Catalysis Today 220–222 (2014) 66–77

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Supported on alumina Co-Mo hydrotreating catalysts: Dependence of

catalytic and strength characteristics on the initial AlOOH particle
morphology
O.V. Klimov a,∗ , K.A. Leonova a,∗∗ , G.I. Koryakina a , E.Yu. Gerasimov a , I.P. Prosvirin a ,
S.V. Cherepanova a , S.V. Budukva a , V.Yu. Pereyma a , P.P. Dik a ,
O.A. Parakhin b , A.S. Noskov a
a
Boreskov Institute of Catalysis SB RAS, Novosibirsk, Russia
b
JSC “NPK Sintez”, Barnaul, Russia

a r t i c l e i n f o a b s t r a c t

Article history: The influence of different initial AlOOH boehmites on the mechanical strength, textural characteristics,
Received 19 April 2013 and activity of diesel fuel hydrotreating catalysts has been studied. The morphology of the initial AlOOH
Received in revised form 30 August 2013 particles was shown to have no significant influence on the composition and structure of the Co-Mo-S
Accepted 2 September 2013
phase, although it does fully define the mechanical strength and textural characteristics of the catalysts.
Available online 25 September 2013
An optimal combination of high mechanical strength and catalytic activity, which depends on a texture
of the catalyst, is attained using initial boehmite with a needle-shaped particle morphology and a particle
Keywords:
size of 150 nm × 8 nm × 8 nm.
Particle morphology
Boehmite © 2013 Elsevier B.V. All rights reserved.
Hydrotreating catalysts
Textural characteristics
Bulk crushing strength

1. Introduction approach to the synthesis of the supported high-activity sulphided

Currently, most diesel fuels with ultra-low sulphur content
(ULSD) are produced in industry using supported Co(Ni)-Mo/Al2 O3
catalysts [1]. Modern catalysts generally contain supported met-
als in the form of highly active and fully sulphided bimetallic
compounds with little interaction with the support. This active
component is often called Co(Ni)-Mo-S phase of Type II [1–5]. The
selective synthesis of Co(Ni)-Mo-S phase of Type II has been suf-
ficiently developed in industry, as confirmed by the presence of
this phase in different industrial catalysts [3]. The basic steps of
the preparation of catalysts that are suitable for industrial appli-
cation and preferentially contain Co(Ni)-Mo-S phase of Type II are
the selection of supports modified with compounds that minimise
chemical interactions between the support and active metals, the
addition of chelating organic compounds into the impregnating
solutions, the drying of catalysts at low temperatures, and the appli-
cation of specific sulphidation techniques [1–8]. Thus, a general
∗ Corresponding author. Tel.: +7 9137238699; fax: +7 3833308056.
∗∗ Corresponding author. Tel.: +7 9231824347; fax: +7 3833308056.
E-mail addresses: klm@catalysis.ru (O.V. Klimov),
lakmallow@catalysis.ru (K.A. Leonova).
compound is defined and can be modified to achieve particular
goals, such as the preparation of hydrotreating catalysts fora spe-
cific type of fuel.
However, the situation for catalyst supports is more com-
plicated. The industry advanced a set of new requirements for
catalysts that are fully defined by the characteristics of the supports
used. Ideally, supports provide the optimal textural characteristics
for the catalyst in combination with high mechanical strength at
minimal granule cross-section. The main textural characteristics
are average surface area, pore volume, and pore size distribution.
High-activity catalysts for the production of ULSD generally contain
10–15% of Mo and 3–5% of Co(Ni), e.g., [8,9]. The optimal surface
metal concentration is 2–3 Co(Ni) atoms and 5–8 Mo atoms per
nm2 [3],which is more than a monolayer filling of the surface by
Co(Ni)-Mo-S or molybdenum oxide [10]. Accordingly, the average
surface area of the catalysts is 150–200 m2 /g. The average surface
area of a catalyst depends on the average pore diameter, which is
in turn defined by the sulphur-containing molecules of the feed.
To achieve maximum conversion of the largest and least reactive
sulphur compounds in diesel fuels, namely, homologs of dibenzoth-
iophene with several alkyl groups [1,11,12], they must be able to
access the active component particles. Thus, a hydrotreating cata-
lyst of middle distillates should contain pores with the diameters of

