CoMoW sulfide nanocatalysts for the HDS of DBT from novel ammonium and alkyltrimethylammonium-thiomolybdatethiotungstate-cobaltate (II) precursors

Article abstract of DOI:10.1016/j.apcata.2014.08.017

Five unsupported, highly active CoMoW trimetallic nanocatalysts were obtained by in situ decomposition from five novel precursors: (NH₄)₂[Co(MoS₄)(WS₄)] and (RN(CH₃)₃)₂[(MoS₄)(WS₄)] (where R = dodecyl, tetradecyl, cetyl and octadecyl), during the HDS of DBT. The catalyst labeled CoMoWS-C14, derived from the precursor containing the tetradecyl group, exhibits the highest catalytic activity (k = 421 × 10^-7mol/gs). N₂ adsorption-desorption shows that the CoMoW catalysts are mesoporous materials with characteristic Type IV isotherms, having surface areas of 11-340m²/g. Elemental analysis by X-ray energy dispersive spectroscopy (EDS) working at STEM mode finds high concentrations of carbon (3.7 < C/Mo < 11.3 and 2.7 < C/W < 9) in all the catalysts except CoMoWS, where carbon was not detected. The XRD patterns show that the catalysts are highly dispersed (less so for the CoMoWS) given the absence of the (00 2) reflection, along with broad and low intensity (101) and (110) reflections. High dispersion is also supported by the STEM micrographs showing unstacked layers. The selectivity of the reaction for all catalysts favors the direct desulfurization pathway. The surface area and high catalytic activity do not show direct correlation with the length of the hydrocarbon chains of the precursors.

Full text of DOI:10.1016/j.apcata.2014.08.017

Accepted Manuscript
Title: CoMoW sulfide nanocatalysts for the HDS of DBT from
novel ammonium and alkyltrimethylammonium-
thiomolybdate-thiotungstate-cobaltate (II)
precursors
Author: Y. Espinoza-Armenta J. Cruz-Reyes F.
Paraguay-Delgado M. Del Valle G. Alonso S. Fuentes R.
Romero-Rivera
PII:
S0926-860X(14)00502-X
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.08.017
APCATA 14954
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
7-6-2014
12-8-2014
18-8-2014
Please cite this article as: Y. Espinoza-Armenta, J. Cruz-Reyes, F. Paraguay-Delgado,
M. Del Valle, G. Alonso, S. Fuentes, R. Romero-Rivera, CoMoW sulfide nanocatalysts
for the HDS of DBT from novel ammonium and alkyltrimethylammonium-
thiomolybdate-thiotungstate-cobaltate (II) precursors, Applied Catalysis A, General
(2014), http://dx.doi.org/10.1016/j.apcata.2014.08.017
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

GRAPHICAL ABSTRACT
Page 1 of 27

HIGHLIGHTS
Highly active CoMoW nanocatalysts were obtained from novel precursors.
These nanocatalysts are mesoporous materials with surface areas up to 340 m2/g.
XRD and TEM analysis finds the catalysts are highly dispersed materials.
Catalysts favor the direct desulfurization pathway.
Page 2 of 27

CoMoW sulfide nanocatalysts for the HDS of DBT from novel
ammonium and alkyltrimethylammonium-thiomolybdate-thiotungstate-
cobaltate (II) precursors
Y. Espinoza-Armenta1, J. Cruz-Reyes1, F. Paraguay-Delgado2, M. Del Valle1, G.
Alonso3, S. Fuentes, and R. Romero-Rivera1*
1) Facultad de Ciencias Químicas e Ingeniería, Universidad Autónoma de Baja
California, Ave. Universidad 14418, Mesa de Otay, Tijuana, C.P. 22390, B.C., Mexico.
2) Depto. De Materiales Nanoestructurados, Centro de Investigación en Materiales
Avanzados, S.C., Miguel de Cervantes 120, Chihuahua, Chih. CP. 31109, Mexico.
3) Centro de Nanociencias y Nanotecnología, Universidad Nacional Autónoma de
México, Km 107 Carretera Tijuana-Ensenada, C. P. 22860, Ensenada, B.C., Mexico.
ABSTRACT
Five unsupported, highly active CoMoW trimetallic nanocatalysts were obtained
by in situ decomposition form five novel precursors: (NH4)2[Co(MoS4)(WS4)] and
(RN(CH3)3)2[(MoS4)(WS4)] (where R=dodecyl, tetradecyl, cetyl and octadecyl), during
the HDS of DBT. The catalyst labeled CoMoWS-C14, derived from the precursor
containing the tetradecyl group, exhibits the highest catalytic activity
(k =421x10-7mol/g.s ) . N2 adsorption-desorption shows that the CoMoW catalysts are
mesoporous materials with characteristic Type IV isotherms, having surface areas of 11-
340 m2/g. Elemental analysis by X-ray energy dispersive spectroscopy (EDS) working at
STEM mode finds high concentrations of carbon (3.7≤C/Mo≤11.3 and 2.7≤C/W≤9) in all
the catalysts except CoMoWS, where carbon was not detected. The XRD patterns show
that the catalysts are highly dispersed (less so for the CoMoWS) given the absence of the
(002) reflection, along with broad and low intensity (101) and (110) reflections. High
dispersion is also supported by the STEM micrographs showing unstacked layers. The
selectivity of the reaction for all catalysts favors the direct desulfurization pathway. The
surface area and high catalytic activity do not show direct correlation with the length of
the hydrocarbon chains of the precursors.
Keywords: Molybdenum disulfide, HDS of DBT, tetraalkylammonium thiomolybdates,
trimetallic catalyst
*Corresponding author
E-mail: dr.raulromero56@gmail.com
Phone: (664) 6822790
Page 3 of 27

