Effect of solid-state synthesized alumina properties on the structure and catalytic performance of NiMo catalyst in hydrodesulfurization

Article abstract of DOI:10.1016/j.apcata.2013.09.008

In this paper, a precursor ammonium aluminum carbonate hydroxide (AACH) was synthesized by a low-temperature solid-state reaction, then alumina labeled as Al₂O₃-AC was obtained by the calcination of AACH, and finally, catalyst labeled as NiMo/Al₂O₃-AC supported on Al₂O₃-AC was prepared using the method of incipient wetness impregnation. The properties of both support and catalyst were investigated by a series of characterization methods. The results indicated that Al₂O₃-AC had unique bimodal porous structure, high acid content and five-fold coordinated aluminum (AlO₅) on the surface. Moreover, NiMo/Al₂O₃-AC exhibited uniform dispersion of active components, higher Bronsted acid sites concentration and weaker metal-support interaction. Encouraged by those excellent properties of Al ₂O₃-AC and NiMo/Al₂O₃-AC, the catalyst NiMo/Al₂O₃-AC was evaluated in the hydrodesulfurization (HDS) of dibenzothiophene (DBT), showing high conversion ratio of DBT.

Full text of DOI:10.1016/j.apcata.2013.09.008

Applied Catalysis A: General 468 (2013) 327–333
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Effect of solid-state synthesized alumina properties on the structure
and catalytic performance of NiMo catalyst in hydrodesulfurization
Min Zhang, Tao Yang, Ruiyu Zhao∗, Chenguang Liu
State Key Laboratory of Heavy Oil Processing, Key Laboratory of Catalysis, China National Petroleum Corporation (CNPC), China University of Petroleum
(East China), Qingdao 266580, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 May 2013
Received in revised form 4 September 2013
Accepted 5 September 2013
Available online 14 September 2013
In this paper, a precursor ammonium aluminum carbonate hydroxide (AACH) was synthesized by a low-
temperature solid-state reaction, then alumina labeled as Al2O3-AC was obtained by the calcination of
AACH, and finally, catalyst labeled as NiMo/Al2O3-AC supported on Al2O3-AC was prepared using the
method of incipient wetness impregnation. The properties of both support and catalyst were investi-
gated by a series of characterization methods. The results indicated that Al2O3-AC had unique bimodal
porous structure, high acid content and five-fold coordinated aluminum (AlO5) on the surface. Moreover,
NiMo/Al2O3-AC exhibited uniform dispersion of active components, higher Bro˝ nsted acid sites concen-
tration and weaker metal-support interaction. Encouraged by those excellent properties of Al2O3-AC
and NiMo/Al2O3-AC, the catalyst NiMo/Al2O3-AC was evaluated in the hydrodesulfurization (HDS) of
dibenzothiophene (DBT), showing high conversion ratio of DBT.
Keywords:
AACH
Hydrodesulfurization catalyst
Structural features
Physicochemical properties
Catalytic performance
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
Traditionally, ␥-Al2O3 is produced by thermal dehydration reac-
tion of well-defined boehmite [9,10]. Compared to the synthesis
It is well documented that the support can significantly influ-
ence the properties and performance of hydrodesulfurization
(HDS) catalysts, such as the reducibility, sulfidability, structure and
dispersion of the deposited metal oxides, as well as the morphol-
ogy of the sulfided active phases [1]. Vissers et al. [2] reported that
carbon-supported CoMoS were more weakly bound to the sup-
port than metal oxide supports, this weak interaction increased the
reducibility of sulfide phase, which further led to a higher catalytic
activity. Liu et al. [3] synthesized a novel alumina with bimodal
mesoporous structure, which contributed to high crystal phase dis-
persion and enhanced the catalytic properties for DBT conversion.
In all, the support may accelerate or slow down the sulfidation
process and determine the final dispersion state [2,4]. Therefore,
the choice of proper supports is very important, because they are
closely related to the progress concerning catalytic performance of
HDS catalysts.
of boehmite, the preparation of AACH by a low temperature solid-
state reaction is relatively simple, efficient and cost-saving [11–15].
Our group [16] has reported the detailed properties of alumina
prepared by thermal decomposition of AACH precursor using low-
temperature solid-state reaction, but some important aspects were
not reported, such as the surface acidity of alumina and reducibil-
ity of deposited metal oxides. As is well-known, those aspects
have close relationship to the interactions between active metal
and alumina surface, which directly influence the dispersion and
composition of active phase, as well as the catalytic performance
thereof.