0920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.09.001
 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77
70–180 Å [13] and major part of these pores should have diameters
in the range of 70–130 Å [14].
The pore volume of a catalyst should be not less than 0.4 cm3 /g to
provide the active component with access to the feed and to localise
the inevitably formed carbon deposits. Catalysts with a pore vol-
ume of more than 0.6 cm3 /g should not be used because this pore
volume decreases the active component concentration per unit vol-
ume of the reactor. Thus, an average surface area of 150–200 m2 /g,
a pore volume of 0.4–0.6 cm3 /g, and an average pore diameter of
10.0 ± 3.0 nm are optimal for catalysts used in the deep hydrotreat-
ment of diesel fuels. We arrived at these conclusions based on a
large amount of modern industrial catalysts. Similar conclusions
were made by Eijsbouts in the review [14].
The supporting 10–15% Mo and 3–5% Co(Ni) using different
methods results in the decrease in an average surface area of
50 ± 10 m2 /g, a pore volume of 0.20 ± 0.05 cm3 /g and an average
pore diameter of 0.5–1.0 nm, e.g., [6,9,10,14,15]. To compensate for
the deterioration of textural characteristics due to the deposition of
Mo, Co, or Ni compounds on the surface, an initial support should
have an average surface area of 200–250 m2 /g, an average pore
diameter of no less than 9.0 nm, and a pore volume of 0.6–0.8 cm3 /g.
Industrial hydrotreating process of middle distillates is usually
performed in trickle-bed reactors with parallel top-down supply
of feedstock and hydrogen [16,17]. Portion of diesel fuel passes
through the reactor as a liquid and forms channels in a catalyst bed.
Large bed height could lead to non-uniform liquid flow along gran-
ules and portion of granule surface will remain dry [17]. To provide
optimal mass transfer through the catalyst bed and diffusion of the
feed along the catalyst granule, it is necessary to use granules with
minimal volume-to-surface ratio and minimal granule diameter.
However, granules could not have sizes less than 1 mm because it
leads to significant pressure drop in the reactor [17]. Consequently,
catalyst granules with trilobe and quadrilobe cross-section and the
diameters of 1.0–1.5 mm are the most effective for deep hydrotreat-
ment, while cylindrical granules with the diameters of more than
2 mm were used previously [14,16–21].
To provide an optimal diffusion of the feed along the catalyst
granule, granules with a minimal volume-to-surface ratio and min-
imal granule diameter should be used. In addition, granules in
the form of trilobe or quadrilobe extrudates with a cross-section
diameter less than 1.5 mm are the most effective in hydrotreat-
ing [14,16–19]. However, the smaller the granule, the lower its
mechanical strength is. In recent years, the bed volume of a cat-
alyst has had to be increased by more than three-fold [12] because
of the conversion to the production of ULSD while preserving the
general productivity of industrial units. Accordingly, requirements
for the mechanical strength of catalysts have increased. Previ-
ously, ASTM D4179-88a (Standard Test Method for Single Pellet
Crush Strength of Formed Catalyst Spheres) and ASTM D6175-
98 (Standard Test Method for Radial Crush Strength of Extruded
Catalyst Particles) had been widely used to define the mechani-
cal strength of hydrotreating catalysts. These methods are based
on strength measurements of single, well-selected granules with
no defects and deformations. Consequently, these methods do not
always provide accurate values for real catalysts, especially cata-
lysts in the form of extrudates with complex cross-sections and
small sizes, which comprise nearly all modern catalysts for ULSD
production.
For the catalysts, which are loaded into the reactor as a dense
bed, mechanical strength is defined using the amount of dust
formed. Dust formation is caused by the destruction of granules
during mechanical impact on the bed. The more dust formed, the
greater the pressure drop along the reactor is [20,22]. Consequently,
for hydrotreating catalysts, methods based on the determination of
a mechanical strength for a large volume of a catalyst are more
appropriate than those for a single granule. Some methods for
67
determining bulk crushing strength and their advantages over tech-
niques for strength measurements of single granules are described
in [22,23]. Currently, industrial plants require the bulk crushing
strength of hydrotreating catalysts to be not less than 1.5 MPa
according to the Shell SMS 1471 or ASTM 7084-4 methods. The
strength of a catalyst depends on many factors, such as the impreg-
nation, drying [24], calcination [25], and sulphidation conditions
[26]. However, the strength of initial supports has the greatest
influence on the final catalyst strength.
Thus, the strength and textural characteristics of catalysts,
which have significant influence on catalytic properties, are
largely defined by the strength and texture of the initial sup-
ports, namely, extrudates of alumina oxide. Different results can
be obtained using the same initial pseudoboehmite subjected to
various natures and concentrations of peptising agents, paste wet-
ness, mixture conditions, granulations, and drying and calcination
conditions [14,24–27]. To produce granulated ␥-Al2 O3 , initial pseu-
doboehmite is often peptised using nitric acid in the mixer with
Z-folded beaters and extruded using a spinneret from polymeric
material at 1.0–3.0 MPa. Next, wet extrudates are dried at a temper-
ature slightly higher than 100 ◦ C and calcined at 550 ◦ C. An increase
in the HNO3 /AlOOH ratio leads to an increase in the mechanical
strength and decrease in the volume and average pore diameter
[14,28,29], and a typical HNO3 /Al2 O3 mole ratio in paste is in the
range of 0.03–0.05 [29]. Thus, the conditions used in the industrial
preparation of ␥-Al2 O3 extrudates are similar, with the texture and
strength of initial supports being generally defined by the prop-
erties of the initial pseudoboehmite. If the chemical and phase
composition of the initial AlOOH powders used for granulation
are similar, the properties of ␥-Al2 O3 will mostly depend on the
morphology of the initial boehmite or pseudoboehmite crystals.
In this study, the influence of the crystalline particle morphol-
ogy of the industrial AlOOH powders on the strength, texture,
and catalytic properties of Co-Mo/Al2 O3 catalysts for the deep
hydrotreatment of diesel fuels was studied.
2. Experimental
2.1. Initial powders and preparation of supports
The industrial AlOOH powders were Pural SB, TH-60, TH-80,
TM-50, and TM-70, produced by Sasol, Germany, synthesised
using alcoholate technology; boehmite produced by “Chimtek-
Engineering”, Chelyabinsk, Russia, synthesised using precipitation
technology; and two boehmites produced by ISCZC, Ishimbay, Rus-
sia, synthesised using the thermochemical activation of gibbsite
followed by plasticisation in autoclave with nitric acid and spray
drying [30] under different plasticisation conditions and designated
as ISCZC-1 and ISCZC-2. The main characteristics of the powders are
shown in Table 1.
All the supports were synthesised using the same method.
Pastes with a mass ratio of AlOOH/H2 O/HNO3 = 1/0.7–0.8/0.014
(HNO3 /Al2 O3 mole ratio = 0.03) were prepared in a Z blade sigma
mixer for 1 h. These ratios were specifically chosen based on a large
amount of experimental data. Therefore, they were well suited for
each powder. These ratios provide production of granules with
specified shape and the highest mechanical strength. Increase of
water amount results in a drastic decrease of mechanical strength
and adhesion of granules. In addition, excess water removes dur-
ing extrusion. In the case of deviation from specified parameters
extrudable pastes could not be obtained.
The paste obtained was extruded using a fluoroplastic spinneret
with trilobe holes on the VINCI extruder at P = 3.5–4.0 MPa and with
the plunger moving at 1.2 mm/s. Extrudates were dried for 4 h at
120 ◦ C in air flow, then heated to 550 ◦ C for 2 h and calcined at
68 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77

Table 1
Characteristics of initial AlOOH powders.

Sample Pural SB TH-60 TH-80 TM-50 TM-70 Chimtek ISCZC-1 ISCZC-2

Formula AlOOH·xH2 O
Content Al2 O3 (%) 72.0 72.5 73.0 72.5 73.0 72.0 72.0 72.5
Crystal structure Boehmite
Admixtures of elements (%) 0.001 Na 0.001 Na 0.001 Na 0.001 Na 0.001 Na 0.07Ca 0.05Ca 0.03Ca
0.01Fe 0.005Fe 0.02Fe
0.005Na 0.04Na 0.02Na
0.002Mg 0.01Mg 0.01Mg
Average crystal particle size (XRD) (nm) 5 8 10 7 8 10 10 17
Powder particle size (d50) (␮m) 58 31 32 45 45 56 57 59

