Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of H₂O on CH₄ oxidation over PdO/Al₂O₃ and CeOx/PdO/Al₂O₃ catalysts Alyani, Mina 2016

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

Item Metadata

Download

Media
24-ubc_2016_september_alyani_mina.pdf [ 3.86MB ]
Metadata
JSON: 24-1.0308724.json
JSON-LD: 24-1.0308724-ld.json
RDF/XML (Pretty): 24-1.0308724-rdf.xml
RDF/JSON: 24-1.0308724-rdf.json
Turtle: 24-1.0308724-turtle.txt
N-Triples: 24-1.0308724-rdf-ntriples.txt
Original Record: 24-1.0308724-source.json
Full Text
24-1.0308724-fulltext.txt
Citation
24-1.0308724.ris

Full Text

  Effect of H2O on CH4 Oxidation over PdO/Al2O3 and CeOx/PdO/Al2O3 Catalysts by Mina Alyani M.Sc., Tarbiat Modares University, 2010 B.Sc., University of Tehran, 2007 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)  July 2016 © Mina Alyani, 2016ii  Abstract  Natural gas is a promising alternative fuel for transportation systems because of reduced CO, CO2, SO2, and NOx emissions into the environment, and its abundance and low cost compared with gasoline and diesel. A significant obstacle in the use of NG for vehicle fuels is that CH4 is difficult to oxidize in the presence of CO2 and H2O and at the low exhaust gas temperature (500-550°C) of natural gas vehicles (NGV). Although Pd is the most active catalyst for CH4 oxidation, the presence of H2O suppresses the catalyst activity.  The effect of H2O on the activity of Pd/Al2O3 catalysts with Pd loadings of 0.3, 2.6 and 6.5Pd (wt.%) and corresponding dispersions of 57%, 48%, and 33%, was well described by a kinetic model that accounted for the effect of H2O. Langmuir adsorption was assumed to determine the amount of H2O adsorbed on active sites for the catalysts with different Pd dispersions under wet and dry reaction conditions. The estimated kinetic parameters of apparent activation energy, Ea of 60.611.5 kJ.mol-1 and heat of H2O adsorption,       of   -81.59.1 kJ.mol-1 indicate that CH4 oxidation is independent of Pd dispersion.   Using different preparation methods and varying Ce:Pd ratios, it was found that sequential impregnation of the Al2O3 support by Ce and Pd, with Ce:Pd ratio of 5, yielded a catalyst that had the least inhibition by H2O. H2O adsorption is the dominant mechanism for activity loss, although some sintering of the support may also occur. In a Time-on-Stream (TOS) study with extra H2O added to the feed gas, the chemical and physical properties of the catalysts showed only small changes before and after use. The less negative effect of H2O at higher temperature and at lower H2O concentration was also confirmed by the kinetic study. iii  The kinetic model is consistent with a Langmuir mechanism in which H2O adsorption suppresses C-H bond activation on the active sites. The kinetic analysis shows that the Ce added to the PdO/Al2O3 catalyst suppresses the amount of H2O adsorbed onto the catalyst, thereby reducing the H2O inhibition effect in the presence of Ce.    iv  Preface  This PhD thesis consists of eight chapters that was conducted by Mina Alyani under the supervision of Professor Kevin J. Smith in the Department of Chemical and Biological Engineering at UBC. A version of Chapter 2 has been published in a reviewed journal and a version of Chapters 5 and 7 has been submitted for publication. Chapters 4 and 6 are in preparation to be submitted for publication. The preparation of this dissertation and the papers were all done by Mina Alyani with the final approval of Professor Kevin J. Smith in the Department of Chemical and Biological Engineering at UBC.  The literature review, reactor set-up, catalyst preparation, catalyst characterization, catalyst testing, reaction modeling, kinetics study, and data analysis were done by Mina Alyani under the direct supervision of Professor Kevin J. Smith. Some parts of the catalyst preparation, catalyst characterization, catalyst testing, and collecting data reported in Appendix I were done by Christoph Heinz.  The publications and conference papers included in this thesis are shown below:  Mina Alyani, Kevin J. Smith, A kinetic analysis of the inhibition of CH4 oxidation by H2O on PdO/Al2O3 and CeO2/PdO/Al2O3 catalysts. Ind. Eng. Chem. Res. (2016). DOI: 10.1021/acs.iecr.6b01881; Publication Date (Web): July 7, 2016  Rahman Gholami, Mina Alyani, Kevin J. Smith, Deactivation of Pd Catalysts by Water during Low Temperature Methane Oxidation Relevant to Natural Gas Vehicle Converters. Catalysts 5 (2015) 561-594.  v  This paper was published in Catalysts journal. Sections 2.3 and 3 were prepared and written by Mina Alyani under revision and final approval of Professor Kevin J. Smith. These sections were included in Chapter 2 and Appendix J. Sections 2, 2.1, 2.2, 2.4, and 4 of this paper were prepared and written by Rahman Gholami Shahrestani under revision and final approval of Professor Kevin J. Smith. Sections 1, 5, and Abstract of this paper were a joint contribution of all authors.    Mina Alyani, Kevin J. Smith, Effect of Pd loading on the activity and stability of Pd/Al2O3 catalysts in the presence of H2O. 24th Canadian Symposium on Catalysis, Ottawa, ON (2016).  Mina Alyani, Kevin J. Smith, Kinetic Study of Effect of Water on the Deactivation of Pd-based Catalysts during Methane Oxidation at Low Temperature. 8th International Conference on Environmental Catalysis, Asheville, NC (2014).  Mina Alyani, Kevin J. Smith, Effect of Water on the Deactivation of Pd-based Catalysts during Methane Oxidation at Low Temperature. 23rd Canadian Symposium on Catalysis, Edmonton, AB (2014).    vi  Table of Contents  Abstract ..................................................................................................................................... ii Preface ..................................................................................................................................... iv Table of Contents ..................................................................................................................... vi List of Tables .......................................................................................................................... xii List of Figures ....................................................................................................................... xvii Nomenclature ....................................................................................................................... xxvi List of Abbreviations ........................................................................................................... xxxi Acknowledgments ............................................................................................................. xxxiii Chapter 1: Introduction ............................................................................................................. 1 1.1 Background ............................................................................................................... 1 1.2 Objectives of the Thesis ............................................................................................ 9 1.3 Approach of the Thesis ........................................................................................... 10 Chapter 2: Literature Review.................................................................................................. 14 2.1 Introduction ............................................................................................................. 14 2.2 Effect of the Support on the Inhibiting Effect of H2O during CH4 Oxidation ........ 14 2.2.1 CH4 Conversion on Different Supports ................................................................ 14 2.2.2 High Oxygen Mobility of Support ........................................................................ 17 2.2.3 Effect of Catalyst Structure on Activity ............................................................... 19 vii  2.2.4 Hydrophobicity of the Support .............................................................................. 21 2.3 The Role of CeO2 .................................................................................................... 23 2.4 Structure Sensitivity of the Pd Based Catalysts ...................................................... 30 2.5 Kinetics of H2O Inhibition in CH4 Oxidation ......................................................... 32 2.6 Summary ................................................................................................................. 41 Chapter 3: Experimental ......................................................................................................... 44 3.1 Catalyst Preparation ................................................................................................ 44 3.2 Catalyst Characterization ........................................................................................ 47 3.2.1 Atomic Absorption Spectroscopy ......................................................................... 48 3.2.2 ICP Analysis ............................................................................................................ 48 3.2.3 N2 Adsorption-desorption ...................................................................................... 48 3.2.4 X-ray Diffraction ............................................................................................. 49 3.2.5 X-ray Photoelectron Spectroscopy .................................................................. 49 3.2.6 Time-of-Flight Secondary Ion Mass Spectrometry ......................................... 50 3.2.7 CO Chemisorption ........................................................................................... 50 3.3 Catalyst Testing ....................................................................................................... 51 3.3.1 Experimental Setup ......................................................................................... 51 3.3.2 Temperature Programmed Oxidation .............................................................. 54 3.3.3 Time-on-Stream Experiments ......................................................................... 55 3.4 Catalyst Activity Calculation .................................................................................. 57 viii  3.4.1 CH4 Conversion Calculation ........................................................................... 57 Chapter 4: Effect of Pd Loading on the Activity and Stability of Pd/Al2O3 Catalysts in the Presence of H2O ...................................................................................................................... 59 4.1 Introduction ............................................................................................................. 59 4.2 Results ..................................................................................................................... 59 4.2.1 Catalyst Properties ........................................................................................... 59 4.2.2 Catalyst Activities ........................................................................................... 63 4.3 Kinetic Model .......................................................................................................... 68 4.4 Discussion ............................................................................................................... 80 4.5 Conclusion ............................................................................................................... 83 Chapter 5: Reduced Inhibition of CH4 Oxidation by H2O with CeO2 Addition to the PdO/Al2O3 Catalyst ................................................................................................................ 85 5.1 Introduction ............................................................................................................. 85 5.2 Results ..................................................................................................................... 85 5.2.1 Catalyst Properties ........................................................................................... 85 5.2.2 Catalyst Activities ........................................................................................... 92 5.2.3 Properties of the Used Catalysts .................................................................... 101 5.3 Discussion ............................................................................................................. 106 5.4 Conclusions ........................................................................................................... 109 ix  Chapter 6: Effect of Preparation Method on the Activity and Stability of CeOx/PdO/Al2O3 Catalysts in the Presence of H2O .......................................................................................... 110 6.1 Introduction ........................................................................................................... 110 6.2 Results ................................................................................................................... 111 6.2.1 Catalyst Properties ......................................................................................... 111 6.2.2 Catalyst Activities ......................................................................................... 127 6.3 Discussion ............................................................................................................. 135 6.4 Conclusion ............................................................................................................. 143 Chapter 7: Kinetics of the Inhibition by H2O ....................................................................... 145 7.1 Introduction ........................................................................................................... 145 7.2 Kinetic Model of H2O Inhibition in a Non-steady State System .......................... 146 7.3 Discussion ............................................................................................................. 149 7.4 Conclusion ............................................................................................................. 150 Chapter 8: Conclusions and Recommendations ................................................................... 152 8.1 Conclusions ........................................................................................................... 152 8.2 Recommendations ................................................................................................. 155 8.2.1 Kinetic Model Applied to Co-impregnated and Sequentially Impregnated Catalysts ................................................................................................................. 155 8.2.2 Studying the Effect of CeO2 on O2 Concentration ........................................ 156 8.2.3 Studying the Effect of Support on H2O Adsorption ...................................... 156 x  8.2.4 Studying the Catalytic Properties during CH4 Oxidation Reaction ............... 157 8.2.5 Studying the Partially Reversible Effect of H2O by TPO ............................. 158 Bibliography ......................................................................................................................... 159 Appendices ........................................................................................................................... 175 Appendix A: Catalyst Preparation ........................................................................................ 176 Appendix B: Catalyst Characterization ................................................................................ 178 B.1 BET ....................................................................................................................... 178 B.2 XRD ...................................................................................................................... 179 B.3 XPS ....................................................................................................................... 180 B.4 CO Chemisorption ................................................................................................ 181 Appendix C: MFC and MS Calibration ................................................................................ 184 C.1 MFC Calibration ................................................................................................... 184 C.2 MS Calibration ..................................................................................................... 185 C.3 Liquid Pump Calibration ...................................................................................... 187 Appendix D: Error Analysis ................................................................................................. 189 Appendix E: Reaction System .............................................................................................. 192 E.1 CH4 Conversion Calculation ................................................................................. 192 Appendix F: Repeatability .................................................................................................... 200 F.1 TPO Reaction Repeatability.................................................................................. 200 F.2 TOS Reaction Repeatability.................................................................................. 201 xi  Appendix G: Supplementary Figures and Tables for Chapter 6........................................... 202 Appendix H: Mass Transfer Effects ..................................................................................... 219 H.1 Internal Mass Transfer Calculation ...................................................................... 219 H.2 External Mass Transfer Calculation ..................................................................... 223 H.3 Pressure Drop Calculation over Catalyst Bed ...................................................... 225 Appendix I: CH4 Oxidation over PdO-ZrOx/Al2O3 in the Presence of H2O ........................ 226 I.1 Catalyst Properties ................................................................................................. 226 I.2 Catalyst Activities .................................................................................................. 229 I.3 Discussion .............................................................................................................. 234 I.4 Conclusion ............................................................................................................. 241 Appendix J: The Effect of Second Metal on Pd Catalysts for CH4 Oxidation ..................... 243 Appendix K: MATLAB M-files Code ................................................................................. 251   xii  List of Tables Table 1.1. Natural gas composition .......................................................................................... 1 Table 1.2. Exhaust emission limits for Light-Duty vehicles .................................................... 2 Table 1.3. Fuel economy and exhaust gas compositions using CNG and gasoline.................. 3 Table 2.1. Comparing light-off temperature (T30) for CH4 oxidation over Pd supported catalysts................................................................................................................................... 15 Table 2.2. Effect of support on properties of 5wt.%Pd catalysts and their CH4 oxidation conversion ............................................................................................................................... 20 Table 2.3. Lattice expansion and oxygen storage capacity of CeO2 as a function of crystallite size .......................................................................................................................................... 26 Table 2.4. The amount of oxygen adsorption/desorption on IWI and SCS samples per gram of catalyst ................................................................................................................................ 27 Table 2.5. Apparent activation energy and order of CH4 combustion reaction over Pd catalysts................................................................................................................................... 33 Table 2.6. Estimated values of Ea and ΔHads .......................................................................... 36 Table 3.1. A comparison of the reaction conditions used in the present study and real NGV operating condition ................................................................................................................. 57 Table 4.1. Properties of PdO catalysts with different loadings of Pd over Al2O3 .................. 60 Table 4.2. Pd 3d spectra for catalysts with different loadings of Pd ...................................... 62 Table 4.3. Light-off temperatures for 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts ................................................................................................................................................ 64 Table 4.4. The constant values of ν0,       , and      used in Equation 4.3 ............................ 70 Table 4.5. Constant values for 6.5Pd/Al2O3 catalyst at T=330°C used in Equation 4.20 ...... 76 xiii  Table 4.6. Estimated values of η, rate constant, equilibrium constant for H2O adsorption, and reaction rate at different temperatures for 6.5Pd/Al2O3 catalyst ............................................ 76 Table 4.7. Estimated values obtained from the design equation for CH4 oxidation over Pd/Al2O3 catalysts with different Pd loadings ........................................................................ 77 Table 4.8. Concentration of H2O per number of active sites as a function of temperature for catalysts with different Pd loadings ........................................................................................ 79 Table 4.9. ∆Xdry-wet at t=24h for 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts ....... 80 Table 5.1. Properties of calcined PdO, CeO2, and co-xCe/yPd catalysts supported on Al2O3 ................................................................................................................................................ 86 Table 5.2. Pd 3d spectra for Pd/Al2O3 and co-xCe/yPd/Al2O3 with different loadings of Ce ................................................................................................................................................ 90 Table 5.3. Light-off temperatures for 6.5Pd/Al2O3 and co-xCe/yPd/Al2O3 catalysts ............ 93 Table 5.4. Constant values for co-2.9Ce/6.5Pd/Al2O3 catalyst at T=330°C used in Equation 4.20 ......................................................................................................................................... 97 Table 5.5. Estimated values of η, rate constant, equilibrium constant for H2O adsorption, and reaction rate at different temperatures for co-2.9Ce/6.5Pd/Al2O3 catalyst ............................. 97 Table 5.6. Compared estimated values obtained from the design equation for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts ....................................................................................... 98 Table 5.7. Rate of deactivation for 6.5Pd/Al2O3 catalyst as a function of temperature ....... 100 Table 5.8. Rate of deactivation for co-2.9Ce/6.5Pd/Al2O3 catalyst as a function of temperature ........................................................................................................................... 100 Table 5.9.     and        ratio for 6.5Pd/Al2O3 catalyst and co-2.9Ce/6.5Pd/Al2O3 catalyst .............................................................................................................................................. 101 xiv  Table 5.10. Properties of fresh and used catalysts after TOS experiment for 24h in wet condition at T=350°C ........................................................................................................... 103 Table 6.1. Effect of adding Ce on the surface composition ratio obtained by ToF-SIMS ... 117 Table 6.2. Ce 3d peaks and               ratio for sequentially impregnated catalysts with different loadings of Ce ........................................................................................................ 122 Table 6.3. CeO2 and PdO crystallite size of calcined co-impregnated and sequentially impregnated catalysts and xCe/Al2O3 supports .................................................................... 126 Table 6.4. O2 chemisorption on CeO2/Al2O3 samples with different loadings of CeO2 ...... 138 Table 7.1. Rate constant for H2O desorption obtained by the proposed kinetic model in Equation 7.4 .......................................................................................................................... 148 Table A.1. Required amounts of N2O6Pd.xH2O salt, Ce(NO3)3.6H2O salt, and Al2O3 support .............................................................................................................................................. 177 Table B.1. Properties of co-2.9Ce/6.5Pd/Al2O3 catalyst used for CO chemisorption analysis .............................................................................................................................................. 182 Table B.2. Cumulative volume of co-2.9Ce/6.5Pd/Al2O3 catalyst during CO chemisorption analysis ................................................................................................................................. 183 Table C.1. CH4/Ar calibration using a bubble flow meter ................................................... 184 Table C.2. He calibration using a bubble flow meter ........................................................... 184 Table C.3. Calibration equations obtained from the data presented in Tables C.1-C.2 ....... 185 Table C.4. MS calibration for 9.97%CH4/He using 185 sccm He and 50 sccm O2 ............. 186 Table C.5. Harvard apparatus syringe pump (Model 44) calibration ................................... 188 Table D.1. Catalyst preparation repeatability ....................................................................... 189 Table D.2. BET analysis repeatability .................................................................................. 190 xv  Table D.3. CO Uptake analysis repeatability ....................................................................... 190 Table D.4. XPS analysis repeatability .................................................................................. 191 Table D.5. XRD analysis repeatability ................................................................................. 191 Table E.1. CH4 conversion calculation for 6.5Pd/Al2O3 catalyst during TPO experiment. Reaction condition: GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar........................................................................................................... 194 Table F.1. TPO Reaction Repeatability. GHSV=180,000 cm3(STP).gcat-1.h-1. 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar ............................................................................. 200 Table F.2. TOS repeatability. GHSV=180,000 cm3(STP).gcat-1.h-1. 5000 ppm CH4, 20(v/v)% O2, and the balance He.......................................................................................................... 201 Table G.1. Properties of calcined catalysts prepared by co-impregnation and sequentially impregnation methods .......................................................................................................... 216 Table G.2. Light-off temperatures for 3.4Pd/Al2O3, co-impregnated and sequentially impregnated catalysts. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar........................................................................................................... 217 Table G.3. Ce 3d peaks and               ratio for co-impregnated catalysts with different loadings of Ce ....................................................................................................................... 218 Table G.4. Ce 3d peaks and               ratio for xCe/Al2O3 supports with different loadings of Ce .......................................................................................................................................... 218 Table H.1. Physical properties of catalyst bed consists of 6.5Pd/Al2O3............................... 220 Table H.2. Operating condition for TOS experiment using 6.5Pd/Al2O3 catalyst ............... 222 Table H.3. Details of calculations for Mears criterion factor for 6.5Pd/Al2O3 catalyst at T=330°C ............................................................................................................................... 224 xvi  Table I.1. Properties of calcined 3.4Pd/Al2O3 and sequential impregnated catalysts with different loadings of Zr ......................................................................................................... 229 Table I.2. Light-off temperatures for 3.4Pd/Al2O3 and sequential impregnated catalysts with different loadings of Zr ......................................................................................................... 230 Table J.1. Changes in Pd and Pt-Pd catalyst properties before and after aging ................... 248 Table J.2. T50 for fresh and steam aged Pd and Pt-Pd catalysts operated in dry and wet feed. Combustion conditions: 4067 vol. ppm CH4; total flow rate of 234.5 cm3.min-1; 500 mg catalyst; 5vol.% water in wet feed ........................................................................................ 250    xvii  List of Figures  Figure 1.1. Average prices for gasoline, diesel, and CNG over time ....................................... 2 Figure 1.2. The three-way catalytic converter installed in the exhaust gas emitted from gasoline engines ........................................................................................................................ 5 Figure 1.3. The concentration of NOx, CO, and HCs as a function of air /fuel ratio operating in gasoline engines .................................................................................................................... 6 Figure 1.4. Possible mechanisms of catalyst activity loss by H2O: (a) H2O adsorption on PdO active sites, (b) formation of inactive Pd(OH)2 ........................................................................ 8 Figure 2.1. Catalytic combustion of CH4 over 1.1wt.%Pd/SnO2 with different amounts of H2O added (vol.%). Reaction conditions: 1vol.%CH4, 20vol.%O2, 0-20vol.%H2O, balanced in N2. ....................................................................................................................................... 16 Figure 2.2. Catalytic combustion of CH4 over 1.1wt.%Pd/Al2O3 with different amounts of H2O added (vol.%). Reaction conditions: 1vol.%CH4, 20vol.%O2, 0-20vol.%H2O, balanced in N2. ....................................................................................................................................... 17 Figure 2.3. Methane conversion for Pd/ZrO2 and Pd/Aerosil130 catalysts. Reaction conditions: 1.5%CH4; 6%O2; total flow=90cm3.min-1, balanced in He; temperature=325°C; catalyst mass= 0.2g ................................................................................................................. 18 Figure 2.4. A) Pd/Aerosil130 catalyst, B) Pd/R972 catalyst. Reaction conditions: total flow=90cm3 (STP).min-1, temperature=325°C; catalyst mass=0.2 g. ..................................... 22 Figure 2.5. Ce2O3 lattice unit cells (a) and CeO2 (b). Blue spheres represent the cerium, red and white spheres are defined as oxygen atoms and vacancies, respectively. ....................... 24 Figure 2.6. The formation of oxygen vacancy for (a) a CeO2 crystal, (b) a pair of Ce3+. Black spheres indicate the Ce3+ and  indicates oxygen vacancy ................................................... 25 V xviii  Figure 2.7. Raman spectra for (a) CeO2, (b) Pd catalyst prepared by impregnation method, and (c) Pd catalyst prepared by deposition-precipitation method .......................................... 29 Figure 2.8. Oxygen exchange mechanism for CH4 oxidation using labeled (18O16O) pulsed experiments ............................................................................................................................. 37 Figure 2.9. Effect of O2 pressure on the CH4 oxidation rate constant over 0.2wt.%Pd/Al2O3 catalyst at 873K (4.8 nm (●, ▲) and 21.3 nm ( , ■) Pd cluster diameter) ........................... 39 Figure 3.1. Schematic diagram of CH4 oxidation setup ......................................................... 53 Figure 4.1. XRD patterns for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, and (c) 6.5Pd/Al2O3.          ∆ PdO, ● Al2O3 ....................................................................................................................... 61 Figure 4.2. XPS Pd 3d spectra measured for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, and (c) 6.5Pd/Al2O3 ............................................................................................................................ 62 Figure 4.3. Temperature Programmed Oxidation profile. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppmv CH4, 20(v/v)% O2, and the balance He and Ar ................................................... 63 Figure 4.4. TOS results for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, (c) 6.5Pd/Al2O3 at T=330°C for dry (open symbol) and wet (closed symbol) conditions. GHSV=180,000 cm3(STP).gcat-1.h-1, 5000 ppmv CH4, 20(v/v)% O2, and the balance He ....................................................... 66 Figure 4.5. TOS results for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, (c) 6.5Pd/Al2O3 at T=350°C for dry (open symbol) and wet (closed symbol) conditions. GHSV=180,000 cm3(STP).gcat-1.h-1, 5000 ppmv CH4, 20(v/v)% O2, and the balance He ....................................................... 67 Figure 4.6. Schematic of the reactor used in the CH4 oxidation process................................ 69 Figure 4.7. Calculated      values from the kinetic model versus measured       values from the experiments for xPd/Al2O3 catalysts ........................................................................ 77 xix  Figure 5.1. XRD patterns for (a) γ-Al2O3 (b) 6.5Pd/Al2O3, (c) co-0.9Ce/6.5Pd/Al2O3, (d) co-2.9Ce/6.5Pd/Al2O3, (e) co-4.8Ce/6.5Pd/Al2O3, (f) co-9.5Ce/6.5Pd/Al2O3, (g) 9.4Ce/Al2O3.   ∆ PdO, ● Al2O3, ○ CeO2, ■ Ce2O3 ......................................................................................... 87 Figure 5.2. Measured Pd atomic percent (○) and Ce atomic percent (■) on the catalyst surface as a function of calculated (Ce/Al)b ........................................................................... 88 Figure 5.3. XPS Pd 3d spectra measured for (a) 6.5Pd/Al2O3, (b) co-0.9Ce/6.5Pd/Al2O3, (c) co-2.9Ce/6.5Pd/Al2O3, (d) co-4.8Ce/6.5Pd/Al2O3, (e) co-9.5Ce/6.5Pd/Al2O3....................... 89 Figure 5.4. XPS Ce 3d spectra measured for (a) co-2.9Ce/6.5Pd/Al2O3, (b) co-4.8Ce/6.5Pd/Al2O3, (c) co-9.5Ce/6.5Pd/Al2O3, (d) 9.4Ce/Al2O3 ........................................... 91 Figure 5.5. Temperature Programmed Oxidation profile. Effect of different loadings of Ce on the initial activity of 6.5Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar ...................... 93 Figure 5.6. TOS results for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts at different temperatures for dry (open symbol) and wet (closed symbol) conditions. GHSV=180,000 cm3(STP).gcat-1.h-1. 5000 ppm CH4, 20(v/v)% O2, and the balance He. Top: 6.5Pd/Al2O3 at (a) T=300°C, (b) T=330°C, (c) T=350°C, Bottom: co-2.9Ce/6.5Pd/Al2O3 at (d) T=330°C, (e) T=350°C, (f) T=380°C ........................................................................................................... 95 Figure 5.7. Calculated  XCH4 values from the kinetic model versus measured  XCH4 values from the experiments for co-2.9Ce/6.5Pd/Al2O3 catalyst ....................................................... 96 Figure 5.8. Calculated        versus      for (a) 6.5Pd/Al2O3 catalyst, and (b) co-2.9Ce/6.5Pd/Al2O3 catalyst ..................................................................................................... 99 xx  Figure 5.9. XRD for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 for both fresh and used catalysts for 24h TOS. (a) Fresh 6.5Pd/Al2O3, (b) Used 6.5Pd/Al2O3, (c) Fresh co-2.9Ce/6.5Pd/Al2O3, (d) Used co-2.9Ce/6.5Pd/Al2O3. ∆ PdO, ● Al2O3 ................................................................ 104 Figure 5.10. XPS binding energy for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 for both fresh and used catalysts for 24h TOS. (a) Fresh 6.5Pd/Al2O3, (b) Used 6.5Pd/Al2O3, (c) Fresh co-2.9Ce/6.5Pd/Al2O3, (d) Used co-2.9Ce/6.5Pd/Al2O3 ........................................................... 105 Figure 6.1. Effect of Ce loading on BET surface area, pore size, and pore volume for co-impregnated catalysts (□), sequentially impregnated catalysts (), and xCe/Al2O3 supports (∆) ......................................................................................................................................... 112 Figure 6.2. Pd atomic percent on the surface of co-impregnated (■) and sequentially impregnated (○) catalysts as a function of (Ce/Al)b ............................................................. 113 Figure 6.3. Ce atomic percent on the surface of co-impregnated catalysts (■), sequentially impregnated catalysts (○), and xCe/Al2O3 supports (∆) as a function of (Ce/Al)b ............... 114 Figure 6.4. Al atomic percent on the surface of co-impregnated catalysts (■), sequentially impregnated catalysts (○), and xCe/Al2O3 supports (∆) as a function of (Ce/Al)b ............... 115 Figure 6.5. ToF-SIMS analysis for seq-17Ce/3.4Pd/Al2O3 catalyst ..................................... 116 Figure 6.6. XPS Pd 3d spectra measured for (a) 3.4Pd/Al2O3 and co-impregnated (b) co-2Ce/3.4Pd/ Al2O3, (c) co-14Ce3.4Pd/Al2O3, and (d) co-47Ce/3.4Pd/Al2O3 catalysts ......... 118 Figure 6.7. XPS Pd 3d spectra measured for (a) 3.4Pd/Al2O3 and sequentially impregnated (b) seq-2Ce/3.4Pd/Al2O3, (c) seq-17Ce/3.4Pd/Al2O3, and (d) seq-57Ce/3.4Pd/Al2O3 catalysts .............................................................................................................................................. 119 Figure 6.8. Ce 3d for sequentially impregnated (a) seq-17Ce/3.4Pd/Al2O3, (b) seq-28Ce/3.4Pd/Al2O3, (c) seq-57Ce/3.4Pd/Al2O3 catalysts....................................................... 121 xxi  Figure 6.9.               ratio obtained by XPS analysis for co-impregnated catalysts (∆), sequentially impregnated catalysts (□), and xCe/Al2O3 supports () as a function of varying loadings of Ce ....................................................................................................................... 123 Figure 6.10. XRD patterns for co-impregnated catalysts (a) 3.4Pd/Al2O3, (b) co-2Ce/3.4Pd/Al2O3, (c) co-14Ce/3.4Pd/Al2O3, and (d) co-47Ce/3.4Pd/Al2O3 catalysts. ∆ PdO, ● Al2O3, ○ CeO2 ................................................................................................................... 124 Figure 6.11. XRD patterns for sequentially impregnated (a) 3.4Pd/Al2O3, (b) seq-2Ce/3.4Pd/Al2O3, (c) seq-17Ce/3.4Pd/Al2O3, and (d) seq-57Ce/3.4Pd/Al2O3 catalysts. ∆ PdO, ● Al2O3, ○ CeO2 .......................................................................................................... 125 Figure 6.12. Temperature Programmed Oxidation profile for co-impregnated catalysts. Effect of different loadings of Ce on the initial activity of 3.4Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .............................................................................................................................................. 127 Figure 6.13. Temperature Programmed Oxidation profile for sequentially impregnated catalysts. Effect of different loadings of Ce on the initial activity of 3.4Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .................................................................................................... 128 Figure 6.14. Rate of catalyst deactivation (     ) as a function of Ce loading for co-impregnated catalysts (Δ), and sequentially impregnated catalysts (●) at T=350°C and (■) at T=320°C. Obtained from dry-TOS results. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar..................................................................... 130 Figure 6.15. Rate of catalyst deactivation (   ) as a function of Ce loading for co-impregnated catalysts (○), and sequentially impregnated catalysts (■). Obtained from wet-xxii  TOS results at T=350°C and 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar..................................................................... 131 Figure 6.16. Wet-TOS results for seq-17Ce/3.4Pd/Al2O3 and 3.4Pd/Al2O3 catalysts with 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar. (a) T=310°C, (b) T=330°C, (c) T=350°C, and (d) T=370°C................ 133 Figure 6.17. Wet-TOS results for seq-17Ce/3.4Pd/Al2O3 and 3.4Pd/Al2O3 catalysts at T=350°C and (a) 1vol.% H2O, (b) 2vol.% H2O, and (c) 5vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .................... 134 Figure 6.18. Oxygen exchange mechanism of the PdOx (PdO phase formed during the temperature programmed isotopic exchange) ....................................................................... 141 Figure 6.19. Possible oxygen exchange mechanism based on the activity results for 3.4Pd/Al2O3 and seq-17Ce/3.4Pd/Al2O3 catalysts. Oxygen exchange between the PdO and oxygen vacancy (1), between the PdO and gas phase (2), and between the oxygen vacancy and bulk oxide support (3) .................................................................................................... 143 Figure 7.1. Fitting Equation 7.4 to the experimental results for the wet-TOS with 5vol.%H2O at T=300°C (■), 330°C (∆), and 350°C (●) for 6.5Pd/Al2O3 catalyst .................................. 147 Figure 7.2. Fitting Equation 7.4 to the experimental results for the wet-TOS with 5vol.%H2O at T=330°C (■), 350°C (∆), and 380°C (●) for co-2.9Ce/6.5Pd/Al2O3 catalyst .................. 148 Figure 7.3. ln   values as a function of       for 0.3Pd/Al2O3 catalyst (▲), 2.6Pd/Al2O3 catalyst (∆), 6.5Pd/Al2O3 catalyst (■), and co-2.9Ce/6.5Pd/Al2O3 catalyst (○) ................... 149 Figure B.1. Isotherm linear plot for calcined 6.5wt.%Pd/Al2O3 catalyst ............................. 179 Figure C.1. MFC Calibration equation obtained for 9.97(v/v)%CH4/Ar ............................. 185 Figure C.2. MS Calibration equation for CH4 ...................................................................... 187 xxiii  Figure G.1. Ce 3d for co-impregnated catalysts (a) co-14Ce/3.4Pd/Al2O3, (b) co-47Ce/3.4Pd/Al2O3 catalysts .................................................................................................. 203 Figure G.2. Ce 3d for (a) 16Ce/Al2O3 (b) 26Ce/Al2O3, and (c) 52Ce/Al2O3 supports ......... 204 Figure G.3. N2 adsorption-desorption isotherms for co-impregnated (a) co-2Ce/3.4Pd/Al2O3, (b) co-14Ce/3.4Pd/Al2O3, and (c) co-47Ce/3.4Pd/Al2O3 catalysts....................................... 205 Figure G.4. N2 adsorption-desorption isotherms for sequentially impregnated (a) seq-2Ce/3.4Pd/Al2O3, (b) seq-6Ce/3.4Pd/Al2O3, (c) seq-17Ce/3.4Pd/Al2O3, (d) seq-28Ce/3.4Pd/Al2O3, and (e) seq-57Ce/3.4Pd/Al2O3 catalysts ................................................ 206 Figure G.5. N2 adsorption-desorption isotherms for (a) 2Ce/Al2O3, (b) 5Ce/Al2O3, (c) 16Ce/Al2O3, (d) 26Ce/Al2O3, and (e) 52Ce/Al2O3 supports................................................. 207 Figure G.6. BJH pore size distribution for co-2Ce/3.4Pd/Al2O3 (○), co-14Ce/3.4Pd/Al2O3 (■), and co-47Ce/3.4Pd/Al2O3 (▲) catalysts ....................................................................... 208 Figure G.7. BJH pore size distribution for (○) seq-2Ce/3.4Pd/Al2O3, (●) seq-6Ce/3.4Pd/Al2O3, (▲) seq-17Ce/3.4Pd/Al2O3, (▼) seq-28Ce/3.4Pd/Al2O3, and (□) seq-57Ce/3.4Pd/Al2O3 catalysts .................................................................................................. 209 Figure G.8. BJH pore size distribution for (○) 2Ce/Al2O3, (●) 5Ce/Al2O3, (▲) 16Ce/Al2O3, (▼) 26Ce/Al2O3, and (□) 52Ce/Al2O3 supports ................................................................... 210 Figure G.9. Dry-TOS results for 3.4Pd/Al2O3 (a) and (b) co-2Ce/3.4Pd/Al2O3, (c) co-14Ce/3.4Pd/Al2O3, and (d) co-47Ce/3.4Pd/Al2O3 at 350°C. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar ........................................... 211 Figure G.10. Dry-TOS results for 3.4Pd/Al2O3 (a) and (b) seq-2Ce/3.4Pd/Al2O3, (c) seq-6Ce/3.4Pd/Al2O3, (d) seq-17Ce/3.4Pd/Al2O3, (e) seq-28Ce/3.4Pd/Al2O3, and (f) seq-xxiv  57Ce/3.4Pd/Al2O3 at 350°C. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar.............................................................................................. 212 Figure G.11. Wet-TOS results for seq-17Ce/3.4Pd/Al2O3 and 3.4Pd/Al2O3 catalysts with 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar. (a) T=310°C, (b) T=330°C, (c) T=350°C, and (d) T=370°C................ 213 Figure G.12. ToF-SIMS analysis for 3.4Pd/Al2O3 catalyst .................................................. 214 Figure G.13. ToF-SIMS analysis for co-14Ce/3.4Pd/Al2O3 catalyst ................................... 215 Figure I.1. XPS analysis, Zr on the surface (square), Pd on the surface (cross) .................. 227 Figure I.2. XRD patterns for 3.4Pd/Al2O3 (a) and sequential impregnated catalysts with different loadings of Zr: (b) seq-1.5Zr/3.4Pd/Al2O3, (c) seq-15Zr/3.4Pd/Al2O3, (d) and seq-25Zr/3.4Pd/Al2O3. ∆ PdO, ● Al2O3, ○ ZrO2 ......................................................................... 228 Figure I.3. Temperature Programmed Oxidation profile. Effect of different loadings of Zr on the initial activity of 3.4Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .................... 229 Figure I.4. Wet-TOS results for (■) 3.4Pd/Al2O3, (∆) seq-1.5Zr/3.4Pd/Al2O3, () seq-15Zr/3.4Pd/Al2O3, and (○) seq-25Zr/3.4Pd/Al2O3 catalysts at 350 °C with 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .............................................................................................................................................. 231 Figure I.5. Wet-TOS results for (■) 3.4Pd/Al2O3, (∆) seq-1.5Zr/3.4Pd/Al2O3, () seq-15Zr/3.4Pd/Al2O3, and (○) seq-25Zr/3.4Pd/Al2O3 catalysts at 350°C with 5vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .............................................................................................................................................. 232 xxv  Figure I.6. Wet-TOS results for (■) 3.4Pd/Al2O3, (∆) seq-1.5Zr/3.4Pd/Al2O3, () seq-15Zr/3.4Pd/Al2O3, and (○) seq-25Zr/3.4Pd/Al2O3 catalysts at 425°C with 10vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar .............................................................................................................................................. 235 Figure I.7. Dry-TOS results for seq-1.5Zr/3.4Pd/Al2O3 catalyst at (a) 310°C, (b) 280°C, (c) 250°C, and (d) 310°C. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar........................................................................................................... 239 Figure J.1. High-temperature in situ XRD profiles of PdPt-Al2O3 during heating .............. 246 Figure J.2. CH4 combustion rate at 800°C with time on stream. Combustion conditions: CH4=1vol.%, air=99vol.%, CH4/air flow= 450 L.h-1, catalyst weight= 0.3g. Catalyst 1, 2, 3, and 4 represent 18.7wt.% Pd, 18.4wt.%Pd-1.6wt.% Pt, 18.1wt.% Pd-3.4wt.% Pt, and 18.0wt.% Pd-3.9wt.% Pt over Al2O3 catalysts ..................................................................... 