novel magnetically recyclable mos catalysts for
TRANSCRIPT
NOVEL MAGNETICALLY RECYCLABLE MOS2 CATALYSTS FOR
HYDRODESULFURIZATION AND HYDRODEOXYGENATION
by
Seyyedmajid Sharifvaghefi
M.Sc. in chemical engineering, Chalmers University of Technology, Sweden
A Dissertation Submitted in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy (PhD)
in the Graduate Academic Unit of Chemical Engineering
Supervisor: Ying Zheng, PhD, Chemical Engineering
Examining Board: Paul A. Arp, PhD, Forestry and Environmental Management
Mladen Eic, PhD, Chemical Engineering
Huining Xiao, PhD, Chemical Engineering
External Examiner: Chunbao Xu, PhD, Chemical and Biochemical Engineering,
University of Western Ontario
This dissertation is accepted by the
Dean of Graduate Studies
THE UNIVERSITY OF NEW BRUNSWICK
May 2017
©Seyyedmajid Sharifvaghefi, 2017
ii
ABSTRACT
Problems associated with hydrotreating of heavy and extra-heavy crude oils in
conventional fixed-bed reactors have led to the design of more innovative reactors such as
slurry reactors. Dispersed fine catalysts are commonly used in slurry reactors. However,
separation of the fine particles with sizes in micrometers or nanometers from the liquid
phase represents a challenge. In this work, novel magnetically recyclable catalysts are
designed to address this issue.
Two different magnetic core materials, Fe3O4 and Fe3S4, were studied. Core-and-shell
composite catalysts MoS2/Fe3O4 and MoS2/Fe3S4 were designed. Both catalysts showed
high activity in the hydrodesulfurization (HDS) of dibenzothiophene (DBT). Former one
showed high affinity towards the direct desulfurization (DDS) pathway while the later one
presented a balanced selectivity between DDS and hydrogenation (HYD) pathways. This
provides the refineries an option to adjust the hydrotreating process based on their needs
using each of the catalysts or a combination of both. Catalyst MoS2/Fe3S4 was further
promoted by nickel and cobalt. The activity of the catalysts could be ranked as
NiMoS/Fe3S4 > CoMoS/Fe3S4 > MoS2/Fe3S4. Addition of Ni to MoS2 catalyst forms the
Ni-Mo-S phase with increased accessible sulfur edge which is responsible for the
enhancement in both hydrogenation and desulfurization. The catalyst MoS2 with greigite
core and its promoted catalyst NiMoS/CoMoS were also applied to the
hydrodeoxygenation (HDO) of stearic acid. HDO was the dominant pathway for all of the
catalysts.
Catalysts MoS2/Fe3O4 and MoS2/Fe3S4 were used to investigate the roles of H2 and H2S
and the involved active sites in the HDS of DBT. Two different active sites for
iii
hydrogenation (HYD) and hydrodesulfurization (HDS) were identified. The MoS2 with
magnetite core demonstrated high selectivity towards the DDS pathway, which was
attributable to the differences in adsorption energy of hydrogen and DBT over
hydrogenation and desulfurization sites. H2S is favored being adsorbed on the S-edge
vacancies to transfer the active sites to HYD/Isomerization sites. Hydrogen plays three
distinctive roles: 1) When adsorbed at Mo edge, hydrogen promotes hydrogenation; 2)
Strongly bonded at S-edge it accounts for direct desulfurization; 3) while loosely bonded
to S-edge it favors HYD and isomerization.
iv
ACKNOWLEDGEMENTS
First and foremost, I offer my sincerest gratitude to my supervisor, Dr. Ying Zheng, who
has supported me throughout my thesis with her patience and knowledge whilst allowing
me the room to work in my own way. Thank you for all the fruitful discussions and
constructive suggestions and thank you for being present every step of this long journey.
Special thanks to my wife and my best friend, Fatemeh Lashkari, for being there whenever
I needed it and her kind words all the way throughout my work. I would not have succeeded
in my work without her constant support.
In my daily work, I have been blessed with a friendly and cheerful group of fellow students
and colleagues. Babak Shirani, Kyle Rogers, Hui Wang, Xue Han, and Sandra Riley. Thank
you all for providing an enjoyable time during my PhD studies.
Finally, I thank my parents for supporting me throughout all my studies an all their
encouragements.
v
Table of Contents
ABSTRACT ........................................................................................................................ ii
ACKNOWLEDGEMENTS ............................................................................................... iv
Table of Contents ................................................................................................................ v
List of Tables ................................................................................................................... viii
List of Figures .................................................................................................................... ix
List of abbreviations ........................................................................................................ xiii
Chapter 1: Introduction ....................................................................................................... 1 1.1 Heavy and extra-heavy crude oil hydro-processing and the plugging problem in fixed-
bed reactors .................................................................................................................................. 1
1.2 Technologies to process heavy and extra-heavy crude oils ............................................. 4
1.3 H2 and H2S dissociation and the active sites of (Ni)MoS2 ............................................... 6
1.4 Hydrodeoxygenation with sulfide catalysts ..................................................................... 8
1.5 Scopes, objectives, and outlines ....................................................................................... 9
Chapter 2: Literature review ............................................................................................. 12 2.1 Hydrotreatment Process ................................................................................................. 12
2.2 Chemistry of hydrodesulfurization ................................................................................ 16
2.3 Hydrodenitrogenation (HDN) ........................................................................................ 21
2.4 Hydrodeoxygenation (HDO) ......................................................................................... 23
2.5 Unsupported hydrotreating catalysts .............................................................................. 26
2.6 Supported catalysts ........................................................................................................ 28
2.7 Nature of active sites in promoted and unpromoted MoS2 ............................................ 30
2.8 Magnetic core and catalyst composite ........................................................................... 33
2.9 Summary ........................................................................................................................ 34
Chapter 3: Analytical procedures for catalyst characterization and evaluation ................ 35 3.1 Characterization techniques for catalysts ....................................................................... 35
3.1.1 Transmission electron microscopy (TEM)............................................................. 36
3.1.2 Scanning electron microscopy (SEM) ................................................................... 37
3.1.3 Nitrogen adsorption-desorption ............................................................................. 37
3.1.4 Elemental analysis ................................................................................................. 37
3.1.5 Magnetization ........................................................................................................ 38
vi
3.1.6 Thermal Gravimetric Analysis (TGA) ................................................................... 38
3.1.7 Surface charge density ........................................................................................... 38
3.1.8 UV-Vis ................................................................................................................... 38
3.2 Evaluation of catalyst performance................................................................................ 39
3.2.1 Hydrodesulfurization of model oil in batch reactor ............................................... 39
3.2.2 Hydrodeoxygenation of model oil in batch reactor ................................................ 40
Chapter 4: Development of novel magnetically recyclable MoS2 catalyst for direct
hydrodesulfurization ......................................................................................................... 42 4.1 Introduction .................................................................................................................... 42
4.2 Catalyst synthesis ........................................................................................................... 43
4.2.1 Preparation of silica coated magnetite (SiO2/Fe3O4) .............................................. 43
4.2.2 Preparation of MoS2/SiO2/Fe3O4 ............................................................................ 43
4.3 Results and Discussion .................................................................................................. 44
4.3.1 Textural and magnetic properties ........................................................................... 46
4.3.2 Effect of CTAC levels on conversion of Mo precursor and structure of catalyst .. 52
4.3.3 Catalytic activity and separation ............................................................................ 54
4.4 Conclusion ..................................................................................................................... 58
Chapter 5: New Insights on the roles of H2 and H2S and the involved active sites of MoS2
in hydrodesulfurization of Dibenzothiophene .................................................................. 60 5.1 Introduction .................................................................................................................... 60
5.2 Catalyst Synthesis .......................................................................................................... 60
5.3 Results and Discussion .................................................................................................. 61
5.3.1 Reaction routes over MoS2 with and without the support ...................................... 65
5.3.2 Inhibiting effect of H2S on DDS pathway .............................................................. 68
5.3.3 H2 and H2S dissociation and involved active sites ................................................. 70
5.3.4 The effect of H2 pressure ....................................................................................... 77
5.4 Conclusion ..................................................................................................................... 80
Chapter 6: Edge activation of H2 and H2S over NiMoS and the effect of Ni loading ...... 82 6.1 Introduction .................................................................................................................... 82
6.2 Catalyst synthesis ........................................................................................................... 82
6.3 Results ............................................................................................................................ 83
6.4 Discussion ...................................................................................................................... 92
vii
6.5 Conclusion ..................................................................................................................... 97
Chapter 7: Sulfided catalysts supported on magnetic greigite: Recyclable catalysts for the
hydrodesulfurization of heavy crude oil ........................................................................... 99 7.1 Introduction .................................................................................................................... 99
7.2 Results and Discussion ................................................................................................ 100
7.3 Conclusion ................................................................................................................... 119
Chapter 8: Hydrodeoxygenation of Stearic acid with unpromoted and Ni(Co)-promoted
MoS2 supported on greigite ............................................................................................ 121 8.1 Introduction .................................................................................................................. 121
8.2 Results and Discussion ................................................................................................ 122
8.3 Conclusion ................................................................................................................... 136
Chapter 9: Conclusions and recommendations ............................................................... 138
Bibliography ................................................................................................................... 141
Curriculum Vitae
viii
List of Tables
Table 2.1: main sulfur compounds in petroleum fractions . ............................................. 17
Table 2.2: Structures of Selected Organic Nitrogen Compounds . ................................... 22
Table 2.3: Composition of different feeds for HDO ......................................................... 24
Table 4.1: Properties of the prepared catalysts. ................................................................ 48
Table 4.2: Effect of the amount of CTAC on the conversion of Mo and the final pH. .... 52
Table 4.3: DBT conversion and selectivity of the prepared catalysts. .............................. 57
Table 5.1: Properties of the fresh and treated MoS2-100M and MoS2 catalysts............... 62
Table 5.2: Product selectivity (%) over MoS2 catalyst for HDS of DBT after 0 and 30 min
of reaction. ........................................................................................................................ 69
Table 6.1:Physical properties of the samples after the treatment. .................................... 85
Table 6.2: Effect of Ni/(Mo + Ni) mole ratio on HDS of DBT over unsupported NiMoS
catalysts at 3.4 MPa and 320 0C for 60 min. ..................................................................... 89
Table 6.3: HDS of DBT over MoS2-M and NiMoS-M catalysts with various Ni/(Ni+Mo)
ratios at 3.4 MPa and 320 0C for 60 min........................................................................... 90
Table 6.4: HDS of DBT over MoS2-I and NiMoS-I catalysts with various Ni/(Ni+Mo)
ratios at 3.4 MPa and 320 0C for 60 min........................................................................... 91
Table 7.1: Properties of the fresh samples. ..................................................................... 101
Table 7.2: DBT conversion and selectivity for 7 cycles in HDS of DBT over
MoS2/Fe3S4 catalyst. ..................................................................................................... 117
Table 7.3: DBT conversion and selectivity for 7 cycles in HDS of DBT over
MoS2/Fe3O4 catalyst...................................................................................................... 117
Table 8.1: Properties of the treated samples. .................................................................. 122
ix
List of Figures
Figure 1.1: Different types of reactors used for hydro-processing of heavy oils. .............. 4
Figure 2.1: A schematic view of a typical High-Conversion Oil Refinery . .................... 13
Figure 2.2: Once-through hydroprocessing unit: two separators, recycle gas scrubber. .. 15
Figure 2.3: GC–FPD chromatograms of gas oil with different levels of HDS. ................ 19
Figure 2.4: Reaction pathways for HDS of DBT. ............................................................. 20
Figure 2.5: HDS of DBT and 4,6-DMDBT through DDS and HYD routs over sulfided
NiMo/Al2O3 (fixed bed reactor, 340 °C, 4.0 MPa). ........................................................ 21
Figure 2.6: Product distributions and selectivities in the hydrodenitrogenation of
quinoline . ......................................................................................................................... 23
Figure 2.7: Reaction routes in conversion of triglycerides into alkanes . ......................... 26
Figure 2.8: Reaction pathway for HDO of 2-ethylphenol on MoS2-based catalysts. ....... 26
Figure 2.9: The "rim-edge" model for MoS2. ................................................................... 31
Figure 2.10: (a) Atom-resolved STM image of a triangular single-layer MoS2
nanocluster. (b) Ball models in top and side view of the edge structure, the Mo edge with
S dimers (S, bright; Mo, dark). (c) Ball model of a MoS2 triangle exposing this type of
edge termination................................................................................................................ 31
Figure 2.11. Schematic representation of the CoMoS model under reaction conditions. Co
is present in three different phases. (1) The active CoMoS nanoparticles; (2) a
thermodynamically stable cobalt sulfide (Co9S8); (3) Co dissolved in the Al2O3 support.
Only the CoMoS particles are catalytically active . .......................................................... 32
x
Figure 3.1: Reactor, stirrer, and furnace set-up for heating and rotating the autoclave
reactor. .............................................................................................................................. 41
Figure 4.1: Magnetization curves for Fe3O4 (circle), SiO2/Fe3O4 (square) and
MoS2/SiO2/Fe3O4 (triangle) measured at 300 K. .............................................................. 49
Figure 4.2: TEM images of a) untreated magnetite (Fe3O4), b) silica coated magnetite
(SiO2/Fe3O4), c) & d) Cat-CTAC, e) & f) Cat-SDS and g) & h) Cat-R. .......................... 51
Figure 4.3: Effect of CTAC levels a, b) 0.1 g and c,d) 0.75 g on the structure of the
synthesised MoS2 after 1 hr of reaction at 200° C. ........................................................... 54
Figure 4.4: Simplified reaction pathways for hydrodesulfurization of DBT. ................... 56
Figure 4.5: prepared catalyst after the HDS reaction, (a) dispersed inside the model oil
and (b) separated from the model oil after 5 seconds of exposure to an external magnet. 58
Figure 5.1: TEM images of MoS2, a, b) fresh and c, d) after treating. ............................. 63
Figure 5.2: TEM images of MoS2-100M, a, b) fresh and c, d) after treating. .................. 64
Figure 5.3: Possible reaction pathways for hydrodesulfurization of DBT. ...................... 65
Figure 5.4: DBT conversion after 2h of reaction and product selectivity at 50%
conversion for the synthesized catalysts. .......................................................................... 67
Figure 5.5: The selectivity of the isomerized products over the synthesized catalysts at
50% conversion of DBT. .................................................................................................. 67
Figure 5.6: DBT conversion for MoS2-100M and MoS2 catalysts at different starting H2
pressures. ........................................................................................................................... 79
Figure 5.7: Product selectivity for MoS2 catalyst at different starting H2 pressures. ....... 79
Figure 5.8: Product selectivity over for MoS2-100M catalyst at different starting H2
pressures. ........................................................................................................................... 80
xi
Figure 6.1: TEM images of the of the freshly synthesized 0.35-NiMoS. ......................... 84
Figure 6.2: TEM images of treated NiMo sulfide samples with various Ni/(Ni+Mo)
loadings, a) 0.1, b) 0.2, c) 0.35, d) 0.5, e) 0.65. ................................................................ 85
Figure 6.3:Isotherm (a) and pore size distribution (b) of unsupported MoS2 and NiMoS
catalysts with different amount of Ni loading................................................................... 87
Figure 6.4: XRD patterns of unsupported MoS2 and NiMoS catalysts with various Ni
loadings. ............................................................................................................................ 88
Figure 6.5: Accessibility of DBT molecule to supported and unsupported NiMoS catalyst
through flat adsorption to the edges. ................................................................................. 94
Figure 7.1: XRD pattern of Fe3O4 and synthesized Fe3S4. ............................................. 100
Figure 7.2: TEM images of the Mo/G (a,b), CoMo/G (c,d), and NiMo/G samples (e,f).
......................................................................................................................................... 103
Figure 7.3: Isotherms (a) and pore size distribution (b) of Mo/G, CoMo/G, and NiMo/G.
......................................................................................................................................... 105
Figure 7.4: Magnetization curves for Fe3S4 and MoS2/Fe3S4 measured at 300 K. ......... 107
Figure 7.5: Mo/G composite after the HDS reaction a) dispersed inside the model oil and
b) separated from the model oil after seconds of exposure to an external magnet. ........ 107
Figure 7.6: DBT conversion and product selectivity for the synthesized catalysts after 2h
of reaction. ...................................................................................................................... 110
Figure 7.7: The selectivity of the isomerized products over the synthesized catalysts after
2 hr of HDS of DBT........................................................................................................ 111
Figure 7.8: Possible reaction pathways in HDS of DMDBT. ......................................... 111
xii
Figure 7.9: DMDBT conversion and product selectivity for the synthesized catalysts after
2h of reaction. ................................................................................................................. 114
Figure 7.10: The selectivity of the isomerized products over the synthesized catalysts
after 2 hr of HDS of DMDBT. ........................................................................................ 115
Figure 7.11: XRD patterns of treated MoS2/Fe3O4 and after cycles 1, 3, 5, and 7. ........ 118
Figure 8.1: TEM images of fresh (left) and treated (right) Mo/G (a,b), CoMo/G (c,d), and
NiMo/G samples (e,f). .................................................................................................... 124
Figure 8.2: XRD patterns of the catalysts. ...................................................................... 125
Figure 8.3: Conversion of stearic acid over the prepared catalysts. ............................... 127
Figure 8.4: The yield of the products from deoxygenation of stearic acid over a) Mo/G, b)
CoMo/G, and C) NiMo/G at different reaction times. .................................................... 128
Figure 8.5: Comparison of the C18/C17 ratio. ............................................................... 132
Figure 8.6: Comparison of the paraffin/olefin ratio. ....................................................... 134
xiii
List of abbreviations
BT: Benzothiophene
CN: Coordination number
DBT: Dibenzothiophene
DDO: Direct deoxygenation
DDS: Direct desulfurization
DFT: Density functional theory
EDX: Energy-dispersive X-ray spectroscopy
FCC: Fluid catalytic cracking
FID: Flame Ionization Detector
FTIR: Fourier transform infrared spectroscopy
GC: Gas chromatography
GC-MS: Gas chromatography and mass spectrometry
HDN: Hydrodenitrogenation
HDO: Hydrodeoxygenation
HDS: Hydrodesulfurization
HYD: Hydrogenation
LCO: Light cycle oil
MDBT: Methyl substituted dibenzothiophene
PFPD: Pulsed flame photometric detector
SEM: Scanning electron microscopy
STM: Scanning tunneling microscope
TEM: Transmission electron microscopy
xiv
TGA: Thermal gravimetric analysis
ULSD: Ultra-low sulfur diesel
1
Chapter 1: Introduction
1.1 Heavy and extra-heavy crude oil hydro-processing and the plugging problem in
fixed-bed reactors
The significant increase in oil demand in this decade and the shift of oil supplies from light
oil to heavier and dirtier oil such as bitumen in addition to stricter legislations for fuel
specifications from governments have made todays refiners to adjust themselves to meet
these new demands [1].
There are two routes commonly used to process heavy and extra-heavy oil, i.e. carbon
rejection and hydrogen addition. Carbon rejection processes such as delayed coking are
popular for their ease of operation and flexibility towards handling different feeds.
However, these processes have low yield of upgraded oil and produce large amounts of
coke. The best known hydrogen addition process for conversion of feed oil to lighter
products is hydroprocessing through fixed-bed reactors [2].
Catalytic hydrotreating is a hydrogenation process that is used for removal of the
contaminants such as nitrogen, sulfur and oxygen, and metals from heteroatom constituents
of petroleum fractions. The reason that these contaminants should be removed from the
petroleum fractions is that they can have detrimental effects on the equipment, the catalysts,
and the quality of the finished product as they pass through the processing units [3]. The
major concern among these impurities inside the products is about sulfur compounds
because of the environmental and health issues that they can cause such as:
2
• Forming poisonous SOX emission after the combustion of the sulfur containing
fuel, which is harmful for human health and also a contributor to acid rain, ozone
depletion, and smog.
• The main petroleum-generated pollutants, esp. diesel, refer to the production of soot
(combustion temperature too low) and NOx (combustion temperature too high).
• In diesel cars, traces of sulfur inside the diesel fuel also have a poisonous effect on
car catalytic convertors, thereby reducing their combustion effectiveness in
removing carbon monoxide and unburned hydrocarbons.
Because of the adverse effects of sulfur compounds in fuels many countries have passed
stringent regulations to reduce the amount of sulfur impurities in fuels, especially in diesel
fuel during the past years. For example, the EU started the regulations for the amount of
sulfur in automotive fuel, including diesel, since 1994 from 2000 ppm and gradually
reduced it to 10 ppm in 2009 [4]. The traditional catalysts for hydrotreating are not active
enough for the removal of the refractory sulphur compounds. Therefore, a great amount of
effort has been spent to develop more active and more efficient catalysts capable of
producing cleaner fuels.
To date, these efforts have resulted in commercially used highly active catalysts such as
NEBULA, STARS or Topsøe BRIM series which can closely satisfy the removal of sulfur
based on the demands by Environmental Protection Agency for sulfur specifications [5].
However, there is a new problem in oil industry that many refiners are facing with. The
problem is that the resources for light petroleum fraction are decreasing and heavy crude
oil resources are becoming extra heavy with higher amount of fine particles and other
3
impurities while the petroleum market is demanding middle distillates with low contents
of sulfur, nitrogen, and other contaminants.
A major problem associated with the high concentration of contaminants and fine solid
particles in extra-heavy oil can block packed reactors. The common solution for this
problem is to use filters to protect the catalyst beds. However, filters are ineffective in
removing complex polymeric iron sulfide gums. These gums are formed from iron
compounds (e.g., iron porphyrins) in response to: (i) their reaction with sulfur in the feed,
(ii) their dissolution by naphthenic acid, and/or (iii) their persistence of particulates smaller
than 5 to 20 microns [6]. The small solid particles are caught by the packed catalysts in the
beds such as hydrotreators. Accumulation of the retained particles can significantly change
the hydrodynamics of the packed-bed reactor by early increasing of the pressure drop and
reducing bed permeability [7]. If the pressure drop builds up beyond the operation limits
allowed for pumps and compressors feeding the packed bed, then the column must be
stopped and the catalytic charge dumped and exchanged with a fresh packing unit. In many
cases, the plugged bed still has catalytic activity so the process economics suffers due to
early catalyst changes [8].
Unless addressed, the problem of plugging packed beds will likely be encountered more
often and in a more accentuated way in the future, as follows:
• High quality (shorter chain) hydrocarbons are the most likely reserves to be
depleted. There is therefore the need to exploit the lower quality reserves [9, 10].
These reserves will have diverse origins and diverse solids content in size and
mineralogy, as is already the case with the Athabasca bituminous sands of Alberta.
4
• With the monotonic decrease in oil reserves, it will become necessary to enhance
the usability of high boiling cuts into products of higher value. Hence, there will be
additional pressure on removing higher sulphur, metals and coke content from raw
refinement material. In the past, high-boiling-point petroleum cuts were discarded
or used in road construction.
1.2 Technologies to process heavy and extra-heavy crude oils
As discussed earlier, using the conventional fixed-bed reactors in hydro-treatment of heavy
and extra-heavy crude oils leads to plugging issues and premature shutdown of oil-refining
processing units. To solve this problem, reactors with different designs were introduced.
Figure 1.1 shows schematic representations of the alternative reactors used for
hydroprocessing of heavy oils [11]. Slurry-phase hydroconversion technology is one of the
developed methods to deal with the problems associated to the processing of heavy crude
oil. Using slurry phase reactors has multiple advantages and disadvantages as follows [12-
14].
Figure 1.1: Different types of reactors used for hydro-processing of heavy oils.
5
Advantages:
• The catalysts in slurry reactors are disposable, and possess high surface areas due
to their micron-scale particle size.
• The smaller size of catalysts used in slurry reactors compared with catalysts utilized
in other reactors results in shorter traveling distances and hence less time for a
reactant molecule to find an active catalyst site.
• Good external mass transfer due to the unsupported nature of the catalysts [6].
• Residue conversion higher than 90% (sometimes very close to 100%) can be
achieved at space velocities much higher than other reactors.
• Due to not needing internal equipment components, the volume of the slurry reactor
is maximized.
• The thermally stable homogeneous-phase operations have low chances of
temperature runaways.
Disadvantages:
• SPR should be carefully designed to maintain a well-mixed three phase slurry of
heavy oil, catalyst, and hydrogen to promote effective contact.
• The efficiency of the hydroprocessing SPR is very dependent upon the type of
catalyst selected.
Well-known processes that are developed based on slurry reactors include Veba combi-
cracking [15], CanMet [16], HDH Plus [17] and Eni slurry technology (EST) [18, 19].
Commonly used catalysts in slurry reactors are cheap and use disposable materials such as
Fe. Application of low activity materials results in low upgrading efficiency. To improve
the quality of products, a known hydrotreating catalyst such as MoS2 is a good candidate.
6
However, the higher cost of MoS2 can compromise the profitability of the process if once-
throughput catalyst is used [20]. Therefore, recyclability of catalyst becomes economically
crucial for processing heavy oil in slurry reactors.
1.3 H2 and H2S dissociation and the active sites of (Ni)MoS2
Many countries have moved to limit the amount of sulfur in the fuels to near zero. Diesel
fuel containing such low levels of sulfur is called ultra-low sulfur diesel (ULSD).
Refineries need to adjust themselves accordingly to reach these new limits for sulfur
content of the fuels they produce. To reach near zero levels of sulfur, nearly all of the sulfur
present in the sulfur-containing molecules should be removed. Removing sulfur from
simple molecules like benzothiophene can be accomplished. However, desulfurization of
more complicated and refractory molecules like dibenzothiophene (DBT) and alkyl-
substituted DBTs can be challenging [1, 21, 22]. Sulfur removal is generally carried out in
refineries through hydrotreating units. Different catalysts have been tested in years for their
application in these units, though, the most abundantly used catalysts are transition metal
sulfides (TMS) due to their high performance and resistance toward poisoning. The most
commonly used TMS catalyst, i.e., molybdenum sulfide, is usually promoted with Co, Ni
and/or W [23, 24].
In detail, MoS2 consists of stacked layers in which each layer of molybdenum is
sandwiched between two layers of sulfur. These layers are joined together by weak van der
Waals forces. Truncated or hexagonal MoS2 nano-particles are expected to contain two
distinct slab terminations, Mo (1010) and S (1010) edges. Regardless of synthesis method,
7
the stoichiometry of MoS2 is mostly identical. However, the textural properties and the
activity of the catalyst can be greatly affected by the varying catalyst synthesis methods.
Various models have been proposed to explain the Co, Ni and/or W promotion effect of
the MoS2 catalyst. The edge-decorated Ni(Co)-Mo-S phase model among these is the most
accepted one [21, 25, 26]. In this model, the promoter atom substitutes the Mo atoms at the
edges of the MoS2 slabs [27-29]. However, the exact edge locations of the promoter atoms
are still unclear, and especially with regard to Ni.
The HDS mechanism and its relation to catalyst active sites involved in the HDS process
over Ni(Co)-promoted and unpromoted MoS2 have been extensively studied using various
techniques. Imaging [30-32], DFT studies [33, 34], spectroscopic studies [35-37], and use
of poisons or inhibitors [38-40] are examples of these techniques.
HDS of DBT is known to proceed either through the direct desulfurization pathway (DDS)
or through the hydrogenation route (HYD). However, there has been two different views
on the active sites associated with the reaction pathways of hydrodesulfurization. Based on
the correlation between the number of edge sites and the activity of the HDS and
hydrogenation reactions, some researchers have claimed that the same type of active sites,
i.e. sulfur vacancies at the edges of the catalyst, is accountable for both DDS and HYD [41,
42]. On the contrary, the other researchers, based on the inhibiting effect of compounds
such as H2S, proposed two different types of sites that are responsible for the two reaction
routes, i.e. sulfur vacancies as DDS sites and metallic-like sites on the basal planes and
near edges (brim sites) as HYD sites (refer to section 2.7 for more details on the nature of
active sites). H2S was identified to have an inhibiting effect on the DDS pathway while it
had only a minor effect on the HYD route [43]. The difference in inhibiting effect of the
8
nitrogen compounds over the DDS and HYD routes is also an evidence for the presence of
two different types of sites [44, 45]. We made an attempt to find an answer to this dilemma
based on our results.
Hydrogen is known to have a two-fold effect at the edges of MoS2. First, by reacting with
the sulfur atoms at the edges. Second, to form -SH groups at the edges through dissociative
adsorption. Both of these processes are considered important for HDS to provide vacancies
and reactive hydrogen supply over the catalyst edges. Hydrogen sulfide can also have two
effects on the edges, i.e., adsorption at the vacancies followed by dissociative adsorption
to form -SH groups. The HYD pathway of the HDS reaction starts with the dissociation of
hydrogen over the edges of MoS2. DFT calculations show that the dissociation of hydrogen
can follow two different pathways [46]: one involving sulfur dimers, and the other via the
interaction with the vacant Mo atoms. However, the difference between the roles of the
dissociated hydrogen at each site within the HDS process remains unclear. Moreover, the
difference between the role of hydrogen and hydrogen sulfide in hydrogenation is yet to be
understood.
1.4 Hydrodeoxygenation with sulfide catalysts
The decrease in the fossil fuel resources and the concerns about the greenhouse gas
emissions has driven the interests toward the use of renewable biofuels. Vegetable oils, i.e.,
triglycerides in the form of fatty acids with 15 to 18 carbons in their chain, however have
some disadvantages due to their immiscibility with standard fuels, and low heating values
due to their high oxygen content. Therefore, to obtain suitable fuel products, the oxygen
should be removed from their structure. The two reaction routes are suggested for the
9
deoxygenation process, i.e. Decarbonisation/Decarboxylation (HDC) and
Hydrodeoxygenation (HDO). With the former, oxygen removal occurs progressively in the
form of C-C bond cleavage and CO or CO2 formation at each step. With the latter, oxygen
is removed in the form of H2O without removing any carbons from the fatty acid [47, 48].
The deoxygenation process in many aspects is similar to the hydrotreating process in the
refineries. It is found that the hydrotreatment process with sulfide catalysts can remove the
oxygenated compounds from bio oils [49-52] . Hence, the existing hydrotreating process
has been used for the co-treatment of fossil fuels and biomass [53-60] . The most common
catalysts used in the hydrotreating process are Ni or Co-promoted and unpromoted MoS2
catalysts. These catalysts have also been applied in the deoxygenation of vegetable oils and
have been found to effectively remove oxygen [61-65]. They showed selectivity for both
HDC and HDO reaction pathways based on the type of catalyst, the preparation method,
and the reaction conditions.
Similar to heavy crude oil, the presence of glycerol and impurities especially in viscous
waste cooking oil can also plug packed-bed reactors [66]. Co-processing of vegetable oils
and heavy oil can therefore become costly unless the plugging effect is reduced or
eliminated altogether.
1.5 Scopes, objectives, and outlines
The scope of this thesis focuses on the synthesis and characterization of nano-sized sulfide
catalysts with magnetic properties. These catalysts are known to be effective in
hydrotreating heavy-feedstock oil recourses, and are thought to become most efficient in
slurry reactors. In this, the synthesis, the effect of type of magnetic core, and the addition
10
of promoters are carefully investigated. This includes examining the H2 and H2S activation
sites and processes over the edges of the promoted and unpromoted MoS2 catalysts.
