Effect of the presence of ionic liquid during the NiMoS bulk preparation in the transformation of decanoic acid

Article abstract of DOI:10.1016/j.apcata.2016.12.020

The impact of the presence and amount of [BMIM][NTf2] ionic liquid during the preparation of bulk NiMoS catalysts was investigated. It was clearly shown that these factors have a strong influence on both the morphology and specific surface area of the obtained NiMoS samples. Most interestingly the catalytic activity for the transformation of decanoic acid increased up to three times when IL was present during synthesis. In the same time, a greater selectivity towards hydrocarbons was observed. On the whole a clear relationship between catalytic activity, selectivity and NiMoS morphology was demonstrated. Consequently, it is possible to modify the morphology of the materials and impact the catalytic properties by changing the synthesis conditions.

Full text of DOI:10.1016/j.apcata.2016.12.020

Accepted Manuscript
Title: Effect of the presence of ionic liquid during the NiMoS
bulk preparation in the transformation of decanoic acid
´
Author: Geraldine Leyral Soizic Brillouet Julie Rousseau
´ ´
´
Frederic Richard Anne Sophie Mamede Laurence Courtheoux
Annie Pradel Michel Ribes Sylvette Brunet
PII:
S0926-860X(16)30623-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.12.020
APCATA 16105
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
30-8-2016
25-11-2016
21-12-2016
´
´ ´
Please cite this article as: Geraldine Leyral, Soizic Brillouet, Julie Rousseau, Frederic
´
Richard, Anne Sophie Mamede, Laurence Courtheoux, Annie Pradel, Michel Ribes,
Sylvette Brunet, Effect of the presence of ionic liquid during the NiMoS bulk
preparation in the transformation of decanoic acid, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2016.12.020
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.

Effect of the presence of ionic liquid during the NiMoS bulk preparation in the
transformation of decanoic acid.
Géraldine Leyral1, Soizic Brillouet2, Julie Rousseau2, Frédéric Richard2, Anne
Sophie Mamede3, Laurence Courthéoux1, Annie Pradel1, Michel Ribes1, and
Sylvette Brunet2*
1
Université de Montpellier, Institut Charles Gerhardt, UMR 5253 CNRS, cc 1503, Place E.
Bataillon, 34095 Montpellier, France
2 Université de Poitiers, Institut de Chimie des Milieux et Matériaux de Poitiers (IC2MP), UMR
7285 CNRS-, 4 rue Michel Brunet, TSA 71106- 86073 Poitiers Cedex9, France
3 Univ. Lille, CNRS, ENSCL, Centrale Lille, Univ. Artois, UMR 8181 – UCCS – Unité de
Catalyse et Chimie du Solide, F-59000 Lille, France
* Corresponding author: email sylvette.brunet@univ-poitiers.fr (Sylvette Brunet)
Tel: (+33) 5 49 45 36 27
Fax: (+33) 5 49 45 38 97
1

Graphical abstract
Highlights

Control of the morphology of unsupported NiMoS catalysts by the presence of [BMIM][NTf2]
ionic liquid



Increase of the specific surface area of unsupported NiMoS catalysts
Increase of the catalytic activity for the transformation of decanoic acid
Higher selectivity towards hydrocarbons
Abstract
The impact of the presence and amount of [BMIM][NTf2] ionic liquid during the preparation
of bulk NiMoS catalysts was investigated. It was clearly shown that these factors have a strong
influence on both the morphology and specific surface area of the obtained NiMoS samples.
Most interestingly the catalytic activity for the transformation of decanoic acid increased up to
three times when IL was present during synthesis. In the same time, a greater selectivity towards
hydrocarbons was observed. On the whole a clear relationship between catalytic activity,
selectivity and NiMoS morphology was demonstrated. Consequently, it is possible to modify
the morphology of the materials and impact the catalytic properties by changing the synthesis
conditions.
Keywords: ionic liquid, [BMIM][NTf2], hydrodeoxygenation, decanoic acid, NiMoS,
characterizations
2

1
Introduction
Ni or Co-promoted MoS2 catalysts used either in a bulk form or in a supported one
mainly on alumina are commonly used in hydrotreament processes (hydrodesulfurization,
hydrodenitrogenation and hydrodeoxygenation) involving the removal of heteroatoms (sulfur,
nitrogen and oxygen) from petroleum feedstocks [1]. These processes need to become more
and more efficient due to continual reinforcement of regulations to obtain cleaner and cleaner
fuels, a consequence of increasing environmental concerns. For example, the reduction of sulfur
content in gasoline and diesel cuts through hydrodesulfurization (HDS) processes is mandatory
to meet present (10 weight ppm of sulfur is imposed by the European Community since 2009
(98/70/EC Directive)) [2] and future worldwide regulations (e.g., forthcoming China V and US
Tier 3 regulations will impose a 10 weight ppm maximum of sulfur in gasoline in 2017) [3].
On the other hand, the diversification of the raw material for the production of fuels
leads to the development of new processes involving lipidic or lignocellulosic biomass, which
are available in large quantities. The bio-oil obtained after the liquefaction step, involving a
pyrolysis step, contains a high level of oxygen compounds (15 – 40 wt %) present in various
functionalities such as carboxylic acids, ketones, ethers, alcohols or phenol [4-5,6,7,8,9]. This
high oxygen content leads to a low stability and a high corrosiveness of the resulting bio-oil
which cannot be used as a gas oil [4]. A deoxygenation process is thus required. It could be
carried out in dedicated units [10-11,12] or benefits from the HDS units used for the
hydrodesulfurization step of crude cuts already existing on refinery sites which, of course,
would be economically more interesting [13-14,15,16]. Performances of NiMoP/Al2O3 and
CoMoP/Al2O3 catalysts were compared for the transformation of fatty acid triglycerides by
hydrotreating a blend of straight-run diesel fraction and sunflower oil [17]. The NiMoP/Al2O3
catalyst was more active than CoMoP/Al2O3. The authors explained this difference of behavior
3

