Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies on lignin model compounds upgrading with in-situ glycerol aqueous phase reforming and the application… Dhewangga Putra, Robertus Dhimas 2019

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

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

Item Metadata

Download

Media
24-ubc_2020_may_dhewanggaputra_robertusdhimas.pdf [ 6.65MB ]
Metadata
JSON: 24-1.0387158.json
JSON-LD: 24-1.0387158-ld.json
RDF/XML (Pretty): 24-1.0387158-rdf.xml
RDF/JSON: 24-1.0387158-rdf.json
Turtle: 24-1.0387158-turtle.txt
N-Triples: 24-1.0387158-rdf-ntriples.txt
Original Record: 24-1.0387158-source.json
Full Text
24-1.0387158-fulltext.txt
Citation
24-1.0387158.ris

Full Text

Studies on Lignin Model Compounds Upgrading with In-situ Glycerol Aqueous Phase Reforming and the Application for Upgrading of Ligneous Material by  Robertus Dhimas Dhewangga Putra  M.Eng., University of Science and Technology (Korea), 2013 B.Eng., Gadjah Mada University, 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)   December 2019  © Robertus Dhimas Dhewangga Putra, 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:  Studies on Lignin Model Compounds Upgrading with In-situ Glycerol Aqueous Phase Reforming and the Application for Upgrading of Ligneous Material  submitted by Robertus Dhimas Dhewangga Putra in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Chemical and Biological Engineering  Examining Committee: Heather Trajano, Chemical and Biological Engineering Co-supervisor Chang Soo Kim, KIST-UBC Biorefinery on-site Laboratory Co-supervisor  Scott Renneckar, Wood Science Supervisory Committee Member Fariborz Taghipour, Chemical and Biological Engineering University Examiner Dana Grecov, Mechanical Engineering  University Examiner    Roberto Rinaldi, Chemical Engineering – Imperial College London  External Examiner  Additional Supervisory Committee Members: Kevin J. Smith, Chemical and Biological Engineering Supervisory Committee Member    iii  Abstract Lignin and glycerol, residues of renewable biomass processing, have significant potential as fuels and chemicals. Lignin is a polymer of phenylpropanoids monomers and is a promising source of renewable hydrocarbons due to its relatively high C/O ratio compared to carbohydrates. However, it also requires hydrogenation for further valorization. Unfortunately, hydrogen currently comes primarily from petroleum, natural gas, and coal. Aqueous phase reforming (APR) of glycerol is a renewable source of hydrogen. This relatively low temperature reforming reaction is thermodynamically possible due to the presence of a C-O bond on every carbon of glycerol. This thesis explores the possibility of lignin depolymerization and fast pyrolysis oil (FPO) hydrogenation using renewable hydrogen from glycerol. This study was conducted with phenol as a model compound. Upgrading more complex materials such as FPO and native lignin from crushed mixed spruce, pine, and fir (SPF) pellets was also tested. Operating conditions were varied in order to understand reaction mechanisms.   First, glycerol APR was conducted with Raney Ni® and it was found that glycerol APR occurred via parallel reactions of 1,2-propylene glycol and ethylene glycol. During glycerol APR, CO2 and CH4 were the dominant gaseous products while the produced hydrogen tended to react with glycerol, glycerol intermediates (direct methanation) or CO2 (Sabatier) to form CH4. The presence of phenol during glycerol APR increased the glycerol reaction rate and CO2/CH4 ratio due to the consumption of hydrogen, and produced cyclohexanol, cyclohexanone, and benzene.  Phenol hydrogenation during in-situ glycerol aqueous phase reforming and phenol hydrogenation (IGAPH) occurred without the formation of molecular hydrogen as the hydrogen produced by glycerol APR was consumed by phenol before molecular hydrogen could form and desorb from iv  the catalyst surface. The mechanism of phenol hydrogenation during IGAPH is hypothesized to follow the Langmuir-Hinshelwood mechanism. Hydrodeoxygenation (HDO) of phenol could be achieved using the combination of hydrogenation (Raney Ni® and Pt/C) and acid catalysts (Amberlyst-15 and H-ZSM-5). During FPO and SPF upgrading with Pt/C and H-ZSM-5, n-decane was used to separate nonpolar deoxygenated products from very reactive carbohydrates derivatives to prevent condensation reactions. Gasoline-like compounds were obtained from FPO and SPF upgrading.   v  Lay Summary  Wood derivatives are promising renewable and carbon neutral substitutes for petroleum. Lignin, a low oxygen polymer in wood, is currently underutilized. Breaking down and removing oxygen from lignin to obtain gasoline-like compounds, known as lignin upgrading, requires substantial amounts of hydrogen, which can be renewably produced through aqueous phase reforming (APR) of glycerol. In this work, a simultaneous process of lignin upgrading and glycerol APR was studied. The initial study with phenol, a model compound for lignin monomer, revealed synergistic effects during the simultaneous reactions. The presence of phenol increased glycerol APR selectivity while phenol hydrogenation could proceed without excess glycerol. Phenol hydrodeoxygenation product selectivity was controllable by altering glycerol loading and the ratio of acid to hydrogenation catalyst. Upgrading fast pyrolysis oil and a mix of spruce, pine and fir required glycerol and a hydrogenation catalyst as well as the addition of a nonpolar co-solvent to obtain gasoline-like compounds.     vi  Preface  This thesis contains 8 chapters. A portion of chapter 5 has been published in a peer reviewed journal. Another portion of chapter 5 and all of chapter 4 are under preparation for publication in a peer reviewed journal. Chapter 6 and 7 are in early preparation for peer reviewed journal publication. Literature review, experimental work, and numerical analysis were conducted by Robertus D. D. Putra under the guidance of Dr. Chang Soo Kim and Dr. Heather L. Trajano. The calibration of GC-FID-TCD for gas analysis was conducted by Shida Liu. All the experimental work was conducted in the University of British Columbia. The thesis and papers were written primarily by Robertus D. D. Putra under supervision and approval of Dr. Chang Soo Kim and Dr. Heather L. Trajano.  The list of publications and author contributions are as follows: Putra, R. D. D.; Trajano, H. L.; Liu, S.; Lee, H.; Smith, K.; Kim, C. S. In-Situ Glycerol Aqueous Phase Reforming and Phenol Hydrogenation over Raney Ni®. Chem. Eng. J. 2018, 350 (May), 181–191.  The manuscript was prepared and written by Robertus D. D. Putra under supervision of Dr. Chang Soo Kim and Dr. Heather L. Trajano. Gas analysis was conducted by Shida Liu. Dr. Kevin Smith contributed to the discussion of thermodynamic prediction of phenol conversion. Dr. Hyunjoo Lee contributed to the GC-MS quantification method.   vii  Table of Contents   Abstract .......................................................................................................................................... iii Lay Summary .................................................................................................................................. v Preface............................................................................................................................................ vi Table of Contents .......................................................................................................................... vii List of Tables ............................................................................................................................... xiv List of Figures ............................................................................................................................. xvii Nomenclatures ............................................................................................................................. xxi List of Abbreviations ................................................................................................................. xxiii Acknowledgements .................................................................................................................... xxvi Chapter 1: Introduction ................................................................................................................... 1 1.1 Objective .................................................................................................................... 5 1.2 Approach ................................................................................................................... 6 Chapter 2: State of the Art .............................................................................................................. 8 2.1 Lignocellulose structure ............................................................................................ 8 2.1.1 Cellulose ................................................................................................................. 9 2.1.2 Hemicellulose ....................................................................................................... 10 2.1.3 Lignin .................................................................................................................... 11 viii  2.2 Lignin isolation ........................................................................................................ 13 2.2.1 Alkaline treatment (Kraft pulping) ....................................................................... 13 2.2.2 Acid hydrolysis ..................................................................................................... 14 2.2.1 Organosolv process (dilute acid process) ............................................................. 15 2.3 Pyrolysis and liquefaction of lignocellulose ............................................................ 17 2.4 Upgrading lignin-derived compounds ..................................................................... 18 2.4.1 The catalyst’s role ................................................................................................. 19 2.4.1.1 Hydrogenation............................................................................................... 19 2.4.1.2 Hydrodeoxygenation ..................................................................................... 23 2.4.2 The role of solvents ............................................................................................... 27 2.4.2.1 Solvent hydrogen donor ................................................................................ 27 2.4.2.2 Enhancement of reaction performance with solvent properties .................... 30 2.4.3 The challenges in lignin and fast pyrolysis oil (FPO) upgrading ......................... 31 2.4.3.1 Thermal condensation of FPO ...................................................................... 31 2.4.3.2 Inhibition of catalytic activity ....................................................................... 34 2.5 Aqueous-phase reforming of glycerol ..................................................................... 35 2.6 Summary .................................................................................................................. 38 Chapter 3: Experimental ............................................................................................................... 40 3.1 Materials .................................................................................................................. 40 3.1.1 Solvents, gas and standard .................................................................................... 40 ix  3.1.2 Ligneous materials and model compounds ........................................................... 40 3.1.3 Catalysts ................................................................................................................ 41 3.2 Experiments procedures .......................................................................................... 42 3.2.1 Reactor configuration............................................................................................ 42 3.2.2 Aqueous phase reforming (APR) .......................................................................... 43 3.2.3 In-situ glycerol APR and phenol hydrogenation (IGAPH)................................... 44 3.2.4 In-situ glycerol aqueous phase reforming and phenol hydrodeoxygenation (IGAPHdo) ............................................................................................................ 46 3.2.5 Reaction with FPO and SPF ................................................................................. 47 3.3 Product analysis of IGAPH and IGAPHdo ............................................................. 48 3.3.1 GC-MS analysis for phenolic compounds ............................................................ 48 3.3.2 HPLC analysis ...................................................................................................... 49 3.3.3 GC-FID-TCD for gas analysis .............................................................................. 49 3.4 Lignin, carbohydrate, and ash analysis .................................................................... 49 3.5 Reaction metrics ...................................................................................................... 51 3.5.1 Reaction metrics for model compounds experiment ............................................ 51 3.5.2 Reaction metrics for the experiment with FPO and SPF ...................................... 55 Chapter 4: Glycerol Aqueous Phase Reforming ........................................................................... 56 4.1 Introduction ............................................................................................................. 56 4.2 Results ..................................................................................................................... 57 x  4.2.1 APR of glycerol and intermediate products .......................................................... 57 4.2.2 Time dependence of glycerol APR ....................................................................... 59 4.3 Discussion ................................................................................................................ 61 4.3.1 Glycerol APR reaction pathway ........................................................................... 61 4.3.2 Kinetics analysis of glycerol APR ........................................................................ 66 4.3.3 Undesired methane formation ............................................................................... 69 4.4 Conclusion ............................................................................................................... 70 Chapter 5: In-situ Glycerol Aqueous Phase Reforming and Phenol Hydrogenation .................... 71 5.1 Introduction ............................................................................................................. 71 5.2 Results and discussion ............................................................................................. 72 5.2.1 Thermodynamic analysis of IGAPH reaction ....................................................... 72 5.2.2 Temperature effect on IGAPH .............................................................................. 74 5.2.3 The role of phenol during IGAPH ........................................................................ 77 5.2.3.1 The effect of phenol during APR of glycerol and glycerol intermediates .... 77 5.2.3.2 The effect of phenol loading on glycerol conversion and product   distribution ................................................................................................... 81 5.2.4 The role of glycerol during IGAPH ...................................................................... 85 5.2.5 The effect of reaction time on IGAPH .................................................................. 88 5.2.6 Determination of mechanism of phenol hydrogenation........................................ 90 5.2.7 Examination of methanation using CO2/CH4 ratio ............................................... 96 xi  5.2.8 Effect of catalyst loading ...................................................................................... 98 5.2.9 Reaction with several lignin model compounds ................................................. 100 5.2.10 Reaction with noble metal catalysts .................................................................... 104 5.2.11 Phenol hydrogenation mechanism and effect of He flow during IGAPH with   Pt/C ..................................................................................................................... 106 5.3 Kinetics of IGAPH ................................................................................................ 108 5.4 Conclusion ............................................................................................................. 111 Chapter 6: In-situ Glycerol Aqueous Phase Reforming and Phenol Hydrodeoxygenation (IGAPHdo) ............................................................................................................ 113 6.1 Introduction ........................................................................................................... 113 6.2 Results ................................................................................................................... 114 6.2.1 Phenol HDO with Raney Ni® and several acid catalysts .................................... 114 6.2.1.1 Acid catalyst screening ............................................................................... 114 6.2.1.2 IGAPHdo with Raney Ni® and H-ZSM-5 .................................................. 117 6.2.1.3 Reaction with the intermediates .................................................................. 119 6.2.2 Controlling phenol HDO products with Pt/C...................................................... 121 6.2.2.1 Targeting benzene production..................................................................... 125 6.2.2.2 Targeting cyclohexane production .............................................................. 128 6.3 Discussion .............................................................................................................. 129 6.3.1 IGAPHdo with Raney Ni® .................................................................................. 129 xii  6.3.2 Controlling selectivity of products with Pt/C ..................................................... 130 6.4 Conclusion ............................................................................................................. 131 Chapter 7: Upgrading of Ligneous Material via In-Situ Glycerol APR ..................................... 133 7.1 Introduction ........................................................................................................... 133 7.2 Results ................................................................................................................... 134 7.2.1 FPO composition analysis................................................................................... 134 7.2.2 FPO upgrading .................................................................................................... 138 7.2.3 Hydrogenolysis of SPF ....................................................................................... 147 7.3 Discussion .............................................................................................................. 158 7.3.1 FPO and SPF composition .................................................................................. 158 7.3.2 Carbohydrate degradation ................................................................................... 159 7.3.3 The effect of a co-solvent ................................................................................... 161 7.3.4 Lignin depolymerization ..................................................................................... 163 7.3.5 Lignin monomers in SPF and BTG FPO ............................................................ 165 7.3.6 HDO of lignin-derived compounds .................................................................... 166 7.3.7 O/C and H/C ratios in the products ..................................................................... 171 7.4 Conclusion ............................................................................................................. 174 Chapter 8: Conclusions and Recommendations ......................................................................... 176 8.1 Conclusions ........................................................................................................... 176 8.2 Recommendations ................................................................................................. 178 xiii  8.2.1 Catalysts improvement........................................................................................ 178 8.2.2 Continuous process ............................................................................................. 179 8.2.3 Renewable co-solvents........................................................................................ 179 Bibliography ............................................................................................................................... 180 Appendices .................................................................................................................................. 198 Appendix A : Theoretical calculations ................................................................................ 198 A.1 Minimum CO2/CH4 during APR ........................................................................ 198 A.2 Thermodynamics calculation .............................................................................. 205 A.3 Thermodynamic prediction of maximum cyclohexanol yield and the MATLAB code ..................................................................................................................... 208 A.4 Predicted saturation pressure by Antoine equation:............................................ 211 A.5 Predicted of potential CO2 that was converted to CH4 ....................................... 212 A.6 Estimation of internal mass transfer effect. ........................................................ 213 Appendix B : Kinetics.......................................................................................................... 215 B.1 Kinetic of glycerol reaction and the MATLAB code ......................................... 215 B.2 Kinetics of IGAPH .............................................................................................. 220 Appendix C : Supplementary data ....................................................................................... 224 Appendix D Supplementary data for FPO upgrading and SPF hydrogenolysis .................. 234 D.1 GC-MS chromatographs ..................................................................................... 234 D.2 Yield of lignin-derived compounds .................................................................... 238 D.3 Selectivity of gas product.................................................................................... 242 D.4 Mass balance for FPO upgrading and SPF hydrogenolysis ................................ 244 Appendix E : Schemes of predicted reaction mechanisms .................................................. 247 xiv  List of Tables Table 2.1  Lignocellulose composition of several types of biomass .......................................... 9 Table 2.2  Examples of upgrading with reduced coke formation ............................................. 33 Table 3.1  Catalysts properties .................................................................................................. 41 Table 4.1  Aqueous phase reforming of glycerol and glycerol APR intermediate products .... 58 Table 4.2  Enthalpy and entropy of formation of compounds involved in glycerol APR reaction .................................................................................................................... 64 Table 4.3  Rate constant of glycerol APR reaction at 220°C with 1st order ............................ 69 Table 5.1  Conversion and product of glycerol and glycerol APR intermediates during  IGAPH ..................................................................................................................... 80 Table 5.2 Phenol hydrogenation products and selectivity by hydrogen source ...................... 80 Table 5.3  Phenol conversion and product selectivity from closed system hydrogenation with formic acid. .............................................................................................................. 92 Table 5.4  Hydrogenation of lignin model compound with in situ glycerol APR .................. 101 Table 5.5.  Reaction metrics of APR and IGAPH in open and closed system .......................... 97 Table 6.1  Reaction of IGAPHdo phenol products.. ............................................................... 120 Table 6.2  IGAPHdo reaction with Pt/C catalyst. ................................................................... 124 Table 6.3  IGAPHdo reaction with Pt/C catalyst to produce benzene. ................................... 126 Table 6.4  IGAPHdo reaction with Pt/C catalyst to produce cyclohexane ............................. 127 Table 7.1  Composition of BTG FPO and pyrolytic lignin .................................................... 137 Table 7.2   Quantitative results for upgraded FPO using Pt/C and H-ZSM-5 catalysts .......... 139 Table 7.3  Quantitative results for upgraded FPO using Pt/C catalysts with the addition of n-decane as a co-solvent ........................................................................................... 143 xv  Table 7.4  Quantitative results for upgraded FPO using Pt/C and H-ZSM-5 catalysts with the addition of n-decane as a co-solvent for 7 h of reaction time ................................ 146 Table 7.5  Compositional analysis of SPF sawdust ................................................................ 147 Table 7.6   Quantitative results of SPF hydrogenolysis using Pt/C and H-ZSM-5 catalysts .. 150 Table 7.7  Quantitative results of SPF hydrogenolysis using Pt/C catalyst ............................ 154 Table 7.8  Sugar yields from the hydrogenolysis of SPF, before and after compositional analysis .................................................................................................................. 155 Table 7.9  Liquid and gas products from cellulose APR reaction .......................................... 157 Table 7.10   Evolution of carbohydrate and lignin constituents during FPO upgrading and SPF hydrogenolysis ....................................................................................................... 165 Table A.1  Thermodynamic properties of possible reactions during glycerol APR. ............... 205 Table A.2  Thermodynamic properties of possible reactions with phenol .............................. 207 Table A.3  Component-specific constant for Antoine equations ............................................. 211 Table B.1  Parameter value of glycerol kinetic model during glycerol APR with several initial guess ...................................................................................................................... 218 Table B.2  Parameter value of glycerol kinetic model during IGAPH with several initial    guess ...................................................................................................................... 219 Table B.3  Reaction rate equations resulting from Langmuir-Hinshelwood model and assumed rate limiting step. ................................................................................................... 220 Table C.1  Complete conversion and yield data ...................................................................... 224 Table C.2  Gas composition on APR and IGAPH in open system in 30 min intervals .......... 225 Table C.3  Phenol carbon balance and hydrogen balance of reaction in Fig 5.2 .................... 226 Table C.4  Phenol carbon balance and hydrogen balance of reaction in Fig 5.5 .................... 226 xvi  Table C.5  Phenol carbon balance and hydrogen balance of reaction in Fig 5.10 .................. 227 Table C.6 Phenol carbon balance and hydrogen balance of reaction in Fig 5.11 .................. 227 Table C.7  Phenol carbon balance and hydrogen balance of reaction in Fig 6.1 .................... 228 Table C.8  Phenol carbon balance and hydrogen balance of reaction in Fig 6.2 .................... 229 Table D.1  Major phenolic compounds in BTG FPO .............................................................. 238 Table D.2   Selectivity of gas products during BTG FPO upgrading ....................................... 242 Table D.3   Selectivity of gas products during SPF hydrogenolysis ........................................ 243 Table D.4  Yield of solid, liquid and gas products of FPO and SPF upgrading ...................... 246  xvii  List of Figures Figure 2.1  Canonical monolignols (a) and non-canonical monolignols (b) with their polymerisation.......................................................................................................... 12 Figure 2.2  Energy barrier during phenol hydrogenation with Pt and Pd .................................. 21 Figure 2.3  Direct hydrogen transfer (a), monohydride mechanism (b), and dihydride mechanism (c). ......................................................................................................... 29 Figure 2.4  The thermodynamic equilibrium constants of reforming-related reactions ............. 37 Figure 4.1  Reaction metrics from glycerol APR at 220 °C as a function of reaction time.. ..... 60 Figure 4.2  Predicted reaction pathways of glycerol APR. Compounds in red are hypothetical compounds that were not detected ........................................................................... 65 Figure 4.3  Searching for reaction order of glycerol APR ......................................................... 66 Figure 4.4  Evolution of glycerol, 1,2-propylene glycol, ethylene glycol, and subsequent products during glycerol APR at 220 °C experimentally and as predicted by Equations 4.5 – 4.9................................................................................................... 68 Figure 5.1  Equilibrium constant (ln K = -ΔG°/(RTnC)) of several reactions involved in IGAPH calculated per mol carbon on the compound. .......................................................... 73 Figure 5.2   Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 180 °C to 240 °C for 2 h ........................................................................................................... 76 Figure 5.3  Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C for 2 h as a function of ratio of phenol-to-glycerol ....................................................... 82 Figure 5.4  Plot searching for glycerol reaction order during IGAPH ....................................... 84 Figure 5.5  Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C for 2 h as a function of ratio of phenol-to-glycerol ....................................................... 86 xviii  Figure 5.6  Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C as a function of reaction time ....................................................................................... 89 Figure 5.7  Comparison of phenol conversion with products selectivity (a) and glycerol conversion and products selectivity (b) during IGAPH in closed and open systems with different phenol to glycerol ratio ..................................................................... 91 Figure 5.8  Pressure and temperature profile of closed system IGAPH with an initial pressure of 0.69 MPa (a) and 3.45 MPa (b). Phenol hydrogenation using formic acid with an initial pressure of 3.45 MPa with catalyst (c) and without catalyst (d). .................. 93 Figure 5.9  Comparison of phenol conversion with products selectivity (a) and glycerol conversion and products selectivity (b) during IGAPH in closed and open systems with different initial and reaction pressures ............................................................. 94 Figure 5.10  Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C for 2 h as a function of Raney Ni® catalyst loading ....................................................... 99 Figure 5.11  IGAPH reaction with different catalysts at 220 °C for 2 h with 3.45 MPa initial He pressure. ................................................................................................................. 105 Figure 5.12  IGAPH reaction with 5 wt% Pt/C in closed and open reactor configurations at 220°C for 2h ........................................................................................................... 107 Figure 5.13  Schematic of in-situ glycerol APR and phenol hydrogenation reaction     mechanism. ............................................................................................................ 111 Figure 6.1  IGAPHdo reaction with several acid catalysts at 240 °C for 2 h with 3.45 MPa initial He pressure .................................................................................................. 116 Figure 6.2  IGAPHdo reaction H-ZSM-5 at 240 °C for 2 h with 3.45 MPa initial He       pressure .................................................................................................................. 118 xix  Figure 6.3  Schematic IGAPHdo reaction with Raney Ni® and H-ZSM-5 catalyst. ................ 121 Figure 6.4  Hydrodeoxygenation reaction of phenol ................................................................ 122 Figure 7.1  Chromatograph and the appearance of BTG FPO. IS: internal standard. .............. 136 Figure 7.2  Chromatographs and appearances of upgraded BTG FPO with Pt/C .................... 142 Figure 7.3  Chromatographs and appearances of upgraded BTG FPO with Pt/C, and                H-ZSM-5 ................................................................................................................ 145 Figure 7.4  Chromatographs and appearance of SPF hydrogenolysis products with Pt/C ....... 152 Figure 7.5  Pressure profile of APR reaction using 5.00 g SPF without Pt/C (a), 3.00 g Avicel® cellulose (b), and 6.00 g Avicel® (c) ...................................................................... 156 Figure 7.6  Composition of BTG FPO (a) and SPF (b) in weight percentages. ....................... 158 Figure 7.7  Yield of p-creosol, p-cresol, 4-methylcyclohexanol, toluene, and methylcyclohexane during HDO of BTG FPO ...................................................... 167 Figure 7.8  Yields of cerulignol, 4-propylphenol, propylbenzene, and propylcyclohexane during SPF hydrogenolysis ............................................................................................... 169 Figure 7.9  Van Krevelen diagram of upgraded BTG FPO...................................................... 172 Figure 7.10  Van Krevelen diagram of SPF hydrogenolysis ...................................................... 174 Figure B.1  Plot of kinetic model of glycerol reaction during IGAPH with pseudo 1st order reaction model ....................................................................................................... 219 Figure D.1  Chromatograph and the appearance of upgraded BTG FPO without n-decane.. .. 234 Figure D.2  Chromatograph and the appearance of upgraded BTG FPO for 2.5 h. ................. 235 Figure D.3  Chromatograph and the appearance of upgraded BTG FPO with acetone. ........... 236 Figure D.4  Chromatograph and the appearance of Avicel® cellulose hydrogenolysis. .......... 237 xx  Figure D.5  Yield of phenolics during BTG FPO upgrading (10.0 g) with 5.0 g n-decane at 300°C. .................................................................................................................... 238 Figure D.6  Yield of oxygenated cycloaliphatics, naphthenes, and aromatics during BTG FPO upgrading (10.0 g) with 5.0 g n-decane at 300°C. ................................................. 239 Figure D.7  Yield of phenolics during SPF hydrogenolysis (5.0 g) upgrading with 5.0 g n-decane and 10 g water at 300°C. ........................................................................... 240 Figure D.8  Yield of oxygenated cycloaliphatics, naphthenes, and aromatics during SPF hydrogenolysis (5.0 g) upgrading with 5.0 g n-decane and 10 g water at 300°C.. 241 Figure E.1  Reactivity of β-O-4 linkage in lignin in alkaline solution ..................................... 247 Figure E.2  Reactivity of β-O-4 linkage of lignin in acidic solution ........................................ 248 Scheme 3.1  Reactor assembly for open system reaction. ........................................................... 43 Scheme 4.1  Reaction model for kinetic analysis ......................................................................... 67    xxi  Nomenclatures  D Deuterium, Da Dalton (1 g mol-1) 𝐷𝑏𝑢𝑙𝑘  Bulk diffusivity (m2 s-1) 𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑    Knudsen effective diffusivity (m2 s-1) 𝐷𝑒𝑓𝑓,𝑏𝑢𝑙𝑘  Bulk effective diffusivity (m2 s-1) 𝐷𝑔𝑙𝑦−𝑤, 𝐷𝑤−𝑔𝑙𝑦  Glycerol/water intradiffusion coefficient (m2 s-1) 𝐷 𝑘𝑛𝑢𝑑   Knudsen diffusivity (m2 s-1) 𝑑𝑝𝑎𝑟  Particle diameter (nm) 𝑑𝑝𝑜𝑟  Pore diameter (nm) K Thermodynamic equilibrium constant   Reaction constant, 1st order reaction: k (min-1 gcat-1) 0.66th order reaction: k (mmol0.34 L-0.34 gcat-1 min-1) L Ligand mf Final weight (g) mo Initial weight (g) n Number of mol (mol) O/C Oxygen to carbon atomic ratio (mol/mol) P/G Molar ratio of phenol to glycerol (mol/mol) Pt/C Platinum on carbon R1, R2 Alkyl xxii  R Ideal gas constant (8.314 J K-1mol-1)  S Selectivity (mol %) SP/G Consumption of hydrogen by phenol hydrogenation relative to glycerol hydrogenolysis STP Standard temperature and pressure (20°C, 1 atm) Vpore Pore volume (mL/g)  W Weight (g) X Conversion (mol %) 𝑥𝑔𝑙𝑦−𝑤, 𝑥𝑤−𝑔𝑙𝑦 Mol fraction of glycerol/ water Y Yield (mol %) ΔGr Gibbs free energy of reaction (kJ mol-1)  Enthalpy of formation at STP (kJ mol-1)  Entropy of formation at STP (kJ mol-1 K-1) Greek Symbol 𝜎  Constriction factor  ∅  Porosity ∅1  Thiele modulus 𝜌  Density (g/mL) 𝜂  Internal effectiveness factor  τ Tortuosity  xxiii  List of Abbreviations  1,2 PG 1,2-Propylene glycol 2-PrOH 2-Propanol AcH Acetic acid APR Aqueous phase reforming CA Compositional analysis CO2/CH4 Molar ratio of carbon dioxide to methane (mol/mol) CTH Catalytic transfer hydrogenation DDO Direct deoxygenation DMSO Dimethyl sulfoxide EG Ethylene glycol EtOH Ethanol FAME Fatty acid methyl ester FDCA 2,5-Furan dicarboxylic acid FPO Fast pyrolysis oil GC-FID-TCD Gas chromatography coupled flame ionization detector and thermal conductivity detector GC-MS Gas chromatography coupled mass spectroscopy detector GTP Glycerol thermal processing GVL γ-Valerolactone H/C Hydrogen to carbon atomic ratio (mol/mol) H-BEA Zeolite β xxiv  HDO Hydrodeoxygenation HMF 5-(Hydroxymethyl)furfural HPLC High performance liquid chromatography IEA International Energy Agency IGAPH In-situ glycerol APR and phenol hydrogenation IGAPHdo In-situ glycerol APR and phenol hydrodeoxygenation LCC Lignin-carbohydrate complex LMW Low molecular weight MeOH Methanol Mo Molybdenum MPV Meerwein–Ponndorf–Verley mechanism MWCNT Multiwall carbon nanotube NAC Nitric acid-treated carbon black NCC Nanocrystalline cellulose Ni Nickel NREL National Renewable Energy Laboratory PEF Polyethylene 2,5-furan dicarboxylate PET Polyethylene terephthalate PLA Polylactic acid P/G Phenol to Glycerol ratio RI Refractive index SMR Steam methane reforming SPF Crushed pellets contain mixture of Spruce, Pine, and Fir xxv  SR Solid residue THF Tetrahydrofuran TOF Turnover frequency UV Ultraviolet WGS Water gas shift WIBO Water insoluble bio-oil XRD X-ray diffraction     xxvi  Acknowledgements   I would like to thank my Ph.D. supervisors, Professor Chang Soo Kim and Professor Heather L. Trajano, for giving me the opportunity to conduct my Ph.D. under their guidance. It was a gratifying experience to have supervisors who were patient, supportive, insightful, and open minded. They taught me to be more independent and creative in research.    I would like to thank my Ph.D. committee, Professor Kevin J. Smith and Professor Scott Renneckar for allowing me to utilize their laboratory equipment to conduct my research and most importantly for their valuable suggestion and discussion.   I would like to thank Marlene Chow, Richard Ryoo, Miles Garcia, Brittany Ji, Kristi Chow, Doug Yuen and all Chemical and Biological Engineering staff for their help during my study.  I would like to express my gratitude to KIST on site laboratory members: Dr. Kwang Ho Kim, Rebecca Leung, Jacqueline Lee, Robin Lai, Andy Chang, Sarah Mahmoud, Oscar Jansen and all KIST laboratory co-op students for helping me during the research.   I would also thank CERC 207 lab members, Jingqian Chen, Varun Rangu, Anuradha Ramachandran, and other members of the CHBE catalysis group, Shida Liu, Haiyan Wang, Cloribel, Majed, and other members for the supportive collaborations.  xxvii  Special thanks for Philip Wijaya and Felix Arie for the discussion in fundamental knowledge both in chemical engineering and life.  Finally, I would also like to thank my parents (Unggul Sudarmo and Nanik Mitayani) and sisters (Dian and Dika) for their wishes, loves, and endless support throughout my study.  1  Chapter 1: Introduction    Global warming is one of the greatest challenges facing the world and fossil fuel combustion is one of the primary sources of greenhouse gases.  It was estimated that 8.9 gigatonnes CO2 was produced annually from 2004 to 2013 while only 2.9 and 2.6 gigatonnes can be naturally absorbed by land and ocean.1 According to the International Energy Agency (IEA), petroleum is the most heavily consumed energy source. The highest consumption is by the transportation sector and the second highest is by non-energy uses (e.g. production of chemical and material).  Several strategies from interrelated sectors, including agriculture and forestry, alternative energy, carbon capture and storage, power generation, and end-user efficiency and conservation, have been proposed to reduce carbon emissions.2 While renewable power sources such as wind, tidal, water turbine, and solar cells can potentially replace petroleum in the energy sector, renewable carbon is still required by the chemical and material sector.  Lignocellulosic biomass, which consists of carbon, hydrogen, and oxygen, is the closest material to replace petroleum, the major constituents of which are hydrogen and carbon. Similar to petroleum, lignocellulosic biomass is a promising source for fuels, chemicals, and materials. In addition to being the most abundant carbon source, lignocellulosic biomass can be considered an efficient natural carbon-recycling agent. Carbon recycling in biomass occurs by adsorption and conversion of CO2 to carbohydrates during photosynthesis. In addition to carbohydrates, other constituents such as lignin and fatty acids are synthesized in the cell wall during growth. Therefore, 2  although the fuel from biomass will produce CO2 during utilization, the total carbon in the atmosphere can be maintained in balance. Finally, minerals that are taken in during plant growth, might be returned to the soil in the form of ash. Lignocellulosic biomass consists of cellulose, hemicelluloses, and lignin; these polymers are extremely recalcitrant to degradation.3 One possible pathway for the conversion of lignocellulose to fuels and chemicals is via thermochemical processes such as gasification, pyrolysis, and liquefaction. Gasification has several technical barriers due to tar formation and a further Fischer-Tropsch process is needed to obtain a valuable product. Pyrolysis and liquefaction are relatively easy to conduct; however, these processes produce bio-oil, a mixture of hundreds of compounds. Separation processes such as liquid-liquid extraction,4,5 distillation,6 or a combination of both are often required before further utilization of bio-oil. Bio-oil contains higher O/C and lower H/C compared to gasoline and thus requires upgrading. The upgrading is generally conducted by means of hydrodeoxygenation reaction (HDO) of the bio-oil.  Another strategy for lignocellulose conversion is the fractionation of carbohydrate and lignin. While carbohydrates have existing markets such as pulp and bio-ethanol, more work is still required for the valorization of lignin. Lignin, which is mostly concentrated in the secondary cell wall and middle lamella, accounts for 20-30 wt% of total biomass weight.7,8 Thus, the valorization of lignin is essential to the biorefinery platform. Although Kraft lignin production, the largest mass of lignin currently produced, can reach around 630 kilotons annually, much of it is burned to produce process heat and electricity and only approximately 100 kilotons are used for other applications.9 Combustion is a relatively low-value use for lignin. Lignin-derived monomers represent up to 21.9 wt% of bio-oil10 and presents a unique fuel feedstock. 3  The low O/C ratio of lignin is closer to gasoline compared to other biomass constituents. However, HDO is still needed to further lower O/C and improve H/C ratio to produce fuel-quality liquids. In fact, hydrogenolysis is the most common reaction conducted with lignin. It has been calculated that the stoichiometric requirement to upgrade 1 kg of bio-oil is approximately 62 g of hydrogen.11  Unfortunately, the majority of hydrogen is currently produced using fossil fuels via methane reforming, naphtha cracking, and coal gasification.12 As can be seen from Fig 1.1, most hydrogen is produced and consumed within a single facility.  Methane reforming is performed to support ammonia production, coal gasification produces syngas for methanol production and naphtha cracking is an integral part of oil refining. Similarly, biomass hydrogenolysis should be conducted with biomass-derived hydrogen to obtain a self-sustaining process, maintain renewability; and carbon neutrality.    Figure 1.1 Hydrogen source and demand by industry. Redrawn with permission from [12] Copyright © 2011 Elsevier.  Aqueous phase reforming (APR) is a promising method for generating hydrogen from lignocellulose biomass since the highly oxygenated carbon in biomass is advantageous. 4  Oxygenated carbon results in a reforming process that is thermodynamically favorable at low temperature.13 Among compounds with O/C = 1, glucose, from cellulose, and glycerol, a biodiesel by-product, are promising candidates with respect to availability and renewability. However, there are two major advantages of glycerol relative to glucose. Glucose can be upgraded to many other products while the supply of low-purity glycerol generated during biodiesel production has far exceeded current demand. Secondly, the lower carbon number of glycerol reduces the probability that it will undergo dehydration and produce alkanes during APR.14  Figure 1.2 Glycerol production and price.15  Glycerol is abundant because it is the byproduct of alkaline transesterification of triglycerides with methanol to produce fatty acid methyl ester (FAME, biodiesel). The production of glycerol has increased in recent decades due to the booming biodiesel industry (Fig 1.2).16 Additional advantages of glycerol relative to other low carbon compounds such as methanol are low toxicity, non-flammability, and high boiling point. These characteristics make glycerol much safer to handle.  5  It is desirable to combine hydrogen production via glycerol APR with lignin hydrogenation for renewable chemicals or fuels production. Conducting in-situ glycerol APR and lignin upgrading will have several advantages. Glycerol’s high hydroxyl content will likely cause a solvolysis effect and facilitate lignin dissolution. The presence of lignin or lignin monomer is expected to improve hydrogen selectivity from glycerol due to rapid consumption of hydrogen. Due to the complexity of lignin, understanding the reaction by directly conducting in-situ glycerol APR and lignin upgrading is a difficult task. Phenol was utilized as a lignin model compound due to the simplicity of potential hydrogenation products. In this thesis, step-by-step understanding was developed by investigating glycerol APR reaction, in-situ glycerol APR and phenol hydrogenation (IGAPH), in-situ glycerol APR and phenol hydrodeoxygenation (IGAPHdo), and finally upgrading of fast pyrolysis oil (FPO) and native wood (mixed spruce, pine and fir, i.e. SPF).    