Moreover, unpromoted MoS2 and Ni and/or Co-promoted MoS2 are used in the
hydrodeoxygenation of stearic acid as model oil. The results to be derived from this
pertaining to catalytic activity and selectivity will then be used to propose and evaluate a
possible step-by-step reaction mechanism.
The objectives of this thesis is as follows:
1. To synthesize of high-performance magnetically recyclable MoS2 using the
hydrothermal method.
2. To study the effect of different magnetic cores and promoters on the activity and
selectivity of the catalyst.
3. To explore the H2 and H2S roles in the hydrodesulfurization process over a
(Ni)MoS2 catalyst with magnetic core, and to locate its active H2 and H2S
dissociation sites.
4. To investigate the activity and selectivity of the recyclable (Ni,Co)MoS2 composite
in deoxygenation.
In Chapter 2, an extensive literature review is provided on the synthesis, properties, and
activities of supported and unsupported Mo-based catalysts. The hydrothermal technique
for the preparation of MoS2 and the effect of the process parameters are described. An
introduction of the catalysis mechanism and the active sites of unpromoted and promoted
11
MoS2 is also provided, as well as their application in hydrodesulfurization and
hydrodeoxygenation.
In Chapter 3, the characterization techniques for the sulfide catalysts are described. This
includes a presentation of the experimental hydrotreatment setup, and the precedures used
for the catalyst activity testing and product analyses.
Chapter 4 describes the step-by-step synthesis of magnetically recyclable nanosized MoS2
composite. This includes addressing the selection of the materials used for anchoring the
MoS2 catalyst, and exploring the relationships between MoS2 catalyst properties and
activities.
In Chapter 5, a series of MoS2 catalysts with and without magnetite particles are described
within the context of their differential desulfurization sites and activities regarding H2 and
H2S hydrogenation.
Chapter 6 builds on Chapter 5 by addressing the loading, role and sites of Ni as H2 and H2S
hydrogenation promoter on the MoS2 catalyst.
Chapter 7 focusses on studying the magnetic core effect on the hydrodesulfurization
process using Ni and Co catalyst promoters.
Chapter 8 builds on Chapter 7 by (i) testing the prepared catalysts in terms of their hydro
deoxygenation effectiveness in terms of activity and selectivity, and (ii) proposing a
hydrodeoxygenation reaction mechanism.
Chapter 9 provides a summary in terms of conclusions and recommendations for future
work.
12
Chapter 2: Literature review
2.1 Hydrotreatment Process
Among different chemicals and materials that catalysts are utilized for their production,
fuels are one of the most important ones that play a vital role in our present economy. Fuels
are mostly produced in refineries by processing Crude oil in several units. One of these
units is hydrotreater which is among the most common process units in modern petroleum
refineries. As it can be seen in Figure 2.1, a typical refinery in a western country usually
has at least three hydrotreaters. Commonly, one hydrotreater unit for naphtha, one or two
for light gas oil, and one or two for heavy gas oil and/or vacuum gas oil are used [67].
Catalytic hydrotreating is a hydrogenation process that is used for removal of the
contaminants such as nitrogen, sulfur, oxygen, and metals from heteroatom constituents of
petroleum fractions. The reason that these contaminates should be removed from the
petroleum fractions is that they can have detrimental effects on the equipment, the catalysts,
and the quality of the finished product as they pass through the processing units [2].
Because of the difference in reactivity and chemistry of each of these contaminants,
specific processes have been developed for their removal. These processes are called
hydrodesulfurization (HDS), hydrodenitrogenation
13
Figure 2.1: A schematic view of a typical High-Conversion Oil Refinery [8].
(HDN), hydrodeoxygenation (HDO) and hydrodemetallation (HDM) for the removal of
S, N, O and metal compounds, respectively.
S, N and O compounds can be found in any type of feedstock but the metal impurities are
usually of concern only in heavier feedstock such as atmospheric and vacuum residues.
The major metal contaminants are usually V and Ni which can cause catalyst deactivation
in long run [68]. Oxygen impurities are not known to have any major environmental
impacts, however, some oxygen compounds such as phenols and naphthenic acids can
cause corrosion problems in equipments like the storage vessels [69].
14
Typically, hydrotreaters are placed prior to processes, like catalytic reforming, to prevent
the contamination of the catalysts by untreated feedstock. Hydrotreaters are also placed
before catalytic cracking units to decrease the amount of sulfur and to improve product
yields, and to upgrade middle-distillate petroleum fractions into finished kerosene, diesel
fuel, and heating fuel oils. Moreover, hydrotreating can result in more saturated compounds
in product by hydrogenating the olefins and aromatics [3].
Hence, the main role of hydrotreating can be summarized as follows [70]:
1. Meeting finished product specification.
• Kerosene, gas oil and lube oil desulphurization.
• Olefin saturation for stability improvement.
• Nitrogen removal.
• De-aromatization for kerosene to improve cetane number.
2. Feed preparation for downstream units:
• Naphtha is hydrotreated for removal of metal and sulphur.
• Sulphur, metal and polyaromatics removal from vacuum gas oil (VGO) to be used
as FCC feed.
• Pretreatment of hydrocracking feed to reduce sulphur, nitrogen and aromatics
which has poisonous effect on its catalyst.
15
Figure 2.2: Once-through hydroprocessing unit: two separators, recycle gas scrubber.
Figure 2.2 shows a simplified flow diagram of a hydrotreating unit which is consisted of
different parts. In this process the feedstock is mixed with hydrogen in the first step,
preheated in a heater (315–430 °C) after that, and then charged under pressure (up to 6.9
MPa) through a fixed-bed catalytic reactor. In the reactor, the sulfur and nitrogen
compounds in the feedstock are converted into H2S and NH3. The reaction products leave
the reactor and then washed with water to convert almost all of the NH3 and some of the
H2S into aqueous ammonium hydrosulfide, NH4HS (aq). The ammonium hydrosulfide is
removed in the low-pressure flash drum as sour water. After washing, the products are
cooled to a low temperature and then enter a two stage (high and low-pressure) liquid/gas
separator [71].
16
The gas stream rich in hydrogen from the high-pressure separation enters a high-pressure
amine absorber to remove H2S. H2S has an inhibiting effect on HDS reactions and lowers
the purity of the recycle gas. The cleaned gas stream is then recycled to combine with the
feedstock. The low-pressure gas stream rich in H2S is sent to a gas-treating unit to remove
H2S. After cleaning, the gas can be used as fuel for the refinery furnaces. The liquid product
from the hydrotreating unit is normally sent to a stripping column where H2S and other
undesirable components are removed [71].
2.2 Chemistry of hydrodesulfurization
The main sulfur compounds that can usually be found in petroleum fractions are thiols
(mercaptans), sulfides, disulfides, thiophenes, benzothiophenes (BTs), dibenzothiophenes
(DBTs), naphthothiophenes (NTs), benzonaphthothiophenes (BNTs), and their alkyl-
substituted derivates, as shown in Table 2.1 [72]. Studies have shown that there is a major
difference between the reactivity of sulfur compounds. The ease of removal of sulfur from
a petroleum stream depends greatly on the structure of the sulfur compound being treated.
Generally, in contact with hydrotreating catalyst, non-cyclic sulfur compounds and
disulfides can react very fast. Six-membered ring sulfur compounds and saturated cyclic
sulfur compounds have also high reactivity. However, as shown in Table 2.2.1, reactivity
of aromatic five membered ring sulfur compounds are much less and their reactivity
reduces as the number of rings increases [73-75].
17
Table 2.1: main sulfur compounds in petroleum fractions [13].
Thiols (mercaptans) R-S-H
Sulfides R-S-R
Disulfides R-S-S-R
Thiophene
Benzothiophenes (BTs)
Dibenzothiophenes (DBTs)
Napthothiophenes (NTs)
Benzonaphthothiophenes (BNTs)
18
Figure 2.3 shows the GC-FPD chromatograms of gas oil with different levels of HDS for
production diesel fuel. After mild HDS, alkyl-substituted BTs are removed from gas oil
while DBT and its alkyl substitutes are remained. After HDS of gas oil, the main remained
sulfur compounds are 4-methyldibenzothiophene (4-MDBT), 4,6-
dimethyldibenzothiophene (4,6-DMDBT), with small traces of 6-ethyl,4-
dimethyldibenzothiophene (6-E,4-DMDBT) and 6-ethyl,2,4-dimethyldibenzothiophene
(6-E,2,4-DMDBT), which shows that these compounds are much less reactive compared
to other sulfur-containing compounds. Generally, only the most refractory sulfur species
remain in the diesel fuel after removing of sulfur to levels below 500 ppm by conventional
hydrodesulfurization [76-79]. It is suggested by Gates and Topspe [78] that the most proper
compounds for testing the HDS catalysts and reaction mechanisms are 4-MDBT and 4,6-
DMDBT.
The low reactivity of 4-MDBT and 4,6- DMDBT has been explained both by steric
hindrance preventing a proper interaction between the molecule and catalyst and by
electronic factors [80, 81]. There are two different set of views about the nature of steric
hindrance in these molecules. One view is that steric hindrance prevents that adsorption of
molecules on the surface of the catalyst [77, 82, 83]. Another view is that steric hindrance
slows the cleavage of C-S bond by responsible sites on catalyst [84, 85].
19
Figure 2.3: GC–FPD chromatograms of gas oil with different levels of HDS.
In general, there are two main reaction pathways for HDS of DBT and alkylated DBTs as
shown in Figure 2.4 [86]. The first pathway is the direct desulfurization (DDS) -also known
as hydrogenolysis- which directly leads to the formation of biphenyl (BP). The second
pathway starts by hydrogenation (HYD) of one aromatic ring to form an intermediate
product, tetrahydrodibenzothiophene (THDBT) which is desulfurized in the next step to
form cyclohexyl benzene (CHB) [78, 87].
Model compound tests have shown that the HDS reaction for DBT preferentially
progresses through the direct desulfurization pathway [82]. When alkyl substituents are
20
attached to DBT, the HDS activity of this the sulfur compounds decreases and the ratio
between DDS and HYD reaction routes changes. The HYD passway becomes the dominant
HDS route for 4and/or 6 alky-substituted DBT [82, 88]. As shown in Figure 2.5, the DDS
route for 4,6-DMDBT is severely reduced compared to DBT while the HYD route is nearly
the same. It has been suggested that the difference in HDS reactivity between DBT and
4,6-DMDBT is mainly because of the selective promoting effect on the DDS pathway [89].
Figure 2.4: Reaction pathways for HDS of DBT.
21
Figure 2.5: HDS of DBT and 4,6-DMDBT through DDS and HYD routs over sulfided
NiMo/Al2O3 (fixed bed reactor, 340 °C, 4.0 MPa).
2.3 Hydrodenitrogenation (HDN)
HDN is generally known to be more difficult to accomplish than HDS, but the relatively
smaller amounts of N-containing compounds in conventional crude oil makes them of less
concern for hydrotreating process [90]. However, the growing interest toward heavier oil
feedstocks, which are richer in nitrogen compounds than the conventional feedstocks, can
be problematic for the refineries as nitrogen removal is essential to prevent catalyst
poisoning in downstream processes, such as hydrocracking, catalytic cracking, and
reforming [40]. Moreover, nitrogen compounds in the feedstocks are known to have a
strong negative impact on other hydrotreating processes especially HDS. Even at low
concentrations, nitrogen compounds inhibit the HDS reaction through competitive
adsorption [91].
22
Nitrogen in petroleum is mostly present as aromatic ring compounds. 6-membered-ring
compounds, such as quinolines and benzoquinolines mostly have basic nature. Non-basic
nitrogen compounds are mainly 5-membered-ring compounds, such as indoles and
carbazoles. Basic nitrogen compounds are considered to be stronger inhibitors for HDS
reaction than the non-basic compounds [92]. Some of the common nitrogen compounds
found in oil feedstocks are shown in table 2.2 [40].
The initial step in the HDN of aromatic nitrogen-containing compounds is the saturation
of the heteroring followed by C–N bond cleavage. The resulting aliphatic or aromaticamine
intermediates are eventually converted to hydrocarbons and ammonia. Therefore, aromatic
nitrogen-containing species exhibit a strong affinity for the active sites associated with
hydrogenation reactions. This affinity is more severe in the case of basic nitrogen
compounds, which therefore results in more severe poisoning effect. Figure 2.6 shows the
reaction pathway and products for HDN of Quinoline (Q) [93].
Table 2.2: Structures of Selected Organic Nitrogen Compounds [40].
Basic Non-Basic
Quinoline
b.p. 238 °C
Indole
b.p. 254 °C
Benzo-
quinoline
b.p. 338 °C
Carbazole
b.p. 355 °C
Acridin
(benzo-
quinoline)
b.p. 346 °C
1,2-
Benzocarbazole
b.p. 411 °C
23
Figure 2.6: Product distributions and selectivities in the hydrodenitrogenation of
quinoline [93].
2.4 Hydrodeoxygenation (HDO)
The importance of hydrodeoxygenation (HDO), which occurs during hydroprocessing
depends on the origin of feeds. HDO plays an insignificant role in the case of the
conventional feeds, whereas for the feeds derived from coal, oil shale, and, particularly
from the biomass, its role can be rather crucial. The oxygen content in different feeds is
shown in Table 2.3 [49].
As mentioned before, the reason for removing sulfur and nitrogen compounds from the
fuel is their environmental impact by production of SOx and NOx poisonous gases during
combustion. However, no such poisonous gas will be emitted due to the combustion of
oxygen containing hydrocarbons. During the HDO process the oxygen is released in the
form of water which is environmentally friendly. Moreover, the oxygen content in the
conventional petroleum derived feed is below 2 wt.% [94]. Therefore, unlike sulfur and
24
nitrogen compounds, no special attention is given to oxygen-containing compounds during
hydrotreatment of conventional crude oil.
Table 2.3: Composition of different feeds for HDO
Convention
al crude
Coal-
derived
naphtha
Oil shale
crude
Bio oils
Liquefied Pyrolyzed
Carbon 85.2 85.2 85.9 74.8 45.3
Hydrogen 12.8 9.6 11.0 8.0 7.5
H/C 1.8 1.4 1.5 1.3 2.0
Sulphur 1.8 0.1 0.5 <0.1 <0.1
Nitrogen 0.1 0.5 1.4 <0.1 <0.1
Oxygen 0.1 4.7 1.2 16.6 46.9
In the case of biomass, the oxygen content can even reach 50 wt.%. The problem associated
with oxygen compounds is that some of these compounds can easily polymerize which can
cause fuel instability and therefore poor combustion performance [49].
Generally there are two types of bio fuel, i.e. vegetable oils and oil derived from the
pyrolysis or liquefaction of the ligno-cellulosic biomass. Vegetable oils are mainly
consisted of triglycerides which are bonded fatty acids with 15 to 18 carbon long chains.
Conversion and further deoxygenation of vegetable oils to bio-fuel can follow different
paths as shown in Figure 2.7 [95]. Triglycerides first decompose into aliphatic acids and
other intermediates and then follow either hydrodeoxygenation (HDO) pathway or
decarbonylation/decarboxylation (HDC) pathway. In HDO pathway, hydrocarbon
products are formed through dehydrogenation of the intermediate acid. HDC pathway
produces hydrocarbons with one carbon number lower that the main oxygenate reactant
due to the formation of CO and CO2 as by-product [96-99]. Since most bio-derived
25
triglycerides contain 18 carbons in branched chains, C18/C17 hydrocarbon ratio is
commonly used to distinguish the primary deoxygenation reaction route (C18/C17 over 1
indicates an HDO dominant deoxygenation reaction) [61, 97, 100].
Different types of catalysts with variety of combinations between active metal and support
have been investigated for the HDO reactions [101-106]. Among these, Mo-based catalysts
have been studied frequently for HDO reactions as these catalysts are also used in
traditional hydrotreating units [51, 105, 107-109]. Whiffen et al. [110] compared low-
surface-area MoO3, MoO2, MoS2 and MoP for HDO of 4-methylphenol at 623 K and 4.40
MPa H2. They found that the turnover frequency (TOF) of these catalysts based on CO
uptake for the HDO of 4-methylphenol decreases in the order MoP > MoS2 > MoO2 >
MoO3. A proposed mechanism for HDO of 2-ethylphenol on MoS2-based catalysts is
shown in Figure 2.8 [85]. It is suggested that 2-ethylphenol is activated over the slab edge
of MoS2 catalyst by adsorption of the oxygen molecule on a vacancy site. The S-H group
present on the edge of the catalyst (formed by hydrogen dissociation over the catalyst)
donates proton to the attached molecule which forms carbocation. The adsorbed compound
then undergoes direct C-O bond cleavage to form the deoxygenated molecule and oxygen
is released in the form of water during this step.
26
Figure 2.7: Reaction routes in conversion of triglycerides into alkanes [95].
Figure 2.8: Reaction pathway for HDO of 2-ethylphenol on MoS2-based catalysts.
2.5 Unsupported hydrotreating catalysts
In recent years, there have been many studies about unsupported transition metal sulfides
and their potential usage for hydrotreating [111-113]. Many mono-metallic sulfides have
27
been tested including MoS2 [114-116], WS2 [117], Ni sulfide [118] and Co sulfide [117,
119] along with a mixed metal sulfides, e.g. Co-Mo-S [120-122], Ni-Mo-S [120], Ni-W-S
[123], Ni-Mo-W-S [124], Fe-Mo-S [125], etc. Several noble metals such as Ru, Rh and Ir
have also shown high HDS activities [126]. Clearly, the high prices of noble metals restrict
their application in commercial hydrotreating. Among these catalysts, the mixed sulfide of
Ni-Mo-W has shown up to 3 times higher activity than the conventional alumina supported
NiMo and CoMo catalysts [127]. This catalyst has commercially been put in use for HDS
of diesel under the trade name of NEBuLa [128].
Among these catalysts MoS2 has been the subject of many studies because of its application
in different areas like lubrication, electrochemistry and most importantly catalysis. Variety
of methods has been used for synthesising MoS2. The most important of these methods are:
sulfidation of oxides, decomposition of precursors, hydrothermal and solvothermal.
Among these techniques, hydrothermal method has shown to be promising specially for
preparing nano-sized crystals because of the ease of process and mild reaction conditions,
which makes it suitable to be industrially applied.
Several groups have applied hydrothermal method for preparation of their catalysts;
however, each has used different precursors and reaction conditions. Peng et al. [129]
prepared their catalyst by reacting ammonium molybdate (NH4)6Mo7O24.4H2O), elemental
sulfur, and hydrazine monohydrate in an autoclave reactor at temperature range of 170-200
°C for reaction times between 72 h to 30 days. In a study by Tian et al. [130], MoO3 was
reacted with KSCN at temperature range 160-220 °C resulting in formation of nanotubes
and nanorods. In another study [131], MoS2 nanowires were formed by reaction between
28
MoO3, Na2S and 0.4mol/l of HCl at 260 °C. The advantageous of this method compared to
others is that it is simpler to perform, the precursors used are convenient and the final MoS2
obtained has good properties. Therefore, this method and materials were chosen for
synthesising MoS2 catalyst in this project.
2.6 Supported catalysts
In different catalysts, usually the active material is supported on a carrier which allows
higher dispersion of active materials and also reduces their susceptibility to sintering. The
main catalysts used currently in petroleum industry for hydrotreating of oil streams are Ni
or Co promoted sulfided Mo, supported on γ-alumina [23, 132]. γ-Alumina is the most
widely used support for commercial hydrotreating catalysts because in addition to the
above advantages it is also highly stable, has excellent mechanical properties and is
relatively inexpensive [133].
It is shown in different studies that the type of support for hydrotreating catalysts can cause
a dramatic change in the rate and selectivity of hydrotreating reactions [134, 135].
Therefore, in the past few years, there has been a growing interest in finding new support
materials for hydrotreating catalysts in order to improve their catalytic performance. These
include SiO2 [136], MgO [137], ZrO2 [138, 139], TiO2 [139, 140], carbon [141-143],
zeolites [144, 145]. For most of these supports, problems such as low surface area, limited
thermal stability and unsuitable mechanical properties prevent their commercial
application. Mesoporous materials such as MCM-41 [146, 147], SBA-15 [148, 149] have
also been mentioned as possible supports.
29
Another effect that the strength of metal-support interaction has is the degree of stacking
of the active phase. Type I sites which have strong interaction with support are mostly
found in monolayer slabs, while Type II sites with weak interaction with support are
usually found in multilayered slabs. Generally, all the slabs are accessible for DBT while
because of the steric hindrance, 4,6-DMDBT molecule cannot interact with the slab
adjacent to the support [88, 150].
The use of mesoporous materials such as SBA-15 as support material has also got some
attention because of its favorable catalytic activities toward bulky molecules, such as those
present in heavy crude oils and heavier petroleum fractions [151]. However, comparing
SBA-15 and zeolites, the acidity and hydrothermal stability of SBA-15 is lower. Bao et al.
[152] tried to solve these problems by incorporating Al into SBA-15 structure thus
replacing some of Si4+ ions with Al3+ ions. By adjusting pH and increasing temperature
during the hydrothermal process, they synthesized Al-SBA-15 with uniformly distributed
mesopores, high hydrothermal stability, and medium BrØnsted/Lewis acidity. They also
performed catalytic activity study on NiW/Al-SBA-15 catalyst and compared it with
activity of SBA-15 and Al2O3 supported catalysts. The results showed much higher HDS
activity for NiW/Al-SBA-15 among the two catalysts.
In addition to traditional hydrotreating catalysts, a new study by Oyama et al. [153] has
shown the high affinity of FeNiP catalyst supported on SiO2 toward removing 4,6 DMDBT
through DDS route. It is known that the main pathway for removal of refractory sulfur
molecules such as 4,6 DMDBT over traditional hydrotreating catalysts is through
hydrogenation due to the presence of methyl groups in this molecule which causes
geometrical restriction for its adsorption over the catalyst. The results also show that their
30
catalyst has higher activity compared to NiMoS/Al2O3 catalyst for removing 4,6 DMDBT.
They suggested that the special reactivity of the bimetallic material was probably due to a
ligand effect of Fe on Ni, which must also account for the high selectivity to DDS.
2.7 Nature of active sites in promoted and unpromoted MoS2
Since the introduction of MoS2 as a candidate for HDS reactions, many researchers have
tried to shed a light on the nature of the active sites in this catalyst. Since the early studies
on MoS2, it has been widely accepted that for unpromoted MoS2, the coordinatively
unsaturated sites (CUS) or exposed Mo ions with sulfur vacancies at the edges and corners
of MoS2 structures are the active sites for hydrogenation and hydrogenolysis reactions. The
basal planes were assumed to be inactive in sulfur removal process. Later, Daage and
Chianelli [80] proposed the rim-edge model in which they suggested that two types of
active sites are available on MoS2 slab, i.e. rim and edge (Figure 2.9). In their model, the
rim sites were considered to be responsible for the hydrogenation while edge sites were
assumed to be both active for hydrogenation and hydrogenolysis. They considered the basal
plane to consist of fully saturated sulfur atoms, making the basal plane almost inactive.
Based on this model, one can change the HYD/DDS selectivity of a catalyst by changing
the stacking height of MoS2 slabs.
31
Figure 2.9: The "rim-edge" model for MoS2.
Figure 2.10: (a) Atom-resolved STM image of a triangular single-layer MoS2
nanocluster. (b) Ball models in top and side view of the edge structure, the Mo edge
with S dimers (S, bright; Mo, dark). (c) Ball model of a MoS2 triangle exposing this
type of edge termination.
Recentely, Topsøe et al. [154, 155] have found new active sites on MoS2 catalyst involved
in hydrogenation reactions using scanning electron microscopy (STM). These so called
"brim sites" were fully saturated and one-dimensional molybdenum sites located on top of
the MoS2 nanocluster adjacent to the edge and extend around the cluster. The STM image
32
and the top and side view ball models of a one layer MoS2 nano-cluster are shown in Figure
2.10. The bright brim sites are easily distinguishable in the STM image.
For Co or Ni promoted catalysts, the most widely accepted model describing the structure
of active sites is the Co (Ni)–Mo–S phase model proposed by Topsøe et al. [27, 156]. In
their model, Co(Ni)–Mo–S structure of the catalyst are consisted of small MoS2-like
nanocrystals with Co or Ni atoms located at the edges of the MoS2 crystal layer. Cobalt in
particular can also be present in another two forms, i.e. thermodynamically stable cobalt
sulfide (Co9S8) and dissolved in the alumina support for supported catalyst. A schematic
representation of this model is shown in Figure 2.11. It was found that the HDS activity of
the catalyst has a linear correlation to the amount of Co(Ni)-Mo-S phase. The most widely
accepted explanation for the promotional effect of Co or Ni is proposed by Harris and
Figure 2.11. Schematic representation of the CoMoS model under reaction
conditions. Co is present in three different phases. (1) The active CoMoS
nanoparticles; (2) a thermodynamically stable cobalt sulfide (Co9S8); (3) Co dissolved
in the Al2O3 support. Only the CoMoS particles are catalytically active [157].
33
Chianelli [158]. They suggested that addition of Co or Ni results in an electron donation
from Co or Ni to Mo which decreases the bond strength of Mo-S and strengthens the Co–
S bond to intermediate metal–sulfur bond strength which is in optimum range for HDS
activity. The studies have also shown that based on the support interaction, two types of
Co–Mo–S structures exist, i.e. type I and type II. Comparing two structures, catalysts with
type II structure are more active than the ones with type I structure. Type I structures were
suggested to be incompletely sulfided with remaining Mo–O–Al linkages to the support.
Such linkages are thought to appear due to the interaction during calcination step between
Mo and OH groups on alumina surface forming oxygen bridged structures that are difficult
to sulfide completely. Type II Co(Ni)–Mo–S phase, on the other hand, are fully sulfided
and have weak interaction with support. This structure can be formed by breaking all the
Mo–O–Al linkages through the use of high temperature sulfidation, additives, chelating
agents or weakly interacting supports such as carbon [159].
2.8 Magnetic core and catalyst composite
Previously, the advantages of using nano-sized catalysts were mentioned, however, some
problems arise with the size of these catalysts. First is the difficulty in separating such
catalysts from the product streams and the second one is preventing agglomeration of these
nano-sized catalysts while performing under severe industrial conditions. To do so, active
material can be deposited on a suitable support. In case of using a nano-sized catalyst, a
suitable support should have magnetic properties which enables the separation of spent
nano-sized catalyst from the oil stream by an external magnetic device. This technique is
34
called magnetic carrier technology (MCT) and has been reported by several research
groups. Among different magnetic materials, magnetite (Fe3O4) has caught a lot of
attention because of the excellent ferromagnetic properties which resulted in wide
application of it for production of magnetic materials and catalysts. As the goal is
synthesizing a nano-sized catalyst, therefore, the size of the magnetic support should also
be in that range.
2.9 Summary
Magnetic carrier technology has been applied in different fields; however, to the best of
our knowledge no work has been done on the applications of the MCT in the hydrotreating
industry. Forming composites containing both a magnetic core with a catalyst layer is
challenging and yet necessary to reach our desired goal for synthesizing a recyclable
catalyst that can be used in the slurry reactors for hydrotreating of heavy and extra-heavy
oils.
Different active sites are recognized over TMS catalysts. Many studies have attempted to
shed light on how H2 and H2S dissociate and active sites on these catalysts. However, the
differences in the roles of H2 and H2S in the formation of active sites is yet to be
understood. Finding the differences between these two molecules and their activation at
the edges of the catalysts can help us to better understand how the HDS mechanism work.
35
Chapter 3: Analytical procedures for catalyst characterization and evaluation
3.1 Characterization techniques for catalysts
The following characterization methods were used in the thesis:
(1) Transmission electron microscopy (TEM), equipped with selected area electron
diffraction (SAD) and energy-dispersive X-ray spectroscopy (EDX). The analysis was
performed at the microscopy facility at University of New Brunswick.
(2) Scanning electron microscopy (SEM). The analysis was performed at the microscopy
facility at University of New Brunswick.
(3) Nitrogen adsorption-desorption. The analysis was performed in the Catalytic Process
Lab at University of New Brunswick.
(4) Elemental analysis. The analysis was performed in the Catalytic Process Lab at
University of New Brunswick.
(5) X-ray diffraction (XRD). The analysis and data processing were performed at
department of forestry at University of New Brunswick.
(6) Magnetization. The analysis was performed at physical properties lab at Dalhousie
university.
(7) Thermal Gravimetric Analysis (TGA), The analysis was performed in the Catalytic
Process Lab at University of New Brunswick.
36
(8) Surface charge density. The analysis was performed in the pulp and paper center at
University of New Brunswick.
(9) UV-Vis. The analysis was performed in the Catalytic Process Lab at University of New
Brunswick.
3.1.1 Transmission electron microscopy (TEM)
Transmission electron microscopy (TEM) was performed on an electron microscope
(JEOL 2011 STEM, JEOL Ltd., Tokyo, Japan) operating at 200 keV. The catalyst powder
was ultrasonically dispersed in ethanol and deposited on a carbon-coated copper grid, then
vacuum-dried for 12 hours before analysis. The layer numbers and length of catalyst
crystalline structure were determined using image analysis software and their average were
calculated according to Equation 3.1 and Equation 3.2 based on at least 100 measurements
from various particles. The standard deviation of the crystalline size parameters is
calculated based on different particles. The atomic ratio of the catalyst was estimated by
energy dispersive X-ray emission (EDX) coupled with TEM.
Average slab length: �� =∑ 𝐿𝑖𝑁𝑖
𝑛𝑖=1,2…𝑛
∑ 𝑁𝑖𝑛𝑖=1,2…𝑛
(Equation 3.1)
Average layer number: �� =∑ 𝑁𝑖
𝑛𝑖=1,2…𝑛
𝑛 (Equation 3.2)
where L, N and n stand for slab length, number of layers in each crystal, and the total
number of crystalline measured, respectively.
37
3.1.2 Scanning electron microscopy (SEM)
The morphologies of catalysts were observed by scanning electron microscopy (SEM,
JEOL JSM6400) operating at 25 kV. Before imaging, samples were mounted on carbon
grids with carbon paste, and sputtered with gold to ensure sufficient conductivity.
3.1.3 Nitrogen adsorption-desorption
Nitrogen adsorption-desorption isotherm was measured at 77 K using Autosorb-1
(Quantachrome Instruments, Florida, US). Desired amount of powder samples were
degassed in a sample preparation station at 200-300 °C for 3 hours prior to measurement,
then switched to the analysis station for adsorption and desorption under liquid nitrogen at
77 K with an equilibrium time of 3 minutes. The specific surface area of the catalyst powder
was calculated using the Brunauer-Emmett-Teller (BET) method with linear region in the
P/Po range of 0.10 to 0.30. The total pore volume was calculated from the volume of
nitrogen adsorbed at the relative pressure p/p0 0.995. Pore size distribution was analyzed
from the adsorption branch of isotherms by the Barrett-Joyner-Halenda (BJH) method.