by an increase of the average particle length of the active phase in the case of CoMoP/Al2O3
compared to NiMoP/Al2O3.
Owing to the new drastic regulations and the large nature and compositions of the new
feedstock, an optimization of the sulfide catalysts based on a clearer understanding of the
underlying properties that control the catalyst activity would be greatly beneficial.
These catalysts are traditionally prepared by incipient wetness impregnation of oxidic
metal precursors on the (alumina) support followed by a sulfidation step. Different parameters
are key factors in the performances of catalysts during the hydrotreatment process: the
preparation, the activation step by a sulfuring agent (H2S, organosulfides) and the support for
the supported catalyst. In conventional catalysts, well dispersed MoS2 nanocristallites with
molybdenum atoms substituted by nickel or cobalt as promoter atoms compose the CoMoS or
NiMoS mixed active phases. DFT calculations (Density Functional Theory) allowed to describe
the structure of non-promoted catalyst MoS2 as having an hexagonal form with two different
types of edges: a sulfur edge (S-edge) and a metal edge (M-edge) [18]. In promoted catalysts,
cobalt is present only on the S edges while nickel is found on both sulfur and metal edges
[19,20,21,22]. The active sites would be sulfur vacancies or CUS (Coordinatively Unsaturated
Sites) present in the mixed CoMoS or NiMoS phases [23]. “Brim sites" proposed by Daage and
Chianelli [24] and Topsoe et al. [25] are supposed to be located on the edges of MoS2 clusters
fully saturated with sulfur which were involved in hydrogenation steps of hydrodesulfurization
reactions.
Raybaud et al. [26,27,28], Paul et al. [29] and Lauritsen et al [30] showed that the
morphology of the MoS2-slabs and the stability of edge sites are sensitive to sulfidation
conditions and more particularly to the PH2S/PH2 partial pressure ratio, which can be used as
adjustable parameters. However, on the whole, the conventional method based upon
impregnation and further sulfidation step leads to poor dispersion and limited metal loading in
4

contradiction with new activity requirements. To overcome these limitations, a direct synthesis
of the sulfide phase with sulfur precursors (i.e. without the sulfidation step of metal oxide
intermediates) can be considered as reported by Fuentes et al for example [31]. However, the
obtained bulk catalysts had low specific surface areas [32].
In order to get larger specific surface areas, the structuration of bulk MoS2 catalysts with
organic compounds such as surfactants, polymers or ionic liquids can be considered. Such
organic compound addition during synthesis of unpromoted-MoS2 has already been
investigated. For example, Ma et al. [33] prepared MoS2 in water in the presence of ionic liquid.
It was shown that the morphology and size of the obtained particles depended on the nature and
concentration of the IL. Spherical particles were indeed obtained when [BMIM][BF4] was used
while flower-like particles were produced in the presence of [BMIM][PF6]. The probable
organization of IL in water due to the presence of hydrophobic BMIMM+ cations would lead
to the formation of vesicles which forms would depend on the IL concentration. The specific
shape of the particles would result from the interaction of Mo-precursors with the positively
charged nitrogen.
Very few papers report on the attempt to produce Ni or Co-promoted MoS2 in the
presence of organic compounds and even less in the presence of IL. The work by Genuit et al
[34] reports on the preparation of efficient bulk Co(Ni)MoS catalysts obtained with nonionic
surface-active agent (poly(ethylene oxide) oligomers with aryl-alkyl tail), in a mixed
water/ethylene glycol solution, leading to a specific surface area of 122 m2/g. A gain in catalytic
activity for the transformation of alkyldibenzothiophene (a model molecule representative of
the most refractory compounds contained in diesel cuts) was observed. A clear description of
the mechanisms at the basis of the polymer-metal precursor interactions was not proposed. The
only paper that reported on the synthesis of NiMoS catalysts in presence of an IL, namely an
alkylpyridinium-type IL, is due to Trikalitis et al. [35]. One can also cite the paper by Sizova et
5

