Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evaluation of alkali and alkaline earth doped samarium oxide catalysts for the oxidative coupling of… Knights, Shanna D. 1993

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

Item Metadata

Download

Media
831-ubc_1993_spring_knights_shanna.pdf [ 9.86MB ]
Metadata
JSON: 831-1.0059294.json
JSON-LD: 831-1.0059294-ld.json
RDF/XML (Pretty): 831-1.0059294-rdf.xml
RDF/JSON: 831-1.0059294-rdf.json
Turtle: 831-1.0059294-turtle.txt
N-Triples: 831-1.0059294-rdf-ntriples.txt
Original Record: 831-1.0059294-source.json
Full Text
831-1.0059294-fulltext.txt
Citation
831-1.0059294.ris

Full Text

EVALUATION OF ALKALI AND ALKALINE EARTH DOPED SAMARIUM OXIDE CATALYSTS FOR THE OXIDATIVE COUPLING OF METHANE by SHANNA DENINE KNIGHTS B.A.Sc., The University of British Columbia, 1988 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Chemical Engineering)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April 1993 ©ShanDeiKnghts,193  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  Chemical Engineering  The University of British Columbia Vancouver, Canada  Date  ^  DE-6 (2/88)  zcc iqq3  ABSTRACT  The catalytic oxidative coupling of methane involves the reaction of methane and oxygen at high temperatures (650°C to 900°C) in the presence of a solid metal oxide catalyst to produce the desirable products, ethane and ethylene, as well as the undesirable products, carbon monoxide and carbon dioxide. Although the homogeneous reactions are well understood, the heterogeneous reactions and their effect on the homogeneous reactions are still the subject of much research and discussion. The effect of the catalyst characteristics on the heterogeneous reactions is also an active area of research. The objective of this thesis was to characterize a series of catalysts, and to determine the effect of various catalyst properties on the oxidative coupling reactions.  The experimental part of this thesis consisted of preparing and testing samarium oxide and alkali (Na and K) and alkaline earth (Mg and Ca) doped samarium oxide catalysts for the oxidative coupling of methane. The effects of the specific dopant used, varying dopant concentration (1:100 and 1:10 dopant:Sm mole ratio), and catalyst preparation were evaluated. The catalysts were tested in a bench scale packed bed reactor under conditions of varying temperature (650°C, 750°C, and 850°C) and methane to oxygen mole ratio (2 to 16). The catalysts were characterized by scanning electron microscopy, powder x-ray diffraction, surface area, estimated basicity, ability to form carbonates, and ionic radius of dopant.  The addition of dopants to the samarium oxide catalyst resulted in changes in catalyst ii  performance. No new phases were observed in the Sm 203 crystal upon addition of the dopant cations, indicating that the cations were dispersed throughout the crystal, although probably not uniformly. The dopant concentration affected the catalyst performance; for example, at 750°C, the C2+ yield increased from 13.1% to 14.3% when the Ca:Sm mole ratio was increased from 1:100 to 10:100. A change in the catalyst preparation procedure resulted in an increase in the crystal dimensions and an improved combustion catalyst (e.g., the methane conversion increased from 9.8% for the standard catalyst to 17.4% for the revised catalyst) with, however, a decrease in the C2 yield from 3.2% to 2.7%. The results of this study indicated that, over the range of surface areas tested (2.0 to 3.1 m 2 /g), surface area did not have a significant effect on the catalyst performance. The basicity of the catalyst appears to have a significant effect on the catalyst performance, with an increase in basicity resulting in an increase in C2+ selectivity (from 54.3% for the least basic catalyst, undoped samarium oxide, to 62.8% for the most basic catalyst, 1:10 mole ratio Na:Sm oxide). The catalysts displayed temperature dependent behaviour, and there existed an optimum temperature for maximum C2+ yield, which is dependent on the amount and nature of the dopant, and is likely associated with the formation of carbonates on the catalyst surface. The ionic radius of the cation dopant must be similar to or smaller than the support cation to achieve effective inclusion in the crystal lattice.  iii  TABLE OF CONTENTS  ABSTRACT ^  ii  TABLE OF CONTENTS ^  iv  LIST OF TABLES ^  viii  LIST OF FIGURES ^  x  NOMENCLATURE ^  xiii  ACKNOWLEDGEMENTS ^  xiv  1.  INTRODUCTION ^  1  2.  LITERATURE REVIEW ^  4  2.1  Choice of Doped Samarium Oxide Catalyst ^  8  2.2  Mechanism of Oxidative Coupling of Methane ^  9  2.2.1^Formation of Methyl Radicals ^  10  2.2.2^Formation of Ethane ^  10  2.2.3^Formation of Non-Selective Products (CO X ) and the Inherent C2+ Yield Limit ^  2.2.4^Gas Phase Reactions ^  12 18  2.2.5^Oxidative Coupling of Methane Under Conditions of High Pressure ^ 2.3  20  The Role of Gas Phase, Surface, and Lattice Oxygen ^ 22 2.3.1^Source of the Active Oxygen Species ^ 22 2.3.2^Nature of the Active Oxygen Species ^ 25 iv  2.3.3^Crystal Lattice Oxygen Mobility ^  31  2.3.4^Concentration of Active Sites on the Catalyst Surface ^ 34 2.4  3.  Samarium Oxide Crystal Structure ^  35  2.4.1^Comparison of Cubic versus Monoclinic Sm 2O3  37  ^  2.5  Effect of Water Addition on the Oxidative Coupling of Methane  38  2.6  Potential Reactor Configurations ^  40  2.7  Engineering and Economic Assessments ^  42  METHODS AND MATERIALS ^  45  3.1  Experimental Equipment ^  45  3.2  Data Acquisition ^  49  3.3  Gas Chromatograph ^  49  3.3.1^GC Calibration ^  50  3.4  Flow Measurement and Control ^  51  3.5  Definitions ^  51  3.6  Catalyst Preparation Procedure ^  52  3.7  Experimental Procedure ^  53  3.7.1^Variation of Catalyst Dopant Concentration ^ 55 3.7.2^Catalyst Preparation Modification ^  4.  55  RESULTS AND DISCUSSION ^  57  4.1  Undoped Sm2 O3 : Conditions of Complete Oxygen Conversion . . . .  57  4.1.1^Effect of Gas Phase Flow Rate ^  61  4.1.2^Carbon Deposition in the Reactor ^  61  v  4.1.3 Comparison of Results for Sm 203 to Other Researchers' Data ^ 4.1.4 Space Time Yield ^  62 65  4.2 Undoped Sm203 : Conditions of Incomplete Oxygen Conversion^65 4.2.1 Effect of Methane to Oxygen Ratio ^ 67 4.2.2 Effect of Reaction Temperature ^  70  4.2.3 Carbon Balance over the Reactor ^  72  4.2.4 Variation in Results ^  75  4.3 Alkali and Alkaline Earth Doped Samarium Oxide Catalysts ^ 77 4.3.1 Results of Catalyst Performance Tests ^ 78 4.4 Modification of Catalyst Preparation Procedure ^ 84 4.4.1 Results of Performance Test for Modified Catalyst ^ 86 4.5 Scanning Electron Microscopy-Electron Dispersive X-Ray ^ 90 4.6 Powder X-Ray Diffraction Analysis (XRD) ^ 94 4.7 Measurement and Effect of Catalyst Surface Area ^ 96 4.8^Basicity of the Catalyst ^  103  4.9 Effect of Carbon Dioxide and Carbonate Formation on Catalyst Performance ^  115  4.9.1 Determination of Carbonate Formation for Sodium and Calcium Doped Samarium Oxide Catalysts ^ 121 4.9.2 Effect of Gas Phase Composition on Carbonate Formation ^ 130 4.10 Ionic Radius of Dopant ^  5.^CONCLUSIONS ^  132  137 vi  6.^REFERENCES ^  139  APPENDIX A: RESULTS OF REACTOR CATALYST TESTS ^ 150  APPENDIX B:^SCANNING ELECTRON MICROSCOPY (SEM) PHOTOGRAPHS ^  APPENDIX C: X-RAY DIFFRACTION (XRD) GRAPHS ^  171  176  vii  LIST OF TABLES  Table 2.1  Comparison of Catalyst Activity and Oxygen Exchange Parameters for Cubic and Monoclinic Sm 203  ^  24  Table 3.1  Conditions Used for Comparison Testing of Doped Catalysts ^ 56  Table 3.2  Conditions Used for Comparison Testing of Catalysts at Higher Dopant Concentrations and Different Preparation Times ^ 56  Table 4.1  Samarium Oxide: Conditions of Complete Oxygen Conversion . 58  Table 4.2  Effect of Total Flow Rate ^  Table 4.3  Samarium Oxide Results for Various Researchers ^ 63  Table 4.4  Average Percent Standard Deviation in GC Calibration ^ 76  Table 4.5  Variation in Results for Two Identical Runs ^ 77  Table 4.6  Doped Samarium Oxide Catalysts ^  78  Table 4.7  Results for all Catalysts, CH 4 /02 = 4 ^  79  Table 4.8  Melting and Decomposition Temperatures of Nitrates used in Catalyst Preparation ^  Table 4.9  61  85  Weight % of Catalyst Surface Components According to SEMEDX ^  92  Table 4.10  Surface Area of Catalysts ^  101  Table 4.11  Selectivity and Acid-Base Properties of Various Catalytic Oxides ^ 104  Table 4.12  Electronegativity of Alkali and Alkaline Earth Elements ^ 110  Table 4.13  Effect of Electronegativity on Catalyst Performance ^ 112  Table 4.14  Decomposition Temperature of Various Carbonates ^ 121  Table 4.15  Estimated Ionic Radii of Selected Elements ^ 133 viii  Table 4.16  Effect of Cation Radius/Charge Ratio on C2+ Selectivity ^ 135  Table A.1  Samarium Oxide Prepared in Oxygen ^  Table A.2  Calcium Doped Samarium Oxide (1:100) Prepared in Air ^ 153  Table A.3  Calcium Doped Samarium Oxide (1:100) Prepared in Oxygen 154  Table A.4  Magnesium Doped Samarium Oxide (1:100) Prepared in Air ^ 156  Table A.5  Magnesium Doped Samarium Oxide (1:100) Prepared in Oxygen ^ 157  Table A.6  Sodium Doped Samarium Oxide (1:100) Prepared in Air ^ 159  Table A.7  Sodium Doped Samarium Oxide (1:100) Prepared in Oxygen ^ 161  Table A.8  Potassium Doped Samarium Oxide (1:100) Prepared in Oxygen . . ^ 163  Table A.9  Calcium Doped Samarium Oxide (1:10) Prepared in Oxygen ^ 165  Table A.10  Sodium Doped Samarium Oxide (1:10) Prepared in Oxygen ^ 167  Table A.11  Calcium Doped Samarium Oxide (1:10) Revised Preparation in Oxygen ^  151  169  ix  LIST OF FIGURES  Figure 2.1  The Crystal Structure of Cubic Samarium Oxide Sm 203  Figure 3.1  Equipment Flow Diagram ^  46  Figure 3.2  Reactor Detail ^  48  Figure 4.1  Methane Conversion for Samarium Oxide ^ 59  Figure 4.2  C2 Selectivity  Figure 4.3  C2 Yield  Figure 4.4  Methane Conversion for Samarium Oxide, All Researchers ^ 64  Figure 4.5  C2 Selectivity  Figure 4.6  C2 Yield  Figure 4.7  STY for Various Catalysts, T=750°C ^  66  Figure 4.8  Samarium Oxide Results ^  68  Figure 4.9  Conversion for Samarium Oxide ^  68  ^  for Samarium Oxide ^  for Samarium Oxide ^  59 60  for Samarium Oxide, All Researchers ^ 64  for Samarium Oxide, All Researchers ^ 66  Figure 4.10 Selectivity for Samarium Oxide ^ Figure 4.11  36  % Yield for Samarium Oxide ^  69 69  Figure 4.12 Product Carbon Output as a Function of Total Carbon ^ 71 Figure 4.13 Output of Total Oxidation Products as a Function of Oxygen Conversion ^  71  Figure 4.14 Conversion as a Function of Temperature for Samarium Oxide . . ^ 73 Figure 4.15 Selectivity as a Function of Temperature for Samarium Oxide ^73 Figure 4.16 Yield as a Function of Temperature for Samarium Oxide ^ 74 Figure 4.17 Methane Conversion for CH 4 /02 = 4 ^ Figure 4.18  C2+ Selectivity  at CH 4 /02 = 4 ^  80 80  Figure 4.19  C2 Yield  for CH4 /02 = 4 ^  81  Figure 4.20 Effect of Catalyst Preparation on % Oxygen Conversion ^ 87 Figure 4.21 Effect of Catalyst Preparation on % Methane Conversion ^ 87 Figure 4.22 Effect of Catalyst Preparation on %  C2 Selectivity  Figure 4.23 Effect of Catalyst Preparation on %  C2 Yield  Figure 4.24  C2+  ^ 89  ^ 89  Selectivity as a Function of Surface Area ^ 102  Figure 4.25 Methane Conversion as a Function of Cation Electronegativity.... 113 Figure 4.26  C2+  Selectivity as a Function of Cation Electronegativity 113  Figure 4.27 Carbonate Formation on Samarium Oxide under a 100% CO 2 Atmosphere ^  122  Figure 4.28 Carbonate Formation on a 1:10 Na:Sm Oxide Catalyst under a 100% CO 2 Atmosphere ^  122  Figure 4.29 Carbonate Formation on 100% Calcium Oxide under a 100% CO 2 Atmosphere ^  124  Figure 4.30 Carbonate Formation on a 1:10 Ca:Sm Oxide Catalyst under a 100% CO2 Atmosphere ^  124  Figure 4.31 Catalyst Performance as a Function of Carbonate Formation (T=650°C) ^  127  Figure 4.32 Catalyst Performance as a Function of Carbonate Formation (T=750°C) ^  127  Figure 4.33 Catalyst Performance as a Function of Carbonate Formation (T=850°C) ^  128  Figure 4.34 Ratio of CO to CO 2 Produced ^  128  Figure 4.35 CO, Yield ^  136 xi  Figure B.1  1:100 Ca:Sm Photograph 192 ^  172  Figure B.2  1:10 Ca:Sm Photograph 193 ^  172  Figure B.3  1:10 Ca:Sm Photograph 194 ^  173  Figure B.4  1:10 Ca:Sm (RP) Photograph 105 ^  173  Figure B.5  1:100 Na:Sm Photograph 195 ^  174  Figure B.6  Sm 203 Photograph 102 ^  174  Figure B.7  1:100 K:Sm Photograph 103 ^  175  Figure B.8  1:100 Mg:Sm Photograph 104 ^  175  Figure C.1  XRD Graph for Cubic Sm 203  177  Figure C.2  XRD Graph for Ca:Sm (1:10) ^  178  Figure C.3  XRD Graph for Na:Sm (1:10) ^  179  Figure C.4  XRD Graph for Ca:Sm (1:10) (RP) ^  180  Figure C.5  XRD Graph for Monoclinic Sm 203  181  ^  ^  xii  NOMENCLATURE  C2^  ethane and ethylene  C3^  propane and propylene  C2+^  C2'S and  C3s  CO ^carbon dioxide and carbon monoxide C2 selectivity^% C2  of reacted methane converted to C 2's  yield^% of methane in feed converted to C2s  C exiting reactor^all carbon species exiting the reactor, as moles carbon (CO, CO2, CH4, C2 H4 , C2H6, and C3) C in products^all carbon containing products, as moles carbon (CO, CO2, C2 H4, C2H6, and C3) CH4 /02^methane to oxygen mole or flow rate ratio AG^Gibbs free energy of reaction OH^heat of reaction methane conversion^% of methane in the feed reacted oxygen conversion^% of oxygen in the feed reacted P*^ fraction, based on pressure or flow, of reacting gases in feed (i.e., (Pai4 FP02)/Ptotal) -  Q  flow rate (mL/min)  STY (space time yield)  the amount of C 2's produced per unit time per unit catalyst weight (p.mol/s/g)  Treaction  reaction temperature  W/F  weight of catalyst divided by feed flowrate (g s/mL)  ACKNOWLEDGEMENTS I am grateful to the various people and institutions who have provided me with the support and assistance necessary to allow me to complete this thesis. There are several people in particular whom I wish to acknowledge. There are numerous people in the Chemical Engineering Department of U.B.C. who provided invaluable assistance throughout the project. In particular, my thesis supervisor, Dr. Clive Brereton, and Dr. Kevin Smith, both provided essential assistance and excellent advice. All the technicians and staff at Chemical Engineering, who provided their work and assistance, are also much appreciated. The support of several B.C. Research personnel was essential to the completion of this thesis. In particular, I would like to thank Mr. Klaus Oehr, who was always generous with his enthusiasm and ideas, and Mrs. Erin Skelton, who prepared the catalysts and was always very helpful and accommodating to my schedule. In addition to providing valuable assistance on the two Autocad drawings, Mr. Donald Livingstone provided his personal support and sometimes much needed encouragement throughout my work on this thesis, for which I am particularly grateful. I also gratefully acknowledge the financial support of British Columbia Research Corporation and CANMET, Energy Mines and Resources, without which this work would not have been possible.  xiv  1.^INTRODUCTION  A process to convert methane directly to ethane and ethylene is desirable in order to avoid the energy intensive steam reforming step currently required for the conversion of methane to more valuable products. The presence of oxygen is required to make the direct conversion of methane to ethane and ethylene thermodynamically favourable. This process, referred to as the oxidative coupling of methane, has received considerable attention in recent years as the most promising approach to add value to natural gas supplies. Currently there are no commercial oxidative coupling reactors. In developmental work, the process is normally carried out in a packed bed bench scale reactor over a metal oxide catalyst, at temperatures of 700°C to 900°C and atmospheric pressure. The progress toward development of a commercial process has been inhibited by the excess production of the non-desirable products, carbon dioxide and carbon monoxide, collectively referred to as CO.. Much work has been carried out on mechanism elucidation and catalyst characterization in order to aid in the understanding of the processes which occur, and to maximize yields and selectivities toward C2 hydrocarbons. However, substantial gaps in the knowledge base still exist, particularly with regard to the effect of catalyst properties on the catalyst mechanism. Although the use of dopants has been found to improve catalyst performance, and the effects have been ascribed to various properties of the dopant and bulk catalyst, there is still much work required on catalyst characterization and the effect on the catalyst mechanism. This thesis concentrated on measurements of the effects of alkali and alkaline earth dopants on the properties of a samarium oxide catalyst, and on the evaluation of the catalyst properties on the oxidative reaction. Samarium oxide has 1  been shown to be a promising catalyst for this reaction.  Several samarium oxide catalysts, doped with alkali, Na and K, and alkaline earth, Mg and Ca, were prepared, with either 1:100 or 1:10 dopant:Sm mole ratio. The dopant, dopant concentration, and catalyst preparation method were varied. The catalysts were tested in a laboratory scale packed bed reactor at conditions of 0.09 seconds residence time in the catalyst bed, atmospheric pressure, reaction temperature of 650°C, 750°C, and 850°C, varying methane to oxygen ratio, and in the presence of a helium diluent. The catalysts were characterized by measuring: their surface area; their surface composition and structure using scanning electron microscopy with electron dispersive x-ray analysis; their bulk crystal type by powder x-ray diffraction analysis; and their ability to form carbonates. The performance of the catalysts was determined based on methane conversion, selectivity to  C2  hydrocarbons, and  C2  yield. These were  correlated with the catalyst surface area, catalyst basicity, carbonate formation tendency, and the ionic radius of the dopant.  1.1^Thesis Objectives  The objectives of this thesis were to accomplish the following:  i)^Design and build a packed bed reactor to minimize pre- and post-catalytic volume, in order to minimize the effect of homogeneous reactions and maximize the sensitivity to catalyst properties.  2  ii)  Test samarium oxide and samarium oxide doped catalysts for the oxidative coupling of methane and compare the effects of various dopants and dopant concentrations.  iii)  Examine the effects of changing the catalyst preparation procedure.  iv)^Characterize the catalysts and determine the effect of various catalyst properties on the catalyst performance, as indicated by methane conversion,  C2  selectivity,  and C2 yield.  3  2.^LITERATURE REVIEW  Methane, the major component of natural gas, is an abundant fossil fuel. With present technology, much of this methane is wasted due to costs associated with transportation and processing. Much of the natural gas is located in remote areas where the costs for transportation to market far exceed its value. However, the reserves of liquid petroleum are diminishing, and the search for alternate sources is increasing.  Of the commercially-established technologies presently available to convert methane to methanol and higher hydrocarbons, the most widely used process is the steam reforming process, which involves the initial formation of synthesis gases (H 2 + CO + CO2):  CH4 + H2O ' 3 H2 + CO AH/ (870°C) = 226 kJ/mol CO + H2O 'rtk H2 + CO2 AH1 (870 ° C) = - 34 kJ/mol  ^  ^  AGf = -63 kJ/mol (2.1)  AG f = 0 kJ/mol^(2.2)  Reaction (2.1) is the main reaction which is highly endothermic and energy intensive. Synthesis gas can be further converted to more valuable products, such as methanol or higher hydrocarbons.  The methods currently available for methane conversion are: natural gas transformation to syngas and syngas conversion to methanol; natural gas transformation to syngas and syngas conversion to hydrocarbons by Fischer-Tropsch (FT) synthesis (a heterogeneous reaction of H2 and CO catalyzed 4  by various metals); and catalytic conversion of methanol to gasoline-type hydrocarbons as well as ethylene and propylene in the MTG (Methanol-To-Gasoline) and MTO (Methanol-to-Olefins) processes.  The above processes are not energy efficient, since they all involve the steam reforming transformation to syngas, which involves the initial cleavage of four C-H bonds per molecule followed by the formation of a C-0 bond. Some of the hydrogens are then restored by further processing, and the oxygen may be removed. A direct methane conversion process which avoids the steam reforming step should be more energy efficient, and therefore is potentially more economically attractive. Capital costs may also be reduced by a reduction in the number of process steps.  The oxidative coupling of methane is a direct conversion process carried out in the presence of oxygen, resulting in the desired products, ethane and ethylene, with small amounts of higher hydrocarbons. Unfortunately, the process also produces substantial amounts of the undesirable products, carbon monoxide and carbon dioxide.  Ethylene is currently produced mainly from the ethane and propane fractions of natural gas, which are recovered from natural gas using cryogenic distillation and other separation techniques. Ethylene is converted to numerous intermediate and end products on a large scale, mainly polymeric materials such as plastics, resins, fibres, and elastomers. Other important products are solvents, surfactants, coatings, plasticizers, and antifreeze.  5  The basic equations for conversion of methane to ethane and ethylene are:  2 CH4 —> C 2 H6 + H2^  AGanc) = +  71 kJ/mol^(2.3)  2 CH4 -4 C 2 H4 + 2 H2^AG(727c) = + 80 kJ/mol^(2.4)  However, these reactions are thermodynamically unfavourable, and the presence of oxygen is required if the reaction is to be carried out at temperatures significantly below 1300°C. When oxygen is added the free energies become strongly negative: ^  AG(7270 = - 121 kJ/mol^(2.5) 2 CH4 + 0.5 02 —) C2H6 ± H20 ^ AG(727c) - 305 kJ/mol^(2.6) 2 CH4 + 02 —) C2H4 4' 2 H20  However, methane, ethane, and ethylene can also be oxidized non-selectively to give carbon oxides. In order to achieve significant conversion to ethane and ethylene, the reaction must be carried out at temperatures in excess of 600°C and either in the presence of a catalyst or under conditions of high pressure. The elevated temperature also encourages formation of CO and CO2 . The maximum C2 yields achievable thus far have been on the order of 30%. The competition between the selective and nonselective reactions, and between heterogeneous and homogeneous reactions, has been extensively studied.  There are two main process methods which have been studied for the oxidative coupling of methane, the cofeed and redox modes. The redox mode involves the reaction of methane with a reducible metal oxide, which donates the oxygen for the 6  oxidative coupling reaction (Kuo, 1991):  2 CH4 + 2/y MO„, y C2 H4 + 2 H2O + 2/y MO,,^(2.7) 2 CH4 + 1/y MO„,), N-• C2H6 + H2O + 1/y MO,,^(2.8)  The reduced metal oxide is subsequently reoxidized by gas phase oxygen:  MO,, + y/2 0 2 vat MO, r+),^  (2.9)  The redox method has been used less than the cofeed method and will not be considered further. The cofeed method has been used for the majority of the studies, and was used in the present study. Methane and the oxidant (most commonly oxygen) are reacted simultaneously, usually in the presence of a solid catalyst which enhances the coupling reaction.  Intensive work has been carried out by many researchers world wide in recent years on the development of the oxidative coupling of methane process. The present work involves evaluation of the catalyst, samarium oxide, promoted with alkali and alkaline earth oxides.  7  2.1 Choice of Doped Samarium Oxide Catalysts for Study  Samarium oxide (Sm 203) and alkali and alkaline earth doped samarium oxide catalysts were chosen for this study because they have been found to be among the most promising active and selective catalysts for the oxidative coupling of methane (Otsuka et  al., 1985, 1986a; Otsuka and Komatsu, 1987; Hutchings et al., 1989c; Korf et a1., 1989).  Samarium is a rare earth element, which is found at its highest natural concentration (3%) in the mineral monazite, in the presence of other rare earth elements (Murthy and Gupta, 1980). Monazite is recovered from placer deposits in Australia, Brazil, India, South Africa and the United States of America, usually as a by-product of rutile, ilmenite and zircon recovery processes. A principle use of samarium is in the production of SmCo 5 permanent magnets.  The catalysis mechanism for the activation of methane is not completely clear. Some form of oxygen species is generally accepted to be the active species but the exact nature of the species is still under discussion, as is the method of catalyst regeneration. The use of dopants has been shown to have a beneficial effect on the performance of the samarium oxide catalyst; however, the important properties of the dopants require further elucidation. The catalytically active sites may be affected by the bulk crystal structure, concentration of surface and bulk oxygen vacancies, the nature of the surface oxygen species, the distribution of dopants within the support crystal, the surface basicity of the catalyst, the presence and/or state of equilibrium of carbonate formation, and the ionic radius of the dopant. 8  2.2 Mechanism of Oxidative Coupling of Methane  For the purpose of designing catalysts and/or reactors with improved yield and selectivity, the mechanism of the reaction has been extensively researched. Although most aspects of the mechanism have been determined, there are still some questions to be answered. Complications stem from difficulties in comparing results from various researchers. Not only are there many different catalysts being used, but the reaction conditions and reactor designs vary widely.  It is generally agreed that the reaction proceeds by hydrogen abstraction from the methane molecule on the catalyst surface, to produce a methyl radical (CH 3 -) which then likely combines with another methyl radical to produce ethane (C2 H6), either on the catalyst surface, or in the gas phase. The ethylene (C 2 H4) may then be produced by dehydrogenation of the ethane, again either on the catalyst surface or in the gas phase. The mechanism of the formation of the non-selective compounds, CO 2 and CO, in the presence of a catalyst has not been clearly defined. The extent of the heterogeneous versus homogeneous reactions to produce CO 2 and CO is a subject of continuing research.  The homogeneous gas phase reactions play a very significant role in the process. This complicates comparison of catalyst performance, because under conditions where the gas phase reactions strongly dominate, the inherent selectivity of the catalyst may be masked. Under such conditions, the product distributions can be very similar over catalysts of very different chemical composition. 9  2.2.1 Formation of Methyl Radicals  The high dissociation energy of the C-H bond (427 kJ/mole) indicates that the abstraction of hydrogen to form the methyl radical is the rate-determining step in the oxidative coupling of methane (Otsuka and Jinno, 1986b). This was confirmed over Sm203 in further studies (Otsuka and Nakajima, 1987), as well as by others over Li/MgO (Cant et a/., 1988), Li/MgO and Sm 203 (Amorebieta and Colussi, 1988, 1989), reducible Na/Mn0x /Si0 2 (Burch et al., 1990), and Li/MgO, SrCO3, and Sm203 (Nelson et al., 1989).  2.2.2 Formation of Ethane  Martin and Mirodatos (1987; Mirodatos and Martin, 1988) reported on evidence of carbene (:CH2) intermediates formed over Li/MgO. The formation of carbene may occur by the abstraction of a proton from methane to form CH3, followed by the abstraction of a hydride ion by Li+ to form carbene. Carbene in the gas phase could insert into a C-H bond of methane to give ethane. An alternative mechanism is based on the catalytic activity of surface peroxide ions, 0 22- . These may be capable of abstracting two H atoms from CH4 to produce carbene. Non-selective oxidation into carbon oxides would proceed on another type of site.  Both of these mechanisms may be disputed based on isotopic studies which were carried out by Otsuka et al. (1989) and Nelson et al. (1989). These indicated that ethane is formed through the coupling of methyl intermediates, either from methyl radicals  10  in the gas phase or methyl groups adsorbed on the surface. Using a feed of a mixture of CH4 and CD4 resulted in ethane only of the forms C 2H6, CD3CH3, and C 2D6 . This rules out the possibility of carbene formation as an intermediate in the reaction. Similar work was also completed by Mims et al. (1989) with similar results. They determined that C3 formation occurs by the terminal addition of methyl radicals to ethylene. Lunsford (1989) determined that the gas phase coupling of surface-generated methyl radicals was a significant pathway for the formation of ethane. They found that the rate of methyl radical formation and the (C 2 yield) "2 followed the same curve.  The net ability of a catalyst to produce methyl radicals depends upon the catalyst's ability to both generate and consume methyl radicals It was estimated that a methyl radical will collide with a surface ca. 105 times before it reacts with another CH 3 radical. Tong et al. (1989) and Tong and Lunsford (1991) studied the reactions of methyl radicals with lanthanide oxides. The methyl radicals reacted extensively with CeO2, Pr 6 0n , and Tb4 07, all of which have multiple cationic oxidation states. The oxides La 203 , Nd 2 03, Sm 203, Eu 203, and Yb203 react with CH 3 - radicals to only a small extent. The former oxides are non-selective for the oxidative coupling of methane reaction. However, the reaction of CeO2 with CH 3 - radicals is strongly inhibited by the addition of Na 2 CO3 and consequently radical production is enhanced.  Feng et al. (1991a, 1991b, 1992) conducted kinetic studies on 1% Sr/La 203 . The results indicated heterogeneous production of CH 3 - and homogeneous loss of CH 3 - by recombination, to C 2 H6 . The small amount of C 2 H4 formed can largely be accounted for by known gas phase processes. No indication was found of any heterogeneous  11  oxidation of methyl radicals to CO ), or of any heterogeneous conversion of methane to C2 compounds.  