UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A study of molybdenum carbide catalysts supported on carbon derived from petroleum coke for hydrotreating Wang, Haiyan 2018

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

Item Metadata

Download

Media
24-ubc_2019_february_wang_haiyan.pdf [ 13.78MB ]
Metadata
JSON: 24-1.0373612.json
JSON-LD: 24-1.0373612-ld.json
RDF/XML (Pretty): 24-1.0373612-rdf.xml
RDF/JSON: 24-1.0373612-rdf.json
Turtle: 24-1.0373612-turtle.txt
N-Triples: 24-1.0373612-rdf-ntriples.txt
Original Record: 24-1.0373612-source.json
Full Text
24-1.0373612-fulltext.txt
Citation
24-1.0373612.ris

Full Text

   A STUDY OF MOLYBDENUM CARBIDE CATALYSTS SUPPORTED ON CARBON DERIVED FROM PETROLEUM COKE FOR HYDROTREATING by  Haiyan Wang  M.Sc., China University of Petroleum (Beijing), 2013 B.Sc., China University of Mining and Technology, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemical and Biological Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  November 2018 © Haiyan Wang, 2018ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: A Study of Molybdenum Carbide Catalysts Supported on Carbon Derived from Petroleum Coke for Hydrotreating  submitted by Haiyan Wang  in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Chemical and Biochemical Engineering  Examining Committee: Kevin J. Smith, Chemical and Biological Engineering Supervisor  Naoko Ellis, Chemical and Biological Engineering Supervisory Committee Member  Keng C Chou, Department of Chemistry Supervisory Committee Member Elod Gyenge, Department of Chemical Engineering University Examiner Jennifer Love, Department of Chemistry University Examiner  Additional Supervisory Committee Members:  Supervisory Committee Member  Supervisory Committee Member iii  Abstract Mo2C catalysts supported on carbon have been investigated for use in hydrotreating reactions that remove S, N and O from oil fractions. The thesis reports on the stability of the catalysts in the presence of different model reactants. The synthesis of mesoporous carbons derived from petroleum coke (petcoke), a by-product of Canadian oilsand upgrading, is described. The impact of the mesoporous carbon as a support of the Mo2C catalysts is also examined.   An activated charcoal (AC) was initially used as the carbon source to prepare Mo2C/AC and Ni-Mo2C/AC catalysts by carbothermal hydrogen reduction (CHR). The most active catalyst for 4-methylphenol (4-MP) hydrodeoxygenation (HDO) was obtained at a CHR temperature of 650 oC. The direct deoxygenation selectivity of this catalyst was > 78%, indicative of high O removal with low H2 consumption. The effect of a Ni promoter on the synthesis and activity of Ni-Mo2C/AC catalysts was also assessed. The presence of Ni significantly reduced the CHR temperature required for Mo2C formation by 100 oC. However, the Ni accelerated catalyst sulfidation during hydrodesulphurization (HDS) and formed a unique core-shell Mo2C-MoS2 structure. Additionally, there was an improved activity in HDS of dibenzothiophene (DBT) in the presence of Ni, provided the Ni:Mo < 0.44.   Extending these results to petcoke, the transition of Mo species and the corresponding changes to the activated petroleum coke (APC) morphology that occur during CHR were determined. A maximum mesoporosity of 37% was achieved for a sample reduced to 750 oC. The activity of the Mo2C/APC catalysts for the HDO of 4-MP was > 3x’s higher than that of Mo2C/AC because of iv  the high surface area (~2000 m2/g) of the Mo2C/APC catalyst, and the high dispersion of the Mo2C nanoparticles.   Finally, the stability of the Mo2C/APC catalysts during the HDS, hydrodenitrogenation and HDO of DBT, carbazole and dibenzofuran, respectively, was determined as a function of the Mo2C average particle size. DFT calculations were combined with experimental data to explain the selectivity change from hydrogenation to DDS observed during the HDS of DBT. Both S and N irreversibly deactivated the catalysts; whereas, the effect of O was reversible.     v  Lay Summary The goal of this study was to prepare new catalysts from Canadian oilsands petroleum coke, a by-product of the oilsands upgrading operations. The catalysts were used to remove contaminants (S, N and O) from oil fractions. Initial work showed that the new catalyst, based on the formation of molybdenum carbide (Mo2C), had good performance in removing O from a model reactant. Nickel added to the catalyst reduced the temperature required for the catalyst synthesis. The mechanism of formation of the Mo2C supported on the activated petroleum coke was also studied, together with the development of porosity within the coke. Finally, the impact of the different heteroatoms (S, N and O) present in the oil on the stability of the catalyst was studied to assess the possibility of using these new catalysts in oil-refining processes.  vi  Preface This Ph.D. dissertation consists of seven chapters. Chapter 2, 3, and 4 have been published previously in peer-reviewed journals. A version of Chapter 5 is in preparation to be submitted for publication. The Ph.D. study was conducted by Haiyan Wang under the direct supervision of Professor Kevin J. Smith in the Department of Chemical and Biological Engineering at UBC. The literature review, catalyst synthesis and characterization, reactor set-up, catalyst testing, data collection and interpretation, kinetic modeling, and preparation of the dissertation were done by Haiyan Wang under the direct supervision of Professor Kevin Smith.   The list of the publications included in this thesis is given below: 1. H. Wang, S. Liu, and K. J. Smith, “Synthesis and hydrodeoxygenation activity of carbon supported molybdenum carbide and oxycarbide catalysts”, Energy and Fuels (2016) 30 (7): 6039-6049. A version of this manuscript is included in Chapter 2.   The carbon supported Mo2C preparation, characterizations, hydrodeoxygenation reactions, kinetic modeling, as well as data analysis and interpretation were done by Haiyan Wang under the direct supervision of Professor Kevin. J. Smith. In addition, the preparation and writing of the manuscript were done by Haiyan Wang with final approval of Professor Kevin J. Smith. The XPS measurements were performed by Dr. Ken Wong from the Interfacial Analysis and Reactivity Laboratory at UBC. The TEM images were done by Shida Liu in FRIPP, China.   vii  2. H. Wang, S. Liu, R. Govindarajan, and K. J. Smith, “Preparation of Ni-Mo2C/carbon catalysts and their stability in the HDS of dibenzothiophene,” Applied Catalysis A: General (2017) 539 (5): 114-127. A version of this manuscript is included in Chapter 3.  The catalysts preparation and characterization, experimental design and set-up, HDS reaction of DBT, sample testing and analysis, and kinetic modeling were performed by Haiyan Wang under the direct supervision of Professor Kevin J. Smith. Shida Liu contributed to the discussion of the results of this work. Some of the catalyst precursors were prepared by Ruben Govindarajan. The XPS and TOF-SIMS measurements were performed by Dr. Ken Wong and Dr. John Kim, respectively from the Interfacial Analysis and Reactivity Laboratory at UBC. STEM measurements were conducted by Dr. Xin Zhang from 4D labs at SFU. Finally, the writing of this manuscript was done by Haiyan Wang with the final approval of Professor Kevin J. Smith.   3. H. Wang, S. Liu., B. Liu, V. Montes, J. M. Hill, and K. J. Smith, “Carbon and Mo transformation during the synthesis of mesoporous Mo2C/carbon catalysts by carbothermal hydrogen reduction,” Journal of Solid State Chemistry (2018) 258: 818-824. A version of this manuscript is included in Chapter 4.  All the catalyst synthesis and characterization, HDO reactions, data collection and interpretation were done by Haiyan Wang under the direct supervision of Professor Kevin J. Smith. The manuscript preparation and writing were done by Haiyan Wang with the final approval of Professor Kevin J. Smith. Shida Liu contributed to the experimental viii  design and data interpretation. Dr. Bing Liu performed the DFT calculations using VASP. Dr. Vicente Montes and Professor Josephine M. Hill contributed to the data interpretation of carbon based materials and manuscript revision. The XPS measurement was performed by Dr. Ken Wong from the Interfacial Analysis and Reactivity Laboratory at UBC.  4. H. Wang, S. Liu, and K. J. Smith, “Understanding selectivity changes during hydrodesulfurization of dibenzothiophene on Mo2C/carbon catalysts”, Journal of Catalysis (2018), in preparation. A version of this manuscript is included in Chapter 5.  The catalyst preparation, characterization, reactor modification, experimental design, sample collection and data analysis were done by Haiyan Wang under the direct supervision of Professor Kevin J. Smith. The effect of S on different particle size Mo2C catalysts were studied in a fixed bed reactor. All the DFT calculations were conducted by Shida Liu using VASP software. The manuscript was prepared and written by Haiyan Wang with final approval of Professor Kevin J. Smith. The XPS measurement was performed by Dr. Ken Wong from the Interfacial Analysis and Reactivity Laboratory at UBC.  Additionally, the following list summarizes the conference papers in proceedings of various international conferences: ix   H. Wang, K. J. Smith, (2015) “Synthesis and hydrodeoxygenation activity of carbon supported molybdenum carbide and oxycarbide catalysts”. 65th Canadian Chemical Engineering (CSChE) Conference, Calgary, Canada.  H. Wang, S. Liu and K. J. Smith, (2016) “Carbon supported Ni-Mo2C for hydrodesulfurization”. 16th International Congress on Catalysis (ICC), Beijing, China.  H. Wang, S. Liu and K. J. Smith, (2017) “The activity and stability of Mo2C supported on activated petroleum coke in hydrotreating reactions”. 2017 American Institute of Chemical Engineers (AIChE) Annual Meeting, Minnesota, USA.  H. Wang, S. Liu and K. J. Smith, (2018) “Transition of Mo2C/carbon catalyst in HDS of dibenzothiophene: the effect of different particle sizes, CHR temperatures and promoters”. 25th Canadian Symposium on Catalysis (CSC), Saskatoon, Canada.  x  Table of Contents Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ...........................................................................................................................x List of Tables ............................................................................................................................ xviii List of Figures .............................................................................................................................xxv List of Nomenclature ............................................................................................................. xxxiv List of Abbreviations ................................................................................................................... iv Acknowledgements .................................................................................................................... viii Chapter 1: Introduction ................................................................................................................1 1.1 Background ..................................................................................................................... 1 1.1.1 Synthesis Methods of Molybdenum Carbide (Mo2C) Catalysts ................................. 1 1.1.2 Carbon Supported Catalysts and their Application in Hydrotreating Reaction .......... 3 1.1.2.1 Hydrotreating Technology .............................................................................. 3 1.1.2.2 Carbon Supported Catalysts in Hydrodesulphurization (HDS) and Hydrodenitrogenation (HDN) ............................................................................................. 4 1.1.2.3 Carbon Supported Catalysts in Hydrodeoxygenation (HDO) ........................ 4 1.1.3 Activation Methods of Carbonaceous Materials ......................................................... 6 1.1.3.1 Physical Activation ......................................................................................... 6 1.1.3.2 Chemical Activation ....................................................................................... 8 1.1.3.3 Catalytic Activation by Metals ..................................................................... 11 1.1.3.4 Petroleum Coke ............................................................................................. 13 xi  1.1.4 Summary ................................................................................................................... 15 1.2 Objectives of the Thesis ................................................................................................ 16 1.3 Approach ....................................................................................................................... 17 1.4 Outline of the Dissertation ............................................................................................ 18 Chapter 2: Synthesis of Mo2C/AC Catalysts and the Application in HDO of 4-methylphenol ........................................................................................................................................................21 2.1 Introduction ................................................................................................................... 21 2.2 Experimental ................................................................................................................. 22 2.2.1 Materials ................................................................................................................... 22 2.2.2 Catalyst Preparation .................................................................................................. 23 2.2.3 Catalyst Characterization .......................................................................................... 24 2.2.3.1 N2 Adsorption and Desorption ...................................................................... 24 2.2.3.2 X-ray Diffraction (XRD) .............................................................................. 24 2.2.3.3 X-ray Photoelectron Spectroscopy (XPS) .................................................... 25 2.2.3.4 CO Chemisorption ........................................................................................ 25 2.2.3.5 O Content Anlysis ......................................................................................... 26 2.2.3.6 Transmission Electron Microscopy (TEM) .................................................. 26 2.2.4 Hydrodeoxygenation of 4-methylphenol .................................................................. 26 2.3 Results ........................................................................................................................... 28 2.3.1 Catalyst Characterization .......................................................................................... 28 2.3.1.1 Textural Properties by BET Analysis and Carbon Loss during CHR .......... 28 2.3.1.2 XRD Analysis ............................................................................................... 30 2.3.1.3 XPS, O Analysis and CO Uptake ................................................................. 32 xii  2.3.1.4 TEM Analysis ............................................................................................... 36 2.3.2 Catalytic Performance in HDO of 4-MP .................................................................. 39 2.4 Discussion ..................................................................................................................... 45 2.5 Conclusions ................................................................................................................... 51 Chapter 3: Preparation of Mo2C/AC and Ni-Mo2C/AC Catalysts and their Stability in HDS of Dibenzothiophene  ...................................................................................................................53 3.1 Introduction ................................................................................................................... 53 3.2 Experimental ................................................................................................................. 55 3.2.1 Catalyst Preparation .................................................................................................. 55 3.2.2 Catalyst Characterization .......................................................................................... 56 3.2.2.1 Elemental Analysis ....................................................................................... 56 3.2.2.2 N2 Adsorption and Desorption ...................................................................... 57 3.2.2.3 X-ray Diffraction (XRD) .............................................................................. 57 3.2.2.4 X-ray Photoelectron Spectroscopy (XPS) .................................................... 57 3.2.2.5 Transmission Electron Microscopy (TEM) .................................................. 57 3.2.2.6 Time of Flight Secondary Ion Mass Spectroscopy (TOF-SIMS) ................. 58 3.2.3 Catalyst Activity Tests .............................................................................................. 59 3.3 Results ........................................................................................................................... 62 3.3.1 Fresh Catalysts Characterization ............................................................................... 62 3.3.2 Catalysts Characterization after HDS Reaction ........................................................ 70 3.3.2.1 XRD and BET Analysis ................................................................................ 71 3.3.2.2 XPS and CHNS Analysis .............................................................................. 72 3.3.2.3 TOF-SIMS Analysis ..................................................................................... 77 xiii  3.3.2.4 TEM Analysis ............................................................................................... 78 3.3.3 Catalyst Performance in HDS of DBT ...................................................................... 82 3.4 Discussion ..................................................................................................................... 93 3.5 Conclusions ................................................................................................................. 100 Chapter 4: Synthesis of Mesoporous Mo2C/Carbon Catalysts by Carbothermal Hydrogen Reduction using Petroleum Coke  ............................................................................................101 4.1 Introduction ................................................................................................................. 101 4.2 Experimental and Computational Methods ................................................................ 102 4.2.1 Catalyst Preparation ................................................................................................ 102 4.2.2 Catalyst Characterization ........................................................................................ 103 4.2.3 Catalyst Activity Tests ............................................................................................ 104 4.2.4 Computational Method ........................................................................................... 104 4.3 Results and Discussion ............................................................................................... 105 4.3.1 Characterization Results ......................................................................................... 105 4.3.2 Mo Species Transformation .................................................................................... 113 4.3.3 Pore Development during Carbothermal Hydrogen Reduction (CHR) .................. 119 4.3.4 Activity Test in HDO of 4-methylphenol ............................................................... 121 4.4 Conclusion .................................................................................................................. 122 Chapter 5: The Effect of S on Mo2C/APC Catalysts with Various Particle Sizes  ..............123 5.1 Introduction ................................................................................................................. 123 5.2 Experimental ............................................................................................................... 124 5.2.1 Preparation of Catalysts .......................................................................................... 124 5.2.2 Catalyst Characterization ........................................................................................ 125 xiv  5.2.3 Catalytic Performance Measurement in HDS ......................................................... 126 5.2.4 Computational Model and Methods........................................................................ 127 5.3 Results ......................................................................................................................... 130 5.3.1 Fresh Catalyst Characterization .............................................................................. 130 5.3.1.1 XRD and Physical Properties Analysis....................................................... 130 5.3.1.2 XPS Analysis .............................................................................................. 132 5.3.1.3 TEM/STEM-EDX Analysis and CO Uptake .............................................. 135 5.3.2 Catalyst Activity and Stability ................................................................................ 138 5.3.3 Used Catalyst Characterization ............................................................................... 142 5.3.3.1 XRD and Physical Properties Analysis....................................................... 142 5.3.3.2 XPS Analysis .............................................................................................. 142 5.3.3.3 TEM/STEM-EDX Analysis ........................................................................ 145 5.4 Discussion and DFT Analysis ..................................................................................... 148 5.5 Conclusions ................................................................................................................. 154 Chapter 6: The Effect of Other Heteroatoms (N and O) on Mo2C/APC Catalysts .............156 6.1 Introduction ................................................................................................................. 156 6.2 Experimental ............................................................................................................... 156 6.3 Results and Discussion ............................................................................................... 157 6.3.1 Catalyst Activities in HDN of Carbazole................................................................ 157 6.3.2 Catalyst Activities in HDO of Dibenzofuran .......................................................... 166 6.4 Conclusion .................................................................................................................. 168 Chapter 7: Conclusions and Recommendations .....................................................................169 7.1 Conclusions ................................................................................................................. 169 xv  7.2 Recommendations ....................................................................................................... 171 7.2.1 Mo2C Catalyst Properties ........................................................................................ 171 7.2.2 Promoter Effect in HDS Reaction ........................................................................... 172 7.2.3 Deactivation of Mo2C and Ni-Mo2C in HDS ......................................................... 173 7.2.4 Effect of N and O on Mo2C Catalysts ..................................................................... 173 7.2.5 Mesoporous Carbon Applications........................................................................... 174 Bibliography ...............................................................................................................................176 Appendices ..................................................................................................................................201 Appendix A Catalyst Preparation ........................................................................................... 202 A.1 Raw Petroleum Coke (PC) and Activated Petcoke (APC) ..................................... 202 A.2 Mo2C/C based Catalyst Precursors and Catalysts Preparation ............................... 204 A.3 Methodology of Petroleum Coke Activation .......................................................... 207 Appendix B Catalyst Characterization.................................................................................... 210 B.1 Physical Properties Test .......................................................................................... 210 B.2 XRD ........................................................................................................................ 214 B.3 XPS ......................................................................................................................... 217 B.4 CO Chemisorption .................................................................................................. 219 B.5 GC-FID/TCD .......................................................................................................... 223 Appendix C Sample Calculation ............................................................................................. 227 C.1 Feed and Products Calculation................................................................................ 227 C.2 Calculation Procedure of MoOxCy Formula ........................................................... 231 C.3 Calculation of Presulfiding Parameters for MoS2/AC Catalyst Preparation .......... 233 C.4 Stacking Degree (N) Calculation for MoS2 from TEM Images ............................. 235 xvi  C.5 Experimental Details for Chapter 4 ........................................................................ 236 C.6 Rate Constants Calculation Reported in Chapter 4................................................. 237 C.7 Calculation of Activation Energy of Carbon Hydrogenation in Chapter 4 ............ 238 C.8 Reaction Phase Determination from ASPEN Plus Calculation .............................. 240 C.9 System Dynamic Response in Fixed-bed Reactor .................................................. 243 Appendix D Kinetic Model Code ........................................................................................... 245 D.1 Matlab Code for HDO of 4-methylphenol in Batch Reactor .................................. 245 D.2 Matlab Code for HDS of Dibenzothiophene in Batch Reactor ............................... 252 D.3 Matlab Code for Deactivation Constant Calculation .............................................. 266 Appendix E Supplementary Figures and Tables ..................................................................... 272 E.1 Supplementary Information for Chapter 2 .............................................................. 272 E.2 Supplementary Information for Chapter 3 .............................................................. 276 E.3 Supplementary Information for Chapter 4 .............................................................. 282 E.4 Supplementary Information for Chapter 5 .............................................................. 289 E.5 Supplementary Information for Chapter 6 .............................................................. 298 Appendix F Error Analysis and Repeatability ........................................................................ 300 F.1 Carbon Balance ....................................................................................................... 300 F.2 Petcoke Activation Repeatability ............................................................................ 301 F.3 Reaction Repeatability ............................................................................................ 301 F.4 Statistical Analysis of Kinetic Model ..................................................................... 306 F.5 Characterization Repeatability ................................................................................ 309 Appendix G Mass Transfers and Heat Transfer Effects ......................................................... 314 G.1 External Mass Transfer Effect ................................................................................ 316 xvii  G.2 Internal Mass Transfer Effect ................................................................................. 319 G.3 Heat Transfer Effect ................................................................................................ 320 G.4 Precheck of Fixed-bed Reactor Operating Condition ............................................. 321 G.5 External and Internal Mass Transfers in Batch Reactor ......................................... 322 Appendix H Additional Calculation and Experimental Data ................................................. 323 H.1 Carbon Efficiency Calculation for Mo2C Formation .............................................. 323 H.2 Comparation between 10%MoS2/Al2O3, 10%MoS2/AC, and 10%Mo2C/AC Catalysts .............................................................................................................................. 323 H.3 Mesoporous Carbon Development by Ni-Mo2C on APC ....................................... 325  xviii  List of Tables Table 1.1 Comparison between bio-oil and crude oil [55]. ............................................................ 4 Table 1.2: Different thermal activation methods for the development of carbon material derived from petroleum coke. .................................................................................................................... 10 Table 1.3: Proximate, ultimate analysis and N2 specific surface area of delayed petroleum coke [99]. ............................................................................................................................................... 14 Table 2.1: Textural properties of Mo-based catalysts prepared by CHR method at different temperatures. ................................................................................................................................. 28 Table 2.2: The crystallite and particle size of Mo/AC prepared at different temperatures. ......... 30 Table 2.3: XPS analysis of Mo (3d) of 10% Mo/AC catalysts prepared at different CHR temperatures. ................................................................................................................................. 34 Table 2.4: The calculated formulas of Mo oxycarbide with different CHR temperatures and the CO uptake measurements ............................................................................................................. 36 Table 2.5: Kinetic rate constants of catalysts prepared at different CHR temperatures at different reaction temperatures. ................................................................................................................... 43 Table 2.6: 4-MP HDO conversion and product selectivity at different reaction temperatures for Mo2C catalysts prepared at different CHR temperatures. ............................................................. 43 Table 2.7: Pre-exponential factors and apparent activation energies extracted from 1st-order rate constants for the DDO and HYD of 4-MP over Mo2C catalysts prepared at different CHR temperatures. ................................................................................................................................. 44 Table 2.8: Kinetic rate constants for the conversion of 4-MP over different Mo-based catalysts at 350 ºC. ........................................................................................................................................... 45 Table 2.9: XPS Analysis of Mo 3d for 10% Mo/AC-650 catalyst after reaction. ........................ 48 xix  Table 3.1: XRD analysis of NixMo2C/AC-600 catalysts and Mo2C/AC-750 catalyst. ................ 65 Table 3.2: Catalyst composition of fresh Mo2C/AC-650 and Ni-Mo2C/AC catalysts. ................ 66 Table 3.3: Physical properties of fresh and used Ni-Mo2C/AC catalysts prepared at reduction temperatures of 550 oC and 600 oC. .............................................................................................. 70 Table 3.4: XPS and CHNS analysis of used (Ni)-Mo2C/AC catalysts with different Ni:Mo ratios and different reduction temperatures. ........................................................................................... 72 Table 3.5: XPS and CHNS analysis of used catalysts (Mo2C/AC-650 and Ni0.19Mo2C/AC-600) after HDS reaction in the batch reactor for different reaction periods. ........................................ 76 Table 3.6: Normalized intensity of selected ions (containing the most abundant isotope 32S) calculated based on TOF-SIMS spectra of used catalysts of 10%Mo2C/AC-650 and Ni0.19Mo2C/AC-600. ..................................................................................................................... 78 Table 3.7: DBT conversion for (Ni)-Mo2C/AC catalysts at 350 oC and an initial pressure of 2.1 MPa. .............................................................................................................................................. 83 Table 3.8: Kinetic model parameters estimated for the HDS of DBT after the 1st hour of the reaction in the batch reactor over Ni-Mo2C/AC catalysts at 350 oC and initial PH2=2.1 MPa. .... 89 Table 4.1: XPS analysis of Mo (3d) for Mo/APC samples prepared at different CHR temperatures. ............................................................................................................................... 108 Table 4.2: Physical properties of Mo2C/APC catalysts produced at different CHR temperatures...................................................................................................................................................... 111 Table 4.3: Bulk kinetic rate constants for the conversion of 4-MP over different Mo-based catalysts at 350 oC. ...................................................................................................................... 122 Table 5.1: Physical properties of catalyst precursors, fresh, and used Mo2C/APC catalysts with different Mo loadings. ................................................................................................................. 131 xx  Table 5.2: XPS analysis of fresh Mo2C/APC catalysts with different Mo loadings. .................. 134 Table 5.3: The particle size and CO uptake of fresh and used Mo2C/APC catalysts. ................ 138 Table 5.4: Conversion and product selectivity for the hydrodesuphfurization of dibenzothiophene over Mo2C/APC and MoS2/APC catalysts at 350 oC and 4.1 MPa after stabilization and TOS > 150 mins. ..................................................................................................................................... 141 Table 5.5: XPS analysis of used Mo2C/APC catalysts with different Mo loadings. .................. 144 Table 5.6: The calculated Gibbs free adsorption energy of DBT on S replaced Mo2C (101) surface with different adsorption angles. .................................................................................... 153 Table 6.1: Identified products from HDN of carbazole with Mo2C/APC catalysts at 350 oC and 4.1 MPa. ...................................................................................................................................... 158 Table 6.2: Conversion and product selectivity for HDN of CBZ over Mo2C/APC catalysts at 350 oC and 4.1 MPa. .......................................................................................................................... 162 Table 6.3: The calculated decay constants (kd) of Mo2C/APC catalysts with various metal loadings in HDN of carbazole at 350 oC and 4.1 MPa. .............................................................. 163 Table 6.4: Identified products from HDO of dibenzofuran with Mo2C/APC catalysts at 350 oC and 4.1 MPa. ............................................................................................................................... 166 Table A.1: CHNS/O wt% analysis of raw petroleum coke and activated petroleum coke. ....... 202 Table A.2: EDX analysis of raw petroleum coke (PC) and activated petroleum coke (APC). .. 203 Table B.1: Isotherm data both from experimentally measured and 2D-NLDFT-HS model fitted...................................................................................................................................................... 212 Table B.2: Peak table of adsorbed CO on in-situ synthesized 10% Mo2C/AC-600 catalyst. ..... 220 Table B.3: Peak table of adsorbed CO on reduced passivated 10% Mo2C/AC-600 catalyst. .... 222 Table B.4: Temperature program used for GC-FID/TCD analysis. ........................................... 223 xxi  Table B.5: The mole composition of the mixture gas. ................................................................ 224 Table B.6: Gas calibration result for carrier gas H2. ................................................................... 224 Table B.7: Calculation of mol. concentration of each gas in gas mixture. ................................. 225 Table B.8: Linear correlation between CH4 mole concentration and measured peak area of CH4...................................................................................................................................................... 226 Table C.1: Feed compositions for HDO of 4-methylphenol. ...................................................... 227 Table C.2: Feed compositions for HDS of dibenzothiophene. ................................................... 228 Table C.3: GC-MS calibration table for DBT concentration with the addition of DPE as internal standard (IS). ............................................................................................................................... 229 Table C.4: The calculation procedure for Mo2C and MoOxCy contents in 10%Mo/AC catalysts...................................................................................................................................................... 231 Table C.5: MoS2 stacking degree (N) of different used Ni-Mo2C/AC catalysts as reported in Chapter 3. .................................................................................................................................... 235 Table C.6: Experimental lists of HDO of 4-MP in Chapter 2. ................................................... 236 Table C.7: Activity, selectivity, and kinetic rate constants for HDO of 4-methylphenol reported in Chapter 2. ................................................................................................................................ 237 Table C.8: Experimental data of in-situ exit gas analysis of Mo800_APC catalyst by carbothermal hydrogenation reduction. ...................................................................................... 238 Table C.9: Mole flow rates of different components in HDS of dibenzothiophene in fixed bed reactor. ........................................................................................................................................ 240 Table C.10: Calculated heat and material balance table. ............................................................ 241 Table D.1: Products distribution of HDO of 4-methylphenol at 350 oC with 10%Mo/AC-650 catalyst. ....................................................................................................................................... 245 xxii  Table D.2: Calculated rated constants of APC supported Mo2C catalysts prepared at 600 and 650 oC................................................................................................................................................. 251 Table D.3: DBT conversion and initial rate of reaction in HDS of DBT for 10%Mo2C/APC catalyst as a function of time on stream. ..................................................................................... 266 Table D.4: Calculated decay constant (kd) from exponential decay rate law for 10Mo2C/APC in HDS of DBT at 350 oC and 4.1 MPa. ......................................................................................... 271 Table E.1: Kinetic parameters measured for thermal reaction of 4-methylphenol. .................... 275 Table E.2: Elemental compositions of fresh and used Mo/APC (after HDO of 4-MP) samples with different CHR temperatures holding for 90 min. ................................................................ 284 Table E.3: H adsorption energy on MoO3 (010), MoO3 (010) with one oxygen vacancy and MoO3C (010). ............................................................................................................................. 288 Table E.4: Gibbs free adsorption energy of DBT on Mo-t1 site of Mo2C (101) clean surface with different orientations. .................................................................................................................. 296 Table E.5: Gibbs free adsorption energy of DBT on Mo-t1 site of S adsorbed Mo2C (101) surface with different orientations. .......................................................................................................... 297 Table F.1: Reactant (DBT) and products concentration for HDS of dibenzothiophene over Ni0.19Mo2C/AC-550 and Ni0.38Mo2C/AC-550 catalysts.............................................................. 300 Table F.2: Petroleum coke activation results from different batches. ........................................ 301 Table F.3: Kinetic parameters for HDS of dibenzothiophene in batch reactor. ......................... 302 Table F.4: Catalyst properties of 2%Mo2C/APC catalyst. .......................................................... 302 Table F.5: Experimental data of 2%Mo2C/APC in HDS of DBT at 350 oC and 4.1 MPa from experiment T8. ............................................................................................................................ 303 xxiii  Table F.6: Experimental data of 2%Mo2C/APC in HDS of DBT at 350 oC and 4.1 MPa from experiment T13. .......................................................................................................................... 304 Table F.7: Comparation between experimental and kinetic model fitted data of different products and reactant in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. .................................................. 306 Table F.8: Summary of ANOVA used calculation formula. ...................................................... 306 Table F.9: ANOVA analysis of DBT concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. ....................................................................................................................................... 307 Table F.10: ANOVA analysis of CHB concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. ................................................................................................................................ 307 Table F.11: ANOVA analysis of BPh concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. ....................................................................................................................................... 307 Table F.12: ANOVA analysis of THDBT concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. ................................................................................................................................ 308 Table F.13: Calculated error associated with the physical properties test of activated petroleum coke (APC).................................................................................................................................. 309 Table F.14: Calculated error associated with the CO uptake. ..................................................... 310 Table F.15: Calculated error associated with CHNS analysis for raw petroleum coke. ............. 310 Table F.16: Calculated error associated with EDX-mappings for raw petroleum coke and APC_800. .................................................................................................................................... 311 Table F.17: Calculated error associated with TOF-SIMS analysis for used 10%Mo2C/AC-650 and Ni0.19Mo2C/AC-600 catalysts after HDS of DBT in batch reactor (Chapter 3). .................. 312 Table F.18: Calculated error associated with in-situ exit gas analysis of 2%Mo2C/APC-700 by carbothermal hydrogen reduction. .............................................................................................. 313 xxiv  Table G.1: The details of catalyst bed, catalyst physical properties and related kinetic parameters...................................................................................................................................................... 315 Table G.2: The details of reaction conditions and feed properties as calculated from Aspen Plus...................................................................................................................................................... 316 Table G.3: The details of external mass transfer calculation by Mears criterion. ...................... 317 Table G.4: The details of internal mass transfer by Weisz-Prater criterion................................ 319 Table G.5: The details of heat transfer calculation by Mears Criterion...................................... 320 Table G.6: The geometry parameters for prechecking of fixed-bed reactor operating condition...................................................................................................................................................... 321 Table G.7: A detailed list of external and internal mass transfer coefficient calculation for Ni0.09Mo2C/AC-550 in HDS of DBT in batch reactor. ............................................................... 322 Table H.1: DBT conversion and products selectivity of 10%MoS2/Al2O3, 10%MoS2/AC, and 10%Mo2C/AC catalysts in HDS of DBT at 350 oC and initial pressure of 2.1 MPa.................. 324 Table H.2: Physical properties of APC supported Ni-Mo2C catalysts. ...................................... 325  xxv  List of Figures Figure 1.1 Reaction network of hydrodesulphurization (HDS) of dimethyldibenzothiophene (DMDBT) [33]. (Copyright © 2007, Elsevier Inc., reproduced with permission) ......................... 1 Figure 1.2: (a) Petroleum coke generation process; (b) physical appearance of raw petcoke. ..... 14 Figure 1.3: Schematic illustration of carbon supported Mo2C catalysts in the application of hydrotreating reactions.................................................................................................................. 20 Figure 2.1: XRD patterns of 10% Mo/AC catalysts prepared at different CHR temperatures under H2 .................................................................................................................................................. 31 ((◊) SiO2; (*) β-Mo2C). ................................................................................................................. 31 Figure 2.2: XPS narrow scan spectra deconvolution of Mo 3d and O1s for fresh 10% Mo/AC catalysts prepared at different temperatures: (a)10%Mo/AC-600; (b)10%Mo/AC-650; and (c)10%Mo/AC-700. ...................................................................................................................... 33 Figure 2.3: TEM images of 10%Mo/AC prepared at different reduction temperatures: (a, b) 10%Mo/AC-650 with insert of lattice fringe d-spacing estimated at 2.28 Å for the (101) plane; (c) 10% Mo/AC-675; (d) 10% Mo/AC-700; (e) 10% Mo/AC-750; and (f) 10% Mo/AC-800. ... 38 Figure 2.4: Experimental and model concentration data versus reaction time of different catalysts at different reaction temperatures: 4-methylphenol (■), DDO product (▲), HYD product (●), kinetic model fit (--). ..................................................................................................................... 40 Figure 2.5: Simplified kinetic steps of 4-methylphenol HDO showing 1st-order reaction paths for DDO and HYD over all Mo2C catalysts prepared at various CHR temperatures. (The product presented in the dashed box is an intermediate-product) .............................................................. 41 Figure 2.6: 1st-order pre-exponential constant for DDO and HYD over all Mo2C catalysts prepared at various CHR temperatures. (The error bar reflects the calculated 95% CI) .............. 49 xxvi  Figure 3.1: Schematic diagram of high pressure fixed-bed reactor for HDS of DBT. ................. 61 Figure 3.2: XRD patterns of Ni-Mo2C/AC catalysts with different ratios of Ni:Mo (0 ~ 0.76) prepared at different reduction temperatures: (a) Reduced at 550 oC; (b) Reduced at 600 oC. (◆) Mo2C; (o) Ni; (*) Ni6Mo6C2. ........................................................................................................ 64 Figure 3.3: Profile of detected CH4 (mol%) during carbothermal hydrogen reduction of the catalyst generation: (a) Mo2C/AC-650 (■); (b) Ni-Mo2C/AC with different ratios of Ni:Mo ((●) Ni0.09Mo2C/AC-600; (▲) Ni0.19Mo2C/AC-600; (▼)Ni0.76Mo2C/AC-600). ................................. 68 Figure 3.4: A correlation between mass burn-off (wt%) and formed CH4 (mol%) during CHR process. The solid line represents the correlation equation: Mass burn-off (wt%) = 7.5847 × CH4 (mol%), Std.Dev.= 0.4055. (R2=0.9918) ...................................................................................... 69 Figure 3.5: X-ray diffraction patterns for fresh and used Ni-Mo2C/AC catalysts. ....................... 71 Figure 3.6: XPS narrow scan spectra of used MoS2/AC, Mo2C/AC and NixMo2C/AC-y catalysts: (a, c) Mo 3d; (b, d) S 2p. (The dashed lines in (a, c) indicate the position of Mo2+ species from Mo2C; the dashed lines in (b, d) indicate the position of S2- species from MoS2.) ....................... 73 Figure 3.7: XPS narrow scan spectra of used Ni0.19Mo2C/AC-550 catalyst after HDS reaction for different reaction periods. (a) Mo 3d; (b) S 2p. ............................................................................ 74 Figure 3.8: TEM images and cluster size distribution of used Ni-Mo2C/AC catalysts with different Ni contents. (a) Ni0.02Mo2C/AC-600; (b) Ni0.09Mo2C/AC-600; (c) Ni0.19Mo2C/AC-600; (d) Ni0.38Mo2C/AC-600; (e) Ni0.44Mo2C/AC-600; and (f) Ni0.76Mo2C/AC-600. .......................... 80 Figure 3.9: Distribution of MoS2 layer numbers for used catalysts. (a) Ni0.09Mo2C/AC-600; (b) Ni0.19Mo2C/AC-600; (c) Ni0.38Mo2C/AC-600; (d) Ni0.44Mo2C/AC-600; (e) Ni0.76Mo2C/AC-600; (f) Enlarged Ni0.19Mo2C/AC-600; and (g) Enlarged Ni0.44Mo2C/AC-600. ................................... 82 xxvii  Figure 3.10: DBT conversion and product selectivity versus time on stream for Ni0.19Mo2C/AC-600 and Mo2C/AC-650 catalyst, measured in the down flow fixed-bed reactor is at 310 oC, 4.1 MPa, H2/feed volumetric ratio = 600 and LHSV = 8 h-1. ............................................................. 86 Figure 3.11: Simplified kinetic steps HDS of DBTshowing 1st order reaction paths over all Ni-Mo2C/AC catalysts with different ratios of Ni:Mo and different reduction temperatures. .......... 87 Figure 3.12: Kinetic parameters k1, k2 vs. Ni:Mo ratio of Ni-Mo2C/AC catalysts: (a) reduced at 550 oC; (b) reduced at 600 oC. (…Trend line) .............................................................................. 90 Figure 3.13: The selectivity to biphenyl (BPh), tetrahydro-dibenzothiophene (THDBT) and cyclohydrobenzene (CHB) vs. DBT conversion: (a-c) reduced at 550 oC; (d-e) reduced at 600 oC........................................................................................................................................................ 92 Figure 3.14: The correlation of Ni:Mo ratio to MoS2 stacking degree (N) (a) and the average particle size (b). ―is the fitted line. .............................................................................................. 96 Figure 3.15: High angle annular dark field scanning (HAADF-STEM) image and energy-dispersive X-ray (EDX) elemental mapping of used Ni0.19Mo2C/AC-600 catalyst for elements Mo, Ni, S, O, C, and Si. ................................................................................................................ 98 Figure 4.1: The CH4 concentration (mol%) in the U-tube reactor exit gas measured during CHR of Mo800_APC (□) and APC support (Δ). Inset: Particle size of generated Mo2C at different CHR temperatures (◆). .............................................................................................................. 