247    xxvi  Nomenclature  A Cross sectional area of the reactor (m2) Als Al surface composition (at.%) C   BET Constant (-) CT Total number of active site (molsite.gcat-1) CCH4                        Bulk concentration of CH4 (mol.m-3) CbCH4                        Bulk concentration of CH4 (mol.m-3) CsCH4 Concentration of CH4 on the catalyst surface (mol.m-3)     Mears criterion factor (-) (Ce/Al)b                  Ce:Al bulk atom ratio (      ) Ces                  Ce surface composition (at.%)  d                             Distance between two planes of atoms in X-ray diffraction analysis (d-spacing) (nm) dp Particle diameter (m)        Internal diameter of bed (cm) dpore Pore size (nm)      Binary bulk diffusivity of components i and j (m2.s-1)         Bulk effective diffusivity of components i and j (m2.s-1)     Knudsen diffusivity (m2.s-1)        Effective Knudsen diffusivity (m2.s-1)        Effective diffusivity at T0=330°C (m2.s-1)       Effective diffusivity (m2.s-1) dWcat                                       Differential  mass (g) Ea                            Apparent activation energy (kJ.mol-1) EB                           Binding energy (eV) Ek                           Kinetic energy (eV) FCH4                        Molar flow rate of CH4 (mol.s-1) F0CH4                       Molar flow rate of CH4 at the inlet (mol.s-1) xxvii             Molar flow rate of CH4 at Wcat (mol.s-1) g   Gram(s)  G Superficial mass velocity (kg.m-2.s-1) h     Hour(s) hν                            Energy of X-ray photon in XPS analysis (eV)        Relative intensity of component i based on He (-)         factor (-)    CH4 reaction rate constant (s-1) kc External mass transfer coefficient (m.s-1)      Pre-exponential factor for CH4 reaction rate constant (mol.gcat.molsite2.s-1.Pa-1)     CH4 reaction rate constant (mol.gcat.molsite-2.s-1.Pa-1) kd                            Rate of catalyst deactivation (min-1) kd,d                          Rate of catalyst deactivation under dry reaction condition (min-1) kd,w                         Rate of catalyst deactivation under wet reaction condition (min-1) kf                            Rate constant H2O adsorption (Pa-1.s-1) kr                            Rate constant for H2O desorption (s-1)        Pre-exponential factor for equilibrium constant of H2O adsorption (Pa-1)                           Equilibrium constant for H2O adsorption (Pa-1)  L                             Crystallite size measured by X-ray diffraction (nm) Lbed Length of catalyst bed (cm) m Total mass of catalyst bed (g) Mwi Molecular weight of component i (g.mol-1) n                            number of active sites involved in the dissociative adsorption of the CH4 (-) n Order of CH4 oxidation reaction (-) n                             Order of diffracted beam in X-ray diffraction (-) NA Avogadro constant (mol-1) O*-O* Two adsorbed oxygen site pair (-) O*-* Adsorbed oxygen and a vacancy site pair (-) P Total pressure (Pa) xxviii  P                             Partial pressure of N2 in surface area measurement (mmHg) P0                            Saturation pressure in surface area measurement (mmHg) Pi                            Partial pressure of component i=CH4, O2, H2O (Pa)      Critical pressure of component i (kPa) Pds                  Pd surface composition (at.%) Pd-*                        Pd vacant site (-) Pd-OH                    Covered active site by OH group (-) R                             Ideal gas constant, 8.3144 (Pa.m3.mol-1.K-1) Re Reynolds number       CH4 reaction rate (mol.(cm3.s)-1)         CH4 reaction rate (mol.gcat-1.s-1)                             CH4 reaction rate (mol.molsite-1.s-1) s             Second(s) Sc Schmidt  Number (-) Sh Sherwood Number (-) SBET BET surface area (m2.g-1) t                              Time (s) T0 Reference temperature (603K)     Critical temperature of component i (K) T                             Temperature (°C, K) T10, T50, T90         Light-off temperatures at 10%, 50%, and 90% CH4 conversion (°C) us Superficial gas velocity (m.s-1) V  Volume of adsorbed gas at a constant pressure P (cm3(STP).gcat-1) Vm  Volume adsorbed at monolayer coverage (cm3(STP).gcat-1) Vbed Volume of bed (cm3) V0 Molar volume of gas (22414 cm3.mol-1) V0 Catalyst pore volume (cm3.g-1) Wcat  Catalyst mass (g) Wsic Mass of SiC (g) XCH4  CH4 conversion (mol.%) xxix  Xs  CH4 conversion at infinite time (mol.%)  yAr     Ar volume fraction (-) yCH4     CH4 volume fraction (-) yHe    He volume fraction (-) yO2     O2 volume fraction (-)        Relative volume fraction of component i based on He (-)   Greek Letters  ρbed Catalyst density (g.cm-3) ρSiC SiC density (g.cm-3) ρcat Catalyst density (g.cm-3) ρbSiC Catalyst bed (g.cm-3) density of both catalyst and SiC ρ   Gas density (kg.m3) ρs Density of solid (g.cm-3) εbSiC Bed porosity (-)     Particle porosity (-) (ε/k)i Lennard-Jones energy/Boltzmann's constant for component i (-) ΔHH2O Enthalpy of H2O adsorption (kJ.mol-1) ΔXdry-wet The difference between CH4 conversion in dry feed gas after 24h  and wet feed gas after removing water ΔXs The difference between Xs for the dry and wet feed η Internal effectiveness factor (-)    Thiele modulus (-) θ Angle of reflection in XRD (°) θv Vacant active site (-) θH2O Fraction of active sites covered by H2O (-) β Full Width at Half Maximum, FWHM (Radians) ν0 Total volumetric flow rate (cm3.s-1) xxx        Initial partial pressure of H2O per initial partial pressure of CH4 (-) λ X-ray wavelength for XRD analysis (Å)    Tortuosity factor (-) σ  The occupied area by adsorption of a single molecule of N2 (Å2) σ Constriction factor (-)     Lennard-Jones characteristic length for component i (i=He or CH4) ΩD Collision integral (-) μ Gas dynamic viscosity (kg.m-1.s-1)    xxxi  List of Abbreviations AAS     Atomic absorption spectroscopy A/F            Air/fuel B.E.         Binding energy (eV) BET         Brunauer-Emmett-Teller BET SA      BET surface area (m2.gcat-1) BJH          Barrett-Joyner-Halenda CeOx   Cerium oxide CNG   Compressed natural gas co-xCe/yPd/Al2O3  A catalyst consists of x wt.%Ce and y wt.%Pd over Al2O3 prepared by  co-impregnation method DFT Density Functional Theory GHG     Greenhouse gas GHSV  Gas hourly space velocity (cm3.gcat-1.h-1) LEV Low emission vehicle LNG  Liquefied natural gas MBtu   Million British thermal units MMT Million metric tons MFC    Mass flow controller MVK   Mars-van Krevelen model MS Mass spectrometer  NG        Natural gas NGV      Natural gas vehicle NOx        Nitrogen oxides ODE       Ordinary differential equation OSC  Oxygen storage capacity PDE       Partial differential equation PdO        Palladium oxide QMS      Quadrupole mass spectrometer RDS       Rate determining step xxxii  seq-xCe/yPd/Al2O3   A catalyst consists of x wt.%Ce and y wt.%Pd over Al2O3 prepared by  sequential impregnation method sccm      cm3(STP).min-1 SOx        Sulfur oxides TLEV Transitional low emission vehicle TOF Turn-over frequency (s-1) TOS       Time on stream (h) TPO       Temperature-programmed oxidation of CH4 TWC      Three-way catalytic converter ULEV Ultra low emission vehicle vol.         Volume  (v/v)       Volume basis wt.          Weight basis XPS       X-ray photoelectron spectroscopy XRD      X-ray diffraction     xxxiii  Acknowledgments  I would like to express my sincerest gratitude to my doctoral supervisor, Professor Kevin J. Smith from the Chemical and Biological Engineering Department at the UBC for his great support. I am thankful to have this opportunity to do my PhD program with a supportive and intelligent supervisor. His wide knowledge, patience, encouragement, and unbounded guidance motivated me to complete my research. I have learnt so much from him and I will try to pursue his advice in my career.   I would like to thank my committee members Dr. Fariborz Taghipour from the Department of Chemical and Biological Engineering and Dr. Patrick Kirchen from the Department of Mechanical Engineering for their helpful comments and advice during my PhD research.  I would also like to thank all of the UBC Catalysis group members and visiting scholars who have been supportive colleagues and friends throughout my PhD appointment: Farnaz Sotoodeh, Rahman Gholami Shahrestani, Shahin Goodarznia, Hooman Rezaei, Ross Kukard, Victoria Whiffen, Pooneh Ghasvareh, Ali Alzaid, Xu Zhao, Siying Bian, Lucie Solnickova, Alex Imbault, Christoph Heinz, Chujie Zhu, Majed Alamoudi, Hamad Almohamadi, Shida Liu, and Haiyan Wang.  I would like to thank all the staff members in the Department of Chemical and Biological Engineering including Joanne Dean, Ivan Leversage, Lori Tanaka, Amber Lee, Marlene Chow, as well as CHBE workshop and CHBE store staff members for their lab assistance and support. In addition, I would like to thank Dr. Ken Wong from the Interfacial Analysis and Reactivity Laboratory at UBC for XPS measurements, Jenny Lai and Dr. Mati xxxiv  Raudsepp from the Department of Earth, Ocean and Atmospheric Sciences at UBC for their help with XRD measurements, and also Canadian Microanalytical Service Ltd. (CMAS) for their help for ICP measurements.  I would like to acknowledge the Natural Sciences and Engineering Research Council of Canada (NSERC), Westport Innovations Inc. and University of British Columbia who have provided project funding over the years that made my PhD research possible.  I would like to thank my loving parents, my brother, and my sister that always support me and motivate me to pursue my desires in the academic path. They have always been a great support in my life and I am so thankful for having them in my life.   Finally I would like to express my thanks to my loving friend, who has been a true support in my PhD program. His unconditionally guidance and love have always been a motivation over the past three years.   xxxv     To those who I love the most, my parents          1  Chapter 1: Introduction  1.1 Background   Natural gas (NG) is a hydrocarbon gas mixture that consists primarily of CH4 and small amounts of other hydrocarbons and impurities, as shown in Table 1.1 [1]. CH4 has the highest H:C ratio between all hydrocarbons that results in the lowest CO2 emissions per unit of energy when compared with gasoline and diesel fuels. The high octane number of natural gas (octane number 130) also improves the combustion efficiency by increasing the compression ratio of the NG engine. Based on the U.S. Energy Information Administration (EIA) report, over 759 million cubic meters of natural gas is consumed in the United States annually [2] and about 150,000 vehicles use natural gas in the U.S. [3].   Table 1.1. Natural gas composition [1] Component Vol.% CH4 70-90 C2H6, C3H8, C4H10 0-20 CO2 0-8.0 O2 0-0.2 N2 0-5.0 H2S 0-5.0 He, Ne, Xe trace  Natural gas can be used as a fuel in the form of compressed natural gas (CNG) or liquefied natural gas (LNG) [3]. The advantages of using natural gas as a fuel in the transportation sector is its low cost and high availability [3]. Figure 1.1 compares natural gas, gasoline, and diesel fuel prices in the U.S over the past years.  2   Figure 1.1. Average prices for gasoline, diesel, and CNG over time (Adopted from [3])  The CO2 equivalent emissions from fossil fuel combustion in the U.S. is about 5,157.7 MMT/yr and 33.3% is from the transportation sector [4]. Emissions from the transport sector are limited by regulations and Table 1.2 reports the exhaust emission limits for light-duty vehicles in the U.S. in 2016. Note that there is no limit legislated for CH4 emission.    Table 1.2. Exhaust emission limits for Light-Duty vehicles (Adopted from [5]) Vehicle  Emissions  Useful Life  NMOGb NOx CO Formaldehyde PMc Type Category Standard g.mi-1 g.mi-1 g.mi-1 g.mi-1 g.mi-1 LDVsa TLEV Intermediate 0.125 0.4 3.4 0.015 - LEV 0.075 0.2 3.4 0.015 - ULEV 0.040 0.2 1.7 0.008 - TLEV Full 0.156 0.6 4.2 0.018 0.08 LEV 0.090 0.3 4.2 0.018 0.08 ULEV 0.055 0.3 2.1 0.011 0.04 a Light-Duty Vehicles, b Non-Methane Organic Gases, c Particulate Matter  $0.00 $0.50 $1.00 $1.50 $2.00 $2.50 $3.00 $3.50 $4.00 $4.50 $5.00 Cost per GGE Gasoline Diesel CNG 3  In 2013 approximately 2.1 MMT CO2 emissions equivalent of CH4 was emitted from natural gas transportation systems [4]. The low H2S concentration in natural gas means that it also has the lowest SO2 emissions from combustion when compared to other fossil fuels. Table 1.3 compares the exhaust emissions using CNG and gasoline fuels and clearly shows the lower emissions associated with CNG.  Table 1.3. Fuel economy and exhaust gas compositions using CNG and gasoline (Adopted from [3]) Vehicle Fuel Economy CO CO2 NMOG NOx Type mi.gal-1 g.mi-1 g.mi-1 g.mi-1 g.mi-1 CNG 11.54 1.99 563.54 0.05 0.54 Gasoline 13.10 5.83 666.85 0.29 0.78  The greenhouse effect or global warming is one of the critical issues facing our environment. The principal greenhouse gases are CH4, CO2, and N2O which serve to retain heat close to the earth by absorbing infrared radiation and slowing down its emitting rate [6]. Increasing the heat trapping potential of the atmosphere by increasing the concentration of greenhouse gases causes harmful effects on human life. The Global Warming Potential (GWP) is an index used to compare the greenhouse gas effect of different gases [7]. The index calculates the ratio of absorbed energy caused by the emission of one ton of a gas such as CH4 or N2O to the absorbed energy caused by CO2 over a specified time period. The GWP value of CH4 and N2O are reported as 28-36 and 265-298, respectively, for a 100 year time horizon, indicating that CH4 and N2O absorb 28 times and 265 times more energy than CO2, regardless of their lifetime [7].   4  A significant barrier for using natural gas vehicles (NGVs) is the unburned CH4 emitted to the atmosphere from the vehicle exhaust gas. Since CH4 has a higher GWP than CO2, the complete combustion of unburned CH4 by a catalytic converter is the main challenge.  Three-way catalytic converters (TWC) were developed to control vehicle emissions using gasoline. As shown in Figure 1.2, a TWC consists of a washcoated ceramic or metal monolith in a honeycomb structure with 1 mm2 channels that permit a high flow rate of exhaust gas with minimal pressure drop [8]. The washcoat is made of high surface area Al2O3 combined with CeO2 and ZrO2 in order to increase the oxygen storage capacity and thermal stability of the monolith [8]. The washcoated monolith is impregnated with active metals such as Pt and Rh. The typical amount of noble metal is 1-2 wt.% of the washcoated monolith with a ratio of Pt:Rh=17:1 [9]. Oxidation of CO and HC to CO2 occurs in the presence of Pt and Rh is used to reduce NOx to N2 [8,10].   5   Figure 1.2. The three-way catalytic converter installed in the exhaust gas emitted from gasoline engines [11]  A high conversion for both oxidation and reduction reactions in a TWC is obtained by controlling the air to fuel ratio (A/F) in an engine operating using gasoline fuel. Figure 1.3 shows the HC, CO, and NOx concentration as a function of A/F. If the engine is operated under rich burn conditions (a lack of air), CO and HC oxidation is unlikely, however, the reduction of NOx is highest as a result of high CO concentration. At stoichiometric conditions in an engine operating using gasoline fuel (A/F=14.6), the CO and HC combustion occurs faster than rich burn conditions because of higher O2 concentration, on the other hand, the increase in temperature and low CO concentration as a result of combustion suppresses NOx reduction (NO+CO→  N2+CO2). Under lean burn conditions (A/F > 14.6), the concentration of CO and NOx decreases, while the HC concentration 6  becomes higher than that under the stoichiometric conditions, as the combustion process is unstable at high A/F ratio [8].   Figure 1.3. The concentration of NOx, CO, and HCs as a function of air /fuel ratio operating in gasoline engines [8] (Copyright © 2010 John Wiley and Sons)   As the exhaust temperature in natural gas vehicles (500-550°C) is lower than gasoline or diesel vehicles, CH4 is not burned completely in NGVs and around 500-1500 ppm of CH4 remains unreacted in the exhaust gas [12]. The unburned CH4 along with 10-15vol.% H2O and 15vol.% CO2 pass through the catalytic converters [12]. Typical TWCs used for gasoline engines are not able to completely oxidize CH4 at the low exhaust gas temperature of NGVs in the presence of high H2O concentrations. CH4 is difficult to oxidize because of the high C-H bond strength (~415 kJ.mol-1) which makes it resistant to oxidation [13,14]. Typical 7  catalytic converters for NGVs use Pd as catalyst, a more affordable option than Pt or Rh and the most active metal for CH4/CO combustion [15,16].   Complete oxidation of CH4 has been widely studied in recent decades [12,16–28] mainly focusing on the effect of different supports (Al2O3, SiO2, ZrO2, and CeO2) and different metals, such as noble metals, rare earth metals, transition metals, and lanthanides. In most studies the loading of Pd was in the range of 0.5-5wt.% in order to simulate the Pd loading in washcoated monolith catalysts of commercial spark ignition natural gas engines [18]. The main focus has been on developing a catalyst that can operate at low temperature (< 400°C) and oxidize CH4 completely (CH4+2O2→CO2+2H2O). Although Pd-based catalysts are active for CH4 oxidation, they lose their activity during reaction. Possible causes of the loss in activity during CH4 oxidation include a reversible loss caused by H2O adsorption (Figure 1.4(a)) or an irreversible loss, caused by, for example, sintering that affects the physical and chemical properties of the catalysts permanently or formation of inactive sites such as Pd(OH)2 (Figure 1.4(b)).        8    Figure 1.4. Possible mechanisms of catalyst activity loss by H2O: (a) H2O adsorption on PdO active sites, (b) formation of inactive Pd(OH)2  H2O plays a crucial role in terms of catalyst activity during CH4 oxidation and is the main product of the complete oxidation of CH4 in addition to CO2. At low temperatures the activity of Pd-based catalysts is mainly affected by H2O. As more H2O is produced during the reaction, the catalytic activity decreases as does the life time of the catalysts [29].   The rate limiting step in CH4 oxidation depends on the H2O concentration and the reaction temperature [17]. At a reaction temperature higher than 500°C along with theH2O produced from the CH4 oxidation reaction (dry feed), the H2O effect is negligible and the rate determining step is C-H bond activation. Therefore, the reaction order with respect to H2O is zero [30]. However, at low reaction temperature (< 500°C) and 3.5vol.% H2O concentration in the wet feed, slow desorption of H2O from the catalyst surface means that the rate limiting step is H2O desorption from the catalyst surface [30]. At conditions where H2O desorption is the rate determining step, the concentration of H2O must be considered in order to calculate (a) (b) 9  the activation energy [17,31]. Proposed reaction steps for the CH4-O2 reaction on Pd-based catalysts assume that CH4 dissociation occurs by the interaction of CH4 on a PdO/Pd-* site pair, where Pd-* refers to an O vacancy [17,32]. Following several surface oxidation reactions, CO2 and Pd-OH species result [17,32]. OH desorption (2Pd-OH→PdO+Pd-*+H2O) leads to the formation of oxygen vacancies on the catalyst surface. These vacancies can be re-oxidized by the oxygen from PdO, oxygen from the support or O2 from the gas phase. However, in the presence of H2O, slower OH migration and consequently slower H2O desorption from the catalyst surface, leads to the suppression of oxygen exchange between the support and the vacant sites [32–34]. In the case of a support with high oxygen storage capacity (OSC) such as CeO2, the oxygen exchange from the support to the oxygen vacancies can be improved. However, the negative effect of H2O is not completely reversed in the presence of high OSC supports. Studies emphasizing the effect of supports with high OSC [33,34], preparation method [35], temperature and H2O concentration [30] on activity suppression by H2O are available. However, a clear comparison between the long term stability of Pd based catalysts with/without a high OSC species (e.g. Pd/Al2O3 vs. Pd/CeO2/Al2O3) in the presence of high H2O concentrations during CH4 oxidation at low temperature as a function of time, has not been reported. Furthermore, the kinetics of the CH4 oxidation reaction in relation to the effect of H2O and the presence of high OSC species has not been reported.  1.2 Objectives of the Thesis  Since H2O has a significant impact on active sites by adsorption and by suppressing the oxygen exchange between Pd vacant sites and the oxide supports, the objectives of this study 10   are formed as below.  To determine the structure sensitivity of Pd based catalysts during CH4 oxidation by investigating the effect of Pd loading on reaction kinetics.  To understand the effect of different preparation methods and various Ce:Pd ratios on the inhibiting effect of H2O during CH4 oxidation at temperatures < 400°C.   To determine the reversible and irreversible effects of H2O on the physical and chemical properties of the catalysts after extended reaction periods.  To develop a unifying kinetic model that quantifies the role of H2O in the inhibition of catalyst activity, with and without CeO2 over a wide range of H2O concentrations. The kinetic model parameters will also be used to quantify the effect of adding CeO2 to the support.      1.3 Approach of the Thesis  In order to meet the stated objectives, a series of experiments along with a kinetic study that accounts for the effect of H2O has been performed. As a first step, a general introduction regarding the negative effects of CH4 as a GHG, removal of unburned CH4 in the exhaust gas of NGVs in the presence of H2O, and the importance of an active catalyst that remains stable at low temperatures (< 500°C), is explained in Chapter 1.   In Chapter 2, a literature review is presented in order to provide a comprehensive review of the effect of H2O on CH4 oxidation over Pd catalysts at low temperatures, as well as to identify knowledge gaps in the mechanism of H2O inhibition.  11  Chapter 3 includes details regarding catalyst preparation with different Pd loadings, Ce:Pd ratios, as well as different preparation methods used in this study. The thermal pre-treatments applied to the catalysts, the experimental equipment, the Temperature Programmed Oxidation (TPO) and Time-on-Stream (TOS) test methodology to measure both initial activity and stability of the catalysts, as well as various characterization techniques applied to the catalysts are explained in this chapter. The calculations related to the effectiveness factor, carbon balance, and CH4 conversion during the TPO and TOS tests are also described in this chapter.  In Chapter 4 the effect of Pd loading (0.3wt.%-6.5wt.%) on the physical and chemical properties of the catalysts, as well as their stability in the presence of H2O is reported. The PdO/Al2O3 catalysts with different Pd loadings are compared in initial activity and stability by TPO and TOS tests, respectively. The effect of Pd dispersion on the amount of H2O adsorbed on active sites is also examined. The values of the rate constant and the equilibrium constant for H2O adsorption were obtained from the reactor design equation to show the structure insensitivity of the catalysts within the narrow Pd dispersion range of 33-57%.  In Chapter 5, the effect of varying Ce:Pd ratio with a constant 6.5wt.% of Pd loading on the initial activity of the catalysts is presented. Same as Chapter 4, the values of the rate constant and the equilibrium constant for H2O adsorption were obtained from the reactor design equation for co-2.9Ce/6.5Pd/Al2O3 catalyst and compared to those reported in Chapter 4 for the 6.5Pd/Al2O3 catalyst in order to show the beneficial effect of CeO2 in suppressing the negative effect of H2O. A dynamic study of the loss in the activity of the catalysts by applying an exponential empirical equation to the TOS results is discussed and the rate of 12  deactivation (  ) as a function of different temperatures (300-380°C) and different H2O concentrations (0 and 5vol.%) is obtained.  In Chapter 6, the effect of catalyst preparation using co-impregnation and sequential impregnation methods on the initial activity and stability of the catalysts is reported. Three catalysts using co-impregnation method and five catalysts using sequential impregnation method are prepared and compared in terms of activity at different temperatures and H2O concentrations. Five xCe/Al2O3 supports with the same Ce loadings as the sequentially impregnated catalysts were also prepared and characterized in order to understand the interaction of Pd with CeO2 and CeO2 with Al2O3. The oxygen exchange mechanism was invoked to emphasize the role of CeO2 on the oxygen storage capacity (OSC) of the support and on the catalyst activity.  Chapter 7 develops a kinetic model to describe the inhibiting effect of H2O on TOS tests in a non-steady state system. A linear regression is applied to the experimental results presented in Chapters 4 and 5 and the rate of H2O desorption (  ) as a function of temperature are calculated for each individual catalyst in order to explain the role of Ce on the H2O desorption.   In Chapter 8 conclusions of this study along with recommendations for future work are presented. Supplementary information is included in the appendices. Appendix A provides the details of catalyst preparation, Appendix B explains different characterization techniques used in this study, and Appendix C provides the information about the unit calibration, including mass flow controllers, mass spectrometers, and liquid pump. In Appendices D and E the error analysis and CH4 conversion calculation procedures are explained. The 13  repeatability of the TPO and TOS reactions are shown in Appendix F. Appendix G contains some supplementary figures and tables for Chapter 6. Mass transfer calculations are shown in Appendix H. The results for PdO/ZrOx/Al2O3 catalysts are presented in Appendix I since ZrO2 promoted catalyst was not the main focus of the study. Supplementary information of bimetallic catalysts are presented in Appendix J. Lastly, the details of the kinetic model MATLAB code are provided in Appendix K.   14  Chapter 2: Literature Review1  2.1 Introduction  In this chapter the effect of various supports, including metal oxides with high oxygen storage capacity, on the inhibiting effect of H2O are reviewed. The molecular structures of cerium oxide, such as CeO2 and Ce2O3 will be compared and their ability to transfer O in order to suppress the negative effect of H2O on the activity of Pd-based catalysts will be explained. In addition, the effect of aging the catalysts with/without extra H2O and its effect on the structure of the catalysts such as BET surface area, Pd dispersion, and the oxidation state of Pd will be considered. The effect of temperature and partial pressure of both reactants and products on the reaction order and the apparent activation energy in CH4 combustion will be reviewed. Lastly, the various kinetic models, such as Langmuir-Hinshelwood, Mars-van Krevelen, and Eley-Rideal will be compared in terms of their response to the inhibiting effect of H2O.   2.2 Effect of the Support on the Inhibiting Effect of H2O during CH4 Oxidation  2.2.1 CH4 Conversion on Different Supports  The data of Table 2.1 show that the inhibition of CH4 oxidation by H2O on Pd catalysts is                                                            1 A version of this chapter was published in "Rahman Gholami, Mina Alyani, Kevin J. Smith, Deactivation of Pd Catalysts by Water during Low Temperature Methane Oxidation Relevant to Natural Gas Vehicle Converters. Catalysts 5 (2015) 561-594.”   15  dependent upon the support. Light-off temperature at 30% CH4 conversion (T30) of Pd catalysts were compared at a constant Gas Hourly Space Velocity (GHSV) and gas feed composition for different supports. Pd/Al2O3 shows significantly more inhibition with 10% H2O added to the feed than either the Pd/SnO2 or Pd/Al2O3-36NiO catalysts.  Table 2.1. Comparing light-off temperature (T30) for CH4 oxidation over Pd supported catalysts [23] (Copyright © 2002 Elsevier) Catalyst 1.1%Pd/Al2O3 1.1%Pd/SnO2 1.1%Pd/Al2O3-36NiO  GHSV, h-1 48,000  Dry feed gas composition, vol.% 1%CH4/20%O2 in N2   T30, C  Added H2O, vol.%     0 345 290 372  1 400 315 372  5 430 335 420  10 460 360 425  20 510 365 445   More detailed data from Kikuchi et al. [23] comparing CH4 light-off curves for a 1.1wt.%Pd/Al2O3 catalyst and a 1.1wt.%Pd/SnO2 catalyst with H2O added to the feed over a range of concentrations (1-20 vol.%), are shown in Figures 2.1 and 2.2. By increasing the H2O concentration, the CH4 light-off curves for both catalysts shift to higher temperatures. However, the temperature shift is larger over the Pd/Al2O3 catalyst than the Pd/SnO2. The authors completed a simplified kinetic analysis of the CH4 oxidation rate data to show that the enthalpy of adsorption of H2O is strongest on the Pd/Al2O3 catalyst (Had ~ -49 kJ.mol-1), from which they concluded that the significant loss in activity of the Pd/Al2O3 in the presence of H2O is due to a high coverage of the active sites by H2O [23]. These results could also be interpreted according to the more recent proposals by Schwartz et al. [33,34], that hydroxyl accumulation on the support hinders oxygen migration and exchange, and 16  hence CH4 oxidation. The strong adsorption of H2O determined by kinetic analysis on the Pd/Al2O3 catalyst [23] is consistent with a large hydroxyl accumulation on the catalyst surface that could inhibit the O exchange.   Figure 2.1. Catalytic combustion of CH4 over 1.1wt.%Pd/SnO2 with different amounts of H2O added (vol.%). Reaction conditions: 1vol.%CH4, 20vol.%O2, 0-20vol.%H2O, balanced in N2. GHSV=48,000h-1 [23] (Copyright © 2002 Elsevier) 17   Figure 2.2. Catalytic combustion of CH4 over 1.1wt.%Pd/Al2O3 with different amounts of H2O added (vol.%). Reaction conditions: 1vol.%CH4, 20vol.%O2, 0-20vol.%H2O, balanced in N2. GHSV=48,000h-1 [23] (Copyright © 2002 Elsevier)  2.2.2 High Oxygen Mobility of Support  The rate of deactivation during CH4 oxidation in the presence of H2O has been shown to be reduced by using a support with high oxygen surface mobility. At temperatures below 450C, Ciuparu et al. [36] reported the inhibition effect of H2O to be dependent upon the oxygen mobility of the support. Comparing PdO supported on oxides with increasing surface oxygen mobility: Al2O3 < ZrO2 < Ce0.1Zr0.9O2, they showed that the resistance to H2O inhibition during CH4 oxidation increased in the same order. The deactivation rate of PdO was also compared over Al2O3, MgO, and TiO2 supports by Schwartz et al. [33,34] at temperatures < 450C. The deactivation was shown to be a consequence of reduced oxygen mobility due to hydroxyl adsorption. It was also shown that a PdO/MgO catalyst had a 18  slower deactivation rate compared with Al2O3 and TiO2 supports because of the higher oxygen surface mobility on the MgO [33,34]. However, Pd catalysts dispersed on other supports such as MCM-41 that have high surface area (1113 m2/g) and lower oxygen mobility than MgO and Al2O3, did not deactivate either, suggesting that other factors also play a role, depending on the catalyst and the support. Another study compared the stability of Pd/SiO2 and Pd/ZrO2 during CH4 oxidation using a dry feed gas [37]. Pd/ZrO2 is stable after 40h TOS, while the CH4 conversion over Pd/SiO2 catalyst increases from 13% to 32% in the first 3h, and then decreases to 22% after 96h (see Figure 2.3). Although the Pd/ZrO2 catalyst is more stable than the Pd/SiO2 catalyst, its conversion is lower than for the Pd/SiO2 catalyst. The lower deactivation rate observed on the Pd/ZrO2 is consistent with the higher oxygen mobility of this catalyst compared to Pd/SiO2, as noted above.    Figure 2.3. Methane conversion for Pd/ZrO2 and Pd/Aerosil130 catalysts. Reaction conditions: 1.5%CH4; 6%O2; total flow=90cm3.min-1, balanced in He; temperature=325°C; catalyst mass= 0.2g [37] (Copyright © 2005 Elsevier)  19  2.2.3 Effect of Catalyst Structure on Activity  Metal-support interactions, support stability and the tendency of the support to encapsulate Pd, may also play a role in the deactivation of Pd catalysts during CH4 oxidation. Gannouni et al. [38] compared Pd catalysts supported on silica and mesoporous aluminosilicas and showed that, according to the light-off curves measured with 1% CH4, 4% O2 in He, the CH4 oxidation activity is enhanced on the pure silica support, whereas on the aluminosilica, the beneficial effect of Al3+ on metal dispersion and catalytic activity is counterbalanced by partial metal encapsulation. Above 500°C in the presence of H2O, the structural collapse of the support, metal sintering, and metal encapsulation by the support all occur [38]. Zhu et al. [39] reported the encapsulation of PdO active sites by SiO2 during CH4 oxidation at 325C and during reduction in H2 at 650C. The authors suggested that the high temperature, the H2O formed during reaction, and the formation of Pd silicide during reduction followed by oxidation in O2, were all important factors promoting the encapsulation of PdO by the SiO2. Migration of SiO2 onto the metal crystallites in other catalyst systems containing H2O has also been reported in the literature [40,41]. Yoshida et al. [42] also examined the effect of various metal oxide supports of Pd on the low temperature oxidation of CH4 (Table 2.2). The catalytic activity varied with the support, but the support oxides with moderate acid strength (Al2O3 and SiO2) gave maximum CH4 conversion. For these catalysts higher activity corresponded to a higher oxidation state of Pd (bulk PdO). The lower activity of Pd on basic supports, was attributed to the formation of binary oxides from PdO and the support (such as Pd/MgOx), in spite of a high Pd oxidation state.  20  Table 2.2. Effect of support on properties of 5wt.%Pd catalysts and their CH4 oxidation conversion (Data adapted with permission from [42]) Support Support Acid Strength Pd Dispersion CH4 Conversiona  (Ho) Fresh Used % MgO 22.3 0.21 0.20 12 ZrO2 9.3 0.41 0.12 3 Al2O3 3.3 0.35 0.20 59 SiO2 -5.6 0.09 0.11 58 SiO2-ZrO2 -8.2 0.16 0.13 20 SiO2-Al2O3 -11.9 0.12 0.06 10 SO42--ZrO2 -13.6 - 0.02 11 aMeasured at 350C in 0.25%CH4/3%O2 in He at GHSV of 1,200,000 h-1  A comparison of initial CH4 oxidation activity as a function of temperature for Pd-Pt catalysts on Al2O3, ZrO2, LaMnAl11O19, and Ce-ZrO2 was reported by Persson et al. [43], with the monolith catalysts tested in a tubular quartz flow reactor at atmospheric pressure in 1.5vol.% CH4 in dry air and a space velocity of 250,000 h-1. For steady-state experiments, the reaction temperature was set at 470°C and then increased to 720°C stepwise in 50C increments, with each temperature held for 1h. The Pd-Pt/Al2O3 catalyst had the highest activity at lower temperatures, while the Pd-Pt/Ce-ZrO2 catalyst had the highest activity between 620°C and 800°C [43]. The authors suggested that the higher surface area of the Al2O3 compared to the other supports (90 m2/g for Al2O3 versus 10 m2/g for Ce-ZrO2), resulted in a more stable catalyst. The authors also noted that the Ce-ZrO2 likely enhances the stability of the PdO, similar to the enhanced stability observed on CeO2 [36]. In addition, ZrO2 has high oxygen mobility [36] and the ability to re-oxidize metallic Pd into PdO was 21  observed to vary between the supports. Pd/alumina is re-oxidized very slowly, whereas Pd supported on ceria-stabilised ZrO2, is re-oxidized more rapidly.  2.2.4 Hydrophobicity of the Support  Since H2O adsorption on the Pd and/or the support is an important step in inhibiting CH4 oxidation over Pd, support hydrophobicity may be expected to impact the inhibition effect of H2O. Araya et al. [37] studied this effect on the deactivation of Pd-based catalysts by preparing 1wt.% Pd on two different commercial silicas, Aerosil130 and Aerosil R972. The Aerosil R972 is hydrophobic since the OH groups have been replaced by methyl groups. Both 1%Pd/A130 and 1%Pd/R972 were reacted at 325°C with 3%H2O, 1.5%CH4, 6%O2 at a total flow rate of 90 cm3.min-1. As shown in Figure 2.4, the effect of H2O addition to the feed gas is approximately the same for both the hydrophobic silica, Pd/R972 and the Pd/A130. In both cases, a large decrease in CH4 conversion is observed with the introduction of H2O to the reactor. The authors reported a reaction order with respect to H2O of -0.25 for both Pd/A130 and Pd/R972, emphasizing that the hydrophobicity of the support does not affect the extent of H2O inhibition observed on either catalyst.  22   Figure 2.4. A) Pd/Aerosil130 catalyst, B) Pd/R972 catalyst.  Reaction conditions: total flow=90cm3 (STP).min-1, temperature=325°C; catalyst mass=0.2 g.  Open symbols: dry feed 1.5%CH4; 6%O2 balance He; closed symbol: wet feed 1.5%CH4; 6%O2 with 3%H2O, balance He [37] (Copyright © 2005 Elsevier)   23  2.3 The Role of CeO2    The oxygen storage capacity plays an important role in the performance of TWC [44–46]. The reactivity of CeO2 originates mainly from its redox chemistry, which allows the storage of O2 under oxidizing conditions (Ce4+) and the release of O2 under reducing conditions (Ce3+) [46]. Among the elements in the periodic table, Ce is known for its unusual electronic structure with the f-orbital partially occupied in Ce [47]. It is also known that the cerium structure remains as face centered cubic, f.c.c during γ-α isostructure transition [47]. However, this structure preservation occurs by a drastic collapse in the volume of the unit cell. The change in volume during the transition is due to delocalization of the 4f electrons [47]. Figure 2.5 shows the cubic unit cell structure of Ce2O3 and CeO2. The Ce2O3 cells are made up of eight unit cells of CeO2, however, the volume of each cell is increased by 3% and 25% of the oxygen atoms are removed [47]. Figure 2.5(a) shows the presence of the oxygen vacancies and larger lattice unit cell for Ce2O3 compared to the CeO2 cells.  24   Figure 2.5. Ce2O3 lattice unit cells (a) and CeO2 (b). Blue spheres represent the cerium, red and white spheres are defined as oxygen atoms and vacancies, respectively [47] (Copyright © 2002 American Physical Society)  In terms of the possibility to form an oxygen vacancy, about 4.55 eV energy is required for the pure CeO2 (Figure 2.6(a)), however, this number reduces to 0.26 eV if an oxygen vacancy forms next to a pair of Ce3+ atoms that are surrounded by the CeO2 matrix (Figure 2.6(b)) [47] indicating higher possibility of formation of oxygen vacancies in the presence of Ce2O3.  25   Figure 2.6. The formation of oxygen vacancy for (a) a CeO2 crystal, (b) a pair of Ce3+. Black spheres indicate the Ce3+ and   indicates oxygen vacancy (Reproduced with permission from [47])  Since O transfer in ceria occurs easily, it can be written as CeO2-x with x in the range of 0-0.5 [46]. Tsunekawa et al. [48] showed that for small ceria particles (≤ 1.5 nm), x is equal to 0.5 and ceria is in a fully reduced form that is presented as cubic Ce2O3 [48]. However, Hailstone et al. [46] showed the formation of Ce2O3 depends on the crystallite size of ceria. It was shown that even for ceria crystallite size < 1 nm, no Ce2O3 was observed. Ideally, decreasing the ceria crystallite size should lead to an increase the OSC because of increasing the surface area/mol of CeO2 [46]. However, Hailstone et al. showed the OSC for ceria with the crystallite size of 11.8 nm is 425 µmol of O2/gCeO2, and this decreases to 349 µmol O2/gCeO2 and 65 µmol O2/gCeO2 as the CeO2 crystallite size decreases to 2.0 nm and 1.1 nm, respectively [46].    V 26  Table 2.3. Lattice expansion and oxygen storage capacity of CeO2 as a function of crystallite size [46] (Copyright © 2009 American Chemical Society) mean diameter Lattice Constant Lattice Expansion OSC nm nm % µmol of O2/gCeO2 1.1±0.3 0.578 6.8 65 2.0±0.5 0.555 2.6 349 11.8±1.2 0.547 1.1 425  In another study, the f.c.c structure of crystalline ceria was confirmed for three samples with different particle diameter of 6.7 nm, 3.8 nm, and 2.1 nm, while the calculated lattice parameter for the samples with different crystallite size decreased from 5.560 Å to 5.453 Å [48].   The presence of noble metals such as Pd improves the oxygen mobility of CeO2 [49]. One of the most active CH4 oxidation catalysts reported in the literature is PdO encapsulated in porous CeO2 [35]. The authors reported that the high activity was due to the strong interaction/contact between the CeO2 and the PdO, combined with the high O exchange mobility of CeO2. Colussi et al. [50,51] showed that only Pd particles in direct contact with the CeO2 re-oxidize at higher temperature and that the re-oxidation of the oxygen vacant site, Pd-*, is kinetically enhanced when in contact with CeO2 [51]. Hence the effect of CeO2 and the optimum loading varies in different studies, depending on the catalyst preparation method [52,53].  Colussi et al. compared the activity of Pd/CeO2 catalysts prepared by two different methods [54]. The Pd/CeO2 catalysts using solution combustion synthesis (SCS) were more active than those prepared by incipient impregnation (IWI), with three to five times higher reaction rate in the former case. The higher activity is ascribed to more stable Pd-O active sites. The 27  ex situ XRD analysis of the fresh and used SCS and IWI catalysts after reaction cycles confirmed the presence of PdO only. However, the in situ XRD analysis revealed a dynamic transformation of PdO→Pd→PdO during heating-cooling cycles for both IWI and SCS catalysts.   The oxygen release and uptake measurements were also performed in order to quantify the PdO→Pd transformation while the catalysts were exposed to 2vol.% O2 in N2 upon heating/cooling cycles by varying the temperature from room temperature to 1000°C. The SCS sample had lower O2 uptake/release than IWI samples indicating the presence of a stable form of Pd-O-Ce sites that do not contribute to the transformation [54]. However, the identical Pd dispersion for both SCS and IWI samples, measured by H2 chemisorption, confirms that the lower O2 uptake and release measured for the SCS sample is only a consequence of stable Pd-O, not as a result of Pd encapsulation [54]. Table 2.4 shows the O2 release and uptake for the catalysts. In another study [55], the presence of Pd2+ cations embedded in CeO2 was observed for the Pd/CeO2 catalysts prepared using solution-combustion method, however, the palladium was in metallic form (Pd0) for the catalyst prepared by the wet impregnation method.  Table 2.4. The amount of oxygen adsorption/desorption on IWI and SCS samples per gram of catalyst (Reproduced with permission from [54]) Sample Pd Loading O2 release O2 uptake wt.% (μmol/gPd)×10-3 (μmol/gPd)×10-3 IWI 1.71 3.76 3.79 SCS 1.71 3.34 3.28  28  The preparation method along with the pretreatment of the Pd/CeO2 catalysts can have a strong impact on the morphology and oxidation state of the catalysts [56]. Stasinska et al. [57] reported no enhancement in the activity of a PdO/Ce-Al2O3 catalyst prepared using a sol-gel method, compared to PdO/Al2O3, except when operated at low temperature (< 427°C). In contrast, Groppi et al. [58] reported that the temperature of PdO reduction and Pd0 re-oxidation increased by about 50-60°C on a 2.5wt.%Pd/11.5wt.%CeO2/Al2O3 catalyst, compared to 2.5wt.%Pd/Al2O3 catalyst, in agreement with the observation that the CH4 combustion was not affected by the CeO2 except at high temperature. Colussi et al. also reported that the oxygen uptake on 10%Pd/15%CeO2/Al2O3 catalyst was higher than other catalysts without CeO2 [59]. The optimum amount of CeO2 varies in different studies [52,53], although high loadings of Ce (50wt.%) are known to suppress catalytic activity [52]. Xiao et al. [60] studied the effect of preparation method on the activity of 2wt.%Pd/CeO2 catalyst. Using impregnation (IM) and deposition-precipitation (DP) methods, they showed that DP is a favorable method in terms of achieving high activity and stability as 50% CH4 conversion occurred at 257°C for the Pd-DP catalyst, about 300°C lower than that for Pd-IM catalyst [60]. The highly dispersed Pd along with the high concentration of oxygen vacancies lead to improved catalytic activity of the Pd-DP.                                                                                                                                                                                                                                                                                                                                                                                                        The Pd-DP catalyst was aged in the presence of 1vol.% CH4 and 99vol.% air with a space velocity of 50,000h-1 at 300°C. The catalytic activity decreased slightly from 100% to 93.4% after 16h. However, the deactivation was completely recovered by reducing the catalyst at 300°C in the presence of H2 for 1h [60]. The same recovery was observed by Bozo et al. [61] as they showed reducing the Pd/Ce0.67Zr0.33O2 catalyst at 300°C leads to formation of oxygen vacancies [61]. Raman spectroscopy of pure CeO2, Pd-DP, and Pd-IM showed there is a 29  main Raman band at 465.5 cm-1 indicating the symmetrical Ce-O bond. This Raman band was observed for all three samples, however, the intensity decreased significantly for Pd-IM and Pd-DP catalysts. This decrease was a consequence of high deficiency of CeO2 due to formation of oxygen vacancies [60]. The Raman spectra of the samples are shown in Figure 2.7.    Figure 2.7. Raman spectra for (a) CeO2, (b) Pd catalyst prepared by impregnation method, and (c) Pd catalyst prepared by deposition-precipitation method [60] (Copyright © 2005 Elsevier)  Misch et al. [62] investigated the C-H bond activity for partial oxidation of CH4 in the presence of Pd substituted CeO2 (Ce1−xPdxO2−δ). The catalyst was prepared using ultrasonic spray pyrolysis (USP) with PdO particle size of 10 nm. The presence of Pd substituted in the 30  lattice cells was proved at x=0.1 in Ce1−xPdxO2−δ. However, it was shown the catalyst is only active for C-H bond activation after reducing the Pd2+ to Pd metal.   A computational density functional theory (DFT) investigation of the complete oxidation of CH4 over CeO2 (111), PdO (100), and PdxCe1-xO2 (111) was presented by Mayernick et al. suggesting the lowest reaction barrier occurs for PdxCe1-xO2 (111) surface [56]. It was also shown that the PdxCe1-xO2 (111) surface is more stable than PdO (100) surface [63]. The rate determining step for complete oxidation of CH4 over Pd/CeO2 catalysts was considered to be C-H activation that is a function of the oxidation state of palladium, morphology, and also Pd, Ce, and O composition of the surface [56]. Comparing the activation barrier between CeO2 (111), PdO (100), and PdxCe1-xO2 (111) surface shows a minimum activation barrier obtained for the PdxCe1-xO2 (111) surface at +0.18 eV as it was +1.65 eV and +1.08 eV for the CeO2 (111) and PdO (100) surface, respectively [56].  2.4 Structure Sensitivity of the Pd Based Catalysts  The structure of Pd-based catalysts is an important factor in determining their activity for CH4 oxidation. The activity of Pd catalysts is strongly affected by the Pd particle size distribution as reported in some studies [64,65]. Hicks et al. [65] examined the effect of metal dispersion on the catalyst activity. When the catalyst was exposed to reaction conditions, the Pd restructured as it converted to PdO and this led to increasing dispersion. The smaller particle size of PdO with higher dispersion showed higher interaction of active sites with the support, which stabilized the PdO and decreased the activity of the catalyst. Otto et al. [66] also showed that for Pd/Al2O3 catalysts with Pd loadings ≤ 0.5wt.