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 SAMARIUMOXIDE CATALYSTS FOR THE OXIDATIVE COUPLING OF METHANEbySHANNA DENINE KNIGHTSB.A.Sc., The University of British Columbia, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF APPLIED SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Chemical Engineering)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1993©Shanna Denine Knights, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of Chemical EngineeringThe University of British ColumbiaVancouver, CanadaDate^zcc iqq3DE-6 (2/88)ABSTRACTThe catalytic oxidative coupling of methane involves the reaction of methane andoxygen at high temperatures (650°C to 900°C) in the presence of a solid metal oxidecatalyst to produce the desirable products, ethane and ethylene, as well as theundesirable products, carbon monoxide and carbon dioxide. Although thehomogeneous reactions are well understood, the heterogeneous reactions and theireffect on the homogeneous reactions are still the subject of much research anddiscussion. The effect of the catalyst characteristics on the heterogeneous reactions isalso an active area of research. The objective of this thesis was to characterize a seriesof catalysts, and to determine the effect of various catalyst properties on the oxidativecoupling reactions.The experimental part of this thesis consisted of preparing and testing samarium oxideand alkali (Na and K) and alkaline earth (Mg and Ca) doped samarium oxide catalystsfor the oxidative coupling of methane. The effects of the specific dopant used, varyingdopant concentration (1:100 and 1:10 dopant:Sm mole ratio), and catalyst preparationwere evaluated. The catalysts were tested in a bench scale packed bed reactor underconditions of varying temperature (650°C, 750°C, and 850°C) and methane to oxygenmole 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 catalystiiperformance. No new phases were observed in the Sm 203 crystal upon addition of thedopant cations, indicating that the cations were dispersed throughout the crystal,although probably not uniformly. The dopant concentration affected the catalystperformance; for example, at 750°C, the C2+ yield increased from 13.1% to 14.3% whenthe Ca:Sm mole ratio was increased from 1:100 to 10:100. A change in the catalystpreparation procedure resulted in an increase in the crystal dimensions and animproved combustion catalyst (e.g., the methane conversion increased from 9.8% for thestandard catalyst to 17.4% for the revised catalyst) with, however, a decrease in the C2yield from 3.2% to 2.7%. The results of this study indicated that, over the range ofsurface areas tested (2.0 to 3.1 m2/g), surface area did not have a significant effect onthe catalyst performance. The basicity of the catalyst appears to have a significanteffect on the catalyst performance, with an increase in basicity resulting in an increasein C2+ selectivity (from 54.3% for the least basic catalyst, undoped samarium oxide, to62.8% for the most basic catalyst, 1:10 mole ratio Na:Sm oxide). The catalysts displayedtemperature dependent behaviour, and there existed an optimum temperature formaximum C2+ yield, which is dependent on the amount and nature of the dopant, andis likely associated with the formation of carbonates on the catalyst surface. The ionicradius of the cation dopant must be similar to or smaller than the support cation toachieve effective inclusion in the crystal lattice.iiiTABLE OF CONTENTSABSTRACT ^  iiTABLE OF CONTENTS^  ivLIST OF TABLES  viiiLIST OF FIGURES^  xNOMENCLATURE  xiiiACKNOWLEDGEMENTS ^ xiv1. INTRODUCTION ^ 12. LITERATURE REVIEW^ 42.1 Choice of Doped Samarium Oxide Catalyst ^ 82.2 Mechanism of Oxidative Coupling of Methane 92.2.1^Formation of Methyl Radicals ^ 102.2.2^Formation of Ethane ^ 102.2.3^Formation of Non-Selective Products (CO X) and the InherentC2+ Yield Limit^ 122.2.4^Gas Phase Reactions 182.2.5^Oxidative Coupling of Methane Under Conditions of HighPressure ^ 202.3 The Role of Gas Phase, Surface, and Lattice Oxygen^ 222.3.1^Source of the Active Oxygen Species ^ 222.3.2^Nature of the Active Oxygen Species 25iv2.3.3^Crystal Lattice Oxygen Mobility ^ 312.3.4^Concentration of Active Sites on the Catalyst Surface ^ 342.4 Samarium Oxide Crystal Structure ^ 352.4.1^Comparison of Cubic versus Monoclinic Sm 2O3 ^ 372.5 Effect of Water Addition on the Oxidative Coupling of Methane 382.6 Potential Reactor Configurations ^ 402.7 Engineering and Economic Assessments ^ 423. METHODS AND MATERIALS ^ 453.1 Experimental Equipment 453.2 Data Acquisition ^ 493.3 Gas Chromatograph 493.3.1^GC Calibration ^ 503.4 Flow Measurement and Control ^ 513.5 Definitions ^ 513.6 Catalyst Preparation Procedure^ 523.7 Experimental Procedure 533.7.1^Variation of Catalyst Dopant Concentration ^ 553.7.2^Catalyst Preparation Modification^ 554. RESULTS AND DISCUSSION ^ 574.1 Undoped Sm2O3: Conditions of Complete Oxygen Conversion . . . . 574.1.1^Effect of Gas Phase Flow Rate ^ 614.1.2^Carbon Deposition in the Reactor 61v4.1.3 Comparison of Results for Sm203 to Other Researchers'Data ^  624.1.4 Space Time Yield ^  654.2 Undoped Sm203: Conditions of Incomplete Oxygen Conversion^654.2.1 Effect of Methane to Oxygen Ratio ^  674.2.2 Effect of Reaction Temperature  704.2.3 Carbon Balance over the Reactor ^  724.2.4 Variation in Results ^  754.3 Alkali and Alkaline Earth Doped Samarium Oxide Catalysts ^ 774.3.1 Results of Catalyst Performance Tests^ 784.4 Modification of Catalyst Preparation Procedure  844.4.1 Results of Performance Test for Modified Catalyst ^ 864.5 Scanning Electron Microscopy-Electron Dispersive X-Ray^ 904.6 Powder X-Ray Diffraction Analysis (XRD) ^ 944.7 Measurement and Effect of Catalyst Surface Area^ 964.8^Basicity of the Catalyst ^  1034.9 Effect of Carbon Dioxide and Carbonate Formation on CatalystPerformance ^  1154.9.1 Determination of Carbonate Formation for Sodium andCalcium Doped Samarium Oxide Catalysts ^ 1214.9.2 Effect of Gas Phase Composition on Carbonate Formation ^ 1304.10 Ionic Radius of Dopant ^  1325.^CONCLUSIONS ^  137vi6.^REFERENCES ^ 139APPENDIX A: RESULTS OF REACTOR CATALYST TESTS ^ 150APPENDIX B:^SCANNING ELECTRON MICROSCOPY (SEM)PHOTOGRAPHS ^ 171APPENDIX C: X-RAY DIFFRACTION (XRD) GRAPHS ^ 176viiLIST OF TABLESTable 2.1Table 3.1Table 3.2Table 4.1Table 4.2Table 4.3Table 4.4Table 4.5Table 4.6Table 4.7Table 4.8Table 4.9Table 4.10Table 4.11Table 4.12Table 4.13Table 4.14Table 4.15Comparison of Catalyst Activity and Oxygen Exchange Parametersfor Cubic and Monoclinic Sm203^  24Conditions Used for Comparison Testing of Doped Catalysts ^ 56Conditions Used for Comparison Testing of Catalysts at HigherDopant Concentrations and Different Preparation Times ^ 56Samarium Oxide: Conditions of Complete Oxygen Conversion . 58Effect of Total Flow Rate ^  61Samarium Oxide Results for Various Researchers ^ 63Average Percent Standard Deviation in GC Calibration ^ 76Variation in Results for Two Identical Runs ^ 77Doped Samarium Oxide Catalysts ^  78Results for all Catalysts, CH4/02 = 4  79Melting and Decomposition Temperatures of Nitrates used inCatalyst Preparation ^  85Weight % of Catalyst Surface Components According to SEM-EDX ^  92Surface Area of Catalysts ^  101Selectivity and Acid-Base Properties of Various Catalytic Oxides ^ 104Electronegativity of Alkali and Alkaline Earth Elements ^ 110Effect of Electronegativity on Catalyst Performance ^ 112Decomposition Temperature of Various Carbonates ^ 121Estimated Ionic Radii of Selected Elements ^ 133viiiTable 4.16Table A.1Table A.2Table A.3Table A.4Table A.5Table A.6Table A.7Table A.8Table A.9Table A.10Table A.11Effect of Cation Radius/Charge Ratio on C2+ Selectivity ^ 135Samarium Oxide Prepared in Oxygen^ 151Calcium Doped Samarium Oxide (1:100) Prepared in Air ^ 153Calcium Doped Samarium Oxide (1:100) Prepared in Oxygen   154Magnesium Doped Samarium Oxide (1:100) Prepared in Air ^ 156Magnesium Doped Samarium Oxide (1:100) Prepared in Oxygen ^ 157Sodium Doped Samarium Oxide (1:100) Prepared in Air ^ 159Sodium Doped Samarium Oxide (1:100) Prepared in Oxygen ^ 161Potassium Doped Samarium Oxide (1:100) Prepared in Oxygen . . ^ 163Calcium Doped Samarium Oxide (1:10) Prepared in Oxygen ^ 165Sodium Doped Samarium Oxide (1:10) Prepared in Oxygen ^ 167Calcium Doped Samarium Oxide (1:10) Revised Preparation inOxygen ^  169ixLIST OF FIGURESFigure 2.1Figure 3.1Figure 3.2Figure 4.1Figure 4.2Figure 4.3Figure 4.4Figure 4.5Figure 4.6Figure 4.7Figure 4.8Figure 4.9Figure 4.10Figure 4.11Figure 4.12Figure 4.13Figure 4.14Figure 4.15Figure 4.16Figure 4.17Figure 4.18The Crystal Structure of Cubic Samarium Oxide Sm 203 ^ 36Equipment Flow Diagram^  46Reactor Detail ^  48Methane Conversion for Samarium Oxide ^ 59C2 Selectivity for Samarium Oxide  59C2 Yield for Samarium Oxide ^  60Methane Conversion for Samarium Oxide, All Researchers ^ 64C2 Selectivity for Samarium Oxide, All Researchers ^ 64C2 Yield for Samarium Oxide, All Researchers  66STY for Various Catalysts, T=750°C ^  66Samarium Oxide Results ^  68Conversion for Samarium Oxide  68Selectivity for Samarium Oxide^  69% Yield for Samarium Oxide  69Product Carbon Output as a Function of Total Carbon ^ 71Output of Total Oxidation Products as a Function of OxygenConversion ^  71Conversion as a Function of Temperature for Samarium Oxide . . ^ 73Selectivity as a Function of Temperature for Samarium Oxide^73Yield as a Function of Temperature for Samarium Oxide ^ 74Methane Conversion for CH 4 /02 = 4 ^  80C2+ Selectivity at CH4 /02 = 4  80Figure 4.19 C2 Yield for CH4 /02 = 4 ^  81Figure 4.20 Effect of Catalyst Preparation on % Oxygen Conversion ^ 87Figure 4.21 Effect of Catalyst Preparation on % Methane Conversion ^ 87Figure 4.22 Effect of Catalyst Preparation on % C2 Selectivity ^ 89Figure 4.23 Effect of Catalyst Preparation on % C2 Yield  89Figure 4.24 C2+ Selectivity as a Function of Surface Area ^ 102Figure 4.25 Methane Conversion as a Function of Cation Electronegativity.... 113Figure 4.26 C2+ Selectivity as a Function of Cation Electronegativity   113Figure 4.27 Carbonate Formation on Samarium Oxide under a 100% CO 2Atmosphere ^  122Figure 4.28 Carbonate Formation on a 1:10 Na:Sm Oxide Catalyst under a100% CO2 Atmosphere ^  122Figure 4.29 Carbonate Formation on 100% Calcium Oxide under a 100% CO2Atmosphere ^  124Figure 4.30 Carbonate Formation on a 1:10 Ca:Sm Oxide Catalyst under a100% CO2 Atmosphere ^  124Figure 4.31 Catalyst Performance as a Function of Carbonate Formation(T=650°C) ^  127Figure 4.32 Catalyst Performance as a Function of Carbonate Formation(T=750°C) ^  127Figure 4.33 Catalyst Performance as a Function of Carbonate Formation(T=850°C) ^  128Figure 4.34 Ratio of CO to CO2 Produced ^  128Figure 4.35 CO, Yield ^  136xiFigure B.1Figure B.2Figure B.3Figure B.4Figure B.5Figure B.6Figure B.7Figure B.8Figure C.1Figure C.2Figure C.3Figure C.4Figure C.51:100 Ca:Sm Photograph 192 ^  1721:10 Ca:Sm Photograph 193  1721:10 Ca:Sm Photograph 194^  1731:10 Ca:Sm (RP) Photograph 105  1731:100 Na:Sm Photograph 195 ^  174Sm203 Photograph 102  1741:100 K:Sm Photograph 103^  1751:100 Mg:Sm Photograph 104  175XRD Graph for Cubic Sm 203^ 177XRD Graph for Ca:Sm (1:10)  178XRD Graph for Na:Sm (1:10) ^  179XRD Graph for Ca:Sm (1:10) (RP)  180XRD Graph for Monoclinic Sm 203 ^  181xiiNOMENCLATUREC2^ ethane and ethyleneC3 propane and propyleneC2+^ C2'S and C3sCO carbon dioxide and carbon monoxideC2 selectivity^% of reacted methane converted to C2'sC2 yield % of methane in feed converted to C2sC exiting reactor^all carbon species exiting the reactor, as moles carbon (CO,CO2, CH4, C2H4, C2H6, and C3)C in products^all carbon containing products, as moles carbon (CO, CO2,C2H4, C2H6, and C3)CH4/02^methane to oxygen mole or flow rate ratioAG Gibbs free energy of reactionOH^heat of reactionmethane conversion^% of methane in the feed reactedoxygen conversion^% of oxygen in the feed reactedP*^ fraction, based on pressure or flow, of reacting gases in feed(i.e., (Pai4 -FP02)/Ptotal)QSTY (space time yield)TreactionW/Fflow rate (mL/min)the amount of C2's produced per unit time per unit catalystweight (p.mol/s/g)reaction temperatureweight of catalyst divided by feed flowrate (g s/mL)ACKNOWLEDGEMENTSI am grateful to the various people and institutions who have provided me with thesupport and assistance necessary to allow me to complete this thesis. There are severalpeople in particular whom I wish to acknowledge.There are numerous people in the Chemical Engineering Department of U.B.C. whoprovided invaluable assistance throughout the project. In particular, my thesissupervisor, Dr. Clive Brereton, and Dr. Kevin Smith, both provided essential assistanceand excellent advice. All the technicians and staff at Chemical Engineering, whoprovided their work and assistance, are also much appreciated.The support of several B.C. Research personnel was essential to the completion of thisthesis. In particular, I would like to thank Mr. Klaus Oehr, who was always generouswith his enthusiasm and ideas, and Mrs. Erin Skelton, who prepared the catalysts andwas always very helpful and accommodating to my schedule.In addition to providing valuable assistance on the two Autocad drawings, Mr. DonaldLivingstone provided his personal support and sometimes much neededencouragement throughout my work on this thesis, for which I am particularly grateful.I also gratefully acknowledge the financial support of British Columbia ResearchCorporation and CANMET, Energy Mines and Resources, without which this workwould not have been possible.xiv1.^INTRODUCTIONA process to convert methane directly to ethane and ethylene is desirable in order toavoid the energy intensive steam reforming step currently required for the conversionof methane to more valuable products. The presence of oxygen is required to make thedirect conversion of methane to ethane and ethylene thermodynamically favourable.This process, referred to as the oxidative coupling of methane, has receivedconsiderable attention in recent years as the most promising approach to add value tonatural gas supplies. Currently there are no commercial oxidative coupling reactors.In developmental work, the process is normally carried out in a packed bed bench scalereactor over a metal oxide catalyst, at temperatures of 700°C to 900°C and atmosphericpressure. The progress toward development of a commercial process has beeninhibited by the excess production of the non-desirable products, carbon dioxide andcarbon monoxide, collectively referred to as CO.. Much work has been carried out onmechanism elucidation and catalyst characterization in order to aid in theunderstanding of the processes which occur, and to maximize yields and selectivitiestoward 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 theeffects have been ascribed to various properties of the dopant and bulk catalyst, thereis still much work required on catalyst characterization and the effect on the catalystmechanism. This thesis concentrated on measurements of the effects of alkali andalkaline earth dopants on the properties of a samarium oxide catalyst, and on theevaluation of the catalyst properties on the oxidative reaction. Samarium oxide has1been shown to be a promising catalyst for this reaction.Several samarium oxide catalysts, doped with alkali, Na and K, and alkaline earth, Mgand 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 weretested in a laboratory scale packed bed reactor at conditions of 0.09 seconds residencetime 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 surfacecomposition and structure using scanning electron microscopy with electron dispersivex-ray analysis; their bulk crystal type by powder x-ray diffraction analysis; and theirability to form carbonates. The performance of the catalysts was determined based onmethane conversion, selectivity to C2 hydrocarbons, and C2 yield. These werecorrelated with the catalyst surface area, catalyst basicity, carbonate formationtendency, and the ionic radius of the dopant.1.1^Thesis ObjectivesThe objectives of this thesis were to accomplish the following:i)^Design and build a packed bed reactor to minimize pre- and post-catalyticvolume, in order to minimize the effect of homogeneous reactions and maximizethe sensitivity to catalyst properties.2ii) Test samarium oxide and samarium oxide doped catalysts for the oxidativecoupling of methane and compare the effects of various dopants and dopantconcentrations.iii) Examine the effects of changing the catalyst preparation procedure.iv)^Characterize the catalysts and determine the effect of various catalyst propertieson the catalyst performance, as indicated by methane conversion, C2 selectivity,and C2 yield.32.^LITERATURE REVIEWMethane, the major component of natural gas, is an abundant fossil fuel. With presenttechnology, much of this methane is wasted due to costs associated with transportationand processing. Much of the natural gas is located in remote areas where the costs fortransportation to market far exceed its value. However, the reserves of liquidpetroleum are diminishing, and the search for alternate sources is increasing.Of the commercially-established technologies presently available to convert methaneto methanol and higher hydrocarbons, the most widely used process is the steamreforming process, which involves the initial formation of synthesis gases (H 2 + CO +CO2):CH4 + H2O ' 3 H2 + CO AH/ (870°C) = 226 kJ/mol^AGf = -63 kJ/mol (2.1)CO + H2O 'rtk H2 + CO2 AH1 (870°C) = -34 kJ/mol^AGf = 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 methanolor 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 byFischer-Tropsch (FT) synthesis (a heterogeneous reaction of H2 and CO catalyzed4by various metals); andcatalytic conversion of methanol to gasoline-type hydrocarbons as well asethylene 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 reformingtransformation to syngas, which involves the initial cleavage of four C-H bonds permolecule followed by the formation of a C-0 bond. Some of the hydrogens are thenrestored by further processing, and the oxygen may be removed. A direct methaneconversion process which avoids the steam reforming step should be more energyefficient, and therefore is potentially more economically attractive. Capital costs mayalso be reduced by a reduction in the number of process steps.The oxidative coupling of methane is a direct conversion process carried out in thepresence of oxygen, resulting in the desired products, ethane and ethylene, with smallamounts of higher hydrocarbons. Unfortunately, the process also produces substantialamounts of the undesirable products, carbon monoxide and carbon dioxide.Ethylene is currently produced mainly from the ethane and propane fractions of naturalgas, which are recovered from natural gas using cryogenic distillation and otherseparation techniques. Ethylene is converted to numerous intermediate and endproducts on a large scale, mainly polymeric materials such as plastics, resins, fibres,and elastomers. Other important products are solvents, surfactants, coatings,plasticizers, and antifreeze.5The basic equations for conversion of methane to ethane and ethylene are:2 CH4 —> C2H6 + H2^ AGanc) = + 71 kJ/mol^(2.3)2 CH4 -4 C2H4 + 2 H2 AG(727c) = + 80 kJ/mol^(2.4)However, these reactions are thermodynamically unfavourable, and the presence ofoxygen is required if the reaction is to be carried out at temperatures significantlybelow 1300°C. When oxygen is added the free energies become strongly negative:2 CH4 + 0.5 02 —) C2H6 ± H20^AG(7270 = - 121 kJ/mol^(2.5)2 CH4 + 02 —) C2H4 4' 2 H20^AG(727c) - 305 kJ/mol^(2.6)However, methane, ethane, and ethylene can also be oxidized non-selectively to givecarbon oxides. In order to achieve significant conversion to ethane and ethylene, thereaction must be carried out at temperatures in excess of 600°C and either in thepresence of a catalyst or under conditions of high pressure. The elevated temperaturealso encourages formation of CO and CO2. The maximum C2 yields achievable thusfar have been on the order of 30%. The competition between the selective and non-selective reactions, and between heterogeneous and homogeneous reactions, has beenextensively studied.There are two main process methods which have been studied for the oxidativecoupling of methane, the cofeed and redox modes. The redox mode involves thereaction of methane with a reducible metal oxide, which donates the oxygen for the6oxidative coupling reaction (Kuo, 1991):2 CH4 + 2/y MO„,y C2H4 + 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 02 vat MO,r+),^ (2.9)The redox method has been used less than the cofeed method and will not beconsidered 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 enhancesthe coupling reaction.Intensive work has been carried out by many researchers world wide in recent yearson the development of the oxidative coupling of methane process. The present workinvolves evaluation of the catalyst, samarium oxide, promoted with alkali and alkalineearth oxides.72.1 Choice of Doped Samarium Oxide Catalysts for StudySamarium oxide (Sm203) and alkali and alkaline earth doped samarium oxide catalystswere chosen for this study because they have been found to be among the mostpromising active and selective catalysts for the oxidative coupling of methane (Otsukaet 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 andGupta, 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 theproduction of SmCo5 permanent magnets.The catalysis mechanism for the activation of methane is not completely clear. Someform of oxygen species is generally accepted to be the active species but the exactnature 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 ofthe samarium oxide catalyst; however, the important properties of the dopants requirefurther elucidation. The catalytically active sites may be affected by the bulk crystalstructure, concentration of surface and bulk oxygen vacancies, the nature of the surfaceoxygen species, the distribution of dopants within the support crystal, the surfacebasicity of the catalyst, the presence and/or state of equilibrium of carbonate formation,and the ionic radius of the dopant.82.2 Mechanism of Oxidative Coupling of MethaneFor the purpose of designing catalysts and/or reactors with improved yield andselectivity, the mechanism of the reaction has been extensively researched. Althoughmost aspects of the mechanism have been determined, there are still some questionsto be answered. Complications stem from difficulties in comparing results fromvarious researchers. Not only are there many different catalysts being used, but thereaction conditions and reactor designs vary widely.It is generally agreed that the reaction proceeds by hydrogen abstraction from themethane molecule on the catalyst surface, to produce a methyl radical (CH 3-) whichthen likely combines with another methyl radical to produce ethane (C2H6), either onthe catalyst surface, or in the gas phase. The ethylene (C 2H4) may then be producedby dehydrogenation of the ethane, again either on the catalyst surface or in the gasphase. 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 theheterogeneous versus homogeneous reactions to produce CO 2 and CO is a subject ofcontinuing research.The homogeneous gas phase reactions play a very significant role in the process. Thiscomplicates comparison of catalyst performance, because under conditions where thegas phase reactions strongly dominate, the inherent selectivity of the catalyst may bemasked. Under such conditions, the product distributions can be very similar overcatalysts of very different chemical composition.92.2.1 Formation of Methyl RadicalsThe high dissociation energy of the C-H bond (427 kJ/mole) indicates that theabstraction of hydrogen to form the methyl radical is the rate-determining step in theoxidative coupling of methane (Otsuka and Jinno, 1986b). This was confirmed overSm203 in further studies (Otsuka and Nakajima, 1987), as well as by others overLi/MgO (Cant et a/., 1988), Li/MgO and Sm203 (Amorebieta and Colussi, 1988, 1989),reducible Na/Mn0x/Si02 (Burch et al., 1990), and Li/MgO, SrCO3, and Sm203 (Nelsonet al., 1989).2.2.2 Formation of EthaneMartin and Mirodatos (1987; Mirodatos and Martin, 1988) reported on evidence ofcarbene (:CH2) intermediates formed over Li/MgO. The formation of carbene mayoccur by the abstraction of a proton from methane to form CH3, followed by theabstraction of a hydride ion by Li+ to form carbene. Carbene in the gas phase couldinsert into a C-H bond of methane to give ethane. An alternative mechanism is basedon the catalytic activity of surface peroxide ions, 0 22-. These may be capable ofabstracting two H atoms from CH4 to produce carbene. Non-selective oxidation intocarbon oxides would proceed on another type of site.Both of these mechanisms may be disputed based on isotopic studies which werecarried out by Otsuka et al. (1989) and Nelson et al. (1989). These indicated that ethaneis formed through the coupling of methyl intermediates, either from methyl radicals10in the gas phase or methyl groups adsorbed on the surface. Using a feed of a mixtureof CH4 and CD4 resulted in ethane only of the forms C 2H6, CD3CH3, and C2D6. Thisrules 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. Theydetermined that C3 formation occurs by the terminal addition of methyl radicals toethylene. Lunsford (1989) determined that the gas phase coupling of surface-generatedmethyl radicals was a significant pathway for the formation of ethane. They found thatthe 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'sability to both generate and consume methyl radicals It was estimated that a methylradical will collide with a surface ca. 105 times before it reacts with another CH3 -radical. Tong et al. (1989) and Tong and Lunsford (1991) studied the reactions ofmethyl radicals with lanthanide oxides. The methyl radicals reacted extensively withCeO2, Pr60n , and Tb407, all of which have multiple cationic oxidation states. Theoxides La203, Nd203, Sm203, Eu203, and Yb203 react with CH3- radicals to only a smallextent. The former oxides are non-selective for the oxidative coupling of methanereaction. However, the reaction of CeO2 with CH3- radicals is strongly inhibited by theaddition of Na2CO3 and consequently radical production is enhanced.Feng et al. (1991a, 1991b, 1992) conducted kinetic studies on 1% Sr/La 203. The resultsindicated heterogeneous production of CH3- and homogeneous loss of CH 3- byrecombination, to C 2H6. The small amount of C2H4 formed can largely be accountedfor by known gas phase processes. No indication was found of any heterogeneous11oxidation of methyl radicals to CO ), or of any heterogeneous conversion of methane toC2 compounds.2.2.3 Formation of Non-Selective Products (CO 3) and the Inherent C2+ Yield LimitThe addition of ethane and ethylene to the gas feed of oxidative coupling reactors haslittle effect on the rate of CO), production, even in the presence of excess oxygen. Thisinitially led investigators to believe that ethane and ethylene were stable under thereaction conditions and did not contribute to CO ), formation. However, the secondaryoxidation of the reaction products was found to contribute significantly to theformation 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 thepartial oxidation was derived from the products, particularly ethylene. The portion ofCO„ formed from the 13C2H4 was 85% when this was present at 8 volume % in the feedgas. This clearly indicates that, as the concentrations of C2 in the reactor increase, theCO„ is increasingly derived from the C2 products.Since the addition of potential reductants in the production of CO„ (ethane andethylene) does not result in a significant effect on the rate of CO ), production, it can beconcluded that the formation of the CO ), is limited by the availability of the oxidant andnot of the reductant. Therefore, the reaction is not simply a gas phase oxidationinvolving molecular oxygen, since this is present in excess, but must involve a speciesformed by the catalyst, which reacts either on the surface or in the gas phase.12A typical feature of the catalytic oxidative coupling of methane is that the maximumC2+ concentration obtainable is similar for many catalysts of very different nature.Attempts to increase the C2+ selectivity by changing reaction conditions invariablyresult in lower methane conversion. The limit in the maximum C2+ concentration maybe similar for so many catalysts because it is ultimately determined by the kinetics ofthe activation of the similar molecules, CH 4, C2H6, and C2H4, by the various catalysts.The reaction rates do not show significant differences for these molecules (Ekstrom etal., 1989b). The c2+ yield of the catalyzed oxidative coupling of methane reaction islimited when the rate of C2+ formation from CH4 is equal to its rate of conversion toCO2. 