1.1 Objective The ultimate purpose of this study is to enable the utilization of carbon from lignin or lignocellulose material and hydrogen from glycerol to obtain higher value chemicals or fuel. To reduce the complexity of potential products, initial studies are conducted using phenol as a lignin model compound. The first objective was to identify the reaction mechanisms and limitations of glycerol APR. The second objective was to improve the selectivity of glycerol APR with the addition of phenol during IGAPH, hydrogenate phenol with hydrogen produced from glycerol, as well as determine the phenol hydrogenation mechanism (i.e. whether molecular hydrogen was formed or not) during IGAPH. The third objective was to deoxygenate phenol and control the product distribution of IGAPHdo and apply the understanding to control selectivity. Insights from these studies are used to guide the upgrading of lignin containing material (FPO and SPF). 6  1.2 Approach 1. Chapter 1 introduces the importance in lignin upgrading and the associated hydrogen demand. This introduction also describes current hydrogen production processes and the motivation for glycerol APR as an alternative.  2. Chapter 2 reviews the current progress on upgrading lignin, bio-oil, and model compounds as well as progress in glycerol APR. 3. Chapter 3 provides the experimental procedures for: glycerol APR, in-situ glycerol aqueous phase reforming and phenol hydrogenation (IGAPH), in-situ glycerol aqueous phase reforming and phenol hydrodeoxygenation (IGAPHdo), and fast pyrolysis oil (FPO) upgrading and SPF hydrogenolysis. The IGAPH procedure includes studies of the hydrogenation mechanism in two different reactor configurations. Product analysis and quantification methods are also detailed. 4. Chapter 4 presents the study of glycerol APR with Raney Ni® catalyst. In this chapter, the thermodynamics, reaction pathways, and kinetics of glycerol APR are examined. The reaction pathways and kinetic analysis were conducted at 220 °C. The selectivities of APR vs methanation in the batch system are discussed in this chapter. This chapter and part of chapter 5 has been prepared as a manuscript for journal publication. 5. Chapter 5 details IGAPH reactions, most were conducted with Raney Ni®. In this chapter reaction thermodynamics were first analyzed. The effect of phenol loading on glycerol APR and the effect of glycerol loading on phenol hydrogenation are discussed with respect to conversion and product selectivity. Hydrogenation mechanism was studied by continuously purging the reactor with helium to remove molecular hydrogen as it formed. Reaction pathways during IGAPH were proposed in this chapter. Several noble metal 7  catalysts were also tested for the IGAPH reaction. Part of this chapter has been published in Chemical Engineering Journal. The remainder of this chapter has been combined with part of chapter 4 as a second publication.  6. Chapter 6 contains IGAPHdo experiments targeting production of deoxygenated compounds. In this chapter, two sets of catalysts were tested: Raney Ni® – H-ZSM-5 and Pt/C – Amberlyst-15. IGAPH products were also subjected to IGAPHdo with Raney Ni® – H -ZSM-5 in order to identify the reaction pathways. The attempt to control product selectivity was conducted using Pt/C. This chapter and some results from chapter 7 are being prepared for journal publication. 7. Chapter 7 reports experiments with ligneous material including FPO and crushed SPF pellet as starting materials. The effect of solvent, catalyst, and reaction condition were observed. 8. Chapter 8 summarizes the whole of the work in unity, provides perspective on the system, and provides recommendations of future work.  9. Appendices provide calculation details.     8  Chapter 2: State of the Art    2.1  Lignocellulose structure Lignocellulosic biomass consists of three major components: cellulose, hemicellulose, and lignin. Depending on the type of lignocellulose biomass, the percentage of these three components can vary. Table 2.1 shows the composition of these three materials in several lignocellulosic biomasses. The three components are bound together via several types of linkages, including phenyl glucoside, benzyl ether, and esters. This bonding between lignin and carbohydrates (cellulose and hemicellulose) is commonly known as a lignin-carbohydrate complex (LCC).17 In the cell wall, hemicelluloses and lignin surround cellulose fibrils; the lignin crosslinks with the hemicelluloses and celluloses. This prevents the polysaccharides from swelling as lignin is hydrophobic unlike the other constituents.18 Given the specific properties of its constituents, lignocellulosic biomass has strong resistance to stressors, including predators, and fractionation.19 The properties of each lignocellulose constituent will be discussed in the following subsections.      9  Table 2.1 Lignocellulose composition of several types of biomass No Biomass Type Cellulose (wt%) Hemicellulose (wt%) Lignin (wt%) Ref. 1 Spruce  Softwood 43 24 28 20 2 Pine Softwood 29 20 33 21 3 Douglas Fir Softwood 49 21 30 22 4 Oak Hardwood 43 16 34 5 5 Empty Fruit Bunch (Palm) Nonwoody 41 22 28 5 6 Sugarcane Bagasse Nonwoody 52 16 26 23 7 Wheat Straw Nonwoody 33 28 22 7   2.1.1 Cellulose Cellulose is a linear polymer of glucose linked via β(1-4) glycosidic bonds in which two anhydro glucose unit (AGU) are repeatedly linked in inverted positions with the respect to the ring’s plane.24 This specific β-glycosidic linkage differentiates cellulose from other carbohydrates, such as amylose, which contain α-glycosidic linkages resulting in a helical structure.25 The degree of cellulose polymerization ranges from 100 to 20,000 AGU.26 This linear polymer structure produces strong intramolecular and intermolecular hydrogen bonds in cellulose. The intramolecular hydrogen bonds occur between hydrogen and oxygen atoms in a single polymer, while the intermolecular hydrogen bonds occur between hydrogen and oxygen atoms in different polymer chains.27–29 It has been found that cellulose with higher molecular weight has greater thermal stability, which may be due to intramolecular hydrogen bonds strengthening the cellulose chain.30 The intermolecular hydrogen bonds produce the cellulose’s crystallinity, which appears 10  at 2 = 22.5 during X-ray diffraction (XRD) analysis.31–33 This crystallinity gives cellulose the strength and compactness that make it difficult for bacteria, enzymes, or chemicals to penetrate. In nature, this recalcitrance is required for a plant to survive from predators. Cellulose is primarily used for pulp production, with approximately 400 Mt of pulp being produced annually.34 It then is made into numerous end-user products, such as paper and packaging. Cellulose is also currently a source of second-generation bioethanol through hydrolysis to produce glucose, which is then followed by fermentation.21 Recently, nanocrystalline cellulose (NCC) has gained significant attention for its usefulness in a wide range of applications, from packaging, to capacitors to biomedical devices.35 The key advantages of NCC as a material are its renewability, high surface area, unique optical properties, biocompatibility, biodegradability, high stiffness and high strength.36–38 Furthermore, lignin removal through low pressure alkaline delignification has been found to enhance the crystalline index of NCC.39  2.1.2 Hemicellulose  Hemicellulose is a heterogeneous branched polymer of sugars that can consist of five-carbon sugars (pentoses), six-carbon sugars (hexoses), or a combination of both. Compared with cellulose, it is a smaller polymer with a narrower range of polymerization, between 100 and 200 anhydro sugar unit.40 Since it is branched and shorter, hemicellulose is easier to degrade than cellulose. Hemicellulose is the weakest constituent of lignocellulose; in the presence of heat alone, it begins to break down at 220°C, compared with 315°C  for cellulose.41 Its weak properties make it an easy target during fractionation. Since some portions of lignin are attached to hemicellulose, breaking hemicellulose under mild fractionation conditions will detach a portion of the lignin from the 11  lignocellulose. Mild conditions can prevent lignin condensation, which can occur under harsh conditions. The type of hemicellulose present in biomass varies with the type of biomass. Xylan, a polymer of xylose, is the most common hemicellulose found in hardwood, while glucomannan, a polymer of glucose and mannose, is the most common in softwood.25 The heterogeneity of hemicellulosic sugars makes them more complicated to utilize than cellulose. One use is for pulp reinforcement, where the re-adsorption of hemicellulose has been shown to increase the strength of paper. As monomers, several hemicellulose sugars are promising as precursors of several renewable polymers. Recently, it was found that polyethylene 2,5-furan dicarboxylate (PEF) can be used as a replacement for polyethylene terephthalate (PET), offering even better properties as a gas barrier.42 PEF results from the polymerization of 2,5-furan dicarboxylic acid (FDCA), which can be synthesized by the oxidation of 5-hydroxymethyl furfural subsequent to the dehydration of fructose.43 Polylactic acid (PLA), the result of lactic acid polymerization, is used to make biodegradable plastic and can be synthesized from several hemicellulose sugars.44   2.1.3 Lignin Lignin is the most heterogeneous polymer among the biomass constituents, consisting of phenylpropanoids units that link together through several types of bonds. There are previously three canonical monolignols, differing only in their number of methoxy groups (Fig. 2.1a): p-coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol. The monolignol units vary depending on the type of lignocellulose biomass. In softwood, the main monolignol unit is coniferyl alcohol, while both coniferyl and sinapyl alcohol can be found in hardwood lignin. In non-woody lignocellulose biomass, such as perennial grasses and empty fruit bunches, all three monolignols 12  can be found.45 In addition to the canonical monolignols, there are several minor monomers that are involved in lignification, either naturally or through engineering, such as caffeyl alcohol, tricin, and piceattannol.46,47    Figure 2.1 Canonical monolignols (a) and non-canonical monolignols (b) with their polymerisation. Adopted from Ralph et al.47  The strength of the lignin polymer is affected by the type of bonding between its monomers. Ether bonds, such as β-O-4 are weaker than, for example, C-C bonds. In native lignin, the type and number of bonds are affected by the monolignol comprising the lignin.45 Softwood lignin, for instance, tends to have more C-C bonds than hardwood lignin, since there is only a single methoxy group on the aromatic ring, leaving one carbon in the meta position (the ortho position relative to the hydroxyl) to attach with a carbon from another phenylpropanoate unit. Recently, many genetic engineering efforts have sought to introduce other monolignols in lignin structure. Monolignol alteration will result in changes such as increase the number of β-O-4 unit and thus improve the yield of desired products during saccharification or lignin hydrogenolysis.48,49 13  Since it has the lowest O:C ratio, lignin is more readily substituted for hydrocarbon than the other lignocellulose biomass constituents. Applications for lignin include fuels, chemicals, and carbonaceous materials (e.g. polyurethane, carbon foam, carbon fiber). When attempting to substitute lignin for petroleum fuels, the lignin must be depolymerized into smaller compounds, in the range of 5–18 carbons, depending on the type of fuel. Another requirement for making lignin an alternative fuel is to reduce the oxygen content in order to increase the resulting fuel’s heating value and decrease its acidity. Oxygen removal from lignin derivatives is usually conducted via HDO so that oxygen can be removed as water.  Although a low carbon number and very low oxygen content are required, lignin does not have to be present as a pure compound to be used as a fuel, which eliminates the burden of purification. When lignin is being made into chemicals, oxygen removal might not be necessary, but purity will be essential. Two approaches are possible to achieve a high purity compound. The first is to attack a weak specific bond so that only specific compounds will form. This approach results in low yield, but the separation process is easier. The second method is to apply very harsh conditions, followed by reconstruction—for example, gasification followed by the Fischer–Tropsch process. Making carbonaceous material is the least complicated process, since lignin depolymerization is not required. However, the value of this material relies heavily on its physical properties.   2.2 Lignin isolation 2.2.1 Alkaline treatment (Kraft pulping) Lignin is known to degrade under alkaline condition. The most well-known reaction of lignin under alkaline conditions is the breaking of the β-O-4 linkage, illustrated in Figure E.1.a (Appendix 14  E). High temperature and high alkali concentration force a hydrogen atom to be released from the hydroxyl group at Cα. The result is a strong nucleophile that breaks the β-O-4 bond, producing a deprotonated phenolic end group.18 When this has a negative charge delocalized on C4 because of the strong electronegativity of O, a double bond with C1, the hydroxyl group or LCC on Cα can leave, forming a quinomethide that has a partial positive charge δ+ at the α position. This Cα can undergo a nucleophilic attack by sulfide, SH–. If SH– is further deprotonated, it then attacks Cβ, breaking the β-O-4 bond. It is important to note that this reaction only happens with sulfide, due to its nucleophilic strength.50 This reaction proceeds until the lignin molecules are small enough to be dissolved in an alkali solution (Figure E.1.b). When quinomethide is formed, Cγ can undergo deprotonation due to the alkaline conditions. The result is a positive charge on Cα and negative charges on the O attached to Cγ and C1, as shown in Figure E.1c. Cγ is then detached to form formic acid and an enol ether structure, which is stable in an alkali.18 This reaction stops further degradation of the lignin fragments. Lignin condensation also occurs during alkaline treatment. Quinomethide produced from coniferyl alcohol through vinyl guaiacol will result in apocynol in the presence of water.51 Vinyl guaiacol and apocynol can further condense together in the presence of OH–/HS–. 52,53  2.2.2  Acid hydrolysis Unlike in alkaline conditions, cellulose undergoes depolymerization into monomeric sugars during acid hydrolysis, which begins with the attachment of a proton to either glycosidic or pyranic ether.26 Lignin can undergo depolymerization and condensation under acidic conditions. Several such reactions are shown in Figure E.2: depolymerization of β-O-4, breakage of LCC, and condensation. Lignin depolymerization starts with the attachment of a hydrogen proton to a 15  hydroxyl group in Cα and further induces an elimination reaction to produce water, formaldehyde, and proton. The presence of acid continues the reaction, as the hydrolysis of β-O-4 produces a phenolic end group (Figure E.2.a). LCC breakage can occur through either hydrolysis or methanolysis; both reactions are promoted by the attachment of acid to the ether bond (Figure E.2.b). Condensation reactions happen in the presence of electron-rich carbon. This reaction leads to the formation of a carbon–carbon bond that is more stable than the ether bond (Figure E.2.c).18 This condensation results in high molecular weight lignin (5000–10000 Da), which makes it more difficult to depolymerize.9 It has also been found that the higher the acidity, the higher the condensation rate; this effect can be avoided by adding alkaline components.54 Another strategy for dampening the condensation reaction is to use short residence times, which can be achieved in a flow-through reactor.55  2.2.1 Organosolv process (dilute acid process) The organosolv process during lignin isolation involves a mixture of organic solvent and water. In some cases, dilute acid (H2SO4, HCl, or organic acids) is incorporated to catalyze the hydrolysis reaction of hemicellulose. Low boiling point alcohols such as methanol and ethanol are widely utilized in the organosolv process, offering easy solvent recovery, while high boiling point alcohols such as glycerol and ethylene glycol make it possible to conduct the reaction at atmospheric pressure.56 It has been suggested that the delignification reaction during the organosolv process occurs through the cleavage of α-O-4 bonds and to a lesser extent β-O-4 bonds.57,58   Pretreatment of Liriodendron tulipifera with a 50 wt% ethanol–water mixture and 1 wt% H2SO4 at 140°C for 10 min was found to remove 43.4 wt% of the total lignin.59 However, organosolv 16  treatment of  Pitch Pine with a similar solution (50 wt% ethanol, 1 wt% H2SO4) at 180°C for 0 min (the reaction was stopped when it reached the target temperature) resulted in negligible lignin removal.60 Aside from the possible differences between the particular lignins, the contrasting results of these two experiments might be due to the condensation and recombination of lignin with lignocellulose due to the higher reaction temperature. A higher acid concentration and pretreatment temperature were found to increase lignin extraction from lodgepole pine that contained lower molecular weight lignin.61 Organosolv pretreatment of wheat straw with 60 wt% ethanol increased lignin removal from 24.4 wt% to 43.8 wt% when the H2SO4 concentration was doubled from 15 mM to 30 mM.62 In these experiments, higher ethanol concentration negatively affected delignification.  Glycerol is known to be helpful in lignin extraction and is widely used during pretreatment to delignify lignocellulose. It was found that the degree of delignification of sugarcane bagasse reached 81.4% at 198°C with an 80% v/v glycerol-water solution.63 Similar experiments with bagasse were conducted and it was found that the presence of glycerol during pretreatment with NaOH preserved xylan while maintaining delignification.64 In both reactions with bagasse, the delignification degree always increased proportionally with glycerol concentration. A delignification process in 75 wt% glycerol solution conducted with hardwood and softwood found that the delignification of softwood was considerably more difficult; less than 20 wt% lignin was removed from the softwood, while more than 60 wt% lignin was removed from the hardwood.65,66 With a high biomass-to-glycerol loading, no delignification occurred during glycerol thermal processing (GTP).67,68 However, lignin removal improved during subsequent extraction with 1,4- dioxane.  17  Besides alcohols, γ-valerolactone (GVL) can dissolve all three lignocellulose constituents.69 More than 80 wt% glucose and xylose was recovered during the treatment of corn stover with an 80 wt% GVL-water solution containing 5 mM of H2SO4, and the lignin could be separated by adding water to dilute the GVL concentration.70 1,4-dioxane is another aprotic solvent commonly used to extract lignin. Recently, simultaneous lignin extraction and modification were conducted using 1,4-dioxane as the solvent. In this reaction, formaldehyde was used to protect the lignin from condensation by attaching formaldehyde to the Cα and Cβ, forming a dioxane-like structure.71 The utilization of formaldehyde also resulted in ring allylation, which could be avoided by using acetaldehyde.72 With this modification, the condensation reaction could be avoided and a high yield of syringyl propanol was obtained from birch wood.   2.3 Pyrolysis and liquefaction of lignocellulose The two most common methods to deconstruct lignin into bio-oil are through fast pyrolysis and liquefaction. To obtain a high yield of bio-oil, fast pyrolysis should be conducted with a rapid heating rate and short residence time and be solvent free.73 Liquefaction, on the other hand, is usually conducted with a lower heating rate, a longer reaction time, and the addition of a solvent. The solvent’s properties at reaction conditions is an important factor during liquefaction.74 The reaction temperature and heating rate affect the products of both fast pyrolysis and liquefaction.  In a batch tubular reactor, a moderate pyrolysis reaction temperature (650–700°C) with a faster heating rate (~100°C/min) resulted in a higher pyrolysis oil yield, while a lower reaction temperature (550°C) with a slower heating rate (~2°C/min) resulted in higher char yield.75 Similar trends were found in pyrolysis with a fluidized bed reactor, where a moderate reaction temperature 18  (500°C) resulted in a higher yield of pyrolysis oil, while a lower reaction temperature (400°C) resulted in a higher yield of char.76,77 The slightly lower reaction temperature in the fluidized bed reactor compared to the tubular reactor for obtaining a high yield of pyrolysis oil might be due to better heat transport in the fluidized bed reactor. The higher reaction temperature resulted in larger amounts of gaseous products, regardless of the type of reactor used. At 450–550°C, when the highest yield of pyrolysis oil was obtained, the oil typically contained 44–47 wt% of carbon, 6–7 wt% of hydrogen, and 46–48 wt% of oxygen.77 Interestingly, pyrolysis oil from woody biomass tended to result in higher oxygen content than microalgae.78 Heating rate and reaction temperature also affect product distribution during the liquefaction of lignocellulose. A faster heating rate with a higher reaction temperature during the liquefaction of sawdust in sub-critical water and super-critical ethanol  resulted in higher bio-oil yield.79 The effect of solvents (acetone, ethanol, and water) during sawdust liquefaction has also been studied between 250 and 450°C, with the highest conversion being obtained in acetone and the highest bio-oil yield in ethanol.80 In addition, liquefaction with ethanol was found to produce bio-oil of a lower molecular weight than with water.79  2.4 Upgrading lignin-derived compounds In this work, only bio-oil from fast pyrolysis was used and therefore is subsequently referred to as fast pyrolysis oil (FPO). Numerous attempts to upgrade FPO and understand the process for doing so have been made, the main focus being to remove oxygen and add hydrogen to the relevant compounds. Due to the complexity of FPO components, most studies in FPO upgrading involve model compounds—monomeric, dimeric, or even bigger compounds. There are two major factors 19  in lignin upgrading to be considered. The first is catalyst performance, which many researchers have explored. Depending on their role, the catalysts are divided into two groups: metal catalysts for hydrogenation and acid catalysts for deoxygenation. The metal catalyst is typically a transition or noble metal, such as Ni, Mo, Pt, Pd, Ru, or Rh. Acid catalysts can be homogenous or solid acids with Lewis and Brønsted sites. Some have other properties, such as molecular scission (in zeolite). Another important factor in lignin or FPO upgrading is solvent compatibility. Lignin and FPO have poor solubility in most commonly used solvents. This poor solubility is primarily due to polar and nonpolar sites on lignin or FPO, which require a solvent with medium polarity to facilitate solubility.81 The capabilities of several polar aprotic, polar protic, and nonpolar solvents in facilitating lignin or FPO upgrading have been studied, and each type of solvent has its own advantages.   This subsection will cover the hydrogenation of lignin model compounds, starting with the conventional HDO reaction, which employs molecular hydrogen and involves hydrogenation and dehydration. The alternative direct deoxygenation (DDO) reaction, which requires less hydrogen, will also be discussed. Catalytic transfer hydrogenation (CTH), which has recently garnered more attention, will also be addressed. Finally, the solvent effect in lignin depolymerization and FPO upgrading will be covered.   2.4.1 The catalyst’s role 2.4.1.1 Hydrogenation Hydrogenation of the aromatic ring has been identified as the rate-limiting step during phenol HDO at 150–220 ℃.82 This can be affected by the exothermic hydrogenation reaction, which favors a lower reaction temperature. Although a low reaction temperature is more beneficial for 20  hydrogenation in terms of overall conversion, the activation energy and the reaction rate usually demand a high reaction temperature. In most studies, highly pressurized hydrogen gas is applied to overcome the thermodynamic boundary, since a high concentration of hydrogen will increase phenol conversion. Hydrogenation of the aromatic ring in phenol has been found to occur at temperatures as low as 50°C.83 In comparison, conversion of vanillin to vanillyl alcohol can be conducted at 45°C under 0.69 MPa H2 with a linearly proportional correlation between the reactant concentration and the TOF (turnover frequency).84 This result indicates that the aldehyde has higher reactivity than the aromatic ring in vanillin. In the hydrogenation of eugenol at 225–325°C, saturation of the allyl group was found to be the first reaction step, indicating this functional group has the highest reactivity.85 In this study, it was also found that cerulignol, the product of eugenol allyl group saturation, further reacts through two routes. The first route is through hydrogenation of the aromatic ring which is the predominant pathway and second route is dehydroxylation and demethoxylation. Furthermore, dehydroxylation on the aromatic ring of eugenol was suggested to occur through the formation of OH radicals.86 Although it might differ slightly, it can be generally said that hydrogenation of the “para” functional group in phenolic monomers is easiest, followed by aromatic ring saturation. The C-O or C-C bond is known to be more stable toward the hydrogenolysis reaction, but this seems to be affected by other structures in the monomer.  21   Figure 2.2 Energy barrier during phenol hydrogenation with Pt (left) and Pd (right).87 Reprinted with permission from [82]. Copyright © 2015, American Chemical Society  In the hydrogenation of phenol at 200°C with 5 MPa H2 (at room temperature), which resulted in ~95% yield of cyclohexanol in water solution, no significant activity difference could be discerned between commercial noble metals supported on carbon, such as palladium (Pd/C), ruthenium (Ru/C), rhodium (Rh/C), and platinum (Pt/C).88 However, it has been found that the adsorption mechanism of phenol varies between one metal and another. For example, the adsorption of phenol on Pd is distinct from its adsorption on Pt. With Pd, phenol undergoes nonplanar adsorption that results in dissociative adsorption, producing cyclohexanone, while in Pt, planar adsorption results in non-dissociative adsorption, generating cyclohexanol.87,89,90 This particular result was only observable under less harsh reaction conditions (< 100°C with limited H2); harsher reaction conditions can overcome all energy barriers and result in less selective products. The energy barrier of phenol hydrogenation with Pd and Pt is shown in Fig 2.2. This reaction is usually conducted to produce ultra-selective cyclohexanone, which has gained attention because it is a precursor of nylon. It was further found that a Lewis acid such as AlCl3, ZnCl2, or InCl2 activates the benzene 22  ring in phenol due to its electron-accepting property, while Pd activates hydrogen, so cyclohexanone can easily form but further hydrogenation to cyclohexanol is prevented due to the acid–base interaction between cyclohexanol and cyclohexanone.91 Beside noble metals, transition metals are also common catalysts for the hydrogenation of lignin model compounds. Among them, Raney Ni® is well known, invented by Murray Raney in 1926 by treating a Ni/Al alloy with a caustic solution of sodium hydroxide. The leaching of aluminum metal in caustic solution renders Ni a porous metal with a high surface area. Some of the aluminum remains to create structural stability and high activity. Upon heating, hydrogen is formed by the reaction between water bound on the alumina and aluminum metal.92  Raney Ni® was found to be active as a hydrogenation catalyst for 4-propylphenol conversion to cycloalkanes, but it was inactive in the presence of several acids, such as H3PO4 and CH3COOH.93 The hydrogenation of diphenyl ether as a lignin model compound with Raney Ni® was achievable at 90C with either polar protic, polar aprotic, or nonpolar solvents. This study suggests that the Lewis basicity of solvents reduces hydrogenation activity due to strong adsorption.81 This study also found that primary alcohols such as methanol and ethanol poisoned the catalyst, while aprotic solvents with Lewis basicity only inhibited the reaction. A similar result was also found in phenol hydrogenation with Raney Ni® at 67C, where the hydrogenation rate of phenol in water was higher than in methanol.94 In contrast, it was found that the addition of methanol during phenol hydrogenation with modified Raney Ni®-Mo at 180C increased phenol conversion.95 Methanol is known to adsorb very quickly on a nickel surface even at a low temperature of –40C.96 Further, methanol can gradually lose its hydrogen, starting from the hydrogen in O-H, to form methoxy and finally carbon monoxide (CO), which binds strongly to the nickel surface.97 In the case of 23  hydrogenation at a low temperature (90°C), the available energy may be insufficient to break the bond between CO and the Ni surface, resulting in catalyst poisoning.   2.4.1.2 Hydrodeoxygenation Hydrodeoxygenation of phenolic monomers probably accounts for the highest number of studies related to lignin upgrading. Strategies for the HDO reaction include using water as an acidic medium at a high temperature, using an acid catalyst, or using a bimetallic catalyst. In the first two approaches, acidic conditions are required to remove oxygen through a dehydration reaction subsequent to the hydrogenation of phenolic to form a saturated cyclic alcohol or ketone. Thus, the HDO of phenolic monomers is technically possible in the presence of water at a relatively high temperature. It is known that the pKw of water decreases at higher pressures and decreases up to 300C, then increases as the temperature is further elevated.46 The acidic conditions can be enhanced by introducing an acid catalyst. This catalyst can be homogenous, such as HCl, H3PO4, or CH3COOH, or heterogeneous, such as Al2O3, SiO2-Al2O3, zeolite, or an acid-treated neutral catalyst. These heterogeneous catalysts are often used as supports to increase metal dispersion. Bimetallic catalysts usually consist of noble–transition or transition–transition metals with a neutral support. The advantages of bimetallic catalysts arise either because the first metal is reduced by the second metal through spillover, or for more complex reasons, such as geometric, electronic, stabilizing, synergetic, and bifunctional effects.98  It was found that the dehydration of cyclohexanol to cyclohexene can be achieved in high-temperature water without any catalyst and occurs through bimolecular elimination, where cyclohexene further forms cyclohexyl cations, rearranges into methyl cyclopentyl cations, and finally forms methyl cyclopentene as a side product.99 The HDO of several phenolic model 24  compounds has also been achieved in the presence of Ru catalyst supported on several carbons, where the dispersion of Ru was found to be the key factor.100 In addition, Pt supported on several types of carbon has been shown to enable the HDO reaction of phenolic compounds.101 DFT calculation suggests that the presence of water, besides conferring acidic properties, also decreases the energy barrier in phenol hydrogenation under Pt and Ni.102  A study to examine the activity of several noble metals (Pt, Rh, Pd, and Ru) on a number of supported solid acid catalysts—alumina (Al2O3), amorphous alumina-silica (SiO2-Al2O3), and nitric acid-treated carbon black (NAC)—in the hydrodeoxygenation of guaiacol under n-decane solution at 250C and 4 MPa H2 was conducted and found that the reaction with Ru/SiO2·Al2O3 has the highest cyclohexane yield. It was suggested this was due to the high acidity of the catalyst compared with the others.103 Besides noble metals, transition metals such as Ni, Co, and Mo are commonly used in the HDO of phenolic compounds, such as guaiacol. However, with these catalysts, demethylation and demethoxylation are the predominant reactions with guaiacol, resulting in catechol or phenol.104,105 A low concentration of a homogeneous acid catalyst was also capable of conducting the HDO of phenol to cyclohexane at 200C.88 However, the homogenous acid catalyst seemed to poison transition metals such as Ni. In addition to the possibility of catalyst poisoning, catalyst separation is another challenge when utilizing a homogenous acid catalyst. Often the homogenous catalyst is supported on a neutral solid catalyst to counteract these challenges.93  In the selection of supports, several parameters should be considered, including physical properties (surface area, pore size), acidity, and stability under the reaction conditions. The highest surface area with a pore width in the mesoporous range will increase the chance of achieving a highly accessible metal catalyst. The acidity of the support will help with dehydroxylation through a 25  dehydration reaction.  It has been suggested that a large pore size in an acid catalyst will result in a high deoxygenation rate and higher levels of coke formation.106–108 A highly accessible catalyst due to pore size will definitely increase accessibility to active sites. However, it is also possible that too high of an acid density in the catalyst will result in coke formation due to condensation reactions.108 In this case, a high degree of hydrogenation should be employed to overcome the condensation reaction.  Zeolite is the most common catalyst found in FPO upgrading or lignin catalytic pyrolysis. In addition to its fine pore structure, which permits selective adsorption on the basis of molecular size—sometimes called molecular sieving—zeolite is also of practical interest because its properties are flexible and can be changed via ion exchange.92 The HDO of guaiacol with Pt on several supports shows the superiority of zeolite compared to other supports, such as TiO2, ZrO2, CeO2, and SiO2.109 In lignin catalytic pyrolysis carried out in the vapor phase at 873 K for 10 min, zeolite with three-dimensional pores (ZSM-5, Y, and β) showed high activity for the production of light oil, with ZSM-5 zeolite improving the decarboxylation reaction, resulting in lignin oil with lower acidity.110 A high ratio of Si to Al on ZSM-5 can improve the cleavage of methoxy, ether, and C-C bonds in aliphatic compounds as well as the dehydration of OH in aliphatic groups, but the decomposition of carboxylic acid is more selective in ZSM-5 with a low Si:Al ratio.111 Catalytic pyrolysis of lignin with Py-GC/MS at 650C showed that H-ZSM-5 increased the yield of deoxygenated aromatic compounds, which were accompanied by several polyaromatic compounds.112 It was also suggested that a large number of acid sites in H-ZSM-5 results in high production of aromatic deoxygenated compounds. Moreover, sufficiently large pore area allows large molecules to pass through and be further converted.113 However, too great a pore size can induce large polyaromatics to polymerize into coke. In this case, too high of an acid density means 26  shorter distances between acid sites, which will lead to polymerization between aromatic molecules and therefore to coke formation.114 To conduct the hydrodeoxygenation reaction, the reaction conditions and reactor system should be considered to select the proper zeolite type. ZSM-5 was found to be stable in liquid water between 150 and 250C, regardless of the SiO2:Al2O3 ratio, while zeolite Y degraded in these conditions through the hydrolysis of Si–O–Si bonds but not through dealumination.115 This degradation was much more pronounced with high SiO2:Al2O3 ratios, indicating that the stability of the materials could be increased with the incorporation of more Al2O3; this was the opposite of the zeolite’s stability’s in steam.116 In addition, it has been found that the amount of water in the hydrophobic reaction can increase or decrease the reaction rate.117  In FPO upgrading, zeolites are often utilized as supports for transition and noble metals. Observation of the HDO reaction with Pt on several zeolites supports—including H-BEA, H-ZSM-5, MMZ, and MCM-48—in decane solution found that acidity and pore size were the keys to high cyclohexane yield.118 FPO upgrading with Ni/H-ZSM-5 increased the hydrogen content and pH of FPO, with the acidity of H-ZSM-5 playing an important role in the catalyst’s activity.119 Major progress has been achieved in upgrading hexane-extracted FPO with Ni/H-ZSM-5 using water as a solvent, resulting in 90% naphthenes and aromatics and 10% paraffin.120 The addition of Al2O3 as a binder for Ni/H-ZSM-5 improved Ni dispersion, which further increased the reaction rate, although the catalyst still suffered from deactivation due to Ni sintering.82 Ni/Al2O3-H-ZSM-5 was further found to have a smaller Ni particle size and generate stronger chemical interactions between the metal and the support through the formation of NiO-Al2O3.121 Besides the employment of acid to improve the dehydration reaction, a bimetallic catalyst is often used in phenolic upgrading. In some cases, the route is through the DDO reaction. It has been 27  suggested that stretching between the aromatic ring that prefers to adsorb on Fe and the oxygen that prefers to adsorb on Pd occurs during the vapor phase of DDO with guaiacol, producing benzene when bimetallic Pd-Fe/C is the catalyst, with the distance between Fe and Pd weakening the C-O bond in phenol.122 The addition of Mo in Pt/TiO2 was also found to increase the HDO of guaiacol to cyclohexane via active hydrogen from Pt-MoOx sites.123 Similar results were found in the HDO of guaiacol with Ni-Fe/CNT, where it was suggested that the aromatic ring preferred to adsorb on Fe, while hydrogen dissociation worked better on Ni.124  2.4.2 The role of solvents With either FPOs or model compounds, upgrading solvents play an important role as reaction media or hydrogen sources. As reaction media, solvents can dissolve the reactant and change the thermodynamics of the transition state due to solvent–solute intramolecular interactions—two major properties desirable for biomass upgrading reaction.125 Other solvents, especially polar protic ones, can act as hydrogen sources for lignin upgrading. Hydrogenation with hydrogen from the solvent can occur either when the donor completely reacts or when the donor loses only a hydrogen pair. The latter mechanism is known as catalytic transfer hydrogenation (CTH).   2.4.2.1 Solvent hydrogen donor The CTH reaction only involves the hydrogen on the donor, without affecting other atoms attached to the donor. One mechanism of the CTH reaction can be through direct hydrogen transfer between a donor and a recipient, known as the Meerwein–Ponndorf–Verley (MPV) reaction, which is reversible.126 Another is the hydride mechanism, which can proceed through monohydride or 28  dihydride.127 In the former, the hydride and proton keep their identity, while in the latter, they lose it.128 The most common solvent used as a hydrogen donor for either model compounds or lignin upgrading is 2-propanol, which results in acetone as the dehydrogenated product. Hydrogenation of phenol and water-insoluble FPO was achieved through the CTH reaction with 2-propanol as a hydrogen donor, employing Raney Ni® at ~80–120 oC.129 The HDO reaction was also conducted with the addition of H-BEA zeolite to produce benzene, and the ratios of hydrogen donor to phenol as well as Raney Ni® to zeolite were suggested to be key for controlling the products’ selectivities.130 It was further found that the adsorption of 2-propanol on Raney Ni® occurs in two different active sites.131 Bimetallic Ru-Re is also able to catalyze the CTH reaction of 2-propanol to convert guaiacol to cyclohexane, where Ru acts as the CTH catalyst while Re provides acidity for the dehydration reaction.132 Another advantage of utilizing 2-propanol as a hydrogen donor in the CTH reaction is that both 2-propanol and acetone are good solvents for dissolving lignin.58,133  29  OHR1OR2MOOR1HR2OR1OHR2-H HOHMeDOMe OHMeDODMeHLnM  Ln-xMDH+++ +-LnMOHR1DR2LnMR1OR2 Ln-xMD+-LnMOHR1DR2+ H(a)(b)(c) Figure 2.3 Direct hydrogen transfer (a), monohydride mechanism (b), and dihydride mechanism (c).127 L: ligand, M: metal, D: deuterium, H: hydrogen, R: alkyl  Methanol is another hydrogen donor typically used during lignin upgrading. The upgrading of native lignin in birch wood points to the fragmentation of lignin macromolecules, which is followed by hydrogenolysis by hydrogen from methanol to produce 4-propylsyringol as a predominant product.134 Depolymerization of organosolv lignin via the CTH of methanol with a copper catalyst under supercritical conditions was found to result in cyclohexyl compounds.135 Another benefit of using methanol as the hydrogen source is the series of CTH and aqueous phase reforming (APR) reactions, starting from methanol to formaldehyde, formic acid, and finally CO2, and producing 3 mol of H2 for each 1 mol of methanol.136  Formic acid is another solvent that can act as a hydrogen donor. The CTH reaction with formic acid can be accommodated by a Pd/AC catalyst at a relatively low temperature (50C).137 However, the catalyst loses activity after several runs, possibly due to blockage by intermediate products, particle agglomeration, or metal leaching. Ni on an aluminum-substituted mesoporous 30  support (Ni/Al-SBA-15) has also been used as a catalyst for the in situ hydrogenation of lignin with formic acid as a hydrogen-donating solvent to produce syringol, syringaldehyde, vanillin, and desaspidinol as the main products.138 The depolymerization of acid hydrolysis lignin with formic acid as a hydrogen source in ethanol solvent has also been conducted using a Ru/C catalyst; the FPO yield was proportional to the amount of formic acid.139  2.4.2.2 Enhancement of reaction performance with solvent properties Numerous solvents, including nonpolar, polar aprotic, and protic, have been utilized as the reaction medium in lignin or FPO upgrading. Lignin, which has nonpolar and polar functionalities, has very poor solubility in a nonpolar solvent. However, nonpolar solvents show the highest performance in the ring saturation of phenolic compounds compared to other types of solvent.81,140 Nonpolar solvents such as n-heptane are also known to be stable under high-temperature hydrogenation reactions.141 However, in limited hydrogen conditions, nonpolar solvents may lose their hydrogen. For example, decaline can be converted to tetralin or even naphthalene.142 In addition, as most FPO upgrading targets nonpolar compounds such as paraffin, naphthenes, and aromatics, this type of solvent can provide good segregation between reactants and products. This segregation can be important to avoid coke formation, such as in the case of HMF production from fructose with biphasic solutions, where the biphasic solution can prevent the formation of humin.143,144  Aprotic solvents such as acetone, THF, and γ-valerolactone (GVL) are widely used for lignin dissolution during lignocellulose fractionation or upgrading. Lignin from corn stover was found to be soluble in GVL solution with dilute acid.69 Furthermore, it was found that the acid-catalyzed reaction is tunable by altering the extent of solvation in the initial and transition states of the catalytic process.145 In mixtures of water–aprotic solvents, such as GVL-H2O and DMSO-H2O, 31  the binding strength of the protons decreased, and thus facilitated the formation and deprotonation of carbenium. An acetone–water solution was also used to extract lignin from lignocellulose materials, leaving cellulose as a solid residue.56,133,146  However, aprotic solvents seem vulnerable to hydrogenolysis.   Besides being known as hydrogen donors, polar protic solvents are also good reaction media for lignin or FPO upgrading. Hydrogenolysis of birch wood with methanol under hydrogen pressure results in high selectivity for propyl syringol and propyl guaiacol.147 Product selectivity can then be shifted to hydroxyl propyl syringol/guaiacol by changing the catalyst from Ru/C to Pd/C.148 Beside methanol, ethylene glycol was also found to be capable of facilitating the depolymerization of a lignosulfonate lignin with Ni-based catalysts.149 It was suggested that polar protic solvents can act as nucleophiles that induce ether bond cleavage via alcoholysis.150   2.4.3 The challenges in lignin and fast pyrolysis oil (FPO) upgrading 2.4.3.1 Thermal condensation of FPO Coke formation is one of the main challenges in FPO upgrading but is rarely seen in model compound studies. Coke can form due to the thermal condensation of FPO. Heating rate and final temperature have been found to affect coke formation during thermal treatment of FPO. A slow heating rate with a low final temperature, and a fast heating rate with a high final temperature, both result in high coke formation.151 Electron spinning resonance suggests that the cause of the latter is radical recombination.152 Another study thermally treated fractionated FPO and found that both water-soluble and water-insoluble fractions of FPO produced coke and that the interaction between carbohydrates and lignin-derived compounds magnified coke formation.153  32  It is clear that during the upgrading of FPO, depolymerization and condensation are in competition. Lowering condensation and increasing the HDO rate are two potential strategies to reduce cooking. Condensation can be lowered by fractionating FPO prior to upgrading. HDO can be accelerated by adding excess catalyst. Conducting the reaction in a continuous mode may also be beneficial, as a shorter residence time with a fast heating rate will reduce the chance of condensation. A highly stable catalyst is the main factor when conducting upgrading in a continuous mode.   33  Table 2.2 Examples of upgrading with reduced coke formation No Catalyst  Solvent   Condition  Starting Material  reactant/ metal (g/g) Major Product Ref Active metal Support Amount of catalyst [active metal] (g)a Type V (mL) T  (°C) P [MPa] Reactant m (g) Major compounds Yield (wt%) 1 Ni HZSM5 2[0.4]  DI water 80  250 H2 [5]  Hexane-extracted FPO 2 5 Naphthenes NA 120 2 Pd/ MoO3 P2O4/SiO2 1.7 ×10-1 [1.3×10-3]  Decalin 7  250 H2 [1]  Water-insoluble FPO 8.6 ×10-2 67.3 Naphthenes and paraffin 46.3 (CB) 142 3 Ru Carbon 1.5 [7.5 ×10-3]  THF & n-heptane 24  150 & 250 H2 [3.45]  GVL lignin 3 40 Alkanes and oxygenated naphthenes 37.7 (CB) 141 4 Raney Ni & HBEA N/A 0.6 [0.48]  n-pentane 7  160 & 220 N2 [1]  Acid-free water-insoluble FPO 0.1 0.21 Aromatics and naphthenes N/A 130 5 Ni Carbon 0.1 [1.5 × 10-2]  Methanol 40  220 Ar [1]  Birch sawdust 2 133 4-Propylguaiacol and 4-propylsyringol 54 (LB) 154 6 Ru MWCNT 0.6 [0.03]  Water 40  300 H2 [5]  Water-soluble FPO 1.3 42.5 Alcohols, paraffin olefin, naphthenes N/A 100 7 Pt NbOPO4 0.2 [0.01]  Cyclohexane 8.29  190 H2 [5]  birch wood 0.2 20 Pentanes, hexanes, naphthenes 28.1 155 anumber outside the bracket indicate the total catalyst loading, number inside the bracket indicate the amount of the active metal; bCB: carbon based;  cLB: lignin based.  34  Table 2.3 presents the findings of several upgrading studies using a batch reactor that resulted in minor coke formation. Most of the upgrading was conducted subsequent to extraction or fractionation (entries 1, 2, 4, and 6).  In addition, almost all of the upgrading of ligneous material (except entry 5) was conducted with a high ratio of catalyst-to-reactant, in contrast to the low ratio of catalyst-to-reactant used in the model-compound upgrading.  