%, a 1.6 eV increase in the Pd binding energy is observed that indicates the formation of highly dispersed 31  PdO that leads to stronger chemical interaction of palladium and the support [66]. In another study by Gigola et al. [67,68], the interaction strength of the Pd and Al2O3 support was distinguished by the precursor concentration. Lower Pd concentration caused stronger interaction between Pd and the Al2O3 support. Ciuparu et al. [30] showed that it is difficult to distinguish the role of PdO particle size and the oxygen mobility of the support on the activity of the catalysts [30]. Castellazzi et al. [69] showed the turn-over frequency (TOF) for the Pd/Al2O3 catalysts with 1wt.%, 2wt.%, and 4wt.%Pd varies as 5.6×10-3 s-1, 1.7×10-2 s-1, and 3.5×10-2 s-1, respectively. The 1wt.%Pd catalyst had a higher dispersion of PdO than the 2wt.% and 4wt.% Pd/Al2O3 catalysts. However, the higher TOF is not related to the PdO dispersion corresponding to the different Pd loading. Since aging the catalysts caused a negligible decrease in PdO dispersion for the 1wt.%Pd/Al2O3 catalyst and it remained constant for the other catalysts, the TOFs increased for all catalysts. They conclude that the TOF is mostly dependent on the Pd-support interaction, not the PdO particle size [69]. In other studies, the change in TOFs was also considered independent of the particle size [70–72]. The activity of 2.7wt.%Pd/γ-Al2O3 catalyst was monitored as a function of particle size. The catalyst was treated under a heating process at 550°C in the presence of air for about 40 days and the activity was measured during this period [71]. The particle size increased significantly from 14 nm to 80 nm, while the activity of the catalyst per gram of palladium remained unaffected by TOS. Ribeiro et al. [72] showed the TOFs for the PdO crystallite size in the range of 2-110 nm changes only from 2×10-2 s-1 to 8×10-2 s-1. Zhu et al. [73] also showed the reaction order is insensitive to the structure of the catalysts. Comparing the TOFs for Pd (111), Pd (100), and Pd (110) show that regardless of the surface structure of the Pd, the reaction rate is only dependent on the oxygen-oxygen interaction as the Pd surface is 32  covered by the adsorbed oxygen [73]. Fujimoto et al. [74] showed the structure sensitivity over Al2O3 is limited to Pd particle size < 7 nm. Since the structure sensitivity of Pd based catalysts for CH4 oxidation is still under debate, the effect of Pd dispersion on the amount of H2O adsorbed on PdO active sites needs to be examined.  2.5 Kinetics of H2O Inhibition in CH4 Oxidation   The rate of reaction for CH4 oxidation is a function of temperature, the CH4 and O2 reactant partial pressures, product H2O and CO2 partial pressures, the oxidation state of Pd at reaction conditions (PdO, Pd0), size and morphology of the catalyst [17]. The effect of catalyst pretreatment on the reaction rate was studied by Muto et al. [75]. The 2wt.%Pd/Al2O3 catalyst was pretreated at 450°C in H2, O2, or N2 before measuring the activity. It was shown that the rate of CH4 oxidation was faster for the catalyst pretreated in N2 rather than O2 or H2. However, the difference in CH4 oxidation rate decreased as the reaction temperature increased to 450°C. The reaction orders with respect to CH4 and O2 were reported by Muto et al. [75] using Equation 2.1.                                                                                                                                       2.1 where r, k,     , and     are defined as the CH4 oxidation reaction rate, the reaction rate constant, and partial pressure of CH4 and O2, respectively. The m and n values in Equation 2.1 are compared with other studies in Table 2.5. The low reaction order with respect to O2 is due to the presence of PdO in the CH4 combustion reaction [75]. In another study by Zhu et al. [73], it was shown that the reaction order for CH4 and O2 is structure insensitive. Pd (111), Pd (100), Pd (110), and Pd foil in both metal or oxide phase have the order of 0.7 < CH4 < 1 33  and -0.1 < O2 < 0.2 [73], indicating that the large single-crystal Pd catalysts result in the same CH4 combustion activity regardless of the oxidation state of Pd [73].   Table 2.5. Apparent activation energy and order of CH4 combustion reaction over Pd catalysts Catalyst Ea Reaction Order Temperature Range Refs kJ.mol-1 CH4 O2 H2O °C  Supported Catalysts             10%Pd/ZrO2 174 1.00 0.00 -1.00 232-360 [72] 1.1%Pd/Al2O3 81 1.00 0.00 -1.00 290-500 [23] 1%Pd/ZrO2 170 1.00 0.00 -1.00 227-441 [37] 1%Pd/SiO2 - 1.00 0.00 -0.25 227-441 [37] 7.3%Pd/Al2O3 86 (151)a 1.00 0.10 -0.80 253-315 [76] 0.5%Pd/Al2O3 60 0.90 0.08 -1.3 to -0.9 240-400 [77] Model Catalysts             Pd foil 125 0.7 -0.10 0.05 296-360 [73] Pd(111) 140 0.7 -0.10 0.05 296-360 [73] Pd(100) 130 0.9 0.01 0.07 296-360 [73] PdO foil 125 0.7 0.20 -0.90 296-360 [73] PdO(111) 140 0.8 -0.10 -0.90 296-360 [73] PdO(100) 125 0.8 0.10 -1.00 296-360 [73] a 86 kJ.mol-1 and 151 kJ.mol-1 for the dry and wet conditions, respectively.  The studies of the inhibition effect of the reaction products confirm the strong negative effect of H2O on CH4 oxidation [72,76,78,79]. Ribeiro et al. [72] investigated the effect of extra H2O added to the feed stream on the reaction order. A mixture of 1%CH4, 0.24%CO2, and H2O in the range of 0.03-0.15% in air was used to measure the TOFs as a function of H2O concentration. The reaction order with respect to H2O was reported as -0.98 [72]. The reaction orders close to -1 indicate strong competition of H2O molecules with CH4 for the surface sites [72]. The inhibiting effect of H2O was explained by the formation of reversible Pd(OH)2 at the surface of PdO active sites [72,80]. The reaction rate order for H2O was determined as a function of temperature by Ciuparu et al. [81]. The H2O inhibition effect is more dominant at temperatures below 450°C [81]. In another study by Hurtado et al. [77] the reaction order with respect to H2O varied from -1.3 to -0.9 while the temperature increased 34  from 300°C to 350°C, indicating the inhibition is the result of H2O adsorption. The activation energy is strongly affected by the presence of H2O [17,31]. As shown by Carsten et al. [31], the apparent activation energy for CH4 combustion over Pd catalysts should be corrected to account for the inhibiting effect of H2O [31]. In the study by van Giezen et al. [76] an empirical equation was derived to define the order of reaction with respect to H2O,  .                                                                                                                                    2.2 In a differential system assuming a small change in CH4 and O2 concentration and a constant temperature, k1 is defined as:                                                                                                                          2.3     is defined as the pre-exponential factor. Equation 2.4 represents the H2O concentration at the bottom of the catalyst bed which is a function of both extra H2O added ([H2O]in) and residence time ().                                                                                                             2.4 By plotting the conversion as a function of residence time, the reaction order of H2O was found to be -0.76 and -0.74 for a wet and dry feed, respectively. It was also shown that the apparent activation energy is dependent on the H2O concentration as Ea was estimated at 86 kJ.mol-1 in dry feed and 151 kJ.mol-1 for a feed with 2vol.% H2O [76]. In spite of the difference in Ea for dry and wet feed, the reaction order was independent of H2O concentration, i.e. there is no significant difference between the values of -0.76 vs. -0.74. The apparent activation energy of 174 kJ.mol-1 reported for 10%Pd/ZrO2 catalyst is in agreement with other studies concerning the effect of H2O on activation energy [31,71,72,74]. Ea values corrected for the effect of H2O are shown in Table 2.5. 35  A kinetic study of the effect of H2O with respect to CH4 adsorption as a rate determining step, was investigated by Kikuchi et al. using Equation 2.5 [23].                                                                                                                                    2.5 The rate of reaction, r, is a function of CH4 partial pressure and the fraction of vacant sites, θv. kr is defined in Equation 2.6 and represents the surface rate of CH4 combustion as a function of apparent activation energy (Ea). It is assumed that adsorption of CH4 molecules on PdO active sites is irreversible and dissociative, therefore, the PdO site coverage is only affected by H2O. The fast desorption of CO2 molecules formed by CH4 oxidation reaction causes a zero coverage of active sites by CO2 [23].                                                                                                                                     2.6 The fraction of vacant sites, defined in Equation 2.7, shows the dependency of vacant sites on the partial pressure of H2O (PH2O) and the adsorption equilibrium constant, KH2O. It is assumed that the H2O adsorption/desorption is at equilibrium [23].                                                                                                                       2.7 By plotting ln kr and ln KH2O versus 1/T, the values for the apparent activation energy and the enthalpy for H2O adsorption were estimated for Pd/Al2O3, Pd/SnO2, and Pd/Al2O3-36NiO catalysts [23]. Pd/Al2O3 catalyst, with the most negative value for the enthalpy of H2O adsorption, is confirmed to have the highest equilibrium H2O coverage of PdO active sites compared to Pd/SnO2 and Pd/Al2O3-36NiO catalysts and in agreement with the experimental results reported in [23]. Table 2.6 compares the apparent activation energy and enthalpy of H2O adsorption reported for the different catalysts.  36  Table 2.6. Estimated values of Ea and ΔHads (Reproduced with permission from [23]) Catalyst Ea ΔHads for H2O kJ.mol-1 kJ.mol-1 1.1%Pd/Al2O3 81 -49 1.1%Pd/SnO2 111 -31 1.1%Pd/Al2O3-36NiO 90 -30   Cullis et al. [78] studied the effect of a large quantity of H2O in the range of 0-5.510-5 mol (sic) added to a system with a mixture of 1.810-6 mol (sic) CH4 and 3.610-6 mol (sic) O2 at 352°C in the presence of 2.7%Pd/Al2O3 catalyst. The conversion of CH4 decreased by increasing the H2O content, emphasizing the inhibiting effect of H2O, however, the H2O produced from the CH4 oxidation reaction is not found to have any significant effect on the activity of the catalyst [78]. van Giezen et al. [76] showed the reaction order for H2O remains constant at -0.8±0.2 independent of different H2O concentrations. However, as shown by Ciuparu et al. the reaction order with respect to H2O is zero in a dry feed stream [30]. At a constant reaction temperature, the order of reaction with respect to H2O varies between 0 and -1 as a function of CH4 conversion [30]. Since a higher H2O content occurs at higher CH4 conversion (it is a product of reaction), the order of reaction has higher negative values with increased conversion.   CH4 oxidation over PdO active sites may occur at different conditions such that either CH4 activation or H2O desorption is the rate determining step [17,82]. During the CH4 oxidation reaction, CH4 and 16O on the PdO surface interact with each other, causing the adsorption of CH4 and the production of CO2 and OH on the surface of the catalyst [17,32]. The oxygen vacancy is rapidly replaced by the diffusion of 18O2 from bulk to the surface of the catalyst. 37  Figure 2.8 shows the schematic of the oxygen exchange during the CH4 oxidation reaction reported by Ciuparu et al. [32].   Figure 2.8. Oxygen exchange mechanism for CH4 oxidation using labeled (18O16O) pulsed experiments [32] (Copyright © 2002 Elsevier)  In another study by Schwartz et al. [34] the oxygen exchange from the gas phase, 18O2, and the oxide support, 16O, to the catalyst surface was studied over a 3wt.%PdO/Al2O3. It was shown that the oxygen exchange in the absence of CH4 becomes significant at temperatures higher than 380°C. The authors suggested that the rate determining step at low temperature (< 500°C) and high H2O content is water desorption from the surface of the catalyst and that the reaction order with respect to H2O is -1. However, at a high temperature and low H2O concentration, the rate determining step is CH4 activation and the reaction order with respect to H2O will be zero [30].  The kinetic mechanism of the CH4-O2 reaction on supported Pd clusters proposed by Chin et al. [82] is given below. Step 1.1. O2 (g)+* O2*  Step 1.2. O2*+* 2O* 38  Step 2.1. CH4+*+*→CH3*+H* Step 2.2. CH4+O*+*→CH3*+OH* Step 2.3. CH4+O*+O*→CH3O*+OH* Step 3. C*+O* CO*+* Step 4. CO*+O* CO2*+* Step 5. 2OH* H2O*+O* Step 6. H2O* H2O+* Step 7. CO2* CO2+* Step 8. CO* CO+* In this mechanism * and O* correspond to Pd-* and PdO, respectively, as used in the present study.  This schematic indicates the dissociation of both CH4 and O2 molecules on the Pd/Al2O3 catalyst surface and reaction of CH4 with O* species that results in the formation of OH* and CO* species. Steps 5-8 are quasi-equilibrated steps that indicate desorption of the produced species. The activation of C-H bond (steps 2.1 to 2.3) is more favorable on the O*-* than O*-O*. Since the oxygen chemical potential, O*-* (an adsorbed oxygen and a vacancy site pair) or O*-O* (two adsorbed oxygen site pair), are influenced by the Pd cluster size and O2 pressure, the C-H bond activation and consequently the TOF are a function of the Pd cluster size and O2 pressure [82]. 39  As shown in Figure 2.9 the CH4 oxidation 1st-order rate constant is higher at lower O2 pressure [82]. For the CH4 combustion reaction occurring over a 0.2wt.%Pd/Al2O3 catalyst at 873K with 4.85 kPa CH4, the O2 pressure varied between 0.3-1.7 kPa. The higher CH4 oxidation rate constant was obtained at an O2 pressure less than 1.2 kPa and 0.7 kPa for catalysts with 21.3 nm and 4.8 nm Pd cluster size, respectively. However, the reactivity of the Pd with 4.8 nm cluster size is lower than those with 21.3 nm and decreasing the O2 pressure caused an increase in the rate constant for both 4.8 nm and 21.3 nm Pd clusters. At a constant O2 pressure, the larger Pd clusters have weaker O* binding that leads to higher vacant site densities which are more effective for C-H bond activation.   Figure 2.9. Effect of O2 pressure on the CH4 oxidation rate constant over 0.2wt.%Pd/Al2O3 catalyst at 873K (4.8 nm (●, ▲) and 21.3 nm ( , ■) Pd cluster diameter) [82] (Copyright © 2011 American Chemical Society)  40  The transition of C-H bond activation (step 2.1 to 2.3) from less reactive O*-O* sites to more active O*-* sites by decreasing O2 pressure, leads to a higher CH4 oxidation rate constant. The same transition mechanism leads to the higher catalyst activity with larger Pd cluster size [82]. Different types of kinetic mechanisms were proposed to model the effect of temperature, H2O concentration, and partial pressure of the reactants and products on the activity of the Pd-based catalysts. Using Langmuir-Hinshelwood [23,30,77,83–86], and Mars-van Krevelen [77] mechanisms it was proposed that the rate limiting step is either the activation of C-H bond [20] or H2O desorption [30], depending on temperature and H2O concentration. However, in the presence of H2O, there is an agreement that OH group adsorption on the catalyst surface is the main cause of loss in catalyst activity [23,77,86]. Cortes et al. [86] defined the decay function that is expressed as Equation 2.8.                                                                                                                               2.8          is the reaction rate at time t and      is the maximum reaction rate achieved at the initial reactivation. A(t), the decay fraction, was functionalized versus TOS using a linear form, a hyperbolic form representing sintering, an exponential equation corresponding to the poisoning of the catalyst surface by chemisorption, and an equation describing coking or fouling of the catalyst surface. The best fit to the experimental results was obtained by the latter equation, indicating that the loss in activity during TOS is related to OH groups on the surface of the catalyst [86].  In another study by Hurtado et al. [77] the kinetics of the CH4-O2 reaction was studied for a 0.5%Pd/Al2O3 catalyst. Eley-Rideal, Langmuir Hinshelwood, and Mars-van Krevelen 41  models were applied to the experimental results. Among these models, the Mars-van Krevelen model that describes the slow desorption of H2O was found to have the best fit. In a modified version of MVK that contains an additional term in order to provide the dependency of reaction rate on the excess amount of H2O, the adsorption of H2O on the oxidized Pd active sites was considered as the inhibiting effect of H2O [77].                                                                                                                           2.9  The parameters are defined as the rate constant for the irreversible adsorbed oxygen (k1), rate constant for the CH4 reaction on the surface (k2), and rate constant for the desorption of products (k3), and KH2O as the equilibrium constant for the adsorption of H2O. The estimated value of       for      in Equation 2.9 for reduced and oxidized 0.5%Pd/Al2O3 catalyst were obtained as -66.1 kJ.mol-1 and -54.5 kJ.mol-1, showing the more inhibiting effect of H2O adsorption on the reduced catalyst.   2.6 Summary  The effect of H2O on the CH4 oxidation reaction using different catalyst supports shows more inhibition by H2O on Pd/Al2O3 than Pd/SnO2 or Pd/ZrO2 catalysts, reflecting the stronger H2O adsorption and consequently less O exchange on the Pd/Al2O3 catalyst than the Pd/SnO2 or Pd/ZrO2 catalysts. Supports with higher oxygen surface mobility, such as CeO2 and ZrO2, show slower deactivation rate compared to Al2O3. Accumulation and slow desorption of OH groups from the support may hinder oxygen exchange with the Pd/PdO, leading to activity loss. The preparation method along with the pretreatment of the Pd/CeO2 catalysts may have an impact on the oxidation state of CeO2 as it was shown that the OSC of 42  CeO2 increases with increasing crystallite size. There is a complex relationship between the role of Pd and the oxygen mobility of the support on the activity of the catalysts. Some studies confirm the high deficiency of CeO2 in the presence of Pd by formation of Ce2O3.  The RDS in the CH4-O2 reaction is considered to be C-H bond activation in most studies. The order of CH4 oxidation reaction with respect to H2O has a negative value, indicating the inhibiting effect of H2O. The negative value has been reported in the literature in the range of 0 to -1, depending on the H2O concentration, type of support, and temperature. The effect of H2O is more dominant at temperatures below 500°C. A Langmuir-Hinshelwood mechanism is assumed in several kinetic studies of CH4-O2 combustion. The slow desorption of H2O inhibits catalyst activity. The accumulation of OH groups and their low tendency to desorb from the catalyst surface suppress the oxygen exchange between the oxide support and the Pd-* vacant sites, causing a loss in the activity of the catalysts. The supports with high OSC such as CeO2, can suppress the negative effect of H2O as CeO2 facilitates oxygen transfer from the oxide support to the vacant sites.   In spite of a broad study in terms of the effect of high OSC species (CeO2) on the activity of Pd supported Al2O3 catalysts, a clear comparison between the long term stability of catalysts with/without CeO2 has not been reported. In addition, the inhibiting effect of H2O during CH4 oxidation at low temperature (< 400°C) on Pd based catalysts with/without CeO2 needs to be studied. The structure sensitivity of Pd based catalysts is still under debate. To understand the effect of Pd loading on the activity and stability, catalysts with different Pd loadings need to be examined for CH4 oxidation reaction in the presence and absence of 43  H2O. Also, the reversible and irreversible effects of H2O need to be investigated quantitatively.   The kinetics of H2O inhibition for different supports are well described in the literature, however, a study of the dynamic response of the catalyst to H2O addition that accounts for the effect of high OSC of the oxide supports and the inhibiting effect of H2O, is not available.  44  Chapter 3: Experimental  This chapter describes the experimental methods and procedures applied in this study and includes catalyst preparation, catalyst characterization, and catalyst activity tests for the CH4 oxidation reaction done with and without H2O added to the feed gas. The catalyst tests were done either by temperature programmed oxidation (TPO) of CH4 and time-on-stream (TOS) tests at fixed reaction conditions using PdO/Al2O3 and PdO/CeOx/Al2O3 catalysts in a fixed bed reactor. The descriptions that follow are augmented with additional details provided in Appendices A, B, and E. Note that a preliminary study of a PdO/ZrOx/Al2O3 catalyst was also completed, but since this catalyst was not the main focus of the study, the results are reported in Appendix I.   3.1 Catalyst Preparation   All of the catalysts studied in this thesis contain Pd, Ce, or Zr on the γ-Al2O3 support. The catalysts are calcined before use and are operated in O2 rich atmospheres so that the metal elements of the catalysts are present in the form of metal oxides. Throughout the thesis, the composition of the catalysts is reported in terms of the metal content (Pd, Ce, and Zr) as a weight % of the calcined catalyst. For example, the 6.5Pd/Al2O3 catalyst has a Pd content of 6.5wt.%, equivalent to a 8.0wt.% PdO.  A typical TWC has a monolith structure, washcoated with a porous material such as γ-Al2O3, and impregnated by a mixture of noble metals such as Pt, Pd, and Rh that provide the active sites for reaction [9]. The total amount of noble metals is 1-2wt.% of the washcoated γ-Al2O3. If Pd is the only noble metal used in the catalytic converter, the loading is typically 5 45  times higher than that used for Pt and Rh based catalysts [9]. Hence, in most studies using Pd supported on Al2O3, the Pd loading is approximately 5wt.%.  In the present study all the catalysts are Pd-based catalysts using γ-Al2O3 as the support. Since some characterization techniques such as XRD are not effective at low Pd loading (e.g. <2wt.%), a higher Pd loading was used to assist in the characterization of the catalysts. Hence, in this study Pd loading was varied from 0.3wt.% to 6.5wt.%.  All of the Pd catalysts studied in this thesis were prepared using incipient wetness impregnation. Generally, a commercial granular γ-Al2O3 (Sasol North America, alumina spheres 2.5/210) was crushed manually a using mortar and pestle and sieved to obtain an alumina powder with particle size in the range 90-354 µm to be used as the oxide support of the catalysts. The alumina powder was dried at 120°C for 24h in ambient air. The γ-Al2O3 had a specific surface area of 224 m2/g, a pore volume of 0.53 cm3/g, and 9.4 nm average pore size.   For the Pd/Al2O3 catalysts (reported in Chapter 4), approximately 2.7 g of the support was impregnated with an aqueous solution of Pd obtained from 0.1N HNO3 and Pd(NO3)2.xH2O (Aldrich ≥ 99% purity) in order to yield the desired loading of Pd (0.3wt.%, 2.6wt.%, and 6.5wt.% Pd). The impregnated catalysts were left at room temperature for 48h and then dried in an oven at 100°C for 8h. The dried catalysts were sieved again in order to obtain the mesh size in the range of 90-354 µm. Finally, the catalysts were calcined in situ in 100 cm3(STP).min-1 air (Praxair extra dry air) while heating from room temperature at 10°C.min-1 to 450°C, with the final temperature held for 15h. The calcined catalysts are identified as xPd/Al2O3, where x is the wt.% of Pd present in the PdO/Al2O3 calcined catalysts. The 46  purpose of the calcination is to remove undesired components originating from the precursors and also to stabilize the PdO phase. The catalysts were subsequently cooled to room temperature before testing.  For the Ce/Pd/Al2O3 catalysts reported in Chapter 5, approximately 2.7 g of the support was co-impregnated with an aqueous premixed solution of Pd and Ce, obtained from 0.1N HNO3, Pd(NO3)2.xH2O (Aldrich ≤ 100% purity) and Ce(NO3)3.6H2O (Aldrich 99% purity) in order to yield the desired loading of Pd and Ce. The thermal treatment of the impregnated Al2O3 was the same as that stated above. The catalysts prepared by this method are identified as co-xCe/yPd/Al2O3 catalysts where "co" stands for co-impregnation, x is the wt.% of Ce with 0.94wt.%, 2.9wt.%, 4.8wt.%, and 9.5%wt.% Ce loadings and y is the wt.% of Pd which is fixed at 6.5wt.% loading.  The catalysts investigated in Chapter 6 were prepared by co-impregnation and sequential impregnation methods. The co-impregnation catalyst followed the same procedure as for the catalysts reported in Chapter 5, however, the Ce loadings were higher (2wt.%, 14wt.% and 47wt.%) and the Pd loading was kept constant at 3.4wt.%. For sequential impregnation, the Al2O3 support was first impregnated with an aqueous solution of Ce obtained from 0.1N HNO3, Ce(NO3)3.6H2O (Aldrich 99% purity) in order to yield the desired loading of Ce. The impregnated supports were left at room temperature for 48h and then dried in an oven at 100°C for 8h and calcined in atmospheric air while heating from room temperature at 10°C.min-1 to 450°C, with the final temperature held for 15h. The new calcined CeO2-Al2O3 supports were impregnated with an aqueous solution of Pd obtained from 0.1N HNO3, Pd(NO3)2.xH2O (Aldrich ≤ 100% purity) then left at room temperature for 48h, dried in an 47  oven at 100°C for 8h and calcined in situ in 100 cm3(STP).min-1 air (Praxair extra dry air) while heating from room temperature at 10°C.min-1 to 450°C, with the final temperature held for 15h. The catalysts were cooled to room temperature before the activity test. The catalysts prepared by this method are identified as seq-xCe/yPd/Al2O3 catalysts where "seq" stands for sequential impregnation, x is the wt.% of Ce with 2wt.%, 6wt.%, 17wt.%, 28wt.%, and 57%wt.% Ce loadings and y is the wt.% of Pd which is fixed at 3.4wt.% loading.  The amount of solvent used to prepare the Pd and Ce solutions was selected based on the solubility of each precursor and also the pore volume of γ-Al2O3 support. As the required volume of solvent to dissolve the Ce salt in the co-47Ce/3.4Pd/Al2O3 catalyst was higher than the pore volume of 1.35 g of γ-alumina support [87], the premixed solution of Pd and Ce salts were added to the support in three steps. Between each step, the catalyst was left at room temperature for 48h in order to complete the impregnation. Then the catalyst was dried in an oven at 100°C for 8h and calcined in air at 450°C for 15h. The same method was applied in order to prepare the seq-57Ce/3.4Pd/Al2O3 catalyst.  The details of the calculations of the required amount of chemicals and solvent needed to prepare the catalysts with different Pd and Ce loadings are reported in Appendix A. Repeatability of the preparation method was confirmed using atomic absorption spectroscopy to determine the Pd content present in a xPd/Al2O3 catalyst prepared in two different batches. The standard deviation of the catalyst preparation is reported in Appendix D.  3.2 Catalyst Characterization  Several characterization techniques such as atomic absorption spectroscopy, ICP analysis, N2  48  adsorption-desorption, XRD, XPS, ToF-SIMS, and CO chemisorption were used in order to identify the properties of the fresh and aged catalysts. Details of these techniques are provided in the following sections.  3.2.1 Atomic Absorption Spectroscopy  The Pd content of the prepared catalysts was determined by Atomic Absorption Spectroscopy (AAS; GBC 904) using an air-acetylene flame with 5.0 mA current, 247.6 nm wavelength, slit width of 0.2 nm, 0.08 μg/mL sensitivity and optimum working range of 4-15 μg/mL of the Pd solution. For the analysis, 10 mg of each catalyst was digested in 2 mL HCl (50wt.%), 2 mL HNO3 (20wt.%) and 2 mL H2SO4 (96wt.%). The resulting solution was diluted with deionized water to obtain approximately 10 μg/mL Pd prior to AAS analysis. Various Pd solutions with different Pd concentrations (0-15 μg/mL) were also prepared using a 1000 μg/mL Pd solution diluted with deionized water. All Pd loadings reported in this thesis are actual loadings after being analyzed by AAS.  3.2.2 ICP Analysis  The Ce content was measured by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) using a Thermo Scientific X-Series II with a Cetac ASX-520 autosampler. The samples were digested in 1 mL concentrated H2SO4/HNO3 and diluted to 50 mL with deionized water. All Ce loadings reported in this thesis are actual loadings after being analyzed by ICP.  3.2.3 N2 Adsorption-desorption  Brunauer-Emmett-Teller (BET) surface area, pore volume, and pore size of the catalysts were determined from N2 adsorption isotherms measured at 77K using a Micromeritics 49  ASAP 2020 analyzer. Approximately 0.1g of the catalyst sample was loaded inside a glass tube and sealed with a plastic frit and degassed at 250°C for 12h in vacuum in order to remove moisture. Then a N2 adsorption-desorption isotherm was obtained as a function of relative pressure at 77K. The BET surface area was calculated using the BET isotherm. The Barrett–Joyner–Halenda (BJH) method was used to estimate the average pore size and the total volume of adsorbed N2 at the relative pressure of P/P0 of 0.995 was used to calculate the pore volume of the catalyst. Details are provided in Appendix B.1.  3.2.4 X-ray Diffraction  X-ray diffraction patterns were collected using a Bruker D8 Focus (LynxEye detector) diffractometer with a CoKα source operating at 40 kV and 40 mA and X-ray wavelength of 1.7902 Å. The analysis was performed using a scan range of 10-80° with a step size of 0.04° and step time of 4 s. The Scherrer equation was used to calculate the crystallite size [8]. Details of the XRD analysis are provided in Appendix B.2.  3.2.5 X-ray Photoelectron Spectroscopy  A Leybold MAX200 X-ray photoelectron spectrometer (XPS) with Al Kα achromatic X-ray source and a survey pass energy of 192 eV and narrow pass energy of 48 eV was used to determine the Pd, Ce, and Zr oxidation states, the Pd/Al atomic ratio as well as the catalyst surface composition. XPSPEAK41 was used to analyze the spectra after background subtraction by the nonlinear Shirley method. Details of XPS analysis are provided in Appendix B.3.   50  3.2.6 Time-of-Flight Secondary Ion Mass Spectrometry  A Physical Electronics TRIFT V nanoTOF instrument was used for ToF-SIMS analysis in order to detect the surface composition of Pd, Al, and Ce. The approximate depth of analysis in ToF-SIMS is 2 nm with the source of 30 keV Au+ pulsed primary ion source in bunched mode. The area of analysis is 400µm400µm using the total ion dose of about 1012 ions/cm2. The mass spectra were collected in the range of 0-1850 m/z (mass-to-charge) for positive polarity.   3.2.7 CO Chemisorption  A Micromeritics AutoChemII 2920 analyzer was used for CO pulse chemisorption of the reduced catalysts in order to determine the dispersion of PdO active sites. In the analysis, the oxidized catalyst was purged in a 50 cm3(STP).min-1 flow rate of Ar (Praxair, UHP) at 200°C for 2h in order to remove moisture. The catalyst was then cooled to 100°C and held for 1h. After degas, a 50 cc.min-1 flow rate of 9.5(v/v)% H2/Ar (Praxair) at 100°C was fed to the catalyst for 1h, and then cooled to 25°C in He [88]. The purpose of flowing H2/Ar is to partially reduce the catalyst so that a thin layer of PdO is transformed to Pd0 that is able to adsorb CO, without affecting the size of the supported PdO particle [74]. The CO uptake was measured by passing pulses of 9.93 (v/v)% CO/He (Praxair) at 25°C over the partially reduced catalyst. The CO pulse injection continued until no additional chemisorption was observed. The CO uptake was measured using a thermal conductivity detector (TCD). Linear CO chemisorption is assumed, so that each surface site (Pd surface atom) is occupied by one CO molecule (stoichiometric factor of 1). Details of the CO uptake and the calculated Pd dispersion is provided in Appendix B.4.  51  Details of the calculations to determine the standard deviations of the BET, XRD, XPS, and CO chemisorption analyses are reported in Appendix D.  3.3 Catalyst Testing  3.3.1 Experimental Setup  The catalyst CH4 oxidation activity was measured in a fixed-bed micro-reactor, operating at atmospheric pressure as shown in Figure 3.1. The stainless steel reactor, with 7.0 mm inner diameter, was placed inside an electric tube furnace with a PID temperature controller. The catalyst (90-354 µm mesh size) was diluted with inert SiC of the same size to ensure isothermal operation through the 4.2 cm length of the catalyst bed. The temperature of the bed was monitored using a K-type thermocouple, located in the middle of the catalyst bed. The desired flow rates of each gas in the inlet feed were controlled using electric mass flow controllers (Brooks 5850 TR). The desired inlet gas mixture was obtained using CH4 (9.93(v/v)%CH4/He, certified purity or 0.76(v/v)%CH4/Ar, certified purity), O2 (Praxair, UHP), He (Praxair, UHP), Ar (Praxair, UHP), and air (Praxair, Extra dry air). He and Ar were used as inert gases to dilute the CH4/O2 mixture and obtain the desired flow rate and CH4 concentration in the inlet gas. Air was used for in situ calcination before the experiment. The outlet gas flow entered a quadrupole mass spectrometer (QMS) to quantify the conversion of CH4.   In cases where water was added to the reactant feed gas, liquid water was pumped into the dry feed gas using a Harvard Apparatus Syringe Pump (Model 44). To ensure the water converts to vapor phase, the feed gas mixture with water passed through a pre-heater held at 52  120°C before entering the reactor. All the feed gas and product lines were heated to the same temperature as the pre-heater in order to ensure that the water remained in the vapor phase. Finally, the outlet gas flow from the reactor passed through a cold-trap followed by a silica gel absorber to remove water before continuous analysis by the mass spectrometer.   53  9.93%CH4 balance He?O2? ?He?AirPressure gaugeFilter? Gate valveCheck valveFlow controller?Heated line  ? CondenserSignalCondensed waterDeionized waterWater pumpPre-heater Tubular furnacePacked bed micro-reactor Mass spectrometer?Gas ventGas ventThermocouple Figure 3.1. Schematic diagram of CH4 oxidation setup 54  The outlet gas flow including the reactants and products were analyzed by a VG ProLab quadrupole mass spectrometer (ThermoFisher Scientific) and a RGA-200 quadrupole mass spectrometer (Stanford Research Systems). The outlet gas was continuously monitored by the mass spectrometer. The signal intensity of the mass peaks for CH4, O2, CO2, He, Ar, and H2O were recorded (see Appendix E.1). The mass spectrometer was calibrated for CH4, CO2 (0.5(v/v)% CO2/Ar or 0.1859(v/v)%CO2/Ar), and O2 in the same concentration range as the CH4 oxidation reaction (see Appendix C.2) and this calibration was used to determine the product gas compositions.  3.3.2 Temperature Programmed Oxidation    Temperature Programmed Oxidation (TPO) tests were done to measure the initial activity of the catalysts. 0.1 g of the dried xCe/yPd/Al2O3 catalyst with 90-354 μm mesh size was diluted with 2.5 g inert SiC with the same mesh size as the dried catalyst and loaded into the reactor. Then the mixture was calcined in situ in 100 cm3(STP).min-1 of dry air heating to 450°C at 10°C.min-1 and holding at this temperature for 15h before cooling to room temperature. Then the gas feed, with a total flow rate of 300 cm3(STP).min-1 and GHSV of 180,000 cm3(STP).gcat-1.h-1, consisting of 0.1(v/v)% CH4, 20(v/v)% O2 balanced with He and Ar was fed over the calcined catalyst. The reactor temperature was increased from room temperature to 450°C at 5°C.min-1 and the outlet gas was continuously monitored by the VG ProLab quadrupole mass spectrometer to measure the CH4 conversion as a function of temperature.    55  3.3.3 Time-on-Stream Experiments  The stability of the catalysts was measured using Time-on-Stream tests in which the CH4 conversion was measured over a 24h period at a constant temperature and H2O concentration. The TOS results reported in Chapters 4 and 5 were generated using 0.0833 g of dried catalyst diluted with 2.1g inert SiC. Similar to the TPO experiment, the catalyst was calcined in situ with 100 cm3(STP).min-1 of dry air at 450°C (10°C.min-1) for 15h and cooled to room temperature. Then the total feed gas flow rate of 250 cm3(STP).min-1 corresponding to a GHSV of 180,000 cm3(STP).gcat-1.h-1 with 0.5(v/v)% CH4, 20(v/v)% O2, 0 or 5vol.% H2O balanced with He was fed to the calcined catalyst. Two different water concentrations (0 and 5vol.% water) in the feed gas are referred to as "dry-TOS" and "wet-TOS" tests, respectively. In the dry-TOS experiments, the temperature was increased from room temperature to the desired temperature (300°C, 330°C, 350°C, 380°C, or 400°C) at 5°C.min-1 in the presence of the reactants. The temperature was then held constant as the reaction proceeded for a period of 24h. A similar heat-up procedure was followed for the wet-TOS experiments, except that water was added to the dry feed using a Harvard Apparatus Syringe pump (model 44) to obtain a 5vol.% H2O in the feed gas once the reaction temperature had stabilized at the desired reaction temperature. The conversion of CH4 at a constant temperature and constant feed gas flow rate was measured and the variation in CH4 conversion with TOS was determined using the RGA-200 quadrupole mass spectrometer. After 24h wet-TOS the syringe pump conveying water to the system was stopped and the experiment was followed by a dry-TOS for 2h. The rate of increase in CH4 conversion upon removing water and conversion in dry-TOS was monitored by the RGA-200 quadrupole 56  mass spectrometer. Once the experiment was completed, the condensate water collected in the cold trap was measured.   The Time-on-Stream was chosen as 24h to ensure enough time to observe loss in CH4 conversion and stability. The temperature and reactant concentrations were chosen based on the NGV exhaust conditions (temperature ≤ 400°C and CH4 concentration < 5000 ppm) and water concentration was set at 5vol.% (lower than the water concentration in NGV exhaust gas) to be able to observe the loss in the activity of the catalysts.  The TOS results reported in Chapter 6 were examined using 0.1g of dried catalyst diluted with 2.5g inert SiC. Similar to the TPO experiment, the catalyst was calcined in situ. Then the total feed gas flow rate of 300 cm3(STP).min-1 corresponding to a GHSV of 180,000 cm3(STP).gcat-1.h-1 containing 0.1(v/v)% CH4, 20(v/v)% O2, 0, 1, 2, or 5vol.% H2O balanced with Ar and He was fed to the calcined catalyst. Similar to the TOS experiments in Chapters 4 and 5, the 0vol.% water in the feed gas is referred as "dry" test and 1, 2, and 5vol.% water in the feed gas is referred as "wet" test. Once the reaction temperature reached the desired temperature (310°, 330°C, 350°C, or 370°C) by heating at 5°C.min-1 the water was added to the dry feed gas and then held for 24h. The CH4 conversion was measured using the VG ProLab quadrupole mass spectrometer. Similar to the wet-TOS in Chapters 4 and 5, the syringe pump was stopped after 24h wet-TOS and a 2h dry-TOS experiment was run. The rate of increase in CH4 conversion upon removing water and conversion in dry-TOS was monitored by the VG ProLab quadrupole mass spectrometer. Once the experiment was completed, the condensate water collected in the cold trap was measured.   57  Table 3.1 summarizes the choice of conditions for the CH4 oxidation tests used in this study. In most cases, the tests have been done at conditions close to those that would be encountered in a real NGV exhaust, except for the O2 content of the feed gas. However, as noted in Chapter 4, the kinetics of the CH4 oxidation reaction is known to be approximately zero order in O2 partial pressure, so this deviation is not expected to have a significant impact on the conclusions drawn from this study.  Table 3.1. A comparison of the reaction conditions used in the present study and real NGV operating condition Parameter Definition This study NGV operating condition [12] T Reaction temperature range (°C) 310-400 500-550 P Total pressure (kPa) 101.325 101.325 yCH4 CH4 concentration (ppm) 1000, 5000 500-1500 yO2 O2 concentration (vol.%) 20 2-12 GHSV Gas hourly space velocity (cm3.gcat-1.h-1) 180,000 200,000 yH2O H2O concentration range (vol.%) 0-5 10-15 XPd Pd loading range (wt.%) 0.3-6.5 0.5-5 yCO2 CO2 concentration (vol.%) - 15 yNOx N2 concentration (vol.%) - Trace ySOx SOx concentration (ppm) - 1  3.4 Catalyst Activity Calculation  3.4.1 CH4 Conversion Calculation  CH4 conversion is calculated based on the mole balance of carbon at the exit of the reactor. Assuming a complete oxidation of CH4 occurring in the reactor, CO2 and H2O are produced according to the reaction:                 . The total moles of carbon is constant through the reactor and can be calculated from the measured composition of the reactor exit gas:  58                                                                                                                              3.1 Analyzing the reactor exit gas by the mass spectrometer confirmed the formation of CO2 as a product of CH4 oxidation (no CO signal was observed by mass spectrometer). In addition, some preliminary CHNS elemental analyses for the used catalysts confirmed no coke deposition on the catalysts, indicating no remaining C during CH4 oxidation reaction. These observations verify the validity of Equation 3.1. Knowing the total carbon and CH4 content exiting the reactor, the CH4 conversion is readily calculated by Equation 3.2.                                                                                                                              3.2 Further details of the calculations are given in Appendix E.  The properties of the fresh catalysts were determined prior to the CH4 oxidation reaction using various characterization techniques described in this chapter. The same characterization techniques were used for the used catalysts in order to verify the effect of aging and H2O adsorption on the properties of the catalysts. Initial activity of the catalysts was determined by Temperature Programmed Oxidation (TPO) tests, while the stability of the catalysts was determined using Time-on-Stream (TOS) tests with/without extra H2O. A reactor design equation combined with the kinetic model were applied to the experimental data obtained from the TOS experiments in order to calculate the apparent activation energy (Ea),      , and kr for PdO/Al2O3 and CeOx/PdO/Al2O3 catalysts.     59  Chapter 4: Effect of Pd Loading on the Activity and Stability of Pd/Al2O3 Catalysts in the Presence of H2O  4.1 Introduction  In this chapter, the effect of different Pd loadings on the activity and stability of the catalysts during CH4 oxidation was investigated. The Pd loading increased from 0.3wt.% to 6.5wt.%. The goal was to study the effect of Pd loading on the physical and chemical properties of the catalysts as well as their initial activity and stability in the presence of H2O. In addition, the effect of Pd dispersion on the amount of H2O adsorbed on active sites was examined in this chapter.  4.2 Results  The effect of varying Pd loading on the properties of the catalysts such as BET surface area, XRD crystallite size, Pd dispersion, Pd oxidation states along with the effect of Pd loading on the catalyst activity for CH4 oxidation reaction are presented in this section.   4.2.1 Catalyst Properties  The properties of the catalysts with 0.3wt.%, 2.6wt.% and 6.5wt.%Pd loadings are summarized in Table 4.1. The BET surface areas of the catalysts show a small increase from 207 m2/g to 218 m2/g by increasing the amount of Pd from 0.3% to 6.5%. The BET surface area of the 6.5Pd/Al2O3 catalyst is close to that of the Al2O3 support (224 m2/g), and the pore volume and pore size decrease with increased Pd loading. The CO uptake results show 60  higher amounts for CO adsorption at higher loadings of Pd. The Pd dispersion obtained by CO chemisorption was 57.0% for the 0.3Pd/Al2O3 catalyst decreasing to 48.0% for the 2.6Pd/Al2O3 and 33.5% for the 6.5Pd/Al2O3 catalyst, indicating lower Pd dispersion obtained at higher Pd loading.   Table 4.1. Properties of PdO catalysts with different loadings of Pd over Al2O3 Catalyst BET Pore Pore CO Pd PdO SAa Volumea Sizea Uptakeb Dispersionb C. Sizec m².gcat-1 cm3.gcat-1 nm μmol.gcat-1 % nm 0.3Pd/Al2O3 207 0.47 9.1 35 57.0 - 2.6Pd/Al2O3 215 0.45 8.3 119 48.0 6 6.5Pd/Al2O3 218 0.43 7.9 204 33.5 6 a Determined by BET            b Obtained by CO chemisorption  c PdO (101) crystallite size obtained by XRD  Figure 4.1(a) shows the XRD analysis for 0.3Pd/Al2O3 catalyst. The main peak for PdO (101) appeared at 2θ=39.59°. This peak is also observed at the same 2θ for 2.6Pd/Al2O3 and 6.5Pd/Al2O3 catalysts in Figure 4.1(b) and 4.1(c). As a result of higher loadings of Pd, a new peak at 2θ=64.50° appears for 2.6Pd/Al2O3 and 6.5Pd/Al2O3 catalysts that corresponds to PdO (112). The PdO (101) crystallite size was calculated by the Scherrer equation and is reported in Table 4.1. Because of the low concentration of the Pd in the 0.3Pd/Al2O3 catalyst, it was not possible to calculate the PdO crystallite size for this catalyst. The crystallite size calculated for the 2.6Pd/Al2O3 and 6.5Pd/Al2O3 catalysts were the same at 6 nm.   Figure 4.2 presents the XPS Pd 3d spectral analysis for the 0.3Pd/Al2O3, 2.6Pd/Al2O3,   61  and 6.5Pd/Al2O3 catalysts. The binding energy (B.E.) of the Pd 3d5/2 and 3d3/2 electrons, the surface compositions of Pd, Al, and O, as well as the Pd/Al ratio on the surface of each catalyst is reported in Table 4.2. The B.E.s for Pd 3d5/2 and Pd 3d3/2 remain almost the same for 0.3Pd/Al2O3, 2.6Pd/Al2O3 and 6.5Pd/Al2O3 catalysts.  10 20 30 40 50 60 70 80  Intensity (a.u.)2theta (°)      (a) (b) (c)      Figure 4.1. XRD patterns for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, and (c) 6.5Pd/Al2O3. ∆ PdO, ● Al2O3  Comparing the surface compositions indicates higher amounts of Pd at higher loadings of Pd and the Pd/Al ratio obtained by XPS analysis also shows an increase from 0.4% to 2.3% with increasing Pd loading from 0.3% to 6.5%. The higher Pd/Al ratio on the surface of the catalysts is a consequence of higher Pd loading. Note that the Pd/Al (%) is calculated from the normalized values of surface concentration, excluded C content, as measured by XPS. 62  Table 4.2. Pd 3d spectra for catalysts with different loadings of Pd Catalyst Binding Energy  Surface Composition Surface eV  (at.%) Pd/Al  Pd 3d5/2 Pd 3d3/2  Pd Al O % 0.3Pd/Al2O3 337.0 342.3  0.16 36.69 63.15 0.4 2.6Pd/Al2O3 336.9 342.1  0.53 35.61 63.86 1.5 6.5Pd/Al2O3 336.8 342.1  0.85 37.30 61.84 2.3  350 348 346 344 342 340 338 336 334 332 330(b)PdOIntensity (a.u.)Binding Energy (eV)(c)3d3/2PdO3d5/2(a) Figure 4.2. XPS Pd 3d spectra measured for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, and (c) 6.5Pd/Al2O3  63  4.2.2 Catalyst Activities  Figure 4.3 reports the initial activity of the calcined 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts using a dry feed consisting of 1000 ppmv CH4 in TPO tests. Higher conversion was obtained for the catalysts with higher Pd loading due to a higher number of active sites within the reactor. Table 4.3 presents T10, T50, and T90 corresponding to the temperature required for 10%, 50%, and 90% CH4 conversion.   