You may notice some images loading slow across the Open Collections website. Thank you for your patience as we rebuild the cache to make images load faster.

Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hydrodesulphurization of dibenzothiophene using carbon supported NiMoS catalysts Alamoudi, Majed 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2016_may_alamoudi_majed.pdf [ 2.84MB ]
Metadata
JSON: 24-1.0228785.json
JSON-LD: 24-1.0228785-ld.json
RDF/XML (Pretty): 24-1.0228785-rdf.xml
RDF/JSON: 24-1.0228785-rdf.json
Turtle: 24-1.0228785-turtle.txt
N-Triples: 24-1.0228785-rdf-ntriples.txt
Original Record: 24-1.0228785-source.json
Full Text
24-1.0228785-fulltext.txt
Citation
24-1.0228785.ris

Full Text

i  Hydrodesulphurization of dibenzothiophene using carbon supported NiMoS catalysts  by  Majed Alamoudi   B.Sc., King Abdulaziz University, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (CHEMICAL AND BIOLOGICAL ENGINEERING)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   April 2016  © Majed Alamoudi, 2016  ii  Abstract Hydrodesulphurization (HDS) is the major process used to remove S from crude oil feedstocks in order to improve fuel quality and meet environmental regulations. The goal of this study was to determine if petroleum coke (petcoke), derived from Alberta oilsands, could be converted into a useful catalyst support. Hence the HDS activity and selectivity of nickel molybdenum sulfided catalysts supported on activated carbon (NiMoS/AC), petroleum coke (NiMoS/PC) and conventional alumina (NiMo/-Al2O3) have been compared using dibenzothiophene (DBT) as a model reactant. The reactions were carried out in a novel slurry-phase batch microreactor at different reaction times (30-120 min) and temperatures (588-638 K) and a fixed H2 pressure (4.8 MPa).  The results showed that NiMoS/PC had higher activity towards the HDS of DBT when compared with NiMoS/AC, although the catalysts had very similar product selectivities. The highest activity for DBT HDS, corresponding to 90% DBT conversion, occurred at 638 K for the NiMoS/AC catalyst and at 623K for the NiMoS/PC catalyst. The reaction proceeded by two pathways: the direct desulphurization (DDS) reaction route and the hydrogenation (HYD) reaction route. The power law pseudo 1st-order kinetic model was applied to the HDS of DBT. The estimated kinetic parameters showed similar magnitudes for the HYD versus the DDS routes over both catalysts, whereas the DDS pathway had higher apparent activation energy compared to the HYD route for both catalysts.    iii  Preface All the work reported in this thesis was performed in the Department of Chemical and Biological Engineering at the University of British Colombia, Vancouver campus. I, Majed Alamoudi, was the investigator of this research and responsible for preparing the catalysts, running the microreactor, obtaining and discussing the results, and stating the major findings as well as constructing this thesis. Kevin J Smith supervised the research, providing guidance, instructions, approving the work, and editing this thesis. Ross Kukard commissioned the novel slurry-phase batch microreactor described in section 2.3 which was used to carry out the experiments and collect the data reported in section 3.3. iv  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables .............................................................................................................................. viii List of Figures .................................................................................................................................x Nomenclature ............................................................................................................................. xvi List of Abbreviations ................................................................................................................ xvii Acknowledgements ......................................................................................................................xx Chapter 1: Introduction ................................................................................................................1 1.1 Background ..................................................................................................................... 1 1.2 Hydrodesulphurization Process ...................................................................................... 3 1.3 Canadian Oil Sands, Petroleum Coke and Activated Carbon: ........................................ 4 1.4 Literature Review............................................................................................................ 7 1.4.1 Comparison of Carbon Supported Catalysts with other Supports .............................. 7 1.4.2 Reaction Kinetics ...................................................................................................... 17 1.5 Literature Review Summary ......................................................................................... 28 1.6 Study Objectives ........................................................................................................... 29 Chapter 2: Experimental .............................................................................................................30 2.1 Catalysts Preparation .................................................................................................... 30 2.2 Catalyst Characterization. ............................................................................................. 31 2.3 Catalyst Activity Measurements ................................................................................... 32 Chapter 3: Results and Discussion .............................................................................................36 v  3.1 Catalyst Characterization .............................................................................................. 36 3.2 Catalyst Characterization Summary ............................................................................. 43 3.3 Catalytic Activity .......................................................................................................... 43 3.3.1 Preliminary Studies to Test the Activity of Different Catalysts ............................... 43 3.3.2 Activity of the NiMoS/AC Catalyst for the HDS of DBT ........................................ 46 3.3.3 Activity of NiMoS/PC Catalyst for HDS of DBT .................................................... 47 3.3.4 Comparison Between the Activity of NiMoS/AC and NiMoS/PC ........................... 48 3.4 Product Distribution ...................................................................................................... 50 3.4.1 Product Distribution over the NiMoS/AC Catalyst .................................................. 50 3.4.2 Product Distribution over NiMoS/PC ....................................................................... 53 3.4.3 Comparison Between the Selectivity of NiMoS/AC and NiMoS/PC ....................... 55 3.5 Summary of Findings .................................................................................................... 57 Chapter 4: HDS Reaction Kinetics of DBT over NiMoS/AC and NiMoS/PC ........................58 4.1 Reaction Mechanism ..................................................................................................... 58 4.2 Kinetic Development .................................................................................................... 59 4.3 Mole Balance ................................................................................................................ 60 4.4 Parameter Estimation .................................................................................................... 61 Chapter 5: Conclusion and Recommendations .........................................................................75 5.1 Conclusion .................................................................................................................... 75 5.2 Recommendations ......................................................................................................... 76 Bibliography .................................................................................................................................77 Appendices ....................................................................................................................................83 Appendix A Calculation of Activation Energies from the Data of Liu et al.24 ......................... 84 vi  Appendix B The Kinetic Model for HDS in Vanrysselberghe and Froment33 ......................... 89 Appendix C Catalyst Characterization...................................................................................... 92 C.1 BET Surface Area Calculations ................................................................................ 92 C.2 BET Analysis for Activated Carbon Before Impregnation ....................................... 93 C.3 for Activated Carbon After Impregnation ................................................................. 95 C.4 BET Analysis for Petcoke Before Impregnation .................................................... 100 C.5 BET Analysis for Petcoke After Impregnation ....................................................... 103 Appendix D C-MS Sample Scan ............................................................................................ 107 D.1 Example of Sample Calculation.............................................................................. 110 Appendix E Repeatability and Analysis Error ........................................................................ 112 E.1 Conversion Calculation ........................................................................................... 112 E.2 Selectivity Calculations .......................................................................................... 113 E.3 Conversion Repeatability ........................................................................................ 113 E.4 Product Selectivity Repeatability ............................................................................ 115 Appendix F Thermal Experiments Data ................................................................................. 116 Appendix G Matlab Codes...................................................................................................... 119 G.1 Main Body Code ..................................................................................................... 119 G.2 Modelmulti Code .................................................................................................... 122 G.3 ODE Codes ............................................................................................................. 122 G.4 Jacobian Matrix Calculation ................................................................................... 123 G.5 Least Square Codes ................................................................................................. 123 Appendix H Calculations of Arrhenius Equations.................................................................. 131 Appendix I Thermodynamic Calculations to Determine the Reaction Phase. ........................ 132 vii  I.1 Reaction Phase Determination ................................................................................ 132 I.2 H2S Partial Pressure Calculation ............................................................................. 134   viii  List of Tables Table 1: Gasoline Sulfur Standards for the USA1 .......................................................................... 2 Table 2: Product yield from Delayed coke and Fluid coke processing of Alberta bitumen9.......... 7 Table 3: Rate constants of the total, DDS and HYD (desulfurization) of DBT and of the desulfurization of tetrahydrodibenzothiophene over NiMo//-Al2O3, CoMo//-Al2O3, and CoMo//-Al2O3 catalysts at different initial partial pressures of H2S and a fixed temperature of 340 oC. Adopted from [22] Copyright © 2004, Elsevier 22 ........................................................... 21 Table 4: Adsorption equilibrium constants and rate coefficients at 573K. Adopted from [33] Copyright © 1996 American Chemical Society33 ........................................................................ 26 Table 5: Comparison of activation energies for HDS of DBT from several literature studies ..... 27 Table 6: BET surface area, pore volume, and pore size for the catalysts and their supports ....... 37 Table 7: Surface composition as defined by XPS of carbon supports NiMoS ............................. 39 Table 8: Chemical states of Mo on activated carbon support from Mo 3d XPS narrow scan ...... 39 Table 9: Chemical states of Mo on petcoke support from Mo 3d XPS narrow scan .................... 39 Table 10: Chemical states of S on activated carbon support from S 2p XPS narrow scan .......... 40 Table 11: Chemical states of S on petcoke support from S 2p XPS narrow scan ........................ 40 Table 12: Estimated reaction rate constants for the thermal reaction of HDS of DBT ................ 66 Table 13: The pre-exponential factors Aj and activation energies for all k1 at the thermal reaction....................................................................................................................................................... 67 Table 14: Estimated reaction rate constants for the HDS of DBT over NiMoS/AC at tested temperature ................................................................................................................................... 68 Table 15: The pre-exponential factors Aj and activation energies for all  kj'  at NiMoS/AC catalysts ......................................................................................................................................... 69 ix  Table 16: Estimated catalytic reaction rate constants for reaction of HDS for DBT over NiMoS/PC at tested temperature .................................................................................................. 70 Table 17: The pre-exponential factors  Aj and activation energies for all kj'  at NiMoS/PC catalysts ......................................................................................................................................... 71 Table 18: Comparison of DBT rates for the total DBT conversion at 623 K and 4.8 MPa with PH2S= 100 kPa with literature data ................................................................................................ 71 Table 19: Comparison of rate of reaction for DBT total, DDS, HYD conversion at 623 K and 4.8 MPa ............................................................................................................................................... 72 Table 20: Comparison of activation energies for the total DBT conversion ................................ 73 Table 21: Conversions and corresponding temperature for NiMoS/γ-Al2O3 ............................... 84 Table 22: Conversions and corresponding temperature for NiMoS/AC ...................................... 85 Table 23: Conversions and corresponding temperature for NiMoS/AAC.................................... 86 Table 24: Conversions and corresponding temperature for NiMoS/AC ...................................... 87 Table 25: Conversions and corresponding temperature for NiMoS/PC ....................................... 88 Table 26 DBT calibration curve calculation ............................................................................... 109 Table 27 : GCMS vail ................................................................................................................. 110 Table 28: Conversions error calculation ..................................................................................... 114 Table 29: Error calculation of biphenyl ...................................................................................... 115 Table 30: Thermal run data ......................................................................................................... 116 Table 31 Aspen result for HDS of DBT at conversion set to zero ............................................. 133 Table 32 Mo, Ni, CS2 molecular weight ..................................................................................... 134 Table 33 Desired metal wt% ....................................................................................................... 134 Table 34 Feed preparation calculation ........................................................................................ 134 x  List of Figures Figure 1: Sulphur content in diesel fuel - trend for all the continents2 ........................................... 2 Figure 2: Schematic diagram of a typical HDS unit in a petroleum refinery7 ................................ 3 Figure 3: HDS activites obtained with different catalysts (FAC:fluid coke, DAC:delayed coke,CAC: commercial activated carbon) under 643 K,3.45 MPa, and 2 hours reaction time. Copyright © 2012, Elsevier5 .......................................................................................................... 8 Figure 4: HDN activities obtained with different catalysts (FAC: fluid coke, DAC: delayed coke, CAC: commercial activated carbon) under 643 K,3.45 MPa, and 2 hours reaction time. Copyright © 2012, Elsevier5 .......................................................................................................... 8 Figure 5: The conversion of DBT over presulfided catalysts NiMoS/-Al2O3, NiMoS/AC and NiMoS/AAC as a function of reaction temperature. Copyright © 2011, Elsevier24 .................... 10 Figure 6: The XRD patterns of NiMoS/γ-Al2O3 (1), NiMoS/AAC (2) NiMoS/AC (3), and AAC (4).Copyright © 2011, Elsevier24 ................................................................................................. 11 Figure 7: Activity of Co-Mo/CMC-1, Co-Mo/AC, and commercial Co-Mo/γ-Al2O3 for the HDS of thiophene in a model gasoline. Reaction conditions: 1.5 MPa, LHSV = 2 h-1, H2/feed volume ratio = 300. The model gasoline contained 0.05% sulfur from thiophene and 20% 1-hexene, and the balance was n-heptane. Copyright © 2010, Elsevier6 ............................................................. 13 Figure 8: Activity of Ni-Mo/CMC-2 and the commercial catalyst FH-98for the HDS of  dibenzothiophene (DBT) in a model diesel. Reaction conditions: 3.1 MPa, LHSV = 2 h−1, H2/feed volume ratio = 500. The model diesel contained 0.3% sulfur from DBT, 0.02% N from quinolone 5% and 0.5% n-octane as an internal standard. Copyright © 2010, Elsevier6 ............. 13 Figure 9: Temperature vs. residual sulfur content in HDS reaction of real gas–oil feedstock using down-flow tubular reactor, with LHSV of 1.5 h-1 under hydrogen pressure of 5MPa in volumetric xi  hydrogen/oil ratio of 250 nl l-1. (_ ) Over the best Ni-Mo/Activated carbon catalyst (k at 330 oC = 1.59 h−1), (_ ) over the conventional Ni-Mo/γ-Al2O3 catalyst (k at 330◦C = 0.68 h-1) Copyright © 2004, Elsevier23. ............................................................................................................................ 15 Figure10: Effect of catalyst amount on the hydrogenation conversion of 1-methylnaphthalene with NiMo/KB or NiMo/γ-Al2O3. Reaction conditions: 1-Methylnaphthalene/decalin= 1/9; reaction temperature 380 oC; reaction pressure 9.5 MPa; reaction time 40min; catalyst NiMo/KB or NiM/γ-Al2O3, 1-5wt% addition to 1-Methylnaphthalene.Adapted from [26].Copyright © 1995, American Chemical Society 26 ............................................................................................ 17 Figure 11: Relative partial pressures of the products in the HDS of dibenzothiophene at 340 oC and 35 kPa H2S over Mo/- Al2O3 as function of weight time.Copyright © 2004, Elsevier 22 .... 18 Figure 12: Relative partial pressures of the products in the HDS of dibenzothiophene at 340 oCand 35 kPa H2S over CoMo/-Al2O3 as function of weight time. Copyright © 2004,Elsevier 22....................................................................................................................................................... 18 Figure 13: Proposed reaction pathway for HDS of DBT. Copyright © 2004, Elsevier 22 ........... 18 Figure 14: Rim/edge model of MoS2 particle adapted from [31]. Copyright © 1994Academic Press31. .......................................................................................................................................... 20 Figure 15: Pseudo first-order rate constants of the total conversion of DBT over CoMo/γ-Al2O3, NiMo/γ-Al2O3, and Mo/γ-Al2O3 catalysts at different partial pressures of H2S and a fixed temperature of 340 oC. Copyright © 2004, Elsevier22 .................................................................. 22 Figure 16: Conversion as function of space time (  ) total conversion of DBT,(    ) conversion of DBT into BP,(     ) conversion of DBT into CHB. Reaction conditions=553K, Pt=60 bar, H2/CH4=6.39.Copyright © 1996 American Chemical Society33 .................................................. 24      xii  Figure 17: Total conversion of DBT as a function of space time at various temperature:  (  ) 513, (    )533, (   ) 553,and (  ) 573K under Pt=80 bar H2/CH4=6.38, and H2/HC=1.33.and H2/HC=1.10 reaction conditions. Copyright © 1996 American Chemical Society33 ........................................ 25 Figure 18: Proposed reaction network, note that here BPH refers to biphenyl (BP)). Copyright © 1996 American Chemical Society33. ............................................................................................. 