106 Figure 4.2: TEM graphs of different samples: (a) Mo600_APC; (b) Mo700_APC; and (c) Mo800_APC. .............................................................................................................................. 109 Figure 4.3: Particle size distribution from TEM micrographs and fitted lognormal distribution: (a) Mo600_APC; (b) Mo650_APC; (c) Mo700_APC; (d) Mo750_APC; and (e) Mo800_APC...................................................................................................................................................... 110 xxviii  Figure 4.4: Arrhenius plot of temperature dependence in the range of 550~700 oC. ................. 117 Figure 4.5: Schematic representation of Mo species transformation during CHR process. (Sizes of Mo species particles are not drawn to scale.) ......................................................................... 119 Figure 4.6: Pore size distribution of Mo2C/APC catalysts prepared at different CHR temperatures calculated from NLDFT model. .................................................................................................. 120 Figure 5.1: XRD diffraction patterns of fresh and used Mo2C/APC catalysts: (◆) carbon support; (*) Mo2C. .................................................................................................................................... 130 Figure 5.2: The deconvolution of Mo 3d of fresh Mo2C/APC catalysts: (a) 2Mo2C/APC; (b) 5Mo2C/APC; and (c) 10Mo2C/APC. .......................................................................................... 134 Figure 5.3: TEM images and cluster size distribution of Mo2C/APC with different metal Mo loadings. (a) 2Mo2C/APC fresh; (b) 2Mo2C/APC used; (c) 5Mo2C/APC fresh; (d) 5Mo2C/APC used; (e) 10Mo2C/APC fresh; and (f) 10Mo2C/APC used.......................................................... 137 Figure 5.4: DBT conversion and selectivity of Mo2C/APC catalysts with different Mo loadings with TOS. (All the experiments were done with the same amount of Mo loading in the reactor): (a) 2Mo2C/APC; (b) 5Mo2C/APC; (c) 10Mo2C/APC; and (d) DBT conversion and HDS conversion for Mo2C/APC catalysts with different Mo loadings as a function of TOS. ............ 139 Figure 5.5: XPS spectra of Mo2C catalysts: (a) Survey scan of fresh and used 10Mo2C/APC catalyst; (b) Narrow scan of Mo 3d spectra for used Mo2C catalysts with different metal loadings and used 5MoS2/APC. ................................................................................................................ 143 Figure 5.6: High angle annular dark field TEM scanning image (HAADF-STEM) of used 2Mo2C/APC catalyst (a); (b-e) Energy dispersive X-ray (EDX) elemental mappings of Mo, S, O and C; and (f) Overlay of C, Mo, and S distributions. ................................................................ 146 xxix  Figure 5.7: (a) TEM image of used 10Mo2C/APC catalyst; (b) High angle annular dark field TEM scanning image (HAADF-STEM); and (c) Line scanning on selected particle. ............... 147 Figure 5.8: Simplified reaction pathway of dibenzothiophene HDS via HYD and DDS routes over all Mo2C/APC catalysts with different Mo loadings. ......................................................... 151 Figure 6.1: Reaction network of hydrodenitrogenation of carbazole over Mo2C/APC catalyst. 158 Figure 6.2: CBZ conversion and selectivity of Mo2C/APC catalysts with different Mo loadings with time on stream. (Mo loading held constant at 0.019 g in the reactor): (a) 2Mo2C/APC; (b) 5Mo2C/APC; (c) 10Mo2C/APC; and (d) CBZ conversion and HDN conversion for Mo2C/APC catalysts with different Mo loadings as a function of TOS. ....................................................... 160 Figure 6.3: Catalyst regeneration (Phase I and II) of 10%Mo2C/APC after HDN reaction of carbazole: (a) Conversion of CBZ; and (b) Selectivity of the products. .................................... 165 Figure 6.4: Reaction network of hydrodeoxygenation of dibenzofuran over Mo2C/APC catalyst...................................................................................................................................................... 166 Figure 6.5: DBF conversion and selectivity of Mo2C/APC catalysts with different Mo loadings with time on stream. (Mo loading held constant at 0.019 g in the reactor): (a) 2Mo2C/APC; (b) 5Mo2C/APC; and (c) 10Mo2C/APC. .......................................................................................... 167 Figure A.1: XRD scan of raw petroleum coke and activated petroleum coke. .......................... 202 Figure A.2: SEM graphs of raw petroleum coke (Left) and activated petroleum coke (Right). 203 Figure A.3: Raman spectroscopy of raw petroleum coke (PC) and activated petroleum coke (APC_800). ................................................................................................................................. 204 Figure A.4: Schematic illustration of the preparation of Mo2C/C catalysts in quartz U-tube. ... 207 Figure A.5: Schematic illustration of the tubular furnace for petroleum coke activation. ......... 209 Figure B.1: Gas physisorption isotherms. (Reprinted with permission from [186]) .................. 211 xxx  Figure B.2: Fitted isotherm by 2D-NLDFT-HS model. (“o” represents measured experimental data)............................................................................................................................................. 213 Figure B.3: Cumulative surface area by 2D-NLDFT-HS model. ............................................... 213 Figure B.4: XRD profile of 10Mo2C/APC_750 catalyst and Gaussian curve fitting for (002) (Left peak) and (101) (Right peak) planes. .......................................................................................... 215 Figure B.5: XPS survey scan of used 1%Ni-10%Mo2C/APC catalyst. ...................................... 218 Figure B.6: Quantification results of used 1%Ni-10%Mo2C/APC catalyst. (Note: the integration of each element was based on Figure B.2) ................................................................................. 218 Figure B.7: TCD signal vs. time for in-situ synthesized 10% Mo2C/AC-600 catalyst at 32 oC. 220 Figure B.8: TCD signal vs. time for reduced passivated 10% Mo2C/AC-600 catalyst at 33 oC. 222 Figure B.9: Linear correlation of set point vs. H2 flow rate. ...................................................... 225 Figure B.10: Linear correlation of measured peak area of CH4 from GC-FID vs. CH4 mole concentration. .............................................................................................................................. 226 Figure C.1: A linear correlation between area ratio of DBT to DPE and wt% of DBT to DPE in high concentration range. ............................................................................................................ 230 Figure C.2: A linear correlation between area ratio of DBT to DPE and wt% of DBT to DPE in low concentration range. ............................................................................................................. 230 Figure C.3: Arrhenius plot of temperature dependence in the range of 550~700 oC for CH4 formation. .................................................................................................................................... 239 Figure C.4: Calculated activation energy (Ea) for carbon hydrogenation. ................................. 239 Figure C.5: Aspen flowsheet for H2 and feedmix in a flash reactor. .......................................... 241 Figure C.6: Total C mol.% in fixed-bed system at 350 oC and LHSV=4 h-1. ............................ 243 xxxi  Figure E.1: XRD pattern of 10% Mo/AC-650 prepared in Ar atmosphere. (◊) - SiO2; (*) - MoO2...................................................................................................................................................... 272 Figure E.2: Product concentrations as a function of reaction time during 4-methylphenol hydrodeoxygenation at 350C and 4.3 MPa H2. () 10%Mo-AC600, (o) 10%Mo-AC-650 and () 10%Mo-AC-700. .................................................................................................................. 273 Figure E.3: Arrhenius plots for 1st-order rate constants of (a) DDO and (b) HYD reactions.  Data centered at T0 = 350C. ............................................................................................................... 274 Figure E.4: XRD patterns of calcined Ni-Mo2C/AC catalyst precursors. .................................. 276 Figure E.5: N2 adsorption-desorption isotherms of nitrogen at -193 oC for the AC support and Ni-Mo2C/AC catalysts...................................................................................................................... 277 Figure E.6: The effect of Ni:Mo on (a) surface area, (b) pore size, (c) Vmeso/Vtotal, for fresh and used Ni-Mo2C/AC catalysts at reduction temperatures of 550 oC and 600 oC. ― Trend line. ... 277 Figure E.7: Experimental and model concentration data versus reaction time of different Ni-Mo2C catalysts prepared at 550 oC: Dibenzothiophene ( DBT, ▼); Biphenyl (BPh, ●); tetrahydro-dibenzothiophene (THDBT, ▲); cyclohexylbenzene (CHB, ■). (a) Ni0.02Mo2C/AC-550; (b) Ni0.09Mo2C/AC-550; (c) Ni0.19Mo2C/AC-550; (d) Ni0.38Mo2C/AC-550; (e) Ni0.44Mo2C/AC-550; (f) Ni0.76Mo2C/AC-550. ............................................................................ 278 Figure E.8: Experimental and model concentration data versus reaction time of different Ni-Mo2C catalysts prepared at 600 oC: Dibenzothiophene ( DBT, ▼); Biphenyl (BPh, ●); tetrahydro-dibenzothiophene (THDBT, ▲); cyclohexylbenzene (CHB, ■). (a) Ni0.02Mo2C/AC-600; (b) Ni0.09Mo2C/AC-600; (c) Ni0.19Mo2C/AC-600; (d) Ni0.38Mo2C/AC-600; (e) Ni0.44Mo2C/AC-600; (f) Ni0.76Mo2C/AC-600. ............................................................................ 279 xxxii  Figure E.9: Correlation of Ni/Mo ratio determined by ICP and adsorbed S wt% determined by CHNS analyzer for Ni-Mo2C/AC reduced at different temperatures (□: reduced at 550 oC; ○: reduced at 600 oC). ...................................................................................................................... 280 Figure E.10: TOF-SIMS spectrum of used Ni0.19Mo2C/AC-550 catalyst based on spot of scan area. ............................................................................................................................................. 281 Figure E.11: Mo (3d) XPS narrow scan spectra deconvolution of APC supported Mo2C catalysts at different CHR temperatures: (a) Mo400_APC; (b) Mo500_APC; (c) Mo550_APC; (d) Mo600_APC; (e) Mo650_APC; (f) Mo700_APC; (g) Mo750_APC; (h) Mo800_APC. ........... 282 Figure E.12: The deconvolution of Mo 3d narrow scan spectra of fresh Mo/APC samples with different CHR temperatures holding for 90 min: (a) survey scan of Mo600_APC; (b) Mo600_APC; (c) Mo650_APC; (d) Mo700_APC. (“―” Mo2+; “―” Mo3+; “―” Mo4+; “―” Mo5+; “―” Mo6+) ........................................................................................................................ 283 Figure E.13: Isotherms of APC supported Mo2C catalysts with different CHR temperatures. .. 285 Figure E.14: Isotherms of APC supported Mo2C catalysts with different CHR temperatures holding for 90 min: (a) Mo600_APC-90; (B) Mo650_APC-90; (c) Mo700_APC-90. .............. 286 Figure E.15: H adsorption energy on MoO3 (010), MoO3 (010) with one oxygen vacancy and MoO3C(010). .............................................................................................................................. 287 Figure E.16: Experimental and model concentration data versus reaction time of different catalysts with different CHR temperatures. (a) Mo600_APC-90; (b) Mo650_APC-90; (c) Mo700_APC-90. (■) Reactant: 4-MP; (●) HYD products; (▲) DDO products..................... 288 Figure E.17: A correlation of Mo loadings for the fresh Mo2C/APC catalysts with various Mo loadings and IMo/IC. ..................................................................................................................... 289 xxxiii  Figure E.18: The deconvolution of S 2p of used Mo2C/APC catalysts: (a) 2Mo2C/APC; (b) 5Mo2C/APC; (c) 10Mo2C/APC; (d) 5MoS2/APC. ..................................................................... 290 Figure E.19: A correlation of Mo, C, and S from HADDF-STEM-EDX mapping with two selected areas. ............................................................................................................................. 291 Figure E.20: EDX mapping for two selected areas of used 2Mo2C/APC catalyst. .................... 292 Figure E.21: Terminal positions of Mo2C (101) surface used in DFT calculation. (Terminal Mo site: Mo-t1; terminal C sites: C-t1 and C-t2, respectively) ......................................................... 293 Figure E.22: Diagram of different S atoms replaced Mo2C (101) surface for DBT adsorption energy calculation (Top and side views). ................................................................................... 294 Figure E.23: Potential energy during dissociative adsorption of H2 on three surfaces: (a) clean Mo2C (101) surface; (b) S adsorbed Mo2C (101) surface; (c) S replaced Mo2C (101) surface. 295 Figure E.24: Designed experiments for N effect study of 10Mo2C/APC catalyst at 350 oC and 4.1 MPa: (a) naphthalene hydrogenation reaction at 250 oC and LHSV = 4 h-1; (b) Part I-naphthalene hydrogenation, Part II-HDN of carbazole at 350 oC and 4.1 MPa, Part III-naphthalene hydrogenation; (c) Part I-naphthalene hydrogenation; Part II-NH3/H2 treatment at 350 oC and 4.1 MPa; Part III-Naphthalene hydrogenation. ................................................................................. 298 Figure F.1: DBT conversion and products selectivity for 2%Mo2C/APC catalyst as a function of time on stream in two trails (T8 and T13). ................................................................................. 305  xxxiv  List of Nomenclature (h, k, l) Miller index A Cross section area of the reactor, m2 B Line broadening at half the maximum intensity, radians CDBT Bulk gas concentration of DBT at 350 oC and 4.1 MPa, kmol/m3 CH2 Bulk gas concentration of H2 at 350 oC and 4.1 MPa, kmol/m3 Ci Atomic concentration of an element, % Ĉp(H2) Heat capacity of H2 at reaction temperature, J/Kg.K Cwp Weisz-Prater criterion d Distance between atomic layers, Å D Mean crystallize size, nm Dbed Fixed-bed tube diameter, m DDBT-H2 Binary bulk phase diffusivity, m2/s DDBT-H2 Binary bulk phase diffusivity, m2/s Deff, DBT-H2 Effective diffusivity, m2/s Deff, rxn Effective diffusivity in this reaction, m2/s Deff,knudsen Effective Knudsen diffusivity, m2/s Dknudsen Knudsen diffusivity, m2/s dp Catalyst particle diameter, m dpore Pore diameter, Å Ea Activation energy, kJ/mol Eb Binding energy, eV xxxv  Ek Kinetic energy of the photoelectron, eV h Heat transfer coefficient, W/(m2.K) ID/IG  The ratio of the relative intensity of D band to G band Ii Peak intensity of element i K Dimensionless shape factor k1 Stabilized kinetic parameter, s-1 kc Mass transfer coefficient, m/s kd Decay constant, h-1 kDDO Kinetic rate constant for DDO route, mL/(gMo.min) kHYD Kinetic rate constant for HYD route, mL/(gMo.min) 𝑘𝐷𝐷𝑂𝑇  Kinetic rate constant for DDO route from thermal reaction, mL/(gMo.min) 𝑘𝐻𝑌𝐷𝑇  Kinetic rate constant for HYD route from thermal reaction, mL/(gMo.min) 𝑘𝐻𝑌𝐷𝑜  Pre-exponential factor for HYD route, mL/(gMo.min) 𝑘𝐷𝐷𝑂𝑜  Pre-exponential factor for DDO route, mL/(gMo.min) kt Thermal conductivity calculated by semi-empirical method for polyatomic gases, W/(m.K) Lbed Length of the catalyst bed, cm kLa Liquid side mass transfer coefficient, s-1 MC Mears’ criterion for external diffusion MC' Mears’ criterion for isothermal operation mcat Mass of loaded catalyst, g MDBT Mole weight of DBT, g/mol xxxvi  MDecalin Mole weight of Decalin, g/mol MH2 Mole weight of H2, g/mol Mmix Feed molecular weight, g/mol msic Mass of SiC, g mtotal Mass of loaded catalyst bed, g n Reaction order Nu Nusselt number Ø Work function from the XPS instrument ø Porosity or void fraction of packed bed  ø1 Thiele modulus for 1st order reaction øp Catalyst particle porosity  PDBT Partial pressure of DBT, atm PDecalin Partial pressure of decalin, atm PH2 Partial pressure of H2, atm Prxn Reaction pressure, atm Pr Prandtl number  Ptotal Total pressure in the system, atm PV Pore volume, cm3/g R Catalyst particle radius, m -rDBT(obs) Observed reaction rate, kmol/gcat.s Re Reynolds number Re’ Reynolds number considering void fraction xxxvii  Rg Gas constant, J/(mol.K) rHYD Rate of reaction for HYD route, min-1 rpore Pore radii of the catalyst, cm SA Surface area, m2/g Sc Schmidt number  Scat. Surface area of the catalyst by NLDFT model, m2/g SDDS/HYD Selectivity between DDS to HYD route Sh Sherwood number  Si Sensitivity factor for peak i T* Dimensionless temperature TDBT,c Critical point temperature of DBT, K To Center temperature of 350 oC Trxn Reaction temperature, K U Superficial gas velocity, m/s Vbed Total volume of catalyst bed, mL Vc Loaded catalyst volume, mL ṼDBT,c Critical volume of DBT, mL/g-mol Vmeso Mesopore volume, cm3/g Vmicro Micropore volume, cm3/g Vo Total pore volume of the catalyst, cm3/g Vsic Loaded inert volume, mL Vtotal Total pore volume, cm3/g iii  X4-MP 4-methylphenol conversion, % ii  Greek letters γ Shape factor  ΔHrxn Heat of reaction, kJ/mol εDBT/ĸ Lennard-Jones parameters for DBT/Boltzmann's constant, K εDBT-H2/ĸ Lennard-Jones parameters for DBT-H2/Boltzmann's constant, K εH2/ĸ Lennard-Jones parameters for H2/Boltzmann's constant, K η Internal effectiveness factor θ Angles of incidence λ Wavelength of the incident X-ray beam, cm-1 νmix Kinetic viscosity of mixture, m2/s ρb Bulk density of catalyst bed, g/cm3 ρbsic Catalyst bed density with SiC and catalyst, g/cm3 ρc Catalyst density, g/cm3 ρmix Density of feed mixturedensity at 350 oC and 4.1 MPa, kg/m3 ρsic SiC density, g/cm3 ρsolid Solid catalyst density, g/cm3 σ DBT Lennard-Jones parameters for DBT/characteristic length, Å σ DBT-H2 Lennard-Jones parameters for DBT-H2/characteristic length, Å σ H2 Lennard-Jones parameters for H2/characteristic length, Å σc Constriction factor τ Tortuosity factor ΩD, DBT-H2 Collision integral, calculated by ignore the last two terms iii  Ω Overall effectiveness factor γmix Volumatic flow rate, cm3/s µmix Dynamic viscosity of the mixture, kg/m.s  iv  List of Abbreviations 10%Mo2C/AC-650 Activated charcoal supported Mo2C with 10 wt% of Mo loadings 2-CHP Phenol, 2-cyclohexyl- 2D-NLDFT 2-D Non-local density functional theory 2D-NLDFT-HS 2-D Non-local density functional theory with heterogeneous surface 4,6-DMDBT 4,6-dimethyldibenzothiophene 4-MP 4-methylphenol AC Activated charcoal AHM Ammonium heptamolybdate tetrahydrate APC Activated petroleum coke APC_800 Activated petroleum coke at 800 oC Ar-OCH3  Methoxy group Ar-OH  Phenolic group  Ave. (µ) Average value B.E. Binding energy, eV BCH Bicyclohexyl BET Brunauer-Emmett-Teller BPh Biphenyl CBZ Carbazole CH Cyclohexane CHB Cyclohexylbenzene CHBA 2-Cyclohexyl-benzenamine v  CHCHE Cyclohexyl-cyclohexene CHR Carbothermal hydrogen reduction CHX Hyrocarbons CNF Carbon nanofiber CNT Carbon nanotubes CPCHX Cyclopentylmethyl-cyclohexane CUS Coordinatively unsaturated sites DBF Dibenzofuran DBT Dibenzothiophene DDS Direct desulfurization dfBG Degrees of freedom between groups DFT Density Functional Theory dfWG Degrees of freedom within groups DHCBZ Dodecahydro-1H-carbazole DI water Deionized water DMBP Dimethylbiphenyl DPM Diphenylmethane EDX Energy dispersive X-ray FANOVA F value by ANOVA GC/FID Gas chromatography-flame ionization detector GC/MS Gas chromatography-mass spectrometry GC/TCD Gas chromatography-thermal conductivity detector vi  GHG Greenhouse gas HAADF-STEM High angle annular dark field-scanning electron microscopy HCH Hexylcyclohexane, C12H24 HDN Hydrodenitrogenation HDO Hydrodeoxygenation HDS Hydrodesulphurization HYD Hydrogenation ICP-OES Inductively coupled plasma-optical emission spectroscopy LHSV Liquid-hourly space velocity L-M model Levenberg-Marquardt nonlinear regression methodology Mo2C Molybdenum carbide MoOxCy Mo oxycarbide MoOxSy Mo oxysulfide MoS2/AC Activated charcoal supported MoS2 MSBG Mean square between groups MSWG Mean square within groups MWNT Multiwalled carbon nanotubes N Stacking degree of MoS2 Ni-Mo2C Ni promoted Mo2C catalyst NLDFT Non-local density functional theory ODEs Ordinary differential equations P/Po Relative pressure vii  PC Raw petroleum coke (petcoke) RI The normalized ion intensity yield RI' The distribution of identified Mo oxysulfide species SEM Scanning electron microscope SEM-EDX Scanning electron microscopy-energy dispersive X-ray spectroscopy SSBG Sum of squares between groups SST Total sum of squares  SSWG Sum of squares within groups Std. Dev. (SD) Standard deviation TEM Transmission electron microscopy THCZ Tetrahydrocarbazole THDBF Tetrahydrodibenzofuran THDBT 1,2,3,4-tetrahydrodibenzothiophene TOF-SIMS Time of flight secondary ion mass spectrometer TPD Temperature programmed desorption TPR Temperature programmed reduction TPR Temperature programmed reduction VASP Vienna ab initio simulation package XPS X-ray photoelectron spectroscopy XRD X-ray diffraction   viii  Acknowledgements First and foremost, I would like to express my deepest and sincerest gratitude to my supervisor, Professor Kevin J. Smith. Thank you for offering me this precious opportunity to study in UBC and explore the catalysis world together with you. Professor Smith’s profound knowledge, professional guidance, and unconditional support have inspired me a lot during my whole PhD study and make me grow into an independent researcher. His great personality and wisdom also positively influence me a lot. I would follow his advice in my future career and life.   Next, I would like to thank my committee members Professor Naoko Ellis from the Department of Chemical and Biological Engineering and Professor Keng C. Chou from the Department of Chemistry at UBC for their valuable comments and kindly support in the completion of this work. Prof. Ellis has kindly granted me access to use her laboratory tubular furnace for petroleum coke activation, which helped me a lot at the initial of my study period. Prof. Chou has provided me valuable points every time on our annul committee meeting, which encourage me to thinking from a new point of view. My doctoral study has been made much easier under your help.  Furthermore, I would like to acknowledge the financial support provided by the Natural Sciences and Engineering Research Council of Canada (NSERC) and SHELL Canada. I am also grateful to the University of British Columbia (UBC) for a four-year fellowship award and the collaboration with the University of Calgary (UC).   ix  Moreover, I would like to thank all individuals who have generously and kindly helped me with during my study with their expertise: Dr. Ken Wong (UBC) and Dr. Philip Kubik (SFU) for XPS analysis and training; Dr. John Kim for TOF-SIMS measurements; Dr. Xin Zhang (SFU) for HRTEM analysis and training; Lan Kato for XRD training; Maureen Soon for ICP measurement. I am also grateful to all CHBE office staffs (Marlene Chow, Amber Lee, Lori Tanaka, Michelle Pang, Kristi Chow, William Wijaya etc.) and technical support personnel (Richard Ryoo, Miles Garcia, Serge Milaire, Doug Yuen, Ken Wong, Richard Zhang etc.) for their assistants. Besides, I wish to thank all CHBE Catalysis group members (Shida Liu, Dr. Ross Kukard, Dr. Pooneh Ghasvareh, Chujie Zhu, Alex Imbault, Lucie Solnickova, Rubenthran Govindarajan, Dr. Mina Alyani, Lingxiu Zhu, Xu Zhao, Yanuar Philip Wijaya, Dr. Ali Alzaid, Majed Alamoudi, Hamad Almohamadi, Abdullah Althobaity, Dr. Rahman Gholami Shahrestani) and visiting scholars (Andreas Geiger, Xin Wang, Yunhua Li, Patrick Neumann and Rubenhran Govindarajan), who helped and supported me with within the past four years.   Last but not least, a special big thank you to my husband, partner, and best friend, Shida Liu, for all the encouragement, supporting and caring during the journey of my PhD life. Without him, I couldn’t get this far. Also, I like to express my gratitude to my beloved parents for their loving consideration and great confidence in me all the time.  ii  Dedication   To my beloved parents  and  my husband, Shida    1  Chapter 1: Introduction 1.1 Background 1.1.1 Synthesis Methods of Molybdenum Carbide (Mo2C) Catalysts Transition metal carbide catalysts, such as W2C, first reported by Levy and Boudart [1], are formed when interstitial C atoms are incorporated into the lattice of transition metals (Mo, Fe, or W) [2]. The presence of C may change the electron distribution of tungsten, making it more like platinum. In Levy and Boudart’s study, the Pt-like behavior of W2C was demonstrated for three reactions: (1) H2 + O2  H2O at room temperature, (2) H2 reduction of WO3 with water, and (3) isomerization of 2, 2-dimethylpropane to 2-methylbutane. Following the discovery by Levy and Boudart [1], metal carbide catalysts have been widely studied. The low cost of Mo compared with noble metals is another advantage. Among metal carbides, molybdenum carbide (Mo2C) shows better catalytic activity and selectivity for hydrogenation [3]. Studies focused on Mo2C as catalyst for isomerization [4, 5], ammonium decomposition [6], the water gas shift reaction (WGS) [7, 8], various electrocatalytic reactions [9], and for hydrotreating [3, 10, 11]. For example, Bouchy et al. [5] pointed out the high selectivity of alkane isomerization on Mo oxycarbide and a detailed explanation of the linkage between Mo oxycarbide structure and catalytic performance was reported. Also, for NH3 decomposition, Mo2C has shown a high H2 production rate compared to other Mo-based catalysts [6]. Sabnis et al. [7] reported a high WGS reaction rate on Pt/Mo2C catalyst because of the synergistic interaction between Pt-Mo alloy and Mo2C domains. The study on Mo2C was also extended to electrocatalysis due to the low cost and less attenuation in acid solution as presented by Chen et al. [9] for the hydrogen evoluation reaction.Mo2C can also be used as a catalyst support [12, 13].  2  The conventional method of preparing metal carbides is by temperature-programmed reduction of a metal oxide precursor employing a mixture of a hydrocarbon gas (CH4, C2H6, C4H10, C6H6 etc.) and hydrogen [14-18]. However, polymeric carbon, a by-product of the hydrocarbon reactions, may contaminate the carbide surface and block active sites, while the high temperature reduction (up to 927 oC) typically yields bulk phase carbides with low intrinsic surface area. Solid carbon can also be used as the carbon source, as reported by Jongerius et al. [19]; wherein, a carbon supported Mo precursor was used for Mo2C synthesis in N2 at temperatures ≥ 1000 oC. However, the high temperature treatment can impact the carbon support, decreasing surface area and collapsing pores. An alternative preparation method uses carbothermal hydrogen reduction (CHR), employing a carbon support and pure H2 to generate the metal carbide at a reduction temperature of 700 ~ 800 oC. This method yields nanoparticles of Mo2C, as first demonstrated by Mordenti et al. [20]. Similar results were reported by Liang et al. [21]. Thus, CHR is a relatively facile method to synthesize carbon supported Mo2C catalysts. This carbothermal hydrogen reduction (CHR) synthesis of Mo2C that uses the solid carbon support as the carburization agent, is more facile than conventional TPR methods using gas phase hydrocarbons. The present study is focused on the synthesis of Mo2C catalysts using the CHR method with the goal of improving the catalyst properties and hence activity. The study also emphasizes the use of different carbons as support material, especially petroleum coke.   The catalytic properties of Mo2C are closely related to the surface state of active sites as affected by the carburization conditions and preparation methods. Several parameters such as metal loading, heating rate, carburization atmosphere and temperature can be adjusted to impact the carburization process [22-25]. As reported by Frank et al.[24], Mo loading and carburization 3  temperature affected the structure of the synthesized catalysts, which in turn impacted their performance in steam reforming of methanol.   Metal promoters can also be introduced to the CHR synthesis process to increase the Mo2C dispersion by reducing the temperature required for the formation of the metal carbide. Liang et al. [26] reported that the presence of Co decreased the CHR temperature for Mo2C formation to 600 oC, at a Co/Mo molar ratio of 1.0. They concluded that the Co increased the formation of CHx species during CHR, facilitating carburization of the Mo. Similar effects have been reported in the synthesis of W2C by CHR using Ni as the promoter [27, 28]. The presence of the 2nd metal has also been shown to provide synergistic effects in the case of Co doped Mo2C catalyst where the chemoselective synthesis of various arylamines from their corresponding nitroarenes was used as the probe reaction, with the Co-Mo2C/AC catalyst having significantly higher activity than the Mo2C/AC alone [29]. Moreover, the bimetallic carbides formed in the presence of Mo and other transition metals [30, 31] such as Fe, Co and Ni, show effective catalytic activity in hydrotreating reactions. In the present study, Ni promotion of a Mo2C catalyst supported on carbon is reported, emphasizing the effect of the promoter on the properties of the catalyst and the resulting activity so as to build catalyst property activity relationships.    1.1.2 Carbon Supported Catalysts and their Application in Hydrotreating Reaction 1.1.2.1 Hydrotreating Technology Hydrotreating is an oil refining technology where the feedstock is thermally treated under H2 environment at a temperature of 300 ~ 450 oC and pressure of 0.7 ~ 15 MPa in the presence of a hydrogenation catalyst. Usually, there are two objectives of hydrotreating. One is to remove 4  heteroatoms (S, N and O) and metals (V or Ni etc.); the other is to stabilize the reactive fragments by terminating many coke-formation reactions in hydrocracking and saturating aromatic compounds. The oil quality can be improved after hydrotreating without significantly changing the boiling range, thus it is often referred to as a nondestructive hydrogenation.   1.1.2.2 Carbon Supported Catalysts in Hydrodesulphurization (HDS) and Hydrodenitrogenation (HDN) Hydrodesulphurization (HDS) and hydrodenitrogenation (HDN) are effective catalytic chemical processes that removes S and N from crude oil by reaction with H2 [32]. Among these two reactions, HDS is usually the main concern of refineries since the concentration of N (< 1 wt%) is relatively low compared to S (0.1~5.0 wt%) in real feedstocks. The aim of HDS is to improve the quality of the product for light feedstocks with S concentrations of ~ 100 ppm to 1 wt%; whereas, for heavy residue, with S content 4x’s that of conventional oil, HDS is accompanied by a high degree of hydrocracking with 70 ~ 90% of the S removed. The produced liquid products can be used for other refining operations.   1    Figure 1.1 Reaction network of hydrodesulphurization (HDS) of dimethyldibenzothiophene (DMDBT) [33]. (Copyright © 2007, Elsevier Inc., reproduced with permission)  Dibenzothiophenes (DBT) and 4, 6-dimethyldibenzothiophene (4, 6-DMDBT) have commonly been used as probe molecules to test the reactivity of HDS catalysts [33-36]. Usually, the HDS reaction occurs in parallel: (I) Direct desulfurization (DDS): which yields biphenyl (BPh) and dimethylbiphenyl (DMBP) as the main products; (II) Hydrogenation (HYD): which gives rise to hydrogenation products tetrahydro- or hexahydro-intermediates followed by desulphurization to cyclohexylbenzenes or bicyclohexyls (Figure 1.1). However, the contribution to the HDS of these two routes is totally different. For 4, 6-DMDBT, HDS usually occurs through the HYD route; whereas the DBT is more likely to convert through the DDS pathway. Also, because of the steric hindrance of alkyl groups, the reactivity of 4, 6-DMDBT is usually less than that of the DBT [36]. Similar to DBT, carbazole (CBZ) is a refractory N compound in HDN and this reaction usually occurs by hydrogenation of the aromatic ring prior to N removal [37].   2  Sulfided Co(Ni)Mo/Alumina catalysts are commonly used commercial catalysts for HDS and HDN, prepared by impregnation of alumina with metal salts that are subsequently converted to metal sulphides by heat treatment in the presence of a S source such as dimethyldisulphide (DMDS) or H2S. However, the results of intensive research [38, 39] showed that the metal-aluminum oxide complexes formed after impregnation and calcination are quite stable and may not be sulfided at all. Hence, catalysts prepared on carbon supports have received much attention as a potential commercial replacement with better metal sulphide formation during their preparation.  Carbon as a non-oxide support has several desirable properties, including high surface area, adjustable pore volume and size and also reduced coking propensity. Furthermore, the interaction between the C support and the active metal is weak and the recovery of active metals from spent catalysts could be more straightforward by burn off of the carbon. A study from Sakanishi et al. [40] demonstrated that the activity of NiMoS/C is higher than the commercial NiMoS prepared on alumina in the temperature range of 340 ~ 380 ºC. The DDS pathway was found to dominate in this temperature range, while the HYD route was preferred when temperature was lower than 340 ºC. A similar phenomenon was also reported by Farag et al. [41, 42] in which carbon and alumina supported CoMoS catalysts were compared for the HDS of DBT and 4, 6-DMDBT. The two supported catalysts showed similar selectivity but a significant difference in catalytic activity. In HDN, different transiton metal sulfides supported on carbon have been extensively studied by using quinolone as reactant at 380 oC and 5.5 MPa as reported by Prins et al. [43] The authors reported that high hydrogenation ability of the catalyst corresponded to high quinoline conversion.  3   Apart from traditional hydrotreating catalysts of sulfided Co(Ni)Mo, the application of Mo2C in HDS and HDN has attracted a lot of attention because of the noble metal-like behavior of the Mo2C [44] and its ability to resist sintering at high temperature [45]. Molybdenum carbide (Mo2C) has been shown to have 5x’s the activity of convential NiMoS/Al2O3 and MoS2/Al2O3 catalysts for the hydrotreating of a coal-derived gas oil, based on the number of active sites measured by CO uptake [3]. Park et al. [46] reported that supported Mo2C possess a higher activity than unsupported Mo2C and pointed out that unlike MoS2, Mo2C was more stable in initial reactivity, which can be attributed to a weakly acidic surface. Moreover, Costa et al. [47] studied the performance of alumina supported Mo2C for 4, 6-DMDBT in deep HDS reaction. The results indicated that the selectivity of DDS to HYD is higher than conventional CoMoS/Al2O3 or NiMoS/Al2O3 catalysts, indicating that the DDS route is the preferred pathway. Hence, the required H2 amount was less for Mo2C than that of sulfided Mo catalysts. Hynaux et al. [48] reported on HNO3 functionalized carbon black supported Mo2C. The prepared Mo2C/C had a CO uptake of 317 μmol/g, which is much higher than Mo2C/alumina of 40 μmol/g as reported by Costa et al. [49]. The increased activities were likely a result of the higher dispersion of molybdenum carbide on carbon, compared to the alumina. The performance of Mo2C catalyst has also been tested in several HDN reactions [50, 51]. As reported by Thompson et al. [50], the synthesized bulk Mo2C catalyst has shown a higher activity in pyridine HDN compare to the commercial CoMo and NiMo sulfides. Also, a detailed kinetic study of carbazole HDN on Mo2C was conducted by Szymanska et al. [51], that pointed out the difference from a commercial catalyst (biphenyl was detected).   4  1.1.2.3 Carbon Supported Catalysts in Hydrodeoxygenation (HDO) Bio-oils, derived from the fast pyrolysis  woody biomass (i.e. thermal cracking in the absence of O2 at 400- 600 oC and short residence time of the order of 1 second), are a promising alternative to fossil fuels since GHG emissions from bio-oils are 77 ~ 99% less than from fossil fuels [52]. Despite the benefits of bio-oil, the presence of high concentrations of oxygen in the form of furans, phenols, acids, and alcohols is problematic (Table 1.1) [53-55]. The oxygenated compounds are responsible for the instability, high viscosity, and high acidity of bio-oils which make them unsuitable for use without upgrading. Thus, hydrodeoxygenation (HDO) is becoming an important step in bio-oil upgrading [56].   Table 1.1 Comparison between bio-oil and crude oil [55].   Bio-oil Crude oil Water (wt%) 15-30 0.1 pH 2.8-3.8 ― ρ (kg/l) 1.05-1.25 0.86 µ50oC (cP) 40-100 180 HHV (MJ/kg) 16-19 44 C (wt%) 55-65 83-86 O (wt%) 28-40 < 1 H (wt%) 5-7 11-14 S (wt%)) < 0.05 < 4 N (wt%) < 0.4 < 1 Ash (wt%) < 0.2 0.1  The aim of HDO is to remove the excess amount of O contained in bio-oils under H2 atmosphere without affecting the boiling range of the oil. The choice of catalyst is crucial for HDO. 5  NiMoS/Al2O3 and CoMoS/Al2O3 were considered first, since they are conventional hydrotreating catalysts used in industry [36, 57]. They have been deemed rational choices for the HDO [58]. However, alumina supported catalysts have high coke yields caused by thermal instability of alumina due to the presence of large amounts of water in bio-oil. In addition, because of the acidity of the support, the HDO selectivity is affected by isomerization reactions that lead to a broad product distribution [59]. Furthermore, the stability of the sulfide catalysts is likely to be reduced due to the absence of S in the bio-oil. Consequently, carbon supported metal catalysts [60, 61] such as Mo, W, Ni and Co [62, 63], and Pd/C, Pt/C, Ru/C [64] have been considered because of the inert support and better water tolerance. Although the carbon support may decrease coking propensity during reaction [62, 65], there remains a need for less expensive catalysts with high HDO activity and low coke yield.   Transition metal carbides have received considerable attention because they have catalytic properties similar to noble metals [2, 14, 58, 66-68]. Mo2C catalysts have been shown to have high selectivities and activities for HDO, and most studies have focused on the HDO of phenolic model compounds, since these are known to be the most refractory species present in bio-oil. For example, in the HDO of guaiacol at 350 C and 5.5 MPa H2, Mo2C catalysts have high selectivity to phenol (45%) and methylated phenols (13%) since removal of the methoxy group (Ar-OCH3 bond strength 376 kJ/mol) is much more facile than the removal of the phenolic group (Ar-OH bond strength 456 kJ/mol) [19]. Consequently, Lee et al. [69] reported the HDO of anisole at 147 C on unsupported Mo2C catalysts, with > 90% selectivity to benzene. The HDO of mono-oxygenated 4-methylphenol on related Mo catalysts such as MoP and NiMoP occurs at higher temperature (up to 375 C and 4.4 MPa) yielding toluene and hydrogenated products such 6  as 4-methyl cyclohexene and 4-methyl cyclohexane, confirming that hydrogenation of the aryl ring also occurs under these more severe reaction conditions. Note that assessing HDO catalysts using the more refractory mono-oxygenated phenols is important because hydrogen consumption during HDO needs to be minimized and the conditions at which O removal occurs without over hydrogenation, needs to be determined for these refractory compounds. Furthermore, phenols are more likely to cleave C-O bonds without consuming significant amounts of H2. As reported by Ren et al. [70], C-O bond scission is much easier than C-C bond cleavage on Mo2C, resulting in less H2 consumption compared to precious metal catalysts. Also, the consistency of DFT and experimental data demonstrated that the activation barrier of C-O scission is less than for the C-C bond scission on Mo2C catalysts.   1.1.3 Activation Methods of Carbonaceous Materials Raw petroleum coke is a non-porous material and consequently, an activation process seems necessary if we want to use it as a high surface area catalyst support. Therefore, different activation methods for carbonaceous materials are reviewed here.   1.1.3.1 Physical Activation Physical activation [71] of carbonaceous materials is a two-step pyrolysis/activation method in which the material is first carbonized under inert atmosphere to order the carbon structure and create some initial pores. Following pyrolysis, the interior of the coke particle is activated at ˃ 800 ºC in the presence of H2O, CO2 or a mixture of them in order to develop a more porous structure with high surface area (See Eq.1-1 and 1-2 [72]). This treatment also influences the surface chemistry by adding functional groups. 7   𝐂 +  𝐇𝟐𝐎  →  𝐂𝐎 +  𝐇𝟐        (Eq.1-1) 𝐂 + 𝐂𝐎𝟐   →  𝟐𝐂𝐎         (Eq.1-2)  DiPanfilo and Egiebor [73] investigated steam activation of fluid coke and showed that a maximum SA (surface area) and PV (pore volume) of 318 m2/g and 0.24 cm3/g, respectively, can develop after the mass burn-off reaches ˃ 50 wt.%. The mass burn-off was mainly from the interior of the particle, thus the particle size distribution remained almost the same or slightly larger than the raw material. The CO2 activation of delayed petcoke was studied by Karimi et al. [74] who reported a maximum SA of 646 m2/g but with a mass burn-off of ~ 80% and an activation time of 12 h. Small et al. [75] studied the properties of two cokes by varying the activation atmospheres. The results indicated that the mixture of CO2 with H2O (0.5 mL/min) is the most effective activation atmosphere, which produced activated delayed petcoke with a SA of 578 m2/g and PV of 0.32 cm3/g, with a Vmeso/Vmicro ratio of 1:1. The mass burn-off at this condition was 60%. The authors also noted that the structure pattern of the raw petcoke affects the activation process, reflected in the different gasification rates between delayed and fluid coke. The same phenomenon was also observed by Wu et al. [76], using steam activation. Their results showed different SA of 800 m2/g and 500 m2/g, respectively, for two petcoke samples derived from different origins.   All of these studies indicate that physical activation of petcoke can generate activated carbon that has the possibility to be used as a catalyst support. However, the maximum SA and PV reported 8  to date were in the lower range when compared with commercial activated carbon as listed in Table 1.2.   1.1.3.2 Chemical Activation Compared with physical activation, chemical activation has more advantages, including relatively low activation temperature (450 ~ 900 ºC), short activation time, highly developed porosity and low mineral content of the resulting petcoke. In chemical activation, the carbon precursor is impregnated with activating agents by a heat treatment in the absence of air. Unlike physical activation, the two steps of carbonization and activation occur simultaneously. The porosity of carbon is affected by several operating variables, such as heating temperature and soaking time. Also, the activating agents can influence the pyrolytic decomposition process. KOH has been reported to be the most widely used alkaline metal compound in petcoke activation by several authors [77-79].  Zhang et al. [78] reported a method of using KOH as reagent for the preparation of adsorbents from petcoke. KOH was chosen as activating agent and mixed with the petcoke. The results indicated that the organic S in petcoke has a high reactivity, and the S could be removed in the temperature range 550 ~ 750 ºC during activation. About 23 ~ 33% of the total S present in the petcoke was captured by the K, and transformed into K2SO4 and K2SO3. Lee and Choi’s study [77] reported that the S removal reached 99.5% at 550 ºC when alkali metals were used to treat high S petroleum cokes, consistent with Zhang et al.’s result. Lee and Choi pointed out that the surface area can be developed during the desulphurization process, with a considerable increase in surface area until S removal exceeds 98%. Also, the results showed a better performance for 9  KOH than NaOH for petcoke activation as illustrated by the improved surface area. Furthermore, the particle size of alkali treated petcoke decreased to about half that of the raw coke, and most particles collapsed after activation, as observed by scanning electron microscopy (SEM). The different mechanisms between KOH and NaOH activation were investigated by Piñero and coworkers [80] based on multiwalled carbon nanotubes (MWNTs). The results indicated that KOH can effectively activate the carbon precursor by the intercalation of metallic K, especially when the carbon precursor has a high crystallinity, while Na is only effective for disordered materials.   Chemical activation under atmospheres other than N2 have also been studied. Xiao et al. [79] used a mixture of N2 and H2 (vol. ratio = 7:3) and enhanced the activation process with KOH by increasing the concentration of -CH- and -CH2- species and removing surface heteroatoms (N, S, O). The presence of H2 increased the yield of metallic K, which further improved intercalation in the carbon precursor. The results showed that both the SA and PV were developed. Moreover, Wu et al. [76] reported the combination of KOH and steam activation for petcoke activation. The results showed that the SA of petcoke can reach ˃ 3000 m2/g after 30 min in the presence of steam and KOH at 800 ºC, but at the expense of less than 30 wt.% yield.   Unlike physical activation, chemical methods present an ability to create a relatively high surface area petcoke (Table 1.2). However, most of the developed surface area resides in the microporous structure of the petcoke, which is undesirable for catalyst support use, especially for reactions with large molecules. Therefore, metal catalytic activation was considered to modify the pore structure for mesopore development.  10  Table 1.2: Different thermal activation methods for the development of carbon material derived from petroleum coke. Author Carbon precursor Reagents Activation condition Textural properties of activated sample Reference Temperature (ºC) Time (h) Atmosphere SA (m2/g) PV (cm3/g) Burn-off (%) DiPanfilo & Egiebor Fluid coke  (38-841 µm) H2O  (5.0 g/h) 850 6 N2 318 0.24 ~50 [73] Karimi et al. Delayed coke (45-90 µm) CO2 900 12 CO2 (150 mL/min) 646 -- 80 [74] Small & Hashisho Delayed coke (75-150 µm) CO2+Steam  (0.3-0.5 mL/min) 900 6 CO2 578 0.32 (Vmeso=0.15) 51-60 [75] Fluid coke (75-150 µm) 533 0.31 (Vmeso=0.16) Wu et al. Petcoke KOH(2:1)+Steam 800 0.42 N2 2500-3000 Ratio of Meso:Micro=1:1 70-75 [76] Zhang et al. High S petcoke (124-150 µm) KOH (2:1) 850 1 N2  (100 mL/min) 1281 0.72 30.6 [78] KOH (3:1) 2111 1.23 43.2 Lee & Choi Delayed coke NaOH (4:1) 550 1-2 N2 1350 0.60 (Vmeso=0.16) -- [77] KOH (4:1) 1980 Xiao et al. Petcoke  (˂150 µm) KOH (2:1) 780 1 N2:H2 =7:3 (140 mL/min) 2477 1.11 (Vmeso=0.19) 59 [79]  11  1.1.3.3 Catalytic Activation by Metals As mentioned in Wu’s report [76], the Ni and Cr originally present in petcoke can act as a catalyst to facilitate activation and pore development. Therefore, the effects of different catalysts on carbon activation for porosity development have been studied.  Transition metals are commonly used for carbon activation. Tomita et al. [81] reported that the presence of Ni accelerates the steam gasification of coal mainly by pitting holes into the char. SEM clearly records the morphology changes by Ni during gasification. However, no specific data on surface area or pore volume were reported by Tomita et al. [81]. Oya et al. [82] compared steam activation of resin-derived carbon fiber with and without Co. The results presented a noticeable development of mesopores in samples containing Co. At 800 ºC, the mesopore volume of carbon fiber increased from 0.03 to 0.34 cm3/g at similar mass burn-off, while the surface area of the non-catalyzed fiber was much higher owing to the large presence of micropores. Liu et al. [83] investigated the difference between Fe catalyzed activation with H2O versus CO2. The results showed that without catalyst, the obtained pitch-based carbon was mainly microporous, while with the help of the catalytic reaction, the average pore size distribution was 2 ~ 4 nm and 10 nm under CO2 and steam, respectively. The mesoporosity from CO2 activation was also higher than that with steam.   The use of Ca in the development of wide pores under steam or CO2 atmosphere was studied by Cazorla-Amoros et al. [84]. The results indicated that the Vmeso of the treated carbon increased from ~ 0.20 cm3/g to ~ 0.70 cm3/g for CO2, and from ~ 0.25 cm3/g to ~ 0.50 cm3/g for steam, indicating that the catalytic effect of Ca in CO2 activation played a much bigger role than that of 12  steam activation. Leboda et al. [85] reported that the ratio of micropore to mesopore volume increased with increased mass burn-off under steam activation without catalyst, while the presence of Ca led to the opposite effect, indicating that increased pore volume by catalytic treatment of Ca is mainly aimed at mesopore development. The improvement of mesopore surface area from 108 to 620 m2/g in the presence of Ca, with a limited increase from 197 to 270 m2/g for the non-catalyzed reaction points to a very significant effect of the Ca. Also, the presence of Ca increased the reaction rate by decreasing the activation energy of carbon gasification from 185 kJ/mol to 164 kJ/mol [85].   The use of rare earth metals, like Cerium (Ce) [86] and Yttrium (Y) [87], was also studied. Although cerium oxide is a seldom used catalyst for carbon gasification, Shen et al.’s research [86] reported CeO as a catalyst for the steam activation of activated carbon. The results showed that CeO was able to increase the mesopore volume when the temperature was < 740 oC and the distribution of pores was mainly between 4 ~ 10 nm. Li et al.’ [88] also reported that Vmeso increased from 0.03 cm3/g to 0.31 cm3/g, while the SA remained at the same level when using Ce. The pores were mainly distributed at 4 nm and 7 nm. The obtained data clearly indicate that the Ce catalyst affects the porosity of the carbon precursor the most. In a subsequent study by Shen et al. [89], metal oxides of Y and Ce were combined for steam activation. They reported that the Y and Ce oxides had a synergistic effect on mesopore development at temperatures below 740 oC.  A further increase in the pore size of activated carbon materials can also be achieved by metal (Ni, Pt, Ir, Ru and Fe) catalysed carbon hydrogenation reactions [90-94]. Studies suggest that the 13  hydrogenation of the carbon occurs through a sequence of steps that include dissolution of carbon into the metal at the carbon/metal interface, diffusion of carbon through the metal particle, and reaction of the carbon with dissociated hydrogen on the metal/gas interface [90, 91, 93, 94]. Two studies have concluded that hydrogenation of the carbon at the gas/metal interface is the rate determining step of the overall carbon hydrogenation reaction [94, 95].  To summarize, catalytic species show great potential to benefit the activation of petcoke as they increase the reaction rate and also change the pore size distribution. The development of selective porosity (mainly mesopores) is of interest to attain an optimum carbon material for application in catalysis.  1.1.3.4 Petroleum Coke Petroleum coke (petcoke) is a by-product of oil refining generated during thermal coking processes. Petcoke consists mostly of carbon (Figure 1.2 (a)) [96]. In the context of upgrading bitumen derived from the Canadian oilsands, there are two coking processes based on different reactors and operating conditions, where two types of coke can be produced: fluid coke and delayed coke. Both types of coke have similar chemical properties with a S content of ~7 wt.%, with 95% of the S present in organic molecules, and a similar content of mineral matter [97, 98]. Delayed coke derived from oil sands bitumen upgrading, supplied by Suncor Energy Inc., Calgary, Alberta, Canada and shown in Figure 1.2 (b) was used in this study. The component analysis of the raw petcoke is listed in Table 1.3 and other measured properties of the raw petcoke are reported in Appendix A.1.  14   Figure 1.2: (a) Petroleum coke generation process; (b) physical appearance of raw petcoke.  Table 1.3: Proximate, ultimate analysis and N2 specific surface area of delayed petroleum coke [99].  (Copyright © 2013, Wiley Online Library, reproduced with permission) Parameter Delayed petroleum coke Proximate analysis (wt.%) Moisture 0.3 Volatile matter 9.4 Fixed carbon 87.1 Ash 3.2 Ultimate analysis (wt.