Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A two-step bio-oil upgrading study using carbon supported molybdenum carbide catalysts Liu, Shida 2019

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_september_liu_shida.pdf [ 11.8MB ]
Metadata
JSON: 24-1.0380462.json
JSON-LD: 24-1.0380462-ld.json
RDF/XML (Pretty): 24-1.0380462-rdf.xml
RDF/JSON: 24-1.0380462-rdf.json
Turtle: 24-1.0380462-turtle.txt
N-Triples: 24-1.0380462-rdf-ntriples.txt
Original Record: 24-1.0380462-source.json
Full Text
24-1.0380462-fulltext.txt
Citation
24-1.0380462.ris

Full Text

    A TWO-STEP BIO-OIL UPGRADING STUDY USING CARBON SUPPORTED MOLYBDENUM CARBIDE CATALYSTS  by  Shida Liu  M.Sc., China University of Petroleum (Beijing), 2013 B.Sc., Dalian University of 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)  August 2019 © Shida Liu, 2019 ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: A Two-step Bio-oil Upgrading Study using Carbon supported Molybdenum Carbide Catalysts  submitted by Shida Liu  in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Chemical and Biological Engineering  Examining Committee: Kevin J. Smith, Chemical and Biological Engineering Supervisor  Heather Trajano, Chemical and Biological Engineering Supervisory Committee Member  Chang Soo Kim, KIST-UBC Biorefinery on-site Laboratory Supervisory Committee Member Naoko Ellis, Chemical and Biological Engineering University Examiner Tom Troczynski, Material Engineering University Examiner   iii  Abstract A two-step bio-oil upgrading process has been investigated using carbon supported molybdenum carbide catalysts.  The waste materials, petcoke (PC) and biochar (BC) were activated to yield the Mo2C/APC and Mo2C/ABC catalysts (APC – activated petcoke and ABC – activated biochar). These catalysts presented very high (approximately 85%) direct deoxygenation (DDO) selectivity in the hydrodeoxygenation (HDO) reaction. Furthermore, the Mo2C/APC catalysts with low Mo loading (1 and 2 wt %) were acid washed in H2SO4 and both the Mo2C/APC catalysts and the acid treated Mo2C/APC catalysts were active for the esterification of acetic acid and 1-butanol. The hydrodeoxygenation (HDO) of 2-methoxyphenol over Pd, Ru and Mo2C catalysts supported on activated charcoal (AC) was compared. The overall 2-methoxyphenol consumption rate decreased in the order Pd > Ru > Mo2C due to Mo2C’s lower hydrogenation activity. Mo2C was the most efficient in terms of O-removal with minimal H2 consumption.  To enhance the hydrogenation activity of the Mo2C catalysts, promotion of a 10 % Mo2C/carbon catalyst with 1%Cu, 1%Ni and 1%Pd was assessed for the HDO of dibenzofuran (DBF). The addition of Ni and Pd decreased the temperature required for the removal of the oxygen layer from the Mo2C that is formed during catalyst passivation. This enhanced ability of the promoted catalyst to hydrogenate surface oxide species suggests that the same catalyst could improve the stability of the catalyst during HDO.    iv  Finally, the catalysts were assessed in a two-step bio-oil upgrading process that combined esterification and hydrodeoxygenation to improve the bio-oil fuel quality. In the first step, the Mo2C catalysts were shown to have sufficient acidity so as to catalyse esterification reactions at 180 °C and 10.3 MPa that stabilized the bio-oil. In the 2nd step of the upgrading process, the best Mo2C catalyst 1Ni-10Mo2C/ABC achieved 69% HDO with an overall carbon yield of 67% when operated at 350 °C and 15.0 MPa. The Ni-Mo2C catalysts are proven to be promising catalysts for the proposed two-step bio-oil upgrading.      v  Lay Summary The goal of this study was to establish an effective method to convert crude bio-oil to a useable fuel that can replace fossil fuels. A two-step process was developed that utilizes catalysts derived from molybdenum and waste carbon materials. The catalyst performance was optimized using model reactions before demonstrating the approach using crude bio-oil. The catalysts were active for bio-oil conversion to fuels and are much cheaper than conventional noble metal catalysts. The two-step process using the newly developed catalysts is a promising approach for the conversion of crude bio-oil into a useable fuel feedstock.  vi  Preface This Ph.D. dissertation has seven chapters. Chapters 2-4 have been published previously in the open literature. Chapters 5 and 6 are in preparation to be submitted for publication. The Ph.D. study was conducted by Shida Liu 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 Shida Liu under the direct supervision of Professor Kevin Smith.   The list of the publications included in this thesis is given below: 1. S. Liu., H. Wang, P. Neumann, C. S. Kim and K. J. Smith, “Esterification over an acid treated mesoporous carbon derived from petroleum coke,” ACS Omega (2019) 4(3): 6050-6058. A version of this manuscript is included in Chapter 2.  Shida Liu conducted all the experimental work associated with carbon supported Mo2C preparation, acid treatment, characterization, hydrodeoxygenation reactions, kinetic modeling, as well as data analysis and interpretation under the direct supervision of Professor Kevin. J. Smith. In addition, the preparation and writing of the manuscript were done by Shida Liu with final approval of Professor Kevin J. Smith. The N2 adsorption measurements were performed by Dr. Haiyan Wang. Patrick Neumann did some of the preliminary work on the chemical activation of petcoke.   vii  2. S. Liu., H. Wang, R. Putra, C. Kim, and K. J. Smith, “Impact of Carbon Properties on Mo2C/Carbon Catalysts for the Hydrodeoxygenation of 4-Methylphenol,” Energy & Fuels (2019). Article ASAP, DOI: 10.1021/acs.energyfuels.9b00531. A version of this manuscript is included in Chapter 3.  The catalysts preparation and characterization, HDO reaction of 4-MP, sample testing and analysis, and kinetic modeling were performed by Shida Liu under the direct supervision of Professor Kevin J. Smith. The writing of this manuscript was done by Shida Liu with the final approval of Professor Kevin J. Smith. The XPS were performed by Dr. Ken Wong from the Interfacial Analysis and Reactivity Laboratory at UBC. Dr. Haiyan Wang performed some of the GCMS measurements. Robertus Putra prepared the biochar used as carbon source. Dr. Chang Soo Kim contributed to the manuscript revision.   3. S. Liu., H. Wang, C. Kim, and K. J. Smith, “Hydrodeoxygenation of 2-Methoxyphenol over Ru, Pd, and Mo2C Catalysts Supported on Carbon,” Energy & Fuels (2017) 31(6): 6378-6388. A version of this manuscript is included in Chapter 4.  Shida Liu conducted the catalyst synthesis and characterization, HDO kinetic modeling, data collection and interpretation under the direct supervision of Professor Kevin J. Smith. The manuscript preparation and writing were done by Shida Liu with the final approval of Professor Kevin J. Smith. Dr. Haiyan Wang performed GC-MS measurement. Dr. Chang Soo Kim contributed to the manuscript revision. The XPS measurement was performed by Dr. Ken Wong from the Interfacial Analysis and Reactivity Laboratory at UBC.  viii   4. S. Liu, H. Wang, Mark Schmiβ, and K. J. Smith, “The Effect of Metal Promoters on Carbon Supported Mo2C Catalysts in Hydrodeoxygenation of Dibenzofuran”, in preparation. A version of this manuscript is included in Chapter 5.  The catalyst preparation, characterization, experimental design, DFT calculation and data analysis were done by Shida Liu under the direct supervision of Professor Kevin J. Smith. Sample collection were conducted by Mark Schmiβ. The manuscript was prepared and written by Shida Liu with final approval of Professor Kevin J. Smith. The XPS measurement was performed by Dr. Dennis Hsiao from 4D-Lab at SFU.  5. Shida Liu, Haiyan Wang, Yiling Dai, Chang Soo Kim, Kevin J. Smith, “A Study of Two-step Bio-oil Upgrading using Carbon Supported Mo2C Catalysts”, in preparation. A version of this manuscript is included in Chapter 6. Shida Liu conducted the catalyst preparation, characterization and experimental design as well as data analysis under the supervision of Professor Kevin J. Smith. Sample separation and collection were done by Dr. Haiyan Wang. The NMR measurement was performed by Yiling Dai. Dr. Chang Soo Kim contributed to the manuscript revision.  ix  Table of Contents Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ......................................................................................................................... ix List of Tables .............................................................................................................................. xvi List of Figures ............................................................................................................................ xxii Nomenclature ......................................................................................................................... xxviii List of Abbreviations ............................................................................................................. xxxiv Acknowledgements .............................................................................................................. xxxviii Dedication ..................................................................................................................................... xl Chapter 1: Introduction ............................................................................................................... 1 1.1 Background ................................................................................................................. 1 1.1.1 Definition of Fast Pyrolysis Oil ...................................................................... 1 1.1.2 Properties of Fast Pyrolysis Oil ...................................................................... 2 1.1.3 Existing Methods of Bio-Oil Upgrading ........................................................ 3 1.1.4 Esterification of Bio-Oil and Catalysts Applied ............................................. 5 1.1.4.1 Conventional Acid Catalysts........................................................................... 6 1.1.4.2 Carbon supported Catalysts for the Esterification of Bio-oil.......................... 7 1.1.5 HDO of Bio-Oil and Catalysts Applied .......................................................... 9 1.1.6 Carbon Supported Mo2C Catalysts ............................................................... 12 1.1.6.1 Carbon Sources and Chemical Activation of Carbon Support ..................... 12 1.1.6.2 Synthesis Method of Mo2C/C Catalyst ......................................................... 13  x  1.1.6.3 Oxygen Effect and Promoter Effect on Mo2C Stability in HDO .................. 14 1.1.7 Summary ....................................................................................................... 16 1.2 Objective of the Thesis ............................................................................................. 18 1.3 Approach ................................................................................................................... 18 1.4 Outline of Dissertation .............................................................................................. 20 Chapter 2: Synthesis of Carbon Supported Mo2C Catalysts and its Application in Esterification ................................................................................................................................................... 23 2.1 Introduction ............................................................................................................... 23 2.2 Experimental ............................................................................................................. 24 2.2.1 Materials ....................................................................................................... 24 2.2.2 Mo2C/APC Catalyst Preparation................................................................... 24 2.2.3 Catalyst Characterization .............................................................................. 26 2.2.3.1 N2 Adsorption and Desorption ...................................................................... 26 2.2.3.2 Transmission Electron Microscopy (TEM) .................................................. 27 2.2.3.3 Elemental Analysis ....................................................................................... 27 2.2.3.4 Ammonium Temperature-Programmed Desorption (NH3-TPD) ................. 27 2.2.3.5 Raman Spectroscopy ..................................................................................... 28 2.2.3.6 Esterification of Mo2C/APC and Acid-treated Mo2C/APC .......................... 28 2.3 Results ....................................................................................................................... 29 2.4 Discussion ................................................................................................................. 37 2.5 Conclusions ............................................................................................................... 44 Chapter 3: Carbon Supported Mo2C Catalysts Derived from Different Carbon Sources for the Hydrodeoxygenation of 4-Methylphenol  ................................................................................. 45  xi  3.1 Introduction ............................................................................................................... 45 3.2 Experimental ............................................................................................................. 47 3.2.1 Materials ....................................................................................................... 47 3.2.2 Mo2C/APC and Mo2C/ABC Catalyst Preparation ........................................ 47 3.2.3 Catalyst Characterization .............................................................................. 49 3.2.3.1 N2 Adsorption and Desorption ...................................................................... 49 3.2.3.2 X-ray Diffraction (XRD) .............................................................................. 49 3.2.3.3 Elemental Analysis ....................................................................................... 49 3.2.3.4 CO Chemisorption ........................................................................................ 49 3.2.3.5 Raman Spectroscopy ..................................................................................... 50 3.2.3.6 DRIFTS ......................................................................................................... 50 3.2.4 Hydrodeoxygenation (HDO) of 4-Methylphenol Mo2C/APC and Mo2C/ABC Catalysts 50 3.3 Results ....................................................................................................................... 52 3.4 Discussion ................................................................................................................. 66 3.5 Conclusions ............................................................................................................... 71 Chapter 4: Hydrodeoxygenation of 2-Methoxyphenol over Ru, Pd, and Mo2C Catalysts Supported on Carbon  ............................................................................................................... 72 4.1 Introduction ............................................................................................................... 72 4.2 Experimental ............................................................................................................. 73 4.2.1 Materials and Catalysts ................................................................................. 73 4.2.2 Catalyst Characterization .............................................................................. 74 4.2.2.1 N2 Adsorption and Desorption ...................................................................... 74  xii  4.2.2.2 CO Chemisorption ........................................................................................ 74 4.2.3 Hydrodeoxygenation (HDO) of 2-Methoxyphenol (GUA) over Ru/AC, Pd/AC and Mo2C/AC catalysts ......................................................................................... 75 4.3 Results ....................................................................................................................... 76 4.3.1 Catalyst Characterization .............................................................................. 76 4.3.2 HDO of 2-Methoxyphenol (GUA) on Pd/AC ............................................... 77 4.3.3 HDO of 2-Methoxyphenol (GUA) on Ru/AC .............................................. 85 4.3.4 HDO of 2-Methoxyphenol (GUA) on Mo2C/AC ......................................... 89 4.4 Discussion ................................................................................................................. 96 4.5 Conclusions ............................................................................................................. 101 Chapter 5: The Effect of Metal Promoters on Carbon Supported Mo2C Catalysts in Hydrodeoxygenation of Dibenzofuran ................................................................................... 102 5.1 Introduction ............................................................................................................. 102 5.2 Experimental ........................................................................................................... 104 5.2.1 Catalyst Preparation .................................................................................... 104 5.2.2 Catalyst Characterization ............................................................................ 105 5.2.2.1 N2 Adsorption and Desorption .................................................................... 105 5.2.2.2 H2 Temperature-Programmed Reduction (H2-TPR) ................................... 105 5.2.2.3 X-Ray Photoelectron Spectroscopy (XPS) ................................................. 106 5.2.2.4 STEM/EDX Analysis.................................................................................. 106 5.2.3 Catalytic Test for Hydrodeoxygenation (HDO) of Dibenzofuran (DBF) ... 106 5.2.4 DFT Calculation using VASP ..................................................................... 108 5.3 Results ..................................................................................................................... 110  xiii  5.3.1 Catalyst Synthesis and Characterization ..................................................... 110 5.3.2 Catalyst Activity and Selectivity in HDO of Dibenzofuran ....................... 125 5.3.3 DFT Calculations ........................................................................................ 129 5.4 Discussion ............................................................................................................... 133 5.5 Conclusions ............................................................................................................. 138 Chapter 6: Bio-oil upgrading .................................................................................................. 140 6.1 Introduction ............................................................................................................. 140 6.2 Experimental ........................................................................................................... 141 6.2.1 Catalyst Preparation and Raw Material ...................................................... 141 6.2.2 Experimental Conditions and Product Analysis ......................................... 141 6.2.3 Catalyst Characterization ............................................................................ 143 6.3 Results ..................................................................................................................... 144 6.4 Discussion ............................................................................................................... 157 6.5 Conclusions ............................................................................................................. 163 Chapter 7: Conclusions and Recommendations ..................................................................... 164 7.1 Conclusions ............................................................................................................. 164 7.2 Recommendations ................................................................................................... 168 7.2.1 Effect of Other Transition Metal Promoter on Mo2C Catalysts in both Esterification and HDO Reaction ................................................................................... 168 7.2.2 Stability Study of Mo2C Catalyst during crude Bio-oil Upgrading ............ 169 7.2.3 DFT Calculations to Identify Mo2C Deactivation Mechanism by O .......... 169 7.2.4 Solvent Effects during Crude Bio-oil Upgrading ....................................... 170 7.2.5 Additional Model Compound Studies ........................................................ 170  xiv  Bibliography ...............................................................................................................................172 Appendix A Catalyst Preparation ....................................................................................... 193 A.1 Activated Petcoke (APC) and Activated Biochar (ABC) ........................... 193 A.2 Catalyst Impregnation and Reduction Process ........................................... 194 Appendix B Bio-oil Analysis .............................................................................................. 196 B.1 Product Work Up ........................................................................................ 196 B.2 Detailed Example for GC-MS Analysis ..................................................... 197 B.3 Detailed Example for 1H-NMR Analysis ................................................... 202 Appendix C Products Calibration ....................................................................................... 204 Appendix D Modeling ........................................................................................................ 209 D.1 Lump Kinetic Models for Chapter 4 ........................................................... 209 D.2 DFT Calculations for Chapter 5 .................................................................. 224 Appendix E Supplementary Figures and Tables ................................................................. 231 E.1 Supplementary Information for Chapter 2 .................................................. 231 E.2 Supplementary Information for Chapter 3 .................................................. 239 E.3 Supplementary Information for Chapter 4 .................................................. 247 E.4 Supplementary Information for Chapter 5 .................................................. 257 E.5 Supplementary Information for Chapter 6 .................................................. 269 Appendix F Mass Transfers and Heat Transfer Effects ...................................................... 273 F.1 External Mass Transfer Effect in Fixed-bed reactor ................................... 273 F.2 Internal Mass Transfer Effect in Fixed-bed reactor .................................... 276 F.3 Heat Transfer Effect in Fixed-bed reactor .................................................. 278 F.4 External and Internal Mass Transfers in Batch Reactor ............................. 279  xv  Appendix G Error Analysis and Repeatability ................................................................... 280 G.1 Carbon Balance ........................................................................................... 280 G.2 CHNS Repeatability.................................................................................... 281 G.3 Water Titration Repeatability ..................................................................... 281 G.4 Statistical Analysis of Kinetic Model ......................................................... 283 Appendix H Reactor Operation Procedures ........................................................................ 285 H.1 Standard Operation Procedure (SOP) of Batch Reactor ............................. 285 H.2 Standard Operation Procedure (SOP) of Trickle Bed Reactor ................... 290   xvi  List of Tables Table 1.1: Typical properties of wood-derived pyrolysis bio-oil and heavy fuel oil [8]. ............... 3 Table 1.2: Comparison between noble metal catalyst and carbide catalyst in HDO reaction. ..... 10 Table 2.1: Textural properties, acidities and reaction rates of the raw petcoke and the activated petcoke before (APC_800) and after acid treatment (Acid-T APC_800) ..................................... 29 Table 2.2: Particle/cluster size of Mo2C/APC samples with different Mo loadings and different reduction temperatures. ................................................................................................................. 33 Table 2.3: Comparison of Mo2C/APC and acid treated Mo2C/APC catalysts prepared at a CHR temperature of 1000 °C: textural properties, acidities and reaction rates. .................................... 34 Table 2.4: Comparison of Mo2C/APC and acid treated Mo2C/APC catalysts prepared at a CHR temperature of 900 °C: textural properties, acidities and reaction rates. ...................................... 35 Table 3.1: Elemental analysis of the as-received raw carbons and the activated carbon supports........................................................................................................................................................ 54 Table 3.2: Textual properties of APC and ABC samples thermochemical activated at different temperatures. ................................................................................................................................. 56 Table 3.3: XPS analysis of Mo 3d species for 10%Mo/APC and 10%Mo/ABC fresh catalysts at different CHR temperatures. ......................................................................................................... 62 Table 3.4: Kinetic rate constants, selectivity, and conversion of 4-methylphenol (4-MP) on Mo2C/APC and Mo2C/ABC catalysts with different CHR temperatures at 350 oC and 4.3 MPa in hydrodeoxygenation (HDO). ........................................................................................................ 65 Table 3.5: Physical properties of Mo2C catalyst supported on APC and ABC with different reduction temperatures. ................................................................................................................. 70  xvii  Table 3.6: Kinetic rate constants for the conversion of 4-MP over Mo2C with different carbon supports at 350 oC. ........................................................................................................................ 71 Table 4.1: Physical and chemical properties of fresh 5%Ru/AC, 5%Pd/AC and 10%Mo2C/AC catalysts. ........................................................................................................................................ 76 Table 4.2: Product distribution from HDO of GUA over Ru/AC, Pd/AC and Mo2C/AC catalysts at 330 oC and 3.4 MPa initial H2 pressure, compared at the same GUA conversion (80% and 100%). ........................................................................................................................................... 77 Table 4.3: Estimated 1st-order rate constants and activation energy (Ea) for lumped model of Figure 4.2 for the hydrodeoxygenation of 2-methoxyphenol over Pd/AC catalyst at different reaction temperatures of 240, 300 and 330 oC. ............................................................................. 84 Table 4.4: Estimated 1st-order rate constants and activation energy (Ea) for lumped model of Figure 4.3 for the hydrodeoxygenation of 2-methoxyphenol over Ru/AC catalyst at different reaction temperatures of 240, 300 and 330 oC. ............................................................................. 87 Table 4.5: Estimated 1st-order rate constants and activation energy (Ea) for lumped model of Figure 4.6 for the hydrodeoxygenation of 2-methoxyphenol over Mo2C/AC catalyst at different reaction temperatures of 330, 350 and 375 oC. ............................................................................. 91 Table 5.1: Physical properties of fresh and used Mo2C/ABC catalysts with different metal promoters. ................................................................................................................................... 113 Table 5.2: H2-TPR analysis of pre-oxidized Mo2C/ABC and M-Mo2C/ABC catalysts. ............ 116 Table 5.3: XPS analysis of both fresh and used Mo2C catalysts with different promoters. ....... 120 Table 5.4: DBF conversion and product selectivites of Ni- and Pd-Mo2C/ABC catalysts at 4.1 MPa and LHSV=4 h-1 with varied temperatures in the range from 230 to 330 oC. .................... 127  xviii  Table 5.5: The calculated Gibbs free energy of DBF adsorption on clean Mo2C (101) and M-Mo2C (101) (M=Ni, Pd or Cu) surface with different adsorption angles at 350 oC. .................. 130 Table 5.6: The potential energy during dissociative adsorption of H2 on M-Mo2C (M=Pd, Ni or Cu) surfaces. ............................................................................................................................... 132 Table 6.1: CHNS/O elemental analysis and water content of oil phase after 1st step esterification reaction at 180 oC and 10.3 MPa using different catalysts. ........................................................ 145 Table 6.2: Oxygenate components distribution after 1st step esterification using different catalysts at 180 oC, 10.3 MPa with a mixing speed of 700 rpm and 1.5 h reaction time. ......................... 147 Table 6.3: CHNS/O elemental analysis and water content of oil phase (decalin excluded) after 2nd-step hydrodeoxygenation (HDO) reaction using different catalysts. .................................... 151 Table 6.4: Oxygenate components distribution after 2nd step HDO by different catalysts......... 153 Table 6.5: Results of 1H-NMR for the bio-oil from 1st step esterification and 2nd step hydrodeoxygenation with different catalysts. ............................................................................. 156 Table 6.6: Summary and comparison of experimental results and literature data for bio-oil upgrading. ................................................................................................................................... 162 Table B.1: Chemicals identified by different retention times in crude bio-oil. .......................... 198 Table B.2: Chemical groups summarized from GCMS data. ..................................................... 201 Table B.3: Product distribution of aqueous and oil phase from 1st step reaction on Mo2C/APC catalyst compared to crude bio-oil. ............................................................................................. 203 Table C.1: GC-MS calibration for reactant DBF. ....................................................................... 205 Table C.2: GC-MS calibration for product BPh. ........................................................................ 205 Table C.3: GC-MS calibration for product CHB. ....................................................................... 206 Table E.1: Ultimate analysis of raw petroleum coke. ................................................................. 231  xix  Table E.2: Physical properties of Mo2C/APC with various Mo loadings and CHR temperatures...................................................................................................................................................... 232 Table E.3: Textural properties of Mo2C/PC catalysts. ................................................................ 233 Table E.4: CHNS analysis of raw petcoke and APC_800 support. ............................................ 233 Table E.5: SEM/EDX analysis of APC_800 and Acid-T APC_800. ......................................... 233 Table E.6: ICP-OES analysis of PC/APC_800 and BC/ABC_700. ........................................... 239 Table E.7: Elemental composition of Mo2C/C catalysts from XPS analysis. ............................ 240 Table E.8: Bulk density of different carbons. ............................................................................. 240 Table E.9: O 1s XPS analysis of Mo2C/C catalysts with different carbon supports and reduction temperatures. ............................................................................................................................... 241 Table E.10: Guaiacol products distribution over Pd/AC catalyst as a function of reaction time at 240 oC.......................................................................................................................................... 248 Table E.11: Guaiacol products distribution over Pd/AC catalyst as a function of reaction time at 300 oC.......................................................................................................................................... 249 Table E.12: Guaiacol products distribution over Pd/AC catalyst as a function of reaction time at 330 oC.......................................................................................................................................... 250 Table E.13: Guaiacol products distribution over Ru/AC catalyst as a function of reaction time at 240 oC.......................................................................................................................................... 251 Table E.14: Guaiacol products distribution over Ru/AC catalyst as a function of reaction time at 300 oC.......................................................................................................................................... 252 Table E.15: Guaiacol products distribution over Ru/AC catalyst as a function of reaction time at 330 oC.......................................................................................................................................... 253  xx  Table E.16: Guaiacol products distribution over Mo2C/AC catalyst as a function of reaction time at 330 oC. ..................................................................................................................................... 254 Table E.17: Guaiacol products distribution over Mo2C/AC catalyst as a function of reaction time at 350 oC. ..................................................................................................................................... 255 Table E.18: Guaiacol products distribution over Mo2C/AC catalyst as a function of reaction time at 375 oC. ..................................................................................................................................... 256 Table E.19: Carbon balance error calculation for reactant (DBF) and products in HDO of dibenzofuran over 10Mo2C/APC catalyst at 350 oC and 4.1 MPa. ............................................ 257 Table E.20: Standard deviation of product concentrations in HDO of dibenzofuran over 10Mo2C/APC catalyst at 350 oC and 4.1 MPa. ........................................................................... 258 Table E.21: Diagram of horizontal and vertical DBF adsorption on clean Mo2C (101) surface for DBF adsorption energy calculation. ........................................................................................... 259 Table E.22: Diagram of four types of horizontal DBF adsorption on Ni-Mo2C surface for DBF adsorption energy calculation. .................................................................................................... 260 Table E.23: Diagram of four types of vertical DBF adsorption on Ni-Mo2C surface for DBF adsorption energy calculation. .................................................................................................... 261 Table E.24: Diagram of four types of horizontal DBF adsorption on Pd-Mo2C surface for DBF adsorption energy calculation. .................................................................................................... 262 Table E.25: Diagram of four types of vertical DBF adsorption on Pd-Mo2C surface for DBF adsorption energy calculation. .................................................................................................... 263 Table E.26: Diagram of four types of horizontal DBF adsorption on Cu-Mo2C surface for DBF adsorption energy calculation. .................................................................................................... 264  xxi  Table E.27: Diagram of four types of vertical DBF adsorption on Cu-Mo2C surface for DBF adsorption energy calculation. .................................................................................................... 265 Table E.28: Textual and Chemical Properties of Catalysts. ....................................................... 269 Table E.29: Identification of types of Protons by Chemical Shift. ............................................. 270 Table F.1: The details of external mass transfer calculation by Mears criterion. ....................... 273 Table F.2: The details of internal mass transfer by Weisz-Prater criterion. ............................... 276 Table F.3: The details of heat transfer calculation by Mears Criterion. ..................................... 278 Table F.4: A detailed list of external and internal mass transfer coefficient calculation for 10Mo2C/APC-R650 in HDO of 4-MP in batch reactor. ............................................................. 279 Table G.1: Reactant (DBF) and products concentration for HDO of dibenzofuran over 10Mo2C/APC-700 catalyst at 350 oC and 4.1 MPa. ................................................................... 280 Table G.2: CHNS results for different samples. ......................................................................... 281 Table G.3: H2O titration results for different samples. ............................................................... 282 Table G.4: Comparation between experimental and kinetic model fitted data of different products and reactant in HDO of 4-MP by 10Mo2C/APC_R600 catalyst................................................. 283 Table G.5: ANOVA analysis of 4-MP concentration data in HDO of 4-MP by 10Mo2C/APC_R600 catalyst. ..................................................................................................... 284 Table G.6: ANOVA analysis of DDO products concentration data in HDO of 4-MP by 10Mo2C/APC_R600 catalyst. ..................................................................................................... 284 Table G.7: ANOVA analysis of HYD products concentration data in HDO of 4-MP by 10Mo2C/APC_R600 catalyst. ..................................................................................................... 284   xxii  List of Figures Figure 1.1: Fast pyrolysis processes for biomass [4]. ..................................................................... 2 Figure 1.2: Schematic illustration of both model compounds and real bio-oil study of bio-oil upgrading process using carbon supported Mo2C catalysts (Mo2C/C). ........................................ 22 Figure 2.1: 3D plot of physical properties of Mo2C/APC samples with various Mo loadings and CHR temperatures: (a) Surface area; (b) total pore volume (Vtotal); (c) mesopore volume (Vmeso)........................................................................................................................................................ 31 Figure 2.2: A correlation between adsorbed NH3 and the 2nd-order esterification reaction rate constants (k): (●) Acid-T Mo2C/APC; (■) Mo2C/APC; (▲) Acid-T APC_800; — trendline .. 37 Figure 2.3: Diagram of pore development process of AHM/APC precursors with different Mo loadings and CHR temperatures. .................................................................................................. 40 Figure 3.1: DRIFTS spectra of raw petcoke, APC_800 and raw bio-char, ABC_700 supports. . 57 Figure 3.2: Profile of detected relative intensity of PCH4/PHe as a function as time for 10%Mo2C/APC_R700 (represented by —) and 10%Mo2C/ABC_R700 samples (represented by —). ................................................................................................................................................ 58 Figure 3.3: XRD patterns of 10%Mo2C/ABC and 10%Mo2C/APC reduced at different CHR temperatures. (♦ identifies C-graphite; * identifies Mo2C) ........................................................... 59 Figure 3.4: XPS spectra of Mo 3d of fresh 10%Mo2C/APC samples reduced at different CHR temperatures: (a) Survey scan of 10%Mo2C/APC_R600; narrow scans of: (b) 10%Mo2C/APC_R600, (c) 10% Mo2C/APC_R650, and (d) 10%Mo2C/APC_R700. ................. 61 Figure 3.5: XPS narrow scan spectra of Mo 3d of fresh 10%Mo2C/ABC samples reduced at different CHR temperatures. (a) 10%Mo2C/ABC_R600; (b) 10% Mo2C/ABC_R650; (c) 10%Mo2C/ABC_R700. ................................................................................................................. 63  xxiii  Figure 4.1: Mole percentage (mol.%) of reactant (2-methoxyphenol) and products as a function of reaction time on Pd/AC at different reaction temperatures ((a) 240 oC; (b) 300 oC; (c) 330 oC). (HYD: 2-methoxycyclohexanone; 2-methoxycyclohexanol; 1-methylcyclohexane-trans-1,2-diol. 1-O removed: cyclohexanol; 1-methoxycyclohexane; cyclopentylmethanol (not detectable at 240 oC). 2-O removed: cyclohexane; cyclopentane; methylcyclopentane.) ........................................ 82 Figure 4.2: Proposed reaction network for HDO of 2-methoxyphenol on Pd/AC. ...................... 84 Figure 4.3: Mole percentage (mol.%) of reactant (2-methoxyphenol) and products as a function of reaction time on Ru/AC at different reaction temperatures ((a) 240 oC; (b) 300 oC; (c) 330 oC). (HYD: 2-methoxycyclohexanone; 2-methoxycyclohexanol; 1-methycyclohexane-trans-1,2-diol, 1-O removed: cyclohexanol and methoxycyclohexane. 2-O removed: cyclohexane, benzene, cyclohexene, cyclopentane and methylcyclopetane. Alkylation&HYC: ethylcyclohexane, methylcyclohexane, 1-ethyl-2-methylbenzene, dimethylcyclohexane and dimethylbenzene etc.)....................................................................................................................................................... 88 Figure 4.4: Proposed reaction network for HDO of 2-methoxyphenol on Ru/AC. ...................... 89 Figure 4.5: Mole percentage (mol.%) of reactant (2-methoxyphenol) and products as a function of reaction time on Mo2C/AC at different reaction temperatures ((a) 330 oC; (b) 350 oC; (c) 375 oC). (1-O removed: phenol; 2-methylphenol; 4-methylphenol and anisole. 2-O removed: toluene and benzene.). ............................................................................................................................... 92 Figure 4.6: Proposed reaction network for HDO of 2-methoxyphenol on Mo2C/AC catalyst. .... 93 Figure 4.7: Experimental and model concentration data versus reaction time of Pd/AC catalyst at (a) 300 oC and (b) 330 oC. ............................................................................................................. 94 Figure 4.8: Experimental and model concentration data versus reaction time of Ru/AC catalyst at (a) 300 oC and (b) 330 oC. ............................................................................................................. 95  xxiv  Figure 4.9: Experimental and model concentration data versus reaction time of Mo2C/AC catalyst at (a) 330 oC, (b) 350 oC and (c) 375 oC. ......................................................................... 96 Figure 5.1: TPR-MS profiles of carbothermal hydrogen reduction (CHR) of Mo2C/ABC and M-Mo2C/ABC catalysts. (a) m/z=2, H2; (b) m/z=15, CH4; (c) m/z=28, CO; (d) m/z=44, CO2. (The signal is presented by the relative intensity of H2, CH4, CO, and CO2 to inert gas He) ............. 111 Figure 5.2: H2-TPR of pre-oxidized Mo2C/ABC and M-Mo2C/ABC (M=Ni, Pd, or Cu) catalysts at different reduction temperatures. (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC. ........................................................................................................................... 114 Figure 5.3: TEM images of fresh catalysts and corresponding particle size distributions: (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC. ................................ 118 Figure 5.4: High angle annular dark field TEM scanning images (HAADF-STEM) and Energy dispersive X-ray (EDX) elemental mappings of fresh Ni-Mo2C/ABC catalyst. ........................ 119 Figure 5.5: XPS (Mo 3d region) narrow scan spectra of fresh Mo2C-typed catalysts: (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC. ................................ 121 Figure 5.6: XPS narrow scan spectra of Mo2C-type catalysts: (a) Ni 2p; (b) Pd 3d; (c) Cu 2p. 123 Figure 5.7: XPS (Mo 3d region) narrow scan spectra of used Mo2C-typed catalysts: (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC. ................................ 124 Figure 5.8: DBF conversion and selectivity of Mo2C/ABC and M-Mo2C/ABC catalysts at350 oC, 4.1 MPa and LHSV = 4 h-1. ........................................................................................................ 126 Figure 5.9: Reaction pathway of dibenzofuran HDO on Mo2C based catalysts. (The products in the dash box represent the intermediates during the reaction) .................................................... 128 Figure 6.1: O/C ratio of feed and esterified bio-oil after 1st step esterification operated at 180 oC. (* represents the experiment was conducted at 240 oC) ............................................................. 146  xxv  Figure 6.2: Properties analysis of 1st esterification products: (a) Oil vs. aqueous phase; (b) Carbon yield, %; (c) Coke yield. ................................................................................................ 149 Figure 6.3: O/C ratio of oil phase products after 2nd step HDO. (a. Reaction Time=1.5 h; Reaction Temperature= 300 oC; b. Reaction Time=4.0 h; Reaction Temperature= 300 oC; c. Reaction Time=4.0 h, Reaction Temperature= 350 oC; a,b,c all used feed from 1st step esterification by Ru/C catalyst; d. Reaction Time=4.0 h, Reaction Temperature= 350 oC, used feed from 1st step esterification by 2Ni-Mo2C/ABC catalyst ) ................................................... 150 Figure 6.4: Properties analysis of 2nd HDO products: (a) Oil vs. aqueous phase; (b) Carbon yield, %; (c) Coke yield. ............................................................................................................. 155 Figure A.1: XRD patterns of activated bio-char (ABC; (a)) and activated petcoke (APC; (b)) samples with different activation temperatures. (Impurities: SiO2 (*); Al2O3 (♦); C peaks at (002), (100) and (110) planes, respectively) ............................................................................... 193 Figure A.2: Schematic illustration of the quartz U-tube applied in CHR process. ..................... 195 Figure B.1: Product work up for bio-oil analysis. ...................................................................... 196 Figure B.2: GC-MS of chemicals in crude bio-oil. ..................................................................... 197 Figure B.3: Product distribution of aqueous and oil phase from 1st step reaction on Mo2C/APC catalyst compared to crude bio-oil. ............................................................................................. 202 Figure C.1: A linear correlation between GC-MS area ratio and DBT. ..................................... 207 Figure C.2: A linear correlation between GC-MS area ratio and BPh. ...................................... 207 Figure C.3: A linear correlation between GC-MS area ratio and CHB. ..................................... 