al [36] who prepared nanocatalysts of nickel tungsten sulfide in ionic liquid 5BMPip°2Ni(WS4)2
and nickel tungstate complex as precursor and obtained catalyst particles of 0.5 nm wide
nanoplates associated in agglomerates with a size of 100-150 nm.
Therefore, very few data on the preparation of Ni-promoted MoS2 catalysts in the
presence of IL was available even though their interest had been demonstrated in the case of
un-promoted catalysts. We recently reported on a first series of data on the preparation of Ni-
promoted MoS2 catalysts in the presence of ionic liquid and their physicochemical
characterization in Leyral et al [37]. At the time the catalytic properties were not reported.
The present paper focus on the characteristics and catalytic properties of unsupported NiMoS
catalysts prepared in the presence of different amounts of ionic liquid. Physicochemical
characterization including SEM, TEM, XPS, BET experiments were carried out and the
transformation of decanoic acid, a model molecule for oxygenated compounds present in the
oil after pyrolysis of lignocellulose materials, in the presence of the NiMoS catalysts, was
investigated.
2
Experimental
2.1
Chemicals
The ionic liquid [BMIM][NTf2] (99%) was purchased at SOLVIONIC. Ni(NO3)2·6H2O
(99.99 %), (NH4)2MoS4 (99.97 %), formamide (99.5 %), and ethanol were obtained from
Aldrich. Decanoic acid (98%) was purchased from SAFC, toluene (99.8%) from Sigma–
Aldrich, n-heptane (99.7%) from Carlo Erba, and dimethyl disulfide (DMDS, 98%) from Fluka.
All products were used without further purification. H2 (99.95%) was provided by Air Liquide.
6

2.2
Catalyst preparation
NiMoS catalysts were prepared as reported previously [37]. (NH4)2MoS4 (4 mmol) and
Ni(NO3)2·6H2O (4 mmol) were separately dissolved in Ar-degassed formamide. When
necessary, the ionic liquid (IL) was added to the (NH4)2MoS4 solution, with various reactant/IL
molar ratios of 1:0.5, 1:1, 1:10. The two solutions were then mixed with soft stirring for 30 s.
The initial formamide volume was adapted to obtain a final volume of 82 mL. A similar
synthesis was performed in pure ionic liquid. However, note that, due to the low solubility of
the inorganic precursors in [BMIM][NTf2], addition of a small amount of formamide (2 mL)
could not be avoided. Also, 2.7 mmol of reactants (compared to 4mmol in the first syntheses)
were used to limit the quantity of solvent (82 mL); that corresponds to a reactant/IL molar ratio
of 1:100. After mixing, the reaction mixture was stored under an atmosphere of argon, at 6°C
for 8 days. The products were then washed by using degassed solvents, that is, first water and
then ethanol. The products were then extracted by centrifugation with a centrifuge 1–6P from
Sigma. For each solvent, three washes were carried out (15 mL of solvent, 5600 rpm, 1 h). All
the products were then dried under vacuum for 24 h. The products were heated in an Ar
atmosphere from room temperature to 500 °C at 1 °C min–1, held 2 h at 500 °C, and cooled
down to room temperature at the same rate. Repeatability of the synthesis was checked by
reproducing the experiments at least three times. A series of similar synthesis carried out while
replacing [BMIM][NTf2] by [BMIM][BF4] was carried out and led to similar results. Thus,
they won’t be discussed any further.
2.3
Characterizations
BET surface areas and textural analyses were determined using a Micromeritics ASAP
2010. Elemental analysis (sulfur and carbon amounts) was carried out using a CE Instruments
NA2100 Protein while the molybdenum and nickel amounts were measured by X fluorescence
(XF) using a Bruker S4 explorer with dispersive wavelengths. X-ray powder diffraction (XRD)
7

patterns were recorded with a Philips X’pert MPD X-ray diffractometer by using Cu radiation
(λ = 1.5406 Å). Raman spectra were obtained with a LabRAM ARAMIS spectrometer (laser
source: 633 nm), and morphologies were observed by using a Hitachi S-4800 scanning electron
microscope (SEM) and a JEOL JEM2100F transmission electron microscope (TEM) equipped
with an EDX analyzer.
XPS spectra were recorded using a Kratos Analytical AXIS UltraDLD spectrometer equipped
with Al K source (hν = 1486.6 eV). The analyser was operated in constant pass energy of 40
eV using an analysis area of approximately 700 m x 300 m. -NiMoS catalysts were packed
in shlenk under argon to avoid sulphate formation. The adventitious C 1s (284.8 eV) binding
energy (BE) was used as internal reference. Quantification and decomposition of the
experimental Ni 2p, Mo 3d and S 2p photopeaks were carried out using CasaXPS software [38]
using a linear background subtraction. The obtained signal can be decomposed into different
contributions relative to the different chemical species. For example, in the case of Ni atoms
detected as Ni(II), NiMoS and NiS species can be identified with the calculation of the
respective contributions leading to the relative proportion of each chemical species. The XPS
elemental composition has an accuracy of 10% and the detection limit is known to be 0.1
atomic %.
2.4
Reaction conditions
The particle size of the catalysts was in the range of 250–315 µm. The catalysts were diluted
with carborundum to keep the volume of the catalyst bed constant and in order to prevent
diffusion limitations.
Catalytic activity measurements were carried out in a fixed-bed reactor at 613 K under a total
pressure of 4 MPa [12]. The model feed was composed of decanoic acid (7.8 wt%), DMDS (1.3
wt%) and n-heptane (2.6 wt%) diluted in toluene. Under these conditions (613 K, 4 MPa), the
8