1. Introduction
The use of transition metal sulfide (TMS) catalysts in oil refinery hydrotreating
has helped reduce the toxic SOx and NOx atmospheric emissions from internal
combustion engines for several decades. Hydrotreating involves a variety of catalytic
reactions, among them, hydrodesulfurization (HDS), hydrodenitrogenation (HDN),
hydrodeoxygenation (HDO), and hydrogenation [1-4].
Typically the hydrotreating process is performed over alumina-supported catalysts
based on molybdenum and tungsten sulfide, promoted with Co or Ni [5-7].
These materials are normally produced by sulfidation of the corresponding metal oxides.
However, it is well known that incomplete sulfidation due to a strong interaction between
metal and support results in a non-optimal hydrotreating activity [8].
Unsupported sulfide catalysts have been synthesized by different methods,
including sulfidation of coprecipitated oxides [7, 9] comaceration [10-12], homogeneous
sulfide precipitation [13], ceramic method [14], impregnated thiosalt decomposition [15],
hydrothermal and solvothermal processes [16-20]. The structures and catalytic properties
of the materials obtained by these methods strongly depend on their preparation and
treatment conditions [12, 19, 21]. The thiosalt decomposition method is simple and
reproducible and has been used in the preparation of binary and ternary sulfides with
controlled stoichiometry, a wide range of specific surface area and improved catalytic
activity [3, 22, 23].
Recently, the development of the NEBULA catalysts, with a HDS activity three
times greater than conventional industrial catalysts, has renewed the interest in research
focusing on unsupported trimetallic catalytic systems [24]. Nanostructured TMS such as
MoS2 -- with their high surface to volume ratio -- are also improving HDS performance
over conventional catalysts [25-27].
This work reports the properties of new unsupported CoMoW sulfide catalysts
obtained by in situ decomposition of novel ammonium and alkyltrimethylammonium-
thiomolybdate-thiotungstate-cobaltate (II) precursors having lauryl (dodecyl, C12),
myristil (tetradecyl, C14), cetyl (hexadecyl, C16), and stearyl (octadecyl, C18) as alkyl
groups, respectively. These new precursors were prepared by reacting equimolar
amounts (1:1:1) of the corresponding ammonium thiomolybdate, ammonium
Page 4 of 27

thiotungstate, alkyltrimethylammonium thiomolybdate, alkyltrimethylammonium
thiotungstate, and Co(NO3)2. The catalytic activity of the catalysts was tested in the HDS
of dibenzothiophene (DBT).
2. Experimental
2.1 Preparation of ammonium thiomolybdate (ATM)
Ammonium thiomolybdate, (NH4)2MoS4, was prepared by bubbling hydrogen
sulfide gas (H2S) for 60 min in an aqueous solution of ammonium heptamolybdate,
(NH4)6Mo7O24, and NH4OH. After this process, red crystals were formed [28]. The
resulting ATM was separated by vacuum filtration, washed with isopropyl alcohol, then
dried at room temperature, and kept under N2 atmosphere.
2.2 Preparation of ammonium thiotungstate (ATT)
Ammoniun thiotungstate (NH4)2WS4 was prepared according to the method
described by Ramanathan and Weller [29] which requires adding ammonium
metatungstate, NH4OH, H2S and water, then heating at 60°C for 4 h to produce a dark
yellow crystalline precipitate. The resulting ATT was separated by vacuum filtration,
washed with isopropyl alcohol, then dried at room temperature, and kept under N2
atmosphere.
2.3 Preparation of dodecyltrimethyl, tetradecyltrimethyl, hexadecyltrimethyl and
octadecyltrimethyl-ammonium thiomolybdates
The alkyltrimethylammonium thiomolybdates were prepared according to the
method reported by Romero et al [30], which requires freshly prepared ATM (1.62 g,
6.22 mmol) dissolved in water (50 mL) with stirring. An aqueous solution (12.44 mmol
in 50 mL water) of alkyltrimethylammonium chloride, R(CH3)3NBr (where R = dodecyl,
tetradecyl, and octadecyl) or R(CH3)3NCl (where R = hexadecyl) was then added and the
mixture stirred for 1 h at room temperature. The resulting precipitate was separated by
vacuum filtration, washed first with cold water and then with isopropyl alcohol. Each
solid was dried at room temperature and kept under N2 atmosphere.
Page 5 of 27