208 Figure E.1: Raman spectra analysis of APC samples: (a) Raw petcoke; (b) APC_800. ............ 234 Figure E.2: Profile of CH4 (mol%) produced during carbothermal hydrogen reduction (CHR): (□) 2Mo2C/APC_R700; (○) 5Mo2C/APC_R700; (Δ) 10Mo2C/APC_R700. .......................... 235  xxvi  Figure E.3: NLDFT pore size distributions derived from N2 adsorption isotherms for APC_800 and Acid-T APC_800. ................................................................................................................ 236 Figure E.4: NLDFT pore size distributions derived from N2 adsorption isotherms for 1Mo2C/APC at different CHR temperatures. (Left: 0-150 nm; Right: 0-50 nm) ....................... 237 Figure E.5: NLDFT pore size distributions derived from N2 adsorption isotherms for 2Mo2C/APC at different CHR temperatures. (Left: 0-150 nm; Right: 0-50 nm) ....................... 238 Figure E.6: Raman spectra analysis of APC samples: (a) Raw petcoke; (b) APC_600; (c) APC_700; (d) APC_800; (e) APC_900. ..................................................................................... 242 Figure E.7: Raman spectra analysis of ABC samples: (a) Raw biochar; (b) ABC_600; (c) ABC_700; (d) ABC_800; (e) ABC_900..................................................................................... 243 Figure E.8: N2 adsorption-desorption isotherms of activated bio-char (ABC) and petcoke (APC) samples prepared by thermochemical method with KOH with different activation temperatures...................................................................................................................................................... 244 Figure E.9: O1s XPS narrow scan spectra deconvolution of different Mo2C catalysts: (a-c) 10%Mo2C/ABC with 600, 650 and 700 oC, respectively; (d-f) 10%Mo2C/APC with 600, 650 and 700 oC, respectively; (g-I) 10%Mo2C/AC with 600, 650 and 700 oC, respectively. .................. 245 Figure E.10: Pore size distribution of ABC and APC supported catalysts. ................................ 246 Figure E.11: High angle annular dark field TEM scanning images (HAADF-STEM) and Energy dispersive X-ray (EDX) elemental mappings of fresh Pd-Mo2C/ABC catalyst. ........................ 266 Figure E.12: The potential energy during dissociative adsorption of H2 on three surfaces: (a, b) Pd-Mo2C; (c, d) Ni-Mo2C; (e, f) Cu-Mo2C surface. ................................................................... 267 Figure E.13: TPR-MS CH4 profile of Ni-Mo2C/ABC and Ni/ABC catalysts during carbothermal hydrogen reduction (CHR). ........................................................................................................ 268  xxvii  Figure E.14: NMR results for feed and 1st esterification oil phase products at 180 oC and 10.3 MPa. ............................................................................................................................................ 271 Figure E.15: NMR results of oil phase products after 2nd step esterification. (a. Reaction Time=1.5 h; Reaction Temperature= 300 oC; b. Reaction Time=4.0 h; Reaction Temperature= 300 oC; c. Reaction Time=4.0 h, Reaction Temperature= 350 oC; a,b,c all used feed from 1st step esterification by Ru/C catalyst; d. Reaction Time=4.0 h, Reaction Temperature= 350 oC, used feed from 1st step esterification by 2Ni-Mo2C/ABC catalyst ) ................................................... 272   xxviii  Nomenclature A Cross sectional area of the reactor, m2 CA Concentration of butanol, mol/L CAO Initial concentration of butanol, mol/L CB Concentration of acetic acid, mol/L CCat Catalyst concentration in the mixed reactor, gMo/mLfeed CDBF Bulk gas concentration of DBF, kmol/m3 CH2 Bulk gas concentration of H2, kmol/m3 Ĉp(H2) Heat capacity of H2 at reaction, J/kg·K Cwp Weisz-Prater criterion DDBF-H2 Binary bulk phase diffusivity, m2/s Deff, DBF-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 E(x/slab) Energy of the adsorbed system with both slab and specie X, eV Ea Apparent activation energy, kJ/mol Ea Energy barrier, eV (Ea = ETS - EIS) Ea Activation energy, J/mol Eads Adsorption energy, eV EFS Final state energy, eV EIS Energy of initial state, eV  xxix  Er Reaction energy, eV (Er = EFS - EIS) Eslab Energy of slab, eV ETS Transition state energy, eV Ex Energy of specie X, eV h Heat transfer coefficient, W/(m2·K) I(D)/I(G) The degree of disordered carbon to graphitic carbon k 2nd-order kinetic rate constant, m3.(mol.s)-1 k1 Stabilized kinetic parameter, s-1 k4-MP Kinetic parameter, s-1 kc Sum of kDDO and kHYD, mL/(gMo.min) kc Mass transfer coefficient, m/s kDDO 1st-order HDO rate constant, mL/(gMo.min) kHYD 1st-order HYD rate constant, mL/(gMo.min) kLa Liquid-side mass transfer coefficient, s-1 kt Thermal conductivity calculated by semiempirical method for polyatomic gases, W/(m.K) kTDDO 1st-order rate constants of the thermal reactions for DDO route (w/o catalyst) kTHYD 1st-order rate constants of the thermal reactions for HYD route (w/o catalyst) Lbed Length of the catalyst bed, cm mafter Mass of carbon or catalyst after KOH activation or CHR treatment, g MC Mears’ criterion for external diffusion  xxx  MC’ Mears’ criterion for isothermal operation mcat Mass of loaded catalyst, g MDBF Mole weight of DBF, g/mol MH2 Mole weight of H2, g/mol minitial Initial mass of the catalyst precursor prior to KOH activation, g Mmix Feed molecular weight, g/mol n Reaction order Nu Nusselt number ø Porosity or void fraction of packed bed  ø1 Thiele modulus for 1st order reaction øp Catalyst particle porosity  P Partial pressure of DBF, kPa PCH4/PHe Relative pressure of CH4 to He PDBF Partial pressure of DBF, atm PDecalin Partial pressure of Decalin, atm PH2 Partial pressure of H2, atm Po Standard state pressure, kPa Prxn Reaction pressure, atm Pt Prandtl number  Ptotal Total pressure in the system, atm r Rate of reaction, mol.gcat-1.s-1 R Catalyst particle radius, m -rDBF(obs) Observed reaction rate, kmol/gcat.s  xxxi  Re Reynolds number Re' Reynolds number considering void fraction  Rg Gas constant, J/(mol·k) ri Internal radius of the reactor, m rpore Pore radii of the catalyst, cm Sc Schmidt number, Sc=νmix/DDBF-H2 Scat. Surface area of the catalyst, m2/g SDDO/HYD Selectivity of DDO to HYD route Sh Sherwood number (Sh=2+0.6Re1/2Sc1/3) T Temperature, K T* Dimensionless temperature, K (T*=ĸTrxn/εDBT-H2) TDBF,b Boiling point of DBF, K TDBF,c Critical point temperature of DBF, K Trxn Reaction temperature, K U Superficial gas velocity, m/s (U=γo/A) Vc Loaded catalyst volume, mL ṼDBF,c Critical volume of DBF, mL/g-mol ṼDBF,liquid Critical volume of DBF at its normal boiling point, 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  xxxii  X Conversion of butanol, % Ƴ Shape factor  yi Mole fraction of the product yoDBF Mole fraction of the DBF in the liquid feed yoi Mole fraction of the product i that contains O Zrot DBT (T) Rotational partition function of DBF Ztrans DBT (T, P) Translational partition function of DBF γo Volumetric flow rate, m3/s ΔG Gibbs free energy of adsorption, eV ΔHrxn Heat of reaction, kJ/mol ΔZPE Zero-point energy difference between adsorbed system and the adsorbate Greek symbols  µmix Dynamic viscosity of the mixture, kg/m.s εDBF/ĸ Lennard-Jones parameters for DBF/Boltzmann's constant, K εDBF-H2/ĸ Lennard-Jones parameters for DBF-H2/Boltzmann's constant, K εH2/ĸ Lennard-Jones parameters for H2/Boltzmann's constant, K η Internal effectiveness factor νmix Kinetic viscosity of mixture, m2/s ρb Catalyst bed density, g/cm3 ρc Catalyst density, g/cm3 ρmix Mixture density, kg/m3 σ DBF Lennard-Jones parameters for DBF/characteristic length, Å  xxxiii  σ DBF-H2 Lennard-Jones parameters for DBF-H2/characteristic length, Å σ H2 Lennard-Jones parameters for H2/characteristic length, Å σc Constriction factor τ Tortuosity factor Ω Overall effectiveness factor ΩD, DBF-H2 Collision integral, calculated by ignore the last two terms   xxxiv  List of Abbreviations 0.5Mo2C/APC_R900 0.5 wt% of Mo loading on activated petroleum coke and CHR reduced at 900 oC 10%Mo2C/APC_R700 10 wt% of Mo loading on activated petroleum coke and CHR reduced at 700 oC 2-CHP 2-Cyclohexylphenol 2D-NLDFT 2-Dimension non-local density functional theory 4-MP 4-Methylphenol ABC Activated bio-char ABC_700 Activated biochar activated at 700 oC AC Activated carbon Acid-T APC_800 Acid washed APC_800 AHM Heptamolybdate tetrahydrate ANOVA Analysis of variance APC Activated petroleum coke APC_800 Activated petroleum coke activated at 800 oC ave Average BC Raw biochar BCH Bicyclohexane BET Brunauer-Emmett-Teller BPh Biphenyl CH Cyclohexane CHB Cyclohexylbenzene  xxxv  CHR Carbothermal hydrogen reduction DBF Dibenzofuran DBF-HX, X=1-4 Horizontal adsorption of DBF DBF-VX, X=1-4 Vertical adsorption of DBF DBT Dibenzothiophene DCM Dichloromethane DDO Direct deoxygenation DFT Density functional theory DRIFTS Diffuse reflectance infrared Fourier transform spectroscopy EDX Energy-dispersive X-ray elemental mapping EG Ethylene glycol EIA Energy Information Administration GC-FID Gas chromatography-Flame Ionization Detector GC-MS Gas chromatography mass spectrometry GC-TCD Gas chromatography-Thermal conductivity detector GGA-PBE Generalized gradient approximation with Perdew-Burke-Ernzerhof GHG Green house gas GUA 2-Mehoxyphenol HAADF High angle annular dark field scanning HDO Hydrodeoxygenation HDS Hydrodesulfurization HHV High heating value  xxxvi  HRTEM High-resolution transmission electron microscopy HYD Hydrogenation route ICP-OES Inductively coupled plasma optical emission spectroscopy IUPAC International union of pure and applied chemistry Mo2C Molybdenum carbide NH3-TPD Ammonium temperature-programmed desorption Ni-Mo2C/ABC Ni promoted Mo2C catalyst supported on activated biochar NLDFT Non-local density functional theory O/C ratio O to C ratio OHE One-step hydrogenation and esterification PAW Projector augmented wave PC Petroleum coke PC Raw petroleum coke Pd/AC Palladium supported on activated carbon Ru/AC Ruthenium supported on activated carbon Ru/SiO2-Al2O3 Ruthenium supported on silica-alumina SA Surface area, m2/g SBET BET specific surface area, m2/g SOP Standard Operation Procedure SPF Spruce, pine, and fir pellets std. dev Standard deviation TEM Transmission electron microscopy THDBF Tetrahydro-dibenzofuran  xxxvii  TOF Turnover frequency TPO-TPR Temperature-programmed oxidation/Temperature programmed reduction TPR Temperature-programmed reduction VASP Vienna Ab initio simulation package WGS Water gas shift reaction XPS X-ray photoelectron spectroscopy XRD X-ray diffraction   xxxviii  Acknowledgements First of all, I would like to thank my Ph.D. supervisor, Professor Kevin J. Smith. His professional guidance and incredible patience walked me through my PhD life. It has been an honor for me to work with Professor Smith and his advice and specific spirit will be part of me in my future career and life.   Secondly, I would like to thank Professor Heather Trajano and Professor Chang Soo Kim for being my committee members. They provided me with their valuable advice which helped me a lot in progressing my PhD research.  Thirdly, I would like to thank all staff at the Department of Chemical and Biological Engineering (CHBE), Marlene Chow, Amber Lee, Lori Tanaka, Kristi Chow, William Wijaya, Miles Garcia, Serge Milaire, Doug Yuen, Ken Wong, Richard Zhang etc. for their help. Special thanks to Mr. Richard Ryoo for being an amazing store manager.  Furthermore, I wish to thank all CHBE Catalysis group members (Dr. Haiyan Wang, Dr. Ross Kukard, Dr. Rahman Gholami Shahrestani, Dr. Pooneh Ghasvareh, Chujie Zhu, Alex Imbault, Lucie Solnickova, Dr. Ali Alzaid, Dr. Mina Alyani, Lingxiu Zhu, Xu Zhao, Majed Alamoudi, Yanuar Philip Wijaya, Hamad Almohamadi) and visiting scholars (Mark Schmiβ, Malte Stelzner, Julian Thomas, Dr. Xin Wang, Dr. Yunhua Li), who helped in the course of my PhD.    xxxix  Last but not least, I would like to thank my wife, Haiyan. She is the reason I can keep moving forward in my PhD life. Also, I like to express my gratitude to my beloved parents for their loving consideration and great confidence in me all the time.   xl  Dedication      To my beloved wife, Haiyan    1  Chapter 1: Introduction 1.1 Background 1.1.1 Definition of Fast Pyrolysis Oil According to the EIA (Energy Information Administration) [1], worldwide energy consumption increased by 14.5% from 2007 to 2017. Due to environmental considerations, renewable energy sources as alternatives to fossil fuels are in demand and are expected to supply a larger fraction of the total world energy needs in the future [2]. Bio-oils (pyrolysis oils) generated by the fast pyrolysis of biomass, are a promising alternative to partially replace fossil liquid fuels such as gasoline and diesel. Green-house gas (GHG) emissions from bio-oil use are 77 ~ 99% less than from fossil fuel combustion [3].  Fast pyrolysis (see Figure 1.1) is a thermochemical process that rapidly converts biomass into three products: bio-oil (65-70 wt.%), bio-char (15-20 wt.%) and a mixture of non-condensable gases. Usually, the reaction occurs in a short period of time (≤ 2 s) at atmospheric pressure and at a temperature of ~500 oC in the absence of oxygen [4]. There are several factors that affect the yield of the product, such as operating conditions (temperature, pressure, residence time etc.) and the composition of the feedstock.  2   Figure 1.1: Fast pyrolysis processes for biomass [4]. (Copyright © 1999, Elsevier Science Ltd., reproduced with permission)  Energy produced from bio-oils has a smaller net CO2 emission per unit of energy than conventional petroleum oil, because the CO2 emitted during bio-oil combustion can be consumed by photosynthesis to generate the feed biomass [5]. Bio-oil combustion reduces NOx emissions by ≥ 50% compared to diesel [6]. Also, SOx does not form because the biomass contains insignificant amounts of sulphur [6, 7]. Consequently bio-oil could be important in securing future energy supplies and in reducing global warming.   1.1.2 Properties of Fast Pyrolysis Oil Bio-oil usually appears as a dark brown liquid with high viscosity and density. It can be mixed with ethanol, methanol or water, but it is difficult to be blend with hydrocarbons. Table 1.1 shows the differences between wood-derived pyrolysis oil and heavy oil. Note that the pH value of bio-oil is low, thus it can cause serious corrosion of vessels and pipes during transportation.  3  The high heating value (HHV) of bio-oil is about 16-19 MJ/kg. It contains 15-30 wt.% water. In addition to the above-mentioned properties, aging propensity is another notable feature of bio-oil. Given these properties, direct utilization of bio-oil as fuel is challenging. Since many of the negative properties of bio-oils can be linked to excessive amounts of O in the oil, removal of O is essential before bio-oil can be effectively utilized as an energy carrier or as a fuel in the transportation sector.  Table 1.1: Typical properties of wood-derived pyrolysis bio-oil and heavy fuel oil [8]. (Copyright © 2006, Elsevier Science Ltd., reproduced with permission) Physical property Bio-oil Heavy fuel oil Moisture content (wt. %) 15–30 0.1 pH 2.5 – Specific gravity 1.2 0.94 Elemental composition (wt. %) C 54–58 85 H 5.5–7.0 11 O 35–40 1 N 0–0.2 0.3 S 0-0.05 1-3 Ash 0–0.2 0.1 HHV (MJ/kg) 16–19 40 Viscosity (at 50 °C) (cP) 40–100 180 Solids (wt.%) 0.2–1 1 Distillation residue (wt.%) up to 50 1  1.1.3 Existing Methods of Bio-Oil Upgrading  There are many different upgrading strategies for bio-oil that have been investigated including hydrodeoxygenation (HDO), cracking, steam reforming, supercritical fluid extraction, esterification, emulsification, among others [6, 9]. They can be summarized into two groups based on the main purpose of the process: phase separation or chemical conversion.  4   The main purpose of phase separation processes is to separate the chemicals with less oxygen into an oil phase and the chemicals with more oxygen into an aqueous water phase. For example, esterification of bio-oil is usually operated at mild temperatures, 47 ~ 55 oC [10]. The main reactants in the bio-oil that participate in the reaction are acids and aldehydes. The majority of the O-functional groups attached to the bio-oil components remain after the esterification reaction. However, most of the highly polar compounds associate with the aqueous phase, resulting in a higher quality bio-oil (organic phase) after esterification. Supercritical fluid extraction operates at critical conditions with a solvent such as ethanol [11] to affect a high efficiency phase separation. In emulsification, bio-oil is blended with diesel with the help of surfactants to increase the ignition point. In an emulsification of bio-oil process, the bio-oil is usually mixed with biodiesel, and the biodiesel absorbs the soluble part of the bio-oil. After the emulsification of bio-oil, the bio-oil will be separated into an organic rich phase and an aqueous phase [12]. In summary, although phase separation processes can be operated at mild conditions with low energy consumption, there is still a considerable amount of oxygen left in the oil. For example, Guo et al. [13] used an ionic liquid to perform an esterification reaction on bio-oil. After removing the aqueous phase from the oil phase, the O content of the upgraded bio-oil remained high (38 wt. %) compared to the O content of 49 wt.% of the crude bio-oil. The application of the upgraded/phase separated bio-oil in engine combustion remains difficult.  The main purpose of chemical conversion processes is the removal of the O-functional groups attached to the bio-oil components and to generate hydrocarbons at high yield. For example, hydrodeoxygenation (HDO) of bio-oil can reduce the oxygen content by removing O-functional  5  groups from acids, aldehydes, ketones and phenols [14]. HDO is a catalytic process operated at ~300 oC under high pressure (7-20 MPa) of H2 aimed at converting bio-oil into liquid phase hydrocarbons [15]. Cracking processes, on the other hand, require temperatures > 350 oC with most of the bio-oil  components converted to gaseous products [16]. Steam reforming of bio-oil is operated at even higher temperature ca. 800~900 oC, with the product target in this process being hydrogen [17].  The deoxygenation ability of chemical conversion processes is very significant. A high degree of deoxygenation requires high temperature. However, the product distribution is not ideal for liquid production if the process temperature is too high and the formation of coke during the process must be avoided. Also, despite the mild temperature required for HDO, the energy consumption is high due to the presence of inert liquid water in the bio-oil.  1.1.4 Esterification of Bio-Oil and Catalysts Applied  Due to its complexity, pyrolysis oil is often unstable during storage let alone during the HDO process. Thus, stabilization of pyrolysis oil by adding a solvent or by pre-treatment such as esterification could be very beneficial for the subsequent HDO reactions. Esterification reduces the polarity of the bio-oil while generating water and this contributes to an oil-water phase separation and the removal of most of the water from the bio-oil [18, 19]. Esterification reactions are usually catalyzed by homogenous acid catalysts or solid acid catalysts.   6  1.1.4.1 Conventional Acid Catalysts Xiong et al. [13] used a dicationic ionic liquid C6(mim)2-HSO4 as catalyst for esterification. Bio-oil (11.4 mL) was mixed with 5.7 mL ethanol and 0.5 g ionic liquid. The reaction occurred at room temperature. The moisture content decreased from 29.8 to 8.2 wt.%, the HHV increased from 17.3 to 24.6 MJ/Kg, and the pH value increased from 2.9 to 5.1 after reaction. The viscosity was also significantly lowered during the process. The ionic liquid ended up in the water phase, making recycle of the ionic liquid very convenient.  Li et al. [20] used a solid commercial Amberlyst-70 acid catalyst to process bio-oil with methanol. The temperature of the reaction varied from 70 to 170 ºC. The authors reported that esterification is a low energy consumption reaction that occurs readily even at low temperature (70 ºC) with catalyst addition. Zhang et al. [21] processed bio-oil with olefin and 1-octene over sulfonic acid resin catalysts at 80 to 120 ºC. In the presence of 1-octene alone, significant coke formation and phase separation occurred at temperatures as low as 80 oC. With 1-butanol addition, however, esters, acetals with carboxylic acids and aldehydes/ketones, were obtained. The addition of the alcohol stabilized the bio-oil resulting in reduced coke formation, and no coke was detected at temperatures as high as 120 oC. The upgraded oil from esterification at 80 oC using olefin and 1-octene mixture as solvent had much less water (6.5 wt%) than the crude bio-oil (37.2 wt%).   A one step hydrogenation and esterification (OHE) process concept has been suggested by Tang et al. [18] In their study, Pt over HZSM-5 or Al2(SiO3)3 catalysts were used at 150 oC and 1.5 MPa with H2. The idea was to utilize the aldehydes in a hydrogenation-esterification reaction to  7  react with the acids of the bio-oil. Since aldehydes exist in the real bio-oil and could cause problems by oligomerization, hydrogenation before oligomerization could be a much more beneficial route. Even though supports with strong acidity were applied, Pt was chosen to increase the hydrogenation ability at low temperature. They used acetaldehyde and butyl aldehyde as model compounds to react with acetic acid in OHE reactions. For Pt/Al2(SiO3)3 catalyst, the selectivity of main products ethyl acetate and butyl acetate reached 72% and 93% respectively, much higher than the 7.5% and 0.3% obtained in blank tests without catalyst addition.  1.1.4.2 Carbon supported Catalysts for the Esterification of Bio-oil Low cost carbon materials, derived from activated petroleum coke (petcoke) and acidified, have been proposed as potential acid catalysts for esterification [22, 23]. Thermochemical treatment of non-porous petcoke [24] is used to generate pores and surface area at relatively low activation temperatures (450~900 ºC) and short activation times (ca. 2 h) [25]. The activated petcoke may have O functional groups present on the surface, depending on the activation conditions. The O functional groups decrease the hydrophobicity of the carbon surface, improving impregnation from aqueous solution when preparing supported metal catalysts [26]. Furthermore, the functional groups can act as nucleation centers for metallic crystallites leading to high metal dispersions [27, 28]. These properties suggest the possibility of metal impregnation onto activated petcoke to yield highly dispersed metals on the carbon that can then be used in catalytic gasification to generate mesoporous carbons. The mechanism of metal catalytic activation of carbon materials is similar to the mechanism of gasification of coal or other carbon materials [29, 30]. Transition metals, such as Ni, Fe, and Cr are usually applied in coal and carbon gasification  8  processes [31, 32]. However, the aim of catalytic activation is to stop the reaction once a high surface area or pore volume is obtained, without gasifying all of the carbon. As noted by Wu et al. [32], Ni and Cr initially present in petcoke, can act as catalysts to promote activation and pore development. Also, Tomita et al. [33] reported that Ni facilitates the steam gasification of coal. Metal sites on the carbon act as active sites for the reaction between C and H2O or CO2. However, for the development of mesopores, the required activation temperature is relatively high, and consequently the carbon yield is low [34, 35]. When using H2 as the environmental gas, pore development occurs because of the reaction between C and H2, also referred to as carbothermal hydrogen reduction (CHR). Accordingly, metals impregnated onto microporous carbons are reacted in a H2 flow while linearly increasing the reaction temperature. Although the process was developed for the synthesis of Mo2C catalysts supported on carbon [36], it also serves to enhance the pore structure of the carbon [37].   Porous carbon can be functionalized by acid treatment to yield acid catalysts [38, 39] or surface functionalized materials [40]. Acid treatment by H2SO4 or HNO3 is common [40-43] and after acid treatment, the carbon can be used as a catalyst support with O-functional groups which improves metal dispersion [43]. They can also be used directly in esterification reactions as acid catalysts. Liu et al. [44] reported that –SO3H functionalized carbon showed good activity for the esterification of oleic acid with methanol compared to Amberlyst-15. Also, Shu et al. [45] have reported using carbon-based solid acid catalysts to convert vegetable oil to fatty acids.    9  1.1.5 HDO of Bio-Oil and Catalysts Applied Oxygenated compounds are responsible for the instability, high viscosity, and high acidity of bio-oils which make them unsuitable for use without upgrading by, for example, hydrodeoxygenation (HDO) [6]. HDO is focused on removing the oxygen from the bio-oil by reaction with H2, leaving the bio-oil with high stability, low viscosity, and low acidity.  Typical catalysts investigated for HDO include conventional metal-sulfide hydrotreating catalysts (CoMoS and NiMoS), noble metals (Pd, Pt, Ru) and solid acids (zeolites) [46]. Novel catalysts such as Mo2C and Mo2N have also been considered [47, 48]. Conventional CoMoS and NiMoS catalysts have reduced stability during HDO because of the absence of S in the feed [49], resulting in O replacing the S present on the CoMoS or NiMoS catalysts, resulting in activity loss. Although noble metals have very high HDO activity, they can sinter in the presence of H2O at high temperature [50] and have relatively high H2 consumption because of their high hydrogenation activity [51]. Despite their high cost, noble metals are promising alternatives for HDO because of their high activity. Transition metal carbide catalysts have better stability compared to metal sulfide catalysts [52] and are known to possess noble metal-like catalytic properties, especially for hydrogenation reactions. The study of Moon et al. [53] shows that the initial activity of Mo2C for the hydrogenation of benzene is even higher than Ru and Pt. In the study of Frauwallner et al. [54], the activation energy for toluene hydrogenation was determined to be 58 kJ/mol, close to the 46 kJ/mol reported on Pt catalysts [55]. Table 1.2 compares the activity of noble metal catalysts and metal carbides for the HDO of various model reactants. Results suggest that metal carbide catalysts have comparable activity for HDO to noble metals. Hence, metal carbides remain an interesting potential alternative class of catalysts for HDO reactions.  10   Table 1.2: Comparison between noble metal catalyst and carbide catalyst in HDO reaction. Catalyst Model compound Main product Conversion% Reaction conditions Reference Ru/C Guaiacol Phenol 100 13.8 MPa, 300 oC, RT= 2 h [56] Pd/C Guaiacol Cyclohexane 99 13.8 MPa, 300 oC, RT= 4 h [56] Mo2C/C Vegetable oil Branched paraffins 80 2.5 MPa, 260 oC, RT= 2 h [57] Mo2C Benzofuran Ethylcyclohexane 97 5.1 MPa, 250 oC, N/A [58] Mo2C/CNF Guaiacol Phenol 99 5.5 MPa, 350 oC, RT= 4 h [59] Note: RT is reaction time in a batch reactor.  Previous studies have demonstrated that the catalyst support has a significant impact on the HDO activity of the catalysts. Several supports have been studied for HDO, including metal oxides (Al2O3, SiO2, ZrO2, TiO2), zeolites (HZSM-5) and carbon [60-63]. Catalysts supported on Al2O3, the most common support, have a relatively low activity and high coke yield. Zeolites have high activity as reported by Boonyasuwat et al. [64] and Bykova et al. [65], but their acidity makes them less favorable due to side reactions, especially oligomerization of the oxygenated compounds present in bio-oils [66]. Zeolites may be a good choice at relatively low temperature (< 200 oC) when the target reactants are acids and acetaldehydes. There remains a need to identify stable supports for HDO that have less impact on catalyst activity and selectivity compared to the conventional oxide supports and zeolites. One such candidate is carbon, a potentially neutral support, stable at high temperature, with high surface area that has been shown to yield catalysts  11  with higher activity than with Al2O3 or SiO2 supports [64, 67]. Alternative metal oxide supports such as ZrO2 and TiO2 also have good performance but are more expensive than Al2O3 and have low surface area making the dispersion of metal catalysts more difficult. The possible sources of carbon include petroleum coke (petcoke) and biochar and these are the focus of the present study.   Studies focused on the fundamentals of HDO catalysis use model reactants in place of crude bio-oil so as to simplify product analysis and data interpretation. For example, guaiacol (2-methoxyphenol), a commonly used model compound and a major component of pyrolysis oil from lignocellulosic biomass, allows the catalyst activity for the removal of O from Ar-OH and Ar-OCH3 to be compared to the hydrogenation of the phenyl group. The selectivity for hydrogenation versus deoxygenation is readily evaluated with this model compound, thereby providing experimental evidence for reaction mechanisms occurring on the catalysts that have also been identified by theoretical studies [68-72].   For noble metal catalysts operated at low pressure, O removal from guaiacol occurs via an initial demethylation to yield catechol, with subsequent Ar-OH bond cleavage to yield phenol and finally benzene. After deoxygenation, hydrogenation of the phenyl ring can also occur to yield cyclohexane. The sequence of the reaction steps has been confirmed experimentally at low reaction pressure [46, 73] and by density functional theory (DFT) studies on Ru [70, 71]. Alternatively, the reactant is hydrogenated first to yield 2-methoxycyclohexan-1-ol which is subsequently deoxygenated to yield cyclohexane [74, 75]. On Mo2C catalysts, however, the removal of one O by scission of the Ar-OCH3 bond always occurs first without the formation of catechol at both high pressure [59] and low pressure [76]. Chang et al. [77] compared the performance of noble  12  metal and carbide catalysts at high reaction temperature. The selectivity to benzene by direct deoxygenation (DDO) was 69.5 and 83.5 wt.% at 380 oC, 4 MPa using Ru/C and Mo2C/C catalysts, respectively. At these conditions the primary products were rapidly consumed, and thermal reactions were emphasized.  1.1.6 Carbon Supported Mo2C Catalysts 1.1.6.1 Carbon Sources and Chemical Activation of Carbon Support Several materials are available that can be used as a source of carbon for porous carbon materials development [78, 79]. One is petroleum coke (petcoke, PC), a by-product of coking processes used to upgrade bitumen derived from the Canadian oil sands [79]. However, this coke is an environmental hazard due to its dust-like property and high S content (~7 wt.%) [80-82]. Bio-char is another useable carbon precursor produced from the pyrolysis of organic matter originating from a diverse range of biomass materials [78, 83]. Utilization of bio-char has begun to draw attention as bio-oil studies proceed. The biochar used in the present study is a by-product of the pyrolysis of wood chips. Unlike petcoke, bio-char does not contain a high content of S (< 0.002 wt.%), instead it contains more O and N that amounts to 5-15 wt.% [84, 85]. Nevertheless, carbon is still the main component of biochar and similar treatments can be used to activate both petcoke and biochar. Therefore, modified petcoke and biochar as valuable porous carbon materials have great potential in catalysis applications [86-88].  The thermochemical treatment [24] of non-porous petcoke and biochar is used to generate pores and surface area at a relatively low activation temperature between 450 and 900 ºC in a short activation time of approximately 2 h [25]. The porosity of the product carbon is affected by  13  several operating variables, including the heating rate, temperature, and the ratio of activating agent to carbon material. Activating agents, such as KOH and K2CO3, play an important role in the pyrolytic decomposition process [89-91]. Among these compounds, KOH is the most widely used in petcoke activation, as reported by several authors [90-93]. Although addition of KOH is highly effective in the production of high surface area activated carbons, most of the developed surface area resides in micropores. The activated sample may also have O functional groups present on the surface, depending on the activation conditions.   1.1.6.2 Synthesis Method of Mo2C/C Catalyst Transition metal carbide catalysts, especially Mo carbide, have shown promising results in HDO reactions as shown in Table 1.2. The method most often used to prepare these catalysts is temperature-programmed reduction (TPR) in a CH4/H2 flow. Accordingly, a continuous gas flow of typically 10% CH4 in H2 is passed over the catalyst precursor as the reaction temperature is increased to as high as 800 oC according to a linear temperature program. The acquired Mo2C is usually highly crystallized. However, the presence of CH4 and a specific CH4 to H2 ratio are required to control the Mo2C formation process [58, 94, 95].   Carbothermal hydrogen reduction (CHR) has received significant attention in the preparation of Mo2C on carbon supports due to the less severe reaction conditions needed to obtain the Mo2C. Typically a carbon material is impregnated with a Mo salt, dried and then reduced in the presence of H2 as the temperature is increased up to 800 °C.  Reduction temperature and type of carbon material are the two most important factors that impact the Mo2C formation process [96, 97]. The impregnated metal sites on the carbon material can act as active sites for the reaction  14  between C and H2. However, to obtain mesoporous structures, the required temperature must be relatively high, and this reduces the carbon yield significantly.  Although the CHR is often used as a catalyst preparation method to synthesize Mo2C supported on carbon [36], it can also serve as a metal activation method to alter pore structure of the carbon material. CHR is rarely considered for the activation of carbon due to the relatively higher cost of hydrogen compared to nitrogen. However, if CHR is used to generate a product of high additional value, such as mesoporous carbons that can be used as catalyst supports or for fuel cell catalyst supports, the cost of hydrogen may be less important.  1.1.6.3 Oxygen Effect and Promoter Effect on Mo2C Stability in HDO One important issue related to Mo2C catalysts is that they are unstable because of the impact of O, where the adsorption and/or exchange of O may change the chemical properties of the active sites. Choi et al. [98, 99] has experimentally studied the O effect on Mo2C by CO titration and diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS). The results show that residual O that remains after carburization has an impact on Mo2C sites, both quantitatively and qualitatively. Moreover, a correlation between the O content of the Mo2C catalyst and the hydrogenation ability has been noted, where increased O adsorption decreased hydrogenation activity. The effect of O from O-containing reactants/products during HDO (anisole [100, 101], isopropanol [102, 103], acetone [104] or H2O [105] etc.), has been studied extensively as well. For example, Lee et al. [100, 101] reported on the impact of O on Mo2C catalyst in HDO of anisole. Their study suggested that the incorporation of O into Mo2C reduced the activity of Mo2C and altered the catalyst acidity, as evidenced by the high benzene selectivity (DDO  15  route >90%). This discovery was further studied by Sullivan et al. [102-104] through tuning metallic-acidic bifunctionality of Mo2C using O2 cofeed for in-situ oxidation that occurs because of the highly oxophilic nature of Mo2C. In Mortensen et al.’s study [105], the rapid deactivation of Mo2C/ZrO2 was discovered with a cofeed of water (30 wt%) and in-situ XRD results indicated the formation of MoO2.The DFT study conducted by Liu and Rodriguez [106] identified the O functionalized Mo2C surface in the water gas shift reaction (WGS) and noted that the O absorbed on C terminated Mo2C (O_C-Mo2C) displays a higher WGS activity than that of O absorbed-Mo terminated Mo2C (O_Mo-Mo2C), indicating that the generated Mo oxycarbide can function as active sites as well. Similar results are also reported by Nagai et al. for Co-Mo2C catalyst [107]. Thus, the impact of O on the Mo2C surface can be determined and in order to improve catalytic activity and stability, the issue of Mo2C modification/deactivation by O must be addressed.    One reasonable approach to improve Mo2C catalyst stability is to provide an environment enriched in activated H* to reduce the oxidation of Mo2C during reaction. The addition of a metal promoter to modify the Mo2C catalyst is one possible approach to achieve this goal. For example, some promoters, such as Ni, have a relatively strong H2 dissociation activity. The activated H* can rapidly remove adsorbed O from metallic active sites. This is why severe reaction conditions (high H2 pressure) are usually required in HDO to reduce catalyst deactivation [6, 108]. Moreover,  one study that reported a zeolite encapsulated Mo2C for the HDO of anisole, showed a higher activity and stability of the Mo2C catalyst for up to 1000 min time-on-stream with the assistance of Brønsted acid sites that surround the Mo2C [109]. In addition, Shi et al. [110] reported the performance of a bi-functional Ni-Mo2C catalyst (Ni:Mo ratio = 1:2) for dry reforming of CH4 with CO2. The results indicate that the Ni addition limits  16  the oxidation of Mo2C active sites by activating CH4 to produce C* and H2. Similarly, a good performance of Ni-Mo2C was identified in Smirnov et al.’s study of the HDO of anisole and ethyl caprate [111].   The presence of a 2nd metal can also be used to tune Mo2C catalyst properties. Among numerous promoters, Ni and Co are frequently used in Mo-based catalysts in industry and many studies have reported these catalysts for hydrotreating [112-114]. Thus, Ni and Co have often been considered as promoter candidates for Mo2C catalysts. As reported previously [115], Ni-Mo2C/AC catalyst has shown a higher activity than unpromoted Mo2C for the hydrodesulfurization (HDS) of dibenzothiophene (DBT). Similarly, Ni promoted W2C has a high yield (~61%) of ethylene glycol (EG) in cellulose conversion due to the synergistic effect between Ni and W2C [116]. Zhao et al. [117] reported the significant higher chemo-selective reduction of aromatic nitro compounds on Co-Mo2C, which is comparable to precious metals. In addition to Ni and Co, other metal promoters have been reported, focusing on carbide catalyst activity and stability improvement in various reactions [118-121]. For example, as mentioned in Leal et al. [118] ‘s study, the presence of Pd can produce a cleaner W2C surface which could in turn enhance the catalyst ability in cellulose conversion. From the above, one can conclude that the Mo2C surface properties can be altered based on the addition of a 2nd metal.  1.1.7 Summary Fast pyrolysis oil, also known as bio-oil, has been identified as an alternative to fossil fuels that has much lower GHG emissions than conventional fossil fuels (Section 1.1.1). However, because of the properties of bio-oil (Section 1.1.2), particularly the presence of excessive amounts of  17  oxygen and water, bio-oil upgrading is required before the oil can be directly combusted in an engine. Crude bio-oil upgrading methods can be summarized in two groups, as reported in Section 1.1.3: (i) phase separation and (ii) chemical conversion. Phase separation processes can be operated with low energy consumption, yet their deoxygenation ability is too low for bio-oil applications. Chemical conversion processes can reduce the oxygen content to the target level for engine combustion. However, chemical conversion requires higher temperature and higher energy consumption compared to phase separation processes. In addition, coke formation during chemical conversion is a significant problem that can limit upgrading potential.  Among the chemical conversion processes, esterification and HDO processes (see Section 1.1.4 and 1.1.5, respectively) are promising. Relative to conventional catalysts applied in esterification reactions, carbon supported catalysts have more resistance to coke formation and can be derived from low cost materials such as petcoke. For the HDO reaction, different types of catalyst have been discussed. Conventional catalysts such as NiMoS/Al2O3 are not stable in the absence of S. Noble metal catalysts have high activity, however, they face sintering problems and are very expensive. Transition metal carbide catalysts have noble metal-like catalytic behavior and are much less expensive than precious metals. Hence, they are promising catalysts for HDO reactions.   The synthesis method for Mo2C catalysts was introduced in Section 1.1.6. Although the TPR-CH4/H2 process is more often studied and forms more crystallized Mo2C, the CHR process is a more practical approach in terms of gas environment. Finally, carbon supported Mo2C catalyst performance for HDO is introduced. Even though the results are promising, most studies are  18  based on model compounds and Mo2C catalyst performance with investigations using real bio-oil needed before application of the technology can occur.    1.2 Objective of the Thesis The overall objective of the present study is to upgrade bio-oil derived from woody biomass, especially aimed at removing oxygen from the bio-oil without coke formation and identifying a cost-effective catalyst for the esterification and HDO reaction. Several key tasks were identified to meet this objective.   • Develop methods of synthesis of Mo2C supported on mesoporous carbon using petcoke and biochar as the carbon source. Assess the Mo2C catalysts supported on activated petcoke and biochar (Mo2C/APC and Mo2C/ABC) for esterification and HDO reactions. • Compare the carbon supported Mo2C catalyst to existing noble metal catalysts for HDO.  • Investigate the effect of oxygen and reaction conditions on the Mo2C catalyst stability and activity. Investigate the role of metal promoters to improve the Mo2C catalyst. • Establish the activity and stability of the most promising Mo2C catalyst formulation using a real feedstock (fast pyrolysis bio-oil) and compare the Mo2C catalyst to noble metal catalysts.  1.3 Approach The approach to achieve the objectives of this study was primarily experimental and is described as follows: 1) Investigate by experiment, the synthesis, characterization and testing of activated bio-char (ABC) and activated petroleum coke (APC) supported Mo2C catalysts. The prepared  19  catalysts were tested for the HDO using the model reactant of 4-methylphenol. The catalyst tests were conducted in an autoclave batch reactor. The prepared APC catalysts were treated with acid and then tested for the esterification reaction using 1-butanol and acetic acid as model reactants. The catalyst tests were conducted in a stirred-batch reactor. Kinetic analysis is used to link the catalyst properties to catalyst performance.  2) Investigate by experiment, the synthesis, characterization and testing of activated charcoal (AC) supported Mo2C catalysts. The prepared catalysts were tested for HDO using the model reactant of 2-methoxyphenol. Commercial Ru/AC and Pd/AC catalyst were tested for comparison. The catalyst tests were conducted in a batch reactor. Kinetic analysis is used to link the catalyst properties to catalyst performance.  3) Explore the effect of O on Mo2C catalysts in a continuous flow fixed bed reactor. The effect of temperature and pressure on the activity and stability of Mo2C catalysts was investigated by experiment and catalyst characterization. The promotion effect of Ni, Pd and Cu were investigated in the same system. DFT calculation using VASP was used to study the effect of surface property changes after promoter addition on the HDO reaction.   4) Prepare Ni promoted Mo2C/C catalysts and assess using real bio-oil, via esterification and HDO reactions in a semi-batch reactor. Ru/C and Ru/SiO2-Al2O3 were also tested under similar conditions for comparison. Products recovered following phase separation were analyzed using several analytical techniques. Catalysts before and after reaction were also characterized.  20   1.4 Outline of Dissertation This dissertation is presented in six chapters, as described in Figure 1.2. The study conducted in Chapter 2 is focused on the synthesis of carbon supported Mo2C catalyst and its application in model compound reactions, as the starting point to test the potential of the catalyst for esterification. Next, in Chapter 3, the effect of two different carbon supports on the catalyst performance in a model compound study of HDO, was investigated. Subsequently, a comparative study between synthesized Mo2C and noble metal catalysts is reported in Chapter 4. Chapters 5 describes further studies of Mo2C/APC catalysts focusing on the effect of O and metal promoter on the HDO reaction. Finally, an improved Mo2C catalyst was utilized in real bio-oil upgrading as described in Chapter 6.   Chapter 1: Provides a brief introduction of bio-oil properties, bio-oil upgrading technologies, petroleum coke/biochar, carbon supported Mo2C catalyst synthesis 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: Presents the chemical activation of petcoke to make initial micropores, the CHR process to make Mo2C sites while generating mesopores, and the application of these prepared catalysts in model compound studies of acetic acid and 1-butanol as testament of catalyst performance in the esterification reaction.  Chapter 3: Compares two different carbon support sources, the surface property changes during chemical activation and the CHR process, and the application of these prepared catalysts in  21  model compound HDO studies of 4-methyl phenol as testament of catalyst performance in the hydrodeoxygenation reaction.  Chapter 4: Compares synthesized Mo2C/AC with Ru/AC and Pd/AC in model compound HDO study of 2-methoxy phenol to determine the differences between Mo2C and noble metal catalysts in terms of activity and selectivity, while gathering the apparent kinetic data as a guideline for the real bio-oil study conducted later. Chapter 5: Presents the effect of promoter such as Ni on the Mo2C catalyst during the HDO reaction using model compound dibenzofuran, based on both kinetic data generated from experiments and calculation results from DFT using VASP, especially on the improvement of hydrogenation ability. Chapter 6: Compares the promoted Mo2C catalyst with a Ru/C catalyst for real bio-oil upgrading. A detailed study of how the complicated oxygenated compounds transform during upgrading processes using multiple analysis techniques such as GC-MS and 1H-NMR are presented. Chapter 7: Summarizes the conclusions of previous chapters and provide the recommendations for future work.   22   Figure 1.2: Schematic illustration of both model compounds and real bio-oil study of bio-oil upgrading process using carbon supported Mo2C catalysts (Mo2C/C).    