initial partial pressures were 53 kPa of decanoic acid, 32 kPa of both H2S and CH4 (generated
by the decomposition of DMDS), 31 kPa of n-heptane (used as an internal standard for
chromatography analysis) and 1.13 MPa of toluene. The volume ratio H2/feed was equal to 487
NL/L.
Owing to the high boiling point of the reactants, on line analysis of the reaction products was
not convenient. Consequently, the reactor effluents were condensed and liquid samples were
periodically collected to be analyzed by gas chromatography. The reactor effluents were
collected at 10 ◦C (using a Minichiller-Huber cryostat). The liquid samples, obtained after
drying over MgSO4, were periodically analyzed on a Varian 430 chromatograph equipped with
a DB1 capillary column (length: 30 m; inside diameter: 0.320 mm; film thickness: 5 µm) and a
flame ionization detector. The samples were silylated prior to their chromatographic analysis,
using N,O-bis(trimethylsilyl)trifluoroacetamide with 1% of chlorotrimethylsilane (BSTFA-
TMCS 99:1, provided by Supelco) as the silylation agent. The reaction products were identified
by GC/MS analysis using either a 1200 TQ mass spectrometer coupled with a Varian 3800
chromatograph or an Agilent 7200 Series Q-TOFGC/MS System coupled with an Agilent
7890A chromatograph and also by adding aliquots of commercial compounds.
The contact time defined as the ratio between the volume of the catalyst and the gas flow of the
oxygenated reactant was adapted in order to obtain a conversion of decanoic acid of 30% for
all the solids. The mass of the catalyst varied from 30 to 240 mg corresponding to a contact
time varying between 0.26 s and 1.29 s or a LHSV varying from 13846 h-1 and 2790 h-1.
Moreover, the various selectivities were compared at iso-conversion (30%). Assuming a first
order reaction, the total activity of each catalyst was calculated according to the following
equation:
퐹
A = − 푊푐푎푡 퐿푛 (1 − 푋)
9

where A is the global activity (mol h−1 g−1), F is the molar flow of the oxygenated reactant (mol
h−1), wcat the weight of the catalyst (g), X is the conversion of decanoic acid measured after 7 h
of reaction.
The reaction scheme shown in
Scheme 1 was well established previously [12].
The products of the transformation of decanoic acid on sulfide catalyst may be classified
according to the various possible ways of transformation. First, the transformation of decanoic
acid leads to the formation of hydrocarbons without loss of carbon atoms (decane and decene),
involving several oxygen (decanal,
1
-decanol) and sulfide (1-decanethiol,
decylméthylthioether) intermediates. This way where oxygen was removed only as water was
named “HDO”. The second way, named decarbonylation way (DEC) leads to the formation of
hydrocarbons (nonane and nonenes), after loss of a carbon atom as carbon monoxide. In this
way, the oxygen atom is removed as water and carbon monoxide. Others products were
classified as “others” (decyle decanoate, decyl thiométhanoate).
3
Results and discussion
3.1
Catalysts properties
Table 1 reports the specific surface area, elemental analysis and Ni/Ni+Mo molar ratio of
NiMoS catalysts with various NiMoS: IL ratios, i.e 1:0, 1:0.5, 1:1, 1:10 and 1:100. The specific
surface area of NiMoS synthesized without and with the lowest amount of ionic liquid (NiMoS
1: 0 and 1: 0.5) are less than 1 m² g-1. The addition of a larger amount of ionic liquid generated
specific surface area up to 15 m² g-1. These solids have specific surface area similar to those
reported in literature [10,32]. The isotherm features obtained by BET experiments correspond
to inter-particle capillary condensation. Carbon deposit is observed to a more or less great extent
for all samples. For the NiMoS sample coming from synthesis without IL the carbon content is
10

equal to 2.5 wt.% which can be explained by the presence of some EtOH residues and a heat-
treatment in an Ar atmosphere. The presence of a noticeable and almost constant content in
carbon, i.e. 6-7 wt%, is observed in all NiMoS samples prepared with IL. The presence of
carbon can be explained by the thermal decomposition of IL trapped within the materials,
during the heat treatment in an Ar atmosphere. The chemical analysis confirmed the presence
of Ni, Mo and S in all materials. While the amount of sulfur is rather constant and close to 35
wt.%, the amounts of molybdenum and nickel are more scattered with a tendency to increase
with the amount of ionic liquid used during the synthesis in the case of molybdenum. The
average Ni / (Ni + Mo) ratio is around 0.3 with a lower value (0.24) for NiMoS 1:0, the poorest
in Ni (5 wt.%), and a larger value (0.37) for NiMoS 1:0.5 which presents the highest content in
Ni (10 wt.%). These values are in agreement with those of the most efficient sulfide phases
generally used in hydrotreatment processes [39].
On the whole, the XRD patterns of the NiMoS samples are similar with the presence of MoS2
and NiS phases observed whatever the amount of IL used during the preparation of the samples
(Figure 1) [35]. However, the peaks attributed to a NiS phase are much weaker in the case of
the 1:0.5 and 1:10 samples while the broad signature attributed to MoS2 is better resolved in the
XRD pattern of the 1:0 sample. The presence of MoS2 phase in all samples was also confirmed
by Raman spectroscopy. Typical Raman spectra are shown in Figure 2 for NiMoS 1:10 and 1:0.
While the characteristic bands of MoS2 in the region 200-900cm-1 alone are observed for the
1:0 sample, additional bands in the region 1300-1600 cm-1 attributed to graphitic carbon were
also observed in the case of the 1:10 sample. Such result agrees with the chemical analysis
reported above which indicated the presence of noticeable amount of carbon in the samples
prepared with IL.
The scanning electron microscopy (SEM) images shown in Figure 3 clearly highlight the impact
of the IL on the morphology of the product. A dense product was obtained when no IL was used
11