2.2.3 Formation of Non-Selective Products (CO 3) and the Inherent  C2+ Yield  Limit  The addition of ethane and ethylene to the gas feed of oxidative coupling reactors has little effect on the rate of CO ), production, even in the presence of excess oxygen. This initially led investigators to believe that ethane and ethylene were stable under the reaction conditions and did not contribute to CO ), formation. However, the secondary oxidation of the reaction products was found to contribute significantly to the formation of the carbon oxides using isotopic techniques (Ekstrom and Lapsezewicz, 1988a, 1989a; Ekstrom et al., 1989b). A large proportion of the CO 2 formed during the partial oxidation was derived from the products, particularly ethylene. The portion of CO„ formed from the 13C2 H4 was 85% when this was present at 8 volume % in the feed gas. This clearly indicates that, as the concentrations of CO„ is increasingly derived from the  C2 in  the reactor increase, the  C2 products.  Since the addition of potential reductants in the production of CO„ (ethane and ethylene) does not result in a significant effect on the rate of CO ), production, it can be concluded that the formation of the CO ), is limited by the availability of the oxidant and not of the reductant. Therefore, the reaction is not simply a gas phase oxidation involving molecular oxygen, since this is present in excess, but must involve a species formed by the catalyst, which reacts either on the surface or in the gas phase.  12  A typical feature of the catalytic oxidative coupling of methane is that the maximum C2+ concentration  obtainable is similar for many catalysts of very different nature.  Attempts to increase the C2+ selectivity by changing reaction conditions invariably result in lower methane conversion. The limit in the maximum  C2+ concentration may  be similar for so many catalysts because it is ultimately determined by the kinetics of the activation of the similar molecules, CH 4, C2 H6, and C2H4, by the various catalysts. The reaction rates do not show significant differences for these molecules (Ekstrom et al., 1989b). The c2 + yield of the catalyzed oxidative coupling of methane reaction is limited when the rate of  C2+ formation from CH4  is equal to its rate of conversion to  CO2 . A typical feature of the oxidative coupling of methane reaction is that the  C2+  selectivity decreases as the methane conversion is increased by increasing the oxygen concentration.  Hutchings et al. (1989a, 1989c) tested several catalysts, including Sm203 and Li/MgO, to study the oxidative coupling of methane. Over Sm 203 at 700°C the CO„ was formed largely via the C2 products and C2 H4 is oxidized somewhat faster than C 2H6 to CO„. Over Li/MgO, at TReadion < 700°C, the oxidation of C2 products accounted for less than 10% of the CO„, but at  TReact ion > 740°C, C2 oxidation was responsible for the formation  of 30-80% of the CO„.  The concept of an inherent limit to the C2+ yield of the oxidative coupling of methane reaction, regardless of catalyst activity and selectivity, has been studied by several researchers, both by experimental studies and modelling studies.  13  Geerts et al. (1989) carried out experimental reactivity studies and found that the conversion of ethane to ethylene is much more rapid than ethane combustion, irrespective of the presence of a catalyst, and the main combustion path is via ethylene. Ethane is consumed much more rapidly than methane. Similar conclusions were reached by Roos et al. (1989a, 1989b), who found that neither ethane nor ethylene are stable under reaction conditions over Li/MgO and Ca/Sm 2 03 . The CO. are formed predominantly from ethylene and the reactions can be represented by a sequential reaction, CH 4 --> C2H6 ---> C2H4 --> CO,„  Nelson et a/. (1989) conducted isotope studies over Li/MgO, SrCO 3, and Sm203 and devised a mechanism for the heterogeneous formation of CO. from methyl radicals, as a result of surface reactions of methyl peroxy radicals (CH 302 ):  C H3(g) . + 02(0 leah CH302(g) .  (2.10)  CH302(6) - + catalyst —> CH3 0 (ads) + 0  (2.11)  CH3O(ads) -f CH 2O (in presence of 02) --) CO, + H2 O  (2.12)  The reaction scheme would explain why selectivity declines with increasing 0 2 pressure and decreasing temperature (longer CH 3 04) - lifetime). However, direct reaction of CH3 - with the surface is another possible route to carbon oxides. A purely homogeneous chain reaction involving CH 302 • is not indicated.  Further work supported the methyl peroxy radical as the source of low temperature CO. from CH 4 , and again indicates that reactions are partly heterogeneous (Nelson et 14  al., 1991). On the other hand, the thermal stability of the ethyl peroxy radical, C2HSO2%  is considerably less than that of CH 302 -, and, at temperatures greater than 327°C, reaction (2.13) is the dominant pathway, resulting in a homogeneous path with high selectivity for the conversion of C 2 H6 to C2H4 .  C2145 + 02 -) C 2 H4 + 1102 .  ^ 4  (2.13)  Peil et al. (1990a, 1990b) also found that the catalyst surface is active in the formation of CO, CO 2, and possibly even C2 H6 . Based on studies over alkali promoted alkaline earth oxide catalysts, Aparicio et al. (1991) concluded that gas phase C., combustion alone cannot account for the observed  C2  yield limit, and surface catalyzed  C2  combustion may play an important role in determining selectivity.  Kennedy and Cant (1991) found that ethane was four times as reactive as methane over the rare earth oxides, La203 , CeO2 , Sm203, and Pr60 11 , under conditions of 13% hydrocarbon, 3.5% oxygen, with the balance helium, at 1 atmosphere pressure, and a gas phase residence time of 0.2 seconds. The selectivities of the catalysts were dependent on the average lifetime of alkyl and alkylperoxy species in the gas phase and the ability of the catalyst to oxidize alkyl radicals. The strong oxidizing power of the Ce and Pr oxides is used to explain the very low ratios of carbon monoxide to carbon dioxide found for these catalysts.  Homogeneous chain-branching reactions, initiated by the reaction of oxygen with heterogeneously formed methyl radicals, may participate in both the selective and non15  selective oxidation of CH 4 (Lunsford, 1990). It is likely that, in the presence of 02, chain-branching reactions in the gas phase result in a multiplication of the surfacegenerated radicals. The chain-branching reactions are likely partially responsible for the conversion of C 2 H6 to C2H4 and for the formation of CO), (Lunsford, 1991). Secondary reactions of CH 3 - radicals with the metal oxide also may contribute to the formation of COX .  Grzybek and Baerns (1991) conclude that the total oxidation of methane occurs at least partly as a surface process in the presence of gas phase oxygen, with the initial step the surface oxidation of methane or methyl radical to  type products, followed by surface oxidation to compounds containing a C=0 group, possibly H2CO, followed by surface oxidation to surface carbonate or CO 2 by oxygen adsorbed from the gas phase.  Many kinetic models have been developed based on different reaction schemes. The model developed by Machin (1992) took into account other models and developed a simplified model consisting of seven elementary steps, 5 homogeneous and 2  heterogeneous, with one adjustable parameter to match the model results with experimental data. An independent correlation between the activation energy of the rate determining step, the adjustable parameter, and the physical-chemical properties 16  of the solid catalyst was also developed. The rate determining step is assumed to be the H abstraction by an active oxygen species leading to a hydroxyl group on the surface of the catalyst and a methyl radical in the gas phase. The formation of ethane is accounted for by the recombination of methyl radicals in the gas phase. Ethylene is formed in the gas phase from radicals produced by active catalyst sites. It should be noted that various mechanisms have been proposed by various researchers, and that all are not consistent with this mechanism. The 7 reactions are as follows:  CH4 + 0: —> CH3 - + Os—H  (2.14)  2 CH3 - —> C 2H6  (2.15)  C2 H6 + 0: —> O S -H+ (C 2 H5 - —> C2H6)  (2.16)  C2H6 + CH3* - CH4 + (C2H3 • ''-4 C2H4)  (2.17)  CH4 + 02 -) CO,  (2.18)  C 2 H 6 + 0 2 4 C 0x  (2.19)  02 -> CO,  (2.20)  -  C2H4 +  A limit to the C2+ yield (<33%) is predicted to exist under conditions of high temperature (>700°C) and one atmosphere pressure. The low C2+ yield is believed to be due to the fact that the temperature of the oxidative coupling process is higher than the autoignition temperature for methane and ethane.  Based on a kinetic model of 9 to 11 equations, Labinger (1988, 1991) found an upper limit of approximately 30% yield of higher hydrocarbons at one atmosphere and about 25% yield at pressures of greater than one atmosphere. Hair et al. (1992) carried out 17  modelling of oxidative coupling of methane, and showed that when the selective catalytic reaction:  CH4 + 1/4 02 4 CH3 + 1/2 H2O^(2.21) -  is added to the gas phase kinetic mechanism, the predicted maximum C2 yield is 84%, which occurs when the rate constant of reaction (2.21) is quite high. This is the scenario which would result for a very active catalyst which selectively activated the methane according to reaction (2.21). However, since the  C2  yield which is actually  achieved in practice is much lower (<30%), the catalysts used must either not achieve a high rate constant, or they must contribute to the non-selective reactions. It is likely that both of these factors contribute.  2.2.4 Gas Phase Reactions  The gas phase reactions have been studied both in the presence and the absence of a catalyst in order to determine the relative importance of these reactions.  Yates and Zlotin (1988) studied the rate of oxidative coupling of methane over MgO and Li/MgO in comparison to blank reactor studies. Significant quantities of ethane, ethylene, and carbon monoxide were formed in the absence of a catalyst (1.0%, 1.4%, and 18.1%, respectively at 700°C, CH 4 /02 = 1.8, and He = 56%). The presence of these catalysts resulted in a change in the principle combustion product from CO to CO 2 , as was also found by Geerts et al. (1989). 18  Hutchings et al. (1989a) found that at temperatures _^ 700°C the gas phase nature of the overall reaction is dominant, at reaction conditions of CH 4 /02 = 3 and total pressure of 85 kPa.  Kalenik and Wolf (1990) studied the effect of gas phase reactions and presented a discussion aimed at clarifying different interpretations of their results. The reaction conditions must be considered carefully to avoid conflicting conclusions. For residence times longer than 0.1 min, methane conversion is significant and may contribute appreciably to the results obtained in the presence of a catalyst. On the other hand, when the feed is diluted with helium such that reactant partial pressure P*=0.3, methane conversion did not exceed 5% even at long residence times. Catalytic studies conducted under highly diluted feeds are therefore not affected by the gas phase initiated reactions.  Olsbye and Desgrandchamps (1991) studied the effects of the pre- and post-catalytic reactor volume, using a 25% Ba/La 2O3 catalyst, at 830°C and CH 4 /02 = 10. The precatalytic volume was of negligible importance under catalytic conditions, indicating that, under these conditions, the conversion of reactant gases is not influenced by the homogeneous gas phase. At complete oxygen conversion, only the conversion of C2 H6 to C2 H4 occurred in the gas phase, while at lower oxygen conversions, the selectivities changed towards CO„ products. A larger post-catalytic volume led to a lower methane conversion. This is probably due to the formation of CH 4 from the products in the post-catalytic gas phase. This interpretation is supported by the observation that the C 2 /COX ratio decreased with increasing post-catalytic volume even at complete oxygen 19  conversion.  2.2.5 Oxidative Coupling of Methane Under Conditions of High Pressure  The commercial implementation of the oxidative coupling of methane process is likely to be carried out at pressures in excess of atmospheric pressure, under which the majority of the testing has been done. Some of the investigations which have been carried out to determine the effect of increased pressures are as follows.  Asami et al. (1987) studied the homogeneous oxidative coupling of methane at pressures up to 1.6 MPa in the temperature range 650°C to 859°C, and under dilute conditions (CH4 : 02: N2 mole ratio = 14: 1.6: 84.4). An increase in pressure resulted in increased methane conversion and decreased hydrocarbon selectivity. A C2+ yield of  ca. 4.5% was obtained. The C2+ selectivity decreased from 60% (at 0.35 MPa) to 45% (at 1.6 MPa).  Ekstrom et al. (1990) found that the importance of the non-catalyzed reactions was significantly increased at higher pressures. At 0.4 to 0.6 MPa and low gas velocities, the only effect of the catalyst was to alter the CO/CO 2 ratio.  Pinabiau-Carlier et al. (1991) studied the effect of total pressure over a strontium doped lanthanum oxycarbonate catalyst, using nondiluted methane/oxygen mixtures at pressures up to 7.5 bar (0.75 MPa). The increased pressure had a negative effect on both methane conversion and C2+ selectivity. The C2+ yield decreased from 9.9% at  20  atmospheric pressure to 3.4% at 7.5 bar. The negative pressure influence on the oxidative coupling of methane reaction can be minimized by appreciably increasing the linear space velocity for pressures up to 3 bar; under these conditions the C2+ yield increased to 10.3%.  At a pressure of 585 kPa and temperatures of 500°C to 575°C, the oxidative coupling of methane gives similar product yields in the presence of Li/MgO or Sm 203, and in the absence of catalyst, indicating that at this pressure, sufficient gas phase radicals to initiate the reaction are readily formed (Hutchings et a/., 1988, 1989a). The role of the oxide catalyst may involve the control of selectivity within the hydrocarbon fraction,  i.e., the alkene/alkane ratio.  Walsh et al. (1992a, 1992b, 1992c) conducted studies at much higher pressures (900 psig or 6.3 MPa) than other studies, and at temperatures of 550°C to 600°C. Increasing the oxygen partial pressure and/or temperature favoured the production of C2+ hydrocarbons, while reducing these parameters favoured the production of methanol.  Experiments conducted at ca. 1 second contact time show that, as pressure increases from 3 to 10 MPa, the temperature required for complete consumption of oxygen declines by more than 100°C (from 630°C to 515°C) Both methane conversion and C2 + selectivity increase with increasing total pressure (up to ca. 40% selectivity at 15% CH 4 conversion). The decreasing influence of a known oxidative coupling catalyst, Sm 203, with increasing pressure is consistent with the view that the catalyst is primarily a radical generator in conventional catalytic oxidative coupling.  21  2.3 The Role of Gas Phase, Surface, and Lattice Oxygen  The active species for methane activation is an oxygen species, the exact nature of which is not conclusively known despite extensive studies.  2.3.1 Source of the Active Oxygen Species  Several studies have shown that the presence of gas phase oxygen is necessary for the activation of the catalyst (Lin et a/., 1986), and that lattice oxygen is not active in the absence of gas phase oxygen (Ekstrom and Lapszewicz, 1988b, 1988c, 1989a; Kalenik and Wolf, 1991a, 1991b). However, the actual active species may be either an adsorbed or lattice species. Amorebieta and Colussi (1988, 1989) determined that, although gas phase oxygen must be present, the active sites are an adsorbed form of oxygen. Ekstrom and Lapszewicz (1988b, 1988c, 1989a) determined that the lattice oxygen atoms exchange easily with the gas phase molecules, and that the lattice oxygen does not participate in the formation of the reaction products in the absence of molecular oxygen. Although the rate of lattice oxygen exchange determines the catalyst activity, the molecular gas phase oxygen is somehow involved. Lo et a/. (1988) concluded that only surface oxygen species are used for the formation of C 2's over Sm 2 03; however, Peil et al. (1989) concluded that lattice oxygen also contributed significantly to the reaction, for both Li/MgO and Sm 2 03 . Otsuka and Said (1987) determined that the activity of the adsorbed oxygen for converting methane was more than three orders of magnitude greater than that of the lattice oxygen.  22  Kalenik and Wolf (1991a, 1991b) conducted isotope studies, in which  1802  replaced the  oxygen in an oxidative coupling reactor over a catalyst which contained 1602 as lattice and surface atoms, for both La 203 and Li/La 203 2Ti0 2 . The surface and bulk oxygen appeared in the reaction products before gas phase 7802, indicating that lattice and adsorbed oxygen were responsible for methane activation. Gas/solid exchange involved over 50% of the lattice oxygen.  Comparison of La 203 and Li/MgO revealed different behaviour when gas phase oxygen was removed after treatment of the catalyst in oxygen (Lin et a/., 1986). For Li/MgO, residual activity remained for a period of minutes after the oxygen was removed, while for La 2 03 , no such residual activity was seen. The difference may be due to the fact that the active species for the Li/MgO may be a part of the lattice; whereas, with La 2 03, the active species may be a surface species (e.g., 02). The La203 is less selective than the Li/MgO, probably due to further oxidation of the  C2  compounds.  Statman et al. (1991) studied the oxidative coupling of methane on Ba/Sr/Sm 2 03 using TAP (Temporal Analysis of Products). Methane was either not adsorbed, or weakly adsorbed on the catalyst. Oxygen was strongly adsorbed at temperatures above 500°C, which suggests incorporation into the lattice with possible formation of surface or subsurface anions. They concluded that the rate of ethane production depends upon the rate at which active surface oxygen species were formed from gas phase molecular oxygen. The formation of ethylene appears to occur in series with ethane, and may possibly involve the same surface oxygen species responsible for the formation of 23  surface CH3 • radicals.  Ekstrom (1992) studied the exchange of lattice oxygen for Sm 203 using isotope switching experiments (see Table 2.1). The results indicate that large amounts of oxygen are exchanged with an oxygen pool present in the catalyst, comparable to the total number of oxygen atoms present in the Sm 203, and that the exchange must occur, at least partially, on the catalyst surface. The presence of LiCO3 on the catalyst surface decreased the rate of the gas-lattice oxygen exchange by about an order of magnitude, with a similar effect observed for the catalyst activity, suggesting a link between catalyst activity and the rate of gas-lattice oxygen exchange.  Table 2.1 Comparison of Catalyst Activity and Oxygen Exchange Parameters for Cubic and Monoclinic Sm 203 Parameter  Cubic  Monoclinic  Total "surface" 160 atoms desorbing per gram of catalyst (x 10 20) 1  13  0.2  Lattice oxygen diffusivity (cm 2 /s x 10 17) 1  13,600  886  CH4 conversion rate (p.mol/s/g) 2  102  6  Fraction of lattice oxygen exchange during 30s 1802 pulse2  0.6  0.09  -  The role of lattice oxygen was investigated by first exchanging the lattice oxygens with 18u,2,followed by introduction of methane into the Sm 2 03 catalyst bed (Ekstrom, 1992). No reaction was observed, indicating that the lattice oxygen atoms played no role in 'Pell et al., 1992 2  Ekstrom, 1992 24  the formation of the reaction products in the absence of molecular oxygen. Addition of 1602 v immediately led to the formation of 180 labelled CO2, indicating that the lattice oxygens are responsible for CO 2 formation.  2.3.2 Nature of the Active Oxygen Species  Several oxygen species have been suggested to be responsible for the activation of methane in the oxidative coupling of methane. Although many researchers agree that O" is the active species, other species, such as 02 or 0 22-, have also been implicated. Another question which is still under debate is whether the active sites are identical for both the selective and non-selective oxidation of methane. A complicating factor when comparing various studies is that different catalysts may react according to different mechanisms, and the method of preparation may even have a significant effect.  The Li/MgO catalyst has been extensively studied and the accepted active species for this catalyst is a. This species may be formed by two methods. In undoped MgO, intrinsic cation vacancies can react with molecular oxygen to give an a centre (Driscoll et al., 1985). However, MgO has low C2+ selectivity and activity and the addition of  substitutional Li+ ions causes a large increase in the activity. The Li+ ions are believed to react with molecular oxygen to form a [Li+0 - ] centre. A good correlation was found between the amount of methyl radicals formed and the [Li+0 - ] centres formed as a function of lithium doping.  Li2 CO3 is formed on the surface of Li/MgO under reaction conditions. Korf et al. 25  (1990a) found that the active sites are created by the gradual loss of carbon dioxide from the surface carbonate species in the presence of oxygen, and that the presence of lithium carbonate on the surface of the catalyst is therefore crucial for its activity. Peng et al. (1990) observed both Li -40- and Li 2CO3 on the surface of Li/MgO under reaction  conditions. The correlation between methane conversion and the surface Li+0  -  concentration demonstrated that [Li+01 species were the active centres for methane conversion.  Hutchings et al. (1989a) used N20 and NO as the oxidant over several catalysts, including Sm203 and Li/MgO, and the results were interpreted to indicate that  o  - ()  is  the oxidizing species responsible for CH 3 • radical formation. A second surface diatomic oxygen species was indicated which was responsible for non-selective oxidation of ethylene and ethane (Hutchings et al., 1989b, 1989c). This species is established more slowly than the selective oxidizing species.  Aparicio et al. (1991) studied alkali promoted alkaline earth oxide catalysts (e.g., Li/MgO) and found that the active sites appear to be produced via two elementary steps. Alkali metals are known to form stable bulk peroxides, and a surface peroxide species formed by adsorption of gaseous diatomic oxygen may dissociate to form the methane-activating sites. The cleavage of the 0-0 bond is thought to determine the overall activity by forming the active G ions.  Otsuka et al. (1986a) conducted studies over Sm 20 3 and initially proposed that the active species were G ions produced on oxygen vacancy sites or basic sites. The deep 26  oxidation of methane may be caused by surface 0 2- or by adsorbed oxygen. The large difference in activation energies for the selective and non-selective reactions supports the idea that the reaction intermediates of the oxygen species responsible for these reaction paths are different. Further kinetic studies supported the mechanism of adsorption of the methane and oxygen independently on different active sites (Otsuka and Jinno, 1986b; Otsuka and Nakajima, 1987). Evidence suggested that adsorbed diatomic oxygen species were more likely responsible for methane activation over rare earth metal oxides and alkali metal promoted oxides (Otsuka et al., 1987). Peroxide anions (022-) were indicated, while 02 ions were not. Studies conducted with sodium peroxide at low temperatures (327°C to 377°C) suggest that the activation of alkanes is caused only by peroxide anions, but the selectivities to further reactions of the surface alkyl groups formed are affected strongly by the presence of gaseous oxygen (Otsuka et al., 1990).  Yamashita et al. (1991) studied the formation and decomposition behaviours of surface oxygen species on BaO/La 203 . They found 0 22- species on the surface of the La 203, and an increased concentration on 15%Ba/La 203 (designated as 15BLO) on which the barium species were well dispersed. The high catalytic ability of 15BLO was ascribed to this dispersed species on the surface. However, increasing the concentration of Ba to 50% resulted in a significant decrease in C2 yield from that of 15BLO, to about the same value as that of undoped La 203 . This may be due to poor dispersion of the barium species. It was concluded that this indicates that the actual active oxygen species is probably formed from the decomposition of the surface 0 2 2", probably to 20 - . The importance of barium dispersion may relate to this step of 0 22- splitting. The 27  temperature region which corresponds to a high catalytic activity roughly coincides with the critical temperature for the formation and decomposition of barium peroxide (ca. 600°C to 800°C). The dispersed Ba0 2 may act as a mediator to transfer 0 22- or a  ions to La203 active sites. 022- on aggregated Ba0 2 may be too stable to be decomposed into active oxygen species.  Spinicci (1991) studied zinc oxide and zinc oxide based catalysts, and concluded that methane reacts through two distinct pathways with different rates, according to the surface oxygen species which contribute to the process of activation. Strong adsorption of CH3 • species, possibly by the less basic a or 022-, may lead to CO. formation at a low rate, whereas weaker adsorption of methane, on the more basic 0 2-, may lead to C2 formation, at a higher rate. The presence of gas phase oxygen is required to slowly  regenerate the active sites via a continuous adsorption of oxygen.  The 02 ion was found on the La203 catalyst, but there was no evidence for the  a ion  (Lin et al., 1986). The active species may be the 02 ion, or it may be a transient  a  species. The proposed sequence of oxygen species is as follows:  ° 2(g) ° 2(ads)  +e +e +2e 20 21—> 20 2  (2.22)  Amorebieta and Colussi (1988, 1989) determined that the kinetically active species are produced by reversible dissociative chemisorption of 0 2(g) on the ionic surface of both Sm203 and Li/MgO. The presence of gaseous oxygen is required to maintain the active sites, and these will disappear upon its removal. Oxide (0) or superoxide (02) ions, 28  as well as [Li+01 centres, are kinetically relevant species on this catalyst, but peroxide ions (022) are specifically excluded.  Ekstrom and Lapszewicz (1988b, 1988c, 1989a; Ekstrom, 1992) conducted isotope studies using labelled gas phase oxygen over Sm 203, Li/Sm203, and Pr60 11 , and found that the rate determining step is the desorption of molecular oxygen. These results suggest that the lattice oxygen atoms themselves cannot be a significant source of reactant oxygen and that molecular, gas phase oxygen is somehow involved. A mechanism based on the formation and reactions of [01 (an oxygen atom trapped in a vacancy) species is proposed for the C2+ products, but a different form of activated oxygen appears to be responsible for the formation of the carbon oxides.  The mechanism developed by Ekstrom and Lapszewicz for the formation of the active oxygen species for the formation of  C2  products is shown below.  anion vacancy [01^oxygen trapped in the vacancy {-0-}^lattice oxygen  18028as +  (surface) ieA 18n `-'2(surface) v.  18 °2(surface) + 2[1^  [ 1801  ±  2[1801  t u }lattice^(- 180—Lattice + [ 160-}  (2.23) (2.24) (2.25)  29  The active oxygen species is [a], which abstracts hydrogen from the methane molecule to form the methyl radical. In this mechanism, the exchange reaction between gas phase and lattice oxygen is necessary for the formation of the active species.  The rate determining step is the desorption of molecular oxygen from the surface (reverse of reaction (2.23)). This mechanism is in accordance with the fact that the [01 species is known to be highly reactive toward C-H bond cleavage, and it provides the necessary link between the rate of oxygen exchange and the rate of reaction. The anion vacancy is regenerated by the formation of water.  Ekstrom and Lapszewicz suggest that two oxygen species, [A] and [B], are formed from 02 adsorbed on the surface, whose concentration is very small. Hydrocarbons react rapidly with species [A] to give the corresponding alkyl radicals. Species [B] reacts rapidly with the alkyl radicals to give carbon oxides. [A] may be [01, but the nature of [B] is not known, but it may be 0 atoms, OH- or H0 2 • radicals.  For Sm203, Peil et al. (1989, 1990b) determined that on the order of 50% of the Sm 203 working surface was involved in the formation of products. Of this 50%, approximately 60% was involved in CO. formation, with the remaining 40% active for C2  formation. The sites active for  C2  formation had a lower activity than sites active  for the formation of CO.. Further studies using steady state isotopic transient kinetic analysis (SSITKA) on MgO, Li/MgO and Sm2 03 indicated that all sites are equally active for both selective and non-selective oxidation, and that whether selective or nonselective oxidation occurs may depend on the oxidative/reductive state of the active 30  site (Peil et a/., 1991a, 1991b).  A model of the catalyst was developed with the oxygen considered as existing in three regions, i) the physical surface at which exchange between the gas phase and the solid occurred, ii) several subsurface atomic layers readily available for exchange, and iii) the bulk oxide.  Otsuka and Hatano (1992) suggest that the adsorbed oxygen species, as opposed to the lattice oxygen, are the active species for the oxidative coupling of methane reaction. They also suggest that the active species is not a monatomic oxygen, as commonly believed, but is diatomic, such as 02, 0 22-, or chemisorbed 02 . Based on electron spin resonance studies (ESR), they conclude that 022- anions must be responsible for the activation of methane over a sodium peroxide (Na 202) sample, and is also likely responsible for the catalytic activity of Sm 203 . However, 022- can be regarded as a dimer of 0-, (i.e., a--co, and it has been observed that a does form from 0 22(Hutchings and Scurrel, 1992). Therefore, several authors have concluded that 0 - is the selective oxidizing species for Sm 203 .  2.3.3 Crystal Lattice Oxygen Mobility  The mobility of the lattice oxygen ions and the rate of lattice oxygen exchange with gas phase oxygen, have been shown to determine the rate of methane conversion (Ekstrom and Lapszewicz, 1988b, 1988c, 1989a).  31  The addition of Li to MgO results in an increase in the lattice oxygen mobility, with a corresponding decrease in the activation energy for diffusion from 63.5 kcal/mole for pure MgO to 14.6 kcal/mole for Li/MgO (Peil et a/., 1991a, 1991b). Similar results were found by Kalenik and Wolf (1991a, 1991b) for Sr doped La 203 . The addition of the lower valence cation to the bulk oxide may create lattice defects in the crystal, which result in higher oxygen ion mobility. The oxygen vacancies formed by this doping can react with gaseous 0 2, resulting in the formation of 0' ions.  