In the present work, we try to get a deep insight into the effect
of solid-state synthesized alumina properties on the structure and
catalytic performance of NiMo catalyst in HDS. In contrast, alu-
mina labeled as Al2O3-SB and catalyst labeled as NiMo/Al2O3-SB
were also prepared using pseudo-boehmite as precursor. The text-
ural and surface properties of alumina were characterized with
scanning electron microscopy, nitrogen adsorption, Fourier trans-
form infrared, temperature-programmed desorption of ammonia
and 27Al MAS NMR. The structural properties of the catalysts were
tested by X-ray powder diffraction, Fourier transform infrared,
H2-temperature programmed reduction, transmission electron
microscopy, and their catalytic performance was tested in the HDS
of DBT.
␥-Al2O3 owing to its low cost, high specific area, good ther-
mal stability, tunable acidity and its interaction with deposited
transition metals, has been widely used as catalyst supports [5–9].
∗
Corresponding author. Tel.: +86 532 86983057; fax: +86 532 86981787.
E-mail addresses: zgsydxzm@163.com (M. Zhang), yang tao9364@126.com
(T. Yang), zhaory@upc.edu.cn (R. Zhao), cgliu@upc.edu.cn (C. Liu).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.09.008

328
M. Zhang et al. / Applied Catalysis A: General 468 (2013) 327–333
2. Experimental
were performed on a Newus Fourier transform infrared spec-
trometer (Nicolet, USA). Prior to measurement, the samples were
degassed in air at 350 ◦C for 3 h, cooled down and then adsorbed
in the saturated pyridine atmosphere at room temperature for 2 h.
After adsorption, the infrared spectrum was recorded with the sam-
ple temperature fixed at 100 ◦C while outgassing.
2.1. Materials and preparation
Pseudo-boehmite (SB) powder was industrial grade reagent pro-
duced by Germany Condea Co., and the other chemicals were of
analytical grade, purchased from National Medicine Group Chemi-
cal Reagent Co., Ltd. The precursor AACH was synthesized according
to the reference [16]. In a typical experiment, 0.16 ml of poly-
glycol (PEG)-400 was mixed with 15.80 g of NH4HCO3 powder in
agate mortar and ground sufficiently. The powder of Al(NO3)3·9H2O
(molar ratio of NH4HCO3/Al = 4/1) was added into mixture and
ground at room temperature for 20 min, then transferred to a
Teflon-lined stainless-steel autoclave and placed in an oven at
100 ◦C. No solvent was used during the process. After aging for
about 12 h, the solid precipitation was filtered off, washed with
deionized water and anhydrous ethanol to remove the impurity
ions, then dried at 100 ◦C in a vacuum oven for 2 h.
Afterwards, 50 g of as-obtained AACH and SB were extruded.
The extrudates were cooled naturally, dried in air at 120 ◦C for 6 h,
and then calcined in muffle furnace at 500 ◦C for 4 h at 2 ◦C/min
heating rate. The as-obtained ␥-Al2O3 samples were labeled as
Al2O3-AC and Al2O3-SB, respectively. Using Al2O3-AC and Al2O3-SB
as supports, the Ni and NiMo supported catalysts were pre-
pared by incipient impregnation with solution of nickel nitrate
(Ni(NO3)2·6H2O) and mixed solution of ammonium heptamolyb-
date ((NH4)6Mo7O24·4H2O) and nickel nitrate (Ni(NO3)2·6H2O),
respectively. Then the impregnated materials were dried in air at
120 ◦C for 6 h, and further calcined in muffle furnace at 500 ◦C for
4 h at 3 ◦C/min heating rate. The as-prepared catalysts were labeled
as Ni/Al2O3-AC, Ni/Al2O3-SB, NiMo/Al2O3-AC and NiMo/Al2O3-SB,
respectively.
2.2.5. 27Al MAS NMR measurements
One-pulse solid-state 27Al MAS NMR spectra were acquired at
a frequency of 104.3 MHz on a Bruker Avance 300 spectrometer
(Switzerland) by using a 5-mm MAS probe with a spinning rate of
10 kHz. Short single pulses (␲/18) were used with a reprocess time
of 1.0 s. The 27Al chemical shifts were referenced to 1 N aqueous
solution of AlCl3.