23  Chapter 2: Synthesis of Carbon Supported Mo2C Catalysts and its Application in Esterification 1 2.1 Introduction The aim of the research reported in Chapter 2 is to transform Canadian oilsands petcoke into a valuable mesoporous carbon with sufficient surface acidity so that it can act as an esterification catalyst. Carbon is chosen as the catalyst support to reduce coke formation during reaction. Esterification is the main reaction of the first step of the proposed two-step bio-oil upgrading process described in Chapter 1. Acidified carbon supports are proven to be effective in catalysing the esterification reaction and the chemical activation and Mo2C synthesis process have the potential to impart acid sites to the carbon surface. A multi-step approach was used in which thermochemical activation of the petcoke with KOH was followed by impregnation with ammonium heptamolybdate prior to carbothermal hydrogen reduction (CHR). An acid wash was then used to reduce the Mo content of the carbon and acidify the catalyst surface. The acid wash could also increase mesoporosity of the catalyst, which could give reactants more access to the active sites, at the cost of surface area loss. The impact of both the Mo loading and the CHR temperature on the carbon porosity is reported and the benefits of using Mo2C in the preparation of the acidified mesoporous carbon is demonstrated. Finally, the catalytic activities of the acidified mesoporous carbons, derived from Mo2C/APC and the APC alone, assessed using acetic acid and 1-butanol as model reactants for esterification, are compared. Hence, results from Chapter 2, demonstrate that the waste carbon (petcoke) can be utilized as a catalyst support                                                  1 A version of this chapter has been published: S. Liu., H. Wang, P. Neumann, C. S. Kim and K. J. Smith, “Esterification over an acid treated mesoporous carbon derived from petroleum coke,” ACS Omega (2019) 4 (3), 6050-6058.  24  following chemical activation. The carbon supported Mo2C catalyst has an ability to catalyze esterification even without acid treatment.  2.2 Experimental 2.2.1 Materials Petroleum coke (PC) recovered from a delayed coker was provided by Suncor Energy Inc. The PC was ground and sieved to 90 ~ 180 μm and then dried overnight at 110 oC before use. Ultimate analysis of the received material is reported in Appendix E – Table E.1. Ammonium heptamolybdate tetrahydrate (AHM; (NH4)6Mo7O24∙4H2O) was used as the Mo precursor. The reactants 1-butanol (Aldrich, 99.9%) and acetic acid (Aldrich, 99%) were used as-received for the esterification reaction.   2.2.2 Mo2C/APC Catalyst Preparation About 3.5 g of the raw petcoke (BET area: ~ 3 m2/g and pore volume:~ 0.01 cm3/g) [122] was dry mixed with KOH in a KOH:petcoke mass ratio of 3:1 and activated in a N2 flow (200 mL(STP)/min) while heating in a tube furnace at a ramp rate of 5 oC/min to the final temperature and then holding this temperature for 2 h. After subsequent cooling to room temperature, the sample was washed with 1M HCl solution to remove excess KOH and then dried at 110 oC for 8 h. The obtained activated petcoke was designated as APC_xxx, where xxx is the final temperature of the activation.  To study the development of mesopores during CHR, a series of Mo2C/APC catalysts were prepared at different CHR temperatures and Mo loadings, using APC_800 as support. For lower  25  Mo loadings (≤ 2 wt%), ca. 3.5 mLsolution/gAPC was used for the wet impregnation and for higher Mo loadings (>2.0 wt%), 5.0 mLsolution/gAPC was used for the impregnation. The impregnating solution was prepared from AHM dissolved in a 10 vol.% acetone/water mixture. After impregnation, the samples were stabilized at room temperature for 3 ~ 4 h prior to drying at 110 oC overnight. The AHM/APC precursors were then carbothermally reduced under continuous H2 flow (100 mL(STP)/min) at a designated temperature in a fixed-bed quartz microreactor to convert the AHM to Mo2C [123]. The temperature was increased from room temperature to the final temperature (600 ~ 1100 oC) at a ramp rate of 1 oC/min. The final temperature was held for a further 90 min before switching to a N2 flow (50 mL(STP)/min) and quenching the sample to room temperature. A GC-FID was connected to the outlet of the CHR microreactor to quantify the CH4 produced during the CHR process. The acquired samples were named as xxMo2C/APC_Ryyy, where xx represents the Mo loading and yyy is the final reduction temperature (ºC) of the CHR.   Several samples were subsequently acid washed to remove the Mo. Accordingly, 10 mL of fuming sulfuric acid (50 wt% SO3) was placed in a 50 mL round bottom flask and heated to 80 oC using a heating mantle. 0.5 g of the selected Mo2C/APC sample was added to the flask and reacted at 80 oC for 4 h under reflux and mixing using a magnetic stirrer. After cooling to room temperature, the acquired mixture was diluted with distilled water. The solid was recovered by filtration after washing with distilled water and dried in an oven for 12 h before use. The collected samples are identified as acid-T xxMo2C/APC_Ryyy. An acid wash of the APC_800 (without Mo2C) was done similarly to determine the impact of Mo on the acid treated carbon catalysts.  26   The carbon loss during the CHR process is defined as: Burn off (%) = 100 ×(𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙−𝑚𝑎𝑓𝑡𝑒𝑟) 𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙      (Eq. 2-1) where 𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙 represents the initial mass of the catalyst precursor prior to the KOH activation or prior to the CHR; 𝑚𝑎𝑓𝑡𝑒𝑟 is the mass of the carbon or catalyst after the KOH activation or CHR treatment, respectively.  The final yield of the Mo2C/APC was calculated as follows: Yield Mo2C APC⁄  (%) =  ((100 − burn off %) × (APC yield %))/100  (Eq. 2-2)  2.2.3 Catalyst Characterization 2.2.3.1 N2 Adsorption and Desorption The N2 physisorption isotherms of the prepared samples were measured at -196 oC using a Micromeritics ASAP 2020 analyzer to determine surface area, pore volume, and pore size distribution. Prior to the analysis, ~ 0.1 g of the sample was degassed at 200 oC (100 mm Hg) for 4 h. The pore size distribution was calculated by the non-local density functional theory (NLDFT) based on a N2-DFT model assuming a pore slit geometry, which is common for carbon materials. The surface area was calculated from the adsorption isotherm in the relative pressure range (p/po) of 0.01-0.30 using the 2D-NLDFT method. The pore volumes were obtained using the 2D-NLDFT method as well.    27  2.2.3.2 Transmission Electron Microscopy (TEM) The particle size of the Mo2C on the Mo2C/APC was determined from transmission electron microscopy analysis (TEM, FEI Tecnai Osiris operated at 200kV with a resolution limit of 1.4 Å). The presence of certain crystal phases was confirmed by d-spacing measurement. The samples were ground to a fine powder and dispersed in ethanol by sonication. One or two drops of the suspension was placed on a 300-mesh lacey carbon film. The TEM images were analysed and > 100 particles measured to determine a particle size distribution that was fitted to a log-normal distribution from which the average size of the metal clusters/particles was estimated.   2.2.3.3 Elemental Analysis  The ultimate analysis of petcoke was conducted in an elemental analyzer (Perkin-Elmer 2400 series II CHNS/O) at a combustion temperature of 975 oC. Inductively coupled plasma optical emission spectroscopy (ICP-OES) was used for elemental analysis of the samples that were acid digested in aqua regia at 110 oC to extract the metals from the support prior to the analysis.  2.2.3.4 Ammonium Temperature-Programmed Desorption (NH3-TPD) Ammonium temperature programmed desorption (NH3-TPD) was used to determine the acidity of both the Mo2C/APC and the corresponding acid-treated samples. The NH3 desorption was measured using the TCD detector of a Micromeritics AutoChem 2920 unit. The sample was treated in a 15 vol% NH3 in He flow of 50 mL(STP)/min at 70 oC for 60 min, followed by flushing in pure He for 2 h at the same temperature to remove physiosorbed NH3 species. The TPD signal was collected as the sample was subsequently heated under the same He flow from 100 oC to 550 oC at a ramp rate of 10 oC/min, holding the final temperature of 550 oC for 30 min.  28   2.2.3.5 Raman Spectroscopy  A Horiba Jobin Yvon LabRAM HR800 operated at a laser power of 10 mW, a grating of 1800 g/m, resolution of 0.65 cm-1 and laser wave length of 532 nm was used for Raman analysis of the carbons. The samples were analysed without any pre-treatment. The analysis was used to determine the degree of disorder of the synthesized carbon materials by comparing the peak area of the disordered carbon (I(D)) to graphitic carbon (I(G)). The Raman spectra peak intensity ratio I(D)/I(G) was calculated, where I(D) is the intensity of the D band peak at 1600 cm-1 and I(G) the peak intensity of the G band at 1350 cm-1.  2.2.3.6 Esterification of Mo2C/APC and Acid-treated Mo2C/APC The Mo2C/APC and the corresponding acid-treated samples were assessed for esterification using the model reactants acetic acid and 1-butanol at a mole ratio of 1:1 reacted at 77 oC for 2 h in a stirred batch reactor under a N2 atmosphere [124]. The catalysts 0.25Mo2C/APC_R1000, 0.50Mo2C/APC_R900, 1Mo2C/APC_R900, 2Mo2C/APC_R900 and the APC_800 were assessed together with their corresponding acid-treated samples. In each case the same amount of catalyst was used (ca. 0.2 grams) and the reactor liquid volume was held constant at 20 mL. After the reaction, the liquid product was separated from the catalyst by centrifuge and diluted 50x’s in acetone prior to quantitative analysis using a Shimadzu (QP-2010-S) GC-MS and a Restek RTX5 30 m × 0.25 mm capillary column. The conversion of 1-butanol after 2 h was taken as a measure of catalyst activity. Repeat experiments showed that measurement error associated with the conversion was ±5% of the measured conversion. The esterification reaction was assumed to proceed through a 2nd-order reaction with the reaction rate [125] given  29  by: 𝑟 (𝑚𝑜𝑙. 𝑔𝑐𝑎𝑡−1. 𝑠−1) = 𝑘∗𝐶𝐴𝐶𝐵 where 𝐶𝐴 is the butanol concentration and 𝐶𝐵 the acetic acid concentration. Assuming a perfectly mixed phase inside the reactor volume and writing the component concentrations in terms of butanol conversion X, one obtains the equation: 𝑑𝑋 𝑑𝑡⁄ =−𝑘𝐶𝐴0(1 − 𝑋)2 with 𝑘 = 𝑘∗𝐶𝑐𝑎𝑡, where 𝐶𝐶𝑎𝑡 is the catalyst concentration in the mixed reactor and 𝐶𝐴0 is the initial concentration of butanol. Hence the 2nd-order rate constant 𝑘 is determined from the equation 𝑘 = 𝑋 ∗ [𝐶𝐴𝑜𝑡(1 − 𝑋)]−1and the butanol conversion data [126]. Finally, the catalyst turnover frequency (TOF) was estimated by dividing the rate 𝑟 measured at the conversion 𝑋 by the NH3 uptake.  2.3 Results The properties of the activated petcoke (APC_800) and the acid washed APC_800 are summarized in Table 2.1, together with the acidity (NH3 uptake) and esterification activity of both materials. From these data, one concludes that the APC_800 is microporous with minimal acidity and esterification activity. However, the Raman data (Table 2.1 and Appendix E.1 - Figure E.1) show that the APC_800 has more disorder (i.e. is more amorphous) than the raw petcoke. Following the acid wash, both the mesoporosity (Figure E.3) and the acidity of the carbon increased significantly and there was a corresponding increase in the esterification activity of the acid washed APC_800 (i.e. Acid-T APC_800).   Table 2.1: Textural properties, acidities and reaction rates of the raw petcoke and the activated petcoke before (APC_800) and after acid treatment (Acid-T APC_800)  Raw petcoke APC_800 Acid-T APC_800 Surface area, m2/g 3 2037 1723 VTotal, cm3/g 0.01 0.84 0.91  30  Vmeso, cm3/g <0.01 0.07 0.20 NH3 uptake, mol/g - < 10 1340 X, mol % - < 1 29.9 k, m3.(mol.s)-1 x10-9 - < 0.01 4.7 TOF, s-1 x10-2 - - 1.5 Raman I(D)/I(G) 1.5 ± 0.01 2.9 ± 0.1 - Note: X is mol % butanol conversion; k is 2nd-order rate constant derived from butanol conversion; TOF is the turnover frequency based on NH3 uptake.   The impact of utilizing CHR and the synthesis of Mo2C to impact the properties of the activated petcoke (APC_800) are shown in the property data of the Mo2C/APC, prepared at different Mo loadings and CHR temperatures, as summarized in the 3D plots of Figure 2.1, with the data values of the AHM/APC precursors and the Mo2C/APC provided in Table E.2. Previously, the effect of CHR temperature at a Mo loading of 10 wt% was reported [37]; whereas, the data reported herein provides information on the interacting effects of Mo loading and CHR temperature on the Mo2C synthesis and the textural properties of the carbon.   The data of Figure 2.1 show that both surface area and total pore volume (Vtotal) of the Mo2C/APC generally decreased with increased Mo loading, regardless of the CHR temperature (Figure 2.1(a, b)). A maximum value occurred at a reduction temperature between 800 and 900 oC at low Mo loading (≤ 2 wt.%). The highest surface area was obtained for the 0.5Mo2C/APC_R900 sample. Once the reduction temperature reached 700 oC, the surface area remained relatively high (> 1800 m2/g) with the Mo loading ≤ 2 wt%.   31        Figure 2.1: 3D plot of physical properties of Mo2C/APC samples with various Mo loadings and CHR temperatures: (a) Surface area; (b) total pore volume (Vtotal); (c) mesopore volume (Vmeso).   32  Figure 2.1(c) shows that the mesopore volume increased with increased CHR temperature at each Mo loading, provided the CHR was conducted at 500 to 900 °C. Above 900 °C the mesopore volume decreased. The maximum mesopore volume occurred for the 1Mo2C/APC_R900 sample. The data of Table E.2 (Appendix E.1) show a significant increase in mesopore volume for all the Mo2C/APC samples compared to their corresponding precursors. The change in textural properties of the Mo2C/APC is a result of the hydrogenation of carbon, yielding mostly CH4 during CHR. The GC-FID results showing the CH4 concentration profiles during CHR of the 2Mo2C/APC_R700, 5Mo2C/APC_R700 and 10Mo2C/APC_R700 samples are reported in Figure E.2. The CH4 concentration in the CHR product gas increased as the CHR temperature increased but then decreased once the final temperature was reached. Also, higher Mo loading resulted in increased carbon hydrogenation, as reflected in the higher CH4 concentration in the CHR product gas.  Finally, the average particle size of the Mo2C present on the Mo2C/APC after CHR is reported in Table 2.2, based on the log-normal distribution applied to the TEM particle size measurements. The data show that increased Mo loading and/or increased CHR temperature increased the average Mo2C size of the Mo2C/APC samples.    33  Table 2.2: Particle/cluster size of Mo2C/APC samples with different Mo loadings and different reduction temperatures. Samples CHR Temperature, °C Mo nominal loading, wt % Ave. particle size (nm) a 1Mo2C/APC_R700 700 1 5.8±0.3 2Mo2C/APC_R700 700 2 6.0±0.2 5Mo2C/APC_R700 700 5 7.0±0.4 10Mo2C/APC_R700 700 10 9.8±0.7 1Mo2C/APC_R800 800 1 9.4±0.2 1Mo2C/APC_R900 900 1 10.4±0.2 1Mo2C/APC_R1000 1000 1 12.6±0.4 0.25Mo2C/APC_R1000 1000 0.25 11.4±0.4 0.50Mo2C/APC_R1000 1000 0.5 10.3±0.5 2Mo2C/APC_R1000 1000 2 22.2±0.4 a. Estimated by analyzing ≥ 100 particles/clusters and fitting size distribution to a lognormal distribution from which the average particle size was determined.   The physical properties of the acid-treated Mo2C/APC samples are compared to the data of the Mo2C/APC prior to acid treatment in Tables 2.3 and 2.4. The data show that the acid wash resulted in 50~70% Mo removal from the catalysts and there was no significant change in surface area after acid treatment, except for the 2Mo2C/APC_R900 sample. Generally, the mesopore volume increased after acid treatment for all samples, presumably due to the removal of Mo by the acid. The acid treated sample Acid T-1Mo2C/APC_R900 had the highest Vmeso (0.70 cm3/g) and Vtotal (1.59 cm3/g).  Tables 2.3 and 2.4 also summarize the NH3-TPD results of the Mo2C/APC and acid-treated Mo2C/APC samples. As expected, the acid treated Mo2C/APC samples had significantly higher  34  acidity in all cases except for the Acid-T 2Mo2C/APC_R900 sample. The Acid-T 1Mo2C/APC_R900 sample had the highest number of acid sites (1942 µmol/g).  Table 2.3: Comparison of Mo2C/APC and acid treated Mo2C/APC catalysts prepared at a CHR temperature of 1000 °C: textural properties, acidities and reaction rates.  0.25Mo2C/APC Acid-T 0.25Mo2C/APC Mo content, wt % 0.34 0.09 Surface area, m2/g 1900 1578 VTotal, cm3/g 1.09 1.32 Vmeso, cm3/g 0.17 0.62 NH3 uptake, mol/g 160 1511 X, mol % 21.2 39.9 k, m3.(mol.s)-1 x10-9 5.6 13.7 TOF, s-1 x10-2 9.7 1.5    35  Table 2.4: Comparison of Mo2C/APC and acid treated Mo2C/APC catalysts prepared at a CHR temperature of 900 °C: textural properties, acidities and reaction rates.   0. 5Mo2C/APC Acid-T 0.5Mo2C/APC  1Mo2C/APC Acid-T 1Mo2C/APC  2Mo2C/APC Acid-T 2Mo2C/APC Mo content, wt % 0.56 0.16 1.22 0.58 2.85 1.23 Surface area, m2/g 2172 1823 1786 1962 1673 908 VTotal, cm3/g 1.19 1.45 1.14 1.59 0.98 1.01 Vmeso, cm3/g 0.34 0.63 0.49 0.70 0.42 0.65 NH3 uptake, mol/g 174 1876 375 1942 701 626 X, mol % 18.5 43.0 17.8 52.6 18.5 18.7 k, m3.(mol.s)-1 x10-9 4.7 15.6 4.5 22.9 4.7 4.8 TOF, s-1 x10-2 8.0 1.2 3.6 1.2 2.0 2.3    36  1-Butanol and acetic acid were chosen as model reactants to determine the esterification activity of the prepared catalysts. The sole product of the reaction is butyl acetate and the selectivity were > 99 wt% for all tests reported herein. The results of the activity tests at 77 oC for the APC_800 and acid treated APC_800 are reported in Table 2.1 and the Mo2C/APC and acid-treated Mo2C/APC catalysts are compared in Tables 2.3 and 2.4. All the acid-treated catalysts showed significantly higher conversion than the untreated samples, while the acid treated Mo2C/APC catalysts with a Mo loading ≤ 1 wt% had higher activity than the acid treated APC_800 (without Mo). The Acid-T 1Mo2C/APC_R900 catalyst had the highest conversion among all samples tested with 52.6% 1-butanol conversion after 2 h reaction. The estimated 2nd-order kinetic rate constants (k), plotted against the number of acid sites as determined by NH3-TPD for both the Mo2C/APC and Acid-T Mo2C/APC catalysts, are shown in Figure 2.2. For all Mo2C/APC catalysts without acid treatment, the kinetic constants are approximately constant 4.9 ± 0.5 x 10-9 m3.(mol.s)-1, independent of the Mo2C loading on the APC. The difference in number of acid sites on these catalysts is relatively small such that the impact on the reaction rate is not evident. However, for the Acid-T Mo2C/APC catalysts, the kinetic constants clearly increase with the number of acid sites present on the catalyst.  37   Figure 2.2: A correlation between adsorbed NH3 and the 2nd-order esterification reaction rate constants (k): (●) Acid-T Mo2C/APC; (■) Mo2C/APC; (▲) Acid-T APC_800; — trendline   2.4 Discussion The data of Table 2.1 show that the activated petcoke is readily converted to an acid catalyst with significant esterification activity following acid washing with fuming H2SO4. Compared to the APC_800 sample, the acid treatment (Acid-T APC_800) increased the porosity of the activated petcoke and generated acid sites on the carbon. The data of Table 2.3 and 2.4 show, however, that the synthesis of Mo2C yields Mo2C/APC catalysts with significantly higher porosity, acidity and esterification activity after acid treatment when compared to the Acid-T APC_800 (i.e. with no Mo2C).    38  The textural properties of the Mo2C/APC (Figure 2.1 and Table E.2) reflect the pore enlargement that occurred during the CHR process. As shown in Figures E.4 and E.5 (Appendix E.1), the pore size distributions, calculated using NLDFT, show a shift from micropores to mesopores after CHR up to a temperature of 900 °C. These data indicate that the micropores are being widened into mesopores during CHR. At a CHR temperature of 1000°C an increase in surface area was observed, indicative of new pores being generated, possibly by pore lengthening. The shape of the isotherms suggest slit-like structures according to the study of Wigmans [127]. A small loss in mesopore volume also occurred at 1000 oC due to the high burn-off at this temperature. Hence, two modes of pore development occur during CHR: (i) enlargement of existing pores and (ii) generation of new pores. The widening of an existing pore results in the replacement of the micropore volume with mesopore volume, without significantly increasing surface area. The generation of new pores results in increased surface area, micropore volume and mesopore volume. The latter mode of pore growth, known as channeling, is mainly due to the high crystallinity of graphite [34, 128], with the size of the resulting pores determined by the particle size of the metal catalyst [128, 129]. Note that the data of Table E.3 show that the same CHR treatment up to 800 oC carried out with the Mo species loaded onto the raw (low surface area) petcoke rather than the APC, did not result in significant increases in surface area or pore volume. This reflects the limited ability of the Mo species to create new pores at the lower CHR temperatures, without some initial porosity generated by the KOH treatment. The Raman data reported in Table 2.1 show an increase in the I(D)/I(G) band intensity ratio after KOH activation, indicative of an increase in disordered (amorphous carbon) formation following activation. Hence, carbon associated with the pore walls activated during the KOH treatment will be less crystalline and more reactive than the carbon associated with the original petcoke that was not  39  activated by the KOH. Consequently, in the initial stage of the catalytic hydrogenation at lower reaction temperatures, amorphous carbon associated with the pore walls is hydrogenated, resulting in the widening of pores. As the amorphous carbon is removed, pore lengthening will also occur as the less reactive, crystalline carbon is hydrogenated at higher temperature. The physical properties of Mo2C/APC generated at temperatures below 900 ºC, reflect the pore enlargement that occurred during the CHR by widening of the micropores present in the APC, rather than through the development of new pores.  The data trend in Figure 2.1 and Table E.2 are consistent with the two pore development modes operating at different temperatures especially at lower Mo loading ( 2 wt%). The hydrogenation of amorphous carbon appears to dominate the pore growth mechanism at lower temperatures ( 900 °C). From 700 to 900 °C, it can be seen that the mesopore volume steadily increased with a small loss in surface area and micropore volume for the catalysts with 1 and 2 wt% Mo. The increase in mesopore volume and loss in micropore volume can be explained by the width of pores. Merging of two pores when the wall between them is too thin will result in a loss of surface area. After the amorphous carbon is consumed at lower temperatures, the Mo species react with the highly crystallized carbon. At 1000 °C, surface area and micropore volume increased with loss of mesopore volume as pore lengthening dominates. The loss of mesopores is due to the collapse of pore walls and the loss of the outer surface of the carbon particles at higher temperature. It can be concluded that for the 1 wt% and 2wt% Mo loading catalysts, the dominating pore development mechanism shifted from widening to lengthening due to the increase in reduction temperature. A diagram of pore development of Mo2C/APC with different Mo loadings and CHR temperatures is illustrated in Figure 2.3.  40   Figure 2.3: Diagram of pore development process of AHM/APC precursors with different Mo loadings and CHR temperatures.  The detailed mechanism of Mo2C generation during CHR has been described elsewhere [37] for the case of 10 wt% Mo loading; whereas, in the present study, different Mo loadings and CHR temperatures were investigated to better understand the generation of the mesoporous carbon. The average Mo2C particle size, as measured by TEM, increased with increased loading at the same CHR temperature. The phenomenon is common to supported catalysts - at low loading the Mo is well dispersed; whereas, at high loading the proximity of particles on the carbon surface can result in aggregation of smaller particles. Consequently, some of the Mo2C or MoOxCy particles grow into larger particles. At high temperature (1000 oC), the Mo2C or MoOxCy become much more mobile on the support surface and there is a higher chance that two or more adjacent  41  particles aggregate upon contact. Alternatively, as the Mo2C or MoOxCy species consume the carbon support adjacent to the particle during CHR, the particles may contact each other and hence aggregate. Moreover, the consumption of the carbon is determined by the overall contact area between the Mo species and carbon support. For example, total carbon loss per gram of Mo for the 1Mo2C/APC_R800 sample (~ 40 gC/gMo) with a Mo2C particle size of 9.4 nm (Table 2.2) is about 5x’s that of the 10Mo2C/APC_R800 sample (7.5 gC/gMo) with a particle size of 10.8 nm and the former has ≥ 10 x’s the mesopore volume generated on a per gram of Mo basis. The larger the Mo2C particle size, the lower the overall Mo/carbon contact area, resulting in less efficient pore generation by the Mo. Large particles also create larger pores which have a higher likelihood of collapse with loss of surface area.   The catalysts with low Mo loading ( 2 wt% Mo) and prepared at CHR temperatures  900°C were selected for acid treatment and the esterification reaction since at higher Mo loading the pore generation efficiency was lower and the resulting materials had fragile pore walls; whereas, at lower CHR temperatures the mesopore volume of the catalyst was relatively low (and similar to that obtained without the Mo – see the Acid-T APC_800). As summarized in Tables 2.3 and 2.4, acidity was present in both the Mo2C/APC catalysts and the acid treated catalysts. Although part of the acidity may come from Mo2C or MoOxCy, the low Mo content indicates that the majority of the acid sites result from functionalization of the carbon support. The Raman spectroscopy (Figure E.1) and CHNS analysis (Table E.4) confirmed that the carbon surface of the APC_800 is functionalized by O that is known to generate acid sites, as has been reported for graphene oxide [130, 131]. In addition, during activation by CHR, defect sites form on the carbon surface as CH4 is generated and these sites can combine with oxygen during passivation.  42  Hence, with increased Mo loading, more defect sites will form, resulting in an increase acid sites (Tables 2.3 and 2.4). From the esterification results we conclude that without acid treatment, the esterification ability of the Mo2C/APC is relatively low, but after acid treatment the catalysts have esterification activity comparable to traditional solid acid catalysts operated at similar temperature. For example, the 2nd-order rate constants (k) of the 1Mo2C/APC_R900 and the Acid-T 1Mo2C/APC_R900 measured at 77 °C were 0.510-8 m3.(mol.s)-1 and 2.3 10-8 m3.(mol.s)-1, respectively. These values compare favourably with the ion-exchange catalysts reported by Peters et al. [132] for the same reaction operated at 75 °C. Analyzing their data [132] using the same methodology as for the Mo2C/APC catalysts (see Section 5.4), values of 2.110-8 m3.(mol.s)-1 and 2.710-8 m3.(mol.s)-1 for Amberlyst 15 and Smopex-101 catalysts, were obtained, respectively.   After acid treatment, the textural properties of catalysts were significantly improved (Tables 2.1, 2.3 and 2.4). Comparing the APC_800 and acid treated APC_800 shows that the acid treatment resulted in increased porosity and acidity of the carbon. The acid treatment also removed > 50% of Mo from the pores of the Mo2C/APC samples, further increasing the mesoporosity of the catalyst. More importantly, the acid treatment introduces oxygen functional groups such as -SO3H and -COOH onto the catalyst [38, 39]. The increase in S content of the Acid-T APC_800 compared to the APC_800 sample (Table E.5) indicates the presence of -SO3H. However, with a S content of only 0.2 wt% on the Acid-T APC_800 sample, the corresponding -SO3H content of the catalyst is 62.5 mol/g. Since this is significantly lower than the acidity determined from NH3 TPD (Table 2.1), we conclude that most of acidity arises from other oxygen functional groups on the acid treated carbon surface, such as –COOH [38, 39]. Consequently, most of the  43  acid washed catalysts had higher acidity and higher esterification activity per gram of catalyst than the corresponding Mo2C/APC. In one case, the 2Mo2C/APC_R900 catalyst loses significant surface area after the acid treatment, presumably due to loss of pore walls. As Figure 2.3 shows, the 2nd-order rate constant increased with increased acidity of the acid treated Mo2C/APC catalysts. Assuming that the NH3 uptake is a measure of the number of active sites of the acid treated catalysts, the turnover frequency (TOF) was calculated for each catalyst and is reported in Tables 2.1, 2.3 and 2.4. We note that the acid treated catalysts have similar TOFs, as expected from Figure 2.2 that shows an approximate linear increase in the 2nd-order rate constant with NH3 uptake. Hence, we conclude that the same type of acid sites (-SO3H, -COOH) are mostly responsible for the esterification reaction on all the acid washed catalysts. For the non-acid washed Mo2C/APC catalysts, the NH3 uptake was significantly lower than for the acid treated catalysts. The NH3 must titrate acid sites not associated with the acid washing (-SO3H, -COOH), but rather those associated with O species of the Mo2C and the activated carbon [101]. The NH3 uptake increased with increased Mo2C content and the corresponding TOF of the non-acid washed Mo2C/APC catalysts generally decreased as the Mo2C loading increased. In this case many factors may impact the calculated TOF including a change in the types of sites as the Mo loading increased, increased mesoporosity and lower Mo2C dispersion. Hence the TOF of the non-acid treated samples cannot be compared directly with the acid treated samples, nor can they be compared directly to one another since the type of acid site is changing (Mo oxycarbide versus O functional group on the carbon), the Mo2C dispersion is decreasing and the internal mass transfer to the acid sites is increasing as the Mo loading increased.  The Acid-T 1Mo2C/APC_R900 had the highest acidity among all catalysts and had about 2x’s higher activity  44  than the acid treated APC_800 catalyst (with no Mo) and the Mo2C/APC catalyst that were not acid washed.  2.5 Conclusions High surface area microporous carbons (APC) were prepared from a Canadian oilsands derived petcoke by thermochemical activation at 800 ºC in the presence of KOH. The subsequent impregnation of the APC with ammonium heptamolybdate and activation by carbothermal hydrogen reduction yields Mo2C supported on a mesoporous carbon with high mesopore volume (Vmeso ~0.4 cm3/g). Mesoporous carbons (Vmeso ~0.7 cm3/g) were subsequently obtained by acid washing. Both CHR temperature and Mo loading impact the textural properties of the carbons. Mo2C/APC catalysts have some esterification activity. Acid treatment significantly improves the activity of all the catalysts. The esterification activity of the acid washed Mo2C/APC catalysts was significantly higher than that of the APC without the Mo. The acid washed Mo2C/APC prepared with low Mo loading and CHR temperature up to 900 ºC yields acidic mesoporous carbon with the highest mesoporosity, acidity and catalyst esterification activity.  45  Chapter 3: Carbon Supported Mo2C Catalysts Derived from Different Carbon Sources for the Hydrodeoxygenation of 4-Methylphenol 2 3.1 Introduction In the previous chapter, the feasibility of a carbon waste material such as petcoke being used as carbon support for Mo2C catalysts was proven. The potential to use acidified carbon derived from petcoke as a catalyst for the esterification reaction was also confirmed. The oxygen functional groups present on the carbon support played an important role in increasing the catalyst acidity. In this chapter, the feasibility of using carbon supported Mo2C catalysts for hydrodeoxygenation reactions is investigated. The effect of oxygen functional groups on the catalyst preparation and the catalyst selectivity in hydrodeoxygenation reactions is studied. Furthermore, the same chemical activation method is applied to activate both petcoke and biochar. These materials, used as a support of Mo2C catalysts for the same hydrodeoxygenation reactions, are also compared. Biochar is chosen as an alternative carbon material because of its high O-content and the known importance of O species in the preparation of the catalyst.  In previous work, Mo2C supported on an activated charcoal (Mo2C/AC) was examined for the HDO of 4-methylphenol [126]. Carbothermal hydrogen reduction (CHR) was used to prepare the carbon supported Mo2C catalysts [36, 37, 126, 133]. CHR refers to a temperature programmed reduction in H2 of the catalyst precursor impregnated onto the carbon support, as first described by Mordenti et al. [134] - CHR has been proven to effectively generate Mo2C and                                                  2 A version of this chapter has been published: S. Liu., H. Wang, R. Putra, C. Kim, and K. J. Smith, “Impact of Carbon Properties on Mo2C/Carbon Catalysts for the Hydrodeoxygenation of 4-Methylphenol,” Energy & Fuels (2019) , Article ASAP DOI: 10.1021/acs.energyfuels.9b00531 (2019)  46  simultaneously enlarge the pore size of the carbon support [37]. Key to the formation of Mo2C is the hydrogenation of the carbon support to yield CH4. The formed Mo2C also contributes to the catalytic hydrogenation of the carbon and the generation of mesopores [37]. Consequently, the properties of the carbon used in the synthesis of Mo2C/carbon catalysts are expected to significantly impact catalyst properties. Santillan-Jiminez et al. [135] studied Mo2C supported on activated carbon, multi-walled carbon nanotubes and carbon nanofibers. They reported that the carbon nanofibers (CNF) yield mostly nano-sized Mo2C particles, smaller than the Mo2C particles on the carbon nanotubes, and that the CNF supported catalysts had the highest phenol yield from guaiacol conversion. Mo2C supported on biochar for the HDO reaction has been reported by Zhang et al. [136] who showed that non-polar solvents such as hexane enhance HDO more than polar solvents such as ethanol.  In the present study, activated carbon derived from petcoke (APC) and biochar (ABC) are used as carbon supports of Mo2C catalysts that are assessed for the HDO of the model reactant 4-methylphenol (4-MP). The impact of the carbon on the catalyst properties and activity is compared with previous work that used activated charcoal as the carbon source [37]. Petroleum coke (or petcoke) is a by-product of crude oil refining. The high carbon content and low ash content make it a good candidate for gasification [137]. Similarly, biochar is the solid product from biomass pyrolysis [138], the properties of which are largely dependent on the pyrolysis processing conditions [139]. Biochar is much less dense than petroleum coke and has significant potential as an adsorbent [140-142]. Both petcoke and biochar also have potential to be used as a catalyst support. The many O functional groups that are known to exist on the biochar surface [143] decrease hydrophobicity of the carbon surface, improving the surface wettability by  47  aqueous solutions [144], thereby improving the dispersion of metal catalysts prepared by impregnation of the carbon support [145]. The Mo2C/ABC catalyst is proven to be a more effective catalyst for HDO than the Mo2C/APC catalyst because of a higher DDO selectivity that results in less H2 consumption during HDO.  3.2 Experimental  3.2.1 Materials The raw petroleum coke (PC) used here is described in Chapter 2 - 2.1. The raw biochar (BC) was prepared by the pyrolysis of SPF (spruce, pine, fir) pellets (moisture content 6 wt%; particle size of 0.5 ~ 1.0 mm) at 450-550 oC in a N2 flow of 5000 L(STP)/h. The PC and BC were ground and sieved to particles with diameters between 90-180 μm before thermochemical activation. Elemental analysis of the PC and BC is reported in Table 3.1.  3.2.2 Mo2C/APC and Mo2C/ABC Catalyst Preparation The thermochemical activation method is described elsewhere in Chapter 2.2. The activated bio-char (ABC) was prepared by the same methodology that used to prepare the activated petroleum coke (APC). The final activation temperature was varied in this study from 600 to 900 oC. The obtained ABC sample is designated as ABC_xxx, where xxx represents as the final temperature of the activation. The ABC prepared at 700 °C (ABC_700) and the APC prepared at 800 °C (APC_800) were selected for use as the supports of the Mo2C catalysts since they had similar surface areas and pore volumes.   48  Catalyst precursors with a nominal loading of 10 wt% Mo on APC and ABC were prepared by wetness impregnation (see preparation details in Chapter 2.2). After impregnation the resulting slurry was transferred to a round bottom flask and sonicated for 1 h before removing the solvent using a rotary evaporator at 35 oC operated under vacuum. The resulting catalyst precursors were dried for 12 h prior to initiating carbothermal hydrogen reduction (CHR) by placing the sample in a tubular reactor under a H2 flow of 100 mL(STP)/min. A different temperature program was used here since the CH4 generated (< 1 mol.%) below 500 oC is low. The reactor temperature was quickly increased from room temperature to 500 oC at a ramp rate of 10 oC/min, followed by slowly increasing the temperature at a ramp rate of 1 oC/min to the final temperature (600 ~ 800 oC), which was then held for a further 90 min before rapidly cooling the sample to room temperature. A mass spectrometer was used to monitor the exit gas from the CHR and quantify the yield of CH4 during CHR. Helium (He) was used as the reference gas during this analysis. The relative pressure of CH4 to He (PCH4/PHe) was plotted as a function of reaction time to provide a comparison between the ABC and APC carbon hydrogenation that occurred during CHR to yield the supported Mo2C catalyst. The catalysts are identified as 10%Mo2C/APC_RXYZ and 10%Mo2C/ABC_RXYZ for the catalyst prepared on the activated petcoke and biochar, respectively, with XYZ representing the final temperature of the CHR process. The loss of carbon during the thermochemical activation with KOH and the CHR is defined as the burn-off as shown in Eq. (2-1).   49  3.2.3 Catalyst Characterization 3.2.3.1 N2 Adsorption and Desorption Textural properties of the carbons and catalysts were measured by N2 physisorption at -196 oC using a Micromeritics ASAP 2020 (see details in Chapter 2.2.3). The textural properties, including surface area (SA), micropore volume (Vmicro), and mesopore volume (Vmeso), were calculated using 2D non-local density functional theory (2D-NLDFT) based on a N2-DFT model.   3.2.3.2 X-ray Diffraction (XRD)  X-ray diffraction (XRD) patterns were obtained at 2θ = 10º ~ 90º using a Bruker D8 Focus (0-20, LynxEye detector) diffractometer. A Co Kα (λ=1.789 Å) source operated at 35 kV and 40 mA was used for the analysis. XPS analysis was done using A Leybold Max200 X-ray photoelectron spectrometer with Mg Kα as the photon source generated at 1253.6 eV. The C1s peak at 284.5 eV was taken as reference to account for charging effects.  3.2.3.3 Elemental Analysis  The ultimate analysis and CHNS content of the petcoke and biochar were measured using an elemental analyzer (Perkin-Elmer 2400 series II CHNS/O), following the procedure described in Chapter 2.2.3.  3.2.3.4 CO Chemisorption Pulsed CO uptake measurements were conducted using a Micromeritics AutoChem 2920 with a TCD detector. All samples were prepared in-situ under a 50 mL(STP)/min flow of 9.5 mol% H2/Ar and the same temperature program used for the corresponding catalyst preparation. After  50  cooling to about 100 oC, the gas was switched to pure He for 2 h. Then the temperature was reduced to 60 oC waiting until the baseline stabilized. Subsequently, 0.5 mL pulses of CO were injected into a flow of He (50 mL(STP)/min) repeatedly until no further CO uptake was observed. The CO uptake is normalized per gram of Mo.  3.2.3.5 Raman Spectroscopy  A Horiba Jobin Yvon LabRAM HR800 was used for Raman analysis as described in Chapter 2.2.3.  3.2.3.6 DRIFTS A Nicolet FT-IR 5700 spectrophotometer was used to conduct DRIFTS (Diffuse Reflectance Infrared Fourier Transform Spectroscopy) tests on the both raw and activated carbon materials. The samples were scanned for 64 times with a resolution of 2 cm-1 at a range between 650-4000 cm-1. All samples were fine powders (dp< 90 μm) and diluted with KBr.  3.2.4 Hydrodeoxygenation (HDO) of 4-Methylphenol Mo2C/APC and Mo2C/ABC Catalysts 4-Methylphenol (4-MP) was chosen as model reactant to determine the catalytic activity of the Mo2C/APC and Mo2C/ABC catalysts for the hydrodeoxygenation (HDO) reaction. The reaction was conducted at 350 oC under 4.3 MPa H2 pressure in a batch reactor with a stirring speed of 1000 rpm. Further details are provided in [126], in which activated charcoal supported Mo2C catalysts were assessed at the same reactions conditions [126]. As noted previously, the main products of 4-MP HDO at 350 oC and 4.3 MPa H2 are toluene, methylcyclohexane, 1- 51  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 kinetics of the DDO and HYD reactions were assessed using the pseudo-1st order rate equations for the DDO and HYD parallel reaction paths, [126] as shown in Eq. 3-1 and Eq. 3-2.   𝑟𝐷𝐷𝑂 = (𝑘𝐷𝐷𝑂𝑇 + 𝑘𝐷𝐷𝑂𝐶𝑐𝑎𝑡)𝐶4−𝑀𝑃        (Eq. 3-1) 𝑟HYD = (𝑘𝐻𝑌𝐷𝑇 + 𝑘𝐻𝑌𝐷𝐶𝑐𝑎𝑡)𝐶4−𝑀𝑃        (Eq. 3-2)  where 𝑘𝐷𝐷𝑂𝑇  and 𝑘𝐻𝑌𝐷𝑇  (min-1) represent the 1st-order rate constants of the thermal reactions (without catalyst), Ccat represents the catalyst concentration (gMo/mLfeed), kDDO and kHYD (mL/(gMo.min)) represent the 1st-order rate constants of the catalytic reaction and 𝐶4−𝑀𝑃 is the concentration of 4-MP. The total kinetic constant kc is the sum of kDDO and kHYD. The kinetic rate constants were estimated using a Levenberg-Marquardt nonlinear regression method described in Appendix D, applied to the measured component concentration versus time batch reactor data. Separate experiments done without catalyst were used to determine 𝑘𝐷𝐷𝑂𝑇 and 𝑘𝐻𝑌𝐷𝑇 . The catalyst turnover frequency (TOF) is calculated by Eq. 3-3:  𝑇𝑂𝐹 =𝑘𝑐∗𝐶4−𝑀𝑃 𝐶𝑂 𝑢𝑝𝑡𝑎𝑘𝑒          (Eq. 3-3) The mass transfers effects in batch reactor is excluded by calculation in Appendix F.4 as an example for the batch reactors.   52  3.3 Results Table 3.1 summarizes the bulk composition of the carbons, both before and after activation. Note that the S content of the PC is extremely high compared to the BC. In addition, the N content of the PC is ~ 10 x’s higher than that of the BC; whereas, the O content of the BC is 5 x’s higher than that of PC. These differences reflect the origin of the materials. Petcoke is generated during the delayed coking of bitumen with high S and N contents; whereas, the biochar is the solid product from the fast pyrolysis of a woody biomass that contains high concentrations of oxygen enriched organic matter. The data show that during activation with KOH, most of the S was removed from the PC, even at 600 oC, where the S content decreased from 6.6 wt% (PC) to 1.3 wt%. No S was detectable in the sample treated at  800 oC and the H decreased to below detection limits at temperatures  700 oC. The BC does not contain any S and the H content decreased to zero at 600 oC. N was not completely removed from any of the activated carbons, but the N content decreased to ~0.6 wt% for the APC and to ~0.4 wt% for the ABC. Note that the data of Table 3.1 show that KOH activation of the PC results in an increase in the O content of the activated carbon, as reflected in the O/C atomic ratio before and after activation of the raw PC. However, as the activation temperature increased to 800 oC, the O content decreased (as shown in Table 3.1, column 7), suggesting the removal of this O at high temperature (likely as CO2). For the BC, the KOH activation introduced oxygen functional groups to the biochar, as shown by the increased O/C ratio for the BC versus the ABC_600 reported in Table 3.1. However, the data also show that the O content of the ABC begins to decline at much lower temperature (700 oC) compared to the APC (800 oC). The ability of KOH to add oxygen to PC during thermochemical activation is reported in other studies [90, 146] and a decrease in the O  53  content of the ABC as temperature is increased above 600 °C has also been observed previously [147, 148].   54   Table 3.1: Elemental analysis of the as-received raw carbons and the activated carbon supports.   CHNS Analysis Raman Analysis Sample C (wt%)a H (wt%)a N (wt%)a S (wt%)a O (wt%)a O/C (at.) I(D)/I(G)c Raw Petcoke (PC) 83.27 3.60 2.02 6.58 4.53b 0.04  1.5 APC_600 77.31 0.88 1.15 1.26 19.41 0.19  2.2 APC_700 71.21 0.00 0.58 0.15 28.07 0.30 2.6 APC_800 89.87 0.00 0.67 0.00 9.46 0.08  2.9 APC_900 92.91 0.00 0.61 0.00 6.48 0.05  1.9 Raw Biochar (BC) 76.20 0.75 0.54 0.00 22.52 0.30 1.5 ABC_600 73.02 0.00 0.30 0.00 26.69 0.37 2.1 ABC_700 80.90 0.00 0.39 0.00 18.72 0.23 2.0 ABC_800 83.23 0.00 0.46 0.00 16.32 0.20 2.4 ABC_900 85.58 0.00 0.39 0.00 14.04 0.16 0.4 a. The error associated with each element: C ± 0.03%; H ± 0.13%; N ± 0.18%; S ± 0.28%; O ± 0.06%. b. O was calculated by 100-(C+H+N+S) (wt%). c. Relative intensity of D (~1350 cm-1) band to G band (~1600 cm-1) for carbon materials. Error is < 0.1.    55  The Raman spectra peak intensity ratio I(D)/I(G), where I(D) is the intensity of the D band peak at 1600 cm-1 and I(G) is peak intensity of the G band at 1350 cm-1, is an indicator of the degree of disorder of carbon materials [149, 150]. The ratio reported in Table 3.1 and the Raman spectra of Figures E.6 and E.7 (See Appendix E.2) indicate that the raw PC had high crystallinity (I(D)/I(G) = 1.5), but the ratio increased gradually with increased activation temperature from 600 oC (I(D)/I(G) = 2.2) to 800 oC (I(D)/I(G) = 2.9), indicative of increased disorder. However, at 900 oC, the ratio decreased (I(D)/I(G) = 1.9). The data indicate that at low temperature KOH activation destroys the carbon crystal structure, yielding amorphous carbon, likely on the pore walls that were exposed to the KOH. However, at higher temperature, graphitization of the carbon dominates. The Raman spectra data of the raw BC and ABC show similar results. As presented in Table 3.1, despite the high oxygen content of the biochar, the crystallinity (I(D)/I(G) = 1.5) is similar to that of the raw petcoke. The degree of disorder of the ABC increased at an activation temperature of 600 to 800 °C and then decreased significantly (I(D)/I(G) = 0.4) as the activation temperature was increased to 900 oC.  Table 3.2 summarizes the textural properties of the APC and ABC, activated with KOH at different temperatures. The APC and ABC isotherms (Appendix E.2 - Figure E.8) are of Type I, based on the IUPAC classification [151] and approach saturation at relative pressure 0.2~0.4, indicative of microporous materials. The data also show that the N2 adsorption capacity increased as the activation temperature increased, in agreement with Mochizuki et al. [152]. A hysteresis loop begins to appear at 700 oC for the APCs while for the ABCs the hysteresis appears at 900 oC. The APC has a marginally higher mesopore volume than the ABC when  56  compared at the same activation temperatures. In general, the surface area and pore volume increased as the activation temperature increased for both APC and ABC.   