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 oxygenconcentration.Hutchings et al. (1989a, 1989c) tested several catalysts, including Sm203 and Li/MgO,to study the oxidative coupling of methane. Over Sm203 at 700°C the CO„ was formedlargely via the C2 products and C2H4 is oxidized somewhat faster than C 2H6 to CO„.Over Li/MgO, at TReadion < 700°C, the oxidation of C2 products accounted for less than10% of the CO„, but at TReact ion > 740°C, C2 oxidation was responsible for the formationof 30-80% of the CO„.The concept of an inherent limit to the C2+ yield of the oxidative coupling of methanereaction, regardless of catalyst activity and selectivity, has been studied by severalresearchers, both by experimental studies and modelling studies.13Geerts et al. (1989) carried out experimental reactivity studies and found that theconversion 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 werereached by Roos et al. (1989a, 1989b), who found that neither ethane nor ethylene arestable under reaction conditions over Li/MgO and Ca/Sm203. The CO. are formedpredominantly from ethylene and the reactions can be represented by a sequentialreaction, CH4 --> C2H6 ---> C2H4 --> CO,„Nelson et a/. (1989) conducted isotope studies over Li/MgO, SrCO 3, and Sm203 anddevised a mechanism for the heterogeneous formation of CO. from methyl radicals, asa result of surface reactions of methyl peroxy radicals (CH 302 ):C H3(g) . + 02(0 leah CH302(g) . (2.10)CH302(6) - + catalyst —> CH30(ads) + 0 (2.11)CH3O(ads) -f CH2O (in presence of 02) --) CO, + H2O (2.12)The reaction scheme would explain why selectivity declines with increasing 0 2 pressureand decreasing temperature (longer CH 304) - lifetime). However, direct reaction ofCH3 - with the surface is another possible route to carbon oxides. A purelyhomogeneous chain reaction involving CH 302 • is not indicated.Further work supported the methyl peroxy radical as the source of low temperatureCO. from CH4, and again indicates that reactions are partly heterogeneous (Nelson et14al., 1991). On the other hand, the thermal stability of the ethyl peroxy radical, C2HSO2%is considerably less than that of CH302-, and, at temperatures greater than 327°C,reaction (2.13) is the dominant pathway, resulting in a homogeneous path with highselectivity for the conversion of C 2H6 to C2H4.C2145 . + 02 -) C2H4 + 1102 4^ (2.13)Peil et al. (1990a, 1990b) also found that the catalyst surface is active in the formationof CO, CO2, and possibly even C2H6. Based on studies over alkali promoted alkalineearth oxide catalysts, Aparicio et al. (1991) concluded that gas phase C., combustionalone cannot account for the observed C2 yield limit, and surface catalyzed C2combustion may play an important role in determining selectivity.Kennedy and Cant (1991) found that ethane was four times as reactive as methane overthe rare earth oxides, La203, CeO2, Sm203, and Pr6011 , under conditions of 13%hydrocarbon, 3.5% oxygen, with the balance helium, at 1 atmosphere pressure, and agas phase residence time of 0.2 seconds. The selectivities of the catalysts weredependent on the average lifetime of alkyl and alkylperoxy species in the gas phaseand the ability of the catalyst to oxidize alkyl radicals. The strong oxidizing power ofthe Ce and Pr oxides is used to explain the very low ratios of carbon monoxide tocarbon dioxide found for these catalysts.Homogeneous chain-branching reactions, initiated by the reaction of oxygen withheterogeneously formed methyl radicals, may participate in both the selective and non-15selective 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 surface-generated radicals. The chain-branching reactions are likely partially responsible forthe conversion of C 2H6 to C2H4 and for the formation of CO), (Lunsford, 1991).Secondary reactions of CH3- radicals with the metal oxide also may contribute to theformation of COX.Grzybek and Baerns (1991) conclude that the total oxidation of methane occurs at leastpartly as a surface process in the presence of gas phase oxygen, with the initial step thesurface oxidation of methane or methyl radical totype products, followed by surface oxidation to compounds containing aC=0group, possibly H2CO, followed by surface oxidation to surface carbonate or CO2 byoxygen adsorbed from the gas phase.Many kinetic models have been developed based on different reaction schemes. Themodel developed by Machin (1992) took into account other models and developed asimplified model consisting of seven elementary steps, 5 homogeneous and 2heterogeneous, with one adjustable parameter to match the model results withexperimental data. An independent correlation between the activation energy of therate determining step, the adjustable parameter, and the physical-chemical properties16of the solid catalyst was also developed. The rate determining step is assumed to bethe H abstraction by an active oxygen species leading to a hydroxyl group on thesurface of the catalyst and a methyl radical in the gas phase. The formation of ethaneis accounted for by the recombination of methyl radicals in the gas phase. Ethylene isformed in the gas phase from radicals produced by active catalyst sites. It should benoted that various mechanisms have been proposed by various researchers, and thatall are not consistent with this mechanism. The 7 reactions are as follows:CH4 + 0: —> CH3- + Os—H (2.14)2 CH3- —> C2H6 (2.15)C2H6 + 0: —> OS -H+ (C2H5 - —> C2H6) (2.16)C2H6 + CH3* - CH4 + (C2H3 • ''-4 C2H4) (2.17)CH4 + 02 -) CO, (2.18)C2H6 + 02 -4 C0x (2.19)C2H4 + 02 -> CO, (2.20)A limit to the C2+ yield (<33%) is predicted to exist under conditions of hightemperature (>700°C) and one atmosphere pressure. The low C2+ yield is believed tobe due to the fact that the temperature of the oxidative coupling process is higher thanthe autoignition temperature for methane and ethane.Based on a kinetic model of 9 to 11 equations, Labinger (1988, 1991) found an upperlimit of approximately 30% yield of higher hydrocarbons at one atmosphere and about25% yield at pressures of greater than one atmosphere. Hair et al. (1992) carried out17modelling of oxidative coupling of methane, and showed that when the selectivecatalytic 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 thescenario which would result for a very active catalyst which selectively activated themethane according to reaction (2.21). However, since the C2 yield which is actuallyachieved in practice is much lower (<30%), the catalysts used must either not achievea high rate constant, or they must contribute to the non-selective reactions. It is likelythat both of these factors contribute.2.2.4 Gas Phase ReactionsThe gas phase reactions have been studied both in the presence and the absence of acatalyst in order to determine the relative importance of these reactions.Yates and Zlotin (1988) studied the rate of oxidative coupling of methane over MgOand 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 thesecatalysts resulted in a change in the principle combustion product from CO to CO 2, aswas also found by Geerts et al. (1989).18Hutchings et al. (1989a) found that at temperatures _^ 700°C the gas phase nature of theoverall reaction is dominant, at reaction conditions of CH 4/02 = 3 and total pressureof 85 kPa.Kalenik and Wolf (1990) studied the effect of gas phase reactions and presented adiscussion aimed at clarifying different interpretations of their results. The reactionconditions must be considered carefully to avoid conflicting conclusions. For residencetimes longer than 0.1 min, methane conversion is significant and may contributeappreciably 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 studiesconducted under highly diluted feeds are therefore not affected by the gas phaseinitiated reactions.Olsbye and Desgrandchamps (1991) studied the effects of the pre- and post-catalyticreactor volume, using a 25% Ba/La2O3 catalyst, at 830°C and CH4/02 = 10. The pre-catalytic volume was of negligible importance under catalytic conditions, indicatingthat, under these conditions, the conversion of reactant gases is not influenced by thehomogeneous gas phase. At complete oxygen conversion, only the conversion of C2H6to C2H4 occurred in the gas phase, while at lower oxygen conversions, the selectivitieschanged towards CO„ products. A larger post-catalytic volume led to a lower methaneconversion. This is probably due to the formation of CH4 from the products in thepost-catalytic gas phase. This interpretation is supported by the observation that theC2 /COX ratio decreased with increasing post-catalytic volume even at complete oxygen19conversion.2.2.5 Oxidative Coupling of Methane Under Conditions of High PressureThe commercial implementation of the oxidative coupling of methane process is likelyto be carried out at pressures in excess of atmospheric pressure, under which themajority of the testing has been done. Some of the investigations which have beencarried out to determine the effect of increased pressures are as follows.Asami et al. (1987) studied the homogeneous oxidative coupling of methane atpressures up to 1.6 MPa in the temperature range 650°C to 859°C, and under diluteconditions (CH4: 02: N2 mole ratio = 14: 1.6: 84.4). An increase in pressure resulted inincreased methane conversion and decreased hydrocarbon selectivity. A C2+ yield ofca. 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 wassignificantly 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 dopedlanthanum oxycarbonate catalyst, using nondiluted methane/oxygen mixtures atpressures up to 7.5 bar (0.75 MPa). The increased pressure had a negative effect onboth methane conversion and C2+ selectivity. The C2+ yield decreased from 9.9% at20atmospheric pressure to 3.4% at 7.5 bar. The negative pressure influence on theoxidative coupling of methane reaction can be minimized by appreciably increasing thelinear space velocity for pressures up to 3 bar; under these conditions the C2+ yieldincreased to 10.3%.At a pressure of 585 kPa and temperatures of 500°C to 575°C, the oxidative couplingof methane gives similar product yields in the presence of Li/MgO or Sm 203, and inthe absence of catalyst, indicating that at this pressure, sufficient gas phase radicals toinitiate the reaction are readily formed (Hutchings et a/., 1988, 1989a). The role of theoxide 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 psigor 6.3 MPa) than other studies, and at temperatures of 550°C to 600°C. Increasing theoxygen 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 increasesfrom 3 to 10 MPa, the temperature required for complete consumption of oxygendeclines 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% CH4conversion). The decreasing influence of a known oxidative coupling catalyst, Sm203,with increasing pressure is consistent with the view that the catalyst is primarily aradical generator in conventional catalytic oxidative coupling.212.3 The Role of Gas Phase, Surface, and Lattice OxygenThe active species for methane activation is an oxygen species, the exact nature ofwhich is not conclusively known despite extensive studies.2.3.1 Source of the Active Oxygen SpeciesSeveral studies have shown that the presence of gas phase oxygen is necessary for theactivation of the catalyst (Lin et a/., 1986), and that lattice oxygen is not active in theabsence of gas phase oxygen (Ekstrom and Lapszewicz, 1988b, 1988c, 1989a; Kalenikand Wolf, 1991a, 1991b). However, the actual active species may be either an adsorbedor lattice species. Amorebieta and Colussi (1988, 1989) determined that, although gasphase oxygen must be present, the active sites are an adsorbed form of oxygen.Ekstrom and Lapszewicz (1988b, 1988c, 1989a) determined that the lattice oxygen atomsexchange easily with the gas phase molecules, and that the lattice oxygen does notparticipate in the formation of the reaction products in the absence of molecularoxygen. Although the rate of lattice oxygen exchange determines the catalyst activity,the molecular gas phase oxygen is somehow involved. Lo et a/. (1988) concluded thatonly surface oxygen species are used for the formation of C2's over Sm203; however,Peil et al. (1989) concluded that lattice oxygen also contributed significantly to thereaction, for both Li/MgO and Sm 203. Otsuka and Said (1987) determined that theactivity of the adsorbed oxygen for converting methane was more than three orders ofmagnitude greater than that of the lattice oxygen.22Kalenik and Wolf (1991a, 1991b) conducted isotope studies, in which 1802 replaced theoxygen in an oxidative coupling reactor over a catalyst which contained 1602 as latticeand surface atoms, for both La203 and Li/La2032Ti02. The surface and bulk oxygenappeared in the reaction products before gas phase 7802, indicating that lattice andadsorbed oxygen were responsible for methane activation. Gas/solid exchangeinvolved over 50% of the lattice oxygen.Comparison of La203 and Li/MgO revealed different behaviour when gas phaseoxygen was removed after treatment of the catalyst in oxygen (Lin et a/., 1986). ForLi/MgO, residual activity remained for a period of minutes after the oxygen wasremoved, while for La203, no such residual activity was seen. The difference may bedue to the fact that the active species for the Li/MgO may be a part of the lattice;whereas, with La203, the active species may be a surface species (e.g., 02). The La203is less selective than the Li/MgO, probably due to further oxidation of the C2compounds.Statman et al. (1991) studied the oxidative coupling of methane on Ba/Sr/Sm 203 usingTAP (Temporal Analysis of Products). Methane was either not adsorbed, or weaklyadsorbed on the catalyst. Oxygen was strongly adsorbed at temperatures above 500°C,which suggests incorporation into the lattice with possible formation of surface orsubsurface anions. They concluded that the rate of ethane production depends uponthe rate at which active surface oxygen species were formed from gas phase molecularoxygen. The formation of ethylene appears to occur in series with ethane, and maypossibly involve the same surface oxygen species responsible for the formation of23surface CH3• radicals.Ekstrom (1992) studied the exchange of lattice oxygen for Sm 203 using isotopeswitching experiments (see Table 2.1). The results indicate that large amounts ofoxygen are exchanged with an oxygen pool present in the catalyst, comparable to thetotal 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 surfacedecreased 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 betweencatalyst activity and the rate of gas-lattice oxygen exchange.Table 2.1Comparison of Catalyst Activity and Oxygen Exchange Parameters for Cubic andMonoclinic Sm203Parameter Cubic MonoclinicTotal "surface" 160 atoms desorbing per gram ofcatalyst (x 10 -20) 113 0.2Lattice oxygen diffusivity (cm2/s x 1017) 1 13,600 886CH4 conversion rate (p.mol/s/g)2 102 6Fraction of lattice oxygen exchange during 30s 1802pulse20.6 0.09The role of lattice oxygen was investigated by first exchanging the lattice oxygens with18,2,u followed by introduction of methane into the Sm203 catalyst bed (Ekstrom, 1992).No reaction was observed, indicating that the lattice oxygen atoms played no role in'Pell et al., 19922Ekstrom, 199224the formation of the reaction products in the absence of molecular oxygen. Additionof 1602v immediately led to the formation of 180 labelled CO2, indicating that the latticeoxygens are responsible for CO 2 formation.2.3.2 Nature of the Active Oxygen SpeciesSeveral oxygen species have been suggested to be responsible for the activation ofmethane in the oxidative coupling of methane. Although many researchers agree thatO" 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 forboth the selective and non-selective oxidation of methane. A complicating factor whencomparing various studies is that different catalysts may react according to differentmechanisms, and the method of preparation may even have a significant effect.The Li/MgO catalyst has been extensively studied and the accepted active species forthis 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 (Driscollet al., 1985). However, MgO has low C2+ selectivity and activity and the addition ofsubstitutional Li+ ions causes a large increase in the activity. The Li+ ions are believedto react with molecular oxygen to form a [Li+0 -] centre. A good correlation was foundbetween the amount of methyl radicals formed and the [Li+0 -] centres formed as afunction of lithium doping.Li2CO3 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 dioxidefrom the surface carbonate species in the presence of oxygen, and that the presence oflithium carbonate on the surface of the catalyst is therefore crucial for its activity. Penget al. (1990) observed both Li -40- and Li2CO3 on the surface of Li/MgO under reactionconditions. The correlation between methane conversion and the surface Li+0-concentration demonstrated that [Li+01 species were the active centres for methaneconversion.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-() isthe oxidizing species responsible for CH 3• radical formation. A second surfacediatomic oxygen species was indicated which was responsible for non-selectiveoxidation of ethylene and ethane (Hutchings et al., 1989b, 1989c). This species isestablished 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 elementarysteps. Alkali metals are known to form stable bulk peroxides, and a surface peroxidespecies formed by adsorption of gaseous diatomic oxygen may dissociate to form themethane-activating sites. The cleavage of the 0-0 bond is thought to determine theoverall activity by forming the active G ions.Otsuka et al. (1986a) conducted studies over Sm203 and initially proposed that theactive species were G ions produced on oxygen vacancy sites or basic sites. The deep26oxidation of methane may be caused by surface 0 2- or by adsorbed oxygen. The largedifference in activation energies for the selective and non-selective reactions supportsthe idea that the reaction intermediates of the oxygen species responsible for thesereaction paths are different. Further kinetic studies supported the mechanism ofadsorption of the methane and oxygen independently on different active sites (Otsukaand Jinno, 1986b; Otsuka and Nakajima, 1987). Evidence suggested that adsorbeddiatomic oxygen species were more likely responsible for methane activation over rareearth metal oxides and alkali metal promoted oxides (Otsuka et al., 1987). Peroxideanions (022-) were indicated, while 02 ions were not. Studies conducted with sodiumperoxide at low temperatures (327°C to 377°C) suggest that the activation of alkanes iscaused only by peroxide anions, but the selectivities to further reactions of the surfacealkyl groups formed are affected strongly by the presence of gaseous oxygen (Otsukaet al., 1990).Yamashita et al. (1991) studied the formation and decomposition behaviours of surfaceoxygen species on BaO/La203. They found 022- species on the surface of the La203, andan increased concentration on 15%Ba/La 203 (designated as 15BLO) on which thebarium species were well dispersed. The high catalytic ability of 15BLO was ascribedto this dispersed species on the surface. However, increasing the concentration of Bato 50% resulted in a significant decrease in C2 yield from that of 15BLO, to about thesame value as that of undoped La 203. This may be due to poor dispersion of thebarium species. It was concluded that this indicates that the actual active oxygenspecies is probably formed from the decomposition of the surface 0 22", probably to 20 - .The importance of barium dispersion may relate to this step of 0 22- splitting. The27temperature region which corresponds to a high catalytic activity roughly coincideswith the critical temperature for the formation and decomposition of barium peroxide(ca. 600°C to 800°C). The dispersed Ba02 may act as a mediator to transfer 0 22- or aions to La203 active sites. 