%) C 84.3 H 3.4 N 1.8 S 6.7 O a 0.6 SBET (m2/g) 0.5 Note: Particle size of delayed petroleum coke is 90~150 µm. O content was calculated by difference: O = 100 - (C+H+N+S+ash)  Significant efforts have been made toward coke utilization in a wide range of technologies. One application is through combustion and gasification of oil sands coke for power generation or synthesis gas production [97]. Also, petcoke is a potential material to produce electrodes, 15  adsorbents, and catalyst supports [78, 100, 101]. However, the high sulfur content and low reactivity have limited its application in energy generation [102]. The aim of present study is to modify petcoke into a carbon support that could be used in catalysis.   1.1.4 Summary  From the above introduction, molybdenum carbide synthesis methods were reviewed, pointing out the advantages of carbothermal hydrogen reduction (CHR) for Mo2C/C catalyst synthesis. However, there is no research to clearly explain the process of Mo2C active site formation, and also no study focused on the simultaneous C hydrogenation. To address this knowledge gap, a more detailed study of Mo2C/C catalyst formation is required to provide insight into CHR so that the synthesis can be controlled to yield desired products.  Also, it is clear that the properties of the raw petcoke will have an impact on the properties of the activated petcoke. Three conventional carbon activation methods have been described although none are able to produce mesoporous carbon in high yield. Consequently, new activation methods need to be developed to enhance yield of mesoporous carbons for applications to catalysts.   Finally, the application of carbon supported catalysts in hydrotreating reactions was reviewed and the potential of Mo2C/C catalysts was described. The link between Mo2C catalyst properties and catalytic activity in HDS and HDO needs to be better understood. In the present work, a focus on the dynamic transition of Mo2C/C catalysts in the presence of S, O and N has been examined so that the stability of these catalysts to real feedstocks can be better understood.  16   1.2 Objectives of the Thesis The overall objective of the present study is to develop Mo2C/carbon catalysts utilizing an activated petroleum coke as the support, especially aimed at Mo2C mesoporous carbon supported catalysts for hydrotreating reactions. Several tasks were identified to meet this objective.    Determine the viability of Mo2C and Ni-Mo2C catalyst synthesis by CHR using activated charcoal (AC) as the carbon source. In the first phase of the study, AC was used as the carbon source since methods of activation of the raw petcoke were under development. Quantify the catalyst activity and selectivity through the analysis of the reaction kinetics of the HDO and HDS of model reactants. Finally, relate the measured reaction kinetics to the catalyst properties so as to build catalyst activity property relationships for these catalysts.  Understand and develop methods of mesoporous carbon formation from the raw petcoke. Determine the mechanism of carbon and Mo transformation during the synthesis of Mo2C on mesoporous activated petcoke (APC) so that Mo2C/APC catalyst properties can be controlled during synthesis.   To investigate the effect of different heteroatoms (S, O and N) on Mo2C/APC catalysts during hydrotreating, with the objective of assessing catalyst stability during reaction and performance of the catalyst with real feedstocks.   17  1.3 Approach The approaches to achieve the objectives of this study were primarily experimental and are described as follows: 1) Investigate by experiment, the synthesis, characterization and testing of AC supported Mo2C and Ni promoted Mo2C (Ni-Mo2C) catalysts. The prepared catalysts will be tested for the HDO and HDS reactions using the model reactants of 4-methylphenol and dibenzothiophene, respectively. The catalyst tests will be conducted in an autoclave batch reactor. Kinetic analysis is used to link the catalyst properties to catalyst performance.  2) Convert raw petcoke into a mesoporous carbon material via a two-step sequential method of thermochemical and catalytic activation (CHR-carbothermal hydrogen reduction) treatment. Several characterization methods will be applied to determine the properties of the obtained catalysts. Additionally, the mechanism of the mesoporous structure development and Mo species transformation will be investigated. The properties of the Mo2C/APC catalysts will also be determined and assessed together with the activity data for the HDO of 4-methylphenol to determine the impact of petcoke properties on the catalyst activity.  3) Explore the effect of different heteroatoms (S, O and N) on Mo2C catalysts in a continuous flow fixed bed reactor. The particle size effect of Mo2C catalysts in the presence of S will also be determined by combining experimental and theoretical methods. A deactivation model will also be developed.   18  1.4 Outline of the Dissertation This dissertation is presented in seven chapters, as described below. The study conducted in Chapters 2 and 3 was based on activated charcoal (AC) as the starting point to test the viability of C supported (Ni)-Mo2C catalysts preparation. At the same time, a study of petcoke activation and the corresponding Mo2C catalyst synthesis was conducted as reported in Chapter 4. Chapters 5 and 6 describe further studies of Mo2C/APC catalysts focusing on the effect of different heteroatoms. Figure 1.3 illustrates the scope of the contents of the dissertation.   Chapter 1: Provides a brief introduction of Mo2C catalyst synthesis, hydrotreating technologies, petroleum coke, different activation methods, and the application of carbon supported catalysts. A brief summary is presented that identifies the knowledge gaps addressed in the present study. Also, the research objectives and research approach are described.  Chapter 2: Reports the catalytic performance of Mo2C/AC catalysts in HDO reaction of 4-methylphenol. A kinetic study was conducted to compare Mo2C catalyst prepared by different CHRs.  Chapter 3: Explores the application of synthesized Mo2C/AC and Ni-Mo2C/AC catalysts in HDS reaction of dibenzothiophene. The catalysts stability is also discussed. The deactivation of Ni-Mo2C/AC was firstly identified and the properties of the used catalyst were analysed.  Chapter 4: Presents the mesoporous carbon development and Mo transformation during the synthesis of Mo2C/APC catalysts by carbothermal hydrogen reduction. A comparation between Mo2C/APC and Mo2C/AC catalyst was conducted in HDO of 4-methylphenol. Chapter 5: Focuses on studying the S effect on Mo2C/APC catalysts with various particle sizes by combining experimental and DFT calculations.  19  Chapter 6: Extends the study to the effect of other heteroatoms (O and N) on Mo2C/APC catalysts with various particle sizes.  Chapter 7: Summarizes the conclusions of previous chapters and provide the recommendations for future work.20    Figure 1.3: Schematic illustration of carbon supported Mo2C catalysts in the application of hydrotreating reactions.  21  Chapter 2: Synthesis of Mo2C/AC Catalysts and the Application in HDO of 4-methylphenol 1 2.1 Introduction Bio-oil as a potential alternative fuel has attracted significant attentation recently. As noted in Chapter 1, Mo2C catalysts show promise for HDO reactions due to their Pt-like behaviour of the Mo2C. Thus, HDO was chosen firstly to test the activity of the synthesized Mo2C/AC catalysts. However, one important issue with respect to Mo2C catalysts for HDO is the role of O and its impact on the catalyst, given that Mo2C is highly oxophilic [66, 69] and that MoO3 and MoO2 are active for the HDO of 4-methylphenol [103]. Choi et al. [66] reported the synthesis of unsupported Mo2C with CH4/H2 gas mixture by variying the reaction temperature and CH4 concentration. The degree of carburization was shown to influence the number and quality of noble metal-like sites of Mo2C, which can be quantified by CO chemisorption [66, 69]. Increased noble-metal like behaviour occurs with increased carbidic carbon content and reduced residual oxygen content of the Mo lattice [66]. The reactive characteristics of Mo2C also change upon O adsorption, with reduced hydrogenation activity reported in the presence of O. Lee et al. [69] reported that the high deoxygenation selectivity (> 90%) and low hydrogenation selectivity of Mo2C for anisole HDO could be ascribed to O adsorption on the Mo2C during reaction. Hence, O content of the catalyst is critical in determining the Mo2C catalyst activity and selectivity.                                                   1 A version of this chapter has been published: H. Wang, S. Liu., and K. J. Smith, “Synthesis and Hydrodeoxygenation Activity of Carbon Supported Molybdenum Carbide and Oxycarbide Catalysts,” Energy & Fuels (2016) 30(7): 6039-6049. 22  Note that a commercial activated carbon (AC) was selected for this first attempt to synthesize Mo2C from a solid carbon source, since it was readily available and the methods used to activate the raw petcoke were not yet developed. In this chapter, CHR at relatively low temperature (600 ~ 800 C) has been used to prepare the Mo2C supported on AC. The transformation of the ammonium heptamolybdate tetrahydrate (AHM, (NH4)6Mo7O24.4H2O) to Mo2C during CHR is examined. At the chosen conditions the resulting catalysts are shown to contain both Mo2C and MoOxCy species and the impact of the residual O on the catalyst activity and selectivity is reported. The catalysts have been evaluated for the HDO of 4-methylphenol (4-MP), a representative phenol and among the most refractory compounds present in bio-oil [104]. Although several other studies have examined di-oxygenated compounds such as 2-methoxy-phenol, the choice of 4-methylphenol allows one to focus on the ability of the catalyst to break the stronger Ar-OH bond (versus the much weaker Ar-OCH3 bond [19, 105]) and hydrogenate the aryl ring. The kinetics of these competing reactions as a function of the O content of the catalyst is reported.  2.2 Experimental  2.2.1 Materials A commercial activated charcoal (Darco, 100 mesh, 1025 m2/g, pore volume: 0.85 cm3/g), denoted as AC, was used as the carbon support. The elemental composition (wt%) of the AC is: C: 84.34 %, H: 0.19 %, N: 1.3%; S: 0.18%; Ash 12.71 %. (Ash and S analyses were performed at ACL Laboratories by methods similar to ASTM 3177 and ASTM D5142, respectively).    23  Ammonium heptamolybdate tetrahydrate (AHM; (NH4)6Mo7O24∙4H2O) was used as the Mo precursor. Decalin (Aldrich, 99%) and p-Cresol (Aldrich, 99%) (also known as 4-methylphenol) were used as received as reaction solvent and model reactant, respectively.  2.2.2 Catalyst Preparation The commercial AC was dried at 110 C for 3 ~ 4 h to remove moisture prior to use. The procedure for preparing the catalyst precursor by wetness impregnation followed Liang et al. [21] with some minor modifications. Ammonium heptamolybdate (AHM) was dissolved in ultra-pure water and added to the dried AC support dropwise (the mass ratio of solution: AC is 5:1) to give a nominal loading of 5 or 10 wt% of Mo. The AHM impregnated AC precursor was equilibrated in an ultrasonic bath for 1 h at room temperature and then transferred to a round bottom flask and vacuum evaporated using a rotary evaporator to remove the water. Finally, the AHM/AC precursor was dried at 110 C overnight. The precursors are referred to as 5% or 10% AHM/AC. The calculation details associated with obtaining the required Mo loading, are provided in Appendix A.2 (I).  The Mo2C/AC catalyst was prepared by CHR of the AHM/AC precursor in a H2 flow of 100 mL (STP)/min. Approximately 0.9 g of the AHM/AC precursor was loaded into a U-tube and held in place with SiC placed on top of the precursor. The temperature was then increased at a rate of 1 ºC/min from room temperature to the final reduction temperature (chosen in the range 600 to 800 ºC) and held at the final temperature for 90 min. Subsequently the sample was quenched to room temperature under N2 to obtain the fresh catalyst that was placed in de-gassed acetone without air exposure, for storage. The catalyst is designated as 5% or 10% Mo/AC-X, where X represents 24  the reduction temperature. Further details on the catalyst preparation are given in Appendix A.2 (III).  2.2.3 Catalyst Characterization 2.2.3.1 N2 Adsorption and Desorption The specific surface area (SBET), pore volume and pore size distribution of the catalysts were determined from N2 adsorption-desorption isotherms measured at -196 C using a Micromeritics ASAP 2020 analyzer. The measurements were made immediately after catalyst synthesis with the samples transferred with minimal air exposure to a sample holder and degassed at 200 C under vacuum (100 µm Hg) for 4 h. The specific surface area was calculated from the measured N2 isotherm using the Brunauer-Emmett-Teller (BET) equation  applied in the relative pressure range (P/Po) of 0.01 ~ 0.20. The total pore volume (Vtotal) was obtained from the N2 uptake at a relative pressure of P/Po = 0.99. The average pore diameter is given by 4Vtotal/SBET. (See Appendix B.1 for details of the relative theory and calculation)  2.2.3.2 X-ray Diffraction (XRD) Powder X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Focus (0-20, LynxEye detector) diffractometer with Co Kα (λ=1.789 Å) radiation operating at 35 kV and 40 mA. In this measurement, the samples were passivated at room temperature prior to the tests. In one case, a fresh sample was analysed without passivation and the XRD pattern was the same as that obtained for the passivated sample, confirming that the air exposure during analysis did not affect the bulk properties of the catalysts and identification of the crystallite structure. (See Appendix B.2 for the calculation details of crystallite size) 25   2.2.3.3 X-ray Photoelectron Spectroscopy (XPS) A Leybold Max200 X-ray photoelectron spectrometer was used for relative abundance and chemical state determination of the surface components of the catalysts. Mg Kα was used as the photon source generated at 1253.6 eV. The pass energy was set at 192 eV for the survey scan and 48 eV for the narrow scan. The vacuum pressure was set at 2x10 -9 torr. The C1s peak at 284.5 eV was taken as reference to calculate binding energies and account for charging effects. Deconvolution of the XPS profiles was done using XPSPEAK41 software. Experimental peaks were decomposed through mixed Gaussian-Lorentzian functions (80%-20%) after Shirley background subtraction. The XPS catalyst sample was pressed onto adhesive tape and placed on the sample holder under ambient conditions prior to being transferred to the vacuum chamber. Exposure of the samples to ambient air during the preparation was minimized to less than 2 minutes. (See Appendix B.3 for details)  2.2.3.4 CO Chemisorption  The CO uptake of the catalysts was measured using a Micromeritics AutoChem II 2920 unit using pulsed chemisorption at 35 C. Preparation of the catalyst was done in-situ in the unit by passing 50 mL(STP)/min of 9.5 mol% H2/Ar while heating to the final temperature at 1C/min, and maintaining the final temperature for 90 min. The Mo2C/AC was then flushed in He at 400 C for 4 h and then cooled to room temperature. Next, 0.5 mL pulses of CO were injected into a flow of He (50 mL(STP)/min) and the CO uptake by the catalyst was measured using a TCD. CO pulses were repeatedly injected until no further CO uptake was observed after consecutive injections. Following this measurement the catalyst was cooled to room temperature and 26  passivated in 1 % O2 in N2 for 2 h. The CO uptake measurement was then repeated on the passivated sample. Prior to the measurement, the sample was pretreated to remove the passivation layer by passing 50 mL(STP)/min of 9.5 mol% H2/Ar while heating to 400 C at 10 C/min, and maintaining the final temperature for 2 h. The flow was then switched to He (50 mL(STP)/min) at 400 C for 4 h in order to remove the adsorbed species. After cooling in a He flow, 0.5 mL pulses of CO were injected into a flow of He (50 mL(STP)/min) and the CO uptake was measured using a TCD as before. (See Appendix B.4 for details)  2.2.3.5 O Content Anlysis A Perkin-Elmer 2400 series II CHNS/O analyzer, operated in O mode and a pyrolysis temperature of 1000 C, was used to determine the O content of the catalysts prepared by CHR. The acquired oxygen amounts were used to estimate the O content in MoOxCy.   2.2.3.6 Transmission Electron Microscopy (TEM) Transmission electron microscope (TEM) images were generated using a (JEOL) JEM 200 scanning transmission electron microscope operating at 200 kV. Samples were prepared by dispersing them in ethanol and sonicating for 15 min, before placing a drop of liquid on a holey-carbon grid and evaporating for 2 ~ 3 minutes in a vacuum chamber.   2.2.4 Hydrodeoxygenation of 4-methylphenol The HDO of 4-MP was carried out in a 300 mL stirred-batch reactor (Autoclave Engineers) to assess the catalytic activity of the fresh catalyst prepared at different reduction temperatures. The reactor was operated at 4.3 MPa H2 and a mixing speed of 1000 rpm. For the catalytic kinetic 27  study, 10%Mo/AC-600, 10%Mo/AC-650, and 10%Mo/AC-700 were tested at 325, 350 and 375 C, respectively, with the same final pressure of 4.3 MPa at each reaction temperature. In each case, 3.1 wt% 4-MP was dissolved in 100 mL decalin, to mimic the phenol content of a biomass-derived fast pyrolysis oil, and the catalyst mass was adjusted to ensure a constant Mo/4-MP mass ratio of 0.026 gMo/g4-MP (See Appendix C.1, Table C.1 for feed calculation). Although the decalin is expected to be inert under the chosen reaction conditions, there is the possibility that decalin can act as an H-donor and/or undergo cracking/isomerization reactions. Experimental analysis of the liquid product showed some cis-decalin/trans-decalin interconversion, but no other significant reactions of decalin occurred. The fresh catalyst was transferred to the liquid reactant in a glove bag under an Ar flow. Before reaction, the reactor was purged with N2 and flushed in ultra-high purity H2 (UHP-H2).   In order to exclude the effect of the reactor heat up period from the catalyst activity assessment, a liquid sample was withdrawn from the reactor and analyzed once the reactor reached the desired temperature, and this concentration is regarded as the concentration at t = 0 min. The H2 pressure, stirrer speed, system and heating jacket temperatures were continuously monitored during each experiment. The identification and quantitative analysis of the products were achieved using a Shimadzu (QP-2010-S) GC/MS analysis with a Restek RTX5 30 m × 0.25 mm capillary column based on an external calibration method.   28  2.3 Results 2.3.1 Catalyst Characterization 2.3.1.1 Textural Properties by BET Analysis and Carbon Loss during CHR Table 2.1 shows that the catalyst BET surface area decreases with increasing CHR temperature, with a significantly higher decrease between 650 and 700 ºC compared to other temperatures. During catalyst preparation by CHR, some portion of the AC is consumed as a consequence of carbon - H2 reactions (xC + 1/2𝑦H2  CxHy) and conversion of the Mo precursor to Mo2C.   Table 2.1: Textural properties of Mo-based catalysts prepared by CHR method at different temperatures. Sample SBET (m2/g) Vtotala (m3/g) Vmicrob (m3/g) Vmesoc (m3/g) Average pore sized (nm) Activated charcoal 1025 0.85 0.23 0.62 3.3 5% AHM/AC 910 0.79 0.19 0.60 3.5 10% AHM/AC 781 0.71 0.16 0.55 3.6 5% Mo/AC-750 955 0.92 0.15 0.77 3.9 10% Mo/AC-600 890 0.77 0.17 0.60 3.5 10% Mo/AC-650 845 0.76 0.14 0.62 3.6 10% Mo/AC-675 669 0.67 0.08 0.59 3.7 10% Mo/AC-700 655 0.61 0.10 0.51 4.0 10% Mo/AC-750 522 0.54 0.07 0.47 4.1 10% Mo/AC-800 283 0.35 0.03 0.32 4.9 a Vtotal was obtained from N2 uptake at P/Po= 0.99;  b Vmicro was obtained by t-plot method;  c Vmeso was calculated by Vtotal-Vmicro. d The average pore size was calculated by cylindrical model and the equation is as follows: Average pore size = 4𝑉𝑇𝑜𝑡𝑎𝑙𝑆𝐵𝐸𝑇.  29  Increasing CHR temperature increases the carbon loss from the AC, as shown by the burn-off mass loss reported in Table 2.2. For 10%Mo/AC-700 catalyst, approximately 44 (± 2) wt% of the dried precursor mass is lost during reduction, leaving 17.8 wt% Mo supported on the carbon. Since higher temperature removes more carbon, a higher concentration of CxHy in the reduction gas results, which increases the possibility of O replacement by C in the Mo precursor. Also, the Mo may facilitate carbon loss during reduction, although temperature plays a much more important role than the Mo:C ratio in the formation Mo2C, at least when the amount of Mo is in the range of 5 ~ 10 wt%. For example, the catalyst prepared using the 5%AHM/AC precursor has 50 (±2)% carbon burn-off at 750 ºC (resulting in a Mo:C mass ratio of  ≈ 0.11:1, Table 2.2); whereas, the catalyst prepared using the 10%AHM/AC precursor at 750 ºC has a 60 (±2) wt% carbon burn-off (mass ratio of Mo:C ratio ≈ 0.30:1, Table 2.2). Even though the ratio of Mo:C in the former catalyst is 3x’s lower than in the latter case, the crystallite size of Mo2C for both catalysts is similar at ~15 nm, as shown in Table 2.2. Hence we conclude that the Mo:C ratio has a negligible effect on the formation of Mo2C compared with the H2 reduction temperature. The most obvious illustration of this point is shown by the 5%Mo/AC-750 and 10%Mo/AC-650 catalysts that have similar Mo:C mass ratio of ~ 0.1:1, but the crystallite size is much larger (~15 nm) for the 5%Mo/AC-750 than the 10%Mo/AC-650 (~5 nm).  30   Table 2.2: The crystallite and particle size of Mo/AC prepared at different temperatures. Catalyst Crystallite sizea Particle/cluster sizeb Burn-off c Mo loading d Mass ratio of Mo:C nm nm wt% wt% 5% Mo/AC-750 14, 15, 15 N/A 50 10.0 0.11:1 10% Mo/AC-600 ―, ―, ―e ―f 11 11.2 0.12:1 10% Mo/AC-650 ―, ―, ―e 4~6 21 12.7 0.15:1 10% Mo/AC-675 12, 12, 11 2~4, 10~12 34 15.2 0.18:1 10% Mo/AC-700 15, 15, 12 10~15 44 17.8 0.21:1 10% Mo/AC-750 16, 15, 14 16~21 60 25.0 0.30:1 10% Mo/AC-800 21, 19, 18 ―, ―, ―g 62 26.3 0.36:1 a. It was calculated by Scherrer equation, representing crystallite sizes corresponding to (100), (002) and (101) planes, respectively; b. It was obtained from TEM; c. Burnoff mass (%) =mbefore CHR−mafter CHRmbefore CHR; standared deviation is around 2%; d. It was calculated by considering the carbon loss during CHR; e. Data were not available since there are no peaks observed from XRD;  f. Data were not available due to the amorphous structure of the sample; g. Data were not available due to the agglomeration of the sample.  2.3.1.2 XRD Analysis Figure. 2.1 shows that the hexagonal close packed (hcp) phase of β-Mo2C (PDF card # 00-035-0787) was successfully prepared by CHR of the 10%AHM/AC precursor reduced at different temperatures, with diffraction peaks of β-Mo2C clearly visible at 2θ = 40.11º, 44.40º, 46.07º, 61.35º, 72.87º, 82.97º, 86.58º and 89.51º and no other bulk phase of MoOx present. Peaks corresponding to the presence of SiO2 (PDF card # 00-046-1045) in the AC (17.1 wt% ash) were 31  also present, and increased in intensity with increased reduction temperature, a consequence of increased C loss at the higher CHR temperatures.    Figure 2.1: XRD patterns of 10% Mo/AC catalysts prepared at different CHR temperatures under H2 ((◊) SiO2; (*) β-Mo2C).  Table 2.2 summarizes the β-Mo2C crystallite size calculated from Scherrer’s equation, showing that the crystallite size increased with increasing CHR temperature and suggesting that higher temperature causes sintering of the β-Mo2C particles. Table 2.2 also shows reasonable agreement between the crystallite size determined by XRD and the cluster size determined by TEM, except for the 10%Mo/AC-650 catalyst, due to the inability of XRD to accurately determine crystallite size of ≤ 5 nm and for the 10%Mo/AC-800 catalyst, since it is difficult to measure the particle 32  size in TEM when the clusters agglomerate. Agglomeration of the β-Mo2C is also demonstrated by the decreased BET surface area and pore volume shown in Table 2.1 for the same catalysts.  2.3.1.3 XPS, O Analysis and CO Uptake Although XRD analysis confirms the presence of β-Mo2C, the presence of amorphous forms of MoOxCy or MoOx cannot be determined by XRD. Hence, XPS and O analysis were used to identify the presence of carbide and oxycarbide surface species and to determine the surface composition of the catalysts. Since all the samples were supported on carbon, it is not possible to obtain useful information from C 1s peaks. Consequently the XPS analysis focused on the Mo 3d spectra that consist of two peaks, because of spin-orbit (j-j) coupling, assigned to Mo3d5/2 and Mo3d3/2  and that are separated by 3.1 eV with a peak area ratio of 3:2 [106, 107].    33   Figure 2.2: XPS narrow scan spectra deconvolution of Mo 3d and O1s for fresh 10% Mo/AC catalysts prepared at different temperatures: (a)10%Mo/AC-600; (b)10%Mo/AC-650; and (c)10%Mo/AC-700.  The measured spectra (Figure 2.2) show no evidence of zero-valent Mo species (B.E. = 227.8 eV [108]) in the synthesized catalysts. Results of curve fitting applied to the linked doublets to decompose the Mo 3d spectra into five different Mo species with B.E. in the range of 228.0 eV to 232.4 eV, are summarized in Table 2.3. Using the approach reported by Izhar et al. [109] and Oshikawa et al. [110] to analyse the Mo 3d spectra, the Mo3d5/2 at B.E.= 228.4 eV with FWHM of 1.4 eV is assigned to a carbide phase. The Mo 3d5/2 spectra is fitted by Mo3+ at 229.0 eV, Mo4+ at 231.1eV, Mo5+ at 231.9 eV, and Mo6+ at 233.1 eV, with FWHM of 1.4 eV, 1.4 eV, 1.4 eV, 1.6 34  eV and 1.7 eV, respectively. These B.E.s are associated with MoOxCy. There are several reasons why the MoOxCy has a wide range of Mo 3d  B.E.s. Firstly, there is no visible peak for MoO3 in the XRD, yet Mo6+ and Mo5+ species are identified by XPS after MoO3 reduction, indicating that surface vacancies are generated. Depending on the nature and composition of the reducing gas, either MoO2 or MoOxCy forms [111]. In the present study, the CH4 formed during CHR is able to partially fill the vacancies and stop the transformation from MoO3 to MoO2, leaving Mo in a relatively stable phase with relatively high valence. However, Mo4+ does exist suggesting the possibility of MoO2, yet the amount is below the detection limit of the XRD. In this study all Mo4+ have been assigned to MoOxCy in order to simplify the analysis. In the absence of CH4, the MoO3 will be reduced directly to MoO2 as shown by the XRD patterns of the same AHM/AC precursor treated in Ar (Appendix E, Figure E.1). There is no obvious MoO2 peak in the XRD either, yet Mo4+, and Mo2+ to Mo4+ assigned to MoOxCy are present in the XPS. Puello-Polo and Breto [31] conclude similarly, although they proposed that the B.E.s from 229.3 eV~230.4 eV to be Mo4+ and the FWHM is larger than that observed in the present study.   Table 2.3: XPS analysis of Mo (3d) of 10% Mo/AC catalysts prepared at different CHR temperatures. Catalyst B.E. (eV) of Mo 3d5/2  Composition (mol. %) Mo2+ Mo3+ Mo4+ Mo5+ Mo6+ Mo2+ Mo3+ Mo4+ Mo5+ Mo6+ 10%Mo/AC-600 228.3 229.1 231.1 232.0 233.1 9.94 25.96 8.65 32.71 22.75 10%Mo/AC-650 228.3 229.0 231.1 231.9 233.1 23.69 18.01 3.79 31.96 22.55 10%Mo/AC-700 228.2 228.8 231.1 231.8 232.9 50.47 17.07 8.56 8.70 15.19  O analysis was used to determine the O content of the Mo oxycarbide species present in the catalysts after CHR by accounting for the O content of the AC (without Mo) after CHR. 35  Although there are a few minites exposure to air due to the sample transfer and preparation, it does not significantly affect the comparison between the samples. The O signal is mainly contributed to by the oxycarbide present in the catalyst samples. The measured O content (Appendix C.2, Table C.4) assigned to MoOxCy was combined with the Mo valence states determined by XPS, to estimate the ratio between oxygen and carbon in the MoOxCy, as shown in Table 2.4. The calculation (see Appendix C.2) assumes that the C has -4 valence, that the CHR at these temperatures does not reduce the SiO2 present in the AC and that the O is uniformly distributed in the bulk and surface of the catalyst. The formula of the MoOxCy reported in Table 2.4 represents the average composition of all the MoOxCy surface species present in the sample, and excludes the surface content of Mo2C. The data show that as the CHR temperature increases the O content of the MoOxCy decreases and the Mo2C content of the catalyst increases. At a CHR of 700 C, the Mo is distributed approximately equally between Mo2C and MoOxCy. The O1s XPS spectra (Figure 2.2) also confirm the presence of multi-oxidation states of Mo. From the report of Delporte et al. [106], the O1s peak at 532.7 eV with FWHM 2.5 eV is attributed to O atoms in carbonate species. The O1s binding energy for Mo-O species present in MoOxCy appears at 530.7 eV with FWHM 2.1 eV. Another O peak at 536.5 eV is assigned to Si-O, and although the position of this peak should be at 532.8 eV, due to charging it occurs at 536.5 eV.   36   Table 2.4: The calculated formulas of Mo oxycarbide with different CHR temperatures and the CO uptake measurements Catalyst O content from MoOxCya Formula Mo distribution CO uptake Mo2C MoOxCy Fresh Passivated wt% Mo mol% mol/gMo 10% Mo/AC-600 2.59 MoO1.64C0.33 9.9 90.1 147 94 10% Mo/AC-650 1.76 MoO1.16C0.61 23.7 76.3 186 98 10% Mo/AC-700 1.00 MoO0.83C0.90 50.5 49.5 156 58 a. O content from MoOxCy = Total O amount in samples – O amount from AC support; see Appendix C.2.  The CO uptake data reported in Table 2.4 also show an increase with increased CHR temperature and hence Mo2C content, however, at a CHR of 700 C the CO uptake decreases, presumably a consequence of the larger Mo2C particles at this temperature, as identified by XRD and reported in Table 2.2.   2.3.1.4 TEM Analysis TEM micrographs of the catalysts reduced at different temperatures are presented in Figure 2.3. At a reduction temperature of 650 ºC, the TEM micrograph shows that there is no clear boundary between carbide particles and the carbon support. However, careful observation of Figure 2.3 (b) shows the characteristic slabs of β-Mo2C with an interplanar distance of 0.228 nm. At 675 ºC, the Mo carbide is clearly identified with a bi-modal size distribution, namely the coexistence of small particles of 2 ~ 4 nm and 10 ~ 12 nm as shown in Figure 2.3 (c). However, distinct fringes of graphitic structures are also observed in this sample, indicating the generation of carbon 37  filaments. Above 675 ºC, the crystallization of the Mo2C increases as does agglomeration of the nanoparticles. At a reduction temperature of 700 ºC, an enlarged image of this sample yields a d-spacing of 0.228 nm, corresponding to the (101) plane of Mo2C and more layers are observed indicating more crystalline Mo2C in this catalyst.  38   Figure 2.3: TEM images of 10%Mo/AC prepared at different reduction temperatures: (a, b) 10%Mo/AC-650 with insert of lattice fringe d-spacing estimated at 2.28 Å for the (101) plane; (c) 10% Mo/AC-675; (d) 10% Mo/AC-700; (e) 10% Mo/AC-750; and (f) 10% Mo/AC-800. 39  2.3.2 Catalytic Performance in HDO of 4-MP 4-Methylphenol was chosen as model reactant to determine the activity of the prepared catalysts. The results of the activity tests at different reaction temperatures for catalysts prepared at different CHR temperatures are shown in Figure 2.4. The main products of 4-MP HDO at 350C and 4.3 MPa H2 are toluene, methylcyclohexane, 1-methylcyclohexene and 4-methylcyclohexene. The primary product is toluene which is produced by Ar-OH bond hydrogenolysis or direct deoxygenation (DDO). Hydrogenated products result from ring hydrogenation and rapid dehydration to produce 4-methylcyclohexene which is rapidly hydrogenated to methylcyclohexane (the hydrogenation route or HYD). The distribution of hydrogenated products is illustrated in Appendix E.1, Figure E.2 that shows that most of the HYD product consists of methylcyclohexane and 4-methylcyclohexene, with the latter product decreasing with increasing reaction time. For clarity of discussion, the total hydrogenated product concentration (HYD) is reported in Figure 2.4, which shows that the 4-MP conversion for 10% Mo/AC-650 is almost the same as that for the 10%Mo/AC-600, especially at low reaction temperatures. However, when the CHR temperature is 700 ºC, agglomeration of Mo2C reduces the catalyst activity.  40   Figure 2.4: Experimental and model concentration data versus reaction time of different catalysts at different reaction temperatures: 4-methylphenol (■), DDO product (▲), HYD product (●), kinetic model fit (--). 41  The kinetics of the reaction were determined for the 10%Mo/AC-600, 10%Mo/AC-650, 10%Mo/AC-700 catalysts operated at 325, 350 and 375 C. Similar to results reported previously on other Mo-based catalysts for the HDO of 4-MP at high temperatures (325 ~ 375 C) [103], the pseudo 1st -order rate equations for the DDO and HYD parallel reaction paths shown in Figure 2.5 were fitted to the data of Figure 2.4. Accordingly, the rate equation for the DDO is written as 𝑟𝐷𝐷𝑂 = (𝑘𝐷𝐷𝑂𝑇 + 𝑘𝐷𝐷𝑂𝐶𝑐𝑎𝑡)𝐶4−𝑀𝑃 and for the HYD 𝑟𝐻𝑌𝐷 = (𝑘𝐻𝑌𝐷𝑇 + 𝑘𝐻𝑌𝐷𝐶𝑐𝑎𝑡)𝐶4−𝑀𝑃 where 𝑘𝐷𝐷𝑂𝑇  and 𝑘𝐻𝑌𝐷𝑇  (min-1) are the 1st-order rate constants of the thermal reaction (without catalyst) (See calculation details in Appendix E.1, Table E.1), 𝐶𝑐𝑎𝑡 is the catalyst concentration in the reaction (gMo/mL) and 𝑘𝐷𝐷𝑂 and 𝑘𝐻𝑌𝐷 (mL/(gMo.min)) are the 1st-order rate constants of the catalytic reaction. Since the two reaction pathways proceed in parallel, according to Whiffen and Smith [112], the DDO/HYD selectivity ratio has been calculated as 𝑆𝐷𝐷𝑂/𝐻𝑌𝐷 = 𝑘𝐷𝐷𝑂/𝑘𝐻𝑌𝐷.   Figure 2.5: Simplified kinetic steps of 4-methylphenol HDO showing 1st-order reaction paths for DDO and HYD over all Mo2C catalysts prepared at various CHR temperatures. (The product presented in the dashed box is an intermediate-product) 42   The rate equations were incorporated into the batch reactor design equation and the parameter values estimated using a Levenberg-Marquardt non-linear regression methodology (See details in Appendix D.1). The rate constants for the thermal reactions (𝑘𝐷𝐷𝑂𝑇  and 𝑘𝐻𝑌𝐷𝑇 ) were estimated from a series of experiments done without catalyst, and the estimated values are reported in Appendix E.1, Table E.3. The estimated values of the catalytic rate constants (kDDO and kHYD) are reported in Table 2.5 and the DDO/HYD selectivity ratio S𝐷𝐷𝑂 𝐻𝑌𝐷⁄  is reported in Table 2.6. Comparison of the rate constants (on a per gram Mo basis) indicates that 10% Mo/AC-650 had the highest activity among the catalysts and that on all catalysts the DDO rate is significantly greater than the HYD rate. The data also show that the reaction temperature has a significant influence on the reaction rate as well and this impact is greater on the DDO route than the HYD. For example, the DDO rate increased 2 times as the reaction temperature increased from 325 oC to 350 oC, and ~3 times as the temperature increased from 350 oC to 375 oC. The data of Table 2.6 suggest a small increase in DDO selectivity with CHR temperature, although at higher reaction temperature there is a significant error associated with the calculated selectivity ratio. Note that the selectivity to toluene is high for the synthesized Mo2C of Table 2.5, which implies that the DDO reaction pathway is dominant and the removal of O from 4-MP occurs prior to ring saturation. The DDO selectivity of the as-synthesized catalyst is much higher than that reported on unsupported MoO2 [103], MoS2 [103, 113], MoP [103], sulfided CoMo/Al2O3 [114] and precious metal catalysts [115]. Wandas et al. [114] reported on the HDO of cresol over CoMo/Al2O3 and showed that the DDO/ HYD selectivity ratio was less than 0.25.  43  Table 2.5: Kinetic rate constants of catalysts prepared at different CHR temperatures at different reaction temperatures. Temperature, C 325  350  375  Catalyst kDDO kHYD  kDDO kHYD  kDDO kHYD   mL/(gMo.min)  mL/(gMo.min)  mL/(gMo.min)  10%Mo/AC-600 0.5140.050a 0.2010.046  1.2560.122 0.4330.110  3.5250.188 0.8120.170  10%Mo/AC-650 0.5210.052 0.2020.052  1.3310.080 0.3930.072  4.0890.742 0.8360.492  10%Mo/AC-700 0.4360.058 0.1290.058  0.9960.118 0.2480.110  3.5190.680 0.5770.468  a. 95% CI estimated as 2(Std. Dev.)  Table 2.6: 4-MP HDO conversion and product selectivity at different reaction temperatures for Mo2C catalysts prepared at different CHR temperatures. Reaction Temperature, C 325  350  375 Catalyst X4-MP S DDO/HYD  X4-MP S DDO/HYD  X4-MP S DDO/HYD 10% Mo/AC-600 25.9a 2.55  0.68b  56.5 2.90  0.80  90.2 4.34  0.94 10% Mo/AC-650 27.5 2.58  0.72  55.3 3.39  0.66  93.2 4.89  3.02 10%Mo/AC-700 22.6 3.34  1.56  50.2 4.02  1.86  87.4 6.09  5.08 a. This value was the conversion of 4-MP after a reaction time of 420 min. b. 95% CI estimated as 2(Std. Dev.). 44  The apparent activation energies (EDDO and EHYD) and pre-exponential factors, as per the equations 𝑘𝐷𝐷𝑂 = 𝑘𝐷𝐷𝑂𝑜 𝑒−𝐸𝐷𝐷𝑂𝑅(1𝑇−1𝑇𝑜) and 𝑘𝐻𝑌𝐷 = 𝑘𝐻𝑌𝐷𝑜 𝑒−𝐸𝐻𝑌𝐷𝑅(1𝑇−1𝑇𝑜) where To is the centre temperature (350 C), were determined from Arrhenius plots of the rate constants (Table 2.7 and Figure E.3). The DDO apparent activation energy (~125 kJ/mol) is significantly greater than that of the HYD route (~89 kJ/mol) and the values are invariant over the 3 catalysts prepared at different CHR reduction temperatures, despite the change in Mo2C and MoOxCy content of the catalysts. Finally we note that the activity of the 10%Mo/AC-650 catalyst for the HDO of 4-MP is lower than that reported for MoO3 but better than MoP, MoO2, MoS2 [103], measured at the same conditions, as those shown in Table 2.8 and reported per gram of Mo.   Table 2.7: Pre-exponential factors and apparent activation energies extracted from 1st-order rate constants for the DDO and HYD of 4-MP over Mo2C catalysts prepared at different CHR temperatures. Catalyst HYD DDO koHYD Ea koDDO Ea  mL/(gMo.min) kJ/mol 10-3 mL/(gMo.min) kJ/mol 10%Mo/AC-600 0.4200.014a 89.85.4 1.3580.128 126.114.2 10%Mo/AC-650 0.3990.024 86.212.8 1.3670.150 124.526.2 10%Mo/AC-700 0.2610.032 90.922.2 1.1030.286 125.050.0 a. – 95% CI estimated as 2(Std. Dev.).  45  Table 2.8: Kinetic rate constants for the conversion of 4-MP over different Mo-based catalysts at 350 ºC. Catalysts k (mL/(min.gMo)) MoS2 unsupported [103] 0.47 MoO2 unsupported [103] 0.55 MoP unsupported [103] 1.64 MoO3 unsupported [103] 3.78 10%Mo/AC-650a 1.72 10%Mo/AC-600a 1.69 10%Mo/AC-700a 1.24 a. k was calculated by formula: k = kDDO + kHYD as reported in Table 2.5.  2.4 Discussion The catalysts prepared from AHM/AC precursors by CHR at relatively low temperature (600 ~ 800 C) are mixtures of Mo2C and MoOxCy. The appearance of Mo2C in the XRD (Figure 2.1) first occurs at a reduction temperature of 650 ºC, where a weak peak at 2θ = 46.07º occurs, indexed to the (101) plane of β-Mo2C. From the trend in diffraction peak intensities, one concludes that the β-Mo2C is poorly crystallized but highly dispersed at this temperature. As the reduction temperature increases, the diffraction peak intensities of β-Mo2C increase, a consequence of both increased β-Mo2C concentration and increased crystallinity.  The TEM results are consistent with the XRD results. The catalyst prepared at a CHR temperature of 650 ºC has a weak diffraction pattern (Figure 2.1) and the TEM micrograph of the same sample (Figure 2.3 (a)) shows that there is no clear boundary between Mo2C particles and 46  the carbon support, indicative of the low crystallinity of the sample and suggesting that some of the Mo remains in an amorphous oxycarbide phase at this temperature. Analysis of the TEM micrographs reveals that the β-Mo2C particle diameters range from 4 ~ 6 nm, indicating a high dispersion on the carbon support. Figure 2.3 (c) shows that carbon filaments are formed around the Mo2C particles (indicated by white arrows). Yoshida et al. [116] found that iron carbide converts C2H2 into CNTs at 600 ºC and CH4/H2 mixtures have been used in CNT synthesis on Fe and Fe-Co catalysts [117, 118]. Under the reduction conditions of the present study, the AC is converted to CH4 and other hydrocarbons in the presence of the added H2, and this gas likely generates carbon filaments over the Mo2C. The formed graphite layers could extend and bend along the facets of the Mo-carbide, and thus cover the Mo-carbide particle leading to lower activity. At a reduction temperature of 700 ºC, the average particle size of the lamellate-like structure is 10 ~ 15 nm, which is larger than that formed at 650 ºC and in good agreement with the XRD analysis showing the same trend. The TEM micrograph also shows that some slabs overlap with each other, suggestive of some agglomeration at 700 ºC. When the CHR temperature rises to 750 or 800 ºC, the agglomeration is severe. The average size for the isolated particle is 16 ~ 21 nm at 750 ºC (Figure 2.3 (e)). As shown in Figure 2.3 (f), the increased temperature results in severe agglomeration with the particle size at 800 ºC difficult to measure.  Aside from the unclear boundary shown in the TEM at 650 ºC, the XPS results shown in Figure 2.2 and Table 2.3 indicate the existence of MoOxCy. Deconvolution of the XPS narrow scan spectra finds that all three samples prepared at 600, 650 and 700 ºC have Mo2C and MoOxCy phases present. Different reduction temperatures influence the relative abundance of each Mo species. As the reduction temperature increases, the amount of Mo2C increases accompanied by 47  a decrease in MoOxCy (Table 2.4). At a reduction temperature of 700 ºC, the carbide phase accounts for 50 % of all Mo species. The composition of different Mo states listed in Table 2.3 shows clearly the transition from MoOxCy to Mo2C. The Mo 3d spectra of Figure 2.2 (a), (b) and (c) show a shift from high B.E. to low B.E. At high temperature, carbon diffusion into the Mo is more likely, while the carbon may also modify the electronic properties of the surface Mo species to increase the reaction between the H2 and carbon, leading to the higher mass burn-off observed with increased temperature. The formula in Table 2.4 indicates that more C is introduced into the MoOxCy structure as the reduction temperature increases. In conclusion, MoO3 gradually converts through high valence states (Mo5+ and Mo6+ ) of MoOxCy to lower valence states (Mo4+ and Mo3+), finally forming the Mo2C. A similar transformation process has been observed by Choi and Thompson [107]. However, further increases in CHR temperature results in agglomeration and carbon deposition that deactivates the catalyst.  Thus there is a narrow CHR temperature range that yields the most active catalyst, even though the resulting catalyst has a surface composed of both Mo2C and MoOxCy.  The analysis of the 4-MP HDO reaction kinetics shows very similar values for both kDDO and kHYD on the 10% Mo/AC-650 and the 10% Mo/AC-600 at low temperature, indicating that both Mo2C and MoOxCy function as active sites for HDO. Reduction at 700 ºC results in Mo2C agglomeration and the formation of filaments that lead to reduced activity. However, the catalyst prepared at higher reduction temperature has a marginally higher DDO selectivity, suggesting that the Mo2C has higher DDO selectivity than the MoOxCy. A similar phenomenon has been reported by Ren et al. [70], indicating that the dominant products of biomass–derived oxygenates from Mo2C contain unsaturated products and the strong oxygen binding energy on Mo2C 48  promotes the scission of C=O/C-O bonds. The O occupied Mo2C surface (denoted as MoOxCy), also has the ability to dissociate hydrogen to form hydroxyl, followed by reaction with another hydroxyl to form water and O vacancies. The DDO route of HDO usually occurs by the absorption of the reactant through O atoms [119]. The formation process of Mo2C and MoOxCy suggests that there might be many vacancies on the catalyst that were originally occupied by O. The catalysts are extremely sensitive to oxygen indicating that the interactions between oxygen atoms and active sites should be facile. Lee et al. [69] reported that high DDO selectivity is due to the suppression of catalyst hydrogenation activity by the presence of O-containing species from the reactants changing the surface properties of Mo2C. Liu et al. [120] also indicate the possibility of changing the selectivity of transition metal carbides by O2 treatment.   Table 2.9: XPS Analysis of Mo 3d for 10% Mo/AC-650 catalyst after reaction. Catalyst B.E. (eV) of Mo 3d5/2  Composition (mol.%) Mo2+ Mo3+ Mo4+ Mo5+ Mo6+ Mo2+ Mo3+ Mo4+ Mo5+ Mo6+ 10%Mo/AC-650 228.3 229.0 231.1 231.9 233.1 23.69 18.01 3.79 31.96 22.55 After 1 h 228.4 229.2 230.6 232.1 233.2 13.12 20.08 17.80 34.47 14.53 After 5 h 228.4 229.2 230.6 232.1 233.2 12.69 20.66 16.93 35.16 14.56  The catalysts of the present study are also seen to be very sensitive to O, even though after CHR the Mo is present as both Mo2C and MoOxCy. Table 2.9 reports a marked increase in the Mo4+ and Mo5+ surface species and a reduction in Mo2+, as estimated from the XPS analysis, after 1 hour reaction compared to the fresh catalyst. After a further 5 h reaction, there is no further change in composition of the Mo species. The sensitivity is further illustrated by the CO uptake data that show a significant drop in uptake after passivation, compared to the un-passivated 49  catalysts (Table 2.4). Nonetheless, several authors have reported that CO uptake is a valid measure of the metallic-like sites of the Mo2C catalysts [66, 69] and the data of Table 2.4 show CO uptake increases with increased Mo2C content, but at a CHR of 700C, CO uptake decreases, likely because of Mo2C agglomeration identified by TEM and XRD analysis.  Figure 2.6: 1st-order pre-exponential constant for DDO and HYD over all Mo2C catalysts prepared at various CHR temperatures. (The error bar reflects the calculated 95% CI)  The apparent activation energy determined form the kinetic analysis shows that both the DDO and HYD reaction pathways are unchanged as the CHR temperature increases. Table 2.7 reports the pre-exponential values for the 1st-order rate constants of the catalysts on a per gram Mo basis. 0 20 40 60 80 1000.00.40.81.21.6  k0 DDO or k0 HYDpre-exponential constant, mL/(min.gMo)CO uptake, mol/gMoHYDDDO50  These values are plotted as a function of the CO uptake (following passivation and re-reduction) in Figure 2.6 and show a reasonable linear correlation for both the DDO and HYD reactions. The fact that HYD pre-exponential rate constant falls on the straight line indicates that the corresponding catalytic reaction is not structure sensitive, since the data were obtained on catalysts of varying Mo2C size. Similarly, the values of the DDO pre-exponential rate constant suggest a linear correlation with CO uptake, from which one concludes that the DDO route is not structure sensitive either. Note that the correlation of the DDO rate constant shows significant more deviation from the linear correlation than the data of the HYD route. However, taking account of the error associated with the rate constant estimates, one concludes that the linear correlation is statistically valid. The rate constant for the sample prepared at a CHR of 700C shows some deviation that may indicate structure sensitivity of this reaction, since the Mo2C particle size was significantly higher for this catalyst (~15 nm) compared to the others of Figure 2.6. The slopes of the lines in Figure 2.6 allow the TOF of the reactions to be calculated based on the kinetic parameters. For the 10%Mo2C-650 sample at 350C the TOF for the DDO of 4-MP is 6.410-2 s-1 and for HYD a value of 1.910-2 s-1 is obtained. These values are about 2 orders of magnitude greater than the values reported for the HDO of anisole at 150 C [69], but extrapolating to 325C yields a TOF of about 1.810-1 s-1, reflecting a more facile Ar-OCH3 bond compared to the Ar-OH bond.   The fact that the kinetics of the reactions are not strong functions of the CHR reduction temperature and hence relative content of Mo2C versus MoOxCy, suggests that the active site of the catalyst is determined by the state of the catalyst under reaction conditions, wherein O 51  adsorption and/or exchange with the catalyst likely occurs, as indicated by the XPS data of Table 2.9 and the CO uptake measurements of the fresh versus passivated catalysts. The resulting active sites can occur on both Mo2C and MoOxCy during the HDO reaction. The data of Table 2.8 also illustrate that the Mo-oxides are active for 4-MP HDO and as noted elsewhere, electrophillic co-ordinatively unsaturated sites present on MoO3 and MoO2 under reaction conditions, result in a high selectivity for C-O hydrogenolysis or DDO [103]. Previous studies have also concluded that oxygen-containing species change the surface properties of the Mo2C during HDO [121]. Lee et al. [69] reported that the presence of D2O suppresses the hydrogenation ability of Mo2C significantly and this change is irreversible suggesting that the loss in activity is not due to competitive adsorption. Liu and coworkers [120] also report the possibility of changing the selectivity of transition metal carbides by O2 treatment.   2.5 Conclusions Molybdenum carbide supported on AC has been synthesized by carbothermal hydrogen reduction. MoO3 transforms from high valence states such as Mo5+ and Mo6+ in MoOxCy, to lower valence states Mo4+ and Mo3+, and finally to Mo2C. The degree of crystallization of β-Mo2C increases and the oxygen in MoOxCy decreases with increased CHR temperature. Increasing CHR temperature also increases the C burn-off rate above 650 ºC, and at 675 ºC some graphitic structures are formed as a consequence of catalytic reactions with the formed β-Mo2C and these are detrimental for catalysis. When the CHR temperature reaches 700 ºC, agglomeration occurs, decreasing the catalyst activity further. Although the activity of the Mo2C/MoOxCy prepared at optimum temperature (650 ºC) is lower than MoO3, it is higher than MoS2, MoO2 and MoP, and the catalyst has higher DDO selectivity. Kinetic analysis shows the 52  DDO activation energy to be significantly higher than the HYD value, and the rate constants per gram Mo are well correlated to the CO uptake data, regardless of the Mo2C content of the catalyst.  53  Chapter 3: Preparation of Mo2C/AC and Ni-Mo2C/AC Catalysts and their Stability in HDS of Dibenzothiophene 2 3.1 Introduction In traditional crude oil upgrading, hydrodesulphurization (HDS) is an important oil refining process to improve the oil quality [122, 123]. Unlike bio-oil, there is not much O contained (< 1 wt%) conventional crude oils ; while the S content is usually high (> 4wt%). Thus, in this chapter, HDS has been used as the probe reaction to test the catalytic activity of Mo2C/AC and Ni-Mo2C/AC catalysts. However, one of the difficulties encountered when applying Mo2C catalysts for HDS is that they are unstable because of surface sulfidation, even in the presence of low concentrations of S (< 0.1 wt%). Aegerter et al. [124] proposed that a thin layer of highly dispersed MoS2 is formed on the surface of Mo2C or Mo2N particles during thiophene HDS, as evidenced by IR and TPD-CO measurement of the catalysts. The authors proposed that Mo2C and Mo2N particles serve as rigid substrates for a sulfided Mo phase, exposing a large number of co-ordinatively unsaturated Mo sites (CUSs), which results in increased activity of the catalyst (by 50%), compared to a conventional MoS2 catalyst. Oyama et al. [125] suggested the formation of a Mo-carbosulfide active site to explain product selectivities observed during the simultaneous conversion of sulfur, oxygen and aromatic compounds over Mo2C, compared to the product selectivities measured with a single reactant, especially for the case of cumene hydrogenation. In another study, Brito and coworkers [126] presulfided a NiMo carbide catalyst prior to HDS and claimed that a Ni-Mo-S phase was responsible for the measured HDS activity.                                                  2 A version of this chapter has been published:  H. Wang, S. Liu., R. Govindarajan, and K. J. Smith, “Preparation of Ni-Mo2C/carbon catalysts and their stability in the HDS of dibenzothiophene,” Applied Catalysis A: General (2017) 539(5): 114-127. 54  Jin et al. [127] reported on NiMo carbide and Mo2C catalyst performance for the HDS of dibenzothiophene (DBT) and pointed out that Ni addition increased the Mo2C activity by 57 %, although there was no conclusion drawn regarding the effect of S on the metal carbides during HDS. The transformation of Ni-Mo2C catalysts in the presence of S-compounds remains somewhat unclear and the application of these catalysts for HDS in high S concentrations (as is the case for gas oils derived from residue oils, for example) has not been reported.  