3.1.4 Elemental analysis
Quantitative elemental analysis was conducted with a CHNS-932 instrument (Leco
Corporation) to measure the sulfur and carbon content of the samples. The weight percent
of carbon in the sample was used to determine the approximate amount of the remaining
surfactant after the catalyst synthesis.
38
3.1.5 Magnetization
Magnetic properties of samples were measured at a temperature of 300 K with a Physical
Property Measurement System (Quantum Design).
3.1.6 Thermal Gravimetric Analysis (TGA)
The decomposition temperature of the surfactants in the structure of the catalysts was tested
by a TA Q500 thermogravimetric analyzer.
3.1.7 Surface charge density
Surface charge density of silica coated magnetite particles dispersed in deionised water
were measured at different pH values by polyelectrolyte titration with a particle charge
detector instrument (Mütek PCD 03). The pH of the solutions was changed with diluted
solutions of HCl and NaOH. PolyDADMAK and potassium polyvinylsulfate (PVSK) were
used as cationic and anionic polyelectrolyte sources, respectively. To identify the total
charge density of particles, the specific charge density values obtained from the instrument
were multiplied by the Faraday's constant.
3.1.8 UV-Vis
To examine the effectiveness of the silica coating, a leaching test in HCl solution was
performed. 15 mg of samples were dispersed in HCl solution (1 M) and left for 15 hours.
The amount of iron ions dissolved in the solution was measured by a UV-Vis instrument
(Shanghai Mapada instruments). To prepare the solutions for UV-Vis measurements,
hydroquinon (2ml, 10 g/L of DIW) and o-phenantroline (2.5 g in 100 mL of ethanol and
39
900 mL of DIW) solutions were added to diluted samples (10 ml) and the pH was adjusted
to 3.5 with a solution of sodium citrate.
3.2 Evaluation of catalyst performance
3.2.1 Hydrodesulfurization of model oil in batch reactor
After the synthesis of the catalysts and before the evaluation tests (from chapter 5), the
catalysts were treated by using the following procedure.
Treating procedure: the treating steps were started by drying the suspension solution of
samples in an oven in nitrogen atmosphere in two consecutive steps. First, the temperature
was increased at 10 °C/min to 85 °C and held at that temperature for 2 hours. The
temperature was increased to 120 °C afterwards and held for 1 hour. The dried catalyst was
mixed with dodecane. The mixture was then added to a 25 ml autoclave reactor and the
reactor was purged and then pressurized with H2 to 21 bar and mounted horizontally in a
furnace and rotated at 200 rpm. The reactor was heated up to 340 °C and kept at that
temperature for 1 hour. The catalysts were then removed and washed with toluene.
The evaluation of the catalysts was performed using model oil which consisted of the
following in each stage:
(chapter 4): 1 wt% DBT, 14 wt% Dodecane and 85 wt% Toluene
(chapter 5, 6): 1 wt% DBT, 99 wt% Dodecane
(chapter 7): a) 1 wt% DBT, 99 wt% Dodecane, b) 0.4 wt% DMDBT, 99.6 wt% Dodecane
Prior to HDS reactions, the catalyst suspension in toluene was dried following the same
steps as described in the treating procedure. The dried catalyst was added to model oil using
the weight ratio of 1/250 (1/100 in chapter 4) for catalysts to model oil. The mixture was
40
added to the autoclave reactor. The reactor was pressurized with H2 to 3.4 MPa and
mounted horizontally in a furnace and rotated at 200 rpm. The HDS reaction was carried
out at 320 °C for 1 or 2 hour after the reactor was heated up to this temperature in 30
minutes. The collected products were analyzed with GC-MS (Shimadzu GCMS-QP5000)
instrument for qualitative and Varian 450-Gas Chromatography equipped with both FID
and PFPD detectors for quantitative analysis.
The DBTs conversion was calculated via the equation 3.3.
DBT conversion (%) =[initial DBT]−[final DBT]
[initial DBT] ×100 (Equation 3.3)
The selectivity of the compounds was defined as the concentration ratio of one specific
product compound to the total product obtained at the end of the reaction. Each of the
experiments were repeated 3 times and the average results are reported here. The error bars
were used to show the standard deviation of the repeating results.
3.2.2 Hydrodeoxygenation of model oil in batch reactor
After the treating and drying steps described earlier, calculated amount of dried catalyst
was added to 6 g of model oil (5 wt% Stearic acid, 85% Toluene, and 10% dodecane) in
an autoclave to reach acive phase (MoS2 with or without Co/Ni promotion) to model oil
ratio of 1/250. The reactor was purged of any residual air, pressurized with H2 to 3.5 MPa,
mounted horizontally in a furnace and rotated at 200 rpm. Figure 3.1 shows the schematic
representaion of the set up. The HDO reaction was carried out at 340 °C for 1 hour after
heating up the reactor for 40 minutes. The components in the oil products were identified
by GC-MS (Shimadzu GCMS-QP5000). The temperature program was as follows: 5
41
minutes at 60 °C, followed by a linear increase at 10 °C /min to 180 °C, kept for 5 min,
then 8 °C /min to 240 °C, maintained
Figure 3.1: Schematic representation of reactor, stirrer, and furnace set-up for heating and rotating
the autoclave reactor.
for 5 min, and another increase at 20 °C/min to 300 °C, kept for 2.5 min. the quntitative
analysis was performed by GC (varian GC-450, USA) using the standard curves of the
compounds.
The conversion of stearic acid (SA), C18/C17 ratio, and ratio of paraffins to olefins is
defined as:
SA conversion (%) =[initial SA]−[final SA]
[initial SA] ×100 (Equation 3.4)
C18/C17 =mass of C18 hydrocarbons
mass of C17 hydrocarbons (Equation 3.5)
Paraffin/Olefin =mass of paraffins
mass of olefins (Equation 3.6)
42
CO and CO2 and other generated gases as products during hydrodeoxygenation were
analyzed using a gas chromatography Varian 3400 coupled to a Mass spectrometer.
Chapter 4: Development of novel magnetically recyclable MoS2 catalyst for
direct hydrodesulfurization
4.1 Introduction
Nanocomposite materials with magnetic properties have been studied extensively and have
found a wide range of applications from drug delivery and magnetic resonance imaging to
separation processes [66, 160-162]. Recently, interest in catalysts with magnetic carriers
has increased in terms of different catalytic purposes, e.g. hydrogenation of ketone [163-
165] and olefin epoxidation [166]. However, to the best of our knowledge, no work has
been done on magnetically separable catalysts for hydrodesulfurization of heavy crude.
Magnetic carriers applied for this purpose are usually in nano-size range and can be
magnetized by applying an external magnetic field. This facilitates the removal and
recycling of the catalysts from the products especially if the catalyst is utilized for the
hydroprocessing of heavy and extra-heavy oil. Therefore, the catalyst can be used in a
slurry reactor without concerns about recycling. In the present study, we report a novel
method to prepare nano-composites in which magnetite is used as a support, while MoS2
catalyst is synthesized through hydrothermal reaction. The resulting catalysts are then
characterized and tested for the hydrodesulfurization (HDS) with dibenzothiophene as a
model sulfur containing compound.
43
4.2 Catalyst synthesis
4.2.1 Preparation of silica coated magnetite (SiO2/Fe3O4)
Magnetite particles (>98%) with mean diameter of 50 nm were purchased from Sigma
Aldrich. These particles covered with a thin layer of silica were prepared through an
aqueous reaction using sodium silicate and sulfuric acid as described by Wu et al.[167]
with some modifications. In a typical procedure, first the magnetite particles (0.35 g) were
dispersed in deionised water (DIW) with the help of ultrasonication and mechanical
mixing. The solution was then moved to a water bath at 85 °C. Sodium silicate (15 ml, 1
M) and an adequate amount of sulfuric acid (0.3 M) were added drop-wisely to the solution
under stirring, while the pH was monitored with a pH meter and kept at 9.5±0.5. The
solution was cooled down at room temperature, while stirring and left for 3 hours. The
silica coated magnetite particles were separated from the solution by a hand-held magnet
(neodymium). Afterwards, the particles were washed 5 times with DIW.
4.2.2 Preparation of MoS2/SiO2/Fe3O4
Cetyltrimethylammonium chloride (CTAC, 25% in water) and sodium dodecyl sulfate
(SDS) were used as surfactants in synthesis of surfactant-assisted nano-catalysts. CTAC
and SDS are cationic and anionic surfactants, respectively. In the CTAC assisted synthesis,
silica-coated magnetite particles and different amounts of CTAC (6.5 g) were introduced
to deionized water (150 ml) in sequence and stirred for 30 min. The resulting catalyst with
the highest Mo conversion was named Cat-CTAC. In the case of SDS assisted synthesis
(Cat-SDS), silica-coated magnetite particles (0.350 g) were added to deionized water and
mixed. pH of water solution was adjusted to 2 by using HCl (4 M) before addition of SDS
44
(4.6 g), and then stirred for 30 min. At the end, the aqueous solution containing extra
surfactant was decanted. Surfactant molecules adsorbed on the surface of coated magnetite
particles can facilitate formation of a bond between MoS2 and carrier surface. It also has a
scaffold effect during the formation of MoS2 crystals. A catalyst sample, without addition
of any surfactant, was also prepared as a reference (Cat-R).
The hydrothermal synthesis method for MoS2 in this study was adopted from ref [168] with
some alterations. Briefly, sodium sulfide (0.21 g of, nonahydrate crystals) was dissolved
in DIW (5 ml). Molybdenum trioxide powder (54 mg) was then dissolved in this solution.
Then, DIW (10 ml) and HCl (0.4 ml, 4 M) were added. This solution was mixed with the
previously prepared particles and transferred to a 25 ml autoclave reactor. The autoclave
was mounted in a furnace horizontally and heated to 200 °C for 30 minutes while rotated
at a speed of 200 rpm. The conversion of Mo precursor was measured by an Inductively
Coupled Plasma (ICP) instrument based on the amount of Mo atoms left in the water
solution at the end of the MoS2 synthesis reaction. MoS2/SiO2/Fe3O4 particles were
removed after the reaction by a hand held magnet from the reaction solution and washed 3
times with ethanol.
4.3 Results and Discussion
To prepare the catalysts, first magnetite particles were coated with a thin silica layer
through an aqueous reaction using sodium silicate and sulfuric acid. A process similar in
concept to admicellar polymerization was adopted to upload MoS2 (surfactant-assisted) on
to the silica coated magnetite. Most particles can obtain a surface charge in aqueous
solutions which usually varies with pH. A charged surface can adsorb an oppositely
45
charged surfactant in a bi-layer form [169]. In admicellar polymerization monomers are
adsorbed in this bilayer to form a polymer. Different from admicellar polymerization, the
formed surfactant bi-layers are utilized as sites in this work to grow MoS2 crystals. Three
different catalysts with and without surfactants were synthesised in this work. The catalysts
prepared by the addition of cetyltrimethylammonium chloride (CTAC) and sodium
dodecyl sulfate (SDS) surfactants were coded as Cat-CTAC and Cat-SDS while the
surfactant-free catalyst was named Cat-R.
High saturation magnetization makes magnetite a suitable choice of core material for
preparation of the HDS catalyst. However, to protect the magnetite particles from harsh
HDS conditions and to provide proper surface for the surfactants, a coating layer around
the magnetite particles is desirable. The thickness of the coating layer should be kept as
small as possible to avoid any significant reduction effect on the saturation magnetization
of the particles.
Silica is an ideal coating choice for magnetite as its thickness can be controlled via
manipulating reaction conditions applied for its synthesis. Silica coating can also reduce
the anisotropic dipolar magnetic attraction between the magnetite particles which can cause
severe agglomeration of magnetite particles [170, 171]. Moreover, charged silanol groups
formed on the surface of the silica layer in aqueous solution can be utilized to attach
oppositely charged functional groups to the surface of particles [167, 172]. Silica coating
was applied to all the catalysts synthesized in this project.
46
4.3.1 Textural and magnetic properties
Table 4.1 shows the properties of the core particles and the prepared catalysts. Carbon
contents of the catalysts are primarily attributed to the presence of surfactants in their
structure. Unlike Cat-SDS (SDS-Assisted) and Cat-R (no surfactant), there is a
considerable amount of carbon left in the structure of Cat-CTAC (CTAC-Assisted) after
the synthesis, which is due to the thermally stable nature of CTAC. The catalysts are
synthesized at 200oC while CTAC is not thermally decomposed until 250oC, as evidenced
by the TGA results. On the other hand, SDS starts to decompose at the synthesis
temperature. At the over one hour synthesis, almost all SDS is thermally decomposed.
Carbon residue left in fresh Cat-CTAC disappeared after the hydrodesufurization reaction,
which is expected. The reaction took place at 320oC, temperature at which CTAC can
easily be de-structured. The carbon content of the samples can also be used to calculate the
amount of surfactant present in the synthesized catalysts.
The sulfur content of the fresh and spent catalysts showed no noticeable loss of sulfur
during the activity test, which indicates that none of the active material was lost. Based on
the sulfur content of the samples and considering the theoretical molybdenum to sulfur
ratio of 1/2, the amount of MoS2 present in each sample can be calculated. Assuming the
weight of the silica layer to be negligible, MoS2 to Fe3O4 ratio was calculated for samples
and results are shown in Table 4.1.
The measured BET surface areas were 35.6, 78.8 and 63.2 m2/g for fresh Cat-CTAC, Cat-
SDS, and Cat-R, respectively. Compared to the coated magnetite core, Cat-SDS and Cat-
R show slightly higher surface area. However, approximately 40% drop in the surface area
is observed in fresh Cat-CTAC. This is due to the presence of CTAC in the structure of
47
Cat-CTAC, which reduces the available surface area. This is confirmed by the increase of
the BET surface area of the Cat-CTAC from 35.6 to 72.2 [m2/g] after the activity test while
Cat-SDS and Cat-R showed 10 and 5% reduction, respectively.
As mentioned before, catalyst preparation through admicellar polymerization requires the
surface of the particles to have an opposite charge to the adsorbing surfactant. Therefore,
the effect of pH on the surface charge of silica coated magnetite particles was investigated
and the isoelectric point for these particles suspended in deionised water was found to be
near 3.7. The pH of the deionised water was measured as 6.5, which suggests that the
prepared silica coated magnetite particles were negatively charged while dispersed in
deionised water. This negative charge allows the assembly of CTAC surfactant molecules
with the positive ends bonded to the surface of particles. In the case of the SDS surfactant,
which is anionic, the pH of the solution was decreased to 2 before the addition of surfactant
to positively charge the silica coated magnetite particles.
To check the effect of silica and MoS2 layers on the magnetic properties of magnetite, a
magnetization test was performed. The results are shown in Figure 4.1. Bare magnetite
particles show the highest saturation value (76 emu/g) followed by SiO2/Fe3O4 (69 emu/g)
and MoS2/SiO2/Fe3O4 (59 emu/g). It is clear that the additional layers covering magnetite
particles have a negative effect on the magnetization of samples. However, this reduction
in magnetization is not substantial and all three samples still have very high saturation
magnetization and negligible coercivity. Moreover, none of the magnetization curves show
any remanence. These are all characteristics of superparamagnetic materials.
48
Table 4.1: Properties of the prepared catalysts.
Sample
Sulfur
[±0.1
wt.%]
Carbon
[±0.1
wt.%]
MoS2/
Fe3O4
ratio
specific
surface area
[m2/g]
Average slab
length [nm]
Average
number
of layers
SiO2/
Fe3O4
- - - 61.2 - -
Cat-
CTAC[a
]
4.9 7.7 0.16 35.6 4.2 ± 2.1 1
Cat-
SDS[b]
5.5 0.2 0.16 78.8 45.2 ± 27.1 3.0
Cat-R[c] 5.4 0.1 0.16 63.2 56.1 ± 31.8 3.8
[a] Cat-CTAC -MoS2/SiO2/Fe3O4 with CTAC; [b] Cat-SDS - MoS2/SiO2/Fe3O4 with SDS;
[c] Cat-R - MoS2/SiO2/Fe3O4 with no surfactant.
49
Figure 4.1: Magnetization curves for Fe3O4 (circle), SiO2/Fe3O4 (square) and
MoS2/SiO2/Fe3O4 (triangle) measured at 300 K.
Figure 4.2 (a, b) shows the TEM image of the magnetite particles before and after silica
coverage. The silica layer is marked by an arrow in the picture. Each magnetite particle has
a separate coating layer of silica 3-6 nm in thickness, however, the silica coatings are
attached together forming agglomerates of coated magnetite particles. As shown in Figure
4.2 (c, d), MoS2 formed with CTAC is dispersed over the surface of these magnetic
agglomerates. MoS2 crystals are in monolayer slabs only and are on average 4.2 nm in
length (refer to Table 4.1 for the average slab length and number of layers for each sample).
50
(c)
(a) (b)
(d)
(e) (f)
51
Figure 4.2: TEM images of a) untreated magnetite (Fe3O4), b) silica coated
magnetite (SiO2/Fe3O4), c) & d) Cat-CTAC, e) & f) Cat-SDS and g) & h) Cat-R.
Figure 4.2 (e, f) shows the SDS-assisted synthesised MoS2. Unlike the CTAC-assisted
sample, the SDS-assisted MoS2 crystals are multilayered and very long. The average slab
length of 45.2 nm and average number of layers of 3.8 were measured for this sample. The
MoS2 structure of Cat-R is shown in Figure 4.2 (g, h). The structure of MoS2 in this sample
is similar to that of Cat-SDS; however, the average length and layers of slabs were found
to be slightly larger. The evidence may suggest that surfactant CTAC provides growing
sites for MoS2 and the scaffold effect of surfactant limits the aggregation of MoS2 slabs.
When CTAC is replaced by SDS, SDS is not fully functional. It is due to the fact that SDS
is quickly thermally decomposed at 200oC during catalyst synthesis, which leads to
disappearance of scaffolding effects and crystal growing sites. Then, MoS2 crystals start to
aggregate and become long and thick slabs.
(g) (h)
52
Table 4.2: Effect of the amount of CTAC on the conversion of Mo and the final pH.
CTAC added [g] Mo conversion [%]
pH of solution after
synthesis
0 61.12 4.24
0.1 75.10 4.41
0.25 84.76 4.95
0.75 89.28 5.25
2.5 95.22 5.36
6.5 99.99 5.42
4.3.2 Effect of CTAC levels on conversion of Mo precursor and structure of
catalyst
Table 4.2 shows the effect of amounts of CTAC on the conversion of Mo precursor (MoO3)
to MoS2 and the final pH of the synthesis solution. Mo conversion and pH of the final
solution were increased as the CTAC content was increased. The increase in pH is directly
related to the consumption of HCl during the conversion of precursors to MoS2 as the
reaction moves forward. CTAC shows a positive effect on the formation of MoS2 crystals
from the precursors. As mentioned before, this positive effect can be obtained by providing
suitable sites for the formation of MoS2 crystals that also enhances the supersaturation state
by decreasing the nucleation energy barrier. Therefore, compared to a surface without the
adsorbed surfactant, more nucleation can occur in the first step of crystal formation for
53
surfaces covered with surfactant. This allows for faster formation of crystals from the
precursors which reduced the overall synthesis time needed for their complete conversion.
Moreover, the long carbon chain of the surfactant works as a barrier preventing
agglomeration of the formed crystals while in the absence of the surfactant, MoS2 crystals
can easily agglomerate to form longer crystals with more layers as evidenced by Figure
4.2.
Figure 4.3 shows the effect of the amount of CTAC on the crystal structure of the formed
MoS2 after 1hr of synthesis reaction. At low levels of CTAC (0.1 g), the formed crystals
show their tendency toward agglomoration while at higher levels (0.75 g), they appear to
be better separated.
Comparing the TEM images in Figure 4.2 and 4.3 for different amounts of CTAC (0.1,
0.75 and 6.5 g), it can be seen that increasing the CTAC increases the number of MoS2
particles crystallized from its amourphous shape. Applying a reduced amount of CTAC
will lead to the formation of fewer nuclei, and monomers are more likely to attach to the
existing nuclei rather than creating new ones. However, crystals will mainly grow through
the continuous growth of individual nuclei. Moreover, in the absence of a suitable amount
of surfactant present over the particles, parts of the MoS2 crystals can form separately from
the particles.
54
Figure 4.3: Effect of CTAC levels a, b) 0.1 g and c,d) 0.75 g on the structure of the
synthesised MoS2 after 1 hr of reaction at 200° C.
4.3.3 Catalytic activity and separation
DBT is desulfurized through two main pathways, i.e. direct desulfurization (DDS) and
hydrogenation (HYD) [82]. A simplified reaction network for these HDS routes is shown
in Figure 4.4. In DDS route, DBT undergoes direct C-S bond cleavage to form biphenyl
(b)
(c) (d)
(a)
55
(BP). In HYD route, cyclohexylbenzene (CHB) is formed via first hydrogenation of DBT
to tetrahydrodibenzothiophen (TH-DBT) leading to further breakage of C-S bond in this
molecule. The preference of HDS reaction for DBT between these two routes mainly
depends on the type of the active catalytic material, additives or promoters to the catalyst
and support [173]. The results for hydrodesulfurization of DBT are shown in Table 4.3.
The main DBT reaction products were BP and CHB with traces of TH-DBT. The
selectivity of the catalysts for DDS and HYD routes (DDS/HYD) can approximately be
calculated by dividing the mole ratio of the products from the direct desulfurization route
and hydrogenation route ([BP]/[CHB+THDBT]) [174]. Comparing the activity of Cat-
CTAC with Cat-SDS and Cat-R, the DBT conversion over Cat-CTAC (80%) is nearly
doubled when compared to the other catalysts (45% for Cat-SDS and 35% for Cat-R),
showing significantly higher activity of Cat-CTAC in HDS of DBT. The small size of
MoS2 crystals and their high dispersion in Cat-CTAC can be the main reason for this high
activity. These properties can provide the catalyst with high number of sulfur vacancies at
the edge sites. These corner sites are believed to be active for breaking the C-S bond of
sulfur containing molecules [175]. The considerably higher amount of DDS product (BP)
over Cat-CTAC also points toward the same reason for higher activity of Cat-CTAC.
Compared to Cat-R, Cat-SDS shows slightly higher activity. SDS is probably playing the
same role as CTAC, however as mentioned before, the starting temperature for SDS
decomposition is 200 °C. Therefore, at the early stages of crystal formation over the
particles SDS was still present but was decomposed quickly. The rapid decomposition of
SDS can also be determined from the TEM images showing similar structure between
56
catalysts 2 and 3, i.e. long and multilayered slabs. This will leave a weak but noticeable
effect on the dispersion and therefore activity of the prepared catalyst.
Figure 4.4: Simplified reaction pathways for hydrodesulfurization of DBT.
The ratio of DDS to HYD reaction routes for Cat-CTAC, Cat-SDS, and Cat-R were 28.4,
8 and 3.7, respectively, showing predominant DDS pathway especially for Cat-CTAC. The
DDS/HYD ratios over MoS2 catalyst, with or without any support, generally fall in the
range of 0.12 to 2 in the literature.[41, 112, 176-179] The ratio for MoS2 prepared by in
situ decomposition of thiosalt was found to be 2.[179] It is shown that agglomeration of
Mo crystals promotes the HYD reaction route.[180, 181] The TEM pictures of the prepared
catalysts show that compared to Cat-CTAC, Cat-SDS and Cat-R exhibit higher degree of
agglomeration with the same Mo loading, which can be the reason for their lower ratio of
DDS to HYD.
57
Table 4.3: DBT conversion and selectivity of the prepared catalysts.
Catalyst
DBT
conversio
n [%][a]
Products
DDS/HY
D
Ratio
BP [CHB+THDBT]
Amoun
t [ppm]
Selectivit
y [%]
Amoun
t [ppm]
Selectivity
[%]
MoS2/SiO2/Fe3
O4 with CTAC
(Cat-CTAC)
80 7468 96.6 263 3.4 28.4
MoS2/SiO2/Fe3
O4 with SDS
(Cat-SDS)
43 3648 88.9 456 11.1 8.0
MoS2/SiO2/Fe3
O4 with no
surfactant
35 2645 78.1 715 21.9 3.7
[a] Reaction conditions: 1 wt.% DBT solution reacted for 1 hour at 320 °C and 3.5 MPa H2.
The magnetization test showed the high magnetization ability of the prepared catalysts
which was further confirmed by exposing the catalysts to an external magnetic field. Figure
4.5 (a, b) shows Cat-CTAC dispersed in model oil after 1 hour of HDS reaction (left) which
is separated afterwards with an external magnet (right). It can be seen that the catalyst can
be easily separated from the reaction media after a very short time (5 seconds). All of the
samples showed the same magnetic effect.
58
Figure 4.5: prepared catalyst after the HDS reaction, (a) dispersed inside the model
oil and (b) separated from the model oil after 5 seconds of exposure to an external
magnet.
4.4 Conclusion
We have demonstrated the preparation of superparamagnetic MoS2-based catalysts through
a novel and easy procedure. Three catalysts were synthesized, two with the help of
surfactants (CTAC and SDS) and one without the addition of any surfactant. Cat-CTAC
prepared through a combination of admicellar polymerization and hydrothermal synthesis
showed the highest activity (nearly twice as high as the other catalysts) in HDS of DBT.
This high activity was related to the size and structure of MoS2 crystals in Cat-CTAC. The
catalysts also showed an exceptional tendency toward the direct desulfurization route for
removing sulfur from DBT. The DDS/HYD was found to be especially high for Cat-
CTAC, showing a significant ratio of 28.4 which is substantially higher than the numbers
reported in the literature. It was found that surfactant plays an important role in the structure
and activity of the synthesized catalysts. Surfactant can also accelerate the speed of MoS2
production from its precursors through providing suitable sites for the crystals to grow.
(a) (b)
59
Being superparamagnetic gives a vital advantage to this catalyst allowing easy separation
and reuse, which can be very useful in processing of heavy and extra-heavy oils in slurry
reactors.
60
Chapter 5: New Insights on the roles of H2 and H2S and the involved active sites
of MoS2 in hydrodesulfurization of Dibenzothiophene
5.1 Introduction
We previously reported a MoS2 catalyst with magnetite particles as core material which
showed excellent capability of hydrodesulfurization via the DDS pathway. We believe that
this catalyst can help us better understand the HDS process over MoS2. In this study, we
used the same catalyst in line with other catalysts to better understand the role of H2 and
H2S in hydrodesulfurization over MoS2. We also made an attempt to address the
disagreement in the literature concerning the types of active sites involved in the HDS
process over MoS2. Furthermore, we investigated the effect of hydrogen pressure on the
activity and selectivity of the catalysts with and without the presence of magnetite as an
adsorbent. To simplify the reaction mechanism, DBT was chosen as the sulfur-containing
model compound.
5.2 Catalyst Synthesis
Briefly, sodium sulfide (0.21 g of, nonahydrate crystals) was dissolved in DIW (5 ml).
Molybdenum trioxide powder (54 mg) was then dissolved in this solution. Then, DIW (10
ml), Cetyltrimethylammonium chloride (200ml, 25% in water) as surfactant, and HCl (0.4
ml, 4 M) were added. Different amount (0, 20, 50, and 100 mg) of magnetite particles
(>98% with mean diameter of 50 nm from Sigma Aldrich) were added to the prepared
solution and sonicated for 5 min. The prepared mixtures were transferred to a 25 ml
autoclave reactor. The autoclave was mounted in a furnace horizontally and heated to 200
°C in 30 minutes while rotated with a speed of 200 rpm. The catalysts were removed after
61
the reaction and washed 3 times with deionized water and ethanol. The prepared samples
were named MoS2, MoS2-20M, MoS2-50M, and MoS2-100M based on the amount of
magnetite added during the preparation. Two other samples were also prepared by simply
mixing 50 mg Fe3O4 and 50 mg Fe powders with the synthesized MoS2 catalyst and called
MoS2-MM and MoS2-MI, respectively.
5.3 Results and Discussion
Table 5.1 shows the properties of the fresh and treated MoS2 and MoS2-100M catalysts.
The freshly prepared catalysts were put through a treating procedure described in the
experimental section to remove the surfactant from the catalysts and to further stabilize the
structure of the catalysts. The carbon content in the fresh catalysts is due to the presence of
the surfactant in the structure of both catalysts. After the catalysts were treated, the
surfactant was mostly removed from their structure. The specific surface area of the
samples shows a great improvement for the treated catalysts compared to the fresh ones
with 19 and 12-fold increase for MoS2 and MoS2-100M, respectively.
62
Table 5.1: Properties of the fresh and treated MoS2-100M and MoS2 catalysts.
Sample
Sulfur
content (±
0.2 wt %)
Carbon
content (±
0.2 wt %)
S/Mo
ratio
Specific
surface area
(m2/g)
Average
slab
length
(nm)
Average
number
of layers
Fresh
MoS2-
100M
9.8 16.4 1.81 18.2 3.5±1.3 1.2±0.2
Treated
MoS2-
100M
16.6 1.7 1.82 216.4 6.5±1.6 2.3±1.5
Fresh
MoS2
21.0 34.0 1.85 24.3 3.6±1.5 1.3±0.2
Treated
MoS2
30.8 2.1 1.87 474.7 6.1±1.7 2.2±0.8
63
Figure 5.1: TEM images of MoS2, a, b) fresh and c, d) after treating.
Figure 5.1 shows the TEM images of MoS2 before (a,b) and after the HDS reaction (c,d).
As the images show, some of the small crystals of the fresh catalyst are joined together to
form longer multilayered slabs. After the treating procedure, the length of the slabs of the
MoS2 approximately increased by 60% and the average number of layers increased from
1.3 to 2.2. A similar increase in the length and thickness was also observed for MoS2-100M
catalysts as shown in Figure 5.2. These treated catalysts were used in the following HDS
activity tests. The treated catalysts show the same average slab length and average number
c
)
d
)
a
)
b
)
64
of layers and during the HDS activity tests the structure of the catalysts remained
unchanged. This means that the amount of catalyst edges exposed for the HDS reactions is
similar. Therefore, studying any difference in the activity of the catalysts, the difference in
the structure of MoS2 catalyst as an interfering factor can be neglected.
Figure 5.2: TEM images of MoS2-100M, a, b) fresh and c, d) after treating.
d
)
c
)
a b
)
65
Figure 5.3: Possible reaction pathways for hydrodesulfurization of DBT.
5.3.1 Reaction routes over MoS2 with and without the support
DBT is desulfurized through two main pathways, i.e. direct desulfurization (DDS) and
hydrogenation (HYD) [82]. Figure 5.3 represents the reaction network for these HDS
routes. In the DDS route, DBT undergoes direct C-S bond cleavage by hydrogenolysis
[182] or by elimination [41] to form bipheny (BP). In the HYD route, cyclohexylbenzene
(CHB) is formed via first hydrogenation of DBT to tetrahydrodibenzothiophen (THDBT)
and/or hexahydrodibenzothiophene (HHDBT), and further breakage of C-S bond in these
molecules. CHB can then be isomerized to form (cyclopentylmethyl)benzene (CPMB).