550 ◦ C for 4 h. The dimensions of the samples decreased to different
degrees; however, in all cases, the cross-section diameter was in
the range of 1.3–1.4 mm. All samples were crushed to an extrudate
size of 3–6 mm. Then these samples were used for the preparation
of the catalysts and measurements of the mechanical strength.
2.2. Preparation and HDS testing of catalysts
The catalysts were prepared using vacuum impregnation.
Alumina extrudates were impregnated with a solution con-
taining the bimetallic complex, which was synthesised from
Co(CH3 COO)2 ·4H2 O and (NH4 )4 [Mo4 O11 (C6 H5 O7 )2 ] according to
the technique described in [31]. The impregnation and drying tech-
niques are described in [32]. In all instances, the concentrations of
impregnating solutions were chosen to obtain 11.0% of Mo and 3.5%
of Co in the final catalysts. The catalysts were calcined at 550 ◦ C
for 4 h before the elemental analysis and measurements of their
textural characteristics. In all other cases, catalysts dried at 120 ◦ C
were used. According to the elemental analysis data obtained using
OPTIMA 4300 PERKIN ELMER equipment, all the catalysts contained
11.5 ± 0.5% Mo and 3.3 ± 0.2% Co.
The catalysts were crushed to a particle size of 0.25–0.50 mm
before sulphiding and hydrotreatment testing. Catalyst sulphiding
was performed in a H2 S (99% purity, admixtures – 1% of H2 ) flow
of 500 h−1 at atmospheric pressure. The temperature was ramped
from 20 to 200 ◦ C over 1 h, maintained for 2 h at 200 ◦ C, ramped to
400 ◦ C over 1 h, and maintained for 2 h.Then, H2 S flow was cut off
and a catalyst was cooled up to room temperature in nitrogen flow
(1000 h−1 ).
After sulphidation, the catalysts were tested in HDS of SRGO
(sulphur concentration, 10,800 wppm, density at 20 ◦ C, 0.855 g/ml;
FBP, 360 ◦ C; nitrogen concentration–138 wppm). In all instances,
the sulphided catalysts were studied in 4-g batches. The catalyst
batch was loaded in fixed bed stainless steel reactor with internal
diameter of 10 mm. Height of the catalyst bed was approximately
50 mm in each case. 5 g of SiC (fraction of 0.25–0.63 mm) was placed
above and below the catalyst bed. The catalyst batch was in the
isothermal zone of the reactor. Whereas the same catalyst frac-
tion with the particle size less than 0.50 cm was used in each case,
diffusion limitations were neglected. The experiments were per-
formed in the following conditions: 3.8 MPa, T = 340 ◦ C, a liquid
hourly space velocity of 2.0 h−1 , and a H2 /feed volume ratio of
300. Under such conditions, the weight liquid product output was
94–96%, mainly due to the efficiency of the apparatus separator. The
sulphur content in the liquid products was measured by a HORIBA
SLFA 2100 X-ray fluorescence analyser (Japan). Samples were not
taken in the first 6 h to allow the catalysts to reach a steady state.
During the next 6 h, samples were analysed and the obtained results
were averaged. As a rule, the differences in the measured sulphur
content did not exceed 5 ppm.
2.3. Characterisation techniques
X-ray diffraction (XRD) was carried out using an HCG 4-C diffrac-
tometer (Freiberger Prazisionsmechanik, Germany) equipped with
a copper anode (Cu K␣) with a wavelength of 1.54 Å.
The textural properties of the catalyst and support were deter-
mined by nitrogen physisorption using an ASAP 2400 (USA)
instrument. Prior to analysis, samples were subjected to a N2 flow
at 200 ◦ C for 2 h. The BET surface areas were calculated from the
nitrogen uptakes at relative pressures ranging from 0.05 to 0.30.
The total pore volume was derived from the amount of nitro-
gen adsorbed at a relative pressure close to unity (in practice,
P/P0 = 0.995) by assuming that all accessible pores had been filled
with condensed nitrogen in the normal liquid state. The pore size
distribution was calculated using the BJH method for the desorption
branch of the isotherm. Equipment and measurement procedure
provided accurate pore size definition with diameters not less than
3 nm. Therefore, micropores present in Table 2 are the pores with
diameters <3 nm, mesopores – 3–50 nm, macropores >50 nm, that
is consistent with generally accepted classifications [33]

Table 2
Pore size distribution for catalysts and supports.

Properties PURAL SB TH-60 TH-80 TM-50 TM-70 Chimtek ISCZC-1 ISCZC-2

Support Catalyst Support Catalyst Support Catalyst Support Catalyst Support Catalyst Support Catalyst Support Catalyst Support Catalyst

Pore availability (%)

Micropores <3 nm 1.2 1.5 0.4 0.5 0.3 0.3 0.7 0.9 0.6 0.7 2.3 2.4 1.1 1.4 1.2 1.3
Mesopores 3–50 nm 97.2 97.3 99.1 97.8 98.5 93.7 96.2 92.7 94.6 97.1 94.6 94.9 93.2 94.3 89.3 88.5
Macropores >50 nm 1.6 1.2 0.5 1.7 1.2 6.0 3.1 6.4 4.8 2.2 3.1 2.7 5.7 4.3 9.5 10.2

Pore size distribution (vol.%)