150 200 250 300 350 400020406080100 0.3% Pd/Al2O3 2.6% Pd/Al2O3 6.5% Pd/Al2O3CH4 Conversion (mol%)Temperature (°C)   Figure 4.3. Temperature Programmed Oxidation profile. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppmv CH4, 20(v/v)% O2, and the balance He and Ar    64  Table 4.3. Light-off temperatures for 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts Catalyst T10 T50 T90 °C °C °C 0.3Pd/Al2O3 267±6 333±4 391±6 2.6Pd/Al2O3 232±6 289±4 332±6 6.5Pd/Al2O3 203±6 249±4 280±6  The catalysts were tested for CH4 oxidation to assess their stability in the presence of extra H2O. Figures 4.4 and 4.5 show the TOS results for a "dry" and "wet" feed gas with 5vol.% extra H2O over a 24h period at temperatures 330°C and 350°C. All catalysts show a slow loss of activity with TOS in the dry feed gas at 330°C, however, the loss is slower for lower Pd loadings. In wet feed gas with 5vol.% extra H2O all catalysts showed a much faster loss of activity than the dry feed gas. An exponential loss in activity occurred immediately after the 5vol.% extra H2O was added for all three catalysts. The exponential loss continued for about 3h, 4h, and 5h for the 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts, respectively, followed by a linear loss up to TOS=24h. The shorter exponential loss in activity that occurred for the 0.3Pd/Al2O3 catalyst, suggesting that the H2O adsorption reaches equilibrium faster in the case of 0.3Pd/Al2O3 catalyst. This is a result of lower Pd/Al on the catalyst surface, indicating a higher possibility of active site coverage by H2O on the 0.3Pd/Al2O3 catalyst than the other catalysts. Once the active sites coverage by H2O reaches equilibrium, the minimum CH4 conversion value is obtained and then stabilizes at that value. The TOS results at 350°C show higher CH4 conversion and less loss in activity of the catalysts compared with those at 330°C. For instance, after 24h TOS at 350°C the CH4 conversion for the 0.3Pd/Al2O3 catalyst decreased from 56.2% to 38.8% in the dry feed gas and to 3.8% in the wet feed gas, comparing the data at 330°C and 350°C shows higher CH4 65  conversion after 24h TOS at 350°C than 330°C for 0.3Pd/Al2O3 catalyst. This observation indicates less H2O adsorption at higher temperatures for all 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts.  66    Figure 4.4. TOS results for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, (c) 6.5Pd/Al2O3 at T=330°C for dry (open symbol) and wet (closed symbol) conditions. GHSV=180,000 cm3(STP).gcat-1.h-1, 5000 ppmv CH4, 20(v/v)% O2, and the balance He 0 5 10 15 20 250204060801000 5 10 15 20 25 0 5 10 15 20 25Remove H2OCH4 Conversion (mol.%)Time on Stream (h)(a) 0.34Pd/Al2O3 Remove H2O(b) 2.6Pd/Al2O3Time on Stream (h)1Pd, 5Pd, 10Pd at T=330 C (dry and wet)Xdry-wet (c) 6.5Pd/Al2O3Time on Stream (h)Remove H2O   0.3Pd/Al2O3                                                                      2.6Pd/Al2O3                                                                    6.5Pd/Al2O3 Remove H2O                                                      Remove H2O                                                      Remove H2O                                                      67     Figure 4.5. TOS results for (a) 0.3Pd/Al2O3, (b) 2.6Pd/Al2O3, (c) 6.5Pd/Al2O3 at T=350°C for dry (open symbol) and wet (closed symbol) conditions. GHSV=180,000 cm3(STP).gcat-1.h-1, 5000 ppmv CH4, 20(v/v)% O2, and the balance He0 5 10 15 20 250204060801000 5 10 15 20 25 0 5 10 15 20 25Remove H2OCH4 Conversion (mol.%)Time on Stream (h) Remove H2OTime on Stream (h) Remove H2O1Pd, 5Pd, 10Pd at T=350 C (dry and wet)Time on Stream (h) 0.3Pd/Al2O3                                                                      2.6Pd/Al2O3                                                                    6.5Pd/Al2O3 Remove H2O                                                      Remove H2O                                                      Remove H2O                                                      68  4.3 Kinetic Model   To quantify the effect of different Pd loadings on the activity and stability of the catalysts, a kinetic analysis of the activity data has been completed. As noted in Chapter 2, the negative effect of H2O on CH4 oxidation has been investigated in several studies [72,76,78,79]. In a study by Ribeiro et al. [72] it was confirmed that the order of the CH4 oxidation reaction is affected by extra H2O added to the feed gas. The reaction order with respect to H2O ranges from -1.3 to -0.9 at low temperatures varying from 300°C to 350°C [77] and the activation energy needs to be corrected in order to account for the inhibiting effect of H2O [31]. van Giezen et al. [76] reported activation energies of 86 kJ.mol-1 and 151 kJ.mol-1 for 7.3%Pd/Al2O3 catalyst in dry feed and wet feed, respectively [76].   The fixed-bed micro-reactor was modeled for the reaction conditions with an inlet feed stream of 5000 ppmv CH4 and 0-5vol.%H2O entering the catalyst bed at a fixed temperature. For a single component (CH4 only), the overall mole balance equation in the reactor is                                [89].              represents the CH4 accumulation per unit volume,       is the porosity of the catalyst bed,         and     are the net rate of CH4 addition per unit volume caused by convection and diffusion, respectively,      is the actual overal rate of CH4 oxidation reaction with the unit of mol.(cm3.s)-1,   is the internal effectiveness factor, and      is the gas molar density or concentration in mol/cm3. A one dimensional packed bed reactor model can be assumed given the high ratio of catalytic bed length to catalyst particle diameter,       =203. Generally, in a packed bed reactor the gas velocity is higher near the wall because of the lower density of the catalyst bed [90,91]. 69  Therefore, the CH4 concentration gradient in the radial direction can be critical. The ratio of the bed diameter to the catalyst particle diameter,       , higher than 10 can reduce the radial direction effect. In the present study, this ratio is 32. Therefore, the radial direction is neglected. In addition, the     term (diffusion transport) is negligible compared to         (convective transport) as a result of the high superficial velocity ( =0.24 m.s-1) of the gas through the reactor, therefore the mole balance for CH4 is simplified as Equation 4.1:                                                                                                                               4.1  Figure 4.6. Schematic of the reactor used in the CH4 oxidation process  Since the CH4 oxidation reaction occurs in the catalyst bed, Equation 4.1 is transformed to Equation 4.2 where the mole balance is a function of mass of catalyst (    ).                                                                                                                                    4.2       and      represent the mass and density of the catalyst as explained in Appendix H.1.    in Equation 4.2 represents the total volumetric flowrate written as    (cm3.s-1). Equation 4.2 is re-written as: 70                   -                                                                                                                   4.3  where t is the time-on-stream (s),               represents the net (inlet-outlet) molar flow rate,      is the CH4 oxidation reaction rate on a mass basis (mol.gcat-1.s-1) and                  is the accumulation term. The constant values of ν0, ε    , and      of Equation 4.3 are reported in Table 4.4. The calculation procedure for ε    and     are presented in Appendix H.1.   Table 4.4. The constant values of ν0,     , and      used in Equation 4.3     ε          cm3(STP).s-1  gcat.(cm3)-1 4.17 0.57 0.051   The calculations for Mears criterion given in Appendix H.2 confirm a negligible external mass transfer effect for the porous catalyst bed emphasizing a rapid mass transfer from the bulk to the catalyst surface, i.e.            , where       and       represent the CH4 concentration in the bulk phase and on the catalyst surface, respectively. The internal mass transfer effect is considered using the effectiveness factor calculated by Equations 4.4-4.6:                                                                                                                                                          4.4 The internal effectiveness factor was calculated based on the 1st-order reaction rate assumption as follows [92]:                                                                                                                                   4.5 where   is Thiele modulus defined as:  71                                                                                                                                        4.6   ,  , and       are the catalyst particle diameter (m), CH4 oxidation reaction rate constant (s-1), and effective diffusivity (m2.s-1), respectively. The rate constant,  , used in Equation 4.6 is calculated from the rate of CH4 oxidation reaction:                                                                                                                                   4.7 where       is the CH4 reaction rate on a volumetric basis (mol.(cm3.s)-1), so that:         ρ                                                                                                                   4.8 Hence we assume that:                                                                                                                                    4.9 By substituting Equation 4.9 in Equation 4.6,   and η  can be calculated provided ks is known. Note that the rate constant in Equation 4.9 is considered to be a function of temperature only. Details of the calculations for effective diffusivity,     , are provided in Appendix H.1.   In the present study, following the report by Kikuchi et al. [22], the rate of CH4 oxidation was assumed 1st-order in CH4 and dependent on the number of active sites as follows:                θ                                                                                                            4.10 where        is the CH4 oxidation reaction rate per mole of active sites as measured by CO uptake (mol.molsite-1.s-1),    is the rate constant (mol.gcat.molsite-2.s-1.Pa-1),      is the partial 72  pressure of CH4 (Pa) defined as                    ,    is the total number of active sites (molsite.gcat-1), and θ  is the fraction of vacant site pairs (both PdO and Pd-*). On a mass basis the rate (mol.gcat-1.s-1) is given by Equation 4.11 as:       =                                                                                                                         4.11  The rate equation follows from several studies that report the order of reaction with respect to CH4 and O2 as 1 and 0, respectively [71,76,93] This implies that the O2 content of the feed gas does not impact the reaction rate significantly and consequently, the relatively high O2 content of the present study (20%) should not impact the kinetic analysis. High O2 partial pressure ensures that the re-oxidation of the catalyst by gas phase O2 is not O2 limited. The reaction order for CO2 is also assumed to be zero, as reported in the literature [76,78] Since the adsorption of CH4 on the active sites is irreversible, as explained by Kikuchi et al. [23] and Ciuparu and Pferfferle [30] CH4 is rapidly consumed, indicating a near zero coverage of active sites by CH4 [30]. In addition, rapid CO2 desorption implies very low coverage of the active sites by CO2 [30]. As discussed by Ciuparu and Pferfferle [30], at low temperatures (< 500°C), and moderate to high H2O content (3.5vol.%), the surface blockage caused by hydroxyl formation is significant. Therefore, the fraction of vacant sites, θ , is given by Equation 4.12:                                                                                                                                4.12  Since the reactor is not at steady state, we further assume that the adsorption-desorption of the hydroxyl/H2O molecules on the catalyst surface is not at equilibrium initially [17] The 73  accumulation of hydroxyl/H2O on the active site pair can be described by Equations 4.13 and 4.14.                                                                                                                          4.13  θ            θ    θ                                                                                                     4.14    and    are the rate constants for H2O adsorption and H2O desorption, respectively.      is determined by stoichiometry as   α                and α    is defined as            at the inlet of the reactor. Substituting Equation 4.12 into Equation 4.14 and integrating, assuming that at t = 0, θ       results in Equation 4.15: θ                                                                                                       4.15                                                                          is the H2O adsorption equilibrium constant, defined as kf kr .  Solving a PDE (Equation 4.3) which is a function of both     and t is readily achieved using the Method of Characteristics [94]. Accordingly, for a PDE such as               with u is defined as         , and a, b, and c are functions of x, y, and u only (              nor        ), the PDE can be re-written as below:                                                                                                                               4.16  Applying the Method of Characteristics [94] to Equation 4.3 results in Equations 4.17 and 4.18:                                                                                                                                         4.17       74                                                                                                                                           4.18                                                                                                             After substituting Equations 4.11, 4.12, and 4.15 into Equation 4.17, and writing                   , where XCH4  is the CH4 conversion, yields Equation 4.19:                    η                                                                           4.19  where               (mol.s-1) is the molar flow rate of CH4 in the inlet of the reactor.  Note that as the TOS increases, i.e. t→∞, which for most of the data reported in the present study corresponds to t  5h, Equation 4.19 reduces to Equation 4.20 when written in terms of  XCH4:                    η                           α                                                                                                  4.20  The temperature dependence of the three unknown parameters of Equations 4.19 and 4.20 are given by:                                                                                                                             4.21                                                                                                                    4.22                                                                                                                             4.23 75  Note that Equation 4.20 represents the steady-state mole balance equation for the reactor system.   The TOS results presented in Figures 4.4 and 4.5 were used to estimate the model parameters    , Ea,      ,       associated with Equation 4.20. Accordingly, the steady-state conversions, taken as Xs (the stable CH4 conversion at t=5h since the CH4 conversion does not change significantly after this time), were regressed onto Equation 4.20 using the Levenberg-Marquardt non-linear regression algorithm [95] combined with a numerical integration of Equation 4.20 using a 4th-order Runge-Kutta algorithm. The Matlab code used to complete these numerical calculations is reported in Appendix K.   The model parameters were estimated from data obtained at different temperatures and different H2O concentrations for all 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts. The different dispersions of the catalysts are accounted for in the model through the measured CO uptakes (CT). The inherent assumption of this approach is that the reaction is not PdO structure sensitive in the narrow range of dispersion since we assume that the same kinetic parameters apply to all three catalysts, despite their varying dispersions. Table 4.5 reports the constant parameters used in Equations 4.6 and 4.20 to estimate η,    and     . Note that the effective diffusivity at different temperatures was calculated assuming that Deff varied at T0.5. Figure 4.7 shows the estimated     values from the model versus the measured     values from Figures 4.4 and 4.5 at t=5h with the adjusted R2=0.98.    76  Table 4.5. Constant values for 6.5Pd/Al2O3 catalyst at T=330°C used in Equation 4.20 Wcat (g) 8.33×10-2 Pd (wt.%) 6.5       (mol.s-1) 373.6        0.005 P (Pa) 101325 GHSV (mol.gcat-1.s-1) 2.23×10-3 CT (molsite.gcat-1) 2.04×10-4       at 330°C (m2.s-1) 1.00×10-6 ρcat (g.cc-1) 1.49 dp (cm) 2.22×10-2      0 and 10   The values of effectiveness factor, rate constant, equilibrium constant for H2O adsorption were estimated from the kinetic model, and the reaction rate values (       ) were calculated by replacing the estimated     values in Equation 4.11. The values at different temperatures for the 6.5Pd/Al2O3 catalyst are presented in Table 4.6. By increasing the temperature from 300°C to 380°C, the reaction rate shows an order of magnitude increase. Decreasing η with increased tremperature reflects the faster increase in intrinsic reaction rate versus actual reaction rate at higher reaction temperatures.  Table 4.6. Estimated values of η, rate constant, equilibrium constant for H2O adsorption, and reaction rate at different temperatures for 6.5Pd/Al2O3 catalyst T η                °C mol.gcat.molsite-2.Pa-1.s-1 Pa-1 mol.(gcat.s)-1 300 0.47 7.2 1.2510-2 1.8410-6 330 0.36 13.5 5.3210-3 6.2110-6 350 0.30 19.9 3.1610-3 1.1710-5 380 0.24 34.1 1.5310-3 1.5810-5  The estimated parameters of Equations 4.21 and 4.22 are reported in Table 4.7.  77  Table 4.7. Estimated values obtained from the design equation for CH4 oxidation over Pd/Al2O3 catalysts with different Pd loadings                    kJ.mol-1 mol.gcat.molsite-2.Pa-1.s-1 kJ.mol-1 Pa-1 60.6±11.5 13.5±1.7 -81.5±9.1 5.310-3±4.310-4  0 20 40 60 80 100020406080100Calculated CH4 Conversion (mol.%)Measured CH4 Conversion (mol.%)   Figure 4.7. Calculated      values from the kinetic model versus measured       values from the experiments for xPd/Al2O3 catalysts  The value obtained for the apparent activation energy for CH4 conversion, 60.6±11.5 kJ.mol-1 is lower than the 151 kJ.mol-1 reported in the literature [76] for a 7.3wt%Pd/Al2O3 catalyst. Since the value of -81.5±9.1 kJ.mol-1 was estimated for      ,  the apparent activation energy considering the H2O effect is calculated as:                                                                                                                                 4.24 78  Hence, the value of 142.1 kJ.mol-1 is obtained which is consistent with other studies [76]. Since the same model parameter values provided a good fit to all the data, regardless of PdO dispersion and H2O content, we conclude that the CH4 oxidation reaction is not structure sensitive.   Upon removing the 5vol.% extra H2O after 24h TOS at both 330°C and 350°C, the CH4 conversion increased to values close to the conversion measured after 24h TOS in the dry feed gas (See Figures 4.4 and 4.5). This observation confirms a partially reversible effect of H2O on the catalyst activity. The        ratio is defined as the H2O concentration at the reactor exit per total number of active sites with units of Pa.gcat.μmolsite-1 with high values implying more coverage of the available active sites by H2O. The amount of H2O results from both       in the inlet of the reactor (0 and 5066.25 Pa for dry and wet feed, respectively) and the amount of H2O produced during the reaction, which varies with CH4 conversion and temperature.    79  Table 4.8. Concentration of H2O per number of active sites as a function of temperature for catalysts with different Pd loadings T=330°C Dry feed  5vol.% extra H2O Catalyst CTa Xb PH2O/CT  Xb PH2O/CT μmolsite.gcat-1 mol.% Pa.gcat.μmolsite-1  mol.% Pa.gcat.μmolsite-1 0.3Pd/Al2O3 35 29.1 8.4  1.9 145.3 2.6Pd/Al2O3 119 53.9 4.6  13.4 43.7 6.5Pd/Al2O3 204 68.5 3.4  27.2 26.2    T=350°C Dry feed  5vol.% extra H2O Catalyst CTa Xb PH2O/CT  Xb PH2O/CT μmolsite.gcat-1 mol.% Pa.gcat.μmolsite-1 mol.% Pa.gcat.μmolsite-1 0.3Pd/Al2O3 35 38.8 11.2  3.8 145.9 2.6Pd/Al2O3 119 64.1 5.5  25.0 44.7 6.5Pd/Al2O3 204 84.3 4.2  42.8 27.8 a Number of active sites      b Conversion measured at t=24h     Comparing the        ratio as a function of Pd loading at a constant temperature and H2O concentration shows higher values of        at lower Pd loadings. The significant decrease in        in the dry feed with increasing Pd loading from 0.3wt.% to 6.5wt.% for both 330°C and 350°C, emphasizes the more significant effect of higher CT values for the 6.5Pd/Al2O3 catalyst. Although higher CH4 conversion for 6.5Pd/Al2O3 catalyst results in higher H2O production (PH2O), the higher CT value of 6.5Pd/Al2O3 catalyst has more impact. The ∆Xdry-wet (the difference between the CH4 conversion in the dry feed after 24h TOS and the wet feed after removing the H2O) values (Table 4.9) are also higher at lower Pd loadings, confirming the negative effect of higher        as shown by the higher values of this ratio at lower Pd loadings (Table 4.8). For both 330°C and 350°C the rate of CH4 conversion recovery varies in this order: 6.5Pd/Al2O3 > 2.6Pd/Al2O3 > 0.3Pd/Al2O3 indicating faster 80  partial recovery at higher Pd loadings as a consequence of lower       . This suggests less Pd-OH bond formation on the PdO crystals having higher Pd loadings.  Table 4.9. ∆Xdry-wet at t=24h for 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts T ∆Xdry-wet °C 0.3Pd/Al2O3 2.6Pd/Al2O3 6.5Pd/Al2O3 330 14.6 6.0 0.1 350 16.7 10.6 5.1  Comparing the        ratio as a function of temperature for each catalyst, shows an increase with both dry feed and 5vol.% extra H2O, however, the increase is more significant for the dry feed, indicative of the larger effect of the produced H2O during the CH4 reaction in the dry feed than the wet feed. The        values are related to the extent of recovery of the catalyst activity when H2O is removed from the feed gas. ∆Xdry-wet for all catalysts is shown in Table 4.9. ∆Xdry-wet at 330°C was 14.6, 6.0, and 0.1 for 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts, respectively. These values increased to 16.7, 10.6, and 5.1 at 350°C. The increase in ∆Xdry-wet values with temperature correlates with the higher        values for 5vol.% extra H2O at higher temperature reported in Table 4.8. Hence with more H2O present there is less recovery of the catalyst activity.   4.4 Discussion  The Pd is present as PdO following calcination of the catalysts in excess O2 at 450°C for 15h. The XRD and XPS analyses confirm the formation of PdO only, since no Pd0 is observed by these analyses. The XPS data show a higher Pd/Al ratio on the surface at higher 81  Pd loadings, as expected [96]. In the study by Cullis et al. [71] the mean diameter of PdO catalysts with 2.7wt.%, 11wt.%, and 25wt.%Pd over γ-Al2O3 increased from 13 nm to 26 nm [71], indicating lower dispersion at higher Pd loading. Stasinska et al. [96] showed the Pd/Al ratio decreased from 0.555 to 0.217 as the Pd crystallite size increased from 4.6 nm to 13 nm on a 0.3Pd/Al2O3 catalyst [96]. They also reported a lower binding energy of Pd 3d in the case of lower Pd dispersion. This could be a result of the presence of more Pd0 formation. As long as the Pd crystallite size increased from 4.6 nm to 9.6 nm, Pd0 surface composition increased from 85.1 (at.%) to 98.0 (at.%) [96]. However, in our study the binding energy of Pd 3d did not show any significant change and no Pd0 was observed. The 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts do not show a significant difference in terms of the BET surface area, PdO crystallite size, or binding energies that is a result of a narrow range of Pd dispersion at different Pd loadings.  The structure sensitivity of Pd based catalysts in CH4 oxidation is not clear. As shown by Castellazzi et al. [69], the TOF is not related to the PdO dispersion as obtained with different Pd loadings. Zhu et al. [73] also showed that the rate of reaction is not structure sensitive and is only dependent on the oxygen-oxygen interaction. Therefore, surface Pd with more adsorbed oxygen improves catalyst activity. Adsorbed hydroxyl groups on the support and PdO correlate with catalyst deactivation [97]. However, the role of Pd dispersion on the inhibiting effect of H2O is not well understood. Stasinska et al. [96] showed the Pd/Al2O3 catalyst with Pd crystallite size smaller than 6.6 nm has the highest activity for CH4 oxidation under lean-burn conditions. For Pd with a crystallite size of 4.6 nm the CH4 conversion reached 100% at 500°C in the absence of extra H2O, however, with 13 nm Pd crystallite size the complete oxidation of CH4 occurred at 705°C in the absence of extra H2O. Upon adding 82  20vol.% H2O, 100% CH4 conversion was reached at 650°C for 4.6 nm Pd and Pd 13 nm did not obtain 100% CH4 conversion even at 750°C [96]. Hence larger crystallites were impacted more by the presence of H2O. However, in the present study it was shown that catalysts with lower Pd dispersion have higher catalyst activity during TPO and TOS experiments due to lower        values and higher Pd loading. Some studies showed the dispersion of Pd decreased after the CH4 oxidation reaction in the presence of H2O [42,75,98–100]. Narui et al. [100] reported a drop in PdO dispersion from 14% to 11% for 0.5%Pd/Al2O3 catalyst after the combustion reaction at 350°C for 6h.   In this study, the effective diffusivity,     , was calculated using the tortuosity factor and constriction factor values as  =3 and σ=0.8, respectively, and the      values for all three catalysts with different Pd loadings were obtained in the order of 10-6    . In a study by Hayes et al. [101] for a washcoated monolith, the tortuosity factor was taken as  =8.1 resulting in a      of 1.710-7    . A value of     =5.610-7     was obtained for  =2.44, showing the sensitivity of the      calculation to the tortuosity factor. In this study, the tortuosity factor was chosen based on the typical value reported in the literature ( =3) [92] and the obtained      value for 6.5Pd/Al2O3 catalyst was 1.0010-6     which is in the range of the reported diffusivity values for gas phase (10-6    ), however, the higher value for  , e.g. 8, results in     =3.7510-7    . Using     =1.0010-6    , low values of the effectiveness factor, η < 1, for the 6.5Pd/Al2O3 catalyst were obtained, indicating internal mass transfer control.   83  The same value of the apparent activation energy and enthalpy of H2O adsorption was applied to all 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts at different reaction temperatures and H2O concentrations, accounting for the number of active sites on each catalyst. Although the initial activity and the loss in CH4 conversion varied between the catalysts with different Pd loadings, the kinetic model that assumes the inhibiting effect of H2O is independent of the Pd dispersion and is governed only by the H2O adsorption equilibrium is shown to describe the catalyst activity data. This confirms the structure insensitivity of the catalysts for the conditions studied.  4.5 Conclusion  The initial CH4 oxidation activity and stability of 0.3Pd/Al2O3, 2.6Pd/Al2O3, and 6.5Pd/Al2O3 catalysts were investigated by TPO, dry-TOS, and wet-TOS experiments. Higher Pd loadings led to a higher Pd/Al ratio on the catalyst surface and lower Pd dispersion. The highest initial activity and the lowest inhibiting effect of H2O in the TOS experiments for the 6.5Pd/Al2O3 catalyst is a result of higher Pd loading. For both dry-TOS and wet-TOS experiments, the amount of H2O adsorbed per active site was the lowest for the 6.5Pd/Al2O3 catalyst. This explains the faster recovery of the 6.5Pd/Al2O3 catalyst compared to the 0.3Pd/Al2O3 and 2.6Pd/Al2O3 catalysts upon removing the extra 5vol.%H2O. H2O adsorption is the main cause of activity loss in CH4 oxidation and the rate of recovery depends on the amount of H2O adsorbed on the active sites. The CH4 oxidation reaction rate constant and the equilibrium constant for H2O adsorption, as well as the correlated apparent activation energy (Ea) and enthalpy of H2O adsorption (       were obtained from the experimental data. Applying the reactor design equation to the values of CH4 conversion 84  (    ) at t=5h with and without H2O in the feed, for all three catalysts, Ea and       were estimated as 60.611.5 kJ.mol-1 and -81.59.1 kJ.mol-1, respectively, indicating that CH4 oxidation reaction is not PdO structure sensitive in the narrow range of studied Pd dispersions (33%-57%). The low values of the effectiveness factors obtained in the range of 300-380°C confirm the slow internal diffusion as a consequence of fast CH4 oxidation reaction.    85  Chapter 5: Reduced Inhibition of CH4 Oxidation by H2O with CeO2 Addition to the PdO/Al2O3 Catalyst  5.1 Introduction  In this chapter, the effect of CeO2 on the inhibition effects of H2O during CH4 oxidation over PdO catalysts at low temperatures has been investigated by monitoring the dynamic response of the CH4 conversion following H2O addition to the CH4/O2/He feed gas. By comparing the observed conversion over a co-Ce/Pd/-Al2O3 and a Pd/-Al2O3 catalyst, combined with catalyst characterization data, the beneficial effects of CeO2 in reducing the inhibition effects of H2O, are demonstrated. Furthermore, the relative importance of H2O adsorption versus PdO sintering or other catalyst deactivation mechanism during CH4 oxidation, is clarified.   5.2 Results   5.2.1 Catalyst Properties  The AAS analysis showed that the average Pd loading of all catalysts was 6.5±0.3wt.%. The measured Ce loadings are reported in Table 5.1, together with other physical properties of the calcined catalysts. Table 5.1 shows that the BET surface area of the 6.5Pd/Al2O3 catalyst (218 m2/g) is close to that of the Al2O3 support (224 m2/g) and that the BET surface area decreases from 208 m2/g to 194 m2/g as the Ce loading increases from 0.9 to 9.5%. The total CO uptake of the 6.5Pd/Al2O3 catalyst is 204 μmol/gcat. However, for the co-0.9Ce/6.5Pd/Al2O3 catalyst the CO uptake is significantly lower (89 μmol/gcat). The reduced uptake of the 0.9Ce/6.5Pd/Al2O3 catalyst may be due to the presence of CeO2 which may 86  limit the PdO→Pd conversion during the reduction step done prior to the CO uptake measurement [74] but then increases with increasing Ce content to 242 μmol/gcat for the co-9.5Ce/6.5Pd/Al2O3 catalyst, confirming increased PdO dispersion (smaller PdO particles) as Ce loading increased. Note that the CO uptake of the 9.4Ce/Al2O3 (no Pd) is 0.39μmol/gcat.   Table 5.1. Properties of calcined PdO, CeO2, and co-xCe/yPd catalysts supported on Al2O3 Catalyst BET SAa m².g-1 Pore Volumea cm3.g-1 Pore Sizea nm CO Uptakeb μmol.gcat-1 Pd Dispersionb % PdO C. Sizec nm CeO2 C. Sizec nm 6.5Pd/Al2O3 218 0.43 7.9 204 33.5 6 - co-0.9Ce/6.5Pd/Al2O3 208 0.42 8.1 89 14.6 6 - co-2.9Ce/6.5Pd/Al2O3 206 0.41 7.9 153 25.1 5 - co-4.8Ce/6.5Pd/Al2O3 196 0.39 8.0 244 37.9 4 - co-9.5Ce/6.5Pd/Al2O3 194 0.36 7.4 242 38.1 4 5 9.4Ce/Al2O3 191 0.41 8.5 0.4 N/A N/A 5 a Determined by N2 adsorption at 77K b Obtained by CO chemisorption c PdO (101) and CeO2 (111) crystallite size obtained by XRD  Figure 5.1b shows the XRD analysis of the 6.5Pd/Al2O3 catalyst with peaks for PdO observed at 2θ = 39.49° and 64.50° corresponding to PdO (101) and PdO (112), respectively. The PdO (101) peak was observed for all Pd-Ce catalysts with different loadings of Ce, however, the peak shifted slightly to lower 2θ as the Ce loading increased from 0.9% to 9.5%. This is a result of an overlap between the PdO (101) peak located at 39.49° and the CeO2 (200) peak at 38.61°. The main peaks for CeO2 appeared at 2θ=33.27°, 38.61°, 55.75° and 66.49° corresponding to CeO2 (111), (200), (220), and (311), respectively. At higher 87  loadings of Ce (4.8 and 9.5%), the peak at 2θ=64.50° appears as an overlap between PdO (112) and CeO2 (311). The PdO crystallite size obtained from XRD analysis is consistent with the CO chemisorption results as a function of Ce loading, since they also show smaller PdO particles for the catalysts with higher loadings of Ce.   10 20 30 40 50 60 70 80(f) (g)(e)(a)(b)(c)(d)Intensity (a.u.)2theta (  )°   Figure 5.1. XRD patterns for (a) γ-Al2O3 (b) 6.5Pd/Al2O3, (c) co-0.9Ce/6.5Pd/Al2O3, (d) co-2.9Ce/6.5Pd/Al2O3, (e) co-4.8Ce/6.5Pd/Al2O3, (f) co-9.5Ce/6.5Pd/Al2O3, (g) 9.4Ce/Al2O3. ∆ PdO, ● Al2O3, ○ CeO2, ■ Ce2O3  Figure 5.2 reports the effect of different Ce loadings on the Pd and Ce surface composition, as measured by XPS. As the catalyst bulk composition increases in Ce (reported as the Ce to Al atom ratio i.e. (Ce/Al)b), the Pd surface concentration Pds increases, indicative of 88  increased Pd dispersion since the Pd bulk composition is relatively constant for these catalysts (the (Pd/Al)b atom ratio varies from 0.033 to 0.037 for the data of Figure 5.2 (not shown in Figure 5.2)). The Ce surface concentration (reported as the surface atom ratio Ces) increases almost linearly up to a (Ce/Al)b ratio of 0.02, indicative of monodispersed Ce species [102]. Further increase in Ce content results in a smaller increase in the Ces ratio, suggesting agglomeration and reduced dispersion of the CeOx. The reported Pds and Ces surface compositions are normalized excluding the C content measured by XPS.  0.00 0.01 0.02 0.03 0.040.00.51.01.52.02.53.03.54.0Surface Composition (at.%)(Ce/Al)bPdsCes   Figure 5.2. Measured Pd atomic percent (○) and Ce atomic percent (■) on the catalyst surface as a function of calculated (Ce/Al)b  Figure 5.3 presents the XPS Pd 3d spectral analysis for the 6.5Pd/Al2O3 and the co-Ce/Pd/Al2O3 catalysts and Table 5.2 summarizes the binding energy (B.E.) of the Pd 3d5/2 89  and 3d3/2 electrons. The B.E.s for all the co-Ce/Pd/Al2O3 catalysts are the same (B.E.=337.0±0.1 eV and 342.3±0.1 eV), with the B.E. of the Pd/Al2O3 marginally lower (336.8 eV and 342.1 eV, respectively), suggesting some charge transfer to the PdO from the CeO2.  350 348 346 344 342 340 338 336 334 332 330PdOBinding Energy (eV)(a)3d3/2(b)(c)3d5/2(e)(d)PdOIntensity (a.u.) Figure 5.3. XPS Pd 3d spectra measured for (a) 6.5Pd/Al2O3, (b) co-0.9Ce/6.5Pd/Al2O3, (c) co-2.9Ce/6.5Pd/Al2O3, (d) co-4.8Ce/6.5Pd/Al2O3, (e) co-9.5Ce/6.5Pd/Al2O3 90  Table 5.2. Pd 3d spectra for Pd/Al2O3 and co-xCe/yPd/Al2O3 with different loadings of Ce Catalyst Pd 3d5/2 Pd 3d3/2               B.E. B.E. % eV eV 6.5Pd/Al2O3 336.8 342.1 - co-0.9Ce/6.5Pd/Al2O3 337.0 342.3 - co-2.9Ce/6.5Pd/Al2O3 337.1 342.4 28.8 co-4.8Ce/6.5Pd/Al2O3 337.0 342.3 23.3 co-9.5Ce/6.5Pd/Al2O3 337.0 342.3 22.2 9.4Ce/Al2O3 - - 17.4   The Ce 3d spectra for the co-4.8Ce/6.5Pd/Al2O3, co-9.5Ce/6.5Pd/Al2O3 and 9.4Ce/Al2O3 catalysts are presented in Figure 5.4. Because of the complexity of the spectra, the XPS analysis of Ce is restricted to the 3d level from 872 to 925 eV and includes a mixture of Ce2O3 and CeO2 oxidation states [103]. Two main peaks for      (Ce2O3) located at 885.8 eV (v') and 903.6 eV (u') are attributed to 3d5/2 and 3d3/2 electrons, respectively. Six main peaks for      (CeO2) are located at 882.7 eV (v), 888.6 eV (v"), 898.3 eV (v‴) for 3d5/2, and 900.8 eV (u), 907.3 eV (u"), and 916.7 eV (u‴) allocated to 3d3/2 [104–109]. The presence of the peak at 916.7 eV (u‴) in all samples indicates that CeAlO3 was not formed during calcination [109]. The fraction of      was determined for each sample from the fitted peak areas. Table 5.2 presents the               ratio for the co-2.9Ce/6.5Pd/Al2O3, co-4.8Ce/6.5Pd/Al2O3, co-9.5Ce/6.5P/Al2O3, and 9.4Ce/Al2O3 catalysts. The               ratio increased with decreased Ce content from 17.4% for the 9.4Ce/Al2O3 to 28.8% for co-2.9Ce/6.5Pd/Al2O3 catalyst. 91  930 920 910 900 890 880 870(d)u"u"' u'v'"v"v' vu(c)(b)Intensity (a.u.)Binding Energy (eV)(a) Figure 5.4. XPS Ce 3d spectra measured for (a) co-2.9Ce/6.5Pd/Al2O3, (b) co-4.8Ce/6.5Pd/Al2O3, (c) co-9.5Ce/6.5Pd/Al2O3, (d) 9.4Ce/Al2O3  This increase is attributed to the transition of      to      following addition of Pd to the Ce/Al2O3 and is indicative of the interaction between Pd and the Ce surface species of the 92  calcined catalysts, with O transfer from the CeO2 to the PdO/Pd-*, consistent with the Pd XPS analysis.                                    5.2.2 Catalyst Activities  Figure 5.5 reports the TPO results for the 6.5Pd/Al2O3 catalyst and the Ce promoted 6.5Pd/Al2O3 catalysts. The 6.5Pd/Al2O3 catalyst had the highest activity for CH4 oxidation with T50 of 251°C. Increased Ce content resulted in reduced activity and the 9.4Ce/Al2O3 catalyst (not shown) was the least active with a T50 of 595°C. Table 5.3 presents the light-off temperatures corresponding to 10%, 50%, and 90% CH4 conversion and these data show similar trends in terms of catalyst activity. Among the Ce-promoted catalysts, the co-2.9Ce/6.5Pd/Al2O3 catalyst has the highest activity.  93  150 200 250 300 350020406080100 6.5% Pd 0.9% Ce-6.5% Pd 2.9% Ce-6.5% Pd 4.8% Ce-6.5% Pd 9.5% Ce-6.5% PdCH4 Conversion (mol.%)Temperature (°C)   Figure 5.5. Temperature Programmed Oxidation profile. Effect of different loadings of Ce on the initial activity of 6.5Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  Table 5.3. Light-off temperatures for 6.5Pd/Al2O3 and co-xCe/yPd/Al2O3 catalysts Catalyst T10 T50 T90 °C °C °C 6.5Pd/Al2O3 142±6 251±4 285±6 co-0.9Ce/6.5Pd/Al2O3 197±6 259±4 298±6 co-2.9Ce/6.5Pd/Al2O3 193±6 253±4 290±6 co-4.8Ce/6.5Pd/Al2O3 208±6 265±4 301±6 co-9.5Ce/6.5Pd/Al2O3 208±6 275±4 325±6 9.4Ce/Al2O3 485±6 595±4 -  94  Consequently, the 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts were selected for assessment of CH4 oxidation stability in the presence of H2O. Figure 5.6 shows the TOS results for a 24h period using “dry” feed gas and “wet” feed gas with 5vol.% H2O, at temperatures 300-380°C. For both the 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts CH4 conversion for the dry-TOS decreased as the catalysts were exposed to the reactants. At 350°C CH4 conversion decreased after 24h from 100% to 84% and 46% for the 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts, respectively, indicative of a slow catalyst deactivation with TOS (Figure 5.6c and 5.6e). With H2O added to the feed gas, both 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts showed a much faster exponential deactivation in the first 5h TOS, followed by a slower, linear deactivation from TOS=5h to TOS=24h. CH4 conversion over the 6.5Pd/Al2O3 catalyst decreased from 100% to 48% in the first 5h and then decreased to 42.8% after 24h. A significant inhibition is also observed for the co-2.9Ce/6.5Pd/Al2O3 catalyst, from 100% to 23% after 5h and then to 17.4% in 24h.   95  0204060801000 5 10 15 20 250204060801000 5 10 15 20 25 0 5 10 15 20 25  (b)-330 °CRemove H2O Remove H2OCH4 Conversion (mol.%)CH4 Conversion (mol.%)Xs(a)-300 °C  Remove H2O(c)-350 °CXsRemove H2O(d)-330 °CTime on Stream (h) Remove H2O(e)-350 °CTime on Stream (h) Remove H2O(f)-380 °CTime on Stream (h) Figure 5.6. TOS results for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts at different temperatures for dry (open symbol) and wet (closed symbol) conditions. GHSV=180,000 cm3(STP).gcat-1.h-1. 5000 ppm CH4, 20(v/v)% O2, and the balance He. Top: 6.5Pd/Al2O3 at (a) T=300°C, (b) T=330°C, (c) T=350°C, Bottom: co-2.9Ce/6.5Pd/Al2O3 at (d) T=330°C, (e) T=350°C, (f) T=380°C 96  Similar to Chapter 4, the rate constant and equilibrium constant for H2O adsorption, as well as the apparent activation energy (Ea) and enthalpy of H2O adsorption (       of the co-2.9Ce/6.5Pd/Al2O3 catalyst were obtained by applying the kinetic model and the reactor design Equation 4.20 to   , the stable CH4 conversion at t=5h. Figure 5.7 shows the estimated      values with the R2=0.98 using the parameters reported in Table 5.4.  0 20 40 60 80 100020406080100Measured CH4 Conversion (mol.%)Calculated CH4 Conversion (mol.%)   Figure 5.7. Calculated  XCH4 values from the kinetic model versus measured  XCH4 values from the experiments for co-2.9Ce/6.5Pd/Al2O3 catalyst     97  Table 5.4. Constant values for co-2.9Ce/6.5Pd/Al2O3 catalyst at T=330°C used in Equation 4.20 Wcat (g) 8.33×10-2 Pd (wt.%) 6.5       (mol.s-1) 373.6        0.005 P (Pa) 101325 GHSV (mol.gcat-1.s-1) 2.23×10-3 CT (molsite.gcat-1) 1.53×10-4       at 330°C (m2.s-1) 9.88×10-7 ρcat (g.cc-1) 1.54 dp (cm) 2.22×10-2      0 and 10   Comparing the estimated kinetic parameters for 6.5Pd/Al2O3 (Table 4.6) and co-2.9Ce/6.5Pd/Al2O3 (Table 5.5) shows smaller η values for the 6.5Pd/Al2O3 catalyst than the co-2.9Ce/6.5Pd/Al2O3 catalyst. This is due to a higher reaction rate constant (ks) on the 6.5Pd/Al2O3 catalyst than the co-2.9Ce/6.5Pd/Al2O3 catalyst. The data also show that KH2O is greater on the 6.5Pd/Al2O3 catalyst than the CeO2 promoted catalyst, confirming the beneficial effect of CeO2 in regards to reducing the suppression of the CH4 oxidation reaction by H2O.  Table 5.5. Estimated values of η, rate constant, equilibrium constant for H2O adsorption, and reaction rate at different temperatures for co-2.9Ce/6.5Pd/Al2O3 catalyst T η                °C mol.gcat.molsite-2.Pa-1.s-1 Pa-1 mol.(gcat.s)-1 330 0.86 1.1 1.2710-3 1.4210-6 350 0.82 1.6 6.6110-4 3.1910-6 380 0.74 2.6 2.6810-4 5.7110-6 400 0.68 3.5 1.5410-4 4.6510-6   98  The values of   and      for the 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts are compared in Table 5.6. The higher negative value of       for co-2.9Ce/6.5Pd/Al2O3 than 6.5Pd/Al2O3 catalyst implies a stronger adsorption of the H2O on the co-2.9Ce/6.5/Pd/Al2O3 catalyst, although the amount adsorbed is lower on this catalyst as shown by the values of      in Figure 5.8, calculated from the estimated parameters. The magnitude of the pre-exponential factor,     , reported in Table 5.6 is related to the entropy change associated with the equilibrium process. Hence the differences in the pre-exponential factors are related to different entropy changes for the water adsorption modes on the Pd/Al2O3 versus the Ce/Pd/Al2O3 catalyst.  Table 5.6. Compared estimated values obtained from the design equation for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts Catalyst                    kJ.mol-1 mol.gcat.molsite-2.Pa-1.s-1 kJ.mol-1 Pa-1 6.5Pd/Al2O3 60.6±11.5 13.5±1.7 -81.5±9.1 5.310-3±4.310-4 co-2.9Ce/6.5Pd/Al2O3 56.1±8.5 1.1±0.1 -101.8±16.3 1.310-3±2.310-4  99  1.45 1.50 1.55 1.60 1.65 1.70 1.75-9.0-8.5-8.0-7.5-7.0-6.5-6.0-5.5-5.0-4.5-4.0ln KH2O calculated1000/T (K-1)(a)(b)   Figure 5.8. Calculated        versus       for (a) 6.5Pd/Al2O3 catalyst, and (b) co-2.9Ce/6.5Pd/Al2O3 catalyst   In another approach to quantify the extent of catalyst activity loss, the conversion data of Figure 5.6 are conveniently correlated to an empirical deactivation equation of the form:                                                                                                                          5.1 where      is the CH4 conversion,    represents the stable CH4 conversion at infinite time (herein taken as TOS = 5h),    is the rate of catalyst deactivation (identified as      for dry feed and     for wet feed) and (    ) represents the initial CH4 conversion. Tables 5.7 and 5.8 summarize the parameter values obtained by non-linear regression of Equation 5.1 using the TOS conversion data for the 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts, measured up to 5h TOS. The fit of the equation to each set of data was good (R2 ≥ 0.92 in all 100  cases) and is shown by the solid lines of Figure 5.6. The values reported in Tables 5.7 and 5.8 clearly show that for both catalysts,     and      decreased with increased temperature i.e. the inhibitory effects of H2O were reduced at higher temperature, as has been reported previously [77,97].  Table 5.7. Rate of deactivation for 6.5Pd/Al2O3 catalyst as a function of temperature   Dry feed   5vol.% extra H2O T, °C    A     R2     A     R2 300 60.9±0.3 24.0±0.7 0.0217±0.0024 0.99  16.9±0.3 68.5±4.2 0.0270±0.0020 0.97 330 83.3±0.5 16.3±0.5 0.0081±0.0007 0.98  33.8±0.8 66.5±1.7 0.0210±0.0040 0.92 350 93.5±0.2 5.8±0.4 0.0019±0.0005 0.99  46.9±2.3 60.9±2.8 0.0120±0.0020 0.98   Table 5.8. Rate of deactivation for co-2.9Ce/6.5Pd/Al2O3 catalyst as a function of temperature   Dry feed  5vol.% extra H2O T, °C    A     R2     A     R2 330 56.1±1.1 39.9±0.2 0.0120±0.0010 0.99  16.6±0.1 84.5±0.3 0.0220±0.0040 0.96 350 59.9±0.4 40.1±0.4 0.0078±0.0002 0.99  23.6±0.5 77.8±0.9 0.0170±0.0020 0.98 380 80.2±0.2 20.5±0.3 0.0048±0.0003 0.99  54.7±0.8 43.8±0.2 0.0120±0.0050 0.99   The impact of Ce addition on the catalyst deactivation in the presence of H2O can be quantified by considering     (the difference between    measured for the dry and wet feed) and the ratio of    measured under wet and dry conditions (      ). These values are reported in Table 5.9 and show that with Ce addition to the Pd catalyst, both     and        decreased, indicating that the inhibition of CH4 conversion by H2O is reduced by the addition of Ce to 101  the 6.5Pd/Al2O3 catalyst, both in terms of the rate of deactivation and the impact of the H2O on the final conversion.  Table 5.9.     and         ratio for 6.5Pd/Al2O3 catalyst and co-2.9Ce/6.5Pd/Al2O3 catalyst T 6.5Pd/Al2O3 catalyst  co-2.9Ce/6.5Pd/Al2O3 °C                         300 44.0 1.2  - - 330 49.5 2.6  39.5 1.8 350 46.6 6.3  36.3 2.2 380 - -  25.5 2.5  Upon removal of the H2O added to the feed gas after 24h TOS, the CH4 conversion increased to a value almost identical to that observed after 24h TOS without H2O added to the feed gas (See Figure 5.6). Hence the inhibition of CH4 conversion by H2O is partially reversible, as has been reported in other studies [29,110]. These factors suggest that the activity loss is mostly a result of H2O adsorption that is reduced at higher temperature and by the presence of CeO2 on the catalyst surface. Note that upon H2O removal for co-2.9Ce/6.5Pd/Al2O3 catalyst at 380°C the CH4 conversion increased to a higher value than conversion at 24h for the dry-TOS (Figure 5.6f). The reason for this increase is not clear but one possible explanation is that at high temperature in the presence of H2O, some CeOx/PdO restructuring occurs, changing the PdO dispersion and this does not occur in the absence of added H2O.    5.2.3 Properties of the Used Catalysts  The properties of the catalysts after the TOS experiments in the presence of 5vol.