2.2.6. Temperature-programmed reduction and desorption of
ammonia
Temperature-programmed reduction (TPR) was conducted with
a Quantachrome CHEMBET-3000 instrument. The sample (0.1 g)
was charged in a loop and heated up to 150 ◦C at the rate of
10 ◦C/min, held at 150 ◦C for 1 h and cooled to room temperature in
He flow to remove the adsorbed materials. The treated sample was
heated to 700 ◦C at a rate of 10 ◦C/min in a mixture of 10 vol% H2/Ar
(100 ml/min). Temperature-programmed desorption of ammonia
(NH3-TPD) was also conducted to determine the total acidity of
supports. The sample (0.1 g) was charged in the loop and heated up
to 500 ◦C at the rate of 10 ◦C/min, held at 500 ◦C for 30 min to ensure
complete removal of impurities. Then the sample was cooled down
to 100 ◦C and saturated with ammonia and was flushed by He flow
for 1 h to remove physically adsorbed ammonia, afterwards the
sample was heated to 500 ◦C at a heating rate of 10 ◦C/min in He
flow (30 ml/min). TCD detector was used to monitor the desorbed
ammonia.
2.2. Characterization
2.3. Catalytic performance test
2.2.1. X-ray powder diffraction
Catalytic activity measurements were carried out in a high-
pressure continuous-flow micro-reactor. Generally, 5 ml of the
catalyst extrudates were loaded in the reactor. Prior to reac-
tion, the catalysts were sulfided at 300 ◦C and under 2.0 MPa H2
pressure for 6 h with a liquid stream containing 3 wt% CS2 in
The X-ray powder diffraction (XRD) patterns of the samples
were recorded at room temperature by a Panalytical X’Pert Pro MPD
diffractometer (the Netherlands) using Cu K␣ radiation at a scan
rate (2Â) of 5◦/min. The accelerating voltage and applied current
were 40 kV and 40 mA, respectively.
cyclohexane solution, liquid hour space velocity (LHSV) 2.0 h−1
,
and H2/feed = 300/1. The model compound (2.06 wt% dibenzoth-
iophene dissolved in toluene) was then pumped into reactor. The
reaction was carried out at a temperature 240 ◦C under the condi-
2.2.2. Electron microscopy
The morphology images of as-prepared samples were obtained
by an S-4800 field emission scanning electron microscopy (FE-SEM,
Hitachi, Japan) with an acceleration voltage of 1.5 kV. The high res-
olution transmission electron microscopy (HRTEM) images of the
samples were taken using a JEOL JEM-2100 UHR microscope. The
catalysts were firstly ground and then suspended in alcohol by an
ultrasonic bath, and finally they were placed in a Cu cellulose coated
grille.
tions of H2 pressure 2.0 MPa, H2/feed = 300/1, and LHSV 4.0 h−1
.
The liquid product was analyzed at Varian 3800 gas chromato-
graph equipped with CP-5 capillary column and flame ionization
detector.
3. Results and discussion
2.2.3. Textural analysis
3.1. Acidity of supports
The specific surface area and pore size distributions were
obtained from nitrogen adsorption–desorption isotherms mea-
sured on a Micromeritics TRISTAR 3020 adsorption analyzer (USA)
at −196 ◦C. All as-prepared samples were degassed at 140 ◦C
in vacuum for 6 h prior to adsorption measurements. The sur-
face area was calculated by the Brunauer–Emmett–Teller (BET)
method, and the pore size distributions were determined by the
Barrett–Joyner–Halenda (BJH) method from the desorption branch
of the isotherms.
In order to detect the surface acidity, pyridine adsorp-
tion/desorption experiments have been carried out on Al2O3-AC
and Al2O3-SB supports, and the results are shown in Fig. 1. The IR
spectra display bands ranging from 1445 cm−1 to 1650 cm−1. The
band at 1615 cm−1 is assigned to pyridine coordinately bonded to
moderate Lewis acid sites of the alumina support and the band at
1593 cm−1 is due to pyridine adsorbed on weak Lewis acid sites.
The band at 1578 cm−1 is also attributed to pyridine coordinated
to Lewis acid sites [17–19]. From the results, it can be found that
there are only weak and moderate Lewis acid sites existing on the
surface of both Al2O3-SB and Al2O3-AC. Besides, no bands around
1540 cm−1 are detected, indicating the absence of Bro˝ nsted acid
2.2.4. Fourier transform infrared (FTIR) spectroscopy
The types of surface acid sites on calcined samples were deter-
mined through FTIR pyridine adsorption technique. Measurements

M. Zhang et al. / Applied Catalysis A: General 468 (2013) 327–333
329
Table 1
Acid content distribution of supports and catalysts.