Table 3.2: Textual properties of APC and ABC samples thermochemical activated at different temperatures. Sample Surface area (m2/g)  Pore volume (cm3/g) Burn-off (%) Micropore volume Mesopore volume Total volume APC samples APC_600 1560 0.53 0.01 0.54 27.2 APC_700  1817 0.62 0.02 0.63 31.2 APC_800  2037 0.78 0.07 0.84 37.5 APC_900  2088 0.77 0.22 0.99 48.2 ABC samples ABC_600 1837 0.70 0.00 0.70 28.1 ABC_700 2096 0.84 0.00 0.84 39.9 ABC_800 2224 0.96 0.03 0.99 42.5 ABC_900 1951 0.82 0.17 0.99 68.8  To compare the two different activated carbons as catalyst supports, APC_800 and ABC_700 were chosen for the subsequent CHR synthesis of the Mo2C catalyst, since these two samples had similar surface areas and total pore volumes. The difference in mesoporosity between APC and ABC is considered of less significance since the CHR develops most of the final catalyst mesoporosity. The composition of the selected supports is reported in Table 3.1 and Table E.5, showing that the APC_800 had significantly less O content than the ABC_700 but the APC contains more trace elements than ABC. The XPS data of Table E.6 and Figure E.9 show the surface O content is also higher for the ABC_700 than the APC_800 sample. This result is consistent with the FT-IR spectra of Figure 3.1, which also show the loss in O species following activation.  57     Figure 3.1: DRIFTS spectra of raw petcoke, APC_800 and raw bio-char, ABC_700 supports.  Figure 3.2 reports the CH4 concentration in the product gas during CHR to 700 °C for the 10%Mo2C/APC_R700 and 10%Mo2C/ABC_R700 catalysts [37]. The data show that the ABC support generates more CH4 during CHR than the APC at the same temperature, and the hydrogenation of the carbon is initiated at lower temperature compared to the APC sample. These differences are important because it has been reported previously that the formation of the Mo2C is related to the formation of CH4 during the CHR process [37].   58   Figure 3.2: Profile of detected relative intensity of PCH4/PHe as a function as time for 10%Mo2C/APC_R700 (represented by —) and 10%Mo2C/ABC_R700 samples (represented by —).  Figure 3.3 summarizes the XRD analysis of the 10%Mo2C/APC and 10%Mo2C/ABC samples, reduced at different temperatures. In the case of APC, distinct peaks at ~15o and ~28o representing graphitic carbon are identified at a CHR temperature of 600 oC. The peak at 46.07o assigned to Mo2C (101) can be observed at 650 oC for the APC supported catalysts, and at 700 oC, the peak intensity increased. XRD patterns for the ABC supported catalysts at all three reduction temperatures, on the other hand, did not show any peaks that could be assigned to Mo2C.  59    Figure 3.3: XRD patterns of 10%Mo2C/ABC and 10%Mo2C/APC reduced at different CHR temperatures. (♦ identifies C-graphite; * identifies Mo2C)    60  Figure 3.4 presents the deconvoluted Mo 3d XPS spectra of the APC supported catalysts, using the same deconvolution method applied in previous work [153]. The deconvoluted peaks at binding energy (B.E) 228.4 eV, 229.4 eV and 232.4 eV are assigned to Mo2+, Mo4+ and Mo6+, respectively. Table 3.3 summarizes the detailed distribution of Mo species identified by XPS. From 600 to 650 oC, the Mo valence decreases from Mo6+ and Mo4+ to Mo2+. The effect is much more significant at a CHR temperature of 700 oC. Mo2+, which can be assigned to Mo2C, reaches 49.6%. The Mo 3d deconvoluted XPS spectra of the ABC supported catalysts are shown in Figure 3.5 with the same deconvolution applied. In this case the Mo2+ surface content is higher than that on APC, at all reduction temperatures, suggesting that the ABC supported catalysts form more Mo2C during the CHR process, compared to the APC supported catalysts.   61   Figure 3.4: XPS spectra of Mo 3d of fresh 10%Mo2C/APC samples reduced at different CHR temperatures: (a) Survey scan of 10%Mo2C/APC_R600; narrow scans of: (b) 10%Mo2C/APC_R600, (c) 10% Mo2C/APC_R650, and (d) 10%Mo2C/APC_R700.   62   Table 3.3: XPS analysis of Mo 3d species for 10%Mo/APC and 10%Mo/ABC fresh catalysts at different CHR temperatures. CHR Temperature R600  R650  R700 Mo species 10%Mo2C/APC 10%Mo2C/ABC  10%Mo2C/APC 10%Mo2C/ABC  10%Mo2C/APC 10%Mo2C/ABC % %  % %  % % Mo2+ 15.9 46.3  30.0 49.5  49.6 63.9 Mo4+ 25.7 27.1  21.0 27.1  11.7 7.9 Mo6+ 58.4 32.2  49.0 23.4  38.7 18.2    63   Figure 3.5: XPS narrow scan spectra of Mo 3d of fresh 10%Mo2C/ABC samples reduced at different CHR temperatures. (a) 10%Mo2C/ABC_R600; (b) 10% Mo2C/ABC_R650; (c) 10%Mo2C/ABC_R700.  Table 3.4 summarizes the kinetic data for the HDO of 4-MP obtained over the APC and ABC supported Mo2C catalysts, prepared at different reduction temperatures. The relevant data for AC supported Mo2C catalysts, reported previously [13], are also listed for comparison. Both the 10%Mo2C/ABC and the 10%Mo2C/APC catalysts show very high DDO selectivity compared to the 10%Mo2C/AC catalysts. Also, both the 10%Mo2C/ABC and the 10%Mo2C/APC catalysts  64  have much higher 1st-order reaction rate constants than the 10%Mo2C/AC. For the Mo2C/APC catalysts, the rate constants (mL/(min. g Mo)) are not significantly different from each other as the CHR temperature increased from 600 to 700 oC. Among the Mo2C/ABC catalysts, the 10%Mo2C/ABC_R650 had the highest activity per gram of Mo and kc was approximately twice that measured on the Mo2C/ABC catalysts prepared at CHR temperatures of 600 and 700 oC. Finally, accounting for the CO uptake on all the catalysts, the turnover frequency (TOF) of the catalysts is reported in Table 3.4. These data suggest that the activity of the both Mo2C/ABC and Mo2C/APC catalysts for the HDO of 4-MP are  approximately the same as the TOFs for the Mo2C/AC catalysts reported previously [126].    65  Table 3.4: Kinetic rate constants, selectivity, and conversion of 4-methylphenol (4-MP) on Mo2C/APC and Mo2C/ABC catalysts with different CHR temperatures at 350 oC and 4.3 MPa in hydrodeoxygenation (HDO). Catalyst Mo load in the reactor  X4-MPa  SDDO/HYDb kDDO kHYD kc CO uptaked  TOF  mgMo/100 mLfeed %  mL/(min.gMo) μmol/gMo s-1 x 10-2 10%Mo2C/APC samples 10%Mo2C/APC_R600 59 66.4 4.9 4.66±1.08e 0.95±0.84 5.62±1.92 661 3.6±1.1 10%Mo2C/APC_R650 58 77.2 6.9 5.40±0.76 0.79±0.58 6.19±1.32 797 3.3±0.5 10%Mo2C/APC_R700 46 70.2 11.0 6.69±1.06 0.61±0.80 7.30±1.86 693 4.5±0.9 10%Mo2C/ABC samples 10%Mo2C/ABC-R600 51 66.7 8.0 4.55±0.60 0.57±0.50 5.12±1.1 593 3.7±0.6 10%Mo2C/ABC-R650 49 84.0 9.7 9.30±1.38 0.96±0.94 10.26±2.32 844 5.2±0.9 10%Mo2C/ABC-R700 54 78.6 9.4 5.92±1.18 0.63±0.88 6.56±2.04 599 4.6±1.3 10%Mo2C/AC samples [126] 10%Mo/AC-600 72 43.2 2.9 1.260.12 0.430.11 1.730.23 147 5.0±1.1 10%Mo/AC-650 72 44.0 3.4 1.330.08 0.390.07 1.720.15 187 3.9±0.5 10%Mo/AC-700 72 36.3 4.0 1.000.12 0.250.11 1.250.23 156 3.4±1.0 a. The conversion of 4-methylphenol after 5 h’s HDO of 4-MP. b. The selectivity of DDO to HYD. c. kc=kDDO+kHYD; unit: (mL/(min.gMo)). d. The associated error for CO uptake is ±5%. e. The associated error for all kinetic constants is reported with the 95% confidence interval (CI; CI= 2 × Std. dev). 66  3.4 Discussion The raw petcoke generated by delayed coking where heavy feed (bitumen) is processed at high temperature [154] is of higher bulk density (Table E.7) than the raw biochar generated in a rapid devolatilization process with high gas yields that generate porosity in the solid biochar [155]. The properties of the activated carbon derived from petcoke and biochar as well as the catalysts synthesized by the CHR process, are affected by the properties of the carbon. The O content of the raw biochar is much higher than the raw petcoke and the CHNS analysis of the activated carbons indicate that the O content of the ABC is higher than the APC. The S present in the raw petcoke is removed during the activation with KOH and consequently, the effect of S on Mo2C synthesis is negligible for both carbon supports investigated herein. The N content, on the other hand, is considerable for both APC and ABC. Although the N may be reduced to NH3 during CHR in H2 at elevated temperature [156] and the NH3 may in turn react with the Mo, the low concentration of N compared to C (N:C ratio is < 0.008) and O (N:O ratio is < 0.08) make such a reaction of much less significance than the direct carburization. The metals present in the raw petcoke and biochar may also impact the activity of the prepared catalyst. However, the metal concentrations are 103 times lower than the Mo, suggesting that the contribution of these metals to the catalyst activity would be minimal compared to the Mo. Also note that the characteristic feature of the raw biochar - the very high oxygen content - remained high after the chemical activation with KOH.   The Raman spectroscopy of the APC and ABC show that KOH chemical activation increased the amorphous carbon content of the support, with the degree of disorder higher for the APC than the ABC. The reaction between KOH and C begins at temperature > 400 oC [157, 158]. As the  67  KOH etches through the carbon, pores develop, as indicated by the isotherm data. The H2O and CO2 generated during the activation with KOH may also  react with the carbon surface leading to the formation of oxygen functional groups [159, 160]. At low temperature ( 700 °C for APC and  600 °C for ABC) the O content increased with increased activation temperature. At higher activation temperature (≥ 800 oC for APC and ≥ 700 oC for the ABC), as more carbon is reacted, the oxygen functional groups are also reacted, leading to a decrease in O content; whereas, graphitization also increases at these high temperatures. Note, however, that because the initial O content of the biochar is very high, after thermochemical activation the O content of the biochar remains relatively high.   The oxygen functional groups make the carbon surface more amorphous and less crystalline [161] and amorphous carbon is more readily hydrogenated in the CHR process [37]. The abundant oxygen in the ABC may also be reduced by H2 during CHR in the presence of Mo, forming surface vacancies that have the ability to activate H2 and further facilitate the CH4 formation reaction [162]. The higher O content of the ABC compared to the APC results in more CH4 generation on the Mo-impregnated ABC_700 than the APC_800. XRD results of the supported catalysts also suggest a higher dispersion of Mo species on the ABC than on APC.  The formation mechanism of Mo2C on APC during CHR has been reported previously and the formation mechanism of Mo2C on ABC is also expected to be strongly dependent on the concentration of CH4 generated during CHR. With higher Mo dispersion and higher CH4 concentration observed from the CHR of the ABC, the generation of Mo2C can be expected at lower temperature. The XPS results confirm this by showing that the ABC supported catalysts  68  contain more Mo2C species on the catalyst surface than the APC supported catalysts prepared at the same reduction temperature.   The data of Table 3.4 show that the Mo2C catalysts supported on the three different carbon supports have similar TOFs for the HDO of 4-MP. Although the characterization data show that the textural properties and Mo2C dispersion of the Mo2C/carbon catalysts are strongly dependent on the carbon support, similar TOFs are obtained on all three carbon supports. Thus, the activity of the catalysts, determined by the number of Mo2C active sites, is dependent upon the Mo2C loading and dispersion. The deconvoluted Mo 3d XPS analysis of the 10%Mo2C/APC catalysts show similar proportions of Mo2C and MoOxCy to that reported for the 10%Mo2C/AC at different reduction temperatures [126]. Yet the 1st-order rate constants reported per gram of Mo (Table 3.4) were significantly different on each of the supports. The 10%Mo2C/APC catalysts have much higher activity per gram of Mo than the 10%Mo2C/AC catalysts because the APC support provides a higher surface area and better Mo2C dispersion (higher CO uptake, Table 3.4). The higher dispersion is a result of oxygen functional groups, such as –OH, on the carbon surface that ensure wetting of the carbon by the solvent during impregnation and anchoring of the metal during heat treatment that reduces sintering [163]. Thus, the properties of the APC support lead to a better dispersion of Mo2C compared to AC. The 10%Mo2C/ABC catalysts, with a higher proportion of Mo2+ (i.e. Mo2C, Table 3.3) and better Mo2C dispersion due to a higher concentration of oxygen surface groups (Table E.8) also has a higher activity (per gram Mo) than the APC supported Mo2C. However, given the fact that very similar TOFs are obtained on all three carbon supports, we conclude that the carbon does not impact the active sites of the Mo2C  69  catalyst, but rather simply determines the number of active sites, based on the activity of the carbon during CHR and the resistance to sintering of the formed Mo2C nanoparticles.  We also note that for the 10%Mo2C/APC catalyst, an increase in reduction temperature from 600 oC to 650 oC increased the number of active sites and the porosity of the resulting catalyst, which resulted in an increase in HDO conversion of 4-MP. However, a further increase in reduction temperature to 700 °C resulted in a drop in conversion (Table 3.4) despite a significant increase in Mo2+ on the catalyst surface (Table 3.3). In this case the loss in conversion is due to significant loss in surface area and pore volume at 700 oC (Table 3.5), which together with the increased burn-off, yields less dispersed Mo2C, as reflected in the CO uptake data of Table 3.4. Similar effects of increased CHR temperature are also apparent for the ABC supported Mo2C. In this case, although the low density and high oxygen content of the biochar results in the generation of more CH4 during CHR, the loss of pore walls means less mesopore growth (Figure E.10). The narrow pore size could make some of the active sites very difficult to access. The kinetic rate constants of the best catalysts in this study (per gram of Mo) are ~5x’s higher (Table 3.6) than the previously reported Mo2C catalyst supported on AC [126].    70  Table 3.5: Physical properties of Mo2C catalyst supported on APC and ABC with different reduction temperatures. Sample Surface area a Pore volume b Burn off c Vtotal Vmeso m2/g cm3/g % APC prepared catalysts 10%Mo – AHM/APC precursor 1281 0.50 0.04 — 10%Mo2C/APC _R600 1372 0.66 0.07 18.3 10%Mo2C/APC_R650 1302 0.72 0.15 35.4 10%Mo2C/APC_R700 1098 0.62  0.21  53.7  ABC prepared catalysts 10%Mo – AHM/ABC precursor 1565 0.66 < 0.01 — 10%Mo2C/ABC _R600 1631 0.73 < 0.01 20.1 10%Mo2C/ABC_R650 1630 0.79 0.04 38.1 10%Mo2C/ABC_R700 948 0.41 < 0.01 55.5 a. The specific surface area was calculated from the measured N2 adsorption isotherm using carbon-N2, 2D-NLDFT-Heterogeneous surface applied in the P/Po range of 0.01~0.30. b. The pore volume data were calculated by 2D-NLDFT method, where the mesopore is defined as 2~50 nm and micropore is ≤ 2 nm.  c. Burn-off%=100 ×𝑚𝐼𝑛𝑖𝑡𝑖𝑎𝑙−𝑚𝑓𝑖𝑛𝑎𝑙𝑚𝐼𝑛𝑖𝑡𝑖𝑎𝑙.   71  Table 3.6: Kinetic rate constants for the conversion of 4-MP over Mo2C with different carbon supports at 350 oC. Catalysts k (mL/(min.gMo)) AC support 10%Mo2C/AC_600 [126] 1.69±0.23 10%Mo2C/AC_650 [126] 1.72±0.15 10%Mo2C/AC_700 [126] 1.24±0.23 APC support 10%Mo2C/APC_600 5.62±0.96 10%Mo2C/APC_650 6.19±0.66 10%Mo2C/APC_700 7.30±0.93 ABC support 10%Mo2C/ABC_600 5.12±0.55 10%Mo2C/ABC_650 12.68±1.03 10%Mo2C/ABC_700 6.56±1.02  3.5 Conclusions Thermochemically activated petcoke and biochar were investigated as supports of Mo2C catalysts, prepared by CHR. The properties of the ABC (high surface O content, higher reactivity in H2) result in higher Mo2C formation and higher Mo2C dispersion on the ABC support compared to the APC support. Nevertheless, when accounting for the Mo2C dispersion, the TOF of the HDO of 4-MP is shown to be very similar on Mo2C supported on activated petcoke, biochar and charcoal. However, compared to the 10%Mo2C/AC catalyst, both 10%Mo2C/APC and 10%Mo2C/ABC catalysts have much higher DDO selectivity. The ABC supported catalysts also have higher DDO selectivity than APC supported catalysts, due in part to the high O content of the ABC surface. The DDO selectivity of ABC supported catalyst reached ~90%.   72  Chapter 4: Hydrodeoxygenation of 2-Methoxyphenol over Ru, Pd, and Mo2C Catalysts Supported on Carbon 3 4.1 Introduction The previous two chapters presented the synthesis process for carbon supported Mo2C catalysts and their application in esterification and hydrodeoxygenation (HDO) reactions. The pore development and changes in surface properties that follow chemical activation and CHR have been discussed. Although the application of Mo2C catalysts in HDO reactions was assessed in the previous chapter, the study focused on the differences between activated carbon derived from petcoke (APC) versus biochar (ABC). Oxygen functional groups provided higher dispersion of Mo species and more amorphous carbon for ABC prior to the CHR process. The O functional groups were also responsible for the higher DDO selectivity identified for the Mo2C/ABC catalyst. In this chapter, the focus is on the active sites of Mo2C and the activity of Mo2C relative to more traditional HDO catalysts such as the noble metals Ru and Pd.  In the present study, Ru, Pd and Mo2C supported on activated charcoal (AC) (Ru/AC, Pd/AC and Mo2C/AC) are compared using 2-methoxyphenol as a model HDO reactant to clarify the differences between the noble metal and Mo2C catalysts, since the latter are often claimed to behave similarly to noble metal catalysts. Activated charcoal was chosen as the carbon support for Mo2C to eliminate support effects when comparing to Ru/AC and Pd/AC, which were commercial formulations purchased and used without further modification. The HDO experiments were conducted in a liquid phase, stirred, batch reactor over a relatively wide range                                                  3 A version of this chapter has been published: S. Liu., H. Wang, C. Kim, and K. J. Smith, “Hydrodeoxygenation of 2-methoxyphenol over Ru, Pd and Mo2C Catalysts Supported on Carbon,” Energy & Fuels (2017) 31 (6), 6378-6388.  73  of temperatures 240, 300 and 330 C for the Pd/AC and Ru/AC catalysts and 330, 350 and 375 C for the Mo2C/AC. The catalysts are assessed based on their selectivities and activities, with the kinetic parameters determined using a lumped kinetic model approach [164, 165] in which the products are grouped according to their oxygen content and reported on a per active site basis. Even though the Mo2C/AC catalyst is not as active as the noble metal catalysts for HDO, it consumes much less H2 in the removal of O because of lower hydrogenation activity. The Mo2C/AC is proven to be a promising catalyst for the HDO reaction, compared to conventional noble metal catalysts.  4.2 Experimental  4.2.1 Materials and Catalysts Palladium supported on activated carbon (Pd/AC) and ruthenium supported on activated carbon (Ru/AC) with a metal loading of 5.0 wt.% were used as supplied by Sigma-Aldrich. The carbon supported molybdenum carbide (Mo2C/AC), was prepared by the carbothermal hydrogen reduction (CHR), as reported in Chapters 2 and 3. [126]. Commercial activated charcoal (Darco, 100 mesh, 1025 m2/g, 0.85 cm3/g) was used both as support and carbon source during CHR. The Mo precursor, ammonium molybdate tetrahydrate (AHM; (NH4)6Mo7O24∙4H2O) was loaded onto the AC support by wetness impregnation and dried prior to CHR in a 100 cm3(STP)/min flow of H2 while heating to the final reduction temperature of 650 oC, to yield a 10 wt% Mo2C/AC catalyst.    74  Decalin (Aldrich, 99%) was chosen as solvent for the reaction and 2-methoxyphenol (Aldrich 99%, recorded as GUA) was used as reactant. Ethanol (Aldrich, 99.8%) was used as received for GC-MS sample preparation.  4.2.2 Catalyst Characterization 4.2.2.1 N2 Adsorption and Desorption The specific surface area (SBET), pore volume and pore size distribution of the three catalysts were determined from N2 adsorption-desorption isotherms measured at -196 oC using a Micromeritics ASAP 2020 analyzer. 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.   4.2.2.2 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 Mo2C/AC catalyst was done in-situ by passing 50 mL(STP)/min of 9.5 mol% H2/Ar while heating to 650C as before. Subsequently, the catalyst was 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 as detector. CO pulses were repeatedly injected until no further CO uptake was observed after consecutive injections. For the Pd/AC and Ru/AC catalysts, the CO uptake was conducted after a pretreatment in He at 110 oC. After cooling to room temperature, CO pulses were injected and the CO uptake was measured using a TCD as  75  before. The metal dispersion was calculated by assuming a stoichiometric ratio of 1:1 between CO and the metal atoms.  4.2.3 Hydrodeoxygenation (HDO) of 2-Methoxyphenol (GUA) over Ru/AC, Pd/AC and Mo2C/AC catalysts  The HDO of 2-methoxyphenol (GUA) was carried out in a 300 mL stirred-batch reactor (Parr Instruments Company), at a mixer speed of 1000 rpm and operated at 3.4 MPa H2 initial pressure and reaction temperatures of 240, 300 and 330 oC for Pd/AC and Ru/AC catalysts, and at 330, 350, and 375 oC for the Mo2C/AC catalysts. The model reactant GUA, was dissolved in 100 mL decalin to yield a 5 wt% reactant solution with 1.28 wt% O. For the Ru/AC or Pd/AC catalysts, 0.33 g catalyst was added to the reaction solution and for Mo2C/AC catalyst, 0.66 g was used. Since the Pd and Ru catalysts are relatively stable in the presence of O2, no further treatment was done prior to the reaction, whereas, due to the oxophilic nature of Mo2C, the catalyst was loaded into the reactor under a N2 blanket using a glove bag to avoid air exposure.  Following heat-up of the reactor, a ca. 1 mL liquid sample was withdrawn from the reactor and this sample represented the initial concentration of components in the reactor at t=0 min. Subsequent samples were withdrawn at 30, 60, 90, 180 and 300 min for the Ru/AC and Pd/AC catalysts, whereas for the Mo2C/AC catalyst, samples were collected at 30, 60, 90, 135, 180, 240 and 300 min. After reaction, all the collected samples were diluted 10 times and then mixed with 50 wt% ethanol before GC-MS analysis. The identification of the products was achieved using a PerkinElmer Clarus 500 GC (Gas Chromatograph) and Clarus 560 S MS (Mass Spectrometer) with a Mandel RSK-5MS (30 m × 0.25 mm × 0.25 µm) capillary column. Quantitative analysis  76  was done by external calibration. Components with concentrations  1 mol.% are reported quantitatively herein, whereas trace components ( 1 mol.%) were identified but not quantified. After completion of the test, the gas phase from the reactor was collected and analyzed by gas chromatograph. For all activity data reported herein, the carbon balance across the reactor was ≥ 95% and several experiments were repeated to estimate the experimental error, determined to be ± 5%. The conversion of decalin was not significant except at 330 C over the Pd and Ru catalysts, where some tetralin was observed in the product.   4.3 Results 4.3.1 Catalyst Characterization Table 4.1 reports the textural properties of the catalysts of this study. The catalysts have similar high surface areas and relatively high mesopore volumes. For the Mo2C/AC and Ru/AC catalysts, mesopores account for ~80% of the total pore volume whereas for the Pd/AC, ~67% of the total pore volume is present as mesopores. The CO uptakes are significantly different on each of the catalysts, with the Ru/AC having the highest dispersion and the Mo2C/AC the lowest.   Table 4.1: Physical and chemical properties of fresh 5%Ru/AC, 5%Pd/AC and 10%Mo2C/AC catalysts.  SBET (m2/g) Vtotal (cm3/g) Vmicro (cm3/g) Vmeso (cm3/g) Pore size (nm) CO uptakea (μmol/g) Metal dispersion, % Mo2C/AC 845  0.76  0.14  0.62  3.60  24  2b Ru/AC  769  0.89  0.19  0.70  4.54  127  26  Pd/AC  1044  0.82  0.27  0.55  3.14  49  10  a. This CO uptake is measured on the fresh catalysts. b. This number is here for reference only since Mo2C does not respond well to CO adsorption the same way as noble metals.  77   4.3.2 HDO of 2-Methoxyphenol (GUA) on Pd/AC Table 4.2 lists the main products identified for each of the catalysts measured at 330 C and 3.4 MPa H2 after reaction times that correspond to GUA conversions of approximately 80 % and 100 %. The products are conveniently grouped according to those that result from hydrogenation reactions without O removal (HYD), those in which one O atom has been removed (1-O removed), mostly by cleavage of the R-OCH3 group, and those where both O atoms have been removed from the product molecule (2-O removed). The detailed product distributions measured over a range of temperatures and reaction times for the Pd/AC catalyst are reported in Appendix E – Tables E.10, E.11, and E.12.    Table 4.2: Product distribution from HDO of GUA over Ru/AC, Pd/AC and Mo2C/AC catalysts at 330 oC and 3.4 MPa initial H2 pressure, compared at the same GUA conversion (80% and 100%).  Grouped products Products 80% conversion (%) 100% conversion (%) Pd/AC Ru/AC Mo2C/AC Pd/AC Ru/AC Mo2C/AC HYD   3 2 0 1 0 0  50 20 0 16 1 0  36 18 0 18 2 0 1-O removed  0 0 7 0 0 7  0 0 7 0 0 7  0 0 77 0 0 73  78   0 0 5 0 0 7  5 36 0 5 11 0  4 3 0 7 0 0  0 0 0 10 0 0 2-O removed  4 11 0 43 57 0 /  0/0 3/3 3/0 0/0 23/1 4/0 / /  0/0/0 2/0/0/ 0/0/1 0/0/0 4/1/0 0/0/2   79   At 240 C the HYD product selectivity (2-methoxycyclohexanone, 2-methoxycyclohexanol and 1-methylcyclohexane-trans-1, 2-diol) was > 94% after 1 h of reaction, with cyclohexanol and methoxycyclohexane (1-O removed) making up the balance (Table E.10). The hydrogenated products 2-methoxycyclohexanone and 2-methoxycyclohexanol are rapidly produced from GUA over the Pd/AC catalyst, as has been reported in other studies [67, 166]; whereas, the 1-methylcyclohexane-trans-1, 2-diol has been observed on Pd catalysts operated at 100 C by Gutierrez et al. [74] and Yakovlev et al. [65] observed similar hydrogenated products with Ni-based catalysts. The Pd/AC has a high hydrogenation activity at this temperature similar to Pt and Fe catalysts [73, 167]. Although 2-methoxycyclohexanone converts to 2-methoxycyclohexanol at 240 C [67, 168], all other products are relatively stable at this temperature since their concentrations remain relatively constant with increased reaction time (Table E.10).   At higher temperature ( 300 C, Tables E.11 and E.12), the HYD products begin to react. The concentration of 2-methoxycyclohexanol and 1-methylcyclohexane-trans-1, 2-diol both decline with increased reaction time. Both may be converted to cyclohexanol through a series of reaction steps [70] that result in cleavage of R-OCH3 and R-OH bonds [67, 168]. They may also convert to cyclopentylmethanol through intermediates 1, 2-cyclohexanediol and 1-cyclopenten-1-ylmethanol, present in trace amounts, similar to the thermal reaction sequence proposed by Lanver et al. [169] Since 2-methoxycyclohexanol is more reactive than 1-methylcyclohexane-cis-1, 2-diol it can directly lose an –OH group [75] to form methoxycyclohexane. Note that the 2-methoxycyclohexanone does not fully convert after 180 min at 300 C, whereas at 240 C the  80  concentration of this component is reduced to zero after 90 min. This is due to competitive HYD and O-removal reactions.   At temperatures  300 C, the products with one O atom also react to yield completely deoxygenated products (2-O removed), cyclohexane (mostly), cyclopentane, and methylcyclopentane (Table E.11). Cyclohexanol is converted to cyclohexene which rapidly hydrogenates to cyclohexane [67, 75] and methoxycyclohexane is converted to cyclohexanol [170] or the R-OCH3 bond breaks directly forming cyclohexane [171]. Cyclopentylmethanol and cyclohexanol can form cyclopentane at high temperatures and reaction times (Tables E.11 & 12) and these reactions also occur in non-catalytic thermal reactions [172]. Note that cyclohexanol is more reactive than methoxycyclohexane or cyclopentylmethanol at high temperature, since the concentration of cyclohexanol increases and then decreases with reaction time (Tables E.10 & 11), whereas the other two components with one O atom removed, continue to increase with reaction time. At 330 C, the intermediate products cyclohexanol and methoxycyclohexane are completely converted within 180 min, whereas cyclopentylmethanol is still present in the product.  Figure 4.1 summarizes the lumped product distributions from the HDO of GUA on Pd/AC as a function of reaction temperature and time. At 240 C, most of the GUA reacts during reactor heat up, leaving only 32.8 wt% GUA after heat-up (Table E.10) and 100% conversion of GUA after 90 min at 240 oC. With the decrease in initial reactant, the hydrogenated products increase rapidly but they are very stable at this temperature and no further reactions occur. The HYD products begin to react at 300 oC, yielding products with one O atom. Finally, at 330 C, reactant  81  and hydrogenated products are totally converted after 180 min. The products with one O atom have a higher reactivity at 330 oC than at 300 C, leaving ≤ 10% after 300 min reaction. The selectivity to the completely deoxygenated products is ≥ 90% at 330 C (Table E.12), with 81% selectivity to cyclohexane.   82    Figure 4.1: Mole percentage (mol.%) of reactant (2-methoxyphenol) and products as a function of reaction time on Pd/AC at different reaction temperatures ((a) 240 oC; (b) 300 oC; (c) 330 oC). (HYD: 2-methoxycyclohexanone; 2-methoxycyclohexanol; 1-methylcyclohexane-trans-1,2-diol. 1-O removed: cyclohexanol; 1-methoxycyclohexane; cyclopentylmethanol (not detectable at 240 oC). 2-O removed: cyclohexane; cyclopentane; methylcyclopentane.)  83  Based on the grouping of the reaction products and the above discussion, Figure 4.2 shows a proposed reaction network for GUA conversion on Pd/AC. Products within the dotted square were not detected by GC-MS but are assumed intermediates of the reaction sequence, based on the literature [70]. For example, although GUA deoxygenation is thought to occur through the intermediate catechol on many noble metals, under the conditions of the present study, catechol was only present as a trace component and could not be quantified. Kinetic parameters for the lumped reaction network, k1, k2, and k3 represent the reaction rate constants for the HYD steps, removal of one O atom, mostly by cleavage of the R-OCH3 bond and finally removal of the second O atom, mainly by cleavage of the R-OH bond, respectively. Table 4.3 summarizes the kinetic constants k1, k2 and k3 estimated using a Levenberg-Marquardt nonlinear regression methodology combined with a Runge-Kutta numerical integration to solve the relevant ODEs, derived from the reaction network shown in Figure 4.2, that describe the product concentrations as a function of reaction time (See Appendix D.1 – Lump kinetic models and Tables E.10-12). The fit of this simplified, lumped kinetic model to the measured data is reported in Figure 4.7 (with R2 > 0.98 in all cases). The analysis shows that the reaction kinetics over the Pd/AC catalyst are dominated by the ring saturation reaction, with kinetic constant k1 increasing from 12.2 x 10-2 mL/molCO.min at 240 C to 46.8 x 10-2 mL/molCO.min at 330 C. The calculated apparent activation energy (Ea) for the ring saturation reaction is 38±2 kJ/mol. The O removal reactions occur at significantly lower rate compared to the HYD reaction on Pd/AC, with values of k2 and k3 only determined at reaction temperatures above 300 oC (Table 4.3).    84   Figure 4.2: Proposed reaction network for HDO of 2-methoxyphenol on Pd/AC.  Table 4.3: Estimated 1st-order rate constants and activation energy (Ea) for lumped model of Figure 4.2 for the hydrodeoxygenation of 2-methoxyphenol over Pd/AC catalyst at different reaction temperatures of 240, 300 and 330 oC. ki (x 10-2mL/molCO.min) Reaction temperature (oC) ki0 (x 10-2mL/molCO.min) Ea (kJ/mol) 240 300 330 k1 12.2±0.9 32.0±2.3 46.8±7.7 31.2±1.0 38±2 k2 <0.01 1.0±0.1 4.6±0.4 - 146±18a k3 <0.01 2.5±0.2 8.8±0.0 - 121±8b a.b These values are calculated based on only two k values at 300 and 330 oC. Note: 𝑘𝑖 = 𝑘𝑖0𝑒−𝐸𝑎𝑅𝑇 85  4.3.3 HDO of 2-Methoxyphenol (GUA) on Ru/AC On the Ru/AC catalyst, the HYD products are more reactive than on the Pd/AC, as shown in Table 4.2. After the heat-up period, the GUA conversion was 85% on the Ru/AC and the concentration of the HYD products was relatively low (20 mol.% 2-methoxycyclohexanol, 18 mol.% 1-methylcyclohexane-trans-1, 2-diol and 2 mol.% 2-methoxycyclohexanone – Table 4.2), because of subsequent reactions of these primary products. The main product with one O atom removed is cyclohexanol (36 mol.%) along with 3 mol.% methoxycyclohexane, while the completely deoxygenated products include cyclohexane (11 mol.%), benzene (3 mol.%), cyclohexene (3 mol.%) and methylcyclopentane (2 mol.%). At 100% GUA conversion, the HYD products are almost completely converted, with 1 mol.% 2-methoxycyclohexanol, 2 mol.% 1-methylcyclohexane-trans-1, 2-diol and 11 mol.% cyclohexanol remaining after 1 h reaction time. Completely deoxygenated products are the major products with 57 mol.% cyclohexane, 23 mol.% benzene, 1 mol.% cyclohexene, 4 mol.% methylcyclopentane, and 1 mol.% cyclopentane. Due to dehydrogenation reactions, benzene selectivity is high.   More detailed analysis of the product distribution from the Ru/AC catalyst at different temperatures (Tables E.13, E.14, and E.15) shows very low concentrations of 2-methoxycyclohexanone initially that is rapidly decreased, indicative of the higher activity of Ru/AC versus Pd/AC for the hydrogenation of the ketonic group. The data also clearly show that the deoxygenation ability of Ru/AC is much higher than Pd/AC, even though conversion of the reactant follows a similar pathway on both catalysts. At 240 C, unlike the Pd/AC, a significant portion of the product included compounds with one O atom removed, especially cyclohexanol (up to 60 %) with small amounts of methoxycyclohexane ( 2%), while the completely  86  deoxygenated product cyclohexane was also detected at this temperature. After 5 h reaction, the HYD products had reacted further, yielding mostly products with one O atom, and these products are very stable at this temperature (details see Table E.13). Cyclopentylmethanol, a product generated from 2-methoxycyclohexanol and 1-methylcyclohexane-trans-1, 2-diol on Pd/AC, is absent in the case of the Ru/AC. At 300 C, aside from the formation of the completely deoxygenated product cyclohexane, notable amounts of benzene also form. Benzene may be produced by the DDO of phenol or anisole, as observed by Ishikawa et al. [173], although only trace amounts (less than 1 wt%) of these compounds were detected over the Ru/AC catalyst. Alternatively, the benzene may be a result of the dehydrogenation of cyclohexane, which increases with temperature. At high temperatures ( 300 C) the HYD products react very fast and are consumed after 180 min. Also, products with one O atom react further such that after 5 h, 80% of the products are completely deoxygenated (2-O removed) and cyclopentane and cyclobutane ( 1%) appear. Furthermore, alkylation (Al) and hydrocracking (HC) reactions of the deoxygenated compounds yield alkylated benzene products reported in Table E.14 & E.15, accounting for 10% of the products at 300 oC, increasing at 330 C. Methanol was detected as one of the gas products, providing further evidence of direct Ar-OCH3 bond cleavage. Ishikawa et al. [173] reported that Ru/AC readily converts methanol to methane which can then act as an alkylating agent.  Figure 4.3 shows the lumped product distribution from the HDO of GUA on Ru/AC as a function of reaction temperature and time. At low temperature (240 C) the GUA was completely converted after 60 min and a significant portion of the product included compounds with one O atom removed. After 5 h reaction, most of the products had one O atom. At 300 oC,  87  the HYD products are consumed after 180 min, such that after 5 h, 80% of the products are completely deoxygenated (2-O removed). Figure 4.4 shows the reaction network for the HDO of GUA on Ru/AC similar to that observed on Pd/AC with ring hydrogenation as the primary reaction on Ru/AC. Table 4.4 summarizes the kinetic constants k1, k2 and k3 estimated from the measured data (Tables E.13-E.15) based on the lumped reaction network of Figure 4.4, using the same methodology as for the Pd/AC (details in Appendix D.2), with the fit to the data reported in  Figure 4.8. The ring saturation reaction constant k1 is higher than that for k2 and k3, but the difference is not as significant as with Pd/AC. The calculated apparent Ea for ring saturation on Ru/AC catalyst is 40±6 kJ/mol, with significantly higher values for the O-removal reactions.   Table 4.4: Estimated 1st-order rate constants and activation energy (Ea) for lumped model of Figure 4.3 for the hydrodeoxygenation of 2-methoxyphenol over Ru/AC catalyst at different reaction temperatures of 240, 300 and 330 oC. ki (x 10-2mL/molCO.min) Reaction temperature (oC) ki0  (x 10 -2 mL/molCO.min) Ea (kJ/mol) 240 300 330 k1 1.8±0.0 4.8±0.8 7.4±0.4 4.8±0.01 40±6 k2 0.5±0.0 2.4±0.3 5.1±0.5 2.4±0.06 68±1 k3 0.1±0.0 0.8±0.1 6.1±0.5 1.3±0.5 110±25 k4 <0.01 0.2±0.0 0.2±0.0 - NA Note: 𝑘𝑖 = 𝑘𝑖0𝑒−𝐸𝑎𝑅𝑇    88    Figure 4.3: Mole percentage (mol.%) of reactant (2-methoxyphenol) and products as a function of reaction time on Ru/AC at different reaction temperatures ((a) 240 oC; (b) 300 oC; (c) 330 oC). (HYD: 2-methoxycyclohexanone; 2-methoxycyclohexanol; 1-methycyclohexane-trans-1,2-diol, 1-O removed: cyclohexanol and methoxycyclohexane. 2-O removed: cyclohexane, benzene, cyclohexene, cyclopentane and methylcyclopetane. Alkylation&HYC: ethylcyclohexane, methylcyclohexane, 1-ethyl-2-methylbenzene, dimethylcyclohexane and dimethylbenzene etc.)   89    Figure 4.4: Proposed reaction network for HDO of 2-methoxyphenol on Ru/AC.   4.3.4 HDO of 2-Methoxyphenol (GUA) on Mo2C/AC The conversion of GUA on Mo2C/AC reached 80% after 135 min at 330 oC, which reflects a significantly slower rate of GUA conversion on the Mo2C/AC compared to Pd/AC or Ru/AC (Table 4.2). More importantly, however, the data of Tables E.16, E.17, & E.18 indicate that the HDO of GUA on the Mo2C/AC does not proceed through an initial HYD reaction, as with the Pd and Ru catalysts. At 330 oC the main product of the reaction is phenol (77 mol.%), with 2-methylphenol and 4-methylphenol produced at the same selectivity (7 mol.%) along with 5 mol.% anisole. There are no ring saturation products generated and the 1-O removed products do not react further, so that the completely deoxygenated products benzene (3 mol.%) and toluene  90  (1 mol.%) are only minor products. When conversion reaches 100%, there is not much change in terms of product distribution compared to 80% conversion, suggesting that the phenols are stable and not very reactive on Mo2C/AC at this temperature (330 oC).  Since GUA is not as reactive on Mo2C as it is on the noble metals, relatively higher reaction temperatures (330 C, 350 C and 375C) were applied to study the obtained products as a function of reaction time on the Mo2C/AC. At 330 C almost all the GUA converts to phenol, a result of the direct dexoygenation reaction (DDO), as has been observed previously on Mo2C [59], transition metal phosphides and Mo2N catalysts [174, 175]. After 90 min reaction anisole, 2-methylphenol and 4-methylphenol also appear and the initial reactant GUA is consumed after 240 min. Most products generated at this temperature retain one O atom with 7% of products completely deoxygenated after 300 min (Table E.18) and more of the completely deoxygenated products benzene and toluene are formed after long reaction times (5 h). Hydrogenated products cyclohexane and cyclohexene are generated in trace amounts, (< 1 wt%). At 350 C, GUA is completely consumed after approximately 180 min and the 1-O removed products dominate with 11% of the products completely deoxygenated after 300 min (Table E.17). Finally, at 375 C, GUA is completely consumed after 135 min, and the completely deoxygenated products and the 1-O removed product yields are almost equal after 300 min (Table E.18). Phenol is converted, 2-methylphenol is relatively more stable than 4-methylphenol, but anisole also reacts at this temperature. The main completely deoxygenated product is benzene.   Figure 4.5 shows the lumped product distribution from the HDO of GUA on Mo2C/AC at different reactions temperatures and times and Figure 4.6 shows the proposed lumped reaction  91  network for the HDO of GUA on Mo2C/AC, based on the lumped product distribution, similar to that used for the Pd/AC and Ru/AC catalysts. However, no HYD products were observed on the Mo2C/AC catalyst and the deoxygenated products (phenol, 2-methylphenol, 4-methylphenol, and anisole) and the completely deoxygenated products (toluene; benzene, with trace amount of ring saturation products such as 1-methylcyclohexene, methylcyclohexane, cyclohexene and cyclohexane) were quite different from that observed with the noble metals, as shown in Figure 4.6. Significantly, there were no HYD products generated before the formation of products with one O atom removed, as was the case on the Pd/AC and Ru/AC catalysts. Table 4.5 summarizes the kinetic constant k2 and k3, obtained using the same Marquardt-Levenburg algorithm (Appendix D.2), representing reaction constants for the removal of the first and second oxygen group, respectively and the fit to the measured data is illustrated in Figure 4.9. The corresponding k2 and k3 are 9.9 x10-2 and 0.2 x10-2 mL/molCO.min at 330 C, 17.5 x 10-2and 0.8 x 10-2 mL/molCO.min at 350 C, 28.4 x 10-2 and 2.1 x 10-2 mL mL/molCO.min at 375 C. The corresponding activation energies were 83±3 and 151±8 kJ/mol, respectively.  Table 4.5: Estimated 1st-order rate constants and activation energy (Ea) for lumped model of Figure 4.6 for the hydrodeoxygenation of 2-methoxyphenol over Mo2C/AC catalyst at different reaction temperatures of 330, 350 and 375 oC. ki  (x 10-2 mL/molCO.min) Reaction temperature (oC) ki0  (x 10-2 mL/molCO.min) Ea  (kJ/mol) 330 350 375 k2 9.9±0.6 17.5±1.4 28.4±6.4 17.4±0.3 83±3 k3 0.2±0.1 0.8±0.1 2.1±0.1 0.8±0.04 151±8 Note: 𝑘𝑖 = 𝑘𝑖0𝑒−𝐸𝑎𝑅𝑇 92   Figure 4.5: Mole percentage (mol.%) of reactant (2-methoxyphenol) and products as a function of reaction time on Mo2C/AC at different reaction temperatures ((a) 330 oC; (b) 350 oC; (c) 375 oC). (1-O removed: phenol; 2-methylphenol; 4-methylphenol and anisole. 2-O removed: toluene and benzene.).   93   Figure 4.6: Proposed reaction network for HDO of 2-methoxyphenol on Mo2C/AC catalyst. 1-O removed 2-O removedReactant: GUAk2Main reaction route2nd reaction routeMinor reaction routek3 94   Figure 4.7: Experimental and model concentration data versus reaction time of Pd/AC catalyst at (a) 300 oC and (b) 330 oC.    95   Figure 4.8: Experimental and model concentration data versus reaction time of Ru/AC catalyst at (a) 300 oC and (b) 330 oC.  96   Figure 4.9: Experimental and model concentration data versus reaction time of Mo2C/AC catalyst at (a) 330 oC, (b) 350 oC and (c) 375 oC.  4.4 Discussion The present study compares the HDO of 2-methoxyphenol over Pd/AC, Ru/AC and Mo2C/AC catalysts, with kinetic data measured in a liquid phase batch reactor over a range of temperatures and at high H2 pressure (initial H2 3.4 MPa). The main products at 330 C and 100 % conversion of GUA on the Pd catalyst are cyclohexane, 2-methoxycyclohexanol and 1-methylcyclohexane-trans-1, 2-diol and on Ru the main products are cyclohexane and benzene. Similar results were reported in the catalyst screening study by Chang et al. [77], done in a fixed-bed reactor in which both a Ni/C and a Pd/C catalyst showed high selectivity to products with phenyl ring saturation.  97  Selectivity data, measured as a function of reaction time in the batch reactor (Tables E.10-18), are consistent with the hydrogenation of the phenyl ring of GUA as the primary reaction on both the Pd and Ru catalysts.   Elliott and Hart [56] reported a product distribution for the HDO of 2-methoxyphenol on Ru/AC at 250 oC and 13.7 MPa, with 2-methoxycyclohexanol and 1, 2-cyclohexanol as primary products that gradually convert to cyclohexanol, cyclohexanone and phenol. At 80% conversion, the hydrogenated products accounted for 30% of the product and approximately 60% of the products retained one O atom. The selectivity is different from that reported herein for the Ru/AC catalyst operated at 240 oC and 3.4 MPa, with hydrogenated compounds accounting for 60% of the product and 30% of the products with one O atom removed at 80% conversion. The differences in selectivity are likely due to the different pressures used in the two studies, with higher H2 pressure   enhancing conversion of hydrogenated intermediates to products with 1-O removed. In both studies, completely deoxygenated products were present in trace amounts and ring saturated products accounted for > 90% product selectivity, showing the significance of the HYD reaction route on Ru.  However, at low pressure (PH2 = 40 kPa), Sun et al. [73] reported  60% selectivity toward DDO products (phenol) on Pd/C and Ru/C catalysts at 250 oC.  Boonyasuwat et al.[64] reported similar results at high temperature (400 C) and low pressure (PH2 = 0.1 MPa), with 66% selectivity to phenol on Ru/C catalyst at 20 % conversion. The low pressure of these studies results in low hydrogenation rates, emphasizing the deoxygenation reaction selectivity. However, Chang et al. [77] reported that at higher temperature (400 C) and high pressure (4 MPa), DDO products were also favored (selectivity 70 %) on Ru/C, with primary products catechol and phenol converted to benzene with increased reactor space time.  98  These results differ from the present study, in part because of the longer reaction times, the impact of products on the reaction kinetics through competitive adsorption [69] and the effect of heat–up encountered with the batch reactor. The fixed-bed reactor used by Chang et al. [77] had much shorter residence times (< 4 min) than those of the batch reactor (up to 5 h). The present study also used decalin as diluent for the reactant and at high temperature (330 C), over the Ru and Pd catalysts, some decalin was converted to tetralin, suggesting the possibility of H-transfer reactions that may enhance the hydrogenation of the reactant at high temperature. However, the amount of H available from decalin dehydrogenation was relatively small compared to the H2 in the gas phase. Furthermore, at the high temperature of their study, the DDO reaction with high activation energy that converts guaiacol to catechol and then phenol, is favored [174]. The high reaction temperature also favors dehydrogenation reactions resulting in benzene (rather than cyclohexane) as the major deoxygenated product.   The data of Figure 4.7 and 4.8 show that in the present study a significant fraction of the GUA reactant was hydrogenated to 2-methoxycyclohexanone, 2-methoxycyclohexanol and 1-methylcyclohexane-trans-1,2-diol during the reactor heat-up period (t = 0). The low Ea for the lumped hydrogenation steps reported for Ru and Pd, indicate that the hydrogenation reaction is relatively facile at low temperature. At higher temperature, DDO via catechol and phenol formation competes with the hydrogenation reaction, but since the starting reactant is already significantly hydrogenated, the main deoxygenated product is cyclohexane, in agreement with the results of Elliott and Hart. [56] Although the hydrogenation of the phenyl ring has been reported as a primary reaction on Pd [176] and Rh [177] catalysts, theoretical studies indicate that on Ru, the reaction proceeds by removing the first O of guaiacol via C–O cleavage of  99  catecholate, followed by removal of the second O, via C–O scission of phenolate, with the much higher barrier energy of the latter step resulting in high phenol selectivity.