniformly monodispersed in the
synthetic solution. However, the TPA-NiMo particles have
short carbon chains, which lead to weak hydrophobic–
hydrophilic repulsion with the surrounding water
molecules.65,66 As a consequence, the particles of TPA-NiMo
are suspended in the synthetic solution in the form of large
aggregates.
γ-Al2O3 in the form of monodispersed nanoparticles after the
addition of γ-Al2O3. During the drying process, the alkyl
carbon chains around TPA-NiMo are relatively short, resulting
in the basically unchanged viscosity of the solution with
vaporization of the solvent;54,64 therefore, the solution is
redistributed, and TPA-NiMo is further aggregated. During
calcination in a N2 atmosphere, it is difficult to effectively
improve the dispersion of NiMo species due to the
aggregation of TPA-NiMo particles during the initial
deposition and subsequent drying process, despite the
carbonization of the carbon chains present in TPA-NiMo.
Therefore, compared with s-NiMo/γ-Al2O3-T, s-NiMo/γ-Al2O3-D
has a higher dispersion of NiMoS phases.
The HDS evaluation results of 4,6-DMDBT and real FCC
diesel have proven that s-NiMo/γ-Al2O3-D has better HDS
activity than s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T. The
XRD and N2 adsorption–desorption characterization results
have shown that s-NiMo/γ-Al2O3-D has smaller metal species
than s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T, which improve
the accessibility of active phases. As demonstrated by the XPS
results, s-NiMo/γ-Al2O3-D has more NiMoS phases than s-
NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T, which enhance the
efficiency of 4,6-DMDBT HDS. The HRTEM results have
proved that s-NiMo/γ-Al2O3-D, s-NiMo/γ-Al2O3-CI and s-NiMo/
γ-Al2O3-T have MoS2 slabs with about 3-layer stackings,
guaranteeing good accessibility of NiMoS active phases for
4,6-DMDBT. s-NiMo/γ-Al2O3-CI and s-NiMo/γ-Al2O3-T with low
dispersion degrees of 0.25 and 0.27, respectively, have
relatively few exposed NiMoS phases, which weakens their
4,6-DMDBT HDS performance. In contrast, s-NiMo/γ-Al2O3-D,
with a high dispersion degree of 0.36 has a sufficient number
of exposed NiMoS phases, which enhances both the HYD and
DDS routes for 4,6-DMDBT HDS. Therefore, s-NiMo/γ-Al2O3-D,
with sufficient and accessible NiMoS active phases, allows
According to the above analyses, the preparation
principles for s-NiMo/γ-Al2O3-D and s-NiMo/γ-Al2O3-T are
proposed in Fig. 12. During the deposition process, the long
alkyl carbon chains around the nanoparticles of DTMA-NiMo
can inhibit the aggregation of NiMo species and DTMA-NiMo
is uniformly monodispersed in the synthetic solution with
fine nanoparticles; after adding γ-Al2O3, these nanoparticles
can be effectively deposited on the surface of γ-Al2O3 with
high dispersion. During the drying process, the interaction
between the long alkyl carbon chains around DTMA-NiMo
nanoparticles leads to an increase in the solution viscosity
with vaporization of the solvent, effectively inhibiting the
redistribution of the solution and thereby restraining the
aggregation
of
DTMA-NiMo
nanoparticles.7
During
calcination in a N2 atmosphere, the carbonization of the long
carbon chains on the outer layer of DTMA-NiMo produces a
barrier for the aggregation of NiMo species and further
improves the dispersion of NiMo species. For s-NiMo/γ-Al2O3-
T, during the deposition process, because TPA-NiMo exists in
the synthetic solution in the form of large aggregates, it is
difficult for TPA-NiMo to accumulate on the surface of
Fig. 11 TEM images of DTMA-NiMo (a and b), size distribution of
DTMA-NiMo (the inset of a) and TEM images of TPA-NiMo (c and d).
Fig. 12 Schematic diagram of the preparation principles of s-NiMo/γ-
Al2O3-D (a) and s-NiMo/γ-Al2O3-T (b).