This effect was also found for perovskite-type catalysts, which are known to be ionic conductors by means of 0' anions. Vermeiren et al. (1991) found this for catalysts consisting of alkaline earth metal elements and Ti, Zr and Ce, and Alcock et a/. (1992) found this for La osSr0.2Y02.9 and La0.9Sro.1 Y01.43 catalysts. Oxides with higher oxygen vacancy concentration were found to be more active for the oxidative coupling of methane. Substitution of part of one of the perovskite component elements considerably improves both the catalytic conversion and C2+ selectivity. The positive charge deficiency in the lattice created by substitution is neutralized by the formation of oxygen anion vacancies, resulting in an increase in the rate of oxygen isotope exchange. This manifests itself in the fast regeneration of the surface oxygen vacancies and surface 0 2" .  Kalenik and Wolf (1992) studied the correlation of oxygen available for isotope exchange with catalyst performance for the oxidative coupling of methane. The effect of doping lanthanum oxide with strontium and doping zirconium and thorium oxides with calcium was studied. An increase in methane conversion resulted from promotion 32  of the oxides with the dopants of lower valence, of similar ionic radii.  A low C2+ selectivity was observed for zirconium dioxide, which the authors concluded was due to the formation of surface carbonates, resulting in a decrease in the number of lattice defects and decreasing the capability of the oxygen atoms to diffuse through the carbonate. The undoped zirconium dioxide exhibited low oxygen exchange capabilities which were significantly increased by doping with calcium. However, in both cases, the C2+ selectivity remained very low, indicating that, in this case, the  C2+  selectivity is determined by more than the oxygen exchange capability of the catalyst.  Borchert, Zhang, and Baerns (1992) investigated the effect of adding cations of different valence to lanthanum oxide. They report that at temperatures greater than 327°C and oxygen partial pressures above 10' bar, lanthanum oxide exhibits conduction consisting of a minor contribution from ionic conductivity due to oxygen vacancies and of a major contribution from defect electron conductivity. They also report on the results of another study, in which the C2 selectivity and oxygen ion conductivity were compared for CaO-Ce02 catalysts of various compositions. It was found that the dependence of oxygen ion conductivity and of  C2 selectivity on CaO content follow a similar pattern,  both having a maximum around 20 to 25 mole % CaO.  The addition of cations of a different valence to a metal oxide should affect the number of oxygen ion vacancies present. Addition of lower valence dopants results in a negative effective charge which may be compensated for by the formation of positive defects, such as oxygen ion vacancies. Higher valence dopants would result in a 33  decrease in oxygen ion vacancies. The increase in oxygen ion vacancies should result in an increase in ionic conductivity. Lanthanum oxide (La 203) was doped with cations of differing valence and tested for the oxidative coupling of methane reaction (Borchert, Zhang, and Baerns, 1992). The addition of Nb sf had no effect, which may be due to lack of incorporation into the lattice, and addition of Ti 4+ decreased selectivity. Addition of Zn' and Sr' both increased selectivity, with Sr" being significantly higher.  The oxygen ion conductivities of the catalysts at reaction temperature were compared to the C2+ selectivities. There appears to be a direct relationship between an increase in oxygen ion conductivity and an increase in C2+ selectivity, based on these four catalysts.  2.3.4 Concentration of Active Sites on the Catalyst Surface  McCarty (1991) reported on a model of the oxidative coupling of methane (kinetic and thermodynamic), which includes 134 reversible homogeneous reactions and a set of heterogeneous reactions. The model indicated that an optimum concentration of active sites exists, and that too many sites is detrimental. The C2+ selectivity increased with the increase in reactive centre concentration to reach a maximum, and then decreased. The methane conversion increased steadily with the increase in active site concentration. Higher than optimal surface concentrations of oxygen result in greater methane conversion rates, but with very low coupling selectivity, due to rapid heterogeneous oxidation of intermediate methyl radicals. Lower than optimal concentrations of reactive surface oxygen leads to both lower conversion and lower C2+  34  selectivity because homogeneous methyl radical oxidation processes compete more effectively with the second order coupling reaction under these conditions. Selectivity is predicted to decline quickly with increased oxygen partial pressure because of direct oxidation of the methyl radicals.  2.4 Samarium Oxide Crystal Structure  Samarium oxide exists in one of two forms: the metastable cubic structure is changed irreversibly into the stable monoclinic structure at around 850°C to 900°C. The cubic structure has been shown to be more active as an oxidative coupling catalyst than the monoclinic structure, and this is believed to be due to the increased oxygen mobility in the cubic structure. In the cubic structure, the samarium atom is surrounded by six oxygen atoms and two lattice oxygen vacancies at the eight equidistant corners of a cube (see Figure 2.1) (Anshits et a1., 1990; Peil et al., 1992). The oxygen vacancies, which are located either on the face diagonals or body diagonals, are ordered in such a way that the oxygen atom vacancies lie in a straight line through the body diagonal of the cubic crystal. This configuration essentially provides a "pipeline" for the oxygen atoms to diffuse into and out of the bulk crystal structure. The monoclinic structure is identical to the cubic structure, with an additional oxygen atom along a threefold axis. This seventh oxygen atom results in a distorted octahedral coordination about the samarium atoms. Not all monoclinic unit cells contain an "extra" oxygen atom, so the structure consists of both sixfold and sevenfold coordination about the metal atoms. The symmetry about the metal atom in the cubic crystal compared to the monoclinic results in a higher degree of lattice oxygen mobility.  35  Figure 2.1 The Crystal Structure of Cubic Samarium Oxide Sm 203 1  0 - Sm 0 - 0  l  ^  Oxygen Vacancy  Anshits et al., 1990 36  2.4.1 Comparison of Cubic versus Monoclinic Sm 203  The ability of the cubic and monoclinic crystals to exchange oxygen has been studied and compared to the catalyst activity (see Table 2.1) (Ekstrom, 1992; Peil et al., 1992). The cubic Sm 203 was capable of faster exchange, and had more oxygen atoms available for exchange. This correlated with the increased catalyst activity observed for the cubic structure.  The two forms have the same basicity, which allows important comparisons to be made without the influence of this variable. The only apparent significant difference between the two lattice structures is that the cubic phase has more oxygen vacancies.  A mixture of cubic and monoclinic phases has been shown to yield a better C2 + selectivity than the cubic or monoclinic alone (Sokolovskii et al., 1990; Anshits et al., 1991). This has been shown both for undoped Sm 203 and Ca/Sm 203 . Electrical conductivity measurements indicated that the oxygen vacancy concentration and their mobility are directly related to catalytic activity and C2+ selectivity in the oxidative coupling of methane. The mixed phase structure had the highest mobility of oxygen vacancies, which may be due to additional defects created in the interface between the cubic and monoclinic phase.  Korf et al. (1991) compared cubic and monoclinic Sm 2 03 and La203 , and developed a possible reaction scheme for the formation of ethane and CO, in the oxidative coupling of methane. In this scheme, two adsorbed methyl groups must combine to form 37  gaseous ethane. The structure of the cubic modification of Sm 203 is less closely packed than that of the monoclinic form and may present a more favourable geometry for this reaction. The ethane must readsorb on the surface, to form the secondary product, ethylene. The CO. is formed by interaction of CH XO species (formed from adsorbed methane and an adsorbed oxygen species) with surface oxygen, rather than directly from methane or from surface CH 3 • species.  Finally, the difference in surface geometry between cubic and monoclinic forms may also give a change in the nature of the oxygen surface species responsible for the formation of the CHX O intermediates and hence a change in the kinetics of this step. The production of CO), probably occurs via different surface species than that of the C2 products.  2.5 Effect of Water Addition on the Oxidative Coupling of Methane  Choudhary et al. (1991) studied the effect of water addition on a non-catalytic oxidative coupling system. Addition of water was highly beneficial to obtaining higher C2 + yield and/or selectivity. At 900°C and CH 4 /02 =3.5, with 50% nitrogen in the feed, the C2+ yield and selectivity were 6.1% and 49.3%, respectively. With 50% steam in the feed, the C 2 + yield and C2+ selectivity both improved, to 8.0% and 69.2%, respectively.  K/Ca/Ni oxide catalysts have attracted some attention due to very high selectivities (90-95%, with a methane conversion of ca. 10%) obtained at low temperatures (600°C) and in the presence of water (Pereira et al., 1990; Rasko et al., 1992). The presence of 38  the water was essential to obtain the high selectivity. Pereira found that, in the presence of steam, considerably higher C2+ selectivity and yield (90 and 9.9% compared to 15 and 0.3% without steam, respectively) were obtained after 400 minutes on stream. The method of catalyst preparation was also found to be important. Dooley and Ross (1992) found that steam addition resulted in an increased yield over a K/Ca/Ni oxide catalyst at 570°C to 650°C. However, they did not obtain the high selectivity obtained by the other researchers, and they obtained higher methane conversion. For some catalyst formulations, they did obtain high selectivities, but these did not last for an extended period.  Buyevskaya et al. (1992) studied the effect of dilution of the feed with steam over several catalysts. Steam mole percentages of 50-70% at 700-850°C and CH 4 /02 =1.5-4.9 were tested. The effect of steam addition was dependent on the methane to oxygen ratio. With 10% Sm203 /Mg0 as the catalyst, at 850°C and CH 4 /02 =4, the C2 yield decreased slightly from 19.6% with helium as the diluent, to 19.3% with steam as the diluent. However, at CH4 /02 =2.3 and with other conditions the same as the previously mentioned case, both ethylene and ethane yields increased in the presence of steam. The ethylene yield was particularly affected, increasing from 9% without steam to 14% with steam.  Olsbye et al. (1992) studied the addition of steam over a 25% Ba/La 203 catalyst and found that C2+ selectivity was not influenced, but the CH 4 conversion fell to 74% of its initial value when 20 mole % H 2 O was added to the feed.  39  The effect of water addition seems to be very dependent on both the catalyst and the reaction conditions. With low methane to oxygen ratios the addition of water appears to positively influence results. Van der Wiele et al. (1992) studied the influence of water for a residence time of 0.032 seconds and a reaction temperature of 800°C. The only effects of the water were to increase the CO2 formation slightly, with a corresponding drop in C2 selectivity.  2.6^Potential Reactor Configurations  Most of the laboratory studies on the oxidative coupling of methane have been carried out in packed bed reactors. This is not a suitable configuration for practical processing due to the high exothermicity of the reactions, which set up large temperature gradients in the catalyst bed. At 25% methane conversion, the oxidative coupling of methane reaction over magnesia based catalysts has an adiabatic temperature rise of 1250°C (Leyshon, 1991). A large-scale fixed-bed reactor would require operation as a multi-bed unit with staged oxygen addition (Edwards and Tyler, 1988). Various reactor configurations have been tested for their efficacy in the oxidative coupling of methane.  Aigler and Lunsford (1991) tested MgO and Li/MgO monolith type catalysts and found that the effectiveness of both types of monolith is not as great as the same materials in a conventional packed-bed.  Fluidized bed reactors are a promising configuration for the reaction due to the ability of the reactor to disperse the heat of the reaction. However, the stability of the 40  catalysts and their tendency to agglomerate cause operational problems with this configuration. The effect of catalyst agglomeration for Na 2CO3 /CaO catalysts was overcome by admixing a-Al 203 particles with the catalyst (Andorf et al., 1991) . The  C2+ selectivity obtained was lower than in previous work for similar catalysts, indicating that further study is necessary to optimize use of the admixed material.  Edwards et al. (1990) compared the oxidative coupling of methane reaction over Li/MgO using both fixed and fluidized bed reactors. In the fixed bed reactor, methane conversion was limited to 15% due to the need to avoid excessive temperature gradients. The fluidized bed reactor operated essentially isothermally at methane conversions in excess of 40%. However, some loss in C2+ selectivity occurred in translating from fixed to fluidized bed reactors. The performance of the fluidized bed reactor with 5-10% ethane in the feed was significantly improved with respect to ethylene production compared with the case where the feed was methane and oxygen alone. These studies led to the development of the 'OXCO' process, which consists of a fluidized bed reactor to which ethane is injected into the oxygen free zone of the fluidized bed (Edwards et al., 1991). Methane coupling and the pyrolysis of higher alkanes are efficiently combined within a single fluidized bed reactor resulting in the total utilization of natural gas. With Australian natural gas (36% of carbon content is ethane and higher alkanes), the overall selectivity is potentially 85% for conversion to unsaturates.  Molten materials have been used as catalysts for the oxidative coupling of methane. Metal oxides dissolved or dispersed in molten metals achieved high C 2 + selectivity  41  (>99%) but with a low yield (<1%) (Fujimoto et al., 1991). The use of molten barium hydroxide at 800°C achieved a methane conversion of ca. 21% and C2+ selectivity of ca. 33-38% (Moneuse et al., 1990).  To overcome the high exothermicity of the reaction a thin bed reactor was used, in which a thin bed of catalyst was placed in a heat conductive catalyst holder and mounted in an adiabatic pressure vessel (Leyshon, 1991). C2+ selectivity of 70 weight% was achieved at 30 psig and 25% methane conversion.  The use of a membrane reactor allows separation of the methane and oxygen streams, and has been shown to achieve high selectivity. A membrane reactor consisting of MgO coated on a porous alumina tube, subsequently coated with lead nitrate and followed by calcination to produce a non-porous lead oxide coating, resulted in greater than 97%  C2+ selectivity (Omata et al., 1989). The separation of the methane and oxygen sources allows the use of air, as the nitrogen does not come into contact with the methane and is not transferred through the membrane. The transfer of the oxygen through the nonporous membrane is a good indication of oxygen transfer through the lead oxide.  2.7 Engineering and Economic Assessments  A number of economic assessments of the oxidative coupling of methane process have been carried out to determine the C 2 + selectivity, yield and methane conversion required to achieve economic viability.  42  A comparison of the direct methane conversion processes considered oxidative coupling, partial oxidation, and oxyhydrochlorination (Fox et al., 1990). Of the three, the oxyhydrochlorination was the most economical, due to higher selectivity, despite the need for special materials of construction.  Another study evaluated direct partial oxidation to methanol and oxidative coupling to ethylene for making liquid fuels, and compared them with a conventional natural gas-to-methanol via steam reforming process (Kuo, 1991). The study indicated that the oxidative coupling to ethylene process could become competitive if >88% selectivity at 35% single-pass conversion was achieved.  Lee and Aitani (1991) based an economic evaluation of the oxidative coupling of methane on results reported by Otsuka in 1988 (30.6% yield with a 64.7% C2+ selectivity). They concluded that these methane conversions and product selectivities are sufficiently high to make the process economically attractive compared with conventional ethylene production.  In a review of the state of process technology for the direct catalytic conversion of methane, the reviewers concluded that direct methane conversion is not yet competitive with conventional processes (Mobil MTG process) (Poirier et al., 1991).  Based on the C2 + yield and selectivity required to achieve an economic process, and considering the possible existence of an inherent yield limit in the oxidative coupling of methane as discussed previously, the possibility of a full scale commercial process  43  based on the oxidative coupling of methane is still questionable. The economics of the process must be improved, with further research aimed at increasing understanding of the catalyst mechanism and improving reactor configurations. The objective of this thesis is to evaluate the effects of the catalyst properties on the oxidative coupling of methane reaction, and to relate these effects to the possible catalytic reaction mechanisms occurring.  44  3. METHODS AND MATERIALS  The experimental part of this thesis consisted of making and testing samarium oxide and alkali and alkaline earth doped samarium oxide catalysts for the oxidative coupling of methane. The catalysts were prepared from samarium and dopant nitrates in a molten state, which decompose at elevated temperatures (up to 600°C) to form the solid oxide. The dopant concentration and the preparation procedure were varied to determine the effects of these variables on the catalyst performance. The catalysts were tested in a bench scale packed bed reactor, into which methane, oxygen, and helium were fed. The exit gas stream was analyzed using an on-line gas chromatograph.  3.1^Experimental Equipment  A schematic flow diagram of the equipment set-up is shown in Figure 3.1. The feed gases consisted of oxygen, ultra high purity (UHP) methane, and diluent helium. The flows were controlled by mass flow controllers, with maximum flow rates of: helium, 300 mL/min; methane, 200 mL/min; and oxygen, 100 mL/min; calibrated at 21.1°C and atmospheric pressure. The gases passed through 5 micron filters prior to the mass flow controllers and a 1 psi check valve immediately after the flow controllers to prevent any reverse flow of gas through the control valves. The methane and helium streams were combined and entered the reactor at the top of the reactor tube. The oxygen stream entered the reactor via a quartz feed tube which ran down the centre of the reactor tube and exited about one half a reactor diameter above the catalyst bed. This appeared to be an adequate distance for sufficient mixing of the methane and oxygen 45  EQUIPMENT FLOW DIAGRAM  •• ^ ••  BUBBLE FLOW METER  REACTOR BYPASS  I*  GAS CHROMATO —OWL -GRAPH  FURNACE MASS FLOW CONTROLLER  CATALYST BED FLOW METER  ddd  r  CONDENSER  z  •  0  II I  TEMPERATURE CONTROLLER  LEGEND  ti  L__  COMPUTER  SIGNALS TO COMPUTER  ,  LL^  0 FLOW MEASUREMENT  z w 0  0 THERMOCOUPLE 0 PRESSURE MEASUREMENT _J  X  o VALVE CHECK VALVE  e FILTER  streams while minimizing the pre-catalytic time for reactions. A pattern of carbon deposition on the catalyst particles would have indicated cracking of the methane in the absence of oxygen. Since this was not apparent, it was assumed that adequate mixing occurred. Nitrogen was available at the bottom of the reactor to quench the exit gas if necessary.  The reactor consisted of a 1 /2 inch quartz tube placed vertically in a tubular furnace, with a fitted quartz disc in place to support the catalyst (see Figure 3.2). Important features of the reactor included the reduction of reactor diameter immediately after the catalyst bed to reduce residence time of the exit gas in the heated zone and the ability to quench the gas with nitrogen if the exit gas leaves the furnace zone at elevated temperature.  To monitor the temperature profile of the reaction gas, the temperature was measured in three places by thermocouples: the entrance to the reactor; immediately below the catalyst bed by a quartz sheathed thermocouple; and the exit of the reactor. In addition, a thermocouple was positioned between the outside wall of the reactor and the furnace interior, to monitor the actual furnace temperature. The thermocouple in the heated zone of the reactor directly below the catalyst bed was protected by a quartz sheath as almost all metals have been shown to be reactive for the oxidative coupling of methane. The temperature measured by this thermocouple was assumed to be the reaction temperature.  A pressure transducer connected to a digital meter was used to monitor the reactor 47  PACKED BED REACTOR FOR THE OXIDATIVE COUPLING OF METHANE 1/8' THERMOCOUPLE  •  NITROGEN  SIDE VIEW  0.125 -II-  CATALYST  SUPPORT \  1/16' THERMOCOUPLE F025  15  METHANE^OXYGEN  I15-1  1/8' QUARTZ THERMOCOUPLE SHEATH  14  4  PLAN VIEW  -0.125  PRODUCT  ALL DIMENSIONS IN INCHES  pressure. The exit gas passed through a glass impinger contained in an ice bath to condense any water out of the gas stream. The gas stream then passed through a rotameter and exited the reactor system to the on-line gas chromatograph (GC). After leaving the GC, the gas stream passed through a bubble flow meter and was exhausted.  3.2 Data Acquisition  The pressure, catalyst bed temperature, and methane, helium, and oxygen flow rates were continuously monitored on a computer using the data acquisition software, "Labtech Notebook", which was interfaced to the equipment with a Das-8 computer interface board.  3.3 Gas Chromatograph  A Shimadzu gas chromatograph (GC), with both a thermal conductivity detector (TCD) and a flame ionization detector (HD), was set up to monitor the product gases using on-line analysis. The TCD was used to measure the carbon monoxide, carbon dioxide, and oxygen. The FID, more accurate for hydrocarbon species, was used to measure the methane, ethylene, ethane, propylene, and propane.  The GC used two columns operated in series and a temperature program to provide separation of the components. The columns used were a Poropak Q (8 feet) followed by a 5A molecular sieve (6 feet). The sequence which was used is as follows: the sample is injected; the lighter components, including methane, carbon monoxide,  49  oxygen, and nitrogen, pass through the Poropak and into the molecular sieve column; the molecular sieve is switched out of the gas flow stream and the gas flows directly from the Poropak to the detectors; the carbon dioxide, ethylene and ethane exit the Poropak and pass through the detectors where they are measured; the molecular sieve is then switched back into line and the components contained in it pass through and into the detectors; the molecular sieve is then switched back out of line, in time for the C3  compounds to exit the Poropak and pass through the detectors. The temperature  program used was as follows: initial temperature of 80°C for 5 minutes; temperature increase at 10°C/minute for seven minutes up to 150°C; followed by a hold at 150°C for three minutes. Approximately 16 minutes was required for each sample cycle of the GC, including complete elution of the columns, cool down, and stabilization of the detector signals. The GC data was recorded and integrated by a Shimadzu Chromatopac integrator.  3.3.1 GC Calibration  The GC was calibrated with a purchased calibration gas containing: 10.00% methane, 2.02% ethylene, 2.00% ethane, 2.01% carbon monoxide, 8.01% carbon dioxide, and 0.10% propane, with the balance helium. No oxygen can be included in the calibration gas as the contents are flammable gases under pressure. The oxygen calibration was carried out by mixing pure oxygen and helium streams in the reactor bypass and directing this mixture into the GC. The composition of this stream was controlled by the mass flow controller, and any errors associated with the mass flow controller calibration will be passed on to the GC calibration for oxygen. 50  The GC columns required regular conditioning at 150°C for at least an hour under a helium flow. This procedure removed substances which tended to accumulate on the columns and interfere with GC operation.  3.4 Flow Measurement and Control  The inlet gases were measured and controlled by the mass flow controllers. These control valves were factory calibrated at 70°F (21.1°C). However, testing of the mass flow controllers indicated that factory calibration was not completely accurate.  An on-line rotameter was initially used to measure the product flow rate. However, this was not accurate enough and a 100 mL bubble flow meter was added to the end of the product gas line after the GC and immediately prior to exhaust. After correcting for temperature, the bubble flow meter was considered accurate.  3.5 Definitions  Throughout this report, some conventions will be used to describe the results obtained in the experiments. These conventions are consistent with the majority of research reported in this field. The term C3 represents  C2 is  used to represent both ethane and ethylene, while  propane and propylene; C2+ represents C 2's + COX represents both  carbon dioxide and carbon monoxide. Methane and oxygen conversion denote the percentage reacted of the original methane and oxygen feed, respectively. C2 yield denotes the percentage of the original methane feed which was converted to C 2 51  molecules. C2 selectivity denotes the percentage of the reacted methane molecules which are converted to C2 molecules. All compounds are measured in moles.  C2 Yield  = 2*(C 2H4 + C2 H6)/ (C exiting reactor)^(3.1)  C2 selectivity  = 2*(C 2 H4 + C2 H6)/ (C in products)^(3.2)  where^(C exiting reactor) = CO + CO 2 + CH4 + 2*(C2H4 + C2 H6) + 3*(C3) (C in products) = CO + CO 2 + 2*(C2 H4 + C2 H6) + 3*(C3)  Similar conventions are used for the oxidation products and C3s. The space time yield (STY) is defined as the amount of C 2's produced per unit time per unit catalyst weight, with the units mol/s/g. Some parameters which are used to describe the reaction conditions include: methane to oxygen mole ratio (CH 4 /02); fraction, based on pressure or flow, of reacting gases in feed,  P*7---(PcH4÷P02)/Ptotal;  and weight of catalyst  over feed flowrate, W/F (g s/mL).  3.6^Catalyst Preparation Procedure  Doped catalysts were prepared by mixing the appropriate weights of dopant nitrate and samarium nitrate and melting in a platinum dish at 95°C. After mixing, the mixture was allowed to cool and the resulting solid was broken into chunks, which were then used as the starting material for the catalyst preparation.  The catalysts were prepared in a modified thermogravimetric analyzer (TGA), in a 52  quartz catalyst preparation boat of 27 mm in length and 11 mm deep. The oxygen flow through the TGA and over the catalyst during preparation was approximately 50 mL/min. The catalysts were prepared using a temperature program which allowed the catalyst material to be heated at a specified rate. The temperature program used, unless otherwise stated, was as follows:  •  initial temperature: 25°C  •  increase temperature by 3°C/min to 325°C  •  hold at 325°C for 25 minutes  •  increase temperature by 4°C/min to 600°C  •  allow to cool  The temperature hold in the middle of the program was to allow nitrogen dioxide to bubble off as the samarium nitrate decomposed.  3.7 Experimental Procedure  The catalysts were sieved and the -16 +32 Tyler mesh fraction (0.5 to 1.0 mm in diameter) was used for catalyst testing. Approximately 50 mg of catalyst was placed on top of the fitted glass disk in the reactor, which resulted in a bed depth of approximately 4 mm. The helium flow was set at 100 mL/min and the furnace was started, with the setpoint at 650°C. The furnace heats up very quickly and was at setpoint temperature in about 10 to 15 minutes.  53  After a total of approximately one hour of helium flow over the catalyst, the conditions for the first run were set (methane and oxygen flows were started and the helium flow adjusted to maintain a constant flow of 100 mL/min). After 15 minutes the initial sample of the product gas stream was sampled by an on-line gas chromatograph. Thereafter, the reaction products were sampled at 16 minute intervals. For the first run, five samples were tested, for the second run, four samples were tested, and thereafter, three samples were tested per run. After the initial two samples for the first run, the results appeared to be reasonably constant, and the results for the last three samples were averaged for each run. After the last sample for each run was removed, the conditions were changed to those for the next run. The first sample time for each run was therefore 15 minutes after the conditions were set.  The conditions used for testing of the catalysts were selected as follows:  i)  Three different reaction temperatures were selected to determine the effect of temperature on the potentially temperature dependent catalyst properties, such as carbonate formation.  ii)  The inherent selectivity and activity of the catalyst is likely to be most apparent at low methane to oxygen ratios; therefore, CH4 /02 ratios of 2 and 4 were tested for comparison of catalyst performance, along with higher ratios to determine the effects of this parameter.  iii)^A univariant experimental design was used, such that only one catalyst 54  parameter (e.g., dopant) was changed at a time. The interpretation of changes in catalyst parameters is fairly complicated. The change in a parameter, such as dopant used, will have a number of effects on the catalyst properties, only some of which can be measured. The change in catalyst properties will in turn have an effect on the oxidative coupling reactions. Therefore, the effect of the change in one catalyst parameter on the oxidative coupling of methane must be interpreted based on several catalyst properties. Therefore, to compare the effects of catalyst properties to that of catalyst performance, the following parameters were varied in a univariant manner: dopant used, dopant concentration, and catalyst preparation procedure. Each catalyst was tested under identical conditions.  