Bicyclohexyl (BCH) is mainly formed through hydrogenation and desulfurization of
HHDBT. Both BCH and CPMB can isomerize to cyclohexane-cyclopentylmethyl
(CHCPM) which can undergo further isomerization to form Dicyclopentylethane (DCPE).
Preference of HDS reaction between these two routes mainly depends on type of the active
catalytic material, additives or promoters to the catalyst and support [183].
66
Figure 5.4 shows the DBT conversion and product distribution of MoS2 catalysts with
different amount of magnetite particles. The mass ratio of MoS2-to-model oil was kept the
same for all the catalysts. The selectivity of the isomerized products is shown in Figure 5.5.
The DBT conversions for MoS2-100M and MoS2 catalysts are 48 and 63 percent,
respectively. The DBT conversion for MoS2-50M and MoS2-20M were found to be 47 and
52 percent, respectively. Overall, the results show that the increase in the amount of
magnetite leads to increase in the selectivity of BP but decrease in the selectivity of CHB
and BCH and therefore causing a great change in the DDS/HYD ratio. The selectivity of
BP decreases from nearly 80% for MoS2-100M to 11% for MoS2 with no magnetite core.
A reverse trend can be observed for hydrogenated products (CHB, BCH, and their
isomerized forms). The selectivity for CHB showed an increase with the decrease in the
amount of magnetite. No BCH was detected for MoS2-100M and MoS2-50M. 3.5% and
5% of selectivity on BCH was observed over MoS2-20M and MoS2 catalysts. This clearly
shows that the presence of magnetite particles in the samples shifts the preference of the
reaction toward DDS. The decrease in the magnetite content of the catalysts (below 50 mg)
increases the selectivity of the isomerized compounds (CPMB, CHCPM, and DCPE).
67
Figure 5.4: DBT conversion after 2h of reaction and product selectivity at 50%
conversion for the synthesized catalysts.
Figure 5.5: The selectivity of the isomerized products over the synthesized catalysts
at 50% conversion of DBT.
48
1.7
19.4
77
47
3.1 2.8
27.5
64.3
52
6.3
2.4
42.7
3.5
31.2
63
9.8
3.3
47.4
4.9
11.8
46
3.11.8
29.6
63.0
67
2.5 1.3
59.9
2.5
25.8
0
10
20
30
40
50
60
70
80
90
DBT conversion THDBT HHDBT CHB BCH BP
DB
T co
nve
rsio
n o
r Se
lect
ivit
y (%
)
MoS2-100M MoS2-50M MoS2-20M MoS2 MoS2-MM MoS2-MI
1.9 2.3
4.7
2.1
7.1
5.9
3.7
13.2
2.5
4.3
0.6
3.1
0
2
4
6
8
10
12
14
16
CPMB DCPE CHCPM
DB
T co
nve
rsio
n o
r Se
lect
ivit
y (%
)
MoS2-100M MoS2-50M MoS2-20M MoS2 MoS2-MM MoS2-MI
68
5.3.2 Inhibiting effect of H2S on DDS pathway
Different amount of H2S gas was detected at the end of reactions for each sample. While
no H2S was found for MoS2-100M, MoS2-50M, MoS2-MM (MoS2 mixed with 50 mg of
magnetite) and MoS2-MI (MoS2 mixed with 50 mg of iron powder) samples, 0.62 and 1.8
µmol of H2S was detected for MoS2-20M and MoS2 samples, respectively. The main
reason for the difference between the H2S content in each sample can be related to the
ability of magnetite particles to adsorb the produced H2S from the gas phase. Considering
the reduction in BP and increase in the HYD selectivity with the decrease in the amount of
magnetite, it can be concluded that H2S has an inhibiting effect on DDS reaction pathway.
The distribution of products in the HDS of DBT over MoS2 catalyst at different reaction
times also confirms this effect of H2S. These results are shown in Table 5.2. At 0 min (after
the heat-up procedure), the main product is BP with selectivity of 58.3%. After 30 min, the
selectivity of all the hydrogenated products showed an increase specifically that of CHB
that was increased to 26%. The yield of BP also increased, nevertheless, the selectivity of
this compound was reduced to 40.8% due to the significantly increased hydrogenated
products. The yield of BP at 30 min of reaction reaches nearly 65% of its final value (after
2h). At the beginning of the reaction when the H2S concentration is low, little hydrogenated
products are generated. As the reaction proceeds the H2S concentration is continuously
increased so is the yield of hydrogenated DBT products (THDBT and HHDBT). After 2-
hour reaction the selectivity for BP turns out to be low. Quick accumulation of
hydrogenated products at the late stage of the reaction when the concentration of H2S
increases, all evidences indicate that H2S has an inhibiting effect on the desulfurization
sites over MoS2 catalyst.
69
Earlier reported results in the literature [41, 43, 184, 185] also demonstrated that H2S is a
strong inhibitor for the DDS pathway in HDS of DBT over MoS2 catalyst while only weak
inhibition effect has been observed for HYD pathway. DFT studies by Logadóttir et al.
[186] showed that the adsorption energy of H2S on both Mo and S edge is similar
suggesting that the adsorption of H2S is equally likely on both edges. However, the
dissociation of H2S on the S-edge was found to be highly exothermic which suggests a
Table 5.2: Product selectivity (%) over MoS2 catalyst for HDS of DBT after 0 and 30
min of reaction.
MoS2-0 min MoS2-30 min MoS2-2 hr
THDBT 20.6 18.4 10.0
HHDBT 12.5 9.4 3.2
CHB 8.6 25.7 47.1
BCH 0 0.5 5.1
CPMB 0 4.5 11.1
DCPE 0 0.1 6.1
CHCPM 0 0.7 3.8
BP 58.3 40.8 13.6
70
strong adsorption on the S-edge sites, explaining the strong inhibition effect by H2S over
the DDS pathway.
The highly favorable DDS pathway over MoS2-100M shows that the HYD sites are
different from DDS sites. If both reaction routes were taking place on the same sites,
without the presence of any synergetic effect, the HYD pathway should have been more
favorable on MoS2-100M. The STM imaging has illustrated the presence of so-called brim
sites with metallic characteristics near the Mo-edge and on the basal plane of MoS2 [27,
31, 187]. These sites were found to be active for hydrogenation and possibly
desulfurization. As the basal plane in MoS2 is fully saturated, the H2S can only weakly
adsorb on brim sites and therefore it does not affect the hydrogenation at this point.
Molecules adsorption to these sites is through π-bonding and because of their position,
even large molecules can easily adsorb to these sites without any steric hindrance effect.
For hydrogenation to take place on the coordinatively unsaturated sites at the edges through
π-bonding, at least two free neighboring sites should be available which is found to be
thermodynamically unfavorable for MoS2 [188-190].
5.3.3 H2 and H2S dissociation and involved active sites
MoS2 consists of layered hexagonal sheets as S–Mo–S sandwiches. In the pure compounds
the sheets are stacked and held together by van der Waals forces. A sheet has two low-
index edge terminations, the (1010) Mo-edge and the (1010 ) S-edge [191]. The results of
a DFT study for HDS of Thiophene on MoS2 catalysts at realistic conditions suggests that
the HYD pathway is initiated by hydrogenation at the Mo edge, because thiophene
preferentially adsorbs at the Mo edge and hydrogenation is energetically unfavorable at the
71
S-edge. In contrast, the S–C scission reaction can occur at both the Mo and S edges; which
edge is preferred depends on the reaction conditions. Specifically, S–C scission preferably
occurs at the Mo-edge at high H2S partial pressure and low H2 pressures (PH2 < 80 bar and
PH2S > 0.1 bar), whereas the S-edge is more active at low H2S partial pressure or high H2
partial pressures. This is due to the presence of different forms of adsorbed hydrogen and
the resulting changes in the availability of S-edge vacancy sites, which exhibit a low barrier
for S–C cleavage [192]. It was also found that the electron back-donation from the
adsorption sites to the adsorbed thiophene molecules is much stronger on S-edge than Mo-
edge. Such electron back-donation is suggested to play a role in the activity of the edge
sites since it may result in a distortion of the thiophene ring, which can strongly activate
the adsorbed thiophene toward the following reactions [37, 193]. Based on these previously
published results and the reaction conditions in this study (low H2S partial pressure), the
active sites for removing sulfur can be considered to be mainly located on the sulfur edge.
As mentioned earlier, removing the magnetite particles from the MoS2 catalysts results in
the decrease of DDS activity and increase of the HYD. The inhibition of DDS pathway
was correlated to the presence of H2S that can be adsorbed at the active sites for DDS which
are the unsaturated sites at S-edge in our reaction conditions based on the above discussion.
However, the drop in HYD activity with the addition of magnetite particles can be due to
two reasons. First, the removal of H2S which can create acid sites over MoS2 catalyst,
through reaction with Fe3O4. This seems unlikely as the findings in literature show that
absence of H2S has a small effect on the HYD pathway. Second, the spill-over of H from
the MoS2 to Fe3O4 particles as these particles can consume hydrogen to reduce to Fe.
Different studies [194, 195] indicate the possibility of migration and/or mobility of the
72
active hydrogen on the surface. This was also confirmed in another study on isotopic
exchange (H2/D2 and H2S/D2S) of hydrogen adsorbed on the surface of the sulfided
NiMo/Al2O3 catalyst [196]. To test this theory, MoS2 catalyst was mixed with iron powder
which can only adsorb H2S, and therefore hydrogen remains available for the
hydrogenation reactions. Figure 5.4 shows the results for HDS of DBT over MoS2 mixed
with 50 mg of Fe (MoS2-MI). As anticipated, in the presence of Fe, the amount of BP
produced was increased compared to MoS2 catalyst. This shows an increase in tendency
toward the DDS pathway which as mentioned before is due to the adsorption of H2S by Fe
particles.
When compared to MoS2-100M, MoS2-MI also shows high affinity toward the HYD
pathway which proves our assumption about the hydrogen as the main source of
hydrogenation over MoS2 catalyst. However, the 7% difference in the amount of the
hydrogenated products (CHB and BCH) between MoS2 and MoS2-MI catalysts and the
difference in the selectivity of the hydrogenated products between MoS2 and MoS2-MI
catalysts, suggests that H2S may also play a role in hydrogenation. Moreover, the
selectivity of the isomerized compounds shows a significant increase with the increase in
the H2S content. H2S can dissociate into –SH and –H species over CUS sites converting
them into Brønsted acid sites [42, 197]. These acid sites can therefore increase the
hydrogenation ability of the catalyst. The role of H2S in hydrogenation is more profound
in the selectivity of BCH and the isomerized products. As the amount of magnetite
decreases, the selectivity and yield of BCH and the isomerized compounds increases.
Either the increase in hydrogen over the surface of the catalyst or presence of more H2S is
73
the reason for this increase. Considering the fact that no trace of BCH was found over
MoS2-100M and MoS2-50M catalysts when H2S is absent and the low selectivity of BCH
over MoS2-MI catalyst when H2S is also absent, it can be concluded that H2S is the main
participant for the production of BCH. The process for BCH production can be both
through the production of higher amounts of HHDBT, which further desulfurizes to BCH
and/or through direct hydrogenation of CHB. Following the same reasoning, it is clear that
CUS sites covered with dissociated H2S are the main sites for isomerization. The only
exception can be the production of CPMB which is detected even over MoS2-100M and
with high selectivity over MoS2-MI. The formation of CPMB can be either over the HYD
sites or parts of the H2S can dissociate over the CUS edges of the catalyst (even in the
presence of magnetite particles) resulting in the formation of this compound from CHB.
Although no H2S is available over MoS2-MI catalyst, the HDS results show the presence
of isomerized products. Aside from the dissociation of H2S, the Brønsted acid sites can be
created through dissociation of H2 over the edges of MoS2. This will create the same –SH
groups that H2S dissociation at the edges would produce. However, there is a noticeable
difference in the selectivity of the isomerized products over MoS2 (23.5%) and MoS2-MI
(8.5%). Considering the 8.5% selectivity through H2-produced acid sites, 15% of the
selectivity for isomerized products can be attributed to the H2S-produced acid sites which
is almost twice the amount for H2-produced acid sites. The first reason for this can be found
in the different dissociation energies of H-H bond (436 KJ/mol) and H-S bond (339
KJ/mol). Higher H-H bond energy makes the dissociation of H2 more difficult. The second
reason is that upon dissociation of H2S, the oxidation state of the adjacent Mo atoms will
be increased to +4. On the contrary, the dissociation of H2 will reduce the oxidation state
74
of the adjacent Mo to +2 which is less favorable. These facts contribute to the difference
in the formation of acid sites through H2 and H2S dissociation.
As mentioned before, the addition of Fe to MoS2 increased the yield of BP, however, the
yield of BP over MoS2-MI was lower than MoS2-100M catalyst. Two explanations can be
found for this difference in the yield of BP. First, the hydrogenated DBT products (THDBT
and HHDBT) compete over the desulfurization sites with DBT, and they are found to be
more reactive than DBT for desulfurization over MoS2 especially in H2S-free environment
[185]. The higher selectivity for hydrogenated products (CHB and BCH) compared to BP
over the MoS2-MI catalyst also suggests that adsorption of THDBT and HHDBT as
hydrogenated intermediates are more favorable than DBT over the desulfurization sites.
Second, part of the produced BP may convert to CHB. To test this theory, 0.5 wt% BP
model oil solution was tested in the same HDS conditions as DBT over MoS2 catalyst. The
results showed 8% conversion of the BP to CHB. Therefore, as the effect of hydrogen in
the reactions increases with the reduction in the amount of magnetite in the catalysts, a
greater part of the produced BP can be converted to CHB. It is clear that in real reaction
conditions, BP molecules have to compete with other molecules which will significantly
reduce the converted percentage of these molecules to CHB.
Considering the inhibiting effect of H2S and the available sites discussed above for the
hydrogenation and desulfurization, the activity of the catalysts toward HDS of DBT can
also be elucidated. The presence of magnetite as the support and its synergetic effect
significantly influences the activity and selectivity of the catalyst. For MoS2-100M and
MoS2-50M catalysts, the adsorption of the dissociated hydrogen atoms from the catalyst
75
by magnetite particles and the low barrier for S-C bond cleavage over the S-edge resulted
in the mainly DDS pathway for this catalyst. The presence of magnetite as the support leads
to the full exposure of the active sites for desulfurization in this catalyst. This feature allows
the catalyst to have high tendency toward DDS pathway while maintaining its high activity.
For MoS2-50M catalyst, the lower amount of the magnetite particles present reduces the
contact area between the MoS2 and the magnetite particles. Therefore, the hydrogen spill-
over effect has less profound effect on the hydrogenation activity of the catalyst which
results in higher CHB formation. In MoS2-20M catalyst, part of the MoS2 is in contact with
the magnetite particles while the other part is mostly free of any contact with the magnetite.
Therefore, the former acts in the same way as the MoS2-50M catalyst and the latter works
in the same way as MoS2 catalyst. The increase in the HYD activity increases the overall
activity of the catalyst. As for MoS2, the brim sites will be fully functionalized by the
complete presence of the dissociated hydrogen, compensating the inhibiting effect of the
H2S on the activity of edge sites.
The DDS pathway on the S-edge works through the adsorption of the DBT via its sulfur
atom [198]. The removal of sulfur from the adsorbed molecule can proceed through
hydrogenolysis or elimination, both of which need hydrogen for the process. The DDS
pathway is the main route of HDS over MoS2-100M catalyst. This catalyst is in a hydrogen
deficient state due to the spill-over effect described before. Despite this and the fact that
catalyst needs hydrogen for the DDS pathway, it is interesting to see that hydrogen is
provided for the DDS pathway while the HYD pathway is retarded. The explanation may
be found in the adsorption energy of the DBT molecule on the main HYD (Mo-edge brim)
76
and DDS (sulfur vacancy at the s-edge) sites. A DFT study by Lauristen et.al [31] shows
that the adsorption energy of DBT through π-bonding over the Mo edge brim sites with S
monomers is nearly -0.1 ev. In another DFT study [199], the adsorption energy of DBT
over sulfur vacancies was found to be -0.5 ev. These numbers clearly show that adsorption
of DBT on sulfur vacancies at the S-edge are more stable than the brim sites. Therefore,
upon the adsorption of the DBT molecules on the catalyst, the ones that are adsorbed at the
DDS sites will stay adsorbed longer and higher percentage of them will be on the catalyst
at a certain time. Meanwhile, the hydrogen is dissociated constantly over the catalyst. In
the absence of any adsorbed DBT molecule, they will be transferred to magnetite particles,
however, in the presence of those molecules they will be consumed to form either a
hydrogenated or desulfurized product based on the adsorption site of the DBT molecule.
The presence of higher percentage of DBT molecules adsorbed on the sulfur vacancies will
therefore result in higher production of BP. Moreover, the adsorption of H2S by magnetite
yields a highly reductive atmosphere. The H bond at the Mo-edge is found to be nearly
twice weaker than the bond at the S-edge at low H2S. This means that hydrogen
displacement on the S-edge in such environment will be slower [192, 200]. More stable
hydrogen on the S-edge increases the probability of its adsorption by DBT rather than
magnetite.
Many studies have focused on finding the nature of the active sites. While some believe
that there is only one type of active site for the DDS and HYD reaction routs, most of the
studies agree with the presence of two different active sites for hydrogenation and
desulfurization. Our results show that in general, there are two different sites available on
the MoS2 catalyst, one for desulfurization and the other one for HYD. However, in the
77
presence of H2S, the desulfurization sites can transform to HYD/isomerization sites due to
their strong acidic nature creating a third type of active sites. Moreover, our results indicate
that while H2 can dissociate on both Mo and S-edges of the MoS2 catalyst, H2S can only
dissociate on the S-edge. Nonetheless, one can expect to see the change in the active sites
based on the type of the catalyst and the operating conditions.
5.3.4 The effect of H2 pressure
Figure 5.6 shows the conversion of DBT for MoS2 and MoS2-100M at different H2
pressures. It is clear that the increase in the H2 pressure results in the higher conversion of
DBT. Figure 5.7 shows the selectivity of products for MoS2 catalyst. As the pressure
increases, the selectivity of the hydrogenated products (CHB and BCH) and their isomers
also increases while the selectivity of BP drops from 25.1% at 7 bar to 11.1% at 41 bar.
The results show a direct relation between the H2 pressure and the activity and selectivity
of the MoS2 catalyst up to 24 bar. This increase can be related to the increase in
hydrogenation ability of the catalysts with more hydrogen dissociation at the Mo-edge and
increase in the vacancy formation by the addition of more hydrogen to the S-edge of the
catalyst. The selectivity of the products for the MoS2-100M catalyst is shown in Figure 5.8.
Unlike MoS2, the selectivity of the products remains constant over the first pressure range
(7-24 bar) with BP as the main product with 90% selectivity. This can be due to the
adsorption of the H atoms by the Fe3O4 particles. Over 24 bar, the hydrogenation activity
of the catalyst increases resulting in higher selectivity for CHB.
The rate of increase for the selectivity of the CHB shows decline for the H2 pressure range
of 24 to 41 bar compared to the 7 to 24 bar range over MoS2. A different trend is observed
78
for MoS2-100M in which the selectivity of the CHB increases after being constant up to 24
bar. These results point to the presence of a different type of hydrogen over MoS2 at higher
pressures (over 24 bar) that is active for hydrogenation/isomerization as well as
desulfurization over the S-edge. It is found that unlike the Mo edge, hydrogen can adsorb
in excess (more than one H atom per S dimer ) on the S edge and the binding energy of H
decreases for the extra H atoms [192, 201]. The required energy for the hydrogenation
reactions or vacancy formation at the S edge depends on the H2 pressure. High H2 pressure
can provide the S edge with large numbers of the weakly bonded H atoms which gives a
lower barrier than the more strongly bonded H for hydrogenation/isomerization reactions
and for vacancy formation. Different steps in the HDS reaction pathways involve H, and
lower barrier is expected for them if the coverage of the weakly bonded H is significant.
Based on our results and the above discussions, it can be concluded that hydrogen can be
activated in three different ways. First, the dissociated hydrogen present at the Mo-edge
can be considered accountable mainly for the hydrogenation process. Second, the loosely
bonded hydrogen at the S-edge can be considered to work for the
hydrogenation/isomerization and possibly desulfurization. Finally, the strongly bonded
hydrogen at the S-edge is mainly responsible for desulfurization.
79
Figure 5.6: DBT conversion for MoS2-100M and MoS2 catalysts at different starting
H2 pressures.
Figure 5.7: Product selectivity for MoS2 catalyst at different starting H2 pressures.
0
10
20
30
40
50
60
70
0 10 20 30 40 50
DB
T co
nve
rsio
n (%
)
H2 pressure (bar)
MoS2 MoS2-100M
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
0 10 20 30 40 50
Sele
ctiv
ity
(%)
H2 Pressure (bar)
THDBT HHDBT BP CHB
BCH CPMB DCPE CHCPM
80
Figure 5.8: Product selectivity over for MoS2-100M catalyst at different starting H2
pressures.
5.4 Conclusion
To better understand the role of H2 and H2S and the involved active sites in the HDS of
DBT over the MoS2, a series of catalysts were synthesized in this study via a hydrothermal
method. The addition of magnetite to the catalysts was found to significantly shifting the
selectivity of the products and the DDS/HYD ratios. Our results evidenced that
desulfurization and hydrogenation took place at two different main sites. The CUS sites
were considered as the main sites for desulfurization and the brim sites were considered as
the major HYD sites.
The high yield of BP for the MoS2-100M catalyst was explained by the difference between
the adsorption energies of DBT over CUS and brim sites as well as the movability of the
0.0
20.0
40.0
60.0
80.0
100.0
0 10 20 30 40 50
Sele
ctiv
ity
(%)
H2 pressure (bar)
THDBT BP CHB
81
dissociated hydrogen on each site. The increase in the selectivity of BP with the increase
in the magnetite content of the catalysts was related to the inhibiting effect of the H2S on
the DDS pathway and the competition between the DBT and its hydrogenated
intermediates for the desulfurization sites. While H2S showed an inhibiting effect on DDS
pathway, it promoted the conversion of CHB to BCH. By increasing the pressure of H2
from 7 to 41 bar, the conversion of DBT increased for both MoS2 and MoS2-100M
catalysts. This was related to the increase in vacancy formation at the S-edge with the
increase in the H2 pressure and the formation of weakly bonded H atoms. H2 was found to
play three distinctive roles. At Mo edge, for hydrogenation, and at S-edge if strongly
bonded as the means for desulfurization and if loosely bonded to the sulfur, for
hydrogenation/isomerization and/or desulfurization and regeneration of the active sites. It
was shown that in the presence of H2S, the desulfurization sites have the ability to transform
to hydrogenation/isomerization sites. Both H2 and H2S showed acid properties over MoS2
capable of isomerizing the main HDS products. However, the results showed that the
creation of acid sites were more facile through H2S dissociation. This was assigned to the
difference in the bond energies for each molecule and the change in oxidation state of the
adjacent Mo atom upon their dissociation.
82
Chapter 6: Edge activation of H2 and H2S over NiMoS and the effect of Ni
loading
6.1 Introduction
We previously reported a MoS2 catalyst with magnetite particles as core material which
showed excellent capability of hydrodesulfurization via the DDS pathway. In the previous
chapter, we studied the activation of H2 and H2S over MoS2 and the involved active sites
using a series of catalysts with and without magnetite core. We believe that this catalyst
can also help us better understand the HDS process over Ni-promoted MoS2. In this study,
we used the same catalyst in line with other catalysts to better understand the promotional
effect of Ni and to find how the Mo substitution with Ni progresses at different edges. We
also made an attempt to investigate the role of H2 and H2S in hydrodesulfurization over
NiMoS.
6.2 Catalyst synthesis
To synthesize the catalysts the same steps as to the previous chapter was followed. In case
of Ni-promoted catalysts, before the hydrothermal synthesis, desirable amount of
Ni(NO3)2 was also added as the Ni precursor to reach different Ni/(Ni+Mo) molar ratios.
The prepared samples were named x-(Ni)MoS2 for samples without magnetite and, x-
(Ni)MoS2-M for samples containing magnetite where x shows the nominal Ni/(Ni+Mo)
ratio. A series of samples (named x-(Ni)MoS2-I) were also prepared by adding 50 mg Fe
powder to the synthesized MoS2 catalyst.
83
6.3 Results
Figure 6.1 shows the TEM images of the freshly synthesized 0.35 NiMoS catalyst. It can
be seen that the catalyst is well dispersed and is consisted of mainly one crystal slabs, and
ca. 3.5 nm in length. Other synthesized samples also showed the same structure. The good
dispersion of the catalyst can be attributed to the presence of surfactant in its structure. We
had found earlier the importance of applying a suitable surfactant which can prevent the
excessive agglomeration of crystals in the catalyst. The surfactant can also assist in
anchoring and dispersing the catalyst to the surface of the magnetite particles. To remove
the surfactant from the catalysts that can block the pores and to stabilize their crystal
structure, the freshly synthesized samples were treated as described in the experimental
section. The TEM images of the samples after the treating procedure are shown in Figure
6.2. All the catalysts showed nearly the same degree of agglomeration after the treatment
procedure increasing their crystal length by 40% and the number of layers by 50%. Further
studies of the catalysts after the HDS reactions showed no significant change in their crystal
structure. This provides a good reference for comparing the catalysts as they all have
similar crystal structures. In all of the catalysts the only detected phase found by EDX
analysis was NiMoS with 0.65 NiMoS as an exception. Ni sulfide phase was detected in
this sample which is circled in the corresponding TEM figure (Figure 6.2e).
84
Figure 6.1: TEM images of the of the freshly synthesized 0.35-NiMoS.
a b
c d
85
Figure 6.2: TEM images of treated NiMo sulfide samples with various Ni/(Ni+Mo)
loadings, a) 0.1, b) 0.2, c) 0.35, d) 0.5, e) 0.65.
Table 6.1:Physical properties of the samples after the treatment.
Sample
Ni/(Ni+Mo)
Specific
surface area
(m2/g)
Pore
volume
(cm3/g)
Average
particle size
(nm)
Average
number of
layers
MoS2 0 474.7 0.86 6.1±2.1 2.2±0.8
0.1-NiMoS 0.1 407.2 0.73 5.9±1.7 1.8±0.8
0.2-NiMoS 0.2 310.5 0.69 6.0±1.7 2.1±0.9
0.35-
NiMoS
0.35 224.5 0.58 5.4±1.7 1.8±0.9
0.5-NiMoS 0.5 138.2 0.41 5.3±1.6 1.7±0.8
0.65-
NiMoS
0.65 195.8 0.78 6.1±2.0 1.7±0.7
The properties of the catalysts after the treating procedure are shown in Table 6.1.
Unpromoted MoS2 showed the highest specific surface area (474.7 m2/g) and pore volume
(0.86 cm3/g). With the addition of the promoter (up to 0.65-NiMoS), the surface area and
pore volume decreased as the amount of promoter in the sample increased. The reduction
e
86
in surface area and pore volume in sulfide catalysts after the addition of promoter was also
reported previously [202, 203].
Based on the isotherm curves (Figure 6.3a), Type I adsorption isotherms are considered for
all of the catalysts and the hysteresis loop between adsorption and desorption isotherms is
associated with mesoporosity of the samples. Isotherms also show that with the amount of
promoter increasing (up to 0.65) the volume of adsorbed nitrogen is reduced which is
consistent with the reduction in their specific surface area and pore volume. Catalysts also
show similar trend in pore size distribution with the narrow distribution centered around 2
nm with a small shoulder around 3 nm (Figure 6.3b).
The XRD patterns of the prepared unsupported NiMoS catalyst series are shown in Figure
6.4. All unsupported Mo-based sulfide catalysts exhibited broad diffraction peaks,
indicating a very poorly crystallized MoS2 structure, particularly when the promoter was
presented. No trace of any Ni sulfide phase was detected in any of the samples except for
the 0.65-NiMoS. Peaks corresponding to hexagonal NiS phase was found in this sample.
In most cases, the ternary Mo–Ni–S phases could not be identified. This could be due to
the fact that there is overlapping of diffraction peaks from MoS2 and Mo–Ni–S phases.
Another possibility could be that the active structures (Mo–Ni–S phase) are possibly nano-
crystallites and very small in size which cannot be characterized by XRD [204].
87
Figure 6.3:Isotherm (a) and pore size distribution (b) of unsupported MoS2 and
NiMoS catalysts with different amount of Ni loading.
a
b
88
Figure 6.4: XRD patterns of unsupported MoS2 and NiMoS catalysts with various Ni
loadings.
The reaction pathways for DBT are shown Figure 5.4 with a detailed description. BP is the
product of DDS pathway and CHB and BCH are the main products of HYD route. CPMB,
CHCPM, and DCPE are also formed as the products of isomerization reactions.
89
Table 6.2: Effect of Ni/(Mo + Ni) mole ratio on HDS of DBT over unsupported NiMoS
catalysts at 3.4 MPa and 320 0C for 60 min.
catalyst MoS2 NiMoS
Ni/(Ni+Mo) 0 0 0.1 0.2 0.35 0.5 0.65
DBT Conversion (%) 63.0a 23.7 47.3 52.4 58.1 66.3 30.8
Selectivity (%)
BP 11.1 29.1 26.3 29.1 32.9 34.4 41.3
CHB 47.1 34.9 44.9 49.3 50.8 52 39.5
BCH 5.1 1.5 2.8 2.6 2.1 1.8 0.2
CPMB 6.1 2.9 5.8 5.4 5.1 4.8 3.4
DCPE 3.8 0 1.3 0.9 0.6 0.4 0
CHCPM 13.6 1.7 4.2 3.6 2.9 2.6 0.3
HHDBT 3.2 9 4.8 3.2 1.7 1.3 5.4
THDBT 10.0 20.9 9.9 5.9 3.9 2.7 9.9
HYD/DDS 8.1 2.4 2.8 2.4 2.0 1.9 1.4
a) After 2 hours of reaction.
Table 6.2 shows the activity and selectivity of the products for the unsupported MoS2 and
NiMoS catalysts with different Ni/(Ni+Mo) ratios in HDS of DBT. The same ratio of
catalyst to model oil was used in all of the experiments. BP is the only product through the
DDS pathway, so to find the HYD to DDS ratio for the catalysts the sum of selectivity of
other products was divided by the selectivity of BP. Results show that the with the addition
of Ni to the catalyst, the DBT conversion was increased from 23.7% for MoS2 to 47.3%
for 0.1-NiMoS catalyst. The activity of the catalysts showed an increasing trend with the
increasing amount of Ni loading up to Ni/(Ni+Mo) ratio of 0.5. However, the conversion
decreased with further increase in Ni loading. The results show a great synergic effect with
the addition of Ni to Mo in HDS of DBT. All of the catalysts show dominant HYD reaction
pathway with the highest HYD/DDS ratio of 2.7 found for 0.1-NiMoS. At higher
90
conversions (after 2 h of reaction), the HYD/DDS ratio over MoS2 was found to be 11.1
when this ratio over 0.5-NiMoS (the same conversion) was 2.0.