2.4 Preparation of dodecyltrimethyl, tetradecyltrimethyl, hexadecyltrimethyl and
octadecyltrimethyl-ammonium thiotungstates
Each alkyltrimethylammonium thiotungstate salt was prepared in a procedure
analogous to the synthesis of the corresponding alkyltrimethylammonium-thiomolybdate
[30]. Thus, freshly prepared ATT (2.17g, 6.25 mmol) was dissolved in water (50 mL)
with stirring, then an aqueous solution (12.50 mmol in 50 mL water) of
alkyltrimethylammonium bromide, R-N(CH3)3Br (where R= dodecyl, tetradecyl, and
octadecyl) or R-N(CH3)3Cl (where R= hexadecyl) was added and the mixture stirred for 1
h at room temperature. The resulting precipitate was isolated by vacuum filtration,
washed first with cold water and then with isopropyl alcohol. The solid was dried at
room temperature and kept under N2 atmosphere.
2.5 Synthesis of (NH4)2[Co(MoS4)(WS4)] precursor.
Building on a procedure used by Müller et al [31-33], ATM (0.260 g, 1 mmol)
was dissolved in 100 mL of a CH3CN:CH3OH (2:3) solvent mixture, at room
temperature. Then ATT (0.350 g, 1 mmol) dissolved in 90 mL of a CH3CN:CH3OH (1:1)
solvent mixture was added to this solution, at room temperature. Under N2 atmosphere,
the thiometallate mixture was added dropwise to Co(NO3)2·6H2O (0.291 g, 1 mmol)
dissolved in 10 mL of a CH3CN:CH3OH (1:1) solvent mixture and acidified with acetic
acid (1 mL), immediately forming a dark brown precipitate. Once the addition of the
thiometallates was completed, the reaction mixture was stirred for 1 h at 0 °C. The
resulting precipitate, labeled as CoMoWP, was filtered, washed with isopropyl alcohol,
then air dried, and kept under N2 atmosphere.
2.6. Preparation of [R-N(CH3)3]2[Co(MoS4)(WS4)] precursors
Following the procedure described in Section 2.5, dodecyltrimethylammonium
thiomolybdate (0.681 g, 1.0 mmol) was dissolved in acetonitrile (30 mL), and added to a
solution of dodecyltrimethylammonium thiotungstate (0.769 g, 1.0 mmol) in acetonitrile
(20 mL), at room temperature. This thiometallate solution mix was then added dropwise
at 0 ºC and under N2 atmosphere to Co(NO3)2·6H2O (0.291 g, 1.0 mmol) dissolved in
Page 6 of 27

acetonitrile (10 mL), then acidified with acetic acid (1 mL); a dark brown precipitate was
produced immediately. Once the addition of the thiometallates was completed, the
reaction mixture was stirred for one hour.
The resulting precipitate,
dodecyltrimethylammonium thiomolybdatethiotungstate cobaltate (II), labeled
CoMoWP-C12, was filtered, washed with isopropyl alcohol, then air-dried, and kept
under N2 atmosphere.
The described procedure for the preparation of CoMoWP-C12 was used for the
synthesis of the CoMoWP-C14, CoMoWP-C16 and CoMoWP-C18 precursors (where
C14, C16 and C18 are the long chain alkyl groups of the ammonium ion), after
modifying with acetonitrile:methanol solvents and increasing the temperature at which
the thiometallate solution mix is prepared (equation 1), as described in Table 1.
CH3CN:CH3OH
N2, 0 °C, 1 h
[RN(CH3)3]2MoS4 + [RN(CH3)3]2WS4 + Co(NO3)2
(RN(CH3)3)2[Co(MoS4)(WS4)] + 2 RN(CH3)3NO3
where R= dodecyl, tetradecyl, hexadecyl and octadecyl
(1)
Table 1. Reaction conditions for the preparation of CoMoWP-Cn trimetallic precursors
[RN(CH3)3]2 MoS4
[RN(CH3)3]2 WS4
Precursor
T (ºC)
w(g) CH3CN:CH3OH / mL w(g) CH3CN:CH3OH / mL
CoMoWP-C12
CoMoWP-C14
CoMoWP-C16
CoMoWP-C18
25
25
50
60
0.681
0.737
0.793
0.849
1:0 / 30
1:1 / 70
7:3 / 100
9:1 / 50
0.769
0.825
0.881
0.937
1:0 / 20
1:1 / 50
1:1 / 50
9:1 / 50
Page 7 of 27