The conditions used for testing of the samarium oxide catalysts doped at a 1:100 mole ratio of dopant to samarium are listed in Table 3.1.  3.7.1 Variation of Catalyst Dopant Concentration  Calcium and sodium doped catalysts were prepared at a dopant to samarium mole ratio of 10:100 in order to investigate the effect of dopant concentration. The conditions used for testing of these catalysts are presented in Table 3.2.  3.7.2 Catalyst Preparation Modification  A revised preparation program was tested. This program omitted the hold midway, 55  and the catalyst temperature was raised from 25°C to 600°C at a rate of 25°C per minute. The catalyst was then held at 600°C for 30 minutes. This catalyst was also tested at the conditions presented in Table 3.2.  Table 3.1 Conditions Used for Comparison Testing of Doped Catalysts Parameter  #1  #2  #3  #4  #5  #6  #7  #8  Temperature (°C)  650  650  650  650  750  750  850  850  Catalyst weight (mg)  50  50  50  50  50  50  50  50  Total flow rate (mL/min)  100  100  100  100  100  100  100  100  % Oxygen (molar)  5  5  5  5  5  5  5  5  % Methane (molar)  10  20  40  80  20  40  20  40  % Helium (molar)  85  75  55  15  75  55  75  55  W/F (g s/mL)  0.03  0.03  0.03  0.03  0.03  0.03  0.03  0.03  CH4 /02 molar ratio  2  4  8  16  4  8  4  8  Table 3.2 Conditions Used for Comparison Testing of Catalysts at Higher Dopant Concentrations and Different Preparation Times Parameter  #1  #2  #3  #4  #5  #6  Temperature (°C)  650  650  750  750  850  850  Catalyst weight (mg)  50  50  50  50  50  50  Total flow rate (mL/min)  100  100  100  100  100  100  % Oxygen (molar)  5  5  5  5  5  5  % Methane (molar)  20  40  20  40  20  40  % Helium (molar)  75  55  75  55  75  55  W/F (g s/mL)  0.03  0.03  0.03  0.03  0.03  0.03  CH4 /02 molar ratio  4  8  4  8  4  8  56  4.^RESULTS AND DISCUSSION  Samarium oxide and alkali and alkaline earth doped samarium oxide catalysts were prepared and tested for the oxidative coupling of methane. The effect of specific dopant used, varying dopant concentration and catalyst preparation method were evaluated. The catalysts were tested in a bench scale packed bed reactor under conditions of varying temperature and methane to oxygen ratio. The catalysts were characterized by scanning electron microscopy, powder x-ray diffraction, BET surface area, estimated basicity, ability to form carbonates, and ionic radius of dopant.  4.1 Undoped Sm 203 : Conditions of Complete Oxygen Conversion  The first samarium oxide catalyst was tested at a temperature of 750°C under various conditions of total flow rate, total partial pressure of reactants, and methane to oxygen ratio. The conditions and results are presented in Table 4.1 and are shown as a function of methane to oxygen ratio in Figures 4.1, 4.2, and 4.3. The product gases were not analyzed for C3+ compounds. The effects due to methane to oxygen ratio are much greater than those due to the amount of diluent used or the total flow rate. The oxygen conversion was greater than 99.8% for all conditions tested. The methane conversion decreased from 41% to 9%, the  C2 selectivity  increased from 33% to 81%,  and the C2 yield decreased from 13.6 to 7.4%, as the methane to oxygen ratio increased from 2 to 16. A lower methane to oxygen ratio results in higher methane conversion, lower C2 selectivity, and higher  C2 yield.  57  Table 4.1 Samarium Oxide: Conditions of Complete Oxygen Conversion Run  #1  #2  #3  #4  #5  #6  #7  #8  Input Flow Rates (mL/min) (at 21.1°C, atmospheric pressure) He  50  150  75  85  55  30  15  50  CH4  40  40  20  10  40  8  80  40  02  10  10  5  5  5  2  5  10  Total  100  200  100  100  100  40  100  100  4  4  4  2  8  4  16  4  0.25  0.25  0.25  0.85  CH4 /02 P"  0.50  0.15  0.45  0.50  W/F g s/mL 0.12  0.06  0.12  0.12  0.12  0.30  0.12  0.12  CO2  10.1  9.3  9.8  22.9  4.2  10.9  1.7  9.8  CO  1.2  1.2  1.1  4.8  0.0  0.8  0.0  1.0  C2 H4  7.5  6.6  6.7  7.7  4.9  6.6  3.1  7.2  C 2 H6  6.1  6.6  6.3  5.9  5.6  5.4  4.3  5.9  Total  24.9  23.7  23.8  41.3  14.7  23.7  9.1  23.8  Yield %  Selectivity % COX  45.5  44.0  45.5  67.1  28.3  49.4  18.6  45.1  C21 S  54.5  56.0  54.5  32.9  71.7  50.7  81.4  54.9  Space Time Yield grnol/g/s C 2 's  8.9  8.7  4.0  2.0  6.3  1.3  9.0  8.0  23.7  23.8  41.2  14.7  23.7  9.1  23.8  100.0  100.0  100.0  100.0  100.0  99.9  Conversion % CH4 02  25.0 99.9  99.8  Reaction conditions: T = 750°C, weight of catalyst = 0.21g  58  Figure 4.1  Methane Conversion for Samarium Oxide Conditions of Complete Oxygen Conversion, T=750C  45.0 c  •  40.0  Eao 35.0  > 30.0 c o 25.0  2  c  20.0 a ..F. 15.0 2  °  • •  10.0 5.0 0.0  0^2^4^6^8^10^12^14^16 CH4/02  Figure 4.2  C2 Selectivity for Samarium Oxide Conditions of Complete Oxygen Conversion, T=750C 90.0 80.0  ^•  •  70.0 :"; 60.0 45 w 50.0  $•  N40.0 csi 0 30.0 9. 20.0  •  10.0 0.0 0  2  4  6  8  10  12  14  16  CH4/02  59  Figure 4.3  C2 Yield for Samarium Oxide Conditions of Complete Oxygen Conversion, T=7500 16.0 14.0 13 12.0 ..-6 >01 10.0 0 °^8.0  •  $ • •  6.0 4.0 0^2^4^6^8  ^ ^ ^ ^ 14 12 16 10  CH4/02  60  4.1.1 Effect of Gas Phase Flow Rate  Keeping all other conditions constant, the total flow rate was varied in order to vary the residence time in the reactor. The results for these runs are presented in Table 4.2. As the residence time was increased, the methane conversion remained constant, and the C2 selectivity decreased slightly with a corresponding small decrease in  C2  yield.  The STY decreased with an increase in residence time. These results indicate that, since the oxygen conversion was 100%, no further activation of methane occurred in the absence of gas phase oxygen. However, it does appear that some C 2's were oxidized. Table 4.2 Effect of Total Flow Rate Run #  Q  (mL/min)  Residence Time (s)  % Methane Conversion  % C2 Yield  % C2  Selectivity  STY p.mol/g/s  2  200  0.2  24  13  56  8.7  3  100  0.4  24  13  55  4.0  6  40  0.9  24  12  51  1.3  Reaction conditions: T=750°C, 20% CH 4 , 5% 02  4.1.2 Carbon Deposition in the Reactor  After the test runs were completed and the catalyst cooled, the catalyst was examined for carbon deposits. There was considerable carbon deposition on both the walls of the reactor and the catalyst. This was likely due to the cracking of the hydrocarbons in the absence of oxygen, to produce carbon and hydrogen. The carbon deposition on the  61  catalyst appeared only on the bottom half of the catalyst and increased in amount down the length of the catalyst bed. A small amount of carbon was deposited on the fritted glass disk.  These observations indicate that oxidative coupling reactions occurred only in the first half of the catalyst bed, after which all of the oxygen was consumed. Carbon is formed at this temperature in the absence of gas phase oxygen. Comparison of catalysts under these conditions is difficult since the lack of oxygen may mask the inherent selectivity of the catalysts.  4.1.3 Comparison of Results for Sm203 to Other Researchers' Data  Table 4.3 is a compilation of results obtained by various researchers for pure samarium oxide catalysts under conditions of 750°C and complete or almost complete oxygen conversion. These results, along with the initial results obtained in the present study, (Table 4.1) are plotted in Figures 4.4, 4.5 and 4.6. The methane conversion,  C2  yield,  and C2 selectivity all show a general correlation when plotted against the methane to oxygen ratio. The results of the present study show the same general trends as the results of other researchers and are of approximately the same values. It should be noted that reactor configurations and reaction conditions vary widely. The methane to oxygen ratio has a significant effect on the determination of methane conversion and C2  selectivity.  62  Table 4.3 Samarium Oxide Results for Various Researchers P(CH4) kPa  CH4 /02  W/F g s/mL  % Methane Conversion  % C2 Selectivity  Yield  % C2  C2 STY innoligis  Ref.  87  6.1  0.012  14  61  8.5  136  3  87  6.1  0.0008  5  60  3  719  18  5.1  2.4  15.1  60.6  9.2  0.15  5  2.5  30.2  41.2  12.4  5  5  2.5  30  40  12  6  68  10.0  0.22  12.2  54  6.6  4.42  51  5.0  0.22  21.9  47  10.3  5.15  13  3.9  0.055  31  87  6.1  2.31  50  81  4.0  0.0014  24  48.9  90  8.1  0.0014  16  30  6.0  0.234  17.9  3  Otsuka and Komatsu, 1987  4  Hutchings et al., 1989a  5  Kaddouri et al., 1989  6  Otsuka and Nakajima, 1987  7  Korf et al., 1989  8  Kennedy and Cant, 1991  3 4  7  8  0.22  9  11.7  62.62  10  62.1  9.8  58.35  55.8  10  2.86  10  ii  Hamid and Moyes, 1991 10  Choudhary and Rane, 1991  11  Deboy and Hicks, 1988 63  ▪  Figure 4.4  Methane Conversion for Samarium Oxide, All Researchers Conditions of Complete Oxygen Conversion, T=7500  45 c o E o C 0 U  0 C  Current re sutts  0  40 35  ■  30 25  her researchers  8  20  •  15  ■ ■  e  2 10  •  0  •  5 0 0.0  ^  2.0^4.0^6.0^8.0^10.0  ^  12.0  ^  14.0  ^  16.0  CH4/02  Figure 4.5  C2 Selectivity for Samarium Oxide, All Researchers Conditions of Complete Oxygen Conversion, 1=7500  90  ^0  80  0  70 :5 60  •  0  rdi 50 ti, 40  • .  f.t)  N 0  Current esutts  0  30  st^  Other re searchers  O 20 10 0 0.0  2.0^4.0^6.0^8.0^10.0^12.0^14.0^16.0 CH4/02  64  4.1.4 Space Time Yield  The space time yield (STY) is often used as a measure of the activity of a catalyst. Some very high STY's have been reported by a few researchers under conditions of high methane to oxygen ratios and low catalyst residence time (Otsuka and Komatsu, 1987; Hutchings et al., 1989a). Amenomiya et al. (1990) plotted the STY divided by the partial pressure of methane in the feed as a function of W/F, for various catalysts. A log-log plot permits a fairly good correlation to a straight line for catalysts of various compositions and reaction conditions. The present results are plotted along with data from other samarium oxide catalysts at 750°C and complete oxygen conversion, and with data for various catalysts at a reaction temperature of 750°C (from the paper by Amenomiya et al.) in Figure 4.7. Amenomiya et al. (1990) interpreted the apparent relationship to indicate that the reaction is controlled by homogeneous reactions under conditions of complete oxygen conversion.  4.2 Undoped Sm 203 : Conditions of Incomplete Oxygen Conversion  The major purpose of this study was to compare an undoped Sm 203 catalyst with Sm203 catalysts doped with alkali and alkaline earth oxides, in order to determine the effect of catalyst properties on the oxidative coupling of methane. Under conditions of complete oxygen conversion, the inherent catalyst activity and selectivity can be masked by the lack of oxygen. Therefore, the samarium oxide catalyst was retested under conditions of incomplete oxygen conversion, which are more suitable for catalyst comparison. The amount of catalyst tested was 50 mg, and the temperatures used were 65  Figure 4.6  C2 Yield for Samarium Oxide, All Researchers Conditions of Complete Oxygen Conversion, T=750C 14  ) —••  12  Curren- results • •  10 8  • '  6 4  •  <  •  N  Other researchers  2 0  0.0^2.0^4.0^6.0^8.0^10.0^12.0^14.0^16.0 CH4/02  Figure 4.7  STY for Various Catalysts, T=750C Other cata ysts^Sm2O3  10  (k  II • IIIIM_IIHI'^ *^IP.  I  Current results  ^•  I^I III  al1I  I  P  ■ 111--7---'  0  0.001 0.0001  0.001  ^  0.01^0.1  ^  1  10  W/F (g s /m1)  66  650°C, 750°C, and 850°C. At 650°C and 750°C, conditions of incomplete oxygen conversion existed. However, at 850°C, oxygen conversion was 100% for some catalysts and approached 100% for all catalysts. Therefore, the inherent selectivity of the catalysts will be somewhat masked at this temperature. The conditions used are fully described in Section 3. The complete results are presented in Table A.1.  4.2.1 Effect of Methane to Oxygen Ratio  The results for each sample collected for the reaction temperature of 650°C are , presented in Figure 4.8. With all other conditions held constant, the helium and methane flows were varied, such that, for a constant oxygen concentration of 5%, the methane to oxygen ratio varied from 2 to 16. This figure shows the change in catalyst performance over time, with the first set of conditions held constant for 80 minutes. No significant change in the exit gas composition was noticed over this period, and the results also appear to be quite steady for each subsequent set of conditions. It is interesting to observe the effect of the change in methane to oxygen ratio. As the ratio is increased, the C2+ selectivity increases with a simultaneous decrease in methane conversion, such that the C2+ yield remains almost constant. This effect is not seen under conditions of 100% oxygen conversion (oxygen limiting), for which the C2 yield decreases significantly for an increase in methane to oxygen ratio (see Figure 4.3).  The average of the data for each set of conditions is used for all further interpretation and discussion. The results are plotted in Figures 4.9 to 4.11 as a function of methane to oxygen ratio for reaction temperature equal to 650°C. It is interesting to note that 67  •  •  Figure 4.8  Samarium Oxide Results Conditions of Incomplete Oxygen Conversion, T= 650C 70.0 ^  60.0 CH4/02=2  50.0  ^  30.0 20.0 10.0 0.0  0  0  n  n  CH4/02=8  ^  0  a  .^  • ^ ■  0^13^o  CH4/02=4  40.0 ^^  ^  0  •  •  ■  •  •  •  •  •  •  •  •  •  •  0^C  •  .^.^• •^• • I^I^I  CH4/02=1  • 1:^ •  .  I^I^I  1^2^3^4^5^6^7^8^9^10 11 12 13 14 15  Sample # • CH4 conversion ° C2+ selectivity ^• C2+ yield  Figure 4.9  Conversion for Samarium Oxide Conditions of Incomplete Oxygen Conversion, T = 650C 80.0  ^^  70.0 O 60.0 0 50.0 40.0  U 30.0 •  20.0 10.0 0.0 0  4^6^8^10  12  14  16  Methane to oxygen ratio •  CH4  ^02  68  ^  Figure 4.10  Selectivity for Samarium Oxide ^80.0  ^  Conditions of Incomplete Oxygen Conversion, 1= 650C % Selectivity  70.0 ^  CO2  60.0 ^  CO  50.0 ^  COx  440.0 ^  C2H4  30.0 ^ a°^' 20.0 ^  C2H6 C3's  10.0 ^  C2+  0.0 ^ 0  5^10^15  20  Methane to oxygen ratio  Figure 4.11  'Ye Yield for Samarium Oxide % Yield  Conditions of Incomplete Oxygen Conversion, 1=6500 14.0  •  •  12.0  CO  10.0 (i)  8.0  e  6.0  CO2  •— COx C2H4 •  4.0  C2H6  __••••••••■••■•■• •  C2  2.0 --41-  0.0 0  5^10^15^20  0  C3's C2+  Methane to Oxygen Ratio  69  the CO and C2H4 yields are virtually identical over the whole range of CH 4 /O2 ratios, with an initial increase up to CH4 /O 2 = 4, and then only a very slight increase as the CH4 /02 ratio increases. The main effects are seen to be the steady increase in C 2 H6 selectivity and the steady decrease in CO 2 yield as the methane to oxygen ratio increases.  The product outputs, in terms of moles of carbon, are plotted as a function of the amount of methane feed in Figure 4.12. Ethane, ethylene, C3 s, and carbon monoxide 1  all appear to be almost linear functions of methane in the feed. The amount of carbon dioxide produced, on the other hand, remains almost constant. The CO and CO2 produced as a function of oxygen conversion are graphed in Figure 4.13. Although CO production varies with oxygen conversion, CO 2 production appears to be independent of oxygen conversion. These results are consistent with results by Ekstrom et al. (1989c), who determined that the rate of CO X production is not a function of the reductant present (e.g., CH4), but of the oxidant. The oxidant is clearly not gas phase molecular oxygen, since this is present in excess, but must be some other form of oxygen species (see Section 2.2.3).  4.2.2 Effect of Reaction Temperature  Due to the exothermicity of the oxidative coupling reaction, the catalyst bed temperature may have a "hot spot" temperature which is greater than the measured gas phase temperature. The measured temperature, which is quoted as the reaction temperature, is actually the gas phase temperature immediately after the catalyst bed. 70  ^  Figure 4.12  Product Carbon Output as a Function of Total Carbon 80.0 ^  Conditions of Incomplete Oxygen Conversion, T = 650C  .---------*  70.0 60.0 ^ 250.0 ^ ^-8'  ■Ai^  40.0 ^  g 30.0 ^ 20.0 ^ 10.0 ^ 0.0 ^ 0  500  1000^1500^2000^2500^3000 Total Carbon x1 0E6 (moles)  CO2 —0-- CO^*— C2H4 —0-- C2H6 • C3's  Figure 4.13  Output of Total Oxidation Products as a Function of Oxygen Conversion 50.0 .o^ 45.0 LAJ o 40.0 7‹ 35.0 32 li; 30.0 S •73 25.0 >.5 ^20.0 =^15.0 13 0^10.0 gr-^5.0 0.0 25.0  Conditions of Incomplete Oxygen Conversion, T=650C •  •  .  n  0^ C^  L 35.0^45.0^55.0^65.0  ^  75.0  85.0  % Oxygen Conversion •  CO2 ° CO  71  The gas phase temperature was also monitored at the entrance and exit of the heated zone of the reactor to determine the temperature profile of the gas stream. The temperature at the entrance to the reactor did not exceed 80°C, and that at the exit to the reactor did not exceed 35°C, for all reaction temperatures. This indicates that the gas phase is sufficiently cooled at the exit of the heated zone such that no further reactions will occur.  The results are plotted for a constant CH 4 /02 ratio as a function of gas phase reaction temperature in Figures 4.14 to 4.16. The methane and oxygen conversions increase with an increase in temperature. The CO X selectivity decreases as the temperature is raised from 650°C to 750°C, and then increases slightly as the temperature is raised to 850°C. The C2+ selectivity is opposite to this, with an increase as the temperature is raised to 750°C and then a small decrease as the temperature is raised to 850°C. The COX yield appears to increase almost linearly with temperature, while the C2+ yield increases considerably between 650°C and 750°C, with almost no increase apparent between 750°C and 850°C.  4.2.3 Carbon Balance over the Reactor  The carbon balance was calculated for each run by comparing the amount of methane input to the reactor with the amount of carbon containing products exiting the reactor. In all cases, the output carbon value was substituted for the input carbon value for calculations, since this was considered to be more accurately measured. A loss of 7 to 14% carbon was observed. If this was due to carbon deposition, over the course of the 72  Figure 4.14  Conversion as a Function of Temperature for Samarium Oxide CH4/02=4 100.0 90.0 80.0 70.0 •r• 60.0 50.0 40.0 00 30.0 20.0 10.0 0.0 600  % Conversion CH4 ^ 02  ^• • 650^700^750^800^850  900  Reaction Temperature (C)  Figure 4.15  Selectivity as a Function of Temperature for Samarium Oxide 65.0  CH4/02 = 4  60.0  z,55.0  % Selectivity COx  f50.0  N  45.0  —0--  C2+  40.0 35.0 30.0 ^ 600^650^700^750^800^850 900 Reaction Temperature (C)  73  Figure 4.16  Yield as a Function of Temperature for Samarium Oxide CH4/02 =4 14.0 12.0 %Yield  10.0 0  5. 8.0 0 -  COx s —0—  C2+  6.0 4.0 2.0 600^650^700^750^800^850  900  Reaction Temperature (C)  74  test period approximately 0.6 g of carbon would be deposited. This amount is 12 times the 50 mg initial mass of the catalyst and would be readily visible at completion of the testing. This amount is not consistent with the small amount of carbon deposition which was observed. If the carbon loss was due to adsorption of carbon dioxide on the catalyst, the weight of carbon dioxide adsorbed would be 2.3 g, or 45 times the original weight of catalyst. This also does not seem reasonable. Although all fittings were tested regularly for gas leaks, it is possible that a small leak in the system may be responsible for the loss of carbon. There were many junctions of various media tubing, such as glass to stainless steel, glass to plastic tubing, and plastic tubing to stainless steel. The other alternatives include errors in the mass flow controllers, bubble flow meter accuracy, and GC calibration.  4.2.4 Variation in Results  In order to determine the significance of the results for comparison of the catalysts, two approaches were used. The first approach included determination of the variation in GC results, based on the calibration procedure. The second approach involved running two identical tests using samarium oxide as the catalyst.  The GC calibration was typically carried out by on-line injection of the calibration gas and analysis of the sample three or four times. This was carried out each day prior to a catalyst run. The results from each of three days, selected at random, were analyzed for standard deviation, and the standard deviation for each component was averaged over the three days. These values are presented in Table 4.4.  75  The average percent standard deviation is an indication of the expected percentage error in the concentration, and is not the absolute deviation. The calibration concentrations are presented in Table 4.4 with the average absolute standard deviation for each component. Table 4.4 Average Percent Standard Deviation in GC Calibration Component  TCD (%)  FID (%)  Calibration Gas % Concentration  carbon dioxide  2.7  -  8.01 +/- 0.22  carbon monoxide  1.4  -  2.01 +/- 0.03  methane  -  0.9  10.00 +/- 0.09  ethane  -  0.3  2.00 +/- 0.01  ethylene  -  0.8  2.02 +/- 0.02  propane  -  0.9  0.10 +/- 0.00  oxygen  4.2  -  -  The average percent standard deviation varied by less than 1% for all of the hydrocarbons. The largest variation was associated with the oxygen concentration, which was calibrated in a different manner than the other gases. These results indicate that the GC results do not vary significantly for the calibration gas. The actual concentration of the components in the calibration gas may vary from that specified and may introduce a systematic error. This error should not affect the comparison of the results.  The variation in the results between catalyst runs for the same catalyst was determined by conducting two runs at identical conditions with two samples of the same catalyst. The variations in results are presented in Table 4.5. The variation between runs under 76  the same conditions establishes a level of confidence for comparing runs under different conditions. Table 4.5 Variation in Results for Two Identical Runs Parameter  Sm203 #1 (%)  Sm203 #2 (%)  Percent Difference  CH4 conversion  6.34  6.36  0.4  CO. selectivity  34.4  32.5  5.4  C2+ selectivity  65.6  67.5  2.8  CO. yield  2.2  2.1  5.0  C 2, yield  4.2  4.3  3.0  4.3 Alkali and Alkaline Earth Doped Samarium Oxide Catalysts  Several doped Sm 203 catalysts were prepared and tested in the reactor. These included samarium oxide doped at a 1:100 dopant to samarium mole ratio with the alkali metals, sodium and potassium, and the alkaline earths, calcium and magnesium. These were all prepared both in oxygen and air atmospheres, with the exception of potassium, which was prepared in oxygen only. Calcium and sodium doped catalysts were also prepared at dopant to samarium mole ratios of 1:10 in an oxygen atmosphere.  The catalysts were prepared in two different oxygen containing atmospheres to determine if the concentration of oxygen available during formation of the catalyst crystals would have any effect. Some effect was noticed during testing of the catalysts and the catalysts prepared in oxygen were generally superior. The catalysts prepared  77  in an oxygen atmosphere are used for the interpretation of the effects of the dopants. A summary of the catalysts which were tested is presented in Table 4.6.  Table 4.6 Doped Samarium Oxide Catalysts Catalyst  Preparation Atmosphere  Dopant:Sm mole ratio  Sm203  Oxygen  1:100  Ca/Sm2O3  Air  1:100  Ca/Sm2O3  Oxygen  1:100  Mg/Sm203  Air  1:100  Mg/ Sm203  Oxygen  1:100  Na/Sm 2O3  Air  1:100  Na/Sm 2O3  Oxygen  1:100  K/Sm203  Oxygen  1:100  Ca/Sm2O3  Oxygen  10:100  Na/Sm 2O3  Oxygen  10:100  4.3.1 Results of Catalyst Performance Tests  The reaction conditions are described in Section 3.7. The complete results are presented in Tables A.2 to A.10. The results are summarized in Table 4.7 for CH 4 /02 = 4, and are compared for all catalysts prepared in oxygen in Figures 4.17 to 4.19. There are significant differences between catalyst performance for different doping agents, different dopant concentrations, and for different reaction temperatures. The performance of the sodium doped catalysts is particularly temperature dependent. There are some significant differences in catalyst performance based on the C2+ yields of the catalysts at each of the reaction temperatures. In this discussion, the doped 78  catalysts will be referred to by their dopant symbol (e.g., 1:100 Na:Sm mole ratio catalyst will be identified by Na where this does not cause confusion), and the undoped catalyst will be referred to by Sm. Unless otherwise specified, the doped catalysts are at 1:100 dopant:Sm mole ratio (e.g., the 1:10 Na:Sm mole ratio catalyst will be identified by Na (1:10)). The performance of the catalysts at 750°C, as defined by  C2+ yield, is in the order Ca (1:10) = Na > Ca > K > Sm = Mg > Na (1:10).  Table 4.7 Results for all Catalysts, CH 4 /02 = 4 % CH4 Conversion  % C2+ Selectivity  % C2+ Yield  Temperature (°C)  Temperature ( °C)  Temperature (°C)  650^750^850  650^750^850  650^750^850  Sm  13.2  23.0  25.0  40.2  54.3  50.6  5.3  12.5  12.6  Ca/Sm  15.0  24.6  25.4  38.2  53.5  51.4  5.7  13.1  13.1  Mg/Sm  13.7  23.2  25.4  40.3  53.6  51.0  5.5  12.4  12.9  Na/Sm  10.0  25.4  25.9  32.6  55.8  51.9  3.3  14.2  13.4  K/Sm  11.7  25.4  26.2  30.0  50.5  48.3  3.5  12.8  12.7  Ca/Sm A  14.3  22.5  23.9  32.1  48.4  48.0  5.7  13.1  13.1  Mg/SmA  13.3  23.2  24.3  39.8  53.6  49.7  5.3  12.4  12.1  Na/Sm A  10.2  26.3  27.9  35.5  51.1  46.1  3.6  13.4  12.9  Ca (1:10)  9.8  25.0  26.8  34.1  57.0  54.7  3.3  14.3  14.6  Na (1:10)  4.6  18.4  25.4  28.6  62.8  49.9  1.3  11.6  12.7  Catalyst  A: Prepared in air.  79  % C2+ Selectivity  5 23 8 6$$ 6  SLO  237176;: 45:Wn 41/  Ca/Sm Mesm7Y. ,  .MtS  ,  746M"V. r,. ,  ;  1:10 Ca :Sm ^ Mg /Sm  tt;  Na/Sm  iniefieNNWFWEr  1:10 Na:Sm  ::::  ::::::::::::::::::  "^ •":.  K/Sm  ^  •  0 0 0  % C2 Yield  P  N A O  90 O  N)  O oo o o b b ob  Sm •• • •P•,:aKM:..A.••••  Ca/Sm 1:10 Ca:Sm ^• M g /S m  C) .;, .§gf•P :  O • ••:•.:  4^•  O  Na/Sm  WagsAvvry  1:10 Na:Sm  a.  ■^  art  ',A*Dwr• • kt•  ''•  K/Sm  ^  •  The following observations can be made regarding the doped catalyst performance as compared to the undoped Sm 203 .  Calcium  The methane conversion was higher at all temperatures for the (1:100) calcium doped catalyst (24.6% compared to 23.0% for Sm 203 at 750°C), but there was no noticeable effect on C2+ selectivity. The C2+ yield was higher at all temperatures (13.1% compared to 12.5% for Sm203 at 750°C). The 1:10 doped catalyst had more effect, with a higher methane conversion at 750°C (25.0%) and 850°C (26.8% compared to 25.0% for Sm 203), but a significantly lower methane conversion at 650°C (9.8% compared to 13.2% for Sm203). The C2+ selectivity was also higher than that of the undoped samarium oxide at 750°C (57.0% compared to 54.3%) and 850°C (54.7% compared to 50.6%), but was lower at 650°C (34.1% compared to 40.2%). The C2+ yield was significantly lower at 650°C (3.3% compared to 5.3%), but in this regard calcium doping gave the best catalyst at 750°C (14.3% compared to 12.5%) and 850°C (14.6% compared to 12.6%).  The concentrations of dopants used in the present study were quite low compared to those used by some researchers. For example, Korf et a/. (1989) used 30 mole % calcium doped samarium oxide and reported a significant effect. This reported effect was an increase in absolute C2 yield of 1% at 780°C (or ca. 10% relative increase in yield), and was less than the increase in absolute C2 yield of 1.8% (or ca. 14% relative increase) obtained in this study for the Ca (1:10) at 750°C.  82  Magnesium  Magnesium showed only a slight increase in methane conversion at all temperatures. There was no significant effect on C2+ selectivity, and a very small increase in the C2+ yield at 850°C.  Sodium  The behaviour of the sodium doped catalysts was very temperature dependent, which is in accordance with another study in which it was observed that the shape of the temperature-yield curves changed with changing sodium concentration (Korf et al., 1989).  The methane conversion at 650°C of the sodium doped catalyst (1:100) was 10.0%, significantly lower than the 13.2% of the undoped samarium oxide, with the effect much more noticeable for the 1:10 catalyst (methane conversion = 4.6%). However, at 750°C the 1:100 doped catalyst produced a higher methane conversion (25.4%) than the undoped (23.0%), while the 1:10 was still much lower (18.4%). At 850°C, both Na doped catalysts produced a slightly higher methane conversion (1:100, 25.9%; 1:10, 25.4%; and undoped, 25.0%). The C2+ selectivity also exhibited temperature dependent behaviour. The doped catalyst gave a significantly lower C2+ selectivity at 650°C (32.6% for 1:100 compared to 40.2% for undoped), again with the 1:10 catalyst being lower than the 1:100 (28.6%). At 750°C the 1:100 gave a slightly higher C2+ selectivity (55.8% compared to 54.3%), and the 1:10 gave a much higher C2+ selectivity (62.8%). There was little effect observed at 850°C on C2+ selectivity. The C2+ yield followed the same trends as that of the methane conversion and C2+ selectivity. The undoped 83  samarium oxide (5.3%) gave a significantly higher yield than the doped catalysts at 650°C (1:100, 3.3%; 1:10, 1.3%). At 750°C the 1:100 sodium doped catalyst gave a significantly higher yield than the undoped catalyst (14.2% compared to 12.5%), with the 1:10 sodium doped giving a lower yield (11.6%). At 850°C, the 1:100 doped catalyst gave a higher yield than the undoped catalyst (13.4% compared to 12.6%), and the 1:10 doped catalyst produced about the same yield as the samarium oxide (12.7%).  Potassium  The potassium doped catalyst also exhibited temperature dependent behaviour, with the methane conversion lower at 650°C (11.7% compared to 13.2% for Sm 203), but higher at 750°C (25.4% compared to 23.0%) and 850°C (26.2% compared to 25.0%). The C2+ selectivity was lower for all temperatures (50.5% compared to 54.3% for Sm 203 at  750°C). The C2+ yield was considerably lower at 650°C (3.5% compared to 5.3%), with that at 750°C slightly higher (12.8% compared to 12.5%) and that at 850°C the same as the Sm203 .  4.4 Modification of Catalyst Preparation Procedure  The preparation of the catalysts involves a temperature hold at 325°C. The process is fully described in Section 3.6. In order to analyze the processes occurring during catalyst preparation, the decomposition temperatures of the samarium nitrate and the dopant nitrates were determined. This was accomplished by observation of the pure nitrates during a TGA run operated at a constant rate of temperature increase. These values are presented in Table 4.8. 