25 Figure 19: Process flow diagram of a novel slurry-phase batch hydroconversion micro-reactor. Copyright ©2015 American Chemical Society.41 ........................................................................ 35 Figure 20: XRD pattrens for AC, NiMoS/AC, NiMoS/PC .......................................................... 38 Figure 21: Narrow scan with peak deconvolution for Mo 3d on activated carbon....................... 41 Figure 22: Narrow scan with peak deconvolution for S 2P on activated carbon .......................... 41 Figure 23: Narrow scan with peak deconvolution for Mo 3d on petcoke .................................... 42 Figure 24: Narrow scan with peak deconvolution for S 2P on Petcoke ....................................... 42 Figure 25: Comparison of DBT conversion between catalysts, NiMoS/γ-Al2O3, NiMoS/AC, NiMoS/PC and thermal reaction at 623 K and 2000RPM at different times................................ 44 Figure 26(a): Selectivity of NiMoS/γ-Al2O3 catalysts at 623 K and 2000RPM at different time 44 Figure 26(b): Selectivity of NiMoS/ PC catalysts at 623 K and 2000RPM at different time……45 Figure 26(c): Selectivity of NiMoS/AC catalyst at 623 K and 2000RPM at different time……..45 Figure 26(d): Selectivity of the thermal reaction at 623 K and 2000RPM at different time…….46 Figure 27: Effect of reaction temperature on the DBT conversion with respect to time over NiMoS/AC .................................................................................................................................... 47 Figure 28: Effect of reaction temperature on the DBT conversion with respect to time over NiMoS/PC ..................................................................................................................................... 48      xiii  Figure 29: Comparison between NiMoS/AC and NiMoS/PC for HDS of DBT at 603 K and 4.8 MPa reaction condition ................................................................................................................. 49 Figure 30: Comparison between NiMoS/AC and NiMoS/PC for HDS of DBT at 623 K and 4.8 MPa reaction condition ................................................................................................................. 49 Figure 31: Selectivity of the product in HDS of DBT at 588K over NiMoS/AC as function of time ............................................................................................................................................... 51 Figure 32: Selectivity of the product in HDS of DBT at 603K over NiMoS/AC as function of time ............................................................................................................................................... 51 Figure 33: Selectivity of the product in HDS of DBT at 623K over NiMoS/AC as function of time ............................................................................................................................................... 52 Figure 34: Selectivity of the product in HDS of DBT at 638K over NiMoS/AC as function of time ............................................................................................................................................... 52 Figure 35: Selectivity of the product in HDS of DBT at 588 K over NiMoS/PC as function of time ............................................................................................................................................... 53 Figure 36: Selectivity of the product in HDS of DBT at 603 K over NiMoS/PC as function of time. .............................................................................................................................................. 54 Figure 37: Selectivity of the product in HDS of DBT at 623 K over NiMoS/PC as function of time. .............................................................................................................................................. 54 Figure 38: Selectivity of the product in HDS of DBT at 638 K over NiMoS/PC as function of time. .............................................................................................................................................. 55 Figure 39: Comparison between NiMoS/AC and NiMoS/PC in term of selectivity towards DDS route at 603 K................................................................................................................................ 56 xiv  Figure 40: Comparison between NiMoS/AC and NiMoS/PC in term of selectivity towards DDS route at 623 K................................................................................................................................ 56 Figure 41: Proposed reaction pathway of HDS of DBT. Copyright © 2006, Springer Science Business Media, Inc.43 .................................................................................................................. 59 Figure 42: Measured (points) and model predicted (line) concentrations as function of time for NiMoS/AC catalyst at different temperatures............................................................................... 64 Figure 43: Measured (points) and model predicted (line) concentrations as function of time for NiMoS/PC catalyst at different temperatures. .............................................................................. 65 Figure 44: Arrhenius plot of ln (kj) versus(1000/T) for the thermal HDS of DBT ...................... 66 Figure 45: Arrhenius plot of ln (kj' ) versus (1000/T) for all reaction temperature using NiMoS/AC catalyst. ...................................................................................................................... 68 Figure 46: The Arrhenius plot of ln (kj' ) versus (1000/T) for all reaction temperature using NiMoS/PC catalyst........................................................................................................................ 70 Figure 47: Arrhenius plot to calculate Ea from Liu et al. work on NiMo/AC .............................. 73 Figure 48: Arrhenius plot to calculate Ea for NiMoS/γ-Al2O3 catalyst ........................................ 84 Figure 49: Arrhenius plot to calculate Ea for NiMoS/AC catalyst ............................................... 85 Figure 50: Arrhenius plot to calculate Ea for NiMoS/AAC catalyst ............................................ 86 Figure 51: Arrhenius plot to calculate Ea for NiMoS/AC catalyst for the current study ............. 87 Figure 52: Arrhenius plot to calculate Ea for NiMoS/PC catalyst for the current study .............. 88 Figure 53: Isotherm Linear Plot .................................................................................................... 95 Figure 54: Isotherm Linear Plot .................................................................................................... 99 Figure 55: Isotherm Linear Plot .................................................................................................. 106 xv  Figure 56: mass spect for a sample collected at 90 min for the HDS of DBT reaction at T=638 and P= 4.8 MPa ........................................................................................................................... 107 Figure 57: DBT calibration curve ............................................................................................... 110 Figure 58: Comparison of different temperature amd reaction tmes for the thermal runs of HDS of DBT reaction .......................................................................................................................... 116 Figure 59: Selectivity of the thermal reaction at 588 K and 2000 RPM at different time .......... 117 Figure 60: Selectivity of the thermal reaction at 603 K and 2000 RPM at different time .......... 117 Figure 61: Selectivity of the thermal reaction at 623 K and 2000 RPM at different time .......... 118 Figure 62: Selectivity of the thermal reaction at 638 K and 2000 RPM at different time .......... 118 Figure 63: Arrhenius plot for NiMoS/AC for k1 at different temperature ................................. 131 Figure 64: HDS of DBT reaction flowchart ............................................................................... 132  xvi  Nomenclature Aj Pre-exponential factor of reaction, 1/s or cm3/(gcats) aj Catalyst deactivation factor of reaction j Ci Concentration of species i, mol/L Eaj Activation energy of reaction j, kJ/mol ΔH Enthalpy of reaction, kJ/mol k Rate constant, 1/s 𝑘𝑗′                  Catalytic rate constant, cm3/(gcats) R2 Degree of explanation v Volumetric flow rate, m3/min V Volume, m3 Greek  Ωj  Catalyst effectiveness factor of reaction j ρ  Density, kg/m3  xvii  List of Abbreviations  AAC  Alumina-activated carbon AC  Activated carbon BET  Brunauer-Emmett-Teller BP  Biphenyl  BCH  Bicyclohexyl  CAC  Commercial active carbon  CHB  Cyclohexylbenzene  CMC  Mesoporous carbon DAC  Delayed coke DBT  Dibenzothiophene DMDBT  Dimethyldibenzothiophene DDS  Direct desulphurization  EXAFS  Extended X-ray absorption fine structure  HC  Hydrocarbon HDS  Hydrodesulphurization  HDN  Hydrodenitrogenation  HYD  Hydrogenation  HVGO  Heavy vacuum gas oil FAC  Fluid coke GC  Gas chromatograph H2S  Hydrogen sulfide gas KB  KetjenBlack xviii  KOH   Potassium hydroxide LHSV   Liquid hourly space velocity L-H   Langmuir-Hinshelwood MS   Mass spectroscopy NiMo/-Al2O3 Nickel molybdenum supported on alumina NiMo/AC     Nickel molybdenum supported on active carbon NiMo/AAC  Nickel moly supported on alumina-active carbon NTA  Nitrilotriacetic  Petcoke           Petroleum coke S  Sulphur  THDBT  Tetrahydrodibenzothiophene XRD   X-ray diffraction  XPS                X-ray photoelectron spectroscopy                   xix   Subscripts  exp   Experimental  pred   Model predicted  j   Reaction j  i   Chemical species i  cat    Catalyst  tot    Total   xx  Acknowledgements I would like to express my sincere gratitude to Prof. Kevin Smith for his tremendous and continued support and guidance throughout this project and for sharing his knowledge with me. Prof. Smith truly expanded my knowledge in the area of catalysis and in scientific research in general.  I also would like to pay very special gratitude to my loving and affectionate parents and to my grandfather for their continuous support and for being on my side through their heart-felt words at the times of difficulty despite being thousands of miles away from me. It is their support that made uncertainty and challenges look easy.    In addition, I would like to thank Mr. Ali Alzaid, Dr. Ross Kukard, and the rest of the catalysis group for sharing their knowledge and expertise with me and for their guidance and their time in teaching and familiarizing me with the equipment that I used throughout my research.  Lastly, I would like to thank the Royal Saudi Government for the scholarship that they have granted me for the opportunity to study at UBC and explore another side of the world. I would also like to thank NSERC and Royal Dutch Shell for the funding they have provided for this project.  Without the help of the aforementioned people and parties, this work would have not been possible.    1  Chapter 1: Introduction 1.1 Background The demand for cleaner fuels in the transportation sector is growing rapidly. At the same time, the supply of aromatic, heavier and more contaminated crude oil is also increasing.  The U.S. EPA is proposing that the S content of gasoline be limited to < 10 part per million (ppm) on a yearly average basis by 1 January 2017, down from the present 30 ppm. Furthermore, the  EPA is considering  whether  to either keep the present 80-ppm refinery gate and 95-ppm downstream S caps or reduce them to 50 and 65 ppm, respectively as described in Table 11. Moreover, according to Figure 1, the current S content in diesel fuel in North America is 20 ppm, and this is expected to decline to 10 ppm by 20252. Therefore, for health and environmental reasons, the refractory species containing S must be removed from the crude oil. An improvement in existing catalyst hydrotreating technology to achieve the removal of S and N from heavier refractory components is needed.3 Hydrodesulphurization (HDS) and hydrodenitrogenation (HDN) are the major processes used to remove S and N, respectively, from oil feed stocks. They are both categorized as a hydrotreating process, defined as a catalytic chemical process used to remove impurities from natural gas, gasoline, or oil feedstock by reacting it with hydrogen. Several studies have indicated that S removal from refractory components such as dibenzothiophene (DBT) and 4,6- dimethyldibenzothiophene (4,6-DMDBT) is difficult to achieve4 because of the C-S  bond strength and steric hindrance effects in the case of 4,6-DMDBT. Many studies conducted to address this issue, show that high activity for HDS can be achieved using sulfided nickel and molybdenum (NiMo) supported on activated carbon or pretreated petcoke3,5,6, but these 2  measurements were made at the laboratory scale, whereas the catalysts may deactivate over extended operation times which has a significant impact on the overall cost of operations.   Table 1: Gasoline Sulfur Standards for the USA1 Standard Cap Option 1 Cap Option 2 Limit Effective Limit Effective Refinery annual average standard 10 ppm 1 Jan 2017 10 ppm 1 Jan 2017 Refinery gate per-gallon cap 80 ppm Already 50 ppm 1 Jan 2020 Downstream per-gallon cap 95 ppm Already 65 ppm 1 March 2020     Figure 1: Sulphur content in diesel fuel - trend for all the continents2    3  1.2 Hydrodesulphurization Process In industry, hydrodesulphurization reactions occur in fixed bed reactors at temperatures ranging from 300 to 400 oC and pressure from 3 to 13 MPa, with Ni-Mo or Co-Mo catalysts supported on -Al2O3. The image shown in Figure 2 is a process flow diagram of a hydrodesulphurization unit.                Figure 2: Schematic diagram of a typical HDS unit in a petroleum refinery7  As shown in Figure 2, the liquid feed is pumped to the desired pressure and mixed with recycled hydrogen-rich gas. The mixture of liquid-gas feed, flows to the heat exchanger to be preheated. The resulting feed passes through a fuel fired heater to be fully vaporized and heated to the desired temperature, prior to entering the reactor and passing through the packed bed of catalysts where the HDS reaction occurs. The heated products from the reactor are cooled by 4  passing through a heat exchanger and a water-cooled heat exchanger prior to passing through a pressure controller (PC) which lowers the pressure to approximately 0.3 to 0.5 MPa. The product mixture then passes to the gas separator vessel at 35 oC and 0.3 - 0.5 MPa pressure. The majority of the hydrogen-rich gas that comes from the gas separator vessel is directed to an amine contactor in order to remove the hydrogen sulfide gas (H2S). The resulting S free gas is recovered for recycle. Some S gas from the gas separator is blended with sour gas from the distillation column. The liquid from the gas vessel is directed to the distillation column. The stripped liquid products in the bottom are the desired desulfurized products.  Ethane, methane, propane and H2S as well as some heavier compounds are the major components of the sour gas from the distillation stripper. The sour gas is directed to a central gas processing plant in the refinery to remove H2S using an amine gas treating unit through a series of distillation columns to recover the heavy components such as propane, butane, and pentane. The H2S extracted by the amine treating unit is then converted to S in a Claus unit7.   1.3 Canadian Oil Sands, Petroleum Coke and Activated Carbon: The Canadian oil sands in Alberta produce about 1.8 million barrels of oil per day. The oil sands is a major source of energy for North America. The Canadian oil sands contain about 10% bitumen. Bitumen undergoes upgrading to a synthetic crude. As a result of this process, petroleum coke (petcoke) is produced as a byproduct in large amounts, ranging between 5-6 million tonnes per year. There are two different types of coke generated in oil sand upgrading: delayed coke which is generated by Suncor and fluid coke which is generated by Syncrude. Delayed coke differs from fluid coke in its properties8. As shown in Table 29, the coke yield 5  from the delayed coker process is 33 wt% whereas from the fluid coker it is 20 wt%. Fluid coke contains 9 wt% of S which is higher than the delayed coke S content of 7 wt%. Since the S content in the coke is high, it needs to undergo pretreatment processes to be purified for use in other processes such as the iron, cement, aluminum, and thermoelectric industries. The limitations to petcoke use are caused by the high S content of 4-9%, and the presence of other metals such as Ni and V. Petcoke also has a low surface area < 5 m2/g8-10. On the other hand,petcoke has a low volatile matter content which results in a high combustion temperature which makes petcoke safe to handle. In addition, petcoke is superior in heat content compared to coal due to a lower ash content of ~7 wt%, and it is not an explosive hazard11. Since petcoke contains many impurities (S, N, V, Ni, and other heavy metals) it must undergo treatment processes to be purified. The present study is aimed at utilizing petcoke as a catalyst support. Since petcoke has a very low surface area (~5 m2/g), it must be processed to increase the surface area. The pre-treating process can also increase the porosity of the petcoke which makes it effective for use in hydrotreating catalysts.5,12-16 Activated carbon is a highly porous material and contains large surface area ranging between 1000-3000 m2/g and pore volumes of 0.8 - 1.2 cm3/g. Activated carbon is an effective adsorbent if it has a high surface area, but if it has a lower surface area and mesopores it could be used as a catalyst or catalyst support. The carbon structure consists of segments of graphene sheets with different sizes and dimensions bonded to each other in an unlimited number of ways.  The applications of activated carbon depend on its physical and chemical properties. The activated carbon can also be used as a catalyst support that aids in the dispersion of active phase due to its large surface area and weak polarity17. For metal sulfide supported on activated carbon, the formation of (Ni,Co)-Mo-S phases are an expected result of the weak interactions between the 6  metal and the support in the case of carbon17. Activated carbons also contain a high number of micropores, but they are usually not effective in hydrotreating processes.18 In the current study, Ni-Mo catalysts were supported either on commercial activated carbon or petcoke. There have been several studies showing that cobalt-molybdenum (Co-Mo) or Ni-Mo supported on activated carbon have promising hydrotreatment activities for removing S from oil feedstocks5,19,20. As aforementioned, there is a large amount of petcoke that has been produced from bitumen upgrading. In the current study, petcoke will be assessed experimentally for use as a support of NiMoS catalysts. The catalysts will be assessed by measuring the conversion and product distribution over a range of conditions and hence the HDS reaction kinetics will be determined. Reaction mechanisms of HDS of DBT have been explored in many studies showing that the reaction occurs by two routes; the direct desulphurization reaction route (DDS) and the hydrogenation reaction route (HYD)3-6,21-23. In the DDS route, the reaction starts by extracting the S from the DBT reactant producing biphenyl (BP). The hydrogenation of BP to form cyclohexylbenzene (CHB) is very slow compared to the other steps and therefore it can be ignored in the kinetic model22. In the HYD route, the reaction starts with the hydrogenation of the DBT to produce tetrahydrodibenzothiophene (THDBT), followed by S removal from the THDBT to produce CHB. Further hydrogenation of CHB leads to bicyclohexyl (BCH).     7  Table 2: Product yield from Delayed coke and Fluid coke processing of Alberta bitumen9 Component /Coke type Delayed Coke Fluid coke H2S (wt%) 1.1 0.7 Light Ends (wt%) 11.1 11.6 Naphtha (vol%) 25.66 20.7 Middle distillate (vol%) 26.4 15.8 Gas Oil (vol%) 13.8 32.5 Coke (wt%) 33 20 S content (wt%) of the coke 7.35 9  1.4 Literature Review  Many previous studies have been conducted on the HDS of model compounds such as DBT as well as real oil feedstock. A variety of feeds, types of catalyst, reactors, and reaction conditions have been investigated3-6,21-23 in academic research and industrial processes, reflecting the commercial importance of  HDS processes. In this chapter, studies related to the current research are reviewed and discussed.  1.4.1 Comparison of Carbon Supported Catalysts with other Supports  Shi et al.5 studied the effect of different types of support on heavy vacuum gas oil (HVGO) upgrading.  The activated carbon used as a support was obtained from Alberta oil sand petroleum coke, either delayed coke or fluid coke. The coke needs to undergo a chemical process to be purified as well as to improve its physical properties. The BET surface area of the coke after chemical treatment was 2194 m2/g for the fluid coke and 2357 m2/g for the delayed coke, and their pore volumes were as high as 1.2 cm3/g. Both supports had a large number of micropores. Ni-Mo was loaded onto the supports by impregnation of Ni(NO3)2 and 8  (NH4)6Mo7O24 followed by calcination in N2 at 773 K. For comparison, two types of catalysts were synthesized using the same aforementioned procedures but using commercial activated carbon and highly porous -Al2O3 supports. The catalysts were sulphided and a magnetically stirred (400 rpm) autoclave reactor along with HVGO feedstock were used to evaluate the hydrotreatment performance of the catalysts at 643 K and 3.45-5.50 MPa, after 2 h of reaction. In addition to the prepared catalysts, two commercial catalysts were also examined and compared under the same aforementioned conditions. As shown in Figures 3 and 4, the catalysts supported on delayed and fluid coke had better hydrotreating performance when compared with the other supports. The higher hydrotreating activity of the catalyst supported on either delayed coke or fluid coke was as a result of high surface area and pore volume of these supports when compared with other supports.  