2.7 Characterization of trimetallic catalysts after HDS Reaction
Specific surface areas were measured with a Quantachrome AUTOSORB-1 by N2
adsorption at 77 K using the BET isotherm and a sample mass of 0.2 – 0.3 g. Samples
were degassed under flowing argon at 473 K for 2 h before N2 adsorption. The pore size
distribution was obtained from the desorption data, following the BJH method. The
mean standard deviation for the surface area measurements was about 2%.
The X-ray diffractograms for the prepared samples were obtained with a Philips
X’Pert analytical diffractometer for powder samples using Cu Kα radiation. Their phases
were identified with reference to the data base from the International Centre for
Diffraction Data.
A Jeol JSM5800 LV scanning electron microscope was used to perform
morphological analysis. Several zones were analyzed at different magnifications in order
to recognize the prevalent features. Scanning Transmission electron microscopy
micrographs and elemental analysis were done on Cs probe corrected Jeol JEM-2200FS
analytical electron microscope operated at 200 kV. All images were recorded and
evaluated (including Fourier analyses) with the Digital Micrograph (Gatan) software.
2.8 Catalytic activity and selectivity
The HDS of DBT was tested in a high pressure 300 mL Parr reactor by placing
4.4 g DBT, 100 mL of decalin and 100 mg of the trimetallic precursor. The reactor was
purged of residual air, pressurized with H2 to 3.1 MPa (450 psi) and then heated to the
reaction temperature of 623 K in about 10 min. A stirring rate of 600 rpm was used. The
progress of the reaction was monitored by gas chromatography in a HP 6890 gas
chromatograph collecting samples every 20 min during the first hour, then every 30 min
for the next four hours. Decrease of reaction mixture volume due to sampling was less
than 5% of total volume. The identity of the reaction products was confirmed by mass
spectrometry in a Agilent 6890 GC-MS, using a HP-5MS capillary column (30m X 0.25
mm X 0.25 μm).
Catalytic activity was expressed in terms of percentage conversion of DBT vs
reaction time, and from this data the reaction rate constant, k’, was calculated for each
catalyst using the integrated form of the first order rate law (equation 2).
Page 8 of 27

- ln (1-X) = k’t
(2)
where X is the DBT conversion fraction and t is the reaction time in minutes. A plot of
[-ln (1-X)] vs time, optimized by linear regression, yields the value of k’ as the slope of
the plot. The specific rate constant, k , for the catalyst was then calculated by equation 3





1
1



k=k' n
= mol/g.s


DBT 



60 wcat


(3)
where nDBT is the initial amount of DBT in moles and wcat is the weight of the recovered
catalyst in grams. The mean standard deviation for catalytic measurements is about
2.5%.
The HDS of DBT yields biphenyl (BP) through the direct desulfurization pathway
(DDS), and cyclohexylbenzene (CHB) and tetrahydrodibenzothiophene (THDBT)
through the hydrogenation pathway (HYD). Since these two pathways are parallel [22],
the selectivity (HYD)/(DDS) was approximately calculated by the equation 4:
HYD DDS  CHB  THDBT
 
BP

 



(4)
3. Results and discussion
3.1 Elemental analysis
The elemental analysis of the in situ-prepared sulfide catalysts was determined by
EDS and results are reported in Table 2. The Co:Mo:W atomic ratios, for all the
materials differ from the nominal 1:1:1 stoichiometry values, with a S/Mo ratio between
3.5 and 5.7 Considerable amounts of carbon (3.6≤C/Mo≤11.3) were also found for all
catalysts except for the CoMoWS, greater than those reported for another trimetallic
catalyst [34]. The presence of carbon is attributed to the decomposition of the aliphatic
chains of the tetraalkylammonium thiosalts (precursors), as suggested by other authors
[34-37]. Besides surface carbon, a sulfur carbide phase could also form -- where the
sulfur atoms are replaced by carbon atoms -- as observed in previous studies [38-40].
Page 9 of 27

Table 2. Specific surface areas, total pore volume and elemental analysis
determined by STEM for trimetallic catalysts CoMoWS-Cn.
Surface area
(m2/g)
Total pore volume
(cm3/g)
Catalyst
Composition
Co0.9MoW0.9S3.5
CoMoWS
12
213
73
0.037
0.053
0.052
0.053
0.047
CoMoWS-C12
CoMoWS-C14
CoMoWS-C16
CoMoWS-C18
Co0.8MoW1.5S5.7C11.3
Co0.9MoW1.4S4.2C3.6
Co0.9MoW1.3S5.0C5.0
Co0.8MoW0.9S3.7C8.4
340
277
3.2 Textural properties
The specific surface areas and total pore volume values of trimetallic catalysts
CoMoWS-Cn are reported in Table 2. Surface areas are between 12 and 340 m2/g, mostly
greater than the 11 – 52 m2/g reported by Huirache, et al [34].
The average total pore volume is around 0.050 cm3/g. Neither surface areas nor total
pore volumes correlate with the size of the precursor’s alkyl group. Average pore
diameters are in the range 24-34 Å. The adsorption isotherms are shown in Fig. 1. All
catalysts present a Type IV isotherm with desorption curves characteristic of mesoporous
materials.
Page 10 of 27