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
efficient adsorption and HDS reactions of refractory sulfur-
containing compounds with steric hindrance.
11 Z. Liu, M. Amani, S. Najmaei, Q. Xu, X. Zou, W. Zhou, T. Yu, C.
Qiu, A. G. Birdwell, F. J. Crowne, R. Vajtai, B. I. Yakobson, Z.
Xia, M. Dubey, P. M. Ajayan and J. Lou, Nat. Commun., 2014, 5,
5246–5254.
12 H. U. Rashid, K. Yu, M. N. Umar, M. N. Anjum, K. Khan, N.
Ahmad and M. T. Jan, Rev. Adv. Mater. Sci., 2015, 40,
235–248.
13 E. Hensen, P. Kooyman, Y. Van der Meer, A. Van der Kraan,
V. De Beer, J. Van Veen and R. Van Santen, J. Catal.,
2001, 199, 224–235.
14 M. S. Rana, J. Ramírez, A. Gutiérrez-Alejandre, J. Ancheyta, L.
Cedeño and S. Maity, J. Catal., 2007, 246, 100–108.
15 P. A. Nikul'shin, Y. V. Eremina, N. N. Tomina and A. A.
Pimerzin, Pet. Chem., 2006, 46, 343–348.
16 N. N. Tomina, P. A. Nikul'shin and A. A. Pimerzin, Pet.
Chem., 2008, 48, 92–99.
17 O. V. Klimov, A. V. Pashigreva, G. A. Bukhtiyarova, S. V.
Budukva, M. A. Fedotov, D. I. Kochubey, Y. A. Chesalov, V. I.
Zaikovskii and A. S. Noskov, Catal. Today, 2010, 150,
196–206.
4. Conclusions
Novel decyltrimethylammonium bromide-dispersed Ni–Mo
sulfide (DTMA-NiMo) was adopted for preparing a highly
efficient NiMoS/γ-Al2O3 HDS catalyst. The Ni and Mo species
were incorporated into the same sulfide with a high Ni/Mo
ratio, which is beneficial for improving the decoration of
MoS2 slabs by Ni to form a high number of NiMoS phases.
The as-synthesized NiMoS phases on the γ-Al2O3 support have
superior active-metal morphology, with a high dispersion of
0.36 and suitable stacking numbers (approximately 3.0),
offering a large number of accessible NiMoS phases. The as-
prepared NiMoS/γ-Al2O3 catalyst exhibits much higher activity
for the HDS of 4,6-DMDBT and FCC diesel than NiMoS/γ-
Al2O3 catalysts prepared by co-impregnation and TPAB-
assisted methods. This method provides a novel and facile
route for the production of sufficient and accessible NiMoS
phases in HDS catalysts for ultra-deep HDS of fuel.
18 J. Liang, M. Wu, P. Wei, J. Zhao, H. Huang, C. Li, Y. Lu, Y.
Liu and C. Liu, J. Catal., 2018, 358, 155–167.
Conflicts of interest
19 J. Liang, M. Wu, J. Wang, P. Wei, B. Sun, Y. Lu, D. Sun,
Y. Liu and C. Liu, Catal. Sci. Technol., 2018, 8,
6330–6345.
There are no conflicts to declare.
20 M. S. Nikulshina, P. Blanchard, A. Mozhaev, C. Lancelot, A.
Griboval-Constant, M. Fournier, E. Payen, O. Mentré, V.
Briois, P. A. Nikulshin and C. Lamonier, Catal. Sci. Technol.,
2018, 8, 5557–5572.
21 A. J. A. Konings, A. Valster, V. H. J. de Beer and R. Prins,
J. Catal., 1982, 76, 466–472.
Acknowledgements
The authors gratefully acknowledge the financial support of
the National Natural Science Foundation of China (Grant No.
21978323).
22 M. Feliz, R. Llusar, S. Uriel, C. Vicent, M. Brorson and K.
Herbst, Polyhedron, 2005, 24, 1212–1220.
References
1 A. Tanimu and K. Alhooshani, Energy Fuels, 2019, 33,
2810–2838.