Figure 3: HDS activites obtained with different catalysts (FAC:fluid coke, DAC:delayed coke,CAC: commercial activated carbon) under 643 K,3.45 MPa, and 2 hours reaction time. Copyright © 2012, Elsevier5  Figure 4: HDN activities obtained with different catalysts (FAC: fluid coke, DAC: delayed coke, CAC: commercial activated carbon) under 643 K,3.45 MPa, and 2 hours reaction time. Copyright © 2012, Elsevier5   9  Liu et al.24 studied the composite alumina-activated carbon (AAC) as a support of Ni-Mo catalyst for HDS of DBT and compared the results with Ni-Mo supported on alumina (NiMoS/-Al2O3) and Ni-Mo supported on activated carbon (NiMoS/AC). The three catalysts were prepared by incipient wetness impregnation with a loading of 7 wt% Ni and 14 wt% Mo. A trickle-bed reactor was used to study the catalyst activity for DBT hydrodesulphurization. The catalysts were sulfided in-situ. The feed contained 1 wt% of DBT in decalin and the reaction was conducted at 260-300 oC with 3 MPa pressure along with 500 Nm3/m3 H2/feed ratio and 0.3-0.6 h-1 LHSV. The investigators showed that the catalyst activity for DBT hydrotreating was significantly increased using the AAC as a support. As shown in Figure 5, 90% DBT conversion was achieved at 260 oC compared with only 58% achieved on the NiMoS/AC and 32% conversion on the NiMoS/-Al2O3. The activation energies for the catalysts were estimated from the data of Figure 5 to be 143 kJ/mol for NiMoS/-Al2O3, 105 kJ/mol for NiMoS/AC, and 89 kJ/mol for NiMoS/AAC (refer to Appendix A for detailed calculation). The significant impact of the aforementioned catalytic activity was due to the higher dispersion of Ni and Mo on the alumina-activated carbon support versus the other supports as shown in Figure 6. The MoO3 (+) disappears from the NiMoS/AAC diffractogram whereas it is present for the NiMoS/-Al2O3, implying that a higher dispersion of Mo occurred on the surface of the alumina-activated carbon than the -Al2O3.24 10   Figure 5: The conversion of DBT over presulfided catalysts NiMoS/-Al2O3, NiMoS/AC and NiMoS/AAC as a function of reaction temperature. Copyright © 2011, Elsevier24 11   Figure 6: The XRD patterns of NiMoS/γ-Al2O3 (1), NiMoS/AAC (2) NiMoS/AC (3), and AAC (4).Copyright © 2011, Elsevier24 12  Shi et al.6 studied how mesoporous carbon as a support for Co-Mo and Ni-Mo catalysts affects the HDS of 4,6-DMDBT and DBT. Mesoporous carbons (CMC) have larger surface area (1400–2000 m2/g), pore diameter (3–7 nm) and pore volume (1.44–2.72 cm3/g) compared to activated carbons (AC) with 936 m2/g surface area, 3.8 nm pore diameter, and 0.34 cm3/g pore volume. An impregnation method was used to prepare the catalysts using different types of chelating agents such as citric acid, tetraacetic acid ethylenediamine, ethylenediamine, and nitrilotriacetic acid (NTA). NTA was found to be the best chelating agent for the preparation of Co-Mo/CMC and Ni-Mo/CMC. Co-Mo/CMC and Ni-Mo/CMC showed better surface area and pore volume than the AC supported catalysts. Consequently, those catalysts which were supported on CMC exhibited much better activity than the AC supported catalysts in the HDS of organic S, as most of the micropores of the CMC were converted to mesopores25 and hence the diffusion rate of the Ni-Mo or Co-Mo species on CMC were faster during the impregnation. According to Figure 7, Co-Mo/CMC showed higher activity than the commercial Co-Mo/ γ-Al2O3 in the HDS of thiophene. Moreover, as shown in Figure 8, Ni-Mo/CMC was significantly more active than the commercial Co-Mo/γ-Al2O3 in the HDS of DBT.6   13   Figure 7: Activity of Co-Mo/CMC-1, Co-Mo/AC, and commercial Co-Mo/γ-Al2O3 for the HDS of thiophene in a model gasoline. Reaction conditions: 1.5 MPa, LHSV = 2 h-1, H2/feed volume ratio = 300. The model gasoline contained 0.05% sulfur from thiophene and 20% 1-hexene, and the balance was n-heptane. Copyright © 2010, Elsevier6  Figure 8: Activity of Ni-Mo/CMC-2 and the commercial catalyst FH-98for the HDS of  dibenzothiophene (DBT) in a model diesel. Reaction conditions: 3.1 MPa, LHSV = 2 h−1, H2/feed volume ratio = 500. The model diesel contained 0.3% sulfur from DBT, 0.02% N from quinolone 5% and 0.5% n-octane as an internal standard. Copyright © 2010, Elsevier6  Shu and Oyama21 studied the hydrotreating activity of carbon-supported transition metal phosphides. Ni2P/C, MoP/C, and WP/C were synthesized through temperature-programmed reduction. BET surface area, CO uptake, extended X-ray absorption fine structure (EXAFS) measurements, and X-ray diffraction (XRD) were used to characterize the above samples. The hydrotreating activity of these catalysts was measured at 613 K and 3.1 MPa and carried out in three different sets of experiments using a packed-bed reactor for HDS and HDN with a liquid feed consisting of (i) 3000 ppm S as CH₃SSCH₃, (ii) 500 ppm S as 4,6-DMDBT and (iii) 200 ppm N in the form of C9H7NO. The best activity based on a constant number of CO chemisorption sites (70 µmol) in the reactor, was found to be the Ni2P/C catalyst. This study found that the best Ni loading for both HDS and HDN was 1.656 mmol/g (11 wt% Ni2P) which 14  contributed significantly to raising the HDN conversion to 100% and HDS to 99% at a molar space velocity of 0.88 h-1. The aforementioned conversion results illustrated that the carbon support played a major role for high HDS and HDN activities in comparison with the commercial Ni-Mo-S/-Al2O3 and Ni2P/SiO2 catalysts, which obtained only 68%-94% conversions. Some exceptional features that make carbon better than silica as a support, include smaller metal phosphide particle size, higher CO uptake, less retention of P on the support, and reduced sulfur deposition.21 Kouzu et al.23 prepared four types of activated carbon, differing in surface area from 916 m2/g to 3075 m2/g, as support for NiMo-sulfide (NiMoS/AC) catalysts to study the effect of properties of supported carbon on the HDS of 4,6-DMDBT. The study was carried out at 320 oC using an autoclave reactor. They reported that an increase in HDS activity was observed as the surface area of the AC increased. By increasing the area, the dispersion of catalytic active components increased. Oxygen functional groups that were present in the activated carbon, restricted the dispersion of the catalytic component. To improve the dispersion, oxygen groups of the activated carbon were reduced, which resulted in an enhancement of dispersion and, hence, the HDS activity. After the reduction treatment, NiMoS was prepared and supported on the reduced AC which catalyzed the hydrogenation reaction path of HDS of 4,6-DMDBT and DBT more than the commercial alumina supported catalysts (NiMoS/-Al2O3). A tubular reactor was also used to carry out the reaction with a real-feedstock under H2 and 5 MPa pressure in 250 nL/L volumetric ratio of H2/Oil. The best result achieved by NiMoS/AC catalyst at 337 oC was a product with 10 ppm-S, whereas for NiMoS/γ-Al2O3 catalyst even at the highest HDS temperature of 350 oC, the catalyst could not achieve the same result as the NiMoS/AC catalyst, as shown in Figure 9.  15   Figure 9: Temperature vs. residual sulfur content in HDS reaction of real gas–oil feedstock using down-flow tubular reactor, with LHSV of 1.5 h-1 under hydrogen pressure of 5MPa in volumetric hydrogen/oil ratio of 250 nl l-1. (_ ) Over the best Ni-Mo/Activated carbon catalyst (k at 330 oC = 1.59 h−1), (_ ) over the conventional Ni-Mo/γ-Al2O3 catalyst (k at 330◦C = 0.68 h-1) Copyright © 2004, Elsevier23.   Sakanishi et al.3 studied the kinetics and mechanism of 4,6-DMDBT HDS over NiMo/C. To establish the kinetics and mechanism of HDS, the feed containing 4,6-DMDBT in decane was hydrodesulfurized (NiMoS/C) using a 50-ml autoclave, equipped with a sampling apparatus, under hydrogen pressure of 3 to 15 MPa. Ni-Mo salts were dissolved in aqueous CH₃OH to impregnate the carbon support by simultaneous impregnation. After impregnation, the oxides on the carbon were presulfided at 360 oC for 2 h .The study compared alumina and carbon as support for NiMo. NiMo/C catalysts showed higher activity than the commercial NiMo/-Al2O3 for HDS of 4,6-DMDBT in the temperature range 613-653 K. It was found that the DDS reaction was dominant in this temperature domain. The study also showed that the HYD route of HDS was the dominant pathway at temperature < 573 K. Another aspect of this study was to show the   16  effect of H2S on the catalyst activity. Despite the effect of temperature, H2S extensively inhibits the DDS route, but on the other hand, H2S significantly enhances the HYD route of the HDS of 4,6-DMDBT.  Sakanishi et al.26studied Ketjen Black (KB) as a support for NiMo catalysts for hydrogenation of 1-methylnaphthalene and compared it in activity and conversion with a commercial NiMo/-Al2O3 catalyst. KetjenBlack (KB) has unique properties such as high surface area >1000 m2/g and low specific gravity around 115 g/L. The reaction was carried out using a 50-mL magnetically-stirred autoclave reactor under the standard conditions of 380 oC, 40 min, and 9.5 MPa of H2 reaction pressure. The catalyst was prepared from [CH3COCH=C(O)CH3]2MoO2 and Ni(CH3CO2)2 in CH3OH solution using successive impregnations of Mo (10 wt %) and Ni (2 wt %), respectively. The catalyst exhibited the highest conversion of 84% to methyltetralins as shown in Figure 10. It is worth mentioning that the merging of organic soluble metal salts with methanol may have great impact on suppressing the agglomeration of KB ultrafine carbon particles due to the balanced hydrophobic nature of the solvent matrix. The study found that NiMo supported on KB is more active than the commercial NiMo/-Al2O3.   17    Figure10: Effect of catalyst amount on the hydrogenation conversion of 1-methylnaphthalene with NiMo/KB or NiMo/γ-Al2O3. Reaction conditions: 1-Methylnaphthalene/decalin= 1/9; reaction temperature 380 oC; reaction pressure 9.5 MPa; reaction time 40min; catalyst NiMo/KB or NiM/γ-Al2O3, 1-5wt% addition to 1-Methylnaphthalene.Adapted from [26].Copyright © 1995, American Chemical Society 26  1.4.2 Reaction Kinetics  Egorova and Prins22 studied the HDS of DBT and 4,6-DMDBT over sulfided CoMo/- Al2O3, NiMo/-Al2O3 and Mo/-Al2O3. The NiMo/-Al2O3 results were obtained from elsewhere.27,28 Since the DBT results are related to the current study, they will be discussed here. The feed in the gas phase contained 130 kPa toluene, 8 kPa dodecane as a reference for DBT and their derivates, 1 kPa DBT , 4 MPa H2 and 0-100 kPa H2S. A continuous fixed- bed Inconel reactor was used to carry out the reactions.27,28 The Ni and Co promoters effectively contributed to the activity of the Mo metal catalyst especially through the DDS route of the HDS of DBT, while the HYD route was to a much lower extent promoted. Several experiments were conducted at 340 oC and 35 kPa H2S with different space velocities over Mo/-Al2O3 and CoMo/γ-Al2O3 as 01020304050607080901000 1 2 3 4 5conversion %Catalyst (wt%)18  shown in Figures 11 and 12. Adding Co as a promotor significantly enhanced the DDS reaction route but the HYD route was not enhanced to any great extent.   Figure 11: Relative partial pressures of the products in the HDS of dibenzothiophene at 340 oC and 35 kPa H2S over Mo/- Al2O3 as function of weight time.Copyright © 2004, Elsevier 22  Figure 12: Relative partial pressures of the products in the HDS of dibenzothiophene at 340 oCand 35 kPa H2S over CoMo/-Al2O3 as function of weight time. Copyright © 2004,Elsevier 22  Egorova and Prins proposed the following reaction sequence for the HDS of DBT  Figure 13: Proposed reaction pathway for HDS of DBT. Copyright © 2004, Elsevier 22 19  It was found that the HDS of DBT over CoMo/-Al2O3 and NiMo/-Al2O3 catalysts proceeded through steps 1, 2, 3 and 4 of the reaction pathway, whereas over Mo/γ-Al2O3 catalyst the reaction occurred through steps 1, 2, 3, 4, 5, 6 and 7. Reaction 7, which was the hydrogenation of BP to CHB, was found to occur over the Mo and NiMo catalysts, but not over the CoMo catalysts. This suggests that the active sites aiding HDS over the NiMo and Mo catalysts are not quite the same as those over the CoMo catalysts. The difference was explained by DFT analysis which showed that the active sites of the Co are at the edge of the S slab of MoS2, whereas the active sites of Mo and NiMo catalysts are at the Mo atom edge29,30. The active phase of MoS2 is always promoted by Ni or Co. As shown in Figure 14, the layered structure of an MoS2 particle results in rim and edge sites. It is reported in the literature31 that the DDS reaction in the HDS of DBT takes place at edge sites, while the HYD reaction takes place on rim sites. In order to improve S removal, it is suggested to increase the height of the MoS2 particle to allow for more DDS reactions to occur31. Moreover, the structure of MoS2 is significantly sensitive to the support where the metal support interaction plays a significant role in MoS2 dispersion. According to Hensen et al.32 a -Al2O3 support results in a lower dispersion of MoS2 compared to other supports such as silica and AC.  20   Figure 14: Rim/edge model of MoS2 particle adapted from [31]. Copyright © 1994Academic Press31.   Egorova and Prins22 also studied the effect of different H2S partial pressures on the HDS of DBT over the three aforementioned catalysts (CoMo/γ-Al2O3, NiMo/γ-Al2O3, Mo/γ-Al2O3,). H2S partial pressures of 0, 10, 35, and 100 kPa were used in the study at 340 oC. In the absence of H2S, the hydrotreatment activities were fast and conversion reached 100% at only τ=(W/F)= 1.9 g.min/mole over both NiMo/γ-Al2O3 and CoMo/γ-Al2O3 catalysts, and 55% conversion over Mo/γ-Al2O3 at τ=3.7 g.min/mol. The DBT conversion was strongly inhibited by H2S on the three types of catalyst. The conversion reduced to 53% over CoMo/γ-Al2O3, 60% over NiMo/γ-Al2O3 and only 24% over Mo/γ-Al2O3 at τ=3.7 gmin/mol and 100 kPa H2S. The pseudo-first-order kinetics was used to describe the observed kinetics in this study. Mo/γ-Al2O3 catalyst was found to be in an agreement with first order kinetics showing that changing the initial pressure does not affect the conversion, whereas over the other two catalysts the conversion changed with initial pressure. The behavior of HDS activity was described by the following equation:  d n Layers    rim edge basal 21  r = 𝑘𝐾𝐷𝐵𝑇 𝑃𝐷𝐵𝑇(1+𝐾𝐷𝐵𝑇 𝑃𝐷𝐵𝑇+𝐾𝐻2𝑆𝑃𝐻2𝑆)  For further simplification, the authors assumed that 𝑃𝐻2𝑆  dominated the surface and the equation was reduced to r = k’PDBT   where k’= 𝑘𝐾𝐷𝐵𝑇 (𝐾𝐻2𝑆𝑃𝐻2𝑆) . The pseudo rate constant includes the effect of 𝑃𝐻2𝑆. Table 3 shows the estimated rate constants for all partial pressures. The product distributions were analyzed at different H2S partial pressures, the rate constants of both routes DDS and HYD (kDDS, kHYD) were measured from ktot and the selectivities at low space velocity. Table 3: Rate constants of the total, DDS and HYD (desulfurization) of DBT and of the desulfurization of tetrahydrodibenzothiophene over NiMo//-Al2O3, CoMo//-Al2O3, and CoMo//-Al2O3 catalysts at different initial partial pressures of H2S and a fixed temperature of 340 oC. Adopted from [22] Copyright © 2004, Elsevier 22  Catalyst           Rate constant, mol/(g min)       P𝐻2 𝑆 (init), kPa 0                10                35              100 NiMo                 𝑘𝑡𝑜𝑡                           𝑘𝐷𝐷𝑆                                        𝑘𝐻𝑌𝐷                            𝑘𝐷𝐸𝑆𝑈𝐿𝐹𝐻𝑌𝐷            CoMo                𝑘𝑡𝑜𝑡                           𝑘𝐷𝐷𝑆                                        𝑘𝐻𝑌𝐷                            𝑘𝐷𝐸𝑆𝑈𝐿𝐹𝐻𝑌𝐷            Mo                     𝑘𝑡𝑜𝑡                           𝑘𝐷𝐷𝑆                                        𝑘𝐻𝑌𝐷                            𝑘𝐷𝐸𝑆𝑈𝐿𝐹𝐻𝑌𝐷               2.33          1.06          0.39          0.25    2.05          0.90          0.32          0.17    0.28          0.16          0.07          0.08    36             15             5.3            2.7      1.42          0.42          0.36          0.20    1.31          0.34          0.28          0.14    0.11          0.08          0.08          0.06    29             9.1            5.1            3.3      0.25          0.09          0.08          0.08    0.19          0.05          0.03          0.03    0.06          0.04          0.05          0.05    4.7            1.3            0.70          0.50 22  In Table 3, kDDS is the rate constant of the DDS route in Figure 13, kHYD is the rate constant of the HYD reaction route in Figure 13, 𝑘𝐷𝑒𝑠𝑢𝑙𝑓𝐻𝑌𝐷  is the rate constant for the HYD reaction followed by the desulphurization reaction pathway (2, 3, and 4) in the case of the NiMo or CoMo catalysts and in the case of the Mo catalyst the reaction follows the pathway (2, 3, 4, 5, and 6) in Figure 13, ktot is the summation of kDDS and kHYD.  Based on the values of rate constants reported in Table 3, the total rate constant ktot of Mo was the least affected by the H2S pressure changes as it follows the pseudo-first order behavior with only 68% reduction compared with the initial rates at zero H2S pressure. The CoMo rate constant was reduced by 85% and the NiMo total reaction rate was the most affected rate constant with 89% reduction. For more illustration, Figure 15 indicates the effect of H2S partial pressure on the estimated rate constants for each of the catalysts studied.   Figure 15: Pseudo first-order rate constants of the total conversion of DBT over CoMo/γ-Al2O3, NiMo/γ-Al2O3, and Mo/γ-Al2O3 catalysts at different partial pressures of H2S and a fixed temperature of 340 oC. Copyright © 2004, Elsevier22  23  Vanrysselberghe and Froment33 studied the kinetics of the HDS of DBT over a commercial CoMo/-Al2O3 catalyst. The feed contained 2 wt% of DBT which has 0.35 wt% S content dissolved in Parapur. Parapur is a mixture of paraffin consisting of 33.66 wt% CH3(CH2)10CH3, 48.42 wt% CH3(CH2)9CH3, 5.23 wt% CH3(CH2)8CH3, 12.52 wt% CH3(CH2)11CH3, and 0.19 wt% CH3(CH2)12CH3. Experiments were conducted in a multiphase continuous reactor under a range of 5-8 MPa total pressure and between 513-573 K temperature, the DBT molar flow rate 𝐹𝐷𝐵𝑇𝑜  was between 1.74x10-6 and 4.04x10-6 kmol/h and the molar H2 to HC (hydrocarbon) ratio was varied between 1.1-4.1. The molar hydrogen to methane ratio was set to be 6.4 for all the experiments. Methane acted as an internal standard for online analysis of the reactor effluent. A commercial CoMo/-Al2O3 catalyst was used with 5-30 wt% MoO3, 0-6 wt% SiO2 1-10 wt% CoO , and 0-10 wt% P2O5 on -Al2O3 support. 𝑊𝐹𝐷𝐵𝑇𝑜 −𝐹𝐷𝐵𝑇𝑔  kgcathr/kmol was defined as the space time where 𝐹𝐷𝐵𝑇𝑜  is the DBT molar flow rate and 𝐹𝐷𝐵𝑇𝑔 is the evaporated fraction  of DBT. Different space times influenced the HDS of DBT to its products at fixed 553 K and 6 MPa condition. Figure 16 shows that the DBT was mostly and rapidly desulphurized to BP but the hydrogenation of BP to CHB was slow. Also, DBT was hydrogenated to THDBT and other partially hydrogenated DBT intermediates, but they were all instantly converted to CHB. The aforementioned observation is in agreement with the study by Egorova and Prins22.  A prime example to explain the above observation is at 600 kgcat hr/kmol: the DBT conversion was 39%, the conversion to BP was 27%, and the conversion to CHB was 8% whereas at 1200 kgcat hr/kmol, the conversion was promoted for DBT, BP and CHB to 58%, 48%, and 14% , respectively as shown in Figure 16. At the same space time, the temperature variations significantly enhanced the conversion of DBT as illustrated in Figure 17. For instance, at 600 24  kgcat hr/kmol and 513 K, DBT conversion was 17% whereas at the same space time of 600 kgcat hr/kmol and 573 K, DBT conversion was 72%. The highest conversion was 87% achieved at 1000 kgcat hr/kmol and 573 K.   Figure 16: Conversion as function of space time (  ) total conversion of DBT,(   ) conversion of DBT into BP,(    ) conversion of DBT into CHB. Reaction conditions=553K, Pt=60 bar, H2/CH4=6.39.Copyright © 1996 American Chemical Society33     25   Figure 17: Total conversion of DBT as a function of space time at various temperature:  (  ) 513, (   )533, (   ) 553,and (   ) 573K under Pt=80 bar H2/CH4=6.38, and H2/HC=1.33.and H2/HC=1.10 reaction conditions. Copyright © 1996 American Chemical Society33  In order to study the reaction kinetics, the authors deduced the following reaction pathway  Figure 18: Proposed reaction network, note that here BPH refers to biphenyl (BP)). Copyright © 1996 American Chemical Society33.      26  A kinetic model for HDS was developed based on σ and τ, the two different types of active sites for hydrogenolysis and hydrogenation reactions, consistent with the idea of rim-edge sites in the Daage and Chianelli31 model. A Langumir-Hinshelwood mechanism was developed for both types of reactions as described below, the rate determining step for all the reactions was the surface reaction step. For instance, the hydrogenation of BP into CHB on the τ sites, the rate determining step was the surface reaction step BP. + 2H. PHCHD.+ 2 (The model description and rate expression are provided in Appendix B).  