360
320
280
240
200
160
120
80
CoMoWS
CoMoWS-C12
CoMoWS-C14
CoMoWS-C16
CoMoWS-C18
40
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Relative pressure (P/Po)
Fig. 1. N2 adsorption-desorption isotherms of the trimetallic
catalysts CoMoWS-Cn
The CoMoWS catalyst has a very narrow hysteresis loop and thus scarcely
contributes to the porosity of the material, as evidenced by its surface area (Table 2),
which is the lowest in the series. The CoMoWS-C12 catalyst also behaves like a poorly
developed porous system, although not as pronounced as CoMoWS, which is consistent
with the presence of a lower amount of carbon. The CoMoWS-C14, CoMoWS-C16 and
CoMoWS-C18 catalysts have better developed porous systems, attributed to the
precursor’s carbon content.
The hysteresis loop of CoMoWS, CoMoWS-C12, CoMoWS-C14 and CoMoWS-C18
is closer to the H3 type characteristic of slit-like pores or spaces formed for plate-like
particles. The CoMoWS-C16 exhibits a hysteresis loop closer to the H2 type, associated
with so called ink-bottle pores, composed of a cylindrical pore closed at one end and with
a narrow neck at the other.
Page 11 of 27

Save for the CoMoWS material, which is scarcely porous, all catalysts are found been
mostly mesoporous, with average pore diameters in the range of 14-42 Å.
3.3 X-ray diffraction
The XRD patterns for the trimetallic catalysts CoMoWS-Cn are shown in Fig 2. For
the CoMoWS, produced from a precursor that does not contain an aliphatic chain, several
broad peaks are observed, with 2θ values at 12.94° (002), 33.79° (101) and 58.67° (110).
The pattern is very similar to that of the hexagonal phases of 2H-MoS2 and 2H-WS2
(JCPDS-ICDD 37-1492, 8-0237). The broad peaks are characteristic of poorly
crystalline materials, such as molybdenum sulfide derived from the decomposition of
thiosalts [41].
(002)
Co9S8
(101)


(110)
a)
b)
c)
d)
e)
5
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80

Fig. 2. XRD patterns of trimetallic catalysts CoMoWS-Cn:
a) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14,
d) CoMoWS-C16 and e) CoMoWS-C18
Page 12 of 27

The low intensity (002) reflection implies a low number of stacking layers in the “c”
direction of the catalyst particles. Moreover, narrow peaks at 2θ = 29.89° (311) and
52.13° (440) which are clearly observed in CoMoWS, are attributed to crystalline cobalt
sulfide phase Co9S8 (JCPDS-ICDD 2-1459).
In the diffraction patterns of the CoMoWS-C12, CoMoWS-C14, CoMoWS-C16 and
CoMoWS-C18 derived from precursors that contain aliphatic chains of 12, 14, 16 and 18
carbons, respectively, the (002) reflection does not appear, suggesting they are highly
dispersed or exfoliated materials [42, 43]. This dispersion could be attributed to the
catalyst’s carbon content derived from the decomposition of the precursor’s aliphatic
chain and the solvent (decalin). An analogous behavior has been reported for other MoS2
and WS2 systems where the carbon is considered to be the dispersing agent [44, 45]. The
destacking effect of the (002) reflection has also been observed in the unsupported
trimetallic sulfide system reported by Nava et al. [5], who find a low dispersion of the
(002) reflection.
All catalysts in the series present a couple of very low-intensity broad peaks, at
around 2θ = 34° and 58°, corresponding to the (101) and (110) reflections of MoS2 and
WS2, which are associated with materials of very low crystallinity. In contrast to
previous studies [46, 47] where binary systems exhibited well-defined cobalt sulfide
diffraction patterns, peaks belonging to the CoS or Co9S8 phases are barely observed,
which suggest that the cobalt content is highly dispersed.
3.4 Transmission electron microscopy
Bright field STEM micrographs of the resulting catalysts (Fig. 3) show nearly
spherical-shaped nanoparticles (<100 nm, according to the IUPAC classification),
forming clusters of an average particle size (APS) between 53 and100 nm, as determined
using the Gatan Digital Micrograph software.
Images at higher magnification of these trimetallic catalysts CoMoWS-Cn are shown
in Fig. 4. These images are similar to that reported for a NiMoW unsupported ex-situ
catalyst [48], showing highly disorganized curved fringes with only a few layers, as is
also seen in the MoS2/WS2 poorly crystalline materials [37, 49]. Some bending of the
Page 13 of 27

layers is observed, which is characteristically due to the structural carbon content of the
catalysts [44, 50].
Fig. 3. STEM micrographs showing clusters of catalyst nanoparticles for:
a) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14, d) CoMoWS-C16,
e) CoMoWS-C18
Table 3. Average particle size for the
CoMoWS-Cn catalysts
CoMoWS-Cn APS (nm)
σ
CoMoWS
100
66
70
84
53
17
17
11
15
9
CoMoWS-C12
CoMoWS-C14
CoMoWS-C16
CoMoWS-C18
Page 14 of 27

Fig. 4. TEM micrographs of the trimetallic catalysts CoMoWS-Cn:
a) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14,
d) CoMoWS-C16 and e) CoMoWS-C18.
3.5 Scanning electron microscopy
The morphology of the trimetallic catalysts CoMoWS-Cn is reported in Fig. 5. In
general, these images show porous materials with irregular agglomerated particles and
small cavities observed, which may result from the decomposition of the precursor’s
alkyl chains during the HDS of DBT, as discussed by Huirache et al. [34].
Page 15 of 27