Supports
Weak acid
Moderate acid
Strong acid
Total acidity
(mmol/g)
(<300 ◦C) (mmol/g)
(300–450 ◦C) (mmol/g)
(450–500 ◦C) (mmol/g)
Al2O3-AC
Al2O3-SB
NiMo/Al2O3-AC
NiMo/Al2O3-SB
0.274
0.222
0.281
0.248
0.136
0.128
0.264
0.169
0
0
0
0
0.410
0.350
0.545
0.417
Table 2
Data of textural properties of Al2O3-SB and Al2O3-AC.
1445
Supports
BET specific surface
area (m2/g)
Pore volume
(ml/g)
Average pore
diameter (nm)
1593
Al2O3-AC
Al2O3-SB
335
283
1.3
0.6
12.6
8.0
1578
1615
1487
(a)
(b)
Furthermore, Al2O3-AC has more weak and medium strong acid
sites than Al2O3-SB, which may influence the activity of the catalyst
in the HDS reaction. The correlation between the acidity nature
and HDS activity has been reported in relevant references [24–26].
3.2. Morphology of the precursors
Fig. 2 shows SEM image of AACH and SB. Compared to the
morphology of SB sample with spherical particles packed closely,
the AACH sample mainly consists of uniform nanorods about
30–50 nm. The difference in morphology of precursors is associated
with the crystal orientation and different planes of crystal parti-
cles. Various proportions of different crystallographic surfaces can
lead to diverse acid properties [17], which further demonstrate the
results of NH3-TPD.
1700
1650
1600
1550
1500
-1
1450
1400
Wavenumber (cm
)
Fig. 1. IR spectra of pyridine adsorbed on support (a) Al2O3-AC and (b) Al2O3-SB.
sites on the surface of those two supports. The acidity of alumina is
originated from the unsaturated aluminum coordination and sur-
face hydroxyl groups [20,21]. Liu [22] found that various OH groups
were adjacent to different Lewis acid sites, which could determine
the interactions between pyridine and OH groups. The author also
assigned these different types of Lewis acid sites to several possible
Al3+ coordination configurations.
Table 1 shows the NH3-TPD data of Al2O3-AC and Al2O3-SB
samples. It is generally accepted that acid sites are classified into
three types: weak (Tdes < 300 ◦C), medium (300 ◦C < Tdes < 450 ◦C)
and strong (Tdes > 450 ◦C) [23]. According to the data of acidic prop-
erties in Table 1, no strong acid sites are detected on both Al2O3-SB
and Al2O3-AC, which is in accordance with the FTIR results.
3.3. Textural property of supports
The specific surface area, pore volume, and average pore diam-
eter of Al2O3-AC and Al2O3-SB samples are listed in Table 2.
Apparently, compared to Al2O3-SB, Al2O3-AC has higher specific
surface area and larger pore volume. Fig. 3 shows the nitrogen
adsorption–desorption isotherms and pore size distributions of
Al2O3-AC and Al2O3-SB. The isotherms for those two samples are of
type IV, suggesting the presence of mesopores. Besides, Al2O3-AC
shows higher saturated adsorption amount of nitrogen than that of
Fig. 2. SEM images of (a) AACH and (b) SB.

330
M. Zhang et al. / Applied Catalysis A: General 468 (2013) 327–333
1000
800
600
400
200
0
Al2O3-AC
Al2O3-SB
(b)
(a)
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure (p/p0)
-70
0
70
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Chemical shift (ppm)
Al2O3-AC
Al2O3-SB
Fig. 4. 27Al MAS NMR patterns of (a) Al2O3-AC and (b) Al2O3-SB.
and 0 ppm, respectively. In contrast, except peaks at 63 ppm and
0 ppm, resonance peak at 34 ppm occurs in the 27Al MAS NMR
spectra of Al2O3-AC, which should be assigned to five-fold coor-
dinated aluminum (AlO5). According to the results in the reference
[26], the five-fold coordinated aluminum species are mainly located
at (1 0 0) crystallographic surfaces in ␥-Al2O3, and their content is
related to the exposure proportion of (1 0 0) surfaces.
3.5. Crystalline structure of catalysts
0
20
40
60
80
100
Fig.