022- on aggregated Ba02 may be too stable to be decomposedinto active oxygen species.Spinicci (1991) studied zinc oxide and zinc oxide based catalysts, and concluded thatmethane reacts through two distinct pathways with different rates, according to thesurface oxygen species which contribute to the process of activation. Strong adsorptionof CH3• species, possibly by the less basic a or 022-, may lead to CO. formation at alow rate, whereas weaker adsorption of methane, on the more basic 02-, may lead toC2 formation, at a higher rate. The presence of gas phase oxygen is required to slowlyregenerate 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 aspecies. The proposed sequence of oxygen species is as follows:+e +e +2e> 2020 21—  (2.22)° 2(g) ° 2(ads)Amorebieta and Colussi (1988, 1989) determined that the kinetically active species areproduced by reversible dissociative chemisorption of 0 2(g) on the ionic surface of bothSm203 and Li/MgO. The presence of gaseous oxygen is required to maintain the activesites, and these will disappear upon its removal. Oxide (0) or superoxide (02) ions,28as well as [Li+01 centres, are kinetically relevant species on this catalyst, but peroxideions (022) are specifically excluded.Ekstrom and Lapszewicz (1988b, 1988c, 1989a; Ekstrom, 1992) conducted isotope studiesusing labelled gas phase oxygen over Sm 203, Li/Sm203, and Pr6011 , and found that therate determining step is the desorption of molecular oxygen. These results suggest thatthe lattice oxygen atoms themselves cannot be a significant source of reactant oxygenand that molecular, gas phase oxygen is somehow involved. A mechanism based onthe formation and reactions of [01 (an oxygen atom trapped in a vacancy) species isproposed for the C2+ products, but a different form of activated oxygen appears to beresponsible for the formation of the carbon oxides.The mechanism developed by Ekstrom and Lapszewicz for the formation of the activeoxygen species for the formation of C2 products is shown below.anion vacancy[01^oxygen trapped in the vacancy{-0-}^lattice oxygen18028as + (surface) ieA 18n`-'2(surface)v.18°2(surface) + 2[1^2[1801[1801 ±t u }lattice^(-180—Lattice + [ 160-}(2.23)(2.24)(2.25)29The active oxygen species is [a], which abstracts hydrogen from the methane moleculeto form the methyl radical. In this mechanism, the exchange reaction between gasphase 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 [01species is known to be highly reactive toward C-H bond cleavage, and it provides thenecessary link between the rate of oxygen exchange and the rate of reaction. The anionvacancy is regenerated by the formation of water.Ekstrom and Lapszewicz suggest that two oxygen species, [A] and [B], are formed from02 adsorbed on the surface, whose concentration is very small. Hydrocarbons reactrapidly with species [A] to give the corresponding alkyl radicals. Species [B] reactsrapidly with the alkyl radicals to give carbon oxides. [A] may be [01, but the natureof [B] is not known, but it may be 0 atoms, OH- or H02• radicals.For Sm203, Peil et al. (1989, 1990b) determined that on the order of 50% of the Sm 203working surface was involved in the formation of products. Of this 50%,approximately 60% was involved in CO. formation, with the remaining 40% active forC2 formation. The sites active for C2 formation had a lower activity than sites activefor the formation of CO.. Further studies using steady state isotopic transient kineticanalysis (SSITKA) on MgO, Li/MgO and Sm203 indicated that all sites are equallyactive for both selective and non-selective oxidation, and that whether selective ornonselective oxidation occurs may depend on the oxidative/reductive state of the active30site (Peil et a/., 1991a, 1991b).A model of the catalyst was developed with the oxygen considered as existing in threeregions, i) the physical surface at which exchange between the gas phase and the solidoccurred, ii) several subsurface atomic layers readily available for exchange, and iii) thebulk oxide.Otsuka and Hatano (1992) suggest that the adsorbed oxygen species, as opposed to thelattice 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 commonlybelieved, but is diatomic, such as 02, 0 22-, or chemisorbed 02. Based on electron spinresonance studies (ESR), they conclude that 022- anions must be responsible for theactivation of methane over a sodium peroxide (Na 202) sample, and is also likelyresponsible for the catalytic activity of Sm203. However, 022- can be regarded as adimer of 0-, (i.e., a--co, and it has been observed that a does form from 022-(Hutchings and Scurrel, 1992). Therefore, several authors have concluded that 0 - is theselective oxidizing species for Sm 203 .2.3.3 Crystal Lattice Oxygen MobilityThe mobility of the lattice oxygen ions and the rate of lattice oxygen exchange with gasphase oxygen, have been shown to determine the rate of methane conversion (Ekstromand Lapszewicz, 1988b, 1988c, 1989a).31The addition of Li to MgO results in an increase in the lattice oxygen mobility, with acorresponding decrease in the activation energy for diffusion from 63.5 kcal/mole forpure MgO to 14.6 kcal/mole for Li/MgO (Peil et a/., 1991a, 1991b). Similar results werefound by Kalenik and Wolf (1991a, 1991b) for Sr doped La 203. The addition of thelower valence cation to the bulk oxide may create lattice defects in the crystal, whichresult in higher oxygen ion mobility. The oxygen vacancies formed by this doping canreact with gaseous 02, resulting in the formation of 0' ions.This effect was also found for perovskite-type catalysts, which are known to be ionicconductors by means of 0' anions. Vermeiren et al. (1991) found this for catalystsconsisting of alkaline earth metal elements and Ti, Zr and Ce, and Alcock et a/. (1992)found this for LaosSr0.2Y02.9 and La0.9Sro.1Y01.43 catalysts. Oxides with higher oxygenvacancy concentration were found to be more active for the oxidative coupling ofmethane. Substitution of part of one of the perovskite component elementsconsiderably improves both the catalytic conversion and C2+ selectivity. The positivecharge deficiency in the lattice created by substitution is neutralized by the formationof oxygen anion vacancies, resulting in an increase in the rate of oxygen isotopeexchange. This manifests itself in the fast regeneration of the surface oxygen vacanciesand surface 02" .Kalenik and Wolf (1992) studied the correlation of oxygen available for isotopeexchange with catalyst performance for the oxidative coupling of methane. The effectof doping lanthanum oxide with strontium and doping zirconium and thorium oxideswith calcium was studied. An increase in methane conversion resulted from promotion32of the oxides with the dopants of lower valence, of similar ionic radii.A low C2+ selectivity was observed for zirconium dioxide, which the authors concludedwas due to the formation of surface carbonates, resulting in a decrease in the numberof lattice defects and decreasing the capability of the oxygen atoms to diffuse throughthe carbonate. The undoped zirconium dioxide exhibited low oxygen exchangecapabilities which were significantly increased by doping with calcium. However, inboth 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 differentvalence to lanthanum oxide. They report that at temperatures greater than 327°C andoxygen partial pressures above 10' bar, lanthanum oxide exhibits conduction consistingof a minor contribution from ionic conductivity due to oxygen vacancies and of a majorcontribution from defect electron conductivity. They also report on the results ofanother study, in which the C2 selectivity and oxygen ion conductivity were comparedfor CaO-Ce02 catalysts of various compositions. It was found that the dependence ofoxygen 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 numberof oxygen ion vacancies present. Addition of lower valence dopants results in anegative effective charge which may be compensated for by the formation of positivedefects, such as oxygen ion vacancies. Higher valence dopants would result in a33decrease in oxygen ion vacancies. The increase in oxygen ion vacancies should resultin an increase in ionic conductivity. Lanthanum oxide (La 203) was doped with cationsof 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 tolack 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 comparedto the C2+ selectivities. There appears to be a direct relationship between an increasein oxygen ion conductivity and an increase in C2+ selectivity, based on these fourcatalysts.2.3.4 Concentration of Active Sites on the Catalyst SurfaceMcCarty (1991) reported on a model of the oxidative coupling of methane (kinetic andthermodynamic), which includes 134 reversible homogeneous reactions and a set ofheterogeneous reactions. The model indicated that an optimum concentration of activesites exists, and that too many sites is detrimental. The C2+ selectivity increased withthe increase in reactive centre concentration to reach a maximum, and then decreased.The methane conversion increased steadily with the increase in active siteconcentration. Higher than optimal surface concentrations of oxygen result in greatermethane conversion rates, but with very low coupling selectivity, due to rapidheterogeneous oxidation of intermediate methyl radicals. Lower than optimalconcentrations of reactive surface oxygen leads to both lower conversion and lower C2+34selectivity because homogeneous methyl radical oxidation processes compete moreeffectively with the second order coupling reaction under these conditions. Selectivityis predicted to decline quickly with increased oxygen partial pressure because of directoxidation of the methyl radicals.2.4 Samarium Oxide Crystal StructureSamarium oxide exists in one of two forms: the metastable cubic structure is changedirreversibly into the stable monoclinic structure at around 850°C to 900°C. The cubicstructure has been shown to be more active as an oxidative coupling catalyst than themonoclinic structure, and this is believed to be due to the increased oxygen mobilityin the cubic structure. In the cubic structure, the samarium atom is surrounded by sixoxygen atoms and two lattice oxygen vacancies at the eight equidistant corners of acube (see Figure 2.1) (Anshits et a1., 1990; Peil et al., 1992). The oxygen vacancies, whichare located either on the face diagonals or body diagonals, are ordered in such a waythat the oxygen atom vacancies lie in a straight line through the body diagonal of thecubic crystal. This configuration essentially provides a "pipeline" for the oxygen atomsto diffuse into and out of the bulk crystal structure. The monoclinic structure isidentical to the cubic structure, with an additional oxygen atom along a threefold axis.This seventh oxygen atom results in a distorted octahedral coordination about thesamarium atoms. Not all monoclinic unit cells contain an "extra" oxygen atom, so thestructure consists of both sixfold and sevenfold coordination about the metal atoms.The symmetry about the metal atom in the cubic crystal compared to the monoclinicresults in a higher degree of lattice oxygen mobility.35Figure 2.1The Crystal Structure of Cubic Samarium Oxide Sm203 10 - Sm 0 - 0^Oxygen VacancylAnshits et al., 1990362.4.1 Comparison of Cubic versus Monoclinic Sm203The ability of the cubic and monoclinic crystals to exchange oxygen has been studiedand compared to the catalyst activity (see Table 2.1) (Ekstrom, 1992; Peil et al., 1992).The cubic Sm203 was capable of faster exchange, and had more oxygen atoms availablefor exchange. This correlated with the increased catalyst activity observed for the cubicstructure.The two forms have the same basicity, which allows important comparisons to be madewithout the influence of this variable. The only apparent significant difference betweenthe 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/Sm203. Electricalconductivity measurements indicated that the oxygen vacancy concentration and theirmobility are directly related to catalytic activity and C2+ selectivity in the oxidativecoupling of methane. The mixed phase structure had the highest mobility of oxygenvacancies, which may be due to additional defects created in the interface between thecubic and monoclinic phase.Korf et al. (1991) compared cubic and monoclinic Sm203 and La203, and developed apossible reaction scheme for the formation of ethane and CO, in the oxidative couplingof methane. In this scheme, two adsorbed methyl groups must combine to form37gaseous ethane. The structure of the cubic modification of Sm 203 is less closely packedthan that of the monoclinic form and may present a more favourable geometry for thisreaction. The ethane must readsorb on the surface, to form the secondary product,ethylene. The CO. is formed by interaction of CHXO species (formed from adsorbedmethane and an adsorbed oxygen species) with surface oxygen, rather than directlyfrom methane or from surface CH3• species.Finally, the difference in surface geometry between cubic and monoclinic forms mayalso give a change in the nature of the oxygen surface species responsible for theformation of the CHXO 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 C2products.2.5 Effect of Water Addition on the Oxidative Coupling of MethaneChoudhary et al. (1991) studied the effect of water addition on a non-catalytic oxidativecoupling system. Addition of water was highly beneficial to obtaining higher C2+ yieldand/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 C2+ 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 of38the water was essential to obtain the high selectivity. Pereira found that, in thepresence of steam, considerably higher C2+ selectivity and yield (90 and 9.9% comparedto 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 oxidecatalyst at 570°C to 650°C. However, they did not obtain the high selectivity obtainedby the other researchers, and they obtained higher methane conversion. For somecatalyst formulations, they did obtain high selectivities, but these did not last for anextended period.Buyevskaya et al. (1992) studied the effect of dilution of the feed with steam overseveral catalysts. Steam mole percentages of 50-70% at 700-850°C and CH 4/02=1.5-4.9were tested. The effect of steam addition was dependent on the methane to oxygenratio. With 10% Sm203/Mg0 as the catalyst, at 850°C and CH 4/02=4, the C2 yielddecreased slightly from 19.6% with helium as the diluent, to 19.3% with steam as thediluent. However, at CH4/02=2.3 and with other conditions the same as the previouslymentioned 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 andfound that C2+ selectivity was not influenced, but the CH 4 conversion fell to 74% of itsinitial value when 20 mole % H 2O was added to the feed.39The effect of water addition seems to be very dependent on both the catalyst and thereaction conditions. With low methane to oxygen ratios the addition of water appearsto positively influence results. Van der Wiele et al. (1992) studied the influence ofwater for a residence time of 0.032 seconds and a reaction temperature of 800°C. Theonly effects of the water were to increase the CO2 formation slightly, with acorresponding drop in C2 selectivity.2.6^Potential Reactor ConfigurationsMost of the laboratory studies on the oxidative coupling of methane have been carriedout in packed bed reactors. This is not a suitable configuration for practical processingdue to the high exothermicity of the reactions, which set up large temperaturegradients in the catalyst bed. At 25% methane conversion, the oxidative coupling ofmethane reaction over magnesia based catalysts has an adiabatic temperature rise of1250°C (Leyshon, 1991). A large-scale fixed-bed reactor would require operation as amulti-bed unit with staged oxygen addition (Edwards and Tyler, 1988). Various reactorconfigurations 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 foundthat the effectiveness of both types of monolith is not as great as the same materials ina conventional packed-bed.Fluidized bed reactors are a promising configuration for the reaction due to the abilityof the reactor to disperse the heat of the reaction. However, the stability of the40catalysts and their tendency to agglomerate cause operational problems with thisconfiguration. The effect of catalyst agglomeration for Na 2CO3/CaO catalysts wasovercome by admixing a-Al 203 particles with the catalyst (Andorf et al., 1991) . TheC2+ 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 overLi/MgO using both fixed and fluidized bed reactors. In the fixed bed reactor, methaneconversion was limited to 15% due to the need to avoid excessive temperaturegradients. The fluidized bed reactor operated essentially isothermally at methaneconversions in excess of 40%. However, some loss in C2+ selectivity occurred intranslating from fixed to fluidized bed reactors. The performance of the fluidized bedreactor with 5-10% ethane in the feed was significantly improved with respect toethylene production compared with the case where the feed was methane and oxygenalone. These studies led to the development of the 'OXCO' process, which consists ofa fluidized bed reactor to which ethane is injected into the oxygen free zone of thefluidized bed (Edwards et al., 1991). Methane coupling and the pyrolysis of higheralkanes are efficiently combined within a single fluidized bed reactor resulting in thetotal utilization of natural gas. With Australian natural gas (36% of carbon content isethane and higher alkanes), the overall selectivity is potentially 85% for conversion tounsaturates.Molten materials have been used as catalysts for the oxidative coupling of methane.Metal oxides dissolved or dispersed in molten metals achieved high C2+ selectivity41(>99%) but with a low yield (<1%) (Fujimoto et al., 1991). The use of molten bariumhydroxide 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, inwhich a thin bed of catalyst was placed in a heat conductive catalyst holder andmounted 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 MgOcoated on a porous alumina tube, subsequently coated with lead nitrate and followedby 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 sourcesallows the use of air, as the nitrogen does not come into contact with the methane andis not transferred through the membrane. The transfer of the oxygen through the non-porous membrane is a good indication of oxygen transfer through the lead oxide.2.7 Engineering and Economic AssessmentsA number of economic assessments of the oxidative coupling of methane process havebeen carried out to determine the C2+ selectivity, yield and methane conversionrequired to achieve economic viability.42A comparison of the direct methane conversion processes considered oxidativecoupling, partial oxidation, and oxyhydrochlorination (Fox et al., 1990). Of the three,the oxyhydrochlorination was the most economical, due to higher selectivity, despitethe need for special materials of construction.Another study evaluated direct partial oxidation to methanol and oxidative couplingto ethylene for making liquid fuels, and compared them with a conventional naturalgas-to-methanol via steam reforming process (Kuo, 1991). The study indicated that theoxidative coupling to ethylene process could become competitive if >88% selectivity at35% single-pass conversion was achieved.Lee and Aitani (1991) based an economic evaluation of the oxidative coupling ofmethane on results reported by Otsuka in 1988 (30.6% yield with a 64.7% C2+selectivity). They concluded that these methane conversions and product selectivitiesare sufficiently high to make the process economically attractive compared withconventional ethylene production.In a review of the state of process technology for the direct catalytic conversion ofmethane, the reviewers concluded that direct methane conversion is not yet competitivewith conventional processes (Mobil MTG process) (Poirier et al., 1991).Based on the C2+ yield and selectivity required to achieve an economic process, andconsidering the possible existence of an inherent yield limit in the oxidative couplingof methane as discussed previously, the possibility of a full scale commercial process43based on the oxidative coupling of methane is still questionable. The economics of theprocess must be improved, with further research aimed at increasing understandingof the catalyst mechanism and improving reactor configurations. The objective of thisthesis is to evaluate the effects of the catalyst properties on the oxidative coupling ofmethane reaction, and to relate these effects to the possible catalytic reactionmechanisms occurring.443. METHODS AND MATERIALSThe experimental part of this thesis consisted of making and testing samarium oxideand alkali and alkaline earth doped samarium oxide catalysts for the oxidative couplingof methane. The catalysts were prepared from samarium and dopant nitrates in amolten state, which decompose at elevated temperatures (up to 600°C) to form the solidoxide. The dopant concentration and the preparation procedure were varied todetermine the effects of these variables on the catalyst performance. The catalysts weretested in a bench scale packed bed reactor, into which methane, oxygen, and heliumwere fed. The exit gas stream was analyzed using an on-line gas chromatograph.3.1^Experimental EquipmentA schematic flow diagram of the equipment set-up is shown in Figure 3.1. The feedgases consisted of oxygen, ultra high purity (UHP) methane, and diluent helium. Theflows 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 andatmospheric pressure. The gases passed through 5 micron filters prior to the mass flowcontrollers and a 1 psi check valve immediately after the flow controllers to preventany reverse flow of gas through the control valves. The methane and helium streamswere combined and entered the reactor at the top of the reactor tube. The oxygenstream entered the reactor via a quartz feed tube which ran down the centre of thereactor tube and exited about one half a reactor diameter above the catalyst bed. Thisappeared to be an adequate distance for sufficient mixing of the methane and oxygen45,LL^SIGNALS TOCOMPUTERXo VALVECHECK VALVEe FILTER0 FLOW MEASUREMENT0 THERMOCOUPLE0 PRESSURE MEASUREMENTMASSFLOWCONTROLLERdddBEDrz0 •TEMPERATURECONTROLLER_JEQUIPMENT FLOW DIAGRAM•^•BUBBLEFLOWMETER•I* •GASCHROMATO-GRAPH —OWLFURNACECATALYSTtiCOMPUTERL _ _FLOWMETERCONDENSERLEGENDREACTORBYPASSzw0II Istreams while minimizing the pre-catalytic time for reactions. A pattern of carbondeposition on the catalyst particles would have indicated cracking of the methane inthe absence of oxygen. Since this was not apparent, it was assumed that adequatemixing occurred. Nitrogen was available at the bottom of the reactor to quench the exitgas 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). Importantfeatures of the reactor included the reduction of reactor diameter immediately after thecatalyst bed to reduce residence time of the exit gas in the heated zone and the abilityto quench the gas with nitrogen if the exit gas leaves the furnace zone at elevatedtemperature.To monitor the temperature profile of the reaction gas, the temperature was measuredin three places by thermocouples: the entrance to the reactor; immediately below thecatalyst bed by a quartz sheathed thermocouple; and the exit of the reactor. Inaddition, a thermocouple was positioned between the outside wall of the reactor andthe furnace interior, to monitor the actual furnace temperature. The thermocouple inthe heated zone of the reactor directly below the catalyst bed was protected by a quartzsheath as almost all metals have been shown to be reactive for the oxidative couplingof methane. The temperature measured by this thermocouple was assumed to be thereaction temperature.A pressure transducer connected to a digital meter was used to monitor the reactor47NITROGEN0.125 -II-SIDE VIEW1514CATALYSTSUPPORT \ METHANE^OXYGEN I15-1F0251/8' QUARTZ THERMOCOUPLESHEATH1/16' THERMOCOUPLE4-0.125PRODUCTPLAN VIEWPACKED BED REACTOR FOR THEOXIDATIVE COUPLING OF METHANE1/8' THERMOCOUPLE•ALL DIMENSIONS IN INCHESpressure. The exit gas passed through a glass impinger contained in an ice bath tocondense any water out of the gas stream. The gas stream then passed through arotameter and exited the reactor system to the on-line gas chromatograph (GC). Afterleaving the GC, the gas stream passed through a bubble flow meter and was exhausted.3.2 Data AcquisitionThe pressure, catalyst bed temperature, and methane, helium, and oxygen flow rateswere continuously monitored on a computer using the data acquisition software,"Labtech Notebook", which was interfaced to the equipment with a Das-8 computerinterface board.3.3 Gas ChromatographA 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 usingon-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 themethane, ethylene, ethane, propylene, and propane.The GC used two columns operated in series and a temperature program to provideseparation of the components. The columns used were a Poropak Q (8 feet) followedby a 5A molecular sieve (6 feet). The sequence which was used is as follows: thesample is injected; the lighter components, including methane, carbon monoxide,49oxygen, 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 directlyfrom the Poropak to the detectors; the carbon dioxide, ethylene and ethane exit thePoropak and pass through the detectors where they are measured; the molecular sieveis then switched back into line and the components contained in it pass through andinto the detectors; the molecular sieve is then switched back out of line, in time for theC3 compounds to exit the Poropak and pass through the detectors. The temperatureprogram used was as follows: initial temperature of 80°C for 5 minutes; temperatureincrease at 10°C/minute for seven minutes up to 150°C; followed by a hold at 150°Cfor three minutes. Approximately 16 minutes was required for each sample cycle ofthe GC, including complete elution of the columns, cool down, and stabilization of thedetector signals. The GC data was recorded and integrated by a ShimadzuChromatopac integrator.3.3.1 GC CalibrationThe 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 gasas the contents are flammable gases under pressure. The oxygen calibration wascarried out by mixing pure oxygen and helium streams in the reactor bypass anddirecting this mixture into the GC. The composition of this stream was controlled bythe mass flow controller, and any errors associated with the mass flow controllercalibration will be passed on to the GC calibration for oxygen.50The GC columns required regular conditioning at 150°C for at least an hour under ahelium flow. This procedure removed substances which tended to accumulate on thecolumns and interfere with GC operation.3.4 Flow Measurement and ControlThe inlet gases were measured and controlled by the mass flow controllers. Thesecontrol valves were factory calibrated at 70°F (21.1°C). However, testing of the massflow 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 endof the product gas line after the GC and immediately prior to exhaust. After correctingfor temperature, the bubble flow meter was considered accurate.3.5 DefinitionsThroughout this report, some conventions will be used to describe the results obtainedin the experiments. These conventions are consistent with the majority of researchreported in this field. The term C2 is used to represent both ethane and ethylene, whileC3 represents propane and propylene; C2+ represents C2's + COX represents bothcarbon dioxide and carbon monoxide. Methane and oxygen conversion denote thepercentage reacted of the original methane and oxygen feed, respectively. C2 yielddenotes the percentage of the original methane feed which was converted to C251molecules. C2 selectivity denotes the percentage of the reacted methane moleculeswhich are converted to C2 molecules. All compounds are measured in moles.C2 Yield = 2*(C2H4 + C2H6)/ (C exiting reactor)^(3.1)C2 selectivity = 2*(C2H4 + C2H6)/ (C in products)^(3.2)where^(C exiting reactor) = CO + CO 2 + CH4 + 2*(C2H4 + C2H6) + 3*(C3)(C in products) = CO + CO 2 + 2*(C2H4 + C2H6) + 3*(C3)Similar conventions are used for the oxidation products and C3s. The space time yield(STY) is defined as the amount of C2's produced per unit time per unit catalyst weight,with the units mol/s/g. Some parameters which are used to describe the reactionconditions include: methane to oxygen mole ratio (CH4/02); fraction, based onpressure or flow, of reacting gases in feed, P*7---(PcH4÷P02)/Ptotal; and weight of catalystover feed flowrate, W/F (g s/mL).3.6^Catalyst Preparation ProcedureDoped catalysts were prepared by mixing the appropriate weights of dopant nitrateand samarium nitrate and melting in a platinum dish at 95°C. After mixing, themixture was allowed to cool and the resulting solid was broken into chunks, whichwere then used as the starting material for the catalyst preparation.The catalysts were prepared in a modified thermogravimetric analyzer (TGA), in a52quartz catalyst preparation boat of 27 mm in length and 11 mm deep. The oxygen flowthrough the TGA and over the catalyst during preparation was approximately 50mL/min. The catalysts were prepared using a temperature program which allowedthe 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 coolThe temperature hold in the middle of the program was to allow nitrogen dioxide tobubble off as the samarium nitrate decomposed.3.7 Experimental ProcedureThe catalysts were sieved and the -16 +32 Tyler mesh fraction (0.5 to 1.0 mm indiameter) was used for catalyst testing. Approximately 50 mg of catalyst was placedon top of the fitted glass disk in the reactor, which resulted in a bed depth ofapproximately 4 mm. The helium flow was set at 100 mL/min and the furnace wasstarted, with the setpoint at 650°C. The furnace heats up very quickly and was atsetpoint temperature in about 10 to 15 minutes.53After a total of approximately one hour of helium flow over the catalyst, the conditionsfor the first run were set (methane and oxygen flows were started and the helium flowadjusted to maintain a constant flow of 100 mL/min). After 15 minutes the initialsample 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 firstrun, five samples were tested, for the second run, four samples were tested, andthereafter, three samples were tested per run. After the initial two samples for the firstrun, the results appeared to be reasonably constant, and the results for the last threesamples 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 eachrun 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 oftemperature on the potentially temperature dependent catalyst properties, suchas carbonate formation.ii) The inherent selectivity and activity of the catalyst is likely to be most apparentat low methane to oxygen ratios; therefore, CH4/02 ratios of 2 and 4 were testedfor comparison of catalyst performance, along with higher ratios to determinethe effects of this parameter.iii)^A univariant experimental design was used, such that only one catalyst54parameter (e.g., dopant) was changed at a time. The interpretation of changesin catalyst parameters is fairly complicated. The change in a parameter, such asdopant used, will have a number of effects on the catalyst properties, only someof which can be measured. The change in catalyst properties will in turn havean effect on the oxidative coupling reactions. Therefore, the effect of the changein one catalyst parameter on the oxidative coupling of methane must beinterpreted based on several catalyst properties. Therefore, to compare theeffects of catalyst properties to that of catalyst performance, the followingparameters were varied in a univariant manner: dopant used, dopantconcentration, and catalyst preparation procedure. Each catalyst was testedunder identical conditions.The conditions used for testing of the samarium oxide catalysts doped at a 1:100 moleratio of dopant to samarium are listed in Table 3.1.3.7.1 Variation of Catalyst Dopant ConcentrationCalcium and sodium doped catalysts were prepared at a dopant to samarium moleratio of 10:100 in order to investigate the effect of dopant concentration. The conditionsused for testing of these catalysts are presented in Table 3.2.3.7.2 Catalyst Preparation ModificationA revised preparation program was tested. This program omitted the hold midway,55and the catalyst temperature was raised from 25°C to 600°C at a rate of 25°C perminute. The catalyst was then held at 600°C for 30 minutes. This catalyst was alsotested at the conditions presented in Table 3.2.Table 3.1Conditions Used for Comparison Testing of Doped CatalystsParameter #1 #2 #3 #4 #5 #6 #7 #8Temperature (°C) 650 650 650 650 750 750 850 850Catalyst weight(mg)50 50 50 50 50 50 50 50Total 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 55W/F (g s/mL) 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03CH4/02 molar ratio 2 4 8 16 4 8 4 8Table 3.2Conditions Used for Comparison Testing of Catalysts at Higher DopantConcentrations and Different Preparation TimesParameter #1 #2 #3 #4 #5 #6Temperature (°C) 650 650 750 750 850 850Catalyst weight (mg) 50 50 50 50 50 50Total 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 55W/F (g s/mL) 0.03 0.03 0.03 0.03 0.03 0.03CH4 /02 molar ratio 4 8 4 8 4 8564.