84  Table 4.8  Melting and Decomposition Temperatures of Nitrates used in Catalyst Preparation Nitrate  Melting Point (°C)  Decomposition Range (°C)  Samarium  85  290  Calcium  44  500-600  Potassium  335  >600  Sodium  300  715  These values indicate that the samarium nitrate decomposed at a much lower temperature than the dopant nitrates. Therefore, the samarium oxide is probably almost completely formed during the 325°C hold, long before the dopant nitrate even starts to decompose. The dopant may therefore not be dispersed uniformly throughout the Sm203 crystal lattice.  An increase in the concentration of catalyst defects, brought about by the improved inclusion of the dopant cations in the samarium oxide crystal, may result in increased catalytic activity. In general, the slower the crystal growth, the fewer the number of defects included. Therefore, decreasing the catalyst preparation time may increase the number of defects.  In order to test the effect of the catalyst preparation time, a modified catalyst preparation program was tested with the calcium dopant. This program omitted the hold midway, and the catalyst temperature was raised from 25°C to 600°C at a rate of 25°C per minute. The catalyst was then held at 600°C for 30 minutes. It was hoped that the samarium oxide would not have time to form prior to the calcium nitrate 85  decomposition. The resulting catalyst would then have a maximum number of dopant cations incorporated into the catalyst crystal.  4.4.1 Results of Performance Test for Modified Catalyst  The calcium doped samarium oxide catalyst, prepared at a 10:100 mole ratio of calcium to samarium, and prepared using a revised catalyst preparation program, was tested according to the conditions presented in Table 3.2. The results obtained are presented in Table A.11 and Figures 4.20 to 4.23. The catalyst preparation had a significant effect on the catalyst performance. The catalyst was designated 1:10 Ca (RP) (for revised program).  Oxygen Conversion  The oxygen conversion was significantly higher for this catalyst, and reached 100% at 750°C. The other catalysts tested did not reach 100% oxygen conversion until 850°C.  Methane Conversion  At 650°C, the methane conversion was significantly higher, particularly at the low methane to oxygen ratio (= 4), where a methane conversion of 17.4% was achieved, compared to 9.8% for the catalyst prepared with the standard program.  C, Selectivity  The C2 selectivity was considerably lower, about half of the previous value (15.8% compared to 32.4%). For all of the catalysts tested prior to this, the  C2  selectivity 86  Figure 4.20  Effect of Catalyst Preparation on % Oxygen Conversion 100.0  ^0  90.0 c 0  80.0  rD c 0  70.0  U N at  1:10 Ca (RP  .-  1: 1 0 Cc  60.0  ..----------.  50.0 40.0 Run 1  Run 2  Run 3  Run 4  Run 5  Run 6  Figure 4.21  Effect of Catalyst Preparation on % Methane Conversion 28.0  •  26.0 c 0 24.0 .E0.° 22.0 u.,  C  0 0 0 c 0 .c 0 2  20.0 18.0 16.0 14.0  ae 12.0  10.0  Li  1:10 Ca^100 02 Conyers' n  8.0 Run 1  Run 2  Run 3  Run 4  Run 5  Run 6  87  showed a major increase between 650°C and 750°C. However, for this catalyst, the  C2  selectivity only increased by a small amount over the same temperature range, and remained low at 850°C. This is due to complete oxygen consumption occurring at 750°C.  C, Yield  The C2 yield of the catalyst was significantly lower than that of the other catalysts, particularly at the higher temperatures.  CO„ Yield  The CO yield was higher at all conditions than for the other catalysts tested. X  This catalyst was a very effective combustion catalyst at 650°C. The observed effects may be due to a large concentration of active sites. According to McCarty (1991), too high a concentration of active sites can result in an effective combustion catalyst (see Section 2.3.4). However, the actual cause of the observed effects can only be speculated upon, due to the reactor configuration used for the tests. Another type of reactor may be more appropriate for interpretation of the results. For example, the use of a membrane reactor, as described in Section 2.6, may facilitate investigative testing. Where the methane and oxygen are separated by the catalyst, an increase in oxygen diffusivity and/or active sites, should result in an increased  C2 yield  with no negative  effects observed for a too active catalyst. If, on the other hand, the increased methane conversion was actually due to an increase in active sites for the undesirable products, no increase in C2 yield would likely be observed with the membrane reactor. 88  Figure 4.22  Effect of Catalyst Preparation on % C2 Selectivity 70.0 65.0 60.0 55.0 50.0  1:10 Ca  g 45.0  i  CNI  4o.o  0 35.0  30.0 25.0  1:10 Ca (RP)  20.0 15.0 Run 1  100% 02 Conversion Run 2  Run 3  Run 4  Run 5  Run 6  Figure 4.23  Effect of Catalyst Preparation on % C2 Yield 14.0 12.0 1:10 C 10.0  100 % 02 Conversion  4.0  I  2.0 Run 1 ^  Run 2  Run 3  Run 4  Run 5  Run 6  89  4.5^Scanning Electron Microscopy-Electron Dispersive X-Ray  The scanning electron microscopy-electron dispersive x-ray (SEM-EDX) technique was used to examine the surface of the catalyst particles and to measure dopant concentrations at and near the surface. The instrument used was a Kevex 8000, with a 40 A diameter beam of electrons.  The SEM photographs are presented in Appendix B. These photographs represent some of the different structures observed on the catalyst particles. All of the samples, with the exception of the catalyst prepared by the revised procedure, 1:10 Ca (RP), were similar. The photographs are of various structures which were found in most samples. The sample which gave one visibly different scan was the 1:100 K:Sm oxide. Although most of the particles examined for this sample were similar to the other samples, the structure revealed in photograph 103, a spheroid type surface, was not found in any of the other samples. This does not necessarily indicate that this structure is unique to the potassium doped sample, only that it was not found in the others during the scanning of a limited number of particles for each. The numbers on the photographs indicate some of the spots which were scanned with the EDX for component analysis.  Overall, the particle surfaces consisted of irregular-shaped and irregular-sized plate-like crystals, with primarily very rough surfaces; although small, smooth surface portions were also found. Based on the EDX analysis, the dopants do not appear to be uniformly distributed on the surface. Hence, it may be assumed that cation dopants 90  were non-uniformly dispersed throughout the Sm 2 03 crystal. However, no crystals containing large concentrations of dopant which were not incorporated into the samarium oxide crystal structure were observed. One generalization which can be made is that the very smooth surfaces appear to contain only samarium oxide. This suggests that one of the functions of the dopant is to increase the roughness of the surface. Unfortunately, a large amount of error is introduced into the EDX analysis for rough surfaces, such as are present in our samples. Therefore, the EDX analyses are approximate, and should be used as indicators of amounts of elements present, rather than as giving absolute numbers. The EDX does not detect elements of atomic number less than that of sodium; therefore, elements such as carbon and oxygen are not detected. The weight percent determined is the weight percent calculated considering only the elements with the atomic number of sodium or greater. The relative mole ratio of dopant to samarium can be calculated and compared to the nominal value for the bulk crystal (i.e., 1:100 or 1:10).  Calcium 1:100 Ca:Sm Photograph 192 The particles appear to consist of flat plates and fragments of plates of varying sizes, with considerable void space between the plates. The results of the EDX analyses are presented in Table 4.9. The presence of neodymium at spot 3 is likely a result of original contamination associated with the samarium.  91  Table 4.9 Weight % of Catalyst Surface Components According to SEM-EDX Element  1:100 Ca:Sm Bulk Mole Ratio  1:10 Ca:Sm Bulk Mole Ratio  192-1  192-2  192-3  193-1  193-2  Sm  75.44  71.61  78.35  82.84  82.81  Ca  0.62  0.11  .83  1.46  2.84  Si  2.01  1.83  0.60  nd  na  Al  0.52  6.82  1.08  1.01  na  Nd  na  na  3.99  nd  nd  Ca:Sm local mole ratio  2.4:100  0.6:100  4.0:100  0.7:10  1.3:10  nd: not detected na: not analyzed 1:10 Ca:Sm Photograph 193, 194 Two different particles were photographed for this catalyst to show some of the different structures present. Photograph 193 is of a sample which consists of fragmented flat plate-like crystals, with many viewed on end. Photograph 194 shows a smooth flat surface with a few small crystal fragments present. The analysis for photograph 193 is presented in Table 4.9.  1:10 Ca:Sm RP^Photograph 105 This sample appeared very different from the others. While still containing some of the plate-like crystals, much of the crystal structure was much smaller and had a feathery appearance. Spots 1 and 2 were both scanned and analyzed. The Ca peak for spot 2 was much larger than for spot 1. The analysis determined 1.36 weight % Ca for spot 2 compared to 0.77 weight % for spot 1, which correspond to Ca:Sm mole ratios  92  of 0.5:10 and 0.3:10, respectively.  Sodium The sodium peak was not observed for any of the samples analyzed due to the fact that it was obscured by one of the Sm 203 peaks.  1:100 Na:Sm Photograph 195 This area had a very rough surface and consisted of many broken plate-like crystals of varying sizes. A second area was scanned (no photograph) which was very different from the first, consisting of a very smooth plane with a few very small particles.  1:10 Na:Sm A sample was scanned and analyzed; however, no photograph was taken and no sodium was detected.  Samarium Oxide Photograph 102 The sample photographed consisted of large flat plates which were broken in irregular patterns at the edges. The area in the centre of the photograph was scanned by EDX and the trace revealed almost pure Sm 203, with small aluminum and silicon peaks.  Potassium Photograph 103 A small potassium peak below the method detection limit was observed on the EDX trace. This sample was different from others in that spheroid formations were seen. The magnification is slightly lower for this photograph in order to see the formations.  93  These formations were observed on only one particle, other particles being similar to other samples, such as that observed in photograph 192.  Magnesium Photograph 104  Photograph 104 shows a good example of needle- and plate-like crystals. This sample was not analyzed. In another sample which was analyzed, the scanned area was a flat plat in an area which was very similar to photograph 193. A small Mg peak was detected, which was calculated to be 4.5 weight% Mg, corresponding to a very high Mg:Sm mole ratio of 29:100.  4.6 Powder X-Ray Diffraction Analysis (XRD)  The samarium oxide catalyst, the 1:10 calcium and sodium catalysts, and the 1:10 Ca prepared by the revised method were examined using powder x-ray diffraction (XRD), on a Siemens, D5000 Diffractometer. In addition, a sample of monoclinic Sm203, which was prepared by treatment of Sm203 in an oxygen atmosphere at 1000°C for 4 hours, was analysed. The XRD traces are included in Appendix C. All of the cubic samples produced a "noisy" trace, possibly indicating that, although the sample was crystalline, the crystals contained numerous defects or dislocations. The noise of the trace may mask small peaks, making detection of small changes in the crystal structure difficult. No new peaks were detected for the samples as compared to Sm 203, although some of the peaks were shifted, indicating a change in distance between atoms in the lattice unit. The monoclinic and cubic traces differed considerably. The monoclinic trace was considerably less "noisy", which may be due to fewer defects and dislocations present 94  in the crystal, as well as larger crystallites, which is consistent with the long preparation time and the crystal structure (Anshits et al., 1990).  The three main peaks in the cubic Sm 203 crystal correspond to spacing between crystal lattice planes, d = 3.16 A for plane (h,k,l) = (2,2,2); d = 1.93 A for plane (4,4,0); and d = 2.73 A for plane (4,0,0). Due to the amount of noise on the XRD trace, it is difficult to detect small shifts in the peaks. There does not appear to be any significant change in the crystal structure for the 1:10 sodium and calcium doped samarium oxide. However, the 1:10 Ca (RP) sample indicated a definite shift of the three most intense peaks compared to that of the Sm 203, which indicates that some of the crystal dimensions were increased. The distance between lattice planes, d, increased from 3.16  A to 3.31 A for plane (2,2,2); from 1.93 A to 2.02 A for plane (4,4,0), and from 2.73 A to 2.86 A for plane (4,0,0). These increases in the crystal dimensions can be used to calculate the lattice constant for the crystal and to give an indication of the crystal structure. The lattice constant a increased from 10.9 A for the cubic Sm 203, to 11.4 A for the 1:10 Ca (RP), while still maintaining the cubic structure.  The lack of new peaks indicates that no new phases were formed within the sensitivity of the method. This indicates that the Na and Ca dopant cations were likely distributed throughout the Sm 2 03 crystal structure. No evidence of the monoclinic crystal structure was seen in the cubic catalysts.  95  4.7 Measurement and Effect of Catalyst Surface Area  Surface area may be an important variable for the catalytic oxidative coupling of methane, the effects of which have not been conclusively determined. Some studies reported in the literature suggest that an increase in surface area is detrimental to the selectivity, while others show the opposite effect. Most studies are not carried out with all other variables held constant, hence the difficulty in interpreting the results.  In a review of catalyst morphology for the oxidative coupling of methane by Martin and Mirodatos (1992), considerable evidence is presented that, for a variety of catalysts, the oxidative coupling of methane reaction is structure sensitive. The catalyst which has been studied the most and which is discussed in greatest detail is MgO. In a study which examined metal doped MgO samples, the specific surface area decreased from 70 m 2 /g for unpromoted MgO down to 1 m2 /g for a sodium content of 20 mole%. The selectivity toward C2 hydrocarbons increased in a parallel way from about 10 to 55% with a sodium content of 15 mole %.  Iwamatsu et al. (1987) also found that the addition of alkali compounds to MgO results in sintering (loss of area) and an improvement in C2 selectivity. It was speculated that the specific rate (per unit weight) of methyl radical formation increases with the surface area, leading to a higher specific rate of reaction. However, the rate of collision of radicals with the surface is increased. This would lead to total oxidation into CO2, thus decreasing the selectivity into C2 hydrocarbons.  96  In an extension of this study it was determined that for compounds with surface area below 100 m 2 /g, a low surface area gave a high  C2  yield (Iwamatsu et al., 1988).  Among the alkali compounds tested, 15 mole % Na/MgO, with a low surface area of 2 m2 /g gave the highest  C2  yield, at 750°C to 800°C. Although both the total  C2  yield  and CO2 yield had a strong dependence on surface area, the ratio of ethylene to ethane changed only slightly at low surface areas, and not at all at surface areas greater than 100 m2 /g, showing that the reaction of ethane to ethylene is not very strongly related to surface area.  In the above studies, the composition of the catalyst was varied to achieve the change in surface area and the observed results may also be due to other effects of the alkali dopants.  In the aforementioned study, Iwamatsu et al. also determined the effect of surface area due to calcination. When the MgO sample was sintered by calcination, the  C2  yield  increased as the surface area decreased. A similar procedure was followed for Na, K, and Cs doped MgO, with similar results. The following reaction scheme was devised in order to explain the apparent detrimental effect of surface area:  1 CH4 ^ CO 2 6  97  Steps 1, 2, 3, 6, and possibly 5 are heterogeneous and steps 4 and 5 are homogeneous. Step 4 should be independent of the surface area and steps 2 and 3 should proceed faster with increased surface area.  Parida and Rao (1991) prepared 15 atom % Li/MgO by various methods, resulting in a variation in surface area while the composition was held constant. The  C2 yield  and  selectivity were found to increase with a decrease in surface area.  The above studies indicate that increased surface area is detrimental to  C2 selectivity  and yield; however, other studies have shown that this is not always the case. For example, among three Li/MgO catalysts prepared by different techniques, the most selective toward C2 hydrocarbons had the greatest specific surface area (Lunsford 1989, 1990). In addition, when the specific surface area of MgO was varied from 70 to 28 m2 /g by sintering at 750°C in the reaction mixture (Martin and Mirodatos, 1992), the C2 selectivity  decreased from 6.2 to 3.8%, with the decrease in surface area.  In another study MgO was prepared by two different methods to compare the structure and catalytic properties (Martin and Mirodatos, 1992). One sample was prepared from the decomposition of basic carbonate and another was prepared by the oxidation of metal ribbons in air. The catalyst prepared from the carbonate gave a higher selectivity to C2 hydrocarbons (typically 60-70%) than that from the metal (about 30%). Analysis of the results based on unit surface area revealed that the total activity of the latter is twice that of the former and the intrinsic activity for the production of C 2 's is three times larger. The researchers concluded that the MgO produced from the basic 98  carbonate had a higher relative concentration of active sites for the oxidative coupling of methane reaction. The catalyst with the higher surface area displayed the higher selectivity for C2 hydrocarbons; therefore, the variation in selectivity can not be due to a possible detrimental effect of surface area.  Another study to examine the effect of surface area concentrated on samarium and lanthanum oxides (Korf et al., 1990b). Samarium oxide undergoes a phase change from the metastable cubic structure to the stable monoclinic structure at about 900°C. Lanthanum undergoes no such phase change. The surface area of samarium oxide was measured at 6.6 m 2 /g for cubic and 2.7 m 2 /g for monoclinic. The specific activity to C2  products based on the surface area of the cubic phase is much higher than that of  the monoclinic phase. The use of the same amount of surface area of the cubic and the monoclinic material led to a lower C2 production in the case of the latter. This shows that the surface area is not a major factor in the change in behaviour during the phase transition from cubic to monoclinic.  The lanthanum oxide was calcined at different temperatures to obtain different surface areas. A higher calcination temperature resulted in a lower surface area. Those samples calcined at low temperatures, and therefore having high surface areas, achieved relatively high C2 selectivity and  C2  yield compared with the results for the  high calcination temperature. The specific activity to  C2  products based on the surface  area was the same for these catalysts. When the same amount of surface area was used in the reactor the  C2  production was the same. In this case, sintering caused  deactivation of the La 2 03 catalysts. 99  The above results may also be affected by variables other than surface area. For example, the heat treatment and the variation in preparation methods may cause changes in the catalyst surface and bulk morphology.  From the above studies it appears that the effects of surface area on the selectivity and activity of the catalyst are still not fully understood. This stems from the difficulty in changing only the surface area of the catalyst without changing other morphologic or chemical properties of the catalyst. The use of dopants to cause sintering in catalysts results in improved catalysts; however, this is likely primarily due to factors other than the decreased surface area, such as increased basicity, increased oxygen diffusivity due to incorporation of defects in the crystal, etc.  The surface areas of the catalysts used in the present study were determined by the single point BET method (using nitrogen gas at liquid nitrogen temperature) for the undoped samarium oxide and the 1:100 mole ratio doped catalysts. There was insufficient sample to determine the surface area for the 1:10 doped catalysts. The surface areas are presented in Table 4.10.  The C2+ selectivities for all catalysts were graphed as a function of surface area to determine if a correlation existed (see Figure 4.24). The existence of a correlation at each temperature was tested by carrying out a population correlation calculation, which returns the covariance of two data sets divided by the product of their standard deviations. For this parameter, a value of 1 indicates a perfect positive correlation and -1 indicates a perfect negative correlation. For a value of zero, no correlation exists. 100  The calculated parameter for the C2+ selectivity was calculated to be -0.46, 0.12, and 0.18 at 650°C, 750°C, and 850°C, respectively. Therefore, it is clear that at 750°C and 850°C, no correlation exists between surface area and C2+ selectivity. However, at 650°C, the analysis is less clear, as a small negative correlation may exist. However, due to the scatter in the results, the only conclusion which can be made based on these data, is that surface area over the range tested (ca. 2 to 3 m2 /g) does not have a significant effect on C2+ selectivity. The methane conversion was also analyzed for correlation with surface area, and none was found to exist. These results are not surprising, based on the previous discussion, which presented several studies with conflicting results.  Table 4.10 Surface Area of Catalysts Catalyst  Surface Area (m 2 /g)  Sm  2.5  Ca/Sm (Oxygen)  2.4  Mg/Sm (Oxygen)  2.0  Na/Sm (Oxygen)  2.6  K/Sm (Oxygen)  3.1  Ca/Sm (Air)  2.0  Mg/Sm (Air)  2.2  Na/Sm (Air)  2.1  101  Figure 4.24  C2+ Selectivity as a Function of Surface Area 60.0 55.0 Z• . 50.0 0 Ci)  (  + CN1 0  0  ^ •  35.0  ^  •  ^ •  Reaction Temperature  o o  • •  n  ■ 650 C  •  •  45.0 40.0  ^  0  ■  •  750 C  • 850 C  •  • ■  ■  30.0 1.8^2^2.2^2.4^2.6^2.8^3  ^  3.2  Surface Area (m2/g)  102  4.8^Basicity of the Catalyst  Catalyst basicity has been shown by several studies in the literature to be an important variable in catalyst performance for the production of C2 hydrocarbons. In general it is found that the catalyst selectivity and yield increase with an increase in basicity.  The metal oxide catalysts, monoclinic (m) Sm 203, cubic (c) Sm 203, MgO and y-Al 203, were studied for the oxidative coupling of methane by Lapszewicz and Jiang (1992). The relative basicities of the compounds were determined by CO 2 temperature programmed desorption profiles (based on the temperature at which 50% of CO  2  desorbed), and decreased in the order m-Sm 203 > c-Sm 203 > MgO > y-Al 203, which is in the same order of decrease in C2 selectivity. The presence of multiple maxima at different temperatures on the desorption curves for all catalysts except y-Al 203 indicates the existence of several types of basic sites of different strength.  The doping of stable oxides by alkali compounds may result in an increase in basicity and an increase in C2 selectivity. The basicity was determined for various catalysts of 15 atom % Li/MgO prepared by various methods (Parida and Rao, 1991). It was found that the C2 yield increased with an increase in basicity, while the CO. formation decreased.  In a study of MgO/CaO catalysts (Philipp et al., 1992), over a wide range of MgO concentrations, the 85% MgO/CaO catalyst was determined to have the highest basicity, as well as being the most active and selective catalyst. 103  Baerns (1992) established the pK A of various catalyst oxides, where the pK A is a measure of the acid-base properties of the material. Table 4.11 illustrates how the C2+ selectivity is dependent on the value of pK A (pKA =7 is neutral, >7 is basic, and <7 is acidic). The results of this study give a clear indication of the positive relationship that is often found to exist between basicity and C2+ selectivity.  Table 4.11 Selectivity and Acid-Base Properties of Various Catalytic Oxides 12 Catalytic Oxide (weight %)  % C2+ Selectivity  pKA  1.5% Li20/Ca0  73  +12.2  3.0% Na20/Ca0  67  +9.3  4.5% K20/Ca0  67  +9.3  SiO2, kieselgel  15  +6.8  y-Al203  3  +3.3/+4.0  Al 203 x SiO2  1  -5.6  Baerns reviewed various studies which were carried out in order to investigate the effect of acidity and basicity. It was found that the addition of alkali in the form of sodium oxide to lead oxide deposited on strongly acidic supports, such as PbO/Al 2O3 SiO2, resulted in an increase in C2+ selectivity. This was interpreted as the neutralization of strong acidic sites by the alkali. An initial increase in methane conversion was also obtained which was reversed by additional sodium oxide; however, this was more than compensated for by the gain in selectivity.  12  Baerns, 1992 104  Increased lead oxide loading resulted in a decrease in the catalyst surface acidity, with a corresponding increase in the C2+ selectivity. Although the results were presented for supported lead oxide catalysts only, the following systems have also been studied by various authors and similar results reported: manganese oxide, nickel oxide, praseodymium oxide, and titanium oxide. Baerns acknowledges that the effect of basicity may not be the only variable affecting the C2+ selectivity and the contribution of various factors has not been elucidated.  The relative selectivities as a function of basicity of the alkaline earth oxides were also investigated (Baerns, 1992). For undoped alkaline earth oxides, the C2+ and C3 + selectivity were lowest for the least basic compound, BeO, and highest for the most basic compounds, SrO and BaO. The catalyst activity also increased from BeO to CaO (the SrO and BaO were not included in the analysis due to hydroxide formation and loss of surface area by sintering). They conclude that this indicates that the 0' anion plays an important role in the selective oxidative coupling of methane reaction.  The effect of various anions was investigated by testing the following calcium compounds: oxide, sulphate, silicate, aluminate, fluoride, and phosphate (Baerns, 1992). A significantly higher selectivity was obtained for CaO and CaSO 4 . The high selectivity obtained with the use of CaSO 4 as the catalyst may be due to at least some decomposition of CaSO4 at the reaction temperature to produce 0' anions.  The highly basic compounds (i.e., SrO and BaO) are not effective catalysts due to the reaction with water vapour which results in partly liquid hydroxides. However, highly  105  basic compounds can be prepared by the incorporation of alkali compounds into the surface of stable alkaline earth oxides like MgO and CaO. It was also found that C2+ selectivity increased with the amount of NaOH that is dispersed on CaO until a final value is reached. The increase in C2+ selectivity was not attributed to the alkali compound itself but to its interaction with CaO.  A catalyst of composition CaO/La 203 was prepared which resulted in a basicity for the compound which was greater than either of CaO or La 203 alone (Becker and Baerns, 1991). This catalyst also had an increased C2 selectivity. However, the maximum selectivity was not found to coincide with the maximum basicity. It was concluded by the investigators that there existed a slight dependency of selectivity on the number of acidic sites. However, the maximum in the acidity is shifted to higher lanthanum concentrations than that of C2+ selectivity. The tentative conclusion reached was that not only the basicity of the catalyst surface, but also its acidity affects C2+ selectivity.  The positive effect of basicity on the C2+ selectivity is widely thought to be due to the presence of negatively charged oxygen ions  (o  -  or 02). Although it is generally  accepted that the oxidative coupling reaction is initiated by the formation of methyl radicals on the catalyst surface, with the C-H bond cleavage supposed to be the rate limiting step, the actual mechanism involved is not clear. The relationship between basicity and reaction mechanisms has been investigated by several researchers.  Dubois (1992) suggests that high (Lewis) basicity is an important factor in the oxidative coupling of methane, due to the fact that basic oxides which possess defect structures  106  and no irreducible cations have very few mechanisms available for the re-equilibration of bulk and surface charge imbalance. The charge density can be decreased by liberating an electron (from 0 2-) to an anion defect site (oxygen vacancy). The new oxygen species  (0) can be viewed as a bulk oxygen possessing a positive charge, as  compared to the 0 2.  The released electron, trapped in an anion vacancy, serves  as the oxygen adsorption site on the catalyst surface. Therefore, the basicity of the bulk oxide is important in the expulsion of single electrons by oxygen (p-type semiconductivity) which is directly related to the capture of gas phase oxygen, a surface phenomenon.  The addition of alkali dopants may lead to an increase in basicity due to the dispersion of univalent alkali ions on the divalent MgO surface matrix (Parida and Rao, 1991). This would lead to the formation and stabilization of  a  ions, which would in turn lead  to a shift in the electrical charge of the catalyst surface. The presence of the 0 ions would also result in high surface basicity.  Lapszewicz and Jiang (1992) carried out tests using deuterium and oxygen isotopes to investigate the mechanism of methyl radical formation. They summarized the various hypotheses concerning the active sites into two concepts.  