5 shows the XRD patterns of oxidic and sulfided
Pore Diameter (nm)
NiMo/Al2O3-AC and NiMo/Al2O3-SB. For oxidic catalysts, very
broad diffraction peaks of ␥-Al2O3 are observed. However, there
is no XRD evidence for the presence of crystalline MoO3 and NiO
phases, suggesting that such phases, if present, have crystal sizes
smaller than 4 nm [3,27]. As to sulfided catalysts, similar peaks
ascribed to ␥-Al2O3 are also found, and very weak signals of MoS2
Fig. 3. The pore size distribution profiles and nitrogen adsorption–desorption
isotherms of Al2O3-AC and Al2O3-SB.
Al2O3-SB, evidencing the larger pore volume of Al2O3-AC. Accord-
ing to the pore size distributions, Al2O3-AC is featured with the
bimodal porous structure, where the smaller pores are narrowly
distributed in 2–8 nm and the larger ones exhibit a broad distri-
bution (10–60 nm), while Al2O3-SB has a fairly uniform pore size
distribution in the range of 3–10 nm. The morphology of Al2O3-
SB with spherical particles can be maintained when it transforms
into alumina during thermal transformation process, which leads
to uniform particles and centered pore size distribution. In con-
trast, Al2O3-AC has the morphology of nanorods, leading to stacking
mode of particles distinct from that of Al2O3-SB. Therefore, there is
a close relationship between pore size distribution of supports and
morphology of precursors. Besides, in the process of calcination,
NH3 and CO2 were released due to the decomposition of Al2O3-
AC, which may contribute to a bimodal pore size distribution of
alumina. During HDS reaction process, the bimodal pore structure
may be beneficial to the reaction performance because the larger
pores can enhance the diffusion of reactants and the smaller one
can provide the active sites [3].
3.4. Surface property of supports
Fig. 4 shows the 27Al MAS NMR spectra of Al2O3-AC and Al2O3-
SB samples. The coordination of surface atoms can be explored by
alumina NMR test. As shown in Fig. 4 for Al2O3-SB, the aluminum
species of ␥-Al2O3 mainly exist in the form of four-fold (AlO4) and
six-fold (AlO6) coordination, with the resonance peak at 63 ppm
Fig. 5. XRD patterns of fresh and sulfided catalysts (a) NiMo/Al2O3-AC; (b)
NiMo/Al2O3-SB; (c) sulfided NiMo/Al2O3-AC; (d) sulfided NiMo/Al2O3-SB; (e) MoS2
(
—MoS2; —␥-Al2O3).

M. Zhang et al. / Applied Catalysis A: General 468 (2013) 327–333
331
Table 3
Product distributions of HDS reaction of DBT.
Catalysts
CHB (%m)
BP (%m)
DBT (%m)
DBT conversiona (%m)
CHB selectivityb
NiMo/Al2O3-SB
NiMo/Al2O3-AC
7.11
11.19
59.86
68.39
33.03
20.42
66.97
79.58
0.12
0.16
CHB: cyclohexylbenzene; BP: biphenyl; DBT: dibenzothiophene.
a
DBT conversion % = wt(BP+CHB)/total amount of DBT.
b
CHB selectivity = wtCHB/wtBP
.
(a)
Ni
1446
381 ºC
358 ºC
1490
1593
570 ºC
1577
1540
(a)
(b)
1606
371 ºC
345 ºC
1700
1650
1600
1550
1500
1450
1400
100
200
300
400
500
600
700
-1
Wavenumber (cm
)
Temperature (cm-1
)
Fig. 6. IR spectra of pyridine adsorbed on (a) NiMo/Al2O3-AC and (b) NiMo/Al2O3-SB
catalysts.
(b)
NiMo
diffraction peaks at 2Â = 33.1◦ and 58.5◦ are identified only in sul-
fided NiMo/Al2O3-AC, indicating a little difference in the MoS2
particle size over these two catalysts. Therefore, MoS2 particles are
almost uniformly dispersed on the surface of both Al2O3-AC and
Al2O3-SB, and the differences in MoS2 particle size may affect the
catalytic performance [28,29].
442 ºC
499 ºC
373 ºC
359 ºC
(b)
3.6. Acid property of catalysts
Fig. 6 shows the IR spectra of pyridine adsorption of NiMo/Al2O3
(a)
catalysts. It can be seen from Fig. 6 that NiMo/Al2O3-AC and
NiMo/Al2O3-SB show similar bands in the region 1700–1400 cm−1
.