The significant increase in the catalytic activity of MoS2 with the addition of Ni promoter
has been reported by many researchers. It was found that with increasing the concentration
of the promoter, the activity of the catalyst increased reaching a maximum. Different
maximum points can be found in the literature for the supported and unsupported catalysts.
In case of the supported catalysts, the optimum Me/(Mo+Me) atomic ratios (Me=Ni or Co)
are generally in the range of 0.2 to 0.4 for Ni(Co)MoS catalysts [21, 205]. For unsupported
catalysts, this ratio is reported to be in the range of 0.3-0.5 for CoMoS [203, 206] and 0.4-
0.5 for NiMoS catalysts [207, 208]. The optimum Ni/(Ni+Mo) ratio found here was 0.5
which is in agreement with the previously reported results for the unsupported NiMoS
catalysts.
Table 6.3: HDS of DBT over MoS2-M and NiMoS-M catalysts with various
Ni/(Ni+Mo) ratios at 3.4 MPa and 320 0C for 60 min.
catalyst MoS2-M NiMoS-M
Ni/(Ni+Mo) 0 0.1 0.2 0.35 0.5 0.65
DBT Conversion (%) 20.2 34.1 27.8 21.6 15.9 24.7
Selectivity (%)
BP 88.7 81.0 85.1 85.5 92.9 85.4
CHB 8.2 14.3 11.5 10.3 3.3 7.6
BCH 0 0 0 0 0 0
CPMB 0 1.6 0.8 1.1 0 1.2
DCPE 0 0 0 0 0 0
CHCPM 0 0 0 0 0 0
HHDBT 0 0.9 0.9 0 0 1.8
THDBT 3.1 2.2 1.7 3.1 3.8 4.0
HYD/DDS 0.1 0.2 0.2 0.2 0.1 0.2
91
Table 6.4: HDS of DBT over MoS2-I and NiMoS-I catalysts with various Ni/(Ni+Mo)
ratios at 3.4 MPa and 320 0C for 60 min.
catalyst MoS2-I NiMoS-I
Ni/(Ni+Mo) 0 0.1 0.2 0.35 0.5 0.65
DBT Conversion (%) 24.5 49.9 54.1 59.6 67.7 31.7
Selectivity (%)
BP 34.8 31.7 32.3 33.9 34.6 42.5
CHB 37.7 54.3 53.8 53.1 52.3 40.9
BCH 0.7 1.6 1.7 1.8 1.8 0
CPMB 2.5 4.1 4.4 4.8 4.7 2.8
DCPE 0 0.2 0.2 0.4 0.4 0
CHCPM 0.7 1.8 2.1 2.4 2.4 0
HHDBT 7.2 2.4 2 1 1.2 4.8
THDBT 16.4 3.9 3.5 2.6 2.6 9
HYD/DDS 1.87 2.2 2.1 1.9 1.9 1.4
The activity and selectivity of the products for MoS2 and NiMoS catalysts supported on
magnetite particles is presented in Table 6.3. The DBT conversion over MoS2-M was found
to be 20.2%. The conversion increased with the addition of the Ni promoter with 0.1-
NiMoS showing the highest activity (44.1%). However, the trend in DBT conversion was
opposite of that for the unsupported catalysts, i.e. the activity of the catalysts was decreased
with the increase in Ni/(Ni+Mo) ratio up to 0.65 were the activity was increased to 24.7%.
These catalysts highly favored the DDS pathway with HYD/DDS ratios between 0.1 and
0.2. DBT conversions over the unsupported MoS2 and NiMoS catalysts after the addition
of iron were found to be slightly increased with the same increasing trend (Table 6.4).
Selectivity for BP and CHB was marginally increased compared to the unsupported
catalysts while the selectivity of the isomerized products was decreased. 0.65-NiMoS
92
sample does not follow the same trend as the other NiMoS catalysts in the above tables due
to the presence of additional NiS in its structure (evidenced by TEM and XRD) which
affects the activity and selectivity of the catalyst.
6.4 Discussion
It is now well accepted that the catalytically active sites in the unpromoted and Ni (Co)-
promoted MoS2 are the exposed Mo (or Ni, CO) cations or coordinatively unsaturated sites
(CUS) that are located at the edges and corners of the catalyst slabs [209, 210]. The
hydrogenation and sulfur removal reactions are proposed to take place on varying number
of adjacent CUS and the hydrogen atoms in close proximity of these sites [41, 211]. The
basal planes of the sulfide catalysts are considered to be mostly inactive for
hydrodesulfurization reactions. The structure of MoS2 has also been found as a determining
factor in HDS of sulfur-containing molecules. This relation between the structure of the
catalyst and its activity was called rim-edge model. It is proposed in this model that rim
sites (top and bottom layers of the MoS2 slabs) are active for both hydrogenation and C-S
bond cleavage but edge sites (outer layers between top and bottom layers) are only active
in hydrogenation [80]. Some groups have also considered the corner sites as hydrogenation
sites where there is a higher probability for the formation of adjacent vacancies [212, 213].
For Ni-promoted MoS2 catalysts, the increase in intrinsic activity of the catalyst can be
mainly correlated to the formation of Ni(Co)-Mo-S phase which is significantly more
reactive than the pure MoS2 phase [21, 157, 214]. In Ni-Mo-S phase, Ni atoms replace Mo
cations at the perimeter of the MoS2 crystal slabs. To explain the synergy effect between
Ni(Co) and Mo in Ni(Co)-Mo-S phase, it was proposed that electron donation from the
93
promoter atom to Mo weakens the Mo-S bond strength to an optimum range for HDS
activity. This electron donation also increases the Ni(CO)-S bond strength and therefore a
sulfur atom at bridging position between the promoter and Mo will have intermediate bond
strength [4, 215, 216]. Recently, the introduction of the so-called brim sites through STM
imaging has improved the insights into the nature of active sites in Mo-based sulfide
catalysts. Atom-resolved images of unpromoted and Ni(Co)-promoted MoS2 showed that
fully saturated sites near the edges of the catalysts have metallic characteristics and can be
active in hydrogenation reactions [31, 32, 217]. A critical difference between Ni and Co
promoters is their preferential location at the edges. While Co atoms are more stable at the
sulfur edge, Ni atoms can be present at both (Mo and S) edges [200, 218].
Direct desulfurization of sulfur-containing molecules proceeds through the sulfur atom of
the molecule in σ adsorption mode. In contrast, hydrogenation proceeds through flat
adsorption of the molecule by π-electron donation from the aromatic ring to the active site
[199, 219]. It is shown that adsorption over brim sites which are considered active in
hydrogenation is also the same (flat adsorption) [31]. Previous findings on the supported
NiMoS catalysts shows that the HYD/DDS ratio for these catalysts is generally below 1
[220-224]. However, this ratio was found to be over 1 for all of the NiMoS catalysts in this
study. This could be due to the special geometry of the supported catalysts where flat
adsorption of bulky molecules such as DBT over the edges of the catalyst for
hydrogenation can be hindered. Figure 6.5 illustrates this geometrical limitation over
supported catalysts. For the flat adsorption of DBT on the edges of a supported catalyst,
the distance between the CUS and the support should be at least 95 nm. however, the
distance between an exposed Mo or Ni in the first slab of NiMoS catalyst to the support is
94
less than 61 nm (the distance between two Mo atoms in adjacent layers of a stacked MoS2).
Hence, unlike unsupported NiMoS catalyst, the flat adsorption of DBT will be hindered at
the bottom slab layer of a supported NiMoS catalyst which results in its lower
hydrogenation ability compared to an unsupported counterpart. The σ adsorption mode
(parallel to the support), however, is not affected in the presence of the support and
therefore the HYD/DDS ratio of supported catalysts is higher for unsupported NiMoS
catalysts. As for brim sites which are active in hydrogenation reactions, the adsorption of
the catalyst on the supports removes accessibility of the molecules to one side of the basal
plane of catalyst and therefore reducing the possibility of molecule adsorption on the brim
sites.
Figure 6.5: Accessibility of DBT molecule to supported and unsupported NiMoS
catalyst through flat adsorption to the edges.
The increase in the concentration of hydrogenated products and the reduction in selectivity
of the intermediates (THDBT and HHDBT) over NiMoS compared to MoS2 shows that
95
both hydrogenation and desulfurization reactions were enhanced over the promoted
catalyst. The increase in BP selectivity with the increase in the Ni loading over unsupported
NiMoS catalysts as evidenced in Table 1 may have two reasons. First, with higher Ni
loading, more CUS sites are formed due to the reduction in Me-S bond strength which also
aids in reduction of poisoning effect of H2S on the catalyst. Second, rate of desulfurization
is higher over Ni-Mo-S phases compared to pure MoS2 meaning that an adsorbed molecule
will undergo desulfurization in less time. In other words, the occupancy time of the active
sites by reactant molecules will be reduced. Therefore, more of the DBT molecules can
convert to BP over a Ni-promoted CUS in the same period of time. For Me-S bonding
energies stronger than the optimal value, one can expect the S-C bond cleavage to be the
rate limiting step in HDS reaction.
As we showed in the previous chapter, the main hydrogen provider for the formation of
CPMB from CHB is dissociated hydrogen while H2S plays only a minor role. The same
conclusion can also be drawn here based on the comparable selectivities for CPMB over
MoS2 and NiMoS catalysts with different Ni loading. The significant difference in the
selectivity of CHCPM and DCPE as isomerised products between MoS2 (after 2h of
reaction and conversion levels close to the promoted catalyst) and NiMoS catalyst and also
the decrease in the amount of isomerised products with the increase in the Ni/(Ni+Mo)
ratio can be attributed the difference in H2S adsorption energies over metallic edges. Lower
acidic functionality of NiMoS catalysts is due to the lower adsorption energy of H2S and
also weaker -SH bond strength over these catalysts as evidenced by different studies (240
j/mol over MoS2 vs -170 j/mol over NiMoS) [225-227]. Therefore, in reaction conditions,
H2S can be more easily desorbed from the surface of the Ni-promoted catalysts, which can
96
be translated as lower concentration of Bronsted acid sites that are formed by H2S over
NiMoS. Lower -SH adsorption energy over NiMoS also results in more facile
recombinative desorption from the surface of catalyst which can also be a contributing
factor to lower acidity of the Ni-promoted catalysts. Among the NiMoS catalysts, higher
Ni loading means that more of Mo atoms at the edges are replaced by the Ni. Higher Ni
dispersion at the edges will continuously remove the sites that have the ability to be
converted to Bronsted acid sites. This is also evident by comparing unsupported NiMoS
with NiMoS-I catalysts. However, the effect of 5-fold increase in Ni loading from 0.1 to
0.5 on acidity of the catalysts is much less noticeable than the acidity changes from MoS2
to NiMoS. This is very interesting as it shows that the promotional electronic effect of Ni
is not localized. If this promotional effect was localized, one would expect to see a 5-fold
decrease in acidity from 0.1-NiMoS to 0.5-NiMoS catalyst.
(Ni)MoS2-M catalysts show high affinity toward DDS pathway due to the ability of
magnetite particles to adsorb loosely bound hydrogens. The DBT conversion was increased
by nearly 14% after the addition of Ni to the MoS2-M catalyst. This increase is mainly
through formation of extra BP as (Ni)MoS2-M catalysts favor the DDS pathway. This
increase in the activity can be due to the change in the structure of the catalyst. It is shown
that the addition of Ni promoter and formation of Ni-Mo-S phase can shift the structure of
MoS2 from truncated shape with mostly Mo edge to more hexagonal shapes with more of
accessible sulfur edge [27, 228]. This change in the structure exposes more sulfur edge of
the catalyst that are considered to be mainly active in desulfurization and therefore results
in higher production of BP. For DBT conversion over MoS2 to be placed between NiMoS
catalysts (lower than 0.1,0.2-NiMoS and higher than 0.5-NiMoS), one can postulate that
97
Ni decoration of edges probably starts from the Mo edge and moves forward toward sulfur
edge as the Ni loading increases. The conversion of DBT over NiMoS-M catalysts with
different Ni/(Ni+Mo) ratios showed a decreasing trend in contrast to what was observed
for unsupported NiMoS catalysts. This could be attributed to the decrease in the hydrogen
bonding energy at the edges of the MoS2 catalyst after the addition of Ni promoter as
evidenced by DFT studies [34, 224]. By increasing the Ni loading, more of Mo atoms are
substituted at the edges of the catalyst and more hydrogen with lower bonding energy to
the surface are formed. These loosely bound hydrogen atoms can be adsorbed by the
magnetite particles and therefore reduce the activity of the catalyst.
6.5 Conclusion
A series of catalysts were synthesised through hydrothermal method in this study to
investigate the effect of H2 and H2S over Ni-promoted MoS2 catalyst. Different Ni/(Ni/Mo)
ratios of the catalyst were prepared and compared here. In comparison to the unsupported
catalysts, the catalysts containing magnetite particles showed a significant shift in
selectivity toward DDS pathway. The addition of Ni to the catalyst was found to enhance
both hydrogenation and desulfurization. The increase in BP selectivity over Ni-promoted
catalysts and its increase with increasing Ni loading was assigned to the increase in the
concentration of CUS sites over the edges of catalysts, lower H2S poisoning effects, and
the enhancement of the S-C bond cleavage through synergic effect of Ni-Mo-S phase.
Significant difference between the acidity of unpromoted and Ni-promoted MoS2 was
observed which was the result of lower H2S bonding energy over the promoted catalyst.
From the small difference in the acidity of the NiMoS catalysts, it was concluded that the
98
promotional electronic effect of Ni atoms is not confined to the adjacent atoms. The
increase in the activity of the MoS2-M after the addition of Ni was suggested to be due to
the structural change in the catalyst to form more hexagonal shape and therefore exposing
more of sulfur edge. Based on the comparison between the DBT conversion for MoS2 and
NiMoS catalysts, it was suggested that the Mo substitution with Ni atoms probably starts
from the Mo edge and progress to the S edge with the increase in Ni loading. The difference
between the HYD/DDS ratio over supported and unsupported NiMoS catalysts was
ascribed to the availability of the edge sites for flat adsorption of DBT and reduction of the
available brim sites.
99
Chapter 7: Sulfided catalysts supported on magnetic greigite: Recyclable
catalysts for the hydrodesulfurization of heavy crude oil
7.1 Introduction
In chapter 4, we succesfully synthesised a magnetically recyclable MoS2 nano-composite
with silica covered magnetite particles as the core and MoS2 catalyst as the top layer
designed for this purpose. The synthesised composite could be separated via a magnetic
field and recycled back to the reactor. The catalyst showed excellent ability in removing
sulfur mainly through direct desulfurization patway (DDS). We showed that synergy effect
between the magenetite core and MoS2 was the reason behind this preferential reaction
pathway. However, based on the feed type, the refineries may need to hydorgenate the feed
to some extent. Therefore, a different core material with similar magnetic properties was
considered to increase the hydrogenation ability of the catalyst. Among different transition
metal calcogenides, Greigite (Fe3S4) has recently found applications in electrochemistry,
biomedicine, and water treatment [229-232]. Because of its magnetic properties and the
presence of sulfur in its structure which can be better adopted with the sulfided catalysts
used in the hydrotreating process, greigite was chosen as the substitute for magetite as the
core material. Here, we report a novel and facile synthesis method to prepare recyclable
nanocomposites in which greigite is used as the core material and Co or Ni promoted MoS2
nanocomposites are synthesized through a hydrothermal reaction. The resulting catalysts
are characterized and tested for the HDS of dibenzothiophene (DBT) and dimethyl
dibenzothiophene (DMDBT) as model sulfur-containing compounds.
100
7.2 Results and Discussion
To prepare the composite samples in this work, Fe3S4 particles were used as the core
material and a blanket-type layer of catalyst was formed around the particles with
assistance from CTAC as a surfactant. To prepare the greigite, magnetite particles were
used as tempelates and were converted to greigite through the procedure described in the
experimental section. During the synthesis procedure, S2- released from the decomposition
of thiourea reacts with the Fe3O4 nanoparticles on the surface of the spheres to produce
Fe3S4 nanoparticles. The XRD pattern of Fe3O4 particles and the as-prepared Fe3S4 product
is shown in Figure 7.1. All the peaks in the XRD pattern can be indexed to the cubic phase
of Fe3S4 and no characteristic peaks from impurities were detected [233-235].
Figure 7.1: XRD pattern of Fe3O4 and synthesized Fe3S4.
101
Table 7.1: Properties of the fresh samples.
Sample
Sulfur
content
(± 0.2 wt
%)
Carbon
content
(± 0.2
wt %)
Me/(M
o +
Me)a
Specific
surface
area
(m2/g)
(Co/Ni)M
oS/Fe3S4
ratio
Average
particle
size (nm)
Averag
e
number
of
layers
Fe3S4 - - - 62.1 - - -
Mo/G 29.2 18.4 0 17.8 0.39 3.5±1.3 1.2±0.3
CoMo/G 22.6 15.8 0.25 19.4 0.45 3.6±1.8 1.3±0.5
NiMo/G 19.8 16.9 0.25 21.5 0.45 3.9±1.8 1.5±0.2
a Me/(Mo + Me) molar ratio based on EDX results (Me = Co or Ni)
Three different samples were prepared and compared in this study. These are MoS2/Fe3S4
(coded as Mo/G), CoMoS/Fe3S4 (coded as CoMo/G), and NiMoS/Fe3S4 (coded as
NiMo/G). Table 7.1 shows the properties of the freshly synthesised catalysts. The carbon
content of the samples can be attributed to the presence of the CTAC surfactant in their
structure. The measured BET surface areas were 17.8, 19.4, and 21.5 m2/g for fresh Mo/G,
CoMo/G, and NiMo/G, respectively. The lower surface area of the samples compared to
the Fe3S4 particles is also due to the presence of CTAC in their structure. This is confirmed
by the increase of the BET surface area of Mo/G from to 216.5 m2/g after the activity test.
CoMo/G and NiMo/G also showed nearly 4 fold increase in their surface areas.
102
a b
c d
103
Figure 7.2: TEM images of the Mo/G (a,b), CoMo/G (c,d), and NiMo/G samples
(e,f).
We had shown before the importance of using a suitable surfactant for sucessful synthesis
of a recyclable composite. Here, the CTAC plays two important role. First, helping the
catalyst to anchor onto the surface of the greigite core and second, an scaffold effect during
synthesis of the catalyst through the hydrothermal process. These effects can be observed
from the TEM images of the catalysts. Figure 7.2 shows the TEM images of the fresh and
HDS treated MoS2/G (a,b), CoMo/G (c,d), and NiMo/G (e,f). From the images, the
structural difference between the synthesised catalysts before and after they are treated at
HDS conditions can be observed.
As the images show, parts of the small crystals of the fresh catalyst join together to form
longer multilayered slabs after the treatment. The length of the slabs approximately
increased by 46% for Mo/G and the average number of layers increased from 1.2 to 2.3.
Similar increase in the length and thickness of the crystals was also observed for other
e f
104
samples. After the treating procedure, the catalysts were found to be structurally stable
during the HDS reactions.
105
Figure 7.3: Isotherms (a) and pore size distribution (b) of Mo/G, CoMo/G, and
NiMo/G.
a
b
106
From the isotherm curves (Figure 7.3 a), all catalysts indicate type IV adsorption isotherms
and exhibit the hysteresis loop between adsorption and desorption isotherms in the range
of (0.4 to 1.0) is indicative of mesoporosity. CoMo/G and NiMo/G catalysts show similar
pore size distribution with bimodal mesopore peaks at around 2 nm and over 4 nm (Figure
7.3 b). Mo/G also shows bimodal mesopore peaks but at about 2 and 3 nm. A small shoulder
is also seen at around 6 nm for Mo/G. These pores are considered to be situated within the
primary particles. This type of pores can easily provide sufficient surface for catalytic
application.
Figure 7.4 shows the magnetization curves measured at 300 K for the Fe3S4 and Mo/G
samples. The curves present no hysteresis loop, suggesting that Fe3S4 particles have
paramagnetic behavior. The saturation magnetization values for the Fe3S4 and MoS2/Fe3S4
were 25 and 22 emu/g, respectively. The small difference between the saturation values
suggests that the addition of the catalyst layer has a negligible effect on the magnetic
properties of the sample. The possibility for magnetic separation of samples was tested by
exposing the catalysts to an external magnetic field. Mo/G dispersed in model oil after 2 h
of HDS reaction is shown in Figure 7.5a and it is separated afterwards with an external
magnet (Figure 7.5b). The catalyst can be separated easily from the reaction medium within
a few seconds. All of the samples showed the same magnetic effect. Therefore, this method
provides an easy and efficient way to separate catalysts from the product stream by an
external magnetic field.
107
Figure 7.4: Magnetization curves for Fe3S4 and MoS2/Fe3S4 measured at 300 K.
Figure 7.5: Mo/G composite after the HDS reaction a) dispersed inside the model oil
and b) separated from the model oil after seconds of exposure to an external
magnet.
The reaction pathways for the HDS of DBT is shown in Figure 5.4. BP is produced through
DDS pathway and CHB is the main product of the HYD route. THDBT and HHDBT are
a b
108
the intermediates in the HYD pathway. CPMB, CHCPM, and DCPE are the products of
the isomerization reactions.
The DBT conversion and product distribution of the prepared catalysts are shown in Figure
7.6. The mass ratio of (Co or Ni)MoS-to-model oil was maintained the same for all the
catalysts. The selectivity of the isomerized products is shown in Figure 7.7. The DBT
conversions for Mo/G, CoMo/G, and NiMo/G are 54, 74, and 77 percent, respectively. As
expected, promoted catalysts show better activity that Mo/G in HDS of DBT with nearly
28% increase in the activity. Considering the uncertainty levels, the activity of CoMo/G
and NiMo/G for DBT conversion is the same. The selectivity of the products show that
Mo/G has higher tendency toward the HYD pathway compared to the promoted catalysts.
The DDS/HYD ratio (calculated by dividing the selectivity of BP by the sum of the
selectivities of other products) for Mo/G was found to be 0.31. Both CoMo/G and NiMo/G
catalysts show the same selectivity for DDS and HYD reaction pathways with nearly 50%
selectivity for each route. The DDS/HYD ratio was 1.19 and 1.11 for the Ni-promoted and
Co-promoted catalysts, respectively. The results suggest that promotional effect of Co and
Ni is mainly through increasing the activity of the catalyst via DDS pathway which shows
to be more selectively enhanced than the HYD route. This is in accordance with the results
reported in the literature describing the same promotional effect for Co and Ni in HDS of
DBT [41, 89].
The active sites for the unpromoted MoS2 are proposed to be the coordinatively unsaturated
sites or exposed Mo ions with sulfur vacancies at the edges of the catalyst. These sites are
considered active for hydrogenation and hydrogenolysis. Recently, unsaturated sites with
metalic charactars at Mo edges of the (Co/Ni)MoS2 catalysts (called brim sites) were also
109
found to be active for hydrogenation reactions. For Co or Ni-promoted catalysts, the Co(or
Ni)–Mo–S phase is considered to be the active phase [27]. In this model, small MoS2
crystals with the promoter atoms located at the edges of the MoS2 layers in the same plane
as Mo are considered to be the building blocks for Co(or Ni)–Mo–S structures. The
promotional effect with the addition of Co or Ni is believed to be due to the electron
donation from the promoting atom to Mo atom, weakening the Mo-S bond strength to an
optimum range for HDS activity [4, 29].
The presence of isomerized products in the samples suggests that acid sites are present on
the surface of the prepared catalysts. The results show that in HDS of DBT, the selectivity
of the isomerized products over Mo/G is higher than the promoted catalysts. The creation
of the acid sites over sulfided catalysts can be through the dissociation of hydrogen over
the sulfurs at the edges of the catalyst. These sites can also be created through dissociation
of hydrogen sulfide at the coordinatively unsaturated (CUS) sites. Considering the
significant hydrogenation ability of the promoted catalysts, the higher acidity of Mo/G can
be attribuated to the dissociation of hydrogen sulfide.The dissociation of H2S can increase
the acidity of the catalyst by converting the CUS sites to –SH. These –SH groups are
considered to be active in hydrogenation and isomerization reactions over the catalysts.
The acidic nature of these groups, however, depends on H+ electron affinity and the bond
strength of metal–sulfide [236]. If the metal-sulfur bond is strong (as in MoS2), the –SH
groups act as Bronsted acid acites while in the presence of a weak metal-sulfur bond they
act as nucleophilic centers [193, 237, 238]. As mentioned before, the promotion effect of
Co or Ni is through reducing the metal-sulfur bond strength. The stronger bonds in Mo/G
implies that the –SH sites created by H2S dissociation are acidic over this catalyst, while
110
the same sites are nucleophilic centers over the promoted catalysts. Thus the Mo/G shows
better acidity than the promoted catalysts in case of DBT.
Figure 7.6: DBT conversion and product selectivity for the synthesized catalysts
after 2h of reaction.
54
6.83.4
49.6
4.0
23.8
74
1.4 0.8
35.5
54.4
77
1.3 0.8
35.3
52.7
0
10
20
30
40
50
60
70
80
90
DBTconversion
THDBT HHDBT CHB BCH BP
Sele
ctiv
ity
or
Co
nve
rsio
n (
%)
Mo/G CoMo/G NiMo/G
111
Figure 7.7: The selectivity of the isomerized products over the synthesized catalysts
after 2 hr of HDS of DBT.
Figure 7.8: Possible reaction pathways in HDS of DMDBT.
Figure 7.8 shows the possible reaction pathway for the HDS of DMDBT. Similar to DBT,
two reaction pathways are considered for the HDS of DMDBT. Dimethyl biphenyl
5.3
1.4
5.7
3.3
1.0
3.6
4.6
1.2
4.1
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
CPMB DCPE CHCPM
Sele
ctiv
ity
(%)
Mo/G CoMo/G NiMo/G
112
(DMBP) is formed through DDS route and the THDMDBT and HHDMDBT are formed
as intermidiates via HYD route. 1-methyl-3-(3-methylcyclohex-1-en-1-yl)benzene
(MMCHB) was detected in this study which is an intermidiate in the formation of MCHT.
This suggests that the same compound withouth the methyl groups ((cyclohex-1-en-1-
yl)benzene) is an intermediate in formation of CHB from THDBT. (cyclohex-1-en-1-
yl)benzene is usually not detected in experiments due to its fast conversion rate, however,
the presence of the methyl substituted compound in HDS of DMDBT strongly suggests
that (cyclohex-1-en-1-yl)benzene plays a role as an intermediate in the formation of CHB.
HHDMDBT can undergo C-S bond cleavage to form MCHT and/or DMBCH after
hydrogenation. MCHT and DMBCH can be isomerized to 1-(cyclohexylmethyl)-4-
methylbenzene (CHMMB) and 1-(cyclohexylmethyl)-3-methylcyclohexane (CHMMCH),
respectively. CHMMB can be further isomerized to 1‐(2‐cyclopentylethyl)‐4‐
methylbenzene (CPEMB) which can be converted to form (3-cyclopentylpropyl)benzene
(CPPB).
Figure 7.9 shows the DMDBT conversion and the selectivity of the main products over the
catalysts. The DMDBT conversion was found highest for NiMo/G with 81% conversion
compared to CoMo/G with 68% and Mo/G with 41% conversion. The reulsts show the
accumulation of THDMDBT for all of the catalysts suggesting the difficulty of the catalysts
in removing the sulfur from the molecules. The DDS/HYD ratio for the catalysts were
calculated as 0.5, 0.25, and 0.19 for Mo/G, CoMo/G, and NiMo/G, respectively. These
ratios show that the HYD pathway is favored for all of the prepared catalysts and Mo/G
has the highest affinity toward the DDS pathway with 34% selectivity for DMBP. The
selectivity of the isomerized products is shown in Figure 7.10. CHMMB is the main isomer
113
compound with 13.0, 17.8, and 23.7% selectivity for the Mo/G, CoMo/G and NiMo/G,
respectively.
Studies with model compounds have shown that DDS pathway is the preferential route in
HDS of DBT with Co and Ni-promoted catalysts [43, 239]. The addition of alkyl
substitutes in 4, 6 positions to the DBT molecule affects the HDS reaction in two different
ways. First, the HDS reactivity of the molecule will be reduced and second, the preferential
reaction pathway will be shifted toward HYD, making it the dominant route [82, 240].
Hydrogenation of DBT and DMDBT is mainly through π-bonding with the catalyst while
the hydrogenolysis is through σ-adsorption via the sulfur atom of the molecule [43]. The
difficulty in converting the alkyl-substituted DBT compounds are known to be related to
the steric hindrance of the alkyl groups positioned close to the sulfur atom and therefore
preventing the interaction of these molecules with the active sites of the catalyst through
σ-adsorption. The partial saturation of one of the rings by π-bonding which is not sterically
hindered in these compounds changes the spatial configuration of them, thus reducing the
hindrance effect and rendering them more flexible and accessible to the reaction sites of
the catalyst [241].
114
Figure 7.9: DMDBT conversion and product selectivity for the synthesized catalysts
after 2h of reaction.
41
9.2
4.12.0
25.6
2.6
34.0
68
10.3
3.5 2.9
25.2
5.3
19.9
79
7.6
2.7 3.5
21.3
6.7
16.2
0
10
20
30
40
50
60
70
80
90
DMDBTConversion
THDMDBT HHDMDBT MMCHB MCHT DMBCH DMBP
Sele
ctiv
ity
or
con
vers
ion
(%
)
Mo/G CoMo/G NiMo/G
115
Figure 7.10: The selectivity of the isomerized products over the synthesized catalysts
after 2 hr of HDS of DMDBT.
It is suggested that the change in the preferential reaction pathway by the addition of alkyl-
substitutes to DBT is mainly due to the sever inhibition on the DDS pathway and the
presence of these alkyl groups hardly affect the HYD pathway. The difference between the
reactivity of DBT and DMDBT was related to the selective promoting effect on DDS
pathway for DBT [41]. While this agrees with our results for the promoted catalysts, Mo/G
shows an opposite trend. The selectivity of the DMDBT through DDS pathway was found
to be higher over this catalyst than DBT. It seems that active sites for hydrogenation of
DBT and DMDBT are different in this catalyst, allowing better hydrogenation of DBT.
This can be due to a special synergic effect between MoS2 and the greigite core. The
addition of the promoter to the Mo enhances the rate of C-S bond cleavage, which is the
rate-limiting step in desulfurization of the DBT and DMDBT over the Mo/G. This results
13.0
3.1 2.73.7
17.8
5.54.3
5.3
23.7
6.95.4 6.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
CHMMB CPEMB CPPB CHMMCH
Sele
ctiv
ity
(%)
Mo/G CoMo/G NiMo/G
116
in higher DDS activity over the promoted catalysts in case of DBT and for DMDBT, it can
also increase the rate of conversion of the hydrogenated intermediates which are more
accessible to the active sites. Considering the overall activity of the catalysts for DBT (1
wt%) and DMDBT (0.4%), it can be concluded that although the addition of promoter
increases the activity of the catalyst, their activity is still lower for DMDBT. This can be
due to the hindrance effect of the DMDBT and its hydrogenated intermediates compared
to DBT due to the presence of the alky-substitutes which makes the final C-S bond cleavage
step more difficult in case of DMDBT.