The present chapter reports on the CHR synthesis of Ni-Mo2C catalysts supported on activated charcoal (AC) and their application to the HDS of DBT at high S concentrations (~ 3500 ppmw) that mimic the S content of heavy oil [32]. New data are reported that capture the effect of Ni content on the formation of the Ni-Mo2C catalysts and their resulting HDS catalytic activities. In particular, the changes in catalyst morphology that occur during reaction, with the formation of a core-shell Mo2C-MoS2structure, is reported for the first time. The structural changes are related to the stability of the Ni-Mo2C/AC and the Mo2C/AC catalysts. Note that the catalysts reported in this chapter were prepared on a commercial activated charcoal rather than the petcoke used in the following Chapters 4 ~ 6. At the time that this study was completed, activated petcoke samples were not yet available. The commercial activated carbon was chosen as a suitable support since it removed any potential impacts of petcoke variability on the comparison between the Mo2C and Ni-Mo2C catalysts.  55  3.2 Experimental 3.2.1 Catalyst Preparation Activated charcoal was impregnated with a solution of ammonium heptamolybdate (AHM) in acetone (10 %) and H2O (90 %) and rotary evaporated under vacuum to remove all the solution. The resulting solid was aged in air for 4 h before drying at 110 oC overnight to yield the precursor. The precursor was then calcined in N2 at 300 oC for 5 h to obtain the oxide states of Mo. The calcined precursor (ca. 1.8 g) was placed in a U-tube reactor and subsequently converted to the Mo2C/AC catalyst by CHR in a continuous H2 flow (200 mL (STP)/min), while increasing temperature from room temperature to 500 oC at 10 oC /min, followed by increasing temperature at 1 oC/min to 650 oC and holding the final temperature for 90 min before quenching to room temperature in N2 [128]. This catalyst (designated as 10%Mo2C/AC-650) was compared with Ni-Mo2C/AC catalysts, prepared similarly but with successive impregnations of the AC with solutions of AHM and Ni(NO3)2, respectively, to obtain precursors with different Ni:Mo ratios. After calcination, the precursors were converted to Ni-Mo2C/AC by CHR as described above but with final reduction temperatures of 550 or 600 oC. The resulting Ni-Mo2C catalysts with different Ni:Mo ratio (from 0.02 ~ 0.76) are designated as NixMo2C-y, where x represents the Ni:Mo atomic ratio and y represents the reduction temperature (oC) (See details in Appendix A.2). The 10% Mo2C/AC-650 and Ni-Mo2C/AC catalysts, assessed in a batch reactor to determine reaction kinetics, were loaded into the reactor without passivation. The catalysts were transferred from the preparation U-tube reactor to the batch reactor in a glove bag under a N2 blanket to ensure minimal air exposure.   56  For the catalyst stability tests conducted in a trickle-bed reactor and for the characterization of the fresh catalysts (XRD, BET), the catalyst samples were first passivated in a flow of 1 vol % O2/N2 at room temperature for 2 h prior to the analysis and a mild pre-reduction of the passivated catalysts was done prior to the reaction test. Carbon supported MoS2 (MoS2/AC) was also used herein as a reference to compare with the used Ni-Mo2C/AC catalysts. The MoS2/AC precursor was prepared similarly to the Mo2C/AC precursor followed by ex-situ presulfiding in a 100 mL decalin for 3 h at 350 oC under approximately 10 vol%H2S/H2 generated from CS2 (See details in Appendix C.3).   3.2.2 Catalyst Characterization 3.2.2.1 Elemental Analysis (1) CHNS analysis The adsorbed S on the used catalysts was determined using a Perkin-Elmer 2400 series II CHNS/O Analyzer. The combustion temperature was set at 975 oC.  (2) ICP-OES analysis Elemental analysis of the catalysts was conducted by inductively coupled plasma optical emission spectroscopy (ICP-OES). All samples were digested in aqua regia to extract the Mo from the carbon support. After metal extraction, the solution was filtered to remove undissolved C residue from the solution before evaporation to remove all aqua regia by heating the sample to 150 oC. Finally, the concentration of all samples was adjusted to the range of 1 ~ 50 ppm in 2 vol% HNO3. An internal standard method based on In was used to quantify the ICP-OES analysis. 57   3.2.2.2 N2 Adsorption and Desorption  The Brunnauer-Emmett-Teller (BET) surface area, pore volume and pore size of the calcined Ni-Mo2C precursors, passivated Ni-Mo2C and used catalysts (following HDS reaction) were determined from N2 adsorption/desorption isotherms measured at -196 oC using a Micromeritics ASAP 2020 analyzer. The used catalysts were washed with acetone several times before analysis to remove the decalin solvent used in the reaction. (See Chapter 2 and Appendix B.1 for Details of the relative theory and calculation)  3.2.2.3 X-ray Diffraction (XRD) The XRD of the fresh and used catalysts was compared to identify bulk phase changes of the catalysts during reaction. A Bruker D8 Focus (0-20, LynxEye detector) diffractometer was used for the analyses. (See Chapter 2 and Appendix B.2 for details)  3.2.2.4 X-ray Photoelectron Spectroscopy (XPS) The details of XPS analysis were described in Chapter 2. Deconvolution of the XPS profiles was done using XPSPEAK41 software. Experimental peaks were decomposed through mixed Gaussian-Lorentzian functions (80% - 20%) after Shirley background subtraction. (See Appendix B.3 for details)  3.2.2.5 Transmission Electron Microscopy (TEM) Transmission electron microscope (TEM) images of the supported catalysts were generated using a (JEOL) JEM 2200FS electron microscope operated at 200 kV, with a 1.9 Å point-to-58  point resolution. Used catalysts were analyzed to identify the particle size distribution and characteristic slabs of the synthesized/formed species. The samples were prepared by dispersion in ethanol and then sonicated for several minutes. A drop of the suspension was placed on a 230 mesh copper grid coated with Formvar-Carbon film and then dried in a vacuum chamber. The clusters/particles identified in the TEM images were measured and counted to yield a size distribution that was fitted to a lognormal distribution to determine the average size of the cluster/particle. The number of formed MoS2 layers was also counted based on the TEM images and the stacking degree (N) of MoS2 was calculated as Eq. 3-1. 𝑵 = ∑ 𝒏𝒊𝑵𝒊/ ∑ 𝒏𝒊𝒊=𝟏..𝒕𝒊=𝟏..𝒕         (Eq. 3-1) where ni is the number of stacks with Ni layers. (See Appendix C.4 for details)  3.2.2.6 Time of Flight Secondary Ion Mass Spectroscopy (TOF-SIMS) A time of flight secondary ion mass spectrometer (TOF-SIMS; a PHI TRIFT V instrument) was used in the static SIMS mode to analyse the used catalysts and to identify changes in catalyst surface properties (with or without Ni as a promoter) following the HDS reaction. TOF-SIMS yields both elemental and molecular ions, and hence it can provide detailed information on the upper layer of the used catalysts. The catalyst samples were sputtered by a primary ion beam of Au+ before the spectra were collected in negative mode. For each analysis the delivered ion dose was 6.3×1011 ion/cm2, and the total sputtering time for each spectrum was approximately 1800 s. The detected mass range was within 0 ~ 1850 amu and triplicate spectra were collected at three different points across the prepared samples with a raster size of 400.0 µm ×400.0 µm as representative sample areas. The obtained spectrum is shown in Appendix E.2, Figure E.10 for the used Ni0.19Mo2C/AC-550 catalyst. 59   3.2.3 Catalyst Activity Tests The HDS of dibenzothiophene (DBT) was carried out in a 300 mL stirred-batch reactor (Autoclave Engineers) to assess the catalytic activities of the as-prepared Ni-Mo2C/AC fresh catalysts with different Ni:Mo ratios (Ni:Mo = 0 ~ 0.76). The experiments were operated at 350 oC with an initial H2 pressure of 2.1 MPa with ca. 2.0 wt% of DBT (0.35 wt% S) as the reactant in 100 mL decalin. Dissolution of DBT in decalin was assisted by sonication, which also removed dissolved air from the prepared liquid feed (Feed preparation is reported Appendix C.1, Table C.2). The catalyst was added to obtain 0.13 ~ 0.14 gmetal/100 mL liquid in the slurry. The fresh catalyst and O-free reactant liquid were placed in the reactor sequentially, under a N2 blanket using a glove bag. The reactor was then sealed and flushed with N2 to remove residual air from the reactor. After a leak test, ultra-high purity H2 was used to first flush and then fill and pressurize the reactor to the desired level, following which the reactor was heated to a pre-determined temperature within 35 mins. After the reaction period (up to 5 h) the reactor was cooled to room temperature and de-pressurized. Liquid samples were recovered periodically during the reaction and analyzed using a Shimadzu (QP-2010-S) GC/MS and RTX5 30 m × 0.25 mm capillary column based on an internal calibration method. The internal standard used in this study was diphenylmethane (DPM) (see Appendix C.1 for details). Overall, the activity data reported herein were measured with a carbon balance ≥ 94% and several experiments were repeated to quantify the error in the activity data (See details in Appendices F.1, F.3 and F.4). Both external and internal mass transfer effects were shown to be minimal at the chosen reaction conditions and details of these calculations are presented in Appendix G.5 Table G.7.  60  A fixed bed reactor (length 50 cm, hot zone 30 cm with an internal diameter of 8.64 mm) operated in a down-flow mode was also used to compare the stability of the Ni0.19Mo2C/AC-600 and 10 % Mo2C/AC-650 catalysts. The schematic of the reactor system is shown in Figure 3.1.  61   Figure 3.1: Schematic diagram of high pressure fixed-bed reactor for HDS of DBT.62   The experiments were operated at 310 oC at a constant pressure of 4.1 MPa. The liquid feed, consisting of ca. 2.0 wt% DBT in decalin, was fed to the reactor by means of a high-pressure piston pump (Gilson model 307). The H2 flow was controlled by a mass flow controller (Brooks 5850TR). All the experiments were operated with a H2/feed volumetric ratio of 600 and a liquid-hourly space velocity (LHSV) of 8 h-1. The liquid product was collected periodically from a condenser placed after the reactor exit and held at room temperature. Finally, the liquid samples were analyzed using gas chromatography (Shimadzu GC) by a capillary column (SHRXI-5MS, 15 m x 0.25 mm x 0.25 um) and a FID detector. Before performing the activity test, the passivated catalyst (ca. 0.77 g, 2.5 mL) was activated in-situ under H2 flow at 400 oC for 2 h to remove the passivation layer from the catalyst surface.  3.3 Results 3.3.1 Fresh Catalysts Characterization XRD analysis of the calcined Ni-Mo2C/AC catalyst precursors showed an amorphous structure, indicating a high dispersion of metals on the support (Appendix E.2, Figure E.4). Figure 3.2 presents the XRD patterns of the Ni-Mo2C/AC catalysts with Ni:Mo ratios of 0 ~ 0.76, and prepared at different CHR temperatures (Figure 3.2 (a): 550 oC; Figure 3.2 (b): 600 oC). Included at the top of both Figures are the spectra for 10%Mo2C/AC-750, the unpromoted Mo2C prepared at a reduction temperature of 750 oC to show the positions of β-Mo2C (PDF card # 00-035-0787). Mo2C is detected in all the diffractograms of Figure 3.2, with diffraction peaks at 2θ = 40.12º, 44.21º, 46.14º, 61.35º, 72.87º, 82.97º, 86.58º and 89.51º and no other bulk phases of MoOx present in the samples. At a reduction temperature of 550 oC, the peak intensities of Mo2C 63  increased with increased Ni:Mo ratio in the range of 0.02 ~ 0.44 and the same phenomenon occurred at a reduction temperature of 600 oC, but for a narrower Ni:Mo ratio of 0.02 ~ 0.19. Figure 3.2 (b) shows that when Ni:Mo ratio is ˃ 0.19, the intensities of the Mo2C reflections do not increase, however other phases are detected. The peaks at 51.28 º and 60.23 ºare assigned to metallic Ni (PDF card # 00-004-0850) and the separation of Ni from the Mo2C can be observed when the Ni:Mo ratio is ˃ 0.38 at both reduction temperatures of 550 oC and 600 oC. A clear phase separation of Ni from the Mo2C was also reported by Puello-Polo et al. [31] and Santillán et al. [30] when synthesizing the catalysts using the same metal salts as those used herein. The Ni-Mo bimetallic carbide (Ni6Mo6C2, PDF card # 01-089-4883) was detected with diffraction angles 2θ = 47.76º, 50.51º and 53.40º when the Ni:Mo ratio reached 0.38 at both reduction temperatures, whereas nickel carbide was not detected in any of the samples.   64   Figure 3.2: XRD patterns of Ni-Mo2C/AC catalysts with different ratios of Ni:Mo (0 ~ 0.76) prepared at different reduction temperatures: (a) Reduced at 550 oC; (b) Reduced at 600 oC. (◆) Mo2C; (o) Ni; (*) Ni6Mo6C2.  65  Analysis of the XRD data for the NixMo2C/AC-600 catalysts (0 ≤ x ≤ 0.19) yields a small decrease in the Mo2C lattice parameters with increased Ni content (Table 3.1). The diffraction angles also shifted to higher diffraction angles for the Ni0.19Mo2C/AC-600 compared to the unmodified catalyst (10%Mo2C/AC-750), implying some incorporation of Ni into the β-Mo2C crystal structure. The presence of Ni means that some of the Ni may substitute Mo atoms in the crystal structure randomly. Thus, the crystallite strain will be changed, leading to a change in the lattice parameters. Similar phenomena have been reported elsewhere [129, 130]. Thus, the interaction between Ni and Mo2C occurs at low Ni loadings. The lattice parameters at higher Ni contents were not determined because of Ni phase separation and the crystallite sizes were not calculated from the XRD data because small particles of Mo2C are air sensitive and re-oxidize during analysis.  Table 3.1: XRD analysis of NixMo2C/AC-600 catalysts and Mo2C/AC-750 catalyst. Catalyst Phases Diffraction peaks, 2θ Lattice parameters a (Å) Mo2C a/b c 100 200 101 10%Mo2C/AC-750 β-Mo2C 40.23 44.30 46.14 3.0050 4.7633 Ni0.09MoC/AC-600 β-Mo2C  40.25 44.23 46.18 3.0032 4.7563 Ni0.19MoC/AC-600 β-Mo2C  40.30 44.41 46.26 3.0000 4.7402 a. The lattice parameters were calculated by Eq.B-4 (Appendix B.2).  The measured Ni:Mo ratios of the catalysts are reported in Table 3.2. The carbon loss (mass burn-off) from the AC support that occurred during CHR (as shown in the last two columns in Table 3.2) increased with increased Ni:Mo ratio and increased reduction temperature.66   Table 3.2: Catalyst composition of fresh Mo2C/AC-650 and Ni-Mo2C/AC catalysts. Ni:Mo a Metal content (wt%) Mass burn off b 550 ºC 600 ºC 550 ºC 600 ºC Ni Mo Ni Mo 0.02 0.12 8.60 0.13 9.41 5 14 0.09 0.49 8.49 0.53 9.26 9 17 0.19 1.01 8.52 1.22 10.35 18 33 0.38 2.06 8.90 2.43 10.52 19 34 0.44 2.90 10.77 3.89 14.43 25 47 0.76 4.78 10.26 6.26 13.44 28 48 a. These values were measured by ICP-OES. b. Mass burn-off (%) was calculated by the mass difference before and after CHR divided by the mass before CHR, error ±2%.  During CHR, the AC support reacts with the H2, mostly forming CH4, as determined by GC-FID analysis of the product gases (Figure 3.3). The maximum concentration of CH4 in the product gas from the CHR was ~2.0 mol% at 650 oC for the unpromoted Mo2C (Figure 3.3 (a)); whereas, in the presence of Ni, the CH4 concentration was significantly higher (up to ~ 6 mol%), even at 600 oC. Figure 3.3 (b) shows that the maximum in the CH4 content of the CHR product gas increased with increased Ni content. Furthermore, the CH4 formation profiles are different between the unpromoted and Ni promoted Mo2C. In the former case, the CH4 content doubles every 50 oC in the temperature interval 550 oC~650 oC and remains approximately constant during the hold period; whereas, for the Ni-Mo2C precursors, the CH4 concentration decreases significantly above the highest reduction temperature of 600 oC. The mass burn-off (carbon loss) for the Mo2C/AC-650 catalyst was ~18 wt%, similar to the Ni0.19Mo2C/AC-550 (reduced at 550 C), clearly indicating that the presence of Ni accelerates the carbon loss at low temperature, 67  which reduces the carburization time required for the Mo precursor. Also note that the mass burn-off is well correlated with CH4 formation during CHR (Figure 3.4). 68    Figure 3.3: Profile of detected CH4 (mol%) during carbothermal hydrogen reduction of the catalyst generation: (a) Mo2C/AC-650 (■); (b) Ni-Mo2C/AC with different ratios of Ni:Mo ((●) Ni0.09Mo2C/AC-600; (▲) Ni0.19Mo2C/AC-600; (▼)Ni0.76Mo2C/AC-600).   69    Figure 3.4: A correlation between mass burn-off (wt%) and formed CH4 (mol%) during CHR process. The solid line represents the correlation equation: Mass burn-off (wt%) = 7.5847 × CH4 (mol%), Std.Dev.= 0.4055. (R2=0.9918)   After CHR, the shape of the N2 adsorption isotherms were similar to the AC support for all prepared catalysts with a H3 type hysteresis loop (Appendix E.2, Figure E.5). Also, it found that the hysteresis loop spanned P/Po = 0.4~1.0, suggesting a wider distribution of the pore size. The surface area of the catalysts prepared at a CHR temperature of 550 oC did not show a significant change with increased Ni content; whereas, at 600 oC, the surface area decreased with increased Ni content (Table 3.3 and Appendix E.2-Figure E.6). Moreover, both catalysts showed a general 70  trend of increased pore size with increased Ni content and the increase was due to an increase in mesopore volume (Appendix E.2, Figure E.6).  Table 3.3: Physical properties of fresh and used Ni-Mo2C/AC catalysts prepared at reduction temperatures of 550 oC and 600 oC. Ni:Mo Surface area (m2/g)  Ave. pore size (nm)  Vmeso/Vtotal (%) Fresh 550 Used 550 Fresh 600 Used 600  Fresh 550 Used 550 Fresh 600 Used 600  Fresh 550 Used 550 Fresh 600 Used 600 0.00a 911 467 911 467  3.70 4.70 3.70 4.70  83 93 83 93 0.02 778 406 887 450  3.58 4.90 3.75 4.90  76 95 80 94 0.09 815 436 774 472  3.67 4.90 4.46 5.50  81 95 85 98 0.19 721 403 701 409  3.88 5.10 5.10 7.10  83 97 90 100 0.38 763 446 650 380  3.86 5.40 6.12 7.50  84 95 93 100 0.44 774 349 647 369  4.50 5.60 6.30 9.00  88 97 95 100 0.76 806 385 525 206  4.70 6.20 6.87 9.40  89 100 98 100 a. This is the unpromoted Mo2C/AC-650 catalyst.  3.3.2 Catalysts Characterization after HDS Reaction All the catalysts were characterized after the HDS reaction to identify potential changes in the catalyst chemical and physical properties. In addition, the MoS2/AC catalyst properties are reported here as a comparator for the used Ni-Mo2C/AC catalyst properties. The catalysts were separated by high speed centrifuge from the slurry recovered from the batch reactor. The solid fraction was then washed in acetone and dried overnight under a N2 blanket in a glove bag. Prior to XPS and TEM analysis, the dried samples were degassed in vacuum at 200 oC for 4 h. The transfer of the samples to the vacuum chamber (XPS, TEM) was done within 5 min to minimize exposure to air.  71  3.3.2.1 XRD and BET Analysis As shown in Figure 3.5, the XRD analysis of the used catalysts showed no change in the Mo2C bulk structure after reaction, indicating that the Mo2C crystallites are resistant to bulk sulfidation under the chosen reaction conditions, consistent with the observation made by Aegerter et al. [124]. There were no peaks that could be assigned to S compounds in the XRD pattern of the used catalysts either.    Figure 3.5: X-ray diffraction patterns for fresh and used Ni-Mo2C/AC catalysts.  72  The BET analysis data (Table 3.3) generally indicate that after the HDS reaction, there was a significant loss in catalyst surface area, whereas the average pore size and mesopore volume fraction increased, implying that mostly micropores were blocked during reaction.   3.3.2.2 XPS and CHNS Analysis The effect of increased Ni:Mo ratio on the XPS spectra of the used catalysts is shown in Table 3.4 and Figure 3.6. The elemental surface compositions clearly show the presence of S on the catalyst surface after the HDS of DBT, consistent with data reported by Sajkowski and Oyama [3]. As illustrated in Table 3.4, the used Ni-Mo2C catalysts reduced at 550 oC have higher S content than the same catalysts reduced at 600 oC. In addition, the Ni-Mo2C catalysts all have higher S content after reaction than the Mo2C catalyst with no Ni addition.  Table 3.4: XPS and CHNS analysis of used (Ni)-Mo2C/AC catalysts with different Ni:Mo ratios and different reduction temperatures. Used catalysta S, At%b Mo, At%b O, At%b C, At%b S/Mo atomic ratio, % Sc, wt% Sd, wt% 10%Mo2C/AC-650 0.43 0.82 4.43 94.32 0.53 1.07 2.00 Ni0.19Mo2C/AC-550 0.90 0.64 3.68 94.78 1.41 2.23 3.24 Ni0.44Mo2C/AC-550 1.10 0.52 5.06 93.32 2.12 2.75 3.84 Ni0.19Mo2C/AC-600 0.67 0.68 4.99 93.67 0.99 1.65 2.22 Ni0.44Mo2C/AC-600 0.97 0.76 4.40 93.87 1.28 2.39 2.85 a. All of these used catalysts listed were recovered following 5 h HDS of DBT in the batch reactor; b. These values were measured by XPS and being normalized; c. It indicated the amount of surface adsorbed S wt% based on XPS measurement; d. It indicated the amount of bulk phase S wt% based on CHNS analyzer.  73  Apart from a clear increase in S content on the catalyst surface with increased Ni addition, the Mo3d XPS spectra of Figure 3.6 (a) and (c) show an obvious shift from lower B.E. to higher B.E. as Ni content increases, indicative of the formation of Mo-S surface species. The Mo3d shift to higher B.E. on the Ni-Mo2C is greater than that of the Mo2C/AC, indicating more sulfidation of the Mo in the presence of the Ni. The presence of Ni likely increased the S adsorption on metal sites, such that the S content of the used catalysts increased with increased Ni addition.    Figure 3.6: XPS narrow scan spectra of used MoS2/AC, Mo2C/AC and NixMo2C/AC-y catalysts: (a, c) Mo 3d; (b, d) S 2p. (The dashed lines in (a, c) indicate the position of Mo2+ species from Mo2C; the dashed lines in (b, d) indicate the position of S2- species from MoS2.) 74   Figure 3.7 compares the Mo 3d and S 2p XPS spectra of the Ni0.19Mo2C/AC-550 catalyst after reaction for different periods to that of the MoS2/AC catalyst. The S 2p peak at B.E. = 162.2 eV confirms the presence of Mo-S surface species, consistent with the results from the Mo 3d spectra. In addition, it is noted that the Mo 3d B.E. of the Ni-Mo2C catalyst surface is relatively stable after 1 h of reaction and clearly different from that of the MoS2/AC surface as shown in Figure 3.7.    Figure 3.7: XPS narrow scan spectra of used Ni0.19Mo2C/AC-550 catalyst after HDS reaction for different reaction periods. (a) Mo 3d; (b) S 2p.  Table 3.5 also shows that the S species accumulate during reaction since the S content (i.e. the atomic ratio of S to Mo or S wt%) increased with increased reaction time for all catalysts. The S content (wt%) measured by CHNS and reported in the last column of Table 3.5, does not show a clear difference between the two catalysts. However, the S wt% calculated based on the XPS 75  data give clear differences between these two catalysts, showing a strong ability of S adsorption in the presence of Ni. For the 10%Mo2C/AC-650 catalyst, the S:Mo ratio is 0.30 after 1 h of HDS, whereas for the Ni-Mo2C it is 0.69, indicating a faster sulfidation of the Ni-Mo2C catalyst at the beginning.   76   Table 3.5: XPS and CHNS analysis of used catalysts (Mo2C/AC-650 and Ni0.19Mo2C/AC-600) after HDS reaction in the batch reactor for different reaction periods. Used catalyst  Reaction time (h) S, At%a Mo, At%a O, At%a C, At% S/Mo atomic ratio, % Sb, wt% Sc, wt% 10%Mo2C/AC-650 1 0.20 0.68 3.73 95.39 0.30 0.51 1.33 3 0.30 0.70 4.08 94.92 0.43 0.75 1.47 5 0.40 0.69 4.44 94.47 0.59 1.01 2.00 Ni0.19Mo2C/AC-600 1 0.47 0.68 4.01 94.83 0.69 1.18 1.40 3 0.58 0.68 3.41 95.33 0.87 1.46 1.83 5 0.67 0.68 4.99 93.67 0.99 1.65 2.22 a. These values reported here were measured by XPS and being normalized; b. It indicated the amount of surface absorbed S wt% based on XPS measurement; c. It indicated the amount of bulk phase S wt% based on CHNS analysis.   77  3.3.2.3 TOF-SIMS Analysis The negative ion mass spectrum of Mo, sulfur or oxysulfide (oxide) spectra were recorded during sputtering from a catalyst sample area of 400.0 ×400.0 µm. Several interesting Mo characteristic fragments containing O, S or both were detected in the used catalysts, such as MoS- or MoO2 (m/z=128), MoOS- or MoO3 (m/z=144), MoO2S- or MoS2 (m/z=160), MoO3S- (m/z=176) and MoO2S2 (m/z=192). The presence of sulfur indicates that the surface of the catalyst has been changed during the HDS reaction with the formation of Mo-O-S or Mo-S bonds on the catalyst surface. In order to eliminate possible variations between samples, the normalized ion intensity yield (i.e. relative intensity, RI) is used to make comparisons. The distribution of identified Mo oxysulfide species (MoOxSy) is represented by the abundance of each Mo oxysulfide species divided by the abundance of all MoOxSy ions (Prime relative intensity, RI’). The identified molecular fragments and their relative abundance are reported in Table 3.6. These data confirm the adsorption of S on the catalyst surface during reaction and indicate that in the presence of Ni, the amount of MoS surface species is almost 2x’s higher than that on the unpromoted Mo2C (Ni:Mo=0), consistent with the XPS analysis results.   78   Table 3.6: Normalized intensity of selected ions (containing the most abundant isotope 32S) calculated based on TOF-SIMS spectra of used catalysts of 10%Mo2C/AC-650 and Ni0.19Mo2C/AC-600. Calculated parameters Used catalyst 10%Mo2C/AC-650  Ni0.19Mo2C/AC-550  RI (S) b (x 10-2) 1.91 ±0.12 a 2.27±0.10 RI (MoOxSy) c (x 10-2) 0.13±0.01 0.25±0.01 RI (MoOxSy)/RI (S) (%) 6.81±0.68 11.01±0.66 Distribution of Identified MoOxSy species (RI') d (%) RI'(MoS) e 13.09±0.87 6.98±0.16 RI'(MoOS) f  52.30±0.46 56.07±0.43 RI'(MoO2S or MoS2) f   24.28±0.45 26.46±0.39 RI'(MoO3S or MoOS2) f 6.60±0.40 4.45±0.05 RI'(MoO2S2 or MoS3) f  3.72±0.28 6.05±0.17 a. This value was estimated as standard deviation; b. The relative intensity (RI) of total identified S to total ions: RI(S)= ƩS-/ƩI+. c. The relative intensity (RI) of total identified MoOxSy- to total ions: RI=ƩMoOxSy-/Ʃ I+. d. This distribution was based on all the identified MoOxSy- species. e. Prime relative intensity (RI’) of MoS ions to total MoOxSy ions: RI’= MoS/ƩMoOxSy-. f. All of these were similar to e’s calculation by changing numerator.   3.3.2.4 TEM Analysis The TEM micrographs of the used Ni-Mo2C/AC catalysts reduced at 600 oC with various Ni:Mo ratios are presented in Figures 3.8 and 3.9. The formation of core-shell like structures is clearly visible, suggesting that a Mo2C core is surrounded by MoS2 generated during the HDS reaction. The size of the core-shell structure increases from 9.8 nm to 20.9 nm as Ni content increases, as shown by the lognormal distribution data of Figure 3.8. The Mo2C-MoS2 clusters become agglomerated when the Ni:Mo ratio ˃ 0.44. At low Ni content, such as Ni0.02Mo2C/AC-600, the 79  contrast between particle and support is low and it is difficult to quantify the particle size from the TEM micrograph. For the 10%Mo2C/AC-650, the characteristic slabs of MoS2 are difficult to identify. Note also that the number of MoS2 layers formed around the Mo2C core increases with increased Ni content, as reported in Figure 3.9.   80   Figure 3.8: TEM images and cluster size distribution of used Ni-Mo2C/AC catalysts with different Ni contents. (a) Ni0.02Mo2C/AC-600; (b) Ni0.09Mo2C/AC-600; (c) Ni0.19Mo2C/AC-600; (d) Ni0.38Mo2C/AC-600; (e) Ni0.44Mo2C/AC-600; and (f) Ni0.76Mo2C/AC-600. 81   82      Figure 3.9: Distribution of MoS2 layer numbers for used catalysts. (a) Ni0.09Mo2C/AC-600; (b) Ni0.19Mo2C/AC-600; (c) Ni0.38Mo2C/AC-600; (d) Ni0.44Mo2C/AC-600; (e) Ni0.76Mo2C/AC-600; (f) Enlarged Ni0.19Mo2C/AC-600; and (g) Enlarged Ni0.44Mo2C/AC-600.  3.3.3 Catalyst Performance in HDS of DBT Dibenzothiophene (DBT) was chosen as the model reactant to determine the activity of the prepared catalysts. The results of the activity tests at 350 oC for Ni-Mo2C/AC catalysts, prepared at different reduction temperatures and Ni:Mo ratios are presented in terms of DBT conversion in Table 3.7.   83  Table 3.7: DBT conversion for (Ni)-Mo2C/AC catalysts at 350 oC and an initial pressure of 2.1 MPa. Catalysts Catalyst weight (g) Burn off% [gcata./ gprecursor] Metal weight in RXN (g) Conversiona (%) Catalysts Catalyst weight (g) Burn off% [gcata./ gprecursor] Metal weight in RXN (g) Conversiona (%) Mo2C/AC-650 1.4029 16.95 0.14 33.67 Mo2C/AC-650 1.4029 16.95 0.14 33.67 Ni0.02Mo2C/AC-550 1.6485 5.17 0.14 41.69 Ni0.02Mo2C/AC-600 1.4920 14.09 0.14 36.79 Ni0.09Mo2C/AC-550 1.5026 8.79 0.13 44.58 Ni0.09Mo2C/AC-600 1.4112 17.13 0.14 35.84 Ni0.19Mo2C/AC-550 1.3144 17.68 0.13 44.34 Ni0.19Mo2C/AC-600 1.1252 33.83 0.13 48.07 Ni0.38Mo2C/AC-550 1.2602 19.10 0.14 50.22 Ni0.38Mo2C/AC-600 1.0326 33.15 0.13 35.52 Ni0.44Mo2C/AC-550 1.2798 25.89 0.18 45.08 Ni0.44Mo2C/AC-600 0.6644 47.63 0.12 31.17 Ni0.76Mo2C/AC-550 1.2081 28.01 0.18 39.58 Ni0.76Mo2C/AC-600 0.6262 48.03 0.12 31.14 a. This conversion was obtained based on HDS reaction of DBT at 350 oC following 5 h.    84  The mass of catalyst used for each activity measurement varied to account for the AC mass loss during the various CHR conditions; however, the total Mo+Ni mass remained relatively constant (~0.14 gmetal/100 mLfeed) for all experiments. The data of Table 3.7 show that the DBT conversion on Ni-Mo2C/AC reduced at 550 oC were generally higher than the DBT conversion measured on the unpromoted Mo2C/AC catalyst and higher than the DBT conversion measured on the Ni-Mo2C/AC reduced at 600 oC, provided the Ni:Mo ratio is < 0.44. The data clearly show improved activity of the Mo2C in the presence of Ni. The conversion over the Ni-Mo2C/AC-550 catalyst increased with increased Ni content for Ni:Mo ratios of 0 ~ 0.38, but decreased when the ratio increased > 0.38. The decreased conversion is likely due to the Ni phase separation that occurs at high Ni content and higher carbon loss (burn-off%) above a Ni:Mo ratio of 0.38 that may also decrease Mo dispersion. Similar observations with respect to increased HDS activity of P-doped Ni-Mo2C/Al2O3 catalysts with increased Ni content, were reported by Sundaramurthy et al. [131], using gas oils derived from Athabasca bitumen.  Kinetic analysis of the batch reactor data is complicated by the fact that the Mo2C-based catalysts undergo sulfidation during reaction and consequently, their activity and selectivity may change with reaction time in the batch reactor. As shown in Table 3.5, the S content of the catalysts (with or without Ni addition) increases with reaction time. However, the difference in the amount of sulfur adsorbed on the 10%Mo2C/AC-650 and Ni0.19Mo2C/AC-600 is constant at ~0.27 At% independent of reaction time (Table 3.5, column 3). This indicates that Ni has the ability to increase S adsorption but the effect of the Ni occurs rapidly within the 1st hour of reaction. The continued adsorption of S beyond this period is a consequence of the carbon 85  support. Furthermore, Figure 3.7 shows that there is no visible difference with respect to the Mo oxidation state of the catalyst after the 1st hour of the reaction.   To confirm the stabilization of the catalysts, Figure 3.10 reports the DBT conversion as a function of time-on-stream (TOS) for the Mo2C and the Ni0.19Mo2C/AC-600 catalysts measured at 310 oC and 4.1 MPa H2 pressure and a LHSV of 8 h-1 in the downflow fixed-bed reactor. A rapid decrease in DBT conversion occurs within the first 60 mins of operation beyond which both the DBT conversion and product selectivities stabilize. Continued operation beyond 5000 min showed that the activity of the Ni0.19Mo2C/AC-600 catalyst remained stable; whereas the DBT conversion over the Mo2C/AC showed a small decline. Hence, to minimize these catalyst stabilization effects on the kinetic analysis of the batch reactor data, only data collected after 1 h operation are included in the analysis. Hence, the kinetic analysis is performed on the partly sulfided, but stabilized Mo2C-based catalysts.  86   Figure 3.10: DBT conversion and product selectivity versus time on stream for Ni0.19Mo2C/AC-600 and Mo2C/AC-650 catalyst, measured in the down flow fixed-bed reactor is at 310 oC, 4.1 MPa, H2/feed volumetric ratio = 600 and LHSV = 8 h-1.  The kinetics of the HDS of DBT on all the synthesized catalysts were determined assuming 1st-order reactions and a reaction pathway [132] illustrated in Figure 3.11.   87    Figure 3.11: Simplified kinetic steps HDS of DBTshowing 1st order reaction paths over all Ni-Mo2C/AC catalysts with different ratios of Ni:Mo and different reduction temperatures.   Direct desulfurization of DBT (the DDS pathway) to form biphenyl (BPh) competes with a hydrogenation pathway (the HYD pathway) in which DBT is converted to1,2,3,4-tetrahydrodibenzothiophene (THDBT) before S-removal to yield cyclohexylbenzene (CHB). The kinetic parameters of this simplified reaction scheme were estimated using a Levenberg-Marquardt nonlinear regression methodology to solve the relevant ODEs (Eq. 3-2 to 3-5) as below.  dCDBTdt= −k1CDBT − k2CDBT        (Eq. 3-2) dCTHDBTdt= k1CDBT − k3CTHDBT       (Eq. 3-3) dCBPhdt= k2CDBT         (Eq. 3-4) dCCHBdt= k3CTHDBT         (Eq. 3-5)  88  The 1st-order estimated kinetic rate constants kj, j=1-3, are reported in Table 3.8 and the model fit to the data is shown in Appendix E.2, Figure E.7 and Figure E.8, for the CHR reduction temperatures of 550 oC and 600 oC, respectively, based only on the data obtained after a reaction time > 1 h. The magnitude of the rate constant for the total consumption of DBT (ktotal=k1+k2) (Table 3.8) indicates that the Ni0.38Mo2C/AC-550 and Ni0.19Mo2C/AC-600 have the highest activity among all the catalysts. Figure 3.12 shows that the rate constants k1 and k2 increase with increased Ni addition up to a maximum value and then decrease at high Ni contents. The thermal reactions are neglected since the thermal related conversions are very low (less than 5%). 89   Table 3.8: Kinetic model parameters estimated for the HDS of DBT after the 1st hour of the reaction in the batch reactor over Ni-Mo2C/AC catalysts at 350 oC and initial PH2=2.1 MPa. NixMo2C/AC-550 Kinetic Parameter (mL /gmetal.min) NixMo2C/AC-600 Kinetic Parameter (mL /gmetal.min) k1 k2 k3b ktotal (k1+k2) SDDS/HYD (k2/k1) k1 k2 k3 ktotal (k1+k2) SDDS/HYD (k2/k1) 0.02  0.12±0.07a 0.62±0.06 2.21±1.48 0.74±0.13 5.00±2.87 0.02  0.12±0.04 0.54±0.04 1.99±1.32 0.67±0.08 4.40±1.59 0.09  0.16±0.05 0.65±0.05 1.78±1.05 0.81±0.10 4.20±1.48 0.09  0.13±0.04 0.48±0.04 1.89±1.11 0.61±0.08 3.55±1.12 0.19  0.22±0.03 0.79±0.03 4.75±1.13 1.01±0.05 3.58±0.47 0.19  0.28±0.12 0.86±0.11 7.77±7.83 1.14±0.22 3.06±1.33 0.38  0.30±0.12 0.81±0.11 4.84±3.79 1.11±0.23 2.71±1.18 0.38  0.17±0.04 0.55±0.04 4.95±2.81 0.72±0.08 3.22±0.88 0.44  0.24±0.11 0.55±0.10 4.12±3.86 0.79±0.21 2.25±1.10 0.44  0.20±0.10 0.40±0.09 6.25±6.76 0.60±0.19 2.06±1.19 0.76  0.22±0.08 0.41±0.07 3.41±2.80 0.63±0.15 1.89±0.78 0.76  0.18±0.11 0.41±0.09 4.71±6.01 0.59±0.20 2.36±1.54 a. Estimated standard deviation. b. k3 is determined by the concentration variation of THDBT. When the conversion of THDBT is too fast, the concentration becomes very low and it is difficult to estimate k3, resulting in a large error in the estimated value of k3, such that k3 is not statistically different from zero.   90   Figure 3.12: Kinetic parameters k1, k2 vs. Ni:Mo ratio of Ni-Mo2C/AC catalysts: (a) reduced at 550 oC; (b) reduced at 600 oC. (…Trend line)   91  The catalyst selectivity to the three main products of BPh, THDBT and CHB is illustrated in Figure 3.13. Biphenyl was the main product resulting from direct hydrogenolysis of the C-S bond of DBT, accounting for ˃ 80% of all products of the Mo2C/AC-650 catalyst, and even higher for the Ni promoted catalysts, as shown from Figures 3.13 (a) and (d). These data imply that the DDS reaction route is dominant and is the preferred route for S removal on these Mo2C-based catalysts. However, the data of Table 3.8, comparing the relative rates of DDS to HYD and reflected in the value of SDDS/HYD =k2/k1, suggests an increase in HYD selectivity with increased Ni/Mo ratio. The hydrogenated product CHB is mainly formed from the hydrogenation of THDBT and the data show that the presence of Ni increases CHB selectivity (Figure 3.13 (b), (c), (d) and (e)). 92   Figure 3.13: The selectivity to biphenyl (BPh), tetrahydro-dibenzothiophene (THDBT) and cyclohydrobenzene (CHB) vs. DBT conversion: (a-c) reduced at 550 oC; (d-e) reduced at 600 oC. 93  3.4 Discussion  The addition of Ni is shown to decrease the CHR temperature required for Mo2C formation from the AHM/AC calcined precursor and this observation is in agreement with that reported previously by Liang et al. [26]. In Chapter 2’s study [128] the optimized reduction temperature for Mo2C formation from AHM supported on AC was reported to be 650 oC; whereas, in the present study, Mo2C formation at significantly lower CHR temperature (550 oC) with Ni addition is demonstrated. Furthermore, the amount of Ni is shown to be critical in terms of the phases generated by CHR and in terms of the catalyst DBT activity. At a reduction temperature of 550 oC, the formation of Mo2C was confirmed by XRD when the Ni:Mo ratio was low (0.09). Since Ni activates H2 at relatively low temperature [133, 134], hydrogenation of the carbon to yield CH4 can occur at low temperature. In addition, by activating H2 at relatively low temperature, Mo oxide is more readily reduced which makes the subsequent conversion to Mo carbide or oxycarbide more facile. As shown in Figure 3.3, the CH4 concentration generated in the product gas from CHR of the Ni0.19Mo2C/AC-600 catalyst is ~ 4x’s higher than that of the 10%Mo2C/AC-650 catalyst when the reduction temperature reaches 600 oC. In general, the concentration of CH4 generated during CHR increases as the reduction temperature increases and as Ni loading increases. The XRD data show that as the amount of Ni increases, the intensity of the Mo2C diffraction peaks increases. Figure 3.3 shows that the CH4 generation rate increases rapidly above about 600 C and Ni can lower this temperature. Hence, although temperature is the most important factor determining the CH4 concentration measured during CHR, the Ni:Mo ratio also plays an important role in the reduction. Hence it is concluded that although the presence of Ni decreases the required reduction temperature for carbide formation, temperature remains the key factor in the CHR synthesis of metal carbides. However, higher temperature 94  results in a higher degree of crystallization and higher carbon loss (burn-off) from the carbon support. The burn-off was 48% for the Ni0.76Mo2C/AC-600 whereas at 550 oC, the burn-off decreased significantly to only 28%. If different Ni:Mo ratios are compared, the difference in mass loss between Ni:Mo ratios of 0.02 and 0.19 is much larger than the mass loss between 0.19 and 0.76, suggesting that the effectiveness of Ni in increasing the CH4 concentration is reduced at high Ni concentrations. When the Ni:Mo ratio reaches 0.38 at both 550 and 600 oC, Ni6Mo6C2 species appear in the XRD, indicating that Ni has been incorporated into the Mo2C structure during the CHR process. When the Ni:Mo ratio is > 0.38, metallic Ni appears. Unlike Mo2C, the formation of Ni6Mo6C2 and Ni are mostly dependent on Ni content rather than CHR temperature. Thus, the formation of the bimetallic carbide and Ni phase separation appears at the same Ni:Mo ratio despite differences in CHR temperature. Table 3.3 and Figure E.9 show that when the catalyst precursors were reduced at 550 oC, the surface areas were approximately constant at ~ 800 m2/g for different Ni:Mo ratios; whereas, at a CHR temperature of 600 oC, there was a clear decrease in surface area with increased Ni content. The differences are due to the fact that the pore widening process that occurs during CHR, removes some pore walls between mircopores causing a loss in surface area at 600 oC; whereas, at 550 oC widening occurs, but since the pores are still developing, the pore walls and the resulting surface area remain intact. In addition, the catalyst pore size increased in the presence of Ni at both CHR temperatures, with most of the micropores of the AC support enlarged into mesopores, which is beneficial for molecular diffusion of large molecules such as DBT.  XRD analysis of the fresh and used catalysts showed that the bulk phase properties of the catalysts were unchanged after reaction with DBT. However, CHNS and XPS analysis both 95  showed that S was present on the surface of the used catalysts. Note that the measured d-spacing from the TEM is ~0.63 nm (Figure 3.9), indicating the formation of MoS2 species on the used Ni-Mo2C/AC catalysts [135]. A similar observation was reported by Zhang et al. [136] on a used Ni2Mo3N catalyst after hydrogenation in the presence of 100 ppm of thiophene. As shown in Table 3.4 and Figure E.9, the amount of S increased with increased Ni content of the catalyst, suggesting that the Ni enhanced S adsorption on the metal sites. In addition, the TEM results confirmed this conclusion showing an increase in the number of MoS2 layers with increased Ni content of the catalyst (Figure 3.9). Also note that the catalysts reduced at 550 oC had higher S content after use than those reduced at 600 oC. These results are consistent with the data obtained by CHNS analysis in which S content of the used catalysts is well correlated with the Ni:Mo ratio. The difference in S content of the catalyst as measured by CHNS (bulk analysis) versus XPS (surface analysis; reported in the last two columns of Table 3.4) can be accounted for by the S adsorbed inside the pores of the carbon support, the carbon “sink effect” proposed by Laine et al. [137]. There is a ~ 1.0 wt% difference between the bulk and surface S content of the Ni-Mo2C catalysts reduced at 550 oC; whereas, the difference is ~ 0.5 wt% for the Ni-Mo2C catalysts reduced at 600 oC. This difference is caused by the pore structure modification during CHR at different reduction temperatures. Since many micropores are destroyed at higher reduction temperature (600 oC), the S adsorption capacity is decreased compared to the catalyst prepared at the lower CHR temperature.   XRD showed that Ni incorporates with Mo species during the CHR process. The calculated lattice parameters of β-Mo2C decrease with increased Ni content before phase separation at high Ni content, which implies that smaller Ni atoms randomly substitute for Mo atoms in the metal 96  carbide crystal structure. A similar conclusion was made by Wan and Leonard [130] in regards to Fe modified Mo2C. XPS shows that the addition of Ni increased the S adsorption on to the catalyst surface after reaction with DBT. TEM analysis of the used Ni-Mo2C/AC-600 catalysts (Figure 3.8) clearly showed that the Mo2C particle size increased with Ni content. The catalyst metal carbide particle size is well correlated with Ni:Mo ratio as shown in Figure 3.14.    Figure 3.14: The correlation of Ni:Mo ratio to MoS2 stacking degree (N) (a) and the average particle size (b). ―is the fitted line.   Comparing the 10%Mo2C/AC-650 and Ni0.09Mo2C/AC-600 catalysts with the same burn off shows an increase in particle size from 5 nm for the 10%Mo2C/AC-650 catalyst to 10 nm for the Ni0.09Mo2C/AC-600 catalyst. Similar results were reported by Yao et al. [138] in their study of a NiMoC catalyst and Leonard and Wan [130] noted that unlike Fe, which decreases the size of Mo2C particles, Ni will increase the size of Mo2C. Figure 3.9 also shows that addition of Ni increases the number of MoS2 layers in the MoS2 shell that surrounds the Mo2C of the used 97  catalyst. Ni increases the stack number of MoS2 as has also been reported on NiMoS/Al2O3 catalysts by Lercher et al. [134]. The STEM mode selected area analysis of the used catalyst shown in Figure 3.15, confirms that the Ni, Mo and S are closely bound in the used catalyst, suggesting that Ni can enhance the S interaction with Mo.      98      Figure 3.15: High angle annular dark field scanning (HAADF-STEM) image and energy-dispersive X-ray (EDX) elemental mapping of used Ni0.19Mo2C/AC-600 catalyst for elements Mo, Ni, S, O, C, and Si.  The study reported in this chapter has shown that Ni addition has a significant impact on both the CHR process and the S adsorption by the catalyst during the HDS reaction, and that both depend on the Ni content of the catalyst. During reaction, Ni enhances the adsorption of S by the Mo to form Ni-Mo-S on the outside of the Mo2C particle, leading to the generation of a core-shell like 99  structure. None of the characterization data, including the TOF-SIMS analysis, identified the presence of Mo carbo-sulfide structures, as proposed by Oyama et al. [125], and although they were not present as bulk species, we cannot exclude the possibility that they exist on the catalyst surface under reaction conditions.  The data of Table 3.8 and Figure E.7 show that Ni added to the Mo2C catalyst increases DBT conversion. A trend notable from Table 3.8 and Figure 3.12 is that k2, representing catalytic ability to perform DDS, increases to a maximum value with increased Ni content. When the Ni concentration is relatively low, the added Ni is incorporated with Mo species into the final metal carbide phase. Thus, in the HDS of DBT, increased Ni increases the MoS2 stacking number and according to the ‘rim-edge’ model of Chianelli and Daage [139], a higher stacking number yields higher DDS selectivity. However, when the Ni concentration is above a critical value (Ni:Mo > 0.38), the added Ni is not incorporated into the Ni-Mo carbide phase; on the contrary, more Ni causes phase separation and enlarges the particle size leading to less active sites and eventually lower activity. Also, available surface area is decreased by increased burn-off with increased Ni content, resulting in decreased dispersion of active sites and reduced activity. Moreover, more reduced single phase Ni contributes to the hydrogenation ability of the catalyst and thus, a decrease in SDDS/HYD is observed. Note that the trend of SDDS/HYD versus Ni content at 600 oC is not very clear, due to the loss of surface area along with Ni addition affecting the catalytic performance. For the above reasons, it is found that the best performing catalysts are Ni0.38Mo2C-550 and Ni0.19Mo2C-600. They have a total kinetic constant for DBT conversion of 1.11 and 1.14 mL/gmetal.min along with a very high SDDS/HYD ratio of 2.7 and 3.1, respectively.  100  3.5 Conclusions Ni addition significantly reduces the temperature required for Mo2C generation by CHR, with the presence of Mo2C confirmed by XRD at a CHR temperature of 550 oC for a catalyst with a Ni:Mo ratio of 0.09. This CHR temperature is about 100 oC lower than that required to form Mo2C in the absence of Ni. XRD and XPS show that Ni is incorporated into the Mo2C structure during the CHR process and TEM showed that increased Ni content increased the Mo2C particle size. However, for Ni:Mo > 0.38 phase separation occurred, decreasing the catalyst activity and the direct desulfurization selectivity. The Ni also increased the rate of S adsorption during reaction, leading to the formation of Mo2C-MoS2 core-shell structures. Increasing Ni content also increased the stack height of the MoS2 shells, leading to a higher DDS selectivity in the HDS of DBT. The best performing catalysts in the HDS of DBT at 350 C are Ni0.38Mo2C/AC-550, with a total kinetic constant of 1.11 mL/gmetal.min and SDDS/HYD of 2.7, and Ni0.19Mo2C/AC-600 with a total kinetic constant of 1.14 mL /gmetal.min and SDDS/HYD around 3.1.  101  Chapter 4: Synthesis of Mesoporous Mo2C/Carbon Catalysts by Carbothermal Hydrogen Reduction using Petroleum Coke 3 4.1 Introduction After the successful preparation of Mo2C from activated charcoal (AC) as reported in Chapters 2 and 3, the study was extended to activated petcoke (APC) as reported in this chapter. A thermochemical activation method was developed, starting with petcoke and using KOH at a temperature of 800 oC to generate pores (> 75% micropores) and a APC surface area of 2028 cm2/g. After the synthesis of the APC, a metal catalyzed carbon hydrogenation was used to further increase the pore size of the APC [77-79, 140-142]. The metal catalyzed method used in the present study is carbothermal hydrogen reduction (CHR), where APC was impregnated with Mo salts, to synthesize Mo2C catalysts supported on APC at relatively low temperature (< 800 oC) [20, 128]. During CHR the carbon of the support and H2 are used to convert the Mo-precursor (usually MoOx) to Mo2C and although several studies of CHR have been reported [21, 143], the sequence of reaction steps that lead to Mo2C remains unclear. Direct reaction between MoOx and carbon occurs at temperatures higher than 900 oC and is therefore only a minor contributor to Mo2C formation during CHR [19, 21]. CH4 from the gas phase, generated by hydrogenation of the carbon support (H2 + C  CH4), may react with the Mo-precursor to produce Mo2C, although the reaction requires a significant CH4 gas phase activity to occur to any significant extent [21, 144]. Furthermore, since dissolution of carbon in metal oxides is orders of                                                  3 A version of this chapter has been published:  H. Wang, S. Liu., B. Liu, V. Montes, J. M. Hill, and K. J. Smith, “Carbon and Mo transformation during the synthesis of mesoporous Mo2C/carbon catalysts by carbothermal hydrogen reduction,” Journal of Solid State Chemistry (2017) 258: 818-824. 102  magnitude less than in metals, diffusion of carbon through the metal oxide and subsequent hydrogenation at the gas/metal oxide interface is unlikely [145].  Consequently, in the present chapter, the conversion processes that occur during CHR were examined to clarify the sequence of reaction steps that lead to Mo2C. The study reports for the first time that simultaneous hydrogenation of the carbon support during CHR, catalysed by the Mo2C, yields mesopores (2~50 nm) within the carbon support. Both the effect of the CHR final temperature and the holding time at this temperature, have been examined. The transition of the AHM precursor to Mo2C and the corresponding changes to the APC morphology that occur during CHR, determined from both characterization data and computational methods, are reported. The Mo species are shown to catalyse the hydrogenation of the carbon support, similar to metals, modifying the pore structure such that Mo2C supported on activated charcoal, with a high mesopore volume, is obtained. Finally, the catalytic activity of the Mo2C/APC catalysts is assessed using 4-methylphenol (4-MP) as a model reactant for the hydrodeoxygenation (HDO) reaction to compare with Mo2C/AC catalysts reported in Chapter 2. The difference between activated petroleum coke (APC) and activated charcoal (AC) supported Mo2C catalyst was explored.   4.2 Experimental and Computational Methods 4.2.1 Catalyst Preparation Petcoke, generated by delayed coking, was used as the carbon source (See Appendix A.1 for details of petcoke properties). Thermo-chemical activation of the petcoke was done at 800 oC for 2 h under N2 flow with a KOH:petcoke mass ratio of 3:1. The petcoke was then wet impregnated 103  with an AHM solution (9:1 mass ratio of water to acetone with sufficient AHM to yield the 10 wt% Mo/APC precursor; AHM was purchased from Sigma-Aldrich, ACS reagent, 81.0-83.0% MoO3 basis) and dried at 120 oC to yield the AHM/APC precursor (See Appendix A.2 and A.3 for details of catalyst preparation; Appendix F.2 for repeatability tests of petcoke activation). CHR of the AHM/APC precursor was done in a quartz U-tube reactor with 0.9 g of sample and a 100 mL (STP)/min H2 flow as the temperature increased at a ramp rate of 1 oC/min from 25 to 500, 550, 600, 650, 700, 750 and 800 oC, respectively (Appendix E.2, Figure A.4 for the illustrion of U-tube reactor). The samples are designated as MoTTT_APC where TTT is the final temperature of the CHR. In three cases, the sample was held at the final CHR temperature (600, 650, and 700 oC) for a further 90 min in the H2 flow and these samples are designated as MoTTT_APC-90.  4.2.2 Catalyst Characterization Physical and chemical properties of the AHM/APC precursor and the CHR produced Mo2C/APC catalysts were determined by N2 adsorption/desorption isotherms (measured using a Micromeritics ASAP 2020 analyzer), X-ray photoelectron spectroscopy (Leybold Max200 XPS), and transmission electron microscopy (Tecnai Osiris TEM). Note that the catalyst samples used for XPS and TEM analysis were transferred from the U-tube reactor into sealed glass vials within a N2 protected glove bag. The samples were subsequently transferred from the glass vial to the XPS or TEM unit as quickly as possible, limiting exposure to the atmosphere to less than 2 min. The evoluted gas from the U-tube reactor during CHR was simultaneously analyzed by an online gas chromatograph using a flame-ionization detector (Shimadzu GC-14B) (See Appendix B.1, B.3 and B.5 for details). 