In addition, the adsorption equilibrium constants and rate coefficients at 573K were estimated and are reported in the following table. Table 4: Adsorption equilibrium constants and rate coefficients at 573K. Adopted from [33] Copyright © 1996 American Chemical Society33   𝐾𝐷𝐵𝑇, = 7.56868 × 101 𝐾𝐻,     = 7.01679 × 10−1  𝐾𝐻2𝑆, =  6.27912 × 101    𝐾𝐵𝑃, =  9.53728  𝑘𝐷𝐵𝑇, = 1.58251 × 10−1 𝐾𝐷𝐵𝑇,𝜏 = 2.52021 𝐾𝐻,𝜏     = 1.41658 × 10−2 𝐾𝐵𝑃,𝜏 = 1.41256 𝑘𝐷𝐵𝑇,𝜏  =  3.08384 × 10−1 𝑘𝐵𝑃,𝜏  = 1.69206 𝑘𝐶𝐻𝐵,𝜏 𝐾𝐶𝐻𝐵,𝜏 = 3.38631 × 10−1  Based on the results above, the adsorption rate constants of DBT, hydrogen (H), and BP on σ active site 𝐾𝐷𝐵𝑇,𝜎, 𝐾𝐻,𝜎 , 𝐾𝐵𝑃,𝜎, with values of 75.7, 0.7 ,and 9.54 m3/kmol were significantly higher than the adsorption constants on τ sites 𝐾𝐷𝐵𝑇,𝜏, 𝐾𝐻,𝜏, and 𝐾𝐵𝑃,𝜏 with values of 2.52, 0.0141, and 1.412 m3/kmol, which explains why HDS of DBT prefers the DDS route over the HYD route. Another observation was that the adsorption of DBT, BP and hydrogen is weaker m3/kmol m3/kmol m3/kmol m3/kmol kmol(kgcat h) m3/kmol m3/kmol m3/kmol kmol(kgcat h) m3/(kgcat h) m3/(kgcat h)    27  on the τ sites than on the σ sites. Moreover, the apparent activation energy for DBT hydrodesulphurization was estimated as 122 kJ/mol which is in good agreement with the results listed in the literature and summarized in the following table.24,33-37  Table 5: Comparison of activation energies for HDS of DBT from several literature studies  The authors Reaction conditions Catalyst Ea kJ/mol Broderick et al34 548-598 K, 3.4-16.2 MPa H2 Co-Mo/-Al2O3 125 Vrinat et al36 473-520 K, 0.5-5 MPa H2 Co-Mo/-Al2O3 96 O'Brien et al37 540-695 K 3.4-12.2 MPa H2 Co-Mo/-Al2O3 138 Vanrysselberghe and Froment33 513-573 K 0.5-0.8 MPa H2 Co-Mo/-Al2O3 122 Liu et al24 533-573 K 3 MPa H2 Ni-Mo/-Al2O3 145    Broderick et al.34 Vrinat et al.36 and O'Brien et al.37 used a Langmuir-Hinshelwood (L-H) kinetic model to estimate the activation energy. Broderick et al.34, Vanrysselberghe and Froment33 reaction conditions are close to each other as a result they obtained more or less the same activation energies. Liu et al24 obtained a high activation energy compared to others but a different catalyst NiMo/-Al2O3 was used in this study. It is observed that the activation energies appear to depend on the reaction temperature and pressure.     28  1.5 Literature Review Summary  As described in the literature review, several studies have discussed HDS reactions of DBT using different types of catalyst support. DBT has been used as a reactant with decalin as a solvent in many of the reports4,21,24,26. Continuous flow reactors as well as stirred batch reactors were used in previous studies to assess catalyst activity and selectivity. Reaction mechanisms of HDS of DBT reported in many studies show that the reaction occurs by two routes - DDS and HYD3-6,21-23. In the DDS route direct extraction of the S from the DBT occurs to produce BP. The subsequent hydrogenation rate of BP to form CHB is very slow compared to the other steps and therefore, it can be ignored in kinetic models.22 For the HYD route, where double bonds and phenyl species are hydrogenated prior to S removal to produce CHB, further hydrogenation will result in BCH. Incipient wetness impregnation is the most common HDS catalyst preparation method38. Many catalysts and supports have been reported in previous studies,3-6,21-23 with the most  active catalysts for the HDS of DBT being  bimetallic NiMo and CoMo sulfides 5,19,20. Ni or Co play a significant role as promoters in enhancing the activity of MoS2. γ-Al2O3 is widely used as a support for many catalysts. However, γ-Al2O3 has a lower surface area than AC. AC has a large surface area and pore volume but it contains a large fraction of micropores which are not utilized in HDS as the diffusion rate of reactants and products are very low through micropores. Moreover, in terms of conversion and selectivity, the best catalyst loading was reported by Kouzu et al.23 to be 15wt% Mo and 3wt% Ni. Nevertheless, NiMo catalysts were inhibited by H2S pressures changes as reported in the study by Egorova and Prins22, but they were chosen for the present research due to their high HDS activity towards DBT. 29  1.6 Study Objectives The aim of the present study is to investigate and compare the HDS activity of Ni-Mo-sulfided catalysts on carbon supports, including petcoke, to the known HDS activity of a conventional -Al2O3 supported catalyst. A major goal of the study is to demonstrate that the petcoke can be converted to a useful support for catalysts in hydrotreating processes. Consequently, hydrotreating technology might be able to use petcoke as a promising alternative support for catalysts instead of commercial -Al2O3. In addition, it is expected that the petcoke may contain promoters or other contaminants which may be beneficial for catalysis. The kinetics of the HDS reaction over the different carbon supports will be evaluated and compared to the known kinetics of Ni-Mo on -Al2O3 supports. The experiments have been performed in a novel slurry-phase batch microreactor. The Levenberg-Marquardt method was used to estimate the kinetic parameters from the measured data.  30  Chapter 2: Experimental  2.1 Catalysts Preparation Two types of carbon support were used to prepare the Ni-Mo catalysts. Activated carbon (DARCO®, −100 mesh particle size, powder, Sigma Aldrich) has a surface area of 1332 m2/g and 0.35 cm3/g pore volume. The petroleum coke (petcoke) was obtained from Canadian oil sands in Alberta, and subjected to a chemical activation process using KOH, to be more effective for the hydrotreating process5. The raw coke was ground and sieved to match the particle size of the commercial activated carbon of 125-149 μm. The ground coke sample was mixed with KOH with mass ratio of KOH/coke = 3/1. The sample was placed in an oven at 393 K and dried for 2 h. Then, the mixture was transferred into a ceramic boat that was inserted into a horizontal furnace with N2 gas flow at 50 mL/min. Subsequently, the mixture was heated to the desired temperature of 1073 K at a ramp rate of 5 K/min and kept at that temperature for 9 h. After the activation process the sample was cooled to room temperature and the product was washed with 1 M diluted HCl and distilled water through filtration paper until a pH of 7 of the product was achieved.5,39 The recovered coke was heated at 393 K overnight in ambient air. The resulting activated coke had a surface area of 1100 m2/g and a pore volume of 0.7 cm3/g. The Ni-Mo was added to the carbon support by wetness impregnation.  Mo with 15 wt% loading was added from a solution of (NH4)6Mo7O24·4H2O (Sigma Aldrich 81.0-83.0% MoO3 basis), and Ni at 3 wt% loading was added from a solution prepared from Ni(OCOCH3)2·4H2O (99.998% trace metal basis,Sigma Aldrich). The impregnation was done using 10 g of either the activated carbon or the petcoke as support. The incipient wetness impregnation method requires that the volume of the metal used in the preparation be the same as the pore volume of the support. The pore volume of the carbon supports was measured by N2 adsorption at 77 K, as 31  described in Section 2.2. The pore volume of the activated carbon was 0.35 cm3/g so the required solution impregnation volume was 3.5 cm3. The required amount of Mo salt was 4.11 g. Since the Mo salt was not soluble at this concentration, the Mo salt was impregnated three times as only 1.37 g of Mo salt was soluble in 3.5 cm3of H2O at a pH of 6.52. For each impregnation step, the sample was kept away from sunlight and aged overnight followed by drying at 393 K for 3 h. In the same manner, the desired amount of Ni was obtained from two impregnations using 0.75 g of Ni salt in 3.5 cm3 of distilled water at a pH of 7. As before, after each impregnation step the sample was aged overnight and dried at 393 K for 3 h. The same procedures were followed to prepare the catalysts supported on petcoke. In this case, 1.4 g of support with a pore volume of 0.7 cm3/g was prepared. One impregnation step was needed for the Mo as 0.58 g was soluble in 1 cm3 distilled water. The sample was aged overnight and dried at 393 K for 3 h. One impregnation step was also needed for Ni since 0.22 g of the Ni salt was soluble in 1 cm3 distilled water. Again the sample was aged overnight and dried at 393 K for 3 h.  The sulfidation of the NiMo catalyst was completed ex-situ in a 300 mL autoclave stirred reactor with 1400 mg catalyst, 1358 mg DBT, 66544 mg decalin, and 697 mg CS2 at 573 K and 4 MPa for 3 h to yield the NiMoS/AC and NiMoS/PC catalysts.    2.2 Catalyst Characterization. The total BET surface area, average pore width, and pore volume of the carbon supports and the prepared catalysts were measured using a Micromeritics ASAP 2020 Accelerated Surface Area and Porosimetry analyzer. Prior to the analysis the sample was degassed at 1.33 kPa/s while heating at a ramp rate of 10 oC/min until 0.04 kPa and 393 K. The temperature was kept at 393 K for 120 min. After degassing, the sample was moved to the analysis section and the 32  N2 adsorption/desorption isotherm was measured at 77 K. (Further details are provided in Appendix C). The catalysts were also characterized by X-ray diffraction (XRD) using a Siemens D500 instrument with Co Kα radiation (λ=0.1789 nm). The XRD patterns for all the samples were scanned from 10-90o with a scan step size of 0.04o. The XRD pattern was used to calculate the crystallite size using the Scherrer formula which relates crystal size to peak width at half maximum height.  X-ray photoelectron spectroscopy (XPS), a surface catalyst characterization technique, was done using a Leybold MAX200 instrument with an Al Kα achromatic X-ray source. XPS was used to confirm the presence of metals on the support. A survey scan with a pass energy of 192 eV, as well as a narrow scan, with a pass energy of 48 eV, were used to analyze the catalyst samples from which the presence of Ni, Mo, and S elements was confirmed.  2.3 Catalyst Activity Measurements The HDS reaction was conducted in a stirred batch microreactor in the presence of pure hydrogen as well as a measured amount of catalyst. Figure 19 shows a process flow diagram of the novel slurry-phase batch hydroconversion micro-reactor that was used in this study. The novelty of this reactor is represented in a micro scale size and the ability to mix the feed at that scale. The reactor consists of a glass insert of height 250 mm with an internal diameter of 4 mm, placed inside a stainless steel shell with glass beads placed inside the shell to keep the insert in the desired heating zone. Kukard has shown that the reactor operates isothermally.40 A vortex mixer placed at the bottom of the reactor was used to mix the feed and limit the effects of 33  external mass transfer. The reaction temperature was monitored by a thermocouple placed co-axially in the glass insert. In order to study the HDS reaction kinetics on the NiMoS catalysts on both the AC and petcoke supports, the effects of reaction time and temperature were examined for each of the different catalyst supports. The reactions were performed for 30, 60, 90 and 120 min at temperatures of 588, 603, 623, and 638 K. Pressure was fixed at a target of 4.8 MPa. At each temperature, two types of supports were examined with different reaction times as aforementioned to determine the activity for the HDS of DBT. At the beginning of each run, the glass insert was loaded with 150 µL of the total feed consisting of 2 wt% of the model compound (DBT), 95 wt% of decalin, 0.99 wt% carbon disulphide and 2 wt% of catalyst, and a total feed weight of 1 g. The insert was placed inside the reactor and sealed, the micro-reactor was purged three times in H2 to insure the removal of air from the system.  After purging, the reactor was pressurized to the target pressure using H2 to 4.28 MPa at room temperature. Prior to the start of the temperature ramp, the system was subjected to a leak test using a leak detector to insure that the system was well sealed. The micro-reactor was then heated to the desired temperature, and kept for the desired reaction time. At the end of the run, the system was depressurized and cooled to room temperature. The liquid products were recovered and stored at 2 oC. To identify the liquid products recovered from the reaction, a Shimadzu QP-2010S GC/MS equipped with Restek RTX5 30 M x 0.25 mm capillary column was used. The calibration curves were constructed by analyzing known concentrations of the DBT and BP. The GCMS reported different corresponding areas of known DBT and BP concentrations. The GCMS area for the unknown sample were determined and the concentration 34  was calculated using the aforementioned calibration curve (refer to Appendix D for more details). To minimize internal and external diffusion, the catalyst particle size was kept between 125 - 150 µm and the vortex mixer speed was set at 2000 rpm, as it was suggested that the short diffusion length and the dynamic mixing in the micro batch reactor promotes the HDS of DBT free of internal and external mass transfer effects.41 In other words, the mixing and the reduced wall effect enhance mass transfer which allows us to measure the intrinsic rate of reaction, as the mass transfer resistance is reduced. The carbon balance was determined by measuring the model compound DBT and all the products (BP, DBT, 1, 2, 3, 4-THDBT, CHB, and BCH) and the closure was better than 95%. To be consistent with the modeling, DBT conversion was calculated from the product measurements. The reproducibility of the micro-reactor was endorsed by repeating several selected runs and the results obtained showed ±11% error in the DBT conversion along with ±10 % of BP selectivity, ±10 % of THDBT selectivity, ±10 % CHB selectivity, and ±10 % BCH selectivity (refer to Appendix E for more details).   35   Figure 19: Process flow diagram of a novel slurry-phase batch hydroconversion micro-reactor. Copyright ©2015 American Chemical Society.41   36  Chapter 3: Results and Discussion  3.1 Catalyst Characterization Table 6 reports the BET surface area, pore volume and pore size of the carbon supports and the sulphided catalysts used in this study. Activated carbon is known to have a high surface area5 but contains a large number of micropores and some mesopores. The micropores are not necessarily useful for HDS of DBT because of their small diameter (< 2nm). After doping the Ni-Mo onto the supports and following sulphidation, the average pore size of the NiMoS/AC catalyst increased significantly because the very small micropores were filled. There was also a corresponding decrease in the surface area by ~90% and pore volume by ~37%. In the case of petcoke, the average pore size decreased by ~15% and the surface area and pore volume decreased by ~97%. The loss in porosity of the NiMoS/PC catalyst compared to the support is likely due to NiMoS that blocks and fills the pores causing the pore volume and surface area to decline, as well as due to the thermal treatments that likely collapse the pore structure of the carbon supports. Both effects result in loss of micropore and mesopore volume. The surface area of the non-sulphided commercial NiMo/-Al2O3 is also reported since the activity of this catalyst will be discussed in Section 3.3.1      37  Table 6: BET surface area, pore volume, and pore size for the catalysts and their supports Supports and supported Catalyst Specific Surface Area (m2/g) Pore volume (cm3/g) Pore size (nm) Metal content (wt%) Activated carbon(AC) 1323 0.35 2.21 - Petcoke (Petcoke) 1140 0.67 2.40 - NiMoS/AC 145 0.22 5.91 15Mo-3Ni NiMoS/PC 37 0.019 2.03 15Mo-3Ni NiMo/-Al2O3 71 - - 8Mo-3Ni   Figure 20 shows the XRD patterns of the activated carbon support, the NiMoS/AC catalyst, and the NiMoS/PC catalyst. The lack of peaks corresponding to MoS2 at 2θ = 40o and 2θ = 60o for the (103) and (110) planes of MoS2, suggests high dispersion of Ni-Mo on the support. The peaks at 2θ=23 to 36o are associated with graphite and SiO2 in the activated carbon. For the NiMoS/PC catalyst, high dispersion of the Ni-Mo on the support is also suggested by the lack of MoS2 peaks. 38                                  Figure 20: XRD pattrens for AC, NiMoS/AC, NiMoS/PC The XPS survey scan data of Table 7 show that the Mo/C and Ni/C surface atom ratio was higher on the petcoke than the AC, indicative of a higher dispersion on the petcoke than the AC. As shown by the narrow scan data of Figures 21, 22, 23, and 24 there were no apparent chemical shifts for the Mo and S species on the AC versus the petcoke. When deconvoluting the peaks in both supports for Mo 3d, MoS2, MoO2, and MoO3 species were identified (Tables 8, 9 and Figures 21, 23). Similarly, as a result of deconvoluting the S peaks for both supports (Tables 10, 11 and Figures 22, 24), a number of compounds were identified such as MoS2, DBT, SOx, dihydrothiophene/Mo, and S. Their binding energies are close to or mostly the same as that reported in the literature.42 The data in Table 8, 9, 10 and 11 show that the most abundant species is MoS2 which could not be identified by the XRD analysis. Some oxide species were also identified by XPS likely because of the exposure of the catalyst to air during sample handling procedures associated with the XPS analysis.  39  Table 7: Surface composition as defined by XPS of carbon supports NiMoS Support/Element Mo At% S At% Ni At% C At% Mo/C % Ni/C % Mo/S % Activated carbon 1.29 2.87 0.19 87.96 1.467 0.22 0.45 Petcoke 2.13 6.37 0.8 78.68 2.707 1.02 0.33    Table 8: Chemical states of Mo on activated carbon support from Mo 3d XPS narrow scan Mo Mo 3d3/2 B.E. Mo 3d5/2 B.E. B.E. Area% MoS2 232.3 229.1 51 MoO2 232.8 229.6 11 MoO3 235.8 232.5 38    Table 9: Chemical states of Mo on petcoke support from Mo 3d XPS narrow scan Mo Mo 3d3/2 B.E. Mo 3d5/2 B.E. B.E. Area % MoS2 232.1 229.1 55 MoO2 233 230 18 MoO3 236.1 233.1 27       40  Table 10: Chemical states of S on activated carbon support from S 2p XPS narrow scan Compound 2p1/2 B.E. 2p3/2 B.E. 2p B.E. S 2s B.E. B.E.Area% MoS2 163.8 162.2 - - 24 S or S8 164.7 163.1 - - 12 SOx - - 169 - 36 dihydrothiophene/Mo 162.8 161.7-161.6 - - 27 S - - - 226.4 0     Table 11: Chemical states of S on petcoke support from S 2p XPS narrow scan Compound 2p1/2 2p3/2 2p S 2s Area % MoS2 163.8 162.2 - - 23 S(C6H5)2 164.6 163 - - 15 SOx - - 169 - 41 dihydrothiophene /Mo 162.8 161.7 - - 21 S - - - 226.4 0   41   Figure 21: Narrow scan with peak deconvolution for Mo 3d on activated carbon   Figure 22: Narrow scan with peak deconvolution for S 2P on activated carbon 0200040006000800010000120001400016000220225230235240245Intensity (CPS)B.E (ev)Raw datasummation of the fitting30003500400045005000550060006500700075008000155160165170175180Intensity(CPS)B.E (eV)Raw dataSummation of the fitting42    Figure 23: Narrow scan with peak deconvolution for Mo 3d on petcoke   Figure 24: Narrow scan with peak deconvolution for S 2P on Petcoke  020004000600080001000012000150155160165170175180Intensity(CPS)B.E (ev)Raw dataSummation of the fitting0500010000150002000025000220222224226228230232234236238240242Intensity (CPS)B.E (ev)Raw datasummation of the fitting43  3.2 Catalyst Characterization Summary There was a loss in surface area and pore volume of the AC and petcoke supports after doping with NiMoS. XRD analysis did not show peaks attributable to MoS2, indictive of a high dispersion of this species on both the petcoke and AC support. XPS analysis indicated a higher dispersion of the Ni and Mo metals on the petcoke support compared to the AC support.  It was also found that there were no differences in the chemical states of both Mo and S on the supports, based on similar binding energies of these species on both supports.  3.3 Catalytic Activity 3.3.1 Preliminary Studies to Test the Activity of Different Catalysts Three types of catalysts were examined at the same reaction conditions in the first part of this study. The NiMoS/PC and NiMoS/AC were compared with a commercial NiMo/-Al2O3 catalyst. An activity test was also done without catalyst to determine the extent of reaction due to thermal reactions and/or reactions catalyzed by the components present in the stainless-steel thermocouple. The HDS reaction was carried out at 623 K and 4.8 MPa at different reaction times. As shown in Figure 25, the NiMoS/PC activity, as reflected in the DBT conversion was promising when compared with the optimized commercial catalyst. Even though the activated carbon has higher surface area than the commercial NiMo/-Al2O3 and the NiMoS/PC catalyst, it showed the lowest DBT conversion activity. This is likely because of inaccessible active sites in the micropores of the AC. In addition, the commercial NiMo/-Al2O3 catalyst contains some other components such as phosphorus to optimize activity. Moreover, the surface composition shows that the low activity of the activated carbon corresponds to a lower Mo/C and Ni/C of the NiMoS/AC compared to the NiMoS/PC catalyst. Other temperatures for the NiMoS/AC and 44  NiMoS/PC catalysts were also examined and the data are shown in the Section 3.3.2 and 3.3.3, while the thermal run data are summarized in Appendix F.  Figure 25: Comparison of DBT conversion between catalysts, NiMoS/γ-Al2O3, NiMoS/AC, NiMoS/PC and thermal reaction at 623 K and 2000RPM at different times  In terms of selectivity of the three catalysts, Figures 26(a-d) show that the NiMoS/PC has almost as high selectivity towards BP as the commercial NiMoS/γ-Al2O3. This indicates that NiMoS/PC is not only an active catalyst but also highly selective towards BP via the DDS route.  