The stability of the MoS2/Fe3S4 composite was investigated by reusing the catalyst for
several cycles of DBT HDS. After each step, the catalyst was collected, washed and then
returned to the reactor to perform another round of reaction. The results of each cycle are
shown in Table 7.2. As a comparison, similar experiment was performed using the
composite containing magnetite cores (MoS2/Fe3O4). Table 7.3 shows the recycling results
for this composite.
The results for MoS2/Fe3S4 shows no significant change after 7 cycles of HDS test. When
compared to the composites prepared with a magnetite core, composites containing a
greigite core show higher hydrogenation selectivity in HDS of DBT for the first 3 cycles.
The difference most likely originates from the difference in core material which can cause
different synergic effects with the sulfide catalysts. For MoS2/Fe3O4, the selectivities
indicate a shift toward HYD reactions after the 3rd cycle. At cycle 4, the DBT conversion
was increased and the selectivity of hydrogenated compounds (CHB and BCH) and
isomerized products was increased. On the other hand, the selectivity of BP was decreased
by 25%. At cycle 5, similar increasing trend for DBT conversion and selectivity of the
117
hydrogenated products and isomers was observed and the values only showed a slight
change after the 5th cycle. After cycle 3, the catalyst had lost its magnetic properties and
could not be collected by the hand-held magnet (the catalyst was separated through
centrifugation).
Table 7.2: DBT conversion and selectivity for 7 cycles in HDS of DBT over
MoS2/Fe3S4 catalyst.
Cycle
1
Cycle
2
Cycle
3
Cycle
4
Cycle
5
Cycle
6
Cycle
7
DBT
conversion 54 49 56 54 49 50 51
THDBT 6.8 5.4 6.1 4.4 3.9 3.5 3.4
HHDBT 3.4 2.5 2.9 2.0 2.3 1.9 2.0
CHB 49.6 48.9 48.2 48.5 49.3 49.9 49.0
BCH 4.0 4.1 4.5 4.5 4.7 5.0 5.1
BP 23.8 25.7 26.0 26.1 24.6 24.2 24.5
CPMB 5.3 5.8 4.9 6.3 6.7 6.8 7.0
DCPE 1.4 1.6 1.8 1.7 1.8 1.8 1.9
CHCPM 5.7 6.0 5.6 6.5 6.7 6.9 7.1
Table 7.3: DBT conversion and selectivity for 7 cycles in HDS of DBT over
MoS2/Fe3O4 catalyst.
Cycle
1
Cycle
2
Cycle
3
Cycle
4
Cycle
5
Cycle
6
Cycle
7
DBT
conversion
48 50 47 58 65 63 62
THDBT 1.7 1.4 1.6 3.6 2.3 2.0 1.4
HHDBT 0.0 0.0 0.0 1.4 1.4 1.1 1.1
CHB 19.4 17.1 18.3 32.0 48.6 47.2 48.4
BCH 0.0 0.0 0.0 0.3 2.4 3.1 3.2
BP 77.0 79.9 78.3 58.1 34.5 34.7 33.5
CPMB 1.9 1.5 1.8 3.5 7.0 7.5 8.0
DCPE 0.0 0.0 0.0 0.5 0.6 0.7 0.7
118
CHCPM 0.0 0.0 0.0 0.5 3.1 3.6 3.7
Figure 7.11: XRD patterns of treated MoS2/Fe3O4 and after cycles 1, 3, 5, and 7.
To investigate the changes in the crystal structure of the MoS2/Fe3O4 sample, XRD test
was performed (Figure 7.11). After cycles 1 and 3, the additional peaks that are observed
can be assigned to iron showing the reduction of part of magnetite particles. Moreover, the
intensity of the magnetite peaks at the 3rd cycle is reduced. At cycle 5, the magnetite peaks
are replaced by pyrrhotite (Fe7S8) indicating the change in the magnetite particles via
119
adsorption of H2S gas produced as the result of HDS reaction. The iron peak can also be
seen as a shoulder to the pyrrhotite peak at 53° (inserted figure) and as a small peak at 65°.
At the same time, the selectivity of the catalyst is shifted toward HYD indicated by higher
selectivity of CHB and BCH (Table 3). At cycle 7, the intensity of the pyrrhotite peaks was
increased and the catalyst activity and selectivity remained mostly unchanged in favor of
HYD reaction pathway. The results clearly show the higher stability of the greigite cores
compared to their magnetite counterparts during our reaction conditions after 7 cycles of
HDS of DBT.
7.3 Conclusion
We have demonstrated the preparation of unpromoted and promoted MoS2-based
composites with magnetic properties through a facile procedure. High activity toward the
refractory sulfur-containing compounds, high stability, and the magnetic properties
associated with these nano-composites provides vital advantage for them as it allows easy
separation and reuse, which can be very useful in the processing of heavy and extra-heavy
crude oils in slurry reactors. Three catalysts were synthesized in this work, two promoted
catalysts (CoMoS/Fe3S4 and NiMoS/Fe3S4) and one unpromoted catalyst (MoS2/Fe3S4).
Greigite (Fe3S4) was used as the core material considering its magnetic properties and easy
adoptability of this material and was covered with catalyst layers as the active phase with
the aid of cetyltrimethylammonium chloride (CTAC) as surfactant.
NiMo/G showed the highest activity in the HDS of DBT and DMDBT. The activity of
CoMo/G in HDS of DBT and the DDS/HYD ratio for this catalyst was found to be similar
120
to the Ni-promoted catalyst. Mo/G showed higher tendency toward the HYD pathway with
DDS/HYD ratio of 0.31. The selectivity of the isomerized products were found to be higher
for Mo/G, an indication of higher acidity of this catalyst. This was assigned to the
difference in the acidic nature of –SH groups formed through the dissociation of H2S over
the CUS sites for the promoted and unpromoted catalysts. In the presence of weak metal-
sulfur bonds as in promoted catalysts, the –SH groups were considered to work as
nucleophilic centers while over Mo/G with strong metal-sulfur bond, they act as Bronsted
acid acites.
All of the repared catalysts showed HYD as the favored route in HDS of DMDBT with
NiMo/G showing the highest HYD activity (81% conversion) and with DDS/HYD ratio of
0.19. This was followed by CoMo/G and Mo/G in activity toward HDS of DMDBT and
DDS/HYD ratios of 0.25 and 0.5, respectively. The higher affinity of the Mo/G toward the
DDS pathway was related to a possible synergic effect between the MoS2 and the Fe3S4
core. The preference for the HYD pathway in HDS of DMDBT for the prepared catalysts
was correlated to the presence of alkyl groups close to the sulfur atom of DMDBT. The
alkyl groups create a hinderence effect, making the interaction between the molecule and
the active sites of the catalysts more difficult.
121
Chapter 8: Hydrodeoxygenation of Stearic acid with unpromoted and Ni(Co)-
promoted MoS2 supported on greigite
8.1 Introduction
In the previous chapter we successfully designed a series of magnetically recyclable MoS2
catalysts, unpromoted or promoted with Co or Ni, with greigite cores for the desulfurization
of heavy oil in slurry reactors. The higher amount of impurities in the heavy and extra-
heavy oil compared to light oil can cause the catalyst-bed in the conventional packed-bed
reactors to easily clog resulting in premature shut down of the unit. Substituting the
conventional packed-bed reactors with slurry reactors can overcome this problem.
However, for the process to be economically viable, the catalysts used in such reactors
should be recyclable. The presence of impurities in the vegetable oils specifically in used
vegetable oils such as waste cooking oil can also cause the same issue in packed-bed
reactors. Another factor that can intensify this problem is the presence of glycerol in
vegetable oils. Accumulation of such a highly viscous liquid may also add to the clogging
problem of the reactor column [66]. Co-processing of vegetable oils and heavy oil can
therefore become very costly in packed-bed reactors which are prone to clogging. Using
magnetically recyclable catalysts in a slurry reactor can be a cost-effective solution for
such problems.
In this study, the focus is on studying the deoxygenation abilities of these catalysts and to
investigate the activity and selectivity of each catalyst. These findings can pave the path to
use these catalysts in simultaneous processing of heavy crude oil and vegetable oils.
122
8.2 Results and Discussion
The properties of the catalysts after the treating procedure is shown in Table 8.1. The pore
size distributions are shown in Figure 8.1. We had shown previously the importance of the
use of a proper surfactant during the synthesis of the composites. The surfactant is removed
during the treating procedure as evidenced by the carbon content of the samples. Mo/G
showed the highest specific surface area (216.5 m2/g). About 62% reduction in the specific
surface area was observed after the addition of the promoters. The results showed that all
of the catalysts are largely mesoporous with a bimodal pore structure.
Table 8.1: Properties of the treated samples.
Sample Sulfur
content (±
0.2 wt %)
Carbon
content (±
0.2 wt %)
Specific
surface area
(m2/g)
Average
particle
size (nm)
Average
number of
layers
Fe3S4 - - 62.1 - -
Mo/G 41.1 1.5 216.5 6.5±2.0 2.3±1.5
CoMo/G 31.1 1.3 83.9 6.9±2.1 2.4±1.1
NiMo/G 27.2 1.8 82.8 7.2±2.7 2.5±1.0
123
a
c
b
d
e f
124
Figure 8.1: TEM images of fresh (left) and treated (right) Mo/G (a,b), CoMo/G (c,d),
and NiMo/G samples (e,f).
The TEM images of the freshly synthesized samples (left) and after treatment (right) are
shown in Figure 8.1. As the images show, parts of the small crystals of the fresh catalyst
join together to form longer multilayered slabs after the treatment. The length of the slabs
approximately increased by 46% for Mo/G and the average number of layers increased
from 1.2 to 2.3. Similar increase in the length and thickness of the crystals was also
observed for other samples. After the treating procedure, the catalysts were found to be
structurally stable during the HDS reactions.
The XRD patterns of the prepared Fe3S4 core, Mo/G, CoMo/G, and NiMo/G are shown in
Figure 8.2. All Mo-based sulfide catalysts exhibited broad diffraction peaks, indicating a
very poorly crystallized MoS2 structure, particularly when the promoter was presented. No
trace of any Ni or Co sulfide phase was detected in any of the samples. In most cases, the
ternary Mo–Co(Ni)–S phases could not be identified. This could be due to the fact that
there is overlapping of diffraction peaks from MoS2 and Mo–Ni–S phases and the nano-
size structure of this phase.
125
Figure 8.2: XRD patterns of the catalysts.
126
C18 fatty acids are the dominant component of the vegetable oils. So, stearic acid was used
as the model compound to study and compare the prepared catalysts for the removal of
oxygen. Figure 8.3 shows the conversion of stearic acid and the yield of the products at
different contact times over Mo/G. NiMo/G showed the highest conversion of stearic acid
at 20 and 40 min followed by CoMo/G and Mo/G. After 60 min, it was fully converted
over all of the catalysts. The yield of the products from the deoxygenation reaction over
the prepared catalysts is depicted in Figure 8.4. Beside the remaining stearic acid,
octadecanal and octadecanol were also detected as the oxygenated compounds.
Octadecane, octadecene, heptadecane, and heptadecene were the main hydrocarbon
products from the deoxygenation reaction. Up to nearly 40 min and before the complete
conversion of stearic acid, the main product was found to be octadecanol. At 60 min and
after the stearic acid was completely converted, the amount of octadecanol was greatly
decreased while a significant increase was observed in the yield of the hydrocarbon
products. Octadecane and octadecane were always found in higher quantities than
heptadecane and heptadecene at different reaction times. Five different isomers of
octadecene were detected by GC-MS but only 1-octadecene and 2-octadecene were
formally identified as the main isomers. Same observations were made for heptadecene.
Trace amounts of hexadecane and hexadecene (0.1% in yield) was also found after 40 min
of reaction. At 60 min, the yield of these hydrocarbons increased to 0.4%, 0.4%, and 0.3%
for NiMo/G, CoMo/G, and Mo/G respectively.
127
Figure 8.3: Conversion of stearic acid over the prepared catalysts.
0.0
20.0
40.0
60.0
80.0
100.0
120.0
0 10 20 30 40 50 60 70
Co
nve
rsio
n (
%)
Time (min)
NiMo/G CoMo/G Mo/G
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
0 1 0 2 0 3 0 4 0 5 0 6 0 7 0
Yiel
d (
%)
Time (min)
a
128
Figure 8.4: The yield of the products from deoxygenation of stearic acid over a) Mo/G,
b) CoMo/G, and C) NiMo/G at different reaction times.
Octadecanal is an aldehyde intermediate formed via the removal of the hydroxy group from
stearic acid as H2O through a hydrogenolysis reaction mechanism involving dehydration
and hydrogenation reactions. Octadecanol (the C18 alcohol intermediate) was another
oxygenate intermediate that was formed during the deoxygenation reactions and was the
most abundant product up to 40 min of reaction. Although the conversion of stearic acid
reaches 100% at 60 min, the yield of octadecanol is higher over Mo/G compared to the
0.0
10.0
20.0
30.0
40.0
50.0
60.0
0 1 0 2 0 3 0 4 0 5 0 6 0 7 0
Yiel
d (
%)
Time (Min)
b
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
0 10 20 30 40 50 60 70
Yiel
d (
%)
Time (min)
Octadecanol octadecanal
Octadecane Octadecens
C
129
promoted catalysts at this time due to the slower reaction rate over this catalyst. It is mainly
believed that the alcohol intermediate is formed through the hydrogenation of the aldehyde
[242-244]. It is also possible that the aldehyde intermediate remains adsorbed on the
surface of the catalyst after its conversion from the carboxylic acid to directly form the
alcohol intermediate. The low yield of octadecanal and high yield of octadecanol in our
experiments support the second scenario. If octadecanal was desorbed from the catalyst,
its adsorption and further conversion would have been restricted by stearic acid as observed
for octadecanol.
The results show that for all the catalysts, the yields of octadecanal and octadecanol were
increased up to 40 min at high conversion levels of the stearic acid. At 60 min and in the
absence of the stearic acid, the yields of both compounds were reduced significantly. This
clearly points to the inhibiting effect of the carboxylic acid on the transformation of the
aldehyde and the alcohol intermediates. This inhibiting effect is evidently reversible as the
removal of the stearic acid from the reaction medium resulted in higher activity of the
intermediates. This can be explained by the competitive adsorption of stearic acid,
octadecanal, and octadecanol on the active sites of the catalyst in favor of the stearic acid.
This inhibiting effect of carboxylic acids is shown in the conversion of decanal in the
presence of decanoic acid [245] and is also confirmed in a DFT study by Dupont et al.
[246] showing stronger adsorption energy for propanoic acid (-0.65 ev) compared to that
of propanal (-0.38 ev) over MoS2 catalyst. This inhibiting effect was also reported in the
transformation of dibenzothiophene in the presence of the carboxylic acid over
CoMoP/Al2O3 catalyst [247].
130
In addition to the above mentioned products, CO and CH4 were detected as the gas
products. Gas analysis showed CO as the main gas product during the transformation of
the carboxylic acid and the quantities of it were dependent on extent of the reaction (higher
quantities for longer reaction times). CH4 was only detected in trace amounts.
Interestingly, no CO2 was detected as a gas product. Two scenarios can be considered for
the lack of presence of CO2 in the gas phase. First, CO2 was produced during the reactions
denoting that no decarboxylation of stearic acid occurred. Second, CO2 was produced but
was converted to CO through the water-gas shift reaction (reaction 1) or into CH4 by
methanation reaction (reaction 2).
CO2 + H2 ↔ CO + H2O (Reaction 1)
CO2 + 4H2 ↔ CH4 + 2H2O (Reaction 2)
The equilibrium reaction constant for equation 1 at our reaction conditions was found to
be 0.057 indicating that the water-gas shift reaction is thermodynamically not favorable.
Considering the very small amount of CH4 produced, it can be concluded that HDC
pathway for the conversion of stearic acid was mainly through decarbonylation by forming
CO. Therefore, the C17 hydrocarbon products were formed either by decarbonylation of
stearic acid and/or octadecanal. It is noteworthy to mention that formation of heptadecane
through the decarbonylation of the aldehyde is found to be solely via thermal
decomposition and non-catalysed [245].
131
Based on the above discussions, the reaction pathways in deoxygention of stearic acid can
be depicted as scheme 1. The deoxygenation of stearic acid follows two main pathways:
the Hydrodeoxygenation route (HDO) and the decarbonylation route (HDC). Through the
HDO pathway, stearic acid is first hydrogenated and then further dehydrated to form
octadecanal producing water as a by-product. Octadecanal can either transform to
octadecanol by hydrogenation or can directly form heptadecane through decarbonylation.
Octadecanol can go through deoxygenation by an elimination step to form 1-octadecene
and water or through hydrogenation and dehydration to form octadecane. 1-octadecene can
undergo through different isomerisation steps by double bond shifting leading to 2-
octadecene and three other isomers, possibly 4, 5, and 7-octadecene. It is also important to
mention that the hydrogenation of octadecenes was inhibited in the presence of stearic acid.
In DOC reaction pathway, stearic acid is converted to heptadecene through dehydration
and decarbonylation reactions forming H2O and CO as by-products. Heptadecene can
further be hydrogenated leading to the formation of heptadecane. As mentioned before, this
final step is greatly influenced by the presence of carboxylic acid.
132
Scheme 1: Possible reaction pathways in deoxygenation of stearic acid over the
prepared catalysts.
Figure 8.5: Comparison of the C18/C17 ratio.
Figure 8.5 compares the C18/C17 ratio of the prepared catalysts. All the catalysts favored
the HDO pathway over HDC in the course of reactions. The ratio was increased for all the
catalysts as the reaction progressed showing higher affinity of the catalysts toward HDO
pathway. The results also demonstrate the higher C18/C17 ratio over Mo/G compared to
0.0
2.0
4.0
6.0
8.0
10.0
0 10 20 30 40 50 60 70
C1
8/C
17
Time (min)
NiMo/G CoMo/G Mo/G
133
the other catalysts which can be related to the higher HDC activity of the promoted
catalysts. This is in agreement with the findings in previous studies showing higher HDO
rate for the unpromoted catalyst [65].
Figure 8.6 shows the paraffin/olefin ratio for the catalysts at different reaction times. The
paraffin/olefin ratio was found to be the same over Mo/G catalyst at 20 and 40 min after
the reaction (close to 1.2), indicating that octadecane was not produced through the
hydrogenation of octadecene. This ratio was increased after the conversion of stearic acid
to 2.1 suggesting that the presence of carboxylic acid has a strong inhibiting effect on the
hydrogenation of octadecene. The same inhibiting effect was observed in conversion of
octadecanol which proceeds through hydrogenation of the C-O bond. This clearly suggests
that the carboxylic acid adsorbs on the hydrogenation sites and prevents the parallel
hydrogenation reactions to proceed at the same time. CoMo/G and NiMo/G also show the
same trend, however, the paraffin/olefin ratio was found to be higher for these catalysts.
NiMo/G showed the highest ratio between the three catalysts with the ratio of 2.7 at the
end of the reaction. The higher alkane/alkene ratio over the promoted catalysts is in
agreement with previous findings showing the promoting effect of Co and Ni in
hydrogenation of various olefins [248].
134
Figure 8.6: Comparison of the paraffin/olefin ratio.
Schematic 2 shows the reaction mechanism over the unpromoted Mo catalyst. As for all
the catalysts HDO was the dominant reaction pathway, the reaction mechanism can follow
the similar way. It is now well accepted that the active sites over the MoS2 phase of sulfided
catalysts are the sulfur vacancies located at the edges of the catalyst slabs. The reacting
molecules adsorb on these sites to undergo further deoxygenation reactions. The first step
in the proposed reaction mechanism is the heterolytic dissociation of hydrogen forming a
hydride ion bonded to a metallic site (Mo-H) and a proton that would bond to an adjacent
sulfur atom creating a sulfhydryl (-SH) group [195, 249, 250]. In the next step, stearic acid
is adsorbed to a sulfur vacancy site through its carbonyl group. Due to the acidic nature of
the sulfhydryl group, the adsorbed stearic acid could then be protonated by the –SH group
and this proton would be added to the hydroxyl (–OH) group. This lowers the bond strength
between the carbon and oxygen in the hydroxyl group. With this reduction in C-OH bond
strength, stearic acid is then dehydrated releasing H2O as a by-product. Hydride ion is
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0 10 20 30 40 50 60 70
Par
affi
n/O
lefi
n
Time (min)
NiMo/G CoMo/G Mo/G
135
added to the carbon atom in the next step forming an adsorbed aldehyde. This aldehyde
can follow three paths. First, it can undergo decarbonylation via C-C bond cleavage to
produce heptadecane. This step might be via an attack from a basic site (-S-2 adjacent to a
vacant site) on the hydrogen located at the β position of the carbon in carbonyl group [251].
The remained CO is desorbed afterwards as a by-product and the vacant sites are
regenerated. Second, it can follow the same proton-hydride addition steps to form
octadecanol which was found in high yields and as a key intermediate in the products.
Octadecanol is further adsorbed on the catalyst and will undergo C-O bond scission to form
C18 hydrocarbons. Third, it can be desorbed from the catalyst surface as it was detected
among the products during the deoxygenation reaction. Desorbed octadecanal can later
adsorb on the catalyst to follow the first and second reaction paths. The accumulation of
octadecanol during the reaction clearly shows that hydrogenation of octadecanal is not rate
determining and is greatly favored over the decarbonylation route.
Although all the catalysts were found to favor the HDO route, the promoted catalysts
showed higher activity for both HDO and HDC pathways. With the addition of Co and/or
Ni promoter atoms, Mo atoms at the edges of the catalyst will be replaced with the promoter
atoms forming Co(Ni)-Mo-S phases that are known for their synergy effect [27, 200, 252].
This synergy effect can play two distinctive roles in the deoxygenation process of
carboxylic acid. First, they can weaken the sulfur-metals bonds and therefore create more
vacant sites which can lead to higher activity of the catalyst [4, 253]. Second, they can
facilitate the crucial C=O hydrogenation and C-C bond cleavage in HDO and HDC
pathways by lowering the energy barriers for these steps [246].
136
Scheme 2: Proposed reaction mechanism in deoxygenation of stearic acid over the
prepared catalysts.
8.3 Conclusion
In this study, the deoxygenation of stearic acid was studied over recyclable sulfided
catalysts supported on greigite. All the catalysts showed high activity in the deoxynation
of stearic acid. NiMo/G was found to be the most active catalyst followed by CoMo/G and
Mo/G. The HDO route was found to be the dominant reaction pathway over all the
catalysts. Mo/G showed the highest C18/C17 ratio (9.2) at the end of reaction compared to
137
the promoted catalysts. The paraffin/olefin selectivity was always over 1 and increased
after the full conversion of stearic acid. NiMo/G showed the highest ratio between the
catalysts. Stearic acid also showed a strong inhibiting effect on the conversion of
octadecanol which proceeds through hydrogenation of C-O bond. Both of these inhibiting
effects indicated that stearic acid easily overwhelms other compounds in the competitive
adsorption over the hydrogenation sites of the catalysts, preventing other hydrogenation
reactions to progress simultaneously.
Three paths were considered for the aldehyde intermediate adsorbed on the catalyst:
desorption and forming octadecanal, conversion to octadecanol by hydride addition, and
forming heptadecene through HDC reaction pathway. The formation of Co(Ni)-Mo-S
species with synergic effects are believed to promote both C-O hydrogenation of alcohol
in HDO route and C-C bond scission of aldehyde in HDC rote. With the magnetic
properties of these catalysts, they can be easily collected and reused. This, in addition to
their high activity in oxygenation reactions makes them suitable catalysts to be used in co-
processing of fossil fuel and biomass.
138
Chapter 9: Conclusions and recommendations
In recent years with the depletion of the conventional light oil resources and the constant
increase in demand, the focuses are turned toward heavy and extra-heavy oil resources as
substitute oil supplies. Heavier oil resources contain higher amount of impurities which
can clog the beds inside the traditional fixed-bed reactors used for hydrotreating. Slurry
reactors are designed to deal with this problem using cheap material in a once-through put
mode as catalyst. However, the low activity of the catalysts employed in these reactors
results in very low quality product. In this work, we designed novel recyclable catalysts
that can be used in the slurry reactors producing higher quality products without
compromising the economics of the process as the catalysts can be collected using a
magnetic field and recycled back to the reactor.
Two different composites were designed. One with magnetite core and MoS2 shell and
the second composite was consisted of greigite core and Ni(Co)-MoS2 shell. Both
catalysts showed high magnetization, easily adsorbed using a hand-held magnet. Both
were tested in the HDS of DBT and found to be highly active. Former one showed high
affinity toward the DDS pathway while the later showed a balanced selectivity between
DDS and HYD pathways. This could give the refineries the option to adjust the
hydrotreating process based on their needs using each of the catalysts or a combination of
them.
NiMoS/Fe3S4 was found to have the highest activity in the hydrodesulfurization (HDS) of
dibenzothiophene (DBT) and DMDBT followed by CoMoS/Fe3S4 and MoS2/Fe3S4. For
the promoted catalysts, the direct desulfurization (DDS) pathway was found to be more
139
selectively enhanced than the HYD route in HDS of DBT. HYD pathway was found to be
the dominant route in HDS of dimethyldibenzothiophene (DMDBT).
The composite with greigite core was also employed in the HDO of stearic acid. The
activity of the catalysts could be ranked as NiMo/G > CoMo/G > Mo/G. HDO was the
dominant pathway for all of the catalysts. As for C18/C17 ratio, the catalysts were found
to be in the order: Mo/G > CoMo/G ≈ NiMo/G. Paraffin/Olefin ratio was over 1 for all of
the catalysts with NiMo/G showing the highest ratio. Stearic acid was found to be dominant
over other reactants in the competitive adsorption over the active sites, inhibiting
simultaneous reactions.
The catalyst containing magnetite core was used to investigate the differences in the roles
of H2 and H2S and the involved active sites in the HDS of DBT over MoS2. A series of
MoS2 catalysts with and without magnetite particles as support were prepared. Two
different active sites for hydrogenation (HYD) and hydrodesulfurization were identified on
MoS2. The core-shell magnetic MoS2 demonstrated high selectivity towards the DDS
pathway, which was attributable to the differences in adsorption energy of hydrogen and
DBT over hydrogenation and desulfurization sites. H2S functioned as a DDS inhibitor and
a HYD/isomerization promoter. H2S is favored being adsorbed on the S-edge vacancies to
transfer the active sites to HYD/Isomerization sites. Hydrogen plays three distinctive roles.
When adsorbed at Mo edge, hydrogen promotes hydrogenation; Strongly bonded at S-edge
it accounts for direct desulfurization while loosely bonded to S-edge it favors HYD and
isomerization. The creation of acid sites was found to be more facile through H2S than H2
dissociation.
140
For the Ni-promoted MoS2, the presence of Ni enhanced the reactions for both
hydrogenation and desulfurization over the catalyst. It was found that Ni addition increases
the concentration of CUS sites, lowers the H2S poisoning effect, and enhances the S-C
bond scission. It was concluded that the significant difference in acidity of Ni-promoted
and unpromoted MoS2 was due to the decrease in the sulfur bonding energy to the active
sites over promoted catalysts. the promotional effect of Ni was found to be a delocalized
effect affecting more than adjacent atoms. It was proposed that the Ni decoration starts at
the Mo edges of the catalyst. The higher HYD/DDS ratio over unsupported NiMoS
compared to the supported counterparts was assigned to the geometrical limitation of DBT
for flat adsorption at the edges of the supported catalyst and also reduction in the number
of available brim sites.
As a future work, different type of magnetic cores can be synthesized, tested, and compared
in their effect on the activity of the catalysts. Moreover, the activity of other type of
catalysts such as phosphide catalysts can be examined and any synergic effect between
catalysts and the magnetic core can be studied. Developing catalysts with intrinsic
magnetic characteristics can also be pursued.
Bibliography
[1] C. Song, An overview of new approaches to deep desulfurization for ultra-clean
gasoline, diesel fuel and jet fuel, Catalysis today, 86 (2003) 211-263.
[2] L. Castañeda, J. Muñoz, J. Ancheyta, Combined process schemes for upgrading of
heavy petroleum, Fuel, 100 (2012) 110-127.
[3] J.G. Speight, in: The Chemistry and Technology of Petroleum, Taylor & Francis
Group, 2007, pp. 605.
[4] R. Chianelli, G. Berhault, P. Raybaud, S. Kasztelan, J. Hafner, H. Toulhoat, Periodic
trends in hydrodesulfurization: in support of the Sabatier principle, Applied Catalysis A:
General, 227 (2002) 83-96.
[5] C. Leyva, M.S. Rana, F. Trejo, J. Ancheyta, On the use of acid-base-supported catalysts
for hydroprocessing of heavy petroleum, Industrial & Engineering Chemistry Research, 46
(2007) 7448-7466.
[6] J. Ancheyta, J.G. Speight, Hydroprocessing of heavy oils and residua, CRC Press, 2007.
[7] C. O’Melia, W. Ali, The role of retained particles in deep bed filtration, Progress in
Water Technology, 10 (1978) 6.
[8] A. Ortiz‐Arroyo, F. Larachi, B. Grandjean, S. Roy, CFD modeling and simulation of
clogging in packed beds with nonaqueous media, AIChE journal, 48 (2002) 1596-1609.
[9] T. P., Engineering of hydrotreating processes, in: H.d.D. Lasa, G.; Ravella, A. (Ed.)
Chemical reactor technology for environmentally safe reactors and products, 1993, pp.
425-442.
[10] C.J.R. Narayan R., Masliyah J.H. and Gray M.R., Particle capture and plugging in
packed-bed reactors, in: M. R.A. (Ed.) Handbook of petroleum refining processes,
McGraw-Hill, New York, 1996, pp. 4620.
[11] B. Delmon, P. Grange, G. Froment, Hydrotreatment and hydrocracking of oil
fractions, Elsevier, 1997.
[12] E. Furimsky, Selection of catalysts and reactors for hydroprocessing, Applied
Catalysis A: General, 171 (1998) 177-206.
[13] P.R. Robinson, G.E. Dolbear, Hydrotreating and hydrocracking: fundamentals, in:
Practical advances in petroleum processing, Springer, 2006, pp. 177-218.
[14] J.G. Speight, B. Ozum, Petroleum refining processes, CRC Press, 2001.
[15] K. Niemann, F. Wenzel, The VEBA-COMBI-CRACKING-Technology: An Update,
Fuel processing technology, 35 (1993) 1-20.
[16] M. Menzies, A. Silva, J. Denis, Hydrocracking without catalysis upgrades heavy oil,
CHEMICAL ENGINEERING, 88 (1981) 46-47.
[17] A. Quitian, C. Leyva, S. Ramirez, J. Ancheyta, An exploratory study for the upgrading
of transport properties of heavy oil by disperse phase hydrocracking, Energy & Fuels, 29
(2015) 9-15.
[18] G. Bellussi, G. Rispoli, A. Landoni, R. Millini, D. Molinari, E. Montanari, D.
Moscotti, P. Pollesel, Hydroconversion of heavy residues in slurry reactors: Developments
and perspectives, Journal of catalysis, 308 (2013) 189-200.