% extra H2O are reported in Table 5.10 for both the 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts. 102  A reduction in BET surface area of both catalysts was observed after reaction (from 218 to 185 m2/g for the 6.5Pd/Al2O3 and from 206 to 152 m2/g for the co-2.9Ce/6.5Pd/Al2O3 catalyst). The decrease in BET surface area was accompanied by a small decrease in pore volume and a decrease in CO uptake, yet XRD analysis showed only a small reduction in PdO crystallite size. The difference in PdO crystallite size of the fresh and used catalysts is within the experimental error associated with the analysis. Together these results suggest some sintering of the Al2O3 occurs following 24h reaction, resulting in collapse of pores, which in turn occludes some of the PdO. Figures 5.9 and 5.10 show the XRD and XPS analysis of the fresh and used 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts. The PdO peaks from XRD appeared at the same 2θ and the two main peaks for Pd 3d from XPS were observed at the same binding energy after the TOS experiment for both catalysts. This confirms the stability of PdO at the reaction conditions even in the presence of H2O and confirms that there is no evidence for the formation of Pd(OH)2 or Pd0 following reaction. Note that the  hydroxide species formed on the surface of the PdO may have decomposed during sample handling. However, if bulk Pd(OH)2 was formed and all of the PdO was converted to Pd(OH)2 then it would be expected to see Pd(OH)2 from the XPS and the XRD since it is thermally stable up to 250 C as a bulk chemical and up to 375 C when supported on carbon [111].   103     Table 5.10. Properties of fresh and used catalysts after TOS experiment for 24h in wet condition at T=350°C Catalyst BET Pore Pore CO Pd PdO SAa Sizea Volumea Uptakeb Dispersionb Crystallite Sizec m²/g nm cm3/g μmol/gcat % nm 6.5Pd/Al2O3-fresh calcined 218 7.9 0.43 204 33.5 6 6.5Pd/Al2O3-wet used 185 8.6 0.40 170 27.8 5 co-2.9Ce/6.5Pd/Al2O3-fresh calcined 206 7.9 0.41 153 25.1 5 co-2.9Ce/6.5Pd/Al2O3-wet used 152 9.6 0.37 150 24.5 4 a Determined by N2 adsorption at 77K   b Obtained by CO chemisorption      c PdO (101) obtained by XRD      104  10 20 30 40 50 60 70 80 (d)(c)(b)Intensity (a.u.)2theta (°)(a)   Figure 5.9. XRD for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 for both fresh and used catalysts for 24h TOS. (a) Fresh 6.5Pd/Al2O3, (b) Used 6.5Pd/Al2O3, (c) Fresh co-2.9Ce/6.5Pd/Al2O3, (d) Used co-2.9Ce/6.5Pd/Al2O3. ∆ PdO, ● Al2O3 105  350 348 346 344 342 340 338 336 334 332 330PdO(b)(d)(c)Intensity (a.u.)Binding Energy (eV)(a)3d3/2 3d5/2PdO Figure 5.10. XPS binding energy for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 for both fresh and used catalysts for 24h TOS. (a) Fresh 6.5Pd/Al2O3, (b) Used 6.5Pd/Al2O3, (c) Fresh co-2.9Ce/6.5Pd/Al2O3, (d) Used co-2.9Ce/6.5Pd/Al2O3 106  5.3 Discussion   In order to oxidize the catalysts prior to reaction, the 6.5Pd/Al2O3 and co-xCe/6.5Pd/Al2O3 catalysts were calcined for 15h in air at 450°C in excess O2 at reaction temperatures significantly below the PdO decomposition temperature [97,110]. Hence Pd and Ce are present as oxidized species at the operating conditions of the present study, as confirmed by XRD and XPS analysis. The XPS data show that the               surface ratio of the 9.4Ce/Al2O3 catalyst is lower than that of the co-9.5Ce/6.5Pd/Al2O3 and the      content increased as the Ce:Pd ratio of the catalyst decreased. These results are consistent with CeO2 strongly promoting the oxidation (or re-oxidation) of the Pd, resulting in some reduction of      to      [50,54]. The data also confirm that co-impregnation of the Pd and Ce species results in an interaction between the Ce and Pd species when supported on Al2O3. The CO uptake data of the Pd/Al2O3 catalyst decreased significantly from 204 mol/gcat to 89 mol/gcat with the addition of 0.9Ce to the Pd/Al2O3 catalyst (Table 5.1), suggesting a decrease in Pd dispersion. However, the XRD data show no change in the PdO crystallite size and the XPS data show only a marginal increase in the Pd surface atom ratio (from 0.8 to 1.1, see Figure 5.2), suggesting a small increase in Pd dispersion. The CO uptake measurements were made following a mild reduction of the calcined catalysts in H2 for 1h at 100C, with the intent to reduce only the outerlayer of the PdO particle [74]. However, in the case of the catalysts with Ce, it is likely that because of the high O exchange associated with the CeO2, the reduction procedure results in significantly less PdO→Pd surface reduction prior to the CO uptake measurement, resulting in a significantly lower CO uptake on the Ce promoted catalyst compared to the Pd/Al2O3 catalyst. Note, however, that for all the Ce-107  containing catalysts, the increase in CO uptake observed with increasing Ce content are in good agreement with the XPS Pds measurements reported in Figure 5.2 and indicates that the PdO dispersion is increased with increased CeO2 content of the catalysts. The addition of CeO2 to the Pd/Al2O3 catalyst also reduces the catalyst surface area and CH4 oxidation activity, similar to results reported in the literature [52].  Several studies have also reported on the effect of H2O on CH4 oxidation [23–25,33,79,88,96,110,112–116]. Kinetic studies show that the rate of CH4 oxidation over Pd catalysts on a wide range of supports is negative 1st-order in H2O partial pressure [97]. Some studies propose that the activity inhibition observed at low temperature (< 500°C) is due to the formation of Pd(OH)2 [25,79,117]. Although Pd(OH)2 decomposes at 250°C, in the presence of large amounts of H2O the formation of Pd(OH)2  can occur at T > 250°C [117]. In addition, the presence of PdO rather than Pd0 favors the formation of inactive Pd(OH)2 since its formation is more likely from PdO than Pd0. However, in the present study, Pd(OH)2 formation was not observed by either XPS or XRD analysis of the used catalysts after 24h reaction in the presence of 5vol.% H2O (Figures 5.9 and 5.10). PdO sintering may be another explanation for the observed inhibition that is exacerbated in the presence of H2O [99]. However, both Ostwald ripening and crystallite migration during a 24h TOS experiment at 350°C is unlikely because the reaction temperature is well below half the melting point of the metal oxides (melting point of PdO ~ 750C) [8]. The data of Table 5.10 confirm this, since they show only minor changes in the PdO crystallite size as measured by XRD and a small reduction in the CO uptake. The drop in CO uptake is likely due to the loss in catalyst surface area, which appears to be a consequence of Al2O3 sintering that may also result in PdO encapsulation. The relatively small decrease in CO uptake and PdO crystallite 108  size are inconsistent with the rapid and significant inhibition in CH4 conversion observed experimentally following H2O addition to the feed gas. Furthermore, the fact that the catalyst activity is recovered once the H2O is removed indicates that the activity loss by H2O addition is not due to a permanent restructuring of the catalyst.  The kinetic analysis showed that the inhibition of CH4 oxidation by H2O is well described by the reversible adsorption of H2O on active sites [23]. The loss of activity caused by H2O adsorption is higher on the Pd/Al2O3 catalyst than the co-2.9Ce/6.5Pd/Al2O3 catalyst and the kinetic analysis shows that this is because of less H2O adsorption on the latter catalyst, despite the strength of the adsorption being higher on the CeO2 promoted catalyst. The high OSC of CeO2 facilitates oxygen transfer between the Pd-* vacancies and the support that reduces the possibility of Pd-OH formation. Hence, the H2O adsorption equilibrium constant for the co-2.9Ce/6.5Pd/Al2O3 catalyst is less than that for the 6.5Pd/Al2O3 catalyst. However, the vacant sites (Pd-*) of the co-2.9Ce/6.5Pd/Al2O3 catalyst have a stronger H2O adsorption (i.e. a higher       ) than the 6.5Pd/Al2O3 catalyst.  Both catalysts showed some loss in activity following 24h reaction in dry and wet feed gas since the conversion after H2O was removed from the wet experiment feed gas was similar to that measured after 24h in the dry feed gas (see Figure 5.6). Hence we conclude that the loss of activity in the absence of added H2O could be a consequence of sintering or adsorption of H2O produced during the reaction. However, during the wet TOS experiments in the presence of 5vol.% H2O, sintering was negligible compared with H2O adsorption effects.   The adsorption of H2O may also occur on the support and one likely consequence is the interruption of the oxygen exchange between the support and Pd-vacancies on the catalyst, as 109  reported by Ciuparu et al. [36]. Hence, at higher temperature, the inhibitory effect of H2O on the CH4 oxidation is reduced, as shown by the data of Figure 5.6. Furthermore, with the addition of CeO2, with a higher oxygen exchange rate compared to the Al2O3 support [50], one would anticipate less of an impact of added H2O because of the high oxygen exchange capacity of the CeO2, consistent with the data reported in Table 5.9.   5.4 Conclusions  Addition of Ce to the 6.5Pd/Al2O3 catalyst decreased the CH4 conversion activity of the catalyst, as determined by TPO. Comparing the apparent activation energy (Ea) and enthalpy of H2O adsorption (       for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts showed similar values of Ea for the catalysts while        is higher for the co-2.9Ce/6.5Pd/Al2O3 catalyst, although      indicates less H2O adsorption on the CeO2 promoted catalyst. Hence the presence of Ce is shown to reduce the inhibition effect of H2O on CH4 oxidation. Catalyst characterization data show minimal changes in catalyst properties after reaction, and removal of H2O from the reactant feed gas results in partial recovery of the catalyst activity. The data are consistent with H2O adsorption on the catalyst/support that may also inhibit O exchange with Pd-*/PdO species, the effect of which is reversible. Addition of CeO2 would be expected to enhance the exchange rate and reduce the extent of inhibition by the adsorbed H2O.  110  Chapter 6: Effect of Preparation Method on the Activity and Stability of CeOx/PdO/Al2O3 Catalysts in the Presence of H2O  6.1 Introduction  In this chapter the effect of different preparation methods on the activity and stability of CeOx/PdO/Al2O3 catalysts is reported. The catalysts used in Chapter 5 were prepared by a co-impregnation method in which the Al2O3 support was impregnated with a premixed solution of Pd and Ce salts. In the present chapter, the Ce and Pd salts were added either by co-impregnation or by sequential impregnation. In co-impregnation, three catalysts with a fixed Pd loading of 3.4wt.% and different Ce:Pd ratios varying in the range of 0-13.8 were prepared. For sequential impregnation, the maximum Ce:Pd ratio considered was 16.7, with a fixed Pd loading of 3.4wt.%. In Chapter 5 the catalysts had 6.5wt.% Pd and the maximum Ce:Pd ratio was selected as 1.46. In this chapter, we focus on a lower Pd loading and higher Ce loadings to investigate the effect of both preparation method and Ce loading. Since the sequentially impregnated catalysts were more promising than the co-impregnated catalysts in both initial activity and stability in the presence of H2O, five sequentially impregnated catalysts were prepared, while only three catalysts were made by the co-impregnation method. The co-impregnated catalysts are identified as co-xCe/yPd/Al2O3 and the sequentially impregnated catalysts as seq-xCe/yPd/Al2O3. Five xCe/Al2O3 supports with the same Ce loading as seq-xCe/yPd/Al2O3 catalysts were also prepared in order to determine the effect of adding Pd on the physical and chemical properties of the CeOx/Al2O3. The OSC of 111  CeO2 and ZrO2 were compared by preparing seq-ZrOx/PdO/Al2O3 catalysts presented in Appendix I.  6.2 Results   6.2.1 Catalyst Properties   Figure 6.1 presents the effect of varying the Ce loading and preparation method on the BET surface area, pore size and pore volume of the prepared catalysts. Detailed values are provided in Table G.1. The decrease in BET surface area, pore size and pore volume with increased Ce loading is observed for both the co-impregnated and sequentially impregnated catalysts and for the xCe/Al2O3 supports. There is no significant difference in the BET surface area, pore size or pore volume among the catalysts prepared by either co-impregnation or sequential impregnation. Figure 6.2 presents the Pd atom % on the surface, Pds, measured by XPS, as a function of (Ce/Al)b for both co-impregnated and sequentially impregnated catalysts, all with 3.4wt.%Pd. The Pds surface concentration of both catalysts increases with increased Ce/Al in the bulk. For (Ce/Al)b ≤ 0.073, the Pds ratio is almost equal for both co-impregnated and sequentially impregnated catalysts. However, for (Ce/Al)b > 0.073, the Pds is higher for the sequentially impregnated catalyst than the co-impregnated catalysts. 112  0 10 20 30 40 50 600601201802400.10.20.30.40.50.6246810BET Surface (m2/gcat)Ce Loading (wt.%)Pore Volume (cm3/gcat)Pore Size (nm) Figure 6.1. Effect of Ce loading on BET surface area, pore size, and pore volume for co-impregnated catalysts (□), sequentially impregnated catalysts (), and xCe/Al2O3 supports (∆)  Figure 6.3 shows the Ces for both co-impregnated and sequentially impregnated catalysts and the xCe/Al2O3 supports as a function of (Ce/Al)b. Clearly Ces increases as the (Ce/Al)b increases. However, the co-impregnated catalysts have higher values than the sequentially impregnated catalysts with the same (Ce/Al)b ratio.  113  0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.350.00.51.01.52.02.53.0Pds (at.%)(Ce/Al)b   Figure 6.2. Pd atomic percent on the surface of co-impregnated (■) and sequentially impregnated (○) catalysts as a function of (Ce/Al)b  Comparing the sequentially impregnated catalysts and xCe/Al2O3 supports shows the same Ce on the surface at (Ce/Al)b ≤ 0.11, however, at (Ce/Al)b > 0.11, the sequentially impregnated catalysts have lower Ces than the xCe/Al2O3 support alone. The lower Ce surface composition of the sequentially impregnated catalysts than the xCe/Al2O3 supports at (Ce/Al)b > 0.11 ratios is due to coverage of the Ces by the Pd impregnation in the sequentially impregnated catalysts. 114  0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.3501234567Ces (at.%)(Ce/Al)b   Figure 6.3. Ce atomic percent on the surface of co-impregnated catalysts (■), sequentially impregnated catalysts (○), and xCe/Al2O3 supports (∆) as a function of (Ce/Al)b  Figure 6.4 shows the Al atom % on the surface of both co-impregnated and sequentially impregnated catalysts and the xCe/Al2O3 supports as a function of (Ce/Al)b. The values for the co-impregnated catalysts are slightly smaller than those for the sequentially impregnated catalysts. As before, the surface composition values reported for Pds, Ces, and Als are normalized, based on the XPS measurement, excluding C. 115  0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35152025303540Al s (at.%)(Ce/Al)b   Figure 6.4. Al atomic percent on the surface of co-impregnated catalysts (■), sequentially impregnated catalysts (○), and xCe/Al2O3 supports (∆) as a function of (Ce/Al)b  Pd, Ce, and Al surface compositions were also determined for a thinner layer of catalyst surface (an approximate depth of 2 nm) using the ToF-SIMS technique. Figure 6.5 shows the distribution of Pd, Ce, and Al on the surface of the seq-17Ce/3.4Pd/Al2O3 catalyst, as an example of the analysis. The scale bar in Figure 6.5 represents the change in color which is due to the relative intensity of the signals.  116   Figure 6.5. ToF-SIMS analysis for seq-17Ce/3.4Pd/Al2O3 catalyst  The TOF-SIMS data reported in Table 6.1 show a higher Pd/Al ratio for seq-17Ce/3.4Pd/Al2O3 and co-14Ce/3.4Pd/Al2O3 catalysts than 3.4Pd/Al2O3, confirming the increased Pd surface concentration that results from Ce addition, as observed from the XPS analysis (Figure 6.2). Higher Pd/Al and Ce/Al ratios for the co-14Ce/3.4Pd/Al2O3 catalyst than the seq-17Ce/3.4Pd/Al2O3 catalyst is in agreement with the higher Pds and Ces values shown in Figures 6.2 and 6.3 (at (Ce/Al)b=0.073). These results show higher surface compositions of Pd and Ce in the case of the co-impregnated catalysts. Total Ion 100 um Al Ce Pd 117  Table 6.1. Effect of adding Ce on the surface composition ratio obtained by ToF-SIMS  Surface ratio (at.%) Pd/Al Ce/Al 3.4PdAl2O3 0.14 - seq-17Ce/3.4Pd/Al2O3 0.38 1.43 co-14Ce/3.4Pd/Al2O3 0.75 2.30  Figure 6.6(a) shows the Pd XPS spectra of the 3.4Pd/Al2O3 catalyst. The Pd3d5/2 and Pd3d3/2 B.E.s are 336.9 eV and 342.2 eV consistent with those reported by Datye et al. [118] and similar values of 336.9±0.1 eV and 342.2±0.1 eV were determined for the co-2Ce/3.4Pd/Al2O3 and co-14Ce/3.4Pd/Al2O3 catalysts (Figure 6.6(b) and (c)). Adding 47wt.%Ce to the catalyst caused the Pd B.E. to increase to 337.3 eV and 342.6 eV for the 3d5/2 and 3d3/2 electrons, respectively, indicating charge transfer from the CeO2 to the PdO by adding 47wt.%Ce to the 3.4Pd/Al2O3 catalyst. The PdO peaks from the XPS analysis for the sequentially impregnated catalysts with 2wt.%, 17wt.%, and 57wt.%Ce are shown in Figure 6.7 with Pd3d5/2 and Pd3d3/2 B.E.s of 336.9±0.2 eV and 342.2±0.2 eV. Note that adding 17wt.%Ce results in a decrease in Pd B.E. (336.6 eV and 341.9 eV), suggesting a weaker oxidation of PdO for the seq-17Ce/3.4Pd/Al2O3 catalyst compared to other catalysts prepared sequentially. The Pd3d5/2 and Pd3d3/2 peaks for Pd0 are at B.E.s 335.4 eV and 340.5 eV, respectively, 1.2 eV and 1.4 eV lower than the PdO B.E.s and not present in the samples shown in Figures 6.6 and 6.7. Note that the 0.4 eV shift in B.E. observed by adding CeO2 to the catalyst is within the B.E. measurement error expected for porous catalysts, so these changes in oxidation state are not definitive since a minimum 1.0 eV shift is probably needed to confirm PdO→Pd0 transformation on these porous catalysts.   118  PdOIntensity (a.u.)Binding Energy (eV)(d)(c)3d5/23d3/2(a)PdO(b)    Figure 6.6. XPS Pd 3d spectra measured for (a) 3.4Pd/Al2O3 and co-impregnated (b) co-2Ce/3.4Pd/ Al2O3, (c) co-14Ce3.4Pd/Al2O3, and (d) co-47Ce/3.4Pd/Al2O3 catalysts                          Binding Energy (eV) 350      348      346      344      342      340      338      336      334     332      330     119  PdOIntensity (a.u.)(b)Binding Energy (eV)Sequential(a)PdO3d5/23d3/2(c)(d) Raw Intensity Peak Sum Background Peak 1 Peak 2 Figure 6.7. XPS Pd 3d spectra measured for (a) 3.4Pd/Al2O3 and sequentially impregnated (b) seq-2Ce/3.4Pd/Al2O3, (c) seq-17Ce/3.4Pd/Al2O3, and (d) seq-57Ce/3.4Pd/Al2O3 catalysts Binding Energy (eV) 350      348      346      344       342       340      338      336       334      332       330     120  The Ce 3d spectra for the sequentially impregnated catalysts are presented in Figure 6.8. Because of the low intensity of Ce 3d peaks of the seq-2Ce/3.4Pd/Al2O3 catalyst, only the Ce 3d spectra of seq-17Ce/3.4Pd/Al2O3, seq-28Ce/3.4Pd/Al2O3, and seq-57Ce/3.4Pd/Al2O3 catalysts are shown in Figure 6.8 and the B.E.s of Ce 3d for all seq-2Ce/3.4Pd/Al2O3, seq-17Ce/3.4Pd/Al2O3, seq-28Ce/3.4Pd/Al2O3, and seq-57Ce/3.4Pd/Al2O3 catalysts are reported in Table 6.2. The two main peaks attributed to      (Ce2O3) are located at 885.0 eV (v') and 903.6 eV (u'), respectively for all catalysts. The three main peaks assigned to 3d3/2 for Ce4+ (CeO2) appeared at 901.0±0.2 eV (u), 906.9 eV (u"), and 917.0 eV (u‴) for all catalysts. There are also three peaks for     (CeO2) 3d5/2 identified as v, v", and v"'. For all sequentially impregnated catalysts, the v and v"' peaks appear at 882.9±0.3 eV and 898.7±0.4 eV, respectively. However, the peak attributed to      (CeO2) 3d5/2 appeared at 887.4 eV (v") for the seq-2Ce/3.4Pd/Al2O3 catalyst but it shifts to 888.6 eV once the Ce loading reaches 17wt.% and remains unchanged for higher Ce loadings (28wt.% and 57wt.%). The Ce 3d spectra for co-impregnated catalysts and xCe/Al2O3 supports are presented in Figures G.1 and G.2 and Tables G.3 and G.4. The B.E.s are similar as the sequentially impregnated catalysts and the peak attributed to      (CeO2) 3d5/2 shifted from 887.4 eV (v") to 888.4 eV and 888.7 eV, respectively, for co-impregnated catalysts and xCe/Al2O3 supports. 121  930 920 910 900 890 880 870Intensity (a.u.)(c)(b)Binding Energy (eV)(a)u'" u" u v"' v" v' v(a) 5Pd over 15Ce+Al2O3(b) 5Pd over 25Ce+Al2O3(c) 5Pd over 50Ce+Al2O3u'   Figure 6.8. Ce 3d for sequentially impregnated (a) seq-17Ce/3.4Pd/Al2O3, (b) seq-28Ce/3.4Pd/Al2O3, (c) seq-57Ce/3.4Pd/Al2O3 catalysts  Binding Energy (eV) 930              920                                90                 8                 8                870 u"'                u"   u'   u   v"'               v"    v'   v (c)      (b)    (a)  Intensity (a.u.) 122      Table 6.2. Ce 3d peaks and               ratio for sequentially impregnated catalysts with different loadings of Ce Catalyst Ce 3d5/2  Ce 3d3/2                                                          eV eV eV eV  eV eV eV eV  % seq-2Ce/3.4Pd/Al2O3 882.9 885.0 887.4 898.7  901.0 903.6 906.9 917.0  31.9 seq-6Ce/3.4Pd/Al2O3 882.9 885.0 887.6 898.7  901.0 903.6 906.9 917.0  29.3 seq-17Ce/3.4Pd/Al2O3 882.9 885.0 888.6 898.7  901.0 903.6 906.9 917.0  13.8 seq-28Ce/3.4Pd/Al2O3 882.6 885.0 888.5 898.3  900.7 903.6 906.9 917.0  10.4 seq-57Ce/3.4Pd/Al2O3 882.7 885.0 888.6 898.3  900.8 903.6 906.9 917.0  10.9 123  Figure 6.9 presents the               ratios for all the catalysts. For the co-impregnated catalysts, sequentially impregnated catalysts, and the xCe/Al2O3 supports, the ratio decreases with increased Ce loading. For instance, the               ratio for the sequentially impregnated catalysts decrease from 31.9% to 10.9% as the loading of Ce increases from 2wt.% to 57wt.%. Similar to Chapter 5, the decrease is attributed to less transition of      to      with higher Ce loading.  0 10 20 30 40 50 600102030409095100Ce3+/(Ce3++Ce4+) (%)Ce Loading (wt.%)   Figure 6.9.               ratio obtained by XPS analysis for co-impregnated catalysts (∆), sequentially impregnated catalysts (□), and xCe/Al2O3 supports () as a function of varying loadings of Ce   124  10 20 30 40 50 60 70 80   Intensity (a.u.)2theta (°)(a)(b)(c)(d)   Figure 6.10. XRD patterns for co-impregnated catalysts (a) 3.4Pd/Al2O3, (b) co-2Ce/3.4Pd/Al2O3, (c) co-14Ce/3.4Pd/Al2O3, and (d) co-47Ce/3.4Pd/Al2O3 catalysts. ∆ PdO, ● Al2O3, ○ CeO2  Figure 6.10(a) shows the XRD analysis of the 3.4Pd/Al2O3 catalyst with peaks for PdO observed at 2θ=39.50° and 64.50° corresponding to PdO (101) and PdO (112), respectively. The PdO (101) peak was observed for all co-impregnated catalysts with different loadings of Ce, however, the peak shifted slightly to lower 2θ as the Ce loading increased from 2wt.% to 47wt.%. This is a result of an overlap between the PdO (101) peak located at 39.50° and the CeO2 (200) peak at 38.61°. The main peaks for CeO2 appear at 2θ=33.27°, 38.61°, 55.75° and 66.49° corresponding to CeO2 (111), (200), (220), and (311), respectively. At higher loadings of Ce (14wt.% and 47wt.%), the peak at 2θ=55.75° has an overlap between Al2O3 125  and CeO2 (220) and the peak at 2θ=64.50° is a result of overlap between the PdO (112) and CeO2 (311) peaks.                                       10 20 30 40 50 60 70 80Intensity (a.u.)2Theta (°)(a)(b)(c)(d)     Figure 6.11. XRD patterns for sequentially impregnated (a) 3.4Pd/Al2O3, (b) seq-2Ce/3.4Pd/Al2O3, (c) seq-17Ce/3.4Pd/Al2O3, and (d) seq-57Ce/3.4Pd/Al2O3 catalysts. ∆ PdO, ● Al2O3, ○ CeO2  Figure 6.11 shows the XRD analysis for sequentially impregnated catalysts. The position of the peaks for PdO and CeO2 are the same as those for the co-impregnated catalysts. The PdO (101) and CeO2 (111) crystallite sizes are reported in Table 6.3. For 3.4Pd/Al2O3 catalyst the PdO (101) crystallite size was calculated as 7 nm. This number is slightly smaller for co-impregnated and sequentially impregnated catalysts. The CeO2 (111) crystallite size increases with increased Ce loading for both co-impregnated and sequentially impregnated catalysts and xCe/Al2O3 supports.  126  Table 6.3. CeO2 and PdO crystallite size of calcined co-impregnated and sequentially impregnated catalysts and xCe/Al2O3 supports Catalyst XRD Crystallite Size CeO2 PdO (111) (101) nm nm 3.4Pd/Al2O3 - 7 co-2Ce/3.4Pd/Al2O3 - 5 co-14Ce/3.4Pd/Al2O3 7 - co-47Ce/3.4Pd/Al2O3 8 - seq-2Ce/3.4Pd/Al2O3 - 5 seq-6Ce/3.4Pd/Al2O3 7 6 seq-17Ce/3.4Pd/Al2O3 7 - seq-28Ce/3.4Pd/Al2O3 9 - seq-57Ce/3.4Pd/Al2O3 9 - 2Ce/Al2O3 - - 5Ce/Al2O3 5 - 16Ce/Al2O3 8 - 26Ce/Al2O3 9 - 52Ce/Al2O3 9 -  The characterization data show very similar physical and chemical properties of the sequentially impregnated and co-impregnated catalysts. However, with increased (Ce/Al)b ratio, the surface area and the               ratio decreased for both preparation methods. The catalyst surface composition analysis showed that the surface Pd concentration (Pds) for (Ce/Al)b > 0.073 was higher for the sequentially impregnated catalyst than co-impregnated catalysts, while the Ces values were smaller. The lower Ces values suggest stronger interaction between Ce and Al2O3 in the sequentially impregnated catalysts that may improve the oxygen exchange capacity of the sequentially impregnated catalysts during the CH4 oxidation reaction.   127  6.2.2 Catalyst Activities  Figures 6.12 and 6.13 compare the TPO results for the 3.4Pd/Al2O3 and the co-impregnated and sequentially impregnated catalysts. The 3.4Pd/Al2O3 catalyst had a T50 of 273°C.  100 150 200 250 300 350020406080100 3.4Pd/Al2O3 co-2Ce/3.4Pd/Al2O3 co-14Ce/3.4Pd/Al2O3 co-47Ce/3.4Pd/Al2O3CH4 Conversion (mol%)Temperature (C)    Figure 6.12. Temperature Programmed Oxidation profile for co-impregnated catalysts. Effect of different loadings of Ce on the initial activity of 3.4Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar 128  100 150 200 250 300 350020406080100 3.4Pd/Al2O3 seq-2Ce/3.4Pd/Al2O3 seq-6Ce/3.4Pd/Al2O3 seq-17Ce/3.4Pd/Al2O3 seq-28Ce/3.4Pd/Al2O3 seq-57Ce/3.4Pd/Al2O3CH4 Conversion (mol%)Temperature (C)   Figure 6.13. Temperature Programmed Oxidation profile for sequentially impregnated catalysts. Effect of different loadings of Ce on the initial activity of 3.4Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  Increased Ce content resulted in reduced activity and the co-47Ce/3.4Pd/Al2O3 and seq-57Ce/3.4Pd/Al2O3 catalysts were the least active catalysts with T50 of 351°C and 288°C, respectively. However, the seq-17Ce/3.4Pd/Al2O3 catalyst showed the highest activity among 3.4Pd/Al2O3 and the other co-impregnated and sequentially impregnated catalysts. The initial activity of the catalysts decreases in the following order as the basis of T50: seq-17Ce/3.4Pd/Al2O3 > seq-2Ce/3.4Pd/Al2O3 > 3.4Pd/Al2O3 > seq-6Ce/3.4Pd/Al2O3 > co-2Ce/3.4Pd/Al2O3  129  Comparing the initial activity of the catalysts shown in Figures 6.12 and 6.13 confirms that the co-impregnated catalysts are less active and also more sensitive to Ce loading compared with the sequentially impregnated catalysts. For the co-47Ce/3.4Pd/Al2O3 catalyst with the highest Ce loading, the T50 was 78°C higher than the 3.4Pd/Al2O3 catalyst, however, the T50 for the seq-57Ce/3.4Pd/Al2O3 was only 15°C higher than the 3.4Pd/Al2O3 catalyst. More details of the light-off temperatures corresponding to 10%, 50%, and 90% CH4 conversion are presented in Table G.2 and these data show similar trends in terms of catalyst activity.   The assessment of the effect of different preparation methods on the stability of the catalysts during CH4 oxidation was done under dry-TOS and wet-TOS experimental conditions for a 24h period at temperatures 310-370°C. At 350°C under dry-TOS conditions, CH4 conversion decreased as the catalysts were exposed to the reactants. A slower loss in CH4 conversion is observed for the co-2Ce/3.4Pd/Al2O3 catalyst compared with the other co-impregnated catalysts with higher Ce loadings (Figure G.9). The sequentially impregnated catalysts also show a loss in the CH4 conversion as a function of TOS (Figure G.10), however, this loss is reduced compared to the co-impregnated catalysts with the same Ce loadings. The rate of catalyst deactivation for the dry-TOS experiment was assessed using the empirical deactivation model discussed previously (Equation 5.1), with the model parameters summarized in Figure 6.14.      values are reported at 350°C and 320°C. At a constant temperature (350°C),      is significantly higher for the co-impregnated catalysts than the sequentially impregnated catalysts. In the case of the 3.4Pd/Al2O3 and the seq-17Ce/3.4Pd/Al2O3 catalysts, the      values are compared at lower temperature (320°C), showing less deactivation in the case of the seq-17Ce/3.4Pd/Al2O3 catalyst.  130  1.60 1.62 1.64 1.66 1.68 1.700.0000.0020.0040.0060.0080.0100.0120.0143.4Pdseq-17Ce/3.4Pdco-47Ce/3.4Pdco-14Ce/3.4Pdseq-2Ce/3.4Pdseq-57Ce/3.4Pdkd,d (min-1)1000/T (K-1)   Figure 6.14. Rate of catalyst deactivation (    ) as a function of Ce loading for co-impregnated catalysts (Δ), and sequentially impregnated catalysts (●) at T=350°C and (■) at T=320°C. Obtained from dry-TOS results. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  The same comparison for the wet-TOS results at 350°C is shown in Figure 6.15 for the co-impregnated and sequentially impregnated catalysts in the presence of 2vol.% H2O. The rate of catalyst deactivation is much lower for the sequentially impregnated catalysts than the co-impregnated catalysts with identical Ce loading. Higher Ce loading results in a higher loss of catalyst activity.  131  0 10 20 30 40 50 600.000.010.020.030.040.050.060.070.080.090.10kd,w (min-1)Ce Loading (wt.%)   Figure 6.15. Rate of catalyst deactivation (   ) as a function of Ce loading for co-impregnated catalysts (○), and sequentially impregnated catalysts (■). Obtained from wet-TOS results at T=350°C and 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  The seq-17Ce/3.4Pd/Al2O3 catalyst was selected as the most active and stable catalyst among the co-impregnated and sequentially impregnated catalysts and was compared to 3.4Pd/Al2O3 catalyst in terms of stability at different temperatures and in the presence of different H2O concentrations under wet-TOS reaction conditions. Figure 6.16 presents the wet-TOS results for both the 3.4Pd/Al2O3 and the seq-17Ce/3.4Pd/Al2O3 catalysts, with 2vol.% H2O and varying temperature between 310°C-370°C. The TOS results show higher CH4 conversion at 132  higher temperature for both catalysts, indicating less H2O adsorption at higher temperatures as discussed in Chapters 4 and 5.  133   0 5 10 15 20 250 5 10 15 20 25020406080100 Time on Stream (h)3.4Pd/Al2O3(c)(b) (a)(d)seq-17Ce/3.4Pd/Al2O3(c)(b)(a)(d)CH4 Conversion (mol.%)Time on Stream (h)     Figure 6.16. Wet-TOS results for seq-17Ce/3.4Pd/Al2O3 and 3.4Pd/Al2O3 catalysts with 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar. (a) T=310°C, (b) T=330°C, (c) T=350°C, and (d) T=370°C 134   0 5 10 15 20 250 5 10 15 20 25020406080100Time on Stream (h)(c)(b)(a)  3.4Pd/Al2O3seq-17Ce/3.4Pd/Al2O3CH4 Conversion (mol.%)Time on Stream (h)(c)(b)(a)  Figure 6.17. Wet-TOS results for seq-17Ce/3.4Pd/Al2O3 and 3.4Pd/Al2O3 catalysts at T=350°C and (a) 1vol.% H2O, (b) 2vol.% H2O, and (c) 5vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar 135  Comparing the TOS results from the two catalysts, as reported in Figure 6.16, shows higher CH4 conversion in the case of seq-17Ce/3.4Pd/Al2O3 catalyst at all temperatures. Upon removal of the 2vol.% extra H2O to the feed gas after 24h, the CH4 conversion increased to a higher value, however, the increase in CH4 conversion was higher at higher reaction temperatures. Wet-TOS results for both the 3.4Pd/Al2O3 and the seq-17Ce/3.4Pd/Al2O3 catalysts with different H2O concentration in the feed gas and constant temperature at 350°C are shown in Figure 6.17. Comparing the TOS results between the two catalysts at the same H2O concentration indicates higher CH4 conversion for the seq-17Ce/3.4Pd/Al2O3 catalyst.  6.3 Discussion  The decrease in BET surface area and pore volume with addition of Ce to the 3.4Pd/Al2O3 catalyst is a result of the partial filling and blocking of the Al2O3 pores by CeO2. The XRD data showed that the CeO2 crystallites are ≤ 9 nm in size, increasing from about 5 nm as the Ce loading increased. The CeO2 and PdO crystallites are smaller than the catalyst pore size (Table G.1), except at the highest Ce loading. Therefore, in the latter case some of the CeO2 may not interact with the Pd located within the pores of the support. The decrease in BET surface area with increased Ce loading indicating coverage of Al2O3 pores by CeO2 is also reported in other studies [51,53,59]. Colussi et al. [59] showed the BET surface area of 10%Pd/Al2O3 and 10%Pd/15%CeO2/Al2O3 catalysts was 124 m2/g and 110 m2/g, respectively, significantly lower than the Al2O3 surface area reported as 148 m2/g [59].   To prepare the catalysts with the highest Ce:Pd ratio, the premixed Pd-Ce solution and the Ce solution for the co-47Ce/3.4Pd/Al2O3 and seq-57Ce/3.4Pd/Al2O3 catalysts, respectively, were added to the Al2O3 support in three steps with a calcination at 450°C for 15h between each 136  step. Hence, the low BET surface area for both co-47Ce/3.4Pd/Al2O3 and seq-57Ce/3.4Pd/Al2O3 catalysts could be a result of partial pore filling because of both the high amount of Ce and also the multiple calcination steps. The effect of different loadings of CeO2 on Al2O3 reported in [119] shows that after calcination under air for 6h at 1273K, the BET surface area decreased from 147 m2/g for the pure Al2O3 to 142 m2/g and 123 m2/g for 5%CeO2/Al2O3 and 15%CeO2/Al2O3 samples, respectively, indicative of the high specific weight and low porosity of ceria.   Figures G.3-G.5 show the N2 adsorption-desorption isotherms for the co-impregnated catalysts, sequentially impregnated catalysts, and the xCe/Al2O3 support of the present study. The hysteresis observed for all samples indicates the presence of mesopores [120], however, the hysteresis decreases with increasing Ce loading indicating a loss of mesopores at high Ce loadings. Decreasing the pore size as a function of increasing the Ce amount could also be explained by the pore filling. Decreasing the pore volume is the consequence of doping metals on the internal surface of the Al2O3 support [121].   The Pd 3d B.E.s show that palladium is present in the form of PdO only and with no other phases, e.g. Pd0, present. The comparison of the Pd 3d spectra of 0.6%Pd/CeO2 and 6.8%Pd/Al2O3 catalysts by Shyu et al. [122] confirmed the role of ceria in increasing the oxidation state of Pd. After calcination at 800°C a peak at 337.0±0.1 eV corresponding to PdO was observed for both 0.6%Pd/CeO2 and 6.8%Pd/Al2O3 catalysts. In the case of reduction at 500°C, a peak at 335.0±0.1 eV appeared for both catalysts indicating the formation of Pd0. By reducing the catalysts at higher temperature (920°C) and then exposing to ambient air, a peak at 337.0 eV corresponding the formation of PdO in the case of 137  0.6%Pd/CeO2 catalyst was observed. However, in the case of 6.8%Pd/Al2O3 catalyst under the same treatment condition the palladium remained as Pd0 [122]. In another study by Xiao et al. [60] it was also claimed that the XPS analysis of the reduced 2wt.%Pd/CeO2 did not show the presence of any Pd0. The Pd3d5/2 B.E. at 336.68±0.3 eV for the 2wt.%Pd/CeO2 catalyst, indicates Pd present as PdO not Pd0 due to the strong Pd-Ce interaction that causes low reducibility of the catalysts under reducing conditions [60].  The ability of ceria to release and store oxygen and also to enhance the thermal stability of Al2O3, is well known [44,45]. However, the oxygen storage capacity of CeO2 is affected by its loading, the presence of precious metals, and also the pretreatment temperature [44]. Yao et al. [44] showed the presence of Pd can increase the oxygen storage capacity of CeO2. A comparison between oxygen chemisorption of CeO2/Al2O3 with different loadings of CeO2 shows less oxygen uptake at higher CeO2 loadings. This behavior was explained by lower CeO2 dispersion at higher CeO2 loadings. The O2 chemisorption was measured for the reduced samples at 500°C for 2h. At low CeO2 loadings (< 2.5μmol CeO2/m2[BET]) the oxygen uptake increases as CeO2 loading increases. However, the O2 uptake per unit weight of CeO2 decreases by increasing the CeO2 loading that indicates lower Ce dispersion at higher CeO2 loadings. Table 6.4 shows the decrease in O2 chemisorption as the CeO2 loading increases from 0.48wt.% to 35.35wt.% [44].  In the present study, Ces increased with increased Ce loading. However, increasing Ces does not necessarily confirm higher Ce dispersion. As shown in Chapter 5, the increase in Ces increases linearly, indicative of a monodispersed Ces. For Ce loading higher than 4.8wt.%, agglomeration of CeOx results in a small increase in Ces (Figure 5.2). However, the opposite 138  behavior is observed for the co-impregnated and sequentially impregnated catalysts and xCe/Al2O3 supports shown in Figure 6.3, where the rate of the increase in Ces increases at (Ce/Al)b > 0.073. The faster increase can be explained by comparing the CeO2 crystal size in Table 6.3 and the catalyst pore size reported in Table G.1. The pore size of the catalysts prepared by co-impregnation and sequential impregnation methods as well as the xCe/Al2O3 supports are smaller than the CeO2 crystal size for (Ce/Al)b > 0.073, resulting in more CeO2 on the outside of the pores of support than inside. This results in an increase in Ces as measured by XPS. The higher Ces at high (Ce/Al)b is in agreement with Yao et al. [44] who reported larger CeO2 crystals at higher CeO2 loadings.   Table 6.4. O2 chemisorption on CeO2/Al2O3 samples with different loadings of CeO2 (Reproduced with permission from [44]) CeO2 Concentration O2 Chemisorptiona  % (μmol O2/μmol CeO2) 0.48 0.27 0.83 0.18 2.04 0.09 3.72 0.06 6.14 0.05 11.69 0.05 21.63 0.06 35.38 0.06 a Reduced CeO2/Al2O3 samples at 500°C for 2h  As shown in Figure 6.9, in the present study the               ratio decreased at higher Ce loading, which together with the XRD data of Table 6.3, showing increased CeO2 crystallite size with increased Ce loading, is consistent with the trend reported by Hailstone et al. [46] that the fraction of Ce3+ decreases as the Ce dispersion decreases or particle size increases. The significant decrease in               surface ratio is also in agreement with the observation 139  by Monteiro et al. [123] that showed u'" peak corresponding to Ce4+ is larger in Pd/20CeO2/Al2O3 catalyst than Pd/3CeO2/Al2O3 catalyst. The CeO2 (111) crystallite size increases from 7 nm for seq-6Ce/3.4Pd/Al2O3 catalyst to 9 nm for seq-57Ce/3.4Pd/Al2O3 catalyst. As shown by Hailstone [46], increasing the crystallite size from 1.1 nm to 11.8 nm facilitates the OSC as a result of lower               surface ratio. In another study the measured crystallite size of CeO2 by XRD analysis for the CeO2/Al2O3 samples with ceria loadings in the 9-27wt.% range shows a constant crystallite size in the range of 55-67Å [44].  By comparing the catalyst activity of the co-impregnated and sequentially impregnated catalysts, it was shown that the sequentially impregnated catalysts are more active than the co-impregnated catalysts with equivalent loadings of Pd and Ce (Figures 6.12 and 6.13). The characterization results show the BET surface area, Pds, Ces, and               ratios vary as a function of Ce loading. However, they are not significantly different between the co-impregnated and sequentially impregnated catalysts. The decrease in BET surface area and consequently the blockage of pore volume of the catalysts has a negative effect in terms of the activity of the Pd based catalysts. On the other hand, the increase in Pd dispersion that is a consequence of the added Ce may or may not improve the activity since the higher Pd-support interaction at higher Pd dispersion suppresses the activity.  However, the effect of CeO2, having higher oxygen storage capacity than Al2O3, can provide the oxygen transfer to the palladium active sites during the CH4 oxidation reaction and improve the catalyst activity. In addition, the increase in Ce4+ obtained by increasing the Ce loading enhances the OSC and should result in more active catalyst. Comparing the catalyst activity of the PdO/Al2O3 and co-impregnated CeOx/PdO/Al2O3 catalysts shows lower 140  activity of the co-impregnated CeOx/PdO/Al2O3 catalysts than the 3.4Pd/Al2O3 catalyst (Figures 6.12). However, in the case of the sequentially impregnated catalysts, seq-17Ce/3.4Pd/Al2O3 is more active than 3.4Pd/Al2O3 catalyst. The difference in the catalyst activity shows the impact of the preparation method, emphasizing that high OSC of Ce is significant for the CH4 oxidation reaction. However, in terms of chemical and physical properties of the catalysts using different preparation methods, a minor difference was observed between the sequentially impregnated and co-impregnated catalysts.   As shown by Fujimoto et al. [74] small PdOx crystals or those PdOx in close contact with the support, have stronger Pd-O bond than larger crystals, which leads to lower oxygen vacancies of the catalyst surface. Ciuparu et al. [124] emphasized the oxygen exchange between the support and PdO for the CH4 combustion at low temperatures. They showed that in the case of the supports with a high oxygen mobility (e.g. CeO2), the oxygen vacancies are partially refilled with the oxygen from the support, however, the surface oxygen vacancies in the case of Al2O3 support are mainly replenished with oxygen from the gas phase. Therefore, in the case of co-impregnated CeO2/PdO/Al2O3 catalysts, PdO is still in a contact with the Al2O3 support and the density of oxygen vacancies is expected to be lower than those prepared sequentially where PdO is in close contact with CeO2 and much less so with the Al2O3.  As shown in other studies the surface oxygen exchange is affected by the hydroxyl desorption [32–34]. The oxygen vacancies formed from H2O desorption are refilled by the oxygen from both PdO and the support:  141  2Pd-OH→H2O+Pd-O+Pd-*                                                                                                   6.1 Pd-O+S-* Pd-*+S-O                                                                                                            6.2 Pd-*+S-Os Pd-Os+S-*                                                                                                          6.3  The recombination of the hydroxyl group is slow and their tendency to migrate on the catalyst surface is higher than desorption. Thus, slow H2O desorption suppresses the oxygen exchange between the support and surface. Figure 6.18 shows the oxygen exchange processes between the gas phase, PdO, and the oxide support proposed by Ciuparu et al. [124]. The oxygen exchange for a reduced Pd catalyst over the oxide support consists of five different steps. The oxygen uptake by Pd0 from the gas phase (1), the oxygen exchange from the new formed PdOx and the vacancy on the catalyst surface (2), the oxygen exchange between the new formed PdOx and the gas phase (3), oxygen exchange between gas phase and the oxygen vacancy on the catalyst surface (4), and finally the equilibrium oxygen exchange between the surface vacancy and the oxygen in the bulk of the catalyst.   Figure 6.18. Oxygen exchange mechanism of the PdOx (PdO phase formed during the temperature programmed isotopic exchange) [124] (Copyright © 2002 American Chemical Society)  142  The isotopic exchange study of oxygen at low temperature (100-500°C) showed an increase in 16O18O concentration for both oxidized and reduced Pd catalysts over either Al2O3 or ZrO2 support, with increasing temperature. On the other hand, no significant difference in 16O18O concentration between the oxidized Pd over Al2O3 or ZrO2 was observed. However, the reduced Pd/ZrO2 catalyst had higher 16O18O concentration than the reduced Pd/Al2O3, indicating the positive effect of ZrO2 support with high oxygen exchange capacity for the reduced Pd catalyst. This difference was a result of formation of double isotopic exchange oxygen, 16O2, that confirms the higher oxygen exchange activity of ZrO2 support for the reduced Pd catalysts [124]. It was also suggested that in the case of reduced Pd supported catalysts, more oxygen from the ZrO2 support is involved in order to reoxidize the Pd0 and form PdO than that used for reduced Pd catalyst over Al2O3. The temperature programmed isotopic exchange analysis also confirmed the oxidation of Pd0 is mostly dependent on steps 1, 2, 3, and 5. Since in our study it is known that palladium is in the oxide phase, the oxygen uptake from the gas phase in order to form PdO is unlikely. Therefore, the most important mechanisms for the oxygen exchange are limited to the three steps shown in Figure 6.19, indicating the importance of oxygen exchange from the bulk oxide support to the PdO active sites. This mechanism can also emphasize the effect of high oxygen capacity of the support to facilitate the oxygen exchange (steps 1 and 3) and suppress the negative effect of H2O.  143   Figure 6.19. Possible oxygen exchange mechanism based on the activity results for 3.4Pd/Al2O3 and seq-17Ce/3.4Pd/Al2O3 catalysts. Oxygen exchange between the PdO and oxygen vacancy (1), between the PdO and gas phase (2), and between the oxygen vacancy and bulk oxide support (3) (Adopted with permission from [124])  In the present study, the higher initial activity of the seq-17Ce/3.4Pd/Al2O3 catalyst than the 3.4Pd/Al2O3 catalyst can be ascribed to the faster oxygen exchange (steps 1 and 3 of Figure 6.19) in the presence of CeO2. In addition, the lower H2O inhibition observed for the seq-17Ce/3.4Pd/Al2O3 catalyst than the 3.4Pd/Al2O3 catalyst at temperatures in the range of 310-370°C and different H2O concentration of 0-5vol.% can be explained by a faster H2O desorption as a result of higher oxygen exchange from CeO2/Al2O3 support to Pd-* sites in the case of seq-17Ce/3.4Pd/Al2O3 catalyst.  6.4 Conclusion  The effect of different preparation methods along with different Ce:Pd ratios on the catalytic properties and catalytic activity of a series of Ce/Pd/Al2O3 catalysts was examined. Catalysts 144  with the Ce:Pd ratio varying in the range of 0-17 using co-impregnation and sequential impregnation methods were prepared. The BET surface area decreased with increased Ce loading, indicating the partial filling and/or blocking of the Al2O3 pores by CeO2 crystals. The Pds surface composition increased with increased Ce loading, indicating higher Pd dispersion at higher Ce loading. The Pd3d B.E. did not change significantly with Ce loading, emphasizing the stable PdO phase and no Pd0 formation in the presence of CeO2. On the other hand,               ratio decreased at higher Ce loading that suggests higher OSC at higher Ce loading. The physical and chemical properties of the catalysts with the same Ce loading prepared by co-impregnation or sequential impregnation methods are similar, confirming that the properties of the CeO2/PdO/Al2O3 catalysts are mostly affected by the presence of Ce not by the preparation methods. The characterization techniques used to observe the properties of the catalysts were not sufficient to differentiate the interaction between the Pd, Ce, and Al2O3 that likely resulted from the different preparation methods. However, the preparation method had an impact on the catalytic activity during CH4 oxidation. Comparing PdO/Al2O3, co-CeOx/PdO/Al2O3, and seq-CeOx/PdO/Al2O3 catalysts for CH4 oxidation showed that the sequentially impregnated catalyst with Ce:Pd ratio of 5 was the most active and stable catalyst among those examined here. The role of CeO2 in increasing the oxygen exchange capacity during the CH4 oxidation reaction is dependent on the catalyst preparation and Ce:Pd ratio. The proposed oxygen exchange mechanism explains the importance of oxygen exchange from the bulk oxide support to the PdO active sites. This mechanism emphasizes the presence of CeO2 with high OSC that facilitates the oxygen transfer from the bulk oxide to the PdO active sites.  