[69, 70] More recently, Lu and Heyden [71] presented a DFT study of the HDO of guaiacol on Ru and concluded that the dominant hydrodeoxygenation pathway was via O–H bond cleavage of guaiacol to yield C6H4(O)(OCH3), followed by dehydrogenation of the methoxy group to C6H4(O)(OC), decarbonylation to C6H4O, and finally hydrogenation to phenol. Although they reported the hydrogenation of the phenyl ring to be unfavorable on Ru, they indicated that –OH removal from phenol could occur by partial phenyl ring hydrogenation at high hydrogen pressures, which leads to benzene and cyclohexane in the product. The experimental results are consistent with the latter case in which a high H2 pressure environment is provided and hydrogenation of phenyl could happen prior to deoxygenation. Furthermore, the DFT study of Rubes et al. [178] points to the strong dependence of selectivity on the H content of the catalyst surface.  The HDO of GUA on Mo2C/AC occurs at a much slower rate and by a different kinetically significant pathway, compared to the Pd or Ru catalysts. The results show that direct deoxygenation (DDO) occurs, yielding phenol as the primary product, with only trace amounts of hydrogenated products identified. During the reactor heat-up to the reaction temperature of 375 C, almost all the GUA reactant is converted to phenol rather than the hydrogenated products and the GUA conversion during heat-up was much lower on the Mo2C than the Pd and Ru catalysts. Jongerius et al. [59] also report that Mo2C/carbon has a very high DDO selectivity in the HDO of GUA, with a product distribution of phenol 57%, methyl phenol 18%, anisole 10% and benzene and toluene 6%, cyclohexane, cyclohexene, and cyclohexanone 6% at nearly 100% conversion after 4 h reaction at 350 C and 5.5 MPa. In the present study, after 4 h at 350  100  C and 3.4 MPa, the product distribution was phenol 63%, methylphenols 12%, anisole 6% and benzene and toluene 10%, cyclohexane, cyclohexene, and cyclohexanone 3%, at 100% conversion and in good agreement with the results of Jongerius et al. [59].   Previous studies have reported that on Mo2C, phenol is the result of the DDO reaction, in which GUA undergoes a direct demethoxylation to form phenol, rather than demethylation to form catechol that subsequently undergoes Ar-OH bond cleavage to yield phenol. Direct demethoxylation reaction routes are also reported on transition metal phosphide catalysts and Mo2N catalyst [174, 175]. Although Mo2C is known to have hydrogenation ability and has been used in benzene and toluene hydrogenation [53, 54], during HDO the Mo2C absorbs oxygen on the surface leading to a change in properties [101] that are known to suppress hydrogenation rates [100]. This adsorbed oxygen is also known to generate acidity [102] which favors the DDO route in HDO reactions [179]. Consequently, deoxygenation reactions occur before saturation of the phenyl ring on the Mo2C catalyst.   The simplified kinetic analysis presented herein provides a quantitative comparison of the GUA consumption rates, on a per active site basis (as estimated from CO uptake data), such that the catalyst activities can be ranked in the order of decreasing activity as Pd  Ru  Mo2C. The hydrogenation activity of Pd is significantly higher than that of Ru, whereas Ru is more active in cleaving the C-O bond of the R-OH and R-OCH3 groups, as has been noted previously [77, 180]. Note however, that the rate constant associated with the removal of the second O atom suggests Pd/AC is more active than Ru/AC for this step, likely due to competitive dehydrogenation and alkylation reactions that occur on the Ru/AC catalyst. On the Mo2C/AC catalyst, the apparent  101  activation energy for removal of the methoxy group is 83 kJ/mol, in good agreement with the data of Lee et al. [100] which show that the apparent activation energy required for breakage of Ar-OCH3 using anisole as model compound on Mo2C is between 67.8~75.8 kJ/mol. However, to break a Ar-OH bond is much more difficult and the apparent activation energy on Mo2C is 151 kJ/mol.   4.5 Conclusions A comparison of the HDO of GUA in the liquid phase over Pd, Ru and Mo2C catalysts supported on activated carbon (AC), using lumped kinetics, shows that that the hydrogenation activity of the Pd catalyst (on a per site basis, as determined from CO uptake measurements), is about 6 times higher than the Ru, but Ru is more active for O removal. Although the Mo2C is the least active, it is most efficient in terms of O-removal with minimal H2 consumption. On both Pd and Ru, hydrogenation of the phenyl ring occurs before O bonds are cleaved; whereas, on Mo2C, HDO of GUA occurs by direct demethoxylation to yield mostly phenol.   102  Chapter 5: The Effect of Metal Promoters on Carbon Supported Mo2C Catalysts in Hydrodeoxygenation of Dibenzofuran 5.1 Introduction In the previous chapter, the activity and selectivity of Mo2C applied to the HDO of guaiacol was compared to noble metal catalysts. The product distributions and lumped kinetics of the catalysts were used to quantify the differences among the catalysts. The Mo2C presented high DDO selectivity and consumed less hydrogen during HDO reactions. However, the activity of the Mo2C was much lower than that of the noble metal catalysts when operated at the same reaction conditions. In this chapter, the possibility of adding a second metal as promoter of the Mo2C catalyst activity and selectivity is investigated, while the role of oxygen during HDO reaction is examed.  In this chapter, the effect of promoters (Ni, Pd, and Cu) on the activity and stability of Mo2C supported on activated carbon derived from bio-char (Mo2C/ABC) during the HDO of dibenzofuran (DBF) is reported. The impact of the promoters on the CHR process is also reported. The performance of different metal promoters is investigated experimentally, using several characterization techniques (BET, TPO-TPR, TPR-MS, HRTEM, and XPS) to study the changes in catalyst surface/morphology during HDO. The structural changes are closely related to the stability of the Mo2C/ABC catalysts. Furthermore, density functional theory (DFT) calculations have been applied to study the impact of promoters on the binding energy of DBF to the M-Mo2C (where M is Ni, Pd or Cu promoter) catalyst surface. These data were used to better understand the function of the promoters of the Mo2C catalyst during the HDO of DBF. Ni as promoter has much better performance than Cu and is less costly than Pd. The data reported in  103  this chapter show that the Ni promoted Mo2C catalyst, supported on ABC, is a candidate catalyst for the bio-oil upgrading process.    104  5.2 Experimental 5.2.1 Catalyst Preparation Bio-char derived activated carbon (referred to as ABC) was prepared by thermochemical activation using the same method described in previous chapters. Then, an incipient wetness impregnation of the ABC was done using ammonium heptamolybdate (AHM) as the Mo precursor. The impregnated ABC was left at room temperature for > 4 h to stabilize and air dry. The obtained precursor was designated as AHM/ABC, with a nominal Mo mass loading of 10 wt%. A second impregnation using relevant solutions of Ni-, Cu- or Pd-nitrate followed, to obtain a nominal 1 wt% loading of the Ni, Cu or Pd. All catalyst precursors were then calcined at 300 oC for 5 h under N2 atmosphere prior to carbothermal hydrogen reduction (CHR) [123, 134] to generate the Mo2C. CHR was done under H2 flow (150 mL(STP)/min) while heating from room temperature to 500 oC at a ramp rate of 5 oC/min, then to 550 ~ 700 oC at 1 oC/min, with the final temperature dependent on the added promoter and held for 90 min before quenching to room temperature in N2. The CHR product gas was analyzed by a quadrupole mass spectrometer (QMS; VG Pro Lab Thermo-Fisher Scientific) with time-on-stream. The resulting catalysts are designated as M-Mo2C/ABC, where M represents the corresponding metal promoter. The relevant C loss during CHR was reported as burn-off % for all prepared catalysts. Finally, the obtained catalysts were passivated in a 1% O2/N2 flow for 2 h at room temperature before removal from the CHR U-tube reactor.    105  5.2.2 Catalyst Characterization 5.2.2.1 N2 Adsorption and Desorption The specific surface area and pore volume (including Vtotal, Vmicro and Vmeso) of both passivated fresh catalysts and used catalysts (after HDO of DBF) were determined using a Micromeritics ASAP 2020 analyzer. The specific surface area was calculated by the 2D-NLDFT method, details of which are given in previous chapters. Further details are provided in Chapter 2.2.3.  5.2.2.2 H2 Temperature-Programmed Reduction (H2-TPR) The effect of the O introduced following passivation of the Mo2C catalysts was studied by a temperature-programmed reduction (TPR) using a Micromeritics Autochem II 2920 analyzer with TCD detector. About 0.1 g of the dried catalyst precursor was loaded in a U-tube reactor and flushed with He for 30 min to remove any potentially adsorbed species. Then, Mo2C/ABC and M-Mo2C/ABC catalysts were prepared in-situ by CHR as described in the catalyst preparation section. Subsequently, the in-situ prepared catalysts were cooled to 100 oC in N2 flow prior to O2 treatment. A 1 vol% O2/N2 was introduced to the fresh Mo2C catalysts at a flow rate of 50 mL (STP)/min for 1 h to mimic the influence of O on Mo2C catalysts during HDO. A temperature-programmed reduction (TPR) in H2 was then performed using 9.5 vol%H2/Ar at a flow rate of 50 mL (STP)/min. The TPR was started by heating the sample to 100 oC, 200 oC and 400 oC, sequentially at a rate of 10 oC/min and holding at each temperature for 1 h. The position of the H2 peak associated with the consumption of H2 by the O of the catalyst is a measure of the difficulty of the reduction on the Mo2C catalysts and the peak area quantifies the H2 consumed during the reduction.    106  5.2.2.3 X-Ray Photoelectron Spectroscopy (XPS) X-ray photoelectron spectroscopy (XPS; Kratos Analytical Axis Ultra DLD) was used to measure the surface properties of the catalysts before and after the HDO reaction. The used catalyst was pretreated with a H2 flow for 2 h at 400 oC to remove the O layer that arises during passivation and storage. The sample preparation for XPS was conducted in a glove box and stored in a vacuum desiccator to limit catalyst oxidation by exposure to the atmosphere. Mono-Al was used as the photon source generated at 150 W (hʋ=1486.6 eV). A pass energy of 160 eV was used for the survey scan and 20 eV was applied for the narrow scan. The C 1s peak at 284.5 eV was used as the reference peak. The peak deconvolution was conducted by XPSPEAKER 41 using a nonlinear least-squares method with a combination of Lorentzian (20%) and Gaussian (80%) peak shapes.   5.2.2.4 STEM/EDX Analysis The morphology of the particle was characterized by STEM/EDX (FEI Tecnai Osiris) at 200 kV with a resolution limit of 0.14 nm. Sample treatment details are same as in previous chapters. High angle annular dark field scanning (HAADF) and energy-dispersive X-ray (EDX) elemental mapping were used to reveal the morphology change of the used Mo2C catalyst.   5.2.3 Catalytic Test for Hydrodeoxygenation (HDO) of Dibenzofuran (DBF) The gas phase HDO of dibenzofuran (DBF) was carried out in a packed bed reactor (length: 500 mm, hot zone: 300 mm; internal diameter: 8.64 mm) to assess the catalytic activities of the Mo2C/ABC and M-Mo2C/ABC catalysts. The experiments were operated by varying reaction temperature from 230 to 350 oC at a constant H2 pressure of 4.1 MPa. The liquid feed, consisting  107  of 2.0 wt% DBF in decalin, was fed to the reactor through a high-pressure piston pump (Gilson model 307) at a volumetric flow rate of 0.1667 mL/min to provide a constant liquid-hourly space velocity (LHSV) of 4 h-1. The H2 flow was controlled by a mass flow controller (Brooks 5850TR) at 100 mL (STP)/min. The product gas passed through a condenser held at room temperature to recover the liquid products of the reaction. Once the reactor temperature reached the set-point, the reaction time was recorded as t = 0 min and a liquid sample was drawn from the condenser. During the HDO reaction, the first three liquid samples were collected every 30 min with subsequent samples drawn periodically at 1 h intervals. The liquid samples were analyzed by GC-MS (Shimadzu (QP-2010-S) GC/MS and RTX5 30 m ×0.25 mm capillary column) based on an external calibration method. The calibration curves were built with a R2 ≥ 99% to ensure the accuracy of the measurement. Overall, the reported activity data have a carbon balance of ≥ 95% and the error in the activity data is within ±8%. (See Table E.19, and E.20 of the Supporting Information). Appendix G.1 has more data for carbon balance as example for error and repeatability throughout the thesis. The mass transfers and heat transfer effects in fix-bed reactor is excluded by calculation in Appendix F.1, F.2 and F.3.  The reactant conversion and relevant products selectivities were calculated as follows:  DBF conversion (%) = 100x∑ 𝑦𝑖𝑦𝐷𝐵𝐹𝑜       (Eq. 5-1) HDO conversion (%) = 100x∑ 𝑦𝑖−∑ 𝑦𝑜𝑖𝑦𝐷𝐵𝐹𝑂      (Eq. 5-2) Selectivity of product 𝑖 (%) = 100x𝑦𝑖∑ 𝑦𝑖     (Eq. 5-3)  108  , where yi is the molar faction of the product i; ∑ 𝑦𝑖is the sum of molar fractions of all products; 𝑦𝐷𝐵𝐹𝑜  is the molar fraction of DBF in the liquid feed; 𝑦𝑜𝑖 is the mole fraction of product i that contains O.   5.2.4 DFT Calculation using VASP Density functional theory (DFT) related calculations were done using Vienna Ab Initio Simulation Package (VASP), which uses a plane-wave based periodic supercell method [181, 182]. The projector augmented wave (PAW) method was applied to describe the electron-ion interaction [183]. The generalized gradient approximation with Perdew–Burke–Ernzerhof (GGA-PBE) was used by considering the van der Waal force (correction method: dDsC dispersion) [184]. A four-layers slab β-Mo2C model was built with a vacuum layer of 15 Å. All calculations were done with cut-off energy of 400 eV and a total convergence of 10-4 can be obtained. In the bulk optimization, the lattice parameters for β-Mo2C were determined by a conjugated gradient algorithm to relax the ions. A 5 × 5 × 5 Monkhorst-Pack k-point grid was used for sampling the Brillouin zone while the 3 × 3 × 1 k-points grid was applied for the periodic slabs. To simulate the effect of metal promoter inside the structure of Mo2C, three similar surfaces were acquired by replacing one of the Mo terminal atoms on the top surface with the metal atoms Cu, Ni or Pd.    After optimization of all four slabs, adsorption calculations were conducted on these surfaces. The adsorbed H2 and DBF on the topmost layer were relaxed to their optimized positions. The DBF adsorption energy Eads was calculated by the following equation: Eads= E(X/slab) – EX – E(slab)       (Eq. 5-4)  109  where E(X/slab) is the energy of the adsorbed system with both slab and specie X, EX is the energy of specie X, and E(slab) is the energy of the slab. Note that the dissociative adsorption of H2 was conducted on both the added metal promoter and on the Mo site near the added metal to study the impact of the addition of the promoter metal on the nearby Mo site.  EIS, ETS, and EFS are used to represent the energies of the initial state, transition state and final state correspondingly in the study of the dissociation adsorption energy of H2. 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 energy of adsorption ΔG is calculated by the following equation [185]: ΔG = 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 DBF; 𝑧𝐷𝐵𝑇𝑟𝑜𝑡 (𝑇) represents the rotational partition function of DBF; T is temperature in Kelvin; P is the partial pressure of DBF 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. The Gibbs free energy of adsorption ΔG were calculated for both clean and metal added surfaces as well as the dissociative adsorption of H2.   110  5.3 Results 5.3.1 Catalyst Synthesis and Characterization The preparation of Mo2C/ABC catalysts was carried out by a controlled temperature-programed reduction in H2. Since the addition of the 2nd metal (Ni, Pd, or Cu) has the potential to influence the carburization process [186] , the most appropriate CHR conditions for each promoted catalyst need to be determined. A previous study [126] demonstrated  that the CHR temperature plays a significant role in determining the properties of the catalysts following CHR. Herein the CHR with product gas analysis by mass spectrometer (MS) for each of the promoted catalysts was conducted to determine the required reduction temperature for each. Figure 5.1 reports the CHR-MS profiles of the M-Mo2C/ABC catalysts. The H2 consumption is given by the relative intensity of the m/z = 2 signal, as shown in Figure 5.1(a). Similarly, the relative intensities of m/z = 15, 28 and 44 represent the CH4, CO and CO2 generated during CHR, as reported in Figure 5.1(b-d), respectively. The reduction of the metal precursor was initiated as indicated by the H2 consumption peak at approximately 400 oC. The reduction was accompanied by the simultaneous generation of CO, CO2 and H2O, although the H2O signal is not reported due to detection limitations, but H2O production has been confirmed in other studies [123, 187]. Compared to the H2 consumption peak for Mo2C/ABC of Figure 5.1(a), it can be clearly seen that Ni addition shifted the peak to lower temperature; whereas Cu showed the opposite trend. Following reduction of the catalyst precursor, carburization was initiated as the CHR temperature increased above 500 oC, with CH4 the main product detected in the product gas. As shown in Figure 5.1(b), the Ni-Mo2C/ABC catalyst generated the most CH4 among all catalysts; whereas Cu-Mo2C/ABC generated the least. The CH4 production mirrors the H2 consumption, with a large H2 consumption occurring on the Ni-Mo2C/ABC at > 500 oC, while H2 consumption was almost  111  constant for the Cu-Mo2C/ABC. For the Mo2C/ABC and Pd-Mo2C/ABC catalysts, both the H2 consumption and CH4 production signals show similar trends and magnitudes. Hence a low CHR temperature of 550 oC was used for the Ni-Mo2C/ABC catalyst preparation while the other catalysts were prepared at 650 oC to ensure the formation of Mo2C species [126, 133].    Figure 5.1: TPR-MS profiles of carbothermal hydrogen reduction (CHR) of Mo2C/ABC and M-Mo2C/ABC catalysts. (a) m/z=2, H2; (b) m/z=15, CH4; (c) m/z=28, CO; (d) m/z=44, CO2. (The signal is presented by the relative intensity of H2, CH4, CO, and CO2 to inert gas He)  Table 5.1 reports the surface area and total pore volume of the catalysts with different promoters. The activated bio-char (ABC) surface area of 2000 m2/g and pore volume associated  with  112  micropores, is in agreement with the results of Azargohar and Dalai [78]. The fresh catalysts all have similar surface areas (~1300 m2/g), while the mesopore volume (Vmeso) is significantly increased after CHR, accounting for ~35% of the total pore volume. The burn-off% is similar for Mo2C/ABC, Pd-Mo2C/ABC and Cu-Mo2C/ABC catalysts due to similar CHR temperatures; whereas, the burn-off% for the Ni-Mo2C/ABC catalyst is significantly lower because of the lower CHR temperature (550 oC).    113  Table 5.1: Physical properties of fresh and used Mo2C/ABC catalysts with different metal promoters. Sample Surface area (m2/g) Pore volume (cm3/g) Burn-off% Vmicro Vmeso Vtotal ABC 2090a 0.84b 0.00c 0.84d 40e Fresh Mo2C/ABC 1485 0.65 0.31 0.96 36 Used Mo2C/ABC  970 0.42 0.30 0.72 — Fresh Ni-Mo2C/ABC 1326 0.44 0.49 0.93 19 Used Ni-Mo2C/ABC 921 0.30 0.31 0.61 — Fresh Pd-Mo2C/ABC 1474 0.64 0.37 1.01 37 Used Pd-Mo2C/ABC 1124 0.51 0.36 0.87 — Fresh Cu-Mo2C/ABC 1220 0.50 0.31 0.81 34 Used Cu-Mo2C/ABC 656 0.28 0.30 0.58 — a. The specific surface area is calculated from the measured N2 absorption isotherm using 2D-NLDFT with P/Po in the range of 0.01~0.30.  b. The micropore volume (Vmicro) is calculated by 2D-NLDFT method for pores ≤ 2 nm. c. The mesopore volume (Vmeso) is calculated by 2D-NLDFT method for pore size between 2~50 nm.  d. The total pore volume (Vtotal) is the sum of Vmicro and Vmeso. e. The burn-off% represent the lost C amount during CHR and it is calculated by: Burn-off% = 100×(1-𝑚𝑓𝑖𝑛𝑎𝑙𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙), where 𝑚𝑓𝑖𝑛𝑎𝑙  is the mass after CHR and 𝑚𝑖𝑛𝑖𝑡𝑖𝑎𝑙  is the mass before CHR.    114   Figure 5.2: H2-TPR of pre-oxidized Mo2C/ABC and M-Mo2C/ABC (M=Ni, Pd, or Cu) catalysts at different reduction temperatures. (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC.   To assess the effect of O on the Mo2C in the presence of the promoters, a 1 h O2 passivation was conducted on freshly prepared Mo2C/ABC and M-Mo2C/ABC catalysts followed by H2-TPR as reported in Figure 5.2(a-d) and Table 5.2. Three H2 reduction steps were employed, with the temperature ramped sequentially to 100, 200 and 400 oC (identified as step I, II and III) and the H2 consumption quantified in each phase. The reduction was done at ≤ 400 oC to minimize impact on the original catalyst. Figure 5.2(a) shows that no H2 consumption occurred during step I and II for the Mo2C/ABC catalyst; whereas a large H2 consumption peak was evident as the temperature increased from 200 to 400 oC, with the peak maximum occurring at 400 oC. In the  115  presence of the 2nd metal, the reduction changed significantly, as shown in Figure 5.2(b-d). On the Ni-Mo2C/ABC catalyst (Figure 5.2(b)) three reduction peaks occurred at ca. 60, 180 and 380 oC. Similarly on the Pd-Mo2C/ABC catalyst three peaks at 90, 150 and 300 oC were evident (Figure 5.2(c)), suggesting that both Ni and Pd enhanced O removal at low temperature, presumably by increasing H2 dissociation [111, 115, 118, 188, 189].  On the Cu-Mo2C/ABC catalyst, two peaks were detected at reduction temperatures of 200 and 400 oC. The peak at 200 oC is attributed to copper oxide reduction, as reported by Dow et al. [190]. The relevant quantity of each reduction peak is reported in Table 5.2. The total H2 consumption (molH2/molmetal) was comparable for the Pd-Mo2C/ABC, Cu-Mo2C/ABC and Mo2C/ABC catalysts. Note that the H2 consumption in step I and II appeared in the presence of promoters. For the Ni-Mo2C/ABC catalyst, 0.2 × 10-2 H2 consumption (molH2/molmetal) occurred in step II; while for the Pd-Mo2C/ABC catalyst the corresponding values was 0.4 × 10-2 H2 consumption (molH2/molmetal). The amount of H2 consumption (molH2/molpromoter) in step I was 0.2 × 10-2 for Ni-Mo2C/ABC catalyst and 1.6× 10-2 for Pd-Mo2C/ABC catalyst. The values in step I is calculated based on mode of promoters since the H2 consumption here is mostly contributed by promoters.   116  Table 5.2: H2-TPR analysis of pre-oxidized Mo2C/ABC and M-Mo2C/ABC catalysts.  Sample Step Ia Step IIa Step IIIa Total H2 consumption (molH2/molpromoter)b × 10-2 Reduced temperaturec (oC) H2 consumption (molH2/molmetal) × 10-2 Reduced temperature (oC) H2 consumption (molH2/molmetal) × 10-2 Reduced temperature (oC) H2 consumption (molH2/molmetal) × 10-2 Mo2C/ABC — — — — 2.2 400 2.2 Ni-Mo2C/ABC 0.2 60 0.2 180 1.3 380 1.5 Pd-Mo2C/ABC 1.6 90 0.4 150 1.3 300 2.0 Cu-Mo2C/ABC — — 0.8 200 1.4 400 2.2 a. Step (I), (II), and (III) represent different reduction temperature at 100, 200, and 400 oC, respectively.  b. The value is calculated based mole of promoter added.  c. The reduction temperature represents the max. reduction temperature of each peak.   117  The average particle size distribution of the Mo2C species, generated during CHR, was calculated from the lognormal distribution of the size distribution data measured by TEM and reported in Figure 5.3. The Mo2C/ABC, Ni-Mo2C/ABC and Pd-Mo2C/ABC were well distributed with average Mo2C particle size of 7.3, 5.3, and 9.3 nm, respectively. The smaller particle size of the Ni-Mo2C/ABC catalyst was because of the lower CHR temperature employed for this catalyst. However, for Cu-Mo2C/ABC, most of the particles were clearly large because of agglomeration and it is difficult to distinguish Cu from Mo2C. The HADDF-STEM image and EDX elemental mapping of the Ni-Mo2C/ABC and Pd-Mo2C/ABC catalysts are presented in Figure 5.4 and Figure E.11, respectively. The analysis shows that the majority of the Ni was closely associated with Mo, although a few isolated Ni particles were detected on the support. Unlike Ni, the atomic radius of Pd is larger than that of Mo, which makes it difficult to insert Pd into the Mo2C crystal structure. However, since the loading amount of Pd is relatively low, the Pd dispersion is high on the high surface area ABC support. As shown in Figure E.11, the Pd and Mo are closely associated with one another.     118      Figure 5.3: TEM images of fresh catalysts and corresponding particle size distributions: (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC.    119          Figure 5.4: High angle annular dark field TEM scanning images (HAADF-STEM) and Energy dispersive X-ray (EDX) elemental mappings of fresh Ni-Mo2C/ABC catalyst.     120  Table 5.3: XPS analysis of both fresh and used Mo2C catalysts with different promoters. Sample Area, % Atomic conc., % Mo2+ (carbide) 228.2 eV Moδ+ (2<δ<4) (oxycarbide)  229.0 eV Mo4+ (oxycarbide) 229.7 eV Mo6+  (oxide) 232.3 eV Mo 3d O 1s C 1s X (X=Ni, Pd, Cu) Fresh catalyst Mo2C/ABC 57 21 0 23 2.44 5.07 92.49 ― Ni-Mo2C/ABC 25 25 0 50 1.32 10.96 87.09 0.63 Pd-Mo2C/ABC 27 35 0 39 0.87 13.69 85.36 0.08 Cu-Mo2C/ABC 27 31 0 43 1.19 13.42 84.71 3.08 Used catalyst Mo2C/ABCa 0 1 27 72 1.37 3.19 95.43 ― Ni-Mo2C/ABCb 0 3 51 45 1.35 5.51 92.61 0.53 Pd-Mo2C/ABCb 0 1 29 70 1.15 9.66 89.08 0.11 Cu-Mo2C/ABCa 0 0 37 63 1.55 6.10 91.89 0.46 a. The catalyst was tested for 24 h at 350 oC, 4 MPa. b. The catalyst was tested for 24 h at 230 oC, 4 MPa.   121   Figure 5.5: XPS (Mo 3d region) narrow scan spectra of fresh Mo2C-typed catalysts: (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC.   The surface composition and oxidation state of the catalyst components was investigated by XPS. The effect of the metal promoters on the XPS spectra of the fresh catalysts is reported in Table 5.3 and Figure 5.5. The distribution of different Mo species was determined by deconvolution of the Mo 3d peaks into molybdenum carbide (Mo2+), oxycarbide (Moδ+, 1 < δ  122  <2), and oxide (Mo6+) species, each with a Mo 3d5/2 and Mo 3d3/2 spin-orbital splitting of 3.1 eV and area ratio of 3 to 2 [153]. The peaks at 228.2 eV and 231.3 eV (± 0.1 eV) can be assigned to Mo2C. The oxycarbide species at a B.E. of 229.0 eV and 232.1 eV were assigned to a lower oxidation state of oxycarbide species; whereas, B.E.s of 229.7 eV and 232.8 eV were assigned to higher oxidation state oxycarbide species [191, 192]. An intensity ratio of ½ and a spin-orbital splitting of 17.4 eV was used to fit the Ni 2p spectra (Figure 5.6a) [193-195]. The peak at 856.1 eV was assigned to Ni in Ni-Mo carbide and the peak at 852.7 eV was assigned to metallic Ni. Also, a strong shake-up satellite of Ni 2p3/2 is shown at 860 eV. For the Pd 3d5/2, both metallic Pd and Pd oxide were identified at 335. 2 eV and 340.1 eV (Figure 5.6b), respectively [196]. For Cu 2p, only metallic Cu was identified with a spin-orbital splitting of 19.9 eV and area ratio of 2 to1. The Cu 2p3/2 and Cu 2p1/2 appeared at 932.2 eV and 952.1 eV, respectively (Figure 5.6c) [119, 197].   123   Figure 5.6: XPS narrow scan spectra of Mo2C-type catalysts: (a) Ni 2p; (b) Pd 3d; (c) Cu 2p.  124   Figure 5.7: XPS (Mo 3d region) narrow scan spectra of used Mo2C-typed catalysts: (a) Mo2C/ABC; (b) Ni-Mo2C/ABC; (c) Pd-Mo2C/ABC; (d) Cu-Mo2C/ABC.  For the fresh catalysts, only the lower oxidation state oxycarbide was present, with peaks at higher B.E. (232.3 eV and 235.4 eV) corresponding to Mo oxide. The relative amount of Mo2C species on the Mo2C/ABC catalyst surface was estimated at 57%. For the Ni-Mo2C/ABC catalyst, the Mo2C amount decreased by about half (to 25 %) since the CHR temperature was lower than that of the Mo2C/ABC catalyst. The Cu-Mo2C/ABC and Pd-Mo2C/ABC catalysts had  125  similar Mo2C contents (27 % each), even though the same CHR temperature (650 oC) was used as in the preparation of the Mo2C/ABC catalyst. One of the possible reasons for the reduced Mo2C content is the phase separation of the promoter from the Mo2C, that results in purely metallic Pd or Cu on the catalyst surface, as shown in Figure 5.6. The XPS analysis indicated about 75% of the Pd was in the metallic state while all the Cu was metallic. The XPS analysis of the used catalysts, provided in Figure 5.7, indicated that none of the Mo2C remained after the reaction in DBF (Table 5.3). Also, except for the Ni-Mo2C catalyst, almost all the low valance Mo oxycarbide was transformed to high valance oxycarbide species with a B.E. of 229.7 eV. The Mo oxide content also increased, indicating the oxidation of Mo2C species under HDO reaction conditions. In addition, there was no significant change in the Ni 2p, Pd 3d, and Cu 2p XPS spectra before and after the HDO reaction (Figure 5.6).  5.3.2 Catalyst Activity and Selectivity in HDO of Dibenzofuran The hydrodeoxygenation of dibenzofuran (DBF; 2 wt%) at 350 oC, 4.1 MPa H2, and a LHSV = 4 h-1 was used to compare the catalytic activity of the Mo2C catalysts with and without promoters. A H2 reduction was conducted at 400 oC for 2 h prior to the performance tests to remove the passivation layer from the catalyst. Both the DBF conversion and product selectivity were measured with time-on-stream (TOS) for 450 min at 60 min time intervals. The activity was assessed with the same mass (ca. 0.72 gcat.) of catalyst loaded in the reactor. Figure 5.8 shows the DBF conversion and product selectivities averaged over the 450 min time-on-stream. During this period the variation in DBF conversion and product selectivity was ± 8 %. The DBF conversion reached 90% for the Mo2C/ABC catalyst and improved to 100% for the Ni- and Pd-Mo2C/ABC catalysts; whereas the DBF conversion for the Cu-Mo2C/ABC was 83%.  126   Figure 5.8: DBF conversion and selectivity of Mo2C/ABC and M-Mo2C/ABC catalysts at350 oC, 4.1 MPa and LHSV = 4 h-1.   127  Table 5.4: DBF conversion and product selectivites of Ni- and Pd-Mo2C/ABC catalysts at 4.1 MPa and LHSV=4 h-1 with varied temperatures in the range from 230 to 330 oC.  Catalyst Temperature, oC DBF conversion, % Product selectivity, % 2-CHP THDBF Cyclohexane BPH CHB BCH Ni-Mo2C/APC 330  100  0  0  36  0  17  48  310  100  0  0  25  0  12  63  290  95  6  0  11  0  9  74  270  75  24  2  5  0  6  64  250  52  38  7  1  0  3  51  Pd-Mo2C/APC 330  100  0  0  20  0  1  79  310  100  0  0  11  0  0  89  290  100  0  0  4  0  0  96  270  100  0  0  3  0  0  97  250  100  0  0  1  0  0  99  230  100  0  0  0  0  0  100    128  The products from the HDO of DBF were mainly cyclohexane, cyclohexylbenzene (CHB), bicyclohexane (BCH), and biphenyl (BPh), along with O-containing products tetrahydro-dibenzofuran (THDBF) and 2-cyclohexylphenol (2-CHP) (detected at low reaction temperatures < 310 °C). A reaction pathway is proposed here based on the report by Infantes-Molina et al. [198]. The reaction pathway for the HDO of DBF, illustrated in Figure 9, has one direct deoxygenation (DDO) pathway, with biphenyl (BPh) the main product and a second hydrogenation reaction (HYD) pathway that generates CHB and BCH. As reported in Figure 5.8, BPh accounts for 53% of the products over the Mo2C/ABC catalyst; whereas, no BPh is present over the Ni- and Pd- Mo2C/ABC, indicating that the DDO pathway is completely suppressed on the promoted Mo2C/ABC catalysts. Instead, hydrogenated products cyclohexylbenzene and bicyclohexane along with the cracked product cyclohexane (CH) are the main products. The selectivity towards cyclohexane is > 30% for Ni- and Pd-Mo2C/ABC catalysts.    Figure 5.9: Reaction pathway of dibenzofuran HDO on Mo2C based catalysts. (The products in the dash box represent the intermediates during the reaction)  129  The effects of reaction temperature (from 230 oC to 330 oC) on the DBF conversion and product selectivity over the Ni- and Pd-Mo2C/ABC catalysts is reported in Table 5.4.  For the Ni-Mo2C/ABC catalyst, complete conversion occurred at temperatures ≥ 310 oC. Below 300 oC, some O-containing products, such as 2-CHP and THDBF were detected. As temperature increased BCH increased to a maximum at 290 °C, whereas both CHB and cyclohexane increased over the entire temperature range. Note that there was no DDO product - BPh was not detected on either catalyst at all reaction temperatures. One possible reason is that all products are generated by the hydrogenation route, initiated through the hydrogenation of the benzene ring of DBF. Also note that the cracked product – cyclohexane increased with increased reaction temperature. For the Pd-Mo2C/ABC complete conversion of DBF occurred at all temperatures and similar to Ni-Mo2C/ABC, BPh was not detected at any temperature.   5.3.3 DFT Calculations The Gibbs free energy of adsorption of DBF on a clean Mo2C (101) surface and a metal-modified Mo2C (101) (metal = Ni, Pd or Cu) surface, was calculated by DFT and the results are reported in Table 5.5. Table E.21 - E.27 shows the corresponding adsorption positions of DBF on the different Mo2C surfaces. For the metal-modified Mo2C surface, the calculation was conducted by replacing one Mo atom with the metal promoter atom to simulate a modified Mo2C surface. Similar to our previous study of dibenzothiophene (DBT) [153], several modes of adsorption were examined  with vertical (V) or horizontal (H) adsorption orientations of the DBF, identified as DBF-V1, DBF-V2, DBF-H1, and DBF-H2 for the Mo2C(101) surface and as DBF-V1-DBF-V4 and DBF H1-DBF-H4 for the M-Mo2C(101) surface. For DBF-V1, DBF-V2, DBF-H1 and DBF-H2, the replaced metal (such as Ni, Pd, or Cu) is close to O from DBF; while  130  for DBF-V3, DBF-V4, DBF-H3 and DBF-H4, the replaced metal is close to the benzene ring of the DBF.  Table 5.5: The calculated Gibbs free energy of DBF adsorption on clean Mo2C (101) and M-Mo2C (101) (M=Ni, Pd or Cu) surface with different adsorption angles at 350 oC. Adsorption orientation Gibbs free energy of adsorption (eV) Mo2C Pd-Mo2C Ni-Mo2C Cu-Mo2C DBF-H1a -1.22 -1.29 -1.52 -1.34 DBF-H2a -1.12 -1.07 -1.12 -1.10 DBF-H3b — -1.10 -1.52 -1.12 DBF-H4b — -1.27 -1.15 -1.25 DBF-V1a -0.64 -0.58 -0.59 -0.84 DBF-V2a -0.81 -0.60 -0.64 -0.62 DBF-V3b — -0.88 -0.94 -0.69 DBF-V4b — -0.85 -0.88 -0.87 a. It represents two types of horizontal and vertical adsorption of DBF on clean Mo2C and M-Mo2C (101) plane where the replaced metal (M=Ni, Pd or Cu) is close to the O from DBF. b. It represents two types of horizontal and vertical adsorption of DBF on clean Mo2C and M-Mo2C (101) plane, where the replace metal (M=Ni, Pd or Cu) is close to the benzene from DBF.  The DBF binding energies were compared by assuming Mo-O binding between Mo on the Mo2C (101) surface and the O from DBF in all cases. The calculated results show that two types of horizontal adsorptions on clean Mo2C (101) surface (DBF-H1 = -1.22 eV; DBF-H2 = -1.12 eV) are stronger than that of the vertical adsorption (DBF-V1 = -0.64 eV; DBF-V2 = -0.81 eV). Thus, the horizontal adsorption is more favorable than vertical adsorption, similar to the results obtained for DBT adsorption on clean Mo2C (101) [153].   131  For the metal promoted Mo2C (101) surfaces, the adsorption between Mo and O mentioned above was studied as well as the adsorption between the O atom of DBF and the promoter metal (Ni, Pd or Cu) on the Mo2C surface. In Table E.22 and E.23, O from DBF was adsorbed on top of the Ni atom as indicated in the first two columns in each figure. The calculated horizontal and vertical ΔG results are DBF-H1 = -1.52 eV, DBF-H2 = -1.12 eV, DBF-V1 = -0.59 eV, and DBF-V2 = -0.64 eV. It is noted that the horizontal adsorption is significantly more favored; whereas the vertical adsorption is weakened by the presence of Ni. Also, DBF was adsorbed on top of a Mo atom near the Ni atom as shown in Table E.22 and E.23 (last two columns). The obtained horizontal and vertical adsorptions are -1.52 eV, -1.16 eV, -0.94 eV and -0.88 eV corresponding to DBF-H3, DBF-H4, DBF-V3, and DBF-V4, respectively. The same trend for horizontal adsorption was observed with a relatively weaker enhancement on vertical adsorption.  In Table E.24 and E.25, the DBF adsorption energy was calculated on the Pd-Mo2C surface by applying the same models used in the Ni-Mo2C calculations. The calculated ΔG for adsorption on top of Pd is DBF-H1 = -1.29 eV, DBF-H2 = -1.07 eV, DBF-V1 = -0.58 eV and DBF-V2 = -0.60 eV. The horizontal adsorption DBF-H2 weakens slightly and both vertical adsorptions (DBF-V1 and DBF-V2) are significantly weakened compared to the results obtained on the clean Mo2C (101) surface. Upon applying the same Pd-Mo2C surface, O from DBF was adsorbed on top of a Mo atom near the Pd atom as shown in Table E.24 and E.25 (last two columns). The calculated ΔG are -1.10 eV, -1.27 eV, -0.88 eV and -0.85 eV corresponding to DBF-H3, DBF-H4, DBF-V3, and DBF-V4, respectively. Different from DBF-V1 and DBF-V2, both vertical adsorptions were slightly enhanced in these two cases, although not much change was observed for the Pd-Mo2C surface.   132   Table E.26 and E.27 reports the horizontal and vertical adsorption of DBF on Cu-Mo2C (101) surface. The calculated adsorption results for adsorption on a Cu site are -1.34 eV, -1.10 eV, -0.84 eV and -0.62 eV corresponding to DBF-H1, DBF-H2, DBF-V1, and DBF-V2, respectively. The calculated adsorption results for adsorption on a Mo site near Cu are -1.05 eV, -1.18 eV, -0.54 eV and -0.77 eV corresponding to DBF-H3, DBF-H4, DBF-V3, and DBF-V4, respectively. These results suggest that on this surface, horizontal adsorption is slightly enhanced, and vertical adsorptions hardly change. Hence, we conclude that Cu as promoter on the Mo2C surface hardly changes the properties of the surface. In summary, Ni addition enhances the horizontal adsorption of DBF and among the three metal promoted surfaces, the Ni promoted surface has the most impact on the calculated G of DBF.  Table 5.6: The potential energy during dissociative adsorption of H2 on M-Mo2C (M=Pd, Ni or Cu) surfaces. Types Surface IS-H2b TSb FS-2Hb Ea (eV)c Er (eV)d Mo-O binding Clean Mo2Ca -0.40 -0.12 -0.93 0.28 -0.53 Mo-O binding Pd-Mo2C -0.40 -0.20 -0.45 0.20 -0.05 Ni-Mo2C -0.40 -0.19 -0.44 0.21 -0.04 Cu-Mo2C      M-O binding (M=Pd, Ni or Cu) Pd-Mo2C -0.03 0.63 -0.14 0.66 -0.11 Ni-Mo2C -0.12 0.42 -0.20 0.54 -0.08 Cu-Mo2C      a. These data were adopted from [153]. b. H2 means the initial state of the H2 molecular; TS is the transition state; 2-H means the final state of the dissociated H. c. Energy barrier (Ea) was calculated by: Ea = ETS - EIS d. Reaction energy (Er) was calculated by: Er = EFS - EIS   133  Dissociative adsorption of H2 on the M-Mo2C (M=Pd, Ni or Cu) surfaces is reported in Table 5.6 and Figure E.12. Previously, the dissociative adsorption of H2 on a clean Mo2C surface was investigated and the energy barrier Eb associated with the adsorption was estimated at 0.28 eV. Table 5.6 shows that the dissociative adsorption energy barrier on a Cu site of the Cu-Mo2C surface is 1.24 eV, but this decreases to 0.22 eV on a Mo site, indicating that Cu should significantly increase the hydrogenation ability of the Mo2C catalyst. For the Ni promoted surface, the H2 dissociative adsorption energy barrier on a Ni site is 0.54 eV and 0.21 eV on a Mo site. For the Pd promoted surface, the corresponding values are 0.66 eV and 0.20 eV, respectively. Overall, the DFT calculations indicate that all the metal promoted Mo2C surfaces should provide an easier path to H2 dissociative adsorption compared to the Mo2C surface alone.  5.4 Discussion The CHR-MS data show that the Ni, Pd and Cu promoters significantly influence Mo2C formation during CHR, increasing the CH4 generation rate and reducing the temperature at which CH4 is formed. Ni and Pd are very effective hydrogenation catalysts [199, 200] that readily dissociate hydrogen [201], a critical step in the generation of CH4. [37] Ni was particularly effective at increasing the amount of CH4 generated and decreasing the temperature required to initiate CH4 generation. A parallel CHR experiment done with a Ni/ABC catalyst (i.e. 1 wt % Ni on ABC without Mo2C) is reported in Figure E.13, showing significantly lower CH4 generation compared to the Ni-Mo2C/ABC. These data suggest that Ni alone cannot account for the increased CH4 production during CHR on the Ni-Mo2C/ABC. Rather the Ni likely increases activation of H2, yielding adsorbed atomic H that interacts with the Mo precursor and the activated carbon support, forming Mo2C and CH4 at lower temperature (550 °C). Note that CHR  134  of the Ni-Mo2C/ABC catalyst at 650 oC, results in high carbon burn-off and pore wall collapse as reported previously [202]. The data of Figure 5.1 also show that Pd does not promote CH4 generation during CHR as much as Ni, mostly likely because Ni is more efficient than Pd in hydrogenation of carbon. The H2 consumption during step I of the TPR was assigned to the reduction of the promoter oxide to the metallic phase. The H2 uptake in step I of the TPR for the Ni-Mo2C/ABC is 0.2× 10-2 (molH2/molpromoter), much less than for 1.6× 10-2 (molH2/molpromoter) the Pd-Mo2C/ABC, confirming that the amount of metallic Ni is much less than the amount of metallic Pd in the corresponding catalysts. The fact that the Ni atom is much smaller than Pd makes it easier to form a solid solution between the Ni and the Mo2C. In addition, the molar concentration of Ni is approximately twice that of the Pd since the promoters were added on a equal weight basis (i.e. 1 wt %), resulting in more active sites associated with the Ni promoter. While Ni and Pd promoted CH4 generation, Cu on the other hand suppressed the formation of CH4, with the CH4 concentration generated significantly lower than on the Mo2C/ABC catalyst.    The textural properties changed before and after reaction, with the M-Mo2C/ABC catalysts losing nearly 30% of both surface area and micropore volume, indicative of pore blocking during HDO. However, the activity data suggest that the catalyst activity was stable during HDO. Hence, the change in catalyst properties must occur rapidly at the beginning of the reaction and then remain relatively constant, as has been observed previously in the case of DBT HDS [126]. The fresh Cu-Mo2C/ABC catalyst pore volume and surface area are lower than the other catalysts because of the suppressed carburization during CHR. The Cu-Mo2C/ABC catalyst lost approximately half the surface area and micropore volume due to pore blocking during HDO.  135  The loss of surface area and micropore volume for the Ni- or Pd-Mo2C/ABC catalysts was less severe than for the Mo2C/ABC catalyst.   TEM images suggest that all the promoters were present as both isolated, reduced metal particles away from Mo2C particles and atoms closely associated with the Mo2C, possibly forming a solid solution in the case of Ni. The H2-TPR data of Figure 5.2 support this conclusion. The three different steps in H2 consumption are determined by the reduction of oxide species generated on the catalysts during the O passivation. The O associated with Mo2C is mostly reduced at high temperature (step III) according to the Mo2C/ABC catalyst TPR results of Figure 5.2a. On the promoted catalysts, the H2 consumption that occurs at lower temperatures (step I and II) is attributed to O bound either to isolated promoter metal sites or sites associated with M-Mo2C. Although it is not possible to identify the O binding site, the lower reduction temperatures in the presence of the Ni and Pd promoters illustrate an improved capability of O removal from the catalyst surface in a reducing environment. More importantly, the peak shift in step III indicates the decreased adsorption strength of the O species that require lower temperatures for removal on the promoted catalysts. The addition of Cu decreased the required temperature for O reduction compared to the Mo2C/ABC catalyst since 1/3 of the H2 consumption occurred in step II and there is a slight shift towards lower temperature in the reduction peak in step III. The O-reduction that occurs in step I and step II for Ni-Mo2C/ABC and Pd-Mo2C/ABC catalysts indicates that some of the O is more easily removed in the presence of Ni and Pd. Also, the peak shifts to lower temperature in step III on both catalysts are even clearer. The calculated total H2 consumption per mole of metal (Table 5.2) shows that the Cu-Mo2C/ABC and Pd-Mo2C/ABC promoted catalysts had approximately the same amount of O that could be reduced (removed) as  136  the Mo2C/ABC catalyst. However, the Ni-Mo2C/ABC catalyst had less O removed, indicating that Ni incorporation into the Mo2C structure increased the catalyst resistance to oxidation.  Unlike the differences in the CHR-MS profiles, the Ni-, Pd- and Cu-Mo2C/ABC catalysts yield approximately the same amount of Mo2+ according to the XPS results (Table 5.3). The Cu- Mo2C/ABC had low CH4 formation during CHR, yet the Mo2C surface concentration is the same as for the Ni and Pd promoted catalysts. For the Pd-Mo2C/ABC about 75% of the Pd is in the metallic state and a large portion of the CH4 could be generated on the metallic Pd sites instead of sites associated with Mo species during CHR. For the Ni-Mo2C/ABC catalyst, a lower CHR temperature was used compared to the other catalysts to avoid excessive carbon loss and consequently, less Mo2C was formed. Compared to the fresh catalysts, the XPS analysis of the used catalysts (after 24 h reaction) show that Mo2C was converted to Mo oxycarbide after reaction, which like MoO3, is also active in HDO reactions. [126] Note that the used Mo2C/ABC and Cu-Mo2C/ABC catalysts had been reacted at 350 oC prior to the XPS analysis; whereas, the Pd-Mo2C/ABC and Ni-Mo2C/ABC catalysts had been reacted at 230 oC, and the catalysts were relatively stable as reflected in the relatively constant selectivities with time-on-stream for the corresponding reaction conditions. Among these catalysts, the Ni-Mo2C/ABC retained more Mo oxycarbide than Mo oxide, consistent with the H2-TPR results that indicated that Ni was closely associated with the Mo, thereby enhancing reduction of Mo species that may have been oxidized during reaction.   Much like Pd, dissociative adsorption of H2 occurs more readily on metallic Ni than an oxidized surface, as shown by the DFT study of Nobuhara et al. [201] that reported the energy barrier for  137  H2 dissociation on metallic Pd or Ni is almost zero. In other words, metallic Ni and Pd both have strong H2 dissociation adsorption ability, which implies a high hydrogenation ability of the catalyst. Although there is no previous DFT study of Ni and Pd promoted Mo2C, the promoter effect of these two metals on other surfaces such as Mg (0001) have been reported by Pozzo et al.[203]. In their study, Ni and Pd decreased the energy barrier required for H2 dissociative adsorption from 0.87 eV to 0.06 and 0.39 eV, respectively. Ni promotion also decreased the energy barrier for H2 dissociative adsorption on MoS2, as reported by Nelson et al. [204] Similar effects of reducing the barrier energy for H2 dissociation on a Mo2C (101) surface are confirmed here as well. On Mo2C (101) the barrier energy decreased from 0.46 eV to a low value of 0.20 eV due to the Ni and Pd. The decrease is one explanation for lower reduction temperature reported for the Ni-Mo2C/ABC and Pd-Mo2C/ABC following partial oxidation (Figure 5.2), compared to Mo2C/ABC. Isolated metallic Ni or Pd, which exists on the corresponding promoted catalysts, could dissociatively adsorb H2 on the (1 1 1) surface even more easily (barrier energy of 0.1 eV). However, the amount of isolated metallic Ni and Pd is relatively low, so the effect is relatively minor.  