The first concept involves an acid-base type reaction, in which the adsorbed methane dissociates on the surface of the basic oxide, yielding W and CH3; subsequent subtraction of the electron from the unstable methyl anion leads to methyl radical formation. For this concept, the rate of methane conversion in the coupling reaction  107  can be expected to correlate with the catalyst's ability to effect heterolytic scission of the C-H bond, as measured by the rates of deuterium exchange.  The second concept involves the formation of a transition complex consisting of the adsorbed methane molecule and the surface oxygen, which subsequently dissociates to yield a methyl radical. The rate of methane conversion would be expected to correlate with the ability of the catalyst to activate oxygen, as measured by the rate of the 0-0 bond scission.  Neither of the above two anticipated correlations existed, indicating that the mechanism of methyl radical formation can not be explained by the ability of the catalyst surface to dissociate either methane or oxygen, or both.  The investigators suggest that the activation of methane can be expected to occur on electron-rich (basic) sites, while the oxygen would be activated on electron-deficient (acidic) sites, and thus the relationship between basicity and  C2  selectivity can be  interpreted in terms of strong activation of methane and weak activation of oxygen. For an acidic surface the opposite is true, making it a good oxidation catalyst. Their experiments with deuterium and oxygen isotopes support this hypothesis; that is, the ratio of the methane activation to oxygen activation increases with basicity and with C2 selectivity.  Baerns (1992) suggests that the negative effect of surface acidity, with the resulting complete oxidation, may be due to the following possible explanation for non-selective 108  oxidation: the Bronsted- or Lewis-type acidic sites cause the formation of a carbonium ion, which is easily attacked by a negatively charged adsorbed oxygen species (e.g., a or 02) leading to total combustion.  Baerns and Becker (1991) interpret their results for La 203 /Ca0 as evidence of a cooperative effect involving acidic sites and basic 0' sites. They suggested the following mechanism. If a heterolytic splitting of the methane molecule occurs, CH; 4- 11" CH4 " CH; + H#  then the CH3 ion may be stabilized by a Lewis acid centre, to which its electron is then transferred, and the CH 3 • species formed may either recombine on the surface, if closely located to each other, or they may desorb. The CH 3+ ion may interact with negatively charged oxygen existing as a or 02 on the surface, leading to nonselective oxidation. Both the electronic and geometric surface properties would affect the C2 + selectivity for such a mechanism.  They concluded that strong acidity favours total oxidation of the hydrocarbons, and that high basicity results in high C2+ selectivity. This may be due to either suppression of the nonselective pathways and/or promotion of the selective routes. Interaction of negatively charged oxygen either with a protonated methane molecule or CH 3+ species, or the reaction between radical-like oxygen and methyl species, may lead to total oxidation.  109  Based on the above studies, it can be concluded that, although the mechanism is not fully understood, high basicity appears to promote the formation of  C2 hydrocarbons,  and high acidity appears to promote the formation of total oxidation products.  The basicity in solids can be measured by various methods which will give an indication of relative basicities. These methods will sometimes produce different relative orders of basicity for different methods. One indication of acid/base strength in a metal oxide is the difference in electronegativity between the cation and the oxygen (Dubois and Cameron, 1990). An oxide having a cation with a low electronegativity will have a high partial charge on oxygen, and therefore will be strongly basic. The electronegativities of group IA and HA cations are presented in Table 4.12. Table 4.12 Electronegativity of Alkali and Alkaline Earth Elements 13 Group IA  Electronegativity  Group HA  Electronegativity  Li  1.0  Be  1.5  Na  0.9  Mg  1.2  K  0.8  Ca  1.0  Rb  0.8  Sr  1.0  Cs  0.7  Ba  0.9  theoretical increase in basicity  —} theoretical decrease in basicity For a binary oxide, the electronegativity x can be calculated as an average of the two  13  Brady and Humiston, 1982 (Originally from Linus Pauling, The Nature of the Chemical Bond). 110  metal cations using Pauling's scale of the electronegativities of metals:  xcaHon  = (1 + z) xmcta i^(4.1)  with z being the valence of the metal cation, and  Xoxide  = 0.5 (Xcation A + Xcation B)  ^  (4.2)  If the factor 0.5 in equation (4.2) is replaced with the atomic fraction of the cation present in the metal, equation (4.3) can be used to estimate the electronegativity of the doped catalysts.  Xoxide  atomic fraction A *  ;  aim A +  atomic fraction B * Xcation B (4.3)  The effect of valence number will be to increase the basicity with a decrease in valence number, for similar electronegativities. Therefore, substitution of Sin' with alkali or alkaline earth elements will increase the basicity.  The electronegativity can be taken as a first approximation of bulk basicity. It should be noted that the actual distribution of dopant is not known, and that the relationship between the bulk basicity and surface basicity is not clear. However, it follows that a catalyst with a high bulk basicity is likely to have a high surface basicity. Therefore, the bulk electronegativity will be used as a first approximation of effective catalyst basicity. 111  According to equation (4.3), the catalysts in the present study should increase in electronegativity in the order Na (1:10) < Ca (1:10) < K < Na < Ca < Mg < Sm, with a corresponding decrease in basicity. Table 4.13 Effect of Electronegativity on Catalyst Performance Catalyst  Na (1:10)  Ca (1:10)  K  Na  Ca  Mg  Sm  Electronegativity*  4.42  4.53  4.65  4.65  4.66  4.67  4.68  % Methane Conversion+  18.4  25  25.4  25.4  24.6  23.2  23  Selectivity  62.8  57.0  50.5  55.8  53.5  53.6  54.3  C2+  * as calculated by equation (4.3). TReadion 750°C, CH4 /02 =4 The methane conversion and  C2+  selectivity as functions of bulk electronegativity are  presented in Figures 4.25 and 4.26 (T Readion = 750°C, CH 4 /02 = 4). The correlation was calculated for these two sets of data in a similar manner to that used for surface area. A small positive correlation of 0.69 was indicated for methane conversion as a function of electronegativity, which is not significant at a 5% uncertainty level. It should be noted that the correlation coefficient is a measure of linearity. Review of Figure 4.25 indicates that if a relationship exists between methane and cation electronegativity, it may pass through a maximum. The correlation coefficient would not be an appropriate measure of such a relationship. A strong negative correlation of -0.89 was indicated for C2+ selectivity, which is significant at a 1% uncertainty level. As observed by other researchers, an increase in  C2+  selectivity was observed for an increase in basicity.  112  Figure 4.25  Methane Conversion as a Function of Cation Electronegativity T= 750C  29 c o 27 c o 23 0 o 19 2 0 17  ■  •^  25  ■  • ■^  Correlatior = 0.69  •  15 4.4^4.45^4.5^4.55^4.6  4.65  4.7  Cation Electronegativity  Figure 4.26  C2+ selectivity as a Function of Cation Electronegativity 65  T = 750C  •  60 .>  •  ■  TD) 55 a,  N  K^  ct, . 50 U 0 45  Cor relation = -0.4 6 (without IC  40  4.4  4.45^4.5^4.55^4.6  4.65  ^  4.7  Cation Electronegativity  113  The C2+ selectivity of the potassium doped catalyst was significantly lower than would be expected due to basicity. This may be due to factors other than basicity, such as carbonate formation or ionic radius. Potassium forms a stable carbonate up to temperatures higher than that of the other dopants used. The effect of carbonate formation is discussed fully in Section 4.9. It has been observed that the formation of very stable carbonates can be detrimental to catalyst performance (Kalenik and Wolf, 1992), possibly by a reduction in the number of lattice defects and a decrease in the capability of the oxygen atoms to diffuse through the carbonate. The ionic radius has also been identified as an important dopant property (Korf et al., 1988), and this is discussed in Section 4.10. Potassium has a larger radius than the other dopant ions, which are all of similar or smaller radius than Sm'. Therefore, the larger potassium ion may not be incorporated into the samarium oxide crystal lattice, or if it is, it may create stress on the lattice, resulting in a destabilization effect and a detrimental effect on catalyst selectivity. It would be reasonable to conclude that a factor other than basicity may be affecting the performance of the potassium doped catalyst.  If the correlation calculation for C2+ selectivity is carried out without the value for potassium included, the parameter calculated approaches that of a perfect negative correlation (-0.96), which is significant at a 0.1% uncertainty level. This is quite a strong correlation, considering that factors other than basicity, which have not been held constant, will undoubtedly have an effect on the catalyst performance. Therefore, it can be concluded that C2+ selectivity appears to be a function of basicity.  114  4.9 Effect of Carbon Dioxide and Carbonate Formation on Catalyst Performance  Several catalysts used for the oxidative coupling of methane are capable of forming high temperature stable carbonates. The amount of carbon dioxide which is produced in the oxidative coupling of methane reaction is more than sufficient to promote this formation. In addition, it is thought that the presence of CO2, even in the absence of carbonate formation, may interact with the catalyst to affect the activity and selectivity. Although these effects have been studied, due to the difficulty in clarifying the mechanisms involved, it is not clear what effect carbonates and carbon dioxide have on the oxidative coupling of methane reaction.  The effect of carbon dioxide addition on gas phase reactions was studied by van der Wiele et al. (1992) in a blank reactor at 800°C, by substituting part of the helium stream with CO2 in the feed gases. They found that carbon dioxide had no effect on the homogeneous reactions. This is in accordance with the fact that the decomposition of CO2 is very slow.  Suzuki et al. (1990) studied the effect of carbon dioxide on the oxidative coupling of methane reaction by introducing CO 2 into the feed. The use of carbon dioxide as a diluent in the feed gas resulted in an increase in ethane and ethylene yields, from a yield of 13.4% to 18.3%, and an increase in  C2 selectivity  C2  from 42.3% to 49.4% (MgO  catalyst, TR = 800°C, CH4 /02 = 2). The addition of carbon dioxide also suppressed the deactivation of the catalyst and decreased the amount of coke deposition. Partial adsorption of carbon dioxide on the surface may inhibit the deep oxidation of methane 115  or C2 products, resulting in increased C2 selectivities.  Several catalysts were tested, including doped and undoped Sm 203 and MgO, with varying preparation methods. For some catalysts the addition of CO 2 had very little effect while for others it was considerable, both on conversion and selectivity. It was apparent that the different methods of catalyst preparation also had a significant effect. The methane conversion decreased for several catalysts, while increasing for others, with the addition of CO2 . The C2 selectivity increased for all catalysts with the exception of CaO at higher concentrations of CO 2, and SrO, for which a considerable decrease was observed.  The authors explain the above as due to the formation of carbonates, according to the following equation:  MO + CO2^MCO3^M = Mg, Ca, Sr^(4.4)  These alkaline earths form carbonates with thermal stability in the order Sr > Ca > Mg. Strontium oxide is highly basic and would readily adsorb CO 2 from the gas phase; SrCO3 is stable up to almost 1900°C. Therefore, even if very small amounts of carbon dioxide are present in the feed gas, SrO would likely react with the carbon dioxide to form stable carbonate species, possibly resulting in the loss of catalytic activity.  Calcium carbonate is stable up to 1339°C, and AG for equation (4.4) is -13.7 kJ/mole. For CaO catalysts, the use of CO 2 was effective only when its partial pressure was very 116  low. The authors suggest that the addition of a small amount of carbon dioxide may cause competitive adsorption of introduced carbon dioxide and oxygen on the catalyst surface. The secondary oxidation of the  C2 or  suppressed, resulting in an increase in the  intermediates produced could be partly C2  yield and C2 selectivity. Higher  concentrations of carbon dioxide resulted in carbonates forming over the entire surface of the oxide and reduced the catalytic activity. Magnesium carbonate decomposes at the low temperature of 350°C. Therefore, the surface would not be covered with inactive carbonate species.  The investigators report that another effect of CO 2 addition may be the oxidation of the methane by the carbon dioxide, resulting in the formation of C2 compounds and carbon monoxide, according to the following equations:  CH4 + CO2 --> 0.5 C 2H4 + CO + H20  AG = 34.7 kJ/mole at 800°C^(4.5)  CH4 + 3 CO 2 ---> 4 CO + 2 H2O^AG = -44.8 kJ/mole at 800°C^(4.6)  The C2 yield may increase due to equation (4.5) by about 2%. However, carbon monoxide production (equation (4.6)) is thermochemically more favoured. This is substantiated by the reaction of methane with carbon dioxide without oxygen over Sm 2 03 /MgO, which gives carbon monoxide as the major product. An increase in carbon monoxide yield was observed over various catalysts when helium was replaced by carbon dioxide.  Peil et al. (1991a) studied the effect of CO 2 on the oxidative coupling of methane, over 117  Li/MgO and Sm 203, using SSITKA (steady-state isotopic transient kinetic analysis). Carbon dioxide addition to the Li/MgO catalyst resulted in a decrease in methane and oxygen conversion, with no change in selectivity, suggesting that all sites are equally active for both selective and nonselective oxidation. No significant effect was observed for the samarium oxide catalyst.  The difference between these two catalysts can be explained by the fact that Li/MgO readily forms surface carbonates, while samarium carbonate is stable only up to 530°C. The lack of effect on product selectivities suggests that carbon dioxide poisons all active sites on Li/MgO equally. This may be due to only one type of active site available on the surface, which may be selective or non-selective, depending on the oxidative/reductive state of the active site.  Dubois and Cameron (1991, 1992) studied the effect of stable carbonate formation on the selectivity of catalysts for the oxidative coupling of methane. They selected several catalysts, some of which form stable carbonates and some of which do not. Among these, yttria and thoria were selected due to their ability to form stable surface superoxides (02), possibly due to the absence of stable carbonate formation. The use of these two catalysts led to relatively high carbon monoxide selectivity, which may be a result of the aforementioned properties. This indicates that the addition of high temperature stable alkaline earth carbonate decreases CO synthesis activity while increasing overall catalytic activity. The decrease in CO formation is thought to be due to the destruction of surface superoxide by surface carbonate formation, possibly through a peroxycarbonate intermediate. 118  The ability of the catalyst to form carbonates is generally through the addition of doping compounds, such as Li, Na, K, La, Sr, and Ba. Of several catalysts tested, the oxides which do not possess stable carbonates and those which are not substantially carbonated at the temperature of the reaction exhibited high CO selectivities. The conversion and C2+ selectivity also tended to be lower for the non-carbonated oxides, although there were exceptions.  The temperature of maximum CO2 thermal decomposition decreased from that of the carbonate forming material, when mixed with a basic non-carbonate forming oxide. This may be due to a transfer of carbonate, or CO 2, from the carbonated species to the oxide, which subsequently decomposes. The authors suggest that the decomposition of surface carbonates should be an advantage under catalytic operation, due to an increase in anion vacancy sites, which are required for oxygen adsorption. As well, the decomposition of bulk forming carbonates would also increase crystal disorder. This is in agreement with Korf et al. (1990a), who found that in the Li/MgO system, the active sites were formed by the decomposition of the Li 2CO3, as discussed in Section 2.3.2.  The higher activity and selectivity observed for the catalysts which form stable carbonates is in disagreement with the hypothesis of the superoxide ion as the active oxygen species for generation of methyl radicals, since the superoxide ion is not found on the carbonate surface. The role of the carbonate has not been proven, but the authors suggest it may be related to its ability to alter the relative stability of the various oxygen species found on the surface, thereby reducing CO formation. 119  Kalenik and Wolf (1992) observed low C2 selectivity for zirconium dioxide, which they concluded was due to the formation of surface carbonates which decrease the number of lattice defects and decrease the capability of the oxygen atoms to diffuse through the carbonate.  It is difficult to draw conclusions about the relative merits of carbonate formation. An additional complication may be that the more basic oxides are also the ones more capable of forming high temperature stable carbonates. In summary, the above studies reveal the following conclusions. Formation of large amounts of carbonates, or very stable carbonates, appears to be detrimental to C2+ selectivity and methane conversion. On the other hand, adsorption of carbon dioxide, or formation of small amounts of carbonates may be beneficial to methane conversion and C2+ selectivity. An optimal ability to form carbonates may be the key to enhancing C2+ selectivity and yield. As indicated by Dubois and Cameron, this may be related to formation of the carbonate on the dopant and subsequent decomposition of the carbonate on the samarium oxide, which may occur on alkali and alkaline earth doped oxides, where the bulk material is incapable of forming stable carbonates.  The decomposition temperatures of the carbonates to the oxides for the elements used in the present study are listed in Table 4.14. The actual amount of carbonate decomposition will depend on the amount of carbon dioxide present, according to the reverse of reaction (4.4). If an increase in carbonate formation were to increase the C2 + yield, the order of the yield obtained with the catalysts doped at 1:100 mole ratio, and at 750°C should be as follows: K > Na > Ca > Mg = Sm. The actual order is Na > Ca 120  > K > Mg = Sm. With the exception of potassium, which forms very stable carbonates, the order follows as expected based on the above hypothesis. The poorer than expected behaviour of potassium may be due to a very strong carbonate formation, which would not decompose at reaction conditions.  Table 4.14 Decomposition Temperature of Various Carbonates' Carbonate TDecomposition  (°C)  Sm  Mg  Ca  Na  K  530  350  1339  851.  891*  melting point, these compounds decompose at a higher temperature.  4.9.1 Determination of Carbonate Formation for Sodium and Calcium Doped Samarium Oxide Catalysts  Tests were carried out using a thermogravimetric analyzer (TGA) in order to investigate the relative stabilities of carbonate formed on the calcium and sodium catalysts. Undoped samarium oxide was heated in a nitrogen stream up to 750°C, the temperature was then held constant and 100% CO 2 was passed over the sample for 25 minutes at 1 atmosphere pressure. The weight of the sample versus time was plotted in Figure 4.27. The shape of the loss in weight over time curve indicates that more than one type of weight loss occurred, which may be due to loss of surface water and bound water. The weight loss ceased when constant temperature was reached. A small steady decrease in weight is noticed after the CO 2 is admitted, indicating that no  14  CRC Handbook of Chemistry and Physics, 1991 121  Figure 4.27 Carbonate Formation on Samarium Oxide Under a 100% CO 2 Atmosphere 104–  – 1400  102 –  – 1200  100  – 1000 Z,  •  -  P  – 800 a 411 CL  E  96  600 z  –  I  1  94–  – 400  92–  92.14 CO2  –^ z 90^  Residua: X (59. 11 ens)^  -4--^-  – 200  0  0. 0^10. 0^20. 0^30. 0^40. 0^50. 0^60. 0^70. 0^BO. 0^90. 0^100. 0^110. 0 Time (min)^  DuPont. 1090  Figure 4.28 Carbonate Formation on a 1:10 Na:Sm Oxide Catalyst under a 100% CO Atmosphere  2  -104  402  100  • 500  94–  —300 tfk if 0.0^20.  90  1–  0^40.0  C  60. 0^80. 0  100.0^120. 0^140.0^160. 0  4  180. 0 200.0 22 . 0  200  Time (min) DuPont. 1090  122  appreciable carbonate formation occurs for the undoped samarium oxide at 750°C. The 1:10 Na:Sm and Ca:Sm catalysts were tested in the TGA for carbonate formation at each reaction temperature: 650°C, 750°C and 850°C.  Sodium  The 1:10 mole ratio sodium doped samarium oxide catalyst was heated in air to 650°C and held at that temperature for approximately 30 minutes (see Figure 4.28). The weight of the sample decreased at a slow rate similar to that observed for the pure samarium oxide. Carbon dioxide was then introduced into the TGA. With the exception of a small increase in weight approximately three minutes after the CO 2 was introduced, the slow loss in weight continued as before, for about 20 minutes. After 20 minutes under the CO 2 atmosphere the weight of the sample started to rapidly increase, which may be attributed to carbonate formation. The temperature was increased to 750°C, which again resulted in a slow decrease in the weight at the same rate as that which occurred at 650°C. When the temperature was increased to 850°C the weight dropped almost instantaneously, likely due to carbonate decomposition. After the period of rapid weight decrease attributed to carbonate decomposition, the weight continued to slowly decrease, at the same rate as observed at 650°C and 750°C. The increase and decrease in weight attributed to carbonate formation and decomposition were the same, indicating that no carbonate remained at 850°C.  Calcium  Commercially obtained calcium oxide was tested in the TGA in order to observe the  behaviour under a carbon dioxide atmosphere at reaction temperatures (see Figure 123  ^  Figure 4.29 Carbonate Formation on 100% Calcium Oxide under a 100% CO2 Atmosphere 150 - 1400 140 - 1200 130  U  - 1000 ° .4)  120  L  _c17)  J  1  V  ^/^  -  -4)  800 0 L  a_  110 - 600 100  /  - 400  /  90 - 200 80  N2 C°2 0 0.0^20.0^40. 0^60. 0^80.0^100.0^120. 0^140. 0^160.0^180. 0 200. 0 220. 0 ^Time (min)^ DuPont 1090  Figure 4.30 Carbonate Formation on a 1:10 Ca:Sm Oxide Catalyst under a 100% CO Atmosphere  2  104  -  1400  102-  -  1200  100  -  1000 p 0  L 0) 98-  - 800^-4) 0 40 /-•  a_  J  - 600  96/  94- -^/ --  - 400 --  - - I ---- - -  —200  92- / /^FA 1 k :^—,. CO —3-^1 90^1^I^i^4^4^4^I^I^I^I^1^1^4^I^I^I^4^I^1 ^0.0^20.0^40:0^60.0^80. 0^100.0^120. 0 , 140. 0^160. 0^180. 0^200. 0^220. 0  -  0  rime (min)  DuPont. 1092  124  4.29). The weight of the sample remained essentially steady up to almost 400°C under a nitrogen atmosphere, at which point the weight dropped suddenly. It levelled off briefly at 500°C and dropped again at a slower rate to 650°C. The carbon dioxide was then introduced. The weight decreased slightly for about 4 minutes, and then increased very rapidly back to 100%, likely due to carbonate formation. The increase in weight continued at a slower rate, again due to carbonate formation. This weight gain increased as the temperature was increased, first to 750°C, and then to 850°C. At each subsequent temperature the rate of weight increase decreased, indicating that less carbonate was being formed.  The 1:10 Ca:Sm oxide was tested in a similar manner to the sodium doped catalyst (see Figure 4.30). Considerably less carbonate formation was observed. The initial shape of the curve up to 650°C appeared to be a combination of the samarium oxide and calcium oxide curves, as would be expected. After the CO 2 was introduced, the weight increased by about 1%, followed by a small decrease of 0.2%, indicative of an overshoot in carbonate formation. The weight then levelled out and stared to increase very slowly, up to the time that the temperature was increased to 750°C. The weight then started to decrease slowly. When the temperature was increased to 850°C, the weight continued to decrease, at a rate of about 3 times the rate at 750°C. Over the time period for the test, it appears that not all of the carbonate formed was decomposed. This is in contrast to the sodium doped catalyst, for which it appears that all of the carbonate was decomposed upon increase of the temperature to 850°C.  The 1:100 Ca and Na doped samples were also tested. However, both showed only a 125  slight increase in weight after introduction of the carbon dioxide. The method is not sensitive enough to determine these low levels of carbonate formation. Therefore, none of the other samples were tested.  Based on the tests performed and information concerning the stability of the carbonates, the relative amount of carbonate formed for each catalyst at each temperature was estimated and graphed (see Figures 4.31 to 4.33). Potassium was not included as the amount of carbonate was difficult to estimate. Calcium and sodium (1:100) doped catalysts were assumed to have one tenth the amount of the 1:10 catalysts. Magnesium and samarium oxide were assumed to have no carbonate formation. The amount of carbonate formed is represented in relative numbers only, and is based on that formed in a 100% carbon dioxide atmosphere. The amounts were corrected based on relative amounts of gas phase CO2 over each catalyst. In actual operation of an oxidative coupling reactor, the concentrations of gas phase oxygen and water would also have an effect on the amount of carbonate formed, but these effects have not been included in this initial study.  As can be seen from these figures, the C2+ selectivity decreases with an increase in carbonate formation at 650°C, while at 750°C and 850°C it actually increases. One possible interpretation is that the state of equilibrium between carbonate formation and decomposition may be important. At 650°C, the carbonate is still forming, particularly for the sodium catalysts, which results in low C2+ selectivity. At 750°C, the carbonate is closer to a state of equilibrium, with both formation and decomposition probably occurring, resulting in an increase in C 2 + selectivity. At 850°C, the calcium catalysts 126  Figure 4.31  Catalyst Performance as a Function of Carbonate Formation Corrected for CO2 Concentration, T=650C 45.0 14,-Sm,Mg 40.0 35.0 30.0 Na^ 25.0 ae 20.0 15.0 10.0 5.0 0.0 0.00  Ca 1:10^Na^  0.50^1.00^1.50  2.00  2.50  Relative Carbonate Formation  C2+ yield^-0-- CH4 conversion --•- C2+ selectivity  Figure 4.32  Catalyst Performance as a Function of Carbonate Formation 70.0 60.0 50.0  Corrected for CO2 Concentration, T=750C Sm, Mg _1:10 Ca ______ _; 1..__  40.0 30.0 20.0 10.0 0.0 0.00  ^. 1:10 Na  ..„  Na  Ca a• -•  •  0.50^1.00^1.50^2.00  2.50  3.00  Relative Carbonate Formation  C2+ yield  CH4 conversion --•- C2+ selectivity  127  Figure 4.33  Catalyst Performance as a Function of Carbonate Formation Corrected for CO2 Concentration, T=850C 60.0  Ca 50.0 k-.-. • Sm, Mg, Na, 1:'0 Na 40.0 30.0 ^  20.0  _  I:10 Ca •  ^  ^  10.0 0.0 0.00^0.01^0.02^0.03^0.04^0.05^0.06^0.07^0.08^0.09 Relative Carbonate Formation  — C2+ yield^—0-- CH4 conversion --•— C2+ selectivity  Figure 4.34  Ratio of CO to CO2 Produced 0.35 Reaction Temperature  0.30 0.25  •  8 0.20 U  650 C  ^ 750 C  8 0.15  la 850 C  0.10 0.05 0.00  E  E  ()  0  c) 0) 2  oN 0  z  o  CEn  z  128  were the only catalysts which were estimated to have carbonate present, which resulted in an increase in C2+ selectivity for the 1:10 catalyst.  The methane conversion appeared to improve with a small amount of carbonate present and then decreased, at both 650°C and 750°C. At 850°C, only a small increase in methane conversion was observed, which is consistent with the results at the lower temperatures for the small amount of carbonate present. The actual amount of carbonate present may affect the methane conversion, with no dependence on whether it is forming or decomposing. A small optimum amount appears to improve methane conversion.  The CO /CO 2 ratios obtained in this study were compared to those obtained by Dubois and Cameron. This ratio is graphed for all catalysts in Figure 4.34. The most significant effect is found for the Na (1:10), for which the CO /CO 2 ratio is considerably lower than for the other dopants. This indicates that the carbonate formation is also much more significant. If the effect for the Na (1:10) may be explained by greater carbonate formation than for the other catalysts, the low methane conversion is consistent with ideas and results reported by Peil et al. (1991a), who suggested that a lower methane conversion is a result of carbonate formation.  Overall, the mechanisms underlying the effects observed due to carbonate formation remain unclear. The improved C2+ selectivity may be due to modification of the ratio of basic to acid sites. This ratio was suggested by Becker and Baerns (1991) to be an important variable. Another possibility is that the physical processes of formation and 129  decomposition of the carbonate may lead to increased crystal disorder, resulting in an increase in oxygen vacancies.  