2.4.3.2 Inhibition of catalytic activity As mentioned earlier, the depolymerization reaction should dominate over the condensation reaction to maintain catalytic activity. However, several factors can cause catalyst deactivation, including catalyst poisoning and leaching. One poisoning mechanism is coking. Slightly different from thermal condensation, catalyst coking occurs due to the deposition of carbon-rich materials on the catalyst’s surface. Furfural, a carbohydrate-derivative compound, was found to be adsorbed strongly on a Ni/Al2O3 catalyst, resulting in high carbon deposits.156 Unfortunately, furfural is more easily adsorbed on the catalyst than guaiacol, a lignin-derived compound, which might be the reason for coke formation during FPO upgrading.157 Moreover, polyoxygenated compounds such as catechol have a higher tendency to form coke, especially on sulfide catalysts.158,159 High H2 pressure has been suggested as a means of lowering coke formation.159 Changes in catalyst structure can also lead to catalyst poisoning. The presence of water was found to affect catalyst structures. As previously mentioned, hot water poisons Y-zeolite through hydrolysis of Si-O-Si bonds.115 It has also been suggested that water removal during FPO upgrading can reduce the risk of catalyst poisoning.160  35  Several factors can affect the activity of hydrogenation catalysts. Deposition of alkali metal, present as ash in FPO, can poison the catalyst.161 The acidity and corrosivity of FPO have been suggested as other factors leading to catalyst deactivation via leaching.162 With transition metal catalysts such as Ni, carboxylic acid was found to deactivate the catalyst.129 Sulfur was also found to poison Pt catalyst through agglomeration, making it difficult to regenerate.163 Sulfur poisoning is an issue when upgrading Kraft lignin and acid hydrolysis lignin since both contain sulfur after being exposed to sodium sulfide or sulfuric acid, respectively.       2.5 Aqueous-phase reforming of glycerol Reforming is a reversible reaction between hydrocarbon compounds and water to produce hydrogen and carbon monoxide, also known as syngas. The most common process is Steam Methane Reforming (SMR). In this process, methane reacts with water to produce CO and H2 in the presence of a metal catalyst, as shown in Equation 2.1.164 As can be seen, this reaction is very endothermic. CO can then react with water, producing CO2 and H2; this is called the Water Gas Shift (WGS) reaction and is shown in Equation 2.2. Since SMR is very endothermic and WGS is slightly exothermic, the two steps are usually performed separately during conventional methane reforming to obtain high H2 yield. First, SMR is conducted in a high temperature reactor (>700 °C) and then the WGS reaction is conducted in a more moderately heated reactor (~500 °C). Another strategy to enhance SMR reforming is to remove one of the products so as to shift the equilibrium, as per Le Chatelier’s principle.165 For example, addition of an absorbent such as calcium carbonate or dolomite can provide in-situ separation by capturing CO2.  CH4 + H2O    CO + 3 H2  ΔH0 = 206 kJ/mol C (2.1) 36  CO + H2O   CO2 + H2 ΔH0 = -41 kJ/mol C (2.2)  13 C3H8O3   CO + 4/3 H2  ΔH0 = 82 kJ/mol C (2.3) CH3OH    CO + 2 H2 ΔH0 = 90 kJ/mol C (2.4) 13 C3H8O3 + H2O  CO2 + 7/3 H2 ΔH0 = 41 kJ/mol C (2.5) CO2 + 4 H2   CH4 + 2 H2O ΔH0 = -165 kJ/mol C (2.6)  Unlike hydrocarbon compounds, the reforming of oxygenated compounds is less endothermic. As can be seen in Equations 2.3 and 2.4, the enthalpies per mol carbon (mol C) of the reforming reactions for glycerol and methanol are lower than for the reforming of methane. The equilibrium constant presented in Figure 2.4 shows that the glycerol APR reaction is thermodynamically plausible at lower reaction temperatures. It can be seen that the equilibrium constants of SMR and WGS intersect near the equilibrium line, meaning that relatively high CO content is likely to be found in SMR. On the other hand, since reforming of oxygenated compounds can be conducted at relatively lower temperature, subsequent WGS is likely to occur resulting in low CO content. The complete glycerol APR reaction is shown in Equation 2.5. However, at low reaction temperatures methane formation is also probable. The methane can be form through reverse SMR reaction or Sabatier reaction, the reaction between CO2 and H2 to produce CH4 as shown in Equation 2.6.  Besides low-temperature operation and low CO content, the reforming of oxygenated compounds has several advantages, including the nonflammability of feedstocks, the possibility of using a single chemical reactor, and the minimization of undesirable reactions at high temperatures.13 The low-temperature reforming of oxygenated compounds is called aqueous-phase reforming (APR) because water is generally in the liquid phase during the reaction. 37   Figure 2.4 The thermodynamic equilibrium constants of reforming-related reactions Glycerol, one of the by-products of triglyceride transesterification in biodiesel production, is a promising feedstock for the APR reaction, as its already affordable price is declining due to overproduction.166 In addition, glycerol is abundant, renewable, and nontoxic.167 The reaction mechanism of APR is complex. First, the dehydrogenation of hydroxyl groups from the primary carbon of glycerol leads to the formation of glyceraldehyde. Glyceraldehyde then undergoes decarbonylation, producing CO and ethylene glycol.168,169 Ethylene glycol APR occurs through C-C cleavage, followed by dehydrogenation, producing CO.13 It was found that shorter molecules with the same C:O molar ratio are more selective for H2 and CO2.14 Active catalysts for the APR reaction are similar to those for the hydrogenation reaction. Pt and Ni were found to be the most active metals, followed by Ru, Rh, Pd, and Ir.170 The addition of a secondary acidic metal has yielded impressive improvement in the APR. With a Pt catalyst, the addition of Re was found to increase the catalytic activity by enhancing the WGS reaction through the promotion of H-OH cleavage.171 Adding Re to Pt, however, decreased the selectivity for H2.172 38  A highly acidic catalyst will result in the dehydration of the oxygenated compounds to form alkanes.173–175 Adding Ni and Co to Pt was also found to increase the APR activity of a catalyst; the suggested mechanism was decreased heat of adsorption of CO and hydrogen through modification of the electronic structure (lowering of the d-band center of Pt).176 The addition of Sn to Raney Ni® increased the selectivity of H2 by inhibiting CO dissociation, which led to the formation of methane.177  2.6 Summary Lignocellulosic biomass consists of cellulose, hemicellulose, and lignin, with the percentage composition very dependent on the type of biomass. The properties of lignin make it the most attractive of the three for use as a renewable fuel. Lignin utilization can be carried out through the isolation of lignin or the pyrolysis and liquefaction of lignocellulose biomass. During isolation, lignin may condense to form a large macromolecule with strong C-C bonds. Pyrolysis and liquefaction produce bio-oil, a mixture of hundreds of different compounds with high O:C and low H:C ratios. Both routes require a further process known as upgrading, which is usually studied using model compounds to reduce the complexity of analysis. Catalysts and solvents play important roles during the upgrading of lignin-derived compounds.  Depending on their functionality, catalysts can be classified into hydrogenation and dehydration catalysts. Hydrogenation catalysts are usually transition metals or noble metals, while dehydration catalysts are acidic catalysts. The hydrogenation of phenolic compounds usually results in combinations of cyclic alcohols and cyclic ketones. In the presence of excess hydrogen, cyclic alcohol is usually the predominant product. Particular catalysts under specific reaction conditions 39  can produce highly selective cyclic ketones, such as cyclohexanone, which is useful as a nylon precursor. Acid catalysts are known to increase the dehydration rate of phenolic compounds, resulting in naphthenes and aromatics. Hydrodeoxygenation with an acid catalyst usually occurs through hydrogenation, which produces cyclohexanols or cyclohexanones, followed by dehydration, producing naphthenes. Another route for upgrading is direct deoxygenation, using a combination of metal catalysts to stretch the bonds between oxygen and aromatics in phenolic compounds.  Solvents are as important as catalysts during upgrading. Several solvents, especially polar protic ones, can act as hydrogen donors during catalytic transfer hydrogenation, which can occur through direct hydrogenation or a hydride mechanism. A solvent’s ability to dissolve the relevant reactant is another important property. In the upgrading process, nonpolar solvents are useful for dissolving the products but have poor solubility toward lignin or FPO. Aprotic solvents are generally very good for dissolving lignin or FPO, but tend to be unstable at high reaction temperatures. Due to their ability to donate hydrogen, protic solvents can induce ether bond cleavage but are poor at dissolving lignin or FPO.  Upgrading the actual FPO or lignin presents other challenges. With FPO, thermal treatment results in a condensation reaction that forms char. Inhibition of the catalysis reaction is another challenge during FPO or lignin upgrading and can occur due to coking when carbon is deposited on the catalyst surface. Changes in the catalyst structure during the reaction can also inhibit catalysis.   40  Chapter 3: Experimental    3.1 Materials 3.1.1 Solvents, gas and standard Dimethyl sulfoxide (DMSO, 99.5%), glycerol (99.5%), 2-propanol (99.5), cyclohexanol (99%), cyclohexanone (99.5%), benzene (99.9%), 1,2-propylene glycol (99.5%), ethylene glycol (99%), ethanol (99.9%), methanol (99.9%), γ-valerolactone (99%), acetic acid (99%), cyclohexane (99.9%), benzene (99.9%), and toluene (99.9%) were purchased from Sigma-Aldrich (Ontario, Canada). Acetone (99.7%), ethyl acetate (99%) and formic acid (99%) were purchased from Fisher Chemicals (Ontario, Canada). Naphthalene (99%), n-heptane (99%), 1-butanol (99%) and n-decane (99%) were purchased from Alfa Aesar (MA, USA). Ultra-high purity helium was purchased from Praxair (Ontario, Canada). All the solvents were used as received.  3.1.2 Ligneous materials and model compounds Lignin monomer and dimer model compounds: eugenol (99%), vanillin (99%),  4-hydroxy-3-methoxy-benzyl alcohol (98%) 2-methoxy-4-methylphenol (98%), 2-methoxypropyl-4-propylphenol (99%), 4-propylphenol (99%), 4-methylphenol (99%), 4-methoxyphenol (98%), benzyl phenyl ether (98%), biphenyl (99%), dibenzyl ether (98%), diphenyl ether (99%), and 4-ethylguaiacol (99%) were purchased from Sigma-Aldrich (Ontario, Canada). Phenol (99%) was 41  purchased from Alfa Aesar (MA, USA). All lignin model compounds were used as received. Fast pyrolysis oil (FPO) was purchased from BTG Bioliquid (Enschede, The Netherland). A mixture of spruce, pine, and fir (SPF) was obtain in the pellet form. The SPF pellets were crushed using a hammer mill (Model SM 100, Retsch Inc., PA, USA) and sieved to obtain 500 – 1000 micron particles.   3.1.3 Catalysts Raney Ni® 2800, 5 wt% Pt/C, 5 wt% Ru/C, 5 wt% Pd/C and Amberlyst-15 were purchased from Sigma Aldrich (Ontario, Canada). These catalysts were used as received. NH4-ZSM-5 was purchased from Alfa Aesar. H-BEA (CP 814) was purchased from Zeolyst with ammonium as the cation. NH4-ZSM-5 and CP 814 were calcined at 550°C for 3 h with a heating rate of 10°C/min under air prior to use. This calcination was conducted to remove ammonia and convert the cation to hydrogen (H-ZSM-5 and H-BEA).  Table 3.1 Catalyst properties Catalyst Number of acid sites (mmol/g) Active particle diameter (nm) Metallic dispersion (%) BET surface Area  (m2 g-1) Average mesopores diameter (nm) Pore volume (cm3 g-1) Raney Ni - 389 0 98.6 <3.8a 0.072 5 wt% Pt/C - 4 28 1823 4.5 1.612 5 wt% Ru/C - 6 22 776 6.4 0.897 5 wt% Pd/C - 4 25 762 5.3 0.758 H-ZSM-5 0.43 - - 462 4.8 0.349 H-BEA 1.23 - - 651 8.1 1.067 Amberlyst-15 4.25 - - 105 11.9 0.430 a BJH curve did not result in full peak  42  Catalyst properties are presented in Table 3.1. The number of acid sites was measured by NH3 desorption. Active particle diameter and metallic dispersion were calculated from CO chemisorption. Surface area and average pore diameter were calculated from N2 adsorption and desorption. The possibility of external diffusion limits when using Raney Ni® was tested by changing the stirring speed (results shown in Fig C.2). The results indicate that the effect of external diffusion can be neglected at stirring speeds above 500 rpm. The potential for internal mass transfer limitations were assessed using Weisz-Prater criterion and it was found that internal diffusivity can be neglected (see Appendix A.6).   3.2 Experiments procedures  3.2.1 Reactor configuration  All experiments were conducted in a 50-mL pressure stirred reactor from Parr Instrument Co.. The open system reaction was conducted by flowing helium gas during the reaction. The outlet of the reactor was connected to a reflux condenser and back pressure regulator in order to maintain an elevated pressure. The flow rate was controlled by adjusting the pressure differential between the He cylinder regulator outlet and back pressure regulator. The outlet of the backpressure regulator was connected to a volatile trap, which contained acetone and DMSO, as an internal standard. This volatile trap was cooled in an ice water bath and was used to trap any condensable volatiles that might have been swept by He. The open system reactor configuration is shown in Scheme 3.1. The system was closed by disconnecting the reflux condenser and installing a plug on the reactor head.  43   Scheme 3.1 Reactor assembly for open system reaction. The figure is not to scale.  3.2.2 Aqueous phase reforming (APR) The glycerol APR mechanism study was conducted by mixing glycerol, 1,2-propylene glycol, ethylene glycol, ethanol, or methanol (2.0 g) with water (30.0 g) and Raney Ni® (1.0 g). Prior to the reaction, the mass of the loaded reactor was recorded (wo) and air was removed by purging the reactor with 0.69 MPa helium four times. During the reaction, the reactor was kept at 220 °C for 120 min with a stirring speed of 500 rpm. Upon completion, the reactor was submerged in an ice bath to quench the reaction. Gas products were collected in a gas bag for further analysis and then the mass of the reactor was measured (wf). The change in reactor mass, wo-wf, was used to assess the total mass of gas products produced. A liquid aliquot (1.5 g) was taken for glycerol products quantification and another liquid aliquot (1.5 g) was taken for GC/MS qualification. The 44  experiments evaluating time dependence of glycerol APR were conducted in a closed system with glycerol (2.0 g), water (30.0 g), and Raney Ni (1.0 g) at 220°C.  Reaction time was varied from 0-120 min in 30 min interval.   3.2.3 In-situ glycerol APR and phenol hydrogenation (IGAPH) The temperature effect during IGAPH was assessed using phenol (1.0 g, 10.6 mmol), glycerol (0.5 g, 5.5 mmol), water (30.0 g), and Raney Ni® (1.0 g) with an initial He pressure of 0.69 MPa. Phenol and glycerol loading were selected so that the potential hydrogen production was only slightly greater than the stoichiometric requirement for phenol hydrogenation to cyclohexanol. The reaction was conducted in the closed system at 180 – 240°C with a stirring speed of 500 rpm. The reaction temperature was selected with consideration of the thermodynamic equilibriums of phenol hydrogenation and glycerol APR reaction. The effect of phenol on aqueous intermediates of glycerol APR experiments was investigated in the closed system with phenol (2.0 g, 21.2 mmol), water (30.0 g), Raney Ni® (1.0 g) at 220 °C for 120 min. Hydrogen was provided by adding glycerol, 1,2-propylene glycol, ethylene glycol, ethanol, or methanol (2.0 g) was to the reactor. The effect of phenol to glycerol ratio was investigated using glycerol (2.0 g, 21.7 mmol), water (30.0 g), and Raney Ni® (1.0 g) at 220°C with a 0.69 MPa He pressure in the closed system. Phenol loading was varied to achieve Phenol to Glycerol ratio (P/G) of 0 – 4.88 mol/mol. The effect of glycerol to phenol ratio was investigated using phenol (1.0 g, 10.6 mmol), water (30.0 g), and Raney Ni® (1.0 g) at 220°C with a 0.69 MPa He pressure in the closed system. Glycerol loading was varied to achieve P/G ratio of 0.1 – 3.86 mol/mol. Time dependence of IGAPH was studied using glycerol (2.0 g, 21.7 mmol), phenol (2.0 g, 21.2 mmol), water (30.0 g), Raney Ni® (1.0 g) at 45  220 °C. Reaction time was varied from 0-120 min in 30 min interval. The effect of catalyst loading was investigated using phenol (5.0 g, 53.1 mmol), glycerol (5.0 g, 54.3 mmol), and of water (30.0 g) at 220 °C. The reaction was conducted in the closed system with a 0.69 MPa He pressure for 120 min.  The study of the hydrogen transfer mechanism was conducted using the closed and open systems loaded with phenol (1.0 g), water (30.0 g), Raney Ni® (1.0 g) at 220 °C for 2 h. Glycerol loading and reaction pressure were varied. In addition, the reaction was also conducted in the open system with continuous helium flow during the reaction (Scheme 3.1) for 120 min while gas samples were taken continuously collected in the gas bag (gas bags were replaced every 30 min interval) once temperature reached 220 °C. In this experiment, phenol (2.0 g), glycerol (2.0 g), water (30.0 g), and Raney Ni® (1.0 g) were reacted at 220 °C under 4.34 MPa (the maximum pressure in the closed system when the initial pressure was 0.69 MPa) and 0.03 – 0.06 L/min helium. The model compound study was conducted with of model compound namely guaiacol, p-cresol, p-creosol, vanillin, eugenol, dibenzyl ether, diphenyl ether, or benzyl phenyl ether (10.6 mmol), glycerol (10.9 mmol), and water (30.0 g) at 220°C. The reaction was conducted in the closed system with a 0.69 MPa He pressure for 2h.  Reactions using noble metal catalysts (Pt/C, Ru/C, or Pd/C) were conducted with phenol (1.0 g, 10.6 mmol), glycerol (2.0 g, 21.7 mmol), catalyst (1.0 g) and water (30.0 g).  Each noble metal catalyst had a metal loading of 5 wt%. The reaction was conducted at 220 °C in the closed system with an initial He pressure of 3.45 MPa for 2 h. IGAPH reaction with Pt/C catalyst was conducted with phenol (1.0 g, 10.3 mmol), glycerol 2.0 g, 21.7 mmol), Pt/C (1.0 g) and water (30.0 g). In the closed system, the reaction was conducted with 46  3.45 MPa He pressure. In the open system, the reaction was conducted with a constant He pressure of 4.34 MPa while the flowrate was varied from 0, 0.03 – 0.06, and 0.15 – 0.25 L/min.   The APR pre-reaction procedure was also carried out for IGAPH experiments. Upon completion, the reactor was submerged in an ice bath to quench the reaction. In the closed system reaction, gas products were collected in a gas bag for further analysis and then the mass of the reactor was measured (mf). The change in reactor mass, mo-mf, was used to assess the total mass of gas products produced. A 1.5 g liquid aliquot was taken for glycerol quantification and the remainder of the liquid products was extracted overnight with 30.0 g of ethyl acetate in a separatory funnel at room temperature for further phenol quantification. Repeatability of the experiments is shown in Fig C.1 and C.2. The greatest standard deviation was for phenol conversion, σ=2.23 %.    3.2.4 In-situ glycerol aqueous phase reforming and phenol hydrodeoxygenation (IGAPHdo) IGAPHdo was conducted with two types of hydrogenation catalyst: Raney Ni® and Pt/C. All reactions with Raney Ni® were conducted in the closed system with an initial He pressure of 3.45 MPa while the reactions with Pt/C were conducted in the open system with 0.03 – 0.06 L/min He flow. The acid catalyst screening reactions were conducted with phenol (1.0 g, 10.6 mmol), glycerol (1.0 g, 10.8 mmol), water (30.0 g), Raney Ni (1.0 g), and acid catalyst (0.1 g). The reactions were conducted at 240 °C for 2h. The effect of H-ZSM-5 loading was investigated using phenol (1.0 g , 10.6 mmol), glycerol (1.0 g, 10.8 mmol), Raney Ni® (1.0 g), and water (30.0 g). In these reactions, the amount of H-ZSM-5 was varied from 0 to 0.1 g. The reactions were conducted at 240°C for 2 h. Reactions using intermediate model compounds were conducted with 47  cyclohexanol, cyclohexene, or phenol (1.0 g) at 240 °C for 2 h under a 3.45 MPa initial He pressure. Unless stated otherwise, attempts to produce benzene were made using phenol (1.0 g, 10.6 mmol), Pt/C (0.1 or 0.5 g), Amberlyst-15 (0.1), and water (30.0 g) at 220°C with a 3.43 MPa constant He pressure in the open system. In the attempt to obtain cyclohexane, the reactions were conducted with phenol (1.0 g, 10.6 mmol)  with glycerol (0.6, 6.5 mmol or 1.0 g, 10.8 mmol), Pt/C (1.0 g), and Amberlyst-15 (0.01 g). The reaction was conducted at 220 °C or 240°C in open system. The pre- and post- IGAPH reaction procedures were also performed before and after the IGAPHdo reactions.  3.2.5 Reaction with FPO and SPF FPO upgrading was conducted with FPO (10.0 g, containing 28.97 wt% water), Pt/C (1.0 g) and  glycerol (10.0 g). As much as 5.0 g additional solvent (n-decane) was added. The reaction was carried out at 300°C in the open system reaction with an initial 10.34 MPa He pressure. The back pressure regulator, however, was set to 19.31 MPa. Although the reaction was conducted without a He flow, the autogenous pressure occurred due to gas production which was released when the pressure exceeded 19.31 MPa (slightly above saturation pressure of methanol at 300°C). SPF hydrogenolysis was conducted with SPF (5.0 g), Pt/C (1.0 g), water (30.0 g), and glycerol (10.0 g). An additional 5.0 g of n-decane was added to the reactor.  The reaction was carried out at 300°C in the open system with an initial 10.34 MPa He pressure. The back pressure regulator, however, was set to 19.31 MPa. The reaction was conducted without a He flow. 48  Upon completion, the mixture was vacuum filtered to separate the catalyst and the solution. A two-phase system formed in the filtrate due to the addition of n-decane. The organic and aqueous phases were separated by cooling the solution to -5 °C in order to freeze the aqueous phase. The weight of each phase was measured and the weight ratio between the two-phases was recorded. A sample of each phase was taken and mixed with an internal standard for quantification. 1-Butanol was used as the internal standard for organic phase while DMSO was used as the internal standard for the aqueous phase. The separated catalyst was washed with acetone and dried overnight in a vacuum oven at 60 °C. The weight of the dried catalyst was measured and recorded.   3.3 Product analysis of IGAPH and IGAPHdo 3.3.1 GC-MS analysis for phenolic compounds Phenol conversion and product yield were analyzed by gas chromatography with a mass spectroscopy detector (Agilent 7820 A GC – 5975 MS, California US). After extraction with ethyl acetate, internal standard equal to the mass of phenol initially loaded was added. Unless otherwise specified, dimethyl sulfoxide (DMSO) was the internal standard. Acetone (30 mL) was added to improve the homogeneity of solutions. Separation of compounds was conducted using an HP-Innowax column (60 m × 0.250 mm × 0.25 µm) and the injector temperature was set at 250 °C with split ratio 100:1.  The GC oven temperature profile during model compounds analysis was: 40 °C for 5 minutes, ramp to 150 °C at 5 °C/min, ramp to 260 °C at 15 °C/min and hold at 260 oC for 5 minutes. GC oven temperature for FPO and upgrading analysis was: 40 °C for 5 minutes, ramp to 260 °C at 5 °C/min and hold for 5 minutes. Prior to quantification, a calibration curve was plotted by comparing the mole and area ratios of reactants and products to the internal standard.  49  3.3.2 HPLC analysis Glycerol conversion and product distribution were analyzed by high-performance liquid chromatography (HPLC, Younglin, Gyeonggi-do, South Korea) using a BioRad HPX-87H column (California, United States). The mobile phase was 0.005 M H2SO4 at 0.6 mL/min at 60 °C. A refractive index (RI) detector at 35 °C was employed. External calibration was conducted for quantification.  3.3.3 GC-FID-TCD for gas analysis Gas products were analyzed with a Shimadzu GC-14B with Shimadzu C-R8A Chromatopac integrator.  Two detectors, FID for hydrocarbon analysis and TCD for CO and CO2, were used in this analysis. The FID detector was equipped with Agilent HP-PLOT U column (19095P-UO4, ID: 30 m × 0.530mm × 20 μm) while the TCD detector was equipped with a packed column (5682PC, 6 ft × 1/8”, 316 stainless steel).  The GC temperature program was as follows: hold at 35 oC for 3 minutes, ramp to 120 oC at 5 oC/min, hold at 120 oC for 2 minutes, ramp to 170 oC at 10 oC/min and hold at 170 oC for 2 minutes. Quantification was conducted with external calibration under the same conditions.  3.4 Lignin, carbohydrate, and ash analysis Compositional analysis was conducted to measure carbohydrate and lignin content in SPF. The compositional analysis was conducted according to the NREL Standard for Determination of Structural Carbohydrates and Lignin in Biomass, with minor  modification.178 Briefly, dried and 50  extractive free SPF (0.3 g) was mixed with 72 wt% H2SO4 (4.92 g) in a serum bottle. The mixture was placed in a water bath at 30 °C for 60 min and stirred with stirring rod periodically. The mixture was then diluted by adding 84 g of water to achieve 30 wt% sulfuric acid. The serum tube was then capped with a Teflon cap, sealed and placed in the autoclave. The autoclave was kept at 121°C for 60 min.  Upon completion, the solution was filtered with a vacuum pump. The sugar content in the hydrolysate was analyzed by high-performance liquid chromatography (HPLC Younglin, Gyeonggi-do, South Korea). Glucose, cellobiose, and xylose were analyzed using a BioRad HPX-87H column (California, United States) with 0.005 M H2SO4 as the mobile phase at 0.6 mL/min at 60 °C. Arabinose, mannose and galactose were analyzed using a BioRad HPX-87P column (California, United States) with water as the mobile phase at 0.6 mL/min at 85 °C. The refractive index (RI) detector was set at 35°C. The amount of acid soluble lignin was analyzed by ultraviolet (UV) spectroscopy (Thermo Scientific) by measuring absorptivity at the wavelength of 240 nm. The ash content in SPF was analyzed using a muffle furnace in accordance with the NREL Standard on Determination of Ash in Biomass179. Briefly, SPF (1.0 g) was placed in a crucible and dried at 105°C until minimum weight change was observed. The oven-dry weight was then used as the initial weight. The sample was then placed in the muffle furnace and the temperature was set at 575°C with 10 °C/min for 18 h. The heating rate was set to avoid pre-ignition. Upon completion, the sample was placed in a desiccator to cool and the weight of samples was measured.  51  3.5 Reaction metrics 3.5.1 Reaction metrics for model compounds experiment In the reaction with model compounds, conversion (X) is defined as the ratio of the number of moles reacted per moles fed to the system: 𝑋 % =  𝑛𝑖 − 𝑛𝑓𝑛𝑖× 100  (3.1) Where ni is the number of moles initially loaded in the reactor and nf is the number of moles after the reaction. In the reaction with model compounds, selectivity of a product (S) is defined as the ratio of number of carbon atoms in the product to total number of carbon atoms in all detected products. Phenol product selectivity (Sip) is defined as: 𝑆𝑖𝑃 % =  𝑛𝐶𝑖∑𝑛𝐶𝑃P× 100% (3.2) where 𝑛𝐶𝑖 is the number of carbon atoms in corresponding products and 𝑛𝐶𝑝𝑝 is the total number of carbon atoms in all detected products from phenol (benzene, cyclohexane, cyclohexanol, and cyclohexanone). Glycerol gaseous product selectivity (𝑆𝑖𝑔) is defined as: 𝑆𝑖𝑔 % =  𝑌𝐶𝐺𝑔𝑎𝑠𝑛𝐶𝑖𝑔𝑎𝑠∑𝑛𝐶𝑔𝑎𝑠𝑝× 100%  (3.3) where 𝑌𝐶𝐺𝑔𝑎𝑠  is the yield of carbon from glycerol that becomes gas; 𝑛𝐶𝑖𝑔𝑎𝑠  is the number of carbon atoms corresponding gaseous product; and ∑𝑛𝐶𝑔𝑎𝑠𝑝 is total number of carbon atoms in gaseous products. Glycerol liquid products selectivity (𝑆𝑖𝑎𝑞) is defined as: 𝑆𝑖𝑎𝑞 % =  𝑌𝐶𝐺𝑎𝑞𝑛𝐶𝑖𝑎𝑞∑𝑛𝐶𝑎𝑞𝑝× 100%  (3.4) 52  where 𝑌𝐶𝐺𝑎𝑞 is the yield of carbon from glycerol present in aqueous products; 𝑛𝐶𝑖𝑎𝑞 is the number of carbon atoms of corresponding aqueous products from glycerol; and ∑𝑛𝐶𝑎𝑞𝑝 is the total number of carbon atoms of all detected aqueous products. Yield of carbon from glycerol products that becomes gas (𝑌𝐶𝐺𝑔𝑎𝑠) is calculated as:  𝑌𝐶𝐺𝑔𝑎𝑠 =𝑛𝐶𝐺0−𝑛𝐶𝐺1−∑𝑛𝐶𝑎𝑞𝑝𝑛𝐶𝐺0 × 100% (3.5) where 𝑛𝐶𝐺0 is the initial number of carbon atoms of glycerol and 𝑛𝐶𝐺1 is the number of carbon atoms of glycerol after the reaction. The yield of carbon from glycerol products that is in the aqueous phase (𝑌𝐶𝐺𝑎𝑞) is calculated as: 𝑌𝐶𝐺𝑎𝑞 =𝑛𝐶𝐺1+∑𝑛𝐶𝑎𝑞𝑝𝑛𝐶𝐺0 × 100%  (3.6)  Selectivity of glycerol to hydrogenolysis, APR, and methanation was calculated using the carbon ratio.   Glycerol carbon selectivity toward hydrogenolysis (𝐶𝐺𝑙) =  𝑁C,Gh3𝑋𝐺𝑛𝑖𝐺 × 100%  (3.7) Glycerol  carbon selectivity toward APR (𝐶𝐺𝑎) =  3𝑋𝐺𝑛𝑖𝐺−𝑁C,Gh3𝑋𝐺𝑛𝑖𝐺 𝑛𝐶𝑂2𝑛𝐶𝑂2+𝑛𝐶𝐻4 × 100%    (3.8) Glycerol carbon selectivity toward methanation  (𝐶𝐺𝑚) =  3𝑋𝐺𝑛𝑖𝐺−𝑁C,Gh3𝑋𝐺𝑛𝑖𝐺 𝑛𝐶𝐻4𝑛𝐶𝑂2+𝑛𝐶𝐻4 × 100%   (3.9) and 𝑁C,Gh = 𝑁C,PG + 𝑁C,EG + 𝑁C,EtOH + 𝑁C,MeOH  (3.10) Where 𝑁C,Gh is the total number of carbon atoms contained in glycerol hydrogenolysis products, i.e. the sum of carbon atoms in 1,2-propylene glycol (𝑁C,PG), ethylene glycol (𝑁C,EG), ethanol (𝑁C,EtOH), and methanol (𝑁C,MeOH). 53  The primary goal of IGAPH is hydrogenation of phenol, therefore consumption of hydrogen by phenol hydrogenation relative to glycerol hydrogenolysis (𝑆𝑃/𝑃𝐺) is a critical metric:   𝑆𝑃/𝑃𝐺 = 𝑁𝐻,𝑃𝑁𝐻,𝑃+𝑁𝐻,𝐺  (3.11) and 𝑁𝐻,𝑃 = 𝑛𝐶𝑙𝜈𝐻,𝐶𝑙 + 𝑛𝐶𝑛𝜈𝐻,𝐶𝑛 + 𝑛𝐵𝜈𝐻,𝐵   (3.12) 𝑁𝐻,𝐺 = 𝑛𝑃𝐺𝜈𝐻,𝑃𝐺  + 𝑛𝐸𝑡𝑂𝐻𝜈𝐻,𝐸𝑡𝑂𝐻 + 𝑛𝑀𝑒𝑂𝐻𝜈𝑀𝑒𝑂𝐻,   (3.13) Where: 𝑛𝑝  : moles of cyclohexanol (Cl); cyclohexanone (Cn); benzene (B); o-cresol (o-c); 1,2-propylene glycol (PG); ethylene glycol (EG); ethanol (EtOH); methanol (MeOH) 𝜈𝐻,𝐶𝑙  : Hydrogen required to convert 1 mol of phenol to cyclohexanol (3 mol)  C6H6O (g) + 3 H2 (g)  C6H12O (g) 𝜈𝐻,𝐶𝑛  : Hydrogen required to convert 1 mol of phenol to cyclohexanone (2 mol) C6H6O (g) + 2 H2 (g)   C6H10O (g) 𝜈𝐻,𝐵  : Hydrogen required to convert 1 mol of phenol to benzene (1 mol) C6H6O (g) + H2 (g)   C6H6 (g) + H2O (g) 𝜈𝐻,𝑃𝐺  : Hydrogen required to convert glycerol to 1 mol of 1,2-propylene glycol (1 mol) C3H8O3 (g) + H2 (g)    C3H8O2 (g) + H2O (g) 𝜈𝐻,𝐸𝑡𝑂𝐻: Hydrogen required to convert glycerol to 1 mol of ethanol (4/3 mol)  2/3 C3H8O3 (g) + 4/3 H2 (g)    C2H5OH (g) + H2O (g) 𝜈𝐻,𝑀𝑒𝑂𝐻 : Hydrogen required to convert glycerol to 1 mol of methanol (2/3 mol)  1/3 C3H8O3 (g) + 2/3 H2 (g)   CH3OH (g)   Potential molecular hydrogen gas production (H2net, mmol) was calculated from the difference between produced hydrogen atoms (Hp) and consumed hydrogen atoms (Hc).   𝐻2 𝑛𝑒𝑡 =𝐻𝑝−𝐻𝑐2  (3.14) 54  The number of produced hydrogen atom (Hp) was calculated from the net number of hydrogen from glycerol (𝐻𝐺) and hydrogen produced from the reaction of carbon with water (𝐻𝑊). Hydrogen produced from the reaction of carbon with water was calculated by based on assumption that carbon in the gas forms CO and produces hydrogen through the reaction: CO + H2O → CO2 + H2 𝐻𝑝 = 𝐻𝐺 +𝐻𝑊  (3.15) 𝐻𝐺 = 𝐻𝐺𝑖 − 𝐻𝐺𝑓 − 𝐻𝐸𝐺 − 𝐻𝑃𝐺 − 𝐻𝐸𝑡𝑂𝐻 − 𝐻𝑀𝑒𝑂𝐻   (3.16) 𝐻𝑊 =  2 𝐶𝐺𝑎  (3.17) For 1,2-propylene glycol the reaction is: 1/3 C3O2 + 4/3 H2O → CO2 + 4/3 H2 𝐻𝑊 =  83 𝐶𝑃𝐺𝑎  (3.18) For ethanol the reaction becomes: ½ C2O + 3/2 H2O → CO2 + 3/2 H2 𝐻𝑊 =  3 𝐶𝐸𝑡𝑂𝐻𝑎  (3.19) The number of consumed hydrogen (Hc) was calculated based on hydrogen consumed by phenol and hydrogen consumed by CO to produce methane. Hydrogen for methane formation was calculated based on the reaction: CO + 3 H2 → CH4 + H2O 𝐻𝑐 = 𝑁𝐻,𝑃 + 𝐻𝑚  (3.20) 𝐻𝑚 = 6𝐶𝐺𝑚   (3.21) For 1,2-propylene glycol the reaction becomes: 1/3 C3O2 + 8/3 H2 → CH4 + 2/3 H2O 55  𝐻𝑚 =163𝐶𝑃𝐺𝑚  (3.22) For ethanol the reaction becomes: ½ C2O + 5/2 H2 → CH4 + ½ H2O 𝐻𝑚 = 5𝐶𝐸𝑡𝑂𝐻𝑚  (3.23) 𝐻𝐺𝑖  : Total moles of hydrogen in the initial glycerol loading (8niG)  𝐻𝐺𝑓  : Total moles of hydrogen in glycerol after reaction (8nfG) 𝐻𝐸𝐺   : Total moles of hydrogen in ethylene glycol in the products (6nEG) 𝐻𝑃𝐺  : Total moles of hydrogen in 1,2-propylene glycol in the products (8nEG) 𝐻𝐸𝑡𝑂𝐻  : Total moles of hydrogen in ethanol in the products (6nEG) 𝐻𝑀𝑒𝑂𝐻 : Total moles of hydrogen in methanol in the products (4nEG)  3.5.2 Reaction metrics for the experiment with FPO and SPF In the reaction with SPF or FPO, yield (Y) is defined as the ratio of products’ weight to the weight of feedstock: 𝑌 % =   𝑊𝑖 𝑊𝑓× 100  (3.24) Where Wf is the weight of feedstock loaded in the reactor (water-free FPO or water and extractive-free for SPF) and Wi is the weight of detected compounds. In the reaction with FPO or SPF, selectivity of a product (S) is defined as the ratio of product weight to total weight of all detected products. Selectivity of product (Sp) is defined as: 𝑆𝑃 % =  𝑊𝑖∑𝑊𝑙× 100%  (3.25) Where Wl corresponds to total weight of detected compounds.     56  Chapter 4: Glycerol Aqueous Phase Reforming    4.1 Introduction  Glycerol Aqueous Phase Reforming (APR) has been chosen as the means of generating hydrogen in this work for several reasons. First, overproduction of glycerol has resulted in a steep price decline. In the United States, it was found that the current glycerol market cannot absorb the abundance of glycerol produced due to increased biodiesel production.180 Second, the C/O ratio of glycerol makes it suitable for APR reaction.14 Glycerol’s physical and chemical properties, such as high boiling point and non-toxicity, also make it highly desirable for APR processing. However, low-temperature APR conditions are also favorable for hydrogenation reaction. This is especially true in a batch process as hydrogen produced during APR has a high chance of reacting with glycerol to produce alkenes. In this chapter, the reaction mechanism of glycerol APR on Raney Ni® will be investigated as a baseline for more complex reactions in the following chapters. The reaction pathways of glycerol APR will be traced by conducting APR of the major intermediate products of glycerol APR. Reaction order analysis and lumped kinetic analysis will be conducted to distinguish reaction pathways.    57  4.2 Results 4.2.1 APR of glycerol and intermediate products Glycerol APR with Raney Ni® was conducted at 220°C for 2 h with an initial He pressure of 0.69 MPa. The results of the reaction are presented in Table 4.1. Moderate glycerol conversion (73 mol%) was achieved and major products included 1,2-propylene glycol, ethylene glycol, ethanol, and methanol. CO2 and CH4 were the primary gaseous products. A negative peak was observed using the TCD detector suggesting the presence of hydrogen gas (Fig. C.4). However, the amount of hydrogen could not be quantified since the equipment could not integrate negative peaks. As oxygen was absent from the reactor, oxidation was considered impossible thus the presence of CO2 is attributed solely to APR reactions. However, since hydrogen is present, several side reactions involving hydrogen might occur and generate hydrocarbons. Methane was the major hydrocarbon observed in this study. During this reaction, the carbon balance of compounds could not be accurately measured as not all gaseous products could be measured by GC-FID-TCD. Differences in the solubility of gaseous products in the aqueous solution would lead to inaccurate calculations. Therefore, the carbon balance was assumed to 100% and selectivity was calculated based on carbon number of liquid and gas products      58  Table 4.1 Aqueous phase reforming of glycerol and glycerol APR intermediate products No Reactant X     (mol%) Selectivity (%) CO2 /CH4  Res P @ 50°C (MPa)h Aqueous  Gaseous H2netg (mmol) 1,2 PGa EGb EtOHc MeOHd AcHe 2-PrOHf  CO CO2 CH4 C2-C6 1 Glycerol 73 16 11 10 2 - -  1 30 28 2 1.10 10 2.92 2 1,2-propylene glycol 47 N/A - 44 - 1 -  - 25 28 2 0.90 9 2.74 3 Ethylene Glycol 36 - N/A 3 7 - -  1 40 45 4 0.90 8 3.25 4 Ethanol 41 - - N/A 6 7 9  - 23 51 2 0.45 7 3.22 5 Methanol 50 - - - N/A - -  1 39 57 3 0.68 20 4.08 a1,2-propylene glycol. bethylene glycol. cethanol. dmethanol. eacetic acid. f2-propanol. g Potentially produced H2 (Eq 3.14). hresidual pressure measured after the reactor was cooled to 50°C. Experiments were conducted with of reactant (2.0 g), water (30.0 g) and Raney Ni® (1.0 g) at 220 °C for 2 h with initial He pressure of 0.69 MPa. 59  APR of 1,2-propylene glycol, ethylene glycol, ethanol, and methanol was conducted to elucidate the complicated pathways of glycerol APR. APR of 1,2-propylene glycol resulted in lower conversion (47 mol%) and ethanol was the primary liquid product accompanied by small amounts of acetic acid. Trace amounts of acetone, methanol, acetaldehyde, and 1-propanol were found during GC analysis but were below the limit of quantification. The ratio of CO2 to CH4 was lower compared to glycerol APR. Ethylene glycol conversion was lower (36 mol%) compared to glycerol or 1,2-propylene glycol and produced small amounts of ethanol and methanol. Methanol was not produced by 1,2-propylene glycol APR. Although the ratio of CO2 to CH4 produced was similar to that from 1,2-propylene glycol APR, the selectivity of gaseous products from ethylene glycol APR was higher. APR of ethanol resulted in 41 mol% conversion and produced methanol, isopropanol, acetone and acetic acid as liquid products. In addition, acetone and acetaldehyde were also found in trace amounts. The selectivity of acetic acid was higher than the selectivity of methanol. The ratio of CO2/CH4 in the reaction with ethanol (0.45) was the lowest of all tested reactants.  The conversion of methanol during APR was the second highest (50 mol%) after glycerol. Only gas phase products were produced and the CO2/CH4 ratio was relatively low (0.68).  4.2.2 Time dependence of glycerol APR  To gain insight into the selectivity of intermediate products and kinetics, time dependent reactions of glycerol APR were conducted and are presented in Fig.4.1. The moment that the reactor reached target temperature was taken as t=0 min. During the course of glycerol APR, glycerol conversion increased steadily. 1,2-propylene glycol, ethylene glycol, and ethanol were detected at all time levels but methanol was only found after 120 min. The selectivity of 1,2-propylene glycol and 60  ethylene glycol consistently decreased as time increased while no clear trend was observed for ethanol.    Figure 4.1 Reaction metrics from glycerol APR at 220 °C as a function of reaction time. Glycerol, water, and Raney Ni® loadings were 2.0 g, 30.0 g, and 1.0 g, respectively. Panel (a) presents glycerol conversion (equation 3.1, right axis), potentially produced H2 (equation 3.14, outer left axis) and the selectivity of aqueous products from glycerol APR (equation 3.4, bars, inner left axis). Panel (b) presents the selectivity of gas products from glycerol APR (equation 3.3, bar, inner left axis), the ratio of CO2 to CH4 (outer left axis), and residual pressure after cooling to 50 °C (right axis).    Gaseous products from glycerol APR are presented as a function of time in Fig 4.1 b. The primary gaseous products were CO2 and CH4. The CO2/CH4 ratio was about 2.91 at 0 min, decreased 61  rapidly to 1.19 after 60 min and then decreased slowly to 1.11 after 120 min. The residual pressure increased continuously during the course of the reaction and reached 2.9 MPa after 120 min. Increasing residual pressure indicates more gas was produced as the reaction proceeded.   4.3 Discussion 4.3.1 Glycerol APR reaction pathway No ethylene glycol was found from the reaction with 1,2-propylene glycol and conversely, no 1,2-propylene glycol was found from the ethylene glycol reaction. This result indicates glycerol conversion occurs via parallel routes to either 1,2-propylene glycol or ethylene glycol. This parallel route was also suggested by previous research on glycerol APR which predicted that glycerol can undergo dehydration to acetol, which is then hydrogenated to produce 1,2-propylene glycol or dehydrogenation to produce glyceraldehyde followed by decarbonylation to produce ethylene glycol.169,172 However, direct experiment with 1,2-propylene glycol was not used in those studies. Very high ethanol selectivity was observed during 1,2-propylene glycol APR; a limited amount of acetic acid was also produced. Ethanol selectivity after ethylene glycol APR was significantly lower than that from 1,2-propylene glycol APR. Indeed, ethanol selectivity after ethylene glycol APR was lower than methanol selectivity. Selectivity after ethanol APR in descending order is 2-propanol, acetic acid, and methanol. These results indicate greater stability of the C-C bond between oxygenated and non-oxygenated carbon. It appears that C-C bond between two oxygenated carbons is easier to break which would explain why ethanol is produced during 1,2-propylene glycol APR instead of ethylene glycol. The breaking of C-C bond between two oxygenated carbon can be initiated by the dehydrogenation of the primary alcohol which then 62  results in aldehyde followed by retro aldol reaction.181 In the case of 1,2-propylene glycol almost half of carbon results in ethanol which arguably could be due to the breaking of bond between the two oxygenated carbons in the compound. Moreover, instead of resulting methanol, ethanol APR resulted acetic acid and 2-propanol as major liquid products. High selectivity of acetic acid from ethanol APR is likely produced from acetaldehyde, formed via dehydration.182 2-Propanol might be formed as the result of hydrogenation of acetone. Acetone may be the product of acetaldehyde reacting with methanol.183 The ratio of CO2/CH4 after ethanol APR was less than 0.5, which suggests that all non-oxygenated carbons are converted to methane. It is known that reforming non-oxygenated carbon, such as methane, is less thermodynamically favorable at low temperatures than reforming oxygenated carbon, such as methanol (Fig 2.4).  Methanol had the second highest conversion after glycerol. Although the CO2/CH4 ratio in methanol reaction is substantially less than 1, there is evidence that hydrogen has been produced. Stoichiometrically, 1 mol of CO2 produced from methanol APR should be accompanied by 3 mol H2 while production of 1 mol CH4 will consume 1 mol H2 therefore the minimum CO2/CH4 that could be obtained is 0.33 (appendix A1). Since the experimental CO2/CH4 ratio is approximately twice this value, this suggests that a portion of produced hydrogen did not react with methanol. In addition, the calculated net produced hydrogen gas (Equation 3.14) is positive for all tests which means that molecular hydrogen gas was produced. There is experimental evidence of hydrogen gas production in Fig C.4, a negative peak in the TCD chromatograms. A reaction network (Fig.4.2) was created based on the results of the experiments reported in Table 4.1. Glycerol is thought to decompose through either dehydration and hydrogenation to produce 1,2-propylene glycol or through dehydrogenation and decarbonylation to produce ethylene glycol. 63  The dehydrogenation and decarbonylation route of glycerol to produce ethylene glycol is more exothermic than the hydrogenation route (Table A.1). The data in Table 4.2 was used to calculate ΔGr for each pathway shown in Fig. 4.2. The ΔGr of glycerol to 1,2-propylene glycol is much lower than ΔGr of glycerol to ethylene glycol, which indicates a more spontaneous reaction, but this reaction requires hydrogen which may limit the extent of reaction. APR of 1,2-propylene glycol and of ethanol produce a very similar product profile suggesting that ethanol is an important intermediate.  APR of 1,2-propylene glycol produced a trace amount of 1-propanol which is not produced during APR of ethanol. Therefore, it is suggested that 1,2-propylene glycol either undergoes dehydrogenation and decarbonylation to produce ethanol or dehydration and hydrogenation to produce 1-propanol. As with the conversion of glycerol to 1,2-propylene glycol, it appears that limited hydrogen availability hinders the more thermodynamically favorable production of 1-propanol from 1,2-propylene glycol. Limited hydrogen results in the production of ethanol as the primary liquid product. Moreover, dehydrogenation and decarbonylation of 1,2-propylene glycol to produce methanol is considerably less exothermic than hydrogenation to produce ethanol (Table A.1 entry 8 and 6).   64  Table 4.2 Enthalpy and entropy of formation of compounds involved in glycerol APR reaction Compound ∆𝑯𝒇𝟎 (kJ/mol) ∆𝑺𝒇𝟎  (J/(K.mol) Ref 1,2-Propylene glycol -435.3 448.3 184 1-Propanol -255.1 322.7 164 2-propanol -272.6 309.2 164 Acetaldehyde -166.1 0.2638 164 Acetic Acid -432.8 283.5 164 Acetone -217.1 295.3 164 Carbon dioxide -393.51 213.8 164 Carbon monoxide -110.53 197.66 164 Ethanol -234.8 281.6 164 Ethylene Glycol -392.2 303.8 164 Glycerol -578.8 382.3 185 Hydrogen 0 130.6 164 Methane -74.6 186.3 164 0 Standard condition (25°C, 1 atm) The least amount of liquid product was produced during ethylene glycol APR. This is in agreement with other ethylene glycol APR studies13,170 and thus similar pathways were suggested. In this reaction, ethylene glycol was dehydrogenated and decarbonylated followed by either further dehydrogenation to produce CO or hydrogenation to produce methanol. Ethylene glycol can also be dehydrated and hydrogenated to produce ethanol. The reaction toward ethanol has the lowest ΔG but requires hydrogen thus the methanol pathway is favored. The reaction toward CO has relatively low ΔG and does not require hydrogen. The experimental result where relatively low selectivity of liquid products (10 %) found in the reaction with ethylene glycol suggests that this route is the most preferable. Once CO is produced, the reaction can further progress to produce CO2 or methane.   65   Figure 4.2 Predicted reaction pathways of glycerol APR. Compounds in red are hypothetical compounds that were not detected. Compounds in blue were found in trace amounts. The Gibbs free energy (ΔG kJ/mol) at 220 °C is in black. Each reaction is numbered using blue.  A bold arrow indicates the primary route.  CO CO2H2O H2CH4H2OH2H2OH2CO BLOCKOHHOC-C clevageMeOH BlockCH2OH* CO BlcokH2EtOH BlockH2H2OH2EG BLOCKEtOH BLOCKOHHOOHOOOHCH3OHH2OCH3OHCH4CO2CH4CO BlockCH4MeOH BlockH2OHH2CO BlockCH4H2OH2MeOH BLOCKOHH2OHOHOEtOH BlockCOH2OH2OH1,2 PG BLOCKOHOHHOOHOHO(see CO Blcok)(see EG Block)CO OHHOH2H2OH2OHOH(see 1,2 PG Block)Glycerol APR PathwaysH2 H2O H2H2-47-160-10210H2+CO-28-67-4813 -6-82-78-932-18-118-21-100-80H2-32H2-67H2123456789101113H2+CO12 14151617H2O66  4.3.2 Kinetics analysis of glycerol APR Kinetic analysis was conducted by first determining the reaction order of glycerol APR using the following equations: −𝑑[𝐺]𝑑𝑡= 𝑘[𝐺]𝑛  (4.1) −[𝐺]−𝑛𝑑[𝐺] = 𝑘 𝑑𝑡  (4.2) If n ≠1 then:  1𝑛+1[𝐺]−𝑛+1 = 𝑘 𝑡 + 𝐶 (4.3) Else: −𝑙𝑛[𝐺] = 𝑘 𝑡 + 𝐶 (4.4) Where [G] is glycerol concentration at reaction time t and n is the reaction order. Polynomial fitting was then conducted with MATLAB® and correlation coefficient (R) was plotted against the reaction order as shown in Fig 4.3. From the Fig 4.3, it can be seen that the reaction order with the highest R2 is 1.06.   Figure 4.3 Searching for reaction order of glycerol APR 67  The kinetic analysis was conducted using the time dependent reaction metric data. The model was developed based on the pathways study which found that glycerol APR proceeds through parallel production of 1,2-propylene glycol and ethylene glycol. Lumped kinetic analysis was then conducted with pseudo 1st order reactions of 1,2-propylene glycol and ethylene glycol as both intermediates ultimately produced similar gaseous products. The schematic of the reaction model is shown in Scheme 4.1. The pseudo 1st order reaction is considered appropriate since the reaction order search found the best-fitting reaction order is 1.06.   Scheme 4.1 Reaction model for kinetic analysis  Based on the model and reaction order analysis, a set of ordinary differential equation (ODE) were posed (equation 4.5-4.9). The ODEs were solved with the Runge-Kutta method and the ODE optimization was conducted by least square method using trust-region-reflective algorithm by MATLAB.  ODE model: 𝑑[𝐺]𝑑𝑡= −𝑘𝐸𝐺1 [𝐺] − 𝑘𝑃𝐺1 [𝐺]    (4.5) 𝑑[𝐸𝐺]𝑑𝑡= 𝑘𝐸𝐺1 [𝐺] − 𝑘𝐸𝐺2 [𝐸𝐺]   (4.6) 𝑑[𝑃𝐺]𝑑𝑡= 𝑘𝑃𝐺1 [𝐺] − 𝑘𝑃𝐺2 [𝑃𝐺]    (4.7) 68  𝑑[𝐹𝑃]𝑑𝑡= 𝑘𝐸𝐺2 [𝐸𝐺] + 𝑘𝑃𝐺2 [𝑃𝐺]    (4.8) [𝐹𝑃] = [𝐺𝑖] − [𝐺] − [𝐸𝐺] − [𝑃𝐺]  (4.9) Where [G], [EG], [PG], and [FP] represent the concentration of glycerol, ethylene glycol, 1,2-propylene glycol, and further degradation products respectively. [Gi] represents initial glycerol loaded to the reactor.   Figure 4.4 Evolution of glycerol, 1,2-propylene glycol, ethylene glycol, and subsequent products during glycerol APR at 220 °C experimentally and as predicted by Equations 4.5 – 4.9. The sum of squared error (R2) for glycerol, ethylene glycol, 1,2-propylene glycol, and further degradation products are 0.9991, 0.9707, 0.8397, and 0.9888 respectively  The model predictions are plotted with the experimental data in Fig 4.4. It can be seen that the model has relatively good coefficient of determination. The estimated reaction constants are presented in Table 4.3. Although the result shows that reaction constant of ethylene glycol 69  formation was slightly lower than propylene glycol formation, the standard error of this estimation is quite high. The model shows that the degradation of ethylene glycol and 1,2-propylene glycol was faster than the formation of both compounds.  