during the synthesis (Figure 3a) while particles were obtained when IL was used (Figures 3c,
d, e). An intermediate morphology was observed for NiMoS 1:0.5 (Figure 3b). Depending upon
the synthesis conditions, the size and shape of the particles changed. With reactant/IL molar
ratios of 1:1 (Figure 3c) and 1:100 (Figure 3e), nanoparticles with sizes of some dozens of
nanometers were obtained. These particles formed aggregates with diameters between 100 and
300 nm. In contrast, the addition of an IL in a molar ratio of 1:10 led to spheres with diameters
ranging from 100 to 500 nm (Figure 3d). This evolution depends on both the amount of IL and
the IL/solvent ratio used during the synthesis. Needle- or hexagonal-shaped crystals or even
tiny crystallites (NiMoS 1/10) attributed to the NiS phase on the basis of the XRD patterns of
the samples were also observed in the SEM images of the NiMoS solids.
Interestedly, the evolution of morphology and size of the particles with the amount of IL
introduced in the reaction medium can be compared with that observed by Lian et al. [40] on
one hand and Fang et al. [41] on the other during the synthesis of Fe2O3 and SnO2, respectively,
in the presence of IL. On the whole, in the presence of small amounts of IL, tiny particles are
obtained which size increases for intermediate IL contents [40] while large amount of IL favors
the precipitation of tiny particles [41].
Many parameters among which low surface tension, self-organization, viscosity, can influence
the synthesis and it is difficult to claim which will be predominant. Antonietti et al. [42]
consider that the low surface tension of IL favors high nucleation rates and leads to the
formation of a great number of tiny particles. It would be the case in the synthesis when IL is
used as the solvent. Lian [40] explained the increase in the size of Fe2O3 particles at intermediate
IL content by the predominance of IL self-organization and consequent increase in the medium
viscosity. In such a case, the hindering diffusion effect of the IL combined with the localized
Ostwald ripening process would be critical for the increase of the size of the microspheres.
12

XPS analysis of all NiMoS samples has been carried out. According to the chemical analysis
presented in Table 2 and to the Mo 3d and S 2p XPS spectra decomposition shown as an
example for NiMoS 1:10 in Figure 4 and to the Ni 2p XPS spectra of all NiMoS samples in
Figure 5, the formation of MoS2, NiMoS, NiS and NiO phases was confirmed. Indeed, the
various MoS2 (Mo 3d5/2 BE = 229,2 eV; S 2p3/2 BE = 162,0 eV), NiS (Ni 2p3/2 BE = 853,4 eV),
NiO (Ni 2p3/2 BE = 856,4 eV) and mixed phase NiMoS (Ni 2p3/2 BE = 854,4 eV) species were
well identified. Quantitative analysis of XPS data helped in evaluating the S/Mo and Ni/Ni+Mo
atomic ratios as well as the sulfidation rate of molybdenum (proportion of MoS2 (%)) and the
amount of NiMoS mixed phase. The MoS2 phase (S/Mo = 2) was indeed formed and the amount
(around 69%) was similar whatever the NiMoS: IL ratio. In the same way, the Ni/Ni+Mo ratio
was close to 0.4, a value close to the one of the conventional NiMo/Al2O3 catalyst used in
hydrotreament processes. The mixed NiMoS phase was formed in a large amount (around 40-
50 %) while NiS and NiOx species were detected in all NiMoS catalysts, their amount
depending on the samples. On the whole, the synthesis method proposed in this work led to an
amount of NiMoS mixed phase in the material similar to the one observed for conventional
catalyst (45%) [32].
NiS and Ni promoted MoS2 formations were also pointed out by TEM analysis for all samples
as shown in Figure 6 and Figure 7 respectively. However, the size and morphology of NiS
particles were different depending on the amount of IL used for the preparation of the NiMoS
samples Table 3, Figure 6). Larger and/or more defined NiS particles were observed for NiMoS
1:1 and 1:100 which can be related to their XRD patterns showing very sharp NiS diffraction
peaks (Figure 1). At the opposite, NiMoS 1:0.5 and 1:10 which XRD patterns comprised very
weak NiS peaks are the samples with the smallest and/or less defined NiS crystals.
Agglomerated slabs of NiMoS2 were identified with an obvious spherical morphology as balls
in the case of NiMoS 1:10 and 1:100.
13