^RESULTS AND DISCUSSIONSamarium oxide and alkali and alkaline earth doped samarium oxide catalysts wereprepared and tested for the oxidative coupling of methane. The effect of specificdopant used, varying dopant concentration and catalyst preparation method wereevaluated. The catalysts were tested in a bench scale packed bed reactor underconditions of varying temperature and methane to oxygen ratio. The catalysts werecharacterized by scanning electron microscopy, powder x-ray diffraction, BET surfacearea, estimated basicity, ability to form carbonates, and ionic radius of dopant.4.1 Undoped Sm203: Conditions of Complete Oxygen ConversionThe first samarium oxide catalyst was tested at a temperature of 750°C under variousconditions of total flow rate, total partial pressure of reactants, and methane to oxygenratio. The conditions and results are presented in Table 4.1 and are shown as afunction of methane to oxygen ratio in Figures 4.1, 4.2, and 4.3. The product gaseswere not analyzed for C3+ compounds. The effects due to methane to oxygen ratio aremuch greater than those due to the amount of diluent used or the total flow rate. Theoxygen conversion was greater than 99.8% for all conditions tested. The methaneconversion 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 increasedfrom 2 to 16. A lower methane to oxygen ratio results in higher methane conversion,lower C2 selectivity, and higher C2 yield.57Table 4.1Samarium Oxide: Conditions of Complete Oxygen ConversionRun #1 #2 #3 #4 #5 #6 #7 #8Input Flow Rates (mL/min) (at 21.1°C, atmospheric pressure)He 50 150 75 85 55 30 15 50CH4 40 40 20 10 40 8 80 4002 10 10 5 5 5 2 5 10Total 100 200 100 100 100 40 100 100CH4/02 4 4 4 2 8 4 16 4P" 0.50 0.25 0.25 0.15 0.45 0.25 0.85 0.50W/F g s/mL0.12 0.06 0.12 0.12 0.12 0.30 0.12 0.12Yield %CO2 10.1 9.3 9.8 22.9 4.2 10.9 1.7 9.8CO 1.2 1.2 1.1 4.8 0.0 0.8 0.0 1.0C2H4 7.5 6.6 6.7 7.7 4.9 6.6 3.1 7.2C2H6 6.1 6.6 6.3 5.9 5.6 5.4 4.3 5.9Total 24.9 23.7 23.8 41.3 14.7 23.7 9.1 23.8Selectivity %COX 45.5 44.0 45.5 67.1 28.3 49.4 18.6 45.1C21 S 54.5 56.0 54.5 32.9 71.7 50.7 81.4 54.9Space Time Yield grnol/g/sC2's 8.9 8.7 4.0 2.0 6.3 1.3 9.0 8.0Conversion %CH4 25.0 23.7 23.8 41.2 14.7 23.7 9.1 23.802 99.9 99.8 100.0 100.0 100.0 100.0 100.0 99.9Reaction conditions: T = 750°C, weight of catalyst = 0.21g58Methane Conversion for Samarium OxideConditions of Complete Oxygen Conversion, T=750C45.0•c 40.0E 35.0ao> 30.0co 25.0c 20.02a..F. 15.02•10.0°•5.00.00^2^4^6^8^10^12^14^16CH4/02Figure 4.1Figure 4.2C2 Selectivity for Samarium OxideConditions of Complete Oxygen Conversion, T=750C90.080.070.0:"; 60.045w 50.0N40.0csi0 30.09.20.010.00.00 ^••$••2 4 6 8 10 12 14 16CH4/0259• $••Figure 4.3C2 Yield for Samarium OxideConditions of Complete Oxygen Conversion, T=750016.014.013 12.0.-6>-01  10.00°^8.06.04.00^2^4^6^8^10^12^14^16CH4/02604.1.1 Effect of Gas Phase Flow RateKeeping all other conditions constant, the total flow rate was varied in order to varythe 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, andthe 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 inthe absence of gas phase oxygen. However, it does appear that some C 2's wereoxidized.Table 4.2Effect of Total Flow RateRun#Q(mL/min)ResidenceTime (s)% MethaneConversion% C2Yield% C2SelectivitySTYp.mol/g/s2 200 0.2 24 13 56 8.73 100 0.4 24 13 55 4.06 40 0.9 24 12 51 1.3Reaction conditions: T=750°C, 20% CH4, 5% 024.1.2 Carbon Deposition in the ReactorAfter the test runs were completed and the catalyst cooled, the catalyst was examinedfor carbon deposits. There was considerable carbon deposition on both the walls of thereactor and the catalyst. This was likely due to the cracking of the hydrocarbons in theabsence of oxygen, to produce carbon and hydrogen. The carbon deposition on the61catalyst appeared only on the bottom half of the catalyst and increased in amountdown the length of the catalyst bed. A small amount of carbon was deposited on thefritted glass disk.These observations indicate that oxidative coupling reactions occurred only in the firsthalf of the catalyst bed, after which all of the oxygen was consumed. Carbon is formedat this temperature in the absence of gas phase oxygen. Comparison of catalysts underthese conditions is difficult since the lack of oxygen may mask the inherent selectivityof the catalysts.4.1.3 Comparison of Results for Sm203 to Other Researchers' DataTable 4.3 is a compilation of results obtained by various researchers for pure samariumoxide catalysts under conditions of 750°C and complete or almost complete oxygenconversion. 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 tooxygen ratio. The results of the present study show the same general trends as theresults of other researchers and are of approximately the same values. It should benoted that reactor configurations and reaction conditions vary widely. The methaneto oxygen ratio has a significant effect on the determination of methane conversion andC2 selectivity.62Table 4.3Samarium Oxide Results for Various ResearchersP(CH4)kPaCH4/02 W/Fg s/mL% MethaneConver-sion% C2Selec-tivity% C2YieldC2 STYinnoligisRef.87 6.1 0.012 14 61 8.5 136 387 6.1 0.0008 5 60 3 719318 5.1 2.4 15.1 60.6 9.2 0.15 45 2.5 30.2 41.2 12.4 55 2.5 30 40 12 668 10.0 0.22 12.2 54 6.6 4.42 751 5.0 0.22 21.9 47 10.3 5.1513 3.9 0.055 31 887 6.1 2.31 50 0.22 981 4.0 0.0014 24 48.9 11.7 62.62 1090 8.1 0.0014 16 62.1 9.8 58.351030 6.0 0.234 17.9 55.8 10 2.86 ii3 Otsuka and Komatsu, 19874 Hutchings et al., 1989a5 Kaddouri et al., 19896 Otsuka and Nakajima, 19877 Korf et al., 19898 Kennedy and Cant, 1991Hamid and Moyes, 199110 Choudhary and Rane, 199111 Deboy and Hicks, 198863esuttsCurrent reher researchers•0•■■Methane Conversion for Samarium Oxide, All ResearchersConditions of Complete Oxygen Conversion, T=7500coEoC0UC024504035■3025 820•1510500.0^2.0^4.0^6.0^8.0^10.0^12.0^14.0^16.0CH4/02C2 Selectivity for Samarium Oxide, All ResearchersConditions of Complete Oxygen Conversion, 1=7500^00Current0.02.0^4.0^6.0^8.0^10.0^12.0^14.0^16.0CH4/02908070:5▪ 60rdi 50ti,f.t) 40N0 30O 201000.0st^••esuttsOther re searchersFigure 4.4Figure 4.5644.1.4 Space Time YieldThe 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 ofhigh 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 thepartial pressure of methane in the feed as a function of W/F, for various catalysts. Alog-log plot permits a fairly good correlation to a straight line for catalysts of variouscompositions and reaction conditions. The present results are plotted along with datafrom other samarium oxide catalysts at 750°C and complete oxygen conversion, andwith data for various catalysts at a reaction temperature of 750°C (from the paper byAmenomiya et al.) in Figure 4.7. Amenomiya et al. (1990) interpreted the apparentrelationship to indicate that the reaction is controlled by homogeneous reactions underconditions of complete oxygen conversion.4.2 Undoped Sm203: Conditions of Incomplete Oxygen ConversionThe major purpose of this study was to compare an undoped Sm 203 catalyst withSm203 catalysts doped with alkali and alkaline earth oxides, in order to determine theeffect of catalyst properties on the oxidative coupling of methane. Under conditionsof complete oxygen conversion, the inherent catalyst activity and selectivity can bemasked by the lack of oxygen. Therefore, the samarium oxide catalyst was retestedunder conditions of incomplete oxygen conversion, which are more suitable for catalystcomparison. The amount of catalyst tested was 50 mg, and the temperatures used were65C2 Yield for Samarium Oxide, All ResearchersConditions of Complete Oxygen Conversion, T=750C)•—• Curren- results••••<'N• Other researchers0.0^2.0^4.0^6.0^8.0^10.0^12.0^14.0^16.0CH4/021412108642010Figure 4.6Figure 4.7(k0.0010.0001STY for Various Catalysts, T=750COther cata ysts^Sm2O3II IIIIM_IIHI' ^II •^IP.* ^• PII0.001^0.01^0.1^1W/F (g s /m1)I^I^ III al1Current results■111---7--'01066650°C, 750°C, and 850°C. At 650°C and 750°C, conditions of incomplete oxygenconversion existed. However, at 850°C, oxygen conversion was 100% for some catalystsand approached 100% for all catalysts. Therefore, the inherent selectivity of thecatalysts will be somewhat masked at this temperature. The conditions used are fullydescribed in Section 3. The complete results are presented in Table A.1.4.2.1 Effect of Methane to Oxygen RatioThe 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 andmethane flows were varied, such that, for a constant oxygen concentration of 5%, themethane to oxygen ratio varied from 2 to 16. This figure shows the change in catalystperformance 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 theresults also appear to be quite steady for each subsequent set of conditions. It isinteresting to observe the effect of the change in methane to oxygen ratio. As the ratiois increased, the C2+ selectivity increases with a simultaneous decrease in methaneconversion, such that the C2+ yield remains almost constant. This effect is not seenunder conditions of 100% oxygen conversion (oxygen limiting), for which the C2 yielddecreases 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 interpretationand discussion. The results are plotted in Figures 4.9 to 4.11 as a function of methaneto oxygen ratio for reaction temperature equal to 650°C. It is interesting to note that67Samarium Oxide ResultsConditions of Incomplete Oxygen Conversion, T= 650C^ 0^CCH4/02=2 CH4/02=4 0^13^o^ n n0CH4/02=8 CH4/02=1^a0^^ ^ 0.• • ■• • • •.^.^••■^• • • • • • • • •^•I^I^I• .1:^•I I^I1^2^3^4^5^6^7^8^9^10 11 12 13 14 15Sample #• CH4 conversion ° C2+ selectivity^• C2+ yield70.060.050.040.030.020.010.00.00Figure 4.8Figure 4.9Conversion for Samarium OxideConditions of Incomplete Oxygen Conversion, T = 650C80.070.0O 60.0050.040.0U• 30.0• 20.010.00.00 ^^4^6^8^10Methane to oxygen ratio12 14 16• CH4 •^0268Selectivity for Samarium OxideConditions of Incomplete Oxygen Conversion, 1= 650C % Selectivity^80.0 ^^70.0 ^60.0 ^50.0 ^440.0 ^30.0 ^a°^'20.0 ^10.0 ^0.0 ^0CO2COCOxC2H4C2H6C3'sC2+5^10^15Methane to oxygen ratio20Figure 4.10Figure 4.11'Ye Yield for Samarium OxideConditions of Incomplete Oxygen Conversion, 1=6500 % Yield14.0 •CO2•12.0CO10.0•— COx(i) 8.0C2H4e 6.0•__••••••••■••■•■•• C2H64.0C22.0C3's--41-0.00 5^10^15^20 0 C2+Methane to Oxygen Ratio69the CO and C2H4 yields are virtually identical over the whole range of CH4 /O2 ratios,with an initial increase up to CH4 /O2 = 4, and then only a very slight increase as theCH4/02 ratio increases. The main effects are seen to be the steady increase in C 2H6selectivity and the steady decrease in CO2 yield as the methane to oxygen ratioincreases.The product outputs, in terms of moles of carbon, are plotted as a function of theamount of methane feed in Figure 4.12. Ethane, ethylene, C31s, and carbon monoxideall appear to be almost linear functions of methane in the feed. The amount of carbondioxide produced, on the other hand, remains almost constant. The CO and CO2produced as a function of oxygen conversion are graphed in Figure 4.13. Although COproduction varies with oxygen conversion, CO2 production appears to be independentof oxygen conversion. These results are consistent with results by Ekstrom et al.(1989c), who determined that the rate of COX production is not a function of thereductant present (e.g., CH4), but of the oxidant. The oxidant is clearly not gas phasemolecular oxygen, since this is present in excess, but must be some other form ofoxygen species (see Section 2.2.3).4.2.2 Effect of Reaction TemperatureDue to the exothermicity of the oxidative coupling reaction, the catalyst bedtemperature may have a "hot spot" temperature which is greater than the measured gasphase temperature. The measured temperature, which is quoted as the reactiontemperature, is actually the gas phase temperature immediately after the catalyst bed.70500 1000^1500^2000^2500^3000Total Carbon x1 0E6 (moles)Product Carbon Output as a Function of Total CarbonConditions of Incomplete Oxygen Conversion, T = 650C^80.0 ^70.060.0 ^250.0 ^^-8' 40.0 ^g 30.0 ^20.0 ^10.0 ^0.0 ^0.---------*■Ai^CO2 —0-- CO^*— C2H4 —0-- C2H6 • C3's••.nLC^0^Figure 4.12Figure 4.1350.0.o^45.0LAJo 40.07‹ 35.032 li; 30.0S •73 25.0>-.5 ^20.013=^15.00^10.0gr-^5.00.025.0Output of Total Oxidation Products as a Function ofOxygen ConversionConditions of Incomplete Oxygen Conversion, T=650C35.0^45.0^55.0^65.0^75.0% Oxygen Conversion85.0• CO2 ° CO71The gas phase temperature was also monitored at the entrance and exit of the heatedzone of the reactor to determine the temperature profile of the gas stream. Thetemperature at the entrance to the reactor did not exceed 80°C, and that at the exit tothe reactor did not exceed 35°C, for all reaction temperatures. This indicates that thegas phase is sufficiently cooled at the exit of the heated zone such that no furtherreactions will occur.The results are plotted for a constant CH 4/02 ratio as a function of gas phase reactiontemperature in Figures 4.14 to 4.16. The methane and oxygen conversions increasewith an increase in temperature. The CO X selectivity decreases as the temperature israised from 650°C to 750°C, and then increases slightly as the temperature is raised to850°C. The C2+ selectivity is opposite to this, with an increase as the temperature israised to 750°C and then a small decrease as the temperature is raised to 850°C. TheCOX yield appears to increase almost linearly with temperature, while the C2+ yieldincreases considerably between 650°C and 750°C, with almost no increase apparentbetween 750°C and 850°C.4.2.3 Carbon Balance over the ReactorThe carbon balance was calculated for each run by comparing the amount of methaneinput 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 forcalculations, since this was considered to be more accurately measured. A loss of 7 to14% carbon was observed. If this was due to carbon deposition, over the course of the72Conversion as a Function of Temperature forSamarium OxideCH4/02=4100.090.080.070.0•r• 60.050.040.000 30.020.010.00.0600 650^700^750^800^850Reaction Temperature (C)900CH4^ 02% Conversion^••CH4/02 = 4% SelectivityCOx—0-- C2+Figure 4.14Figure 4.15Selectivity as a Function of Temperature forSamarium Oxide65.060.0z,55.0f50.0N 45.040.035.030.0600^650^700^750^800^850^900Reaction Temperature (C)732.0600^650^700^750^800^850Reaction Temperature (C)900Yield as a Function of Temperature for Samarium OxideCH4/02 =414.012.010.005.- 8.00COx—0— C2+%Yields6.04.0Figure 4.1674test period approximately 0.6 g of carbon would be deposited. This amount is 12 timesthe 50 mg initial mass of the catalyst and would be readily visible at completion of thetesting. This amount is not consistent with the small amount of carbon depositionwhich was observed. If the carbon loss was due to adsorption of carbon dioxide onthe catalyst, the weight of carbon dioxide adsorbed would be 2.3 g, or 45 times theoriginal weight of catalyst. This also does not seem reasonable. Although all fittingswere tested regularly for gas leaks, it is possible that a small leak in the system maybe responsible for the loss of carbon. There were many junctions of various mediatubing, such as glass to stainless steel, glass to plastic tubing, and plastic tubing tostainless steel. The other alternatives include errors in the mass flow controllers,bubble flow meter accuracy, and GC calibration.4.2.4 Variation in ResultsIn order to determine the significance of the results for comparison of the catalysts, twoapproaches were used. The first approach included determination of the variation inGC results, based on the calibration procedure. The second approach involved runningtwo identical tests using samarium oxide as the catalyst.The GC calibration was typically carried out by on-line injection of the calibration gasand analysis of the sample three or four times. This was carried out each day prior toa catalyst run. The results from each of three days, selected at random, were analyzedfor standard deviation, and the standard deviation for each component was averagedover the three days. These values are presented in Table 4.4.75The average percent standard deviation is an indication of the expected percentageerror in the concentration, and is not the absolute deviation. The calibrationconcentrations are presented in Table 4.4 with the average absolute standard deviationfor each component.Table 4.4Average Percent Standard Deviation in GC CalibrationComponent TCD(%)FID(%)Calibration Gas %Concentrationcarbon dioxide 2.7 - 8.01 +/- 0.22carbon monoxide 1.4 - 2.01 +/- 0.03methane - 0.9 10.00 +/- 0.09ethane - 0.3 2.00 +/- 0.01ethylene - 0.8 2.02 +/- 0.02propane - 0.9 0.10 +/- 0.00oxygen 4.2 - -The average percent standard deviation varied by less than 1% for all of thehydrocarbons. The largest variation was associated with the oxygen concentration,which was calibrated in a different manner than the other gases. These results indicatethat the GC results do not vary significantly for the calibration gas. The actualconcentration of the components in the calibration gas may vary from that specifiedand may introduce a systematic error. This error should not affect the comparison ofthe results.The variation in the results between catalyst runs for the same catalyst was determinedby 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 under76the same conditions establishes a level of confidence for comparing runs underdifferent conditions.Table 4.5Variation in Results for Two Identical RunsParameter Sm203 #1(%)Sm203 #2(%)Percent DifferenceCH4 conversion 6.34 6.36 0.4CO. selectivity 34.4 32.5 5.4C2+ selectivity 65.6 67.5 2.8CO. yield 2.2 2.1 5.0C2, yield 4.2 4.3 3.04.3 Alkali and Alkaline Earth Doped Samarium Oxide CatalystsSeveral doped Sm203 catalysts were prepared and tested in the reactor. These includedsamarium 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 wereall prepared both in oxygen and air atmospheres, with the exception of potassium,which was prepared in oxygen only. Calcium and sodium doped catalysts were alsoprepared at dopant to samarium mole ratios of 1:10 in an oxygen atmosphere.The catalysts were prepared in two different oxygen containing atmospheres todetermine if the concentration of oxygen available during formation of the catalystcrystals would have any effect. Some effect was noticed during testing of the catalystsand the catalysts prepared in oxygen were generally superior. The catalysts prepared77in 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 4.6Doped Samarium Oxide CatalystsCatalyst Preparation Atmosphere Dopant:Sm mole ratioSm203 Oxygen 1:100Ca/Sm2O3 Air 1:100Ca/Sm2O3 Oxygen 1:100Mg/Sm203 Air 1:100Mg/ Sm203 Oxygen 1:100Na/Sm2O3 Air 1:100Na/Sm2O3 Oxygen 1:100K/Sm203 Oxygen 1:100Ca/Sm2O3 Oxygen 10:100Na/Sm2O3 Oxygen 10:1004.3.1 Results of Catalyst Performance TestsThe reaction conditions are described in Section 3.7. The complete results are presentedin Tables A.2 to A.10. The results are summarized in Table 4.7 for CH 4/02 = 4, andare compared for all catalysts prepared in oxygen in Figures 4.17 to 4.19. There aresignificant differences between catalyst performance for different doping agents,different dopant concentrations, and for different reaction temperatures. Theperformance of the sodium doped catalysts is particularly temperature dependent.There are some significant differences in catalyst performance based on the C2+ yieldsof the catalysts at each of the reaction temperatures. In this discussion, the doped78catalysts will be referred to by their dopant symbol (e.g., 1:100 Na:Sm mole ratiocatalyst will be identified by Na where this does not cause confusion), and theundoped catalyst will be referred to by Sm. Unless otherwise specified, the dopedcatalysts are at 1:100 dopant:Sm mole ratio (e.g., the 1:10 Na:Sm mole ratio catalyst willbe identified by Na (1:10)). The performance of the catalysts at 750°C, as defined byC2+ yield, is in the order Ca (1:10) = Na > Ca > K > Sm = Mg > Na (1:10).Table 4.7Results for all Catalysts, CH4/02 = 4Catalyst% CH4 Conversion % C2+ Selectivity % C2+ YieldTemperature (°C)650^750^850Temperature (°C)650^750^850Temperature (°C)650^750^850Sm 13.2 23.0 25.0 40.2 54.3 50.6 5.3 12.5 12.6Ca/Sm 15.0 24.6 25.4 38.2 53.5 51.4 5.7 13.1 13.1Mg/Sm 13.7 23.2 25.4 40.3 53.6 51.0 5.5 12.4 12.9Na/Sm 10.0 25.4 25.9 32.6 55.8 51.9 3.3 14.2 13.4K/Sm 11.7 25.4 26.2 30.0 50.5 48.3 3.5 12.8 12.7Ca/SmA 14.3 22.5 23.9 32.1 48.4 48.0 5.7 13.1 13.1Mg/SmA 13.3 23.2 24.3 39.8 53.6 49.7 5.3 12.4 12.1Na/SmA 10.2 26.3 27.9 35.5 51.1 46.1 3.6 13.4 12.9Ca (1:10) 9.8 25.0 26.8 34.1 57.0 54.7 3.3 14.3 14.6Na (1:10) 4.6 18.4 25.4 28.6 62.8 49.9 1.3 11.6 12.7A: Prepared in air.79237176;: 45:Wn 41/Mes,m7Y. .MtS, 746M"V.,r,. ;iniefieNNWFWErtt;Mg /Sm% C2+ Selectivity5 23 8 6$$ 6SLO Ca/Sm1:10Ca :Sm ^Na/Sm1:10Na:SmK/Sm:::: :::::::::::: :::::"^ •":.^ •0 0 0Sm•• • •P•,:aKM:..A.••••Ca/Sm1:10Ca:Sm •^Mg/Sm • ••:•.:4^•Na/SmWagsAvvry1:10Na:SmK/Sm■^art•''',A*Dwr• • kt•^ •.;, :.§gf•P% C2 YieldP N A O 90 O N)O oo o o b b obC)a.OOThe following observations can be made regarding the doped catalyst performance ascompared to the undoped Sm203 .Calcium The methane conversion was higher at all temperatures for the (1:100) calcium dopedcatalyst (24.6% compared to 23.0% for Sm 203 at 750°C), but there was no noticeableeffect on C2+ selectivity. The C2+ yield was higher at all temperatures (13.1% comparedto 12.5% for Sm203 at 750°C). The 1:10 doped catalyst had more effect, with a highermethane 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% forSm203). The C2+ selectivity was also higher than that of the undoped samarium oxideat 750°C (57.0% compared to 54.3%) and 850°C (54.7% compared to 50.6%), but waslower at 650°C (34.1% compared to 40.2%). The C2+ yield was significantly lower at650°C (3.3% compared to 5.3%), but in this regard calcium doping gave the best catalystat 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 tothose used by some researchers. For example, Korf et a/. (1989) used 30 mole %calcium doped samarium oxide and reported a significant effect. This reported effectwas an increase in absolute C2 yield of 1% at 780°C (or ca. 10% relative increase inyield), and was less than the increase in absolute C2 yield of 1.8% (or ca. 14% relativeincrease) obtained in this study for the Ca (1:10) at 750°C.82Magnesium 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, whichis in accordance with another study in which it was observed that the shape of thetemperature-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 effectmuch more noticeable for the 1:10 catalyst (methane conversion = 4.6%). However, at750°C the 1:100 doped catalyst produced a higher methane conversion (25.4%) than theundoped (23.0%), while the 1:10 was still much lower (18.4%). At 850°C, both Nadoped 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 dependentbehaviour. 