Table 4.3 Rate constant of glycerol APR reaction at 220°C with 1st order   k (min-1 gcat-1) kEG1 0.006 (0.002) kEG2 0.016 (0.009) kPG1 0.006 (0.002) kPG2 0.021 (0.010) *parentheses indicate standard error calculated from 95 % confidence interval.    4.3.3 Undesired methane formation Besides being a good catalyst for APR, Ni is also known as a good catalyst for hydrogenation. Since the reaction was conducted in a batch reactor, it is possible that hydrogenation will occur and produce methane. The first possible pathway is direct methanation in which hydrogen reacts with intermediates (methanol, ethanol, or CO) to produce methane and the second possible pathway is hydrogenation of CO2 to produce methane. The latter reaction is known as the Sabatier reaction.186 From the glycerol and intermediate products APR study, it can be seen that the selectivity of CO2, an indicator of APR, was low and methane selectivity was high. From thermodynamics, the methanation reactions of CO, CO2, and methanol are very exothermic therefore, whenever hydrogen is available, the methanation reaction will occur. Removal of hydrogen is required to prevent methanation. In situ utilization of hydrogen from glycerol APR 70  may be beneficial as hydrogen will be scavenged and the chances of hydrogen reacting with glycerol or intermediates will be suppressed.   4.4 Conclusion  It was confirmed that glycerol APR on Raney Ni® occurs through parallel reactions of dehydration and hydrogenation to produce 1,2-propylene glycol and of decarbonylation to produce ethylene glycol. This parallel reaction is attributed to the greater difficulty of cleaving the bond between hydroxylated carbon and non-oxygenated carbon relative to the bond between two hydroxylated carbons.  This disparity prevents the formation of ethylene glycol from 1,2-propylene glycol. The APR reaction through 1,2-propylene glycol mostly proceeds through the formation of ethanol while the reaction through ethylene glycol mostly produces gaseous products. It was found that production of 1,2-propylene glycol from glycerol is slightly preferable over the production of ethylene glycol. Direct methanation of glycerol and intermediates and the Sabatier reaction between CO2 and hydrogen are two possible CH4 formation pathways. Conducting in-situ hydrogenation might potentially limit the production of methane from glycerol.    __________________ *A version of this Chapter has been published:  Putra, R. D. D.; Trajano, H. L.; Liu, S.; Lee, H.; Smith, K.; Kim, C. S. In-situ glycerol aqueous phase reforming and phenol hydrogenation over Raney Ni®. Chemical Engineering Journal 2018, 350 (May), 181–191.   71  Chapter 5: In-situ Glycerol Aqueous Phase Reforming and Phenol Hydrogenation*    5.1 Introduction Lignin hydrogenolysis to monomers has been proposed as an alternative for clean energy and renewable chemicals. This process consumes significant quantities of hydrogen; by one estimate it will require 62 g of hydrogen to upgrade 1 kg of lignin-derived biooil11. Unfortunately, hydrogen is mostly produced from nonrenewable sources such as natural gas, petroleum, and coal.12,187 In Chapter 4, it was shown that APR of glycerol produced hydrogen but hydrogen selectivity and yield was reduced due to methanation. In this chapter, in-situ aqueous phase reforming and phenol hydrogenation (IGAPH) was investigated. In addition to evaluating the possibility of upgrading FPO or lignin with hydrogen from glycerol APR, this study also evaluated the changes to APR selectivity due to the addition of phenol as a hydrogen scavenger.  First, thermodynamic analysis was applied to understand the complex multi-reaction system and to select the range of reaction temperatures. The effect of phenol on glycerol APR was observed to understand the synergistic effects of in-situ reaction. The effect of reactant concentration and reaction time was investigated in order to decouple the two reactions and examine the combined reaction mechanism. IGAPH was conducted in an open system with flowing He in order to understand the hydrogen transfer mechanism during IGAPH. Process parameters including initial pressure, catalyst loading, and 72  stability were also studied. Additional model compounds, both monomers and dimers, were tested to better understand the reactivity of common lignin derivatives. In addition, IGAPH reaction with several noble metal catalyst was also conducted. Finally, a kinetic analysis was conducted to determine the rate-limiting step.    5.2 Results and discussion 5.2.1 Thermodynamic analysis of IGAPH reaction Glycerol APR is an endothermic reaction thus increasing reaction temperature will increase the concentration of products at equilibrium. In contrast, phenol hydrogenation is an exothermic reaction, which is less favorable at elevated temperatures, especially when phenol or hydrogen is limited. Figure 5.1 shows the plot of the equilibrium constant (ln K = -ΔG°/RT) of several reactions involved in the IGAPH system. From Fig. 5.1, it can be seen that above 250°C the equilibrium constant of phenol hydrogenation is negative indicating that substantial hydrogen is required to increase phenol conversion. However, increasing glycerol to increase hydrogen supply for phenol hydrogenation will not help since it was shown in Chapter 4 that the selectivity to hydrogen from glycerol APR with Raney Ni® was low at 220℃.   73   Figure 5.1 Equilibrium constant (ln K = -ΔG°/(RTnC)) of several reactions involved in IGAPH calculated per mol carbon on the compound.     From the equilibrium constants, glycerol hydrogenolysis to produce methane or methanation is the most favorable reaction. Thermodynamically, in the absence of phenol, once hydrogen is produced by glycerol APR there is a strong possibility of this hydrogen reacting with glycerol itself to eventually produce methane. Glycerol to ethylene glycol, the route which yields the greater amount of hydrogen, is endothermic while glycerol to 1,2-propylene glycol and hydrogenation of glycerol derivatives are exothermic. Although the production of 1,2-propylene glycol has a higher equilibrium constant and lower Gibbs free energy (Fig. 4.2) compared to the production of ethylene glycol, low availability of hydrogen may hinder this reaction, as discussed in section 4.3.1. APR of phenol is unlikely to occur below 300°C due to the relatively low equilibrium constant and relatively high Gibbs free energy compared to glycerol APR (Table A1 and A2, Appendix A). In addition, results in Chapter 4 indicated that C-C cleavage between oxygenated 74  carbon atoms is the easiest. Therefore, APR of glycerol is favored relative to phenol since all carbon in glycerol is attached to a hydroxyl group.  Inspection of the equilibrium constants confirms that increasing reaction temperature favors the forward reaction of endothermic APR reactions. Unfortunately, in the case of limited hydrogen availability the higher reaction temperature may limit the maximum conversion of phenol since the thermodynamic equilibrium of phenol hydrogenation to cyclohexanol will decrease.   5.2.2 Temperature effect on IGAPH The effect of temperature (180 °C – 240 °C) on glycerol conversion, selectivity of non-condensable products from glycerol, carbon selectivity, and residual pressure during IGAPH are presented in Fig. 5.2(a). Glycerol conversion increased with increasing reaction temperature. Glycerol conversion was 92 mol% at 180 °C and reached 100 mol% at 200 °C. Several glycerol liquid intermediate products were found: 1,2-propylene glycol, ethylene glycol, methanol, and ethanol. The selectivity of glycerol liquid intermediate products, shown by the bar graph in Fig. 5.2 (a), decreased from 21% to 3% as temperature increased from 180 °C to 240 °C.  The selectivity to 1,2-propylene glycol and ethylene glycol decreased considerably from 9% to 0% and from 5% to 0%, respectively.  Ethanol selectivity decreased from 6% at 180 °C to 2% at 240 °C.  However, selectivity to methane increased from 13% at 180 °C to 28% at 240 °C. Glycerol APR carbon selectivity increased from 65% at 180 °C to a maximum of 73% at 220 °C and then decreased to 67% at 240 °C.  The decrease in APR selectivity has been observed by others, for example, increasing the temperature from 225 °C to 265°C lowered H2 selectivity from 74% to 68%.177 The trends in selectivity are reflected by the increase in residual pressure at 50 °C. Increased production 75  of non-condensable gases, such as CO2 and CH4, resulted in an increase in residual pressure from 1.27 MPa after IGAPH at 180 °C to 2.25 MPa after IGAPH at 240 °C.  Phenol conversion, the selectivity of phenol hydrogenation products, hydrogen selectivity to phenol hydrogenation over glycerol hydrogenolysis (𝑆𝑃/𝑃𝐺), and the ratio of CO2-to-CH4 are shown in Fig. 5.2(b).  Phenol conversion increased from 65% at 180 °C to 90 mol% at 200 °C before decreasing to 79 mol% at 240 °C.  𝑆𝑃/𝑃𝐺 increased from 0.97 mol/mol at 180 °C to 1.00 mol/mol at 220 °C but then decreased to 0.99 mol/mol at 240 °C. The phenol hydrogenation products such as cyclohexanol, cyclohexanone, benzene, and o-cresol were observed. The production of o-cresol was unexpected; however, it might result from reaction of phenol with formaldehyde, a possible degradation product of glycerol. The highest selectivity of benzene (12%) and o-cresol (6%) were observed at 240 °C. The highest selectivity of cyclohexanol (60%) and cyclohexanone (38%) were observed at 180 °C and 220 °C, respectively; a greater decrease in cyclohexanol selectivity with increasing temperature was observed. Residual pressure increased as reaction temperature increased, indicating that the total amount of non-condensable gas products increased. This was accompanied by a decrease in the ratio of CO2-to-CH4. If phenol hydrogenation reaches equilibrium, then the ability of phenol to react with hydrogen would be limited thus resulting in glycerol hydrogenation followed by methane formation.  76   Figure 5.2  Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 180 °C to 240 °C for 2 h. Phenol, glycerol, water and Raney Ni® loadings were 10.6 mmol, 5.5 mmol, 30.0 g, and 1.0 g, respectively.  Panel (a) presents glycerol conversion (equation 3.1, outer left axis), Selectivity of glycerol hydrogenolysis products (equation 3.4, bars, inner left axis), glycerol carbon selectivity (equation 3.7 – 3.9, inner right axis), and residual pressure after cooling to  50 °C (outer right axis).  Panel (b) presents phenol conversion (equation 3.1, outer left axis), the selectivity of phenol hydrogenation products (equation 3.2, bars, inner left axis), hydrogen selectivity (inner right axis), and the ratio of CO2(g) to CH4(g) (outer right axis).   There are several possibilities regarding the decrease in phenol conversion and selectivity of glycerol APR at temperatures above 200 oC (Fig 5.2). The first possibility is that equilibrium is almost achieved during the reaction with dilute phenol loading (1.0 g phenol in 30.0 g water). As 77  can be seen in Fig 5.1, the equilibrium constant of phenol hydrogenation decreases as temperature increased. The decrease of equilibrium constant with low hydrogen availability could explain the decreased phenol conversion at higher reaction temperature. As the result, the available hydrogen is taken by either glycerol or other oxygenated products to produce methane. The second possibility is that at high reaction temperature, glycerol methanation occurs faster than glycerol APR reforming or phenol hydrogenation resulting in high methane selectivity, low selectivity of glycerol APR and low phenol conversion. At higher temperature, both glycerol methanation and glycerol APR reaction are more spontaneous, as indicated by the low ΔGr (Table A1 in parentheses). Phenol hydrogenation, on the other hand, becomes less spontaneous (Table A2 in parentheses).  5.2.3 The role of phenol during IGAPH 5.2.3.1 The effect of phenol during APR of glycerol and glycerol intermediates To assess the effect of phenol on glycerol APR, the experiments reported in Table 4.1 were conducted in the presence of phenol. Conversion and product selectivity from these experiments are presented in Table 5.1.  Compared to the reaction without phenol (Table 4.1), glycerol conversion increased by 30 mol% to 95 mol%. In the presence of phenol, 1,2-propylene glycol and ethanol became the major products of glycerol hydrogenolysis. During IGAPH, the selectivity of all liquid intermediates decreased relative to glycerol APR. The decrease in ethylene glycol selectivity was the greatest. Among the gaseous products, CO2 selectivity increased by 81% while selectivity to CH4 decreased by 13% resulting in a CO2/CH4 ratio of 2.28. Phenol conversion during this reaction (Table 5.2) 78  was 84 mol% while the selectivity of benzene, cyclohexanol, and cyclohexanone were 2%, 69%, and 30%, respectively. Conversion of 1,2-propylene glycol increased from 47 mol% to 76 mol% in the presence of phenol. The addition of phenol also increased the overall gas selectivity. Although both CO2 and CH4 increased, the increase in CO2 selectivity was greater resulting in a higher CO2/CH4 ratio (1.06). Overall liquid product selectivity in this reaction decreased and ethanol was found to be the most diminished of the products (44% without phenol to 18% with phenol). Selectivity of acetic acid slightly increased (3%) while no methanol was detected. As in Chapter 4, no ethylene glycol was detected. Phenol conversion was lower (72 mol%) relative to the reaction with glycerol. The selectivity of benzene, cyclohexanol, and cyclohexanone were 6%, 69%, and 25% respectively. IGAPH with ethylene glycol also increased ethylene glycol conversion from 36 mol% (Table 4.1) to 67 mol%. The overall selectivity of gaseous products also increased. CO2 selectivity increased to 62% while CH4 selectivity slightly decreased to 29% giving a CO2/CH4 ratio of 2. Ethanol and methanol selectivity decreased to 1% and 3%, respectively. No acetic acid was detected. 77 mol% of phenol was converted during this reaction; the selectivity to cyclohexanol and cyclohexanone was 84% and 16%, respectively. No benzene was detected. Similar to the other systems when phenol is present, ethanol conversion increased to 69 mol%. Overall selectivity of gas products increased due to a substantial increase of CO2 selectivity (31%). The selectivity to CH4 slightly decreased to 50% resulting in a CO2/CH4 of 0.62. Methanol, acetic acid, and 2-propanol selectivity decreased to 1%, 8%, and 4% respectively. Phenol conversion during this reaction was 75 mol%, which is the lowest phenol conversion amongst the tested hydrogen sources. The selectivity to benzene, cyclohexanol, and cyclohexanone were 11%, 60%, and 29% respectively. 79  Methanol conversion increased to 76 mol% in the presence of phenol. The selectivity to CO2 increased to 64% while the selectivity to methane decreased to 33%. Therefore, the ratio of CO2/CH4 increased to 2. No liquid products were found. IGAPH with methanol as the hydrogen source resulted in the highest phenol conversion, 95 mol%.  Almost all of this phenol was converted to cyclohexanol (91%), but a small amount of cyclohexanone was also recovered (9%). No benzene was detected. It was shown in Chapter 4 that ethylene glycol and 1,2-propylene glycol are the major products of glycerol APR and represent two separate APR pathways. The ethylene glycol pathway favored APR while the 1,2-propylene glycol pathway resulted in greater alkanes production (Fig 4.2). The addition of phenol during APR of ethylene glycol and 1,2-propylene glycol increased their conversion by 84% and 62%, respectively. The greater improvement of ethylene glycol conversion compared to 1,2-propylene glycol in the presence of phenol may explain the higher ratio of 1,2-propylene glycol yield to ethylene glycol yield in IGAPH than in glycerol APR (Table 4.1 and Table 5.1). The increase of CO2/CH4 also indicates that the presence of phenol promotes glycerol APR over methanation.    80  Table 5.1 Conversion and product of glycerol and glycerol APR intermediates during IGAPH No Reactant X     (mol%) Selectivity (%) CO2 /CH4 H2netg (mmol) Res P @ 50°C (MPa)h Aqueous   Gaseous 1,2 PGa EGb EtOHc MeOHd AcHe 2-PrOHf  CO CO2 CH4 C2-C6 1 Glycerol 95 7 3 7 1 - -  - 55 24 3 2.28 9 3.65 2 1,2-propylene glycol 76 N/A 0 18 - 3 -  - 39 37 2 1.06 18 3.43 3 Ethylene Glycol 67 - N/A 1 3 - -  - 62 29 4 2.18 17 3.50 4 Ethanol 69 - - N/A 1 8 4  - 31 50 7 0.62 7 3.37 5 Methanol 76 - - - N/A - -  - 64 33 3 1.93 78 3.46 a1,2-propylene glycol. bethylene glycol. cethanol. dmethanol. eacetic acid. f2-propanol. gPotentially produced H2 (Eq 3.14). hresidual pressure measured after reactor was cooled to 50 °C.  Experiments were conducted with reactant (2.0 g), phenol (2.0 g), and Raney Ni® (1.0 g) at 220 °C for 2 h with an initial He pressure of 0.69 MPa. Phenol conversion and products selectivity is presented in Table 5.2  Table 5.2 Phenol hydrogenation products and selectivity by hydrogen sourcea No Hydrogen source X (mol %) C-balance Selectivity (%) Benzene Cyclohexanol Cyclohexanone 1 Glycerol 84 0.98 2 69 30 2 1,2-propylene glycol 72 0.75 6 69 25 3 Ethylene Glycol 77 0.79 - 84 16 4 Ethanol 75 0.75 11 60 29 5 Methanol 95 0.79 - 91 9 a Experiments were conducted with reactant (2.0 g), phenol (2.0 g), and Raney Ni® (1.0 g) at 220 °C for 2 h with an initial He pressure of 0.69 MPa 81  5.2.3.2 The effect of phenol loading on glycerol conversion and product distribution To understand the effects of hydrogen availability on phenol hydrogenation and the distribution of products, the IGAPH reaction was conducted at 220oC with different phenol loadings. The temperature of 220 °C was chosen as the glycerol APR carbon selectivity and hydrogen selectivity to phenol hydrogenation were highest at this temperature (section 5.2.2).  The results are presented in Fig 5.3.  In the absence of phenol, glycerol conversion was 73 mol% and increased with increased phenol loading to a maximum 95% at P/G= 0.98 mol/mol. Glycerol conversion then decreased to 89% when P/G was further increased to 4.88 mol/mol. This result emphasizes the previous observation that the presence of phenol, a hydrogen scavenger, accelerates glycerol conversion. However, a surplus of phenol may dominate catalyst active sites and inhibit glycerol APR. Selectivity of liquid intermediate products from glycerol decreased continuously from 39% without phenol to 15% at P/G= 4.88 mol/mol. 1,2-Propylene glycol was always present as the major liquid intermediate product from glycerol. Ethylene glycol selectivity was second highest in the absence of phenol but decreased significantly when phenol was added. Although ethanol selectivity also decreased as the phenol-to-glycerol ratio increased, it was less affected by phenol loading relative to ethylene glycol.  APR selectivity consistently increased as phenol-to-glycerol ratio increased. In contrast, the selectivity of methane decreased with increased phenol loading. The residual pressure changed slightly as the phenol-to-glycerol ratio increased from 0 to 0.49 mol/mol but increased by 14% at P/G=0.98 mol/mol. This increase was followed by a decrease when the phenol-to-glycerol ratio was increased to 4.88 mol/mol.  82   Figure 5.3* Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C for 2 h as a function of ratio of phenol-to-glycerol. Glycerol, water, and Raney Ni® loadings were 2.0 g (21.7 mmol), 30.0 g, and 1.0 g, respectively.  Panel (a) presents glycerol conversion (outer left axis), selectivity of glycerol hydrogenolysis products (bars, inner left axis), glycerol carbon selectivity (inner right axis), and residual pressure after cooling to 50 °C (outer right axis).  Panel (b) presents phenol conversion (outer left axis), selectivity of phenol hydrogenation products (bars, inner left axis), hydrogen selectivity (inner right axis), and ratio of CO2(g) to CH4(g) (outer right axis). *Horizontal axis is not to scale.   The conversion of phenol decreased from 90 mol% at P/G=0.24 mol/mol to 35 mol% at P/G=4.88 mol/mol.  This was anticipated due to the greater mass of phenol added to the system. The primary 83  phenol hydrogenation product was cyclohexanol, with selectivity of 81.7% at P/G= 0.24 mol/mol. The selectivity decreased, however, to only 35.0% at the highest phenol loading (P/G = 4.88). Conversely, the selectivity to cyclohexanone increased as the phenol-to-glycerol ratio increased. The lower cyclohexanol and higher cyclohexanone selectivity indicates a lack of hydrogen thus phenol tends to form compounds with lower hydrogen demand. With increasing phenol-to-glycerol ratio, more hydrogen reacts with phenol than glycerol as indicated by the increasing SP/PG. The synergistic effect during IGAPH was shown by the increase in glycerol conversion and CO2-to-CH4 ratio upon the addition of phenol (Fig. 5.3 and Table 5.1). Although the identity of the major liquid products did not change, the presence of phenol increased the conversion of glycerol. Moreover, glycerol reaction order and rate analysis of IGAPH determined that the apparent reaction order of glycerol was 0.66 while the rate constant was 0.0114 mmol0.34 L-0.34 gcat-1 min-1(R2= 0.996) as shown in Fig 5.4. Since glycerol reaction order decreased to 0.66, using a pseudo 1st order reaction would not produce an accurate prediction of glycerol APR kinetics during IGAPH (Table B.2, Appendix B.1). The IGAPH glycerol reaction order decreased and the rate constant increased compared to the reaction order (1.06) and rate constant (0.0079 min-1 gcat-1) during APR. The reduced reaction order indicates that glycerol concentration has less effect on the glycerol reaction rate during IGAPH compared to during glycerol APR.   84   Figure 5.4 Plot searching for glycerol reaction order during IGAPH  Assuming that the reactions occur on the catalyst surface, the surface reaction rate and the glycerol concentration gradient between the bulk liquid and the catalyst surface are important factors. In glycerol APR, the glycerol reaction order was approximately 1 indicating that the concentration gradient between glycerol in bulk solution and on catalyst surface controls the reaction rate. It was previously found that active sites are blocked by adsorbed H* or CO* during ethylene glycol APR.188 This H* can also be produced during glycerol APR and must be removed either by reacting with another H* and desorbing or reacting with another reactant. During glycerol APR, the only available reactants were glycerol or glycerol intermediates thus the amount of glycerol influenced the rate of removal of H*. Beside higher glycerol conversion, the presence of phenol during IGAPH also resulted in higher CO2/CH4 ratio and fewer liquid intermediate products. Faster hydrogen uptake by phenol than by glycerol seems to be the cause of increased CO2/CH4 ratio and reduced glycerol hydrogenolysis products. Further evidence for this hypothesis is provided in 85  section 5.2.6. This explanation also agrees with the decrease in apparent reaction order of glycerol as the clearing of H* from active sites would be mostly by phenol.   5.2.4 The role of glycerol during IGAPH The effect of glycerol loading on conversion and selectivity of phenol during IGAPH was observed by conducting the reaction with constant phenol loading and changing glycerol loading, as presented in Fig 5.5. In this reaction, 1.0 g phenol was loaded in the reactor and the glycerol loading was varied from 0 to 4 mol/mol. At P/G=0.1, phenol (17 mol%) and glycerol (29 mol%) conversion were very low. At this condition, most glycerol undergoes hydrogenolysis to 1,2-propylene glycol (49%) and ethanol (15%) while only a small amount forms ethylene glycol (11%) and methanol (1%). The selectivity to gaseous products was very low, as indicated by the low glycerol carbon selectivity toward APR (15%) and methane (10%). The selectivity to cyclohexanone (59%) was higher than cyclohexanol (38%) indicating poor hydrogenation performance. It is clear that at very low phenol-to-glycerol ratio, reactions only occur in the immediate vicinity of glycerol to produce intermediate products.  86    Figure 5.5*. Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C for 2 h as a function of ratio of phenol-to-glycerol.  Phenol, water, and Raney Ni® loadings were 1.0 g (10.6 mmol), 30.0 g, and 1.0 g, respectively.  Panel (a) presents glycerol conversion (outer left axis), selectivity of glycerol hydrogenolysis products (bars, inner left axis), glycerol carbon selectivity (inner right axis), and residual pressure after cooling to 50 °C (outer right axis).  Panel (b) presents phenol conversion (outer left axis), selectivity of phenol hydrogenation products (bars, inner left axis), hydrogen selectivity (inner right axis), and ratio of CO2(g) to CH4(g) (outer right axis). *Horizontal axis is not to scale.  87  Phenol conversion significantly increased when the phenol-to-glycerol ratio increased to 0.49 mol/mol and reached a maximum (95 mol%) for a phenol-to-glycerol ratio of 0.98 mol/mol. Phenol conversion decreased when the phenol-to-glycerol ratio was further increased to 1.95. Highest cyclohexanol selectivity (74%) was obtained at P/G= 0.49 mol/mol and continuously decreased as the phenol-to-glycerol ratio increased. In contrast, the lowest cyclohexanone selectivity was obtained at P/G= 0.49 mol/mol and continuously increased as the phenol-to-glycerol ratio was increased. This observation is consistent with previous results (Fig 5.3) which indicated that the amount of glycerol controls the amount of hydrogen available, which in turn affects phenol conversion and product selectivity.  Exploration of very high and very low P/G systems provides insight into active site occupation by phenol and glycerol. In the reaction with P/G = 4.88 and 10.0 g of phenol (Fig 5.3), glycerol conversion decreased. However, the CO2/CH4 ratio was still high and SP/PG was almost 1 indicating that most hydrogen was utilized to hydrogenate phenol. This suggests that the adsorption of glycerol was hindered by the high concentration of phenol. On the other hand, in the reaction with P/G = 0.1 and 10 g of glycerol (Fig 5.5), phenol conversion was very low and the primary product was cyclohexanone. The selectivity of 1,2-propylene glycol was very high and SP/PG was very low indicating that the high concentration of glycerol hindered phenol adsorption to the catalyst surface. Based on phenol and glycerol conversion at very high or very low P/G, it appears that phenol and glycerol compete for the same active sites.    Phenol hydrogenation can be conducted with very limited glycerol as shown by P/G = 3.66 (Fig 5.5), which is close to stoichiometric conditions, indicating high hydrogen efficiency. By controlling the ratio of phenol-to-glycerol, it is possible to adjust the product distribution. This 88  control over product distribution and the finding that phenol increases glycerol APR are synergies unique to the IGAPH system. At very low or very high ratios of phenol-to-glycerol, however, these synergistic effects are diminished.   5.2.5 The effect of reaction time on IGAPH The effect of reaction time toward phenol and glycerol was observed from 0 to 120 min in 30 min intervals; 0 min indicates the time when the reaction reached the desired temperature. In this set of reactions, 2.0 g (21.7 mmol) of glycerol and 2.0 g (21.3 mmol) of phenol were loaded with 1.0 g of Raney Ni® and 30.0 g of water. Glycerol conversion increased continuously as the reaction proceeded from 32 mol% at 0 min to 95 mol% after 120 min reaction as shown in Fig 5.5. 1,2-Propylene glycol and ethanol were always the primary liquid intermediates and were more stable than ethylene glycol and methanol. Carbon selectivity from glycerol indicates that at 30 min the majority of carbon from glycerol was converted to liquid intermediates (71%) and the balance (29%) formed gaseous products. However, the selectivity of APR and methanation increased significantly between 30 min and 60 min and then slowly continued to increase.  89   Figure 5.6 Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C as a function of reaction time.  Phenol, glycerol, water, and Raney Ni® loadings were 2.0 g (21.3 mmol), 2.0 g (21.7 mmol), 30.0 g, and 1.0 g, respectively.  Panel (a) presents glycerol conversion (outer left axis), selectivity of glycerol hydrogenolysis products (bars, inner left axis), glycerol carbon selectivity (inner right axis), and residual pressure after cooling to 50 °C (outer right axis).  Panel (b) presents phenol conversion (outer left axis), selectivity of phenol hydrogenation products (bars, inner left axis), hydrogen selectivity (inner right axis), and ratio of CO2(g) to CH4(g) (outer right axis).   90  Phenol conversion increased continuously during the course of reaction from 21 mol% at 0 min to 84 mol% after 120 min reaction. At 0 min, cyclohexanone was the primary product, but its selectivity decreased as time increased. Cyclohexanol selectivity increased continuously with reaction time.  The ratio of CO2-to-CH4 decreased from 2.99 mol/mol at 0 min to 2.13 mol/mol at 60 min and then plateaued at 2.30 mol/mol. Hydrogen selectivity (SP/PG) indicates more hydrogen was consumed by phenol hydrogenation than by side reactions with glycerol.  5.2.6 Determination of mechanism of phenol hydrogenation Hydrogen transfer during IGAPH could occur through the formation and desorption of molecular hydrogen followed by re-adsorption or via a hydrogen atom on the catalyst surface. To explore these possibilities, a set of experiments comparing the reaction in a stirred reactor without (closed system) and with He flow (open system) were conducted.  In addition to identifying the hydrogenation mechanism of phenol, these experiments also provide guidance for optimum reactor design.  The experiments were conducted with two different phenol-to-glycerol ratio (0.98 and 1.96 for both closed and open system), two different initial pressures (0.69 and 3.45 MPa) for the closed system, and three different reaction pressures (2.41, 4.34, and 10.34 MPa) for the open system. Fig. 5.7(a) shows phenol conversion and product selectivity for these reactions. Phenol conversion in the closed system was 91 mol% and 87 mol% when P/G= 0.98 mol/mol and 1.96 mol/mol, respectively. The lower phenol conversion at P/G = 1.96 mol/mol was expected since hydrogen is limited. In the open system, phenol conversion was 98 mol% at P/G= 0.98 mol/mol.  Even with limited hydrogen (P/G= 1.96 mol/mol), phenol conversion in the open system (89 mol%) was slightly higher than that in the closed system. Glycerol conversion (Fig 5.7 b) reached 100% in all 91  reactions. Higher liquid APR product selectivity was found in the closed system compared to the open system. In the open system with P/G= 1.96 mol/mol, glycerol was completely converted to gaseous products. Since phenol conversion is higher in the open system than in the closed system, it can be inferred that gas sweeping does not affect hydrogen availability. Therefore, phenol hydrogenation occurs without the formation of molecular hydrogen. Helium might however, sweep gaseous byproducts such as CO or CO2, potential inhibitors, from the catalyst surface. In the case of limited glycerol (e.g. P/G =1.96) this flushing effect is reduced as glycerol, the source of CO2 and CH4, is restricted.  Figure 5.7 Comparison of phenol conversion with products selectivity (a) and glycerol conversion and products selectivity (b) during IGAPH in closed and open systems with different phenol to glycerol ratio. All experiments were conducted with phenol (1.0 g), water (30.0 g), and Raney Ni® (1.0 g) at 220 °C for 2 h. Initial He pressure for closed systems was 0.69 MPa while 4.34 MPa constant pressure for open system. Unless otherwise stated, He flow for open system was 0.03-0.06 L/min.   Additional evidence that molecular hydrogen gas was not produced during the process is the lack of a decrease in pressure during closed system operation (Fig 5.8 a-b). This can be confirmed by 92  comparison to the results of phenol hydrogenation with and without Raney Ni® with formic acid as the hydrogen source. The reaction with Raney Ni® shows a decline in reaction pressure (Fig 5.8 c) indicating readsorption and consumption of H2 by phenol (Table 5.3). On the other hand, the pressure in the reaction without Raney Ni® continuously increased, although more slowly than the reaction with Raney Ni®. The slowly increasing pressure in the reaction without Raney Ni® (Fig 5.8 d) indicates some degradation of formic acid to H2 in the absence of catalyst, as confirmed by others.189 The absence of a pressure drop indicates that hydrogen cannot directly react with phenol. In contrast to the reaction with formic acid, the pressure was never observed to decrease during IGAPH with glycerol (Fig 5.8a-b) therefore it seems that hydrogen production only occurs on the catalyst surface and this hydrogen is consumed by phenol and other reactants without the formation of molecular hydrogen. The proposed non-molecular hydrogen mechanism further reinforces the claim that the presence of phenol accelerates removal of adsorbed hydrogen thus freeing active sites. Table 5.3 Phenol conversion and product selectivity from closed system hydrogenation with formic acid.  No Catalyst  Phenol    XFA (mol %) XP  (mol %) Selectivity (mol%)   Benzene Cyclohexanol Cyclohexanone   1 Raney Ni®  28  6 57 37  100 2 N/A  0  - - -  25 Experiments were conducted with formic acid (2.0 g), water (3.0 g), phenol (1.0 g), Raney Ni® (1.0 g) at 220°C for 2 h with 3.45 MPa He initial pressure.   93   Figure 5.8 Pressure (solid line) and temperature profile (dash line) of closed system IGAPH with an initial pressure of 0.69 MPa (a) and 3.45 MPa (b). Phenol hydrogenation using formic acid with an initial pressure of 3.45 MPa with catalyst (c) and without catalyst (d).  Since phenol hydrogenation occurs without the formation of molecular hydrogen gas, the gaseous products can be released. During large scale operation, gaseous products may lead to prohibitively high pressures. By releasing gaseous by-products, the requirement for high pressure conditions can be eliminated. To understand the effect of pressure on the degree of phenol hydrogenation, experiments in open and closed system configurations at different pressures were conducted. 94  Closed system reactions were conducted at 0.69 MPa and 3.45 MPa He. The open system reactions were conducted at 2.41, 4.34 and 10.34 MPa He, representing the saturation pressure of water at 220°C, maximum pressure achieved in the closed system with an initial pressure of 0.69 MPa, and an elevated pressure at which all intermediate products of glycerol (e.g methanol, ethanol) will be condensed, respectively.   Figure 5.9 Comparison of phenol conversion and products selectivity (a) and glycerol conversion and products selectivity (b) during IGAPH in closed and open systems with different initial and reaction pressures. All experiments were conducted with phenol (1.0 g), glycerol (1.0 g) water (30.0 g), and Raney Ni® (1.0 g) at 220 °C for 2 h. He pressures for closed systems indicates the initial pressure while for open system indicate reaction pressure. He flows for open system was 0.03-0.06 L/min.  As shown in Fig 5.9a in the closed system reaction, the higher initial pressure resulted in lower phenol conversion and degree of hydrogenation. In the open system reaction, on the other hand, the conversion and degree of phenol hydrogenation increased with reaction pressure. Phenol conversion increased from 84% to 98% when the reaction pressure was increased from 2.41 MPa 95  to 4.34 MPa and remained as 98% at 10.34 MPa. In addition, selectivity to cyclohexanol increased with reaction pressure from 39% at 2.41 MPa, to 62% at 4.34 MPa and further to 74% at 10.34 MPa.  At lower reaction pressure, more glycerol is converted into gaseous products in both open/closed systems. At an initial pressure of 0.69 MPa in the closed system, 1,2-propylene glycol and ethanol were the primary liquid APR products. When initial pressure was raised to 3.45 MPa, glycerol conversion decreased by 4% with more liquid APR products (125% higher). Similar trends were also found in the open system where higher reaction pressure tended to result in more liquid intermediates.  In open system reaction, phenol conversion behaves differently than glycerol. In the reaction with 2.41 MPa He, phenol conversion was low with relatively low cyclohexanol selectivity which indicates hydrogen deficiency. On the other hand, glycerol conversion was 100% with low selectivity of liquid products. This suggests that at 2.41 MPa low-boiling glycerol liquid intermediates such as methanol and ethanol were vaporized. The saturation pressure of methanol and ethanol at 220 °C are 5.40 MPa and 4.01 MPa (Antoine equation), respectively. A more efficient reflux condenser system should be installed or the reaction pressure should be maintained above the vapor pressure of the most volatile intermediate, which is likely methanol. At 10.34 MPa both ethanol and methanol will remain in the liquid phase. Although phenol conversion was similar at 4.34 MPa and 10.34 MPa, cyclohexanol selectivity was slightly higher at 10.34 MPa, indicating greater hydrogenation.  The evaporation of glycerol APR liquid intermediate products is more difficult to observe in the closed system reaction since the reactor head space was small. Pressurization of the reactor by gaseous byproducts would increase the liquid/vapor ratio of low 96  boiling point compounds and a higher liquid ratio of these compounds would increase the likelihood of these compounds undergoing APR. Higher phenol conversion even with high reaction pressure in the open system could be due to the flushing of catalyst inhibitors such as CO or CO2 by He. The removal of CO2 might also have increased the selectivity of glycerol APR. It was previously found that the removal of CO2 by introducing CaO shifts the equilibrium reaction forward, suppressing CH4 formation.190 Another possibility is that CO2 occupied the same  Raney Ni® active sites as phenol and glycerol thus lowering the reaction rate of glycerol. It was previously found that CO2 dissociates when adsorbed on Ni,191 which is an indication of occupation of active site by CO2. Increased pressure is likely to increase the amount of CO2 dissolved in solution since the Henry’s Law constant of CO2 is almost a hundred times higher than that of He.  5.2.7 Examination of methanation using CO2/CH4 ratio   Experiments with and without phenol, as well as in open and closed systems, show that the presence of phenol increases glycerol conversion and selectivity toward APR reaction while removal of gaseous products increases the selectivity of the APR reaction. The effects of each factor after a reaction time of 120 min are summarized in Table 5.4. The ratio of CO2/CH4 in the open system reaction was calculated by summing the gaseous products from all sample intervals from 0-120 min (Table E.3). The sweeping of gaseous products did not significantly increase glycerol conversion (5%) but significantly increased the CO2/CH4 ratio (49%). The addition of phenol to the closed system increased glycerol conversion and CO2/CH4 ratio by 31% and 107%, 97  respectively. Comparison between IGAPH in the closed and open system shows that the removal of gaseous products increased glycerol conversion by 3% while CO2/CH4 ratio increased by 20%.  Table 5.4.  Reaction metrics of APR and IGAPH in open and closed system   Closeda Openb APR Glycerol conversion (mol%) : 73 Glycerol conversion (mol%) : 76 CO2/CH4 (mol/mol) : 1.10 CO2/CH4 (mol/mol) : 1.65 IGAPHc Glycerol conversion (mol%) : 95 Glycerol conversion : 98 CO2/CH4 (mol/mol) : 2.28 CO2/CH4 : 2.74 a Experiments were conducted in closed system with 2.0 g glycerol, 30.0 g of water, and 1.0 g Raney Ni® at 220 °C for 2 h with initial He pressure of 100 psig. b Experiments were conducted in continuous 0.03-0.06 L/min He flow at 630 psig. c Experiments were conducted with addition of 2.0 g phenol.   The CO2/CH4 ratio during glycerol APR indicates that 48% of potential CO2 that can be produced is converted into CH4 while in open system this decreases to 38% (calculation in Appendix A.5). This difference indicates that removing gaseous products during the reaction reduces the possibility of the Sabatier reaction. Assuming that the Sabatier reaction did not occur during open system reaction, it can be inferred that 38% of carbon undergoes direct methanation during glycerol APR. This suggests that the remaining 10% of reacting carbon undergoes the Sabatier reaction in the closed system. During IGAPH the potential CO2 converted to CH4 decreased to 30% in the closed system and 27% in open system. By the same logic, it can be inferred that 27% of carbon undergoes direct methanation during glycerol APR and the remaining 3% of reacting carbon undergoes the Sabatier reaction in the closed system. This suggests that direct methanation is suppressed by the presence of phenol. Finally, the decrease in the extent of the Sabatier reaction 98  between glycerol APR and IGAPH can also be attributed to the hydrogen scavenging effect of phenol.  5.2.8  Effect of catalyst loading To understand the effect of catalyst loading, reactions with 54.3 mmol glycerol (5.0 g), 53.1 mmol phenol (5.0 g), and water (30.0 g) were conducted at 220 °C for 2 h with 3 different catalyst loadings (0.5, 1.0, and 5.0 g). Fig. 5.9 shows increasing catalyst loading resulted in higher glycerol conversion. It can be seen that glycerol conversion increased from 47 mol% with 0.5 g Raney Ni® to 67 mol% with 1.0 g loading, and 91 mol% with 5.0 g loading. The selectivity of 1,2-propylene glycol and ethylene glycol, decreased from 25% and 8% to 21% and 7%, respectively. Hydrogenolysis products selectivity decreased to 2%, 1%, and 3% for 1,2-propylene glycol, ethylene glycol, and ethanol, respectively, when the catalyst loading was increased from 1.0 g to 5.0 g. This indicates that the hydrogenolysis products react more quickly, possibly to gaseous products, with higher catalyst loading.  99   Figure 5.10* Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 220 °C for 2 h as a function of Raney Ni® catalyst loading.  Phenol, glycerol, and water loadings were 5.0 g (53.1 mmol), 5.0 g (54.3 mmol), and 30.0 g respectively.  Panel (a) presents glycerol conversion (outer left axis), the selectivity of glycerol hydrogenolysis products (bars, inner left axis), glycerol carbon selectivity (inner right axis), and residual pressure after cooling to 50 °C (outer right axis).  Panel (b) presents phenol conversion (outer left axis), the selectivity of phenol hydrogenation products (bars, inner left axis), hydrogen selectivity (inner right axis), and the ratio of CO2(g) to CH4(g) (outer right axis). *Horizontal axis is not to scale.   Phenol conversion also increased with increasing catalyst loading: 22 mol%, 46 mol%, and 87 mol% with Raney Ni® loadings of 0.5, 1.0, and 5.0 g, respectively. The ratio of cyclohexanol to 100  cyclohexanone increased with higher catalyst loading indicating that the rate of hydrogenation is proportional to catalyst loading. The ratio of CO2-to-CH4 tended to decrease with higher catalyst loading. This trend in combination with the increasing residual pressure and changes in glycerol carbon selectivity shows that more glycerol was converted to gaseous products at higher catalyst loading. The amount of catalyst loading is not linearly proportional to glycerol or phenol conversion as the use of 5.0 g catalyst did not result in total conversion of glycerol and phenol.   5.2.9 Reaction with several lignin model compounds  Several lignin monomers (eugenol, vanillin, p-creosol, guaiacol, p-cresol) and dimers (dibenzyl ether, diphenyl ether, benzyl phenyl ether), were hydrogenated by in-situ glycerol APR. The results are presented in Table 5.5. In this reaction, 10.6 mmol of a model compound was reacted with 10.9 mmol glycerol together with 1.0 g Raney Ni® and 30.0 g water at 220°C for 2h. Guaiacol hydrogenation generated the highest selectivity of saturated compounds with cyclohexanol and cyclohexanone present as the chief products. In addition, 2-methoxycyclohexanol and phenol were found among the products. Cyclohexanol can form from the hydrogenation of phenol or the demethoxylation of 2-methoxycyclohexanol while cyclohexanone seems to occur only through hydrogenation of phenol.192     101  Table 5.5 Hydrogenation of lignin model compounds with in situ glycerol APR  Experiments were conducted with model compound (10.6 mmol), glycerol (10.9 mmol), water (3.0 g), Raney Ni® (1.0 g) at 220°C for 2 h with 0.69 MPa He initial pressure. a quantified using cyclohexanone as standard; b quantified using 4-propyl guaiacol as standard; c quantified using 4-propyl phenol as standard. ReactantX (mol C %)C-balance8522 68 4 680a7 26 6864a5 14 53 281005 23 4 21 30 5 1199b c c1 4 13 47 33 11003 10 83 44731 14 55 1973 48 18 26 50.990.871.03Product selectivity (%)0.930.980.991.000.77HOOCH3HOOCH3HOHOOCH3OO OHOHHOOCH3HOOCH3HO HOHOOCH3O OHHOHOOCH3 O OHHOHOO OHHOOCH3OHOCH3OOO OHHOOO OHHOH3CO102  p-Cresol had the second highest selectivity of saturated compounds; 4-methylcyclohexanol, 4-methylcyclohexanone, and toluene were the main products. This selectivity distribution is similar to the product distribution from phenol with the methyl group attached in the para position. Of the monomers, p-creosol was the least reactive (X=64 mol C%). The product distribution of the reaction with p-creosol is similar to the product distribution from p-cresol. Therefore, the reaction of p-creosol may occur through demethoxylation to form p-cresol since p-cresol was also a product.  Complete conversion of vanillin was achieved; this was the highest of all tested monomers.  However, vanillin’s aromatic ring is stable as 67% of detected products are phenolics. p-Creosol was the product with the highest selectivity indicating high reactivity of the aldehyde group in vanillin. The high aldehyde reactivity was also observed during the hydrogenation of vanillin at 65oC with Ru; vanillyl alcohol was the predominant product.84 The reaction of vanillin may occur through aldehyde hydrogenation to form vanillyl alcohol. Vanillyl alcohol could subsequently undergo dehydration forming water and p-creosol or losing methanol to form guaiacol. The second route seems dominant as the selectivity to guaiacol, phenol, cyclohexanol, and cyclohexanone are higher than p-creosol, p-cresol, and 4-methylcyclohexanol. Near complete conversion of eugenol was achieved but most of it formed cerulignol (47%) indicating hydrogenation of allyl group. The aromatic ring in eugenol seems to be more stable than the hydroxyl group as deoxygenated compounds, such as 4-propylbenzene and 4-propylcyclohexene, were detected in the products. In this reaction, most products (99%) have the propyl group attached. Less than one percent of the products had an ethyl group, which was probably formed through demethylation of the propyl group. The formation of propylbenzene and 103  propylcyclohexene suggests a similarity to the formation of toluene from p-cresol. The attachment of an alkyl group in the para position may weaken the hydroxyl group in the phenolic compound. Reactions with dimeric lignin model compounds were also conducted. Complete conversion of dibenzyl ether was obtained and the dominant product was toluene (83%). Benzene, methylcyclohexane and cyclohexane were also found with selectivities of 10%, 4%, and 3% respectively. The absence of saturated ether dimers and the high selectivity of toluene indicate that the ether bond is the most reactive functional group in dibenzyl ether. In addition, the absence of oxygenated products indicated that the ether bond might be dehydrated. Although limited, demethylation and hydrogenation of the ring also occurred, which was indicated by the presence of benzene, cyclohexane, and methylcyclohexane. Only 47% of diphenyl ether was converted during the reaction, mostly to cyclohexanol (55%). Benzene, cyclohexanone, and phenol were also formed with selectivities of 31%, 14%, and 1% respectively. The low conversion of diphenyl ether might due to the stearic hindrance as the dimer consists of two twisted aromatic rings. The high selectivity of cyclohexanol and the low selectivity of phenol suggests two possible reaction pathways. One pathway is the hydrogenation of an aromatic ring followed by very fast O-C bond breaking to produce equimolar amounts of benzene and cyclohexanol. An alternate path is O-C bond breaking to produce benzene and phenol, followed by hydrogenation of phenol to cyclohexanol.  Benzyl phenyl ether was almost fully converted (97%) and produced toluene (48%), cyclohexanol (26%), cyclohexanone (18%), phenol (26%), and benzene (3%). The product distribution indicates cleavage of the ether bond between aliphatic carbon and oxygen was the primary reaction 104  producing toluene and phenol in equal amounts. Phenol hydrogenation likely resulted in formation of saturated compounds such as cyclohexanol and cyclohexanone.  