23 A. Griboval, P. Blanchard, E. Payen, M. Fournier and J. L.
Dubois, Catal. Today, 1998, 45, 277–283.
2 A. Stanislaus, A. Marafi and M. S. Rana, Catal. Today,
2010, 153, 1–68.
3 S. K. Bej, S. K. Maity and U. T. Turaga, Energy Fuels,
2004, 18, 1227–1237.
4 Q. Zhang, J. Zhang, H. Yang, Y. Dong, Y. Liu, L. Yang, D.
Wei, W. Wang, L. Bai and H. Chen, Catal. Sci. Technol.,
2019, 9, 2915–2922.
24 A. Griboval, P. Blanchard, L. Gengembre, E. Payen, M.
Fournier, J. L. Dubois and J. R. Bernard, J. Catal., 1999, 188,
102–110.
25 A. Griboval, P. Blanchard, E. Payen, M. Fournier, J. L. Dubois
and J. R. Bernard, Appl. Catal., A, 2001, 217, 173–183.
26 C. I. Cabello, I. L. Botto and H. J. Thomas, Appl. Catal., A,
2000, 197, 79–86.
5 M. Zuo, X. Huang, J. Li, Q. Chang, Y. Duan, L. Yan, Z. Xiao,
S. Mei, S. Lu and Y. Yao, Catal. Sci. Technol., 2019, 9,
2923–2930.
6 M. Breysse, C. Geantet, P. Afanasiev, J. Blanchard and M.
Vrinat, Catal. Today, 2008, 130, 3–13.
7 A. J. van Dillen, R. J. A. M. Terörde, D. J. Lensveld, J. W.
Geus and K. P. de Jong, J. Catal., 2003, 216, 257–264.
8 J. A. Bergwerff, T. Visser, G. Leliveld, B. D. Rossenaar, K. P.
de Jong and B. M. Weckhuysen, J. Am. Chem. Soc., 2004, 126,
14548–14556.
27 C. I. Cabello, F. M. Cabrerizo, A. Alvarez and H. J. Thomas,
J. Mol. Catal. A: Chem., 2002, 186, 89–100.
28 M. Taniguchi, D. Imamura, H. Ishige, Y. Ishii, T. Murata, M.
Hidai and T. Tatsumi, J. Catal., 1999, 187, 139–150.
29 K. Herbst, M. Monari and M. Brorson, Inorg. Chem.,
2001, 40, 2979–2985.
30 T. K. T. Ninh, L. Massin, D. Laurenti and M. Vrinat, Appl.
Catal., A, 2011, 407, 29–39.
31 P. A. Nikulshin, D. I. Ishutenko, A. A. Mozhaev, K. I.
Maslakov and A. A. Pimerzin, J. Catal., 2014, 312, 152–169.
32 J. Xu, T. Huang and Y. Fan, Appl. Catal., B, 2017, 203,
839–850.
9 J. A. Bergwerff, T. Visser and B. M. Weckhuysen, Catal.
Today, 2008, 130, 117–125.
10 P. Serp, P. Kalck and R. Feurer, Chem. Rev., 2002, 102,
3085–3128.
33 S. Liu, X. Liang, J. Zhang and B. Chen, Catal. Sci. Technol.,
2017, 7, 466–480.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Catalysis Science & Technology
Paper
34 T. Weber, J. C. Muijsers and J. W. Niemantsverdriet, J. Chem.
Phys., 1995, 99, 9194–9200.
50 W. Lai, L. Pang, J. Zheng, J. Li, Z. Wu, X. Yi, W. Fang and L.
Jia, Fuel Process. Technol., 2013, 110, 8–16.
35 A. Müller, E. Diemann, R. Jostes and H. Bögge, Angew.
Chem., Int. Ed. Engl., 1981, 20, 934–955.
51 W. Zhou, M. Liu, Q. Zhang, Q. Wei, S. Ding and Y. Zhou, ACS
Catal., 2017, 7, 7665–7679.
36 R. J. H. Clark and J. R. Walton, J. Chem. Soc., Dalton Trans.,
1987, 1535–1544.