The length and stacking of slabs of MoS2 were estimated from 500 particles for each TEM
image for all NiMoS samples. Row data are shown in Figure 8 and Figure 9 while the average
length (L), average stacking (S) and Ni/(Ni+Mo) atomic ratio (measured by EDX) are reported
in Table 3. The average length is similar for all solids formed with a value between 6.1 and 7.1
nm. The average stacking also appears unaffected by the addition of ionic liquid during slab-
the synthesis. However, the distribution in different stacking with a larger number of 1 and 2
stacking is broader for samples prepared with IL as shown in Figure 9. It is in agreement with
a dismantling of the MoS2 slab-stacking in presence of IL as reported earlier [31]. The
calculated Ni/(Ni + Mo) ratios are close to 0.3 for all samples but NiMoS 1:0 in agreement with
the results obtained by fluorescence X and XPS and close to results obtained previously by
EDX [37].
3.2
Decanoic acid transformation over NiMoS catalysts
The catalytic performances of the NiMoS samples, i.e. their global activity and selectivities
(towards hydrocarbons, HDO and DEC pathways), for the transformation of decanoic acid
under hydrotreatment operating conditions (613 K, 4 MPa, H2/feed = 470 NL/L) were
compared. All the catalysts were stable under operating conditions in the presence of H2S
(generated by DMDS decomposition) and water, the main by-product of the transformation of
decanoic acid. No change, either in global activity or selectivity towards hydrocarbons i.e. the
final products (nonane, nonene, decene and decane) of deoxygenation, during time on stream
was noticed (Figure 10 Figure 11).
The presence of IL during the NiMoS preparation clearly improved the catalytic activity of the
obtained products (Figure 10). Indeed, the global activity increased with the amount of IL used
during the synthesis of NiMoS materials (corresponding to higher NiMoS :IL ratios), the
highest activity was obtained for NiMoS 1:100 (three time higher than that obtained for NiMoS
1:0). It also improved the selectivity towards the formation of hydrocarbons (total products of
14

deoxygenation: nonanes, nonenes, decenes and decane) with a selectivity in the range of 65-75
mol.% for NiMoS 1:1, 1:10 and 1:100 (Figure 11, Figure 12). The formation of other products,
either sulfides or oxygenated products, were mainly intermediates in the HDO way (Scheme
1). Compared to NiMoS 1:0 and 1:0.5, NiMoS 1:1, 1:10 and 1:100 clearly favoured the
formation of hydrocarbons to the detriment of sulfides. Two ways for the formation of
hydrocarbons, i.e. DEC and HDO, possibly occur during the transformation of decanoic acid
as shown in scheme 1. As seen in Figure 13, the DEC way was favoured for the formation of
the total deoxygenation products when the NiMoS catalysts were used. In the same time, the
formation of intermediates almost completely disappeared. This data, i.e. enhancement of
activity and selectivity for the catalysts prepared in the presence of IL, is interesting. According
to [12], the active sites are sulphur vacancies. Then their number and/or properties are probably
affected when IL is added during the preparation of NiMoS catalysts. In the following, we
attempted to correlate the catalyst performance with the characteristics of the NiMoS samples.
It is well known that the catalytic performance of NiMoS samples are often related to the
amount of mixed NiMoS phase present in the catalyst. In the present case and taking into
account the uncertainty in XPS measurements, the quantity of mixed phase can be considered
similar and large for all samples but NiMoS 1:10. So the differences in activity of NiMoS
prepared with different amounts of IL during preparation cannot be accounted for by increasing
amount of mixed NiMoS phase. Taking into account the specific surface area, a slight difference
was noticed between NiMoS 1:1 with the lowest activity per square meter and NiMoS 1:10 and
1:100. In the same way, considering molybdenum atom as the active site, the intrinsic activity
was higher for NiMoS 1:100 than for NiMoS 1:1 and 1:10.
Finally, if we refer to both the total and intrinsic activities (Figure 10 and Table 4) the most
active catalysts, i.e. NiMoS 1:1 and 1:100, are those which morphology corresponded to tiny
particles (in the range of dozens nm in diameter) and highest catalytic surface area. Intermediate
15