145  Chapter 7: Kinetics of the Inhibition by H2O   7.1 Introduction  In this chapter the inhibition of the CH4 oxidation kinetics by H2O on Pd based catalysts is analyzed under non-steady state reaction conditions. As explained in Chapter 4, the packed bed reactor is modeled assuming one dimensional plug flow. The external mass transfer is neglected and the internal mass transfer effect is considered in the reactor model. For the kinetic model, it is assumed that the CH4-O2 reaction follows a Langmuir-Hinshelwood mechanism as presented in the literature [82,85,86]. The dissociation of CH4 takes place on active sites and C-H bond activation on the active sites is the rate determining step. The oxygen transfer from the oxide support to the Pd-* can provide the active sites for C-H bond activation [82]. The mechanistic steps can be written as follows [82]: Step 1.1. O2 (g)+* O2*  Step 1.2. O2*+* 2O* Step 2.1. CH4+*+*→CH3*+H* Step 2.2. CH4+O*+*→CH3*+OH* Step 2.3. CH4+O*+O*→CH3O*+OH* Step 3. C*+O* CO*+* Step 4. CO*+O* CO2*+* Step 5. 2OH* H2O*+O* 146  Step 6. H2O* H2O+* Step 7. CO2* CO2+* Step 8. CO* CO+*  7.2 Kinetic Model of H2O Inhibition in a Non-steady State System  The fixed-bed micro-reactor operating under non-steady state conditions, was modeled for the reaction conditions with an inlet feed stream of 5000 ppmv CH4 with 0vol.% and 5vol.%H2O entering the catalyst bed at a fixed temperature as shown in Equation 4.19.  By replacing      with                   . Equation 4.19 is converted to:                                                                                                             7.1 where   η          ,                     , and                 . η,   , and      values for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts were obtained in Chapters 4 and 5 at different temperatures and applied to Equation 7.1 to calculate   and  .   As defined in Chapter 4,                        results in             . However,            with H2O added to the feed gas, so   and   could be assumed to be independent of     .  Using these assumptions, integration of Equation 7.1 results in:                                                                                                          7.2 147  Defining                and           , Equation 7.2 is written as:                                                                                                                  7.3                                                                                                                              7.4 A linear regression of Equation 7.4 was applied to the experimental data of CH4 conversion as a function of TOS from t=0h to t=5h for the 0.3Pd/Al2O3, 2.6Pd/Al2O3, 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts. The linear regression at different reaction temperatures (300°-380°C) and 5vol.%H2O are shown in Figures 7.1 and 7.2. The obtained   values from the linear regression were used to determine the H2O desorption rate constant values (  ) at different temperatures. The calculated values are reported in Table 7.1.  0 2000 4000 6000 8000 10000 12000 14000 16000LHS of Equation 7.4Time on Stream (s)   Figure 7.1. Fitting Equation 7.4 to the experimental results for the wet-TOS with 5vol.%H2O at T=300°C (■), 330°C (∆), and 350°C (●) for 6.5Pd/Al2O3 catalyst 148  0 2000 4000 6000 8000 10000 12000 14000 16000 18000LHS of Equation 7.4Time on Stream (s)co-2.9Ce/6.5Pd   Figure 7.2. Fitting Equation 7.4 to the experimental results for the wet-TOS with 5vol.%H2O at T=330°C (■), 350°C (∆), and 380°C (●) for co-2.9Ce/6.5Pd/Al2O3 catalyst  Table 7.1. Rate constant for H2O desorption obtained by the proposed kinetic model in Equation 7.4 6.5Pd/Al2O3   co-2.9Ce/6.5Pd/Al2O3 T (°C)    (s-1) R2   T (°C)    (s-1) R2 300 3.9410-6 ± 3.8310-7 0.95  300 - - - - 330 1.2110-5 ± 1.8910-6 0.80  330 4.8010-5 ± 6.3710-6 0.86 350 1.3910-5 ± 1.1410-6 0.94  350 1.2510-4 ± 5.6010-6 0.98 380 4.2810-5 ± 4.1810-6 0.95    380 1.8910-4 ± 2.1510-5 0.92     values increase with increasing reaction temperature for both 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts, indicating a higher rate of H2O desorption from active sites at higher temperatures. Comparing the   values for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts show larger    for co-2.9Ce/6.5Pd/Al2O3 than 6.5Pd/Al2O3, indicating faster H2O desorption for co-2.9Ce/6.5Pd/Al2O3 catalyst than 6.5Pd/Al2O3 catalyst. The activation energy of the H2O desorption rate constant (    ) for both 6.5Pd/Al2O3 and co-149  2.9Ce/6.5Pd/Al2O3 catalysts were obtained from Figure 7.3 by plotting         values as a function of      .    values were obtained as 89.6±7.9 kJ.mol-1 and 87.2±28.8 kJ.mol-1 for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts showing almost identical values of    , while    values are higher in the case of co-2.9Ce/6.5Pd/Al2O3 catalyst.  1.50 1.55 1.60 1.65 1.70 1.75-17-16-15-14-13-12-11-10-9-8-7ln kr1000/T (K-1)   Figure 7.3. ln   values as a function of       for 0.3Pd/Al2O3 catalyst (▲), 2.6Pd/Al2O3 catalyst (∆), 6.5Pd/Al2O3 catalyst (■), and co-2.9Ce/6.5Pd/Al2O3 catalyst (○)  7.3 Discussion  The effect of H2O is significant at low temperatures (< 500°C) and at high concentrations of H2O. Following the study by Kikuchi et al. [23], and as presented in Chapters 4 and 5, it is 150  assumed that H2O adsorption-desorption on active sites is the cause of catalyst deactivation. Since the CH4 dissociation on the Pd-*/PdO surface is affected by H2O [30,77,86], it is more likely that the rate of H2O adsorption is faster than the rate of desorption. In addition, based on the TOS results presented in the previous chapters, the CH4 conversion at a constant temperature with H2O added to the CH4/O2 feed varies with TOS, indicating that the H2O adsorption-desorption needs a significant time period to reach equilibrium. The data presented in previous chapters suggest that a TOS of 5h is needed for the catalyst activity to reach steady state after the addition of H2O to the feed gas. Therefore, in this chapter the effect of H2O was accounted for through a non-equilibrium adsorption of H2O applied to the CH4 conversion data measured within the first 5h TOS, prior to steady state activity being achieved. The obtained    values using the linear regression of Equation 7.4 indicates higher    values for co-2.9Ce/6.5Pd/Al2O3 catalyst than 6.5Pd/Al2O3 catalyst which is in agreement with the results presented in Chapter 5, showing H2O adsorption is reduced by the presence of CeO2 on the catalyst surface compared to the catalysts without CeO2.   7.4 Conclusion  The negative effect of H2O on the CH4-O2 reaction under wet-TOS reaction conditions at constant temperature was modeled by extending the kinetic model with H2O inhibition to the case of non-steady state operation. A Langmuir-Hinshelwood mechanism along with C-H bond activation as a rate determining step in the presence of H2O adsorption-desorption was assumed to model the negative effect of H2O. The active sites are mainly covered by H2O molecules, assuming negligible coverage by CH4, O2, and CO2. The non-equilibrium H2O adsorption indicates a higher rate of adsorption than desorption, confirming less inhibiting 151  effect of H2O by increasing the temperature. The number of active sites, n, was considered as 1, assuming the dissociation of a CH4 molecule takes place on a unit Pd-*/PdO site pair. The non-steady state mole balance equation was fitted to the experimental data for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts in the range of 300-380°C in order to explain the exponential activity loss for the first 5h of TOS. The H2O desorption rate constants,   , obtained by this linear regression fitting showed the higher values for co-2.9Ce/6.5Pd/Al2O3 catalyst than 6.5Pd/Al2O3 catalyst and at higher temperatures concluding the positive effect of CeO2 and temperature to increase the H2O desorption from the catalyst surface. 152  Chapter 8: Conclusions and Recommendations  8.1 Conclusions  This study focuses on the inhibiting effects of H2O during CH4 oxidation over PdO/Al2O3 and CeO2/PdO/Al2O3 catalysts. A kinetic analysis of the steady-state and dynamic response of the catalyst activity to H2O addition to the feed gas is reported. This approach provides a kinetic interpretation of the role of CeO2 in reducing the inhibiting effects of H2O on the CH4 oxidaton reaction. A comparison of the kinetic parameters in the presence of H2O for the PdO/Al2O3 and CeO2/PdO/Al2O3 catalysts studied in this thesis also confirms the role of high oxygen storage capacity materials such as CeO2 on the catalyst activity.  The inhibiting effects of H2O on the CH4 oxidation activity of PdO/Al2O3 catalysts at temperatures in the range of 300-380°C were examined. Increasing the loading of Pd from 0.3wt.% to 6.5wt.% resulted in increasing the Pd/Al ratio on the catalyst surface and decreasing Pd dispersion. A higher initial activity and smaller activity loss was observed for the 6.5Pd/Al2O3 catalyst compared to the 0.3Pd/Al2O3 and the 2.6Pd/Al2O3 catalysts as a result of having a higher number of active sites. H2O adsorption is determined to be the main cause of activity loss in CH4 oxidation and the rate of recovery when the H2O is removed from the feed depends on the amount of H2O adsorbed on active sites. The H2O partial pressure per number of active sites (      ) was lowest for the 6.5Pd/Al2O3 catalyst compared with 0.3Pd/Al2O3 and 2.6Pd/Al2O3 catalysts, indicating a faster recovery of catalyst activity for the 6.5Pd/Al2O3 catalyst upon removal of the extra 5vol.%H2O from the feed gas.   153  The rate constant and equilibrium constant for H2O adsorption were obtained from kinetic analysis. Applying the design equation to the values of CH4 conversion (    ) at t=5h with and without H2O in the feed, for three catalysts with different Pd loadings, Ea and       were estimated as 60.611.5 kJ.mol-1 and -81.59.1 kJ.mol-1, respectively. Since the same kinetic parameters were able to describe the measured activity data, one concludes that the CH4 oxidation reaction is not structure sensitive. The low values of effectiveness factors obtained in the range of 300-380°C confirm the slow internal diffusion as a consequence of a fast CH4 oxidation reaction.   The effect of Ce loading with a constant Pd loading of 6.5wt.% was examined using catalysts prepared by the co-impregnation method. The initial activity of the catalysts decreased with increased Ce loading, because of a loss in BET area and consequently PdO encapsulation at higher Ce loading. However, by comparing the ratio of the deactivation rate (      ) between the 6.5Pd/Al2O3 and the co-2.9Ce/6.5Pd/Al2O3 catalysts, showed that CeO2 addition reduced the inhibiting effect of H2O adsorption on the active sites. The lower magnitude of      for the co-2.9Ce/6.5Pd/Al2O3 catalyst comparing to the 6.5Pd/Al2O3 catalyst, confirmed less H2O adsorption on the CeO2 promoted catalyst. However, a higher        for the co-2.9Ce/6.5Pd/Al2O3 catalyst than the 6.5Pd/Al2O3 catalyst implies stronger H2O adsorption on the active sites of the CeO2 promoted catalyst. The small loss in BET surface area and CO uptake for the used 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts following the wet-TOS experiments, confirms some permanent change in physical and chemical properties of the catalysts. The permanent change of the used catalysts properties is attributed to sintering of the catalysts that leads to activity loss. In addition, the inhibiting effect of H2O adsorption is 154  another possible mechanism for loss in catalyst activity. In the dry-TOS experiment, both sintering and H2O adsorption are significant mechanisms of activity loss. However, the effect of H2O adsorption on the active sites is more dominant than sintering in the wet-TOS experiments, as a fast drop in activity is observed once H2O is added to the feed. Upon removal of H2O after 24h wet-TOS, the CH4 conversion is almost identical to that obtained in the dry-TOS, indicating the reversible effect of H2O adsorption caused by extra H2O addition. The reversible effect of H2O is also confirmed by the decrease in the      and    values as a function of temperature, showing less H2O adsorption at higher temperature and hence lower activity loss.   The effect of preparation methods was also examined in order to investigate the effect of CeO2 on the activity of CeOx/PdO/Al2O3. The Al2O3 support was impregnated using different loadings of Ce at a constant Pd loading of 3.4wt.%, using two different preparation methods, co-impregnation and sequential impregnation. Higher catalyst activity and lower rate of deactivation were found for the sequentially impregnated catalysts compared to the co-impregnated catalysts at the same Ce loading. However, the physical and chemical properties of the catalysts prepared by co-impregnation and sequential impregnation methods were not significantly different, confirming that the OSC of CeO2 influences the catalytic properties during the CH4 oxidation reaction. Higher activity and stability of the seq-17Ce/3.4Pd/Al2O3 catalyst than the 3.4Pd/Al2O3 catalyst was explained by the importance of oxygen exchange from the bulk oxide support to the active sites and the role of CeO2 to facilitate this mechanism.  155  A Langmuir adsorption of H2O along with C-H bond activation as a rate determining step were assumed to model the inhibiting effect of H2O on the CH4 oxidation rate. The non-steady state mole balance equation was applied to a set of experimental data obtained with a feed gas of 5000 ppm CH4, 5vol.% H2O and temperature in the range of 300-380°C. The H2O desorption rate constants,   , showed higher values at higher temperatures concluding the positive effect of temperature to increase the H2O desorption from the catalyst surface. In addition, the higher values of    for co-2.9Ce/6.5Pd/Al2O3 catalyst than 6.5Pd/Al2O3 catalyst quantifies the beneficial effect of CeO2 through the increase in the H2O desorption rate constant.  8.2 Recommendations  8.2.1 Kinetic Model Applied to Co-impregnated and Sequentially Impregnated Catalysts   In this study a reactor model was applied to the TOS experimental data for 6.5Pd/Al2O3 and co-2.9Ce/6.5Pd/Al2O3 catalysts in order to calculate the activation energy and enthalpy of H2O adsorption. Since the number of experimental data points reported in Chapter 6 considering both dry-TOS and wet-TOS for co-impregnated and sequentially impregnated catalysts were insufficient for kinetic analysis, calculating the activation energy and enthalpy of H2O adsorption was not possible. It is recommended that the reactor model be extended to the co-impregnated and sequentially impregnated catalysts in order to quantify the effect of preparation method on the activity and stability of the catalysts. In addition, the presented reactor model can be applied to the expremintal data in order to predict the kinetic 156  parameters for the co-impregnated and sequentially impregnated catalysts. It is anticipated that the estimated parameters for the co-impregnated catalysts presented in Chapter 6 are similar to the co-2.9Ce/6.5Pd/Al2O3 catalyst presented in Chapter 5 with the same Ce:Pd ratio.  8.2.2 Studying the Effect of CeO2 on O2 Concentration  The purpose of this study was to investigate the initial activity and stability of the catalysts in a lean-burn condition. Therefore, an excess amount of O2 (20(v/v)%) was fed to the reactor, however, it was kept constant. Groppi et al. [58] showed that a large excess of O2 in CH4 oxidation, as used in the present study, does not significantly affect the activity of the catalysts at low temperature, even in the case of a CeO2/PdO/Al2O3 catalyst. Furthermore, the reported zero-order dependence in oxygen of the rate of reaction means that the high O2 content used in the present study does not impact the kinetics significantly. To confirm the zero order of the reaction rate with respect to O2 (either the O2 in the gas phase or in the bulk of the support), it is suggested that the effect of different preparation methods, the role of different Ce loadings on oxygen exchange capacity of the support, and the inhibiting effect of H2O be investigated as a function of various O2 concentrations. This can lead to a better understanding of the role of the oxygen transfer mechanism during the CH4 oxidation reaction. These experiments should also be extended to examine the degree of CH4 oxidation in the absence of gas phase O2. 8.2.3 Studying the Effect of Support on H2O Adsorption Enthalpy of H2O adsorption and the rate constant for H2O desorption were compared for   157  PdO/Al2O3 and CeOx/PdO/Al2O3 catalysts to emphasize the effect of CeO2 promoted Al2O3 on the inhibiting effect of H2O. However, the role of Al2O3 on H2O adsorption is not clear. Hence, is it suggested that the       value for the Al2O3 support should also be obtained using FTIR and Raman spectroscopy. Comparing the       values of the Al2O3 support, PdO/Al2O3 catalyst, and CeOx/PdO/Al2O3 catalyst can help to better understanding of the inhibiting effect of H2O. The same approach could also be applied to different types of supports (CeO2 or ZrO2) to study the inhibiting effect of H2O on different supports. It is also recommended to investigate the effect of hydrophobicity of the support on H2O adsorption.   8.2.4 Studying the Catalytic Properties during CH4 Oxidation Reaction  The effect of different Pd loadings and Ce:Pd ratios on the physical and chemical properties of the catalysts were confirmed using different characterization techniques such as BET, XPS, XRD, CO chemisorption, and ToF-SIMS. However, the effect of different preparation methods on the catalyst properties was not significant based on these analyses. Hence the preparation methods have an influence on the properties of the catalysts during the CH4 oxidation reaction that are not observable by ex situ characterization techniques. It is recommended that the surface properties of the catalysts be determined by online reaction monitoring. The in situ techniques should include FTIR and Raman spectroscopy to monitor the O transfer between the oxide support, Pd-*, and gas phase. Interaction of PdO and Al2O3 in the presence and absence of CeOx could also be monitored in order to determine the effect of CeO2 promoted support to facilitate H2O desorption.  158  8.2.5 Studying the Partially Reversible Effect of H2O by TPO  The partially reversible effect of H2O was confirmed upon removing the extra H2O after 24h wet-TOS experiments. Since a slight activity loss was observed after 24h for both dry-TOS and wet-TOS, it is concluded that a permanent change in properties of the catalyst occurs during TOS experiments. This permanent change was also confirmed by characterization of the used catalysts. However, the small activity loss could also be the effect of H2O produced during the CH4 oxidation reaction. To understand this, the activity of the used catalysts can be measured by TPO following purging the catalyst bed with an inert gas (e.g. He) for 2h to remove all the reactants and produced H2O. Comparing the CH4 conversion obtained by fresh and used catalysts can explain if the activity loss is permanent or only caused by the presence of H2O.   159  Bibliography  [1] NaturalGas.org, http://naturalgas.org/overview/background/ (Accessed July 20, 2016). [2] Natural Gas Vehicles for America, http://www.ngvamerica.org/natural-gas/ (Accessed July 20, 2016). [3] Alternative Fuels Data Center, http://www.afdc.energy.gov/vehicles/natural_gas.html (Accessed July 20, 2016). [4] Inventory of U.S. Greenhouse Gas Emissions and Sinks, http://www3.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2015-Main-Text.pdf (Accessed July 20, 2016). [5] United States Environmental Protection Agency, https://www.epa.gov/sites/production/files/2016-03/documents/420b16006.pdf (Accessed July 20, 2016). [6] Livescience, http://www.livescience.com/32691-what-are-greenhouse-gases-and-how-do-they-warm-the-earth.html (Accessed July 20, 2016). [7] United States Environmental Protection Agency, http://www3.epa.gov/climatechange/science/causes.html (Accessed July 20, 2016). [8] C.H. Bartholomew, R.J. Farrauto, Fundamentals of industrial catalytic processes, 2nd edition, A JOHN WILEY & SONS, INC, Hoboken, New Jersey, 2006. [9] I. Chorkendorff, J.W. Niemantsverdriet, Concepts of Modern Catalysis and Kinetics, WILEY-VCH Verlag GmbH & Co., 2003. [10] P.G. Boulter, J. Borken-Kleefeld, L. Ntziachristos, The Evolution and Control of NOx Emissions from Road Transport in Europe, 2012.  [11] The Three-way Catalytic Converter, http://find.jorum.ac.uk/resources/10949/950 160  (Accessed July 20, 2016). [12] P. Gélin, M. Primet, Complete oxidation of methane at low temperature over noble metal based catalysts: a review, Appl. Catal. B Environ. 39 (2002) 1–37.  [13] J.-W. Snoeck, G.F. Froment, M. Fowles, Kinetic Study of the Carbon Filament Formation by Methane Cracking on a Nickel Catalyst, J. Catal. 169 (1997) 250–262.  [14] M. Aryafar, F. Zaera, Kinetic study of the catalytic oxidation of alkanes over nickel, palladium, and platinum foils, Catal. Letters. 48 (1997) 173–183. [15] R.J. Farrauto, M.C. Hobson, T. Kennelly, E.M. Waterman, Catalytic chemistry of supported palladium for combustion of methane, Appl. Catal. A Gen. 81 (1992) 227–237. [16] J.K. Lampert, M.S. Kazi, R.J. Farrauto, Palladium catalyst performance for methane emissions abatement from lean burn natural gas vehicles, Appl. Catal. B Environ. 14 (1997) 211–223.  [17] D. Ciuparu, M.R. Lyubovsky, E. Altman, L.D. Pfefferle, A. Datye, Catalytic Combustion of Methane Over Palladium-Based Catalysts, Catal. Rev. 44 (2002) 593–649.  [18] R.G. Silver, J.C. Summers, Catalysts for Natural Gas Emission Control Applications, Catal. Automot. Pollut. Control III Stud. Surf. Sci. Catal., Elsevier Science 96 (1995) 871–884. [19] D. Bounechada, G. Groppi, P. Forzatti, K. Kallinen, T. Kinnunen, Effect of periodic lean/rich switch on methane conversion over a Ce–Zr promoted Pd-Rh/Al2O3 catalyst in the exhausts of natural gas vehicles, Appl. Catal. B Environ. 119-120 (2012) 91–99.  [20] T.V. Choudhary, S. Banerjee, V.R. Choudhary, Catalysts for combustion of methane 161  and lower alkanes, Appl. Catal. A Gen. 234 (2002) 1–23.  [21] G. Centi, Supported palladium catalysts in environmental catalytic technologies for gaseous emissions, J. Mol. Catal. A Chem. 173 (2001) 287–312. d [22] P. Forzatti, G. Groppi, Catalytic combustion for the production of energy, Catal. Today. 54 (1999) 165–180.  [23] R. Kikuchi, S. Maeda, K. Sasaki, S. Wennerström, K. Eguchi, Low-temperature methane oxidation over oxide-supported Pd catalysts: inhibitory effect of water vapor, Appl. Catal. A Gen. 232 (2002) 23–28.  [24] C.L. Pieck, C.R. Vera, E.M. Peirotti, J.C. Yori, Effect of water vapor on the activity of Pt-Pd/Al2O3 catalysts for methane combustion, Appl. Catal. A Gen. 226 (2002) 281–291.  [25] D. Roth, P. Gélin, M. Primet, E. Tena, Catalytic behaviour of Cl-free and Cl-containing Pd/Al2O3 catalysts in the total oxidation of methane at low temperature, Appl. Catal. A Gen. 203 (2000) 37–45.  [26] G. Lapisardi, L. Urfels, P. Gélin, M. Primet, A. Kaddouri, E. Garbowski, et al., Superior catalytic behaviour of Pt-doped Pd catalysts in the complete oxidation of methane at low temperature, Catal. Today. 117 (2006) 564–568.  [27] P. Euzen, J.-H. Le Gal, B. Rebours, G. Martin, Deactivation of palladium catalyst in catalytic combustion of methane, Catal. Today. 47 (1999) 19–27.  [28] M. Ferrandon, J. Carnö, S. Järås, E. Björnbom, Total oxidation catalysts based on manganese or copper oxides and platinum or palladium II: Activity, hydrothermal stability and sulphur resistance, Appl. Catal. A Gen. 180 (1999) 153–161.  [29] P. Gélin, L. Urfels, M. Primet, E. Tena, Complete oxidation of methane at low 162  temperature over Pt and Pd catalysts for the abatement of lean-burn natural gas fuelled vehicles emissions: influence of water and sulphur containing compounds, Catal. Today. 83 (2003) 45–57.  [30] D. Ciuparu, L. Pfefferle, Support and water effects on palladium based methane combustion catalysts, Appl. Catal. A Gen. 209 (2001) 415–428.  [31] J.N. Carstens, S.C. Su, A.T. Bell, Factors Affecting the Catalytic Activity of Pd/ZrO2 for the Combustion of Methane, J. Catal. 176 (1998) 136–142.  [32] D. Ciuparu, L. Pfefferle, Contributions of lattice oxygen to the overall oxygen balance during methane combustion over PdO-based catalysts, Catal. Today. 77 (2002) 167–179.  [33] W.R. Schwartz, D. Ciuparu, L.D. Pfefferle, Combustion of Methane over Palladium-Based Catalysts: Catalytic Deactivation and Role of the Support, J. Phys. Chem. C. 116 (2012) 8587–8593.  [34] W.R. Schwartz, L.D. Pfefferle, Combustion of Methane over Palladium-Based Catalysts: Support Interactions, J. Phys. Chem. C. 116 (2012) 8571–8578.  [35] M. Cargnello, J.J.D. Jaén, J.C.H. Garrido, K. Bakhmutsky, T. Montini1, J.J.C. Gámez, et al., Exceptional Activity for Methane Combustion over Modular Pd@CeO2 Subunits on Functionalized Al2O3, Science 337 (2012) 713–717.  [36] D. Ciuparu, E. Perkins, L. Pfefferle, In situ DR-FTIR investigation of surface hydroxyls on γ-Al2O3 supported PdO catalysts during methane combustion, Appl. Catal. A Gen. 263 (2004) 145–153.  [37] P. Araya, S. Guerrero, J. Robertson, F.J. Gracia, Methane combustion over Pd/SiO2 catalysts with different degrees of hydrophobicity, Appl. Catal. A Gen. 283 (2005) 163  225–233.  [38] A. Gannouni, B. Albela, M.S. Zina, L. Bonneviot, Metal dispersion, accessibility and catalytic activity in methane oxidation of mesoporous templated aluminosilica supported palladium, Appl. Catal. A Gen. 464-465 (2013) 116–127.  [39] G. Zhu, K. Fujimoto, D.Y. Zemlyanov, A.K. Datye, F.H. Ribeiro, Coverage of palladium by silicon oxide during reduction in H2 and complete oxidation of methane, J. Catal. 225 (2004) 170–178.  [40] R. Lamber, N. Jaeger, G. Schulz-Ekloff, Metal-support interaction in the Pd/SiO2 system: Influence of the support pretreatment, J. Catal. 123 (1990) 285–297.  [41] C.R.F. Lund, J.A. Dumesic, Strong oxide-oxide interactions in silica-supported Fe3O4: III. Water-induced migration of silica on geometrically designed catalysts, J. Catal. 72 (1981) 21–30.  [42] H. Yoshida, T. Nakajima, Y. Yazawa, T. Hattori, Support effect on methane combustion over palladium catalysts, Appl. Catal. B Environ. 71 (2007) 70–79.  [43] K. Persson, A. Ersson, S. Colussi, A. Trovarelli, S.G. Järås, Catalytic combustion of methane over bimetallic Pd–Pt catalysts: The influence of support materials, Appl. Catal. B Environ. 66 (2006) 175–185.  [44] H.C. Yao, Y.F.Y. Yao, Ceria in automotive exhaust catalysts: I. Oxygen storage, J. Catal. 86 (1984) 254–265.  [45] E.C. Su, C.N. Montreuil, W.G. Rothschild, Oxygen storage capacity of monolith three-way catalysts, Appl. Catal. 17 (1985) 75–86.  [46] R.K. Hailstone, A.G. DiFrancesco, J.G. Leong, T.D. Allston, K.J. Reed, A Study of Lattice Expansion in CeO2 Nanoparticles by Transmission Electron Microscopy, J. 164  Phys. Chem. C. 113 (2009) 15155–15159.  [47] N. V. Skorodumova, S.I. Simak, B.I. Lundqvist, I.A. Abrikosov, B. Johansson, Quantum Origin of the Oxygen Storage Capability of Ceria, Phys. Rev. Lett. 89 (2002) 166601–1–166601–4.  [48] S. Tsunekawa, R. Sivamohan, S. Ito, A. Kasuya, T. Fukuda, Structural study on monosize CeO2-x nano-particles, Nanostructured Mater. 11 (1999) 141–147.  [49] L. Kȩpiński, J. Okal, Occurrence and Mechanism of Formation of CeOCl in Pd/CeO2 Catalysts, J. Catal. 192 (2000) 48–53.  [50] S. Colussi, A. Trovarelli, G. Groppi, J. Llorca, The effect of CeO2 on the dynamics of Pd–PdO transformation over Pd/Al2O3 combustion catalysts, Catal. Commun. 8 (2007) 1263–1266.  [51] S. Colussi, A. Trovarelli, E. Vesselli, A. Baraldi, G. Comelli, G. Groppi, et al., Structure and morphology of Pd/Al2O3 and Pd/CeO2/Al2O3 combustion catalysts in Pd–PdO transformation hysteresis, Appl. Catal. A Gen. 390 (2010) 1–10.  [52] R. Ramírez-Lópeza, I. Elizalde-Martineza, L. Balderas-Tapia, Complete catalytic oxidation of methane over Pd/CeO2–Al2O3: The influence of different ceria loading, Catal. Today. 150 (2010) 358–362. doi:10.1016/j.cattod.2009.10.007. [53] S. Fouladvand, S. Schernich, J. Libuda, H. Gronbeck, T. Pingel, E. Olsson, et al., Methane Oxidation Over Pd Supported on Ceria–Alumina Under Rich/Lean Cycling Conditions, Top. Catal. Catal. 56 (2013) 410–415.  [54] S. Colussi, A. Gayen, M.B. Matteo Farnesi Camellone, J. Llorca, S. Fabris, A. Trovarelli, Nanofaceted Pd-O Sites in Pd-Ce Surface Superstructures: Enhanced Activity in Catalytic Combustion of Methane, Angew. Chem. Int. Ed. 48 (2009) 8481 165  –8484.  [55] K.R. Priolkar, P. Bera, P.R. Sarode, M.S. Hegde, S. Emura, R. Kumashiro, et al., Formation of Ce1-xPdxO2-δ Solid Solution in Combustion-Synthesized Pd/CeO2 Catalyst: XRD, XPS, and EXAFS Investigation, J. Chem. Mater. 14 (2002) 2120–2128.  [56] A.D. Mayernick, M.J. Janik, Methane oxidation on Pd–Ceria: A DFT study of the mechanism over PdxCe1-xO2, Pd, and PdO, J. Catal. 278 (2011) 16–25.  [57] B. Stasinska, W. Gac, T. Ioannides, A. Machocki, Complete Oxidation of Methane over Palladium Supported on Alumina Modified with Calcium, Lanthanum, and Cerium Ions, J. Nat. Gas Chem. 16 (2007) 342–348.  [58] G. Groppi, C. Cristiani, L. Lietti, C. Ramella, M. Valentini, P. Forzatti, Effect of ceria on palladium supported catalysts for high temperature combustion of CH4 under lean conditions, Catal. Today. 50 (1999) 399–412.  [59] S. Colussi, A. Trovarelli, C. Cristiani, L. Lietti, G. Groppi, The influence of ceria and other rare earth promoters on palladium-based methane combustion catalysts, Catal. Today. 180 (2012) 124–130.  [60] L. Xiao, K. Sun, X. Xu, X. Li, Low-temperature catalytic combustion of methane over Pd/CeO2 prepared by deposition–precipitation method, Catal. Commun. 6 (2005) 796–801.  [61] C. Bozo, N. Guilhaume, J.-M. Herrmann, Role of the Ceria–Zirconia Support in the Reactivity of Platinum and Palladium Catalysts for Methane Total Oxidation under Lean Conditions, J. Catal. 203 (2001) 393–406.  [62] L.M. Misch, J.A. Kurzman, A.R. Derk, Y.-I. Kim, R. Seshadri, H. Metiu, et al., C–H 166  Bond Activation by Pd-substituted CeO2: Substituted Ions versus Reduced Species, Chem. Mater. 23 (2011) 5432–5439.  [63] A.D. Mayernick, M.J. Janik, Ab initio thermodynamic evaluation of Pd atom interaction with CeO2 surfaces, J. Chem. Phys. 131 (2009) 84701–1–84701–12.  [64] R.F. Hicks, H. Qi, M.L. Young, R.G. Lee, Structure sensitivity of methane oxidation over platinum and palladium, J. Catal. 122 (1990) 280–294.  [65] R.F. Hicks, H. Qi, M.L. Young, R.G. Lee, Effect of catalyst structure on methane oxidation over palladium on alumina, J. Catal. 122 (1990) 295–306.  [66] K. Otto, L.P. Haack, J.E. DeVries, Identification of two types of oxidized palladium on γ-alumina by X-ray photoelectron spectroscopy, Appl. Catal. B Environ. 1 (1992) 1–12.  [67] J. Goetz, M.A. Volpe, A.M. Sica, C.E. Gigola, R. Touroude, Low-Loaded Palladium on α-Alumina Catalysts: Characterization by Chemisorption, Electron-Microscopy, and Photoelectron Spectroscopy, J. Catal. 153 (1995) 86–93.  [68] V.H. Sandoval, C.E. Gigola, Characterization of Pd-Pb/Al2O3 catalysts. A TPR-TPD study, Appl. Catal. A Gen. 148 (1996) 81–96.  [69] P. Castellazzi, G. Groppi, P. Forzatti, A. Baylet, P. Marécot, D. Duprez, Role of Pd loading and dispersion on redox behaviour and CH4 combustion activity of Al2O3 supported catalysts, Catal. Today. 155 (2010) 18–26.  [70] T.R. Baldwin, R. Burch, Remarkable activity enhancement in the catalytic combustion of methane on supported palladium catalysts, Catal. Letters. 6 (1990) 131–138. [71] C.F. Cullis, B.M. Willatt, Oxidation of methane over supported precious metal catalysts, J. Catal. 83 (1983) 267–285.  167  [72] F.H. Ribeiro, M. Chow, R.A. Dallabetta, Kinetics of the Complete Oxidation of Methane over Supported Palladium Catalysts, J. Catal. 146 (1994) 537–544.  [73] G. Zhu, J. Han, D.Y. Zemlyanov, F.H. Ribeiro, The Turnover Rate for the Catalytic Combustion of Methane over Palladium is Not Sensitive to the Structure of the Catalyst, J. Am. Chem. Soc. 126 (2004) 9896–9897.  [74] K. Fujimoto, F.H. Ribeiro, M. Avalos-Borja, E. Iglesia, Structure and Reactivity of PdOx/ZrO2 Catalysts for Methane Oxidation at Low Temperatures, J. Catal. 179 (1998) 431–442.  [75] K. Muto, N. Katada, M. Niwa, Complete oxidation of methane on supported palladium catalyst: Support effect, Appl. Catal. A Gen. 134 (1996) 203–215.  [76] J.C. van Giezen, F.R. van den Berg, J.L. Kleinen, A.J. van Dillen, J.W. Geus, The effect of water on the activity of supported palladium catalysts in the catalytic combustion of methane, Catal. Today. 47 (1999) 287–293.  [77] P. Hurtado, S. Ord  ez, H. Sastre, F. V. D  ez, Development of a kinetic model for the oxidation of methane over Pd/Al2O3 at dry and wet conditions, Appl. Catal. B Environ. 51 (2004) 229–238.  [78] C.F. Cullis, B.M. Willatt, The inhibition of hydrocarbon oxidation over supported precious metal catalysts, J. Catal. 86 (1984) 187–200.  [79] C.F. Cullis, T.G. Nevell, D.L. Trimm, Role of the Catalyst Support in the Oxidation of Methane over Palladium, J. Chem. Soc., Faraday Trans. 1. 68 (1972) 1406–1412.  [80] R.J. Card, J.L. Schmitt, J.M. Simpson, Palladium-carbon hydrogenolysis catalysts: The effect of preparation variables on catalytic activity, J. Catal. 79 (1983) 13–20.  [81] D. Ciuparu, N. Katsikis, L. Pfefferle, Temperature and time dependence of the water 168  inhibition effect on supported palladium catalyst for methane combustion, Appl. Catal. A Gen. 216 (2001) 209–215.  [82] Y.-H. (Cathy) Chin, E. Iglesia, Elementary Steps, the Role of Chemisorbed Oxygen, and the Effects of Cluster Size in Catalytic CH4-O2 Reactions on Palladium, J. Phys. Chem. C. 115 (36) (2011) 17845–17855.  [83] G. Groppi, Combustion of CH4 over a PdO/ZrO2 catalyst: an example of kinetic study under severe conditions, Catal. Today. 77 (2003) 335–346.  [84] J. Cortés, E. Valencia, Configuration of Adsorbed Phases and Their Evolution to Absorbent States in the CH4–O2 Catalytic Reaction, Bull. Chem. Soc. Jpn. 82 (2009) 683–688.  [85] J. Cortés, E. Valencia, P. Araya, Monte Carlo simulation studies of the catalytic combustion of methane, Catal. Letters. 112 (2006) 121–128.  [86] J. Cortés, E. Valencia, M. Pineda, P. Araya, Temporal decay of the catalytic oxidation of methane over palladium supported on silica, Sci. China Chem. 56 (2013) 1601–1607. [87] Perry’s Chemical Engineers' Handbook, 7th edition, McGraw Hill, 1999. [88] R. Gholami, K.J. Smith, Activity of PdO/SiO2 catalysts for CH4 oxidation following thermal treatments, Appl. Catal. B Environ. 168–169 (2015) 156–163.  [89] R.B. Bird, W. E.Stewart, E. N.Lightfood, Transport Phenomena, 2nd edition, John Wiley & Sons, Inc., 2001. [90] D.E. Mears, Tests for Transport Limitations in Experimental Catalytic Reactors, Ind. Eng. Chem. Process Des. Dev. 10 (1971) 541–547.  [91] R. Abbasi, L. Wu, S.E. Wanke, R.E. Hayes, Kinetics of methane combustion over Pt 169  and Pt–Pd catalysts, Chem. Eng. Res. Des. 90 (2012) 1930–1942.  [92] H.. Fogler, Elements of Chemical Reaction Engineering, Prentice-Hall, NJ, 1992. [93] R. Rudham, M.. Sanders, The catalytic properties of zeolite X containing transition metal ions: Part 2—Methane oxidation, J. Catal. 27 (1972) 287–292.  [94] D.L. Simpson, A numerical method of characteristics for solving hyperbolic partial differential equations, Iowa State University, 1967. [95] W.H. Press, A.A. Teukolsky, W.T. Vetterling, B.P. Flannery, Numerical Recipes in Fortran 77, 2nd edition, Press Syndicate of the University of Cambridge, 1992. [96] B. Stasinska, A. Machocki, K. Antoniak, M. Rotko, J.L. Figueiredo, F. Gonçalves, Importance of palladium dispersion in Pd/Al2O3 catalysts for complete oxidation of humid low-methane–air mixtures, Catal. Today. 137 (2008) 329–334.  [97] R. Gholami, M. Alyani, K.J. Smith, Deactivation of Pd Catalysts by Water during Low Temperature Methane Oxidation Relevant to Natural Gas Vehicle Converters, Catalysts. 5 (2015) 561–594.  [98] B. Zhang, X. Wang, O. M’Ramadj, D. Li, H. Zhang, G. Lu, Effect of water on the performance of Pd-ZSM-5 catalysts for the combustion of methane, J. Nat. Gas Chem. 17 (2008) 87–92.  [99] Q. Xu, K.C. Kharas, B.J. Croley, A.K. Datye, The Sintering of Supported Pd Automotive Catalysts, Chem Cat Chem. 3 (2011) 1004–1014.  [100] K. Narui, H. Yata, K. Furuta, A. Nishida, Y. Kohtoku, T. Matsuzaki, Effects of addition of Pt to PdO/Al2O3 catalyst on catalytic activity for methane combustion and TEM observations of supported particles, Appl. Catal. A Gen. 179 (1999) 165–173.  [101] R.E. Hayes, S.T. Kolaczkowskib, P.K.C. Li, S. Awdry, Evaluating the effective 170  diffusivity of methane in the washcoat of a honeycomb monolith, Appl. Catal. B Environ. 25 (2000) 93–104.  [102] F.P.J.M. Kerkhof, J.A. Moulijn, Quantitative Analysis of XPS Intensities for Supported Catalysts, J. Phys. Chem. 83 (1979) 1612–1619.  [103] A. Trovarelli, Catalysis by Ceria and Related Materials, Imperial College Press, London, 2001. [104] W.H. Cassinelli, L.S.F. Feio, J.C.S. Arau´jo, C.E. Hori, F.B. Noronha, C.M.P. Marques, et al., Effect of CeO2 and La2O3 on the Activity of CeO2–La2O3/Al2O3- Supported Pd Catalysts for Steam Reforming of Methane, Catal Lett. 120 (2008) 86–94.  [105] P.W. Park, J.S. Ledford, Effect of Crystallinity on the Photoreduction of Cerium Oxide: A Study of CeO2 and Ce/Al2O3 Catalysts, Langmuir. 12 (1996) 1794–1799.  [106] M.H. Zoellner, P. Zaumseil, H. Wilkens, S. Gevers, J. Wollschläger, M. Bäumer, et al., Stoichiometry–structure correlation of epitaxial Ce1−xPrxO2−δ (x=0−1) thin films on Si(111), Cryst. Growth. 355 (2012) 159–165.  [107] S.A. Acharya, V.M. Gaikwad, S.W. D’Souza, S.R. Barman, Gd/Sm dopant-modified oxidation state and defect generation in nano-ceria, Solid State Ionics. 260 (2014) 21–29.  [108] NIST X-ray Photoelectron Spectroscopy Database, http://srdata.nist.gov/xps/selEnergyType.aspx (Accessed July 20, 2016). [109] J.Z. Shyu, W.H. Weber, H.S. Gandhi, Surface characterization of alumina-supported ceria, J. Phys. Chem. 92 (1988) 4964–4970.  [110] K. Persson, L.D. Pfefferle, W. Schwartz, A. Ersson, S.G. Järås, Stability of palladium-171  based catalysts during catalytic combustion of methane: The influence of water, Appl. Catal. B Environ. 74 (2007) 242–250.  [111] Rahman Gholami Shahrestani, Kinetic and Deactivation Studies of Methane Oxidation over Palladium Catalysts in the Presence of Water, University of British Columbia, 2015. [112] S. Eriksson, M. Boutonnet, S. Järås, Catalytic combustion of methane in steam and carbon dioxide-diluted reaction mixtures, Appl. Catal. A Gen. 312 (2006) 95–101.  [113] D. Gao, S. Wang, C. Zhang, Z. Yuan, S. Wang, Methane Combustion over Pd/Al2O 3 Catalyst: Effects of Chlorine Ions and Water on Catalytic Activity, Chinese J. Catal. 29 (2008) 1221–1225.  [114] J.-H. Park, B. Kim, C.-H. Shin, G. Seo, S.H. Kim, S.B. Hong, Methane Combustion over Pd Catalysts Loaded on Medium and Large Pore Zeolites, Top Catal. 52 (2009) 27–34.  [115] Y. Liu, S. Wang, D. Gao, T. Sun, C. Zhang, S. Wang, Influence of metal oxides on the performance of Pd/Al2O3 catalysts for methane combustion under lean-fuel conditions, Fuel Process. Technol. 111 (2013) 55–61.  [116] G. Zhu, J. Han, D.Y. Zemlyanov, F.H. Ribeiro, Temperature dependence of the kinetics for the complete oxidation of methane on palladium and palladium oxide, J. Phys. Chem. B. 109 (2005) 2331–2337.  [117] R. Burch, F.J. Urbano, P.K. Loader, Methane combustion over palladium catalysts: The effect of carbon dioxide and water on activity, Appl. Catal. A Gen. 123 (1995) 173–184.  [118] A.K. Datye, J. Bravo, T.R. Nelson, P. Atanasova, M. Lyubovsky, L. Pfefferle, 172  Catalyst microstructure and methane oxidation reactivity during the Pd↔PdO transformation on alumina supports, Appl. Catal. A Gen. 198 (2000) 179–196.  [119] A. Piras, A. Trovarelli, G. Dolcetti, Remarkable stabilization of transition alumina operated by ceria under reducing and redox conditions, Appl. Catal. B Environ. 28 (2000) L77–L81.  [120] A. Barrera, J.A. Montoya, P. del Angel, J. Navarrete, M.E. Cano, F. Tzompantzi, et al., Surface properties of palladium catalysts supported on ternary ZrO2–Al2O3–WOx oxides prepared by the sol–gel method: Study of the chemical state of the support, J. Phys. Chem. Solids. 73 (2012) 1017–1025.  [121] Sonia Gil, J.M. Garcia-Vargas, L.F. Liotta, G. Pantaleo, M. Ousmane, L. Retailleau, et al., Catalytic Oxidation of Propene over Pd Catalysts Supported on CeO2, TiO2, Al2O3 and M/Al2O3 Oxides (M = Ce, Ti, Fe, Mn), Catalysts. 5 (2015) 671–689.  [122] J.Z. Shyu, K. Otto, W.L.H. Watkins, G.W. Graham, R.K. Belitz, H.S. Gandhi, Characterization of Pd/γ-alumina catalysts containing ceria, J. Catal. 114 (1988) 23–33.  [123] R. de S. Monteiro, F.B. Noronha, L.C. Dieguez, M. Schmal, Characterization of Pd-CeO2 interaction on alumina support and hydrogenation of 1,3-butadiene, Appl. Catal. A Gen. 131 (1995) 89–106.  [124] D. Ciuparu, F. Bozon-Verduraz, L. Pfefferle, Oxygen Exchange between Palladium and Oxide Supports in Combustion Catalysts, J. Phys. Chem. B. 106 (2002) 3434–3442.  [125] R. Klaewkla, M. Arend, W.F. Hoelderich, A Review of Mass Transfer Controlling the Reaction Rate in Heterogeneous Catalytic Systems, 2011.  173  [126] M.M.V.M. Souza, D.A.G. Aranda, M. Schmal, Reforming of Methane with Carbon Dioxide over Pt/ZrO2/Al2O3 Catalysts, J. Catal. 204 (2001) 498–511.  [127] M.M.V.M. Souza, D.A.G. Aranda, C.A.C. Pérez, M. Schmal, Surface Characterization of Zirconia-Coated Alumina as Support for Pt Particles, Phys. Status Solidi. 187 (2001) 297–303.  [128] Y. Guo, G. Lu, Z. Zhan, L. Jiang, X. Wang, S. Li, et al., Effects of ZrO2/Al2O3 properties on the catalytic activity of Pd catalysts for methane combustion and CO oxidation, Catal. Today. 126 (2007) 441–448.  [129] C. Amairia, S. Fessi, A. Ghorbel, A. Rîves, Methane oxidation behaviour over sol–gel derived Pd/Al2O3-ZrO2 materials: Influence of the zirconium precursor, J. Mol. Catal. A Chem. 332 (2010) 25–31.  [130] D. Ciuparu, E. Altman, L. Pfefferle, Contributions of Lattice Oxygen in Methane Combustion over PdO-Based Catalysts, J. Catal. 203 (2001) 64–74.  [131] D. Duprez, Study of surface mobility by isotopic exchange: recent developments and perspectives, Stud. Surf. Sci. Catal. 112 (1997) 13–28.  [132] C. Amairia, S. Fessi, A. Ghorbel, Sol gel derived Pd/Al2O3–ZrO2 as catalysts for methane combustion: effect of zirconium loading, J. Sol-Gel Sci. Technol. 54 (2010) 29–35.  [133] K. Sekizawa, H. Widjaja, S. Maeda, Y. Ozawa, K. Eguchi, Low temperature oxidation of methane over Pd catalyst supported on metal oxides, Catal. Today. 59 (2000) 69–74.  [134] S. Yang, A. Maroto-Valiente, M. Benito-Gonzalez, I. Rodriguez-Ramos, A. Guerrero-Ruiz, Methane combustion over supported palladium catalysts: I. Reactivity and active 174  phase, Appl. Catal. B Environ. 28 (2000) 223–233.  [135] J. Au-Yeung, K. Chen, A.T. Bell, E. Iglesia, Isotopic Studies of Methane Oxidation Pathways on PdO Catalysts, J. Catal. 188 (1999) 132–139.  [136] T.P. Beebe Jr, D.W. Goodman, B.D. Kay, J.T. Yates Jr, Kinetics of the activated dissociative adsorption of methane on the low index planes of nickel single crystal surfaces, J. Chem. Phys. 87 (1987) 2305–2315.  [137] H. Yamamoto, H. Uchid, Oxidation of methane over Pt and Pd supported on alumina in lean-burn natural-gas engine exhaust, Catal. Today. 45 (1998) 147–151. [138] K. Persson, A. Ersson, K. Jansson, J.L.G. Fierro, S.G. Järås, Influence of molar ratio on Pd–Pt catalysts for methane combustion, J. Catal. 243 (2006) 14–24.  [139] K. Persson, A. Ersson, K. Jansson, N. Iverlund, S. Järås, Influence of co-metals on bimetallic palladium catalysts for methane combustion, J. Catal. 231 (2005) 139–150.  [140] K. Persson, K. Jansson, S.G. Järås, Characterisation and microstructure of Pd and bimetallic Pd–Pt catalysts during methane oxidation, J. Catal. 245 (2007) 401–414.  [141] W.J. Kuper, M. Blaauw, F. van der Berg, G.H. Graaf, Catalytic combustion concept for gas turbines, Catal. Today. 47 (1999) 377–389.  [142] Y. Ozawa, Y. Tochihara, A. Watanabe, M. Nagai, S. Omi, Deactivation of Pt·PdO/Al2O3 in catalytic combustion of methane, Appl. Catal. A Gen. 259 (2004) 1–7.  [143] T. Ishihara, H. Shigematsu, Y. Abe, Y. Takita, Effects of Additives on the Activity of Palladium Catalysts for Methane Combustion, Chem. Lett. 22 (1993) 407–410.   175             Appendices   176  Appendix A: Catalyst Preparation  The required amounts of chemicals to prepare the catalysts with different loadings of Pd and Ce are reported in Table A.1. The initial calculations for the nominal 10wt.%Pd/Al2O3 are as follows: 1. Total amount of 10wt.%Pd/Al2O3=3 g 2. Amount of Pd in 10wt.%Pd/Al2O3 catalyst                                                                                                                        3. Total mass of N2O6Pd.xH2O salt (Palladium (II) Nitrate Hydrate)                                                                                                                           4. Required amount of Al2O3                                                                                                      5. Required amount of 0.1N HNO3 solution                                                                                                                    The same calculations are applied for 10wt.%Ce/Al2O3 catalyst using Cerium (III) Nitrate Hexahydrate salt, Ce(NO3)3.6H2O, that the molecular weight for Ce and Ce(NO3)3.6H2O are 140.116 g/mol and 434.23 g/mol, respectively. Table A.1 also reports the required amounts of salts and Al2O3 support for co-impregnated and sequentially impregnated catalysts consisting of both Pd and Ce. jkjkjkjkjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjkjjjjjjjjjjjjjjjj                                                         177      Table A.1. Required amounts of N2O6Pd.xH2O salt, Ce(NO3)3.6H2O salt, and Al2O3 support Catalyst  Pd Loading  Ce Loading  Total Mass Mass of  Mass of  Mass of   of Catalyst N2O6Pd.xH2O Salt Ce(NO3)3.6H2O Salt Al2O3 (wt.%) (wt.%) (g) (g) (g) (g) 10Pd/Al2O3 10 0 3 0.65 0.00 2.70 co-1Ce/10Pd/Al2O3 10 1 3 0.65 0.09 2.67 co-3Ce/10Pd/Al2O3 10 3 3 0.65 0.28 2.61 co-5Ce/10Pd/Al2O3 10 5 3 0.65 0.46 2.55 co-10Ce/10Pd/Al2O3 10 10 3 0.65 0.93 2.40 10Ce/Al2O3 0 10 3 0 0.93 2.70 1Pd/Al2O3  1 0 3 0.06 0.00 2.97 5Pd/Al2O3  5 0 3 0.32 0.00 2.85 co-1.5Ce/5Pd/Al2O3 5 1.5 3 0.32 0.14 2.80 co-15Ce/5Pd/Al2O3 5 15 3 0.32 1.39 2.40 co-50Ce/5Pd/Al2O3 5 50 3 0.32 4.65 1.35 seq-1.5Ce/5Pd/Al2O3 5 1.5 3 0.32 0.14 2.80 seq-5Ce/5Pd/Al2O3 5 5 3 0.32 0.46 2.70 seq-15Ce/5Pd/Al2O3 5 15 3 0.32 1.39 2.40 seq-25Ce/5Pd/Al2O3 5 25 3 0.32 2.32 2.10 seq-50Ce/5Pd/Al2O3 5 50 3 0.32 4.65 1.35 178  Appendix B: Catalyst Characterization  B.1 BET   Surface area, pore size, and pore volume are important properties of a catalyst. The most common method for measuring the surface area and pore size of a mesoporous material is by gas adsorption (normally N2). The amount of N2 adsorbed as a function of pressure at 77K is measured volumetrically. The BET (Brunauer-Emmett-Teller) isotherm describes multilayer gas adsorption on the surface of a solid.                                                                                                                                        B.1 The BET equation in linear form is expressed as:                                                                                                                                 B.2        ,                                                                                                                    B.3 So,                                                                                                                             B.4  P is the partial pressure of N2, P0 is the saturation pressure at the adsorption temperature, V is the volume of adsorbed gas at P, Vm is volume adsorbed at monolayer coverage, and C is a constant. By calculating Vm from the isotherm fit to the experimental adsorption data and knowing the area occupied by each N2 molecule, the surface area is obtained.  