Based on the DBF HDO experimental data, Ni and Pd addition changed the properties of the catalyst significantly by strengthening the catalyst hydrogenation activity. H2-TPR results suggest that Ni and Pd interact with Mo sites of the promoted catalysts and DFT calculation confirmed that these interactions could improve the catalyst hydrogenation activity and thereby shift the selectivity of the catalysts from DDO to HYD. Both Ni-Mo2C/ABC and Pd-Mo2C/ABC catalysts had no DDO products (Figure 5.8) at temperatures from 230 to 350 oC. The non-deoxygenated products generated on the Ni-Mo2C/ABC catalyst suggest that most products were  138  hydrogenated prior to deoxygenation. The product distribution of the Ni-Mo2C/ABC and Pd-Mo2C/ABC catalysts are consistent with the DFT results. Even though Ni and Pd could both significantly increase the hydrogenation ability of the catalysts, Pd performed better than Ni at lower temperatures. The DBF was completely converted on the Pd-Mo2C/ABC catalyst, even at 230 oC while the Ni-Mo2C/ABC required temperatures > 290 oC. Pd in metallic form prefers the hydrogenation route in HDO of the furan ring while Ni prefers ring opening. [205] In this study, furan ring opening is much more difficult due to the two attached benzo-rings. Thus, the metallic Pd increased catalyst activity more than metallic Ni. Pd is also capable of removing oxygen from Mo2C at lower temperatures than the Ni, according to the H2 TPR results. Hence, the Ni-Mo2C/ABC catalyst required higher temperature to maintain activity during HDO compared to the Pd-Mo2C/ABC catalyst.  5.5 Conclusions Addition of Ni and Pd significantly reduced the temperature required for CH4 generation during CHR of AHM/ABC precursors. The Ni-Mo2C/ABC and Pd-Mo2C/ABC catalysts fully converted DBF at temperatures as low as 310 and 230 oC, respectively, while without the metal promoter, the Mo2C/ABC catalyst achieved 90% DBF conversion at 350 oC. Ni and Pd decreased the temperature required to remove oxygen from Mo2C, potentially reducing the oxidation of Mo during the HDO reaction. Ni and Pd also enhanced the hydrogenation activity of the Mo2C/ABC catalyst based on measured selectivities. The experimental observations are consistent with DFT calculations, that indicate a preference for horizontal adsorption of DBF with promoter addition to Mo2C. The Gibbs free energy of DBF adsorption was -1.52 eV on the Ni promoted Mo2C surface, 0.3 eV higher than on Mo2C/ABC surface. Most importantly, simulation of H2  139  dissociative adsorption on the promoted Mo2C surface, suggests that addition of Ni and Pd lowers the energy barrier for H2 dissociative adsorption from 0.46 eV to 0.20 eV. Pd and Cu did not incorporate into the Mo2C as readily as Ni. Cu-Mo2C/ABC performed poorly because of the suppressed CH4 generation during CHR. Pd-Mo2C/ABC had better HDO activity because of the existence of reduced Pd. Ni-Mo2C/ABC also had high activity because of the incorporation of Ni into Mo2C, increasing the hydrogenation ability of the catalyst and partially preventing oxidation of the Mo2C during HDO.  140  Chapter 6: Bio-oil upgrading 6.1 Introduction In previous chapters, a series of carbon supported Mo2C catalysts were synthesized and investigated for their potential application in bio-oil upgrading, including esterification and hydrodeoxygenation. These experiments were done using various model compounds so that product analyses was simplified and the link between catalyst properties and catalyst activity and selectivity could be more easily understood. The catalyst activity and selectivity were determined and the impact of the properties of the carbon support were discussed. Finally, addition of Ni to the Mo2C catalyst is shown to improve the activity of the Mo2C catalyst. The results from these studies showed that oxygen adsorption and incorporation into the Mo2C structure during the HDO reaction has a significant impact on catalyst activity and selectivity. In this chapter, the developed catalysts are used in real bio-oil upgrading and compared to Ru/C.   This study assessed a two-step upgrading process that combines esterification and hydrodeoxygenation. Ru/C and Ru/SiO2-Al2O3 were selected as representative noble metal catalysts for the 1st step esterification reaction. A series of Mo2C based catalysts were also tested in this step. Ru/SiO2-Al2O3 had the best performance among all catalysts. All carbide catalysts presented sufficient acidity to generate esters and stabilize the bio-oil. Among them, 2Ni-10Mo2C/ABC performed better because of higher hydrogenation activity. In the 2nd step HDO process, a series of carbide catalysts were tested as well, and Ru/C catalyst was tested for comparison. The Ru/C catalyst achieved 78% HDO with an overall carbon yield of 70%. The best Mo2C catalyst achieved 69% HDO with an overall carbon yield of 67%. The results are comparable with the Pd/C catalyst in a one step HDO reaction. Even though the coke formation  141  is higher on the Mo2C-based catalysts than the noble metal catalysts, the (Ni)Mo2C catalyst is proven to be a promising catalyst that can be utilized in the two-step bio-oil upgrading process. 6.2 Experimental 6.2.1 Catalyst Preparation and Raw Material Ruthenium supported on carbon (Ru/C) was purchased from Sigma-Aldrich with a metal loading of 5 wt%. Ruthenium supported on silica-alumina (Ru/SiO2-Al2O3) was prepared in-house by incipient wetness impregnation with a Ru loading of 5 wt%. For other carbon supported catalysts, raw petcoke and bio-char were used as starting materials to prepare the C support. The petcoke and bio-char supported catalysts were prepared and named using the same method described in Chapter 5. The bio-oil used was directly purchased from Biomass Technology Group (BTG) and the feed for the 1st step esterification reaction was a mixture of the BTG purchased crude bio-oil (80 wt%) and 1-butanol (20 wt%).  6.2.2 Experimental Conditions and Product Analysis The esterification of bio-oil was carried out in a 300 mL stirred, semi-batch reactor (Parr Instruments Company) to assess the catalytic activity of the catalysts. The reactor was operated at 10.3 MPa and 180 oC for 1.5 h with a mixing speed of 700 rpm under a continuous 1000 mL(STP)/min H2 flow. The feed for the esterification was a mixture of crude bio-oil (80 wt%) and butanol (20 wt%). Following reaction, the product was centrifuged at 12000 rpm for 30 min at room temperature to obtain separate liquid and solid products. The solid product was treated in a Soxhlet extractor using dichloromethane (DCM) as solvent to remove soluble contaminants. The recovered oil phase was mixed with an equivalent mass of decalin to obtain the feed used for the 2nd hydrodeoxygenation (HDO) step of the proposed 2-step upgrading process. Using the  142  same 300 mL stirred reactor, but now operated in batch mode, the HDO was conducted with an initial H2 pressure of 6.9 ~ 8.3 MPa at 300 or 350 oC for 4 h with a mixing speed of 700 rpm. The reactor maximum pressure reached 15.2 MPa when the temperature reached 300 or 350 oC. After the HDO reaction, the liquid, gas and solid products were recovered.  Each phase required specific analytical methods to quantify the extent of the HDO.  The liquid samples were analyzed by a Shimadzu (QP-2010-S) GC/MS with a RTX5 30 m × 0.25 mm capillary column. Due to the complexity of the bio-oil, many components were identified by GC-MS and were accounted for quantitatively based simply on signal area percentage. The components were grouped according to their functional groups as alcohols, acids, ketones plus aldehydes, phenols, heterocyclic compounds, sugars, hydrocarbons, esters, and ethers. The ketones and aldehydes were grouped together as they both contain C=O bonds. Components containing 6 carbon rings were classified as phenols and other ringed compounds such as furans were classified as heterocyclic compounds. Chemicals like D-Glucopyranoside were classified as sugars due to an intact ring. The water content of the oil was measured using standard Karl Fischer titration.  The solid phase recovered included the solid catalyst. CHNS content of both the liquid phase and solid phase were determined using an elemental analyzer (Perkin-Elmer 2400 series II CHNS/O) operated at a combustion temperature of 975 oC.   The gas product from the reactor was collected in a gas bag and analysed by gas chromatography (GC) using a Shimadzu GC-14B equipped with two distinct analytical paths. The first utilised an  143  Agilent Technologies Inc. (Agilent) HP-PLOT U column (19095P-UO4, ID 0.530 mm, length 30 m, film 20.00 µm) with FID as detector; the second used an Alltech Carbosphere 80/100 packed column (5682PC, OD 1/8’’, length 6’, 316 stainless steel) with thermal conductivity detector (TCD) as detector.  The yield of separated products was calculated as follows Oil phase yield= moil/ mfeed         (Eq. 6-1) Aqueous phase yield= maqueous/ mfeed        (Eq. 6-2) Carbon yield= (Coil × moil)/ (Cfeed × mfeed)       (Eq. 6-3) O/C ratio= Ooil/ Coil          (Eq. 6-4)  where moil is the mass of oil phase; mfeed is the mass of feed; maqueous is the mass of aqueous phase; Coil is the wt% of C in the oil phase based on the CHNS analysis; Cfeed is the wt% of C in the feed based on the CHNS analysis; Ooil is the wt% of O in oil phase based on the CHNS analysis. The examples of data repeatability for CHNS and water titration are presented in Appendix G2 and G3.  6.2.3 Catalyst Characterization Textural properties of the prepared and used catalysts were measured by N2 physisorption at -196 oC using a Micromeritics ASAP 2020 analyzer. Same as in previous chapters, 2D non-local density functional theory (2D-NLDFT) based on a N2-DFT model was used for calculations assuming the pores have a slit geometry. Further details are provided in Chapter 2.2.3.   144  Pulsed CO uptake measurements were conducted using a Micromeritics AutoChem 2920 with a TCD detector. as described in Chapter 3.2.3. The acidity. of the prepared catalyst was determined by ammonium temperature programed desorption (NH3-TPD), following the methodology described in Chapter 2.  6.3 Results Table 6.1 summarizes the CHNS elemental analysis and water content of the oil phase recovered after esterification of the crude bio-oil with 20 wt% butanol, using various catalysts. The results show that for all catalysts the C content of the product oil increased and the O content decreased following esterification. The water content measured by Karl Fisher titration, decreased significantly from ca. 26 wt% to ca. 15 wt%. Among all the catalysts, the Ru/SiO2-Al2O3 had the lowest H2O yield. The O/C atom ratios of the product oil are compared in Figure 6.1, indicating that after esterification, the O/C ratio decreased to about half that of the feed. The highest O/C ratio was obtained with the Ru/C catalyst at 180 oC; whereas, the lowest occurred for the same catalyst operated at 240 oC.    145   Table 6.1: CHNS/O elemental analysis and water content of oil phase after 1st step esterification reaction at 180 oC and 10.3 MPa using different catalysts.  C H N S O Water     wt%   wt% Feeda 43.1 8.6 0.3 1.0 47.0 20.5 Catalyst       Ru/C 50.7 9.5 0.5 1.2 38.1 16.7 Ru/SiO2-Al2O3 54.8 8.7 0.4 0.8 35.3 13.7 Mo2C/APC 53.9 7.2 0.7 0.7 37.6 14.9 Ru/Cb 55.0 8.6 0.6 0.9 34.9 15.0 1Ni-Mo2C/ABC 52.2 7.5 1.1 0.7 38.5 16.2 1Ni-Mo2C/APC 53.7 7.5 0.2 0.9 37.7 16.7 2Ni-10Mo2C/ABC 51.7 8.6 0.9 1.6 37.2 16.8 a. Feed oil is mixture of 80 wt% of crude bio-oil and 20 wt% of 1-butanol. b. This reaction was conducted at 240 oC.   146   Figure 6.1: O/C ratio of feed and esterified bio-oil after 1st step esterification operated at 180 oC. (* represents the experiment was conducted at 240 oC)  Table 6.2 reports the oxygenated compound product distribution after esterification over the various catalysts. As expected, esters were the majority product, especially butyl acetate. The ester content is a measure of the degree of esterification. At 180 oC, the ester content over the Ru/C was ~44 %, while over the Ru/SiO2-Al2O3 the ester content was ~70%. The ester content over the Ru/C decreased to ~29 % when the reaction temperature increased to 240 oC. For the Mo2C/APC and the Ni-Mo2C/APC catalysts, the ester content was > 60%.  147  Table 6.2: Oxygenate components distribution after 1st step esterification using different catalysts at 180 oC, 10.3 MPa with a mixing speed of 700 rpm and 1.5 h reaction time. Component groups Crude bio-oilb Catalysts Ru/C Ru/SiO2-Al2O3 Mo2C/APC Ru/Ca 1Ni-10Mo2C/APC 2Ni-10Mo2C/ABC Alcohols 3b  2  0  2  25  5  6  Acids 15  0 2  2  0  0  1  Ketones; aldehyde  9  12  8  9 10 6  11  Phenols 29  28  18 22  29  19  27  Heterocyclic compounds 6  5  1  3  2  0  2  Sugars 35  6  0  1  0  2  0  Hydrocarbons 0  1  0  0  0  4  6  Esters 0  44  70  60 29  65  35  Ethers 4  1  1  1  7  0  12  a. This experiment is conducted at 240 oC. b. Water is excluded.   148  Figure 6.2(a) reports the oil phase versus aqueous phase yields after esterification over the various catalysts. There are no significant differences in yields among the catalysts, although the oil yield is marginally highest over the Ru/C catalyst. Figure 6.2(b) reports the carbon yield after esterification, defined as the carbon mass in the product oil compared to the crude bio-oil. The Ru/SiO2-Al2O3 catalyst had the highest yield of 95% and the Ru catalysts all had higher yields than the Mo2C and Ni-Mo2C catalysts. Figure 6.2(c) presents the coke yield from esterification, showing that the Ru/C catalysts had less coke formation than the Mo2C based catalysts. Among all the catalysts, the un-promoted Mo2C/APC catalyst had the highest coke yield.  149   Figure 6.2: Properties analysis of 1st esterification products: (a) Oil vs. aqueous phase; (b) Carbon yield, %; (c) Coke yield. 150  Table 6.3 summarizes the CHNS results and water content of the deoxygenated oil that results from the HDO of the esterified bio-oil. Note that the reported C content excludes the decalin solvent. Clearly, after HDO, the C content of the product oil increased significantly while the O content decreased. Also, by comparing the data of Table 6.1, HDO decreased the water content of the deoxygenated oil, from ~15 wt% (average H2O content of the esterified oil), to about 2 wt%. The water content of the deoxygenated oil was lowest on the Ru/C. Figure 6.3 shows that the O/C ratio of the deoxygenated oil decreased to an average level of approximately 0.25.     Figure 6.3: O/C ratio of oil phase products after 2nd step HDO. (a. Reaction Time=1.5 h; Reaction Temperature= 300 oC; b. Reaction Time=4.0 h; Reaction Temperature= 300 oC; c. Reaction Time=4.0 h, Reaction Temperature= 350 oC; a,b,c all used feed from 1st step esterification by Ru/C catalyst; d. Reaction Time=4.0 h, Reaction Temperature= 350 oC, used feed from 1st step esterification by 2Ni-Mo2C/ABC catalyst ) 151  Table 6.3: CHNS/O elemental analysis and water content of oil phase (decalin excluded) after 2nd-step hydrodeoxygenation (HDO) reaction using different catalysts. Catalyst  Ru/C Ru/C 1Pd-Mo2C/ABC 1Ni-10Mo2C/ABC 1Ni-10Mo2C/ABC 2Ni-10Mo2C/ABC Esterified feed – catalyst and esterification reaction conditions Esterification catalysta Ru/C Ru/C Ru/C Ru/C 2Ni-10Mo2C/ABC 2Ni-10Mo2C/ABC Temperature (oC) 300 350 350 350 350 350 Time (h) 1.5 4 4 4 4 4 HDO product analysis C (wt%) 71.7 76.1 67.4 69.6 70.6 62.2 H (wt%) 9.3 11.9 10.3 12.9 10.8 12.7 N (wt%) 0.9 0.2 0.2 1.4 0.2 1.6 S (wt%) 1.4 1.6 0.7 1.4 1.4 1.0 O (wt%) 16.7 10.2 21.4 14.7 17.0 22.5 H2O (wt%) 2.1 1.1 2.2 1.8 2.0 1.5 a. This row represents catalyst used in 1st step for the corresponding feed source. b. The reported C data exclude the decalin solvent. 152  Table 6.4 summarizes the oxygenated component product distribution after HDO. As reported in Table 6.2, the main components of the 1st step esterification step are esters. However, after the HDO, most of the esters were converted to other components, with hydrocarbons the most abundant among all other products. The data reported in columns of 2 to 5 of Table 6.4 were obtained using a feed oil derived from the esterification of the crude bio-oil using the Ru/C catalyst. The last two columns report results using a feed derived from the esterification of the crude bio-oil using the 2Ni-10Mo2C/ABC catalyst. The phenols present before and after HDO indicate that phenol conversion was minimal during HDO at the chosen reaction conditions.    153   Table 6.4: Oxygenate components distribution after 2nd step HDO by different catalysts. 2nd step HDO Feeda Ru/C Ru/C Pd-Mo2C/ABC 1Ni-10Mo2C/ABC HDO Feedb 2Ni-10Mo2C/ABC 1Ni-10Mo2C/ABC HDO reaction conditions Feed source - a a a a - b b Temperature (oC) - 300 350 350 350 - 350 350 Reaction Time (h) - 1.5 4 4 4 - 4 4 HDO product analysis Alcohols 1.98 10.85 19.47 12.48 8.22 6.15 5.97 2.12 Acids 0 0 0 0.68 0.43 1.47 1.16 0.75 Ketones; Aldehydes  12.25 0.82 9.67 16.68 26.54 10.7 17.94 24.51 Phenols 28.38 25.63 23.22 17.16 21.01 26.59 20.48 24.26 Heterocyclic compounds 5.39 2.07 2.03 0.98 0.54 2.2 1.01 1.03 Sugars 5.88 0 0 0 0 0 0 0 Hydrocarbons 1.29 2.21 28.56 22.9 14.12 6.14 24.32 15.11 Esters 43.91 55.05 17.05 29.13 28.17 34.86 26.68 20.33 Ethers 0.91 3.37 0 0 0.97 11.9 2.44 11.88 a. This feed from 1st step esterification using Ru/C catalyst at 180 oC and 10.3 MPa.  b. This feed from 1st step esterification using 2Ni-Mo2C/ABC catalyst at 180 oC and 10.3 MPa. c. Quantity of products are determined by signal area percentage.   154  Figure 6.4(a) reports the oil and aqueous phase yields after HDO and the data indicate that there were no significant differences among the different catalysts. Overall, the 1Ni-10Mo2C/ABC catalyst had the highest oil yield using an esterified bio-oil feed derived from the 2Ni-10Mo2C/ABC catalyst. Figure 6.4(b) shows the carbon yield from the 2nd step HDO for each catalyst tested. The Ru/C had higher carbon yield (approximately 86%) than the 1Ni-10Mo2C/ABC catalyst when using the same feed (i.e. feed from the 1st step esterification using Ru/C). Also, when using the same esterified bio-oil from the 2Ni-10Mo2C/ABC, 1Ni-10Mo2C/ABC had a higher C yield than that from the 2Ni-10Mo2C/ABC catalyst. Figure 6.4(c) shows the coke yield of each experiment after HDO. Similar to the esterification reaction, Ru catalysts had less coke formation while all Mo2C-based catalysts had more. 2Ni-Mo2C/ABC catalyst had the highest coke yield after HDO.  Figures E.14 and E.15 report the 1H-NMR analysis of the oil product from the esterification and HDO reactions, respectively. 1H-NMR is widely used to identify chemical groups in bio-oils [206-208] as summarized in Table 6.5. The detailed classification of protons based on different chemical shifts is reported in the SI. One of the obvious changes before and after the esterification reaction is that most of the aldehydic compounds were converted during this process on different catalysts. Most importantly, the amount of aliphatic group compounds (Chemical shift: 0.5-1.8 ppm) were significantly increased. After the HDO reaction, the number of protons with shifts of 1.8-3.0 ppm and 3.0-4.2 ppm were significantly reduced, indicating that part of the aromatic groups was hydrogenated, and the removal of O occurred during the HDO reaction.  155   Figure 6.4: Properties analysis of 2nd HDO products: (a) Oil vs. aqueous phase; (b) Carbon yield, %; (c) Coke yield.   156  Table 6.5: Results of 1H-NMR for the bio-oil from 1st step esterification and 2nd step hydrodeoxygenation with different catalysts.                     Catalyst Chemical shift (ppm) 0.5-1.8  1.8-3.0  3.0-4.2  4.2-6.5  6.5-9.0  9.0-10.0   Area, % Crude bio-oil 14.35 32.14 16.36 29.56 6.74 0.86 1st step: Esterification Ru/C (180 oC) 52.63 20.00 12.11 0.53 14.74 0.00 Ru/SiO2-Al2O3 (180 oC) 94.81 2.85 0.97 0.63 0.72 0.02 Mo2C/APC (180 oC) 42.92 36.05 14.16 0.00 6.87 0.00 Ru/C (240 oC) 49.02 29.90 15.20 0.00 5.88 0.00 1Ni-10Mo2C/ABC  (180 oC) 42.74 35.47 14.10 1.71 5.98 0.00 1Ni-10Mo2C/APC (180 oC) 43.86 35.53 14.47 0.88 5.26 0.00 2Ni-10Mo2C/ABC (180 oC) 49.88 36.24 11.76 0.00 1.88 0.24 2nd step: Hydrodeoxygenation Ru/C (300 oC, 1.5 h) 98.04 0.00 0.98 0.00 0.98 0.00 Ru/C (350 oC, 4.0 h) 96.67 1.88 0.63 0.12 0.68 0.02 1Pd-10Mo2C/ABC (350 oC, 4 h) 91.32 5.50 1.26 0.75 1.12 0.05 1Ni-10Mo2C/ABCa (350 oC, 4 h) 87.90 6.48 1.06 1.08 3.46 0.02 1Ni-10Mo2C/ABCb (350 oC, 4 h) 91.73 4.81 1.52 0.15 1.77 0.01 2Ni-10Mo2C/ABC (350 oC, 4 h)  89.38 6.56 1.72 0.10 2.24 0.01 a. Feed was from 1st step esterification by Ru/C catalyst b. Feed was from 1st step esterification by 2Ni-10Mo2C/ABC catalyst  157  6.4 Discussion From the oxygenated component distribution data (Table 6.2, Column 2), it can be seen that crude bio-oil contains many polar hydrophilic compounds such as acids, aldehydes and sugars. The 1H NMR data also show that crude bio-oil contained significant quantities of double bonds such as C=O and C=C. These components contribute to coke formation and catalyst deactivation during HDO [209, 210].  After esterification, most of the water was removed from the bio-oil. Upon addition of 50% of decalin as solvent, the esterified bio-oil was much less likely to deactivate the catalyst by coking or oxidation, even under the severe reaction conditions of HDO.  After the 1st step esterification, the product distribution on different catalysts all showed a decrease in acid and sugar content and increased ester content. The decrease in acid content was because of the esterification reaction between acids and alcohols. Sugars in the bio-oil were also significantly reduced, likely because they undergo hydrolysis reactions and then react with alcohols, as suggested by the model compound study of Li et al. [211]. Any un-reacted sugars remained in the aqueous phase due to their extreme polarity. Although the NMR results indicated a significant decrease in aldehyde content, the content of the chemical group ketones plus aldehydes barely changed before and after esterification. Part of the reason is that sugars were removed from the oil phase, so that the relative number of remaining components increased. The other reason is that ketones do not convert during esterification. For all the catalysts, the major product from the 1st processing step was esters. The amount of ester formation is determined by the acidity of the catalysts. Thus, Ru/SiO2-Al2O3 had the highest yield of esters. Furthermore,  158  higher temperature is not suitable for ester formation because esters are not stable at higher temperature. On the Ru/C catalyst, reaction at 240 oC produced fewer esters than at 180 °C. Mo2C/APC, 1Ni-10Mo2C/APC and 1Ni-10Mo2C/ABC catalysts have more acidity than Ru/C, and they yield more esters. However, the 2Ni-10Mo2C/ABC catalyst yields less esters than Ru/C because some of the ester product was converted into hydrocarbons and ethers. Comparing the NMR results among 1Ni-10Mo2C/APC, 1Ni-10Mo2C/ABC and 2Ni-10Mo2C/ABC catalysts, it is clear that 2Ni-10Mo2C/ABC catalyst had much more H in aliphatic groups, indicating a higher degree of deoxygenation. The results are consistent with the further conversion of esters on 2Ni-10Mo2C/ABC catalyst. Ni as promoter gives Mo2C more hydrogenation ability. For the Ru/C catalyst, even though the hydrocarbons and ethers in the esterified oils were not as high as over the 2Ni-10Mo2C/ABC catalyst, the amount of H in aliphatic groups was higher. The reason is that the hydrogenation ability of Ru/C is higher than the Mo2C-based catalysts and more double bonds in the oil components were saturated.   During the esterification process, part of the O is removed from the bio-oil, as reflected in the decreased O/C ratio. Ru/C at 240 oC had the lowest O/C ratio and at the same conditions, the aqueous phase yield was much higher than Ru/C at 180 oC. Since esters are not stable at the higher reaction temperature 240 oC, some alcohols remain in the esterified oil, but a fraction is removed in the aqueous phase of the product, leading to a lower oil phase yield. Thus, despite the O/C ratio difference, the carbon yields at both temperatures are approximately the same at ~80%. The esterified bio-oil produced using the Ru/SiO2-Al2O3 catalyst, had low O/C ratio and high carbon yield suggesting that this catalyst is a very good candidate for esterification. Compared to the Ru/C catalyst, the Mo2C-based catalysts had comparable O/C ratio in the esterified bio-oil.  159  However, the Mo2C-based catalysts had higher coke yield than the Ru catalysts because the Ru catalysts have higher hydrogenation activity to saturate double bonds and avoid the polymerization/condensation reactions.   After esterification, the partially hydrogenated esterified oil was separated from the aqueous phase that included dissolved sugars. The Mo2C-based catalysts were then assessed for HDO using the esterified product oil generated from the 1st step esterification over the Ru/C and 2Ni-10Mo2C/ABC catalysts. These two esterified oils were selected because they had similar O/C ratios and the NMR results suggested that their hydrogenation degree was also very similar.  The GCMS results from the analysis of the deoxygenated oil products of HDO suggest that the esters in the oil phase were mostly converted to hydrocarbons on all the catalysts. Phenols were the most stable components and did not completely convert at the given reaction conditions. Although partial deoxygenation could help remove part of the oxygen functional groups from phenols, the Ar-OH bond cannot be cleaved on the Mo2C-based catalysts and were only partially converted on the Ru/C catalyst. The NMR results suggest that most of the 1H in the bio-oil components arises from aliphatic groups. Ru/C had the best performance, as inferred from the NMR results that showed the highest quantity of H in aliphatic groups. Among the Mo2C catalysts, 2Ni-10Mo2C/ABC had the highest quantity of aliphatic H. The results are consistent with the degree of deoxygenation inferred from the O/C ratio of the deoxygenated oil products, with the lowest value on Ru/C and second lowest value on 1Ni-10Mo2C/ABC. Compared to 1Pd-10Mo2C/ABC and 2Ni-10Mo2C/ABC catalysts, 1Ni-10Mo2C/ABC catalyst had less H in aliphatic groups. According to the GC-MS results, 1Ni-10Mo2C/ABC had the highest conversion  160  of esters, some of which were converted to ethers and some to hydrocarbons. The hydrogenation ability of the 1Ni-10Mo2C/ABC catalyst was not as good as the other Mo2C catalysts. The 1Pd-10Mo2C/ABC catalyst had much higher hydrogenation ability. For the 2Ni-10Mo2C/ABC catalyst, the promoter effect of Ni on the catalyst is stronger than 1Ni-10Mo2C/ABC. However, the deoxygenation ability of the 1Ni-10Mo2C/ABC catalyst is higher. The deoxygenated oil acquired using 1Ni-10Mo2C/ABC catalyst reached a similar level of deoxygenation as observed with the Ru/C catalyst, with less H2 consumption in saturation of the C=C bonds. Furthermore, as a result of stronger hydrogenation ability, 1Pd-10Mo2C/ABC and 2Ni-10Mo2C/ABC catalysts all have more alcohols in their products from ester conversion compared to 1Ni-10Mo2C/ABC. Products such as alcohols readily dissolve in the aqueous phase, causing a loss in C yield in the oil phase. Hence 1Ni-10Mo2C/ABC has a higher yield of C in its deoxygenated oil product.   The overall process C yield, using Ru/C catalyst for both esterification and HDO, was 70%; whereas, using 2Ni-10Mo2C/ABC in the esterification step plus 1Ni-10Mo2C/ABC in the HDO, the C yield was 66%. The results suggest that the Ni-Mo2C catalyst shows great potential for bio-oil upgrading. The two-step process allows the catalysts to be utilized for HDO without severe deactivation. In terms of coke formation, Ru/C had the best performance and 1Pd-10Mo2C/ABC was the best among the Mo2C catalysts. It can be concluded that saturation of double bonds is still the key factor that avoids coke formation. Table 6.6 summarizes the overall carbon yield and O removal of two representative catalysts and compares them with a Pd/C catalyst used for one step HDO upgrading. The results show that the two-step approach has lower carbon yield because some of the carbon reports to the aqueous phase during the esterification process as water soluble products are formed. However, the Ru/C catalyst has a higher HDO than the one  161  step process. The HDO on the 1Ni-10Mo2C/ABC catalyst was relatively close to the HDO on the Pd/C catalyst. The bio-oil processing ability of the two-step method is much higher than the one step method. For noble metal catalysts, the hydrogenation of the ring usually happens before deoxygenation. In the study of Wang et al. [14], the ring containing model compounds dibenzofuran, cresol and guaiacol all had ring saturation prior to deoxygenation. On the other hand, Mo2C catalyst have notable DDO selectivity. 162  Table 6.6: Summary and comparison of experimental results and literature data for bio-oil upgrading. a. C element content in the oil phase product after excluding diluent decalin. b. C recovery rate after 2nd step treatment after excluding diluent decalin.  c. Overall carbon yield was calculated including both the 1st and 2nd step of the overall process. d. Two-step upgrading method: 2nd step - Feed for Ru/C is from 1st step esterification using Ru/C catalyst; Feed for 1Ni-10Mo2C/ABC is from 1st step esterification by 2Ni-10Mo2C/ABC catalyst  Catalyst HDO oil yield, g/gdry feed C,  wt.% a C yield 2nd step,  % b Overall C yield, % c O removal, % Process capacity Operation condition Ru/Cd 0.62 80.5 86.0 70.1 78 4 h  20 L oil/Lcat 15.5 MPa; 350 oC batch reactor 1Ni-10Mo2C/ABCd 0.60 77.3 86.26 66.7 69 4 h 40 L oil/Lcat 15.5 MPa; 350 oC batch reactor Pd/C [212] 0.62 75.5 N/A 81.1 75 4 h 1 L oil/Lcat 13.8 MPa; 340 oC fix-bed reactor  163  6.5 Conclusions A two-step bio-oil upgrading process was investigated in this study. During the 1st step esterification, the bio-oil was partially hydrogenated and most of the water was separated from the bio-oil with some of the highly polar chemicals. Among all the catalysts tested, Ru/SiO2-Al2O3 showed the best performance for esterification. Among the Mo2C catalysts, 2Ni-10Mo2C/ABC presented good hydrogenation ability to saturate double bonds and reduce potential condensation reactions that lead to polymerized products. The catalyst had sufficient acidity to convert most of the alcohols and acids into esters. In the 2nd step HDO, Ru/C catalyst showed the best performance of all catalysts with higher carbon yield and higher extent of HDO. However, the Ni-10Mo2C/ABC catalyst presented comparable performance in both carbon yield and extent of HDO. In both processes, coke formation is more of a problem with Mo2C-based catalysts than Ru catalysts. Compared to direct HDO of crude bio-oil, the processing capacity of the two-step method is higher. The two-step process has low coke yields and even though coke formation on Ni-Mo2Cwas higher than on Ru, the coke yield was acceptable at <3 wt% in each step.  164  Chapter 7: Conclusions and Recommendations 7.1 Conclusions This thesis has investigated a two-step approach to bio-oil upgrading based on esterification and hydrodeoxygenation over a series of novel Mo2C catalysts supported on activated carbons. The main contributions of the study are in understanding the synthesis of the Mo2C catalysts by carbothermal hydrogen reduction of waste carbons (petcoke and biochar) and the application of the catalysts to bio-oil upgrading. In particular, the use of model reactants to assess catalyst activities and selectivities is complemented by additional measurements of catalyst performance with crude bio-oil.  The results provide the data needed to compare the performance of the Mo2C catalysts to conventional noble metal catalysts at conditions that are relevant to potential commercial operations.  The first part of the study focused on the synthesis of the catalyst and the utilization of the waste carbon (petcoke). A multi-step activation of a Canadian oilsands petroleum coke that yields an acidified mesoporous carbon catalyst, is reported. Microporous activated carbon (APC; ~ 2000 m2/g), obtained by thermochemical activation of petroleum coke using KOH, was impregnated with ammonium heptamolybdate and activated by carbothermal hydrogen reduction (CHR). The resulting Mo2C supported on high mesopore volume (Vmeso ~0.4 cm3/g) carbon, yielded the desired mesoporous carbon catalyst (Vmeso ~0.7 cm3/g) following acid washing. The effect of CHR temperature and the benefit of Mo2C loading on mesopore development is reported and pore development models are discussed. The mesoporous carbons were active for the esterification of acetic acid and 1-butanol at 77 oC and the butanol conversion correlated with the catalyst acidity as measured by NH3-TPD. Acid-T 1Mo2C/APC reduced at 900 oC yielded acidic  165  mesoporous carbon with the highest mesoporosity, acidity and catalyst esterification activity. More importantly, the carbon supported Mo2C itself had sufficient acidity to catalyse the esterification reaction.   Subsequently, the synthesis of Mo2C on activated petcoke was extended to the use of biochar. Petcoke and biochar were chemically activated with KOH at 800 and 700 °C, respectively to acquire microporous activated petcoke (APC) and activated bio-char (ABC) with large surface areas (~ 2000 m2/g). The O content of the petcoke increased while that of the biochar decreased following activation. Nevertheless, ABC had a higher O content on the surface and in the bulk compared to the APC. The APC and ABC precursors were wet impregnated with ammonium heptamolybdate (AHM), dried and reduced in H2 to yield Mo2C/APC and Mo2C/ABC catalysts with a nominal loading of 10 wt% Mo. These catalysts were assessed for the hydrodeoxygenation (HDO) of 4-methylphenol (4-MP). The Mo2C/ABC catalyst had higher activity (per gram Mo) and direct deoxygenation (DDO) selectivity than that of the Mo2C/APC due to the higher dispersion of Mo species and higher O content on the ABC surface. The 10%Mo2C/ABC_R650 catalyst prepared at CHR of 650 °C had the highest activity per gram Mo among all catalysts, whereas turnover frequencies on all catalysts were similar.  The focus of the study then shifted to active sites of the carbon-supported Mo2C catalysts. The hydrodeoxygenation (HDO) of 2-methoxyphenol (or guaiacol, GUA) over Pd, Ru and Mo2C catalysts supported on activated carbon (AC) were compared. The activities of the catalysts for hydrogenation versus deoxygenation on a per site basis, measured over a range of temperatures in a liquid phase batch reactor at high H2 pressure (3.4 MPa), were quantified using lumped  166  kinetics. The overall GUA consumption rate decreased in the order Pd > Ru > Mo2C. Hydrogenation of the phenyl ring of GUA occurred at low temperature (240 C) on both the Pd/AC and Ru/AC catalysts. At higher temperature ( 300 C) the R-OCH3 and R-OH bonds of the hydrogenated products were cleaved yielding cyclohexanol, cyclohexane (Pd and Ru) and benzene (Ru) as major products. On the Mo2C/AC catalyst, HDO of GUA occurred by direct demethoxylation yielding phenol followed by Ar-OH bond cleavage to ultimately yield benzene at high temperature. The lumped kinetics indicate that the hydrogenation activity of the Pd catalyst (on a per site basis, as determined from CO uptake measurements), was about 6 times higher than the Ru, but Ru was more active for O removal. Although the Mo2C was the least active, it was the most efficient in terms of O-removal with minimal H2 consumption.  The application of Mo2C catalyst in HDO was then extended to determine the effect of promoters on catalyst activity. Hence, 1% Cu-, 1%Ni- and 1%Pd-promoted Mo2C catalysts supported on carbon, with 10% Mo loading, were synthesized by carbothermal hydrogen reduction. The product gases from the CHR process suggested that addition of Ni and Pd significantly reduced the temperature required for the generation of CH4. 1% O2 passivation at 100 oC was applied to the promoted Mo2C catalysts and the unpromoted Mo2C catalyst. The study showed that addition of Ni and Pd reduced the temperature required for removal of the oxygen layer formed during passivation, potentially also enhancing the stability of the catalyst during HDO. Cu on the other hand, suppressed the formation of CH4 and the Mo carburization process. These observations are consistent with the HDO data obtained from the HDO of DBF, where the Pd and Ni promoted Mo2C catalyst show good performance and stability and the Cu promoted catalyst showed lower activity than the un-promoted Mo2C catalyst. The 10% Mo2C  167  converted 90% DBF at 350 oC, 4.1 MPa and LHSV=4 h-1, while the Ni and Pd promoted Mo2C catalyst achieved full conversion at 310 oC and 230 oC, respectively. DFT calculations confirmed that addition of Ni and Pd improved the horizontal adsorption of DBF on a Mo2C surface. Dissociative adsorption of H2 on Ni and Pd promoted catalyst required only 0.20 eV to overcome the energy barrier, lower than that on clean Mo2C catalyst, which is 0.28 eV.  Results from the study have shown the importance of O-functional groups in the preparation of the catalysts and their resulting activity and selectivity during esterification and hydrodeoxygenation reactions. O-functional groups on the carbon support contribute to the acidity of the catalyst that impacts the CHR process and the catalyst acidity, which in turn impacts the esterification reaction. O species may also be incorporated into the catalyst structure during HDO, yielding Mo-oxycarbide species that also change the HDO reaction activity and selectivity. Furthermore, the Ni promoter is shown to provide enhanced hydrogenation activity of the catalyst that reduces the impact of O species on the Mo2C during HDO.  Finally, the improved catalysts were applied to crude bio-oil upgrading via the two-step process of esterification followed by hydrodeoxygenation. Ru/C and Ru/SiO2-Al2O3 were selected as representative noble metal catalysts for the 1st step esterification reaction. A series of Mo based carbide catalysts were also tested in this step. Ru/ SiO2-Al2O3 had the best performance among all the catalysts. All carbide catalysts had sufficient acidity to generate esters and stabilize the bio-oil. Among them, the 2Ni-10Mo2C/ABC catalyst performed better because of higher hydrogenation activity. In the 2nd HDO step, a series of carbide catalysts was also tested and compared to the Ru/C catalyst. The Ru/C catalyst achieved 78% HDO with an overall carbon  168  yield of 70% and the best carbide catalyst (Ni-10Mo2C/ABC) achieved 69% HDO with an overall carbon yield of 67%. The results are comparable to the Pd/C catalyst and a one-step HDO. Even though the coke formation is more problematic for carbide catalysts than noble metal catalysts, (Ni)Mo2C catalysts are proven to be promising catalysts that can be utilized in the bio-oil upgrading process.   Although carbon supported Mo2C catalysts prepared by the CHR process have been described in previous studies, the originality of this study is in the synthesis of Mo2C catalysts using waste carbon materials as the carbon support. In addition, the O content of the waste materials is shown to provide additional benefits for both the esterification and hydrodeoxygenation reactions. Ni as a promoter of Mo2C has been reported in studies of HDS. However, this study demonstrated the use of Ni promoted Mo2C catalyst for crude bio-oil upgrading and proved that it is a promising catalyst for this application.  7.2 Recommendations 7.2.1 Effect of Other Transition Metal Promoter on Mo2C Catalysts in both Esterification and HDO Reaction Ni addition has significant beneficial effects on the Mo2C catalyst performance, especially in enhancing hydrogenation activity. Other transition metals are also potential candidates as promoters. For example, Co is a common promoter of conventional hydrotreating catalysts (CoMoS/Al2O3) and although some model compound work has been done [213], the use of a Co promoter for crude bio-oil upgrading remains to be proven. The promoter could potentially improve the hydrogenation ability even more than Ni and convert more ketone/aldehydes to  169  acids and alcohols in the esterification step. Furthermore, it is possible that a promoter could improve the deoxygenation ability of the catalyst to increase the O removal.  7.2.2 Stability Study of Mo2C Catalyst during crude Bio-oil Upgrading Even though a two-step upgrading method and Ni promoter were applied to reduce the tendency for coke formation during crude bio-oil upgrading, deactivation and coke deposition on the Mo2C catalysts remains an issue. The chemical environment in crude bio-oil upgrading is much more complex than that present during the model compound studies. Consequently, catalyst deactivation processes during crude bio-oil upgrading involve not only the effects of O on the catalyst as identified from the model compound studies, but also carbon deposition on the catalyst surface. The stability of the Mo2C catalysts needs to be assessed during HDO of crude bio-oil using a continuous H2 and bio-oil flow reactor system in which the gas and liquid composition could be monitored with time-on-stream. In this way details of the catalyst deactivation rate and mechanism could be understood for crude bio-oil upgrading and this information can be used to inform further catalyst development research to improve the Ni-Mo2C/ABC catalyst.  7.2.3 DFT Calculations to Identify Mo2C Deactivation Mechanism by O During HDO, Mo2C deactivation caused by oxidation of the Mo2C surface also occurred. Although the mechanism of this deactivation has been examined experimentally, theoretical calculation would be helpful to gain further insight into the oxidation. Using DFT calculations through VASP software to model different stages of oxygen coverage of the Mo2C, the evolution of deactivation during HDO would be clarified. Following Mo2C layer optimization, O atoms  170  can be adsorbed onto the top layer and the change in surface energetics calculated. Replacement of surface C atoms by O atoms would also be worth investigating. Also, similar DFT calculations based on Ni promoted Mo2C surface is worth studying.  7.2.4 Solvent Effects during Crude Bio-oil Upgrading The solvent (decalin) used in the thesis for the crude bio-oil studies was set at 50:50 by mass for all the HDO experiments. Further studies of the effect of solvent concentration need to be completed to reduce the amount of solvent in the process. Furthermore, instead of using a specific alcohol namely 1-butanol and a specific hydrocarbon namely decalin, the application of an alcohol mixture and hydrocarbon mixture coming from a cost-effective source such as fermentation of biomass and petroleum thermal cracking, need to be investigated.   7.2.5 Additional Model Compound Studies This study has used several representative model compounds as reactants. For lower temperature reactions, acetic acid and 1-butanol were studied for the esterification reaction. For higher temperature reactions, 4-MP and 2-methoxy phenol were studied to represent phenolic groups. DBF was studied to represent larger molecules in bio-oil. However, furans as a major group of chemicals present in bio-oil, was not examined. Furfural as a representative chemical for furans is widely studied [214] and would be an interesting candidate to determine the HDO of furans using the catalyst developed in this study. The selective deoxygenation of furfural [215] could help to understand how to remove as much oxygen as possible without converting chemicals into gas products for a higher overall carbon efficiency in the oil.   171  In addition, the use of a mixture of model compounds (rather than the single component systems used herein) would be worth analysing. Furthermore, addition of compounds such as dienes to the 4-methyl phenol model compound would provide information regarding possible oligomerization reactions that lead to gums and coke during HDO reaction.  172  Bibliography  1. U.S.E.I. Administration. International Energy Statistics. https://www.eia.gov/ 2017. 2. N.L. Panwar, S.C. Kaushik, and S. Kothari, Role of renewable energy sources in environmental protection: A review. Renewable and Sustainable Energy Reviews, 2011. 15(3): p. 1513-1524. 3. 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. 4. A. Bridgwater, D. Meier, and D. Radlein, An overview of fast pyrolysis of biomass. Organic Geochemistry, 1999. 30(12): p. 1479-1493. 5. M. Stöcker, Biofuels and biomass‐to‐liquid fuels in the biorefinery: Catalytic conversion of lignocellulosic biomass using porous materials. Angewandte Chemie International Edition, 2008. 47(48): p. 9200-9211. 6. S. Xiu and A. Shahbazi, Bio-oil production and upgrading research: A review. Renewable and Sustainable Energy Reviews, 2012. 16(7): p. 4406-4414. 7. G. Kumar, A.K. Panda, and R. Singh, Optimization of process for the production of bio-oil from eucalyptus wood. Journal of Fuel Chemistry and Technology, 2010. 38(2): p. 162-167. 8. Q. Zhang, J. Chang, T. Wang, and Y. Xu, Review of biomass pyrolysis oil properties and upgrading research. Energy Conversion and Management, 2007. 48(1): p. 87-92. 9. L. Zhang, R. Liu, R. Yin, and Y. Mei, Upgrading of bio-oil from biomass fast pyrolysis in China: A review. Renewable and Sustainable Energy Reviews, 2013. 24: p. 66-72. 10. X. Junming, J. Jianchun, S. Yunjuan, and L. Yanju, Bio-oil upgrading by means of ethyl ester production in reactive distillation to remove water and to improve storage and fuel characteristics. Biomass and Bioenergy, 2008. 32(11): p. 1056-1061. 11. S. Xiu, A. Shahbazi, L. Wang, and C.W. Wallace, Supercritical ethanol liquefaction of swine manure for bio-oils production. American Journal of Engineering Applied Sciences, 2010. 3(2): p. 494-500. 12. M. Garcia-Perez, J. Shen, X.S. Wang, and C.-Z. Li, Production and fuel properties of fast pyrolysis oil/bio-diesel blends. Fuel Processing Technology, 2010. 91(3): p. 296-305.  173  13. W.-M. Xiong, M.-Z. Zhu, L. Deng, Y. Fu, and Q.-X. Guo, Esterification of organic acid in bio-oil using acidic ionic liquid catalysts. Energy & Fuels, 2009. 23(4): p. 2278-2283. 14. Y. Wang, T. He, K. Liu, J. Wu, and Y. Fang, From biomass to advanced bio-fuel by catalytic pyrolysis/hydro-processing: Hydrodeoxygenation of bio-oil derived from biomass catalytic pyrolysis. Bioresource Technology, 2012. 108: p. 280-284. 15. E. Furimsky, Catalytic hydrodeoxygenation. Applied Catalysis A: General, 2000. 199(2): p. 147-190. 16. X. Xiwei, J. Enchen, M. Wang, and L. Bosong, Rich hydrogen production from crude gas secondary catalytic cracking over Fe/γ-Al2O3. Renewable Energy, 2012. 39(1): p. 126-131. 17. D. Wang, S. Czernik, D. Montane, M. Mann, and E. Chornet, Biomass to hydrogen via fast pyrolysis and catalytic steam reforming of the pyrolysis oil or its fractions. Industrial & Engineering Chemistry Research, 1997. 36(5): p. 1507-1518. 18. Y. Tang, W. Yu, L. Mo, H. Lou, and X. Zheng, One-step hydrogenation− esterification of aldehyde and acid to ester over bifunctional Pt catalysts: A model reaction as novel route for catalytic upgrading of fast pyrolysis bio-oil. Energy & Fuels, 2008. 22(5): p. 3484-3488. 19. J.-J. Wang, J. Chang, and J. Fan, Upgrading of bio-oil by catalytic esterification and determination of acid number for evaluating esterification degree. Energy & Fuels, 2010. 24(5): p. 3251-3255. 20. X. Li, R. Gunawan, C. Lievens, Y. Wang, D. Mourant, S. Wang, H. Wu, M. Garcia-Perez, and C.-Z. Li, Simultaneous catalytic esterification of carboxylic acids and acetalisation of aldehydes in a fast pyrolysis bio-oil from mallee biomass. Fuel, 2011. 90(7): p. 2530-2537. 21. Z. Zhang, Q. Wang, P. Tripathi, and C.U. Pittman Jr, Catalytic upgrading of bio-oil using 1-octene and 1-butanol over sulfonic acid resin catalysts. Green Chemistry, 2011. 