Figure 26(a): Selectivity of NiMoS/γ-Al2O3 catalysts at 623 K and 2000RPM at different time  01020304050607080901000 20 40 60 80 100Conversion % Time (min)NiMoS/Al2O3NiMoS/ACNiMoS/PCThermal0102030405060700 30 60 90 120 150Selectivity, Wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene, 1,2,3,4-tetrahydro-45  Figure 26(b): Selectivity of NiMoS/ PC catalysts at 623 K and 2000RPM at different time   Figure 26(c): Selectivity of NiMoS/AC catalyst at 623 K and 2000RPM at different time  01020304050600 20 40 60 80 100 120 140Selectivity, Wt%Time(min)1,1'-Bicyclohexyl1,1'-BicyclohexylBiphenylDibenzothiophene,1,2,3,4-tetrahydro-010203040506020 30 40 50 60 70 80 90 100Selectivity, Wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene, 1,2,3,4-tetrahydro-46   Figure 26(d): Selectivity of the thermal reaction at 623 K and 2000RPM at different time  3.3.2 Activity of the NiMoS/AC Catalyst for the HDS of DBT  Reactions were conducted at different temperatures and reaction times with temperatures varying between 588-638 K and reaction times of 30-120 minutes at fixed pressure of 4.83 MPa.   The experiments were designed to obtain the highest conversion in the most optimal reaction times. As shown in Figure 27, as the temperature increased the activity of NiMoS/AC catalyst for the hydrotreating of DBT increased and therefore the DBT conversion increased. 0102030405060700 20 40 60 80 100Selectivity, Wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene, 1,2,3,4-tetrahydro-47    Figure 27: Effect of reaction temperature on the DBT conversion with respect to time over NiMoS/AC  3.3.3 Activity of NiMoS/PC Catalyst for HDS of DBT  At the same reaction conditions and procedures stated above, experiments were conducted on the NiMoS/PC catalyst. As expected, the DBT conversion increased as temperature of reaction increased as shown in Figure 28. Interestingly, the highest DBT conversion was achieved at 623 K which indicated that petcoke is a promising support for the hydrotreating process because of the high dispersion of Ni and Mo on the petcoke support. No data could be obtained at 638 K because all the DBT was converted in less than 30 min.  01020304050607080901000 20 40 60 80 100 120 140Conversion%Time (min)588 K603 K623 K638 K48   Figure 28: Effect of reaction temperature on the DBT conversion with respect to time over NiMoS/PC  3.3.4 Comparison Between the Activity of NiMoS/AC and NiMoS/PC Figures 29 and 30 show the HDS of DBT for both catalysts at 603 K and 623 K. In both figures, NiMoS/PC has higher activity than NiMoS/AC for the HDS of DBT, as reflected in the DBT conversion. This is in agreement with the chemical surface compositions since the Ni/C and Mo/C atom ratio on petcoke is higher than on the AC support.  01020304050607080901000 20 40 60 80 100 120 140ConversionTime (min)588 K603 K623 K49   Figure 29: Comparison between NiMoS/AC and NiMoS/PC for HDS of DBT at 603 K and 4.8 MPa reaction condition   Figure 30: Comparison between NiMoS/AC and NiMoS/PC for HDS of DBT at 623 K and 4.8 MPa reaction condition  01020304050607080901000 30 60 90 120 150DBT Conversion%Time (min)NiMoS/ACNiMoS/PC01020304050607080901000 20 40 60 80 100DBT Conversion%Time (min)NiMoS/ACNiMoS/PC50  3.4 Product Distribution  The major DBT products were BP, THDBT, CHB, and BCH. Although other products were identified their quantities were < 10 ppm and are not considered further. The following section discusses the product distribution for the different supports.  3.4.1 Product Distribution over the NiMoS/AC Catalyst Figures 31-34 illustrate the product distribution of the HDS of DBT over NiMoS/AC at different temperatures. The selectivity to BP, which represents the DDS of DBT, varies between 43 wt% and 60 wt% at different reaction times and temperatures. At all temperatures the the selectivity to BP remained high as a function of reaction time, with only a small decline. In contrast, the selectivity towards 1, 2, 3, 4-THDBT significantly declined while there was a correspondingly significant increase in the selectivity to CHB with reaction time. These results are indicative of the hydrogenation of 1, 2, 3, 4-THDBT leading to the formation of CHB as shown, for instance, in Figure 34, with an insignificant hydrogenation of BP to CHB. This finding is in agreement with Figure 13 presented by Egorova and Prins22. The data also show that 1,1-bicyclohexyl was a minor product, requiring long reaction times (120 min) and higher temperature to be observed in the product. The maximum bicyclohexyl formation was 4 wt% at 638 K.  51   Figure 31: Selectivity of the product in HDS of DBT at 588K over NiMoS/AC as function of time   Figure 32: Selectivity of the product in HDS of DBT at 603K over NiMoS/AC as function of time 010203040506040 50 60 70 80 90 100 110 120 130Selectivity, wt%Time (min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene, 1,2,3,4-tetrahydro-01020304050600 10 20 30 40 50 60 70 80 90 100 110 120 130Selectivity, Wt%Time (min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-52   Figure 33: Selectivity of the product in HDS of DBT at 623K over NiMoS/AC as function of time   Figure 34: Selectivity of the product in HDS of DBT at 638K over NiMoS/AC as function of time  010203040506020 30 40 50 60 70 80 90 100Selectivity, Wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene, 1,2,3,4-tetrahydro-0102030405060700 10 20 30 40 50 60 70 80 90 100Selectivity, Wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-Biphenyl1,2,3,4THDBT53  3.4.2 Product Distribution over NiMoS/PC Figures 35-38 illustrate the product distributions of the HDS of DBT over NiMoS/PC at different temperatures. The results are very similar to those obtained for the NiMoS/AC catalyst. The selectivity to BP changes between 40 wt% and 60 wt% at different reaction times and temperature. As before, the selectivity towards 1, 2, 3, 4-THDBT dramatically decreased while the selectivity to CHB increased, consistent with the view that the formation of CHB is primarily from the hydrogenation of 1, 2, 3, 4-THDBT. This finding is in agreement with Figure 13 in the literature review. BCH was difficult to observe, requiring more than 120 min reaction time and higher temperature to be observed in the product. The maximum BCH formation was 6.8 wt% at 638 K.  Figure 35: Selectivity of the product in HDS of DBT at 588 K over NiMoS/PC as function of time  010203040506040 60 80 100 120 140Selectivity, Wt %Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-54   Figure 36: Selectivity of the product in HDS of DBT at 603 K over NiMoS/PC as function of time.   Figure 37: Selectivity of the product in HDS of DBT at 623 K over NiMoS/PC as function of time.  01020304050600 20 40 60 80 100 120 140 160 180 200Selectivity, Wt %Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-01020304050600 20 40 60 80 100 120 140Selectivity, Wt %Time(min)1,1'-Bicyclohexyl1,1'-BicyclohexylBiphenylDibenzothiophene,1,2,3,4-tetrahydro-55   Figure 38: Selectivity of the product in HDS of DBT at 638 K over NiMoS/PC as function of time.  3.4.3 Comparison Between the Selectivity of NiMoS/AC and NiMoS/PC Figures 39 and 40 show the selectivity of HDS of DBT reaction for both catalysts at 603 K and 623 K. Taking account of the experimental error associated with the measured selectivities (Appendix E), one concludes that there is minimal difference in selectivity between the NiMoS supported on petcoke versus the AC.  010203040506050 60 70 80 90 100 110 120Selectivity,Wt %Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-56   Figure 39: Comparison between NiMoS/AC and NiMoS/PC in term of selectivity towards DDS route at 603 K   Figure 40: Comparison between NiMoS/AC and NiMoS/PC in term of selectivity towards DDS route at 623 K  303540455055600 30 60 90 120 150Selectivity, Wt %Time(min)BP-NiMoS/ACBP-NiMoS/PC01020304050600 20 40 60 80 100Selectivity, Wt %Time(min)BP-NiMoS/ACBP-NiMoS/PC57  3.5 Summary of Findings  Among the two catalysts prepared on either the AC or the pretreated petcoke, the NiMoS/PC catalyst showed higher HDS activity, although product selectivities were the same.  Both catalysts had about 50% selectivity to DDS products. In terms of properties of the different supports, the XPS narrow scan analysis showed that the petcoke exhibited the highest concentration of metal on the surface, corresponding with the higher hydrotreating activity of the NiMoS/PC catalyst towards DBT, achieving 90% conversion at 623K compared with NiMoS/AC, which achieved 90% conversion at 638 K. In terms of selectivity, both catalysts showed comparable trends over the reaction times, with NiMoS/PC resulting in ~20% higher formation of CHB than the catalyst supported on NiMoS/AC at 638 K. The selectivity of the HDS of DBT measured in the current study are in an agreement with the literature.22 58  Chapter 4: HDS Reaction Kinetics of DBT over NiMoS/AC and NiMoS/PC 4.1 Reaction Mechanism  From Figure 25-40 along with the discussion of the results in the previous chapter, it is concluded that the HDS of DBT proceeds by two routes, the DDS route and the HYD route.  In the DDS route, the reaction starts by extracting S from DBT to produce BP which may then undergo further hydrogenation to form CHB.  However the product selectivity data suggest the hydrogenation of BP is relatively slow and it can therefore be ignored in the kinetic model.22 In the HYD route, the reaction starts with the hydrogenation of DBT to produce THDBT, followed by S removal from the THDBT to produce CHB. Further hydrogenation of CHB leads to BCH. The aforementioned conclusion is in agreement with several studies in the literature.3-6,21-23. The reaction pathways for the HDS of DBT are illustrated in Figure 41.43   59   Figure 41: Proposed reaction pathway of HDS of DBT. Copyright © 2006, Springer Science Business Media, Inc.43  4.2 Kinetic Development  From the liquid sample analysis, the concentrations of the model compound DBT and reaction products BP, CHB, THDBT, and BCH as shown in Figure 41 were determined as a function of time. Based on the reaction studies of Tye and Smith20 and Egorova and Prins 22,43, the rate constants for each step can be determined using a power law model assuming the reaction is 1st order and accounting for both the catalytic and non-catalytic reactions. The rate constant for the formation of CHB from BP through the DDS reaction route was eliminated as it is relatively slow and can be ignored.22 Hence for each of the reaction steps of Figure 41 we can write: 𝑟1= (𝑘1, 𝐶𝑐𝑎𝑡+ 𝑘1) 𝐶𝐷𝐵𝑇                                                     (1)                                                3 4 1 2 60  𝑟2= (𝑘2, 𝐶𝑐𝑎𝑡+ 𝑘2) 𝐶𝐷𝐵𝑇                                                      (2) 𝑟3= (𝑘2, 𝐶𝑐𝑎𝑡+ 𝑘2) 𝐶𝐷𝐵𝑇+ (𝑘3, 𝐶𝑐𝑎𝑡+ 𝑘3) 𝐶𝑇𝐻𝐷𝐵𝑇               (3) 𝑟4= (𝑘4, 𝐶𝑐𝑎𝑡+ 𝑘4) 𝐶𝐶𝐻𝐵                                                    (4) The rate of reaction has units of 𝑟𝑛=mol/(cm3s), where 𝑘𝑗,  is the catalytic rate constant with units of cm3/(gcats) and 𝑘𝑛 is the thermal or non-catalytic rate constant with unit of s-1. 𝐶𝑐𝑎𝑡   is the catalyst concentration in the reaction fluid (g/cm3) defined as the mass of the catalyst inside the reactor divided by the liquid volume (V).  4.3 Mole Balance  The general mole balance equation for any system is   {[𝑣𝑖 𝐶𝑖]-[ 𝑣𝑓  𝐶𝑖𝑓]}+ [∑ 𝛿𝑗𝑖𝑁𝑗=1 Ω 𝑎𝑗  𝑟𝑗]= [V 𝑑𝐶𝑖𝑑𝑡+𝐶𝑖  𝑑𝑉𝑑𝑡]               (5) where is 𝑣𝑖  the influent volumetric flow rate, 𝑣𝑓  is the effluent volumetric flow rate. V is the total volume, 𝐶𝑖 is the concentration of species i, 𝛿𝑖 is the stoichiometric coefficient of species i in reaction j, Ω is the overall effectiveness factor, 𝑎𝑗 is the catalyst activity, rj is rate of generation of component j, and N is the number of reactions. The mole balance equation can be simplified for the current study with the following assumptions:  There is no input and output: {[𝑣𝑖 𝐶𝑖]-[ 𝑣𝑓  𝐶𝑖𝑓]} = 0 because the unit is operated in batch mode. 61   A catalyst effectiveness of Ω = 1 since the catalyst particles are very small.   No catalysts deactivation:  𝑎𝑗=1  Liquid density changes are negligible: 𝑑𝑉𝑑𝑡 = 0 After applying the above assumptions the general mole balance equation is reduced to        𝑑𝐶𝑖𝑑𝑡= ∑ 𝑣𝑗𝑖𝑁𝑗=1 𝑟𝑗                                                         (6) After combining the mole balance equations with the reaction rate equations we obtain:   𝑑𝐶𝐷𝐵𝑇𝑑𝑡=-(𝑘1, 𝐶𝑐𝑎𝑡+ 𝑘1) 𝐶𝐷𝐵𝑇 - (𝑘2, 𝐶𝑐𝑎𝑡+ 𝑘2) 𝐶𝐷𝐵𝑇                                                       (7)                     𝑑𝐶𝐵𝑃𝑑𝑡 =  (𝑘1, 𝐶𝑐𝑎𝑡+ 𝑘1) 𝐶𝐷𝐵𝑇                                                                                        (8)                          𝑑𝐶𝑇𝐻𝐷𝐵𝑇𝑑𝑡 = (𝑘2, 𝐶𝑐𝑎𝑡+ 𝑘2) 𝐶𝐷𝐵𝑇 - (𝑘3, 𝐶𝑐𝑎𝑡+ 𝑘3) 𝐶𝑇𝐻𝐷𝐵𝑇                                              (9)                𝑑𝐶𝐶𝐻𝐵𝑑𝑡= (𝑘3, 𝐶𝑐𝑎𝑡+ 𝑘3) 𝐶𝑇𝐻𝐷𝐵𝑇 - (𝑘4, 𝐶𝑐𝑎𝑡+ 𝑘4) 𝐶𝐶𝐻𝐵                                                                 (10)      𝑑𝐶𝐵𝐶𝐻𝑑𝑡 = (𝑘4, 𝐶𝑐𝑎𝑡+ 𝑘4) 𝐶𝐶𝐻𝐵                                                                                    (11)  4.4 Parameter Estimation The parameters of the kinetic equations were estimated by minimizing the objective function using the sum of least-squares method. The objective function was defined as the sum of squares of the difference between the experimental concentrations and the model calculated concentration for each chemical species i at each reaction time t as shown in Equation 12:             OBJ=∑ ∑(𝐶𝑒𝑥𝑝,𝑖,𝑡 − 𝐶𝑝𝑟𝑒𝑑,𝑖,𝑡)2                                (12) The above equation was minimized using the Levenberg-Marquardt method implemented in MATLAB. At each iteration, the ordinary differential equations (ODEs) obtained in equations 62  7-10 were solved simultaneously using a Runge-Kutta 4th order numerical integration, except Equation 11 which was replaced by mass balance, in order to estimate the model calculated concentrations 𝐶𝐷𝐵𝑇, 𝐶𝑇𝐻𝐷𝐵𝑇, 𝐶𝐵𝑃, 𝐶𝐶𝐻𝐵, and 𝐶𝐵𝐶𝐻. A series of thermal experiments were conducted at each reaction temperature and their data were collected to estimate 𝑘1- 𝑘4 for the non-catalytic reaction using the aforementioned numerical method. The same procedures were followed to calculate the catalytic rate constants 𝑘1′ − 𝑘4′  for NiMoS/AC and NiMoS/PC catalysts. (Refer to appendix G for ODE MATLAB codes) According to Figure 25 and 26 (d) the thermal reaction becomes significant after 60 min at 623 K with the DDS reaction and HYD reactions most significant. Hence the thermal rate constants 𝑘1 and 𝑘2 in equation 1 and 2 were estimated for these data sets. Due to the low concentrations and selectivity of the other reactions and the significant error associated with the low concentration measurements, the model parameters 𝑘3 and 𝑘4 were ignored and set to zero.After determining the rate constants for the thermal reaction, the catalytic rate constants 𝑘1′ − 𝑘4′   in equations 7-11 were estimated.  The Arrhenius equation was used to obtain the reaction apparent activation barrier energy as follows:                        𝑘𝑗= 𝐴𝑗*exp(−𝐸𝑎𝑗𝑅∗𝑇)                              (13) The pre-exponential factors 𝐴𝑗 as well as the activation energy 𝐸𝑎𝑗 can be determined from the intercept and the slope of ln(𝑘𝑗) versus (1000𝑇) (Detailed calculations are given in Appendix H).Table 12 and 13 report the fitted parameter values of the thermal reactions. Table 14 - 17 report the parameter estimates for the NiMoS/AC and NiMoS/PC catalysts. The fit of the kinetic model is shown in Figures 42 and 43, with the degree of explanation (R2) determined to be 0.92 for both catalysts. This result is acceptable and the reaction fits the pseudo first order 63  reaction assumption. The Arrhenius plots are given in Figure 44 - 46. Note that 𝑘4′  is not plotted in the Arrhenius diagram of NiMoS/AC and NiMoS/PC (Figures 45, 46, respectively) because of the large parameter errors, a consequence of  the low concentration of the product BCH and few experimental data points where the concentration of this component was significant.  64                                                                                                                                                                                Figure 42: Measured (points) and model predicted (line) concentrations as function of time for NiMoS/AC catalyst at different temperatures. 65                                                                              Figure 43: Measured (points) and model predicted (line) concentrations as function of time for NiMoS/PC catalyst at different temperatures.66  Table 12: Estimated reaction rate constants for the thermal reaction of HDS of DBT Parameter                                                        Reaction temperature, K                        588 K                                   603 K                                        623 K                              638 K            Value  Std.    error Value  Std. error Value  Std.    error Value  Std.    error 𝑘1, 1/s 4.6E-04 1.08E-04 4.5E-04 2.9E-05 1.37E-03 7.6E-05 1.50E-03 8.058E-05 𝑘2, 1/s 2.47E-04 1.31E-04 2.65E-04 3.52E-05 6.33E-04 9.12E-05 6.05E-04 9.71E-05    Figure 44: Arrhenius plot of ln (kj) versus(1000/T) for the thermal HDS of DBT        67  Table 13: The pre-exponential factors 𝑨𝒋 and activation energies for all 𝒌𝟏 at the thermal reaction                 Parameter                     ln(𝑨𝒋), 1/s  Value Std. err 𝑬𝒂𝒋, kJ/mol std.err 𝑘1 13.5 ±6 105 ±32 𝑘2 8 ±5.3 81 ±27 68  Table 14: Estimated reaction rate constants for the HDS of DBT over NiMoS/AC at tested temperature    Figure 45: Arrhenius plot of ln (𝒌𝒋′) versus (1000/T) for all reaction temperature using NiMoS/AC catalyst.       Parameter                                                        Reaction temperature, K              588K                                   603K                                  623K                                 638K             Value Std.    error Value Std.    error Value Std.    error Value Std.    error 𝑘1′ , cm3/gcat*s 1.7E-01 5.9E-04 2.6E-01 3.5E-04 5.2E-01 7.1E-04 8.3E-01 1.4E-03 𝑘2′ , cm3/gcat*s 1.8E-01 7.4E-04 2.8E-01 4.4E-04 4.8E-01 8.4E-04 6.5E-01 1.4E-03 𝑘3′ , cm3/gcat*s 1.783 2.6E-02 2.493 1.5E-02 3.36 2.4E-02 4.811 4.9E-02 𝑘4′ , cm3/gcat*s 4.3E-02 3.4E0-3 6.81E-02 1.2E-03 4.54E-02 1.3E-03 8.337E-02 1.5E-03  69  Table 15: The pre-exponential factors 𝑨𝒋 and activation energies for all 𝒌𝒋′ at NiMoS/AC catalysts  Parameter ln(𝑨𝒋), cm3/gcat*s  Value Std. err Eaj, kJ/mol std. err 𝑘1′  19.37 ±0.92 104 ±5 𝑘2′  14.5 ±0.71 79 ±4 𝑘3′ 12.3 ±1.1 57 ±6                    70  Table 16: Estimated catalytic reaction rate constants for reaction of HDS for DBT over NiMoS/PC at tested temperature   1.60 1.62 1.64 1.66 1.68 1.70-1.8-1.6-1.4-1.2-1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.01.21.41.6  k3k2lnk1000/Tk1 Figure 46: The Arrhenius plot of ln (𝒌𝒋′) versus (1000/T) for all reaction temperature using NiMoS/PC catalyst.   Parameter                                                        Reaction temperature, K    588 K                                    603 K                                                   623 K  Value Std. err Value Std. err Value Std. error     𝑘1′ , cm3/gcat*s 1.822E-01 1.437E-04 5.40E-01 7.704E-04 9.65E-01 2.16E-03 𝑘2′ , cm3/gcat*s 2.17E-01 1.842E-04 5.40E-01 9.079E-04 1.02E+00 2.48E-03 𝑘3′ , cm3/gcat*s 2.212 7.206E-03 3.534 2.486E-02 4.72 4.27E-02 𝑘4′ , cm3/gcat*s 8.305E-02 6.905E-04 5.201E-02 8.074E-04 0 1.10E-03 71  Table 17: The pre-exponential factors 𝑨𝒋 and activation energies for all 𝒌𝒋′ at NiMoS/PC catalysts  Parameter ln(𝑨𝒋), cm3/gcat*s  Value std. err Eaj, kJ/mol std. err 𝑘1′  31.6 ±7.3 163 ±36 𝑘2′  28.22 ±4.5 146 ±22 𝑘3′  15.3 ±2.5 71 ±13  The H2S partial pressure in the reactor of the current study is between 150 – 300 kPa. Hence we can compare the rate of consumption of DBT (rDBT) in Table 3 at PH2S = 100 kPa with the rDBT in Tye’s work44 along with the rDBT in the current study as shown in Table 18. It is concluded that the rDBT in this study is lower by a magnitude of 10 compared to Prins’s work.  The lower rate observed in the present study is likely due to the fact the present study was conducted in the liquid phase at relatively high H2S partial pressure (>100 kPa) which significantly inhibits the DBT consumption rate of DBT according to Figure 15 (Appendix I provides detailed calculations of the phases present during reaction and the H2S partial pressure in the reactor). On the other hand rDBT for the current study is higher by a magnitude of 100 compared to Tye’s work. This difference is likely due to the significant effect of the promoter and the support used in the current study whereas Tye’s catalysts were neither promoted nor supported. Table 18: Comparison of DBT rates for the total DBT conversion at 623 K and 4.8 MPa with 𝐏𝐇𝟐𝐒= 100 kPa with literature data  Literature Catalysts 𝐫𝐃𝐁𝐓, mol/(g Mo s) Prins's work22 NiMoS/-Al2O3 0.05 Tye's work44 Unsupported MoS2 8.00E-05 Current study NiMoS/AC 0.004  NiMoS/PC 0.008  72  According to Table 19, the HDS of DBT reaction in Prins’s work prefers the DDS route over the HYD route as indicated by the rate constants. Whereas, in both catalysts in the current study, the relative rates of HYD versus DDS are approximately equal. The metal support interaction is a suggested reason for these findings. Since the support used in Prins’s work was -Al2O3, with a known ability to interact with the metal and form a different MoS2 stacking structure. In addition, a strong metal support interaction results in the precursor being converted to MoOx that is not easily reduced. In contrast, the carbon support is known to have a weak metal support interaction resulting in a uniform dispersion and hence better MoS2 formation along with reduced MoOx.31,32 XPS data showed good dispersion of NiMo on AC and petcoke indicating that the MoS2 are well dispersed, indicative of a high and probably equal distribution of edge and rim sites.  