The band at 1540 cm−1, 1490 cm−1 and 1450 cm−1 are associated
with pyridine adsorbed on Brønsted acid sites, pyridine adsorbed
on both Brønsted and Lewis acid sites, and pyridine adsorbed
on Lewis acid sites [30], respectively. Besides, the band inten-
sity of NiMo/Al2O3-AC at 1450 cm−1 is almost the same as that of
NiMo/Al2O3-SB while the bands of NiMo/Al2O3-AC at 1540 cm−1
and 1490 cm−1 are more intensive than those of NiMo/Al2O3-SB,
indicating higher Bro˝ nsted acid sites concentration of NiMo/Al2O3-
AC [31].
The NH3-TPD data of the oxidic catalysts are also listed in Table 1.
The total acidity is increased by the impregnation of molybdenum
and nickel ions. Furthermore, NiMo/Al2O3-AC contains more weak
and medium strong acid sites than NiMo/Al2O3-SB, which may be
related to the content of Bro˝ nsted acid sites [32]. Duan et al. [33]
reported that HDS activity could be enhanced by higher Bro˝ nsted
acid sites concentration, which is consistent with the results of our
following HDS test.
100
200
300
400
500
600
700
Temperature (ºC)
Fig. 7. TPR patterns of Ni and NiMo supported on (a) Al2O3-AC and (b) Al2O3-SB.
three reduction peaks in the 200–700 ◦C region. The first peak cen-
tered at 345 ◦C may be related to the reduction of NiO to Ni0, and
the second reduction peak at 371 ◦C may be due to the interac-
tion of Ni2+ species with alumina [34]. According to the reference
[35], the third peak at 570 ◦C can be attributed to the reduction
of nickel–aluminum species. In contrast, for Ni/Al2O3-SB, the first
and second reduction peaks are about 10 ◦C higher than those of
Ni/Al2O3-AC, indicating that Ni2+ species have relatively strong
interaction with alumina and are more difficult to be reduced than
Ni2+ supported on Al2O3-AC. Furthermore, the third reduction peak
of Ni/Al2O3-AC at 570 ◦C is less intensive than that of Ni/Al2O3-SB,
evidencing that the metal dispersion of Ni/Al2O3-AC is superior to
that of Ni/Al2O3-SB. Therefore, Al2O3-AC may promote the forma-
tion of NiMoS active phase, which would lead to the enhancement
of catalytic activity over NiMo catalysts.
3.7. Temperature-programmed reduction (TPR) for catalysts
Fig. 7 represents the TPR patterns of Ni and NiMo supported on
Al2O3-AC and Al2O3-SB. For Ni/Al2O3-AC, the TPR pattern exhibits

332
M. Zhang et al. / Applied Catalysis A: General 468 (2013) 327–333
Fig. 8. HRTEM micrographs of sulfided (a) NiMo/Al2O3-AC and (b) NiMo/Al2O3-SB catalysts.
It appears three reduction peaks in the TPR profile recorded on
the NiMo/Al2O3-AC. The first peak at about 359 ◦C and the sec-
ond peak at about 373 ◦C are related to the reduction of Ni species
and partial reduction of octahedrally coordinated polymeric Mo
species. The third peak at about 499 ◦C is attributed to the fur-
ther reduction of polymolybdate, which exerts stronger interaction
with the support. In the case of NiMo/Al2O3-SB, the reduction
peak of Ni and Mo oxides occurs at about 442 ◦C and further
reduction of polymolybdate is not observed until 700 ◦C, indicating
that NiMo/Al2O3-SB exhibits stronger metal-support interaction.
It is generally known that the catalyst pre-sulfurization operation
would be difficult if Ni and Mo species have strong interaction with
alumina support, which may lead to poor catalytic performance
of HDS. Therefore, the TPR results reveal that the active species of
NiMo/Al2O3-AC are more easily sulfided than those of NiMo/Al2O3-
SB, indicating that Al2O3-AC favors the sulfurization of active metal
[36,37], which can be well verified in the following HDS activity test.
60
50
40
30
20
10
0
NiMo/Al2O3-SB
NiMo/Al2O3-AC
4-5
>6
2-3
3-4
0-2
3.8. HDS activity
Length (nm)
Product distributions of the HDS reaction of DBT over those two
catalysts are listed in Table 3. As seen from Table 3, the conversion of
DBT for NiMo/Al2O3-AC is 12.61%, higher than that for NiMo/Al2O3-
SB. Besides, the CHB/BP values of those two catalysts are less than
1, indicating that the direct hydrogenolysis (DDS) reaction occurs
faster than the hydrogenation (HYD) reaction.