The dimer reactions show that the ether bond between aliphatic carbon and oxygen is more reactive than the bond between carbons in the aromatic rings, as illustrated by the reaction of dibenzyl ether. The bond between aliphatic carbon and oxygen seems more reactive than the bond between aromatic carbon and oxygen, as illustrated by the reaction of benzyl phenyl ether. The bonds between carbons within aromatic rings seem to be the least reactive as indicated by the absence of saturated dimers in the products.   5.2.10 Reaction with noble metal catalysts Pt/C, Ru/C, and Pd/C (5 wt% metal loading) were used to conduct IGAPH. These reactions were conducted using 1.0 g of glycerol, 1.0 g of phenol, 30.0 g of water and 1.0 g of carbon supported metal catalyst at 220°C for 2 h with an initial He pressure of 3.45 MPa. The result of the reactions is presented in Fig 5.10. Pt/C and Ru/C show potential as an IGAPH catalyst but Pd/C does not. The highest glycerol conversion, 60 mol%, was obtained using Ru/C. During this reaction, 1,2-propylene glycol was observed as the primary liquid product with 26% selectivity followed by ethanol, methanol, and ethylene glycol with selectivities of 7%, 6%, and 4%, respectively. The phenol conversion in the reaction with Ru/C was 47 mol%; benzene was the major product (37%) followed by cyclohexanone (35%) and cyclohexanol (28%). The CO2-to-CH4 ratio was very low, 0.8 mol/mol, which reflects the high selectivity of low hydrogen-content compounds. The reaction with Pt/C results in 50 mol% conversion of glycerol with 1,2-propylene glycol present as the major product (28%) followed by ethanol (16%), methanol (9%), and ethylene glycol (5%). 105  A high ratio of CO2-to-CH4 was found in this reaction (3.72 mol/mol). Phenol conversion was 47%, the highest conversion of the tested noble metals, and the primary products were cyclohexanol (83%) and cyclohexanone (13%). Cyclohexane selectivity was very low, only 4%.   Figure 5.11* IGAPH reaction with different catalysts at 220 °C for 2 h with 3.45 MPa initial He pressure. Phenol, glycerol, catalyst and water loadings were 1.0 g, 1.0 g, 1.0 g, and 30.0 g respectively. Panel (a) present glycerol conversion (right axis), liquid products selectivity from glycerol (left axis), and the ratio of CO2(g) to CH4(g) (outer left axis). Panel (b) present phenol conversion (right axis) and products selectivity from phenol (left axis). *Raney Ni® contains approximately 80 wt% Ni loading and carbon supported noble metal contains 5 wt% of active metal.   The selectivities of liquid products from glycerol were similar for Pt/C and Ru/C but are slightly different from Raney Ni®.  Ethanol was the primary glycerol product when Raney Ni® was used. 106  Glycerol conversion in the presence of noble metals was lower than when Raney Ni® was used. However, the CO2/CH4 ratio after reaction with Pt/C is significantly higher than when Raney Ni® was used indicating high selectivity of glycerol APR. Phenol conversion in the presence of noble metals was also lower than when Raney Ni® was used. The phenol product distribution, mostly saturated compounds, from the reaction with Pt/C was similar to the reaction with Raney Ni®. Considering the differences in metal content of Raney Ni® and Pt/C, the phenol conversion in the reaction with Pt/C is not unreasonably low. Unlike Raney Ni® which loses its catalytic activity, Pt/C is known to remain active under acidic conditions.88,93 Given that FPO contains organic acids, further investigation of Pt/C is merited.    5.2.11 Phenol hydrogenation mechanism and effect of He flow during IGAPH with Pt/C The reaction of Pt/C was next conducted in the open system. In these reactions, 1.0 g of glycerol, 1.0 g of phenol, 30.0 g of water and 1.0 g of Pt/C were heated in the reactor at 220°C for 2 h. The open system configuration was similar to the configuration used in section 5.2.6. The reaction was conducted at 4.34 MPa He and He flowrate was varied.  Both glycerol and phenol conversion increased when the reaction was conducted in the open system. As shown in Fig. 5.12, glycerol conversion increased from 50 mol% in the closed system to 63 mol% in the open system without He flow. The He flow was found to increase glycerol conversion to 72 mol% at 0.03-0.06 L/min and then to 94 mol% at 0.15-0.25 L/min. The selectivity of liquid intermediates decreased with higher He flow. The selectivity of liquid intermediate products decreased considerably in the reaction with He flow.  107   Figure 5.12 IGAPH reaction with 5 wt% Pt/C in closed and open reactor configurations at 220°C for 2h. Closed system reaction was conducted with 3.45 MPa initial He pressure. Open system reactions were conducted with constant 4.34 MPa He pressure. The He flow rate was varied from 0 L/min to 0.15-0.25 L/min.  Phenol, glycerol, Pt/C and water loadings were 1.0 g, 1.0 g, 1.0 g, and 30.0 g respectively. Panel (a) present glycerol conversion (right axis) and liquid products selectivity from glycerol (left axis). Panel (b) present phenol conversion (right axis) and products selectivity from phenol (left axis)   Phenol conversion increased from 47 mol% in the closed system to 65 mol% in the open system without flow and further increased to 100 mol% with 0.03-0.06 L/min He flow. Selectivity of products was altered by flowrate. In most cases, cyclohexanol was the major product, however at 0.03-0.06 L/min cyclohexane became the major product with selectivity of 38%. This flowrate also produced the highest selectivities of cyclohexanone and benzene of all tested conditions. The higher phenol conversion in the open system configuration shows that phenol hydrogenation 108  during IGAPH with Pt/C occurs without the formation of molecular hydrogen gas, as was observed with Raney Ni®. In addition, the He flow significantly increases phenol conversion demonstrating that removal of CO2 more strongly affects the performance of Pt/C than of Raney Ni®.    5.3 Kinetics of IGAPH In order to understand the complex IGAPH reaction, which involves three reactants and Raney Ni®, an initial rate model was developed.  Water participates as a reactant in the system as well as a process fluid used to reduce system viscosity, therefore in this study, an excess of water was used for all experiments. Consequently, glycerol and phenol are the only reactants considered when identifying the rate-limiting step of the reaction mechanism. The results presented in Fig 5.3 and 5.4 show that both phenol and glycerol are adsorbed to the catalyst.  Therefore, the Langmuir-Hinshelwood mechanism was selected to interpret the IGAPH reaction mechanism.       A possible IGAPH reaction mechanism was adopted based on the mechanism of ethylene glycol APR on Pt and Ni188,193 and hydrogenation of naphthalene on Ni and Ru.194 The data was taken from Table C.1. The reaction mechanism is outlined by equations 5.1 to 5.9 and illustrated in Fig 5.13. For the sake of clarity, glycerol APR is shown as occurring before phenol hydrogenation in Fig 5.13, however, in reality, adsorption occurs competitively between phenol, glycerol, and water. The APR reaction begins with molecular adsorption of glycerol and dissociative adsorption of water (i) followed by dehydrogenation (ii-iii) and C-C cleavage (iv-v) of glycerol to produce CO* and H*.  CO* will then react with OH* to produce CO2 and additional H*.  Meanwhile, C2O2Hx* will undergo similar dehydrogenation and C-C cleavage (v-vii). Adsorbed phenol will then 109  irreversibly react with H* pairs to produce cyclohexanol or cyclohexanone (vii-xi).  The cycle will then restart through competitive adsorption of phenol, glycerol, and water.   C3H8O3 + *      𝐾𝑔   ↔     C3H8O3*  - adsorption of glycerol (5.1) C3H8O3* +(8-x)* 𝑘𝑑ℎ→  C3HxO3* + (8-x)H*   - irreversible dehydrogenation of glycerol (5.2) C3HxO3* + * 𝑘𝑐−𝑐→   C2O2Hx* + CO*  - irreversible reaction leading to CO* (5.3) H2O + 2*     𝐾𝑤  ↔    OH* + H* - dissociative adsorption of water (5.4) CO* + OH* 𝑘𝑤𝑔𝑠→   CO2* + H* - irreversible water-gas shift reaction (5.5) CO2*    1/𝐾𝑑  ↔      CO2 + * - CO2 desorption (5.6) C6H6O + *       𝐾𝑝  ↔    C6H6O*  - phenol adsorption (5.7) C6H6O* + 2H*    𝑘ℎ𝑦→    C6H8O* + 2* - irreversible phenol hydrogenation leading to   cyclohexanone or cyclohexanol (5.8) C6H10O* or C6H12O*     1/𝐾𝑐 ↔    C6H10O or C6H12O + *   - desorption of cyclohexanone or cyclohexanol (5.9) Five rate equations models were developed by assuming a different rate-limiting step: glycerol adsorption (Eqn. 5.1), phenol adsorption (Eqn. 5.7), water adsorption (Eqn. 5.4), glycerol dehydrogenation (Eqn.5.2), and phenol hydrogenation (Eqn. 5.8). Initial rate models were applied to each rate equation assuming negligible product concentration (𝐶𝐶6𝐻12𝑂 and 𝐶𝐶𝑂2) and constant reactant concentration for the initial concentration. This assumption was taken with consideration 110  that the concentration of products was much less than that of the reactants and therefore had negligible effect during the reaction.   The rate of the reaction, r, was calculated for phenol conversion in a batch reactor: 𝑟 = −[𝑃0]𝑑𝑋𝑃𝑑𝑡×1𝑚𝑐𝑎𝑡  (5.10)  where [P0], Xp, t, and mcat denote initial phenol concentration, phenol conversion, reaction time, and weight of catalyst respectively. Equilibrium and rate constants for each elementary reaction were determined by iterating with the Levenberg-Marquardt algorithm in Origin Lab®. The results for the five aforementioned models including parameter values, adjusted R2 and residual sum of square (SSR) are summarized in Table B.3.  Assuming phenol hydrogenation as the rate determining step resulted in the highest adjusted R2, 0.989. The corresponding rate equation is shown in equation 5.11, where 𝑘’ =  𝐾𝑝𝑘ℎ𝑦.  The initial rate expression simplifies to equation 5.12. 𝑟 =  𝑘′𝐶𝐶6𝐻6𝑂(𝐾𝑔𝐶𝐶3𝐻8𝑂3+(𝐾𝑤𝐶𝐻2𝑂)12)2 [1+2𝐾𝑔𝐶𝐶3𝐻8𝑂3+2(𝐾𝑤𝐶𝐻2𝑂)12+𝐾𝑝𝐶𝐶6𝐻6𝑂+𝐾𝑐𝐶𝐶6𝐻12𝑂+𝐾𝑑𝐶𝐶𝑂2  ]3  (5.11) 𝑟𝑖 = 𝑘′𝐶𝐶6𝐻6𝑂(𝐾𝑔𝐶𝐶3𝐻8𝑂3+(𝐾𝑤𝐶𝐻2𝑂)12)2 [1+2𝐾𝑔𝐶𝐶3𝐻8𝑂3+2(𝐾𝑤𝐶𝐻2𝑂)12+𝐾𝑝𝐶𝐶6𝐻6𝑂]3  (5.12)   111   Figure 5.13 Schematic of in-situ glycerol APR and phenol hydrogenation reaction mechanism.  5.4 Conclusion This study was conducted to evaluate the possibility to upgrade lignin or FPO with hydrogen formed from in-situ glycerol APR and understand the synergistic effects of IGAPH. In the batch reactor, phenol hydrogenation on Raney Ni® using hydrogen produced in-situ by glycerol APR results in the production of cyclohexanone, cyclohexanol, benzene, and o-cresol. These are promising results for future lignin or FPO upgrading.  In-situ reactions involving phenol, glycerol, and glycerol liquid intermediates showed that the presence of phenol enhanced the conversion and the APR selectivity of glycerol and glycerol intermediate products. It was found that the glycerol reaction order decreased and the apparent rate constant increased with the presence of phenol. These results indicate a synergistic effect on the in-situ reaction. Furthermore, by employing an open system and sweep gas, it was determined that 112  phenol hydrogenation likely proceeds without the formation of molecular hydrogen. This result suggests that phenol accelerates the removal of hydrogen from the catalyst surface. Initial rate kinetic analysis suggests that IGAPH follows a Langmuir-Hinshelwood mechanism with phenol hydrogenation as the rate-limiting step.  The source of methane during the reaction was explored. Direct methanation of glycerol and intermediates and the Sabatier reaction of CO2 are two possible CH4 formation pathways. Comparing the CO2/CH4 ratio from glycerol APR and IGAPH in the open and closed system, it was found that the presence of phenol significantly reduces methane formation by scavenging hydrogen. The open system reaction removes CO2 and thus reduces the extent of the Sabatier reaction although direct methanation still occurs. The open system reaction is also advantageous from a practical perspective as it enables a decrease in reactor pressure. However, the reaction pressure of the open system must be sufficient to ensure that the intermediate products of glycerol APR remain in the liquid phase. In addition, the reaction was successfully performed using other model lignin monomers and dimers indicating the reaction has wide applicability. Finally, the high selectivity of APR with Pt/C makes it worthy of further study.   113  Chapter 6: In-situ Glycerol Aqueous Phase Reforming and Phenol Hydrodeoxygenation (IGAPHdo)    6.1 Introduction Hydrodeoxygenation of lignin-derived compounds will result in naphthenes and aromatics which can be used as fuel additives as well as chemical precursors.  As a continuation of the previous chapter, hydrodeoxygenation of phenol is performed in this Chapter. Hydrodeoxygenation is conducted via sequential reactions, hydrogenation and dehydration. Two sets of reactions are conducted in this chapter: the first uses Raney Ni® and H-ZSM-5 while the second set is conducted with Pt/C and Amberlyst-15. Raney Ni® and Pt/C were demonstrated to be active catalysts for the IGAPH reactions in Chapter 5 and will be used as hydrogenation catalysts. In this chapter, the effect of H-ZSM-5 loading on the production of non-oxygenated compounds was observed. The mechanism of benzene formation (whether through DDO or sequential HDO and dehydrogenation) was observed by conducting the HDO reaction with intermediates from phenol. The Pt/C and Amberlyst-15 study was conducted in order to determine if the distribution of phenol HDO products could be controlled by changing the ratio of metal and acid catalyst. Pt was selected for this purpose because it was previously found to accelerate phenol hydrogenation relative to glycerol APR during IGAPH.   Chapter 5 showed that the IGAPH reaction can be conducted with Raney Ni® and Pt/C catalyst to produce cyclohexanol and cyclohexanone. HDO products such as benzene in the reaction with 114  Raney Ni® and cyclohexane in the reaction with Pt/C, were also produced, although in minor amounts. It has been previously suggested that HDO reactions can occur in the presence of acid catalysts.88,130 Water at elevated temperature has a high dissociation constant, creating a low pH environment.195 This dissociation constant is also affected by pressure.46 IGAPH conditions make it possible that the acidity of water at high temperature promotes the deoxygenation reaction. Fig 5.2 showed that the selectivity of benzene increased with temperature. Limited hydrogen (from glycerol) relative to phenol seems to increase the probability of HDO reactions as could be seen from Fig. 5.3 and 5.4.  There is, however, a thermodynamic constraint to increasing temperature of HDO reactions. The equilibrium constant of phenol hydrogenation decreases at higher temperature as the reaction is exothermic, while the equilibrium constant of glycerol APR increases at higher temperature as the reaction is endothermic. The combined effects will likely result in low hydrogenation performance. The addition of a solid acid catalyst while conducting the reaction at thermodynamically favorable temperature (Fig. 5.1) is likely more feasible.   6.2 Results 6.2.1 Phenol HDO with Raney Ni® and several acid catalysts 6.2.1.1 Acid catalyst screening  Phenol HDO was conducted at 240°C with several acid catalysts: Amberlyst-15, H-BEA, and H-ZSM-5. The results are shown in Fig. 6.1. Glycerol was almost fully converted except when Amberlyst-15 was used. In all reactions, ethanol had the highest selectivity followed by 1,2-propylene glycol. However, the low selectivity of liquid products derived from glycerol indicates 115  that the majority of glycerol was converted to gaseous products. At 240°C, the ratio of CO2/CH4 was generally low especially in the reaction with zeolites (0.75 and 0.81 mol/mol for H-BEA and H-ZSM-5 respectively). Relatively higher CO2/CH4 (1.28) was found in the reaction with Amberlyst-15.  Phenol conversion varied with the acid catalyst. In the presence of Amberlyst-15 and H-BEA, conversions were less than in the absence of acid catalyst (X = 89 mol%). In the presence of H-ZSM-5, phenol conversion was slightly higher (X = 97 mol%). Higher benzene and cyclohexane selectivity were achieved when an acid catalyst was added. The highest selectivity to non-oxygenated products was obtained using H-ZSM-5. The selectivity of benzene, cyclohexane, and toluene were 66%, 27%, and 4% respectively. The respective selectivities of cyclohexanol and cyclohexanone were 1% and 2%. Despite having the lowest glycerol conversion, phenol conversion in the presence of Amberlyst-15 was 88% and the selectivity of non-oxygenated products was relatively high. In this system, the selectivity of benzene and cyclohexane were both 34% while the selectivity of oxygenated compounds such as cyclohexanol and cyclohexanone, were both 15%. The reaction with H-BEA resulted in higher selectivity of oxygenated compounds compared to non-oxygenated compounds.  116   Figure 6.1 IGAPHdo reaction with several acid catalysts at 240 °C for 2 h with 3.45 MPa initial He pressure. Phenol, glycerol, Raney Ni®, acid catalyst and water loading were 1.0 g, 1.0 g, 1.0 g, 0.1 g, and 30.0 g respectively. Panel (a) present glycerol conversion (equation 3.1, right axis) and selectivity of glycerol hydrogenolysis products (equation 3.4, bars, inner right axis). Panel (b) phenol conversion (equation 3.1, inner right axis), liquid products selectivity from phenol (equation 3.2, bars, inner left axis), O/C ratio in phenol products (outer left axis), and H/C ratio in phenol products (outer right axis).  117  The low O/C ratio, a measure of degree of deoxygenation, shows that H-ZSM-5 and Amberlyst-15 provided the highest degree of deoxygenation. The H/C ratio, a measure of hydrogenation, conversely shows that these two catalysts also produced the lowest degree of hydrogenation. It was previously found in IGAPH (Chapter 5) that selectivity of cyclohexanone, which has lower degree of hydrogenation than cyclohexanol, increased when hydrogen availability was low.  It can be seen that the liquid products selectivity of glycerol in the reaction at 240°C was dominated by ethanol rather than 1,2-propylene glycol or ethylene glycol. In addition, lower CO2/CH4 ratio (0.93) compared to the reaction at 220°C (1.50) was found and this ratio decreased even more in the presence of zeolites. The high selectivity of deoxygenated products indicated that Amberlyst-15 and H-ZSM-5 catalyst were able to promote HDO of phenol. Although the use of H-BEA resulted in low selectivity of deoxygenated products from phenol, glycerol conversion was relatively high with low amount of liquid intermediates. The CO2/CH4 was very low compared to the other IGAPHdo tests. Low phenol HDO, low CO2/CH4, and high glycerol conversion indicate that H-BEA is ineffective under aqueous condition. It seems that it is primarily glycerol which is dehydrated resulting in alkane production.   6.2.1.2 IGAPHdo with Raney Ni® and H-ZSM-5 Since the greatest degree of deoxygenation was obtained using H-ZSM-5, the effect of H-ZSM-5 loading was assessed and results are shown in Fig. 6.2. Glycerol conversion and product distribution were similar for all loadings. Ethanol exhibited the highest selectivity followed by 1,2-propylene glycol. Phenol conversion increased with increasing H-ZSM-5 loading. In the absence of H-ZSM-5, cyclohexanol and cyclohexanone were the dominant products with selectivities of 48 % and 45% respectively. Benzene selectivity was only 7%. The selectivity of benzene increased 118  to 28 % and cyclohexane appeared with selectivity of 4% when 0.05 g of H-ZSM-5 was added. This reaction indicates that the phenol HDO reaction increased with increasing acid catalyst loading.  Figure 6.2 IGAPHdo reaction H-ZSM-5 at 240 °C for 2 h with 3.45 MPa initial He pressure. Phenol, glycerol, Raney Ni®, and water loading were 1.0 g, 1.0 g, 1.0 g, and 30.0 g respectively. Panel (a) present glycerol conversion (equation 3.1, right axis) and selectivity of glycerol hydrogenolysis products (equation 3.4, bars, inner right axis). Panel (b) phenol conversion (equation 3.1, inner right axis), liquid products selectivity from phenol (equation 3.2, bars, inner left axis), O/C ratio in phenol products (outer left axis), and H/C ratio in phenol products (outer right axis). 119  The O/C ratio decreased with increasing H-ZSM-5 loading, indicating a greater degree of deoxygenation of phenol. However, it appeared that hydrogenation occurs to a reduced degree with increasing H-ZSM-5 loading. In addition, ethanol selectivity, a glycerol dehydration product, had increasing selectivity with increasing H-ZSM-5 loading. The dehydration of glycerol may result in less hydrogen being produced.   6.2.1.3 Reaction with the intermediates Intermediate products from previous tests were used as IGAPHdo reactants. Conversion and products selectivity are shown in Table 6.1. Reaction of cyclohexanol without glycerol produced cyclohexane and benzene, with selectivities of 61% and 40%, respectively. Cyclohexanol conversion was 86 mol%. Dehydration of cyclohexanol was then conducted using only H-ZSM-5; this produced cyclohexene (S = 56%), cyclohexane (S = 28%) and benzene (S = 16%). Cyclohexanol conversion was 92 mol% indicating that dehydration of cyclohexanol was unaffected by Raney Ni®.   Since it was found as an intermediate, cyclohexene was used as the starting reactant. Reaction of cyclohexene without any catalyst resulted in 60 mol% conversion and generated cyclohexane (S = 70%) and benzene (S = 30%). Similar selectivity was observed when Raney Ni® was added to the system but cyclohexene conversion increased to 100 mol%. Product selectivity changed slightly when glycerol was added to the reaction with cyclohexene (entry 5). In this reaction, cyclohexane and benzene selectivities were 82% and 18%, respectively, and cyclohexene conversion was 79 mol%.  120  Table 6.1 Reaction of IGAPHdo phenol products. The reaction was conducted with 1.0 g of reactant and 30.0 g of water at 240 °C for 2 h under 3.45 MPa initial He pressure. No Reactant  Catalyst (g) X (mol%) Products selectivity (%) Raney Ni®  H-ZSM-5 C-xane1 C-xene2 Benzene C-nol3  C-one4 1 Cyclohexanol 1 0.05 89 60 - 40 - - 2 Cyclohexanol N/A 0.05 92 28 58 14 - - 3 Cyclohexene  N/A N/A 60 70 - 30 - - 4 Cyclohexene  1 N/A 100 58 - 42 - - 5 Cyclohexene5 1 N/A 79 82 - 18 - - 6 Phenol + cyclohexene6,7 1 N/A 57 (97) 26 - 42 13 19 1 Cyclohexane, 2Cyclohexene, 3Cyclohexanol, 4Cyclohexanone. 5 The experiment was conducted with the addition of 1.0 g glycerol, glycerol conversion was 98 mol%. 6 The experiment was conducted with the addition of 1.0 g cyclohexene, cyclohexene conversion was 97 mol%. 7 The selectivity was calculated from the total moles of compounds detected after the reaction minus the moles of phenol and cyclohexene.   Simultaneous reaction of phenol and cyclohexene, without glycerol, was then performed using Raney Ni®. In this system, phenol conversion was 57 mol% while cyclohexene conversion was 97 mol%. Cyclohexane, benzene, cyclohexanol, and cyclohexanone were found among the products with selectivities of 26%, 42%, 13%, and 19%, respectively. This result indicates that cyclohexene can act as a hydrogen donor for phenol hydrogenation.  121   Figure 6.3 Schematic IGAPHdo reaction with Raney Ni® and H-ZSM-5 catalyst.  Reaction of cyclohexanol with Raney Ni® without glycerol as a hydrogen source produced cyclohexane and benzene. Without Raney Ni®, reaction of cyclohexanol produced cyclohexene. These two results indicate that cyclohexanol dehydration produces cyclohexene which can react with itself to produce benzene and cyclohexane, as illustrated in Fig. 6.3. It seems that cyclohexene can also donate hydrogen to phenol as shown in Table 6.1 entry 6. Hydrogen donation from cyclohexene results in formation of benzene.    6.2.2 Controlling phenol HDO products with Pt/C In Chapter 5, Pt/C was found to be a promising catalyst for IGAPH. Despite lower phenol conversion in the closed system, IGAPH with Pt/C in open system reaction resulted in high phenol conversion. Interestingly, during this reaction glycerol conversion was lower than phenol conversion which indicates a more efficient IGAPH reaction. Thus, there is a strong possibility to 122  control product selectivity of phenol HDO during IGAPHdo by employing Pt/C instead of Raney Ni®.  OH O OHH2 H2H2OH2H2-4 -27-49-22-16H2-27Pt/C Pt/C AmberlystPt/C Figure 6.4 Hydrodeoxygenation reaction of phenol. The Gibbs free energy (ΔG kJ/mol) at 220 °C is given for each reaction.  Fig. 6.4 shows predicted pathways of HDO of phenol to produce benzene or cyclohexane. In the attempt to produce benzene, dehydrogenation of cyclohexene should exceed the hydrogenation of cyclohexene. Therefore, there should be sufficient phenol available to serve as a hydrogen scavenger for cyclohexene dehydrogenation. In order to preserve phenol, hydrogenation of phenol should occur more slowly than dehydration of cyclohexanol so that when cyclohexene is produced, phenol is still available. In order to achieve this condition, the ratio of Amberlyst-15 to Pt/C should be high. In addition, the hydrogen source, in this case glycerol, should be limited so that cyclohexene hydrogenation to cyclohexane does not occur. Instead, cyclohexene should donate hydrogen to phenol and form benzene. On the other hand, to obtain high selectivity of cyclohexane the ratio between Amberlyst-15 to Pt/C should be low so that phenol will be saturated before cyclohexene is formed. Consequently, there will be a reduced chance of cyclohexene losing hydrogen and forming benzene. Glycerol loading should also be sufficiently high so that as cyclohexene forms, it can be hydrogenated to cyclohexane. 123  A series of experiments were conducted to test this proposal; results are presented in Table 6.2. Without an acid catalyst (entries 1-4), Pt/C can promote HDO reaction of phenol to produce cyclohexane and benzene. At high Pt/C and glycerol loading (1.0 g of Pt/C and 1.0 g of glycerol), cyclohexane is a major product (entry 1 to 3). Although the selectivity of products changes slightly with reaction pressure, the ratio of O/C and H/C were similar. Glycerol conversion varied with reaction pressure but not proportionally. Methanol and ethanol were the major liquid products at 4.34 MPa while 1,2-propylene glycol was found as the primary liquid product at 9.65 MPa.  Low glycerol loading (0.6) resulted in high selectivity to benzene (41%), cyclohexanone (21%), cyclohexane (19%), and cyclohexanol (19%). Phenol conversion (100 mol%) was greater than glycerol conversion (86 mol%). During this reaction, the O/C ratio was similar to that obtained after the reaction with 1.0 g glycerol at 4.34 MPa (entry 1). The H/C ratio of entry 4 was lower than entry 1 due to the limited hydrogen.  Phenol (96 mol%) and glycerol (83 mol%) conversion in the reaction with Pt/C and H-ZSM-5 are similar to those without an acid catalyst. Cyclohexane was the primary product followed by benzene, cyclohexanone, and cyclohexanol with selectivities of 51%, 32%, 11%, and 6%, respectively. As expected, the acid catalyst increased dehydration leading to a very low O/C ratio (0.03). Surprisingly, higher H/C ratio was obtained from this reaction.  A similar result was also observed when Amberlyst-15 was used.    124  Table 6.2 IGAPHdo reaction with Pt/C catalyst. The reactions were conducted with 1 g of phenol at 220 ℃ in an open system reaction with He flow of 0.03 – 0.06 L/min for 2h. Acid catalyst loading was 0.01 g.  No Glycerol (g) Catalyst P (MPa) Phenol conversion  Glycerol Pt (g) Acid catalyst X (mol%) Selectivity (%) C- balance O/C H/C  X (mol%) Selectivity (%) C-xanea Benzene C-nolb C-onec  EGd PGe MeOHf EtOHg 1 1 1 N/A 4.34 100 38 4 37 21 0.96 0.10 1.96  72 2 5 6 5 2 1 1 N/A 6.89 98 47 8 27 17 1.06 0.07 1.92  58 2 3 - - 3 1 1 N/A 9.65 100 38 3 44 15 0.93 0.10 1.97  68 2 3 - - 4 0.6 0.5 N/A 4.34 100 19 41 19 21 0.90 0.07 1.59  86 1 1 - - 5 0.6 0.5 H-ZSM-5 4.34 96 51 32 6 11 0.94 0.03 1.68  83 - - - - 6 0.6 0.5 Amberlyst-15 4.34 96 52 20 12 16 0.94 0.05 1.80  89 1 2 - - a Cyclohexane, bCyclohexanol, cCyclohexanone, dethylene glycol, e1,2-propylene glycol, fmethanol, gethanol     125  6.2.2.1 Targeting benzene production Since high selectivity of benzene was obtained without any acid catalyst, several reactions without any acid catalyst were conducted. In this reaction a very limited amount of glycerol (0.3 g) was used to convert phenol (1.0 g) to benzene. The amount of Pt/C was reduced to 0.1 g to slow the hydrogenation reaction. Results are presented in Table 6.3, entries 1-3. Longer reaction times resulted in higher phenol conversion and benzene selectivity. Phenol conversion increased from 30 mol% after 120 min reaction to 86 mol% after 360 min reaction. At 120 min, cyclohexanone and cyclohexanol were the major products with the selectivity of 51% and 30%, respectively. The selectivity of cyclohexanone was higher than cyclohexanol since glycerol, the primary hydrogen source, was very limited. With longer reaction times, benzene and cyclohexane selectivity increased while cyclohexanol and cyclohexanone selectivity decreased, indicating progression of HDO. The longer reaction time decreased both O/C and H/C ratios. Unfortunately, the longer reaction time also resulted in very low mole balance of phenol (42 mol%). In this set of reactions, glycerol was fully converted after 240 min. Addition of 0.01 g H-ZSM-5 for 240 min slightly increased phenol conversion to 79 mol%. Cyclohexane selectivity (11%) also increased but benzene selectivity was similar (66%). Unfortunately, this reaction also resulted in low mole balance of phenol. The reaction was then conducted for 120 min but with higher Pt/C loading (0.5 g) to compensate for the shorter reaction time (entry 5). Amberlyst-15 loading was increased to 0.1 g. Phenol conversion increased significantly (86 mol%) as did the selectivity of deoxygenated compounds. 126  Table 6.3 IGAPHdo reaction with Pt/C catalyst to produce benzene. Unless otherwise stated, the reactions were conducted with 1.0 g of phenol, 0.1 g acid catalyst at 220 ℃ in an open system reaction with He flow of 0.03 – 0.06 L/min under constant pressure of 4.34 MPa. No Glycerol (g) Catalyst t (min) Phenol conversion  Glycerol Pt (g) Acid catalyst X (mol%) Selectivity (%) Carbon balance O/C H/C X (mol%) Selectivity (%) C-xanea Benzene C-nolb C-onec EGd PGe MeOHf EtOHg AcHh 1 0.3 0.1 N/A 120 30 4 15 30 51 1.03 0.13 1.85  91 2 12 5 7 0 2 0.3 0.1 N/A 240 65 6 68 7 20 0.82 0.04 1.32  100 - 3 - - - 3 0.3 0.1 N/A 360 86 11 89 - - 0.42 0.00 1.11  100 - 3 - - - 4 0.3 0.1 H-ZSM-5 (0.01) 240 79 13 66 6 15 0.74 0.03 1.34  100 - - - - - 5 0.3 0.5 Amberlyst-15 120 86 1 99 - - 0.89 0.00 1.01  69 - - - - - 6j 0.3 0.5 Amberlyst-15 120 99 4 96 - - 0.98 0.00 1.04  100 - - - - 1 7j 0.15 0.5 Amberlyst-15 120 68 - 100 - - 0.74 0.00 1.00  100 - - - - - a Cyclohexane, bCyclohexanol, cCyclohexanone, dethylene glycol, e1,2-propylene glycol, fmethanol, gethanol. hacetic acid. iReaction was conducted with 0.01 g of H-ZSM-5. j Reactions were conducted at 240 ℃ with constant He pressure of 7.58 MPa.   127  Table 6.4 IGAPHdo reaction with Pt/C catalyst to produce cyclohexane. Unless otherwise stated, the reactions were conducted with 1.0 g of phenol, 0.01 g acid catalyst for 2h in an open system reaction with He flow of 0.03 – 0.06 L/min under constant pressure of 4.34 MPa. No Glycerol (g) Catalyst T (℃) Phenol conversion  Glycerol Pt (g) Amberlyst-15 X (mol%) Selectivity (%) Carbon balance O/C H/C  X (mol%) Selectivity (%) C-xanea Benzene C-nolb C-onec  EGd PGe MeOHf EtOHg AcHh 1 0.6 1 0.01 220 99 74 21 2 3 1.15 0.01 1.79  95 - - - 3 - 2 1.0 1 0.01 220 100 31 3 42 24 0.90 0.11 1.97  78 1 3 2 0 1 3 1.0 1 0.02 220 100 50 3 34 13 1.08 0.08 1.97  82 1 4 - - 1 4 0.6 1 0.01 240 100 73 27 - - 0.96 0.00 1.73  100 - - - - - 5 1.0 1 0.01 240 100 47 13 27 14 0.81 0.07 1.87  86 1 2 - - - a Cyclohexane, bCyclohexanol, cCyclohexanone, dethylene glycol, e1,2-propylene glycol, fmethanol, gethanol. hacetic acid. i Reaction was conducted under constant 7.58 MPa He pressure.   128  All the reactions with Amberlyst-15 removed oxygen (O/C =0). The reaction with Amberlyst-15 at 240 °C (entry 6) resulted in higher phenol conversion as well as slightly higher H/C ratio since cyclohexane selectivity increased. Glycerol was fully converted. Lowering glycerol loading (entry 7) resulted in benzene as the sole product from phenol HDO. However, the mole balance was decreased to 74 mol%.   6.2.2.2 Targeting cyclohexane production In an attempt to obtain high selectivity of cyclohexane, Pt/C and glycerol loadings were increased. These results are presented in Table 6.4. The reaction with 0.6 g glycerol and 0.01 g Amberlyst-15 resulted in 99 mol% phenol conversion and 95 mol% glycerol conversion. Cyclohexane had the highest selectivity (74%) followed by benzene (21%), cyclohexanone (3%), and cyclohexanol (2%). The O/C ratio was 0.01 while H/C ratio was 1.79. Unfortunately, cyclohexane selectivity did not increase when glycerol loading was increased to 1.0 g (entry 2), instead fewer deoxygenated products were found; cyclohexanol was the primary product with selectivity of 42%. Although higher glycerol loading (entry 2) slightly increased the H/C ratio, the O/C ratio was also increased. Increasing Amberlyst-15 loading to 0.02 g (entry 3) decreased O/C ratio and maintained H/C ratio. However, the selectivity of cyclohexane was less than in the reaction with low glycerol loading (0.01 g).  The reaction at 240 °C (entry 4) resulted in higher selectivity of deoxygenated compounds. In the reaction with 0.6 g of glycerol, cyclohexane selectivity was 73% and benzene selectivity was 27%.  Cyclohexanol and cyclohexanone were not detected in the products (O/C = 0). The H/C ratio after this reaction (entry 4) is slightly lower than the similar reaction at 220 °C (entry 1). All glycerol was converted into gaseous products during this reaction. A similar effect was observed where the 129  addition of glycerol (entry 5) decreased the selectivity of cyclohexane and benzene to 47% and 13%, respectively. Higher cyclohexanol selectivity (27%) compared to cyclohexanone (14%) was observed in this reaction. Both O/C and H/C ratios during this reaction were slightly lower than the identical reaction at 220°C (entry 2). Glycerol conversion was 86 mol% and 1,2-propylene glycol and ethylene glycol were observed as liquid intermediates.   6.3 Discussion 6.3.1 IGAPHdo with Raney Ni® From the reactions in Table 6.1, cyclohexanol dehydration results in higher selectivity of cyclohexane. In the reaction with Raney Ni® and H-ZSM-5, the ratio of cyclohexane and benzene were close to the theoretical stoichiometry. In the ideal reaction, cyclohexanol will be dehydrated to cyclohexene. Cyclohexene can then react with itself to produce cyclohexane and benzene in a ratio of 2 to 1. This mechanism is in agreement with previous work by Wang et al.130 The closest result to stoichiometric behavior was obtained in the absence of Raney Ni®. However, without Raney Ni®, the reaction of cyclohexene proceeds more slowly. The slow reaction of cyclohexene was also reflected in Table 6.1 entry 3 where cyclohexene conversion was only 54 mol%. The reactivity of cyclohexene increased when Raney Ni® was added (Table 6.1, entry 4). Cyclohexene with and without Raney Ni® produced similar selectivity of cyclohexane and benzene.  Hydrogenation of cyclohexene with glycerol resulted in lower conversion of cyclohexene but higher cyclohexane selectivity compared to the reaction of cyclohexene without glycerol. This result demonstrates the ability of glycerol APR to donate hydrogen to cyclohexene to produce cyclohexane. However, low cyclohexene conversion in the presence of glycerol indicates a competitive reaction between both compounds. Given the ratio of glycerol to cyclohexene, the 130  amount of hydrogen consumed by cyclohexene to produce cyclohexane was very low compared to the hydrogen that could be produced by glycerol. After phenol hydrogenation with cyclohexene, the selectivity of cyclohexane was lower than the selectivity of benzene. Assuming that cyclohexane and benzene came from cyclohexene, this result indicates that cyclohexene can donate hydrogen to phenol.  The high selectivity of cyclohexane from the cyclohexanol reaction was contrary to most IGAPHdo results in which benzene was the major product. Relatively high benzene selectivity was found in most IGAPHdo with Raney Ni® and H-ZSM-5. In addition, glycerol conversion was relatively high with relatively low CO2/CH4 ratio. The low CO2/CH4 ratio and the high selectivity of benzene indicate greater dehydration of glycerol to alkanes relative to APR of glycerol to CO2 and H2. However, the low CO2/CH4 ratio was not proportional to the acid loading as shown in Figure 6.2a. High glycerol conversion but low aromatic ring saturation (high benzene selectivity) reflects the inefficient IGAPHdo with Raney Ni® and H-ZSM-5.  6.3.2 Controlling selectivity of products with Pt/C Pt/C demonstrated the ability to conduct HDO without the addition of an acid catalyst. The high selectivity of glycerol APR with Pt/C (shown by high CO2/CH4 in Fig 5.10) may result in high CO2 concentration in solution which creates additional acidity relative to the acidity of high temperature water. Low glycerol loading seems to be the most important factor to obtain high benzene selectivity. Although no acid catalyst is needed, phenol HDO reaction should be conducted at very low Pt/C loading so that phenol hydrogenation is not too fast relative to dehydration. Consequently, a longer reaction time is required but, unfortunately, this results in a 131  low mole balance. Simultaneous use of Pt/C and acid catalyst seems to accelerate the reaction and maintain a high mole balance.  Producing cyclohexane requires a substantial amount of H2. This is difficult to obtain from in-situ hydrogen generation by glycerol APR. High glycerol loading can result a disproportionate amount of self-hydrogenation of glycerol relative to phenol hydrogenation as shown in Fig. 5.5 (P/G=0.1). Unexpectedly, the addition of glycerol lowered the degree of HDO and generated more cyclohexanone and cyclohexanol. Under these conditions, the higher glycerol loading appears to reduce cyclohexanol dehydration instead of reducing phenol hydrogenation. Reduced dehydration might occur due to occupation of acid sites or by reduction of the dissociation constant of water. Thus, the main challenge in the production of cyclohexane is to preserve sufficient glycerol concentration as the reaction proceeds while simultaneously maintaining cyclohexanol dehydration. Although the sulfonate group of Amberlyst-15 may leach at temperatures above 190°C,196 it displayed good activity as a dehydration catalyst during the first run. The resistance of Pt/C to acidic media might explain the high degree of HDO observed after reaction with Pt/C and Amberlyst-15. In the case of Raney Ni®, acid poisoning of Ni may explain the slightly lower phenol conversion relative to the reaction without an acid catalyst.   6.4 Conclusion IGAPHdo can be conducted by simultaneously employing hydrogenation and acid catalysts. The formation of benzene and cyclohexane occurs through dehydration of cyclohexanol or cyclohexanone to produce cyclohexene. Cyclohexene can either lose hydrogen to produce benzene 132  or receive hydrogen to form cyclohexane. In the presence of Raney Ni®, the high glycerol APR rate and low cyclohexene hydrogenation rate resulted in high benzene selectivity. In the presence of Pt/C, phenol hydrogenation likely occurs more rapidly than glycerol APR therefore high selectivity of benzene can be obtained with little glycerol. Although high selectivity of cyclohexane (~74%) can be obtained using moderate glycerol loading and a low ratio of Amberlyst-15-to-Pt/C, identifying an ultra-selective cyclohexane condition (>95%) still remains a challenge.  133  Chapter 7: Upgrading of Ligneous Material via In-Situ Glycerol APR    7.1 Introduction FPO (fast pyrolysis oil) is a high energy-density product from renewable lignocellulosic biomass. The flexibility of the feedstock and the short reaction time are the key advantages of fast pyrolysis over other processes, such as enzymatic hydrolysis or liquefaction. FPO, however, is not a drop-in fuel, so further upgrading is required. Experiments with phenol upgrading demonstrated that Pt/C is an active catalyst for IGAPH and IGAPHdo reactions. Therefore, in this chapter, Pt/C will be used as the catalyst for FPO upgrading via in-situ glycerol APR.   There are differences between FPO and model compounds, which make FPO upgrading via glycerol APR more challenging. First, the thermal instability of FPO can result in char formation due to condensation reactions.151 The interaction between water-soluble compounds and water-insoluble compounds in FPO can also promote char formation.153 A second difference is the insolubility of FPO in water. Unlike phenols and other model compounds such as vanillin and guaiacol, which are slightly soluble in water, FPO is very insoluble in water. The presence of polymeric lignin might be the reason, although several of the monomers in FPO may also be very insoluble.   The hydrogenolysis of SPF (a mix of spruce, pine and fir) will also be conducted to understand the differences between the thermally deconstructed carbohydrate and lignin in FPO and the native form of carbohydrates and lignin in SPF during hydrogenolysis via glycerol APR. Unlike FPO, 134  which contains some monomeric compounds, lignin and carbohydrates in SPF are in polymeric form. In addition, these carbohydrates and lignin have not undergone thermal treatment above 500°C. Dehydration, depolymerization, and condensation products are therefore not expected to appear in SPF. However, both FPO and SPF lignin require depolymerization, since they are mostly in polymeric form. Once lignin monomers from FPO or SPF are formed, different behaviors between various lignin monomers are expected, as was observed from the hydrogenation of model compounds via glycerol APR described in Chapter 5.  Higher glycerol loading, with limited water loading compared to the reactions with model compounds, was applied to reactions with FPO and SPF. Limited water loading was applied to prevent the precipitation of ligneous material in excess water. This Chapter will investigate the effects of additional solvents, the presence of an acid catalyst, and reaction time.  7.2 Results  7.2.1 FPO composition analysis FPO was purchased from BTG Bioliquid (Netherlands) and contained 29.0 wt% water (Karl Fischer titration). The pyrolytic lignin content was measured by performing the precipitation of FPO according to Meier et.al.197 Briefly, 60.0 g BTG FPO was added dropwise to 650.0 g stirred ice-cold water. Upon precipitation, the solution was filtered and the residue was regarded as pyrolytic lignin, which was then added to 1.0 kg water and stirred for 4 h, followed by centrifugation to further wash out the water-soluble compounds. Up to 16.2 wt% dark-brown dried pyrolytic lignin from BTG FPO was obtained after vacuum evaporation at room temperature for 4 days.   135  GC-MS analysis of BTG FPO was conducted by diluting the FPO with acetone in a 1:1 ratio; the chromatograph is shown in Fig 7.1. FPO contains mostly organic acids, oxygenated aliphatics, methoxyphenols, furanics, and oxygenated cycloaliphatics. It can be seen that most of the phenolic compounds appear after a retention time of 25 min. The quantification of BTG FPO was carried out by diluting the FPO with acetone and by adding butanol as an internal standard. Identified compounds accounted for 38 wt% of the FPO (Table 7.1, column 1).  The monomers in the FPO—and later in the products of FPO upgrading and SPF hydrogenolysis—were classified based on the corresponding source of the compounds, whether it was a carbohydrate derivative or a lignin derivative. For paraffins and oxygenated aliphatics (i.e. non-cyclic aliphatic compounds containing oxygen— e.g. ethanol, acetol, hexanone), if the backbone of a compound had 5 or fewer carbons, it was considered a carbohydrate derivative. Similarly, for naphthenes and oxygenated cycloaliphatics (i.e. cyclic aliphatics with oxygen functional groups—e.g.  cyclohexanol, cyclopentanone), if the carbon number of a cyclic compound was 5 or lower, it was considered a carbohydrate derivative. All of the furanics/pyranics and organic acids were considered carbohydrate derivatives, while all of the aromatics, naphthalenes, non-methoxylated phenols, and methoxyphenols were considered lignin derivatives.  136   Figure 7.1 Chromatograph and the appearance of BTG FPO. IS: internal standard.  As much as 15.2 wt% (S = 39.6%) of the water-free BTG FPO was organic acids, with acetic acid (11.8 wt%) being the most abundant compound in this group. Oxygenated aliphatics were the second highest constituent of FPO, at 11.8 wt% (S = 30.6%), while acetol (7.8 wt%) was the third. Furanics/pyranics compounds contributed 4.9 wt% (S = 12.7%). Only a few lignin derivatives could be detected, with all of them in the form of methoxyphenols. As much as 3.3 wt% (S = 8.5%) of the methoxyphenols could be detected from the BTG FPO, with p-creosol as the most abundant product. Oxygenated cycloaliphatics contributed 3.3 wt%. Accompanying all of them were carbohydrate derivatives such as cyclopentanones. In total, 38.5 wt% monomers were detected in BTG FPO, with 35.2 wt% these being carbohydrate derivatives and 3.3 wt% lignin derivatives.    137  Table 7.1 Composition of BTG FPO and pyrolytic lignin   BTG FPO   Pyrolytic Lignin Carbohydrate Derivatives Lignin Derivatives S2 (%) Carbohydrate Derivatives Lignin Derivatives S2 (%) Yt 1 (wt%) Yt 1 (wt%) Yt 1 (wt%) Yt 1 (wt%) Furanics/pyranics 4.9 - 12.7  0.2 - 2.9 Oxygenated aliphatics 11.8 - 30.6  4.4 - 52.3 Oxygenated cycloaliphatics 3.3 - 8.5  0.2 - 2.6 Non-methoxylated phenols - - -  - 0.6 7.2 Methoxyphenols - 3.3 8.5  - 2.7 32.6 Organic acids 15.2 - 39.6  0.2 - 2.5 Total yield each (wt%)1 35.2 3.3 100.0  5.0 3.3 100.0 Total yield of all (wt%)1 38.5   8.3  1Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water-free FPO (Equation 3.14). Selectivity was calculated based on Equation 3.15.  The monomeric content in pyrolytic lignin was also analyzed by dissolving pyrolytic lignin in acetone in a 1:1 ratio. A monomer content of up to 8.3 wt% was detected from completely dissolved pyrolytic lignin, which consisted of 5.0 wt% carbohydrate derivatives and 3.3 wt% lignin derivatives (Table 7.1 column 2). Except for non-methoxylated phenols, which had a selectivity of only 7.2% (0.6 wt% yield) in the total detected compounds, most of the monomers detected in pyrolytic lignin were also detected in whole FPO. With respect to whole FPO, the yield of non-methoxyphenol would be less than 0.1 wt% and thus could not be detected. Despite the presence of non-methoxylated phenols, the composition of pyrolytic lignin indicated residual monomers, which had been detected in whole BTG-FPO. Since 8.3 wt% of pyrolytic lignin was detected as monomers, this leaves 14.8 wt% of whole FPO as an undetected fraction of pyrolytic lignin, which 138  is considered as polymeric lignin. Humin was not considered part of the undetected fraction, since pyrolytic lignin was completely soluble in acetone while only 11.0 wt% humin can be dissolved in acetone.198   7.2.2 FPO upgrading BTG FPO upgrading was first conducted without any additional solvent. Initially, 10.0 g BTG FPO was mixed with 10.0 g glycerol, 1.0 g Pt/C, and 0.25 g H-ZSM-5 at 300°C for 7 h with 10.34 MPa initial He pressure. The reaction was carried out in a semi-open type of reactor with a refluxed condenser and backpressure regulator to achieve a maximum pressure as high as 19.31 MPa (slightly above the saturation vapor pressure of methanol at 300°C). Since the BTG FPO was used without any treatment and contained 29.0 wt% water, additional water was not added to the reaction. Upon the completion of the reaction, a clear solution with a small portion of black insoluble oil was produced, as shown in Fig D.1. A single phase of solution was obtained after filtration. GC-MS quantification using DMSO as an internal standard showed a low yield of monomeric compounds. Organic acids were the dominant compounds in this solution (excluding water). Despite being the dominant compound in the aqueous phase, the yield of organic acids was only 1.0 wt%. In total, 1.5 wt% monomers could be detected from the aqueous phase, with the remaining 0.4 wt% in the form of oxygenated aliphatics and 0.1 wt% as non-methoxylated phenol.   