Hence the rate constants in both catalysts of both routes are statistically the same. Table 19: Comparison of rate of reaction for DBT total, DDS, HYD conversion at 623 K and 4.8 MPa  Rates  Prins's work NiMoS/γ-Al2O3 Current work NiMoS/AC Current work NiMoS/PC  rtot, mol/(g Mo s) 0.05 0.004 0.008 rDDS, mol/(g Mo s) 0.035 0.002 0.004 rHYD, mol/(g Mo s) 0.016 0.0018 0.0038  The activation energy is defined as the barrier energy between the reactants and products of each rate step. From Table 15 and 17, it is concluded that for the NiMoS/AC catalyst, the activation energy of the DDS route is significantly higher (>20 kJ/mol) than the activation energy for the HYD route. In the case of the NiMoS/PC catalyst, both routes have similar activation energy, taking account of the error associated with the activation energy estimates for these two routes.  73  The activation energy for NiMoS in this study can be compared to the data from Liu et al.24 and Tye and Smith44 for the HDS of DBT reaction at the same assumptions and conditions, by extracting the activation energy of Liu’s work from Figure 47 as reported in Table 20.   Figure 47: Arrhenius plot to calculate Ea from Liu et al. work on NiMo/AC Table 20: Comparison of activation energies for the total DBT conversion  Authors Type of Catalyst Ea (kJ/mol) Current work NiMoS/AC 83  NiMoS/PC 90 Liu et al.24 NiMo/AC 105 Tye and Smith44 Unsupported MoS2 149   Tye and Smith catalyst was unsupported and unpromoted but Liu et al. and the current study catalysts were all supported on activated carbon and petcoke: NiMo/AC, NiMoS/AC, NiMoS/PC respectively. The activation energies for the current study are the lowest (refer to Appendix A for detailed calculation) indicative that the NiMoS/AC and NiMoS/PC catalysts y = 2E+10e-12.63xR² = 0.99890.11101.72 1.74 1.76 1.78 1.8 1.82 1.84 1.86 1.88 1.9ln(ln(1-x))1000/T74  that were prepared in the current study have the lowest activation barrier, which means NiMoS/AC and NiMoS/PC have higher activity at lower temperature. Broderick et al.34 Vrinat et al.36, Liu et al24, and O'Brien et al31 reported the activation energies for the same reaction mechanism used in this study, using Co-Mo/-Al2O3 and NiMo/AC catalysts for Liu et al.24 as indicated in Table 5 in the literature review section. By comparing the values with the current study activation energies, it can be concluded that the present study’s activation energies are in agreement with the literature values in Table 5. This shows that the reaction mechanism developed in the current study fits the pseudo first order assumption.    75  Chapter 5: Conclusion and Recommendations 5.1 Conclusion  Petcoke, which was obtained from Canadian Oil sand from Alberta, was successfully treated in this study using KOH reagent. Petcoke after treatment has a surface area of 1140 m2/g and a pore volume of 0.67 cm3/g. The treated petcoke was used as a support for the NiMoS catalysts. Hence, the HDS activity and selectivity of 3 catalyst supports were compared: NiMoS/AC, NiMoS/PC, and NiMo/-Al2O3 using DBT as a model compound reactant. The reactions were conducted in a novel slurry-phase batch microreactor at different reaction times (30-120 min) and temperatures (588-638 K) and a fixed H2 pressure of 4.8 MPa.  The NiMoS catalysts used in this study showed higher dispersion on petcoke compared to AC. The results also showed that NiMoS/PC had a promising activity and selectivity when compared with the optimized NiMo/-Al2O3 catalyst, indicative that petcoke can be used as a promising support for catalysts. The HDS of DBT proceeded by two routes: the DDS reaction route and the HYD reaction route.  The pseudo 1st order kinetic power law provided a suitable approximation of the reaction kinetics of the HDS reactions. The rate constants for both DDS and HYD routes showed similar magnitude on both NiMoS/AC and NiMoS/PC catalysts, whereas for the apparent activation energy, the DDS route activation energy was higher than the HYD route activation energy on the NiMoS/AC catalyst, but they were statistically equal on the NiMoS/PC catalyst. The key contributions of this study are firstly, demonstrating that raw petcoke can be converted into a useful catalyst support for hydroteating processes. Secondly, after examining the catalyst supported on petcoke for S removal from DBT and obtaining a relatively high 76  activity, we can conclude that the treated petcoke is a promising support for hydrotreating catalysts. 5.2 Recommendations  The experiments reported herein were done at relatively high concentrations of H2S, which is known to suppress catalytic activity. Additional experiments are needed using both NiMoS/AC and NiMoS/PC catalyst at lower H2S partial pressure and in the gas phase to assess the performance of these catalysts at conditions comparable to literature reports.  The carbon supported  catalysts need to be assessed using a real feedstock for a period of at least one month to observe how the catalysts behave in terms of activity, selectivity, and deactivation under industrial conditions. The present study limited the catalyst activity measurements to the HDS of DBT. Use of more refractory model compounds such as 4,6-DMDBT will provide further clarification as to the relative activity and selectivity of the catalysts. The petcoke support properties can be influenced by varying the pretreatment processes and conditions, which in turn will impact the catalyst performance.  Further studies with petcoke prepared under different conditions to control pore volume, pore size and surface area are needed. The kinetic modeling can be improved by using CHB as a reactant over both catalysts to produce BCH. Data form the present study showed that it is difficult to form BCH at the reaction conditions studied.       77  Bibliography 1. greencarcongress. EPA proposes tier 3 standards for gasoline sulfur content and vehicle emissions; harmonized with california LEV III   . http://www.greencarcongress.com/2013/03/tier3-20130329.html. Accessed October/22, 2015. 2. UOP Honeywell company. Sulphur content in diesel fuel. http://www.honeywell-uop.cn/more-diesel-reflect-changing-market. Updated 2015. Accessed 11/01, 2015. 3. Sakanishi K, Nagamatsu T, Mochida I, Whitehurst DD. Hydrodesulfurization kinetics and mechanism of 4,6-dimethyldibenzothiophene over NiMo catalyst supported on carbon. J Mol Catal A: Chem. 2000;155(1-2):101-109. 4. Ali SA, Ahmed S, Ahmed KW, Al-Saleh MA. Simultaneous hydrodesulfurization of dibenzothiophene and substituted dibenzothiophene over phosphorus modified CoMo/-Al2O3 catalysts. Fuel Process Technol. 2012;98:39-44. 5. Shi Y, Chen J, Chen J, MacLeod RA, Malac M. Preparation and evaluation of hydrotreating catalysts based on activated carbon derived from oil sand petroleum coke. Appl Catal , A. 2012;441-442:99-107. 6. Shi G, Zhao Y, Huang Y, Shen J. Mesoporous carbon supported Co-Mo and Ni-Mo catalysts for hydrodesulfurization. Cuihua Xuebao. 2010;31(8):961-964. 7. Milton Beychok. Hydrodesulfurization. http://www.eoearth.org/view/article/171121/. Updated 2013. Accessed Nov. 22, 2014. 78  8. Hill JM, Karimi A, Malekshahian M. Characterization, gasification, activation, and potential uses for the millions of tonnes of petroleum coke produced in canada each year. The Canadian Journal of Chemical Engineering. 2014;92(9):1618-1626. 9. R. Murray Gray. Upgrading petroleum residues and heavy oils. . 1994:244. 10. DiPanfilo R, Egiebor NO. Activated carbon production from synthetic crude coke. Fuel Process Technol. 1996;46(3):157-169. 11. PEC Consulting Group. Comparative properties of bituminous coal and petroleum coke as fuels in cement kilns. http://pecconsultinggroup.com/newsflash/comparative-properties-of-coal-and-petcoke. Accessed october/8, 2015. 12. PEC Consulting Group. Comparative properties of bituminous coal and petroleum coke as fuels in cement kilns. http://pecconsultinggroup.com/newsflash/comparative-properties-of-coal-and-petcoke. Updated 2015. Accessed March, 03, 2015. 13. Liang C, Li Z, Dai S. Mesoporous carbon materials: Synthesis and modification. Angew Chem , Int Ed. 2008;47(20):3696-3717. 14. Shen WZ, Zheng JT, Zhang YL, Wang JG, Qin ZF. Preparation of mesoporous carbon by steam activation of commercial activated carbon in the presence of yttrium oxide. Studies in Surface Science and Catalysis. 2003;146:767-770. 15. Marsh H, Yan DS, O'Grady TM, Wennerberg A. Formation of active carbons from cokes using potassium hydroxide. Carbon. 1984;22(6):603-611. doi: http://dx.doi.org/10.1016/0008-6223(84)90096-4. 79  16. Karimi A, Thinon O, Fournier J, Hill JM. Activated carbon prepared from canadian oil sands coke by CO2 activation: I. trends in pore development and the effect of pre-oxidation. The Canadian Journal of Chemical Engineering. 2013;91(9):1491-1499. 17. Rodríguez-reinoso F. The role of carbon materials in heterogeneous catalysis. Carbon. 1998;36(3):159-175. 18. Marsh H, Rodríguez-Reinoso F. Chapter 2 - activated carbon (origins). In: Rodríguez-Reinoso HM, ed. Activated carbon. Oxford: Elsevier Science Ltd; 2006:13-86. http://dx.doi.org/10.1016/B978-008044463-5/50016-9. 19. Glasson C, Geantet C, Lacroix M, Labruyere F, Dufresne P. Beneficial effect of carbon on hydrotreating catalysts. Journal of Catalysis. 2002;212(1):76-85. 20. Vázquez P, Pizzio L, Blanco M, et al. NiMo(W)-based hydrotreatment catalysts supported on peach stones activated carbon. Applied Catalysis A: General. 1999;184(2):303-313. 21. Shu Y, Oyama ST. Synthesis, characterization, and hydrotreating activity of carbon-supported transition metal phosphides. Carbon. 2005;43(7):1517-1532. 22. Egorova M, Prins R. Hydrodesulfurization of dibenzothiophene and 4,6-dimethyldibenzothiophene over sulfided NiMo/γ-Al2O3, CoMo/γ-Al2O3, and Mo/γ-Al2O3 catalysts. Journal of Catalysis. 2004;225(2):417-427. 23. Kouzu M, Kuriki Y, Hamdy F, Sakanishi K, Sugimoto Y, Saito I. Catalytic potential of carbon-supported NiMo-sulfide for ultra-deep hydrodesulfurization of diesel fuel. Appl Catal , A. 2004;265(1):61-67. 80  24. Liu F, Xu S, Chi Y, Xue D. A novel alumina-activated carbon composite supported NiMo catalyst for hydrodesulfurization of dibenzothiophene. Catalysis Communications. 2011;12(6):521-524. 25. - Huang Y, - Hu S, - Zuo S, - Xu Z, - Han C, - Shen J. - Mesoporous carbon materials prepared from carbohydrates with a metal chloride template. - J Mater Chem. (- 41):- 7759. 26. Sakanishi K, Hasuo H, Mochida I, Okuma O. Preparation of highly dispersed NiMo catalysts supported on hollow spherical carbon black particles. Energy Fuels. 1995;9(6):995-8. 27. Egorova M, Prins R. Mutual influence of the HDS of dibenzothiophene and HDN of 2-methylpyridine. Journal of Catalysis. 2004;221(1):11-19. 28. Egorova M, Prins R. Competitive hydrodesulfurization of 4,6-dimethyldibenzothiophene, hydrodenitrogenation of 2-methylpyridine, and hydrogenation of naphthalene over sulfided NiMo/γ-Al2O3. Journal of Catalysis. 2004;224(2):278-287. 29. Byskov LS, Nørskov JK, Clausen BS, Topsøe H. DFT calculations of unpromoted and promoted MoS2-based hydrodesulfurization catalysts. Journal of Catalysis. 1999;187(1):109-122. 30. Schweiger H, Raybaud P, Toulhoat H. Promoter sensitive shapes of Co(Ni)MoS nanocatalysts in sulfo-reductive conditions. Journal of Catalysis. 2002;212(1):33-38. 31. Daage M, Chianelli RR. Structure-function relations in molybdenum sulfide catalysts: The "rim-edge" model. Journal of Catalysis. 1994;149(2):414-427. 81  32. Hensen EJM, Kooyman PJ, van der Meer Y, et al. The relation between morphology and hydrotreating activity for supported MoS2 particles. Journal of Catalysis. 2001;199(2):224-235. 33. Vanrysselberghe V, Froment GF. Hydrodesulfurization of dibenzothiophene on a CoMo/-Al2O3 catalyst: Reaction network and kinetics. Ind Eng Chem Res. 1996;35(10):3311-3318. 34. Broderick DH, Sapre AV, Gates BC, Kwart H, Schuit GCA. Hydrogenation of aromatic compounds catalyzed by hydrogen sulfide-cobalt(II) oxide-molybdenum oxide/alumina. J Catal. 1982;73(1):45-49. 35. Vanrysselberghe V, Le Gall R, Froment GF. Hydrodesulfurization of 4-methyldibenzothiophene and 4,6-dimethyldibenzothiophene on a CoMo/-Al2O3 catalyst: Reaction network and kinetics. Ind Eng Chem Res. 1998;37(4):1235-1242. 36. Vrinat. M.L and de Mourgues. L. Kinetic study of DBT on CoMo/-Al2O3 catalyst. . 1982;45:79. 37. O'Brien. W, Chen. R, Carr. G. Kinetics study of DBT HDS over CoMo/-Al2O3. 1979;661:76. 38. Campanati M, Fornasari G, Vaccari A. Fundamentals in the preparation of heterogeneous catalysts. Catalysis Today. 2003;77(4):299-314. 39. Kawano T, Kubota M, Onyango MS, Watanabe F, Matsuda H. Preparation of activated carbon from petroleum coke by KOH chemical activation for adsorption heat pump. Appl Therm Eng. 2008;28(8–9):865-871. 82  40. Kukard R. Catalytic hydroconversion of diphenylmethane with unsupported MoS2. [PhD]. university of british columbia; 2014. 41. Kukard RS, Smith KJ. Slurry-phase batch microreactor for hydroconversion studies. Energy Fuels. 2015;29(8):5274-5281. 42. Whiffen VML, Smith KJ. Hydrodeoxygenation of 4-methylphenol over unsupported MoP, MoS2, and MoOx catalysts. Energy Fuels. 2010;24(9):4728-4737. 43. Tye CT, Smith KJ. Catalytic activity of exfoliated MoS2 in hydrodesulfurization, hydrodenitrogenation and hydrogenation reactions. Top Catal. 2006;37(2-4):129-135. 44. Tye CT, Smith KJ. Catalytic activity of exfoliated MoS2 in hydrodesulfurization, hydrodenitrogenation and hydrogenation reactions. Top Catal. 2006;37(2-4):129-135.    83  Appendices                     84  Appendix A   Calculation of Activation Energies from the Data of Liu et al.24 After extracting the conversions from Figure 5 of each catalyst and fitting them with the temperatures, we assume 1st-order kinetics so that k = (1/)(-ln(1-X), and hence a plot of lnk versis 1/T will yield a slope proportional to the Ea. the activation energies will be  NiMoS/-Al2O3 Table 21: Conversions and corresponding temperature for NiMoS/γ-Al2O3 x ln(1-x) T 1000/T 0.32 0.385662481 260 1.876173 0.59 0.891598119 270 1.841621 0.78 1.514127733 280 1.808318 0.91 2.407945609 290 1.776199 0.98 3.912023005 300 1.745201   Figure 48: Arrhenius plot to calculate Ea for NiMoS/γ-Al2O3 catalyst Ea = -8.314*-17.22 =143.16 kJ/mol   y = 5E+13e-17.22xR² = 0.98780.11101.72 1.74 1.76 1.78 1.8 1.82 1.84 1.86 1.88 1.9Ln(ln(1-x))Temperature 1000/T (1000/K)85   NiMoS/AC Table 22: Conversions and corresponding temperature for NiMoS/AC x ln(1-x) T 1000/T 0.58 0.867500568 260 1.876173 0.74 1.347073648 270 1.841621 0.86 1.966112856 280 1.808318 0.95 2.995732274 290 1.776199 0.99 4.605170186 300 1.745201   Figure 49: Arrhenius plot to calculate Ea for NiMoS/AC catalyst   Ea = -8.314*-12.63  =105 kJ/mol   y = 2E+10e-12.63xR² = 0.99890.11101.72 1.74 1.76 1.78 1.8 1.82 1.84 1.86 1.88 1.9Ln(Ln(1-x))Temperature 1000/T (1000/K)86   NiMoS/AAC  Table 23: Conversions and corresponding temperature for NiMoS/AAC x ln(1-x) T 1000/T 0.9 2 260 1.9 0.97 4 270 1.8 0.99 5 280 1.8 0.999 7 290 1.8   Figure 50: Arrhenius plot to calculate Ea for NiMoS/AAC catalyst  Ea = -8.314*10.71 = 89 kJ/mol      y = 1E+09e-10.71xR² = 0.99381101.8 1.8 1.8 1.8 1.8 1.9 1.9 1.9Ln(Ln(1-x))Temperature 1000/T (1000/K)87   Current work NiMoS/AC Table 24: Conversions and corresponding temperature for NiMoS/AC                  Ea = -8.314*10.04 = 83 kJ/mol   x ln(1-x) 1000/T 0.61 0.94 1.70 0.71 1.24 1.66 0.912 2.43 1.61 y = 2E+07e-10.04xR² = 0.97020.101.0010.001.60 1.62 1.64 1.66 1.68 1.70 1.72Ln(Ln(1-x))1000/TFigure 51: Arrhenius plot to calculate Ea for NiMoS/AC catalyst for the current study 88   NiMoS/PC Table 25: Conversions and corresponding temperature for NiMoS/PC x ln(1-x) 1000/T 0.6547 1.063342 1.700 0.97 3.506558 1.605 0.99 4.60517 1.560   Figure 52: Arrhenius plot to calculate Ea for NiMoS/PC catalyst for the current study Ea = -8.314*10.79 = 90 kJ/mol        y = 1E+08e-10.79xR² = 0.97891101.540 1.560 1.580 1.600 1.620 1.640 1.660 1.680 1.700 1.720Ln(Ln(1-x))1000/T89  Appendix B  The Kinetic Model for HDS in Vanrysselberghe and Froment33  The study was developed based on σ and τ, the two different types of active sites for hydrogenolysis and hydrogenation reactions. A Langmuir-Hinshelwood mechanism was developed for both types of reactions as described below:    90   91  The rate expressions were obtained as follow                92  Appendix C  Catalyst Characterization  C.1 BET Surface Area Calculations The N2 adsoprtion isotherm is measured at 77K using the Micromeritics ASAP 2020.  The volume of N2 adsorpbed as a function of pressure ratio (P/p0) is measured.  Hence, the BET surface area can be estimated through the following equation 𝑃𝑃𝑜⁄𝑉(1 −𝑃𝑃𝑜)=  𝐶 − 1𝑉𝑚𝐶(𝑃𝑃𝑜) +1𝑉𝑚𝐶 Where 𝑃 is the equilibrium pressure             𝑃𝑜 is the saturation pressure of the adsorbate pressure            𝑉 is the adsorbed volume             𝑉𝑚 is the adsorbed volume at the monolayer coverage             𝐶 is constant                      93  C.2 BET Analysis for Activated Carbon Before Impregnation  Full Report Set  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 1  Sample: activated carbon new bottle (again) Operator: Majed Submitter: UBC File: C:\...\MAJED\000-175.SMP  Started: 01/05/2014 6:02:29PM Analysis Adsorptive: N2 Completed: 01/05/2014 8:44:23PM Analysis Bath Temp.: -195.850 °C Report Time: 02/05/2014 12:21:33PM Thermal Correction: No Sample Mass: 0.1200 g Warm Free Space: 28.4000 cm³ Measured Cold Free Space: 90.4938 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes     Summary Report  Surface Area  Single point surface area at p/p° = 0.312549128: 1326.0480 m²/g  BET Surface Area: 1322.5673 m²/g  t-Plot Micropore Area: 652.5297 m²/g  t-Plot External Surface Area: 670.0375 m²/g  BJH Adsorption cumulative surface area of pores between 1.7000 nm and 300.0000 nm diameter: 436.478 m²/g  Pore Volume  t-Plot micropore volume: 0.349497 cm³/g                                    BJH Adsorption cumulative volume of pores                                    between 1.7000 nm and 300.0000 nm diameter: 0.241134 cm³/g  Pore Size                                BJH Adsorption average pore diameter (4V/A): 2.2098 nm       94  Full Report Set  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 2  Sample: activated carbon new bottle (again) Operator: Majed Submitter: UBC File: C:\...\MAJED\000-175.SMP  Started: 01/05/2014 6:02:29PM Analysis Adsorptive: N2 Completed: 01/05/2014 8:44:23PM Analysis Bath Temp.: -195.850 °C Report Time: 02/05/2014 12:21:33PM Thermal Correction: No Sample Mass: 0.1200 g Warm Free Space: 28.4000 cm³ Measured Cold Free Space: 90.4938 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes       Isotherm Tabular Report  Relative Absolute Quantity Elapsed Time Saturation Pressure (p/p°) Pressure Adsorbed (h:min) Pressure   (mmHg) (cm³/g STP)   (mmHg)                01:22 755.363525 0.049434026 37.340660 356.1179 01:50  0.120518469 91.035255 391.4967 01:59  0.181500324 137.098724 410.8322 02:06  0.245724202 185.611099 427.4877 02:12  0.312549128 236.088211 443.1075 02:19     Full Report Set  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 3  Sample: activated carbon new bottle (again) Operator: Majed Submitter: UBC File: C:\...