4.9.2 Effect of Gas Phase Composition on Carbonate Formation  The level of carbonate formation will also depend on the amount of oxygen and water in the gas phase, according to equations such as:  Na2 02 + CO 2 .r." Na 2 CO3 + 1 /2 02  (4.7)  Na2CO3 + H2O i(- 2 NaOH + CO 2  (4.8)  2 NaOH + 1 /2 0 2 4- Na2 02 + H2 O  (4.9)  The amount of each solid species in the oxide/carbonate/hydroxide system present under equilibrium conditions can be estimated by free energy minimization if it is assumed that the solids have unit activity. This was simulated using Aspen Plus software, for pure sodium oxide in the presence of a dilute gas phase mixture of oxygen, water, and carbon dioxide. Using this thermodynamic approach and assuming all species to be solids, approximately twice as much NaOH was formed as Na 2CO3 or Na2O at each of the reaction temperatures (650°C, 750°C, and 850°C). The CO 2 and water were almost completely consumed. At such temperatures pure NaOH would in practise be present as a liquid, so the thermodynamics are in fact much more complicated than a simple model can represent. Further complications arise because the sodium is assumed to be dispersed throughout the Sm 2 03 crystal, perhaps in a solid  130  solution; again the actual behaviour can not be determined from the study of the pure Na20. For example, it has been shown that the decomposition temperature of stable carbonates decreases when the carbonate forming material is mixed with a basic noncarbonate forming oxide (Dubois and Cameron, 1991, 1992). Therefore, the sodium oxide is likely significantly stabilized by the samarium oxide. The formation of carbonate and hydroxide will likely occur, but to an unknown extent which is certainly less than that for pure sodium oxide. The effect of the gas phase composition can be generalized, as in the following discussion.  At lower oxygen conversion, the oxygen concentration in the gas phase will be higher, which will inhibit carbonate formation. The carbonate forming catalysts generally had lower oxygen conversions. This may be due to the formation of a carbonate layer, which would interfere with the gas phase/surface oxygen exchange; this in turn would result in a slower regeneration of active sites and therefore a lower methane conversion. The effect on the C2+ selectivity would depend on the nature of the active sites and whether the carbonate layer affects this concentration negatively.  The net effect of the oxygen concentration would appear to be to limit the carbonate formation according to reaction (4.7), as the carbonate forming catalysts result in lower oxygen conversion, and therefore, higher oxygen concentrations.  Lower oxygen conversion should also result in lower water concentrations. The effect of water on the carbonate equilibrium should be to inhibit carbonate formation and increase the formation of MOH or M(OH) 2 , which will result in different active sites, 131  the effect of which is not known.  Therefore, those catalysts which tend to lower degrees of carbonate formation, have gas phase compositions with lower oxygen concentration and higher water concentration, resulting in higher hydroxide concentrations. Those with a greater tendency toward carbonate formation, have gas phase compositions with higher oxygen concentration and lower water concentration, which tends to limit the amount of carbonate formation and hydroxide formation in favour of the oxide.  4.10 Ionic Radius of Dopant  It has been suggested that the ionic radius of the dopant material may have an affect on the ability of the dopant to promote the oxidative coupling reaction. For example, IC' has a large ionic radius compared to Sin' (see Table 4.15). This may result in difficulty in incorporation of the potassium ion into the samarium oxide crystal lattice. This is consistent with the results in this study, which show that the potassium doped catalyst was the least selective catalyst at 750°C and 850°C. The other dopants used in this study all have ionic radii which are similar to or smaller than that of samarium. Therefore, these cations should be incorporated into the crystal lattice.  The magnesium cation is smaller than either the sodium or calcium. Doping samarium oxide with magnesium actually had a detrimental effect on C2+ selectivity. A possible explanation is that the small cation incorporated into the crystal lattice distorts the crystal, resulting in a less selective catalyst, possibly due to lower oxygen mobility, 132  which has been indicated as a factor in selective catalysts.  Table 4.15 Estimated Ionic Radii of Selected Elements 15 Group IIIB and Lanthanides  Ionic Radius (A)  0.99-1.10  Sclf  0.81-0.91  Sr+  1.12-1.27  Y3+  0.92-1.07  Ba2+  1.34-1.35  La3+  1.14-1.22  Sm3+  1.0-1.13  Group IIA  Ionic Radius (A)  0.95-0.98  Mg2+  0.65-0.8  1.33  Ca2+  Group IA  Ionic Radius (A)  Li+  0.6-0.78  Na+ K+  The effect of ionic radius of dopants has been discussed with regard to samarium oxide as a means of explaining catalyst performance (Korf  et al., 1988). The addition of Li has  a detrimental effect on the catalytic performance, which has been shown by XRD to be due to a destabilizing effect which favours the cubic-monoclinic transformation. Ca and Na have been shown to have a beneficial effect. The tendency of the additives to stabilize the cubic phase is reported to be in the order: Ca > Mg > La. Na has also been shown to induce a decrease in stability. These effects may be due to the ionic radii of the different ions and the structure of the pure promoter phase that would be present under these conditions. The CaO and MgO are cubic, and their ions fit into the surface of Sm2 03 without stress due to the size of their ions. La' has a larger radius and a smaller stabilising effect. Na 2 CO3, Li 2 CO3, and LiOH have a monoclinic structure, which encourages the phase transition of Sm 2 03 from cubic to monoclinic.  'Dubois and Cameron, 1990 133  Of the dopant cations used in this study, potassium (K+) alone has a larger radius than Sm3+, and is even larger than that of La', which was observed to reduce the stability of the Sm 203 . In addition, due to the much larger radius, the K+ ions may not fit into the crystal structure and therefore may not have the desired doping effect. This may account for the low C2+ selectivity obtained with the potassium doped catalyst, which could not be explained based on basicity (see Section 4.8).  Ohno and Moffat (1991) determined a dependence of selectivity on the ratio of the cation radius to charge, for alkali and alkaline earth phosphate catalysts. They plotted the C2+ selectivities obtained for various catalysts, over a range of 0% to 75%, against the radius/charge ratio, which was found to pass through a maximum. The lowest radius/charge ratio is for the smallest cation with a +2 valence (Mg). The C2 + selectivity is lowest for this catalyst, and increases with cation radius. The highest selectivity was obtained with the smallest cation with +1 valence (Li), and the C2 + selectivities decreased as the radius/charge ratio increased.  They interpret this by viewing the ion radius/ion charge as an approximation of the one-dimensional space occupied per unit charge by the particular cation. The cation may enhance the ability of the surface oxygen atom to extract a hydrogen atom from methane through a polarization effect. If so, this effect may be related to the charge density. However, too high a polarization could lead to undesirable results; that is, a decrease in selectivity.  Comparison of Ohno and Moffat's study to the present results reveals a similar effect, 134  although the cation concentrations in this study are much lower, resulting in small differences between cations (see Table 4.16). There is no significant difference in the C2+ selectivity for Mg and Ca dopants.  Table 4.16 Effect of Cation Radius/Charge Ratio on C2+ Selectivity Dopant  Mg2+  Cat'  Na+  K+  Cation radius/charge  0.40  0.55  0.98  1.33  % C2+ Selectivity  53.6  53.5  55.8  50.5  + TR = 750°C, CH 4 /02 =4  This effect may also be explained as a result of electronegativity, which was discussed previously. As the cation radius increases, the positive charge has less effect on the oxygen ion, causing the oxygen to have a higher partial charge, and in effect be more basic. At high values of cation radius to charge ratio, as in the case of potassium, the basicity may be too high. This may result in a change in the oxygen species, which may lead to increased COX yield. As can be seen in Figure 4.35, the potassium catalyst is the most active in CO, production at 750°C and 850°C.  135  Figure 4.35  136  5. CONCLUSIONS  The addition of alkali and alkaline earth dopants to the samarium oxide catalyst resulted in changes in catalyst performance. Based on the work conducted in the project, the following conclusions were made:  i)  The concentration of dopant used had a significant effect on catalyst performance.  ii)  No new phases were observed in the Sm 2O3 crystal upon addition of the sodium and calcium dopant cations, indicating that the cations were dispersed throughout the crystal, although possibly not uniformly.  iii)  The catalyst preparation procedure used in this study was different than that used by other researchers. It was found that the catalyst preparation can have a significant effect on the catalyst performance, which has also been observed by others. The results were interpreted in this case to indicate that there may be an optimum number of active sites, with too many active sites resulting in an effective combustion catalyst.  iv)  The effect of catalyst surface area has been studied by many researchers with varying conclusions as to the effect on the oxidative coupling of methane. The results of this study indicated that over the surface area tested, this variable did not have a significant effect on the catalyst performance. 137  v)  The basicity of the catalyst appears to have a significant effect on the catalyst performance, with an increase in basicity resulting in an increase in C2 + selectivity. These results confirm those found by other researchers.  vi)  The performance of the catalysts is temperature dependent. There is an optimum temperature for maximum C2+ yield, which is dependent on the amount and nature of the dopant. The formation of carbonates on the catalyst surface is likely partially responsible for the temperature dependent behaviour of the catalysts. The mechanism of carbonate interaction with the catalyst and reaction molecules is not known. However, the process of decomposition of the carbonate may create active sites, thereby increasing the C2+ selectivity. An attempt was made to quantify the carbonate formation on the sodium and calcium doped catalysts, for which little information was available in the literature.  vii) The results support the hypothesis that the ionic radius of the cation dopant must be similar to or smaller than the support cation to achieve effective inclusion in the crystal lattice.  138  6.^REFERENCES Aigler, J.M. and J.H. Lunsford (1991), "Oxidative Dimerization of Methane over MgO and Li+/Mg0 Monoliths", Applied Catalysis, 70, 1991, pp. 29-42. Alcock, C.B., J.J. Carberry, R. Doshi, and N. Gunasekaran (1992), "Coupling Reactions on Oxide Solid Solution Catalysts", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 123-129. Amenomiya, Y., V.I. Birss, M. Goledzinowski, and J. Galuszka (1990), "Conversion of Methane by Oxidative Coupling", Catalysis Review-Science Engineering, 32, 1990, pp. 163-227. Amorebieta, V.T. and A.J. Colussi (1988), "Kinetics and Mechanism of the Catalytic Oxidation of Methane over Lithium-Promoted Magnesium Oxide", Journal of Physical Chemistry, 92, 1988, pp. 4576-4578. Amorebieta, V.T. and A.J. Colussi (1989), "Mass Spectrometric Studies of the Low-Pressure Oxidation of Methane on Samarium Sesquioxide", Journal of Physical Chemistry, 93, 1989, pp. 5155-5158. Andorf, R., L. Mleczko, D. Schweer and M. Baerns (1991), "Oxidative Coupling of Methane in a Bubbling Fluidized Bed Reactor", Canadian Journal of Chemical Engineering, Vol. 69, August, 1991. Anshits, A.G., E.N. Voskresenskaya and L.I. Kurteeva (1990), "Role of Defect Structure of Active Oxides in Oxidative Coupling of Methane", Catalysis Letters 6, 1990, pp. 67-76. Anshits, A.G., E.N. Voskresenskaya and A.N. Shigapov (1991), "Prediction of the Oxide Systems Catalytic Properties in Methane Oxidative Coupling", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 49-55. Aparicio, L.M., S.A. Rossini, D.G. Sanfilippo, J.E. Rekoske, A.A. Trevino, and J.A. Dumesic (1991), "Microkinetic Analysis of Methane Dimerization Reaction", Industrial Engineering Chemistry Research, 30, 1991, pp. 2114-2123. Asami, K., K. Omata, K. Fujimoto, and H. Tominaga (1987), "Oxidative Coupling of Methane in the Homogeneous Gas Phase under Pressure", Journal of the Chemical Society, Chemical Communications, 1987, pp. 1287-1288. Baerns, M. "Basic Solids as Catalysts for the Oxidative Coupling of Methane", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 382-402.  139  Becker, S. and M. Baerns (1991), "Oxidative Coupling of Methane over La 203-Ca0 Catalysts, Effect of Bulk and Surface Properties on Catalytic Performance", Journal of Catalysis 128, 1991, pp. 512-519. Borchert, H., Z.L. Zhang, and M. Baerns (1992), 'The Effect of Oxygen Ion Conductivity of Catalysts for their Performance in the Oxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 111-116. Brady, J.E. and G.E. Humiston (1982), Third Edition, General Chemistry, Principles and Structure, John Wiley & Sons, Inc., 1982. Burch, R., S.C. Tsang, C. Mirodatos, and J.G. Sanche (1990), "Kinetic Isotope Effects in Methane Coupling of a Reducible Oxide Catalyst", Catalysis Letters, 7, 1990, pp. 423-430. Buyevskaya, 0.A., M.P. Vanina, N.F. Saputina, and V.D. Sokolovskii (1992), "Effect of Steam in the Oxidative Coupling of Methane over MgO and CaO Based Catalysts", Catalysis Today, 13, 1992, pp. 589-592. Cant, N.W., C.A. Lukey, P.F. Nelson, and R.J. Tyler (1988), 'The Rate Controlling Step in the Oxidative Coupling of Methane over a Lithium-promoted Magnesium Oxide Catalyst", Journal of the Chemical Society, Chemical Communications, 1988, pp. 766-768. Choudhary, V.R. and V.H. Rane (1991), "Acidity/Basicity of Rare-Earth Oxides and Their Catalytic Activity in Oxidative Coupling of Methane to C2-Hydrocarbons", Journal of Catalysis, 130, 1991, pp. 411-422. Deboy, J.M. and R.F. Hicks (1988), Industrial Engineering Chemistry Research, 27, 1988, p. 1577. Dooley, K.M. and J.R.H. Ross (1992), "Potassium/Calcium/Nickel Catalysts for Oxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 93-97. Driscoll, D.J., W. Martir, J.X. Wang, and J.H. Lunsford (1985), "Formation of Gas-Phase Methyl Radicals over MgO", Journal of the American Chemical Society, 107, 1985, pp. 58-63. Dubois, J.-L. and C.J. Cameron (1990), "Common Features of Oxidative Coupling of Methane Cofeed Catalysts", Applied Catalysis, 67, 1990, pp. 49-71. Dubois, J.-L. and C.J. Cameron (1991), "Synergy between Stable Carbonates and Yttria in Selective Catalytic Oxidation of Methane", Chemistry Letters, 1991, pp. 1089-1092.  140  Dubois, J.-L. and C.J. Cameron (1992), "Surface Carbonate on Methane Coupling Catalysts: A Poison or a Catalyst Itself?", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 85-88. Edwards, J.H. and R.J. Tyler (1988), "The Production of Liquid Fuels via the Catalytic Oxidative Coupling of Methane", Eds. D.M. Bibby et al., Methane Conversion, Elsevier Science Publishers B.V., Amsterdam, 1988, pp. 395-401. Edwards, J.H., R.J. Tyler and S.D. White (1990), "Oxidative Coupling of Methane over Lithium-Promoted Magnesium Oxide Catalysts in Fixed-Bed and Fluidized-Bed Reactors", Energy & Fuels, 4, 1990, pp. 85-93. Edwards, J.H., K.T. Do, and R.J. Tyler (1991), "The 'OXCO' Process for Natural Gas Conversion Via Methane Oxidative Coupling", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 489-495. Ekstrom, A. and J.A. Lapszewicz (1988a), "Methane Adsorption on a Working Samarium Oxide Catalyst and its Role in Hydrocarbon Formation during High Temperature Partial Oxidation.", Journal of the Chemical Society, Chemical Communications, 1988, pp. 797-799. Ekstrom, A. and J.A. Lapszewicz (1988b), "Investigation of the Mechanism of the Partial Oxidation of Methane over Samaria Catalysts", Preprints, Symposia, American Chemical Society, September 25-30, 1988, pp. 430-436. Ekstrom, A. and J.A. Lapszewicz (1988c), "The Role of Oxygen in the Partial Oxidation of Methane over a Samarium Oxide Catalyst", Journal of the American Chemical Society, 110, 1988, pp. 5226-5228. Ekstrom, A. and J.A. Lapszewicz (1989a), "A Study of the Mechanism of the Partial Oxidation of Methane over Rare Earth Oxide Catalysts Using Isotope Transient Techniques.", Journal of Physical Chemistry, 93, 1989, pp. 5230-5237. Ekstrom, A., J.A. Lapszewicz and I. Campbell (1989b), "Origin of the Low Limits in the Higher Hydrocarbon Yields in the Oxidative Coupling Reaction of Methane", Applied Catalysis, 56, 1989, pp. L29-L34. Ekstrom, A., R. Regtop and S. Bhargava (1990), "Effect of Pressure on the Oxidative Coupling Reaction of Methane", Applied Catalysis, 63, 1990, pp. 253-269. Ekstrom, A. (1992), "The Oxidative Coupling of Methane: Reaction Pathways and Their Process Implications", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 99-137. Feng, Y., J. Niiranen, and D. Gutman (1991a), "Kinetic Studies of the Catalytic  Oxidation of Methane. 1. Methyl Radical Production of 1% Sr/La 2 0 3 ", Journal 141  of Physical Chemistry, 95, 1991, pp. 6558-6563. Feng, Y., J. Niiranen, and D. Gutman (1991b), "Kinetic Studies of the Catalytic Oxidation of Methane. 2. Methyl Radical Recombination and Ethane Formation over 1% Sr /La 203 ", Journal of Physical Chemistry, 95, 1991, pp. 6564-6568. Feng, Y., J. Niiranen, L.N. Krasnoperov, and D. Gutman (1992), "Fundamental Studies of Gas/Surface Processes Occurring during Methane coupling on 1% Sr/La 203 ",  Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 161-166.  Fox, J.M., III, T.P. Chen, and B.D. Degen (1990), "An Evaluation of Direct Methane Conversion Processes", Chemical Engineering Progress, April 1990, pp. 42-50. Fujimoto, K., K. Omata, and J. Yoshihara (1991), "Selective Oxidation of Methane to Ethane and Ethene with Metal Oxides in Molten Metals", Applied Catalysis, 67, 1991, pp. L21-L24. Geerts, J.W.M.H., J.M.N. van Kasteren, and K. van der Wiel (1989), 'The Investigation of Individual Reaction Steps in the Oxidative Coupling of Methane over Lithium Doped Magnesium Oxide", Catalysis Today, 4, 1989, pp. 453-461. Grzybek, T. and M. Baerns (1991), "Surface Interaction between Methane and Alkali/Alkaline-Earth Oxide Catalysts", Journal of Catalysis 129, 1991, pp. 106-113. Hair, L.M., W.J. Pitz, M.W. Droege, and C.K. Westbrook (1992), "Modelling of Catalytic Oxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 188-195. Hamid, H.B.A. and R.B. Moyes (1991), "Oxidative Coupling of Methane on Sodium Aluminate- and Magnesium Oxide-Supported Samarium Oxide", Catalysis Today, 10, 1991, pp. 267-273. Hutchings, G.J., M.S. Scurrel, and J.R. Woodhouse (1988), "The Role of Gas Phase Reaction in the Selective Oxidation of Methane", Journal of the Chemical Society, Chemical Communications, 1988, pp. 253-255. Hutchings, G.H., J.R. Woodhouse, and M.S. Scurrell (1989a), "Partial Oxidation of Methane over Oxide Catalysts", Journal of the Chemical Society, Faraday Transactions, 1, 85(8), 1989, pp. 2507-2523. Hutchings, G.H., M.S. Scurrell, and J.R. Woodhouse (1989b), "Selective Oxidation of Methane in the Presence of NO: New Evidence on the Reaction Mechanism", Journal of the Chemical Society, Chemical Communications, 1989, pp. 765-766. Hutchings, G.J., M.S. Scurrell, and J.R. Woodhouse (1989c), "Partial Oxidation of Methane over Samarium and Lanthanum Oxides: A Study of the Reaction Mechanism", Catalysis Today, 4, 1989, pp. 371-381. 142  Hutchings, G.J. and M.S. Scurrell (1992), "Studies of the Mechanism of the Oxidative Coupling of Methane Using Oxide Catalysts", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 200-258. Iwamatsu, E., T. Moriyama, N. Takasaki, and K. Aika (1987), "Importance of the Specific Surface Area of the Catalyst in Oxidative Dimerization of Methane over Promoted Magnesium Oxide", Journal of the Chemical Society, Chemical Communications, 1987, pp. 19-20. Iwamatsu, E., T. Moriyama, N. Takasaki, and K. Aika (1988), "Oxidative Coupling of Methane over Promoted Magnesium Oxide Catalysts; Relation between Activity and Specific Surface Area", Eds. D.M. Bibby et al., Methane Conversion, 1988, pp. 373-382. Kaddouri, A. et al. (1989), "Oxidative Coupling of Methane over LnLiO 2 Compounds (Ln = Sm, Nd, La)", Applied Catalysis, 51, 1989, pp. L1-L6. Kalenik, Z. and E.E. Wolf (1990), "Comments on the Effect of Gas-Phase Reactions on Oxidative Coupling of Methane", Journal of Catalysis, 124, 1990, pp. 566-569. Kalenik, Z. and E.E. Wolf (1991a), "Transient and Isotopic Studies of the Oxygen Transport and Exchange During Oxidative Coupling of Methane on Sr Promoted La203 ", Catalysis Letters 9, 1991, pp. 441-450. Kalenik, Z. and E.E. Wolf (1991b), "Methane Oxidative Coupling over Lithium Promoted Lanthanum-Titanate Oxide", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 97-105. Kalenik, Z. and E.E. Wolf (1992), "Isotopic Studies of the Effect of Promotion on the Activity of La2O3, ThO2, and ZrO 2 Catalysts during Oxidative Dimerization of Methane", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 117-122. Kennedy, E.M. and N.W. Cant (1991), "Comparison of the Oxidative Dehydrogenation of Ethane and Oxidative Coupling of Methane over Rare Earth Oxides", Applied Catalysis, 75, 1991, pp. 321-330. Korf, S.J. et al. (1988), "The Effect of Promoters on the Behaviour of Sm 2 03 Catalysts for the Oxidative Coupling of Methane", Preprints, Symposia, American Chemical Society, September 25-30, 1988, pp. 437-442. Korf, S.J., J.A. Roos, J.M. Diphoorn, R.H.J. Veehof, J.G. van Ommen, and J.R.H. Ross (1989), "The Selective Oxidation of Methane to Ethane and Ethylene over Doped and Un-doped Rare Earth Oxides", Catalysis Today, 4, 1989, pp. 279-292. Korf, S.J., J.A. Roos, N.A. De Bruijn, J.G. van Ommen, and J.R.H. Ross (1990a), "Lithium Chemistry of Lithium Doped Magnesium Oxide Catalysts Used in the Oxidative 143  Coupling of Methane", Applied Catalysis, 58, 1990, pp. 131-146. Korf, S.J., J.G. van Ommen, and J.R.H. Ross (1990b), 'The Oxidative Coupling of Methane over Sm 203 and La 203 ", Symposium on Structure-Activity Relationships in Heterogeneous Catalysts, Division of Petroleum Chemistry, Inc., American Chemical Society, 1990, pp. 54-57. Korf, S.J., J.G. van Ommen and J.R.H. Ross (1991), "The Oxidative Coupling of Methane over Sm203 and La203 ", Eds. Grasselli, R.K. and A.W. Sleight, Structure-Activity and Selectivity Relationships in Heterogeneous Catalysis, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 117-126. Kuo, J.C.W. (1991), "Evaluation of Direct Methane Conversion Processes", Eds. de Lasa, H.I., G. Dogu, and A. Ravella, Chemical Reactor Technology for Environmentally Safe Reactors and Products, Series E: Applied Sciences - Volume 225, Kluwer Academic Publishers. Labinger, J.A. (1988), "Oxidative Coupling Of Methane: An Inherent Limit to Selectivity?", Catalysis Letters, 1, 1988, pp. 371-376. Labinger, J.A. (1991), "Mechanism-Imposed Limitations on the Yield of Higher Hydrocarbons from the Oxidative Coupling of Methane, and Alternate Approaches to Methane Conversion", Symposium on Methane Upgrading, Division of Petroleum Chemistry, American Chemical Society, April 14-19, 1991, pp. 151-154. Lapszewicz, J.A. and X.-Z. Jiang (1992), "Activation of Reactants by Metal Oxide Catalysts - A Key to Selectivity of Oxidative Methane Coupling?", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 102-110. Lee, J.S. and S.T. Oyama (1990), "Comments on the Mechanism of Oxidative Coupling of Methane", Reaction Kinetics Catalysis Letters, Vol. 41, No. 2, 1990, pp. 257-263. Lee, A.K.K. and A.M. Aitani (1991), "Methane Conversion Technology and Economics", Fuel Science and Technology International, 9(2), 1991, pp. 137-158. Leyshon, D.W. (1991), "Thin Bed Reactor for Conversion of Methane to Higher Hydrocarbons", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 497-507. Lide, D.R. Ed. (1991), CRC Handbook of Chemistry and Physics, CRC Press, 1991. Lin, C.H., K.D. Campbell, J.X. Wang, and J.H. Lunsford (1986), "Oxidative Dimerization of Methane over Lanthanum Oxide", Journal of Physical Chemistry, 90, 1986, pp. 534-537. Lo, M.Y., S.N. Kamat, and G.L. Schrader (1988), "Kinetics and Mechanism of Methane Oxidative Coupling over Samarium Oxide", Preprints of Papers Presented at the 144  196th ACS National Meeting, 33, 3, 1988, pp. 378-386. Lunsford, J.H. (1989), "The Role of Surface-Generated Gas-Phase Radicals in Catalysis", Langmuir, 5, 1989, pp. 12-16. Lunsford, J.H. (1990), 'The Catalytic Conversion of Methane to Higher Hydrocarbons", Catalysis Today, Vol. 6(3), 1990, pp. 235-259. Lunsford, J.H. (1991), 'The Catalytic Conversion of Methane to Higher Hydrocarbons", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 3-13. Machin,I., P. Pereira, V. de Gouveia, and F. Rosa (1992), "Modelling of the Catalytic Oxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 173-179. Martin, G.A. and C. Mirodatos (1987), "Evidence of Carbene Formation in Oxidative Coupling of Methane over Lithium Promoted Magnesium Oxide", Journal of the Chemical Society, Chemical Communications, 1987, pp. 1393-1394. Martin, G.A. and C. Mirodatos (1992), "Morphological Aspects of Catalysts for Oxidative Coupling of Methane", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 351-381. McCarty, J.G. (1991), "Kinetic and Thermodynamic Descriptions of Co-oxidative Methane Dimerization", Symposium on Methane Upgrading, Division of Petroleum Chemistry, American Chemical Society, April 14-19, 1991, pp. 142-150. Mims, C., R.B. Hall, K.D. Rose, and G.R. Myers (1989), "Oxidative Dimerization of CH4 /CD4 Mixtures: Evidence for Methyl Intermediate", Catalysis Letters, 2, 1989, pp. 361-368. Mirodatos, C. and G.A. Martin (1988), "Oxidative Coupling of Methane on Li-Mg Oxides: Kinetics and Mechanism Involving Carbenic Intermediates", Proceedings, 9th International Congress on Catalysis, Vol. 2, 1988, pp. 899-906. Moneuse, C., M. Cassir, C. Piolet, and J. Devynck (1990), "Oxidative Coupling of Methane in Molten Barium Hydroxide at 800°C", Applied Catalysis, 63, 1990, pp. 67-76. Murthy, T.K.S. and C.K. Gupta (1980), "Rare Earth Resources, Their Extraction and Purification", Eds. E.C. Subbarao and W.E. Wallace, Science and Technology of Rare Earth Materials, Academic Press, 1980, pp. 3-24. Nelson, P.F., C.A. Lukey, and N.W. Cant (1989), "Measurements of Kinetics Isotope Effects and Hydrogen/Deuterium Distributions over Methane Oxidative Coupling Catalysts", Journal of Catalysis, 120, 1989, pp. 216-230. 145  Nelson, P.F., E.M. Kennedy and N.W. Cant (1991), "Isotopic Labelling Studies of the Mechanism of the Catalytic Oxidative Coupling of Methane", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 89-95. Ohno, T. and J.B. Moffat (1991), "The Oxidative Coupling of Methane on Alkali- and Alkaline Earth-Phosphate Catalysts", Catalysis Letters, 1991, pp. 23-34. Olsbye, U. and G. Desgrandchamps (1991), "Preliminary Tests for a Kinetic Study of the Oxidative Coupling of Methane over a 25% Ba/La203 Catalyst", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 131-137. Olsbye, U., G. Desgrandchamps, K.-J. Jens, and S. Kolboe (1992), "A Kinetic Study of the Oxidative coupling of Methane over a BaCO3 /La20.(CO3)3, Catalyst", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 180-187. Omata, K., S. Hashimoto, H. Tominaga and K. Fugimoto (1989), "Oxidative Coupling of Methane using a Membrane Reactor", Applied Catalysis, 52, 1992, pp. L1-L4. Otsuka, K., K. Jinno, and A. Morikawa (1985), "The Catalysts Active and Selective in Oxidative Coupling of Methane", Chemistry Letters, The Chemical Society of Japan, 1985, pp. 499-500. Otsuka, K., K. Jinno, and A. Morikawa (1986a), "Active and Selective Catalysts for the Synthesis of C 2 H4 and C 2H6 via Oxidative Coupling of Methane", Journal of Catalysis, 100, 1986, pp. 353-359. Otsuka, K. and K. Jinno (1986b), "Kinetic Studies on Partial Oxidation of Methane over Samarium Oxides", Inorganica Chimica Acta, 121, 1986, pp. 237-241. Otsuka, K. and A.A. Said (1987), "Role of Lattice Oxygen Atoms in Partial Oxidations of Methane, Ethane and Ethylene over Samarium Oxides", Inorganica Chimica Acta, 132, 1987, pp. 123-128. Otsuka, K. and T. Komatsu (1987), "High Catalytic Activity of Sm 203 for Oxidative Coupling of Methane into Ethane and Ethylene", Chemistry letters, The Chemical Society of Japan, 1987, pp. 483-484. Otsuka, K. and T. Nakajima (1987), "Oxidative Coupling of Methane over Samarium Oxides using N 20 as the Oxidant", Journal of the Chemical Society, Faraday Transactions, 1, 83, pp. 1315-1321. Otsuka, K., A.A. Said, K. Jinno and T. Komatsu (1987), "Peroxide Anions as Possible Active Species in Oxidative Coupling of Methane", Chemistry Letters, 1987, pp. 77-80. 146  Otsuka, K., M. Inaida, Y. Wada, T. Komatsu, and A. Morikawa (1989), "Isotopic Studies on Oxidative Methane Coupling over Samarium Oxide", Chemistry letters, The Chemical Society of Japan, 1989, pp. 1531-1534. Otsuka, K., Y. Murakami, Y. Wada, A.A. Said, and A. Morikawa (1990), "Oxidative Coupling of Methane, Ethane, and Propane with Sodium Peroxide at Low Temperatures", Journal of Catalysis 121, 1990, pp. 122-130. Otsuka, K. and M. Hatano (1992), "Partial Oxidation of Methane over Metal Oxides: Reaction Mechanism and Active Species", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 78-98. PaHda, K.M. and S.B. Rao (1991), "Importance of Specific Surface Area and Basic Sites of the Catalyst in Oxidative Coupling of CH4 over LiO/MgO Catalysts Prepared by Precipitation Methods", Reaction Kinetics Catalysis Letters, Vol. 44, No. 1, 1991, pp. 95-101. Peil, K.P., J.G. Goodwin, Jr., and G. Marcelin (1989), "An Examination of the Oxygen Pathway during Methane Oxidation over a Li/MgO Catalyst", Journal of Physical Chemistry, 93, 1989, pp. 5977-5979. Peil, K.P., Marcelin G., J.G. Goodwin, Jr., and A. Kiennemann (1990a), "Transient Kinetic Analysis of the Oxidative Coupling of Methane over Sm 203", Ed. Dekker, Chemical Industries, 46, Novel Production Methods of Ethylene and Light Hydrocarbons, 1990, pp. 61-74. Peil, K.P., J.G. Goodwin, Jr., and G. Marcelin (1990b), "Surface Concentrations and Residence Times of Intermediates on Sm 2O3 during the Oxidative Coupling of Methane", Journal of the American Chemical Society, 112, 1990, pp. 6129-6130. Peil, K.P., J.G. Goodwin, Jr., and G. Marcelin (1991a), "Influence of Product CO 2 on the Overall Reaction Network in the Oxidative Coupling of Methane", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 73-79. Peil, K.P., J.G. Goodwin, Jr., and G. Marcelin (1991b), "Surface Phenomena during the Oxidative Coupling of Methane over Li/MgO", Journal of Catalysis, 131, 1991, pp. 143-155. Peil, K., G. Marcelin, and J.G. Goodwin, Jr. (1992), "The Role of Lattice Oxygen in the Oxidative Coupling of Methane", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 138-167. Peng, X.D., D.A. Richards, and P.C. Stair (1990), "Surface Composition and Reactivity of Lithium-Doped Magnesium Oxide Catalysts for Oxidative Coupling of Methane", Journal of Catalysis, 121, 1990, pp. 99-109. 