Fig. 5. SEM micrographs of the trimetallic catalysts CoMoWS-Cn:
a) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14
d) CoMoWS-C16 and e) CoMoWS-C18.
3.6 Catalytic activity and selectivity
The results of the catalytic test for the trimetallic catalysts CoMoWS, CoMoWS-C12,
CoMoWS-C14, CoMoWS-C16 and CoMoWS-C18 are reported in the Table 4. The
analysis of the kinetic data finds that the reaction of HDS of DBT closely follows a first
order model during the first 5 h reaction time.
According to the GC analysis, the HDS of DBT yields mainly biphenyl (BP) through
the direct desulfurization (DDS) pathway, and cyclohexylbenzene
(CHB) and
tetrahydrodibenzothiophene (THDBT) through the hydrogenation (HYD) pathway.
Page 16 of 27

Cyclohexylcyclopentylmethane (CHCPM) and dicyclohexyl (DCH) can also be detected
in the HYD pathway.
Table 4. Gas chromatography analysis* of HDS reaction products, HYD/DDS selectivity
and kinetic data for trimetallic catalysts CoMoWS-Cn
CHCPM DCH
CHB
(%)
BP
(%)
THDBT
(%)
DBT
(%)
Conversion [HYD]
k x107
(mol/g.s)
Catalyst
CoMoWS
(%)
1.2
1.3
1.5
1.7
1.9
(%)
0.6
0.6
4.1
1.2
1.0
DBT (%)
[DDS]
10.8
8.1
22.8
28.2
51.3
44.4
43.4
4.8
4.8
2.0
2.4
2.8
59.8
57.0
25.1
32.7
36.6
40.2
0.7
164
172
421
377
235
CoMoWS-C12
CoMoWS-C14
CoMoWS-C16
CoMoWS-C18
43.0
0.5
16.0
17.5
14.1
74.9
0.4
67.3
0.4
63.4
0.4
* After 5h of reaction
The experimental selectivity of all the catalysts favors DDS (HYD/DDS < 1) and
it is mostly independent of the carbon content of the precursor. The highest DBT
conversion was attained with the CoMoWS-C14 catalyst, resulting from the in situ
decomposition of the CoMoWP-C14 precursor.
According to Table 4, the reference catalyst with no carbon (CoMoWS) has the
lowest specific rate constant, k = 164x10-7 mol/g·s. The specific rate constant increases
with the size of the alkyl groups of the precursors, resulting in a maximum k value
(421x10-7 mol/g·s) for the CoMoWS-C14 catalyst, after which the k value decreases.
The increase in k value from CoMoWS to CoMoWS-C14, does not correlate with an
analogous increase in surface area over the same catalysts. In fact, the surface area of the
CoMoWS-C14 catalyst is considerably less (73 m2/g) than any of the other catalysts
derived from the hydrocarbon sources. Moreover, no direct relation is found between
the surface area and the catalytic activity as is generally the case among anisotropic
transition metal sulfides [1], and bimetallic sulfides [51].
The increase in catalytic activity, especially in the case of CoMoWS-C14, is
attributed to the presence of a mix of CoMoS and CoWS phases at the border faces of the
Page 17 of 27

microcrystals [52], which result in a high dispersion of cobalt; another factor is the
particular nanostructure of the catalysts, which includes the presence of curved layers in
the catalyst due to structural carbon from the precursors, as described by several authors
[39, 53, 54]. The resulting specific rate constants are one order of magnitude greater than
those reported by Huirache et al [34], Gochi et al [55], and Olivas et al [35].
CONCLUSIONS
The in situ decomposition, during the HDS of DBT, of five novel trimetallic
precursors, one of them (NH4)2[(MoS4)(WS4)] and the rest of the type
(RN(CH3))2[Co(MoS4)(WS4)], yields the corresponding trimetallic CoMoW
nanostructured catalysts.
Their N2 adsorption-desorption isotherms are Type IV, characteristic of
mesoporous materials, with high surface areas.
XRD and TEM analysis find them to be highly dispersed materials.
Catalytic activity is not influenced by the size of the n-alkyl group of the
precursor, having the CoMoWS-C14 catalyst as the most active at k = 421x 10-7 mol/g.s.
The increase in catalytic activity is attributed to the presence of a mixture of
CoMoS and CoWS phases in the edges of the faces of the microcrystals, the dispersion of
which increased due to the structural carbon derived from the precursors. Microcrystals
with curved layers are also seen. The CoMoWS-C14 catalyst is a good candidate for the
deep HDS of other sulfur-containing compounds in the oils process.
Acknowledgements
We gratefully acknowledge technical assistance of C. Ornelas, E. Torres and L. de
la Torre at the National Nanotechnology Lab of CIMAV-Chihuahua, Mexico. Y.
Espinoza express gratefully for doctoral scholarship from CONACYT. G. Alonso
acknowledges funding from CONACyT Project 155388 and PAPIIT IN104714. J. Cruz-
Reyes and M. Del Valle acknowledge the economic support of UABC (13° y 14°
Convocatoria Interna de Apoyo a Proyectos de Investigación).
Page 18 of 27