activity values were obtained for NiMoS 1:10 which comprised spherical particles of 100-300
nm in diameter. Finally, NiMoS 1:0 and 1:0.5 prepared in the absence –or in the presence of
very small amount- of IL were (quasi-) bulk products and their catalytic performances were
poor. Therefore, we have to conclude that the enhancement of activity relies mainly on the
morphology and the surface of the obtained NiMoS catalysts. These results are in agreement
with those obtained by Nikulshin et al [43] who found that the activity of CoMo/Cx/Al2O3
catalyst for the transformation of the HDO of guaiacol and oleic acid increased with the
decrease of size of particles of CoMoS phase.
A morphology with the presence of tiny particles could lead to higher specific surface area and
favor the accessibility of the active sites, and the formation of hydrocarbons via the DEC route.
4
Conclusion
Ni-promoted MoS2 catalysts are commonly used in hydrotreament processes. Owing to the new
drastic regulations and the large nature and compositions of the new feedstock (biomass), it
would be important to improve their performance. The aim of the work was to investigate the
influence of the presence and amount of [BMIM][NTf2] IL during the preparation of NiMoS
bulk catalysts on their physico-chemical and catalytical properties. It has been clearly shown
that the morphology and specific surface area of the NiMoS products depended on the presence
and amount of IL. Indeed in the absence or presence of very small amount of IL, (quasi)-bulk
solids with negligeable specific surface area were obtained while tiny or spherical particles were
obtained when larger amount of IL were used. Most interestly, catalytic tests based upon the
transformation of decanoic acid, a model molecule for the hydrotreatment of bio-oils showed a
higher catalytic activity and a larger selectivity towards hydrocarbons for the NiMoS catalysts
prepared in the presence of IL, the catalyst prepared with the largest amount of IL (NiMoS
1:100) having a catalytic activity three times larger than that of the catalyst prepared in the
16

absence of IL (NiMoS 1: 0). Finally, these results demonstrated that depending on the operating
conditions of the preparation of NiMoS phase it is possible to modify the morphology of the
material and to impact the catalytic performances (in terms of activity and selectivity).
Acknowledgements
This work was funded by the French ANR Programme National de Recherche - PASSCATA,
a joint project between the Centre National de la Recherche Scientifique (CNRS), the
universities of Montpellier and Poitiers and IFPEN Lyon. The Fonds Européen de
Développement Régional (FEDER), CNRS, Région Nord Pas-de-Calais and Ministère de
l’Education Nationale de l’Enseignement Supérieur et de la Recherche are acknowledged for
funding of XPS/LEIS/ToF-SIMS spectrometers within the Pôle Régional d’Analyses de
Surface.
17

References
1) T. Chapuis, D. Hudebine, V. Souchon,;1; Catalysis by transition metal sulfides : from
molecular theory to industrial application IFPEN-Lyon, Technip Edition, Paris, Chapter
3.2 (2013) 547-578.
2) Directive 2009/30/CE L140/88.
3) http://transportpolicy.net/index.php?title=Global_Comparison:_Fuels.
4) E. Furimsky, ;1; Appl. Catal. A Gen. 199 (2000) 147-190.
5) G.W. Huber, S. Iborra, A. Corma, ;1; Chem. Rev., 106 (2006) 4044-4098.
6) A.V. Bridgwater, ;1; Appl. Catal. A Gen. 116 (1994) 5-47.
7) D. Mohan, C.U. Pittman, P. Steele, ;1;Energy Fuels 20 (2006) 849-889.
8) L. Ingram, D. Mohan, M. Bricka, P. Steele, D. Strobel, D. Crocker, B. Mitchell, J.
Mohammad, K. Cantrell, C. Pittman, ;1; Energy Fuels 22 (2008) 614-625.
9) J. Zhang, H. Toghiani, D. Mohan, C. Pittman, R. Toghiani, ;1; Energy Fuels 21 (2007)
2373-2385.
10) M. Ruinart de Brimont, C. Dupont, A. Daudin, C. Geantet, P. Raybaud. ;1; J. Catal. 286
(2012) 153–164.
11) C. Bouvier, Y. Romero, F. Richard, S. Brunet;1;, Green Chemistry 14 (2011) 2441–
2451.
12) S. Brillouet, E. Baltag, S. Brunet, F. Richard, ;1; Appl. Catal. B Environmental. 148–
149 (2014) 201– 211.
13) A.Pinheiro, D. Hudebine, N. Dupassieux, C. Geantet, Energy Fuels 25 (2011) 804-812.
14) M.Philippe, F. Richard, D. Hudebine, S. Brunet, Appl. Catal. B Environmental.132–
133 (2013) 493– 498.
15) F. Pelardy, C. Dupont, C. Fontaine, E. Devers, A. Daudin, F. Bertoncini, P. Raybaud,
S. Brunet, Appl. Catal. B: Environmental 97 (2010) 323–332.
18