52 T. Shimizu, K. Hiroshima, T. Honma, T. Mochizuki and M.
Yamada, Catal. Today, 1998, 45, 271–276.
37 T. Ito, K. Sawada and T. Yamase, Chem. Lett., 2003, 32,
938–939.
53 Y. Okamoto, K. Ochiai, M. Kawano and T. Kubota, J. Catal.,
2004, 222, 143–151.
38 G. Huang, T. Chen, W. Chen, Z. Wang, K. Chang, L. Ma, F.
Huang, D. Chen and J. Y. Lee, Small, 2013, 9, 3693–3703.
39 E. Diemann and A. Müller, Coord. Chem. Rev., 1973, 10, 79–122.
40 E. Koeniger-Ahlborn, A. Mueller, A. D. Cormier, J. D. Brown
and K. Nakamoto, Inorg. Chem., 1975, 14, 2009–2011.
41 I. Søtofte, Acta Chem. Scand., Ser. A, 1976, 30, 157–162.
42 W.-H. Pan, D. C. Johnston, S. T. McKenna, R. R. Chianelli,
T. R. Halbert, L. L. Hutchings and E. I. Stiefel, Inorg. Chim.
Acta, 1985, 97, L17–L19.
54 W. Han, P. Yuan, Y. Fan, G. Shi, H. Liu, D. Bai and X. Bao,
J. Mater. Chem., 2012, 22, 25340–25353.
55 C. J. H. Jacobsen, E. Törnqvist and H. Topsøe, Catal. Lett.,
1999, 63, 179–183.
56 P. Afanasiev, Appl. Catal., A, 2006, 303, 110–115.
57 B. Liu, Y. Chai, Y. Li, A. Wang, Y. Liu and C. Liu, Appl.
Catal., A, 2014, 471, 70–79.
58 P. Li, X. Liu, C. Zhang, Y. Chen, B. Huang, T. Liu, Z. Jiang
and C. Li, Appl. Catal., A, 2016, 524, 66–76.
43 C. Liu, J. Li, Y. Zhang, S. Chen, J. Zhu and K. Liew, J. Mol.
Catal. A: Chem., 2012, 363-364, 335–342.
44 E. Payen, R. Hubaut, S. Kasztelan, O. Poulet and J. Grimblot,
J. Catal., 1994, 147, 123–132.
45 A. M. Venezia, Catal. Today, 2003, 77, 359–370.
46 A. D. Gandubert, C. Legens, D. Guillaume and E. Payen, Surf.
Interface Anal., 2006, 38, 206–209.
47 P. A. Nikulshin, A. Mozhaev, K. Maslakov, A. Pimerzin and V.
Kogan, Appl. Catal., B, 2014, 158, 161–174.
48 T. Huang, J. Chang, S. He, C. Wen and Y. Fan, Energy Fuels,
2019, 33, 135–147.
49 X. Li, Y. Chai, B. Liu, H. Liu, J. Li, R. Zhao and C. Liu, Ind.
Eng. Chem. Res., 2014, 53, 9665–9673.
59 K. Sakanishi, T. Nagamatsu, I. Mochida and D. D.
Whitehurst, J. Mol. Catal. A: Chem., 2000, 155, 101–109.
60 M. Egorova and R. Prins, J. Catal., 2004, 225, 417–427.
61 H. Wang and R. Prins, J. Catal., 2009, 264, 31–43.
62 X. Wang, Z. Zhao, P. Zheng, Z. Chen, A. Duan, C. Xu, J. Jiao,
H. Zhang, Z. Cao and B. Ge, J. Catal., 2016, 344, 680–691.
63 A. Niquille-Röthlisberger and R. Prins, J. Catal., 2006, 242,
207–216.
64 S. Paria and K. C. Khilar, Adv. Colloid Interface Sci.,
2004, 110, 75–95.
65 Z. Zhang, M. A. Horsch, M. H. Lamm and S. C. Glotzer,
Nano Lett., 2003, 3, 1341–1346.
66 R. B. Grubbs, Polym. Rev., 2007, 47, 197–215.
This journal is © The Royal Society of Chemistry 2020
Catal. Sci. Technol.