References
[1] O. Weisser, S. Landa (1973) Sulfide Catalysts: Their Properties and Applications.
Pergamon Press, Oxford.
[2] P. Grange, Catal. Rev. Sci. Eng. 21 (1980) 135-181.
[3] M. Zdrazil, Catal. Today 3 (1988) 269-365.
[4] H. Topsoe, B.S Clausen, F.E Massoth, Hydrotreating Catalysis Science and Tech-
nology (Anderson JR, Boudart M, Eds.), Springer-Verlag, Germany (1996) Vol. 11.
[5] H. Nava, F. Pedraza, G. Alonso, Catal. Lett. 99 (2005) 65-71.
[6] T. Kabe, W. Qian, A. Ishihara, J. Catal. 149 (1994) 171-180.
[7] W. Qian, A. Ishihara, S. Ogawa, T. Kabe, J. Phys. Chem. 98 (1994) 907-911.
[8] R. Prins, Adv. Catal. 46 (2001) 399-464.
[9] K.C Pratt, J.V Sanders, N. Tamp, J. Catal. 66 (1980) 82-92.
[10] J.V Sanders, K.C Pratt, J. Catal. 67 (1981)331-347.
[11] G. Hagenbach, Ph. Courty, B. Delmon, J. Catal. 23 (1971) 295-300.
[12] H. Ishihara, T Itoh., Y. Hino, M. Nomura, P. Qi, T. Kabe, J. Catal. 140 (1993) 184
-189.
[13] G. Hagenbach, Ph. Courty, B. Delmon, J. Catal. 31 (1973) 264-273.
[14] R. Candia, B.J. Clausen, H. Topsoe, Bull. Soc. Chim. Belg. 90 (1981) 1225-1232.
[15] A.A. Al-Hilli, B.L. Evans, J. Crystal Growth 15 (1972) 93-101.
[16] Y.Y. Peng, Z.Y. Meng, C. Zhong, J. Lu , W.C. Yu, Z.P. Yang, Y.T. Qian, J. Solid
State Chem. 159 (2002) 170-173.
[17] Y.Y. Peng, Z.Y. Meng, C. Zhong , J. Lu, W.C. Yu, Y.B. Jia, Y.T. Qian, Chem.
Lett. 30 (2001) 772-773.
[18] W.J. Li, E.W. Shi, J.M. Ko, H. Ogino, T. Fukuda, J. Cryst. Growth 250 (2003) 418-
422.
[19] E. Devers , P. Afanasiev, B. Jouguet, M. Vrinat, Catal. Lett. 82 (2002)13-17.
[20] N. Rueda, R. Bacaud, M. Vrinat, J. Catal. 169 (1997) 404-406.
[21] K.S. Liang, R.R. Chianelli, F.Z. Chien, S.C. Moss, J Non-Cryst Solids 79 (1986)
251-273.
[22] S. Fuentes, G. Diaz , F. Pedraza, H. Rojas, N. Rosas, J. Catal 113 (1988) 535-539.
[23] A.W. Naumann, A.S. Behan, U.S. Patent 4,243,554 (1981).
Page 19 of 27

[24] S.L. Soled, S. Miseo, R. Krikak, H. Vroman , T.C. Ho, K.L. Riley, U.S. Patent
6,299,760 (2001)
[25] M.M. Mdleleni, T. Hyeon, K.S. Suslick, J. Am. Chem. Soc. 120 (1998) 6189-6190.
[26] Y. Iwata, K. Sato, T. Yoneda, Y. Miki, Y. Sugimoto, A. Nishijima, H. Shimada,
Catal. Today 45 (1998) 353-359.
[27] G.A. Camacho-Bragado, J.L. Elechiguerra, A. Olivas, S. Fuentes, D. Galvan,
M.J. Yacaman, J. Catal. 234 (2005) 182-190.
[28] G. Kruss, Justus Liebigs Ann. Chem. 225 (1884) 1-57.
[29] K. Ramanathan, W.S. Weller, J. Catal. 95 (1985) 249-259.
[30] R. Romero-Rivera, A.G. Camacho, M. Del Valle, G. Alonso, S. Fuentes, J. Cruz-
Reyes, Top. Catal. 54 (2011) 561-567.
[31] A. Müller, E. Diemann, J. Chem. Soc. D (1971) 65a-65a.
[32] A. Müller, E. Diemann, H. Heinsen, Chem. Ber. 104 (1971) 975-980.
[33] A. Müller, M.C. Chakravorti, H. Dornfeld, Z. Natureforsch 30B (1975) 162-164.
[34] R. Huirache-Acuña, M.A. Albiter, C. Ornelas, F. Paraguay-Delgado, R. Martínez-
Sánchez, G. Alonso-Núñez, App. Catal. A 308 (2006) 134-142.
[35] A. Olivas, D.H Galván, G. Alonso, S. Fuentes, App. Catal. A 352 (2009) 10-16.
[36] G. Alonso, M.H Siadati, G. Berhault, A. Aguilar, S. Fuentes, R.R Chianelli, Appl.
Catal. A 263 (2004) 109-117.
[37] R. Romero-Rivera, G. Berhault, G. Alonso-Nuñez, M. Del Valle, F. Paraguay-
Delgado, S. Fuentes, S. Salazar, A. Aguilar, J. Cruz-Reyes, Appl. Catal. A 433
(2012) 115-121.
[38] A) G. Alonso, M. Del Valle, J. Cruz, A. Licea Claverie, V. Petranovskii, S. Fuentes,
Catal. Lett. 52 (1998) 55-61;
[39] G. Berhault, A. Mehta, A.C. Pavel, J. Yang, L. Rendon, M.J. Yacaman, L. Cota, A.
Duarte, R.R. Chianelli, J. Catal. 198 (2001) 9-19;
[40] V. Schwartz, V. Texeira da Silva, S.T. Oyama, J. Mol. Cat. A 163 (2000) 251-
268.
[41] K.S. Liang, R.R. Chianelli, F.Z. Chien, S.C. Moss, J. Non-Cryst. Solids 79 (1986)
251-273.
[42] M. Del Valle, J. Cruz-Reyes, M. Avalos-Borja, S. Fuentes, Catal. Lett. 54 (1998) 59-
Page 20 of 27