104   The pore size distribution was determined from the Non-Local Density Functional Theory (NLDFT) analysis of the N2 isotherm data [146] and the Mo2C particle size distribution was determined by counting > 100 particles from the TEM micrographs and applying a lognormal distribution to the particle count data. A relatively high loading of 10 wt% Mo was chosen so as to obtain XPS data with high signal intensities and clear information on the transition process of Mo species during the CHR process.  4.2.3 Catalyst Activity Tests The catalyst activity tests follow those described in Chapter 2 [128]. The reaction was conducted at 350 oC and 4.3 MPa with a stirring speed of 1000 rpm in a batch reactor. The same concentration of 4-methylphenol (3.1 wt%) was used here to make it comparable with Mo2C/AC catalysts. The obtained Mo2C/APC catalyst was added to the reactor with minimal air exposure using a glove bag under Ar flow and the catalyst mass was adjusted to ensure a constant Mo/4-MP mass ratio of 0.021 gMo/g4‑MP (calculation details see Appendix C.5, Table C.6). First-order kinetic analysis of the 4-MP conversion versus time data was used to extract the apparent 1st-order rate constant of the HDO reactions and these rate constants were used to compare catalyst activities (See Appendix C.6 and Appendix D.1 for details).   4.2.4 Computational Method Density functional theory (DFT) calculations were performed to provide insight into the activation of H2 during the CHR process as the AHM was converted to Mo2C. The calculations 105  were completed using the Vienna ab initio simulation package (VASP) [147]. The projector-augmented wave method was used to represent core-valence interactions [148, 149]. Valence electrons were described by a plane wave basis with an energy cutoff of 400 eV. The generalized gradient approximation with the Perdew-Burke-Ernzerhof functional was used to model electronic exchange and correlation [150]. The Brillouin zone was sampled at the Gamma point. Optimized structures were obtained by minimizing the forces on each ion until they were less than 0.05 eV/Å. Gaussian smearing with a width of 0.05 eV was used to improve convergence of states near the Fermi level. The adsorption energy was defined as [151]: Eads = E(adsorbate+surface) - E(adsorbate) - E(surface) where E(adsorbate+surface) is the total energy of the adsorbate interacting with the surface; E(adsorbate) and E(surface) are the energies of the free adsorbate in gas phase and the bare surface, respectively. A negative value corresponds to exothermic adsorption, with a more negative value corresponding to stronger adsorption.  4.3 Results and Discussion 4.3.1 Characterization Results The previous study [128] of Mo2C synthesis on AC supports by CHR as described in Chapter 2, reported that during the temperature ramp, H2 reacts with the C support generating CH4, while at the same time converting the Mo precursor to Mo2C. Figure 4.1 shows the CH4 concentration measured at the reactor exit during CHR of the AHM/APC precursor, to a maximum temperature of 800 oC. From 25 oC to 350 oC there was no measurable quantity of CH4 produced; whereas, from 350 to 550 C, the CH4 concentration slowly increased. From 550 to 700 oC the CH4 106  concentration increased exponentially before declining again above 700 oC. CHR of the APC alone (no Mo addition) showed minimal reaction between the APC and the H2 with < 0.5 % CH4 in the produced gas at 800 oC. Hence, the presence of Mo species is critical for the generation of most of the CH4 at these relatively low CHR temperatures.   Figure 4.1: The CH4 concentration (mol%) in the U-tube reactor exit gas measured during CHR of Mo800_APC (□) and APC support (Δ). Inset: Particle size of generated Mo2C at different CHR temperatures (◆).  107  Table 4.1 reports the distribution of Mo species, identified by XPS analysis, present after CHR to final temperatures between 400 and 800 oC. A peak deconvolution of Mo 3d is shown in Appendix E.3, Figure E.11 and E.12. The valence state of Mo decreased from Mo6+ to Mo2+ as the final CHR temperature increased.   108   Table 4.1: XPS analysis of Mo (3d) for Mo/APC samples prepared at different CHR temperatures.  Activated Samples Composition (%) Mo2+  Mo3+ ~ 4+  Mo5+ ~ 6+ Mo400_APC 0.0 a 29.9 b 70.1 c Mo500_APC 0.0  47.3  52.7  Mo550_APC  1.7  51.9  46.5  Mo600_APC  10.7  46.5  42.8  Mo650_APC  14.3  36.9  48.8  Mo700_APC  22.3  29.4  48.3  Mo750_APC  38.0  26.4  35.6  Mo800_APC 49.2 20.9  29.9  Holding 90 min at final temperature Mo600_APC-90 13.9 30.8 55.3 Mo650_APC-90 15.0  36.9  48.1  Mo700_APC-90 40.9 24.8 34.3 a. The deconvoluted peak position of Mo2+ was 228.3 eV. b. The deconvoluted peak positions of Mo3+ and Mo4+ were 229.1 and 230.3 eV, respectively. c. The deconvoluted peak positions of Mo5+ and Mo6+ were 231.9 and 233.1 eV, respectively.  Furthermore, Figures 4.2, and 4.3, and the inset of Figure 4.1 show that the Mo2C particle size, as measured by TEM analysis, increased with increased CHR temperature, especially above 650 oC. From 650 to 700 oC, the average particle size increased significantly from 5.8 to 7.9 nm and above 700 oC the particle size increased by ~1.5 nm per 50 oC. The physical properties of the Mo2C/APC also varied with the CHR temperature as shown in Table 4.2, Figure E.13 and Figure E.14. Between 550 and 800 oC, the mesopore volume increased to 0.34 cm3/g at 750 oC and then decreased to 0.20 cm3/g at 800 oC. The data also show that CHR of the APC alone (no Mo) at 109  800 oC, resulted in minimal change in both mesopore volume and the catalyst surface area (APC w/o Mo but CHR to 800 oC ― Table 4.2).    Figure 4.2: TEM graphs of different samples: (a) Mo600_APC; (b) Mo700_APC; and (c) Mo800_APC.  110   Figure 4.3: Particle size distribution from TEM micrographs and fitted lognormal distribution: (a) Mo600_APC; (b) Mo650_APC; (c) Mo700_APC; (d) Mo750_APC; and (e) Mo800_APC.  111  Table 4.2: Physical properties of Mo2C/APC catalysts produced at different CHR temperatures. Sample Surface area (m2/g) Pore Volume (cm3/g) Mesopore Volume (cm3/g) Micropore Volume (cm3/g) Mesopore Volume (%) Yield (%) Mo loading (%) Starting Materials APC 2028 a 0.93 0.12 b 0.81 b 12.90 66.14 d ― AHM/APC 1520 0.69 0.08 0.61 11.59 ― 8.23 e Activated Samples APC_w/o Mo c 1957 0.91 0.13 0.78 14.29 95.18 8.65 f Mo400_APC 1832 0.80 0.09 0.71 11.25 86.67 9.50 Mo500_APC  1794 0.78 0.09 0.69 11.54 84.52 9.74 Mo550_APC  1763 0.77 0.08 0.69 10.39 85.10 9.67 Mo600_APC  1751 0.78 0.09 0.69 11.54 83.22 9.89 Mo650_APC  1709 0.76 0.09 0.67 11.84 82.35 9.99 Mo700_APC  1613 0.82 0.19 0.63 23.17 69.82 11.79 Mo750_APC  1563 0.91 0.34 0.57 37.36 62.75 13.12 Mo800_APC  1717 0.86 0.20 0.66 23.26 54.99 14.97 Mo600_APC-90 1911 1.02 0.10 0.92 9.94 79.90 10.30 Mo650_APC-90 1759 1.02 0.17 0.84 17.11 64.58 12.74 Mo700_APC-90 1651 1.16 0.42 0.74 36.21 46.33 17.76 a. The specific surface area was calculated from the measured N2 adsorption isotherm using 2D-NLDFT applied in the P/Po range of 0.01~0.30. 112  b. The pore volume was reported based on NLDFT method provided by Micromeritics, where mesopore volume is between 2-50 nm and micropore is ≤ 2 nm. c. It represted the CHR treated APC at 800 oC without Mo loading.  d. Yield was calculated by the final mass of the catalyst relative to the initial mass of the precursor. e. This value was measured by ICP-MS. f. The Mo loading of activated samples calculated as (8.23%)/[1-(1-cata. yield after CHR)].  113  The effect of holding the final reaction temperature (600, 650 and 700 oC) for a further 90 min on the properties of the Mo2C/APC catalyst are summarized in Table 4.1 and 4.2 as well. The XPS data of Table 4.1 and Figure E.12 of Appendix E.3 show that conversion of the Mo-precursor to Mo2C increased with holding time. Comparing the corresponding samples prepared with no holding time shows that the holding time increased the Mo2+ content of the catalysts. In addition, the 90 min holding time increased the extent of carbon hydrogenation, resulting in a lower catalyst yield, and higher total and mesopore volumes (Table 4.2).   4.3.2 Mo Species Transformation  The transformation of Mo species and mesopore development can be grouped into four steps based on different temperature ranges referenced in Figure 4.1:  (I) CHR temperature ≤ 350 oC Since no detectable CH4 was produced during CHR between 25 and 350 oC, the main transformation that occurred in this temperature range was the decomposition of AHM to MoO3. Normally the decomposition of AHM to MoO3 occurs at temperatures in the range of 310 ~ 350 oC [152]; although, Thomazeau et al. [153] report that the carbon support can increase the AHM decomposition temperature, 350 oC was sufficient to ensure conversion of the AHM to MoO3.  (II) CHR temperature: 350 ~ 550 oC Between 350 and 550 oC, the conversion of MoO3 to MoOxCy was initiated (Table 4.1) and the production of CH4 was apparent as shown in Figure 4.1. A commonly acknowledged theory of the reduction of MoO3 by hydrogen is that the reduction starts with a hydrogen molecule directly 114  attacking O atoms [154, 155]. After the initial reaction, O vacancy sites are generated as water is produced. Chen et al. [156] used DFT analysis to show that vacancy sites on MoO3 enhance the dissociative adsorption of H2 on the surface. Similar calculations have been reported for NiO by Rodriguez et al. [157], who also concluded that such vacancy sites assist the dissociative adsorption of H2. Furthermore, in the study by Chen et al. [156], the dissociative adsorption of methyl groups was shown to benefit from vacancy sites. H species can migrate to the carbon support and react, producing CHx (and CH4). The CHx species may also migrate and interact with other vacancy sites, hence replacing oxygen with carbon, yielding the MoOxCy. Alternatively, the produced CH4 may adsorb on the active sites of the MoOx, decomposing to hydrogen; and an adsorbed, highly reactive carbon that reduces the MoOx to yield the MoOxCy [144]. The O is removed as H2O, CO or CO2, confirmed in the present study from the mass spectrometer analysis of the exit gas from the CHR. However, the generated amounts of these species were too low to be accurately quantified. Consequently, only the main product CH4 was reported here. Although it is well known that hydrocarbons can facilitate the carburization reaction on Mo oxides [14-18, 66], the relatively low temperature and CH4 concentration implies a low carbon activity at these conditions, limiting the reaction [144]. The XPS data confirm the gradual conversion of MoO3 to MoOxCy with only 1.7 % Mo2C on the surface of the catalyst after CHR to 550 C (Table 4.1). Furthermore, the properties of the Mo2C/APC do not change significantly from 350 ~ 550 oC (Table 4.2), indicative of the low reactivity of the carbon at these temperatures.  DFT calculations were performed to elucidate the influence of O vacancies and C on the adsorption of H in the transformation of MoO3 to MoOxCy. The adsorption of H on an O vacancy 115  (-1.21 eV) was stronger than that on a stoichiometric surface of MoO3 (-1.10 eV), indicating that O vacancy sites promoted the adsorption of hydrogen (Appendix E.3: Figure E.15 and Table E.3). MoOxCy also promoted the adsorption of hydrogen with a high binding energy of -2.46 eV, indicating that the interaction between H and C can create more and more vacancy sites through oxygen removal and water formation, consistent with the XPS results shown in Table 4.1. The oxidation state of Mo gradually changed towards lower valency, indicating that more and more vacancy sites were being generated and O was being removed.  (III) CHR temperature: 550 ~ 700 oC As the CHR temperature increased from 550 to 700 oC, the CH4 concentration increased rapidly as the transition of MoOxCy to Mo2C occurred. The CH4 production observed during CHR is a consequence of two processes occurring simultaneously such that the net rate of CH4 production depends on the rate at which the carbon support is hydrogenated minus the rate at which the CH4 is consumed by the carburization of the Mo. However, the portion of CH4 consumed by the Mo precursor was small (< 1%) compared to the total CH4 produced, so that the measured CH4 production rate was taken as a measure of the carbon hydrogenation rate. Since, in the absence of Mo there was minimal CH4 produced during CHR, the rate of carbon hydrogenation is assumed to be determined by the reaction temperature, the concentration of H activated by the Mo and the carbon in contact with the Mo particle or in close proximity to the particle, such that the H is able to interact with the C at the metal/carbon interface or spillover onto the carbon prior to reaction.  The observed rate of carbon hydrogenation during CHR, calculated from the exit CH4 concentration and exit gas flowrate is well described by the Arrhenius equation in the 116  temperature range 550 to 700 oC (Figure 4.4; see Appendix C.7 for calculation details). The apparent activation energy for hydrogenation of the APC was determined to be approximately 120 kJ/mol, compared to a value of 105 kJ/mol reported for a Ni impregnated carbon [92]. Note, however, that as the temperature increased from 550 to 700 oC, the concentration of Mo2C increased and the number and type of sites available for hydrogen activation also increased because of the transition of MoOxCy to Mo2C and the fact that Mo2C has strong hydrogen activation ability [158]. The fact that the carbon hydrogenation rate follows an Arrhenius dependency, despite these changes in the catalyst, implies that carbon hydrogenation is the slow step of the reaction, whereas hydrogen activation and spillover to the carbon surface are much faster.   117   Figure 4.4: Arrhenius plot of temperature dependence in the range of 550~700 oC. 118   The maximum carbon hydrogenation rate on the Mo2C/APC catalyst was estimated at 650 ºC to be approximately 1013 molecules/(cm2.s), based on a Mo2C particle diameter of 5.8 nm. Comparable values have been reported for Ni hydrogenation of graphite at 800 ºC (of the order 1014 molecules/(cm2.s)) [91].  (IV) CHR temperature: > 700 oC The CH4 generation rate decreased when the CHR temperature was above 700 oC, even though more Mo2C was generated at higher temperatures as shown in Table 4.1. The decline in activity can be explained by two phenomena. Firstly, in the catalytic hydrogenation of the APC, highly reactive amorphous carbon will react first so that above 700 C, the MoOxCy and Mo2C particles are now more likely in contact with less reactive, highly crystalline graphite [92]. Secondly, Figure 4.3 shows that the MoOxCy and Mo2C particles grow significantly larger above 700 oC and hence, the total interfacial area between these particles and the carbon support will decrease, resulting in a reduced rate of carbon hydrogenation. The transition process that occurred with increased temperature can be illustrated as in Figure 4.5.  119    Figure 4.5: Schematic representation of Mo species transformation during CHR process. (Sizes of Mo species particles are not drawn to scale.)  4.3.3 Pore Development during Carbothermal Hydrogen Reduction (CHR) The data of Table 4.2 show that the Mo2C and MoOxCy had no significant impact on the physical properties of the materials synthesized by CHR at temperatures below 650 oC. The total pore volume remained approximately constant at 0.80 cm3/g and the mesopore volume was 0.09 cm3/g. Above 650 oC, both the total pore volume and the mesopore volume increased significantly, consistent with the rapid increase in carbon hydrogenation as reflected in the rapid increase in the CH4 concentration in the gas exiting the reactor. The pore development is further illustrated in the pore size distribution plots (Figure 4.6); the number of pores of size between 1.5 and 20 nm increased. With 10 wt% Mo loading at 750 oC, the mesopore volume reached a maximum of 0.34 cm3/g. Above 750 oC, pore volume decreased, mostly likely a consequence of excessive C removal from the APC that resulted in pore wall collapse. In the absence of the Mo2C, the data of Table 4.2 show that almost no pore development occurred despite CHR at 800 120  oC. Hence, the hydrogenation of the carbon support, catalyzed by Mo2C, showed more pore development potential given suitable reaction conditions, which in this case required temperatures from 650 to 750 oC.     Figure 4.6: Pore size distribution of Mo2C/APC catalysts prepared at different CHR temperatures calculated from NLDFT model. 121   Much like the growth in catalyst particles that is observed when metal oxides are used in the metal-catalyzed steam activation of carbons [142], growth of Mo2C particles during CHR was also observed in this study. The average particle size of the Mo2C increased from 5.8 ± 0.3 nm after CHR at 650 C to 9.4 ± 1.9 nm at 750 C, as determined from the TEM analysis (Figure 4.2). The average Mo2C particle size was larger than the average pore size of the APC. However, both the pore size distribution and the Mo2C size distribution data show that small Mo2C particles could reside within the pores of the APC. As the CHR temperature increased, the Mo2C particle size increased (Figure 4.3). Hence it is assumed that the pore growth was due to smaller Mo2C particles within the pores of the APC catalyzing hydrogenation of the pore walls, resulting in increased pore width. Furthermore, growth of the particles as the CHR temperature increased would also result in increased pore size.  4.3.4 Activity Test in HDO of 4-methylphenol The catalysts of Table 4.1 were assessed for the HDO of 4-MP and the results are reported in Table 4.3 and Figure E.16. The catalysts supported on APC prepared by CHR (Mo2C/APC) with a 90 min holding time were more active than Mo2C/AC catalysts as reported in Chapter 2 [103, 128]. In general, the bulk kinetic rate constant of Mo2C/APC catalyst is > 3x’s higher than that of Mo2C/AC due to the high surface area of APC and different C sources. A relatively high surface area of APC (>1500 m2/g) facilitates the metal dispersion of Mo2C. Also, due to the high thermal resistance of the APC, the generated pore structure does not collapse easily. Furthermore, increased activity (increase in bulk kinetic rate constant k) was associated with increased mesopore volume (Tables 4.1 and 4.2). 122   Table 4.3: Bulk kinetic rate constants for the conversion of 4-MP over different Mo-based catalysts at 350 oC.  Catalysts k  (mL/(min.gMo)) MoP unsupported [103] MoO3 unsupported [103] 10Mo%/AC-650 [128] Mo600_APC-90 1.64 3.78 1.72 5.62  6.19  7.30  Mo650_APC-90 Mo700_APC-90   4.4 Conclusion The transition of Mo species during CHR was reported as a function of temperature up to 800 oC. The generation of CH4 from the APC was critical for the transformation of MoO3  MoOxCy  Mo2C and the rate of CH4 generation increased on Mo2C. CHR of the Mo-loaded APC had a significant impact on the physical properties of the APC. Mesoporous Mo2C/APC resulted when CHR was done at temperatures to 750 oC, above which significant Mo2C sintering and APC pore collapse occurred. A 90 min holding time at the final reduction temperature increased the total pore volume and the mesopore volume fraction of the catalyst. The activity of the Mo2C/APC catalysts for the HDO of 4-MP was > 3x’s higher than that of Mo2C/AC as reported in Chapter 2, and the activity increased with increased mesoporosity of the carbon support.   123  Chapter 5: The Effect of S on Mo2C/APC Catalysts with Various Particle Sizes 4 5.1 Introduction S present in many oil feedstocks has a detrimental effect on noble metal catalysts [98]. Similarly, S has been shown to modify Mo2C catalysts during reaction [10, 124, 125, 159-161]. The study of Chapter 3 reported the deactivation of Ni-Mo2C/AC (AC– activated charcoal) catalyst during HDS [160]. However, the transformation of the Ni-Mo2C in the presence of S occurs rapidly and consequently, the relationship between the changing composition and morphology of the catalyst surface and the catalyst activity remain unclear. Also, the introduction of Ni makes the study of the nature of Mo2C catalyst during the transformation much more complex. Therefore, this chapter focused on studying the effects of S on Mo2C/APC catalysts by performing the reactions in a packed-bed reactor so that the dynamics of the transformation could be quantified. The impact of S on Mo2C catalysts was first reported by Lee and Boudart [10] for the hydrodesulphurization (HDS) of thiophene. Aegerter et al. [124] proposed a model of Mo2C/Al2O3 catalyst after HDS, where a thin layer of MoS2 covered Mo2C species, as evidenced by CO-IR spectroscopy. Oyama et al. [125] subsequently reported the formation of carbosulfide sites on a Mo2C surface during HDS, that promoted HDS but not hydrogenation reactions. CO-DRIFT spectroscopy was also used to study the sulfidation of a Ni-Mo2C surface [159]. The similar position of the CO adsorption band on a H2S/H2 treated Mo-carbide catalyst compared to                                                  4 A version of this chapter is in preparation to be submitted for publication. H. Wang, S. Liu., and K. J. Smith, “Understanding selectivity changes during hydrodesulfurization of dibenzothiophene on Mo2C/carbon catalysts”. 124  a MoS2/Al2O3 catalyst, suggested the partial sulfidation of Mo carbide under an H2S environment.   In this chapter, the activity and stability of Mo2C supported on activated petroleum coke (Mo2C/APC) during the HDS of dibenzothiophene (DBT) is reported. In particular, the dynamic transition of the Mo2C catalyst during the initial stages of HDS is examined. Although previous studies have reported that sulfidation of Mo2C occurs during HDS, none have examined the dynamic transition at conditions that allow one to capture the changes in activity, and selectivity with time-on-steam (TOS). Experiments with relatively low sulfur concentration in the feed were done to capture the transition process. Since the dynamic of the sulfidation of the Mo2C catalyst during HDS will depend on the Mo2C particle size, the Mo2C particle size was varied by varying the Mo2C catalyst loading. The effect of S on the Mo2C catalyst surface has been determined experimentally using several characterization methods (BET, XRD, XPS, HRTEM) applied to the catalysts before and after reaction. The data allow one to provide a more accurate description of the formation of MoS2 covering the Mo2C. Furthermore, density functional theory (DFT) calculations have been used to determine the impact of S on the binding energy of the reactant to the Mo2C catalyst surface. These data have been used to better understand the correlation between Mo2C catalyst structure and catalyst activity/stability as the Mo2C is exposed to S.   5.2 Experimental 5.2.1 Preparation of Catalysts Both APC and the catalyst precursors were prepared as described in Chapter 4. Subsequently catalysts with Mo loadings of 2%, 5% and 10 wt% were synthesized by CHR to a final reduction 125  temperature of 700 oC (See Appendix A.2 and A.3 for details). The resulting catalysts with different Mo loadings are designated as xxMo2C/APC, where xx represents the nominal Mo loading (wt%). The synthesized catalysts were passivated in a flow of 1 vol%O2/N2 at room temperature for 2 h. The Mo loading of the prepared catalysts was confirmed by ICP-OES and all calculations were based on the measured Mo loading. In addition, a reference MoS2 catalyst, also supported on APC (MoS2/APC), was synthesized with a nominal Mo loading of 5 wt% to compare with the used Mo2C/APC catalyst. The MoS2/APC was prepared by wetness impregnation of AHM, and dried at 110 oC overnight. An in-situ presulfiding was conducted at 370 oC for 3 h under H2 flow of 50 mL (STP)/min and 6.24 wt% CS2 in decalin at a liquid flow rate of 0.83 mL/min and 3 MPa (with further details provided in Appendix C.3). Afterwards, a solvent wash was applied to remove excess H2S adsorbed on the catalyst.  5.2.2 Catalyst Characterization The specific surface area and pore volume (Vmicro, Vmeso, and VTotal) of the catalysts were measured using a Micromeritics ASAP 2020 analyzer (See details in Chapter 2 and Appendix B.1). The 2D-NLDFT methodology was applied in the P/Po range of 0.01~0.30 to obtain the specific surface area. The pore volume was reported using the same model with the definition of Vmeso between 2 ~ 50 nm and Vmicro ≤ 2 nm and the total pore volume (VTotal) calculated as the sum of Vmeso and Vmicro.  The catalyst properties of both fresh and used catalysts were measured by X-ray diffraction (XRD) and XPS (X-ray photoelectron spectroscopy) (See details in Chapter 2, Appendix B.2 and B.3). The number of exposed Mo atoms was estimated from the CO uptake amount by assuming 126  one chemisorbed CO molecule titrates one active site (See details in Appendix B.4). The elemental quantification of Mo was done by inductively coupled plasma optical emission spectroscopy (ICP-OES) as described in Chapter 2.  The particle morphology was characterized by HRTEM/STEM/EDX (FEI Tecnai Osiris) under 200 kV with a resolution limit of 0.14 nm. The samples were ground, dispersed in ethanol and sonicated for 2 ~ 3 min to obtain a suspension. A drop of the suspension was placed on a 300 mesh lacey carbon film with Cu grid for analysis. The particle size was calculated by analyzing > 100 particles/clusters and then fitting the measured size to a lognormal distribution to determine the average particle size. High angle annular dark field scanning (HAADF) and energy-dispersive x-ray (EDX) elemental mapping were used to reveal the morphology and elemental composition of the used Mo2C.   5.2.3 Catalytic Performance Measurement in HDS The HDS of dibenzothiophene (DBT) was carried out in a packed bed reactor (length 500 mm, hot zone 300 mm and internal diameter of 8.64 mm) to assess the catalytic activities of the as-prepared Mo2C/APC catalysts with different Mo loadings. Both the mass transfer and heat transfer effect were eliminated (see details of this calculation in Appendix G.1 ~ G.4). The experiments were operated at 350 oC and constant pressure of 4.1 MPa. The liquid feed, consisting of ca. 0.2 wt% DBT (340 ppmw S) in decalin, was fed to the reactor by means of a high-pressure piston pump (Gilson model 307). The H2 flow was controlled by a mass flow controller (Brooks 5850TR). All the experiments were operated with a H2/feed volumetric ratio of 600 and a liquid-hourly space velocity (LHSV) of 4 h-1. An Aspen calculation demonstrated 127  that the reaction occurred in the gas phase at the chosen reaction conditions (see Appendix C.8). Following reactor heat up and stabilization period of 45 mins (the system dynamic response was measured as 41 mins; Appendix C.9 for calculation), liquid product was collected periodically from a condenser held at room temperature and placed after the rector exit. Finally, all liquid samples were analyzed by gas chromatography - mass spectroscopy (Shimadzu GC/MS) using a capillary column (RTX5 30 m x 0.25 mm) and an external calibration method. Before performing the activity test, the passivated catalyst (containing 0.019 gMo, 2.5 mL) was activated in-situ under H2 flow at 400 oC for 2 h to remove the passivation layer from the catalyst surface. Overall, the activity data reported herein were measured with a carbon balance ≥ 93% and several experiments were repeated to quantify the error in the activity data (Appendix F.3).  The reactant conversion and relevant product selectivities are calculated as follows:   DBT conversion (%) = 100 ×∑ 𝑚𝑖 𝑚𝐷𝐵𝑇𝑜       (Eq. 5-1) HDS conversion (%) = 100 ×∑ 𝑚𝑖−∑ 𝑚𝑆𝑖 𝑚𝐷𝐵𝑇𝑜       (Eq. 5-2) Selectivity of product 𝑖 (%) = 100 ×m𝑖 ∑ m𝑖      (Eq. 5-3) where 𝑚𝑖 is the molar faction of product i; ∑ 𝑚𝑖 is the sum of molar fraction of all products; 𝑚𝐷𝐵𝑇𝑜  is the molar fraction of DBT in the liquid feed; 𝑚𝑆𝑖 is the molar fraction of product i that contains S.  5.2.4 Computational Model and Methods Since there is no uniform model established for β-Mo2C, the eclipsed configuration has often been applied in the building of bulk β-Mo2C [162, 163]. In this part of the study, the eclipsed 128  configuration used in Wang et al.’s work [164] to generate bulk hexagonal Mo2C was used. The calculated lattice parameters of the optimized bulk are a=5.994 Å, b=5.994 Å, and c=4.727 Å, consistent with the Kuo et al.’s XRD results for β-Mo2C [165]. Since the surface termination (101) plane is the most stable, it has been modeled by periodic slabs with p(22) unit cells and used as the adsorption surface for DBT. The thickness of the four-layer slab is 5.0 Å with the top two layers relaxed and the bottom two layers fixed. A conjugated-gradient algorithm is used for ion relaxation. The vacuum layer between slabs is 15 Å to avoid interaction between the periodic slabs.   All density functional theory (DFT) related calculations were done using the plane-wave based periodic method in the Vienna ab initio simulation package (VASP) [147, 149]. The projector-augmented wave (PAW) method was used to describe the core-valance interaction while the generalized gradient approximation with Perdew-Burke-Ernzerhof (GGA-PBE) was used for the electron exchange correlation energy. All the calculations were done with cut-off energy of 400 eV with a total convergence of above 10-4. A 5  5  5 Monkhorst-Pack k-point grid for sampling the Brillouin zone was applied to bulk optimization. A 3  3  1 Monkhorst-Pack k-point grid set was used for supercell optimization. In these calculations, Van der Waals’ force is considered with dDsC dispersion correction method to correct the conventional DFT calculations.  The adsorbed H2 and DBT on the topmost layer were relaxed to their optimized positions. The DBT adsorption energy Eads was calculated by the following equation: Eads= E(X/slab) – EX – E(slab)         (Eq. 5-4) 129  Where E(X/slab) is the energy of the adsorbed system with both slab and adsored X, EX is the energy of specie X, and E(slab) is the energy of the slab.   EIS, ETS, and EFS are used to represent the energies of the initial state, transition state and final state, respectively, in the study of the dissociation adsorption energy of H2. The energy barrier (Ea) and reaction energy (Er) were calculated by the equations as follows: Ea= ETS – EIS          (Eq. 5-5) Er= EFS – EIS          (Eq. 5-6)  The Gibbs free adsorption energy ΔGAds is calculated by the following equation [166]: 𝚫𝐆𝑨𝒅𝒔(𝐓, 𝐏) = 𝐄𝐚𝐝𝐬 − 𝟑𝐤𝐓 + 𝚫𝐙𝐏𝐄 − 𝐤𝐓 𝐥𝐧 (𝟏𝐙𝐃𝐁𝐓𝐫𝐨𝐭 (𝐓)∗𝐙𝐃𝐁𝐓𝐭𝐫𝐚𝐧𝐬(𝐓,𝐏)) − 𝐤𝐓𝐥𝐧(𝐏𝐏𝟎)   (Eq. 5-7) Where 𝑧𝐷𝐵𝑇𝑡𝑟𝑎𝑛𝑠(𝑇, 𝑃) represents the translational partition function of DBT; 𝑧𝐷𝐵𝑇𝑟𝑜𝑡 (𝑇) represents the rotational partition function of DBT; T is temperature in Klevin; P is the partial pressure of DBT in the system, Pa; Po is the standard state pressure, Pa; ΔZPE is the zero point energy difference between adsorbed system and the adsorbate; k is the Boltzmann constant, 1.38 x 10-23 J/K.    130  5.3 Results 5.3.1 Fresh Catalyst Characterization 5.3.1.1 XRD and Physical Properties Analysis  Figure 5.1: XRD diffraction patterns of fresh and used Mo2C/APC catalysts: (◆) carbon support; (*) Mo2C.  Figure 5.1 shows the XRD patterns of the Mo2C/APC catalysts with different Mo loadings. No peaks were detected for the freshly-synthesized 2Mo2C/APC, probably due to the low metal 131  loading and small particle size. However, for the 5Mo2C/APC and 10Mo2C/APC catalysts, there is a small peak located at 2θ=46.07 o, corresponding to the (101) plan of β-Mo2C (PDF card #: 00-035-0787) and indicative of Mo2C formation. There are also two distinct broad diffraction peaks at ~14.8 o and ~30.2 o, corresponding to the (001) and (002) diffraction planes of the carbon support, respectively (PDF card #: 00-041-1487).   Table 5.1: Physical properties of catalyst precursors, fresh, and used Mo2C/APC catalysts with different Mo loadings. Samples Pore volume (cm3/g) Surface area (m2/g) Vtotal  (cm3/g) Mesopore (%) Burn-off% Final yield (%) VMicro VMeso Catalyst Precursors APC support 0.93a 0.09a 2028b 1.02c 9.05 34.50d 65.50e 2AHM/APC 0.90 0.05 1994 0.95 5.49 ― ― 5AHM/APC 0.84 0.08 1832 0.93 9.05 ― ― 10AHM/APC 0.69 0.07 1520 0.76 9.74 ― ― Fresh Catalysts 2Mo2C/APC 0.84 0.34 1829 1.17 28.67 22.87 46.26f 5Mo2C/APC 0.80 0.32 1722 1.12 28.91 32.37 37.66 10Mo2C/APC 0.74 0.42 1651 1.16 36.21 53.67 30.35 Used Catalysts 2Mo2C/APC 0.86 0.31 1818 1.17 26.27 ― ― 5Mo2C/APC 0.64 0.29 1362 0.93 30.75 ― ― 10Mo2C/APC 0.60 0.19 1269 0.79 23.88 ― ― 5MoS2/APC 0.73 0.11 1566 0.79 7.82 ― ― a. The pore volume was reported based on NLDFT method provided by Micromeritics, where mesopore volume is between 2-50 nm and micropore is ≤ 2 nm. b. The specific surface area was calculated from the measured N2 adsorption isotherm using 2D-NLDFT-HS applied in the P/Po range of 0.01~0.30. c. The total pore volume is the sum of Vmeso and Vmicro. d. The carbon burn-off (%) during the CHR was calculated as: Burn-off (%) = 100 x [1- 𝑚𝐶𝐻𝑅𝑚𝑜]. e. Final yield of activated petcoke (APC) was calculated by 100 minus burn-off. f. The final yield of the catalyst was calculated based on the utilization of raw petcoke: Final yield% =(100 − 𝐵𝑢𝑟𝑛 𝑜𝑓𝑓) × 65.50100  132  Table 5.1 shows that the total pore volume (VTotal) and surface area of the catalyst precursors decreased with increased Mo loading. The micropore volume (Vmicro) of the precursors decreased with Mo loading, indicating that some of the AHM blocked the micropores following impregnation and drying of the APC support. Following CHR, the catalysts all exhibited a significant increase in mesopore volume [167] with the 10Mo2C/APC having the highest mesopore volume. Note, however, that a higher carbon burn-off was obtained during the synthesis with increased Mo loading so that the 10Mo2C/APC had the lowest mass yield from the synthesis. The fresh catalysts all have similar surface areas (1734 ± 73 m2/g) and total pore volumes (1.15 ± 0.02 cm3/g).   5.3.1.2 XPS Analysis The compositions and surface electronic state of the fresh Mo2C/APC catalysts were further investigated by XPS. For the fresh catalysts, Mo, O, and C were detected and a detailed Mo 3d deconvolution of the spectra is reported in Table 5.2 and Figure 5.2. The Mo 3d spectra were deconvoluted into five different peaks attributed to Mo2+, Mo3+, Mo4+, Mo5+ and Mo6+ [128], each with Mo 3d5/2 and Mo3d3/2 spin-orbital splitting of 3.1 eV and an area ratio of 3:2. The peaks at 228.3 eV and 231.4 eV can be assigned to Mo2C. The Mo3+, Mo4+, and Mo5+ species are ascribed to MoOxCy species, while the peaks at higher B.E. correspond to Mo6+ attributed to Mo(VI) oxide species [168, 169]. The ratio of Mo carbide plus oxycarbide species to oxide species on the synthesized fresh catalyst was ~85:15, indicative of a significant formation of Mo2C related species at 700 oC. Moreover, there was no obvious difference in the Mo species composition as the Mo loading increased, which indicates that the CH4 produced at 700 oC during CHR is sufficient to convert all Mo precursors to the carbide for Mo nominal loadings of 133  ≤ 10 wt%. Furthermore, a good correlation between Mo loading and the XPS intensity ratio (IMo/IC) for the fresh Mo2C/APC catalysts, shown in Appendix E.4, Figure E.17, indicates monodispersion of Mo species on APC [170] under the present CHR temperature of 700 oC when the nominal Mo (wt.%) content was ≤ 10%.  134   Figure 5.2: The deconvolution of Mo 3d of fresh Mo2C/APC catalysts: (a) 2Mo2C/APC; (b) 5Mo2C/APC; and (c) 10Mo2C/APC.  Table 5.2: XPS analysis of fresh Mo2C/APC catalysts with different Mo loadings. Samples Mo 3d, At% O 1s, At% C 1s, At% Mo loading by ICP, wt% O/Mo atomic ratio IMo/Ic Mo2+ Mo3+ Mo4+ Mo5+ Mo6+ (B.E.=228.3 eV) (B.E.=228.8 eV) (B.E.=230.4 eV) (B.E.=232.0 eV) (B.E.=233.1 eV) 2Mo2C/APC 0.55  2.91  96.54  2.47 a  5.29  0.08 b  35.39 c  23.13 c 3.98 c 21.03 c  16.48 c 5Mo2C/APC 1.03  2.86  96.11  5.90  2.78  0.19  40.08  21.27  4.30  20.04  14.32  10Mo2C/APC 2.90  5.96  91.14  13.79  2.06  0.55  38.55  19.60  7.36  21.08  13.40  a. This value was measured by ICP-MS. b. The relative intensity of Mo to carbon obtained from XPS data. c. The deconvoluted peaks of Mo of different valences.  135  5.3.1.3 TEM/STEM-EDX Analysis and CO Uptake TEM micrographs of both fresh and used catalysts with different metal loadings are presented in Figure 5.3. The particle size distribution of Mo species was taken from multiple TEM images, where the particle had a lattice fringe spacing of 0.23 nm. The average particle size increased from 6.1 nm to 10.3 nm as Mo content increased, as shown by the lognormal distribution of Figure 5.3 (a, c, and e). The analysis also showed that the particles were well-distributed under present preparation conditions with Mo loading ≤ 10 wt%. In addition, the standard deviation of the particle size increased with increased Mo loading, indicating that at higher loading there was less control of the particle size.   136   137    Figure 5.3: TEM images and cluster size distribution of Mo2C/APC with different metal Mo loadings. (a) 2Mo2C/APC fresh; (b) 2Mo2C/APC used; (c) 5Mo2C/APC fresh; (d) 5Mo2C/APC used; (e) 10Mo2C/APC fresh; and (f) 10Mo2C/APC used.  138   The CO uptake data of the fresh catalyst (Table 5.3) also showed increased CO uptake as the Mo loading of the Mo2C/APC catalysts decreased, consistent with the decreased particle size measured by TEM.   Table 5.3: The particle size and CO uptake of fresh and used Mo2C/APC catalysts. Samples Fresh catalyst Used catalysts Identified phase a  Particle size (nm)b CO uptake, µmol/gmetalc Particle size (nm) b CO uptake, µmol/gmetalc 2Mo2C/APC-700 ― 6.1±1.5 137 9.3±2.2 1 5Mo2C/APC-700 β-Mo2C 7.3±2.0 126 9.6±1.5 5 10Mo2C/APC-700 β-Mo2C 10.3±3.2 116 16.0±12.0 4 a. From XRD analysis; b. The average particle size of Mo2C is reported by counting ≥ 100 particles and applying a lognormal distribution to the particle count data to obtain mean and standard deviation values. c. This data was obtained based on reduced passivated catalysts.  5.3.2 Catalyst Activity and Stability Hydrodesulphurization of dibenzothiophene (350 oC, 4.1 MPa, H2: feed = 600, LHSV= 4 h-1) was used to determine the catalytic activity of the carbide catalysts. Before performing the tests, the passivated catalysts were reduced under H2 flow at 400 oC for 2 h to remove the passivation layer. Subsequently DBT conversion data were collected in 30 to 40 min intervals until steady state was achieved. All the experiments were conducted for the same time on stream (TOS) of 600 min.  139   Figure 5.4: DBT conversion and selectivity of Mo2C/APC catalysts with different Mo loadings with TOS. (All the experiments were done with the same amount of Mo loading in the reactor): (a) 2Mo2C/APC; (b) 5Mo2C/APC; (c) 10Mo2C/APC; and (d) DBT conversion and HDS conversion for Mo2C/APC catalysts with different Mo loadings as a function of TOS.  To assess the effect of Mo2C particle size, the catalysts were tested with the same amount of Mo (0.019 g Mo) in the reactor. The conversion and selectivity data are presented in Figure 5.4, showing the changes in conversion and selectivity with TOS and Mo catalyst loading. All of the Mo2C catalysts deactivated in the presence of S, as shown in Figure 5.4 (a-c). Note that the DBT conversion decreased from 98% to 60% in 600 min on larger Mo2C particles (10Mo2C/APC); whereas, on smaller Mo2C particles (2Mo2C/APC) the loss in conversion is much less severe (from 100 to 95 % conversion). The relative decay constants (kd) associated with the loss in 140  conversion are reported in Table 5.4 (see details in Appendix D.3), and show increased deactivation rate with increased Mo2C particle size. The DBT conversion and product selectivities reached a steady state after ~150 min TOS for all three catalysts. The averaged steady state DBT conversion and product selectivity are reported in Table 5.4, based on the last five data points collected during the experiment. The difference between the DBT conversion and HDS conversion reported in Figure 5.4(d) and Table 5.4, show that there was no significant loss in HDS conversion relative to DBT conversion.  The primary products of DBT conversion, as determined by GCMS, were biphenyl (BPh), cyclohexylbenzene (CHB) and bicyclohexane (BCH), with less significant quantities of hydrocarbons and minor amounts of S-containing product—tetrahydrodibenzothiophene (THDBT). The change in selectivity with TOS is shown in Figure 5.4 (a-c). All Mo2C catalysts with different Mo loadings and hence particle sizes, showed very high selectivity towards CHB and BCH, at the beginning of the reaction (60% and 20%, respectively); whereas, both decreased with TOS until they reached a similar level (CHB ≈ 20%; BCH < 10%) and BPh became the primary product with selectivity > 70%. Importantly, this transformation occurred at a similar time (~150 min) for all Mo2C catalysts, independent of the Mo2C particle size. Note that over the 5MoS2/APC catalyst, the DBT conversion and product selectivity did not undergo any measurable change with TOS and selectivity to BPh was 73% (Table 5.4).  141   Table 5.4: Conversion and product selectivity for the hydrodesuphfurization of dibenzothiophene over Mo2C/APC and MoS2/APC catalysts at 350 oC and 4.1 MPa after stabilization and TOS > 150 mins. Catalyst Loading (g) Conversion of DBT (%)a HDS conversion of DBT (%)b Selectivity (%)a kd (h-1)c R2 THDBT BPh CHB BCH Mo2C of different particle sizes 2Mo2C/APC 0.77 94.89 93.19b 1.79a 76.02a 18.11a 4.08a 2.60E-3±1.28E-4 0.98 5Mo2C/APC 0.32 70.08 65.86 6.02 73.01 16.48 4.49 9.65E-3±1.71E-3 0.90 10Mo2C/APC 0.14 60.77 56.49 7.05 73.92 14.43 4.61 2.53E-2±5.78E-3 0.93 5MoS2/APC  0.77 100.00 100.00 0.00 72.97 23.35 3.68 ― ― a. The conversion and selectivity were obtained as the average of the last 5 data points collected after 400 min TOS period. b. The HDS conversion of DBT reported the HDS conversion to products free of S.  c. The decay constant was calculated based on exponential decay equation: −𝑑𝑎𝑑𝑡= 𝑘𝑑𝑎.   142  5.3.3 Used Catalyst Characterization 5.3.3.1 XRD and Physical Properties Analysis The XRD analysis of the used catalysts is reported in Figure 5.1. The used catalysts showed no visible phase change from Mo2C, suggesting that bulk sulfidation did not occur under the present reaction conditions. Similar phenomenon were also observed by Aegerter et al. [124]. Note that there are no peaks associated with MoS2 for the used 5MoS2/APC catalyst, probably due to a high MoS2 dispersion on this catalyst.   Table 5.1 shows that there was no noticeable change in surface area and pore volume for the used 2Mo2C/APC catalyst; whereas, a small decrease occurred for the used 5 and 10Mo2C/APC catalysts, when compared to the fresh catalyst. This decrease is probably due to the increased particle size blocking micropores.  5.3.3.2 XPS Analysis The surface composition of the used catalysts included the presence of S as shown by the survey scan in Figure 5.5 (a) and the component concentrations reported in Table 5.5. The narrow scan of the S 2p region suggests the existence of S12-, S22- and SO42- species [171, 172]. A detailed deconvolution of the S 2p signal is shown in Appendix E.4, Figure E.18. The peaks at 162.0 eV and 163.2 eV are assigned to S2- 2p3/2 and S2- 2p1/2, respectively, with a spin-orbit splitting of 1.2 eV and indicative of the presence of MoS2 species. Moreover, the S 2p peaks at 163.6 eV and 164.7 eV represent the formation of S22- of MoS3 species. The presence of SO42- species is also identified on the used Mo2C catalysts. To identify the difference between the sulfided Mo2C 143  surface and the MoS2 catalyst, the 5%MoS2/APC catalyst was analyzed as reported in Figure 5.5 (b). The same deconvolution method was applied and the same S species present as MoS2, MoS3, and SO42-were identified, indicating that sulfidation of Mo does occur on the Mo2C catalyst surface during the HDS reaction. Similar observations have been reported in previous studies [31, 124, 158]. The relative composition of the different S species on the used catalysts is reported in Table 5.5. The ratio of MoS2 (S2-) species to MoS3 (S22-) species is about 6 for the MoS2 catalyst; whereas, the ratio is about 1 (S2-: S22-≈1) for the used Mo2C catalysts. Furthermore, it is interesting to note that the S/Mo atom ratio (Table 5.5 column 8, 9) decreased with increased Mo loading.   Figure 5.5: XPS spectra of Mo2C catalysts: (a) Survey scan of fresh and used 10Mo2C/APC catalyst; (b) Narrow scan of Mo 3d spectra for used Mo2C catalysts with different metal loadings and used 5MoS2/APC.  144   Table 5.5: XPS analysis of used Mo2C/APC catalysts with different Mo loadings. Samples Mo 3d, At% O 1s, At% C 1s, At% S 2p, At% S, wt% S, wt% S/Mo atomic ratio, %c S/Mo atomic ratio, %d S12- 2p (B.E.=162.0 eV) S22- 2p (B.E.=163.6 eV) SO42- (B.E.=169.0 eV) 2Mo2C/APC 0.50a  3.99a  95.04a  0.47a  1.19a  0.22b  0.94c  0.27d  21.66  35.17  43.17  5Mo2C/APC 0.89  4.30  94.29  0.52  1.28  0.39  0.58  0.20  38.54  26.00  35.46  10Mo2C/APC 2.78  5.53  90.95  0.74  1.61  0.65  0.27  0.14  37.20  26.97  35.83  5MoS2/APC 0.62  4.16  94.35  0.87  2.16 0.31 1.40  0.19 76.84  12.94  10.23  a. Surface elemental (Mo, O, C, S) contents were measured by XPS. b. Bulk phase S content (wt%) was obtained by CHNS analyzer. c. The atomic ratio of S/Mo from XPS measurement. d. The atomic ratio of S/Mo from CHNS analyzer.    145  5.3.3.3 TEM/STEM-EDX Analysis As shown in Figure 5.3(b, d, and f), the bulk phase of Mo2C was still present on the used catalysts after the HDS reaction. However, the presence of MoS2 layers on top of the Mo2C particles, was confirmed by the lattice fringe spacing of 0.62 nm. This result is consistent with the XPS data, which demonstrated the formation of MoS2 on the used catalyst.  As part of the transition of Mo carbide species during the HDS reaction, the particle size distribution also changed. All the used catalysts showed an increase in particle size after reaction. For the 2Mo2C/APC and 5Mo2C/APC catalysts, the particle size increased to ~ 9.5 nm; whereas, it increased to 16.0 nm for the 10Mo2C/APC catalyst. The HADDF-STEM image and EDX elemental mapping of the used 2Mo2C/APC catalyst are presented in Figure 5.6, Figure E.19 and Figure E.20. This analysis shows that S is closely associated with the Mo, with the Mo and S elemental composition map of the selected area almost identical.   146   Figure 5.6: High angle annular dark field TEM scanning image (HAADF-STEM) of used 2Mo2C/APC catalyst (a); (b-e) Energy dispersive X-ray (EDX) elemental mappings of Mo, S, O and C; and (f) Overlay of C, Mo, and S distributions. 147   The line scan image (Figure 5.7) of the same particles shows the relationship more clearly. As the Mo concentration increased, the S concentration increased. Moreover, the S content is much higher in the presence of Mo than in its absence, indicating that most S remained on the surface of Mo after reaction.   Figure 5.7: (a) TEM image of used 10Mo2C/APC catalyst; (b) High angle annular dark field TEM scanning image (HAADF-STEM); and (c) Line scanning on selected particle. 148  5.4 Discussion and DFT Analysis The Mo2C/APC catalysts with 2, 5, and 10% Mo, prepared by CHR, showed similar surface compositions with ~85% of the Mo present as Mo carbide or oxycarbide species. The average particle size of the Mo species increased from 6.1 to 10.3 nm as the Mo loading increased from 2 to 10 wt%. Hence the catalysts are of similar composition but varying size, allowing one to examine the transition of Mo2C during HDS, accounting for particle size effects.  Although the Mo2C/APC catalysts of the present study showed different levels of DBT conversion, they all declined in activity within the first 600 min of the reaction, with the product selectivity also changing significantly. The data provide evidence that the transition of the Mo2C caused by S is one possible reason for the change in activity and selectivity during the HDS reaction. To exclude the possibility of carbon deposition or other mechanism that may block pores leading to a loss in activity, the textural properties of the used catalysts were measured and compared to the fresh catalysts. Both the specific surface area and pore volume (Table 5.1) do not change significantly before and after reaction, indicative of no pore blockage, even though the size of the Mo2C particles was observed to increase after reaction (Table 5.3).   The transition of Mo2C during the HDS reaction with MoS2 phase formation was also observed by Aegerter et al. [124] In the present study, both XPS and TEM analyses provide clear evidence of the transformation of Mo2C to MoS2 because of S species on the catalyst surface. However, the bulk phase of Mo2C did not change, as confirmed by XRD. Since the catalyst activity and selectivity stabilized after approximately 150 min TOS, we conclude that the outer surface of the 149  Mo2C particles are transformed to MoS2 which then act as primary active sites for HDS reactions.   Characterization of the fresh and used catalysts has provided clear evidence of Mo2C particle size varying with Mo loading of the Mo2C/APC catalysts, while maintaining constant Mo carbide content. As expected, S incorporation increased the particle size of the Mo species following HDS, since the Mo molar density of MoS2 (0.0316 mol/cm3) is lower than that of Mo2C (0.0873 mol/cm3) [173]. For the 2Mo2C/APC catalyst, the particle size increased ~50%; whereas, for the 5Mo2C/APC catalyst a ~30% increase in size was observed. This result is consistent with the S/Mo ratio from XPS (Table 5.5, column 8), which demonstrates that smaller Mo2C particles (with higher surface area) incorporated more S during HDS than larger Mo2C particles. The degree of particle size growth for the 10Mo2C/APC catalyst was ~56% and this relatively high value may be a consequence of agglomeration of adjacent particles. Agglomeration of particles would further decrease the S content of the used 10Mo2C/APC catalyst since the agglomerated particles would reduce the number of surface Mo2C sites exposed to S. Also, there are more edge sites on smaller particles, which will enhance S adsorption and incorporation on the catalyst surface [174]. The line scan and EDX mapping (Figure 5.7) show the close correlation between Mo and S concentration. However, from the TEM image the MoS2 layers appear as patches on top of Mo2C particles, rather than fully developed overlayers of MoS2 that cover the Mo2C, as proposed by Aegerter et al. [124]. Also, the MoS2-Mo2C morphology is quite different from the core-shell (Mo2C-MoS2) structure on Ni-Mo2C catalyst, reported previously in Chapter 3 [160].   150  From Figure 5.4, we observe that the activity change can be divided into two stages: (i) an initial deactivation period of about 150 mins; and (ii) a stabilized period where conversion and selectivity remain constant with TOS. This study reveals that the active sites of the Mo2C catalysts in stage (ii) must be similar to those present on MoS2 catalysts, since the catalyst selectivities were identical. As reported in the literature [103, 175], MoS2 catalysts usually have a low CO uptake, which is consistent with the used Mo2C catalyst CO uptake reported in Table 5.3.  Since the deactivation stage is relatively short and not easy to associate with catalyst characterization measurements, the effect of sulfur residue and C-S replacement on the catalyst surface was simulated through DFT calculation. Firstly, the adsorption energy of DBT on clean Mo2C (101) surface has been calculated. Figure E.21 (Appendix E.4) shows one terminal Mo site (labeled as Mo-t1) and two terminal C sites (C-t1 and C-t2, respectively) used in the analysis. Table E.4 (Appendix E.4) summarizes the adsorption of DBT on the clean Mo2C (101) surface, the most stable surface of Mo2C. Two modes of adsorption have been examined; vertical adsorption (V) and horizontal adsorption (H) each with two different orientations relative to the surface. These modes are identified DBT-V1, DBT-V2, DBT-H1, and DBT-H2 respectively. To better compare the DBT binding energies, we assume Mo-S bonding between Mo-t1 and the S from DBT in all cases. The calculation results show that DBT-H1 and DBT-H2 have stronger adsorption energies, with values of -1.53 eV and -1.58 eV, respectively, than that of DBT-V1 and DBT-V2, with values of -0.94 eV and -1.01 eV, respectively. Hence horizontal adsorption of DBT on the clean Mo2C (101) surface is most favourable. This result is consistent with the experimental selectivity data at the beginning of the reaction that show no BPh formation, a 151  product of the DDS of DBT (Figure 5.8). Secondly, at the beginning of the reaction, the high activity of Mo2C results in rapid conversion of DBT and further reaction of primary products. At the beginning of the reaction period, the Mo2C's (101) surface is free of S and horizontal adsorption of DBT is favoured over vertical adsorption, according to the DFT analysis. Horizontal adsorption of DBT means that ring hydrogenation will occur prior to S removal from DBT, i.e., the DBT conversion through the HYD route will be preferred. Vertical adsorption of DBT will yield direct breakage of the C-S bond and DBT conversion by the DDS route [176]. The DFT analysis shows that the horizontal adsorption energy for DBT is as high as -1.58 eV, higher than -1.3 eV for DBT adsorption on MoS2 surface [176].    