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 beinglower 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 thesame trends as that of the methane conversion and C2+ selectivity. The undoped83samarium oxide (5.3%) gave a significantly higher yield than the doped catalysts at650°C (1:100, 3.3%; 1:10, 1.3%). At 750°C the 1:100 sodium doped catalyst gave asignificantly higher yield than the undoped catalyst (14.2% compared to 12.5%), withthe 1:10 sodium doped giving a lower yield (11.6%). At 850°C, the 1:100 doped catalystgave a higher yield than the undoped catalyst (13.4% compared to 12.6%), and the 1:10doped catalyst produced about the same yield as the samarium oxide (12.7%).Potassium The potassium doped catalyst also exhibited temperature dependent behaviour, withthe methane conversion lower at 650°C (11.7% compared to 13.2% for Sm 203), buthigher at 750°C (25.4% compared to 23.0%) and 850°C (26.2% compared to 25.0%). TheC2+ selectivity was lower for all temperatures (50.5% compared to 54.3% for Sm 203 at750°C). The C2+ yield was considerably lower at 650°C (3.5% compared to 5.3%), withthat at 750°C slightly higher (12.8% compared to 12.5%) and that at 850°C the same asthe Sm203 .4.4 Modification of Catalyst Preparation ProcedureThe preparation of the catalysts involves a temperature hold at 325°C. The process isfully described in Section 3.6. In order to analyze the processes occurring duringcatalyst preparation, the decomposition temperatures of the samarium nitrate and thedopant nitrates were determined. This was accomplished by observation of the purenitrates during a TGA run operated at a constant rate of temperature increase. Thesevalues are presented in Table 4.8.84Table 4.8Melting and Decomposition Temperatures of Nitrates used in CatalystPreparationNitrate Melting Point(°C)Decomposition Range(°C)Samarium 85 290Calcium 44 500-600Potassium 335 >600Sodium 300 715These values indicate that the samarium nitrate decomposed at a much lowertemperature than the dopant nitrates. Therefore, the samarium oxide is probablyalmost completely formed during the 325°C hold, long before the dopant nitrate evenstarts to decompose. The dopant may therefore not be dispersed uniformly throughoutthe Sm203 crystal lattice.An increase in the concentration of catalyst defects, brought about by the improvedinclusion of the dopant cations in the samarium oxide crystal, may result in increasedcatalytic activity. In general, the slower the crystal growth, the fewer the number ofdefects included. Therefore, decreasing the catalyst preparation time may increase thenumber of defects.In order to test the effect of the catalyst preparation time, a modified catalystpreparation program was tested with the calcium dopant. This program omitted thehold midway, and the catalyst temperature was raised from 25°C to 600°C at a rate of25°C per minute. The catalyst was then held at 600°C for 30 minutes. It was hopedthat the samarium oxide would not have time to form prior to the calcium nitrate85decomposition. The resulting catalyst would then have a maximum number of dopantcations incorporated into the catalyst crystal.4.4.1 Results of Performance Test for Modified CatalystThe calcium doped samarium oxide catalyst, prepared at a 10:100 mole ratio of calciumto samarium, and prepared using a revised catalyst preparation program, was testedaccording to the conditions presented in Table 3.2. The results obtained are presentedin Table A.11 and Figures 4.20 to 4.23. The catalyst preparation had a significant effecton the catalyst performance. The catalyst was designated 1:10 Ca (RP) (for revisedprogram).Oxygen ConversionThe oxygen conversion was significantly higher for this catalyst, and reached 100% at750°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 lowmethane 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 selectivity86Effect of Catalyst Preparation on % Oxygen Conversionc0rDc0UNat100.090.080.070.060.050.040.0Run 1 Run 2 Run 3 Run 4 Run 5 Run 6^01: 1 0 Cc..----------.1:10 Ca (RP .-c0.E.°0Cu.,000c0.c02ae28.026.024.022.020.018.016.014.012.010.08.0Run 1 Run 2 Run 3 Run 4 Run 5 Run 6•Li1:10 Ca^100 02 Conyers' nFigure 4.20Figure 4.21Effect of Catalyst Preparation on % Methane Conversion87showed a major increase between 650°C and 750°C. However, for this catalyst, the C2selectivity only increased by a small amount over the same temperature range, andremained low at 850°C. This is due to complete oxygen consumption occurring at750°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 COX yield was higher at all conditions than for the other catalysts tested.This catalyst was a very effective combustion catalyst at 650°C. The observed effectsmay be due to a large concentration of active sites. According to McCarty (1991), toohigh a concentration of active sites can result in an effective combustion catalyst (seeSection 2.3.4). However, the actual cause of the observed effects can only be speculatedupon, due to the reactor configuration used for the tests. Another type of reactor maybe more appropriate for interpretation of the results. For example, the use of amembrane reactor, as described in Section 2.6, may facilitate investigative testing.Where the methane and oxygen are separated by the catalyst, an increase in oxygendiffusivity and/or active sites, should result in an increased C2 yield with no negativeeffects observed for a too active catalyst. If, on the other hand, the increased methaneconversion 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.88Effect of Catalyst Preparation on % C2 Selectivity70.065.060.055.050.0ig 45.04o.oCNI0 35.030.025.020.015.0Run 1 Run 21:10 Ca (RP)1:10 CaRun 3100% 02Run 4ConversionRun 5 Run 610.0100 % 02 Conversion4.0Run 1 Run 2 Run 3 Run 4 Run 5 Run 614.012.01:10 C2.0 I^Figure 4.22Figure 4.23Effect of Catalyst Preparation on % C2 Yield894.5^Scanning Electron Microscopy-Electron Dispersive X-RayThe scanning electron microscopy-electron dispersive x-ray (SEM-EDX) technique wasused to examine the surface of the catalyst particles and to measure dopantconcentrations at and near the surface. The instrument used was a Kevex 8000, witha 40 A diameter beam of electrons.The SEM photographs are presented in Appendix B. These photographs representsome 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 mostsamples. 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 othersamples, the structure revealed in photograph 103, a spheroid type surface, was notfound in any of the other samples. This does not necessarily indicate that this structureis unique to the potassium doped sample, only that it was not found in the othersduring the scanning of a limited number of particles for each. The numbers on thephotographs indicate some of the spots which were scanned with the EDX forcomponent analysis.Overall, the particle surfaces consisted of irregular-shaped and irregular-sized plate-likecrystals, with primarily very rough surfaces; although small, smooth surface portionswere also found. Based on the EDX analysis, the dopants do not appear to beuniformly distributed on the surface. Hence, it may be assumed that cation dopants90were non-uniformly dispersed throughout the Sm 203 crystal. However, no crystalscontaining large concentrations of dopant which were not incorporated into thesamarium oxide crystal structure were observed. One generalization which can bemade is that the very smooth surfaces appear to contain only samarium oxide. Thissuggests that one of the functions of the dopant is to increase the roughness of thesurface. Unfortunately, a large amount of error is introduced into the EDX analysis forrough surfaces, such as are present in our samples. Therefore, the EDX analyses areapproximate, and should be used as indicators of amounts of elements present, ratherthan as giving absolute numbers. The EDX does not detect elements of atomic numberless than that of sodium; therefore, elements such as carbon and oxygen are notdetected. The weight percent determined is the weight percent calculated consideringonly the elements with the atomic number of sodium or greater. The relative moleratio of dopant to samarium can be calculated and compared to the nominal value forthe bulk crystal (i.e., 1:100 or 1:10).Calcium 1:100 Ca:Sm Photograph 192The 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 arepresented in Table 4.9. The presence of neodymium at spot 3 is likely a result oforiginal contamination associated with the samarium.91Table 4.9Weight % of Catalyst Surface Components According to SEM-EDXElement1:100 Ca:Sm Bulk Mole Ratio 1:10 Ca:Sm Bulk Mole Ratio192-1 192-2 192-3 193-1 193-2Sm 75.44 71.61 78.35 82.84 82.81Ca 0.62 0.11 .83 1.46 2.84Si 2.01 1.83 0.60 nd naAl 0.52 6.82 1.08 1.01 naNd na na 3.99 nd ndCa:Sm localmole ratio2.4:100 0.6:100 4.0:100 0.7:10 1.3:10nd: not detectedna: not analyzed1:10 Ca:Sm Photograph 193, 194Two different particles were photographed for this catalyst to show some of thedifferent structures present. Photograph 193 is of a sample which consists offragmented flat plate-like crystals, with many viewed on end. Photograph 194 showsa smooth flat surface with a few small crystal fragments present. The analysis forphotograph 193 is presented in Table 4.9.1:10 Ca:Sm RP^Photograph 105This sample appeared very different from the others. While still containing some ofthe plate-like crystals, much of the crystal structure was much smaller and had afeathery appearance. Spots 1 and 2 were both scanned and analyzed. The Ca peak forspot 2 was much larger than for spot 1. The analysis determined 1.36 weight % Ca forspot 2 compared to 0.77 weight % for spot 1, which correspond to Ca:Sm mole ratios92of 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 thatit was obscured by one of the Sm 203 peaks.1:100 Na:Sm Photograph 195This area had a very rough surface and consisted of many broken plate-like crystals ofvarying sizes. A second area was scanned (no photograph) which was very differentfrom the first, consisting of a very smooth plane with a few very small particles.1:10 Na:SmA sample was scanned and analyzed; however, no photograph was taken and nosodium was detected.Samarium Oxide Photograph 102The sample photographed consisted of large flat plates which were broken in irregularpatterns at the edges. The area in the centre of the photograph was scanned by EDXand the trace revealed almost pure Sm 203, with small aluminum and silicon peaks.Potassium Photograph 103A small potassium peak below the method detection limit was observed on the EDXtrace. 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.93These formations were observed on only one particle, other particles being similar toother samples, such as that observed in photograph 192.Magnesium Photograph 104Photograph 104 shows a good example of needle- and plate-like crystals. This samplewas not analyzed. In another sample which was analyzed, the scanned area was a flatplat in an area which was very similar to photograph 193. A small Mg peak wasdetected, which was calculated to be 4.5 weight% Mg, corresponding to a very highMg: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 Caprepared by the revised method were examined using powder x-ray diffraction (XRD),on a Siemens, D5000 Diffractometer. In addition, a sample of monoclinic Sm203, whichwas 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 samplesproduced a "noisy" trace, possibly indicating that, although the sample was crystalline,the crystals contained numerous defects or dislocations. The noise of the trace maymask small peaks, making detection of small changes in the crystal structure difficult.No new peaks were detected for the samples as compared to Sm203, although some ofthe peaks were shifted, indicating a change in distance between atoms in the latticeunit. The monoclinic and cubic traces differed considerably. The monoclinic trace wasconsiderably less "noisy", which may be due to fewer defects and dislocations present94in the crystal, as well as larger crystallites, which is consistent with the longpreparation time and the crystal structure (Anshits et al., 1990).The three main peaks in the cubic Sm203 crystal correspond to spacing between crystallattice 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 todetect small shifts in the peaks. There does not appear to be any significant change inthe 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 intensepeaks compared to that of the Sm 203, which indicates that some of the crystaldimensions were increased. The distance between lattice planes, d, increased from 3.16A 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 to2.86 A for plane (4,0,0). These increases in the crystal dimensions can be used tocalculate the lattice constant for the crystal and to give an indication of the crystalstructure. The lattice constant a increased from 10.9 A for the cubic Sm203, to 11.4 Afor 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 sensitivityof the method. This indicates that the Na and Ca dopant cations were likelydistributed throughout the Sm 203 crystal structure. No evidence of the monocliniccrystal structure was seen in the cubic catalysts.954.7 Measurement and Effect of Catalyst Surface AreaSurface area may be an important variable for the catalytic oxidative coupling ofmethane, the effects of which have not been conclusively determined. Some studiesreported in the literature suggest that an increase in surface area is detrimental to theselectivity, while others show the opposite effect. Most studies are not carried out withall other variables held constant, hence the difficulty in interpreting the results.In a review of catalyst morphology for the oxidative coupling of methane by Martinand Mirodatos (1992), considerable evidence is presented that, for a variety of catalysts,the oxidative coupling of methane reaction is structure sensitive. The catalyst whichhas been studied the most and which is discussed in greatest detail is MgO. In a studywhich examined metal doped MgO samples, the specific surface area decreased from70 m2/g for unpromoted MgO down to 1 m2/g for a sodium content of 20 mole%. Theselectivity 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 resultsin sintering (loss of area) and an improvement in C2 selectivity. It was speculated thatthe specific rate (per unit weight) of methyl radical formation increases with the surfacearea, leading to a higher specific rate of reaction. However, the rate of collision ofradicals with the surface is increased. This would lead to total oxidation into CO2, thusdecreasing the selectivity into C2 hydrocarbons.966In an extension of this study it was determined that for compounds with surface areabelow 100 m2/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 of2 m2/g gave the highest C2 yield, at 750°C to 800°C. Although both the total C2 yieldand CO2 yield had a strong dependence on surface area, the ratio of ethylene to ethanechanged only slightly at low surface areas, and not at all at surface areas greater than100 m2/g, showing that the reaction of ethane to ethylene is not very strongly relatedto surface area.In the above studies, the composition of the catalyst was varied to achieve the changein surface area and the observed results may also be due to other effects of the alkalidopants.In the aforementioned study, Iwamatsu et al. also determined the effect of surface areadue to calcination. When the MgO sample was sintered by calcination, the C2 yieldincreased 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 devisedin order to explain the apparent detrimental effect of surface area:1CH4 ^ CO297Steps 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 proceedfaster with increased surface area.Parida and Rao (1991) prepared 15 atom % Li/MgO by various methods, resulting ina variation in surface area while the composition was held constant. The C2 yield andselectivity were found to increase with a decrease in surface area.The above studies indicate that increased surface area is detrimental to C2 selectivityand yield; however, other studies have shown that this is not always the case. Forexample, among three Li/MgO catalysts prepared by different techniques, the mostselective 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 28m2/g by sintering at 750°C in the reaction mixture (Martin and Mirodatos, 1992), theC2 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 structureand catalytic properties (Martin and Mirodatos, 1992). One sample was prepared fromthe decomposition of basic carbonate and another was prepared by the oxidation ofmetal ribbons in air. The catalyst prepared from the carbonate gave a higher selectivityto C2 hydrocarbons (typically 60-70%) than that from the metal (about 30%). Analysisof the results based on unit surface area revealed that the total activity of the latter istwice that of the former and the intrinsic activity for the production of C 2's is threetimes larger. The researchers concluded that the MgO produced from the basic98carbonate had a higher relative concentration of active sites for the oxidative couplingof methane reaction. The catalyst with the higher surface area displayed the higherselectivity for C2 hydrocarbons; therefore, the variation in selectivity can not be due toa possible detrimental effect of surface area.Another study to examine the effect of surface area concentrated on samarium andlanthanum oxides (Korf et al., 1990b). Samarium oxide undergoes a phase change fromthe metastable cubic structure to the stable monoclinic structure at about 900°C.Lanthanum undergoes no such phase change. The surface area of samarium oxide wasmeasured at 6.6 m 2/g for cubic and 2.7 m2/g for monoclinic. The specific activity toC2 products based on the surface area of the cubic phase is much higher than that ofthe monoclinic phase. The use of the same amount of surface area of the cubic and themonoclinic material led to a lower C2 production in the case of the latter. This showsthat the surface area is not a major factor in the change in behaviour during the phasetransition from cubic to monoclinic.The lanthanum oxide was calcined at different temperatures to obtain different surfaceareas. A higher calcination temperature resulted in a lower surface area. Thosesamples calcined at low temperatures, and therefore having high surface areas,achieved relatively high C2 selectivity and C2 yield compared with the results for thehigh calcination temperature. The specific activity to C2 products based on the surfacearea was the same for these catalysts. When the same amount of surface area was usedin the reactor the C2 production was the same. In this case, sintering causeddeactivation of the La203 catalysts.99The above results may also be affected by variables other than surface area. Forexample, the heat treatment and the variation in preparation methods may causechanges in the catalyst surface and bulk morphology.From the above studies it appears that the effects of surface area on the selectivity andactivity of the catalyst are still not fully understood. This stems from the difficulty inchanging only the surface area of the catalyst without changing other morphologic orchemical properties of the catalyst. The use of dopants to cause sintering in catalystsresults in improved catalysts; however, this is likely primarily due to factors other thanthe decreased surface area, such as increased basicity, increased oxygen diffusivity dueto incorporation of defects in the crystal, etc.The surface areas of the catalysts used in the present study were determined by thesingle point BET method (using nitrogen gas at liquid nitrogen temperature) for theundoped samarium oxide and the 1:100 mole ratio doped catalysts. There wasinsufficient sample to determine the surface area for the 1:10 doped catalysts. Thesurface areas are presented in Table 4.10.The C2+ selectivities for all catalysts were graphed as a function of surface area todetermine if a correlation existed (see Figure 4.24). The existence of a correlation ateach temperature was tested by carrying out a population correlation calculation, whichreturns the covariance of two data sets divided by the product of their standarddeviations. 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.100The calculated parameter for the C2+ selectivity was calculated to be -0.46, 0.12, and0.18 at 650°C, 750°C, and 850°C, respectively. Therefore, it is clear that at 750°C and850°C, no correlation exists between surface area and C2+ selectivity. However, at650°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 thesedata, is that surface area over the range tested (ca. 2 to 3 m2/g) does not have asignificant effect on C2+ selectivity. The methane conversion was also analyzed forcorrelation with surface area, and none was found to exist. These results are notsurprising, based on the previous discussion, which presented several studies withconflicting results.Table 4.10Surface Area of CatalystsCatalyst Surface Area (m2/g)Sm 2.5Ca/Sm (Oxygen) 2.4Mg/Sm (Oxygen) 2.0Na/Sm (Oxygen) 2.6K/Sm (Oxygen) 3.1Ca/Sm (Air) 2.0Mg/Sm (Air) 2.2Na/Sm (Air) 2.1101ReactionTemperature■ 650 C0 750 C• 850 CFigure 4.24C2+ Selectivity as a Function of Surface Area60.055.0Z•. 50.00Ci)( 45.0+CN10 40.0035.030.0o^• ^^ ^•o•• n•••■•••■ ■1.8^2^2.2^2.4^2.6^2.8^3^3.2Surface Area (m2/g)1024.8^Basicity of the CatalystCatalyst basicity has been shown by several studies in the literature to be an importantvariable in catalyst performance for the production of C2 hydrocarbons. In general itis found that the catalyst selectivity and yield increase with an increase in basicity.The metal oxide catalysts, monoclinic (m) Sm203, cubic (c) Sm203, MgO and y-Al203,were studied for the oxidative coupling of methane by Lapszewicz and Jiang (1992).The relative basicities of the compounds were determined by CO 2 temperatureprogrammed desorption profiles (based on the temperature at which 50% of CO 2desorbed), and decreased in the order m-Sm 203 > c-Sm203 > MgO > y-Al 203, which isin the same order of decrease in C2 selectivity. The presence of multiple maxima atdifferent temperatures on the desorption curves for all catalysts except y-Al 203 indicatesthe existence of several types of basic sites of different strength.The doping of stable oxides by alkali compounds may result in an increase in basicityand an increase in C2 selectivity. The basicity was determined for various catalysts of15 atom % Li/MgO prepared by various methods (Parida and Rao, 1991). It was foundthat the C2 yield increased with an increase in basicity, while the CO. formationdecreased.In a study of MgO/CaO catalysts (Philipp et al., 1992), over a wide range of MgOconcentrations, the 85% MgO/CaO catalyst was determined to have the highestbasicity, as well as being the most active and selective catalyst.103Baerns (1992) established the pK A of various catalyst oxides, where the pK A is ameasure of the acid-base properties of the material. Table 4.11 illustrates how the C2+selectivity is dependent on the value of pKA (pKA=7 is neutral, >7 is basic, and <7 isacidic). The results of this study give a clear indication of the positive relationship thatis often found to exist between basicity and C2+ selectivity.Table 4.11Selectivity and Acid-Base Properties of Various Catalytic Oxides 12Catalytic Oxide(weight %)% C2+ Selectivity pKA1.5% Li20/Ca0 73 +12.23.0% Na20/Ca0 67 +9.34.5% K20/Ca0 67 +9.3SiO2, kieselgel 15 +6.8y-Al203 3 +3.3/+4.0Al203 x SiO2 1 -5.6Baerns reviewed various studies which were carried out in order to investigate theeffect of acidity and basicity. It was found that the addition of alkali in the form ofsodium oxide to lead oxide deposited on strongly acidic supports, such asPbO/Al2O3SiO2, resulted in an increase in C2+ selectivity. This was interpreted as theneutralization of strong acidic sites by the alkali. An initial increase in methaneconversion was also obtained which was reversed by additional sodium oxide;however, this was more than compensated for by the gain in selectivity.12Baerns, 1992104Increased lead oxide loading resulted in a decrease in the catalyst surface acidity, witha corresponding increase in the C2+ selectivity. Although the results were presentedfor supported lead oxide catalysts only, the following systems have also been studiedby various authors and similar results reported: manganese oxide, nickel oxide,praseodymium oxide, and titanium oxide. Baerns acknowledges that the effect ofbasicity may not be the only variable affecting the C2+ selectivity and the contributionof various factors has not been elucidated.The relative selectivities as a function of basicity of the alkaline earth oxides were alsoinvestigated (Baerns, 1992). For undoped alkaline earth oxides, the C2+ and C3+selectivity were lowest for the least basic compound, BeO, and highest for the mostbasic 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 andloss of surface area by sintering). They conclude that this indicates that the 0' anionplays an important role in the selective oxidative coupling of methane reaction.The effect of various anions was investigated by testing the following calciumcompounds: oxide, sulphate, silicate, aluminate, fluoride, and phosphate (Baerns, 1992).A significantly higher selectivity was obtained for CaO and CaSO 4. The high selectivityobtained with the use of CaSO 4 as the catalyst may be due to at least somedecomposition 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 thereaction with water vapour which results in partly liquid hydroxides. However, highly105basic compounds can be prepared by the incorporation of alkali compounds into thesurface 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 finalvalue is reached. The increase in C2+ selectivity was not attributed to the alkalicompound itself but to its interaction with CaO.A catalyst of composition CaO/La203 was prepared which resulted in a basicity for thecompound 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 maximumselectivity was not found to coincide with the maximum basicity. It was concluded bythe investigators that there existed a slight dependency of selectivity on the number ofacidic sites. However, the maximum in the acidity is shifted to higher lanthanumconcentrations than that of C2+ selectivity. The tentative conclusion reached was thatnot 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 thepresence of negatively charged oxygen ions (o- or 02). Although it is generallyaccepted that the oxidative coupling reaction is initiated by the formation of methylradicals on the catalyst surface, with the C-H bond cleavage supposed to be the ratelimiting step, the actual mechanism involved is not clear. The relationship betweenbasicity and reaction mechanisms has been investigated by several researchers.Dubois (1992) suggests that high (Lewis) basicity is an important factor in the oxidativecoupling of methane, due to the fact that basic oxides which possess defect structures106and no irreducible cations have very few mechanisms available for the re-equilibrationof bulk and surface charge imbalance. The charge density can be decreased byliberating an electron (from 0 2-) to an anion defect site (oxygen vacancy). The newoxygen species (0) can be viewed as a bulk oxygen possessing a positive charge, ascompared to the 02.  