16) M.Philippe, F. Richard, D. Hudebine, S. Brunet, Appl. Catal. A Gen. 383 (2010) 14-23.
17) P. A. Nikul’shin, V. A. Sal’nikov, Al. A. Pimerzin, Yu. V. Eremina, A. S.
Koklyukhin,V. S. Tsvetkov, and A. A. Pimerzin. Petroleum Chemistry. 56 (2016) 56–
61
18) P. Raybaud, Appl. Catal. A : General 322 (2007) 76.
19) H. Schweiger, P. Raybaud, G. Kresse, H. Toulhoat, J. Catal. 207 (2002) 76.
20) H. Schweiger, P. Raybaud, H. Toulhoat, J. Catal. 212 (2002) 33.
21) M. Sun, J. Adjaye, A.E. Nelson, Appl. Catal. A : General 263 (2004) 131.
22) A.D. Gandubert, E. Krebs, C. Legens, D. Costa, D. Guillaume, P. Raybaud, Catal.
Today 130 (2008) 149.
23) P.G. Moses, B. Hinnemann, H. Topsoe, J. K. Norskov, J. Catal. 268 (2009) 201.
24) M. Daage, R.R. Chianelli, Journal of Catalysis 149 (1994) 414-427
25) P.G. Moses, B. Hinnemann, H. Topsoe, J.K. Norskov, Journal of Catalysis 248 (2007)
188-203
26) P. Raybaud, J. Hafner, G. Kresse, S. Kasztelan, H. Toulhoat, J. Catal. 189 (2000) 129.
27) H. Schweiger, P. Raybaud, G. Kresse, H. Toulhoat, J. Catal. 207 (2002) 76.
28) H. Schweiger, P. Raybaud, H. Toulhoat, J. Catal. 212 (2002) 33.
29) J.F. Paul, S. Cristol, E. Payen, Catal. Today 130 (2008) 139.
30) J.V. Lauritsen, M. V. Bollinger, E. Laegsgaard, K.W. Jacobsen, J. K. Nørskov, B.S.
Clausen, H. Topsøe, F. Besenbacher, J. Catal. 221 (2004) 510.
31) S. Fuentes, G. Diaz, F. Pedraza, H. Rojas, N. Rosas, Journal of Catalysis 113 (1988)
535-539.
32) A.-F. Lamic, A. Daudin , S. Brunet , C. Legens, C. Bouchy , E. Devers , Applied
Catalysis A: General 344 (2008) 198–204.
33) L. Ma, W.-X. Chen, H. Li, Z.-D. Zu, Mater. Chem. Phys. 116 (2009) 400-405.
19

34) D. Genuit, P. Afanasiev, M. Vrinat, J. Catal. 235 (2005) 302-317.
35) P.N. Trikalitis, T.A. Kerr, M.G. Kanatzidis, Microprous and Mesoporous Materials 88,
(2006) 187-190.
36) I.A. Sizova, S.I. Serdyukov, A.L. Maksimov, N.A. Sinikova, Petroleum Chemistry. 55
(2015) 38-44
37) G. Leyral, M. Ribes. L. Courthéoux, D. Uzio. A. Pradel, Eur. J. Inorg. Chem. (2012)
4967-4971.
38) http://www.casaxps.com.
39) C. Legens, P. Raybaud, Catalysis by transition metal sulfides : from molecular theory
to industrial application IFPEN-Lyon, Technip Edition, Paris, Chapter 2.4 (2013) 259-
300.
40) J. Lian, X. Duan, J. Ma, P. Peng, T. Kim, W. Zheng, ACSNano, 3 (11) (2009) 3749-
3761.
41) B. Fang, J. Yu, X. Ge, C. Yang, Materials Letters 73 (2012) 229-231
42) M. Antonietti, D. Kuang, B. Smarsly, Y. Zhou, Angew. Chemie. Int. Ed. 43 (2004)
4988-4992.
43) P.A. Nikulshina,b, V.A. Salnikova, A.N. Varakina, V.M. Kogan. Catalysis Today 271
(2016) 45–55
20

Figure 1 : Diffractograms of NiMoS bulk catalysts prepared with various amounts of IL. The
broad peaks can be attributed to MoS2 phase ( ; reference JCPDS : 010751539) while the
sharp peaks are due to NiS phase ( ; reference JCPDS : 010771624) .
21

Figure 2 : Raman spectra of the 1:0 and 1:100 NiMoS catalysts.
22

Figure 3 : SEM images of NiMoS samples focused on NiS particles : NiMoS 1:0 (a)
NiMoS 1:0,5 (b) ; NiMoS 1:1 (c) ; NiMoS 1:10 (d) ; NiMoS 1:100 (e))
23

A
B
Figure 4 : S 2p (A) and Mo 3d (B) XPS spectra of NiMoS 1:10
24

Figure 5 : Ni 2p XPS spectra of : NiMoS 1:0 (A) NiMoS 1:0.5 (B) ; NiMoS 1:1 (C) ;
NiMoS 1:10 (D) and NiMoS 1:100 (E)
25

B
A
C
D
E
Figure 6: TEM images of NiMoS samples focused on NiS particles : NiMoS 1:0 (A)
NiMoS 1:0,5 (B) ; NiMoS 1:1 (C) ; NiMoS 1:10 (D) ; NiMoS 1:100 (E))
26

b
d
c
e
Figure 7 : TEM images of NiMoS samples focused on Ni promoted MoS2 slabs : NiMoS 1:0
(a) NiMoS 1:0,5 (b) ; NiMoS 1:1 (c) ; NiMoS 1:10 (d) ; NiMoS 1:100 (e))
27

Figure 8: Histograms of the slab distribution as a function of length determined by TEM for
various NiMoS samples., (A) : 1:0, (B) : 1:0,5, (C) : 1:1, (D) : 1:10, (E) : 1:100
28

Figure 9 : Distribution of slabs as a function of their stacking determined by TEM for various
NiMoS samples :, (A) : 1:0, (B) : 1:0,5, (C) : 1:1, (D) : 1:10, (E) : 1:100
29