147  Pereira, P., S.H. Lee, G.A. Somorjai, and H. Heinemann (1990), "The Conversion of Methane to Ethylene and Ethane with Near Total Selectivity by Low Temperature (<610°C) Oxyhydrogenation over a Calcium-Nickel-Potassium Oxide Catalyst", Catalysis Letters, 6, 1990, pp. 255-262. Philipp, R., K. Omata, A. Aoki, and K. Fujimoto (1992), "On the Active Site of MgO/CaO Mixed Oxide for Oxidative Coupling of Methane", Journal of Catalysis, 134, 1992, pp. 422-433. Pinabiau-Carlier, M., A. Ben Hadid, and C.J. Cameron (1991), "The Effect of Total Pressure on the Oxidative Coupling of Methane Reaction Under Cofeed Conditions", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 183-190. Poirier, M.G., A.R. Sanger, and K.J. Smith (1991), "Direct Catalytic Conversion of Methane", The Canadian Journal of Chemical Engineering, Vol. 69, October 1991, pp. 1027-1035. Rasko, J., G.A. Somorjai, and H. Heinemann (1992), "Catalytic Low-Temperature Oxydehydrogenation of Methane to Higher Hydrocarbons with very High Selectivity at 8-12% Conversion", Applied Catalysis A: General, 84,1992, pp. 57-75. Roos, J.A., S.J. Korf, R.H.J. Veehof, and J.G. van Ommen (1989a), "Reaction Path of the Oxidative Coupling of CH 4 over a Li/MgO Catalyst, Factors affecting the Rate of Total Oxidation of C 2H6 and C2 H4 .", Applied Catalysis, 52, 1989, pp. 147-156. Roos, J.A., S.J. Korf, R.H.J. Veehof, and J.G. van Ommen (1989b), "An Investigation of the Comparative Reactivities of Ethane and Ethylene in the Presence of Oxygen over Li/MgO and Ca/Sm 203 Catalysts in Relation to the Oxidative Coupling of Methane", Catalysis Today, 4, 1989, pp. 441-452. Sokolovskii, V.D., O.V. Buyevskaya, L.M. Plyasova, and G.S. Litvak (1990), "Structure and Catalytic Properties of Samarium-Calcium Oxide Catalysts in Methane Oxidative Dehydrodimerization", Catalysis Today, 6, 1990, pp. 489-495. Spinicci, R. (1991), "Properties of Zinc Oxide Based Catalysts Towards Methane Coupling as Studied by Transient Response Method", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 173-181. Statman, D.J., J.T. Cleaves, D. McNamara, P.L. Mills, G. Fornasari, and J.R.H. Ross (1991), "TAP Reactor Investigation of Methane Coupling over Samarium Oxide Catalysts", Applied Catalysis, 77, 1991, pp. 45-53. Suzuki, T., K. Wada and Y. Watanabe (1990), "Effects of Carbon Dioxide and Catalyst Preparation on the Oxidative Dimerization of Methane", Applied Catalysis, 59, 1990, pp. 213-225. 148  Tong, Y., M.P. Rosynek, and J.H. Lunsford (1989), "Secondary Reactions of Methyl Radicals with Lanthanide Oxides: Their Role in the Selective Oxidation of Methane", The Journal of Physical Chemistry, Vol. 93, No. 8, 1989. Tong, Y. and J.H. Lunsford (1991), "Mechanistic and Kinetic Studies of the Reactions of Gas-Phase Methyl Radicals with Metal Oxides", Journal of the American Chemical Society, Vol. 113, No. 13, 1991, pp. 4741-4746. van der Wiele, K., J.W.M.H. Geerts, and J.M.N. van Kasteren (1992), "Elementary Reactions and Kinetic Modelling of the Oxidative Coupling of Methane", Ed. E.E. Wolf, Methane Conversion By Oxidative Processes Fundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992, pp. 259-319. Vermeiren, W.J.M., I.D.M.L. Lenotte, J.A. Martens, and P.A. Jacobs (1991), "Perovskite-Type Complex Oxides as Catalysts for the Oxidative Coupling of Methane", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 33-40. Walsh, D.E., D.J. Martenak, S. Han, and R.E. Palermo (1992a), "Direct Oxidative Methane Conversion at Elevated Pressure and Moderate Temperatures", Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 147-152. Walsh, D.E., D.J. Martenak, S. Han, and R.E. Palermo (1992b), "Direct Oxidative Methane Conversion at Elevated Pressure and Moderate Temperatures", Industrial Engineering Chemistry Research, 31, 1992, pp. 1259-1262. Walsh, D.E., D.J. Martenak, S. Han, R.E. Palermo, J.N. Michaels, and D.L. Stern (1992c), "Pressure, Temperature, and Product Yield Relationships in Direct Oxidative Methane Conversion at Elevated Pressures and Moderate Temperatures", Industrial Engineering Chemistry Research, 31, 1992, pp. 2422-2425. Yamashita, H., Y. Machida, and A. Tomita (1991), "Oxidative Coupling of Methane with Peroxide Ions over Barium-Lanthanum-Oxygen Mixed Oxide", Applied Catalysis A: General, 79, 1991, pp. 203-214. Yates, D.J.C. and N.E. Zlotin (1988), "Blank Reactor Corrections in Studies of the Oxidative Dehydrogenation of Methane", Journal of Catalysis, 111, pp. 317-324.  149  APPENDIX A: RESULTS OF REACTOR CATALYST TESTS  150  Table A.1 Samarium Oxide Prepared in Oxygen Run 1 Temperature ( °C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He CH4/02  Run 2  Run 3  Run 4  Run 5  Run 6  Run 8  Run 7  650  650  650  650  750  750  850  850  10 5 85  20 5 75  40 5 55  80 5 15  20 5 75  40 5 55  20 5 75  40 5 55  2  4  8  16  4  8  4  8  Average Average Average Average Average Average Average Average Run 7 Run 8 Run 2 Run 3 Run 4 Run 5 Run 6 Run 1 Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) gmo1/min Output (p.mol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out p.mol Carbon lost % Carbon lost  6.96 2.73 1.04 0.10 0.05 0.12 0.00 8.45  15.34 1.77 1.15 0.24 0.12 0.32 0.02 17.66  33.04 1.41 1.15 0.38 0.21 0.63 0.04 36.36  72.08 1.21 1.09 0.55 0.27 1.03 0.06 76.51  14.12 0.37 1.55 0.38 0.51 0.58 0.04 18.34  31.81 0.30 1.29 0.53 0.76 1.04 0.08 37.46  13.87 0.00 1.70 0.58 0.62 0.50 0.03 18.48  31.73 0.00 1.48 0.71 0.93 0.84 0.06 37.63  30.04 22.27 99.49 4119  30.16 22.87 98.88 4094  30.93 23.47 96.23 3985  32.46 23.87 91.62 3794  30.86 24.00 96.28 3986  31.52 23.83 94.30 3905  31.05 23.50 95.84 3968  31.58 23.50 94.22 3901  287 112 43 4 2 5 0 348  628 72 47 10 5 13 1 723  1316 56 46 15 8 25 2 1449  2734 46 41 21 10 39 2 2902  563 15 62 15 20 23 2 731  1242 12 51 21 30 40 3 1463  550 0 68 23 25 20 1 733  1238 0 58 28 36 33 2 1468  31 8  122 14  125 8  232 9  114 11  111 7  112 11  105 7  151  Table A.1 (continued) Samarium Oxide Prepared in Oxygen Average Average Average Average Average Average Average Average Run 7 Run 6 Run 8 Run 3 Run 4 Run 5 Run 1 Run 2 STY tunol/g/s C2s  2.4  6.1  11.1  16.5  14.4  23.4  14.9  23.0  % Conversion CH4 02  17.6 44.6  13.2 64.6  9.1 71.9  5.8 75.8  23.0 92.6  15.1 94.0  25.0 100.0  15.7 100.0  % Selectivity CO2 CO COX selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity  69.6 6.7 76.3 7.1 16.5 23.7 0.0 23.7  49.5 10.3 59.8 10.6 27.5 38.1 2.1 40.2  34.6 11.4 46.1 12.4 37.9 50.3 3.6 53.9  24.6 12.4 37.1 12.2 46.7 58.9 4.1 62.9  36.7 9.0 45.7 24.2 27.3 51.5 2.8 54.3  22.9 9.4 32.2 26.9 36.7 63.5 4.2 67.8  36.9 12.5 49.4 26.9 21.8 48.7 1.9 50.6  25.1 12.0 37.1 31.4 28.5 59.9 3.0 62.9  % Yield CO2 CO CO, yield C2H4 C2H6 C2 yield C3's C2+ yield  12.3 1.2 13.5 1.3 2.9 4.2 0.0 4.2  6.5 1.4 7.9 1.4 3.6 5.0 0.3 5.3  3.2 1.0 4.2 1.1 3.5 4.6 0.3 4.9  1.4 0.7 2.1 0.7 2.7 3.4 0.2 3.6  8.5 2.1 10.5 5.6 6.3 11.9 0.7 12.5  3.5 1.4 4.9 4.1 5.5 9.6 0.6 10.2  9.2 3.1 12.3 6.7 5.4 12.2 0.5 12.6  3.9 1.9 5.8 4.9 4.5 9.4 0.5 9.9  152  Table A.2 Calcium Doped Samarium Oxide (1:100) Prepared in Air Average Average Average Average Average Average Average Average Run 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Temperature (°C)  650  650  650  650  750  750  850  850  Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  6.61 3.31 1.16 0.27 0.05 0.12 0.00 8.37  15.27 2.65 1.27 0.47 0.11 0.30 0.00 17.82  33.64 2.07 1.24 0.67 0.18 0.58 0.02 37.13  72.23 1.33 1.15 0.87 0.25 0.97 0.04 76.81  14.43 0.80 1.62 0.54 0.45 0.53 0.02 18.62  32.81 0.71 1.35 0.67 0.66 0.93 0.03 38.11  14.14 0.06 1.77 0.54 0.57 0.47 0.02 18.59  32.73 0.03 1.61 0.62 0.82 0.75 0.05 38.27  % Conversion CH4 02  21.1 33.7  14.3 46.9  9.4 58.6  6.0 73.5  22.5 84.1  13.9 85.7  23.9 98.9  14.5 99.5  % Selectivity CO2 CO COx selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity  65.8 15.3 81.1 5.7 13.2 18.9 0.0 18.9  49.6 18.3 67.9 8.9 23.2 32.1 0.0 32.1  35.4 19.2 54.6 10.5 33.2 43.7 1.7 45.4  25.2 19.0 44.2 10.9 42.2 53.2 2.6 55.8  38.6 13.0 51.6 21.7 25.3 47.0 1.4 48.4  25.5 12.7 38.2 24.9 35.2 60.1 1.7 61.8  39.8 12.2 52.0 25.7 21.0 46.7 1.4 48.0  29.0 11.3 40.3 29.8 27.2 57.0 2.7 59.7  % Yield CO2 CO CO x yield C2H4 C2H6 C2 yield C3's C2+ yield  13.9 3.2 17.1 1.2 2.8 4.0 0.0 4.0  7.1 2.6 9.7 1.3 3.3 4.6 0.0 4.6  3.3 1.8 5.1 1.0 3.1 4.1 0.2 4.3  1.5 1.1 2.6 0.7 2.5 3.2 0.2 3.3  8.7 2.9 11.6 4.9 5.7 10.6 0.3 10.9  3.6 1.8 5.3 3.5 4.9 8.4 0.2 8.6  9.5 2.9 12.4 6.1 5.0 11.2 0.3 11.5  4.2 1.6 5.8 4.3 3.9 8.2 0.4 8.6  153  Table A.3 Calcium Doped Samarium Oxide (1:100) Prepared in Oxygen Run 1 Temperature ( °C) Input flows (mL/min) Nominal (at 21.1°C) CH4  02  He CH4/02  Run 2  Run 3  Run 4  Run 5  Run 6  Run 7  Run 8  650  650  650  650  750  750  850  850  10 5 85  20 5 75  40 5 55  80 5 15  20 5 75  40 5 55  20 5 75  40 5 55  2  4  8  16  4  8  4  8  Average Average Average Average Average Average Average Average Run 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  6.26 0.99 1.06 0.21 0.06 0.13 0.00 7.91  14.54 1.64 1.19 0.40 0.14 0.34 0.01 17.12  32.73 1.27 1.17 0.59 0.22 0.67 0.03 36.38  70.63 1.01 1.14 0.78 0.31 1.12 0.06 75.59  13.67 0.30 1.58 0.49 0.55 0.58 0.04 18.12  31.57 0.26 1.33 0.61 0.81 1.03 0.07 37.41  13.53 0.00 1.71 0.53 0.65 0.49 0.03 18.14  31.34 0.00 1.54 0.64 0.95 0.79 0.05 37.15  29.62 25.93 99.67 4127  30.08 25.50 98.27 4069  30.62 25.50 96.54 3998  31.67 26.00 93.19 3859  30.43 26.00 96.96 4015  31.00 26.00 95.20 3942  30.51 26.00 96.73 4005  31.29 26.00 94.31 3905  Output (pinol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  249 39 45 9 2 5 0 318  592 67 48 16 6 14 0 696  1309 51 47 24 9 27 1 1454  2725 39 44 30 12 43 2 2917  549 12 63 20 22 23 2 727  1244 10 52 24 32 41 3 1475  542 0 68 21 26 20 1 727  1224 0 60 25 37 31 2 1451  psnol Carbon lost % Carbon lost  60.9 16.1  148.6 17.6  119.0 7.6  217.8 6.9  117.6 13.9  98.9 6.3  118.6 14.0  122.7 7.8  Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) ilmol/min  154  Table A.3 (continued) Calcium Doped Samarium Oxide (1:100) Prepared in Oxygen Average Average Average Average Average Average Average Average Run 7 Run 8 Run 6 Run 5 Run 4 Run 2 Run 3 Run 1 STY Lunol/g/s C2s  2.6  65  11.9  18.4  15.1  24.2  15.2  22.6  % Conversion CH4 02  21.0 80.2  15.0 67.1  10.0 74.6  6.6 79.8  24.6 93.9  15.6 94.9  25.4 100.0  15.6 100.0  % Selectivity CO2 CO CO X selectivity C2H4 C2H6 C2 selectivity Cg's C2+ selectivity  64.2 12.9 77.1 6.8 16.1 22.9 0.0 22.9  46.2 15.6 61.8 10.6 26.4 37.0 1.2 38.2  32.2 16.3 48.4 12.2 36.9 49.1 2.5 51.6  22.9 15.8 38.7 12.4 45.3 57.6 3.6 61.3  35.5 11.0 46.5 24.7 26.1 50.8 2.7 53.5  22.8 10.5 33.3 27.7 35.4 63.1 3.6 66.7  37.1 11.6 48.6 28.2 21.2 49.4 2.0 51.4  26.5 11.0 37.4 32.9 27.1 60.0 2.6 62.6  % Yield CO2 CO CO X yield C2H4 C2H6 C2 yield Cg's C2+ yield  13.5 2.7 16.2 1.4 3.4 4.8 0.0 4.8  7.0 2.3 9.3 1.6 4.0 5.6 0.2 5.7  3.2 1.6 4.9 1.2 3.7 4.9 0.2 5.2  1.5 1.0 2.5 0.8 3.0 3.8 0.2 4.0  8.7 2.7 11.4 6.1 6.4 12.5 0.7 13.1  3.6 1.6 5.2 4.3 5.5 9.9 0.6 10.4  9.4 2.9 12.4 7.2 5.4 12.6 0.5 13.1  4.1 1.7 5.9 5.1 4.2 9.4 0.4 9.8  155  Table A.4 Magnesium Doped Samarium Oxide (1:100) Prepared in Air Average Average Average Average Average Average Average Average Run 5 Run 6 Run 7 Run 8 Run 1 Run 2 Run 3 Run 4 Temperature (°C)  650  650  650  650  750  750  850  850  Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  6.47 2.34 0.97 0.17 0.06 0.12 0.00 7.96  14.87 1.93 1.07 0.30 0.13 0.30 0.01 17.14  32.54 1.56 1.05 0.45 0.21 0.59 0.03 35.73  70.31 1.20 1.00 0.66 0.32 1.01 0.06 74.83  13.68 0.45 1.47 0.45 0.50 0.54 0.04 17.81  31.50 0.29 1.22 0.57 0.74 0.96 0.07 36.90  13.69 0.03 1.66 0.55 0.58 0.47 0.03 18.09  31.96 0.00 1.50 0.68 0.83 0.75 0.05 37.44  % Conversion CH4 02  18.8 53.2  13.3 61.3  8.9 68.9  6.0 75.9  23.2 91.0  14.6 94.2  24.3 99.5  14.6 100.0  % Selectivity CO2 CO COx selectivity C2H4 C2H6 C2 selectivity Cs's C2+ selectivity  64.7 11.2 75.9 7.6 16.5 24.1 0.0 24.1  46.9 13.2 60.2 11.4 26.7 38.1 1.7 39.8  32.9 14.2 47.1 13.4 36.7 50.1 2.8 52.9  22.2 14.5 36.8 14.3 44.9 59.2 4.0 63.2  35.5 10.8 46.4 24.4 26.3 50.7 2.9 53.6  22.6 10.6 33.2 27.4 35.5 62.9 3.9 66.8  37.8 12.5 50.3 26.4 21.2 47.6 2.0 49.7  27.5 12.4 39.8 30.2 27.3 57.5 2.7 60.2  % Yield CO2 CO CO x yield C2H4 C21-16 C2 yield Cs's C2+ yield  12.1 2.1 14.2 1.4 3.1 4.5 0.0 4.5  6.2 1.8 8.0 1.5 3.5 5.1 0.2 5.3  2.9 1.3 4.2 1.2 3.3 4.5 0.3 4.7  1.3 0.9 2.2 0.9 2.7 3.6 0.2 3.8  8.2 2.5 10.7 5.7 6.1 11.8 0.7 12.4  3.3 1.5 4.9 4.0 5.2 9.2 0.6 9.8  9.2 3.0 12.2 6.4 5.2 11.6 0.5 12.1  4.0 1.8 5.8 4.4 4.0 8.4 0.4 8.8  156  Table A.5 Magnesium Doped Samarium Oxide (1:100) Prepared in Oxygen Run 1 Temperature (°C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He CH4 /02  Run 2  Run 3  Run 4  Run 5  Run 6  Run 7  Run 8  650  650  650  650  750  750  850  850  10 5 85  20 5 75  40 5 55  80 5 15  20 5 75  40 5 55  20 5 75  40 5 55  2  4  8  16  4  8  4  8  Average Average Average Average Average Average Average Average Run 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Products (%) CH4 02 CO2 CO C2H4 C2116 C3's Total Carbon out  6.35 2.34 0.98 0.13 0.05 0.12 0.00 7.80  14.50 1.70 1.10 0.27 0.13 0.32 0.01 16.79  31.77 1.39 1.10 0.41 0.21 0.63 0.04 35.08  69.37 1.14 1.05 0.62 0.30 1.06 0.06 73.95  13.68 0.45 1.47 0.45 0.50 0.54 0.04 17.81  30.87 0.33 1.25 0.56 0.77 1.01 0.08 36.48  13.35 0.00 1.65 0.57 0.63 0.49 0.03 17.89  31.18 0.00 1.45 0.70 0.91 0.79 0.06 36.92  30.16 23.67 98.62 4084  30.08 24.00 98.75 4089  30.65 24.57 96.75 4006  32.21 24.93 91.95 3807  30.73 24.93 96.36 3990  31.28 25.23 94.57 3916  30.84 25.13 95.97 3974  31.53 24.90 93.94 3890  Output (i.unol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  259 96 40 5 2 5 0 318  593 70 45 11 5 13 1 687  1273 56 44 17 9 25 1 1405  2641 44 40 23 11 40 2 2815  530 13 60 17 20 22 2 697  1209 13 49 22 30 40 3 1429  530 0 66 23 25 19 1 711  1213 0 57 27 35 31 2 1436  grnol Carbon lost % Carbon lost  60.5 16.0  158.4 18.7  168.3 10.7  319.1 10.2  148.5 17.6  144.8 9.2  134.2 15.9  137.6 8.7  Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) gmolimin  157  Table A.5 (continued) Magnesium Doped Samarium Oxide (1:100) Prepared in Oxygen Average Average Average Average Average Average Average Average Run 7 Run 8 Run 3 Run 4 Run 5 Run 6 Run 1 Run 2  STY timol/g/s C2's  2.5  5.4  8.4  11.6  12.4  17.4  15.9  20.9  % Conversion CH4 02  18.6 53.2  13.7 65.9  9.4 72.3  6.2 77.1  23.2 91.0  15.4 93.4  25.4 100.0  15.5 100.0  % Selectivity CO2 CO COX selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity  67.5 9.0 76.5 6.9 16.6 23.5 0.0 23.5  48.1 11.6 59.7 11.0 27.6 38.6 1.7 40.3  33.3 12.5 45.8 12.9 37.9 50.9 3.3 54.2  22.9 13.5 36.4 13.1 46.4 59.5 4.1 63.6  35.5 10.8 46.4 24.4 26.3 51.1 2.9 53.6  22.3 10.0 32.3 27.5 36.0 63.5 4.3 67.7  36.4 12.6 49.0 27.6 21.4 49.0 2.0 51.0  25.3 12.2 37.6 31.7 27.6 59.3 3.1 62.4  % Yield CO2 CO CO, yield C2H4 C2H6  12.5 1.7 14.2 1.3 3.1  6.6 1.6 8.2 1.5 3.8  3.1 1.2 4.3 1.2 3.6  1.4 0.8 2.3 0.8 2.9  8.2 2.5 10.7 5.7 6.1  3.4 1.5 5.0 4.2 5.5  9.2 3.2 12.4 7.0 5.4  3.9 1.9 5.8 4.9  C2 yield C3's C2+ yield  4.4 0.0 4.4  5.3 0.2 5.5  4.8 0.3 5.1  3.7 0.3 3.9  12.2 0.7 12.4  9.8 0.7 10.4  12.4 0.5 12.9  4.3 9.2 0.5 9.7  158  Table A.6 Sodium Doped Samarium Oxide (1:100) Prepared in Air Run 1 Temperature (°C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He CH4/02  Run 2  Run 4  Run 3  Run 5  Run 7  Run 6  Run 8  650  650  650  650  750  750  850  850  10 5 85  20 5 75  40 5 55  80 5 15  20 5 75  40 5 55  20 5 75  40 5 55  2  4  8  16  4  8  4  8  Average Average Average Average Average Average Average Average Run 7 Run 8 Run 5 Run 6 Run 2 Run 3 Run 4 Run 1 Products (%) CH4 02 CO2 CO C2H4 C2H6 C3s Total Carbon out Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) pmol/min Output (p.mol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out pinol Carbon lost % Carbon lost  7.55 3.39 0.63 0.21 0.02 0.08 0.00 8.60  16.12 2.80 0.84 0.34 0.06 0.25 0.01 17.95  33.79 2.40 0.97 0.50 0.12 0.53 0.02 36.61  72.49 2.02 1.05 0.80 0.18 0.94 0.04 76.68  14.15 0.78 2.07 0.40 0.56 0.65 0.05 19.20  31.78 0.59 1.87 0.50 0.82 1.15 0.09 38.37  13.84 0.06 2.38 0.51 0.66 0.53 0.03 19.21  31.86 0.03 2.10 0.64 0.93 0.86 0.06 38.37  29.18 25.57 101.26 4193  29.53 26.07 99.91 4137  29.78 26.57 98.91 4096  32.46 23.87 91.62 3794  30.50 27.57 96.25 3985  30.91 27.13 95.11 3938  30.63 27.37 95.91 3971  31.06 27.10 94.65 3919  317 142 27 9 1 3 0 361  667 116 35 14 3 10 0 742  1384 98 40 20 5 22 1 1500  2750 77 40 30 7 36 2 2909  564 31 82 16 22 26 2 765  1252 23 74 20 32 45 4 1511  550 3 94 20 26 21 1 763  1248 1 82 25 37 34 2 1504  18 5  103 12  74 5  226 9  80 8  62 4  82 8  70 4  159  Table A.6 (continued) Sodium Doped Samarium Oxide (1:100) Prepared in Air Average Average Average Average Average Average Average Average Run 1 Run 2 Run 7 Run 3 Run 4 Run 6 Run 8 Run 5 STY Lunol/g/s C2's  1.4  4.3  8.8  14.1  16.1  25.9  15.8  23.4  % Conversion CH4 02  12.2 29.8  10.2 44.0  7.7 52.0  5.5 59.7  26.3 84.4  17.2 88.3  27.9 98.7  17.0 99.4  % Selectivity CO2 CO CO, selectivity C2H4 C2H6 C2 selectivity Cis C2+ selectivity  60.3 20.0 80.3 3.8 15.9 19.7 0.0 19.7  45.8 18.6 64.5 6.9 27.0 33.9 1.6 35.5  34.5 17.8 52.3 8.3 37.3 45.6 2.1 47.7  25.0 19.1 44.0 8.4 44.7 53.1 2.9 56.0  40.9 8.0 48.9 22.2 25.9 48.1 3.0 51.1  28.4 7.6 36.0 25.0 34.9 59.9 4.1 64.0  44.3 9.6 53.9 24.6 19.9 44.5 1.7 46.1  32.3 9.9 42.2 28.7 26.4 55.1 2.8 57.8  7.4 2.4 9.8 0.5 1.9 2.4 0.0 2.4  4.7 1.9 6.6 0.7 2.7 3.5 0.2 3.6  2.7 1.4 4.0 0.6 2.9 3.5 0.2 3.7  1.4 1.0 2.4 0.5 2.4 2.9 0.2 3.1  10.8 2.1 12.9 5.8 6.8 12.6 0.8 13.4  4.9 1.3 6.2 4.3 6.0 10.3 0.7 11.0  12.4 2.7 15.0 6.9 5.6 12.4 0.5 12.9  5.5 1.7 7.2 4.9 4.5 9.3 0.5 9.8  % Yield CO2 CO COX yield C2H4 C2H6 C2 yield C3's C2+ yield  160  Table A.7 Sodium Doped Samarium Oxide (1:100) Prepared in Oxygen Run 1 Temperature ( ° C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He CH4/02  Run 2  Run 3  Run 4  Run 5  Run 6  Run 8  Run 7  650  650  650  650  750  750  850  850  10 5 85  20 5 75  40 5 55  80 5 15  20 5 75  40 5 55  20 5 75  40 5 55  2  4  8  16  4  8  4  8  Average Average Average Average Average Average Average Average Run 8 Run 4 Run 5 Run 6 Run 7 Run 1 Run 2 Run 3 Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out Output flows Time for 50m1 (s) Temperature ( ° C) mL/min (at 21.1°C) gmol/min Output (p.mol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out iimol Carbon lost % Carbon lost  7.59 3.97 0.80 0.19 0.02 0.09 0.00 8.81  16.20 3.43 0.93 0.28 0.06 0.23 0.01 18.00  34.09 2.87 1.12 0.50 0.11 0.50 0.02 37.01  73.03 2.39 1.26 0.54 0.19 0.96 0.04 77.23  14.30 0.65 1.80 0.36 0.60 0.68 0.05 19.18  32.51 0.43 1.54 0.47 0.88 1.19 0.10 38.95  14.21 0.01 1.85 0.54 0.69 0.56 0.03 19.17  32.68 0.01 1.62 0.68 1.01 0.90 0.06 38.97  29.42 23.60 101.12 4187  29.58 24.57 100.25 4151  30.11 25.13 98.30 4070  31.34 25.20 94.42 3910  30.84 24.07 96.31 3988  31.59 23.63 94.16 3899  31.22 23.33 95.38 3949  31.65 24.10 93.82 3885  318 166 34 8 1 4 0 369  672 142 39 12 2 9 0 747  1388 117 46 20 5 20 1 1506  2855 93 49 21 7 37 1 3020  570 26 72 14 24 27 2 765  1267 17 60 18 34 46 4 1518  561 0 73 21 27 22 1 757  1270 0 63 27 39 35 2 1514  10 3  98 12  67 4  115 5  80 8  55 3  88 9  60 4  161  Table A.7 (continued) Sodium Doped Samarium Oxide (1:100) Prepared in Oxygen Average Average Average Average Average Average Average Average Run 5 Run 6 Run 7 Run 8 Run 1 Run 2 Run 3 Run 4 STY gmol/g/s C2's  1.5  3.9  8.4  14.9  17.0  26.8  16.4  24.6  % Conversion CH4 02  13.8 17.9  10.0 31.5  7.9 42.5  5.4 52.3  25.4 87.0  16.5 91.4  25.9 99.8  16.1 99.8  % Selectivity CO2 CO COx selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity  66.1 15.7 81.8 3.9 14.3 18.2 0.0 18.2  51.6 15.8 67.4 6.3 25.2 31.5 1.1 32.6  38.5 17.1 55.6 7.8 34.5 42.3 2.1 44.4  30.0 12.8 42.8 9.0 45.6 54.6 2.6 57.2  36.9 7.4 44.2 24.6 27.9 52.5 3.3 55.8  23.9 7.3 31.2 27.2 36.9 64.1 4.7 68.8  37.3 10.8 48.1 27.7 22.4 50.1 1.8 51.9  25.7 10.9 36.6 32.0 28.5 60.6 2.9 63.4  % Yield CO2 CO CO X yield C2H4 C2H6 C2 yield C3's C2+ yield  9.1 2.2 11.3 0.5 2.0 2.5 0.0 2.5  5.2 1.6 6.8 0.6 2.5 3.1 0.1 3.3  3.0 1.4 4.4 0.6 2.7 3.3 0.2 3.5  1.6 0.7 2.3 0.5 2.5 3.0 0.1 3.1  9.4 1.9 11.3 6.3 7.1 13.3 0.8 14.2  4.0 1.2 5.2 4.5 6.1 10.6 0.8 11.4  9.7 2.8 12.5 7.2 5.8 13.0 0.5 13.4  4.1 1.8 5.9 5.2 4.6 9.8 0.5 10.2  162  Table A.8 Potassium Doped Samarium Oxide (1:100) Prepared in Oxygen Run 1 Temperature ( °C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He CH4/02  Run 2  Run 3  Run 4  Run 5  Run 6  Run 7  Run 8  650  650  650  650  750  750  850  850  10 5 85  20 5 75  40 5 55  80 5 15  20 5 75  40 5 55  20 5 75  40 5 55  2  4  8  16  4  8  4  8  Average Average Average Average Average Average Average Average Run 7 Run 1 Run 2 Run 4 Run 5 Run 6 Run 8 Run 3 Products (%) CH4 02 CO2 CO C2H4 C2H6 C3s Total Carbon out  7.79 4.04 0.94 0.20 0.02 0.09 0.00 9.16  16.66 3.23 1.21 0.34 0.07 0.25 0.01 18.87  34.95 2.47 1.42 0.49 0.15 0.58 0.03 38.39  75.52 1.91 1.51 0.59 0.27 1.09 0.06 80.51  14.44 0.63 2.01 0.42 0.54 0.62 0.05 19.35  32.95 0.53 1.77 0.55 0.80 1.08 0.09 39.30  14.35 0.04 2.10 0.54 0.66 0.53 0.03 19.45  32.96 0.03 1.91 0.69 0.97 0.87 0.06 39.43  29.55 24.60 100.32 4154  29.89 25.00 99.06 4102  30.34 25.20 97.51 4038  31.73 26.00 92.99 3850  30.57 26.00 96.52 3997  31.18 26.40 94.50 3913  30.67 26.87 95.94 3972  31.29 27.00 94.00 3892  Output (grnol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  324 168 39 8 1 4 0 380  683 133 50 14 3 10 0 774  1411 100 57 20 6 23 1 1550  2908 74 58 23 10 42 2 3100  577 25 80 17 22 25 2 773  1289 21 69 22 31 42 3 1538  570 1 83 21 26 21 1 772  1283 1 74 27 38 34 2 1535  innol Carbon lost % Carbon lost  -1.6 -0.4  71.0 8.4  23.6 1.5  34.5 1.5  71.9 7.3  35.7 2.3  72.6 7.3  38.8 2.5  Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) p.mol/min  163  Table A.8 (continued) Potassium Doped Samarium Oxide (1:100) Prepared in Oxygen Average Average Average Average Average Average Average Average Run 8 Run 6 Run 7 Run 4 Run 5 Run 2 Run 3 Run 1 STY gmol/g/s C2's  1.5  4.3  9.7  17.4  15.5  24.5  15.7  23.9  % Conversion CH4 02  14.9 17.0  11.7 35.3  8.9 50.5  6.2 61.8  25.4 87.4  16.1 89.5  26.2 99.3  16.4 99.4  % Selectivity CO2 CO CO, selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity  69.0 14.9 83.9 2.9 13.2 16.1 0.0 16.1  54.8 15.2 70.0 6.0 22.6 28.7 1.4 30.0  41.3 14.3 55.6 8.5 33.6 42.1 2.3 44.4  30.2 11.9 42.1 10.7 43.7 54.3 3.6 57.9  41.0 8.5 49.5 22.1 25.3 47.4 3.1 50.5  27.9 8.7 36.6 25.2 34.1 59.3 4.1 63.4  41.1 10.5 51.7 25.8 20.8 46.6 1.8 48.3  29.6 10.6 40.2 30.1 26.9 57.0 2.8 59.8  % Yield CO2 CO CO X yield C2H4 C2H6 C2 yield C3's C2+ yield  10.3 2.2 12.5 0.4 2.0 2.4 0.0 2.4  6.4 1.8 8.2 0.7 2.6 3.4 0.2 3.5  3.7 1.3 5.0 0.8 3.0 3.8 0.2 4.0  1.9 0.7 2.6 0.7 2.7 3.4 0.2 3.6  10.4 2.2 12.6 5.6 6.4 12.0 0.8 12.8  4.5 1.4 5.9 4.1 5.5 9.6 0.7 10.2  10.8 2.8 13.5 6.8 5.5 12.2 0.5 12.7  4.9 1.7 6.6 4.9 4.4 9.4 0.5 9.8  164  Table A.9 Calcium Doped Samarium Oxide (1:10) Prepared in Oxygen Run 1 Temperature (°C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He  Run 2  Run 3  Run 4  Run 5  Run 6  650  650  750  750  850  850  20 5 75  40 5 55  20 5 75  40 5 55  20 5 75  40 5 55  4  8  4  8  4  8  CH4/02 Average Run 1  Average Run 2  Average Run 3  Average Run 4  Average Run 5  Average Run 6  Products (%) CH4 02 CO2 CO C2H4 C2H6 Cg's Total Carbon out  16.73 2.07 0.90 0.30 0.06 0.24 0.01 18.54  35.06 1.77 1.16 0.50 0.12 0.54 0.02 38.11  14.41 0.55 1.68 0.39 0.60 0.69 0.05 19.22  32.76 0.35 1.48 0.49 0.91 1.22 0.10 39.27  14.14 0.01 1.84 0.50 0.78 0.57 0.04 19.31  32.75 0.00 1.62 0.61 1.11 0.90 0.07 39.21  Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) gmo1/min  30.20 21.73 99.14 4105  30.38 22.73 98.20 4066  30.74 23.00 96.98 4016  31.36 23.43 94.92 3930  30.69 23.47 96.96 4015  31.48 23.50 94.52 3914  687 85 37 12 2 10 0 761  1426 72 47 20 5 22 1 1549  579 22 68 16 24 28 2 772  1287 14 58 19 36 48 4 1544  568 0 74 20 31 23 2 775  1282 0 63 24 43 35 3 1535  84 10  24 2  73 9  30 2  70 8  39 2  Output (p.mol/min) CH4 02 CO2 CO C2H4 C2H6 Cg's Total Carbon out iimol Carbon lost % Carbon lost  165  Table A.9 (continued) Calcium Doped Samarium Oxide (1:10) Prepared in Oxygen Average Run 1 STY mol/g/s C2s  Average Run 2  Average Run 3  Average Run 4  Average Run 5  Average Run 6  4.0  8.9  17.4  27.8  18.1  26.2  % Conversion CH4 02  9.8 58.1  8.0 64.5  25.0 88.9  16.6 92.9  26.8 99.8  16.5 99.9  % Selectivity CO2 CO COX selectivity C2H4 C21-1 16 C2 selectivity C3s C2+ selectivity  49.4 16.5 65.9 6.3 26.1 32.4 1.7 34.1  38.1 16.5 54.6 8.1 35.3 43.4 2.0 45.4  35.0 8.0 43.0 25.1 28.8 53.9 3.1 57.0  22.7 7.5 30.2 27.8 37.4 65.2 4.6 69.8  35.7 9.6 45.3 30.2 22.2 52.4 2.3 54.7  25.1 9.5 34.6 34.2 28.0 62.2 3.2 65.4  4.8 1.6 6.5 0.6 2.6 3.2 0.2 3.3  3.0 1.3 4.4 0.6 2.8 3.5 0.2 3.6  8.8 2.0 10.8 6.3 7.2 13.5 0.8 14.3  3.8 1.2 5.0 4.6 6.2 10.8 0.8 11.6  9.5 2.6 12.1 8.1 5.9 14.0 0.6 14.6  4.1 1.6 5.7 5.6 4.6 10.3 0.5 10.8  % Yield CO2 CO CO X yield C2H4 C2H6 C2 yield Cg's C2+ yield  166  Table A.10 Sodium Doped Samarium Oxide (1:10) Prepared in Oxygen Run 1 Temperature (°C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He  Run 2  Output (gmol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out gmol Carbon lost % Carbon lost  Run 5  Run 6  650  750  750  850  850  20 5 75  40 5 55  20 5 75  40 5 55  20 5 75  40 5 55  4  8  4  8  4  8  Average Run 1  Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) gmol/min  Run 4  650  CH4/02  Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  Run 3  Average Run 2  Average Run 3  Average Run 4  Average Run 5  Average Run 6  17.15 4.29 0.52 0.08 0.01 0.11 0.00 17.98  35.23 3.82 0.47 0.07 0.03 0.24 0.00 36.31  15.09 2.15 1.17 0.10 0.40 0.63 0.03 18.50  32.60 1.07 1.31 0.16 0.72 1.16 0.08 38.08  14.17 0.00 2.02 0.40 0.63 0.53 0.03 19.01  32.36 0.00 1.68 0.64 0.90 0.81 0.06 38.28  28.76 26.00 102.61 4249  29.44 26.00 100.23 4150  29.82 26.17 98.90 4095  30.80 26.17 95.76 3965  30.33 26.17 97.25 4027  30.91 26.13 95.43 3952  729 182 22 3 0 5 0 764  1462 159 19 3 1 10 0 1507  618 88 48 4 16 26 1 758  1293 42 52 6 29 46 3 1510  571 0 81 16 26 21 1 765  1279 0 67 25 36 32 2 1513  81 10  66 4  87 10  64 4  80 9  61 4  167  Table A.10 (continued) Sodium Doped Samarium Oxide (1:10) Prepared in Oxygen Average Run 1 STY j.unol/g/s C2's % Conversion CH4 02 % Selectivity CO2 CO COx selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity % Yield CO2 CO COx yield C2H4 C2H6 C2 yield C3's C2+ yield  Average Run 2  Average Run 3  Average Run 4  Average Run 5  Average Run 6  1.7  3.8  14.0  24.8  15.6  22.6  4.6 10.0  3.0 23.6  18.4 56.9  14.4 78.6  25.4 100.0  15.5 100.0  62.4 9.0 71.4 2.5 26.2 28.6 0.0 28.6  42.1 6.9 48.9 5.6 45.5 51.1 0.0 51.1  34.4 2.8 37.2 23.4 36.7 60.2 2.6 62.8  23.9 3.0 26.9 26.3 42.4 68.7 4.4 73.1  41.8 8.3 50.1 26.2 21.8 48.0 1.9 49.9  28.4 10.9 39.3 30.5 27.3 57.8 2.9 60.7  2.9 0.4 3.3 0.1 1.2 1.3 0.0 1.3  1.3 0.2 1.5 0.2 1.3 1.5 0.0 1.5  6.3 0.5 6.9 4.3 6.8 11.1 0.5 11.6  3.4 0.4 3.9 3.8 6.1 9.9 0.6 10.5  10.6 2.1 12.8 6.7 5.5 12.2 0.5 12.7  4.4 1.7 6.1 4.7 4.2 9.0 0.4 9.4  168  Table A.11 Calcium Doped Samarium Oxide (1:10) Revised Preparation (RP) in Oxygen Run 1 Temperature (°C) Input flows (mL/min) Nominal (at 21.1°C) CH4 02 He  Run 2  Run 3  Run 4  Run 5  Run 6  650  650  750  750  850  850  20 5 75  40 5 55  20 5 75  40 5 55  20 5 75  40 5 55  4  8  4  8  4  8  CH4 /02 Average Run 1  Average Run 2  Average Run 3  Average Run 4  Average Run 5  Average Run 6  Products (%) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out  15.65 0.99 2.16 0.60 0.06 0.20 0.00 18.94  34.19 0.61 2.06 0.87 0.08 0.38 0.01 38.07  15.26 0.00 2.55 0.55 0.18 0.30 0.01 19.36  34.00 0.00 2.43 0.76 0.26 0.52 0.01 38.78  15.25 0.00 2.41 0.47 0.26 0.29 0.01 19.25  34.28 0.00 2.23 0.62 0.37 0.48 0.01 38.85  Output flows Time for 50m1 (s) Temperature (°C) mL/min (at 21.1°C) umol/rnin  30.02 24.93 98.66 4085  30.13 26.00 97.94 4055  30.22 26.37 97.52 4038  30.63 26.63 96.15 3981  30.32 26.63 97.13 4022  30.62 26.40 96.25 3985  650 41 90 25 2 8 0 786  1383 25 83 35 3 15 0 1540  614 0 103 22 7 12 0 779  1334 0 95 30 10 21 0 1522  609 0 96 19 10 12 0 768  1341 0 87 24 14 19 0 1521  59 7  34 2  66 8  52 3  77 9  53 3  Output (urnol/min) CH4 02 CO2 CO C2H4 C2H6 C3's Total Carbon out gmol Carbon lost % Carbon lost  169  Table A.11 (continued) Calcium Doped Samarium Oxide (1:10) Revised Preparation (RP) in Oxygen Average Run 1  Average Run 2  Average Run 3  Average Run 4  Average Run 5  Average Run 6  STY wriol/g/s C2's  3.6  6.2  6.5  10.2  7.3  11.0  % Conversion CH4 02  17.4 79.8  10.2 87.9  21.2 100.0  12.3 100.0  20.8 100.0  11.8 100.0  % Selectivity CO2 CO CO,, selectivity C2H4 C2H6 C2 selectivity C3's C2+ selectivity  65.7 18.2 83.9 3.4 12.4 15.8 0.3 16.1  53.2 22.3 75.5 4.3 19.4 23.7 0.8 24.5  62.3 13.4 75.7 8.9 14.6 23.6 0.7 24.3  50.8 15.8 66.6 10.9 21.9 32.8 0.6 33.4  60.1 11.8 72.0 12.8 14.5 27.3 0.7 28.0  48.8 13.5 62.3 16.2 20.8 37.0 0.7 37.7  % Yield CO2 CO COx yield C2H4 C21-16 C2 yield C3's C2+ yield  11.4 3.2 14.6 0.6 2.1 2.7 0.1 2.8  5.4 2.3 7.7 0.4 2.0 2.4 0.1 2.5  13.2 2.8 16.0 1.9 3.1 5.0 0.2 5.1  6.3 2.0 8.2 1.3 2.7 4.0 0.1 4.1  12.5 2.5 15.0 2.7 3.0 5.7 0.2 5.8  5.7 1.6 7.3 1.9 2.5 4.4 0.1 4.4  170  APPENDIX B: SCANNING ELECTRON MICROSCOPY (SEM) PHOTOGRAPHS  171  Figure B.1 1:100 Ca:Sm Photograph 192  Figure B.2 1:10 Ca:Sm Photograph 193  172  Figure B.3 1:10 Ca:Sm Photograph 194  Figure B.4 1:10 Ca:Sm (RP) Photograph 105 ----..... e^—^do^w c ,  .1,4 •  •^CU 1.91•4  1 °^4^  t* CU  Nwso.W,^It*  rt. troll^"""  i^41  ." •  116.01111ime  173  Figure B.5 1:100 Na:Sm Photograph 195  Figure B.6 Sm2 03 Photograph 102  • •  • ICIU • ••■41 •  .114 •a ,  a  •C • C  eg)  Cu  Cu 4.-11  in Cu  174  Figure B.7 1:100 K:Sm Photograph 103  Figure B.8 1:100 Mg:Sm Photograph 104  175  APPENDIX C: X-RAY DIFFRACTION (XRD) GRAPHS  176  2-Theta - Scale 0 0  1  UNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:56  5^10^15^20^25^  30  35  40  45  SO  SS  2-Theta - Scale  UNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:39  0  -4-)  c  O  0  cs)  -^•^r^. 30^35  i .^1^.^•^•^1^. 40^45^50^55  2-Theta - Scale  10  UNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:46  I^'^'^I  30^35  •  40^45  •  50^SS  •  2-Theta - Scale  UNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:43  N  a 0  S^10^15^20^25^30^3S^  40^4I5  2-Theta  -  Scale^  UNIVERSITY OF BRITISH COLUMBIA 16-Mar-1993 12:40  0 "  U)  U) N  CO  fD  01 fa  N  c)  JJ ri  0 If nzr  )  0  z  0  z  01  s)  Dj  Ui 0 0  i  .1.4:11:41is j 'a l g'^.ig& ..l 1 li^,.44, , I..: i.^,„; .6. Ir.  10  Li  III i i..., .^'..A ;...i. , Al, ; ;,..41.!^4 i^ '^o ..,^ ^44  15^20^25^30^35^40^45  'lit ii l '4 % ■ 3  ILI i  SO  SS  

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059294/manifest

Comment

Related Items