63.
[43] P. Afanasiev, G.F. Xia, G. Berhault, B. Jouguet, M. Lacroix, Chem. Mater.
11 (1999) 3216-3219.
[44] S.P. Kelty, G. Berhault, R.R. Chianelli, Appl. Catal. A 322 (2007) 9-15.
[45] X. Jin, C. Ma, Y. Yi, Q. Zhang, J. Qiu, C. Liang, J. Phys. Chem. Solids 71
(2010) 642-646.
[46] R.W. Phyllips, A.A. Fote, J. Catal. 41 (1976) 168-172.
[47] S. Gobolos, Q. Wu, J. Ladriere, F. Delannay, B. Delmon, Bull. Soc. Chim. Belg.
93 (1984) 687-696.
[48] B.S. Zhang, Y.J. Yi, W. Zhang, C.H. Liang, D.S. Su, Mater. Charact.62 (2011) 684-
690.
[49] R.R. Chianelli, E.B. Presstidge, T.A. Pecoraro, J.P. Deneufville, Science 203 (1979)
1105-1107.
[50] G. Berhault, L. Cota Araiza, A. Duarte Moller, A. Metha, R.R. Chianelli, Catal.
Lett. 78 (2002) 81-90.
[51] R.R. Chianelli, G. Berhault, Catal. Today 53 (1999) 357-366.
[52] C. Wivel, R. Candia, B.S. Clausen, S. Morup, H.Topsoe, J. Catal. 68 (1981) 453-
463.
[53] A. Nogueira, R. Znaiguia, D. Uzio, P. Afanasiev, G. Berhault, Appl. Catal. A
429 (2012) 92-105.
[54] W. Iwata, Y. Araki, K. Honna, Y.Miki, K. Sato, H. Shimada, Catal. Today 65 (2001)
335-341
[55] Y. Gochi, C. Ornelas, F. Paraguay, S. Fuentes, L. Alvarez, J.L.Rico, G. Alonso-
Núñez G, Catal. Today 107 (2005) 531-536.
Page 21 of 27

FIGURES
360
320
280
240
200
160
120
80
CoMoWS
CoMoWS-C12
CoMoWS-C14
CoMoWS-C16
CoMoWS-C18
40
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Relative pressure (P/Po)
Y. Espinoza et al
Fig. 1
Page 22 of 27

(002)
Co9S8
(101)


(110)
a)
b)
c)
d)
e)
5
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80

Y. Espinoza et al
Fig. 2
Page 23 of 27

Y. Espinoza et al
Fig. 3
Page 24 of 27

Y. Espinoza et al
Fig. 4
Page 25 of 27

Y. Espinoza et al
Fig. 5
Page 26 of 27

FIGURE CAPTIONS
Fig. 1. N2 adsorption-desorption isotherms of the trimetallic catalysts CoMoWS-Cn
Fig. 2. XRD patterns of trimetallic catalysts CoMoWS-Cn:
b) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14,
d) CoMoWS-C16 and e) CoMoWS-C18
Fig. 3. TEM micrographs showing clusters of catalyst nanoparticles for:
b) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14, d) CoMoWS-C16,
e) CoMoWS-C18
Fig. 4. TEM micrographs of the trimetallic catalysts CoMoWS-Cn:
a) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14,
d) CoMoWS-C16 and e) CoMoWS-C18.
Fig. 5. SEM micrographs of the trimetallic catalysts CoMoWS-Cn:
a) CoMoWS, b) CoMoWS-C12, c) CoMoWS-C14
d) CoMoWS-C16 and e) CoMoWS-C18.
Page 27 of 27