The released electron, trapped in an anion vacancy, servesas the oxygen adsorption site on the catalyst surface. Therefore, the basicity of the bulkoxide is important in the expulsion of single electrons by oxygen (p-typesemiconductivity) which is directly related to the capture of gas phase oxygen, asurface phenomenon.The addition of alkali dopants may lead to an increase in basicity due to the dispersionof 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 leadto a shift in the electrical charge of the catalyst surface. The presence of the 0 ionswould also result in high surface basicity.Lapszewicz and Jiang (1992) carried out tests using deuterium and oxygen isotopes toinvestigate the mechanism of methyl radical formation. They summarized the varioushypotheses concerning the active sites into two concepts.The first concept involves an acid-base type reaction, in which the adsorbed methanedissociates on the surface of the basic oxide, yielding W and CH3; subsequentsubtraction of the electron from the unstable methyl anion leads to methyl radicalformation. For this concept, the rate of methane conversion in the coupling reaction107can be expected to correlate with the catalyst's ability to effect heterolytic scission ofthe C-H bond, as measured by the rates of deuterium exchange.The second concept involves the formation of a transition complex consisting of theadsorbed methane molecule and the surface oxygen, which subsequently dissociatesto yield a methyl radical. The rate of methane conversion would be expected tocorrelate with the ability of the catalyst to activate oxygen, as measured by the rate ofthe 0-0 bond scission.Neither of the above two anticipated correlations existed, indicating that the mechanismof methyl radical formation can not be explained by the ability of the catalyst surfaceto dissociate either methane or oxygen, or both.The investigators suggest that the activation of methane can be expected to occur onelectron-rich (basic) sites, while the oxygen would be activated on electron-deficient(acidic) sites, and thus the relationship between basicity and C2 selectivity can beinterpreted 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. Theirexperiments with deuterium and oxygen isotopes support this hypothesis; that is, theratio of the methane activation to oxygen activation increases with basicity and withC2 selectivity.Baerns (1992) suggests that the negative effect of surface acidity, with the resultingcomplete oxidation, may be due to the following possible explanation for non-selective108oxidation: the Bronsted- or Lewis-type acidic sites cause the formation of a carboniumion, which is easily attacked by a negatively charged adsorbed oxygen species (e.g., aor 02) leading to total combustion.Baerns and Becker (1991) interpret their results for La203/Ca0 as evidence of acooperative effect involving acidic sites and basic 0' sites. They suggested thefollowing 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 thentransferred, and the CH3• species formed may either recombine on the surface, ifclosely located to each other, or they may desorb. The CH 3+ ion may interact withnegatively charged oxygen existing as a or 02 on the surface, leading to nonselectiveoxidation. 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, andthat high basicity results in high C2+ selectivity. This may be due to either suppressionof the nonselective pathways and/or promotion of the selective routes. Interaction ofnegatively charged oxygen either with a protonated methane molecule or CH3+ species,or the reaction between radical-like oxygen and methyl species, may lead to totaloxidation.109Table 4.12Electronegativity of Alkali and Alkaline Earth Elements 13Group IA Electro-negativityGroup HA Electro-negativityLi 1.0 Be 1.5theoreticalincrease inbasicityNa 0.9 Mg 1.2K 0.8 Ca 1.0Rb 0.8 Sr 1.0Cs 0.7 Ba 0.9—} theoretical decrease in basicityBased on the above studies, it can be concluded that, although the mechanism is notfully 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 anindication of relative basicities. These methods will sometimes produce differentrelative orders of basicity for different methods. One indication of acid/base strengthin a metal oxide is the difference in electronegativity between the cation and theoxygen (Dubois and Cameron, 1990). An oxide having a cation with a lowelectronegativity will have a high partial charge on oxygen, and therefore will bestrongly basic. The electronegativities of group IA and HA cations are presented inTable 4.12.For a binary oxide, the electronegativity x can be calculated as an average of the two13 Brady and Humiston, 1982 (Originally from Linus Pauling, The Nature of theChemical Bond).110metal 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, andXoxide = 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 cationpresent in the metal, equation (4.3) can be used to estimate the electronegativity of thedoped 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 valencenumber, for similar electronegativities. Therefore, substitution of Sin' with alkali oralkaline earth elements will increase the basicity.The electronegativity can be taken as a first approximation of bulk basicity. It shouldbe noted that the actual distribution of dopant is not known, and that the relationshipbetween the bulk basicity and surface basicity is not clear. However, it follows that acatalyst 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 catalystbasicity.111According to equation (4.3), the catalysts in the present study should increase inelectronegativity in the order Na (1:10) < Ca (1:10) < K < Na < Ca < Mg < Sm, witha corresponding decrease in basicity.Table 4.13Effect of Electronegativity on Catalyst PerformanceCatalyst Na(1:10)Ca(1:10)K Na Ca Mg SmElectro-negativity*4.42 4.53 4.65 4.65 4.66 4.67 4.68% MethaneConversion+18.4 25 25.4 25.4 24.6 23.2 23C2+ Selectivity 62.8 57.0 50.5 55.8 53.5 53.6 54.3* as calculated by equation (4.3).TReadion 750°C, CH4/02=4The methane conversion and C2+ selectivity as functions of bulk electronegativity arepresented in Figures 4.25 and 4.26 (TReadion = 750°C, CH4/02 = 4). The correlation wascalculated 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 functionof electronegativity, which is not significant at a 5% uncertainty level. It should benoted that the correlation coefficient is a measure of linearity. Review of Figure 4.25indicates that if a relationship exists between methane and cation electronegativity, itmay pass through a maximum. The correlation coefficient would not be an appropriatemeasure of such a relationship. A strong negative correlation of -0.89 was indicatedfor C2+ selectivity, which is significant at a 1% uncertainty level. As observed by otherresearchers, an increase in C2+ selectivity was observed for an increase in basicity.1124.4^4.45^4.5^4.55^4.6Cation Electronegativity4.65 4.7Methane Conversion as a Function of CationElectronegativityT= 750C29oc 2725c0o 23o 1920 1715■• ^ ■• ■ ^= 0.69Correlatior•6560.>TD) 55a,N.ct, 50U045404.4T = 750C4.45^4.5^4.55^4.6 4.65^4.7•• ■K^relation = -0.4Cor 6 (without ICFigure 4.25Figure 4.26C2+ selectivity as a Function of Cation ElectronegativityCation Electronegativity113The C2+ selectivity of the potassium doped catalyst was significantly lower than wouldbe expected due to basicity. This may be due to factors other than basicity, such ascarbonate formation or ionic radius. Potassium forms a stable carbonate up totemperatures higher than that of the other dopants used. The effect of carbonateformation is discussed fully in Section 4.9. It has been observed that the formation ofvery 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 thecapability of the oxygen atoms to diffuse through the carbonate. The ionic radius hasalso been identified as an important dopant property (Korf et al., 1988), and this isdiscussed 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 potassiumion may not be incorporated into the samarium oxide crystal lattice, or if it is, it maycreate stress on the lattice, resulting in a destabilization effect and a detrimental effecton catalyst selectivity. It would be reasonable to conclude that a factor other thanbasicity may be affecting the performance of the potassium doped catalyst.If the correlation calculation for C2+ selectivity is carried out without the value forpotassium included, the parameter calculated approaches that of a perfect negativecorrelation (-0.96), which is significant at a 0.1% uncertainty level. This is quite astrong correlation, considering that factors other than basicity, which have not beenheld 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.1144.9 Effect of Carbon Dioxide and Carbonate Formation on Catalyst PerformanceSeveral catalysts used for the oxidative coupling of methane are capable of forminghigh temperature stable carbonates. The amount of carbon dioxide which is producedin the oxidative coupling of methane reaction is more than sufficient to promote thisformation. In addition, it is thought that the presence of CO2, even in the absence ofcarbonate formation, may interact with the catalyst to affect the activity and selectivity.Although these effects have been studied, due to the difficulty in clarifying themechanisms involved, it is not clear what effect carbonates and carbon dioxide haveon the oxidative coupling of methane reaction.The effect of carbon dioxide addition on gas phase reactions was studied by van derWiele et al. (1992) in a blank reactor at 800°C, by substituting part of the helium streamwith CO2 in the feed gases. They found that carbon dioxide had no effect on thehomogeneous reactions. This is in accordance with the fact that the decomposition ofCO2 is very slow.Suzuki et al. (1990) studied the effect of carbon dioxide on the oxidative coupling ofmethane reaction by introducing CO 2 into the feed. The use of carbon dioxide as adiluent in the feed gas resulted in an increase in ethane and ethylene yields, from a C2yield of 13.4% to 18.3%, and an increase in C2 selectivity from 42.3% to 49.4% (MgOcatalyst, TR = 800°C, CH4 /02 = 2). The addition of carbon dioxide also suppressed thedeactivation of the catalyst and decreased the amount of coke deposition. Partialadsorption of carbon dioxide on the surface may inhibit the deep oxidation of methane115or C2 products, resulting in increased C2 selectivities.Several catalysts were tested, including doped and undoped Sm 203 and MgO, withvarying preparation methods. For some catalysts the addition of CO 2 had very littleeffect while for others it was considerable, both on conversion and selectivity. It wasapparent 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 theexception of CaO at higher concentrations of CO 2, and SrO, for which a considerabledecrease was observed.The authors explain the above as due to the formation of carbonates, according to thefollowing 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 carbondioxide are present in the feed gas, SrO would likely react with the carbon dioxide toform 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 very116low. The authors suggest that the addition of a small amount of carbon dioxide maycause competitive adsorption of introduced carbon dioxide and oxygen on the catalystsurface. The secondary oxidation of the C2 or intermediates produced could be partlysuppressed, resulting in an increase in the C2 yield and C2 selectivity. Higherconcentrations of carbon dioxide resulted in carbonates forming over the entire surfaceof the oxide and reduced the catalytic activity. Magnesium carbonate decomposes atthe low temperature of 350°C. Therefore, the surface would not be covered withinactive carbonate species.The investigators report that another effect of CO2 addition may be the oxidation of themethane by the carbon dioxide, resulting in the formation of C2 compounds and carbonmonoxide, according to the following equations:CH4 + CO2 --> 0.5 C2H4 + CO + H20 AG = 34.7 kJ/mole at 800°C^(4.5)CH4 + 3 CO2 ---> 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, carbonmonoxide production (equation (4.6)) is thermochemically more favoured. This issubstantiated by the reaction of methane with carbon dioxide without oxygen overSm203 /MgO, which gives carbon monoxide as the major product. An increase incarbon monoxide yield was observed over various catalysts when helium was replacedby carbon dioxide.Peil et al. (1991a) studied the effect of CO2 on the oxidative coupling of methane, over117Li/MgO and Sm203, using SSITKA (steady-state isotopic transient kinetic analysis).Carbon dioxide addition to the Li/MgO catalyst resulted in a decrease in methane andoxygen conversion, with no change in selectivity, suggesting that all sites are equallyactive for both selective and nonselective oxidation. No significant effect was observedfor the samarium oxide catalyst.The difference between these two catalysts can be explained by the fact that Li/MgOreadily 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 activesites on Li/MgO equally. This may be due to only one type of active site available onthe surface, which may be selective or non-selective, depending on theoxidative/reductive state of the active site.Dubois and Cameron (1991, 1992) studied the effect of stable carbonate formation onthe selectivity of catalysts for the oxidative coupling of methane. They selected severalcatalysts, some of which form stable carbonates and some of which do not. Amongthese, yttria and thoria were selected due to their ability to form stable surfacesuperoxides (02), possibly due to the absence of stable carbonate formation. The useof these two catalysts led to relatively high carbon monoxide selectivity, which may bea result of the aforementioned properties. This indicates that the addition of hightemperature stable alkaline earth carbonate decreases CO synthesis activity whileincreasing overall catalytic activity. The decrease in CO formation is thought to be dueto the destruction of surface superoxide by surface carbonate formation, possiblythrough a peroxycarbonate intermediate.118The ability of the catalyst to form carbonates is generally through the addition ofdoping compounds, such as Li, Na, K, La, Sr, and Ba. Of several catalysts tested, theoxides which do not possess stable carbonates and those which are not substantiallycarbonated at the temperature of the reaction exhibited high CO selectivities. Theconversion 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 thecarbonate 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 theoxide, which subsequently decomposes. The authors suggest that the decompositionof surface carbonates should be an advantage under catalytic operation, due to anincrease in anion vacancy sites, which are required for oxygen adsorption. As well, thedecomposition of bulk forming carbonates would also increase crystal disorder. Thisis in agreement with Korf et al. (1990a), who found that in the Li/MgO system, theactive sites were formed by the decomposition of the Li 2CO3, as discussed in Section2.3.2.The higher activity and selectivity observed for the catalysts which form stablecarbonates is in disagreement with the hypothesis of the superoxide ion as the activeoxygen species for generation of methyl radicals, since the superoxide ion is not foundon the carbonate surface. The role of the carbonate has not been proven, but theauthors suggest it may be related to its ability to alter the relative stability of thevarious oxygen species found on the surface, thereby reducing CO formation.119Kalenik and Wolf (1992) observed low C2 selectivity for zirconium dioxide, which theyconcluded was due to the formation of surface carbonates which decrease the numberof lattice defects and decrease the capability of the oxygen atoms to diffuse through thecarbonate.It is difficult to draw conclusions about the relative merits of carbonate formation. Anadditional complication may be that the more basic oxides are also the ones morecapable of forming high temperature stable carbonates. In summary, the above studiesreveal the following conclusions. Formation of large amounts of carbonates, or verystable carbonates, appears to be detrimental to C2+ selectivity and methane conversion.On the other hand, adsorption of carbon dioxide, or formation of small amounts ofcarbonates may be beneficial to methane conversion and C2+ selectivity. An optimalability to form carbonates may be the key to enhancing C2+ selectivity and yield. Asindicated by Dubois and Cameron, this may be related to formation of the carbonateon 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 materialis incapable of forming stable carbonates.The decomposition temperatures of the carbonates to the oxides for the elements usedin the present study are listed in Table 4.14. The actual amount of carbonatedecomposition will depend on the amount of carbon dioxide present, according to thereverse 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, andat 750°C should be as follows: K > Na > Ca > Mg = Sm. The actual order is Na > Ca120> 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 thanexpected behaviour of potassium may be due to a very strong carbonate formation,which would not decompose at reaction conditions.Table 4.14Decomposition Temperature of Various Carbonates'Carbonate Sm Mg Ca Na KTDecomposition (°C) 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 DopedSamarium Oxide CatalystsTests were carried out using a thermogravimetric analyzer (TGA) in order toinvestigate the relative stabilities of carbonate formed on the calcium and sodiumcatalysts. Undoped samarium oxide was heated in a nitrogen stream up to 750°C, thetemperature was then held constant and 100% CO2 was passed over the sample for 25minutes at 1 atmosphere pressure. The weight of the sample versus time was plottedin Figure 4.27. The shape of the loss in weight over time curve indicates that morethan one type of weight loss occurred, which may be due to loss of surface water andbound water. The weight loss ceased when constant temperature was reached. Asmall steady decrease in weight is noticed after the CO 2 is admitted, indicating that no14CRC Handbook of Chemistry and Physics, 1991121-104402100•50094–—300tfk90 if0.0^20. 0^40.0C60. 0^80. 0 100.0^120. 0^140.0^160. 04 200180. 0 200.0 22 . 010096Figure 4.27 Carbonate Formation on Samarium Oxide Under a 100% CO 2 Atmosphere104– – 1400102 – – 1200– 1000 Z,•– 800 -Pa411CLE600 1–– Iz194–Residua:92–   92.14 X(59. 11 ens)^ – 200–^CO2 z90 -4--^-    00. 0^10. 0^20. 0^30. 0^40. 0^50. 0^60. 0^70. 0^BO. 0^90. 0^100. 0^110. 0Time (min) DuPont. 1090Figure 4.28 Carbonate Formation on a 1:10 Na:Sm Oxide Catalyst under a 100% CO 2AtmosphereTime (min)DuPont. 1090– 400122appreciable carbonate formation occurs for the undoped samarium oxide at 750°C. The1:10 Na:Sm and Ca:Sm catalysts were tested in the TGA for carbonate formation at eachreaction 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°Cand held at that temperature for approximately 30 minutes (see Figure 4.28). Theweight of the sample decreased at a slow rate similar to that observed for the puresamarium oxide. Carbon dioxide was then introduced into the TGA. With theexception of a small increase in weight approximately three minutes after the CO2 wasintroduced, the slow loss in weight continued as before, for about 20 minutes. After20 minutes under the CO 2 atmosphere the weight of the sample started to rapidlyincrease, which may be attributed to carbonate formation. The temperature wasincreased to 750°C, which again resulted in a slow decrease in the weight at the samerate as that which occurred at 650°C. When the temperature was increased to 850°Cthe weight dropped almost instantaneously, likely due to carbonate decomposition.After the period of rapid weight decrease attributed to carbonate decomposition, theweight 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 anddecomposition 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 thebehaviour under a carbon dioxide atmosphere at reaction temperatures (see Figure123104102-1000) 98-/-•/94- -^/ --J-- - - I ---- - -- /^/^FA 1 k :^—,. CO —3-^190 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. 0rime (min)DuPont. 109212496-92-- 1400- 1200- 1000 p0L- 800^-4)04-0a_- 600- 400—2000Figure 4.29 Carbonate Formation on 100% Calcium Oxide under a 100% CO2Atmosphere150- 1400140- 1200130U- 1000 °10080//90N2LJ-4)- 800 0^/ La_- 600- 400- 200120.4)_c17)1V110C°200.0^20.0^40. 0^60. 0^80.0^100.0^120. 0^140. 0^160.0^180. 0 200. 0 220. 0^Time (min) DuPont 1090Figure 4.30 Carbonate Formation on a 1:10 Ca:Sm Oxide Catalyst under a 100% CO 2Atmosphere4.29). The weight of the sample remained essentially steady up to almost 400°C undera nitrogen atmosphere, at which point the weight dropped suddenly. It levelled offbriefly at 500°C and dropped again at a slower rate to 650°C. The carbon dioxide wasthen introduced. The weight decreased slightly for about 4 minutes, and thenincreased very rapidly back to 100%, likely due to carbonate formation. The increasein weight continued at a slower rate, again due to carbonate formation. This weightgain increased as the temperature was increased, first to 750°C, and then to 850°C. Ateach subsequent temperature the rate of weight increase decreased, indicating that lesscarbonate was being formed.The 1:10 Ca:Sm oxide was tested in a similar manner to the sodium doped catalyst (seeFigure 4.30). Considerably less carbonate formation was observed. The initial shapeof the curve up to 650°C appeared to be a combination of the samarium oxide andcalcium oxide curves, as would be expected. After the CO 2 was introduced, the weightincreased by about 1%, followed by a small decrease of 0.2%, indicative of an overshootin carbonate formation. The weight then levelled out and stared to increase veryslowly, up to the time that the temperature was increased to 750°C. The weight thenstarted to decrease slowly. When the temperature was increased to 850°C, the weightcontinued to decrease, at a rate of about 3 times the rate at 750°C. Over the timeperiod 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 thecarbonate 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 a125slight increase in weight after introduction of the carbon dioxide. The method is notsensitive enough to determine these low levels of carbonate formation. Therefore, noneof the other samples were tested.Based on the tests performed and information concerning the stability of thecarbonates, the relative amount of carbonate formed for each catalyst at eachtemperature was estimated and graphed (see Figures 4.31 to 4.33). Potassium was notincluded 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 1:10catalysts. Magnesium and samarium oxide were assumed to have no carbonateformation. 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 werecorrected based on relative amounts of gas phase CO2 over each catalyst. In actualoperation of an oxidative coupling reactor, the concentrations of gas phase oxygen andwater would also have an effect on the amount of carbonate formed, but these effectshave not been included in this initial study.As can be seen from these figures, the C2+ selectivity decreases with an increase incarbonate formation at 650°C, while at 750°C and 850°C it actually increases. Onepossible interpretation is that the state of equilibrium between carbonate formation anddecomposition may be important. At 650°C, the carbonate is still forming, particularlyfor the sodium catalysts, which results in low C2+ selectivity. At 750°C, the carbonateis closer to a state of equilibrium, with both formation and decomposition probablyoccurring, resulting in an increase in C2+ selectivity. At 850°C, the calcium catalysts126C2+ yield^-0-- CH4 conversion --•- C2+ selectivityCatalyst Performance as a Function of CarbonateFormationCorrected for CO2 Concentration, T=650C45.040.035.030.025.020.015.010.05.00.00.00 0.50^1.00^1.50Relative Carbonate Formationae2.00 2.501:10 N^a ^Na^14,-Sm,MgCa70.060.050.040.030.020.010.00.00.00Sm, Mg Ca_1:10 ^.1..________..„_; 1:10 NaCa  Naa •-• •0.50^1.00^1.50^2.00Relative Carbonate Formation2.50 3.00C2+ yield   CH4 conversion --•- C2+ selectivityFigure 4.31Figure 4.32Catalyst Performance as a Function of CarbonateFormationCorrected for CO2 Concentration, T=750C127E()0zCEnoNozc)0)2E00.350.300.258 0.20U8 0.150.100.050.00Figure 4.33Catalyst Performance as a Function of CarbonateFormationCorrected for CO2 Concentration, T=850C60.050.040.030.0^20.010.00.0Ca _ Ca •I:10•k-.