<3 nm 1.2 1.5 0.4 0.5 0.3 0.3 0.7 0.9 0.6 0.7 2.3 2.4 1.1 1.4 1.2 1.3
3–5 nm 18.0 23.1 4.6 4.8 3.3 3.8 9.1 10.5 6.2 7.3 24.3 37.1 11.1 12.0 10.3 12.2
5–10 nm 76.1 71.0 28.4 36.2 20.4 25.3 54.7 59.7 36.1 45.8 28.8 23.2 32.7 33.4 26.0 24.4
10–25 nm 2.7 2.8 64.9 55.8 73.5 64.4 32.1 22.2 51.3 42.7 37.6 30.6 47.2 45.0 35.4 36.4
25–50 nm 0.4 0.4 1.2 1.0 1.3 0.2 0.3 0.3 1.1 1.3 3.9 4.0 2.2 3.9 17.6 15.5
>50 nm 1.6 1.2 0.5 1.7 1.2 6.0 3.1 6.4 4.8 2.2 3.1 2.7 5.7 4.3 9.5 10.2
7–13 nm 26.5 22.3 49.9 54.2 60.1 51.7 66.1 58.7 58.1 59.5 27.1 23.1 30.7 33.4 22.7 21.7
 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77
HRTEM images were obtained on a JEM-2010 electron micro-
scope (JEOL, Japan) with a lattice-fringe resolution of 0.14 nm at
an accelerating voltage of 200 kV. The high-resolution images of
the periodic structures were analysed by the Fourier method. Sam-
ples for HRTEM examination were prepared on a perforated carbon
film mounted on a copper grid. The particle size of the initial AlOOH
powders and the slab length of the sulphided active components
were defined using the average data for at least 500 particles.
X-ray photoelectron spectra (XPS) were recorded using a
SPECS spectrometer (Germany) with a PHOIBOS-150 hemispherical
energy analyser and AlK␣ irradiation (h␯ = 1486.6 eV, 200 W). The
binding energy scale was preliminarily calibrated using the peak
positions of the Au4f7/2 (84.0 eV) and Cu2p3/2 (932.67 eV) core lev-
els. The samples were supported using conductive scotch tape. The
internal reference method was used for the correct calibration of
the photoelectron peaks. The C1s peak (Eb = 284.8 eV) corresponded
to surface hydrocarbon-like deposits (C C and C H bonds) accu-
mulated on the surface during the storage in the atmosphere. A
low-energy electron gun (FG-15/40, SPECS) was used for the sam-
ple charge neutralisation. Peak decomposition was made using XPS
Peak Fitting Program Version 4.1.
The bulk crushing strength (BMS) was measured using SMS 1471
or the analogous ASTM method 7084-4 (Standard Test Method for
Determination of Bulk Crushing Strength of Catalysts and Catalyst
Carriers) was carried out using a bulk crushing strength instru-
ment, VINCI Technologies, France. Sulphur content in catalysts was
measured using VarioELIII elemental analyser (ELEMENTAR Analy-
sensysteme GmbH). S/Mo ratio was in the range of 2.1 ± 0.1 for all
the catalysts.
3. Results and discussion
All the initial powders had similar Al2 O3 contents and small
amounts of admixtures (Table 1). The X-ray diffraction spectra of
all the studied powders contain a set of peaks characteristic of
the structure of boehmite (ICDD PDF No. 21-1307). No peaks for
other phases were observed. The main differences between pow-
ders were the size and form of crystalline AlOOH particles. The size
of boehmite crystals is often calculated using the Scherrer formula
based on the XRD data, e.g., [29,34–36]. However, the crystallite size
defined using XRD data often differs from the particle size deter-
mined from HRTEM microphotographs. For example, according to
the XRD data, the particle size is not more than 3 nm, whereas trans-
mission electron microscopy data reports a length of 30 nm and
a width of 10 nm [35]. In [29], the HRTEM patterns show a pref-
erential particle length of approximately 50 nm for the boehmite
Fig. 1. XPS and HRTEM patterns of AlOOH for ISCZC-1 and ISCZC-2.
69
sample, which has a maximum particle size of 3.6 nm according
to the XRD data. One study investigated the differences in parti-
cle size calculated from XRD and HRTEM, and the XRD method
was proven to be especially efficient for the smallest nanoparti-
cles (approximately 3–7 nm) [36]. In our study, according to XRD,
only one sample, Pural SB, has a particle size in this range. For all
the other samples, the particle sizes calculated using XRD are in
the range of 7–17 nm (Table 1). The non-applicability of the XRD
method in this case is illustrated in Fig. 1.
The HRTEM patterns clearly show that the ISCZC-1 and ISCZC-
2 particles are needle-shaped. However, the average length of
the ISCZC-2 needles is approximately 1000 nm, which is several
times higher than that of the ISCZC-1 particles, with an average
length of approximately 150 nm. The X-ray diffraction patterns
of these samples are similar, and the maximum size calculated
using the Scherrer formula (plane 002) is 17 nm, which is clearly
incorrect. Accordingly, all the conclusions about the size and form
of the boehmite particles have been based on the HRTEM data.
Typical HRTEM patterns for the boehmite samples are shown in
Fig. 2. Boehmite particles can have different morphologies, and the
increase in the particle size results in the transformation of the
deformed hexagonal morphology into diamond-shaped platelets
[36]. Nevertheless, the particles with a length of approximately
10 nm have a complicated morphology, and their form is similar to
a rectangular cuboid. Many papers have been devoted to the mor-
phology of boehmite, describing particles with rectangular cuboid
or cuboid forms with truncated corners [29,35–39].
Sizes, values of external surface and volumes of typical AlOOH
crystals that were determined using HRTEM data in the approx-
imation of rectangular cuboid form are given in Table 3. General
crystal surface area was calculated using Sc = 4 × L × B + 2 × B × H
formula. Volume of particles was calculated using Vc = L × B × H for-
mula, where L, B, H are length, width and height of the crystal that
were defined by HRTEM data.
The dimensions, surface, and volume defined using the HRTEM
data, approximating a rectangular cuboid form, are shown in
Table 3. According to particle form, all the studied samples fall
into three groups: (1) laminar crystals, (2) prolonged lamina, and
(3) needles and thin plates. The samples produced by Sasol have
a particle morphology that is typical for samples prepared by
alkoxide hydrolysis [27], and their particle form corresponds to
laminar crystals and prolonged lamina. The Chimtek sample is a
mixture of thin plates and needles that is typical for precipitate
boehmite [29,34,36,37]. The ISCZC samples contain needle-shaped
particles that are typical for boehmite prepared in acidic conditions
[36].
70 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77

Fig. 2. HRTEM patterns of the typical fragments of initial AlOOH powders.

Table 3
Characteristics of AlOOH particles based on HRTEM data. Length (L), width (B) and heigth (H) of crystals, Sc – crystal surface area, Vc – crystal volume.

Sample Pural SB TH-60 TH-80 TM-50 TM-70 Chimtek ISCZC-1 ISCZC-2

Average particle size 15 × 10 × 5 20 × 10 × 8 40 × 20 × 10 25 × 10 × 8 35 × 8 × 8 110 × 5 × 5/110 × 40 × 5 150 × 8 × 8 1000 × 10 × 10

(L nm × B nm × H nm)
Sc (nm2 ) 550 880 2800 1060 1680 2250/10,300 4928 40,200
Vc (nm3 ) 750 1600 8000 2000 2248 2750/22,000 9600 100,000
Sc /Vc 0.73 0.55 0.35 0.53 0.75 0.82/0.47 0.51 0.40
Particle shape Laminar Laminar Laminar Prolonged Prolonged Thin plates and needles Needles Long needles
crystals crystals crystals lamina lamina

Table 4
Textural characteristics and mechanical strength of the supports and catalysts.

AlOOH Sample Surface area (m2 /g) Pore volume (cm3 /g) Average pore diameter (nm) Bulk crushing strength (MPa)