139  Table 7.2  Quantitative results for upgraded FPO using Pt/C and H-ZSM-5 catalysts1   Carbohydrate Derivatives  Lignin Derivatives S (%)  Ya2 (wt%) Yw2 (wt%) Yt2 (wt%) Ya2 (wt%) Yw2 (wt%) Yt2 (wt%) Paraffins - - -  - 0.1 0.1 1.8 Naphthenes - - -  - 0.2 0.2 5.5 Aromatic - - -  - 1.1 1.1 24.7 Naphthalenes - - -  - 0.5 0.5 12.1 Oxygenated aliphatics 0.4 - 0.4  - - - 9.4 Non-methoxylated phenols - - -  0.1 0.9 1.0 23.5 Organic acids 1.0 - 1.0  - - - 23.1 Total yield of each (wt%)2 1.4  2.9 100 Total yield of all (wt%)2 4.4 Char (wt%)3 6.3 1 The reactions were conducted with 10.0 g BTG FPO, 1.0 g Pt/C, and 0.25 g H-ZSM-5 at 300°C for 7 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas formation and was released at 19.31 MPa. 2 Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water-free FPO (Ya: yield in aqueous phase; Yw: yield in washed catalyst solution; Yt: total yield in all phases). 3 Char was determined from the weight difference of the catalyst after being washed with acetone and dried in a vacuum oven at 50°C overnight, and it was calculated based on water-free FPO.  To recover water-insoluble compounds, the catalyst was washed with acetone. Several water-insoluble compounds were found in the washing solution, with yields of 0.9, 1.1, and 0.5 wt% for non-methoxylated phenols, aromatics, and naphthalenes, respectively. Paraffins and naphthenes amounted to 0.1 and 0.2 wt%, respectively, giving a total 2.9 wt% lignin-derived compounds (Table 7.2). Only 6.3 wt% char was formed after the reaction. Non-methoxylated phenols, aromatics, and naphthalenes, which are less soluble in water, were found only in the catalyst washing solution. This result indicates the importance of reactant and product solubility in the reaction solvents.  140  To improve the solubility of lignin derivatives, which are less polar than water or glycerol, n-decane was added as a co-solvent. BTG FPO upgrading was then conducted with 5.0 g glycerol as a hydrogen source, along with 1.0 g Pt/C catalyst and 5.0 g n-decane. The reactor was pressurized with 10.34 MPa of He and was heated to 300°C for 2.5 h. Since n-decane was employed in the reaction, two liquid phases were formed from the beginning to the end of the reaction, as shown in Table D.2. The compounds in the organic phase were quantified with butanol as an internal standard, while the compounds in the aqueous phase were quantified with DMSO as an internal standard. Only 12.4 wt% (Table 7.3, column 1) of the monomeric compounds were detected upon completion of the reaction, a lower amount than the total detected in raw BTG FPO (38.5 wt%).  Organic acids were the predominant compounds, with a yield of 4.5 wt% relative to water-free FPO and were mostly found in the aqueous phase. Acetic acid was still the predominant compound in this group, along with propanoic, butanoic, and pentanoic acids. Non-methoxylated phenols were the predominant group in the organic compounds, with a 1.9 wt% yield. Cresol isomers were the dominant products in this group, along with ethyl and propyl phenols, and were mostly found in the organic phase. Oxygenated aliphatics were mostly found in the aqueous phase and accounted for 2.4 wt%. Among the compounds in oxygenated aliphatics, 2-hexanone dominated in the organic phase, while acetone dominated in the aqueous phase. Aromatics such as ethylbenzene and toluene were found in amounts as much as 1.0 wt% and were the second largest group in the organic phase. Naphthene accounted for as much as 1.3 wt% and consisted only of methyl cyclopentane, cyclohexane, and methylcyclohexane. In total, 7.8 wt% carbohydrate derivatives and 4.6 wt% lignin derivatives could be recovered from the reaction.  In an attempt to depolymerize more polymeric lignin, the upgrading was then conducted with a higher glycerol loading (10.0 g) and a longer reaction time (7 h). After the reaction, as much as 141  20.3 wt% water-free BTG FPO was recovered. Oxygenated aliphatics were the highest-yield products, at 8.5 wt%, and most of them were 2-pentanone and 2-hexanone in the organic phase, as well as acetone and 2-propanol in the aqueous phase. Paraffins were the second major products from this reaction, with a yield of 3.2 wt%, and were recovered in the organic phase. Among the paraffins, pentane and hexane were two major compounds. The naphthenes yield was 3.1 wt%, with methyl cyclopentane and cyclohexane being the two major compounds in this group. All of the naphthenes were recovered in the organic phase. Organic acids contributed 1.9 wt%, which was found in both the aqueous and the organic phases. Acetic acid and propanoic acid were two major compounds in this group. Non-methoxylated phenols contributed 1.3 wt% and were mostly in the organic phase. Oxygenated cycloaliphatics contributed 0.9 wt% and were present in the organic phase. Aromatics contributed 0.9 wt%, majority being toluene and indanes. Furanics, naphthalenes, and methoxyphenols were also found in the products, with yields of 0.2, 0.1, and 0.1 wt%, respectively.  The chromatograph and appearance of upgraded BTG FPO with 10.0 g glycerol reacting for 7 h are presented in Fig. 7.2. It can be seen that several high-intensity peaks appear in the 1–2.5 min range, most of them paraffins and naphthenes. Most of the aromatics appear between 2.5 and 15.5 min, followed by several oxygenated cycloaliphatics and naphthalenes (15–30 min). Most of the phenolic compounds appear after 30 min and with relatively low intensity. Several oxygenated aliphatics appear across a wide range of retention times in the organic phase. The aqueous phase of upgraded FPO contained only oxygenated aliphatics, organic acids, and a few oxygenated cycloaliphatics. Table 7.3, column 2 shows the quantitative results for raw BTG FPO and upgraded FPO using a Pt/C catalyst.   142                   Figure 7.2 Chromatographs and appearances of upgraded BTG FPO: (a–c) chromatographs of the organic phase, (d) chromatograph of the aqueous phase. The insert in (a) shows the whole spectrum of the organic phase. The experiment was conducted with 10.0 g BTG FPO, 5.0 g n-decane, 10.0 g glycerol, and 1.0 g Pt/C at 300°C for 7 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.      (a) (b) (c) (d) 143  Table 7.3 Quantitative results for upgraded FPO using Pt/C catalysts with the addition of n-decane as a co-solvent1   Glycerol (5.0 g), 2.5 h (I)  Glycerol (10.0 g), 7 h (II) Carbohydrate Derivatives Lignin Derivatives S (%) Carbohydrate Derivatives Lignin Derivatives S (%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Paraffins - 0.2 0.2 - 0.1 0.1 2.5 - 2.2 2.2 - 1.0 1.0 15.6 Naphthenes - 0.7 0.7 - 0.6 0.6 10.2 - 1.5 1.5 - 1.6 1.6 15.3 Aromatic - - - - 1.0 1.0 8.3 - - - - 0.9 0.9 4.6 Naphthalenes - - - - 0.2 0.2 1.9 - - - - 0.1 0.1 0.7 Furanics/pyranics - - - - - - - - 0.2 0.2 - - - 1.2 Oxygenated aliphatics 1.6 0.5 2.1 - 0.3 0.3 19.6 1.4 3.8 5.2 - 3.3 3.3 42.0 Oxygenated cycloaliphatics - 0.1 0.1 0.1 0.4 0.5 4.9 - 0.1 0.1 - 0.8 0.8 4.5 Non-methoxylated phenols - - - 0.1 1.8 1.9 15.0 - - - - 1.3 1.3 6.3 Methoxyphenols - - - - 0.1 0.1 0.9 - - - - 0.1 0.1 0.5 Organic acids 3.2 1.4 4.5 - - - 36.6 0.7 1.2 1.9 - - - 9.3 Total yield each (wt%)2 7.8 4.6 100.0 11.0 9.3 100.0 Total yield of all (wt%)2 12.4 20.3 Char (wt%)3 2.7 0.9 1 The reactions were conducted with 10.0 g BTG FPO, 5.0 g n-decane, and 1.0 g Pt/C at 300°C. In experiment (I), 5.0 g glycerol was added and the reaction was conducted for 2.5 h. In experiment (II), 10.0 g glycerol was added and the reaction was conducted for 7.0 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa. 2 Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water-free FPO (Ya: yield in aqueous phase; Yo: yield in organic (n-decane) phase; Yt: total yield in all phases). 3 Char was calculated from the weight difference of the catalyst after being washed with acetone and dried in a vacuum oven at 50°C overnight and was calculated based on water-free FPO.  144  Solid acid catalyst (H-ZSM-5, 0.25 g) was added to improve the HDO reaction. In total, 13.4 wt% monomeric compounds were recovered from the upgraded FPO (Table 7.4, column 1). Naphthenes were the main products, with a yield of 4.3 wt% (S =31.7%), and methylcyclohexane was the dominant compound in this group. The yield of paraffins was 3.0 wt% (S = 22.6%), with hexane dominating. Aromatics accounted for 3.2 wt% (S = 23.5%) and naphthalenes for 0.4 wt% (3.4% selectivity). Toluene and ethylbenzene dominated in the aromatics group and methyl naphthalenes in the naphthalenes groups. Non-methoxylated phenol in the amount of 0.6 wt% was obtained, with 2,4-dimethyl-phenol as the major compound. The upgraded FPO contained only organic acids, yielding 1.4 wt% dried BTG FPO in which acetic acid was the major compound.  Figure 7.3 shows the chromatograph of FPO upgraded with Pt/C and H-ZSM-5. Several high-intensity peaks appeared early on, between 1 and 4 min, and they had higher intensity than in the reaction with only Pt/C. Most of these peaks were paraffins and naphthenes. Several aromatic compounds appeared between 3 and 23 min, followed by naphthalenes. Phenolic compounds appeared at a higher retention time. 145                 Figure 7.3 Chromatographs and appearances of upgraded BTG FPO. (a–b) chromatographs of the organic phase, (c) chromatograph of the aqueous phase. The insert in (a) shows the whole spectrum of the organic phase. The experiment was conducted with 10.0 g BTG FPO, 10.0 g glycerol, 5.0 g n-decane, 1.0 g Pt/C, and 0.25 g H-ZSM-5 at 300°C for 7 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.      (a) (b) (c) 146  Table 7.4 Quantitative results for upgraded FPO using Pt/C and H-ZSM-5 catalysts with the addition of n-decane as a co-solvent for 7 h of reaction time1    Pt/C (1.0 g) + H-ZSM-5 (0.25 g) (III)  Pt/C (2.0 g) + H-ZSM-5 (0.50 g) (IV) Carbohydrate Derivatives Lignin Derivatives S (%)  Carbohydrate Derivatives Lignin Derivatives S (%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%)  Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Paraffins - 1.1 1.1 - 1.9 1.9 22.6  - 0.7 0.7 - 1.2 1.2 16.3 Naphthenes - 1.4 1.4 - 2.9 2.9 31.7  - 0.8 0.8 - 3.8 3.8 39.8 Aromatic - - - - 3.2 3.2 23.5  - - - - 3.5 3.5 30.0 Naphthalenes - - - - 0.4 0.4 3.3  - - - - 0.5 0.5 4.5 Oxygenated aliphatics 0.2 0.3 0.5 - - - 4.0  0.2 - 0.2 - - - 1.4 Non-methoxylated phenols - - - - 0.6 0.6 4.3  - - - - - - 0.3 Organic acids 1.4 - 1.4 - - - 10.5  0.9 0.0 0.9 - - - 7.8 Total yield each (wt%)2 4.4 9.0 100.0  2.6 9.0 100.0 Total yield of all (wt%)2 13.4  11.6 Char (wt%)3 2.6  4.7 1 The reactions were conducted with 10.0 g BTG FPO, 10.0 g glycerol, and 5.0 g n-decane at 300°C for 7 h. In experiment (III), 1 g Pt/C and 0.25 g H-ZSM-5 were used. In experiment (IV), 2 g Pt/C and 0.50 g H-ZSM-5 were used. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa. 2 Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water-free FPO (Ya: yield in aqueous phase; Yo: yield in organic (n-decane) phase; Yt: total yield in all phases). 3 Char was calculated from the weight difference of the catalyst after being washed with acetone and dried in a vacuum oven at 50°C overnight and was calculated based on water-free FPO.  147  To further increase the reaction rate and monomeric lignin yield, BTG upgrading was conducted with double the initial catalyst loading (2.0 g Pt/C and 0.5 g H-ZSM-5). However, only 11.6 wt% monomer was recovered, along with 2.6 wt% carbohydrate-derived compounds and 9.0 wt% lignin-derived compounds. A relatively high yield of naphthenes (4.6 wt%) was obtained under these reaction conditions, with hexane as the major compound in this group (Table 7.3, column 2). The aromatics yield was 3.5 wt%, with propylbenzene and ethylbenzene as the major compounds. Paraffins, naphthenes, and aromatics contributed 86.1% of the total detected compounds in the products. Paraffin yield was 1.9 wt%, with n-hexane as the major compound. Organic acids and naphthalenes yielded 0.9 and 0.5 wt%, respectively while the yield of oxygenated aliphatics was 0.2 wt%. Neither furanics, oxygenated cycloaliphatics, nor methoxyphenols could be detected in the upgraded FPO.   7.2.3 Hydrogenolysis of SPF  Prior to hydrogenolysis reaction tests, a compositional analysis of SPF was conducted to obtain carbohydrate and lignin content; the results are shown in Table 7.5. Glucan and acid-insoluble lignin were two major components. SPF upgrading was conducted with 5.0 g SPF in a glycerol- water solution. Unlike the upgrading of BTG FPO, the hydrogenolysis of SPF was conducted with the addition of water, since SPF has a very low water content.  Table 7.5 Compositional analysis of SPF sawdust Water (wt%) Extractive (wt%) Dry and Extractive Free Composition (wt%) Glucan Xylan Galactan Arabinan Mannan ASL1 AIL2 Ash 5.9 3.9 39.7 6.2 2.9 1.1 9.5 3.69 35.4 0.4 1 ASL: acid-soluble lignin. 2 AIL: acid-insoluble lignin.  148  The first hydrogenolysis of SPF was conducted with 5.0 g SPF, 20.0 g water, 10.0 g glycerol, 1.0 g Pt/C, and 0.25 g H-ZSM-5 at 300 ℃ for 5 h. Upon reaction completion, a single aqueous phase of solution was obtained. Relatively high yields of organic acids (11.6 wt%) and oxygenated aliphatics (6.8 wt%) were found in the solution (Table 7.6, column 1). Acetic acid and propanoic acid were the two major compounds in the group, along with longer-chain organic acids, such as hexanoic acid. In the oxygenated aliphatics, ethanol and 1-propanol were the two major products. Several methoxyphenols and non-methoxylated phenols were also found in the aqueous solution, with yields of 0.3 and 0.1 wt%, respectively. Guaiacol and propyl guaiacol were found to be the major compounds in the group, while phenol, p-cresol, and 4-propyl phenol were the only non-methoxylated phenols found. As much as 1.6 wt% solid residue was found after the reaction.  Upon filtration, the catalyst was washed with acetone to ascertain the existence of more water-insoluble compounds. Methoxyphenols and non-methoxylated phenol were found to be the major lignin derivatives in the washing solution, both with a yield of 0.8 wt%. Furanics/pyranics were the major carbohydrate derivatives in the washing solution (2.2 wt%), followed by organic acids (1.3 wt%). Oxygenated aliphatics were also found (0.2 wt%). In addition, naphthalenes, naphthenes, aromatics, furanics/pyranics, and paraffins were found in very small amounts.   To improve the solubility of less-polar compounds, the reaction was then conducted with the addition of n-decane and a slightly longer reaction time (7 h). More specifically, 5.0 g SPF, 10.0 g water, 10.0 g glycerol, 5.0 g n-decane, 1.0 g Pt/C, and 0.25 g H-ZSM-5 were loaded into the reactor and heated at 300°C for 7 h with an initial pressure of 10.34 MPa. A total of 14.2 wt% monomeric compounds could be detected from the product, where 9.0 wt% were carbohydrate derivatives and 5.1 wt% were lignin derivatives. No solid residue from SPF was noticeable after this reaction.  149  Organic acids had the highest yield (7.2 wt%), with acetic acid and pentanoic acid being the predominant compounds in the aqueous and organic phases, respectively. Naphthenes were the second highest yield (3.9 wt%), consisting of 1.4 wt% carbohydrate derivatives and 2.5 wt% lignin derivatives. The yield of paraffins was 1.2 wt%, with hexane as the major product. The methoxyphenols and non-methoxyphenols yields were 0.9 and 0.6 wt%, respectively. 4-propylguaiacol and 4-propylphenol were the major methoxyphenol and non-methoxyphenol compounds, respectively. Ethanol, the only oxygenated aliphatic, was also found (0.2 wt%), as were aromatics and naphthalenes (0.1 wt% each); 6-methyltetralin and 9-ethyl-9-methylfluorene were the major products in each group. The quantities of monomeric compounds are presented in Table 7.6, column 2. 150  Table 7.6  Quantitative results of SPF hydrogenolysis using Pt/C and H-ZSM-5 catalysts1   Pt/C + H-ZSM-5, no decane (A)   Pt/C + H-ZSM-5, with n-decane1 (B)   Carbohydrate Derivatives Lignin Derivatives S (%)  Carbohydrate Derivatives Lignin Derivatives S (%) Ya2 (wt%) Yw2 (wt%) Yt2 (wt%) Ya2 (wt%) Yw2 (wt%) Yt2 (wt%)  Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%)  Paraffins - - - - 0.0 0.0 0.2  - 0.2 0.2 - 1.0 1.0 9.0 Naphthenes - - - - 0.0 0.0 0.2  - 1.4 1.4 - 2.5 2.5 26.9 Aromatic - - - - 0.1 0.1 0.3  - - - - 0.1 0.1 0.5 Naphthalenes - - - - 0.1 0.1 0.2  - - - - 0.1 0.1 0.8 Furanics/pyranics 0.2 2.2 2.4 - - - 9.6  - - - - - - - Oxygenated aliphatics 6.8 0.2 7.0 - - - 28.6  0.2 - 0.2 - - - 1.2 Oxygenated cycloaliphatics - - - - - - 0.1  - - - - - - 0.0 Non-methoxylated phenols - - - 0.1 0.8 0.9 3.9  - - - - 0.6 0.6 4.3 Methoxyphenols - - - 0.3 0.8 1.1 4.4  - - - - 0.9 0.9 6.1 Organic acids 11.6 1.3 12.9 - - - 52.5  5.6 1.6 7.2 - - - 51.1 Total yield each phase (wt%)2 22.2 2.2 100  9.0 5.1 100 Total yield of all phases (wt%)2 24.5  14.2 Solid residue (wt%)3 1.6  0.0 1 Unless otherwise mentioned, the reactions were conducted with 5.0 g SPF sawdust, 10.0 g glycerol, 20.0 g water, 1 g Pt/C, and 0.25 g H-ZSM-5 at 300°C. Reaction (A) was conducted for 5 h. Reaction (B) was conducted for 7 h with the addition of 5 g n-decane. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa. 10.0 g water was used in the reaction with the addition of 5.0 g n-decane, and the reaction was conducted for 7 h. 2 Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water- and extractive-free SPF (Ya: yield in aqueous phase; Yw: yield in washed catalyst solution (acetone); Yo: yield in organic (n-decane) phase; Yt: total yield in all phases). 3 Solid residue was calculated from the weight difference of the catalyst after being washed with acetone and dried in a vacuum oven at 50°C overnight and was calculated based on water- and extractive-free SPF. 151  To understand the effect of Pt/C and H-ZSM, SPF hydrogenolysis was then conducted without H-ZSM-5. During this reaction, 5.0 g SPF was mixed with 10.0 g water, 10.0 g glycerol, 5.0 g n-decane, and 1.0 g Pt/C loading, then heated to 300°C for 7 h with an initial He pressure of 10.34 MPa. 17.0 wt% monomeric compounds were obtained, 11.6 wt% of which were carbohydrate derivatives and 5.4 wt% lignin derivatives. Upon reaction completion, the solution was filtered and 0.1 g (2.6 wt%) solid residue was detected.  Oxygenated aliphatics were the highest-yield product at 4.7 wt% (4.3 wt% were found in the aqueous phase and 0.4 wt% in the organic phase), with ethanol predominating. Naphthenes were next at 4.2 wt%, with 1.4 wt% carbohydrate derivatives and 2.8 wt% lignin derivatives. Propyl cyclohexane was the major compound in this group. Organic acids (3.7 wt%) were found in the upgraded oil, especially in the aqueous phase, acetic acid being the major compound. Paraffins were 1.6 wt%, with carbohydrate and lignin derivatives contributing 1.0 and 0.6 wt%, respectively. Hexane was the major product in this group. The non-methoxylated phenols yield was 1.0 wt%, and the methoxyphenols yield was 0.8 wt%. 4-propyl phenol and 4-propyl guaiacol were the predominant compounds in non-methoxylated phenols and methoxyphenols, respectively. The aromatics yield was 0.2 wt%, with propyl benzene as the major product. Chromatographs of SPF hydrogenolysis are shown in Fig. 7.4, and quantities are presented in Table 7.7, column 1.    152                   Figure 7.4 Chromatographs and appearance of SPF hydrogenolysis products. (a) chromatograph of the organic phase, (b) chromatograph of the aqueous phase. The experiment was conducted with 5.0 g SPF, 10.0 g water, 10.0 g glycerol, 5.0 g n-decane, and 1.0 g Pt/C at 300°C for 7 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.  The reaction was then conducted with lower glycerol loading and a shorter reaction time to further understand the compounds’ degradation steps. The loadings of SPF, water, glycerol, n-decane, and Pt/C were 5.0, 10.0, 5.0, 5.0, and 1.0 g, respectively. The reaction was conducted at 300°C for 2.5 h with an initial He pressure of 10.34 MPa. Upon completion, 21.4 wt% monomeric compound   (b) (a) 153  was detected with 8.5 wt% solid residue. The monomer yield broke down into 19.9 wt% carbohydrate derivatives and only 1.5 wt% lignin derivatives.  Among the 19.9 wt% carbohydrate derivatives, oxygenated aliphatics contributed 10.0 wt%, with ethanol being the major compound in the group. Organic acids were the second highest group (9.2 wt%), with acetic acid being the predominant compound. Furanics were also found (0.4 wt%), and as much as 1.1 wt% methoxyphenols, with 4-propyl-2-methoxyphenol being the major compound. Naphthenes were present in small amounts (0.6 wt%). The quantitative results for this reaction are presented in Table 7.7, column 2.       154  Table 7.7 Quantitative results of SPF hydrogenolysis using Pt/C catalyst1   Pt/C (C)  Pt/C short (D)  Carbohydrate Derivatives Lignin Derivatives S (%)  Carbohydrate Derivatives Lignin Derivatives S (%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%)  Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Ya2 (wt%) Yo2 (wt%) Yt2 (wt%) Paraffins - 1.0 1.0 - 0.6 0.6 9.6  - - - - 0.1 0.1 0.4 Naphthenes - 1.4 1.4 - 2.8 2.8 24.2  - 0.3 0.3 - 0.3 0.3 2.9 Aromatic - - - - 0.2 0.2 1.0  - - - - - - 0.0 Furanics/pyranics 0.1 0.8 0.8 - - - 4.9  0.2 0.1 0.4 - - - 1.6 Oxygenated aliphatics 4.3 0.3 4.6 - 0.1 0.1 28.0  10.0 - 10.0 - - - 46.8 Non-methoxylated phenols - - - - 0.9 1.0 5.7  - - - - - - - Methoxyphenols - - - - 0.8 0.8 4.8  - - - 0.3 0.8 1.1 5.2 Organic acids 3.7 - 3.7 - - - 21.9  9.0 0.2 9.2 - - - 43.1 Total yield each phase (wt%)2 11.6 5.4 100.0  19.9 1.5 100.0 Total yield of all phases (wt%)2 17.0  21.4 Solid residue (wt%)3 2.6  8.5 1 Unless otherwise mentioned, the reactions were conducted with 5.0 g SPF sawdust, 10.0 g water, and 1 g Pt/C at 300°C. Reaction (C) was conducted with 10.0 g glycerol for 7 h. Reaction (D) was conducted with 5.0 g glycerol for 2.5 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa. 2 Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water- and extractive-free SPF (Ya: yield in aqueous phase; Yo: yield in washed catalyst solution (acetone); Yo: yield in organic (n-decane) phase; Yt: total yield in all phases). 3 Solid residue was calculated from the weight difference of the catalyst after being washed with acetone and dried in a vacuum oven at 50°C overnight and was calculated based on water- and extractive-free SPF. 155  To ascertain the carbohydrates degradation during SPF hydrogenolysis, sugar analysis of the aqueous phase was conducted for all the SPF hydrogenolysis reactions. Compositional analysis was also conducted using the aqueous phase of the SPF hydrogenolysis products to detect oligomeric sugars. The compositional analysis was conducted by mixing the aqueous phase of hydrogenolysis with sulfuric acid and heating it to 121°C, according to the NREL procedure.199 Table 7.8 shows the results for monomeric sugars in the aqueous phase before and after the compositional analysis. Only glucose and arabinose could be detected from the solution, in very low yields.  After the compositional analysis, sugars were found only in the aqueous solution of the reaction after 2.5 h (reaction D), which might indicate incomplete degradation of cellulose with a shorter reaction time and low glycerol loading. In most of the reactions, the yield of monomeric sugars decreased after compositional analysis due to the degradation of monomeric sugars during that analysis. Table 7.8 Sugar yields from the hydrogenolysis of SPF, before and after compositional analysis   Pt/C + H-ZSM-51 (A)   Pt/C + H-ZSM-5 (B)   Pt/C (C)   Pt/C2 (D) Before CA (wt%) After CA (wt%)  Before CA (wt%) After CA (wt%)  Before CA (wt%) After CA (wt%)  Before CA (wt%) After CA3 (wt%) Glucose - -  0.24 -  - -  0.07 0.06 Arabinose - -   - -   - -   0.30 0.26 Unless otherwise mentioned, the hydrogenolysis reactions were conducted with 5.0 g SPF, 5.0 g n-decane, 10.0 g water, and 10.0 g glycerol at 300°C for 7 h. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa. “Pt/C + H-ZSM-5” indicates that 1.00 g Pt/C and 0.25 g H-ZSM-5 were used. “Pt/C” indicates that only 1.00 g Pt/C was used. “Before CA” indicates the monomeric sugar yield before compositional analysis. “After CA” indicates the monomeric sugar yield after compositional analysis. 1 Reaction was conducted with 5.0 g SPF, 20.0 g water, and 10.0 g glycerol at 300°C for 5 h. 2 Reaction was conducted with 5.0 g glycerol for 2.5 h; other variables were kept the same. 3 The value was determined after adjustment using sugar recovery standard.   156    Figure 7.5 Pressure profile of APR reaction using 5.00 g SPF without Pt/C (a), 3.00 g Avicel® cellulose (b), and 6.00 g Avicel® (c). Reaction conditions: 5.0 g SPF, 0.25 g H-ZSM-5, 10.0 g glycerol, and 20.0 g water for 5 h (a); 3.0 g Avicel®, 1.0 g Pt/C, 0.25 g H-ZSM-5, and 20.0 g water for 5 h (b); 6.0 g Avicel®, 1.0 g Pt/C, 0.25 g H-ZSM-5, and 20.0 g water for 0.5 h (c).  A cellulose APR reaction was conducted to further understand the possibility of cellulose conversion to gaseous products. Two sets of reactions were conducted with different cellulose loadings and reaction times. The first was conducted with 3.0 g Avicel® cellulose for 5 h, the second with 6.0 g Avicel® cellulose for 0.5 h. Glycerol was not added in either reaction. Fig. 7.5 shows the pressure profiles for the cellulose APR and SPF reactions without Pt/C. The pressure profile of the SPF reaction without Pt/C can be considered a baseline when the APR reaction does not occur. It shows that the pressure of the APR reaction with 3.0 g cellulose was higher than the reaction of SPF without Pt/C (no APR reaction). The pressure was even higher when the reaction was conducted with 6.0 g cellulose.  The product yields and gas composition of the reactions with cellulose are presented in Table 7.9; a chromatograph can be found in Fig D.4. The reaction with 3.0 g cellulose for 5 h resulted in 157  mostly gaseous products and very low yields of organic acids (1.5 wt%) and oxygenated aliphatics (1.5 wt%). The reaction with 6.0 g cellulose for 0.5 h resulted in several organic acids (22.5 wt%), furanics (2.9 wt%), and oxygenated aliphatics (7.2 wt%). The gas analysis shows the high selectivity of CO2 in both reactions, along with several hydrocarbons ranging from C1 to C6 (compounds with 1 to 6 carbons). Sugars were not found in the aqueous products of cellulose APR before or after compositional analysis. Table 7.9 Liquid and gas products from cellulose APR reaction   3.0 g, 5 ha   6.0 g, 0.5 hb Carbohydrate Derivatives S (%)  Carbohydrate Derivatives S (%) Yac (wt%) Ywc (wt%) Ytc (wt%)  Yac (wt%) Ywc (wt%) Ytc (wt%) Furanics/pyranics - - - -  1.4 1.4 2.9 8.9 Oxygenated aliphatics - 1.5 1.5 49.4  3.6 3.6 7.2 22.1 Organic acids 0.8 0.7 1.5 50.6  11.2 11.2 22.5 69.0 Total yield (wt%)c 3.0 100  32.6 100.0 Char (wt%)d 0.0  0.0  S (%)e  S (%)e CO -  - CO2 79.6  88.3 C1 13.1  5.2 C2 6.5  3.4 C3 0.7  1.9 C4 0.1  0.9 C5 0.1  0.2 C6 -  0.1 Reaction conditions: a 3.0 g Avicel®, 1.0 g Pt/C, 0.25 g H-ZSM-5, and 20.0 g water for 5 h ;b 6.0 g Avicel®, 1.0 g Pt/C, 0.25 g H-ZSM-5, and 20.0 g water for 0.5 h. c Only identified compounds with a quality match of 80% or above were taken into account, and yield was calculated based on water- and extractive-free SPF (Ya: yield in aqueous phase; Yo: yield in washed catalyst solution (acetone); Yo: yield in organic (n-decane) phase; Yt: total yield in all phases). d Solid residue was calculated from the weight difference of the catalyst after being washed with acetone and dried in a vacuum oven at 50°C overnight and was calculated based on water- and extractive-free SPF. e Selectivity was calculated based on the mol ratio of gaseous products.   158  7.3 Discussion  7.3.1 FPO and SPF composition          Figure 7.6 Composition of BTG FPO (a) and SPF (b) in weight percentages.  BTG FPO composition is presented in Fig. 7.6(a). Lignin-derived monomers (methoxyphenols) contributed only 2.3 wt%, while 14.8 wt% was polymeric lignin. Identified carbohydrate derivatives consisted of organic acids (10.8 wt%), oxygenated cycloaliphatics (2.3 wt%), oxygenated aliphatics (8.4 wt%), and furanics (3.5 wt%). Water content in the BTG FPO was 29.0 wt%. This composition is similar to the FPO content from red oak, which contains 26.9 wt% water, 4.4 wt% monomeric phenols, 16.1 wt% pyrolytic lignin, 2.4 wt% furanics, 7.0 wt% organic acids (by IC, ionic chromatography), 2.4 wt% cyclics, 16.7 wt% low-molecular-weight (LMW) compounds (derived from carbohydrates), and 15.2 wt% sugars (by HPLC), with around 8.9 wt% unidentified compounds.200 The unidentified constituents (28.9 wt%) in Fig. 7.6(a) are possibly sugar compounds, which require HPLC to be detected.200 In a past study of red oak FPO, the sugars were dominated by levoglucosan.200   (a) (b) 159  The SPF composition is presented in Fig. 7.6(b) Two types of lignin were present: acid-soluble (3.2 wt%) and acid-insoluble (31.9 wt%). The former was believed to be composed of LMW degradation products and hydrophilic derivatives of lignin.201 The carbohydrate content of SPF was 53.7 wt%, with a cellulose content of no more than 35.8 wt% as glucan may also derive from hemicellulose.  7.3.2 Carbohydrate degradation  BTG FPO quantification showed that acetic acid and acetol were two major compounds. Similarly, organic acids and oxygenated aliphatics were major compounds in cellulose APR after 0.5 h (Table 7.9, column 2), which indicates that the acetic acid and acetol might come from the degradation of carbohydrates. Several furanics also existed in the FPO, which may also have come from the degradation of carbohydrate compounds. The different reaction conditions for cellulose APR and lignocellulose FPO, however, might have resulted in differences in the specific organic acids, oxygenated aliphatics, or furanics/pyranics. Several oxygenated cycloaliphatics were also found, with hydroxy cyclopentenones being predominant, which might have come from the further reaction of furanics.  The reaction with 0.5 g glycerol after 2.5 h (Table 7.3, column 1) resulted in a decrease in carbohydrate-derived compounds from 35.2 to 7.8 wt%. Oxygenated aliphatics and organic acids underwent the most significant decreases. A further increase in the reaction time (to 7 h) and glycerol loading (to 10 g; see Table 7.3, column 2) resulted in a slight increase in carbohydrate derivatives (11.0 wt%). Although the yield of carbohydrate derivatives increased, the yield of organic acids continued to decrease. The increase in the former arose from the increase in 160  oxygenated aliphatics, which might have been due to the higher glycerol loading. The further addition of H-ZSM-5 decreased the yield of carbohydrate derivatives. The yield of oxygenated aliphatics decreased considerably from 5.2 to 0.5 wt%, and pentanone and pentanols could no longer be found. In addition, furanics and oxygenated cycloaliphatics could not be found after upgrading under these conditions. The yields of paraffins and organic acids from carbohydrates decreased from 1.9 to 0.7 wt% with the presence of H-ZSM-5. The yield of carbohydrate-derived compounds further decreased to 2.6 wt% when the catalyst loading was doubled.  Similar phenomena occurred during the hydrogenolysis of SPF. After compositional analysis, sugars were detected only in the reaction conducted for 2.5 h (reaction D), as shown in Table 7.8. This result indicated less carbohydrate degradation occurred in reaction D compared to the other SPF hydrogenolysis reactions. In the presence of n-decane, the yield of carbohydrate derivatives decreased from 19.9 to 11.6 wt% after the reaction time was extended from 2.5 to 7 h, regardless of the glycerol loading. Further addition of H-ZSM-5 decreased the carbohydrate derivatives to 9.0 wt%. The solid residue also decreased from 8.5 to 2.6 wt% when the reaction was extended from 2.5 to 7 h and further decreased to 0 wt% with the addition of H-ZSM-5. It seems that in both FPO and SPF, more carbohydrates were degraded with longer reaction times and the presence of H-ZSM-5. Given that carbohydrates can undergo APR, the presence of an acid catalyst apparently increased the APR reaction, particularly the WGS step through the dissociation of water.171  FPO carbohydrates presented as degradation products, including oxygenated aliphatics/cycloaliphatics, furanics/pyranics, organic acids, and pyrolytic lignin, while SPF carbohydrates were in the native forms of cellulose and hemicellulose. Hence, under similar reaction conditions, more carbohydrate-derived compounds can be preserved from SPF, since the depolymerization of carbohydrate polymers needs to occur before the APR reaction. APR of 161  Avicel® cellulose and compositional analysis of the liquid products showed that the depolymerization of carbohydrates occurs very rapidly. In this case, the absence of lignin might result in rapid cellulose APR, since lignin, which is hydrophobic, limits water access. Char was not noticeable after the APR reaction of Avicel®. Given this observation, the solid residue from SPF hydrogenolysis might have been unreacted lignin fragments. Since the carbohydrates were not likely to produce char, and humin is not present in FPO, it seems that the char formation during FPO upgrading might have come from the condensation of lignin fractions. Although the reforming of carbohydrates under these reaction conditions was not the desired outcome, this result presents the opportunity for converting carbohydrates to non-sugar compounds through in-situ reforming by optimizing the reaction conditions. It can be seen that shortening the reaction time resulted in a higher yield of liquid monomers (Table 7.9).   7.3.3 The effect of a co-solvent  Although limited, the HDO reaction seemed to occur in the BTG FPO even without a co-solvent, as indicated by the presence of aromatics, naphthalenes, naphthenes, and paraffins in the products. However, these compounds seem to remain on the catalyst surface, and washing with acetone was required to recover the compounds due to their insolubility in water. BTG FPO contains 29.0 wt% water as a single-phase liquid. It seems that the high concentration of oxygenated compounds, such as organic acids (acetic acid) and oxygenated aliphatics (acetol), improved the solubility of less-polar compounds (phenolics) in water. However, when these oxygenated compounds were further degraded to less-polar compounds, possibly through the APR reaction, this may have caused phase separation between the less-polar compounds and water. In addition, the HDO products of 162  methoxyphenols such as aromatics and naphthalenes are even less polar and consequently less soluble in water.  Monomeric yield slightly decreased from 3.3 wt% in raw BTG FPO to 2.9 wt% after the reaction with Pt/C and H-ZSM-5 for 7 h without n-decane as the co-solvent. Although oxygen removal might have contributed to the lower monomeric yield (since the yield was calculated based on wt%), monomeric condensation might have occurred, indicated by the formation of naphthalenes (0.5 wt%). The amount of naphthalenes in this reaction was the highest among all of the FPO upgrading and SPF hydrogenolysis reactions. The presence of n-decane seems to have isolated the nonpolar products of FPO upgrading and prevented condensation.  In SPF hydrogenolysis, the effect of a co-solvent was less significant than the effect of longer reaction time and higher glycerol loading. Comparison of reaction A, 10.0 g glycerol for 5 h without n-decane (Table 7.6, column 1) and reaction D, 5.0 g glycerol for 2.5 h with n-decane (Table 7.7, column 2) shows that higher lignin monomers were recovered from reaction A (2.2 wt%) than reaction D (1.5 wt%), with less solid residue. The different form of carbohydrates in BTG FPO and SPF might be the reason. Organic acids and furanics/pyranics were found largely in raw BTG FPO (Table 7.1), while in SPF, the carbohydrates were initially in native polymer form. The polymeric carbohydrates likely have less acidity than carbohydrate degradation products such as organic acids therefore separation of lignin fragments during SPF upgrading is less essential than during FPO upgrading. However, since carbohydrates are in polymer form and limit lignin accessibility, a longer reaction time is required to break both carbohydrate and lignin polymers. Another speculation is that the SPF lignin polymer required a longer reaction time and a hydrogen donor to break into smaller polymers or monomers compared to the FPO lignin 163  polymer. Since fewer monomers were initially available, condensation between monomers would be less.  7.3.4 Lignin depolymerization Although depolymerization of ether dimeric lignin was possible via glycerol APR, the presence of carbohydrate-derived compounds in FPO may have promoted condensation of the less stable degradation fragments, which was indicated by the low lignin monomer yield and relatively high char yield (Table 7.2). It was found that the polymerization of lignin-derived oligomers occurred during thermal treatment at 250–400 °C and was enhanced by the presence of carbohydrate-derived species.153 The two-phase reaction seemed to isolate the nonpolar products of FPO upgrading from other FPO compounds, such as organic acids, resulting in higher lignin monomeric products. This was confirmed by the observation that most of the nonpolar compounds were found in the organic phase. The depolymerization of native lignin seems to have occurred to some extent without the presence of a co-solvent, as 2.2 wt% lignin-derived monomers could be recovered from SPF hydrogenolysis without a co-solvent.  The evolution of lignin and carbohydrate derivatives during FPO upgrading and SPF hydrogenolysis are presented in Table 7.10. Polymeric lignins are lignin polymers small enough to be dissolved in either water, n-decane, or acetone but still too big to be detected by GC-MS. Char or solid residue (SR) are bigger lignin polymers that are not soluble in water, n-decane, or acetone and are therefore noticeable as SR after catalyst filtration and drying. Lignin monomers are lignin-derived compounds that can be detected by GC-MS. The SR in SPF was accounted from the total lignin in SPF based on water and extractive-free weight. 164  In the presence of a co-solvent, lignin derivatives increased with longer reaction time and higher glycerol loading. In FPO upgrading, lignin-derived monomers increased from 4.6 to 9.3 wt% when the reaction was extended from 2.5 h with 5.0 g glycerol to 7 h with 10.0 g glycerol. Similarly, lignin derivatives increased from 1.5 wt% in the 2.5 h reaction with 5.0 g glycerol to 5.4 wt% in the 7 h reaction with 10.0 g glycerol during SPF hydrogenolysis. This condition (10.0 g glycerol for 7 h without H-ZSM-5) resulted in the highest lignin monomer yield from either FPO upgrading or SPF hydrogenolysis. Omitting water content, the total lignin content in FPO was 24.2 wt%, which means 38.3 wt% lignin monomers were recovered after FPO upgrading. This result is comparable to that achieved by HDO of water insoluble bio-oil (WIBO), which produced 29.6 wt% liquid alkane after reaction with 1MPa H2 at 250 °C in decaline for 15 h.142 The total lignin content in dry, extractive-free SPF was 39.0 wt%, which means that 14.0 wt% lignin monomers were recovered from SPF. This result was comparable to the 24 h hydrogenolysis of Pinus radiata with Pd/C in dioxane/water and 3.4 MPa H2 which yielded 1.6 wt% monomers and 26 wt% dimers.202 The higher monomeric recovery from FPO upgrading compared to SPF hydrogenolysis may be due to the larger lignin polymers in SPF, which require longer reaction time and a greater amount of hydrogen to be depolymerized. Although C-C coupling may be present in FPO, very large condensed lignin molecules would have formed as char during the fast pyrolysis process. Gas analysis and the higher CO2/CH4 ratio (Table D.2 and D.3) showed that under similar conditions, SPF hydrogenolysis produced more CO2 compared with FPO upgrading. The addition of H-ZSM-5 did not increase the yield of lignin-derived monomers in either FPO upgrading or SPF hydrogenolysis. In FPO upgrading, the addition of H-ZSM-5 seemed to increase 165  the amount of char, while in SPF hydrogenolysis, the SR decreased with the addition of H-ZSM-5, although the amount of lignin monomers did not increase. Table 7.10  Evolution of carbohydrate and lignin constituents during FPO upgrading and SPF hydrogenolysis1     Condition Lignin (wt%) Carbohydrates (wt%) Poly2 Mono Char/ SR3 Poly Mono FPO Upgrading4 BTG FPO (control) 20.9 3.3 - N/A 35.2 Pt/C (1.0 g) + Glycerol (5.0 g), 2.5 h (I) 16.9 4.6 2.7 N/A 7.8 Pt/C (1.0 g) + Glycerol (10.0 g), 7 h (II) 14.0 9.3 0.9 N/A 11.0 Pt/C (1.0 g) + H-ZSM-5 (0.25 g), 7 h (III) 12.7 9.0 2.6 N/A 4.4 Pt/C (2.0 g) + H-ZSM-5 (0.50 g), 7 h (IV) 10.6 9.0 4.7 N/A 2.6        SPF Hydrogenolysis5 SPF (control) - - 39.0 59.5 - Pt/C (1.0 g) + Glycerol (5.0 g), 2.5 h (D) 29.0 1.5 8.5 0.3 20.3 Pt/C (1.0 g) + Glycerol (10.0 g), 7 h (C) 30.9 5.4 2.6 - 11.6 Pt/C (1.0 g) + H-ZSM-5 (0.25 g), 7 h (III) 33.8 5.1 - - 9.3 1 Yield was calculated based on water- and extractive-free FPO and SPF. Poly refers to polymers and mono refers to monomers. 2 Calculated based on total lignin polymers and monomers in control minus lignin monomers and char/SR after the reaction. 3 SR refers to solid residue. 4 Unless otherwise stated, the reaction was conducted with 10.0 g BTG FPO, 10.0 g glycerol, 5.0 g n-decane, and 1.0 g Pt/C at 300°C. 5 Unless otherwise stated, the reaction was conducted with 5.0 g SPF, 10.0 g water, 10.0 g glycerol, 5.0 g n-decane, and 1.0 g Pt/C at 300°C. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.  7.3.5 Lignin monomers in SPF and BTG FPO Hydrogenolysis of SPF results in the deconstruction of the lignocellulose structure, with most of the biomass composition becoming liquefied. In the hydrogenolysis with a short reaction time (2.5 166  h), 4-propyl, 2-methoxyphenol (cerulignol) was found to be a major phenolic compound. This condition resulted in the highest yield of methoxyphenols among other SPF hydrogenolysis and FPO upgrading reactions, but it was lower than from raw BTG FPO. The high yield of methoxyphenols, dominated by cerulignol, indicates that cerulignol is the major lignin depolymerization product from SPF hydrogenolysis.   On the other hand, in raw BTG FPO, p-creosol is the major lignin-derived monomer from lignocellulose fast pyrolysis; p-creosol has also been found to be the major product during the thermolysis of coniferyl alcohol.203 The difference between FPO resulting in p-creosol and hydrogenolysis resulting in cerulignol might be due to the much higher reaction temperature used with FPO (~550 °C) compared to hydrogenolysis (300°C).  7.3.6 HDO of lignin-derived compounds Since p-creosol was the major monomer detected in BTG FPO, HDO of p-creosol is here considered representative of phenolic monomers in BTG FPO. It was shown in Table 5.4 that hydrogenation of p-creosol will produce p-cresol (4-methylphenol), 4-methylcyclohexanol, and toluene. Since Pt/C was used as a catalyst instead of Raney Ni®, methylcyclohexane production via a similar path as cyclohexane production from phenol is possible (Table 6.2). Fig. 7.7 shows the yield of p-creosol, p-cresol, 4-methylcyclohexanol, toluene, and methylcyclohexane during HDO of BTG FPO with Pt/C and H-ZSM-5 and with Pt/C only.    167    Figure 7.7 Yield of p-creosol, p-cresol, 4-methylcyclohexanol, toluene, and methylcyclohexane during HDO of BTG FPO. The reaction was conducted with 10.0 g BTG FPO and 5.0 g n-decane at 300°C. The amounts of glycerol and catalyst and the reaction times are stated in the graph. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.    Demethoxylation of phenolic compounds seemed to have been the major reaction in the earlier upgrading reaction, as can be seen in reaction (I), which used 5.0 glycerol for 2.5 h, as shown in Fig. 7.7; the p-creosol yield decreased while 4-methylphenol had the highest yield. In general, the shifting of phenolic compounds from methoxyphenol to non-methoxylated phenols can be seen by comparing Table 7.1 and Table 7.3, column 1. Methoxyphenol (Table 7.3, column 1) as well as 4-methylphenol (Fig. 7.7) can still be found during this reaction, although with very low yields. Benzenes and cyclohexanes were also found as products with low selectivity. It was shown in Chapter 6 that the IGAPH reaction with Pt/C can further proceed to the IGAPHdo reaction without acid.  168  The hydrogenation reaction of 4-methylphenol proceeded further with the higher glycerol loading (10 g), producing 4-methylcyclohexanol. Further demethoxylation seemed to occur with a longer reaction time (7 h, reaction II), and the p-creosol yield further decreased. 4-Methylphenol started to decrease. At the same time, the HDO reaction progressed further to produce a higher yield of methylcyclohexane and toluene, since the reaction was conducted at a relatively high temperature. In general, a similar trend was found for methoxyphenols, non-methoxylated phenols, naphthenes, and aromatics, as can be seen in Table 7.3, column 2.  The HDO reaction of p-creosol compounds increased with the addition of H-ZSM-5, which was shown by the increases in methylcyclohexane and toluene. Under these conditions, p-creosol and 4-methylcyclohexanol could no longer be found. The fast HDO reaction of 4-methylcyclohexanol to produce either methylcyclohexane or toluene in the presence of H-ZSM-5 might have been the reason for the absence of 4-methylcyclohexanol under these reaction conditions. In general, the yield of paraffins, naphthenes, aromatics, and naphthalenes significantly increased with the addition of H-ZSM-5 (Table 7.4, column 1). Conversely, the yield of non-methoxylated phenols, oxygenated aliphatics, and oxygenated cycloaliphatics considerably decreased with the presence of H-ZSM-5.  The further addition of both Pt/C and H-ZSM-5 decreased the yields of methylcyclohexane and toluene. In general, doubling the catalyst loading did not change the yield of lignin-derived compounds but did change the selectivity of the products. The yields of naphthenes and naphthalenes further increased while those of paraffins and non-methoxylated phenols decreased. These results are consistent with those from the reaction with model compounds (see Chapter 6), where the addition of the acid catalyst resulted in more HDO reaction. Interestingly, the yield of char increased with the presence of H-ZSM-5, from 1 wt% without H-ZSM-5 to 3 and 5 wt% with 169  0.25 and 0.5 g H-ZSM-5, respectively. It seems that the presence of H-ZSM-5 promoted a condensation reaction; this suggestion is supported by the increased yield of naphthalenes.  Figure 7.8 Yields of cerulignol, 4-propylphenol, propylbenzene, and propylcyclohexane during SPF hydrogenolysis. The reaction was conducted with 5.0 g SPF sawdust, 10.0 g water, and 5.0 g n-decane at 300°C. The amounts of glycerol and catalyst and the reaction times are stated in the graph. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.  It has been discussed that SPF hydrogenolysis with a shorter reaction time (2.5 h), lower glycerol loading (5.0 g), and no H-ZSM-5 (reaction D) resulted in the lowest recovery of lignin-derived monomers of all the reactions. However, the yield of methoxyphenol was the highest of all the reactions, and cerulignol constituted the highest proportion of methoxyphenol. This result might have indicated an early process in the depolymerization of lignin, so cerulignol was taken as a representative compound to observe the HDO reaction during SPF hydrogenolysis. Fig. 7.8 shows the yields of cerulignol, 4-propyl phenol, propylbenzene, and propylcyclohexane.  170  A longer reaction time (7 h) with a higher glycerol loading (10.0 g) (reaction C) resulted in higher yields of cerulignol, 4-propylphenol, propylbenzene, and propylcyclohexane. However, the overall yield of methoxyphenol decreased, due to the decrease in p-cresol and other methoxyphenols (Fig D.7). This result indicates a higher degree of lignin depolymerization, as can be seen in Table 7.10 where higher monomeric yields were obtained. Although the polymeric lignin amount appeared to be higher than from reaction (D), which used 5.0 g glycerol for 2.5 h, the amount of SR was lower, indicating more depolymerization. No 4-propylcyclohexanol was found in the products, yielding a product distribution similar to that of eugenol hydrogenation (Table 5.4). In the hydrogenation of eugenol, 4-propylcyclohexanol was not found in the products but propylcyclohexene and propylbenzene were. There are two possible reasons for the absence of 4-propylcyclohexanol. The first is that unlike in other phenolic monomers, such as p-cresol, vanillin, phenol, or guaiacol,  deoxygenation of cerulignol occurs before ring saturation. The second is that a similar reaction occurs, starting with saturation of the aromatics rings and followed by deoxygenation of hydroxyl groups. In this case, dehydrogenation of 4-propylcylohexanol must have proceeded very fast under the SPF hydrogenolysis conditions, making it impossible to observe 4-propylcyclohexanol. The yield of 4-propylphenol in this reaction was higher than the yield of cerulignol. This result confirmed the demethoxylation of cerulignol to be the first reaction step in lignin monomers, as shown in Table 5.4, where 4-propylphenol was the second most predominant product. In general, a relatively high yield of naphthenes was found in this reaction (Table 7.7, column 1).   