-.Sm, Mg, Na, 1:'0 Na^^0.00^0.01^0.02^0.03^0.04^0.05^0.06^0.07^0.08^0.09Relative Carbonate Formation— C2+ yield^—0-- CH4 conversion --•— C2+ selectivityFigure 4.34Ratio of CO to CO2 ProducedReactionTemperature• 650 C^ 750 Cla 850 C128were the only catalysts which were estimated to have carbonate present, which resultedin an increase in C2+ selectivity for the 1:10 catalyst.The methane conversion appeared to improve with a small amount of carbonatepresent and then decreased, at both 650°C and 750°C. At 850°C, only a small increasein methane conversion was observed, which is consistent with the results at the lowertemperatures for the small amount of carbonate present. The actual amount ofcarbonate present may affect the methane conversion, with no dependence on whetherit is forming or decomposing. A small optimum amount appears to improve methaneconversion.The CO /CO2 ratios obtained in this study were compared to those obtained by Duboisand Cameron. This ratio is graphed for all catalysts in Figure 4.34. The mostsignificant effect is found for the Na (1:10), for which the CO /CO 2 ratio is considerablylower than for the other dopants. This indicates that the carbonate formation is alsomuch more significant. If the effect for the Na (1:10) may be explained by greatercarbonate formation than for the other catalysts, the low methane conversion isconsistent with ideas and results reported by Peil et al. (1991a), who suggested that alower methane conversion is a result of carbonate formation.Overall, the mechanisms underlying the effects observed due to carbonate formationremain unclear. The improved C2+ selectivity may be due to modification of the ratioof basic to acid sites. This ratio was suggested by Becker and Baerns (1991) to be animportant variable. Another possibility is that the physical processes of formation and129decomposition of the carbonate may lead to increased crystal disorder, resulting in anincrease in oxygen vacancies.4.9.2 Effect of Gas Phase Composition on Carbonate FormationThe level of carbonate formation will also depend on the amount of oxygen and waterin the gas phase, according to equations such as:Na202 + CO2 .r." Na2CO3 + 1 /2 02 (4.7)Na2CO3 + H2O i(- 2 NaOH + CO 2 (4.8)2 NaOH + 1 /2 02 4- Na202 + H2O (4.9)The amount of each solid species in the oxide/carbonate/hydroxide system presentunder equilibrium conditions can be estimated by free energy minimization if it isassumed that the solids have unit activity. This was simulated using Aspen Plussoftware, for pure sodium oxide in the presence of a dilute gas phase mixture ofoxygen, water, and carbon dioxide. Using this thermodynamic approach and assumingall species to be solids, approximately twice as much NaOH was formed as Na 2CO3 orNa2O at each of the reaction temperatures (650°C, 750°C, and 850°C). The CO 2 andwater were almost completely consumed. At such temperatures pure NaOH would inpractise be present as a liquid, so the thermodynamics are in fact much morecomplicated than a simple model can represent. Further complications arise becausethe sodium is assumed to be dispersed throughout the Sm 203 crystal, perhaps in a solid130solution; again the actual behaviour can not be determined from the study of the pureNa20. For example, it has been shown that the decomposition temperature of stablecarbonates decreases when the carbonate forming material is mixed with a basic non-carbonate forming oxide (Dubois and Cameron, 1991, 1992). Therefore, the sodiumoxide is likely significantly stabilized by the samarium oxide. The formation ofcarbonate and hydroxide will likely occur, but to an unknown extent which is certainlyless than that for pure sodium oxide. The effect of the gas phase composition can begeneralized, 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 hadlower 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 wouldresult in a slower regeneration of active sites and therefore a lower methaneconversion. The effect on the C2+ selectivity would depend on the nature of the activesites and whether the carbonate layer affects this concentration negatively.The net effect of the oxygen concentration would appear to be to limit the carbonateformation according to reaction (4.7), as the carbonate forming catalysts result in loweroxygen conversion, and therefore, higher oxygen concentrations.Lower oxygen conversion should also result in lower water concentrations. The effectof water on the carbonate equilibrium should be to inhibit carbonate formation andincrease the formation of MOH or M(OH) 2, which will result in different active sites,131the effect of which is not known.Therefore, those catalysts which tend to lower degrees of carbonate formation, have gasphase compositions with lower oxygen concentration and higher water concentration,resulting in higher hydroxide concentrations. Those with a greater tendency towardcarbonate formation, have gas phase compositions with higher oxygen concentrationand lower water concentration, which tends to limit the amount of carbonate formationand hydroxide formation in favour of the oxide.4.10 Ionic Radius of DopantIt has been suggested that the ionic radius of the dopant material may have an affecton 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 indifficulty 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 dopedcatalyst was the least selective catalyst at 750°C and 850°C. The other dopants used inthis 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 samariumoxide with magnesium actually had a detrimental effect on C2+ selectivity. A possibleexplanation is that the small cation incorporated into the crystal lattice distorts thecrystal, resulting in a less selective catalyst, possibly due to lower oxygen mobility,132which has been indicated as a factor in selective catalysts.Table 4.15Estimated Ionic Radii of Selected Elements 15Group IA IonicRadius(A)Group IIA IonicRadius(A)Group IIIBandLanthanidesIonicRadius(A)Li+ 0.6-0.78Na+ 0.95-0.98 Mg2+ 0.65-0.8K+ 1.33 Ca2+ 0.99-1.10 Sclf 0.81-0.91Sr+ 1.12-1.27 Y3+ 0.92-1.07Ba2+ 1.34-1.35 La3+ 1.14-1.22Sm3+ 1.0-1.13The effect of ionic radius of dopants has been discussed with regard to samarium oxideas a means of explaining catalyst performance (Korf et al., 1988). The addition of Li hasa detrimental effect on the catalytic performance, which has been shown by XRD to bedue to a destabilizing effect which favours the cubic-monoclinic transformation. Caand Na have been shown to have a beneficial effect. The tendency of the additives tostabilize the cubic phase is reported to be in the order: Ca > Mg > La. Na has alsobeen shown to induce a decrease in stability. These effects may be due to the ionicradii of the different ions and the structure of the pure promoter phase that would bepresent under these conditions. The CaO and MgO are cubic, and their ions fit into thesurface of Sm203 without stress due to the size of their ions. La' has a larger radiusand a smaller stabilising effect. Na 2CO3, Li2CO3, and LiOH have a monoclinic structure,which encourages the phase transition of Sm 203 from cubic to monoclinic.'Dubois and Cameron, 1990133Of the dopant cations used in this study, potassium (K+) alone has a larger radius thanSm3+, and is even larger than that of La', which was observed to reduce the stabilityof the Sm203. In addition, due to the much larger radius, the K+ ions may not fit intothe crystal structure and therefore may not have the desired doping effect. This mayaccount for the low C2+ selectivity obtained with the potassium doped catalyst, whichcould not be explained based on basicity (see Section 4.8).Ohno and Moffat (1991) determined a dependence of selectivity on the ratio of thecation radius to charge, for alkali and alkaline earth phosphate catalysts. They plottedthe C2+ selectivities obtained for various catalysts, over a range of 0% to 75%, againstthe radius/charge ratio, which was found to pass through a maximum. The lowestradius/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 highestselectivity 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 theone-dimensional space occupied per unit charge by the particular cation. The cationmay enhance the ability of the surface oxygen atom to extract a hydrogen atom frommethane through a polarization effect. If so, this effect may be related to the chargedensity. However, too high a polarization could lead to undesirable results; that is, adecrease in selectivity.Comparison of Ohno and Moffat's study to the present results reveals a similar effect,134although the cation concentrations in this study are much lower, resulting in smalldifferences between cations (see Table 4.16). There is no significant difference in theC2+ selectivity for Mg and Ca dopants.Table 4.16Effect of Cation Radius/Charge Ratio on C2+ SelectivityDopant 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, CH4/02=4This effect may also be explained as a result of electronegativity, which was discussedpreviously. As the cation radius increases, the positive charge has less effect on theoxygen ion, causing the oxygen to have a higher partial charge, and in effect be morebasic. At high values of cation radius to charge ratio, as in the case of potassium, thebasicity may be too high. This may result in a change in the oxygen species, whichmay lead to increased COX yield. As can be seen in Figure 4.35, the potassium catalystis the most active in CO, production at 750°C and 850°C.135Figure 4.351365. CONCLUSIONSThe addition of alkali and alkaline earth dopants to the samarium oxide catalystresulted in changes in catalyst performance. Based on the work conducted in theproject, the following conclusions were made:i) The concentration of dopant used had a significant effect on catalystperformance.ii) No new phases were observed in the Sm 2O3 crystal upon addition of the sodiumand calcium dopant cations, indicating that the cations were dispersedthroughout the crystal, although possibly not uniformly.iii) The catalyst preparation procedure used in this study was different than thatused by other researchers. It was found that the catalyst preparation can havea significant effect on the catalyst performance, which has also been observedby others. The results were interpreted in this case to indicate that there maybe an optimum number of active sites, with too many active sites resulting inan effective combustion catalyst.iv) The effect of catalyst surface area has been studied by many researchers withvarying conclusions as to the effect on the oxidative coupling of methane. Theresults of this study indicated that over the surface area tested, this variable didnot have a significant effect on the catalyst performance.137v) The basicity of the catalyst appears to have a significant effect on the catalystperformance, 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 anoptimum temperature for maximum C2+ yield, which is dependent on theamount and nature of the dopant. The formation of carbonates on the catalystsurface is likely partially responsible for the temperature dependent behaviourof the catalysts. The mechanism of carbonate interaction with the catalyst andreaction molecules is not known. However, the process of decomposition of thecarbonate may create active sites, thereby increasing the C2+ selectivity. Anattempt was made to quantify the carbonate formation on the sodium andcalcium doped catalysts, for which little information was available in theliterature.vii) The results support the hypothesis that the ionic radius of the cation dopantmust be similar to or smaller than the support cation to achieve effectiveinclusion in the crystal lattice.1386.^REFERENCESAigler, J.M. and J.H. Lunsford (1991), "Oxidative Dimerization of Methane over MgOand Li+/Mg0 Monoliths", Applied Catalysis, 70, 1991, pp. 29-42.Alcock, C.B., J.J. Carberry, R. Doshi, and N. Gunasekaran (1992), "Coupling Reactionson 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 ofMethane 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 CatalyticOxidation of Methane over Lithium-Promoted Magnesium Oxide", Journal ofPhysical Chemistry, 92, 1988, pp. 4576-4578.Amorebieta, V.T. and A.J. Colussi (1989), "Mass Spectrometric Studies of theLow-Pressure Oxidation of Methane on Samarium Sesquioxide", Journal ofPhysical Chemistry, 93, 1989, pp. 5155-5158.Andorf, R., L. Mleczko, D. Schweer and M. Baerns (1991), "Oxidative Coupling ofMethane in a Bubbling Fluidized Bed Reactor", Canadian Journal of ChemicalEngineering, Vol. 69, August, 1991.Anshits, A.G., E.N. Voskresenskaya and L.I. Kurteeva (1990), "Role of Defect Structureof 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 OxideSystems Catalytic Properties in Methane Oxidative Coupling", Eds. A. Holmenet 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 ofMethane in the Homogeneous Gas Phase under Pressure", Journal of the ChemicalSociety, 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 EngineeringAspects, Van Nostrand Reinhold, New York, 1992, pp. 382-402.139Becker, S. and M. Baerns (1991), "Oxidative Coupling of Methane over La 203-Ca0Catalysts, 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 Conductivityof Catalysts for their Performance in the Oxidative Coupling of Methane",Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, AmericanChemical Society, April 5-10, 1992, pp. 111-116.Brady, J.E. and G.E. Humiston (1982), Third Edition, General Chemistry, Principles andStructure, John Wiley & Sons, Inc., 1982.Burch, R., S.C. Tsang, C. Mirodatos, and J.G. Sanche (1990), "Kinetic Isotope Effects inMethane 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 ofSteam in the Oxidative Coupling of Methane over MgO and CaO BasedCatalysts", Catalysis Today, 13, 1992, pp. 589-592.Cant, N.W., C.A. Lukey, P.F. Nelson, and R.J. Tyler (1988), 'The Rate Controlling Stepin the Oxidative Coupling of Methane over a Lithium-promoted MagnesiumOxide 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 andTheir 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 forOxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Divisionof 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-PhaseMethyl 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 ofMethane Cofeed Catalysts", Applied Catalysis, 67, 1990, pp. 49-71.Dubois, J.-L. and C.J. Cameron (1991), "Synergy between Stable Carbonates and Yttriain Selective Catalytic Oxidation of Methane", Chemistry Letters, 1991, pp.1089-1092.140Dubois, J.-L. and C.J. Cameron (1992), "Surface Carbonate on Methane CouplingCatalysts: A Poison or a Catalyst Itself?", Symposium on Natural Gas UpgradingII, 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 CatalyticOxidative 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 overLithium-Promoted Magnesium Oxide Catalysts in Fixed-Bed and Fluidized-BedReactors", Energy & Fuels, 4, 1990, pp. 85-93.Edwards, J.H., K.T. Do, and R.J. Tyler (1991), "The 'OXCO' Process for Natural GasConversion Via Methane Oxidative Coupling", Eds. A. Holmen et al., Natural GasConversion, Elsevier Science Publishers B.V., Amsterdam, 1991, pp. 489-495.Ekstrom, A. and J.A. Lapszewicz (1988a), "Methane Adsorption on a WorkingSamarium Oxide Catalyst and its Role in Hydrocarbon Formation during HighTemperature Partial Oxidation.", Journal of the Chemical Society, ChemicalCommunications, 1988, pp. 797-799.Ekstrom, A. and J.A. Lapszewicz (1988b), "Investigation of the Mechanism of the PartialOxidation of Methane over Samaria Catalysts", Preprints, Symposia, AmericanChemical Society, September 25-30, 1988, pp. 430-436.Ekstrom, A. and J.A. Lapszewicz (1988c), "The Role of Oxygen in the Partial Oxidationof Methane over a Samarium Oxide Catalyst", Journal of the American ChemicalSociety, 110, 1988, pp. 5226-5228.Ekstrom, A. and J.A. Lapszewicz (1989a), "A Study of the Mechanism of the PartialOxidation of Methane over Rare Earth Oxide Catalysts Using Isotope TransientTechniques.", Journal of Physical Chemistry, 93, 1989, pp. 5230-5237.Ekstrom, A., J.A. Lapszewicz and I. Campbell (1989b), "Origin of the Low Limits in theHigher 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 OxidativeCoupling Reaction of Methane", Applied Catalysis, 63, 1990, pp. 253-269.Ekstrom, A. (1992), "The Oxidative Coupling of Methane: Reaction Pathways and TheirProcess Implications", Ed. E.E. Wolf, Methane Conversion By Oxidative ProcessesFundamental and Engineering Aspects, Van Nostrand Reinhold, New York, 1992,pp. 99-137.Feng, Y., J. Niiranen, and D. Gutman (1991a), "Kinetic Studies of the CatalyticOxidation of Methane. 1. Methyl Radical Production of 1% Sr/La 203", Journal141of Physical Chemistry, 95, 1991, pp. 6558-6563.Feng, Y., J. Niiranen, and D. Gutman (1991b), "Kinetic Studies of the CatalyticOxidation of Methane. 2. Methyl Radical Recombination and Ethane Formationover 1% Sr /La203", Journal of Physical Chemistry, 95, 1991, pp. 6564-6568.Feng, Y., J. Niiranen, L.N. Krasnoperov, and D. Gutman (1992), "Fundamental Studiesof Gas/Surface Processes Occurring during Methane coupling on 1% Sr/La203",Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, AmericanChemical Society, April 5-10, 1992, pp. 161-166.Fox, J.M., III, T.P. Chen, and B.D. Degen (1990), "An Evaluation of Direct MethaneConversion Processes", Chemical Engineering Progress, April 1990, pp. 42-50.Fujimoto, K., K. Omata, and J. Yoshihara (1991), "Selective Oxidation of Methane toEthane 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 Investigationof Individual Reaction Steps in the Oxidative Coupling of Methane over LithiumDoped Magnesium Oxide", Catalysis Today, 4, 1989, pp. 453-461.Grzybek, T. and M. Baerns (1991), "Surface Interaction between Methane andAlkali/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 CatalyticOxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Divisionof 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 SodiumAluminate- 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 PhaseReaction 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 ofMethane over Oxide Catalysts", Journal of the Chemical Society, FaradayTransactions, 1, 85(8), 1989, pp. 2507-2523.Hutchings, G.H., M.S. Scurrell, and J.R. Woodhouse (1989b), "Selective Oxidation ofMethane 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 ofMethane over Samarium and Lanthanum Oxides: A Study of the ReactionMechanism", Catalysis Today, 4, 1989, pp. 371-381.142Hutchings, G.J. and M.S. Scurrell (1992), "Studies of the Mechanism of the OxidativeCoupling of Methane Using Oxide Catalysts", Ed. E.E. Wolf, Methane ConversionBy Oxidative Processes Fundamental and Engineering Aspects, Van NostrandReinhold, New York, 1992, pp. 200-258.Iwamatsu, E., T. Moriyama, N. Takasaki, and K. Aika (1987), "Importance of theSpecific Surface Area of the Catalyst in Oxidative Dimerization of Methane overPromoted Magnesium Oxide", Journal of the Chemical Society, ChemicalCommunications, 1987, pp. 19-20.Iwamatsu, E., T. Moriyama, N. Takasaki, and K. Aika (1988), "Oxidative Coupling ofMethane over Promoted Magnesium Oxide Catalysts; Relation between Activityand 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 onOxidative Coupling of Methane", Journal of Catalysis, 124, 1990, pp. 566-569.Kalenik, Z. and E.E. Wolf (1991a), "Transient and Isotopic Studies of the OxygenTransport and Exchange During Oxidative Coupling of Methane on Sr PromotedLa203", Catalysis Letters 9, 1991, pp. 441-450.Kalenik, Z. and E.E. Wolf (1991b), "Methane Oxidative Coupling over LithiumPromoted Lanthanum-Titanate Oxide", Eds. A. Holmen et al., Natural GasConversion, 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 theActivity of La2O3, ThO2, and ZrO2 Catalysts during Oxidative Dimerization ofMethane", 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 Dehydrogenationof Ethane and Oxidative Coupling of Methane over Rare Earth Oxides", AppliedCatalysis, 75, 1991, pp. 321-330.Korf, S.J. et al. (1988), "The Effect of Promoters on the Behaviour of Sm 203 Catalysts forthe Oxidative Coupling of Methane", Preprints, Symposia, American ChemicalSociety, 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 Dopedand 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), "LithiumChemistry of Lithium Doped Magnesium Oxide Catalysts Used in the Oxidative143Coupling 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 ofMethane over Sm203 and La203", Symposium on Structure-Activity Relationships inHeterogeneous Catalysts, Division of Petroleum Chemistry, Inc., American ChemicalSociety, 1990, pp. 54-57.Korf, S.J., J.G. van Ommen and J.R.H. Ross (1991), "The Oxidative Coupling of Methaneover Sm203 and La203", Eds. Grasselli, R.K. and A.W. Sleight, Structure-Activityand Selectivity Relationships in Heterogeneous Catalysis, Elsevier Science PublishersB.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 SafeReactors and Products, Series E: Applied Sciences - Volume 225, Kluwer AcademicPublishers.Labinger, J.A. (1988), "Oxidative Coupling Of Methane: An Inherent Limit toSelectivity?", Catalysis Letters, 1, 1988, pp. 371-376.Labinger, J.A. (1991), "Mechanism-Imposed Limitations on the Yield of HigherHydrocarbons from the Oxidative Coupling of Methane, and AlternateApproaches to Methane Conversion", Symposium on Methane Upgrading, Divisionof 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 OxideCatalysts - A Key to Selectivity of Oxidative Methane Coupling?", Symposium onNatural Gas Upgrading II, Division of Petroleum Chemistry, American ChemicalSociety, April 5-10, 1992, pp. 102-110.Lee, J.S. and S.T. Oyama (1990), "Comments on the Mechanism of Oxidative Couplingof 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 HigherHydrocarbons", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier SciencePublishers 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 Dimerizationof 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 MethaneOxidative Coupling over Samarium Oxide", Preprints of Papers Presented at the144196th 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 CatalyticOxidative Coupling of Methane", Symposium on Natural Gas Upgrading II, Divisionof Petroleum Chemistry, American Chemical Society, April 5-10, 1992, pp. 173-179.Martin, G.A. and C. Mirodatos (1987), "Evidence of Carbene Formation in OxidativeCoupling of Methane over Lithium Promoted Magnesium Oxide", Journal of theChemical Society, Chemical Communications, 1987, pp. 1393-1394.Martin, G.A. and C. Mirodatos (1992), "Morphological Aspects of Catalysts forOxidative Coupling of Methane", Ed. E.E. Wolf, Methane Conversion By OxidativeProcesses Fundamental and Engineering Aspects, Van Nostrand Reinhold, NewYork, 1992, pp. 351-381.McCarty, J.G. (1991), "Kinetic and Thermodynamic Descriptions of Co-oxidativeMethane Dimerization", Symposium on Methane Upgrading, Division of PetroleumChemistry, 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 ofCH4/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-MgOxides: 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 ofMethane 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 andPurification", Eds. E.C. Subbarao and W.E. Wallace, Science and Technology of RareEarth Materials, Academic Press, 1980, pp. 3-24.Nelson, P.F., C.A. Lukey, and N.W. Cant (1989), "Measurements of Kinetics IsotopeEffects and Hydrogen/Deuterium Distributions over Methane OxidativeCoupling Catalysts", Journal of Catalysis, 120, 1989, pp. 216-230.145Nelson, P.F., E.M. Kennedy and N.W. Cant (1991), "Isotopic Labelling Studies of theMechanism of the Catalytic Oxidative Coupling of Methane", Eds. A. Holmenet 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- andAlkaline Earth-Phosphate Catalysts", Catalysis Letters, 1991, pp. 23-34.Olsbye, U. and G. Desgrandchamps (1991), "Preliminary Tests for a Kinetic Study ofthe 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 ofthe Oxidative coupling of Methane over a BaCO3/La20.