Pural SB Carrier 270 0.49 7.3 0.45

Catalyst 206 0.33 6.5 0.86
TH-60 Carrier 235 0.70 11.8 0.35
Catalyst 155 0.43 11.0 0.70
TH-80 Carrier 214 0.68 12.7 0.24
Catalyst 142 0.44 12.3 0.68
TM-50 Carrier 263 0.65 9.9 0.33
Catalyst 193 0.45 9.3 0.66
TM-70 Carrier 254 0.68 10.8 0.48
Catalyst 162 0.40 10.0 0.70
Chimtek Carrier 216 0.45 7.8 0.77
Catalyst 147 0.31 8.3 1.53
ISCZC-1 Carrier 219 0.55 10.0 0.77
Catalyst 183 0.44 9.5 1.57
ISCZC-2 Carrier 230 0.53 9.2 1.01
Catalyst 145 0.37 10.3 1.80
 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77
Fig. 3. HRTEM of the typical fragments of sulphided catalysts.
According to the XRD data, supports containing only ␥-
Al2 O3 were obtained in all the cases after peptisation, extrusion,
drying and calcinations. No other phases were observed. These
supports differed significantly in textural characteristics and
mechanical strength (Table 4). According to the HRTEM and XRD
data for all the supports and catalysts, the shape of ␥-Al2 O3 parti-
cles is similar to that of initial boehmite particles (Fig. 3), whereas
the initial boehmite particles fully converted into ␥-Al2 O3 after cal-
cination. The interdependence of the form of the initial boehmite
particle and the ␥-Al2 O3 obtained from this boehmite was previ-
ously described in [35–37,39].
In contrast to the initial AlOOH powders, it is not possible to cal-
culate the precise size of ␥-Al2 O3 particles using the HRTEM data
for the supports and catalysts due to the sintering and agglomerate
formation. However, it is possible to determine the most preva-
lent particle shape in a sample. For example, laminar crystals with
shapes and sizes comparable to those of the initial AlOOH particles
are clearly observed in the microphotographs of sulphided catalysts
prepared from Pural SB, TH-60, and TH-80. HRTEM patterns of the
catalysts prepared from TM-50 and TM-70 primarily contain lam-
ina particles. Finally, microphotographs of the catalysts based on
the Chimtek and ISCZC samples contain needles and agglomerates,
which are typical of initial boehmites.
The textural characteristics of the supports are similar (Table 4):
an average surface area of 216–270 m2 /g, a pore volume of
0.45–0.70 cm3 /g, and an average pore diameter of 7.3–12.7 nm.
However, the pore size distribution and mechanical strength differ
significantly. For example, the mechanical strength of the ISCZC-2
support differs from that of TH-80 by more than a factor of 4.
The obtained supports can be divided in two groups: weakly
resistant (<0.48 MPa) and strongly resistant (>0.77 MPa). The
weakly resistant supports are those prepared from Sasol Alkoxide
powders and feature a laminar form with a length not exceeding
40 nm. Strongly resistant supports were prepared from Chimtek
and ISCZC boehmites, which have needle-shaped particles with
lengths of no more than 110 nm. The mechanical strength of the
supports prepared from Sasol powders with laminar particles are
well correlated with the Sc /Vc ratio for the particles of the ini-
tial boehmite, yielding the empirical formula BMS ≈ 0.65 × (Sc /Vc ).
There is no similar correlation for boehmites with needle-shaped
particles, and the mechanical strength of these supports increases
with the average needle length. The mechanical strength of the
supports for all the studied boehmites is defined by the num-
ber of contacts between particles per unit volume of the support
for similar peptisation, granulation, and calcination conditions.
71
Meanwhile, the number of contacts depends on the Sc /Vc ratio
of the initial particles. Interweaving of agglomerates, which is
observed in the HRTEM patterns (Fig. 3), contributes to the mechan-
ical strength of the supports for boehmites with needle-shaped
morphology and long lengths.
The pore size distributions and nitrogen adsorption/desorption
isotherms for the supports and catalysts are shown in Figs. 4 and 5.
The morphology of the initial boehmite particles defines both
the pore size distribution and shape of pores. The average pore
diameter increases with the particle size for the supports prepared
using powders with similar particle shapes. For example, the aver-
age pore diameter increased from 7.3 to 12.7 in the Pural SB, TH-60,
and TH-80 line and from 9.9 to 10.8 nm for TM-50 and TM-70. A nar-
row pore size distribution and essentially no micro- or macropores
according to the classification used in [33,40,41] were reported for
the supports based on the Sasol and ISCZC-1 boehmites. The sup-
ports based on Chimtek and ISCZC-2 powders have a broad pore
size distribution and contain both micro- and macropores.
The pore shape of the supports can be defined in terms of the
nitrogen adsorption/desorption isotherms. According to classifica-
tions [33,40,41], the supports prepared from Sasol and ISCZC-1
powders exhibit type IV isotherms containing H1 and H2 hys-
teresis loops, which are common for mesoporous supports and
catalysts. The pore shape is approximately cylindrical, which is typ-
ical for materials containing particles with similar dimensions for
each axis in a three-dimensional coordinate system. Both Chimtek
and ISCZC-2 supports exhibit the type IV isotherms with H3 hys-
teresis loops. Consequently, these supports contain preferentially
slit-shaped pores between long needles and laminas.
Both the bulk crushing strength and textural characteristics
(Table 4; Fig. 4) change considerably after the supporting of 11.0% of
Mo and 3.5% of Co. However, these changes depend on the support’s
characteristics and, therefore, on the morphology of the initial alu-
mina hydroxide particles. The change in the average surface area
and pore volume is not affected by the nature of the support. The
decrease in the average surface area is in the range of 64–92 m2 /g
and that in the pore volume is approximately 0.11–0.28 cm3 /g.
These changes are typical for hydrotreating catalysts and have been
described for many times in the literature [6,9,10,14,15]. By com-
paring the pore size distribution of supports and catalysts (Table 2;
Fig. 4), it was noticed there is no significant changes in portion of
any pores after supporting of active metals. Thus, active metals are
uniformly distributed over the surface of each support and present
at both wide and narrow pores. The shape of the nitrogen adsorp-
tion/desorption isotherms does not change after supporting of Co
72 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77

Fig. 4. Pore size distribution for supports and catalysts.

 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77 73

Fig. 5. Nitrogen adsorption/desorption isotherms for supports and catalysts.