[19] A. Delbianco, S. Meli, L. Tagliabue, N. Panariti, Eni Slurry Technology: A New
Process For Heavy Oil Upgrading, in: 19th World Petroleum Congress, World Petroleum
Congress, 2008.
[20] H. Rezaei, S.J. Ardakani, K.J. Smith, Study of MoS2 Catalyst Recycle in Slurry-Phase
Residue Hydroconversion, Energy & Fuels, 26 (2012) 6540-6550.
[21] H. Topsøe, B.S. Clausen, F.E. Massoth, Hydrotreating catalysis, in: Catalysis,
Springer, 1996, pp. 1-269.
[22] F. Cui, G. Li, X. Li, M. Lu, M. Li, Enhancement of hydrodesulfurization of 4, 6-
dimethyldibenzothiophene catalyzed by CoMo catalysts supported on carbon-covered γ-
Al 2 O 3, Catalysis Science & Technology, 5 (2015) 549-555.
[23] F. Van Looij, P. Van der Laan, W. Stork, D. DiCamillo, J. Swain, Key parameters in
deep hydrodesulfurization of diesel fuel, Applied Catalysis A: General, 170 (1998) 1-12.
[24] A. Stanislaus, A. Marafi, M.S. Rana, Recent advances in the science and technology
of ultra low sulfur diesel (ULSD) production, Catalysis Today, 153 (2010) 1-68.
[25] H. Topsøie, R. Candia, N.Y. Topsøe, B.S. Clausen, H. Topsøe, On The State of the
Co‐MO‐S Model, Bulletin des Sociétés Chimiques Belges, 93 (1984) 783-806.
[26] H. Tops⊘ e, B.S. Clausen, Importance of Co-Mo-S type structures in
hydrodesulfurization, Catalysis Reviews Science and Engineering, 26 (1984) 395-420.
[27] J.V. Lauritsen, J. Kibsgaard, G.H. Olesen, P.G. Moses, B. Hinnemann, S. Helveg, J.K.
Nørskov, B.S. Clausen, H. Topsøe, E. Lægsgaard, Location and coordination of promoter
atoms in Co-and Ni-promoted MoS 2-based hydrotreating catalysts, Journal of Catalysis,
249 (2007) 220-233.
[28] S. Bouwens, J. Van Veen, D. Koningsberger, V. De Beer, R. Prins, Extended X-Ray
absorption fine structure determination of the structure of cobalt in carbonsupported Co
and Co-Mo sulfide hydrodesulfurization catalysts, J. Phys. Chem, 95 (1991) 123-134.
[29] P. Raybaud, J. Hafner, G. Kresse, S. Kasztelan, H. Toulhoat, Structure, energetics,
and electronic properties of the surface of a promoted MoS 2 catalyst: an ab initio local
density functional study, Journal of Catalysis, 190 (2000) 128-143.
[30] Y. Zhu, Q.M. Ramasse, M. Brorson, P.G. Moses, L.P. Hansen, C.F. Kisielowski, S.
Helveg, Visualizing the Stoichiometry of Industrial‐Style Co‐Mo‐S Catalysts with Single‐
Atom Sensitivity, Angewandte Chemie International Edition, 53 (2014) 10723-10727.
[31] A.K. Tuxen, H.G. Füchtbauer, B. Temel, B. Hinnemann, H. Topsøe, K.G. Knudsen,
F. Besenbacher, J.V. Lauritsen, Atomic-scale insight into adsorption of sterically hindered
dibenzothiophenes on MoS 2 and Co–Mo–S hydrotreating catalysts, Journal of Catalysis,
295 (2012) 146-154.
[32] J.V. Lauritsen, J. Kibsgaard, S. Helveg, H. Topsøe, B.S. Clausen, E. Lægsgaard, F.
Besenbacher, Size-dependent structure of MoS2 nanocrystals, Nature nanotechnology, 2
(2007) 53-58.
[33] T.F. Jaramillo, K.P. Jørgensen, J. Bonde, J.H. Nielsen, S. Horch, I. Chorkendorff,
Identification of active edge sites for electrochemical H2 evolution from MoS2
nanocatalysts, science, 317 (2007) 100-102.
[34] P.G. Moses, L.C. Grabow, E.M. Fernandez, B. Hinnemann, H. Topsøe, K.G. Knudsen,
J.K. Nørskov, Trends in hydrodesulfurization catalysis based on realistic surface models,
Catalysis Letters, 144 (2014) 1425-1432.
[35] N. Topsoe, H. Topsoe, FTIR studies of Mo/Al 2 O 3-based catalysts: II. Evidence for
the presence of SH groups and their role in acidity and activity, Journal of Catalysis, 139
(1993) 641-651.
[36] F. Maugé, J. Lamotte, N. Nesterenko, O. Manoilova, A. Tsyganenko, FT-IR study of
surface properties of unsupported MoS 2, Catalysis today, 70 (2001) 271-284.
[37] J. Chen, F. Maugé, J. El Fallah, L. Oliviero, IR spectroscopy evidence of MoS 2
morphology change by citric acid addition on MoS 2/Al 2 O 3 catalysts–A step forward to
differentiate the reactivity of M-edge and S-edge, Journal of Catalysis, 320 (2014) 170-
179.
[38] H. Farag, A.-N.A. El-Hendawy, K. Sakanishi, M. Kishida, I. Mochida, Catalytic
activity of synthesized nanosized molybdenum disulfide for the hydrodesulfurization of
dibenzothiophene: Effect of H 2 S partial pressure, Applied Catalysis B: Environmental,
91 (2009) 189-197.
[39] H. Farag, K. Sakanishi, M. Kouzu, A. Matsumura, Y. Sugimoto, I. Saito, Investigation
of the influence of H2S on hydrodesulfurization of dibenzothiophene over a bulk MoS2
catalyst, Industrial & engineering chemistry research, 42 (2003) 306-310.
[40] P. Wiwel, B. Hinnemann, A. Hidalgo-Vivas, P. Zeuthen, B.O. Petersen, J.Ø. Duus,
Characterization and identification of the most refractory nitrogen compounds in
hydroprocessed vacuum gas oil, Industrial & Engineering Chemistry Research, 49 (2010)
3184-3193.
[41] F. Bataille, J.-L. Lemberton, P. Michaud, G. Pérot, M. Vrinat, M. Lemaire, E. Schulz,
M. Breysse, S. Kasztelan, Alkyldibenzothiophenes hydrodesulfurization-promoter effect,
reactivity, and reaction mechanism, Journal of catalysis, 191 (2000) 409-422.
[42] M. Vrinat, R. Bacaud, D. Laurenti, M. Cattenot, N. Escalona, S. Gamez, New trends
in the concept of catalytic sites over sulfide catalysts, Catalysis today, 107 (2005) 570-577.
[43] M. Egorova, R. Prins, Hydrodesulfurization of dibenzothiophene and 4, 6-
dimethyldibenzothiophene over sulfided NiMo/γ-Al 2 O 3, CoMo/γ-Al 2 O 3, and Mo/γ-
Al 2 O 3 catalysts, Journal of Catalysis, 225 (2004) 417-427.
[44] M. Breysse, P. Afanasiev, C. Geantet, M. Vrinat, Overview of support effects in
hydrotreating catalysts, Catalysis Today, 86 (2003) 5-16.
[45] T.C. Ho, Inhibiting effects in hydrodesulfurization of 4, 6-diethyldibenzothiophene,
Journal of Catalysis, 219 (2003) 442-451.
[46] L.S. Byskov, M. Bollinger, J.K. Nørskov, B.S. Clausen, H. Topsøe, Molecular aspects
of the H 2 activation on MoS 2 based catalysts—the role of dynamic surface arrangements,
Journal of Molecular Catalysis A: Chemical, 163 (2000) 117-122.
[47] N. Arun, R.V. Sharma, A.K. Dalai, Green diesel synthesis by hydrodeoxygenation of
bio-based feedstocks: Strategies for catalyst design and development, Renewable and
Sustainable Energy Reviews, 48 (2015) 240-255.
[48] R.W. Gosselink, S.A. Hollak, S.W. Chang, J. van Haveren, K.P. de Jong, J.H. Bitter,
D.S. van Es, Reaction pathways for the deoxygenation of vegetable oils and related model
compounds, ChemSusChem, 6 (2013) 1576-1594.
[49] E. Furimsky, Catalytic hydrodeoxygenation, Applied Catalysis A: General, 199
(2000) 147-190.
[50] T. Choudhary, C. Phillips, Renewable fuels via catalytic hydrodeoxygenation,
Applied Catalysis A: General, 397 (2011) 1-12.
[51] D.C. Elliott, Historical developments in hydroprocessing bio-oils, Energy & Fuels, 21
(2007) 1792-1815.
[52] P.M. Mortensen, J.-D. Grunwaldt, P.A. Jensen, K. Knudsen, A.D. Jensen, A review
of catalytic upgrading of bio-oil to engine fuels, Applied Catalysis A: General, 407 (2011)
1-19.
[53] G. Fogassy, N. Thegarid, G. Toussaint, A.C. van Veen, Y. Schuurman, C. Mirodatos,
Biomass derived feedstock co-processing with vacuum gas oil for second-generation fuel
production in FCC units, Applied Catalysis B: Environmental, 96 (2010) 476-485.
[54] A. Lappas, S. Bezergianni, I. Vasalos, Production of biofuels via co-processing in
conventional refining processes, Catalysis Today, 145 (2009) 55-62.
[55] G.W. Huber, P. O’Connor, A. Corma, Processing biomass in conventional oil
refineries: production of high quality diesel by hydrotreating vegetable oils in heavy
vacuum oil mixtures, Applied Catalysis A: General, 329 (2007) 120-129.
[56] C. Templis, A. Vonortas, I. Sebos, N. Papayannakos, Vegetable oil effect on gasoil
HDS in their catalytic co-hydroprocessing, Applied Catalysis B: Environmental, 104
(2011) 324-329.
[57] A. Pinheiro, D. Hudebine, N. Dupassieux, C. Geantet, Impact of oxygenated
compounds from lignocellulosic biomass pyrolysis oils on gas oil hydrotreatment, Energy
& Fuels, 23 (2009) 1007-1014.
[58] V.N. Bui, G. Toussaint, D. Laurenti, C. Mirodatos, C. Geantet, Co-processing of
pyrolisis bio oils and gas oil for new generation of bio-fuels: Hydrodeoxygenation of
guaïacol and SRGO mixed feed, Catalysis Today, 143 (2009) 172-178.
[59] R. Kumar, B.S. Rana, R. Tiwari, D. Verma, R. Kumar, R.K. Joshi, M.O. Garg, A.K.
Sinha, Hydroprocessing of jatropha oil and its mixtures with gas oil, Green Chemistry, 12
(2010) 2232-2239.
[60] M. Philippe, F. Richard, D. Hudebine, S. Brunet, Inhibiting effect of oxygenated
model compounds on the HDS of dibenzothiophenes over CoMoP/Al 2 O 3 catalyst,
Applied Catalysis A: General, 383 (2010) 14-23.
[61] M. Toba, Y. Abe, H. Kuramochi, M. Osako, T. Mochizuki, Y. Yoshimura,
Hydrodeoxygenation of waste vegetable oil over sulfide catalysts, Catalysis Today, 164
(2011) 533-537.
[62] P. Priecel, D. Kubička, L. Čapek, Z. Bastl, P. Ryšánek, The role of Ni species in the
deoxygenation of rapeseed oil over NiMo-alumina catalysts, Applied Catalysis A: General,
397 (2011) 127-137.
[63] H. Zhang, H. Lin, Y. Zheng, The role of cobalt and nickel in deoxygenation of
vegetable oils, Applied Catalysis B: Environmental, 160 (2014) 415-422.
[64] S.K. Kim, J.Y. Han, H.-s. Lee, T. Yum, Y. Kim, J. Kim, Production of renewable
diesel via catalytic deoxygenation of natural triglycerides: Comprehensive understanding
of reaction intermediates and hydrocarbons, Applied Energy, 116 (2014) 199-205.
[65] D. Kubička, L. Kaluža, Deoxygenation of vegetable oils over sulfided Ni, Mo and
NiMo catalysts, Applied Catalysis A: General, 372 (2010) 199-208.
[66] S. Hama, S. Tamalampudi, A. Yoshida, N. Tamadani, N. Kuratani, H. Noda, H.
Fukuda, A. Kondo, Enzymatic packed-bed reactor integrated with glycerol-separating
system for solvent-free production of biodiesel fuel, Biochemical engineering journal, 55
(2011) 66-71.
[67] G.E.D. Paul R. Robinson, HYDROTREATING AND HYDROCRACKING:
FUNDAMENTALS, in: P.R.R. Chang S. Hsu (Ed.) Practical Advances in Petroleum
Processing, Springer, 2006, pp. 178.
[68] T.H.-S.T. Takatsuka, Kabe, T. (Ed.), IPC: Tokyo, 2000.
[69] E. Furimsky, Applied Catalysis A, 199 (2000) 147.
[70] T.A.A.-S. Mohamed A. Fahim, Haitham M. S. Lababidi, Amal Sayed Elkilani,
Fundamentals of Petroleum Refining, 1st ed., Elsevier, 2009.
[71] P.R. Nicholas P. Cheremisinoff, Best practices in petroleum industry, 1st ed., Elsevier,
2009.
[72] C. Song, X. Ma, Ultra-deep desulfurization of liquid hydrocarbon fuels: Chemistry
and process, International journal of green energy, 1 (2004) 167-191.
[73] M.J. Girgis, B.C. Gates, Reactivities, reaction networks, and kinetics in high-pressure
catalytic hydroprocessing, Industrial & Engineering Chemistry Research, 30 (1991) 2021-
2058.
[74] M. Houalla, N. Nag, A. Sapre, D. Broderick, B. Gates, Hydrodesulfurization of
dibenzothiophene catalyzed by sulfided CoO‐MoO3γ‐Al2O3: The reaction network,
AIChE Journal, 24 (1978) 1015-1021.
[75] M.K. Andari, F. Abu-Seedo, A. Stanislaus, H.M. Qabazard, Kinetics of individual
sulfur compounds in deep hydrodesulfurization of Kuwait diesel oil, Fuel, 75 (1996) 1664-
1670.
[76] T. Kabe, A. Ishihara, H. Tajima, Hydrodesulfurization of sulfur-containing
polyaromatic compounds in light oil, Industrial & Engineering Chemistry Research, 31
(1992) 1577-1580.
[77] X. Ma, K. Sakanishi, I. Mochida, Hydrodesulfurization reactivities of various sulfur
compounds in vacuum gas oil, Industrial & Engineering Chemistry Research, 35 (1996)
2487-2494.
[78] B. Gates, H. Topsøe, Reactivities in deep catalytic hydrodesulfurization: challenges,
opportunities, and the importance of 4-methyldibenzothiophene and 4, 6-
dimethyldibenzothiophene, Polyhedron, 16 (1997) 3213-3217.
[79] D.D. Whitehurst, T. Isoda, I. Mochida, Present state of the art and future challenges
in the hydrodesulfurization of polyaromatic sulfur compounds, Advances in catalysis, 42
(1998) 345-471.
[80] M. Daage, R. Chianelli, Structure-function relations in molybdenum sulfide catalysts:
The" Rim-Edge" model, Journal of Catalysis, 149 (1994) 414-427.
[81] X. Ma, K. Sakanishi, T. Isoda, I. Mochida, Quantum chemical calculation on the
desulfurization reactivities of heterocyclic sulfur compounds, Energy & Fuels, 9 (1995)
33-37.
[82] K.G. Knudsen, B.H. Cooper, H. Topsøe, Catalyst and process technologies for ultra
low sulfur diesel, Applied Catalysis A: General, 189 (1999) 205-215.
[83] H. Schulz, W. Böhringer, P. Waller, F. Ousmanov, Gas oil deep hydrodesulfurization:
refractory compounds and retarded kinetics, Catalysis today, 49 (1999) 87-97.
[84] V. Meille, E. Schulz, M. Lemaire, M. Vrinat, Hydrodesulfurization of
Alkyldibenzothiophenes over a NiMo/Al< sub> 2</sub> O< sub> 3</sub> Catalyst:
Kinetics and Mechanism, Journal of Catalysis, 170 (1997) 29-36.
[85] J. Mijoin, G. Pérot, F. Bataille, J. Lemberton, M. Breysse, S. Kasztelan, Mechanistic
considerations on the involvement of dihydrointermediates in the hydrodesulfurization of
dibenzothiophene-type compounds over molybdenum sulfide catalysts, Catalysis letters,
71 (2001) 139-145.
[86] M. Houalla, D. Broderick, A. Sapre, N. Nag, V. De Beer, B. Gates, H. Kwart,
Hydrodesulfurization of methyl-substituted dibenzothiophenes catalzed by sulfided Co-
MO/. gamma.-Al/sub 2/O/sub 3, J. Catal.;(United States), 61 (1980).
[87] R. Shafi, G.J. Hutchings, Hydrodesulfurization of hindered dibenzothiophenes: an
overview, Catalysis today, 59 (2000) 423-442.
[88] B.S.C. H. Topsoe, F.E. Massoth, in: M.B. J.R. Anderson (Ed.) Hydrotreating Catalysis
science and Technology, Springer Verlag, Berlin, 1996.
[89] G. Pérot, Hydrotreating catalysts containing zeolites and related materials—
mechanistic aspects related to deep desulfurization, Catalysis today, 86 (2003) 111-128.
[90] H. Topsøe, B.S. Clausen, F.E. Massoth, Hydrotreating catalysis, Springer, 1996.
[91] R. Prins, M. Jian, M. Flechsenhar, Mechanism and kinetics of hydrodenitrogenation,
Polyhedron, 16 (1997) 3235-3246.
[92] C. Kwak, J.J. Lee, J.S. Bae, S.H. Moon, Poisoning effect of nitrogen compounds on
the performance of CoMoS/Al< sub> 2</sub> O< sub> 3</sub> catalyst in the
hydrodesulfurization of dibenzothiophene, 4-methyldibenzothiophene, and 4, 6-
dimethyldibenzothiophene, Applied Catalysis B: Environmental, 35 (2001) 59-68.
[93] H. Farag, M. Kishida, H. Al-Megren, Competitive hydrodesulfurization of
dibenzothiophene and hydrodenitrogenation of quinoline over unsupported MoS< sub>
2</sub> catalyst, Applied Catalysis A: General, 469 (2014) 173-182.
[94] J.G. Speight, The chemistry and technology of petroleum, CRC press, 2014.
[95] G.W. Huber, A. Corma, Synergies between Bio‐and Oil Refineries for the Production
of Fuels from Biomass, Angewandte Chemie International Edition, 46 (2007) 7184-7201.
[96] B. Donnis, R.G. Egeberg, P. Blom, K.G. Knudsen, Hydroprocessing of bio-oils and
oxygenates to hydrocarbons. Understanding the reaction routes, Topics in Catalysis, 52
(2009) 229-240.
[97] P. Šimáček, D. Kubička, G. Šebor, M. Pospíšil, Hydroprocessed rapeseed oil as a
source of hydrocarbon-based biodiesel, Fuel, 88 (2009) 456-460.
[98] P. Šimáček, D. Kubička, G. Šebor, M. Pospíšil, Fuel properties of hydroprocessed
rapeseed oil, Fuel, 89 (2010) 611-615.
[99] J. Walendziewski, M. Stolarski, R. Łużny, B. Klimek, Hydroprocesssing of light gas
oil—rape oil mixtures, Fuel Processing Technology, 90 (2009) 686-691.
[100] M. Krár, S. Kovács, D. Kalló, J. Hancsók, Fuel purpose hydrotreating of sunflower
oil on CoMo/Al 2 O 3 catalyst, Bioresource technology, 101 (2010) 9287-9293.
[101] A. Gutierrez, R. Kaila, M. Honkela, R. Slioor, A. Krause, Hydrodeoxygenation of
guaiacol on noble metal catalysts, Catalysis today, 147 (2009) 239-246.
[102] J. Wildschut, F.H. Mahfud, R.H. Venderbosch, H.J. Heeres, Hydrotreatment of fast
pyrolysis oil using heterogeneous noble-metal catalysts, Industrial & Engineering
Chemistry Research, 48 (2009) 10324-10334.
[103] V. Yakovlev, S. Khromova, O. Sherstyuk, V. Dundich, D.Y. Ermakov, V.
Novopashina, M.Y. Lebedev, O. Bulavchenko, V. Parmon, Development of new catalytic
systems for upgraded bio-fuels production from bio-crude-oil and biodiesel, Catalysis
today, 144 (2009) 362-366.
[104] J. Wildschut, J. Arentz, C. Rasrendra, R. Venderbosch, H. Heeres, Catalytic
hydrotreatment of fast pyrolysis oil: model studies on reaction pathways for the
carbohydrate fraction, Environmental progress & sustainable energy, 28 (2009) 450-460.
[105] V.N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Hydrodeoxygenation of guaiacol
with CoMo catalysts. Part I: Promoting effect of cobalt on HDO selectivity and activity,
Applied Catalysis B: Environmental, 101 (2011) 239-245.
[106] R.J. French, J. Stunkel, S. Black, M. Myers, M.M. Yung, K. Iisa, Evaluate Impact of
Catalyst Type on Oil Yield and Hydrogen Consumption from Mild Hydrotreating, Energy
& Fuels, 28 (2014) 3086-3095.
[107] M. Badawi, J. Paul, S. Cristol, E. Payen, Y. Romero, F. Richard, S. Brunet, D.
Lambert, X. Portier, A. Popov, Effect of water on the stability of Mo and CoMo
hydrodeoxygenation catalysts: A combined experimental and DFT study, Journal of
Catalysis, 282 (2011) 155-164.
[108] V.N. Bui, D. Laurenti, P. Delichère, C. Geantet, Hydrodeoxygenation of guaiacol:
Part II: Support effect for CoMoS catalysts on HDO activity and selectivity, Applied
Catalysis B: Environmental, 101 (2011) 246-255.
[109] I. Gandarias, V. Barrio, J. Requies, P. Arias, J. Cambra, M. Güemez, From biomass
to fuels: Hydrotreating of oxygenated compounds, International Journal of Hydrogen
Energy, 33 (2008) 3485-3488.
[110] V.M. Whiffen, K.J. Smith, Hydrodeoxygenation of 4-Methylphenol over
Unsupported MoP, MoS2, and MoO x Catalysts†, Energy & Fuels, 24 (2010) 4728-4737.
[111] S. Eijsbouts, S. Mayo, K. Fujita, Unsupported transition metal sulfide catalysts: From
fundamentals to industrial application, Applied Catalysis A: General, 322 (2007) 58-66.
[112] B. Yoosuk, J.H. Kim, C. Song, C. Ngamcharussrivichai, P. Prasassarakich, Highly
active MoS2, CoMoS2 and NiMoS2 unsupported catalysts prepared by hydrothermal
synthesis for hydrodesulfurization of 4, 6-dimethyldibenzothiophene, Catalysis today, 130
(2008) 14-23.
[113] T.C. Ho, Hydroprocessing catalysis on metal sulfides prepared from molecular
complexes, Catalysis today, 130 (2008) 206-220.
[114] Y. Iwata, K. Sato, T. Yoneda, Y. Miki, Y. Sugimoto, A. Nishijima, H. Shimada,
Catalytic functionality of unsupported molybdenum sulfide catalysts prepared with
different methods, Catalysis today, 45 (1998) 353-359.
[115] R. Chianelli, M. Daage, M. Ledoux, Transition-Metal Sulfide Catalytic, Advances
in catalysis, 40 (1994) 177.
[116] M. Kouzu, K. Uchida, Y. Kuriki, F. Ikazaki, Micro-crystalline molybdenum sulfide
prepared by mechanical milling as an unsupported model catalyst for the
hydrodesulfurization of diesel fuel, Applied Catalysis A: General, 276 (2004) 241-249.
[117] C. Moreau, J. Joffre, C. Saenz, J.C. Afonso, J.L. Portefaix, Mechanism of carbon sp<
sup> 2</sup>-heteroatom bond cleavage in hydroprocessing of substituted benzenes over
unsupported transition metal sulfides, Journal of Molecular Catalysis A: Chemical, 161
(2000) 141-147.
[118] A. Olivas, J. Cruz-Reyes, V. Petranovskii, M. Avalos, S. Fuentes, Synthesis and
characterization of nickel sulfide catalysts, Journal of Vacuum Science & Technology A:
Vacuum, Surfaces, and Films, 16 (1998) 3515-3520.
[119] E.T. C.J.H. Jacobsen, H. Topsøe, catalysis letters, 63 (1999) 179.
[120] L.S. Byskov, J.K. Nørskov, B.S. Clausen, H. Topsøe, DFT Calculations of
Unpromoted and Promoted MoS< sub> 2</sub>-Based Hydrodesulfurization Catalysts,
Journal of Catalysis, 187 (1999) 109-122.
[121] K. Inamura, R. Prins, The role of Co in unsupported Co-Mo sulfides in the
hydrodesulfurization of thiophene, Stud. Surf. Sci. Catal, 92 (1995) 401.
[122] M. Karroua, H. Matralis, E. Sham, P. Grange, B. Delmon, Synergy between the
CoMoS Phase and Supported or Unsupported Cobalt Sulfide. Existence of a Remote
Control Effect, Bulletin of the Chemical Society of Japan, 68 (1995) 107-119.
[123] M. Lacroix, M. Vrinat, M. Breysse, Unsupported nickel tungsten sulfide catalysts:
Part 1: Catalytic behaviour in hydrogenation and hydrodesulfurization reactions, Applied
catalysis, 21 (1986) 73-83.
[124] Y. Gochi, C. Ornelas, F. Paraguay, S. Fuentes, L. Alvarez, J. Rico, G. Alonso-Núñez,
Effect of sulfidation on Mo-W-Ni trimetallic catalysts in the HDS of DBT, Catalysis today,
107 (2005) 531-536.
[125] M. Garcia, J. Lindner, A. Sachdev, J. Schwank, Model hydrodesulfurization
catalysts: Solid state synthesis and characterization of iron containing molybdenum
sulphide, Applied catalysis, 56 (1989) 281-295.
[126] A.P. Raje, S.J. Liaw, R. Srinivasan, B.H. Davis, Second row transition metal sulfides
for the hydrotreatment of coal-derived naphtha I. Catalyst preparation, characterization and
comparison of rate of simultaneous removal of total sulfur, nitrogen and oxygen, Applied
Catalysis A: General, 150 (1997) 297-318.
[127] K.L. Riley, D.P. Klein, Z. Hou, S.L. Soled, M.C. Kerby, G.B. McVicker, E.S. Ellis,
M.S. Touvelle, S. Miseo, Hydrotreating using bulk multimetallic catalysts, in, Google
Patents, 2004.
[128] S.L. Soled, S. Miseo, R. Krycak, H. Vroman, T.C. Ho, K.L. Riley, Nickel
molybodtungstate hydrotreating catalysts (law444), in, Google Patents, 2001.
[129] Y. Peng, Z. Meng, C. Zhong, J. Lu, W. Yu, Z. Yang, Y. Qian, Hydrothermal
Synthesis of MoS< sub> 2</sub> and Its Pressure-Related Crystallization, Journal of Solid
State Chemistry, 159 (2001) 170-173.
[130] Y. Tian, Y. He, Y. Zhu, Low temperature synthesis and characterization of
molybdenum disulfide nanotubes and nanorods, Materials Chemistry and Physics, 87
(2004) 87-90.
[131] W.J. Li, E.W. Shi, J.M. Ko, Z. Chen, H. Ogino, T. Fukuda, Hydrothermal synthesis
of MoS< sub> 2</sub> nanowires, Journal of Crystal Growth, 250 (2003) 418-422.
[132] H. Shimada, T. Sato, Y. Yoshimura, J. Hiraishi, A. Nishijima, Support effect on the
catalytic activity and properties of sulfided molybdenum catalysts, Journal of Catalysis,
110 (1988) 275-284.
[133] H. Farag, I. Mochida, K. Sakanishi, Fundamental comparison studies on
hydrodesulfurization of dibenzothiophenes over CoMo-based carbon and alumina
catalysts, Applied Catalysis A: General, 194 (2000) 147-157.
[134] F. Luck, A review of support effects on the activity and selectivity of hydrotreating
catalysts, Bulletin des Sociétés Chimiques Belges, 100 (1991) 781-800.
[135] G.M. Dhar, B. Srinivas, M. Rana, M. Kumar, S. Maity, Mixed oxide supported
hydrodesulfurization catalysts—a review, Catalysis today, 86 (2003) 45-60.
[136] D.C. Phillips, S.J. Sawhill, R. Self, M.E. Bussell, Synthesis, characterization, and
hydrodesulfurization properties of silica-supported molybdenum phosphide catalysts,
Journal of Catalysis, 207 (2002) 266-273.
[137] K. Chary, H. Ramakrishna, K. Rama Rao, G. Murali Dhar, P. Kanta Rao,
Hydrodesulfurization on MoS 2/MgO, Catalysis letters, 10 (1991) 27-33.
[138] S. Maity, M. Rana, B. Srinivas, S. Bej, G. Murali Dhar, T. Prasada Rao,
Characterization and evaluation of ZrO< sub> 2</sub> supported hydrotreating catalysts,
Journal of Molecular Catalysis A: Chemical, 153 (2000) 121-127.
[139] Y. Ji, P. Afanasiev, M. Vrinat, W. Li, C. Li, Promoting effects in hydrogenation and
hydrodesulfurization reactions on the zirconia and titania supported catalysts, Applied
Catalysis A: General, 257 (2004) 157-164.
[140] J. Ramirez, G. Macias, L. Cedeño, A. Gutierrez-Alejandre, R. Cuevas, P. Castillo,
The role of titania in supported Mo, CoMo, NiMo, and NiW hydrodesulfurization catalysts:
analysis of past and new evidences, Catalysis today, 98 (2004) 19-30.
[141] H. Farag, D. Whitehurst, K. Sakanishi, I. Mochida, Carbon versus alumina as a
support for Co-Mo catalysts reactivity towards HDS of dibenzothiophenes and diesel fuel,
Catalysis today, 50 (1999) 9-17.
[142] B. Pawelec, R. Mariscal, J. Fierro, A. Greenwood, P. Vasudevan, Carbon-supported
tungsten and nickel catalysts for hydrodesulfurization and hydrogenation reactions,
Applied Catalysis A: General, 206 (2001) 295-307.
[143] A. Hynaux, C. Sayag, S. Suppan, J. Trawczynski, M. Lewandowski, A. Szymanska-
Kolasa, G. Djéga-Mariadassou, Kinetic study of the deep hydrodesulfurization of
dibenzothiophene over molybdenum carbide supported on a carbon black composite:
Existence of two types of active sites, Catalysis today, 119 (2007) 3-6.
[144] A.D. G. Wan, Y. Zhang, Z. Zhao, G. Jiang, D. Zhang, Z. Gao, Catalysis today, 149
(2010) 69-75.
[145] S. Mohanty, K. Chandra Mouli, K. Soni, J. Adjaye, A. Dalai, Catalytic
hydrotreatment using NiMo/MAS catalysts synthesized from ZSM-5 nano-clusters,
Applied Catalysis A: General, (2011).