80
70
60
50
40
30
20
10
0
NiMo/Al2O3-SB
NiMo/Al2O3-AC
Fig. 8 shows the HRTEM images of sulfided NiMo/Al2O3-AC and
NiMo/Al2O3-SB. It can be seen that the dispersion of visible MoS2
particles on the surface of Al2O3-AC is better than that of Al2O3-SB.
Fig. 9 shows the length and layer stacking distributions of MoS2
crystallites in sulfided NiMo catalysts supported on Al2O3-AC and
Al2O3-SB. In Fig. 9, the number of layer stacking of MoS2 particles
of NiMo/Al2O3-AC is in the range of 1–3 with two layer stacking
accounting for 50%, while NiMo/Al2O3-SB has single layer which
occupies 75%. Besides, the length of MoS2 particles of NiMo/Al2O3-
SB is longer than that of NiMo/Al2O3-AC. According to relevant
references [38–40], there are two types of Ni–Mo–S active phases:
single layer (type I) and multilayer (type II). Specifically, type I
has stronger interaction with alumina support while type II has
weaker interaction with alumina support. Furthermore, the desul-
furization activity of type II almost doubles that of type I and the
1
2
3
4
Number of layers
Fig. 9. Length and layer stacking distributions of MoS2 crystallites in sulfided NiMo
catalysts supported on Al2O3-AC and Al2O3-SB.

M. Zhang et al. / Applied Catalysis A: General 468 (2013) 327–333
333
type of active phase has close relationship with the surface prop-
erties of alumina support. Combined with Fig. 9, it reveals that the
active phase of NiMo/Al2O3-AC mainly exists as type II while the
active phase of NiMo/Al2O3-SB mainly exists as type I. Therefore,
the results of HRTEM and MoS2 stacking layers distribution are in
good agreement with TPR results and the HDS performance of NiMo
catalysts.
[5] L. Ji, J. Lin, K.L. Tan, H.C. Zeng, Chem. Mater. 12 (2000) 931–939.
[6] K. Hellgardt, D. Chadwick, Ind. Eng. Chem. Res. 37 (1998) 405–411.
[7] Z. Zhang, T.J. Pinnavaia, J. Am. Chem. Soc. 124 (2002) 12294–12301.
[8] W.Q. Cai, Y.Z. Hu, J. Chen, G.X. Zhang, T. Xia, CrystEngComm 14 (2012) 972–977.
[9] Z.R. Zhang, T.J. Pinnavaia, Langmuir 26 (2010) 10063–10067.
[10] G.C. Li, Y.Q. Liu, D. Liu, L.H. Liu, C.G. Liu, Mater. Res. Bull. 45 (2010) 1487–1491.
[11] S.J. Wilson, J. Solid State Chem. 30 (1979) 247–255.
[12] G.C. Li, Y.Q. Liu, L.L. Guan, X.F. Hu, C.G. Liu, Mater. Res. Bull. 47 (2012)
1073–1079.
[13] Y.T.O. S.W. Kim, D.C. Shin, Colloids Surf., A 313–314 (2008) 415–418.
[14] W.L. Wang, J.Q. Bi, Y.Q. Qi, Z. Zhang, Z. Xing, H.L. Zhu, D. Guo, J.J. Xu, Y.J. Bai,
Powder Technol. 201 (2010) 273–276.
4. Conclusion
[15] J.G. Li, X.D. Sun, M. Zhang, X.D. Li, H.Q. Ru, J. Inorg. Mater. 13 (1998) 803–807.
[16] X.F. Hu, Y.Q. Liu, Z. Tang, G.C. Li, R.Y. Zhao, C.G. Liu, Mater. Res. Bull. 47 (2012)
4271–4277.
[17] G.C. Li, Y.Q. Liu, Z. Tang, G.C. Liu, Appl. Catal., A: Gen. 437–438 (2012) 79–89.
[18] E.P. Parry, J. Catal. 2 (1963) 371–379.