Figure 5.8: Simplified reaction pathway of dibenzothiophene HDS via HYD and DDS routes over all Mo2C/APC catalysts with different Mo loadings.  Shortly after the beginning of the reaction, either S adsorption on the Mo2C surface or S incorporation occur and result in a change in product selectivity as shown by the experimental 152  data of Figure 5.4. This change was completed within about 150 mins; whereas, the analysis of the used catalyst was done after a TOS of about 600 mins. Hence DFT was used to assess the impact of S deposition on the Mo2C catalyst that would occur during the very initial stages of the DBT reaction. Since transition of the Mo2C structure to MoS2 is unlikely to occur at the beginning of the reaction, it is reasonable to assume that the selectivity change initially is caused by adsorption of sulfur from DBT on the Mo2C surface. Table E.5 presents the adsorption of DBT on the same Mo site (Mo-t1) of the Mo2C (101) surface but with a pre-adsorbed S atom located on-top the C-t1 site. Note that because of steric hindrance effects, only DBT-V1 and DBT-H1 adsorption orientations were evaluated. The results show that the adsorption energy of horizontal position is significantly lower than that of the clean Mo2C surface, which may explain the loss in activity observed during the HDS reaction; whereas, the energy of the vertical adsorption is slightly higher than that of the clean Mo2C surface. This indicates that the direct desulphurization route (DDS) is more favoured on the S-Mo2C (101) adsorbed surface, as observed experimentally and reported in Figure 5.4.   After 600 min TOS and stabilized selectivity and conversion, MoS2 formation is confirmed by multiple characterization methods. Since the catalyst characterization confirmed that S can change the Mo2C structure and form MoS2 species, S replacement of the carbon of Mo2C must occur. Hence, DFT calculations to assess DBT adsorption on the Mo2C (101) surface with different degrees of S replacement have been completed. Figure E.22 presents the C atoms replaced by S on the Mo2C (101) surface with different degrees of S replacement. Three different replacements of S were assessed, as shown in Table 5.6 and Figure E.22 (Appendix E.4), where X indicates the number of S atoms’ replaced on the surface. The calculations indicate that the S 153  replacement decreases the DBT adsorption strength on the Mo2C (101) surface. In the case where C is replaced by S at the C-t1 position, even though the activity is reduced, horizontal adsorption still dominates (Table 5.6, 1S and 2S replacement). However, at the C-t2 position, with one more carbon being replaced, the S atom is raised from the surface making the horizontal adsorption impossible; whereas, it enhances the vertical adsorption of DBT (Appendix E.4-Table E.4 and Table 5.6, 3 S replacement). Hence, the DDS selectivity is favoured as the S incorporation increases above a certain concentration.  Table 5.6: The calculated Gibbs free adsorption energy of DBT on S replaced Mo2C (101) surface with different adsorption angles.  Adsorption orientation  Gibbs free Adsorption energy with X (0~3) S replacement (eV) 0Sa 1S 2S 3S DBT-H1b -1.53 -1.41 -1.34 -0.89 DBT-H2 -1.58 -1.55 -1.32 -0.30 DBT-V1 -0.94 -1.07 -1.27 -0.53 DBT-V2 -1.01 -0.86 -0.67 -1.10 a. 0S represents clean (101) Mo2C surface; 1S means one C-t1 was replaced by one S atom; 2S means two C-t1 were replaced by two S atoms; 3S means two C-t1 and one C-t2 were replaced by three S atoms. b. Adsorption of DBT in a horizontal 1 (DBT-H1); Adsorption of DBT in a vertical 1 (DBT-V1).  Dissociative adsorption of H2 on the Mo2C (101) surface was also assessed on the Mo-t1. Figure E.23 shows the molecular adsorption of H2 on Mo-t1 site, followed by dissociative adsorption on a nearby bridge site of the clean Mo2C (101) surface, with S adsorbed on the C-t1 site, and with C-t1 site replaced by a S atom. The potential energy trends of dissociative adsorption of H2 on 154  these three surfaces is also reported. The calculated energy barrier (Ea) for H2 dissociative adsorption on clean Mo2C (101) surface is about 0.28 eV (calculated by -0.12-(-0.40) = 0.28), much lower than the barrier energy on conventional hydrotreating catalysts, MoS2 (101̅0) Mo-edge site reported as 0.91 eV [177].  In addition to relatively low energy barrier for H2 dissociative adsorption, the overall dissociative process (Er) is exothermic at -0.53 eV (Calculated by -0.93-(-0.40) = -0.53), higher than most cases on pure metal surfaces such as Co (0001) with -1.07 eV [178]; whereas, it is better than NiMoS, which is slightly endothermic (0.17 eV), or CoMoS, which is -0.15 eV. Figure E.23 shows that during the HDS reaction the energy barrier for dissociative adsorption of H2 becomes higher with either S atom adsorbed on top of C or S replacing C. Either way the reaction becomes less likely, especially in the S adsorbed case where the dissociative adsorption state is unstable (> 0 eV). As a result, the hydrogenation ability of Mo2C decreases during the HDS reaction possibly acting as one of the factors that cause the observed activity drop.  5.5 Conclusions Different Mo loading Mo2C/APC catalysts prepared at the same reduction temperature have different Mo2C particle sizes, yet similar concentrations of Mo surface species. Reduction temperature is the crucial factor that determines the degree of Mo2C formation from the precursor; whereas, the particle size primarily depends on the catalyst Mo loading. At the beginning of the HDS of DBT, the catalyst activity is very high and the HYD selectivity changes due to the adsorption of S species on Mo2C surface. This change continues to develop as the adsorption and replacement of C by S continues. Finally, both selectivity and activity reach a stable state, while selectivity tends towards DDS products. Faster deactivation occurs on larger 155  particles. The Mo2C  MoS2 transition is simulated with DFT calculation, and the change in DBT adsorption geometry follows the observed changes in product selectivites. After reaching a stable state, the Mo2C surfaces are mostly covered by MoS2 species according to XPS and TEM results.   156  Chapter 6: The Effect of Other Heteroatoms (N and O) on Mo2C/APC Catalysts 6.1 Introduction Following Chapter 5, this chapter was conducted as an exploratory study of the effect of other heteroatoms (N and O) on the performance of Mo2C/APC catalysts during hydrodenitrogenation (HDN) and hydrodeoxygenation (HDO) reactions, respectively. For the HDN study, carbazole (CBZ) was chosen as a model reactant because it has a similar structure to DBT, providing an opportunity to directly compare the impact of the N versus S on the reaction and the catalyst stability. Similarly, for the HDO reaction, dibenzofuran (DBF) was chosen as reactant, consistent with the reactants used for HDS and HDN.   6.2 Experimental The experimental methods follow those described in Chapter 5. The reaction was conducted at 350 oC and with a constant pressure of 4.1 MPa in a packed bed reactor. The synthesis of Mo2C/APC catalysts used in this chapter was the same as that described in Chapter 5. In addition, catalyst characterization methods followed those described in Chapter 5. The model compounds used in this chapter were carbazole (CBZ) and dibenzofuran (DBF), representing the N and O containing reactants typical of HDN and HDO processes, respectively.  157  6.3 Results and Discussion 6.3.1 Catalyst Activities in HDN of Carbazole The Mo2C/APC catalysts described in Chapter 5 with various particle sizes were tested for the HDN of carbazole with a carbazole concentration of ~ 0.2 wt.%, equivalent to 150 ppm N in the feed oil. As reported in the literature [51, 179], carbazole is a refractory N-containing compound.   Several products and intermediates were detected from the GC-MS product analysis as shown in Figure 6.1 and Table 6.1. For simplicity, the products are grouped into five categories, based on the reaction that generated the product. The product generated from direct denitrogenation (DDN) route is cyclohexylbenzene (CHB), which is produced following the hydrogenation of one aromatic ring of CBZ. The hydrogenation (HYD) route proceeds by hydrogenation of two rings, where cyclohexyl-cyclohexene (CHCHE) and bicyclohexyl (BCH) are the main products. The subsequent reaction is the isomerization of BCH to cyclopentylmethyl-cyclohexane (CPCHX) and hexylcyclohexane (HCH) along with the cracking reaction.  158   Figure 6.1: Reaction network of hydrodenitrogenation of carbazole over Mo2C/APC catalyst.   Table 6.1: Identified products from HDN of carbazole with Mo2C/APC catalysts at 350 oC and 4.1 MPa. Symbol Product name Reaction route CHB Cyclohexylbenzene Route of DDN  THCZ Tetrahydrocarbazole N-containing product CHBA 2-Cyclohexyl-benzenamine N-containing product DHCBZ Dodecahydro-1H-carbazole N-containing product CHCHE Cyclohexyl-cyclohexene HYD BCH Bicyclohexyl HYD CPCHX Cyclopentylmethyl-cyclohexane Isomerization HCH Hexylcyclohexane Isomerization Benzene, Cyclohexane, Toluene Cracking products  159  Similar to the results reported by Szymanska et al. [51], there was no BPh detected in this study. Almost all the obtained products were generated from the hydrogenation of the aromatic rings followed by ring opening and cracking reactions to produce the relevant isomerization and cracked products. There were also some products detected that retained the N-atom, indicating incomplete denitrogenation.   160   Figure 6.2: CBZ conversion and selectivity of Mo2C/APC catalysts with different Mo loadings with time on stream. (Mo loading held constant at 0.019 g in the reactor): (a) 2Mo2C/APC; (b) 5Mo2C/APC; (c) 10Mo2C/APC; and (d) CBZ conversion and HDN conversion for Mo2C/APC catalysts with different Mo loadings as a function of TOS.161  Figure 6.2 reports the carbazole conversion and product selectivity of the Mo2C/APC catalysts with various Mo loadings and corresponding particle sizes, measured at 350 oC and 4.1 MPa in the packed bed reactor. The data clearly show that the conversion of CBZ was relatively high in the initial stages of the reaction for all catalysts. However, the conversion decreased with time on stream (TOS). This observation is similar to what was observed during the HDS of DBT and reported in Chapter 5. The CBZ conversion decreased from 98% to 90% on the 2%Mo2C/APC catalyst; whereas, it decreased from 95% to 60% for the 10%Mo2C/APC catalyst, suggesting that the smaller the particle size, the higher the stabilized CBZ conversion.   Figure 6.2 (d) clearly shows a difference between the CBZ conversion and HDN conversion at steady-state; where, this difference between the two represents the N-containing reaction products generated during the HDN reaction. For example, the difference is about 8% (90−8390%) for the 2%Mo2C/APC catalyst; whereas, it increased to 54% (59−2759%) for the 10%Mo2C/APC catalyst. The higher HDS activity compared to the HDN activity of the Mo2C/APC catalysts, reflects the difficulty of removing N completely compared to S because of the higher bond energy of C-N versus C-S [180]. In addition, unlike the HDS reaction, HDN usually proceeds through a hydrogenation route prior to C-N bond cleavage. The presence of N may decrease the hydrogenation ability of Mo2C, suggesting a detrimental effect of N on the Mo2C surface. Furthermore, it is noted that the required time to reach a steady-state is independent with Mo2C catalyst particle size, which is similar to the HDS reaction.   162   Table 6.2: Conversion and product selectivity for HDN of CBZ over Mo2C/APC catalysts at 350 oC and 4.1 MPa. Catalyst Conversion of CBZ (%) a HDN conversion of CBZ (%) b Selectivity (%)a N-contained products DDN route HYD products Isomerization Cracking 2Mo2C/APC  89.75 83.26b 7.23a 3.92a 29.74a 28.86a 30.25a 5Mo2C/APC 78.49 65.64 16.38 0.12 14.62 6.19 62.69 10Mo2C/APC 58.67 27.45 53.21 0.55 30.62 6.65 8.97 a. The conversion and selectivity were obtained by the average number at steady-state.  b. The HDN conversion of CBZ represented the percentage of N removed products from the initial reactant.   163  Figure 6.2 (a-c) also shows the selectivity change of Mo2C/APC catalysts with time on stream. Taking 2Mo2C/APC catalyst as an example, the data clearly show a decrease in cracking products with TOS compared to the beginning of the reaction. The increase in N-containing products stabilized after ~250 min, indicating the complete transformation of the 2Mo2C/APC catalyst. On the 10Mo2C/APC catalyst, cracking, isomerization and HYD products all decreased and there was an obvious increased in N-containing products at the very beginning of the reaction (~100 min).   Table 6.3: The calculated decay constants (kd) of Mo2C/APC catalysts with various metal loadings in HDN of carbazole at 350 oC and 4.1 MPa.  Catalyst kd (h-1)a R2 2Mo2C/APC  3.70E-03±5.83E-04 0.97 5Mo2C/APC 5.62E-03±3.14E-04 0.98 10Mo2C/APC 2.26E-02±2.72E-03 0.90 a. The decay constant was calculated based on exponential decay equation: −𝑑𝑎𝑑𝑡= 𝑘𝑑𝑎  Similar to the effect of S reported in Chapter 5, the relative decay constants (kd) were calculated for the HDN reaction and these are reported in Table 6.3. The effect of N showed an increased deactivation rate with increased Mo2C particle size, similar to the trend observed during HDS. Moreover, the N poison effect is much stronger than that of S for smaller particle size Mo2C catalysts, including 2Mo2C/APC; whereas, there is no significant difference in the HDN and HDS deactivation rates on 10Mo2C/APC catalyst.  164  After observing the serious drop of catalyst activity in HDN of CBZ, a H2 reduction process was applied on the used Mo2C/APC catalyst as a way to regenerate the used catalyst. Considering the properties of the synthesized catalyst, the regeneration was conducted under pure H2 flow of 100 mL (STP)/min at 400 oC for 2 h, consistent with the process conditions used to remove the catalyst passivation layer after synthesis. The corresponding experimental data are reported in Figure 6.3. It can be observed that after the 1st regeneration, the catalyst performance improved marginally at the beginning of the regeneration (Figure 6.3, phase I regeneration). However, the catalyst activity did not return to the activity of the fresh 10Mo2C/APC catalyst. Furthermore, the CBZ conversion decreased to the same level as that of the stabilized fresh Mo2C catalyst. After the phase Ⅱ regeneration period shown in Figure 6.3, there was no further recovery of the catalyst performance, indicative of the irreversible effects of N on Mo2C/APC catalyst surface modification. Also, the product selectivity at steady-state did not change significantly among the fresh catalyst and after the two regeneration periods, as shown in Figure 6.3 (b). As reported in Zheng et al. [6], the N atoms tend to occupy some subsurface sites of Mo2C, thus the nitridation of carbide can be initiated during ammonia decomposition.    165   Figure 6.3: Catalyst regeneration (Phase I and II) of 10%Mo2C/APC after HDN reaction of carbazole: (a) Conversion of CBZ; and (b) Selectivity of the products.   166  6.3.2 Catalyst Activities in HDO of Dibenzofuran The O effect on the Mo2C/APC catalyst has also been studied using dibenzofuran (DBF) as the model reactant. The same reaction conditions as those used for the HDS and HDN reactions were applied for the HDO. The HDO reaction usually occurs through two kinetically significant pathways: DDO and HYD as shown in Figure 6.4. The obtained products can be grouped into several categories as listed in Table 6.4.   Figure 6.4: Reaction network of hydrodeoxygenation of dibenzofuran over Mo2C/APC catalyst.  Table 6.4: Identified products from HDO of dibenzofuran with Mo2C/APC catalysts at 350 oC and 4.1 MPa. Symbol Product name Grouped route BPh Biphenyl DDO CHB Cyclohexylbenzene HYD BCH Bicyclohexyl HYD CH Cyclohexane HYD 2-CHP Phenol, 2-cyclohexyl- O-containing products THDBF Tetrahydrodibenzofuran O-containing products Benzene, Cyclohexane Cracking products 167    Figure 6.5: DBF conversion and selectivity of Mo2C/APC catalysts with different Mo loadings with time on stream. (Mo loading held constant at 0.019 g in the reactor): (a) 2Mo2C/APC; (b) 5Mo2C/APC; and (c) 10Mo2C/APC.  Figure 6.5 shows the performance of Mo2C/APC catalysts with various Mo loadings for the HDO of dibenzofuran (DBF; equv. 170 ppm O). The data show that DBF is completely converted for all three catalysts at 350 oC, 4.1 MPa and LHSV= 4 h-1. However, the selectivity shows a different trend. For the 2%Mo2C/APC catalyst (Figure 6.5 (a)), the HYD product-CHB decreased from ~55% to ~38%, while there was an obvious increase in BPh from ~15% to 27%. 168  Also note that the selectivity reached a relatively stable status after 150 min of HDO reaction. For the 10%Mo2C/APC catalyst (Figure 6.5 (c)), the time taken to reach steady-state is longer at ≥ 400 min. Moreover, it is interesting to observe that all catalysts reach the similar selectivity, independent of the Mo2C particle size. This result shows that larger particles are less prone to surface oxidation and hence have the ability to reduce oxidation. A similar phenomenon was also reported by Stellwagen et al. [168].   6.4 Conclusion The effect of reactants containing N and O atoms on Mo2C catalysts with different particle sizes have been studied during hydrotreating. N has a detrimental effect on Mo2C performance and results in irreversible catalyst deactivation; whereas, the effect of O is reversible. Moreover, the impact of these three atoms on catalyst deactivation severity can be arranged in decreasing order as N > S > O. 169  Chapter 7: Conclusions and Recommendations 7.1 Conclusions The main contribution from this study is the synthesis of a mesoporous carbon supported Mo2C catalyst, derived from Canadian oil sands petroleum coke and the successful application of these catalysts in several hydrotreating reactions.   The first part of the study was conducted using a commercial activated charcoal (AC) to assess the viability of carbon supported Mo2C catalyst synthesis. This study showed the successful preparation of Mo2C-MoOxCy/AC catalysts at relatively low temperature (600~800 oC). The corresponding catalysts prepared at different reduction temperatures were tested for the HDO of 4-methlyphenol at 350 oC and 4.3 MPa H2. Catalyst screening identified the optimal Mo2C/AC catalyst with a CHR temperature of 650 oC based on catalytic activity. Also, both Mo2C and MoOxCy function as active sites and the formation of these active sites is the result of O adsorption and/or exchange with the catalyst during HDO. The kinetics was closely related to the amount of Mo2C/MoOxCy active sites. Moreover, a high DDO selectivity was identified.   The application of Mo2C/AC catalysts was extended to HDS and the effect of Ni addition to the Mo2C/AC was determined. The addition of Ni reduced the CHR temperature required for Mo2C generation by ~100 oC. A phase separation occurred when the Ni:Mo ratio was > 0.38, resulting in decreased catalyst activity. The highest HDS activity occurred for the catalysts with Ni:Mo ratios of 0.38 and 0.19, when prepared at the relatively low CHR temperature of 550 and 600 oC. Sulfidation of the Ni-Mo2C/AC catalysts occurred during reaction in 2.0 wt% dibenzothiophene (DBT) at 350 oC and 2.1 MPa H2. The uptake of S was enhanced with increased Ni content, 170  resulting in the formation of Mo2C-MoS2 core-shell structures. The stacking height of the MoS2 shell increased with increased Ni content and promoted the direct desulfurization of DBT.  The synthesis of Mo2C on activated petroleum coke (APC) was also investigated, with a focus on mesopore development. From the previous study using AC, it was known that CHR temperature is an important variable for Mo2C synthesis. Therefore, a series of APC supported catalysts were prepared by varying the CHR temperature from 500 to 800 oC, to yield mesoporous Mo2C/APC catalysts. The transition of Mo species (MoOx  MoOxCy  Mo2C) was identified as a function of CHR temperature. The generated Mo2C species had a significant effect on CH4 generation, which in turn modified the physical structure of APC. The maximum mesopore volume was achieved at a CHR temperature of 750 oC, with 37% of the total pore volume appearing as mesopores. A longer CHR hold time facilitates total pore/mesopore volume development. The 4-methylphenol HDO activity of the synthesized Mo2C/APC catalysts was relatively high compared with Mo2C/AC catalysts and the activity increased with increased mesoporosity.   The impact of S the on Mo2C/APC catalyst activity was also investigated for a series of catalysts with varying Mo2C loadings and hence particle size. XPS results of the fresh catalysts showed an invariant Mo composition with different Mo catalyst loading, where a constant ~40% of Mo2C species was achieved. Analysis of the used catalysts after the HDS of DBT showed that a patchy MoS2 surface structure was formed on top of the Mo2C. Initial catalyst deactivation was observed for all catalysts and the HDS activities decreased with increased particle size of the Mo2C. DFT was used to demonstrate the significant effect of S adsorbed on Mo2C (101) and S substituted on 171  the Mo2C (101) surface, on the adsorption of DBT. The calculated results were consistent with the observed selectivity shift during DBT HDS from HYD to DDS products.   Finally, an exploratory study was conducted of the effect of the two heteroatoms N and O. The model reactants carbazole (CBZ) and dibenzofuran (DBF), having similar structure to DBT but for the presence of N or O in place of the S atom, were used to assess the impact of the heteroatom on the activity and stability of the Mo2C/APC catalysts. N has a stronger deactivation effect than S, and the deactivation was irreversible even after H2 regeneration. Most of the active sites for hydrogenation were poisoned by the presence of N. For HDO higher activity (almost 100% conversion) was observed and the selectivity changed with time on stream, with decreasing hydrogenation and increasing direct deoxygenation. Hence one concludes that the O can modify the catalyst surface but the modification is reversible and the formed species (MoOxCy) are active for HDO reaction.   7.2 Recommendations 7.2.1 Mo2C Catalyst Properties Since Mo2C is air-sensitive, all the catalyst samples of the study were handled in two ways: (a) passivation; (b) fast transfer in a N2 glove bag, to minimize the effects of air exposure. As reported in Chapter 2, there is a large difference between passivated Mo2C catalysts and fresh catalysts, based on the measured CO uptake data. Thus, it is important to consider in-situ catalyst characterization techniques to monitor the catalyst surface change during reaction, without exposure to air, which will provide more insight into catalyst property-activity relationships. Examples include FT-IR analysis of probe molecule adsorption (CO, for example) using a 172  DRIFTS reaction cell that would allow reactions on samples to be followed by adsorption experiment without exposing the catalyst to the atomosphere. In addition, an XPS unit with a high pressure chamber and sample transfer station would allow surface analysis of fresh catalysts and catalysts after exposure to different reactants to be completed without air exposure of the catalyst sample. Furthermore, electron energy loss spectroscopy (EELS) is a good characterization method to determine the oxide state of a material. Thus, we can further apply EELS to clearly identify the Mo oxycarbide species present in the catalysts.   7.2.2 Promoter Effect in HDS Reaction Ni addition was shown to significantly decrease the CHR temperature required for Mo2C formation by 100 oC. In future work, other promoters should be explored, especially Co, since it is another commonly used promoter in conventional Co (Ni)MoS/Al2O3 hydrotreating catalysts. In preliminary work completed during the present study, Co was less effective than Ni for Mo2C catalyst generated by CHR at 550 oC, but other synthesis conditions need to be explored to determine the best synthesis conditions.   Also, since the data reported in Chapter 3 were based on activated charcoal (AC), it is recommended that APC be used as a support of the Ni-Mo2C catalyst, to determine the impact of the carbon supports. In addition, based on the data reported in Appendix H-Table H.2, Ni-Mo2C catalyst has a significant effect on the APC mesopore development. A further study should be conducted to determine how the various metal/metal carbide active sites contribute to the pore development.   173  7.2.3 Deactivation of Mo2C and Ni-Mo2C in HDS  As reported in Chapters 3 and 5, both Mo2C and Ni-Mo2C catalysts supported on carbon deactivated in the presence of S. For the Mo2C/APC catalyst, patches of MoS2 were formed on top of the Mo2C during HDS; whereas, for Ni-Mo2C/AC, a core-shell structure was formed. Both structures have catalytic activity in HDS, although all are less effective than the fresh Mo2C. One possible approach to limit the transformation of the Mo2C is to encapsulate the Mo2C into a frame/shell structure that would prevent direct S contact with the Mo2C, thereby improving catalyst stability [181].   In Chapter 5, the transition of Mo2C/APC catalyst in the presence of S was studied. In future work, the deactivation study should be expanded to Ni-Mo2C/APC to compare the performance between these two APC supported catalysts. Considering the results in Chapter 3, it was shown that Ni-Mo2C/AC deactivated faster than Mo2C/AC in HDS of DBT. If it also occurs on Ni-Mo2C/APC, some experiments need to be designed to capture the faster transition, such as reducing the reactant concentration and decreasing the reaction temperature.   7.2.4 Effect of N and O on Mo2C Catalysts In Chapter 6, the deactivation of Mo2C catalysts in the presence of N during HDN was demonstrated. However, no relevant characterization data were obtained that could clarify the changes in catalyst morphology and/or the surface chemical and physical properties that could account for the deactivation. Several characterization methods have been proposed here which can be applied in future work: (1) CHNS analyzer can used to measure the bulk concentration of N on the used Mo2C/APC catalyst, which provides an idea of how much N is left on the catalyst. 174  (2) Pyridine can be used as the N reactant in FT-IR measurement since it is easier to be evaporated/removed. After pyridine treatment, a high vacuum can be applied to remove all the physically adsorbed N species. Followed by this, CO can be used as the probe molecule to measure the N-modified Mo2C/APC surface. (3) NH3-TPD is another potential method to measure the N on Mo2C/APC catalyst. It will provide an idea as to how N interacts with Mo2C species. (4) Electron energy-loss spectroscopy (EELS) will be a potential technique as well to study the N modification on Mo2C catalyst [6]. (5) DFT calculation can be applied to the N adsorbed Mo2C catalyst, similar to the S study as illustrated in Chapter 5, to explore the catalyst poisoning by N. For the O study, since we didn’t see any deactivation at present conditions, the experimental conditions should be adjusted to capture this change. Besides, some small molecules should be considered, such as pyridine and furan to see if this deactivation is related to molecular structure.   Additionally, it is recommended that to apply Mo2C/C catalyst into real bio-oil upgrading study due to the good performance and coke resistant ability.  7.2.5 Mesoporous Carbon Applications From the study of Chapter 2 that demonstrated the successful preparation of a mesoporous carbon material and an application in HDO of 4-methylphenol reaction, further study to utilize the mesoporous carbon as an adsorbent or electrode material is worth investigating. Also, this can be extended to other catalyst (MoS2/APC) synthesis based on the mesoporous structure. Due to the coke resistant ability of carbon supported catalyst, it is worth to utilize it in heavy oil upgrading. The used C based catalyst should be easily recycled and regenerated after several 175  cycles’ utilization. The properties of the used catalyst can be compared with the coke-catalyst mixture formed in slurry-phase residue hydroconversion as reported in our group’s previous work [182].   176  Bibliography 1. R.B. Levy and M. Boudart, Platinum-like behavior of tungsten carbide in surface catalysis. science, 1973. 181(4099): p. 547-549  2. S.T. Oyama, Preparation and catalytic properties of transition metal carbides and nitrides. Catalysis Today, 1992. 15(2): p. 179-200. 3. D.J. Sajkowski and S.T. Oyama, Catalytic hydrotreating by molybdenum carbide and nitride: unsupported Mo2N and Mo2C/Al2O3. Applied Catalysis A: General, 1996. 134(2): p. 339-349. 4. F.H. Ribeiro, M. Boudart, R.A. Dalla Betta, and E. Iglesia, Catalytic reactions of n-Alkanes on β-W2C and WC: The effect of surface oxygen on reaction pathways. Journal of Catalysis, 1991. 130(2): p. 498-513  5. C. Bouchy, C. Pham-Huu, B. Heinrich, C. Chaumont, and M.J. Ledoux, Microstructure and Characterization of a Highly Selective Catalyst for the Isomerization of Alkanes: A Molybdenum Oxycarbide. Journal of Catalysis, 2000. 190(1): p. 92-103. 6. W. Zheng, T.P. Cotter, P. Kaghazchi, T. Jacob, B. Frank, K. Schlichte, W. Zhang, D.S. Su, F. Schüth, and R. Schlögl, Experimental and Theoretical Investigation of Molybdenum Carbide and Nitride as Catalysts for Ammonia Decomposition. Journal of the American Chemical Society, 2013. 135(9): p. 3458-3464. 7. K.D. Sabnis, M.C. Akatay, Y. Cui, F.G. Sollberger, E.A. Stach, J.T. Miller, W.N. Delgass, and F.H. Ribeiro, Probing the active sites for water–gas shift over Pt/molybdenum carbide using multi-walled carbon nanotubes. Journal of Catalysis, 2015. 330: p. 442-451  8. J.A. Schaidle, A.C. Lausche, and L.T. Thompson, Effects of sulfur on Mo2C and Pt/Mo2C catalysts: Water gas shift reaction. Journal of Catalysis, 2010. 272(2): p. 235-245. 177  9. W.F. Chen, C.H. Wang, K. Sasaki, N. Marinkovic, W. Xu, J.T. Muckerman, Y. Zhu, and R.R. Adzic, Highly active and durable nanostructured molybdenum carbide electrocatalysts for hydrogen production. Energy & Environmental Science, 2013. 6(3): p. 943-951. 10. J.S. Lee and M. Boudart, Hydrodesulfurization of thiophene over unsupported molybdenum carbide. Applied Catalysis, 1985. 19(1): p. 207-210. 11. A. Hynaux, C. Sayag, S. Suppan, J. Trawczynski, M. Lewandowski, A. Szymanska-Kolasa, and G. Djéga-Mariadassou, Kinetic study of the hydrodesulfurization of dibenzothiophene over molybdenum carbides supported on functionalized carbon black composite: Influence of indole. Applied Catalysis B: Environmental, 2007. 72(1-2): p. 62-70. 12. J.A. Schaidle, N.M. Schweitzer, O.T. Ajenifujah, and L.T. Thompson, On the preparation of molybdenum carbide-supported metal catalysts. Journal of catalysis, 2012. 289: p. 210-217  13. Y. Chen, S. Choi, and L.T. Thompson, Low temperature CO2 hydrogenation to alcohols and hydrocarbons over Mo2C supported metal catalysts. Journal of Catalysis, 2016. 343: p. 147-156  14. J. Lee, S. Oyama, and M. Boudart, Molybdenum carbide catalysts: I. Synthesis of unsupported powders. Journal of Catalysis, 1987. 106(1): p. 125-133. 15. T. Xiao, A.P.E. York, V.C. Williams, H. Al-Megren, A. Hanif, X. Zhou, and M.L.H. Green, Preparation of Molybdenum Carbides Using Butane and Their Catalytic Performance. Chemistry of Materials, 2000. 12(12): p. 3896-3905. 16. P.D. Costa, C. Potvin, J.-M. Manoli, B. Genin, and G. Djéga-Mariadassou, Deep hydrodesulphurization and hydrogenation of diesel fuels on alumina-supported and bulk molybdenum carbide catalysts. Fuel, 2004. 83(13): p. 1717-1726. 178  17. T. Xiao, A.P.E. York, H. Al-Megren, C.V. Williams, H. Wang, and M.L.H. Green, Preparation and characterisation of bimetallic cobalt and molybdenum carbides. Journal of Catalysis, 2001. 202(1): p. 100-109  18. Q. Zhu, Q. Chen, X. Yang, and D. Ke, A new method for the synthesis of molybdenum carbide. Materials Letters, 2007. 61(29): p. 5173-5174. 19. A.L. Jongerius, R.W. Gosselink, J. Dijkstra, J.H. Bitter, P.C.A. Bruijnincx, and B.M. Weckhuysen, Carbon nanofiber supported transition-metal carbide catalysts for the hydrodeoxygenation of guaiacol. ChemCatChem, 2013. 5(10): p. 2964-2972. 20. D. Mordenti, D. Brodzki, and G. Djéga-Mariadassou, New synthesis of Mo2C 14 nm in average size supported on a high specific surface area carbon material. Journal of solid state chemistry, 1998. 141(1): p. 114-120  21. C. Liang, P. Ying, and C. Li, Nanostructured β-Mo2C prepared by carbothermal hydrogen reduction on ultrahigh surface area carbon material. Chemistry of materials, 2002. 14(7): p. 3148-3151. 22. M. Kaewpanha, G. Guan, Y. Ma, X. Hao, Z. Zhang, P. Reubroychareon, K. Kusakabe, and A. Abudula, Hydrogen production by steam reforming of biomass tar over biomass char supported molybdenum carbide catalyst. International journal of hydrogen energy, 2015. 40(25): p. 7974-7982  23. J. Han, J. Duan, P. Chen, H. Lou, X. Zheng, and H. Hong, Carbon-supported molybdenum carbide catalysts for the conversion of vegetable oils. ChemSusChem, 2012. 5(4): p. 727-733  179  24. B. Frank, K. Friedel, F. Girgsdies, X. Huang, R. Schlögl, and A. Trunschke, CNT‐Supported MoxC Catalysts: Effect of Loading and Carburization Parameters. ChemCatChem, 2013. 5(8): p. 2296-2305  25. T. Mo, J. Xu, Y. Yang, and Y. Li, Effect of carburization protocols on molybdenum carbide synthesis and study on its performance in CO hydrogenation. Catalysis Today, 2016. 261(Supplement C): p. 101-115. 26. C. Liang, W. Ma, Z. Feng, and C. Li, Activated carbon supported bimetallic CoMo carbides synthesized by carbothermal hydrogen reduction. Carbon, 2003. 41(9): p. 1833-1839. 27. C. Liang, F. Tian, Z. Li, Z. Feng, Z. Wei, and C. Li, Preparation and adsorption properties for thiophene of nanostructured W2C on ultrahigh-surface-area carbon materials. Chemistry of materials, 2003. 15(25): p. 4846-4853  28. N. Ji, T. Zhang, M. Zheng, A. Wang, H. Wang, X. Wang, and J.G. Chen, Direct Catalytic Conversion of Cellulose into Ethylene Glycol Using Nickel-Promoted Tungsten Carbide Catalysts. Angewandte Chemie, 2008. 120(44): p. 8638-8641  29. Z. Zhao, H. Yang, Y. Li, and X. Guo, Cobalt-modified molybdenum carbide as an efficient catalyst for chemoselective reduction of aromatic nitro compounds. Green chemistry, 2014. 16(3): p. 1274-1281. 30. L.A. Santillán-Vallejo, J.A. Melo-Banda, A.I.R. de la Torre, G. Sandoval-Robles, J.M. Domínguez, A. Montesinos-Castellanos, and J.A. de los Reyes-Heredia, Supported (NiMo, CoMo)-carbide, -nitride phases: Effect of atomic ratios and phosphorus concentration on the HDS of thiophene and dibenzothiophene. Catalysis today, 2005. 109(1): p. 33-41. 180  31. E. Puello-Polo and J.L. Brito, Effect of the type of precursor and the synthesis method on thiophene hydrodesulfurization activity of activated carbon supported Fe-Mo, Co-Mo and Ni-Mo carbides. Journal of Molecular Catalysis A: Chemical, 2008. 281(1): p. 85-92  32. J.G. Speight, The desulfurization of heavy oils and residua1999: CRC Press. 33. X. Li, A. Wang, M. Egorova, and R. Prins, Kinetics of the HDS of 4,6-dimethyldibenzothiophene and its hydrogenated intermediates over sulfided Mo and NiMo on γ-Al2O3. Journal of Catalysis, 2007. 250(2): p. 283-293. 34. M. Houalla, D.H. Broderick, A.V. Sapre, N.K. Nag, V.H.J. de Beer, B.C. Gates, and H. Kwart, Hydrodesulfurization of methyl-substituted dibenzothiophenes catalyzed by sulfided CoMo/γ-Al2O3. Journal of Catalysis, 1980. 61(2): p. 523-527. 35. F. Bataille, J.-L. Lemberton, P. Michaud, G. Pérot, M. Vrinat, M. Lemaire, E. Schulz, M. Breysse, and S. Kasztelan, Alkyldibenzothiophenes Hydrodesulfurization-Promoter Effect, Reactivity, and Reaction Mechanism. Journal of Catalysis, 2000. 191(2): p. 409-422. 36. M. Egorova and R. Prins, 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): p. 417-427  37. D. Ferdous, A.K. Dalai, and J. Adjaye, Comparison of hydrodenitrogenation of model basic and nonbasic nitrogen species in a trickle bed reactor using commercial NiMo/Al2O3 Catalyst. Energy & Fuels, 2003. 17(1): p. 164-171  38. H. Topsøe, B.S. Clausen, and F.E. Massoth, Hydrotreating catalysis1996: Springer. 39. S. Eijsbouts, On the flexibility of the active phase in hydrotreating catalysts. Applied Catalysis A: General, 1997. 158(1–2): p. 53-92. 181  40. K. Sakanishi, T. Nagamatsu, I. Mochida, and D.D. Whitehurst, Hydrodesulfurization kinetics and mechanism of 4, 6-dimethyldibenzothiophene over NiMo catalyst supported on carbon. Journal of Molecular Catalysis A: Chemical, 2000. 155(1): p. 101-109. 41. H. Farag, I. Mochida, and K. Sakanishi, Fundamental comparison studies on hydrodesulfurization of dibenzothiophenes over CoMo-based carbon and alumina catalysts. Applied Catalysis A: General, 2000. 194–195: p. 147-157. 42. H. Farag, D.D. Whitehurst, K. Sakanishi, and I. Mochida, Carbon versus alumina as a support for Co–Mo catalysts reactivity towards HDS of dibenzothiophenes and diesel fuel. Catalysis Today, 1999. 50(1): p. 9-17. 43. S. Eijsbouts, V.H.J. De Beer, and R. Prins, Hydrodenitrogenation of quinoline over carbon-supported transition metal sulfides. Journal of Catalysis, 1991. 127(2): p. 619-630. 44. M. Pang, C. Liu, W. Xia, M. Muhler, and C. Liang, Activated carbon supported molybdenum carbides as cheap and highly efficient catalyst in the selective hydrogenation of naphthalene to tetralin. Green chemistry, 2012. 14(5): p. 1272-1276. 45. I. Kojima and E. Miyazaki, Catalysis by transition metal carbides: V. Kinetic measurements of hydrogenation of CO over TaC, TiC, and Mo2C catalysts. Journal of Catalysis, 1984. 89(1): p. 168-171. 46. H.K. Park and K.L. Kim, Hydrodesulfurization of dibenzothiophene over supported and unsupported molybdenum carbide catalysts. Korean Journal of Chemical Engineering, 1998. 15(6): p. 625-630. 47. P. Da Costa, C. Potvin, J.M. Manoli, J.L. Lemberton, G. Pérot, and G. Djéga-Mariadassou, New catalysts for deep hydrotreatment of diesel fuel: Kinetics of 4,6-182  dimethyldibenzothiophene hydrodesulfurization over alumina-supported molybdenum carbide. Journal of Molecular Catalysis A: Chemical, 2002. 184(1–2): p. 323-333. 48. A. Hynaux, C. Sayag, S. Suppan, J. Trawczynski, M. Lewandowski, A. Szymanska-Kolasa, and G. Djéga-Mariadassou, Kinetic study of the hydrodesulfurization of dibenzothiophene over molybdenum carbides supported on functionalized carbon black composite: Influence of indole. Applied Catalysis B: Environmental, 2007. 72(1–2): p. 62-70. 49. P. Da Costa, J.-M. Manoli, C. Potvin, and G. Djéga-Mariadassou, Deep HDS on doped molybdenum carbides: From probe molecules to real feedstocks. Catalysis today, 2005. 107: p. 520-530  50. J.-G. Choi, J.R. Brenner, and L.T. Thompson, Pyridine hydrodenitrogenation over molybdenum carbide catalysts. Journal of Catalysis, 1995. 154(1): p. 33-40  51. A. Szymańska, M. Lewandowski, C. Sayag, and G. Djéga-Mariadassou, Kinetic study of the hydrodenitrogenation of carbazole over bulk molybdenum carbide. Journal of Catalysis, 2003. 218(1): p. 24-31. 52. J. Fan, T.N. Kalnes, M. Alward, J. Klinger, A. Sadehvandi, and D.R. Shonnard, Life cycle assessment of electricity generation using fast pyrolysis bio-oil. Renewable Energy, 2011. 36(2): p. 632-641  53. D. Mohan, C.U. Pittman, and P.H. Steele, Pyrolysis of wood/biomass for bio-oil: a critical review. Energy & Fuels, 2006. 20(3): p. 848-889  54. Q. Lu, W.-Z. Li, and X.-F. Zhu, Overview of fuel properties of biomass fast pyrolysis oils. Energy Conversion and Management, 2009. 50(5): p. 1376-1383. 183  55. P.M. Mortensen, J.D. Grunwaldt, P.A. Jensen, K.G. Knudsen, and A.D. Jensen, A review of catalytic upgrading of bio-oil to engine fuels. Applied Catalysis A: General, 2011. 407(1): p. 1-19. 56. S. Xiu and A. Shahbazi, Bio-oil production and upgrading research: A review. Renewable and Sustainable Energy Reviews, 2012. 16(7): p. 4406-4414. 57. J.A.R. Van Veen, H.A. Colijn, P. Hendriks, and A.J. Van Welsenes, On the formation of type I and type II NiMoS phases in NiMo/Al2O3 hydrotreating catalysts and its catalytic implications. Fuel Processing Technology, 1993. 35(1): p. 137-157  58. Q. Bu, H. Lei, A.H. Zacher, L. Wang, S. Ren, J. Liang, Y. Wei, Y. Liu, J. Tang, Q. Zhang, and R. Ruan, A review of catalytic hydrodeoxygenation of lignin-derived phenols from biomass pyrolysis. Bioresource Technology, 2012. 124: p. 470-477. 59. A.L. Jongerius, R. Jastrzebski, P.C.A. Bruijnincx, and B.M. Weckhuysen, CoMo sulfide-catalyzed hydrodeoxygenation of lignin model compounds: An extended reaction network for the conversion of monomeric and dimeric substrates. Journal of Catalysis, 2012. 285(1): p. 315-323. 60. E. Furimsky, Carbons and Carbon-supported Catalysts in Hydroprocessing2008: Royal Society of Chemistry. 61. F. Luck, A review of support effects on the activity and selectivity of hydrotreating catalysts. Bulletin des Sociétés Chimiques Belges, 1991. 100(11‐12): p. 781-800  62. G. De La Puente, A. Gil, J.J. Pis, and P. Grange, Effects of support surface chemistry in hydrodeoxygenation reactions over CoMo/activated carbon sulfided catalysts. Langmuir, 1999. 15(18): p. 5800-5806. 184  63. M. Ferrari, B. Delmon, and P. Grange, Influence of the active phase loading in carbon supported molybdenum–cobalt catalysts for hydrodeoxygenation reactions. Microporous and mesoporous materials, 2002. 56(3): p. 279-290  64. J. Wildschut, M. Iqbal, F.H. Mahfud, I.M. Cabrera, R.H. Venderbosch, and H.J. Heeres, Insights in the hydrotreatment of fast pyrolysis oil using a ruthenium on carbon catalyst. Energy & Environmental Science, 2010. 3(7): p. 962-970. 65. J.M. Solar, F.J. Derbyshire, V.H.J. de Beer, and L.R. Radovic, Effects of surface and structural properties of carbons on the behavior of carbon-supported molybdenum catalysts. Journal of Catalysis, 1991. 129(2): p. 330-342. 66. J.-S. Choi, G. Bugli, and G. Djéga-Mariadassou, Influence of the Degree of Carburization on the Density of Sites and Hydrogenating Activity of Molybdenum Carbides. Journal of Catalysis, 2000. 193(2): p. 238-247. 67. B. Dhandapani, T. St. Clair, and S.T. Oyama, Simultaneous hydrodesulfurization, hydrodeoxygenation, and hydrogenation with molybdenum carbide. Applied Catalysis A: General, 1998. 168(2): p. 219-228. 68. G.M. Dolce, P.E. Savage, and L.T. Thompson, Hydrotreatment activities of supported molybdenum nitrides and carbides. Energy & fuels, 1997. 11(3): p. 668-675. 69. W.-S. Lee, Z. Wang, R.J. Wu, and A. Bhan, Selective vapor-phase hydrodeoxygenation of anisole to benzene on molybdenum carbide catalysts. Journal of Catalysis, 2014. 319: p. 44-53. 70. H. Ren, W. Yu, M. Salciccioli, Y. Chen, Y. Huang, K. Xiong, D.G. Vlachos, and J.G. Chen, Selective hydrodeoxygenation of biomass-derived oxygenates to unsaturated hydrocarbons using molybdenum carbide catalysts. ChemSusChem, 2013. 6(5): p. 798-801. 185  71. F. Rodriguez-Reinoso, M. Molina-Sabio, and M.T. González, The use of steam and CO2 as activating agents in the preparation of activated carbons. Carbon, 1995. 33(1): p. 15-23  72. J.M. Hill, A. Karimi, and M. Malekshahian, 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): p. 1618-1626  73. R. Dippanfilo and N.O. Egiebor, Structural Features of Activated Carbon Produced From Oil Sands Coke. Developments in Chemical Engineering and Mineral Processing, 1995. 3(1): p. 3-13. 74. A. Karimi, O. Thinon, J. Fournier, and J.M. Hill, 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): p. 1491-1499  75. C.C. Small, Z. Hashisho, and A.C. Ulrich, Preparation and characterization of activated carbon from oil sands coke. Fuel, 2012. 92(1): p. 69-76. 76. M. Wu, Q. Zha, J. Qiu, X. Han, Y. Guo, Z. Li, A. Yuan, and X. Sun, Preparation of porous carbons from petroleum coke by different activation methods. Fuel, 2005. 84(14–15): p. 1992-1997. 77. S.H. Lee and C.S. Choi, Chemical activation of high sulfur petroleum cokes by alkali metal compounds. Fuel Processing Technology, 2000. 64(1-3): p. 141-153  78. H. Zhang, J. Chen, and S. Guo, Preparation of natural gas adsorbents from high-sulfur petroleum coke. Fuel, 2008. 87(3): p. 304-311. 79. R. Xiao, S. Xu, Q. Li, and Y. Su, The effects of hydrogen on KOH activation of petroleum coke. Journal of Analytical and Applied Pyrolysis, 2012. 96: p. 120-125  186  80. E. Raymundo-Pinero, P. Azais, T. Cacciaguerra, D. Cazorla-Amorós, A. Linares-Solano, and F. Béguin, KOH and NaOH activation mechanisms of multiwalled carbon nanotubes with different structural organisation. Carbon, 2005. 43(4): p. 786-795. 81. A. Tomita, K. Higashiyama, and Y. Tamai, Scanning electron microscopic study on the catalytic gasification of coal. Fuel, 1981. 60(2): p. 103-114  82. A. Oya, S. Yoshida, J. Alcaniz-Monge, and A. Linares-Solano, Formation of mesopores in phenolic resin-derived carbon fiber by catalytic activation using cobalt. Carbon, 1995. 33(8): p. 1085-1090. 83. Z.L. Liu, Licheng; Liu, Lang, Effect of CO2 activation on the formation of mesopore of pitch-based carbon sphere containing iron. Coal Conversion, 1999. 22(2): p. 71-75. 84. D. Cazorla-Amorós, D. Ribes-Pérez, M. Roman-Martinez, and A. Linares-Solano, Selective porosity development by calcium-catalyzed carbon gasification. Carbon, 1996. 34(7): p. 869-878. 85. R. Leboda, J. Skubiszewska-Zięba, and W. Grzegorczyk, Effect of calcium catalyst loading procedure on the porous structure of active carbon from plum stones modified in the steam gasification process. Carbon, 1998. 36(4): p. 417-425. 86. W. Shen, J. Zheng, Z. Qin, and J. Wang, Preparation of mesoporous carbon from commercial activated carbon with steam activation in the presence of cerium oxide. Journal of Colloid and Interface Science, 2003. 264(2): p. 467-473. 87. H. Yasuda, H. Tamai, M. Ikeuchi, and S. Kojima, Extremely large mesoporous carbon fibers synthesized by the addition of rare earth metal complexes and their unique adsorption behaviors. Advanced Materials, 1997. 9(1): p. 55-58  187  88. Y. Li, Z.-h. Huang, F.-y. Kang, and B.-h. Li, Preparation of activated carbon microspheres from phenolic resin with metal compounds by sub-and supercritical water activation. New carbon materials, 2010. 25(2): p. 109-113. 89. W. Shen, J. Zheng, Z. Qin, J. Wang, and Y. Liu, The effect of temperature on the mesopore development in commercial activated carbon by steam activation in the presence of yttrium and cerium oxides. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2003. 229(1): p. 55-61  90. R.T.K. Baker, R.D. Sherwood, and J.A. Dumesic, Catalytic hydrogenation of graphite by platinum, iridium, and platinum-iridium. Journal of Catalysis, 1980. 66(1): p. 56-64  91. C.W. Keep, S. Terry, and M. Wells, Studies of the nickel-catalyzed hydrogenation of graphite. Journal of Catalysis, 1980. 66(2): p. 451-462. 92. A. Tomita, N. Sato, and Y. Tamai, Hydrogenation of carbons catalyzed by nickel, platinum and rhodium. Carbon, 1974. 12(2): p. 143-149  93. P.J. Goethel and R.T. Yang, Mechanism of graphite hydrogenation catalyzed by nickel. Journal of Catalysis, 1987. 108(2): p. 356-363  94. P.J. Goethel and R.T. Yang, Mechanism of graphite hydrogenation catalyzed by ruthenium particles. Journal of Catalysis, 1988. 111(1): p. 220-226  95. P.J. Goethel and R.T. Yang, Platinum-catalyzed hydrogenation of graphite: Mechanism studied by the rates of monolayer channeling. Journal of Catalysis, 1986. 101(2): p. 342-351. 96. L. Stockman, Petroleum coke: The coal hiding in the tar sands, 2013, Oil Change International: Washington DC, USA. 188  97. E. Furimsky, Gasification of oil sand coke: Review. Fuel Processing Technology, 1998. 56(3): p. 263-290  98. J.G. Speight and B. Özüm, Petroleum refining processes2001: CRC Press. 99. M. Malekshahian and J.M. Hill, Effect of pyrolysis and CO2 gasification pressure on the surface area and pore size distribution of petroleum coke. Energy & Fuels, 2011. 25(11): p. 5250-5256  100. R. Alcántara, P. Lavela, G.F. Ortiz, J.L. Tirado, R. Menéndez, R. Santamarı́a, and J.M. Jiménez-Mateos, Electrochemical, textural and microstructural effects of mechanical grinding on graphitized petroleum coke for lithium and sodium batteries. Carbon, 2003. 41(15): p. 3003-3013. 101. J. Choi, Z.G. Barnard, S. Zhang, and J.M. Hill, Ni catalysts supported on activated carbon from petcoke and their activity for toluene hydrogenation. The Canadian Journal of Chemical Engineering, 2012. 90(3): p. 631-636. 102. R.K. Lattanzio, Petroleum coke: Industry and environmental issues, 2013. 103. V.M.L. Whiffen and K.J. Smith, Hydrodeoxygenation of 4-methylphenol over unsupported MoP, MoS2, and MoOx catalysts. Energy & Fuels, 2010. 24(9): p. 4728-4737  104. B.S. Gevert, J.E. Otterstedt, and F.E. Massoth, Kinetics of the HDO of methyl-substituted phenols. Applied Catalysis, 1987. 31(1): p. 119-131  105. H. Wang, J. Male, and Y. Wang, Recent advances in hydrotreating of pyrolysis bio-oil and its oxygen-containing model compounds. Acs Catalysis, 2013. 3(5): p. 1047-1070  106. P. Delporte, C. Pham-Huu, P. Vennegues, M.J. Ledoux, and J. Guille, Physical characterization of molybdenum oxycarbide catalyst; TEM, XRD and XPS. Catalysis today, 1995. 23(3): p. 251-267. 189  107. J.G. Choi and L.T. Thompson, XPS study of as-prepared and reduced molybdenum oxides. Applied Surface Science, 1996. 93(2): p. 143-149. 108. Y. Li, Y. Fan, J. He, B. Xu, H. Yang, J. Miao, and Y. Chen, Selective liquid hydrogenation of long chain linear alkadienes on molybdenum nitride and carbide modified by oxygen. Chemical Engineering Journal, 2004. 99(3): p. 213-218. 109. S. Izhar, H. Kanesugi, H. Tominaga, and M. Nagai, Cobalt molybdenum carbides: Surface properties and reactivity for methane decomposition. Applied Catalysis A: General, 2007. 317(1): p. 82-90  110. K. Oshikawa, M. Nagai, and S. Omi, Characterization of molybdenum carbides for methane reforming by TPR, XRD, and XPS. The Journal of Physical Chemistry B, 2001. 105(38): p. 9124-9131  111. P. Delporte, C. Pham-Huu, P. Vennegues, M.J. Ledoux, and J. Guille, Physical characterization of molybdenum oxycarbide catalyst; TEM, XRD and XPS. Catalysis today, 1995. 23(3): p. 251-267  112. V.M.L. Whiffen and K.J. Smith, A comparative study of 4-methylphenol hydrodeoxygenation over high surface area MoP and Ni2P. Topics in Catalysis, 2012. 55(14-15): p. 981-990  113. Y.Q. Yang, C.T. Tye, and K.J. Smith, Influence of MoS2 catalyst morphology on the hydrodeoxygenation of phenols. Catalysis Communications, 2008. 9(6): p. 1364-1368. 114. R. Wandas, J. Surygala, and E. Śliwka, Conversion of cresols and naphthalene in the hydroprocessing of three-component model mixtures simulating fast pyrolysis tars. Fuel, 1996. 75(6): p. 687-694. 190  115. C.R. Lee, J.S. Yoon, Y.-W. Suh, J.-W. Choi, J.-M. Ha, D.J. Suh, and Y.-K. Park, Catalytic roles of metals and supports on hydrodeoxygenation of lignin monomer guaiacol. Catalysis Communications, 2012. 17(0): p. 54-58. 116. H. Yoshida, S. Takeda, T. Uchiyama, H. Kohno, and Y. Homma, Atomic-scale in-situ observation of carbon nanotube growth from solid state iron carbide nanoparticles. Nano letters, 2008. 8(7): p. 2082-2086  117. A. Okita, Y. Suda, A. Ozeki, H. Sugawara, Y. Sakai, A. Oda, and J. Nakamura, Predicting the amount of carbon in carbon nanotubes grown by CH4 rf plasmas. Journal of applied physics, 2006. 99(1): p. 1-7. 118. A.W. Orbaek, A.C. Owens, and A.R. Barron, Increasing the efficiency of single walled carbon nanotube amplification by Fe–Co catalysts through the optimization of CH4/H2 partial pressures. Nano letters, 2011. 11(7): p. 2871-2874  119. F. Massoth, P. Politzer, M. Concha, J. Murray, J. Jakowski, and J. Simons, Catalytic hydrodeoxygenation of methyl-substituted phenols: Correlations of kinetic parameters with molecular properties. The Journal of Physical Chemistry B, 2006. 110(29): p. 14283-14291. 120. N. Liu, S.A. Rykov, and J.G. Chen, A comparative surface science study of carbide and oxycarbide: the effect of oxygen modification on the surface reactivity of C/W(1 1 1). Surface Science, 2001. 487(1–3): p. 107-117. 121. M.M. Sullivan, J.T. Held, and A. Bhan, Structure and site evolution of molybdenum carbide catalysts upon exposure to oxygen. Journal of Catalysis, 2015. 326: p. 82-91. 122. S. Badoga, R.V. Sharma, A.K. Dalai, and J. Adjaye, Synthesis and characterization of mesoporous aluminas with different pore sizes: Application in NiMo supported catalyst for hydrotreating of heavy gas oil. Applied Catalysis A: General, 2015. 489: p. 86-97  191  123. Y. Sun, H. Wang, and R. Prins, Hydrodesulfurization with classic Co–MoS2 and Ni–MoS2/γ-Al2O3 and new Pt–Pd on mesoporous zeolite catalysts. Catalysis Today, 2010. 