\MAJED\000-175.SMP  Started: 01/05/2014 6:02:29PM Analysis Adsorptive: N2 Completed: 01/05/2014 8:44:23PM Analysis Bath Temp.: -195.850 °C Report Time: 02/05/2014 12:21:33PM Thermal Correction: No Sample Mass: 0.1200 g Warm Free Space: 28.4000 cm³ Measured Cold Free Space: 90.4938 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes    95   450 activated carbon - Adsorption            400        350       STP) 300              Adsorbed (cm³/g 250       200       Quantity 150               100        50        0        0.00 0.05 0.10 0.15 0.20 0.25 0.30                                                                          Relative Pressure (p/p°)                Figure 53: Isotherm Linear Plot  C.3 BET Analysis for Activated Carbon After Impregnation   ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 1  Sample: Majed active carrbon with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-213.SMP  Started: 10/2/2015 8:14:25PM Analysis Adsorptive: N2 Completed: 10/3/2015 3:40:00AM Analysis Bath Temp.: 77.332 K Report Time: 10/3/2015 12:25:48PM Thermal Correction: No Sample Mass: 0.2532 g Warm Free Space: 27.9507 cm³ Measured Cold Free Space: 87.7548 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes Summary Report  96  Surface Area                            Single point surface area at p/p° = 0.200429005: 145 .1785 m²/g  BET Surface Area: 151.5212 m²/g  Langmuir Surface Area: 210.7268 m²/g  t-Plot Micropore Area: 7.4690 m²/g  t-Plot External Surface Area: 144.0522 m²/g                                  BJH Adsorption cumulative surface area of pores between 1.7000 nm and 300.0000 nm diameter: 137.813 m²/g                                   BJH Desorption cumulative surface area of pores between 1.7000 nm and 300.0000 nm diameter: 155.1126 m²/g  Pore Volume  Single point adsorption total pore volume of pores less than 69.8048 nm diameter at p/p° = 0.971478049 : 0.224022 cm³/g  t-Plot micropore volume: 0.001476 cm³/g  BJH Adsorption cumulative volume of pores between 1.7000 nm and 300.0000 nm diameter: 0.279339 cm³/g  BJH Desorption cumulative volume of pores between 1.7000 nm and 300.0000 nm diameter: 0.291424 cm³/g  Pore Size  Adsorption average pore width (4V/A by BET): 5.91395 nm  BJH Adsorption average pore diameter (4V/A): 8.1078 nm  BJH Desorption average pore diameter (4V/A): 7.5152 nm 97  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 2  Sample: Majed active carrbon with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-213.SMP  Started: 10/2/2015 8:14:25PM Analysis Adsorptive: N2 Completed: 10/3/2015 3:40:00AM Analysis Bath Temp.: 77.332 K Report Time: 10/3/2015 12:25:48PM Thermal Correction: No Sample Mass: 0.2532 g Warm Free Space: 27.9507 cm³ Measured Cold Free Space: 87.7548 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes       Isotherm Tabular Report  Relative Absolute Quantity Elapsed Time Saturation Pressure (p/p°) Pressure Adsorbed (h:min) Pressure   (mmHg) (mmol/g)   (mmHg)                00:56 758.652283 0.010382216 7.875463 1.08181 01:25  0.032893834 24.950260 1.28613 01:38  0.063011019 47.792900 1.43257 01:45  0.078819094 59.781738 1.49573 01:50  0.100150859 75.959824 1.57006 01:54  0.120399813 91.316078 1.63536 01:58  0.140421447 106.499374 1.69537 02:02  0.160496667 121.722763 1.75284 02:06  0.180567602 136.942352 1.80785 02:10  0.200429005 152.002487 1.86087 02:14  0.248805871 188.687408 1.98238 02:18  0.302402132 229.329178 2.11621 02:22  0.352944079 267.653259 2.24424 02:26  0.400619577 303.800873 2.36924 02:31  0.450603358 341.698792 2.50491 02:35  0.500351338 379.414825 2.64792 02:40  0.550429819 417.381653 2.80260 02:44  0.600453128 455.303223 2.97111 02:49  0.650656097 493.357086 3.15840 02:55  0.700422891 531.080566 3.36770 03:00        03:02 758.221619 0.750149216 568.770081 3.61392 03:07  0.799867692 606.453247 3.90913 03:14  0.821166088 622.591370 4.05878 03:19  0.850433824 644.771057 4.29215 03:24  0.875074646 663.439941 4.52301 03:30  0.899858366 682.216431 4.80343 03:36  0.924451643 700.845520 5.15902 03:43  0.948511361 719.064575 5.65505 03:52  0.971478049 736.449158 6.46154 04:03  0.982321010 744.647034 7.05668 04:12  0.991222915 751.370605 7.66950 04:22  0.995201275 754.361694 8.08246 04:32  0.983216772 745.260437 7.80971 04:39  0.974600286 738.710022 7.50400 04:47  0.949100563 719.339966 6.49375 05:05        05:07 757.912598 0.918061725 695.810547 5.67629 05:19  98  0.906172026 686.799194 5.47083 05:26  0.879712248 666.744995 5.10673 05:33  0.850498001 644.603149 4.79631 05:41  0.826554563 626.456116 4.58688 05:46  0.800885917 607.001526 4.38956 05:51  0.750738949 568.994507 4.06545 05:58     ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 3  Sample: Majed active carrbon with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-213.SMP  Started: 10/2/2015 8:14:25PM Analysis Adsorptive: N2 Completed: 10/3/2015 3:40:00AM Analysis Bath Temp.: 77.332 K Report Time: 10/3/2015 12:25:48PM Thermal Correction: No Sample Mass: 0.2532 g Warm Free Space: 27.9507 cm³ Measured Cold Free Space: 87.7548 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes                           Isotherm Tabular Report  Relative Absolute Quantity Elapsed Time Saturation Pressure (p/p°) Pressure Adsorbed (h:min) Pressure   (mmHg) (mmol/g)   (mmHg)          0.700553889 530.958618 3.79193 06:05  0.650729849 493.196350 3.55783 06:10  0.600612677 455.211914 3.34915 06:16  0.550798783 417.457336 3.16189 06:21  0.500647104 379.446747 2.98217 06:26  0.455938729 345.561707 2.67826 06:35  0.392350076 297.367065 2.42916 06:40  0.337293922 255.639313 2.28509 06:45  0.303055210 229.689362 2.19977 06:49  0.250721433 190.024933 2.07105 06:53  0.200388298 151.876816 1.94583 06:57  0.143300176 108.609009 1.79527 07:01  99  Figure 54: Isotherm Linear Plot  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 4  Sample: Majed active carrbon with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-213.SMP  Started: 10/2/2015 8:14:25PM Analysis Adsorptive: N2 Completed: 10/3/2015 3:40:00AM Analysis Bath Temp.: 77.332 K Report Time: 10/3/2015 12:25:48PM Thermal Correction: No Sample Mass: 0.2532 g Warm Free Space: 27.9507 cm³ Measured Cold Free Space: 87.7548 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes   Majed active carrbon with cata.2015 - Adsorption Majed active carrbon with cata.2015 - Desorption   8            7            6           (mmol/g) 5                      Quantity Adsorbed 4           3            2            1            0            0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0   100  C.4 BET Analysis for Petcoke Before Impregnation  Quantachrome NovaWin2 - Data Acquisition and Reduction  for NOVA instruments  ©1994-2006, Quantachrome Instruments   version 2.2   Analysis  Report   Operator:1 Date:2013/07/08 Operator:1 Date:9/18/2015 Sample ID: Nova 2200E Filename: D:\QCDATA\Physisorb\Majed sample from Dr Sharif.qps Sample Desc:  Comment:   Sample weight: 0.0749 g Sample Volume: 0 cc  Outgas Time: 3.0 hrs OutgasTemp:  300.0 C  Analysis gas: Nitrogen Bath Temp: 77.3 K   Press. Tolerance:0.100/0.100 (ads/des)Equil time: 60/60 sec (ads/des) Equil timeout: 240/240 sec (ads/des)  Analysis Time: 467.4 min End of run:  2013/07/08 16:11:16 Instrument:  Nova Station A Cell ID: 33   Adsorbate Nitrogen Temperature 77.350K Molec. Wt.: 28.013 g  Cross Section: 16.200 Ų  Liquid Density: 0.808 g/cc Relative Volume@STP 1 / [ W((Po/P) - 1) ] Pressure   P/Po cc/g  6.27140e-02 373.1782 1.4346e-01 8.53920e-02 380.7947 1.9617e-01 1.05149e-01 385.5174 2.4387e-01 1.25188e-01 389.1453 2.9423e-01 1.42403e-01 391.6825 3.3920e-01 101  1.66811e-01 394.6003 4.0595e-01 BET summary  Slope = 3.155 Intercept = -9.918e-02 Correlation coefficient, r = 0.995518 C constant= -30.810 Surface Area = 1139.612 m²/g 1.85318e-01 396.4666 4.5906e-01 2.05435e-01 398.1982 5.1951e-01 2.26038e-01 399.7163 5.8460e-01 2.46790e-01 401.1205 6.5356e-01 2.67668e-01 402.5154 7.2653e-01 2.88879e-01 403.6585 8.0521e-01 3.08826e-01 404.9227 8.8289e-01 3.29542e-01 405.9670 9.6872e-01 3.49814e-01 407.1694 1.0572e+00 102  Analysis  Report   Operator:1 Date:2013/07/08 Operator:1 Date:9/18/2015 Sample ID: Nova 2200E Filename: D:\QCDATA\Physisorb\Majed sample from Dr Sharif.qps Sample Desc:  Comment:   Sample weight: 0.0749 g Sample Volume: 0 cc  Outgas Time: 3.0 hrs OutgasTemp:  300.0 C  Analysis gas: Nitrogen Bath Temp: 77.3 K   Press. Tolerance:0.100/0.100 (ads/des)Equil time: 60/60 sec (ads/des) Equil timeout: 240/240 sec (ads/des)  Analysis Time: 467.4 min End of run:  2013/07/08 16:11:16 Instrument:  Nova Station A Cell ID: 33  Adsorbate Nitrogen Temperature  77.350K Molec. Wt.: 28.013 g Cross Section: 16.200 Ų  Liquid Density: 0.808 g/cc  Total Pore Volume summary  Total Pore Volume  Total pore volume = 6.647e-01 cc/g for pores smaller than -33116.1 Å (Radius)  at P/Po = 1.00029 V-t method summary  Thickness method: DeBoer  Slope = 3.409  Intercept = 381.600  Correlation coefficient, r = 0.747996  Micropore volume = 0.590 cc/g  Micropore area = 1086.874 m²/g  External surface area = 52.737 m²/g     103  C.5 BET Analysis for Petcoke After Impregnation   ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 1  Sample: Majed petcoke with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-214.SMP  Started: 10/3/2015 12:43:10PM Analysis Adsorptive: N2 Completed: 10/3/2015 5:41:51PM Analysis Bath Temp.: 77.333 K Report Time: 10/4/2015 2:58:26PM Thermal Correction: No Sample Mass: 0.1786 g Warm Free Space: 28.0776 cm³ Measured Cold Free Space: 90.0115 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes    Summary Report  Surface Area  Single point surface area at p/p° = 0.201020610: 37. 1338 m²/g  BET Surface Area: 37.7058 m²/g  Langmuir Surface Area: 51.5974 m²/g  t-Plot Micropore Area: 16.5836 m²/g  t-Plot External Surface Area: 21.1222 m²/g  BJH Adsorption cumulative surface area of pores between 1.7000 nm and 300.0000 nm diameter: 7.325 m²/g  BJH Desorption cumulative surface area of pores between 1.7000 nm and 300.0000 nm diameter: 0.3850 m²/g  Pore Volume  Single point adsorption total pore volume of pores less than 79.0758 nm diameter at p/p° = 0.974895587 : 0.019137 cm³/g  t-Plot micropore volume: 0.007163 cm³/g  BJH Adsorption cumulative volume of pores between 1.7000 nm and 300.0000 nm diameter: 0.009936 cm³/g  BJH Desorption cumulative volume of pores between 1.7000 nm and 300.0000 nm diameter: 0.004822 cm³/g  Pore Size  Adsorption average pore width (4V/A by BET): 2.03012 nm  BJH Adsorption average pore diameter (4V/A): 5.4258 nm  BJH Desorption average pore diameter (4V/A): 50.1030 nm 104  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 2  Sample: Majed petcoke with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-214.SMP  Started: 10/3/2015 12:43:10PM Analysis Adsorptive: N2 Completed: 10/3/2015 5:41:51PM Analysis Bath Temp.: 77.333 K Report Time: 10/4/2015 2:58:26PM Thermal Correction: No Sample Mass: 0.1786 g Warm Free Space: 28.0776 cm³ Measured Cold Free Space: 90.0115 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes       Isotherm Tabular Report  Relative Absolute Quantity Elapsed Time Saturation Pressure (p/p°) Pressure Adsorbed (h:min) Pressure   (mmHg) (mmol/g)   (mmHg)                00:57 758.769897 0.011118213 8.431630 0.28729 01:52  0.033384891 25.313129 0.34831 02:11  0.067763153 51.374428 0.39892 02:21  0.080428217 60.974022 0.41285 02:25  0.100738627 76.369453 0.42930 02:28  0.120600228 91.423752 0.44244 02:31  0.140728105 106.679016 0.45322 02:34  0.160837576 121.919434 0.46238 02:37  0.180754113 137.012711 0.46967 02:40  0.201020610 152.371857 0.47633 02:42  0.250876640 190.156708 0.48278 02:45  0.301199392 228.293091 0.48648 02:48  0.350938401 265.984802 0.48760 02:51  0.420191391 318.463989 0.48600 02:54  0.470328854 356.452820 0.48558 02:57        03:00 757.857727 0.520108618 394.168335 0.48451 03:02  0.569574335 431.656311 0.48286 03:05  0.619481835 469.479095 0.47978 03:08  0.669333523 507.259583 0.47632 03:10  0.719225922 545.070923 0.47339 03:12  0.769237113 582.972290 0.47336 03:15  0.819302746 620.914917 0.47456 03:19  0.820136057 621.546448 0.47843 03:22  0.850385298 644.471069 0.48072 03:24  0.893875937 677.430786 0.48552 03:27  0.900123309 682.165405 0.48963 03:29  0.925090811 701.087219 0.49913 03:32  0.949866877 719.863953 0.51564 03:35  0.974895587 738.832153 0.55197 03:37  0.980442533 743.035950 0.57231 03:41  0.990431789 750.606384 0.61668 03:43  0.994777937 753.900146 0.65891 03:46  0.973572051 737.829102 0.58787 03:49  0.932772376 706.908752 0.54413 03:52  0.906777068 687.208008 0.53522 03:55  0.881638427 668.156494 0.53232 03:58  0.856448645 649.066223 0.53198 04:00  105  0.831387560 630.073486 0.53269 04:03  0.806519118 611.226746 0.53544 04:06  0.781294947 592.110413 0.53777 04:09  0.731389501 554.289185 0.54387 04:12  0.681134204 516.202820 0.54942 04:14  0.630978692 478.192078 0.55567 04:17      ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 3  Sample: Majed petcoke with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-214.SMP  Started: 10/3/2015 12:43:10PM Analysis Adsorptive: N2 Completed: 10/3/2015 5:41:51PM Analysis Bath Temp.: 77.333 K Report Time: 10/4/2015 2:58:26PM Thermal Correction: No Sample Mass: 0.1786 g Warm Free Space: 28.0776 cm³ Measured Cold Free Space: 90.0115 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes       Isotherm Tabular Report  Relative Absolute Quantity Elapsed Time Saturation Pressure (p/p°) Pressure Adsorbed (h:min) Pressure   (mmHg) (mmol/g)   (mmHg)          0.580980670 440.300690 0.56247 04:20  0.530948217 402.383209 0.56862 04:22  0.481197643 364.679352 0.56879 04:25  0.431341767 326.895691 0.57126 04:28  0.380895599 288.664673 0.57583 04:31  0.330801013 250.700104 0.57890 04:34  0.281113438 213.043991 0.58074 04:36  0.231036026 175.092438 0.57973 04:39  0.200614578 152.037308 0.57734 04:42  0.141047498 106.893936 0.56466 04:45  106  Figure 55: Isotherm Linear Plot  ASAP 2020 V3.01 H Unit 1 Serial #: 336 Page 4  Sample: Majed petcoke with cata.2015 Operator: MAJED oct.2, 2015 Submitter: File: C:\DATA\MAJED\000-214.SMP  Started: 10/3/2015 12:43:10PM Analysis Adsorptive: N2 Completed: 10/3/2015 5:41:51PM Analysis Bath Temp.: 77.333 K Report Time: 10/4/2015 2:58:26PM Thermal Correction: No Sample Mass: 0.1786 g Warm Free Space: 28.0776 cm³ Measured Cold Free Space: 90.0115 cm³ Equilibration Interval: 10 s Low Pressure Dose: None Automatic Degas: Yes  Majed petcoke with cata.2015 - Adsorption Majed petcoke with cata.2015 - Desorption     0.6            0.5           Adsorbed (mmol/g) 0.4           0.3           Quantity             0.2            0.1            0.0            0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0  Relative Pressure (p/p°)  107  Appendix D  C-MS Sample Scan  Some of the products found in the GCMS sample Figure 56: mass spect for a sample collected at 90 min for the HDS of DBT reaction at T=638 and P= 4.8 MPa 108  As shown in Figure 56 the GCMS was used to determine the concentration of the reaction species. In order to find the exact concentration, the GCMS should be calibrated. The calibration curve was constructed by analyzing the known concentrations of the feed and the products species. The GCMS showed the different areas of different known species’ concentrations. The unknown sample’s concentration was determined by fitting the sample’s area in the calibration curve. More elaboration will follow. As shown in Table 25 the calibration curve was constructed at a range of DBT concentration 0.1 – 2 wt%. The liquid inside the insert is 150 µL and we diluted it to 15 µL to maintain the GCMS column not to be damaged. Average density was calculated as follows 𝜌𝑎𝑣𝑒 = (%100)DBT 𝜌𝐷𝐵𝑇 + (%100)Decalin 𝜌𝐷𝑒𝑐𝑎𝑙𝑖𝑛  Target initial sample (mg) =      𝜌𝑎𝑣𝑒*diluted volume (15 µL)  Target DBT conc (wt%) = (𝐶𝐷𝐵𝑇∗𝑇𝑎𝑟𝑔𝑒𝑡 𝑖𝑛𝑡𝑖𝑎𝑙 𝑠𝑎𝑚𝑝𝑙𝑒)100(𝑇𝑎𝑟𝑔𝑒𝑡 𝑖𝑛𝑡𝑖𝑎𝑙 𝑠𝑎𝑚𝑝𝑙𝑒+𝑇𝑎𝑟𝑔𝑒𝑡 𝐷𝑃𝐸+𝑇𝑎𝑟𝑔𝑒𝑡 𝐷𝑒𝑐𝑎𝑙𝑖𝑛)100 Target DPE conc. (wt%) = (𝑇𝑎𝑟𝑔𝑒𝑡 𝐷𝑃𝐸)(𝑇𝑎𝑟𝑔𝑒𝑡 𝑖𝑛𝑡𝑖𝑎𝑙 𝑠𝑎𝑚𝑝𝑙𝑒+𝑇𝑟𝑔𝑒𝑡 𝐷𝑃𝐸+𝑇𝑎𝑟𝑔𝑒𝑡 𝐷𝑒𝑐𝑎𝑙𝑖𝑛)*100  109  Table 26 DBT calibration curve calculation 110  By plotting actual DBT conc Vs DBT area we get                         Figure 57: DBT calibration curve  Once we obtain the area of the unknown sample we lay it on Figure 57 to find the diluted conc. Of DBT. D.1 Example of Sample Calculation           Table 27 : GCMS vail  GCMS vail contains weight (mg) DPE 0.25 Decalin  71.8 Diluted sample 13.45  85.5  From the GCMS sample DBT weight = conc. Wt% of  DBT (obtained from the curve after knowing its area) * Total weight  = 0.2119100*85.5 = 0.181 mg DBT diluted conc. =𝐷𝐵𝑇 𝑤𝑒𝑖𝑔ℎ𝑡𝐷𝑖𝑙𝑢𝑡𝑒𝑑 𝑣𝑜𝑙𝑢𝑚𝑒 = 0.181 𝑚𝑔15 µ𝐿 = 0.01208 mg/µL  y = 8E-08x + 0.0018R² = 0.98650.000.050.100.150.200.250.300.350 1000000 2000000 3000000 4000000 5000000diluted conc. DBTGCMS area(DBT)111                                                              = 1.207𝐸−5 𝑔/µL184.26 𝑔/𝑚𝑜𝑙                                                                             = 6.554E-8 mol/µL                                                                               = 6.55E-2 mol/L  To convert the diluted concentration to the actual concentration it has to be as follow 𝐷𝑖𝑢𝑡𝑒 𝐷𝐵𝑇 ∗ 𝑇𝑜𝑡𝑎𝑙 𝐺𝐶𝑀𝑆 𝑣𝑎𝑖𝑙 𝑣𝑜𝑙𝑢𝑚𝑒𝑉𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 𝑖𝑛 𝐺𝐶𝑀𝑆 𝑣𝑎𝑖𝑙                                              = 6.55𝐸−2∗(85.50.9)(13.450.9) = 0.416               112  Appendix E  Repeatability and Analysis Error E.1 Conversion Calculation After each reaction, we find the GCMS areas correspond to the DBT, BP, CHB, THDBT, and BCH by locate the areas in the calibration curves to obtain the concentration weight percent. To find the weight of each component, for instance, DBT, we multiply DBT conc.% into total weight of (DPE + Decalin +Product). Followed by that, we calculate the diluted moles of each component. Finally, to find the conversion, we divide the sum of the number of moles of the products by the total number of moles of the products + the remaining DBT in the reaction as shown in the following calculations: BCH Conc% 0.011650668 DBT 248695 DBT conc% 0.021621908 CHB Conc% 0.114441108 TetraDBT 0.004039297 BP Conc% 0.124547969 DPE 0.248mg   Decalin 71.004mg  Product 14.058mg    0.29070 <-- DPE wt%   0.01845 < -- mg DBT  1.00107E-07 < --mole DBT  BP 0.10625 < -- mg BP  6.8901E-07 < --mole BP in product CHB 9.7630E-02 < -- mg BCH  6.0920E-07 < --mole BCH in product Tetra-DBT 3.4459E-03 < -- mg Tetra-DBT  1.8301E-08 < -- mole Tetra DBT in product BCH 9.939E-03   5.9767E-08  %Conv 93.219                            sum of moles (prod.+remaining DBT)   1.4764E-06 113  E.2 Selectivity Calculations To calculate the selectivity of any reaction, we collect the concentration weight percent of the products and normalize them to 100 and then we can get the selectivity. A calculation example will follow  component conc.Wt% normnalizing BCH 0.002 0.025 CHB 0.032 0.342 BP 0.046 0.49 THDBT 0.013 0.143  E.3 Conversion Repeatability The standard error was used to determine the variability of the measured data. The standard error equation is                                     Se = √∑ 𝑆𝑆𝑄∗𝑑𝑓𝑚𝑖=1∑ 𝑑𝑓                                         (E.1 1) Where Xi is the sample variable, Xavg is the average sample value, and df is the degree of freedom which is the number of samples minus one.  The repeatability runs for DBT conversions are given in the following table      114  Table 28: Conversions error calculation          Temperature experiments  Average SSQ df at 588 K 27.92      36.60      47.02 37.178 183.0553 2 at 603 K 60.23      49.09      81.97      55.11 61.602 615.5213 3 at 623 K 84.15      107.14      97.75      110.95 99.997 427.0217 3 at 638 K 119.07      105.26      120.94 115.09 146.6377 2                        ∑ 𝑆𝑆𝑄𝑚𝑖=1 *df             1372.236                          ∑ 𝑑𝑓 10                           Se^2   137.2236                            Se 11.71425   115  E.4 Product Selectivity Repeatability  The standard error was calculated for biphenyl as an example. The other products followed the same procedure. Table 29: Error calculation of biphenyl     Average SSQ df BP at 603 K 1.93E-01       2.15E-01       1.94E-01 2.01E-01 0.000315 2        BP at 623 K 3.23E-01       2.80E-01       2.85E-01       2.69E-01 2.89E-01 0.001672 3    ∑ 𝑆𝑆𝑄𝑚𝑖=1 *df 0.001987      ∑ 𝑑𝑓  5      Se^2 0.000397       Se 0.019936               116  Appendix F  Thermal Experiments Data Many thermal runs were conducted at different temperature as shown in Table 30 Table 30: Thermal run data  Reaction time(min) RPM Temp carbon Balance% Conversion (Prod accum) 60 2000 588 104.26 2 90 2000 588 112.93 3 120 2000 588 100.94 3 30 2000 603 105.65 3 60 2000 603 100.85 3 90 2000 603 107.51 7 120 2000 603 101.60 7 30 2000 623 96.42 5 60 2000 623 103.95 10 90 2000 623 105.05 18 30 2000 638 96.64 6 60 2000 638 97.25 10 90 2000 638 101.23 18   Figure 58: Comparison of different temperature amd reaction tmes for the thermal runs of HDS of DBT reaction 024681012141618200 20 40 60 80 100 120 140conversion%Time (min)588 K603 K623 K638 K117   Figure 59: Selectivity of the thermal reaction at 588 K and 2000 RPM at different time   Figure 60: Selectivity of the thermal reaction at 603 K and 2000 RPM at different time     00.10.20.30.40.50.60.70 20 40 60 80 100 120 140Selectivity, wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-00.10.20.30.40.50.60.70.80 20 40 60 80 100 120 140 160Selectivity, wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-118   Figure 61: Selectivity of the thermal reaction at 623 K and 2000 RPM at different time   Figure 62: Selectivity of the thermal reaction at 638 K and 2000 RPM at different time     0102030405060700 30 60 90 120 150Selectivity, Wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene, 1,2,3,4-tetrahydro-010203040506070800 20 40 60 80 100 120 140Selectivity, wt%Time(min)1,1'-BicyclohexylBenzene, cyclohexyl-BiphenylDibenzothiophene,1,2,3,4-tetrahydro-119  Appendix G   Matlab Codes These codes were used to model the HDS of DBT over NiMoS/AC and NiMoS/PC catalysts at different temperature 588 – 638 K. Least square function using Levenberg-Marquardt was used to estimate the kinetic parameters as well as their error. G.1 Main Body Code clear all global nvar nx x0 y0  global verbose global n1 n2 n3 n4 H2   verbose(1:2) = 1; %  % x is the indep varaibale vector e.g. time measurements % y is matrix of responses % columns of y are responses y1, y2 (e.g. mol frac of component 1 and 2) % rows of y are y values at the value of the indep variable (time) in x % first row of y is initial value of response % the program uses the Levenberg-Marquardt method to estimate parameters % and calc statistics - done in leasqr and dfdp % these two matlab m-files are designed for single repsonse % the input data is re-aarnaged to yoied a single respone vector y % the L-M requires the model to be calculated -this is done in modelmulti.m % and assume sthe model is a series of ODEs, with the number of odes equalt  % to the number of responses.  The ODEs are calcualte din ODEfunm.  Note that % this function must use teh correct model for each y % % INPUT % % input number of responses % % Order of reactions: n1=1;      % order of CA n2=1;      % order of CB n3=1;      % order of CC n4 = 1; H2=1;   % Concentration of Hydrogen in liquid.  % % Majed's data as of Novemver 25 2015 % time datq min) T  =[0 60 90 120]'; 44 nt=length (T) x(1:nt-1)=T(2:nt) x nx=length(x)   CDBT=[5.98E-01 2.957E-01    2.545E-01   1.891E-01]'; 120    CBCH= [0 1.111E-01  1.474E-01   1.454E-01]';   CBP=[0 7.456E-02    1.177E-01   1.471E-01]';   CCHB=[0 2.658E-02   2.552E-02   2.322E-02]';   CTHDBT=[0 5.222E-03 8.772E-03   1.481E-02]';     for j=1:nt-1     y1(j)=CAX(j+1);     y2(j)=CBX(j+1);     y3(j)=CCX(j+1);     y4(j)=CDX(j+1);     y5(j)=CEX(j+1);    end   nvar=5; x0=0.;   oldx=x; nx = length(x);  y = [y1' y2' y3' y4' y5'];  newy=y(:);  oldy=reshape(newy,nx,nvar);  x=x';  newx=[x;x;x;x;x];   % y01(1:nx)=CAX(1);  y02(1:nx)=CBX(1);  y03(1:nx)=CCX(1);  y04(1:nx)=CDX(1);  y05(1:nx)=CEX(1);  y01(1)=5.13E-01  y01(2)=5.54E-01  y01(3)=5.20E-01  newy0=[y01';y02';y03'; y04'; y05']; % %INPUT DATA NOW IN CORRECT COLUMN FORMAT %  y0=newy0  x=newx  y=newy %        %  provide initial parameter guesses % theta=[     5.0000e-003     8.0000e-003     2.0000e-003     4.0000e-003     8.0000e-003] ;                                             np=length(theta) pin=theta   % 121  % Begin calculation by calling L-M leat squares routine % [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]=leasqr(x,y,pin,'modelmulti'); disp('RESPONSE:') if kvg ==1     disp ('PROBELM CONVERGED')     elseif kvg == 0     disp('PROBLEM DID NOT CONVERGE') end format shortEng oldf=reshape(f,nx,nvar); oldr=reshape(y-f, nx, nvar);   % model value calculation %ODEfunm(xatx,yatx,p,knt) tspan=0:.1:120; C0 = [y0(1);CBX(1);CCX(1);CDX(1);CEX(1)]; [t,Y]=ode45(@ODEfunm,tspan,C0,[],p);   CA  =Y(:,1); CB  =Y(:,2); CC  =Y(:,3); CD  =Y(:,4); CE  =Y(:,5);        %    figure; %    %       subplot(2,4,1)                                                   %               plot(T,CA_X_150_day2,'bo',T,CB_X_150_day2,'md',T,CC_X_150_day2,'r+',m,C_4MS_T150_day2,'b--',m,C_HYD_T150_day2,'m--',m,C_C18_T150_day2,'r--')  %               title('T=150C') %               ylabel({'Day 2';'Concentration (mol/L)'}) %               legend('4MS','HYD','C18',1)         disp ('X-values:')     disp (oldx')      disp ('Y-values')     disp(oldy)      disp('f-values - i.e. model calculated responses')     disp(oldf)     disp('Residuals:')     disp (oldr) %     disp ('Standarddized residuals:') %     disp (stdresid)     disp ('Estimated parameter values are;')     disp (p)     disp ('Covariance of estimated parameters - sqrt of diagonal gives CL')     disp (covp)     disp('R2 values is:')     disp (r2)     plot (oldx,oldy,'d'), hold, plot (oldx,oldf) 122     figure   plot(oldx(:),oldy(:,1),'bx',oldx(:),oldy(:,2),'ko',oldx(:),oldy(:,3),'r+',oldx(:),oldy(:,4),'gd',oldx(:),oldy(:,5),'mo',tspan,CA,'b--',tspan,CB,'k--',tspan,CC,'r--',tspan,CD,'g--',tspan,CE,'m--')            title('T=588, NiMoS/PC')               xlabel('Time (min)')               ylabel('Concentration (mol/L)')               legend('DBT','biphenyl','cyclohexylBenzene','1,2,3,4THDBT','1,1-Bicyclohexy')                   G.2 Modelmulti Code function f = modelmulti (x,pin) % Solve a simple system of 2 ODE's  - 2 response variables % find the solution at sepcified x values - corresponding to measured data % first data point in x corresponds to initial condition global nvar nx x0 y0  global verbose global n1 n2 n3 n4 H2    nxx=length(x); yzero=reshape(y0,nx,nvar);       for i = 1:nx      xf = x(i);      xoo=x0;      yzed=yzero(i,:);      [xmodel,ymodel] = ode45 (@ODEfunm,[xoo,xf], yzed,[],pin);      yfinal(i,:)=ymodel(end,:);     end  f = yfinal(:);  G.3 ODE Codes function yprime=ODEfunm(xatx,yatx,p,knt) global nvar nx x0 y0  global verbose global n1 n2 n3 n4 H2   %    disp('*****************************YPRIME') %    disp (knt) %   nx k1=p(1); k2=p(2); k3=p(3); k4=p(4); % k5=p(5);   %     yp(1)=-k1*yatx(1)-k2*yatx(1); %     yp(2)=-k3*yatx(2)+k1*yatx(1); %     yp(3)=-k5*yatx(3)+k4*yatx(4)+k3*yatx(2); %     yp(4)=-k4*yatx(4)+k2*yatx(1); %     yp(5)=-yp(1)-yp(2)-yp(3)-yp(4); 123      yp(1)=-k1*yatx(1)-k2*yatx(1);     yp(2)=k1*yatx(1);     yp(3)=k3*yatx(4)-k4*yatx(3);     yp(4)=-k3*yatx(4)+k2*yatx(1);     yp(5)=-yp(1)-yp(2)-yp(3)-yp(4);   yprime =[yp(1);yp(2)';yp(3)';yp(4)';yp(5)']; G.4 Jacobian Matrix Calculation  function prt=dfdp(x,f,p,dp,func) % numerical partial derivatives (Jacobian) df/dp for use with leasqr % --------INPUT VARIABLES--------- % x=vec or matrix of indep var(used as arg to func) x=[x0 x1 ....] % f=func(x,p) vector initialsed by user before each call to dfdp % p= vec of current parameter values % dp= fractional increment of p for numerical derivatives %      dp(j)>0 central differences calculated %      dp(j)<0 one sided differences calculated %      dp(j)=0 sets corresponding partials to zero; i.e. holds p(j) fixed % func=string naming the function (.m) file %      e.g. to calc Jacobian for function expsum prt=dfdp(x,f,p,dp,'expsum') %----------OUTPUT VARIABLES------- % prt= Jacobian Matrix prt(i,j)=df(i)/dp(j) %================================ m=length(x);n=length(p);      %dimensions ps=p; prt=zeros(m,n);del=zeros(n,1);       % initialise Jacobian to Zero for j=1:n       del(j)=dp(j) .*p(j);    %cal delx=fract(dp)*param value(p)            if p(j)==0            del(j)=dp(j);     %if param=0 delx=fraction            end       p(j)=ps(j) + del(j);       if del(j)~=0, f1=feval(func,x,p);            if dp(j) < 0, prt(:,j)=(f1-f)./del(j);            else            p(j)=ps(j)- del(j);            prt(:,j)=(f1-feval(func,x,p))./(2 .*del(j));            end       end       p(j)=ps(j);     %restore p(j) end return    G.5 Least Square Codes function [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]= ...       leasqr(x,y,pin,F,stol,niter,wt,dp,dFdp,options) %function[f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]= %                   leasqr(x,y,pin,F,{stol,niter,wt,dp,dFdp,options}) % % Version 3.beta %  {}= optional parameters % Levenberg-Marquardt nonlinear regression of f(x,p) to y(x), where: 124  % x=vec or mat of indep variables, 1 row/observation: x=[x0 x1....xm] % y=vec of obs values, same no. of rows as x. % wt=vec(dim=length(x)) of statistical weights.  These should be set %   to be proportional to (sqrt of var(y))^-1; (That is, the covariance %   matrix of the data is assumed to be proportional to diagonal with diagonal %   equal to (wt.^2)^-1.  The constant of proportionality will be estimated.), %   default=ones(length(y),1). % pin=vector of initial parameters to be adjusted by leasqr. % dp=fractional incr of p for numerical partials,default= .001*ones(size(pin)) %   dp(j)>0 means central differences. %   dp(j)<0 means one-sided differences. % Note: dp(j)=0 holds p(j) fixed i.e. leasqr wont change initial guess: pin(j) % F=name of function in quotes,of the form y=f(x,p) % dFdp=name of partials M-file in quotes default is prt=dfdp(x,f,p,dp,F) % stol=scalar tolerances on fractional improvement in ss,default stol=.0001 % niter=scalar max no. of iterations, default = 20 % options=matrix of n rows (same number of rows as pin) containing %   column 1: desired fractional precision in parameter estimates. %     Iterations are terminated if change in parameter vector (chg) on two %     consecutive iterations is less than their corresponding elements %     in options(:,1).  [ie. all(abs(chg*current parm est) < options(:,1)) %      on two consecutive iterations.], default = zeros(). %   column 2: maximum fractional step change in parameter vector. %     Fractional change in elements of parameter vector is constrained to be %     at most options(:,2) between sucessive iterations. %     [ie. abs(chg(i))=abs(min([chg(i) options(i,2)*current param estimate])).], %     default = Inf*ones(). % %          OUTPUT VARIABLES % f=vec function values computed in function func. % p=vec trial or final parameters. i.e, the solution. % kvg=scalar: =1 if convergence, =0 otherwise. % iter=scalar no. of interations used. % corp= correlation matrix for parameters % covp= covariance matrix of the parameters % covr = diag(covariance matrix of the residuals) % stdresid= standardized residuals % Z= matrix that defines confidence region % r2= coefficient of multiple determination   % All Zero guesses not acceptable % Richard I. Shrager (301)-496-1122 % Modified by A.Jutan (519)-679-2111 % Modified by Ray Muzic 14-Jul-1992 %       1) add maxstep feature for limiting changes in parameter estimates %          at each step. %       2) remove forced columnization of x (x=x(:)) at beginning. x could be %          a matrix with the ith row of containing values of the %          independent variables at the ith observation. %       3) add verbose option 125  %       4) add optional return arguments covp, stdresid, chi2 %       5) revise estimates of corp, stdev % Modified by Ray Muzic 11-Oct-1992 %   1) revise estimate of Vy.  remove chi2, add Z as return values % Modified by Ray Muzic 7-Jan-1994 %       1) Replace ones(x) with a construct that is compatible with versions %          newer and older than v 4.1. %       2) Added global declaration of verbose (needed for newer than v4.x) %       3) Replace return value var, the variance of the residuals with covr, %          the covariance matrix of the residuals. %       4) Introduce options as 10th input argument.  Include %          convergence criteria and maxstep in it. %       5) Correct calculation of xtx which affects coveraince estimate. %       6) Eliminate stdev (estimate of standard deviation of parameter %          estimates) from the return values.  The covp is a much more %          meaningful expression of precision because it specifies a confidence %          region in contrast to a confidence interval..  If needed, however, %          stdev may be calculated as stdev=sqrt(diag(covp)). %       7) Change the order of the return values to a more logical order. %       8) Change to more efficent algorithm of Bard for selecting epsL. %       9) Tighten up memory usage by making use of sparse matrices (if %          MATLAB version >= 4.0) in computation of covp, corp, stdresid. % Modified by Sean Brennan 17-May-1994 %          verbose is now a vector: %          verbose(1) controls output of results %          verbose(2) controls plotting intermediate results % % References: % Bard, Nonlinear Parameter Estimation, Academic Press, 1974. % Draper and Smith, Applied Regression Analysis, John Wiley and Sons, 1981. % %set default args   % argument processing %   plotcmd='plot(x(:,1),y,''+'',x(:,1),f); shg'; %if (sscanf(version,'%f') >= 4), vernum= sscanf(version,'%f'); if vernum(1) >= 4,   global verbose   plotcmd='plot(x(:,1),y,''+'',x(:,1),f); figure(gcf)'; end; if (exist('OCTAVE_VERSION'))   global verbose end;   if(exist('verbose')~=1), %If verbose undefined, print nothing     verbose(1)=0    %This will not tell them the results     verbose(2)=0    %This will not replot each loop end; if (nargin <= 8), dFdp='dfdp'; end; if (nargin <= 7), dp=.001*(pin*0+1); end; %DT 126  if (nargin <= 6), wt=ones(length(y),1); end;    % SMB modification if (nargin <= 5), niter=20; end; if (nargin == 4), stol=.0001; end; %   y=y(:); wt=wt(:); pin=pin(:); dp=dp(:); %change all vectors to columns % check data vectors- same length? m=length(y); n=length(pin); p=pin;[m1,m2]=size(x); if m1~=m ,error('input(x)/output(y) data must have same number of rows ') ,end;   if (nargin <= 9),   options=[zeros(n,1) Inf*ones(n,1)];   nor = n; noc = 2; else   [nor noc]=size(options);   if (nor ~= n),     error('options and parameter matrices must have same number of rows'),   end;   if (noc ~= 2),     options=[options(noc,1) Inf*ones(noc,1)];   end; end; pprec=options(:,1); maxstep=options(:,2); %   % set up for iterations % f=feval(F,x,p); fbest=f; pbest=p; r=wt.*(y-f); sbest=r'*r; nrm=zeros(n,1); chgprev=Inf*ones(n,1); kvg=0; epsLlast=1; epstab=[.1 1 1e2 1e4 1e6];   % do iterations % for iter=1:niter,   pprev=pbest;   prt=feval(dFdp,x,fbest,pprev,dp,F);   r=wt.*(y-fbest);   sprev=sbest;   sgoal=(1-stol)*sprev;   for j=1:n,     if dp(j)==0,       nrm(j)=0;     else       prt(:,j)=wt.*prt(:,j);       nrm(j)=prt(:,j)'*prt(:,j);       if nrm(j)>0,         nrm(j)=1/sqrt(nrm(j)); 127        end;     end     prt(:,j)=nrm(j)*prt(:,j);   end; % above loop could ? be replaced by: % prt=prt.*wt(:,ones(1,n)); % nrm=dp./sqrt(diag(prt'*prt)); % prt=prt.*nrm(:,ones(1,m))';   [prt,s,v]=svd(prt,0);   s=diag(s);   g=prt'*r;   for jjj=1:length(epstab),     epsL = max(epsLlast*epstab(jjj),1e-7);     se=sqrt((s.*s)+epsL);     gse=g./se;     chg=((v*gse).*nrm); %   check the change constraints and apply as necessary     ochg=chg;     for iii=1:n,       if (maxstep(iii)==Inf), break; end;       chg(iii)=max(chg(iii),-abs(maxstep(iii)*pprev(iii)));       chg(iii)=min(chg(iii),abs(maxstep(iii)*pprev(iii)));     end;      if (verbose(1) & any(ochg ~= chg)),        disp(['Change in parameter(s): ' ...           sprintf('%d ',find(ochg ~= chg)) 'were constrained']);      end;     aprec=abs(pprec.*pbest);       %--- % ss=scalar sum of squares=sum((wt.*(y-f))^2).     if (any(abs(chg) > 0.1*aprec)),%---  % only worth evaluating function if       p=chg+pprev;                       % there is some non-miniscule change       f=feval(F,x,p);       r=wt.*(y-f);       ss=r'*r;       if ss<sbest,         pbest=p;         fbest=f;         sbest=ss;       end;       if ss<=sgoal,         break;       end;     end;                          %---   end;   epsLlast = epsL; %   if (verbose(2)), %     eval(plotcmd); %   end;   if ss<eps,     break;   end   aprec=abs(pprec.*pbest); %  [aprec chg chgprev]   if (all(abs(chg) < aprec) & all(abs(chgprev) < aprec)),     kvg=1; 128      if (verbose(1)),       fprintf('Parameter changes converged to specified precision\n');     end;     break;   else     chgprev=chg;   end;   if ss>sgoal,     break;   end; end;   % set return values % p=pbest; f=fbest; ss=sbest; kvg=((sbest>sgoal)|(sbest<=eps)|kvg); if kvg ~= 1 , disp(' CONVERGENCE NOT ACHIEVED! '), end;   % CALC VARIANCE COV MATRIX AND CORRELATION MATRIX OF PARAMETERS % re-evaluate the Jacobian at optimal values jac=feval(dFdp,x,f,p,dp,F); msk = dp ~= 0; n = sum(msk);           % reduce n to equal number of estimated parameters jac = jac(:, msk);  % use only fitted parameters   %% following section is Ray Muzic's estimate for covariance and correlation %% assuming covariance of data is a diagonal matrix proportional to %% diag(1/wt.^2). %% cov matrix of data est. from Bard Eq. 7-5-13, and Row 1 Table 5.1   if vernum(1) >= 4,   Q=sparse(1:m,1:m,(0*wt+1)./(wt.^2));  % save memory   Qinv=inv(Q); else   Qinv=diag(wt.*wt);   Q=diag((0*wt+1)./(wt.^2)); end; resid=y-f;                                    %un-weighted residuals covr=resid'*Qinv*resid*Q/(m-n);                 %covariance of residuals Vy=1/(1-n/m)*covr;  % Eq. 7-13-22, Bard         %covariance of the data   jtgjinv=inv(jac'*Qinv*jac);         %argument of inv may be singular covp=jtgjinv*jac'*Qinv*Vy*Qinv*jac*jtgjinv; % Eq. 7-5-13, Bard %cov of parm est d=sqrt(abs(diag(covp))); corp=covp./(d*d');   covr=diag(covr);                 % convert returned values to compact storage stdresid=resid./sqrt(diag(Vy));  % compute then convert for compact storage Z=((m-n)*jac'*Qinv*jac)/(n*resid'*Qinv*resid);   129  %%% alt. est. of cov. mat. of parm.:(Delforge, Circulation, 82:1494-1504, 1990 %%disp('Alternate estimate of cov. of param. est.') %%acovp=resid'*Qinv*resid/(m-n)*jtgjinv   %Calculate R^2 (Ref Draper & Smith p.46) % r=corrcoef(y,f); if (exist('OCTAVE_VERSION'))   r2=r^2; else   r2=r(1,2).^2; end   % if someone has asked for it, let them have it %  if (verbose(2)), eval(plotcmd); end,  if (verbose(1)),    disp(' Least Squares Estimates of Parameters')    disp(p')    disp(' Correlation matrix of parameters estimated')    disp(corp)    disp(' Covariance matrix of Residuals' )    disp(covr)    disp(' Correlation Coefficient R^2')    disp(r2)    sprintf(' 95%% conf region: F(0.05)(%.0f,%.0f)>= delta_pvec''*Z*delta_pvec',n,m-n)    Z %   runs test according to Bard. p 201.   n1 = sum((f-y) < 0);   n2 = sum((f-y) > 0);   nrun=sum(abs(diff((f-y)<0)))+1;   if ((n1>10)&(n2>10)), % sufficent data for test?     zed=(nrun-(2*n1*n2/(n1+n2)+1)+0.5)/(2*n1*n2*(2*n1*n2-n1-n2)...       /((n1+n2)^2*(n1+n2-1)));     if (zed < 0),       prob = erfc(-zed/sqrt(2))/2*100;       disp([num2str(prob) '% chance of fewer than ' num2str(nrun) ' runs.']);     else,       prob = erfc(zed/sqrt(2))/2*100;       disp([num2str(prob) '% chance of greater than ' num2str(nrun) ' runs.']);     end;   end; end   % A modified version of Levenberg-Marquardt % Non-Linear Regression program previously submitted by R.Schrager. % This version corrects an error in that version and also provides % an easier to use version with automatic numerical calculation of % the Jacobian Matrix. In addition, this version calculates statistics % such as correlation, etc.... % 130  % Version 3 Notes % Errors in the original version submitted by Shrager (now called version 1) % and the improved version of Jutan (now called version 2) have been corrected. % Additional features, statistical tests, and documentation have also been % included along with an example of usage.  BEWARE: Some the the input and % output arguments were changed from the previous version. % %     Ray Muzic     <rfm2@ds2.uh.cwru.edu> %     Arthur Jutan  <jutan@charon.engga.uwo.ca>                                       131  Appendix H  Calculations of Arrhenius Equations   Arrhenius expression can be shown as follows kj= Aj*exp(−𝐸𝑎𝑗𝑅∗𝑇)                              (G1) National logarithm of G1 gives                                            ln(kj) = ln(Aj) + −𝐸𝑎𝑗𝑅* 1𝑇                     (G2)   By plotting ln(kj) Vs 1𝑇 as illustrated in Figure 77 the slope and the interception can be used to obtain the pre-exponetioal factor Aj and the activation energy Ea as follow:  Figure 63: Arrhenius plot for NiMoS/AC for k1 at different temperature  The slope = −𝐸𝑎𝑗𝑅=-11.88 Ea1 = -11.88*8.314 = 104 kJ/mol Intercept = ln(A)=19.37 A = 2.5*108 cm3/(gcat*s y = -12.5x + 19.37R² = 0.9949-2-1.5-1-0.501.55 1.6 1.65 1.7 1.75LnK1000/T132  Appendix I  Thermodynamic Calculations to Determine the Reaction Phase. I.1 Reaction Phase Determination Aspen software was use to simulate the current study reaction and determine the phase of the reaction inside the reactor. As shown in Figure 64 and Table 31, after setting the conversion to zero to determine the DBT phase inside the reactor, it was found to be liquid (refer to Table 30 LIQPRO column):                Figure 64: HDS of DBT reaction flowchart 133  Table 31 Aspen result for HDS of DBT at conversion set to zero   FEED H2 H2PURGE LIQPRO OUTRECT SPEINLET VAPRO   B1 B1     B2 B3         B2 B3 B1 B2 B3   LIQUID VAPOR VAPOR LIQUID MIXED LIQUID MIXED                 Substream: MIXED                       Mole Flow   kmol/hr                      DIBENZOT                 0.0108541 0 0 0.0108432 0.0108432 0.010843 0   DECALIN                  0.7088458 0 0 0.7088458 0.7088458 0.708846 0   CYCLO-01                 0 0 0 0 1.81E-06 1.81E-06 1.81E-06   DIPHE-01                 0 0 0 7.24E-06 7.24E-06 7.24E-06 0   HYDRO-01                 0 0 0 0 1.09E-05 1.09E-05 1.09E-05   BICYC-01                 0 0 0 0 1.81E-06 1.81E-06 1.81E-06   C12H12S                  0 0 0 0 0 0 0   SULFUR                   0 0 0 0 0 0 0   H2                       0 0.1488184 0.1487804 0 0.1487804 0 0 Total Flow  kmol/hr        0.7196999 0.1488184 0.1487804 0.7196963 0.8684912 0.719711 1.45E-05 Total Flow  kg/hr          100 0.3 0.2999234 99.99912 100.3 100.0001 9.61E-04 Total Flow  l/min          1.855203 1.445154 2.740963 2.925255 6.639883 2.925296 2.47E-04 Temperature C              25 24 365 365 365 365 365 Pressure    bar            42.40276 42.40276 48 48 48 48 48 Vapor Frac                 0 1 1 0 0.2495916 0 0.9045203 Liquid Frac                1 0 0 1 0.7504084 1 0.0954796 Solid Frac                 0 0 0 0 0 0 0 Enthalpy    cal/mol        -50932.84 -6.872138 2372.776 -21945.11 -17344.26 -21944.8 -2503.228 Enthalpy    cal/gm         -366.5636 -3.409002 1177.042 -157.9396 -150.1828 -157.939 -37.70937 Enthalpy    cal/sec        -10182.32 -0.2840835 98.06183 -4387.171 -4184.259 -4387.19 -0.0100631 Entropy     cal/mol-K      -228.5736 -7.438328 -2.356278 -168.05 -138.8462 -168.047 -32.62025 Entropy     cal/gm-K       -1.645044 -3.689867 -1.168858 -1.20946 -1.20226 -1.20945 -0.4914011 Density     mol/cc         6.47E-03 1.72E-03 9.05E-04 4.10E-03 2.18E-03 4.10E-03 9.75E-04 Density     gm/cc          0.8983741 3.46E-03 1.82E-03 0.5697458 0.2517615 0.569743 0.0647494 Average MW                 138.9468 2.01588 2.01588 138.9463 115.4876 138.9448 66.38213 Liq Vol 60F l/min          1.844838 0.1328398 0.1328058 1.844828 1.977654 1.844848 2.05E-05   134  I.2 H2S Partial Pressure Calculation First of all let’s show the feed calculations: We want 1000 uL, or roughly 1 g of total mixed feed with 3600 ppm Mo (or 0.36wt%) Table 32 Mo, Ni, CS2 molecular weight Model compound DBT  Unit Desired conc. 2 wt% Mr Mo 95.96 g/mol Mr CS2 76.131 g/mol Mr Ni 58.693 g/mol  Table 33 Desired metal wt% Catalyst 18wt% NiMo/C  unit Mo conc in cat 15 wt% Ni conc in cat 3 wt% Desired Metal conc. 0.36 wt%  Table 34 Feed preparation calculation          Catalyst conc 0.36 wt%         Total mass 1 g         DBT mass 0.019400616 g 19.40 mg 15.5 uL Decalin mass 0.950630202 g 950.63 mg 1061.0 uL Metal mass 0.0036 g         Catalyst mass 0.02 g 20.00 mg     Mo moles 3.75156E-05 mol         Ni mass 0.00072 g         Ni moles 1.22672E-05 mol         Stoich S required 8.72985E-05 mol         CS2 for 3 x stoich S 0.000130948 mol           0.009969181 g 9.97 mg 7.9 uL Total additives 0.049369798 g               1000.00       135  From Table 34 we conclude: Mass of DBT + Decalin 970.03mg Mass of Mo 3.6 DBT conc 2.000mg wt% Mo loading 3.6000E-01 wt% catalyst mass  Metal 3.6 mg Support 16.4 TOTAL CATALYST MASS 20 mg  Now let’s calculate number of moles of S needed to sulphide the catalysts From Table 34 number of moles of Mo = 37.52 micromoles                         Number of S needed to sulphide the catalysts = 37.52*2=75.05 micromoles                        We are using ~ 10 mg of CS2 which equals 131.34 micromoles If                  CS2    2H2S which means 2 moles of H2S will be generated for 1 mole of CS2 Now H2S number of moles = 2*131.34 =262.67 micrometers  H2S remaining after MoS2 formation = 262.67-75.05 =187.63 Reactor volume = 𝜋4*d2*h = 𝜋4*0.0062*0.25 =7.07E-0.6 m3 PH2S = 187.63∗0.000001∗8.314∗6137.07E−6/1000 = 135.28 kPa  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0228785/manifest

Comment

Related Items