74 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77

Table 5 161.7 ± 0.1 eV and a shoulder at 162.8 ± 0.1 eV are observed in all
Particle morphology of type II Co-Mo-S according to the HRTEM data.
S2p spectra. The intensity ratio of the main peak to the shoulder
Catalyst Average slab Average slab Average slab number is approximately 6:1. These binding energies and intensities are
length (nm) number per 1000 nm2 typical for S2− and S2 2 in Co-Mo catalysts [42–46]. The binding
Pural SB 3.04 1.8 41 energy for Co2p is 778.9 ± 0.1 eV, which is characteristic of Co2+
TH-60 3.36 1.9 60 in sulphur surrounding in the composition of the Co-Mo-S phase
TH-80 3.37 2.0 52 [42–46]. To confirm the presence of elements in the composition of
TM-50 3.31 1.8 39
a bimetallic Co-Mo-S phase and exclude the presence of less active
TM-70 3.27 1.8 48
Chimtek 3.25 1.7 45 individual sulphides, the binding energies of Co2p, Mo3d, and S2p
ISCZC-1 3.16 1.9 46 are compared, as proposed in [45]. For example, BE = Co2p-S2p is
ISCZC-2 3.26 1.8 47 616.8–617.1 eV for Co-Mo-S phase [43–45] and 616.5 [44] or 616.2
[45] for Co9 S8 . The experimental results obtained for BE are in
good agreement with the data for the Co-Mo-S phase described in
and Mo. Therefore an alteration of pore shapes and plugging does [43–45]. The Co/S and Mo/S atomic ratios obtained from the XPS
not occur (Fig. 5). All the catalysts could be fall into three groups spectra and S/Mo = 2.1 ± 0.1 ratio obtained from elemental analysis
different in pore size distribution: (1) catalysts with high content confirm full sulphidation of the metals. The Co/Mo ratio obtained
of narrow pores with the diameters <5 nm (Pural SB, Chimtek); (2) from XPS is higher than that based on the elemental analysis data
catalysts, in which pores with the diameters 7–13 nm prevail (TH- due to the cobalt localisation on the edges of MoS2 and the screen-
60, TH-80, TM-50, TM-70, ISCZC-1); and (3) catalysts with wide ing of molybdenum by cobalt.
pore size distribution (ISCZC-2). On the other hand, decomposition of the Mo3d and Co2p spec-
Supporting of Co and Mo results in the increase in the bulk crush- tra is widely used to define active metal sulphidation degree and
ing strength by 0.22–0.44 MPa for the supports prepared using Sasol presence of Co and Mo in the composition of Co-Mo-S phase. Cal-
boehmite and 0.75–0.83 MPa for those prepared using Chimtek culations are usually performed using three different Mo (Mo4+ ;
and ISCZC boehmite. This change could most likely be attributed Mo5+ and Mo6+ ) states [42–44,46] and two (Co-S and Co-O) [43,46]
to the localisation of active metals either near the point of con- or three (Co-Mo-S, Co9 S8 and Co2+ ) [42,44] Co states. Some works
tact between separate ␥-Al2 O3 particles or in narrow slot-like studied hydrotreating catalysts with different metal content that
pores. Co and Mo compounds form bridges between ␥-Al2 O3 par- were prepared using the same support, e.g. ␥-alumina extrudates
ticles, increasing the mechanical strength of the catalyst relative [44], or nanotubular titania [46]. In this case the dependence
to the initial supports. Because narrow slot-like pores are con- of catalytic activity in toluene-hydrogenation or HDS of diben-
siderably prevalent in the supports containing parallel-oriented zothiophene on metal sulphidation degree and Co-Mo-S phase
conjugate ␥-Al2 O3 particles with needle or prolonged morpholo- concentration was defined. In the case of different supports and
gies (Fig. 3), these supports are characterised by a greater increase preparation techniques dependence of catalytic activity on metal
in the mechanical strength after supporting of Co and Mo. sulphidation degree and Co-Mo-S phase concentration is absent
All the sulphided catalysts contain Co-Mo-S particles with simi- [42].
lar morphologies. The morphology and surface metal concentration The data obtained on decomposition of Mo3d spectra are suf-
are typical for well-described catalysts containing Co-Mo-S phase ficiently similar for all the catalysts studied. The content of each
of Type II [1–6,42,43]. Mo state is within following limits, %: Mo4+ – 79.1 ± 5.5; Mo5+
In our case, Co-Mo-S particles have an average slab length of – 12.6 ± 4.1; Mo6+ – 8.2 ± 1.4. Two cobalt states were defined by
approximately 3.20 ± 0.15 nm and an average stacking number of decomposition of Co2p spectra using decomposition on two states,
1.85 ± 0.15, which are typical for CoMo/Al2 O3 catalysts sulphided %: CoS – 77.6 ± 1.5; CoO – 22.4 ± 1.5. If decomposition on three
under H2 S flow [42]. It should be noted that no correlation between states is used, it is not possible to obtain exact fit between experi-
the average surface area and slab number per 1000 nm2 for cata- mental and calculated spectra. Therefore, decomposition of Co2p3/2
lysts with different surface areas was observed (Table 5). line was performed similarly to the one described in [42]. In this
All the catalysts in the sulphided state were studied using XPS. case values were in the following ranges for all the catalysts, %:
Thus, the differences in the binding energies for each catalyst did Co-Mo-S – 77.0 ± 3.0; Co9 S8 – 8.0 ± 2.0; Co2+ – 16.6 ± 2.5. Decom-
not exceed 0.1 eV, and the discrepancies in the atomic ratios for position of Mo3d and Co2p XPS spectra for two catalysts having the
the elements were not more than 15%. Therefore, XPS data for the highest and the lowest sulphidation degree are shown in Fig. 6 as
catalysts are given in Table 6. an example. In addition, these catalysts have the highest (Catalyst
One peak with a binding energy of Eb = 228.8 eV is observed in ISCZC-2) and the lowest (Catalyst TH-60) Co-Mo-S phase concen-
the Mo3d spectra for all samples. This binding energy is charac- tration. XPS method has a definite accuracy. Thus, the data for Co
terised for Mo4+ in the sulphided state in the composition of the and Mo states are given with either 1% accuracy [42] or with sta-
Co-Mo-S phase [42–46]. A main peak with a binding energy at tistical straggling, which is about 10% [44].

Table 6
Average XPS data for sulphided catalysts.

Element Al2p S2p Mo3d C1s O1s Co2p

Binding energy, eV 74.6 161.7 ± 0.1 228.8 ± 0.1 284.8 531.4 778.9 ± 0.1
162.8 ± 0.1
Element Co/Mo Co/Al Mo/Al Co/S Mo/S O/Al
Atomic ratio 0.65 ± 0.10 0.11 ± 0.01 0.15 ± 0.02 0.25 ± 0.03 0.31 1.7 ± 0.1

BE (eV) for Co-Mo-S phase Co2p-Mo3d Mo3d-S2p Co2p-S2p

Our experiments 550.0 ± 0.2 67.1 ± 0.1 617.2 ± 0.1

Ref. [43] 550.0 67.2 616.8
Ref. [44] 549.8 617.1
Ref. [45] 66.8 ± 0.1 617.0
 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77 75

Fig. 6. Decomposition of Mo3d and Co2p XPS spectra for two catalysts having the highest (ISCZC-2) and the lowest (TH-60) sulphidation degree.

Portion of Co and Mo in each of the catalyst studied is in the
range of statistical straggling [44]. Consequently, metal sulphida-
tion degree and portion of Co and Mo states for different catalysts
are similar within the limits of XPS method accuracy. In our case,
sulphidation degree of molybdenum and cobalt is close to 80%
and 78%, respectively. Notably, 77% of general cobalt concentra-
tion present in the composition of Co-Mo-S phase. Catalysts that
have XPS characteristics similar to the ones obtained in this work
showed high activity in HDS of thiophene, dibenzothiophene and
straight run gas oil and hydrogenation of toluene [42–44,46].
The HRTEM and XPS data for sulphided catalysts suggest that
the catalysts preferentially contain supported metals in the sul-
phided form. Moreover, the morphology of the active component
is similar for all samples and is typical for high active Co-Mo HDS
catalysts. Recently, it was reported that bimetallic complex com-
pounds with citrate ligands are formed in solutions containing Co,
76 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77

Fig. 7. (a) Relative catalytic activity in HDS versus the nature of initial AlOOH; (b) relative catalytic activity versus average catalyst pore diameter.