150(3): p. 213-217  124. P.A. Aegerter, W.W.C. Quigley, G.J. Simpson, D.D. Ziegler, J.W. Logan, K.R. McCrea, S. Glazier, and M.E. Bussell, Thiophene hydrodesulfurization over alumina-supported molybdenum carbide and nitride catalysts: adsorption sites, catalytic activities, and nature of the active surface. Journal of Catalysis, 1996. 164(1): p. 109-121. 125. B. Dhandapani, T.S. Clair, and S.T. Oyama, Simultaneous hydrodesulfurization, hydrodeoxygenation, and hydrogenation with molybdenum carbide. Applied Catalysis A: General, 1998. 168(2): p. 219-228  126. E. Puello-Polo, M. Ayala-G, and J.L. Brito, Sulfidability and thiophene hydrodesulfurization activity of supported NiMo carbides. Catalysis Communications, 2014. 53: p. 9-14  127. G. Jin, J. Zhu, X. Fan, G. Sun, and J. Gao, Effect of Ni Promoter on Dibenzothiophene Hydrodesulfurization Performance of Molybdenum Carbide Catalyst. Chinese Journal of Catalysis, 2006. 27(10): p. 899-903. 128. H. Wang, S. Liu, and K.J. Smith, Synthesis and Hydrodeoxygenation Activity of Carbon Supported Molybdenum Carbide and Oxycarbide Catalysts. Energy & Fuels, 2016. 30(7): p. 6039-6049  129. X. Liu, A. Wang, X. Wang, C.-Y. Mou, and T. Zhang, Au–Cu Alloy nanoparticles confined in SBA-15 as a highly efficient catalyst for CO oxidation. Chemical Communications, 2008(27): p. 3187-3189. 192  130. C. Wan and B.M. Leonard, Iron-doped molybdenum carbide catalyst with high activity and stability for the hydrogen evolution reaction. Chemistry of Materials, 2015. 27(12): p. 4281-4288  131. V. Sundaramurthy, A.K. Dalai, and J. Adjaye, HDN and HDS of different gas oils derived from Athabasca bitumen over phosphorus-doped NiMo/γ-Al2O3 carbides. Applied Catalysis B: Environmental, 2006. 68(1–2): p. 38-48. 132. G.H. Singhal, R.L. Espino, J.E. Sobel, and G.A. Huff, Hydrodesulfurization of sulfur heterocyclic compounds: kinetics of dibenzothiophene. Journal of Catalysis, 1981. 67(2): p. 457-468  133. A.A. Smirnov, S.A. Khromova, D.Y. Ermakov, O.A. Bulavchenko, A.A. Saraev, P.V. Aleksandrov, V.V. Kaichev, and V.A. Yakovlev, The composition of Ni-Mo phases obtained by NiMoOx-SiO2 reduction and their catalytic properties in anisole hydrogenation. Applied Catalysis A: General, 2016. 514: p. 224-234  134. E. Schachtl, E. Kondratieva, O.Y. Gutiérrez, and J.A. Lercher, Pathways for H2 Activation on (Ni)-MoS2 Catalysts. The journal of physical chemistry letters, 2015. 6(15): p. 2929-2932. 135. P.A. Nikulshin, D.I. Ishutenko, A.A. Mozhaev, K.I. Maslakov, and A.A. Pimerzin, Effects of composition and morphology of active phase of CoMo/Al2O3 catalysts prepared using Co2Mo10–heteropolyacid and chelating agents on their catalytic properties in HDS and HYD reactions. Journal of Catalysis, 2014. 312: p. 152-169. 136. L. Zhang, J. Feng, Q. Chu, W. Li, K. Xu, and T.S. Wiltowski, Effect of potassium on the catalytic performance of Ni2Mo3N catalyst during hydrogenation of thiophene-containing benzene. Catalysis Communications, 2015. 66: p. 50-54. 193  137. J. Laine, M. Labady, F. Severino, and S. Yunes, Sink effect in activated carbon-supported hydrodesulfurization catalysts. Journal of Catalysis, 1997. 166(2): p. 384-387  138. Z. Yao, J. Jiang, Y. Zhao, F. Luan, J. Zhu, Y. Shi, H. Gao, and H. Wang, Insights into the deactivation mechanism of metal carbide catalysts for dry reforming of methane via comparison of nickel-modified molybdenum and tungsten carbides. RSC Advances, 2016. 6(24): p. 19944-19951. 139. M. Daage and R.R. Chianelli, Structure-function relations in molybdenum sulfide catalysts: the" rim-edge" model. Journal of Catalysis, 1994. 149(2): p. 414-427  140. T. Kawano, M. Kubota, M.S. Onyango, F. Watanabe, and H. Matsuda, Preparation of activated carbon from petroleum coke by KOH chemical activation for adsorption heat pump. Applied Thermal Engineering, 2008. 28(8-9): p. 865-871. 141. L.D. Virla, V. Montes, J. Wu, S.F. Ketep, and J.M. Hill, Synthesis of porous carbon from petroleum coke using steam, potassium and sodium: Combining treatments to create mesoporosity. Microporous and Mesoporous Materials, 2016. 234: p. 239-247. 142. V. Montes and J.M. Hill, Pore enlargement of carbonaceous materials by metal oxide catalysts in the presence of steam: Influence of metal oxide size and porosity of starting material. Microporous and Mesoporous Materials 2017. 256: p. 91-101. 143. X. Li, D. Ma, L. Chen, and X. Bao, Fabrication of molybdenum carbide catalysts over multi-walled carbon nanotubes by carbothermal hydrogen reduction. Catalysis letters, 2007. 116(1-2): p. 63-69. 144. O. Ostrovski and G. Zhang, Reduction and carburization of metal oxides by methane‐containing gas. AIChE journal, 2006. 52(1): p. 300-310  194  145. I. Wolf and H.J. Grabke, A study on the solubility and distribution of carbon in oxides. Solid state communications, 1985. 54(1): p. 5-10  146. R.G. Kaldenhoven and J.M. Hill, Determining the pore structure of activated carbon by nitrogen gas adsorption, in Catalysis2018. p. 41-63. 147. G. Kresse and J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Computational Materials Science, 1996. 6(1): p. 15-50. 148. P.E. Blöchl, Projector augmented-wave method. Physical review B, 1994. 50(24): p. 17953. 149. G. Kresse and J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Physical Review B, 1996. 54(16): p. 11169-11186. 150. J.P. Perdew, K. Burke, and M. Ernzerhof, Generalized gradient approximation made simple. Physical review letters, 1996. 77(18): p. 3865. 151. B. Liu, Z. Zhao, G. Henkelman, and W. Song, Computational design of a CeO2-supported Pd-based bimetallic nanorod for CO oxidation. The Journal of Physical Chemistry C, 2016. 120(10): p. 5557-5564  152. Z.M. Hanafi, M.A. Khilla, and M.H. Askar, The thermal decomposition of ammonium heptamolybdate. Thermochimica Acta, 1981. 45(3): p. 221-232. 153. C. Thomazeau, V. Martin, and P. Afanasiev, Effect of support on the thermal decomposition of (NH4)6Mo7O24·4H2O in the inert gas atmosphere. Applied Catalysis A: General, 2000. 199(1): p. 61-72  195  154. J. Dang, G.-H. Zhang, and K.-C. Chou, Phase transitions and morphology evolutions during hydrogen reduction of MoO3 to MoO2. High Temperature Materials and Processes, 2014. 33(4): p. 305-312  155. E. Lalik, W.I.F. David, P. Barnes, and J.F.C. Turner, Mechanisms of reduction of MoO3 to MoO2 reconciled ? The Journal of Physical Chemistry B, 2001. 105(38): p. 9153-9156  156. M. Chen, C.M. Friend, and E. Kaxiras, The chemical nature of surface point defects on MoO3 (010): adsorption of hydrogen and methyl. Journal of the American Chemical Society, 2001. 123(10): p. 2224-2230  157. J.A. Rodriguez, J.C. Hanson, A.I. Frenkel, J.Y. Kim, and M. Pérez, Experimental and theoretical studies on the reaction of H2 with NiO: role of O vacancies and mechanism for oxide reduction. Journal of the American Chemical Society, 2002. 124(2): p. 346-354. 158. E. Furimsky, Metal carbides and nitrides as potential catalysts for hydroprocessing. Applied Catalysis A: General, 2003. 240(1): p. 1-28  159. V. Sundaramurthy, A.K. Dalai, and J. Adjaye, Effect of phosphorus addition on the hydrotreating activity of NiMo/Al2O3 carbide catalyst. Catalysis today, 2007. 125(3): p. 239-247  160. H. Wang, S. Liu, R. Govindarajan, and K.J. Smith, Preparation of Ni-Mo2C/carbon catalysts and their stability in the HDS of dibenzothiophene. Applied Catalysis A: General, 2017. 539(5): p. 114-127. 161. T.-C. Xiao, H.-T. Wang, A.P. York, and M.L. Green, Effect of sulfur on the performance of molybdenum carbide catalysts for the partial oxidation of methane to synthesis gas. Catalysis letters, 2002. 83(3-4): p. 241-246. 196  162. X.-R. Shi, S.-G. Wang, H. Wang, C.-M. Deng, Z. Qin, and J. Wang, Structure and stability of β-Mo2C bulk and surfaces: A density functional theory study. Surface Science, 2009. 603(6): p. 852-859  163. J.W. Han, L. Li, and D.S. Sholl, Density functional theory study of H and CO adsorption on alkali-promoted Mo2C surfaces. The Journal of Physical Chemistry C, 2011. 115(14): p. 6870-6876  164. T. Wang, Y.-W. Li, J. Wang, M. Beller, and H. Jiao, Dissociative hydrogen adsorption on the hexagonal Mo2C phase at high coverage. The Journal of Physical Chemistry C, 2014. 118(15): p. 8079-8089  165. K. KuO and G. HÄGg, A New Molybdenum Carbide. Nature, 1952. 170: p. 245. 166. D. Loffreda, F. Delbecq, and P. Sautet, Adsorption thermodynamics of acrolein on Pt (1 1 1) in realistic temperature and pressure from first-principle calculations. Chemical physics letters, 2005. 405(4-6): p. 434-439. 167. H. Wang, S. Liu, B. Liu, V. Montes, J.M. Hill, and K.J. Smith, Carbon and Mo transformations during the synthesis of mesoporous Mo2C/carbon catalysts by carbothermal hydrogen reduction. Journal of Solid State Chemistry, 2018. 258: p. 818-824. 168. D.R. Stellwagen and J.H. Bitter, Structure–performance relations of molybdenum-and tungsten carbide catalysts for deoxygenation. Green Chemistry, 2015. 17(1): p. 582-593. 169. Y. Qin, L. He, J. Duan, P. Chen, H. Lou, X. Zheng, and H. Hong, Carbon-supported molybdenum-based catalysts for the hydrodeoxygenation of maize oil. ChemCatChem, 2014. 6(9): p. 2698-2705. 170. J.C. Rivière and S. Myhra, Handbook of surface and interface analysis: methods for problem-solving2009: CRC press. 197  171. C. Tang, W. Wang, A. Sun, C. Qi, D. Zhang, Z. Wu, and D. Wang, Sulfur-Decorated Molybdenum Carbide Catalysts for Enhanced Hydrogen Evolution. ACS Catalysis, 2015. 5(11): p. 6956-6963. 172. Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong, and H. Dai, MoS2 Nanoparticles Grown on Graphene: An Advanced Catalyst for the Hydrogen Evolution Reaction. Journal of the American Chemical Society, 2011. 133(19): p. 7296-7299. 173. NIST Chemistry WebBook, NIST Standard Reference Database Number 69, ed. P.J. Linstrom and W.G. Mallard2005: National Institute of Standards and Technology. 174. P. Liu, J.A. Rodriguez, and J.T. Muckerman, Sulfur adsorption and sulfidation of transition metal carbides as hydrotreating catalysts. Journal of Molecular Catalysis A: Chemical, 2005. 239(1-2): p. 116-124. 175. D. Wang, Z. Wang, C. Wang, P. Zhou, Z. Wu, and Z. Liu, Distorted MoS2 nanostructures: An efficient catalyst for the electrochemical hydrogen evolution reaction. Electrochemistry Communications, 2013. 34: p. 219-222. 176. S. Cristol, J.-F. Paul, E. Payen, D. Bougeard, F. Hutschka, and S. Clémendot, DBT derivatives adsorption over molybdenum sulfide catalysts: a theoretical study. Journal of Catalysis, 2004. 224(1): p. 138-147  177. M. Sun, A.E. Nelson, and J. Adjaye, Ab initio DFT study of hydrogen dissociation on MoS2, NiMoS, and CoMoS: mechanism, kinetics, and vibrational frequencies. Journal of Catalysis, 2005. 233(2): p. 411-421  178. B. Jiang, X. Hu, S. Lin, D. Xie, and H. Guo, Six-dimensional quantum dynamics of dissociative chemisorption of H2 on Co(0001) on an accurate global potential energy surface. Phys Chem Chem Phys, 2015. 17(36): p. 23346-23355. 198  179. A. Szymańska-Kolasa, M. Lewandowski, C. Sayag, D. Brodzki, and G. Djéga-Mariadassou, Comparison between tungsten carbide and molybdenum carbide for the hydrodenitrogenation of carbazole. Catalysis today, 2007. 119(1): p. 35-38. 180. I.I. Abu and K.J. Smith, HDN and HDS of model compounds and light gas oil derived from Athabasca bitumen using supported metal phosphide catalysts. Applied Catalysis A: General, 2007. 328(1): p. 58-67. 181. T. Iida, M. Shetty, K. Murugappan, Z. Wang, K. Ohara, T. Wakihara, and Y. Román-Leshkov, Encapsulation of Molybdenum Carbide Nanoclusters inside Zeolite Micropores Enables Synergistic Bifunctional Catalysis for Anisole Hydrodeoxygenation. ACS Catalysis, 2017. 7(12): p. 8147-8151. 182. H. Rezaei and K.J. Smith, Catalyst deactivation in slurry-phase residue hydroconversion. Energy & Fuels, 2013. 27(10): p. 6087-6097. 183. A. Calafat, J. Laine, A. López-Agudo, and J.M. Palacios, Effect of Surface Oxidation of the Support on the Thiophene Hydrodesulfurization Activity of Mo, Ni, and NiMo Catalysts Supported on Activated Carbon. Journal of Catalysis, 1996. 162(1): p. 20-30. 184. J. Matos, J.L. Brito, and J. Laine, Activated carbon supported Ni-Mo: effects of pretreatment and composition on catalyst reducibility and on ethylene conversion. Applied Catalysis A: General, 1997. 152(1): p. 27-42  185. S.P.A. Louwers and R. Prins, Ni EXAFS studies of the Ni-Mo-S structure in carbon-supported and alumina-supported Ni-Mo catalysts. Journal of Catalysis, 1992. 133(1): p. 94-111. 199  186. K.S.W. Sing, Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity (Recommendations 1984). Pure and applied chemistry, 1985. 57(4): p. 603-619  187. J.M.D. Tascón, Novel carbon adsorbents2012: Elsevier. 188. N.A. Seaton, J.P.R.B. Walton, and N. quirke, A new analysis method for the determination of the pore size distribution of porous carbons from nitrogen adsorption measurements. Carbon, 1989. 27(6): p. 853-861. 189. C. Lastoskie, K.E. Gubbins, and N. Quirke, Pore size distribution analysis of microporous carbons: a density functional theory approach. The journal of physical chemistry, 1993. 97(18): p. 4786-4796. 190. J. Jagiello and J.P. Olivier, 2D-NLDFT adsorption models for carbon slit-shaped pores with surface energetical heterogeneity and geometrical corrugation. Carbon, 2013. 55: p. 70-80. 191. J. Jagiello and J.P. Olivier, Carbon slit pore model incorporating surface energetical heterogeneity and geometrical corrugation. Adsorption, 2013. 19(2-4): p. 777-783. 192. A. Hayter, Probability and statistics for engineers and scientists2012: Nelson Education. 193. H.S. Fogler, Elements of chemical reaction engineering1999. 194. R.B. Bird, W.E. Stewart, and E.N. Lightfoot, Transport Phenomena, 2nd Edition2002: New York: Wiley. 195. M.C. Tsai, Y.W. Chen, B.C. Kang, J.C. Wu, and L.J. Leu, Hydrodesulfurization and hydrodemetalation reactions of residue oils over cobalt-molybdenum/aluminum borate catalysts in a trickle bed reactor. Industrial & engineering chemistry research, 1991. 30(8): p. 1801-1810  200  196. V. Whiffen, A study of metal phosphides for the hydrodeoxygenation of phenols and pyrolysis oil, 2013, University of British Columbia.    201  Appendices      202  Appendix A  Catalyst Preparation A.1 Raw Petroleum Coke (PC) and Activated Petcoke (APC) The petroleum coke used in this study was provided by Suncor Energy Inc. (Calgary, AB, Canada) generated in a delayed coking process. Several characterization methods have been used to determine the properties of raw petcoke, as listed below.  Elemental analysis Table A.1: CHNS/O wt% analysis of raw petroleum coke and activated petroleum coke. Sample C H N S O O/C (at.) Raw petcoke 83.27±0.01 3.60±0.01 2.02±0.05 6.58±0.00 4.53±0.06 0.04  APC_800 a 89.87±0.00 0.00±0.00 0.67±0.02 0.00±0.00 9.46±0.03 0.08  Note: a. It was prepared by activating raw petcoke with KOH in a mass ratio of 1:3 at 800 oC for 2h.   XRD  Figure A.1: XRD scan of raw petroleum coke and activated petroleum coke. 203   SEM-EDX       Figure A.2: SEM graphs of raw petroleum coke (Left) and activated petroleum coke (Right).  Table A.2: EDX analysis of raw petroleum coke (PC) and activated petroleum coke (APC).  Elements (wt%) Sample C O S Al Si PC-Raw Petroleum Coke Ave. 89.96 5.91 3.58 0.20 0.28 Std. Dev. 0.57 0.43 0.66 0.03 0.07 APC-Activated Petroleum Coke at 800 oC (APC_800) Ave. 93.04 6.78 0.00 0.18 0.00 Std. Dev. 0.61 0.47 0.00 0.05 0.00  204   Raman Spectroscopy  Figure A.3: Raman spectroscopy of raw petroleum coke (PC) and activated petroleum coke (APC_800).  The peaks at 1360 and 1590 cm-1 are characteristic peaks of carbon materials, represented as the D band and G band, respectively. The ratio of the relative intensity of ID/IG can clearly present the degree of disorder. Figure A.3 shows that ID/IG = 1.62 for raw petroleum coke which is lower than that of APC_800 sample. It suggests that the disorder of the carbon material was increased during activation due to the formation of pores during activation. This is consistent with the XRD results.  A.2 Mo2C/C based Catalyst Precursors and Catalysts Preparation (I) Preparation of Mo2C/C catalyst precursors: sample calculations for the required chemicals for preparing 10 wt% Mo on Carbon (carbon refers to AC or APC; the catalyst is designated as 10 wt.%Mo/C) are as follows: 1. Required amount of 10 wt.%Mo/C catalyst = 10.0 g 205  2. Amount of Mo in 10 wt.%Mo/C catalyst =𝟏𝟎 𝐠 𝐌𝐨𝟏𝟎𝟎 𝐠 𝐌𝐨/𝐂𝐚𝐫𝐛𝐨𝐧× 𝟏𝟎𝐠 𝐌𝐨𝐂𝐚𝐫𝐛𝐨𝐧= 𝟏. 𝟎 𝐠       (Eq. A-1) 3. Required amount of ammonium heptamolybdate (AHM) =𝟏𝟐𝟑𝟓.𝟖𝟔𝐠𝐦𝐨𝐥𝐀𝐇𝐌/𝟕𝟗𝟓.𝟗𝟒 𝐠 𝐌𝐨× 𝟏. 𝟎 𝐠 𝐌𝐨 = 𝟏. 𝟖𝟒 𝐠       (Eq. A-2) 4. Required amount of Carbon = 𝟏𝟎. 𝟎 𝐠 (𝟏𝟎𝐰𝐭. %𝐌𝐨/𝐂) − 𝟏. 𝟎 𝐠 𝐌𝐨 = 𝟗 𝐠        (Eq.A-3) 5. Required amount of acetone to DI water solution with a mass ratio of 1:9 = 𝟓. 𝟎𝐠𝐠 𝐂𝐚𝐫𝐛𝐨𝐧∗ 𝟗 𝐠 𝐂𝐚𝐫𝐛𝐨𝐧 = 𝟒𝟓. 𝟎 𝐠        (Eq.A-4)  (II) Preparation of Ni-Mo2C/C catalyst precursors The calculation of required chemicals for 1 wt% Ni and 10 wt% of Mo on Carbon (designated as 1 wt.%Ni-10 wt%Mo/C) are as follows: 1. Required amount of 1wt.%Ni-10 wt.%Mo/C catalyst = 10.0 g 2. Amount of Mo is 1.0 g as calculated by Eq. A-1. 3. Amount of Ni is calculated by  =𝟏 𝐠 𝐍𝐢𝟏𝟎𝟎 𝐠 𝐦𝐞𝐭𝐚𝐥𝐂𝐚𝐫𝐛𝐨𝐧× 𝟏𝟎 𝐠 𝐦𝐞𝐭𝐚𝐥𝐂𝐚𝐫𝐛𝐨𝐧= 𝟎. 𝟏 𝐠        (Eq. A-5) 4. Required amount of Carbon  = 𝟏𝟎. 𝟎 𝐠 𝐌𝐞𝐭𝐚𝐥𝐂𝐚𝐫𝐛𝐨𝐧− 𝟏. 𝟎 𝐠 𝐌𝐨 − 𝟎. 𝟏 𝐠 𝐍𝐢 = 8.9 g      (Eq. A-6) 5. A successive impregnation method is used for bimetallic catalyst preparation [183-185]. The catalyst is prepared by first impregnated with Mo and then by Ni.  6. Required amount of AHM in the 1st step of impregnation =𝟏𝟐𝟑𝟓.𝟖𝟔𝐠𝐦𝐨𝐥𝐀𝐇𝐌/𝟕𝟗𝟓.𝟗𝟒 𝐠 𝐌𝐨× 𝟏. 𝟎 𝐠 𝐌𝐨 = 𝟏. 𝟖𝟒 𝐠       (Eq. A-7) 206  Note: The obtained precursor is designated as 10AHM/C.  7. Apply a drying process for 10AHM/C precursor before 2nd step impregnation. According to Hanafi et al.’s study [152], it knows that AHM could decompose in air by losing H2O and NH3 at 110 oC as shown by the following formula:  3(NH4)2O • 7MoO3 • 4H2O (M1=1235.86 g/mol)  (NH4)2O • 2.5MoO3 (M2=417.85 g/mol) 8. Calculate the weight of obtained and dried 10AHM/C precursor =𝟏.𝟖𝟒 𝐠𝐌𝟏∗𝟕.𝟎𝟐.𝟓× 𝐌𝟐 + 𝟖. 𝟗 𝐠 = 1.74 g + 8.9 g =10.64 g      (Eq. A-8) Note: this calculated mass is designated as m2. 9. Take m1 (g) of step 8 prepared precursor and apply for the 2nd step impregnation 10. Calculate the weight of carbon contained in m1 (g) 10AHM/C precursor =  𝐦𝟏𝐦𝟐× 𝟖. 𝟗 𝐠          (Eq. A-9) Note: this calculated mass is recorded as m3. 11. Required amount of Ni, recorded as m4 =  𝐦𝟑 (𝐠)𝟖.𝟗 𝐠∗ 𝟎. 𝟏 𝐠           (Eq. A-10) 12. Required amount of Ni(NO3)2·6H2O =  𝐦𝟒 (𝐠)𝟓𝟖.𝟔𝟗 𝐠/𝐦𝐨𝐥∗ 𝟐𝟗𝟎. 𝟕𝟗 𝐠/𝐦𝐨𝐥        (Eq. A-11)  (III) Carbothermal hydrogen reduction (CHR) 1. Place 0.9 g catalyst precursor in the U-tube reactor, as shown in Figure A.4. 207   Figure A.4: Schematic illustration of the preparation of Mo2C/C catalysts in quartz U-tube.  2. Apply a continuous H2 flow of 100 mL (STP)/min through the catalyst bed from bottom to top.  3. Set up a temperature program from room temperature to the target temperature by 1 oC/min, holding at the final temperature for 90 min.  Note: the temperature program may vary based on different requirements.  4.  After CHR, the catalyst was quickly quenched to room temperature in N2 flow.  5. Immerse the tube into water bath for a further cooling down when temperature < 50 oC. 6. Apply a passivation process by 1%O2/N2 at 15 mL (STP)/min for 2 h.  A.3 Methodology of Petroleum Coke Activation In general, the activation of petroleum coke can be grouped into three parts: I) Preparation of petcoke powder 208  1) Add raw petcoke into the mortar  2) Crush the petcoke slowly using the pestle until all pieces become smaller 3) Use the sieve tray to sieve the petcoke into the desired particle diameter of 90-180 um  4) Repeat the process by returning the larger diameter petcoke to the mortar for further grinding to produce finer particles II) Activation  5) Apply drying process to remove the contained moisture from raw petcoke 6) Weigh x* amount of prepared petcoke powder  7) Weigh y* amount of KOH pellets, the mass ratio of y*to x* was set at ~3.0 (An exact value z* can be obtained based on y*to x* ratio) 8) Add the KOH pellets in the mortar and crush it until it turns into a fine powder 9) Add the petcoke powder into the mortar and mix the two powders together 10) Place a clean ceramic boat on the scale, tare the balance and then add the mixture from step 9) into the boat 11) Record the number from step 10) as quickly as possible to avoid the moisture adsorption from the air 12) Calculated the real amount of petcoke (recorded as m1) from the ratio z* and the actual mixture mass in ceramic boat  13) Place the ceramic boat to the center of the tubular furnace and then place thermos blocks on both ends of the tube (See illustration: Figure A.1) 14) Secure the ends of the ceramic tube by using the metal connections 15) Open the N2 valve and purge the tube for several minutes prior heating up 209  16) Set the temperature, activation time and ramping rate on the controller and then start the program 17) After activation, a natural cooling down is proceed under N2 before taking the sample out   Figure A.5: Schematic illustration of the tubular furnace for petroleum coke activation.  III) Washing activated petroleum coke 18) Remove the activated petcoke sample into the filter funnel with DI water  19) Use DI and 1 M HCl solution to wash the samples in turn for at least 3 times until the pH=7 20) Dry the sample in the oven at 110 oC overnight  21) Weigh the obtained activated petcoke on scale (recorded as m2) and then a burn-off rate (%) can be calculated by the following equation: Burnoff rate % =𝑚1 − 𝑚2𝑚1   210  Appendix B  Catalyst Characterization B.1 Physical Properties Test  Two models have been listed here for the study of pore structure: (1) Brunauer, Emmett and Teller (BET) method The BET theory applies to multilayer adsorption systems. The isotherm measurement is used to quantify the specific surface area of the testing materials at a constant temperature. The standard BET test is often operated in N2 at boiling point temperature of -196 oC and the relative equation can be expressed as follows: 𝑷/𝑷𝒐𝑽(𝟏−𝒑𝒑𝒐⁄ )=𝒄−𝟏𝑽𝒎𝑪(𝑷𝑷𝒐) +𝟏𝑽𝒎𝑪        (Eq. B-1) where C is the BET constant, representing the attraction between the adsorbate and adsorbent, Vm is the monolayer coverage, V is the adsorbed gas quantity, P is the equilibrium pressure of the adsorbate, and Po is the saturation pressure of the adsorbate. A plot of x=P/Po vs. y=𝑃/𝑃𝑜𝑉(1−𝑝 𝑝𝑜⁄ ) can generate a linear curve with a slop of 𝐶−1𝑉𝑚𝐶 and an intercept of 1𝑉𝑚𝐶. Thus, Vm and C can be obtained.   Furthermore, the BET surface area can be calculated by 𝑆𝐵𝐸𝑇 = 4.35𝑉𝑚 for N2 isotherms. The valid linear range of this measurement is between P/Po = 0.05 ~ 0.35. In the meanwhile, the shape of the isotherm can be used to determine the adsorption process and generally it is classified into six types as shown in Figure B1.   211   Figure B.1: Gas physisorption isotherms. (Reprinted with permission from [186])  (II) Non-local density functional theory (NLDFT) The surface area determined from BET equation is often being overestimated, especially for microporous materials, based on Bottani and Tascon’s study [187]. Thus, it was considered worthwhile to assess several NLDFT models in this study.   The Density Functional Theory model was firstly developed by Seaton et al. [188] combining micropore filling and capillary condensation. After that, it has been improved by Lastoskie et al. [189] through a non-local mean field theory (NLDFT), mainly for slit-like porous activated carbons. Furthermore, it has been modified into a 2-D Non-Local Density Functional Theory with finite pores (2D-NLDFT) and 2-D Non-Local Density Functional Theory with heterogeneous surfaces (2D-NLDFT-HS). The DFT models can be applied to accurately determine the surface area and pore size distribution of microporous carbon materials.   212  Example for data processing: Sample: activated petroleum coke at 800 oC (APC_800) Method: Carbon-N2, 2D NLDFT-HS [190, 191]  Table B.1: Isotherm data both from experimentally measured and 2D-NLDFT-HS model fitted. P/Po Experimental adsorbed amount (cm3/g) Fitted adsorbed amount (cm3/g) Difference between experimental and fitted data (cm3/g) 0.009902 461.39811 457.55926 3.838851879 0.019757 496.70208 492.94613 3.755953429 0.030420 523.66742 530.92839 -7.260961022 0.046162 554.46160 559.68722 -5.225615080 0.058677 574.21795 580.89948 -6.681529288 0.081822 603.23738 598.90861 4.328762393 0.101198 621.79555 620.02272 1.772833683 0.122760 638.01360 633.90901 4.104588236 0.144364 650.54058 646.90584 3.634732701 0.166597 660.56502 657.41622 3.148807393 0.189013 668.38688 667.76454 0.622341025 0.209578 674.04333 674.69169 -0.648361622 0.253308 682.81875 684.75733 -1.938574522 0.306232 689.78627 692.94519 -3.158917368 Ave. std. dev. of the fit 4.0525 Note: The specific surface area was calculated from the measured N2 adsorption isotherm using 2D-NLDFT-HS applied in the P/Po range of 0.01~0.30.  213  The calculated surface area is 2028 m2/g, which is smaller than the BET method calculated (2097 m2/g; calculated in the range of P/Po = 0.05 ~ 0.30).  Figure B.2: Fitted isotherm by 2D-NLDFT-HS model. (“o” represents measured experimental data)   Figure B.3: Cumulative surface area by 2D-NLDFT-HS model. 214   B.2 XRD Power X-ray diffraction (XRD) is used to determine the bulk properties of the samples. From this measurement, several pieces of information can be obtained: (I) Identify the bulk phases present in the catalyst by Bragg’s law (usually the required conc. of measured components is ≥ 5 wt%; as shown in Eq. (B-2)); (II) Estimate the crystallite or grain size of the particles by Scherrer equation (Eq. (B-3)); (III) Determine the lattice parameters of the crystals.  Bragg’s law: 𝒅 =𝒏𝝀𝟐𝒔𝒊𝒏𝜽         (Eq. B-2) where n is the integer, λ is the wavelength of the incident X-ray beam, θ is the angles of incidence, and d is the distance between atomic layers in a crystal.   Scherrer equation: 𝑫 =𝒌𝝀𝑩𝒄𝒐𝒔𝜽        (Eq. B-3) where k is the dimensionless shape factor (it varies with the actual shape of the crystallite, typically 0.9 or 1), λ is the X-ray wavelength, B is the line broadening at half the maximum intensity (FWHM) in radians (This number can be obtained after subtracting the instrumental line broadening), θ is the angle of incidence, and D is the mean crystallize size.  For example, the β-Mo2C we obtained in this research has a hexagonal crystal structure, where a=b≠c, α=γ=90o, β=120o. The relative correlation is shown in Eq. B-4. 𝟏𝒅𝒉𝒌𝒍𝟐 =𝟒𝟑(𝒉𝟐+𝒉𝒌+𝒌𝟐)𝒂𝟐+𝟏𝒄𝟐         (Eq. B-4)  where (h, k, l) is the Miller index, d is the interplanar lattice spacing of the (h, k, l) plane, a, b, and c are lattice parameters.  215   An example of particle size calculation by Scherrer equation: Sample: 10Mo2C/AC_750 Fitting method: Gaussian fit by the equation (Eq. B-5) below.         (Eq. B-5)   Figure B.4: XRD profile of 10Mo2C/APC_750 catalyst and Gaussian curve fitting for (002) (Left peak) and (101) (Right peak) planes.   216   The calculated result for (101) plane is as follows: 10Mo2C/AC_750 (101) plane Parameter Value Error (±) yo (PSD) 94.82 19.19 xc (2θ) 46.14 0.077 w (2θ) 0.88 0.019 FWHM (2θ) 1.04 0.022   D (nm) 19.33 0.41 Note: λ=0.17902 nm (for Co source); k=0.9; 1o = 0.0175 radian.  217  B.3 XPS X-ray photoelectron spectroscopy (XPS) works by irradiating the sample surface with a soft X-ray. If the binding energy (Eb) of the sample atoms is lower than the X-ray (hν), it will be excited and emitted from the parent atom as photoelectron (Ek). It is a surface technique since only the photoelectrons from the outer surface (1 ~ 10 nm) has the capability to escape. XPS is usually used to measure the elemental composition of the surface and the electronic state of each element in the surface. The calculation formula of binding energy (Eb) and atomic concentration of an element (Ci) are given be the following equations: 𝑬𝒃 = 𝒉𝝂 −  𝑬𝒌 − ∅         (Eq. B-6) 𝑪𝒊 =𝑰𝒊/𝑺𝒊∑ 𝑰𝒊/𝑺𝒊𝒊          (Eq. B-7) Where h is the X-ray photo energy (for monochromatic Mg kα, h = 1253.6 eV); Ek is the kinetic energy of the photoelectron, and ∅ is the work function from the instrument; Ii is the peak intensity of element i, and Si is the sensitivity factor for the peak i.  In the present study, there is no charging effect due to the good conductivity of the carbon support. The C 1s peak located at 284.5 eV is used as the internal reference peak for all elements analysis. Both elements quantification, survey scan, and narrow scan were conducted (See Figure B.5 and B.6). All the experimental data were proceed by Vision Processing and XPSPEAKER 41 softwares. 218   Figure B.5: XPS survey scan of used 1%Ni-10%Mo2C/APC catalyst.   Figure B.6: Quantification results of used 1%Ni-10%Mo2C/APC catalyst. (Note: the integration of each element was based on Figure B.2)219  B.4 CO Chemisorption  The CO uptake of the catalysts was carried on using a Micromeritics AutoChem II 2920 unit using pulsed chemisorption at 35 oC. Two types of catalysts were tested here: one is in-situ synthesized Mo2C catalyst with no passivation process; another one is passivated Mo2C catalyst with a TPR prior the CO uptake test. The calculation process is as follows. (1) In-situ synthesized Mo2C/C catalysts   Preparation of the in-situ synthesized catalyst: a. Synthesis the catalyst by 50 mL(STP)/min of 9.5 mol% H2/Ar while heating to the final temperature at 1C/min, and maintaining the final temperature for 90 min; b.  The obtained Mo2C/C was then flushed in He at 400 C for 4 h and then cooled to room temperature.  c. Next 0.5 mL pulses of CO were injected into a flow of He (50 mL(STP)/min) and the CO uptake was measured using a TCD. CO pulses were repeatedly injected until no further CO uptake was observed after consecutive injections.  Obtain the result from Micromeritics instrument  Example: 10%Mo2C/AC-600 fresh catalyst synthesized in-situ 220   Figure B.7: TCD signal vs. time for in-situ synthesized 10% Mo2C/AC-600 catalyst at 32 oC.  Table B.2: Peak table of adsorbed CO on in-situ synthesized 10% Mo2C/AC-600 catalyst. Peak number Temperature (oC) Quantity Adsorbed (µmol/gcat) Cumulative Quantity (µmol/gcat) 1 32.5 6.78554 6.78554 2 32.2 4.64694 11.43248 3 32.5 2.96738 14.39986 4 32.4 1.97577 16.37563 5 32.2 0.34299 16.71862 6 32.3 0.00000 16.71862 7 32.4 0.00000 16.71862 So, the CO uptake is 16.7 µmol/gcat. 221   Calculate the CO uptake based on per gram of Mo a. 10 wt%Mo/AC precursor was used for 10% Mo2C/AC-600 catalyst synthesis, thus there is 0.1 𝑔𝑀𝑜𝑔𝑝𝑟𝑒𝑐𝑢𝑟𝑠𝑜𝑟; b. The calculated Mo amount in catalyst can be calculated by0.1𝑔𝑀𝑜𝑔𝑃𝑟𝑒𝑐𝑢𝑟𝑠𝑜𝑟(1−𝐵𝑢𝑟𝑛𝑜𝑓𝑓%); in this case, the burn-off rate is 11%, thus the result is 0.1𝑔𝑀𝑜𝑔𝑃𝑟𝑒𝑐𝑢𝑟𝑠𝑜𝑟(1−0.11)= 0.1124𝑔𝑀𝑜𝑔𝑐𝑎𝑡; c. Combined with the measured CO uptake, the adsorbed CO based on per gram of Mo is 16.7µ𝑚𝑜𝑙𝑔𝑐𝑎𝑡0.1124𝑔𝑀𝑜𝑔𝑐𝑎𝑡= 148 µ𝑚𝑜𝑙𝑔𝑀𝑜  (2) Passivated Mo2C/C catalyst Prior to the measurement, the sample was pretreated to remove the passivation layer by passing 50 mL(STP)/min of 9.5 mol% H2/Ar while heating to 400 C at 10 C/min, and maintaining the final temperature for 2 h. The flow was then switched to He (50 mL(STP)/min) at 400 C for 4 h in order to remove the adsorbed species. After cooling in a He flow, 0.5 mL pulses of CO were injected into a flow of He (50 mL(STP)/min) and the CO uptake was measured using a TCD as before. Here, passivated 10% Mo2C/AC-600 catalyst is put here as an example. 222  Table B.3: Peak table of adsorbed CO on reduced passivated 10% Mo2C/AC-600 catalyst. Peak number Temperature (oC) Quantity Adsorbed (µmol/gcat) Cumulative Quantity (µmol/gcat) 1* 33.7 2.95816 9.61747 2 33.5 0.01899 9.63646 3 33.4 0.00000 9.63646 4 33.3 0.37331 10.00977 5 33.4 0.00000 10.00977 6 33.1 0.44632 10.45609 7 33.2 0.00000 10.45609 8 33.1 0.18901 10.64510 9 33.2 0.00000 10.64510 * 1st dose was completely adsorbed with a quantity of 6.65931 µmol/g.  Figure B.8: TCD signal vs. time for reduced passivated 10% Mo2C/AC-600 catalyst at 33 oC.223  B.5 GC-FID/TCD  The GC used in the present study was a Shimadzu GC-14B equipped with C-R8A Chromatopac integrator. There are two columns and two detector equipped in this equipment. One is flame ionization detector (FID) with Agilent HP-PLOT U capillary column (19095P-UO4, ID: 0.530 mm, Length: 30 m, film: 20 µm); another one is thermal conductivity detector (TCD) with a packed column (5682PC, OD: 3.175 mm, Length: 0.154 m, 316 stainless steel).  In Chapter 4’s study, the outlet of the U-tube is connected with the GC-FID/TCD inlet, thus an in-situ analysis can be applied. The parameters set for the analysis are as follows: For FID, three gases were used, which are H2, Air, and He, functionalized as combustion, oxidant, and carrier/make-up gas, representatively. The gas pressure has been set as: Carrier P = 40 kPa, Carrier M = 40 kPa, H2 = 60 kPa, and Air = 25 kPa. The temperature program used in this study is shown in Table B.2.  Table B.4: Temperature program used for GC-FID/TCD analysis.  Program rate (ºC/min) Temperature (ºC) Time (min) Initial × 35 3 1st step 5 120 2 2nd step 10 170 2  For TCD, only He gas is needed, where the pressure of Carrier P = 160 kPa and Carrier M = 100 kPa. The detector temperature is set at 200 oC.   A gas calibration was conducted prior to the in-situ analysis. The procedure is as follows: 224  1. Identify the mixture gas composition: Table B.5: The mole composition of the mixture gas.  2. Chose H2 as the diluting gas. 3. Apply gas calibration both for H2 and mixture gas, respectively (See H2 calibration as an example: Table B.6).  Table B.6: Gas calibration result for carrier gas H2.  Where x represents the Set point (S.P.) of MFC and y represents the actual flow rate. The linear correlation can be obtained as shown in Figure B.9. Calibration Gas Mol%Butane 4Carbon Dioxide 8Ethane 8Ethylene 2Hydrogen 25Isobutane 2Methane 10Nitrogen 10Propane 6Carbon Monoxide 25H2 calibration-- initial P:130 psiS.P. Ave. Time (s) Ave. time (min) cc/min10 14.84 14.65 14.68 14.88 14.85 14.66 14.76 0.2460 20.325215 9.88 9.84 9.93 9.94 9.69 9.84 9.85 0.1642 30.4465518 8.18 8.38 8.06 8.31 8.31 8.31 8.26 0.1376 36.3269420 7.44 7.37 7.47 7.22 7.53 7.43 7.41 0.1235 40.4858325 5.91 6.09 6.09 5.97 5.85 5.94 5.98 0.0996 50.2092130 5.09 4.85 4.97 4.84 4.94 4.91 4.93 0.0822 60.8108140 3.65 3.78 3.75 3.72 3.73 0.0621 80.53691Time (s)225  Figure B.9: Linear correlation of set point vs. H2 flow rate.   4. The calibration was run by adjusting the flow rate of diluting gas H2. (See Table B.7)  Table B.7: Calculation of mol. concentration of each gas in gas mixture.  Note: CH4 has been highlighted since it is the main generated gas during CHR; only three points were listed here as an example. Point 1 Point 2 Point 3MFC for Mixture gasMFC for diluting H2 UnitMFC for diluting H2 UnitMFC for diluting H2 UnitSetpoint 14 0 ― 10 20Actual Flow 20.2983 0 Sccm 20.331 Sccm 40.424 SccmCalibration Gas Mol% Diluted Gas Mol% Diluted Gas Mol% Diluted Gas Mol%Butane 4 Butane 4 Butane 1.998390324 Butane 1.337123264Carbon Dioxide 8 Carbon Dioxide8 Carbon Dioxide 3.996780648 Carbon Dioxide 2.674246529Ethane 8 Ethane 8 Ethane 3.996780648 Ethane 2.674246529Ethylene 2 Ethylene 2 Ethylene 0.999195162 Ethylene 0.668561632Hydrogen 25 Hydrogen 25 Hydrogen 12.48993953 Hydrogen 8.357020403Isobutane 2 Isobutane 2 Isobutane 0.999195162 Isobutane 0.668561632Methane 10 Methane 10 Methane 4.995975811 Methane 3.342808161Nitrogen 10 Nitrogen 10 Nitrogen 4.995975811 Nitrogen 3.342808161Propane 6 Propane 6 Propane 2.997585486 Propane 2.005684897Carbon Monoxide 25 Carbon Monoxide25 Carbon Monoxide 12.48993953 Carbon Monoxide 8.357020403Hydrogen 0 Hydrogen 50.04024189 Hydrogen 66.57191839100 100 100226  5. Build up a correlation between peak area and mole concentration.  Table B.8: Linear correlation between CH4 mole concentration and measured peak area of CH4.    Figure B.10: Linear correlation of measured peak area of CH4 from GC-FID vs. CH4 mole concentration. H2 flow rate (cc/min)Measured peak area Meaured peak area/10^5Mole(%)0 2253701 22.53701 1010 1021335 10.21335 4.99597640 651707 6.51707 3.34280816160 491258 4.91258 2.50546094380 398489 3.98489 2.01155901100 322833 3.22833 1.674746413227  Appendix C  Sample Calculation C.1 Feed and Products Calculation  (1) Feed Calculation for HDO of 4-methylphenol 4-Methylphenol was chosen as the reactant to mimic the phenol content of a biomass-derived fast pyrolysis oil. This model component was used both in the study of Chapter 2 and Chapter 4. The compositions of the feed are listed in Table C.1:  Table C.1: Feed compositions for HDO of 4-methylphenol. Chemicals Mass (g) 4-methylphenol (4-MP) 2.76 Decalin 87.5 (~ 100 mL) Calculated wt.% of 4-MP 3.1 Note: after dissolving 4-MP in decalin, a sonication was applied to facilitate the dissolving process and remove the dissolved O2 from the solvent.  (2) Feed Calculation for HDS of Dibenzothiophene  Dibenzothiophene (DBT) is one of the refractory S components contained in the heavy oil. It was chosen as the model component for synthesized catalysts testing both in Chapter 3 and Chapter 5. The concentration of DBT is 10x’s lower in the study of Chapter 5 than in Chapter 3, since the aim of Chapter 5 is to study the S effect on Mo2C based catalyst and a low concentration could help for capturing this phenomenon.  The concentration of the prepared feed are present in Table C.2.  228   Table C.2: Feed compositions for HDS of dibenzothiophene. Chemicals Mass (g) a Mass (g) b Dibenzothiophene (DBT) 1.79 0.179 Decalin 87.5 (~ 100 mL) 87.5 (~100 mL) Calculated wt.% of DBT 2.0 0.2 a. This feed was used in Chapter 3’s study. b. This feed was used in Chapter 5’s study.  (3) Products Calculation by GC-MS Measurement Two types of calibration methods were used in the present study for the calculation of reaction conversion and product selectivity. In Chapter 3, an internal calibration method was applied by using diphenylmethane (DPM) as the internal standard. Both the reactant DBT and two main products biphenyl (BPh) and cyclohexylbenzene (CHB) were prepared into different concentrations for the accuracy of the experiment with the addition of DPM. Also, the obtained samples as drawn from the reactor were prepared by the same methodology as the standards.   Here, a calibration table was built for all standards, as shown in Table C.3 (a calibration for DBT with two different concentration ranges).   229   Table C.3: GC-MS calibration table for DBT concentration with the addition of DPE as internal standard (IS).  a. DBT (mg) was calculated by 𝑠𝑎𝑚𝑝𝑙𝑒 (𝑚𝑔)×𝑅𝑒𝑎𝑙 𝑤𝑡%100; b. wt% of DBT was calculated by 𝐷𝐵𝑇 (𝑚𝑔)𝐷𝑃𝐸 (𝑚𝑔)+𝑠𝑎𝑚𝑝𝑙𝑒 (𝑚𝑔); c. wt% of DPE was calculated by 𝐷𝑃𝐸 (𝑚𝑔)𝐷𝑃𝐸(𝑚𝑔)+𝑠𝑎𝑚𝑝𝑙𝑒 (𝑚𝑔).  No.aim wt% DBT(g) Decalin(g)real wt% of DBT―Area of DPE Area  of DBT Area Ratio of DBT:DPE DPE(mg)sample (mg)Real wt% of DBTDBT (mg)a wt% of DBT in prepared sample with IS bwt% of DPE cwt% ratio of DBT to DPE1 1 0.0958 8.9273 1.061719365 ― 3647782 9488086 2.601056203 0.612 191.21 1.06172 2.030134832 1.058332029 0.319042456 3.3172138Aim wt%1 wt% DBT solvent(g)Diluted to xx greal DBT (g)Real wt% of DBTArea of DPE Area of DBT Area Ratio of DBT:DPE DPE(mg)sample (mg)Real wt% of DBTDBT (mg)wt% of DBT in prepared sample with ISwt% of DPEwt% ratio of DBT to DPE2 0.1 1.0957 10.1299 0.011633259 0.114840813 4022566 1116961 0.277673754 0.612 185.78 0.11484 0.213350113 0.114463742 0.328342034 0.34861133 0.3 1.0639 3.3738 0.011295633 0.334804455 5168344 3230321 0.625020509 0.606 184.51 0.3348 0.617751048 0.333708437 0.327360534 1.01939124 0.5 1.0248 2.1398 0.0108805 0.50848212 4736680 4738576 1.00040028 0.602 187.09 0.50848 0.951324284 0.506851232 0.320736522 1.58027295 0.7 0.9989 1.4204 0.010605515 0.746656934 3785651 6895621 1.821515243 0.611 185.59 0.74666 1.385735536 0.74420688 0.328136496 2.2679796 Based on 0.1148408 wt% DBT sample (Sample #: 2)No.Aim wt%1 wt% DBT solvent(g)Diluted to xx greal DBT (g)Real wt% of DBTArea of DPE Area  of DBT Area Ratio of DBT:DPE DPE(mg)sample (mg)Real wt% of DBTDBT (mg)wt% of DBT in prepared sample with ISwt% of DPEwt ratio of DBT to DPE6 0.01 1.0892 10.0412 0.001250846 0.012457137 5639137 91056 0.016147152 0.598 186.76 0.01246 0.023264699 0.012417376 0.319178462 0.03890427 0.03 1.1641 3.3819 0.001336862 0.039529902 6231039 367535 0.058984545 0.614 182.61 0.03953 0.072187135 0.039397437 0.335101622 0.11756868 0.0076 1.0715 13.0998 0.001230519 0.00939342 4314765 74821 0.017340689 0.624 190.8 0.00939 0.017922175 0.009362798 0.325986449 0.0287214230  For sample No.1~No.5, the obtained calibration curve is shown in Figure C.1:  Figure C.1: A linear correlation between area ratio of DBT to DPE and wt% of DBT to DPE in high concentration range.  For sample No.5~No.8, the obtained calibration curve is shown in Figure C.2:  Figure C.2: A linear correlation between area ratio of DBT to DPE and wt% of DBT to DPE in low concentration range.y = 1.2031x + 0.1834R² = 0.983900.511.522.533.50 1 2 3wt% of DBT to DPEArea ratio of DBT to DPEArea ratio of DBT to DPE to wt% of DBT to DPESeries1Linear (Series1)y = 1.2005x + 0.0192R² = 0.984200.10.20.30.40 0.1 0.2 0.3wt.% of DBT to DPEArea ratio of DBT to DPEArea ratio of DBT to DPE to wt% of DBT to DPESeries1Linear (Series1)231  C.2 Calculation Procedure of MoOxCy Formula   Table C.4: The calculation procedure for Mo2C and MoOxCy contents in 10%Mo/AC catalysts. Catalyst Burn-off Mo loading Mo2+ by XPS Mo2C MoOxCy O content x y mol/g cat mol% mol/gcat mol/gcat wt% mol/gcat 10%Mo/AC-600 11.96 1095.2 9.94 108.9 986.3 2.59 1618.8 1.64 0.33 10%Mo/AC-650 21.34 1225.4 23.69 277.7 947.7 1.76 1099.3 1.16 0.61 10%Mo/AC-700 43.01 1619.0 50.47 853.9 752.1 1.00 624.2 0.83 0.90  Calculation process: For 10 g of 10%Mo/AC precursor, it contains 1 g of Mo (eqv. 1.81 g AHM) and 9 g of AC. The total amount of catalyst precursor: 1.81g + 9.0 g = 10.81 g The mass loss of 10%Mo/AC-650 catalyst: 10.81 g x 21.34 % = 2.31 g The remaining mass of 10%Mo/AC-650 catalyst: 10.81 g - 2.31 g = 8.50 g The Mo amount as per gram of synthesized catalyst: 1 𝑔𝑀𝑜8.50 𝑔𝑐𝑎𝑡. = 0.1176𝑔𝑀𝑜𝑔𝑐𝑎𝑡. = 1225.4 𝜇𝑚𝑜𝑙𝑀𝑜𝑔𝑐𝑎𝑡. The Mo composition (%) of different valences as measured by XPS are as follows: 232  Mo2+: 23.69; Mo3+: 18.01; Mo4+: 3.79; Mo5+: 31.96: Mo6+: 22.55 The Mo amount from Mo2C: 1225.4 𝜇𝑚𝑜𝑙𝑀𝑜𝑔𝐶𝑎𝑡. × 23.69% = 277.68 𝜇𝑚𝑜𝑙𝑀𝑜𝑔𝐶𝑎𝑡. The Mo amount from MoOxCy: 1225.4 𝜇𝑚𝑜𝑙𝑀𝑜𝑔𝐶𝑎𝑡.− 277.68𝜇𝑚𝑜𝑙𝑀𝑜𝑔𝐶𝑎𝑡.= 947.72 𝜇𝑚𝑜𝑙𝑀𝑜𝑔𝐶𝑎𝑡. The O content for 10%Mo/AC-650 is 5.68 wt% from O analyzer; the O content from SiO2 is 3.12%(1−21.4%)= 3.92%; the O content from MoOxCy: 5.68%-3.92% = 1.76% = 1.76 𝑤𝑡%16 𝑔/𝑚𝑜𝑙= 1099.34 𝜇𝑚𝑜𝑙𝑂𝑔𝐶𝑎𝑡. So, x= 1099.24947.72= 1.16. The formula of MoOxCy becomes MoO1.16Cy, where y can be calculated as below: y = 18.01%×3+3.79%×4+31.96%×5+22.55%×6100%−23.69%−1.16×24= 0.61 The formula of MoOxCy of 10%Mo/AC-650 is MoO1.16C0.61.  233  C.3 Calculation of Presulfiding Parameters for MoS2/AC Catalyst Preparation (I) The presulfiding was carried out in a batch reactor as reported in Chapter 3 and the details are as follows:  Calculate the amount of Mo in 10%Mo/AC precursor; For example: in each batch of presulfiding, 5 g of 10%Mo/AC precursor was used. The relative amount of Mo in 10%Mo/AC precursor is 0.5 g, which is equivalently to 5.2*10-3 mol.  Calculate the required amount CS2 to sulfide Mo; In order to provide enough S to sulfide Mo, the added CS2 amount is 1.5x’s higher than Mo in present study.  𝐂𝐒𝟐(𝐦𝐋) =𝟏.𝟓∗𝐌𝐨(𝐦𝐨𝐥)∗(𝟕𝟔.𝟏𝟒 𝐠𝐦𝐨𝐥)(𝟏.𝟐𝟔𝟔 𝐠𝐦𝐋)        (Eq. C-1) So, for 5 g of precursor, the required CS2 is 0.469 mL.  Calculate the required amount of H2 pressure for 10 vol%H2S/H2; Since CS2+2H2→2H2S, the required amount of H2 to convert CS2 into H2S is 2x’s of CS2 mole amount. The total required H2 amount is roughly calculated as follows: 𝐇𝟐(𝐦𝐋) = 𝟏. 𝟓 ∗ 𝐌𝐨(𝐦𝐨𝐥) ∗ 𝟐 ∗ (𝟐𝟐.𝟒 𝐋𝐦𝐨𝐥) (𝟏𝟎𝟎𝟎 𝐦𝐋𝐋) ∗ 𝟏𝟎     (Eq. C-2) So, for 5 g of precursor, the total required H2 is 3494.4 mL.  Calculate the required initial pressure of H2. 100 mL decalin was used as solvent during presulfiding. The left volume in 300 mL autoclave is 200 mL. The required initial pressure of H2 was estimated by the ideal gas law (PV=nRT) of P1V1=P2V2, where P1=1 atm, V1=3494.4 mL, V2=200 mL. 𝐏𝟐 =𝐏𝟏𝐕𝟏𝐕𝟐=(𝟏 𝐚𝐭𝐦)(𝟑𝟒𝟗𝟒.𝟒 𝐦𝐋)(𝟐𝟎𝟎 𝐦𝐋)= 𝟏𝟕. 𝟒𝟕 𝐚𝐭𝐦 = 𝟐𝟓𝟑. 𝟑𝟒 𝐩𝐬𝐢     (Eq. C-3) 234   (II) In Chapter 5, the presulfiding was carried out in a continuous flow fixed bed reactor to prepare MoS2/APC in-situ.  Presulfiding conditions:  3.0 MPa and 370 oC for 3 hrs from under 50 cc (STP)/min H2 with 0.0833 cc (STP)/min CS2 in decalin to prepare 5.75 vol% H2S/H2. Calculation process: H2 flow rate = 50 [𝑚𝐿𝑚𝑖𝑛] = 3 [𝐿ℎ] = 3 [𝐿ℎ]22.4 [𝐿𝑚𝑜𝑙]= 0.1339 [𝑚𝑜𝑙ℎ] CS2 + 4H2 = 2H2S + CH4 Mass flow rate of CS2 = 0.1339 [𝑚𝑜𝑙ℎ] × 5.75 % ÷ 2 = 3.84 𝑒 − 3 [𝑚𝑜𝑙ℎ] = 0.2923 [𝑔ℎ] Volumetic flow rate of mixture (CS2 and decalin) = 0.0833 [𝑐𝑐𝑚𝑖𝑛] = 5.0 [𝑚𝐿ℎ]  Set mass concentration of CS2 in decaline as x, The density of the mixture (CS2 and decalin) ρmix = 𝜌𝑐𝑠2*x+ 𝜌𝐷𝑒𝑐𝑎𝑙𝑖𝑛*(1-x) = 1.266 x + 0.896 * (1-x) [𝑔𝑐𝑚3] The mass flow rate of the mixture (CS2 + Decalin) = 5.0 [𝑚𝐿ℎ] x ρmix =5 * (1.266 x +0.896 (1-x)) The mass flow rate of mixure * x = CS2 mass flow rate  x=6.24% So, the calculated concentration of CS2 in decalin is 6.24 wt%.  235  C.4 Stacking Degree (N) Calculation for MoS2 from TEM Images As reported in Chapter 3, the used Ni-Mo2C/AC catalyst could form some MoS2 layers. The number of slabs per stack was recorded by an average stacking degree (N), calculated by 𝑁 =∑ 𝑛𝑖𝑁𝑖/ ∑ 𝑛𝑖𝑖=1..𝑡𝑖=1..𝑡 , where ni is the number of stacks with Ni layers.   For example: For Ni0.19MoC-600 catalyst (In Chapter 3), there are 39 particles being calculated, with 7 of one layer, 28 of two layers and 4 of three layers. So, the calculation is N=28×2+1×7+3×428+7+4=1.9230. The same methodology was applied on other catalysts and the results are summarized in Table C.5.  Table C.5: MoS2 stacking degree (N) of different used Ni-Mo2C/AC catalysts as reported in Chapter 3. Used catalyst N Ni0.09Mo2C/AC-600 0.4324 Ni0.19Mo2C/AC-600 1.9230 Ni0.38Mo2C/AC-600 2.7800 Ni0.44Mo2C/AC-600 4.2000 Ni0.76Mo2C/AC-600 4.8182  236  C.5 Experimental Details for Chapter 4 A detailed experimental recording of activated petroleum coke (APC) supported Mo2C catalysts prepared at three different CHR temperatures with 90 min holding time as reported in Chapter 2 are listed in Table C.6:  Table C.6: Experimental lists of HDO of 4-MP in Chapter 2. Catalyst Burn-off % gMo/gcata a Added cata. amount (g) Calc Mo amount in added cata.(g) b Mo wt% measured by ICP  Real Mo amount in added cata. (g) c Feed gMo/ g4-MP d gMo/ mol4-MP e gMo/ mL4-MP f 4-MP (g) Decalin(g) Mo600_APC-90 20.10 0.1252  0.5776 0.0723  10.23  0.0591  2.7839  87.3272 0.0212 2.2951 0.000591 Mo650_APC-90 35.42 0.1548  0.4583 0.0710  12.59  0.0577  2.7720  87.0618 0.0208 2.2508 0.000577 Mo700_APC-90 53.67 0.2158  0.3323 0.0717  17.76  0.0590 2.7741  86.5691 0.0213 2.3004 0.000590 a. It was calculated by 0.1(1−𝑏𝑢𝑟𝑛 𝑜𝑓𝑓%100); b. The calculated Mo amount in added catalyst was calculated by 𝑔𝑀𝑜𝑔𝑐𝑎𝑡𝑎.× 𝑎𝑑𝑑𝑒𝑑 𝑐𝑎𝑡𝑎. 𝑎𝑚𝑜𝑢𝑛𝑡; c. The real added Mo amount in added catalyst was calculated by 𝑀𝑜 𝑤𝑡% 𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑏𝑦 𝐼𝐶𝑃100× 𝑎𝑑𝑑𝑒𝑑 𝑐𝑎𝑡𝑎. 𝑎𝑚𝑜𝑢𝑛𝑡; d. It was calculated by 𝑅𝑒𝑎𝑙 𝑀𝑜 𝑎𝑚𝑜𝑢𝑛𝑡 𝑖𝑛 𝑎𝑑𝑑𝑒𝑑 𝑐𝑎𝑡𝑎𝑙𝑦𝑠𝑡 (𝑔)𝑚𝑎𝑠𝑠 𝑜𝑓 4−𝑀𝑃 𝑖𝑛 𝐹𝑒𝑒𝑑 (𝑔); e. It was calculated by gMogcata.× 108.13 (gmol); f. It was calculated by 𝑅𝑒𝑎𝑙 𝑀𝑜 𝑎𝑚𝑜𝑢𝑛𝑡 𝑖𝑛 𝑎𝑑𝑑𝑒𝑑𝑐𝑎𝑡𝑎𝑙𝑦𝑠𝑡 (𝑔)100 (𝑚𝐿). 237  C.6 Rate Constants Calculation Reported in Chapter 4 All the experiment were run in a batch reactor at 350 oC and 4.3 MPa for 5 h. In order to exclude the reactor heat up period from the experiment, a liquid was drawn from the side line of the reactor and recorded as the first point (t=0 min). Also, other three samples were drawn at t = 60, 180, and 300 min, respectively. An identification and quantitative analysis were achieved using a Shimadzu GC/MS analyzer. The rate constants were calculated by assuming pseudo 1st order reaction. The used matlab code is shown in Appendix D.1.   Table C.7: Activity, selectivity, and kinetic rate constants for HDO of 4-methylphenol reported in Chapter 2.  Catalyst SDDO/HYD b  kDDO (mL/gMo.min) c kHYD (mL/gMo.min) c k (mL/gMo.min) d Mo600_APC-90 4.89 4.66±0.53 0.95±0.42 5.62±0.95 Mo650_APC-90 6.88 5.40±0.38 0.79±0.29 6.19±0.67 Mo700_APC-90 8.23 6.69±0.53 0.61±0.40 7.30±0.93 a. Reaction conversion (χ) was obtained by running experiment for 5 h.  b. Selectivity of DDO to HYD was calculated by 𝑆𝐷𝐷𝑂/𝐻𝑌𝐷 =𝑘𝐷𝐷𝑂𝑘𝐻𝑌𝐷. c. Both DDO (kDDO) and HYD (kHYD) rate constants were calculated by assuming pseudo 1st order reaction. d. It was calculated by 𝐾 = 𝑘𝐷𝐷𝑂 + 𝑘𝐻𝑌𝐷.238  C.7 Calculation of Activation Energy of Carbon Hydrogenation in Chapter 4 Table C.8: Experimental data of in-situ exit gas analysis of Mo800_APC catalyst by carbothermal hydrogenation reduction. Temp.(°C) Time Time interval (min) CH4 area  (RT = 1.45 min) CH4 mole% 1000/T  (1/K) CH4 rate (µmol/s)a 62 3/8/2017 13:03 37  0 0 2.98507 0 110 3/8/2017 13:50 84  0 0 2.61097 0 174 3/8/2017 14:54 148  0 0 2.23714 0 251 3/8/2017 16:11 225  130 0.00416 1.9084 0.00309 297 3/8/2017 16:57 271  161 0.005152 1.75439 0.00383 341 3/8/2017 17:42 316  390 0.01248 1.62866 0.00928 381 3/8/2017 18:21 355  2514 0.0804238 1.52905 0.0598 412 3/8/2017 18:52 386  2950 0.0877813 1.45985 0.06527 450 3/8/2017 19:30 424  3460 0.0963875 1.38313 0.07167 488 3/8/2017 20:08 462  4883 0.1204006 1.31406 0.08953 520 3/8/2017 20:47 501  7937 0.1719369 1.26103 0.12785 553 3/8/2017 21:13 527  12919 0.2560081 1.21065 0.19036 588 3/8/2017 21:50 564  32346 0.5481261 1.16144 0.40758 623 3/8/2017 22:23 597  108809 0.8765347 1.11607 0.65178 659 3/8/2017 22:59 633  378360 2.0342562 1.07296 1.51264 703 3/8/2017 23:44 678  793443 3.8170377 1.02459 2.83828 737 3/9/2017 0:17 711  729335 3.5416938 0.9901 2.63354 768 3/9/2017 0:48 742  631151 3.1199935 0.96061 2.31997 800 3/9/2017 1:21 775  537979 2.7198198 0.93197 2.02241 a. CH4 rate =𝐶𝐻4100 𝑚𝑜𝑙% ×100 𝑚𝐿/𝑚𝑖𝑛22414 𝑚𝑜𝑙𝑚𝐿×60𝑠𝑚𝑖𝑛 × 106µ𝑚𝑜𝑙𝑚𝑜𝑙. 239  A linear regression can be obtained in the range of T = 550 ~ 700 oC with a R2 = 0.9950.  Figure C.3: Arrhenius plot of temperature dependence in the range of 550~700 oC for CH4 formation.   The calculated results are shown in Figure C.4:  Figure C.4: Calculated activation energy (Ea) for carbon hydrogenation.   Then, Ea = b*(-8.314)=120.1 kJ/mol; the pre-exponential factor ko = a = 7.7.240  C.8 Reaction Phase Determination from ASPEN Plus Calculation A phase determination was simulated by ASPEN plus in a flash to see if the reaction was running in one phase or two phases in Chapter 5 and Chapter 6.  As an example, HDS of 0.2 wt% DBT in a fixed bed reactor was put here: Simulation conditions: 350 oC (662 F), 4.1 MPa (600 psi), and H2/Feedmix = 600 Reactant: 0.2 wt% DBT in decalin (equv. to 0.1541 mol%) Thermodynamics method: Wilson Steams:   Table C.9: Mole flow rates of different components in HDS of dibenzothiophene in fixed bed reactor.  Component Name Flow rate (mol/h) H2 H2 414.0 DBT Feedmix 0.1541 Decalin 99.8459  The calculation details for mole flow rates are as follows:  Density of Feedmix: 𝜌𝑚𝑖𝑥 =0.896×1.2520.896×0.2%+1.252×99.8%= 0.8965 𝑔/𝑐𝑚3 Mole weight of the Feedmix: 𝑀𝑚𝑖𝑥 = 138.25𝑔𝑚𝑜𝑙× 99.8459% + 184.26𝑔𝑚𝑜𝑙× 0.1541% = 138.32𝑔𝑚𝑜𝑙 Assume there is 1 L of feedmix, Mole of Feedmix: 𝑛𝑚𝑖𝑥 =𝜌𝑚𝑖𝑥×𝑉𝑚𝑖𝑥𝑀𝑚𝑖𝑥=0.8965𝑔𝑐𝑚3×1.0 𝐿×1000 𝑐𝑚3𝐿138.32𝑔𝑚𝑜𝑙=6.48 mol Since H2/Feedmix (v/v) = 600, 1 L of feedmix needs 600 L H2, 241  Mole of H2: 𝑛𝐻2 =600 𝐿22.4𝐿𝑚𝑜𝑙= 26.79 𝑚𝑜𝑙 The mole ratio between Feedmix and H2 is: 6.48 𝑚𝑜𝑙26.79 𝑚𝑜𝑙=14.14  Figure C.5: Aspen flowsheet for H2 and feedmix in a flash reactor.   Table C.10: Calculated heat and material balance table.   LIQUIDFLASHVAPORH2FEEDMIXHeat and Material Balance TableStream ID FEEDMIX H2 LIQUID VAPORTemperature F      662.0      662.0       662.0Pressure psia     600.00     600.00     600.00     600.00Vapor Frac      0.000      1.000       1.000Mole Flow lbmol/hr      0.220      0.913      0.000      1.133Mass Flow lb/hr     30.495      1.840      0.000     32.335Volume Flow cuft/hr      0.832     18.311      0.000     22.734Enthalpy MMBtu/hr     -0.010      0.004      -0.004Mole Flow lbmol/hr       H2                0.913                0.913  C10H1-01      0.220                          0.220  C12H8-01    < 0.001                        < 0.001242  The simulation shows that every component will go into vapor phase. So, it demonstrates that this reaction happened in one phase. The same simulation has been applied in HDN of carbazole (CBZ) and HDO of dibenzofuran (DBF). All the results show the gas phase reaction.  243  C.9 System Dynamic Response in Fixed-bed Reactor The system dynamic response time was measured to make sure all data reported in Chapter 5 and Chapter 6 are effective to show the transition/deactivation of Mo2C/APC catalyst. 2% DBT in decalin was used as feed to pump into the reactor with a LHSV of 4 h-1. The feed was injected into the reactor once the furnace started heating. SiC was used here to represent the catalyst bed. The reactor temperature can reach to 350 oC after 31 min. Then, the 1st sample was drawn from the reactor. The experiment was keep running by drawing samples from the condenser every 20 min. The total C mol.% was calculated and shown in Figure C.6.    Figure C.6: Total C mol.% in fixed-bed system at 350 oC and LHSV=4 h-1.  It shows that the system took 72 min to reach a C mol.