[146] A. Wang, Y. Wang, T. Kabe, Y. Chen, A. Ishihara, W. Qian, Hydrodesulfurization
of dibenzothiophene over siliceous MCM-41-supported catalysts: I. sulfided Co–Mo
catalysts, Journal of Catalysis, 199 (2001) 19-29.
[147] K.M. Reddy, B. Wei, C. Song, Mesoporous molecular sieve MCM-41 supported Co–
Mo catalyst for hydrodesulfurization of petroleum resids, Catalysis today, 43 (1998) 261-
272.
[148] G. Muthu Kumaran, S. Garg, K. Soni, M. Kumar, L. Sharma, G. Murali Dhar, K.
Rama Rao, Effect of Al-SBA-15 support on catalytic functionalities of hydrotreating
catalysts: I. Effect of variation of Si/Al ratio on catalytic functionalities, Applied Catalysis
A: General, 305 (2006) 123-129.
[149] T.I. Korányi, Z. Vít, D.G. Poduval, R. Ryoo, H.S. Kim, E.J.M. Hensen, SBA-15-
supported nickel phosphide hydrotreating catalysts, Journal of Catalysis, 253 (2008) 119-
131.
[150] S. Eijsbouts, On the flexibility of the active phase in hydrotreating catalysts, Applied
Catalysis A: General, 158 (1997) 53-92.
[151] A. Corma, From microporous to mesoporous molecular sieve materials and their use
in catalysis, Chemical reviews, 97 (1997) 2373-2420.
[152] Y. Li, D. Pan, C. Yu, Y. Fan, X. Bao, Synthesis and hydrodesulfurization properties
of NiW catalyst supported on high-aluminum-content, highly ordered, and hydrothermally
stable Al-SBA-15, Journal of Catalysis, (2011).
[153] S. Ted Oyama, H. Zhao, H.-J. Freund, K. Asakura, R. Włodarczyk, M. Sierka,
Unprecedented selectivity to the direct desulfurization (DDS) pathway in a highly active
FeNi bimetallic phosphide catalyst, Journal of Catalysis, 285 (2012) 1-5.
[154] J. Lauritsen, M. Nyberg, J.K. Nørskov, B. Clausen, H. Topsøe, E. Lægsgaard, F.
Besenbacher, Hydrodesulfurization reaction pathways on MoS< sub> 2</sub>
nanoclusters revealed by scanning tunneling microscopy, Journal of Catalysis, 224 (2004)
94-106.
[155] H. Topsøe, B. Hinnemann, J.K. Nørskov, J.V. Lauritsen, F. Besenbacher, P.L.
Hansen, G. Hytoft, R.G. Egeberg, K.G. Knudsen, The role of reaction pathways and
support interactions in the development of high activity hydrotreating catalysts, Catalysis
today, 107 (2005) 12-22.
[156] N.-Y. Topsøe, H. Topsøe, Characterization of the structures and active sites in
sulfided Co MoAl2O3 and Ni MoAl2O3 catalysts by NO chemisorption, Journal of
Catalysis, 84 (1983) 386-401.
[157] H. Topsøe, The role of Co–Mo–S type structures in hydrotreating catalysts, Applied
Catalysis A: General, 322 (2007) 3-8.
[158] S. Harris, R. Chianelli, Catalysis by transition metal sulfides: A theoretical and
experimental study of the relation between the synergic systems and the binary transition
metal sulfides, Journal of Catalysis, 98 (1986) 17-31.
[159] R. Candia, O. Sorensen, J. Villadsen, N.-Y. Topsøe, B. Clausen, H. Topsøe, Active
phase in sulfided NiMo/γ-Al2O3 catalyst, Bull Soc Chim Belg, 93 (1984) 763-773.
[160] Q. Liu, Z. Xu, J. Finch, R. Egerton, A novel two-step silica-coating process for
engineering magnetic nanocomposites, Chemistry of materials, 10 (1998) 3936-3940.
[161] M.F. Casula, P. Floris, C. Innocenti, A. Lascialfari, M. Marinone, M. Corti, R.A.
Sperling, W.J. Parak, C. Sangregorio, Magnetic resonance imaging contrast agents based
on iron oxide superparamagnetic ferrofluids, Chemistry of materials, 22 (2010) 1739-1748.
[162] O. Veiseh, J.W. Gunn, M. Zhang, Design and fabrication of magnetic nanoparticles
for targeted drug delivery and imaging, Advanced drug delivery reviews, 62 (2010) 284-
304.
[163] F.A. Khan, G. Süss‐Fink, Superparamagnetic Core‐Shell‐Type Fe3O4/Ru
Nanoparticles as Catalysts for the Selective Hydrogenation of an Unconstrained α, β‐
Unsaturated Ketone, European Journal of Inorganic Chemistry, 2012 (2012) 727-732.
[164] C.-H. Pélisson, L.L. Vono, C. Hubert, A. Denicourt-Nowicki, L.M. Rossi, A.
Roucoux, Moving from surfactant-stabilized aqueous rhodium (0) colloidal suspension to
heterogeneous magnetite-supported rhodium nanocatalysts: Synthesis, characterization
and catalytic performance in hydrogenation reactions, Catalysis today, 183 (2012) 124-
129.
[165] Q.M. Kainz, R. Linhardt, R.N. Grass, G. Vilé, J. Pérez‐Ramírez, W.J. Stark, O.
Reiser, Palladium Nanoparticles Supported on Magnetic Carbon‐Coated Cobalt
Nanobeads: Highly Active and Recyclable Catalysts for Alkene Hydrogenation, Advanced
Functional Materials, 24 (2014) 2020-2027.
[166] M. Shokouhimehr, Y. Piao, J. Kim, Y. Jang, T. Hyeon, A magnetically recyclable
nanocomposite catalyst for olefin epoxidation, Angewandte Chemie, 119 (2007) 7169-
7173.
[167] P. Wu, J. Zhu, Z. Xu, Template‐Assisted Synthesis of Mesoporous Magnetic
Nanocomposite Particles, Advanced Functional Materials, 14 (2004) 345-351.
[168] H. Zhang, H. Lin, Y. Zheng, Y. Hu, A. MacLennan, Understanding of the effect of
synthesis temperature on the crystallization and activity of nano-MoS 2 catalyst, Applied
Catalysis B: Environmental, 165 (2015) 537-546.
[169] J. Texter, Reactions and synthesis in surfactant systems, CRC Press, 2011.
[170] Z. Ma, Y. Guan, H. Liu, Superparamagnetic silica nanoparticles with immobilized
metal affinity ligands for protein adsorption, Journal of magnetism and magnetic materials,
301 (2006) 469-477.
[171] V. Malik, K.J. Suthar, D.C. Mancini, J. Ilavsky, Magnetic-field-dependent assembly
of silica-coated magnetite nanoclusters probed by Ultra-Small-Angle X-ray Scattering
(USAXS), Journal of magnetism and magnetic materials, 354 (2014) 70-75.
[172] Z.-Y. Ma, X.-Q. Liu, Y.-P. Guan, H.-Z. Liu, Synthesis of magnetic silica nanospheres
with metal ligands and application in affinity separation of proteins, Colloids and Surfaces
A: Physicochemical and Engineering Aspects, 275 (2006) 87-91.
[173] D. Valencia, T. Klimova, Citric acid loading for MoS< sub> 2</sub>-based catalysts
supported on SBA-15. New catalytic materials with high hydrogenolysis ability in
hydrodesulfurization, Applied Catalysis B: Environmental, (2012).
[174] T.C. Ho, Deep HDS of diesel fuel: chemistry and catalysis, Catalysis today, 98
(2004) 3-18.
[175] A.K. Tuxen, H.G. Füchtbauer, B. Temel, B. Hinnemann, H. Topsøe, K.G. Knudsen,
F. Besenbacher, J.V. Lauritsen, Atomic-scale insight into adsorption of sterically hindered
dibenzothiophenes on MoS< sub> 2</sub> and Co–Mo–S hydrotreating catalysts, Journal
of Catalysis, (2012).
[176] F. Bataille, J. Lemberton, G. Perot, P. Leyrit, T. Cseri, N. Marchal, S. Kasztelan,
Sulfided Mo and CoMo supported on zeolite as hydrodesulfurization catalysts:
transformation of dibenzothiophene and 4, 6-dimethyldibenzothiophene, Applied Catalysis
A: General, 220 (2001) 191-205.
[177] M. Albiter, R. Huirache-Acuna, F. Paraguay-Delgado, J. Rico, G. Alonso-Nunez,
Synthesis of MoS2 nanorods and their catalytic test in the HDS of Dibenzothiophene,
Nanotechnology, 17 (2006) 3473.
[178] R. Romero-Rivera, A. Camacho, M. Del Valle, G. Alonso, S. Fuentes, J. Cruz-Reyes,
HDS of DBT with Molybdenum Disulfide Catalysts Prepared by In Situ Decomposition of
Alkyltrimethylammonium Thiomolybdates, Topics in Catalysis, 54 (2011) 561-567.
[179] W. Trakarnpruk, B. Seentrakoon, Hydrodesulfurization activity of MoS2 and
bimetallic catalysts prepared by in situ decomposition of thiosalt, Industrial & Engineering
Chemistry Research, 46 (2007) 1874-1882.
[180] O.Y. Gutiérrez, T. Klimova, Effect of the support on the high activity of the (Ni)
Mo/ZrO< sub> 2</sub>–SBA-15 catalyst in the simultaneous hydrodesulfurization of
DBT and 4, 6-DMDBT, Journal of Catalysis, 281 (2011) 50-62.
[181] Z.-D. Huang, W. Bensch, L. Kienle, S. Fuentes, G. Alonso, C. Ornelas, SBA-15 as
support for MoS2 and Co-MoS2 Catalysts Derived from Thiomolybdate Complexes in the
reaction of HDS of DBT, Catalysis letters, 122 (2008) 57-67.
[182] T. Todorova, R. Prins, T. Weber, A density functional theory study of the
hydrogenolysis reaction of CH 3 SH to CH 4 on the catalytically active (100) edge of 2H-
MoS 2, Journal of Catalysis, 236 (2005) 190-204.
[183] D. Valencia, T. Klimova, Citric acid loading for MoS 2-based catalysts supported on
SBA-15. New catalytic materials with high hydrogenolysis ability in hydrodesulfurization,
Applied Catalysis B: Environmental, 129 (2013) 137-145.
[184] E.O. Orozco, M. Vrinat, Kinetics of dibenzothiophene hydrodesulfurization over
MoS 2 supported catalysts: modelization of the H 2 S partial pressure effect, Applied
Catalysis A: General, 170 (1998) 195-206.
[185] H. Wang, R. Prins, Hydrodesulfurization of dibenzothiophene and its hydrogenated
intermediates over sulfided Mo/γ-Al 2 O 3, Journal of Catalysis, 258 (2008) 153-164.
[186] Á. Logadóttir, P.G. Moses, B. Hinnemann, N.-Y. Topsøe, K.G. Knudsen, H. Topsøe,
J.K. Nørskov, A density functional study of inhibition of the HDS hydrogenation pathway
by pyridine, benzene, and H 2 S on MoS 2-based catalysts, Catalysis today, 111 (2006) 44-
51.
[187] S. Helveg, J.V. Lauritsen, E. Lægsgaard, I. Stensgaard, J.K. Nørskov, B. Clausen, H.
Topsøe, F. Besenbacher, Atomic-scale structure of single-layer MoS 2 nanoclusters,
Physical review letters, 84 (2000) 951.
[188] J.-F. Paul, E. Payen, Vacancy formation on MoS2 hydrodesulfurization catalyst:
DFT study of the mechanism, The Journal of Physical Chemistry B, 107 (2003) 4057-4064.
[189] M. Sun, A.E. Nelson, J. Adjaye, Adsorption and dissociation of H 2 and H 2 S on
MoS 2 and NiMoS catalysts, Catalysis today, 105 (2005) 36-43.
[190] S. Cristol, J. Paul, E. Payen, D. Bougeard, J. Hafner, F. Hutschka, Theoretical study
of benzothiophene hydrodesulfurization on MoS 2, Studies in surface science and catalysis,
127 (1999) 327-334.
[191] B. Hinnemann, P.G. Moses, J.K. Nørskov, Recent density functional studies of
hydrodesulfurization catalysts: insight into structure and mechanism, Journal of Physics:
Condensed Matter, 20 (2008) 064236.
[192] P.G. Moses, B. Hinnemann, H. Topsøe, J.K. Nørskov, The hydrogenation and direct
desulfurization reaction pathway in thiophene hydrodesulfurization over MoS 2 catalysts
at realistic conditions: a density functional study, Journal of Catalysis, 248 (2007) 188-203.
[193] J. Lauritsen, M. Nyberg, J.K. Nørskov, B. Clausen, H. Topsøe, E. Lægsgaard, F.
Besenbacher, Hydrodesulfurization reaction pathways on MoS 2 nanoclusters revealed by
scanning tunneling microscopy, Journal of Catalysis, 224 (2004) 94-106.
[194] J. Leglise, L. Finot, J. Van Gestel, J. Duchet, Conversion of model sulfur compounds
to characterize hydrodesulfurization CoMo/Al 2 O 3 catalysts, Studies in surface science
and catalysis, 127 (1999) 51-65.
[195] M. Sun, A.E. Nelson, J. Adjaye, Ab initio DFT study of hydrogen dissociation on
MoS 2, NiMoS, and CoMoS: mechanism, kinetics, and vibrational frequencies, Journal of
Catalysis, 233 (2005) 411-421.
[196] C. Thomas, L. Vivier, J. Lemberton, S. Kasztelan, G. Pérot, Deuterium Tracer
Studies on Hydrotreating Catalysts—Isotopic Exchange between Hydrogen and Hydrogen
Sulfide on Sulfided NiMo/Al 2 O 3, Journal of Catalysis, 167 (1997) 1-11.
[197] E. Schachtl, E. Kondratieva, O.Y. Gutiérrez, J.A. Lercher, Pathways for H2
Activation on (Ni)-MoS2 Catalysts, The journal of physical chemistry letters, 6 (2015)
2929-2932.
[198] R. Prins, M. Egorova, A. Röthlisberger, Y. Zhao, N. Sivasankar, P. Kukula,
Mechanisms of hydrodesulfurization and hydrodenitrogenation, Catalysis today, 111
(2006) 84-93.
[199] S. Cristol, J.-F. Paul, E. Payen, D. Bougeard, F. Hutschka, S. Clémendot, DBT
derivatives adsorption over molybdenum sulfide catalysts: a theoretical study, Journal of
Catalysis, 224 (2004) 138-147.
[200] J.-F. Paul, S. Cristol, E. Payen, Computational studies of (mixed) sulfide
hydrotreating catalysts, Catalysis Today, 130 (2008) 139-148.
[201] B. Hinnemann, J.K. Nørskov, H. Topsøe, A density functional study of the chemical
differences between type I and type II MoS2-based structures in hydrotreating catalysts,
The Journal of Physical Chemistry B, 109 (2005) 2245-2253.
[202] F. Pedraza, S. Fuentes, Ni–Mo and Ni–W sulfide catalysts prepared by
decomposition of binary thiometallates, Catalysis letters, 65 (2000) 107-113.
[203] K. Inamura, R. Prins, The role of Co in unsupported Co-Mo sulfides in the
hydrodesulfurization of thiophene, Journal of Catalysis, 147 (1994) 515-524.
[204] L.S. Byskov, B. Hammer, J.K. Nørskov, B. Clausen, H. Topsøe, Sulfur bonding in
MoS2 and Co-Mo-S structures, Catalysis letters, 47 (1997) 177-182.
[205] T. Isoda, S. Nagao, X. Ma, Y. Korai, I. Mochida, Catalytic activities of NiMo and
CoMoAl2O3 of variable Ni and Co contents for the hydrodesulfurization of 4, 6-
dimethyldibenzothiophene in the presence of naphthalene, Applied Catalysis A: General,
150 (1997) 1-11.
[206] M. Vrinat, L. De Mourgues, On the role of cobalt in sulfided unsupported Co-Mo
hydrodesulfurization catalysts: Kinetic studies and scanning electron microscopic
observations, Applied Catalysis, 5 (1983) 43-57.
[207] K.C. Pratt, J.V. Sanders, N. Tamp, The role of nickel in the activity of unsupported
Ni Mo hydrodesulfurization catalysts, Journal of Catalysis, 66 (1980) 82-92.
[208] C. Gachet, R. Paulus, L. De Mourgues, C. Durand, H. Toulhoat, Unsupported Ni‐mo
Sulfide Catalysts: Relation Between the HDS and Hydrogenation Activities and Some
Structural Properties, Bulletin des Sociétés Chimiques Belges, 93 (1984) 681-686.
[209] M. Sun, J. Adjaye, A.E. Nelson, Theoretical investigations of the structures and
properties of molybdenum-based sulfide catalysts, Applied Catalysis A: General, 263
(2004) 131-143.
[210] C. Moreau, J. Joffre, C. Saenz, P. Geneste, Hydroprocessing of substituted benzenes
over a sulfided CoO-MoO33/γ-A12O3catalyst, Journal of Catalysis, 122 (1990) 448-451.
[211] J. Van Gestel, C. Dujardin, F. Maugé, J.-C. Duchet, Direct aromatic C–N bond
cleavage evidenced in the hydrodenitrogenation of 2, 6-dimethylaniline over cobalt-
promoted Mo/Al 2 O 3 sulfide catalysts: A reactivity and FT-IR study, Journal of Catalysis,
202 (2001) 78-88.
[212] E. Hensen, P. Kooyman, Y. Van der Meer, A. Van der Kraan, V. De Beer, J. Van
Veen, R. Van Santen, The relation between morphology and hydrotreating activity for
supported MoS 2 particles, Journal of Catalysis, 199 (2001) 224-235.
[213] P. Da Silva, N. Marchal, S. Kasztelan, Influence of high Mo loading on the
HYD/HDS selectivity of alumina supported MoS 2 catalysts, Studies in Surface Science
and Catalysis, 106 (1997) 353-360.
[214] M. Breysse, G. Djega-Mariadassou, S. Pessayre, C. Geantet, M. Vrinat, G. Pérot, M.
Lemaire, Deep desulfurization: reactions, catalysts and technological challenges, Catalysis
Today, 84 (2003) 129-138.
[215] L.P. Nielsen, S.V. Christensen, H. Topsøe, B.S. Clausen, Changes in metal–sulfur
bond energy in promoted and unpromoted molybdenum catalysts, Catalysis letters, 67
(2000) 81-85.
[216] R. Chianelli, Periodic trends transition metal sulfide catalysis: intuition and theory,
Oil & Gas Science and Technology-Revue de l'IFP, 61 (2006) 503-513.
[217] J.V. Lauritsen, M. Nyberg, R.T. Vang, M. Bollinger, B. Clausen, H. Topsøe, K.W.
Jacobsen, E. Lægsgaard, J. Nørskov, F. Besenbacher, Chemistry of one-dimensional
metallic edge states in MoS2 nanoclusters, Nanotechnology, 14 (2003) 385.
[218] E. Krebs, B. Silvi, P. Raybaud, Mixed sites and promoter segregation: A DFT study
of the manifestation of Le Chatelier's principle for the Co (Ni) MoS active phase in reaction
conditions, Catalysis Today, 130 (2008) 160-169.
[219] C. Song, X. Ma, New design approaches to ultra-clean diesel fuels by deep
desulfurization and deep dearomatization, Applied Catalysis B: Environmental, 41 (2003)
207-238.
[220] T. Zepeda, B. Pawelec, J. Fierro, A. Olivas, S. Fuentes, T. Halachev, Effect of Al
and Ti content in HMS material on the catalytic activity of NiMo and CoMo hydrotreating
catalysts in the HDS of DBT, Microporous and Mesoporous Materials, 111 (2008) 157-
170.
[221] N. Kagami, B.M. Vogelaar, A.D. van Langeveld, J.A. Moulijn, Reaction pathways
on NiMo/Al 2 O 3 catalysts for hydrodesulfurization of diesel fuel, Applied Catalysis A:
General, 293 (2005) 11-23.
[222] Z. Huang, W. Bensch, A. Lotnyk, L. Kienle, S. Fuentes, J. Bocarando, G. Alonso, C.
Ornelas, SBA-15 as support for NiMo HDS catalysts derived from sulfur-containing
molybdenum and nickel complexes: Effect of activation mode, Journal of Molecular
Catalysis A: Chemical, 323 (2010) 45-51.
[223] P. Nikulshin, V. Salnikov, A. Mozhaev, P. Minaev, V. Kogan, A. Pimerzin,
Relationship between active phase morphology and catalytic properties of the carbon–
alumina-supported Co (Ni) Mo catalysts in HDS and HYD reactions, Journal of Catalysis,
309 (2014) 386-396.
[224] E.M. Fernández, P.G. Moses, A. Toftelund, H.A. Hansen, J.I. Martínez, F. Abild‐
Pedersen, J. Kleis, B. Hinnemann, J. Rossmeisl, T. Bligaard, Scaling relationships for
adsorption energies on transition metal oxide, sulfide, and nitride surfaces, Angewandte
Chemie, 120 (2008) 4761-4764.
[225] T. Kabe, Y. Aoyama, D. Wang, A. Ishihara, W. Qian, M. Hosoya, Q. Zhang, Effects
of H 2 S on hydrodesulfurization of dibenzothiophene and 4, 6-dimethyldibenzothiophene
on alumina-supported NiMo and NiW catalysts, Applied Catalysis A: General, 209 (2001)
237-247.
[226] A. Daudin, A.-F. Lamic, G. Perot, S. Brunet, P. Raybaud, C. Bouchy, Microkinetic
interpretation of HDS/HYDO selectivity of the transformation of a model FCC gasoline
over transition metal sulfides, Catalysis Today, 130 (2008) 221-230.
[227] E. Krebs, B. Silvi, A. Daudin, P. Raybaud, A DFT study of the origin of the
HDS/HydO selectivity on Co (Ni) MoS active phases, Journal of Catalysis, 260 (2008)
276-287.
[228] J. Lauritsen, S. Helveg, E. Lægsgaard, I. Stensgaard, B. Clausen, H. Topsøe, F.
Besenbacher, Atomic-scale structure of Co–Mo–S nanoclusters in hydrotreating catalysts,
Journal of Catalysis, 197 (2001) 1-5.
[229] Y.-S. Chang, S. Savitha, S. Sadhasivam, C.-K. Hsu, F.-H. Lin, Fabrication,
characterization, and application of greigite nanoparticles for cancer hyperthermia, Journal
of colloid and interface science, 363 (2011) 314-319.
[230] F. Cao, W. Hu, L. Zhou, W. Shi, S. Song, Y. Lei, S. Wang, H. Zhang, 3D Fe 3 S 4
flower-like microspheres: high-yield synthesis via a biomolecule-assisted solution
approach, their electrical, magnetic and electrochemical hydrogen storage properties,
Dalton Transactions, (2009) 9246-9252.
[231] M. Feng, Y. Lu, Y. Yang, M. Zhang, Y.-J. Xu, H.-L. Gao, L. Dong, W.-P. Xu, S.-H.
Yu, Bioinspired greigite magnetic nanocrystals: chemical synthesis and biomedicine
applications, Scientific reports, 3 (2013).
[232] X. Wang, W. Cai, G. Wang, Z. Wu, H. Zhao, One-step fabrication of high
performance micro/nanostructured Fe 3 S 4–C magnetic adsorbent with easy recovery and
regeneration properties, CrystEngComm, 15 (2013) 2956-2965.
[233] Y. Zhang, Y. Du, H. Xu, Q. Wang, Diverse-shaped iron sulfide nanostructures
synthesized from a single source precursor approach, CrystEngComm, 12 (2010) 3658-
3663.
[234] L. Kong, L. Yan, Z. Qu, N. Yan, L. Li, β-Cyclodextrin stabilized magnetic Fe 3 S 4
nanoparticles for efficient removal of Pb (ii), Journal of Materials Chemistry A, 3 (2015)
15755-15763.
[235] P. Li, C. Xia, Q. Zhang, Z. Guo, W. Cui, H. Bai, H.N. Alshareef, X.-x. Zhang,
Fabrication and characterization of nanostructured Fe3S4, an isostructural compound of
half-metallic Fe3O4, Journal of Applied Physics, 117 (2015) 223903.
[236] S.H. Yang, C.N. Satterfield, Some effects of sulfiding of a NiMoAl2O3 catalyst on
its activity for hydrodenitrogenation of quinoline, Journal of Catalysis, 81 (1983) 168-178.
[237] H. Topsøe, B.S. Clausen, R. Candia, C. Wivel, S. Mørup, In situ Mössbauer emission
spectroscopy studies of unsupported and supported sulfided Co-Mo hydrodesulfurization
catalysts: Evidence for and nature of a Co-Mo-S phase, Journal of Catalysis, 68 (1981)
433-452.
[238] M. Breysse, E. Furimsky, S. Kasztelan, M. Lacroix, G. Perot, Hydrogen activation
by transition metal sulfides, Catalysis Reviews, 44 (2002) 651-735.
[239] B. Yoosuk, J.H. Kim, C. Song, C. Ngamcharussrivichai, P. Prasassarakich, Highly
active MoS 2, CoMoS 2 and NiMoS 2 unsupported catalysts prepared by hydrothermal
synthesis for hydrodesulfurization of 4, 6-dimethyldibenzothiophene, Catalysis Today, 130
(2008) 14-23.
[240] M. Landau, Deep hydrotreating of middle distillates from crude and shale oils,
Catalysis Today, 36 (1997) 393-429.
[241] T. Kabe, A. Ishihara, Q. Zhang, Deep desulfurization of light oil. Part 2:
hydrodesulfurization of dibenzothiophene, 4-methyldibenzothiophene and 4, 6-
dimethyldibenzothiophene, Applied Catalysis A: General, 97 (1993) L1-L9.
[242] M.R. de Brimont, C. Dupont, A. Daudin, C. Geantet, P. Raybaud, Deoxygenation
mechanisms on Ni-promoted MoS 2 bulk catalysts: a combined experimental and
theoretical study, Journal of catalysis, 286 (2012) 153-164.
[243] K.A. Rogers, Y. Zheng, Selective Deoxygenation of Biomass‐Derived Bio‐oils
within Hydrogen‐Modest Environments: A Review and New Insights, ChemSusChem, 9
(2016) 1750-1772.
[244] H. Zhang, H. Lin, W. Wang, Y. Zheng, P. Hu, Hydroprocessing of waste cooking oil
over a dispersed nano catalyst: Kinetics study and temperature effect, Applied Catalysis B:
Environmental, 150 (2014) 238-248.
[245] S. Brillouet, E. Baltag, S. Brunet, F. Richard, Deoxygenation of decanoic acid and
its main intermediates over unpromoted and promoted sulfided catalysts, Applied Catalysis
B: Environmental, 148 (2014) 201-211.
[246] C. Dupont, R. Lemeur, A. Daudin, P. Raybaud, Hydrodeoxygenation pathways
catalyzed by MoS 2 and NiMoS active phases: a DFT study, Journal of catalysis, 279
(2011) 276-286.
[247] M. Philippe, F. Richard, D. Hudebine, S. Brunet, Transformation of
dibenzothiophenes model molecules over CoMoP/Al 2 O 3 catalyst in the presence of
oxygenated compounds, Applied Catalysis B: Environmental, 132 (2013) 493-498.
[248] M. Badawi, L. Vivier, G. Pérot, D. Duprez, Promoting effect of cobalt and nickel on
the activity of hydrotreating catalysts in hydrogenation and isomerization of olefins,
Journal of Molecular Catalysis A: Chemical, 293 (2008) 53-58.
[249] A. Travert, H. Nakamura, R.A. van Santen, S. Cristol, J.-F. Paul, E. Payen, Hydrogen
activation on Mo-based sulfide catalysts, a periodic DFT study, Journal of the American
Chemical Society, 124 (2002) 7084-7095.
[250] P.-Y. Prodhomme, P. Raybaud, H. Toulhoat, Free-energy profiles along reduction
pathways of MoS 2 M-edge and S-edge by dihydrogen: a first-principles study, Journal of
catalysis, 280 (2011) 178-195.
[251] J. Mijoin, V. Thévenin, N.G. Aguirre, H. Yuze, J. Wang, W. Li, G. Pérot, J.
Lemberton, Thioreduction of cyclopentanone and hydrodesulfurization of
dibenzothiophene over sulfided nickel or cobalt-promoted molybdenum on alumina
catalysts, Applied Catalysis A: General, 180 (1999) 95-104.
[252] P. Raybaud, Understanding and predicting improved sulfide catalysts: Insights from
first principles modeling, Applied Catalysis A: General, 322 (2007) 76-91.
[253] V. Kogan, P. Nikulshin, On the dynamic model of promoted molybdenum sulfide
catalysts, Catalysis Today, 149 (2010) 224-231.
Curriculum Vitae
Candidate’s full name: Seyyedmajid Sharifvaghefi
Universities attended (with dates and degrees obtained): Master of Science in Innovative
and Sustainable Chemical Engineering, Chalmers University of Technology, Gothenburg,
Sweden, 2008-2011.
Bachelor of Science in Chemical Engineering, Sharif University of Technology, Tehran,
Iran, 2003-2008.
Publications:
1. H. Harelind, F. Gunnarsson, S.M Sharifvaghefi, M. Skoglundh, P.-A. Carlsson,
Influence of the Carbon–Carbon Bond Order and Silver Loading on the Formation of
Surface Species and Gas Phase Oxidation Products in Absence and Presence of NOX over
Silver-Alumina Catalysts, ACS Catalysis, 2 (2012) 1615-1623.
2. S. Sharifvaghefi, Y. Zheng, Development of a Magnetically Recyclable Molybdenum
Disulfide Catalyst for Direct Hydrodesulfurization, ChemCatChem, 7 (2015) 3397-3403.
3. S. Sharifvaghefi, Y. Zheng, New Insights on the roles of H2 and H2S and the involved
active sites of MoS2 in hydrodesulfurization of DBT. Submitted.
4. S. Sharifvaghefi, Y. Zheng, A novel dispersed catalyst with magnetic greigite as a core:
Recyclable catalysts for hydrodesulfurization. Submitted.
5. S. Sharifvaghefi, Y. Zheng, Edge activation of H2 and H2S over NiMoS catalyst and the
effect of Ni loading. To be submitted.
6. S. Sharifvaghefi, Y. Zheng, Hydrodeoxygenation of Stearic acid with unpromoted and
Ni(Co)-promoted MoS2 supported on magnetic greigite. To be submitted.
Conference Presentations:
1. Development of a novel magnetically recyclable nano-MoS2 catalyst for
hydrodesulfurization, 24th North American Catalysis Society Meeting, Jun 14-19, 2015,
Pittsburgh, USA
2. Hydrodesulfurization of DBT over MoS2 catalyst: H2 and H2S activation pathways and
the involved active sites, 24th Canadian Symposium on Catalysis, May 8-11, 2016,
University of Ottawa, Ottawa, Ontario, Canada