13(4): p. 940. 22. I. Hita, R. Palos, J.M. Arandes, J.M. Hill, and P. Castaño, Petcoke-derived functionalized activated carbon as support in a bifunctional catalyst for tire oil hydroprocessing. Fuel Processing Technology, 2016. 144: p. 239-247.  174  23. D. Zeng, S. Liu, W. Gong, G. Wang, J. Qiu, and Y. Tian, Acid properties of solid acid from petroleum coke by chemical activation and sulfonation. Catalysis Communications, 2013. 40: p. 5-8. 24. M. Molina-Sabio and F. Rodrıguez-Reinoso, Role of chemical activation in the development of carbon porosity. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2004. 241(1): p. 15-25. 25. A. Ahmadpour and D. Do, The preparation of active carbons from coal by chemical and physical activation. Carbon, 1996. 34(4): p. 471-479. 26. F. Rodríguez-Reinoso, The role of carbon materials in heterogeneous catalysis. Carbon, 1998. 36(3): p. 159-175. 27. E. Auer, A. Freund, J. Pietsch, and T. Tacke, Carbons as supports for industrial precious metal catalysts. Applied Catalysis A: General, 1998. 173(2): p. 259-271  28. G. Goncalves, P.A.A.P. Marques, C.M. Granadeiro, H.I.S. Nogueira, M.K. Singh, and J. Gracio, Surface modification of graphene nanosheets with gold nanoparticles: the role of oxygen moieties at graphene surface on gold nucleation and growth. Chemistry of Materials, 2009. 21(20): p. 4796-4802  29. X. Huang, F. Zhang, M. Fan, and Y. Wang, Catalytic coal gasification. Sustainable Catalytic Processes, 2015: p. 179. 30. D. Kawashima, T. Aihara, Y. Kobayashi, T. Kyotani, and A. Tomita, Preparation of mesoporous carbon from organic polymer/silica nanocomposite. Chemistry of Materials, 2000. 12(11): p. 3397-3401. 31. W.J. Lee and S.D. Kim, Catalytic activity of alkali and transition metal salt mixtures for steam-char gasification. Fuel, 1995. 74(9): p. 1387-1393. 32. 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): p. 1992-1997. 33. A. Tomita, K. Higashiyama, and Y. Tamai, Scanning electron microscopic study on the catalytic gasification of coal. Fuel, 1981. 60(2): p. 103-114  34. H. Tamai, T. Kakii, Y. Hirota, T. Kumamoto, and H. Yasuda, Synthesis of extremely large mesoporous activated carbon and its unique adsorption for giant molecules. Chemistry of Materials, 1996. 8(2): p. 454-462.  175  35. Z. Hu and M. Srinivasan, Mesoporous high-surface-area activated carbon. Microporous and Mesoporous Materials, 2001. 43(3): p. 267-275. 36. 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. 37. 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. 38. S. Kang, J. Ye, and J. Chang, Recent advances in carbon-based sulfonated catalyst: preparation and application. International Review of Chemical Engineering, 2013. 5(2): p. 133-144. 39. M. Kitano, K. Arai, A. Kodama, T. Kousaka, K. Nakajima, S. Hayashi, and M. Hara, Preparation of a sulfonated porous carbon catalyst with high specific surface area. Catalysis Letters, 2009. 131(1-2): p. 242-249. 40. H.-Z. Geng, K.K. Kim, K.P. So, Y.S. Lee, Y. Chang, and Y.H. Lee, Effect of acid treatment on carbon nanotube-based flexible transparent conducting films. Journal of the American Chemical Society, 2007. 129(25): p. 7758-7759. 41. Y. Otake and R.G. Jenkins, Characterization of oxygen-containing surface complexes created on a microporous carbon by air and nitric acid treatment. Carbon, 1993. 31(1): p. 109-121. 42. A. Afkhami, T. Madrakian, and Z. Karimi, The effect of acid treatment of carbon cloth on the adsorption of nitrite and nitrate ions. Journal of Hazardous Materials, 2007. 144(1): p. 427-431. 43. M. Trépanier, A. Tavasoli, A.K. Dalai, and N. Abatzoglou, Fischer–Tropsch synthesis over carbon nanotubes supported cobalt catalysts in a fixed bed reactor: Influence of acid treatment. Fuel Processing Technology, 2009. 90(3): p. 367-374. 44. T. Liu, Z. Li, W. Li, C. Shi, and Y. Wang, Preparation and characterization of biomass carbon-based solid acid catalyst for the esterification of oleic acid with methanol. Bioresource Technology, 2013. 133: p. 618-621.  176  45. Q. Shu, J. Gao, Z. Nawaz, Y. Liao, D. Wang, and J. Wang, Synthesis of biodiesel from waste vegetable oil with large amounts of free fatty acids using a carbon-based solid acid catalyst. Applied Energy, 2010. 87(8): p. 2589-2596. 46. A.H. Zacher, M.V. Olarte, D.M. Santosa, D.C. Elliott, and S.B. Jones, A review and perspective of recent bio-oil hydrotreating research. Green Chemistry, 2014. 16(2): p. 491-515. 47. S. Boullosa-Eiras, R. Lødeng, H. Bergem, M. Stöcker, L. Hannevold, and E.A. Blekkan, Catalytic hydrodeoxygenation (HDO) of phenol over supported molybdenum carbide, nitride, phosphide and oxide catalysts. Catalysis Today, 2014. 223: p. 44-53. 48. S.K. Kim, D. Yoon, S.-C. Lee, and J. Kim, Mo2C/Graphene Nanocomposite As a Hydrodeoxygenation Catalyst for the Production of Diesel Range Hydrocarbons. ACS Catalysis, 2015. 5(6): p. 3292-3303  49. I. Gandarias, V.L. Barrio, J. Requies, P.L. Arias, J.F. Cambra, and M.B. Güemez, From biomass to fuels: Hydrotreating of oxygenated compounds. International Journal of Hydrogen Energy, 2008. 33(13): p. 3485-3488  50. J. Barbier and D. Duprez, Steam effects in three-way catalysis. Applied Catalysis B: Environmental, 1994. 4(2-3): p. 105-140  51. P. Grange, E. Laurent, R. Maggi, A. Centeno, and B. Delmon, Hydrotreatment of pyrolysis oils from biomass: reactivity of the various categories of oxygenated compounds and preliminary techno-economical study. Catalysis Today, 1996. 29(1): p. 297-301. 52. W. Zhang, Y. Zhang, L. Zhao, and W. Wei, Catalytic activities of NiMo carbide supported on SiO2 for the hydrodeoxygenation of ethyl benzoate, acetone, and acetaldehyde. Energy & Fuels, 2010. 24(3): p. 2052-2059. 53. J.S. Lee, M.H. Yeom, K.Y. Park, I.-S. Nam, J.S. Chung, Y.G. Kim, and S.H. Moon, Preparation and benzene hydrogenation activity of supported molybdenum carbide catalysts. Journal of Catalysis, 1991. 128(1): p. 126-136. 54. M.-L. Frauwallner, F. López-Linares, J. Lara-Romero, C.E. Scott, V. Ali, E. Hernández, and P. Pereira-Almao, Toluene hydrogenation at low temperature using a molybdenum carbide catalyst. Applied Catalysis A: General, 2011. 394(1): p. 62-70.  177  55. J.W. Thybaut, M. Saeys, and G.B. Marin, Hydrogenation kinetics of toluene on Pt/ZSM-22. Chemical Engineering Journal, 2002. 90(1-2): p. 117-129. 56. D.C. Elliott and T.R. Hart, Catalytic hydroprocessing of chemical models for bio-oil. Energy & Fuels, 2008. 23(2): p. 631-637. 57. J. Han, J. Duan, P. Chen, H. Lou, X. Zheng, and H. Hong, Nanostructured molybdenum carbides supported on carbon nanotubes as efficient catalysts for one-step hydrodeoxygenation and isomerization of vegetable oils. Green Chemistry, 2011. 13(9): p. 2561-2568. 58. B. Dhandapani, T.S. Clair, and S. Oyama, Simultaneous hydrodesulfurization, hydrodeoxygenation, and hydrogenation with molybdenum carbide. Applied Catalysis A: General, 1998. 168(2): p. 219-228. 59. A.L. Jongerius, R.W. Gosselink, J. Dijkstra, J.H. Bitter, P.C. Bruijnincx, and B.M. Weckhuysen, Carbon nanofiber supported transition‐metal carbide catalysts for the hydrodeoxygenation of guaiacol. ChemCatChem, 2013. 5(10): p. 2964-2972. 60. N. Li and G.W. Huber, Aqueous-phase hydrodeoxygenation of sorbitol with Pt/SiO2–Al2O3: Identification of reaction intermediates. Journal of Catalysis, 2010. 270(1): p. 48-59. 61. X. Zhang, Q. Zhang, T. Wang, L. Ma, Y. Yu, and L. Chen, Hydrodeoxygenation of lignin-derived phenolic compounds to hydrocarbons over Ni/SiO2–ZrO2 catalysts. Bioresource Technology, 2013. 134: p. 73-80. 62. R.C. Nelson, B. Baek, P. Ruiz, B. Goundie, A. Brooks, M.C. Wheeler, B.G. Frederick, L.C. Grabow, and R.N. Austin, Experimental and theoretical insights into the hydrogen-efficient direct hydrodeoxygenation mechanism of phenol over Ru/TiO2. ACS Catalysis, 2015. 5(11): p. 6509-6523. 63. C. Zhao and J.A. Lercher, Selective hydrodeoxygenation of lignin‐derived phenolic monomers and dimers to cycloalkanes on Pd/C and HZSM‐5 catalysts. ChemCatChem, 2012. 4(1): p. 64-68. 64. S. Boonyasuwat, T. Omotoso, D.E. Resasco, and S.P. Crossley, Conversion of guaiacol over supported Ru catalysts. Catalysis Letters, 2013. 143(8): p. 783-791. 65. M.V. Bykova, D.Y. Ermakov, V.V. Kaichev, O.A. Bulavchenko, A.A. Saraev, M.Y. Lebedev, and V.А. Yakovlev, Ni-based sol–gel catalysts as promising systems for crude  178  bio-oil upgrading: Guaiacol hydrodeoxygenation study. Applied Catalysis B: Environmental, 2012. 113-114: p. 296-307. 66. T.M. Sankaranarayanan, A. Berenguer, C. Ochoa-Hernández, I. Moreno, P. Jana, J.M. Coronado, D.P. Serrano, and P. Pizarro, Hydrodeoxygenation of anisole as bio-oil model compound over supported Ni and Co catalysts: Effect of metal and support properties. Catalysis Today, 2015. 243: p. 163-172. 67. C. Zhao, J. He, A.A. Lemonidou, X. Li, and J.A. Lercher, Aqueous-phase hydrodeoxygenation of bio-derived phenols to cycloalkanes. Journal of Catalysis, 2011. 280(1): p. 8-16. 68. C. Liu, Y. Zhang, and X. Huang, Study of guaiacol pyrolysis mechanism based on density function theory. Fuel Processing Technology, 2014. 123: p. 159-165. 69. K. Lee, G.H. Gu, C.A. Mullen, A.A. Boateng, and D.G. Vlachos, Guaiacol hydrodeoxygenation mechanism on Pt (111): Insights from density functional theory and linear free energy relations. ChemSusChem, 2015. 8(2): p. 315-322. 70. C.-c. Chiu, A. Genest, A. Borgna, and N. Rösch, Hydrodeoxygenation of Guaiacol over Ru (0001): A DFT Study. ACS Catalysis, 2014. 4(11): p. 4178-4188  71. J. Lu and A. Heyden, Theoretical investigation of the reaction mechanism of the hydrodeoxygenation of guaiacol over a Ru(0001) model surface. Journal of Catalysis, 2015. 321: p. 39-50. 72. D. Gao, Y. Xiao, and A. Varma, Guaiacol hydrodeoxygenation over platinum catalyst: Reaction pathways and kinetics. Industrial & Engineering Chemistry Research, 2015. 54(43): p. 10638-10644. 73. J. Sun, A.M. Karim, H. Zhang, L. Kovarik, X.S. Li, A.J. Hensley, J.-S. McEwen, and Y. Wang, Carbon-supported bimetallic Pd–Fe catalysts for vapor-phase hydrodeoxygenation of guaiacol. Journal of Catalysis, 2013. 306: p. 47-57. 74. A. Gutierrez, R.K. Kaila, M.L. Honkela, R. Slioor, and A.O.I. Krause, Hydrodeoxygenation of guaiacol on noble metal catalysts. Catalysis Today, 2009. 147(3-4): p. 239-246. 75. Y.-C. Lin, C.-L. Li, H.-P. Wan, H.-T. Lee, and C.-F. Liu, Catalytic Hydrodeoxygenation of Guaiacol on Rh-Based and Sulfided CoMo and NiMo Catalysts. Energy & Fuels, 2011. 25(3): p. 890-896.  179  76. C.-J. Chen, W.-S. Lee, and A. Bhan, Mo2C catalyzed vapor phase hydrodeoxygenation of lignin-derived phenolic compound mixtures to aromatics under ambient pressure. Applied Catalysis A: General, 2016. 510: p. 42-48. 77. J. Chang, T. Danuthai, S. Dewiyanti, C. Wang, and A. Borgna, Hydrodeoxygenation of guaiacol over carbon-supported metal catalysts. ChemCatChem, 2013. 5(10): p. 3041-3049. 78. R. Azargohar and A. Dalai, Steam and KOH activation of biochar: Experimental and modeling studies. Microporous and Mesoporous Materials, 2008. 110(2): p. 413-421. 79. L. Stockman, D. Turnbull, and S. Kretzmann, Petroleum coke: The coal hiding in the tar sands. Oil Change International: January, 2013. 80. J.G. Speight and B.X. Ozum, Petroleum refining processes2001: CRC Press. 81. E. Furimsky, Gasification of oil sand coke: Review. Fuel Processing Technology, 1998. 56(3): p. 263-290  82. A. Andrews and R.K. Lattanzio, Petroleum coke: industry and environmental issues. Congressional Research Service, 2014. 83. N. Borchard, A. Wolf, V. Laabs, R. Aeckersberg, H. Scherer, A. Moeller, and W. Amelung, Physical activation of biochar and its meaning for soil fertility and nutrient leaching–a greenhouse experiment. Soil Use and Management, 2012. 28(2): p. 177-184. 84. L.E. Hernandez-Mena, A.A. Pécoraa, and A.L. Beraldob, Slow pyrolysis of bamboo biomass: Analysis of biochar properties. Chemical Engineering, 2014. 37. 85. C.E. Brewer, K. Schmidt‐Rohr, J.A. Satrio, and R.C. Brown, Characterization of biochar from fast pyrolysis and gasification systems. Environmental Progress & Sustainable Energy, 2009. 28(3): p. 386-396. 86. Z.B. Shen, M. Ke, P. Yu, S.D. Liu, Z.Z. Song, and Q.Z. Jiang. Study on Mo-Ni/Al2O3 Catalyst for Conversion of Mercaptans in the FCC Gasoline. 2012. Trans Tech Publications. 87. 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  88. R. Azargohar and A. Dalai. Biochar as a precursor of activated carbon. in Twenty-Seventh Symposium on Biotechnology for Fuels and Chemicals. 2006. Springer.  180  89. B.S. Girgis, S.S. Yunis, and A.M. Soliman, Characteristics of activated carbon from peanut hulls in relation to conditions of preparation. Materials Letters, 2002. 57(1): p. 164-172. 90. H. Zhang, J. Chen, and S. Guo, Preparation of natural gas adsorbents from high-sulfur petroleum coke. Fuel, 2008. 87(3): p. 304-311. 91. S.H. Lee and C.S. Choi, Chemical activation of high sulfur petroleum cokes by alkali metal compounds. Fuel Processing Technology, 2000. 64(1): p. 141-153. 92. 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  93. 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. 94. G. Dolce, P. Savage, and L. Thompson, Hydrotreatment activities of supported molybdenum nitrides and carbides. Energy & Fuels, 1997. 11(3): p. 668-675. 95. 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. 96. E. Ochoa, D. Torres, R. Moreira, J. Pinilla, and I. Suelves, Carbon nanofiber supported Mo2C catalysts for hydrodeoxygenation of guaiacol: The importance of the carburization process. Applied Catalysis B: Environmental, 2018. 239: p. 463-474. 97. H. Guo, Y. Song, P. Chen, H. Lou, and X. Zheng, Effects of graphitization of carbon nanospheres on hydrodeoxygenation activity of molybdenum carbide. Catalysis Science & Technology, 2018. 8(16): p. 4199-4208. 98. 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. 99. J.-S. Choi, J.-M. Krafft, A. Krzton, and G. Djéga-Mariadassou, Study of residual oxygen species over molybdenum carbide prepared during in situ DRIFTS experiments. Catalysis Letters, 2002. 81(3): p. 175-180   181  100. 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. 101. W.-S. Lee, A. Kumar, Z. Wang, and A. Bhan, Chemical Titration and Transient Kinetic Studies of Site Requirements in Mo2C-Catalyzed Vapor Phase Anisole Hydrodeoxygenation. ACS Catalysis, 2015. 5(7): p. 4104-4114. 102. 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. 103. M.M. Sullivan and A. Bhan, Acid site densities and reactivity of oxygen-modified transition metal carbide catalysts. Journal of Catalysis, 2016. 344: p. 53-58  104. M.M. Sullivan and A. Bhan, Acetone Hydrodeoxygenation over Bifunctional Metallic–Acidic Molybdenum Carbide Catalysts. ACS Catalysis, 2016. 6(2): p. 1145-1152. 105. P.M. Mortensen, H.W.P. de Carvalho, J.-D. Grunwaldt, P.A. Jensen, and A.D. Jensen, Activity and stability of Mo2C/ZrO2 as catalyst for hydrodeoxygenation of mixtures of phenol and 1-octanol. Journal of Catalysis, 2015. 328: p. 208-215. 106. P. Liu and J.A. Rodriguez, Water-gas-shift reaction on molybdenum carbide surfaces: essential role of the oxycarbide. The Journal of Physical Chemistry B, 2006. 110(39): p. 19418-19425  107. M. Nagai and K. Matsuda, Low-temperature water–gas shift reaction over cobalt–molybdenum carbide catalyst. Journal of Catalysis, 2006. 238(2): p. 489-496. 108. E. Furimsky, Metal carbides and nitrides as potential catalysts for hydroprocessing. Applied Catalysis A: General, 2003. 240(1): p. 1-28  109. 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. 110. C. Shi, A. Zhang, X. Li, S. Zhang, A. Zhu, Y. Ma, and C. Au, Ni-modified Mo2C catalysts for methane dry reforming. Applied Catalysis A: General, 2012. 431-432: p. 164-170. 111. A. Smirnov, Z. Geng, S. Khromova, S. Zavarukhin, O. Bulavchenko, A. Saraev, V. Kaichev, D.Y. Ermakov, and V. Yakovlev, Nickel molybdenum carbides: Synthesis,  182  characterization, and catalytic activity in hydrodeoxygenation of anisole and ethyl caprate. Journal of Catalysis, 2017. 354: p. 61-77. 112. R. Prins, V.H.J. De Beer, and G.A. Somorjai, Structure and function of the catalyst and the promoter in Co-Mo hydrodesulfurization catalysts. Catalysis Reviews-Science and Engineering, 1989. 31(1-2): p. 1-41. 113. M. Egorova and R. Prins, The role of Ni and Co promoters in the simultaneous HDS of dibenzothiophene and HDN of amines over Mo/γ-Al2O3 catalysts. Journal of Catalysis, 2006. 241(1): p. 162-172. 114. 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(1-2): p. 85-92. 115. 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. 116. 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. 117. 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. 118. G.F. Leal, S.F. Moya, D.M. Meira, D.H. Barrett, E. Teixeira-Neto, A.A.S. Curvelo, V. Teixeira da Silva, and C.B. Rodella, Promotion effects of Pd on tungsten carbide catalysts: physiochemical properties and cellulose conversion performance. RSC Advances, 2016. 6(90): p. 87756-87766. 119. Y. Ma, G. Guan, X. Hao, Z. Zuo, W. Huang, P. Phanthong, K. Kusakabe, and A. Abudula, Highly-efficient steam reforming of methanol over copper modified molybdenum carbide. RSC Advances, 2014. 4(83): p. 44175-44184. 120. H.C. Woo, K.Y. Park, Y.G. Kim, I.-S. Namau, Jong ShikChung, and J.S. Lee, Mixed alcohol synthesis from carbon monoxide and dihydrogen over potassium-promoted molybdenum carbide catalysts. Applied Catalysis, 1991. 75(1): p. 267-280.  183  121. 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. 122. S.C. Liew and J.M. Hill, Impacts of vanadium and coke deposits on the CO2 gasification of nickel catalysts supported on activated carbon from petroleum coke. Applied Catalysis A: General, 2015. 504: p. 420-428. 123. 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. 124. T. Peters, N. Benes, A. Holmen, and J. Keurentjes, Comparison of commercial solid acid catalysts for the esterification of acetic acid with butanol. Applied Catalysis A: General, 2006. 297(2): p. 182-188. 125. J. Sepulveda, J. Yori, and C. Vera, Repeated use of supported H3PW12O40 catalysts in the liquid phase esterification of acetic acid with butanol. Applied Catalysis A: General, 2005. 288(1-2): p. 18-24. 126. 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. 127. T. Wigmans, Industrial aspects of production and use of activated carbons. Carbon, 1989. 27(1): p. 13-22. 128. C. Keep, S. Terry, and M. Wells, Studies of the nickel-catalyzed hydrogenation of graphite. Journal of Catalysis, 1980. 66(2): p. 451-462. 129. R. Baker, R. Sherwood, and J. Dumesic, Catalytic hydrogenation of graphite by platinum, iridium, and platinum-iridium. Journal of Catalysis, 1980. 66(1): p. 56-64. 130. W. Xia, Y. Wang, R. Bergsträßer, S. Kundu, and M. Muhler, Surface characterization of oxygen-functionalized multi-walled carbon nanotubes by high-resolution X-ray photoelectron spectroscopy and temperature-programmed desorption. Applied Surface Science, 2007. 254(1): p. 247-250.  184  131. A.M. Dimiev, L.B. Alemany, and J.M. Tour, Graphene oxide. origin of acidity, its instability in water, and a new dynamic structural model. ACS Nano, 2013. 7(1): p. 576-588. 132. T.A. Peters, N.E. Benes, A. Holmen, and J.T.F. Keurentjes, Comparison of commercial solid acid catalysts for the esterification of acetic acid with butanol. Applied Catalysis A: General, 2006. 297(2): p. 182-188. 133. S. Liu, H. Wang, K.J. Smith, and C.S. Kim, Hydrodeoxygenation of 2-methoxyphenol over Ru, Pd, and Mo2C catalysts supported on carbon. Energy & Fuels, 2017. 31(6): p. 6378-6388. 134. 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. 135. E. Santillan-Jimenez, M. Perdu, R. Pace, T. Morgan, and M. Crocker, Activated carbon, carbon nanofiber and carbon nanotube supported molybdenum carbide catalysts for the hydrodeoxygenation of guaiacol. Catalysts, 2015. 5(1): p. 424-441. 136. B. Zhang, L. Wang, R. Li, Q.M. Rahman, and A. Shahbazi, Catalytic conversion of chlamydomonas to hydrocarbons via the ethanol-assisted liquefaction and hydrotreating processes. Energy & Fuels, 2017. 31(11): p. 12223-12231. 137. B.N. Murthy, A.N. Sawarkar, N.A. Deshmukh, T. Mathew, and J.B. Joshi, Petroleum coke gasification: A review. The Canadian Journal of Chemical Engineering, 2014. 92(3): p. 441-468. 138. D.A. Laird, R.C. Brown, J.E. Amonette, and J. Lehmann, Review of the pyrolysis platform for coproducing bio‐oil and biochar. Biofuels, Bioproducts and Biorefining, 2009. 3(5): p. 547-562. 139. E.W. Bruun, P. Ambus, H. Egsgaard, and H. Hauggaard-Nielsen, Effects of slow and fast pyrolysis biochar on soil C and N turnover dynamics. Soil Biology and Biochemistry, 2012. 46: p. 73-79. 140. D. Mohan, A. Sarswat, Y.S. Ok, and C.U. Pittman, Organic and inorganic contaminants removal from water with biochar, a renewable, low cost and sustainable adsorbent–a critical review. Bioresource Technology, 2014. 160: p. 191-202.  185  141. Z. Jiyi, P. Lijun, and L. Gen, Preparation of biochar adsorbent from straw and its adsorption capability. Transactions of the Chinese Society of Agricultural Engineering, 2011. 2011(14). 142. Y. Ma, W.-J. Liu, N. Zhang, Y.-S. Li, H. Jiang, and G.-P. Sheng, Polyethylenimine modified biochar adsorbent for hexavalent chromium removal from the aqueous solution. Bioresource Technology, 2014. 169: p. 403-408. 143. M. Ahmad, A.U. Rajapaksha, J.E. Lim, M. Zhang, N. Bolan, D. Mohan, M. Vithanage, S.S. Lee, and Y.S. Ok, Biochar as a sorbent for contaminant management in soil and water: A review. Chemosphere, 2014. 99: p. 19-33. 144. C. Moreno-Castilla, Adsorption of organic molecules from aqueous solutions on carbon materials. Carbon, 2004. 42(1): p. 83-94. 145. S. Dong Jin, P. Tae-Jin, and I. Son-Ki, Effect of surface oxygen groups of carbon supports on the characteristics of Pd/C catalysts. Carbon, 1993. 31(3): p. 427-435. 146. T. Otowa, Y. Nojima, and T. Miyazaki, Development of KOH activated high surface area carbon and its application to drinking water purification. Carbon, 1997. 35(9): p. 1315-1319. 147. T. Tay, S. Ucar, and S. Karagöz, Preparation and characterization of activated carbon from waste biomass. Journal of Hazardous Materials, 2009. 165(1): p. 481-485. 148. W.T. Tsai, C.Y. Chang, S.Y. Wang, C.F. Chang, S.F. Chien, and H.F. Sun, Cleaner production of carbon adsorbents by utilizing agricultural waste corn cob. Resources, Conservation and Recycling, 2001. 32(1): p. 43-53. 149. M.S. Dresselhaus, A. Jorio, M. Hofmann, G. Dresselhaus, and R. Saito, Perspectives on carbon nanotubes and graphene Raman spectroscopy. Nano Letters, 2010. 10(3): p. 751-758  150. A.C. Ferrari and J. Robertson, Resonant Raman spectroscopy of disordered, amorphous, and diamondlike carbon. Physical Review B, 2001. 64(7): p. 075414. 151. M.D. Donohue and G.L. Aranovich, Classification of Gibbs adsorption isotherms. Advances in Colloid and Interface Science, 1998. 76–77: p. 137-152. 152. T. Mochizuki, M. Kubota, H. Matsuda, and L.F. D'Elia Camacho, Adsorption behaviors of ammonia and hydrogen sulfide on activated carbon prepared from petroleum coke by KOH chemical activation. Fuel Processing Technology, 2016. 144: p. 164-169.  186  153. H. Wang, S. Liu, and K.J. Smith, Understanding selectivity changes during hydrodesulfurization of dibenzothiophene on Mo2C/carbon catalysts. Journal of Catalysis, 2019. 369: p. 427-439. 154. H. Predel, Petroleum coke. Ullmann's Encyclopedia of Industrial Chemistry, 2000. 155. B.C. E., S.R. Klaus, S.J. A., and B.R. C., Characterization of biochar from fast pyrolysis and gasification systems. Environmental Progress & Sustainable Energy, 2009. 28(3): p. 386-396. 156. L. Volpe and M. Boudart, Compounds of molybdenum and tungsten with high specific surface area: I. Nitrides. Journal of Solid State Chemistry, 1985. 59(3): p. 332-347. 157. M. Lillo-Ródenas, D. Cazorla-Amorós, and A. Linares-Solano, Understanding chemical reactions between carbons and NaOH and KOH: an insight into the chemical activation mechanism. Carbon, 2003. 41(2): p. 267-275. 158. J. Wang and S. Kaskel, KOH activation of carbon-based materials for energy storage. Journal of Materials Chemistry, 2012. 22(45): p. 23710-23725. 159. S.-J. Park and W.-Y. Jung, Effect of KOH activation on the formation of oxygen structure in activated carbons synthesized from polymeric precursor. Journal of Colloid and Interface Science, 2002. 250(1): p. 93-98. 160. C. Peng, X.-b. Yan, R.-t. Wang, J.-w. Lang, Y.-j. Ou, and Q.-j. Xue, Promising activated carbons derived from waste tea-leaves and their application in high performance supercapacitors electrodes. Electrochimica Acta, 2013. 87: p. 401-408. 161. I.D. Rosca, F. Watari, M. Uo, and T. Akasaka, Oxidation of multiwalled carbon nanotubes by nitric acid. Carbon, 2005. 43(15): p. 3124-3131. 162. Y. Gao, D. Ma, C. Wang, J. Guan, and X. Bao, Reduced graphene oxide as a catalyst for hydrogenation of nitrobenzene at room temperature. Chemical Communications, 2011. 47(8): p. 2432-2434. 163. C. Prado-Burguete, A. Linares-Solano, F. Rodriguez-Reinoso, and C.S.-M. De Lecea, The effect of oxygen surface groups of the support on platinum dispersion in Pt/carbon catalysts. Journal of Catalysis, 1989. 115(1): p. 98-106. 164. S. Zhang, Y. Yan, T. Li, and Z. Ren, Lumping kinetic model for hydrotreating of bio-oil from the fast pyrolysis of biomass. Energy Sources, Part A, 2009. 31(8): p. 639-645   187  165. S.M. Jacob, B. Gross, S.E. Voltz, and V.W. Weekman, A lumping and reaction scheme for catalytic cracking. AIChE Journal, 1976. 22(4): p. 701-713. 166. Y. Nakagawa, M. Ishikawa, M. Tamura, and K. Tomishige, Selective production of cyclohexanol and methanol from guaiacol over Ru catalyst combined with MgO. Green Chemistry, 2014. 16(4): p. 2197. 167. T. Nimmanwudipong, R.C. Runnebaum, D.E. Block, and B.C. Gates, Catalytic conversion of guaiacol catalyzed by platinum supported on alumina: Reaction network including hydrodeoxygenation reactions. Energy & Fuels, 2011. 25(8): p. 3417-3427. 168. B. Güvenatam, O. Kurşun, E.H.J. Heeres, E.A. Pidko, and E.J.M. Hensen, Hydrodeoxygenation of mono- and dimeric lignin model compounds on noble metal catalysts. Catalysis Today, 2014. 233: p. 83-91. 169. A. Lanver, K. Ebel, R. Klopsch, W. Bertleff, R. Dehn, J.H. Teles, H. Kronemayer, and M.G. Schrems, Process for the preparation of 1-methylcyclopentane derivatives, 2012, Google Patents. 170. X. Zhu, L.L. Lobban, R.G. Mallinson, and D.E. Resasco, Bifunctional transalkylation and hydrodeoxygenation of anisole over a Pt/HBeta catalyst. Journal of Catalysis, 2011. 281(1): p. 21-29. 171. V.N. Bui, G. Toussaint, D. Laurenti, C. Mirodatos, and C. Geantet, Co-processing of pyrolisis bio oils and gas oil for new generation of bio-fuels: Hydrodeoxygenation of guaïacol and SRGO mixed feed. Catalysis Today, 2009. 143(1-2): p. 172-178. 172. V. Haensel and V. Ipatieff, Cracking cyclohexane. thermal and catalytic decomposition at high pressures. Industrial & Engineering Chemistry, 1943. 35(6): p. 632-638. 173. M. Ishikawa, M. Tamura, Y. Nakagawa, and K. Tomishige, Demethoxylation of guaiacol and methoxybenzenes over carbon-supported Ru–Mn catalyst. Applied Catalysis B: Environmental, 2016. 182: p. 193-203. 174. H.Y. Zhao, D. Li, P. Bui, and S.T. Oyama, Hydrodeoxygenation of guaiacol as model compound for pyrolysis oil on transition metal phosphide hydroprocessing catalysts. Applied Catalysis A: General, 2011. 391(1-2): p. 305-310. 175. C. Sepúlveda, K. Leiva, R. García, L.R. Radovic, I.T. Ghampson, W.J. DeSisto, J.L.G. Fierro, and N. Escalona, Hydrodeoxygenation of 2-methoxyphenol over Mo2N catalysts supported on activated carbons. Catalysis Today, 2011. 172(1): p. 232-239.  188  176. J. Wildschut, F.H. Mahfud, R.H. Venderbosch, and H.J. Heeres, Hydrotreatment of fast pyrolysis oil using heterogeneous noble-metal catalysts. Industrial & Engineering Chemistry Research, 2009. 48(23): p. 10324-10334. 177. H. Wang, F. Zhao, S.-i. Fujita, and M. Arai, Hydrogenation of phenol in scCO2 over carbon nanofiber supported Rh catalyst. Catalysis Communications, 2008. 9(3): p. 362-368  178. M. Rubeš, J. He, P. Nachtigall, and O. Bludský, Direct hydrodeoxygenation of phenol over carbon-supported Ru catalysts: A computational study. Journal of Molecular Catalysis A: Chemical, 2016. 423: p. 300-307. 179. V.M. 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. 180. C.A. Teles, R.C. Rabelo-Neto, J.R. de Lima, L.V. Mattos, D.E. Resasco, and F.B. Noronha, The Effect of Metal Type on Hydrodeoxygenation of Phenol Over Silica Supported Catalysts. Catalysis Letters, 2016. 146(10): p. 1848-1857  181. 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. 182. 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. 183. G. Kresse, G. Kresse and D. Joubert, Phys. Rev. B 59, 1758 (1999). Phys. Rev. B, 1999. 59: p. 1758. 184. J.P. Perdew, K. Burke, and M. Ernzerhof, Generalized gradient approximation made simple. Physical Review Letters, 1996. 77(18): p. 3865. 185. 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. 186. S.T. Oyama, C.C. Yu, and S. Ramanathan, Transition metal bimetallic oxycarbides: synthesis, characterization, and activity studies. Journal of Catalysis, 1999. 184(2): p. 535-549.  189  187. A. Hanif, T. Xiao, A.P.E. York, J. Sloan, and M.L.H. Green, Study on the structure and formation mechanism of molybdenum carbides. Chemistry of materials, 2002. 14(3): p. 1009-1015  188. Y. Hong, H. Zhang, J. Sun, K.M. Ayman, A.J.R. Hensley, M. Gu, M.H. Engelhard, J.-S. McEwen, and Y. Wang, Synergistic catalysis between Pd and Fe in gas phase hydrodeoxygenation of m-cresol. ACS Catalysis, 2014. 4(10): p. 3335-3345  189. J.A. Cecilia, A. Infantes-Molina, J. Sanmartín-Donoso, E. Rodríguez-Aguado, D. Ballesteros-Plata, and E. Rodríguez-Castellón, Enhanced HDO activity of Ni2P promoted with noble metals. Catalysis Science & Technology, 2016. 6(19): p. 7323-7333. 190. W.-P. Dow, Y.-P. Wang, and T.-J. Huang, Yttria-Stabilized Zirconia Supported Copper Oxide Catalyst: I. Effect of Oxygen Vacancy of Support on Copper Oxide Reduction. Journal of Catalysis, 1996. 160(2): p. 155-170. 191. 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. 192. 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. 193. J.-M. Manoli, P. Da Costa, M. Brun, M. Vrinat, F. Maugé, and C. Potvin, Hydrodesulfurization of 4, 6-dimethyldibenzothiophene over promoted (Ni, P) alumina-supported molybdenum carbide catalysts: activity and characterization of active sites. Journal of Catalysis, 2004. 221(2): p. 365-377. 194. A. Galtayries and J. Grimblot, Formation and electronic properties of oxide and sulphide films of Co, Ni and Mo studied by XPS. Journal of Electron Spectroscopy and Related Phenomena, 1999. 98: p. 267-275. 195. J. Cheng and W. Huang, Effect of cobalt (nickel) content on the catalytic performance of molybdenum carbides in dry-methane reforming. Fuel Processing Technology, 2010. 91(2): p. 185-193. 196. Y. Dai, V. Pavan Kumar, C. Zhu, M.J. MacLachlan, K.J. Smith, and M.O. Wolf, Mesoporous Silica-Supported Nanostructured PdO/CeO2 Catalysts for Low-Temperature Methane Oxidation. ACS Applied Materials & Interfaces, 2018. 10(1): p. 477-487.  190  197. X. Zhang, X. Zhu, L. Lin, S. Yao, M. Zhang, X. Liu, X. Wang, Y.-W. Li, C. Shi, and D. Ma, Highly dispersed copper over β-Mo2C as an efficient and stable catalyst for the reverse water gas shift (RWGS) reaction. ACS Catalysis, 2016. 7(1): p. 912-918. 198. A. Infantes-Molina, E. Gralberg, J.A. Cecilia, E. Finocchio, and E. Rodríguez-Castellón, Nickel and cobalt phosphides as effective catalysts for oxygen removal of dibenzofuran: role of contact time, hydrogen pressure and hydrogen/feed molar ratio. Catalysis Science & Technology, 2015. 5(6): p. 3403-3415. 199. J. Rabo, A. Risch, and M. Poutsma, Reactions of carbon monoxide and hydrogen on Co, Ni, Ru, and Pd metals. Journal of Catalysis, 1978. 53(3): p. 295-311. 200. K. Nagaveni, A. Gayen, G. Subbanna, and M. Hegde, Pd-coated Ni nanoparticles by the polyol method: an efficient hydrogenation catalyst. Journal of Materials Chemistry, 2002. 12(10): p. 3147-3151. 201. K. Nobuhara, H. Kasai, W.A. Diño, and H. Nakanishi, H2 dissociative adsorption on Mg, Ti, Ni, Pd and La surfaces. Surface Science, 2004. 566-568: p. 703-707. 202. S. Liu, H. Wang, P. Neumann, C.S. Kim, and K.J. Smith, Esterification over acid-treated mesoporous carbon derived from petroleum coke. ACS Omega, 2019. 4(3): p. 6050-6058. 203. M. Pozzo and D. Alfè, Hydrogen dissociation and diffusion on transition metal (=Ti, Zr, V, Fe, Ru, Co, Rh, Ni, Pd, Cu, Ag)-doped Mg(0001) surfaces. International Journal of Hydrogen Energy, 2009. 34(4): p. 1922-1930. 204. 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. 205. S. Sitthisa and D.E. Resasco, Hydrodeoxygenation of furfural over supported metal catalysts: a comparative study of Cu, Pd and Ni. Catalysis Letters, 2011. 141(6): p. 784-791. 206. X. Jiang, N. Ellis, and Z. Zhong, Fuel properties of bio-oil/bio-diesel mixture characterized by TG, FTIR and 1 H NMR. Korean Journal of Chemical Engineering, 2011. 28(1): p. 133-137. 207. N.S. Tessarolo, R.V. Silva, G. Vanini, A. Casilli, V.L. Ximenes, F.L. Mendes, A. de Rezende Pinho, W. Romão, E.V. de Castro, and C.R. Kaiser, Characterization of thermal  191  and catalytic pyrolysis bio-oils by high-resolution techniques: 1H NMR, GC×GC-TOFMS and FT-ICR MS. Journal of analytical and applied pyrolysis, 2016. 117: p. 257-267. 208. M.I. Schnitzer, C.M. Monreal, G.A. Facey, and P.B. Fransham, The conversion of chicken manure to biooil by fast pyrolysis I. Analyses of chicken manure, biooils and char by 13C and 1H NMR and FTIR spectrophotometry. Journal of Environmental Science and Health Part B, 2007. 42(1): p. 71-77. 209. M. Bertero, G. de la Puente, and U. Sedran, Products and coke from the conversion of bio-oil acids, esters, aldehydes and ketones over equilibrium FCC catalysts. Renewable Energy, 2013. 60: p. 349-354. 210. D.M. Ginosar and B. Subramaniam, Olefinic oligomer and cosolvent effects on the coking and activity of a reforming catalyst in supercritical reaction mixtures. Journal of Catalysis, 1995. 152(1): p. 31-41. 211. R. Gunawan, X. Li, A. Larcher, X. Hu, D. Mourant, W. Chaiwat, H. Wu, and C.-Z. Li, Hydrolysis and glycosidation of sugars during the esterification of fast pyrolysis bio-oil. Fuel, 2012. 95: p. 146-151. 212. D.C. Elliott, T.R. Hart, G.G. Neuenschwander, L.J. Rotness, and A.H. Zacher, Catalytic hydroprocessing of biomass fast pyrolysis bio‐oil to produce hydrocarbon products. Environmental Progress & Sustainable Energy: An Official Publication of the American Institute of Chemical Engineers, 2009. 28(3): p. 441-449. 213. V.N. Bui, D. Laurenti, P. Delichère, and C. Geantet, Hydrodeoxygenation of guaiacol: Part II: Support effect for CoMoS catalysts on HDO activity and selectivity. Applied Catalysis B: Environmental, 2011. 101(3-4): p. 246-255. 214. J.P. Lange, E. Van Der Heide, J. van Buijtenen, and R. Price, Furfural—a promising platform for lignocellulosic biofuels. ChemSusChem, 2012. 5(1): p. 150-166. 215. K. Xiong, W.S. Lee, A. Bhan, and J.G. Chen, Molybdenum carbide as a highly selective deoxygenation catalyst for converting furfural to 2‐methylfuran. ChemSusChem, 2014. 7(8): p. 2146-2149. 216. N. Yoshizawa, K. Maruyama, Y. Yamada, E. Ishikawa, M. Kobayashi, Y. Toda, and M. Shiraishi, XRD evaluation of KOH activation process and influence of coal rank. Fuel, 2002. 81(13): p. 1717-1722.  192  217. H.S. Fogler, Elements of chemical reaction engineering1999. 218. R.B. Bird, W.E. Stewart, and E.N. Lightfoot, Transport Phenomena, 2nd Edition2002: New York: Wiley. 219. V. Whiffen, A study of metal phosphides for the hydrodeoxygenation of phenols and pyrolysis oil, 2013, University of British Columbia. 220. A. Hayter, Probability and statistics for engineers and scientists2012: Nelson Education.   193  Appendix A  Catalyst Preparation Details of the catalyst preparation procedures and the carbon support activation methodology are reported in the main body of the thesis (see Chapter 2, section 2.2.2). A.1 Activated Petcoke (APC) and Activated Biochar (ABC)   Figure A.1: XRD patterns of activated bio-char (ABC; (a)) and activated petcoke (APC; (b)) samples with different activation temperatures. (Impurities: SiO2 (*); Al2O3 (♦); C peaks at (002), (100) and (110) planes, respectively)   194  XRD results in Figure A.1 show the change in the petcoke and biochar following chemical activation. For the petcoke activated different temperatures, distinct peaks at ~15o and ~28o can be assigned to the (001) and (002) diffraction planes of graphitic carbon (PDF card #: 00-041-1487). The intensity of the peaks gradually shifts from higher angle to lower angle with the increase of temperature. A similar trend was reported by Yoshizawa et al. [216] in their study of the activation of coal with KOH. For the bio-char samples ABC, the intensity of higher angle peak representing (002) plane is lower from the beginning. Nevertheless, the demolish of (002) peak as the temperature increase is the same as APC.   A.2 Catalyst Impregnation and Reduction Process (I) Preparation of Mo2C/C catalyst precursors Sample calculation for 10 wt% Mo on carbon (carbon could be APC, ABC or AC) is as follows: 1. Target for 10 wt% Mo/C catalyst is set to 5 g. 2. The amount of Mo required = 𝟏𝟎% × 𝟓 𝐠 = 𝟎. 𝟓 𝐠 3. The amount of ammonium heptamolybdate (AHM) required =𝟏𝟐𝟑𝟓. 𝟖𝟔 𝐠 𝐀𝐇𝐌/𝟕𝟗𝟓. 𝟗𝟒 𝐠 𝐌𝐨× 𝟎. 𝟓 𝐠 𝐌𝐨 = 𝟎. 𝟗𝟐 𝐠 𝐀𝐇𝐌 4. The amount of Carbon required = 𝟓 𝐠 − 𝟎. 𝟓 𝐠 = 𝟒. 𝟓 𝐠    (II) Preparation of promoted Mo2C/C catalyst precursors Sample calculation of 1 wt% Ni and 10 wt% of Mo on carbon is as follows: 1. Same as part A.2 (I) 2. Same as part A.2 (I) 3. Same as part A.2 (I)  195  4. The amount of Ni required = 𝟏% × 𝟓 𝐠 = 𝟎. 𝟎𝟓 𝐠 5. The amount of Ni(NO3)2·6H2O required =𝟐𝟗𝟎. 𝟕𝟗 𝐠 𝟓𝟖. 𝟔𝟗 𝐠 × 𝟎. 𝟎𝟓 𝐠 = 𝟎. 𝟐𝟓 𝐠 6. Required amount of Carbon = 𝟓 𝐠 − 𝟎. 𝟓 𝐠 − 𝟎. 𝟎𝟓 𝐠 = 𝟒. 4𝟓 𝐠  (III) Carbothermal hydrogen reduction (CHR) The CHR process was conducted in a U-tube like in Figure A.2:  Figure A.2: Schematic illustration of the quartz U-tube applied in CHR process.  . 196  Appendix B  Bio-oil Analysis B.1 Product Work Up  Figure B.1: Product work up for bio-oil analysis.   197  B.2 Detailed Example for GC-MS Analysis Here is an example of GC-MS data for bio-oil products. Figure B.2 is products identified in crude bio-oil   Figure B.2: GC-MS of chemicals in crude bio-oil.   198  Table B.1 summaries the chemical identified in the bio-oil and the area% they have.  Table B.1: Chemicals identified by different retention times in crude bio-oil. RT(min) Area (%) Name 1.524 1.11 Formic acid 1.560 0.3 1-Propanol, 2,2-dimethyl- 1.712 0.09 Propanedioic acid 1.721 0.33 Leucinol 1.766 0.08 Butanal 1.874 0.25 Ethyal Acetate 1.924 0.44 Lactic acid 2.264 11.46 Acetic acid 2.405 1.03 Propanoic acid 2.657 0.06 2,3-pentanedione 2.686 0.02 3-pentanone 2.927 0.28 2-Butanone, 3-hydroxy- 3.097 0.04 Butanoic acid, methyl ester 3.285 0.08 2-Propenoic acid 3.418 0.03 3-Penten-2-one 3.489 0.1 4-methoxymethoxy-hex-1-ene 3.662 0.07 2-ethoxypentane 3.707 0.01 2-methoxytetrahydrofuran 3.765 0.02 1-penten-3-one 4.018 0.11 Toluene 4.138 0.64 2-Butanone 4.249 0.09 2-Methoxy-1,3-dioxolane 4.511 0.26 Pentanal 4.688 0.11 Cyclopentanone 5.119 0.14 diglycolic acid 5.510 0.05 3-Furaldehyde  199  RT(min) Area (%) Name 5.740 0.06 2-Pentanone, 4-hydroxy- 5.975 0.04 Pentanoic acid, 2-methyl- 6.141 2.11 Furan, 2,5-dimethyl- 6.329 0.03 3-Penten-2-one 6.375 0.02 2-cyclopentene-1,4-dione 6.507 0.08 1-Ethylpentyl acetate 6.626 0.03 3-Pentenoic acid, methyl ester 6.899 0.44 Butanal 7.168 0.05 p-xylene 7.203 0.16 Oxirane,(methoxymethyl)- 7.294 0.14 3-(Diethylamino)-1,2-propanediol 7.509 0.51 Ethylene glycol 7.955 0.04 Furan,tetrahydro-2,5-dimethoxy- 8.163 0.46 2-cyclopenten-1-one,2-methyl- 8.306 0.12 Ketone, 2-furyl methyl 8.468 1.77 2-Furanone 8.740 0.25 3,4-dihydro-5,5-dimethyl-4-ethoxycarbonyloxazole 8.886 0.07 Cyclopentane,1,1,3-trimethyl- 8.995 0.22 2-Furanone,5-methyl- 9.105 0.09 3,4Dehydro-dl-proline 9.205 0.09 Cyclohexanone,2-methyl- 9.303 0.04 2-Propenoic acid,2-methylpropyl ester 9.430 0.18 Benzaldehyde 9.542 0.77 Furo[3,4-b]furan-2,6(3H,4H)-dione 9.625 0.05 2-Butanone,1-(acetyloxy)- 9.701 0.06 2H-Pyran-2-one 9.863 0.39 2-Furanone,3-methyl- 10.036 0.57 Phenol 10.259 0.08 Spiro[2,4]heptan-4-one  200  RT(min) Area (%) Name 10.343 0.62 3-Deoxyglucose 10.606 0.05 6-Nonen-1-ol 10.710 0.29 2-Cyclopenten-1-one 10.982 2.48 1,2-cyclopentanedione,3-methyl- 11.075 0.21 2-Cyclopenten-1-one,2,3-dimethyl- 11.221 0.73 4-Methyl-5H-furan-2one 11.439 0.79 Phenol,2-methyl- 11.528 0.12 Cyclohexene,3,3,5-trimethyl- 11.811 0.71 Phenol,4-methyl- 12.045 3.74 Phenol,2-methoxyl- 12.519 0.24 Maltol 12.582 0.38 2-Cyclopenten-1-one 13.029 0.35 Phenol,2,4-dimethyl- 13.327 0.39 Phenol,3-methoxy-2-methyl- 13.519 0.22 2-Methoxy-6-methylphenol 13.771 4.23 Phenol,2-methoxy-4-methyl- 13.996 3.13 1,2-Benzenediol 14.500 0.75 2-Furancarboxaldehyde,5-(hydroxymethyl)- 14.860 0.58 1,2-Benzenediol,3-methyl- 15.055 1.21 Phenol,4-ethyl-2-methoxy- 15.300 1.81 1,2-Benzenediol,4-methyl- 15.556 0.47 2-Methoxy-4-vinylphenol 16.081 0.41 Phenol,2,6-dimethoxy- 16.170 1.32 Phenol,2-methoxy-3-(2-propenyl)- 16.296 0.4 Phenol,2-methoxy-4-propyl- 16.643 3.84 1-butene,3-btoxy-2-methyl- 16.804 1.77 Vanillin 16.855 0.36 Phenol,2-methoxy-4-(1-propenyl)- 16.965 0.85 1-Octanol,2-methyl- 17.345 0.13 Phenol,2-methoxy-4-(1-propenyl)-  201  RT(min) Area (%) Name 17.397 0.44 Phenol,2-methoxy-4-(1-propenyl)- 17.570 0.36 Phenol,2-methoxy-4-propyl- 17.730 0.78 2,5-Pryidinedicarboxylic acid 17.934 2.02 Ethanone,1-(4-hydroxy-3-methoxyphenyl)- 19.087 34.18 Levoglucosan 19.867 1.07 Methyl-(2-hydroxy-3-ethoxy-benzyl)ether 20.840 1.35 3-Methoxycinnamic acid 21.822 1.10 1-Butanol,2,3-dimethyl-  Table B.2 presents the area% based on groups of chemical, the chemicals got summaries into 9 groups.  Table B.2: Chemical groups summarized from GCMS data. Chemical groups Area % Alcohols 2.77 Acids 14.48 Ketones 8.80 Phenols 28.79 Heterocyclic compounds 5.56 Sugars 34.8 Hydrocarbons 0.35 Esters 0.44 Ethers 4.01 Total 100   202  B.3 Detailed Example for 1H-NMR Analysis The analysis of 1H-NMR follows identification of types of protons by chemical shift. Here is an example of NMR analysis presented in Figure B.3 and Table B.3.  Catalyst: Mo2C/APC (BO7) 1st step upgrading-esterification Rxn conditions: w/ butanol, 180 oC, 1500 psi, 7000 rpm, 1000 cc/min H2, 1.5 h.   Figure B.3: Product distribution of aqueous and oil phase from 1st step reaction on Mo2C/APC catalyst compared to crude bio-oil.   203  Table B.3: Product distribution of aqueous and oil phase from 1st step reaction on Mo2C/APC catalyst compared to crude bio-oil. Chemical shift (ppm) 0.5-1.8 1.8-3.0 3.0-4.2  4.2-6.5  6.5-9.0  9.0-10.0  BO7_Aqueous phase 28.99 19.71 33.33 15.94 1.16 0.87 BO7_Oil phase 45.40 35.54 15.83 0.13 3.05 0.04 Crude Bio-oil 14.35 32.14 16.36 29.56 6.74 0.86     204  Appendix C  Products Calibration Among all the model compounds used in this thesis, DBF is chosen here as example. An external calibration method was applied for the reactant DBF and all the products. Here two main products biphenyl (BPh) and cyclohexylbenzene (CHB) are presented as examples. Note that the calibration curve is valid for corresponding level of concentration. For example, if the concentration of DBF in product is estimated around 0.2 wt%, the curve range should be from 0.1 - 0.5 wt%.    205   Table C.1: GC-MS calibration for reactant DBF.   Table C.2: GC-MS calibration for product BPh.   Aim wt%1.0 wt% DBF solution (g)Diluted to xx gReal wt% of DBFArea1 of DBF Area2 of DBF Ave. Area0.30 1.03 3.06 0.379 7534264 7470362 75023130.20 1.05 5.03 0.234 4630595 4733434 46820150.10 1.02 8.58 0.133 2657155 2656528 26568420.05 1.08 2.06 0.070 1386361 1412575 13994680.03 1.04 5.08 0.027 534881 540756 5378190.01 1.04 12.12 0.011 225594 224579 225087Aim wt%1.0 wt% BPh solution (g)Diluted to xx gReal wt% of BPhArea1 of BPh Area2 of BPh Ave. Area0.50 1.02 2.04 0.617 13630527 13960011 137952690.30 1.00 3.03 0.408 9399232 9449751 9424491.50.20 1.09 5.04 0.265 6305345 6227224 6266284.50.10 1.10 8.56 0.158 3858540 3817135 3837837.50.05 1.16 2.06 0.089 2296649 2204308 2250478.50.03 1.02 5.00 0.032 799360 805293 802326.5 206   Table C.3: GC-MS calibration for product CHB.   Aim wt% CHB (g) Decalin (g) wt% Area1 of CHB Area2 of CHB Ave. Area0.50 1.02 2.03 0.642 14040811 14034473 140376420.30 1.05 3.00 0.447 10173236 10153500 101633680.20 1.10 5.06 0.279 6494138 6522605 6508371.50.10 1.05 8.52 0.158 3740279 3605860 3673069.50.05 1.05 2.06 0.081 1967259 1968679 19679690.03 1.07 5.09 0.033 789485 789947 789716 207  Data from sample No.4~No.9 in Table C.1 generated calibration curve shown in Figure C.1:  Figure C.1: A linear correlation between GC-MS area ratio and DBT.  