Mo and citric acid. Moreover, citrate ligands minimize interactions
between the support and active metals [47]. Accordingly, in spite
of the possible differences in acidic characteristics of the supports
used, sulphided active component with similar composition and
structure was obtained in all the catalysts in present work.
Consequently, if the preparation method, active metal content
in the catalyst, electronic states for Co, Mo, and S in the active com-
ponent and morphology of Co-Mo-S phase particles are the same
(or very similar), the differences in the catalytic properties should
be mostly determined by the textural characteristics of the initial
supports and catalysts. Although the catalysts prepared using the
method described are suitable for the production of diesel fuels
containing not more than 10 ppm S [31,32,47], the aim of this work
was not to attain the maximum desulphurisation degree but to
illustrate the differences in catalytic properties. A HORIBA SLFA
2100 X-ray fluorescence analyser allowed the measurement of the
sulphur content with an accuracy of within 5 ppm. Thus, condi-
tions similar to those for ULSD production (e.g., [1,3,8,43,48,49])
were chosen. These conditions results in the production of diesel
fuels with residual sulphur content of not more than 50–100 ppm.
All the catalysts were tested under the same conditions: 3.8 MPa,
340 ◦ C, a liquid hourly space velocity of 2.0 h−1 , and a H2 /feed vol-
ume ratio of 300. The results of the catalyst testing are shown in
Fig. 7a and b.
The catalysts were compared in terms of relative catalytic activ-
ity, which was defined as the ratio of the residual sulphur content
in the reaction product for the most active catalyst to the residual
sulphur content in the reaction product of one of the catalysts. On
top of that, residual sulphur content for each catalyst is given in
Fig. 7a.
The activity of the most active catalyst, which prepared from
TH-60 boehmite, was normalised to 1. Under testing conditions
used in this work, a residual sulphur content in diesel fuel dif-
fered in the range of 30–40 ppm. These are accurate and repeatedly
reproduced results obtained for each catalyst. For example, a resid-
ual sulphur content of 100 ppm was obtained for the least active
catalyst (Chimtek), while the most active catalysts (TH-60, TH-80,
TM-50, TM-70, ISCZC-1) provided the production of diesel fuel with
a residual sulphur content of 50–60 ppm. Absolute difference in
the residual sulphur content seems to be not high for the different
catalysts studied in present work. It could be compensated by the
increase of process temperature on 10–20 ◦ C. On the other hand,
temperature decrease by 5–10 ◦ C to achieve specified sulphur con-
tent is so important that new kinds and generations of catalysts are
developed, e.g. C-605A and C-606A [8] and TK-574 and TK-576BRIM
[49].
As in the case of mechanical strength, the catalysts can be
divided into two groups based on the morphology of the initial
boehmites. For the catalysts based on the laminar-shaped Sasol
boehmites and needle-shaped ISCZC-1 boehmites, possessing a
narrow pore size distribution and preferentially cylindrical pores,
the activity increases with the average pore diameter, peaking
at 11 nm and then decreasing (Fig. 7b). There is no correlation
between the activity and average pore diameter for the cata-
lysts prepared using boehmites with long needle-shaped (ISCZC-2)
or wide laminar-shaped (Chimtek) particles containing a large
amount of small slot-like pores. The activity of the catalysts based
on Chimtek and ISCZC-2 supports was significantly inferior relative
to that of the catalysts based on TM-50 and TM-70 supports with
approximately the same average pore diameter. The cause of these
differences is the localisation of a considerable amount of active
metals in narrow pores with diameters not less than 5 nm in the
Chimtek and ISCZC-2 catalysts because the active component parti-
cles are not available for sulphur-containing molecules in the feed,
namely, 4,6-dibenzothiophene homologs, which are prevalent in
diesel fuels with less than 350 ppm S after hydrotreating and have
low reactivity [1,11,12,48]. This conclusion is confirmed by Fig. 8.
There is shown the dependence between catalytic activity and con-
tent of pores with diameters of 7–13 nm in the catalysts (Table 2).
Pores with such diameters are preferential for hydrotreatment of
distillates [14]. The volume of pores with diameters of 7–13 nm
does not exceed 25% in the catalysts based on Pural SB, ISCZC-2 and
Fig. 8. The dependence between catalytic activity and content of pores with diam-
eters of 7–13 nm in the catalysts.
 O.V. Klimov et al. / Catalysis Today 220–222 (2014) 66–77
Chimtek boehmites. Consequently, these catalysts have low activity
that could not provide the production of ULSD. Volume portion of
pores with diameters of 7–13 nm is more than 50% in the catalysts
based on TH-60, TH-80, TM-50 and TM-70 boehmites. These cata-
lysts have high activity. However, mechanical strength values for
these catalysts are insufficient for industrial application (Table 4).
The catalyst based on ISCZC-1 support contains about 33% of
pores with the diameters of 7–13 nm. It was not inferior in activ-
ity to that of most of the samples based on Sasol supports and
was much greater in activity than that of the samples based on
Chimtek and ISCZC-2 supports. High mechanical strength value of
this catalyst allows one to recommend it for industrial application.
Obviously, particle morphology of ISCZC-1 powder (needles with
the length of 150 nm × 8 nm × 8 nm) produces the support and then
catalyst with an optimal combination of mechanical strength and
texture. We do not consider that particle morphology of AlOOH
powders optimal for preparation of hysrotreating catalysts of diesel
fuel should be identical to ISCZC-1 powder. An average crystallite
size of AlOOH could deviate within several limits along each axis.
However, main tendency should be kept–the powder should con-
tain needles with the length of 100–200 nm, width and height of
8–10 nm. The increase of needle length will result in the increase of
amount of slit-like narrow pores that are undesirable for catalysis.
Decrease of needle length inevitably decreases catalyst mechanical
strength.
4. Conclusions
The influence of the initial boehmite particle morphology on the
mechanical strength, textural characteristics, and catalytic activ-
ity of Co-Mo/Al2 O3 catalysts for the deep hydrotreatment of diesel