By plotting y versus x for the P/P0 < 0.3, a line with a specific slope and intercept in which C and Vm would be obtained. Hence the BET surface area is calculated as:                                                                                                                                     B.5 179  σ is the occupied area by adsorption of a single molecule of N2 (16.2 Å2) and NA  is the Avogadro constant. 0.0 0.2 0.4 0.6 0.8 1.0050100150200250300Vadsorbed (cc STP/gcat)Relative Pressure (P/P0) Adsorption Isotherm Desorption Isotherm   Figure B.1. Isotherm linear plot for calcined 6.5wt.%Pd/Al2O3 catalyst  B.2 XRD  X-Ray Diffraction analysis was used to identify the bulk crystalline phases present in the catalysts. The structure of crystalline materials, crystallite size, and degree of crystallinity are determined by XRD. When an X-ray beam hits a crystalline material it is diffracted and the distance between the two different planes of the crystal is described by Bragg's law.                                                                                                                                     B.6   180  where n is the order of the diffracted beam,   is the X-ray wavelength, d is the distance between two planes of atoms (d-spacing), and   is the angle of reflection. d-spacing is calculated by knowing   and  . By comparing the obtained data with the standard reference patterns, the material can be identified.  The crystallite size of a material can be determined from the Scherrer equation.                                                                                                                                         B.7 L is the crystallite size,   is an instrument constant (usually 1),   is the X-ray wavelength (1.7902 Å), β is the width of the peak (full width at half maximum-FWHM), and   is the angle of reflection. If an amorphous material is present in a sample, a broad peak will be observed in the XRD pattern.  B.3 XPS  X-ray photoelectron spectroscopy is a surface technique which is used to measure the oxidation state of the active species and the interaction between a metal and the oxide support. In addition, the detection of small crystallites or thin films which is difficult by XRD, can be determined by XPS. In this technique, an electron from the surface of the catalyst is excited when photons from an X-ray source hit it. When the X-ray photon with the energy of hν hits the surface of the catalyst, an electron with the binding energy of EB is adsorbed on the surface that causes the emission of the core electron of the catalyst with a specific amount of kinetic energy of Ek. A+hν A++e- Ek=hν-EB 181  By measuring Ek and having hν, the binding energy (EB) can be determined. The electron emitted from each component has a specific amount of binding energy. So, EB is used to demonstrate the atom that the electron comes from and also the oxidation state of that atom. For instance, in Pd/Al2O3 catalyst, X-ray hits the surface of the catalyst and the electrons of both palladium and Al2O3 leave the surface with different signals. The intensity of each signal depends on the number of electrons that escape the sample of palladium versus Al2O3. This intensity demonstrates the indication of concentration of Pd on the surface versus Al2O3 on the surface. If the Pd/Al2O3 ratio is high, it indicates high dispersion of Pd and if this ration is low, we have low Pd dispersion on the surface of the catalyst.  B.4 CO Chemisorption  Gas adsorption is commonly used to determine the dispersion of metals for fresh and used catalysts. Metal dispersion is defined as the number of metal atoms on the surface of the catalyst which are exposed to the adsorbate species divided by the total number of metal atoms in the catalyst.                                                                     CO is of the most common used gas for metal dispersion for different types of supported catalysts. In this technique, an inert gas passed through a bed of catalyst which is kept at a specific temperature. Then a pulse of CO is injected into the inert gas. The adsorbed CO is measured by having the amount of CO injected into the system and measuring the amount of gas which is not adsorbed. When the monolayer coverage is obtained, no more CO is adsorbed on the surface from the gas. The amount of adsorbed CO represents the number of metal surface atoms. Unfortunately, the dispersion obtained based on CO adsorption is not 182  consistent because a CO molecule is able to bind with more than one metal atom. Hence, it may not give the actual sites of the catalyst exposed to the reaction species.   A Micromeritics AutoChemII 2920 analyzer was used for CO pulse chemisorption of the reduced catalysts in order to determine the dispersion of active sites. In the analysis, the oxidized catalyst was purged in a 50 cm3(STP).min-1 flow rate of Ar (Praxair, UHP) at 200°C for 2h in order to remove moisture. The catalyst was then cooled to 100°C and held for 1h. After degas, a 50 cc.min-1 flow rate of 9.5(v/v)% H2/Ar (Praxair) at 100°C was fed to the catalyst for 1h, and then cooled to 25°C in He [88]. The purpose of flowing H2/Ar is to partially reduce the catalyst so that a thin layer of PdO is transformed to Pd0 that is able to adsorb CO, without affecting the size of the supported PdO particle. The CO uptake was measured by passing pulses of 9.93 (v/v)% CO/He (Praxair) at 25°C over the partially reduced catalyst. The CO pulse injection continued until no additional chemisorption was observed. The CO uptake was measured using a thermal conductivity detector (TCD).   The CO chemisorption analysis for the co-2.9Ce/6.5Pd/Al2O3 catalyst is described below:  Table B.1. Properties of co-2.9Ce/6.5Pd/Al2O3 catalyst used for CO chemisorption analysis  Sample mass (g) 0.1167 Active loop volume at 108.4°C (cc STP) 4.0410-2 Peak area at zero CO uptake 7.9410-3 Fc (volume-to-area factor) (cc/peak area) 5.09 Stoichiometry factor 1 Pd molecular weight (g/μmol) 106.4210-6     183  Table B.2. Cumulative volume of co-2.9Ce/6.5Pd/Al2O3 catalyst during CO chemisorption analysis Peak Quantity Peak ∆Arean Volume Cumulative Number Adsorbed Arean Adsorbed Quantity  μmol/g  (cc STP) (cc STP) 1 - 0 7.94E-03 4.04E-02 4.04E-02 2 - 0 7.94E-03 4.04E-02 8.09E-02 3 46.37 0 7.94E-03 4.04E-02 1.21E-01 4 15.12 1.80E-04 7.76E-03 3.95E-02 1.61E-01 5 14.03 7.90E-04 7.15E-03 3.64E-02 1.97E-01 6 13.75 9.40E-04 7.00E-03 3.56E-02 2.33E-01 7 13.69 9.80E-04 6.96E-03 3.54E-02 2.68E-01 8 12.89 1.42E-03 6.52E-03 3.32E-02 3.01E-01 9 10.86 2.54E-03 5.40E-03 2.75E-02 3.29E-01 10 9.91 3.06E-03 4.88E-03 2.48E-02 3.54E-01 11 8.82 3.67E-03 4.27E-03 2.17E-02 3.76E-01 12 6.45 4.97E-03 2.97E-03 1.51E-02 3.91E-01 13 4.45 6.08E-03 1.86E-03 9.46E-03 4.00E-01 14 1.47 7.72E-03 2.18E-04 1.11E-03 4.01E-01  The total CO uptake is calculated is as follow:                                                                                                                                                   B.8                  -                 -                                                                                                                                                       B.9                                           -                                    184  Appendix C: MFC and MS Calibration   C.1 MFC Calibration  Mass flow controllers were used to direct the flow rate of CH4/Ar, He, O2, and Ar gases to the reactor separately. Each MFC was calibrated for the wide range of flow rate using a bubble flow meter. Tables C.1 and C.2 present the 9.97(v/v)%CH4/Ar and He MFC calibration results. Table C.1. CH4/Ar calibration using a bubble flow meter  Set Point Volume Time Flow rate % cc s cc(STP).min-1 10 5 38.0 7.3 20 10 34.7 15.9 30 10 24.1 22.9 40 15 26.6 31.1 60 15 17.0 48.8   Table C.2. He calibration using a bubble flow meter  Set Point Volume Time Flow rate % cc s cc(STP).min-1 30 5 34.4 8.0 40 5 25.8 10.7 50 5 20.6 13.4 60 5 17.2 16.0 65 5 15.7 17.6 75 5 13.7 20.1 85 5 12.3 22.4 185  0 10 20 30 40 50 60 7001020304050Flowrate (cc(STP).min-1)Set Point (%)Flowrate=0.7949*Set point          R2=0.997   Figure C.1. MFC Calibration equation obtained for 9.97(v/v)%CH4/Ar  Table C.3. Calibration equations obtained from the data presented in Tables C.1-C.2 Gas Calibration equation R2 CH4/Ar Flow rate=0.7949SP 0.997 He Flow rate=0.2673SP 0.999 O2 Flow rate=2.525SP 0.995 Ar Flow rate=7.112SP 0.996   C.2 MS Calibration  To calculate CH4 conversion, the change in the concentration of CH4 and CO2 during the reaction needs to be measured over time. Therefore, the MS is calibrated for CH4 and CO2. For both TPO and TOS experiments the signal intensities of CH4 and CO2 based on the mass signal of He are recorded. 186  Table C.4 presents the change in         ratio as a function of         molar fraction in a constant flow rate of He and O2 at 185 sccm and 60 sccm, respectively.  Table C.4. MS calibration for 9.97%CH4/He using 185 sccm He and 50 sccm O2 IHe ICH4 ICH4/IHe YCH4/ YHe Torr Torr     1.0110-4 3.0010-7 0.0033 0.000000 5.6610-5 1.8010-6 0.0318 0.002371 5.9410-5 2.1710-6 0.0356 0.003240 5.8310-5 2.8010-6 0.0480 0.004145 5.4610-5 3.2410-6 0.0593 0.005280 5.5010-5 3.8010-6 0.0691 0.006582 5.8610-5 5.1010-6 0.0870 0.008223  Figure C.2 presents the calibration equation for         as a function of         molar ratio. 187  0.000 0.002 0.004 0.006 0.0080.000.020.040.060.080.10I CH4/IHeYCH4/YHeICH4/IHe=10.88*(YCH4/YHe)                R2=0.996   Figure C.2. MS Calibration equation for CH4  C.3 Liquid Pump Calibration  In a wet-TOS experiment with 5vol.% H2O in the feed (with the total inlet flow rate of 300 sccm) , the flow rate of H2O is 15 sccm.                                                                                          In order to obtain 15 cm3(STP).min-1 in gas feed, 0.013 cm3.min-1 of H2O in liquid phase in ambient temperature needs to be injected to the system.  Table C.5 shows the calibration for Harvard Apparatus Syringe Pump (Model 44) at 25°C. 188  Table C.5. Harvard apparatus syringe pump (Model 44) calibration  Infuse Rate H2O Collected  Time Flow rate cc.min-1 cc min cc.min-1 0.002 1.1 720 0.0015 0.005 2.8 720 0.0039 0.010 7.0 720 0.0097 0.020 13.8  720 0.0190    189  Appendix D: Error Analysis   In order to confirm the repeatability of the characterization, the standard deviation (SD) is reported for BET, CO chemisorption, XPS, and XRD analyses. The standard deviation is calculated in Equation D.1.                 -                                                                                                                            D.1     is the summation of the squares of each variable (yj) and the average ( ):         -                                                                                                                           D.2 n is the number of repeats and k is the number of    .  Table D.1. Catalyst preparation repeatability Sample Pd Content wt.% 6.5Pd/Al2O3-Batch#1 6.49 6.5Pd/Al2O3-Batch#2 6.97 Average 6.73 St1 0.11 co-2.9Ce/6.5Pd/Al2O3-Batch#1 6.68 co-2.9Ce/6.5Pd/Al2O3-Batch#1 6.90 Average 6.79 St2 0.02 SD 0.26      190  Table D.2. BET analysis repeatability Sample Surface Area Pore Size Pore Volume m2.g-1 nm cm3.g-1 3.4Pd/Al2O3-Run#1 216 8.3 0.45 3.4Pd/Al2O3-Run#2 225 8.4 0.47 Average 220 8.4 0.46 St1 46.2 1.810-4 2.2310-4 9.5Ce/Al2O3-Run#1 191 8.5 0.41 9.5Ce/Al2O3-Run#2 191 8.5 0.41 Average 191 8.5 0.41 St2 0.2 3.910-4 1.2510-8 SD 4.8 1.710-2 1.110-2  Table D.3. CO Uptake analysis repeatability Sample CO Uptake Pd Dispersion μmol.gcat-1 % 6.5Pd/Al2O3-Run#1 204 33.5 6.5Pd/Al2O3-Run#2 198 32.4 Average 201 32.9 St1 18 0.6 co-9.5Ce/6.5Pd/Al2O3-Run#1 228 36.5 co-9.5Ce/6.5Pd/Al2O3-Run#2 242 38.1 Average 235 37.3 St2 98 1.3 SD 7.6 0.9             191  Table D.4. XPS analysis repeatability Sample Pd Ce O Al at.% at.% at.% at.% 3.4Pd/Al2O3-Run#1 0.41 - 62.25 37.35 3.4Pd/Al2O3-Run#2 0.37 - 63.65 35.97 Average 0.39 - 62.95 36.66 St1 4.9410-4 - 0.99 0.94 seq-5Ce/3.4Pd/Al2O3-Run#1 1.02 0.13 61.98 36.88 seq-5Ce/3.4Pd/Al2O3-Run#2 1.08 - 63.30 35.62 Average 1.05 0.13 62.64 36.25 St2 0.002 - 0.87 0.78 seq-17Ce/3.4Pd/Al2O3-Run#1 1.31 0.51 62.00 36.17 seq-17Ce/3.4Pd/Al2O3-Run#2 1.26 0.46 62.13 36.16 Average 1.28 0.49 62.07 36.17 St3 0.001 0.002 0.01 0.00 co-14Ce/3.4Pd/Al2O3-Run#1 1.78 0.85 64.89 32.48 co-14Ce/3.4Pd/Al2O3-Run#2 1.08 0.69 63.06 35.16 Average 1.43 0.77 63.98 33.82 St4 0.25 0.01 1.67 3.60 6.5Pd/Al2O3-Run#1 0.99 - 61.63 37.38 6.5Pd/Al2O3-Run#2 1.04 - 63.48 35.48 6.5Pd/Al2O3-Run#3 0.85 - 61.84 37.30 Average 0.96 - 62.32 36.72 St5 0.02 - 2.04 2.30 SD 0.21 0.07 0.96 1.23  Table D.5. XRD analysis repeatability Sample Obs. Max d (@ Obs. Max) FWHM Crystallite Size (nm) 6.5Pd/Al2O3- Run#1 39.50 2.65 1.92 5.68 6.5Pd/Al2O3- Run#2 39.50 2.65 1.76 6.18 6.5Pd/Al2O3- Run#3 39.43 2.65 1.91 5.71 Average 39.48 2.65 1.86 5.86 SD 0.04 0.00 0.09 0.28 Error (%) 0.10 0.09 4.69 4.85     192  Appendix E: Reaction System  E.1 CH4 Conversion Calculation   In order to measure the initial activity and loss of activity CH4 conversion needs to be calculated during both TPO and TOS experiments. To calculate the CH4 conversion, the total mole balance of carbon, e.g. (         ) is calculated. The mass signal intensities (  ) are monitored by a quadrupole mass spectrometer showing the relative intensities of both CH4 and CO2 to the He intensity. A constant He flow rate (20 cc(STP).min-1) was fed to the system so the change in signal intensities of CH4 and CO2 during the reaction can be recorder based on the He signal intensity. Prior to the reaction, the reactant gas feed consists of CH4, O2, and He passes through the catalyst bed and the MS in order to measure the relative signal intensities of the reactants at the beginning of the reaction. However, a small amount of CO2 is observed by MS, indicating the CO2 from the environment. Therefore, the exact amount of CO2 as a product of the reaction is obtained by subtracting the CO2 of the environment from the measured CO2 at each specific time by MS. For this purpose, the mass signal intensities are recorded for 15 min at ambient temperature and the relative intensities are recorded.           , i=CH4 or CO2                                                                                                        E.1 During the TPO reaction, the change in signal intensities of CH4 and CO2 are recorded every 40 s to be consistent with the reaction temperature recording with the same time step. For the TOS experiments, with a constant reaction temperature, the change in signal intensities of each component are recorded every 40 s for 24h.  193  As shown in Appendix C.2, the relative flow rate of each component is calculated using the calibration equation presented in Figure C.2.                                                                                                                                        E.2                                                                                                                                          E.3 The flow rate of component i (e.g. CH4 or CO2) is calculated considering the He flow rate,    , remains unchanged during the reaction. Corrected      is calculated as below:                                                                                                                           E.4 where       is the CO2 flow rate at ambient temperature. To measure the CH4 conversion, total carbon mole balance should be calculated.                                                                                                                             E.5 As the reaction stoichiometry of complete oxidation of CH4 shows, the volume basis corresponds to the mole basis, so the CH4 conversion is calculated as below:                                                                                                                    E.6                                                                                                                                     E.7 Details of CH4 conversion calculations in a TPO experiment for 6.5Pd/Al2O3 catalyst using 0.1 g catalyst with 5°C.min-1 from ambient temperature to 450°C are reported in Table E.1.  194  Table E.1. CH4 conversion calculation for 6.5Pd/Al2O3 catalyst during TPO experiment. Reaction condition: GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar T IHe ICH4 ICO2 ICH4/IHe ICO2/IHe YCH4/YHe YCO2/YHe He flow CH4 flow CO2 flow              X °C Torr Torr Torr cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 mol.% 60.1 6.85E-09 4.77E-10 8.09E-11 6.96E-02 -2.12E-04 1.50E-02 -2.48E-05 20 3.01E-01 -4.96E-04 0.3 -0.17 64.9 6.84E-09 4.76E-10 8.07E-11 6.96E-02 -2.17E-04 1.50E-02 -2.55E-05 20 3.00E-01 -5.09E-04 0.3 -0.17 71.1 6.81E-09 4.72E-10 8.07E-11 6.92E-02 -1.87E-04 1.49E-02 -2.19E-05 20 2.99E-01 -4.38E-04 0.3 -0.15 75.3 6.84E-09 4.72E-10 8.18E-11 6.90E-02 -7.22E-05 1.49E-02 -8.46E-06 20 2.98E-01 -1.69E-04 0.3 -0.06 79.6 6.81E-09 4.68E-10 8.06E-11 6.87E-02 -2.03E-04 1.48E-02 -2.38E-05 20 2.96E-01 -4.76E-04 0.3 -0.16 83.9 6.82E-09 4.67E-10 8.16E-11 6.84E-02 -6.12E-05 1.48E-02 -7.17E-06 20 2.95E-01 -1.43E-04 0.3 -0.05 88.3 6.75E-09 4.62E-10 8.11E-11 6.85E-02 -1.66E-05 1.48E-02 -1.95E-06 20 2.96E-01 -3.90E-05 0.3 -0.01 92.5 6.73E-09 4.60E-10 8.05E-11 6.84E-02 -6.96E-05 1.48E-02 -8.16E-06 20 2.95E-01 -1.63E-04 0.3 -0.06 96.5 6.71E-09 4.55E-10 7.99E-11 6.79E-02 -1.10E-04 1.46E-02 -1.29E-05 20 2.93E-01 -2.58E-04 0.29 -0.09 100.4 6.75E-09 4.56E-10 8.04E-11 6.76E-02 -1.19E-04 1.46E-02 -1.39E-05 20 2.92E-01 -2.78E-04 0.29 -0.1 104 6.76E-09 4.55E-10 7.98E-11 6.72E-02 -2.21E-04 1.45E-02 -2.59E-05 20 2.90E-01 -5.18E-04 0.29 -0.18 107.5 6.70E-09 4.51E-10 8.03E-11 6.74E-02 -3.72E-05 1.45E-02 -4.36E-06 20 2.91E-01 -8.72E-05 0.29 -0.03 110.8 6.69E-09 4.50E-10 7.94E-11 6.72E-02 -1.67E-04 1.45E-02 -1.95E-05 20 2.90E-01 -3.91E-04 0.29 -0.13 114 6.66E-09 4.45E-10 7.97E-11 6.69E-02 -5.18E-05 1.44E-02 -6.07E-06 20 2.89E-01 -1.21E-04 0.29 -0.04 117 6.69E-09 4.47E-10 7.91E-11 6.67E-02 -2.04E-04 1.44E-02 -2.39E-05 20 2.88E-01 -4.77E-04 0.29 -0.17 119.9 6.63E-09 4.43E-10 7.99E-11 6.67E-02 2.30E-05 1.44E-02 2.69E-06 20 2.88E-01 5.38E-05 0.29 0.02 122.7 6.66E-09 4.42E-10 8.00E-11 6.63E-02 -1.07E-06 1.43E-02 -1.26E-07 20 2.86E-01 -2.52E-06 0.29 0 125.5 6.64E-09 4.38E-10 8.14E-11 6.61E-02 2.46E-04 1.43E-02 2.88E-05 20 2.85E-01 5.76E-04 0.29 0.2 128.2 6.68E-09 4.36E-10 8.22E-11 6.53E-02 2.72E-04 1.41E-02 3.18E-05 20 2.82E-01 6.37E-04 0.28 0.23 130.9 6.60E-09 4.32E-10 8.38E-11 6.55E-02 6.69E-04 1.41E-02 7.84E-05 20 2.83E-01 1.57E-03 0.28 0.55 133.6 6.58E-09 4.32E-10 8.67E-11 6.56E-02 1.15E-03 1.42E-02 1.35E-04 20 2.83E-01 2.69E-03 0.29 0.94               195  T IHe ICH4 ICO2 ICH4/IHe ICO2/IHe YCH4/YHe YCO2/YHe He flow CH4 flow CO2 flow              X °C Torr Torr Torr cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 mol.% 136.4 6.60E-09 4.34E-10 8.88E-11 6.57E-02 1.43E-03 1.42E-02 1.68E-04 20 2.84E-01 3.35E-03 0.29 1.17 139.2 6.60E-09 4.30E-10 8.88E-11 6.51E-02 1.43E-03 1.40E-02 1.68E-04 20 2.81E-01 3.35E-03 0.28 1.18 142.1 6.58E-09 4.27E-10 8.86E-11 6.49E-02 1.43E-03 1.40E-02 1.68E-04 20 2.80E-01 3.36E-03 0.28 1.19 145 6.52E-09 4.25E-10 8.88E-11 6.51E-02 1.59E-03 1.41E-02 1.86E-04 20 2.81E-01 3.72E-03 0.28 1.3 148 6.53E-09 4.23E-10 8.96E-11 6.47E-02 1.69E-03 1.40E-02 1.98E-04 20 2.79E-01 3.96E-03 0.28 1.4 151.1 6.49E-09 4.20E-10 9.05E-11 6.46E-02 1.91E-03 1.39E-02 2.24E-04 20 2.79E-01 4.47E-03 0.28 1.58 154.1 6.48E-09 4.19E-10 9.18E-11 6.47E-02 2.14E-03 1.40E-02 2.51E-04 20 2.79E-01 5.02E-03 0.28 1.77 157.2 6.49E-09 4.15E-10 9.43E-11 6.39E-02 2.50E-03 1.38E-02 2.93E-04 20 2.76E-01 5.86E-03 0.28 2.08 160.3 6.50E-09 4.14E-10 9.61E-11 6.37E-02 2.74E-03 1.37E-02 3.21E-04 20 2.75E-01 6.43E-03 0.28 2.29 163.4 6.54E-09 4.14E-10 9.92E-11 6.33E-02 3.13E-03 1.37E-02 3.67E-04 20 2.73E-01 7.34E-03 0.28 2.62 166.5 6.50E-09 4.08E-10 1.03E-10 6.27E-02 3.83E-03 1.35E-02 4.49E-04 20 2.71E-01 8.99E-03 0.28 3.21 169.6 6.46E-09 4.02E-10 1.05E-10 6.23E-02 4.27E-03 1.34E-02 5.01E-04 20 2.69E-01 1.00E-02 0.28 3.59 172.7 6.42E-09 4.01E-10 1.09E-10 6.25E-02 5.03E-03 1.35E-02 5.89E-04 20 2.70E-01 1.18E-02 0.28 4.19 175.7 6.42E-09 3.97E-10 1.12E-10 6.18E-02 5.41E-03 1.33E-02 6.34E-04 20 2.67E-01 1.27E-02 0.28 4.54 178.8 6.42E-09 3.92E-10 1.15E-10 6.11E-02 5.86E-03 1.32E-02 6.87E-04 20 2.64E-01 1.37E-02 0.28 4.95 181.9 6.39E-09 3.89E-10 1.14E-10 6.09E-02 5.88E-03 1.31E-02 6.89E-04 20 2.63E-01 1.38E-02 0.28 4.98 184.9 6.44E-09 3.91E-10 1.18E-10 6.07E-02 6.28E-03 1.31E-02 7.36E-04 20 2.62E-01 1.47E-02 0.28 5.32 187.9 6.40E-09 3.85E-10 1.20E-10 6.02E-02 6.73E-03 1.30E-02 7.88E-04 20 2.60E-01 1.58E-02 0.28 5.72 191 6.39E-09 3.84E-10 1.22E-10 6.01E-02 7.12E-03 1.30E-02 8.34E-04 20 2.59E-01 1.67E-02 0.28 6.05 194.1 6.43E-09 3.82E-10 1.28E-10 5.94E-02 7.94E-03 1.28E-02 9.31E-04 20 2.56E-01 1.86E-02 0.27 6.77 197.2 6.44E-09 3.78E-10 1.34E-10 5.87E-02 8.85E-03 1.27E-02 1.04E-03 20 2.54E-01 2.08E-02 0.27 7.57 200.2 6.43E-09 3.72E-10 1.42E-10 5.79E-02 1.01E-02 1.25E-02 1.18E-03 20 2.50E-01 2.37E-02 0.27 8.64 203.3 6.42E-09 3.64E-10 1.51E-10 5.67E-02 1.15E-02 1.22E-02 1.35E-03 20 2.45E-01 2.70E-02 0.27 9.93 206.3 6.36E-09 3.56E-10 1.61E-10 5.59E-02 1.33E-02 1.21E-02 1.55E-03 20 2.41E-01 3.11E-02 0.27 11.4 209.4 6.38E-09 3.49E-10 1.74E-10 5.48E-02 1.53E-02 1.18E-02 1.79E-03 20 2.37E-01 3.58E-02 0.27 13.16              196  T IHe ICH4 ICO2 ICH4/IHe ICO2/IHe YCH4/YHe YCO2/YHe He flow CH4 flow CO2 flow              X °C Torr Torr Torr cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 mol.% 212.5 6.40E-09 3.39E-10 1.87E-10 5.30E-02 1.72E-02 1.14E-02 2.01E-03 20 2.29E-01 4.03E-02 0.27 14.96 215.5 6.39E-09 3.31E-10 2.01E-10 5.17E-02 1.94E-02 1.12E-02 2.27E-03 20 2.23E-01 4.54E-02 0.27 16.9 218.5 6.42E-09 3.24E-10 2.17E-10 5.05E-02 2.18E-02 1.09E-02 2.55E-03 20 2.18E-01 5.10E-02 0.27 18.96 221.5 6.39E-09 3.16E-10 2.32E-10 4.94E-02 2.42E-02 1.07E-02 2.84E-03 20 2.13E-01 5.68E-02 0.27 21.04 224.4 6.35E-09 3.05E-10 2.47E-10 4.80E-02 2.69E-02 1.04E-02 3.15E-03 20 2.07E-01 6.30E-02 0.27 23.31 227.3 6.36E-09 2.94E-10 2.64E-10 4.63E-02 2.95E-02 9.99E-03 3.46E-03 20 2.00E-01 6.93E-02 0.27 25.75 230.3 6.34E-09 2.82E-10 2.82E-10 4.44E-02 3.24E-02 9.59E-03 3.80E-03 20 1.92E-01 7.60E-02 0.27 28.37 233.2 6.33E-09 2.68E-10 3.03E-10 4.23E-02 3.58E-02 9.13E-03 4.20E-03 20 1.83E-01 8.40E-02 0.27 31.5 236.2 6.37E-09 2.58E-10 3.25E-10 4.05E-02 3.90E-02 8.74E-03 4.57E-03 20 1.75E-01 9.15E-02 0.27 34.35 239.2 6.35E-09 2.45E-10 3.47E-10 3.86E-02 4.26E-02 8.32E-03 4.99E-03 20 1.66E-01 9.98E-02 0.27 37.47 242.2 6.34E-09 2.29E-10 3.72E-10 3.62E-02 4.66E-02 7.80E-03 5.46E-03 20 1.56E-01 1.09E-01 0.27 41.18 245.2 6.36E-09 2.17E-10 3.98E-10 3.41E-02 5.06E-02 7.36E-03 5.93E-03 20 1.47E-01 1.19E-01 0.27 44.6 248.3 6.37E-09 2.00E-10 4.24E-10 3.14E-02 5.46E-02 6.78E-03 6.40E-03 20 1.36E-01 1.28E-01 0.26 48.57 251.4 6.33E-09 1.87E-10 4.51E-10 2.95E-02 5.92E-02 6.37E-03 6.94E-03 20 1.27E-01 1.39E-01 0.27 52.17 254.4 6.33E-09 1.68E-10 4.78E-10 2.66E-02 6.34E-02 5.73E-03 7.44E-03 20 1.15E-01 1.49E-01 0.26 56.48 257.4 6.39E-09 1.55E-10 5.06E-10 2.42E-02 6.71E-02 5.23E-03 7.86E-03 20 1.05E-01 1.57E-01 0.26 60.05 260.4 6.37E-09 1.37E-10 5.30E-10 2.15E-02 7.12E-02 4.64E-03 8.35E-03 20 9.28E-02 1.67E-01 0.26 64.28 263.4 6.38E-09 1.20E-10 5.59E-10 1.89E-02 7.56E-02 4.07E-03 8.86E-03 20 8.14E-02 1.77E-01 0.26 68.53 266.4 6.39E-09 1.07E-10 5.86E-10 1.67E-02 7.96E-02 3.60E-03 9.33E-03 20 7.20E-02 1.87E-01 0.26 72.15 269.5 6.38E-09 9.00E-11 6.09E-10 1.41E-02 8.34E-02 3.04E-03 9.78E-03 20 6.09E-02 1.96E-01 0.26 76.26 272.5 6.41E-09 7.77E-11 6.37E-10 1.21E-02 8.73E-02 2.61E-03 1.02E-02 20 5.23E-02 2.05E-01 0.26 79.64 275.6 6.37E-09 5.53E-11 6.58E-10 8.67E-03 9.12E-02 1.87E-03 1.07E-02 20 3.74E-02 2.14E-01 0.25 85.1 278.7 6.40E-09 4.47E-11 6.79E-10 6.99E-03 9.41E-02 1.51E-03 1.10E-02 20 3.02E-02 2.20E-01 0.25 87.97 281.7 6.39E-09 2.96E-11 6.99E-10 4.62E-03 9.73E-02 9.97E-04 1.14E-02 20 1.99E-02 2.28E-01 0.25 91.96 284.7 6.39E-09 2.17E-11 7.14E-10 3.39E-03 9.97E-02 7.32E-04 1.17E-02 20 1.46E-02 2.34E-01 0.25 94.1              197  T IHe ICH4 ICO2 ICH4/IHe ICO2/IHe YCH4/YHe YCO2/YHe He flow CH4 flow CO2 flow              X °C Torr Torr Torr cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 mol.% 287.7 6.38E-09 1.11E-11 7.27E-10 1.74E-03 1.02E-01 3.74E-04 1.19E-02 20 7.49E-03 2.39E-01 0.25 96.96 290.6 6.37E-09 2.98E-12 7.40E-10 4.67E-04 1.04E-01 1.01E-04 1.22E-02 20 2.02E-03 2.44E-01 0.25 99.18 293.6 6.41E-09 -2.43E-12 7.50E-10 -3.79E-04 1.05E-01 -8.17E-05 1.23E-02 20 -1.63E-03 2.46E-01 0.24 100.67 296.5 6.38E-09 -8.01E-12 7.56E-10 -1.26E-03 1.07E-01 -2.71E-04 1.25E-02 20 -5.42E-03 2.50E-01 0.24 102.22 299.5 6.37E-09 -1.16E-11 7.62E-10 -1.82E-03 1.07E-01 -3.93E-04 1.26E-02 20 -7.86E-03 2.52E-01 0.24 103.22 302.5 6.34E-09 -1.57E-11 7.63E-10 -2.47E-03 1.08E-01 -5.32E-04 1.27E-02 20 -1.06E-02 2.54E-01 0.24 104.38 305.4 6.35E-09 -1.46E-11 7.70E-10 -2.30E-03 1.09E-01 -4.97E-04 1.28E-02 20 -9.94E-03 2.56E-01 0.25 104.04 308.4 6.34E-09 -1.57E-11 7.71E-10 -2.48E-03 1.09E-01 -5.35E-04 1.28E-02 20 -1.07E-02 2.57E-01 0.25 104.35 311.4 6.37E-09 -1.77E-11 7.69E-10 -2.78E-03 1.09E-01 -6.00E-04 1.27E-02 20 -1.20E-02 2.55E-01 0.24 104.94 314.4 6.30E-09 -1.97E-11 7.67E-10 -3.12E-03 1.10E-01 -6.73E-04 1.28E-02 20 -1.35E-02 2.57E-01 0.24 105.53 317.4 6.31E-09 -2.05E-11 7.69E-10 -3.25E-03 1.10E-01 -7.01E-04 1.29E-02 20 -1.40E-02 2.58E-01 0.24 105.76 320.3 6.31E-09 -1.83E-11 7.67E-10 -2.90E-03 1.10E-01 -6.26E-04 1.28E-02 20 -1.25E-02 2.57E-01 0.24 105.12 323.3 6.29E-09 -1.90E-11 7.68E-10 -3.02E-03 1.10E-01 -6.51E-04 1.29E-02 20 -1.30E-02 2.58E-01 0.25 105.31 326.4 6.26E-09 -2.32E-11 7.68E-10 -3.71E-03 1.11E-01 -8.00E-04 1.30E-02 20 -1.60E-02 2.59E-01 0.24 106.58 329.4 6.27E-09 -2.42E-11 7.64E-10 -3.85E-03 1.10E-01 -8.31E-04 1.29E-02 20 -1.66E-02 2.57E-01 0.24 106.91 332.5 6.25E-09 -2.39E-11 7.65E-10 -3.82E-03 1.10E-01 -8.24E-04 1.29E-02 20 -1.65E-02 2.59E-01 0.24 106.81 335.5 6.26E-09 -2.29E-11 7.67E-10 -3.66E-03 1.11E-01 -7.89E-04 1.30E-02 20 -1.58E-02 2.59E-01 0.24 106.48 338.6 6.23E-09 -1.97E-11 7.66E-10 -3.17E-03 1.11E-01 -6.84E-04 1.30E-02 20 -1.37E-02 2.60E-01 0.25 105.55 341.7 6.26E-09 -1.99E-11 7.67E-10 -3.19E-03 1.11E-01 -6.87E-04 1.30E-02 20 -1.37E-02 2.59E-01 0.25 105.61 344.6 6.27E-09 -2.37E-11 7.63E-10 -3.79E-03 1.10E-01 -8.17E-04 1.29E-02 20 -1.63E-02 2.57E-01 0.24 106.79 347.6 6.25E-09 -2.51E-11 7.61E-10 -4.02E-03 1.10E-01 -8.67E-04 1.28E-02 20 -1.73E-02 2.57E-01 0.24 107.24 350.6 6.22E-09 -2.04E-11 7.60E-10 -3.28E-03 1.10E-01 -7.08E-04 1.29E-02 20 -1.42E-02 2.58E-01 0.24 105.8 353.7 6.24E-09 -2.10E-11 7.59E-10 -3.36E-03 1.10E-01 -7.25E-04 1.28E-02 20 -1.45E-02 2.57E-01 0.24 105.98 356.7 6.20E-09 -2.16E-11 7.60E-10 -3.48E-03 1.11E-01 -7.51E-04 1.30E-02 20 -1.50E-02 2.59E-01 0.24 106.15 359.8 6.15E-09 -2.51E-11 7.57E-10 -4.08E-03 1.11E-01 -8.81E-04 1.30E-02 20 -1.76E-02 2.60E-01 0.24 107.26              198  T IHe ICH4 ICO2 ICH4/IHe ICO2/IHe YCH4/YHe YCO2/YHe He flow CH4 flow CO2 flow              X °C Torr Torr Torr cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 mol.% 362.9 6.16E-09 -2.47E-11 7.56E-10 -4.01E-03 1.11E-01 -8.65E-04 1.30E-02 20 -1.73E-02 2.59E-01 0.24 107.15 366 6.17E-09 -1.94E-11 7.58E-10 -3.15E-03 1.11E-01 -6.80E-04 1.30E-02 20 -1.36E-02 2.60E-01 0.25 105.52 368.9 6.16E-09 -2.09E-11 7.59E-10 -3.39E-03 1.11E-01 -7.32E-04 1.30E-02 20 -1.46E-02 2.60E-01 0.25 105.96 371.9 6.18E-09 -2.27E-11 7.56E-10 -3.68E-03 1.10E-01 -7.94E-04 1.29E-02 20 -1.59E-02 2.59E-01 0.24 106.54 374.9 6.18E-09 -2.20E-11 7.58E-10 -3.56E-03 1.11E-01 -7.67E-04 1.30E-02 20 -1.53E-02 2.59E-01 0.24 106.3 377.8 6.22E-09 -1.97E-11 7.57E-10 -3.17E-03 1.10E-01 -6.84E-04 1.29E-02 20 -1.37E-02 2.57E-01 0.24 105.62 380.9 6.14E-09 -2.26E-11 7.57E-10 -3.68E-03 1.11E-01 -7.94E-04 1.30E-02 20 -1.59E-02 2.61E-01 0.25 106.48 383.9 6.14E-09 -2.04E-11 7.53E-10 -3.33E-03 1.11E-01 -7.18E-04 1.30E-02 20 -1.44E-02 2.59E-01 0.24 105.86 386.9 6.12E-09 -2.42E-11 7.54E-10 -3.96E-03 1.11E-01 -8.54E-04 1.30E-02 20 -1.71E-02 2.60E-01 0.24 107.02 390 6.13E-09 -2.06E-11 7.52E-10 -3.36E-03 1.11E-01 -7.25E-04 1.30E-02 20 -1.45E-02 2.59E-01 0.24 105.93 393 6.13E-09 -2.34E-11 7.55E-10 -3.81E-03 1.11E-01 -8.23E-04 1.30E-02 20 -1.65E-02 2.60E-01 0.24 106.75 396 6.15E-09 -2.08E-11 7.54E-10 -3.38E-03 1.11E-01 -7.30E-04 1.30E-02 20 -1.46E-02 2.59E-01 0.24 105.96 398.9 6.16E-09 -2.41E-11 7.53E-10 -3.92E-03 1.10E-01 -8.45E-04 1.29E-02 20 -1.69E-02 2.58E-01 0.24 107.01 401.8 6.12E-09 -2.01E-11 7.53E-10 -3.28E-03 1.11E-01 -7.07E-04 1.30E-02 20 -1.41E-02 2.60E-01 0.25 105.75 404.9 6.13E-09 -2.00E-11 7.51E-10 -3.26E-03 1.10E-01 -7.03E-04 1.29E-02 20 -1.41E-02 2.59E-01 0.24 105.75 407.9 6.13E-09 -2.08E-11 7.53E-10 -3.40E-03 1.11E-01 -7.33E-04 1.30E-02 20 -1.47E-02 2.60E-01 0.25 105.97 410.9 6.09E-09 -2.57E-11 7.49E-10 -4.22E-03 1.11E-01 -9.10E-04 1.30E-02 20 -1.82E-02 2.60E-01 0.24 107.53 414 6.08E-09 -2.47E-11 7.51E-10 -4.07E-03 1.11E-01 -8.78E-04 1.31E-02 20 -1.76E-02 2.61E-01 0.24 107.2 417 6.10E-09 -1.95E-11 7.49E-10 -3.20E-03 1.11E-01 -6.90E-04 1.30E-02 20 -1.38E-02 2.60E-01 0.25 105.61 420 6.09E-09 -2.41E-11 7.51E-10 -3.95E-03 1.11E-01 -8.52E-04 1.30E-02 20 -1.70E-02 2.61E-01 0.24 107 422.9 6.15E-09 -1.98E-11 7.50E-10 -3.22E-03 1.10E-01 -6.94E-04 1.29E-02 20 -1.39E-02 2.58E-01 0.24 105.7 425.9 6.07E-09 -2.48E-11 7.50E-10 -4.09E-03 1.11E-01 -8.82E-04 1.31E-02 20 -1.76E-02 2.61E-01 0.24 107.24 428.9 6.10E-09 -2.25E-11 7.52E-10 -3.68E-03 1.11E-01 -7.95E-04 1.30E-02 20 -1.59E-02 2.61E-01 0.24 106.5 432 6.09E-09 -2.27E-11 7.51E-10 -3.73E-03 1.11E-01 -8.04E-04 1.30E-02 20 -1.61E-02 2.61E-01 0.24 106.57 435.1 6.07E-09 -2.15E-11 7.50E-10 -3.54E-03 1.12E-01 -7.63E-04 1.31E-02 20 -1.53E-02 2.61E-01 0.25 106.2              199  T IHe ICH4 ICO2 ICH4/IHe ICO2/IHe YCH4/YHe YCO2/YHe He flow CH4 flow CO2 flow              X °C Torr Torr Torr cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 cc(STP).min-1 mol.% 438.1 6.02E-09 -2.55E-11 7.49E-10 -4.23E-03 1.12E-01 -9.13E-04 1.32E-02 20 -1.83E-02 2.64E-01 0.25 107.45 441.1 6.06E-09 -2.26E-11 7.47E-10 -3.73E-03 1.11E-01 -8.05E-04 1.30E-02 20 -1.61E-02 2.61E-01 0.24 106.59 444.1 6.05E-09 -2.11E-11 7.49E-10 -3.50E-03 1.12E-01 -7.55E-04 1.31E-02 20 -1.51E-02 2.62E-01 0.25 106.11 446.9 6.06E-09 -2.14E-11 7.48E-10 -3.53E-03 1.11E-01 -7.62E-04 1.31E-02 20 -1.52E-02 2.61E-01 0.25 106.2 450 6.03E-09 -2.15E-11 7.45E-10 -3.56E-03 1.12E-01 -7.69E-04 1.31E-02 20 -1.54E-02 2.62E-01 0.25 106.25 200  Appendix F: Repeatability  F.1 TPO Reaction Repeatability  Table F.1. TPO Reaction Repeatability. GHSV=180,000 cm3(STP).gcat-1.h-1. 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar Sample Light-off Temperature (°C) T10 T50 T90 3.4Pd/Al2O3-Run#1 223 278 317 3.4Pd/Al2O3-Run#2 214 268 301 Average 219 273 309 St1 45 45 125 seq-2Ce/3.4Pd/Al2O3-Run#1 222 273 309 seq-2Ce/3.4Pd/Al2O3-Run#2 207 271 307 Average 214 272 308 St2 110 3 3 seq-17Ce/3.4Pd/Al2O3-Run#1 207 262 295 seq-17Ce/3.4Pd/Al2O3-Run#2 226 271 299 seq-17Ce/3.4Pd/Al2O3-Run#3 212 260 294 Average 215 264 296 St3 10 6 3 seq-28Ce/3.4Pd/Al2O3-Run#1 226 283 322 seq-28Ce/3.4Pd/Al2O3-Run#2 221 278 315 Average 223 281 318 St4 11 11 25 SD 6 4 6       201  F.2 TOS Reaction Repeatability  Table F.2. TOS repeatability. GHSV=180,000 cm3(STP).gcat-1.h-1. 5000 ppm CH4, 20(v/v)% O2, and the balance He Sample   TOS (h)  1 6 12 24   Conversion (mol.%) 6.5Pd/Al2O3, dry-TOS, 350°C Run#1 100 95 90 84 Run#2 100 97 93 87 Average  100 96 92 86 St1  0 2 4.5 4.5 6.5Pd/Al2O3, wet-TOS, 350°C Run#1 82 48 45 43 Run#2 78 49 48 46 Average  80 49 47 45 St2  8 0.5 4.5 4.5 co-2.9Ce/6.5Pd/Al2O3, dry-TOS, 350°C Run#1 84 61 53 47 Run#2 87 64 58 51 Average  86 63 56 49 St3  4.5 4.5 12.5 8 co-2.9Ce/6.5Pd/Al2O3, wet-TOS, 380°C Run#1 76 54 53 50 Run#2 80 52 49 48 Average  78 53 51 49 St4  8 2 8 2 SD  2 1 3 2    202  Appendix G: Supplementary Figures and Tables for Chapter 6  The Ce 3d spectra for co-impregnated catalysts are presented in Figure G.1. Because of the low intensity of Ce 3d peaks of the co-2Ce/3.4Pd/Al2O3 catalyst, only the Ce 3d spectra of co-14Ce/3.4Pd/Al2O3 and co-47Ce/3.4Pd/Al2O3 catalysts are shown in Figure G.1 and the B.E.s of Ce 3d for all co-2Ce/3.4Pd/Al2O3, co-14Ce/3.4Pd/Al2O3, and co-47Ce/3.4Pd/Al2O3 catalysts are reported in Table G.2. The two main peaks attributed to 3d5/2 and 3d3/2 electrons for Ce3+ (Ce2O3) are located at 885.0 eV (v') and 903.6 eV (u'), respectively for all three catalysts. The three main peaks assigned to 3d3/2 for Ce4+ (CeO2) occur at 901.0 eV (u), 906.9 eV (u"), and 917.0 eV (u‴) for all three catalysts. There are also three peaks for Ce4+ (CeO2) 3d5/2 identified as v, v", and v"'. For all co-impregnated catalysts, the v and v"' peak appear at 882.9 eV and 898.7 eV, respectively, however, the peak corresponded to v" appears at 887.4 eV for co-2Ce/3.4Pd/Al2O3 and the B.E. increases to 888.1 eV and 888.4 eV for co-14Ce/3.4Pd/Al2O3 and co-47Ce/3.4Pd/Al2O3 catalysts, respectively.  203  930 920 910 900 890 880 870Intensity (a.u.)Binding Energy (eV)v'(b)(a)vv"v"'uu'u"u"' Figure G.1. Ce 3d for co-impregnated catalysts (a) co-14Ce/3.4Pd/Al2O3, (b) co-47Ce/3.4Pd/Al2O3 catalysts  For the xCe/Al2O3 support shown in Figure G.2 the same peaks appear for Ce 3d at the same positions as those for the sequentially impregnated catalysts, except for the v" and u" peaks. The peak for Ce4+ (CeO2) 3d5/2 appears at 887.4 eV (v") for seq-2Ce/Al2O3 support and shifts to 888.7 eV as the Ce loading increases to 52wt.%. Also the peak assigned to Ce4+ (CeO2) 3d3/2 (u") appeared at 906.9 eV for 2Ce/Al2O3 support and moved to 907.8 eV by increasing the Ce loading to 52wt.%.    204  930 920 910 900 890 880 870Intensity (a.u.)(a)Binding Energy (eV)(a) 5Ce+Al2O3 support(b) 15Ce+Al2O3 support(c) 25Ce+Al2O3 support(d) 50Ce+Al2O3 support(b)vv'v"v"'uu'u"u"'(c) Figure G.2. Ce 3d for (a) 16Ce/Al2O3 (b) 26Ce/Al2O3, and (c) 52Ce/Al2O3 supports  205  0.0 0.2 0.4 0.6 0.8 1.002468101214161820Quantity Adsorbed (mmol/g)Relative Pressure (p/p°)co-impregnated(c)(b)(a)   Figure G.3. N2 adsorption-desorption isotherms for co-impregnated (a) co-2Ce/3.4Pd/Al2O3, (b) co-14Ce/3.4Pd/Al2O3, and (c) co-47Ce/3.4Pd/Al2O3 catalysts    206  0.0 0.2 0.4 0.6 0.8 1.0051015202530Quantity Adsorbed (mmol/g)Relative Pressure (p/p°)sequential(e)(d)(c)(b)  (a) Figure G.4. N2 adsorption-desorption isotherms for sequentially impregnated (a) seq-2Ce/3.4Pd/Al2O3, (b) seq-6Ce/3.4Pd/Al2O3, (c) seq-17Ce/3.4Pd/Al2O3, (d) seq-28Ce/3.4Pd/Al2O3, and (e) seq-57Ce/3.4Pd/Al2O3 catalysts 207  0.0 0.2 0.4 0.6 0.8 1.0051015202530Quantity Adsorbed (mmol/g)Relative Pressure (p/p°)(e)(d)(c)(b)(a)   Figure G.5. N2 adsorption-desorption isotherms for (a) 2Ce/Al2O3, (b) 5Ce/Al2O3, (c) 16Ce/Al2O3, (d) 26Ce/Al2O3, and (e) 52Ce/Al2O3 supports   208  0 50 100 150 200 2500.00.20.40.60.81.01.21.40 5 10 15 20 25 300.00.20.40.60.81.01.21.4dV/d(logD) Pore Volume (cm3/g-1.Å-1)Pore Diameter (nm)co-impregnated     Figure G.6. BJH pore size distribution for co-2Ce/3.4Pd/Al2O3 (○), co-14Ce/3.4Pd/Al2O3 (■), and co-47Ce/3.4Pd/Al2O3 (▲) catalysts  209  0 50 100 150 200 2500.00.20.40.60.81.01.21.40 5 10 15 20 25 300.00.20.40.60.81.01.21.4dV/d(logD) Pore Volume (cm3/g-1.Å-1)Pore Diameter (nm)sequential     Figure G.7. BJH pore size distribution for (○) seq-2Ce/3.4Pd/Al2O3, (●) seq-6Ce/3.4Pd/Al2O3, (▲) seq-17Ce/3.4Pd/Al2O3, (▼) seq-28Ce/3.4Pd/Al2O3, and (□) seq-57Ce/3.4Pd/Al2O3 catalysts   210  0 50 100 150 200 2500.00.20.40.60.81.01.21.40 5 10 15 20 25 300.00.20.40.60.81.01.21.4dV/d(logD) Pore Volume (cm3/g-1.Å-1)Pore Diameter (nm)    supports   Figure G.8. BJH pore size distribution for (○) 2Ce/Al2O3, (●) 5Ce/Al2O3, (▲) 16Ce/Al2O3, (▼) 26Ce/Al2O3, and (□) 52Ce/Al2O3 supports  211  0 5 10 15 20 25020406080100(b)CH4 Conversion (mol%)Time on Stream (h)(a)(c)(d)   Figure G.9. Dry-TOS results for 3.4Pd/Al2O3 (a) and (b) co-2Ce/3.4Pd/Al2O3, (c) co-14Ce/3.4Pd/Al2O3, and (d) co-47Ce/3.4Pd/Al2O3 at 350°C. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar 212  0 5 10 15 20 25020406080100CH4 Conversion (mol%)Time on Stream (h)  (a)(b)(c),(d)(e)(f) Figure G.10. Dry-TOS results for 3.4Pd/Al2O3 (a) and (b) seq-2Ce/3.4Pd/Al2O3, (c) seq-6Ce/3.4Pd/Al2O3, (d) seq-17Ce/3.4Pd/Al2O3, (e) seq-28Ce/3.4Pd/Al2O3, and (f) seq-57Ce/3.4Pd/Al2O3 at 350°C. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar   213   0 5 10 15 20 250 5 10 15 20 25020406080100 Time on Stream (h)3.4Pd/Al2O3(c)(b) (a)(d)seq-17Ce/3.4Pd/Al2O3(c)(b)(a)(d)CH4 Conversion (mol.%)Time on Stream (h)   Figure G.11. Wet-TOS results for seq-17Ce/3.4Pd/Al2O3 and 3.4Pd/Al2O3 catalysts with 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar. (a) T=310°C, (b) T=330°C, (c) T=350°C, and (d) T=370°C 214   Figure G.12. ToF-SIMS analysis for 3.4Pd/Al2O3 catalyst      Total Ion 100 um Al Pd 215   Figure G.13. ToF-SIMS analysis for co-14Ce/3.4Pd/Al2O3 catalyst   Total Ion 100 um Al Ce Pd 216  Table G.1. Properties of calcined catalysts prepared by co-impregnation and sequentially impregnation methods Catalyst BET SA Pore Pore Size Volume m²/g nm cm3/g 3.4Pd/Al2O3 215 8.3 0.45 co-2Ce/3.4Pd/Al2O3 218 8.2 0.45 co-14Ce/3.4Pd/Al2O3 170 7.7 0.33 co-47Ce/3.4Pd/Al2O3 93 6.2 0.14 seq-2Ce/3.4Pd/Al2O3 205 8.6 0.44 seq-6Ce/3.4Pd/Al2O3 166 9.3 0.38 seq-17Ce/3.4Pd/Al2O3 143 9.5 0.34 seq-28Ce/3.4Pd/Al2O3 152 7.1 0.27 seq-57Ce/3.4Pd/Al2O3 91 6.3 0.14 2Ce/Al2O3 180 9.8 0.44 5Ce/Al2O3 196 9.4 0.46 16Ce/Al2O3 161 9.0 0.36 26Ce/Al2O3 137 8.7 0.29 52Ce/Al2O3 76 7.8 0.15    217  Table G.2. Light-off temperatures for 3.4Pd/Al2O3, co-impregnated and sequentially impregnated catalysts. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar Catalyst T10 T50 T90 °C °C °C 3.4Pd/Al2O3 219 273 309 co-2Ce/3.4Pd/Al2O3 221 278 314 co-14Ce/3.4Pd/Al2O3 254 334 395 co-47Ce/3.4Pd/Al2O3 263 351 446 seq-2Ce/3.4Pd/Al2O3 214 271 307 seq-6Ce/3.4Pd/Al2O3 225 276 310 seq-17Ce/3.4Pd/Al2O3 215 264 296 seq-28Ce/3.4Pd/Al2O3 223 281 318 seq-57Ce/3.4Pd/Al2O3 227 288 332 218  Table G.3. Ce 3d peaks and               ratio for co-impregnated catalysts with different loadings of Ce Catalyst Ce 3d5/2  Ce 3d3/2                                                          eV eV eV eV  eV eV eV eV  % co-2Ce/3.4Pd/Al2O3 882.9 885.0 887.4 898.7  901.0 903.6 906.9 917.0  37.8 co-14Ce/3.4Pd/Al2O3 882.9 885.0 888.1 898.7  901.0 903.6 906.9 917.0  18.4 co-47Ce/3.4Pd/Al2O3 882.9 885.0 888.4 898.7  901.0 903.6 906.9 917.0  8.5           Table G.4. Ce 3d peaks and               ratio for xCe/Al2O3 supports with different loadings of Ce Catalyst Ce 3d5/2  Ce 3d3/2                                                          eV eV eV eV  eV eV eV eV  % 2Ce/Al2O3 882.9 885.0 887.4 898.7  901.0 903.6 906.9 917.0  32.1 5Ce/Al2O3 882.9 885.0 887.4 898.7  901.0 903.6 906.9 917.0  27.7 16Ce/Al2O3 882.9 885.0 888.3 898.7  901.0 903.6 906.9 917.0  14.4 26Ce/Al2O3 882.9 885.0 888.5 898.7  901.0 903.6 906.9 917.0  12.0 52Ce/Al2O3 882.9 885.0 888.7 898.7  901.0 903.6 907.8 917.0  8.9    219  Appendix H: Mass Transfer Effects  H.1 Internal Mass Transfer Calculation  In order to determine if the kinetic model is controlled by internal mass transfer, the theoretical calculations are presented in this appendix. Table H.1 presents the physical properties of the catalyst bed for 6.5Pd/Al2O3 catalyst.      , the average pore size of the catalyst is calculated as follows:                                                                                                                                         H.1   , particle porosity is the ratio of the catalyst pore volume to the total volume of the catalyst that is defined in Equation H.2.                                                                                                                                        H.2 The bed density,     , is calculated as the mass of catalyst per total volume of the catalyst bed. The bed volume is the total bed of both catalyst and SiC diluent.                                                                                                                                           H.3                                                                                                                                        H.4 m is the total mass of bed as:                                                                                                                               H.5                                                                                                                                        H.6        is defined as the total density of both porous catalyst and non-porous SiC:                                                                                                                                                 H.7  220  Table H.1. Physical properties of catalyst bed consists of 6.5Pd/Al2O3 Parameter Definition Value Wcat Catalyst mass (g) 0.0833 Wsic Mass of SiC (g) 2.1 Lbed Length of bed (cm) 4.2        Internal diameter of bed (cm) 0.703 Vbed Volume of bed (cm3) 1.63 m mass of bed (g) 2.1833 SBET BET surface area (m2.g-1) 2.180105 V0 Total pore volume of catalyst (cm3.g-1) 0.433 dpore Pore size (cm) 7.8910-7 ρs 6.5%Pd/Al2O3 density (g.cm-3)  4.13 ρbed Bed density (g.cm-3) 0.051 ρSiC SiC density (g.cm-3) 3.21 ρcat Catalyst density (g.cm-3) density of both pores of catalyst and solid particles 1.488 ρbSiC Catalyst bed (g.cm-3) density of both catalyst and SiC 1.337 εp Particle porosity 0.64 dp Particle diameter (cm) 2.22010-2  a Tortuosity factor 3 σa Constriction factor 0.8 εbSiC Bed porosity 0.57 a Typical values obtained from [92]  Table H.2 presents the reaction operating condition for TOS experiment using 5000 ppmv CH4, 20(v/v)% O2, and the balance He for GHSV=180,000 cm3(STP).gcat-1.h-1. The effective diffusivity,     , is calculated at T=330°C and P=1 atm.   , the Lennard-Jones characteristic length for component i (i=He or CH4) is defined in Equation H.8 [89].                                                                                                                                      H.8                                                                                                                         H.9                                                                                                                                      H.10                                                                                                                               H.11 221                                                                                                                                      H.12                                                                                                                                    H.13                                                                                                                H.14 where P is in atm, T is in K, and         is in cm2.s-1.                                                                                                                             H.15                                                                                                                               H.16 where T is in K,       in kg.mol-1, and       in m.                                                                                                                                    H.17                                                                                                                                     H.18 Equations H.8-H.18 are presented in [87,89,92].   222  Table H.2. Operating condition for TOS experiment using 6.5Pd/Al2O3 catalyst Parameter   Definition   Value T   Reaction temperature (K) 603 P  Total pressure (Pa) 101325 R  Gas constant (Pa.m3.mol-1.K-1) 8.314 yCH4  CH4 volume fraction (-) 0.005 yO2  O2 volume fraction (-) 0.200 yHe  He volume fraction (-) 0.793 MwHe  He molecular weight (g.mol-1) 4        He critical pressure (kPa) 229          He critical temperature (K) 5.2 MwCH4  CH4 molecular weight (g.mol-1) 16.04         CH4 critical pressure (kPa) 4640         CH4 critical temperature (K) 190.7 Mwfeed  Feed molecular weight (g.mol-1) 9.65             Feed CH4 partial pressure (Pa) 506.62 ν0  Total volumetric flow rate (m3.s-1) at T=330°C 9.20E-06            Feed CH4 molar flow (mol.s-1 (STP)) 9.29E-07 rt  Radius of reactor (m) 3.52E-03 A  Cross sectional area of the reactor (m2) 3.88E-05 V  Superficial gas velocity (m.s-1) at T=330°C 0.237 G  Superficial mass velocity (kg.m-2.s-1) 0.046 (ε/k)He  He Lennard-Jones energy/Boltzmann's constant (-) 4.00 σHe  He Lennard-Jones characteristic length (Å) 3.22 (ε/k)CH4  CH4 Lennard-Jones energy/Boltzmann's constant (-) 146.84 σCH4  CH4 Lennard-Jones characteristic length (Å) 3.93 (ε/k)CH4-He   Lennard-Jones energy/Boltzmann's constant (-) 24.25 σCH4-He  Lennard-Jones characteristic length (Å) 3.57 T*  (-) 24.87 Ω  Collision integral (-) 0.64     Tortuosity factor (-) 3 σ  Constriction factor (-) 0.8           Binary bulk diffusivity (m2.s-1) 1.87E-04              Bulk effective diffusivity (m2.s-1) 3.20E-05      Knudsen diffusivity (m2.s-1) 6.06E-06         Effective Knudsen diffusivity (m2.s-1) 1.03E-06        Effective diffusivity (m2.s-1) 1.00E-06  223  Applying the calculated      value into kinetic equations in Chapters 4 and 5,   values were obtained in the range of 0.24-0.86 emphasizing the internal mass transfer control.  H.2 External Mass Transfer Calculation  Particle Reynolds number is calculated as follows:     ρ     μ   ε                                                                                                                          H.19 where   is the particle diameter, μ is the dynamic gas viscosity, and ε    is the bed porosity. ρ , the gas density and    the superficial gas velocity are defined as below:  ρ                                                                                                                                   H.20    ν                                                                                                                                               H.21 Schmidt number is calculated using Equation H.22.    μρ                                                                                                                             H.22 The external mass transfer coefficient (  ) is calculated using Equation H.23. The Sherwood number is obtained using Equations H.24-H.25.                                                                                                                                       H.23 In a gas phase system with Re < 2000 and 0.416 < ε     < 0.788,    factor is calculated using Equation H.24 [125].   ε                                                                                                                        H.24                                                                                                                                        H.25 Finally, Mears criterion is calculated as follows: 224                                                                                                                                        H.26 The calculated numbers are reported in Table H.3.  