% balance. Excluded the 31 min from heating up phase, it only takes 41 min to get mass balance. This measured experimental data is 244  accordance with the calculated response time (usually it is 3x’s residence time; 45 min). So, all the experimental data reported in Chapter 5 and Chapter 6 is after 41 min.   245  Appendix D  Kinetic Model Code D.1 Matlab Code for HDO of 4-methylphenol in Batch Reactor This code has been used for many years by this research group. In the present study, it is used to calculate the rate constants of HDO of 4-methylphenol presented in Chapter 2 and Chapter 4. All the products have been grouped into HYD and DDO.  For example, the product distribution of 10%Mo/AC-650 catalyst in HDO of 4-methylphenol at 350 oC is reported in Table D.1. There data shown here were used for the modeling as below. Since modelmulti code, Jacobian matrix and least square code won’t change for HDO of 4-methylphenol and HDS of dibenzothiophene, they are present one time in this thesis in Appendix D.2.  Table D.1: Products distribution of HDO of 4-methylphenol at 350 oC with 10%Mo/AC-650 catalyst. Time (min) 4-methylphenol DDO product (mol/L) HYD products (mol/L) Toluene Methyl-cyclohexane 1-Methylcyclohexene 4-Methylcyclohexene Sum of HYD products 60 0.226123 0.024646849 0.001851564 0.001820452 0.002706077 0.006378 180 0.185071 0.057222816 0.005373534 0.004825215 0.005081097 0.01528 300 0.143882 0.087081926 0.009559886 0.008039904 0.008519898 0.02612 420 0.114825 0.109486825 0.013872935 0.006260651 0.01324843 0.033382  (1) Main body clear all global nvar nx x0 y0 global verbose 246  global n1 n2 n3 n4 H2 verbose(1:2)=1;   % x is the indep variable vector e.r. time measurements % y is the matrix of responses % columns of y are responses y1, y2(e.g. mol fraction of component 1 and 2) % rows of y are y values at the values of the indep variable (time) in x % first row of y is initial value of response % the program uses the Levenbreg-Marquart method to estimate parameters % and calculate -done in leasqr and dfdp % the L-M requires the model to be calculated- this is done in modelmulti.m % and assumes the model is a series of ODEs, with the number of ODEs is % equal to the number of responses. The ODEs are calculated in ODEfunm. % Note that this function must use the correct model for each y % time data (min) T=[0 60 180 300]   nt=length (T) x(1:nt-1)=T(2:nt) x nx=length(x)   % 4-methylphenol concentration data (mol/L) CAX=[0.226123 0.185071 0.143882 0.114825];  % DDO concentration data (mol/L) CBX=[0.024646849 247  0.057222816 0.087081926 0.109486825];   % HYD concentration data (mol/L) CCX=[0.006378 0.01528 0.02612 0.033382];  for j=1:nt-1     y1(j)=CAX(j+1);     y2(j)=CBX(j+1);     y3(j)=CCX(j+1);      end   nvar=3; x0=0;   oldx=x; nx= length(x); y=[y1';y2';y3']; %y=[y1';y2']; newy=y(:); oldy=reshape(newy,nx,nvar); x=x'; newx=[x;x;x;]; %newx=[x;x;] y01(1:nx)=CAX(1); 248  y02(1:nx)=CBX(1); y03(1:nx)=CCX(1);    newy0=[y01';y02';y03']; %newy0=[y01';y02']; %INPUT DATA NOW IN CORRECT COLUMN FORMAT y0=newy0 x=newx y=newy   %provide initial parameter guesses theta=[0.001 0.009]; np=length(theta) pin=theta nxcheck=size(x) nycheck=size(y)   %Begin calculation by calling L-M least squares routine   [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]=leasqr(x,y,pin,'modelmulti',0.00001,100000); disp('RESPONSE:') if kvg==1     disp('PROBLEM CONVERGED') elseif kvg==0     disp('PROBLEM DID NOT CONVERGE') end format shortEng oldf=reshape(f,nx,nvar); oldr=reshape(y-f,nx,nvar); disp('X-values:') 249  disp(oldx') disp('Y-values') disp(oldy) disp('f-values - i.e. model calculated responses') disp(oldf) disp('Residuls:') disp(oldr) % disp('Standardized 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) figure subplot(3,3,1) plot(oldx(:),oldy(:,1),'o',oldx(:),oldf(:,1),'--') title('4MP') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,2) plot(oldx(:),oldy(:,2),'o',oldx(:),oldf(:,2),'--') title('DDO') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,3) plot(oldx(:),oldy(:,3),'o',oldx(:),oldf(:,3),'--') title('HYD') 250  xlabel('Time (min)') ylabel('Concentration (mol/L)')  (2) ODE code function yprime=ODEfunm(xatx,yatx,pin) global nvar nx x0 y0 xstep   % disp('*****************YPRIME') % disp(knt) % nx k1=pin(1); k2=pin(2);   %ntest=(knt/nx);   yp(1)=-k1*yatx(1)-k2*yatx(1); yp(2)=k1*yatx(1); yp(3)=k2*yatx(1);   yprime=[yp(1)';yp(2)';yp(3)'];   251  An example of rate constant calculation results by matlab code in Appendix D.1:  Table D.2: Calculated rated constants of APC supported Mo2C catalysts prepared at 600 and 650 oC.   Where k’DDO- rate constant for DDO route; Δk’DDO- standard deviation of rate constant for DDO route; k’HYD- rate constant for HYD route; Δk’HYD- standard deviation of rate constant for HYD route; kTDDO- rate constant for DDO route from thermal reaction; ΔkTDDO- standard deviation of rate constant for DDO route from thermal reaction; kTHYD- rate constant for HYD route from thermal reaction; ΔkTHYD- standard deviation of rate constant for HYD route from thermal reaction; kDDO- rate constant for DDO route excluding thermal reaction; ΔkDDO- standard deviation of rate constant for DDO route excluding thermal reaction; kHYD- rate constant for HYD route excluding thermal reaction; ΔkHYD- standard deviation of rate constant for HYD route excluding thermal reaction. Note: The unit of all the rate constants reported here is [1/min].  After unit conversion:  Note: The unit of all rate constants reported here is [mL/gMo.min] Catalyst k'DDO Δk'DDO kTDDO Δ kTDDO kDDO ΔkDDO k'HYD Δk'HYD kTHYD Δ kTHYD kHYD ΔkHYD K' ΔK'Mo600_APC-90 0.003 3E-04 5E-04 1.0095E-05 0.0028 0.0003 7E-04 2E-04 0.0002 1.047E-05 0.0006 0.0002 0.003 6E-04Mo650_APC-90 0.004 2E-04 5E-04 1.0095E-05 0.0031 0.0002 6E-04 2E-04 0.0002 1.047E-05 0.0005 0.0002 0.004 4E-04Mo700_APC-90 0.004 2E-04 5E-04 1.0095E-05 0.0031 0.0002 4E-04 2E-04 0.0002 1.047E-05 0.0003 0.0002 0.003 4E-04Catalyst kDDO ΔkDDO kHYD ΔkHYD K ΔKMo600_APC-90 4.665 0.534 0.954 0.422 5.618 0.956Mo650_APC-90 5.404 0.376 0.786 0.286 6.189 0.662Mo700_APC-90 5.192 0.414 0.475 0.311 5.667 0.725252  D.2 Matlab Code for HDS of Dibenzothiophene in Batch Reactor The following codes were used to model the HDS of DBT in batch reactor. The activity data of Ni0.19Mo2C/AC-550 catalyst were used as an example to show the process of this calculation. Since the catalyst could go through a sulfidation in the 1st hour of the HDS reaction, the experimental data obtained after 1 h were used for kinetic analysis. The ODE code was written based on reaction pathways. ===================================================================== (1) Main Body >> clear all global nvar nx x0 y0 global verbose global n1 n2 n3 n4 H2 verbose(1:2)=1;   % x is the indep variable vector e.r. time measurements % y is the matrix of responses % columns of y are responses y1, y2 (e.g. mol fraction of component 1 and 2) % rows of y are y values at the values of the indep variable (time) in x % first row of y is initial value of response % the program uses the Levenbreg-Marquart method to estimate parameters % and calculate done in m-files of leasqr and dfdp % these two m-files are designed for single response % the L-M requires the model to be calculated- this is done in modelmulti.m % and assumes the model is a series of ODEs, with the number of ODEs is % equal to the number of responses.  % The ODEs are calculated in ODEfunm. % Note that this function must use the correct model for each y % the calculated rate constants are in the unit of [1/min]   % Time data (min) T=[60 180 300]   nt=length (T) 253  x(1:nt-1)=T(2:nt)   nx=length(x)   % DBT concentration data (mol/L) CAX=[8.1175E-02 6.3547E-02 5.6010E-02];   % THDBT concentration data (mol/L) CBX=[1.0873E-03 2.0758E-03 2.6117E-03];   % BPh concentration data (mol/L) CCX=[1.3937E-02 2.6506E-02 3.3690E-02];   % CHB concentration data (mol/L) CDX=[7.5116E-04 3.1119E-03 4.2710E-03];   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); end   nvar=4; x0=0;   oldx=x; nx= length(x); y=[y1';y2';y3';y4']; %y=[y1';y2';y3']; newy=y(:); 254  oldy=reshape(newy,nx,nvar); x=x'; newx=[x;x;x;x;]; %newx=[x;x;x;] y01(1:nx)=CAX(1); y02(1:nx)=CBX(1); y03(1:nx)=CCX(1); y04(1:nx)=CDX(1);   newy0=[y01';y02';y03';y04']; %newy0=[y01';y02';y03']; %INPUT DATA NOW IN CORRECT COLUMN FORMAT y0=newy0 x=newx y=newy   %provide initial parameter guesses theta=[0.001 0.001 0.00001]; np=length(theta) pin=theta nxcheck=size(x) nycheck=size(y)   %Begin calculation by calling L-M least squares routine   [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]=leasqr(x,y,pin,'modelmulti',0.00001,10000); disp('RESPONSE:') if kvg==1     disp('PROBLEM CONVERGED') elseif kvg==0     disp('PROBLEM DID NOT CONVERGE') end format shortEng oldf=reshape(f,nx,nvar); oldr=reshape(y-f,nx,nvar); disp('X-values:') disp(oldx') disp('Y-values') disp(oldy) 255  disp('f-values - i.e. model calculated responses') disp(oldf) disp('Residuls:') disp(oldr) % disp('Standardized 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) figure subplot(3,3,1) plot(oldx(:),oldy(:,1),'o',oldx(:),oldf(:,1),'--') title('DBT') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,2) plot(oldx(:),oldy(:,2),'o',oldx(:),oldf(:,2),'--') title('THDBT') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,3) plot(oldx(:),oldy(:,3),'o',oldx(:),oldf(:,3),'--') title('BPh') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,4) plot(oldx(:),oldy(:,3),'o',oldx(:),oldf(:,4),'--') title('CHB') xlabel('Time (min)') ylabel('Concentration (mol/L)')  (2) Modelmulti code function f= modelmulti (x,pin) % Solve a simple system of ODE's - response variables % find the solution at sepcified x values - corresponding to measured data 256  % first data point in x corresponds to initial condition global nvar nx x0 y0 global verbose   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(:);  (3) ODE code function yprime=ODEfunm(xatx,yatx,pin) global nvar nx x0 y0 xstep   % disp('*****************YPRIME') % disp(knt) % nx k1=pin(1); k2=pin(2); k3=pin(3);    %ntest=(knt/nx);   yp(1)=-k1*yatx(1)-k2*yatx(1); yp(2)=k1*yatx(1)-k3*yatx(2); yp(3)=k2*yatx(1); yp(4)=k3*yatx(2);   yprime=[yp(1)';yp(2)';yp(3)';yp(4)'];  (4) Calculation of Jacobian matrix 257  function prt=dfdp(x,f,p,dp,func) % numerical partial derivatives (Jacobian) df/dp for use with leasqr   m=length(x); n=length(p); %dimensions ps=p;prt=zeros(m,n);del=zeros(n,1); % initialize 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, fl=feval(func,x,p);         if dp(j)<0, prt(:,j)=(fl-f)./del(j);         else             p(j)=ps(j)-del(j);             prt(:,j)=(fl-feval(func,x,p))./(2.*del(j));         end     end     p(j)=ps(j);  %restore p(j) end  return  (5) Least square code 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}) % % Levenberg-Marquardt nonlinear regression of f(x,p) to y(x). % % Version 3.beta % Optional parameters are in braces {}. % x = column vector or matrix of independent variables, 1 row per %   observation: x = [x0 x1....xm]. % y = column vector of observed values, same number of rows as x. % wt = column vector (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 258  %   diagonal with diagonal equal to (wt.^2)^-1.  The constant of %   proportionality will be estimated.); default = ones(length(y),1). % pin = column vec of initial parameters to be adjusted by leasqr. % dp = fractional increment of p for numerical partial derivatives; %   default = .001*ones(size(pin)) %   dp(j) > 0 means central differences on j-th parameter p(j). %   dp(j) < 0 means one-sided differences on j-th parameter p(j). %   dp(j) = 0 holds p(j) fixed i.e. leasqr wont change initial guess: pin(j) % F = name of function in quotes; the function shall be of the form y=f(x,p), %   with y, x, p of the form y, x, pin as described above. % dFdp = name of partial derivative function in quotes; default is "dfdp", a %   slow but general partial derivatives function; the function shall be %   of the form prt=dfdp(x,f,p,dp,F) (see dfdp.m). % stol = scalar tolerance on fractional improvement in scalar sum of %   squares = sum((wt.*(y-f))^2); default stol = .0001; % niter = scalar maximum number 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 = column vector of values computed: f = F(x,p). % p = column vector trial or final parameters. i.e, the solution. % kvg = scalar: = 1 if convergence, = 0 otherwise. % iter = scalar number of iterations 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 (see comments in the source). % r2 = coefficient of multiple determination. 259  % % All Zero guesses not acceptable   % 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.... % % 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>   % 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 %       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. 260  %       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 Francesco Potort? %       for use in Octave % % References: % Bard, Nonlinear Parameter Estimation, Academic Press, 1974. % Draper and Smith, Applied Regression Analysis, John Wiley and Sons, 1981. % %set default args   % argument processing %   %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)'; else   plotcmd='plot(x(:,1),y,''+'',x(:,1),f); shg'; end; if (exist('OCTAVE_VERSION'))   global verbose   plotcmd='plot(x(:,1),y,"+;data;",x(:,1),f,";fit;");'; end;   if(exist('verbose')~=1), %If verbose undefined, print nothing     verbose=0;       %This will not tell them the results 261  end;   if (nargin <= 8), dFdp='dfdp'; end; if (nargin <= 7), dp=.001*(pin*0+1); end; %DT if (nargin <= 6), wt=ones(length(y),1); end;    % SMB modification if (nargin <= 5), niter=20; end; if (nargin == 4), stol=.00001; 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(:,1), Inf*ones(nor,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); ss=r'*r; sbest=ss; nrm=zeros(n,1); chgprev=Inf*ones(n,1); kvg=0; 262  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));       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;     idx = ~isinf(maxstep);     limit = abs(maxstep(idx).*pprev(idx)); 263      chg(idx) = min(max(chg(idx),-limit),limit);     if (verbose & 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),     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;     if (verbose),       fprintf('Parameter changes converged to specified precision\n');     end;     break;   else     chgprev=chg; 264    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 exist('sparse')  % save memory   Q=sparse(1:m,1:m,1./wt.^2);   Qinv=sparse(1:m,1:m,wt.^2); else   Q=diag((0*wt+1)./(wt.^2));   Qinv=diag(wt.*wt); 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 265  d=sqrt(abs(diag(covp))); corp=covp./(d*d');   if exist('sparse')   covr=spdiags(covr,0);   stdresid=resid./sqrt(spdiags(Vy,0)); else   covr=diag(covr);                 % convert returned values to compact storage   stdresid=resid./sqrt(diag(Vy));  % compute then convert for compact storage end Z=((m-n)*jac'*Qinv*jac)/(n*resid'*Qinv*resid);   %%% 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(:)]); r2=r(1,2).^2;   % if someone has asked for it, let them have it % if (verbose),    eval(plotcmd);   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; 266    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;  D.3 Matlab Code for Deactivation Constant Calculation The exponential decay rate law (a=𝑒−𝑘𝑑𝑡) was used to study the deactivation of Mo2C/APC catalysts in the presence of S and N. The calculated decay constant (kd) can be used to describe how fast the deactivation could be. Take 10%Mo2C/APC catalyst in HDS of DBT as an example. The transition period data is reported in Table D.3.  Table D.3: DBT conversion and initial rate of reaction in HDS of DBT for 10%Mo2C/APC catalyst as a function of time on stream.  TOS (sec) χDBT (%) -In(1- χDBT) k' [molDBT/gcata.s] Initial-rDBT (gDBT/gcata.s) Initial-rDBT (molDBT/gcata.s) 3600 98.02 3.9222 1.0586E-06 1.9539E-04 1.0604E-06 5400 83.76 1.8180 3.5049E-07 6.4689E-05 3.5108E-07 7500 69.06 1.1730 2.2614E-07 4.1738E-05 2.2652E-07 9120 65.45 1.0628 2.0490E-07 3.7818E-05 2.0524E-07 10980 64.73 1.0420 2.0089E-07 3.7078E-05 2.0123E-07  (1) Main body 267  clear all clc global  x0 y0 a global verbose verbose(1:2) = 1; % Activity term parameter estimation % input number of responses % format shortEng nvar=1; x0=0.; y0=0.;   %--------------------------------------------------------- % t,sec   x_s=[3600 5400 7500 9120 10980];   oldx=x_s; [nx,mx] = size(x_s)   %--------------------------------------------------------------------- %DBT consumption rate   yT2=[1.0604E-06 3.51076E-07 268  2.26517E-07 2.05244E-07 2.01226E-07];   %--------------------------------------------------------------------- y = yT2; oldy=y;   %kd   A0 %1st order theta=[.0001];   %2nd order %theta=[1e-1]; %reciprocal power/Modified 1st order with H2S/NH3 conc %theta=[0.0001 0.00001]; %coke %theta=[5];   np=length(theta); pin=theta;   % Begin calculation by calling L-M least squares routine [a_s,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]=leasqr(x_s,y,pin,'KS_modelmulti_HW',0.00001,10000);   disp('RESPONSE:') if kvg ==1     disp ('PROBELM CONVERGED')     elseif kvg == 0     disp('PROBLEM DID NOT CONVERGE') 269  end   r=y-a_s;     disp ('X-values:')     disp (oldx')      disp ('Y-values:')     disp (oldy')      disp ('actvity values for all of run')     disp (a')     disp ('a-values - i.e. model calculated responses')     disp (a_s)     disp ('Residuals:')     disp (r)     disp ('Standardized residuals')     disp (stdresid)     disp ('Estimated parameter values are:')     disp (p)     disp ('Covariance of estimated parameters:')     disp (covp)     disp ('R2 values is:')     disp (r2)     disp ('No. of interations:')     disp (iter)       subplot(2,1,1);    plot (oldx,a_s,'--R','linewidth',2)    hold on    plot(oldx,oldy,'ok','MarkerSize',10) %plot the data using open black circles    xlabel('time (sec)')    ylabel('DBT consumption rate, mol/(sec.gcat)') 270         subplot(2,1,2);    plot (oldy,a_s,'o'), hold, plot(oldy,oldy,'black')    xlabel('Experimental, mol/(sec.gcat)')    ylabel('Calculated, mol/(sec.gcat)')  (2) Modelmulti code % Effect of S/N activity model. % Modified by Haiyan Wang-20170416. function a_s = KS_modelmulti_HW (x_s,pin) global x0 y0 a  %n=3; %T,K  t,h x=x_s;  %--------------------------------------------------------- % H2S or NH3 concentration, mmol   s=[0.01071639 0.008991737 0.007185979 0.00674708 0.006654635];    %-------------------------------------------------------- a(1)=1.0; ro=1.0604E-06; %Initial experimental rate data   [nxx,mxx]=size(x);    271   for i = 2:nxx;      z(1,1)=0;      z(i,1)=(x(i,1)-x(1,1));% Divide 0.8 before            kd=pin(1);  %1st order  a(i)=a(i-1)*exp(-kd*(z(i,1)-z(i-1,1)));    %H2S modified 1st order  %a(i)=a(i-1)*exp(-kd*(s(i)^n*z(i,1)-s(i-1)^n*z(i-1,1))) end    [nxs,mxs]=size(x_s);       for j=1:nxs     a_s(j)=ro*a(j); end      a_s = a_s';  Then calculated results are presented as below:  Table D.4: Calculated decay constant (kd) from exponential decay rate law for 10Mo2C/APC in HDS of DBT at 350 oC and 4.1 MPa.   kd 4.2211E-04 s-1 2.5327E-02 h-1Covp 9.2679E-09SD 9.6270E-05 s-15.7762E-03 h-1R29.3116E-01272  Appendix E  Supplementary Figures and Tables  E.1 Supplementary Information for Chapter 2   Figure E.1: XRD pattern of 10% Mo/AC-650 prepared in Ar atmosphere. (◊) - SiO2; (*) - MoO2.  273   Figure E.2: Product concentrations as a function of reaction time during 4-methylphenol hydrodeoxygenation at 350C and 4.3 MPa H2. () 10%Mo-AC600, (o) 10%Mo-AC-650 and () 10%Mo-AC-700.  0 1 2 3 4 5 6 7 80.000.020.040.060.080.100.12  Concentrtaion, mol/L Toluene0 1 2 3 4 5 6 7 80.0000.0020.0040.0060.0080.0100.0120.0140.0160.0180.020    Methyl-cyclohexane0 1 2 3 4 5 6 7 80.0000.0020.0040.0060.0080.0100.0120.0140.0160.0180.020  Concentration, mol/L Reaction time, h1-methyl-cyclohexene0 1 2 3 4 5 6 7 80.0000.0020.0040.0060.0080.0100.0120.0140.0160.0180.020   Reaction time, h4-methyl-cyclohexene274   Figure E.3: Arrhenius plots for 1st-order rate constants of (a) DDO and (b) HYD reactions.  Data centered at T0 = 350C. 275   Table E.1: Kinetic parameters measured for thermal reaction of 4-methylphenol.   𝑘𝐷𝐷𝑂𝑇      𝑘𝐻𝑌𝐷𝑇   Temperature, C x104 min-1  325 1.5050.140a 0.7510.014  350 5.1750.200 1.5510.200  375 13.940.700 4.0500.788   276  E.2 Supplementary Information for Chapter 3  Figure E.4: XRD patterns of calcined Ni-Mo2C/AC catalyst precursors.  277   Figure E.5: N2 adsorption-desorption isotherms of nitrogen at -193 oC for the AC support and Ni-Mo2C/AC catalysts.  Figure E.6: The effect of Ni:Mo on (a) surface area, (b) pore size, (c) Vmeso/Vtotal, for fresh and used Ni-Mo2C/AC catalysts at reduction temperatures of 550 oC and 600 oC. ― Trend line. 278    Figure E.7: Experimental and model concentration data versus reaction time of different Ni-Mo2C catalysts prepared at 550 oC: Dibenzothiophene ( DBT, ▼); Biphenyl (BPh, ●); tetrahydro-dibenzothiophene (THDBT, ▲); cyclohexylbenzene (CHB, ■). (a) Ni0.02Mo2C/AC-550; (b) Ni0.09Mo2C/AC-550; (c) Ni0.19Mo2C/AC-550; (d) Ni0.38Mo2C/AC-550; (e) Ni0.44Mo2C/AC-550; (f) Ni0.76Mo2C/AC-550.279    Figure E.8: Experimental and model concentration data versus reaction time of different Ni-Mo2C catalysts prepared at 600 oC: Dibenzothiophene ( DBT, ▼); Biphenyl (BPh, ●); tetrahydro-dibenzothiophene (THDBT, ▲); cyclohexylbenzene (CHB, ■). (a) Ni0.02Mo2C/AC-600; (b) Ni0.09Mo2C/AC-600; (c) Ni0.19Mo2C/AC-600; (d) Ni0.38Mo2C/AC-600; (e) Ni0.44Mo2C/AC-600; (f) Ni0.76Mo2C/AC-600. 280   Figure E.9: Correlation of Ni/Mo ratio determined by ICP and adsorbed S wt% determined by CHNS analyzer for Ni-Mo2C/AC reduced at different temperatures (□: reduced at 550 oC; ○: reduced at 600 oC).   The solid lines represent the correlation equation:  for (□), S (wt%) = (2.357±0.082a) x Ni:Mo ratio + (2.801±0.036);  for (○), S (wt%) = (3.164±0.548) x Ni:Mo ratio + (1.865±0.219). a represents standard deviation.   281    Figure E.10: TOF-SIMS spectrum of used Ni0.19Mo2C/AC-550 catalyst based on spot of scan area. 282  E.3 Supplementary Information for Chapter 4   Figure E.11: Mo (3d) XPS narrow scan spectra deconvolution of APC supported Mo2C catalysts at different CHR temperatures: (a) Mo400_APC; (b) Mo500_APC; (c) Mo550_APC; (d) Mo600_APC; (e) Mo650_APC; (f) Mo700_APC; (g) Mo750_APC; (h) Mo800_APC. 283  A XPS analysis of Mo 3d peak deconvolution of fresh Mo/APC samples prepared at different CHR temperatures with and without holding for 90 min are presented in Figure E.11 and E.12. Also, an atomic percentage of Mo 3d, O 1s and C 1s as obtained from XPS survey scans are reported in Table E.2.   Figure E.12: The deconvolution of Mo 3d narrow scan spectra of fresh Mo/APC samples with different CHR temperatures holding for 90 min: (a) survey scan of Mo600_APC; (b) Mo600_APC; (c) Mo650_APC; (d) Mo700_APC. (“―” Mo2+; “―” Mo3+; “―” Mo4+; “―” Mo5+; “―” Mo6+)  284   Table E.2: Elemental compositions of fresh and used Mo/APC (after HDO of 4-MP) samples with different CHR temperatures holding for 90 min. At% Mo600_APC-90 Mo650_APC-90 Mo700_APC-90 Fresh Used Fresh Used Fresh Used Mo3d 1.41 1.30 1.84 1.80 2.90 2.99 C1s 94.18 94.06 92.54 92.73 91.14 90.99 O1s 4.40 4.64 5.62 5.47 5.96 6.01 Sum 99.99 100.00 100.00 100.00 100.00 99.99   285  The isothermal curves of APC supported Mo samples are presented as below:   Figure E.13: Isotherms of APC supported Mo2C catalysts with different CHR temperatures. 286    Figure E.14: Isotherms of APC supported Mo2C catalysts with different CHR temperatures holding for 90 min: (a) Mo600_APC-90; (B) Mo650_APC-90; (c) Mo700_APC-90.  287  Molecular models for DFT calculation:               MoO3 (010)                                  MoO3 (010) with one oxygen vacancy  MoO3C (010) ==============================================================        H-MoO3                             H-Ov-MoO3  H-MoO3C Figure E.15: H adsorption energy on MoO3 (010), MoO3 (010) with one oxygen vacancy and MoO3C(010). 288  Table E.3: H adsorption energy on MoO3 (010), MoO3 (010) with one oxygen vacancy and MoO3C (010). Status MoO3 (010) MoO3 (010) with O vacancy MoO3C H adsorption energy (eV) -1.10  -1.21  -2.46   The results of activity tests are shown in Figure E.16.   Figure E.16: Experimental and model concentration data versus reaction time of different catalysts with different CHR temperatures. (a) Mo600_APC-90; (b) Mo650_APC-90; (c) Mo700_APC-90. (■) Reactant: 4-MP; (●) HYD products; (▲) DDO products. 289  E.4 Supplementary Information for Chapter 5   Figure E.17: A correlation of Mo loadings for the fresh Mo2C/APC catalysts with various Mo loadings and IMo/IC.  290   Figure E.18: The deconvolution of S 2p of used Mo2C/APC catalysts: (a) 2Mo2C/APC; (b) 5Mo2C/APC; (c) 10Mo2C/APC; (d) 5MoS2/APC.  291   Figure E.19: A correlation of Mo, C, and S from HADDF-STEM-EDX mapping with two selected areas.   292    Figure E.20: EDX mapping for two selected areas of used 2Mo2C/APC catalyst.   293   Top view Side view    Figure E.21: Terminal positions of Mo2C (101) surface used in DFT calculation. (Terminal Mo site: Mo-t1; terminal C sites: C-t1 and C-t2, respectively)  294  XS X=1 X=2 X=3 Top view    Side view    Figure E.22: Diagram of different S atoms replaced Mo2C (101) surface for DBT adsorption energy calculation (Top and side views). 295            Figure E.23: Potential energy during dissociative adsorption of H2 on three surfaces: (a) clean Mo2C (101) surface; (b) S adsorbed Mo2C (101) surface; (c) S replaced Mo2C (101) surface.  296  Orientation name DBT-V1 DBT-V2 Top view  & Side view     Gibbs Free Adsorption Energy (eV) -0.94 -1.01 Orientation name DBT-H1 DBT-H2 Top view  & Side view     Gibbs Free Adsorption Energy (eV) -1.53 -1.58 Table E.4: Gibbs free adsorption energy of DBT on Mo-t1 site of Mo2C (101) clean surface with different orientations. 297  Orientation name DBT-H1 DBT-V1 Top view  & Side view     Gibbs Free Adsorption Energy (eV) -1.21 -0.99 Table E.5: Gibbs free adsorption energy of DBT on Mo-t1 site of S adsorbed Mo2C (101) surface with different orientations.  298  E.5 Supplementary Information for Chapter 6  Figure E.24: Designed experiments for N effect study of 10Mo2C/APC catalyst at 350 oC and 4.1 MPa: (a) naphthalene hydrogenation reaction at 250 oC and LHSV = 4 h-1; (b) Part I-naphthalene hydrogenation, Part II-HDN of carbazole at 350 oC and 4.1 MPa, Part III-naphthalene hydrogenation; (c) Part I-naphthalene hydrogenation; Part II-NH3/H2 treatment at 350 oC and 4.1 MPa; Part III-Naphthalene hydrogenation.  The N could seriously affect the Mo2C/APC catalyst as reported in Chapter 6. Several experiments were designed to see how does the catalyst has been modified in the presence of N. Figure E.24 (a) has clearly shown the strong hydrogenation ability of naphthalene on 299  10Mo2C/APC catalyst with 93% conversion at 250 oC and 4.0 MPa. In Figure E.24 (b), the experiment has been designed into three phases, first with a hydrogenation of naphthalene, then HDN of carbazole and finally a hydrogenation of naphthalene again. After the 1st phase of hydrogenation reaction, it found that the catalyst is still active for HDN reaction (almost 100% conversion), indicative of the unchanged catalyst surface. Followed by HDN, another hydrogenation of naphthalene was conducted, where a huge drop on naphthalene conversion was observed in phase III. It suggests that most of the hydrogenation ability is lost after HDN reaction. In order to see if this change is caused by the released product-NH3, NH3/H2 has been introduced in the 2nd phase of the experiment as shown in Figure E.24 (c). An equivalent amount of NH3 (150 ppm) was used here with similar treatment time. It is interested to see that the conversion of naphthalene could fall to 65%, which is not severe than HDN of carbazole reaction. This phenomenon indicates the effect of N as NH3, but it is not all the sources for this deactivation.   300  Appendix F  Error Analysis and Repeatability  F.1 Carbon Balance The carbon balance should be 100% for each experiment. However, duo to sampling and experimental errors (GC-MS, operation etc.), it is hard to get a 100% carbon balance. It needs to be monitored during the data processing to see the accuracy of the experiment. Table F.1 presents the reactant and products concentration of Ni0.19Mo2C/AC-550 and Ni0.38Mo2C/AC-550 catalysts over time for the reaction of HDS of DBT in batch reactor as reported in Chapter 3.   Table F.1: Reactant (DBT) and products concentration for HDS of dibenzothiophene over Ni0.19Mo2C/AC-550 and Ni0.38Mo2C/AC-550 catalysts. Ni0.19Mo2C/AC-550 catalyst      Time (h) CHB(mol/L) BPh (mol/L) THDBT (mol/L) DBT (mol/L) Total mole of C (mol) Total mole of C in Feed (mol) Carbon balance error 1 7.5116E-04 1.3937E-02 1.0873E-03 8.1175E-02 1.1634E-01 1.1657E-01 0.20% 3 3.1119E-03 2.6506E-02 2.0758E-03 6.3547E-02 1.1429E-01 1.1657E-01 1.96% 5 4.2710E-03 3.3690E-02 2.6117E-03 5.6010E-02 1.1590E-01 1.1657E-01 0.58%  Ni0.38Mo2C/AC-550 catalyst      Time (h) CHB(mol/L) BPh (mol/L) THDBT (mol/L) DBT (mol/L) Total mole of C (mol) Total mole of C in Feed (mol) Carbon balance error 1 8.0949E-04 1.3931E-02 1.4353E-03 7.7898E-02 1.1289E-01 1.1657E-01 3.16% 3 2.2574E-03 2.3836E-02 2.2044E-03 6.1945E-02 1.0829E-01 1.1657E-01 7.11% 5 6.1666E-03 3.5773E-02 2.3100E-03 4.5395E-02 1.0757E-01 1.1657E-01 7.72%  301  F.2 Petcoke Activation Repeatability Table F.2: Petroleum coke activation results from different batches. Batch No. Ratio of KOH to Petcoke Mixture (KOH + Petcoke) Weight in boat (g) Petcoke in ceramic boat (g) Recoverd Petcoke (g) Burn-off % Batch1 2.96 13.51 3.41 2.25 34.06  Batch2 2.96 14.30 3.61 2.36 34.58  Batch3 3.00 13.18 3.29 2.08 36.87  Batch4 3.00 14.62 3.65 2.45 32.97  Batch5 2.97 13.97 3.52 2.21 37.28  Batch6 2.97 14.47 3.65 2.38 34.61  Batch7 2.97 13.70 3.45 2.20 36.22  Batch8 2.97 13.78 3.47 2.17 37.38  Batch9 3.00 13.39 3.35 2.13 36.34  Batch10 3.00 14.06 3.52 2.39 32.17  Batch11 3.00 14.38 3.60 2.19 39.06  Batch12 3.00 13.72 3.43 2.60 24.12  Batch13 3.00 14.36 3.59 2.53 29.51  Batch14 3.00 13.20 3.30 2.53 23.29  Batch15 3.00 13.54 3.39 2.15 36.42  Batch16 3.00 14.26 3.57 2.25 36.85  Average 2.31 33.86 Std. dev. (SD) 0.16 4.46  F.3 Reaction Repeatability To observe the repeatability of the experiments, several experiments were repeated to quantify the error associated with the experiments. 302  (1) Batch reactor  Here, Ni0.19Mo2C/AC-550 and Ni0.38Mo2C/AC-550 catalysts tested in HDS of DBT were put here as an example. The obtained rate constants were reported in Table F.2.  Table F.3: Kinetic parameters for HDS of dibenzothiophene in batch reactor. Catalysts Run No. k1 (mL/min.gmetal) k2 (mL/min.gmetal) k3 (mL/min.gmetal) Ni0.19Mo2C/AC-550 1 2.1956E-01 7.8671E-01 4.7529E+00  2 2.7313E-01 7.8574E-01 3.6545E+00 Average  2.4635E-01 7.8622E-01 4.2037E+00 Std.dev. (SD)  2.6787E-02 4.8463E-04 5.4921E-01  Ni0.38Mo2C/AC-550 1 2.9860E-01 8.1009E-01 4.8360E+00  2 2.4624E-01 7.1025E-01 3.4768E+00 Average  2.7242E-01 7.6017E-01 4.1564E+00 Std.dev. (SD)  2.6179E-02 4.9917E-02 6.7962E-01  (2) Fixed bed reactor  2% Mo2C/APC catalyst was used in the HDS reaction of DBT (0.2 wt%, eqv. 340 ppm S) in fixed bed reactor. The reaction was conducted at 350 oC, 600 psi, LHSV = 4 h-1 with 0.77 g catalyst loading.  Table F.4: Catalyst properties of 2%Mo2C/APC catalyst.  Catalyst Exp. No. Burn-off rate of the catalyst, % 2%Mo2C/APC T8 29.66 2%Mo2C/APC T13 28.43 303  The experimental data of T8 is reported as follows:  Table F.5: Experimental data of 2%Mo2C/APC in HDS of DBT at 350 oC and 4.1 MPa from experiment T8.  TOS (min) Conversion of DBT (%) Selectivity (%) THDBT BPh CHB BCH 30 99.56 0.00 31.25 53.05 15.70 63 99.65 0.00 41.03 51.68 7.29 90 99.63 0.00 61.34 33.47 5.19 120 99.40 0.00 74.44 21.37 4.19 152 99.27 0.17 77.34 18.45 4.04 177 98.97 0.48 77.18 18.69 3.66 213 98.40 1.02 76.50 18.18 4.30 244 97.77 1.44 76.36 18.20 3.99 268 96.24 1.60 76.65 17.85 3.89 306 95.42 1.71 76.35 17.90 4.03 332 95.22 1.73 76.24 18.35 3.68 362 95.18 1.72 76.27 17.92 4.09 389 94.93 1.79 76.45 17.89 3.87 418 95.01 1.79 76.07 18.39 3.76 444 94.68 1.81 75.83 18.04 4.32 476 94.92 1.79 75.96 18.12 4.14 Ave. value after 300 min  97.00 1.02 73.62 20.43 4.93   304  The experimental data of T13 is reported in Table F.6:  Table F.6: Experimental data of 2%Mo2C/APC in HDS of DBT at 350 oC and 4.1 MPa from experiment T13. TOS (min) Conversion of DBT (%) Selectivity (%) THDBT BPh CHB BCH 30 98.79 1.83 34.31 48.61 15.25 58 98.95 0.20 31.13 57.95 10.71 92 99.76 0.08 41.01 51.13 7.78 123 99.89 0.00 54.54 38.92 6.54 154 99.91 0.04 69.91 24.40 5.64 193 99.93 0.06 73.40 21.22 5.32 227 99.80 0.40 73.60 20.72 5.29 255 98.31 0.79 74.02 20.69 4.50 288 97.54 0.94 73.31 20.53 5.22 312 97.16 1.01 73.44 20.17 5.39 346 97.18 1.00 73.46 20.51 5.03 376 97.11 1.01 73.60 20.38 5.00 404 96.84 1.05 73.96 20.32 4.68 434 97.02 1.04 73.37 20.54 5.05 467 96.98 1.02 73.64 20.61 4.73 496 96.71 1.05 73.84 20.46 4.65 Ave. value after 300 min  95.05 1.76 76.17 18.09 3.98  97.00a 1.02a 73.62a 20.43a 4.93a Ave.b 96.03 1.39 74.89 19.26 4.46 Std. dev. (SD) b 0.98 0.37 1.27 1.17 0.47 a. These data were adopted from Table F.4. b. These values were calculated by comparing T8 and T13.   305    Figure F.1: DBT conversion and products selectivity for 2%Mo2C/APC catalyst as a function of time on stream in two trails (T8 and T13).  From Table F.5, Table F.6 and Figure F.1, it can be observed that the experiment is repeatable.   306  F.4 Statistical Analysis of Kinetic Model One way ANOVA analysis and F statistic test were conducted to see if the kinetic model fitted well with the experimental data. The probability (P) was set at 0.05. Both the experimental and kinetic model fitted data of Ni0.19Mo2C/AC-550 catalyst in HDS of DBT at 350 oC and initial pressure of 2.1 MPa were reported in Table F.7.  Table F.7: Comparation between experimental and kinetic model fitted data of different products and reactant in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst.  Time (h) CHB (mol/L) BPh (mol/L) THDBT (mol/L) DBT (mol/L) Experimental Model Exp. Model Exp. Model Exp. Model 1 7.51E-04 4.00E-04 1.39E-02 1.00E-02 1.09E-03 2.30E-03 8.12E-02 9.10E-02 3 3.11E-03 2.73E-03 2.65E-02 2.65E-02 2.08E-03 4.43E-03 6.35E-02 7.00E-02 5 4.27E-03 5.80E-03 3.37E-02 3.92E-02 2.61E-03 4.79E-03 5.60E-02 5.39E-02 Ave. 2.71E-03 2.98E-03 2.47E-02 2.52E-03 1.92E-03 3.84E-03 6.69E-02 7.16E-02  The used formulas are summarized in Table F.8. Table F.8: Summary of ANOVA used calculation formula. Symbol Definition Formula SSWG Sum of squares within groups SSWG = ∑ (𝑋𝑖2𝑚𝑖=1 + 𝑥𝑖2) −(∑ 𝑋𝑖)2+(∑ 𝑥𝑖)2𝑚𝑖=1𝑚𝑖=1𝑚 SST Total sum of squares  SST = ∑ (𝑋𝑖2𝑚𝑖=1 + 𝑥𝑖2) −(∑ 𝑋𝑖+𝑥𝑖)2𝑚𝑖=1𝑚, m is the number of data SSBG Sum of squares between groups SSBG = SST - SSWG dfWG Degrees of freedom within groups dfWG = ∑ (𝑚 − 1)𝑖𝑧𝑖=1 , z means the set of data dfBG Degrees of freedom between groups dfWG = z - 1 MSWG Mean square within groups MSWG = 𝑆𝑆𝑊𝐺𝑑𝑓𝑊𝐺 MSBG Mean square between groups MSBG = 𝑆𝑆𝐵𝐺𝑑𝑓𝐵𝐺 FANOVA F value by ANOVA FANOVA = 𝑀𝑆𝐵𝐺𝑀𝑆𝑊𝐺  307  The ANOVA analysis of the experimental and model data of different chemicals (DBT, CHB, BPh and THDBT) are reported in the Table F.9, Table F.10, Table F.11 and Table F.12.   Table F.9: ANOVA analysis of DBT concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst.  DBT SS df MS FANOVA F(1,4) Between groups 3.35E-05 1 3.35E-05 1.31E-01 7.709 Within groups 1.02E-03 4 2.56E-04     Sum 1.06E-03 5       Note: F(1,4), P=0.05 was obtained from reference [192].   Table F.10: ANOVA analysis of CHB concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. CHB SS df MS FANOVA F(1,4) Between groups 1.06E-07 1 1.06E-07 2.01E-02 7.709 Within groups 2.11E-05 4 5.28E-06     Sum 2.12E-05 5        Table F.11: ANOVA analysis of BPh concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. BPh SS df MS FANOVA F(1,4) Between groups 4.14E-07 1 4.14E-07 2.64E-03 7.709 Within groups 6.28E-04 4 1.57E-04     Sum 6.29E-04 5       308   Table F.12: ANOVA analysis of THDBT concentration data in HDS of DBT by Ni0.19Mo2C/AC-550 catalyst. THDBT SS df MS FANOVA F(1,4) Between groups 5.49E-06 1 5.49E-06 4.58E+00 7.709 Within groups 4.79E-06 4 1.20E-06     Sum 1.03E-05 5        From all these calculation, it found that all the FANOVA < F(1,4, P=0.05), which means there is no significant difference between these two set of data, the model fitted data is fitted well with experimental data.  309  F.5 Characterization Repeatability The sample standard deviation (SD) is used to describe the scatter of the obtained data. It can be calculated by Eq. F-1. Also, the mean value (µ) of the replicate measurements can be calculated as given in Eq. F-2. In the present study, both of them have been applied in various characterization data of BET, CO uptake, CHNS, EDX mapping, TOF-SIMS and GC/FID for in-situ CHR process as shown in the following tables. 𝑺𝑫 = √∑ (𝒙𝒊−𝝁)𝟐𝒏𝒊=𝟏𝒏−𝟏         (Eq. F-1) µ =∑ 𝒙𝒊𝒏𝒊=𝟏𝒏          (Eq. F-2) %𝐄𝐫𝐫𝐨𝐫 =𝐒𝐃µ          (Eq. F-3) where 𝑥𝑖 is the sample variable, µ is the average number of the samples’ value, and n is the number of the samples.   Table F.13: Calculated error associated with the physical properties test of activated petroleum coke (APC). Sample Repeat No. Surface area based on BET model (m2/g) Total pore volume  (cm3/g) Ave. pore size  (nm) Activated petcoke 1 2240 0.99 1.8  2 2385 1.14 1.9  3 2343 1.12 1.9 Mean value (µ)  2322 1.08 1.9 Std. dev. (SD)  61 0.07 0.1 % Error  2.6 6.5 5.3  310   Table F.14: Calculated error associated with the CO uptake. Sample Repeat No. CO uptake (µmol/gcata.) 10%Mo2C/AC-600 fresh catalyst synthesized in-situ 1 16.7 2 20.7 Mean value (µ)  18.7 Std. dev. (SD)  2.0 % Error  10.7 10%Mo2C/APC-650 passivated catalyst  1 27.6 2 22.4 Mean value (µ)  25.0 Std. dev. (SD)  2.6 % Error  10.4  Table F.15: Calculated error associated with CHNS analysis for raw petroleum coke. Element Repeated No. Carbon Hydrogen Nitrogen Sulfur Oxygen Raw petcoke 1 83.26 3.59 1.97 6.59 4.59  2 83.28 3.61 2.07 6.57 4.47 Mean value (µ)  83.27 3.60 2.02 6.58 4.53 Std. dev. (SD)  0.01 0.01 0.05 0.01 0.06 % Error  0.012 0.28 2.48 0.15 1.32  311   Table F.16: Calculated error associated with EDX-mappings for raw petroleum coke and APC_800.  Elements Sample C O S Al Si PC-Raw Petroleum Coke PC1 90.06 6.48 3.03 0.18 0.24 PC2 89.21 5.44 4.51 0.24 0.38 PC3 90.60 5.82 3.19 0.18 0.21 Ave. 89.96 5.91 3.58 0.20 0.28 Std. Dev. 0.57 0.43 0.66 0.03 0.07 % Error 0.63 7.28 18.44 15.00 25.00 APC-Activated Petroleum Coke at 800 oC (APC_800) APC1 92.87 6.92 0.00 0.22 0.00 APC2 93.86 6.14 0.00 0.00 0.00 APC3 92.40 7.27 0.00 0.33 0.00 Ave. 93.04 6.78 0.00 0.18 0.00 Std. Dev. 0.61 0.47 0.00 0.05 0.00 % Error 0.66 6.93 0.00 27.78 0.00    312   Table F.17: Calculated error associated with TOF-SIMS analysis for used 10%Mo2C/AC-650 and Ni0.19Mo2C/AC-600 catalysts after HDS of DBT in batch reactor (Chapter 3).  Note: Ni: Mo=0.19 stands for Ni0.19Mo2C/AC-600 catalyst; Ni: Mo=0.00 stands for 10%Mo2C/AC-650 catalyst. 130 146 162 178 194MoS MoOS MoO2S MoO3S MoO2S2MoO2 MoO3 MoS2 MoOS2 MoS3Ni:Mo=0.19 651 5242 2547 416 585 3541990 9441 76159 2.15E-02 2.67E-03 6.90 55.52 26.98 4.41 6.20Ni:Mo=0.19 617 5099 2376 398 523 3509877 9013 79214 2.26E-02 2.57E-03 6.85 56.57 26.36 4.42 5.80Ni:Mo=0.19 598 4663 2163 375 510 3506899 8309 84121 2.40E-02 2.37E-03 7.20 56.12 26.03 4.51 6.14AVE 622 5001 2362 396 539 3519589 8921 79831 2.27E-02 2.53E-03 6.98 56.07 26.46 4.45 6.05Std. Dev. 22 246 157 17 33 15887 467 3280 1.02E-03 1.23E-04 0.16 0.43 0.39 0.05 0.17Ni:Mo=0.00 594 2356 1092 326 183 3840918 4551 76854 2.00E-02 1.18E-03 13.05 51.77 23.99 7.16 4.02Ni:Mo=0.00 767 2825 1294 340 181 4219830 5407 73420 1.74E-02 1.28E-03 14.19 52.25 23.93 6.29 3.35Ni:Mo=0.00 602 2643 1245 318 190 3536102 4998 70630 2.00E-02 1.41E-03 12.04 52.88 24.91 6.36 3.80AVE 654 2608 1210 328 185 3865617 4985 73635 1.91E-02 1.29E-03 13.09 52.30 24.28 6.60 3.72Std. Dev. 80 193 86 9 4 279677 350 2545 1.22E-03 9.37E-05 0.87 0.46 0.45 0.40 0.28Normalized MoS (m/z=130) (RI'=100MoS/ƩMoOxSy-)Normalized MoOS (m/z=146) (RI'=100MoOS/ƩMoOxSy-)Normalized MoO2S or MoS2 (m/z=162) (RI'=100MoO2S/ƩMoOxSy-)Normalized MoO3S or MoOS2 (m/z=178) (RI'=100MoO3S/ƩMoOxSy-)Normalized MoO2S2 or MoS3 (m/z=194) (RI'=100MoO2S2/ƩMoOxSy-)SampleNormalized  Identified total MoS species( RI=ƩMoOxSy-/Ʃ I+)Total ion        (Ʃ I+)Total Identified MoS species (ƩMoOxSy-)S (ƩS-)Normalized S (ƩS-/Ʃ I+)313  Table F.18: Calculated error associated with in-situ exit gas analysis of 2%Mo2C/APC-700 by carbothermal hydrogen reduction. Temperature (oC) Exp1 Exp2 Mean value (µ) Std.dev. (SD) %Error Time interval (min) CH4 mol% Time interval (min) CH4 mol% 91 66 0 66 0 0 0 ― 265 240 0 240 0 0 0 ― 340 315 0 315 0 0 0 ― 421 396 0.105056 396 0.104056 0.104556 0.0005 0.478213 464 439 0.164288 439 0.165288 0.164788 0.0005 0.30342 487 462 0.224928 462 0.223928 0.224428 0.0005 0.222789 558 533 0.502131 535 0.502238 0.502185 5.37E-05 0.010699 588 563 0.62465 566 0.653348 0.638999 0.014349 2.245537 626 601 1.04215 605 1.104975 1.073563 0.031413 2.926027 652 627 1.615816 631 1.579067 1.597441 0.018374 1.150245 678 653 2.463443 655 2.479983 2.471713 0.00827 0.334587 700 675 3.080209 670 2.9567 3.018454 0.061754 2.045897 700 700 3.017785 699 3.0249 3.021343 0.003557 0.117739 700 720 2.921921 726 2.943344 2.932633 0.010712 0.36526 700 736 2.864935 740 2.8954 2.880167 0.015233 0.528876 700 750 2.834814 755 2.8644 2.849607 0.014793 0.519122 700 765 2.802529 770 2.7095 2.756014 0.046514 1.687738    314  Appendix G  Mass Transfers and Heat Transfer Effects In order to show if the experiments obtained in fixed-bed reactor have been controlled by mass transfer, several calculations have been applied to illustrate the external and internal mass diffusion phenomenon. Also, a heat transfer calculation was conducted by Mears’ Criterion. As an example, 10%Mo2C/APC catalyst in HDS of DBT (0.2 wt%) at 350 oC and P=600 psi has been put here. The details of the catalyst bed, reaction condition, and catalyst properties are shown in Table G.1.  315  Table G.1: The details of catalyst bed, catalyst physical properties and related kinetic parameters. Symbol Definition Calculation formula/Source  Value Unit -rDBT(obs) Observed reaction rate Calculated from experimental data 1.81E-10 [kmol/gcat·s] k1 Stabilized kinetic parameter Calculated from experimental data 1.67E-02 [s-1] Vc Loaded catalyst volume Measured 0.50  [mL] Vsic Loaded inert volume Measured 2.00 [mL] Vbed Total volume of catalyst bed Vbed=Vc+Vsic 2.50 [mL] Lbed Length of the catalyst bed Measured 4.20 [cm] ρbsic catalyst bed density with SiC and catalyst ρbsic=mtotal/Vbed 1.4680  [g/cm3] ρc Catalyst density Calculated by 1.48 [g/cm3] x 80% + 8.90 [g/cm3] x 20%, where ρAC =1.48 [g/cm3]; ρMo2C = 8.90 [g/cm3] 2.9640  [g/cm3] ρsic SiC density — 3.2100  [g/cm3] ρsolid Solid catalyst density ρsolid=(ρcVc+ρsicVsic)/(Vc+Vsic) 3.1608  [g/cm3] mcat Mass of loaded catalyst Measured 0.16 [g] msic Mass of SiC Measured 3.51  [g] mtotal Mass of loaded catalyst bed mtotal= mcat + msic 3.67 [g] ø Porosity or void fraction of packed bed  ø=1-ρbsic/ρsolid 0.5356  — ρb Bulk density of catalyst bed ρb=ρc*(1-ø) 1.3766  [g/cm3] dp Catalyst particle diameter Measured 1.35E-04 [m] R Catalyst particle radius R=dp/2 6.75E-05 [m] n reaction order — 1 — Vo Total pore volume of the catalyst Measured  1.16 [cm3/g] Scat. Surface area of the catalyst Measured 1651 [m2/g] rpore Pore radii of the catalyst Measured 1.40521E-07 [cm] øp Catalyst particle porosity  øp=Vo/(Vo+1/ρc) 0.7747 — τ Tortuosity factor — 3 — σc Constriction factor — 0.8 — γ Shape factor  γ = 𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙 𝑠𝑢𝑟𝑓𝑎𝑐𝑒 𝑎𝑟𝑒𝑎𝜋𝑑𝑝2 1.00E+00 —  316  G.1 External Mass Transfer Effect  The details of feed properties were calculated by ASPEN Plus with Wilson method as reported in Table G.2.  Table G.2: The details of reaction conditions and feed properties as calculated from Aspen Plus.  Symbol Definition Calculation formula/Source  Value Unit Trxn Reaction temperature — 623.15 [k] Prxn Reaction pressure — 40.83 [atm] MDBT Mole weight of DBT — 184.256 [g/mol] MH2 Mole weight of H2 — 2.016 [g/mol] MDecalin Mole weight of Decalin — 138.25 [g/mol] TDBT,c Critical point temperature of DBT — 897 [k] ṼDBT,c Critical volume of DBT — 525 [mL/g-mol] ρmix Density of feed mixturedensity at 350 oC and 600 psi 0.1667 cc/min 0.2 wt% DBT in decalin, 100 cc/min H2 2.07 [kg/m3] µmix Dynamic viscosity of the mixture — 1.51E-05 [kg/m.s] νmix kinetic viscosity of mixture — 7.32E-06 [m2/s] Mmix Feed molecular weight — 2.24E+00 [g/mol] PH2 Partial pressure of H2 — 41.93 [atm] PDBT Partial pressure of DBT — 1.08E-04 [atm] PDecalin Partial pressure of Decalin — 6.98E-02 [atm] Ptotal Total pressure in the system at 350 oC and 600 psi — 41.9999 [atm] CH2 Bulk gas concentration of H2 at 350 oC and 600 psi — 1.0333 [kmol/m3] CDBT Bulk gas concentration of DBT at 350 oC and 600 psi — 2.89E-08 [kmol/m3] 317  Table G.3: The details of external mass transfer calculation by Mears criterion. Symbol Definition Calculation formula/Source  Value Unit U Superficial gas velocity U=𝛾𝑚𝑖𝑥/𝐴 2.85E-02 [m/s] 𝛾𝑚𝑖𝑥  Volumatic flow rate 100 [cc/min H2] + 0.1667 [cc/min] Feed 1.67E-06 [cm3/s] A Cross section area of the reactor A=π𝑟𝑏𝑒𝑑2  (𝛾𝑏𝑒𝑑 = 0.432 𝑐𝑚) 5.86E05 [m2] σ DBT Lennard-Jones parameters for DBT/characteristic length σDBT=0.841*ṼDBT,c1/3 6.7845 [Å] σ H2 Lennard-Jones parameters for H2/characteristic length — 2.915 [Å] σ DBT-H2 Lennard-Jones parameters for DBT-H2/characteristic length σ DBT-H2 = ½ (σ DBT+ σ H2) 4.8497 [Å] εDBT/ĸ Lennard-Jones parameters for DBT/Boltzmann's constant εDBT/ĸ=0.77TDBT,C 690.69 [k] εH2/ĸ Lennard-Jones parameters for H2/Boltzmann's constant — 38 [k] εDBT-H2/ĸ Lennard-Jones parameters for DBT-H2/Boltzmann's constant 𝜀𝐷𝐵𝑇−𝐻2ĸ=√ε𝐷𝐵𝑇−𝐻2ĸ×ε𝐻2ĸ 162.00 [k] T* Dimensionless temperature T*=ĸTrxn/εDBT-H2 3.8464 [k] ΩD, DBT-H2 Collision integral, calculated by ignore the last two terms 𝛺𝐷,𝐷𝐵𝑇−𝐻2 =1.06036𝑇∗0.15610+0.19300exp (0.47635𝑇∗ 0.8902 — DDBT-H2 Binary bulk phase diffusivity 𝐷𝐶𝐻4−𝐻𝑒=0.0018583𝑃𝑟𝑥𝑛𝜎𝐷𝐵𝑇−𝐻22 𝛺𝐷,𝐷𝐵𝑇−𝐻2√𝑇3(1𝑀𝐷𝐵𝑇+1𝑀𝐻2) 2.39E-06 [m2/s] Sc Schmidt number  Sc=νmix/DDBT-H2 3.06E+00 — Sh Sherwood number  Sh=2+0.6Re1/2Sc1/3 2.93E+00 — Re Reynolds number Re = 𝑈𝜌𝑑𝑝𝜇 5.26E-01 — Re’ Reynolds number considering void fraction Re’= 𝑅𝑒(1−∅)𝛾 1.13E+00 — kc Mass transfer coefficient kc=𝑆ℎ×𝐷𝐷𝐵𝑇−𝐻2𝑑𝑃 5.19E-02 [m/s] MC Mears’ criterion for external difussion — 1.12E-11 —  318  Based on Mears’ criterion [193], the calculated value is accorded with −𝑟𝐷𝐵𝑇(𝑂𝑏𝑠)𝜌𝑏𝑅𝑛𝑘𝐶𝐶𝐷𝐵𝑇 ≪ 0.15, thus the external mass transfer effects can be neglected. 319  G.2 Internal Mass Transfer Effect  Table G.4: The details of internal mass transfer by Weisz-Prater criterion. Symbol Definition Calculation formula/Source  Value Unit DDBT-H2 Binary bulk phase diffusivity 𝐷𝐶𝐻4−𝐻𝑒=0.0018583𝑃𝑟𝑥𝑛𝜎𝐷𝐵𝑇−𝐻22 𝛺𝐷,𝐷𝐵𝑇−𝐻2√𝑇3(1𝑀𝐷𝐵𝑇+1𝑀𝐻2) 2.39E-06 [m2/s] Deff, DBT-H2 Effective diffusivity 𝐷𝑒𝑓𝑓,𝐷𝐵𝑇−𝐻2 =𝐷𝐷𝐵𝑇−𝐻2∅𝑃𝜎𝐶𝜏 4.95E-07 [m2/s] Dknudsen Knudsen diffusivity 𝐷𝑘𝑛𝑢𝑑𝑠𝑒𝑛 =2𝑟𝑝𝑜𝑟𝑒3√(8 ∙ 𝑅𝑔) ∙ 𝑇𝜋 ∙ 𝑀𝑚𝑖𝑥 2.27E-06 [m2/s] Deff,knudsen Effective knudsen diffusivity 𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑𝑠𝑒𝑛 =𝐷𝑘𝑛𝑢𝑑𝑠𝑒𝑛∅𝑃𝜎𝐶𝜏 4.69E-07 [m2/s] Deff, rxn Effective diffusivity in this reaction 1𝐷𝑒𝑓𝑓,𝑟𝑥𝑛=1𝐷𝑒𝑓𝑓,𝐷𝐵𝑇−𝐻2+1𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑𝑠𝑒𝑛 2.41E-07 [m2/s] ø1 Thiele modulus for 1st order reaction ∅1 = 𝑅√𝑘1𝐷𝑒𝑓𝑓,𝑟𝑥𝑛 1.78E-02 — η Internal effectiveness factor η =3∅12 (∅1𝑐𝑜𝑡ℎ∅1 − 1) 1.00E+00 — Cwp  Weisz-Prater criterion — 3.17E-04 —  Based on Weisz-Prater criterion [193], the calculated value of 𝐶𝑊𝑃 = η∅12=−𝑟𝐷𝐵𝑇(𝑜𝑏𝑠)𝜌𝑐𝑅2𝐷𝑒𝑓𝑓,𝑟𝑥𝑛𝐶𝐴𝑆 ≪1, thus there are no internal diffusion limitations.   320  G.3 Heat Transfer Effect  Table G.5: The details of heat transfer calculation by Mears Criterion.  Symbol Definition Calculation formula/Source  Value Unit kt Thermal conductivity calculated by semiempirical method for polyatomic gases kt=(Ĉ𝑝,(𝐻2)+5𝑅4𝑀𝐻2)µ𝑚𝑖𝑥 2.98E-01 [W/(m.K)] Ĉp(H2) Heat capacity of H2 at Trxn — 14500 [J/Kg·K] ΔHrxn Heat of reaction Primary reaction: C12H8S + 2H2  C12H10 + H2S -45.5 KJ/mol Ea Activation energy Estimated based on literature 150000 [J/mol] Pt Prandtl number  Pr=µmix.Ĉp/kt 7.38E-01 — Nu Nusselt number Nu=2+0.6Re1/2Pr1/3 2.58E+00 — h Heat transfer coefficient h=𝑁𝑢∙𝑘𝑡𝑑𝑝 5.68E+03 [W/(m2·K)] Rg Gas constant — 8.314 [J/(mol·k)] MC’ Mears’ criterion for isothermal operation — 6.25E-18 —  Based on Mears’ criterion [193, 194], the calculated value of |−∆𝐻𝑟𝑥𝑛(−𝑟𝐷𝐵𝑇,𝑜𝑏𝑠)𝜌𝑏𝑅𝐸ℎ𝑇2𝑅𝑔| ≪ 0.15, thus there is no temperature gradient during reaction.   321  G.4 Precheck of Fixed-bed Reactor Operating Condition In order to ensure the kinetic data are free from the influence of unwanted transport effects, including wall effect and axial dispersion, some criteria have been applied to evaluate the designed reactor bed [195].   Table G.6: The geometry parameters for prechecking of fixed-bed reactor operating condition. Symbol Definition Value Unit 𝑑𝑝 Catalyst particle diameter 1.35E-04 m 𝐿𝑏𝑒𝑑 Length of catalyst bed 4.20E-02 m 𝐷𝑏𝑒𝑑 Fixed-bed tube diameter 8.64E-01 m 𝐷𝑏𝑒𝑑𝑑𝑝 ― 6.40E+03 > 10 ― 𝐿𝑏𝑒𝑑𝑑𝑝 ― 3.11E+02 > 50 ―  From Table G.6, it find that the value of 𝐷𝑏𝑒𝑑𝑑𝑝> 10, suggesting that the reactor wall effect can be eliminated. The value of 𝐿𝑏𝑒𝑑𝑑𝑝 > 50 indicates that the axial effect can be ignored as well.     322  G.5 External and Internal Mass Transfers in Batch Reactor  All the kinetic data reported in Chapter 2 and 3 were obtained from a batch reactor operation after excluding external and internal mass transfer’s effects. Ni0.09Mo2C/AC-550 catalyst was put here as an example.   Table G.7: A detailed list of external and internal mass transfer coefficient calculation for Ni0.09Mo2C/AC-550 in HDS of DBT in batch reactor. Symbol Definition Value Unit 𝑘𝐿𝑎 Liquid-side mass transfer coefficienta 4.37E-02 s-1 𝑘𝐷𝐵𝑇 Kinetic parameter 1.77E-05 s-1 R Catalyst particle radius 6.75E-05 m ø1 Thiele modulus for 1st order reaction 4.04E-04 ― Deff, DBT-H2 Effective diffusivity 4.95E-07 m2/s η Internal effectiveness factor 1.00E+00 ― Ω Overall effectiveness factorb 1.00E+00 ― a. This number was adopted from [196] since the operating system was the same as the previous group member’s. b. 𝛺 =𝜂(1+𝜂𝑘𝐷𝐵𝑇𝑘𝐿𝑎)  As shown in Table G.7, kLa >> kDBT and η=Ω=1, thus the external and internal mass transfer can be neglected under present conditions.  323  Appendix H  Additional Calculation and Experimental Data H.1 Carbon Efficiency Calculation for Mo2C Formation Take 10%Mo2C/APC-700 as an example (The carbon burn-off% is 30.35%): For 1 g of 10%AHM/APC precursor, The mole of Mo is 0.1 𝑔𝑀𝑜95.94 𝑔/𝑚𝑜𝑙= 1.04 × 10−3 𝑚𝑜𝑙 The required amount of C for Mo2C formation = 1.04×10−3𝑚𝑜𝑙2× 12𝑔𝑚𝑜𝑙= 6.25 × 10−3 𝑔 The lost C calculated from burn-off% = 0.9 𝑔 × 30.35% = 0.27 𝑔 The carbon efficiency is 6.25 ×10−30.27= 2.32%  H.2 Comparation between 10%MoS2/Al2O3, 10%MoS2/AC, and 10%Mo2C/AC Catalysts The performance of 10%MoS2/Al2O3, 10%MoS2/AC, and 10%Mo2C/AC catalysts were tested in HDS of DBT at 350 oC with an initial pressure of 2.1 MPa in a batch reactor. The DBT conversion and products selectivity were reported in Table H.1.  324   Table H.1: DBT conversion and products selectivity of 10%MoS2/Al2O3, 10%MoS2/AC, and 10%Mo2C/AC catalysts in HDS of DBT at 350 oC and initial pressure of 2.1 MPa. Catalysts DBT Conv. (%) Time (h) Product selectivity (%) BPh THDBT CHB 10%MoS2/Al2O3 8.87 1  34.92  54.28  10.80  25.38 3  34.80  49.84  15.36  34.68 5  34.19  47.05  18.76   10%MoS2/AC 10.68 1 76.28 21.21 2.51 33.03 3 70.15 14.34 15.51 44.04 5 69.87 15.41 14.72  10%Mo2C/AC 14.87 1 83.69 11.55 4.76 27.68 3 82.67 13.15 4.18 40.31 5 78.91 14.25 6.71 325  H.3 Mesoporous Carbon Development by Ni-Mo2C on APC  Table H.2: Physical properties of APC supported Ni-Mo2C catalysts.  Sample BET Surface area (m2/g) BJH surface area (m2/g) Total Pore volume (cm3/g) t-plot micro pore volume (cm3/g) BJH pore volume (cm3/g) VBJH/ Vtotal ratio Ave Pore size (nm) BJH pore size (nm) Yield (%) APC_800  2413 447 1.17 0.78 0.30 25.99 1.9 2.7 66.14 APC_800-redH2 500-600C 2420 512 1.17 0.78 0.33 28.23 1.9 2.7 86.95 1%Ni/APC precursor 2180 402 1.05 0.71 0.27 25.98 1.9 2.7  N/A 1%Ni/APC-redH2 500-600 2394 434 1.15 0.76 0.29 25.33 1.9 2.7 90.16 1%Ni-10%Mo2C/APC  2014 911 1.52 0.41 1.07 70.54 3.0 4.7 64.47 2%Ni_10%Mo2C/APC  1690 995 1.89 0.23 1.68 88.62 4.5 6.7 38.04  In order to study the effect of Ni on pore development, APC_800 support with 1 wt% of Ni was prepared (recorded as 1%Ni/APC precursor). From Table H.2, it found that the CHR could bring a similar pore development effect with or without Ni at 600 oC (Compare sample APC_800-redH2 500-600C and 1%Ni/APC-redH2 500-600C). Ni alone couldn’t bring a big influence on pore development. However, the BJH pore volume was significant increased for 1%Ni-Mo2C/APC sample with account of 70% BJH porosity by 1 wt.% of Ni. It indicates that Ni may immerse into the structure of Mo2C, accelerating the C burn-off rate by generating more mesopores.  

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}]}"
                            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.24.1-0373612/manifest

Comment

Related Items