In conclusion, the precursor AACH consisted of uniform
nanorods was successfully prepared by a solid-state reaction, then
Al2O3-AC was obtained after the calcination of AACH. Compared
to Al2O3-SB prepared by the calcination of commercial SB, Al2O3-
AC had higher specific surface area, larger pore volume, unique
bimodal porous structure, and more weak and moderate Lewis acid
sites. Interestingly, Al2O3-AC had five-fold coordinated aluminum
(AlO5) on the surface while Al2O3-SB only had four-fold (AlO4)
and six-fold coordinated aluminum (AlO6). Furthermore, XRD pat-
terns showed that active components of NiMo/Al2O3-AC supported
on Al2O3-AC were uniformly dispersed. Remarkably, NiMo/Al2O3-
AC exhibited more Bro˝ nsted acid sites and weaker metal-support
interaction than NiMo/Al2O3-SB supported on Al2O3-SB. Therefore,
catalytic evaluation results showed that NiMo/Al2O3-AC had higher
conversion ratio of DBT for HDS reaction than NiMo/Al2O3-SB.
[19] G. Busca, Catal. Today 41 (1998) 191–206.
[20] X.S. Liu, R.E. Truitt, J. Am. Chem. Soc. 119 (1997) 9856–9860.
[21] C. Morterra, G. Magnacca, Catal. Today 27 (1996) 497–532.
[22] X.S. Liu, J. Phys. Chem. C 112 (2008) 5066–5073.
[23] J.R. Grzechowiak, J. Rynkowski, I. Wereszczako-Zielinska, Catal. Today 65
(2001) 225–231.
[24] M. Lewandowski, Z. Sarbak, Fuel 79 (2000) 487–495.
[25] U. Usman, M. Takaki, T. Kubota, Y. Okamoto, Appl. Catal., A: Gen. 286 (2005)
148–154.
[26] M. Digne, P. Sautet, P. Raybaud, P. Euzen, H. Toulhoat, J. Catal. 226 (2004)
54–68.
[27] D. Solís, A.L. Agudo, J. Ramírez, T. Klimova, Catal. Today 116 (2006) 469–477.
[28] B. Liu, Y.M. Chai, Y.J. Wang, T.T. Zhang, Y.Q. Liu, G.C. Liu, Appl. Catal., A: Gen.
388 (2012) 248–255.
[29] G. Berhault, M.P.D.I. Rosa, A. Mehta, M.J. Yácaman, R.R. Chianelli, Appl. Catal.,
A: Gen. 345 (2008) 80–88.
[30] M.E. Cervantes-Gaxiola, M. Arroyo-Albiter, A. Pérez-Larios, P.B. Balbuena, J.
Espino-Valencia, Fuel 113 (2013) 733–743.
[31] H. Kim, J.J. Lee, S.H. Moon, Appl. Catal., B: Environ. 44 (2003) 287–299.
[32] D. Ferdous, A.K. Dalai, J. Adjaye, Appl. Catal., A: Gen. 260 (2004) 137–151.
[33] A.J. Duan, G.F. Wan, Z. Zhao, C.M. Xu, Y.Y. Zheng, Y. Zhang, T. Dou, X.J. Bao, K.
Chung, Catal. Today 119 (2007) 13–18.
Acknowledgment
Financial support from “the Fundamental Research Funds for the
Central Universities” 13CX06052A is gratefully acknowledged.
[34] S.B. Ren, J.S. Qiu, C.Y. Wan, B.L. Xu, Y.N. Fan, Y. Chen, Chin. J. Catal. 28 (2007)
651–656.
References
[35] L.F. Zhang, J.F. Lin, Y. Chen, J. Chem. Soc., Faraday Trans. 88 (1992) 497–502.
[36] A. Dandekar, R.T.K. Baker, M.A. Vannice, Carbon 36 (1998) 1821–1831.
[37] H. Shang, C. Liu, Y. Xu, J. Qiu, F. Wei, Fuel Process. Technol. 88 (2007) 117–123.
[38] M. Daage, R.R. Chianelli, J. Catal. 149 (1994) 414–427.
[39] Y.M. Chai, G.J. An, Y.Q. Liu, C.G. Liu, Prog. Chem. 19 (2007) 234–242.
[40] J. Liu, Y.S. Zhao, Y. Liu, R.Y. Zhao, C.G. Liu, Acta Petrolei Sin. 26 (2010)
517–524.
[1] Y. Okamoto, M. Breysse, G.M. Dhar, C. Song, Catal. Today 86 (2003) 1–4.
[2] J.P.R. Vissers, B. Scheffer, V.H.J. de Beer, J.A. Moulijin, R. Prins, J. Catal. 105 (1987)
277–284.
[3] X.M. Liu, X. Li, Z.F. Yan, Appl. Catal., B: Environ. 121–122 (2012) 50–56.
[4] M. Breysse, P. Afanasiev, C. Geantet, M. Vrinat, Catal. Today 86 (2003) 5–16.