Table H.3. Details of calculations for Mears criterion factor for 6.5Pd/Al2O3 catalyst at T=330°C Parameter Definition Value T Reaction temperature (K) 603 P Total pressure (Pa) 101325         Feed molecular weight (g.mol-1) 9.65 R Gas constant (Pa.m3.mol-1.K-1) 8.314 ν0 Total volumetric flow rate (m3.s-1) 9.20E-06 A Cross sectional area of the reactor (m2) 3.88E-05     Particle diameter (m) 2.22E-04 us Superficial gas velocity (m.s-1) at T=330°C 0.237 ε      Bed porosity (-) 0.57          Binary bulk diffusivity (m2.s-1) 1.87E-04 ρ   Gas density (kg.m3) 1.95E-01 μ Gas dynamic viscosity (kg.m-1.s-1) 3.22E-05 Re Reynolds Number (-) 0.74 Sc Schmidt  Number (-) 0.88        factor (-) 0.67 Sh Sherwood Number (-) 0.49     External mass transfer coefficient (m.s-1) 0.417       Bulk concentration of CH4 (mol.m-3) 0.101         CH4 reaction rate (mol.kg-1.s-1) 6.21E-03        Catalyst bed density of both catalyst and SiC (kg.m-3) 1.337E+03 n Order of CH4 oxidation reaction (-) 1     Mears criterion factor (-) 0.02  For    < 0.15 the mass transfer from the bulk gas phase to the surface of the catalyst is negligible. In our study, is obtained as 0.02, indicating no external mass transfer control.    225  H.3 Pressure Drop Calculation over Catalyst Bed         ρ                   ε                 ρ                                                                  H.27 where G and are defined as superficial mass velocity (kg.m-2.s-1) and cross sectional area of the reactor (m2). So,    over the entire catalyst bed is 2.4 kPa.     226  Appendix I: CH4 Oxidation over PdO-ZrOx/Al2O3 in the Presence of H2O   I.1 Catalyst Properties  The Pd content measured by AAS analysis was obtained 3.4wt.% as an average loading for all catalysts. Figure I.1 shows the effect of different Zr loadings on the Pd/Al and Zr/Al surface composition, measured by XPS. As the Zr bulk composition increases (reported as the Zr to Al atom ratio i.e. (Zr/Al)b), the (Zr/Al)s increases. A higher amount of Zr in the catalyst causes more blockages of alumina pores consequently more Zr stays on the surface [126]. To identify the crystal size of ZrO2 at different Zr loadings, a XRD analysis was performed. Almost the same quantity of Pd appears on the surface of 0wt.% and 1.5wt.%Zr-catalysts. Since the Zr loading is very low, not many Al2O3 pores are blocked [127]. Thus the conditions for Pd to enter into the pores or to stay on the surface are very similar in both catalysts. The slight decrease of Pd on the surface (1.5wt%Zr) is maybe due to the coverage through ZrO2, so that not all of the Pd could be detected on the surface. Increasing the Zr ratio to 25wt.% leads to a high dispersion of Pd on the surface (higher (Pd/Al)s). Amaira et al., Souza and Gou argued that the Zr particles block the Al2O3 pores and prevent the penetration of Pd particles into the pores [126,128,129].   227  0 5 10 15 20 250,000,050,100,150,200,250,300,35Zr-loading [wt%](Zr/Al)s [%]0,000,010,020,030,040,050,060,07 (Pd/Al)s [%] Figure I.1. XPS analysis, Zr on the surface (square), Pd on the surface (cross)  (Pd/Al)s decreased as the Zr loading increased from 0wt.% to 15wt.% which is due to some Pd blockage caused by Zr. However, higher loading of Zr (25wt.%) caused the opposite effect that can be explained by blocking Al2O3 pores due to high Zr loading (see explanation before).   Figure I.2 shows the XRD analysis for 3.4Pd/Al2O3 and sequential impregnated catalysts with different loadings of Zr. Major peaks were found at 2θ=35.55° and 59.96° and 71.55° for tetragonal ZrO2 and at 2θ=39.49° and 64.55° for tetragonal PdO, respectively.  228  10 20 30 40 50 60 70 80(d)(c)(b)Intensity (a.u.)2theta (°)(a)   Figure I.2. XRD patterns for 3.4Pd/Al2O3 (a) and sequential impregnated catalysts with different loadings of Zr: (b) seq-1.5Zr/3.4Pd/Al2O3, (c) seq-15Zr/3.4Pd/Al2O3, (d) and seq-25Zr/3.4Pd/Al2O3. ∆ PdO, ● Al2O3, ○ ZrO2  Table I.1 shows the crystal size of PdO and of ZrO2. The PdO crystal size does not change by adding different loadings of Zr as it is measured between 6-7 nm for all catalysts. The ZrO2 crystal size measured for the seq-25Zr/3.4Pd/Al2O3 catalyst that shows ZrO2 crystals are 2.8 bigger than PdO crystals.    229  Table I.1. Properties of calcined 3.4Pd/Al2O3 and sequential impregnated catalysts with different loadings of Zr Catalysts BET Pore Pore PdO ZrO2 SAa Volumea Sizea Crystallite Sizeb Crystallite Sizeb m²/g cm³/g nm nm nm 3.4Pd/Al2O3 215 0.45 8.3 7 - seq-1.5Zr/3.4Pd/Al2O3 192 0.41 8.5 6 - seq-15Zr/3.4Pd/Al2O3 139 0.31 8.6 6 - seq-25Zr/3.4Pd/Al2O3 83 0.18 8.6 7 17 a Determined by N2 adsorption at 77K   b PdO (1 0 1) and ZrO2 (111) obtained by XRD    I.2 Catalyst Activities  Figure I.3 shows the TPO results for the 3.4Pd/Al2O3 and sequential impregnated catalysts with different loadings of Zr. 150 200 250 300 350020406080100 seq-1.5Zr/3.4Pd/Al2O3 seq-15Zr/3.4Pd/Al2O3  seq-25Zr/3.4Pd/Al2O3CH4 Conversion (mol.%)Temperature (°C)   Figure I.3. Temperature Programmed Oxidation profile. Effect of different loadings of Zr on the initial activity of 3.4Pd/Al2O3 as a function of temperature. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar 230  The seq-1.5Zr/3.4Pd/Al2O3 catalyst shows the highest initial activity. Increasing the Zr content resulted in less active catalysts as the loading of Zr increased. The light-off temperatures corresponding to 10%, 50%, and 90% CH4 conversion are shown in Table I.2. Since the standard deviation for T50 for seq-15Zr/3.4Pd/Al2O3, seq-25Zr/3.4Pd/Al2O3, and 3.4Pd/Al2O3 catalysts is between 2.5 and 6.9°C, no significant different between the initial activity of the three catalysts can be observed.  Table I.2. Light-off temperatures for 3.4Pd/Al2O3 and sequential impregnated catalysts with different loadings of Zr Catalyst T10 T50 T90 °C °C °C 3.4Pd/Al2O3 226±5 277±2.5 316.2±2 seq-1.5Zr/3.4Pd/Al2O3 214±5 262.4±5 295.6±5 seq-15Zr/3.4Pd/Al2O3 226±1 274.1±4 310.2±8 seq-25Zr/3.4Pd/Al2O3 221±4 276±7 315.2±7  Effect of H2O concentration on the stability of Zr-supported catalysts in CH4 oxidation was performed using 2vol.% and 5vol.% H2O into the feed stream. Figures I.4 and I.5 show the wet-TOS results at 350°C for 3.4P/Al2O3 and sequential catalysts with different loadings of Zr. 231  0 5 10 15 20 25020406080100CH4 Conversion (mol.%)Time on Stream (h)   Figure I.4. Wet-TOS results for (■) 3.4Pd/Al2O3, (∆) seq-1.5Zr/3.4Pd/Al2O3, () seq-15Zr/3.4Pd/Al2O3, and (○) seq-25Zr/3.4Pd/Al2O3 catalysts at 350 °C with 2vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  All catalysts showed a fast decrease in CH4 conversion once H2O was injected. An exponential decrease followed by a linear decrease was observed for all catalysts. seq-1.5Zr/3.4Pd/Al2O3 catalyst showed the highest activity over 24h compared to seq-15Zr/3.4Pd/Al2O3 and seq-25Zr/3.4Pd/Al2O3 catalysts.   However, the performance of the 3.4%Pd catalyst is very close to 1.5%Zr and after 20h both catalysts showed the same conversion at around 27 mol%. The CH4 conversion for 15%Zr and 25%Zr is much lower than for 1.5%Zr and 3.4%Pd. At t=24h, the difference in CH4 conversion is about 15mol% and 11mol%, respectively. Upon removing H2O from the system, an increase of catalytic activity can be seen for every catalyst. 1.5%Zr and 3.4%Pd 232  regain high conversion (94 mol% and 90 mol% conversion, respectively), whereas 15%Zr and 25%Zr only achieve 80 mol% of conversion after removing H2O. Further experiments with 5vol.% H2O were done to better understanding the impact of H2O on different supports. The results are shown in Figure I.5.  0 5 10 15 20 25020406080100CH4 Conversion (mol.%)Time on Stream (h)   Figure I.5. Wet-TOS results for (■) 3.4Pd/Al2O3, (∆) seq-1.5Zr/3.4Pd/Al2O3, () seq-15Zr/3.4Pd/Al2O3, and (○) seq-25Zr/3.4Pd/Al2O3 catalysts at 350°C with 5vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  With 5vol.% H2O, the sequential catalysts show a better stability than 3.4Pd/Al2O3 catalyst in the following order: 1.5Zr/3.4Pd/Al2O3 > 15Zr/3.4Pd/Al2O3 > 25Zr/3.4Pd/Al2O3 for the first 5h. The CH4 conversion for both seq-15Zr/3.4Pd/Al2O3 and seq-25Zr/3.4Pd/Al2O3 catalysts reached to 18% after 5h and continued to drop to 12% from t=5h to t=24h. In the case of 3.4Pd/Al2O3 catalyst a fast decrease in CH4 conversion was observed in the first 2h. Then the conversion remained constant at 12% from t=2h to t=24h. The seq-15Zr/3.4Pd/Al2O3 and 233  seq-25Zr/3.4Pd/Al2O3 catalysts were more stable than 3.4Pd/Al2O3 catalyst only at the beginning of TOS experiment, as all three catalysts reached the same conversion at t=24h. The seq-1.5Zr/3.4Pd/Al2O3 catalyst showed highest activity for the whole 24h TOS, however, it was more significant for the first 10h. Upon removing H2O, the CH4 conversion reached to 98% for 3.4Pd/Al2O3 and seq-1.5Zr/3.4Pd/Al2O3 catalysts, 92% and 85% for seq-15Zr/3.4Pd/Al2O3 and seq-25Zr/3.4Pd/Al2O3 catalysts, respectively.   Experiments were performed with 10vol.% extra water at 425°C simulating the real condition in a catalytic converter (10-15vol.% water vapor, < 500°C). This setting was chosen, since no deactivation could be detected neither with 5vol.% extra water at 425°C nor with 10vol.% extra water at 450°C. After 24h the water feeding was stopped. The aim was to investigate the faster oxygen mobility of ZrO2 at high temperatures published in the literature [17,130]. Figure I.6 presents the results for 3.4%Pd with 1.5%Zr,15% Zr and 25% =Zr at this condition. 3.4%Pd shows with 78 mol% of CH4 conversion (after 24 hours) the highest catalytic activity and the slowest deactivation during the stability test. The conversion of CH4 for the catalysts with Zr content decreases in the following order: 1.5% Zr > 15% Zr > 25% Zr. After removing the water, all catalysts immediately regain 100 mol% of conversion.   This outcome is unexpected, since ZrO2 is well known for its high thermal stability and for its high oxygen mobility at higher temperatures [17,130]. Moreover Ciuparu published that the effect of support becomes significant having temperatures of 700K (or above) [130]. However, in the experiment faster deactivation and lower conversion occurs for the ZrO2 supported catalysts than for 3.4%Pd catalyst. Consequently other aspects like structural properties have an influence on the catalytic activity at higher temperature. Amairia and 234  Souza argued that the surface area of the catalyst decreases having a higher Zr loading due to the ZrO2 blockage of the pores. Smaller surface area leads to a lower catalytic activity [127,129]. By looking at Figure I.6 it can be seen that the catalytic activity matches with the decreasing magnitude of surface areas (from 3.4%Pd to 3.4%Pd/25%Zr). Sintering occurs for all catalysts, but having a bigger surface area more palladium active sites are available and losing some of them is less effective than for catalysts with a smaller surface area. Considering the result with 2vol.% water at 350°C (Figure I.4), a further suggestion can be made. At this setting 3.4%Pd was as good as 1.5%Zr and better than 15%Zr and 25%Zr. The reason is mentioned to be due to low water inhibition and good oxygen exchange of the catalyst with the gas phase. The oxygen exchange rate of the support with PdO is slower, thus the support effect is not significant at this condition. The same behavior can cause the results with 10vol.% water at 425°C. At this temperature the water amount (10vol.%) may be low as well as the hydroxyl coverage of palladium active sites. The oxygen exchange between gas phase and catalyst is fast, making the oxygen mobility of the support negligible. Consequently the larger surface area of 3.4%Pd leads to the better catalytic activity seen in Figure I.6. To verify this explanation, further investigation with higher amounts of water has to be done. The activity of 3.4%Pd should decrease more than the activities of the ZrO2 supported catalysts.   I.3 Discussion  The higher initial activity of seq-1.5Zr/3.4Pd/Al2O3 catalyst than 3.4Pd/Al2O3 catalyst is consistent with the results reported in the literature [128,129]. Gou et al. showed the highest catalytic activity was found with a Zr content of 2wt.% and the activity decreased by 235  increasing the Zr loading (> 10wt.%) [128]. ZrO2 is reported to have the oxygen mobility three times higher than Al2O3 [131] and a small amount of ZrO2 may enhance the oxygen exchange and consequently improve the catalytic activity.   0 5 10 15 20 25020406080100CH4 Conversion (mol.%)Time on Stream (h)   Figure I.6. Wet-TOS results for (■) 3.4Pd/Al2O3, (∆) seq-1.5Zr/3.4Pd/Al2O3, () seq-15Zr/3.4Pd/Al2O3, and (○) seq-25Zr/3.4Pd/Al2O3 catalysts at 425°C with 10vol.% H2O. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  On the other hand, increasing the Zr content leads to the formation of higher amount of metastable tetragonal ZrO2, which decreases the catalytic activity [129,132,133]. The XRD results shown in Figure I.2 confirms the presence of tetragonal ZrO2 in seq-25Zr/3.4Pd/Al2O3 catalyst. The BET results may also play a role in terms of the catalyst activity. The lower 236  surface area and lower pore volume at higher loadings of Zr explain the lower CH4 conversion for seq-15Zr/3.4Pd/Al2O3 and for seq-25Zr/3.4Pd/Al2O3 [127,128].   Increasing the amount H2O in the feed stream leads to a lower conversion of CH4. According to Schwartz et al. the accumulation of hydroxyl groups occurs on the surface of the catalyst leads to blockage of the active sites [34]. Having more H2O in the reaction, more active sites are not available, which causes the lower catalytic activity. On the other hand, Burch et al. [117] claimed that the catalytic deactivation is due to the reaction of H2O with active palladium and form inactive Pd(OH)2. Moreover, this reaction is favored at lower temperature (< 450°C), whereas the reverse reaction takes place above 450°C regaining catalytic active sites [117]. As explained in Chapter 5, the formation of Pd(OH)2 is unlikely in this study. The experiments proved however the partial reversibility of H2O on the active sites.   Comparing the wet-TOS results of 3.4Pd/Al2O3 and seq-1.5Zr/3.4Pd/Al2O3 catalysts at different H2O concentrations leads to a conclusion concerning the impact of ZrO2. The influence of the support on catalytic activity gets more significant at higher H2O concentration in the system (i.e. more H2O inhibition on the surface). 3.4%Pd along with 1.5%Zr were the best catalyst at a low H2O amount (2vol.%), however 3.4%Pd shows the worst performance at higher H2O inhibition (5vol.%) (See Figures I.4 and I.5). ZrO2 is known for its high oxygen mobility (three times higher than Al2O3) [131]. Schwartz and Ciuparu published that supports with high oxygen mobility are more resistant to H2O inhibition confirming our results obtained in the experiments [33,36]. During CH4 oxidation the active sites get reduced by methane and can get re-oxygenized (Pd +         PdO) not 237  only by the gas phase oxygen but also by the oxygen of the support (lattice oxygen) [34,130]. Yang et al. also reported that a high oxygen mobility of lattice oxygen improves the re-oxidation of the palladium active sites and enhances CH4 combustion [134]. Since the oxygen exchange of the gas phase with the catalyst is slow at low temperatures [130,135] and is limited by the surface reaction and H2O inhibition respectively, ZrO2 partly provides the oxygen for the reaction and consequently improves the catalytic activity. ZrO2 has the ability to form oxygen vacancies. These vacancies can get refilled quickly by migration of oxygen from the support or the gas phase [34,124]. That is the reason why the catalysts with Zr content show a better performance than 3.4% Pd, when having high hydroxyl coverage on the surface of the catalyst (Figure I.5).   Less H2O vapor in the system (2vol.%) leads to a better catalytic activity of 3.4%Pd (Figure I.4). The H2O inhibition is weaker (compared to 5vol.%) at this temperature and affects less the oxygen exchange between the surface of the catalyst and the gas phase. The activity of 3.4%Pd and 1.5%Zr is almost the same for this condition concluding that the effect of support is less significant for a lower amount of H2O (less H2O inhibition) in the reaction.  Moreover Ciuparu published that the effect of support becomes significant having temperatures of 700K (or above) [130]. However, in the experiment faster deactivation and lower conversion occurs for the ZrO2 supported catalysts than for 3.4%Pd catalyst. Consequently other aspects like structural properties have an influence on the catalytic activity at higher temperature. Amairia and Souza argued that the surface area of the catalyst decreases having a higher Zr loading due to the ZrO2 blockage of the pores. Smaller surface area leads to a lower catalytic activity [127,129].  238  Considering the result with 2vol.% H2O at 350°C (Figure I.4), a further suggestion can be made. At this setting 3.4%Pd was as good as 1.5%Zr and better than 15%Zr and 25%Zr. The reason is mentioned to be due to low H2O inhibition and good oxygen exchange of the catalyst with the gas phase. The oxygen exchange rate of the support with PdO is slower, thus the support effect is not significant at this condition. The same behavior can cause the results with 10vol.% H2O at 425°C. At this temperature the H2O amount (10vol.%) may be low as well as the hydroxyl coverage of palladium active sites. The oxygen exchange between gas phase and catalyst is fast, making the oxygen mobility of the support negligible. Consequently the larger surface area of 3.4%Pd leads to the better catalytic activity seen in Figure I.6. To verify this explanation, further investigation with higher amounts of H2O has to be done. The activity of 3.4%Pd should decrease more than the activities of the ZrO2 supported catalysts.  One dry experiment (with 1.5%Zr) was performed to identify the influence of temperature on the catalytic activity. Therefore the temperature was changed between 280°C and 330°C. In Figure I.7 the results are presented for four different temperatures.  239  0 5 10 15 20 25020406080100CH4 Conversion (mol.%)Time on Stream (h)(a)(b)(c)(d)   Figure I.7. Dry-TOS results for seq-1.5Zr/3.4Pd/Al2O3 catalyst at (a) 310°C, (b) 280°C, (c) 250°C, and (d) 310°C. GHSV=180,000 cm3(STP).gcat-1.h-1, 1000 ppm CH4, 20(v/v)% O2, and the balance He and Ar  The deactivation of the catalyst started immediately having reached the temperature of 310°C. The conversion decreased from 99 mol% to 93 mol% while the temperature was kept at T=310°C for about 3h. At the temperature of 280°C the conversion decreased from 60% to 56% after 15h of reaction. For 250°C the conversion stayed constant at 25% for 8h. However, increasing the temperature to 310°C, the deactivation restarted. The deactivation of the catalyst could be due to the sintering caused by temperature. For example, smaller PdO particles combine to bigger PdO particles, which are supposed to be less active. Moreover, pores of alumina melt together that prevents the usability of PdO particles inside the pores and reduces the surface area of the catalyst. All of these effects are irreversible and cause the decrease of conversion over time. More sintering occurs for higher temperature that is why the deactivation restarted, increasing the temperature from 250°C to 310°C. Since H2O 240  inhibition occurs below 450°C, the coverage of PdO active sites through hydroxyl groups can also affect the conversion of CH4 [34]. Thus, hhigher deactivation of the catalyst at higher temperature could also be due to higher H2O production during CH4 oxidation reaction. Hydroxyl groups may adsorb on PdO active sites or sinter the catalyst. Figure I.7 shows clearly the impact of temperature on the catalytic performance. Increasing the temperature leads to higher conversion in CH4 oxidation. The difference in conversion between 310°C and 250°C is around 60 mol% after a certain time of reaction.  The initial points of 280°C and 250°C obtained from TPO experiments are included in Figure I.7. They represent the maximum achievable conversion when using a fresh catalyst. The strong influence of temperature in CH4 oxidation can be seen comparing the initial points of conversion at 310, 280, 250°C. They are 99, 72 and 38 mol% of conversion, respectively.   Considering the surface reaction of the catalyst the impact of temperature in CH4 oxidation is explainable. Beebe et al. studied the adsorption process of CH4 and showed that one C-H bond breaks leading to adsorbed hydrogen and methyl radicals [136]. Due to the high binding energy of the C-H bond in CH4 [12], higher temperature favors the activation of CH4 and the formation of these radicals. Consequently increasing the temperature enhances the adsorption process of CH4 and the conversion of CH4 to CO2 and H2O on the surface of the catalyst. Another explanation for the less catalytic activity at lower temperature is made by Ciuparu and Au-Yeung [130,135]. They found out that the overall oxygen exchange of the gas phase with the catalyst is very slow at lower temperature. Thus slower oxygen exchange can also cause the decrease of conversion when reducing the temperature. 241  I.4 Conclusion   The seq-1.5Zr/3.4Pd/Al2O3 catalyst demonstrates the highest catalytic activity compared to 3.4Pd/Al2O3, 15% Zr and 25% Zr. When comparing the results at 350°C between low amount of H2O (2vol.%) with high amount of H2O (5vol.%), it can be concluded that the effect of support gets more significant having more H2O inhibition. At 2vol.% of H2O the unsupported and 1.5%Zr supported catalysts, both have with 27mol.% of conversion the highest activity. This can be explained due to the larger surface area of 3.4%Pd and the higher oxygen mobility of ZrO2 (monoclinic). However, increasing the H2O vapor to 5vol.% the unsupported catalyst (3.4%Pd) shows the worst performance compared to all ZrO2 supported catalysts especially during the first 10h. After 24h the 1.5%Zr catalyst was with 16 mol% conversion around 4 mol% better than 3.4%Pd. Thus a higher H2O amount leads to more hydroxyl blockage of the palladium active sites and limits/decelerates the oxygen exchange from the gas phase to the catalyst. In this condition the oxygen mobility of ZrO2 (and the oxygen exchange from the support to the surface respectively) helps to improve the conversion of CH4.   After 24h the feeding of H2O was stopped in all wet experiments to investigate the reversibility of H2O inhibition. The conversion went up immediately proving the reversible hydroxyl coverage on the surface of the catalyst. However, the final conversion depends on the H2O amount utilized in the 24h before. Using more extra H2O in the experiment, the conversion regains to a higher value. As it was explained before more extra H2O leads to more hydroxyl coverage of palladium active sites. Due to the coverage more PdO sites are 242  protected against sintering and stay unused. After removing the H2O more fresh PdO sites are available again causing a higher final conversion.   The faster oxygen mobility of ZrO2 at higher temperature could not be proved, because the unsupported catalyst presents the best catalytic activity at 425°C and 10vol.% of H2O. The catalytic activity at this condition decreases in the following order: 3.4%Pd > 1.5%Zr > 15%Zr > 25% Zr with the corresponding conversions of 76% > 68% > 48% > 35% after 2h. It can be suggested that 10vol.% at 425°C leads to a low H2O inhibition like 2vol.% H2O at 350°C. The oxygen exchange between gas phase and catalyst is fast, making the impact of the support negligible. The catalytic activity may depend on the surface area, since the activity decreases in accordance with the decrease of the surface areas (3.4%Pd (215m2/g) > 1.5%Zr (192m2/g) > 15%Zr (139m2/g) > 25%Zr).           243  Appendix J: The Effect of Second Metal on Pd Catalysts for CH4 Oxidation (Bimetallic)  The effect of metal oxides added to Pd/Al2O3 to improve the hydrothermal stability has been reported by Liu et al. [115] who showed in particular, that the addition of NiO or MgO improved the hydrothermal stability of Pd/Al2O3 through the formation of NiAl2O4 and MgAl3O4 spinel structures. According to the authors, the spinel results in weakened support acidity that suppresses the formation of Pd(OH)2 during hydrothermal aging.   Pd-bimetallic catalysts have also been studied to improve stability of Pd catalysts for CH4 oxidation [26,91,100,137]. Pd-bimetallic catalysts are usually less active than Pd alone [43,138–140] simply because they contain less Pd, the most active metal for CH4 oxidation [29]. The lower activity of the bimetallic compared to Pd alone may also be due to the presence of smaller amounts of PdO as a result of alloy formation between Pd and Pt [43], or the transformation of PdO to Pd metal [141]. According to Ozawa et al. [142] the addition of Pt improves PdO/Al2O3 catalyst stability by preventing the growth of PdO and Pd–Pt particles during CH4 oxidation at high temperature (800C) [142]. Several studies have reported higher initial activity of Pd-bimetallic catalysts compared to Pd alone [24,26,100,143]. These researchers suggest that the second metal added to Pd dissociates O2 and the resulting O atoms are adsorbed by Pd, helping to maintain PdO active sites. Ishihara et al. [143] reported T50 for a 1wt.% Pd/Al2O3 catalyst to be 533°C, whereas for a Pd-Ni/Al2O3 catalyst (Pd:Ni= 9:1) T50 was found to be 380°C. In another study, it was found that the higher dispersion of PdO on a PdO-Pt/α-Al2O3 catalyst (27%) compared to PdO/α-Al2O3 (14%) results in higher initial activity and higher stability of the bimetallic catalyst 244  [100]. After exposing the PdO/α-Al2O3 catalyst to the reaction feed stream for 6h at 350°C, an increase in average particle size from 8 to 11 nm was observed, whereas the average particle size did not change significantly for the PdO-Pt/α-Al2O3 catalyst [100]. Persson et al. [139] examined a series of Pd-bimetallics supported on Al2O3 finding that the metallic phase structure had a significant influence on the catalyst stability. For example, in several bimetallic systems (PdAg, PdCu, PdRh, and PdIr) spinel phases enhanced catalyst stability, whereas formation of Co or Ni aluminate spinels in PdCo and PdNi bimetallics did not improve catalyst stability. Alloy formation in PdPt and PdAu on Al2O3 was found to increase hydrothermal stability in the presence of 15%H2O/air at 1000C for 10h. In another study by Persson et al. [43], Pd-Pt bimetallic catalysts on various supports (Al2O3, ZrO2) also yielded better thermal stability than monometallic Pd during CH4 oxidation in dry air (1.5%CH4 in air at a GHSV 250,000h-1). The stability of the Pd-Pt catalysts was better at lower temperatures (up to 620°C). At temperatures of 520°C and 570°C CH4 conversion of the Pd-Pt catalysts increased with time-on-stream. Above 620°C (especially at 670°C and 720°C) conversion decreased with time-on-stream. Those catalysts with higher initial activity also showed higher deactivation rates. The deactivation could not be attributed to PdO decomposition because the initial activity test showed that PdO decomposition started at higher temperature (770°C with 1.5vol.% CH4 in air). The XRD results also confirmed that no PdO decomposition was observed at temperatures below 800°C for the Pd/Al2O3.  The amount of second metal added to the Pd can also affect the stability of the bimetallic catalyst. Persson et al. [138] showed that Pd-Pt bimetallic catalysts with Pt:Pd ratios of 0.33:0.67 and 0.5:0.5 were stable catalysts. Time-on-stream experiments for both a 5wt.% Pd/Al2O3 and a 2:1Pd:Pt/Al2O3 bimetallic with total metal loading of 5wt.% were studied 245  over a wide range of temperatures (470-720°C) [43]. The temperature was increased from 470°C to 720°C stepwise by 50°C and was held for 1h at each temperature. CH4 conversion over the Pd/Al2O3 and Pd-Pt/Al2O3 catalyst decreased during the 1h reaction time at each temperature. However, the decrease in conversion was lower for the bimetallic catalyst compared to the Pd catalyst. The decrease in activity was higher at higher temperatures (670°C and 720°C), especially for the Pd catalyst. In situ XRD spectra of the Pd-Pt bimetallic catalysts are shown in Figure J.1 at room temperature, a sharp peak corresponding to Pd-Pt (111) and a small peak corresponding to PdO (101) are observed for the PdPt-Al2O3 catalyst. By increasing the temperature to 300°C, the PdO peak disappears and then reappears at 500°C. The Pd-Pt peak intensity reaches a maximum at 700°C while the PdO peak disappears at this temperature. The formation of Pd-Pt instead of PdO is consistent with deactivation of the bimetallic catalyst at high temperature (700°C).  246   Figure J.1. High-temperature in situ XRD profiles of PdPt-Al2O3 during heating [43] (Copyright © 2006 Elsevier)  Steady-state experiments using a 18.7wt.%Pd/Al2O3 catalyst with different loadings of Pt (1.6, 3.1 and 3.9wt.%) (Figure J.2) reported by Ozawa et al. [142], also provide some insight into the improved stability of these bimetallic catalysts as Pt content is increased. In this study, reaction temperature was held at 800°C and CH4 combustion rate was measured over a 10h period using a 1%CH4 in air feed gas at a GHSV of 1,500,000 mL/(gcat.h). Deactivation rate is shown to decrease as the Pt loading of the Pd-Pt bimetallics increase. For example, the combustion rate for the 18wt.% Pd-3.9wt.% Pt/Al2O3 decreased from 710 μmol.s-1.g-1 to 460 μmol.s-1.g-1 after 10h, whereas it decreased to 400 μmol.s-1.g-1 for the 18.4wt.% Pd-1.6wt.% Pt/Al2O3 catalyst. 247   Figure J.2. CH4 combustion rate at 800°C with time on stream. Combustion conditions: CH4=1vol.%, air=99vol.%, CH4/air flow= 450 L.h-1, catalyst weight= 0.3g. Catalyst 1, 2, 3, and 4 represent 18.7wt.% Pd, 18.4wt.%Pd-1.6wt.% Pt, 18.1wt.% Pd-3.4wt.% Pt, and 18.0wt.% Pd-3.9wt.% Pt over Al2O3 catalysts [142] (Copyright © 2004 Elsevier)  XRD analysis of the catalysts studied by Ozawa et al. [142] after 10h reaction indicates the PdO to be present in the Pt-doped catalysts while no Pd0 is observed. However, Pd0 was present in the Pd monometallic catalyst, likely because of the decomposition of PdO at the high temperature of the reaction (800C). In addition, the crystallite size of the PdO (101) in the Pd catalyst was larger than for the Pd-Pt catalysts. Table J.1 compares changes in PdO particle size and BET surface area before and after 10h reaction for the same Pd and Pd-Pt catalysts. From these data it is clear that the extent of sintering of the Pd catalyst is greater than for the Pd-Pt catalysts. The time-on-stream conversion data reported by Ozawa et al. 248  [142] (Figure J.2) were fitted to a deactivation equation with two terms, the first representing rapid transformation of PdO to Pd0 of the Pd-Pt alloy phase, and the second associated with the slow growth of the PdO crystallite [142]. The deactivation was affected more by the second term suggesting that particle growth of the PdO is the main cause of catalyst deactivation at the chosen reaction conditions [142].   Table J.1. Changes in Pd and Pt-Pd catalyst properties before and after aging (Adapted with permission from [142]) Catalyst, wt % on Al2O3 18.7%Pd 18.4%Pd-1.6%Pt 18.1%Pd-3.1%Pt 18.0%Pd-3.9%Pt BET area, m2/g Fresh 56 51 51 52 Aged 46 46 46 46 PdO size, nm Fresh 12.5 15.3 15.2 14.7 Aged 17.9 18.0 16.7 16.2  These results are in a good agreement with the results reported by Yamamoto et al. [137] in which a Pd-Pt bimetallic catalyst was more active for CH4 conversion (in terms of 50% CH4 conversion) than Pd and the conversion was maintained following 2500h time-on-stream at 385°C. XRD analyses showed that the crystallite growth as a function of time for both Pd (111) and PdO (101) was faster on the Pd (10 g/l)/Al2O3 catalyst than the Pd (10 g/l)-Pt (10 g/l)/Al2O3 catalyst. Hence one concludes that the presence of Pt retards the sintering of PdO. Effects of H2O on deactivation of Pt versus Pt-Pd catalysts have also been reported, at both thermal and hydrothermal aging conditions [24,26,91]. Pieck et al. [24] reported that the T50 of a 0.4%Pt-0.8%Pd/Al2O3 catalyst after thermal treatment at 600°C for 4h in wet air (60 cm3.min-1 air flow with 0.356 cm3.h-1 water), was~50C lower than that obtained over a Pd catalyst. Lapisardi et al. [26] reported that a fresh Pd0.93-Pt0.07/Al2O3 catalyst (total metal 249  loading 2.12wt.% with Pd:Pt molar ratio of 0.93:0.07) was as active as a fresh Pd/Al2O3 catalyst in a dry feed [26]. Interestingly, the Pd0.93-Pt0.07/Al2O3 catalyst was less affected by addition of 10vol.% steam to the feed stream than the 2.2wt.%Pd/Al2O3 catalyst. The T50 for the Pd-Pt bimetallic increased from 320°C to 400°C when 10vol.% steam was added to the feed stream, whereas the corresponding increase in T50 for the Pd/Al2O3 catalyst was from 320°C to 425°C. Thus, the Pd-Pt bimetallic, containing only 0.26wt.%Pt was more active and stable than the Pd catalyst for CH4 oxidation in the presence of steam.  The stability of Pt and Pt-Pd catalysts loaded on a washcoated monolith has also been reported [91]. A feed stream with 4067 ppmv CH4 in air was reacted over these catalysts as reaction temperature increased from 300°C to 700°C stepwise in 50°C increments. CH4 conversion was monitored for a period of 1h at each temperature. Subsequently the temperature was decreased to 300°C also in 50°C steps, again holding at each temperature for 1h. The conversion of CH4 was compared for both heating and cooling cycles. The results showed that the Pt-Pd catalyst was more active than the Pt catalyst. The comparison between the heating and cooling cycles was also done for steam-aged catalysts, in which the catalysts were exposed to the feed stream at 650°C with 5vol.% water for 20h. Table J.2 lists the T50 for both fresh Pt and Pd-Pt catalysts, the steam-aged catalysts during tests in a dry feed and the steam-aged catalysts tested in a wet feed, containing 5wt.%H2O. The data show that the fresh Pd-Pt catalyst is more active than the fresh Pt catalyst. Higher activities were also observed for steam-aged Pd-Pt catalysts tested in dry or wet feed gas.   250  Table J.2. T50 for fresh and steam aged Pd and Pt-Pd catalysts operated in dry and wet feed. Combustion conditions: 4067 vol. ppm CH4; total flow rate of 234.5 cm3.min-1; 500 mg catalyst; 5vol.% water in wet feed (Adapted with permission from [91])   Temperature at 50% CH4 conversion (T50), C Catalyst Fresh Dry feed Steam-aged Dry feed Steam-aged Wet feed Pt 540 610 610 4:1 Pt-Pd 400 470 535    251  Appendix K: MATLAB M-files Code  A non-linear regression MATLAB program using Levenberg-Marquardt method written by R. Schrager and A. Jutan and made available through an open-access MATLAB users group was combined with simple MATLAB code containing calculations to solve an ODE using a 4th-order Runge-Kutta algorithm.   Least Square:  function [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]= ...       leasqr(x,y,pin,F,stol,niter,wt,dp,dFdp,options) plotcmd='plot(x(:,1),y,''+'',x(:,1),f); shg'; %if (sscanf(version,'%f') >= 4), vernum= sscanf(version,'%f'); if vernum(1) >= 4,   global verbose   plotcmd='plot(x(:,1),y,''+'',x(:,1),f); figure(gcf)'; end; if (exist('OCTAVE_VERSION'))   global verbose end;  if(exist('verbose')~=1), %If verbose undefined, print nothing     verbose(1)=0    %This will not tell them the results     verbose(2)=0    %This will not replot each loop end; if (nargin <= 8), dFdp='dfdp'; end; if (nargin <= 7), dp=.001*(pin*0+1); end; %DT if (nargin <= 6), wt=ones(length(y),1); end;    % SMB modification if (nargin <= 5), niter=50; end; if (nargin == 4), stol=.0001; end; 252   y=y(:); wt=wt(:); pin=pin(:); dp=dp(:); %change all vectors to columns % check data vectors- same length? m=length(y); n=length(pin); p=pin;[m1,m2]=size(x); if m1~=m ,error('input(x)/output(y) data must have same number of rows') ,end;  if (nargin <= 9),   options=[zeros(n,1) Inf*ones(n,1)];   nor = n; noc = 2; else   [nor noc]=size(options);   if (nor ~= n),     error('options and parameter matrices must have same number of rows'),   end;   if (noc ~= 2),     options=[options(noc,1) Inf*ones(noc,1)];   end; end; pprec=options(:,1); maxstep=options(:,2);  % set up for iterations% f=feval(F,x,p); fbest=f; pbest=p; r=wt.*(y-f); sbest=r'*r; nrm=zeros(n,1); chgprev=Inf*ones(n,1); kvg=0; epsLlast=1; epstab=[.1 1 1e2 1e4 1e6];  % do iterations% for iter=1:niter,   pprev=pbest;   prt=feval(dFdp,x,fbest,pprev,dp,F); 253    r=wt.*(y-fbest);   sprev=sbest;   sgoal=(1-stol)*sprev;   for j=1:n,     if dp(j)==0,       nrm(j)=0;     else       prt(:,j)=wt.*prt(:,j);       nrm(j)=prt(:,j)'*prt(:,j);       if nrm(j)>0,         nrm(j)=1/sqrt(nrm(j));       end;     end     prt(:,j)=nrm(j)*prt(:,j);   end; % above loop could ? be replaced by: % prt=prt.*wt(:,ones(1,n)); % nrm=dp./sqrt(diag(prt'*prt)); % prt=prt.*nrm(:,ones(1,m))';   [prt,s,v]=svd(prt,0);   s=diag(s);   g=prt'*r;   for jjj=1:length(epstab),     epsL = max(epsLlast*epstab(jjj),1e-7);     se=sqrt((s.*s)+epsL);     gse=g./se;     chg=((v*gse).*nrm); %   check the change constraints and apply as necessary     ochg=chg;     for iii=1:n,       if (maxstep(iii)==Inf), break; end; 254        chg(iii)=max(chg(iii),-abs(maxstep(iii)*pprev(iii)));       chg(iii)=min(chg(iii),abs(maxstep(iii)*pprev(iii)));     end;      if (verbose(1) & any(ochg ~= chg)),        disp(['Change in parameter(s): sprintf('%d ',find(ochg ~= chg)) 'were constrained']);      end;     aprec=abs(pprec.*pbest);        % ss=scalar sum of squares=sum((wt.*(y-f))^2).     if (any(abs(chg) > 0.1*aprec)),%---  % only worth evaluating function if       p=chg+pprev;                       % there is some non-miniscule change       f=feval(F,x,p);       r=wt.*(y-f);       ss=r'*r;       if ss<sbest,         pbest=p;         fbest=f;         sbest=ss;       end;       if ss<=sgoal,         break;       end;     end;                              end;   epsLlast = epsL; %   if (verbose(2)), %     eval(plotcmd); %   end;   if ss<eps,     break;   end   aprec=abs(pprec.*pbest); 255  %  [aprec chg chgprev]   if (all(abs(chg) < aprec) & all(abs(chgprev) < aprec)),     kvg=1;     if (verbose(1)),       fprintf('Parameter changes converged to specified precision\n');     end;     break;   else     chgprev=chg;   end;   if ss>sgoal,     break;   end; end; % set return values% p=pbest; f=fbest; ss=sbest; kvg=((sbest>sgoal)|(sbest<=eps)|kvg); if kvg ~= 1 , disp(' CONVERGENCE NOT ACHIEVED! '), end; % CALC VARIANCE COV MATRIX AND CORRELATION MATRIX OF PARAMETERS % re-evaluate the Jacobian at optimal values% jac=feval(dFdp,x,f,p,dp,F); msk = dp ~= 0; n = sum(msk);           % reduce n to equal number of estimated parameters jac = jac(:, msk);  % use only fitted parameters  %% following section is Ray Muzic's estimate for covariance and correlation %% assuming covariance of data is a diagonal matrix proportional to %% diag(1/wt.^2). if vernum(1) >= 4, 256    Q=sparse(1:m,1:m,(0*wt+1)./(wt.^2));  % save memory   Qinv=inv(Q); else   Qinv=diag(wt.*wt);   Q=diag((0*wt+1)./(wt.^2)); end; resid=y-f;                                    %un-weighted residuals covr=resid'*Qinv*resid*Q/(m-n);                 %covariance of residuals Vy=1/(1-n/m)*covr;  % Eq. 7-13-22, Bard         %covariance of the data  jtgjinv=inv(jac'*Qinv*jac);         %argument of inv may be singular covp=jtgjinv*jac'*Qinv*Vy*Qinv*jac*jtgjinv;  d=sqrt(abs(diag(covp))); corp=covp./(d*d');  covr=diag(covr);                 % convert returned values to compact storage stdresid=resid./sqrt(diag(Vy));  % compute then convert for compact storage Z=((m-n)*jac'*Qinv*jac)/(n*resid'*Qinv*resid); %%% alt. est. of cov. mat. of parm.:(Delforge, Circulation, 82:1494-1504, 1990 %%disp('Alternate estimate of cov. of param. est.') %%acovp=resid'*Qinv*resid/(m-n)*jtgjinv  %Calculate R^2 (Ref Draper & Smith p.46)% r=corrcoef(y,f); if (exist('OCTAVE_VERSION'))   r2=r^2; else   r2=r(1,2).^2; end  % if someone has asked for it, let them have it %  if (verbose(2)), eval(plotcmd); end,  if (verbose(1)),    disp(' Least Squares Estimates of Parameters') 257     disp(p')    disp(' Correlation matrix of parameters estimated')    disp(corp)    disp(' Covariance matrix of Residuals' )    disp(covr)    disp(' Correlation Coefficient R^2')    disp(r2)    sprintf(' 95%% conf region: F(0.05)(%.0f,%.0f)>= delta_pvec''*Z*delta_pvec',n,m-n)    Z   n1 = sum((f-y) < 0);   n2 = sum((f-y) > 0);   nrun=sum(abs(diff((f-y)<0)))+1;   if ((n1>10)&(n2>10)), % sufficient data for test?     zed=(nrun-(2*n1*n2/(n1+n2)+1)+0.5)/(2*n1*n2*(2*n1*n2-n1-n2)...       /((n1+n2)^2*(n1+n2-1)));     if (zed < 0),       prob = erfc(-zed/sqrt(2))/2*100;       disp([num2str(prob) '% chance of fewer than ' num2str(nrun) ' runs.']);     else,       prob = erfc(zed/sqrt(2))/2*100;       disp([num2str(prob) '% chance of greater than ' num2str(nrun) ' runs.']);     end;   end; end  % A modified version of Levenberg-Marquardt % Non-Linear Regression program previously submitted by R.Schrager. % This version corrects an error in that version and also provides % an easier to use version with automatic numerical calculation of % the Jacobian Matrix. In addition, this version calculates statistics % such as correlation, etc.... % Errors in the original version submitted by Shrager (now called version 1) 258  % and the improved version of Jutan (now called version 2) have been corrected. % Additional features, statistical tests, and documentation have also been % included along with an example of usage.  BEWARE: Some the the input and % output arguments were changed from the previous version. % %     Ray Muzic     <rfm2@ds2.uh.cwru.edu> %     Arthur Jutan  <jutan@charon.engga.uwo.ca>  dfdp: function prt=dfdp(x,f,p,dp,func) % numerical partial derivatives (Jacobian) df/dp for use with leasqr % --------INPUT VARIABLES--------- % x=vec or matrix of indep var(used as arg to func) x=[x0 x1 ....] % f=func(x,p) vector initialised by user before each call to dfdp % p= vec of current parameter values % dp= fractional increment of p for numerical derivatives %      dp(j)>0 central differences calculated %      dp(j)<0 one sided differences calculated %      dp(j)=0 sets corresponding partials to zero; i.e. holds p(j) fixed % func=string naming the function (.m) file %      e.g. to calc Jacobian for function expsum prt=dfdp(x,f,p,dp,'expsum') %----------OUTPUT VARIABLES------- % prt= Jacobian Matrix prt(i,j)=df(i)/dp(j) %================================ m=length(x);n=length(p);      %dimensions ps=p; prt=zeros(m,n);del=zeros(n,1);       % initialise Jacobian to Zero for j=1:n       del(j)=dp(j) .*p(j);    %cal delx=fract(dp)*param value(p)            if p(j)==0            del(j)=dp(j);     %if param=0 delx=fraction            end       p(j)=ps(j) + del(j); 259        if del(j)~=0, f1=feval(func,x,p);            if dp(j) < 0, prt(:,j)=(f1-f)./del(j);            else            p(j)=ps(j)- del(j);            prt(:,j)=(f1-feval(func,x,p))./(2 .*del(j));            end       end       p(j)=ps(j);     %restore p(j) end return Modelmulti: function f = modelmulti (x,pin) % Solve a simple system of 2 ODE's  - 2 response variables % find the solution (f) at sepcified x values - corresponding to measured data % first data point in x corresponds to initial condition % global tempK thetaW ncount ya0 PT rhocat Rp tau mPd nx nvar CT; global verbose for knt = 1:ncount; % data sets W0=0.0; WF=x(knt);   initial = 0; [W,Xa] = ode45(@P2b, [W0,WF], initial, [],pin,knt);     %x is the solution matrix% yfinal(knt,:)=Xa(end,:); end f = yfinal(:) pin end     260  P2B:  %USS FBR % Program contains calculations for odes which are to be solved by ODE45%  function xprime = P2bwet(W,Xa,pin,knt) global tempK thetaW ncount ya0 PT rhocat Rp tau mPd nx nvar CT Deff global verbose T tempK(knt) Tbar = 603.; Deff=6.726e-8; % m2/s at 330°C DeffT=Deff*((tempK(knt)/(273+330))^0.5); P=PT(knt); T=tempK (knt); km1=exp((-pin(1)/8.314)*(1/tempK(knt)-1./Tbar)); km=pin(2)*km1; Kwater1=exp(-pin(3)/8.314*(1/tempK(knt)-1./Tbar)); Kwater=pin(4)*Kwater1; k1=km*CT(knt)*CT(knt)*8.314*tempK(knt)*1000*rhocat;% OPTION 1 ratio = k1/DeffT; thiele=Rp*sqrt(ratio); eta=(3/thiele)*((1/tanh(thiele))-1/thiele); % Mina's model xprime=eta*km*CT(knt)*CT(knt)*ya0(knt)*(1-Xa)*P/(1+Kwater*ya0(knt)*P*(thetaW(knt)+2*Xa));  Multi_realfit:  clear all global tempK thetaW ncount ya0 PT rhocat Rp tau mPd nx nvar CT Deff global verbose   verbose(1:2) = 1; 261  % This program does non-linear regression using the lsqr program...a Levenberg-Marquardt nonlinear regression% % Start by generating some phoney data for the test% % x is the indep variable vector e.g. time measurements % y is matrix of responses % columns of y are responses y1, y2 (e.g. mol frac of component 1 and 2) % rows of y are y values at the value of the indep variable (time) in x % first row of y is initial value of response % the program uses the Levenberg-Marquardt method to estimate parameters % and calc statistics - done in leasqr and dfdp % these two matlab m-files are designed for single response % the input data is re-arranged to yield a single response vector y % the L-M requires the model to be calculated -this is done in modelmulti.m % and assumes the model is a series of ODEs, with the number of odes equal  % to the number of responses. The ODEs are calculated in ODEfunm. Note that this function must use the correct model for each y% % Generate INPUT data % In this demo the data are generated from the known problem solution % input number of responses nvar=1;  %This example is single response (conversion versus time)% % Arbitrary system properties %             %  INPUT DATA format longE rhocat=1.49e3; % Catalyst density kg/m3 Rp = (2.22e-4/2); % Particle radius, m measured=xlsread('PdCh4-1.xlsx') % Order of input file: % Temp(K)   ThetaW   CONV %  GHSV (ml/g h)  YaO  PT Pd-loading CT tempK = 273.+ measured (:,1)% reaction T thetaW = measured(:,2) % water to methane feed molar ratio conv = measured(:,3)./100.0 %CH4 conv 262  ghsv = measured (:,4)./22414./60./60. % units are mol/(g.s) ya0=measured(:,5)  %CH4 inlet mole fraction ghsvA=ghsv.*ya0 tau=1./ghsvA PT=measured(:,6)  % Total pressure mPd=measured(:,7)  % mass fraction of Pd catalyst uCT = measured(:,8)  % Total sites micromole/g %Deff = measured(:,9) CT = uCT./1e6; [tcount,nnn]= size(tempK) Xcount=tcount W0=0.0; ncount=tcount; %  INPUT DATA - re-formatted% nx=ncount y=conv' newy=y(:) oldy=reshape(newy,nx,nvar) x=tau' newx=x(:); oldx=x(:); % plot (tempK,y, 'o')%  %INPUT DATA NOW IN CORRECT COLUMN FORMAT%  y=newy  x=newx     %  provide initial parameter guesses% pin=[1.15E5 17 -40000 1.5E-3]                                              np=length(pin) % Begin calculation by calling L-M least squares routine% [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]=leasqr(x,y,pin,'modelmulti'); disp('RESPONSE:') 263  if kvg ==1     disp ('PROBELM CONVERGED')     elseif kvg == 0     disp('PROBLEM DID NOT CONVERGE') end oldf=reshape(f,nx,nvar); oldr=reshape(y-f, nx, nvar); disp ('X-values:')     disp (oldx')      disp ('Y-values')     disp(oldy)      disp('f-values - i.e. model calculated responses')     disp(oldf)     disp('Residuals:')     disp (oldr)     disp ('Final SSQ')     disp (stdresid)     disp ('Estimated parameter values are;')     disp (p)     disp ('Covariance of estimated parameters')     disp (covp)     disp('R2 values is:')     disp (r2)     disp (p)     disp (pin) Calc the eta and theile modulus   Tbar = 603.;   for knt=1:ncount       tempK(knt)       Deff=6.726e-8; % m2/s at 330°C DeffT=Deff*((tempK(knt)/(273+330))^0.5) 264  P=PT(knt); %DeffT=Deff(knt); T=tempK (knt); %T=(1./tempK(knt)-1./Tbar) disp (p(1)) km1=exp((-p(1)/8.314)*(1/tempK(knt)-1./Tbar)) km=p(2)*km1 Kwater1=exp(-p(3)/8.314*(1/tempK(knt)-1./Tbar)) Kwater=p(4)*Kwater1 %k1=km*CT(knt)*CT(knt)*8.314*tempK(knt)*1000*rhocat/(1+Kwater*ya0(knt)*P*(thetaW(knt)+2*f(knt))) %k1=(km*CT(knt)*CT(knt)*8.314*tempK(knt)*1000*rhocat) % OPTION 1 k1=(km*CT(knt)*CT(knt)*8.314*tempK(knt)*1000*rhocat) % OPTION 1 ratio = k1/DeffT; thiele=Rp*sqrt(ratio); TM(knt)=thiele; EffectFactor(knt)=(3/thiele)*((1/tanh(thiele))-1/thiele);   end disp ('thiele modulus') disp (TM) disp ('Effectiveness factors') disp (EffectFactor)    % plot (oldx,oldy,'d'), hold, plot (oldx,oldf)   % plot (oldy,oldf, 'o'),hold, plot (oldy,oldy) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0308724/manifest

Comment

Related Items