Data from sample No.3~No.8 in Table C.2 generated calibration curve shown in Figure C.2:  Figure C.2: A linear correlation between GC-MS area ratio and BPh.  y = 19.841x + 0.0097R² = 0.99990123456780 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4DBF y = 22.173x + 0.2627R² = 0.999202468101214160 0.1 0.2 0.3 0.4 0.5 0.6 0.7BPh 208   Data from sample No.3~No.8 in Table C.3 generated calibration curve shown in Figure C.3:  Figure C.3: A linear correlation between GC-MS area ratio and CHB.  y = 21.814x + 0.226R² = 0.998902468101214160 0.1 0.2 0.3 0.4 0.5 0.6 0.7CHB 209  Appendix D  Modeling D.1 Lump Kinetic Models for Chapter 4 The kinetic modeling was done by first lumping the obtained products into four different groups: products with 1-O removed, 2-O removed, hydrogenation and alkylation products (Ru/AC only). Based on these grouped products and proposed reaction network, a series of ODEs was derived to account for the change in component concentration as a function of time. The kinetic parameters were estimated using a Levenberg-Marquardt nonlinear regression methodology combined with a Runge-Kutta numerical integration to solve the relevant ODEs (Eq. D.1-8) as below:  For Pd/AC and Ru/AC catalysts:  𝑑𝐶𝐺𝑈𝐴𝑑𝑡= −𝑘1𝐶𝐺𝑈𝐴         (Eq. D.1) 𝑑𝐶𝐻𝑌𝐷𝑑𝑡= 𝑘1𝐶𝐺𝑈𝐴 − 𝑘2𝐶𝐻𝑌𝐷        (Eq. D.2) 𝑑𝐶1−𝑂𝑑𝑡= 𝑘2𝐶𝐻𝑌𝐷 − 𝑘3𝐶1−𝑂        (Eq. D.3) 𝑑𝐶2−𝑂𝑑𝑡= 𝑘3𝐶1−𝑂 − 𝑘4𝐶2−𝑂        (Eq. D.4) 𝑑𝐶𝐴𝑙𝑘𝑑𝑡= 𝑘4𝐶2−𝑂         (Eq. D.5)  For Mo2C/AC catalyst: 𝑑𝐶𝐺𝑈𝐴𝑑𝑡= −𝑘2𝐶𝐺𝑈𝐴         (Eq. D.6) 𝑑𝐶1−𝑂𝑑𝑡= 𝑘2𝐶𝐺𝑈𝐴 − 𝑘3𝐶1−𝑂        (Eq. D.7) 𝑑𝐶2−𝑂𝑑𝑡= 𝑘3𝐶1−𝑂         (Eq. D.8)  210   A program used for Mo2C/AC catalyst using lumped kinetic is presented below and run by MATLAB. ===================================================================== (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=[0 30 60 90 135 180 240 300]  nt=length (T) x(1:nt-1)=T(2:nt) x nx=length(x)  % 2-methoxy-phenol concentration data (mol/L) CAX=[0.077847 0.029656  211  0.0140866 0.00763642 0 0 0 0 ]; % 1-O products concentration data (mol/L) CBX=[0.29003568 0.3173192 0.3276988 0.3006377 0.2672747 0.2290926 0.2031436 0.1657029 ];  % 2-O products concentration data (mol/L) CCX=[0.011121 0.025949 0.040777 0.0648725 0.085261 0.118624 0.137159 0.177936 ];    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;  212   oldx=x; nx= length(x); y=[y1';y2';y3']; %y=[y1';y2';y3']; newy=y(:); oldy=reshape(newy,nx,nvar); x=x'; newx=[x;x;x;]; %newx=[x;x;x;] y01(1:nx)=CAX(1); y02(1:nx)=CBX(1); y03(1:nx)=CCX(1);   newy0=[y01';y02';y03']; %newy0=[y01';y02';y03']; %INPUT DATA NOW IN CORRECT COLUMN FORMAT y0=newy0 x=newx y=newy  %provide initial parameter guesses theta=[0.0002 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',0.00001,10000); disp('RESPONSE:') if kvg==1     disp('PROBLEM CONVERGED') elseif kvg==0     disp('PROBLEM DID NOT CONVERGE') end oldf=reshape(f,nx,nvar); oldr=reshape(y-f,nx,nvar);  213  disp('X-values:') 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('2MP') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,2) plot(oldx(:),oldy(:,2),'o',oldx(:),oldf(:,2),'--') title('2-O products ') xlabel('Time (min)') ylabel('Concentration (mol/L)') subplot(3,3,3) plot(oldx(:),oldy(:,3),'o',oldx(:),oldf(:,3),'--') title('2-O products ') 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 % first data point in x corresponds to initial condition global nvar nx x0 y0  214  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);   %ntest=(knt/nx);  yp(1)=-k1*yatx(1); yp(2)=k1*yatx(1)-k2*yatx(2); yp(3)=k2*yatx(2);  yprime=[yp(1)';yp(2)';yp(3)'];  (4) Calculation of Jacobian matrix 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  215  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 %   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))  216  %   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. % % 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  217  % 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. %       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,  218  %          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 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; %  219    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; 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);  220    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));     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);  221        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;   end;   if ss>sgoal,     break;   end; end;   % set return values % p=pbest; f=fbest; ss=sbest;  222  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 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);    223  %%% 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;   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;   224  D.2 DFT Calculations for Chapter 5 Here, a DBF molecule adsorbing on NiMo2C surface is used as an example to present the POSCAR, INCAR, KPOINTS used in the calculations. POSCAR: RNi-opted 1.0        15.3479995728         0.0000000000         0.0000000000        -4.7989432216        11.1476285816         0.0000000000         0.0000000000         0.0000000000        18.8577003479    Mo   Ni    C    O    H    39    1   44    1    8 Direct      0.212210000         0.229780003         0.844370008      0.371019989         0.185509995         0.789590001      0.077179998         0.288280010         0.760219991      0.350369990         0.175180003         0.928139985      0.013910000         0.131960005         0.870310009      0.211469993         0.478760004         0.844489992      0.370860010         0.431860000         0.789059997      0.079609998         0.039799999         0.761129975      0.350369990         0.425179988         0.928139985      0.013910000         0.381960005         0.870310009      0.710600019         0.230140001         0.844839990  225       0.871010005         0.186770007         0.789960027      0.576780021         0.288399994         0.760450006      0.850369990         0.175180003         0.928139985      0.513909996         0.131960005         0.870310009      0.710600019         0.480470002         0.844839990      0.870639980         0.435319990         0.789979994      0.577170014         0.036680002         0.759590030      0.850369990         0.425179988         0.928139985      0.513909996         0.381960005         0.870310009      0.211469993         0.732699990         0.844489992      0.376610011         0.688310027         0.790120006      0.077179998         0.788919985         0.760219991      0.350369990         0.675180018         0.928139985      0.013910000         0.631959975         0.870310009      0.212210000         0.982429981         0.844370008      0.370860010         0.939000010         0.789059997      0.075719997         0.537869990         0.760649979      0.350369990         0.925180018         0.928139985      0.013910000         0.881959975         0.870310009      0.708850026         0.728330016         0.844709992      0.871010005         0.684249997         0.789960027      0.850369990         0.675180018         0.928139985      0.513909996         0.631959975         0.870310009  226       0.708850026         0.980520010         0.844709992      0.874249995         0.937129974         0.791130006      0.577170014         0.540509999         0.759590030      0.850369990         0.925180018         0.928139985      0.513909996         0.881959975         0.870310009      0.575320005         0.787670016         0.763080001      0.227339998         0.113679998         0.765510023      0.000000000         0.000000000         0.946969986      0.472299993         0.110820003         0.756860018      0.250000000         0.000000000         0.946969986      0.225459993         0.362300009         0.767530024      0.000000000         0.250000000         0.946969986      0.472299993         0.361490011         0.756860018      0.250000000         0.250000000         0.946969986      0.725600004         0.113760002         0.767549992      0.500000000         0.000000000         0.946969986      0.969950020         0.109820001         0.758379996      0.750000000         0.000000000         0.946969986      0.726209998         0.363119990         0.768530011      0.500000000         0.250000000         0.946969986      0.970279992         0.359789997         0.757059991      0.750000000         0.250000000         0.946969986      0.225979999         0.613009989         0.772270024  227       0.000000000         0.500000000         0.946969986      0.470209986         0.609539986         0.761439979      0.250000000         0.500000000         0.946969986      0.225459993         0.863189995         0.767530024      0.000000000         0.750000000         0.946969986      0.470209986         0.860679984         0.761439979      0.250000000         0.750000000         0.946969986      0.725600004         0.611869991         0.767549992      0.500000000         0.500000000         0.946969986      0.970279992         0.610499978         0.757059991      0.750000000         0.500000000         0.946969986      0.719560027         0.859799981         0.777729988      0.500000000         0.750000000         0.946969986      0.969950020         0.860140026         0.758379996      0.750000000         0.750000000         0.946969986      0.473049998         0.762589991         0.630299985      0.376359999         0.686140001         0.630400002      0.347200006         0.561770022         0.631269991      0.416660011         0.513930023         0.632059991      0.513170004         0.593020022         0.631940007      0.543669999         0.717090011         0.631079972      0.512459993         0.410710007         0.633459985      0.542140007         0.316430002         0.634410024  228       0.470880002         0.200739995         0.634970009      0.374410003         0.181659997         0.634590030      0.346109986         0.277599990         0.633629978      0.416180015         0.394439995         0.633050025      0.572019994         0.531170011         0.632790029      0.493889987         0.859669983         0.629610002      0.323190004         0.724889994         0.629779994      0.271910012         0.503260016         0.631330013      0.618820012         0.775780022         0.631009996      0.617129982         0.332080007         0.634700000      0.491019994         0.123949997         0.635730028      0.320730001         0.090060003         0.635049999      0.270969987         0.261649996         0.633340001    229  INCAR: ISTART=0 #0 fresh start #ICHARG=0 default 2 when ISTART=0 else 0 ISPIN=2 #spin polarized calculation ENCUT = 400 eV #Default value from POTCAR determined by PREC unless use value from literature here C is 400 ev PREC=Normal # better to use accurate and normal to replace high and medium ISIF=2 #3 for optimization of cell combined with IBRION=2 IBRION=2 # important parameter determines how ion move 2 is Conjugate gradient method POTIM=0.2 #default value 0.5 when IBRION=1,2,3 ALGO=Fast # equals IALGO = 48 EDIFF=1E-04 #Energy difference convergence limit for electronic optimization EDIFFG=-0.05 #Energy difference convergence limit for ionic optimization default=10*EDIFF ISMEAR = -5 #Use 0 for KPOINTS less than 4 otherwise -5, 1 for metals #KPOINTS=5X5X5=125 SIGMA = 0.2 #default 0.2 NSW =400 #Total number of ionic steps LREAL=A #depend on PREC #use FALSE for reciprocal space projection  #ISYM = 2 # if PAW default 2 #VOSKOWN =0    # 0 for PBE 1 for PW91 NELMDL =-5 #to lower calculation steps #NSIM=4 # para calculation parameter #NFREE=10 #  230  #NELM=100 # default=60 Electron self consistent iteration times LORBIT=10 # requires RWIGS line when =0,1,2,5 SYMPREC=0.001 IDIPOL=3 LWAVE = .FALSE. LCHARG = .FALSE. IVDW=4  KPOINTS: Auto 0 M 3 3 1 0.  0.  0.  231  Appendix E  Supplementary Figures and Tables  E.1 Supplementary Information for Chapter 2  Table E.1: Ultimate analysis of raw petroleum coke.  Ultimate analysis (wt.%) Raw  Petroleum coke C 83.27±0.01 H 3.60±0.01 N 2.02±0.05 S 6.58±0.00 Oa 4.53±0.06 a. O content is calculated by difference (O = 100 – C – H – N – S)    232  Table E.2: Physical properties of Mo2C/APC with various Mo loadings and CHR temperatures. Sample Surface area  (m2/g) Pore volume (cm3/g) Burn off % Final yield% Vtotal Vmeso 0.25Mo – AHM/APC precursor 2080 0.99 0.24 — — 0.25Mo2C/APC_R1000 1900 1.09  0.17  27.7  47.3   0.5Mo – AHM/APC precursor 2128 1.03 0.15 — — 0.5Mo2C/APC_R900 2172 1.19  0.34  21.2  51.5  0.5Mo2C/APC_R1000 1766 0.94  0.25  28.3  46.9   1Mo – AHM/APC precursor 1965 0.83 0.07 — — 1Mo2C/APC_R700 2191 1.11  0.25  22.0  51.1  1Mo2C/APC_R800 1911 1.08  0.33  40.0  39.3  1Mo2C/APC_R900 1786 1.14  0.49  51.5  31.8  1Mo2C/APC_R1000 1808 1.04  0.35  53.7  30.3   2Mo – AHM/APC precursor 1981 0.85 0.09 — — 2Mo2C/APC_R600 1839 0.74  0.05  10.2  58.8  2Mo2C/APC_R700 1846 1.04  0.33  29.4  46.3  2Mo2C/APC_R800 1916 1.06  0.32  45.3  35.8  2Mo2C/APC_R900 1673 0.98  0.42  54.7  29.7  2Mo2C/APC_R1000  1798 0.98 0.29 57.8 27.6  5Mo – AHM/APC precursor 1843 0.85 0.11 — — 5Mo2C/APC_R700 1760 0.99  0.32  42.5  37.7  5Mo2C/APC_R800 1674 0.99  0.38  62.5  24.5  5Mo2C/APC_R1000 1517 0.96  0.40  69.2  20.2   7.5Mo – AHM/APC precursor 1611 0.65 0.05 — — 7.5Mo2C/APC_R700 1331 0.74  0.25  53.9  30.2  7.5Mo2C/APC_R800 1419 0.76  0.23  55.8  29.0   10Mo – AHM/APC precursor 1281 0.50 0.04 — — 10Mo2C/APC _R700 1098 0.62  0.21  53.7  30.4  10Mo2C/APC_R800 1316 0.74  0.26  75.0  16.4   233  Table E.3: Textural properties of Mo2C/PC catalysts. Sample Surface area (m2/g) Pore volume (cm3/g) Burn-off % Vtotal Vmeso 5Mo/PC_R700a 13 0.01 0 53.6 5Mo/PC_R800 27 0.02 0 47.3 2Mo/PC_R800 34 0.02 0 48.1 a. It is prepared based on raw petcoke, represented by PC.  Table E.4: CHNS analysis of raw petcoke and APC_800 support.  CHNS Analysis Sample C (wt%)a H (wt%)a N (wt%)a S (wt%)a O (wt%)a O/C (at.) Raw Petcoke 83.27 3.60 2.02 6.58 4.53b 0.04  APC_800 89.87 0.00 0.67 0.00 9.46 0.08   Table E.5: SEM/EDX analysis of APC_800 and Acid-T APC_800.  EDX Analysis Sample C (wt%) O (wt%) S (wt%) APC_800 95.4 4.2 0.0  Acid-T APC_800 93.9 2.9 0.2   234    Figure E.1: Raman spectra analysis of APC samples: (a) Raw petcoke; (b) APC_800.   235    Figure E.2: Profile of CH4 (mol%) produced during carbothermal hydrogen reduction (CHR): (□) 2Mo2C/APC_R700; (○) 5Mo2C/APC_R700; (Δ) 10Mo2C/APC_R700.   236   Figure E.3: NLDFT pore size distributions derived from N2 adsorption isotherms for APC_800 and Acid-T APC_800.    237     Figure E.4: NLDFT pore size distributions derived from N2 adsorption isotherms for 1Mo2C/APC at different CHR temperatures. (Left: 0-150 nm; Right: 0-50 nm)   238     Figure E.5: NLDFT pore size distributions derived from N2 adsorption isotherms for 2Mo2C/APC at different CHR temperatures. (Left: 0-150 nm; Right: 0-50 nm)   239  E.2 Supplementary Information for Chapter 3  Table E.6: ICP-OES analysis of PC/APC_800 and BC/ABC_700. Sample Al, ppm Ba, ppm Cr, ppm Co, ppm Cu, ppm Fe, ppm Mg, ppm Mn, ppm Ni, ppm V, ppm Zn, ppm PC 1545 24 33 22 29 1694 141 81 391 955 0 APC_800 1089 15 39 18 55 452 48 23 185 216 73 BC 748 57 36 16 101 192 990 445 40 44 93 ABC_700 1262 15 0 0 78 54 72 26 58 46 70    240  Table E.7: Elemental composition of Mo2C/C catalysts from XPS analysis. Catalyst C1s Mo3d O1s ABC support 10%Mo2C/ABC_R600 93.83 1.60 4.56 10%Mo2C/ABC_R650 92.62 2.04 5.34 10%Mo2C/ABC_R700 91.28 2.75 5.97 APC support 10%Mo2C/APC_R600 94.18 1.41 4.40 10%Mo2C/APC_R650 92.54 1.84 5.62 10%Mo2C/APC_R700 91.14 2.90 5.96 AC supporta 10%Mo2C/AC_R600 94.01 0.84 4.54 10%Mo2C/AC_R650 93.49 1.10 4.61 10%Mo2C/AC_R700 92.10 1.67 5.06 a. The elemental composition is normalized by excluding Si content.  Table E.8: Bulk density of different carbons. Samples Bulk density, g/cm3 PC 0.67 BC 0.46    241  Table E.9: O 1s XPS analysis of Mo2C/C catalysts with different carbon supports and reduction temperatures. Catalyst Mo-O bond (B.E.=530.7 eV) C-O bond (B.E.=532.7 eV) Si-O bond (B.E.=536.5 eV) ABC support 10%Mo2C/ABC_R600 49.04 50.96 — 10%Mo2C/ABC_R650 39.02 60.98 — 10%Mo2C/ABC_R700 28.64 71.36 — APC support 10%Mo2C/APC_R600 59.03 40.97 — 10%Mo2C/APC_R650 56.33 43.67 — 10%Mo2C/APC_R700 63.60 36.40 — AC support 10%Mo2C/AC_R600 32.46 31.56 35.98 10%Mo2C/AC_R650 32.55 25.68 21.53 10%Mo2C/AC_R700 31.59 31.63 46.43    242   Figure E.6: Raman spectra analysis of APC samples: (a) Raw petcoke; (b) APC_600; (c) APC_700; (d) APC_800; (e) APC_900.   243   Figure E.7: Raman spectra analysis of ABC samples: (a) Raw biochar; (b) ABC_600; (c) ABC_700; (d) ABC_800; (e) ABC_900.   244     Figure E.8: N2 adsorption-desorption isotherms of activated bio-char (ABC) and petcoke (APC) samples prepared by thermochemical method with KOH with different activation temperatures.   245   Figure E.9: O1s XPS narrow scan spectra deconvolution of different Mo2C catalysts: (a-c) 10%Mo2C/ABC with 600, 650 and 700 oC, respectively; (d-f) 10%Mo2C/APC with 600, 650 and 700 oC, respectively; (g-I) 10%Mo2C/AC with 600, 650 and 700 oC, respectively.    246   Figure E.10: Pore size distribution of ABC and APC supported catalysts.    247  E.3 Supplementary Information for Chapter 4 Table E.10, E.11, and E.12 represent the reactant and products distribution (both based on mol.%) of 2-methoxy-phenol HDO over Pd/AC catalyst as a function of reaction time at different reaction temperatures.  248  Table E.10: Guaiacol products distribution over Pd/AC catalyst as a function of reaction time at 240 oC. Time (min) 0 30 60 90 180 300 Reactant (mol%) GUA 32.8 8.5 2.4 0.0 0.0 0.0 HYD (mol.%)  1.5 0.0 0.0 0.0 0.0 0.0  38.1 52.1 55.3 57.0 56.0 55.2  24.8 36.3 38.9 39.5 40.5 41.1 sum  64.4 88.3 94.2 96.5 96.5 96.3 1-O removed (mol.%)  1.5 1.8 2.0 2.0 2.1 2.2  1.3 1.3 1.4 1.5 1.4 1.5  0.0 0.0 0.0 0.0 0.0 0.0 sum  2.8 3.1 3.4 3.5 3.5 3.7 2-O removed (mol.%)  0.0 0.0 0.0 0.0 0.0 0.0  0.0 0.0 0.0 0.0 0.0 0.0  0.0 0.0 0.0 0.0 0.0 0.0 sum  0.0 0.0 0.0 0.0 0.0 0.0  249  Table E.11: Guaiacol products distribution over Pd/AC catalyst as a function of reaction time at 300 oC. Time (min) 0 30 60 90 180 300 Reactant (mol%) GUA 22.4 6.1 0.0 0.0 0.0 0.0 HYD (mol.%)  3.9 1.9 1.8 1.6 1.2 0.0  43.0 46.7 45.8 39.9 26.5 13.7  25.6 32.7 32.7 32.5 26.7 15.6 sum  72.5 81.4 80.3 74.0 54.4 29.3 1-O removed (mol.%)  1.7 3.9 5.0 6.0 5.7 4.6  1.4 3.9 5.4 6.1 9.2 9.7  0.0 1.0 1.9 2.7 5.2 8.9 sum  3.1 8.7 12.2 14.8 20.1 23.3 2-O removed (mol.%)  2.0 3.8 6.5 9.2 22.8 43.4  0.0 0.0 1.0 2.0 2.8 3.5  0.0 0.0 0.0 0.0 0.0 0.5 sum  2.0 3.8 7.5 11.2 25.5 47.5    250  Table E.12: Guaiacol products distribution over Pd/AC catalyst as a function of reaction time at 330 oC. Time (min) 0 30 60 90 180 300 Reactant (mol%) GUA 15.5 5.0 0.0 0.0 0.0 0.0 HYD (mol.%)  2.1 1.0 1.4 1.6 0.0 0.0  41.2 32.2 16.5 8.3 0.0 0.0  29.7 27.7 17.5 9.2 0.0 0.0 sum  73.0 60.9 35.3 19.0 0.0 0.0 1-O removed (mon.%)  4.4 4.0 5.1 5.2 0.0 0.0  3.6 5.1 7.0 7.3 0.0 0.0  0.0 6.4 9.6 10.7 12.9 9.6 sum  8.0 15.5 21.7 23.1 12.9 9.6 2-O removed (mol.%)  3.5 16.8 38.7 52.8 80.4 80.5  0.0 1.8 4.3 4.1 5.1 6.8  0.0 0.0 0.0 1.0 1.5 3.1 sum  3.5 18.7 43.0 57.8 87.1 90.4  251  Table E.13: Guaiacol products distribution over Ru/AC catalyst as a function of reaction time at 240 oC. Time (min) 0 30 60 90 180 300 Reactant (mol.%) GUA 35.4 14.7 3.1 0.0 0.0 0.0 HYD (mol.%)  0.0 0.0 0.0 0.0 0.0 0.0  28.2 30.4 31.2 28.0 22.0 13.3  19.8 28.5 29.1 29.0 27.4 16.6 sum 48.0 58.9 60.4 57.0 49.5 29.9 1-O removed (mol.%) /  14.0/1.1 23.8/1.1 33.3/1.1 37.8/1.7 43.8/1.9 59.6/1.9 sum 15.1 24.9 34.4 39.5 45.6 61.5 2-O removed (mol.%) / /  1.5/0.0 /0.0 1.4/0.0 /0.0 2.1/0.0 /0.0 3.5/0.0 /0.0 4.9/0.0 /0.0 8.6/0.0 /0.0 /  0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 sum 1.5 1.4 2.1 3.5 4.9 8.6 Alkylation (mol.%) /  0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 / /  0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /0.0  Sum 0.0 0.0 0.0 0.0 0.0 0.0 Hydrocracking (mol.%) 0.0 0.0 0.0 0.0 0.0 0.0  252  Table E.14: Guaiacol products distribution over Ru/AC catalyst as a function of reaction time at 300 oC. Time (min) 0 30 60 90 180 300 Reactant (mol.%) GUA 20.8 5.4 0.0 0.0 0.0 0.0 HYD (mol.%)  1.1 0.0 0.0 0.0 0.0 0.0  27.2 22.4 11.9 5.5 0.0 0.0  19.0 16.8 10.7 6.4 0.0 0.0 sum 47.3 39.2 22.6 11.9 0.0 0.0 1-O removed (mol.%) /  28.4/1.3 47.0/1.5 63.1/2.4 63.9/2.4 36.1/2.4 14.2/0.0 sum 29.7 48.5 65.5 66.3 38.3 14.2 2-O removed (mol.%) / /  2.3/0.0 /0.0 6.9/0.0 /0.0 11.9/0.0 /0.0 19.0/1.0 /0.0 42.5/4.6 /0.0 56.4/6.6 /0.0 /  0.0/0.0 0.0/0.0 0.0/0.0 2.0/0.0 3.8/1.1 5.5/2.7 sum 2.3 6.9 11.9 21.9 52.0 71.3 Alkylation (mol.%) /  0.0/0.0 0.0/0.0 0.0/0.0 0.0/0.0 1.9/0.0 3.2/0.0 / /  0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /0.0 3.6/0.0 /3.0 4.0/0.0 /5.6  Sum 0.0 0.0 0.0 0.0 8.5 12.8 Hydrocracking (mol.%) 0.0 0.0 0.0 0.0 1.1 1.7  253  Table E.15: Guaiacol products distribution over Ru/AC catalyst as a function of reaction time at 330 oC. Time (min) 0 30 60 90 180 300 Reactant (mol.%) GUA 18.6 3.3 0.0 0.0 0.0 0.0 HYD (mol.%)  2.0 0.0 0.0 0.0 0.0 0.0  16.3 8.8 1.2 0.0 0.0 0.0  15.2 9.5 1.6 0.0 0.0 0.0 sum 33.4 18.3 2.8 0.0 0.0 0.0 1-O removed (mol.%) /  29.8/2.4 21.2/0.0 11.2/0.0 4.0/0.0 0.0/0.0 0.0/0.0 sum 32.2 21.2 11.2 4.0 0.0 0.0 2-O removed (mol.%) / /  8.9/2.3/2.7 38.1/13.8/2.1 56.2/22.5/1.1 67.6/17.4/0.0 63.6/8.4/0.0 49.2/2.5/0.0 /  2.0/0.0 3.2/0.0 3.9/1.1 4.0/1.9 5.4/5.2 6.0/6.1 sum 15.8 57.2 84.8 90.9 82.5 63.4 Alkylation (mol.%) / / 0.0/0.0 0.0/0.0 0.0/0.0 0.5/0.5 1.5/1.4 3.7/3.4 / /  0.0/0.0 /0.0 0.0/0.0 /0.0 0.0/0.0 /1.1 0.7/0.4 /2.4 4.3/1.6 /7.2 8.2/3.4 /14.4  Sum 0.0 0.0 1.1 4.6 15.9 33.0 Hydrocracking (mol.%) 0.0 0.0 0.0 0.5 1.5 3.1  254  Table E.16: Guaiacol products distribution over Mo2C/AC catalyst as a function of reaction time at 330 oC. Time (min) 0 30 60 90 135 180 240 300 Reactant (%) GUA 95.0 70.2 52.8 37.4 20.0 11.5 0.0 0.0 1-O removed  5.0 22.75 37.5 49.5 62.0 66.3 73.5 72.9  0.0 2.1 2.8 3.6 5.3 6.4 7.3 7.3  0.0 2.9 3.4 4.0 5.5 6.4 7.1 6.1  0.0 1.4 2.5 3.3 4.0 4.9 6.5 6.4 sum 5.0 29.1 46.1 60.5 76.8 84.2 94.3 92.7 2-O removed  0.0 0.0 0.0 0.6 0.8 1.1 1.6 2.1  0.0 0.7 1.0 1.6 2.4 3.2 4.2 5.2 sum 0.0 0.7 1.0 2.2 3.2 4.4 5.7 7.3  255  Table E.17: Guaiacol products distribution over Mo2C/AC catalyst as a function of reaction time at 350 oC. Time (min) 0 30 60 90 135 180 240 300 Reactant (%) GUA 59.0 38.1 17.4 4.2 2.9 1.9 0.0 0.0 1-O removed  32.3 46.6 62.9 72.2 70.9 69.8 69.8 68.3  3.1 4.5 5.4 6.1 6.6 7.5 7.5 7.7  3.3 5.8 6.3 6.6 6.1 5.7 5.5 5.2  1.3 2.0 3.7 5.0 6.2 6.3 6.4 6.4 sum 40.0 58.9 78.3 89.8 89.8 89.3 89.3 87.7 2-O removed  0.0 1.0 1.5 2.0 2.3 2.7 3.3 3.8  1.0 2.0 2.8 4.0 5.0 6.0 7.4 8.5 sum 1.0 3.0 4.3 6.0 7.3 8.8 10.7 12.3  256  Table E.18: Guaiacol products distribution over Mo2C/AC catalyst as a function of reaction time at 375 oC. Time (min) 0 30 60 90 135 180 240 300 Reactant (%) GUA 20.5 8.0 3.7 2.0 0.0 0.0 0.0 0.0 1-O removed  62.6 69.6 68.8 63.6 59.9 51.2 46.8 37.8  5.6 5.9 6.3 7.0 7.3 7.2 7.5 7.3  6.3 5.7 5.1 4.7 4.2 3.7 3.2 1.4  2.1 4.0 5.4 5.4 4.4 3.7 2.2 1.7 sum 76.5 85.1 85.7 80.6 75.8 65.9 59.7 48.2 2-O removed  0.0 1.0 1.9 2.5 3.2 4.3 5.4 6.5  2.9 6.0 8.7 14.9 21.0 29.9 34.9 45.3 sum 2.9 7.0 10.7 17.4 24.2 34.1 40.3 51.8  257  E.4 Supplementary Information for Chapter 5  Table E.19: Carbon balance error calculation for reactant (DBF) and products in HDO of dibenzofuran over 10Mo2C/APC catalyst at 350 oC and 4.1 MPa. Time (min) CHB (mol/L) CHY (mol/L) BPh (mol/L) BCH (mol/L) DBF (mol/L) Total mole of C (mol) Total mole of C in Feed (mol) Carbon balance error % 144.00 6.68E-04 2.84E-04 3.06E-04 1.16E-04 0.00E+00 1.48E-02 1.55E-02 4.52  165.00 6.50E-04 2.92E-04 3.27E-04 1.15E-04 1.25E-06 1.49E-02 1.55E-02 3.96  188.00 6.60E-04 2.98E-04 3.47E-04 1.17E-04 2.40E-06 1.53E-02 1.55E-02 1.20  202.00 6.67E-04 2.71E-04 3.42E-04 1.20E-04 1.49E-06 1.52E-02 1.55E-02 1.93  254.00 6.49E-04 2.91E-04 3.40E-04 1.20E-04 2.93E-06 1.51E-02 1.55E-02 2.51      258  Table E.20: Standard deviation of product concentrations in HDO of dibenzofuran over 10Mo2C/APC catalyst at 350 oC and 4.1 MPa. Time (min) CHB (mol/L) Std. dev % CHY (mol/L) Std. dev % BPh (mol/L) Std. dev % BCH (mol/L) Std. dev % 144.00 6.68E-04 1.37 2.84E-04 3.60  3.06E-04 4.96  1.16E-04 1.96 165.00 6.50E-04 2.92E-04 3.27E-04 1.15E-04 188.00 6.60E-04 2.98E-04 3.47E-04 1.17E-04 202.00 6.67E-04 2.71E-04 3.42E-04 1.20E-04 254.00 6.49E-04 2.91E-04 3.40E-04 1.20E-04    259  Table E.21: Diagram of horizontal and vertical DBF adsorption on clean Mo2C (101) surface for DBF adsorption energy calculation.    Orientation DBF adsorption on clean Mo2C (101) DBF-H1 DBF-H2 DBF-V1 DBF-V2 Top view     Side view     ΔG (eV) -1.22 -1.12 -0.64 -0.81  260  Table E.22: Diagram of four types of horizontal DBF adsorption on Ni-Mo2C surface for DBF adsorption energy calculation. Orientation DBF horizontal adsorption on Ni-Mo2C (101) DBF-H1 DBF-H2 DBF-H3 DBF-H4 Top view     Side view     ΔG (eV) -1.52 -1.12 -1.52 -1.15   261  Table E.23: Diagram of four types of vertical DBF adsorption on Ni-Mo2C surface for DBF adsorption energy calculation.  Orientation DBF vertical adsorption on Ni-Mo2C (101) DBF-V1 DBF-V2 DBF-V3 DBF-V4 Top view     Side view     ΔG (eV) -0.59 -0.64 -0.94 -0.88  262  Table E.24: Diagram of four types of horizontal DBF adsorption on Pd-Mo2C surface for DBF adsorption energy calculation. Orientation DBF horizontal adsorption on Pd-Mo2C (101) DBF-H1 DBF-H2 DBF-H3 DBF-H4 Top view     Side view     ΔG (eV) -1.29 -1.07 -1.10 -1.27  263  Table E.25: Diagram of four types of vertical DBF adsorption on Pd-Mo2C surface for DBF adsorption energy calculation. Orientation DBF vertical adsorption on Pd-Mo2C (101) DBF-V1 DBF-V2 DBF-V3 DBF-V4 Top view     Side view     ΔG (eV) -0.58 -0.60 -0.88 -0.85  264  Table E.26: Diagram of four types of horizontal DBF adsorption on Cu-Mo2C surface for DBF adsorption energy calculation. Orientation DBF horizontal adsorption on Cu-Mo2C (101) DBF-H1 DBF-H2 DBF-H3 DBF-H4 Top view     Side view     ΔG (eV) -1.34 -1.10 -1.12 -1.25   265  Table E.27: Diagram of four types of vertical DBF adsorption on Cu-Mo2C surface for DBF adsorption energy calculation. Orientation DBF vertical adsorption on Cu-Mo2C (101) DBF-V1 DBF-V2 DBF-V3 DBF-V4 Top view     Side view     ΔG (eV) -0.69 -0.53 -0.54 -0.77  266         Figure E.11: High angle annular dark field TEM scanning images (HAADF-STEM) and Energy dispersive X-ray (EDX) elemental mappings of fresh Pd-Mo2C/ABC catalyst.  267   Figure E.12: The potential energy during dissociative adsorption of H2 on three surfaces: (a, b) Pd-Mo2C; (c, d) Ni-Mo2C; (e, f) Cu-Mo2C surface.     268   Figure E.13: TPR-MS CH4 profile of Ni-Mo2C/ABC and Ni/ABC catalysts during carbothermal hydrogen reduction (CHR).    269  E.5 Supplementary Information for Chapter 6  Table E.28: Textual and Chemical Properties of Catalysts. Catalyst CO uptake  (µmol/g) NH3-TPD  (µmol/g) Surface area  (m2/g) Pore volume  (cm3/g) Ru/C 122 258 775 0.87 Ru/SiAl 27 690 513 0.58 Mo2C/APC 51 260 1302 0.72 1Ni-Mo2C/ABC 67 400 1326 0.93 1Ni-Mo2C/APC 67 685 1527 0.7 2Ni-10Mo2C/ABC 77 965       270   Table E.29: Identification of types of Protons by Chemical Shift. Chemical Shift (ppm) Type of protons 1.8-0.5 Aliphatic C bond to aliphatic groups (-CH3-, -CH2-) 3.0-1.8 Aliphatic adjacent to aromatic groups or alkene groups (Ar-CH2-, Ar-CH3, -CH3-C=O, -CH3-N=)  4.2-3.0 Aliphatic adjacent to O (-CH3O-, -CH2O-, -CH2-N=) 6.5-4.2 Phenolic or non-conjugated olefin proton  9.0-6.5 Aromatic 10.0-9.0 Aldehydic compounds (R-(H-)C=O)  The type of protons can be classified into several different groups based on different chemical shifts. Typically, the chemical shift between 0.5 to 1.8 ppm belongs to aliphatic C bound to aliphatic groups (-CH3-, -CH2-); the shift between 1.8-3.0 ppm can be identified as aliphatic C adjacent to aromatic groups or alkene groups (Ar-CH2-, Ar-CH3, CH3-C=O, CH3-N); the chemical shift of 3.0-4.2 ppm is accounted as aliphatic adjacent to O (CH3O-, -CH2O-, -CH2-N-); for phenolic or non-conjugated olefin proton (Ar-OH, HC=C-), the chemical shift appears at 4.2-6.5 ppm; the shift of 6.5-9.0 belongs to aromatic; the chemical shift appeared at 9.0-10.0 is aldehydic compounds (R-(H-)C=O).    271    Figure E.14: NMR results for feed and 1st esterification oil phase products at 180 oC and 10.3 MPa. *represent the reaction was operated at 240 oC.  272    Figure E.15: NMR results of oil phase products after 2nd step esterification. (a. Reaction Time=1.5 h; Reaction Temperature= 300 oC; b. Reaction Time=4.0 h; Reaction Temperature= 300 oC; c. Reaction Time=4.0 h, Reaction Temperature= 350 oC; a,b,c all used feed from 1st step esterification by Ru/C catalyst; d. Reaction Time=4.0 h, Reaction Temperature= 350 oC, used feed from 1st step esterification by 2Ni-Mo2C/ABC catalyst )   273  Appendix F  Mass Transfers and Heat Transfer Effects F.1 External Mass Transfer Effect in Fixed-bed reactor 10%Mo2C/ABC catalyst in HDO of DBF (2 wt%) at 330 oC and P=600 psi has been put here as an example. The details are shown in Table F.1.  Table F.1: The details of external mass transfer calculation by Mears criterion. Symbol Definition/Unit  Value -rDBF(obs) Observed reaction rate, [kmol/gcat.s] 5.64E-10 k1 Stabilized kinetic parameter, [s-1] 4.74E-03 Vc Loaded catalyst volume, [mL] 2.50E+00 Vsic Loaded inert volume, [mL] 0.00E+00 ρc Catalyst density, [g/cm3] 2.96E+00 mcat mass of loaded catalyst, [g] 7.20E-01 Lbed Length of the catalyst bed, [cm] 4.20E+00 ρb catalyst bed density  2.88E-01 ø porosity or void fraction of packed bed  9.03E-01 dp catalyst particle diameter, [m] 1.35E-04 R catalyst particle radius, [m] 6.75E-05 n reaction order 1.00E+00 Vo Total pore volume of the catalyst, [cm3/g] 9.60E-01 Scat. Surface area of the catalyst, [m2/g] 1.49E+03 rpore Pore radii of the catalyst, [cm] 1.29E-07 øp Catalyst particle porosity  7.40E-01 τ Tortuosity factor 3.00E+00 σc Constriction factor 8.00E-01 Ƴ Shape factor  1.00E+00 ρmix Mixture density, [kg/m3] 2.06E+00 µmix Dynamic viscosity of the mixture, [kg/m.s] 1.51E-05 νmix Kinetic viscosity of mixture, [m2/s] 7.33E-06 Mmix Feed molecular weight, [g/mol] 2.59E+00 PH2 Partial pressure of H2, [atm] 4.07E+01 PDBF Partial pressure of DBF, [atm] 3.27E-02 PDecalin Partial pressure of Decalin, [atm] 1.31E-01 Ptotal Total pressure in the system, [atm] 4.08E+01  274  Symbol Definition/Unit  Value CH2 Bulk gas concentration of H2 [kmol/m3] 1.03E+00 CDBF Bulk gas concentration of DBF [kmol/m3] 8.99E-06 ri Internal radius of the reactor, [m] 4.32E-03 A Cross sectional area of the reactor, [m2] 5.86E-05 γo Volumetic flow rate, [m3/s] 1.67E-06 U Superfical gas velocity, [m/s] (U=γo/A) 2.85E-02 Re Reynolds number 5.26E-01 Re' Reynolds number considering void fraction  5.41E+00 ṼDBF,c Critical volume of DBF, [mL/g-mol] 7.50E+02 ṼDBF,liquid Critical volume of DBF at its normal boiling point, [mL/g-mol] 2.94E+02 σ DBF Lennard-Jones parameters for DBF/characteristic length, [Å]  7.64E+00 σ H2 Lennard-Jones parameters for H2/characteristic length, [Å] 2.92E+00 σ DBF-H2 Lennard-Jones parameters for DBF-H2/characteristic length, [Å] 5.28E+00 TDBF,b Boiling point of DBF, [k] 5.60E+02 TDBF,c Critical point temperature of DBF, [k] 8.24E+02 εDBF/ĸ Lennard-Jones parameters for DBF/Boltzmann's constant, [k]  6.34E+02 εH2/ĸ Lennard-Jones parameters for H2/Boltzmann's constant, [k] 3.80E+01 εDBF-H2/ĸ Lennard-Jones parameters for DBF-H2/Boltzmann's constant, [k] 1.55E+02 Trxn Reaction temperature, [k] 6.03E+02 T* Dimensionless temperature, [k] (T*=ĸTrxn/εDBT-H2) 3.88E+00 ΩD, DBF-H2 Collision integral, calculated by ignore the last two terms 8.88E-01 Prxn Reaction pressure, [atm] 4.08E+01 MDBF Mole weight of DBF, [g/mol] 1.68E+02 MH2 Mole weight of H2, [g/mol] 2.02E+00 DDBF-H2 Binary bulk phase diffusivity, [m2/s] 1.93E-06 Sc Schmidt number (Sc=νmix/DDBF-H2) 3.80E+00 Sh Sherwood number (Sh=2+0.6Re1/2Sc1/3) 4.18E+00 kc Mass transfer coefficient, [m/s] 5.97E-02 MC Mears’ criterion for external difussion 2.04E-14  The important formulas involved in this calculation are listed below: σDBF=0.841*ṼDBF,c1/3        (Eq. F.1)  275  σDBF-H2 = ½ (σ DBF+ σ H2)      (Eq. F.2) εDBF/ĸ=0.77*TDBF,C        (Eq. F.3) 𝜀𝐷𝐵𝐹−𝐻2ĸ=√ε𝐷𝐵𝐹−𝐻2ĸ×ε𝐻2ĸ      (Eq. F.4) 𝛺𝐷,𝐷𝐵𝐹−𝐻2 =1.06036𝑇∗0.15610+0.19300exp (0.47635𝑇∗      (Eq. F.5) 𝐷𝐷𝐵𝐹−𝐻2 =0.0018583𝑃𝑟𝑥𝑛𝜎𝐷𝐵𝐹−𝐻22 𝛺𝐷,𝐷𝐵𝐹−𝐻2√𝑇3(1𝑀𝐷𝐵𝐹+1𝑀𝐻2)   (Eq. F.6) Sc=νmix/DDBF-H2        (Eq. F.7) Sh=2+0.6Re1/2Sc1/3        (Eq. F.8) Re = 𝑈𝜌𝑑𝑝𝜇        (Eq. F.9) Re’= 𝑅𝑒(1−∅)𝛾        (Eq. F.10) kc=𝑆ℎ×𝐷𝐷𝐵𝐹−𝐻2𝑑𝑃        (Eq. F.11) 𝑀𝐶 =−𝑟𝐷𝐵𝐹(𝑂𝑏𝑠)𝜌𝑏𝑅𝑛𝑘𝐶𝐶𝐷𝐵𝐹       (Eq. F.12)  The calculated MC is far less than 0.15, which means the external mass transfer effects can be neglected.   276  F.2 Internal Mass Transfer Effect in Fixed-bed reactor  Table F.2: The details of internal mass transfer by Weisz-Prater criterion. Symbol Definition/Unit  Value DDBT-H2 Binary bulk phase diffusivity [m2/s] 4.94E-02 Deff, DBT-H2 Effective diffusivity [m2/s] 9.75E-03 Dknudsen Knudsen diffusivity [m2/s] 1.92E-06 Deff,knudsen Effective knudsen diffusivity [m2/s] 3.78E-07 Deff, rxn Effective diffusivity in this reaction [m2/s] 3.78E-07 ø1 Thiele modulus for 1st order reaction 7.56E-03 η Internal effectiveness factor 1.00E+00 Cwp  Weisz-Prater criterion 5.71E-05  The important formulas involved in this calculation are listed below: 𝐷𝑒𝑓𝑓,𝐷𝐵𝑇−𝐻2 =𝐷𝐷𝐵𝑇−𝐻2∅𝑃𝜎𝐶𝜏     (Eq. F.13) 𝐷𝑘𝑛𝑢𝑑𝑠𝑒𝑛 =2𝑟𝑝𝑜𝑟𝑒3√(8∙𝑅𝑔)∙𝑇𝜋∙𝑀𝑚𝑖𝑥     (Eq. F.14) 𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑𝑠𝑒𝑛 =𝐷𝑘𝑛𝑢𝑑𝑠𝑒𝑛∅𝑃𝜎𝐶𝜏     (Eq. F.15) 1𝐷𝑒𝑓𝑓,𝑟𝑥𝑛=1𝐷𝑒𝑓𝑓,𝐷𝐵𝑇−𝐻2+1𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑𝑠𝑒𝑛    (Eq. F.16) ∅1 = 𝑅√𝑘1𝐷𝑒𝑓𝑓,𝑟𝑥𝑛       (Eq. F.17) η =3∅12 (∅1𝑐𝑜𝑡ℎ∅1 − 1)      (Eq. F.18) 𝐶𝑊𝑃 = η∅12=−𝑟𝐷𝐵𝑇(𝑜𝑏𝑠)𝜌𝑐𝑅2𝐷𝑒𝑓𝑓,𝑟𝑥𝑛𝐶𝐴𝑆     (Eq. F.19)   277  According to Weisz-Prater criterion [217], the calculated value of 𝐶𝑊𝑃 ≪1, which means internal mass transfer effects can also be neglected.    278  F.3 Heat Transfer Effect in Fixed-bed reactor  Table F.3: The details of heat transfer calculation by Mears Criterion.  Symbol Definition  Value kt Thermal conductivity calculated by semiempirical method for polyatomic gases [W/(m.K)] 2.97E-01 Ĉp(H2) Heat capacity of H2 at Trxn [J/Kg·K] 1.45E+04 ΔHrxn Heat of reaction [kJ/mol] 1.05E+02 Ea Activation energy [J/mol] 6.70E+04 Pt Prandtl number  7.38E-01 Nu Nusselt number 3.26E+00 h Heat transfer coefficient [W/(m2·K)] 7.18E-03 Rg Gas constant [J/(mol·k)] 8.31E+00 MC’ Mears’ criterion for isothermal operation -3.54E-18  The important formulas involved in this calculation are listed below: kt=(Ĉ𝑝,(𝐻2)+5𝑅4𝑀𝐻2)µ𝑚𝑖𝑥       (Eq. F.20) Pr=µmix.Ĉp/kt        (Eq. F.21) Nu=2+0.6Re1/2Pr1/3       (Eq. F.22) h=𝑁𝑢∙𝑘𝑡𝑑𝑝         (Eq. F.23) 𝑀𝐶 = |−∆𝐻𝑟𝑥𝑛(−𝑟𝐷𝐵𝑇,𝑜𝑏𝑠)𝜌𝑏𝑅𝐸ℎ𝑇2𝑅𝑔|      (Eq. F.24) According to Mears’ criterion [217, 218], the calculated value is far less than 0.15, thus heat transfer effect during the reaction can be neglected.    279  F.4 External and Internal Mass Transfers in Batch Reactor  All the kinetic data reported in Chapters 3 and 4 were obtained from a batch reactor operation after excluding external and internal mass transfer’s effects. 10Mo2C/APC-R650 catalyst was put here as an example.   Table F.4: A detailed list of external and internal mass transfer coefficient calculation for 10Mo2C/APC-R650 in HDO of 4-MP in batch reactor. Symbol Definition/Unit Value 𝑘𝐿𝑎 Liquid-side mass transfer coefficienta, [s-1] 4.37E-02 𝑘4−𝑀𝑃 Kinetic parameter, [s-1] 6.70E-05 R Catalyst particle radius, [m] 6.75E-05 ø1 Thiele modulus for 1st order reaction 2.49E-06 Deff, DBF-H2 Effective diffusivity, [m2/s] 4.94E-02 η Internal effectiveness factor 1.00E+00 Ω Overall effectiveness factor 9.99E-01 a. This number was adopted from [219] since the operating system was the same.  The important formulas involved in this calculation are listed below: ø1 = 𝑅(𝑘4−𝑀𝑃𝐷eff,DBF−H2)0.5      (Eq. F.25) 𝛺 =𝜂(1+𝜂𝑘𝐷𝐵𝑇𝑘𝐿𝑎)       (Eq. F.26) η = (3ø12)(ø1𝑐𝑜𝑡ℎø1 − 1)      (Eq. F.27)  As shown in Table F.4, ø1 < 0.4 and Ω ≈η = 1, both external and internal mass transfer can be neglected.  280  Appendix G  Error Analysis and Repeatability  G.1 Carbon Balance The carbon balance should be 100% for each experiment. Table G.1 presents the reactant and products concentration of 10Mo2C/APC-700 catalyst over time for the reaction of HDO of DBF in a fixed-bed reactor. It can be seen that the carbon balance error is within 5%  Table G.1: Reactant (DBF) and products concentration for HDO of dibenzofuran over 10Mo2C/APC-700 catalyst at 350 oC and 4.1 MPa. Time (min) CHB (mol/L) CHY (mol/L) BPh (mol/L) BCH (mol/L) DBF (mol/L) Total mole of C (mol) Total mole of C in Feed (mol) Carbon balance error % 61.00 7.45E-04 2.73E-04 2.67E-04 1.25E-04 4.95E-06 1.53E-02 1.55E-02 0.92  83.00 7.17E-04 2.60E-04 2.87E-04 1.18E-04 2.47E-06 1.51E-02 1.55E-02 2.74  103.00 6.93E-04 2.75E-04 3.00E-04 1.17E-04 1.70E-06 1.50E-02 1.55E-02 3.21  123.00 7.01E-04 2.86E-04 2.98E-04 1.22E-04 8.32E-07 1.52E-02 1.55E-02 1.99  144.00 6.68E-04 2.84E-04 3.06E-04 1.16E-04 0.00E+00 1.48E-02 1.55E-02 4.52  165.00 6.50E-04 2.92E-04 3.27E-04 1.15E-04 1.25E-06 1.49E-02 1.55E-02 3.96  188.00 6.60E-04 2.98E-04 3.47E-04 1.17E-04 2.40E-06 1.53E-02 1.55E-02 1.20  202.00 6.67E-04 2.71E-04 3.42E-04 1.20E-04 1.49E-06 1.52E-02 1.55E-02 1.93  254.00 6.49E-04 2.91E-04 3.40E-04 1.20E-04 2.93E-06 1.51E-02 1.55E-02 2.51   281  G.2 CHNS Repeatability Table G.2 presented several examples for CHNS data.  Table G.2: CHNS results for different samples.  Carbon% Hydrogen% Nitrogen% Sulfur% BO-14 oil phase-1 53.46 14.09 0.17 0.51 BO-14 oil phase-2 53.98 15.82 0.14 1.24 Ave. 53.72 14.96 0.16 0.88 std.dev 0.26 0.87 0.02 0.37 BO-14 aqueous phase-1 18.66 20.83 0.09 1.11 BO-14 aqueous phase-2 18.86 15.95 0.14 0.56 Ave. 18.76 18.39 0.12 0.84 std.dev 0.10 2.44 0.03 0.28 BO-17 oil phase-1 73.78 12.30 0.24 0.09 BO-17 oil phase-2 80.40 13.40 0.15 1.23 Ave. 77.09 12.85 0.20 0.66 std.dev 3.31 0.55 0.05 0.57 BO-17 aqueous phase-1 6.36 24.52 0.02 1.52 BO-17 aqueous phase-2 6.27 22.83 0.03 1.86 Ave. 6.32 23.68 0.03 1.69 std.dev 0.05 0.85 0.01 0.17 BO-19 oil phase-1 78.76 13.13 0.16 1.37 BO-19 oil phase-2 78.66 13.11 0.16 1.46 Ave. 78.71 13.12 0.16 1.42 std.dev 0.05 0.01 0.00 0.04 BO-19 aqueous phase-1 6.27 18.85 0.08 0.76 BO-19 aqueous phase-2 6.79 24.90 0.02 1.37 Ave. 6.53 21.88 0.05 1.07 std.dev 0.26 3.03 0.03 0.31  G.3 Water Titration Repeatability Table G.3 presented several examples for H2O titration.   282  Table G.3: H2O titration results for different samples.   H2O titration results, wt% Sample name 1 2 Ave. Std. Dev. Crude bio-oil 25.67 25.67 25.67 0.00 Bo1 oil phase 17.16 16.19 16.68 0.49 Bo1 aqueous phase 52.71 50.40 51.56 1.16 Bo2 oil phase 43.39 43.50 43.45 0.06 Bo2 aqueous phase 55.29 52.98 54.14 1.16 Bo3 oil phase 0.77 0.52 0.65 0.13 Bo3 aqueous phase 86.48 85.71 86.10 0.39 Bo4 oil phase 13.34 14.05 13.70 0.36 Bo4 aqueous phase 66.05 67.41 66.73 0.68 Bo5 oil phase 17.16 16.19 16.68 0.49 Bo5 aqueous phase 52.71 50.4 51.56 1.16 Bo6 oil phase (upper) 0.85 0.20 0.53 0.33 Bo6 aqueous phase (middle) 75.95 74.46 75.21 0.75 Bo6 aqueous phase (down) 8.82 10.51 9.67 0.85 Bo7 oil phase 14.32 15.38 14.85 0.53 Bo7 aqueous phase 64.68 59.69 62.19 2.50 Bo8 oil phase 15.32 14.78 15.05 0.27 Bo8 aqueous phase 62.74 63.19 62.97 0.23 Bo9 oil phase 11.13 10.46 10.80 0.34 Bo9 aqueous phase 70.20 74.25 72.23 2.03  283  G.4 Statistical Analysis of Kinetic Model ANOVA analysis and F statistic test were done to check if the kinetic model fitted well with the experimental data. The probability (P) was set at 0.05. Table G.4 presents both experimental and kinetic model fitted data of different products and reactant in HDO of 4-MP by 10Mo2C/APC_R600 catalyst.  Table G.4: Comparation between experimental and kinetic model fitted data of different products and reactant in HDO of 4-MP by 10Mo2C/APC_R600 catalyst.  Time (min) 4-MP (mol/L) DDO products (mol/L) HYD products (mol/L) Experimental Model Exp. Model Exp. Model 60 2.16E-01 2.05E-01 3.83E-02 3.88E-02 3.22E-03 6.09E-03 180 1.42E-01 1.38E-01 1.02E-01 9.68E-02 1.32E-02 1.52E-02 300 8.31E-02 9.25E-02 1.49E-01 1.36E-01 2.49E-02 2.13E-02  The ANOVA analysis of 4-MP, DDO products and HYD products are reported in the Table G.5, Table G.6, and Table G.7 respectively. SS represents sum of squares (SSWG = ∑ (𝑋𝑖2𝑚𝑖=1 + 𝑥𝑖2) −(∑ 𝑋𝑖)2+(∑ 𝑥𝑖)2𝑚𝑖=1𝑚𝑖=1𝑚 ; SST = ∑ (𝑋𝑖2𝑚𝑖=1 + 𝑥𝑖2) −(∑ 𝑋𝑖+𝑥𝑖)2𝑚𝑖=1𝑚, m is the number of data; SSBG = SST - SSWG ). Df represents degree of freedom (dfWG = ∑ (𝑚 − 1)𝑖𝑧𝑖=1 , z means the set of data; dfWG = z – 1). MS represents Mean square (MSWG = 𝑆𝑆𝑊𝐺𝑑𝑓𝑊𝐺; MSBG = 𝑆𝑆𝐵𝐺𝑑𝑓𝐵𝐺 ). FANOVA = 𝑀𝑆𝐵𝐺𝑀𝑆𝑊𝐺.   284   Table G.5: ANOVA analysis of 4-MP concentration data in HDO of 4-MP by 10Mo2C/APC_R600 catalyst.  4-MP SS df MS FANOVA F(1,4) Between groups (BG) 6.98E-06 1 6.98E-06 1.84E-03 7.709 Within groups (WG) 1.52E-02 4 3.80E-03   Sum 1.52E-02     Note: F(1,4), P=0.05 was obtained from reference [220].   Table G.6: ANOVA analysis of DDO products concentration data in HDO of 4-MP by 10Mo2C/APC_R600 catalyst. DDO products SS df MS FANOVA F(1,4) Between groups 5.30E-05 1 5.30E-05 1.94E-02 7.709 Within groups 1.09E-02 4 2.74E-03   Sum 1.10E-02      Table G.7: ANOVA analysis of HYD products concentration data in HDO of 4-MP by 10Mo2C/APC_R600 catalyst. HYD products SS df MS FANOVA F(1,4) Between groups 2.46E-07 1 2.46E-07 2.79E-03 7.709 Within groups 3.54E-04 4 8.84E-05   Sum 3.54E-04      All the FANOVA is far less than F(1,4, P=0.05), it means that no significant difference between these two set of data can be observed, and model acquired data fitted well with experimental data. 285  Appendix H  Reactor Operation Procedures H.1 Standard Operation Procedure (SOP) of Batch Reactor Pre-start up checklist: • Wear appropriate PPE prior to operate the unit.  • Ensure all electrical connections are connected with the computer. • Ensure all gas cylinders (H2, N2) are in open status and also have enough pressure inside. • Ensure all flow controllers are set to be fully closed. • A leak test is required before starting the reaction.   Operation procedures: Leak test should be performed prior to each experiment. 1. Filled feed and catalyst into the reactor. 2. Close the reactor tightly by torque wrench to 30-40 ft lbs. 3. Flush the reactor with N2 before switching with H2. 4. Close all the valves connected with the reactor. 5. Pressurized the reactor to the operating pressure. 6. Leave for 30 min to observe if there is a pressure change. If there is no or few pressure change (ca. 5-10 psi), the operation can be proceed. Otherwise, leak should be checked for every fitting, valve and the gasket and redo step 2.  Experimental running procedure: 7. Prepare 100 mL liquid feed and catalyst. Notice: catalyst loading amount depends on experimental design. 8. Load the catalyst and liquid feed into the reactor vessel. Notice: Inspect and clean both the cover and vessel body seal grooves with soft cloth or tissue, or suitable solvent such as acetone. Sealing surface should be free of all foreign materials.  286  9. Place a clean seal gasket into the body seal groove. Notice: gasket needs to be changed if it is worn. 10. Place the cover onto seal carefully; align the threaded holes in the flange with the cap screw holes in the cover. 11. Thread the socket head cap screws into the flange through the cover about 4 turns. 12. Finger tightens all socket head cap screws with the appropriate hex Allen wrench. Notice: Keep the top of the vessel and flange parallel with the bottom of the cover. 13. Start with the torque wrench by setting at 5 ft lbs. Torque each cap screw to the new wrench setting in cross order. 14. Repeat step 13 till the torque wrench set at 40 ft lbs. 15. Close all valves indicated below by the red arrows.  16. Close the cooling water valve at the back of the reactor.  287   Notice: Prior to run the experiment, turn off the valve to ensure no cooling water flows inside the reactor.  17. Flush the reactor system with N2 under 50 psi for 10 min to remove the inside air. 18. Pressurized the reactor with N2 to the aimed pressure for a leak test (Part 3.2.1). If no leak happens, it can be proceed to the following steps. Otherwise, go back to Part 3.2.1. 19. Open the vent valve and release the pressure.  20. Flush the system with H2 gas in place of N2 for 15 min at ca. 20 psi. 21. Increase reactor pressure to the desired value. 22. Turn on cooling water of Magnetic Drive. 23. Set the heating mantle temperature from control tower.  288  Notice: Usually the set point of heater’s temperature is 15-25 oC higher than that of the process temperature.  24. Set the mixer speed from control tower.  25. Wear thermal jacket on the outside of the vessel. 26. Open software  from the computer desktop.    Notice: Ensure the software is communicated with the control tower.  27. Start the experiment by flipping the two red switches (as indicated in step 23). 28. Enter a running number to record your data.  29. Draw samples from the sideline of the reactor through operate the 4 valves below during reaction.   289   30. Open valve 1.  31. Turn valve 2 towards the reactor side to make liquid fill inside the small vessel. 32. Turn valve 2 off quickly. 33. Open valve 3 slowly to draw sample out. Notice: 1st and 2nd times of drawing samples are used to clean up the small vessel and lines. 34. Close valve 3. 35. Open valve 4 to the right-hand side and valve 2 to the opposite of the reactor direction. 36. Open valve 3 to use N2 blowing out the left samples. 37. Repeat step 31-36 after clean up the lines to draw a sample. 38. Record the temperature and pressure before and after drawing sample. 39. Weight the drawn sample mass.  Shut off procedures: 40. Turn off the mantle heater and take it off the reactor vessel. 41. Open cooling water valve as indicated in step 16. 42. Slow down the stirring speed to 500 rpm. 43. Notice: Once the temperature is completely cooling down, turn off the stirring speed. 44. Release the reactor pressure and collect gas sample. 45. Open reactor by torque wrench and recover the catalyst.  290  46. Generate a report after done with the experiment by .   47. Check cooling water, electricity and heater to ensure all of them are turned off.  H.2 Standard Operation Procedure (SOP) of Trickle Bed Reactor Pre-Start up checklist: • Wear appropriate PPE prior to operate the unit.  • Ensure H2S scrubber is filled with 1M NaOH fresh solution of 250-300 mL before HDS experiment. • Ensure all gas cylinders (H2, He) are in open status and also have enough pressure to operate the reaction (Pgas bottle  > Prxn). • Ensure all flow controllers are set to be fully closed. • Ensure the feed level is higher than the pump filter and no bubble in the feed inlet line. • A leak test is required before starting the reaction   Operation procedures: Catalyst bed packing: 1. Take the reactor off the furnace by disconnecting 1/8’’ fittings both on the top and bottle of the reactor. 2. Cut the used reactor in the labeled part to recover the used catalyst. Notice: The cut should extremely follow the labeled part to avoid damage the thermocouple.   291  3. Prepare a new 3/8’’ Swagelok stainless steel tube with 50 cm for reaction use. 4. Load different stuffs at different layers to increase the liquid distribution.  Stuff materials Volume (mL) Glass beads 3 30 mesh SiC 2 46 mesh SiC 2 80 mesh SiC 2 100 mesh SiC 1-1.5 Catalyst bed 2.5 100 mesh SiC 1-1.5 80 mesh SiC 2 46 mesh SiC 2 30 mesh SiC 2 Glass beads 3 Notice: (a) Quartz wools is inserted between each layer; (b) The amount of catalyst is varied due to different experimental design. 5. Connect the reactor back to the system. 6. Ensure the thermocouple is connected to the top of the reactor. 7. Put quartz wools on the top and bottom of the reactor after leak test.  Leak test: (Method 1) 8. Open BPR thoroughly;  9. Open V2, V6, V5, V7, V8; then set a flow rate on MFC-He (software: Brooks Smart Interface) to make He flush the whole system for 5 mins at least; 10. Adjust BPR to the operation pressure; then close V2 and set 0 SCC/min on the interface and leave there for 15 min to check the leakage.   292  If there is no leakage, open BPR to release the system pressure. 11. Open V1 and use H2 to flush the system for 5-10 mins;  12. Adjust BPR to make the pressure increase to the operation pressure to start the experiment; close V5 and V8.  (Method 2) 12. Repeat Step 8 and 9 in Method 1 to flush the system with He. 13. Adjust BPR to the operation pressure. 14. Ensure to open V6, V7; close V1, V2, V5, V9. 15. Open V3 gently and slowly to pressurize the reactor to the operation pressure and then close it quickly. Notice: The real-time pressure can be presented on the pressure gauge (P01). 16. Apply leak detector to check the joint parts of every fitting Notice: Special attention should be paid on the parts which connected with the reactor.  Hydrotreating experiments: 17. Pressurized the reactor with H2 to the operation pressure before pumping the liquid solution inside the reactor.  18. Open  software to monitor the pressure. If recording is needed, goes to “data log”→”Start”       19. Open  software to monitor the pressure drop. Start the real time monitoring by pressing .   293  20. Communicate the furnace with the software . 21. Persse “go online now”→ →start setting the temperature program.   22. Open V1 and start  software. 23. Open “Controls”→”Initialize”→”Use previous configuration”  24. Set the used flow rate in the red rectangle and press Enter.   25. Open liquid pump by pressing Pump and Run; Also, open the three valve connected with the reactor as well.  294   Notice: An appropriate flow rate should be entered before pumping feed. The maximum flow rate is 5.0 mL/min. In the meanwhile, flush the injecting line by feeds before the experiment.  26. The frequent of sample drawing depends on your experimental design. Notice: the starting weight of feed should be read from the balance. 27. Start running the experiment by initiating the three parts of FURNACE, PUMP and GAS mentioned above. 28. Open V5 and V8; then close V6 and V7; open V9 to make the sample flow from condenser1 to condenser2.  29. Close V9 quickly. 30. Open V3 to make H2 fill into condenser1 and close V3 until P1 reach to the same pressure of P2. 31. Open V6 and V7 and close V5, V8 to make condenser1 connect into the system again 32. Open V10 slowly and gently to collect the liquid samples. Notice: A plastic tube near condensers will be applied to ventilate the gas.  295    The operation of step 28-31 can also be done by step 33-38. 33.  Close H2 flow rate from the software; 34.  Stop the pump by pressing “Pump” and “Run” buttons. 35.  Close V6 and V7; then open V9 quickly. 36.  Close V9; open V3 to fill Condenser1 with H2 to the system pressure. 37.  Close V3 and open V6 and V7. 38.  Start the pump and H2 flow rate. 39.  Read and record the mass of feed from the balance.  Shut off procedures: 40.  Close liquid pump. 41.  Turn off the furnace heating from the software.  42.  Pull out quartz wools from top and bottom of the reactor. 43.  Set H2 flow rate to zero and then close V1.  296  44.  Open V13 slowly to release all the waste sample from bypass line if step 31-34 is proceed.  Close BPR to release the system pressure. 

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.24.1-0380462/manifest

Comment

Related Items