Unlike in FPO upgrading, where p-cresol, the major monomeric lignin in FPO, could not be found after upgrading in the presence of H-ZSM-5 (Fig. 7.8), cerulignol, the major phenolic compound in SPF, could still be found, even after a long reaction time in the presence of H-ZSM-5 (reaction B, Fig. 7.8). Moreover, the yield of cerulignol was relatively consistent compared to the yields of 171  4-propyl phenol, propylbenzene, and propylcyclohexane. The HDO of cerulignol seems to be more difficult than that of other phenolics compounds, where it has been shown that the upgrading of eugenol mostly produces cerulignol (Table 5.4). Although the yield of monomeric lignin generally remained similar to the yield in reaction C (Table 7.10), in reaction B, where H-ZSM-5 was added, the product distribution changed slightly, with naphthalene starting to appear and the naphthene yield decreasing. This result indicates that H-ZSM-5 might have promoted a condensation reaction, which was also noticeable in FPO upgrading.    7.3.7 O/C and H/C ratios in the products O/C and H/C ratios are the simplest indicators of hydrogenation and deoxygenation. They are also some of the key parameters for the utilization of upgraded FPO; the lower the O/C and the higher the H/C of upgraded FPO, the closer the product is to conventional fuels. Although a very low oxygen content does not guarantee a high octane number, it does guarantee a higher heating value.204 The upgrading of BTG FPO can result in low O/C and high H/C ratios (e.g. paraffins and naphthenes), low O/C and low H/C ratios (e.g. aromatics and naphthalenes), and high O/C and high H/C ratios (e.g. alcohols). The Van Krevelen diagram in Fig. 7.9 presents the O/C and H/C ratios of raw BTG FPO and upgraded FPO. One of the advantages of this process is that the reaction occurs in two phases, with the less polar compounds tending to present in the organic (n-decane) phase while the more polar compounds exist in the aqueous (water) phase.  The red diamonds in Fig. 7.9 show the O/C and H/C ratios for the combined phases (organic and aqueous) of upgraded FPO. The reactions with the co-solvent show that the longer the reaction time and the higher the catalyst loading, the lower the O/C ratio. For reactions with the same 172  glycerol loading (II, III, and IV), greater H-ZSM-5 loading resulted in lower O/C and H/C ratios. The low glycerol loading in reaction (I) might have been the reason for the low H/C ratio. All of the reactions with the co-solvent resulted in lower O/C and higher H/C ratios compared to the initial BTG FPO, indicating the occurrence of HDO. In the reaction without the co-solvent (red star), the O/C and H/C ratios were both lower than in the initial BTG FPO, indicating the ineffectiveness of the hydrogenation reaction. The lower O/C ratio in this reaction was mostly the result of the demethoxylation of methoxyphenol and the degradation of both organic acids and oxygenated aliphatics, indicating a low hydrogenation reaction.  Figure 7.9 Van Krevelen diagram of upgraded BTG FPO. Green circles refer to the organic phase, blue squares refer to the aqueous phase, red diamonds refer to the combination of both phases, and the black triangle refers to the raw BTG FPO. The stars refer to the reactions without co-solvents.  The green circles in Fig. 7.9 shows the composition of the products in the organic phase of upgraded FPO. Most of the compounds in the organic phase, especially from reactions II, III, and 173  IV, have O/C and H/C ratios close to those in gasoline. A similar trend to the combined phase was found in the organic phase, where higher H-ZSM-5 loading resulted in a lower O/C ratio and a slightly lower H/C ratio. In the reaction without the co-solvent, although the compounds in the catalyst wash (green star) contained a considerably lower O/C ratio, the H/C ratio was also low. The aqueous composition (blue squares) shows that all of the products had a higher H/C ratio than raw BTG FPO. The O/C trend in the aqueous phase was different from the O/C trend in the combined phases or the organic phase, where the reaction with H-ZSM-5 tended to have a higher O/C in the aqueous phase. Figure 7.10 shows the Van Krevelen diagram of SPF hydrogenolysis. Overall, the HDO reaction of SPF seems to have occurred because the H/C ratio of the combined phase was higher and the O/C ratio was lower than in common biomass. It seems that in the case of SPF hydrogenolysis, longer reaction time and higher glycerol loading are more important than the presence of the co-solvent. It can be seen that the overall O/C ratio of the reaction without the co-solvent with 10.0 g glycerol for 5 h (red star A) is lower than the overall O/C ratio of the reaction with the co-solvent and 5.0 g glycerol for 2.5 h (red diamond D). The presence of n-decane as the co-solvent in the reaction with 10.0 g glycerol for 7 h (reaction C) resulted in compounds with high H/C and low O/C ratios. Moreover, the organic phase of these products was in the gasoline range. Although the organic phase of the reaction with H-ZSM-5 was also in the gasoline range, the presence of H-ZSM-5 during SPF hydrogenolysis (reaction B) did not improve HDO performance. As can be seen in Fig. 7.10, reaction B (with H-ZSM-5) had a slightly higher O/C ratio and lower H/C ratio compared to reaction C (without H-ZSM-5) in the organic phase (green), aqueous phase (blue), and combined phases (red). In addition, Fig. 7.8 shows that the yield of deoxygenated compounds in reaction B was also lower than in reaction C. 174  .  Figure 7.10 Van Krevelen diagram of SPF hydrogenolysis. Green circles refer to the organic phase, blue squares refer to the aqueous phase, and red diamonds refer to the combination of both phases. The stars refer to the reactions without co-solvents.   7.4 Conclusion BTG FPO was found to consist of 17.2 wt% lignin derivatives, 29.0 wt% water, 25.0 wt% carbohydrates derivatives, and 28.9 wt% unidentified compounds. Of the lignin derivatives, 14.8 wt% was polymeric and only 2.3 wt% was monomeric. SPF consisted of 35.1 wt% lignin and 53.7 wt% carbohydrate. During FPO upgrading and SPF hydrogenolysis, higher carbohydrate degradation was found with a longer reaction time and the presence of H-ZSM-5. Indication of carbohydrates and carbohydrate derivatives reforming was found during SPF hydrogenolysis and further confirmed by APR of Avicel® cellulose. No sign of char formation from carbohydrates was found.  175  The addition of n-decane as a co-solvent seemed to prevent lignin condensation during FPO upgrading by separating polar carbohydrate derivatives from less polar lignin derivatives. In the presence of n-decane as a co-solvent, the higher glycerol loading and the longer reaction time resulted in higher lignin depolymerization. The presence of H-ZSM-5 did not increase the depolymerization of lignin. Lignin condensation, especially during FPO upgrading, seemed to increase with the presence of H-ZSM-5. Overall, as much as 38.3 wt% monomeric lignin was recovered from whole FPO lignin and 14.0 wt% from whole SPF lignin. The primary lignin monomer in BTG FPO was p-creosol and in SPF hydrogenolysis was cerulignol.  In FPO upgrading, p-creosol, the main monomer, underwent demethoxylation to produce p-cresol, followed by hydrogenation to produce methyl cyclohexanol, and further deoxygenation and hydrogenation/dehydrogenation to produce methylcyclohexane and toluene. In SPF hydrogenation, where cerulignol was found to be the primary monomer, 4-propyl-cyclohexanol could not be observed. However, other HDO and intermediate products, such as propylcyclohexane, propylbenzene, and 4-propylphenol, were found. HDO of p-creosol and cerulignol occurred in similar steps to the reaction with model compounds. In the case of FPO upgrading, the presence of H-ZSM-5 increased the HDO reaction, with the O/C ratio of the product decreasing considerably. However, the effect of H-ZSM-5 was not noticeable in SPF hydrogenolysis. The use renewable hydrogen from glycerol APR and products in the range of gasoline, make BTG FPO upgrading and SPF hydrogenolysis with in situ glycerol APR a promising strategy for utilizing lignin fractions derived from FPO and lignocellulose. However, the monomeric lignin yield still needs to be improved.  176  Chapter 8: Conclusions and Recommendations    8.1 Conclusions This thesis describes the study of a novel strategy for ligneous material upgrading using glycerol APR to provide hydrogen. This study was first conducted with phenolics as lignin model compounds and examined each reaction starting from glycerol APR, in-situ glycerol APR phenol hydrogenation, in-situ glycerol APR phenol hydrodeoxygenation, and finally upgrading of ligneous material via in-situ glycerol APR.   Low selectivity of H2 and CO2 and high selectivity of CH4 were observed during glycerol APR with Raney Ni® in the batch reactor. The mechanistic study of glycerol APR indicates that glycerol produces 1,2-propylene glycol and ethylene glycol which react in parallel to produce gaseous products. During the mechanistic study, it was also found that the cleavage of the C-C bond between oxygenated carbons is easier than the C-C bond between oxygenated and non-oxygenated carbon. In-situ phenol hydrogenation and glycerol aqueous phase reforming was conducted in a batch reactor at 180 – 240°C. The presence of phenol increased glycerol conversion. Hydrogen scavenging by phenol decreased direct methanation which increased glycerol APR selectivity. Phenol hydrogenation produced cyclohexanone and cyclohexanol as the major products, relative selectivity was affected by the availability of hydrogen from glycerol APR. In-situ hydrogenation 177  and glycerol aqueous phase reforming was applicable to other phenolic compounds and resulted in demethoxylation, hydrogenation, and hydrogenolysis of ether bonds.  IGAPH proceeds without the formation of gaseous hydrogen as indicated by the slightly increased conversion of phenol when IGAPH was performed with a continuous He flush (open system). The slight increase in phenol conversion was attributed to a decreased in the Sabatier reaction. Minimum reaction pressure in the open system should exceed the saturation pressure of methanol to avoid loss of this intermediate. Excessively high reaction pressure, especially in closed system, will also negatively affect phenol conversion. Further, the Langmuir-Hinshelwood mechanism was proposed as the reaction mechanism for IGAPH with Raney Ni®; phenol hydrogenation was the rate-limiting step. During this study, it also found that Pt/C was an active catalyst for IGAPH.  Hydrodeoxygenation (HDO) of phenol was performed with the addition of a solid acid catalyst; H-ZSM-5 and Amberlyst-15 were the most active catalysts. The phenol HDO mechanism progressed through hydrogenation to produce saturated compounds such as cyclohexanone and cyclohexanol, dehydration to produce cyclohexene and finally hydrogenation or dehydrogenation to produce benzene or cyclohexane. Although dehydration can occur with only the presence of water at high temperature, the presence of acid catalyst increased the dehydration reaction rate. The selectivity of cyclohexane and benzene could be controlled by changing phenol to glycerol loading ratio and the ratio of hydrogenation catalyst (Raney Ni® or Pt/C) to acid catalyst (H-ZSM-5 or Amberlyst-15). Increasing the phenol to glycerol ratio and lowering the hydrogenation catalyst to acid catalyst ratio resulted in high selectivity of benzene.  Polymeric lignin was found in both FPO and SPF thus depolymerization into lignin monomers followed by HDO of monomers to paraffins, naphthenes, and aromatics occurred during 178  upgrading. The addition of a nonpolar co-solvent (n-decane) isolated the HDO products from FPO and prevented condensation. Increasing glycerol loading and reaction time increased the degree of depolymerization. Once phenolic monomers were produced, the HDO mechanism was similar to that observed for the model compounds. Carbohydrate derivatives are suspected to contribute to the increase of condensation of lignin derived compounds. The condensation reaction also increased with the presence of H-ZSM-5. During FPO upgrading H-ZSM-5 decreased the O/C ratio and the H/C ratio. Carbohydrates may undergo reforming reactions.   8.2  Recommendations 8.2.1 Catalysts improvement The thermodynamic analysis shows that IGAPH reaction can be conducted at a lower temperature. However, the reaction temperature study in Chapter 5 shows that both phenol and glycerol conversion was low at 180°C, possibly due to the low reaction rate of glycerol APR. Modification of catalyst to lower the activation energy of glycerol APR will likely benefit the IGAPH reaction. Bimetallic catalysts, such as Pt-Ni or Ru-Ni, can separately improve glycerol APR and phenolic hydrogenation and thus should be investigated as IGAPH catalysts. In addition, incorporating an acid catalyst as a support for hydrogenation catalyst might benefit the process by improving the ease of catalyst recovery. Another advantage of incorporating a hydrogenation catalyst on the acidic support is the shorter distance between active catalysts sites which can potentially increase the reaction rate, since desorption of intermediate compounds would not be necessary, or even change the reaction mechanism.   179  8.2.2 Continuous process There are indications that the Sabatier reaction and hydrogenation of glycerol intermediates result in methane production. These reactions can be avoided if the reaction is conducted continuously to reduce the possibility of CO2 and H2 reacting. In the case of upgrading FPO or SPF, the continuous process (i.e CSTR type of reactor) will limit the possibility of condensation reactions and further degradation of FPO into gaseous compounds.   8.2.3 Renewable co-solvents During the upgrading of ligneous material, a co-solvent is required. This co-solvent will likely degrade due to the complexity of the reaction. The effectiveness of a renewable co-solvent with low polarity (e.g. fatty acid, carbohydrate hydrogenation products or phenolics HDO products) must be investigated.   180  Bibliography (1)  Le Quéré, C.; Moriarty, R.; Andrew, R. M.; Peters, G. P.; Ciais, P.; Friedlingstein, P.; Jones, S. D.; Sitch, S.; Tans, P. P.; Arneth, A. Global carbon budget 2014. Earth System Science Data 2015, 7 (1), 47–85. (2)  Socolow, R. H.; Pacala, S. W. A plan to keep carbon in check. Scientific American 2006, 295 (3), 50–57. (3)  Himmel, M. E.; Ding, S.-Y.; Johnson, D. K.; Adney, W. S.; Nimlos, M. R.; Brady, J. W.; Foust, T. D. Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 2007, 315 (5813), 804–807. (4)  Huber, G. W.; Vispute, T. P.; Routray, K. Production of hydrogen, liquid fuels, and chemicals from catalytic processing of bio-oils. US 8,741,258 B2, 2014. (5)  Amen-Chen, C.; Pakdel, H.; Roy, C. Separation of phenols from Eucalyptus wood tar. Biomass and Bioenergy 1997, 13 (1–2), 25–37. (6)  Zhang, X.-S.; Yang, G.-X.; Jiang, H.; Liu, W.-J.; Ding, H.-S. Mass production of chemicals from biomass-derived oil by directly atmospheric distillation coupled with co-pyrolysis. Scientific Reports 2013, 3, 1120. (7)  Fergus, B. J.; Procter, A. R.; Scott, J. A. N.; Goring, D. A. I. The distribution of lignin in sprucewood as determined by ultraviolet microscopy. Wood Science and Technology 1969, 3 (2), 117–138. (8)  Hon, D. N.-S.; Shiraishi, N. Wood and cellulosic chemistry, revised, and expanded; CRC Press, 2000. (9)  Alexey Vishtal; Kraslawski, A. Challenges in industrial applications of technical lignins. BioResources 2011, 6 (3), 3547–3568. (10)  Mullen, C. A.; Boateng, A. A. Chemical Composition of Bio-oils Produced by Fast Pyrolysis of Two Energy Crops. Energy & Fuels 2008, No. 22, 2104–2109. (11)  Bridgwater, A. V. Production of high grade fuels and chemicals from catalytic pyrolysis of biomass. Catalysis Today 1996, 29 (1), 285–295. (12)  Lan, R.; Irvine, J. T. S.; Tao, S. Ammonia and related chemicals as potential indirect hydrogen storage materials. International Journal of Hydrogen Energy 2012, 37 (2), 1482–1494. (13)  Davda, R. R.; Shabaker, J. W.; Huber, G. W.; Cortright, R. D.; Dumesic, J. A. A review of 181  catalytic issues and process conditions for renewable hydrogen and alkanes by aqueous-phase reforming of oxygenated hydrocarbons over supported metal catalysts. Applied Catalysis B: Environmental 2005, 56 (1-2 SPEC. ISS.), 171–186. (14)  Cortright, R. D.; Davda, R. R.; Dumesic, J. A. Hydrogen from catalytic reforming of biomass-derived hydrocarbons in liquid water. 2002, 418 (August), 54–57. (15)  Nomanbhay, S.; Hussein, R.; Ong, M. Y. Sustainability of biodiesel production in Malaysia by production of bio-oil from crude glycerol using microwave pyrolysis: A review. Green Chemistry Letters and Reviews 2018, 11 (2), 135–157. (16)  Quispe, C. A. G.; Coronado, C. J. R.; Carvalho, J. A. Glycerol : Production, consumption, prices, characterization and new trends in combustion. Renewable and Sustainable Energy Reviews 2013, 27 (November 2013), 475–493. (17)  Yuan, T. Q.; Sun, S. N.; Xu, F.; Sun, R. C. Characterization of lignin structures and lignin-carbohydrate complex (LCC) linkages by quantitative 13C and 2D HSQC NMR spectroscopy. Journal of Agricultural and Food Chemistry 2011, 59 (19), 10604–10614. (18)  Henriksson, G.; Li, J.; Zhang, L.; Lindström, M. E. Lignin utilization. In Thermochemical Conversion of Biomass to Liquid Fuels and Chemicals; RSC, 2010. (19)  Rubin, E. M.; Himmel, M. E.; Ding, S.; Johnson, D. K.; Adney, W. S. Biomass Recalcitrance : Science 2007, 454 (February), 804–807. (20)  Várnai, A.; Siika-aho, M.; Viikari, L. Restriction of the enzymatic hydrolysis of steam-pretreated spruce by lignin and hemicellulose. Enzyme and Microbial Technology 2010, 46 (3–4), 185–193. (21)  Wijaya, Y. P.; Dhimas Dhewangga Putra, R.; Widyaya, V. T.; Ha, J.-M.; Suh, D. J.; Kim, C. S. Comparative study on two-step concentrated acid hydrolysis for the extraction of sugars from lignocellulosic biomass. Bioresource Technology 2014, 164, 221–231. (22)  Boussaid, A.; Robinson, J.; Cai, Y. J.; Gregg, D. J.; Saddler, J. N. Fermentability of the hemicellulose-derived sugars from steam-exploded softwood (Douglas fir). Biotechnology and Bioengineering 1999, 64 (3), 284–289. (23)  Gupte, A.; Madamwar, D. Solid state fermentation of lignocellulosic waste for cellulase and β-glucosidase production by cocultivation of Aspergillus ellipticus and Aspergillus fumigatus. Biotechnology Progress 1997, 13 (2), 166–169. (24)  Vaca-Garcia, C. Biomaterials. In Introduction to Chemicals from Biomass; John Wiley & 182  Sons, Ltd, 2008; pp 103–142. (25)  Sjöström, E. Wood chemistry: fundamentals and applications; Academic Pr, 1993. (26)  Klemm, D.; Philipp, B.; Heinze, T.; Heinze, U.; Wagenknecht, W. General Considerations on Structure and Reactivity of Cellulose: Section 2.1–2.1. 4; Wiley Online Library, 2004. (27)  Jarvis, M. Chemistry: cellulose stacks up. Nature 2003, 426 (6967), 611–612. (28)  Rinaldi, R.; Schüth, F. Design of solid catalysts for the conversion of biomass. Energy & Environmental Science 2009, 2 (6), 610–626. (29)  Klemm, D.; Heublein, B.; Fink, H.; Bohn, A. Cellulose: fascinating biopolymer and sustainable raw material. Angewandte Chemie International Edition 2005, 44 (22), 3358–3393. (30)  Calahorra, M. E.; Cortazar, M.; Eguiazabal, J. I.; Guzman, G. M. Thermogravimetric Analysis of Cellulose : Effect of the Molecular Weight on Thermal Decomposition. Journal of Applied Polymer Science 1989, 37, 3305–3314. (31)  Oh, S. Y.; Dong, I. Y.; Shin, Y.; Hwan, C. K.; Hak, Y. K.; Yong, S. C.; Won, H. P.; Ji, H. Y. Crystalline structure analysis of cellulose treated with sodium hydroxide and carbon dioxide by means of X-ray diffraction and FTIR spectroscopy. Carbohydrate Research 2005, 340 (15), 2376–2391. (32)  Kim, S. J.; Dwiatmoko, A. A.; Choi, J. W.; Suh, Y. W.; Suh, D. J.; Oh, M. Cellulose pretreatment with 1-n-butyl-3-methylimidazolium chloride for solid acid-catalyzed hydrolysis. Bioresource Technology 2010, 101 (21), 8273–8279. (33)  Park, S.; Baker, J. O.; Himmel, M. E.; Parilla, P. A.; Johnson, D. K. Cellulose crystallinity index: Measurement techniques and their impact on interpreting cellulase performance. Biotechnology for Biofuels 2010, 3, 1–10. (34)  Food and Agriculture Organization of the United Nations. Pulp and Paper Capacities -Survey 2016-2021; 2017. (35)  Shopsowitz, K. E.; Hamad, W. Y.; MacLachlan, M. J. Chiral nematic mesoporous carbon derived from nanocrystalline cellulose. Angewandte Chemie - International Edition 2011, 50 (46), 10991–10995. (36)  Henriksson, M.; Berglund, L. A.; Isaksson, P.; Lindström, T.; Nishino, T. Cellulose nanopaper structures of high toughness. Biomacromolecules 2008, 9 (6), 1579–1585. 183  (37)  Mittal, N.; Ansari, F.; Gowda Krishne, V.; Brouzet, C.; Chen, P.; Larsson, P. T.; Roth, S. V.; Lundell, F.; Wågberg, L.; Kotov, N. A.; et al. Multiscale Control of Nanocellulose Assembly: Transferring Remarkable Nanoscale Fibril Mechanics to Macroscale Fibers. ACS Nano 2018, 12 (7), 6378–6388. (38)  Peng, B. L.; Dhar, N.; Liu, H. L.; Tam, K. C. Chemistry and applications of nanocrystalline cellulose and its derivatives: A nanotechnology perspective. Canadian Journal of Chemical Engineering 2011, 89 (5), 1191–1206. (39)  Islam, M. S.; Kao, N.; Bhattacharya, S. N.; Gupta, R.; Bhattacharjee, P. K. Effect of low pressure alkaline delignification process on the production of nanocrystalline cellulose from rice husk. Journal of the Taiwan Institute of Chemical Engineers 2017, 80, 820–834. (40)  Klass, D. L. Biomass for Renewable Energy, Fuels, and Chemicals; Academic Press. San Diego California, 1998. (41)  Yang, H.; Yan, R.; Chen, H.; Lee, D. H.; Zheng, C. Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel 2007, 86 (12–13), 1781–1788. (42)  Papageorgiou, G. Z.; Papageorgiou, D. G.; Terzopoulou, Z.; Bikiaris, D. N. Production of bio-based 2,5-furan dicarboxylate polyesters: Recent progress and critical aspects in their synthesis and thermal properties. European Polymer Journal 2016, 83, 202–229. (43)  Kröger, M.; Prüße, U.; Vorlop, K. D. A new approach for the production of 2,5-furandicarboxylic acid by in situ oxidation of 5-hydroxymethylfurfural starting from fructose. Topics in Catalysis 2000, 13 (3), 237–242. (44)  Mäki-Arvela, P.; Simakova, I. L.; Salmi, T.; Murzin, D. Y. Production of lactic acid/lactates from biomass and their catalytic transformations to commodities. Chemical Reviews 2014, 114 (3), 1909–1971. (45)  Sakakibara, A.; Sano, Y. Chemistry of lignin. Wood and cellulosic chemistry 2000, 109–174. (46)  Bandura, A. V.; Lvov, S. N. The ionization constant of water over wide ranges of temperature and density. Journal of Physical and Chemical Reference Data 2006, 35 (1), 15–30. (47)  Ralph, J.; Lapierre, C.; Boerjan, W. Lignin structure and its engineering. Current Opinion in Biotechnology 2019, 56, 240–249. (48)  Mottiar, Y.; Vanholme, R.; Boerjan, W.; Ralph, J.; Mansfield, S. D. Designer lignins: 184  Harnessing the plasticity of lignification. Current Opinion in Biotechnology 2016, 37, 190–200. (49)  Mahon, E. L.; Mansfield, S. D. Tailor-made trees: engineering lignin for ease of processing and tomorrow’s bioeconomy. Current Opinion in Biotechnology 2019, 56, 147–155. (50)  Fengel, D.; Wegener, G. Wood: chemistry, ultrastructure, reactions; 1984; Vol. 613. (51)  Gierer, J.; Lindeberg, O.; Kulonen, E.; Deutsch, A.; Matsuno, T. Reactions of Lignin During Sulfate Pulping. Part XV. The Behaviour of Intermediary Coniferyl Alcohol Structures. Acta Chemica Scandinavica B 1978, 32, 577–587. (52)  Gellerstedt, G. Softwood kraft lignin: Raw material for the future. Industrial Crops and Products 2015, 77, 845–854. (53)  Kanungo, D.; Francis, R. C.; Shin, N.-H. Mechanistic Differences Between Kraft and Soda/AQ Pulping. Part 2: Results from Lignin Model Compounds. Journal of Wood Chemistry and Technology 2009, 29 (3), 227–240. (54)  Li, J.; Henriksson, G.; Gellerstedt, G. Lignin depolymerization/repolymerization and its critical role for delignification of aspen wood by steam explosion. Bioresource Technology 2007, 98 (16), 3061–3068. (55)  Trajano, H. L.; Engle, N. L.; Foston, M.; Ragauskas, A. J.; Tschaplinski, T. J.; Wyman, C. E. The fate of lignin during hydrothermal pretreatment. Biotechnology for Biofuels 2013, 6 (1), 1–16. (56)  Zhao, X.; Cheng, K.; Liu, D. Organosolv pretreatment of lignocellulosic biomass for enzymatic hydrolysis. Applied Microbiology and Biotechnology 2009, 82 (5), 815–827. (57)  McDonough, T. J. The Chemistry of Organosolv Delignification. Institute of Paper Science and Technology 1992. (58)  Ferrini, P.; Rinaldi, R. Catalytic biorefining of plant biomass to non-pyrolytic lignin bio-oil and carbohydrates through hydrogen transfer reactions. Angewandte Chemie - International Edition 2014, 53, 8634–8639. (59)  Koo, B. W.; Min, B. C.; Gwak, K. S.; Lee, S. M.; Choi, J. W.; Yeo, H.; Choi, I. G. Structural changes in lignin during organosolv pretreatment of Liriodendron tulipifera and the effect on enzymatic hydrolysis. Biomass and Bioenergy 2012, 42, 24–32. (60)  Park, N.; Kim, H. Y.; Koo, B. W.; Yeo, H.; Choi, I. G. Organosolv pretreatment with 185  various catalysts for enhancing enzymatic hydrolysis of pitch pine (Pinus rigida). Bioresource Technology 2010, 101 (18), 7046–7053. (61)  Park, N.; Kim, H. Y.; Koo, B. W.; Yeo, H.; Choi, I. G. Organosolv pretreatment with various catalysts for enhancing enzymatic hydrolysis of pitch pine (Pinus rigida). Bioresource Technology 2010, 101 (18), 7046–7053. (62)  Wildschut, J.; Smit, A. T.; Reith, J. H.; Huijgen, W. J. J. Ethanol-based organosolv fractionation of wheat straw for the production of lignin and enzymatically digestible cellulose. Bioresource Technology 2013, 135, 58–66. (63)  Pereira, L.; Vinícius, L.; Gurgel, A.; Marabezi, K.; Aprigio, A. Delignification of sugarcane bagasse using glycerol – water mixtures to produce pulps for saccharification. Bioresource Technology 2011, 102, 10040–10046. (64)  Martín, C.; Puls, J.; Saake, B.; Schreiber, A. Effect of glycerol pretreatment on component recovery and enzymatic hydrolysis of sugarcane bagasse. Cellulose Chemistry and Technology 2011, 45, 487–494. (65)  Demirbas, A. Aqueous glycerol delignification of wood chips and ground wood. Bioresource Technology 1998, 63, 179–185. (66)  Demirbas, A. Mathematical modeling on thermal degradation of wood chips using glycerol and alkaline glycerol. Energy sources 2004, 26 (12), 1165–1175. (67)  Zhang, W.; Sathitsuksanoh, N.; Barone, J. R.; Renneckar, S. Enhanced enzymatic saccharification of pretreated biomass using glycerol thermal processing (GTP). Bioresource Technology 2015, 199, 148–154. (68)  Zhang, W.; Barone, J. R.; Renneckar, S. Biomass fractionation after denaturing cell walls by glycerol thermal processing. ACS Sustainable Chemistry and Engineering 2015, 3 (3), 413–420. (69)  Luterbacher, J. S.; Rand, J. M.; Alonso, D. M.; Han, J.; Youngquist, J. T.; Maravelias, C. T.; Pfleger, B. F.; Dumesic, J. A. Nonenzymatic Sugar Production from Biomass Using Biomass-Derived γ-Valerolactone. Science 2014, 343 (6168), 277–281. (70)  Alonso, D. M.; Hakim, S. H.; Zhou, S.; Won, W.; Hosseinaei, O.; Tao, J.; Garcia-Negron, V.; Motagamwala, A. H.; Mellmer, M. A.; Huang, K.; et al. Increasing the revenue from lignocellulosic biomass: Maximizing feedstock utilization. Science Advances 2017, 3 (5), e1603301. 186  (71)  Shuai, L.; Amiri, M. T.; Questell-Santiago, Y. M.; Héroguel, F.; Li, Y.; Kim, H.; Meilan, R.; Chapple, C.; Ralph, J.; Luterbacher, J. S. Formaldehyde stabilization facilitates lignin monomer production during biomass depolymerization. Science 2016, 354 (6310), 329 LP – 333. (72)  Lan, W.; Amiri, M. T.; Hunston, C. M.; Luterbacher, J. S. Protection Group Effects During α,γ-Diol Lignin Stabilization Promote High-Selectivity Monomer Production. Angewandte Chemie - International Edition 2018, 57 (5), 1356–1360. (73)  Bridgwater, A. V. Review of fast pyrolysis of biomass and product upgrading. Biomass and Bioenergy 2012, 38, 68–94. (74)  Toor, S. S.; Rosendahl, L.; Rudolf, A. Hydrothermal liquefaction of biomass: A review of subcritical water technologies. Energy 2011, 36 (5), 2328–2342. (75)  Demirbas, A. Determination of calorific values of bio-chars and pyro-oils from pyrolysis of beech trunkbarks. Journal of Analytical and Applied Pyrolysis 2004, 72 (2), 215–219. (76)  Horne, P. A.; Williams, P. T. Influence of temperature on the products from the flash pyrolysis of biomass. Fuel 1996, 75 (9), 1051–1059. (77)  Bridgwater, A. V.; Meier, D.; Radlein, D. An overview of fast pyrolysis of biomass. Organic Geochemistry 1999, 30, 1479–1493. (78)  Mohan, D.; Pittman, C. U.; Steele, P. H. Pyrolysis of Wood / Biomass for Bio-oil : A Critical Review. Energy & Fuesl 2006, 20 (4), 848–889. (79)  Brand, S.; Hardi, F.; Kim, J.; Suh, D. J. Effect of heating rate on biomass liquefaction: Differences between subcritical water and supercritical ethanol. Energy 2014, 68, 420–427. (80)  Liu, Z.; Zhang, F. S. Effects of various solvents on the liquefaction of biomass to produce fuels and chemical feedstocks. Energy Conversion and Management 2008, 49 (12), 3498–3504. (81)  Wang, X.; Rinaldi, R. Solvent Effects on the Hydrogenolysis of Diphenyl Ether with Raney Nickel and their Implications for the Conversion of Lignin. ChemSusChem 2012, 5 (8), 1455–1466. (82)  Zhao, C.; Kasakov, S.; He, J.; Lercher, J. A. Comparison of kinetics, activity and stability of Ni/HZSM-5 and Ni/Al2O3-HZSM-5 for phenol hydrodeoxygenation. Journal of Catalysis 2012, 296, 12–23. 187  (83)  Ertas, I. E.; Gulcan, M.; Bulut, A.; Yurderi, M.; Zahmakiran, M. Metal-organic framework (MIL-101) stabilized ruthenium nanoparticles: Highly efficient catalytic material in the phenol hydrogenation. Microporous and Mesoporous Materials 2016, 226, 94–103. (84)  Bindwal, A. B.; Vaidya, P. D. Reaction Kinetics of Vanillin Hydrogenation in Aqueous Solutions Using a Ru/C Catalyst. Energy & Fuels 2014, 28 (5), 3357–3362. (85)  Bjelić, A.; Grilc, M.; Likozar, B. Catalytic hydrogenation and hydrodeoxygenation of lignin-derived model compound eugenol over Ru/C: Intrinsic microkinetics and transport phenomena. Chemical Engineering Journal 2018, 333 (July 2017), 240–259. (86)  Huš, M.; Bjelić, A.; Grilc, M.; Likozar, B. First-principles mechanistic study of ring hydrogenation and deoxygenation reactions of eugenol over Ru(0001) catalysts. Journal of Catalysis 2018, 358, 8–18. (87)  Li, G.; Han, J.; Wang, H.; Zhu, X.; Ge, Q. Role of dissociation of phenol in its selective hydrogenation on Pt(111) and Pd(111). ACS Catalysis 2015, 5 (3), 2009–2016. (88)  Zhao, C.; He, J.; Lemonidou, A. A.; Li, X.; Lercher, J. A. Aqueous-phase hydrodeoxygenation of bio-derived phenols to cycloalkanes. Journal of Catalysis 2011, 280 (1), 8–16. (89)  Nelson, N. C.; Manzano, J. S.; Sadow, A. D.; Overbury, S. H.; Slowing, I. I. Selective Hydrogenation of Phenol Catalyzed by Palladium on High-Surface-Area Ceria at Room Temperature and Ambient Pressure. ACS Catalysis 2015, 5 (4), 2051–2061. (90)  Wang, Y.; Yao, J.; Li, H.; Su, D.; Antonietti, M. Highly selective hydrogenation of phenol and derivatives over a Pd@carbon nitride catalyst in aqueous media. Journal of the American Chemical Society 2011, 133 (8), 2362–2365. (91)  Liu, H.; Jiang, T.; Han, B.; Liang, S.; Zhou, Y. Selective phenol hydrogenation to cyclohexanone over a dual supported Pd-Lewis acid catalyst. Science 2009, 326 (5957), 1250–1252. (92)  Satterfield, C. N. Heterogeneous catalysis in industrial practice. Heterogeneous catalysis in industrial practice 1991. (93)  Zhao, C.; Kou, Y.; Lemonidou, A. A.; Li, X.; Lercher, J. A. Hydrodeoxygenation of bio-derived phenols to hydrocarbons using Raney Ni® and Nafion/SiO2 catalysts. Chemical communications 2010, 46 (3), 412–414. 188  (94)  Xiang, Y.; Ma, L.; Lu, C.; Zhang, Q.; Li, X. Aqueous system for the improved hydrogenation of phenol and its derivatives. Green Chemistry 2008, 10 (9), 939–943. (95)  Xu, Y.; Zhang, L.; Chang, J.; Zhang, X.; Ma, L.; Wang, T.; Zhang, Q. One step hydrogenation-esterification of model compounds and bio-oil to alcohols and esters over Raney Ni catalysts. Energy Conversion and Management 2016, 108, 78–84. (96)  Al-Mawlawi, D.; Saleh, J. M. Interaction of alcohols with evaporated metal films. Part 1.-Adsorption and decomposition of methanol on metal films of nickel, palladium and aluminium. Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases 1981, 77 (12), 2965–2976. (97)  Zhou, Y.-H.; Lv, P.-H.; Wang, G.-C. DFT studies of methanol decomposition on Ni (100) surface: Compared with Ni (111) surface. Journal of Molecular Catalysis A: Chemical 2006, 258 (1), 203–215. (98)  Alonso, D. M.; Wettstein, S. G.; Dumesic, J. A. Bimetallic catalysts for upgrading of biomass to fuels and chemicals. Chemical Society Reviews 2012, 41 (24), 8075. (99)  Akiya, N.; Savage, P. E. Kinetics and mechanism of cyclohexanol dehydration in high-temperature water. Industrial and Engineering Chemistry Research 2001, 40 (8), 1822–1831. (100)  Dwiatmoko, A. A.; Zhou, L.; Kim, I.; Choi, J.-W.; Suh, D. J.; Ha, J.-M. Hydrodeoxygenation of lignin-derived monomers and lignocellulose pyrolysis oil on the carbon-supported Ru catalysts. Catalysis Today 2015, 265, 192–198. (101)  Ohta, H.; Kobayashi, H.; Hara, K.; Fukuoka, A. Hydrodeoxygenation of phenols as lignin models under acid-free conditions with carbon-supported platinum catalysts. Chemical Communications 2011, 47 (44), 12209–12211. (102)  Yoon, Y.; Rousseau, R.; Weber, R. S.; Mei, D.; Lercher, J. A. First-principles study of phenol hydrogenation on pt and ni catalysts in aqueous phase. Journal of the American Chemical Society 2014, 136 (29), 10287–10298. (103)  Lee, C. R.; Yoon, J. S.; Suh, Y.-W.; Choi, J.-W.; Ha, J.-M.; Suh, D. J.; Park, Y.-K. Catalytic roles of metals and supports on hydrodeoxygenation of lignin monomer guaiacol. Catalysis Communications 2012, 17, 54–58. (104)  Bui, V. N.; Laurenti, D.; Delichère, P.; Geantet, C. Hydrodeoxygenation of guaiacol. Part II: Support effect for CoMoS catalysts on HDO activity and selectivity. Applied Catalysis 189  B: Environmental 2011, 101 (3–4), 246–255. (105)  Mora-Vergara, I. D.; Hernández Moscoso, L.; Gaigneaux, E. M.; Giraldo, S. A.; Baldovino-Medrano, V. G. Hydrodeoxygenation of guaiacol using NiMo and CoMo catalysts supported on alumina modified with potassium. Catalysis Today 2018, 302 (July 2017), 125–135. (106)  Park, H. J.; Heo, H. S.; Jeon, J. K.; Kim, J.; Ryoo, R.; Jeong, K. E.; Park, Y. K. Highly valuable chemicals production from catalytic upgrading of radiata pine sawdust-derived pyrolytic vapors over mesoporous MFI zeolites. Applied Catalysis B: Environmental 2010, 95 (3–4), 365–373. (107)  Yu, Y.; Li, X.; Su, L.; Zhang, Y.; Wang, Y.; Zhang, H. The role of shape selectivity in catalytic fast pyrolysis of lignin with zeolite catalysts. Applied Catalysis A: General 2012. (108)  Jae, J.; Tompsett, G. A.; Foster, A. J.; Hammond, K. D.; Auerbach, S. M.; Lobo, R. F.; Huber, G. W. Investigation into the shape selectivity of zeolite catalysts for biomass conversion. Journal of Catalysis 2011, 279 (2), 257–268. (109)  Hellinger, M.; Carvalho, H. W. P.; Baier, S.; Wang, D.; Kleist, W.; Grunwaldt, J. D. Catalytic hydrodeoxygenation of guaiacol over platinum supported on metal oxides and zeolites. Applied Catalysis A: General 2015, 490, 181–192. (110)  Ben, H.; Ragauskas, A. J. One step thermal conversion of lignin to the gasoline range liquid products by using zeolites as additives. RSC Advances 2012, 2 (33), 12892–12898. (111)  Ben, H.; Ragauskas, A. J. Influence of Si/Al Ratio of ZSM ‑ 5 Zeolite on the Properties of Lignin Pyrolysis Products. ACS Sustainable Chemistry & Engineering 2013, 1, 316–324. (112)  Mullen, C. A.; Boateng, A. A. Catalytic pyrolysis-GC/MS of lignin from several sources. Fuel Processing Technology 2010, 91 (11), 1446–1458. (113)  Ma, Z.; Troussard, E.; van Bokhoven, J. A. Controlling the selectivity to chemicals from lignin via catalytic fast pyrolysis. Applied Catalysis A: General 2012, 423–424, 130–136. (114)  Yu, Y.; Li, X.; Su, L.; Zhang, Y.; Wang, Y.; Zhang, H. The role of shape selectivity in catalytic fast pyrolysis of lignin with zeolite catalysts. Applied Catalysis A: General 2012, 447–448, 115–123. (115)  Ravenelle, R.; Schüßler, F.; Amico, A. D.; Danilina, N.; Van, J. A.; Lercher, J. A.; Jones, C. W.; Sievers, C. Stability of Zeolites in Hot Liquid Water. Jounal Physycal Chemistry C 2010, 114, 19582–19595. 190  (116)  Hicks, J. C. Advances in C-O Bond Transformations in Lignin-Derived. The Journal of Physical Chemistry Letters 2011, 2, 2280–2287. (117)  Chen, K.; Damron, J.; Pearson, C.; Resasco, D.; Zhang, L.; White, J. L. Zeolite Catalysis: Water Can Dramatically Increase or Suppress Alkane C–H Bond Activation. ACS Catalysis 2014, 4 (9), 3039–3044. (118)  Lee, E. H.; Park, R. su; Kim, H.; Park, S. H.; Jung, S. C.; Jeon, J. K.; Kim, S. C.; Park, Y. K. Hydrodeoxygenation of guaiacol over Pt loaded zeolitic materials. Journal of Industrial and Engineering Chemistry 2016, 37, 18–21. (119)  Zhang, X.; Wang, T.; Ma, L.; Zhang, Q.; Jiang, T. Hydrotreatment of bio-oil over Ni-based catalyst. Bioresource Technology 2013, 127, 306–311. (120)  Zhao, C.; Lercher, J. A. Upgrading Pyrolysis Oil over Ni/HZSM-5 by Cascade Reactions. Angewandte Chemie - International Edition 2012, 124 (24), 6037–6042. (121)  Zhao, C.; Yu, Y.; Jentys, A.; Lercher, J. A. Understanding the impact of aluminum oxide binder on Ni/HZSM-5 for phenol hydrodeoxygenation. Applied Catalysis B: Environmental 2013, 132–133, 282–292. (122)  Sun, J.; Karim, A. M.; Zhang, H.; Kovarik, L.; Li, X. S.; Hensley, A. J.; McEwen, J. S.; Wang, Y. Carbon-supported bimetallic Pd-Fe catalysts for vapor-phase hydrodeoxygenation of guaiacol. Journal of Catalysis 2013, 306, 47–57. (123)  He, Z.; Hu, M.; Wang, X. Highly effective hydrodeoxygenation of guaiacol on Pt/TiO2: Promoter effects. Catalysis Today 2018, 302, 136–145. (124)  Fang, H.; Zheng, J.; Luo, X.; Du, J.; Roldan, A.; Leoni, S.; Yuan, Y. Product tunable behavior of carbon nanotubes-supported Ni–Fe catalysts for guaiacol hydrodeoxygenation. Applied Catalysis A: General 2017, 529, 20–31. (125)  Shuai, L.; Luterbacher, J. Organic Solvent Effects in Biomass Conversion Reactions. ChemSusChem 2016, 9 (2), 133–155. (126)  Ponndorf, W. Der reversible Austausch der Oxydationsstufen zwischen Aldehyden oder Ketonen einerseits und primären oder sekundären Alkoholen anderseits. Angewandte Chemie 1926, 39 (5), 138–143. (127)  Samec, J. S. M.; Bäckvall, J.-E.; Andersson, P. G.; Brandt, P. Mechanistic aspects of transition metal-catalyzed hydrogen transfer reactions. Chemical Society Reviews 2006, 35 (3), 237. 191  (128)  Laxmi, Y. R. S.; Bäckvall, J. E. Mechanistic studies on ruthenium-catalyzed hydrogen transfer reactions. Chemical Communications 2000, No. 7, 611–612. (129)  Wang, X.; Rinaldi, R. Exploiting H-transfer reactions with Raney Ni® for upgrade of phenolic and aromatic biorefinery feeds under unusual, low-severity conditions. Energy & Environmental Science 2012, 5, 8244. (130)  Wang, X.; Rinaldi, R. A route for lignin and bio-oil conversion: Dehydroxylation of phenols into arenes by catalytic tandem reactions. Angewandte Chemie - International Edition 2013, 52, 11499–11503. (131)  Kennema, M.; De Castro, I. B. D.; Meemken, F.; Rinaldi, R. Liquid-Phase H-Transfer from 2-Propanol to Phenol on Raney Ni: Surface Processes and Inhibition. ACS Catalysis 2017, 7 (4), 2437–2445. (132)  Kim, M.; Ha, J. M.; Lee, K. Y.; Jae, J. Catalytic transfer hydrogenation/hydrogenolysis of guaiacol to cyclohexane over bimetallic RuRe/C catalysts. Catalysis Communications 2016, 86, 113–118. (133)  Hasegawa, I.; Tabata, K.; Okuma, O.; Mae, K. New Pretreatment Methods Combining a Hot Water Treatment and Water / Acetone Extraction for Thermo-Chemical Conversion of Biomass. Energy & Fuels 2004, 45 (6), 755–760. (134)  Song, Q.; Wang, F.; Cai, J.; Wang, Y.; Zhang, J.; Yu, W.; Xu, J. Lignin depolymerization (LDP) in alcohol over nickel-based catalysts via a fragmentation–hydrogenolysis process. Energy & Environmental Science 2013, 6 (3), 994. (135)  Barta, K.; Matson, T. D.; Fettig, M. L.; Scott, S. L.; Iretskii, V.; Ford, P. C. Catalytic disassembly of an organosolv lignin via hydrogen transfer from supercritical methanol. Green Chemistry 2010, 12, 1640–1647. (136)  Gandarias, I.; Arias, P. L.; Fernández, S. G.; Requies, J.; El Doukkali, M.; Güemez, M. B. Hydrogenolysis through catalytic transfer hydrogenation: Glycerol conversion to 1,2-propanediol. Catalysis Today 2012, 195 (1), 22–31. (137)  Zhang, D.; Ye, F.; Xue, T.; Guan, Y.; Wang, Y. M. Transfer hydrogenation of phenol on supported Pd catalysts using formic acid as an alternative hydrogen source. Catalysis Today 2014, 234, 133–138. (138)  Toledano, A.; Serrano, L.; Labidi, J.; Pineda, A.; Balu, A. M.; Luque, R. Heterogeneously Catalysed Mild Hydrogenolytic Depolymerisation of Lignin Under Microwave Irradiation 192  with Hydrogen-Donating Solvents. ChemCatChem 2013, 5 (4), 977–985. (139)  Kristianto, I.; Limarta, S. O.; Lee, H.; Ha, J. M.; Suh, D. J.; Jae, J. Effective depolymerization of concentrated acid hydrolysis lignin using a carbon-supported ruthenium catalyst in ethanol/formic acid media. Bioresource Technology 2017, 234, 424–431. (140)  Feng, G.; Liu, Z.; Chen, P.; Lou, H. Influence of solvent on upgrading of phenolic compounds in pyrolysis bio-oil. RSC Advances 2014, 4 (91), 49924–49929. (141)  Luterbacher, J.; Azarpira, A.; Motagamwala, A. H.; Lu, F.; Ralph, J.; Dumesic, J. Lignin monomer production integrated into the g-valerolactone sugar platform. Energy & Environmental Science 2015, 8, 2657–2663. (142)  Duan, H.; Dong, J.; Gu, X.; Peng, Y.; Chen, W.; Issariyakul, T.; Myers, W. K.; Li, M.; Yi, N.; Kilpatrick, A. F. R.; et al. Hydrodeoxygenation of water-insoluble bio-oil to alkanes using a highly dispersed Pd – Mo catalyst. Nature Communications 2017, 8 (1), 591. (143)  Roman-Leshkov, Y.; Chheda, J. N.; Dumesic, J. A. Production of Hydroxymethylfurfural from Fructose. Science 2006, 312 (4), 1933. (144)  Crossley, S.; Faria, J.; Shen, M.; Resasco, D. E. Solid nanoparticles that catalyze biofuel upgrade reactions at the water/oil interface. Science 2010, 327 (5961), 68–72. (145)  Mellmer, M. A.; Sanpitakseree, C.; Demir, B.; Bai, P.; Ma, K.; Neurock, M.; Dumesic, J. A. Solvent-enabled control of reactivity for liquid-phase reactions of biomass-derived compounds. Nature Catalysis 2018, 1 (March), 199–207. (146)  Araque, E.; Parra, C.; Freer, J.; Contreras, D.; Rodríguez, J.; Mendonça, R.; Baeza, J. Evaluation of organosolv pretreatment for the conversion of Pinus radiata D. Don to ethanol. Enzyme and Microbial Technology 2008, 43 (2), 214–219. (147)  Van den Bosch, S.; Schutyser, W.; Vanholme, R.; Driessen, T.; Koelewijn, S.-F.; Renders, T.; De Meester, B.; Huijgen, W. J. J.; Dehaen, W.; Courtin, C. M.; et al. Reductive lignocellulose fractionation into soluble lignin-derived phenolic monomers and dimers and processable carbohydrate pulps. Energy & Environmental Science 2015, 8 (6), 1748–1763. (148)  Van Den Bosch, S.; Schutyser, W.; Koelewijn, S. F.; Renders, T.; Courtin, C. M.; Sels, B. F. Tuning the lignin oil OH-content with Ru and Pd catalysts during lignin hydrogenolysis on birch wood. Chemical Communications 2015, 51 (67), 13158–13161. 193  (149)  Song, Q.; Wang, F.; Xu, J. Hydrogenolysis of lignosulfonate into phenols over heterogeneous nickel catalysts. Chemical Communications 2012, 48 (56), 7019. (150)  Xu, C.; Arancon, R. A. D.; Labidi, J.; Luque, R. Lignin depolymerisation strategies: Towards valuable chemicals and fuels. Chemical Society Reviews 2014, 43 (22), 7485–7500. (151)  Xiong, Z.; Wang, Y.; Syed-Hassan, S. S. A.; Hu, X.; Han, H.; Su, S.; Xu, K.; Jiang, L.; Guo, J.; Berthold, E. E. S.; et al. Effects of heating rate on the evolution of bio-oil during its pyrolysis. Energy Conversion and Management 2018, 163 (December 2017), 420–427. (152)  Zhou, Q.; Liu, Q.; Shi, L.; Yan, Y.; Liu, Z. Behaviors of coking and radicals during reaction of volatiles generated from fixed-bed pyrolysis of a lignite and a subbituminous coal. Fuel Processing Technology 2017, 161, 304–310. (153)  Wang, Y.; Mourant, D.; Hu, X.; Zhang, S.; Lievens, C.; Li, C. Z. Formation of coke during the pyrolysis of bio-oil. Fuel 2013, 108, 439–444. (154)  Song, Q.; Wang, F.; Cai, J.; Wang, Y.; Zhang, J.; Yu, W.; Xu, J. Lignin depolymerization (LDP) in alcohol over nickel-based catalysts via a fragmentation-hydrogenolysis process. Energy & Environmental Science 2013, 6 (3), 994–1007. (155)  Qineng, X.; Chen, Z.; Shao, Y.; Gong, X.; Wang, H.; Liu, X.; Parker, S. F.; Han, X.; Yang, S.; Wang, Y. Direct hydrodeoxygenation of raw woody biomass into liquid alkanes. Nature Communications 2016, 7, 11162. (156)  Zhang, X.; Wang, T.; Ma, L.; Wu, C. Aqueous-phase catalytic process for production of pentane from furfural over nickel-based catalysts. Fuel 2010, 89 (10), 2697–2702. (157)  Dwiatmoko, A. A.; Lee, S.; Ham, H. C.; Choi, J.; Suh, D. J.; Ha, J. Effects of Carbohydrates on the Hydrodeoxygenation of Lignin- Derived Phenolic Compounds. ACS Catalysis 2015, 1–5. (158)  Furimsky, E. Catalytic hydrodeoxygenation. Applied Catalysis A: General 2000, 199 (2), 147–190. (159)  Ruddy, D. A.; Schaidle, J. A.; Ferrell III, J. R.; Wang, J.; Moens, L.; Hensley, J. E. Recent advances in heterogeneous catalysts for bio-oil upgrading via “ex situ catalytic fast pyrolysis”: catalyst development through the study of model compounds. Green Chemistry 2014, 16 (2), 454. (160)  Jacobson, K.; Maheria, K. C.; Kumar Dalai, A. Bio-oil valorization: A review. Renewable 194  and Sustainable Energy Reviews 2013, 23, 91–106. (161)  Bridgwater, A. V. Upgrading Biomass Fast Pyrolysis Liquids. Environmental Progress & Sustainable Energy 2012, 31 (2), 1944–7442. (162)  Wang, H.; Male, J.; Wang, Y. Recent advances in hydrotreating of pyrolysis bio-oil and its oxygen-containing model compounds. ACS Catalysis 2013, 3 (5), 1047–1070. (163)  Chang, J. R.; Chang, S. L.; Lin, T. B. γ-Alumina-supported Pt catalysts for aromatics reduction: A structural investigation of sulfur poisoning catalyst deactivation. Journal of Catalysis 1997, 169 (1), 338–346. (164)  Speight, J. Lange’s handbook of chemistry; McGraw Hill Professional, 2016. (165)  Barelli, L.; Bidini, G.; Gallorini, F.; Servili, S. Hydrogen production through sorption-enhanced steam methane reforming and membrane technology: A review. Energy 2008, 33 (4), 554–570. (166)  Anitha, M.; Kamarudin, S. K.; Kofli, N. T. The potential of glycerol as a value-added commodity. Chemical Engineering Journal 2016, 295, 119–130. (167)  Quispe, C. A. G.; Coronado, C. J. R.; Carvalho Jr, J. A. Glycerol: Production, consumption, prices, characterization and new trends in combustion. Renewable and Sustainable Energy Reviews 2013, 27, 475–493. (168)  Ciftci, A.; Ligthart, D. A. J. M.; Hensen, E. J. M. Aqueous phase reforming of glycerol over Re-promoted Pt and Rh catalysts. Green Chemistry 2014, 16 (2), 853–863. (169)  Wawrzetz, A.; Peng, B.; Hrabar, A.; Jentys, A.; Lemonidou, A. A.; Lercher, J. A. Towards understanding the bifunctional hydrodeoxygenation and aqueous phase reforming of glycerol. Journal of Catalysis 2010, 269 (2), 411–420. (170)  Davda, R. R.; Shabaker, J. W.; Huber, G. W.; Cortright, R. D.; Dumesic, J. A. Aqueous-phase reforming of ethylene glycol on silica-supported metal catalysts. Applied Catalysis B: Environmental 2003, 43, 13–26. (171)  Ciftci, A.; Ligthart, D. A. J. M.; Hensen, E. J. M. Aqueous phase reforming of glycerol over Re-promoted Pt and Rh catalysts. Green Chemistry 2014, 16 (2), 853–863. (172)  King, D. L.; Zhang, L.; Xia, G.; Karim, A. M.; Heldebrant, D. J.; Wang, X.; Peterson, T.; Wang, Y. Aqueous phase reforming of glycerol for hydrogen production over Pt-Re supported on carbon. Applied Catalysis B: Environmental 2010, 99 (1–2), 206–213. (173)  Soares, R. R.; Simonetti, D. A.; Dumesic, J. A. Glycerol as a source for fuels and 195  chemicals by low-temperature catalytic processing. Angewandte Chemie - International Edition 2006, 45 (24), 3982–3985. (174)  Huber, G. W.; Cortright, R. D.; Dumesic, J. A. Renewable Alkanes by Aqueous-Phase Reforming of Biomass-Derived Oxygenates. Angewandte Chemie - International Edition 2004, 116 (12), 1575–1577. (175)  Huber, G. W.; Chheda, J. N.; Barrett, C. J.; Dumesic, J. A. Production of Liquid Alkanes by Aqueous-Phase Processing of Biomass-Derived Carbohydrates. Science 2005, 308 (5727), 1446 LP – 1450. (176)  Huber, G. W.; Shabaker, J. W.; Evans, S. T.; Dumesic, J. A. Aqueous-phase reforming of ethylene glycol over supported Pt and Pd bimetallic catalysts. Applied Catalysis B: Environmental 2006, 62 (3–4), 226–235. (177)  Huber, G. W.; Shabaker, J. W.; Dumesic, J. A. Raney Ni-Sn Catalyst for H2 Production from Biomass-Derived Hydrocarbons. Science 2003, 300 (June), 2075–2078. (178)  Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D.; Nrel, D. C. Determination of Structural Carbohydrates and Lignin in Biomass Determination of Structural Carbohydrates and Lignin in Biomass: Laboratory Analytical Procedure (LAP); 2011; Vol. 2011. (179)  Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D. Determination of Ash in Biomass: Laboratory Analytical Procedure (LAP); 2008. (180)  Johnson, D. T.; Taconi, K. A. The Glycerin Glut: Options for the Value-Added Conversion of Crude Glycerol Resulting from Biodiesel Production. Joint Commission Journal on Quality and Patient Safety 2007, 26 (4), 338–348. (181)  Mainil, R. I.; Paksung, N.; Matsumura, Y. Determination of retro-aldol reaction type for glyceraldehyde under hydrothermal conditions. Journal of Supercritical Fluids 2019, 143 (June 2018), 370–377. (182)  Nozawa, T.; Mizukoshi, Y.; Yoshida, A.; Naito, S. Aqueous phase reforming of ethanol and acetic acid over TiO2 supported Ru catalysts. Applied Catalysis B: Environmental 2014, 146, 221–226. (183)  Roy, B.; Martinez, U.; Loganathan, K.; Datye, A. K.; Leclerc, C. A. Effect of preparation methods on the performance of Ni/Al2O3 catalysts for aqueous-phase reforming of ethanol: Part I-catalytic activity. International Journal of Hydrogen Energy 2012, 37 (10), 196  8143–8153. (184)  Chemeo. Chemical Properties of Propylene Glycol https://www.chemeo.com/cid/23-447-0/Propylene Glycol. (185)  Verevkin, S. P.; Zaitsau, D. H.; Emel’yanenko, V. N.; Zhabina, A. A. Thermodynamic properties of glycerol: Experimental and theoretical study. Fluid Phase Equilibria 2015, 397, 87–94. (186)  Sabatier, P. Process of Manufacturing Methane or of Mixture of Methane and Hydrogen. 