(CO3)3, Catalyst",Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, AmericanChemical Society, April 5-10, 1992, pp. 180-187.Omata, K., S. Hashimoto, H. Tominaga and K. Fugimoto (1989), "Oxidative Couplingof 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 inOxidative 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 theSynthesis of C2H4 and C2H6 via Oxidative Coupling of Methane", Journal ofCatalysis, 100, 1986, pp. 353-359.Otsuka, K. and K. Jinno (1986b), "Kinetic Studies on Partial Oxidation of Methane overSamarium Oxides", Inorganica Chimica Acta, 121, 1986, pp. 237-241.Otsuka, K. and A.A. Said (1987), "Role of Lattice Oxygen Atoms in Partial Oxidationsof Methane, Ethane and Ethylene over Samarium Oxides", Inorganica ChimicaActa, 132, 1987, pp. 123-128.Otsuka, K. and T. Komatsu (1987), "High Catalytic Activity of Sm 203 for OxidativeCoupling of Methane into Ethane and Ethylene", Chemistry letters, The ChemicalSociety of Japan, 1987, pp. 483-484.Otsuka, K. and T. Nakajima (1987), "Oxidative Coupling of Methane over SamariumOxides using N20 as the Oxidant", Journal of the Chemical Society, FaradayTransactions, 1, 83, pp. 1315-1321.Otsuka, K., A.A. Said, K. Jinno and T. Komatsu (1987), "Peroxide Anions as PossibleActive Species in Oxidative Coupling of Methane", Chemistry Letters, 1987, pp.77-80.146Otsuka, K., M. Inaida, Y. Wada, T. Komatsu, and A. Morikawa (1989), "Isotopic Studieson Oxidative Methane Coupling over Samarium Oxide", Chemistry letters, TheChemical Society of Japan, 1989, pp. 1531-1534.Otsuka, K., Y. Murakami, Y. Wada, A.A. Said, and A. Morikawa (1990), "OxidativeCoupling of Methane, Ethane, and Propane with Sodium Peroxide at LowTemperatures", 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 ByOxidative 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 Sitesof the Catalyst in Oxidative Coupling of CH4 over LiO/MgO Catalysts Preparedby 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 OxygenPathway during Methane Oxidation over a Li/MgO Catalyst", Journal of PhysicalChemistry, 93, 1989, pp. 5977-5979.Peil, K.P., Marcelin G., J.G. Goodwin, Jr., and A. Kiennemann (1990a), "TransientKinetic Analysis of the Oxidative Coupling of Methane over Sm 203", Ed. Dekker,Chemical Industries, 46, Novel Production Methods of Ethylene and LightHydrocarbons, 1990, pp. 61-74.Peil, K.P., J.G. Goodwin, Jr., and G. Marcelin (1990b), "Surface Concentrations andResidence Times of Intermediates on Sm 2O3 during the Oxidative Coupling ofMethane", 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 theOverall 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 theOxidative 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 theOxidative Coupling of Methane", Ed. E.E. Wolf, Methane Conversion By OxidativeProcesses Fundamental and Engineering Aspects, Van Nostrand Reinhold, NewYork, 1992, pp. 138-167.Peng, X.D., D.A. Richards, and P.C. Stair (1990), "Surface Composition and Reactivityof Lithium-Doped Magnesium Oxide Catalysts for Oxidative Coupling ofMethane", Journal of Catalysis, 121, 1990, pp. 99-109.147Pereira, P., S.H. Lee, G.A. Somorjai, and H. Heinemann (1990), "The Conversion ofMethane to Ethylene and Ethane with Near Total Selectivity by LowTemperature (<610°C) Oxyhydrogenation over a Calcium-Nickel-PotassiumOxide Catalyst", Catalysis Letters, 6, 1990, pp. 255-262.Philipp, R., K. Omata, A. Aoki, and K. Fujimoto (1992), "On the Active Site ofMgO/CaO Mixed Oxide for Oxidative Coupling of Methane", Journal ofCatalysis, 134, 1992, pp. 422-433.Pinabiau-Carlier, M., A. Ben Hadid, and C.J. Cameron (1991), "The Effect of TotalPressure on the Oxidative Coupling of Methane Reaction Under CofeedConditions", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier SciencePublishers B.V., Amsterdam, 1991, pp. 183-190.Poirier, M.G., A.R. Sanger, and K.J. Smith (1991), "Direct Catalytic Conversion ofMethane", The Canadian Journal of Chemical Engineering, Vol. 69, October 1991, pp.1027-1035.Rasko, J., G.A. Somorjai, and H. Heinemann (1992), "Catalytic Low-TemperatureOxydehydrogenation of Methane to Higher Hydrocarbons with very HighSelectivity 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 theOxidative Coupling of CH4 over a Li/MgO Catalyst, Factors affecting the Rateof Total Oxidation of C2H6 and C2H4.", 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 ofthe Comparative Reactivities of Ethane and Ethylene in the Presence of Oxygenover Li/MgO and Ca/Sm203 Catalysts in Relation to the Oxidative Coupling ofMethane", Catalysis Today, 4, 1989, pp. 441-452.Sokolovskii, V.D., O.V. Buyevskaya, L.M. Plyasova, and G.S. Litvak (1990), "Structureand Catalytic Properties of Samarium-Calcium Oxide Catalysts in MethaneOxidative Dehydrodimerization", Catalysis Today, 6, 1990, pp. 489-495.Spinicci, R. (1991), "Properties of Zinc Oxide Based Catalysts Towards MethaneCoupling 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 OxideCatalysts", Applied Catalysis, 77, 1991, pp. 45-53.Suzuki, T., K. Wada and Y. Watanabe (1990), "Effects of Carbon Dioxide and CatalystPreparation on the Oxidative Dimerization of Methane", Applied Catalysis, 59,1990, pp. 213-225.148Tong, Y., M.P. Rosynek, and J.H. Lunsford (1989), "Secondary Reactions of MethylRadicals with Lanthanide Oxides: Their Role in the Selective Oxidation ofMethane", The Journal of Physical Chemistry, Vol. 93, No. 8, 1989.Tong, Y. and J.H. Lunsford (1991), "Mechanistic and Kinetic Studies of the Reactionsof Gas-Phase Methyl Radicals with Metal Oxides", Journal of the AmericanChemical 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), "ElementaryReactions and Kinetic Modelling of the Oxidative Coupling of Methane", Ed. E.E.Wolf, Methane Conversion By Oxidative Processes Fundamental and EngineeringAspects, 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 ofMethane", Eds. A. Holmen et al., Natural Gas Conversion, Elsevier SciencePublishers B.V., Amsterdam, 1991, pp. 33-40.Walsh, D.E., D.J. Martenak, S. Han, and R.E. Palermo (1992a), "Direct OxidativeMethane Conversion at Elevated Pressure and Moderate Temperatures",Symposium on Natural Gas Upgrading II, Division of Petroleum Chemistry, AmericanChemical Society, April 5-10, 1992, pp. 147-152.Walsh, D.E., D.J. Martenak, S. Han, and R.E. Palermo (1992b), "Direct OxidativeMethane 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 OxidativeMethane 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 Methanewith Peroxide Ions over Barium-Lanthanum-Oxygen Mixed Oxide", AppliedCatalysis A: General, 79, 1991, pp. 203-214.Yates, D.J.C. and N.E. Zlotin (1988), "Blank Reactor Corrections in Studies of theOxidative Dehydrogenation of Methane", Journal of Catalysis, 111, pp. 317-324.149APPENDIX A: RESULTS OF REACTOR CATALYST TESTS150Table A.1Samarium Oxide Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Temperature (°C) 650 650 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 10 20 40 80 20 40 20 4002 5 5 5 5 5 5 5 5He 85 75 55 15 75 55 75 55CH4/02 2 4 8 16 4 8 4 8Average Average Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Products (%)CH4 6.96 15.34 33.04 72.08 14.12 31.81 13.87 31.7302 2.73 1.77 1.41 1.21 0.37 0.30 0.00 0.00CO2 1.04 1.15 1.15 1.09 1.55 1.29 1.70 1.48CO 0.10 0.24 0.38 0.55 0.38 0.53 0.58 0.71C2H4 0.05 0.12 0.21 0.27 0.51 0.76 0.62 0.93C2H6 0.12 0.32 0.63 1.03 0.58 1.04 0.50 0.84C3's 0.00 0.02 0.04 0.06 0.04 0.08 0.03 0.06Total Carbon out 8.45 17.66 36.36 76.51 18.34 37.46 18.48 37.63Output flowsTime for 50m1 (s) 30.04 30.16 30.93 32.46 30.86 31.52 31.05 31.58Temperature (°C) 22.27 22.87 23.47 23.87 24.00 23.83 23.50 23.50mL/min (at 21.1°C) 99.49 98.88 96.23 91.62 96.28 94.30 95.84 94.22gmo1/min 4119 4094 3985 3794 3986 3905 3968 3901Output (p.mol/min)CH4 287 628 1316 2734 563 1242 550 123802 112 72 56 46 15 12 0 0CO2 43 47 46 41 62 51 68 58CO 4 10 15 21 15 21 23 28C2H4 2 5 8 10 20 30 25 36C2H6 5 13 25 39 23 40 20 33C3's 0 1 2 2 2 3 1 2Total Carbon out 348 723 1449 2902 731 1463 733 1468p.mol Carbon lost 31 122 125 232 114 111 112 105% Carbon lost 8 14 8 9 11 7 11 7151Table A.1 (continued)Samarium Oxide Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8STY tunol/g/sC2s 2.4 6.1 11.1 16.5 14.4 23.4 14.9 23.0% ConversionCH4 17.6 13.2 9.1 5.8 23.0 15.1 25.0 15.702 44.6 64.6 71.9 75.8 92.6 94.0 100.0 100.0% SelectivityCO2 69.6 49.5 34.6 24.6 36.7 22.9 36.9 25.1CO 6.7 10.3 11.4 12.4 9.0 9.4 12.5 12.0COX selectivity 76.3 59.8 46.1 37.1 45.7 32.2 49.4 37.1C2H4 7.1 10.6 12.4 12.2 24.2 26.9 26.9 31.4C2H6 16.5 27.5 37.9 46.7 27.3 36.7 21.8 28.5C2 selectivity 23.7 38.1 50.3 58.9 51.5 63.5 48.7 59.9C3's 0.0 2.1 3.6 4.1 2.8 4.2 1.9 3.0C2+ selectivity 23.7 40.2 53.9 62.9 54.3 67.8 50.6 62.9% YieldCO2 12.3 6.5 3.2 1.4 8.5 3.5 9.2 3.9CO 1.2 1.4 1.0 0.7 2.1 1.4 3.1 1.9CO, yield 13.5 7.9 4.2 2.1 10.5 4.9 12.3 5.8C2H4 1.3 1.4 1.1 0.7 5.6 4.1 6.7 4.9C2H6 2.9 3.6 3.5 2.7 6.3 5.5 5.4 4.5C2 yield 4.2 5.0 4.6 3.4 11.9 9.6 12.2 9.4C3's 0.0 0.3 0.3 0.2 0.7 0.6 0.5 0.5C2+ yield 4.2 5.3 4.9 3.6 12.5 10.2 12.6 9.9152Table A.2Calcium Doped Samarium Oxide (1:100) Prepared in AirAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8Temperature (°C) 650 650 650 650 750 750 850 850Products (%)CH4 6.61 15.27 33.64 72.23 14.43 32.81 14.14 32.7302 3.31 2.65 2.07 1.33 0.80 0.71 0.06 0.03CO2 1.16 1.27 1.24 1.15 1.62 1.35 1.77 1.61CO 0.27 0.47 0.67 0.87 0.54 0.67 0.54 0.62C2H4 0.05 0.11 0.18 0.25 0.45 0.66 0.57 0.82C2H6 0.12 0.30 0.58 0.97 0.53 0.93 0.47 0.75C3's 0.00 0.00 0.02 0.04 0.02 0.03 0.02 0.05Total Carbon out 8.37 17.82 37.13 76.81 18.62 38.11 18.59 38.27% ConversionCH4 21.1 14.3 9.4 6.0 22.5 13.9 23.9 14.502 33.7 46.9 58.6 73.5 84.1 85.7 98.9 99.5% SelectivityCO2 65.8 49.6 35.4 25.2 38.6 25.5 39.8 29.0CO 15.3 18.3 19.2 19.0 13.0 12.7 12.2 11.3COx selectivity 81.1 67.9 54.6 44.2 51.6 38.2 52.0 40.3C2H4 5.7 8.9 10.5 10.9 21.7 24.9 25.7 29.8C2H6 13.2 23.2 33.2 42.2 25.3 35.2 21.0 27.2C2 selectivity 18.9 32.1 43.7 53.2 47.0 60.1 46.7 57.0C3's 0.0 0.0 1.7 2.6 1.4 1.7 1.4 2.7C2+ selectivity 18.9 32.1 45.4 55.8 48.4 61.8 48.0 59.7% YieldCO2 13.9 7.1 3.3 1.5 8.7 3.6 9.5 4.2CO 3.2 2.6 1.8 1.1 2.9 1.8 2.9 1.6COx yield 17.1 9.7 5.1 2.6 11.6 5.3 12.4 5.8C2H4 1.2 1.3 1.0 0.7 4.9 3.5 6.1 4.3C2H6 2.8 3.3 3.1 2.5 5.7 4.9 5.0 3.9C2 yield 4.0 4.6 4.1 3.2 10.6 8.4 11.2 8.2C3's 0.0 0.0 0.2 0.2 0.3 0.2 0.3 0.4C2+ yield 4.0 4.6 4.3 3.3 10.9 8.6 11.5 8.6153Table A.3Calcium Doped Samarium Oxide (1:100) Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Temperature (°C) 650 650 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 10 20 40 80 20 40 20 4002 5 5 5 5 5 5 5 5He 85 75 55 15 75 55 75 55CH4/02 2 4 8 16 4 8 4 8Average Average Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Products (%)CH4 6.26 14.54 32.73 70.63 13.67 31.57 13.53 31.3402 0.99 1.64 1.27 1.01 0.30 0.26 0.00 0.00CO2 1.06 1.19 1.17 1.14 1.58 1.33 1.71 1.54CO 0.21 0.40 0.59 0.78 0.49 0.61 0.53 0.64C2H4 0.06 0.14 0.22 0.31 0.55 0.81 0.65 0.95C2H6 0.13 0.34 0.67 1.12 0.58 1.03 0.49 0.79C3's 0.00 0.01 0.03 0.06 0.04 0.07 0.03 0.05Total Carbon out 7.91 17.12 36.38 75.59 18.12 37.41 18.14 37.15Output flowsTime for 50m1 (s) 29.62 30.08 30.62 31.67 30.43 31.00 30.51 31.29Temperature (°C) 25.93 25.50 25.50 26.00 26.00 26.00 26.00 26.00mL/min (at 21.1°C) 99.67 98.27 96.54 93.19 96.96 95.20 96.73 94.31ilmol/min 4127 4069 3998 3859 4015 3942 4005 3905Output (pinol/min)CH4 249 592 1309 2725 549 1244 542 122402 39 67 51 39 12 10 0 0CO2 45 48 47 44 63 52 68 60CO 9 16 24 30 20 24 21 25C2H4 2 6 9 12 22 32 26 37C2H6 5 14 27 43 23 41 20 31C3's 0 0 1 2 2 3 1 2Total Carbon out 318 696 1454 2917 727 1475 727 1451psnol Carbon lost 60.9 148.6 119.0 217.8 117.6 98.9 118.6 122.7% Carbon lost 16.1 17.6 7.6 6.9 13.9 6.3 14.0 7.8154Table A.3 (continued)Calcium Doped Samarium Oxide (1:100) Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8STY Lunol/g/sC2s 2.6 65 11.9 18.4 15.1 24.2 15.2 22.6% ConversionCH4 21.0 15.0 10.0 6.6 24.6 15.6 25.4 15.602 80.2 67.1 74.6 79.8 93.9 94.9 100.0 100.0% SelectivityCO2 64.2 46.2 32.2 22.9 35.5 22.8 37.1 26.5CO 12.9 15.6 16.3 15.8 11.0 10.5 11.6 11.0COX selectivity 77.1 61.8 48.4 38.7 46.5 33.3 48.6 37.4C2H4 6.8 10.6 12.2 12.4 24.7 27.7 28.2 32.9C2H6 16.1 26.4 36.9 45.3 26.1 35.4 21.2 27.1C2 selectivity 22.9 37.0 49.1 57.6 50.8 63.1 49.4 60.0Cg's 0.0 1.2 2.5 3.6 2.7 3.6 2.0 2.6C2+ selectivity 22.9 38.2 51.6 61.3 53.5 66.7 51.4 62.6% YieldCO2 13.5 7.0 3.2 1.5 8.7 3.6 9.4 4.1CO 2.7 2.3 1.6 1.0 2.7 1.6 2.9 1.7COX yield 16.2 9.3 4.9 2.5 11.4 5.2 12.4 5.9C2H4 1.4 1.6 1.2 0.8 6.1 4.3 7.2 5.1C2H6 3.4 4.0 3.7 3.0 6.4 5.5 5.4 4.2C2 yield 4.8 5.6 4.9 3.8 12.5 9.9 12.6 9.4Cg's 0.0 0.2 0.2 0.2 0.7 0.6 0.5 0.4C2+ yield 4.8 5.7 5.2 4.0 13.1 10.4 13.1 9.8155Table A.4Magnesium Doped Samarium Oxide (1:100) Prepared in AirAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8Temperature (°C) 650 650 650 650 750 750 850 850Products (%)CH4 6.47 14.87 32.54 70.31 13.68 31.50 13.69 31.9602 2.34 1.93 1.56 1.20 0.45 0.29 0.03 0.00CO2 0.97 1.07 1.05 1.00 1.47 1.22 1.66 1.50CO 0.17 0.30 0.45 0.66 0.45 0.57 0.55 0.68C2H4 0.06 0.13 0.21 0.32 0.50 0.74 0.58 0.83C2H6 0.12 0.30 0.59 1.01 0.54 0.96 0.47 0.75C3's 0.00 0.01 0.03 0.06 0.04 0.07 0.03 0.05Total Carbon out 7.96 17.14 35.73 74.83 17.81 36.90 18.09 37.44% ConversionCH4 18.8 13.3 8.9 6.0 23.2 14.6 24.3 14.602 53.2 61.3 68.9 75.9 91.0 94.2 99.5 100.0% SelectivityCO2 64.7 46.9 32.9 22.2 35.5 22.6 37.8 27.5CO 11.2 13.2 14.2 14.5 10.8 10.6 12.5 12.4COx selectivity 75.9 60.2 47.1 36.8 46.4 33.2 50.3 39.8C2H4 7.6 11.4 13.4 14.3 24.4 27.4 26.4 30.2C2H6 16.5 26.7 36.7 44.9 26.3 35.5 21.2 27.3C2 selectivity 24.1 38.1 50.1 59.2 50.7 62.9 47.6 57.5Cs's 0.0 1.7 2.8 4.0 2.9 3.9 2.0 2.7C2+ selectivity 24.1 39.8 52.9 63.2 53.6 66.8 49.7 60.2% YieldCO2 12.1 6.2 2.9 1.3 8.2 3.3 9.2 4.0CO 2.1 1.8 1.3 0.9 2.5 1.5 3.0 1.8COx yield 14.2 8.0 4.2 2.2 10.7 4.9 12.2 5.8C2H4 1.4 1.5 1.2 0.9 5.7 4.0 6.4 4.4C21-16 3.1 3.5 3.3 2.7 6.1 5.2 5.2 4.0C2 yield 4.5 5.1 4.5 3.6 11.8 9.2 11.6 8.4Cs's 0.0 0.2 0.3 0.2 0.7 0.6 0.5 0.4C2+ yield 4.5 5.3 4.7 3.8 12.4 9.8 12.1 8.8156Table A.5Magnesium Doped Samarium Oxide (1:100) Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Temperature (°C) 650 650 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 10 20 40 80 20 40 20 4002 5 5 5 5 5 5 5 5He 85 75 55 15 75 55 75 55CH4 /02 2 4 8 16 4 8 4 8Average Average Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Products (%)CH4 6.35 14.50 31.77 69.37 13.68 30.87 13.35 31.1802 2.34 1.70 1.39 1.14 0.45 0.33 0.00 0.00CO2 0.98 1.10 1.10 1.05 1.47 1.25 1.65 1.45CO 0.13 0.27 0.41 0.62 0.45 0.56 0.57 0.70C2H4 0.05 0.13 0.21 0.30 0.50 0.77 0.63 0.91C2116 0.12 0.32 0.63 1.06 0.54 1.01 0.49 0.79C3's 0.00 0.01 0.04 0.06 0.04 0.08 0.03 0.06Total Carbon out 7.80 16.79 35.08 73.95 17.81 36.48 17.89 36.92Output flowsTime for 50m1 (s) 30.16 30.08 30.65 32.21 30.73 31.28 30.84 31.53Temperature (°C) 23.67 24.00 24.57 24.93 24.93 25.23 25.13 24.90mL/min (at 21.1°C) 98.62 98.75 96.75 91.95 96.36 94.57 95.97 93.94gmolimin 4084 4089 4006 3807 3990 3916 3974 3890Output (i.unol/min)CH4 259 593 1273 2641 530 1209 530 121302 96 70 56 44 13 13 0 0CO2 40 45 44 40 60 49 66 57CO 5 11 17 23 17 22 23 27C2H4 2 5 9 11 20 30 25 35C2H6 5 13 25 40 22 40 19 31C3's 0 1 1 2 2 3 1 2Total Carbon out 318 687 1405 2815 697 1429 711 1436grnol Carbon lost 60.5 158.4 168.3 319.1 148.5 144.8 134.2 137.6% Carbon lost 16.0 18.7 10.7 10.2 17.6 9.2 15.9 8.7157Table A.5 (continued)Magnesium Doped Samarium Oxide (1:100) Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8STY timol/g/sC2's 2.5 5.4 8.4 11.6 12.4 17.4 15.9 20.9% ConversionCH4 18.6 13.7 9.4 6.2 23.2 15.4 25.4 15.502 53.2 65.9 72.3 77.1 91.0 93.4 100.0 100.0% SelectivityCO2 67.5 48.1 33.3 22.9 35.5 22.3 36.4 25.3CO 9.0 11.6 12.5 13.5 10.8 10.0 12.6 12.2COX selectivity 76.5 59.7 45.8 36.4 46.4 32.3 49.0 37.6C2H4 6.9 11.0 12.9 13.1 24.4 27.5 27.6 31.7C2H6 16.6 27.6 37.9 46.4 26.3 36.0 21.4 27.6C2 selectivity 23.5 38.6 50.9 59.5 51.1 63.5 49.0 59.3C3's 0.0 1.7 3.3 4.1 2.9 4.3 2.0 3.1C2+ selectivity 23.5 40.3 54.2 63.6 53.6 67.7 51.0 62.4% YieldCO2 12.5 6.6 3.1 1.4 8.2 3.4 9.2 3.9CO 1.7 1.6 1.2 0.8 2.5 1.5 3.2 1.9CO, yield 14.2 8.2 4.3 2.3 10.7 5.0 12.4 5.8C2H4 1.3 1.5 1.2 0.8 5.7 4.2 7.0 4.9C2H6 3.1 3.8 3.6 2.9 6.1 5.5 5.4 4.3C2 yield 4.4 5.3 4.8 3.7 12.2 9.8 12.4 9.2C3's 0.0 0.2 0.3 0.3 0.7 0.7 0.5 0.5C2+ yield 4.4 5.5 5.1 3.9 12.4 10.4 12.9 9.7158Table A.6Sodium Doped Samarium Oxide (1:100) Prepared in AirRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Temperature (°C) 650 650 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 10 20 40 80 20 40 20 4002 5 5 5 5 5 5 5 5He 85 75 55 15 75 55 75 55CH4/02 2 4 8 16 4 8 4 8Average Average Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Products (%)CH4 7.55 16.12 33.79 72.49 14.15 31.78 13.84 31.8602 3.39 2.80 2.40 2.02 0.78 0.59 0.06 0.03CO2 0.63 0.84 0.97 1.05 2.07 1.87 2.38 2.10CO 0.21 0.34 0.50 0.80 0.40 0.50 0.51 0.64C2H4 0.02 0.06 0.12 0.18 0.56 0.82 0.66 0.93C2H6 0.08 0.25 0.53 0.94 0.65 1.15 0.53 0.86C3s 0.00 0.01 0.02 0.04 0.05 0.09 0.03 0.06Total Carbon out 8.60 17.95 36.61 76.68 19.20 38.37 19.21 38.37Output flowsTime for 50m1 (s) 29.18 29.53 29.78 32.46 30.50 30.91 30.63 31.06Temperature (°C) 25.57 26.07 26.57 23.87 27.57 27.13 27.37 27.10mL/min (at 21.1°C) 101.26 99.91 98.91 91.62 96.25 95.11 95.91 94.65pmol/min 4193 4137 4096 3794 3985 3938 3971 3919Output (p.mol/min)CH4 317 667 1384 2750 564 1252 550 124802 142 116 98 77 31 23 3 1CO2 27 35 40 40 82 74 94 82CO 9 14 20 30 16 20 20 25C2H4 1 3 5 7 22 32 26 37C2H6 3 10 22 36 26 45 21 34C3's 0 0 1 2 2 4 1 2Total Carbon out 361 742 1500 2909 765 1511 763 1504pinol Carbon lost 18 103 74 226 80 62 82 70% Carbon lost 5 12 5 9 8 4 8 4159Table A.6 (continued)Sodium Doped Samarium Oxide (1:100) Prepared in AirAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8STY Lunol/g/sC2's 1.4 4.3 8.8 14.1 16.1 25.9 15.8 23.4% ConversionCH4 12.2 10.2 7.7 5.5 26.3 17.2 27.9 17.002 29.8 44.0 52.0 59.7 84.4 88.3 98.7 99.4% SelectivityCO2 60.3 45.8 34.5 25.0 40.9 28.4 44.3 32.3CO 20.0 18.6 17.8 19.1 8.0 7.6 9.6 9.9CO, selectivity 80.3 64.5 52.3 44.0 48.9 36.0 53.9 42.2C2H4 3.8 6.9 8.3 8.4 22.2 25.0 24.6 28.7C2H6 15.9 27.0 37.3 44.7 25.9 34.9 19.9 26.4C2 selectivity 19.7 33.9 45.6 53.1 48.1 59.9 44.5 55.1Cis 0.0 1.6 2.1 2.9 3.0 4.1 1.7 2.8C2+ selectivity 19.7 35.5 47.7 56.0 51.1 64.0 46.1 57.8% YieldCO2 7.4 4.7 2.7 1.4 10.8 4.9 12.4 5.5CO 2.4 1.9 1.4 1.0 2.1 1.3 2.7 1.7COX yield 9.8 6.6 4.0 2.4 12.9 6.2 15.0 7.2C2H4 0.5 0.7 0.6 0.5 5.8 4.3 6.9 4.9C2H6 1.9 2.7 2.9 2.4 6.8 6.0 5.6 4.5C2 yield 2.4 3.5 3.5 2.9 12.6 10.3 12.4 9.3C3's 0.0 0.2 0.2 0.2 0.8 0.7 0.5 0.5C2+ yield 2.4 3.6 3.7 3.1 13.4 11.0 12.9 9.8160Table A.7Sodium Doped Samarium Oxide (1:100) Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Temperature ( °C) 650 650 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 10 20 40 80 20 40 20 4002 5 5 5 5 5 5 5 5He 85 75 55 15 75 55 75 55CH4/02 2 4 8 16 4 8 4 8Average Average Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Products (%)CH4 7.59 16.20 34.09 73.03 14.30 32.51 14.21 32.6802 3.97 3.43 2.87 2.39 0.65 0.43 0.01 0.01CO2 0.80 0.93 1.12 1.26 1.80 1.54 1.85 1.62CO 0.19 0.28 0.50 0.54 0.36 0.47 0.54 0.68C2H4 0.02 0.06 0.11 0.19 0.60 0.88 0.69 1.01C2H6 0.09 0.23 0.50 0.96 0.68 1.19 0.56 0.90C3's 0.00 0.01 0.02 0.04 0.05 0.10 0.03 0.06Total Carbon out 8.81 18.00 37.01 77.23 19.18 38.95 19.17 38.97Output flowsTime for 50m1 (s) 29.42 29.58 30.11 31.34 30.84 31.59 31.22 31.65Temperature (°C) 23.60 24.57 25.13 25.20 24.07 23.63 23.33 24.10mL/min (at 21.1°C) 101.12 100.25 98.30 94.42 96.31 94.16 95.38 93.82gmol/min 4187 4151 4070 3910 3988 3899 3949 3885Output (p.mol/min)CH4 318 672 1388 2855 570 1267 561 127002 166 142 117 93 26 17 0 0CO2 34 39 46 49 72 60 73 63CO 8 12 20 21 14 18 21 27C2H4 1 2 5 7 24 34 27 39C2H6 4 9 20 37 27 46 22 35C3's 0 0 1 1 2 4 1 2Total Carbon out 369 747 1506 3020 765 1518 757 1514iimol Carbon lost 10 98 67 115 80 55 88 60% Carbon lost 3 12 4 5 8 3 9 4161Table A.7 (continued)Sodium Doped Samarium Oxide (1:100) Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8STY gmol/g/sC2's 1.5 3.9 8.4 14.9 17.0 26.8 16.4 24.6% ConversionCH4 13.8 10.0 7.9 5.4 25.4 16.5 25.9 16.102 17.9 31.5 42.5 52.3 87.0 91.4 99.8 99.8% SelectivityCO2 66.1 51.6 38.5 30.0 36.9 23.9 37.3 25.7CO 15.7 15.8 17.1 12.8 7.4 7.3 10.8 10.9COx selectivity 81.8 67.4 55.6 42.8 44.2 31.2 48.1 36.6C2H4 3.9 6.3 7.8 9.0 24.6 27.2 27.7 32.0C2H6 14.3 25.2 34.5 45.6 27.9 36.9 22.4 28.5C2 selectivity 18.2 31.5 42.3 54.6 52.5 64.1 50.1 60.6C3's 0.0 1.1 2.1 2.6 3.3 4.7 1.8 2.9C2+ selectivity 18.2 32.6 44.4 57.2 55.8 68.8 51.9 63.4% YieldCO2 9.1 5.2 3.0 1.6 9.4 4.0 9.7 4.1CO 2.2 1.6 1.4 0.7 1.9 1.2 2.8 1.8COX yield 11.3 6.8 4.4 2.3 11.3 5.2 12.5 5.9C2H4 0.5 0.6 0.6 0.5 6.3 4.5 7.2 5.2C2H6 2.0 2.5 2.7 2.5 7.1 6.1 5.8 4.6C2 yield 2.5 3.1 3.3 3.0 13.3 10.6 13.0 9.8C3's 0.0 0.1 0.2 0.1 0.8 0.8 0.5 0.5C2+ yield 2.5 3.3 3.5 3.1 14.2 11.4 13.4 10.2162Table A.8Potassium Doped Samarium Oxide (1:100) Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Temperature (°C) 650 650 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 10 20 40 80 20 40 20 4002 5 5 5 5 5 5 5 5He 85 75 55 15 75 55 75 55CH4/02 2 4 8 16 4 8 4 8Average Average Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8Products (%)CH4 7.79 16.66 34.95 75.52 14.44 32.95 14.35 32.9602 4.04 3.23 2.47 1.91 0.63 0.53 0.04 0.03CO2 0.94 1.21 1.42 1.51 2.01 1.77 2.10 1.91CO 0.20 0.34 0.49 0.59 0.42 0.55 0.54 0.69C2H4 0.02 0.07 0.15 0.27 0.54 0.80 0.66 0.97C2H6 0.09 0.25 0.58 1.09 0.62 1.08 0.53 0.87C3s 0.00 0.01 0.03 0.06 0.05 0.09 0.03 0.06Total Carbon out 9.16 18.87 38.39 80.51 19.35 39.30 19.45 39.43Output flowsTime for 50m1 (s) 29.55 29.89 30.34 31.73 30.57 31.18 30.67 31.29Temperature (°C) 24.60 25.00 25.20 26.00 26.00 26.40 26.87 27.00mL/min (at 21.1°C) 100.32 99.06 97.51 92.99 96.52 94.50 95.94 94.00p.mol/min 4154 4102 4038 3850 3997 3913 3972 3892Output (grnol/min)CH4 324 683 1411 2908 577 1289 570 128302 168 133 100 74 25 21 1 1CO2 39 50 57 58 80 69 83 74CO 8 14 20 23 17 22 21 27C2H4 1 3 6 10 22 31 26 38C2H6 4 10 23 42 25 42 21 34C3's 0 0 1 2 2 3 1 2Total Carbon out 380 774 1550 3100 773 1538 772 1535innol Carbon lost -1.6 71.0 23.6 34.5 71.9 35.7 72.6 38.8% Carbon lost -0.4 8.4 1.5 1.5 7.3 2.3 7.3 2.5163Table A.8 (continued)Potassium Doped Samarium Oxide (1:100) Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6AverageRun 7AverageRun 8STY gmol/g/sC2's 1.5 4.3 9.7 17.4 15.5 24.5 15.7 23.9% ConversionCH4 14.9 11.7 8.9 6.2 25.4 16.1 26.2 16.402 17.0 35.3 50.5 61.8 87.4 89.5 99.3 99.4% SelectivityCO2 69.0 54.8 41.3 30.2 41.0 27.9 41.1 29.6CO 14.9 15.2 14.3 11.9 8.5 8.7 10.5 10.6CO, selectivity 83.9 70.0 55.6 42.1 49.5 36.6 51.7 40.2C2H4 2.9 6.0 8.5 10.7 22.1 25.2 25.8 30.1C2H6 13.2 22.6 33.6 43.7 25.3 34.1 20.8 26.9C2 selectivity 16.1 28.7 42.1 54.3 47.4 59.3 46.6 57.0C3's 0.0 1.4 2.3 3.6 3.1 4.1 1.8 2.8C2+ selectivity 16.1 30.0 44.4 57.9 50.5 63.4 48.3 59.8% YieldCO2 10.3 6.4 3.7 1.9 10.4 4.5 10.8 4.9CO 2.2 1.8 1.3 0.7 2.2 1.4 2.8 1.7COX yield 12.5 8.2 5.0 2.6 12.6 5.9 13.5 6.6C2H4 0.4 0.7 0.8 0.7 5.6 4.1 6.8 4.9C2H6 2.0 2.6 3.0 2.7 6.4 5.5 5.5 4.4C2 yield 2.4 3.4 3.8 3.4 12.0 9.6 12.2 9.4C3's 0.0 0.2 0.2 0.2 0.8 0.7 0.5 0.5C2+ yield 2.4 3.5 4.0 3.6 12.8 10.2 12.7 9.8164Table A.9Calcium Doped Samarium Oxide (1:10) Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6Temperature (°C) 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 20 40 20 40 20 4002 5 5 5 5 5 5He 75 55 75 55 75 55CH4/02 4 8 4 8 4 8Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6Products (%)CH4 16.73 35.06 14.41 32.76 14.14 32.7502 2.07 1.77 0.55 0.35 0.01 0.00CO2 0.90 1.16 1.68 1.48 1.84 1.62CO 0.30 0.50 0.39 0.49 0.50 0.61C2H4 0.06 0.12 0.60 0.91 0.78 1.11C2H6 0.24 0.54 0.69 1.22 0.57 0.90Cg's 0.01 0.02 0.05 0.10 0.04 0.07Total Carbon out 18.54 38.11 19.22 39.27 19.31 39.21Output flowsTime for 50m1 (s) 30.20 30.38 30.74 31.36 30.69 31.48Temperature (°C) 21.73 22.73 23.00 23.43 23.47 23.50mL/min (at 21.1°C) 99.14 98.20 96.98 94.92 96.96 94.52gmo1/min 4105 4066 4016 3930 4015 3914Output (p.mol/min)CH4 687 1426 579 1287 568 128202 85 72 22 14 0 0CO2 37 47 68 58 74 63CO 12 20 16 19 20 24C2H4 2 5 24 36 31 43C2H6 10 22 28 48 23 35Cg's 0 1 2 4 2 3Total Carbon out 761 1549 772 1544 775 1535iimol Carbon lost 84 24 73 30 70 39% Carbon lost 10 2 9 2 8 2165Table A.9 (continued)Calcium Doped Samarium Oxide (1:10) Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6STY mol/g/sC2s 4.0 8.9 17.4 27.8 18.1 26.2% ConversionCH4 9.8 8.0 25.0 16.6 26.8 16.502 58.1 64.5 88.9 92.9 99.8 99.9% SelectivityCO2 49.4 38.1 35.0 22.7 35.7 25.1CO 16.5 16.5 8.0 7.5 9.6 9.5COX selectivity 65.9 54.6 43.0 30.2 45.3 34.6C2H4 6.3 8.1 25.1 27.8 30.2 34.2C21-1 16 26.1 35.3 28.8 37.4 22.2 28.0C2 selectivity 32.4 43.4 53.9 65.2 52.4 62.2C3s 1.7 2.0 3.1 4.6 2.3 3.2C2+ selectivity 34.1 45.4 57.0 69.8 54.7 65.4% YieldCO2 4.8 3.0 8.8 3.8 9.5 4.1CO 1.6 1.3 2.0 1.2 2.6 1.6COX yield 6.5 4.4 10.8 5.0 12.1 5.7C2H4 0.6 0.6 6.3 4.6 8.1 5.6C2H6 2.6 2.8 7.2 6.2 5.9 4.6C2 yield 3.2 3.5 13.5 10.8 14.0 10.3Cg's 0.2 0.2 0.8 0.8 0.6 0.5C2+ yield 3.3 3.6 14.3 11.6 14.6 10.8166Table A.10Sodium Doped Samarium Oxide (1:10) Prepared in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6Temperature (°C) 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 20 40 20 40 20 4002 5 5 5 5 5 5He 75 55 75 55 75 55CH4/02 4 8 4 8 4 8Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6Products (%)CH4 17.15 35.23 15.09 32.60 14.17 32.3602 4.29 3.82 2.15 1.07 0.00 0.00CO2 0.52 0.47 1.17 1.31 2.02 1.68CO 0.08 0.07 0.10 0.16 0.40 0.64C2H4 0.01 0.03 0.40 0.72 0.63 0.90C2H6 0.11 0.24 0.63 1.16 0.53 0.81C3's 0.00 0.00 0.03 0.08 0.03 0.06Total Carbon out 17.98 36.31 18.50 38.08 19.01 38.28Output flowsTime for 50m1 (s) 28.76 29.44 29.82 30.80 30.33 30.91Temperature (°C) 26.00 26.00 26.17 26.17 26.17 26.13mL/min (at 21.1°C) 102.61 100.23 98.90 95.76 97.25 95.43gmol/min 4249 4150 4095 3965 4027 3952Output (gmol/min)CH4 729 1462 618 1293 571 127902 182 159 88 42 0 0CO2 22 19 48 52 81 67CO 3 3 4 6 16 25C2H4 0 1 16 29 26 36C2H6 5 10 26 46 21 32C3's 0 0 1 3 1 2Total Carbon out 764 1507 758 1510 765 1513gmol Carbon lost 81 66 87 64 80 61% Carbon lost 10 4 10 4 9 4167Table A.10 (continued)Sodium Doped Samarium Oxide (1:10) Prepared in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6STY j.unol/g/sC2's 1.7 3.8 14.0 24.8 15.6 22.6% ConversionCH4 4.6 3.0 18.4 14.4 25.4 15.502 10.0 23.6 56.9 78.6 100.0 100.0% SelectivityCO2 62.4 42.1 34.4 23.9 41.8 28.4CO 9.0 6.9 2.8 3.0 8.3 10.9COx selectivity 71.4 48.9 37.2 26.9 50.1 39.3C2H4 2.5 5.6 23.4 26.3 26.2 30.5C2H6 26.2 45.5 36.7 42.4 21.8 27.3C2 selectivity 28.6 51.1 60.2 68.7 48.0 57.8C3's 0.0 0.0 2.6 4.4 1.9 2.9C2+ selectivity 28.6 51.1 62.8 73.1 49.9 60.7% YieldCO2 2.9 1.3 6.3 3.4 10.6 4.4CO 0.4 0.2 0.5 0.4 2.1 1.7COx yield 3.3 1.5 6.9 3.9 12.8 6.1C2H4 0.1 0.2 4.3 3.8 6.7 4.7C2H6 1.2 1.3 6.8 6.1 5.5 4.2C2 yield 1.3 1.5 11.1 9.9 12.2 9.0C3's 0.0 0.0 0.5 0.6 0.5 0.4C2+ yield 1.3 1.5 11.6 10.5 12.7 9.4168Table A.11Calcium Doped Samarium Oxide (1:10) Revised Preparation (RP) in OxygenRun 1 Run 2 Run 3 Run 4 Run 5 Run 6Temperature (°C) 650 650 750 750 850 850Input flows(mL/min)Nominal(at 21.1°C)CH4 20 40 20 40 20 4002 5 5 5 5 5 5He 75 55 75 55 75 55CH4 /02 4 8 4 8 4 8Average Average Average Average Average AverageRun 1 Run 2 Run 3 Run 4 Run 5 Run 6Products (%)CH4 15.65 34.19 15.26 34.00 15.25 34.2802 0.99 0.61 0.00 0.00 0.00 0.00CO2 2.16 2.06 2.55 2.43 2.41 2.23CO 0.60 0.87 0.55 0.76 0.47 0.62C2H4 0.06 0.08 0.18 0.26 0.26 0.37C2H6 0.20 0.38 0.30 0.52 0.29 0.48C3's 0.00 0.01 0.01 0.01 0.01 0.01Total Carbon out 18.94 38.07 19.36 38.78 19.25 38.85Output flowsTime for 50m1 (s) 30.02 30.13 30.22 30.63 30.32 30.62Temperature (°C) 24.93 26.00 26.37 26.63 26.63 26.40mL/min (at 21.1°C) 98.66 97.94 97.52 96.15 97.13 96.25umol/rnin 4085 4055 4038 3981 4022 3985Output (urnol/min)CH4 650 1383 614 1334 609 134102 41 25 0 0 0 0CO2 90 83 103 95 96 87CO 25 35 22 30 19 24C2H4 2 3 7 10 10 14C2H6 8 15 12 21 12 19C3's 0 0 0 0 0 0Total Carbon out 786 1540 779 1522 768 1521gmol Carbon lost 59 34 66 52 77 53% Carbon lost 7 2 8 3 9 3169Table A.11 (continued)Calcium Doped Samarium Oxide (1:10) Revised Preparation (RP) in OxygenAverageRun 1AverageRun 2AverageRun 3AverageRun 4AverageRun 5AverageRun 6STY wriol/g/sC2's 3.6 6.2 6.5 10.2 7.3 11.0% ConversionCH4 17.4 10.2 21.2 12.3 20.8 11.802 79.8 87.9 100.0 100.0 100.0 100.0% SelectivityCO2 65.7 53.2 62.3 50.8 60.1 48.8CO 18.2 22.3 13.4 15.8 11.8 13.5CO,, selectivity 83.9 75.5 75.7 66.6 72.0 62.3C2H4 3.4 4.3 8.9 10.9 12.8 16.2C2H6 12.4 19.4 14.6 21.9 14.5 20.8C2 selectivity 15.8 23.7 23.6 32.8 27.3 37.0C3's 0.3 0.8 0.7 0.6 0.7 0.7C2+ selectivity 16.1 24.5 24.3 33.4 28.0 37.7% YieldCO2 11.4 5.4 13.2 6.3 12.5 5.7CO 3.2 2.3 2.8 2.0 2.5 1.6COx yield 14.6 7.7 16.0 8.2 15.0 7.3C2H4 0.6 0.4 1.9 1.3 2.7 1.9C21-16 2.1 2.0 3.1 2.7 3.0 2.5C2 yield 2.7 2.4 5.0 4.0 5.7 4.4C3's 0.1 0.1 0.2 0.1 0.2 0.1C2+ yield 2.8 2.5 5.1 4.1 5.8 4.4170APPENDIX B: SCANNING ELECTRON MICROSCOPY (SEM) PHOTOGRAPHS171Figure B.11:100 Ca:Sm Photograph 192Figure B.21:10 Ca:Sm Photograph 193172Figure B.41:10 Ca:Sm (RP) Photograph 105----..... e^—^do^w c ,.1,4 •1 °^4^t *Nwso.W,^It*•^CU1.91•4CU." •116.01111imeFigure B.31:10 Ca:Sm Photograph 194rt. troll^"""i^41173Figure B.51:100 Na:Sm Photograph 195Figure B.6Sm203 Photograph 102••• ICIU• ••■41•.114•a,a• C• Ceg)CuCu4.-11inCu174Figure B.71:100 K:Sm Photograph 103Figure B.81:100 Mg:Sm Photograph 104175APPENDIX C: X-RAY DIFFRACTION (XRD) GRAPHS1762-Theta - Scale00UNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:565^10^15^20^25^30 35 40 45 SO SS12-Theta - Scale UNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:390-4-)cO0cs)-^•^r^.30^35 .^1^. •^•^1^.40^45 50^55i2-Theta - ScaleUNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:461 0 I^'^'^I30 35•40^45• •50^SS2-Theta - ScaleUNIVERSITY OF BRITISH COLUMBIA 26-Mar-1993 09:43Na0S^10^15^20^25^30^3S^40^4I501 DjUi002-Theta0U)U)N- Scale^ UNIVERSITY OF BRITISH COLUMBIA 16-Mar-1993 12:40CO"faNriJJfD01c)00z0zs)i.1.4:11:41is j 'la..lg'^.ig& 1 li^,.44, , I..: i.^,„; .6. Ir. i i...,^ '..A ;...i. , Al, ; ;,..41.!^4 '^o ..,^LiIII 4^4i 'lit ii l '4 % ■ 3 ILI iSO10 15^20^25^30^35^40^45 S SIf)nzr

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