fuels was studied. The catalysts were prepared using bimetallic
complexes with citrate ligands.
The active component in all the catalysts was shown to have
similar structure and composition. The nature and morphology of
the initial boehmite particles have little influence on particle mor-
phology of the active component (length of 3.20 ± 0.15 nm and
average stacking number of 1.85 ± 0.15), and Mo and Co sulphi-
dation degree and portion of cobalt in the composition of CoMoS
phase (79.1 ± 5.5; 77.6 ± 1.5; 77.0 ± 3.0%, respectively). However,
the mechanical strength, textural characteristics, localisation of the
active component, and catalytic activity in the hydrotreatment of
diesel fuels are defined by the morphology of the initial boehmite
particles.
The catalysts based on Sasol boehmites with laminar-shaped
particles, e.g., TH-60, have an optimal pore size and form for
hydrotreatment and, accordingly, have high catalytic activity. How-
ever, their mechanical strength is not sufficient for use in industry.
The catalysts based on Chimtek (a mixture of needles and large
laminas) and ISCZC-2 (very long needles) boehmites contain large
amounts of narrow pores. The Co-Mo-S phase, which is localised
in these pores, is not available for the large sulphur-containing
molecules in the feed; therefore, the catalysts have poor activity.
Accordingly, despite their high mechanical strength, Chimtek and
ISCZC-2 boehmites are not optimal for the preparation of industrial
catalysts for deep diesel fuel hydrotreatment.
Among all the powders studied the optimal material for
the preparation of industrial supports and catalysts is ISCZC-1
boehmites containing 150 nm × 8 nm × 8 nm needles. Obviously,
this morphology for the initial AlOOH particles provides the optimal
combination of high mechanical strength and textural characteris-
tics for the deep hydrotreatment of diesel fuels.
77
Acknowledgements
The authors are grateful to Mrs. Tatyana Ya. Efimenko for mea-
surements of the textural parameters of the samples.
References
[1] A. Stanislaus, A. Marafi, M.S. Rana, Catal. Today 153 (2010) 1.
[2] H. Topsoe, Appl. Catal. A 322 (2007) 3.
[3] S. Eijsbouts, L.C.A. Van den Oetelaar, R.R. Van Puijenbroek, J. Catal. 229 (2005)
352.
[4] E.J.M. Hensen, V.H.J. de Beer, J.A.R. van Veen, R.A. van Santen, Catal. Lett. 84
(2002) 59.
[5] J.A.R. Van Veen, E. Gerkema, A.M. Van der Kraan, A. Knoestera, J. Chem. Soc.
Chem. Commun. (1987) 1684.
[6] F. Dumeignil, K. Sato, M. Imamura, N. Matsubayashi, E. Payen, H. Shimada, Appl.
Catal. A 315 (2006) 18.
[7] M. Sun, D. Nicosia, R. Prins, Catal. Today 86 (2003) 173.
[8] T. Fujikawa, J. Jpn. Petrol Inst. 50 (2007) 249.
[9] A. Duan, G. Wan, Z. Zhao, C. Xu, Y. Zheng, Y. Zhang, T. Dou, X. Bao, K. Chung,
Catal. Today 119 (2007) 13.
[10] E. Hillerova, H. Morishige, K. Inamura, M. Zdrazil, Appl. Catal. A 156 (1997) 1.
[11] M.J. Girgis, B.C. Gates, Ind. Eng. Chem. Res. 30 (1991) 2021.
[12] C. Song, X. Ma, Appl. Catal. B 41 (2003) 207.
[13] J. Ancheyta, M.S. Rana, E. Furimsky, Catal. Today 109 (2005) 3.
[14] S. Eijsbouts, Hydrotreating catalysts, in: K.P. de Jong (Ed.), Synthesis of Solid
Catalysts, VILEY-VCH Verlag GmbH & Co, KGa A, Weinheim, 2009, p. 302.
[15] M.S. Rana, J. Ancheyta, P. Rayo, S.K. Maity, Catal. Today 98 (2004) 151.
[16] J. Ancheyta, G. Marroqui, M.J. Angeles, M.J. Macias, I. Pitault, M. Forissier, R.D.
Morales, Energy Fuels 16 (2002) 1059.
[17] F.S. Mederos, J. Ancheyta, J. Chen, Appl. Catal. A 355 (2009) 1.
[18] M.J. Macias, J. Ancheyta, Catal. Today 98 (2004) 243.
[19] B.H. Cooper, B.B.L. Donnis, B. Moyse, Oil Gas J. 84 (1986) 39.
[20] S.T. Sie, Chem. Eng. J. 53 (1993) 1.
[21] J.D. Carruthers, D.J. Di Camillo, Appl. Catal. 43 (1988) 253.
[22] D. Wu, L. Song, B. Zhang, Y. Li, Chem. Eng. Sci. 58 (2003) 3995.
[23] Y. Li, D. Wu, Y.S. Lin, China Particuol. 2 (2004) 53–62.
[24] D. Wu, Y. Li, Stud. Surf. Sci. Catal. 143 (2000) 101–109.
[25] D. Wu, Y. Li, Y. Shi, Z. Fang, D. Wu, L. Chang, Chem. Eng. Sci. 57 (2002) 3495.
[26] D. Wu, J. Zhou, Y. Li, Chem. Eng. Sci. 64 (2009) 198.
[27] A. Danner, K.K. Unger, Stud. Surf. Sci. Catal. 31 (1987) 343.
[28] V. Kaloidas, A.M. Thanos, D.C. Tsamatsoulis, N.G. Papayannakos, Chem. Eng.
Proc. 39 (2000) 407.
[29] F. Karouia, M. Boualleg, M. Digne, P. Alphonse, Powder Technol. 237 (2013) 602.
[30] Russian Patent No. 2167818 (1999).
[31] O.V. Klimov, A.V. Pashigreva, M.A. Fedotov, D.I. Kochubey, Yu.A. Chesalov, G.A.
Bukhtiyarova, A.S. Noskov, J. Mol. Catal. A 322 (2010) 80.
[32] O.V. Klimov, A.V. Pashigreva, G.A. Bukhtiyarova, S.V. Budukva, M.A. Fedotov, D.I.
Kochubey, Yu.A. Chesalov, V.I. Zaikovskii, A.S. Noskov, Catal. Today 150 (2010)
196.
[33] G. Leofanti, M. Padovan, G. Tozzola, B. Venturelli, Catal. Today 41 (1998) 207.
[34] D. Panias, A. Krestou, Powder Technol. 175 (2007) 163.
[35] R. Petrovic, S. Milonjic, V. Jokanovic, Lj. Kostic-Gvozdenovic, I. Petrovic-Prelevic,
Dj. Janackovic, Powder Technol. 133 (2003) 185.
[36] D. Chiche, M. Digne, R. Revel, C. Chaneac, J.-P. Jolivet, J. Phys. Chem. C 112 (2008)
8524.
[37] G. Li, Y. Liu, Z. Tang, C. Liu, Appl. Catal. A 437–438 (2012) 79.
[38] P. Raybaud, M. Digne, R. Iftimie, W. Wellens, P. Euzen, H. Toulhoat, J. Catal. 201
(2001).
[39] M. Digne, P. Sautet, P. Raybaud, P. Euzen, H. Toulhoat, J. Catal. 226 (2004) 54.
[40] K. Kaneko, Membr. Sci. 96 (1994) 59.
[41] S. Brunauer, L.S. Deming, W.E. Deming, E. Teller, J. Am. Chem. Soc. 62 (1940)
1723.
[42] D. Laurenti, Bo Phung-Ngoc, C. Roukoss, E. Devers, K. Marchand, L. Massin, L.
Lemaitre, C. Legens, A.-A. Quoineaud, M. Vrinat, J. Catal. 297 (2013) 165.
[43] N. Frizi, P. Blanchard, E. Payen, P. Baranek, C. Lancelot, M. Rebeilleau, C. Dupuy,
J.P. Dath, Catal. Today 130 (2008) 32.
[44] A.D. Gandubert, E. Krebs, C. Legens, D. Costa, D. Guillaume, P. Raybaud, Catal.
Today 130 (2008) 149.
[45] I. Alstrup, I. Chorkendorff, R. Candia, B.S. Clausen, H. Topsøe, J. Catal. 77 (1982)
397.
[46] J.A. Toledo-Antonio, M.A. Cortes-Jacome, C. Angeles-Chavez, J. Escobar, M.C.
Barrera, E. Lopez-Salinas, Appl. Catal., B 90 (2009) 213.
[47] A.V. Pashigreva, O.V. Klimov, G.A. Bukhtiyarova, M.A. Fedotov, D.I. Kochubey,
Yu.A. Chesalov, V.I. Zaikovskii, I.P. Prosvirin, A.S. Noskov, Stud. Surf. Sci. Catal.
175 (2010) 109.
[48] K.G. Knudsen, B.H. Cooper, H. Topsoe, Appl. Catal., A 189 (1999) 205.
[49] H. Topsoe, B. Hinnemann, J.K. Norskov, J.V. Lauritsen, F. Besenbacher, P.L.
Hansen, G. Hytoft, R.G. Egeberg, K.G. Knudsen, Catal. Today 107–108 (2005)
12.