9560734, 1910. (187)  Nikolaidis, P.; Poullikkas, A. A comparative overview of hydrogen production processes. Renewable and Sustainable Energy Reviews 2017, 67, 597–611. (188)  Shabaker, J. W.; Davda, R. R.; Huber, G. W.; Cortright, R. D.; Dumesic, J. A. Aqueous-phase reforming of methanol and ethylene glycol over alumina-supported platinum catalysts. Journal of Catalysis 2003, 215 (2), 344–352. (189)  Yu, J.; Savage, P. E. Decomposition of Formic Acid under Hydrothermal Conditions. Industrial and Engineering Chemistry Research 1998, 37 (1), 2–10. (190)  Xie, J.; Su, D.; Yin, X.; Wu, C.; Zhu, J. Thermodynamic analysis of aqueous phase reforming of three model compounds in bio-oil for hydrogen production. International Journal of Hydrogen Energy 2011, 36 (24), 15561–15572. (191)  Norton, P. R.; Kcy, A. Photoelectron spectroscopic studies of the adsorption of CO and CO2 on Ni, Pt and Cu. Chem. Phys. Lett. 1976, 38 (2), 207. (192)  Nakagawa, Y.; Ishikawa, M.; Tamura, M.; Tomishige, K. Selective production of cyclohexanol and methanol from guaiacol over Ru catalyst combined with MgO. Green Chemistry 2014, 16 (4), 2197–2203. (193)  Shabaker, J. W.; Dumesic, J. A. Kinetics of Aqueous-Phase Reforming of Oxygenated Hydrocarbons: Pt/Al2O3 and Sn-Modified Ni Catalysts. Industrial & Engineering Chemistry Research 2004, No. 43, 3105–3112. (194)  Monteiro-Gezork, A. C. A.; Natividad, R.; Winterbottom, J. M. Hydrogenation of naphthalene on NiMo- Ni- and Ru/Al2O3 catalysts: Langmuir–Hinshelwood kinetic modelling. Catalysis Today 2008, 130 (2–4), 471–485. (195)  Sweeton, F. H.; Mesmer, R. E.; Baes, C. F. Acidity measurements at elevated temperatures. VII. Dissociation of water. Journal of Solution Chemistry 1974, 3 (3), 191–197  214. (196)  Bothe, N.; Döscher, F.; Klein, J.; Widdecke, H. Thermal stability of sulphonated styrene-divinylbenzene resins. Polymer 1979, 20 (7), 850–854. (197)  Scholze, B.; Meier, D. Characterization of the water-insoluble fraction from pyrolysis oil (pyrolytic lignin). Part I. PY-GC/MS, FTIR, and functional groups. Journal of Analytical and Applied Pyrolysis 2001, 60 (1), 41–54. (198)  Hoang, T. M. C.; Van Eck, E. R. H.; Bula, W. P.; Gardeniers, J. G. E.; Lefferts, L.; Seshan, K. Humin based by-products from biomass processing as a potential carbonaceous source for synthesis gas production. Green Chemistry 2015, 17 (2), 959–972. (199)  Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D. Determination of Sugars, Byproducts, and Degradation Products in Liquid Fraction Process Samples: Laboratory Analytical Procedure (LAP); 2006. (200)  Choi, Y. S.; Johnston, P. A.; Brown, R. C.; Shanks, B. H.; Lee, K. H. Detailed characterization of red oak-derived pyrolysis oil: Integrated use of GC, HPLC, IC, GPC and Karl-Fischer. Journal of Analytical and Applied Pyrolysis 2014, 110 (1), 147–154. (201)  Yasuda, S.; Fukushima, K.; Kakehi, A. Formation and chemical structures of acid-soluble lignin I: Sulfuric acid treatment time and acid-soluble lignin content of hardwood. Journal of Wood Science 2001, 47 (1), 69–72. (202)  Torr, K. M.; van de Pas, D. J.; Cazeils, E.; Suckling, I. D. Mild hydrogenolysis of in-situ and isolated Pinus radiata lignins. Bioresource Technology 2011, 102 (16), 7608–7611. (203)  Masuku, C. P. Thermolytic decomposition of coniferyl alcohol. Journal of Analytical and Applied Pyrolysis 1992, 23 (2), 195–208. (204)  Lynskey, M. G.; Foster, M.; Bizub, J. J.; Anderson, J. E.; Polovina, D.; Shelby, M. H.; Leone, T. G.; Wallington, T. J. Octane Numbers of Ethanol-Gasoline Blends: Measurements and Novel Estimation Method from Molar Composition. SAE Technical Paper Series 2012, 1. (205)  D’Errico, G.; Ortona, O.; Capuano, F.; Vitagliano, V. Diffusion coefficients for the binary system glycerol + water at 25 °C. A velocity correlation study. Journal of Chemical and Engineering Data 2004, 49 (6), 1665–1670. 198  Appendices Appendix A  : Theoretical calculations  A.1 Minimum CO2/CH4 during APR  1. Methanol 1.1 In the case of CH4 only produced by Sabatier  CH3OH + H2O → CO2 + 3 H2 Initial 1    0  0 Reaction (-)1    (+)1  (+)2 Final 0    1  3 Since water is abundant, the amount of water will not be considered, thus the following Sabatier depending on the percentage of hydrogen will be:  CO2 + 4 H2 → CH4 + 2H2O Initial 1  3     Reaction (-)3a/4  (-)3a  (+)3a/4   Final (4-3a)/4  3(1-a)  3a/4   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 4−3𝑎3𝑎= 𝑅; where R is ratio. The minimum ratio will be obtained when a =1 with R = 1/3.  1.2 In the case of CH4 only produced by direct methanation  CH3OH + H2O → CO2 + 3 H2 Initial 1       Reaction (-)x    (+)x  (+)2x Final 1-x    x  3x          CH3OH +  H2 → CH4 + H2O Initial 1-x  3x     199  Reaction (-)3xa  (-)3xa  (+)3xa   Final 0  3x(1-a)  3xa   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 13𝑎= 𝑅 ; where R is the ratio. The minimum ratio will be obtained when a =1 with R = 1/3. The minimum x will be: (1-x-3x(1) = 0); which results x = 0.25 . It means that in the minimum ratio of CO2/CH4 will be obtain when 25% of CH4O converted to CO2 and H2 while the rest 75% of CH4O converted to CH4.   2. Ethanol 2.1 In the case of CH4 only produced by Sabatier  C2H5OH + 3H2O → 2CO2 + 6 H2 Initial 1    0  0 Reaction (-)1    (+)2  (+)6 Final 0    2  6 Since water is abundant, the amount of water will not be considered, thus the following Sabatier depending on the percentage of hydrogen will be:  CO2 + 4 H2 → CH4 + 2H2O Initial 2  6     Reaction (-)6a/4  (-)6a  (+)6a/4   Final (8-6a)/4  6(1-a)  6a/4   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 4−3𝑎3𝑎= 𝑅; where R is ratio. The minimum ratio will be obtained when a =1 with R = 1/3.    200  2.2  In the case of CH4 only produced by direct methanation  C2H5OH + 3H2O → 2CO2 + 6 H2 Initial 1       Reaction (-)x    (+)2x  (+)6x Final 1-x    2x  6x          C2H6O +  2H2 → 2CH4 + H2O Initial 1-x  6x     Reaction (-)3xa  (-)6xa  (+)6xa   Final 0  6x(1-a)  6xa   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 13𝑎= 𝑅 ; where R is the ratio. The minimum ratio will be obtained when a =1 with R = 1/3. The minimum x will be: (1-x-3x(1) = 0); which result x = 0.25. It means that in the minimum ratio of CO2/CH4 will be obtain when 25% of C2H6O converted to CO2 and H2 while the rest 75% of C2H6O converted to CH4.  3. Ethylene Glycol 3.1 In the case of CH4 only produced by Sabatier  C2H6O2 + 2H2O → 2CO2 + 5 H2 Initial 1    0  0 Reaction (-)1    (+)2  (+)5 Final 0    2  5 Since water is abundant, the amount of water will not be considered, thus the following Sabatier depending on the percentage of hydrogen will be:  CO2 + 4 H2 → CH4 + 2H2O Initial 2  5     Reaction (-)5a/4  (-)5a  (+)5a/4   201  Final (8-5a)/4  5(1-a)  5a/4   Where a is the percentage of hydrogen that further react to produce CH4 ; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 8−5𝑎5𝑎= 𝑅; where R is ratio The minimum ratio will be obtained when a =1 with R = 3/5.  3.2  In the case of CH4 only produced by direct methanation  C2H6O2 + 2H2O → 2CO2 + 5 H2 Initial 1       Reaction (-)x    (+)2x  (+)5x Final 1-x    2x  5x          C2H6O2 +  3H2 → 2CH4 + 2H2O Initial 1-x  5x     Reaction (-)5xa/3  (-)5xa  (+)10xa/3   Final 0  6x(1-a)  10xa/3   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 35𝑎= 𝑅 ; where R is the ratio  The minimum ratio will be obtained when a =1 with R = 3/5 The minimum x will be: (1-x-(5x(1)/3) = 0); which result x = 0.375 It means that in the minimum ratio of CO2/CH4 will be obtain when 37.5% of C2H6O2 converted to CO2 and H2 while the rest 62.5% of C2H6O2 converted to CH4   4. 1,2-propylene glycol 4.1 In the case of CH4 only produced by Sabatier  C3H8O2 + 4H2O → 3CO2 + 8 H2 Initial 1    0  0 Reaction (-)1    (+)3  (+)8 202  Final 0    3  8 Since water is abundant, the amount of water will not be considered, thus the following Sabatier depending on the percentage of hydrogen will be:  CO2 + 4 H2 → CH4 + 2H2O Initial 3  8     Reaction (-)2a  (-)8a  (+)2a   Final (3-2a)/4  8(1-a)  2a/4   Where a is the percentage of hydrogen that further react to produce CH4 ; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 3−2𝑎2𝑎= 𝑅; where R is ratio The minimum ratio will be obtained when a =1 with R = ½.  4.2  In the case of CH4 only produced by direct methanation  C3H8O2 + 4H2O → 3CO2 + 8 H2 Initial 1       Reaction (-)x    (+)3x  (+)8x Final 1-x    3x  8x          C3H8O2 +  4H2 → 3CH4 + 2H2O Initial 1-x  8x     Reaction (-)2xa  (-)8xa  (+)6xa   Final 0  8x(1-a)  6xa   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 12𝑎= 𝑅 ; where R is the ratio  The minimum ratio will be obtained when a =1 with R = ½. The minimum x will be: (1-x-2x(1) = 0) ; which result x = 0.33 It means that in the minimum ratio of CO2/CH4 will be obtain when 33.3% of C3H6O2 converted to CO2 and H2 while the rest of 77.7% of C3H6O2 converted to CH4  203  5. Glycerol 5.1 In the case of CH4 only produced by Sabatier  C3H8O3 + 3H2O → 3CO2 + 7 H2 Initial 1    0  0 Reaction (-)1    (+)3  (+)7 Final 0    3  7 Since water is abundant, the amount of water will not be considered, thus the following Sabatier depending on the percentage of hydrogen will be:  CO2 + 4 H2 → CH4 + 2H2O Initial 3  7     Reaction (-)7a/4  (-)7a  (+)7a/4  (+)2a Final 3-(7a/4)  7(1-a)  7a/4  7a/2 Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 𝐶𝑂2𝐶𝐻4= 12−7𝑎7𝑎= 𝑅 ; where R is ratio The minimum ratio will be obtained when a =1 with R = 5/7.  5.2 In the case of CH4 only produced by direct methanation  C3H8O3 + 3H2O → 3CO2 + 7 H2 Initial 1       Reaction (-)x    (+)3x  (+)7x Final 1-x    3x  7x          C3H8O3 + 5 H2 → 3CH4 + 3H2O Initial 1-x  7x     Reaction (-)7xa/5  (-)7xa  (+)21xa/5   Final (1-x)-(7xa/5)  7x(1-a)  21xa/5   Where a is the percentage of hydrogen that further react to produce CH4; 0 ≤ a ≤ 1 Thus, the ratio of CO2/CH4 will be: 204  𝐶𝑂2𝐶𝐻4= 57𝑎= 𝑅 ; where R is the ratio  The minimum x will be: (1-x-(7x(1)/5) = 0); which result x = 0.41 It means that in the minimum ratio of CO2/CH4 will be obtain when 41.7% of C3H6O3 converted to CO2 and H2 while 58.3% of C3H6O3 converted to CH4   205  A.2 Thermodynamics calculation Gibbs free energy for particular reaction (ΔGr) was calculated as follow: ∆𝐺𝑟 = ∆𝐻𝑟0 − 𝑇∆𝑆𝑟0    (A.1) Where ∆𝐻𝑟0 is standard enthalpy reaction (kJ), ∆𝑆𝑟0 is standard entropy reaction (kJ.K-1), and T (K) is temperature of reaction. Natural logarithmic of reaction constant (ln k) was calculated as: ln 𝑘 =  −∆𝐺𝑟𝑅𝑇  (A.2) Where R is Boltzmann gas constant (8.314 kJ. mol-1.K-1). The enthalpy, entropy, and natural logarithmic of reaction constant ln k is presented in Table A.1  Table A.1 Thermodynamic properties of possible reactions during glycerol APR.1 No Reaction ∆𝐻𝑟0 (kJ/mol) ∆𝑆𝑟0 (kJ/mol K) ln k 25 °C 180 °C 240 °C 300 °C  Liquid Intermediate formationa       1 C3H8O3 (g) + H2 (g)    C3H8O2 (g) +  H2O (g) -98.3 0.12412 54.6 (-135.29) 41.03 (-154.53) 37.98 (-161.97) 35.56 (-169.42) 2 C3H8O3 (g) + ½ H2 (g)    3/2 C2H6O2 (g) -9.5 -0.18796 -18.77 (46.51) -20.09 (75.65) -20.38 (86.92) -20.61 (98.20) 3 C3H8O3 (g)    C2H6O2 (g)+ H2 (g) + CO(g) 76.07 0.24984 -0.65 (2) 9.85 (-37.11) 12.21 (-52.10) 14.08 (-67.09) 4 C3H8O3 (g) + 2 H2 (g)    3/2 C2H5OH (g) + 3/2 H2O (g) -136.1 0.06194 62.38 (-154.56) 43.59 (-164.16) 39.36 (-167.88) 36.02 (-171.59) 5 C3H8O2 (g)    C2H5OH (g) + H2 (g) + CO (g) 89.97 0.16164 -16.87 (41.80) -4.45 (16.74) -1.65 (7.05) 0.56 (-2.64) 6 C2H6O2 (g) +  H2 (g)   C2H5OH (g) +  H2O (g)  -84.4 0.03592 38.39 (-95.10) 26.73 (-100.67) 24.11 (-102.83) 22.04 (-104.98) 7 C3H8O3 (g) + 2 H2 (g)    3 CH3OH (g) -24.2 0.07604 18.91 (-46.86) 15.57 (-58.65) 14.82 (-63.21) 14.23 (-67.77) 206  No Reaction ∆𝐻𝑟0 (kJ/mol) ∆𝑆𝑟0 (kJ/mol K) ln k 25 °C 180 °C 240 °C 300 °C 8 C3H8O2 (g) + H2O (g)   2 CH3OH (g) + CO (g) + H2 (g) 164.57 0.17104 -45.85 (113.60) -23.12 (87.09) -18.01 (76.83) -13.97 (66.56) 9 C2H6O2 (g)+ H2    2 CH3OH (g)  -9.8 0.04532 9.41 (-23.31) 8.05 (-30.33) 7.75 (-33.05) 7.51 (-35.77) 10 C2H5OH (g) +  H2 (g)   CH3OH (g) + CH4 (g)  -40.8 0.01392 18.14 (-44.95) 12.51 (-47.11) 11.24 (-47.94) 10.24 (-48.78)  Glycerol APR       11 C3H8O3 (g) + 3 H2O (g)    3 CO2 (g) + 7 H2 (g) 123.64 0.60743 23.16 (-57.37) 40.01 (-151.53) 44.07 (-187.97) 47.11 (-224.41)  Intermediate products APR (formation of CO2 and H2)       12 CH4 (g) + 2H2O (g)   CO2 (g) + 4 H2 (g) 164.95 0.17025 -46.1 (114.22) -23.32 (87.83) -18.2 (77.61) -14.15 (67.40) 13 CO (g) + H2O (g)   CO2 (g) +  H2 (g) -41.18 -0.04198 11.57 (-28.67) 5.88 (-22.16) 4.61 (-19.644) 3.59 (-17.13) 14 C3H8O2 (g) + 4H2O (g)   3 CO2 (g) + 8 H2 (g) 221.94 0.48331 -31.45 (77.91) -0.8 (3.00) 6.1 (-26.00) 11.54 (-55.00) 15 C2H5OH (g) + 3 H2O (g)   2 CO2 (g) + 6 H2 (g) 173.16 0.36366 -26.15 (64.79) -2.24 (8.42) 3.14 (-13.40) 7.39 (-35.22) 16 CH3OH (g) + H2O (g)   CO2 (g) + 3 H2 (g) 49.28 0.17713 1.41 (-3.50) 8.22 (-30.96) 9.75 (-41.59) 10.96 (-52.22) 17 C2H6O2 (g) + 2 H2O (g)   2 CO2 (g) + 5 H2 (g) 88.76 0.39958 12.24 (-30.31) 24.49 (-92.25) 27.25 (-116.23) 29.43 (-140.20)  Glycerol methanation       18 C3H8O3 (g) + 5 H2 (g)    3 CH4 (g) + 3 H2O (g) -371.21 0.09668 161.46 (-400.02) 110.19 (-415.01) 98.66 (-420.81) 89.55 (-426.61)  Intermediate product methanation       19 CO (g) + 3 H2 (g)   CH4 (g) + H2O (g) -205.87 -0.2146 57.28 (-141.92) 28.85 (-108.66) 22.46 (-95.78) 17.40 (-82.90) 20 C3H8O2 (g) + 4H2(g)   3 CH4 (g) + 2H2O (g) -272.91 -0.02744 106.85 (-264.73) 69.16 (-260.48) 60.69 (-258.83) 53.99 (-257.19) 21 C2H5OH (g) + 2 H2 (g)   2 CH4 (g) + H2O (g) -156.74 0.02316 66.05 (-163.64) 44.4 (-167.23) 39.54 (-168.62) 35.69 (-170.01) 22 CH3OH (g) + H2 (g)   CH4 (g) + H2O (g) -115.67 0.00688 47.51 (-117.72) 31.54 (-118.79) 27.95 (-119.20) 25.11 (-119.61) 23 C2H6O2 (g) + 3 H2 (g)   2 CH4 (g) + 2 H2O (g) -241.14 0.05908 104.44 (-258.75) 71.13 (-267.90) 63.64 (-271.45) 57.72 (-274.99) 0 Standard condition (25°C, 1 atm), 1number in parentheses indicate the ΔGr (kJ mol-1)   207  Table A.2 Thermodynamic properties of possible reaction with phenol No Reaction ∆𝐻𝑟0 (kJ/mol) ∆𝑆𝑟0 (kJ/mol K) ln k  25 °C 180 °C 240 °C 300 °C  Phenol hydrogenation       1 C6H6O (g) + 3 H2 (g)  C6H12O (g) -194.4 -0.35384 35.90 (-88.96) 9.06 (-34.11) 3.01 (-12.89) -1.75 (8.35) 2 C6H6O (g) + 2 H2 (g)  C6H10O (g) -129.7 -0.25476 21.71 (-53.78) 3.8 (-14.29) -0.23 (0.99) -3.42 (16.28) 3 C6H6O (g) + H2 (g)  C6H6 (g)+ H2O (g) -62.8 0.01172 26.76 (-66.29) 18.08 (-68.11) 16.13 (-68.81) 14.59 (-69.52)  Phenol APR       4  C6H6O (g) + 11 H2O (g)  6 CO2 (g) + 14 H2 (g) 395.08 0.71986 -72.88 (180.56) -18.32 (68.98) -6.05 (25.79) 3.65 (-17.40)  APR of phenol hydrogenation products       5 C6H12O (g) + 11 H2O (g)  6 CO2 (g) + 17 H2 (g) 589.48 1.0737 -108.78 (269.52) -273.37 (103.09) -9.07 (38.67) 5.41 (-25.75) 6 C6H10O (g) + 11 H2O (g)  6 CO2 (g) + 16 H2 (g) 524.78 0.97462 -94.59 (234.34) -22.11 (83.27) -5.81 (24.80) 7.07 (-33.68) 7 C6H6 (g) + 12 H2O (g)  6 CO2 (g) + 12 H2 (g) 457.88 0.3161 -146.79 (363.68) -83.55 (314.68) -69.33 (295.72) -58.09 (276.75)  Phenol methanation       8 C6H6O (g) + 10 H2 (g)  6 CH4 (g) + H2O (g) -594.62 -0.30164 203.72 (-504.73) 121.6 (-457.98) 103.13 (-439.88) 88.54 (-421.78)  Methanation of phenol hydrogenation products       9 C6H12O (g) + 7 H2 (g)  6 CH4 (g) + H2O (g) -400.22 0.0522 167.82 (-415.78) 112.54 (-423.87) 100.12 (-427.00) 90.29 (-430.13) 10 C6H10O (g) + 8 H2 (g)  6 CH4 (g) + H2O (g) -464.92 -0.04688 182.01 (-450.95) 117.81 (-443.68) 103.37 (-440.87) 91.95 (-438.06) 11 C6H6 (g) + 9 H2 (g)  6 CH4 (g)  -531.82 -0.31336 176.96 (-438.44) 103.52 (-389.87) 87.00 (-371.07) 73.94 (-352.27) 0 Standard condition (25°C, 1 atm) 1number in parentheses indicate the ΔGr (kJ mol-1)    208  A.3 Thermodynamic prediction of maximum cyclohexanol yield and the MATLAB code Thermodynamic prediction of phenol conversion and cyclohexanol yield at 180 °C to 240 °C with phenol, glycerol and water loading of 10.6 mmol, 5.5 mmol, and 10 g respectively was conducted with several assumptions: 1. All glycerol converted into H2 and CO2 in the reaction: C3H8O3 + 3H2O → 3CO2 + 7H2 2. Cyclohexanol is the only product from phenol hydrogenation 3. Phenol hydrogenation occurs in gas phase. Phenol hydrogenation reaction: C6H6O(g) + 3 H2(g)  C6H12O(g)   ∆𝐻𝑟0 = −194 𝑘𝐽/𝑚𝑜𝑙 ∆𝑆𝑟0 = −0.35384𝑘𝑗𝑚𝑜𝑙 𝐾 𝐾 = [𝐶6𝐻12O][𝐶6𝐻6𝑂] [𝐻2]3  𝐾(𝑇) = 𝑒−∆𝐺(𝑇)/𝑅𝑇  where ∆𝐺 =  ∆𝐻𝑟0 − 𝑇 ∆𝑆𝑟0 Since the only products from phenol conversion is only cyclohexanol then: [𝐶6𝐻12O] = 𝑋 [𝐶6𝐻6𝑂]𝑖    [𝐶6𝐻12𝑂]𝑓 = (1 − 𝑋) [𝐶6𝐻6𝑂]𝑖 [𝐻2]𝑖 = 7 [𝐶6𝐻8𝑂3]𝑖  with assumption that all glycerol converted into H2 and CO2 [𝐻2]𝑓 = [𝐻2]𝑖 − 3 𝑋 [𝐶6𝐻6𝑂]𝑖 Thus, the equation becomes: 𝐾(𝑇){(1 − 𝑋) [𝐶6𝐻6𝑂]𝑖} {[𝐻2]𝑖 − 3 𝑋 [𝐶6𝐻6𝑂]𝑖} = 𝑋 [𝐶6𝐻6𝑂]𝑖 X(T) was obtained by residual minimization as follow 𝐾(𝑇){(1 − 𝑋) [𝐶6𝐻6𝑂]𝑖} {[𝐻2]𝑖 − 3 𝑋 [𝐶6𝐻6𝑂]𝑖} = 𝑋 [𝐶6𝐻6𝑂]𝑖 + res; where res is residual Thus res = 𝐾(𝑇){(1 − 𝑋) [𝐶6𝐻6𝑂]𝑖} {[𝐻2]𝑖 − 3 𝑋 [𝐶6𝐻6𝑂]𝑖} − 𝑋 [𝐶6𝐻6𝑂]𝑖 209  Then X = argminX (res2)  Figure A.1 Predicted maximum cyclohexanol yield and measured cyclohexanol yield for in-situ glycerol APR and phenol hydrogenation. Phenol (10.6 mmol), glycerol (5.5 mmol), water (30.0 g), and Raney Ni® (1.0 g) at 180-240 °C for 120 min. Phenol/Glycerol of 1.9.  Calculation was conducted in Matlab as follow: clear all; clc; p0=0.339131010172987;   %initial concentration of phenol H0=1.21777882426078;    %initial concentration of H2 T0 = 180;               %lower range of temperature T1 = 240;              %upper range of temperature dT = 0.5;               %sensitivity or spacing dH = -194.4;         %enthalpy of phenol to cyclohexanol dS = -0.35384;      %entropy of phenol to cyclohexanol R = 0.008314;       %ideal gas constant  %finding the equilibrium constant noT = (T1-T0)/dT; for a = 1:noT     T(1) = T0;     T(a+1) =T(a)+dT; 210      dG(a) = dH-(T(a)+273)*dS;     dG(a+1)= dH-(T(a+1)+273)*dS;     k(a) = exp(-dG(a)/(R*(T(a)+273)));     k(a+1) = exp(-dG(a+1)/(R*(T(a+1)+273))); end  %finding thermodynamic conversion for i=1:noT+1 objective = @(x) (p0*x-k(i)*(p0-x*p0)*(H0-3*x*p0)^3)^2; x0=0.001; A=[]; B=[]; Aeq=[]; Beq=[]; lb = 0.001; ub = 1; x = fmincon(objective,x0,A,B,Aeq,Beq,lb,ub); X(i) = x; R(i) =(objective(x))^0.5; end k = k.*10^-3; Ans =[T;dG;k;X]; Ans = Ans' plot (T,X)    211  A.4 Predicted saturation pressure by Antoine equation: 𝑙𝑜𝑔10(𝑃) = 𝐴 − (𝐵𝑇+𝐶)  (A.3) Where P is pressure (bar) and T is temperature (K). A, B, and C are component-specific constant.  A, B, and C values as well as P (predicted by Antoine equation) for compounds in IGAPH at 220°C presented in Table A.3  Table A.3 Component-specific constant for Antoine equations Compounds A B C log P P(bar) Methanol 5.15853 1569.613 -34.846 1.7326 54.02 Ethanol 4.92531 1432.526 -61.819 1.6030 40.08 2propanol 4.57795 1221.423 -87.474 1.5660 36.81 Benzene 4.60362 1701.073 20.806 1.2929 19.63 Cyclohexane 4.13983 1316.554 -35.581 1.2616 18.26 1,2-propylene glycol 6.07936 2692.187 -17.94 0.4123 2.58 Ethylene glycol 4.97012 1914.951 -84.996 0.2767 1.89    212  A.5 Predicted of potential CO2 that was converted to CH4   Percentage of potential CO2 that becomes CH4 in section 5.3.6 is calculated as follow:  Consider the carbon balance is 1 and neglecting the amount of CO, and other gaseous alkanes then,  CO2 + CH4 = 1 In the case of closed system APR: 𝐶𝑂2𝐶𝐻4= 1.1  𝐶𝑂21−𝐶𝑂2= 1.1  𝑠𝑜 𝐶𝑂2 = 0.5238; 𝐶𝐻4 = 0.4762  Similarly, in case of open system APR Open APR  : 𝐶𝑂21−𝐶𝑂2= 1.65  𝑠𝑜 𝐶𝑂2 = 0.6226; 𝐶𝐻4 = 0.3774  Closed IGAPH  :  𝐶𝑂21−𝐶𝑂2= 2.28 𝑠𝑜 𝐶𝑂2 = 0.6951; 𝐶𝐻4 = 0.3049 Open IGAPH   : 𝐶𝑂21−𝐶𝑂2= 2.74 𝑠𝑜 𝐶𝑂2 = 0.7326; 𝐶𝐻4 = 0.2674   213  A.6 Estimation of internal mass transfer effect. The catalyst properties for the calculation were taken from data in Table 3.1. The reaction properties were taken from the reaction with 1.0 g phenol, 0.5 g glycerol, 30 g water, and 1.0 g Raney Ni at 200°C (also used to assess external mas transfer effects, Fig C.2). The rate constant for calculating Thiele modulus with respect to glycerol-water binary diffusion was taken from Table B.3. Intradiffusion data of glycerol water was taken from elsewhere.205 Considering glycerol-water binary phase: Known/assumed parameter: 𝜎𝑅𝐴𝑁𝐼 = 0.8 𝜏𝑅𝐴𝑁𝐼 = 3.0 ∅𝑅𝐴𝑁𝐼 = 𝜌𝑅𝐴𝑁𝐼 × 𝑉𝑝𝑜𝑟𝑒,𝑅𝐴𝑁𝐼 = 3.46𝑔𝑚𝐿 × 0.072𝑚𝐿𝑔= 0.249  𝑑𝑝𝑜𝑟,𝑅𝐴𝑁𝐼 = 3.8 𝑛𝑚  𝑑𝑝𝑎𝑟,𝑅𝐴𝑁𝐼 = 389 𝑛𝑚  𝑥𝑔𝑙𝑦−𝑤 = 0.003233  𝑥𝑤−𝑔𝑙𝑦 = 0.990453  𝐷𝑔𝑙𝑦−𝑤 = 7.4 × 10−10 m2 s-1  𝐷𝑤−𝑔𝑙𝑦 = 1.506 × 10−11 m2 s-1  𝐷𝑏𝑢𝑙𝑘,𝑔𝑙𝑦−𝑤 = (𝑥𝑔𝑙𝑦−𝑤𝐷𝑤−𝑔𝑙𝑦) + ( 𝑥𝑤−𝑔𝑙𝑦𝐷𝑔𝑙𝑦−𝑤)  =  7.329 × 10−10 m2 s-1  𝐷𝑒𝑓𝑓,𝑏𝑢𝑙𝑘 = 𝐷𝑏𝑢𝑙𝑘,𝑔𝑙𝑦−𝑤∅𝑅𝐴𝑁𝐼 𝜎𝑅𝐴𝑁𝐼𝜏𝑅𝐴𝑁𝐼= 4.869 × 10−10 m2 s-1  𝐷𝑘𝑛𝑢𝑑𝑠𝑒𝑛,𝑔𝑙𝑦−𝑤 = 𝑑𝑝𝑜𝑟,𝑅𝐴𝑁𝐼3 √(8 𝑅).𝑇𝜋 𝑀𝑤,𝑔= 2.7195 × 10−8  m2 s-1 𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑 = 𝐷𝑘𝑛𝑢𝑑𝑠𝑒𝑛,𝑔𝑙𝑦−𝑤 ∅𝑅𝐴𝑁𝐼 𝜎𝑅𝐴𝑁𝐼𝜏𝑅𝐴𝑁𝐼= 1.8066 × 10−9  m2 s-1 214  1𝐷𝑒𝑓𝑓,𝑡𝑜𝑡 =1𝐷𝑒𝑓𝑓,𝑏𝑢𝑙𝑘+1𝐷𝑒𝑓𝑓,𝑘𝑛𝑢𝑑= 5.2142 × 10−10 m2 s-1 ∅1 =𝑑𝑝𝑎𝑟,𝑅𝐴𝑁𝐼2√𝑘1𝐷𝑒𝑓𝑓,𝑡𝑜𝑡 = 1.61612× 10−4   𝜂 =  3∅12  (∅1𝑐𝑜𝑡ℎ∅1 − 1) = 1.000 𝐶𝑊𝑃 =  𝜂∅12 = 2.611 × 10−8   Since 𝐶𝑊𝑃 ≪ 1, Internal diffusion can be neglected.              215  Appendix B  : Kinetics B.1 Kinetic of glycerol reaction and the MATLAB code 1. Searching of reaction order  MATLAB code for the reaction order searching is as follow: clear all; clc; t = [0 30  60  90  120]; Cgt = xlsread('kineticdata.xlsx','sheet2','w5:w9'); %reading glycerol concentration data Cg = Cgt'; y0 =0.49; %initial power dx = 0.01; % sesitivity x0 = 0.5; %lowest power x1 = 1.5; %highest power nox= (x1-x0)/dx; %number of power   for b = 1:nox+1     y(b) = y0+(dx*b);          if y(b) == 1          v(b,:)=-log(Cg);         p(b,:) = polyfit(t,-log(Cg),1);          else     v(b,:) = (1/(-y(b)+1))*(-Cg.^(-y(b)+1));      p(b,:) = polyfit (t,(1/(-y(b)+1))*(-Cg.^(-y(b)+1)),1);          end     a(b,:) = polyval (p(b,:),t);     r(b,:) = corr2(v(b,:),a(b,:)); %% has been check with excell corr2 equal to linear R2 in excell end nox; y; rr=r; A =[y' rr] plot (y',rr,'-'); xlabel('order'); ylabel('rsquared'); B = max(rr) [I,J] = find(A==B); U= A(I,1) This MATLAB code also applicable for order searching of glycerol during IGAPH.  216  2. Pseudo 1st order kinetics Glycerol kinetics was conducted with pseudo 1st order reaction assuming that all the reactions proceeded with 1st order reaction.  The reaction model is presented in chapter 4 sub chapter 4.3.2. MATLAB code for glycerol kinetics is as follow: clear all clc k0 =[0.00001    0.00001    0.00001    0.00001]; % inital test for initial guess of k km = zeros(4,4); a =1; global CG0 CEG0 CPG0 Cres0 Cex C tspan CG0 = xlsread('kineticdata.xlsx','sheet2','w5'); % data reading for  glycerol conentration at t = 0 CEG0 = xlsread('kineticdata.xlsx','sheet2','x5'); % data reading for ethylene glycol conentration at t = 0  CPG0 = xlsread('kineticdata.xlsx','sheet2','z5'); % data reading for 1,2 porpylene glycol conentration at t = 0  Cres0 = xlsread('kineticdata.xlsx','sheet2','y5');% data reading for further degradation products conentration at t = 0  CexG = xlsread('kineticdata.xlsx','sheet2','w5:w9');% data reading for  glycerol conentration at all t   CexEG = xlsread('kineticdata.xlsx','sheet2','x5:x9');% data reading for ethylene glycol conentration at all t CexPG = xlsread('kineticdata.xlsx','sheet2','z5:z9');% data reading for 1,2 porpylene glycol conentration at all t Cexres = xlsread('kineticdata.xlsx','sheet2','y5:y9');% data reading for urther degradation product conentration at all t Cex = [CexG CexEG Cexres CexPG]; % bundeling the data tspan = (0:30:120); % range of reaction time tested   for a =1:4  % looping for several initial guess of k     k0 = k0.*10; k0 = [0.01 0.01 0.01 0.01]; lb = zeros (1,4); ub = 1000*ones(1,4);   options = optimoptions('lsqnonlin','Algorithm','levenberg-marquardt'); [k,resnorm,res,exitflag,output,lambda,J]=lsqnonlin(@optim,k0,lb,ub,options) Ci = nlparci(k,res,'Jacobian',J) 217  Cex; C; t1=tspan'; plot(t1,C(:,1),'b-',t1,C(:,2),'r--',t1,C(:,3),'k-.',t1,C(:,4),'m:'); xlabel('t (min)'); xticks([0:30:120]); ylabel('Concentration (mmol/mL)'); hold on; plot(t1,Cex(:,1),'ob',t1,Cex(:,2),'+r',t1,Cex(:,3),'*k',t1,Cex(:,4),'^m'); legend('glycerol', 'ethylene glycol', 'further degradation products','1,2-propylene glycol','location','northeast') %hold off; km(a,:) = k; a=a+1; sumCex=Cex(5,1)+Cex(5,2)+Cex(5,3)+Cex(5,4); sumC = C(5,1)+C(5,2)+C(5,3)+C(5,4); end hold off km   function sse=optim(k); global CG0 CEG0 CPG0 Cres0 Cex C tspan [t,C] = ode45(@bal,tspan,[CG0, CEG0, Cres0, CPG0]);   function dCdt=bal(t,C) dCdt=zeros(4,1);   CT=C(1)+C(2)+C(3)+C(4);   CG = CG0*(C(1)/CT); CEG = CG0*(C(2)/CT); CC = CG0*(C(3)/CT); CPG = CG0*(C(4)/CT);   dCdt(1) = -k(1)*CG-k(3)*CG; %k1 = kEG1; k2 = kEG2; k3 = kPG1; k4 = kPG2 dCdt(2) = k(1)*CG-k(2)*CEG; dCdt(3) = k(2)*CEG+k(4)*CPG; dCdt(4) = k(3)*CG-k(4)*CPG;   end Cmod = C; err = (Cmod-Cex).^2; sse = vertcat(err); end 218  The MATLAB code contain looping of several initial guess of reaction constant (k0) to test the robustness of the model. The overlap plot of model and matrix of k depending on the initial guess of reaction guess (k0) will be produced. For glycerol kinetics during APR the robustness of model was shown by the consistent result of k regardless the initial guess of k0 which is shown in Table B.1  Table B.1 Parameter value of glycerol kinetic model during glycerol APR with several initial guess  k value  Initial guess k0 kEG1 kEG2 kPG1 kPG2  k0EG1 k0EG2 k0PG1 k0PG2 0.0056 0.0157 0.0059 0.0206  0.001 0.001 0.001 0.001 0.0056 0.0157 0.0059 0.0206  0.01 0.01 0.01 0.01 0.0056 0.0157 0.0059 0.0206  0.1 0.1 0.1 0.1 0.0056 0.0157 0.0059 0.0206  1 1 1 1  Since the k produce the consistent value, the figure plot also shows the consistent plot as shown in Fig 4.4. The sum of squared error (R2) for glycerol, ethylene glycol, 1,2-propylene glycol, and further degradation products are 0.9991, 0.9707, 0.8397, and 0.9888 respectively. In the case of glycerol kinetic during IGAPH, pseudo 1st order reaction model is not applicable to predict the reaction constant.  The trial of several k0 resulted in the changing of final k value as shown in Table B.2 and Fig B.1. The changing glycerol reaction order to 0.66 might be the reason for the inability of pseudo 1st order reaction to model glycerol reaction during IGAPH.      219  Table B.2 Parameter value of glycerol kinetic model during IGAPH with several initial guess   k value  initial value k 0 kEG1 kEG2 kPG1 kPG2  k0EG1 k0EG2 k0PG1 k0PG2 0.0128 0.0476 0.0138 0.0365  0.00001 0.00001 0.00001 0.00001 0.0052 0.0439 0.02 0.0432  0.0001 0.0001 0.0001 0.0001 0.0066 0.0398 0.0189 0.0432  0.001 0.001 0.001 0.001 0.014 0.0562 0.0113 0.0341  0.01 0.01 0.01 0.01     Figure B.1 Plot of kinetic model of glycerol reaction during IGAPH with pseudo 1st order reaction model.   220  B.2 Kinetics of IGAPH  Table B.3 Reaction rate equations resulting from Langmuir-Hinshelwood model and assumed rate limiting step.    Limiting Step Rate Expression Initial Rate Expression k', Kp, Kg, Kw (depend on the equation)* Adj. R2 for initial Rate SSR Glycerol Adsorption** 𝑟 =  𝐾𝑔 (𝐶𝐶3𝑂3𝐻8 −𝐾𝑑𝐾𝑔𝐶𝐶𝑂2)1 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 𝐾𝑝𝐶𝐶6𝐻6𝑂 +𝐾𝑐𝐶𝐶6𝐻12𝑂 + 3𝐾𝑑𝐶𝐶𝑂2  𝑟 =  𝐾𝑔𝐶𝐶3𝑂3𝐻81 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 𝐾𝑝𝐶𝐶6𝐻6𝑂  Kg =  8.19×102 (8.5×107) Kp = 1×10-14  Kw = 3.3×105(6.8×1010)  -2.762 0.3 Phenol Adsorption 𝑟 =  𝐾𝑝(𝐶𝐶6𝐻6𝑂 −𝐾𝑐𝐾𝑝𝐶𝐶6𝐻12𝑂) 1 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8 + 2𝐾𝑐𝐶𝐶6𝐻12𝑂 + 𝐾𝑑𝐶𝐶𝑂2  𝑟 =  𝐾𝑝(𝐶𝐶6𝐻6𝑂)1 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8  Kg =  4.7×10-2 (3.2×10-1) Kp = 2×10-1 (1.0) Kw = 0 (4.7×10-5)  0.746 2.0×10-2 Water Adsorption 𝑟 =  𝐾𝑤(𝐶𝐻2𝑂 −𝐾𝑑2𝐾𝑤𝐶𝐶𝑂22 ) [1 + 𝐾𝑝𝐶𝐶6𝐻6𝑂 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8 +𝐾𝑐𝐶𝐶6𝐻12𝑂 + 3𝐾𝑑𝐶𝐶𝑂2]2 𝑟 =  𝐾𝑤𝐶𝐻2𝑂[1 + 𝐾𝑝𝐶𝐶6𝐻6𝑂 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8]2  Kg =  2.7×10- 3(1.1×10-1) Kp = 7.7×10-17  Kw = 5.5×10-3 (3.6×10-3)  -1.169 0.2 Glycerol dehydrogenation 𝑟 =  𝑘ℎ𝑦𝐾𝑔𝐶𝐶3𝑂3𝐻81 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8 + 𝐾𝑝𝐶𝐶6𝐻6𝑂 +𝐾𝑐𝐶𝐶6𝐻12𝑂 + 𝐾𝑑𝐶𝐶𝑂2  𝑟 =  𝑘"𝐶𝐶3𝑂3𝐻81 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8 + 𝐾𝑝𝐶𝐶6𝐻6𝑂  k” = 5.8×10-1 (3.7×10-1) Kg =  6.7×102 (1.4×107) -1.062 0.1 221  Limiting Step Rate Expression Initial Rate Expression k', Kp, Kg, Kw (depend on the equation)* Adj. R2 for initial Rate SSR Kp = 1×10-14  Kw = 13.8 (5.8×105)  Phenol Hydrogenation 𝑟=  𝑘′𝐶𝐶6𝐻6𝑂(𝐾𝑔𝐶𝐶3𝑂3𝐻8 + (𝐾𝑤𝐶𝐻2𝑂)12)2 [1 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8 + 2(𝐾𝑤𝐶𝐻2𝑂)12 + 𝐾𝑝𝐶𝐶6𝐻6𝑂 +𝐾𝑐𝐶𝐶6𝐻12𝑂 +𝐾𝑑𝐶𝐶𝑂2 ]3 𝑟 =  𝑘′𝐶𝐶6𝐻6𝑂(𝐾𝑔𝐶𝐶3𝑂3𝐻8 + (𝐾𝑤𝐶𝐻2𝑂)12)2 [1 + 2𝐾𝑔𝐶𝐶3𝑂3𝐻8 + 2(𝐾𝑤𝐶𝐻2𝑂)12 +𝐾𝑝𝐶𝐶6𝐻6𝑂]3 k’ = 85.6 (24.1) Kg =  21.6 (10.2) Kp = 6.2 (1.6) Kw = 2.6×10-3 (0.1) 0.989 5.8×10-4 *Standard errors are shown in parentheses ** Calculation did not converge   222  Derivatization of rate equation: Phenol hydrogenation is limiting step: Other steps reach equilibrium and therefore: Concentration of adsorbed glycerol,  𝑟𝑔 = 𝑘𝑔[𝐶3𝐻8𝑂3] [∗] − 𝑘−𝑔[𝐶3𝐻8𝑂3 ∗] = 0  𝐾𝑔 =[𝐶3𝐻8𝑂3∗][𝐶3𝐻8𝑂3] [∗]  [𝐶3𝐻8𝑂3 ∗] = 𝐾𝑔 [𝐶3𝐻8𝑂3] [∗]  Similarly, Concentration of adsorbed phenol [𝐶6𝐻6O ∗] =  𝑘𝑝[𝐶6𝐻6O][∗]  Concentration of adsorbed hydrogen From water: [H ∗] = 𝐾𝑊12[𝐻2𝑂]12[∗]   From glycerol: [H ∗] = 𝐾𝑔 [𝐶3𝐻8𝑂3] [∗]    Total adsorbed hydrogen: [H ∗] = (𝐾𝑊12[𝐻2𝑂]12 + 𝐾𝑔 [𝐶3𝐻8𝑂3]) [∗]   Total site: 𝐶𝑇 = [∗] + [𝐻 ∗] + [OH ∗] + [𝐶3𝐻8𝑂3 ∗] + [𝐶6𝐻6O ∗] + [𝐶𝑂2 ∗]  𝐶𝑇 = [∗] + 𝐾𝑊12[𝐻2𝑂]12[∗] + 𝐾𝑔 [𝐶3𝐻8𝑂3][∗] + 𝐾𝑊12[𝐻2𝑂]12[∗] + 𝐾𝑔 [𝐶3𝐻8𝑂3][∗] +𝑘𝑝[𝐶6𝐻6O][∗]  + 𝐾𝑑[𝐶𝑂2][∗]  223  𝐶𝑇 = [∗](1 + 2𝐾𝑊12[𝐻2𝑂]12 + 2𝐾𝑔 [𝐶3𝐻8𝑂3] + 𝑘𝑝[𝐶6𝐻6O] + 𝐾𝑑[𝐶𝑂2])  = 1 [∗] =1(1+(2𝐾𝑊[𝐻2𝑂])12+2𝐾𝑔 [𝐶3𝐻8𝑂3]+𝑘𝑝[𝐶6𝐻6O]+𝐾𝑑[𝐶𝑂2])    Phenol hydrogenation is limiting step: 𝑟𝑟𝑥𝑛 =  𝑟𝑃𝐻 = 𝑘ℎ𝑦[𝐶6𝐻6O ∗] [𝐻 ∗]2   𝑟𝑟𝑥𝑛 = 𝑘ℎ𝑦𝑘𝑝[𝐶6𝐻6O][∗] (𝐾𝑊12[𝐻2𝑂]12 + 𝐾𝑔 [𝐶3𝐻8𝑂3])2[∗]2    Rate equation: 𝑟𝑟𝑥𝑛 =  𝑘′[𝐶6𝐻6O] (𝐾𝑔[𝐶6𝐻6O]+(𝐾𝑊[𝐻2𝑂])12)2(1+(2𝐾𝑊[𝐻2𝑂])12+2𝐾𝑔 [𝐶3𝐻8𝑂3]+𝑘𝑝[𝐶6𝐻6O]+𝐾𝑑[𝐶𝑂2])3   Initial rate equation: [𝐶𝑂2] = 0 𝑟𝑟𝑥𝑛 =  𝑘′[𝐶6𝐻6O] (𝐾𝑔[𝐶6𝐻6O]+(𝐾𝑊[𝐻2𝑂])12)2(1+(2𝐾𝑊[𝐻2𝑂])12+2𝐾𝑔 [𝐶3𝐻8𝑂3]+𝑘𝑝[𝐶6𝐻6O])3 224  Appendix C  : Supplementary data  Table C.1 Complete conversion and yield data No Input    Products from glycerol   Products from phenol   Gaseous products Conva (mol%) Yield (molC%)  Conva (mol%) Yield (mol%)  C-balance  Composition (mol%) Res Pres @ 50°C (MPa)j CO2/ CH4 H2netk (mmol) Phenol (g) Glycerol (g) Water (g) Catalyst (g) EGb PGc MOHd EOHe Bf Clg Cnh o-Ci   CO CO2 C1 C2-C6 1 - 2 30 1 73 8 12 1 7 - - - - - 0 1 23 21 1 1 3.02 11 2 0.5 2 30 1 87 2 3 - 2 90 1 75 16 - 102 0 30 21 1 1 3.13 12 3 1 2 30 1 91 1 3 - 2 87 1 60 20 - 94 1 36 20 1 2 3.12 19 4 2 2 30 1 94 1 2 - 2 77 1 49 27 1 101 1 44 18 1 2 3.54 17 5 10 2 30 1 89 1 2 - 1 35 3 14 21 2 105 - 51 16 1 3 3.02 -26 6 1 10 30 1 29 1 4 - 1 17 1 8 13 - 105 2 31 20 1 2 2.83 7 7 1 1 30 1 100 - - - 2 91 3 53 26 - 91 - 31 20 1 2 2.47 -1 8 1 0.5 30 1 100 - - - 0 87 9 39 30 1 92 - 27 9 - 3 1.79 2 9 1 0.25 30 1 100 - - - 2 68 5 28 24 5 94 - 22 4 - 6 1.41 -2 10 0.1 0.1 30 0.1 96 - - - 4 51 4 9 17 - 79 - 6 1 - 7 1.16 4 11 0.5 0.5 30 0.5 99 - 1 - 2 88 4 57 24 - 97 - 22 10 - 2 1.94 4 12 2 2 30 2 99 - 1 - 1 90 2 53 26 1 92 - 45 23 1 2 3.9 13 13 5 5 30 5 91 - 1 - 1 87 11 44 25 1 94 - 48 26 2 2 7.08 23 14 0.5 0.5 30 1 100 - - - - 97 10 43 20 - 76 - 23 13 1 2 2.01 5 15 5 5 30 1 67 2 4 - 1 46 1 20 19 - 94 1 49 19 1 2 3.75 28 16 7.5 7.5 22.5 1 60 2 8 - 3 29 - 3 6 - 80 2 40 20 1 2 3.05 11 17 5 5 30 0.5 47 1 3 - 1 22 - 4 8 - 90 2 34 14 1 2 2.51 34 a conversion b ethylene glycol. c 1,2-propylene glycol. d methanol. e ethanol. f benzene. g cyclohexanol. h cyclohexanone. i o-cresol.  j residual pressure measured after cooling to 50°C. k Potentially produced H2 (Eq 3.14). Reactions were conducted at 220°C for 2h under 690 kPa He pressure.  225  Table C.2 Gas composition on APR and IGAPH in open system in 30 min intervals. Experiments were conducted in continuous 0.03-0.06 L/min He flow at 630 psig with 2.0 g glycerol, 30.0 g of water, and 1.0 g Raney Ni® at 220 °C for 2 h. a Experiments were conducted with addition of 2.0 g phenol. Intervals (min) APR  IGAPHa Composition (mol%) CO2/CH4  Composition (mol%) CO2/CH4 CO CO2 CH4 C2-C6  CO CO2 CH4 C2-C6 0-30 - - 0.22 0.02 -  - - 0.15 0.15 - 30-60 0.39 8.03 5.53 0.12 1.45  - 11.34 5.20 5.38 2.18 60-90 0.39 8.97 5.14 0.11 1.75  - 10.36 3.00 3.18 3.45 90-120 0.31 7.31 3.88 0.07 1.89  - 7.95 2.48 2.48 3.21 average 0.27 6.08 3.69 0.08 1.65  - 7.41 2.70 2.80 2.74    226  Table C.3 Phenol carbon balance and hydrogen balance of reaction in Fig 5.2 T (°C) Phenol  Glycerol H2netj (mmol) Xa  (mol %) Selectivity (%)  C-balance Xa (mol%) Selectivity (%) Bb Clc Cnd o-Ce EGf PGg MOHh EOHi 180 65 2 47 28 1 1.12 92 4 7 1 4 -2 200 90 5 52 32 1 1.02 100 - 1 - 3 -2 220 87 8 40 31 2 0.95 100 - - - 1 0 240 79 8 31 25 4 0.89 100 - - - 1 1 In-situ glycerol APR and phenol hydrogenation at 180 °C to 240 °C for 2 h. Phenol, glycerol, water and Raney Ni® loadings were 10.6 mmol, 5.5 mmol, 30.0 g, and 1.0 g, respectively. Phenol, glycerol, water and Raney Ni® loadings were 10.6 mmol, 5.5 mmol, 30.0 g, and 1.0 g, respectively. a conversion. bbenzene. ccyclohexanol. dcyclohexanone. eo-cresol.  fethylene glycol. g1,2-propylene glycol. hmethanol. iethanol. jPotentially produced H2 (Eq 3.14).   Table C.4 Phenol carbon balance and hydrogen balance of reaction in Fig 5.5 t (min) Phenol  Glycerol H2neti (mmol)  Xa (mol%) Selectivity (%) C-balance Xa (mol%) Selectivity (%) Bb Clc Cnd EGe PGf MOHg EOHh 0 21 - 40 60 0.89 32 12 40 2 13 2 30 42 1 52 48 0.95 52 11 36 4 19 -6 60 51 - 61 39 0.96 75 6 16 2 12 2 90 73 1 63 35 0.88 90 3 8 1 6 35 120 84 2 69 30 0.98 95 3 7 1 7 9 In-situ glycerol APR and phenol hydrogenation at 220 °C as a function of reaction time. Phenol, glycerol, water, and Raney Ni® loading were 2.0 g (21.3 mmol), 2.0 g (21.7 mmol), 30.0 g, and 1.0 g, respectively. a conversion. bbenzene. ccyclohexanol. dcyclohexanone.  eethylene glycol. f1,2-propylene glycol. gmethanol. hethanol. iPotentially produced H2 (Eq 3.14).  227  Table C.5 Phenol carbon balance and hydrogen balance of reaction in Fig 5.10  Catalyst Phenol  Glycerol H2netj (mmol) Xa (mol%) Selectivity (%) C-bal Xa (mol%) Selectivity (%) Bb Clc Cnd o-Ce EGf PGg MOHh EOHi  Raney Ni (1 g)  90 - 2 68 29 0.94 96 4 8 4 10 -17 (5wt%) Pt/C (1 g)  47 4 - 83 13 1.01 50 5 28 9 16 -6 (5wt%) Pd/C (1 g)  3 - - - 100 1.00 - - - - - -5 (5wt%) Ru/C (1 g) 38 - 37 28 35 0.91 60 4 26 6 7 -6 IGAPH reaction with different catalysts at 220 °C for 2 h with 3.45 MPa initial He pressure. Phenol, glycerol, catalyst and water loading were 1.0 g, 1.0 g, 1.0 g, and 30.0 g respectively. a conversion. bbenzene. ccyclohexanol. dcyclohexanone. eo-cresol.  fethylene glycol. g1,2-propylene glycol. hmethanol. iethanol. jpotentially produced H2 (Eq 3.14).   Table C.6 Phenol carbon balance and hydrogen balance of reaction in Fig 5.11 Closed/open Helium  Phenol conversion  C-bal P (psi) Flow (L/min) Xa (mol%) Selectivity (%) Bb Clc Cnd o-Ce Closed 500 - 47 4 - 83 13 1.01 Open 630 - 65 - 2 79 19 0.93 Open 630 0.02-0.06 100 38 4 37 21 0.96 Open 630 0.1-0.2 100 3 3 76 18 1.05 IGAPH reaction with 5 wt% Pt/C in closed and open reactor configurations at 220 °C for 2h. Closed system reaction was conducted with 3.45 MPa initial He pressure. Open system reactions were conducted with constant 4.34 MPa He pressure. The He flow rate was varied from 0 L/min to 0.15-0.25 L/min.  Phenol, glycerol, Pt/C and water loading were 1.0 g, 1.0 g, 1.0 g, and 30.0 g respectively. aconversion. bbenzene. ccyclohexanol. dcyclohexanone. eo-cresol.  228   Table C.7 Phenol carbon balance and hydrogen balance of reaction in Fig 6.1 Catalyst (Acid) Phenol  Glycerol H2netl (mmol) Xa  (mol %) selectivity (mol%) C balance Xa  (mol%) Yield (molC%) Cxb Bc H2netj Cne o-Cf Tg EGh PGi MOHj EOHk  - 89 - 7 48 45 - - 0.79 99 - 1 - 2 -10 Amberlyst-15 88 34 34 15 15 2 - 0.65 88 4 7 3 9 6 H beta E 85 3 24 32 38 3 - 0.68 99 - 1 - 2 -10 H-ZSM-5 air 97 27 66 1 2 - 4 0.71 100 - 1 - 3 -5 IGAPHdo reaction with several acid catalysts at 240 °C for 2 h with 3.45 MPa initial He pressure. Phenol, glycerol, Raney Ni®, acid catalyst and water loading were 1.0 g, 1.0 g, 1.0 g, 0.1 g, and 30.0 g respectively. aconversion. bcylohexane. cbenzene. dcyclohexanol. ecyclohexanone. fo-cresol. gtoluene. hethylene glycol. i1,2-propylene glycol. jmethanol. kethanol. lpotentially produced H2 (Eq 3.14).     229  Table C.8 Phenol carbon balance and hydrogen balance of reaction in Fig 6.2 H-ZSM-5 loading (g) Phenol  Glycerol H2netl (mmol) Xa (mol %) Selectivity (%) C balance  Xa (mol %) Selectivity (%) Cxb Bc Td Cle Cnf o-Cg EGh PGi MOHj EOHk  - 89 - 7 - 48 45 - 0.79 99 - 1 - 2 -10 0.025 85 - 10 - 41 47 2 0.68 100 - 1 - 2 -12 0.050 89 4 28 - 30 38 - 0.90 100 - 1 - 2 -8 0.100 97 27 66 4 1 2 - 0.71 99 - 1 - 3 -13 IGAPHdo reaction H-ZSM-5 at 240 °C for 2 h with 3.45 MPa initial He pressure. Phenol, glycerol, Raney Ni®, and water loading were 1.0 g, 1.0 g, 1.0 g, and 30.0 g respectively. aconversion. bcylohexane. cbenzene. dtoluene. ecyclohexanol. fcyclohexanone. go-cresol. hethylene glycol. i1,2-propylene glycol. jmethanol. kethanol. jpotentially produced H2 (Eq 3.14).230   Figure C.1 Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 180 °C to 240 °C for 2 h. Phenol, glycerol, water and Raney Ni® loadings were 10.6 mmol, 5.5 mmol, 30.0 g, and 1.0 g, respectively.  Panel (a) presents glycerol conversion (equation 3.1, outer left axis), Selectivity of glycerol hydrogenolysis products (equation 3.4, bars, inner left axis), and residual pressure after cooling to 50 °C (right axis).  Panel (b) presents phenol conversion (equation 3.1, outer left axis), the selectivity of phenol hydrogenation products (equation 3.2, bars, inner left axis), and the ratio of CO2(g) to CH4(g) (right axis). 231   Figure C.2  The effect of stirring speed in the IGAPH with Raney Ni. Phenol, glycerol, water and Raney Ni® loadings were 10.6 mmol, 5.5 mmol, 30.0 g, and 1.0 g, 200 ℃ for 2 h   232   Figure C.3 Reaction metrics from in-situ glycerol APR and phenol hydrogenation at 240 °C for 2 h as a function of catalyst stability.  Phenol, glycerol, water and Raney Ni® loadings were 10.6 mmol, 5.5 mmol, 30.0 g, and 1.0 g, respectively. Panel (a) presents glycerol conversion (equation 3.1, outer left axis), Selectivity of glycerol hydrogenolysis products (equation 3.4, bars, inner left axis), and residual pressure after cooling to 50 °C (right axis).  Panel (b) presents phenol conversion (equation 3.1, outer left axis), the selectivity of phenol hydrogenation products (equation 3.2, bars, inner left axis), and the ratio of CO2(g) to CH4(g) (right axis).   233                     Figure C.4 Chromatograph of TCD analysis of gaseous products from glycerol (a), 1,2-propylene glycol (b), ethylene glycol (c), and methanol (d) APR. Experiments were conducted with 2.0 g of reactant, 30.0 g of water and 1.0 g Raney Ni® at 220 °C for 2 h with initial He pressure of 0.69 MPa.          (a) (b) (d) (c) 234  Appendix D  : Supplementary data for FPO upgrading and SPF hydrogenolysis D.1 GC-MS chromatographs                   Figure D.1 Chromatograph and the appearance of upgraded BTG FPO. a chromatograph of the aqueous phase, b chromatograph of the wash solution. The experiment was conducted with 10.0 g of BTG bio-oi, 10.0 g of glycerol, 0.25 f of H-ZSM-5, and 1.0 g of Pt/C at 300°C for 7.0 h. He initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa (a)   Fig. D-1  (b) 235                   Figure D.2 Chromatograph and the appearance of upgraded BTG FPO. a-c chromatograph of the organic phase, d chromatograph of the aqueous phase. The experiment was conducted with 10.0 g of BTG FPO, 5.0 g n-decane, 5.0 g of glycerol, and 1.0 g of Pt/C at 300°C for 2.5h. He initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa     (a) (b) (c) (d) 236                Figure D.3 Chromatograph and the appearance of upgraded BTG FPO. a-c chromatograph of the organic phase, d chromatograph of the aqueous phase. The experiment was conducted with 10.0 g of BTG FPO, 5.0 g acetone, 10.0 g of glycerol, 0.25 g H-ZSM-5 and 1.0 g of Pt/C at 300°C for 7h. He Initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa     (a) (b) (c) (d) 237                      Figure D.4 Chromatograph and the appearance of Avicel® cellulose hydrogenolysis. The experiment was conducted with : 3.00 g Avicel®, 1.00 g Pt/C, 0.25 g H-ZSM-5, 20.00 g water, 5 h (a); 6.00 g Avicel®, 1.00 g Pt/C, 0.25 g H-ZSM-5, 20.00 g water, 0.5 h (b)He Initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa.   (a) (b)   238  D.2 Yield of lignin-derived compounds 1. BTG FPO  Table D.1 Major phenolic compounds in BTG FPO Compounds Wt.% Guaiacol 0.60 p-Creosol 0.68 4-Ethyl guaiacol 0.22 Cerulignol 0.18 Eugenol 0.33 Isoeugenol 0.37   Figure D.5 Yield of phenolics during BTG FPO upgrading (10.0 g) with 5.0 g n-decane at 300°C. He Initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa 239   Figure D.6 Yield of oxygenated cycloaliphatics, naphthenes, and aromatics during BTG FPO upgrading (10.0 g) with 5.0 g n-decane at 300°C. He Initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa  240   Figure D.7 Yield of phenolics during SPF hydrogenolysis (5.0 g) with 5.0 g n-decane and 10 g water at 300°C. He Initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa 241   Figure D.8 Yield of oxygenated cycloaliphatics, naphthenes, and aromatics during SPF hydrogenolysis (5.0 g) with 5.0 g n-decane and 10 g water at 300°C. He Initial pressure was 10.34 MPa, autogenous pressure occurred due to gas production and released on 19.31 MPa   242  D.3 Selectivity of gas product  Table D.2  Selectivity of gas products during BTG FPO upgrading     Selectivity (%)  Pt/C (1.0 g) + H-ZSM-5 (0.25 g) + Glycerol (10.0g), no n-decane (X)  Pt/C (1.0 g) + Glycerol (5.0g), 2.5 h (I)  Pt/C (1.0 g)+ Glycerol (10.0g), 7 h (II)  Pt/C (1.0 g) + H-ZSM-5 (0.25 g)+ Glycerol (10.0g), (III)  Pt/C (2.0 g) + H-ZSM-5 (0.50 g)+Glycerol (10.0g), (IV) CO  -  -  -  -  - CO2  67.5  71.6  72.4  66.5  70.0 C1  21.6  24.7  23.4  25.7  21.3 C2  8.0  2.9  3.3  6.4  6.9 C3  1.9  0.6  0.6  1.1  1.2 C4  0.6  0.1  0.1  0.3  0.4 C5  0.3  -  0.1  -  0.2 C6  0.1  -  -  -  - CO2/CH4   3.1   2.9   3.1   2.6   3.3 Unless otherwise mentioned, the reactions were conducted with 10.0 g BTG FPO and 50.0 g n-decane. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.    243  Table D.3  Selectivity of gas products during SPF hydrogenolysis     Selectivity (%)  Pt/C (1.0 g) + H-ZSM-5 (0.25 g) + Glycerol (10.0g), no n-decane (A)  Pt/C (1.0 g) + H-ZSM-5 (0.25 g) + Glycerol (10.0g), (B)  Pt/C (1.0 g) + Glycerol (10.0g), 7 h (C)  Pt/C (1.0 g) + Glycerol (5.0g), 2.5 h (D) CO  -  -  -  - CO2  82.02  79.56  78.88  74.30 C1  11.16  13.09  17.34  20.63 C2  5.37  6.45  3.15  4.47 C3  0.81  0.67  0.50  0.44 C4  0.49  0.12  0.07  0.12 C5  0.02  0.10  0.03  0.04 C6  0.12  0.01  0.03  - CO2/CH4   7.35   6.08   4.55   3.60 Unless otherwise mentioned, the reactions were conducted with 5.0 g SPF sawdust, 10.0 g water and 50.0 g n-decane. The He initial pressure was 10.34 MPa; autogenous pressure occurred due to gas production and was released at 19.31 MPa.    244  D.4 Mass balance for FPO upgrading and SPF hydrogenolysis  Mass balance for FPO upgrading  Yield of solid from FPO (WS) was calculated based on solid recovered after filtration and overnight drying subtract by initial catalyst loading. Yield of liquid from FPO (WL) was calculated based on the assumption that all glycerol was converted into gaseous products. This assumption was taken since no glycerol was detected in GC-MS and HPLC analysis cannot be conducted. Yield of gas (WG) was calculated by difference.  Solid product (𝑊𝑆) =𝑊𝑆𝐹−𝑊𝐶𝐼𝑊𝐹𝐼× 100%  Liquid product (𝑊𝐿) =𝑊𝐿𝐹−𝑊𝐷𝐼𝑊𝐿𝐹𝐼× 100%  𝑊𝐿𝐹 =𝑊𝑓−𝑊𝑒−𝑊𝑆𝐹𝑊𝐹𝐼  Gas product (𝑊𝐺) = 100 −𝑊𝐿− 𝑊𝑆    Mass balance for SPF hydrogenolysis and Avicel® APR Mass balance of SPF and Avicel® were calculated with similar method. Yield of solid (WS) was calculated based on solid recovered after filtration and overnight drying subtract by initial catalyst loading. Yield of liquid from FPO (WL) was calculated based on the assumption that water was not consumed during the reaction. Yield of gas (WG) was calculated by difference.  Solid product (𝑊𝑆) =𝑊𝑆𝐹−𝑊𝐶𝐼𝑊𝐹𝐼× 100%  Liquid product (𝑊𝐿) =𝑊𝐿𝐹−𝑊𝐷𝐼−𝑊𝑊𝐼−𝑊𝐺𝐹𝑊𝐿𝐹𝐼× 100%  𝑊𝐿𝐹 =𝑊𝑓−𝑊𝑒−𝑊𝑆𝐹𝑊𝐹𝐼  Gas product (𝑊𝐺) = 100 −𝑊𝐿− 𝑊𝑆   245  Where   𝑊𝑆𝐹  : Weight of final solid after vacuum drying at 50°C overnight (g). 𝑊𝐶𝐼  : Weight of initial catalyst loading (g). 𝑊𝐿𝐹 : Weight of final liquid (g). 𝑊𝐷𝐼 : Weight of initial n-decane loading (g).  𝑊𝑓  : Final weight of reactor and the content after the reaction and releasing gas (g). 𝑊𝑒  : Weight of empty reactor (g). 𝑊𝐹𝐼  : Weight of initial FPO loading (g). 𝑊𝑊𝐼 : Weight initial water (g). 𝑊𝐺𝐹 : Weight residual glycerol and liquid products (g).    246  Table D.4 Yield of solid, liquid and gas products of FPO and SPF upgrading   Condition Yield (wt%) Solid Liquid  Gas FPO upgrading Glycerol (10.0g), 7 h no Decane 4 82 13 Glycerol (5.0g), 2.5 h (I) 2 77 21 Glycerol (10.0g), 7 h (II) 1 71 29 Pt/C (1.0 g) + H-ZSM-5 (0.25 g) (III) 2 80 18 Pt/C (2.0 g) + H-ZSM-5 (0.50 g) (IV) 3 66 31      SPF Hydrogenation Glycerol (10.0g), 5 h no Decane 2 62 37 Glycerol (5.0g), 2.5 h (D) 0 65 35 Glycerol (10.0g), 7 h (C) 3 53 45 Pt/C (1.0 g) + H-ZSM-5 (0.25 g) (III) 8 71 22      Avicel®3 APR Avicel (3 g)®, 3 h  0 2 98 Avicel (6 g)®, 0.6  h  0 36 64 1Unless otherwise stated, the reaction was conducted with 10.0 g BTG FPO, 10.0 g glycerol, 5.0 g n-decane, and 1.0 g Pt/C at 300°C. The He initial pressure was 10.34 MPa; autogenous  2Unless otherwise stated, the reaction was conducted with 5.0 g SPF, 10.0 g water, 10.0 g glycerol, 5.0 g n-decane, and 1.0 g Pt/C at 300°C. The He initial pressure was 10.34 MPa; autogenous 3 Unless otherwise stated, the reaction was conducted with 1.0 g Pt/C, 0.25 g H-ZSM-5, and 20.0 g water for 5 h. The He initial pressure was 10.34 MPa; autogenous         247  Appendix E  : Schemes of predicted reaction mechanisms   OOCH3CHCHCH2OHO OCH3HOOOCH3CHCHCH2OHO OCH3OOOCH3CHCHCH2OHO O OCH3H+OH-++   OOCH3CHCHCH2OHHOOOOCH3CHCHCH2OHROOOOCH3HCCHCH2OHO+− OOCH3CHCHCH2OHHSO OCH3OHOCH3CHCHCH2OHSO OCH3OHOCH3CHCHCH2OHSOHOCH3HCCHCH2OHSH+O OCH3SnH2O   OOCH3CHCHCH2OHOHOH2OOOCH3HCCHCH2OHOHOOCH3HCCHCH2OOOOCH3HCHC OCHH O  Figure E.1 Reactivity of β-O-4 linkage in lignin in alkaline solution starting from initiation reaction of non-phenolic end group (a), depolymerization of phenolic end group in the presence of sulphide (b), and enol ether formation that stop the depolymerization (c); adopted from Henriksson.18 b) c) a) 248    HOOOCH3OHOH2OOOCH3OOHOOCH3OOHOCH3OH2CHCOOHOOCH3H+H+ H2O H2COH+H2O H+HCOOOH+HCOOHOH2OHCOOHOOHHHCOOHHOOOCH3OCH3 OCH3H+HCOOHOOHH3CHCOOCH3 HOOOCH3 OCH3H+HOCH3CHHO CHCH2OOCH3OHOH+H2OHCCHCH2OOCH3OHOH+HOOCH3CHCHCH2OOCH3OHOHOH3CO   Figure E.2 Reactivity of β-O-4 linkage of lignin in acidic solution includes depolymerization (a), breaking of LCC (b), and condensation (c); adopted from Henriksson.18    b) c) a) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0387158/manifest

Comment

Related Items